Optimization of the C-terminus of an autonomous human IgG1 CH2

CAS Key Laboratory of Special Pathogens and Biosafety. Wuhan Institute of Virology. Page 1 of 43. ACS Paragon Plus Environment. Molecular Pharmaceutic...
1 downloads 0 Views 1MB Size
Subscriber access provided by BUFFALO STATE

Article

Optimization of the C-terminus of an autonomous human IgG1 CH2 domain for stability and aggregation resistance Xinyu Gao, Alex Conard, Chunpeng Yang, Yancheng Zhan, Fang Zeng, Jian Shi, Wei Li, Dimiter S. Dimitrov, and Rui Gong Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.9b00544 • Publication Date (Web): 08 Jul 2019 Downloaded from pubs.acs.org on July 18, 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 43 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

Molecular Pharmaceutics

Optimization of the C-terminus of an autonomous human IgG1 CH2 domain for stability and aggregation resistance

Xinyu Gao1,2, Alex Conard3, Chunpeng Yang1,2, Yancheng Zhan1,2, Fang Zeng1,2, Jian Shi1, Wei Li3, Dimiter S. Dimitrov3, Rui Gong1*

1CAS

Key Laboratory of Special Pathogens and Biosafety, Wuhan Institute of Virology,

Chinese Academy of Sciences, Wuhan, Hubei 430071, China; 2University of Chinese Academy of Sciences, Beijing 100049, China; 3Center for Antibody Therapeutics, University of Pittsburgh Medical School, Pittsburgh, Pennsylvania 15261, USA.

*To whom correspondence should be addressed: Rui Gong Center for Emerging Infectious Diseases CAS Key Laboratory of Special Pathogens and Biosafety Wuhan Institute of Virology

ACS Paragon Plus Environment

1

Molecular Pharmaceutics 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 43

Chinese Academy of Sciences No.44 Xiao Hong Shan Wuhan, Hubei 430071, P. R. China E-mail: [email protected] Tel: +86-27-87003936, +86-27-87199331 Fax: +86-27-87806108

KEYWORDS. CH2 domain; phage display; stability; aggregation resistance; physicochemical property; CH2-based therapeutics; Ig-like protein

ABSTRACT: The IgG1 CH2 domain is involved in Fc-mediated effector functions and is a promising scaffold for development of novel therapeutics. We previously reported that removal of seven unstructured N-terminal residues of an autonomous human IgG1 CH2 domain significantly increased its stability and aggregation resistance. However, the way in which the C-terminal residues affect folding is unclear. Here, we found that the

ACS Paragon Plus Environment

2

Page 3 of 43 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

Molecular Pharmaceutics

CH2 C-terminus is highly sensitive to truncation although these residues adopt random coil conformation according to the crystal structure of a CH2 domain. To optimize the Cterminus, we used phage display platform for high-throughput screening of mutants with improved physicochemical properties. After panning of the CH2 mutant library at high temperature against a CH2-specific antibody recognizing conformational epitope, we obtained two candidates, B3 and D9, with markedly increased thermal stability. We found that substitution of K338 (EU numbering) by isoleucine is crucial for the increased stability, which might be due to enhanced hydrophobic interactions involving W313. However, the aggregation propensity was also increased. To reduce the aggregation propensity, we mutated several residues in the same region by rational design and identified a mutant, CH2-IKS (K338I, A339K and K340S), with high stability and aggregation resistance. In summary, the C-terminus of CH2 is important for its folding and could be further optimized toward better potential application for CH2-based therapeutics. Our strategy might be also useful for stabilization of other Ig-like proteins.

ACS Paragon Plus Environment

3

Molecular Pharmaceutics 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 43

Introduction There are more than 70 therapeutic monoclonal antibodies (mAbs) on the market for treatment of cancer, autoimmune disease, infection, and other diseases1. However, due to their large size ( ~ 150 kDa), they demonstrate poor tissue penetration (e.g., solid tumor2) and awful accessibility to sterically restricted epitopes on the surface of some molecules (e.g., viral envelope protein3). Meanwhile, inter- and intramolecular disulfide bonds render these molecules vulnerable to environmental changes4. These limitations hinder their clinical use. An attractive alternative to overcome these drawbacks is to develop immunoglobulin (Ig) domain-derived5 and non-Ig domain-based binders6 with reduced size. Among these binders, antibody heavy chain variable domain (VH)-based single domain antibodies (V-sdAbs) are popular as next generation antibody therapeutics7. We have proposed that the IgG1 second constant domain (CH2 domain) (~12.5 kDa) from the antibody Fc fragment could be used as a novel scaffold for the development

of

C-based

single

domain

antibodies

(C-sdAbs,

also

termed

nanoantibodies) and as therapeutic candidates8, 9. Construction of large phage display libraries and subsequent expression in prokaryotic expression systems are relatively

ACS Paragon Plus Environment

4

Page 5 of 43 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

Molecular Pharmaceutics

easy due to the small molecular weight of CH2. In addition, it exhibits relatively long serum half-life (~10 hours) in different animal models compared to VH (within minute range) due to residual binding to neonatal Fc receptor (FcRn)10-12.

Unlike other IgG domains, CH2 domain, as a typical constant 1 set (C1-set) immunoglobulin-like (Ig-like) domain13, is an independent folding domain that only interacts with the partner CH2 domain weakly by its sugar moieties at N297 position14 (all the CH2 residues here are numbered according to EU numbering15). CH2 domain exhibits a sandwich structure containing seven β-strands connected by three flexible loops and two helices16,

17.

These β-strands constitute two β-sheets to further form a

compact barrel covalently stabilized by a central, conserved disulfide bridge18. Autonomous CH2 is a monomer which possesses a structure like that found in an intact IgG15. The flexible loops in CH2 are functionally similar to the complementaritydetermining regions (CDRs) in VH. Hence, they could be mutated to other residues with sequence and length diversities to mimic the CDRs. Indeed, several binders specifically targeting viral antigens (e.g, envelope protein of HIV-119,

20)

or tumor antigens (e.g.,

ACS Paragon Plus Environment

5

Molecular Pharmaceutics 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 43

EphA221 and nucleolin22) have been selected from different libraries, which shows the proof-of-concept of the potentials of C-sdAbs as CH2-based therapeutics.

However, autonomous wildtype CH2 domain is relatively unstable compared to other Ig domains and single domain scaffolds23-26. This drawback limits CH2 to be a qualified antigen recognition unit. To make it a more suitable monomeric scaffold, a disulfide bond was introduced between strands A and G to stabilize human IgG1 CH2 (m01)25. In the same study, it was also found that the N-terminus is relatively flexible whereas the C-terminus is more rigid as measured by nuclear magnetic resonance (NMR). Following this study, a shortened CH2 (CH2s), with improved stability and aggregation resistance, was identified via removal of seven unstructured residues at the N-terminus27. Unfortunately, this strategy could not be directly adapted to engineer the C-terminus although four residues seem to be also in random states according to the crystal structure of an autonomous CH2 domain. Removal of even a single C-terminal amino acid residue led to dramatic loss of soluble expression, indicating that these C-terminal residues are critical for the structural stability of the CH2 domain. As direct truncation

ACS Paragon Plus Environment

6

Page 7 of 43 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

Molecular Pharmaceutics

did not work, we thought that the C-terminal residues could be substituted by other residues to improve physiochemical properties.

Here, we constructed a phage-display library with random mutation on the C-terminal residues of CH2 and screened of mutants with increased stability under high temperature selection pressure28. We obtained a mutant (CH2-IKS) with triple mutations (K338I, A339K and K340S) that displayed increased both stability and aggregation resistance compared to CH2 after panning and rational site mutagenesis. Our results indicate that the C-terminus of the CH2 domain is important for stability of the molecule. Better physiochemical properties of CH2 domain as scaffold may be achieved through further manipulation of its C-terminus.

Materials and methods Design, construction and expression of the CH2 and its truncated mutants Based on the structure of CH2, we chose seven residues at C-terminus from 334 to 340 to truncate. These mutants with one to seven residues truncated named 1 to 7. CH2

ACS Paragon Plus Environment

7

Molecular Pharmaceutics 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 43

and the truncated mutants with 6×His tag at the N-terminus and termination codon at the end of C-terminus were cloned in phagemid vector pComb3XSS (Addgene, USA) by PCR. The recombinant plasmids were transformed into Escherichia coli (E. coli) strain HB2151 for expression in periplasm.

Library construction Primers used in construction of the library containing CH2 mutants with random mutations at C-terminus are listed in Table S1. DNA fragments containing mutated genes were amplified by PCR, and then were cloned into pComb3XSS plasmid for phage display library construction as described previously29.

Selection of CH2 mutants The panning step was performed under high-temperature selection pressure as previously reported methods modified by us28, 30. Briefly, 500 μL of phage library (1012 colony-forming units) in 1.5 mL Eppendorf Tube was heated at 80°C for 10 min to denature phage-displayed CH2 mutants and then cooled at 4°C for 20 min for refolding.

ACS Paragon Plus Environment

8

Page 9 of 43 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

Molecular Pharmaceutics

After centrifugation for 1 min at 13,000 g, 450 μL supernatant was taken out, mixed with 50 μL of phosphate buffer saline [PBS (NaCl: 137 mM; KCl: 2.683 mM; Na2HPO4: 8.1 mM; KH2PO4: 1.76 mM), pH 7.4] + 10% skim milk, and distributed into five wells in StripwellTM Microplate (Corning, USA). These wells were pre-coated with bovine serum albumin (BSA, 0.5 μg/well, overnight at 4°C) and blocked by PBS + 3% skim milk (incubation at 37°C for 1 h) before adding the supernatant. After incubation at 37°C for 1 h, the nonspecific binding clones were absorbed into the wells. Then, the supernatant was transferred to another five wells which were pre-coated with an anti-CH2 monoclonal antibody (0.5 μg/well, overnight at 4°C) as previously used31 and blocked as described above. The wells were incubated at 37°C for 2 h and washed ten times with PBS + 0.1% Tween-20 (PBST) followed by two times with PBS. An amount of 2 mL E.

coli TG1 at exponential phase was added into wells for infection. The phage containing CH2 mutants was rescued by M13KO7 helper phage. Panning was repeated four times. The candidate clones were subjected to monoclonal phage ELISA.

Monoclonal phage ELISA

ACS Paragon Plus Environment

9

Molecular Pharmaceutics 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 43

Monoclonal phage ELISA was performed to select candidate CH2 mutants with correct conformation after panning. Hundreds of colonies were picked up for analysis. TG1 only was used as negative control. The anti-CH2 monoclonal antibody (2 μg/mL) used in panning was coated in CostarTM Assay Plate (Corning, USA). A parallel plate coated with BSA (2 μg/mL) was used for detection of nonspecific binding. The horseradish peroxidase (HRP)-conjugated anti-M13 monoclonal antibody (GE Healthcare, USA) was used as secondary antibody. The clones with positive binding signal to the anti-CH2 antibody (reading > 2 at OD405) and negative binding signal to BSA (reading < 0.1 at OD405) were selected for sequencing and expression.

Expression and purification of CH2 All the positive clones obtained from monoclonal phage ELISA were re-cloned into the same vector with 6×His tag at N-terminus and a termination codon (TAG) at the end of C-terminus as described above. The plasmids were transformed into E. coli strain HB2151 for expression and purification through Ni-NTA resin as previously reported32. Purified proteins were dialyzed against PBS and stored at -80ºC.

ACS Paragon Plus Environment

10

Page 11 of 43 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

Molecular Pharmaceutics

Size exclusion chromatography (SEC) The purified protein samples were loaded into Superdex 75 10/300 GL column (GE Healthcare, USA) connected to an ÄKTA pure system (GE Healthcare, USA) at a concentration of 1.0 mg/mL in PBS to assess possible oligomer formation. PBS was selected as the mobile phase at the flow rate of 0.8 mL/min. The molecular mass standards including bovine serum albumin (67 kDa), β-lactoglobulin (35 kDa), cytochrome C (13.6 kDa), aprotinin (6.512 kDa) and vitamin B12 (1.355 kDa) were used to calculate the molecular weight of the proteins.

Circular dichroism (CD) spectroscopy measurements The secondary structure and thermal stability of proteins were determined by CD spectroscopy. The purified proteins (0.2 mg/mL) were in PBS, and the CD spectra were recorded on Applied Photophysics Chirascan-SF.3 spectrofluorometer (Applied Photophysics Ltd, UK). The secondary structures of these proteins were analyzed by using a 0.1 cm path length cuvette at 25°C. Temperature-induced unfolding was

ACS Paragon Plus Environment

11

Molecular Pharmaceutics 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 43

monitored at 216 nm by recording the CD signal in the temperature range of 30 - 85°C with a heating rate of 0.5°C/min to determine the melting temperature (Tm). The fraction folded (ff) of the protein is calculated as ff = ([θ] − [θu])/([θf] − [θu]). [θf] and [θu] are the mean residue ellipticities at 216 nm of the folded state at 30°C and the unfolded state of 85°C. The experiments were repeated twice.

Turbidity assays The optical density (OD) at 320 nm is a typical index to detect protein aggregation27. To perform the turbidity assay, purified protein samples were concentrated and filtered through a 0.22 μm filter, and then diluted in PBS at the final concentration of 1 mg/mL and the working solution volume was 500 μL. After incubation at 50°C, samples were put into a cuvette to record the OD320 values at different time points as required with ultraviolet spectrophotometer (Beijing Liuyi Biotechnology Co., Ltd., China). The experiments were repeated three times.

Urea-induced denaturation experiments

ACS Paragon Plus Environment

12

Page 13 of 43 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

Molecular Pharmaceutics

Urea was used as the chemical denaturant and intrinsic fluorescence was used to monitor protein unfolding33. Purified CH2 and its mutants at a concentration of 0.1 mg/mL in PBS with urea from 0 to 10 M were used to measure intrinsic fluorescence with excitation wavelength at 280 nm and emission wavelength at 340 nm at 25°C. After incubation at 4°C overnight, the intrinsic fluorescence of all samples was recorded on a EnVision™ (PerkinElmer, USA). The background fluorescence intensity of the solution (buffer + denaturant) was deducted from the sample fluorescence intensity. The fraction folded (ff) of the protein is calculated as ff = ([F] − [Fu])/([Ff] − [Fu]). [Ff] and [Fu] are the intrinsic fluorescence with excitation wavelength at 280 nm and emission wavelength at 340 nm at 25°C of the folded state in the absence of urea and the unfolded state in the presence of urea concentration of 10 M. The experiments were repeated twice.

Dynamic light scattering (DLS) measurements For comparison of aggregation tendency, both CH2 and CH2-IKS protein were prepared as described in “turbidity assay” section. Protein samples were incubation at 37°C in seven day, which were detected on day 0, day 1 and day 7. DLS measurements were

ACS Paragon Plus Environment

13

Molecular Pharmaceutics 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 43

performed using the Zetasizer Nano S (Malvern Panalytical Ltd, UK). 450 μL of sample (1 mg/mL) was added to 1.5 mL narrow chamber disposable cuvette. The cuvette was placed in the instrument and allowed to calibrate at 25°C for 180 sec. The selected dispersant was ICN PBS Tablets (RI = 1.330; viscosity = 0.8882 cP). Three measurements were performed or each sample with “Measurement Duration” set to “Automatic”. The experiments were repeated three times.

Trypsin digestion assays The purified CH2 and CH2-IKS were subjected to trypsin digestion at 1:20 (molar ratio, trypsin: protein) for 0, 30, 60 and 90 min at 37°C. After digestion, loading buffer was added to samples and boiled immediately to terminate digestion. All samples were examined by SDS-PAGE.

Results The C-terminal residues of CH2 are critical for its soluble expression

ACS Paragon Plus Environment

14

Page 15 of 43 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

Molecular Pharmaceutics

In our previous study, we removed the first seven unstructured N-terminal residues of CH2 to obtain a truncated CH2 mutant (CH2s) with significantly improved stability27. Here, we found that, like the N-terminus, the C-terminus of CH2 also has several unstructured residues according to the crystal structure of an isolated CH2 domain (PDB 3D9J)

16

(Figure 1a). We removed one, two …, and seven residues from the C-

terminus of CH2 resulting in seven shortened CH2 mutants (Figure 1b). However, soluble expression in E. coli periplasm could not be achieved for any of the seven mutants (Figure 1c) indicating that the C-terminal residues are indispensable for the soluble expression of CH2 and may influence the stability of CH2 by a different mechanism from the N-terminal residues.

ACS Paragon Plus Environment

15

Molecular Pharmaceutics 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 43

Figure 1. The design and analysis of CH2 C-terminal truncation mutants. (a) Crystal structure of autonomous CH2 (PDB: 3DJ9). (b) Sequence alignment of the seven truncation mutants (c) Expression of wild-type and mutant CH2. Only CH2 could be expressed as indicated by arrow. M: protein molecular mass marker (from top to bottom: 25 kDa, 15 kDa and 10 kDa).

Identification of high stability mutants from a randomly mutated library Although the C-terminal residues could not be simply truncated, we wondered whether the C-terminus could be optimized for stability by substitution of the amino acids. Since the phage display technique has been widely used in protein engineering, a phage-

ACS Paragon Plus Environment

16

Page 17 of 43 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

Molecular Pharmaceutics

display library of CH2 mutants with random mutation of the C-terminus from position 332 to 340 was constructed with an estimated size of 5×108. To obtain mutants with improved stability, a heat denaturation panning strategy28, utilizing an antibody specific to a conformational epitope of CH2 as bait was used31 (Figure S1). Several mutants with binding activity were obtained by monoclonal phage ELISA. Interestingly, mutations mainly occurred at the last four, unstructured residues (S337, K338, A339 and K340) at the C-terminus (Figure S2). Among these candidates, the soluble expression levels in

E. coli of two clones, B3 (approximately 30 mg/L) and D9 (approximately 10 mg/L) were higher than or equal to that of CH2 (approximately 10 mg/L). Hence, these two mutants were selected for further characterization.

Both of B3 and D9 were monomeric in PBS at pH 7.4 as evaluated by SEC (Figure 2a). The apparent molecular mass of CH2, B3 and D9 were 12.2 kDa, 12.8 kDa and 12.6 kDa calculated by the standard curve of SEC (Figure S3). Both exhibited typical CH2 βstrand structure as determined by CD at 25°C (Figure 2b), suggesting that no obvious change of the secondary structure of B3 and D9 with mutation. Temperature-induced

ACS Paragon Plus Environment

17

Molecular Pharmaceutics 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 43

unfolding of B3 and D9 were also monitored by CD (Figure 2c). The melting temperatures (Tms) of B3 and D9 were 61.7 ± 0.3ºC and 62.0 ± 0.3ºC, respectively, higher than that of CH2 (54.7 ± 0.1ºC). A urea-induced denaturation assay was performed to detect resistance to chemical denaturation33. The concentrations of urea when 50% protein unfolded (CUU) were 4.7 ± 0.2 M and 4.3 ± 0.1 M for B3 and D9 respectively, which were higher than that of CH2 (CUU = 3.9 ± 0.1 M) (Figure 2d). In addition, turbidity assays performed to detect the formation of aggregates in samples during 50ºC incubation showed that the aggregation resistance of B3 and D9 was decreased compared to that of CH2, possibly due to the introduction of the hydrophobic residue isoleucine (Table 1).

ACS Paragon Plus Environment

18

Page 19 of 43 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

Molecular Pharmaceutics

Figure 2. The physicochemical properties of CH2 (black), B3 (red) and D9 (blue). (a) SEC analysis. (b) Secondary structure at 25°C determination by CD. A major negative peak at 216 nm in each protein indicates they mainly consist of β-strands and no obvious conformational change occurs after mutation. (c) Temperature-induced unfolding measured by CD. (d) Urea-induced unfolding.

Table 1. The turbidity of CH2, B3 and D9.

ACS Paragon Plus Environment

19

Molecular Pharmaceutics 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 43

Variant

Mutation

0 min

30 min

60 min

120 min

180 min

CH2

/

0.01 ± 0.01

0.04 ± 0.01

0.04 ± 0.01

0.06 ± 0.01 0.07 ± 0.02

B3

K338I A339L

0.01 ± 0.01

0.35 ± 0.01

0.70 ± 0.02

0.90 ± 0.03 0.92 ± 0.03

0.02 ± 0.01

0.16 ± 0.01

0.36 ± 0.03

0.51 ± 0.02 0.58 ± 0.02

S337K K338I D9

A339S K340S

White shading represents clear; light gray shading represents mild turbid; deeper gray represents more severe turbid; dark gray shading represents very severe turbid. The OD320 value in the table are presented as mean ± SD.

K338I is a critical point mutation in both B3 and D9 Since both B3 and D9 share the K338I mutation, we hypothesized that this point mutation is crucial for the observed increase in stability. As a C1-set Ig-like domain, there is a hydrophobic cavity formed by the conserved strands B, C, E and F, which have a significant impact on structural stability34. W313 is an important conserved residue in all CH2 domains across different human Ig classes as well as among other species (Figure S4). According to the crystal structure of an autonomous CH2 domain (PDB: 3DJ9)16, it coordinates with a cluster of amino acids such as P244, P245, V259,

ACS Paragon Plus Environment

20

Page 21 of 43 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

Molecular Pharmaceutics

V308, Y319 and I336 in the hydrophobic cavity. The strong interactions (e.g., hydrophobic interaction) among these residues and W313 stabilize the molecular structure13 (Figure 3a). The C-terminal residues may provide protection of the hydrophobic cavity via their hydrophilic side chains (Figure 3a). The I-TASSER server35 was used to predict the structure of CH2 with the K338I mutation (CH2-IAK) (Figure 3b). According to the prediction, this point mutation could enhance the hydrophobicity of the cavity and lead to an increased interaction between the cavity and W313, subsequently stabilizing the C-terminus and the whole CH2 molecule. Simultaneously, the predicted computational ΔΔG between CH2-IAK and CH2 was 0.457 kcal/mol based on MUpro36 indicating that the stability of CH2-IAK was improved compared with CH2.

ACS Paragon Plus Environment

21

Molecular Pharmaceutics 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 43

Figure 3. The comparison of structure of (a) CH2 (PDB: 3DJ9) and (b) CH2-IAK (predicted). The amino acids including P244, P245, V259, V308, Y319 (green, in both CH2 and CH2-IAK) that form hydrophobic surface interact with W313 (orange). Replacement of K338 (blue) in CH2 by I338 (green) in CH2-IAK could enhance the interaction with W313 in CH2-IAK.

To assess the impact of K338I, two mutants, CH2-IAK and CH2-AAK (K338A) were constructed and expressed in a bacterial expression system with soluble expression level of about 20 mg/L and 5 mg/L, respectively. They were evaluated by the same

ACS Paragon Plus Environment

22

Page 23 of 43 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

Molecular Pharmaceutics

methods as used for B3 and D9. Both CH2-IAK (12.8 kDa) and CH2-AAK (12.0 kDa) were monomeric and mainly consisted of β-strands (Figure 4a, 4b). The stability of CH2-IAK against heat (Tm = 61.2 ± 0.3°C) and urea (CUU = 4.5 ± 0.1 M) was very similar to that of B3 and D9, whereas CH2-AAK (Tm = 54.9 ± 0.2°C and CUU = 3.9 ± 0.1 M, respectively) behaved much like CH2 (Figure 4c, 4d). Not surprisingly, the aggregation resistance of CH2-IAK and CH2-AAK was still lower than that of CH2 as measured by turbidity (Table 2). These data provided direct evidence in support of our hypothesis. In addition, we noticed that, besides mutation of K338 to isoleucine in B3, A339 was mutated to leucine, another hydrophobic residue. Therefore, we also constructed a mutant CH2-KLK (A339L), which resulted in the loss of soluble expression (data not shown). Therefore, K338L, not A339L, is crucial for the increase of the stability of CH2.

ACS Paragon Plus Environment

23

Molecular Pharmaceutics 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 43

Figure 4. Biophysical characteristics of CH2-AAK (green) and CH2-IAK (wine). (a) SEC analysis. (b) Secondary structure determination by CD at 25°C. (c) Temperatureinduced unfolding. (d) Urea-induced unfolding.

Table 2. The turbidity of CH2, CH2-IAK and CH2-AAK.

Variant

Mutation

0 min

30 min

60 min

120 min

180 min

CH2

/

0.01 ± 0.01

0.03 ± 0.01

0.06 ± 0.01

0.07 ± 0.01

0.13 ± 0.03

ACS Paragon Plus Environment

24

Page 25 of 43 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

Molecular Pharmaceutics

CH2-IAK

K338I

0.03 ± 0.01

0.11 ± 0.01

0.35 ± 0.02

0.46 ± 0.02

0.54 ± 0.04

CH2-AAK

K338A

0.01 ± 0.01

0.28 ± 0.01

0.39 ± 0.03

0.46 ± 0.03

0.51 ± 0.02

White shading represents clear; light gray shading represents mild turbid; deeper gray represents more severe turbid. The OD320 value in the table are presented as mean ± SD.

Further optimization of the C-terminal residues in CH2-IAK It would be interesting to know if the decrease in aggregation resistance caused by K338I mutation could be reversed by introducing hydrophilic residues at the C-terminus of CH2-IAK. Lysine is the most universal, charged residue that has been widely used in the elimination of surface hydrophobicity. For example, it has been shown to protect the edge strands of natural proteins from aggregation37. Serine is a typical hydrophilic, neutral amino acid, which is more prone to forming β-sheet than lysine38. These two residues were selected for incorporation into the C-terminus of CH2 for further optimization. Four mutants CH2-ISS (K338I, A339S and K340S), CH2-ISK (K338I, A339S), CH2-IKK (K338I, A339K) and CH2-IKS (K338I, A339K and K340S) were

ACS Paragon Plus Environment

25

Molecular Pharmaceutics 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 43

constructed based on CH2-IAK by replacement of the last two residues (A339 and K340) with lysine and serine in different combinations.

The respective soluble expression levels of CH2-ISS, CH2-ISK, CH2-IKK and CH2-IKS were approximately 15 mg/L, 20 mg/L, 20 mg/L and >30 mg/L. All mutants were monomeric and mainly composed of β-strands (Figure 5a, 5b). The apparent molecular mass of CH2-ISS, CH2-ISK, CH2-IKK and CH2-IKS were 12.6 kDa, 13.0 kDa, 13.0 kDa and 12.6 kDa calculated by the standard curve of SEC (Figure S3). The Tm values of CH2-ISS, CH2-ISK, CH2-IKK and CH2-IKS were 60.4 ± 0.2ºC, 62.3 ± 0.3°C, 60.3 ± 0.7°C and 63.0 ± 0.2ºC, respectively (Figure 5c). The CUU values of CH2-ISS, CH2ISK, CH2-IKK and CH2-IKS were 4.7 ± 0.1 M, 4.8 ± 0.1 M, 4.4 ± 0.1 M and 4.9 ± 0.1 M, respectively (Figure 5d). Therefore, all the mutants were as stable as CH2-IAK which exhibited improved stability relative to CH2. Importantly, in a time-extended turbidity assay, the aggregation resistance of CH2-IKS is the best among all tested candidates according to the measurement of OD320 values (Figure 6). Since CH2-IKS displays the highest stability and aggregation resistance among the candidate clones, it was

ACS Paragon Plus Environment

26

Page 27 of 43 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

Molecular Pharmaceutics

selected for further size analysis by dynamic light scattering (DLS) following incubation at 37ºC, as previously reported27. Soluble oligomers were observed in the case of CH2 after a seven-day incubation, while very little to no soluble oligomers formed in the case of CH2-IKS (Figure 7).

Figure 5. Biophysical characteristics of CH2-IKS (orange), CH2-ISS (violet), CH2-IKK (pink) and CH2-ISK (dark yellow). (a) SEC analysis. (b) Secondary structure

ACS Paragon Plus Environment

27

Molecular Pharmaceutics 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 43

determination by CD at 25ºC. (c) Temperature-induced unfolding. (d) Urea-induced unfolding.

Figure 6. The turbidity assays. During incubation at 50°C for 600 min, the OD320 values of CH2-IKS and CH2-IKK increase slower than CH2 obviously. Data are presented as mean ± SD.

ACS Paragon Plus Environment

28

Page 29 of 43 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

Molecular Pharmaceutics

Figure 7. DLS assay. A second peak with large size is observed on day 7 during incubation at 37°C in the case of CH2 whereas no additional peak appears in the case of CH2-IKS.

ACS Paragon Plus Environment

29

Molecular Pharmaceutics 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 43

After stabilization, the CH2 molecule could be more resistant to protease digestion31. Therefore, a trypsin-digestion experiment was performed at 37ºC to further validate the stability of CH2-IKS. The result showed that CH2 was significantly digested during a 90min incubation whereas CH2-IKS was only partially digested, suggesting that CH2-IKS is more resistant to protease digestion and therefore it has more compact structure relative to CH239, 40, likely due to the enhanced interactions at the C-terminus (Figure 8).

ACS Paragon Plus Environment

30

Page 31 of 43 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

Molecular Pharmaceutics

Figure 8. Trypsin digestion assay. The digestion of CH2-IKS is much slower than that of CH2. M: protein molecular mass marker (from top to bottom: 35 kDa, 25 kDa, 15 kDa and 10 kDa).

Discussion It has been proven that some small, independent folding proteins (e.g., Ig-like domains) can serve as scaffolds for the development of next-generation therapeutic candidates41. As a typical Ig-fold domain, CH2 is such a promising scaffold. Multiple strategies have been employed to engineer autonomous CH2 domain for improved better stability and aggregation resistance22, 27, 31. However, little work focused on its C-terminus.

Our initial attempt of truncation at the C-terminus led to loss of soluble expression as the protection from C-terminal residues for the hydrophobic cavity around W313 might have been abolished. We utilized phage display, an efficient high throughput method widely used in protein engineering42, for selection of stabilized mutants under heat selection pressure. Although our library covered C-terminal residues from position 332

ACS Paragon Plus Environment

31

Molecular Pharmaceutics 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 43

to 340 in CH2, most of the mutants after panning only contained mutations on the last four residues from 337 to 340, which fell into the C-terminal unstructured region of CH2. Hence, other residues from 332 to 336 could not be simply substituted.

The panning resulted in the identification of a critical point mutation K338I involved in increase of the stability, which was spatially adjacent to W313. According to sequence alignment, there are two conserved tryptophan residues (W277 and W313) in different CH2 domains. A conserved disulfide bond between C261 in the B strand and C321 in the F strand interact with the highly conserved residue W277 to form a hydrophobic core13. Meanwhile, the conserved aromatic residues including Y319 (also called Ycorner43) and W313 occupy the appropriate volume to fill in the gaps between β-strands to stabilize the structure34. In our opinion, replacement of K338 by isoleucine might result in a novel hydrophobic interaction with W313 that further fill the gaps, and leads to a more compact structure. In addition, it has been proven that increased ratio of residues with aliphatic side chains in globular proteins could improve their thermostability44, which could also give an explanation for this result. Interestingly, a

ACS Paragon Plus Environment

32

Page 33 of 43 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

Molecular Pharmaceutics

recent study shows the C-terminal residues “SKTK” in a mouse IgG1 CH2 are important for the protection of the hydrophobic core around tryptophan, which corresponds to W313 in human IgG1 CH245.

However, the increase in hydrophobicity by the introduction of the K338I mutation might not be compensated by the hydrophilicity of other C-terminal residues, which could be the reason for the decrease in aggregation resistance observed in B3, D9, and CH2IAK. To eliminate the adverse effect caused by K338I mutation, charged (lysine) and hydrophilic (serine) residues were used to replace the last two C-terminal residues in CH2-IAK. Interestingly, CH2-IKS exhibits highest stability and aggregation resistance among four combinations. The possible reasons could be: 1. Lysine could provide largest compensation for the increased hydrophobicity; 2. the nonpolar side chain of I338 could interact with the nonpolar neck of K33946 to stabilize C-terminus, and simultaneously the hydrophilic side chain of S340 might further shield this structure and result in decreased aggregation propensity; 3. the introduced serine might be involved in the extension of β-strand (G-strand).

ACS Paragon Plus Environment

33

Molecular Pharmaceutics 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 34 of 43

To some extent, due to its relatively low stability and high aggregation propensity, the CH2 domain is the weak point of an antibody or Fc fragment47-49. In several studies, the stability or aggregation resistance of full-length antibodies or Fc fragments are improved by site directed mutagenesis in the CH2 domain47, 50-52. It will be interesting to see how our strategies work in the context of full-length antibodies or Fc fragments in future work.

Engineering of natural β-sheet proteins is a high-risk task as they are sophisticated molecules and evolution has selected for optimized β-strand interactions37, 53. Therefore, we adopted a combination of phage display technology and rational site directed mutagenesis as our engineering strategy, which resulted in successful identification of a CH2 mutant (CH2-IKS) with improved physicochemical properties. CH2-IKS should be more suitable as scaffold for the development of C-sdAbs therapeutics and applicable for Fc engineering. Modification of C-terminal residues using a combination of directed

ACS Paragon Plus Environment

34

Page 35 of 43 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

Molecular Pharmaceutics

evolution via display and rational mutagenesis used here might also be adopted for the modification of other Ig-like proteins.

Supporting Information The following files are available free of charge (PDF) on the ACS Publications website. Experimental details including the flow chart of panning (Figure S1), sequence results before and after panning (Figure S2), the standard curve of SEC (Figure S3), the comparison of CH2 from human, mouse and rat Igs (Figure S4) and primers are used in constructing library (Table S1).

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Funding Sources

ACS Paragon Plus Environment

35

Molecular Pharmaceutics 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 36 of 43

This work was funded by the National Natural Science Foundation of China (Grant No. 31870926) and the “One-Three-Five” Strategic Programs of Wuhan Institute of Virology, Chinese Academy of Sciences (Grant No. Y605221SA1) ACKNOWLEDGMENT We thank the Core Facility and Technical Support, Wuhan Institute of Virology, Chinese Academy of Sciences; Wuhan Institute of Biotechnology; Wuhan Key Laboratory on Emerging Infectious Diseases and Biosafety; and Wuhan National Bio-Safety Level 4 Lab of the Chinese Academy of Sciences for their support. ABBREVIATIONS mAb (monoclonal antibody), Ig (immunoglobulin G), VH (heavy chain variable domain), C-sdAb (C-based single domain antibody), FcRn (neonatal Fc receptor), C1-set (constant 1 set), Ig-like (immunoglobulin-like), SEC (size exclusion chromatography), CD (circular dichroism), Tm (melting temperature), CUU (the concentration of urea at 50% protein unfolding), DLS (dynamic light scattering), OD (optical density). REFERENCES

ACS Paragon Plus Environment

36

Page 37 of 43 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

Molecular Pharmaceutics

1.

Elgundi, Z.; Reslan, M.; Cruz, E.; Sifniotis, V.; Kayser, V. The state-of-play and

future of antibody therapeutics. Advanced drug delivery reviews 2017, 122, 2-19. 2.

Jain, M.; Chauhan, S. C.; Singh, A. P.; Venkatraman, G.; Colcher, D.; Batra, S.

K. Penetratin improves tumor retention of single-chain antibodies: a novel step toward optimization of radioimmunotherapy of solid tumors. Cancer research 2005, 65, (17), 7840-6. 3.

Labrijn, A. F.; Poignard, P.; Raja, A.; Zwick, M. B.; Delgado, K.; Franti, M.; Binley,

J.; Vivona, V.; Grundner, C.; Huang, C. C.; Venturi, M.; Petropoulos, C. J.; Wrin, T.; Dimitrov, D. S.; Robinson, J.; Kwong, P. D.; Wyatt, R. T.; Sodroski, J.; Burton, D. R. Access of antibody molecules to the conserved coreceptor binding site on glycoprotein gp120 is sterically restricted on primary human immunodeficiency virus type 1. Journal

of virology 2003, 77, (19), 10557-65. 4.

Liu, H.; May, K. Disulfide bond structures of IgG molecules: structural variations,

chemical modifications and possible impacts to stability and biological function. mAbs 2012, 4, (1), 17-23. 5.

Muyldermans, S. Nanobodies: natural single-domain antibodies. Annual review

of biochemistry 2013, 82, 775-97. 6.

Skrlec, K.; Strukelj, B.; Berlec, A. Non-immunoglobulin scaffolds: a focus on their

targets. Trends in biotechnology 2015, 33, (7), 408-18. 7.

Khodabakhsh, F.; Behdani, M.; Rami, A.; Kazemi-Lomedasht, F. Single-Domain

Antibodies or Nanobodies: A Class of Next-Generation Antibodies. International reviews

of immunology 2018, 37, (6), 316-322. 8.

Dimitrov, D. S. Engineered CH2 domains (nanoantibodies). mAbs 2009, 1, (1),

26-8. 9.

Gong, R.; Xiao, G.

Engineered antibody variable and constant domains as

therapeutic candidates. Pharmaceutical patent analyst 2013, 2, (5), 637-46. 10.

Yasmeen, D.; Ellerson, J. R.; Dorrington, K. J.; Painter, R. H. The structure and

function of immunoglobulin domains. IV. The distribution of some effector functions among the Cgamma2 and Cgamma3 homology regions of human immunoglobulin G1.

Journal of immunology 1976, 116, (2), 518-26.

ACS Paragon Plus Environment

37

Molecular Pharmaceutics 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

11.

Page 38 of 43

Gehlsen, K.; Gong, R.; Bramhill, D.; Wiersma, D.; Kirkpatrick, S.; Wang, Y.;

Feng, Y.; Dimitrov, D. S.

Pharmacokinetics of engineered human monomeric and

dimeric CH2 domains. mAbs 2012, 4, (4), 466-74. 12.

Holliger, P.; Hudson, P. J. Engineered antibody fragments and the rise of single

domains. Nature biotechnology 2005, 23, (9), 1126-36. 13.

Bork, P.; Holm, L.; Sander, C. The immunoglobulin fold. Structural classification,

sequence patterns and common core. Journal of molecular biology 1994, 242, (4), 30920. 14.

Krapp, S.; Mimura, Y.; Jefferis, R.; Huber, R.; Sondermann, P.

Structural

analysis of human IgG-Fc glycoforms reveals a correlation between glycosylation and structural integrity. Journal of molecular biology 2003, 325, (5), 979-89. 15.

Edelman, G. M.; Cunningham, B. A.; Gall, W. E.; Gottlieb, P. D.; Rutishauser, U.;

Waxdal, M. J. The covalent structure of an entire gammaG immunoglobulin molecule.

Proceedings of the National Academy of Sciences of the United States of America 1969, 63, (1), 78-85. 16.

Prabakaran, P.; Vu, B. K.; Gan, J.; Feng, Y.; Dimitrov, D. S.; Ji, X. Structure of

an isolated unglycosylated antibody C(H)2 domain. Acta crystallographica. Section D,

Biological crystallography 2008, 64, (Pt 10), 1062-7. 17.

Saphire, E. O.; Parren, P. W.; Pantophlet, R.; Zwick, M. B.; Morris, G. M.; Rudd,

P. M.; Dwek, R. A.; Stanfield, R. L.; Burton, D. R.; Wilson, I. A. Crystal structure of a neutralizing human IGG against HIV-1: a template for vaccine design. Science 2001,

293, (5532), 1155-9. 18.

Goto, Y.; Hamaguchi, K.

The role of the intrachain disulfide bond in the

conformation and stability of the constant fragment of the immunoglobulin light chain.

Journal of biochemistry 1979, 86, (5), 1433-41. 19.

Xiao, X.; Feng, Y.; Vu, B. K.; Ishima, R.; Dimitrov, D. S. A large library based on

a novel (CH2) scaffold: identification of HIV-1 inhibitors. Biochemical and biophysical

research communications 2009, 387, (2), 387-92. 20.

Gong, R.; Wang, Y.; Ying, T.; Dimitrov, D. S. Bispecific engineered antibody

domains (nanoantibodies) that interact noncompetitively with an HIV-1 neutralizing epitope and FcRn. PloS one 2012, 7, (8), e42288.

ACS Paragon Plus Environment

38

Page 39 of 43 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

Molecular Pharmaceutics

21.

Ullman, C.; Mathonet, P.; Oleksy, A.; Diamandakis, A.; Tomei, L.; Demartis, A.;

Nardi, C.; Sambucini, S.; Missineo, A.; Alt, K.; Hagemeyer, C. E.; Harris, M.; Hedt, A.; Weis, R.; Gehlsen, K. R. High Affinity Binders to EphA2 Isolated from Abdurin Scaffold Libraries; Characterization, Binding and Tumor Targeting. PloS one 2015, 10, (8), e0135278. 22.

Li, D.; Gong, R.; Zheng, J.; Chen, X.; Dimitrov, D. S.; Zhao, Q. Engineered

antibody CH2 domains binding to nucleolin: Isolation, characterization and improvement of aggregation. Biochemical and biophysical research communications 2017, 485, (2), 446-453. 23.

Feige, M. J.; Walter, S.; Buchner, J. Folding mechanism of the CH2 antibody

domain. Journal of molecular biology 2004, 344, (1), 107-18. 24.

Demarest, S. J.; Rogers, J.; Hansen, G. Optimization of the antibody C(H)3

domain by residue frequency analysis of IgG sequences. Journal of molecular biology 2004, 335, (1), 41-8. 25.

Gong, R.; Vu, B. K.; Feng, Y.; Prieto, D. A.; Dyba, M. A.; Walsh, J. D.;

Prabakaran, P.; Veenstra, T. D.; Tarasov, S. G.; Ishima, R.; Dimitrov, D. S. Engineered human antibody constant domains with increased stability. The Journal of biological

chemistry 2009, 284, (21), 14203-10. 26.

Jacobs, S. A.; Diem, M. D.; Luo, J.; Teplyakov, A.; Obmolova, G.; Malia, T.;

Gilliland, G. L.; O'Neil, K. T. Design of novel FN3 domains with high stability by a consensus sequence approach. Protein engineering, design & selection : PEDS 2012,

25, (3), 107-17. 27.

Gong, R.; Wang, Y.; Ying, T.; Feng, Y.; Streaker, E.; Prabakaran, P.; Dimitrov, D.

S. N-terminal truncation of an isolated human IgG1 CH2 domain significantly increases its stability and aggregation resistance. Molecular pharmaceutics 2013, 10, (7), 264252. 28.

Jespers, L.; Schon, O.; Famm, K.; Winter, G.

Aggregation-resistant domain

antibodies selected on phage by heat denaturation. Nature biotechnology 2004, 22, (9), 1161-5. 29.

Chen, W.; Zhu, Z.; Xiao, X.; Dimitrov, D. S. Construction of a human antibody

domain (VH) library. Methods in molecular biology 2009, 525, 81-99, xiii.

ACS Paragon Plus Environment

39

Molecular Pharmaceutics 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

30.

Page 40 of 43

Arbabi-Ghahroudi, M.; MacKenzie, R.; Tanha, J. Selection of non-aggregating

VH binders from synthetic VH phage-display libraries. Methods in molecular biology 2009, 525, 187-216, xiii. 31.

Gong, R.; Wang, Y.; Feng, Y.; Zhao, Q.; Dimitrov, D. S. Shortened engineered

human antibody CH2 domains: increased stability and binding to the human neonatal Fc receptor. The Journal of biological chemistry 2011, 286, (31), 27288-93. 32.

Gong, R.; Chen, W.; Dimitrov, D. S.

Expression, purification, and

characterization of engineered antibody CH2 and VH domains. Methods in molecular

biology 2012, 899, 85-102. 33.

Pace, C. N. Determination and analysis of urea and guanidine hydrochloride

denaturation curves. Methods in enzymology 1986, 131, 266-80. 34.

Wang, J. H. The sequence signature of an Ig-fold. Protein & cell 2013, 4, (8),

569-72. 35.

Zhang, Y.

I-TASSER server for protein 3D structure prediction. BMC

bioinformatics 2008, 9, 40. 36.

Cheng, J.; Randall, A.; Baldi, P. Prediction of protein stability changes for single-

site mutations using support vector machines. Proteins 2006, 62, (4), 1125-32. 37.

Richardson, J. S.; Richardson, D. C. Natural beta-sheet proteins use negative

design to avoid edge-to-edge aggregation. Proceedings of the National Academy of

Sciences of the United States of America 2002, 99, (5), 2754-9. 38.

Minor, D. L., Jr.; Kim, P. S. Measurement of the beta-sheet-forming propensities

of amino acids. Nature 1994, 367, (6464), 660-3. 39.

Riahi-Madvar, A.; Hosseinkhani, S. Design and characterization of novel trypsin-

resistant firefly luciferases by site-directed mutagenesis. Protein engineering, design &

selection : PEDS 2009, 22, (11), 655-63. 40.

Vandermarliere, E.; Mueller, M.; Martens, L. Getting intimate with trypsin, the

leading protease in proteomics. Mass spectrometry reviews 2013, 32, (6), 453-65. 41.

Gebauer, M.; Skerra, A.

Engineered protein scaffolds as next-generation

antibody therapeutics. Current opinion in chemical biology 2009, 13, (3), 245-55. 42.

Smith, G. P. Filamentous fusion phage: novel expression vectors that display

cloned antigens on the virion surface. Science 1985, 228, (4705), 1315-7.

ACS Paragon Plus Environment

40

Page 41 of 43 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

Molecular Pharmaceutics

43.

Hemmingsen, J. M.; Gernert, K. M.; Richardson, J. S.; Richardson, D. C. The

tyrosine corner: a feature of most Greek key beta-barrel proteins. Protein science : a

publication of the Protein Society 1994, 3, (11), 1927-37. 44.

Ikai, A.

Thermostability and aliphatic index of globular proteins. Journal of

biochemistry 1980, 88, (6), 1895-8. 45.

Weber, B.; Brandl, M. J.; Pulido Cendales, M. D.; Berner, C.; Pradhan, T.; Feind,

G. M.; Zacharias, M.; Reif, B.; Buchner, J. A single residue switch reveals principles of antibody domain integrity. The Journal of biological chemistry 2018, 293, (44), 1710717118. 46.

Dyson, H. J.; Wright, P. E.; Scheraga, H. A. The role of hydrophobic interactions

in initiation and propagation of protein folding. Proceedings of the National Academy of

Sciences of the United States of America 2006, 103, (35), 13057-61. 47.

Chennamsetty, N.; Voynov, V.; Kayser, V.; Helk, B.; Trout, B. L.

Design of

therapeutic proteins with enhanced stability. Proceedings of the National Academy of

Sciences of the United States of America 2009, 106, (29), 11937-42. 48.

Chennamsetty, N.; Voynov, V.; Kayser, V.; Helk, B.; Trout, B. L. Prediction of

aggregation prone regions of therapeutic proteins. The journal of physical chemistry. B 2010, 114, (19), 6614-24. 49.

Latypov, R. F.; Hogan, S.; Lau, H.; Gadgil, H.; Liu, D. Elucidation of acid-induced

unfolding and aggregation of human immunoglobulin IgG1 and IgG2 Fc. The Journal of

biological chemistry 2012, 287, (2), 1381-96. 50.

Chen, W.; Kong, L.; Connelly, S.; Dendle, J. M.; Liu, Y.; Wilson, I. A.; Powers, E.

T.; Kelly, J. W.

Stabilizing the CH2 Domain of an Antibody by Engineering in an

Enhanced Aromatic Sequon. ACS chemical biology 2016, 11, (7), 1852-61. 51.

Jacobsen, F. W.; Stevenson, R.; Li, C.; Salimi-Moosavi, H.; Liu, L.; Wen, J.; Luo,

Q.; Daris, K.; Buck, L.; Miller, S.; Ho, S. Y.; Wang, W.; Chen, Q.; Walker, K.; Wypych, J.; Narhi, L.; Gunasekaran, K. Engineering an IgG Scaffold Lacking Effector Function with Optimized Developability. The Journal of biological chemistry 2017, 292, (5), 18651875. 52.

Zeng, F.; Yang, C.; Gao, X.; Li, X.; Zhang, Z.; Gong, R.

Comprehensive

elucidation of the structural and functional roles of engineered disulfide bonds in

ACS Paragon Plus Environment

41

Molecular Pharmaceutics 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 42 of 43

antibody Fc fragment. The Journal of biological chemistry 2018, 293, (49), 1912719135. 53.

Wang, W.; Hecht, M. H. Rationally designed mutations convert de novo amyloid-

like fibrils into monomeric beta-sheet proteins. Proceedings of the National Academy of

Sciences of the United States of America 2002, 99, (5), 2760-5.

ACS Paragon Plus Environment

42

Page 43 of 43 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

Molecular Pharmaceutics

Optimization of the C-terminus of an autonomous human IgG1 CH2 domain for stability and aggregation resistance Xinyu Gao1,2, Alex Conard3, Chunpeng Yang1,2, Yancheng Zhan1,2, Fang Zeng1,2, Jian Shi1, Wei Li3, Dimiter S. Dimitrov3, Rui Gong1*

For Table of Contents Use Only

ACS Paragon Plus Environment

43