Generation of monoclonal antibodies via phage display and in vitro affinity maturation using activation induced deoxycytidine deaminase and DNA polymerase eta

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

GENERATION OF MONOCLONAL ANTIBODIES VIA PHAGE DISPLAY AND
IN VITRO AFFINITY MATURATION USING
ACTIVATION INDUCED DEOXYCYTIDINE DEAMINASE AND DNA POLYMERASE
ETA

by

Soo Lim Jeong

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirement for the Degree
DOCTOR OF PHILOSOPHY
(MOLECULAR BIOLOGY)

May 2022

Copyright 2022 Soo Lim Jeong

ii
Acknowledgements
I thank my mom, Gi Jung, dad, Cheol Hyeun, and my sister, Soo Min, for their
unconditional love and support throughout my life. They are my everything.
I am extremely grateful to have Dr. Myron Goodman as my advisor, who has provided
me with the tools and guidance for my research endeavor. I was constantly inspired by his
endless aspiration and enthusiasm for science, and his sense of humor made our lab a fun place
to do research.
I am truly indebted to Dr. Phuong Pham, who has taught me so much about this project
and science in general. His guidance and mentorship throughout my PhD journey has been
invaluable.
I thank the members of Goodman lab, Adi, Hongyu, Gosia, Debika, and Aeriel, for their
support and feedback. I appreciate Hongyu for his contribution to the Ab project.
I would like to thank Dr. David McKemy and members of McKemy lab for their
collaborative efforts and support throughout the years for this project. I would like to thank Dr.
Raymond Stevens and members of Stevens lab for the provision of GLP-1R, and Dr. Michael
Lieber for the provision of tonsils.
I would like to express my gratitude to the esteemed members of my dissertation
committee, Dr. David McKemy, Dr. Michael Lieber, and Dr. Steven Finkel, for their time,
feedback, and guidance for my dissertation efforts.
I express my thanks to the USC Dornsife Chemical Biology Training Program, for
financially supporting this project.

iii
I am grateful to my dear friends, Janet, Garim, Agatha, and Anne, who have been
tremendously supportive and loving when times were rough. They made me realize what good
people I have in my life, and love myself for who I am.
Lastly, I thank God, for this precious opportunity, this life, and this world full of
wonders.

iv
TABLE OF CONTENTS
Acknowledgements……………………………………………………………………………………….ii
List of Tables……………………………………………………………………………………………..vi
List of Figures………………………………………………………..…………………………..……….vii

Abstract………………………………………………………………...……………………………….…ix

Chapter 1. Introduction ................................................................................................................. vii
1.1 Role of Somatic Hypermutation in generating Ab diversity during an immune response ... 1
1.2 Role of AID in SHM: AID’s catalytic specificities (WRC) on IgV in vivo and in vitro ..... 4
1.3 Role of Polh in SHM: Polh (WA) catalytic specificities on IgV in vivo and in vitro .......... 5
1.4 Current methods in generation of monoclonal antibodies and phage display ...................... 7
Conclusion ................................................................................................................................ 12

Chapter 2. Analysis of in vitro AID and Polh mutation spectrum ............................................... 13
2.1 Introduction ......................................................................................................................... 13
2.2 Materials and Methods ........................................................................................................ 14
2.3 Results ................................................................................................................................. 20
1. Mutation spectrum of combined AID and Polh actions on IGVH323-01 gapDNA
construct shows that AID and Polh retain their signature catalytic specificities in vitro ..... 20
2. Application of AID and Polh in affinity maturation in vitro successfully diversifies a
synthetic scFv gene repertoire .............................................................................................. 26
2.4 Discussion ........................................................................................................................... 38

Chapter 3. Construction of AID and Polh - diversified human-scFv and llama-VHH phage
display libraries ............................................................................................................................. 40
3.1 Introduction ......................................................................................................................... 40
3.2 Materials & Methods .......................................................................................................... 42
3.3 Results ................................................................................................................................. 48
1. Construction of a mutagenized human scFv phage display library .................................. 48
1

v
2. Construction of mutagenized llama VHH-phage display library ..................................... 52
3.4 Discussion ........................................................................................................................... 53

Chapter 4. Application of AID and Polh affinity maturation in vitro to isolate Abs targeting
specific antigens: Target 1 – GLP-1R ........................................................................................... 55
4.1 Introduction ......................................................................................................................... 55
4.2 Materials & Methods .......................................................................................................... 57
4.3 Results ................................................................................................................................. 64
1. Eight high-affinity scFv binders to GLP-1R were isolated by biopanning of human scFv-
phage library and characterized ............................................................................................ 64
2. Further in vitro affinity maturation of eight scFvs using AID and Polh generated Ab
variants with enhanced affinity to GLP-1R .......................................................................... 77
4.4 Discussion ........................................................................................................................... 87

Chapter 5. Application of AID and Polh affinity maturation in vitro to isolate Abs targeting
specific antigen: Target 2 - Artemin ............................................................................................. 90
5.1 Introduction (Artemin) ........................................................................................................ 90
5.2 Materials & Methods .......................................................................................................... 91
5.3 Results ................................................................................................................................. 94
1. Four anti-Artemin VHH nanobodies were isolated by biopanning of llama VHH-phage
display libraries and characterized ........................................................................................ 94
5.4 Discussion ......................................................................................................................... 105

References ................................................................................................................................... 107

Supplementary Figures ............................................................................................................... 113
List 1. .................................................................................................................................. 113
List 2. .................................................................................................................................. 119

vi
List of Tables

Table 2-1. Comparison of deamination frequency in WRC/WGCW motifs within each CDR and
FW region and total number of deaminations occurring in each region of IGVH3-23*01 in
M13mp2……………………………………………………………………………………...……2
Table 2-2. Number of colonies on agar plate supplemented with ampicillin upon transformation of
10 pg of linearized pADL20c vector, linearized pADL20c-scFv plasmid, and double-gapped
pADL20c-scFv construct into E.coli TG1………………………………………………………..28
Table 2-3. A. B. Number of mutations in each sub-region of Ig variable heavy and light chains of
20 representative scFv clones……………………………………………………………………31
Table 2-4. A.B. % mutation frequency representing the number of mutations per base pair in each
sub-region of Ig variable heavy and light chain of 20 representative scFvs……………….……..34
Table 3-1. Sequencing analysis of 16 representative clones randomly selected from mutagenized
human f3TR1-scFv phage display library………………………………………………………..50
Table 4-1. Antibody sequence analysis of 8 anti-GLP-1R scFv clones isolated from the AID and
Polη diversified human scFv-phage library………………………………………………….…...70
Table 4-2. % mutation frequencies representing the number of mutations per base pair in the
designated sub-regions of Ig variable heavy and light chains of eight GLP-1R specific scFvs….71
Table 4-3. Kinetic constants and equilibrium dissociation constants (KD) for binding interactions
between 6 GLP1R-specific scFv peptides and GLP1R-ECD determined by Surface Plasmon
Resonance……………………………………………………………………………….………..76
Table 4-4. Kinetic and equilibrium dissociation constants of scFv 25 and its affinity-matured
scFv 33’ variant to GLP-1R-ECD from SPR…………………………………………......…..….86
Table 5-1. Kinetic and equilibrium dissociation constant values of purified llama-VHH
nanobodies targeting artemin, determined by SPR………………………………………………99

vii
List of Figures

Figure 1-1. Structure of a full-length human IgG and antibody display formats; Fab, scFv and
VHH……………………………………………………..…………………………………...…...2
Figure 1-2. Schematic diagram of a typical phage-antibody library biopanning process……..….11
Fig 2-1. Analysis of AID- and polh -catalyzed mutations on human IgVH3-23 gap substrate…23
Fig 2-2. Compilation of 862 AID and Polh-catalyzed mutations across the IgVH-323 region
obtained from a subset of 150 individual M13 mutant DNA clones……………………………..24
Figure 2-3. Comparison of % deamination frequencies occurring within the WRC/WGCW AID
hotspot motifs in each CDR or FW sub-region of IGVH3-23*01 in M13mp2 gapped
construct……………………………………………………………………………….………….25
Figure 2-4. Schematic diagram illustrating the synthesis of scFv-double gap DNA constructs
and in vitro mutagenesis of the resultant constructs by AID and Polh…………………………..28
Figure 2-5. Pairwise sequence alignment of representative scFv clone (10-15), and heavy (Hu-
VH 932) and light chain (Hu-VL914)……………………………………………………….…...37
Figure 3-1. Pairwise alignment of DNA sequences of a scFv clone (F3) from mutagenized human
scFv-phage library and human germline IgV heavy and light chain genes……………………….51
Figure 3-2. Sequence alignment for 27 scFv clones randomly selected from mutagenized human
scFv-phage library………………………………………………………………………………..52
Figure. 4-1. ELISA- binding of 31 individual human scFv-phages to purified human GLP-
1R…………………………………………………………………………………..…………….66
Fig 4-2. Amino acid sequence identities of GLP-1R-specific 8 scFvs…………………………..67
Figure 4-3. GLP-1R binding by eight scFvs determined by ELISA assay………………………73
Figure 4-4. SPR sensorgrams showing relative binding of purified scFv-His6 peptides to N-
terminal GLP-1R extracellular domain over time………………………………………………….76
Figure 4-5. Schematic diagram describing steps involved in “gapDNA” method and in ssDNA
treatment method………………………………………………………………………………...79
Figure 4-6. Biopanning enriches target-specific scFv-phage population…………………..……..…..81
Fig 4-7. DNA sequence alignment of GLP-1R-specific scFv 25 and scFv 33’.…………………82
Fig 4-8. Alignment of GLP1R-specific scFv amino acid sequences………………………….….82

viii
Fig 4-9. Phage-ELISA based binding of original (GHM25) and affinity-matured (GHM33’)
human-scFv to purified GLP1-R whole membrane protein………………….………………….83
Figure 4-10. Concentration-dependent binding of original and affinity-matured scFv-peptides to
isolated GLP1-R measured by ELISA……………………………………….…………………..85
Figure 4-11. SPR sensorgrams showing relative binding of purified scFv-His6 peptides to N-
terminal GLP-1R extracellular domain…………………………………….…………………….86
Figure 5-1. Enrichment of artemin-specific f3TR1-VHH-phage population for successive
biopanning………………………………………………………………….…………………….95
Figure 5-2. Bar graph showing the absorbance at 450 nm of twenty individual monoclonal VHH
- phage binding to purified mouse artemin in phage-ELISA……………….……………………96
Figure 5-3. DNA sequence alignment of LU5-VHH and LM52-VHH and LU68-VHH and
LM41-VHH………………………………………………………………………………………98
Figure 5-4. Amino acid sequence alignment of LU5-VHH and LM52-VHH and LU68-VHH and
LM41-VHH…………………………………………………………………………………....…98
Figure. 5-5. Sensorgrams depicting the binding of the mouse artemin and purified llama-VHH by
Surface Plasmon Resonance (SPR) for the first trial..………………………………………….100
Figure 5-6: Inflammatory cold allodynia is inhibited by both LM52-VHH and anti-artemin
Mab1085) ………………………………………………………………………………………103
Figure 5-7: Bar graph showing the inhibition of inflammatory cold allodynia by LM52-VHH and
anti-artemin Mab1085 in 5 male mice…………………………………………………………..104

ix
Abstract
Current advances in medicine feature intensive attempts at antibody (Ab) engineering to target
a diverse array of antigens (Ag). For in vitro screening and isolation of antibodies that specifically
target a chosen Ag, a library of highly diverse antibodies must be created. The primary goal of this
project was to generate human and llama monoclonal Abs in vitro by utilizing the unique
mechanism of mutagenic diversification of antibody-encoded immunoglobulin (Ig) genes by
activation-induced cytidine deaminase (AID) and error-prone DNA polymerase eta (pol h).
During an immune response, Abs with increased affinity toward Ags are generated in activated
B-cells through mutagenic diversification processes of Ig genes. One of these processes is somatic
hypermutation (SHM) of variable (V) regions of Ig genes. SHM is generated at G-C sites by
enzyme activation induced deoxycytidine deaminase (AID) preferentially at WRC (W = A/T, R =
A/G) hotspot motifs and at A-T sites by polh at WA motifs. Affinity maturation occurs by the
repetitive action of AID and error-prone polymerase h (polh) during a series of cell divisions
accompanied by clonal selection for high affinity Abs. This study describes how we developed
highly diversified phage display Ab libraries by performing affinity maturation in vitro using
purified human AID and polh. The scientific rationale for the application of this novel method in
Ab discovery was based on the earlier studies and our recent findings showing that in vitro
mutation spectrum of purified AID and Polh on IgV gene resembles in vivo SHM mutation spectra
on the same V region. Since AID and Pol h retain their catalytic specificities in vitro, we
hypothesized that the diversified Ab library would mimic products of SHM in B-cells with the
characteristic WRC (AID) and WA (pol h) hotspot mutations.
This project applied the AID- and pol h -diversified Ab libraries to isolate Abs
specifically targeting GLP-1R, a G-protein coupled receptor protein that plays an essential role in

x
insulin production during glucose metabolism, and artemin neurotrophic factor, which is
implicated in human cold hypersensitivity. Here, we report the isolation and characterization of
target-specific monoclonal Ab clones derived from in vitro affinity maturation of Ab- repertoires
prior to exposure to Ag, and of individual Ab clones after enrichment in vitro via Ag.

1
Chapter 1. Introduction
1.1 Role of Somatic Hypermutation in generating Ab diversity during an immune response
Vertebrate immune systems have evolved complex molecular mechanisms to generate a
diverse array of antibodies (Ab) that can recognize and neutralize a broad range of invading
foreign pathogens. In mice and humans, a large repertoire of pre-immune Abs of low affinity
(IgM isotype) are generated in pro- and pre-B cells of bone marrow prior to antigen recognition
through VDJ recombination. VDJ recombination rearranges the germ line immunoglobulin (Ig)
variable (V), diversity (D), and joining (J) segments to form the heavy (H) and light (L) chain of
Ig genes. This random rearrangement of Ig gene segments generates a diversity which ensures
there will be an antigen binding site to fit almost any antigen, albeit with low binding capability.

Upon exposure to an antigen, mature B cells proliferate and migrate to the germinal
centers in peripheral lymphoid organs. The activated B-cells then undergo affinity maturation, an
iterative process by which secondary Ab gene diversification and clonal selection generate high
affinity Abs (typically of isotypes IgG, IgA, or IgE), toward a specific antigen. A functional Ab
is composed of four Ig polypeptides, two heavy chains (H) and two light chains (L) of either
lambda (Igl) or kappa (Igk) type. The antigen-binding region of Abs, the variable (V) domain, is
composed of amino acid sequences of variable heavy chain (VH) and variable light chain (VL),
which, combined via disulfide bonds, create the binding site for an antigen and hence determine
the antigen binding specificity of an Ab (Fig 1.1). More specifically, Each VH and VL contains
three complementarity determining regions (CDR) that form variable loops responsible for
binding to the antigen. Affinity maturation involves highly localized and targeted mutagenesis of

2
VL and VH regions of immunoglobulin (Ig) genes to generate antigen-binding sites with varying
affinities, through a process called somatic hypermutation (SHM).

Figure 1-1. (A) Structure of a full-length human IgG and (B) antibody display formats;
Fab, scFv and VHH.

SHM is initiated by enzyme activation-induced deoxycytidine deaminase (AID), which
introduces cytosine to uracil (C®U) deaminations preferentially at WRC (W= A/T, R = A/G , C
= C) (Rogozin and Kolchanov, 1992; Rogozin et al., 2001) hotspot motifs on a single-stranded V
region of Ig genes (Bransteitter et al. 2003., Pham et al. 2003., Chaudhuri et al., 2003., Dickerson
et al., 2003) during active transcription of B-cells (Di Noia & Neuberger, 2007). The C®U
deaminations on IgV region occur at a rate of ~10
-3
– 10
-4
/base pair/cell division, which is a
million times higher than the average rate of somatic mutations on non-Ig genes (McKean et
al.,1984; Rajewsky, 1996). Moreover, it has been found that the WRC motifs are highly
concentrated in the three complementarity determining regions (CDRs) within the Ig V
A B

3
sequences, which are the regions that make up the antigen binding pocket of Abs (Wei., et al.,
2015). Hence AID’s preferential deaminations of WRC hotspots on the CDR regions of Ig V
gene (Rogozin et al., 1992; Shapiro et al., 2002) ensures the mutagenic diversity of Ab paratopes
in B-cells during immune response.
During the subsequent phase of SHM, an AID-induced G:U mismatch either gets
replicated over, generating a C®T transition mutation on the newly synthesized strand (or G®A
if the opposite strand is targeted), or triggers a series of error-prone repair machineries such as
base excision repair (BER) or mismatch repair (MMR) to add additional mutations. In BER,
uracil DNA glycosylase removes uracil from DNA, creating an abasic site that can be filled with
A, G, C, or T and consequently resulting in either a transition or transversion mutation. In MMR,
the U:G mismatch is recognized and excised by a complex of MutS homologs, MSH2/MSH6.
The MutS homologs recruit an error-prone DNA polymerase eta (Pol h), which introduces
additional mutations at neighboring A:T bases preferentially at WA motifs (W = A/T)(Di Noia &
Neuberger, 2002). It was discovered that more than 60% of the SHM induced mutations in mice
and humans were in A:T bases, and approximately half of those are transversions rather than
transitions (Peled., 2008). Thus the interplay between AID, a SHM initiating factor that creates a
G:U mismatch on IgV, and the error-prone repair machineries including pol h assembled upon
the AID deamination event, affects a robust SHM to create Ab mutagenic diversity in activated
B-cells.

4
1.2 Role of AID in SHM: AID’s catalytic specificities (WRC) on IgV in vivo and in vitro
The requirement of AID in SHM was confirmed by earlier evidence that genetically
engineered mice lacking AID and AID-deficient patients with type II hyper-IgM
immunodeficiency syndrome (HIGM-2) were unable to carry out SHM (Muramatsu et al, 1999).
To further assess the AID action regarding its intrinsic mutational preference at WRC motifs
during SHM, intensive studies have been carried out on the deamination profiles of AID on V
gene in B cells and in multiple in vitro biochemical systems. The comparison of in vivo and in
vitro mutation profiles have been crucial in elucidating the signature AID specificities and the
sequence context for AID targeting.
In 2009, our lab in collaboration with the Matthew Scharff lab (Albert Einstein College of
Medicine, NY) have compared the AID activity in vitro and in vivo by comparing the
deamination profile of purified human recombinant AID on a murine Ig V gene (VHJ558-Jh4
intron) with AID-catalyzed mutations on the same Ig region in mice that lack both BER and
MMR (MacCarthy et al., 2009). It was discovered that the AID activity yields a site-by-site
distribution of C®T mutations with hot spots (WRC, W = A/T, R = A/G) and cold spots (SYC,
S = G/C, Y = C/T) similar to those observed for the in vivo SHM profile (Rada & Neuberger.,
2004), with deaminations accumulating on the hot spots and away from the cold spots. In 2015,
the spatial distribution of AID deaminations in the human IgVH3-23*01 region in unselected
human memory B-cells (Ohm-Laursen et al., 2007), Ramos B-cell lines, and in vitro
reconstituted system (Wei et al., 2015) was examined. It was discovered that the distributions of
mutations in both G:C and A:T bases throughout the V region in vivo and in a Ramos cell line
were quite similar to one another, with the overlapping AGCT (WRCW) hotspot motifs focusing
the AID-induced mutations primarily in CDR regions, especially in CDR 2. Further analysis of

5
the mutations added in an in vitro system on a single-stranded IgV substrate again showed
AGCT in CDR1 and CDR2 accumulates deaminations, suggesting that AID-deamination is
targeted towards the parts of the V region that affect Ag binding, specifically the CDRs. These
findings were further supported when in 2019, an AID deamination profile in conjunction with
DSIF transcription factor and RNA Polymerase II on IgHV3-23*01 during Pol II transcription
elongation was investigated (Pham et al., 2019). The transcriptional elongation of V gene by
RNAP II was reconstituted in vitro using purified human AID, Pol II and DSIF, and it was found
that the frequency of AID-Pol II mutations was elevated in WRC and WGCW overlapping hot
motifs in CDR 1 and CDR2 domains on both transcribed (TS) and non-transcribed strands
(NTS). Site-by-site comparisons for the in vitro reconstituted AID-Pol II system and human
memory B-cell mutation spectra in an IGHV3-23 target (Ohm-Laursen et al., 2007) exhibited
strong bias for deaminations by the antigen-binding CDR regions compared to the FW regions,
which can be explained by the statistically significant enrichment of AID hotspots in CDRs
versus FWs. Approximately 50 % of G:C bases in CDR1 and CDR2 were found in AID hot
spots, while less than 22 % of G:C bases in the FWs are in AID hotspots (Wei et al., 2015). In
conclusion, the consistency in AID-induced deamination patterns in B-cells and in in vitro
systems reinforces the inherent nature of AID’s signature catalytic specificities, consequently
demonstrating that a significant portion of the B-cell SHM can be captured in a biochemically
reconstituted in vitro system.
1.3 Role of Polh in SHM: Polh (WA) catalytic specificities on IgV in vivo and in vitro
Polymerase eta (Polh) is a low fidelity enzyme lacking the proofreading 3’-5’ exonuclease
activity and hence highly mutagenic when copying undamaged DNA (Yang W., Woodgate R.
2007). It is only at sites of UV-induced T-T cyclobutene pyrimidine dimers that Polh carries out

6
translesion synthesis in an error-free manner (Masutani., 2000; Jonhson., 2000). During the
second phase of SHM, an AID-induced U:G mismatch is recognized by components of the
mismatch repair machinery, the homologs MSH2 and MSH6. Once a repair patch is generated,
the low fidelity polh is recruited by the mismatch repair homologs and starts to incorporate point
mutations at nearby A and T bases (Rogozin et al.,2001). It has been discovered that human
Polh, depending on the mismatch, incorporates one base substitution error for every 100
nucleotides synthesized while favoring the WA motifs (Rogozin et al., 2001; Pavlov et al.,
2002), thus highlighting the highly error-prone nature of its polymerase catalysis.
Numerous studies have confirmed that Polh predominantly adds substitutions at A:T bases
on V genes preferentially at the WA hotspot motifs during somatic hypermutation. The analysis
of mutation patterns on Ig genes of Polh-deficient mice exhibited fewer mutations at A:T (Zhao
et al., 2013; Longerich et al., 2006). Likewise, studies done on patients suffering from xeroderma
pigmentosum variant disease (XPV), a condition that produces defective Polh, showed a similar
suppression in mutations at A:T pairs and a concomitant increase in mutations at C:G pairs
(Pavlov et al., 2002). Furthermore, it was reported that a double mutant mouse deficient in
MSH2 and Polh exhibited a complete abolishment of A:T mutagenesis, marking Polh as the sole
contributor of A:T mutations (Delbos et al, 2007), and the interaction of MMR machinery and
Polh a necessary component of a robust SHM. It was also found that the Polh’s substitution rates
correlated inversely with the distance to the nearest 3’ WRC hotspot on NTS and TS, suggesting
that phase II SHM occurs in close proximity to the AID deamination site (Ohm-Laursen et al.,
2007)

7
To further examine the Polh’s role in SHM, the A:T mutation profiles on IgV regions in vitro
were closely investigated by many groups. Pavlov and colleagues showed that there was a
significant correlation between SHM at A-T bases on a mouse Ig light chain transgene, VkOx, and
substitutions generated by mouse Polh on the same gene in vitro (Pavlov et al. 2002), with
mutation skewed on WA hotspots both in vivo and in vitro. Although the AID deamination-
induced recruitment of the MMR machinery for Polh catalysis in B cells is absent in an in vitro
system, it is apparent that Polh retains its catalytic specificity for WA hotspots. A recent analysis
on human IGHV sequence profiles revealed that there were more WA Polh hotspots in CDR1 and
CDR2 than in FW1 and FW2; approximately 45% of the A:T residues in CDR and CDR2 were in
WA hot spots, whereas the percentage of A:T was much lower in FW1 and FW2, explaining the
much lower mutation frequencies in FW regions. (Tang et al., 2020). In conclusion, the in vivo
and in vitro A:T mutation profiles on Polh’s hotspot motifs WA bear significant resemblance.
Moreover, there is an enrichment of hotspots in the strategic regions of CDR, again strengthening
the point that the effects of B-cell SHM for the generation of diverse Ab paratopes can be captured
in vitro.

1.4 Current methods in generation of monoclonal antibodies and phage display
The two most prominently used methods in human antibody development are the traditional
hybridoma technology and the more recent in vitro display techniques. The former involves the
fusion of B-cells of mice immunized with a target antigen and immortal B-cell cancer cells
called myeloma, to produce a hybrid cell line that can produce Abs with the longevity and
reproductivity of myeloma (Hnasko et al., 2015). Although successful in generating many human
mAbs, this method carries disadvantages such as the inherent immunogenicity elicited by the

8
murine Abs, and the use of live animals and several rounds of immunization processes that are
expensive and time-consuming. The latter, most notable of which is the phage display method,
utilizes the recombinant DNA technique in which the Ig gene encoding the Ag-interacting
domain of an Ab (paratope) is fused to a bacteriophage gene encoding a phage coat protein, such
that after packaging, the Ag-interacting domain is “displayed” on the surface of the phage
particles (Smith. G.P., 1985). In an Ab-phage display library, each phage particle in a library of
billions of Ab-expressing phages encapsulates a distinct recombinant Ab-phage gene, thereby
expressing a distinct Ab paratope different from one another. The ability to assemble a large
collection of unique antibody clones enables the phage display method to rapidly screen and
identify specific Abs toward a target Ag after several rounds of affinity-based selection. There
are other in vitro Ab display methods that utilize the physical linkage of the displayed Ab protein
and the DNA encoding, examples are yeast display, ribosome display, mRNA display, bacterial
surface display, and mammalian cell display (H. Shim, 2016).

A. Phage-antibody display
Bacteriophage, or phage, is a type of virus that can infect bacteria. The most widely used
bacteriophage for antibody phage display is M13 filamentous phage. M13 phage consists of a
single-stranded phage genome encapsulated by ~2700 copies of p8 major coat proteins and 5
copies each of minor coat proteins p3, p6, and p9 at the ends of phage surface (Smith,G.P.,
1985). Upon non-lytic infection of E. coli expressing F pili, the phage single-stranded DNA is
replicated, transcribed, and translated so that the phage proteins are synthesized and packaged
using the host’s machinery and finally secreted into the bacterial periplasm. The p3 coat proteins
are most frequently used as a fusion peptide for Ab display.

9
There are two ways to incorporate an Ab gene into a phage system. In the first method, Ab
genes are introduced next to a gene encoding a phage coat protein (pIII) in a phage vector, which
contains all the phage genes required for infection, replication, and assembly of phage virions.
Due to the presence of 3-5 copies of pIII coat proteins on each phage particle, the Ab-G3P fusion
proteins are displayed as multiple copies on the phage surface – resulting in a multivalent
display. One downside of a multivalent phage display library is that the binding of multiple
copies of Ab to an Ag can lead to an avidity effect, which may result in the enrichment of weak
binding clones with propagative advantages. Another method is to use a phagemid vector, which
only carries genes for an antibiotic marker, the phage replication origin, and the recombinant Ab-
phage gene for Ab display. This simple assembly of genes significantly reduces the vector size,
increasing the transformation efficiency and consequently the size of the phage-Ab library. Since
a phagemid lacks genes encoding the necessary components for phage assembly, a helper phage
carrying the complete set of M13 genome is used to infect E. coli harboring the recombinant
phagemid (termed “superinfection”). The wild type G3P from the helper phage competes with
the Ab-G3P fusion protein for incorporation into phage, ultimately resulting in only 10% of the
phage population displaying the Ab-fusion protein as a single copy on their proximal ends
(monovalent display). The mono-valency of a phagemid-Ab library ensures that the apparent
affinity of the displayed Ab to target Ag is representative of 1:1 binding model. Our project
implemented both phage and phagemid vector systems to give rise to multivalent and
monovalent Ab-phage display libraries (Chapter 3, 4). The phage-Ab library was used in the
earlier rounds of affinity selection to weed out non-specific binders and select for medium to
high-affinity target binders, while the phagemid-Ab library was employed in later rounds of

10
panning for the isolation of a few tightest binding Abs. Further details will be described in the
following chapters.
Once displayed on the phage particles, Abs are subjected to an affinity-based in vitro
selection procedure called “biopanning,” illustrated in Figure 1-2. Biopanning is performed by
incubating the library of phage-displayed Abs with a target antigen immobilized on a solid
support (e.g., tubes and plates), then washing away the unbound phages, and finally eluting the
Ag-bound phages by disrupting the Ag-Ab interactions using low or high pH, enzyme
hydrolysis, or high-frequency agitation. The eluted phages therefore are presumed to be specific
to the target Ag. The eluted phages are then amplified by infection into E. coli, and then used for
subsequent rounds of panning. Subjecting the binder-enriched Ab-phage library to several rounds
of target binding and washing results in a stepwise enrichment of the phage pool in favor of the
tightest binding clones. Usually after 3 rounds of selection/amplification, individual clones are
characterized by ELISA and DNA sequencing. The phage output from each panning or the final
round of panning can be subjected to further mutagenesis in vitro to incorporate mutations into
the Ab clones presumed to have substantial affinities to the Ag. This project employs the phage-
Ab display method to display the Abs that have undergone affinity maturation in vitro. This will
allow two things: First, the rapid isolation of an Ab specifically binding to a target Ag via
biopanning. Second, the assessment of the effects of in vitro affinity maturation on select Abs by
comparing sequences and target binding capacities of affinity-matured variants and their original
counterparts.

11

Figure 1-2. Schematic diagram of a typical phage-antibody library biopanning process.
B. Antibody display formats
Phage-displayed Ab libraries can be in several formats, Fragment-Antigen Binding (Fab), single-
chain variable fragment (scFv), or a VHH antibody (or a nanobody). The Fab fragment on an Ab
consists of one constant and one variable domain of each of the VH and VL to form the paratope
at either amino terminus (the arms of the Y) of an Ab. The two variable domains bind the Ag.
scFv is a fusion of two variable regions, VH and VL of immunoglobulins via a flexible linker
peptide of ~10-25 amino acids rich in Serine and Glycine (Fig 1-1). Since the Ag binding site of
an Ab is retained, the Ag-binding specificity is also retained in scFvs, despite the introduction of
the linker and removal of the constant region. scFvs are generally expressed better than Fab in E.
coli, making the library generation and initial screening process more efficient (Holliger P,
Hudson PJ. 2005). Also, its relatively shorter length (~800 bp) compared to Fab (~1.5 kbp) can
be covered by a single Sanger sequencing read, making sequencing analysis of isolated clones
easier (H. Shim. 2015). In addition to conventional Abs, llamas produce functional Abs that lack

12
light chains and only contain the heavy chain. The variable domain of such a heavy chain Ab
(VHH) contains 3 CDR regions and fully capable of antigen recognition
(Muyldermans, 2013;Hassanzadeh-Ghassabeh, 2013). VHH is small (~15 kD compared to 25 kD
of a scFv and 50 kD of a Fab) has high sequence similarity to human VH3. The smaller size of
VHH enables llama Abs to fit into epitopes that are normally not accessible to traditional
antibodies. Studies have found VHH Abs to be very stable and highly soluble, which, together
with their small size, makes them an ideal candidate for the generation of Ab libraries
(Muyldermans, 2013;Hassanzadeh-Ghassabeh, 2013). Based on the above discussions, two Ab
formats will be used in this project, scFv and VHH.
Conclusion
Based on the scientific rationales of these earlier discoveries and our findings (Chapter 2),
this research extends the application of recombinant antibody technology to incorporate in vitro
Ab affinity maturation using human AID and Pol h. This approach will be thoroughly explored
in the following chapters. A standard phage display panning assay will be used to screen and
select for clones that bind to chosen Ags. To optimize Ab-Ag binding affinities, the DNA will be
recovered and subjected to further rounds of in vitro affinity maturation, by repeated exposure to
AID and pol h.
In this project, we plan to apply AID- and pol h-diversified Ab libraries to isolate Abs that
target two important antigens, 1) Glucagon-like peptide 1-receptor, whose activation stimulates
insulin secretion during glucose metabolism, and 2) glial-cell line derived family of ligands
(GDNF) neurotrophic factor artemin, known as the mediator of cold pain in humans. The
availability of Abs to these two targets will provide a potential medical breakthrough aimed at
treating type II diabetes and cold hypersensitivity.

13
Chapter 2. Analysis of in vitro AID and Polh mutation spectrum
2.1 Introduction
In chapter 1, we briefly mentioned several earlier papers from our lab in collaboration with
Matthew Scharf’s lab for which AID deamination spectra of human and murine IgV sequences
were compared to SHM spectra of B-cells (MacCarthy et al., 2009; Wei et al., 2015., Pham et al.,
2019). A consensus was established that the AID deamination patterns during SHM, which
exhibit a sequence specific mutation bias at WRC hotspots and WGCW overlapping hotspot
motifs especially in the CDR regions, are preserved in the reconstituted biochemical system that
utilizes purified AID and single-stranded IgV DNA substrates in the absence of downstream
repair machinery. Additionally, earlier studies showed that there is a significant correlation
between the SHM profiles of a mouse gene at A-T bases and substitutions generated by Polh on
the same gene in vitro (Pavlov et al. 2002), suggesting that Polh also retains its intrinsic catalytic
specificity.
Yet, the combined action of purified AID and Polh on human IgV gene have not been studied
in a cell-free system, which raises a question whether the application of AID and Polh for Ab gene
mutagenesis in vitro would produce similar effects to those of B-cell SHM. In this project, we have
examined the frequency and pattern of substitutions carried out by both AID and Polh on human
V genes in two of our biochemically reconstituted systems. This chapter describes the design of
these systems in the following order, along with the analyses of AID and Polh mutation spectra
derived from each.
1) Human IGVH3-23*01, prepared as an M13 lacZ-IgV gap DNA substrate
2) Human scFv variants made by a fusion of 18 VH genes and 20 VL genes joined by a
(G4S)4 linker via overlapping PCR, and presented as double-gapDNA substrates

14
2.2 Materials and Methods
M13mp2 phage and E.coli CSH50 and MC1061 ung- strains are from our lab collection (Pham,
2003). pADL20c phagemid vector, E. coli TG1 strain were purchased from Antibody Design
Labs (San Diego, CA). A derivative of M13 mp2 phage with two unique restriction endonuclease
sites for Pst1 and BglII downstream of lacZa gene (positions +281 and +293, respectively) was
constructed by site directed mutagenesis. Human IgHV3-23*01 sequence was cloned into the
Pst1 site in the forward orientation downstream of the lacZa gene (Wei, 2015). M13 phage were
propagated in CSH50 host cells in 2X YT medium at 37 °C. M13 phage dsDNA and ssDNA
were purified using Maxiprep plasmid and QIAprep M13 ssDNA kits (Qiagen). DNAs were
stored in 10 mM Tris (pH 8.5) at -20 °C. Minimal medium plates for M13 phage were prepared
by addition of 20 ml of 50X VB salt (MgSO4.7H2O -10 g, Citric Acid - 100 g, K2HPO4- 500 g,
Na2(NH4)HPO4.4H2O - 175 g dissolved in 1 L of H2O), 20 ml of 20% glucose, 1 ml of Thiamine
(5 mg/ml) and 2 ml of of Isopropyl β-D-1-thiogalactopyranoside (IPTG, 100 mM) to 1 L of
autoclaved agar (15 g of Bacto-agar in 1 L of H2O).

AID Purification. Human AID was expressed as a glutathione S-transferase (GST) fusion
protein in Sf9 insect cells using a baculovirus expression system (Pham, 2003) (Bransteitter,
2003). Sf9 insect cells were infected with the recombinant baculovirus expressing GST-AID with
a MOI of 3 and harvested after 72 h. Collected cells were resuspended in the insect cell lysis
buffer (10mM Tris pH 7.5, 130mM NaCl, 1% Triton X-100, 10mM Sodium fluoride, 10mM
sodium phosphate, 10mM Sodium pyrophosphate, 1mM EDTA, 1 mM DTT and protease
inhibitor) and lysed on ice for 30 min. Crude extract, containing soluble GST-AID was collected
after centrifugation of the lysed cells at 15,000 rpm for 1 h. Crude extract was incubated with
Glutathione Sepharose resin (GE Healthcare) at 4
°
C for 4h. After extensive washing with PBS

15
buffer, GST-AID (~50 kDa) was eluted from the resin using an elution buffer 10 mM Tris (pH
9.8), 500 mM NaCl, 1 mM EDTA, 1 mM DTT and 10 mM reduced glutathione. Eluted samples
were dialyzed overnight in 20mM Tris pH 7.5, 50 mM NaCl, 0.1mM dithiothreitol, 1 mM
EDTA, 10% glycerol and stored at –70 °C.
Human Pol h purification. N-terminal His-tagged full-length human polη was expressed in E.
coli (Frank, 2012). Collected cells from 5 L cultures (~25 grams) were resuspended in 180 ml of
lysis buffer (50 mM Tris pH 7.5; 500 mM NaCl, 20 mM Imidazole, 10% glycerol, 10 mM 2-
mercaptoethanol and 4 tablets of complete protease inhibitor (Roche) and lysed by French press.
After centrifugation at 20,000 x g for 45 min, the supernatant was incubated with 7.5 ml Ni-NTA
resin (Qiagen) for 30 min at 4oC on a rotating platform for 30 min. The NTA resin was washed
with 50 ml of wash buffer 1 (50 mM Tris pH 7.5; 1 M NaCl, 20 mM Imidazole, 10% glycerol,
10 mM 2-mercaptoethanol) followed by 50 ml of wash buffer 2 (10 mM Na-Phosphate pH 7.7,
500 mM NaCl, 20 mM Imidazole, 10% glycerol, 10 mM 2-mercaptoethanol). His-tagged polh
was eluted with elution buffer (10 mM Sodium Phosphate pH 7.7, 500 mM NaCl, 200 mM
Imidazole, 10% glycerol, 10 mM 2-mercaptoethanol). Pol h fractions were pooled and applied to
a Superdex G200 26/60 gel-filtration column (GE Healthcare) using running buffer (20 mM Tris
pH 7.5; 500 mM NaCl, 1 mM EDTA, 1mM DTT, 5% glycerol). Peak fractions containing
monomeric His-tagged pol h were collected and diluted with 4 volumes of a dilution buffer (20
mM Sodium Phosphate pH 7.3, 10% glycerol, 1 mM DTT). The diluted pool was loaded into an
1 ml mono S ion-exchange column equilibrated with Buffer A (20 mM Na-Phosphate pH7.3;
100 mM NaCl, 10% glycerol, 10 mM 2-mercaptoethanol). After washing the column with 20 ml
buffer A, his-pol h was eluted using a 20 ml gradient of 100 ml NaCl to 1000 mM NaCl in

16
buffer A. Pure his-pol h fractions were pooled, dialyzed overnight in a dialysis buffer (20 mM
Tris, pH 7.5; 50 mM NaCl, 1 mM EDTA, 1 mM DTT, 10% glycerol) and stored at -70 °C.

Construction of M13 gap construct with the lacZa - IgHV3-23*01 ssDNA region. Close
circular DNA gapped substrates with the lacZ - IgHV3-23*01 region as ssDNA were constructed
as follows (Pham, 2003; Wei, 2015). M13mp2 dsDNA was digested with PvuII and BglII
restriction enzymes (New England Biolabs), separated by 0.7% agarose gel electrophoresis and
~6.8 kb PvuII - BglII fragment was extracted and purified by Qiaquick Gel Extraction Kit
(Qiagen). 500 ng of the PvuII - BglII fragment of M13mp2 was denatured in a PCR tube with 45
µl of H2O at 70 °C for 5 min, followed by a quick addition of 250 ng of purified ssDNA of
M13mp2 lacZa - IgHV3-23*01 and 5 ml of 20X SSC buffer (3 M NaCl, 300 mM Sodium
citrate, pH 7.0). The mixture was incubated at 60 °C for 5 min and placed on ice. Gapped DNA
from 8 to 16 PCR tubes were pooled, desalted 3 times with H2O using Amicon Ultra-0.5 10 kDa
(Millipore) centrifugal filter unit and stored in 1 mM Tris (pH 8.0) and 0.1 mM EDTA at -20 °C.

AID deamination and pol h error-prone gap-filling synthesis in vitro. AID deamination
reactions (30 ml total volume), containing GST-AID (100 ng), RNase A (100 ng) and a gapped
DNA substrate (500 ng) dissolved in a reaction buffer (10 mM Tris-HCl, pH 8.0, 1mM EDTA, 1
mM dithiothreitol), were carried out at 37
o
C for 5 min and terminated by twice extracting the
DNA product with phenol:chloroform:isoamyl alcohol (25:24:1). The deaminated gap DNA (1
µg) were subjected to in pol h gap-filling synthesis at 37 °C for 2 h in the presence of 40 mM
Tris–HCl (pH 8.0), 50 mM NaCl, 2.5% glycerol, 10 mM Dithiothreitol, 2.5 mM MgCl2, 500 µM
each of the four dNTPs and 300 ng of human pol h. Synthesis reaction was terminated and pol h

17
was removed by twice extracting the DNA product with phenol:chlorophorm:isoamyl alcohol
(25:24:1). AID and Pol h treated DNA were desalted 4 times with H2O using Amicon Ultra-0.5
10 kDa (Millipore) centrifugal filter unit.

Analysis of AID and pol h-induced mutations in IgHV3-23*01 sequence in vitro. 50 ng of
desalted DNA were incubated with 50 µl of uracil glycosylase deficient (ung
-
) MC1061
competent cells and transformation was carried out by electroporation using a BioRad
electroporator. Following addition of 1 ml of SOC medium and incubation at 37 °C for 30 min,
aliquots of electroporated cells (5 to 200 µl) were added to a tube containing 3 ml of soft agar
(7.5 g of bacto-agar in 1 L of H2O; autoclave and keep at 42 °C), 250 µl of mid-log CSH50 a-
complementation host cells , 50 µl of 5-bromo-4-chloro-3-indolyl- beta-D-galactopyranoside (X-
gal, 50 mg/ml), 50 µl of Isopropyl β-D-1-thiogalactopyranoside (IPTG, 100 mM), mixed and
poured on the top of a minimal medium plate (Benenek, 1995; Pham, 2003). After plate
incubation overnight at 37 °C, wild-type (colorless) and mutant (light or dark blue) M13 phage
plaques were counted. DNAs from mutant M13 phages were isolated and the entire IgHV3-
23*01 region was sequenced by standard Sanger sequencing.

Construction of a mini synthetic scFv library. Synthetic human genes, VH (18 genes) and VL
(20 genes) (Suppl. Figures. List 1) were synthesized by Integrated DNA Technologies IDT
(Coralville, Iowa). Each VH gene contains a left portion of (G4S)4 linker sequence:
GGTGGTGGTGGTTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC at the 3’ end, and
each VL gene contains a right portion of (G4S)4 linker sequence:
GGCGGCGGCGGCTCCGGTGGTGGTGGATCC at the 5’ end. V gene fragments also contain

18
a BglI restriction site for cloning into pADL20c phagemid vector. Random fusion of VH to VL
via a flexible (G4S)4 linker was carried out using overlapping PCR. PCR reaction (50 µl
volume) was assembled with 5 ng of VH + VL gene pool, 100 ng each of VH-F
(TACTCGCGGCCCAGCCGGCCA) and VL-R (TGG TGT TGG CCT CCC GGG CCA)
primer, and 25 µl of 2x PCR master mix (Promega). After 25 cycles of PCR (94 °C- 1 min; 94
°C – 30 s, 55 °C – 30 s, 72 °C – 1 min for 25 cycles; 72 °C for 2 min), scFv PCR products were
purified, digested with BglI (New England Biolabs) and ligated with dephosphorylated BglI-
digested pADL-20c vector using T4 DNA ligase. Ligated DNA were transformed into E. coli
TG1 cells and plated on LB plates with 0.2% glucose and ampicillin (100 µg/ml). After
incubation overnight at 37 °C, TG1 transformant colonies were harvested, resuspended in LB
medium in the presence of 15% glycerol and stored at -80 °C.

Phagemid scFv “double gap” construction. To make the scFv double-gap constructions,
dsDNA phagemid pADL-20c and pADL20-scFv library were first digested with BglI and PvuI
restriction enzymes, respectively. Linearized dsDNAs were cleaned and purified by QIAquick
PCR purification kit (Qiagen). Linear pADL20c and pADL20-scFv dsDNAs (0.5 µg each in 45
µl of H2O) were heat denatured at 70 °C for 5 min in separate PCR tubes and combined. After
addition of 10 ul of 20X SSC buffer (3 M NaCl, 300 mM Sodium citrate, pH 7.0), the mixture
was incubated at 60 °C for 5 min and placed on ice. Double gapped DNA from 8 tubes were
pooled, desalted 3 times with H2O using Amicon Ultra-0.5 10 kDa (Millipore) centrifugal filter
unit and stored in 1 mM Tris (pH 8.0) and 0.1 mM EDTA at -20 °C. To verify the efficiency of
gap formation, 10 pg of each linear pADL20c, pADL20-scFv, and the double gap constructs
were used to transform 50 µl of TG1 electrocompetent cells by electroporation. The

19
transformation mixtures were plated on LB plates in the presence of 100 µg/ml of ampicillin and
TG1 bacterial colonies were counted after incubation at 37 °C overnight.

AID and pol h mutagenesis for scFv double-gap constructs. To maximize the scFv diversity,
the gapped constructs were incubated with AID in a series of individual tubes for different
incubation times (30 s, 45 s, 1 min, 2 min, 5 min, 10 min, 20 min and 30 min). Each AID
deamination tube (30 µl total volume), contains 500 ng of scFv double gap constructs, GST-AID
(100 ng), RNase (100 ng) in a reaction buffer (10 mM Tris-HCl, pH 8.0, 1mM EDTA, 1 mM
dithiothreitol) at 37 °C. At the indicated time point for each tube (30 s, 45 s, 1 min, 2 min, 5
min, 10 min, 20 min or 30 min), AID deamination was stopped by twice extracting the reaction
mixture with phenol:chloroform:isoamyl alcohol (25:24:1). Deaminated scFv DNAs were
combined and purified by QIAquick PCR purification kit (Qiagen). The deaminated scFv
double-gap DNA (1 µg) were subjected to pol h gap-filling synthesis at 37 °C for 2 h in the
presence of 40 mM Tris–HCl (pH 8.0), 50 mM NaCl, 2.5% glycerol, 10 mM Dithiothreitol, 2.5
mM MgCl2, 1 mM dNTPs and 300 ng of human pol h. The synthesis reaction was terminated
and pol h was removed by twice extracting the DNA product with phenol:chloroform:isoamyl
alcohol (25:24:1). AID and Pol h treated DNA were desalted 4 times with H2O using an Amicon
Ultra-0.5 10 kDa (Millipore) centrifugal filter unit and stored in 1 mM Tris (pH 8.0), 0.1 mM
EDTA at -20 oC. To analyze AID and pol h actions on the scFv sequence, the purified double
gap constructs were used to transform 50 µl of TG1 electrocompetent cells by electroporation.
The transformation mixtures were plated on LB plates in the presence of 100 µg/ml of ampicillin
and incubated at 37 °C overnight. pADL20-scFv phagemid from individual transformants were

20
isolated and scFv regions were sequenced by Sanger sequencing. Sequence alignment and
mutation analysis were performed using Sequencer program (Gene Code Corp.).
2.3 Results
1. Mutation spectrum of combined AID and Polh acting on IGVH323-01 gapDNA
construct shows that AID and Polh retain their signature catalytic specificities in vitro
To explore the application of AID and polh in IgV affinity maturation in a test tube, we
have examined their combined mutagenic actions on a human IgHV3-23*01 gene. IGHV3-23*01
is the most commonly used variable region during normal immune responses (Brezinschek,
1995) and in chronic lymphocytic leukemia (Dal-Bo, 2011).
A M13mp2 phage derivative with IgVH3-23 inserted downstream of lacZ gene was constructed
(Fig.2-1). Since AID only works on a ssDNA substrate (Bransteitter et al., 2003), partial DNA
gapped substrates with the lacZ – IgVH3-23 region exposed as ssDNA were constructed by
annealing purified M13mp2 ssDNA with a ~6.8 kb PvuII - BglII fragment of M13 dsDNA using
protocol described previously (Pham et al., 2003). The lacZ – IgVH3-23 region was first exposed
to AID deamination action for 5 min at 37
0
C followed by gap-filling synthesis by purified human
polh for 2 h in the presence of 1 mM dNTP. DNA products were transfected into E. coli cells
and plated on a-complementation host cells (CSH50 strain) in the presence of X-gal and IPTG.
AID deamination within the lacZ – IgVH3-23 ssDNA gap region causes C ® T mutations,
whereas polh causes other mutations during gap-filling synthesis. Mutation(s) in lacZ gives rise
to either white or light blue mutant M13 plaques. Since AID acts processively on ssDNA
(Chelico et al., 2009), individual substrates with mutations in lacZ reporter gene are also
expected to have mutations in the downstream IgVH3-23 region within the ssDNA gap (Pham et

21
al., 2003). DNAs from mutant M13 phages were isolated and the entire IgVH3-23 insert region
was sequenced.

A mutation spectrum of AID and Polh was obtained by sequencing 150 mutant phage clones
and compiling a total of 872 individual mutations (Fig. 2-2). C®T mutations were presumed to
have been induced by AID deamination, while the rest the of mutations were presumed to have
been added by Polh. The spectrum revealed that AID and Pol h add base substitution mutations
throughout the IgV region, most predominantly at WRC hotspot sites for AID and WA sites for
Polh. A majority of AID-induced C→T deaminations occurring within the CDR regions were
concentrated on WRC/WGCW motifs, given that 75%, 88% and 65 % of all C→T mutations in
each CDR1, CDR2, and CDR3 region, respectively, occurred within the hotspots (Table 2-1).
This observation was in agreement with the earlier findings that mutations at hot motifs are
favored in both CDR domains for B-cells and in vitro AID-RNA polymerase II system (Pham et
al., 2019), and that WRC motif is a preferred site of AID-deaminations (Rogozin et al.,1992,
2001). Furthermore, the overlapping AGCT hot motif at the 5’ end of CDR2 (nt 144) in IgVH3-
23*01 was mutated to a significant extent with our AID+Polh system, as was observed in
memory B-cells and Ramos B-cells (Wei et al., 2015). Compared to the CDR regions,
framework (FW) regions had relatively fewer deaminations occurring on the WRC motifs,
shown by the lower percentages (less than 40%) of C®T mutations confined within the WRC
hotspots relative to the total number of deaminations within each FW region (Table 2-1). This
observation reaffirmed the findings that CDR regions, which contain large numbers of WRC and
WGCW hot motifs, are strategically more favored sites of AID deamination compared to the FW
regions (Wei et al., 2015; Pham et al., 2019., Tang et al., 2020), and that this characteristic is

22
retained in the in vitro reconstituted system. Since CDR regions form the antigen binding site of
an Ab, it is expected to be susceptible to more mutations than FW regions- which encode the
overall structure of the Ab and hence tend to be more conserved (Goodman et al., 2007).

Another feature to be noted from the mutation spectrum is the accumulation of T®C mutations
at WA hotspot motifs for Polh (highlighted in yellow, Fig. 2-2). This is presumed to have arisen
from the Polh’s preferential misincorporation of dGTP opposite dT in WA (AA or TA) hotspot,
thereby generating an A®G transition on a newly synthesized strand (Johnson RE, 2000; Matsuda
T. 2000) and consequently, T®C mutation after DNA replication (Zhao Y et al., 2013).
There are also mutations found at other non-canonical hotspot sites (Fig. 2-2). Clustered mutations
(i. e. multiple mutations in a small region) in individual clones were also observed, as well as a
wide variety in the distribution and number of mutations - both of which are telling features of
AID’s processivity (Pham et al., 2003; Bransteitter et al., 2004). The formation of mutation clusters
at canonical WRC and WA somatic hypermutation hotspot motifs and the broad distribution of
mutations across entire IgVH3-23 sequence indicate that both AID and Pol h retain their signature
activities, and hence can be used for in vitro mutagenesis of Abs to mimic in vivo IgV
diversification in B-cells.

23

Figure 2-1. Analysis of AID- and polh -catalyzed mutations on human IgVH3-23 gap
substrate shows signature AID somatic hypermutation (SHM) spectrum, with C®T
deaminations preferentially occurring on WRC hotspot motifs and mutations by polh on
WA motifs.
Sketch of protocol to detect AID-catalyzed C®T mutations and polh-catalyzed mutations on individual
ssDNA substrate molecules in a gapped M13DNA construct. The ssDNA gap region is 323 nucleotides
long and contains 230 nucleotides of lacZa reporter sequence. Mutated products are analyzed after
transfecting the DNA into E. coli, followed by plating on an a-complementation strain of E. coli in the
presence of IPTG and X-gal. Mutations occurring on individual DNA substrates are detected in the lacZa
gene of mutant M13 phages (white or light blue plaques)

24

Figure 2-2. Compilation of 862 AID and Polh-catalyzed mutations across the IgVH-323
region obtained from a subset of 150 individual M13 mutant DNA clones.
Each letter denotes either a transition or a transversion mutation occurring in M13 DNA isolated from
either white or light-blue plaques. CDR 1, 2 and 3 regions are indicated with red underlines and FW 1,2,3
and 4 regions with green underlines. The locations of CDR and FW regions have been identified by Kabat
database. The AID (WRC & WGCW) and Polh (WA) hotspots are highlighted in red and yellow,
respectively. C®T deamination clusters are observed in WRC hotspots within CDR 1 and CDR 2
regions. T®C transition mutations, presumably done by pol h, accumulate on TA (WA) hotspot motifs.

25

IgVH3-23*01
Sub-region
Frequency of C®T
deamination within
WRC/WGCW
Total frequency of C®T
deamination
% deamination frequency
in WRC/WGCW
FW1 19 48 40
CDR1 65 87 75
FW2 0 15 0
CDR2 21 24 88
FW3 32 99 32
CDR3 40 62 65
FW4 7 21 33

Table 2-1. Comparison of deamination frequency in WRC/WGCW motifs within each
CDR and FW region and total number of deaminations occurring in each region of
IGVH3-23*01 in M13mp2 gapped construct. % C ®T deaminations in WRC/WGCW was
calculated by dividing the number of deaminations occurring within WRC hotspots in the designated
region by the total number of deamination occurring in that region and multiplying by 100. Both CDR1
and CDR2 exhibit the largest accumulations of deaminations within the hotspot motifs (75% and 88%,
respectively). All FW regions display equal or less than 40% of total deaminations occurring within the
AID hotspots. % deaminations in each CDR and FW region are also represented in bar graphs below
(figure 2-3).

Figure 2-3. Comparison of % deamination frequencies occurring within the WRC/WGCW
AID hotspot motifs in each CDR or FW sub-region of IGVH3-23*01 in M13mp2 gapped
construct
40
75
0
88
32
65
33
FW1 CDR1 FW2 CDR2 FW3 CDR3 FW4 % C®T IN WRC MOTIFS RELATIVE TO
TOTAL C®T IN EACH SUB REGION
POSITION ON IGV3-23*01
% DEAMINATION IN WRC

26
2. Application of AID and Polh in affinity maturation in vitro successfully diversifies a
synthetic scFv gene repertoire
The combined actions of AID and pol h on IgHV3-23*01 in the earlier section demonstrated
the applicability of AID and Polh in Ab gene diversification in vitro. As a proof of principle for
the application of AID and pol h to diversify an existing Ab gene library or sub-library, we have
constructed a “mini” synthetic scFv library using humanized 18 heavy chains (VH) and 20 light
chains (VL) of known sequence identities. These chains were synthesized by grafting CDR
cassettes from rabbit mAbs (“CDR grafting”) and integrating them into human heavy and light
chain frameworks. A total of 360 possible combinations of VHs and VLs to form scFvs was
achieved by overlapping PCR; 15 oligonucleotides complementary to the 3’ end of VHs and 5’
end of VLs were used to form the linker joining VHs to VLs. The resulting pool of scFvs was
cloned into pADL20c phagemid vector and pADL20c-scFv gene library was obtained by
transformation of ligated DNA into E. coli TG1 cells. The combined pool of recombinant
pADL20c-scFv dsDNAs was extracted by Maxiprep.
Since AID only works on a ssDNA substrate, the reconstitution of AID catalysis in a
biochemical system involves a partially gapped DNA construct exposing the target DNA
substrate as a ssDNA, as shown in the previous section. In our attempt to subject both DNA
strands of IgV to AID treatment, we have developed a “double-gapped” method– to construct
partially gapped dsDNA in which a single-stranded region exposing the scFv insert is either on a
top or bottom strand of the phagemid DNA (Fig. 2-4A). In this protocol, a double-stranded
phagemid not containing the scFv (pADL-20c) was linearized with BglI restriction enzyme and
the DNA containing a scFv insert (pADL20-scFv library) was linearized with PvuI. The
linearized DNAs were denatured in water at high temperature (70 °C), quickly mixed together,
and cooled down in the presence of salt to allow annealing to occur, leading to the formation of

27
circular gapped constructs with the scFv region exposed as ssDNA on the sense or anti-sense
strand. The efficiency of gapped construct formation was assessed by transforming 10 pg of
DNA into E. coli competent cells by electroporation. As shown in Table 2-3, linearized DNA
(both pADL20 and pADL20-scFv) only gave rise to ~ 9 to 53 colonies, whereas following
denaturing and annealing, there are 30-fold increase in the number of colonies (1724), suggesting
a very efficient circular gapped scFv formation.
To diversify the scFv repertoire, we performed in vitro mutagenesis on the resultant “double-
gapped” pADL20-scFv library using AID and Polh (Fig. 2-4B). Previous biochemical studies
have shown that AID is a very inefficient ssDNA scanning enzyme with the deamination rate of
~1 deamination per 40 s and that the average number of deaminations on individual substrates
increases proportionally with increasing incubation times (Pham, 2011; Mak, 2013). However,
there is a wide variation in the distribution regarding the numbers of mutations per clone at any
incubation time, ranging from 2 to 70 mutations per clone for the lacZ target (Pham, 2003.,
Pham, 2011). Therefore, to maximize the scFv diversity, the gapped constructs were incubated
with AID for different incubation times (from 1 to 30 min). The deaminated gapped DNAs were
then subjected to pol h error-prone gap filling synthesis for 2 h at 37°C. Pol h is a low-fidelity
polymerase that on average makes 1 mutation per 100 nucleotide incorporation while favoring
WA hotspot motifs (Pavlov, 2002). Following AID and pol h treatment, all DNAs were pooled
and used to transform E. coli cells (NR9404 ung- strain) to make AID and pol h-mutagenized
scFv gene library of size ~ 5.4 x 10
5
.

28

Figure 2-4. Schematic diagram illustrating the synthesis of scFv-double gap DNA constructs
(A) and in vitro mutagenesis of the resultant constructs by AID and Polh (B).

Table 2-2. Number of colonies on agar plate supplemented with ampicillin upon
transformation of 10 pg of linearized pADL20c vector, linearized pADL20c-scFv plasmid,
and double-gapped pADL20c-scFv construct into E.coli TG1.

No. of transformants (~10 pg DNA)
Linearized pADL20c 9
Linearized pADL20c-scFv 53
Double-gapped pADL20c-scFV 1724

29
Since the sequence identities of heavy (VH) and light chains (VL) used to generate the
synthetic scFv library were known, we were able to compare the DNA sequences of
representative scFv clones picked from the mutagenized scFv library and the original 18 VHs
and 20 VLs to assess the effects of in vitro application of AID and Polh. DNA sequences of 20
randomly mutagenized clones were aligned to the original VH sequences and mutations in each
sub-region were counted. The same procedure was carried out with another 20 scFv clones and
20 VL sequences. The results in Table 2-4. A. and B. exhibited a broad distribution in mutation
frequencies for individual clones, ranging from 1 to 14 mutations for the VH and 1 to 29
mutations for VL, consistent with the previous findings that the AID action shows a highly
diverse clonal pattern of mutations (Bransteitter et al., 2004). Additionally, clustered
deaminations in individual clones were observed, especially in scFvs H4, H5, H16, H19, L3, L5,
L14 and L20, suggesting a non-distributive and sequence-specific action by AID and Polh.

To determine whether AID and Polh's in vitro actions on synthetic scFvs are consistent
with the previous findings, we compared the average mutation frequencies of each sub-region for
20 scFv clones. As shown in Table 2-5, the highest average percent (%) mutation frequency per
nucleotide occurred in CDR regions rather than in FW regions, especially in CDR 2 for VH
(6.2 %), and CDR 2 for VL (7.4%). The highest % mutation frequencies for FW regions in VH
and VL were 3.3% in FW2 and 1.2% in FW1, respectively. This is in agreement with earlier
discoveries that AID’s deaminations were concentrated particularly in CDR 1 and 2 regions of
IGVH3-23*01 both in vivo (human memory & Ramos B-cells) and in vitro (RNAPII-
transcription system) due to the presence of overlapping AID’s WGCW hotspots (Pham et al.,

30
2019, Wei et al., 2015). Our analysis of the synthetic scFvs again showed that the combined
action of AID and Polh accumulates more mutations in the CDR than in FW.

In addition, there were clones with multiple AID-induced C to T mutations and G to A
mutations (Fig. 2-5), as well as mutations that were not C to T, or G to A, suggesting that both
AID and pol h had access to both the sense and anti-sense strands and made a wide range of
mutations on scFv gene fragments. Pairwise alignment between a representative scFv clone (10-
15) and its original VH 932 and VL 914 sequences displays the AID’s signature processive
action on both strands of scFv (Fig. 2-5); one C to T and nine G to A deaminations were
introduced in proximity on VH 932, while 16 transition and transversion mutations were added
by Polh near the deamination sites as well as other sites throughout the scFv.

31
Table 2-3. A.

Mutation frequency in each region

Clone number Clone identity Hu-VH identity
Human germline framework
identity FW1 CDR1 FW2 CDR2 FW3 CDR3

Total
H1 30-13 1080 IGVH 3-53*04 3 0 0 2 0 0

5
H2 1~1 930 IGVH 3-64*D06 1 0 0 0 0 0
1

H3 10~5 932 IGVH 4-4*02 0 0 1 0 1 0
2
H4 20-13 932 IGVH 4-4*02 3 1 9 10 6 0
29
H5 10~15 932 IGVH 4-4*02 2 5 6 10 3 0
26
H6 1~7 933 IGVH 4-4*08 0 0 0 0 1 0
1
H7 30~5 933 IGVH 4-4*08 1 0 0 0 1 0
2
H8 20-4 933 IGVH 4-4*08 2 0 0 0 3 0
5
H9 5~12 933 IGVH 4-4*08 4 1 1 3 2 0
11
H10 30-6 983 IGVH 3-23*01 1 0 0 0 0 0
1
H11 1~3 984 IGVH 3-23*01 4 0 0 0 0 0
4
H12 30~12 1059 IGVH 3-23*03 0 0 0 0 0 2
2
H13 30~13 1080 IGVH 3-53*04 3 0 0 1 0 0
4
H14 1~4 1086 IGVH 3-15*06 0 0 0 0 1 0
1
H15 10~2 1086 IGVH 3-15*06 1 0 2 0 1 1
5
H16 30~1 1085 IGVH 3-53*04 1 0 1 0 0 0
2
H17 1~16 1085 IGVH 3-53*04 4 4 5 1 8 3
25
H18 5~10 1085 IGVH 3-53*04 4 4 5 1 2 0
16
H19 20-14 985 IGVH 3-23*03 0 0 0 1 0 0
1
H20

5~15 1084 IGVH 3-33*02 9 1 4 0 4 2

20
Sum 43 16 34 29 33 8 43
31

32
Table 2-3. B.

Mutation frequency in each region
Clone
number
Clone
identity
Hu-VL
identity
Human germline
framework identity FW1 CDR1 FW2 CDR2 FW3 CDR3 Total
L1 30-16 930 IGKV1D-16*01 0 0 0 0 1 0 1
L2 5~14 930 IGKV1D-16*01 1 0 0 0 0 0 1
L3 5~8 1084 IGKV1-5*01 0 8 0 1 1 2 12
L4 10~15 914 IGKV1-5*04 0 0 0 0 2 0 2
L5 30-15 985 IGKV1-5*01 5 7 2 0 0 0 14
L6 10~4 985 IGKV1-5*01 0 0 0 0 0 1 1
L7 30~14 1066 IGKV1-27*01 1 0 1 2 1 1 6
L8 30-15 915 IGKV1-12*01 0 1 0 1 0 0 2
L9 20-11 983 IGKV1-5*01 4 0 0 0 0 0 4
L10 20-12 983 IGKV1-5*01 1 3 0 0 1 2 7
L11 1~12 983 IGKV1-5*01 0 3 0 0 0 2 5
L12 30~9 983 IGKV1-5*01 0 3 0 0 0 2 5
L13 30~11 1059 IGKV1-27*01 2 0 0 0 0 0 2
L14 5~15 1085 IGKV1-12*01 0 0 0 0 7 1 8
L15 30~4 933 IGKV1-5*04 0 2 0 1 0 1 4
L16 10~1 933 IGKV1-5*04 0 0 0 1 0 0 1
L17 30-8 1060 IGKV1-27*01 1 0 0 0 2 1 4
L18 30-14 1060 IGKV1-27*01 0 0 0 2 0 5 7
L19 30-3 1082 IGKV1-5*04 4 0 0 0 0 0 4
L20 5~16 983 IGKV1-5*01 0 8 0 2 1 2 13
Sum

19 35 3 10 16 20 103
32

33
Table 2-4. A. B. Number of mutations in each sub-region of Ig variable heavy and light chains of 20 representative scFv clones.
Substitution mutations on heavy chains were identified by pairwise alignments between DNA sequences of 20 randomly mutagenized clones
selected from the mini-synthetic library and 18 humanized heavy chain sequences originally used for library construction. The identities of human
germline IgV framework and lengths of framework (FW) and complementarity determining regions (CDR) were identified by NCBI IgBlast. For
each clone, mutations occurring in each sub-region was counted and added together to obtain a total number of mutations accumulated in each
chain. The numbers were used to calculate the normalized mutation frequency per base pair in sub-region in tables 3 & 4.
33

34

% Mutation frequency per nucleotide in each sub-region of humanized
immunoglobulin gene variable heavy chain (IgVH)
Clone no.
Hu-VH
identity
Human germline IgV framework
identity FW1 CDR1 FW2 CDR2 FW3 CDR3
H1 1080 IGVH3-53*04 4.0 0.0 0.0 9.5 0.0 0.0
H2 930 IGVH3-64*D06 1.3 0.0 0.0 0.0 0.0 0.0
H3 932 IGVH4-4*02 0.0 0.0 2.0 0.0 0.9 0.0
H4 932 IGVH4-4*03 4.0 3.7 17.6 41.7 5.3 0.0
H5 932 IGVH4-4*04 2.7 18.5 11.8 41.7 2.6 0.0
H6 933 IGVH4-4*08 0.0 0.0 0.0 0.0 0.9 0.0
H7 933 IGVH4-4*09 1.3 0.0 0.0 0.0 0.9 0.0
H8 933 IGVH4-4*10 2.7 0.0 0.0 0.0 2.6 0.0
H9 933 IGVH4-4*11 4.0 4.2 2.0 12.5 1.8 0.0
H10 983 IGVH3-23*01 1.3 0.0 0.0 0.0 0.0 0.0
H11 984 IGVH3-23*02 5.3 0.0 0.0 0.0 0.0 0.0
H12 1059 IGVH3-23*03 0.0 0.0 0.0 0.0 0.0 5.3
H13 1080 IGVH3-53*04 4.0 0.0 0.0 4.8 0.0 0.0
H14 1086 IGVH3-15*06 0.0 0.0 0.0 0.0 0.9 0.0
H15 1086 IGVH3-15*07 1.3 0.0 3.9 0.0 0.9 3.1
H16 1085 IGVH3-53*04 1.3 0.0 2.0 0.0 0.0 0.0
H17 1085 IGVH3-53*05 5.3 16.7 9.8 4.8 7.0 11.5
H18 1085 IGVH3-53*06 5.3 16.7 9.8 4.8 1.8 0.0
H19 985 IGVH3-23*03 0.0 0.0 0.0 4.8 0.0 0.0
H20 1084 IGVH3-33*02 12.0 4.2 7.8 0.0 3.5 6.3
Average per clone

2.8 3.2 3.3 6.2 1.4 1.3

Table 2-4. A.

34

35

Table 2-4. B.

% Mutation frequency per nucleotide in each sub-region of humanized
immunoglobulin gene variable light chain (IgVk)
Clone no.
Hu-VL
identity Human germline framework identity FW1 CDR1 FW2 CDR2 FW3 CDR3
L1 930 IGKV1D-16*01 0.0 0.0 0.0 0.0 0.9 0.0
L2 930 IGKV1D-16*01 1.3 0.0 0.0 0.0 0.0 0.0
L3 1084 IGKV1-5*01 0.0 33.3 0.0 11.1 0.9 6.7
L4 914 IGKV1-5*04 0.0 0.0 0.0 0.0 0.9 0.0
L5 985 IGKV1-5*01 6.4 29.2 3.9 0.0 0.0 0.0
L6 985 IGKV1-5*01 0.0 0.0 0.0 0.0 0.0 3.3
L7 1066 IGKV1-27*01 1.3 0.0 2.0 22.2 0.9 2.6
L8 915 IGKV1-12*01 0.0 5.6 0.0 11.1 0.0 0.0
L9 983 IGKV1-5*01 5.1 0.0 0.0 0.0 0.0 0.0
L10 983 IGKV1-5*01 1.3 12.5 0.0 0.0 0.9 5.1
L11 983 IGKV1-5*01 0.0 12.5 0.0 0.0 0.0 5.1
L12 983 IGKV1-5*01 0.0 12.5 0.0 0.0 0.0 5.1
L13 1059 IGKV1-27*01 2.6 0.0 0.0 0.0 0.0 0.0
L14 1085 IGKV1-12*01 0.0 0.0 0.0 0.0 6.5 2.8
L15 933 IGKV1-5*04 0.0 8.3 0.0 11.1 0.0 2.8
L16 933 IGKV1-5*04 0.0 0.0 0.0 11.1 0.0 0.0
L17 1060 IGKV1-27*01 1.3 0.0 0.0 0.0 1.9 2.6
L18 1060 IGKV1-27*01 0.0 0.0 0.0 22.2 0.0 12.8
L19 1082 IGKV1-5*04 5.1 0.0 0.0 0.0 0.0 0.0
L20 983 IGKV1-5*01 0.0 33.3 0.0 22.2 0.9 5.1
Average per clone

1.2 7.4 0.3 5.6 0.7 2.7
35

36
Table 2-5. A.B. % mutation frequency representing the number of mutations per base pair in each sub-region of Ig variable
heavy and light chain of 20 representative scFvs. Substitution mutations on heavy chains were identified by pairwise alignments between
DNA sequences of 20 randomly mutagenized clones selected from the mini-synthetic scFv library and 18 humanized heavy chain sequences
originally used for library construction. The identities of human germline IgV framework and lengths of framework (FW) and complementarity
determining regions (CDR) were identified by NCBI IgBlast. For each clone, mutation frequencies were first normalized for each sub-region by
dividing the total number of mutations accumulated in a sub-region by the length (nt) of the sub-region, then multiplied by 100 to calculate the
final % mutation frequency per base pair. Same procedure was applied to obtain the % mutation frequencies in humanized light chains of 20 scFvs.
The average % mutation frequencies for sub-region was calculated by dividing the sum of normalized mutation frequencies in each region by the
total number of clones analyzed (n = 20).
36

37

Figure 2-5. Pairwise sequence alignment of representative scFv clone (10-15), which has
been treated with 1mM of AID for 10 mins and 5.7 mM of Polh for 2 h at 37°C, and its
original humanized heavy (Hu-VH 932) and light chain (Hu-VL914).
Consensus sequence is represented in red, while the position of mutations is indicated by blue letters in
the consensus. The black dotted line represents the linker sequence incorporated during the overlapping
PCR to join the mutagenized heavy and light chain.

38
2.4 Discussion
A primary goal for investigating the combined AID and Polh mutation profiles on human
IgVH3-23*01 and humanized scFv gene repertoire was to assess whether the application of two
SHM enzymes, AID and Polh, in biochemically reconstituted systems captures the salient features
of B-cell SHM. Earlier studies show that during SHM, AID and Polh 1) introduce mutations on
IgV at a high rate, approximately million-fold higher than the basal level of mutation elsewhere in
the genome (McKean et al.,1984;Rajewsky, 1996), 2) introduce mutations preferentially on WRC
and WA hotspot motifs (Rogozin et al., 1992; Di Noia & Neuberger., 2002), particularly in the
CDR regions (Wei et al., 2015; Pham et al., 2019., Tang et al., 2020), and, 3) retain their catalytic
specificities in both B cells and in cell-free systems, as evidenced by their similar mutation profiles
in vivo and in vitro (MacCarthy et al., 2009, Wei et al., 2015., Pham et al., 2019). The mutation
profile on human IgVH3-23*01 region generated by the combined AID and Polh actions exhibited
these signature characteristics, demonstrated by the broad spectrum of substitutions covering the
entire IgV region, and accumulation of AID-induced deaminations in the WRC hotspot motifs and
Polh-induced substitutions in WA motifs, particularly in the CDR regions. Despite the absence of
MMR machinery that is required for the recruitment of Polh in our system, Polh still introduced
mutations preferentially at WA motifs on A:T bases near AID-induced deamination sites on IgV
gene, indicating that its catalytic specificity was retained in vitro. Analysis of AID and Polh-
induced mutations on the synthetic scFv gene library confirmed that the AID and Polh’s catalytic
specificities are preserved during in vitro mutagenesis, shown by the high mutation frequencies in
the CDR regions compared to the FW regions in a sample of 40 clones. It also suggested that the
application of AID and Polh could successfully diversify a V gene repertoire, as indicated by the
diversity in mutation patterns of the randomly selected scFv clones.

39
In summary, the combined AID and Polh actions on human IgV 3-23*01 gene and the scFv
gene repertoire in our in vitro system exhibited consistency with B-cell SHM. This provided the
impetus and scientific rationale to implement AID and Polh in our Ab diversification strategies in
the next chapters. We expected that the application of AID and Polh in vitro would facilitate the
same kind of mutagenesis on IgV to simulate the effects of B-cell SHM, especially in the strategic
location of CDR regions.

40
Chapter 3. Construction of AID and Polh - diversified human-scFv and llama-VHH phage
display libraries
3.1 Introduction
In Chapter 1, we discussed that the development of phage antibody display technique has made it
possible to achieve the rapid generation of monoclonal antibodies to target any desired antigens.
To increase the chance of successfully selecting monoclonal Abs against specific antigens, it is
crucial to set up a large, highly diverse repertoire of antibody genes before the production of
antibody-displaying phages. There are many ways to increase Ab repertoire diversity. The most
predominantly used approach is the mutagenesis of CDR regions by random combinations of
mono- or trinucleotide units during PCR, performed usually on synthetic Ab libraries (Shim. H,
2015). Naïve Ab libraries rely solely on natural B-cell sources for their sequence diversity if no
human input is introduced (Shim. H, 2015).

In Chapters 1 and 2, it was concluded that the key enzymes of somatic hypermutation
(SHM), AID and Polh, retain their catalytic specificities in both B cells and in cell-free systems,
shown by the good agreement between their in vitro and in vivo mutational spectra. Based on
this conclusion, we sought to maximize the diversity of naïve Ab-gene repertoires by
incorporating the AID and Polh - mediated V-gene diversification step in the construction of
human and llama Ab gene libraries. Applying in vitro affinity maturation to diversify Ab gene
libraries offers several advantages over the usual methods used in the generation of Ab libraries
such as random mutagenesis and synthetic Ab libraries; first, it can quickly increase the size and
complexity of any naïve or immune Ab libraries by a simple, one-step treatment using purified
AID and Polh enzymes. Also, previous studies suggest that IgV gene sequences have evolved to
optimize efficient mutagenesis by AID and Pol h (Tang et al., 2020; Wei et al., 2015; Shapiro et

41
al., 2002; Rada & Neuberger., 1998; Rogozin et al., 1992). Therefore, it is likely that the
application of AID and pol h to diversify V genes will allow for the production of minimal
numbers of non-productive V gene variants. Third, since AID and Pol h-generate IgV mutations
that mimic natural SHM mutations in activated B-cells, Abs generated by AID and Pol h are
likely to be highly tolerated by the human immune system. In contrast, synthetic libraries and
random mutagenesis introduce many Ab variants that might elicit human immune response. All
in all, the ability to hyper-diversify the variable region in test tubes provides a natural way (i.e.,
similar to IgV diversification in B-cells) to increase the size and complexity of existing or “to-
be-generated” human Ab libraries. Such libraries would have acquired enough complexity to
yield Abs with a high specificity against any chosen antigen, thus providing a powerful general
“toolkit” for important research, diagnostic and therapeutic purposes. This chapter describes how
we designed a novel platform for the in vitro mutagenesis of Ab libraries prior to the target-
specific selection, generating a relatively large human-scFv phage display library (~1 x 10
9
independent scFv phage clones) and llama-VHH library (~2.8 x 10
8
independent VHH clones).

42
3.2 Materials & Methods
Materials. f3TR1 phage display vector and E. coli strain K91BK (Thomas, 2010) were obtained
from George P. Smith (University of Missouri, Columbia, MO). A naïve VHH library was
purchased from Abcore, Inc. (Ramona, CA). VHH repertoire in this library was prepared from
Peripheral Blood Mononuclear cells derived from 24 non-immunized Llamas and cloned into the
pADL20c phagemid vector.

cDNA preparation. Tonsils (provided by Dr. Michael Lieber, USC Norris Cancer and Keck
Medical center, Los Angeles, CA) collected from tonsillectomies of 7 individuals were frozen in
liquid nitrogen and pulverized using pestle and mortar. Total RNA were extracted using TRIzol
reagent (Thermo Fisher Scientific) according to the manufacturer protocol. ProtoScript II First
Strand cDNA synthesis kit (New England Biolabs, MA) was used for cDNA preparation. 16
PCR tubes (40 µl volume each), containing 15 mg of RNA and the random primer mix (oligo-
dT18 and random hexamers, 6 mM) was denatured at 65 °C for 5 min and placed on ice. After
addition of the ProtoScript reaction mix and enzyme mix, cDNA synthesis was carried out by
incubation 25 °C for 5 min followed by 42 °C for 60 min and enzyme inactivation at 80 °C for 5
min. cDNA were pooled and stored at -70 °C.

1
st
PCR Amplification of variable regions of Ab heavy chains (VH) and light chain (VL)
repertoire. In order to reduce amplification bias, 1
st
PCR amplification was carried out in
independent PCR tubes to amplify individual V gene segments, using all possible combinations
with VH and VL forward and reverse primers. The primer sequences, allowing amplification of
the entire repertoire of human antibody genes (Hust, 2012) are listed Supplementary Figure. List

43
2. Each PCR reaction (50 µl volume) was carried out in the presence of 375 ng of cDNA, 100 ng
of each forward and reverse primers and 25 ml of 2X PCR master mix (Promega Corp.)
using following PCR program: 94 °C- 2 min; 94 °C – 1 min, 55 °C – 1 min, 72 °C – 2 min for
30 cycles; 72 °C for 10 min. PCR products were separated by 1.2% TAE agarose gel
electrophoresis and PCR bands corresponding to V genes were cut out (VH: ~380
bp, kappa/lambda: ~650 bp), purified by QIAquick Gel Extraction Kit (Qiagen) and eluted in 10
mM Tris, pH 8.5. PCR products from each subfamily (VH, kappa, lambda) were pooled
separately and stored at – 20 °C.
The second round of PCR introduced BglI restriction sites at the 5’-end of VH and at the 3’-end
of Vk and Vl genes. Reverse primers for VH and forward primers for Vk and Vl also contain
additional sequences to form a flexible (G4S)4 linker between VH and VL to constitute scFv. The
primers for 2nd PCR amplification are listed in Table 3. PCR reactions (50 µl volume) were set
up for all pair of forward and reverse primers in the presence of 10-20 ng of a purified PCR
product from the first PCR, 100 ng of each forward and reverse primers and 25 ml of 2X PCR
master mix (Promega Corp.). PCR program is as follow: 94 °C- 1 min; 94 °C – 30 s, 55 °C – 30
s, 72 °C – 1 min for 25 cycles; 72 °C for 5 min. The 2
nd
PCR amplification produced single VH
or VL product bands (~430-450 bp), which were combined for each subfamily and purified by
Qiaquick PCR purification kit (Qiagen).

scFv diversification by AID and pol h VH, Vk and Vl PCR products were mixed together at
molar ratios of 2 VH:1 Vk:1 Vl to constitute a V gene mixture. AID deamination was carried out
in a series of PCR tubes using a thermocycler. Each tube, containing 200 ng of the V gene mix in
H2O (50 µl volume) was heat-denatured at 94 °C for 2 min and quickly cooled down to 37 °C,

44
followed by an immediate addition of 4 µl of pre-activated GST-AID (100 ng). Pre-activated
AID was prepared by incubation of GST-AID (100 ng) and RNase (100 ng) for 3 min at 37
°C. After incubation at 37 °C for predetermined time (30 s, 45 s, 1 min, 2 min, 5 min, 10 min or
20 min), AID deamination was stopped by twice extracting the reaction mixture with
phenol:chloroform:isoamyl alcohol (25:24:1). Deaminated V gene DNAs were combined,
desalted 3 times with H2O and concentrated using Amicon Ultra-0.5 10 kDa centrifugal filter
unit (Millipore).
Pol h error-prone synthesis was performed as follows. A PCR tube (50 ml volume),
containing 50 - 100 ng of the V gene mix (AID-treated or non-AID treated), 10 ng of a forward
and reverse primers (scFv-F: TACTCGCGGCCCACGCGGCCA, scFv-R:
TGGTGTTGGCCTCAGCGGCACT) in pol h reaction buffer 40 mM Tris–HCl (pH 8.0), 50
mM NaCl, 2.5% glycerol, 10 mM Dithiothreitol, 2.5 mM MgCl2 and 500 mM each of the four
dNTPs was heat denatured and cooled to 4 °C to allow the annealing of the primers to V
genes. Since the 3’-end of VH and the 5-end of Vk/Vl PCR products contain a complementary
sequence (GGC GGC GGC GGC TCC), annealed top strands of VH and bottom strands of
Vk/Vl can also serve as primers for pol h extension. DNA synthesis was initiated by addition of
300 ng of purified human pol h and incubated for 2 h at 37 °C. The synthesis reaction was
terminated and pol h was removed by twice extracting the reaction mixture with
phenol:chlorophorm:isoamyl alcohol (25:24:1). Pol h-treated DNAs from 20 – 24 tubes were
combined, desalted 4 times with H2O using Amicon Ultra-0.5 10 kDa centrifugal filter unit
(Millipore) and stored in 1 mM Tris (pH 8.0), 0.1 mM EDTA at -20 °C.

45
Generation of scFv repertoire by overlapping PCR. V genes treated with both AID and pol h
were combined with V genes treated with AID alone and V genes treated with pol h alone and
non-treated V genes at an equal concentration to use as templates to generate mutagenized scFv
repertoire. The 3’-end of VH and the 5-end of Vk/Vl contain complementary sequences allowing
a fusion of VH and VL and forming a (G4S)4 linker for scFv by overlapping PCR. 96 PCR tubes
(50 µl volume), each contains 10 ng of the V gene mixture, 100 ng of a forward primer (scFv-F:
TACTCGCGGCCCACGCGGCCA), 100 ng of reverse primer (scFv-R:
TGGTGTTGGCCTCAGCGGCACT) and 25 ml of 2 X PCR master mix (Promega) were
subjected to 30 cycles of PCR: 94 °C- 1 min; 94 °C – 30 s, 55 °C – 30 s, 72 °C – 1 min for 30
cycles; 72 °C for 2 min. Overlapping PCR produced a single PCR product (~850-900 bp)
corresponding to scFv composition VH-(G4S)4-VL. PCR products were combined, extracted
twice with phenol:chloroform:isoamyl alcohol (25:24:1) and ethanol precipitated. The
mutagenized scFv gene repertoire was resuspended in 10 mM (Tris pH 8.5) and stored at -20
°C.

Cloning of scFv into f3TR1 phage display vector and generation of f3TR1-scFv phage display
library. The mutagenized scFv gene repertoire and f3TR1 vector dsDNA were digested with
BglI (New England Biolabs) for 3 h at 37 °C. Digested scFv DNA were purified using QIAquick
PCR purification kit (Qiagen) and eluted in 10 mM Tris (pH 8.5). The digested f3TR1 was de-
phosphorylated with Shrimp Alkaline Phosphatase (New England Biolabs), purified using a
Qiagen-tip 500 column (Qiagen) and resuspended in 10 mM Tris (pH 8.5). Ligation of scFv into
f3TR1 vector was carried out at 16 °C for 16 h by T4 DNA ligase (New England Biolabs) using

46
scFv : f3TR1 molar ratio of ~2:1. Ligated DNAs were desalted 4 times with H2O and
concentrated using Amicon Ultra-15 10 kDa centrifugal unit (Millipore) and stored at -20
o
C.
The ligated DNAs were electroporated into E. coli competent cells MC1061 using a Bio-Rad
electroporator. To generate the mutagenized scFv library, a total of 240 electroporation reactions
were carried out, in a batch of 20 cuvettes. About 50 ng of ligated DNA was used for a
transformation of 50 µl of the competent cells. After electroporation each cuvette was flushed
with 1 ml of SOC medium and transformed cells from all 20 cuvettes were combined in a 100 ml
flask. The flask was incubated at 37°C in a New Brunswick shaker (200 rpm) for 30-45 min and
the entire content was transferred into a prewarmed 2 L flask containing 1 L of 2x YT medium
supplemented with 20 µg/ml tetracycline. Aliquots (10 µl to 200 µl) from the flask with 2X YT
medium were plated immediately on LB plates containing tetracycline (20 µg/ml). After
incubation overnight at 37 °C, the number of colonies was used to determine the number of
independent transformants for each batch. Numbers of independent transformants from all 12
electroporation batches were combined to calculate the size of the mutagenized scFv library
(~1.1 x 10
9
).

Purification of primary f3TR1-scFv phage library. Flasks with transformants in 2x YT medium
were incubated at 37 °C for 16 h to allow the production of primary f3TR1-scFv phages. The
culture supernatants (total of 12 L) were collected after centrifugation and phages were
precipitated by adding 15% volume of PEG/NaCl solution (16.6% PEG 8000 MW, 3.3 M NaCl)
for 4 h at 4 °C. Precipitated phages were harvested by centrifugation at 10,000 x g for 20 min.
The phage pellets were resuspended in TBS buffer (1/30 volume of supernatant) and subjected to
a second round of PEG/NaCl precipitation. Phage pellet from the second precipitation was

47
resuspended in TBS buffer and 50% glycerol (typically 1 ml for each liter of supernatant) and
stored at -20 °C.
Titration of f3TR1-scFv phage tetracycline transducing units (TU). Aliquots (10 ml each) of
serially diluted phage library (10
-2
, 10
-4
, 10
-6
, 10
-8
, 10
-10
dilution in TBS buffer) were mixed with
0.5 ml of mid-log K91BK cells in 2x YT medium and incubated at 37 °C to allow infection.
After 30 min, infected cells (100 ml) were plated on LB plates supplemented tetracycline (20
µg/ml) and kanamycin (20 µg/ml). After incubation at 37 °C overnight, phage library titer of
transducing unit (TU) was calculated based on numbers of colonies on plates and dilution factor.

Preparation of high-titer phage stocks for bio-panning. K91BK cells were grown in 1 L flask
containing 500 ml of 2x YT at 37 °C until OD600 of 1/10 dilution is 0.2 (~ 2.5 x 10
12
cells) and
infected with primary scFv phage (7.5 x 10
11
TU). The flask was incubated without shaking for
15 min and six 90 ml each of the culture were distributed to six flasks each contains 1 L of pre-
warmed 2x YT medium supplemented with 0.22 µg/ml tetracycline (6 flasks total). After shaking
at 37 °C for 35 min, tetracycline was added to a final concentration of (20 µg/ml), and phage
production was continued for 16 h. A high titer human scFv phage library stock (5 x 10
12
TU/ml)
was harvested and purified as described above.

Generation and preparation of mutagenized Llama f3TR1-VHH phage library. Llama VHH
repertoire was PCR amplified from Abcore’s naïve VHH library using a forward
(TATTACTCGCGGCCCACGCGGCCATGGCT) and a reverse
(GGTGATGGTGTTGGCCCCAGGGGCTGAGGAGACGGTGAC) primers, which incorporate
BglI restriction sites for cloning into f3TR1. VHH repertoire PCR products were subjected to

48
AID and pol h diversification using procedures described for the mutagenized human-scFv
library. 160 transformations by electroporation were carried out to obtain a mutagenized Llama
VHH library (2.8 x 10
8
independent clones). A high titer VHH phage library stock (1.9 x 10
13

TU/ml) was prepared for bio-panning.

3.3 Results
1. Construction of a mutagenized human scFv phage display library
A human-scFv library in f3TR1 phage display vector (Thomas, 2010) was constructed in 4
steps: i, Heavy chain VH and light chain Vk and Vl gene repertoires were PCR amplified from
tonsil B-cells cDNA using a set of subfamily-specific forward and reverse primers (Hust,
2012)(Suppl. Fig. List 2); In the second PCR, adaptor sequences were incorporated into the ends
of VH and VL, which allow random fusion of VH and VL via a flexible (G4S)4 linker by
overlapping PCR in a later step. BglI restriction sites are also added at the 5’ end of VH and 3’-
end of VL for cloning into the phage vector; ii, V gene repertoires were denatured by heating at
95 °C for 2 min, quickly cooled down and incubated with purified AID at 37 °C to allow dC
deaminations on both denatured ssDNA strands resulting in C to T or G to A mutations. Since a
longer incubation time lead to a higher average number of AID-induced mutations on ssDNA
substrates (Pham, 2011;Mak, 2013), various AID incubation times (30 s, 45 s, 1 min, 2 min, 5
min, 10 min and 20 min) were done in separate tubes and subsequently combined to increase the
ranges of mutations on individual scFv; iii, AID-treated V genes were annealed to primers and
one round of pol h extension synthesis was carried out to introduce mutations at A and T sites;
iv, Overlapping PCR was used to generate scFv gene repertoire. scFv PCR products were
digested with BglI and directionally ligated into BglI sites of f3TR1 phage vector. Ligated DNA

49
were transformed into MC1061 E. coli cells by electroporation. A total of 240 independent
electroporations were carried out to generate a human f3TR1-scFv library composed of 1.1 x 10
9

independent phage clones.
The diversity of the scFv repertoire and the quality of the primary library were examined by
PCR analysis and by sequencing DNA segments encoding the scFv genes from 16 randomly
picked f3TR1-scFv clones. Analysis of 16 clones by PCR using primers flanking the BglI sites
showed that 15 clones (94%) contain an insert corresponding to the expected scFv
size. Sequencing analysis and comparison to germline V-genes (IMGT/V-QUEST) of 16 scFv
clones showed that all clones have different combination of heavy VH and light VL chains
(Table 1). The variable regions were derived from 12 different V gene families, including five
VH gene families (VH1, VH2, VH3, VH4, and VH6) and seven VL gene families of both kappa
and lambda light chain (Vκ1, Vκ2, Vκ3, Vκ5 and Vλ1, Vλ3, Vλ6). The CDR3 of VH sequences
were highly diverse, with lengths between 9 to 19 amino acids. The VL sequences had between 9
and 12 amino acids in their CDR3 regions (Table 3-1). Thus, the scFv gene fragments were
distributed across the full repertoire of antibody germ line genes. The number of different bases
from germ line genes varied from 0-32 mutations for VH and 0–48 mutations for VL. Only two
V regions have no mutations indicating these V genes have not undergone affinity maturation in
B-cells or in vitro (Table 3-1). Most of VH and VL in the library contain multiple mutations,
suggesting that these clones arose from antibodies produced in B-cells during secondary immune
responses and/or during AID and Pol h affinity maturation in vitro. A representative scFv clone
(F3) displaying a clear signature of AID-induced C to T and G to A mutations in vitro is shown
in Fig. 3-1. Compared to the germ line genes, the heavy chain of this clone has 4 mutations, all C
to T (Fig. 3-1.A). The light chain contains 19 AID-induced C to T or G to A mutations along

50
with other mutations, some of which can be attributed to pol h (Fig. 3-1B). Actions of AID and
pol h on scFv in vitro are also seen by the presence of mutations in the linker (G4S)4 sequences,
as shown in Fig 3-2. Since this linker sequence was added to the 2
nd
PCR primer set to fuse VH
and VL chains in the scFv library, mutations in the sequence were likely to be caused by AID
and pol h treatment during the mutagenesis steps. Among 16 sequenced clones, 9 did not have
any mutations in the linker sequence, while 5 clones had an AID-induced C to T or G to A
mutation and two clones contained pol h-induced mutations that were not C to T or G to A.

Table 3-1. Sequencing analysis of 16 representative clones randomly selected from
mutagenized human f3TR1-scFv phage display library

51
(A)

(B)

Figure 3-1. Pairwise alignment of DNA sequences of a scFv clone (F3) randomly picked
from the mutagenized human scFv-phage library and human germline IgV heavy (A) and
light chain (B) genes. Germline sequence identification was done by aligning the scFv sequence to
IMGT database using NCBI IgBlast.

52

Figure 3-2. Sequence alignment for 27 scFv clones randomly selected from mutagenized
human scFv-phage library, showing parts of the heavy chain and Serine-Glycine linker
regions. AID and polh induced mutations on scFv in vitro are indicated by the presence of mutations in
the linker (G 4S) 4 sequences (red), represented by black letters.

2. Construction of mutagenized llama VHH-phage display library
In addition to conventional Abs, llamas produce functional Abs that lack light chains and
only contain the heavy chain. The variable domain of such a heavy chain Ab (VHH) contains 3
CDR regions and fully capable of antigen recognition (Muyldermans, 2013;Hassanzadeh-
Ghassabeh, 2013). VHH is small (~15 kD) has high sequence similarity to human VH3. The
smaller size of VHH enables llama Abs to fit into epitopes that are normally not accessible to
traditional antibodies. Studies have found VHH Abs to be very stable and highly soluble, which,
together with their small size, makes them an ideal candidate for the generation of Ab libraries
(Muyldermans, 2013;Hassanzadeh-Ghassabeh, 2013).

53
A llama mutagenized f3TR1-VHH phage library was constructed using an available naïve
Llama VHH phage library (Abcore, Ramona, CA). Llama VHH repertoire was amplified from
the naïve Llama VHH library using forward and reverse primers, which contain BglI restriction
sites to allow subcloning VHH genes into f3TR1 phage vector. VHH gene library was
sequentially mutagenized by AID and pol h as described for human scFv library. PCR amplified
VHH pool was digested with BglI and directionally cloned into the BglI sites of f3TR1 phage
vector. Ligated DNAs were transformed into E. coli MC1061 by electroporation (160
transformations) to obtain a moderate size mutagenized f3TR1-VHH phage library containing
~2.8 x 10
8
independent clones. PCR analyses showed that over 90% of phage clones contain an
insert consistent with VHH size (350 – 380 bp).
3.4 Discussion
We have developed a novel platform for in vitro Ab library diversification to mimic the
effects of B-cell SHM by utilizing the SHM enzymes AID and Polh. Purified human AID and
Polh were used to mutagenize naive human-scFv and llama-VHH libraries. Sequence
characterization of the scFv clones selected from the mutagenized human-scFv library showed
that mutations have been successfully introduced in vitro, indicated by the substitution mutations
in the consensus linker sequence of known sequence identity. Since the linker sequence was
added during the overlapping PCR step to join the heavy and light chains, performed after cDNA
synthesis following mRNA extraction from B-cells, mutations in this sequence were likely
caused by AID and Polh treatment in vitro. However, because these scFv clones originated from
naive B-cells and their sequence identities prior to AID and Polh mutagenesis were unknown,
we could not distinguish mutations that had been incorporated during B-cell immune responses
from mutations that were induced by AID and Polh during our library diversification process.

54
Therefore, further analysis of mutagenized scFv/VHH genes with known sequence identities
would have to be carried out to assess the effectiveness of in vitro application of AID and Polh
in Ab optimization protocol.

55
Chapter 4. Application of AID and Polh affinity maturation in vitro to isolate Abs
targeting specific antigens: Target 1 – GLP-1R
4.1 Introduction
From chapters 2 and 3, we have shown that the key features of in vivo SHM, such as the
mutation bias on WRC and WA hotspot motifs, and relatively higher frequency of substitutions
in CDR regions rather than in FW regions, can be translated in reconstituted biochemical
systems via the application of AID and Polh in vitro. However, our attempts to apply in vitro
mutagenesis so far have been limited to Abs that have not yet encountered any specific Ags, such
as those in naïve scFv/VHH libraries. Since affinity maturation of B cells involves multiple
rounds of SHM in IgV regions after antigen exposure, followed by the clonal selection of target-
specific Abs, we sought to simulate this in vivo process by focusing now on single Ab clones
selected against a specific target. First, we isolated a subset of antibody candidates from phage-
displayed Ab libraries exhibiting target-specific affinities towards a chosen antigen via
biopanning. Then we subjected the isolated Ab clones to further in vitro affinity maturation using
AID and Polh, followed by additional rounds of biopanning of the ‘affinity-matured’ sub-library.
By further diversifying Abs that already show target-specific affinity towards an antigen, we
expected to be able to acquire Ab variants that have 'evolved' and developed higher
affinities. Finally, all Ab clones were purified for characterization assays, and analysis was done
to assess the effects of in vitro affinity maturation. Thus, sequences and binding constants of
affinity-matured Abs and their parental clones were compared. In this and the next chapter, we
describe our endeavor in detail with emphasis on the specific targets we selected, GLP-1R
(Chapter 4) and Artemin (Chapter 5).

56
Glucagon-like peptide-1 receptor (GLP-1R)
Type II diabetes is a multifactorial disease characterized by the impairment in the way the
body regulates and metabolizes glucose. Its causes can either be the pancreatic beta cells’ low
production of insulin, a hormone that regulates glucose metabolism, or the body cells’
insensitivity to insulin. Both can lead to chronic hyperglycemia that can eventually cause
disorders of the circulatory, nervous, and immune system. Insulin secretion is stimulated by the
increase in cAMP level upon binding of glucagon-like peptide-1 (GLP-1) to its receptor, GLP-
1R, in a glucose-dependent manner (Nauck et al., 1993; Coopman et al., 2010). Glucagon-like
peptide-1 receptor (GLP-1R), a transmembrane protein widely expressed in the pancreas, lungs,
heart, kidney, stomach, and brain, belongs to a Family B G-Protein Coupled Receptor (GPCR)
proteins with a characteristic 7 transmembrane structure (Palczewski et al, 2000). Due to its role
in glucose metabolism, GLP-1R has been a major therapeutic target for the treatment of type II
diabetes. Although GLP-1 acts as a natural ligand for GLP-1R, its half-life in the body is limited
due to its degradation by the enzyme Dipeptidyl Peptidase IV (DPPIV)(Deacon et al., 1995).
Consequently, developing long-lasting, DPPIV resistant agonists of the GLP-1R has been
proposed as a potential clinical solution for treating type II diabetes.

57
4.2 Materials & Methods
Antigens. Live cultures of Sf9 insect cells expressing a recombinant human His6-tagged
Glucagon-like peptide-1 receptor (GLP-1R) and biomass for purification (Wu, 2020) were
obtained from Raymond Stevens’ lab (Bridge Institute, University of Southern California, CA).
Recombinant GLP-1R protein was purified as described (Wu, 2020). Recombinant histidine-
tagged human GLP-1R N-terminal domain expressed in yeast (aa24-145, LS-G22934) was
purchased from LSBio, Seattle, WA, and reconstituted in water to a concentration of 1 mg/mL.
5% glycerol (final concentration) was added for storage at -80°C.

Selection of phage antibody library by Bio-panning
Sf9 insect cell panning was carried out at 4 °C, human f3TR1-scFv phage (1 x 10
11
TU) was pre-
incubated with 2 × 10
7
control Sf9 cells in PBS + 2.5% BSA for 1 h to deplete non-specific
binding phages. After centrifugation of control Sf9 cells at 300 x g for 5 minutes, the unbound
phage supernatant was incubated with 1 × 10
7
GLP-1R expressing cells pre-blocked in PBS +
2.5% BSA for 1 h at 4C. Ice cold PBS was used for washing (10 times) by centrifugation at 300
x g for 5 min. Bound phages were eluted by resuspending cells in 500 µl of PBS containing 10
μg/ml trypsin and incubating at room temperature (RT) for 30 min.

Biopanning on purified protein is done as follows; A polystyrene immunotube (Nunc, Rochester,
NY) was coated with 500 nM of human GLP1-R-His6 in 100 mM sodium bicarbonate buffer (pH
9.6). The immunotube was incubated overnight on a labquake at 4°C. The tube was washed two
times with 4.5 mL of 0.1% Tween PBS (pH7.4) and two times with PBS on the next day. The
tube was filled with 2% non-fat dried milk PBS (2% MPBS) and incubated for two hours at
room temperature for blocking. Excess blocker was removed by rinsing the tube three times with

58
4.5 mL of PBS. The initial f3TR1-scFv library (for first round of panning) or amplified eluted
phages (for subsequent rounds of panning) was added to the immunotube at a concentration of 1
x 10
11

TU in 1 ml of 2% MPBS. The library was incubated for 2 h at RT on a labquake. The
unbound library was removed by aspiration, and the weakly bound phages were removed by
washing the tube 9 times with 0.1% Tween-PBS and 2 times with PBS, at room temperature for
5 minutes per wash.
The bound phages were eluted with 1 mL of 10 µg/ml trypsin in PBS, pH 7.4, for 30 minutes at
RT on a labquake. 10 µl of the eluted phages were immediately titered, while the rest of the
eluted phages were infected into 20 mL of E. coli K91BK at an OD600 of 0.8-1.0 for phage
amplification. The infected E. coli was incubated for 30 minutes at 37°C. Following incubation,
the infected cells were diluted in an additional 80 mL of 2xTY medium supplemented with
tetracycline (20 µg/ml) and grown overnight at 37 in a shaker (220 rpm). The culture was
centrifuged the next day at 8000 rpm for 10 minutes at 4°C. Phage-containing supernatant was
precipitated with 1/10 volume of 16.7% PEG/3.3M NaCl for 4 hours at 4°C. The PEG-
precipitated supernatant was centrifuged at 10,000 rpm for 10 minutes at 4°C. The pellet was
suspended in 5 ml TBS and precipitated again with 1/10 volume of 16.7% PEG/3.3M NaCl for 1
hour at 4°C. Phage precipitate was centrifuged at 5000 rpm for 30 minutes at 4°C, then the pellet
was resuspended in 2 mL of TBS in 50% glycerol and centrifuged at 13000 rpm for 5 minutes at
room temperature to remove bacterial debris. The supernatant was transferred to a new tube to be
titered and stored at 4°C until used in further rounds of panning or screening.

59
Screening for antigen-specific clones by phage ELISA
Individual K91BK colonies on titration plates from the last panning were distributed into 96-
deep well plates containing 1.5 ml of 2x YT + 20 µg/ml tetracycline and grown overnight in a
shaker at 37
o
C to produce scFv phage particles. Phages from 1 ml of supernatant were
precipitated with 150 µl of the PEG/NaCl solution (15% v/v) and resuspended in 200 µl of PBS,
pH 7.4.
Each well of Nunc Maxisorp 96-well microplates was coated with 100 µl of 5 μg/ml of
recombinant human GLP-1R in 100 nM carbonate buffer (pH 9.6). After overnight incubation at
4 °C, plates were blocked with 2% MPBS for 1.5 hour at RT followed by 3 washed with PBS-T
and 3 washes with PBS. All washings were performed on AquaMax 2000 plate washer
(Molecular Devices, LLC). The selected phage preparation was diluted 1:2 in 4% MPBS before
adding 100 µl into each well, and incubated for 1.5 hour at RT. The plates were washed four
times with PBS-T, followed by four times with PBS, and incubated with 100 µl of a 1: 4000
dilution of anti-M13-HRP (Sino Biological US, Inc., Wayne, PA) in 2% MPBS for 1h at RT.
Plates were washed 4 times with PBS-T and 4 times with PBS. TMB substrate solution 100 µl
(Thermo Fisher Scientific) was added to each well (for 2 min to 30 min). After adding 100 µl of
2 M sulfuric acid stop solution, the absorbance was read at 450 nm, using SpectraMax iD5
microplate reader (Molecular Devices, LLC). The same process was carried out for the
recombinant A2a-His6 background binding control Ag.

Construction of f3TR1-scFv gapDNA construct with the scFv ssDNA region
Closed circular DNA gapped substrates with the scFv region as ssDNA were constructed as
follows. f3TR1-dsDNA vector was linearized with BglI restriction enzyme (New England

60
Biolabs), separated by 0.7% agarose gel electrophoresis and ~9 kb fragment was extracted and
purified by Qiaquick Gel Extraction Kit (Qiagen). 500 ng of the BglI cut f3TR1-dsDNA was
denatured in a PCR tube with 45 µl of H2O at 70 °C for 5 min, followed by a rapid addition of
~270 ng of purified ssDNA of f3TR1-scFv phages and 5 µl of 20X SSC buffer (3 M NaCl, 300
mM Sodium citrate, pH 7.0). The mixture was incubated at 60 °C for 5 min and placed on ice for
30 min. Gapped DNA from 8 to 16 PCR tubes were pooled, followed by the addition of P3
buffer (1/10 reaction volume) and 100% ethanol (2 x reaction volume) and incubation on ice for
30 mins. After centrifugation at 13000 rpm for 10 min, the gapped DNA was washed with 70%
ethanol and dried in a SpeedVac vacuum concentrator (Thermo Fischer Scientific). The final
product was resuspended and stored in 1 mM Tris (pH 8.0) and 0.1 mM EDTA at -20 °C.

AID and pol h mutagenesis for f3TR1-scFv gapDNA constructs. The gapped constructs were
incubated with purified human AID in a series of individual tubes for different incubation times
(30 s, 1 min, 2 min, 5 min, 10 min, 20 min and 60 min). Each AID deamination tube (50 µl total
volume), contains 500 ng of scFv double gap constructs and AID3A variant in a reaction buffer
(10 mM Tris-HCl, pH 8.0, 1mM EDTA, 1 mM dithiothreitol) at 37 °C. At the indicated time
point for each reaction tube (30 s, 1 min, 2 min, 5 min, 10 min, 20 min or 1 hr), AID deamination
was stopped by twice extracting the reaction mixture with phenol:chloroform:isoamyl alcohol
(25:24:1). Deaminated scFv DNAs were combined and purified by Monarch DNA Clean-up kit
(NEB). The deaminated scFv-gap DNA (1 µg) were subjected to in pol h gap-filling synthesis at
37 °C for 2 h in the presence of 40 mM Tris–HCl (pH 8.0), 50 mM NaCl, 2.5% glycerol, 10 mM
Dithiothreitol, 2.5 mM MgCl2, 1 mM dNTPs and 0.4 mM of human pol h. Synthesis reaction

61
was terminated and pol h was removed by twice extracting the DNA product with
phenol:chloroform:isoamyl alcohol (25:24:1). AID and Pol h treated gapDNAs were desalted
2 times with H2O using Amicon Ultra-0.5 10 kDa (Millipore) centrifugal filter unit and stored in
1 mM Tris (pH 8.0), 0.1 mM EDTA at -20 °C.

AID and pol h mutagenesis for f3TR1-scFv single-stranded DNAs.
f3TR1-scFv ssNDAs were extracted by first diluting an aliquot of f3TR1-scFv phage library in 3
times volume of 10 mM Tris-HCl (pH 8.0) containing 1 mM EDTA, then extracting twice with
equal volume of phenol:chloroform:isoamyl alcohol (25:24:1). ssDNA were purified using a
Monarch DNA clean-up kit (NEB), analyzed on 0.7% agarose gel and quantified with Qubit 4
Fluorometer (Thermo Fischer Scientific). 0.3 µg of f3TR1 scFv-ssDNA was used in 50 µl total
reaction. 10 µl of 1 µM, 5 µM, and 10 µM of human AID3A variant in reaction buffer (10 mM
Tris-HCl, pH 8.0, 1mM EDTA, 1 mM dithiothreitol) was used for each of 7 reactions. Each
reaction was incubated for different durations of time at 37C; 30s, 1 min, 2 min, 5 min, 10 min,
20 min, and 60 mins. Reactions were quenched and twice extracted by
phenol:chloroform:isoamyl alcohol (25:24:1), and the resulting DNA was cleaned up using the
Monarch DNA clean-up kit (NEB). AID-deaminated f3TR1 scFv-ssDNA was mixed with 1 x
Polh buffer (40 mM Tris–HCl (pH 8.0), 50 mM NaCl, 2.5% glycerol, 10 mM Dithiothreitol, 2.5
mM MgCl2), 1 mM dNTPs, and VHH-ftopADL reverse primer (molar ratio ss-DNA : reverse
primer = 1:2) and briefly heated at 95 °C for 2 min, followed by cooling down on ice for 5 min
to allow the primer annealing on ssDNA. 0.4µM (final conc) human polh was added and the
mixture was incubated at 37 °C for 2h. The AID and Polh mutagenized scFv-dsDNA region was
PCR amplified using pADL20c vector-specific forward and reverse primers, and cleaned-up

62
using Monarch kit. After BglI digestion, the resultant scFv pool was ligated into pADL20c-
vector using 1:6 vector:insert molar ratio, and transformed into TG1 electrocompetent cells.

Amplification of mutagenized pADL20c-scFv phage library
OD600 of the 100-fold dilution of the TG1 phage-scFv library stock (pooled transformants) was
measured to quantify the number of colonies present in the inoculum. At least 10
10
colonies were
added into 100 mL of 2xYT with 100 µg/mL carbenicillin to ensure sufficient diversity of the
initial amplified phage-scFv library. When TG1 growth reached OD600 of 0.4~0.5 in a 37C
shaker, 1µl of CM13d3 helper phages per 1 mL of culture was added according to the
manufacturer’s protocol (Antibody Design Labs, San Diego, CA). After incubation for 1 hour at
250 rpm at 37C, 20 µg/mL kanamycin was added and the culture was incubated at 30C overnight
in a shaker. Purification of the pADL20c phagemid-Ab library followed the same protocol used
to produce f3TR1-scFv/VHH libraries (Chapter 3).

Expression and purification of soluble scFvs. Coding sequences for GLP-1R-specific scFv were
subcloned into the BglI sites of pADL-20c phagemid vector. The constructs were transformed
into E. coli CSH50 cells and Abs were expressed as His6-tagged soluble proteins in the
periplasm. Cells were grown at 37
o
C in LB + 0.2% glucose + Ampicillin (100 µg/ml). When the
culture OD600 reaches ~0.7, IPTG (1 mM) was added to induce scFv expression at 30
o
C
overnight. Cells from 1 L culture were harvested by centrifugation at 7000 rpm for 7 minutes,
then resuspended in 400 ml ice-cold TES buffer (30 mM Tris (pH 8.0), and 20% sucrose). EDTA
was added dropwise to 1 mM. After incubation at 4 °C for 10 min, cells were spun down at 8000
x g for 20 min. The pellet was resuspended in 400 ml of ice cold 5 mM MgSO4 and incubated

63
for 10-15 min at 4
o
C. The supernantant containing a soluble scFv protein was collected by
centrifugation at 8000 x g for 20 min. His6-tagged scFv proteins were purified by affinity
chromatography using 2-3 ml of Ni-NTA resin (Qiagen) using PBS as a wash buffer and eluted
in PBS containing 300 mM Imidazole. scFvs were further purified by gel filtration using a
Superdex 75 column (GE Healthcare). Proteins were concentrated using Amicon Ultra-15 10
kDa centrifugal filter unit (Millipore) and stored in PBS at -70 °C.

Titration ELISA. Titration ELISA was used to measure binding of purified scFv to GLP-1R. A
96-well Maxisorp plate (Nunc) was coated with 100 µl of 10 µg/ml GLP-1R (without His-tag)
and same amount of A2a (without His-tag) negative control overnight at 4
o
C, blocked in 2%
MPBS for 2h the next morning, then washed three times with TPBS and three times with PBS.
A purified scFv was prepared at varying concentrations (from 1 nM to 2 µM) in 2% MPBS. 100
µl of scFv aliquot of each concentration was added to a well and incubated at RT for 1.5 h. The
plate was then washed with TPBS and PBS three times each. Primary anti His-tag Mouse mAb
(Cell signaling Technology, Inc.) was diluted 1:250 in 2% MPBS and added into each well. After
1.5 h incubation, wells were washed and secondary anti-mouse (goat-anti mouse-HRP; Santa
Cruz Biotechnology) diluted 1:1000 in 2% MPBS was added and incubated for 1 h. After
washing, TMB substrate solution was added for color development (15 to 30 min) and 2M
sulfuric acid was added to stop the reaction. The absorbance was read at 450 nm, using
SpectraMax iD5 microplate reader (Molecular Devices, LLC). The half-maximal binding
concentration, EC50, was calculated according to a published protocol (Eble, 2018).

64
Surface plasmon resonance (SPR). Binding of purified scFvs to N-terminal GLP-1R
extracellular domain peptides (aa 24-145, yeast, LSBio) was determined using Biacore T100
instrument (GE Healthcare). GLP-1R ECD was suspended in 10 mM sodium acetate (pH 4.5)
and immobilized on a CM5 Series S sensor chip (GE Healthcare) at 250 RU (response unit)
using amine coupling chemistry according to the manufacturer’s protocol (GE Healthcare).
Purified scFv at a concentration ranging from 0-500 nM in flow buffer (HBS-EP: 10 mM
HEPES, pH 7.4, 0.15M NaCl, 3 mM EDTA, 0.005% v/v Surfactant P20) was injected onto the
flow cells at a flow rate of 30 ml/min for 120 seconds, and allowed to dissociate for 600 seconds.
A blank cell was used as a non-binding reference. The sensor chip surface was regenerated using
4 mM NaOH. Kinetic constants for binding interaction were determined by fitting the
sensorgrams with 1:1 binding model after blank subtraction, using Biacore T100 Evaluation
Software, version 2.0 (GE healthcare).

4.2 Results
1. Eight high-affinity scFv binders to GLP-1R were isolated by biopanning of the human
scFv-phage library and characterized
To select for scFvs specific to a native state GLP-1R, Sf9 insect cells expressing GLP-1R
and purified GLP-1R- His6 whole membrane protein (provided by Raymond Stevens Lab, USC
Bridge Institute) were used alternately as the Ag targets for biopanning. First, the human
mutagenized scFv-phage library was used in 4 rounds of biopanning against GLP-1R-expressing
Sf9 cells. 1 x 10
11
scFv-phages were pre-incubated with 2 x 10
7
control Sf9 cells to deplete non-
specific binding phages, then mixed with 1 x 10
7
GLP-1R expressing Sf9 cells. After 1h
incubation, phages that were not bound to the GLP-1R expressing cells were removed by
addition of PBS to the cells, followed by centrifugation at 300 x g for 5 minutes and discarding

65
of the supernatant. scFv-phages that have high affinities to GLP-1R would remain bound to the
cell pellet even after stringent washes. After 10 washes, the cell-bound phages were eluted by
incubation in trypsin-PBS solution at room temperature for 30 minutes. Phage titration and
amplification were done according to the procedure described in Chapter 3, Materials &
Methods. Clones retrieved from successive panning were expected to be increasingly enriched
for GLP-1R-specific scFv encoding sequences. Since both Ag specificity and genetic diversity
were desired, eluted phage output from both 3
rd
and 4
th
panning round were combined and
amplified.
Two additional rounds of biopanning were performed on the combined scFv-phage
library using purified c-terminal poly-histidine tagged GLP-1R protein immobilized on a
polystyrene immunotube. 1 x 10
12
TU of scFv-phages were added into the immunotube and
incubated for 2h to allow Ab-Ag interaction. Non-bound phages were removed from the tube by
washing 10 times with PBS-Tween, followed by 2 times washing with PBS. GLP-1R bound
phages were eluted by incubating with trypsin solution at room temperature for 30 minutes.
Eluted phages were amplified according to the previously described phage amplification
protocol, and the final 2
nd
panning output was used to infect E. coli at an exponentially growing
phase. The infected bacterial cells were spread on agar plates supplemented with carbenicillin to
only select for clones carrying the recombinant phage-scFv gene. 96 individual colonies were
randomly picked for screening by phage-ELISA. Another GPCR protein, recombinant
adenosine-2-receptor (A2a-His6), was used as an off-target control to assess the background
binding by the scFv-phages. The ratio of GLP-1R binding signal and background signal (A450
nm) was calculated for each phage-scFv clone, and positive binders were identified as those

66
giving A450 nm signals 3-fold above background (Fig 4-1). Using this criterion, 8 scFvs were
isolated and their sequences were identified by Sanger sequencing (Fig 4-2).

Figure. 4-1. ELISA based binding of 31 individual human scFv-phages to purified human
GLP1-R.
1 µg of recombinant GLP1-R and A2a negative control was immobilized on each well of a Nunc 96-well
Maxisorp plate and incubated with 50 µl of amplified 31 scFv-phage overnight cultures diluted 2-fold in
2% milk-PBS. For this assay, 25 scFv-phage clones were selected from 2
nd
panning output against
purified GLP1-R (labeled GHM), and 6 scFv-phage clones were selected from 4
th
panning output against
Sf9 cells expressing GLP1-R (labeled GC) – both groups of clones were picked from earlier 96 x well
phage-ELISAs. Binding of scFv-phages was followed by immunodetection of the phage coat protein p8
by anti-M13 p8 conjugated with HRP (Sinobiological). Assay was done once (n = 1), and background
binding of scFv-phage to A2a is represented as orange bars, while binding to GLP1-R is represented as
blue bars on the graph. Positive binders were defined as those having A450 (GLP1-R): A450 (A2a) ratio
of 3:1. Some of GHM clones, 25, 58, 66, 94 (shown on graph), and 71, 75, 79, 95 (not shown on graph)
were sequenced and purified for further binding studies.

GHM 2
GHM 9
GHM 25
GHM 33
GHM41
GHM 58
GHM59
GHM 60
GHM 61
GHM 63
GHM 64
GHM 66
GHM70
GHM 72
GHM 73
GHM 74
GHM 75
GHM 82
GHM 84
GHM 87
GHM 90
GHM 92
GHM 93
GHM 94
GC 1
GC 14
GC 16
GC 30
GC 32
GC 77
ABS (450 NM)
INDIVIDUAL SCFV-PHAGE CLONES
BINDING OF SCFV PHAGES TO GLP-1R (BLUE)
& A2A (ORANGE)
Abs (GLP1R) Abs (A2a)

67

Figure 4-2. Amino acid sequence identities of GLP-1R-specific 8 scFvs isolated from AID
and Polh-mutagenized human-scFv phage library that has undergone biopanning against
GLP-1R expressed on Sf9 cells and purified GLP-1R whole-membrane protein. Framework
and CDR boundaries are in accordance with the IMGT database. Each sequence is unique from one
another.

The alignment of the isolated scFv sequences to the human germline V gene database by IMGT/
V-QUEST (Table 4-1.) indicated that each scFv contained a unique combination of heavy chain
(VH) and light chain (VL & VK), confirming that the scFv-library we generated represents the
full repertoire of human antibody germline genes. A diverse array of substitution mutations had
been added across all scFvs, with the numbers of mutation varying from 0 to 6 for VH and 7 to
24 for VL, reflecting the effects of B-cell SHM and/or AID and Polh-induced in vitro
mutagenesis during the library construction. Analysis of mutation frequencies for eight GLP-1R
specific-scFv sequences in Table 4-2. corroborates earlier evidence that both in vitro and in vivo

68
SHM introduces mutations preferentially in CDR regions rather than in FW regions. While the %
normalized mutation frequencies for FW ranged in 0.25 % - 1.3 % for VH and 2.2% - 5.4 % for
VL, those of CDR regions ranged in 0 – 1.5 % for VH and 7.5 % - 12.5 % for VL. It was also
found that for both VH and VL, CDR2 was the most mutated region. These results show that Ab
clones after diversification process both in vivo and in vitro have acquired mutations in the Ag-
binding region of CDR, further supporting the applicability of AID and Polh to facilitate the Ab
optimization that mimics the in vivo affinity maturation.

69

Types of substitution mutations & number of mutations in each sub-region
scFv clone # Human IgV Germ
line Identity
FR1 CDR1 FR2 CDR2 FR3 CDR3 Total # mutations
7 IGHV1-18*01 T®G
(1)
- - - -- - 1
IGKV1D-16*01 C®G
A®T
A®C
T®C
G®A
(5)
- G®A
A®G
T®C
C®T
(4)

-
C®T
A®C
G®A
G®A
T®A
T®A
(6)
A®G
(1)
16
25 IGHV1-69*01 - - - - - - 0
IGLV3-21*02 A®G
C®G
(2)
C®G
A®T
A®C
(3)
G®C
C®T
G®C
A®T
C®T
(5)
C®T
G®A
(2)
A®G
(1)
A®T
G®C
G®C
G®C
C®G
(5)
17
58 IGHV1-18*01 T®G
(1)
- - - C®T
C®T
(2)
- 3
IGKV1-5*01 C®T
(1)
- G®A
(1)
G®C
(1)
G®C
A®G
A®T
G®A
T®A
T®C
A®T
C®T
(8)
T®A
A®G
A®T
T®C
(4)
15
66 IGHV1-69*10 C®G
(1)
G®A
(1)
G®A
(1)
G®A
G®A
(2)
A®C
(1)
- 6
IGLV6-57*03 A®G
C®G
(2)
A®G
(1)
A®T
G®C
A®T
(3)
A®G
(1)
C®T
C®T
A®G
T®A
C®T
(5)
- 12
71 IGHV1-69*21 - - - - - - 0
IGKV1-12*01 - G®A
(1)
- - T®A
G®C
A®G
(2)
A®C
C®T
T®C
(3)
7
74
IGVH6-1*01

-

-

-
A®G
(1)
G®C
G®T
G®T
(3)

-

4
IGLV3-19*01 G®C
A®G

(2)
G®A
C®T

(2)
A®G
A®G
A®T
(3)
A®T
A®G
A®G
C®T
(4)
A®G
T®G
A®G
A®C
C®G
A®G
A®G
(7)
A®G
C®T
G®T
G®C
G®C
A®T
(6)
24
75 IGHV1-69*12 A®G
A®G
A®G
(3)
3
IGKV1-12*01 A®G
(1)
A®T
C®G
G®A
(3)
C®A
T®C
C®T
(3)
7
95 IGHV1-18*01 C®T 1
IGKV1-39*01 T®A
C®A
(2)
G®A
G®A
(2)
G®T
G®A
G®A
T®G
A®G
T®C
(6)
A®C
C®T
A®G
(3)
C®G
A®G
A®T
C®A
C®A
(5)
T®A
T®A
T®A
T®C
T®C
(5)
23

70
Table 4-1. Antibody sequence analysis of 8 anti-GLP-1R scFv clones isolated from the AID
and Polh diversified human scFv-phage library. Types and number of substitution mutations
introduced on the germline IgV heavy (VH) and light (VL) chains of eight scFvs were identified by
comparison to IMGT database (NCBI IgBlast).

71

Table 4-2. % mutation frequencies representing the number of mutations per base pair in the designated sub-regions of Ig
variable heavy and light chains of eight GLP-1R specific scFvs isolated from the AID and Polh mutagenized human scFv-
phage library.
The identities of human germline IgV framework and lengths of framework region (FW) and complementarity determining regions (CDR) for each
scFv, as well as the mutations, were identified by NCBI IgBlast. For each clone, mutation frequencies were first normalized for each sub-region by
dividing the total number of mutations in a sub-region by the sub-region length (nt), then multiplied by 100 to calculate the final % mutation
frequency per base pair. The average % mutation frequencies for sub-region was calculated by dividing the sum of % mutation frequencies by the
total number of isolated clones (n = 8).
Heavy chain, VH
scFv
clone #
Germ line
Identity
Normalized % mutation frequency for IgV sub-
regions
FW1 CDR1 FW2 CDR2 FW3 CDR3
7 IGHV1-
18*01
1.4 0 0 0 0 0
25 IGHV1-
69*01
0 0 0 0 0 0
66 IGHV1-
69*10
2.7 4.2 2.0 8.3 0.9 0
58 IGHV1-
18*01
1.3 0 0 0 1.8 0
71 IGHV1-
6-1*01
0 0 0 0 0 0
74 IGHV6-
69*21
0 0 0 3.7 2.6 0
75 IGHV1-
69*21
4 0 0 0 0 0
95 IGHV1-
18*01
1.3 0 0 0 0 0
Average 1.3 0.5 0.25 1.5 0.7 0
Light chain, VL
Germ line
Identity
Normalized % mutation frequency for IgV sub-
regions
FW1 CDR1 FW2 CDR2 FW3 CDR3
IGKV1D-
16*01
6.4 0 7.8 0 5.5 3.7
IGLV3-
21*02
1.4 16.7 9.8 11.1 1.9 15.2
IGLV6-
57*03
2.9 4.2 5.9 11.1 5.3 0
IGKV1-
5*01
1.5 0 2.0 0 8.3 14.8
IGKV1-
12*01
0 5.6 0 0 2.8 11.1
IGLV3-
19*01
2.7 11.1 5.9 44.4 6.5 18.2
IGKV1-
12*01
0 5.6 0 0 2.8 11.1
IGKV1-
39*01
2.6 11.1 11.8 33.3 4.6 18.5
Average 2.2 7.5 5.4 12.5 4.7 11.5
71

72
To verify the specificity of isolated scFv peptides towards GLP-1R, all scFvs (labeled
with a prefix, GHM or scFv, in all figures) were expressed in E. coli CSH50, where they were
exported to the bacterial periplasm via a PelB leader peptide. Following periplasmic extraction
by osmotic shock, the peptides were purified by nickel affinity chromatography and their purities
were analyzed by SDS-PAGE. Each scFv was expressed as a monomer with an apparent
molecular weight of ~37 kD, which corresponds to an expected weight for a scFv (not shown).
Their concentration-dependent, specific binding to GLP-1R (1µg) was verified by peptide-
ELISA. A2a-receptor was used as a non-specific binding control and the background binding
signal at Absorbance 450 nm was subtracted from the read-outs for scFv-GLP-1R binding. The
plots of binding signal (A450 nm) versus. scFv concentration exhibit the signature hyperbolic
curves indicative of equilibrium binding between the scFv to GLP-1R (Figure. 4-3.), suggesting
that all isolated scFv peptides specifically bind to GLP-1R. Most scFvs reach an apparent
saturation at a concentration over 2 mM, except for GHM 58 and 95. GHM 25-scFv displayed
the best binding performance, with its apparent half-maximal binding (EC50) to GLP-1R
occurring at ~100 nM. Further estimations of the binding constants were done using the surface
plasmon resonance assay (SPR).

73

A.
B.
C.
D.
73

74
Figure 4-3. GLP-1R binding by eight scFvs determined by ELISA assay. Different
concentrations of purified scFvs were subjected to binding to 1 µg of GLP-1R immobilized in each well
of a 96-well plate (n=2). Same concentration of recombinant 6 x histidine-tagged adenosine A2a receptor
protein (A2a-His6) was used as a non-specific binding control. The absorbance (A450nm) of each set of
scFv after normalizing with background values are presented as average ± STDEV (n = 2) in the binding
curves of GHM-scFvs; (A) scFv 25 & 71, (B) scFv 58 & 95, (C) scFv 7, 66, 74, and (D) scFv 75. Assays
for all scFvs were performed at the same time. Binding curves are presented in separate panels for
presentation purpose.

Next, we examined whether the purified scFv peptides specifically bind to the N-terminal
extracellular domain (ECD) of GLP-1R. The hallmark structural feature of GPCR receptors is a
large N-terminal extracellular domain (ECD) (Watkins et al., 2012). Numerous studies have
proposed the GLP-1R N-terminal ECD as the initial binding site for known GLP-1R agonists
(GLP-1 and Exendin 4), and plays a crucial role in the receptor activation (Heller et al., 1996,
Wilmen et al., 1996; Al-Sabah and Donnelly, 2003; Runge et al., 2008; Zhang et al., 2010).
Hence in this project, a purified peptide corresponding to the isolated N-terminal domain of
human GLP-1R (aa 24-145, LSBio) was used in SPR studies instead of GLP-1R whole
membrane, which is more difficult to immobilize on SPR sensor chip due to its requirement for
solubilization in detergent and notorious instability once extracted from lipid membranes
(Sarramegn V. et al. 2006.). Further, since N-glycosylation in GLP-1R is essential in the correct
trafficking and processing of GLP-1R to reach the plasma membrane in vivo (Huang et al.,
2010), we chose the GLP-1R ECD expressed in yeast over E. coli, so that the glycosylation on
the ECD is preserved. All six scFvs tested (GHM 7, 25, 58, 66, 74, 75) displayed a monovalent
interaction with GLP-1R-ECD and fit the Langmuir-equation based 1:1 binding model (Fisher
and Fivash, 1994) (Fig.4-4, Table 4-3.) The SPR results demonstrated that the isolated scFvs
bind specifically to GLP-1R N-terminal ECD with equilibrium dissociation constants (KD)
ranging in a single-digit nanomolar to double-digit nanomolar range (8~72 nM, Table 4-3.),
identifying the region as an epitope for the scFv binding.

75

A.

B.

C.

D.

Resonance (RU)
Resonance (RU)
Time (seconds)
Time (seconds)

76

Figure 4-4. SPR sensorgrams showing relative binding of purified scFv-His6 peptides to N-
terminal GLP-1R extracellular domain (aa 24-145, LSBio) over time (seconds). Each panel
represents binding between GLP-1R-ECD and (A) scFv-7, (B) scFv-25, (C) scFv 58, (D)
scFv-66, (E) scFv-74, and (F) scFv-75 from a single run.
All scFv peptides were purified by periplasmic extraction and nickel column chromatography. SPR
measurements were determined by injecting 1-500 nM of each scFv diluted in HBS-EP (10mM HEPES,
0.15M NaCl, 3 mM EDTA, 0.005% P20) running buffer at a flow rate of 30μl/min onto a sensor chip
surface immobilized with 150 RU of GLP1R-ECD, followed by dissociation for 600s. 4mM NaOH was
used to regenerate the surface. All sensorgram data were fitted to 1:1 binding model to calculate the
binding kinetic constants by Biacore T200 evaluation software after referencing with a blank cell. The
calculated results are presented in table 4-3.

scFv KD (nM) kon (M
-1
S
-1
)E
+6
koff (S
-1
)E
-2

7 22 0.19 0.41
25 8 2.80 2.18
58 15 1.62 2.45
66 43 1.07 4.67
74 72 0.27 1.98
75 45 0.95 4.27

Table 4-3. Kinetic constants (kon & koff) and equilibrium dissociation constants (KD) for
binding interactions between 6 GLP1R-specific scFv peptides and GLP1R-ECD
determined by Surface Plasmon Resonance.
E.

F.

Resonance (RU)
Time (seconds)

77
2. Further in vitro affinity maturation of eight scFvs using AID and Polh generated Ab
variants with enhanced affinity to GLP-1R
In order to accommodate the iterative nature of B-cell affinity maturation into our Ab
optimization strategy, we subjected the 8 GLP-1R-specific scFv clones to further AID and Polh-
in vitro affinity maturation. In doing so, we applied a modified version of the gap DNA method
(Chapter 2). First, the phage vector f3TR1-dsDNAs were linearized with BglI restriction enzyme
and denatured by heat, followed by annealing with circular single-stranded DNAs (ssDNAs)
extracted from the 8 scFv-phage clones, as illustrated in Figure 4-5. A. Single-stranded gapped
regions of the resultant gapDNA constructs would expose a combination of ~800 bp scFv genes
of 8 clones to be targeted by AID and Polh. After treatment with AID at 3 different
concentrations (0.1 mM, 0.5mM, and 1 mM) and 7 time points (30 s – 1h) at 37C, followed by
Polh treatment for 2h, the mutagenized products were pooled together and transformed into E.
coli to obtain a sub-library of mutagenized scFv-phages.
Another method of treating DNA substrate using AID and Polh was adopted to avoid
some of the pitfalls involved in the gap DNA method (Chapter 2) and in the use of f3TR1 M13
phage vector (Chapter 3). First, it was observed that the gapDNA synthesis resulted in low
substrate DNA yields compared to the input amount of ssDNA and linearized dsDNA,
presumably due to the low annealing efficiency and the loss of DNA during the clean-up
processes. The large size of M13 phage vector f3TR1 (8958 bp) also gave rise to low
transformation efficiency, contributing to reduced diversity of the phage-Ab sub-library which
was generated by amplifying the pool of E. coli transformants. Additionally, the fact that f3TR1
is a polyvalent display vector, which displays the Ab fragments fused to multiple (up to 5) copies
of pIII coat proteins can lead to the avidity effect during Ab-phage selection. The avidity effect

78
amplifies the overall binding of the polyvalent Ab-displaying phage to the target antigen, and
consequently may result in the isolation of low affinity Ab-phage binders.
Hence we tested a new method for treating the substrate DNA with AID and Polh using a
smaller sized phagemid vector, pADL20c (3930 bp), which was used for the generation of mini-
synthetic scFv library in Chapter 2. pADL20c is a type 3+3 monovalent phage display vector
which displays the antibody fragment fused to one of the five P3 coat proteins at the tip of the
phage surface. Its smaller size increases the transformation efficiency, which in turn increases
the size of the packaged Ab-phage library. The mono-valency of the displayed Ab fragment also
ensures that the selected Ab-phage after affinity selection and screening procedures are truly
high-affinity binders. This method is illustrated in Figure 4-5.B. Here, ssDNAs were extracted
from the combined pool of 8 f3TR1 scFv-phages. An equal volume of three different
concentrations (1µM, 5 µM, 10 µM) of purified AID was then added at different time intervals
(30 s – 1h) to the ssDNA to diversify the AID deamination patterns on the substrate DNA. After
cleaning-up of AID-treated f3TR1-scFv ssDNA, reverse primers specifically designed for
cloning the scFv gene from f3TR1 vector into pADL20c vector were annealed to the ssDNAs to
allow pol h extension synthesis on the free 3’-OH of the annealed primers. In the presence of
dNTPs, Polh polymerization converted the single-stranded scFv region to double stranded DNA.
AID and Polh-treated scFvs were then amplified via PCR using forward and reverse primers
designed for pADL20c vector cloning. PCR products were pooled together and digested with
BglI restriction enzyme, followed by ligation into pADL20c vector. The resultant AID and Polh
treated scFv-pADL20c recombinant plasmids were then transformed into TG.1 E.coli cells to
generate the secondary in vitro affinity-maturated, pADL20c-scFv monovalent phage library.
Since pADL20c phagemid lacks the viral genes required for synthesizing peptides that are

79
needed for phage reproduction and coat packaging, helper phage carrying the necessary viral
genes was added to assist the virion production. The size of the resultant affinity-maturated,
pADL20c-scFv phage sub-library was 6 x 10
7
independent clones, which was approximately 10
times larger than that of the f3TR1-scFv phage sub-library (~5 x 10
6
).

Figure 4-5. (A) Schematic diagram describing steps involved in “gapDNA” method
BglI cut f3TR1 phage vector is denatured by heat and annealed to f3TR1-scFv phage ssDNA. The
resultant gapDNA with the scFv exposed as a single-stranded region is targeted by AID deamination,
followed by Polh’s gap-filling synthesis in the presence of excess dNTPs.

(B) Schematic diagram describing steps in ssDNA treatment method
Phage-scFv ssDNA is extracted from scFv-phage pool and incubated with AID. AID-mutagenized
ssNDA is then annealed to a reverse primer to provide 3’-OH end for Pol h to extend on. AID and Pol h
treated scFv region is amplified with PCR. PCR products are cut with BglI and ligated into pADL20c
phagemid vector to be transformed into E. coli.

A.
B.

80
The sub-library (labelled GHM’; Fig 4-6. onwards) was used in 3 rounds of panning on
the purified GLP-1R protein, in parallel with the original scFv-phage library (labelled GHM)
from which the 8 clones were derived. The pH of wash buffer was gradually decreased (7.4 ®
6.5 ® 5) in subsequent panning round to exert stronger selection pressure, thus increasing the
likelihood of obtaining strong binders in the final panning output. The percentage recovery
increased with each round of selection for the phage sub-library of 8 clones, despite the
increasingly harsher condition for successive panning (Fig. 4-6), showing that target-specific
phages are increasingly dominating the population of eluted phages. scFv-phages that underwent
affinity maturation twice until second panning show lower % recovery compared to scFv-phages
that were affinity matured one time during the library construction (Chapter 3), indicating that a
relatively lower proportion of specific binders were recovered from the input phages. This is
probably due to the lower diversity of input phages, since this library was amplified from the
pool of only eight scFv clones, despite having been further diversified. After 3rd panning
onward, the % recovery of the affinity-matured sub-library was larger compared to that of the
original library indicating that the eluted phage population is dominated by more selectively
binding phages.

The final 3
rd
panning output was used to infect E. coli at an exponentially growing phase,
and infected bacterial cells were plated on the selective medium. 96 individual clones were
randomly picked for screening and initial selection of high affinity phage-scFvs. The selected
phage-Abs were analyzed for their GLP1-R-specific binding again via phage-ELISA (results not
shown), and their sequences were characterized by Sanger sequencing (Retrogen). One scFv
clone (scFv 33’) displaying high affinity was isolated, and it was revealed that this clone

81
originated from scFv-25, one of the eight clones selected prior to further in vitro affinity
maturation. (Fig. 4-7) scFv-33’ contains T220M and S243C amino acid replacements (Fig 4-8),
which are caused by C®T mutation in the light chain VL- FW3 region done by AID, and C®G
in VL- CDR3 region by Polh (Fig 4-7).

Figure 4-6. Biopanning enriches target-specific scFv-phage population
Enrichment of 1 x affinity matured (GHM, yellow) and 2x affinity matured (GHM’, blue) human f3TR1-
scFv-phage population after successive biopanning round with decreasing wash pH, pH 7.4 → 6.5 → 5,
for successive panning round is represented. The percentage recovery of phage input was determined by
dividing the number of phage eluted after panning by the number of phage input used for each panning.
The percentage recovery increased with each round of selection for 2 x affinity matured scFv phages,
despite the increasingly harsher condition for panning, showing that target-specific phages are
increasingly dominating the population of eluted phages regardless of panning conditions. 2 x affinity
matured scFv-phages until 2nd panning show lower % recovery compared to 1 x affinity matured scFv-
phages, indicating that a relatively lower proportion of specific binders were recovered among the input
phages (perhaps due to lower diversity of input phages i.e 8 individual scFv clones were affinity matured
and amplified, resulting in lower diversity of scFvs compared to 1st affinity matured phage-library). After
3rd panning onward, the % recovery of 2 x affinity matured scFv-phages is larger compared to that of 1 x
affinity matured scFv-phages, indicating that now the eluted phage population is dominated by more
selectively binding phages.

82
-

Fig 4-7. DNA sequence alignment of GLP-1R-specific scFv 25 (original scFv selected from a
mutagenized library) and scFv 33’ (affinity matured variant of 25). Framework (FW) and
Complementarity Determining Region (CDR) boundaries are identified by comparison to human
germline Ig genes in IMGT database and represented by arrows. The alignment reveals that a C®T
mutation has been added by AID in the VL FW 3 region, and a C®G mutation has been added by Polh in
VL CDR 3 region.

Fig 4-8. Alignment of GLP1R-specific scFv amino acid sequences. scFv 33’ clones has a
T220M mutation in the FW 3 region of VL, and S243C in CDR3.

VH-FW1 VH-CDR1
VH-FW2

VH-FW2
VH-CDR2

VH-FW3

VH-FW3

VH-CDR3

VL-FW1

Linker
Linker
VL-CDR1 VL-FW2

VL-CDR2 VL-FW3

VL-FW3

VL-CDR3
VL-FW3

VL-CDR3

83
The binding capacities of scFv 25 and its affinity-matured variant, scFv 33’ were
compared via phage-ELISA (Fig. 4-9.), peptide-ELISA (Fig. 4-10), and SPR (Fig. 4-11). In
phage-ELISA, 10
8
TU of f3TR1-scFv phages of each clone were added into a set of four wells
coated with a constant concentration of either GLP-1R or the background binding control, A2a
receptor. Background subtracted absorbance at 450 nm upon immunodetection with anti-M13
pIII detection antibody (Santa Cruz) revealed that GHM33’-scFv phage clone had approximately
two-fold higher binding (A450 nm~ 0.6) compared to GHM25-scFv phage (A450 nm~ 0.3, Fig.
4-9).

Fig 4-9. Phage-ELISA based binding of original (GHM25) and affinity-matured (GHM33’)
human-scFv to purified GLP1-R whole membrane protein. ELISA based binding of 2 clones of
human scFv-phages to purified GLP1-R coated on a microtiter plate reveals higher target affinity by the
affinity matured clones, GHM33’, compared to the original clones, GHM25. Human GLP1-R was
immobilized on each well of a Nunc 96-well Maxisorp plate and incubated with 10
8
TU of GHM33’ &
GHM25-scFv phages. Binding of scFv-phages was followed by immunodetection of the phage coat
protein pIII by anti-M13 pIII mAb conjugated with HRP (Sinobiological). Assay was repeated 4 times,
and background binding of scFv-phage to A2a-receptor protein was also repeated 4 times (n = 4) and
subtracted from the Abs(450nm) for GLP-1R binding. Error bars represent the standard deviation (SD) of
4 trials.

84
In order to characterize two scFvs as peptides, we expressed the scFvs in fusion with C-
terminal C-myc and Histidine-tags. After nickel column purification, the histidine tag was
removed by protease treatment. Binding of scFv 25 and scFv’ 33 -peptides to GLP1R were
demonstrated by peptide-ELISA using anti-Cmyc antibody conjugated with HRP as the detection
Ab. Purified GLP-1R-His6 and A2a-His6 were immobilized on 96 well microtiter plate and were
incubated with increasing concentrations of C-myc tagged-scFvs. The half-maximal effective
concentrations (EC50) estimated from the ELISA results showed that the purified scFv-33’
protein binds to GLP-1R with higher affinity than its parental clone, scFv-25; its estimated EC50
of 100 nM was ~3-fold lower than the EC50 of scFv-25, which was ~317 nM. (Fig 4-10). Both
scFvs did not bind to the A2a negative control nor the milk-PBS blocked polystyrene wells.

Since the original scFv-25 peptide exhibited specific binding to the N-terminal GLP-1R
extracellular domain (ECD, aa 24-145)(Table. 4-3) in the earlier SPR assay, the same epitope
was used to compare the binding properties of the original and affinity-matured scFvs via SPR
under the same conditions as previously described (Fig. 4-11). Results showed that scFv-33' had
a slightly enhanced affinity towards GLP-1R-ECD, with an average equilibrium dissociation
constant (KD) of 14 compared to scFv-25, with an average KD of 16 (Table 4-4.). However, this
difference (~2) is within the error range of each measurement, indicating that both clones have
similar binding affinities to the GLP-1R ECD.

85

Figure 4-10. Concentration-dependent binding of original and affinity-matured scFv-
peptides to isolated GLP1-R measured by ELISA. GLP1-R solubilized in detergent n-Dodecyl-B-
D-Maltoside (DDM) was immobilized on Nunc maxisorp plate and incubated with increasing
concentration (1 nM, 10 nM, 100 nM, 500 nM, 1µM, 5 µM, 10 µM) of each C-myc tagged scFvs, scFv-
25 (blue) and scFv-33’ (green). Binding of scFv was followed by immunodetection of the C-myc tag
using anti-Cmyc antibody conjugated with HRP. Error bars represent the standard deviation (SD) of
duplicates. Apparent EC50 was estimated by first extrapolating the log [scFv] at y = half-maximal
binding signal (OD450 nm, 0.55 for scFv33’ and 0.45 for scFv 25), taking the anti-log of log half-
maximal binding scFv concentration. Apparent EC50 (nM) of scFv 25 occurs at scFv concentration of
~320 nM (anti-log 2.5 = 316), while that of scFv 33’ occurs at 100 nM (anti-log 2).

86

Figure 4-11. SPR sensorgrams showing relative binding of purified scFv-His6 peptides to
N-terminal GLP-1R extracellular domain (aa 24-145, expressed in yeast, LSBio) from trial
2. Left panel represents concentration-dependent binding (0-500 nM) of scFv 25 isolated from human
mutagenized scFv-phage library (1 x affinity maturation), while right panel shows that of scFv 33’,
isolated from the scFv-phage sub-library that has been amplified from the in vitro affinity-maturated 8
scFvs (2 x affinity maturations). Each set of scFv dilutions was flowed over GLP-1R ECD immobilized
on sensor chip at a flow rate of 30ml/min and then allowed to dissociate for 600 seconds. 1:1 curve fitting
and calculation of kinetic constants were done by Biacore T200 evaluation software after referencing with
a blank cell.

Clone kon (M
-1
S
-1
)E
+6
± STDEV koff (S
-1
)E
-2
± STDEV KD (nM) ± STDEV
scFv 25 1.45 ± 0.48 2.24 ± 0.14 16.2 ± 4.49
scFv 33’ 1.81 ± 0.85 2.43 ± 0.46 14.4 ± 4.19

Table 4-4. Kinetic and equilibrium dissociation constants of scFv 25 and its affinity-
matured scFv 33’ variant to purified glucagon-like peptide-1 receptor N-terminal
extracellular domain (aa 24-145), determined by Surface Plasmon Resonance (n = 2)
Resonance (RU)
Time (seconds)

87
In summary, the affinity-matured scFv-33’ demonstrated a 2~3 fold improved affinity
towards GLP-1R whole membrane protein in both phage and peptide form compared to its
original clones, scFv-25. Although SPR assays showed similar affinities between scFv-33 and
scFv-25 for the N-terminal extracellular domain of GLP-1R, one round of in vitro affinity
maturation of anti-GLP-1R scFv clearly enhanced target binding affinity.

4.4 Discussion
To develop scFv-Abs specific to GLP-1R, as well as to assess the effects of in vitro
affinity maturation, AID and Polh-mutagenized human naïve-scFv phage library has been used
for biopanning against purified GLP-1R whole membrane protein. 8 unique scFv clones were
successfully isolated, and all clones exhibited specific binding toward GLP-1R and N-terminal
GLP-1R extracellular domain (amino acid 24 – 145). Sequence characterization of these scFvs
revealed the signature SHM bias in the CDR region, suggesting that IgV diversification has been
carried out in vivo and/or in vitro before and/or during library construction. Titration-ELISA
revealed that all 8 scFv peptides possess specific binding toward GLP-1R, and not to the off-
target GPCR protein, A2a receptor. SPR experiments indicated that all tested scFvs (6 out of 8)
possess dissociation constants in the single-digit nanomolar to double-digit nanomolar range for
binding to GLP-1R extracellular domain, identifying the ECD as the epitope targeted by GLP-1R
specific scFvs. Since we did not know whether these binders originated from the naïve scFv
library prior to the in vitro affinity maturation or from the mutagenized library after the in vitro
affinity maturation, one more round of affinity maturation was performed on the 8 identified
scFv clones. Construction and panning of the resultant sub-library of mutagenized 8 scFvs
yielded one clone that is an affinity-matured variant of one of the 8 scFvs, scFv-25.

88
To compare the GLP-1R binding affinities of the original and affinity-matured clones, a
number of binding assays were performed. The results of phage- and peptide-ELISA using GLP-
1R whole membrane protein as the target showed ~2-fold and ~3-fold improvement in binding
affinity in the affinity-matured scFv compared to its parental scFv. However, these were not
reflected in the SPR data, which showed similar KD values for both scFvs (with a difference
within the error range) for binding to GLP-1R-ECD. This phenomenon where a positive phage
ELISA result does not correlate with SPR results has been observed in other groups (R.W Gene
et al., 2015) as well as in our lab. It is not clear what causes this, but it can be presumed that the
change in display format of scFv from phage-display to a purified protein may affect the scFv
folding, consequently altering the binding capacity. Also, different Ags were used for the binding
measurement. The reconstitution and immobilization methods for these Ags are different, with
the GLP-1R whole protein requiring the addition of detergent n-Dodecyl-B-D-Maltoside (DDM)
for reconstitution, which may influence the Ag immobilization and Ag-Ab interaction in binding
assays. The signal amplification during ELISA detection and inaccurate estimation of phage
titers may also lead to over-amplified signal reads that may misrepresent the actual binding.
Nonetheless, all binding assays showed enhanced affinity by the affinity-matured scFv (scFv-
33’) toward GLP-1R. This suggests that key elements of adaptive immunity, which involves the
progressive improvement in Abs via further diversification of IgV and selection of specific
binders, have been successfully reconstituted in our in vitro platform through iterative rounds of
AID and Polh-mediated affinity maturation and biopanning.

In the germinal center, multiple rounds of Ab diversification and clonal selection are
carried out to generate Abs with higher affinities after an encounter with an Ag. Since only one

89
round of affinity maturation was performed on the isolated clones, we expect even greater Ag-
binding affinity from these clones with further rounds of affinity maturation and selection by
biopanning. However, we also speculate that affinity maturation in vitro may have less effects on
clones with an already high affinity toward the target antigen, as was the case for the scFv-25. It
has been reported by Poulsen and colleagues that Ag-driven affinity maturation and repertoire
diversification have biological limits, as evidenced by little changes in the distribution of
mutations and binding rate constants in Abs in individuals exposed to a target Ag multiple time
(3 vaccinations) (Poulsen et al., 2011). Also, we do not know the effects of the in vitro
diversification that was carried out on the human-scFv repertoire before the exposure to GLP-1R
(Chapter 3), except that mutations had been introduced in vitro. Therefore, in the next chapter,
we conducted parallel analysis on the naïve Ab library and AID and Polh-treated, mutagenized
Ab library to address the effect of in vitro affinity maturation on Ab repertoire prior to antigen
exposure.

90
Chapter 5. Application of AID and Polh affinity maturation in vitro to isolate Abs
targeting specific antigen: Target 2 - Artemin
5.1 Introduction (Artemin)
Cold hypersensitivity is clinically manifested as pain in response to noxious (hyperalgesia) or
innocuous (allodynia) stimuli (Baustista et al., 2009). Cold allodynia is a frequent symptom in
patients suffering from nerve injury, fibromyalgia, multiple sclerosis, or chemotherapeutic
polyneuropathies. (Lippoldt et al., 2016.) Previous studies done in the McKemy lab at USC
Neuroscience revealed that glial cell line-derived neurotrophic factor family of receptors
(GFRa3), were co-expressed in a subset of TRPM8 (transient receptor potential melastatin 8)
sensory neurons, which act as the principal cold sensor in the peripheral nervous system
(Baustista et al., 2007; Knowlton et al., 2010) and are implicated in pathological cold pain. Their
recent findings showed that cold allodynic response after injuries was abolished in transgenic
mice deficient in GFRa3, indicating that this receptor was crucial in the development of cold
pain (Lippoldt et al., 2016). In addition, in vivo evidence indicated that the ligand for GFRa3,
glial cell line-derived neurotrophic factor, artemin, could induce cold responses in a TRPM8-
dependent manner (Lippoldt et al., 2013). The role of artemin as the mediator of cold pain was
further substantiated when the injection of an artemin-neutralizing antibody attenuated the
established cold allodynia in mice, suggesting that artemin inhibition abolishes the artemin-
GFRa3 interaction pathway to suppress cold pain (Lippoldt et al., 2016). Thus we hypothesized
that developing monoclonal antibodies targeting artemin may provide a novel therapeutic
solution to alleviating cold hypersensitivity. In collaboration with the McKemy lab, we
developed a llama monoclonal VHH nanobody that binds specifically to artemin, and tested its
inhibitory effects on the cold pain signaling via mice behavioral experiments.

91
5.2 Materials & Methods

Antigens. Purified recombinant mouse artemin protein (Ala112-Gly224 fragment, 1085 AR) was
purchased as lyophilized powder from R&D Systems, Inc., was reconstituted at 100 mg/mL in 4
mM HCl and stored at -20°C.

Selection of phage-VHH library by Bio-panning
Biopanning of llama-VHH phages specific for mouse artemin was performed on MaxiSorp
Nunc-Immunotubes. 500 nM of mouse-artemin in 0.5 ml of 100 mM sodium bicarbonate buffer
(pH 9.6) was coated on the tube surface overnight at 4°C. After blocking the tube with 2% (w/v)
skimmed milk powder in PBS (2% MPBS) for 2 h, an aliquot of llama f3TR1-VHH phage-
library containing 1 x 10
12
TU of phages was diluted in 2% MPBS to a total volume of 0.5 mL,
then added to the tube in the 1
st
round of panning. For the subsequent rounds of panning, less
phages (2 x 10
11
TU) were used. After 2 h incubation at room temperature, non-bound phages
were removed by washing 10 times with PBS containing 0.1% Tween 20 (PBS-T), followed by
10 times washing with PBS. The bound phages were eluted by incubation with 1 ml of 10 μg/ml
trypsin in PBS for 30 min at RT. Phage titration and amplification were carried out according to
the previous protocol (Chapter 4, Materials & Methods).

Screening for antigen-specific clones by phage ELISA. Individual K91BK colonies on titration
plates from the 3rd artemin panning were pick into 96-deep well plates containing 1.5 ml of 2x
YT + 20 µg/ml tetracycline and grown overnight in a shaker at 37 oC to produce VHH-phage
particles. Phages from 1 ml of supernatant were precipitated with 150 µl of the PEG/NaCl
solution and resuspended in 200 µl of PBS.

92
Each well of Nunc Maxisorp 96-well microplates was coated with 100 µl of 1.2 μg/ml of mouse-
artemin (100 nM per well) in 0.1M carbonate buffer (pH 9.6). After overnight incubation at 4 °C,
plates were blocked with 2% MPBS for 1.5 hour at RT followed by 3 washes with PBS-T and 3
washes with PBS. All washings were performed on AquaMax 2000 plate washer (Molecular
Devices, LLC). The VHH-phage supernatant was diluted 1:2 in 4% MPBS before adding 100 µl
into each well, and incubated for 1.5 hour at RT. The plates were washed three times with PBS-
T, followed by three times with PBS, and incubated with 100 µl of a 1: 4000 dilution of anti-
M13-HRP (Sino Biological US, Inc., Wayne, PA) in 2% MPBS for 1 h at RT. Plates were
washed 4 times with PBS-T and 4 times with PBS. TMB substrate solution 100 µl (Thermo
Fisher Scientific) was added to each well (for 2 min to 10 min). After adding 100 µl of 2 M
sulfuric acid stop solution, the absorbance was read at 450 nm, using SpectraMax iD5 microplate
reader (Molecular Devices, LLC).

Expression and purification of soluble VHH nanobodies. Coding sequences for artemin-
specific VHHs were subcloned into the BglI sites of pADL-20c phagemid vector. The constructs
were transformed into E. coli CSH50 cells and Abs were expressed as His6-tagged soluble
proteins in the periplasm. Cells were grown at 37 °C in LB + 0.2% glucose + Ampicillin (100
µg/ml). When the culture OD600 reaches ~0.7, IPTG (1 mM) was added to induce VHH
expression at 30 °C overnight. Cells from 1 L culture were harvested by centrifugation,
resuspended in 400 ml ice-cold wash buffer (30 mM Tris pH 8.0 and 20% sucrose) and EDTA
was added to 1 mM. After incubation at 4 °C for 10 min, cells were centrifuged at 8000 x g for
20 min, resuspended in 400 ml of ice cold 5 mM MgSO4 and incubated for 10 min at 4 °C. The
supernatant containing a soluble VHH protein was collected by centrifugation at 8000 x g for 20

93
min. His6-tagged VHH peptides were purified via nickel affinity chromatography using 2 ml of
Ni-NTA resin (Qiagen) and PBS as a wash buffer. The VHH was eluted in PBS containing 300
mM Imidazole, then further purified by gel filtration using Superdex 75 column (GE
Healthcare). Proteins were concentrated using Amicon Ultra-15 10 kDa centrifugal filter unit
(Millipore) and stored in PBS at -70 °C.

Surface plasmon resonance (SPR). Binding of purified VHH nanobodies to mouse artemin
(Ala112-Gly224, R&D Systems) was determined using Biacore T100 instrument (GE
Healthcare). Artemin was diluted to 10 µg/mL in 10 mM sodium acetate (pH 4.5) and
immobilized on a CM5 Series S sensor chip (GE Healthcare) at 150 RU (response unit) using
amine coupling chemistry according to the manufacturer’s protocol (GE Healthcare). Purified
VHHs at concentrations ranging from 1-500 nM in flow buffer (PBS-Tween, 0.005%, pH 7.4)
was injected onto the flow cells at a flow rate of 30 ml/min for 120 seconds, and allowed to
dissociate for 600 seconds. The sensor chip surface was regenerated using 6 mM NaOH. Kinetic
constants for binding interaction were determined by fitting the sensorgrams with 1:1 binding
model using Biacore T100 evaluation software, version 2.0 (GE healthcare).

Cold plantar assay on mice (Performed in the McKemy Lab, USC Neurosciences). The cold
sensitivity of the hind paws of mice injected with purified llama-VHH and Mab1085 positive
control (R&D Systems) was measured in the McKemy lab (Department of Biological Sciences,
Neuroscience, University of Southern California, CA) using the previously described protocol
(Knowlton et al., 2013; Lippoldt et al., 2016). Mice were allowed to acclimate in Plexiglas
chambers for 2 hours prior to cold plantar testing performed at room temperature or 30 min prior

94
to cold plantar testing performed at a glass surface of 30°C. A dry ice pellet was then applied to
the hind paw of the mice through a glass at a thickness of 6mm. Hind paw withdrawal latencies
(in seconds) were recorded for a total of 3 trials per paw for each time point.

5.3 Results
1. Four anti-Artemin VHH nanobodies were isolated by biopanning of llama VHH-phage
display libraries and characterized

To further evaluate the effects of AID and Polh-mediated in vitro diversification that was carried
out on the naïve Ab repertoire before the antigen exposure, we performed parallel analysis on the
individual clones derived from llama naïve-VHH library and AID and Polh-treated,
mutagenized-VHH library.
E. coli derived mouse Artemin peptide (Ala112-Gly224, 12 kD) was purchased from R&D
Systems and used as a target Ag for biopanning. Naïve llama-VHH and AID and Polh-treated
llama-mutagenized VHH-phage libraries (Chapter 3) were used in parallel to select VHH
nanobodies specific to a purified mouse artemin. To enrich potent Artemin binders, each llama-
VHH library was subjected to three rounds of panning against purified mouse artemin. After 3
rd

panning, the phage population of each library had been enriched with artemin-binding VHHs,
shown by the high percent phage recovery of 3
rd
panning output compared to those of previous
panning rounds (Fig 5-1).

95

Figure 5-1. Enrichment of artemin-specific f3TR1-VHH-phage population for successive
biopanning. The percentage of input llama f3TR1-VHH phages recovered increased with each round of
selection, showing that phages are selectively enriched after successive panning. The Wash condition
remained constant throughout 3 panning rounds (PBS, pH 7.4). The percentage recovery of phage input
was determined by comparing the number of phages eluted after panning in transducing units (TU) and
the number of phage input in TU used for each panning round. Phage number was calculated by making a
100 fold serial dilutions of M13 phage pool and infecting exponentially growing bacterial culture and
plating an aliquot of infected culture on an agar plate supplemented with selective antibiotic
(tetracycline). Panning of the same phage input against non-coated immunotubes were done in parallel as
the negative control and their % phage recoveries are indicated alongside with llama-mutagenized (LM)
and naïve (LU) libraries.

Phages eluted from the final panning were used to infect E.coli at exponentially growing
phase, and infected bacterial cells were plated on the selective agar plate supplemented with
tetracycline. 96 individual clones were randomly picked for screening by ELISA, using
recombinant GLP-1R-His6 (~50 kD) as a negative control antigen for background binding. The
ratio of phage-artemin binding signals at 450 nm absorbance (A450 nm) compared to
background binding was calculated for each clone, and positive binders were defined as having
0
0.5
1
1.5
2
2.5
3
1
LM
LU
LM control
LU control
2
LM
LU
LM control
LU control
3
LM
LU
LM control
LU control
% RECOVERY
PANNING ROUND
LM : LLAMA-MUTAGENIZED VHH LIBRARY
LU: LLAMA-UNTREATED (NAIVE) VHH LIBRARY
% Llama-VHH phage recovery after each panning aginst
mouse artemin protein
(Phage Input/Output ratio)

96
ratios greater than 3. 20 VHH-phage clones, 10 from llama-naive and 10 from mutagenized
libraries, were picked from the phage-ELISA as part of the initial selection. To select for the
highest affinity binders, the artemin-binding capacities of these VHH-phages were compared by
ELISA again (Fig 5-2); equal titer (10
9
TU) of each phage was added in duplicates to artemin
and GLP-1R-coated wells. Background subtracted A450 nm for 20 clones were compared to
A450 nm of anti-mouse artemin mAb (Mab1085, R&D Systems), identifying 7 high-affinity
clones with A450 nm comparable (A450 nm > 1.4) to that of the positive control (A450 nm =
1.61): LM 22, 25, 41, 52 from the llama-mutagenized library, and LU 9, 68, and 75 from llama
untreated/naïve library.

Figure 5-2. Bar graph showing the absorbance at 450 nm of twenty individual monoclonal
VHH - phage binding to purified mouse artemin (R&D system) in phage-ELISA. 1 x 10
9
TU
of each VHH-phage was used for binding against 100 ng of artemin coated in each well of a Nunc 96 well
Maxisorp. Detection done using anti-M13 pIII mAb conjugated with HRP (Sinobiological). Absorbances
were normalized by subtracting the A450 nm of background binding of phage-VHH to GLP-1R. Positive
control (MAB1085, anti-mouse artemin mAb; R&D system) is indicated in orange. Error bars indicate the
standard deviation for 2 trials (n = 2).

0
0.5
1
1.5
2
2.5
3
LM 16
LM 21
LM 22
LM 23
LM 25
LM 37
LM 41
LM 47
LM 50
LM 52
LU 5
LU 8
LU 9
LU 19
LU 35
LU 66
LU 67
LU 68
LU 70
LU 75
Anti-artemin mAb
OD 450 nm
VHH-phage clone #
LM : Llama mutagenized VHH library
LU : Llama naive VHH library

97
Sequence characterization of 20 VHH clones identified 2 pairs of related VHH
nanobodies from mutagenized and naïve VHH libraries. Two clones, LM41 and LM52, from the
mutagenized library likely represent AID and Polh-induced mutants of LU68 and LU5 clones
from the naïve VHH library. LU5 from the naïve library and LM52 from the mutagenized library
had identical DNA sequences except that LM52 contains 2 G®A, 1 C®T mutations, indicative
of AID deaminations, and T®C and T®A mutations, presumed to have arisen from Polh
actions (Fig 5-3. A). All deaminations were 5~7 bases apart, which reflected the processive
action of AID (Pham et al., 2003; Bransteitter et al., 2004). The resulting changes in amino acid
compositions on LM52 were Y62H, A78T, T80M, and N89K (Fig 5-4A), all in the framework 3
(FW3) region of VHH. Although mutations in the FW domain are strongly suppressed in B-cells
(Ohm-Laursen et al., 2007), it has been observed that in vitro, AID deamination does occur more
frequently on FW domains than it does in vivo (Pham et al., 2019). This difference may be
because the U:G mispairs arising from the AID actions in vivo are processed by BER and MMR
machinery in B-cells, reducing the mutational levels in the FW regions that are important in
maintaining the overall Ab structure, whereas in vitro, no such repair machineries are recruited.
Another pair, LU68 and LM41, share identical sequences except at ~150 bp downstream of the
start codon, where two T® C substitutions have been added 9 bases apart from each other in
WA (W = A/T) sites, which correspond to the Polh hot motifs (Fig 5-3.B). One mutation was
found in CDR2 region while the other was in FW3. The resulting change in amino acid was
Y62H on LM41 in the FW3 region (Fig 5-4.B).

98
(A)

(B)

Figure 5-3. (A) DNA sequence alignment of LU5-VHH (top) and LM52-VHH (bottom). LU5
is isolated from the naïve llama-VHH library, while LM52 is isolated from the llama-mutagenized VHH
library that has been treated with AID and Polh in vitro(Chapter 3).
(B) DNA sequence alignment of LU68-VHH (top) and LM41-VHH (bottom). LU68 and
LM41 are isolated from the naïve llama-VHH and llama-mutagenized VHH library, respectively.
FW and CDR regions are identified using IMGT definitions and indicated with arrows. Locations of
mutations are indicated with blue letters in the consensus sequence and stars on the top of the naïve VHH
sequences.

(A)

(B)

Figure 5-4. Amino acid sequence alignment of (A) LU5-VHH and LM52-VHH and (B)
LU68-VHH and LM41-VHH. FW and CDR regions are identified using IMGT definitions and
indicated with arrows. The location of amino acid substitution is noted by a blue letter in the consensus
and a star on the top of naïve VHH sequence.
VH-FW3
VH-FW1 VH-CDR1
VH-FW2
VH-FW2 VH-CDR2 VH-FW3
VH-CDR3 VH-FW3
VH-CDR2
VH-FW1 VH-CDR1
VH-FW2
VH-FW2 VH-CDR2 VH-FW3
VH-CDR3 VH-FW3
VH-FW3
VH-FW4
VH-FW4

99
The 4 VHHs isolated from both naïve and mutagenized libraries were expressed as
monomers and purified by periplasmic extraction. Their binding to artemin was characterized by
surface plasmons resonance (SPR). All four VHHs displayed a monovalent interaction with
mouse artemin and fit the Langmuir-equation based 1:1 binding model (Fig. 5-5)(Fisher and
Fivash, 1994). Overall, the mutagenized VHHs exhibited stronger binding affinities towards
artemin than their parental clones, as indicated by their lower equilibrium dissociation constant
(KD) values (Table 5-1). The mutagenized VHH nanobody, LM41, showed about four times
stronger binding affinity (KD ~ 109 nM) than its parental clone, LU68 (KD ~435 nM). Another
mutagenized nanobody, LM52, displayed a slightly stronger binding to artemin (KD ~ 88 nM)
compared to LU5 (KD ~ 135 nM)(Table 5-1). Based on the higher affinities demonstrated by
LM41 and LM52 selected from the AID and Polh- mutagenized VHH library, it is evident that
the in vitro affinity maturation process has improved the affinity of naïve VHHs toward artemin.

Clone name kon (M
-1
S
-1
)E
+6
±
STDEV
koff (S
-1
)E
-2
±
STDEV
KD (nM) ± STDEV Amino acid
replacement
LU68-VHH
(Original) 0.51 ± 0.13 22.0 ± 0.99 435 ± 30
Y62H

LM41-VHH
(mutant) 0.45 ± 0.13 4.93 ± 0.33 109 ± 3
LU5-VHH
(original)

3.08 ± 0.71
8.63 ± 0.17 135 ± 60 A78T, T80M,
N89K

LM52-VHH
(mutant)
3.27 ± 0.63 5.16 ± 1.63 88 ± 20

Table 5-1. Kinetic and equilibrium dissociation constant values of purified llama-VHH
nanobodies targeting artemin, determined by SPR (nM) (n = 2), and corresponding amino
acid changes. Mouse Artemin (150 RU, R&D Systems) was immobilized on CM5 sensor chip via
amine coupling, and different concentrations (10-500 nM) of each llama-VHH nanobody in PBS-0005%
Tween buffer (pH 7.4) was injected at a rate of 30 µl/min, then allowed to dissociate for 600 s. The
binding curves were fitted to 1:1 binding model using Biacore T100 Evaluation Software, to yield fitted
kinetic association constant Ka, dissociation constant Kd, and the equilibrium dissociation constants, KD.
Standard deviations were calculated for n = 2. All nanobodies were purified by periplasmic extraction,
and nickel affinity chromatography. kon = association constant. koff = dissociation constant. KD =
Equilibrium dissociation constant.

100
A.

B.

C.

Resonance
Time (s)
Resonance Resonance
Time (s)
Time (s)

101
D.

Figure. 5-5. Sensorgrams depicting the binding of the mouse artemin and purified llama-
VHH by Surface Plasmon Resonance (SPR) for the first trial. Different concentrations (0-500
nM) of the parental llama VHH-LU5 (A) and LU68(C) and affinity-matured VHH LM52 (B) and
LM41(D) were flowed over a CM5 sensor chip coated with 150 RU of mouse artemin, and the change in
response unit (RU) on the sensor chip surface has been recorded against time (second). A blank flow-cell
has been used as a background control. A total of two SPR assays has been conducted to yield fitted
kinetic constants in Table 5-1.

To determine whether the isolated VHHs specifically inhibit artemin in vivo to attenuate
the artemin-mediated cold hypersensitivity, mice behavioral assays were performed by McKemy
lab (USC Neuroscience) using one of the 4 VHH peptides, LM52, along with the positive
control, anti-mouse artemin monoclonal antibody, Mab1085 (R&D Systems)(Fig 5-5). In this
experiment, wild-type mice were given a unilateral hind paw injection of complete Freund's
adjuvant (CFA, Sigma-Aldrich, St. Louis, MO) to induce cold allodynia. After baseline testing
(BL) for cold sensitivity using the cold plantar assay previously described (Knowlton et al.,
2013; Lippoldt et al., 2016).), mice were re-tested 48 hours later (post-CFA). After a
subcutaneous nape injection of either LM52-VHH or Mab1085 (10mg/kg body weight) (Arrow,
Figure 5-5) into mice, the initial inflammation-induced response time to cold stimuli was
recorded. Cold-sensitivity in mice was tested again at 1, 2, and 4 hours post Mab1085 injection
Resonance
Time (s)

102
(n = 5, male mice)(Fig 5-6.A. & 5-7) and 1, 2, 4, 8, and 12 hours post LM52 injection (n = 9, 5
male mice and 4 female mice)(Fig 5-6.B. & 5-7). Reduction in cold sensitivity was observed in
both groups of mice injected with LM52 or Mab1085, as illustrated by the increase in withdrawal
latencies above the post-CFA level 1 to 4 hour post injection (Fig 5-5, 5-6). Although LM52-
VHH took longer (4 h post-injection, Fig 5-5.B) to fully restore the cold response back to the
pre-CFA level compared to the positive control (1h post-injection, Fig 5-5.A), its inhibitory
effect on artemin-mediated cold sensitivity was evident from the increasing trend in the
withdrawal latency post-nanobody injection. For LM52, the inhibitory effect appeared to be
waning by 12 hours (Mab-1085 injected mice were not tested beyond 4 hr time point, hence no
comparison was done), indicating the discontinuation in the binding interaction between the
nanobody and artemin. In 24 hours, cold allodynia was fully recovered in both Mab1085 and
LM52-VHH injected mice (Fig 5-7, 3 days post-CFA). The mice studies revealed that LM52-
VHH is not only specific to artemin, but also inhibits artemin function to suppress cold allodynia
in vivo.

103

Figure 5-6: Inflammatory cold allodynia is inhibited by both LM52-VHH and anti-artemin
Mab1085. Decreased cold-evoked hind paw withdrawal latencies 2 days after injection of the
inflammatory agent, CFA, were significantly increased after injection with (A) mab1085 and (B) LM52-
VHH. Contra-indicates the uninjected, control hind paw. Withdrawal latency (seconds) is the time
between when the cold stimulus was applied and when the mouse removed its hind paw. Arrow denotes
the initial inflammation-induced response time to cold stimuli at the time of mAB and VHH injection. A
shorter latency means the animal was more sensitive to cold. (A) For MAB1085, n=5 male mice; (B) For
LM52 n=9 at BL thru 4hrs (5 males, 4 females); n=4 at 8 and 12hrs (females). ns p>0.05, ***p<0.001,
**p<0.01,*p<0.05
Experiments performed by Shanni Yamaki and Chenyu Yang in the McKemy Lab, USC
Neuroscience.

0
5
10
15
MAB1085-ipsi
MAB1085-contra
BL 1 2 4
post-MAB1085 (Hrs)
ns
***
ns
ns
ns
MAB1085
0
5
10
15
LM52-ipsi
LM52-contra
BL 1 2 4 8 12
post-LM52 (Hrs)
LM52
ns
***
***
*
ns ns
**
A. B.

104

Figure 5-7: Bar graph showing the inhibition of inflammatory cold allodynia by LM52-
VHH (green) and anti-artemin Mab1085 (black) in 5 male mice (n = 5), up to 4 hrs post
injection (2d CFA) and after 24 hrs post injection (3d CFA). Cold-evoked hind paw withdrawal
latencies 2 days (2d CFA) and 3 days (3d CFA) after injection of the inflammatory agent are shown along
with the baseline level (Pre-CFA). Withdrawal latency (seconds) was recorded for each group of 5 male
mice injected with either Mab1085 or LM52, and individual data points were presented for each group,
with bars indicating the average response time (s), and error bars indicating the standard deviations (n =
5). Withdrawal latencies increased for both groups post 1 hour injection, with LM52 reaching the
effectiveness of Mab1085 2 hours post injection. (n = 5). At 24 hrs post-injection point (3d CFA), cold
allodynia is restored in both groups of mice.

Experiments performed by Shanni Yamaki in the McKemy Lab, USC Neuroscience.

105
5.4 Discussion
Llama mutagenized and naïve VHH libraries were used in parallel for biopanning against
mouse artemin (Ala112-Gly224 fragment, R&D system) and four artemin-specific VHH
nanobodies were isolated. After 3 rounds of panning and sequence analysis of isolated clones, 2
pairs of related VHHs from mutagenized and naïve libraries were identified. Sequence analysis
of these clones revealed that LM41 and LM52 from the mutagenized VHH library likely
represent AID and pol h-induced variants of LU68 and LU5 clones from the naïve VHH library.
SPR measurements showed that the mutant VHHs have relatively lower KD values, and thus
higher affinities, for binding to artemin compared to their original counterparts. The mice
behavioral studies done on one of the affinity-matured VHHs, LM52, revealed that the VHH
binds to artemin to reduce the artemin-mediated cold sensitivity in mice. This demonstrated not
only the specific binding of the isolated VHH toward artemin, but also its ability to influence the
target function in vivo. To compare the in vivo effectiveness of the original and affinity-matured
VHHs, further testing on mice will have to be conducted by injecting the original and mutant
VHH peptides simultaneously into two groups of mice.
Although it is likely that both LM 41 and LM52 from the llama mutagenized library
originated from LU68 and LU5 from the llama naïve library, based on the significant match
between the pairs’ sequences and the observations that their mutations are characteristic of AID
and Polh signature actions, we could not rule out the possibility that LM41 and LM52 clones
already existed in the llama naïve library as separate clones to the presumed parental clones in
the same library. Hence we plan to perform further rounds of affinity maturation using purified
AID and Polh on the isolated clones to further assess the effects of AID and Polh diversification
on individual Abs. AID and Polh mutagenesis is to be carried out following the two methods

106
described earlier in Chapter 4. Either gapped DNA constructs from the isolated 4 VHH phage
clones are to be prepared or single-stranded DNAs are to be extracted from the combined pool of
4 VHH-phages for the initial treatment with AID (Chapter 4, Materials & Methods). The
resultant f3TR1-VHH repertoire (produced from gapDNA method) and pADL20c-VHH
repertoire (generated from ssDNA treatment) will be transformed into appropriate E. coli strain
K91BK and TG1, respectively, to obtain two sub-libraries of mutagenized VHHs. These sub-
libraries will be used in 3-5 rounds of biopanning on the purified mouse artemin, followed by
screening and selection of affinity-matured, high-affinity artemin binders. Further analysis on
these affinity-matured variants is required to confirm the Ab-enhancing effects of in vitro affinity
maturation on artemin-specific VHH nanobodies.

107
References
Al-Sabah, S. and D. Donnelly (2003). "The positive charge at Lys-288 of the glucagon-like
peptide-1 (GLP-1) receptor is important for binding the N-terminus of peptide agonists." FEBS
Lett 553(3): 342-346.

Bautista, D. M., J. Siemens, J. M. Glazer, P. R. Tsuruda, A. I. Basbaum, C. L. Stucky, S. E. Jordt
and D. Julius (2007). "The menthol receptor TRPM8 is the principal detector of environmental
cold." Nature 448(7150): 204-208.

Bebenek, K. and T. A. Kunkel (1995). "Analyzing fidelity of DNA polymerases." Methods
Enzymol 262: 217-232.

Bransteitter, R., P. Pham, P. Calabrese and M. F. Goodman (2004). "Biochemical analysis of
hypermutational targeting by wild type and mutant activation-induced cytidine deaminase." J
Biol Chem 279(49): 51612-51621.

Bransteitter, R., P. Pham, M. D. Scharff and M. F. Goodman (2003). "Activation-induced
cytidine deaminase deaminates deoxycytidine on single-stranded DNA but requires the action of
RNase." Proc Natl Acad Sci U S A 100(7): 4102-4107.

Bransteitter, R., J. L. Sneeden, S. Allen, P. Pham and M. F. Goodman (2006). "First AID
(activation-induced cytidine deaminase) is needed to produce high affinity isotype-switched
antibodies." J Biol Chem 281(25): 16833-16836.

Brezinschek, H. P., R. I. Brezinschek and P. E. Lipsky (1995). "Analysis of the heavy chain
repertoire of human peripheral B cells using single-cell polymerase chain reaction." J Immunol
155(1): 190-202.

Chaudhuri, J., M. Tian, C. Khuong, K. Chua, E. Pinaud and F. W. Alt (2003). "Transcription-
targeted DNA deamination by the AID antibody diversification enzyme." Nature 422(6933):
726-730.

Chelico, L., P. Pham, J. Petruska and M. F. Goodman (2009). "Biochemical basis of
immunological and retroviral responses to DNA-targeted cytosine deamination by activation-
induced cytidine deaminase and APOBEC3G." J Biol Chem 284(41): 27761-27765.

Coopman, K., Y. Huang, N. Johnston, S. J. Bradley, G. F. Wilkinson and G. B. Willars (2010).
"Comparative effects of the endogenous agonist glucagon-like peptide-1 (GLP-1)-(7-36) amide
and the small-molecule ago-allosteric agent "compound 2" at the GLP-1 receptor." J Pharmacol
Exp Ther 334(3): 795-808.

Dal-Bo, M., I. Del Giudice, R. Bomben, D. Capello, F. Bertoni, F. Forconi, L. Laurenti, D.
Rossi, A. Zucchetto, G. Pozzato, R. Marasca, D. G. Efremov, A. Guarini, G. Del Poeta, R. Foa,
G. Gaidano and V. Gattei (2011). "B-cell receptor, clinical course and prognosis in chronic

108
lymphocytic leukaemia: the growing saga of the IGHV3 subgroup gene usage." Br J Haematol
153(1): 3-14.

Deacon, C. F., M. A. Nauck, M. Toft-Nielsen, L. Pridal, B. Willms and J. J. Holst (1995). "Both
subcutaneously and intravenously administered glucagon-like peptide I are rapidly degraded
from the NH2-terminus in type II diabetic patients and in healthy subjects." Diabetes 44(9):
1126-1131.

Delbos, F., S. Aoufouchi, A. Faili, J. C. Weill and C. A. Reynaud (2007). "DNA polymerase eta
is the sole contributor of A/T modifications during immunoglobulin gene hypermutation in the
mouse." J Exp Med 204(1): 17-23.

Di Noia, J. M., G. T. Williams, D. T. Chan, J. M. Buerstedde, G. S. Baldwin and M. S.
Neuberger (2007). "Dependence of antibody gene diversification on uracil excision." J Exp Med
204(13): 3209-3219.

Dickerson, S. K., E. Market, E. Besmer and F. N. Papavasiliou (2003). "AID mediates
hypermutation by deaminating single stranded DNA." J Exp Med 197(10): 1291-1296.

Fisher, R. J. and M. Fivash (1994). "Surface plasmon resonance based methods for measuring
the kinetics and binding affinities of biomolecular interactions." Curr Opin Biotechnol 5(4): 389-
395.

Fleetwood, F., N. Devoogdt, M. Pellis, U. Wernery, S. Muyldermans, S. Stahl and J. Lofblom
(2013). "Surface display of a single-domain antibody library on Gram-positive bacteria." Cell
Mol Life Sci 70(6): 1081-1093.

Goodman, M. F., M. D. Scharff and F. E. Romesberg (2007). "AID-initiated purposeful
mutations in immunoglobulin genes." Adv Immunol 94: 127-155.

Habib, I., D. Smolarek, C. Hattab, M. Grodecka, G. Hassanzadeh-Ghassabeh, S. Muyldermans,
S. Sagan, C. Gutierrez, S. Laperche, C. Le-Van-Kim, Y. C. Aronovicz, K. Wasniowska, S.
Gangnard and O. Bertrand (2013). "V(H)H (nanobody) directed against human glycophorin A: a
tool for autologous red cell agglutination assays." Anal Biochem 438(1): 82-89.

Hassanzadeh-Ghassabeh, G., N. Devoogdt, P. De Pauw, C. Vincke and S. Muyldermans (2013).
"Nanobodies and their potential applications." Nanomedicine (Lond) 8(6): 1013-1026.

Hnasko, R. M. and L. H. Stanker (2015). "Hybridoma Technology." Methods Mol Biol 1318:
15-28.

Hust, M., A. Frenzel, T. Meyer, T. Schirrmann and S. Dubel (2012). "Construction of human
naive antibody gene libraries." Methods Mol Biol 907: 85-107.

109
Knowlton, W. M., A. Bifolck-Fisher, D. M. Bautista and D. D. McKemy (2010). "TRPM8, but
not TRPA1, is required for neural and behavioral responses to acute noxious cold temperatures
and cold-mimetics in vivo." Pain 150(2): 340-350.

Lippoldt, E. K., R. R. Elmes, D. D. McCoy, W. M. Knowlton and D. D. McKemy (2013).
"Artemin, a glial cell line-derived neurotrophic factor family member, induces TRPM8-
dependent cold pain." J Neurosci 33(30): 12543-12552.

Lippoldt, E. K., S. Ongun, G. K. Kusaka and D. D. McKemy (2016). "Inflammatory and
neuropathic cold allodynia are selectively mediated by the neurotrophic factor receptor
GFRalpha3." Proc Natl Acad Sci U S A 113(16): 4506-4511.

Longerich, S., U. Basu, F. Alt and U. Storb (2006). "AID in somatic hypermutation and class
switch recombination." Curr Opin Immunol 18(2): 164-174.

MacCarthy, T., S. L. Kalis, S. Roa, P. Pham, M. F. Goodman, M. D. Scharff and A. Bergman
(2009). "V-region mutation in vitro, in vivo, and in silico reveal the importance of the enzymatic
properties of AID and the sequence environment." Proc Natl Acad Sci U S A 106(21): 8629-
8634.

Maccarthy, T., S. Roa, M. D. Scharff and A. Bergman (2009). "SHMTool: a webserver for
comparative analysis of somatic hypermutation datasets." DNA Repair (Amst) 8(1): 137-141.

Mak, C. H., P. Pham, S. A. Afif and M. F. Goodman (2013). "A mathematical model for
scanning and catalysis on single-stranded DNA, illustrated with activation-induced
deoxycytidine deaminase." J Biol Chem 288(41): 29786-29795.

Masutani, C., R. Kusumoto, A. Yamada, M. Yuasa, M. Araki, T. Nogimori, M. Yokoi, T. Eki, S.
Iwai and F. Hanaoka (2000). "Xeroderma pigmentosum variant: from a human genetic disorder
to a novel DNA polymerase." Cold Spring Harb Symp Quant Biol 65: 71-80.

Matsuda, T., K. Bebenek, C. Masutani, F. Hanaoka and T. A. Kunkel (2000). "Low fidelity DNA
synthesis by human DNA polymerase-eta." Nature 404(6781): 1011-1013.

McCafferty, J., K. J. Fitzgerald, J. Earnshaw, D. J. Chiswell, J. Link, R. Smith and J. Kenten
(1994). "Selection and rapid purification of murine antibody fragments that bind a transition-
state analog by phage display." Appl Biochem Biotechnol 47(2-3): 157-171; discussion 171-153.

McKean, D., K. Huppi, M. Bell, L. Staudt, W. Gerhard and M. Weigert (1984). "Generation of
antibody diversity in the immune response of BALB/c mice to influenza virus hemagglutinin."
Proc Natl Acad Sci U S A 81(10): 3180-3184.

Muramatsu, M., V. S. Sankaranand, S. Anant, M. Sugai, K. Kinoshita, N. O. Davidson and T.
Honjo (1999). "Specific expression of activation-induced cytidine deaminase (AID), a novel
member of the RNA-editing deaminase family in germinal center B cells." J Biol Chem 274(26):
18470-18476.

110
Muyldermans, S. (2013). "Nanobodies: natural single-domain antibodies." Annu Rev Biochem
82: 775-797.

Nauck, M. A., E. Bartels, C. Orskov, R. Ebert and W. Creutzfeldt (1993). "Additive
insulinotropic effects of exogenous synthetic human gastric inhibitory polypeptide and glucagon-
like peptide-1-(7-36) amide infused at near-physiological insulinotropic hormone and glucose
concentrations." J Clin Endocrinol Metab 76(4): 912-917.

Palczewski, K., T. Kumasaka, T. Hori, C. A. Behnke, H. Motoshima, B. A. Fox, I. Le Trong, D.
C. Teller, T. Okada, R. E. Stenkamp, M. Yamamoto and M. Miyano (2000). "Crystal structure of
rhodopsin: A G protein-coupled receptor." Science 289(5480): 739-745.

Pavlov, Y. I., I. B. Rogozin, A. P. Galkin, A. Y. Aksenova, F. Hanaoka, C. Rada and T. A.
Kunkel (2002). "Correlation of somatic hypermutation specificity and A-T base pair substitution
errors by DNA polymerase eta during copying of a mouse immunoglobulin kappa light chain
transgene." Proc Natl Acad Sci U S A 99(15): 9954-9959.

Peled, J. U., F. L. Kuang, M. D. Iglesias-Ussel, S. Roa, S. L. Kalis, M. F. Goodman and M. D.
Scharff (2008). "The biochemistry of somatic hypermutation." Annu Rev Immunol 26: 481-511.

Pham, P., R. Bransteitter, J. Petruska and M. F. Goodman (2003). "Processive AID-catalysed
cytosine deamination on single-stranded DNA simulates somatic hypermutation." Nature
424(6944): 103-107.

Pham, P., P. Calabrese, S. J. Park and M. F. Goodman (2011). "Analysis of a single-stranded
DNA-scanning process in which activation-induced deoxycytidine deaminase (AID) deaminates
C to U haphazardly and inefficiently to ensure mutational diversity." J Biol Chem 286(28):
24931-24942.

Pham, P., S. Malik, C. Mak, P. C. Calabrese, R. G. Roeder and M. F. Goodman (2019). "AID-
RNA polymerase II transcription-dependent deamination of IgV DNA." Nucleic Acids Res
47(20): 10815-10829.

Poulsen, T. R., A. Jensen, J. S. Haurum and P. S. Andersen (2011). "Limits for antibody affinity
maturation and repertoire diversification in hypervaccinated humans." J Immunol 187(8): 4229-
4235.

Prakash, S., R. E. Johnson, M. T. Washington, L. Haracska, C. M. Kondratick and L. Prakash
(2000). "Role of yeast and human DNA polymerase eta in error-free replication of damaged
DNA." Cold Spring Harb Symp Quant Biol 65: 51-59.

Rada, C., J. M. Di Noia and M. S. Neuberger (2004). "Mismatch recognition and uracil excision
provide complementary paths to both Ig switching and the A/T-focused phase of somatic
mutation." Mol Cell 16(2): 163-171.

111
Rajewsky, K. (1996). "Clonal selection and learning in the antibody system." Nature 381(6585):
751-758.

Rogozin, I. B. and N. A. Kolchanov (1992). "Somatic hypermutagenesis in immunoglobulin
genes. II. Influence of neighbouring base sequences on mutagenesis." Biochim Biophys Acta
1171(1): 11-18.

Rogozin, I. B., Y. I. Pavlov, K. Bebenek, T. Matsuda and T. A. Kunkel (2001). "Somatic
mutation hotspots correlate with DNA polymerase eta error spectrum." Nat Immunol 2(6): 530-
536.

Runge, S., H. Thogersen, K. Madsen, J. Lau and R. Rudolph (2008). "Crystal structure of the
ligand-bound glucagon-like peptide-1 receptor extracellular domain." J Biol Chem 283(17):
11340-11347.

Sarramegna, V., I. Muller, A. Milon and F. Talmont (2006). "Recombinant G protein-coupled
receptors from expression to renaturation: a challenge towards structure." Cell Mol Life Sci.

Shapiro, G. S. and L. J. Wysocki (2002). "DNA target motifs of somatic mutagenesis in antibody
genes." Crit Rev Immunol 22(3): 183-200.

Shim, H. (2016). "Therapeutic antibodies by phage display." Curr Pharm Des 22(43): 6538-
6559.

Shim, H. (2017). "Antibody Phage Display." Adv Exp Med Biol 1053: 21-34.

Smith, G. P. (1985). "Filamentous fusion phage: novel expression vectors that display cloned
antigens on the virion surface." Science 228(4705): 1315-1317.

Tang, C., D. Bagnara, N. Chiorazzi, M. D. Scharff and T. MacCarthy (2020). "AID Overlapping
and Poleta Hotspots Are Key Features of Evolutionary Variation Within the Human Antibody
Heavy Chain (IGHV) Genes." Front Immunol 11: 788.

Thomas, W. D., M. Golomb and G. P. Smith (2010). "Corruption of phage display libraries by
target-unrelated clones: diagnosis and countermeasures." Anal Biochem 407(2): 237-240.

Wang, Y., R. Woodgate, T. P. McManus, S. Mead, J. J. McCormick and V. M. Maher (2007).
"Evidence that in xeroderma pigmentosum variant cells, which lack DNA polymerase eta, DNA
polymerase iota causes the very high frequency and unique spectrum of UV-induced mutations."
Cancer Res 67(7): 3018-3026.

Watkins, H. A., M. Au and D. L. Hay (2012). "The structure of secretin family GPCR peptide
ligands: implications for receptor pharmacology and drug development." Drug Discov Today
17(17-18): 1006-1014.

112
Wei, L., R. Chahwan, S. Wang, X. Wang, P. T. Pham, M. F. Goodman, A. Bergman, M. D.
Scharff and T. MacCarthy (2015). "Overlapping hotspots in CDRs are critical sites for V region
diversification." Proc Natl Acad Sci U S A 112(7): E728-737.

Wilmen, A., B. Goke and R. Goke (1996). "The isolated N-terminal extracellular domain of the
glucagon-like peptide-1 (GLP)-1 receptor has intrinsic binding activity." FEBS Lett 398(1): 43-
47.

Wu, F., L. Yang, K. Hang, M. Laursen, L. Wu, G. W. Han, Q. Ren, N. K. Roed, G. Lin, M. A.
Hanson, H. Jiang, M. W. Wang, S. Reedtz-Runge, G. Song and R. C. Stevens (2020). "Full-
length human GLP-1 receptor structure without orthosteric ligands." Nat Commun 11(1): 1272.

Yuan, C., C. C. Chu, X. J. Yan, D. Bagnara, N. Chiorazzi and T. MacCarthy (2017). "The
Number of Overlapping AID Hotspots in Germline IGHV Genes Is Inversely Correlated with
Mutation Frequency in Chronic Lymphocytic Leukemia." PLoS One 12(1): e0167602.

Zhang, Y., B. Sun, D. Feng, H. Hu, M. Chu, Q. Qu, J. T. Tarrasch, S. Li, T. Sun Kobilka, B. K.
Kobilka and G. Skiniotis (2017). "Cryo-EM structure of the activated GLP-1 receptor in complex
with a G protein." Nature 546(7657): 248-253.

Zhao, Y., M. T. Gregory, C. Biertumpfel, Y. J. Hua, F. Hanaoka and W. Yang (2013).
"Mechanism of somatic hypermutation at the WA motif by human DNA polymerase eta." Proc
Natl Acad Sci U S A 110(20): 8146-8151.

113
Supplementary Figures
List 1. Humanized synthetic 18 VH and 20 VL sequences used for the generation of scFv
synthetic library (Chapter 2)

Humanized VH Gene

>Hu-VH914
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGGTGGAGTCTGGAGGAGGCTTGATCCAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGAATCGACCTCAGTCGCAATGCAATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGTTATTAATACTAGGGGTGACACATACTACGCAGACTCCGTGAAGGGCCGATTCACCA
TCTCCAGAGACAATTCCAAGAACACGCTGTATCTTCAAATGAACAGCCTGAGAGCCGAGGACACGGCCGTGTATTAC
TGTGTTCATGGTTATAATCCCTGTAAGTTGTGGGGCCAGGGCACCCTGGTCACCGTCTCCGGTGGTGGTGGTTCTGG
TGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH930
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGGTGGAGTCTGGGGGAGGCTTGGTCCAGCCTGGGGGGTCC
CTGAGACTCTCCTGTTCAGCCTCTGGAATCGACCTCAGTTACTATGCTATGCACTGGGTCCGCCAGGCTCCAGGGAA
GGGACTGGAATATGTTTCAGCTATTGGTAGTAGTGGTAGAACATACTACGCAGACTCCGTGAAGGGCAGATTCACCA
TCTCCAGAGACAATTCCAAGAACACGCTGTATCTTCAAATGAGCAGTCTGAGAGCTGAGGACACGGCTGTGTATTAC
TGTGTCAGGGGTGGTCCTACTTCTAGTCCTAGTTTGTGGGGCCAGGGCACCCTGGTCACCGTCTCCGGTGGTGGTGG
TTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC
>Hu-VH931
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGGTGGAGTCTGGGGGAGGCTTGGTCCAGCCTGGGGGGTCC
CTGAAACTCTCCTGTGCAGCCTCTGGAATCGACCTCAGTAGAAATATGCACTGGGTCCGCCAGGCTTCCGGGAAAGG
GCTGGAGTGGGTTGGCCGTATTGCGAGTCGTGGTAACATATGGTTCAGGGCATATGCTGCGTCGGTGAAAGGCAGGT
TCACCATCTCCAGAGATGATTCAAAGAACACGGCGTATCTGCAAATGAACAGCCTGAAAACCGAGGACACGGCCGTG
TATTACTGTGGCAGATTTCTTGTAGTATCTGGGGTGGGGACTTTTGATCCCTGGGGCCAAGGAACCCTGGTCACCGT
CTCCGGTGGTGGTGGTTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH932
TACTCGCGGCCCAGCCGGCCATGGCTCAGGTGCAGCTGCAGGAGTCGGGCCCAGGACTGGTGAAGCCTTCGGGGACC
CTGTCCCTCACCTGCGCTGTCTCTGGATTCTCCCTCAGTCGCTACTGGTGGAGTTGGGTCCGCCAGCCCCCAGGGAA
GGGGCTGGAGTGGATTGGGGAAATTGGTGGTGGTAGTGGTAGTACAAACTACAACCCGTCCCTCAAGAGTCGAGTCA
CCATATCAGTAGACAAGTCCAAGAACCAGTTCTCCCTGAAGCTGAGCTCTGTGACCGCCGCGGACACGGCCGTGTAT
TACTGTGGCAGATATGTTAAAAATGGTGGTGGTTATCGGTTGGATCTTTGGGGCCAGGGCACCCTGGTCACCGTCTC
CGGTGGTGGTGGTTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH933
TACTCGCGGCCCAGCCGGCCATGGCTCAGGTGCAGCTGCAGGAGTCGGGCCCAGGACTGGTGAAGCCTTCGGAGACC
CTGTCCCTCACCTGCACTGTCTCTGGATTCTCCCTCAGTAGCTATGCATGGAGCTGGATCCGGCAGCCCCCAGGGAA
GGGACTGGAGTGGATTGGGTATATTAACACTATTACTGGTGGCACAAACTACAACCCCTCCCTCAAGAGTCGAGTCA
CCATATCCGTAGACACGTCCAAGAACCAGTTCTCCCTGAAGCTGAGCTCTGTGACCGCCGCAGACACGGCCGTGTAT
TACTGTGCGAGAAGTGGTGCCTACTTTGACTTGTGGGGCCGTGGCACCCTGGTCACTGTCTCCGGTGGTGGTGGTTC
TGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH983
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGTTGGAGTCTGGGGGAGGCTTGGTACAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGATTCTCCCTCAGTAACTATGCAATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGCTATTGGTAAAAGTGGTAGTACGTACTACGCAGACTCCGTGAAGGGCCGGTTCACCA
TCTCCAGAGACAATTCCAAGAACACGCTGTATCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCCGTATATTAC

114
TGTGTCAGATTTGTGCTCTTGTGGGGCCGTGGCACCCTGGTCACTGTCTCCGGTGGTGGTGGTTCTGGTGGTGGTGG
TTCTGGCGGCGGCGGCTCC

>Hu-VH984
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGTTGGAGTCTGGGGGAGGCTTGGTACAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGAATTCTCCCTCAGTGACTATATAATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGCTATCATGGGTACTAGTGGTACCGCATACTACGCAGACTCCGTGAAGGGCCGGTTCA
CCATCTCCAGAGACAATTCCAAGAACACGCTGTATCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCCGTATAT
TACTGTGCCAGAGGGGGTGTTGCTACTTCTAATTTCTGGGGCCAGGGCACCCTGGTCACCGTCTCCGGTGGTGGTGG
TTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH985
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGTTGGAGTCTGGGGGAGGCTTGGTACAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGATTCTCCCTCAACGACTACGACATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAACCATTTATGTTAGTGGTAACACATACTATGCAGACTCCGTGAAGGGCCGGTTCACCA
TCTCCAGAGATAATTCCAAGAACACGCTGTATCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCCGTATATTAC
TGTGCCAGAGCGGTTCCTGGTAGTGGTAAGGGGTTGTGGGGCCGTGGCACCCTGGTCACTGTCTCCGGTGGTGGTGG
TTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1059
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGTTGGAGTCTGGGGGAGGCTTGGTACAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGAATCGACCTCAGTAGCTATGCAATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGTTATTATTTATGCTAGTGATACATACTATGCAGACTCCGTGAAGGGCCGGTTCACCA
TCTCCAGAGATAATTCCAAGAACACGCTGTATCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCCGTATATTAC
TGTGCCAGAGGAGATTCTACTAGTGGTTTTCATTTGACTTTGTGGGGCCAGGGCACCCTGGTCACCGTCTCCGGTGG
TGGTGGTTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1066
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGTTGGAGTCTGGGGGAGGCTTGGTACAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGATTCGACCTCAGTAGCTATGCAATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGTTATTATTTATGGTACTGAACTCACATACTATGCAGACTCCGTGAAGGGCCGGTTCA
CCATCTCCAGAGATAATTCCAAGAACACGCTGTATCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCCGTATAT
TACTGTGCCAGAGGAGATGCTACTAGTGGTTTTCATTTGACGTTGTGGGGCCAGGGCACCCTGGTCACCGTCTCCGG
TGGTGGTGGTTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1068
TACTCGCGGCCCAGCCGGCCATGGCTCAGGTACAGCTGGTGGAGTCTGGGGGAGGCGTGGTCCAGCCTGGGAGGTCC
CTGAGACTCTCCTGTGCAGCGTCTGGATTCTCCTTCAGTAGCAGCTACTGGATGCACTGGGTCCGCCAGGCTCCAGG
CAAGGGGCTGGAGTGGGTGGCAGTTATTTATGGTGGTAGTATTAATATTATTTACTATGCAGACTCCGCGAAGGGCC
GATTCACCATCTCCAGAGACAATTCCACGAACACGCTGTTTCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCT
GTGTATTACTGTGCGAGAGTACAGGGTGGTAATAGTGGTGGTTGGTACTTTGACTTGTGGGGCCGTGGCACCCTGGT
CACTGTCTCCGGTGGTGGTGGTTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1080
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGGTGGAGTCTGGAGGAGGCTTGGTCCAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGAATCGACCTCAGTAAGTGGCCAATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGTTATTGGTAGGAGTGGTAGCACGTACTACGCAGACTCCGTGAAGGGCCGATTCACCA
TCTCCAGACACAATTCCAAGAACACGCTGTATCTTCAAATGAACAGCCTGAGAGCTGAGGACACGGCCGTGTATTAC
TGTGCCAGAGGTGGTAGTTATTATGATTTGTGGGGCCGTGGCACCCTGGTCACTGTCTCCGGTGGTGGTGGTTCTGG
TGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1081
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGTTGGAGTCTGGGGGAGGCTTGGTACAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGATTCTCCCTCAGTACCTATGCAATGAGCTGGGTCCGCCAGGCTCCAGGGAA

115
GGGGCTGGAGTGGGTCTCAGCTATCGTTGGAAAGAGTGGTATTATATACTACGCAGACTCCGTGAAGGGCCGGTTCA
CCATCTCCAGAGACAATTCCAAGAACACGCTGTATCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCCGTATAT
TACTGTGCCAGACTATGGAGCTTGTGGGGCCAGGGCACCCTGGTCACCGTCTCCGGTGGTGGTGGTTCTGGTGGTGG
TGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1082
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGTTGGAGTCTGGGGGAGGCTTGGTACAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGAATCGACCTCAGTAGTTATGCAATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGCTGTTCGTCGTAGTGGTACCACATACTACGCAGACTCCGTGAAGGGCCGGTTCACCA
TCTCCAGAGACAATTCCAAGAACACGCTGTATCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCCGTATATTAC
TGTGCCAGATGTGATAATAGTGCTGGTGACTGGAGTTACGGCATGGACCTCTGGGGCCAGGGCACCCTGGTCACCGT
CTCCGGTGGTGGTGGTTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1083
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGGTGGAGTCTGGAGGAGGCTTGGTCCAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGATTCTCCCTCAGTACCAATGCAATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGTTATTGCTGGTAGTGGTAGCACATACTACGCAGACTCCGTGAAGGGCCGATTCACCA
TCTCCAGACACAATTCCAAGAACACGCTGTATCTTCAAATGAACAGCCTGAGAGCTGAGGACACGGCCGTGTATTAC
TGTGCCAGAGGGGGTTGGGTTAGTGGTCCGGAGAGCTTGTGGGGCCAGGGCACCCTGGTCACCGTCTCCGGTGGTGG
TGGTTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1084
TACTCGCGGCCCAGCCGGCCATGGCTCAGGTACAGCTGGTGGAGTCTGGGGGAGGCGTGGTCCAGCCTGGGAGGTCC
CTGAGACTCTCCTGTGCAGCGTCTGGATTCTCCCTCAATAACTACGACATGCACTGGGTCCGCCAGGCTCCAGGCAA
GGGGCTGGAGTGGGTGGCAGTTATTTTTGTTAGTGGTAATATATACTATGCAGACTCCGCGAAGGGCCGATTCACCA
TCTCCAGAGACAATTCCACGAACACGCTGTTTCTGCAAATGAACAGCCTGAGAGCCGAGGACACGGCTGTGTATTAC
TGTGCCAGAGCAATTCTTGGTAGTAGTAAGGGGTTGTGGGGCCGTGGCACCCTGGTCACTGTCTCCGGTGGTGGTGG
TTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1085
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGGTGGAGTCTGGAGGAGGCTTGGTCCAGCCTGGGGGGTCC
CTGAGACTCTCCTGTGCAGCCTCTGGATTCCCCCTCAGCATCTACGACATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCTCAGTTATTTATGTTAGTGGTAATATATACTACGCAGACTCCGTGAAGGGCCGATTCACCA
TCTCCAGACACAATTCCAAGAACACGCTGTATCTTCAAATGAACAGCCTGAGAGCTGAGGACACGGCCGTGTATTAC
TGTGCCAGAGCGGTTCCTGGTAGTAGTAAGGGGTTGTGGGGCCAGGGCACCCTGGTCACCGTCTCCGGTGGTGGTGG
TTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

>Hu-VH1086
TACTCGCGGCCCAGCCGGCCATGGCTGAGGTGCAGCTGGTGGAGTCTGGGGGAGGCTTGGTAAAGCCTGGGGGGTCC
CTTAGACTCTCCTGTGCAGCCTCTGGATTCTCCCTCGATAACTACCACATGAGCTGGGTCCGCCAGGCTCCAGGGAA
GGGGCTGGAGTGGGTCGGCCGTATTACTCGTGGTGGTACCACAAACTACGCTGCACCCGTGAAAGGCAGATTCACCA
TCTCAAGAGATGATTCAAAAAACACGCTGTATCTGCAAATGAACAGCCTGAAAACCGAGGACACAGCCGTGTATTAC
TGTGCCAGAGGAAGTGGCGCTAGCGGCTTTTACTTGTGGGGCCGTGGCACCCTGGTCACTGTCTCCGGTGGTGGTGG
TTCTGGTGGTGGTGGTTCTGGCGGCGGCGGCTCC

Humanized VL Gene

>Hu-VL914
GGCGGCGGCGGCTCC
GGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCTTCCACCCTGTCTGCATCTGTAGGAGACAGAGTCACCAT
CACTTGCCGGGCCAGTGAGAGCATTGGCACTGCATTGGCCTGGTATCAGCAGAAACCAGGGAAAGCCCCTAAGCTCC
TGATCTATAAGGCATCCAGTTTAGAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGTGGATCTGGGACAGAATTCACT

116
CTCACCATCAGCAGCCTGCAGCCTGATGATTTTGCAACTTATTACTGCCAACAGGGTGAAACTGCAAATAGAATTGA
TAATGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACCA

>Hu-VL915
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCTTCCGTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGTCGGGCGAGTGAGAGCATTAACACTGCATTAGCCTGGTATCAGCAGAAACCAGGGA
AAGCCCCTAAGCTCCTGATCTATGCTGCCTCCAGTTTGCAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGTGGATCT
GGGACAGATTTCACTCTCACCATCAGCAGCCTGCAGCCTGAAGATTTTGCAACTTACTATTGTCAACAGGGTTATAC
TGCAAATAATATTGATAATGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACC
A

>Hu-VL930
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCCTCACTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGTCGGGCGAGTCAGAATGTTGTTAATAACAACTGGTTAGCCTGGTTTCAGCAGAAAC
CAGGGAAAGCCCCTAAGTCCCTGATCTATTTTGTATCCAGTTTGCAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGGACAGATTTCACTCTCACCATCAGCAGCCTGCAGCCTGAAGATTTTGCAACTTATTACTGTGGAGGCGG
TTATAGTGATAATATTCTCACTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACC
A

>Hu-VL931
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGCCATCCAGATGACCCAGTCTCCATCCTCCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCAAGTCAGAGTGTTTATGGGACCAACCGTTTAGGCTGGTATCAGCAGAAAC
CAGGGAAAGCCCCTAAGCTCCTGATCTATGGTGCATCCAGTTTACAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGCACAGATTTCACTCTCACCATCAGCAGCCTGCAGCCTGAAGATTTTGCAACTTATTACTGTCTAGGCGG
TTGGTTTGAAAGTAGTAGTAGTCTTGATTGGGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGG
GAGGCCAACACCA

>Hu-VL932
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCTTCCACCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTCAGAGTGTTTATAATAACAACGAATTGGCCTGGTATCAGCAGAAAC
CAGGGAAAGCCCCTAAGCTCCTGATCTATGATGCATCCAGTTTGGAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGGACAGAATTCACTCTCACCATCAGCAGCCTGCAGCCTGATGATTTTGCAACTTATTACTGCCTAGGCGG
TTATAATGATGATACTAATAGATGGGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCC
AACACCA

>Hu-VL933
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCTTCCACCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTGAGAGTGTTGCTAATAACAACTGGTTGGCCTGGTATCAGCAGAAAC
CAGGGAAAGCCCCTAAGCTCCTGATCTATAAGGCATCCAGTTTAGAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGGACAGAATTCACTCTCACCATCAGCAGCCTGCAGCCTGATGATTTTGCAACTTATTACTGCGCAGGATA
TAAAAGTAGTACTACTGATGCTGTTGCTTTCGGCCAAGGGACCAAGGTGGAAATCAAAACTAGTGGCCCGGGAGGCC
AACACCA

>Hu-VL983
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCTTCCACCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTCAGAGTGTTTATAAGAACAACGACTTGGCCTGGTATCAGCAGAAAC
CAGGGAAAGCCCCTAAGCTCCTGATCTATTATGCATCCAGTTTAGAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGGACAGAATTCACTCTCACCATCAGCAGCCTGCAGCCTGATGATTTTGCAACTTATTACTGCCTAGGTAG
TTATGATTGTAGTAGTGCTGATTGTAATGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAG
GCCAACACCA

117
>Hu-VL984
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCCTCCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCGAGTCAGAGTGTTAATAACAACAACTTCTTAGCCTGGTATCAGCAGAAAC
CAGGGAAAGTTCCTAAGCTCCTGATCTATAGGGCTTCCACTTTGCAATCAGGGGTCCCATCTCGGTTCAGTGGCAGT
GGATCTGGGACAGATTTCACTCTCACCATCAGCAGCCTGCAGCCTGAAGATGTTGCAACTTATTACTGTGCAGGCGG
TTATAGTGGTAATATTTATGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACC
A

>Hu-VL985
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCTTCCACCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTCAGAGCGTTTATGGTAACAATTGGTTGGCCTGGTATCAGCAGAAAC
CAGGGAAAGCCCCTAAGCTCCTGATCTATTCTGCATCTAGTTTGGAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGGACAGAATTCACTCTCACCATCAGCAGCCTGCAGCCTGATGATTTTGCAACTTATTACTGCGTAGGCGG
GTATAGTGGTAATATTCATGTTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACC
A

>Hu-VL1059
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCCTCCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCGAGTCAGAGTCTTTATAATAAGAAAAATTTAGCCTGGTATCAGCAGAAAC
CAGGGAAAGTTCCTAAGCTCCTGATCTATTATGCATCCACTTTGCAATCAGGGGTCCCATCTCGGTTCAGTGGCAGT
GGATCTGGGACAGATTTCACTCTCACCATCAGCAGCCTGCAGCCTGAAGATGTTGCAACTTATTACTGTCAAGGCGA
ATTTAGTTGCAGCAGTGTTGATTGGACGTTCGGCCAAGGGACCAAGGTGGAAATCAAAACTAGTGGCCCGGGAGGCC
AACACCA

>Hu-VL1060
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCCTCCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCGAGTCAGAGTCTTTATAATAAGAAAAATTTAGCCTGGTATCAGCAGAAAC
CAGGGAAAGTTCCTAAGCTCCTGATCTATTTTGCATCCACTTTGCAATCAGGGGTCCCATCTCGGTTCAGTGGCAGT
GGATCTGGGACAGATTTCACTCTCACCATCAGCAGCCTGCAGCCTGAAGATGTTGCAACTTATTACTGTCAGGGCGA
GTTTATTTGCAGCAGTGGTGATTGTGTTGCTTTCGGCCAAGGGACCAAGGTGGAAATCAAAACTAGTGGCCCGGGAG
GCCAACACCA

>Hu-VL1066
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCCTCCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCGAGTCAGAGTCTTTATAATAAGAAAAATTTAGCCTGGTATCAGCAGAAAC
CAGGGAAAGTTCCTAAGCTCCTGATCTATTTTACGTCCACTTTGCAATCAGGGGTCCCATCTCGGTTCAGTGGCAGT
GGATCTGGGACAGATTTCACTCTCACCATCAGCAGCCTGCAGCCTGAAGATGTTGCAACTTATTACTGTCAAGGCGA
ATTTAGTTGTAGCAGTGGTGATTGTCTTGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAG
GCCAACACCA

>Hu-VL1068
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGTTGACCCAGTCTCCATCCTTCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTCAGACCATTTATAGTGGTTTAGCCTGGTATCAGCAAAAACCAGGGA
AAGCCCCTAAGCTCCTGATCTATAGGGCATCCACTTTGCAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGTGGATCT
GGGACAGAATTCACTCTCACAATCAGCAGCCTGCAGCCTGAAGATTTTGCAACTTATTACTGTCAAGTTGGTGGTTA
TGGTGTTAGTTATGACCATGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACC
A

>Hu-VL1080
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCTTCCACCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTCAGAGTGTTTGGAAGAATAACGACTTGGCCTGGTATCAGCAGAAAC
CAGGGAAAGCCCCTAAGCTCCTGATCTATTATGCATCCAGTTTAGAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGGACAGAATTCACTCTCACCATCAGCAGCCTGCAGCCTGATGATTTTGCAACTTATTACTGCGTAGGCAG
TTATGATTGTAGTAGTGCTGATTGTAATGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAG
GCCAACACCA

118

>Hu-VL1081
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCCTCCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCAGGCGAGTCAGAGTGTTGATAATAACAACTACTTAAATTGGTATCAGCAGAAAC
CAGGGAAAGCCCCTAAGCTCCTGATCTACGATGCATCCAATTTGGAAACAGGGGTCCCATCAAGGTTCAGTGGAAGT
GGATCTGGGACAGATTTTACTTTCACCATCAGCAGCCTGCAGCCTGAAGATATTGCAACATATTACTGTGCAGGCGG
TTATATAACTAGTAGTGATATTTTTTATGATTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAG
GCCAACACCA

>Hu-VL1082
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCTTCCACCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTCAGAGCATTAGCAACTGGTTGGCCTGGTATCAGCAGAAACCAGGGA
AAGCCCCTAAGCTCCTGATCTATAGGGCATCCAGTTTAGAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGTGGATCT
GGGACAGAATTCACTCTCACCATCAGCAGCCTGCAGCCTGATGATTTTGCAACTTATTACTGCCAAAGCGATTATGG
TATAGATACTTATGGAAGTGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACC
A

>Hu-VL1083
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCTGCCATGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGTCGGGCGAGTCAGAGTGTTTATCAGAACAACTACTTAGCCTGGTTTCAGCAGAAAC
CAGGGAAAGTCCCTAAGCGCCTGATCTATTCTGCATCCAGTTTGCAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGGACAGAATTCACTCTCACAATCAGCAGCCTGCAGCCTGAAGATTTTGCAACTTATTACTGTCTGGGCGC
CTATGATTGTAGTGGTGTTGATTGTAGTGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAG
GCCAACACCA

>Hu-VL1084
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCTTCCACCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTCCGAGTGTTTATGGTAATAACTGGTTGGCCTGGTATCAGCAGAAAC
CAGGGAAAGCCCCTAAGCTCCTGATCTATTCTGCATCCAGTTTGGAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGT
GGATCTGGGACAGAATTCACTCTCACCATCAGCAGCCTGCAGCCTGATGATTTTGCAACTTATTACTGCGCAGGCGG
TTATAGTGGTAATATTCATGTTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACC
A

>Hu-VL1085
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGATGACCCAGTCTCCATCTTCCGTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGTCGGGCGAGTCAGAGCATTTACAGCTATTTAGCCTGGTATCAGCAGAAACCAGGGA
AAGCCCCTAAGCTCCTGATCTATTCTGCATCCAGTTTGCAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGTGGATCT
GGGACAGATTTCACTCTCACCATCAGCAGCCTGCAGCCTGAAGATTTTGCAACTTACTATTGTCAACACGGGTACAT
TAGTGGTAATGTTGATAATGCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACC
A

>Hu-VL1086
GGCGGCGGCGGCTCCGGTGGTGGTGGATCCGACATCCAGTTGACCCAGTCTCCATCCTTCCTGTCTGCATCTGTAGG
AGACAGAGTCACCATCACTTGCCGGGCCAGTCAGAGTATTTATACTAACTACTTAGCCTGGTATCAGCAAAAACCAG
GGAAAGCCCCTAAGCTCCTGATCTATTCTGCATCCACTTTGCAAAGTGGGGTCCCATCAAGGTTCAGCGGCAGTGGA
TCTGGGACAGAATTCACTCTCACAATCAGCAGCCTGCAGCCTGAAGATTTTGCAACTTATTACTGTCAAGCCTATTT
TACTGGTGAGATTTTTCCTTTCGGCGGAGGGACCAAGGTGGAGATCAAAACTAGTGGCCCGGGAGGCCAACACCA

119
List 2. Primers used for first and second PCR of antibody genes for antibody gene library
construction using phagemids like pHAL14
Primer 5’ –3’ sequence
First antibody gene PCR VH
MHVH1_f cag gtb cag ctg gtg cag tct gg
MHVH1/7_f car rts cag ctg gtr car tct gg
MHVH2_f cag rtc acc ttg aag gag tct gg
MHVH3_f1 sar gtg cag ctg gtg gag tct gg
MHVH3_f2 gag gtg cag ctg ktg gag wcy sg
MHVH4_f1 cag gtg car ctg cag gag tcg gg
MHVH4_f2 cag stg cag ctr cag sag tss gg
MHVH5_f gar gtg cag ctg gtg cag tct gg
MHVH6_f cag gta cag ctg cag cag tca gg
MHIgMCH1_r aag ggt tgg ggc gga tgc act
MHIgGCH1_r gac cga tgg gcc ctt ggt gga
MHIgECH1_r tgg gct ctg tgt gga gg

First antibody gene PCR kappa
MHVK1_f1 gac atc cag atg acc cag tct cc
MHVK1_f2 gmc atc crg wtg acc cag tct cc
MHVK2_f gat rtt gtg atg acy cag wct cc
MHVK3_f gaa atw gtg wtg acr cag tct cc
MHVK4_f gac atc gtg atg acc cag tct cc
MHVK5_f gaa acg aca ctc acg cag tct cc
MHVK6_f gaw rtt gtg mtg acw cag tct cc
MHkappaCL_r aca ctc tcc cct gtt gaa gct ctt

First antibody gene PCR lambda
MHVL1_f1 cag tct gtg ctg act cag cca cc
MHVL1_f2 cag tct gtg ytg acg cag ccg cc
MHVL2_f cag tct gcc ctg act cag cct
MHVL3_f1 tcc tat gwg ctg acw cag cca cc
MHVL3_f2 tct tct gag ctg act cag gac cc
MHVL4_f1 ctg cct gtg ctg act cag ccc
MHVL4_f2 cag cyt gtg ctg act caa tcr yc
MHVL5_f cag sct gtg ctg act cag cc
MHVL6_f aat ttt atg ctg act cag ccc ca
MHVL7/8_f cag rct gtg gtg acy cag gag cc
MHVL9/10_f cag scw gkg ctg act cag cca cc
MHlambdaCL_r tga aca ttc tgt agg ggc cac tg
MHlambdaCL_r2 tga aca ttc cgt agg ggc aac tg

Second antibody gene PCR VH for pADL20c plasmid
VH1fN TA CTC GCG GCC CAG CCG GCC ATG GCT cag gtb cag ctg gtg cag tct gg
VH1/7fN TA CTC GCG GCC CAG CCG GCC ATG GCT car rts cag ctg gtr car tct gg
VH2fN TA CTC GCG GCC CAG CCG GCC ATG GCT cag rtc acc ttg aag gag tct gg
VH3f1N TA CTC GCG GCC CAG CCG GCC ATG GCT sar gtg cag ctg gtg gag tct gg
VH3f2N TA CTC GCG GCC CAG CCG GCC ATG GCT gag gtg cag ctg ktg gag wcy sg
VH4f1N TA CTC GCG GCC CAG CCG GCC ATG GCT cag gtg car ctg cag gag tcg gg
VH4fN TA CTC GCG GCC CAG CCG GCC ATG GCT cag stg cag ctr cag sag tss gg
VH5fN TA CTC GCG GCC CAG CCG GCC ATG GCT gar gtg cag ctg gtg cag tct gg
VH6fN TA CTC GCG GCC CAG CCG GCC ATG GCT cag gta cag ctg cag cag tca gg

120

IgMscFvR GGA GCC GCC GCC GCC AGA ACC ACC ACC ACC AGA ACC ACC ACC ACC ggt tgg
ggc gga tgc act

IgGscFvR GGA GCC GCC GCC GCC AGA ACC ACC ACC ACC AGA ACC ACC ACC ACC gac
cga tgg gcc ctt ggt gga

IgEscFvR GGA GCC GCC GCC GCC AGA ACC ACC ACC ACC AGA ACC ACC ACC ACC tgg gct
ctg tgt gga gg

Second antibody gene PCR kappa
VK1Linkf1 GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC gac atc cag atg acc cag tct cc
VK1Linkf2 GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC gmc atc crg wtg acc cag tct cc
VK2Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC gat rtt gtg atg acy cag wct cc
VK3Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC gaa atw gtg wtg acr cag tct cc
VK4Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC gac atc gtg atg acc cag tct cc
VK5Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC gaa acg aca ctc acg cag tct cc
VK6Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC gaw rtt gtg mtg acw cag tct cc

scFvkappa_r TG GTG TTG GCC TCC CGG GCC ACT AGT gaa gac aga tgg tgc agc cac agt

Second antibody gene PCR lambda
VL1Linkf1 GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC cag tct gtg ctg act cag cca cc
VL1Linkf2 GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC cag tct gtg ytg acg cag ccg cc
VL2Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC cag tct gcc ctg act cag cct
VL3Linkf1 GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC tcc tat gwg ctg acw cag cca cc
VL3Linkf2 GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC tct tct gag ctg act cag gac cc
VL4Linkf1 GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC ctg cct gtg ctg act cag ccc
VL4Linkf2 GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC cag cyt gtg ctg act caa tcr yc
VL5Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC cag sct gtg ctg act cag cc
VL6Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC aat ttt atg ctg act cag ccc ca
VL7/8Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC cag rct gtg gtg acy cag gag cc
VL9/10Linkf GGC GGC GGC GGC TCC GGT GGT GGT GGA TCC cag scw gkg ctg act cag cca cc

scFvlambda_r TG GTG TTG GCC TCC CGG GCC ACT AGT aga gga sgg ygg gaa cag agt gac

Second antibody gene PCR VH for f3TR1 plasmid
F3VH1fN TA CTC GCG GCC CAC GCG GCC ATG GCT cag gtb cag ctg gtg cag tct gg
F3VH1/7fN TA CTC GCG GCC CAC GCG GCC ATG GCT car rts cag ctg gtr car tct gg
F3VH2fN TA CTC GCG GCC CAC GCG GCC ATG GCT cag rtc acc ttg aag gag tct gg
F3VH3f1N TA CTC GCG GCC CAC GCG GCC ATG GCT sar gtg cag ctg gtg gag tct gg
F3VH3f2N TA CTC GCG GCC CAC GCG GCC ATG GCT gag gtg cag ctg ktg gag wcy sg
F3VH4f1N TA CTC GCG GCC CAC GCG GCC ATG GCT cag gtg car ctg cag gag tcg gg
F3VH4fN TA CTC GCG GCC CAC GCG GCC ATG GCT cag stg cag ctr cag sag tss gg
F3VH5fN TA CTC GCG GCC CAC GCG GCC ATG GCT gar gtg cag ctg gtg cag tct gg
F3VH6fN TA CTC GCG GCC CAC GCG GCC ATG GCT cag gta cag ctg cag cag tca gg

121
Reverse primers for Vkappa and Vlambda for f3TR1 cloning
F3scFvkappa_r T GGT GTT GGC CTC AGC GGC ACT AGT gaa gac aga tgg tgc agc cac agt

F3scFvlambda_r T GGT GTT GGC CTC AGC GGC ACT AGT aga gga sgg ygg gaa cag agt gac

Forward and Reverse primers for scFv cloning into F3TR1
F3scFv-For TAC TCG CGG CCC ACG CGG CCA

F3scFv-Rev TGG TGT TGG CCT CAG CGG CAC T

Abstract (if available)

Abstract Current advances in medicine feature intensive attempts at antibody (Ab) engineering to target a diverse array of antigens (Ag). For in vitro screening and isolation of antibodies that specifically target a chosen Ag, a library of highly diverse antibodies must be created. The primary goal of this project was to generate human and llama monoclonal Abs in vitro by utilizing the unique mechanism of mutagenic diversification of antibody-encoded immunoglobulin (Ig) genes by activation-induced cytidine deaminase (AID) and error-prone polymerase eta (pol h). ❧ During an immune response, Abs with increased affinity toward Ags are generated in activated B-cells through mutagenic diversification processes of Ig genes. One of these processes is somatic hypermutation (SHM) of variable (V) regions of Ig genes. SHM is generated at G-C sites by enzyme activation induced deoxycytidine deaminase (AID) preferentially at WRC (W = A/T, R = A/G) hotspot motifs and at A-T sites by Polh at WA motifs. Affinity maturation occurs by the repetitive action of AID and error-prone polymerase eta (polh) during a series of cell divisions accompanied by clonal selection for high affinity Abs. This study describes how we developed highly diversified phage display Ab libraries by performing affinity maturation in vitro using purified human AID and polh. The scientific rationale for the application of this novel method in Ab discovery was based on the earlier studies and our recent findings showing that in vitro mutation spectrum of purified AID and polh on IgV gene resembles in vivo SHM mutation spectra on the same V region. Since AID and polh retain their catalytic specificities in vitro, we hypothesized that the diversified Ab library would mimic products of SHM in B-cells with the characteristic WRC (AID) and WA (pol h) hotspot mutations. ❧ This project applied the AID- and polh-diversified Ab libraries to isolate Abs specifically targeting GLP-1R, a G-protein coupled receptor protein that plays an essential role in insulin production during glucose metabolism, and artemin neurotrophic factor, which is implicated in human cold hypersensitivity. Here, we report the isolation and characterization of target-specific monoclonal Ab clones derived from in vitro affinity maturation of Ab- repertoires prior to exposure to Ag, and of individual Ab clones after enrichment in vitro via Ag.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

AID scanning & catalysis and the generation of high-affinity antibodies

Asset Metadata

Creator Jeong, Soo Lim (author)

Core Title Generation of monoclonal antibodies via phage display and in vitro affinity maturation using activation induced deoxycytidine deaminase and DNA polymerase eta

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Molecular Biology

Degree Conferral Date 2022-05

Publication Date 02/09/2022

Defense Date 12/10/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Ab discovery,activation induced deoxycytidine deaminase,affinity-matured antibodies,artemin,biopanning,error-prone DNA polymerase eta,glucagon-like peptide 1-receptor,immunoglobulin genes,in vitro affinity maturation,in vitro antibody diversification,monoclonal antibodies,mutation spectra,novel method in antibody discovery,OAI-PMH Harvest,phage display,scFv,somatic hypermutation,VHH

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Goodman, Myron (committee chair), Finkel, Steven (committee member), Lieber, Michael (committee member), McKemey, David (committee member)

Creator Email jsl3431@hotmail.com,soolimje@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC110702256

Unique identifier UC110702256

Legacy Identifier etd-JeongSooLi-10388

Document Type Dissertation

Format application/pdf (imt)

Rights Jeong, Soo Lim

Type texts

Source 20220214-usctheses-batch-912 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu