Production and validation of anti-HCV antibodies for viral neutralization

STOCKHOLM SWEDEN 2020,

Production and validation of anti- HCV antibodies for viral

neutralization

MSc Medical Biotechnology Degree Project MEGAN OLGA GARCÍA-MARK

KTH ROYAL INSTITUTE OF TECHNOLOGY

SCHOOL OF ENGINEERING SCIENCES IN CHEMISTRY, BIOTECHNOLOGY AND HEALTH

1. Table of Contents

2. Abstract 3. Key Words 4. Introduction

4.1. Hepatitis C Virus

4.2. Project Background and Objectives 4.3. Applying the Project to SARS-CoV-2 5. Materials and Methods

5.1. Plasmid Design & Cloning Strategy

5.2. Sub-cloning of Commercial Backbone Plasmids 5.3. Restriction Digestion & Ligation

5.3.1. Restriction Mapping for Plasmid Verification and to Generate the Cut for Cloning

5.4. Restriction Digestion of FAbs & Ligation 5.5. Transformation into E. coli

5.6. PCR Colony Screen

5.7. Preparing Candidate Colonies 5.8. Transfection of HEK cells 5.9. Harvesting

5.10. Western Blot 6. Results

6.1. Plasmid Design 6.1.1. IGBlast Results

6.1.2. The unique sequences (FAbs) for each mAb 6.1.3. Plasmid Insert Components

6.1.4. SignalP 3.0 for TPA Signal Sequence 6.1.5. Theoretical Plasmid

6.1.6. Sub-cloning of Commercial Backbone

6.1.7. Mapping for Plasmid Verification and to Generate the Cut for Cloning

6.2. PCR Colony Screen 6.3. Western Blot

7. Discussion

8. Acknowledgements 9. References

10. Appendix

2. Samanfattning

Hepatit-C (HCV) är fortsatt den enskilt största orsaken till levertransplantationer med uppskattningsvis 71 miljoner infekterade globalt sett, enligt världshälsoorganisationen (WHO).Ett vaccin mot HCV skulle drastiskt minska trycket på global hälso- och

sjukvård. Syftet med detta projekt är att producera antikroppar (igG) mot HCV. Projektet fokuserar på HEPC3, AR3C, HEPC74 och HCV1 som är monoklonala antikroppar (mAbs). Dessa antikroppsvarianter har isolerats från patienter som tillfrisknat från infektion. Både DNA-sekvenser och strukturer av antikropparna finns offentligt tillgängliga. Dessa fyra antikroppar har också visats kunna binda till E2 som är ett membranbundet glykoprotein hos HCV som är centralt för viral adhesion och fusion.

Interaktionen mellan dessa mAbs och E2 har visat sig neutralisera virulens, vilket gör dem till lovande kandidater för denna studie. Tre av fyra mAbs kunde klonas och produceras framgångsrikt. Försöket med HEPC74 misslyckades på grund av ett fel i plasmidsekvensen och just som western blot skulle genomföras för att bekräfta

sekretion av en alternativ klon avslutades the praktiska arbetet med anledning av Covid- 19 utbrottet. Resultaten visar entydigt att tre av fyra mAb producerades framgångsrikt.

Det går dock att anta att det andra försöket med HEPC74 sannolikt också lyckades pga perfekt sekventiell matchning. Då det huvudsakliga syftet med projektet var att framgångsrikt klona och producera dessa fyra antikroppar så kan studien anses vara framgångsrik. Slutligen så undersöktes huruvida samma förfarande kunde appliceras mot SARS-CoV-2 genom kloning och produktion av anti-RBD IgG och tester av viral neutralisering.

Abstract

Hepatitis-C Virus (HCV) remains the leading cause of liver transplant in the US and the UK, and the World Health Organization (WHO) estimates that 71 million people are infected worldwide. A vaccine would drastically impact the healthcare-associated burdens that HCV causes globally. The objective of this master’s thesis project is to produce human antibody (IgG) against HCV. This project will focus on the monoclonal antibodies (mAbs) HEPC3, AR3C, HEPC74, and HCV1. These four antibodies have been isolated from patients who have successfully cleared the infection, and their sequences and structures are available in the public domain. These four antibodies have also shown to bind to E2, a glycoprotein on the surface HCV that is crucial for viral binding and entry. This interaction of the mABs with E2 has been implicated in viral neutralization, making them promising choices for this study. Overall, 3 out of 4 mAbs were successfully cloned and produced. The unsuccessful antibody, HEPC74, was discovered to have failed due to an error in the plasmid sequence. Just as the western blot to confirm secretion was ready to be run, the laboratory closed due the Covid-19 outbreak. Therefore, the data can officially declare a ¾ mAb production

success, however it is safe to assume that the alternative clone for HEPC74 was also a success due to a perfect sequence match. Since the primary objective of this project was to successfully clone and produce these four antibodies, then this study is considered an overall success. Lastly, this study examined how the same protocol

could be applied the SARS-CoV-2 outbreak, by the cloning and production of anti-RBD IgG and testing them for viral neutralization.

3. Key Words

Cloning, antibody, infection, HCV, IgG production, molecular biology, cell factory, plasmid design, immunology

4. Introduction

4.1 Hepatitis C Virus

Despite the advancement of specific pharmaceuticals, there are 2-4 million new Hepatitis C Virus (HCV) infections per year (WHO 2019). The virus can cause symptoms ranging from acute and mild to a severe, chronic illness. The Hepatitis C Virus is blood-born, with modes of transmission including injection drug use, unsafe medical injection practices, transfusion of unscreened blood, and unsafe sexual practices. Globally, the World Health Organization (WHO) estimates that 71 million people are infected worldwide. A vaccine is available for Hepatitis A and B, leaving Hepatitis C the last major challenge in the fight against chronic infection of the liver.

Research for an HCV vaccine is extremely important, since a significant number of people with HCV develop cirrhosis or hepatocellular carcinoma (primary liver cancer).

Therefore, it is the HCV-associated diseases that leads to a significant global burden (i.e. liver transplant) on healthcare (WHO 2019).

According to the WHO, approximately 30% of people will spontaneously clear the

infection within 6 months without any treatment. The remaining 70% will develop chronic HCV infection (Ploss 2012). There is a wealth of published data containing monoclonal antibody (mAb) sequences isolated from HCV+ patients’ blood. Published structural data in the Protein Data Bank (PDB) has demonstrated that these antibodies bind to HCV’s structural glycoprotein E2 (Figure 1). E2 is on the surface of HCV, which has been implicated as a crucial protein for HCV binding and entry (Ploss 2012). E2 binds CD81 on host cells, implicating CD81 as an important host molecule for viral entry and subsequent infection (Ploss 2012). Current evidence suggests that antibodies targeting E2 are involved in the immune response that clears the infectionby blocking the E2:

CD81 interaction (Li 2015). Therefore, the working theory is that these monoclonal antibodies (mAbs) have successfully neutralized the virus within the patients of which they were developed, as supported by the E2: mAb crystal structure interaction (Figure 1). This makes E2 an attractive target for vaccine development, and the corresponding antibodies implicated in this interaction the key components for an effective immune response.

4.2 Project Background and Objectives

This project will focus on the mAbs HEPC3, AR3C, HEPC74, and HCV1. These four antibodies have been shown to bind to E2, implicating them as clinically relevant mAbs for this study(Figure 1). We can use the available sequence data on these mAbs to isolate the complementarity determining regions (CDRs), and synthesize the antigen binding fragments (FAbs), and clone them using commercial plasmids. The mechanism for an antibody-dependent immune response involves the CDRs, which contain a

Figure 1 The four monoclonal antibodies (mAbs) HEPC3, HCV1, HEPC74, and AR3C that were cloned in this study. Heavy chain= green, light chain= orange, blue= full E2 protein, and purple = contact residues involved in hydrogen bonding (red lines) with E2. Evidence suggests that this mAb: E2 interaction plays a role in the immune response and clears the infection (Stejskal 2020). A. HEPC3 + the full E2 protein B. HEPC3 + full E2 contact residues C. Close-up of HEPC3 + E2 contact residues interaction site D. HCV1 + E2 contact residues E.

HEPC74 + E2 contact residues F. AR3C+ E2 contact residues

unique sequence that binds to E2 in a highly specific manner (Figure 1). Although commercial antibodies against Hepatitis C are available for purchase, a more efficient and cost-effective and versatile approach would be to produce the antibodies in-house in mammalian cells. Therefore, this project entails the establishment of an effective protocol for antibody production via an in-house recombinant technique. Advantages of this approach include not relying on the extremely long wait times for custom-made antibodies (up to 6 months), the high cost associated with purchasing them, the ability to test several different antibodies at various times, the potential to co-express them for structural studies (i.e. crystallization), and avoiding the consequential costs and wait times associated with a failed antibody neutralization of the virus. Dr Joe Grove of University College London has been studying Hepatitis C for over a decade and is determined to contribute to the search for a plausible vaccine. By having a steady bank of antibodies available for experiments, the ability to discard and test new sequences at will with little to no financial or loss of large amounts of time makes this an efficient approach to bioproduction and an excellent opportunity for a medical biotechnology student.

Firstly, I will design the plasmids that encode for all the elements necessary to express the antibodies of interest (Figure 2). The plasmid insert includes the antibodies’ unique hypervariable sequences (CDRs), the constant region (Fc), and other relevant

sequences that ensure the effective transcription (i.e. promoter), translation (i.e. KOZAK sequence), and secretion of antibody (i.e. TPA signal sequence). This plasmid insert was sent to Life Technologies™ for synthesis. Due to the distinct genetic origins of the

heavy and light chain during the generation of antibody diversity (Alberts 1970), two distinct commercial backbones, one for the heavy chain and one for the light chain, will be ordered. The commercial backbone that was utilized

corresponds to HuIgG1, or human IgG. This is because IgG is the antibody class associated with long term protection against viruses and is the antibody class choice in immunological studies (ThermoFisher 2020).

Next, the restriction digestion of the circular commercial plasmid and ligation to the FAbs for the full, correct plasmid (Figure 3) will be the first

key step in the lab. The plasmid will then be transformed into the sub-cloning expression host Escherichia coli. Successful clones will then be screened and selected for further growth, allowing for a significant scale-up of the amount of available plasmid. These plasmids, after passing various quality control steps (i.e. PCR colony screen and sequencing) will then be transfected into the IgG expression host human embryonic kidney (HEK293T) cells, effectively establishing cell factories for IgG production (Figure 3). The antibodies produced could then be tested for their biological activity i.e. neutralization against HCV

Figure 2 The desired plasmid construct with all the components for successful mAb secretion.

pseudoparticles (HCVpp), and/or recognition of viral glycoprotein E2 via ELISA. This testing of the cloned IgG’s biological activity was not possible due to the COVID-19 pandemic which led to the closure of the laboratory before biological activity tests could be performed. However, this pipeline can allow researchers to produce various mAbs simultaneously for future biological activity experiments.

4.3 Applying the Project to SARS-CoV-2

Finally, due to this project being completed during the midst of a coronavirus pandemic, known as COVID-19 or SARS-CoV-2, this project will also demonstrate how the same protocol can be applied to SARS-CoV-2 research. As with the antibodies against HCV, the sequences for antibodies isolated from Covid-19-infected patients are becoming rapidly available (Ju 2020). The viral entry of SARS-CoV-2 depends on the binding of the viral spike protein receptor-binding domain (RBD) with the angiotensin converting enzyme 2 (ACE2) target receptor (Wang2020). Angiotensin II, the most important component of the renin-angiotensin (RAS) system, promotes atherosclerosis (Wang 2020). ACE2 in turn degrades the pro-atherosclerotic angiotensin II and converts it to anti-atherosclerotic angiotensin 1-7. The establishment of ACE2’s important role in protecting people against atherosclerosis suggests that blocking ACE2 as part of a novel coronavirus treatment may lead to undesirable side effects associated with disruptions in the RAS system. Therefore, anti-RBD antibodies pose a more favorable

Figure 3 The pipeline for the synthesis, cloning, and detection of the 4 antibodies of interest.

Asterisks indicate where QC was performed.

option to treatment or the development of a vaccine (Ju 2020). This is analogous to strategy against HCV which involves antibody-mediated blocking of the E2: CD81 interaction.

The potential application of this project’s objective to the coronavirus outbreak would include obtaining the anti-SARS-CoV-2 antibody sequences for both the heavy and light chains. Then, a plasmid would be designed in Serial Cloner that encodes all aspects necessary for the transcription, translation, and secretion of these antibodies; identical to the steps outlined for antibodies against HCV (Figure 2). Then, just as outlined in the pipeline in Figure 3, the plasmid would be transformed into E. coli to bulk-up the

plasmid, the plasmid will be purified, and transfected into a mammalian cell line for IgG production. Finally, the anti-SARS-CoV-2 antibodies would then undergo various biological activity assays and possible structural analysis to further understand and characterize the immune response against SARS-CoV-2.

Overall, this pipeline encompasses plasmid design, plasmid construction, cloning, cell culturing, virology, and hands-on immunology. It is a promising opportunity to apply the skills taught within the MSc in medical biotechnology at KTH.

5. Materials and Methods

5.1.1. Plasmid Design & Cloning Strategy

The antibodies that will be produced are: AR3C, HEPC3, HCV1, and HEPC74. There will be one heavy chain plasmid and one light chain plasmid for each antibody, for a total of 8 plasmids. This is to ensure maximum efficiency, as plasmids that are too large, or with too many components, increase chances of error during transfection and IgG production.

The first step was to download the antibody crystal structure from the Protein Data Bank (PDB) and open the sequence in UGene. Then the sequences pertaining to the heavy and light chain were extracted. The heavy and light chain amino acid sequences for each antibody were ran in IgBLAST. This step is necessary in order to be certain of the genetic segment origin of the chain. For example, light chains are produced either via a kappa or lambda light chain gene segment during the generation of antibody diversity (Alberts 1970). Therefore, by running the heavy and light chain sequences through IGBlast, we receive the corresponding genotypic origins, allowing for the selection of the correct kappa or lambda commercial backbone.

The software that was used for plasmid design is Serial Cloner. The heavy and light chain sequences for each antibody were pasted into a unique sequence window. Next, I performed an alignment in Serial Cloner of each heavy and light chain sequences to a reference. Although the sequences downloaded from PDB should correlate to the unique hypervariable regions, there remains residual sequences from the constant

regions. These Fc sequences must be removed to ensure that when ligated to the commercial backbone containing the Fc region, there is no overlap or introductions of mutations (frameshift, early stop codon, etc.). The alignment allows for the

determination of when the CDRs end and the Fc region begins for the correct nucleotide deletions in the sequence window. Finally, the other components required for successful production and secretion of antibody (i.e. signal sequence, Kozak sequence, start

codon, etc.) are added to the sequence window. The final construct was then codon optimized using Integrated DNA Technology™ Codon Optimization Tool.

The plasmid insert sequence window therefore contains all aspects needed to encode for antibody, except the Fc region which will be encoded in the commercial backbone.

These plasmid insert constructs were sent to Life Technologies™ for synthesis. The commercial backbone containing the Fc region and Kanamycin resistance cassette was purchased from Oxford Genetics™. Plasmid design is demonstrated in detail in Section 6.1.3. Plasmid Insert Components.

The plasmid maps and sequences provided by Oxford Genetics™ for the heavy, light (κ or ƛ) allowed for the selection of restriction sites for a harmonious ligation of the final constructs to the commercial backbone. Unique restriction enzyme sites are desirable (over the double sites) to ensure a cut at just one site and are visualized in Serial Cloner by Find Restriction Site.

The cloning strategy that was chosen was Restriction Cloning (Addgene 2016). This entails cutting open the plasmid backbone and inserting a linear fragment of DNA, known as the plasmid insert, which has been cut by restriction enzymes for compatible ends (Figure 4). DNA ligase then ligates the two segments together, resulting in the complete circular plasmid. This plasmid can then be transfected into live cells and is easily maintained within these biological systems.

5.2. Sub-cloning of Commercial Backbone Plasmids

The objective is to “bulk-up” the number of commercial plasmids from Oxford

Genetics™, in order to have plenty available in case there are problems with the ligation

Figure 4 The restriction cloning method taken from Addgene. The plasmid backbone (blue) and plasmid insert (green) are cut by restriction enzymes that were chosen in Serial Cloner in order to generate compatible ends whose overhang allows for binding. DNA ligase then ligates the plasmid insert and backbone together.

with the plasmid inserts. Agar plates were poured using the protocol given by the external supervisor.

Sub-cloning was achieved via the NEB5a (sub-cloning efficiency) transformation protocol (New England Biolabs) using sub competent E. coli. Various dilutions of DNA (i.e. 1:10, 1:100, 1:1000, etc.) were used in order to maximize distinct colonies on the plates for effective selection of colonies.

After 24 hours in the incubator, 1 colony was chosen from each plate and placed in starter culture of 5mL to grow throughout the day. Then, the 5mL cultures were added to large tissue culture (TC) flasks for overnight culture in the shaking incubator at 37°C.

The following morning, the DNA corresponding to the commercial plasmids was extracted via MidiPrep. The QIAGEN® MidiPrep Protocol (QIAGEN 2020) for Plasmid Purification was used.

The concentration of DNA as well as the A260/280 were determined via Nanodrop and recorded. The now amplified and pure plasmids were stored in the refrigerator for future experiments.

5.3. Restriction Digestion & Ligation

5.3.1. Restriction Mapping for Plasmid Verification and to Generate the Cut for Cloning

Although the commercial backbones have been increased in number with acceptable concentrations and purity, they must be linearized with their identity confirmed before ligation to the plasmid insert. Thus, restriction mapping was used for plasmid

verification.

Restriction enzymes were chosen by analyzing the restriction sites present on the plasmid maps given by Oxford Genetics™ (Appendix). Then, theoretical cuts using various combinations of the chosen restriction sites in Serial Cloner’s Virtual Cutter was completed. The objective was to ensure that the pattern of commercial backbone

fragments produced in Virtual Cutter can be used to determine its correct identity (to rule out any error in sub-cloning), as well as to extract the desired linearized backbone segment for subsequent ligation.

Two digests were performed, each containing a pair of restriction enzymes to produce the desired pattern of generated fragments. Table 1 describes incubation with the restriction enzymes BamHI + SacI. Table 2 describes incubation with PvuII + HindIII.

Samples given identifying numbers 1 or 2, depending on the enzyme pair used. The letters A, B, and C in sample names refer to the Heavy, Kappa, and Lambda Chains, respectively.

The buffers were vortexed before use, and the restriction enzymes were added last.

When all components were added, the microcentrifuge tubes were briefly vortexed and spun. All six reactions were placed in the heat box at 37°C for one hour.

All six reactions were then run in the gel electrophoresis to visualize the fragments generated. The theoretical fragment generated in Virtual Cutter was confirmed with the fragment visualized via gel electrophoresis. The gel was placed under UV light in order to cut out the band with a scalpel.

Finally, a DNA gel extraction was performed following the Monarch® DNA Gel Extraction Kit Protocol Card (New England Biolabs). This excised the linearized commercial

backbone.

5.4. Restriction Digestion of FAbs & Ligation Restriction Digestion of FAbs

1. BamHI + SacI 1A 1B 1C

Component Heavy Kappa Lambda

[DNA] 1474.8 ng/ µL 1950.7 ng/ µL 2240.0 ng/ µL

Buffer 1.5µL 1.5µL 1.5µL

DNA ~ 2µg 1.4µL 1.1µL 1.0µL

SacI 1.5µL 1.5µL 1.5µL

BamHI 1.5µL 1.5µL 1.5µL

H20 to have 15µL 9.1µL 9.4µL 9.5µL

2. PvuII + HindIII 2A 2B 2C

Component Heavy Kappa Lambda

[DNA] 1474.8 ng/µL 1950.7 ng/ µL 2240.0 ng/ µL

Buffer 1.5µL 1.5µL 1.5µL

DNA ~ 2µg 2µL 1.5µL 1.3µL

PvuII 2µL 2µL 2µL

BamHI 2µL 2µL 2µL

H20 to have 15µL 7.5µL 8µL 8µL

Table 1 The restriction digest reaction using restriction enzyme pair BamHI + SacI. 1.= incubation with BamHI + SacI and A=Heavy Chain, B=Kappa Chain, C=Lambda Chain.

Table 2 The restriction digest reaction using restriction enzyme pair PvuII + HindIII. 2. = incubation with PvuII + HindIII and A=Heavy Chain, B=Kappa Chain, C=Lambda Chain.

In order to prepare the FAbs for restriction digestion and ligation, they must first be reconstituted because they arrive in a lyophilized state. All FAbs were reconstituted by adding sufficient dH20 to have a concentration of 50ng/µL for each FAb (Table 3).

FAb Mass (ng) Amount of dH²0 added (µL) to reach [50ng/

µL]

AR3C-H 1650 33

AR3C-L 1700 34

HCV1-H 1950 39

HCV1-L 1750 35

HEPC3-H 1900 38

HEPC3-L 1850 37

HEPC74-H 1300 26

HEPC74-L 1500 30

A restriction digest was carried out on all FAbs in order to create compatible ends for subsequent ligation (Table 4). The restriction enzymes PvuII and HindIII were used to create compatible ends with the plasmid insert.

After the restriction digest, DNA cleanup followed using the Monarch® DNA Cleanup Protocol (New England Biolabs), and the concentration of FAbs measured via Nanodrop.

The Ligation Reaction

The required mass of the plasmid insert was calculated for the molar insert: vector ratio of 1:3, backbone: insert, using NEB Ligation Calculator. The formula used by the

Ligation Calculator is:

required mass insert (g) = desired insert to vector molar ratio x mass of vector (g) x ratio of insert to vector lengths

Component x1 rxn (

x8 rxn (

x1.1 (

Buffer 2 16 17.6

H

0 7 56 61.6

HindIII 0.5 4 4.4

PvuII 0.5 4 4.4

Then add 10

Table 3 The amount of dH20 added (µL) to reconstitute the FAbs to 50ng/ µL.

Table 4 The restriction digest master mix used to digest all FAbs to crate compatible ends for ligation.

The amounts of components were multiplied by 1.1 to allow for sufficient amount in case of pipetting error.

Before starting, a 1:20 dilution of all FAbs was performed. DNA ligase was the enzyme used to ligate the plasmid insert to the commercial backbone.

The ligation reaction was performed on all FAbs.

FAbs AR3 C-H

AR3C -L

HEPC3 -H

HEPC3 -L

HEPC74 -H

HEPC74 -L

HCV1 -H

HCV1 -L Buffer

(µL)

10 10 10 10 10 10 10 10

Vector (µL)

1 1.3 `1 1.34 1 1.34 1 1.34

Insert (µL)

2.5 2.4 1.8 1.7 3.4 2.3 1.7 1.8

H20(µL) 5.5 5.3 6.2 5.96 4.6 5.36 1.3 5.86 Enzyme

(µL)

1 1 1 1 1 1 1 1

For a total of 20 µL each

5.5. Transformation into E. Coli

The expression host used in this transformation were NEB 5-alpha competent Escherichia. coli bacterial cells. The NEB High Efficiency Transformation Protocol (C2992) was used for optimal transformation. Various dilutions of the mixture were utilized for plating (i.e. 1:10, 1:100. 1:100, etc.) in order to maximize distinct single colonies and minimize lawn growth.

5.6. PCR Colony Screen

After incubation at 30°C for 24 hours post-transformation, the PCR Colony Screen was performed. Suitable colonies are defined as not touching other colonies and large enough to ensure clonal homogeneity. First, suitable colonies from each transformation plate were selected via a wooden toothpick and streaked onto a grid-labelled plate. By streaking several colonies from each transformation plate onto their numbered grid- labelled plates, this increases the chance of encountering a positive clone. A positive clone is defined as a clone demonstrating fragments of desired size as predicted by the theoretical plasmids.

A PCR Colony Screen Heavy Chain Master Mix as well as Light Chain Master Mix was made (Table 6). This is because the primers for the PCR reaction were identical for the forward primer, but the reverse primers were specific to either heavy or light chain. The number of reactions performed depended on the number of colonies selected for each mAb. OneTaq Quick-Load 2x Master Mix was provided by the laboratory, which

Table 5 The ligation reaction for each Fab calculated at a ratio of 1:3 backbone: insert.

contains the DNA polymerase and the dye for gel visualization. The reaction proceeded in the PCR Thermal Cycler and was then loaded for gel electrophoresis and visualized.

Positive clones were allowed to progress to the next phase, whereas negative clones underwent troubleshooting such as a repeated digest and ligation under different insert:

vector ratios, transformation at different dilutions, and Sanger sequencing.

5.7. Preparing Candidate Colonies

Positive clones determined via the PCR colony screen were prepared for the next phase. Positive clones were selected from the grid-labelled plates and grown in overnight cultures of 100mL Terrific Broth with 100µL Kanamycin antibiotic. Only plasmids that contain the vector encoding for Kanamycin can grow in the culture.

The following day, The QIAGEN^® MiniPrep Protocol (QIAGEN)was used to extract and purify the assumed desired plasmids. The concentration and purity were measured via Nanodrop. A sample of each plasmid was sent to Eurofins Genomics for sequencing.

The sequencing data confirms if the clones are truly positive prior to transfection.

Clones that pass the sequencing test were then transfected.

5.8. Transfection of HEK cells

Human endothelial kidney (HEK) cells were transfected with positive clones who passed QC. HEK cells are essentially the “golden standard” human expression system because they are easy to grow, transfect, and they produce high yields of proteins (Baldi 2008).

Each well of HEK cells was transfected with one heavy chain plasmid and one light chain plasmid for each mAb. Controls included a well with only the heavy chain of a known correct plasmid (AR3C), a well with only the light chain of a known correct plasmid, and a well with no plasmid. The mAb AR3C was used as a positive control because it successfully produced IgG by passing all QC steps and was brought forward in the pipeline before all other mABs. Two separate positive clones for AR3C, AR3CH-5 and AR3CL-13, were included again (Table 7) in order to verify the transfection protocol and provide a reference for the remaining mABs.

Component x1 rxn (µL) Multiply x1 rxn by the

number of desired reaction x 1.1.

15 µL of mix per PCR tube

Forward Primer (10µM) 0.5 Heavy or Light Chain

Reverse Primer (10µM)

0.5

DNA Single colony

OneTaq Quick-Load 2x Master Mix

12.5

dH²0 1.5

Table 6 The PCR Colony Screen Method. A master mix is created, placed into each PCR tube, and a single colony is placed in each PCR tube. Several colonies from each plate were selected to increase the chances of finding a positive clone.

First, 4mL of media from each well was removed and replaced with 2mL of fresh media.

The wells were labeled as in Table 8, and 100µL of transfection mix (Table 9) was added dropwise and clockwise to the well. The cells were gently swirled and placed in an incubator for ~72 hours.

Plasmids Plasmid Code Concentration (ng/ul)

5x dilution

HEPC74H-5 A 1121.2 224.24

HEPC74L-7 B 767.17 153.4

AR3CH-5 C 477 95.4

AR3CL-13 D 821.4 164.28

HCV1H-11 E 790 158

HCV1L-3 F 1090.7 218.14

HEPC3H-3 G 1259.5 251.9

HEPC3L-3 H 1536.9 307.4

Well 1 A+B HEPC74

Well 2 C+D AR3C

Well 3 E+F HCV1

Well 4 G+H HEPC3H

Well 5 C only AR3C Heavy Chain

Well 6 D only AR3C Light Chain

Well 7 No plasmid N/A

6 well 1 2 3 4 5 6 7

Plasmid H (300ng)

1.3 3.1 1.9 1.2 3.1 N/A N/A Plasmid L

(300ng)

2.0 1.8 1.4 1.0 N/A 1.8 N/A Water 16.7 15.0 16.7 17.8 16.9 18.2 20 FuGENE

(3µl/µg)

1.8 1.8 1.8 1.8 0.9 0.9 0.9

Optimem 78.2 78.2 78.2 78.2 79.1 79.1 79.1

Table 7 The plasmids, plasmid code, concentration after bulking-up and purifying, and the 5x dilution used in the procedure. The plasmid names end in H (heavy chain) or L (light chain) and the following number corresponds to the clone number designated at the PCR colony screen.

Table 8 The plasmids that were transfected in each well, and the corresponding mAB that the HEK cells should be producing.

Table 9 The calculated amounts for all transfection components in order to transfect HEK cells with the desired plasmids. The plasmids transfected into each well can be decoded in Table 8.

5.9. Harvesting

After 72 hours post-transfection, the HEK cells were harvested for their supernatant and whole cell lysate (WCL). The supernatant is defined as the culture medium, where secreted proteins are to be found. 5mL of media was removed via a syringe and inserted into the Millex syringe filter. This filter sterilizes cell culture reagents (i.e.

antibiotics, culture media, impurities, etc.). The supernatant was then filtered by the Millex filter into a labelled tube. It was imperative to separate the supernatant from the WCL, since that will give insight as to what is occurring in the cell. For example, if antibody is seen in the WCL but not the supernatant, this would indicate an issue with secretion.

In order to harvest the WCL, each well containing the transfected HEKs was washed with 500mL PBS and then the PBS was removed. 250mL RIPA lysis buffer + protease inhibitor was added into each well and incubated on ice for 5 minutes. Harvest lysates were transferred into 0.5ml Eppendorf tubes via a pipette boy and stored at -20°C.

5.10. Western Blot

The western blot is arguably the most crucial milestone in the project, since it indicates the presence of IgG in the supernatant and WCL. A wet system for western blot was used, with TruPage loaded into the middle compartment. 6µL of the sample ladder (color protein standard broad range) and 10µL of the sample was loaded. A space between each well was desired for clear visualization, therefore 6µL Takasuki loading buffer (without reducing agent) was loaded in the spaces between each sample.

The voltage was set to 110V for one hour at room temperature. Filters were then soaked in the transfer buffer (350mL dH20 + 100mL methanol + 50mL 10x Tris/Gly).

The membranes were cut and added under the filter until completely soaked. The membranes and filters were placed in the Trans-Blot Turbo. The Trans-Blot Turbo was run twice at 7 minutes each. The membrane was blocked with 2% milk PBS Tween for one hour on a rocker. Then, the membrane was incubated with 1% milk (~10mL) in a 50mL tube. Meanwhile, the secondary antibody solution was made by combining 20mL 1% milk with 2µL α-HuIgG (donkey anti-human IgG) and performing a 1/10,000 dilution.

Since the target of the western blot is an antibody, a primary antibody was not

necessary. The secondary antibody solution was added to the tube with the membrane and put on a roller mixer for one hour. The membrane was then washed with PBS- Tween (~10mL) three times for 10 minutes. Finally, the membrane was visualized in the Bio-Rad ChemiDoc Imaging System to detect the cloned IgG. Western blots were completed for both the supernatant and WCL.

6. Results

6.1. Plasmid Design

6.1.1. IGBlast Results

The purpose of the IGBlast is to determine whether the heavy and light chain for each antibody is of heavy, kappa, or lambda origin in order to order the correct commercial backbone for ligation. Table 10 demonstrates the results of the IGBlast thus

categorizing the genetic origin of the heavy and light chains of each mAb. As seen in Table 10, all the light chains for the 4 mAbs in this study are of kappa (rather than lambda) origin, and the heavy chains are of heavy chain origin. The latter may seem obvious; however, it was important to verify that the published data was correct. Table 10 allowed for the purchase of the correct commercial backbone plasmid with

confidence.

Antibody Name (PDB ID)

Chain Genotype

Genotypic Origin

IMGT Gene Name

HepC3 (6mei) Chain A IGHV1-69*01 Heavy Homo sapiens IGHV1-69 HepC3 (6mei) Chain B IGKV1-39*01 Kappa Homo sapiens

IGKV1-39 HCV1 (4dgy) Chain H IGHV3-33*01 Heavy Homo sapiens

IGHV3-33 HCV1 (4dgy) Chain L IGKV3-11*01 Kappa Homo sapiens

IGKV3-11 HepC74 (6MEE) Chain A IGHV1-69*01 Heavy Homo sapiens

IGHV1-69 HepC74 (6MEE) Chain B IGKV1-5*03 Kappa Homo sapiens

IGKV1-5 AR3C (4MWF) Chain H IGHV1-69*06 Heavy Homo sapiens

IGHV1-69 AR3C (4MWF)

Chain L IGKV3-15*01 Kappa Homo sapiens IGKV3-15

Table 10 IGBlast Results for the determination of genotypic origin (heavy, kappa, or lambda) of the 4 antibodies of interest. Chains A and B correspond to heavy and light chains.

6.1.2. The unique sequences (CDRs) for each mAb

The sequences shown in Table 11 correspond to the unique hypervariable regions, known as CDRs. for the heavy and light chain of each antibody. These sequences are the result of extracting the sequences from PDB, removal of residual lateral

sequences, and codon optimization.

Antibody Name

Codon Optimized Nucleotide Sequence Amino Acid Sequence HCV1-H

GGGAGGAGCCTGAGGCTCAGTTGCACCGCTTCTGGTTTCACC TTCAACAACTATGGTATGCACTGGGTAAGACAGACCCCAGGT AAGGGGCTCGAGTGGTTGGCAGTGATCTGGTTCGATGAGAAC AATAAGTACTATGCTGACTCAGTCAGGGGGAGGTTTACAATC TCTCGAGACAACTCAAAAAATACTCTTTTCCTGCAAATGAAC AGCCTTAAAACAGAGGACACTGCTATGTACTATTGCGCTCGA GACATAAGCCTTGTCCGGGATGCTTTTATATATTTCGATTTC TGGGGCTTGGGGACGCTCGTCACAGTAAGTTCA

EVQLLESGGGVVQPGRSLR LSCTASGFTFNNYGMHWVR QTPGKGLEWLAVIWFDENN KYYADSVRGRFTISRDNSK NTLFLQMNSLKTEDTAMYY CARDISLVRDAFIYFDFWG LGTLVTVSS

HCV1-L

ELTLTQSPATLSLSPGERA TLSCRASQSVSSYLAWYQQ KPGQAPRLLIYDASNRATG IPARFSGSGSGTDFTLTIS SLEPEDFAVYYCQQRSNWI TFGQGTRLEIK

HEPC3-H

QVQLVQSGAEVKKPGSSVK VSCKASGGTLNSYEITWVR QAPGQGLEWMGGITPIFET TYAQKFQGRVTITADESTS TTYMELSSLRPEDTAVYYC ARDGVRYCGGGRCYNWFDP WGQGTLVTVSS

HEPC3-L

DIQMTQSPSSLSASVGDRV TITCRAGQNINNYLNWYQQ KPGKAPKVLIYAASNLQSG VPSRFSGSGSGTDFTLTIS SLQPEDFATYYCQQSHSTV RTFGQGTKVEIK

HEPC74- H

GTCCAACTGGTTCAGTCTGGAGCCGAGGTCAAAAAGCCGGGGAGCTCCGTA AAGGTCAGTTGCACTACCTCAGGTAATTATGCGATTTCCTGGGTTCGGCAA GCTCCAGGTCAGGGGCTCGAGTGGGTTGGCGGTATGAGCCCAATCTCCAAC ACACCGAAGTACGCGCAGAAGTTTCAAGGGAGGGTTACCATCACAGCAGAC GAGTCTACATCAACAACGTACATGGAACTGAGTAGTTTGAGACCGGAGGAC ACGGCAGTGTATTATTGCGCTAGGGACCTTCTGAAATATTGTGGGGGAGGC AACTGTCATTCTCTTCTGGTAGACCCGTGGGGGCAAGGTACTTTGGTTACG GTGTCATCT

VQLVQSGAEVKKPGSSVKV SCTTSGNYAISWVRQAPGQ GLEWVGGMSPISNTPKYAQ KFQGRVTITADESTSTTYM ELSSLRPEDTAVYYCARDL LKYCGGGNCHSLLVDPWGQ GTLVTVSS

HEPC74- L

GACATCGTGATGACCCAGAGCCCCAGCACCCTGAGCGCCAGCGTGGGCGAC CGCGTGACCATCTCGTGCCGCGCCAGCCAGAGCATCAGCTCGTGGCTGGCC TGGTACCAGCAGAAGCCCGGCCGCGCCCCCAAGCTGCTGATCTACAAGGCC AGCAGCCTGGAGACCGGCGTGCCCAGCCGCTTCAGCGGCAGCGGCAGCGGC ACCGAGTTCACCCTGACCATCAGCAGCCTGCAGCCCGACGACTTCGCCACC TACTACTGCCAGCACTACAACACCTACCTGTTCACCTTCGGCCCCGGCACC AAGGTGGACCTGAAG

DIVMTQSPSTLSASVGDRV TISCRASQSISSWLAWYQQ KPGRAPKLLIYKASSLETG VPSRFSGSGSGTEFTLTIS SLQPDDFATYYCQHYNTYL FTFGPGTKVDLK

AR3C-H

EVQLLEQSGAEVKKPGSSV KVSCETSGGTFDNYALNWV RQAPGQGLEWIGGVVPLFG TTRNAQKFQGRVTISDDKS TGTGHMELRSLRSEDTAVY YCVRSVTPRYCGGGFCYGE FDYWGQGTLVTVSS

AR3C-L

EIELTQSPATLSVSPGERA TLSCRASQSVSSNLAWYQQ KPGQAPRLLIYGASTRATG IPARFSGSGSGTEFTLTVS RLEPEDSAVYFCQQYYRSP LTFGGGTKVEIK

6.1.3. Plasmid Insert Components

The plasmid insert components were synthesized by Life Technologies™. Since they are in linear form, they must be ligated to the commercial backbone to form the fully constructed, circular plasmid. The plasmid insert components that are essential for the successful transcription, translation, and secretion of antibody are listed in Table 12.

Component Function Nucleotide sequence

Codon Optimized Amino Acid

Sequence

HindIII site Restriction site for compatible ends

AAGCTT Untranslated

Kozak Sequence

Initiation of translation in vertebrates, known as Shine-Dalgarno sequence in prokaryotes (Kozak 1986)

AGGAGGTACCCAC C

Untranslated

Start Codon Initiation of translation on the ribosome (Lind 2016)

ATG M

Tissue

Plasminogen Activator (TpA) Signal

Directs protein to the cellular secretion pathway leading to higher expression and secretion of protein

GACGCCATGAAGC GGGGCCTGTGCTG CGTGCTGCTGCTG TGTGGCGCCGTGT

DAMKRGLCC VLLLCGAVFV DSVTG

Table 11 The CDRs for each monoclonal antibody: each nucleotide sequence encodes for the hypervariable regions or Complementarity Determining Regions (CDRs) for each mAb of whose CDRs engage E2 of HCV for viral

neutralization. H= heavy chain and L=light chain.

Sequence (Kou 2017) TCGTGGATAGCGT GACCGGT

Component Function Nucleotide sequence

Codon Optimized Amino Acid

Sequence

FAbs Encode for each unique mAb

Please see Table 11

Please see Table 11

PvuII Restriction site for compatible ends

CAGCTG Untranslated

Fc Overlap Due to the location of PvuII within the Fc region, the first 80 nucleotides of the Fc region are included in the insert to ensure perfect ligation after restriction digestion

GCCAGCACCAAGG GACCTAGCGTGTT CCCTCTGGCCCCT TCCTCCAAGAGCA CCAGCGGCGGAAC AGCTGCCCTGGGA TG

*includes PvuII site

ASTKGPSVFPLAPSSK STSGGTAALGX

*includes PvuII site

The complete plasmid insert results from the sum of the plasmid insert components in Table 12. For example, this can be visualized by the sequence window encoding the plasmid insert for mAb AR3C in Figure 5 below. This can also be seen as a graphic map via the Graphic Map feature in Serial Cloner. Each plasmid insert component is color-coded in the sequence window as well as in the graphic map (Figure 5).

Table 12 The linear DNA insert to be ligated to the commercial backbone via restriction cloning. The unique hypervariable sequences encoding each unique mAb can be found in Table 11.

Figure 5 An example of a complete plasmid insert with all of its components color-coded and ready for ligation to the commercial backbone. This plasmid insert encodes for the FAb heavy chain of mAb AR3C.

6.1.4. SignalP 3.0 for TPA Signal Sequence

The Technical University of Denmark provides the SignalP 3.0 Server, which predicts cleavage sites in amino acid sequences and the resulting peptide via a combination of neural networks and hidden Markov Models. In this case, the cleavage site of interest is the signal peptidase in the ER. This server provides the prediction of peptides based on the signal sequence, in this case, the Tissue Plasminogen Activator (TPA) signal

sequence. Since the TPA signal sequence gets cleaved after successfully directing the transport of the protein to the Endoplasmic Reticulum, we can use SignalP to ensure that this cleavage occurs in the correct location, resulting in the correct peptide. The signal sequence is essential for all proteins destined to be secreted, and the SignalP 3.0 Server provides a quality control step prior to sending the Plasmid Insert to Life

Technologies™ for synthesis. Peaks that surpass the score of 0.5 on the y-axis, also visualized as a pink horizontal line, are assumed a true cleavage signal. The SignalP results for the heavy and light chain plasmid insert for HEPC74 are shown in Figure 6.

As seen in Figure 6, the plasmid insert encoding the heavy chain for the mAb HEPC74 demonstrates a true cleavage signal at position 26 beginning at the amino acid codon VQL.

Figure 6 The SignalP prediction for the plasmid insert encoding HEPC74’s heavy chain demonstrates cleavage at the correct site.

When the position of this cleavage is compared to the sequence of HEPC74-H in Serial Cloner, it appears directly after the TPA signal sequence as intended. The TPA signal sequence can be seen as the pink sequence prior to the highlighted nucleotides in Figure 7. This indicates that the TPA signal sequence should be cleaved at the correct site, thus allowing the nascent protein to be fed into the ER and successfully destined for secretion. This was performed for the heavy and light chain for each antibody. Since the SignalP outputs for all mAbs look quite similar, the remaining will be provided in the Appendix.

6.1.5. Theoretical Plasmids

The theoretical plasmid describes the intended plasmid sequence based on the cloning strategy whose electronic sequence is used as a reference to the final plasmid

construct made in the lab (Figure 8). The final plasmid construct, generated by the successful ligation of the plasmid insert with the commercial backbone, should match the theoretical plasmid sequence prior to proceeding to transfection. It also provides a convenient reference for intended fragment size when performing a digest for quality

Figure 7 The SignalP prediction for cleavage at the correct site is verified by matching the predicted cleavage site from Signal P to the intended cleavage site in Serial Cloner. The TPA signal sequence is shown in pink, the proceeding teal sequence is HEPC74-H’s CDRs, and the highlighted portion indicates where SignalP predicted the nascent protein to begin post-cleavage. This is a correct match.

control purposes (Appendix). An example of the annotated theoretical plasmid encoding for the mAb AR3C is shown in Figure 8.

The remainder of the theoretical plasmids and their sequences for each mAb heavy and light chain can be found in the Appendix.

6.1.6. Sub-cloning of Commercial Backbone Plasmids

The purpose of sub-cloning was to increase the availability of commercial backbone available for future experiments. The concentrations obtained after DNA extraction for plasmid purification are well within the concentration expectations. The concentrations listed in Table 13 provide an excellent amount of commercial plasmid for future

experiments. The A260/280, an indicator of DNA purity, is also well within the

acceptable range. Although none of the 4 antibodies of interest are derived from the lambda locus, the lambda commercial backbone was ordered prior to deciding on which mAbs to clone. Since future mAbs studies will most likely be carried out in the Grove Lab, the lambda backbone was still sub-cloned for future experiments.

Figure 8 The Theoretical Plasmid for the heavy chain of the mAb AR3C. This represents the desired final plasmid construct that is to be transfected into HEK cells for mAb secretion if it matches the theoretical plasmid sequence.

Commercial Backbone

Common Name [DNA] ng/µL A260/280

OGS527 IgG Heavy Chain 1474.8 1.86

OGS528 IgG Kappa Light Chain

1950.7 1.84

OGS529 IgG Lambda Light Chain

224.0 1.76

6.1.7. Restriction Mapping for Plasmid Verification and to Generate the Cut for Cloning

Restriction mapping was performed in order to verify that sub-cloning was successful, verify the identity of each backbone, and to generate the cut in the commercial

backbones for cloning. This cut also created compatible ends for subsequent ligation with the plasmid insert. Serial Cloner’s Virtual Cutter feature predicts the fragments generated when cut with a particular pair of restriction enzymes. The enzyme pairs BamHI and SacI, and PvuII and HindIII, were chosen because they generate distinct patterns that will allow for plasmid verification. The enzyme pair PvuII and HindIII created compatible ends for ligation and allowed for the extraction of the desired backbone fragment. The results produced by Serial Cloner’s Virtual Cutter are shown in Figure 9.

Table 13 The results of sub-cloning the commercial backbone plasmids from Oxford Genetics™. The nomenclature OGD52X was given by Oxford Genetics™, however the column named Type contains the nomenclature used throughout the project. The concentration and purity of DNA obtained is well within acceptable limits.

As seen in Figure 10, each lane of the gel showed 2 fragments that corresponded to the theoretical cut patterns generated in Virtual Cutter. The fragments showing at approx.

4000 kb, or the highest band in each lane, corresponds to the desired linearized

commercial backbone. These fragments were excised as described in 5.3.1 in Methods.

The smaller fragments that traveled farther are the portions that were intended to be excised and removed. Since all fragments generated match what was predicted in the theoretical cuts made in Virtual Cutter, the sub-cloning step preserved the identity of the plasmids, and the fragments circled in red were excised followed by DNA gel extraction.

The combination of patterns produced in Figure 9 allowed for the confirmation of the correct identity of the commercial backbone after sub-cloning. This method of restriction mapping has also produced the linear commercial backbones now with compatible ends for subsequent ligation to produce the full plasmid construct (Figure 10 2A, 2B, and 2C).

Figure 10 The Gel for backbone DNA gel extraction. 1.= incubation with BamHI + SacI 2. = incubation with PvuII + HindIII and A=Heavy Chain, B=Kappa Chain, C=Lambda Chain. The patterns of fragments produced match the predicted theoretical cuts made in Virtual Cutter. 1A, 1B, and 1C were used for plasmid verification, and 2A, 2B, and 2C produced linearized commercial backbone with compatible ends for cloning.

Figure 9 A. The generated fragments produced by Virtual Cutter for the Heavy Chain commercial backbone incubated with SacI + BamHI restriction enzymes (left) and HindII + PvuII restriction enzymes (right). B. The generated fragments produced by Virtual Cutter for the Kappa (light) Chain commercial backbone incubated with SacI + BamHI restriction enzymes (left) and HindII + PvuII restriction enzymes (right). C. The generated fragments produced by Virtual Cutter for the Lambda (Light) Chain commercial backbone incubated with SacI + BamHI restriction enzymes (left) and HindII + PvuII restriction enzymes (right).

6.2 PCR Colony Screen

The PCR Colony Screen indicated if ligation and transformation were successful. It also demonstrated which clones of E. coli contain the correct plasmid. Clones whose

plasmids did not ligate properly, i.e. have re-ligated to the backbone, not ligated the insert to the plasmid, etc., will demonstrate fragment sizes outside of the expected when visualized on the gel. Several colonies were chosen for each mAb, as to increase the chances of having positive candidates to bring forward for transfection. The mAb AR3C was brought forward before the rest of the antibodies, to ensure technical mastery prior to scaling-up the protocol with the remaining mAbs.

As seen in Figure 11, there were several positive colonies for both the heavy and light chain of AR3C. Positive colonies for the heavy chain include colonies 4,5,11,14,15, as they generated fragments at the correct size. As for the light chain of AR3C, positive colonies identified were 10,13,16. All other colonies were deemed negative, as they did not contain the desired PCR product. Positive colonies were then subjected to a culture and MiniPrep, then sent for sequencing to confirm whether they were truly positive.

Truly positive clones were then selected for overnight culture, the plasmids were purified via MidiPrep, and transfected into the HEK cells.

When attempting to bring the remainder of the mAbs forward, I encountered the need for troubleshooting, as all colonies for all mABs in the PCR colony screen turned out negative (Figure 12). This is most likely due to a failed ligation reaction or

transformation. Several attempts were made to correct this including: increasing the ratio of insert: vector in the ligation reaction calculation, using several ratios of dilution during transformation (i.e. 1:10, 1:100, 1:1000, etc.), repeating the ligation reaction, and

Figure 11 The PCR Colony Screen results for the mAb AR3C. AR3C Heavy chain (left) demonstrated 4,5,11,14,15, & 16 as positive colonies. AR3C Light chain (right) demonstrated colonies 10, 13, and 16 as positive colonies.

only selecting colonies to screen that were distinct, large, and not touching other colonies on the plate.

Thankfully, identifying all sources of potential error worked, and positive clones were finally seen for each plasmid category (Figure 13). AR3C was used as a positive

control, since it was shown to work previously (Figure 11). Therefore, the original AR3C positive clones, referred to as later figures as AR3CHO in Figure 18, as well as a “new”, repeated ligation experiment for AR3C, AR3HN, was used as a point of reference.

Figure 12 The PCR Colony Screen results for HEPC3-H, HEPC3- L, HEPC74-H, HEPC74- L, and HCV1-L where troubleshooting was required. All colonies were negative thus requiring deep troubleshooting.

The positive colonies were selected for overnight culture, their DNA then extracted via MiniPrep, and sent for sequencing. Positive clones whose sequencing data matched the sequence from the theoretical plasmids were brought forward for transfection. This segment of the experiment took up the largest amount of time.

6.3 Western Blot

The purpose of performing a western blot is to determine if IgG has been produced and secreted, which is the primary objective of this project. After transfecting and allowing the cells ~72 hours, supernatant and whole cell lysate (WCL) were harvested.

Essentially, the supernatant is the cell culture media, where all secreted and

extracellular proteins, including the cloned IgG, are to be found. The detection of the cloned IgG in the supernatant indicates successful bioproduction.

In addition to the supernatant, it was also important to run a western blot on the whole cell lysate (WCL), as this would provide insight as to what is occurring inside the cell.

For example, if IgG is detected in the WCL but not in the supernatant, that would insinuate that expression has taken place, but that there is a problem with secretion. In contrast, if no IgG is detecting in the supernatant nor the WCL, this would indicate an issue with expression, and the plasmid sequence would then need to be reviewed.

Troubleshooting can then take place if needed.

AR3C was the first to be brought forward, due to its rapid passing of all laboratory and QC steps, and also to allow me to familiarize myself with the protocols and techniques so as not to make errors when scaling-up the remaining mABs. Since there were a few positive candidates for AR3C, they were transfected in various combinations as to see if there were any differences in expression (Figure 14). Not only did the AR3C plasmid pass all previous laboratory tasks with flying colors, it also produced antibody from all of its positive colonies (Figure 14).

Figure 14 Left: the combination of heavy and light plasmid candidates that were transfected into each well. Since there was more than one positive clone for AR3C heavy and light chain, all were transfected as to see if there were any differences in expression. Right: The transfected cells from each well were harvested and run on a Western Blot. IgG production and secretion were seen in all wells. Positive control is adjacent to lane 6.

The western blot does not demonstrate clean, smudge-free blots. The research

assistant for the Grove lab advised me that the human secondary antibody can be quite

“sticky”, thus requiring several washing steps, and is prone to produce the smudges seen in Figure 14. However, I was reassured by experienced researchers that it is acceptable by normal standards. Therefore, we still concluded that IgG was produced, and this was a fantastic green light to bring the remainder of the mAbs up to this point.

After transfecting the remaining mAbs, the western blot in Figure 15 demonstrated IgG production (WCL) and secretion (SN) for HCV1, HEPC3, AR3C, but not for HEPC74.

This is seen by the presence of two bands in both the WCL and SN for these mAbs. We see two bands on the western blot due to the use of a reducing agent, which reduces the disulfide bonds between the heavy and light chain. This therefore allows us to distinguish the heavy and light chain in case troubleshooting was required. Without the use of this reducing agent, the mAb would appear as one single mass, making it

impossible to distinguish the protein chains. Since the heavy and light chain were encoded on separate plasmids, this distinction was necessary. Although a western blot was already completed for AR3C, it was included again here as a point of reference, since it was shown to work previously.

Figure 15 The images at the top and bottom correspond to the same Western Blot taken at different exposures. IgG production and secretion were seen in all mABs except for HEPC74. Sequencing data confirmed that this was due to an error in HEPC74’s sequence.

HEPC74 demonstrated the production and secretion of its light chain (Figure 15), but the heavy chain was absent from both, indicating an issue at the expression stage.

Analysis of the sequencing data for HEPC74-H showed an error in the sequence to be the cause of the lack of expression. The sequencing data was analyzed after the western blot because it is important to transfect when the HEK cells are at an optimal confluency and health, which is time dependent. Another positive clone for HEPC74-H was selected, its sequence was verified as a perfect identity match to the theoretical plasmid and was transfected with the same light chain plasmid as in Figure 15. After harvesting the transfected cells, the laboratory was shut down due to the covid-19 outbreak prior to being able to run the western blot. Therefore, I was able to officially produce 3 out of 4 mABs, however since the sequencing data for the remaining HEPC74-H plasmid is correct, I believe it is highly likely that the harvested cells transfected with HPEC74 will demonstrate IgG production and secretion.

7. Discussion

The purpose of this project was to clone and produce human IgG against Hepatitis-C Virus (HCV) using an in-house recombinant technique. This in-house recombinant technique for expression of antibody is advantageous because it is more time and cost effective than purchasing custom-made antibodies from a manufacturer. It allows the researcher to test as many antibodies against the desired target without fear of wasting valuable time and money.

With vaccines available for Hep-A and Hep-B, Hep-C is the last challenge of the hepatitis viruses. Hep-C infection leads to a chronic, lifelong disease whose HCV- related diseases pose a large burden on healthcare. Research for an HCV vaccine is extremely important, since a significant number of people with HCV develop cirrhosis or hepatocellular carcinoma (WHO 2019).

The mAb’s target in this study is the viral glycoprotein E2, which is present on the surface of HCV and acts as the key molecule for viral fusion and subsequent entry into the host cell (Ploss 2012). The structures and sequences of antibodies isolated from patients who have successfully cleared the infection are published in the public domain and have been shown to bind to E2(RSCB PDB). The binding of these mAbs to E2 leads to viral neutralization and successful clearing of the infection, making this a desired immune response for potential vaccine development.

Overall, the four mAbs cloned in this study were HCV1, HEPC3, AR3C, and HEPC74.

HCV1, HEPC3, and AR3C were successfully produced (Figure 14, 15). The

sequencing data for HEPC74 showed that failed secretion seen in Figure 15 was due to an error in its sequence. Another clone for HEPC74 was confirmed to have an exact sequence match to the theoretical plasmid and was transfected then harvested. Just as the western blot to confirm secretion was ready to be run, the laboratory closed due the Covid-19 pandemic. Therefore, the data can officially declare a 75% mAb success rate, although it is safe to assume that there is antibody present in the HEPC74 WCL and SN since the clone used for transfection has a perfect sequence match to the

theoretical plasmid. Since the primary objective of this project was to successfully cone and produce these four antibodies, then this study is considered a satisfactory

success.

Due to the Covid-19 pandemic, further testing and experiments on these mAbs was not possible but my results provide for meaningful future perspectives. This includes the upscaling of the antibodies to have amounts comparable to a working supply for future experiments. The upscaling of the mAbs would have required upscaling the number of HEK cells while maintaining confluency, and mAb purification using a protein G column on an ÄKTA pure protein purification system.

In addition, it would have been fascinating to test the biological activity of all four antibodies to demonstrate that not only has antibody been successfully produced, but that they also actively engage HCV’s E2 viral glycoprotein. The Grove Lab uses HCV pseudoparticles (HCVpp), whose genes encode for all required elements (including E2) for HCV study while eliminating the risk of infectious exposure to researchers (Bailey 2018). Thus, neutralization assays could be performed to examine antibody binding to HCVpp E2. The HCVpp encodes a reporter luciferase gene that generates a signal with viral entry via E2. Thus, if one of the four mABs successfully neutralizes E2, it can be quantified (Bailey 2020). This is visualized specifically via a reduction in the

luciferase signal (Bailey 2020).

It would also be necessary to demonstrate that the mAbs have the correct specificity, which can be determined by an enzyme-linked immunosorbent assay (ELISA). This would entail putting soluble E2 in wells using lectin which binds to the glycans on E2.

Then the four mAbs would be allowed to bind, and a donkey anti-human IgG

secondary antibody would be used to detect the bound mAbs. This would confirm that the mAbs in question are indeed specific to E2.

Another test for the biological activity could be to compare the mAbs’ affinities to E2.

The affinity between the four antibodies and E2 can be quantitatively described by performing a surface plasmon resonance (SPR) test (Homburger 2008). SPR is a time- dependent method for characterizing ligand binding, including antibody-antigen

interaction. The Grove Lab also has a confocal microscope, which provides a resolution at 0.8µm (Berkeley 2007). It perhaps would have also been possible to examine the mAb: E2 interaction under the confocal microscope to characterize this interaction further.

In addition to testing for biological activity, the mAb production protocol itself could be further improved for efficiency and commercial scale-up by testing different signal sequences for optimal secretion efficiency. The signal sequence is important for directing nascent peptides destined for secretion to the endoplasmic reticulum (Kou 2017). This study used the tissue plasminogen activator (TPA) signal sequence, which has proved reliable and successful in the Grove Lab. However, other types of

optimized signal sequences have been described for IgG production (Haryadi 2015) and would prove useful if this study were to be implemented on an industrial scale.