EXAMENSARBETE INOM BIOTEKNIK, AVANCERAD NIVÅ, 30 HP
STOCKHOLM, SVERIGE 2018
Generation of affinity maturation libraries of PSMA targeting
affibody molecules and selections to find improved binders
REBECKA ELIN LOUISE STOCKGARD
KTH
SKOLAN FÖR KEMI, BIOTEKNOLOGI OCH HÄLSA
2
Generation of affinity maturation libraries of PSMA targeting
affibody molecules and selections to find improved binders
Rebecka Elin Louise Stockgard
19941215-6581 2018-08-07
Project location: KTH, CBH School, Department of Protein Science, AlbaNova University Center, Roslagstullsbacken 21, 10691 Stockholm
Supervisor: Hanna Lindberg Examiner: Stefan Ståhl
Abstract
Prostate specific membrane antigen (PSMA) is a membrane protein expressed in both prostate cancer cells and in the neovasculature of many other solid tumors. It can act as a therapeutic target and as a target for an imaging agent for various cancers. Earlier, three different PSMA targeting affinity molecules, called Affibody molecules, were generated through phage selection.
The aim of this thesis was to produce three corresponding affinity maturation libraries based on the three potential PSMA binders and select for improved binders through phage display.
The affinity maturation libraries were produced, and two to four selection cycles were performed with yet insufficient enrichment. Therefore, more cycles must be performed in order to discover potential PSMA-binders with sufficient affinity.
Sammanfattning
Prostataspecifikt membranantigen (PSMA) är ett membranprotein som uttrycks i både prostatacancerceller såväl som i nybildade blodkärl kring en tumör. Cancercellers uttryck av PSMA kan användas för att både behandla och upptäcka tumörer i kroppen. Tidigare har tre PSMA-sökande affinitetsmolekyler tagits fram, kallade Affibodymolekyler, genom fagdisplay.
Syftet med detta examensarbete var att producera tre korresponderande affinitetsmatureringsbibliotek baserat på de tidigare potentiella PSMA-bindarna, och att selektera via fagdisplay för att hitta bindare med högre affinitet. De tre affinitetsmatureringsbiblioteken producerades och två till fyra selektionscykler har genomförts utan att se tillräcklig anrikning av bindare. Därför måste fler cykler genomföras i syfte att hitta potentiella PSMA-bindare med tillräckligt god affinitet.
3
Keywords
PSMA - Affibody - Phage display
Introduction
Prostate cancer accounts for 25 % of all new cancer diagnoses in men, and a reliable screening method is needed.
Of all new cancer diagnoses in men 2017, 25
% were a diagnosis of prostate cancer (PCa).
Between the years 2010 and 2013, there was a decrease in incidence of PCa, however, this is thought to be due to a decrease in Prostate Specific Antigen (PSA) screening. [1] PSA is an enzyme that is produced in the prostate, and a portion of this is leaked out in the bloodstream. Thus, a simple blood test can determine the PSA levels in a person.
Normally, men with PCa will have higher levels of PSA in their bloodstream than healthy men. However, increased level of PSA can be subject to not only PCa, in fact, most of the men with a high level of PSA does not have PCa. Due to the unreliability of the PSA-test it was earlier this year (2018) decided to not screen for PSA-levels in men in Sweden. [2] PSA-testing is often combined with other tests, for example physical examination, to study any enlargements or other suspicious bumps, but this method is not decisive. Other imaging methods are also common, such as ultrasound or MRI.
However, it is hard to distinguish between normal and abnormal tissue with ultrasound, [3] and MRI is not a cost-efficient method. [4]
Therefore, a reliable screening method is needed for an accurate diagnosis of PCa.
Prostate Specific Membrane Antigen (PSMA) is an attractive target to detect not only PCa but other neoplastic tissues too.
In healthy prostate epithelia, Prostate Specific Membrane Protein (PSMA) is mainly
expressed as a cytoplasmic protein named PSM [5]. In PCa cells, differential mRNA splicing creates a type II non-covalently bound homo-dimeric membrane protein.
PSMA consists of 19 amino acids (aas) in the cytoplasm and 707 aas extracellularly, connected through a membrane-spanning part of 22 aas, as can be seen in Figure 1.
PSMA has a known peptidase function [5] [6]
and is also known as glutamate carboxypeptidase II. [7] PSMAs expression is positively correlating with cancer progression. [5] When the PCa progresses, the cells become increasingly independent of hormones required for male development and characteristics (androgens), the more activity of PSMA is detected. [6] [8] Despite its known enzymatic activity, PSMAs role in cancer progression is unknown. [5]
Figure 1. A 3D-structure of the extracellular part of PSMA where the two homo-dimers are shown, one i grey and one in colours. [7]
PSMA was first described in the human prostate cancer cell line derived from a lymph node, called LNCaP. It was characterized by a monoclonal antibody (mAb) from mouse, which recognized an intracellular epitope of PSMA. [6] Since its discovery, PSMA has been shown to be expressed in other solid cancer tumors and their connected neovasculature as well [5] [6], such as non- Hodgkin’s lymphoma (cancer in the lymph system), meningioma (cancer in the meningioma), [6] renal cell carcinoma (kidney cancer), breast cancer, colorectal cancer
4 (colon cancer), gastric adenocarcinoma
(stomach cancer) and glioblastoma (brain cancer). [9] Something to note is that PSMA has not been shown at all in healthy adjacent epithelial cells. [5] [9]
PSMA could function as a potential therapeutic target.
Interestingly, it has been shown that PSMA holds a motif for internalization at the N- terminal, in the cytoplasm. This motif enables the endocytic pathway that internalizes the entire receptor-ligand complex. Rajasekaran et al. [8] showed that the receptor-ligand complex anti-PSMA mAb J591-PSMA was internalized in LNCaP cells. J591 was dissociated from PSMA in the endosome where PSMA either ended up in the lysosome for degradation or was re-cycled to the surface of the cell. [8] In other studies, not only internalization of anti-PSMA mAbs has been shown [6] [9] but also that anti-PSMA mAbs coupled to toxins can eliminate PCa tumors in mice without any signs of toxicity.
[6] Due to PSMAs internalization feature it can deliver toxins directly into the cancer cell which is vital for development of affinity- based drugs. [8]
Thus, PSMA could not only function as an accurate target for diagnosis of PCa, and give information about the degree of disease, PSMA can also be employed to detect many other solid cancers. Furthermore, PSMA has the ability to internalize mAbs which is important for the basis of developing affinity- based treatments.
Affibody molecules as a small imaging agent with high binding affinity
There are around 75 Food and Drug Administration (FDA) and European Medicines Agency (EMA) approved mAbs on the market today [10] and have shown successes in their use in therapy. However, mAbs themselves are limited in their use as
therapeutics. mAbs are large, around 150 kDa, highly post-translational modified and contains a lot of disulphide bonds. They are produced in mammalian cells. This is just a few remarks that are limiting their use as therapeutics. [11]
Hence, there is a need for smaller proteins, with few modifications, that are easily produced in an efficient and cheap way.
furthermore, they need to be easily purified and have the property of high affinity and specificity that mAbs possesses.
There are a lot of different scaffold proteins developed today that holds these qualities.
One of them is called affibody molecules, see Figure 2A.
Affibody molecules are small proteins, consisting of 58 aas which creates a three- helix bundle of 6.5 kDa derived from the IgG-binding Z domain derived from the B domain of protein A at the surface of Staphylococcus aureus. [12]
The Z-domain is stable without any cysteines incorporated and have a high solubility, which makes it possible for it to be produced cost efficiently in prokaryotic hosts. [13]
Since affibody molecules have a small size, it makes them attractive for diagnostic imaging, where the quick biodistribution allows for good tumor binding and penetration. Their small size also enables a fast blood clearance, which can give rise to high-contrast images of the tumor. [14]
To create a high affinity and selective affibody molecule binder against a given target, the binding surface of the affibody is altered. The binding surface of normally 13 to 14 aas are randomized, to find a unique combination that allows binding to the target.
However, should the affinity between a lead affibody molecule and the target not be sufficient, typically nanomolar affinity, an affinity maturation of the lead can be performed. Then, some positions are locked,
5 while other positions are varied in order to
find a better match and reach nanomolar or even sub-nanomolar affinity toward the target. [12]
Their small size, the ability to be produced recombinantly in many hosts, together with their high thermal stability and solubility makes them suitable for being used as an imaging agent.
A. B.
Display of affibody molecules at the surface of filamentous phages and selection against target displayed on cell
There are a lot of different display methods, including mammalian cell display [15], yeast display [16] and phage display [17]. All of the different display methods possess the trait of connecting the genotype to the phenotype. In this thesis phage display will be employed as the selection method.
At the surface of a filamentous phage, a protein called pIII is displayed in three to five copies, see Figure 2B. The N-terminal of pIII is used for binding to gram-negative bacteria through F pilus for infection. The C-terminal of pIII is buried into the phage. George P.
Smith showed 1985 that a peptide can be inserted between the N-terminal and the C- terminal, neither disrupting any function of
the inserted protein or the infectious phage function. pIII is the far most used display protein for mAbs. [17]
A phagemid is a plasmid containing origin of replication (ORI) for both plasmid replication and an ORI for phages. A phagemid that contains the pIII gene disrupted by the gene of interest is one way of displaying peptides at the surface of the phage. When using a phagemid, the bacteria that holds the phagemid has to contain a helper phage, which provides all the other necessary proteins for the assembly of a phage displaying the wanted peptide.
Normally, the antigen selected against is an immobilized recombinant antigen. However, if the recombinant antigen is malfunctioning, the antigen selected against can be present on a mammalian cell surface. If there are phages that display a protein that have affinity to the antigen, it will be eluted, and non-binders will be washed off.
The eluted phages will be added to F pilus- containing bacteria and will be let to infect.
Subsequently, helper phages are added which results in assembled phages. However, only 1-10 % of the produced phages will display the protein of interest. For this reason, the phages need to be amplified. The phages displaying protein, see Figure 2B, are let to bind to the antigen again through several cycles. [18]
To find the binders with the highest affinity and selectivity, the conditions are changed to become more stringent per cycle. When performing selection against an antigen located on the surface of mammalian cells, the number of antigens displayed for the phage can be decreased i.e. the number of cells, or the decreased amount of incubation time. The washing is carried out by cautiously spinning down cells and resuspension in new buffer. [18]
Figure 2. A) The affibody molecule, helix 1 to the left and helix 2 to the right, with red spots indicating the usually randomized positions.
Picture from [14]. B) The display of 0061n affibody, connected to pIII, at the surface of a filamentous phage
Filamentous phage particle
pIII protein
6
Affibody molecules toward PSMA was earlier generated with micromolar affinity and the aim of this thesis
Earlier, PSMA targeting affibody molecules were generated using a randomized affibody library selected against PSMA expressing cells.
Three different variants, here referred to as H05, H09 and G03, were shown to have affinity toward PSMA positive cell lines in a low micromolar range (unpublished data).
The aim of this thesis is to produce affinity maturation libraries of three promising PSMA targeting affibody molecules, H05, H09 and G03, and select for improved candidates targeting PSMA by phage display.
The recombinant produced PSMA did not show the same properties as PSMA located on the surface of mammalian cells. The possible reason for this is that the non- covalently bound homo-dimeric structure of PSMA is hard to express recombinantly (unpublished data). The phage particles with affinity toward PSMA will be eluted with two different strategies, lowering pH and by adding trypsin.
Materials and methods
An illustration of the methods for library preparation can be found in appendix A, supplemental figures, Figure 17, at p. 22. All chemicals and reagents used are presented in appendix B, Table 6 at p. 23. All prepared buffers, medium composition and compositions of plates used are presented in appendix B, Table 7, at p. 23.
Library Design
Three affinity maturation libraries were already designed based on sequence evaluation of previously potential affibody molecules targeting PSMA. Out of the 13 positions that were randomized before, 6
were locked while 7 positions were randomized with NNK-codons. This resulted in three affinity maturation libraries, each the size of 109 possible variants.
Preparation of library vectors
Amplification of oligonucleotidesPCR with forward primer LIHA48 (CCGGCGCTCGAGGTAGATGCCAAA TACGCCAAAGAA) and reverse primer LIHA79
(ATACATGTCGACTTTCGGCGCTG) was performed to amplify affibody genes from three affinity maturation libraries. The PCR amplification were analyzed on a 1 % agarose gel [150 V, 400 mA, 15 minutes] and purified using a Qiagen PCR-purification kit according to manufacturer’s recommendations. The elution was performed with 50 µl pre-heated (65 oC) Elution Buffer (EB), incubated [10 min, Room Temperature (RT)] and finally centrifuged [full speed, 1 min, RT] in a table top centrifuge. Concentration of purified products was determined on the nanodrop (NanoDrop technologies, Model: ND-1000, Serial No. 0678) at a wavelength of 260 nm with EB as blank.
Restriction Enzyme (RE) digestion and purification of PCR product for each library The PCR amplified affibody library fragments was heated to 70 oC in 10x CutSmart buffer and MilliQ at 70 oC for 15 minutes followed by cooling at 4 oC for 30 minutes. 160 U of each restriction enzyme XhoI [20 U/µl] and NheI-HF [20 U/µl] was added to a total volume of 1000 µl and incubated for 5 hours at 37 oC. The restricted library inserts were purified using Qiagen PCR-Purification kit using the same elution strategy as before. The purified fragments were analyzed on an 1 % agarose gel [150 V, 400 mA, 1h].
The three libraries were re-restricted in 10x CutSmart Buffer, with 40 U of the
7 corresponding restriction enzymes, to a total
volume of 300 µl, and incubated [2 hours, 37
oC]. After incubation, the mixtures were analyzed on an agarose gel [150 V, 400 mA, 30 minutes].
Preparative gel extraction
The re-restricted libraries were run on a 1 % agarose gel together with 5 µl glycerol and 50 µl Loading Dye and extracted with a QIA Quick Gel Extraction Kit and eluted in MilliQ. The concentration of the gel extracted library inserts were determined on the nanodrop at a wavelength of 260 nm with MilliQ as blank.
Test ligation of PCR product and phagemid vector pAffi1
Test ligations of all three library constructs were performed, using both a 4-fold and a 5- fold molar excess of construct to the phagemid vector pAffi1, using T4 DNA Ligase. First, the fragments were incubated with the restricted phagemid vector pAffi1, as well as a background control with only the restricted phagemid pAffi1, in 50 oC for 10 minutes, and then let to cool in RT for 5 minutes. Subsequentially, 10x Ligase Buffer and T4 DNA ligase were added to a total volume of 20 µl, where 400 U was wanted for a reaction volume of 20 µl, incubated in RT for 4 hours and then in 4 oC over-night (ON).
Transformation of test ligated product into E. coli TOP10 to evaluate ratio between test ligated affibody affinity maturation libraries and phagemid vector pAffi1
100 ng of the test ligated pAffi1-affibody- library constructs were used in both 4-fold and a 5-fold molar excess to the phagemid vector pAffi1. 5X KCM stock was added to each ligation mix to a total volume of 10 µl which was then chilled on ice for 5 minutes.
The mixtures were then transformed into chemically competent E. coli (TOP10, in- house produced) using heat-shock. After transformation, Tryptic Soy Broth (TSB) was
added. The mixture was incubated at 37 oC for 1 hour and then plated onto TBAB-plates supplemented with Ampicillin (100 µg/µl) (Amp-plates) and placed in RT over weekend (OW). Colony PCR were performed followed by a gel-screening on a 1 % agarose gel [150 V, 400 mA, 15 minutes]. 64 clones per library were sent to MicroSynth for sequencing.
Library ligation
The ligation of the different libraries into the phagemid vector pAffi1 was performed with a 4-fold excess using T4 DNA Ligase. The ligations were performed in the same manner as before, however, the total reaction volume was expanded into 2000 µl, and 1600 U of the T4 DNA Ligase were used for this reaction volume.
The T4 DNA ligase were heat-inactivated at 65 oC for 20 minutes, and then let to cool in RT for 15 minutes before purifying the plasmid with Qiagen gel purification kit. The plasmids were eluted in pre-heated MilliQ (60
oC) and examined on an 1 % agarose gel [150 V, 400 mA, 30 minutes]. The resulting phagemid containing libraries concentration were determined on the nanodrop at a wavelength of 260 nm with MilliQ as blank.
Library transformations into electrocompetent E. coli ER2732
Test electroporation of library ligation to evaluate transformation frequencyA test electroporation of the phagemid pAffi1-affibody-H05 library was performed using electrocompetent E. coli (ER2738) cells, testing different amounts of DNA testing different amounts of DNA (118 ng DNA and 235 ng) together with 25 µl cells.
Electroporation was performed using electrocompetent E. coli ER2738 from Lucigen (Catalog number: 60522-2) for the three pAffi1-affibody-libraries. 25 µl cells per 1 µl un-purified test-ligation mix was used.
The ligation mixes and cells were mixed and put on ice for 5 minutes before
8 electroporation [1.8 kV, 0.1 cm cuvette].
After electroporation, 975 µl Tryptic Soy Broth supplemented with Yeast extract (TSB + Y) was added and each sample was incubated at 150 rpm at 37 oC. After 1 hour the incubated cells were diluted in TSB+Y in 101, 102, and 103 dilutions and 100 µl of the dilutions were spread onto Amp-plates placed at 37 oC ON. The transformation frequency was evaluated based on number of colony forming units (cfu). A PCR-screening was made and analyzed on a 1 % agarose gel [150 V, 400 mA, 15 minutes].
Subsequently, the transformation frequency for pAffi1-affybody-H09/G03 libraries were established with 238 ng DNA (H09) and 232 ng DNA (G03) 25 µl cells using the same electroporation protocol as before. The transformation frequency was evaluated in the same manner as for H05.
Library transformation into E. coli ER2738 The ligation products for each library were transformed into electrocompetent E. coli (ER2738). The electroporated cells were incubated [37 oC, 150 rpm, 1 h] supplemented with recovery media from Lucigen.
After 1 h, 10 µl cells were taken out for titration of the library and kept on ice until timepoint of titration, whereas the rest of the cells where transferred into a 5 L baffled flask containing TSB + Y supplemented with Ampicillin (100 µg/ml), 2 % Glucose and Tetracycline (10 µg/ml), to a total volume of 500 ml and amplified for 18 h (G03), and 16 h for H05 and H09.
Harvesting of the library transformants
OD600 was measured on the ON-cultures for determination of cell density, then pelleted [4000 rpm, 15 min, 4°C] and resuspended in the residual media. Glycerol stocks were made and stored in -80 oC.
Viability of frozen library glycerol stocks
The viability of the frozen library glycerol stocks was estimated using titration. 10 µl of the frozen glycerol stocks were diluted in TSB + Y, spread on Amp-plates in duplicates, and incubated 37 oC ON.
Preparation of affinity maturation library phage stocks
Cultivation of library
A 50-fold of the library size were inoculated to a starting OD600 of 0.1 to each two shake flasks, one for each track (either elution by lowering the pH, or elution by trypsin), containing TSB + Y supplemented with 1 % glucose, tetracycline (10 µg/ml) and carbenicillin (100 µg/ml).
Infection of library with helper phages M13K07 When the cultures had reached an OD600 of 0.4, each culture was infected with helper phage M13K07 in a 10-fold excess. At the time of infection and after 20 minutes the cultures were carefully swirled. After incubation for 45 minutes in RT, the cultures were centrifuged [6000 xg, 20 min, 4°C] and the pellets were resuspended in 15 ml TSB+Y. The resuspended bacteria were added to baffled shake flasks containing 500 ml TSB+Y supplemented with Carbenicillin (100 µg/ml), Kanamycin (25 µg/ml), and IPTG (1 µg/ml). The cultures were incubated at 150 rpm in 37 oC for 18 hours for the G03 library, and for 16 hours for H05 and H09 libraries.
Harvesting of phage containing supernatant The ON-cultures were centrifuged [6000 xg, 20 min, 4°C]. The phage containing supernatant were precipitated by adding ¼ of the total volume (500 ml) of [20 % PEG8000 in 2.5 M NaCl] and incubated on ice for 45 minutes. The precipitated phages were centrifuged [15,000 xg, 30 min, 4°C] and resuspended in 120 ml MilliQ, followed by another precipitation by adding ¼ of the
9 volume of [20 % PEG8000 in 2.5 M NaCl]
and incubation on ice for 45 minutes. Once again, the precipitated phages were pelleted [15,000 xg, 30 min, 4°C] and resuspended in 2 ml PBS, followed by another centrifugation [full speed, 10 min, RT] in a table top centrifuge. The phage containing supernatant were transferred to a pre-blocked eppendorf tube and the cell debris were saved in glycerol stocks.
Determination of phage concentration by titration A ON-culture for ER2738 was started 1 day ahead, in TSB+Y supplemented with tetracycline (10 µg/ml) at 37 oC. The ON- cultures were used for inoculation of new cultures of 100 ml TSB+Y supplemented with tetracycline (10 µg/ml) to an initial OD600 of 0.1. The culture were cultivated in 37 oC at 150 rpm until log phase (OD600
between 0.5-0.8). 100 µl of E. coli ER2738 cells in log phase was added into phage dilutions between 101 – 1010 to a total volume of 200 µl and incubated for 30-45 minutes at 37 oC standing still before being plated onto TBAB-plates supplemented with carbenicillin (100 µg/ml) and TBAB-plates supplemented with kanamycin (50 µg/ml). The plates were incubated at 37 oC ON or at RT OW.
24 colonies of each track were PCR screened and verified by gel electrophoresis [150 V, 400 mA, 30 minutes] on an 1 % agarose gel.
Cell lines and cell culture
The human embryonic kidney cell line HEK293T and HEK293 cell line transfected with PSMA were sustained in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10 % (v/v) fetal bovine serum (FBS), 1 mM Sodium Pyruvate, 2 mM L-Glutamine, at 37 °C, 5 % CO2. The transfected HEK293 cell line were cultivated with a supplement of 100 µg/ml Zeocine as selective antibiotic.
Evaluation of PSMA expression on target cells
The growth medium was removed, and cells were washed in PBS. Seeded cells were detached by pipetting up and down carefully, and washed again with PBSB1%, before being centrifuged (1700 RPM, 3 min, RT).
Pelleted cells were dissolved in PBSBPlus and counted. 300 000 cells per labelling experiment was washed twice (centrifuged 1700 RPM, 3min, RT, resuspended in PBSB), followed by adding 50 µL primary reagent, according to Table 1, followed by incubation in 4ºC for 45 min.
Table 1. Schematic overview of primary and secondary antibodies used for evaluation of PSMA expression on target cells
Cell line Primary reagent Secondary reagent HEK293
(PSMA+)
50 µl of a-hPSMA Sheep IgG [2 µg/ml]
50 µl of Donkey a-Sheep Alexa- 488 [2 µg/ml]
HEK293
(PSMA+) 70 µl PBSBPlus 50 µl of Donkey a-Sheep Alexa- 488 [2 µg/ml]
HEK293 50 µl of a-hPSMA Sheep IgG [2 µg/ml]
50 µl of Donkey a-Sheep Alexa- 488 [2 µg/ml]
HEK293 70 µl PBSBPlus
50 µl of Donkey a-Sheep Alexa- 488 [2 µg/ml]
After incubation the cells were centrifuged [1700 rpm, 3 min, RT] and resuspended in the secondary reagent, according to Table 1, and incubated at 4oC with rotation for 45 minutes.
After incubation, the cells were washed two times in 100 µl PBSB (1700 rpm, 3 min, 22
oC) and finally dissolved in 400 µl PBSB and kept on ice in darkness.
Selection of affinity maturation libraries against HEK293 transfected cells
Cycle 1
Cycle 1 is illustrated in Figure 3. All selections were carried out in two tracks differing in the elution strategy, one by lowering the pH and the other by adding trypsin.
10 Preparation of mammalian cells for selection
HEK293T cells and HEK293 transfected with PSMA were detached from the culture dish using PBS. The cells are washed three times in mammalian cell buffer (PBSB0.1%) [1700 rpm, 3 min, RT], before being dissolved in selection buffer (PBSTB1%plus).
Selection
All tubes used during the selections were pre- blocked with 1 ml blocking buffer (PBSTB 3%).
Negative pre-selections, for each elution strategy, were carried out by adding 800 µl of the produced phage library stocks (dissolved in selection buffer) to the 100 µl HEK293T mammalian cells and incubating at 4 oC for 30 min at 12 rpm in a rotamixer. The cells were centrifuged [1700 rpm, 3 min, RT], and the pre-selected phage containing supernatants were transferred to 100 µl PSMA transfected cells. The selection mixes were incubated at 4
oC for 2h, at 12 rpm in a rotamixer, and then washed two times in 1 ml washing buffer (PBSTB1%plus), followed by removal of the supernatants.
Elution
Elution by lowering the pH was performed by adding 500 µl 50 mM glycine-HCl pH 2.2 and incubating at RT for 10 minutes, before adding 50 µl Tris-HCl pH 8.0 and 450 µl PBS.
The solution was mixed by pipetting up and down, centrifuged [2000 rpm, 2 min, RT] and the supernatant was transferred to a new tube.
Elution by trypsin was performed by adding 500 µl 1x Trypsin in PBS and incubating for 30 min at 37 oC, followed by addition of 500 µl PBS and pipetting up and down and centrifugation [2000 rpm, 2 min, RT]. The supernatant was transferred to a new tube.
Amplification of eluted phages
980 µl of each eluate was added to separate tubes with 40 ml E. coli (ER2738) in log phase
and were incubated at 37°C for 45 min standing still, and then centrifuged [3500 xg, 15 min, RT]. The pellets were resuspended in residual media and each track was plated onto a large TYE plate supplemented with 10 µg/ml tetracycline, 100 µg/ml carbenicillin, and 1 % glucose. The plates were incubated ON at 37 oC.
Phage stock culturing
TSB was added to the plates and the cells were scraped of the ON-plates. Glycerol stocks were prepared on the cell suspensions, with approximately 20 % glycerol in every cell stock. The cell density was measured through OD600 measurement.
TSB supplemented with 2 % glucose, 100 µg/ml carbenicillin, 10 µg/ml tetracycline was inoculated with 1010 cells to a starting OD600 of 0.1. The cells were grown to log- phase at 37 oC with 150 rpm shaking, where 1010 cells were taken out and added to a 50 ml tube. At least 1.25 times as much helper phages, M13K07 concentration of > 1x1011 helper phages, were added and incubated at 37 oC for 30 min standing still. The cells were centrifuged [3300 xg, 10 min, RT] and then resuspended in TSB + Y supplemented with 100 µg/ml carbenicillin, 25 µg/ml kanamycin, and 0.1 mM IPTG and incubated at 37 oC ON.
Precipitation of phages
The ON-cultured cells were centrifuged [3300 xg, 10 min, RT] in GSA tubes. The supernatant was transferred into a new GSA tube. ¼ volume [20 % PEG8000 in 2.5 M NaCl] was added, mixed gently, and incubated on ice for 1.5 h. The mixture was centrifuged [10500 xg, 30 min, 4 oC], and the phage pellet was dissolved in 10 ml MilliQ and transferred to SS34 tubes. Again, ¼ volume [20 % PEG8000 in 2.5 M NaCl] was added, and incubated on ice for 45 min. After incubation, the mixture was centrifuged [17500 xg, 30 min, 4 oC], and the phage pellet was resuspended in 800 µl selection buffer.
11 Phage titration
The phage stock and eluted phages were titered in order to estimate the phage concentration. A dilution series was prepared in technical duplicates in the range between 101-1010. 10 µl phages, 90 µl MQ and 100 µl E. coli ER2738 in log-phase was added to each tube, followed by mixing and incubation [30 - 45 min, 37 oC]. 100 µl from each tube was plated onto agar plates supplemented with 100 µg/ml carbenicillin, and 50 µl from each tube was plated onto agar plates supplemented with 25 µg/ml kanamycin. The plates were incubated at 37 oC ON.
PCR-screening of individual clones
Single colonies were picked from the carbenicillin plate, dipped onto a new carbenicillin plate, and put into a PCR-plate with 30 µl MilliQ in each well. The plate was sealed, vortexed and incubated at 95 oC for 10 min in order to lyse the cells. Subsequently, the plate was vortexed and 1 µl from each well was transferred to a new PCR-plate and run in order to amplify the affibody fragments. The PCR products were run on a 1 % agarose gel [150 V, 400 mA, 20 minutes].
Subsequent cycles
The following phage selection cycles were performed as described above, with a few changes, see Table 2. After the second cycle, the material was not plated onto TYE plates, instead, amplified ON as following: 750 µl of the eluate was let to infect 7.5 ml E. coli ER2738 cells in log-phase at 37 oC for 30 minutes standing still. The rest of the eluate was mixed 1:1 with 87 % glycerol. 7.5 ml TSB + Y supplemented with 2 % glucose, 200 µg/ml carbenicillin was added into the infected cells and was incubated 30-60 minutes with 100 rpm shaking at 37 oC.
Then, 5 times excess of helper phage M13K07 was added in comparison to the amount of E. coli cells, and then incubated at 37 oC for 30 minutes standing still. The cells were centrifuged [3300 xg, 10 min, RT], resuspended in 100 ml TSB + Y supplemented with 100 µg/ml carbenicillin, 25 µg/ml kanamycin and 0.1 mM IPTG and incubated at 37 oC, 150 rpm ON. Then, precipitation of phages was carried out in the same manner as for cycle 1.
Figure 3. An overview of the first phage cycle. After harvesting of phages, the next cycle is started by pre-selecting the harvested phages against non-expressing PSMA cells. The number of cycles depends on the degree of enrichment, which can be evaluated by titering and sequencing.
12
Table 2. Variation of parameters in the phage selection cycles.
Cycle 1 2 3 4
Pre- selection (estimated
no. cells)
106 106 106 5*105
Selection (estimated
no. cells) 106 106 106 5*105 Selection
time (h) 2 2 2 1
Helper phage excess (times)
1.25 5 2 2
Plated X - X -
Amplified
ON - X - X
After three cycles 48 clones per track from the G03 library were sent to MicroSynth for sequencing.
Results
Coverage and yields during library vector preparation
All yields during each step is included in Figure 6, as well as coverage of library transformation, the final library sizes, and the size and coverage of the phage libraries.
The amplification of the library oligonucleotides was successful, and the product was purified. The purified library fragments were double digested for 5 hours, the digested library fragments were purified and analyzed on a 1 % agarose gel, see Figure 4A. However, there was fragments under 100 bp remaining in the samples, which can be seen in Figure 4A.
Figure 4. Abbreviations, L: Ladder, where the corresponding masses are labelled to the left of the figures, Ref: Reference of un-digested library.
The digested library fragments have a size of 143 bp, whereas the undigested library fragments a size of 294 bp. A) Gel-verification after the first digestion. H05 was divided into two wells, whereas H09 and G03 were put in 1 each, marked in the figure. It can be seen that there are products with a lower size than 100 bp for all libraries, indicating that the purification was not successful, marked with a box, as well fragments that is still not digested, marked with an arrow. B) The gel- verification after the re-digestion. Nearly all un-digested sample is gone after the re-digestion, for all three libraries, in comparison to where the reference band.
According to Figure 4A there were still non- digested library fragments (294 bp, compared to digested library fragments of 143 bp).
Consequentially, it was decided to re-digest the library fragments for another 2 hours.
Figure 4B is a gel-verification of the re- digested libraries. No non-digested library fragments can be seen, indicating a successful digestion. However, since fragments under 100 bp remained in the samples, it was decided to purify the library fragments through gel-extraction.
Figure 5. Abbreviations, L: Ladder, corresponding masses are labelled to the left of the figure. This figure shows the gel verification of the gel- extracted library fragments. H05 and H09 was purified with no seen impurities, as no band of lower mass is seen. However, a weaker band is seen under the G03 library fragments, marked with an arrow, indicating that some impurities is still remaining in the sample.
Figure 5 shows a gel electrophoresis of purified library fragments. A band of 143 bp indicated library fragments of H05 and H09 and no impurities can be seen. However, a weaker band is seen under the G03 library fragments of 143 bp, marked with an arrow.
This indicates that impurities are still remaining in the G03 library after gel extraction and purification.
13 To confirm the best ratio of phagemid vector
pAffi1 to the library inserts which results in the highest transformation frequency between ratio 1:4 and 1:5 (vector: library), test ligations were performed in. The ligated products were transformed into E. coli TOP10. Ratio 1:4 gave the highest transformation frequency together with the lowest percentage of background ligations
Evaluation of library design based on sequencing data
64 individual colonies of each library were sequenced for evaluation of the library design, 56 clones from E. coli TOP10 and 8 clones from E. coli ER2738. Figure 7 (H05), Figure 9 (H09), and Figure 11 (G03), were made to illustrate the sequence analysis.
Figure 6. Abbreviations: cfu: colony forming unit. Aff. Mat. Library:
Affinity maturation library. All yields in the figure are calculated based on concentration measurement, volumes, and length of insert. In the end, the produced affinity maturation libraries for H05, H09 and G03 had a coverage of respectively 54 %, 24 % and 51 % and were covered 2.1 times, 6.5 times and 11.9 times in the produced phage libraries.
Assessment of the affinity maturation library H05 66 % (42 out of 64) of the sequenced oligos did not contain errors such as mutations, insertions or deletions. 3 % of sequences contained insertions (2 out of 64 clones). 1 %
of all sequences (1/64) had 1 mutation outside of the randomized positions. 17 % (11 out of 64 clones) contained a pre-mature stop-codon, which yields undesirable truncated proteins. 6 % (4 out of 64 clones) was sequenced with low quality. % (3 out of 64 clones) did not show any sequence at all. 2
% had several errors, 1 clone had 2 deletions and 1 mutation. This is illustrated Figure 7.
Figure 7. A circle diagram illustrating the evaluation of the library design of H05 with respect to the percentage of high fidelity affibody inserts, mutations, insertions, stop codons, low quality sequences.
All the full-length sequences were analyzed.
The relative frequency of all aas were calculated, and the difference in relative frequency in comparison to the basic frequency (Every aas distributed equally) were calculated, and two heat-maps were made based on these calculations. The heat- maps can be seen in Figure 8.
The difference between the relative frequency and the basic frequency is calculated by subtracting the observed from the basic frequency. In Figure 8B there can be seen that a few amino acids are overrepresented (positive value, color red), consequentially most of the amino acids are underrepresented (negative value, color purple).
Correct oligos
66 %
Insertions 3%
Mutations 1%
Affibody containing Stop codon
17%
Low Quality Sequencing
6%
No sequences
5%
Multiple:
2%
H05 Sequence Analysis
14 Assessment of the affinity maturation library
H09
72 % (46 out of 64) of the sequenced oligos did not contain errors such as mutations, insertions or deletions. 1 % (1 out of 64 clones) had a deletion. 19 % (12 out of 64 clones) contained a pre-mature stop-codon, which yields undesirable truncated proteins. 2
% of the sequence had several errors, 1 clone containing both 1 deletion and 1 mutation. 6
% (4 out of 64 clones) did not show any sequence at all. This is illustrated in Figure 9.
There was no insertion nor any sequence that had low quality.
Figure 9. A circle diagram illustrating the evaluation of the library design of H09 with respect to the percentage of high fidelity affibody inserts, mutations, insertions, stop codons, low quality sequences.
The H09 library followed the same distribution as for the H05 library, see Figure 10.
Correct oligos Deletions 72 %
1%
Affibody containing Stop codon
19%
Multiple:
2%
No sequences
6%
H09 Sequence Analysis
Figure 10. Assessment of H09 library design. A) The relative frequency of every aa at each randomized position. Spanning from 0 (white) where none of the aa is found in that position, to 0.2 (yellow) where an aa is found in that position in 20 % of the sequences. B) The relative frewuqncy of every aa at each randomized position in comparison to the basic frequency. A value negative value, in purple, indicates a lower occurrence the basic frequency, while a positive value, in red, indicates a higher occurance than the basic freuqncy, a value of zero (white) indicates that the prevalence is correlating to the design. There are many negative values, while a few postive ones, demonstrating that many aas are underrepresented in the library design and a only few overrepresented.
Figure 8. Assessment of H05 library design. A) The relative frequency of every aa at each randomized position. Spanning from 0 (white) where none of the aa is found in that position, to 0.2 (yellow) where an aa is found in that position in 20 % of the sequences.
B) The relative frewuqncy of every aa at each randomized position in comparison to the basic frequency. A value negative value, in purple, indicates a lower occurrence the basic frequency, while a positive value, in red, indicates a higher occurance than the basic freuqncy, a value of zero (white) indicates that the prevalence is correlating to the design.
There are many negative values, while a few postive ones,
demonstrating that many aas are underrepresented in the library design and a only few overrepresented.
15 Assessment of the affinity maturation library
G03
81 % (51 out of 64) of the sequenced oligos did not contain errors such as mutations, insertions or deletions. 5 % (3 out of 64 clones) had a deletion. 2 % (1 out of 64 clones) had an insertion. 11 % (7 out of 64 clones) contained a pre-mature stop-codon, which yields undesirable truncated proteins. 3
% (2 out of 64 clones) did not show any sequence at all. This is illustrated in a circle diagram, see Figure 11.
Figure 11. A circle diagram illustrating the evaluation of the library design with respect to the percentage of high fidelity affibody inserts, mutations, insertions, stop codons, low quality sequences.
In the same matter as for the H05 and H09 library. The distribution of aas in the G03 library followed the same distribution as the H05 and H09 library, see Figure 12.
Affinity maturation libraries were transformed into E. coli ER2738 and infected with helper phages
When the better ratio between phagemid pAffi1 and the library inserts had been established to 1:4, the large library ligations were performed followed by purification of the ligated plasmids. After purification, a gel electrophoresis was performed to confirm that ligation was successful, the insert at 143 bp, and successful ligation product over 4000 bp.
Figure 12 Assessment of G03 library design. A) The relative frequency of every aa at each randomized position. Spanning from 0 (white) where none of the aa is found in that position, to 0.2 (yellow) where an aa is found in that position in 20 % of the sequences. B) The relative frequency of every aa at each randomized position in comparison to the basic frequency. A value negative value, in purple, indicates a lower occurrence the basic frequency, while a positive value, in red, indicates a higher occurrence than the basic frequency, a value of zero (white) indicates that the prevalence is correlating to the design. There are many negative values, while a few positive ones, demonstrating that many aas are underrepresented in the library design and a only few overrepresented.
All ligations were successful, with only some weak bands indicating remains of library inserts still in samples H09 and G03, see Figure 13.
Figure 13. Abbreviations, L: Ladder, where the corresponding masses are labelled to the left of the figures. Gel verification of library ligations into pAffi1. H05 library shows no remains of the library inserts.
However, there are weak bands between 100-200 bp, marked with an arrow, for both H09 and G03 library ligations, indicating remains of library inserts.
In order to establish what amount of DNA to add into the electrocompetent E. coli ER2738 to achieve the highest transformation frequency, a test-electroporation was
Correct oligos
81 % Affibody
containing Stop codon
11%
Deletions 5%
Insertions
2 % No
sequence 3 %
G03 Sequence Analysis
16 performed. Ligation amounts of both 118 ng
DNA and 235 ng DNA of library H05 was used. When using 118 ng DNA, 5 times more transformants were achieved (2091 cfu/µg DNA versus 413 cfu/µg DNA).
After this, the transformation frequency was determined for H09 and G03 when using 238 ng DNA and 232 ng DNA ligation mix.
Approximately the same transformation frequency was achieved for library H09 (3564 cfu/µg DNA) and G03 (1553 cfu/µg DNA).
When transformation-frequency had been established, the libraries were transformed into E. coli ER2738. Since there only was 64 aliquots of 25 µl cells, the aliquots were distributed over the libraries accordingly to the transformation frequency. 6 electroporations were not successful and was therefore excluded in library amplification.
H05 library transformation
H05 library was transformed with 20 electroporations (eps), which resulted in a 3.0 times coverage of the whole library with a total number of 8.9*1010 colony forming units (cfu). The library was amplified 500 times during ON-culturing.
H09 library transformation
H09 library was transformed with 16 eps, which resulted in a 4.7 times coverage of the whole library with a total number of 4.7*1010 cfu. The library was amplified 850 times.
G03 library transformation
G03 library was transformed with 22 eps, which resulted in a coverage of 3.4 times, with a total number of 1.0*1011 cfu. The library was amplified 425 times.
Phage libraries
The three affinity maturation libraries H05, H09, G03 libraries were infected with helper phage M13K07 in order to produce phage libraries.
H05 phage library was estimated through titering to cover the corresponding affinity maturation library by 4 % with a size of 2.3*1010 cfu/ml. H09 phage library was estimated to cover of 7 % with a size of 4.1*1010 cfu/ml. G03 phage library was estimated to cover 12 % with a total size of 7.2*1010 cfu.
Affinity maturation libraries displayed on phages selected against PSMA transfected HEK293 cells
Phage Display of H05
Two cycles were conducted with H05.
Through PCR-screening there could be confirmed that a high percentage of the screened colonies carried an affibody inserts for both elution strategies. After the first round, pH elution track had 96 % and trypsin elution track had 91 % affibody inserts.
However, after the second cycle, the pH- elution track only had 52 % and trypsin elution track had 15 % affibody inserts. The estimated elution-titer had not increased much after two cycles. Data available in Table 3.
Table 3. Abbreviations: pH/T: indicates which method of elution that was conducted. Cfu: colony forming units. % Z: Percentage of colonies that holds an affibody-domain according to PCR-screening. % Dummy:
Percentage of colonies that holds a dummy-domain, remaining from not successful restriction of the pAffi1 vector. N/A: No data available yet.
Titers performed before and after each cycle, as well as the result of PCR screening of single colonies. High percentage of affibody insert in the screened colonies after cycle 1, however, a decrease of colonies containing affibody inserts to cycle 2.
Cycle pH/
T In
[cfu] Out [cfu] %
Z % Dummy %
Empty 1 pH 2.3T 2.3**101099 5.26.1**101066 96 91 4 4 0 4
2 pH T N/AN/A 4.3 8.7*107 52 25 23
*107 15 8 77
Phage Display of H09
Two cycles were conducted with H09. After the first round, a relative high percentage of the screened colonies carried affibody inserts.
The pH elution track had 75 % and the trypsin elution track had 83 % affibody
17 inserts. After cycle 2, the percentage of
affibody inserts had decreased to 38 % for the pH elution track, and to 25 % for the trypsin elution track. The estimated elution-titer had not increased much after two cycles. Data available in Table 4.
Table 4. Abbreviations: see table 3. High percentage of affibody insert in the screened colonies after cycle 1, however, a decrease of colonies containing affibody inserts after cycle 2. *: In cycle 1, trypsin elution track, there was a PCR product of un-expected size around 400bp.
Another band at 400 bp can be seen in combination with a dummy- vector.
Cycle pH/
T In
[cfu] Out [cfu] %
Z % Dummy %
Empty 1 pH 7.3T 7.3**101099 3.04.3**101066 75 83 8 4 17 8 2 pH T N/AN/A 5.2 8.8*107 38 40 23
*107 25 25 50
Phage Display of G03
Four cycles were conducted with the G03 library. The percentage of colonies that held affibody inserts in is high throughout every cycle 1-4, with exception of cycle 4, trypsin elution track, seen in Table 5.
Table 5. Abbreviations: pH/T: indicates which method of elution that was conducted. Cfu: colony forming units. % Z: Percentage of colonies that holds an affibody inserts according to PCR-screening. % Dummy:
Percentage of colonies that holds a dummy-domain, remaining from not successful restriction of the pAffi1 vector. N/A: No data available yet.
Titers performed before and after each cycle, as well as the result of PCR screening of single colonies. High percentage of affibody insert after every cycle, with exception of cycle 4, trypsin elution track. The titer decreases in the eluate between cycle 1-3 and decreases after in the eluate of cycle 4.
After cycle 3, the estimated titer in the eluate had increased 1000-fold in comparison to the estimated titer in the eluate from cycle 1. The titer decreased again after cycle 4. This can be seen in Figure 14.
Sequence analysis of G03 cycle 3
After three cycles 48 clones for each elution strategy were sent for sequencing. Every sequenced clone was unique. The relative frequency of each aa were calculated and illustrated in Figure 13. Any pattern in preferred amino acids in a certain position cannot be seen in either elution strategy, see Figure 15 and Figure 16.
Figure 15. Amino acid distribution of sequenced clones from the G03 cycle with lowering pH as elution strategy. No pattern of a preferred amino acid can be seen.
1 100 10000 1000000 100000000 1E+10 1E+12
In cycle 1 Out
cycle 1 Out cycle 2 Out
cycle 3 Out cycle 4
Estimated no. of cfu
G03 library elution curve
pH Elution Trypsin elution
0 10 20 30 40 50 60 70 80 90 100
1 2 3 4 5 6 7
Amino Acid Distribution (%)
Randomized positions Amino Acid distribution of G03, cycle 3, acid Elution
Figure 14. An elution curve illustrating the titer, an estimation of phages, in the eluates in comparison to what we added to the first cycle.
In the unit cfu (colony forming unit). The Y-axis is a logarithmic scale of the titers in every cycle.
18
Figure 16. Amino acid distribution of sequenced clones from the G03 cycle with trypsin as elution strategy. No pattern of a preferred amino acid can be seen.
Discussion
The aim of this thesis was to produce and select for improved binders of three designed affinity maturation libraries, denoted H05, H09 and G03. The libraries were generated, 2 selection cycles (H05 and H09) and 4 selection cycles (G03) respectively were performed, so far with a too low enrichment of potential binders. Therefore, more cycles have to be conducted in order to find potential PSMA binders.
The library preparation started with an amplification of the fragments, followed by a purification and digestion to expose the RE- sites. The purification of the digested library was unsuccessful and there were still un- digested library fragments remaining. Hence, a re-digestion was performed, followed by a gel extraction, to fully purify of the fragments from remains.
The H05 and H09 libraries were found to be pure. Despite some impurities in the G03 library it was considered an insignificant amount in comparison to the library fragments.
Different ligation ratios between library insert and phagemid pAffi1 were tested, with the
aim of finding a high transformation frequency. Molar excesses of 4 and 5 were tried, and it was determined that a 4-fold excess (library insert to phagemid) was the more optimal between the two tested ratios.
The transformation frequency was established and the big library transformations into E. coli ER2738 was conducted. The affinity maturation libraries were estimated to be covered to 3.0 times (H05), 1.6 times (H09), and 3.4 times (G03).
This means that all possible combinations of the 7 randomized positions is present in more than 1 copy. This is required when displaying the libraries at phages, with the aim of displaying every sequence combination when selecting for binders.
The obtained libraries contained a high percentage of sequences in accordance to the design, with only some percentage of clones containing mutations, deletions and insertions. The errors can be due to mistakes in the production of the library oligonucleotides, and specifically that all oligos containing errors is not removed when purifying the library fragments. Some errors can also be due to faulty sequencing data.
Some percentage of the clones had low quality. This can be due to errors when sequencing, as well as due to one clone containing more than one sequence, giving rise to two or more signals when sequencing.
Certain aas were overrepresented in each randomized position and this is characteristic for NNK-codons [14]. Some aas were not present in every position, however this is probably due to sequencing too few clones (64 clones of a library with a size of 3*109 different combinations).
The affinity maturation libraries were infected with helper phages in order to assemble and produce corresponding phage libraries. Approximately 1-10 % of the phages displays a protein on the surface when expressed from a phagemid [18],
0 10 20 30 40 50 60 70 80 90 100
1 2 3 4 5 6 7
Amino Acid Distribution (%)
Randomized positions Amino Acid distribution of G03, cycle 3, trypsin elution
19 consequentially it is assumed that there are
similar conditions for an affibody. Therefore, the produced phage libraries ought to cover the libraries at least 10-100 times in order to display every affibody molecule present in the transformed library.
The prepared phage stocks covered the affinity maturation libraries with 4 % (H05), 7 % (H09), and 12 % (G03), assuming that 5
% of the phages are displaying the affibody of interest on its surface. The phage stocks did not cover the affinity maturation libraries, however, since all sequences are based on potential binders, the potential of a sequence being a binder is higher. Therefore, the libraries were further used for selections against PSMA.
For both the H05 and H09 selections, the percentage of affibody inserts was high in the eluate from cycle 1 but decreased in the eluate from cycle 2. This indicates that a lot of material was lost in both the selections during cycle 2.
Titers are only an estimation of a phage concentration, and concentrations of different phage solutions are only comparable if the titering is performed consistently.
The G03 selection had high percentage of affibody inserts throughout all cycles, with exception of cycle 4 with trypsin elution, were only 54 % held an affibody insert. The high percentage of affibody inserts is an indicator that the binders are still selected and enriched for in every cycle. After cycle 3, the titer in the eluate increased 1000-fold in comparison to the titer in eluate from cycle 1, which is an indication that potential binders are enriched for. Therefore, 48 clones from each track (pH elution and trypsin elution) were sent for sequencing from cycle 3. The titer decreased in the eluate from cycle 4 for both tracks.
However, since cycle 4 were performed with more stringent conditions, it is expected that the titer decreases.
Even though some PCR-products of screened colonies in the phage selections cannot be seen in some cases it does not imply that the affibody insert is not present in the clone, since the plasmid holds the antibiotic resistance. There could simply be a problem with amplifying the DNA.
Out of the 48 sequenced clones for each track, every clone was unique. Since three cycles were performed some level of convergence in the sequences is expected.
However, nothing precise can be said about the convergence. This is also an indicator that more selection cycles have to be performed.
In the nearest future, more selection cycles are going to be conducted with the libraries until an enrichment can be observed through sequencing. Cycle 2 for both H05 and H09 libraries are going to be done again since the percentage of affibody inserts was low in the eluate of cycle 2.
Acknowledgements
First, I would like to thank my supervisor Hanna Lindberg. Your guidance and support through this project has been the very best, thank you for always keeping the door open.
To work with you has been a real pleasure.
Stefan Ståhl, thank you for giving me the opportunity to do my master’s thesis in your research group. It has been a couple of very welcoming and fun months.
Thanks, VBG, for being the world’s best group, in many aspects.
Charlie, thank you for assisting me with library transformations.
I want to extend my gratitude to everyone attending the phage-meetings for all encouragement and feedback.
Finally, I want to thank all co-workers and new-found friends on Floor 3, AlbaNova. It would not have been the same without this warm atmosphere.
