The ETH-2 human antibody phage library

Schematic representation of the selection of antibodies from phage display libraries, showing the enrichment of an antigen-specific phage-antibody (circle) from a background of non specific phage-antibodies (square).

The ETH-2 library is a modified version of the synthetic antibody library of:   Pini, A., Viti, F., Santucci, A., Carnemolla, B., Zardi, L., Neri, P., Neri, D. "Design and use of a phage display library: human antibodies with subnanomolar affinity against a marker of angiogenesis eluted from a two-dimensional gel"  J.Biol. Chem. (1998)  273, 21769-21776.

Version: 30 September 1999

1) Foreword

Thanks for your interest in the ETH-2 synthetic human antibody library.
From this library, monoclonal antibody fragments against a virtually infinite number of different antigens can be produced. The antibodies can be expressed in bacteria (typical yields: 1-50 mg/litre in shaker flasks) and affinity-purified on Protein A Sepharose. They can be used for practically all standard antibody-based assays (western blotting, ELISA, immunohistochemistry, immunoprecipitation, etc.).
Very limited equipment (normally available in Biochemistry or Molecular Biology laboratories) is required. Typically, 1-2 weeks of (limited amount of) work are necessary to produce antibodies against a purified antigen, by a normally skilled scientist.
The ETH requests that you sign a suitable Transfer Agreement.
In order to obtain large dimensionality libraries, several electroporations were needed, resulting in "sub-libraries". The ETH-2 library consists of four sub-libraries, prepared in identical fashion but on different days: ETH-1, ETH-1a, ETH-1b, ETH-1c. More electroporations of the library are to be expected, so that in the future sub-libraries ETH-1d, ETH-1e may be added to the collection. You may want to contact us for updated versions of the library.

The sub-libraries should be mixed and used together.

If commercially relevant antibodies arise from your research and you want to exploit them, you should contact the ETH Transfer office to discuss proprietary aspects.

2) Description of the Library

Thanks to the pioneering work of Dr. Greg Winter (Cambridge Centre for Protein Engineering, MRC Centre, Cambridge UK), antibody fragments can nowadays be isolated from naive phage display libraries without immunisation, by-passing hybridoma technology (Winter et al. (1994) Annu. Rev. Immunol. 12, 433).

We have aimed at producing a robust and flexible synthetic human antibody phage display library, which features a number of technical improvements, developed in our laboratory in the last few years.

The ETH-2 library is a phage disply library of human recombinant antibodies in scFv format. Antibodies in scFv format consist of a single polypeptide chain, comprising an antibody heavy chain variable domain (VH) linked by a flexible polypeptide linker to a light chain variable domain (VL) (see Fig. 1).

Figure 1: Schematic representation of different formats of antibodies and antibody fragments.

Antibodies are assembled in humans from approximately 50 different VH and 70 VL germline genes. However, only a few of them dominate the functional repertoire.
Since our purpose was to generate a highly diverse library of functional antibodies with similar characteristics, we opted to utilise only three antibody germline gene segments (DP-47 for the heavy chain, DPK-22 and DPL-16 for the light chain; Tomlinson et al. (1996) J. Mol. Biol. 256, 813) and to produce a large repertoire by appending short variable complementarity-determining regions 3 (CDR3) onto them.
The CDR3 of the heavy chain is the largest and most diverse loop of the antigen recognition site of the antibody. We appended at position 95 of the VH (see Chothia and Lesk (1987) J. Mol. Biol., 196, 901 for numbering) random loops of 4, 5 or 6 amino acids.

....... C⁹² A R (X)_4-6 F D Y ......

Similarly, we introduced diversity in six aminoacid positions in the CDR3 of light chains assembled with DPL-16 or DPK-22 germline genes (Figure 2):

Figure 2: Antigenís eye view of an antibody fragment of the ETH-2 library.

A phagemid vector, pDN 332 (Figure 3), derived from the phagemid pHEN1 (Hoogenboom et al., 1991), was used for cloning of the library.
pDN332 has the phage origin of replication, M13 ori (allowing the genome to be packaged into the phage particles), the E. coli origin of replication, colE1 ori (enabling the phagemid to replicate in the E. coli host cells), an ampicillin resistance (for selection of phagemid-containing colonies), a peptide leader, pel B (for secretion of phage protein III fused with the antibody fragment to the periplasm of the bacteria) and the lacZ -promotor (for regulation of expression with promotion by IPTG and inhibition by glucose). Two restriction sites (NcoI and NotI) were used for cloning.  Furthermore, an amber codon between the tag sequence (described below) and the phage protein III provides the opportunity to either produce the antibody fused to the coat protein of the phage, using a E. coli suppressor (supE) strain, for example TG1, or to produce soluble antibodies by expressing the vector in a non-suppressor strain, such as HB 2152.

Figure 3: Schematic representation of the antibody library cloned in pDN332.

The pDN332 phagemid vector appends at the C-terminal extremity of the recombinant antibody a D₃SD₃-FLAG-HIS₆  versatile tag which provides:
- a phosphorylation site (Neri et al. (1996) Nature Biotechnol., 14, 485), for the antibody labeling with radioactive P-32.
- the FLAG-tag sequence for detecting the antibody with the anti-FLAG-tag M2 antibody
-  the (His)₆ sequence, allowing rapid purification by nickel-chelate chromatography, and the detection and immobilisation with suitable nickel chelating reagents (Qiagen).
- the recognition site for the endopeptidase enterokinase.

The sequence of pDN332 between the NotI site and the amber codon preceeding the gene III, coding for the C-terminal antibody peptidic tag, is depicted below:

  Not 1        Phosphorylation site
        R    A       A       D  D  D  S    D    D
5í-GCG GCC GCA GAT GAC GAT TCC GAC GAT
FLAG-tag
D      Y      K      D        D        D      D      K   #  H      H       H       H      H       H  amber
GAC TAC AAG GAC  GAC GAC GAC AAG# CAC CAT CAC CAT CAC CAT TAG-3í

All antibody fragments of the library contain a VH DP-47 germline segment, and can be purified on Protein A sepharose (Hoogenboom and Winter (1992) J. Mol. Biol. 227, 381). Antibody purification by immobilised metal chelate affinity chromatography (Qiagen) is also possible.
Furthermore, antibodies can be purified from bacterial supernatants on antigen columns, or on columns with an immobilised M2 anti-FLAG antibody (Sigma).

3) How to store and use the library

The libraries are generally stored at -70°C as bacteria harbouring phagemid or phage DNA (at least 10⁸ bacteria per 10 µl in 2xTY-15% glycerol). To make phage for selection, the primary library is grown, infected with helper phage and the phage harvested from bacterial supernatant.

IT IS RECOMMENDED THAT A LARGE NUMBER OF PHAGE ALIQUOTS IS PREPARED FROM THE PRIMARY ALIQUOT, AND THAT SECONDARY STOCKS ARE NOT USED

For a library of 5x10⁸ clones, inoculation with at least 5x10⁸ bacteria (50 µl stock, and preferably more) is essential.

ETH-2 library

1. Inoculate each of the glycerol stocks into 50 ml 2xTY-AMP-GLU, until OD = 0.1
2. Grow to OD 0.4 - 0.5 at 37°C (about 1-1.5 hrs). An aliquot can be taken at this stage to make a secondary stock (see below).
3. Infect each of the 50ml cultures with helper phage (0.5 ml of >10¹² tu/ml VCS-M13) (The bacterial concentration is = 8x10⁸ bacteria/ml with an OD₆₀₀= 1.0 ). Infections are carried out at 37°C in water bath for at least 30 min.
4. Spin down the infected bacteria at 3,300 g for 10 min. Gently resuspend the pellet in  500 ml of 2xTY-AMP-KAN. (Total volume: 2 litres in four 2 l flasks).
   Sub-libraries can be mixed at this point.
5. Incubate with shaking 30°C overnight.
6. Spin down the culture at 10,800 g for 10 min and immediately PEG precipitate the  phage from the supernatant (as in "SECTION 4.2.3")
7. Resuspend phages in 15 ml PBS + 10% glycerol. Titrate phages. Typically 30 aliquots of 0.5 ml each can be produced. Each aliquot will be sufficient for at least one selection. (use at least 10¹¹ (and ideally 10¹²) phages for selection)
8. Store aliquots at -20°C. These aliquots will be all of the same quality and will ensure the reproducibility of the selection procedure.

MAKE A TEST SELECTION USING BSA AS ANTIGEN: THIS WILL GIVE YOU CONFIDENCE WITH THE PROCEDURES, AND WILL ENABLE TO CHECK WHETHER YOUR LIBRARY PREPARATION IS OK.

Secondary Stocks.

An excellent secondary stock is the frozen phage obtained from the primary library stock. Antibodies on phage may be proteolysed or get denatured, but phage particles are resistant and maintain a good infectivity. A titre of phage larger than the library size can therefore be used to infect exponential TG1 and produce a second generation of phage library.
Secondary bacterial glycerol stocks can also be prepared from the primary library culture described.

4)Protocols for the isolation and use of antibodies from phage display libraries

4.1 Materials

4.1.1 Buffers and solutions

• PBS - 50 mM phosphate, pH 7.2, 100 mM NaCl in water
• TBS - 50 mM Tris, pH 7.4, 100 mM NaCl in water
• TBS; - 50 mM Tris, pH 7.4, 100 mM NaCl, 1 mM CaCl2 in water
• HB; - 50 mM HEPES, pH 7.0, 100 mM NaCl in water
• X% MPBS - PBS containing X% (w/v) dried skimmed milk, for example Marvel)
• 10% Tween 20-(SIGMA)

4.1.2 Growth media

• 2xTY  -  see Nagai and Thogersen (1987) Methods. Enzymol. 153, 461.
• TYE - see Miller (1992) "A short course in bacterial genetics: a laboratory manual and handbook for E.coli and related bacteria. Cold Spring Harbour Laboratory Press, New York, 1992.
• minimal  - see Miller (1992)
• H-top - see Miller (1992)
• AMP - medium contains 100 µg/ml ampicillin.
• KAN - medium contains 25 µg/ml kanamycin.
• TET - medium contains 12.5  µg/ml tetracycline
• GLU - medium contains glucose to a final concentration of 1%. If different concentrations of glucose are to be used, this will be indicated (e.g. 0.1%- GLU = final concentration of glucose is equal to 0.1%).

4.1.3  Other reagents

• Streptavidin coupled M-280 Dynabeads (Dynal).
• NHS-SS-Biotin (Pierce)
• Horse radish peroxidase (HRP) conjugated anti-M13 monoclonal antibody (Pharmacia)
• HRP substrate solution, e.g. BM blue POD substrate (Boehringer).
• anti-FLAG-M2 antibody to detect flag -tagged antibody fragments (Sigma)
• HRP conjugated monoclonal anti-mouse IgG Fc secondary antibody (e.g. Cat No. A-2554, Sigma).
• Nitrocellulose filter (Schleicher & Schuell) <
• 96well microtiterplate for cell cultures (Corning)
• Detection reagents for Western Blot and Dot blot using Chemiluminiscence : ECL (Amersham)
• MicroTest III flexible assay plates (Falcon)
• BM blue POD substrate (Boehringer).

4.1.4  Bacteria and Phage

Two bacterial strains are used: E. coli suppressor strain TG1 (K12, D(lac-pro), supE, thi, hsdD5/F'traD36, proA+B+, lacIq, lacZDM15) for propagation of phage particles; and HB2151 non-supressor strain (K12, ara, D(lac-pro), thi/F'proA+B+, lacIqZDM15) for expression of antibody fragments. Phage/phagemid infect F+-E. coli via the sex pili. For sex pili production and efficient infection E. coli must be grown at 37°C and be in log phase (OD at 600 nm of 0.4-0.5; see Methods).
A variety of helper phages are available for the rescue of phagemid libraries. VCS M13 (Stratagene) and M13 K07 (Pharmacia) can be purchased in small aliquots: large quantities for rescue of phagemid libraries can be prepared as described in the Methods section.
For growth of pHEN1-derived phagemids use AMP-GLU media.
For long term storage at -70°C, bacteria containing plasmids or phage replicative form are resuspended in 2xTY-15% glycerol in 2ml plastic vials. 4.2 Methods

4.2.1 General

Phage contaminations in flasks, bottles etc. may accumulate as one performs selections from phage display libraries. Precautions need to be taken throughout the protocol to avoid any carry over of phage. Autoclaving alone is not sufficient enough to remove all phage contamination. Wherever possible use devoted pipettes and disposable plasticware. Ensure that all non-disposable plasticware and glassware (e.g. centrifuge bottles) is completely phage free by soaking them for 1 hr in 2% (v/v) sodium hypochlorite, followed by extensive washing and then autoclaving. The use of polypropylene tubes is recommended as phage may adsorb non-specifically to other types of plastic.

Unless otherwise stated in the text:

1. Bacteria and phage are grown in liquid media at 37°C in an orbital shaker at 250-300 rpm.
2. Agar plates are grown at 37°C. However, if time allows it, it is advisable to grow  plates always at 30°C. Two sizes of agar plates are used: round 9 cm diameter for titre determination, or large (20 cm diameter) round plates for rescue of bacteria infected by phages at the end of a selection round.
3. Phage infections are carried out for 30 min in a 37°C water bath without shaking,  using exponentially growing bacteria.
4. Absorbance of bacterial cultures is measured at 600 nm.
5. Centrifugations are carried out at 4°C.

4.2.2 Bacteria and helper phage

Exponential bacterial cultures

1. Transfer a bacterial colony from a minimal media plate into 5 ml of 2xTY medium and grow overnight.
2. Next day subculture by diluting 1:100 (OD 0.1) into fresh 2xTY medium, grow until OD 0.4-0.5 and then infect with phage.

Efficiency of infection is greatly reduced above OD = 0.5. Preparation of helper phage

1. Infect 200 µl E. coli TG1 (or other suitable strain) at OD 0.2 with 10 µl serial dilutions of helper phage (in order to get well separated plaques). Add to 3 ml H-top agar (42°C) and pour onto warm TYE plates. Allow to set and then incubate overnight.
2. Pick a small plaque into 3-4 ml of an exponentially growing culture of TG1 (see above). Grow for about 2 hr.
3. Inoculate into 500 ml 2xTY in a 2 litre flask and grow for 1 hr and then add kanamycin  (25 mg/ml in water) to a final concentration of 50-70 µg/ml. Grow for a further 8 - 16  hr.
4. Spin down bacteria at 10,800 g for 15 min. To the phage supernatant add 1/4 volume PEG/NaCl (20% polyethylene glycol 6000-2.5 M NaCl) and incubate for a minimum of 30 min on ice. Spin 10,800 g for 15 min.
5. Resuspend pellet in 2 ml TE and filter sterilize the stock through a 0.45 µm filter  (Minisart NML; Sartorius).
6. Determine the titre of the stock and then dilute to about 1x10¹² p.f.u./ml. Store  aliquots at -20°C.

4.2.3 Purification of phage

The phage can be concentrated (and any soluble antibodies removed) by precipitation with polyethylene glycol (PEG) 6000. The protocol described here is valid for 300 ml phage-containing supernatant, but can be up- or down- scaled proportionally.

1. Transfer the phage supernatant to a tube and add 75 ml PEG/NaCl (20% polyethylene  glycol 6000-2.5 M NaCl).
2. Mix well and leave for a minimum of 1 hr at 4°C or at least 40 min on ice.
3. Spin 10,800 g for 30 min.
4. Resuspend the pellet in around 40 ml water, and add 1/5 volume PEG/NaCl (e.g. 10  ml PEG/NaCl to 40 ml. Mix and leave for a minimum of 20 min at 4°C.
5. Spin 10,800 g for 30 min and aspirate off the supernatant.
6. Respin briefly and aspirate off any remainings of PEG/NaCl.
7. Resuspend the pellet in 2 ml PBS . Phage yields are normally 1-5 x 10¹³ t.u./ml  phage suspension.
8. Spin 3,300 g for 10 min or 11,600 g for 2 min to remove any residual bacterial cell  debris.
9. Store the phage supernatant either at 4°C for short term storage or in PBS-15%  glycerol for longer term storage at -70°C.

4.2.4 Selection of Phage.

The phage libraries can be selected using immunotubes, biotinylated antigen in solution or affinity chromatography. Here, details are present for the first two methods, which are the most commonly used. The stringency of selection conditions can be adjusted; particularly during later rounds of selection, the stringency can be increased, for example by reducing the density of coating of antigen to solid phase. To avoid non-specific binding of phage to surfaces, powder milk can be used as a blocking agent.

Selection on Immunotubes.

Coating Immunotube with Antigen and Blocking.

1. Coat a Nunc-immunotube (5 ml tube, Maxisorp) overnight using 4 ml of antigen.
   The efficiency of coating can depend on the antigen concentration, the buffer and the  temperature. Usually 10-100 µg/ml antigen in PBS or 50 mM sodium hydrogen  carbonate, pH 9.6 at room temperature (rt), is used.
2. Rinse the tube 3 times with PBS and then block with 4 ml 2% MPBS at rt for 2 hr.
3. Rinse the tube 3 times with PBS.

First Round of Selection.

1. Add 10¹²-10¹³ t.u. phage library in 2 ml PBS to immunotube containing 2 ml of 4%  MPBS (to give a final concentration of 2% MPBS).
2. Seal the tube with parafilm and mix by repeated inversion at rt for 30 min.
3. Allow the immunotube to stand upright at rt for 1.5 hr and then throw away the unbound phage in the supernatant.
4. Rinse the tube 10 times with PBS-0.1% Tween 20 and then 10 times with PBS. Each washing step is performed by pouring buffer in and then immediately out.
5. Shake out excess PBS from the tube and elute the bound phage by adding 1 ml 100 mM triethylamine (700 µl triethylamine (7.18 M) in 50 ml water, diluted on the day use).
6. Mix by repeated inversion at rt for 5-10 min (but not longer! phage looses infectivity!).
7. Pour into a microfuge tube containing 0.5 ml 1 M Tris/Cl, pH 7.4 and vortex to  neutralise the triethylamine. Keep on ice until the TG1 bacteria reach OD = 0.4-0.5.
8. Infect 10 ml TG1 culture with 1.5 ml of collected phage (at further rounds of selection, 1.4 ml should be used to infect TG1, while keeping 100 µl for trace infections of HB2151. Typically, 10 µl collected phage can be used to infect 1 ml exponential HB2151 for soluble ELISA screening).
   Some people want to ensure that they recover also some phage that may have not been eluted from the immunotube by the triethylamine, so they add to the immunotube 200 µl of 1 M Tris, pH 7.4 followed by 4 ml TG1 culture and incubate 30 min at 37°C for infection, then pool the fractions. In general, this precaution is not needed.
9. Make a series of 2-3 100-fold serial dilutions in 2xTY using new pipette tips each  time.
   We usually plate 10 µl of the infected 10 ml TG1 onto selective TYE-AMP-GLU plate (titre = colonies x 10³) and 10 µl of a 1:100 dilution (titre = colonies x 10⁵).
10. Spin down the remaining infected TG1 at 3,300 g for 10 min.
11. Resuspend the pelleted bacteria in 0.5-1 ml 2xTY and spread on large agar plates of  the appropriate selective TYE agar.
12. Incubate at 30°C overnight, or until colonies are visible.

Further Rounds of Selection (Phagemid Vectors).

1. Next day, monitor the titre of the selection by counting colonies on the plates from the dilution series experiment. Add 5-10 ml 2xTY-15% glycerol to the large agar plates and GENTLY loosen the bacteria with a glass spreader until a homogenous suspension is obtained.
2. Inoculate 50 ml of 2xTY-AMP-GLU with enough bacterial suspension (typically 20 80 µl) to yield an OD(600nm) = 0.05-0.1.
   Store the remaining bacteria at -70°C in 15% glycerol.
3. Rescue as in "Growth of libraries after selection" (see the end of this section).
4. PEG precipitate (as in "Purification of Phage"), and resuspend the second phage pellet in 2 ml of PBS.
5. Store a 1 ml aliquot of the phage at 4°C and use the other 1 ml aliquot for the next  round of selection. Repeat the selection for another 3 to 4 rounds.

Selection using Biotinylated Antigens.

Here the phage antibodies react with biotinylated antigen in solution, and the complex is then captured using streptavidin coupled to Dynabeads (Dynal).

Binding Phage to Soluble Biotinylated Antigen..

The antigen first needs to be biotinylated with NHS-SS-Biotin (Pierce) according to the manufacturer instructions (e.g. for FITC-BSA and NIP-BSA see Griffiths et al. 1994), and is not described here. The phage is then bound to the biotinyated antigen.
Sometimes, biotinylating reagents may loose their reactivity, typically by hydrolysis of the NHS-ester moiety. It is good practice to check the quality of the reagent by biotinylating 1 mg hen egg lysozyme at (3:1) biotin-protein ratio. The biotinylated lysozyme can be analysed on a native polyacrylamide gel (a conventional Laemmli gel, but: i. without stacking gel; ii. omitting SDS in all the buffers; iii. using 80 mM GABA + 20 mM acetic acid, pH 4.8, as gel and running buffer) run reversing the polarity of the electrodes (lysozyme is positively charged at pH 4.8). The unmodified lysozyme and lysozyme with 1, 2, 3 etc. biotin molecules will migrate as distinct bands.

1. Mix together:
   • 2.5 ml of phage in PBS (10¹²-10¹³ t.u)
   • 2.5 ml 4% MPBS
   • 50 µl 10% Tween 20 (0.1% final concentration)
   • S-S-biotinylated antigen to give a final concentration of no more than 50 nM (for  excess of streptavidin-Dynabeads over antigen)
2. Gently rotate on an inclined wheel for one hr at rt (but shorter incubation times favour the selection of antibodies with fast k_on)

Blocking the Dynabeads.

1. Block 100 µl (per selection) streptavidin M-280 dynabeads by adding 600 µl 5% MPBS for > 15 min at RT.
2. Separate the beads from the MPBS with a MPC-E (Magnetic Particle Concentrator  from Dynal) for 4-5 min.
3. Resuspend beads in 200 µl PBS.

Selection.

1. Add 150 µl of blocked streptavidin Dynabeads to the phage bound to the S-S biotinylated antigen.
2. Gently rotate for 15 min at rt.
3. Separate with a MPC-E and pour off the supernatant.
4. Wash five times with 1 ml PBS 0.1% Tween 20.
5. Wash five times with 1 ml PBS. For each wash, resuspend the beads and bring down with a MPC-E.
6. Finally elute the phage from the beads by resuspending in 300 µl PBS-50 mM  dithiothreitol. If standard NHS-LC-biotin is used as biotinylating reagent, the elution  can be performed using 100 mM triethylamine (but antibodies against streptavidin will also be selected!!).
7. Stand for 5 min at rt and separate the beads again with the MPC-E.
8. Use 150 µl of the phage supernatant to infect 10 ml TG1, and store the remaining  phage at 4°C.
9. Remove a small aliquot of the infected TG1 and make a series of 4-5 100-fold serial  dilutions in 2xTY and plate 100 µl each on the appropriate selective TYE plates.
10. Pellet the remaining bacteria by spinning for 10 min at 3,300 g.
11. Resuspend in 1 ml 2xTY and spread on a large agarplate with appropriate selective  TYE agar (amp)
12. Incubate at 30°C overnight, or until colonies are visible and then repropagate phage as in "Selection on Immunotubes: Further Rounds of Selection".
13. Repeat the selection procedure another 3 to 4 times (use a 1 ml aliquot for selection and store the remaining phage at 4°C). For the second and subsequent rounds of selection,
    a) mix 1 ml phage from the previous round of selection with:
    • 0.5 ml 6% MPBS
    • 10 µl Tween 20
    • Biotinylated antigen to give a final concentration of 50 nM.

    
    b) use only 150 µl of blocked streptavidin-coated Dynabeads.
    

    Growth of libraries after selection

    1. Inoculate 20-80 µl of bacterial library stock (approx. 5x10⁸) clones) into 50 ml 2xTY-AMP-GLU (you should aim at OD = 0.05 - 0.1).
    2. Grow to OD 0.4-0.5. at 37°C (about 1-1.5 hrs).
    3. Infect 10 ml of this culture with helper phage in a ratio of around 20:1 phage:bacteria for 30 min. at 37°C.
       (The bacterial concentration is = 8x10⁸ bacteria/ml with an OD₆₀₀=1.0 ).
       Typically we use 100 µl VCS-M13 (>10¹² tu/ml) per 10 ml culture.
    4. Spin down the infected bacteria at 3,300 g for 10 min. Gently resuspend the pellet in  100 ml of 2xTY-AMP-KAN.
    5. Incubate at 30°C overnight, shaking.
    6. Spin down the culture at 10,800 g for 10 min and immediately PEG precipitate the  phage from the supernatant (as in "Purification of phage"). The phage,  representing the library, should be used within few days (proteases!!), or can be stored in aliquots at -20°C.

    4.2.5 Screening Phage by ELISA.

    After the third round of panning, it is usually good practice to monitor the progress of the selection by ELISA, either with soluble antibodies, or using antibodies on phage.

    When performing phage ELISA, one often chooses a polyclonal phage ELISA protocol, using phage produced at the end of each round of panning. The amount of antigen needed is minimal, since only few wells have to be coated. Alternatively, individual colonies of TG1 bacteria infected with phage from the desired round of panning can be used to produce phage particles in bacterial supernatants, which are then tested in ELISA.
    However, since for most applications you will be using soluble scFv fragments, you may as well directly screen by ELISA soluble antibodies in bacterial supernatants.
    The diversity of the selected phage monoclonal antibodies may be assessed by sequencing of the vector DNA.

    ELISA wells are usually coated with antigen overnight at rt with 10-100 µg/ml of antigen in either PBS or 50 mM sodium hydrogen carbonate, pH 9.6. Binding of phage is detected by HRP-anti-M13 conjugate (Pharmacia). Unless otherwise stated in the text, for the ELISA, "incubations" are for 60 min at rt, using 100 µl samples in 2% MPBS. "Washing" involves flipping over the ELISA plates to discard excess liquid, washing three times with PBS-0.1% Tween 20, then 3 times with PBS.

    Polyclonal Phage ELISA.

    1. Coat MicroTest III flexible assay plates (Falcon) with 100 µl per well of protein  antigen.
    2. Rinse wells 3 times with PBS, and block with 200 µl per well of 2% MPBS for 2 hr at  rt.
    3. Rinse wells 3 times with PBS. Add 10 µl PEG precipitated phage from the stored aliquot of phage at the end of each round of selection (about 10¹⁰ t.u.). Make up to  100 µl with 2% MPBS. Alternatively, 20 µl 10% MPBS and 80 µl phage-containing  supernatants can be used. Incubate and wash.
    4. Add appropriate dilution (try 1:1000) of HRP-anti-M13. Incubate and wash.
    5. If sheep anti-M13 antisera is used in step 4, incubate with HRP-anti-sheep antisera and  wash.
    6. Develop with HRP substrate solution, e.g. with ready-to-use BM blue POD substrate  (Boehringer).
    7. Add 100 µl to each well and leave at rt for 10 min. A blue colour should develop.
    8. Stop the reaction by adding 60 µl 1 M sulphuric acid. The colour should change to  yellow.
    9. Read the OD at 650 nm and at 450 nm. Subtract OD 650 from OD 450.

    Monoclonal Phage ELISA.

    For better aeration during phage growth or induction in 96-well plates, the plate lid may be removed.

    1. Inoculate individual colonies from the plates of phage infected TG1 spread after the  desired round of selection into 150 µl 2xTY-AMP-GLU into 96-well plates. As  inoculating device, use a toothpick or a yellow pipette tip. After inoculation, use the  same device to touch a replica 2xTY-AMP-GLU plate with a numbered grid attached at  the bottom. The colonies from the replica plate corresponding to positive clones in  ELISA can be used to prepare a glycerol culture or to PCR amplify the antibody gene.
    2. Grow for 3 hrs.
    3. To each well add 25 µl 2xTY-AMP-GLU containing 10⁹ p.f.u. of helper phage.
    4. Stand for 30 min at 37°C. Spin at 1,800 g for 10 min and then aspirate off the  supernatant.
    5. Resuspend bacterial pellet in 200 µl 2xTY-AMP-KAN.  Grow overnight at 30°C.
    6. Spin at 1,800 g for 10 min and use 50-100 µl of the supernatant in phage ELISA, as  above.

    4.2.6 Screening Antibody Fragments by ELISA

    ELISA for the antibody fragments either detects a peptide tag (for our library we use the anti-FLAG antibody M2) or VH of VH3 family (we would use in this case Protein A - horseradish peroxidase).
    We here provide a protocol for the assay using the M2 antibody.

    Antibody Fragments from Phagemid Vectors.

    1. Infect 1 ml culture of HB2151 with 10 µl (about 10⁵ t.u.) of the phage eluted and  neutralised at the last round of selection. Plate 0.1, 1, 10, 100 µl on TYE-AMP GLU. Incubate overnight.
    2. Inoculate individual colonies from the plates of phage infected TG1 spread after the  round of selection of interest into 180 µl 2xTY-AMP-0.1%GLU in 96-well plates.
       As inoculating device, use a toothpick or a yellow pipette tip.
    3. Grow for 2 hrs at 37°C.
    4. Transfer with a multichannel pipette 50 µl from each well onto a replica microtitre plate conatining 50 µl of 40% glycerol. The resulting glycerol stock can be frozen and used at later stage to rescue clones of interest.
    5. Add 50 µl 2xTY-AMP containing > 5 mM IPTG (to give a final concentration > 1 mM  IPTG). Continue shaking at 30°C for a further 16 to 24 hr.
    6. Spin at 1,800 g for 10 min and use 100 µl of the supernatant in ELISA.

    NOTES:
    1) Make glycerol stocks of positive clones at a very early stage, avoid propagating clones on agar plates or with long cultivation times: some antibodies are toxic to the bacteria and clones may loose antibody expression properties.
    2) As some antibodies are preferentially found in bacterial supernatants, some people freeze and thaw the microtitre plate before centrifugation, to release antibody from the periplasmic space.
    
    

    ELISA of Supernatants.
    Washing and coating is as in Screening Phage by ELISA. This is a 1-step procedure which works very well.

    1. Coat MicroTest III flexible assay plates (Falcon) with 100 µl per well of protein  antigen.
    2. Rinse wells 3 times with PBS, and block with 200 µl per well of 2% MPBS for 2 hr at rt.
    3. Rinse wells 3 times with PBS. Add 30 µl 10% MPBS and 80 µl supernatant to each  well.
    4. Immediately afterwards, add with a multi-channel pipette 11 µl freshly prepared  developing mix (e.g.,  anti-FLAG M2 (1:400 of a 4 mg/ml stock; Sigma) + anti- mouse HRP (1:100 of Sigma A-2554) in 2%MPBS.
    5. Incubate for 20-60 min and wash.
    6. Develop with HRP substrate solution, e.g. with ready-to-use BM blue POD substrate (Boehringer).
    7. Add 100 µl to each well and leave at rt for 10 min. A blue colour should develop.
    8. Stop the reaction by adding 60 µl 1 M sulphuric acid. The colour should change to  yellow.
    9. Read the OD at 650 nm and at 450 nm. Subtract OD 650 from OD 450.

    4.2.7 Preparation of Antibody fragments.
    After identifying polyclonal or monoclonal antibody fragments with binding activities it is usually necessary to grow a larger culture in order to make enough antibody for use as a reagent. The fragments can be purified or used directly.

    Cultures.

    • For polyclonal antibody fragments from phagemid vectors use an aliquot of pooled, scraped HB2151 bacteria infected with the selected phage from the last round of selection.
    • For monoclonal antibody fragments use individual colonies.
    1. Inoculate either 10 µl of pooled scraped bacteria or single colonies into 2xTY-AMP-0.1% GLU (1 ml to 1 litre medium, depending on the size of preparation needed).
    2. Grow to OD 0.8 and induce with 1 mM IPTG.
    3. Incubate with shaking:
       • overnight at 30°C to harvest antibody fragments secreted into culture supernatant.
       • for 3-16 hr at 30°C to harvest antibody fragments from E. coli periplasm. Sucrose in the medium may increase the proportion of antibody in the periplasmic space.

    Harvesting Fragments from the Culture Supernatant.

    1. Spin the induced bacterial culture at 10,800 g for 15 min and collect the antibody  containing supernatant.
    2. For small scale preparations (original culture volumes of less than 100 ml) the  supernatant can either be:
       • taken directly for further purification.
       • or concentrated about 10 fold by dialysis against dry granular PEG-6000 and used  without further purification.
    3. For large scale preparations the antibody containing supernatant can be purified further. However, beforehand it is recommended that the supernatant is either:
       • centrifuged very well to remove bacterial debris;
         or
       • supernatants from an original culture volume of up to 1 litre are filtered through  disposable 0.45 µM filters inserted into a 500 ml housing device (Millipore).
         or
       • supernatants from an original culture volume of over 1 litre are ultrafiltered through a 16 µm filter (e.g. Filtron Technology Corporation 'Minisette' system) and then  concentrated 5-10 fold using the same system equipped with a 10 kDa cut off filter.

    Harvesting Fragments from the Periplasm.

    1. Spin the induced bacterial culture at 10,800 g for 15 min.
    2. Resuspend in 1/20 the original volume of 30 mM Tris, pH 7.0-20% sucrose-1 mM  EDTA and leave for 20 min on ice.
    3. Spin at 10,800 g for 15 min and collect supernatant (periplasmic fraction) into a new  tube. <
    4. Resuspend pellet in 50 ml 5 mM MgSO4 and incubate for 20 min on ice.
    5. Take the supernatant and spin at 10,800 g for 15 min. Take the supernatant (osmotic  shock fraction) and add it to the periplasmic fraction.
    6. Concentrate periplasmic preparation or filter depending on the initial culture volume (as with fragments from the culture supernatant), or dialyse in suitable buffer for further purification.

    Purification of Antibody Fragments.
    Antibody fragments from the E. coli periplasm or supernatant can be purified using a range of different columns, the choice of which depends on the type of antibody fragment and the tag attached to it.. Immobilised metal affinity chromatography is incompatible with direct loading of antibodies in supernatants onto the column. The supernatant has to be exchanged into loading buffer, for example by ammonium sulfate precipitation or using an ultra-filtration device.
    Antibody fragments can be subcloned into other expression vectors which use other tags. Throughout the purification try to keep the protein at 4°C or on ice all the time and if possible perform the purification in a cold room.

    Purification using a Metal Affinity Column.
    Metal affinity columns can be used for the purification of antibodies with a hexa- Histidine tag.

    1. A) For periplasmic preparations:
       Dialyse the antibody preparation (10 kDa cut off) against loading buffer (50 mM   phosphate buffer, pH 7.5-500 mM NaCl-20 mM imidazole) overnight at 4°C.
       B) For bacterial supernatants:
       Exchange (by ammonium sulfate precipitation followed by dialysis, or using ultra- filtration techniques) the supernatant into loading buffer at 4°C.
    2. Fill a column with 5 ml Ni-NTA resin (Diagen).
    3. Equilibrate the column with 50 ml loading buffer.
    4. Load the antibody preparation onto column.
    5. Wash the column with 50 ml washing buffer (50 mM phosphate buffer, pH 7.5-500  mM NaCl-20 mM imidazole).
    6. Elute the protein with 20 ml elution buffer (50 mM phosphate buffer, pH 7.5-500 mM  NaCl-100 mM imidazole) collecting 1 ml fractions. The antibody fragments elute  between 4-10 ml.
    7. Dialyse the antibody fractions (detected by ELISA) against PBS or HPS to remove  imidazole and NaCl (typical yields are 0.2-20 mg per litre of culture). Some antibodies  tend to precipitate during dialysis. The precipitation is associated with the presence of  the His-tag. Addition of EDTA into the antibody sample in  many cases solves the problem.
    8. To regenerate the column wash with 50 ml of 50 mM phosphate buffer, pH 7.5, 500  mM NaCl-250 mM imidazole and equilibrate column with 50 ml of loading buffer.
       For small scale rapid purification Ni-NTA resin spin columns can be used (Qiagen).

    Purification using a Protein A Column.
    Protein A-Sepharose can be used to purify antibody fragments encoded by VH segments from the VH3 family (Hoogenboom and Winter 1992).

    1. Pre-swell the Protein A-Sepharosein PBS.
    2. Fill a column with 1-5 ml Protein A-Sepharose (depending on culture size and resin capacity) and equilibrate column with PBS.
    3. Load antibody preparation onto column.
    4. Wash with 5 column volumes of firstly PBS, then with PBS-0.5 M NaCl.
    5. Elute the protein with about 3 column volumes of either 0.2 M glycine, pH 3.0, or 100 mM triethylamine, collecting 1 ml fractions into 0.2 ml 1 M Tris-HCl, pH 7.4 and immediately mixing.
    6. Measure the OD at 280 nm (1 OD = 0.8 mg/ml) and dialyse sample overnight against  PBS (10 kDa cut off).
       For small scale purification, mix 5-10 ml of antibody fragments with 100-200 µl of pre-swollen Protein A-Sepharose. Purify using ultrafree-MC filters (Millipore) following the above protocol, except washing with 1 ml and eluting with 100 µl.

    Purification on an anti-FLAG M2-Sepharose column (Kodak or other suppliers)
    - Follow manufacturersí instructions
    

    4.2.8 Antibody Fragments or Phage as Reagents.

    Antibody fragments  preparations or PEG precipitated phage can be used as reagents.

    Western Blotting.
    Polyclonal or monoclonal phage antibodies can be used as reagents to detect antigens on Western blots. Here the detection of purified antigen is described, but the same protocol can be used for antigen present in complex protein mixtures.

    1. Run 0.2-2 µg of each purified antigen on a 10% or 15% polyacrylamide gel (Laemmli 1970) and then electrophoretically transfer proteins onto a nitrocellulose filter.
    2. Block the filter for 1 hr at rt in 2% MPBS-0.05% Tween 20 (for antibody fragments) or  5-10% MPBS-0.5% Tween 20 (for phage). The antibodies are generally dissolved in  2% MPBS-0.05% Tween 20. However, according to the signal and background the  percentage Marvel and Tween 20 can be varied.
    3. Add the antibody fragments (1-10 µg/ml ofpurified antibody, or 1:2 dilution of 10x  concentrated supernatant) or phage (10¹¹ t.u./ml) to the filter
    4. A) For soluble antibodies, add the antibodies needed for detection to the primary antibody  solution for a 1-step procedure. Choose the secondary antibodies as follows:
        one of the following reagents:
        a. HRP-protein A (Sigma) in a 1:1000 dilution, for antibody fragments of the human  VH3 family.
        b. Monoclonal antibody M2 (1 µg/ml) + by HRP-goat anti-mouse Fc specific   antibody  (1:1000; Sigma) to detect antibody fragments with a myc tag.
       B) For phage, wash, then detect the binding by incubating the filter with HRP-anti-M13  (Pharmacia) in a 1:1000 dilution.
    5. Wash.
    6. Peroxidase activity is detected using an ECL kit (Amersham) by mixing equal amounts of reagent 1 with reagent 2. Pour onto the blot, allow to react detect with a light  sensitive film.

    Oligos for sequencing

    The structure of the library is:
    
    

    

    For sequencing your antibodies, primers C and D are sufficient, since diversity is concentrated only in the CDR3 regions:

    fdseq1 (priming at the beginning of gIII)   <---------

    5í - GAA TTT TCT GTA TGA GG - 3í  (Tm = 45.5 C)
    

    DP47CDR2back  (priming in the VH germline gene, before the VH CDR3)  ---------->

    5í - TAC TAC GCA GAC TCC GTG AAG - 3í (Tm = 59.8)

    You may find useful also the following primers:

    PelBback  (priming on the PelB leader) ------------->

    5í - AGC CGC TGG ATT GTT ATT AC - 3í (Tm 55.3)
    

    LMB3 (priming before the PelB leader) ------------->

    5í - CAG GAA ACA GCT ATG AC - 3í