Scalable High Throughput Selection From Phage-displayed Synthetic Antibody Libraries

1. Antigen Generation

NOTE:  Antigen domains can be synthesized and cloned by a variety of commercial vendors into an appropriate IPTG-inducible expression construct.

1. Transform expression constructs into chemically competent, T1-phage resistant, BL21 E. coli cells by mixing 10 ng of encoding DNA in 20 µl of 1X KCM on ice in a 96 well PCR plate followed by the addition of 20 L of chemically competent cells.
2. Incubate the cell/DNA mixture on ice for 20 min, at RT for 10 min and then on ice again for 2 min before rescuing with 100 µl of pre-warmed super optimal broth with glucose (SOC) media, covering with a breathable plate seal and shaking at 37 °C for 1 hr at 200 rpm in a 2.5 cm orbital shaker.
3. Plate 5 µl of transformed cells on to Luria-Bertani (LB)/agar plates supplemented with carbenicillin (100 g/ml) and grow O/N to obtain single colonies used for generating glycerol stocks.
4. Inoculate 1 ml of 2YT/carb media with 5 µl glycerol stock and grow for 12 hr at 37 °C in a 96 well deep-well block with shaking at 200 rpm in a 2.5 cm orbital shaker.
5. Inoculate NZY media15 (supplemented with 0.05 % glucose, 2% lactose, and 100 g/ml carbenicillin) with a 1:40 dilution of this culture, then shake for 6-8 hr at 30 °C and 200 rpm in a 2.5 cm orbital shaker.
6. Pellet bacteria and purify antigenic proteins in high-throughput in a 96-well filter plate containing 100 µl of Ni-NTA resin as per previously published protocols16,17.  
7. Determine the concentration of purified proteins by Bradford dye-binding assay18 and characterize for size and purity by separation and visualization of 10-50 µg with Coomassie stain on an SDS-page gel (See Figure 2).
   NOTE: At this stage proteins can be further characterized by methods that assess protein aggregation¹⁹ or folding²⁰ depending on the nature of the antigen and the requirements of the selection pipeline.

2. Preparation and Titration of the Phage Library

1. Pick a single colony of T1 phage-resistant E. coli cells into 1 ml of 2YT media supplemented with 50 μg/ml tetracycline.
2. Once cell growth is visually established, dilute the culture in a larger volume of 2YT supplemented with 50 μg/ml tetracycline to ensure a sufficient volume of cells for infection (generally 45 µl per library dilution, see below.).
3. Prepare the library for use by precipitating the phage from storage buffer with 1/5th volume of autoclaved 20% PEG-8000, 2.5 M NaCl (PEG-NaCl) incubating on ice for 30 min and pelleting precipitated phage by centrifugation at 12,000 x g.
4. Discard the supernatant and resuspend precipitated phage in an appropriate volume of 1x PBS pH 7.4, 0.2% BSA and 0.05% Tween (PBT), adjusting to an appropriate phage concentration for selections (if necessary) to ensure adequate library coverage. See Step 2.9 and Discussion.
5. Dilute a 5 μl phage aliquot into 45 μl of 1x PBS pH 7.4 in a sterile multi-well plate, mix, and create a series of 10 fold dilutions in the same buffer.
6. Add 5 µl of each phage dilution to 45 µl of T1 phage-resistant E . coli cells in log phase growth (OD₆₀₀ = 0.4-0.8) in another multiwell plate, cover with a breathable seal and incubate at 37 °C with shaking at 200 rpm for 30 min in a 2.5 cm orbital shaker.
7. Plate 5 μl of infected cells from each of the wells on to a pre-warmed agar plate supplemented with 100 g/ml carbenicillin and grow O/N at 37 °C until colonies are visible.
8. Calculate the total phage/ml as follows:
   Phage/ml = number of colonies on a carbenicillin plate in the most dilute sample x 200 x 10ⁱ (where i = the maximum number of dilutions in which colonies are apparent).
9. To determine the number of coated antigen wells to ensure coverage of the library diversity, calculate as follows:
   Number of coated antigen wells required =
   Desired fold coverage * library diversity
   # of phage per 100 µl

3. Antibody Selection from Phage-displayed Libraries

3.1) Antigen Immobilization

1. To avoid freeze-thaw cycles that can affect antigen quality and to facilitate dilutions, prepare a stock antigen protein plate normalized to 100x working concentrations (e.g., 500 g/ml) and aliquot in to non-binding 96-well plates.  
2. Prepare antigen-coated plates from stock protein solutions one day prior to each round of selections by diluting from stock plates with PBS.
3. Coat 96-well high protein-binding polystyrene plates with 100 μl of GST-tagged antigen protein or negative control protein (GST, BSA, neutravidin, etc.) at 5 μg/ml in PBS. Shake at 250 rpm 4 °C O/N on a microplate shaker.
4. Remove protein solution from the coated microplates, block the antigen-coated and negative selection plates with 200 μl blocking buffer (0.2% BSA in 1x PBS pH 7.4) and shake at 250 rpm at RT for 1-2 hr.
5. After blocking, wash the blocked protein-coated plates four times with 200 µl of 1x PBS pH 7.4 with 0.05% Tween (PT) buffer either manually or using an automated microplate washer.

3.2) Library Pre-clearance and Antigen-phage Incubation

1. Deplete the library of non- (or tag-) specific binding phage-displayed antibody fragment clones via transfer of 100 µl (10¹² phage) of prepared phage library in PBT buffer to a number of wells coated with negative control protein or free affinity tag protein (e.g., GST) equivalent to the number of targets and incubate phage in wells for 1-2 hr at RT with shaking at 250 rpm.
   NOTE: As an alternate or additional means of negative selection, incubation can also be performed in the presence of excess (5-25 μM) soluble protein affinity tag (e.g., GST) to further remove tag-binding clones and enhance recovery of specific target binding clones.
2. Transfer phage supernatant from negative selection to antigen-coated plates and incubate for 1–2 hr at RT with shaking at 250 rpm for capture of specific binding clones.
3. Remove unbound phage and wash wells of the microplate 8-15 times with PT buffer either manually or using a microplate washer. Remove excess wash buffer just prior to elution.

3.3) Elution, Infection and Phage Amplification

1. Early on the day of selections, pick a single colony of T1 phage-resistant E. coli cells into 1 ml of 2YT media supplemented with 50 μg/ml tetracycline and grow at 37 °C with shaking at 200 rpm in a 2.5 cm orbital shaker or equivalent.
2. Once cell growth is established and can be seen by eye, increase the volume of media with 2YT supplemented with 50 ug/ml tetracycline to ensure a sufficient volume of cells for infection (generally 100 µl per selection well).
3. Grow the cells until an OD₆₀₀ of 0.4–0.8 is reached (approximately 5-7 hr), then to elute bound phage add 100 μl of cells to the washed antigen plate and shake at 37 °C for 30 min.
4. Add M13K07 helper phage to a final concentration of 1 x 10¹⁰ cfu/ml and incubate at 37 °C for 45 min with shaking at 200 rpm.
5. Transfer cells to 1.2 ml of 2YT supplemented with 150 μg/ml carbenicillin and 75 μg/ml kanamycin in a 96-well deep well block with V-bottom and grow O/N at 37 °C with shaking at 200 rpm in a 2.5 cm orbital shaker.

3.4) Phage Preparation and Repeated Rounds of Selection

1. Pellet bacteria by centrifugation at 4,000 x g for 15 min at 4 °C.
2. Mix equal volumes of phage supernatants from replicate plates in a mini-tube 96-well sterile microplate blue box, neutralize with 1/10^th volume of 10X PBT and mix. Repeat rounds of selection (starting from  Step 3.1.1) for 3-4 rounds or until enrichment is observed (See Sections 4. and 7.) compared to negative control protein.
   NOTE: In the early rounds of selection when significant / detectable enrichment is not expected (i.e., rounds 1 and 2), quantification of input and output phage titres (described in Section 7 and representative results shown in Figure 3) may be helpful in ensuring minimum phage concentrations for successful selections (described in Discussion) and are more appropriate than pooled ELISAs.

4. Characterization by Pooled ELISA

1. Coat 96-well high protein-binding polystyrene plates with 100 μl of GST-tagged antigen protein or negative control protein (GST, BSA, neutravidin etc.) at 2 μg/ml as per Section 3.1 and shake at 4 °C O/N.
2. Remove excess antigen, block the plate with 200 μl of blocking buffer with shaking for 1–2 hr at RT at 250 rpm on a microplate shaker and wash four times with 200 μl PT buffer either by hand or by automated plate washer.
3. Apply 100 μl of phage supernatant (diluted 1:10-1:20 in PBT buffer), shake at 250 rpm for 15 min at RT, and wash the plate eight times with 200 μl PT buffer.
4. Add 100 μl of anti-M13-HRP (1:5,000 in PBT buffer). Shake at 200 rpm for 30 min at RT, then wash the plate six times with 200 μl PT buffer and twice with 200 μl of PBS.
5. Apply 100 μl of 1:1 TMB substrate and develop for 5–15 min (or until substantial color has developed), then stop the reaction by addition of 100 μl of 1 M H₃PO₄ and read the plate at 450 nm.
6. In general, enrichment in phage pools can be observed in rounds 3–4 as a ratio of specific binding versus non-specific binding of >2 (See Figure 4)
   NOTE: It is informative to note that high absolute binding signal on the affinity tag protein indicates enrichment of tag-specific binders (rather than target binders), and that high absolute binding signal on the negative control protein indicates enrichment of non-specific or “sticky” binders.

5. Clone Selection, Sequencing and Characterization

5.1) Isolation of Single Fab-phage Clones

1. To isolate individual clones for sequencing and characterization, infect 1/10^th volume of the amplified phage pool in to T1 phage-resistant E. coli cells in log phase of growth (OD₆₀₀ = 0.4-0.8) in a round-bottom microtiter plate, cover with a breathable plate seal and incubate with shaking at 200 rpm for 30 min at 37 °C.  
2. Prepare 10-fold serial dilutions in 2YT media and plate out on separate agar plates supplemented with 100 g/ml carbenicillin.
3. Pick single colonies in to 450 µl of 2YT/carb media supplemented with 1 x 10⁹ M13K07 phage/ml and grow O/N in a mini-tube 96 well sterile microplate blue box incubating at 37 °C with shaking at 200 rpm.
4. Pellet bacteria by spinning at 4,000 x g for 10 min at 4 °C.
5. Clonal phage supernatant can be used for sequencing of Fab-phage clones (as described in Section 5.4) or for clonal direct-binding ELISAs to determine specificity against immobilized antigens (as described in Sections 5.3 and ²¹) for which representative results are shown in Figure 5.

5.2) Expression of Fab clones for characterization

1. Following cloning in to an appropriate expression vector (see Discussion), pick single E. coli colonies transformed with expression constructs in to 1 ml of 2YT/carb media in a 96 well deep-well plate using a sterile toothpick or pipette tip and grow for 12-16 hr at 37 °C with shaking at 200 rpm. This can be done either manually or with an automated colony picker.
2. Transfer 40 µl of growing culture to 1.5 ml of NZY media supplemented with glucose/lactose/glycerol in a 96 well deepwell block as per the methodology of Studier et al.¹⁵
   NOTE: Alternately, cells can be grown for 3-4 hr (OD 0.8-1.0) in non-supplemented and protein expression induced by the addition of 1 mM IPTG.
3. Cover the culture block(s) with a breathable seal and express Fab clones for 6-8 hr at 30 °C with shaking at 200 rpm. Spin plates at 4,000 x g for 15 min to pellet bacterial cells, remove the supernatant, cover plates with a foil seal and freeze pellets at -20 °C until lysis.
4. Liberate expressed Fabs from collected pellets with 100 µl of lysis buffer (Tris HCl 50 mM, 200 mM NaCl, 1.25% Triton X-100 pH 8.0 with 1mg/ml lysozyme and 10 U benzonase/ml culture and protease inhibitors) and shaking for 2 hr at 4 °C.
5. Remove cell debris by centrifugation at 4,000 x g for 15 min at 4 °C and collect crude lysate supernatant for use in initial characterization of specificity by ELISA plates (see Section 5.3).

5.3) Clone Characterization by Direct Binding Clonal Fab ELISA

1. Immobilize antigens as described in Section 3.1 instead aliquotting 30 µl of 2 µg/ml antigen in PBS to wells of a 384 well plate. Quad-based antigen layouts in the plate can facilitate reagent transfer when using 96 channel-based dispensing heads. Wash and block ELISA plates as per Section 3.1.
2. Cleared lysates collected from Section 5.2 can be applied directly to immobilized antigen for evaluation of binding.  Incubate 30 µl of lysate per well for 15 min at RT with shaking at 200 rpm.
   NOTE: Purified Fab clone protein can alternately be used by diluting to 10 µg/ml in PBT buffer and applying 30 µl of each clone to blocked wells containing antigen or negative control protein (GST, BSA etc.) and developing in the same manner described below. Alternately, Fab-phage can also be used by diluting 10 µl of phage supernatant (described in Section 5.1) in 20 µl PBT buffer (1:3) and detecting with anti-M13-HRP antibodies (described in Steps 4.4 – 4.5)
3. Remove excess lysate and wash wells either manually or using an automated plate washed 8x with 100 µl of PT buffer and remove excess buffer.
4. Incubate 30 µl of a 1:5,000 dilution of secondary anti-FLAG antibody in PBT buffer for 30 min at RT with shaking at 200 rpm.
5. Wash, develop and quantify as per Sections 4.3 – 4.5.
   NOTE: Representative results illustrating both specific and non-specific binding clones are shown in Figure 5.

5.4) Clone Sequencing

1. Prepare a PCR master mix as indicated in Table 2.
2. Add 2 µl of the appropriate PCR template (either phage supernatant or re-suspended single phagemid-containing bacterial colony) to 20 µl of PCR master mix prepared with the appropriate primers in wells of a PCR plate.
   NOTE: Separate master mixes are required for sequencing of both the heavy and light chain genes.
3. Run the PCR reaction using the parameters listed in Table 2 – PCR Settings.
4. Verify success of the PCR reaction by mixing 5 µl of the completed reaction with loading dye on a 1% agarose gel supplemented with DNA visualization stain and imaging on a 300 nm transilluminator.
5. Add 2 µl of PCR product to 10 µl of sterile water containing 0.2 µl each of exonuclease and shrimp alkaline phosphatase and incubate at 37 °C for 20 min followed by enzyme inactivation at 80 °C for 10 min to remove excess primers in preparation for sequencing.   

6. Scaling for Automated Selections

6.1) Pipette-based Liquid Handling

1. Prepare a 1% (w/v) bromophenol blue in water and remove insoluble particles using a 0.45 ml filter.
2. To determine the linear range of absorbance on the plate reader, prepare a 1:1,000 dilution of the 1% bromophenol blue solution in water, and continue with two-fold serial dilutions (16 dilutions total).
3. Transfer 100 µl of each dilution to a flat bottom 96 well plate and read the absorbance at 590 nm. Plot absolute absorbance versus dilution factor and choose a dilution at the mid to high end of the linear range for subsequent tests.
4. Add bromophenol blue to water or to the actual solution that will be pipetted based on the dilution chosen in the previous step in sufficient volume for pipetting to three 96 well plates, plus the dead volume in the reservoir.
5. Calibrate a single-channel pipettor to deliver the test volume by pipetting water onto an analytical balance (100 µl of water weighs 100 mg at RT.)
6. Using the calibrated pipettor, transfer the test volume of the dyed solution to at least three wells of a flat bottom 96 well plate and read their absorbances at 590 nm, in triplicate.
7. Weigh three empty flat-bottom 96 well plates on an analytical balance and record their masses.
8. Pipette test volume from reservoir to each of three 96 well plates and measure their absorbances at 590 nm, in triplicate.
9. Weigh each plate and calculate the difference in mass.
10. First, evaluate whether the absorbances in each well are uniform within the range of tolerance (e.g., +/- 5%) and whether they are consistent with the absorbances obtained by hand-pipetting with the calibrated pipetman. If particular channels are consistently over- or under-delivering (for example see Figure 6), follow repair procedures and repeat the evaluation procedure until all channels deliver uniform volumes.
11. Second, once all channels are confirmed to deliver uniform volumes, check the accuracy of that volume by calculating the average mass difference, per well. For instance, a perfectly calibrated liquid handler set for 100 µl will deliver 9.60 g of water per plate, or 0.10 g of water per well. If the real volume delivered is consistently over or under the intended volume, then a correction factor can be applied, i.e., programming 110 µl in order to actually deliver 100 µl.
    NOTE: For small volumes, e.g., 10 µl, use ten-fold more bromophenol blue in the dyed solution, and pre-aliquot 90 µl of water to the 96 well plates by hand, to ensure that the absorbances are measureable and fall in the linear range.

6.2) Automated Plate Washing

1. Immobilize positive target and negative control proteins in separate wells of 96 well microplates (two plates for each condition to be tested) as described in Section 3.1.
2. Block non-specific binding sites by incubation with blocking solution for one hr at RT.
3. Add identical 100 µl aliquots of a 10¹³ cfu/ml target-binding Fab-phage solution in PBT to all wells in all plates and incubate with gentle agitation at RT for 2 hr.
4. Wash two plates according to each condition, one plate for infection and titration and one plate for ELISA.
5. Evaluate the efficacy of the wash by direct infection into bacteria and titration (as described in Section 3.3 (without M13K07 addition) and 7.2.1) or development with anti-M13 HRP antibody and colorimetric substrate (as described in Section 4).
6. Phage titres from specific clones will often yield in the range of 10⁵–10⁷ phage/ml from an immobilized protein against which the clone binds specifically. Some residual binding to negative control protein (e.g., BSA) is to be expected and generally 10²-10⁵ phage/ml are observed, however very low titres (i.e., <10¹) can signal overly stringent washing, which can compromise a selection.
7. For phage binding determined by ELISA, compare the absolute binding signals for positive and negative control proteins to determine if robotic washing is equivalent to established hand washing results (Figure 7).

6.3) Plate Washer Decontamination

1. To start, run the decontamination protocol to be tested.
   NOTE: We recommend using at least double the total line volume, and prime and soak the washer head for 5 min in 0.5% sodium hypochlorite, freshly diluted from commercial bleach (typically 6% sodium hypochlorite).
2. Prime and soak the washer head for 5 min in sterile (autoclaved) water and finally, prime the system until the lines are full of air.
3. Prime the lines with sterile water or wash buffer.
4. Wash a sterile 96 well microplate once, remove from washer and seal with sterile plate seal. This is the pre-contamination control plate.
5. Aliquot 100 µl of O/N amplified phage culture supernatant to all wells of a 96 well plate.
6. Wash the phage-containing plate once, then remove from the washer and seal with a plastic or foil plate seal. This is the contamination control plate.
7. Carry out the decontamination protocol being tested. In this example, repeat steps 6.3.1 and 6.3.2.
8. Wash sterile microplate once, then remove from washer and seal. This is the decontamination test plate.
9. Plate 50 µl of log phase E. coli to titration plates; this is the pre-infection control plate.
10. Unseal pre-contamination, contamination and decontamination 96-well plates and add 100 µl of E. coli in the log-phase growth stage to all wells, using new tips for each well.
11. Seal and incubate at 37 °C for 30 min at 200 rpm.
12. Plate 50 µl of infected cells from three random wells of each 96-well plate to 10 cm 2YT/carb plates and incubate at 37 °C O/N.
13. Transfer 50 µl of infected cells from all wells of each plate to deep well plates containing 1 ml 2YT plus 100 g/ml carbenicillin per well and grow O/N at 37 °C.
14. Repeat the decontamination protocol and leave the washer lines dry.
    NOTE: No O/N growth on plates or in liquid medium should be observed on the pre-infection control, pre-contamination control or decontamination test plates. Many colonies and growth should be observed on the contamination control plate; if not, no conclusions about the efficacy of the decontamination protocol can be made. Ideally, the decontamination plates will yield no colonies and no O/N growth in any of the wells in the O/N plates. The stringency of the decontamination protocol can be increased with higher bleach concentrations, longer soak times and additional cycles.

7. Input/Output Titrations

7.1) Input Phage Titrations

1. Dilute a 5 µl aliquot of phage to 50 µl with sterile-filtered PBS and mix.
2. Prepare serial 10-fold dilution in sterile-filtered PBS to 10¹⁰ using multi-channel pipette or a 96 channel pipettor.
3. Inoculate 5 µl of diluted phage in to 45 µl of T1 phage-resistant E. coli cells in log-phase of growth (OD 0.6-0.8) and incubate covered with a breathable seal for 30 min at 37 °C.
4. Plate 5 µl of infected cells on to an LB/agar plate supplemented with 100 µg/ml carbenicillin and incubate O/N at 37 °C.
5. The phage concentration can be estimated from the most dilute sample in which colonies appear according to the calculations detailed in Section 2.8.

7.2) Output Phage Pitrations

1. Following elution of plate-bound phage by the addition of E. coli cells, dilute a 5 µl aliquot of phage-infected cells to 50 µl with autoclaved 2YT media
2. Prepare serial 10-fold dilution in autoclaved 2YT media to 10⁶ using a multi-channel pipette or 96 channel pipettor.
3. Plate 5 µl of infected cells on to an agar plate supplemented with 100 µg/ml carbenicillin and incubate O/N at 37 °C.
4. The phage concentration can be estimated from the most dilute sample in which colonies appear according to the calculations detailed in Section 2.8.
   NOTE: In both cases (i.e., input and output phage titres, comparison with colony numbers on both tetracycline (total cell number) and kanamycin (helper phage titre) can also be informative.