Journal of Microbiological Methods 120 (2016) 6–14

Contents lists available at ScienceDirect

Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth

Application of streptavidin mass spectrometric immunoassay tips for immunoaffinity based antibody phage display panning Chai Fung Chin a, Lian Wee Ler b, Yee Siew Choong a, Eugene Boon Beng Ong a, Asma Ismail a, Gee Jun Tye a, Theam Soon Lim a,⁎ a b

Institute for Research in Molecular Medicine, Universiti Sains Malaysia, 11800 Minden, Penang, Malaysia Fisher Scientific (M) Sdn Bhd, No. 3, Jalan Sepadu 25/123, Taman Perindustrian Axis, Seksyen 25, 40400 Shah Alam, Selangor Darul Ehsan, Malaysia

a r t i c l e

i n f o

Article history: Received 18 August 2015 Received in revised form 9 November 2015 Accepted 9 November 2015 Available online 12 November 2015 Keywords: Domain antibodies Disposable Automation Research Tips (D.A.R.T's®) Mass spectrometry immunoassay (MSIA™) Semi-automated panning Phage display

a b s t r a c t Antibody phage display panning involves the enrichment of antibodies against specific targets by affinity. In recent years, several new methods for panning have been introduced to accommodate the growing application of antibody phage display. The present work is concerned with the application of streptavidin mass spectrometry immunoassay (MSIA™) Disposable Automation Research Tips (D.A.R.T's®) for antibody phage display. The system was initially designed to isolate antigens by affinity selection for mass spectrometry analysis. The streptavidin MSIA™ D.A.R.T's® system allows for easy attachment of biotinylated target antigens on the solid surface for presentation to the phage library. As proof-of-concept, a domain antibody library was passed through the tips attached with the Hemolysin E antigen. After binding and washing, the bound phages were eluted via standard acid dissociation and the phages were rescued for subsequent panning rounds. Polyclonal enrichment was observed for three rounds of panning with five monoclonal domain antibodies identified. The proposed method allows for a convenient, rapid and semi-automated alternative to conventional antibody panning strategies. © 2015 Elsevier B.V. All rights reserved.

1. Introduction The mass spectrometry immunoassay (MSIA™) system is based on the principles of affinity separation of antibodies and antigen for mass spectrophotometry (MS) analysis (Nelson et al., 1995; Parker et al., 2010). The efficiency of this technique was first demonstrated with the capture and concentration of antigen to overcome signal repression of analytes for MS. MSIA™ has been used for fast, selective and quantitative screening of human blood for the presence of myotoxin a, and Mojave toxin from the venoms of Crotalus viridis viridis, and Crotalus scutulatus scutulatus (Nelson et al., 1995). Antibodies are bound to the modified matrix of the distal tips to facilitate selective protein capture to enrich and purify the target protein for subsequent MS analysis (Niederkofler et al., 2001; Niederkofler et al., 2008; Lopez et al., 2010). MSIA™ Streptavidin D.A.R.T's® (Disposable Automation Research Tips) are MSIA™ tips that contain covalently linked streptavidin to the porous monolithic solid support as a capture molecule. The streptavidin molecules introduced to the D.A.R.T's® MSIA™ platform functions for

Abbreviations: dAb, domain antibody; D.A.R.T's, Disposable Automation Research Tips; HlyE, Hemolysin E; MS, mass spectrophotometry; MSIA, mass spectrometry immunoassay; scFv, single-chain variable fragment; Ub, ubiquitin. ⁎ Corresponding author. E-mail address: [email protected] (T.S. Lim).

http://dx.doi.org/10.1016/j.mimet.2015.11.007 0167-7012/© 2015 Elsevier B.V. All rights reserved.

easy capture of biotinylated targets due the strong streptavidin affinity to biotin. MSIA™ Streptavidin D.A.R.T's® takes advantage of the high affinity between the biotin and streptavidin tetramer for rapid, specific and strong capture of target molecules (Diamandis and Christopoulos, 1991). The conventional application of MSIA™ Streptavidin D.A.R.T's® platform which employs biotin-conjugated antibody to capture and enrich endogenous or clinically relevant antigens is done for rapid detection and analysis by MS. The MSIA™ tips serve as a convenient solid phase for target presentation for accessible target-ligand interaction (Niederkofler et al., 2008; Parker et al., 2010). As the design is based on the pipette tip, the physical motion of liquid movement through the tips allows a higher interaction of molecules to the solid phase to be achieved. The pipetting motion allows for all conventional stages of work to take place including incubation, washing and elution. The packing material of the tips also allows large amounts of protein to be captured for presentation. Subsequently, the MSIA™ tips are rinsed to remove any unbound antibodies and non-specific proteins while bound antibodies are eluted using suitable elution buffer. This allows for simultaneous target protein concentration and purification for downstream applications (Kiernan et al., 2002; Kiernan et al., 2003). Phage display technology allows the physical display of peptides or proteins on the surface of bacteriophage (Smith, 1985). Antibody phage display panning involves the attachment of target proteins to solid phase for binding with specific antibodies displayed by

C.F. Chin et al. / Journal of Microbiological Methods 120 (2016) 6–14

bacteriophages (Hairul Bahara et al., 2013; Hammers and Stanley, 2014). This is commonly done using high protein binding microtiter plates (Krebs et al., 2001), immunotubes (De Kruif et al., 1995), magnetic particles (Walter et al., 2001) and even an affinity column (Noppe et al., 2009) to physically present target antigens for binding with antibody presenting phages. The steps involved in conventional panning are similar to the processes involved in the MSIA™ protocol with an initial incubation step followed by incubation, washing and final elution (Coomber, 2002). The basic principle of antibody phage display panning is similar to that of MSIA™ with selection and concentration of antibodies by affinity pressure. Here, we propose a new method for antibody phage display panning using the MSIA™ technology. This MSIA™ protocol was modified and customized to allow for antibody phage display panning based on conventional panning methods (Hammers and Stanley, 2014). The initial target binding is carried out to physically attach the target antigen on the MSIA™ tips. The incubation step allows for the antibody phage library to bind to the target molecule. The wash step will remove any unbound phages followed by the final elution. The elution step of MSIA™ will allow for phage recovery from each round of panning. Bound phages are subsequently amplified and used for subsequent panning rounds in order to obtain clonal enrichment. A proof-of-concept experiment using synthetic single-domain antibody (dAb) phage library for panning to generate monoclonal antibodies against the target antigen is performed. The proposed method provides an attractive alternative for rapid recombinant antibody development.

7

resuspended in lysis buffer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, pH 8.0) with 20 mg/ml lysozyme. The cells were then chilled on ice for 30 min and the lysate was sonicated for 3 min continuously on ice. Subsequently, cell lysate was centrifuged and the supernatant containing the soluble proteins was collected. IMAC purification was performed for both recombinant antigens using the 1 ml Ni-NTA Agarose fast flow column (GE healthcare, Uppsala, Sweden) according to the manufacturer's instructions. Purified fractions were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis. 2.3. Preparation of antibody phage library An in-house synthetic domain antibody phagemid library stock (Hairul Bahara et al., 2015) with library size of 109 in XL1-Blue was inoculated into 500 ml 2YT broth containing ampicillin and 2% glucose with initial OD600nm around 0.1–0.2. The cells were further grown to OD600nm = 0.5 and co-infected with M13KO7 helper phage at 37 °C, static for 30 min. Subsequently, the culture was centrifuged and resuspended with 500 ml 2YT containing ampicilin, 60μg/ml kanamycin and 0.1% glucose. The library phage was propagated at 30 °C, 180 rpm for overnight. The culture was then centrifuged and the phagecontaining supernatants were collected and precipitated by incubation with 20% (wt/vol) polyethylene glycol 6000 for 1 h at 4 °C. The supernatant was removed by centrifugation and the phage pellet was suspended in 1 ml phosphate buffer saline (PBS, pH 7.4) and stored at 4 °C.

2. Materials and methods 2.4. MSIA™ streptavidin D.A.R.T's® loading of biotinylated antigen 2.1. Materials Chloramphenicol and isopropyl β-D-1-thiogalactopyranoside (IPTG) were purchased from Calbiochem (San Diego, California). Glucose and polyethylene glycol 6000 were obtained from R&M chemicals (Essex, UK). Bovine serum albumin (BSA) was purchased from Nacalai Tesque (Kyoto, Japan). Ampicillin was obtained from Fisher Scientific (Pittsburgh, Pa.). Kanamycin and 2,2′-Azinobis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) were purchased from Amresco (OH, USA). 2YT was obtained from Merck (Darmstadt, Germany). Escherichia coli strain BL21 (DE3) competent cell was purchased from Stratagen (California, USA). TG1 and XL1Blue competent cells were obtained from Agilent Technologies (Santa Clara, CA). M13KO7 helper phage was purchased from NEB (MA, USA). Helper plasmid pRARE3 encoding the biotin ligase and rare tRNAs together with the pRSET-BH6 plasmid were obtained from Dr. Zoltan Konthur, Max Planck Institute for Molecular Genetics (Berlin, Germany). Anti-M13 horseradish peroxidise (HRP)-conjugated monoclonal antibody was purchased from GE healthcare (NJ, USA). MSIA™ Streptavidin D.A.R.T's® and Finnpipette™ Novus i Electronic 12-channel Pipette were obtained from Thermo scientific (USA). QIAprep Spin Miniprep Kit was purchased from Qiagen (CA, USA). Costar flat bottom high protein binding microtiter plate was purchased from Corning (NY, USA) and BluElf prestained protein ladder was obtained from GeneDireX. Horseradish peroxidase-anti-c-Myc antibody was purchased from Abcam (MA, USA). 2.2. Expression of biotinylated antigens For the production of biotinylated antigens, the E. coli strain BL21 (DE3) was transformed with the vector pRSET-BH6 and the helper plasmid, pRARE. The hemolysin E and ubiquitin genes in pRSET-BH6 allows for N-terminal fusion of the avi-tag. The transformed cells were grown at 37 °C to OD600nm = 0.6 in 2YT broth containing 100 μg/ml ampicillin, 17 μg/ml chloramphenicol and 0.1% glucose. The T7 promoter was induced by 1 mM IPTG and further expressed overnight (o/n) at 25 °C with 160 rpm agitation. At the end of protein induction, cells were

MSIA™ Streptavidin D.A.R.T's® were mounted to a Finnpipette™ Novus i Electronic 12-channel Pipette for antigen binding. Biotinylated recombinant antigen at 100 μg in bicarbonate buffer (0.1 M NaHCO3, pH 8.6) was loaded in MSIA™ Streptavidin D.A.R.T's® by continuous aspiration and dispensing. The electronic pipette program was set for 999 cycles with a moderate speed (Speed Setting 5). Then, the MSIA™ Streptavidin D.A.R.T's® were washed twice (20 cycles, Speed Setting 8) with PBS-T (PBS in 0.5% Tween 20) and finally (20 cycles, Speed Setting 8) with PBS. The antigen captured MSIA™ tip is ready for use. 2.5. MSIA™ streptavidin D.A.R.T's® antibody phage library panning The recombinant Ubiquitin (rUb) antigen coupled tip was first used to enrich monoclonal antibodies by affinity selection before it was used for panning and recombinant Hemolysin E (rHlyE) and MSIA™ Streptavidin D.A.R.T's® were also passed through non-specific antirUb monoclonal phage for background system control. Then, in the panning process, the antigen-coupled tip, i,e. rHlyE coupled tip was blocked with 3% skimmed milk in PBS-T (PTM) (500 cycles, speed setting 5). A total of 1012 phage particles of the antibody library was pre-incubated with PTM before used for panning. Antibody phage capture was done by performing repetitive pipetting with a fixed volume of 150 μl, with 999 cycles repeat and a speed setting of 5. Then, the MSIA™ Streptavidin D.A.R.T's® were rinsed 5 rounds with PBS-T and 5 rounds with PBS. Each wash cycle constitutes 20 cycles of aspirating and dispensing with a faster flow (Speed Setting 8). The bound phages were eluted by using 100 μl of 0.2 M glycine-HCl, pH 2.2 with 300 cycles of aspiration and dispensing with a slow speed (Speed Setting 3). The eluted fraction is then immediately neutralized with 1 M Tris HCl, pH 9.1. The eluted phages were used to infect an exponentially growing TG1 culture (OD600 of 0.5) for phage rescue and phage titer was also done by serial dilution. The culture was spun down and resuspended with 20 ml 2YT containing ampicilin. The cells were further grown for 3.5 h and infected with helper phage, M13KO7 for 30 min at 37 °C. Packaging and amplification of successive enriched phage was done by culturing the cells (30 °C, 180 rpm) o/n with ampicillin and kanamycin. The culture was then

8

C.F. Chin et al. / Journal of Microbiological Methods 120 (2016) 6–14

centrifuged and the phage-containing supernatants were collected and precipitated by incubation with 20% (wt/vol) polyethylene glycol 6000 for 1 h at 4 °C. The supernatant was removed by centrifugation at 9000×g and the phage pellet was resuspended in 300 μl PBS and stored at 4 °C for further panning rounds or analysis. Fig. 1 illustrates the antibody phage library panning.

control well as a negative control. The bound phages were detected by anti-M13 horseradish peroxidise (HRP)-conjugated monoclonal antibody (1:5000). Lastly, 150 ul of ABTS developing solution was used to detect bound phage by reacting with the peroxidase enzyme and subsequently, absorbance reading at 405 nm (OD405) was taken with a microplate reader (Thermo Scientific Multiskan Spectrum).

2.6. Microtiter plate based panning of antibody phage library

2.8. Monoclonal phage ELISA

A conventional plate panning was performed with slight modifications (Frenzel et al., 2014). 10 μg of rHlyE was coated to the surface of each well of the high protein absorption flat bottom Costar microtiter plate for the first panning round followed by 1 μg for subsequent panning rounds. After overnight (o/n) incubation at 4 °C, the wells were washed three times with PBS-T and blocked with 300 μlof blocking buffer (5% skimmed milk in PBS-T) for 2 h with gentle agitation. This was continued with the standard panning process. Packaging and amplification of successive rounds of enriched phage was done for 3 rounds and the phage-containing supernatants were collected.

Putative binding phage clones obtained from enriched polyclonal rounds were randomly picked and propagated. Binding activity was confirmed by monoclonal phage ELISA according to previous report with slight modifications (Casey et al., 2004). The positive control was anti-enhanced green fluorescent protein (eGFP) scFv against recombinant eGFP whereas the negative control was M13K07 helper phage. Briefly, two 96-well microtiter plates were coated with 10 μg of recombinant antigen in bicarbonate buffer o/n at 4 °C. The plates were washed thrice with 300 μl PBS-T and blocked with 2% BSA for 1 h at RT. Washing with 300 μl PBS-T for 3 times was done before the addition of 50ul of monoclonal phage. Subsequently, the plates were incubated for 1 h at RT followed by 3 times washing with PBS-T. 100 μl of HRP conjugated anti-M13 diluted to 1:5000 in blocking agent was added to each well and further incubated for 1 h at RT with shaking at 700 rpm. The plates were washed 3 times with PBS-T. Bound-phage carrying the peroxidase enzyme catalyzed the conversion of ABTS to a color product and absorbance readout at 405 nm was measured.

2.7. Phage polyclonal ELISA The polyclonal enrichment from the panning rounds were evaluated by ELISA. Briefly, 10 μg of recombinant antigen in bicarbonate buffer was coated on microtiter plate with o/n incubation at 4 °C. The next day, the plates were washed 3 times with PBS-T and tap dried. The microtiter wells were then blocked with blocking buffer (2% BSA in PBS-T). 10 μg of streptavidin was coated on the plate as control to examine crossreactivity between phage and streptavidin. After washing three times with 300 μl of PBS-T, the wells were blocked with 2% BSA for 1 h at room temperature (RT) (700 rpm) to reduce non-specific binding. After washing thrice with PBS-T, a total of 109 amplified phage particles from each round of panning was incubated for 1 h and washed 3 times with 300 μl of PBS-T. A total of 109 CFU M13K07 was added to the

2.9. DNA sequencing Mimiprep of phage clones that showed positive binding activity was prepared using QIAprep Spin Miniprep Kit and sequenced (1st Base, Malaysia). The inserts were sequenced using the LMB3 Fw sequencing primer (CAGGAAACAGCTATGAC) and pIII Rv sequencing primer (GTTAGCGTAACGATCTAA).

Fig. 1. Illustration of MSIA™ streptavidin D.A.R.T's® applied for antibody phage display panning.

C.F. Chin et al. / Journal of Microbiological Methods 120 (2016) 6–14

2.10. Prepararion of soluble dAb formats and soluble ELISA detection The monoclonal antibodies were cultured in a 100 ml culture with an inoculum ratio of 1:100. The culture was grown to OD600nm = 0.6, and 1 mM IPTG was added to induce o/n expression at 25 °C with 160 rpm agitation. Cells were harvested and resuspended in l ml cold 1X TES buffer and hypotonic shock was induced by adding 1.5 ml 1:5 dilution of 1× TES buffer to release proteins from cells. The cells were then chilled on ice for 1 h, centrifuged and the supernatant containing the soluble proteins were collected. Specificity of the soluble dAbs were assessed by soluble ELISA. Briefly, 10 μg/well of rHlyE was coated on microtiter plate o/n at 4 °C in PBS buffer and 3% PTM as background control. After washing three times with PBS-T, the wells were blocked with 3% PTM (1 h, 700 rpm) to reduce non-specific binding. The plate was washed thrice with PBS-T before the addition of soluble dAbs and were incubated for 1 h. Bound proteins were then detected by anti-cMyc-HRP antibody (1:2500 in PTM). Lastly, ABTS developing solution was used to detect bound antibody carrying the peroxidase enzyme that covert the substrate into color product. Absorbance reading (OD405nm) was recorded using a microplate reader. 3. Results 3.1. Expression and purification for biotinylated antigens Biotinylated antigens, i.e. rHlyE and rUb were successfully expressed and purified using Ni-NTA protein purification with molecular mass of approximately 34 kDa and 16 kDa respectively as shown in Fig. 2(a) and (b). The purified antigens were used for subsequent experiments. 3.2. Initial assessment of MSIA™ streptavidin D.A.R.T's® panning In order to examine the feasibility of the proposed panning method, loading capacity of MSIA™ tips was first taken into consideration. The evaluation of the loading capacity of the MSIA™ tips was carried out by determining the amount of protein captured from a fixed amount of antigen loaded by Bradford assay. The analysis shows a binding capacity of 0.157 mg (R2 = 0.985) for biotinylated rHlyE antigen onto the MSIA™ tips. Then, the initial panning assessment of the in-house anti-rUb single-chain variable fragment (scFv) against rUb target was done by phage titer (Table 1) and ELISA (Fig. 3) -. Table 1 indicates 1.1 × 102 CFU of rescued phage binding against rUb whereas none of the phage were non-specically bound to rHlyE and negative MSIA™ tip. Fig. 3 shows the anti-rUb scFv phage was able to bind with rUb (OD405nm = 2.56) with low background and exhibit low non–specific binding with streptavidin (OD405nm = 0.1). A control using M13KO7 helper phage against rUb also shows low binding (OD405nm = 0.09). This indicates that the panning protocol capable to capture, elute phage bound to target antigen and ready for antibody phage library panning.

9

Table 1 Titer of anti-rUb scFv phage against rUb, rHlyE and negative MSIA™ streptavidin tips.

rUb rHlyE Negative

Input (CFU)

Output (CFU)

6.0 × 1010 6.0 × 1010 6.0 × 1010

1.1 × 102 0 0

3.3. Phage dAb library panning against rHlyE The panning experiment with the synthetic dAb library against rHlyE target showed an enrichment of specific antibodies in the library pool to rHlyE. Fig. 4a illustrates polyclonal ELISA result of rHlyE panning with OD405 nm readouts at 0.21, 0.14 and 0.62 for the first to third round of panning. Result shows a slight decrease from round 1 to 2 but start to increase from round 2 to round 3. This is common during the panning process as initial panning round constituted from a pool of specific and non-specific binders whereby enrichment of specific binders against target commence after the second round. The microtiter based panning was also performed as a comparison to MSIA™ Streptavidin D.A.R.T's® panning. The polyclonal phage ELISA result of conventional panning (Fig. 4b) shows a slight enrichment from the first to third round, i.e. OD405 nm readouts of 0.23, 0.28, 0.29. The polyclonal ELISAs from both panning methods were further supported by phage titer enrichments shown in Tables 2 and 3. Approximate 160 fold and 90 fold enrichment were observed for third round panning from MSIA™ Streptavidin D.A.R.T's® and conventional panning method respectively. The enrichment pattern using the MSIA™ Streptavidin D.A.R.T's® panning was higher in comparison to the conventional method. The negative control with M13KO7 was 0.12 and 0.10 for MSIA™ Streptavidin D.A.R.T's® and conventional panning method respectively. 3.4. Monoclonal phage ELISA After enrichment of specific antibody phage binders against rHlyE from previous panning, 185 monoclones were randomly picked and propagated from third panning round from both MSIA™ Streptavidin D.A.R.T's® panning and conventional panning. The monoclonal ELISA results show 9 anti-rHlyE monoclones from MSIA™ Streptavidin D.A.R.T's® panning and 1 monoclone from conventional plate panning with OD405 nm exceeding 0.2 after normalizing with the background readout (Fig. 5a b, c and d) from a total 185 selected monoclones. 3.5. DNA sequencing for possible anti-rHlyE monoclonal dAbs The 9 anti-rHlyE monoclonal dAbs were sequenced for clonal identification. The sequencing results identified 5 unique monoclonal dAbs with proper sequences that are in-frame. The clones are dAb-HlyE 13, dAb-HlyE 38, dAb-HlyE 69, dAb-HlyE 127 and dAb-HlyE 176 (Table 4). Nonetheless, a total of 4 from the 5 clones showed the presence of an amber stop in the complementarity determining regions (CDRs) and 2

Fig. 2. Coomassie blue staining of a 12% SDS-polyacrylamide gel showing (a) rUb and (b) rHlyE with M indicates BLUelf prestained protein marker; lane 1, crude of lysate protein; lane 2, flow-through of the lysate protein; lane 3, first wash, L4-7, lysates after passage through affinity column.

10

C.F. Chin et al. / Journal of Microbiological Methods 120 (2016) 6–14

4. Discussion

Fig. 3. ELISA result of MSIA™ Streptavidin D.A.R.T's® based on rUb monoclonal antibody enrichment. Streptavidin and M13K07 are negative controls. All values represent means ± SD. n = 2.

clones showed mutations in the framework. Nonetheless, dAb monoclones are successfully being enriched using MSIA™ Streptavidin D.A.R.T's® panning. Only one potential monoclonal dAbs, ie. monoclone 32 from Fig. 5c with OD405 nm = 0.36 was enriched from conventional panning method. Nonetheless, after sequenced, the monoclone discussed was not domain antibody, or in other words, no monoclone was selected from conventional panning method as compared to 5 monoclones selected from MSIA™ Streptavidin D.A.R.T's® panning method.

3.6. Soluble dAb ELISA detection After sequencing, the clones, i.e. dAb-HlyE 13, dAb-HlyE 38, dAbHlyE 69, dAb-HlyE 127 and dAb-HlyE 176 were further expressed and soluble ELISA was also performed of the crude fraction. The expression did not produce high yield of proteins, but sufficient for ELISA analysis. The result shows after taking into account the background readout, OD405nm readout of dAb_HlyE 13 was 0.24, dAb_HlyE 38 and dAb_HlyE 176 was 0.1. The remaining clones dAb_HlyE 69 and dAb_HlyE 127 were less than 0.1. The readings of clones dAb_HlyE 69 and dAb_HlyE 127 were found to be lower than the negative control which indicates that these two clones may not be functional in soluble form.

The MSIA™ method is an immunoaffinity approach initially designed for protein analyte purification for MS detection (Nelson et al., 1995; Kiernan, 2007). Nonetheless, the principles involved in the conventional MSIA™ method can be modified for use in antibody phage display. The application of MSIA™ Streptavidin D.A.R.T's® with a standard electronic multichannel pipettor and adjustable pipette stand is a cost-effective semi-automated solution for phage display panning. During phage panning experiments, antigens are anchored to various types of solid supports, such as magnetic beads (Walter et al., 2001), column matrix (Noppe et al., 2009), nitrocellulose (Hawlisch et al., 2001) or plastic surfaces in the form of polystyrene tubes (Hust et al., 2002) and 96 well polystyrene microtiter plates (Krebs et al., 2001). The different solid phases used for panning is unique in terms of solid surface characteristics, loading capacity, accessible surface area as well as resistance to elution medias. The MSIA™ Streptavidin D.A.R.T's® take advantage of the natural affinity between streptavidin and biotin by utilizing a recombinant form of streptavidin originally isolated from Streptomyces avidinii to allow easy immobilization of biotinylated antigens onto the MSIA™ Streptavidin D.A.R.T's® solid matrix. This allows for rapid and specific presentation of biotinylated antigens on the surface of the MSIA™ solid phase. The packing material on the MSIA™ tips allows for high antigen binding with low background. This is shown by binding capacity of 0.157 mg for biotinylated rHlyE antigen onto the MSIA™ tips. The pore size of the packing material allows it to withstand high pressure and flow rates besides providing a higher surface area for improved target binding to the solid phase. This would permit more target proteins to be accessible to the phage antibodies. The pore size of the MSIA™ tip has never been reported to allow the flow of phage particles across the packing material. The main concern surrounding the use of flow-through type solid phase materials is sample clogging that can result in the loss of either the ligand or analyte during the process. The initial assessment for antibody phage binding was done with the MSIA™ tips to determine the suitability of the method for application with phage particles. This was done by using rUb as the target antigen for capture and elution of an in-house anti-rUb scFv. The monoclonal ELISA result shows the anti-rUb scFv was able to bind with rUb with low background. The recovery of phage by elution shows the ability of the phage particles to move freely in the MSIA™ tips. The surface material of the MSIA™ tips demonstrates low non–specific binding between anti-rUb scFv with streptavidin. This can be assessed by coating streptavidin as a control to the plate instead of rUb to examine the binding of phage presenting antibodies to streptavidin. A control using M13KO7 helper phage against rUb also shows low binding reflecting the suitability of the packing material for panning as it does not bind to blank phage particles non-specifically. M13KO7 helper phage provides structural and functional proteins for assembly of full functional antibody-displayed

Fig. 4. Polyclonal ELISA result of (a) MSIA™ Streptavidin D.A.R.T's® based panning (b) conventional microtiter plate panning against rHlyE using synthetic domain antibody library. Experiment was performed in duplicates.

C.F. Chin et al. / Journal of Microbiological Methods 120 (2016) 6–14 Table 2 Enrichment of phage from each round of MSIA™ Streptavidin D.A.R.T's panning.

Round 1 Round 2 Round 3

Input (CFU)

Output(CFU)

Ratio (ouput/input)

2.1 × 1012 5.0 × 1010 2.0 × 1010

2.0 × 104 2.0 × 102 3.0 × 104

9.5 × 10−9 4.0 × 10−9 1.5 × 10−6

Table 3 Enrichment of phage from each round of conventional panning.

Round 1 Round 2 Round 3

Input (CFU)

Output(CFU)

Ratio (ouput/input)

1.0 × 1012 1.4 × 1012 1.0 × 1010

1.0 × 106 2.0 × 106 9.0 × 105

1.0 × 10−6 1.4 × 10−6 9.0 × 10−5

phage. This is because phagemids are used to introduce the gene of the foreign antibody sequence for packaging as phage (Qi et al., 2012). The application of the M13KO7 in the immunoassay for analysis is to ensure

11

that no non-specific binding of blank phage particles are binding to the antigen or surface. This is to reduce the possibility of any false positivity from the selection. Streptavidin is a robust molecule making it ideal for the disruption of biotinylated ligands-analyte binding without having to destroy the streptavidin–biotin interaction during elution (Wilchek et al., 2006; Laitinen et al., 2007). This makes it suitable for panning experiments as the ability to withstand harsh conditions allows for customization of phage elution conditions such as acidic glycine-HCL buffer (Dottavio, 1996; Hoogenboom et al., 1998). Furthermore, the MSIA™ tips allows for a low background recovery of non-specific binders which is shown in Table 1 where only target specific phage will bind to the target (rUb) but not to non- specific target antigen (rHlyE) and MSIA™ tip alone (negative). Therefore, the protocol was then applied to conduct a proper panning experiment using a biotinylated antigen (rHlyE) with an in-house synthetic dAb library with the variable domain of heavy chain whereby in the absence of light chain. In the panning experiment, the protocol used for the single scFv enrichment against rUb was used as the starting guide. We noticed that the complexity of the domain library was able to generate background

Fig. 5. Monoclonal ELISA result of domain antibody phage raised against rHlyE by MSIA™ Streptavidin D.A.R.T's® panning with (a) positive (po) binding of anti-eGFP scFv monoclonal phage against recombinant eGFP, negative (neg) control of binding of blank M13K07 against rHlyE and monoclones 1–90; (b) monoclones 91–185; monoclonal ELISA result of domain antibody phage raised against rHlyE by conventional panning with (c) positive (po) binding of anti-eGFP scFv monoclonal phage against recombinant eGFP, negative (neg) control of binding of blank M13K07 against rHlyE and monoclones 1–90; (d) monoclones 91–185.

12

C.F. Chin et al. / Journal of Microbiological Methods 120 (2016) 6–14

Table 4. Amino acid sequences for 5 unique anti-rHlyE monoclonal dAbs. Green letter indicates substitution of amber stop codon with glutamine (Q). Monoclone

Amino acid sequence

dAb–HlyE 13

dAb–HlyE 38

dAb–HlyE 69

dAb–HlyE 127

dAb–HlyE 176

signals during enrichment applying the method used for the single scFv enrichment. This is expected as the use of a single target specific scFv greatly reduces the possibility for non-specific binding. However, when applying a diverse naïve library, the sheer complexity of the clones in the library can generate binders against the surface material. Therefore, several modifications were introduced to improve the library panning protocol. Additional pipetting cycles were performed to ensure optimal coupling between the biotinylated rHlyE with the MSIA™ Streptavidin D.A.R.T's®. An additional blocking step with skimmed milk was applied to reduce non-specific binding between library phage with the rHlyE and MSIA™ tip packing material. Higher speed settings are needed for washing steps to wash off low affinity and non-specific bound phage which could further enhance the affinity of bound antibody phage against the target rHlyE. The panning experiment with the library showed an enrichment of specific antibodies to rHlyE. In phage display, panning is an iterative process that continuously enriches and multiplies specific binders from a pool of predominantly non-binders until the specific binders becomes the majority population (Konthur and Crameri, 2003). However, the slight decrease in readout for round 2 could be attributed to a higher pool of unspecific binding in round 1 as the stringency was not high in round 1. This can result in a higher readout as a higher population of the non-specific binders is not completely removed. Therefore, when the selection stringency is increased in round 2, this facilitates the removal of more non-specific binders from the enriched pool. This will leave behind a lower population of specific binding phage resulting in

a decrease in the readout. The increase in readout from round 2 to 3 indicates successful enrichment of positive binders. This is further supported by enrichment of phage titer from MSIA™ Streptavidin D.A.R.T's® panning (Table 2) which indicates 160 fold enrichment of the third round panning to the first round panning. A comparison of MSIA™ Streptavidin D.A.R.T's® panning with the microtiter plate panning method would allow an understanding on the potential use of the MSIA™ Streptavidin D.A.R.T's® panning method for antibody isolation. The polyclonal phage ELISA result of the microtiter plate panning showed a slight increment in OD405nm values indicating successful enrichment of antibodies against rHlyE. This is further supported by phage titers from each microtiter plate panning round (Table 3) which shows 90 fold enrichment from first round to third round panning. Nonetheless, the OD405nm readout of the microtiter plate panning is relatively lower compared to that of MSIA™ Streptavidin D.A.R.T's® panning as the OD405nm readout from the microtiter plate panning from round 1 to round 3 after subtracting the background were 0.23, 0.28, 0.29; whereas for MSIA™ Streptavidin D.A.R.T's® panning were 0.21, 0.14 and 0.62. This could be subjected to the amount of phage particles successfully rescued during the panning process. Even so, it is also likely that different antigens may result in different panning results with different panning strategies due to the nature of the protein. An optimized higher blocking condition with 5% PTM was required for the conventional panning method due to the presence of a higher background reading. Initially, both panning methods were carried out using acid elution method. However, the high background for non-specific binding was still observed only for conventional panning but not MSIA™ Streptavidin D.A.R.T's® panning. Therefore, trypsin elution was used for the conventional panning to reduce the background from non-specific binding as suggested by Lee et al. whereby trypsin or proteolytic elution can yield lower nonspecific phage as compared to acid or non-specific elution (Lee et al., 2007). However, in the context of the applicability of MSIA™ Streptavidin D.A.R.T's® as a panning method, the method was successful at enriching monoclonal antibodies against rHlyE at satisfactory rates. This makes MSIA™ Streptavidin D.A.R.T's® panning method a reliable alternative method to enrich target antibodies. We further screened for monoclonal domain antibodies from the third panning round. A cut-off emission of 0.2 after taking into consideration background interference was employed for monoclonal ELISA. This is because the monoclonal phage packaging was carried out independently in each well, thus quality control to ensure that equal number of phage particles presenting the phage was used in each well is difficult to be ascertained. In addition to differences in affinities, the variation in phage packaging efficiency could also contribute to lower OD405nm readings. This is why the cut-off emission of 0.2 was used. This is to elevate any possibility of losing any potential monoclonal antibody during selection. There were 9 potential monoclonal antibodies that showed positive binding against rHlyE based on monoclonal ELISA of MSIA™ Streptavidin D.A.R.T's® panning method. The clones were confirmed by sequencing which identified only 5 unique clones with proper sequences that are in-frame. The clones are dAb-HlyE 13, dAb-HlyE 38, dAb-HlyE 69, dAb-HlyE 127 and dAb-HlyE 176. A total of 4 clones of the 5 showed the presence of amber stop codons in the complementarity determining regions (CDRs). This is expected as the synthetic dAb library used was designed with NNK degeneracy in the CDRs. Although amber stop clones were isolated, this phenomenon is possible as the stop codon would read as glutamine when using amber suppressor strains such as TG1 (Weigert et al., 1965). The 2 clones that showed mutations in the framework is likely due to large scale gene assembly during the cloning process in library preparation. The mutations that occurred resulted mainly as substitution which did not affect the binding capabilities of these clones. Nevertheless, the application of the proposed MSIA™ Streptavidin D.A.R.T's® for antibody phage display panning was still able to enrich monoclonal dAbs against the target antigen. On the other hand, no positive enrichment was obvious with the

C.F. Chin et al. / Journal of Microbiological Methods 120 (2016) 6–14

13

Acknowledgments The authors would like to acknowledge support by the Malaysian Ministry of Education through the Higher Institution Centre of Excellence (HICoE) Grant (Grant No. 311/CIPPM/44001005) and Universiti Sains Malaysia Research University Cluster Grant (Grant No. 1001/PSKBP/8630015).

References

Fig. 6. Soluble ELISA of 5 monoclonal dAbs against rHlyE with positive (po) indicates binding of eGFP soluble antibody against recombinant GFP, negative (neg) control binding of anti-eGFP soluble antibody against rHlyE. Experiment was performed with triplicate.

conventional panning method. The only positive clone obtained was determined to be truncated after sequencing. The lack of positive clones could be due to the high background during panning that complicates the monoclonal isolation process. The 5 anti-rHlyE monoclonal dAbs were subsequently expressed and soluble ELISA was performed to examine the functionality and binding efficiency of the soluble dAb against rHlyE. Results shown in Fig. 6 indicate the soluble dAb fragment, i.e. dAb-HlyE 13 was able to bind to the target antigen. However, the OD405nm readout was not as good when compared to the readout in phage form. This is likely due to the higher amount of antibody fragments being presented on phage in comparison to the soluble fraction. The low protein yield from the expression is likely due to poor solubility and expression of the clones. This could also be due to the presence of amber stop codons within the clones, although amber suppressor strains were used. Clones dAb_HlyE 69 and dAb_HlyE 127 were found to be non-functional in soluble form with a loss in binding. When compared to the conventional panning which requires 2 days to accomplish a single round, the proposed MSIA™ Streptavidin D.A.R.T's® panning take precedence for ease and rapidity. A single selection round can be performed within a day, thus allowing the entire panning process to be completed in a week. In the MSIA™ Streptavidin D.A.R.T's® workflow, the panning process can also be customized to yield antibodies of specific characteristics like any other panning method. The increase in stringency during incubation and wash steps are easily applied with the integration of an electronic pipettor. This allows for a cost effective semi-automated panning solution without the need of a large and expensive instrumentation. The use of streptavidin in this set up allows for less contamination as biotin-streptavidin system is specific and robust, allowing modifications to elution strategies for customization. Here, the proposed method allows for convenient, rapid and specific panning of antibodies with the customizations in panning design much like conventional methods. The MSIA™ Streptavidin D.A.R.T's® antibody panning method is an attractive alternative for semi-automated recombinant antibody generation processes. Conflict of interest LWL is an employee of Fisher Scientific (M) Sdn Bhd, the authorized distributor for Thermo. The MSIA™ Streptavidin D.A.R.T's® were provided by Fisher Scientific (M) Sdn Bhd.

Casey, J.L., Coley, A.M., Anders, R.F., Murphy, V.J., Humberstone, K.S., Thomas, A.W., Foley, M., 2004. Antibodies to malaria peptide mimics inhibit Plasmodium falciparum invasion of erythrocytes. Infect. Immun. 72, 1126–1134. Coomber, D.W., 2002. Panning of antibody phage-display libraries. In: O'Brien, P.M., Aitken, R. (Eds.), Antibody Phage Display. Humana Press, Totowa, pp. 133–145. De Kruif, J., Boel, E., Logtenberg, T., 1995. Selection and application of human single chain Fv antibody fragments from a semi-synthetic phage antibody display library with designed CDR3 regions. J. Mol. Biol. 248, 97–105. Diamandis, E.P., Christopoulos, T.K., 1991. The biotin-(strept) avidin system: principles and applications in biotechnology. Clin. Chem. 37, 625–636. Dottavio, D., 1996. Phagemid-displayed peptide libraries. In: Kay, B.K., Winter, J., McCafferty, J. (Eds.), Phage Display of Peptides and Proteins. A Laboratory Manual. Academic Press, San Diego, pp. 113–125. Frenzel, A., Kügler, J., Wilke, S., Schirrmann, T., Hust, M., 2014. Construction of human antibody gene libraries and selection of antibodies by phage display. In: Steinitz, M. (Ed.), Human Monoclonal Antibodies. Humana Press, Totowa, pp. 215–243. Hairul Bahara, N.H., Choong, Y.S., Lim, T.S., 2015. Construction of a semisynthetic human VH single-domain antibody library and selection of domain antibodies against αcrystalline of Mycobacterium tuberculosis. J. Biomol. Screen. Hairul Bahara, N.H., Tye, G.J., Choong, Y.S., Ong, E.B.B., Ismail, A., Lim, T.S., 2013. Phage display antibodies for diagnostic applications. Biologicals 41, 209–216. Hammers, C.M., Stanley, J.R., 2014. Antibody phage display: technique and applications. J. Investig. Dermatol. 134, e17. Hawlisch, H., Müller, M., Frank, R., Bautsch, W., Klos, A., Köhl, J., 2001. Site-specific antiC3a receptor single-chain antibodies selected by differential panning on cellulose sheets. Anal. Biochem. 293, 142–145. Hoogenboom, H.R., de Bruïne, A.P., Hufton, S.E., Hoet, R.M., Arends, J.-W., Roovers, R.C., 1998. Antibody phage display technology and its applications. Immunotechnology 4, 1–20. Hust, M., Maiss, E., Jacobsen, H.-J., Reinard, T., 2002. The production of a genus-specific recombinant antibody (scFv) using a recombinant potyvirus protease. J. Virol. Methods 106, 225–233. Kiernan, U.A., 2007. Quantitation of target proteins and post-translational modifications in affinity-based proteomics approaches. Expert Rev. Proteomics 4, 421–428. Kiernan, U.A., Tubbs, K.A., Nedelkov, D., Niederkofler, E.E., Nelson, R.W., 2002. Comparative phenotypic analyses of human plasma and urinary retinol binding protein using mass spectrometric immunoassay. Biochem. Biophys. Res. Commun. 297, 401–405. Kiernan, U.A., Tubbs, K.A., Nedelkov, D., Niederkofler, E.E., Nelson, R.W., 2003. Detection of novel truncated forms of human serum amyloid A protein in human plasma. FEBS Lett. 537, 166–170. Konthur, Z., Crameri, R., 2003. High-throughput applications of phage display in proteomic analyses. Targets 2, 261–270. Krebs, B., Rauchenberger, R., Reiffert, S., Rothe, C., Tesar, M., Thomassen, E., Cao, M., Dreier, T., Fischer, D., Höß, A., 2001. High-throughput generation and engineering of recombinant human antibodies. J. Immunol. Methods 254, 67–84. Laitinen, O.H., Nordlund, H.R., Hytönen, V.P., Kulomaa, M.S., 2007. Brave new (strept) avidins in biotechnology. Trends Biotechnol. 25, 269–277. Lee, C.M., Iorno, N., Sierro, F., Christ, D., 2007. Selection of human antibody fragments by phage display. Nat. Protoc. 2, 3001–3008. Lopez, M.F., Rezai, T., Sarracino, D.A., Prakash, A., Krastins, B., Athanas, M., Singh, R.J., Barnidge, D.R., Oran, P., Borges, C., 2010. Selected reaction monitoring-mass spectrometric immunoassay responsive to parathyroid hormone and related variants. Clin. Chem. 56, 281–290. Nelson, R.W., Krone, J.R., Bieber, A.L., Williams, P., 1995. Mass spectrometric immunoassay. Anal. Chem. 67, 1153–1158. Niederkofler, E.E., Kiernan, U.A., O'Rear, J., Menon, S., Saghir, S., Protter, A.A., Nelson, R.W., Schellenberger, U., 2008. Detection of endogenous B-type natriuretic peptide at very low concentrations in patients with heart failure. Circ. Heart Fail. 1, 258–264. Niederkofler, E.E., Tubbs, K.A., Gruber, K., Nedelkov, D., Kiernan, U.A., Williams, P., Nelson, R.W., 2001. Determination of β-2 microglobulin levels in plasma using a high-throughput mass spectrometric immunoassay system. Anal. Chem. 73, 3294–3299. Noppe, W., Plieva, F., Galaev, I.Y., Pottel, H., Deckmyn, H., Mattiasson, B., 2009. Chromato-panning: an efficient new mode of identifying suitable ligands from phage display libraries. BMC Biotechnol. 9, 21. Parker, C.E., Pearson, T.W., Anderson, N.L., Borchers, C.H., 2010. Mass-spectrometry-based clinical proteomics—a review and prospective. Analyst 135, 1830–1838. Qi, H., Lu, H., Qiu, H.-J., Petrenko, V., Liu, A., 2012. Phagemid vectors for phage display: properties, characteristics and construction. J. Mol. Biol. 417, 129–143.

14

C.F. Chin et al. / Journal of Microbiological Methods 120 (2016) 6–14

Smith, G.P., 1985. Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228, 1315–1317. Walter, G., Konthur, Z., Lehrach, H., 2001. High-throughput screening of surface displayed gene products. Comb. Chem. High Throughput Screen. 4, 193–205. Weigert, M.G., Lanka, E., Garen, A., 1965. Amino acid substitutions resulting from suppression of nonsense mutations: II. Glutamine insertion by the Su-2 gene; tyrosine insertion by the Su-3 gene. J. Mol. Biol. 14, 522–527.

Wilchek, M., Bayer, E.A., Livnah, O., 2006. Essentials of biorecognition: the (strept) avidin– biotin system as a model for protein–protein and protein–ligand interaction. Immunol. Lett. 103, 27–32.