Comparison of Solid-Phase and Liquid-Phase Antigen-Based Fab Library Screening Methods
2026-10-25
IntroductionANTIBODY
Antibody phage display ties a binding phenotype to the gene that encodes it, and more than three decades on it remains the workhorse of in vitro antibody discovery. A library holding 1011 Fab variants is only as useful as the selection step that interrogates it, and every selection begins with one deceptively simple decision: should the antigen be fixed to a solid surface, or left free in solution? This review walks through both formats, reports what published head-to-head comparisons actually measured, and closes with the practical options available at Alpha Lifetech for teams that would rather outsource the pipeline than build it in house.
Why the Fab Format?ANTIBODY
In 1985 George Smith inserted foreign DNA into gene III of a filamentous phage and showed that the resulting fusion phage could be enriched more than a thousand-fold by affinity capture[1]. Five years later McCafferty and colleagues displayed complete antibody variable domains on fd bacteriophage and recovered rare binders — roughly one in a million — by affinity chromatography[2]. That pairing of a physical binder with its own coding sequence is the whole engine of antibody phage display, and it now delivers products: more than seventy phage-derived antibodies have entered clinical studies and fourteen have reached approval[3].
Most of those campaigns display the antigen-binding fragment, or Fab. A phage-displayed human Fab pairs the variable domains VH and VL (Vκ or Vλ) with the constant domains CH1 of IgG1 and CL, giving a monovalent heterodimer anchored to the virion through pIII[4]. Because the two chains are held by their native constant-domain interface rather than an engineered peptide linker, a Fab behaves more like the arm of a natural antibody than a single-chain format does: it resists aggregation better, preserves the VH/VL pairing the immune system selected, and converts to full-length IgG without re-engineering the variable regions. Protocols built on the pComb3 phagemid family scale to 108–1011 independent Fab clones[4].
Most of those campaigns display the antigen-binding fragment, or Fab. A phage-displayed human Fab pairs the variable domains VH and VL (Vκ or Vλ) with the constant domains CH1 of IgG1 and CL, giving a monovalent heterodimer anchored to the virion through pIII[4]. Because the two chains are held by their native constant-domain interface rather than an engineered peptide linker, a Fab behaves more like the arm of a natural antibody than a single-chain format does: it resists aggregation better, preserves the VH/VL pairing the immune system selected, and converts to full-length IgG without re-engineering the variable regions. Protocols built on the pComb3 phagemid family scale to 108–1011 independent Fab clones[4].
Send Inquiry
Contact Us For Best Would you Like to Know more We can Give you the answer, For inquiries about our products and services. please leave your e-mail to us and will reply within 24 hours.
Click for inquiryBefore Any Panning: Build the LibraryANTIBODY
Selection can only recover what the library already contains. A naïve Fab library is usually assembled from the IgM repertoire, whereas an immune library draws on the IgG repertoire of donors or immunised animals whose response the target has already shaped[5]. Either way the antibody genes are amplified by PCR and recombined combinatorially, so heavy and light chains are shuffled relative to their original pairing. That expands accessible diversity well beyond the in vivo repertoire and means the final binder need not resemble any single B cell clone[5].
What Every Panning Round Has in CommonANTIBODY
Biopanning is an iterative loop: expose the library to the target, remove what does not bind, recover what does, amplify the survivors in E. coli, and repeat[6]. The in vitro setting is what makes the method attractive, because antigen concentration, wash stringency and elution conditions can be set deliberately rather than being dictated by an animal's immune system[6]. Three presentation modes account for most campaigns — antigen immobilised on a solid support, biotinylated antigen in solution, and antigen on a cell surface — and all three are compatible with Fab libraries[7]. Whichever mode is chosen, enrichment accumulates across rounds, each cycle raising the fraction of specific binders[8].
Solid-Phase Panning: Antigen on a SurfaceANTIBODY
In solid-phase panning the antigen is adsorbed onto a polystyrene immunotube, a microplate well or a membrane, then blocked, incubated with the library, washed and eluted. It is the simplest format to run: no chemical modification of the antigen is needed, coating density can be tuned, and the infrastructure is a refrigerator and a shaker.
The drawbacks follow from the same chemistry. Adsorption to a hydrophobic plastic surface can distort a protein, so some epitopes are lost while others — epitopes that only exist on a flattened antigen — are created. Binders recovered this way sometimes recognise the coated target but not its native form. Steric crowding on the surface can also mask epitopes that are perfectly accessible in solution.
Solid-phase selection tolerates imperfect material better than one might expect. Lim and colleagues deliberately coated crude, unpurified antigen and saturated the library with blocking agents before binding — a strategy they called Yin-Yang panning — and still isolated three unique human antibodies against the MERS-CoV nucleoprotein from a 10⁹-member library[9]. For difficult or low-yield targets that tolerance is often decisive[9].
The drawbacks follow from the same chemistry. Adsorption to a hydrophobic plastic surface can distort a protein, so some epitopes are lost while others — epitopes that only exist on a flattened antigen — are created. Binders recovered this way sometimes recognise the coated target but not its native form. Steric crowding on the surface can also mask epitopes that are perfectly accessible in solution.
Solid-phase selection tolerates imperfect material better than one might expect. Lim and colleagues deliberately coated crude, unpurified antigen and saturated the library with blocking agents before binding — a strategy they called Yin-Yang panning — and still isolated three unique human antibodies against the MERS-CoV nucleoprotein from a 10⁹-member library[9]. For difficult or low-yield targets that tolerance is often decisive[9].
Liquid-Phase Panning: Antigen in SolutionANTIBODY
Liquid-phase, or solution-phase, panning reverses the logic. The antigen — usually biotinylated — is mixed with the library so that binding happens in three dimensions, and the complexes are then captured, most often on streptavidin-coated magnetic beads[10-11].
Three consequences follow. The antigen never touches plastic, so conformational epitopes survive and recovered binders are more likely to recognise the target as it actually exists. The interaction is monovalent and antigen concentration is a free variable, so selection pressure can be tightened round by round simply by using less antigen. And because capture is fast and the beads magnetic, washes are quicker and more reproducible than in a tube.
The concentration lever is the real advantage. Hawkins, Russell and Winter showed that phage can be mixed with sub-microgram amounts of biotinylated antigen held below the dissociation constant of the antibody, so that only the tightest binders capture antigen at all — an arrangement that distinguishes antibodies whose affinities differ only modestly[12]. They extended the idea to off-rate selection, preloading phage with labelled antigen and challenging them with excess unlabelled antigen so that slow-releasing clones survive[12]. Neither experiment is feasible when the antigen sits on a wall at unknown local density.
Three consequences follow. The antigen never touches plastic, so conformational epitopes survive and recovered binders are more likely to recognise the target as it actually exists. The interaction is monovalent and antigen concentration is a free variable, so selection pressure can be tightened round by round simply by using less antigen. And because capture is fast and the beads magnetic, washes are quicker and more reproducible than in a tube.
The concentration lever is the real advantage. Hawkins, Russell and Winter showed that phage can be mixed with sub-microgram amounts of biotinylated antigen held below the dissociation constant of the antibody, so that only the tightest binders capture antigen at all — an arrangement that distinguishes antibodies whose affinities differ only modestly[12]. They extended the idea to off-rate selection, preloading phage with labelled antigen and challenging them with excess unlabelled antigen so that slow-releasing clones survive[12]. Neither experiment is feasible when the antigen sits on a wall at unknown local density.
Head-to-Head Comparisons: What the Data ShowANTIBODY
Comparative studies are rarer than one would like, but the published ones are informative.
Lai and colleagues built three libraries from VEGF-immunised mice — scFv-M13KO7, Fab-M13KO7 and scFv-Hyperphage — and ran each through both solid-phase and solution-phase selection, giving six panning combinations[13]. Sixty-four anti-VEGF antibodies emerged, and the six arms did not deliver interchangeable outputs: format and selection strategy jointly shaped the epitope coverage and functional profile of the panel, and three antibodies outperformed the benchmark in a xenograft model[13]. The lesson is not that one arm won, but that running both roughly doubles the epitope space sampled from the same library.
Lakzaei and colleagues compared three strategies against diphtheria toxoid over four rounds: soluble antibody-capturing, pH-stepwise elution, and conventional solid-phase panning[14]. Solution capture yielded fifteen positive clones against nine for conventional panning and five for pH-stepwise elution, and two of the three neutralising clones came from the solution-capture arm[14].
Panagides and colleagues posed a related question about presentation rather than phase, comparing a surface-tethered recombinant extracellular domain with cell-based biopanning against a type I membrane receptor; the soluble recombinant form gave the more diverse repertoire[15]. Antigen presentation, not just binding chemistry, drives what you get out.
Membrane proteins, unstable complexes and targets that fold only in a cellular context defeat both formats. The field has accumulated alternatives — depletion and selection on whole cells, panning on tissue sections, in vivo selection, fluorescence-activated sorting, and solution pull-down variants — each with its own handling requirements[16]. The trade-off is consistent: formats that best preserve the native target tend to be slower and to enrich more irrelevant binders, so they need rigorous counter-selection.
Lai and colleagues built three libraries from VEGF-immunised mice — scFv-M13KO7, Fab-M13KO7 and scFv-Hyperphage — and ran each through both solid-phase and solution-phase selection, giving six panning combinations[13]. Sixty-four anti-VEGF antibodies emerged, and the six arms did not deliver interchangeable outputs: format and selection strategy jointly shaped the epitope coverage and functional profile of the panel, and three antibodies outperformed the benchmark in a xenograft model[13]. The lesson is not that one arm won, but that running both roughly doubles the epitope space sampled from the same library.
Lakzaei and colleagues compared three strategies against diphtheria toxoid over four rounds: soluble antibody-capturing, pH-stepwise elution, and conventional solid-phase panning[14]. Solution capture yielded fifteen positive clones against nine for conventional panning and five for pH-stepwise elution, and two of the three neutralising clones came from the solution-capture arm[14].
Panagides and colleagues posed a related question about presentation rather than phase, comparing a surface-tethered recombinant extracellular domain with cell-based biopanning against a type I membrane receptor; the soluble recombinant form gave the more diverse repertoire[15]. Antigen presentation, not just binding chemistry, drives what you get out.
Membrane proteins, unstable complexes and targets that fold only in a cellular context defeat both formats. The field has accumulated alternatives — depletion and selection on whole cells, panning on tissue sections, in vivo selection, fluorescence-activated sorting, and solution pull-down variants — each with its own handling requirements[16]. The trade-off is consistent: formats that best preserve the native target tend to be slower and to enrich more irrelevant binders, so they need rigorous counter-selection.
Choosing a Format: A Practical ViewANTIBODY
-- Solid-phase suits an abundant, stable antigen, an exploratory campaign, or throughput across many targets.
-- Solution-phase suits membrane proteins and fragile complexes, conformational or neutralising antibodies, and campaigns that need to steer affinity by cutting antigen concentration round by round.
-- Run both in parallel when epitope breadth matters more than speed; the VEGF comparison above is the clearest published argument.
-- Add NGS when colony picking keeps returning the same dominant clones, or when you suspect binders are lost to sampling rather than selection.
-- Solution-phase suits membrane proteins and fragile complexes, conformational or neutralising antibodies, and campaigns that need to steer affinity by cutting antigen concentration round by round.
-- Run both in parallel when epitope breadth matters more than speed; the VEGF comparison above is the clearest published argument.
-- Add NGS when colony picking keeps returning the same dominant clones, or when you suspect binders are lost to sampling rather than selection.
Alpha Lifetech supports the whole chain described above, from target material to characterised binders.
-- Custom Fab library construction. Naïve, immune and synthetic human Fab libraries, plus chimeric rabbit/human and mouse/human Fab formats, delivered with diversity and insertion-rate QC.
-- Biopanning services. Solid-phase, liquid-phase, cell-based and tissue-based selection on the same library, with antigen concentration and wash stringency designed around the target.
-- Antigen preparation. Recombinant protein and peptide expression, biotinylation and quality control — the step most campaigns underestimate.
-- Antibody engineering. Conversion of Fab to full-length IgG or other formats, affinity maturation, humanisation, and sequence-defined antibody production.
-- Complementary routes. Hybridoma development, single B cell screening, and VHH/nanobody library construction and screening for targets that Fab formats handle poorly.
Keeping antigen, library and selection records with a single partner also means a campaign that stalls at round three can be diagnosed from the data rather than repeated from scratch.
-- Custom Fab library construction. Naïve, immune and synthetic human Fab libraries, plus chimeric rabbit/human and mouse/human Fab formats, delivered with diversity and insertion-rate QC.
-- Biopanning services. Solid-phase, liquid-phase, cell-based and tissue-based selection on the same library, with antigen concentration and wash stringency designed around the target.
-- Antigen preparation. Recombinant protein and peptide expression, biotinylation and quality control — the step most campaigns underestimate.
-- Antibody engineering. Conversion of Fab to full-length IgG or other formats, affinity maturation, humanisation, and sequence-defined antibody production.
-- Complementary routes. Hybridoma development, single B cell screening, and VHH/nanobody library construction and screening for targets that Fab formats handle poorly.
Keeping antigen, library and selection records with a single partner also means a campaign that stalls at round three can be diagnosed from the data rather than repeated from scratch.
FAQsANTIBODY
-
1. Does liquid-phase panning always give higher-affinity antibodies?
-
2. How many rounds of panning are enough?
-
3. Can I pan if my antigen is difficult to purify?
-
4. Must the Fab be converted into full-length IgG afterwards?
-
5. Is next-generation sequencing worth the cost on a modest campaign?
ReferenceANTIBODY
[1]Smith GP. Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science. 1985 Jun 14;228(4705):1315-7. doi: 10.1126/science.4001944. PMID: 4001944.
[2]McCafferty J, Griffiths AD, Winter G, Chiswell DJ. Phage antibodies: filamentous phage displaying antibody variable domains. Nature. 1990 Dec 6;348(6301):552-4. doi: 10.1038/348552a0. PMID: 2247164.
[3]Alfaleh MA, Alsaab HO, Mahmoud AB, Alkayyal AA, Jones ML, Mahler SM, Hashem AM. Phage Display Derived Monoclonal Antibodies: From Bench to Bedside. Front Immunol. 2020 Aug 28;11:1986. doi: 10.3389/fimmu.2020.01986. PMID: 32983137; PMCID: PMC7485114.
[4]Peng H, Rader C. Generation of Antibody Libraries for Phage Display: Human Fab Format. Cold Spring Harb Protoc. 2026 Jan 5;2026(1):pdb.prot108597. doi: 10.1101/pdb.prot108597. PMID: 39406530.
[5]Lai JY, Lim TS. Construction of Naïve and Immune Human Fab Phage Display Library. Methods Mol Biol. 2023;2702:39-58. doi: 10.1007/978-1-0716-3381-6_3. PMID: 37679614.
[6]Shim H. Antibody Phage Display. Adv Exp Med Biol. 2017;1053:21-34. doi: 10.1007/978-3-319-72077-7_2. PMID: 29549633.
[7]Peng H, Rader C. Phage Display Selection of Antibody Libraries: Panning Procedures. Cold Spring Harb Protoc. 2026 Jan 5;2026(1):pdb.prot108602. doi: 10.1101/pdb.prot108602. PMID: 39406533.
[8]Saw PE, Song EW. Phage display screening of therapeutic peptide for cancer targeting and therapy. Protein Cell. 2019 Nov;10(11):787-807. doi: 10.1007/s13238-019-0639-7. Epub 2019 May 28. PMID: 31140150; PMCID: PMC6834755.
[9]Lim CC, Woo PCY, Lim TS. Development of a Phage Display Panning Strategy Utilizing Crude Antigens: Isolation of MERS-CoV Nucleoprotein human antibodies. Sci Rep. 2019 Apr 15;9(1):6088. doi: 10.1038/s41598-019-42628-6. PMID: 30988390; PMCID: PMC6465254.
[10]Wenzel EV, Roth KDR, Russo G, Fühner V, Helmsing S, Frenzel A, Hust M. Antibody Phage Display: Antibody Selection in Solution Using Biotinylated Antigens. Methods Mol Biol. 2020;2070:143-155. doi: 10.1007/978-1-4939-9853-1_8. PMID: 31625094.
[11]Heine PA, Ruschig M, Langreder N, Wenzel EV, Schubert M, Bertoglio F, Hust M. Antibody Selection in Solution Using Magnetic Beads. Methods Mol Biol. 2023;2702:261-274. doi: 10.1007/978-1-0716-3381-6_13. PMID: 37679624.
[12]Hawkins RE, Russell SJ, Winter G. Selection of phage antibodies by binding affinity. Mimicking affinity maturation. J Mol Biol. 1992 Aug 5;226(3):889-96. doi: 10.1016/0022-2836(92)90639-2. PMID: 1507232.
[13]Lai YD, Wu YY, Tsai YJ, Tsai YS, Lin YY, Lai SL, Huang CY, Lok YY, Hu CY, Lai JS. Generation of Potent Anti-Vascular Endothelial Growth Factor Neutralizing Antibodies from Mouse Phage Display Library for Cancer Therapy. Int J Mol Sci. 2016 Feb 5;17(2):214. doi: 10.3390/ijms17020214. PMID: 26861297; PMCID: PMC4783946.
[14]Lakzaei M, Rasaee MJ, Fazaeli AA, Aminian M. A comparison of three strategies for biopanning of phage-scFv library against diphtheria toxin. J Cell Physiol. 2019 Jun;234(6):9486-9494. doi: 10.1002/jcp.27636. Epub 2018 Nov 11. PMID: 30417355.
[15]Panagides N, Zacchi LF, De Souza MJ, Morales RAV, Karnowski A, Liddament MT, Owczarek CM, Mahler SM, Panousis C, Jones ML, Fercher C. Evaluation of Phage Display Biopanning Strategies for the Selection of Anti-Cell Surface Receptor Antibodies. Int J Mol Sci. 2022 Jul 30;23(15):8470. doi: 10.3390/ijms23158470. PMID: 35955604; PMCID: PMC9369378.
[16]França RKA, Studart IC, Bezerra MRL, Pontes LQ, Barbosa AMA, Brigido MM, Furtado GP, Maranhão AQ. Progress on Phage Display Technology: Tailoring Antibodies for Cancer Immunotherapy. Viruses. 2023 Sep 9;15(9):1903. doi: 10.3390/v15091903. PMID: 37766309; PMCID: PMC10536222.
[2]McCafferty J, Griffiths AD, Winter G, Chiswell DJ. Phage antibodies: filamentous phage displaying antibody variable domains. Nature. 1990 Dec 6;348(6301):552-4. doi: 10.1038/348552a0. PMID: 2247164.
[3]Alfaleh MA, Alsaab HO, Mahmoud AB, Alkayyal AA, Jones ML, Mahler SM, Hashem AM. Phage Display Derived Monoclonal Antibodies: From Bench to Bedside. Front Immunol. 2020 Aug 28;11:1986. doi: 10.3389/fimmu.2020.01986. PMID: 32983137; PMCID: PMC7485114.
[4]Peng H, Rader C. Generation of Antibody Libraries for Phage Display: Human Fab Format. Cold Spring Harb Protoc. 2026 Jan 5;2026(1):pdb.prot108597. doi: 10.1101/pdb.prot108597. PMID: 39406530.
[5]Lai JY, Lim TS. Construction of Naïve and Immune Human Fab Phage Display Library. Methods Mol Biol. 2023;2702:39-58. doi: 10.1007/978-1-0716-3381-6_3. PMID: 37679614.
[6]Shim H. Antibody Phage Display. Adv Exp Med Biol. 2017;1053:21-34. doi: 10.1007/978-3-319-72077-7_2. PMID: 29549633.
[7]Peng H, Rader C. Phage Display Selection of Antibody Libraries: Panning Procedures. Cold Spring Harb Protoc. 2026 Jan 5;2026(1):pdb.prot108602. doi: 10.1101/pdb.prot108602. PMID: 39406533.
[8]Saw PE, Song EW. Phage display screening of therapeutic peptide for cancer targeting and therapy. Protein Cell. 2019 Nov;10(11):787-807. doi: 10.1007/s13238-019-0639-7. Epub 2019 May 28. PMID: 31140150; PMCID: PMC6834755.
[9]Lim CC, Woo PCY, Lim TS. Development of a Phage Display Panning Strategy Utilizing Crude Antigens: Isolation of MERS-CoV Nucleoprotein human antibodies. Sci Rep. 2019 Apr 15;9(1):6088. doi: 10.1038/s41598-019-42628-6. PMID: 30988390; PMCID: PMC6465254.
[10]Wenzel EV, Roth KDR, Russo G, Fühner V, Helmsing S, Frenzel A, Hust M. Antibody Phage Display: Antibody Selection in Solution Using Biotinylated Antigens. Methods Mol Biol. 2020;2070:143-155. doi: 10.1007/978-1-4939-9853-1_8. PMID: 31625094.
[11]Heine PA, Ruschig M, Langreder N, Wenzel EV, Schubert M, Bertoglio F, Hust M. Antibody Selection in Solution Using Magnetic Beads. Methods Mol Biol. 2023;2702:261-274. doi: 10.1007/978-1-0716-3381-6_13. PMID: 37679624.
[12]Hawkins RE, Russell SJ, Winter G. Selection of phage antibodies by binding affinity. Mimicking affinity maturation. J Mol Biol. 1992 Aug 5;226(3):889-96. doi: 10.1016/0022-2836(92)90639-2. PMID: 1507232.
[13]Lai YD, Wu YY, Tsai YJ, Tsai YS, Lin YY, Lai SL, Huang CY, Lok YY, Hu CY, Lai JS. Generation of Potent Anti-Vascular Endothelial Growth Factor Neutralizing Antibodies from Mouse Phage Display Library for Cancer Therapy. Int J Mol Sci. 2016 Feb 5;17(2):214. doi: 10.3390/ijms17020214. PMID: 26861297; PMCID: PMC4783946.
[14]Lakzaei M, Rasaee MJ, Fazaeli AA, Aminian M. A comparison of three strategies for biopanning of phage-scFv library against diphtheria toxin. J Cell Physiol. 2019 Jun;234(6):9486-9494. doi: 10.1002/jcp.27636. Epub 2018 Nov 11. PMID: 30417355.
[15]Panagides N, Zacchi LF, De Souza MJ, Morales RAV, Karnowski A, Liddament MT, Owczarek CM, Mahler SM, Panousis C, Jones ML, Fercher C. Evaluation of Phage Display Biopanning Strategies for the Selection of Anti-Cell Surface Receptor Antibodies. Int J Mol Sci. 2022 Jul 30;23(15):8470. doi: 10.3390/ijms23158470. PMID: 35955604; PMCID: PMC9369378.
[16]França RKA, Studart IC, Bezerra MRL, Pontes LQ, Barbosa AMA, Brigido MM, Furtado GP, Maranhão AQ. Progress on Phage Display Technology: Tailoring Antibodies for Cancer Immunotherapy. Viruses. 2023 Sep 9;15(9):1903. doi: 10.3390/v15091903. PMID: 37766309; PMCID: PMC10536222.



