Guided Antibody Selection Strategies

HuCAL® antibody generation is performed entirely in vitro, allowing Bio-Rad’s scientists to flexibly direct antibody specificities towards a range of desired outcomes. This is a feature no animal-based antibody technology can offer.

We use the term “guided selection” to refer to the various strategies, shown in the examples below, that drive HuCAL antibody generation toward chosen specificities. It is also possible to directly select matched sandwich pairs or to isolate HuCAL antibodies while using buffers or conditions that may be necessary for the final assay.

For example, we are able to produce antibodies that specifically recognize:

  • A single amino acid modification such as phosphorylation or oxidation (Fig 1, Fig 2)
  • Amino acid mutations (Fig 3, Fig 4)
  • Two individual, non-overlapping 10aa cleavage peptides, but not the 20aa full length peptide (Fig 5).

Epitope-specific and phospho-specific antibodies

The HuCAL library is ideal for selecting antibodies against specific epitopes such as phosphorylation or oxidation sites. The antibody library is ‘pre-adsorbed’ with the non-phosphorylated peptide, to remove all antibodies that might bind to it, before it is used for selection against the phosphorylated peptide. Further counter-screening and QC ensure that the antibody is phospho-specific (Fig 1, Fig 2). A similar procedure was used to develop antibodies specific for a single amino acid mutation (Fig 3).

Antibodies that don’t cross-react with closely related proteins

A subtraction strategy can be used to eliminate cross reacting antibodies from the selection process. Pre-adsorption steps and intelligent counter-screening drive the antibody library selection towards the unique epitopes on the antigen protein.

For example, HuCAL antibodies were developed to specifically recognize a 10 amino acid peptide, but not an identical 8 amino acid peptide lacking the 2 N-terminal amino acids (Fig 4), or to recognize each of two 10 amino acid peptides but not the 20 amino acid full length peptide (Fig 5). This procedure is also ideal for the generation of anti-idiotype antibodies.

Antibodies that bind to homologous proteins

When both proteins are supplied, an alternate screening strategy allows us to identify HuCAL antibodies reacting with both antigens. This technique can also be useful for generating antibodies recognizing two different isoforms of a protein, or identifying anti-peptide antibodies that also react with the parental protein (Fig 6, Fig 7).

Antibodies that bind antigens under special conditions

We can alter the screening conditions to ensure that antigen is presented to the antibody library in the form in which will be assayed - denatured, native, captured, soluble, or masked. We can even alter the binding buffer conditions to simulate your assay buffer!

See other examples:

  • HuCAL Antibody Specificity Profiling on a Microarray (Fig 8)
  • IHC Staining of Human Tonsil using a HuCAL Bivalent Fab Antibody (Fig 9)
  • Western Blot of Mouse Tissue using a HuCAL Fab Antibody (Fig 10)

Figure 1: ELISA QC of a Phospho-specific Antibody

Exquisite Figure 01

Phospho-specific antibodies were generated against a 9 amino acid peptide phosphorylated at ser-6 (P-Peptide X1). The library was pre-blocked with the non-phosphorylated peptide and then panned against the phosphorylated peptide coupled to BSA and Transferrin. ELISA QC shows that the antibody specifically recognizes the phosphorylated peptide and does not react with either the non-phosphorylated peptide, an unrelated phosphorylated peptide, or the carrier proteins. Antibody generation took 8 weeks.

Figure 2: ELISA QC of an Oxidation-specific Antibody

Exquisite Figure 02

Antibodies were generated against oxidized DJ-1 protein using a 12 amino acid DJ-1 peptide containing the C106 oxidation site. The library was pre-blocked with the non-oxidized peptide (DJ-2 peptide) and then panned against the oxidized DJ-1 peptide coupled to BSA and Transferrin. ELISA QC shows that all 3 selected antibodies specifically recognize the oxidized peptide and do not react with the non-oxidized peptide or the carrier proteins. Antibody generation took 8 weeks. Antibody AbyD3055 was subsequently shown to bind to full length oxidized DJ-1 (Ooe et al., Neurosci Lett. 2006 Oct 9; 406(3); 165-8.)

Figure 3

Exquisite Figure 03

Antibodies were generated against an x amino acid peptide with a single amino acid mutation at position x. The library was pre-blocked with the wild-type peptide and then panned against the peptide with the single amino acid mutation coupled to both BSA and Transferrin. ELISA QC shows that two of the three antibodies developed specifically recognize the mutant peptide and do not react with either the wild type peptide, or the carrier proteins. The third antibody recognizes both wild-type and mutant peptides. Antibody generation took 8 weeks.

Figure 4: HuCAL Antibody Distinguishes between 8 and 10 aa Peptides

Exquisite Figure 04

The Fab antibody was generated against the 10 aa peptide. The library was subtracted with the 8 aa peptide coupled to BSA and then panned against the 10 aa peptide coupled to BSA and Transferrin. The HuCAL Fab clearly distinguishes between 10aa peptide and 8aa peptide (difference: lack of 2aa at the N-terminus). Antibody generation time was 8 weeks.

Figure 5: Direct Selection of a Sequence- and Phosphorylation-specific Antibody

Exquisite Figure 05

Unwanted antibody specificities were first removed by preadsorbing the HuCAL library with related peptides which were either not-phosphorylated (Peptides B and D), or phosphorylated at the same amino acid but with an alternate flanking sequence (Peptide C). The depleted library was then screened against Peptide A to generate phospho- and sequence-specific antibodies. The target peptide was conjugated to both transferrin and BSA carriers for screening, to eliminate carrier specific antibodies. Three antibodies recognizing only the phosphorylated Peptide A were isolated and checked by ELISA.

Figure 6: Cross-reactivity Screening After Panning Against Peptide Derived from Mac-1 (CD11b)

Exquisite Figure 06

The M18 peptide was panned against the HuCAL library and 11 binders were expressed as Bivalent Mini-Antibodies. ELISA QC shows that all 11 antibodies bind specifically to the M18 peptide but not to the BSA or Transferrin carrier proteins, or to an unrelated peptide bound to Transferrin. The I-domain of the parental Mac1 protein is also recognized by 8 of the 11 antibodies.

Figure 7: FACS – Homologous Antigens

Exquisite Figure 07

Antibodies 1 and 2 were generated against a human protein which has a highly homologous mouse counterpart. FACS analysis of CHO cells expressing both the human and mouse proteins shows that Antibody 1 recognizes both the human and the mouse proteins, while Antibody 2 recognizes only the human form. (Data kindly provided by an academic researcher wishing to remain anonymous.)

Figure 8: Specificity Profiling on 96 Antigen Microarray

Exquisite Figure 08

Anti-Cyclophilin A HuCAL-antibody tested on 96 antigens (spotted on micro array). Data kindly provided by Protagen, Germany

Figure 9: IHC with HuCAL Bivalent Fab Antibody

Exquisite Figure 09

The HuCAL AgX system was used to generate an MHCII antigen from a cDNA fragment. This antigen was used to pan the HuCAL library for anti-MHCII antibodies. Formalin-fixed, paraffin-embedded human tonsil tissue was treated with a bivalent Fab antibody against MHCII, which was detected using mouse anti-His antibody for bridging and subsequent biotinylated anti mouse - ABC-DAB-detection system. (Data kindly provided by Prof. H. Merz, University of Lübeck, Germany).

Figure 10: Antibody Specificity - Western Blot

Exquisite Figure 10

Western blot of mouse tissues and cell lines using a HuCAL Fab generated against an EST-fragment of mKIAA0232. Data courtesy of Dr. O. Ohara, Kazusa DNA Research Institute, Kisarazu, Japan