|
Abstract INTRODUCTION:
The current antibody market includes over six million available antibodies for uses in quantification, detection, enrichment and/or perturbation of a target protein. Despite this, it is estimated that approx. 50% of commercial antibodies do not meet basic standards for characterisation, with many of these defective antibodies being found in published studies. Binding characterisation of antibodies is typically done using labelling techniques (e.g., ELISA) that signal substrate binding, or in vitro with optical sensors (e.g., SPR) that study binding in environments outside of those to be used. Immunoprecipitation and capture- based workflows are becoming increasingly common in clinical mass spectrometry (e.g., SISCAPA). We therefore developed a simple, depletion-based LC-MS/MS workflow that monitored the loss of SARS-CoV-2 nucleocapsid (NCAP) peptides in the addition of SISCAPA antibodies. Calculating the bound proportion, we successfully determined antibody saturation, rate kinetics, and selectivity between polyclonal and monoclonal antibodies in a method that can be easily used to discern in situ binding characteristics. Quantifying binding characteristics also allows for optimisation of workflows and limits the quantities of costly antibodies needed.
OBJECTIVES:
We sought to develop a new method by which binding kinetics between antibodies and COVID peptides could be measured, with the intention of carrying this method over to validate antibody binding specificity for laboratory and commercial use. Our aim was to demonstrate a workflow that was simple and scalable, while providing greater characterisation than previously existing methods. This method should be repeatable and applicable to any laboratory with a LC-MS/MS.
METHODS:
Monoclonal (MAb) and a polyclonal (PAb) SISCAPA antibodies for SARS-CoV-2 NCAP peptides (AYN, DGI, and ADE) were tested in a series of depletion experiments to characterise binding. Saturation was investigated by incubating 5 µL of either peptide at a variable concentration (0.05 - 2500 fmol/µL) with 5 µL of respective SISCAPA MAb or PAb ( 0.05 – 0.25 µg/µL) and 15 µL ammonium acetate (pH 8.5) for 30 minutes at room temperature. SISCAPA Abs were removed using a magnetic array, before the unbound fraction was analysed on a Waters Xevo TQ-XS, C18 column over a 5-minute gradient ( Mobile phase A, 0.1% formic acid. Mobile phase B, acetonitrile 0.1% formic acid). Triplicate, Ab-free controls were analysed in parallel to allow for calculation of total bound peptide.
RESULTS:
Adsorption models were explored to describe antibody-peptide interactions, revealing that the Langmuir model, which assumes a homogenous surface with finite binding sites, provided the best fit for monoclonal antibodies (MAbs), while the Freundlich model, which assumes a heterogeneous surface with infinite binding sites, was more suitable for polyclonal antibodies (PAbs). The saturation capacity (Bmax) varied across antibodies, with MAbs exhibiting higher binding capacities: MAb-AYN (27.0 pmol/µg), MAb-DGI (18.1 pmol/µg), and MAb-ADE (10.0 pmol/µg). In contrast, PAbs demonstrated lower Bmax values: PAb-DGI (3.1 pmol/µg) and PAb-ADE (7.4 pmol/µg), with PAb-AYN failing to
converge. This suggests that MAbs have a greater maximum binding capacity per µg compared to PAbs. Kinetic modelling revealed that first-order kinetics were observed for MAb-AYN, MAb-ADE, and PAb-ADE, with half-times of 0.07, 0.07, and 0.25 min, respectively. Second-order kinetics were observed for PAb-AYN, MAb-DGI, and PAb-DGI, with half-times of 0.12, 0.17, and 0.53 min, respectively. The association rate constant (K) for MAb-AYN (13.03) was significantly higher than for MAb-ADE (3.63) and MAb-DGI (5.25). Similarly, the dissociation constant (tau) for MAb-AYN (0.07) was lower than that of MAb-ADE (0.27) and MAb-DGI (0. 19), indicating stronger binding stability. For PAbs, higher affinity was observed for PAb-AYN (4.63) compared to PAb-ADE (2.77) and PAb-DGI (1.26), with PAb-AYN also exhibiting the lowest dissociation rate (0.21) relative to PAb-ADE (0.36) and PAb-DGI (0.79). Selectivity experiments showed that MAbs exhibited high specificity (>90% depletion) towards their corresponding peptides at 0.1 pmol/µL, though peptide AYN displayed some degree of off-target binding across all MAbs (18.3–97.0%). In contrast, PAbs showed lower binding efficiency (<52.9%) and lower selectivity, with peptides DGI and ADE being recognised only by their respective PAbs. Peptide AYN bound both PAb-AYN (38.9%) and PAb-DGI (51.0%), but not PAb-ADE, differing from the monoclonal results. Notably, MAb- AYN exhibited selectivity for both AYN and DGI antibodies, whereas DGI and ADE peptides showed high specificity for their corresponding MAbs. Overall, MAbs displayed higher binding capacities and a discrete number of binding sites, while PAbs had lower binding capacities with a continuous range of binding sites. The strong selectivity of MAbs for their target peptides was contrasted by the lower selectivity observed in PAbs, with some off-target binding of AYN detected across antibodies. This mass spectrometry assay effectively quantified and characterised antibody binding, providing a novel method for optimising antibody use in immunoprecipitation and capture assays, ultimately reducing the consumption of costly antibodies.
CONCLUSION:
Using mass spectrometry, we developed a novel, label free method by which the binding characteristics of Ab could be elucidated using LC-MS/MS, this method is repeatable and could be applied to other antibodies including those used for immunoprecipitation assays for mass spectrometry. This method provides an in-situ view of binding, without having to rely on other detection systems such as ELISA or SPR.
|
= Discovery stage. (53.14%, 2025)
= Translation stage. (22.33%, 2025)
= Clinically available. (24.53%, 2025)
