Inline Liquid Chromatography-Fast Photochemical Oxidation of Proteins for Targeted Structural Analysis of Conformationally Heterogeneous Mixtures

Abstract

Structural analysis of proteins in a conformationally heterogeneous mixture has long been a difficult problem in structural biology. In structural analysis by covalent labeling mass spectrometry, conformational heterogeneity results in data reflecting a weighted average of all conformers, complicating data analysis and potentially causing misinterpretation of results. Here, we describe a method coupling size-exclusion chromatography (SEC) with hydroxyl radical protein footprinting using inline fast photochemical oxidation of proteins (FPOP). Using a controlled synthetic mixture of holomyoglobin and apomyoglobin, we validate that we can achieve accurate footprints of each conformer using LC-FPOP when compared to offline FPOP of each pure conformer. We then applied LC-FPOP to analyze the adalimumab heat-shock aggregation process. We found that the LC-FPOP footprint of unaggregated adalimumab was consistent with a previously published footprint of the native IgG. The LC-FPOP footprint of the aggregation product indicated that heat-shock aggregation primarily protected the hinge region, suggesting that this region is involved with the heat-shock aggregation process of this molecule. LC-FPOP offers a new method to probe dynamic conformationally heterogeneous mixtures that can be separated by SEC such as biopharmaceutical aggregates and to obtain accurate information on the topography of each conformer.

Figure 1.. Schematic of inline FPOP.

Protein mixture of heterogeneous conformations is separated on size exclusion chromatography. FPOP reagent is injected with T- junction, just before laser irradiation. FPOP reagent contains phosphate buffer, adenine and glutamine. After the laser irradiation, the samples are collected in a quench solution containing catalase and methionine amide. Adenine absorbance is measured to make sure that all replicates have similar amount of free radicals. After the protein digestion, the samples are run onto mass spectrometer, and footprint of the individual protein conformers are calculated and plotted onto protein structure.

Figure 2.. Comparison of normalized inline LC-FPOP (blue) and off-line FPOP (orange).

The mixture of holomyoglobin and apomyoglobin was separated using SEC. Comparison is shown of peptide oxidation normalized to total oxidation of all peptides quantified for that sample between FPOP performed inline and off-line of (left) holomyoglobin and (right) apomyoglobin. Error bars represent one standard deviation from a triplicate data set. The footprint was equivalent between the inline and off-line FPOP (t-test α=0.05).

Figure 3.. Size exclusion chromatography of adalimumab.

(A) The native adalimumab gives a single peak around 30 min. The adalimumab was heated at 80 °C for 15 min to make aggregates. (B) Separation of monomeric and aggregated adalimumab by SEC.

Figure 4.. FPOP of the native and aggregated adalimumab.

Adalimumab heated at 80 °C for 15 min results in its aggregation. A mixture of monomer and aggregated adalimumab was separated by SEC followed directly by in-line LC-FPOP of the monomer and aggregate peaks. (A) The comparison of the footprint of (A) heavy chain and (B) light chain peptides of monomeric (blue) and aggregated (orange) adalimumab. Error bars represent one standard deviation from triplicate measurements. Asterisks label peptides that significantly change in oxidation between the monomer and aggregate (p ≤ 0.05). (C) LC-FPOP data are mapped onto a model of adalimumab. The model is colored to represent the change in oxidation in aggregate compared to monomer. Blue: >5x decrease; cyan: <5x decrease; magenta: <5x increase; green: no change (heavy chain); yellow: no change (light chain); grey: peptide not observed.