Development of in silico models to predict viscosity and mouse clearance using a comprehensive analytical data set collected on 83 scaffold-consistent monoclonal antibodies.

Biologic drug discovery pipelines are designed to deliver protein therapeutics that have exquisite functional potency and selectivity while also manifesting biophysical characteristics suitable for manufacturing, storage, and convenient administration to patients. The ability to use computational methods to predict biophysical properties from protein sequence, potentially in combination with high throughput assays, could decrease timelines and increase the success rates for therapeutic developability engineering by eliminating lengthy and expensive cycles of recombinant protein production and testing. To support development of high-quality predictive models for antibody developability, we designed a sequence-diverse panel of 83 effector functionless IgG1 antibodies displaying a range of biophysical properties, produced and formulated each protein under standard platform conditions, and collected a comprehensive package of analytical data, including in vitro assays and in vivo mouse pharmacokinetics. We used this robust training data set to build machine learning classifier models that can predict complex protein behavior from these data and features derived from predicted and/or experimental structures. Our models predict with 87% accuracy whether viscosity at 150 mg/mL is above or below a threshold of 15 centipoise (cP) and with 75% accuracy whether the area under the plasma drug concentration-time curve (AUC0-672 h) in normal mouse is above or below a threshold of 3.9 × 106 h x ng/mL.

Prediction of the Hydrogen Peroxide-Induced Methionine Oxidation Propensity in Monoclonal Antibodies.

Methionine oxidation in therapeutic antibodies can impact the product's stability, clinical efficacy, and safety and hence it is desirable to address the methionine oxidation liability during antibody discovery and development phase. Although the current experimental approaches can identify the oxidation-labile methionine residues, their application is limited mostly to the development phase. We demonstrate an in silico method that can be used to predict oxidation-labile residues based solely on the antibody sequence and structure information. Since antibody sequence information is available in the discovery phase, the in silico method can be applied very early on to identify the oxidation-labile methionine residues and subsequently address the oxidation liability. We believe that the in silico method for methionine oxidation liability assessment can aid in antibody discovery and development phase to address the liability in a more rational way.

Phosphoprotein isotope-coded affinity tags: application to the enrichment and identification of low-abundance phosphoproteins.

The use of a phosphoprotein isotope-coded affinity tag (PhIAT), which employs differential isotopic labeling and biotinylation, has been shown capable of enriching and identifying mixtures of low-abundance phosphopeptides. A denatured solution of beta-casein was labeled using the PhIAT method, and after proteolytic digestion, the labeled peptides were isolated using immobilized avidin. The recovered peptides were separated by capillary reversed-phase liquid chromatography and identified by tandem mass spectrometry. PhIAT-labeled peptides corresponding to known O-phosphorylated peptides from beta-casein were identified along with the phosphorylated peptides from alphas1-casein and alphas2-casein, known low-level (<5%) contaminants of commercially available beta-casein. All of the casein-phosphorylated residues identified by the present PhIAT approach correspond to previously documented sites of phosphorylation. The results illustrate the efficacy of the PhIAT-labeling strategy to not only enrich mixtures for phosphopeptides but also, more importantly, permit the detection and identification of low-level phosphopeptides. In addition, the differences in the phosphorylation state could be determined between phosphopeptides in comparative samples by stoichiometric conversion using the light and heavy isotopic versions of the PhIAT reagents. Overall, our results exemplify the application of the PhIAT approach and demonstrate its utility for proteome-wide phosphoprotein identification and quantitation.

Toward a human blood serum proteome: analysis by multidimensional separation coupled with mass spectrometry.

Blood serum is a complex body fluid that contains various proteins ranging in concentration over at least 9 orders of magnitude. Using a combination of mass spectrometry technologies with improvements in sample preparation, we have performed a proteomic analysis with submilliliter quantities of serum and increased the measurable concentration range for proteins in blood serum beyond previous reports. We have detected 490 proteins in serum by on-line reversed-phase microcapillary liquid chromatography coupled with ion trap mass spectrometry. To perform this analysis, immunoglobulins were removed from serum using protein A/G, and the remaining proteins were digested with trypsin. Resulting peptides were separated by strong cation exchange chromatography into distinct fractions prior to analysis. This separation resulted in a 3-5-fold increase in the number of proteins detected in an individual serum sample. With this increase in the number of proteins identified we have detected some lower abundance serum proteins (ng/ml range) including human growth hormone, interleukin-12, and prostate-specific antigen. We also used SEQUEST to compare different protein databases with and without filtering. This comparison is plotted to allow for a quick visual assessment of different databases as a subjective measure of analytical quality. With this study, we have performed the most extensive analysis of serum proteins to date and laid the foundation for future refinements in the identification of novel protein biomarkers of disease.