Category: Automation and High-Throughput Technologies
Routine characterization of N-glycan profiles is critical during manufacturing of biotherapeutic proteins. Traditional labeling methods such as 2-aminobenzamide (2-AB), procainamide and Instant AB limit sample throughput and detection as a result of lengthy preparations and weak mass spectrometric (MS) detection capabilities. Recently, the GlycoWorks N-glycan analysis kit using RapiFluor-MS (RFMS) was introduced and provides a solution to these problems through rapid labeling in conjunction with enhanced fluorescence (FLR) and MS signaling. We have developed a validated automation solution for this N-glycan labeling method using an Andrew Alliance pipetting robot. The aforementioned robotic solution promises further gains in data reproducibility while eliminating the need for tedious manual pipetting. Using a set of standard manual pipettors and an optically guided arm the entire N-glycan labeling protocol was automated and optimized. In this poster, the N-glycan total area counts for three proteins with unique N-glycan profiles were prepared using the RFMS labeling procedure. These samples were then analyzed via hydrophilic interaction liquid chromatography (HILIC) couple to FLR detection, over a range of denaturation and de-N-glycosylation temperatures. Denaturation and de-N-glycosylation of samples normally occurs at 90°C and 50°C in pre-heated blocks when manually performing the method, after which they are immediately placed at room temperature. The robot is unable to remove sample tubes from the custom Peltier effect heating device, requiring that samples be exposed to a range of temperatures over a much longer period of time. The optimal temperatures found when analyzing these proteins using the Peltier device were 75°C and 55°C respectively. Following re-optimization of the two heating steps, the automated solution was validated in comparison to the manual preparation method using a murine immunoglobulin G1 (IgG1) monoclonal antibody (mAb). N-glycan yields (total area) and relative abundance (% area) were monitored over multiple automated preparations and found to be comparable to the manual method. Moreover, N-glycan total and % areas were reproducible between preparations, showing average relative standard deviations (RSDs) of 12.08% for total area and 4.75% for % area. The quality control (QC) automation method provided exceptional pipetting accuracy and results comparable to the manual method in terms of N-glycan yield. Overall, the N-glycan preparation solution proposed here can relieve researchers of hours of monotonous every day pipetting while providing a robust and reliable method for the monitoring of N-glycans.
Jennifer Fournier– Director, Product Marketing Chemistry, Waters Corporation, Milford, MA
Director, Product Marketing Chemistry
I spent 4 years at the University of Massachusetts Amherst for my undergraduate degree in Biochemistry and Molecular Biology. During my time as an undergraduate I was fortunate enough to receive several internship opportunities, the first of which found me at the University of Connecticut Health Center working in Dr. Suzy Torti's lab in the Molecular Biology and Biophysics department. My second internship was in Dr. Jennifer Ross's lab at UMass where I studied the biophysics of cellular protein movements. I was placed in my third internship at Waters Corporation (where I currently work) through the Integrated Concentration in Science (iCons) program at UMass. Following a 1 year Masters program in Molecular and Cellular Biology (in Dr. Ross's lab) at UMass Amherst I accepted a full time position at Waters Corporation and began working on the automation of our rapid N-glycan labeling kit with the Andrew Alliance automated pipetting robot.