Insulin is a peptide hormone which plays a significant role in metabolism.
In its monomer form, insulin can partially misfold, causing a cascade of non-native cross assembly of beta-pleated sheets or amyloid-like fibrils. For the insulin analog manufacturer, this is especially problematic, especially since insulin analogs can have a higher propensity for aggregation than native insulin.
As such, when a new formulation or delivery device is used, insulin fibrillation, commonly known as high molecular weight proteins (HMWP), should be assessed. A common method for analyzing insulin HMWP is size exclusion chromatography (SEC). The European Pharmacopoeia (Ph. Eur.) prescribes a 300 x 7.5 mm hydrophilic silica column. However, the method requires a lengthy 35-minute run time, which is time prohibitive for any laboratory with many samples to run and data points to generate.
This application will show the method transfer of the Ph. Eur. method from a traditional 10 µm hydrophilic silica 300 x 7.8 mm column, to a high performance 1.8 µm hydrophilic silica 150 x 4.6 mm column.
Human insulin solution was prepared as per the Ph. Eur. and the sample was diluted in 0.01 M hydrochloric acid to a concentration of approximately 4 mg/mL. Insulin analog was prepared by stressing at 37 °C and diluted 0.01 M hydrochloric acid to ~4 mg/mL prior to injection.
HPLC was performed using a 10 µm hydrophilic silica 300 x 7.8 mm column as prescribed by the Ph. Eur. and a Yarra™ 1.8 µm SEC-X150 150 x 4.6 mm column. Both columns were run on a Waters® ACQUITY® UPLC® Instrument (Waters Corporation, Milford, MA). Running conditions were as per the method conditions listed below.
Figures 1 and 2 shows the chromatograms for system suitability with Ph. Eur. insulin standards. Monomer and dimer for the 10 µm column have a resolution of 2.15, which meets the system suitability requirement of ≥ 2.0. However, with the Yarra 1.8 µm SEC-X150 column, monomer and dimer have a resolution of 2.86, which far exceeds the resolution requirement of the monograph. Peak areas for dimer and HMWP for both columns are within the < 1 % system suitability requirement. Note total analysis time for system suitability is approximately 12 minutes for Yarra 1.8 µm SEC-X150 column and 35 minutes for the traditional 10 µm hydrophilic silica column.
Data for system suitability is summarized in Table 1. Note the higher efficiency (Plate Count, Ph. Eur.) for all peaks using the Yarra™ 1.8 µm SEC-X150 column. This higher efficiency not only gives better resolution for monomer and dimer, but also allows for the resolution calculation for HMWP and dimer.
Figures 3 and 4 show the degraded insulin analog sample. At 1.96, resolution is acceptable between monomer and dimer for the 10 µm column. This is in comparison to 2.06 for Yarra 1.8 µm SEC-X150 column. Sum of areas of HMWP and analog exceeds 1 % for both columns, indicating the sample would not pass the test limits of <1 % for insulin HMWP. Again, note the difference in total run time between the two methods.
Data is summarized in Table 2. Improvements are seen in efficiency for monomer and dimer peaks in the Yarra 1.8 µm SEC-X150 column method, again lending to the increased resolution seen with the updated method.
Find the full technical note here.
Yang, Y., A. Petkova, K. Huang, B. Xu, Q.-X. Hua, I-J. Ye, Y.-C. Chu, S.-Q. Hu, N. B. Phillips, J. Whittaker, F. Ismail-Beigi, R. B. Mackin, P. G. Katsoyannis, R. Tycko, and M. A. Weiss. “An Achilles’ Heel in an Amyloidogenic Protein and Its Repair: INSULIN FIBRILLATION AND THERAPEUTIC DESIGN.” Journal of Biological Chemistry 2010: 10806-0821. Print.
Librizzi, Fabio, and Christian Rischel. “The Kinetic Behavior of Insulin Fibrillation Is Determined by Heterogeneous Nucleation Pathways.” Protein Science : A Publication of the Protein Society 14.12 (2005): 3129–3134. PMC. Web. 5 Oct. 2015.
European Pharmacopeia 8.6, Human Insulin Monograph, 2015:5299-5301