LU LIU,1 THEODORE W. RANDOLPH,2 JOHN F. CARPENTER1
1Department of Pharmaceutical Sciences, University of Colorado, Denver
2Department of Chemical and Biological Engineering, University of Colorado, Boulder
Posted with permission from Lu Liu and John Carpenter,
Purpose:
To investigate: 1) if syringe filters used to remove particles from protein solutions actually shed particle; 2) if pre-flushing filter with buffer reduces particle shedding into solution during filtration; 3) the effects of particles on agitation-induced protein aggregation.
Methods:
Various types of syringe filters (0.2µm) were tested without or with pre-flushing using sodium nitrate buffer (10mM, pH6.2). Particle counts and size distribution (>1µm) in buffer alone or in solutions of keratinocyte growth factor 2 (KGF2, 0.5 mg.ml) were determined with a ProteinSimple Micro-Flow Imaging (MFI) instrument. Protein samples filtered with a polyether sulfone (PES) membrane were incubated with and without agitation for one day at room temperature. To determine protein loss during filtration or incubation, the concentration of soluble KGF2 was measured by UV and bicinchoninic acid protein assay (BCA).
Results:
For all three different membrane materials, i.e., PES, polyvinylidene difluoride (PVDF), and cellulose acetate (CA), syringe filters with multi-layers of borosilicate glass microfibers on the top of membrane shed up to tens of thousands of particles when filtering with buffer or KGF-2 solution, either without or with pre-flushing. Also, 20-80% of KGF2 was lost during filtration due to KGF-2 binding to glass microfibers and membranes. In contrast, filters with membranes alone showed 0-20% loss of KGF-2, and an equal or a slightly decreased particle number. An exception was on an PES filter which shed particles. During incubation of KGF-2, all filtered and agitated samples showed significantly more particles than unfiltered control sample. There was no substantial difference among non-agitated samples. There was up to 20% loss of KGF2 for filtered samples during agitation, whereas there was no detectable loss for non-filtered samples.
Conclusions:
Particles shedding from syringe filters varied greatly among the filters types from the three manufacturers. Furthermore, particles shed from the filters stimulate protein aggregation during agitation. It is important to check if syringe filters or other filtration devices shed particle during filtration of protein solutions because these particle may impact stability of the protein in filtered solutions.
PRESENTED AT:
2010 AAPS National Biotechnology Conference, San Francisco CA, USA
2010 AAPS NBC_UofC_Effect of syringe filter particles on aggitation induced protein aggregation.pdf
2010 AAPS NBC UofC Effect of syringe filter