Purchase this article with an account.
Charlotte Fischer, Viktoria Mans, Nicolas Feltgen, Hans Hoerauf, Christian van Oterendorp; The antiproliferative effect of bevacizumab on human tenon fibroblasts is presumably not through VEGF inhibition.. Invest. Ophthalmol. Vis. Sci. 2016;57(12):2931. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Bevacizumab has previously been shown to reduce proliferation and myofibroblast transdifferentiation of human tenon fibroblasts (hTF). Data from previous publications suggest a direct toxic effect of bevacizumab. We have investigated whether bevacizumab acts on hTF via inhibition of VEGF or rather through a non-antigen dependent effect.
Primary cultures of hFT were treated with bevacizumab (2.5 to 10 mg/ml), ranibizumab (2,5mg/ml) or aflibercept (5mg/ml to 10mg/ml) for 24h. Proliferation was quantified by cell counts and staining for viable and dead cells (LIVE/DEAD ® cell imaging kit, Life Technologies). Immunostainings and western blot were performed for detection of intracellular IgG. Bevacizumab-Antibody and its solvent buffer were separated by membrane filtering (Amicon® Ultra-30k, Merck Millipore).
Bevacizumab significantly inhibited cell proliferation in concentrations of 5mg/ml and 10mg/ml (1,2±0,8 (mean and SEM) and 0,001±0,0005 cells/mm2, respectively), in comparison to untreated controls (32,4±2,3 cells/mm2; p<0,0001), whereas 10mg/ml aflibercept (53,4±4,8 cells/mm2) and 2,5mg/ml ranibizumab (35,4±2,8 cells/mm2) did not decrease proliferation. Staining for dead cells showed a significant increase 24h after incubation with 5mg/ml bevacizumab (21,1±2,5 cells/mm2) compared to untreated cells (4,3±0,7 cells/mm2; p<0,0001), or aflibercept (4,7±2,0 cells/mm2) or ranibizumab (4,6±0,8cells/mm2) treated cells in concentrations bioequivalent to 5mg/ml bevacizumab. Exposure of hTF with either the bevacizumab solvent buffer without protein or the pure antibody in PBS did not inhibit cell proliferation, however, the number of dead cells was increased with the solvent buffer (16,3±3,4 vs 4,6±0,9 cells/mm2; p=0,0085). IgG was detected in intracellular vesicles after incubation with bevacizumab in a dose depended manner. The same was observed with pure bevacizumab antibody in PBS as well as aflibercept but without the antiproliferative and cell death inducing effect.
Bevacizumab, but not aflibercept or ranibizumab at equivalent concentration, inhibits cell proliferation and promotes cell death in hTF. This suggests a non-VEGF-dependent effect of bevacizumab, which might, in part, be mediated by its solvent buffer.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only