Correlating Internalization and Potency to Accelerate Antibody Discovery and Development
Abstract #2009—Keystone Symposium on Molecular and Cellular Biology, Antibodies as Drugs (2016) British Columbia, Canada
by C. Langsdorf, R. Aggeler, B. Agnew, D. Beacham, J. Berlier, K. Chambers, A. Chen, A. Dix, N. Dolman, K. Gee, M. Janes, L. Smolenska, W. Zhou; Thermo Fisher Scientific, Eugene, OR, USA - 03/10/16
Abstract and Introduction
The basic cellular mechanisms of internalization and trafficking are important to many areas of cell biology, and especially to the proper function of therapeutic antibodies. Those antibodies intended for use as antibody drug conjugates should specifically bind to target cells and rapidly internalize into acidic compartments. Conversely, antibodies intended to kill cells via direct cell death, complement cascade, or effector cell killing should remain bound to the external surface of target cells as long as possible. However, the ability to study these internalization processes has historically been limited by the lack of tools to directly monitor the internalization and subsequent acidification of extracellular material.
Here we present experimental data demonstrating the use of pH-sensitive fluorophores to monitor the three primary mechanisms of endocytosis in live cells. Additionally we demonstrate three approaches to directly visualize internalization of therapeutic antibodies and ADCs in live cells. Two of these techniques involve site-specific antibody labeling to screen for and further characterize internalizing clones without affecting antibody structure or function. These techniques are demonstrated using quantitative and widefield imaging to study anti-Her2 antibodies and ADCs in breast cancer models. The internalization of ADCs into Her2-positive breast cancer cells shows a strong correlation with target cell killing. We believe the insights provided by these approaches have the potential to accelerate therapeutic antibody lead generation and development.
Figure 1. pHrodo pH sensor dyes. (A) pH response profile of pHrodo Red and pHrodo Green dextrans. (B) pHrodo fluorescence dynamically responds to pH changes. NIH-3T3 cells loaded with 5 µM Invitrogen™ pHrodo™ Red AM intracellular pH indicator were washed with a series of HEPES-based pH standard buffers containing 10 µM nigericin and 10 µM valinomycin to clamp intracellular pH to the indicated values.
Figure 3. Monitoring pinocytosis with fluorescently-labeled dextran. HeLa cells were treated overnight with Invitrogen™ CellLight™ Early Endosome GFP (green fluorescence), then loaded with 10 µg/mL pHrodo Red 10k dextran (red fluorescence) to visualize pinocytosis. pHrodo Red dextran appeared in GFP+ vesicles 5 minutes after loading, indicating both internalization of dextran and acidification of early endosomes.
Figure 4. Fluorogenic indication of EGF endocytosis. A431 cells were labeled with 5 µg/mL of fluorescein- (green fluorescence) or pHrodo Red-conjugated Epidermal Growth Factor (red fluorescence) and Invitrogen™ NucBlue™ Live Cell Stain (blue fluorescence) at 37°C for 30 minutes. (A) FITC fluorescence is seen on the cell surface and throughout the endocytic pathway. (B) pHrodo Red fluorescence is limited to punctate spots, indicating endocytic vesicles with low pH.
Figure 5. Live-cell visualization of phagocytosis. The dynamic pH-response of pHrodo dyes enable visualization of acidification processes such as the phagocytosis of bacteria by macrophages. 100 µg of Invitrogen™ pHrodo™ Red E. coli BioParticles™ conjugate (red fluorescence) were added to MMM cells along with NucBlue Live Cell Stain (blue fluorescence) and imaged after 0, 30, and 60 minutes. The pHrodo BioParticles conjugates dramatically increase in fluorescence upon internalization and acidification.
Figure 7. Conventional antibody labeling strategies. Amine- and thiol-reactive dyes are commonly used to label antibodies. However, the lack of specificity of these bioconjugation reactions can threaten immunoreactivity and lead to poorly defined constructs.
Figure 8. Internalization of amine-conjugated antibodies. Trastuzumab was labeled with amine-reactive pHrodo Red (the degree of labeling was 2.8). Her2+ SK-BR-3 cells were plated in DMEM + 10% FBS in a 35 mm glass-bottom dish. Cells were treated with 10 nM pHrodo Red dye–labeled trastuzumab for 16 hours at 37°C, 5% CO2. Media was removed and replaced with Invitrogen™ Live Cell Imaging Solution supplemented with 1% BSA and NucBlue Live Cell Stain for 30 minutes at 37°C. Images were captured on an EVOS FL Auto imaging system with 60x Olympus Plan Apochromat oil immersion objective and DAPI and TRITC light cubes. Red spots indicate internalization of trastuzumab into acidic vesicles.
Figure 10. Fast, scalable screening of antibody internalization with pHrodo Fab fragments. Any IgG antibody can be non-covalently coupled with fluorescent dyes using the Zenon™ labeling system. Unconjugated primary antibodies are incubated with fluorescently-labeled Fc-specific Fab fragments. This labeling reaction leaves the target antibody intact and free of antigen-binding site obstruction, while providing a consistent degree of labeling between 3 and 5. The Zenon system is useful for labeling small amounts of antibodies (≤20 µg).
Figure 11. Trastuzumab labeled with pHrodo Red dye using Zenon technology is internalized and trafficked to lysosomes. 5 µg aliquots of trastuzumab were labeled with Invitrogen™ Alexa Fluor™ 594 or pHrodo Red human IgG Zenon Fab conjugates. HER2+ SK-BR-3 cells were loaded with 1 µg/mL of fluorescently labeled trastuzumab and 50 nM Invitrogen™ LysoTracker™ Deep Red for 30 minutes at 37°C. (A) Alexa Fluor 594 Zenon–trastuzumab complexes are visible both on the cell surface and where they have trafficked to lysosomes. (B) pHrodo Red Zenon–trastuzumab complexes are brightly fluorescent when internalized to acidic lysosomes but not on the cell surface.
Figure 12. Site-specific labeling of heavy-chain N-linked glycans. IgG antibodies contain two Fc N-glycans that provide an ideal target for conjugation. The Invitrogen™ SiteClick™ labeling system involves three steps: 1, enzymatic removal of terminal galactose residues on heavy chain glycans; 2, enzymatic addition of galactose-azide (GalNAz); 3, the covalent click conjugation of fluorophore-modified dibenzocyclooctynes to the azide-modified sugars. This system allows efficient site-selective attachment of one or multiple fluorescent dyes, radiometal chelators, or small-molecule drugs to antibodies.
Figure 13. SiteClick antibody labeling with pHrodo dye for site-specific labeling and antibody internalization. Trastuzumab was azide-activated with the SiteClick labeling system, then click conjugated with pHrodo Red-DIBO. Cells were treated with 10 nM pHrodo Red–labeled trastuzumab for 16 hours at 37°C, 5% CO2. Media was replaced with Live Cell Imaging Solution with 1% BSA, NucBlue Live Cell Stain, and 50 nM LysoTracker Deep Red for 30 minutes at 37°C. Red spots indicate internalization of trastuzumab into acidic intracellular vesicles that are positive for LysoTracker Deep Red.
Figure 14. Direct Monitoring of ADC internalization in live cells using dual-labeled SiteClick Alexa Fluor 647/MMAE trastuzumab conjugates. (A) Trastuzumab was azide-activated with the SiteClick labeling system, then conjugated with DIBO–Alexa Fluor 647 and DIBO-MMAE (monomethyl auristatin E). HER2+ SK-BR-3 cells and HER2- MDA-MB-231 cells were treated with this fluorescently-labeled antibody-drug conjugate for one hour. (B) Membrane-bound and internalized antibody was visible in SK-BR-3 cells. (C) No signal was observed in MDA-MB-231 cells.
Figure 15. Specific internalization and cell killing response of trastuzumab–Alexa Fluor 647/MMAE conjugates. HER2+ SK-BR-3 cells and HER2- MDA-MB-231 cells in 96 well plates were treated with trastuzumab–Alexa Fluor 647/MMAE conjugate for 72 hours. Cells were then labeled with Invitrogen™ CellEvent™ Caspase-3/7 Green detection reagent and analyzed on a Thermo Scientific™ ArrayScan VTI HCS reader (A) SK-BR-3 cells internalized antibody-drug conjugate, while MDA-MB-231 cells had minimal internalization. (B) Apoptosis was observed in SK-BR-3 cells at higher ADC concentrations, while MDA-MB-231 cells had minimal apoptosis.
Studies of endocytosis, pinocytosis, and phagocytosis benefit from the ability to monitor pH changes in live cells with fluorogenic, pH-sensitive pHrodo dyes. Additionally, the internalization of antibodies can be analyzed using pHrodo dyes attached via three mechanisms.