Novel Approaches to Measure Receptor-Mediated Endocytosis of Herceptin

In this series "From Image to Results", explore various case studies explaining how to reach results from your demanding samples and acquired images in an efficient way. For each case study, we highlight different samples, imaging systems, and research questions.

In this case study, we sought to develop a proof-of-concept assay that could quantify Herceptin plasma-membrane binding, internalization, and trafficking to late endosomes.

In vitro characterization of complex medicines

In recent years, the focus of medical discovery has shifted from traditional small molecule approaches to complex therapies such as nanomedicines and biologics. Prior to application in vivo, the molecular dynamics and mechanism of these novel therapies must be thoroughly characterized in vitro. This not only helps to ensure efficacy and safety, but also advances the predictive capability of in vitro models and provides an avenue for the rational design of new medicines. In this study, Medicines Discovery Catapult (MDC), a non-profit medicines discovery organization in the United Kingdom, and ZEISS have developed an approach that combines genome engineering, advanced microscopy, and computational image analysis to quantify the plasma membrane binding, internalization and trafficking of Herceptin, an antibody-based complex medicine. We note that this approach is broadly applicable to other classes of complex medicines, such as silica nanoparticles, lipid nanoparticles, nucleic acids, and antibody-drug conjugates.

Antibody-based therapeutics have made valuable contributions to the treatment of cancer, autoimmune conditions, and aging-related disorders. Key to antibody-based approaches is the existence of a plasma membrane-bound or extracellular protein target whose function is critical to disease etiology.

Antibodies exert several effects when targeted to plasma membrane-bound receptors. In addition to directly blocking the receptor’s natural ligand binding and preventing activation of the receptor’s signaling cascade, receptor-bound antibodies can also provide an epitope for subsequent binding of immune cells and stimulate the Antibody-Dependent Cell-mediated Cytotoxicity (ADCC), whereby the target cell (e.g., a cancer cell) is destroyed. Antibody-bound receptors can also be internalized from the cell’s surface by receptor-mediated endocytosis and then trafficked into the endolysosomal network. The latter process is key for a group of complex medicines called antibody-drug conjugates (ADC), which exploit the specificity of antibody-epitope binding to deliver a therapeutic payload intracellularly.
 

For sample preparation, BT-474 GFP-Rab7a cells were seeded on 8-well chamber slides (ibidi 80826) and grown for at least 24 hours.

For RNAi, 10 pmol of Her2 or control siRNA (Thermo Fisher) was added to 50 µl OptiMEM containing 0.3 µl Lipofectamine RNAiMAX (Invitrogen), incubated at room temperature for 20 minutes then added to cells for 48 hours.

For priming with Trastuzumab-AF647 (TZ-AF647; Novus Biologicals), cells were washed with cold serum-free DMEM and incubated with 10 µg/ml TZ-AF647 in serum-free DMEM at room temperature for 15 minutes. As controls, 10 µg/ml Transferrin-AF647 (Jackson ImmunoResearch) or 20 µg/ml of AF647-labelled goat anti-human secondary antibody (Invitrogen A-21445) was used in place of TZ-AF647. Clathrin-mediated endocytosis was inhibited with 30 µM Pitstop 2 (Abcam ab120687) for 30 minutes prior to addition of TZ-AF647. After antibody incubation, cells were washed once and cultured with warm complete DMEM.

Imaging was performed using Airyscan in confocal mode (increased sensitivity to detect weak signals) on the ZEISS LSM 880 confocal microscope equipped with a Plan-Apochromat 20x 0.8 NA objective, humidified chamber and incubator that maintained 37°C and 5% CO2. For live imaging, Definite Focus 2 was used to avoid focus drift. A single plane was acquired every 7 minutes for 4 hours. For spheroid imaging, Z-stacks were acquired on the LSM 880 confocal system described above with optimized Nyquist sectioning (0.5 µm for high-resolution image data).

Quantifying Trastuzumab-AF647 Association with Late Endosomes

To establish an optimal read-out for TZ-AF647 association with late endosomes, we treated endogenous eGFP-Rab7a expressing BT-474 cells with TZ-AF647 or Transferrin-AF647 for 30min prior to fixation. To confirm that TZ-AF647 and Transferrin-AF647 were internalized by endocytosis, we incubated cells with Pitstop 2, an inhibitor of clathrin-mediated endocytosis (Figure 6). We then sought to develop an automated image analysis pipeline to distinguish plasma membrane-bound from intracellular TZ-AF647 and then quantify the colocalization of TZ-AF647 and GFP-Rab7a.

In this case study we present a method to computationally quantify antibody internalization and trafficking to late endosomes from confocal microscopy images. We have applied this to real-time live imaging and to 3D cancer spheroids, demonstrating its general feasibility and the appropriateness of the ZEISS LSM 880 confocal microscope and ZEISS arivis Pro software for such experiments. If you want to reproduce the analysis or use this use case to learn more about arivis Pro, we have prepared some data sets and associated analysis pipeline for download below.

To finalize, we present some discussion points on how to further this experimental workflow.

A critical aspect of an antibody internalization assay is correctly identifying plasma membrane-bound vs. endosomal antibodies. Creating cells stably expressing endogenously-tagged GFP-Rab7a was key to obtain a reliable quantitative read-out of endocytosed and trafficked Trastuzumab. This could be furthered by multiplexing with markers that label the plasma membrane and additional endocytic compartments, thus enabling simpler plasma membrane-bound segmentation and increasing the scope of trafficking analysis. Furthermore, the different morphological appearance of plasma membrane-bound vs. endosomal antibody makes it a good candidate for a Deep Learning segmentation approach.

We also were able to quantify antibody internalization in cancer spheroids. This is noteworthy as 3D model systems like cancer spheroids can yield valuable data that directs in vivo translation, a major bottleneck in drug development. Moreover, many spheroids can be grown as co-cultures and therefore allow the study of processes involving cell-cell interactions. A drawback for imaging spheroids is the limited depth of imaging. Here, with the ZEISS LSM 880 confocal microscope, we were able to reach a depth of ~30 µm with sufficient resolution. If a greater imaging depth was required, we advise research to investigate using non-linear optics (also known as multiphoton), which is available with the ZEISS LSM 980 NLO confocal system.