Single-Molecule Localization Microscopy of Membrane Proteins using Single-Antibody Labeling

My research focuses on the membrane of T cells and aims to reveal new nanoscale insights relevant to therapeutic targeting. Existing imaging methods do not typically directly visualize nanoscale antibody-epitope interactions in cells. This protocol overcomes that limitation by enabling direct observation within the cellular environment.

After mounting the sample on the microscope and depositing gold colloids onto the surface, use bright field illumination to verify that enough colloids have been deposited within the region of interest. Confirm that the ideal density of 5 to 10 gold colloids is present within the region of interest before proceeding. For the optical configuration of the antibody probe fluorophore, open the imaging software and access the optical configuration settings.

Select the appropriate dichroic mirror and set the laser excitation wavelength and power density. Adjust the camera integration time and select the emission filter. Then, set the camera pixel binning to achieve a pixel size of 150 to 160 nanometers.

Now, create an optical configuration for the photo bleaching step by selecting a photo bleaching mode in the imaging software. Set the laser power to 100%and adjust the camera integration time to one second. Set the imaging parameters in the image acquisition software by defining the non-illuminating interval, number of frames, and frequency of the photo bleaching step.

Next, prepare the imaging buffer by diluting the CD81 antibody to a concentration of 1 microgram per milliliter in 200 microliters of 5%BSA made in DPBS. The final recommended antibody concentration is 0.5 nanomolar for Jurkat T cells, or 2 nanomolar for U-2 OS cells. Add the prepared imaging buffer to the well containing the cells and start the image acquisition in the software to begin data collection.

Use the live view function in the imaging software configured for membrane single-antibody labeling to monitor antibody binding in real time. Configure the imaging parameters in the image acquisition software to obtain at least a 10-frame acquisition and set the non-illuminating interval to five seconds. Start the acquisition and evaluate the sparsity of single molecule localizations in the recorded movie.

If multiple overlapping antibody binding events are observed immediately after adding the imaging buffer, or if single-molecule localizations accumulate over time, abort the acquisition. Reduce the antibody concentration by twofold and repeat the acquisition on a new sample. Continue reducing the antibody concentration by twofold and reevaluating the number of events per unit area until the desired single-molecule localization density is achieved.

If the event density is sparse, increase the antibody concentration in the imaging buffer by twofold. Repeat the acquisition on a new sample and continue increasing the antibody concentration by twofold while reevaluating the number of binding events per unit area until the desired single-molecule localization density is achieved. For non-illuminating interval optimization, configure the imaging parameters in the image acquisition software to obtain a 50-frame acquisition.

After adding the antibody and starting acquisition, evaluate the density and sparsity of single-molecule localizations in the recorded movie. Determine the lowest non-illuminating interval that yields an optimal single-molecule event density. Finally, if spatially overlapping binding events occur as the image acquisition progresses, introduce a photo bleaching step at regular intervals.

Adjust the photo bleaching frequency so that fluorescence from bound antibodies is removed before overlapping events accumulate. Set the file path in the MATLAB software to the folder that contains the reconstructed M cell comma-separated values file. Run the code by clicking Run in the MATLAB toolbar.

When prompted, select the M cell CSV file and click Open to load it into the program. Run the analysis using the previously defined inputs as a starting point. Evaluating the scatter plot containing the unclustered localization data.

Choose a minimum point value that is higher than the number of noise localizations but lower than the number of localizations on the cell. Evaluate the window that depicts the clustered data. Confirm that clusters are retained on the cell with the expected phenotype while localizations outside the cell are minimized.

If the clustering parameters are too restrictive, decrease the minimum points value and increase epsilon. If the clustering parameters are too inclusive, increase the minimum points value and decrease epsilon. Finally, after adjusting the minimum points value and epsilon, rerun the code to obtain optimal clustering.

Jurkat T cells were immobilized with sufficient spacing to preserve membrane integrity and allow antibody access to membrane epitopes. Gold colloids were visualized and used as fiducial markers for lateral drift correction during M cell image acquisition. Optimization of the non-illumination interval showed that a five-second interval produced antibody binding events with minimal spatial overlap compared to shorter or longer intervals at one nanomolar antibody concentration.

Application of drift correction improved the reconstructed M cell image compared to the uncorrected reconstruction. Density-based clustering of M cell localizations reduced low-density background interactions and highlighted clustered CD81 binding events. We can observe antibodies interacting with their membrane epitopes in a cellular environment, providing insight relevant to therapeutic antibodies.

The antibody concentration, non-illumination interval, and photo bleaching steps are critical to optimize. Suboptimal parameters will compromise the results. Single-antibody labeling can be extended to simultaneously evaluate membrane protein binding by multiple antibodies within the same cellular environment.