Sample preparation for dSTORM
1. Grow E.coli cells to mid-log phase (O.D. 600 ~ 0.6) in LB media (Miller) at 37°C with shaking (200 rpm) in an Erlenmeyer flask.
Note - At Mid-log phase, protein synthesis and trafficking is at maximum to compensate the rapid cell growth.
2. Collect aliquots (500 µL) of cells by centrifugation (4000 x g, 5 min, 4°C) and then wash twice in PBS before being resuspended in 500 µL of PBS. Coat the cover glass bottom chambers of 8-well plates (Sarstedt®) with 0.01% (v/v) poly-L-lysine (Sigma-Aldrich®) for 10 min at room temperature and remove excess poly-L-lysine.
Note - It is necessary to optimise the concentration of poly-L-lysine, as higher concentrations would result in multilayers of bacteria, which hinders Total internal reflection fluorescence (TIRF) imaging.
3. Afterwards, immobilize 200 µL of E. coli cells in to each coated well chamber. To ensure the formation of a monolayer of bacteria, centrifuge the chamber slides (4000 x g, 3 min, 4 ° C) and remove the excess cell suspension followed by three brief washes with PBS.
Note - Ensuring the formation of a monolayer of bacteria is critical to perform TIRF imaging.
4. Fix the monolayer of bacteria with a mixture of paraformaldehyde (2% w/v) and glutaraldehyde (0.2% v/v) in PBS for 5 min at 4°C followed by PBS washing steps to remove excess fixatives.
5. To minimise auto-fluorescence caused by Schiff's bases in glutaraldehyde, treat the samples with freshly prepared 0.1% (w/v) NaBH4 dissolved in PBS for 15 min, followed by two washing steps with PBS.
6. Where applicable the samples can be permeabilized either with Digitonin (0.001% or 0.01% w/v in PBS) or TritonX-100 (0.001% or 0.01% v/v in PBS) followed by three washing steps with PBS.
Note - Depending on the localization of a particular protein in the cell, reagents for permeabilization may differ. For example, antibody accessibility to inner membrane environment can be increased by using TritonX-100, a much harsher detergent than digitonin.
Immunofluorescent labelling of E.coli membrane proteins
1. The specificity of an antibody should be tested with a negative control.
Note - In this instant, we have used a monoclonal antibody raised against E.coli membrane protein BamC. The absence of fluorescent signal on ΔbamC E.coli mutants confirms the specificity of anti-BamC antibody (Figure 1).
2. Blocking of non-specific binding sites is achieved by incubation in 5% w/v BSA in PBS for 1 h at room temperature, followed by incubation of antisera diluted to 1:1000 in 5% w/v BSA in PBS, for 1 h mixing by rotary inversion at room temperature.
3. Remove excess ubound primary antibodies by washing the samples thrice with PBS followed by addition of secondary anti-mouse IgG-Alexa 647 (ThermoFisher®) conjugated antibody (diluted 1:1000 in 5% BSA in PBS) for 45 min at room temperature. Follow the same washing steps to remove excess unbound secondary antibody. Store the samples at 4°C in PBS prior to microscopy imaging.
Note - Prior to imaging, a post fixation step can be carried out (with paraformaldehyde, 2% w/v) which ensures the preservation of antibodies in the sample for a long period of time.
4. Prior to imaging add fiducial markers (TetraSpeck™ Microspheres, 0.1 µm ; ThermoFisher T7279) to the sample.
dSTORM image acquisition and processing
1. Collect super-resolution images for Alexa-647 labelled samples by illuminating continuously with 638 nm laser (1-5% laser power) at an appropriate TIRF angle.
Note - When performing dSTROM imaging, take necessary steps to adhere to laser safety guidelines enlisted in your imaging facility.
2. Drive dyes into the dark state (this takes about 30-60 s for Alexa-647, under 80-90% laser power), and then single molecule blinking time series can be captured. This can be under 10,000-20,000 frames at an exposure time of 20 ms and an electron multiplying gain of 50. Raw image pixel size with 100x objective and 1.6x magnifier engaged is 100 nm x 100 nm.
Note - Laser illumination parameters may change depending on the miscrocopy system and the type of fluorophore in use. Therefore it is paramount to optimise these parameters before acquiring images.
3. To reconstruct the series of to super-resolved images, an out-put pixel size of
10 nm is recommended. For this we have used the open-source software rapidSTORM version 3.3.1 (Wolter et al., 2012).
Note - You can also use other localization microscopy evaluation programs such as ThunderSTORM (https://zitmen.github.io/thunderstorm/) to achieve the desired resolution output.
4. Here, we have set the full width of half maximum (FWHM) of the fitting point spread function to 350 nm and fitted point spread functions (PSF) with a local signal-to-noise ratio (SNR) < 120. Images are first colour coded for temporal appearance of blinks to detect sample drift, then (if applicable) drift correction is acheived by using the linear drift correction available in rapidSTORM. If fiducial markers are used, use the fiducial based drift correction option in rapidSTORM. Finally, export the file as 8-bit greyscale images for further image processing.
5. Extract the corresponding localisations files for the single-molecule images from rapidSTORM software listing all detected localisations in x,y coordinates and use it for quantification and data analysis.
Cluster analysis of Super resolution images
Cluster analysis procedure can be found in the following link at Git repository; https://github.com/monashmicroimaging/autoclustr.