Test Samples for Optimizing STORM Super-Resolution Microscopy

The overall goal of this procedure is to demonstrate how to acquire high quality storm super resolution images. This is accomplished by first preparing a fluorescently labeled test sample. The second step is to prepare switching buffer and then fill the imaging chamber with the buffer.

Next, the raw data is acquired on a storm microscope. The final step is to reconstruct super resolution images from raw data using rainstorm processing software. Ultimately, the test samples are used to show how to acquire high quality raw data, which can then be processed to generate storm super resolution images in cells with resolutions between 30 and 50 nanometers, representing a five to tenfold resolution enhancement over traditional fluorescence microscopy techniques including turf and confocal.

Generally, individuals new to this method will struggle because they don't realize the importance of collecting lots of high quality sparse links. This is made much easier by using Alexa 6 4 7 dye in the correct switching buffer. A visual demonstration of this technique is essential because acquiring high quality raw data is critical in producing SPR resolution images.

The raw video sequences of this technique look completely different to any other form of fluorescence microscopy. To begin the procedure for coating glass with fluorescent dextrin at 200 microliters of 0.01%polylysine solution to each well of an eight chamber cover glass and incubate for 10 minutes. After 10 minutes, remove the polylysine solution with a pipette to ensure that no liquid remains dilute.

0.25 microliters of a two milligram per milliliter stock solution of dextrin. Alexa 6 47 into 25 microliters of deionized water to create a 20 microgram per milliliter solution to create a high density coating of dextrin. Alexa 6 47 solution.

Dilute 20 microliters of the 20 microgram per milliliter solution into a total of 200 microliters of deionized water. For a medium density solution, dilute two microliters in 200 microliters of deionized water. For a low density solution dilute 0.2 microliters.

In 200 microliters of deionized water, add 200 microliters of the diluted dextrin Alexis X 47 solutions to each well and incubate for 10 minutes. Longer incubation times can be used, which may result in a denser dextrin coating. Finally, remove Dexter Xis X 47 solutions and wash with water three times.

To prepare fluorescent actin filaments on glass first, add 200 microliters of 0.01%polylysine solution to each well of an eight chamber. Cover glass and incubate for 10 minutes. Remove with a pipette to ensure that no liquid remains.

Add to each chamber 90 microliters of general actin buffer previously reconstituted according to the manufacturer's instructions. Then add 10 microliters of 10 micromolar preformed actin filament solution and one microliter foid and Alexa 6 47 and gently pipet up and down to mix incubate for 30 minutes at room temperature to prepare microsphere fluorescent beads as fiducial markers to dilute one microliter of tepec beads into 200 microliters PBS and mix. Add the diluted beads into the imaging chamber and wait for 15 minutes.

After 15 minutes, remove the solution and wash three times with PBS prior to imaging he a cells cultured to approximately 80%co fluency are used in this experiment. Remove the culture medium and wash with PBS. Then add 500 microliters of 4%formaldehyde solution for 10 minutes.

Remove the formaldehyde and wash three times with PBS. Do not allow the cells to remain dry and always use PBS buffered solutions. To avoid hypotonic stress, add two microliters of a 50 microgram per milliliter stock solution of EGF Alexa, 6 47 to 200 microliters of 0.1%BSA solution, and then add this solution to the hela cells.

Incubate for 30 minutes at room temperature. Remove the solution and wash three times with PBS. Add a blocking solution of 0.1%BSA and leave for 15 minutes at room temperature.

After that, remove the blocking solution and wash a further three times with PBS prior to imaging. Just prior to imaging, prepare the switching buffer by mixing the enzyme glucose and reducing agent stock solutions together In a ratio of 50 microliters, 400 microliters and 100 microliters, add PBS to make up the final volume to one milliliter. Remove the PBS from the imaging chamber and then fill to the top with the switching buffer.

Carefully place a cover slip over the top of the chamber, making sure that there are no bubbles at all and the entire chamber is covered. If any bubbles are seen, top up with additional buffer or PBS. Otherwise, buffer performance can decline.

Collecting high quality, sparse linking data is essential for the success of this procedure, so make sure the microscopes properly focused and the correct acquisition settings are used. Locate a fluorescent structure of interest to be imaged. Select the desired angle of illumination.

In this case, a turf or highly inclined angle is recommended as this improves the signal to noise ratio by minimizing out of focus light, which is particularly of benefit for cell samples. Take a reference to fraction limited image of the structure prior to storm imaging. Increase the excitation laser to a high power corresponding to approximately two kilowatts per centimeter squared while imaging with camera settings of 10 millisecond exposure and a cycle time of approximately 50 hertz or frames per second.

Once the flora fours have transitioned into a sparse blinking pattern, collect 10, 000 frames. To begin this reconstruction. Open the RAINSTORM M file from within MATLAB and run it in the rainstorm window.

Select the image file to be analyzed using the browse button. This should be a TIF file. Alter the pixel width to match the raw data of the microscope system used to acquire the data.

Leave the other values to default. Press the process images button and wait until a preliminary image appears. The duration of the processing will depend on the file size computer specification and the number of candidate links within the raw data.

To generate an updated super resolution image, press the open reviewer button and a second window will appear. Press the adjust contrast button in order to change the image display as desired. In some cases where the image is very dark, the maximum value should be decreased.

Use the reviewer window to generate useful image quality metrics and refine the super resolution image further. Leave the first three quality control parameters to default values of 5, 000 0.1 and 0.8 to 3.5 respectively. Update the counts per photon value.

Alter the localization precision cutoff to trade off between optimizing mean localization, precision against localization number. In the final image, values of 30 to 50 nanometers are recommended depending on the quality of the data and the sample. The reconstruction scale factor default is five, which will generate super resolution pixels of 32 nanometers, assuming that the pixel within the original images is 160 nanometers.

If the raw images have a pixel size of 100 nanometers, this will produce super resolution images with pixels of 20 nanometers. Increase the scale factor to produce smaller super resolution pixels. In the final image, press the run reviewer button to generate an updated super resolution image.

Press the view histograms button in order to display image quality metrics. Lastly, press the save image button in the reviewer in order to save all of this data. Begin this procedure by generating an image in the same way as demonstrated previously.

Press the box tracking button in the reviewer and click and drag a box over the fiducial marker on the reviewed image. Wait until a new image appears, which will display the boxed positions. Save this image if desired, if the object of interest is a fiducial marker.

As is the case here, perform drift correction by clicking the set anchor button followed by the subtract drift button. Finally, click run reviewer again to generate a new storm image. A successful coating of dextrin should produce a fluorescent monolayer and typical dextrin data is shown in this figure.

The diffraction limited images show different concentrations of dextrin prior to storm imaging. Super resolution reconstructions have a pixel size of 25 nanometers. 10, 000 images were collected with a 1 28 by 1 28 pixel frame size and individual frames are shown from that sequence.

At a high dextrin concentration. The fluorescent monolayer should appear relatively uniform as illustrated in these images and this video segment. At lower concentrations, the blinks will appear sparser and in focus.

With no obvious background during this image acquisition phase at constant laser illumination, the number of blinks will decrease with time as illustrated by a comparison between frame 2000 and frame 8, 000 in the high density sample. The main difference between the super resolution images of the high, medium and low dextrin concentrations is the decreasing number of localizations. This graph shows an average of three 10, 000 frame sequences of each concentration of dextrin where the error bars indicate the standard deviation.

However, this proportional relationship between the concentration of the fluorescent molecules number of blinks in the raw data and the number of localizations is not a simple linear one as illustrated in this plot of the number of accepted localizations per frame. Using a rolling 100 frame average at very high molecule densities, the software is unable to successfully localize molecules when imaging any sample. A common problem can be the presence of bright but unfocused fluorescent haze across the image caused by fluoro force diffusing through the medium.

These detached fluorescent molecules can be prevented or removed by increasing the number of wash steps with PBS prior to adding the switching buffer or by adding fresh switching buffer to the chamber processing and comparing the data from each image sequence results in very different super resolution images. Panel C and D are super resolution imagery constructions corresponding to the data collected in panels A and B respectively. This graph plots the number of accepted localizations per frame using a rolling 100 frame average.

The red line corresponds to storm sequence A and C where there is a high background, the blue line corresponds to B and D where the background is low. The accepted localization number for the images corresponding to high and low background data is shown in the second graph. These three diffraction limited images show typical data of preformed actin filaments stuck to the surface of the glass prior to addition of switching buffer and storm imaging.

Variable lengths of filaments can be seen. Very bright filaments are often several filaments tangled together. Selecting single relatively bright filaments rather than tangled areas results in better quality images during the acquisition phase.

Bright in focus blinks should be seen along the length of the filament. Sparse blinking should be seen during the acquisition phase and subsequently in the process data. There should be a thin continuous filament in the localizations per frame should show a gradual decline.

Typically, the full with half maximum or FWHM of a filament is given as an empirical estimate of resolution by drawing a straight line through a zoomed in region of the actin filament using the plot profile feature in image J and subsequently performing a Gaussian fit. The FWHM is calculated as 43.2 nanometers. A mis localization problem can occur where a number of actin filaments are branching and or crossing each other by processing subsets of frames and comparing the first and last 5, 000 frames.

A different image is produced in the super resolution image using the first 5, 000 frames. Many localizations are seen in the middle of the image. However, when using the last 5, 000 frames, very few of these localizations are apparent and only the filaments are left albeit somewhat as continuous owing to the low number of localizations in the image.

If a too high blinking density is suspected comparison of the images with the localization per frame data can strongly suggest that this problem is occurring during the first set of frames. There is an average number of localizations per frame of over 10 compared with the last set of frames where it is approximately four. Another potential problem in storm microscopy is drift, which occurs when the sample moves in relation to the objective lens Through the data acquisition phase, although very difficult to detect during the image acquisition phase, drift can be detected in the super resolution images of known structures such as actin filaments.

The first sign that lateral drift may have occurred is that the structure is larger than expected. For example, with a relatively large full with half maximum compared with the precision limit data from rainstorm, in this case 90 nanometers compared with 67 nanometers. A better way to detect drift is by comparing the localizations as a function of time IE, seeing if the localizations in the later frames are displaced compared with those in early frames.

This can be clearly seen in the case of actin filaments when displayed with a color code. Using the box tracking feature in rainstorm as shown in panels B and c localizations are displayed with a color corresponding to the frame number of when they were acquired. For example, localizations from early in the acquisition sequence are blue while localizations from late in acquisition sequence are red.

The displaced colors indicate that drift has occurred in order to measure and correct. For drift fluorescent beads of 100 nanometer diameter can be used as fiducial markers numbers one through four show cropped and zoomed beads individually. In an example where there is relatively severe drift of approximately 100 nanometers over the course of three minutes, 10 seconds during the acquisition phase, the same box tracking feature can be used to color code and confirm the drift occurred as all four beads in this example shown near identical drift, it is possible to select one of them in this case bead two to use as a reference and to subtract the drift from the other beads.

A comparison with panel E shows that the lateral drift has been corrected. Finally, epidermal growth factor stained heela cells can be used to give a realistic example of image resolution. In cells panel A shows a diffraction limited image of part of a hela cell focused at the basal cell surface where the yellow boxes indicate zoomed in regions of interest shown in panels B and C in contrast to the indistinct zoomed in regions of interest in the diffraction limited image.

The super resolution images should show a mixture of clusters occasional isolated single pixels. The clusters will be approximately 100 nanometers in diameter and are likely to correspond to forming pits and vesicles. The pathway via which EGF is predominantly downregulated and endocytosed localizations per frame data from Panel D are presented in graph G.After watching this video, you should have a good understanding of how to make some simple test samples, acquire data and reconstruct high quality storm, see resolution images.

These methods will help avoid some common fit pitfalls such as drift, and help optimize your storm. See resolution experiments.