Spectroscopic Super-resolution Imaging of DNA Molecules using Intrinsic Contrast

Our research presents the first protocol for measurement intrinsic fluorescent spectra of DNA molecules, enabling investigation of their native optical properties and sequence confirmation-related characteristics. This protocol exploits endogenous fluorescence for DNA spectroscopic imaging, avoiding labeling induced disruption, linkage error, and inconsistent labeled density in existing methods. To begin, dissolve the DNA sample in nuclease-free deionized water to a final concentration of 100 millimolar.

Prepare five microliter aliquots of the DNA stock solution to minimize repeated freeze-thaw cycles. Store the extra tubes containing stock solution at 4 degrees Celsius. Using a laboratory wipe, gently clean a glass slide.

Pipette five microliters of the 100 millimolar DNA solution onto the freshly cleaned glass slide and let the sample dry for one to three hours to form a DNA gel. In a fume hood, prepare either the beta-mercaptoethanol buffer or the DABCO imaging buffer to support stochastic photo switching of DNA molecules. Pipette five microliters of the imaging buffer onto the DNA gel and place a cover slip on top of the sample.

If bubbles are present inside the covered sample, use a tweezer to remove them. Obtain a fabricated nanohole array containing 100 nanometer diameter holes, spaced four micrometers apart, arranged in eight rows and five columns. Turn on the sSMLM system and insert narrow bandwidth filters directly in front of the camera.

Place the calibration sample under a microscope and turn on the microscope lamp to illuminate the sample. Set the exposure time to 200 milliseconds for all filters, but change it to 2, 000 milliseconds for the 750 nanometer filter. Go to acquire and fast time lapse to adjust the number of frames to 10 for each filter.

After calibration, load the DNA gel sample onto the microscope stage with the cover slip side facing the objective and open the camera software for data acquisition. Identify the appropriate Z position using the automatic focus tracking system and adjust the focus until clear stochastic blinking is visible. Ensure that the objective does not touch the cover slip edge to prevent buffer leakage into the immersion oil and interference with tracking.

Acquire around 10, 000 to 30, 000 frames with the desired illumination power and an exposure time of 30 milliseconds. After data acquisition, allow the software to save the data in nd2 format by default. Dispose of the glass slide in an approved waste bin designated for broken glass or sharp items.

Discard all the remaining buffers into the designated chemical liquid waste container according to the institutional safety guidelines. Open Fiji and use the bio formats plugin to load the acquired raw image stack of the DNA sample. Select image followed by stacks, Z Project, and choose minimum intensity as the projection type to generate the minimum projection of the raw image stack.

Select the image calculator and subtract the minimum projection from the raw image stack. Save the background subtracted dataset as a TIF file for spectral analysis. Next, perform localization identification on the zeroth order raw image stack using the ThunderSTORM plugin version 1.3.

Set the threshold to 1.2 times the standard deviation wave. F1, fitting radius to three, and sigma to 1.5 to reconstruct the zeroth order image and extract emission event coordinates. For the cropped zeroth order raw image stack, select plugin followed by ThunderSTORM and run analysis.

Use Fiji and the open source software RainbowSTORM with MATLAB version newer than 2020b to process the calibration data and obtain the relation between horizontal and vertical shifts and wavelengths. Perform localization identification on the calibration images of the five wavelengths using the ThunderSTORM plugin. Set the threshold to five times the standard deviation, wave.

F1, fitting radius to three, and sigma to 1.5 to reconstruct nanohole localizations on the zeroth order and corresponding spectral localizations on the first order. Obtain an average image of each calibration data and adjust the threshold within a range of five to 10 times the standard deviation wave. F1 until 80 localizations are obtained for each calibration image.

Run the RainbowSTORM and click browse under spectral calibration to load the CSV files for each calibration wavelength. Then click save calibration file to store the calibration file. Use customized Python script version 3.11.7 to extract the full spectrum data.

Then set the correct file paths in the Python script for the background subtracted image stack in TIF format from the previous background subtraction step. Set the file path for the localization CSV file produced from the localization analysis step and the calibration file in MAT format generated from the calibration step. Then set the horizontal shift to 925 to 984 pixels for the 532 nanometer data acquired with a 3D DWP corresponding to a spectral window of 525 to 750 nanometers to ensure coverage beyond the 590 to 650 nanometer filter range.

Adjust the vertical shift to 23 pixels for both the 488 nanometer and 532 nanometer data acquired with the 3D DWP based on the spectral calibration curve obtained using RainbowSTORM. Finally, linearize the non-linear spectrum using a third order polynomial fitting for subsequent comparison and classification analysis. The spectral calibration was performed, enabling measurement of the corresponding horizontal and vertical pixel shifts between the zeroth order and the first order images.

This allowed the mapping of a calibration curve between the pixel position and the wavelength. The non-linear and the converted linear emission spectra under 532 nanometer excitation were obtained for GC alter, which refers to a type of double stranded DNA molecule. The green curve represented the average spectrum.

The nonlinear and converted linear emission spectra of A10C10, a type of single stranded DNA molecule, were obtained under 488 nanometer excitation with the blue curve representing the average spectrum. This protocol allows researchers to measure intrinsic fluorescent spectra of synthesized single and double stranded DNA molecules with spectroscopic super resolution imaging. Precisely locating the correct Z position is critical, as DNA gel samples lack structural information and require careful focus to observe the stochastic blinkings.

Future work can extend to measuring intrinsic fluorescence of the RNA and protein, integrating computational models for sequence prediction and label free chromatin imaging.