A hydrophilic gel matrix for single-molecule super-resolution microscopy

Background: Novel microscopic techniques which bypass the resolution limit in light microscopy are becoming routinely established today. The higher spatial resolution of super-resolution microscopy techniques demands for
Background: Novel microscopic techniques which bypass the resolution limit in light microscopy are becoming routinely established today. The higher spatial resolution of super-resolution microscopy techniques demands for precise correction of drift, spectral and spatial offset of images recorded at different axial planes.
Methods: We employ a hydrophilic gel matrix for super-resolution microscopy of cellular structures. The matrix allows distributing fiducial markers in 3D, and using these for drift correction and multi-channel registration. We demonstrate single-molecule super-resolution microscopy with photoswitchable fluorophores at different axial planes. We calculate a correction matrix for each spectral channel, correct for drift, spectral and spatial offset in 3D.
Results and discussion: We demonstrate single-molecule super-resolution microscopy with photoswitchable fluorophores in a hydrophilic gel matrix. We distribute multi-color fiducial markers in the gel matrix and correct for drift and register multiple imaging channels. We perform two-color super-resolution imaging of click-labeled DNA and histone H2B in different axial planes, and demonstrate the quality of drift correction and channel registration quantitatively. This approach delivers robust microscopic data which is a prerequisite for data interpretation.
show moreshow less

Download full text files

  • Additional file 1: Figure S1: Sample preparation and data processing. (a) Fiducial markers are added to the hydrophilic gel matrix (extracellular matrix (ECM)) and applied to the specimen. (b) Drift correction and image registration workflow.

  • Additional file 2: Figure S2: Non-linear drift correction in 2D. We demonstrate the robustness of drift correction with a sample containing 13 fiducial markers which exhibit non-linear drift. (a) Coordinates (before (black) and after (red) drift correction) and calculated centre (triangle, before (blue) and after (magenta) drift correction) of a single fiducial marker. (b) Trajectories (coordinates of fiducial markers over time) were split in x and y and (c) convolved with a Gaussian low pass filter to eliminate high frequencies. (d) Smoothed trajectories of 12 fiducial markers were used to calculate the transfer matrix and applied to correct for the non-linear drift of the selected fiducial marker shown in (a) (the fiducial marker in (a) was not used to generate the transfer matrix). This procedure was repeated for all 13 fiducial markers. (e) Distance distribution (left) from all fiducial marker coordinates to its center shown in (a) before (black) and after (red) drift correction. The peak of the distance distribution, which is a measure of the quality of drift control, moved from 40.1 nm (s.d., uncorrected) to 6.2 nm (corrected) (respective values for each dimension: 23.5 nm (before) and 4.5 nm (after drift correction) in x-direction, 32.5 nm and 4.2 nm in y-direction). A further and different measure for the quality of drift correction is the covariance matrix (right, distribution of covariances for x (dashed line) and y (straight line) for both uncorrected (black) and corrected (red) coordinates of all fiducial markers). The average covariance decreases from 360.7 nm2 (x, uncorrected) and 698 nm2 (y, uncorrected) to 31.2 nm2 and 24.8 nm2 (corrected).

  • Additional file 3: Figure S3: Stability of fiducial markers in gel matrix. The stability of fiducial markers (N = 31) in the gel matrix was measured over 10.000 frames. The distance variation of a single fiducial marker with respect to the centre of gravity of all other fiducial markers (i.e. 30) over time was calculated. This procedure was repeated for each marker and histogrammed over time, which results in the distributions shown for x (top), y (middle) and z (bottom) (the centre position was set to zero for better comparison). The histograms (right) show the distance variation summed over all frames. The histograms were approximated with a Gaussian function and yielded an s.d. of 3.5 nm (x), 4.9 nm (y) and 11.3 nm (z).

  • Additional file 4: Figure S4: Drift correction of single fiducial markers. Coordinates of four fiducial markers obtained from one super-resolution image (x (red), y (blue) and z (green)) are shown before (left panel) and after (right panel) drift correction. Drift in x and y was corrected as described, drift in z was corrected by hardware stabilization (for details, see Materials and Methods).

  • Additional file 5: Figure S5: Super-resolution imaging of DNA and histone H2B. Replicated DNA (EdU-Alexa Fluor 647, magenta) and H2B (mEos2-H2B, green) were imaged (a) at the bottom and (b) at the middle of a HeLa cell. Drift was not corrected here (see Figure 4 for drift corrected images) (scale bar 5 μm).

Export metadata

  • Export Bibtex
  • Export RIS
Author:Patrick J. M. Zessin, Carmen L. Krüger, Sebastian Malkusch, Ulrike Endesfelder, Mike Heilemann
Parent Title (English):Optical Nanoscopy
Publisher:Springer Open
Place of publication:Berlin ; Heidelberg [u. a.]
Document Type:Article
Year of Completion:2013
Date of first Publication:2013/09/10
Publishing Institution:Universitätsbibliothek Johann Christian Senckenberg
Release Date:2013/09/18
Tag:Drift correction; Registration; Single-molecule localization microscopy; Super-resolution microscopy
First Page:1
Last Page:8
© 2013 Zessin et al.; licensee Springer. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
HeBIS PPN:353192600
Institutes:Biochemie und Chemie
Dewey Decimal Classification:540 Chemie und zugeordnete Wissenschaften
Licence (German):License LogoCreative Commons - Namensnennung 2.0

$Rev: 11761 $