to BioTechniques free email alert service to receive content updates.
Correlative light microscopy for high-content screening
 
Benjamin Flottmann*1, Manuel Gunkel*1, Tautvydas Lisauskas*1, Mike Heilemann1,2, Vytaute Starkuviene1, Jürgen Reymann1, and Holger Erfle1
1BioQuant Centre, Heidelberg University, Heidelberg, Germany
2Insitute of Physical & Theoretical Chemistry, Goethe University Frankfurt, Frankfurt, Germany


*B.F., M.G., and T.L. contributed equally to this work
BioTechniques, Vol. 55, No. 5, November 2013, pp. 243–252
Full Text (PDF)
Abstract

High-throughput microscopy is an effective tool for rapidly collecting data on a large scale. However, high throughput comes at the cost of low spatial resolution. Here we introduce correlative light microscopy by combining fast automated widefield imaging, confocal microscopy and super-resolution microscopy. We demonstrate the potential of this approach for scalable experiments. The workflow consists of a robust approach for selecting cells of interest on a wide-field screening microscope at low resolution and subsequently re-localizing those cells with micrometer precision for confocal and super-resolution imaging. As a case study, we visualized and quantified cis- and trans-Golgi markers at increasing resolution.

Introduction

High-throughput widefield or spinning disk confocal screens can process large numbers of targets. For example, RNAi screens can be performed with a sample number sufficient to cover the whole human genome (~23,000 genes). To obtain a reasonable number of cells per field of view and to ensure fast imaging, low to medium spatial resolution is usually chosen (10× air or 40× water objective) (1, 2).

Method Summary

Quantitative description of biological systems is the major challenge of biological research. Fluorescence microscopy plays an important role because it allows researchers to not only visualize but also evaluate virtually every cellular process of interest. Fluorescence microscopy, however, comes with a trade-off between throughput and spatial/temporal resolution. Due to these technological limitations, imaging biological samples is generally constrained to either high throughput but low resolution or high resolution but low throughput.

Point scanning confocal microscopy is applicable to small-scale experiments and is often used following data collection by high-throughput methods (3). In order to increase the information content of small-scale confocal screens, automated correlative screening techniques (4), which combine low with high-resolution point scanning, may be used for targeting relevant phenotypes and zooming closer into interesting events by imaging in a mode of high information context.

The spatial resolution in both widefield and confocal microscopy is limited by diffraction. In recent years, a new generation of microscopy techniques has been developed that bypass this limit in resolution (5-7). Among these are single-molecule techniques, which were demonstrated to reach near molecular spatial resolution (8-12). Additionally, co-localization analysis can be refined by using super-resolution data. For example, co-localization analysis at the molecular level was demonstrated, attributing a co-localization value to each single biomolecule detected (13, 14).

In order to combine fast low-resolution widefield methods with slow high- or super-resolution microscopy, we have developed a workflow for correlative microscopy that combines the best features of each. By a fast widefield screen of a large sample, representative cells showing defined phenotypes are selected and subsequently imaged in confocal and super-resolution modes. To be able to image in the different microscopy modes, we have chosen fluorophores that can be used in all three imaging platforms. In this study, we use correlative light microscopy to analyze the distribution of proteins residing at different cisternae in the Golgi complex. We demonstrate significantly improved spatial resolution of GalT (trans-Golgi marker) and GM130 (cis-Golgi marker) under diverse treatments when applying correlative microscopy for the same cell.

Materials and methods

Cell preparation

NRK (normal rat kidney, CRL-6509; ATCC, Manassas, VA) cells that had been stably transfected with the Golgi enzyme GalT tagged with CFP (NRK-GalT-CFP) were cultured in DMEM (GIBCO/Invitrogen, CA) containing 5% fetal bovine serum (FBS), 1% non-essential amino acids (GIBCO/Invitrogen), 2 mM glutamine, 100 U/mL penicillin and 100 μg/mL streptomycin. For experiments cells were plated on μ-Slide 8-well slides (Ibidi, Martinsried, Germany) at densities of 15,000 cells/well 24 h before drug treatment. Poly-L-lysine(Sigma-Aldrich, M0) was applied to μ-Slides for 15 min before cell plating.

Cell treatment

BFA and nocodazole (Calbiochem/Merck Chemicals, Darmstadt, Germany) treatment was performed at concentrations of 5 μg/mL and 1 μg/mL, respectively, for 20 min. Cycloheximide (0.1 mg/mL; Sigma-Aldrich) was added to abolish protein synthesis.

Sample preparation and immunofluorescence

Cells were fixed incubating in 3% PFA for 20 min. TetraSpeck microspheres (0.1 μm; Invitrogen) were applied and incubated for 15 min, and cells were permeabilized by incubation with 0.1% Triton-X-100 for 5 min. GalT-CFP was counter-stained with anti-GFP (6556, polyclonal rabbit; Abcam, Cambridge, UK) and anti-rabbit Alexa Fluor 532 (Invitrogen). GM130 was stained with anti-GM130 (monoclonal mouse; BD Biosciences, Franklin Lakes, NJ) and anti-mouse Alexa Fluor 647 (Invitrogen).

Widefield microscopy

Widefield images were acquired with an Olympus IX81 microscope with a 10× objective lens (Olympus UPlanSApo, NA 0.4; Olympus, Tokyo, Japan) and a field of view of 866 μm × 660 μm. The sample was illuminated by an 150 W Hg/Xe mixed gas arc burner together with appropriate filter combinations for DAPI, Alexa Fluor 532, and Alexa Fluor 647, respectively. Integration times were set to 200 ms (20 ms for DAPI staining), which were the best match to the dynamic range of the CCD camera. In total, 50 fields of view were recorded for each cell selection.

  1    2    3    4