Javascript is turned off in your web browser.

To take full advantage of this site, please turn it on and then refresh the page.

Fluorescence Microscopy Techniques
Frequently Asked Questions (FAQ)

Show All

Return to Main FAQ List

Even without the "optimal" filter sets, it is possible to visualize many fluorophores. All of our microscopes are equipped with the basic sets to image the common fluorophores. Such fluorophores include the standard fluorescent proteins: GFP, YFP, CFP and the standard fluorescent conjugates: Fluorescein (e.g. FITC) and Rhodamine (e.g. TRITC). For technical reasons (for example, dearth of laser-lines for excitation), not all microscopes can visualize DAPI/Hoechst staining of cell nuclei. Wavelength information for the confocal microscopes are tabulated below.

For less common fluorophores, we are building a comprehensive database of fluorescence filter sets and corresponding microscopes. Until the data are available online, feel free to ask the Facility staff, preferably via email to , about availability for your application. In particular, we are struggling with the database structure to record swappability/compatibility between micrscopes. Although we own multiple filter sets, there is no assurance that the filter set will work on a particular microscope. Even with the correct excitation filter, the wrong light source which lacks sufficient energy in the appropriate wavelengths won't generate usable images. The same considerations are operant for the low-light camera used to visualize samples. Different cameras and detectors have different spectral sensitivities.

Excitation Laser-Lines Emission
 405458473477488514543561 568633647 411–753525568700
Zeiss 510 Meta-2 X X X X X X X X X X
3i SDC X X X X X X

Both Zeiss 510 Meta Confocals are appropriate for almost all fluorophores used in life sciences. Filters and photomultiplier tube detectors are available for the standard blue, green and red emissions of standard fluorophores. The two Zeiss 510 Meta confocals are slightly different in 543nm/561nm laser-lines, where 561nm excitation (Meta2) is better for Texas Red fluorophore. In addition, both Meta confocals have an exclusive spectrometer mode ("lambda fingerprint" mode). At each voxel/pixel of the image, a full emission spectrum ranging 411 – 753 nm, separable into 10nm-wide bins, allows extended detection range and the descrimination of overlapping emission of multiple fluorophores. For example, after appropriate calibration, contributions from autofluorescence can be eliminated using the Meta emission spectrometers. On the newest machine, Meta2, the spectrometer is just as sensitive as the photomultiplier tubes, but have higher dark-noise (only relevant for dim labelling).

The Ultraview Spinning Disk Confocal (SDC) is appropriate for imaging emissions at 525nm (FITC, GFP, etc), 568nm (TRITC, Texas Red, etc) and 700nm (Cy-5, YoPro, etc). Because it lacks near-UV excitation, it is NOT appropriate for DAPI/Hoechst stains of the cell nuclei. Alternative fluorescent nuclear stains are SYTO-Green (Invitrogen, formerly Molecular Probes) and Draq5 (Biostatus Ltd, or Axxora LLC, formerly Alexis).

The 3i Spinning Disk Confocal (SDC) is appropriate for imaging with the standard fluorophores popular in the life sciences. Unlike gas lasers, its solid-state lasers have an extremely long lifetime (25,000 hrs vs 2,000 hrs), guaranteeing constant power between typical experimental sessions.

For standard imaging , there are two considerations to balance when choosing between spinning disk (Ultraview or 3i SDC) vs point-scan (Zeiss 510 Metas) laser confocals: cell viability/bleaching vs image resolution and subregion scan control.

The spinning disk confocals (UltraView or 3i) are best for retaining cell viability and minimize photobleaching. With a spinning disk, multiple laser spots scan the full field of view at speeds much faster than a point-scan confocal. Images of the full field of view are acquired using very fast, sensitive CCD cameras. In particular, the 3i system can produce extremely fast confocal images at 83 Hz (12 ms) and has a regulated CO2/temperature/humidity enclosure for live-cell studies. Originating from the non-adjustinghe downside of spinning disk confocals are two-fold. The z-resolution is fixed and is not as good as point-scan confocals (0.8-1.0 µm vs.0.5 µm).

The Zeiss 510 Metas are point-scan confocals with either photomultiplier tube detectors or spectrometer as the detector. The computer-controlled scan pattern is ideal for "zooming" into arbitrarily rectangular regions of interest, or trading xy-spatial resolution for temporal resolution. The z-spatial resolution can be adjusted by altering the confocal pinhole aperture, potentially acquiring very thin optical sections with the theoretical limit of ~0.5 µm resolution in z. Unfortunately, the laser illumination is sufficiently high that certain cell types die during imaging, whereas other cell types do fine. Instead of live cells, fixed samples seem to work consistently, but photobleaching is always a concern on any fluorescence microscope.

For specialty applications, such as FRAP or FRET, while acquiring confocal images, only the Zeiss 510 Metas provide all options. The UltraView SDC provides some rudimentary FRET capabilities, but no FRAP, and the 3i SDC lacks hardware for either FRET or FRAP. Of course, confocal imaging is not required for many applications, and Olympus Station 2 with its motorized filters can perform FRET (as well as ratio imaging) in a wide-field (non-confocal) modality.

The z-resolution is the optical thickness of the optical z-plane, typically controlled by the size of the pinhole of the confocal. The resolution in the xy-directions are typically the same as wide-field epifluorescence microscopy (~0.25 µm)

Because high-resolution imaging requires immersion objectives (either water or oil immersion), your sample should have a glass coverslip between the specimen and the microscope objective. The maximum thickness of the coverslip is #1 (#0 are acceptable, but tend to be more fragile, often breaking due to thermal stresses when "flaming" them after dipping into 70% ethanol). If you culture you cells in 35mm dishes, use glass-coverslip-mounted dishes (Mattek). If your cells are fixed, please use the appropriate ProLong or SlowFade mounting media (both Invitrogen) or a homemade mounting medium (often glycerol and n-propyl gallate).  You can use nail polish to seal the coverslip (not necessary with ProLong).  Keep such slides stored in the dark at 4° C (although cold-sensitive structures, such as microtubules, might not last if fixation were incomplete).

For multiply labelled samples, there is a worry that one fluorescent dye "bleeds" through the optical filters used for another dye, thus producing artefactual co-localization. The amount of "bleed-through" depends on the optical filters of the microscope and on the amount of dye. In general, exciting/observing multiple fluorophores simultaneously can't evaluate the risk of artefactual co-localization. A quick assessment can be accomplished by exciting only one fluorophore at a time, while monitoring emission settings of all the other fluorophores (e.g. on the confocals, exciting with only one laser at a time and look at all emission channels). Ideally, the "wrong" fluorophore channels should appear dark (no signal, no cross-talk).

For absolute rigor, you should prepare samples (controls) which are labelled with only a single fluorophore. With these singly-labelled controls, you can directly quantify the fraction of "bleed-through" through the "wrong" optical filters compared to "correct" signal ("bleed-through" will always be proportional to "real"). When you examine your double/multiply labelled specimen with the same optical filters, you can use the proportions to computationally remove all "bleed-through" and compute the real distribution/co-localization of fluorescent dyes.

On the Zeiss 510 Meta’s you have another, more sophisticated option. Because the Meta feature is a full spectrometer, signal at multiple emission wavelengths are collected simultabeously. These spectra are a more thorough version of "correct" and "wrong" filter channels. The spectra are characteristic of each fluorophore (similar to the proportionality described above). Using these control spectra, the Zeiss 510 Meta has built-in software to compute the "real" distribution of fluorescent dyes in a multiply labelled sample. Ask us for details.

Yes, the Zeiss 510 Metas easily perform FRAP photobleaching experiments. With the Zeiss 510 software, you choose a small rectangular subregion to photobleach, and then rescan at lower power levels. The protocol is very much like time-series/time-lapse functions. We'd be happy to train you in the technique. At the Facility, time constants as short as ~0.25s have been measured with excellent statistical rigor (faster is possible, but require considerably more statistics).

Yes, the automated Olympus Station 2 and the Zeiss 510 Metas easily perform FRET. In both instruments, fluorescence imaging is performed using donor, acceptor and FRET filter settings, but interpretation requires individual donor and individual acceptor samples for calibration/comparison to the FRET sample. On the Zeiss 510 Meta, an additional modality where photobleaching of the FRET sample (bleach donor only) allows calibration within the same FRET sample. We would suggest that you do the controls (i.e. donor-only and acceptor-only samples) before relying on the short-cut using the photobleaching trick. Please make an appointment to discuss your strategy and options before committing to a protocol.

Last modified