Information Panel

Selection of Appropriate Imaging Equipment and Methodology for Live Cell Imaging in Drosophila

This information panel was adapted from "Time-Lapse Cinematography in Living Drosophila Tissues," Chapter 21, in Live Cell Imaging: A Laboratory Manual (eds. Goldman and Spector). Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA, 2005.

INTRODUCTION

The wide range of microscopy equipment and techniques are truly bewildering and ever-expanding. In the past two decades, there have been many revolutions in light microscopy techniques made possible by improvements in optics, detector technology, and computers. Furthermore, there is no indication that the rate of development of new equipment is slowing down. Here we attempt to provide an overview of available options and important considerations applicable to imaging Drosophila cells and tissues.

RELATED INFORMATION

See Table 1 for an overview of imaging Drosophila cells and tissues.

Evaluating a Microscope System

When deciding which kind of microscope system to use or purchase, the first consideration is what is most important for the imaging: speed, sensitivity, resolution, multiwavelength discrimination, or cell viability. Many of these requirements are mutually exclusive, so compromises must usually be made. The second consideration is whether the system should be tailored to a specialized application or be a general-purpose instrument. In general, systems with more flexibility in the choice of hardware may be better customized for specialized tasks, but they are likely to have more "technical" problems. Choosing "off-the-shelf solutions" with software that has been through its first teething problems can be more time-efficient. Whatever type of system is being considered, the need to test the instruments using a particular experimental material cannot be overemphasized. Company representatives demonstrating equipment should be advised about the choice of fluorochromes and magnifications, so that they can bring the appropriate filter sets, excitation sources, and objective lenses. In addition to the biological material of interest, it is worth using sets of slides of fluorescent beads such as Inspec and Tetraspec (from Molecular Probes) to provide quantitative data for support (Davis 2000).

The Microscope

The first decision is whether to use an upright ("view from above") or an inverted ("view from below") microscope. In general, for live cell work, an inverted microscope (Fig. 1 ) offers more advantages if high resolution and multidimensional imaging are required. On an inverted microscope, the specimen is generally more accessible, simplifying the microinjection process and the use of growth chambers and environmental control chambers. Inverted microscopes also offer better mechanical stability for mounting CCD cameras.

View larger version (105K): [in this window] [in a new window]

Figure 1. An Olympus IX70 inverted microscope/Applied-Precision DeltaVision motorized stage. Note the condenser and the micromanipulator, which can be rotated out of the way to accommodate a stage cover (inset).

Contrast-Enhancing Bright-Field Methods

Bright-field imaging is desirable in addition to fluorescence as it provides a reference for the location of the fluorescent signal. Contrast-enhancing bright-field methods such as Nomarski, also known as differential interference contrast (DIC), and phase contrast are very useful in specific cases. For example, DIC can be used to observe lipid droplet motility or patterning in living embryos, and phase contrast works well to image mitosis in the developing testis. There are also several ways of "combining" bright-field and fluorescent imaging. If the fluorescence signal is very bright, then one can keep the DIC polarizer and Wollaston prism in place, while imaging fluorescence, and accept a large loss in fluorescence intensity. However, there are specialized filter cubes that allow simultaneous acquisition of DIC images and fluorescent images without loss of fluorescence intensity (e.g., custom cube U-MWIB/DIC-SP from Olympus, made by Chroma).

Choosing a Fluorescence Imaging System to Use with Drosophila Tissues

The two predominant types of epifluorescence imaging used in biological research--wide-field systems and "optical sectioning" systems, in their varying forms--have all been used to image Drosophila tissues (Table 1). In general, optical sectioning techniques (including multiphoton and spinning-disc confocals) perform better than wide-field systems on bright signals with a lot of blur in thick specimens, such as embryos (Fig. 2 ; Fig. 3 ; Fig. 4 ; Movie 1 ) and tissues dissected from larvae and adults. Wide-field deconvolution tends to outperform confocal microscopy on fainter signals and more sensitive material (such as following mRNA dynamics) where an image, however hazy, is still visible on the original unprocessed data.

View larger version (67K): [in this window] [in a new window]

Figure 2. Comparison of confocal and wide-field images of stage-4 syncytial blastoderm embryos (approximately cycles 14-15) expressing GFP in the nuclei. (A) Confocal images (20X water immersion objective 0.7 NA): bright-field, GFP fluorescence, merged. (B) Undeconvolved wide-field image (20X dry objective 0.75 NA): bright-field, GFP fluorescence. (C) Image in B after deconvolution: bright-field, GFP fluorescence. Bar, 50 µm.

View larger version (99K): [in this window] [in a new window]

Figure 3. Comparison of wide-field deconvolution and spinning-disc confocal images. An Olympus 60X NA 1.2 water-immersion objective was used throughout. (A,B) Stage-8 egg chambers expressing Tau-GFP associated with the microtubule cystoskeleton. The location of the oocyte nucleus is indicated by a white arrowhead. Egg chambers are oriented with the oocyte posterior to the right and dorsal side up. (C,D) Particle tracks of fluorescently labeled ftz mRNA after injection of labeled mRNA into stage-4 syncytial blastoderm embryos (approximately cycles 14-15). Particle movement is represented by the projected time-series images (image/2 sec). The paths of two individual particles are indicated by white arrowheads. Particles move from the site of injection, near the center of the embryo (bottom of the image, indicated by an asterisk), toward the peripherally located nuclei (top of the image, not shown). RNA eventually accumulates on the side of the nuclei toward the outer membrane. Bars: (A,B) 50 µm; (C,D) 10 µm.

View larger version (68K): [in this window] [in a new window]

Figure 4. Comparison of confocal and multiphoton imaging of the same stage-4 syncytial blastoderm embryo (cycles 14-15) expressing GFP in the nuclei. Single median xy views 75 nm deep and xz line scan views up to 100 µm deep are compared for confocal (excitation 476 nm) and multiphoton (excitation 870 nm) imaging using a Bio-Rad Radiance 2000 confocal/multiphoton. The same 60X 1.2 NA water-immersion objective was used throughout. Multiphoton emission was detected using the direct detector system (DDS), avoiding the need for a confocal pinhole. Note the faster rate of signal fall-off with depth using confocal compared to multiphoton. The vitteline membrane appears to be more autofluorescent at the multiphoton excitation wavelengths. Bar, 50 µm.

View as movie: (148K): [in QuickTime]

Movie 1. Multiphoton and confocal imaging deep into a late stage (12-18 hr after laying) Drosophila histone-2A GFP embryo. Confocal excitation of the GFP was at 475 nm and multiphoton excitation was at 870 nm. Emission was collected above 520 nm. (Top) Imaging using a 20X 0.75 NA multi-immersion objective with water. (Bottom) Imaging of the same embryo with a 60X 1.2 NA water immersion objective. Note the improved imaging depth possible with multiphoton but the poorer lateral resolution and optical sectioning capabilities.

In practice, it has been observed that wide-field combined with deconvolution can be more sensitive and quantitative and achieve higher resolution than confocal imaging. Confocals are "more convenient" because the results are ready without lengthy processing and a need to image many z sections. However, for rapid image capture of sensitive processes, for example, cytoskeleton dynamics, or following fast RNA particle movements in Drosophila embryos and egg chambers, only spinning-disc confocals can really compete with wide-field. Figure 3 shows a comparison of wide-field and spinning-disc confocal microscopy. Using Tau-GFP associated with the microtubule cytoskeleton in egg chambers as the test specimen, we found that it was possible to image for longer with a PerkinElmer UltraView spinning disc and an Orca-ER-cooled CCD (Hamamatsu) than with PMT-based scanning confocals, or using a wide-field system with a similar CCD camera (Coolsnap-HQ, Roper). However, it was necessary to use maximum laser power and bin 2 x 2 pixels to obtain good image quality. After a couple of minutes of continual imaging, photobleaching and cytoskeletal disruption were evident.

Optimizing Excitation and Emission for Imaging Drosophila Tissues

Optimizing the collection of photons for a given dose of illumination light is largely a matter of selecting appropriate combinations of dye, excitation source, filter sets, detector, and objective. Failure to do this impacts image quality, dye bleaching, and cell viability. Drosophila cells and tissues are no exception, so the rules of good imaging practices apply (for an overview, see Pawley 1995).

In general, longer excitation wavelengths cause less damage to biological specimens and induce less autofluorescence, but the exact characteristics depend on the Drosophila tissue being imaged. For example, yolk is particularly fluorescent in UV and blue. It is better to replace UV excitation with excitation at 405-440 nm, use alternative dyes, or use multiphoton excitation. We tend to use green-excited red-emitting dyes (Rhodamine-like spectral characteristics) when imaging in the presence of yolk.

If the cellular process being imaged is particularly susceptible to damage from imaging, then try to determine the cause of this perturbation.

• Photodamage is caused by the excitation light itself, in the absence of the fluorescent molecule. It is generally not significant, except with UV light or the high-level infrared irradiation used in multiphoton imaging.
  
  
• Phototoxicity is caused when the fluorescent molecule itself (upon irradiation) causes the damage, by generating free radicals or localized heating. This can be particularly significant and is a major reason for keeping the dye concentration and irradiation to a minimum.
  
  
• Photobleaching is the process by which the excited light damages the fluorescent molecule, so that it is no longer able to emit light. Photobleaching is dependent on the exact fluorescent molecule used and the conditions within the cells. All dyes possess a limit to the number of times they can be excited, known as the quantum yield, which is believed to be independent of whether excitation is continuous or intermittent. However, nonlinear photobleaching and phototoxic effects may also occur and could account for some of the benefits experienced with spinning-disc confocals.
  
  

To avoid the effects of photodamage, phototoxicity, and photobleaching when imaging live cells, it is best to minimize exposure times and attenuate the excitation power whenever possible, for example, by using neutral-density filters. It is important to maximize the efficiency of the imaging by matching the choice of fluorescent molecule with the excitation and emission filter sets, the bright peaks of the illumination source (http://www.olympusmicro.com/primer), and the quantum efficiency of the detector. Protective infrared and UV filters can also be useful to reduce the damaging effects of excitation light. Filter sets, particularly excitation filters (UV to blue) should be treated as consumables and checked annually for damage. It is important to note that the excitation and emission curves of fluorescent molecules reported in a company’s product information may be generated under very different conditions compared to in vivo imaging conditions. This is particularly important with multiple dyes when trying to ensure that their spectra are sufficiently separated to allow them to be covisualized. The spectral detection options now available on several confocal systems (such as the Leica SP2 AOBS) offer increased light efficiency and much greater freedom to optimize excitation and emission.

The choice of objective and its use is paramount for image quality and is considered again in more detail below with reference to specific Drosophila tissues. The next most important hardware choice that determines image quality is the detector. Point-scanning confocal systems generally come with optimized photomultipliers so the choice is already made. Spinning-disc confocals and wide-field systems which rely on CCD cameras offer more freedom of choice but may be limited by the availability of software drivers and capture cards for the operating software. A detailed comparison of different CCD cameras is beyond the scope of this chapter and has been covered elsewhere (Amos 2000). Where a choice is available, do not skimp on the cost of the detector! For live cell work, where speed (for following developments) and sensitivity (for reducing the necessity for damaging excitation exposure) are generally paramount, a high-quality cooled CCD is the usual choice for wide-field fluorescence systems.

REFERENCES

Amos W.B. 2000. Instruments for fluorescence imaging. In Protein localization by fluorescence microscopy (ed. V.J. Allan), pp. 67-108. Oxford University Press, England.

Davis I. 2000. Visualising fluorescence in Drosophila--Optimal detection in thick specimens. In Protein localization by fluorescence microscopy: A practical approach (ed. V.J. Allan), pp. 131-162. Oxford University Press, England.

Diaspro A. 2001. Confocal and two-photon microscopy: Application and advances. Wiley-Liss, New York.

Murphy D.B. 2001. Fundamentals of light microscopy and electronic imaging, Wiley-Liss, New York.

Parton R.M. and Davis I. 2004. Lifting the fog: Image restoration by deconvolution. In Cell biology: A laboratory handbook, 3d ed. (ed. J.E. Celis). Academic, San Diego.

Pawley J.B. 1995. Handbook of biological confocal microscopy, 2d ed. Plenum, New York.

Swedlow J.R., Hu K., Andrews, P.D., Roos D.S., and Murray J.M. 2002. Measuring tubulin content in Toxoplasma gondii: A comparison of laser-scanning confocal and wide-field fluorescence microscopy. Proc. Natl. Acad. Sci. 2014-2019. 99::.

Wallace W., Schaefer L.H., and Swedlow J.R. 2001. A workingperson’s guide to deconvolution in light microscopy. BioTechniques 1076-1078, 1080, 1082. 31::.