SummaryA novel compact illumination device in variable-angle total internal reflection fluorescence microscopy (VA-TIRFM) is described. This device replaces the standard condensor of an upright microscope. Light from different laser sources is delivered via a monomode fibre and focused onto identical parts of a sample under variable angles of total internal reflection. Thus, fluorophores in close proximity to a cell-substrate interface are excited by an evanescent wave with variable penetration depth, and localized with high (nanometre) axial resolution. In addition to quantitative measurements in solution, fluorescence markers of the cytoplasm and the plasma membrane, i.e. calcein and laurdan, were examined using cultivated endothelial cells. Distances between the glass substrate and the plasma membrane were determined using the mathematical algorithm of a four-layer model, as well as a Gaussian-shaped intensity profile of the illumination spot on the samples. Distances between 0 and 30 nm in focal contacts and between 100 and 300 nm in other parts of the cell were thus determined. In addition to measurements of cell-substrate topology, the illumination device appears appropriate for numerous applications in which high axial resolution is required, e.g. experiments on endocytosis or exocytosis, as well as measurements of ion concentrations proximal to the plasma membrane. The compact illumination device is also suitable for combining TIRFM with further innovative techniques, e.g. time-resolved fluorescence spectroscopy, fluorescence lifetime imaging (FLIM) or fluorescence resonance energy transfer (FRET).

The plasma membrane of Chinese hamster ovary cells was made permeable using the focused beam of an argon ion laser (488 nm) and phenol red as a light absorbing dye. Small circular dark spots on the cell surface appeared immediately after laser irradiation and disappeared within about 5 min. They were related to transient changes in membrane properties, which could be visualized using the fluorescent marker laurdan, and were probably due to a local increase in temperature. According to a colony forming assay, cell viability was maintained by using light doses up to 2.5 MJ/cm(2) applied for 1 s. In addition to measurements of the efflux of the cytoplasmic marker calcein, cell transfection using a green fluorescent protein (GFP) coding plasmid was studied: brightly fluorescent GFP with an emission maximum around 510 nm was observed within part of the cells after 24 h. The transfection rates after laser irradiation were around 30% for younger subcultures and less than 10% for aging cells. This may be due to age dependent changes in the phase transition of membrane lipids from gel phase to liquid crystalline phase. High transfection rates, visual control and universality towards various cell lines are possibly the main advantages of laser-assisted optoporation in comparison with presently existing methods of cell transfection.

A test system for cell viability based on colony formation has been established and applied to high resolution fluorescence microscopy and single molecule detection. Living cells were irradiated either by epi-illumination or by total internal reflection (TIR) of a laser beam, and light doses where at least 90% of irradiated cells survived were determined. These light doses were in the range of a few J/cm2 up to about 200 J/cm2 depending on the wavelength of illumination as well as on the presence or absence of a fluorescent dye (e.g., the membrane marker laurdan). In general, cells were less sensitive to TIR than to epi-illumination. However, comparably high light doses needed for repetitive excitation of single molecules limit the application of super-resolution microscopy to living cells.

Lifetime images of autofluorescence of cultivated endothelial cells were recorded using a novel picosecond laser diode in the near ultraviolet range (375 nm). In contrast to existing picosecond light sources this wavelength permits efficient excitation of the free and protein bound coenzyme NADH with fluorescence lifetimes of 0.4-0.5 ns and 2.0-2.5 ns, respectively. The effective fluorescence lifetime tau(eff) (depending on both lifetimes) was homogenously distributed over the cells with some shortening in the perinuclear region, possibly close to mitochondria. A slight decrease of tau(eff) was observed after inhibition of the mitochondrial respiratory chain, whereas a slight increase was observed after inhibition of the glycolytic pathway, thus indicating variations of the ratio of free and protein bound NADH. Although present applications are still limited by their low pulse energy (< or = 5 pJ), uv picosecond laser diodes have a large potential in high resolution fluorescence microscopy and fluorescence lifetime endoscopy.

SummaryTest systems for measuring cell viability in optical microscopy (based on colony formation ability or lysosomal integrity) were established and applied to native cells as well as to cells incubated with fluorescence markers or transfected with genes encoding for fluorescent proteins. Human glioblastoma and Chinese hamster ovary cells were irradiated by various light doses, and maximum doses where at least 90% of the cells survived were determined. These tolerable light doses were in the range between 25 J cm −2 and about 300 J cm −2 for native cells (corresponding to about 250−3000 s of solar irradiance and depending on the wavelength as well as on the mode of illumination, e.g. epi-or total internal reflection illumination) and decreased to values between 50 J cm −2 and less than 1 J cm −2 upon application of fluorescent markers, fluorescent proteins or photosensitizers. In high-resolution wide field or laser scanning microscopy of single cells, typically 10−20 individual cell layers needed for reconstruction of a 3D image could be recorded with tolerable dose values. Tolerable light doses were also maintained in fluorescence microscopy of larger 3D samples, e.g. cell spheroids exposed to structured illumination, but may be exceeded in superresolution microscopy based on single molecule detection.