Introduction

Fluorescent imaging of live cells is one of the most important strategies to study real-time cellular responses. Besides conventional small fluorescent indicators, such as fura-2 (1) and fluo-3 (2) to measure Ca^2 + concentration, there are now many fluorescence resonance energy transfer (FRET)-based fluorescent indicators to measure biological parameters like cyclic nucleotides (3,4) and several other metabolites (5). To capture a clear fluorescent image with a high signal-to-noise ratio, it is important to detect emitted fluorescence light as efficiently as possible utilizing a high numerical aperture objective lens, a wide band-pass filter, and a sensitive charge-coupled device (CCD) camera (6). For further improvement of the fluorescent signal, a commonly used strategy is to increase the intensity of excitation light or the concentration of the fluorophores. A significant side effect of these procedures is that they increase the generation of reactive oxygen species (ROS), such as singlet oxygen (6), an excess of which will damage the cell (phototoxicity) and increase the rate of photobleaching of the fluorophore itself (6). Indeed, fluorescence detection of ROS often suffers from the signal amplification because of self-generation of ROS by the indicator (7).

We recently developed a stroboscopic fluorescence microscopy technique using a high power light-emitting diode (LED) to study the relationship between cytoplasmic free Ca^2 + concentration ([Ca^2 +]i) and flagellar form in swimming sea urchin sperm (8). In our system, a fluorescent Ca^2 + indicator, in this case fluo-4, was excited by short pulses of light (1 ms) generated by the LED. This enabled us to capture sharp and clear fluorescence images from the rapidly moving flagellum (about 40 Hz in the case of sea urchin sperm). The stroboscopic illumination system is not just useful for capturing “frozen” images of a rapidly beating flagellum, however, as the reduced illumination time should decrease the generation of ROS and offer protection from phototoxicity and photobleaching during live cell imaging. In this study, we use motile human sperm as a model cell system and demonstrate how the LED-based pulsed illumination system has advantages against conventional illumination system for fluorescence imaging of live cells.

Materials and Methods

Materials

Semen was obtained from volunteer donors by masturbation after at least 2 days of abstinence. After liquefaction, 1.5 mL Ham's F-10 was applied to 1 mL semen to allow the motile sperm to swim up into the upper layer of the suspension (1 h at 36°C, 5% CO₂). Swim-up sperm were collected and washed twice by centrifugation (750× g for 5 min at room temperature) and suspended in the experimental medium: 120 mM NaCl, 4 mM KCl, 15 mM NaHCO₃, 1 mM MgCl₂, 0.3 mM CaCl₂, 10 mM HEPES, 10 mM D-glucose, 1 mM sodium pyruvate, and 5 mg/mL bovine serum albumin (BSA), pH 7.4, by NaOH. Fluo-3 AM, Fura Red AM™, BCECF-AM, and Pluronic® F-127 were from Molecular Probes™ (Invitrogen, Carlsbad, CA, USA). Poly-L-lysine was from Sigma (St. Louis, MO, USA).

Loading of Fluorescent Indicators Into Sperm and Immobilization of Sperm Onto a Coverslip

Human sperm (1–8 × 10⁷ cell/mL) were incubated with 2 µM fluo-3 AM, 10 µM Fura Red AM, or 2 µM BCECF AM plus 0.2% (w/v) Pluronic F-127 for 1 h at 36°C. To remove an excess of the fluorescent indicator and Pluronic F-127, sperm were washed by centrifugation (750× g for 5 min at room temperature) and resuspended in the same volume of the medium.

A poly-L-lysine-coated (20 µg/mL) round coverslip (25 mm diameter) was mounted in a temperature-controlled chamber (Harvard Apparatus, Holliston, MA, USA), and 50 µL human sperm suspensions were deposited on the coverslip and left for 5 min. Sperm unattached to the coverslip were removed by washing, and the chamber was filled with 250 µL medium. Regions where 50% of sperm are attached to the coverslip only at the head region (with their flagella moving freely) were selected for imaging. After the temperature of the chamber reached equilibrium (around 30°C in the observing area), fluorescence imaging was initiated.

Fluorescence Imaging System

Figure 1 shows a diagram of the imaging equipment used in this study. Epifluorescence images were collected with a Nikon PlanApo 60× (1.4 NA oil immersion; Nikon, Melville, NY, USA) objective using a Chroma filter set (Ex, HQ470/40×; DC, 505DCXRU; Em, HQ510LP; Chroma Technology, Rockingham, VT, USA) mounted on a TE300 Eclipse microscope (Nikon). The LED (Luxeon V Star Lambertian Cyan LED; Lumileds Lighting LLC, San Jose, CA, USA) was attached on a custom-built aluminum holder and mounted in a FlashCube40 assembly (Rapp OptoElectronic GmbH, Hamburg, Germany) connected to rear epifluorescence illumination port of the microscope. The LED was controlled by custom-built stroboscopic power supply that provided 3 A current pulses of 1 ms duration, which was synchronously triggered by the Photometrics® Quantix® 57 CCD camera (Roper Scientific, Tucson, AZ, USA). Images were collected and analyzed with Andor™ iQ software (Andor Technology, Belfast, Northern Ireland). For dual-emission imaging using fluo-3 and Fura Red, we used an Optosplit image splitter (Cairn Research, Kent, UK) with a Chroma filter set (DC, Q595LP; Em1, HQ535/50m fluo-3; Em2, HQ665/65m Fura Red).