The holographic microscope (Fig. 1a ) was mounted around a commercial epi-fluorescence upright microscope (Olympus BX50WI). As a source for uncaging laser we used a 405nm diode laser (CUBE 405−50, Coherent).
The output beam was expanded (25x) to match the input window of a PALM-SLM (PPM X8267, Hamamatsu) which operates in reflection mode. A 4f telescope (L1 = 750mm, L2 = 300mm) was used to image via a dichroic mirror (DM) (Chroma Technology 425DCXR) the plane of the SLM to the rear aperture of the objective (LUMPLFL60XW/IR). The main source of power loss with an SLM device is related to the portion of un-diffracted light that forms an unwanted central spot (zero order) in the excitation field and to the light diffracted into higher orders (about 40% in total). The diffracted beam was spatially displaced from the zero order spot by introducing a phase grating in the hologram. The zero order spot, the higher orders and the ghost image were eliminated by placing a beam blocker and a diaphragm in the intermediate Fourier plane. This reduced the excitation field to a square of approximately 50×50µm2 (link). Alexa 594 was excited with a 75-W Xenon arc source coupled to a monochromator (Optoscan, Cairn Research) (centre-wavelength at 540nm, slit width 30nm) and imaged using an emission filter Chroma Technology HQ 600/40M. The laser intensity and duration were controlled using a 1MHz Digital-to-Analog Converter (DAC) (National Instruments 6713), whose output clock was linked to the clock of the Analog-to-Digital Converter (ADC), in order to synchronize the illumination pulse and voltage-clamp acquisition. Sample fluorescence was captured on a CCD camera (CoolSNAP HQ2, Roper Scientific), at the upper port of the microscope. For the experiments on cerebellum slices a similar setup was integrated into a modified photolysis system (Prairie Technologies, WI, USA) on an upright microscope (Nikon) equipped with a 100x water immersion objective (Nikon, Plan 1.1 NA) and a 405 nm diode laser (Deep Star, Omicron, Germany). For a description of preparation of brain slices, recording conditions and data analysis seeSupplementary Method online.
The output beam was expanded (25x) to match the input window of a PALM-SLM (PPM X8267, Hamamatsu) which operates in reflection mode. A 4f telescope (L1 = 750mm, L2 = 300mm) was used to image via a dichroic mirror (DM) (Chroma Technology 425DCXR) the plane of the SLM to the rear aperture of the objective (LUMPLFL60XW/IR). The main source of power loss with an SLM device is related to the portion of un-diffracted light that forms an unwanted central spot (zero order) in the excitation field and to the light diffracted into higher orders (about 40% in total). The diffracted beam was spatially displaced from the zero order spot by introducing a phase grating in the hologram. The zero order spot, the higher orders and the ghost image were eliminated by placing a beam blocker and a diaphragm in the intermediate Fourier plane. This reduced the excitation field to a square of approximately 50×50µm2 (link). Alexa 594 was excited with a 75-W Xenon arc source coupled to a monochromator (Optoscan, Cairn Research) (centre-wavelength at 540nm, slit width 30nm) and imaged using an emission filter Chroma Technology HQ 600/40M. The laser intensity and duration were controlled using a 1MHz Digital-to-Analog Converter (DAC) (National Instruments 6713), whose output clock was linked to the clock of the Analog-to-Digital Converter (ADC), in order to synchronize the illumination pulse and voltage-clamp acquisition. Sample fluorescence was captured on a CCD camera (CoolSNAP HQ2, Roper Scientific), at the upper port of the microscope. For the experiments on cerebellum slices a similar setup was integrated into a modified photolysis system (Prairie Technologies, WI, USA) on an upright microscope (Nikon) equipped with a 100x water immersion objective (Nikon, Plan 1.1 NA) and a 405 nm diode laser (Deep Star, Omicron, Germany). For a description of preparation of brain slices, recording conditions and data analysis see
