Experiments were undertaken in mice engineered to express GFP in microglia under the control of the fractalkine receptor promoter CX
3CR
1 (B6.129P-Cx3cr1tm1Litt/J; Jackson Laboratory, Bar Harbor, ME), which is expressed on microglia, blood monocytes, natural killer, and dendritic cells.
27 Homozygous CX
3CR
1-GFP mice were crossed with the C57BL/6 parental strain to maintain one functional copy of the fractalkine receptor allele in microglia.
28 Heterozygous mice were lethally irradiated with a single dose of 11.5-Gy gamma radiation from a Cesium-137 source (Gammacell 40 Exactor; MDS Nordion, Ottawa, Ontario, Canada). Five hours after the lethal exposure, mice received a whole BMT of 5 × 10
6 cells harvested from age-matched homozygous actin-DsRed donor mice (B6.Cg-Tg[CAG-DsRed*MST]1Nagy/J; Jackson Laboratory).
To investigate the influence of anti-inflammatory therapy on microglial turnover, six mice were treated with dexamethasone after irradiation and BMT. A dose schedule that tapers the dexamethasone concentration during the treatment time was adopted to minimize the side effects of the steroid
14 (
Fig. 1). Mice received initial intraperitoneal (IP) injections of 3 mg/kg dexamethasone (Dexamethasone Sodium Phosphate Injection, USP, 10 mg/mL; West-Ward Pharmaceuticals, Eatontown, NJ) for 2 days, tapering to 1.5 mg/kg IP for 2 more days. A final dose of 0.75 mg/kg was delivered in drinking water for the remainder of the study. Dexamethasone concentration in the drinking water was adjusted twice weekly while monitoring the average daily water consumption and body weight for each mouse. A control group of four mice received no dexamethasone treatment.
The populations of resident GFP+ microglia and DsRed+ BMDCs were tracked by in vivo retinal imaging with our confocal SLO that we developed specifically for multicolor imaging of the murine retina.
26,29 Our custom-built SLO is a confocal microscope that uniquely enables simultaneous excitation and detection of multiple fluorescent markers. To this end, a triple-edge dichroic beam splitter separates the emitted fluorescence from the incident laser light of up to three excitation lasers. Two dichroic beam splitters (FF560-Di01 and FF650-Di01, Semrock, Lake Forest, IL, USA) further split the fluorescence into three spectrally distinct detectors (red = 650–825 nm, green = 550–650 nm, and blue = 500–550 nm). The SLO uses large pinholes (3.2 to 4.1 times the Airy disc size, depending on the wavelength), generating an optical section that is approximately 40 μm thick, capturing a thick slice of the retina at once. A source telescope at each excitation laser introduces slight beam divergence to compensate for chromatic aberrations of the mouse eye. This design enables acquisition of up to three channels simultaneously and, thus, allows detection of fluorescence from multiple distinct cell populations. For the experiments described here, GFP fluorescence of microglia was excited with the 491-nm laser, detected through 525/50 bandpass filter and assigned green color. DsRed fluorescence from BMDCs was elicited with the 532-nm laser, detected through a 561 long-pass filter and assigned red color in the RGB images, respectively.
Because an SLO uses the cornea and lens of the eye as the objective, the image quality and resolution depend critically on the optical characteristics of the eye and the incident laser beam. The resolution an SLO can obtain critically depends on the size of the incident laser beam on the cornea and the quality of the eye itself. Although adaptive optics correction has been used to significantly improve image resolution in the aberrated mouse eye,
30,31 we chose an incident laser beam that underfills the mouse eye pupil, aiming for a suitable balance between optical resolution and aberration that both increase with the incident beam diameter.
31 With an incident laser beam of 1.1 mm in diameter, the diffraction-limited resolution is calculated to be 1.3 μm, sufficient to detect cells that are several microns in size, such as microglia and BMDCs.
28,32 To characterize the optical resolution, the intensity profiles of 12 microglial processes and 22 spines from four different mice were fitted to a normal distribution. The full width at half maximum (FWHM) diameter of the measured microglial structure was compared to literature.
A heated holder that integrates a nose cone for inhalation anesthesia (1%–2% isoflurane in oxygen) was mounted on a six-axis stage to position the dilated pupil into the SLO imaging beam. A contact lens (diameter 2.5 mm, base curvature 1.65 mm, power: +12D, material PMMA; Unicon Corporation, Osaka, Japan) was placed on the mydriatic eye and a drop of GenTeal eye gel (Alcon, Fort Worth, TX, USA) prevented the cornea from drying. In vivo images were recorded before the irradiation and BMT (baseline) and at days 7, 14, 28, 42, and 70 after the irradiation and BMT.
At each time point, the numbers of resident GFP+ cells and DsRed+ bone marrow–derived cells were evaluated. The total retinal immune cell population, defined as the sum of remaining microglia and BMDCs, was computed and tracked. To identify extravasated BMDCs, a 10-minute time-lapse stack was generated at each time point, where one 10-frame mean image was taken every 30 seconds. Cells that were stationary for the duration of the time-lapse were considered as extravasated BMDCs. All animal procedures were approved by the Massachusetts General Hospital Institutional Animal Care and Use Committee and were consistent with the ARVO statement for the Use of Animals in Ophthalmic and Vision Research.