Sections were imaged with a wide-field epifluorescence microscope (DM5000B; Leica Biosystems, Inc., Buffalo Grove, IL, USA) and Leica DC500 digital camera.
16 Slides were first visualized with the ×10 objective to orient the investigator to the sections (
Fig. 1), and the ×20 objective was used to obtain images for analysis. High-resolution images (3795 pixels/mm) of each stain were taken and then merged with Adobe Photoshop 6.0 (San Jose, CA, USA) as previously described.
16 All SMA and vinculin images were taken on the same day with the same microscope and camera settings. Draq5 images were taken in a similar manner; however, the camera exposure was varied to allow for the single-nuclei resolution needed for cell counting. Merged images were imported into Stereo Investigator software (MBF Biosciences, Williston, VT, USA) and analyzed. In brief, the ciliary muscle was manually outlined with Stereo Investigator software by tracing the ciliary muscle perimeter using a stylus and touch-screen computer display (
Fig. 1); this procedure is repeatable to within ± 0.012 mm
2.
11 Stereo Investigator uses this information to determine both ciliary muscle cross-sectional area and ciliary muscle length. Mean SMA brightness was recorded from Stereo Investigator after correcting for background brightness by analyzing only the green channel images (unmerged image); unmerged images were required because Stereo Investigator was unable to uncouple the brightness values obtained from the multichannel images. Smooth muscle actin brightness was used as a measure of the overall concentration of smooth muscle present. Numbers of cell nuclei were obtained from Stereo Investigator using optimized cell nuclei size detection parameters (
Fig. 1). Mean values for each eye were then used to calculate cell area (cross-sectional area / mean cell number) and ciliary muscle volume (cross-sectional area × limbal circumference).
11 Nasal and temporal ciliary muscle cross-sectional areas, lengths, cell numbers, and cell area differences were determined by comparing the corresponding regions of each eye. Representative confocal microscopy images were also obtained with a Zeiss LSM 510 (Pleasanton, CA, USA) confocal microscope and ZEN 2009 software to determine if ciliary muscle cells were mono- or multinuclear.