Confocal autofluorescence and Rhodamine B fluorescence 2D images of enamel and dentin for 19 y, 25 y and 55 years-old human teeth

Three human third molars were used in this study following surgical extraction from patients aged 55, 25 and 19 years old (designated by 19y, 25y and 55y). The samples were cleaned with an ultrasound bath and disinfected in a Chloramine-T (0.5% - 5g/L) solution, fixed in 70% ethanol for more than 48 hours and embedded in an athermal curing epoxy resin. Four slices were sectioned in the mesiodistal plane using a diamond saw with water lubrication, cleaned and dried for 24H. The sections were hand-polished to a final thickness of 200 ±10 μm.

The sections from each patient were finally divided into two groups: 1) stained and mounted in a glycerol solution with 0.02 wt% Rhodamine B;  2) without staining (unstained) and mounted in pure glycerol.

Autofluorescence images were acquired using a confocal laser scanning microscope (CLSM) with an oil immersion objective with 40x 1.3 NA (Leica TCS SP8), at an excitation of 405 nm, by integrating into a wavelength range of 410-650 nm.

To obtain a global overview, large 2D mosaic images of transmitted light and autofluorescence were simultaneously acquired on the unstained samples. The 25y sample was thus imaged over a field of view (FOV) of 10 393 x 5 727 μm² with a mosaic consisting of 31 x 17 tiles with 15 % overlap, each covering an FOV of 388.2 x 388.2 μm² with a pixel size of 758 nm at 2.8 µW laser power.

Higher resolution analysis was then performed by acquiring autofluorescence images from five structurally distinct regions of interest (ROIs) in the 19y, 25y and 55y unstained samples: enamel, dentin-enamel junction (DEJ), dentin, predentin and interglobular dentin (IGD). Images were acquired with a FOV of 72.66 x 72.66 μm² and a pixel size of 35 nm and a FOV of 48.44 x 48.44 μm² with a pixel size of 24 nm at a laser power of 7.17 μW.

To compare the autofluorescence with fluorescence staining, simultaneous acquisitions of autofluorescence and fluorescence imaging were acquired in five similar ROIs in the 19y and 55y stained sample. Two lasers were used simultaneously to excite both the endogenous fluorophores at 405 nm with 14.4 μW power and Rhodamine B at 561 nm with 7 μW. For IGD and predentin, the 561 nm laser was constrained to 2.1 μW. The separation of the two signals was achieved by limiting the detection wavelength range of the two channels to 410–510 nm for autofluorescence and 575 – 600 nm for Rhodamine B, respectively.