Abstract
Purpose:
2-photon excitation fluorescence imaging (TPEF) permits tissue-based subcellular visualization of the human trabecular meshwork (TM). Optimal excitation by this non-linear technique to simultaneously observe multiple tissue fluorophores in tissue is hard to predict and needs to be determined empirically.
Methods:
Human corneoscleral donor tissue was fixed and stained with Hoechst 33342 for nuclei and Alexa 568-conjugated phalloidin for F-actin. Autofluorescence from the structural extracellular matrix (ECM) and Hoechst signals were captured with a green (525/50nm) filter. F-actin epifluorescence was captured with near-red (585/40nm) or far-red (635/90nm) emission filters. TPEF excitation wavelengths of 750nm, 800nm, 850nm and 900nm were tested.
Results:
Shorter excitation wavelengths yielded the highest emission intensities for all fluorophores. Autofluorescence was barely evident relative to Hoechst at 750nm and 800nm excitations. Hoechst-to-autofluorescence intensity ratios (reflecting signal balance) were: 1.0 at 900nm; 2.7 at 850nm; and 9.2 at 800nm. Bleed-through of green (Hoechst and autofluorescence) signals into the red emission channel increased with longer excitation wavelengths, but a far-red filter reduced this bleed-through. F-actin red intensity-to-green fluorescence ratios increased with longer excitation wavelength up to 900nm excitation.
Conclusions:
850nm excitation provided the optimal balance in emission intensities for TPEF triple fluorophore visualization in the human TM. At 850nm, Hoechst 33342-nuclear labeling was clearly visible without overpowering the ECM’s autofluorecsent signal. A far-red filter permitted Alexa-568 epifluorescence with the highest signal-to-noise ratio.