Abstract
Purpose :
Aqueous humor outflow through the trabecular meshwork (TM) is non-uniform, or segmental, such that some regions of TM experience higher local outflow than others. Our aims were (i) to develop a quantitative approach to analyze segmental outflow patterns using spatial autocorrelation and cluster analysis, and (ii) to apply these techniques to determine how segmental outflow patterns change over time in living mice.
Methods :
Fluorescent tracer microspheres (0.2 µm; 108 µl-1) were infused under anesthesia into the anterior chamber of C57BL/6 mice to label segmental outflow patterns in the TM. To visualize the change in segmental outflow patterns over time within individual eyes, each eye was infused twice with a different tracer color separated by a delay of 2, 7 or 14 days (n=10 mice). Mice were culled 48 hrs after the second infusion. Eyes were enucleated and prepared for flat mount imaging. We analyzed the spatial autocorrelation of tracer intensity patterns within the TM using Moran’s I, which yields statistically defined clusters of high, intermediate, and low tracer-labelling (Figure).
Results :
The global Moran’s I statistic indicated significantly more tracer clustering within the TM than expected by chance (0.67±0.08; mean±SD; N=16 eyes; p<0.01). High tracer-labelled clusters were estimated to occupy 26±2% of the TM, while low and intermediate clusters were estimated to occupy 34±6% and 40±7% respectively. There was no change in the total extent of each cluster type over time. However, with increasing delay between the two infusions, the patterns of tracer labelling changed significantly, as indicated by a decreasing spatial correlation coefficient (r=0.69 [0.52,0.86] at 2 days vs. 0.06 [-0.17,0.28] at 14 days; mean [95% CI]; p<0.01).
Conclusions :
Moran’s I provides an objective and reproducible method to identify high, intermediate and low tracer-labelled clusters within the TM. Over 2 weeks in young adult mice, segmental outflow patterns redistribute spatially within the TM. However, despite this redistribution, the TM appears to preserve the overall extent of high, intermediate, and low filtration-active clusters. This suggest that dynamic changes in the spatial distribution of segmental outflow are not random but appear to be regulated by active processes within the outflow pathway.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.