Telmisartan Reduces Axon Degeneration in Mice With Experimental Glaucoma

Abstract

Purpose: The purpose of this study was to determine if treatment with telmisartan, an angiotensin II type 1 receptor blocker (ARB), protects against retinal ganglion cell (RGC) degeneration in a mouse glaucoma model with induced elevation of intraocular pressure (IOP).

Methods: IOP elevation was induced by injection of polystyrene microbeads into the anterior chamber of the right eye of 3-month-old C57BL/6J mice, with the left eye serving as contralateral control. Starting the day of microbead injection, mice were maintained on solid food pellets with or without incorporated telmisartan. IOP was measured by Tono Lab tonometry prior to and weekly after microbead injection. Twelve weeks postinjection, mice were euthanized to obtain optic nerves for analysis of RGC axons. The total numbers of optic nerve axons were determined manually and automatedly using AxonJ. Degenerating axons were counted manually.

Results: IOP elevation induced by microbead injection was similar in magnitude and duration in vehicle and telmisartan-fed mice, although IOP was reduced 5.8% in uninjected mice treated with telmisartan (P = 0.0027). Axon loss determined by manual and automated methods was greater in vehicle compared to telmisartan-treated mice (manual: 9.5% vs. 1.8%, P = 0.044; automated: 14.2% vs. 2.9%, P = 0.0375). An increase in the percent of axons undergoing degeneration was observed in nerves from microbead-injected eyes that was greater in vehicle-treated compared to telmisartan-treated mice (49.0% vs. -0.58%, P = 0.0019).

Conclusions: Elevation of IOP by microbead injection led to loss of RGC axons in vehicle-treated mice that was largely prevented by telmisartan treatment, suggesting a neuroprotective effect of telmisartan.

Figure 1.

Chow consumption. The rate of chow consumption normalized to mouse body weight was determined at 12, 23, 36, 46, 57, and 77 days after starting administration of telmisartan-containing or normal (vehicle) chow (A). At each time point, there was no significant difference in chow consumption between vehicle and telmisartan-treated mice (P > 0.3, Student's t-test, n = 19 and 18, respectively). The body weights of male (gray and yellow lines) and female (blue and orange lines) fed telmisartan (yellow and orange lines) or vehicle (gray and blue lines) were determined weekly (B). Mice fed telmisartan tended to have lower body weight compared to vehicle-treated mice (1.3–8.4% lower), with differences reaching statistical significance at time points indicated in the figure (*P < 0.05, ⁺P < 0.01, Student's t-test performed at each time point). Symbols represent group averages at each time point, with error bars indicating +/− SEM. Numbers of mice in each group were 11 males and 8 females treated with vehicle; and 11 males and 7 females treated with telmisartan.

Figure 2.

IOP responses to microbead injection. Microbeads were injected into the anterior chamber of the right eye (filled symbols) of mice treated with telmisartan (red symbols, n = 18) or vehicle (black symbols, n = 19), while the left eye was not injected (open symbols), resulting in similar elevations of IOP in microbead-injected compared to uninjected eyes (A). Average IOP for individual mice over the 77-day period showed that compared to vehicle, telmisartan treatment reduced IOP in uninjected eyes (B, P = 0.0027, Student's t-test) but did not reduce the magnitude of the IOP response in microbead-injected eyes (C, p = 0.41, Student's t-test). Symbols represent either group means at each time point +/− SEM A or means for individual eyes over the study period, with larger symbols to the right of each group representing mean with error bars representing the 95% CI B and C. Vehicle: eyes from mice treated with vehicle; Telmi: eyes from mice treated with telmisartan.

Figure 3.

Manual axon counting. Images of PPD-stained optic nerve cross-sections were overlaid with a counting masks consisting of 24 boxed regions (A, red boxes) and both normal and degenerating axons were counted in each box to determine axon density, which was multiplied by nerve area to obtain the total number of axons. Examples of normal and degenerating axons are indicated by yellow and orange arrows, respectively in zoomed-in images (B). Manual axon counts of nerves (C) from vehicle-treated (black symbols) and telmisartan-treated mice (red symbols) revealed a significant decrease in the number of axons from microbead-injected eyes (closed symbols) as compared to their contralateral uninjected eyes (open symbols) for vehicle-treated (P = 0.0012, paired t-test), but not for telmisartan-treated mice (P = 0.39, paired t-test, lines connect paired eyes from individual mice). The percentage axon loss (D) in microbead-injected eyes compared to contralateral control eyes was significantly greater for vehicle-treated mice (black symbols) as compared to telmisartan-treated mice (red symbols, P = 0.0396), with larger symbols with error bars representing mean and 95% CI. Vehicle-treated: n = 13, telmisartan-treated, n = 17.

Figure 4.

Automated axon counting. Plotting the number of axons determined by AxonJ on the vertical axis and the number determined manually on the horizontal axis for both eyes from vehicle-treated and telmisartan-treated mice (n = 60 eyes from 30 mice) revealed that AxonJ under-counted axons compared to the gold standard of manual counting, as indicated by data points below the dashed black equivalence line (A). Plotting the ratio of AxonJ counts to manual counts versus axon density for each nerve (B) shows that under-counting by AxonJ is correlated with axon density (R² = 0.70667, P < 0.0001). Adjusting AxonJ counts for axon density (C) aligned the best-fit line of the corrected AxonJ versus manual counts data (red dotted line) with the equivalence line (black dashed line). Using corrected AxonJ counts to evaluate axon loss (D) in nerves from mice treated with vehicle (black symbols) or telmisartan (red symbols) showed a significant decrease in microbead-injected eyes (closed symbols) as compared to their contralateral uninjected eyes (open symbols) for vehicle-treated mice (P = 0.0027, paired t-test, lines connect paired eyes from individual mice), but not for telmisartan-treated mice (P = 0.26, paired t-test). The percentage axon loss (E) in microbead-injected eyes compared to contralateral control eyes was significantly greater for vehicle-treated mice (black symbols) as compared to telmisartan-treated mice (red symbols, P = 0.0375), with larger symbols with error bars representing mean and 95% CI. Vehicle-treated: n = 13, telmisartan-treated, n = 17. Equations of the best-fit lines (red dotted line in A and C, orange dotted line in B) along with R² and P values are shown in each figure A–C.

Figure 5.

Degenerating axons. The percentage of axons undergoing degeneration was determined by manual counting (A) of nerves from vehicle-treated (black symbols) and telmisartan-treated mice (red symbols) and revealed a significant increase in the percentage of axons undergoing degeneration from microbead-injected eyes (closed symbols) as compared to their contralateral uninjected eyes (open symbols) for vehicle-treated mice (P = 0.0044, paired t-test), but not for telmisartan-treated mice (P = 0.36, paired t-test, lines connect paired eyes from individual mice). The percentage increase in % degenerating axons (B) in microbead-injected eyes compared to contralateral control eyes was significantly greater for vehicle-treated mice (black symbols) compared to telmisartan-treated mice (red symbols, P = 0.0019), with larger symbols with error bars representing mean and 95% CI. Vehicle-treated: n = 13, telmisartan-treated, n = 17.