TRPV1 Supports Axogenic Enhanced Excitability in Response to Neurodegenerative Stress

Abstract

Early progression in neurodegenerative disease involves challenges to homeostatic processes, including those controlling axonal excitability and dendritic organization. In glaucoma, the leading cause of irreversible blindness, stress from intraocular pressure (IOP) causes degeneration of retinal ganglion cells (RGC) and their axons which comprise the optic nerve. Previously, we discovered that early progression induces axogenic, voltage-gated enhanced excitability of RGCs, even as dendritic complexity in the retina reduces. Here, we investigate a possible contribution of the transient receptor potential vanilloid type 1 (TRPV1) channel to enhanced excitability, given its role in modulating excitation in other neural systems. We find that genetic deletion of Trpv1 (Trpv1 ^-/-) influences excitability differently for RGCs firing continuously to light onset (αON-Sustained) vs. light offset (αOFF-Sustained). Deletion drives excitability in opposing directions so that Trpv1 ^-/- RGC responses with elevated IOP equalize to that of wild-type (WT) RGCs without elevated IOP. Depolarizing current injections in the absence of light-driven presynaptic excitation to directly modulate voltage-gated channels mirrored these changes, while inhibiting voltage-gated sodium channels and isolating retinal excitatory postsynaptic currents abolished both the differences in light-driven activity between WT and Trpv1 ^-/- RGCs and changes in response due to IOP elevation. Together, these results support a voltage-dependent, axogenic influence of Trpv1 ^-/- with elevated IOP. Finally, Trpv1 ^-/- slowed the loss of dendritic complexity with elevated IOP, opposite its effect on axon degeneration, supporting the idea that axonal and dendritic degeneration follows distinctive programs even at the level of membrane excitability.

Keywords: TRPV1; axon; dendritic pruning; glaucoma; neurodegeneration; retinal ganglion cells.

Figure 1

Intraocular pressure (IOP) elevation in Trpv1^−/− mice following microbead occlusion. (A) PCR product of wild type (WT) Trpv1 in C57 mice at 289 bp (lane 2–3) and mutant Trpv1 at 176 bp (lanes 4–5) compared to no-template control (NTC, lane 1). (B) A single unilateral injection of microbeads (1.5 μl) elevates IOP for 2 Weeks in WT (+33%) and Trpv1^−/− (+36%) eyes compared to an equivalent volume saline injection (Ctrl; *p ≤ 0.003). Post-injection IOP in each eye did not differ between strains (p = 0.99; n = 18 animals for each. Statistic: one-way ANOVA, Tukey post hoc (B). Data = mean ± SEM.

Figure 2

Comparison of WT and Trpv1^−/− αON-sustained retinal ganglion cells (RGCs). (A) Confocal micrograph of WT αON-S RGC following intracellular filling with Alexa555 dye (AL555) shows strong localization of non-phosphorylated neurofilament H (SMI32, green). Orthogonal rotation (inset) shows dendrites ramifying narrowly in the ON region of the inner plexiform layer (IPL) relative to the label for choline acetyltransferase (ChAT, green). (B) WT αON-S RGC responds to light (365 nm, 3 s; dashed line) with a sustained train of action potentials during whole-cell current-clamped conditions that preserved resting membrane potential (RMP; 0 pA). (C) Trpv1^−/− αON-S RGC has similar morphology and branching pattern and a more robust sustained response to light (D). When averaged across cells (E,F), the response of Trpv1^−/− αON-S RGCs exceeded WT for both mean (53.4 ± 7.7 vs. 35.5 ± 4.8 spikes/s; *p < 0.001) and peak firing rate (81.4 ± 9.9 vs. 56.2 ± 7.2 spikes/s; *p = 0.02), with more depolarized (RMP; −57.9 ± 1.1 mV vs. −60.6 ± 0.6 mV, *p = 0.03). (G,H) For naïve WT αON-S RGCs (n = 7), bath application of IRTX (100 nM) significantly increased the mean light response histogram (+40%; *p < 0.001) and depolarized the RMP (*p = 0.009). Peak response was unaffected (p = 0.77). WT control group contains 14 cells from naïve eyes and 29 cells from saline-injected eyes. (A,C) Scale bar = 40 μm. Statistics: Student’s t-tests (F), paired t-tests (H). Data = mean ± SEM.

Figure 3

Opposing influence of elevated IOP on WT vs. Trpv1^−/− αON-S RGCs. (A) The spontaneous firing rate of WT and Trpv1^−/− αON-S RGCs is not affected by elevated IOP (p ≥ 0.07). (B) Response to light (dashed line) increases for WT αON-S RGCs with elevated IOP but decreases for Trpv1^−/−. (C) Elevated IOP increased the mean and integrated light response for WT αON-S RGCs (*p ≤ 0.04), but decreased both for Trpv1^−/− (*p ≤ 0.05). The integrated response for Trpv1^−/− Ctrl cells exceeded that for WT (^#p = 0.05), as did mean and peak (#, see Figure 2). (D) RMP for WT αON-S RGCS becomes more depolarized with elevated IOP (−53 ± 1.4 vs. −60.6 ± 0.5 mV; *p < 0.001); for control, Trpv1^−/− was more depolarized (#, see Figure 2). (E) The voltage response of WT and Trpv1^−/− αON-S RGCs following brief (1 s) pulses of depolarizing current (0–280 pA; 2 s inter-stimulus interval). (F) Elevated IOP significantly increased voltage response averaged across current pulses for WT αON-S RGCs (*p = 0.01) but decreased it for Trpv1^−/− compared to Ctrl (*p = 0.002). Statistics: Mann–Whitney test (A), Student’s t-tests (C,D), Kruskal–Wallis test, Dunn’s post hoc (F). Data = mean ± SEM.

Figure 4

Comparison of WT and Trpv1^−/− αOFF-sustained RGCs. (A) WT αOFF-S RGC following intracellular filling (AL555) and labeled for SMI-32 shows dendrites ramifying narrowly in the OFF region of the IPL proximal to ChAT labeling (inset). (B) The voltage response of WT αOFF-S RGC increases and is sustained at the light offset, while excitation diminishes during light stimulation (dashed line). (C) Trpv1^−/− αOFF-S RGC has similar morphology and response to light offset, though less robust than WT (D). When averaged across cells (E,F), the mean response of Trpv1^−/− αOFF-S RGCs to light offset was less than WT (9.5 ± 0.6 vs. 15.3 ± 0.8 spikes/s; *p < 0.001), though the RMP was slightly more depolarized (−54.3 ± 0.8 vs. −56.0 ± 0.6 mV; p = 0.07). (G,H) For WT αOFF-S RGCs (n = 5), bath application of IRTX (100 nM) reduced the mean response histogram to light offset (−34%; *p < 0.001), though peak off response and RMP were not affected (p ≥ 0.14). WT control group consists of 10 cells from naïve eyes and 24 cells from saline-injected eyes. (A,C) Scale = 40 μm. Statistics: Mann–Whitney tests (F) and paired t-tests (H). Data = mean ± SEM.

Figure 5

Elevated IOP similarly influences WT and Trpv1^−/− αOFF-S RGCs. (A) Spontaneous spike activity of Trpv1^−/− αOFF-S RGCs is significantly less than WT for control cells (^#p = 0.025) but increased with elevated IOP (*p = 0.022). (B,C) Response to light offset increases for both WT and Trpv1^−/− αOFF-S RGCs, including peak response for WT (^#p = 0.05) and mean and integrated response for Trpv1^−/− (*p ≤ 0.05); the mean response for Trpv1^−/− control cells was less than WT (#, see Figure 4). (D) Elevated IOP depolarized RMP for WT (*p ≤ 0.001) but not Trpv1^−/− αOFF-S cells (p = 0.81). (E) The voltage response of WT and Trpv1^−/− αOFF-S RGCs following brief (1 s) pulses of depolarizing current (0–280 pA; 2 s inter-stimulus interval). (F) Elevated IOP significantly increased response averaged across current pulses for both genotypes compared to respective control cells (*p ≤ 0.002), which did not differ by genotype (p = 0.21). Statistics: Student’s t-tests (A,D), Student’s t-tests (peak response) or Mann–Whitney tests (mean and integrated responses, C), Kruskal–Wallis, Dunn’s post hoc (F). Data presented as mean ± SEM.

Figure 6

Trpv1^−/− does not significantly influence excitatory synaptic mechanisms. (A) Light-evoked currents (voltage-clamp at −65 mV) of WT and Trpv1^−/− Ctrl and 2 Week αON-S RGCs following application of tetrodotoxin (TTX, 1 μM). Elevated IOP did not affect the peak (B, left) or area of the inward current (B, right) for either WT or Trpv1^−/− 2 Week RGCs compared to their respective Ctrl cells, which also did not differ between WT and Trpv1^−/− (p ≥ 0.31). (C) For αOFF-S RGCs, light offset elicited a transient inward peak followed by a slower recovery. Elevated IOP did not affect the peak (D, left) or area (D, right) of the inward current for either WT or Trpv1^−/− 2 Week RGCs compared to their respective Ctrl cells, which did not differ between WT and Trpv1^−/− (p ≥ 0.35). Statistics: Kruskal–Wallis tests, Dunn’s post hoc tests (B,D). Data = mean ± SEM.

Figure 7

Trpv1^−/− RGCs are less susceptible to pruning. (A) Skeletonized dendritic arbors of αON-S RGCs from Trpv1^−/− retinas are more compact than WT, which lost branching following 2 weeks of elevated IOP. (B) Compared to WT, the mean soma area was smaller (*p = 0.03) and total dendrite length less (*p = 0.002) for Trpv1^−/− control αON-S RGCs. (C) Averaged Sholl analysis for Trpv1^−/− control αON-S RGCs shows reduced dendritic complexity between 90–200 μm from the soma compared to WT (*p ≤ 0.01). (D) Two weeks of elevated IOP reduced the mean number of branch points for WT (p = 0.03) but not Trpv1^−/− αON-S RGCs (p = 0.68). (E) Dendritic arbors for Trpv1^−/− αOFF-S RGCs are more compact than WT. (F) Compared to WT, total dendrite length was less for Trpv1^−/− control αOFF-S RGCs (*p < 0.001). (G) Sholl analysis for Trpv1^−/− control αOFF-S RGCs demonstrates reduced dendritic complexity from 90 to 230 μm from the soma (*p ≤ 0.05). (H) Elevated IOP again reduced number of dendritic branch points for WT αOFF-S RGCs (p = 0.03) but not for Trpv1^−/− (p = 0.48). (A,E) Scale = 50 μm. Statistics: Mann–Whitney tests (soma area, B,F) and Student’s t-tests (B,D,F,H), Two-Way Repeated Measures ANOVA on Ranks, Dunn’s post hoc (C,G).