The Nociceptor Primary Cilium Contributes to Mechanical Nociceptive Threshold and Inflammatory and Neuropathic Pain

Abstract

The primary cilium, a single microtubule-based organelle protruding from the cell surface and critical for neural development, also functions in adult neurons. While some dorsal root ganglion neurons elaborate a primary cilium, whether it is expressed by and functional in nociceptors is unknown. Recent studies have shown the role of Hedgehog, whose canonical signaling is primary cilium dependent, in nociceptor sensitization. We establish the presence of primary cilia in soma of rat nociceptors, where they contribute to mechanical threshold, prostaglandin E₂ (PGE₂)-induced hyperalgesia, and chemotherapy-induced neuropathic pain (CIPN). Intrathecal administration of siRNA targeting Ift88, a primary cilium-specific intraflagellar transport (IFT) protein required for ciliary integrity, resulted in attenuation of Ift88 mRNA and nociceptor primary cilia. Attenuation of primary cilia was associated with an increase in mechanical nociceptive threshold in vivo and decrease in nociceptor excitability in vitro, abrogation of hyperalgesia, and nociceptor sensitization induced by both a prototypical pronociceptive inflammatory mediator PGE₂ and paclitaxel CIPN, in a sex-specific fashion. siRNA targeting Ift52, another IFT protein, and knockdown of NompB, the Drosophila Ift88 ortholog, also abrogated CIPN and reduced baseline mechanosensitivity, respectively, providing independent confirmation for primary cilia control of nociceptor function. Hedgehog-induced hyperalgesia is attenuated by Ift88 siRNA, supporting the role for primary cilia in Hedgehog-induced hyperalgesia. Attenuation of CIPN by cyclopamine (intradermal and intraganglion), which inhibits Hedgehog signaling, supports the role of Hedgehog in CIPN. Our findings support the role of the nociceptor primary cilium in control of mechanical nociceptive threshold and inflammatory and neuropathic pain, the latter Hedgehog-dependent.

Keywords: chemotherapy-induced neuropathy; Hedgehog; hyperalgesia; inflammation; nociceptor; primary cilium.

Figure 1.

Rat DRG neurons elaborate primary cilia, in vivo. A–D, Immunohistofluorescence analysis of adult rat DRGs in vivo, using confocal microscopy. Histological sections were labeled with antibodies recognizing ARL13B (A–D, green), ɣ-tubulin (A–D, red), and Fox-3 (A, NeuN, grayscale), TrkA (B, grayscale), or CGRP (C, grayscale). D, IB4 staining is indicated in grayscale. A–D, Cell nuclei are marked by DAPI (blue). Red and yellow arrowheads indicate neuronal and non-neuronal primary cilia, respectively. A–D, Scale bar, 3 µm. See Extended Data Figure 1-1 for further cilium analysis.

Figure 2.

Targeted knockdown of Ift88 in rat DRG leads to a loss of neuronal primary cilia and a decrease in the average length of retained cilia. A, Ift88 siRNA specifically reduces Ift88 mRNA levels in the rat DRG. B–G, Reduction in the fraction of ciliated cells and the length of primary cilia in DRG cells in vivo (B–D), and in ex vivo (E–G) acute cultures of DRG neurons exposed, in vivo, to siRNA targeting the rat Ift88 gene, 7 d after final siRNA injection. B, E, Confocal immunofluorescence images of L4 rat DRGs stained for the ciliary axoneme (αArl13b, green), the basal body (αγtub, red), and the nucleus (DAPI, blue). Left panels, Control siRNA. Right panels, Ift88-targeted siRNA. Bottom panels, Magnifications of green boxes (green star) in the top panels. Magnification/zoom factor indicated. Scale bars: 10 µm, top panels; 2 µm, bottom panels. E, IB4-stained cultured DRG neurons. Arrowheads indicate primary cilia. C, D, F, G, Quantification of primary cilium parameters. Percentage ciliation (C, F) and cilium length (D, G) in L4 rat DRGs injected with control versus Ift88-targeted (knockdown) siRNA, both (C, D) in vivo and (F, G) ex vivo neurons cultured from anti-Ift88 siRNA-treated DRGs. **p < 0.01, ***p < 0.001, ****p < 0.00001.

Figure 3.

Effect of siRNA targeting Ift88 on baseline mechanical nociceptive threshold, nociceptor rheobase, and paclitaxel CIPN. A, Male rats were treated intrathecally for 3 d with siRNA targeting Ift88 or its negative control siRNA. Mechanical nociceptive threshold (presented in grams) was evaluated before and 72 h after the last siRNA treatment. At this time, nociceptive threshold was significantly increased by 2, 5, and 10 µg Ift88 siRNA compared with negative control siRNA (p < 0.02, 0.0005, and 0.0001, respectively). B, Ift88-targeting siRNA administered in vivo produced a significant elevation in rheobase in nociceptors cultured from Ift88 siRNA-treated rats, compared with rheobase in nociceptors cultured from control and negative siRNA-treated rats (****p < 0.0001, t₍₃₆₎= 5.1; ^#p = 0.03, t₍₃₆₎= 2.3, Šidák's multiple-comparisons test for ANOVA; effect of treatment is significant: p < 0.0001, F_(2,36)= 12.0; n = 19 control, 13 Ift88 siRNA-treated and 7 negative siRNA-treated groups). C, Example traces of depolarization-induced whole-cell current in rat small-diameter (<35 µm) DRG neurons cultured from control (naive) and in vivo Ift88 siRNA-treated rats. The stimulus current was elicited by step depolarization from a holding potential of −70 to 0 mV for 20 ms using voltage-clamp mode of the whole-cell patch-clamp configuration. The biphasic current with rapid-onset inward phase (driven by sodium voltage-gated ion channels) transitioning into a slower outward phase (driven by potassium voltage-gated ion channels). D, Peak current density of the inward (magnitude; current is negative) and outward phases (positive, measured at the end of depolarization). While there was no significant change in inward current, the outward current was upregulated by 63% (*p = 0.011, t₍₆₀₎= 2.9, Šidák's multiple-comparisons test for two-way ANOVA; effect of treatments is significant: p = 0.004, F_(1,60)= 8.9). E, F, Alleviation of CIPN hyperalgesia by intrathecal administration of Ift88 siRNA. Seventy-two hours after the last administration of anti-Ift88 siRNA (A), paclitaxel (1 mg/kg, i.p.) was administered (Days 0, 2, 4, and 6). Mechanical nociceptive threshold was evaluated before siRNA treatment was started (baseline) and again on Days 7, 14, 21, and 28 after the last dose. siRNA treatment significantly attenuated CIPN hyperalgesia (two-way repeated-measures ANOVA; E, p < 0.0001, F_(9,80)= 11.23; F, p < 0.0001, F_(9,80)= 11.37 followed by Bonferroni’s multiple-comparisons test: **p < 0.01; ***p < 0.005; ****p < 0.0001, n = 6 paws of each group).

Figure 4.

Effect of siRNA targeting Ift52 on paclitaxel-induced mechanical hyperalgesia (CIPN). A, siRNA-mediated knockdown of Ift52 mRNA. The histogram depicts the level of Ift52 mRNA in DRG from rats treated with either siRNA targeting Ift52 or control siRNA. Gene expression was determined with SYBR green RT-PCR and values (mean ± 95% CI) represent expression relative to GAPDH (the housekeeping gene). Rats treated with siRNA targeting Ift52 demonstrate a significant reduction in Ift52 mRNA levels in their lumbar DRG compared with those treated with the control siRNA (unpaired two-tailed Student's t test; p < 0.0001; t₍₁₀₎ = 11.21; n = 6). B, C, Male rats were treated with siRNA targeting Ift52 (10 µg/day, i.t.) for 3 consecutive days. Seventy-two hours after the last siRNA injection, paclitaxel (1 mg/kg, i.p.) was administered, Days 0, 2, 4, and 6. Mechanical nociceptive threshold was evaluated before siRNA treatment was started (baseline), and again on Days −3, −2, −1, 3, 5, 7, 14, 21, and 28 d after paclitaxel injection. B, Magnitude of hyperalgesia is expressed as absolute values of mechanical nociceptive threshold, in grams. Ift52 siRNA attenuates paclitaxel induced CIPN. Data are expressed as mean ± SEM, n = 6 paws in each group. F_(9,90)= 29.38, ****p < 0.0001, ***p < 0.0002, **p < 0.0028; when Ift52 siRNA group was compared with control siRNA group; two-way repeated-measures ANOVA followed by Bonferroni’s multiple-comparisons test. C, Hyperalgesia magnitude is expressed as percentage reduction from the baseline mechanical nociceptive thresholds. Ift52 siRNA attenuates paclitaxel induced CIPN. Data are expressed as means ± SEM, n = 6 paws in each group. F_(8,80)= 12.87, ***p = 0.0005 **p < 0.0089 *p = 0.0143; when the Ift52 siRNA group was compared with the control siRNA group; two-way repeated-measures ANOVA followed by Bonferroni’s multiple-comparisons test.

Figure 5.

siRNA targeting Ift88 attenuates PGE₂-induced hyperalgesia. Male rats were treated with Ift88 siRNA for 3 consecutive days, followed 72 and again 144 h later with PGE₂ (100 ng/5 µl, i.d.), administered on the dorsal surface of one hindpaw. Mechanical nociceptive threshold was evaluated before treatment with siRNA (baseline), 72 h after the last siRNA treatment, and then 30 min after each of two sequential administrations of PGE₂ separated by 72 h. A–C, Ift88 siRNA led to a significant increase in (A) baseline nociceptive threshold and (B, C) reduction in PGE₂ hyperalgesia, measured as % reduction from baseline mechanical nociceptive threshold (B) 72 h and then again (C) 144 h after the last treatment with siRNA.

Figure 6.

siRNA targeting Ift88 increases action potential threshold, increasing peak outward currents, and attenuates PGE₂-induced nociceptor sensitization. A–D, PGE₂-induced reduction of rheobase in putative C-type rat nociceptors cultured from control (A, naive), experimental (B, treated in vivo with siRNA targeting Ift88), and negative control (C, treated in vivo with negative siRNA) rats. Electrophysiological traces show APs generated in response to stimulation of small-diameter DRG neurons (inset) with a square wave current pulse (below AP recordings). The height of the pulse represents rheobase. Dotted line shows membrane potential of 0 mV. D, Graphical representation of reduction in rheobase, comparing control rats to ones treated in vivo with Ift88 siRNA and negative siRNA. Symbols show effect in individual neurons. *p = 0.02, t₍₂₆₎= 2.6; ^#p = 0.02, t₍₂₆₎= 2.8, Šidák's multiple-comparisons test for ANOVA; effect of treatment is significant: p = 0.0017, F_(2,26)= 4.8 (n = 15 control, 8 Ift88 siRNA-treated and 6 negative siRNA-treated groups).

Figure 7.

In female rats Ift88 siRNA does not affect mechanical threshold and PGE₂ or CIPN hyperalgesia. A, Female rats were treated with siRNA for Ift88 mRNA for 3 consecutive days in a dose of 10 µg/20 µl/day, intrathecally. Mechanical nociceptive threshold (in grams) was evaluated before siRNA treatment was started (baseline) and again 72 h after the last siRNA treatment. Measured 72 h after the last dose of siRNA targeting Ift88, there was no significant change in mechanical nociceptive threshold (data shown as means ± SEM, time F_(1,10)= 0.1832, p = 0.6777: siIft88 vs siRNA neg control; two-way repeated-measures ANOVA followed by Bonferroni's multiple post hoc comparisons test, n = 6 paws of each group). B, Seventy-two hours after the last siRNA injection, PGE₂ was administered intradermally (100 ng/5 µl/paw, i.d.), and mechanical nociceptive threshold evaluated 30 min later; magnitude of hyperalgesia is expressed as percentage reduction from the baseline mechanical nociceptive threshold before siRNA treatment. In female rats, Ift88 siRNA did not affect PGE₂-induced hyperalgesia (data shown as mean ± SEM, unpaired Student's t test: t = 0.3576, df = 10; p = 0.7281: siIft88 vs siRNA neg control, n = 6 paws of each group). C, Female rats were treated with siRNA targeting Ift88 (10 µg/20 µl/day, i.t.) daily for 3 consecutive days or with its negative control siRNA (10 µg/20 µl/day). Seventy-two hours after the last injection of siRNA, paclitaxel was administered intraperitoneally (1 mg/kg, i.p.) on Days 0, 2, 4, and 6. Mechanical nociceptive threshold was evaluated before starting siRNA treatment (baseline), and again on Days 0 and 1, 3, 5, 7, 14, 21, and 28 after paclitaxel injection. Hyperalgesia induced by paclitaxel was unaffected in both the negative control and Ift88 siRNA-treated groups (data are expressed as means ± SEM, n = 6 paws in each group; F_(1,10)= 0.9413, p = 0.3548, when the Ift88 siRNA group was compared with the negative control siRNA group; two-way repeated-measures ANOVA followed by Bonferroni's multiple post hoc comparisons test).

Figure 8.

Pronociceptive effect of Shh is primary cilium dependent. A, B, Male rats were treated with siRNA targeting Ift88 (10 µg/day, i.t.) for 3 consecutive days. Seventy-two hours after the last injection of siRNA, rats received intradermal or intraganglion injections of recombinant Shh (200 ng) and mechanical nociceptive threshold was assessed at 0.5, 1, 3, and 5 h after injection. A, The hyperalgesic effect of Shh injected intradermally was significantly reduced by Ift88 siRNA. Data shown as mean ± SEM, treatment F_(1,40) = 73.26, time F_(4,50)= 6.861, ****p < 0.01: siIft88, Shh intradermally versus siRNA neg, Shh intradermally. B, The hyperalgesic effect of intraganglion Shh was also significantly reduced by Ift88 siRNA. Data shown as mean ± SEM, treatment F_(1,40) = 48.13, time F_(3,40) = 2.286, ****p < 0.01: Ift88 siRNA, Shh intraganglion versus siRNA neg, Shh intraganglion Two-way repeated-measures ANOVA followed by Bonferroni's test, n = 6 paws in each group. C, D, Male rats received paclitaxel every other day for a total of four injections, on Days 0, 2, 4, and 6 (1 mg/kg × 4, i.p.). Seven days after the first administration of paclitaxel, rats were treated with intradermal (10 µg/5 µl) or intraganglion (10 µg/5 µl) cyclopamine. As a control, the contralateral paw (i.d.) or contralateral lumbar DRG (i.gl.) received vehicle (saline plus DMSO 2%). Mechanical nociceptive threshold was evaluated before and 7 d after paclitaxel injection (before cyclopamine injection, 0 h), and again 0, 0.5, 1, 3, and 5 h after intradermal or intraganglion cyclopamine injection. C, Paclitaxel-induced hyperalgesia was markedly attenuated in the male rats treated intradermally with cyclopamine. Data shown as mean ± SEM, treatment F_(1,50) = 28.28, time F_(4,50)= 4.154, **p < 0.01: cyclopamine intradermal versus vehicle intradermal groups. D, The magnitude of paclitaxel-induced hyperalgesia was markedly attenuated in male rats treated with intraganglion cyclopamine. Data shown as mean ± SEM, treatment F_(1,50) = 57.38, time F_(4,50)= 4.748, **p < 0.01: cyclopamine intradermal versus vehicle intradermal.

Figure 9.

Nociceptor-targeted RNAi-mediated knockdown of the Ift88 ortholog NompB in Drosophila larvae reduces sensitivity to noxious mechanical stimulation. A, Immunohistofluorescence analysis of larval pickpocket-positive neurons in vivo. Fillet preparations of third instar ppk1.9-tdTomato larvae were labeled with antibodies recognizing a GFP::NompB fusion protein (GFP::NompB+; green). Pickpocket-expressing neurons (ppk/Td+) are indicated by Td-Tomato signal in red. White arrowhead indicates presumed neuronal primary cilium. Scale bar, 10 µm. B, When third instar larvae of genotype ppk1.9-Gal4/UAS-NompB-RNAi were stimulated with a 2,346 kPa von Frey filament, the frequency of their escape response was significantly lower than that of normal controls “No UAS” (ppk1.9-Gal4/y¹v¹; p < 0.001) and “No Gal4” (w¹¹¹⁸/UAS-NompB-RNAi; p < 0.05). N values (left to right) were 170, 161, and 176 animals. Response frequencies were compared by chi-square test.