Journal Logo

Full Report

Ciguatera poisoning: the role of high-voltage-activated and store-operated calcium channels in ciguatoxin-induced sensory effects

L’Herondelle, Killian PhDa; Misery, Laurent MD, PhDa,b; Le Gall-Ianotto, Christelle PhDa; Philippe, Reginald PhDc; Talagas, Matthieu MD, PhDa,b; Mignen, Olivier PhDc,d; Lewis, Richard J. PhDe; Le Garrec, Raphaele PharmD, PhDa,

Author Information
doi: 10.1097/itx.0000000000000043
  • Open


Ciguatera fish poisoning (CFP) is caused by the consumption of fish contaminated with ciguatoxins (CTXs)1. The most characteristic symptoms are cutaneous sensory disorders, including peri-oral and acral paresthesia, cold dysesthesia and intense pruritus. CFP is commonly named gratte (itch) or “grattelle” in French-speaking Pacific and Caribbean countries due to the widespread occurrence of pruritus2,3. CFP pruritus is localized to the palms and soles or generalized4–6 and occasionally associated with skin rash7,8. This symptom commonly persists for weeks or months after acute CTX exposure, recurring especially after consuming alcohol or meat products, or exercising8–13. No specific treatment exists.

CTXs potently activate voltage-gated sodium (Nav) channels and inhibit voltage-gated potassium (Kv) channels in sensory neurons, thereby increasing their excitability14–16. The downstream molecular events that lead to CFP cutaneous sensory disturbances remain poorly defined. In vivo experiments are limited by the low availability of CTXs. We previously showed that Pacific ciguatoxin-2 (P-CTX-2) induces the release of substance P (SP) and calcitonin gene-related peptide (CGRP) in co-cultured rat dorsal root ganglion (DRG) neurons and human epidermal keratinocytes17. Peptidergic DRG neurons were identified as major targets of CTXs18. Pacific ciguatoxin-1 (P-CTX-1) triggered CGRP release from rat skin flaps and neurogenic flare in humans after intracutaneous injection19,20. Thus, the release of neuropeptides from our co-culture model is a relevant in vitro endpoint for exploring the molecular processes that lead to CFP sensory disturbances.

Here, we studied the molecular mechanisms involved in the SP release induced by P-CTX-2. SP is released from peptidergic primary afferents in response to noxious or irritating stimuli, proteases or depolarization21. SP participates in itch22–24 and pain25,26 by contributing to cutaneous neurogenic inflammation and spinal transmission of pain and nonhistaminergic itch25,27–30.

An increased concentration of cytosolic calcium ([Ca2+]i) is a key factor for the SP release from DRG neurons31,32. Our previous work17 suggested that calcium influx plays a critical role in P-CTX-2-evoked SP release. Among the molecular candidates likely to support Ca2+ influx into sensory neurons, L-type and N-type voltage-gated calcium (Cav) channels mediate the release of neuropeptides from peripheral or spinal DRG afferents33–37. Calcium entry can also occur through the Na+/Ca2+ exchanger (NCX) functioning in its reverse mode, which contributed to the P-CTX-1-elicited acetylcholine release in Torpedo synaptosomes38,39. The calcium-permeable transient receptor potential ankyrin 1 (TRPA1) and vanilloid 1 (TRPV1) channels are expressed in subpopulations of peptidergic nociceptors and are directly or indirectly activated by painful stimuli and pruritogens40,41. In mice, P-CTX-1-induced cold allodynia involves TRPA1 sensitization to cold in a peptidergic subpopulation of C-type primary sensory neurons18. Store-operated calcium entry (SOCE) is activated by the depletion of the endoplasmic reticulum (ER) calcium stores and is supported in nonexcitable cells by the highly Ca2+-selective channel Orai calcium release-activated calcium modulator 1 (ORAI1). Recently, SOCE involving notably ORAI1 was also shown in sensory neurons42,43, driving their hyperexcitability44. The present study sought to explore the involvement of the aforementioned calcium entry actors in the [Ca2+]i increase and SP release elicited by P-CTX-2 in primary sensory neurons.


Single-cell Ca2+ imaging experiments were conducted in cultured rat DRG neurons, while SP release experiments were performed in an optimized co-culture model of rat DRG neurons and human epidermal keratinocytes. We showed in a recent study45 that, although SP release was significantly increased in response to P-CTX-2 in monocultured DRG neurons but not keratinocytes, the SP levels released in the co-culture were 6-fold higher than those released in monocultured neurons, becoming sufficient to study the molecular players involved.

Monocultures of primary rat DRG neurons and co-cultures with human keratinocytes

The rat sensory neurons were obtained as previously described17. Briefly, the sacral, lumbar, thoracic, and cervical dorsal root ganglia were extracted from newborn male and female (2- to 5-day-old) Wistar rats, chemically digested by collagenase IV (200 U/mL) and mechanically dissociated with a fire-polished Pasteur pipette. The neuronal cell suspension was filtered through a 100-µm nylon cell strainer, centrifuged at 150 g for 5 minutes and used for the cytosolic Ca2+ imaging experiments or co-culture. The animal experimental procedures were performed in accordance with the French Ministry of Agriculture and the European Communities Council Directive 2010/63/UE and were approved by the local veterinary authority.

Human skin samples were obtained from healthy adult donors undergoing abdominal reduction surgery after obtaining informed consent. Dissociated human epidermal keratinocytes were obtained as previously described46. Briefly, the epidermal layer was separated after digestion with dispase, and the cells were dissociated using a 0.05% trypsin-EDTA (ethylenediaminetetraacetic acid) solution. The epidermal suspension was filtered through a 100-µm nylon cell strainer and centrifuged at 400 g for 5 minutes. The cells were cultured in complete keratinocyte serum-free medium and detached using trypsin-EDTA for subculture.

The co-culture model reported in our previous study17 was subjected to some modifications to obtain better neurite outgrowth and potentiated neuropeptide release. The dissociated DRG neurons were seeded on 96-well plates at a rate of one newborn rat per 12–14 wells in DMEM:DMEM/F12 1:1 supplemented with Normocin (100 µg/mL), B27 (20 µL/mL), nerve growth factor (100 ng/mL), insulin (4 µg/mL), brain-derived neurotrophic factor (20 ng/mL), and hydrocortisone (10 ng/mL); then, the neurons were incubated at 37°C in a 5% CO2 humidified atmosphere. After 5–7 days of culture, the medium was gently removed from the wells, and keratinocytes from 2 to 3 and 8–9 passages were seeded at a density of 20,000–25,000 cells per well in complete keratinocyte serum-free medium. The co-cultures were maintained for 24 hours at 37°C in a 5% CO2 humidified atmosphere to allow for keratinocyte attachment; then, the medium was replaced with DMEM:DMEM/F12 1:1 to induce keratinocyte differentiation for 12–16 hours.

Culture treatments

P-CTX-2 (52-epi-54-deoxyCTX1B), purified from Gymnothorax javanicus livers, as previously described47,48, was the purest P-CTX-2 available (>90% purity) and did not contain substances other than P-CTXs. P-CTX-2 was dissolved in methanol (MeOH):water (1:1) to prepare a 10 mM stock solution.

The monocultured or co-cultured cells were treated for 90 minutes at 37°C in a 5% CO2 humidified atmosphere with 10 nM P-CTX-2 or its vehicle control (MeOH 0.05%) in a final volume of 100 µL DMEM:DMEM/F12 1:1 supplemented with Normocin. In some experiments, the external calcium was omitted from the medium, and the calcium chelator ethylene glycol tetraacetic acid (EGTA, 1 mM) was added. The pharmacological antagonists nifedipine, KB-R7943, benzamil hydrochloride, AMG 9810 and HC-030031 (Sigma-Aldrich, St. Quentin Fallavier, France), ω-conotoxin GVIA, (Smartox, Saint Martin d’Hères, France) and Synta-66 (Aobious, Gloucester, USA) were applied (at the concentrations specified in the figure legends) 15 minutes before the exposure to P-CTX-2. The co-culture supernatants from each condition were collected in the presence of a protease inhibitor cocktail (Roche, Meylan, France), centrifuged to remove the floating cells, and stored at −20°C until SP measurements.

Cytosolic Ca2+ imaging in DRG neurons

The changes in the cytosolic free Ca2+ concentration were measured using the Ca2+-sensitive fluorescent probe Fura-2 (Thermo Fisher Scientific, Molecular probes). The neuronal cell suspension was plated on poly-L-lysin (PLL)-coated glass coverslips. One day after plating, the cells were loaded with 4 µM Fura-2/AM plus 2 µM pluronic acid (Gibco) for 45 minutes in the dark at 37°C in a medium containing the following (in mM): 135 NaCl, 5 KCl, 1 MgCl2, 1.8 CaCl2, 10 HEPES, and 10 glucose and a pH adjusted to 7.45 with NaOH. The cells were then treated with 10 nM P-CTX-2 with or without a 15-minute pretreament with an antagonist (at the concentrations specified in the figure legends). Ratiometric images of the Ca2+ signals (ratios of the emitted fluorescence measured at 510 nm when cells were excited at 340 and 380 nm) were obtained under a microscope IX71 (Olympus, Tokyo, Japan) equipped with a monochromator illumination system (Polychrome V, TILL Photonics). Emitted fluorescence was collected through a 415DCLP dichroic mirror by a 14-bit CCD camera (EXiBlue, Qimaging). Image acquisition and analysis were performed with Metafluor 6.3 software (Universal Imaging, West Chester, PA). The regions of interest (ROIs) were placed on the cell soma of clearly defined single DRG neurons, and a dim region was selected as the background. The neurons were selected according to their morphology, and the neuronal identity was assessed at the end of the experiments by applying KCl, which induces neuron depolarization and a subsequent increase in [Ca2+]i. The experiments were performed at room temperature in HEPES-buffered solution (pH 7.4).

The cytosolic Ca2+ imaging data are expressed as the mean±SEM of at least 3 separate experiments. The calcium responses are presented as the 340/380 nm fluorescence ratio normalized to the initial ratio F0 using the formula ΔF/F0=(F−F0)/F0, where F is the ratiometric value at a given moment, and F0 is the average of the values before any cell treatments (baseline). The ΔF/F0 amplitude values at each time point measurement of the P-CTX-2-elicited calcium response in sensory neurons were reported using the maximum ΔF/F0 for the first and second transients and the average ΔF/F0 obtained between 35 and 40 minutes after P-CTX-2 application for the plateau. For this heterogeneous population, the cells were considered to respond if they exhibited a ΔF/F0 increase of at least 0.15, as previously described in DRG neurons18, and 2 parameters were reported: the percentages of cells responding to P-CTX-2 and the amplitude values ≥0.15. Antagonist inhibitory effects on the two parameters were assessed for each response phase using normalized data; the averaged data obtained from independently repeated experiments with the application of P-CTX-2 alone were considered 100%.

SP enzyme immunoassay (EIA)

The levels of SP in the co-culture supernatants were quantified using a SP EIA kit (Cayman chemical, Bertin Pharma, Montigny le Bretonneux, France) following the manufacturer’s instructions as previously described49. The levels of the neuropeptide were obtained in pg/mL.

The SP quantification data are expressed as the mean±SEM of at least 4 separate experiments. The values obtained under the vehicle control conditions (MeOH 0.05%) were subtracted from those obtained under the 10 nM P-CTX-2 conditions. The SP levels (in pg/mL) obtained under the antagonist conditions were normalized as previously described17,45, that is, expressed as a percentage of those obtained under the control conditions using the following formula: % of CTX-induced release=(SPCTX+x–SPMeOH+x)/(SPCTX–SPMeOH)×100, where x is an antagonist. Given the interexperimental variability in SP release, this was done experiment by experiment, leading to a value of 100% without SEM for the P-CTX-2 control condition.

Statistical analysis

Statistical analyses were conducted with GraphPad Prism 6.0 (San Diego, CA), and the details are provided in the figure legends. The differences were considered statistically significant if the P-value was <0.05.


P-CTX-2 induces a complex calcium signal in DRG neurons

Our single-cell Ca2+ imaging results show that the responses induced by 10 nM P-CTX-2 were heterogeneous, probably involving different neuron subpopulations50. However, the calcium responses of DRG neurons were typically biphasic and long-lasting, as shown by the patterns presented in Figure 1A. This typical response can be divided into 3 components: (1) a first immediate acute and transient [Ca2+]i increase, called the “first transient,” which ended with a partial return to the baseline level; (2) a second rapid and more sustained [Ca2+]i increase called the “second transient”; and (3) a long-lasting calcium “plateau” that almost never returned to the baseline level at the end of the recording period.

Figure 1
Figure 1:
In dorsal root ganglion (DRG) neurons, Pacific CTX-2 (P-CTX-2) evokes a biphasic and sustained calcium response. A, Representative patterns of calcium signal responses evoked by P-CTX-2 (10 nM) in DRG neurons. The calcium response was decomposed into the following 3 phases: (1) a first immediate acute [Ca2+]i increase called the “first transient,” followed by (2) a second [Ca2+]i increase called the “second transient,” and (3) a long-lasting calcium “plateau” that almost never returned to the baseline level. Vehicle control (MeOH 0.05%) had no effect on [Ca2+]i levels (data not shown). The ΔF/F0 amplitude values during each phase of the P-CTX-2-elicited calcium response were reported using the maximum ΔF/F0 for the first and second transients and the average ΔF/F0 obtained between 35 and 40 minutes after P-CTX-2 application for the plateau. The neurons were considered to respond if they exhibited a ΔF/F0 increase of at least 0.15, and 2 parameters were reported: the percentages of cells responding to P-CTX-2 and the amplitude values ≥0.15. B, Summary panel of both average parameters for each phase of the P-CTX-2-evoked calcium signal obtained on a total of 1720 neuronal cells analyzed.

P-CTX-2 elicited a calcium response in half of the rat sensory neurons (871/1720 analyzed cells, ie, 50.6±4.3%, responding with the first transient). In responding neurons, the average amplitude of the P-CTX-2-elicited first transient was 1.00±0.08. Most DRG neurons responding to P-CTX-2 with the first transient (83.2%, ie, 42.1±4.7% of the studied population) displayed the second transient, which reached an average amplitude of 0.75±0.06. After treatment with P-CTX-2, the [Ca2+]i level in sensory neurons did not return to basal level after the second transient, and a long-lasting increase in [Ca2+]i (plateau) was observed. Two-thirds (67.8%) of the DRG neurons responding to P-CTX-2 with a second transient, and half (56.3%) of those responding with a first transient, displayed a plateau. Of the entire rat DRG neuron population, 28.5±3.4% displayed this plateau, and the average amplitude value was 0.45±0.03. These values are summarized in the Figure 1B.

Calcium influx through N-type Cav, and then ORAI1 channels is involved in the P-CTX-2-induced calcium response in DRG neurons

To explore the temporal sequence of molecular events involved in this complex calcium signal, the inhibitory effect of specific pharmacological antagonists on the following two parameters, expressed as a percentage of values obtained in cells treated with P-CTX-2 alone (ie, normalized to this control condition), was assessed during each phase of the signal: (1) the percentage of neurons responding with such a phase pattern and (2) the associated amplitude values measured in the responding cells. The antagonists used included nifedipine and ω-conotoxin GVIA, which block L-type and N-type Cav channels, respectively; KB-R7943, used at 0.5 µM to assess the involvement of NCX in its reverse mode51; AMG 9810, a specific inhibitor of TRPV1 channel52; and Synta-66, an inhibitor of ORAI1-supported SOCE53.

Cell incubation in calcium-free medium significantly and strikingly reduced the percentage of DRG neurons responding to P-CTX-2 with the first transient (18.5±6.5% of the control value; Fig. 2A), which was also significantly decreased by the N-type blocker ω-conotoxin GVIA (51.9±15.9%). In the responding neurons, the average amplitude of the first transient was significantly decreased in calcium-free medium (37.7±5.2% of the amplitude value measured in control condition; Fig. 2B) and by ω-conotoxin GVIA (57.9±10.0%). No significant changes in the percentage of responding cells (Fig. 2A) and amplitude values (Fig. 2B) of the first calcium transient were observed after pretreatment with nifedipine, KB-R7943, AMG 9810 or Synta-66. These results support the involvement of N-type Cav channels, but not L-type Cav channels, NCX, TRPV1 and ORAI1, in the generation of the first phase of the calcium response of sensory neurons to P-CTX-2.

Figure 2
Figure 2:
N-type Cav and ORAI1 channels mediate the P-CTX-2-evoked calcium response in DRG neurons. The antagonist inhibitory effects on the 2 parameters were assessed for each phase using normalized data; the averaged data obtained from independently repeated experiments with the application of P-CTX-2 alone were considered 100%. Normalized percentages of responding cells and amplitude values in responding cells recorded during the first transient (A, B, respectively), the second transient (C, D), or the plateau (E, F) of the calcium signal in response to 10 nM P-CTX-2 alone (control condition, n=26) or after pretreatment with 1 mM EGTA in calcium-free medium (n=9), 1 µM nifedipine (n=11), 1 µM ω-conotoxin GVIA (n=9), 0.5 µM KB-R7943 (n=6), 1 µM AMG 9810 (n=4) or 10 µM Synta-66 (n=3), expressed as the mean±standard error of the mean (SEM). n represents the number of experiments in which 40–100 neuronal cells were included in the analysis. The data were statistically analyzed using 1-way analysis of variance, followed by Dunnett post hoc test for multiple comparisons to control. * P<0.05, ** P<0.01, **** P<0.0001.

The percentage of cells responding to P-CTX-2 with a second transient (Fig. 2C) was significantly reduced in the absence of extracellular Ca2+ (39.5±15.0% of the control value) and virtually abolished by the Synta-66 pretreatment (6.5±4.4%). The other antagonists did not have a significant effect on this parameter. In the responding cells, none of the studied conditions significantly changed the average amplitude of the second Ca2+ transient induced by P-CTX-2 (Fig. 2D). This parameter was nevertheless reduced by almost half in the absence of extracellular Ca2+ (45.7±5.0% of the control amplitude value) and after the Synta-66 pretreatment (54.6±12.9%). These results suggest that a Ca2+ influx supported by ORAI1 and, to a lesser extent, N-type Cav channels may be involved in the generation of the second calcium transient elicited by P-CTX-2 in sensory neurons. The involvement of ORAI1 strongly suggests that SOCE supported by this channel occurs in the calcium signal induced by P-CTX-2.

None of the studied conditions significantly affected the third phase (plateau) of the Ca2+ response induced by P-CTX-2. However, as in the case of the second transient, the absence of extracellular Ca2+ strongly reduced the percentage of cells responding with a plateau (to 25.8±12.0%; Fig. 2E) and the associated amplitude (to 60.6±4.1%; Fig. 2F). The percentage of cells displaying a plateau was also reduced (to 36.3±24.7%) by Synta-66.

Representative traces of the calcium response to P-CTX-2 after each pretreatment condition are shown in Figure 3. In summary, our results show that calcium influx plays a major role in the calcium signal elicited by P-CTX-2 in sensory neurons. Different molecular actors drive the early and late events of this response. N-type Cav channels contribute to its initiation, while the subsequent stages of the calcium signal involve ORAI1-supported SOCE.

Figure 3
Figure 3:
Representative traces of the calcium response of rat DRG neurons to P-CTX-2 after pretreatment for 15 minutes with 1 mM EGTA in calcium-free medium (A), 1 µM nifedipine (B), 1 µM ω-conotoxin GVIA (C), 0.5 µM KB-R7943 (D), 1 µM AMG 9810 (E) or 10 µM Synta-66 (F).

L-type, N-type Cave and ORAI1 channels are crucial molecular effectors of P-CTX-2-induced SP release in the co-culture

We previously reported a novel in vitro model in which co-cultured DRG neurons and primary keratinocytes released SP and CGRP after exposure to P-CTX-217. In the present study, we modified the co-culture model by adding and differentiating keratinocytes after a large neuritis network had been formed in the DRG neuron culture and then treating with P-CTX-2 in a medium containing a higher (i.e., physiological) external calcium concentration. These modifications led to an 8-fold increase in the SP levels released in response to P-CTX-2 compared with those released in the prototypical model45, resulting in SP levels high enough to allow for the robust investigation of the molecular mechanisms involved by using pharmacological antagonists. The selected antagonists included those used in the cytosolic Ca2+ imaging experiments, that is, nifedipine, ω-conotoxin GVIA, KB-R7943, AMG 9810, and Synta-66. In addition, the effects of HC-030031, a TRPA1 blocker, and benzamil hydrochloride, a less specific NCX antagonist than KB-R7943, were also investigated.

The results presented in Figure 4 show that the P-CTX-2-induced SP release was significantly decreased by cell incubation in calcium-free medium (25.1±10.6% compared with P-CTX-2 alone) and Synta-66 pretreatment (17.4±12.3% compared with the control value). Pretreatment with the Cav channel blockers nifedipine and ω-conotoxin GVIA significantly decreased the induced SP release to the same extent (64.0±18.9% and 60.1±13.2%, respectively). The other antagonists used did not significantly change SP release. These data indicate that calcium influx involving L-type and N-type Cav channels and ORAI1-supported SOCE plays a major role in the SP release induced by P-CTX-2 in the co-culture.

Figure 4
Figure 4:
L-type, N-type Cav and ORAI1 channels are major molecular players in P-CTX-2-induced substance P (SP) release in the co-culture. Effect of selected antagonists on P-CTX-2-evoked SP release in the co-culture. Normalized SP levels after P-CTX-2 exposure without (n=23) or with pretreatment with 1 mM EGTA in calcium-free medium (n=6), 1 µM nifedipine (n=10), 1 µM ω-conotoxin GVIA (n=9), 0.5 µM KB-R7943 (n=4), 1 µM benzamil hydrochloride (n=12), 1 µM AMG 9810 (n=15), 30 µM HC-030031 (n=16) or 10 µM Synta-66 (n=4) were statistically analyzed using 1-way analysis of variance with Dunnett post hoc test for multiple comparisons to control. P-CTX-2 (10 nM) was applied for 90 minutes with or without preatreatment with pharmacological antagonists. The SP levels were measured in the supernatants by enzyme immunoassay assay. The values obtained under the vehicle control conditions (MeOH 0.05%) were subtracted from those obtained under the 10 nM P-CTX-2 conditions, and then, the average SP levels in pg/mL were normalized. The normalized values are expressed as the mean±SEM, and n is the number of experiments. * P<0.05, ** P<0.01.


In this study, we used in vitro models to explore the molecular mechanisms involved in 2 sensory effects of CTXs. First, single-cell Ca2+ imaging data show that P-CTX-2 induced a multiphasic and long-lasting calcium response in neonatal rat DRG neurons, with 50.6% of neuronal cells responding to P-CTX-2. In a previous work18, 51% of adult mouse DRG neurons responded to P-CTX-1, which suggests a similar proportion of responding neurons regardless of rodent species or age, and of the CTX congener. Adult mouse DRG neurons responding to P-CTX-1 were A- and C-type, mostly medium sized, and virtually all CGRP-and TRPA1-positive. It is not known whether neonatal rat neurons responding to P-CTX-2 have the same features because we have not characterized their size and immunochemical profile. It cannot be excluded that the molecular mechanisms involved differ depending on the stage of neuronal maturation. The phenotypic profile of DRG neurons indeed evolves between the postnatal and adult periods, especially in the subpopulation binding isolectin B454, which is poorly targeted by CTXs in adult sensory neurons18. After pretreatment with specific antagonists, a separate analysis of their effects on each phase of the P-CTX-2-elicited calcium response allowed us to determine the phase(s) in which a molecular target was involved. Because each phase was strikingly inhibited in the absence of extracellular Ca2+, calcium entry appears to be required, at least for the initiation of the calcium response.

Second, we refined our prototypical co-culture model17 to optimize P-CTX-2-elicited SP release, allowing a better exploration of the mechanisms involved. In this model, our results obtained in calcium-free medium confirm the previously suggested critical involvement of calcium influx in the SP release elicited by CTXs17,20.

The role of N-type and L-type Cav channels

We investigated whether certain HVA Cav channels drive the calcium influx involved in these effects. Although the P/Q-, R- and T-type Cav subtypes are also expressed in DRG neurons55–57, we focused on the L-type (Cav1) and N-type (Cav2.2) subtypes due to their known involvement in neuropeptide release from peripheral or spinal DRG afferents33–37.

Our results indicate that N-type Cav channels significantly contribute to the initiation of the P-CTX-2-evoked calcium response in sensory neurons. This finding suggests that CTXs, which activate Nav channels and inhibit Kv channels in these cells14,16, induce a plasma membrane depolarization strong enough to activate N-type Cav channels. Our results show that N-type Cav channels are involved in the P-CTX-2-evoked SP release. This is consistent with the expression of these subtypes in peptidergic sensory afferents55,58 and previous works showing that N-type Cav channels are involved in the release of SP from both peripheral35,59 and terminal33,36 sensory afferents.

Although L-type Cav channels are weakly involved in the P-CTX-2-evoked calcium response, they appear to contribute equally with N-type Cav channels to the SP release. This discrepancy could be related to the different neuronal subpopulations studied. L-type Cav channels are mainly expressed in small-sized neurons, which include the SP-expressing subpopulation, but also in larger neurons60. The L-type Cav channel contribution in peptidergic neurons, the only cellular actors in the induced SP release, may have gone unnoticed when analyzing the calcium response of the whole DRG population.

In SH-SY5Y neuroblastoma cells, P-CTX-1-induced calcium responses involved N-type and, to a greater extent, L-type Cav channels19. The different contribution of these Cav subtypes could stem from differential expression and functionality of Cav subtypes in SH-SY5Y cells compared with DRG neurons.

L-type and T-type Cav antagonists did not significantly reduce the P-CTX-1-induced CGRP release in adult mouse skin flaps20. The role of N-type Cav channels was not explored. Thus, our study is the first to identify N-type and L-type Cav channels as molecular effectors of neuropeptide release induced by a CTX.

Nifedipine was used in one patient suffering from CFP, with limited success on sensory disturbances61. In contrast, gabapentinoids rapidly improved paresthesia, cold dysesthesia and pruritus in 4 CFP patients, including disturbances that persisted for several weeks after the toxic meal62,63. Gabapentinoids inhibit HVA calcium currents, especially N-type64, by binding to the α2δ subunits of HVA Cav channels65. Gabapentin and pregabalin have also been effective in the treatment of pruritus refractory to antihistamines66,67. No specific N-type Cav blocker has been used in humans or experimental animals exposed to CTXs. Among such blockers, ziconotide has been approved for the clinical treatment of severe chronic pain, and leconotide (also known as AM336 and CVID) is promising36,68. Spinal N-type Cav antagonism was effective in preventing scratching behavior in several in vivo pruritus models69. Our results suggest that these inhibitors may be of interest in mitigating the sensory effects occurring in the CFP context.

NCX, TRPV1, and TRPA1 are not involved in the sensory responses to P-CTX-2

NCX did not significantly contribute to the P-CTX-2-induced neuronal calcium response or SP release, in contrast to the acetylcholine release elicited by P-CTX-1 in Torpedo synaptosomes38,39. NCX is mainly expressed by non-peptidergic DRG neurons70 and is therefore unlikely involved in the release of neuropeptides. The nonsignificant decrease by KB-R7943 of the first and second transients of the P-CTX-2-induced calcium response could reflect a regulatory mechanism by NCX occurring in non-peptidergic neurons.

TRPV1 and TRPA1 are widely involved in pain perception and transmission41 and are downstream effectors of a large number of pruritogens40. Our results rule out their involvement in the P-CTX-2-evoked SP release. This is consistent with the unaffected P-CTX-1-induced CGRP release in skin flaps from TRPV1 or TRPA1 knockout mice compared with wild-type mice20. TRPA1 sensitization to cold in peptidergic primary sensory neurons was nevertheless shown to play a pivotal role in the cold allodynia induced by P-CTX-118 and the P-CTX-2-induced hypersensitivity to cold of spinal neurons measured in vivo after subcutaneous injection in rodents48. One explanation for our result is that, in the absence of a cooling stimulus, CTX-induced TRPA1 sensitization may not be sufficiently marked to be involved in the SP release in our in vitro experiments performed at 37°C.

ORAI1 channel is a key effector of CTX effects in sensory neurons

ORAI1 is a calcium channel supporting SOCE, which allows the replenishment of ER calcium stores after depletion but also contributes to cell signaling71,72. ORAI1 plays a pivotal role in the late phase of the calcium signal and subsequent SP release elicited by P-CTX-2 in DRG neurons, suggesting that P-CTX-2 induces SOCE in sensory neurons. SOCE was recently shown to be a major player in calcium signaling in neurons72,73. In DRG neurons, SOCE contributes to maintaining cytosolic calcium concentration and replenishing calcium stores in resting neurons42, and increases excitability44.

Two major mechanisms that drive ER calcium store depletion mediate SOCE in sensory neurons. The first is activation of Gαq/11-protein-coupled receptors, which through phospholipase C leads to inositol-triphosphate receptor (IP3R)-dependent store depletion74,75. In trigeminal sensory neurons, bradykinin elicits ORAI1-supported SOCE through such a pathway43. The second is the process termed calcium-induced calcium release (CICR), which depletes ryanodine-dependent stores76. Cav channel-driven CICR has been well characterized in sensory neurons77,78 and we demonstrate that Cav channels are involved in the P-CTX-2-induced [Ca2+]i increase. Thus, CICR appears to be a plausible mechanism linking Cav channel-induced [Ca2+]i increase and the ER calcium store depletion that drives ORAI1-supported SOCE. However, the contribution of IP3Rs cannot be excluded. Indeed, IP3Rs are expressed in sensory neurons79, and IP3R-dependent calcium stores have been previously involved in calcium responses to P-CTX-1 in neuronal cell lines and myotubes80,81.

Our results reveal that ORAI1, as a critical molecular effector of the late phase of the calcium signal and the subsequent SP release elicited by P-CTX-2 in DRG neurons, is a potential target for treating SP-related sensory disturbances occurring during CFP. To our knowledge, this is the first report involving ORAI1 in neuropeptide release from mammalian DRG neurons. Because SP released from these neurons is involved in neurogenic inflammation and spinal pain and itch transmission, our results suggest that ORAI1-supported SOCE could be targeted to dampen these processes.


In this study, we confirm the critical involvement of calcium influx in P-CTX-2-elicited SP release. We demonstrate that N-type Cav channels contribute to the initiation of the calcium response to P-CTX-2 in the whole DRG neuron population, while N-type and L-type Cav channels equally contribute to the SP release elicited in peptidergic neurons. In contrast, NCX, TRPV1 and TRPA1 do not participate in the calcium response or the SP release evoked by P-CTX-2. We are the first to show that ORAI1-driven SOCE mediates neuropeptide release from mammalian DRG neurons and is a signaling molecule in CTX sensory effects. This key component of SOCE critically controls the late phase of the calcium signal and the subsequent SP release elicited by P-CTX-2. Because peptidergic neurons are major players in CFP sensory disturbances, identifying the critical molecular actors in the SP release induced by a CTX is of major therapeutic interest. While CTX availability limits in vivo experiments, our in vitro findings highlight the Cav and ORAI1 channels as promising pharmacological targets for specifically relieving the sensory effects of CTXs.

Conflict of interest statement

The authors declare that they have no financial conflict of interest with regard to the content of this report.


The authors thank the French Society of Dermatology (Paris, France) for financial support for this study.


1. Isbister GK, Kiernan MC. Neurotoxic marine poisoning. Lancet Neurol 2005;4:219–28.
2. Stinn JF, De Sylva DP, Fleming LE, et al. Geographical information systems and ciguatera fish poisoning in the tropical Western Atlantic region. Proceedings of the 1998 Geographic Information Systems in Public Health Conference, Third National Conference. 2000:223–33.
3. Friedman MA, Fernandez M, Backer LC, et al. An updated review of Ciguatera fish poisoning: clinical, epidemiological, environmental, and public health management. Mar Drugs 2017;15:72. doi: 10.3390/md15030072.
4. Banner AH, Shaw SW, Alender CB, et al. Fish Intoxication; notes on Ciguatera, Its mode of action and a suggested therapy. South Pacific Commission; 1963.
5. Russell FE. Ciguatera poisoning: a report of 35 cases. Toxicon 1975;13:383–5.
6. Zimmermann K, Eisenblätter A, Vetter I, et al. Vergiftung durch Tropenfisch: Ciguatera-Epidemie in Deutschland [Imported tropical fish causes ciguatera fish poisoning in Germany]. DMW. Dtsch Med Wochenschr 2015;140:125–30.
7. Lawrence DN, Enriquez MB, Lumish RM, et al. Ciguatera fish poisoning in Miami. JAMA 1980;244:254–8.
8. Gillespie NC, Lewis RJ, Pearn JH, et al. Ciguatera in Australia. Occurrence, clinical features, pathophysiology and management. Med J Aust 1986;145:584–90.
9. Bagnis R, Legrand A-MGopalkrishnakone P, Tan CK. Clinical features on 12,890 cases of ciguatera (fish poisoning) in French Polynesia. Progress in Venom and Toxin Research: Proceedings of the First Asia-Pacific Congress on Animal, Plant and Microbial Toxins Held in Singapore, June 24–27, 1987. Singapore: University of Singapore; 1987:372–84.
10. Frenette C, MacLean JD, Gyorkos TW. A large common-source outbreak of ciguatera fish poisoning. J Infect Dis 1988;158:1128–31.
11. Katz AR, Terrell-Perica S, Sasaki DM. Ciguatera on Kauai: investigation of factors associated with severity of illness. Am J Trop Med Hyg 1993;49:448–54.
12. Chateau-Degat M-L, Huin-Blondey M-O, Chinain M, et al. Prevalence of chronic symptoms of ciguatera disease in French Polynesian adults. Am J Trop Med Hyg 2007;77:842–6.
13. Baumann F, Bourrat M-B, Pauillac S. Prevalence, symptoms and chronicity of ciguatera in New Caledonia: results from an adult population survey conducted in Noumea during 2005. Toxicon 2010;56:662–7.
14. Strachan LC, Lewis RJ, Nicholson GM. Differential actions of pacific ciguatoxin-1 on sodium channel subtypes in mammalian sensory neurons. J Pharmacol Exp Ther 1999;288:379–88.
15. Inserra MC, Israel MR, Caldwell A, et al. Multiple sodium channel isoforms mediate the pathological effects of Pacific ciguatoxin-1. Sci Rep 2017;7:42810.
16. Birinyi-Strachan LC, Gunning SJ, Lewis RJ, et al. Block of voltage-gated potassium channels by Pacific ciguatoxin-1 contributes to increased neuronal excitability in rat sensory neurons. Toxicol Appl Pharmacol 2005;204:175–86.
17. Le Garrec R, L’herondelle K, Le Gall-Ianotto C, et al. Release of neuropeptides from a neuro-cutaneous co-culture model: a novel in vitro model for studying sensory effects of ciguatoxins. Toxicon 2016;116:4–10.
18. Vetter I, Touska F, Hess A, et al. Ciguatoxins activate specific cold pain pathways to elicit burning pain from cooling. EMBO J 2012;31:3795–808.
19. Zimmermann K, Deuis JR, Inserra MC, et al. Analgesic treatment of ciguatoxin-induced cold allodynia. Pain 2013;154:1999–2006.
20. Touska F, Sattler S, Malsch P, et al. Ciguatoxins evoke potent CGRP release by activation of voltage-gated sodium channel subtypes NaV1.9, NaV1.7 and NaV1.1. Mar Drugs 2017;15:269.
21. Gouin O, L’Herondelle K, Lebonvallet N, et al. TRPV1 and TRPA1 in cutaneous neurogenic and chronic inflammation: pro-inflammatory response induced by their activation and their sensitization. Protein Cell 2017;8:644–661.
22. Andoh T, Nagasawa T, Satoh M, et al. Substance P induction of itch-associated response mediated by cutaneous NK1 tachykinin receptors in mice. J Pharmacol Exp Ther 1998;286:1140–5.
23. Ständer S, Siepmann D, Herrgott I, et al. Targeting the neurokinin receptor 1 with aprepitant: a novel antipruritic strategy. PloS One 2010;5:e10968.
24. Azimi E, Reddy VB, Pereira PJS, et al. Substance P activates Mas-related G protein-coupled receptors to induce itch. J Allergy Clin Immunol 2017;140:447–53.e3.
25. Nichols ML, Allen BJ, Rogers SD, et al. Transmission of chronic nociception by spinal neurons expressing the substance P receptor. Science 1999;286:1558–61.
26. Gautam M, Prasoon P, Kumar R, et al. Role of neurokinin type 1 receptor in nociception at the periphery and the spinal level in the rat. Spinal Cord 2016;54:172–82.
27. Rogoz K, Andersen HH, Lagerström MC, et al. Multimodal use of calcitonin gene-related peptide and substance P in itch and acute pain uncovered by the elimination of vesicular glutamate transporter 2 from transient receptor potential cation channel subfamily V member 1 neurons. J Neurosci 2014;34:14055–68.
28. Carstens EE, Carstens MI, Simons CT, et al. Dorsal horn neurons expressing NK-1 receptors mediate scratching in rats. Neuroreport 2010;21:303–8.
29. Akiyama T, Tominaga M, Davoodi A, et al. Roles for substance P and gastrin-releasing peptide as neurotransmitters released by primary afferent pruriceptors. J Neurophysiol 2013;109:742–8.
30. Green DP, Limjunyawong N, Gour N, et al. A mast-cell-specific receptor mediates neurogenic inflammation and pain. Neuron 2019;101:412–20.e3.
31. Huang LY, Neher E. Ca(2+)-dependent exocytosis in the somata of dorsal root ganglion neurons. Neuron 1996;17:135–45.
32. Zupanc GK. Peptidergic transmission: from morphological correlates to functional implications. Micron Oxf Engl 1993 1996;27:35–91.
33. Santicioli P, Del Bianco E, Tramontana M, et al. Release of calcitonin gene-related peptide like-immunoreactivity induced by electrical field stimulation from rat spinal afferents is mediated by conotoxin-sensitive calcium channels. Neurosci Lett 1992;136:161–4.
34. Evans AR, Nicol GD, Vasko MR. Differential regulation of evoked peptide release by voltage-sensitive calcium channels in rat sensory neurons. Brain Res 1996;712:265–73.
35. Kress M, Izydorczyk I, Kuhn A. N- and L- but not P/Q-type calcium channels contribute to neuropeptide release from rat skin in vitro. Neuroreport 2001;12:867–70.
36. Smith MT, Cabot PJ, Ross FB, et al. The novel N-type calcium channel blocker, AM336, produces potent dose-dependent antinociception after intrathecal dosing in rats and inhibits substance P release in rat spinal cord slices. Pain 2002;96:119–27.
37. Takasusuki T, Yaksh TL. Regulation of spinal substance p release by intrathecal calcium channel blockade. Anesthesiology 2011;115:153–64.
38. Molgó J, Gaudry-Talarmain YM, Legrand AM, et al. Ciguatoxin extracted from poisonous moray eels Gymnothorax javanicus triggers acetylcholine release from Torpedo cholinergic synaptosomes via reversed Na(+)-Ca2+ exchange. Neurosci Lett 1993;160:65–68.
39. Gaudry-Talarmain YM, Molgo J, Meunier FA, et al. Reversed mode Na(+)-Ca2+ exchange activated by ciguatoxin (CTX-1b) enhances acetylcholine release from Torpedo cholinergic synaptosomes. Ann N Y Acad Sci 1996;779:404–6.
40. Luo J, Feng J, Liu S, et al. Molecular and cellular mechanisms that initiate pain and itch. Cell Mol Life Sci CMLS 2015;72:3201–23.
41. Patapoutian A, Tate S, Woolf CJ. Transient receptor potential channels: targeting pain at the source. Nat Rev Drug Discov 2009;8:55–68.
42. Gemes G, Bangaru MLY, Wu H-E, et al. Store-operated Ca2+ entry in sensory neurons: functional role and the effect of painful nerve injury. J Neurosci 2011;31:3536–49.
43. Szteyn K, Gomez R, Berg KA, et al. Divergence in endothelin-1- and bradykinin-activated store-operated calcium entry in afferent sensory neurons. ASN Neuro 2015;7:1759091415578714.
44. Wei D, Mei Y, Xia J, et al. Orai1 and Orai3 mediate store-operated calcium entry contributing to neuronal excitability in dorsal root ganglion neurons. Front Cell Neurosci 2017;11:400.
45. L’Herondelle K, Pierre O, Fouyet S, et al. PAR2, keratinocytes and cathepsin S mediate the sensory effects of ciguatoxins responsible for ciguatera poisoning. J Invest Dermatol 2020. In Press.
46. Le Gall-Ianotto C, Andres E, Hurtado SP, et al. Characterization of the first coculture between human primary keratinocytes and the dorsal root ganglion-derived neuronal cell line F-11. Neuroscience 2012;210:47–57.
47. Lewis RJ, Sellin M, Poli MA, et al. Purification and characterization of ciguatoxins from moray eel (Lycodontis javanicus, Muraenidae). Toxicon 1991;29:1115–27.
48. Patel R, Brice NL, Lewis RJ, et al. Ionic mechanisms of spinal neuronal cold hypersensitivity in ciguatera. Eur J Neurosci 2015;42:3004–11.
49. Chéret J, Lebonvallet N, Buhé V, et al. Influence of sensory neuropeptides on human cutaneous wound healing process. J Dermatol Sci 2014;74:193–203.
50. Mohammed ZA, Doran C, Grundy D, et al. Veratridine produces distinct calcium response profiles in mouse dorsal root ganglia neurons. Sci Rep 2017;7:45221.
51. Amran MS, Homma N, Hashimoto K. Pharmacology of KB-R7943: a Na+-Ca2+ exchange inhibitor. Cardiovasc Drug Rev 2003;21:255–76.
52. Gavva NR, Tamir R, Qu Y, et al. AMG 9810 [(E)-3-(4-t-Butylphenyl)-N-(2,3-dihydrobenzo[b][1,4] dioxin-6-yl)acrylamide], a novel vanilloid receptor 1 (TRPV1) antagonist with antihyperalgesic properties. J Pharmacol Exp Ther 2005;313:474–84.
53. Di Sabatino A, Rovedatti L, Kaur R, et al. Targeting gut T cell Ca2+ release-activated Ca2+ channels inhibits T cell cytokine production and T-box transcription factor T-bet in inflammatory bowel disease. J Immunol Baltim Md 1950 2009;183:3454–62.
54. Luo W, Wickramasinghe SR, Savitt JM, et al. A hierarchical NGF signaling cascade controls Ret-dependent and Ret-independent events during development of nonpeptidergic DRG neurons. Neuron 2007;54:739–54.
55. Westenbroek RE, Hoskins L, Catterall WA. Localization of Ca2+ channel subtypes on rat spinal motor neurons, interneurons, and nerve terminals. J Neurosci 1998;18:6319–30.
56. Fang Z, Hwang JH, Kim JS, et al. R-type calcium channel isoform in rat dorsal root ganglion neurons. Korean J Physiol 2010;14:45–49.
57. Rose KE, Lunardi N, Boscolo A, et al. Immunohistological demonstration of CaV3.2 T-type voltage-gated calcium channel expression in soma of dorsal root ganglion neurons and peripheral axons of rat and mouse. Neuroscience 2013;250:263–74.
58. Nieto-Rostro M, Ramgoolam K, Pratt WS, et al. Ablation of α 2 δ-1 inhibits cell-surface trafficking of endogenous N-type calcium channels in the pain pathway in vivo. Proc Natl Acad Sci 2018;115:E12043–52.
59. Maggi CA, Tramontana M, Cecconi R, et al. Neurochemical evidence for the involvement of N-type calcium channels in transmitter secretion from peripheral endings of sensory nerves in guinea pigs. Neurosci Lett 1990;114:203–6.
60. Scroggs RS, Fox AP. Calcium current variation between acutely isolated adult rat dorsal root ganglion neurons of different size. J Physiol 1992;445:639–58.
61. Calvert GM, Hryhorczuk DO, Leikin JB. Treatment of ciguatera fish poisoning with amitriptyline and nifedipine. J Toxicol Clin Toxicol 1987;25:423–8.
62. Perez CM, Vasquez PA, Perret CF. Treatment of ciguatera poisoning with gabapentin. N Engl J Med 2001;344:692–3.
63. Brett J, Murnion B. Pregabalin to treat ciguatera fish poisoning. Clin Toxicol Phila Pa 2015;53:588.
64. Sutton KG, Martin DJ, Pinnock RD, et al. Gabapentin inhibits high-threshold calcium channel currents in cultured rat dorsal root ganglion neurones. Br J Pharmacol 2002;135:257–65.
65. Alles SRA, Smith PA. Etiology and pharmacology of neuropathic pain. Pharmacol Rev 2018;70:315–47.
66. Matsuda KM, Sharma D, Schonfeld AR, et al. Gabapentin and pregabalin for the treatment of chronic pruritus. J Am Acad Dermatol 2016;75:619–25.e6.
67. Akiyama T, Andoh T, Ohtsuka E, et al. Peripheral gabapentin regulates mosquito allergy-induced itch in mice. Eur J Pharmacol 2018;833:44–49.
68. Patel R, Montagut-Bordas C, Dickenson AH. Calcium channel modulation as a target in chronic pain control. Br J Pharmacol 2018;175:2173–84.
69. Maciel IS, Azevedo VM, Pereira TC, et al. The spinal inhibition of N-type voltage-gated calcium channels selectively prevents scratching behavior in mice. Neuroscience 2014;277:794–805.
70. Scheff NN, Yilmaz E, Gold MS. The properties, distribution and function of Na+–Ca2+ exchanger isoforms in rat cutaneous sensory neurons. J Physiol 2014;592:4969–93.
71. Lewis RS. The molecular choreography of a store-operated calcium channel. Nature 2007;446:284–7.
72. Majewski L, Kuznicki J. SOCE in neurons: signaling or just refilling? Biochim Biophys Acta 2015;1853:1940–52.
73. Putney JW. Capacitative calcium entry in the nervous system. Cell Calcium 2003;34:339–44.
74. Putney JW. TRP, inositol 1,4,5-trisphosphate receptors, and capacitative calcium entry. Proc Natl Acad Sci U S A 1999;96:14669–71.
75. Taylor CW, Machaca K. IP3 receptors and store-operated Ca2+ entry: a license to fill. Curr Opin Cell Biol 2019;57:1–7.
76. Usachev YM, Thayer SA. Ca2+ influx in resting rat sensory neurones that regulates and is regulated by ryanodine-sensitive Ca2+ stores. J Physiol 1999;519(pt 1):115–30.
77. Usachev YM, Thayer SA. All-or-none Ca2+ release from intracellular stores triggered by Ca2+ influx through voltage-gated Ca2+ channels in rat sensory neurons. J Neurosci 1997;17:7404–14.
78. Solovyova N, Veselovsky N, Toescu EC, et al. Ca(2+) dynamics in the lumen of the endoplasmic reticulum in sensory neurons: direct visualization of Ca(2+)-induced Ca(2+) release triggered by physiological Ca(2+) entry. EMBO J 2002;21:622–30.
79. Solovyova N, Verkhratsky A. Neuronal endoplasmic reticulum acts as a single functional Ca2+ store shared by ryanodine and inositol-1,4,5-trisphosphate receptors as revealed by intra-ER [Ca2+] recordings in single rat sensory neurones. Pflugers Arch 2003;446:447–54.
80. Molgó J, Shimahara T, Legrand AM. Ciguatoxin, extracted from poisonous morays eels, causes sodium-dependent calcium mobilization in NG108-15 neuroblastoma x glioma hybrid cells. Neurosci Lett 1993;158:147–50.
81. Hidalgo J, Liberona JL, Molgó J, et al. Pacific ciguatoxin-1b effect over Na+ and K+ currents, inositol 1,4,5-triphosphate content and intracellular Ca2+ signals in cultured rat myotubes. Br J Pharmacol 2002;137:1055–62.

Ciguatoxin; Ciguatera; Sensory; Itch; Substance P; Calcium

Copyright © 2020 The Authors. Published by Wolters Kluwer Health, Inc. on behalf of The International Forum for the Study of Itch.