Abstract
Oxytocinergic transmission blocks nociception at the peripheral, spinal, and supraspinal levels through the oxytocin receptor (OTR). Indeed, a neuronal pathway from the hypothalamic paraventricular nucleus (PVN) to the spinal cord and trigeminal nucleus caudalis (Sp5c) has been described. Hence, although the trigeminocervical complex (TCC), an anatomical area spanning the Sp5c, C1, and C2 regions, plays a role in some pain disorders associated with craniofacial structures (e.g., migraine), the role of oxytocinergic transmission in modulating nociception at this level has been poorly explored. Hence, in vivo electrophysiological recordings of TCC wide dynamic range (WDR) cells sensitive to stimulation of the periorbital or meningeal region were performed in male Wistar rats. PVN electrical stimulation diminished the neuronal firing evoked by periorbital or meningeal electrical stimulation; this inhibition was reversed by OTR antagonists administered locally. Accordingly, neuronal projections (using Fluoro-Ruby) from the PVN to the WDR cells filled with Neurobiotin were observed. Moreover, colocalization between OTR and calcitonin gene–related peptide (CGRP) or OTR and GABA was found near Neurobiotin-filled WDR cells. Retrograde neuronal tracers deposited at the meningeal (True-Blue, TB) and infraorbital nerves (Fluoro-Gold, FG) showed that at the trigeminal ganglion (TG), some cells were immunopositive to both fluorophores, suggesting that some TG cells send projections via the V1 and V2 trigeminal branches. Together, these data may imply that endogenous oxytocinergic transmission inhibits the nociceptive activity of second-order neurons via OTR activation in CGRPergic (primary afferent fibers) and GABAergic cells.
Significance Statement
This study sheds light on the mechanisms involved in the regulation of trigeminal nociception, which is crucial for understanding the pathophysiology of primary headaches, such as migraine. Current evidence suggests that the hypothalamus plays a role in controlling the nociceptive activity at the trigeminal level. The present study found that electrical stimulation of the hypothalamic paraventricular nucleus (PVN) inhibited the nociceptive activity of trigeminal second-order wide dynamic range cells through oxytocinergic mechanisms. Furthermore, we found that oxytocin receptors (OTRs) are located in the primary peptidergic afferent fibers and GABAergic cells. These findings support the idea of a direct oxytocinergic pathway between the PVN and trigeminocervical complex and highlight the potential of OTRs as a target for migraine pain and other primary headaches.
Introduction
In the pathophysiology of migraine (and other primary headache-related disorders), the hypothalamus seems to play a crucial role in modulating the nociception in the trigeminocervical complex (TCC; Alstadhaug, 2009; Goadsby et al., 2017; May and Burstein, 2019). The TCC is an anatomical area spanning the trigeminal nucleus pars caudalis (Sp5c) and upper cervical region (C1 and C2), where the diencephalic and brainstem structures modulate via top–down the incoming peripheral nociception (Akerman et al., 2011). In this regard, human imaging studies suggest that the hypothalamus may play a role not only in premonitory symptoms (e.g., food craving) but also in the initiation of migraine attacks (Denuelle et al., 2007; Maniyar et al., 2014; Schulte and May, 2016; Schulte et al., 2020).
Accordingly, the hypothalamus contains a plethora of neuromediators and neuroanatomical connections to “painful” modulatory systems (Mayanagi et al., 1978; Bartsch et al., 2005; Condés-Lara et al., 2015; May and Burstein, 2019). Hence, an imbalance in these modulatory systems affects the sensory processing in the TCC (Goadsby et al., 2017; Stankewitz et al., 2021). Incidentally, in addition to the hypothalamic neuropeptides, such as orexins, somatostatin, neuropeptide Y, and pituitary adenylate cyclase–activating protein (Akerman et al., 2011; Strother et al., 2018), oxytocin has recently been proposed to act at the TCC level (García-Boll et al., 2018; Strother et al., 2018; González-Hernández et al., 2021; Krause et al., 2021).
Briefly, oxytocin is synthesized in the hypothalamic paraventricular nucleus (PVN), and direct neuronal projections from the PVN to the spinal dorsal horn and Sp5c have been described (Condés-Lara et al., 2007; Abdallah et al., 2013; Robert et al., 2013). Furthermore, using a competitive binding autoradiography in human tissues, the presence of the oxytocin receptor (OTR) in the spinal trigeminal nucleus has been suggested (Freeman et al., 2017). Besides, using an electrophysiological migraine model (García-Boll et al., 2020), it was shown that exogenous oxytocin applied locally inhibits neuronal firing in wide dynamic range (WDR) cells, an effect reversed by an OTR antagonist. Certainly, at the spinal lumbar level, electrical or optogenetic stimulation of the PVN evokes antinociception mediated by the oxytocinergic mechanisms (Miranda-Cardenas et al., 2006; Condés-Lara et al., 2008; Eliava et al., 2016; Poisbeau et al., 2018). Coupled with evidence showing that OTR is expressed in trigeminal ganglion (TG) cells (Tzabazis et al., 2016), a presynaptic inhibitory mechanism in the TCC diminishing the firing of WDR cells is suggested. Nonetheless, this hypothesis has not been directly tested, and it is unknown whether PVN stimulation can block TCC nociception through oxytocinergic mechanisms.
This study examined the influence of endogenous oxytocinergic neurotransmission (from the PVN) on TCC. In vivo extracellular recordings of WDR cells at TCC were used to assess the effects of periorbital nociceptive stimulation before and after PVN stimulation and how OTR antagonist pretreatment altered them. Additionally, an animal model of acute dural nociceptive activation of the trigeminovascular system (an established migraine model; Akerman et al., 2013) was used to assess the effect of PVN stimulation on TCC WDR cells with convergent receptive fields from the meningeal and periorbital regions (areas mainly innervated by the V1 branch of the trigeminal nerve). Immunofluorescence experiments were performed using confocal microscopy to delineate the potential TCC oxytocinergic circuit that modulates trigeminal nociception. These results support the contention that endogenous oxytocinergic transmission reduces nociceptive activity by activating OTR in CGRPergic or GABAergic cells. Preliminary data were presented in abstract form at the 16th European Headache Congress (González-Hernández et al., 2022).
Materials and Methods
Experimental animals and ethical standards
A total of 59 male Wistar rats (260–300 g) from the main bioterium of our institute were housed in the satellite bioterium of our laboratory and used in this study. The animals were housed in pairs in acrylic cages with wood-based bedding and controlled temperature (23 ± 2°C) and humidity (50%) on a 12 h light/dark cycle with food (LabDiet 5001) and tap water ad libitum. The rodents were allowed to acclimatize to their new environment for at least 72 h before being handled for the experiments. The local animal welfare committee approved all the procedures in accordance with the NIH Guide for the Care and Use of Laboratory Animals (80-23, revised in 1996).
General procedures
Surgical procedures for in vivo electrophysiological recording of trigeminal WDR cells and PVN electrical stimulation
These experiments were carried out in a total of 56 rats. The animals were introduced into a chamber and anesthetized with 6% (v/v) sevoflurane (in a mixture of three-fourth N2O and one-fourth O2) delivered through a vaporizer. Then, an intratracheal cannula was inserted for artificial ventilation (72 strokes/min) and to maintain the anesthesia throughout the experiment. Subsequently, the animals were mounted onto a stereotaxic frame (David Kopf Instruments). Next, a craniotomy on the left side of the skull was performed, and a concentric stimulation electrode (stainless steel Teflon–isolated microwire, Ø 279 µm, catalog #7920; A-M Systems) was placed into the PVN (1 MΩ; 7.2 mm AP, 0.3 mm lateral, and 2 mm height, from interaural reference; Paxinos and Watson, 1998). The muscles of the dorsal neck were separated to access the TCC, and cervical (C1) laminectomy was performed to allow access to the caudal brainstem/cervicomedullary junction. The overlying dura was then removed.
Extracellular recordings of WDR cells were performed under 3–3.5% sevoflurane in nonparalyzed animals (García-Boll et al., 2018, 2020). The core body temperature was maintained at 37°C using a circulating water pad. End-tidal CO2 was monitored using a CO2 analyzer (Taema Artema MM206, Artema Medical Group) and maintained between 2.0 and 3.0% by adjusting the stroke volume.
Under these conditions, the effects produced by PVN electrical stimulation on the WDR cell firing elicited by periorbital and/or supratentorial dura mater electrical stimuli were investigated. Moreover, the pharmacological OTR blockade was tested. In animals with meningeal electrical stimulation, a craniotomy (∼3 × 3 mm) near the lambdoid suture on the left side of the skull was performed to expose the transverse sinus (TS). Mineral oil was applied to the dural tissue to avoid desiccation.
These experiments were conducted on 56 rats divided into three main sets (n = 24, 20, and 12 rats); see Protocols 1, 2, and 3 in Experimental design for in vivo electrophysiological recordings. At the end of the experiments, the animals were perfused transcardially to collect tissues for further histological analysis (see below, Histological reconstruction of recording sites, Juxtacellular labeling of trigeminal WDR cells inhibited by PVN electrical simulation, and Neuronal tracing, immunofluorescence, and confocal microscopy).
Extracellular recordings of trigeminal WDR cells
Quartz-insulated Pt-W microelectrodes (4–10 MΩ) or glass micropipettes (5–9 MΩ) were used to record the neuronal activity of WDR cells with input from the periorbital and/or meningeal receptive field (RF). These WDR cells are second-order neurons that receive peripheral input from the non-nociceptive and nociceptive fibers (Akerman et al., 2013). The quartz-insulated Pt-W microelectrodes were mounted in a motorized multichannel microdrive with an integrated preamplifier (Eckhorn System, Thomas Recording), whereas the glass microelectrodes were manipulated using a hydraulic microdrive (model 607, David Kopf Instruments). The microelectrodes were positioned into the dorsolateral segment of the TCC and lowered (400–1,000 µm from the surface) in small steps (2–5 µm/s) to search for single-unit discharges.
A first search was performed for cells responding to gentle tactile stimulation of the periorbital dermatome (mainly innervated by the ophthalmic branch of the trigeminal nerve; V1). The RF was assessed for nonnoxious (brushing) and noxious (pinching) input. Then, to verify that the peripheral input relay was in a WDR cell, we applied an electrical pulse stimulus by two electrodes (27 G) inserted into the periorbital RF. The electrodes were attached to a stimulus isolator unit (PSIU6, Grass Instruments) and connected to an electrical stimulator (S88 Stimulator, Grass Instruments). In the case of electrical stimulation of the dural tissue, a second electrode (bipolar) was placed near TS and superior sagittal sinus (SSS).
The evoked second-order WDR cell activity was amplified ×1,000. In the case of quartz-insulated Pt-W microelectrodes, a differential AC amplifier (model 1700, A-M Systems) was used, whereas, for glass micropipettes, the signals were amplified using a KS-700 amplifier (World Precision Instruments) connected to a P511 series AC preamplifier (Grass Instruments). In the latter case, the assembly allows us to perform the current injections of Neurobiotin to stain the recorded cells (see Juxtacellular labeling of trigeminal WDR cells inhibited by PVN electrical stimulation). The multiunit cluster signals were digitalized, discriminated, and stored on a computer disk using CED hardware and Spike2 v5.15 software (Cambridge Electronic Design). Raw and discriminated signals were fed through an audio monitor and displayed on an oscilloscope (TDS 420A, Tektronix).
Baselines and evoked activities of trigeminal WDR cells were analyzed as peri-stimulus time histograms (PSTHs). Considering the distance between the RF and recording electrode, the spike latencies observed correspond to peripheral conduction velocities within the Aδ- (3–25 ms) and C-fibers (26–100 ms; Levy and Strassman, 2002). On this basis, the threshold to evoke action potentials and their frequency of occurrence, resulting from the stimulation of the periorbital region and/or meninges, were attributed to the recruitment of Aδ- and C-fibers. Thus, the number of action potentials elicited by the 20 RF stimuli was compared before (basal) and after treatment. The basal response was set up after an identified neuron had ≤10% variation in the evoked neuronal responses in at least three consecutive tests.
Experimental design for in vivo electrophysiological recordings
Protocol 1: electrical stimulation of periorbital dermatome and hypothalamic PVN and the effects of spinal OTR antagonist
This set of rats (N = 24) was divided into three groups (n = 8 rats for each group). In one group, the effect of periorbital electrical stimulation upon WDR activity was evaluated before (basal) and after PVN electrical stimuli (at t = 5, 10, 15, and 20 min post-PVN stimuli). The peripheral electrical stimulation consisted of 20 square-wave stimuli at 0.5 Hz with a 1 ms pulse duration at 0.1–1.0 mA to evoke an Aδ- and C-fiber response. As previously established for the spinal cord (Yirmiya et al., 1990; Miranda-Cardenas et al., 2006; Condés-Lara et al., 2015), the PVN was electrically stimulated to induce antinociception using a train of 6 s at 60 Hz, pulse duration of 1 ms, and 300 µA.
A similar procedure was done in another group of rats pretreated with
Protocol 2: electrical stimulation of periorbital/meningeal dermatome and hypothalamic PVN
This experimental part (N = 20 rats) was completed using a well-known migraine pain model to evaluate potential drugs with antimigraine actions (Bergerot et al., 2006; Akerman et al., 2013). The glass microelectrodes were filled with 1 M KCl solution of 4% pontamine sky blue to stain the location of the cells recorded or 2% Neurobiotin (in NaCl 0.9%) to identify the recorded cells by juxtacellular injections (see Juxtacellular labeling of trigeminal WDR cells inhibited by PVN electrical stimulation for details). Moreover, four animals previously injected at the PVN with the anterograde tracer Fluoro-Ruby (FR; see Neuronal tracing, immunofluorescence, and confocal microscopy for details) were also used.
As described above, a RF was identified in the periorbital region. Then, the effect of periorbital electrical stimulation on WDR cell responses was evaluated before (basal), immediately after PVN electrical stimulation (t = 0), and every 5 min until the recovery or stabilization of neuronal-evoked responses. After the periorbital nerve activity was recovered, convergent meningeal input at the trigeminal WDR cell was assessed by electrical stimulation of the dural tissue. The effect of meningeal electrical stimulation on the WDR cell response was also evaluated before (basal), immediately after PVN electrical stimulation (t = 0), and every 5 min until the recovery or stabilization of neuronal responses. The meningeal electrical stimulation consisted of 20 square-wave stimuli (1 Hz; 1 ms pulse; at 0.1–1.0 mA), whereas the PVN electrical stimulation was given with the abovementioned parameters.
Protocol 3: effect of pharmacological blockade of OTR at TCC on the acute dural nociceptive activation of the trigeminovascular system
In N = 12 rats with localized WDR cells responding to meningeal electrical stimulation, the neuronal activity related to activation of Aδ- and C-fibers was assessed before and after PVN stimuli with or without OTR blockade using L-368,899 (10−5 M, 20 µl; pKi: 7.6 ± 1.8; 40-fold selectivity over V1A receptors; Williams et al., 1994; Jasper et al., 1995). This antagonist was topically delivered using a Hamilton syringe at the TCC 10 min before PVN stimulation. The PVN and dural electrical stimulation parameters used are described above. The ongoing spontaneous neuronal activity (spikes/second, Hz) was recorded continuously for 40 s preceding the dural stimulation.
Histological reconstruction of recording sites
At the end of the recordings, the location of the stimulating electrode into the PVN was labeled with an electrolytic lesion (100 µA, 5 s). In the case of WDR cells recorded with glass micropipettes filled with pontamine, the position was iontophoretically marked using a cathodic current (15–20 µA, 30 min). Lastly, the animals received an overdose of pentobarbital and were perfused through the ascending aorta with isotonic saline solution (0.9% NaCl), followed by 10% formaldehyde (≈200 ml each). The brain and the TCC region were removed and postfixed in 10% formaldehyde. Coronal sections (40 µm) were obtained using a freezing microtome, and the electrolytic and microinjection sites were verified using a light microscope (Leica SM2000R, Leica Biosystems). In the case of glass microelectrodes filled with Neurobiotin, labeling of WDR cells was performed before perfusion (see below, Juxtacellular labeling of trigeminal WDR cells inhibited by PVN electrical stimulation).
Juxtacellular labeling of trigeminal WDR cells inhibited by PVN electrical stimulation
Neurobiotin was used to label some electrophysiological characterized trigeminal convergent WDR cells inhibited by PVN electrical stimulation. In this case, at the end of the extracellular recordings of the WDR cells, anodal current (2 s ON, 2 s OFF, at 3–6 nA) was passed for 10–15 min through the glass micropipette filled with Neurobiotin. The current injection was adjusted for each cell recorded to produce bursts of action potentials; alternatively, the peripheral RF was electrically stimulated to maintain neuronal firing of the WDR cell. At the end of the labeling process (≈2 h), the animals received an overdose of sevoflurane and were perfused through the ascending aorta with a saline solution (0.9% NaCl), followed by a fixing solution [4% paraformaldehyde in PBS (0.1 M; PH, 7.4); ≈250 ml each]. The brain, brainstem, and C1–C2 spinal cord levels were collected, postfixed for 2 h, and stored in a cryoprotectant solution (30% sucrose in PBS).
Neuronal tracing, immunofluorescence, and confocal microscopy
PVN stereotaxic microinjection of the anterograde tracer FR
The rats (n = 4) were anesthetized by the intraperitoneal injection of ketamine plus xylazine (70/6 mg/kg) and mounted in a stereotaxic apparatus. Next, the skull was exposed, the periosteum was removed, and a craniotomy (using a dental drill) on the left side of the skull was performed to position a glass micropipette (≈10 µm Ø tips) into the PVN (7.2 mm AP interaural, 0.3 mm lateral, and 2 mm height). The micropipettes were filled with an isotonic saline solution of 10% tetramethylrhodamine 10,000 MW lysine-fixable dextran (n = 4; FR catalog #D1817, Invitrogen, Thermo Fisher Scientific). A total volume of 60 nl was microinjected. Then, the surgical wounds were sutured. Each animal was kept in an individual cage for 10 d to allow axonal transport from the PVN to the TCC region, reanesthetized with pentobarbital (45 mg/kg), and perfused with isotonic saline solution (≈200 ml) followed by 4% paraformaldehyde in PBS 0.1 M (≈300 ml; pH 7.4). The brain was removed, kept in the same fixative solution for 2 h, and cryoprotected in a 30% sucrose solution at 4°C. Two days later, 40-µm-thick coronal sections were obtained to verify the FR injection site and the FR–positive fibers in the TCC region.
Microinjection of retrograde tracers FG or TB in the peripheral trigeminal branches (V1 and V2 region)
The rats (n = 3) were anesthetized by intraperitoneal injection of ketamine plus xylazine (70/6 mg/kg) and mounted in a stereotaxic apparatus. A craniotomy (on the left side of the skull) near the lambda suture and an incision (≈7 mm) below the infraorbital ridge to expose the infraorbital nerve were performed. Under the microscopic control, the tracer was placed on the meninges using a stainless steel wire charged with tiny pallets of TB attached to its blunt tip. In the case of the infraorbital nerve, micropipettes filled with a solution of FG (10%) were carefully injected into the nerve. Then, the surgical wounds were sutured. Each animal was kept in an individual cage for 10 d to allow axonal transport from the peripheral trigeminal region to the TG, reanesthetized with pentobarbital (45 mg/kg), and perfused with isotonic saline solution (≈200 ml) followed by 4% paraformaldehyde in PBS 0.1 M (≈300 ml; pH 7.4). The TG was removed and kept in the same fixative solution for 2 h and cryoprotected in a 30% sucrose solution at 4°C. Two days later, 40-µm-thick coronal sections were obtained to analyze the colocalization of TB with FG.
Tissue processing for immunofluorescence
To study the relationship between oxytocinergic projections from the PVN with some relevant neuromediators at TCC (i.e., CGRP, GABA, and OTR), the PVN region and TCC were washed with PBS (0.1 M, pH 7.4). Later, the slices were incubated for 24 h in PBS with the following antibody cocktails: (1) anti-neurophysin (PS60 mouse monoclonal, 1:250 diluted; a gift from Harold Gainer, National Institutes of Health); (2) anti-CGRP (polyclonal, rabbit, diluted 1: 2,000; catalog #AB15360 Chemicon International); (3) anti-OTR (polyclonal, goat or mouse, diluted 1:400, catalog #sc-8103 or sc-515809; Santa Cruz Biotechnology); and (4) anti-GABA (polyclonal rabbit, 1:1,500 diluted; catalog #A2052; Sigma-Aldrich). The following day, the tissue was washed with PBS for 10 min and then incubated with the secondary antibody cocktail (from Invitrogen): (1) donkey anti-rabbit IgG (Alexa Fluor 555 conjugate, catalog #A31572) or (2) donkey anti-mouse IgG (Alexa Fluor 647 conjugate, catalog #A31571). Finally, the sections were washed for 5 min in PBS, mounted on glass slides, and coverslipped with p-xylene-bis-pyridinium bromide (DPX). It should be noted that the combination of antibodies was selected to avoid cross-species reactivity.
Image analysis
Confocal images of the TCC region were acquired using a Zeiss Laser Scanning Microscopy (model LSM 780, upright configuration) with a motorized stage and Zeiss software V4.2. The pinhole and z-sectioning intervals were kept constant for all the images. About 25–35 optical z-sections of 1 µm thickness were obtained from 40-µm-thick tissue for each image stack. Optical sections were acquired at a digital size of 1,024 × 1,024 pixels and averaged four times to reduce the noise. In all cases, the image obtained was improved (brightness and gamma) and analyzed using the Zen Blue 3.7 software (v3.7.3 Carl Zeiss Microscopy). The color for each channel was selected as follows: (1) green for Neurobiotin; (2) red for FR, CGRP, GABA, or oxytocin; (3) blue for TB, CGRP, or OTR; and (4) yellow for FG. In all cases, a two-dimensional projection image or a single-optical section image was imported in the Tag Image File Format and was used to compose the multi-paneled figures. In one case, as Extended Data, a 3D projection reconstruction using the z-stack file was rendered. In the case of TG tissue, images were acquired using an inverted Olympus IX81 microscope equipped with an OH-ORCA-FLASH4.0 digital camera and fluorescence illumination system.
Data presentation and statistical evaluation
One trigeminal WDR cell was studied per animal. To analyze the impact of an OTR antagonist on PVN-induced antinociception, we normalized the data from the electrophysiological experiments (i.e., the number of evoked potentials) and expressed them as a percent change from the respective baseline (the figures represent the mean ± SEM). In this case, first, the stability of the recorded neurons (only for the control group) across the experiment (time) was verified using a one-way repeated measure analysis of variance (RM ANOVA). The effect of PVN and the OTR antagonist on the neuronal firing of periorbital or trigeminovascular WDR cells was analyzed using a two-way repeated measure analysis of variance (2W-RM ANOVA). In the case of ANOVAs, the sphericity was not assumed, and degrees of freedom were corrected according to the Greenhouse–Geisser method. Furthermore, if applicable, a Tukey's post hoc test was performed. Statistical significance was set at p < 0.05. Data were analyzed using the GraphPad Prism Software (v7; GraphPad Software).
Results
Generalities about electrophysiological data
Figure 1A illustrates the electrophysiological setup used to record trigeminal WDR cells. Extracellular unitary recordings were obtained from 56 WDR neurons. As shown in Figure 1B, the histological analysis showed that recorded cells were situated between laminae III and V in the dorsal horn of the cervicomedullary junction. Comparable findings were obtained using the Thomas RECORDING software data indicating that the microelectrodes were positioned 780 ± 41 µm from the surface. At this level, it has been shown that second-order WDR cells receive concomitant input from non-nociceptive and nociceptive fibers (Burstein et al., 1998; Storer et al., 2004). The cells recorded had an average baseline firing latency after peripheral electrical stimulation between 3 and 25 ms for the Aδ- and 25 and 100 ms for the C-fibers. Figure 1, C and D, depicts the electrophysiological response of a single WDR cell responsive to PVN stimulation (i.e., inhibition of evoked neuronal firing of Aδ- and C-fibers). Among the 56 cells recorded, 30 cells had only a periorbital RF, while 26 cells received convergent input from both the periorbital and meningeal receptive fields. It is worth mentioning that the C-firing discharge was the most sensitive to the inhibition caused by PVN stimulation.
Electrophysiological setup to record trigeminal second-order WDR cells. A, Experimental setup illustrating the site of electrophysiological recording of the WDR cell at the TCC, the PVN stimulating electrode, and the location sites of the stimulating electrodes for periorbital and/or meningeal receptive fields. The OTR antagonist was directly administered to the TCC region. SSS, superior sagittal sinus; TCC, trigeminocervical complex; TG, trigeminal ganglion; TS, transverse sinus. B, Electrophysiological recording sites in the TCC region and histological photomicrography of the electrothermolytic lesion site (white arrow) in the dorsal horn at the cervical C1 level; the locations refer to WDR cells with concomitant RF input from the meningeal and periorbital region. C, Raw tracing of a typical second-order WDR neuron cluster response induced by one periorbital electrical shock before (blue) and after (green) PVN electrical stimulation. D, The corresponding raster plot (top) and PSTH constructed from 20 neuronal responses elicited by periorbital electrical stimulation; the time latencies of Aδ-fiber and C-fiber components are depicted. Note that PVN stimulation diminishes the neuronal response associated with peripheral stimulation.
Effect of PVN electrical stimulation and dOVT (OTR antagonist) on trigeminal WDR cell activity evoked by periorbital electrical stimulation
In this case, 24 WDR cells were electrophysiologically recorded. First, the effect of time on the neuronal responses was analyzed to exclude that no time-dependent changes in neuronal responses occurred during the experimental protocols. Accordingly, in the control group (n = 8), time did not affect Aδ-fiber (F(4, 20.35) = 1.658, p = 0.199) or C-fiber firing (F(4, 12.34) = 1.754, p = 0.201).
Figure 2, A and B, shows the analysis of the neuronal activity related to Aδ- and C-fiber activation elicited by periorbital electrical stimulation in (1) control cells (n = 8); (2) under PVN electrical stimulation (n = 8); and (3) the effect of the OTR antagonist (dOVT, 10−5 M) on PVN stimulation (n = 8). The 2W-RM ANOVA revealed that PVN stimulation reduced the periorbital-evoked neuronal firing of WDR cells (F(2,105) = 11.6, p < 0.0001 for Aδ-fibers and F(2,105) = 98.3, p < 0.0001 for C-fibers) in a time-dependent manner (F(4,105) = 4.10, p = 0.004 for Aδ-fibers and F(4,105) = 25.2, p < 0.0001 for C-fibers). Tukey's post hoc test suggests that the PVN-induced inhibition is prevalent over C-fibers rather than Aδ-fibers. Notably, pretreatment with dOVT (OTR antagonist) at a dose validated to block oxytocin function at the spinal cord level (Miranda-Cardenas et al., 2006) attenuated the PVN-induced inhibition of trigeminal WDR cells.
PVN-induced inhibition of trigeminal nociceptive responses (elicited by periorbital RF stimulation) is blocked by an OTR antagonist. A, B, Time course changes in the percentage average of the Aδ- (left) and C-fiber (right) neuronal firing elicited by periorbital electrical stimulation in a control situation (blue circles) or before (basal) and after PVN electrical stimulation (PVN st) with (red circles) or without (blue circles) OTR antagonist pretreatment (PVN st + dOVT). Note that the PVN-induced inhibition of both cellular components (Aδ- and C-fiber) was reversed when animals were pretreated with dOVT. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs control; §p < 0.05, §§p < 0.01 vs PVN st. C, Histological photomicrograph showing an electrolytic lesion in the rostral part of the PVN (white arrow). D, An Sp5c histological photomicrograph showing the location of the recorded WDR cell using pontamine sky blue (white arrow). E, An example of the raster plots of a single WDR cell sensitive to PVN stimulation obtained before (control) and after PVN electrical stimulation at 5, 10, and 20 min poststimulation; at the bottom, the raw tracing of the WDR neuron cluster response to one peripheral electrical pulse is illustrated.
In Figure 2, C and D, photomicrography of the site of PVN electrical stimulation (an electrolytic lesion) and trigeminal TCC recording are shown. Figure 2E illustrates, in a raster display, the WDR cells' responses before (basal response) and after PVN electrical stimulation (at 5, 10, and 20 min after PVN stimulation); note that the main effect relies upon the unitary cell response's late component (i.e., C-fibers).
Effect of PVN electrical stimulation in an electrophysiological migraine pain model: recordings of convergent trigeminal WDR cells
Extracellular recordings of 20 trigeminal WDR cells were performed. All cells had a periorbital RF, but only 14 cells were inhibited by the PVN electrical stimulation. Notably, these WDR cells recorded, sensitive to PVN electrical stimulation, have concomitant input from the periorbital and meningeal receptive fields (i.e., convergent trigeminal WDR cells; Fig. 1B).
In this regard, Figure 3, A and B, illustrates the impact of PVN electrical stimulation upon the evoked neuronal firing of a single convergent WDR trigeminal cell (Fig. 3C). Note that PVN electrical stimulation suppressed neuronal firing associated with electrical activation of the Aδ- and C-fibers (i.e., the stimuli at the periorbital or meningeal RF); this inhibition persisted for >5 min. At the end of this experimental protocol, some convergent WDR cells (n = 9) were injected with Neurobiotin. Figure 3D shows the current injection's effect on the WDR cell's activity, whereas Figure 3E is a confocal image illustrating the WDR cell recorded filled with Neurobiotin. Note that neither CGRP nor OTR colocalized with the Neurobiotin-positive cells.
Neuronal firing of a single WDR cell sensitive to PVN electrical stimulation with a RF in the periorbital and meningeal regions. A, B, Raster plots of the WDR neuronal responses evoked by periorbital (left) or meningeal (right) electrical stimulation before (basal, bottom), immediately after PVN electrical stimulation (middle panel), and after PVN electrical stimulation (5–20 min; top panel); note that the PVN-induced inhibition of WDR responses is transient and returns to basal response after 5–20 min. C, The wave average of the activity of the WDR neuron recorded is illustrated; note the similarity between the spikes evoked by periorbital or meningeal RF stimulation, indicating that the same WDR cell is recorded. D, Raw tracing of the recorded cell (top trace) during a juxtacellular current injection of Neurobiotin; the bottom trace originates from the extracellular DC activity. E, Confocal photomicrograph illustrating the WDR cell filled with Neurobiotin (asterisk, *), in which the immunofluorescence against CGRP and OTR was visualized.
As shown in Figure 4, the one-way RM ANOVA revealed that PVN electrical stimulation diminished the neuronal firing of Aδ- and C-fibers evoked by peripheral electrical stimuli of the periorbital (Aδ-fibers, F(1.54,20.00) = 20.00, p < 0.0001; C-fibers, F(1.06,13.84) = 47.69, p < 0.0001; Fig. 4A) and meningeal region (Aδ-fibers, F(1.52,19.72) = 7.64, p = 0.006; C-fibers, F(1.39,18.14) = 8.60, p = 0.005; Fig. 4B). This inhibition of the firing of WDR cells was reversible, as no difference was found after >5 min post-PVN stimuli (p > 0.1). The exact locations of the electrodes in the PVN are shown in Figure 4C. In certain cases (Fig. 4D), electrical stimulation did not seem to affect the peripheral evoked firing of the Aδ-fibers (t(5) = 2.11, n = 6, p = 0.09) or C-fibers (t(5) = 0.16, n = 6, p = 0.9), an effect linked to the fact that the electrode was located outside the PVN.
Convergent trigeminal WDR cells are inhibited by the PVN electrical stimulation. The effects of the PVN electrical stimulation on the neuronal WDR firing evoked by (A) periorbital or (B) meningeal sensory electrical stimulation. The Aδ- and C-fiber activities were reduced immediately after PVN activation in both cases (t = 0). The red circles correspond to the animals previously injected at PVN with FR, a neuronal tracer used to dissect the PVN → TCC pathway. C, Histological photomicrographs of electrolytic lesions of the PVN, where the stimulation electrode was positioned in the PVN. Using the Paxinos and Watson (1998) atlas, the location of the electrolytic lesion (blue dots) was shown along the rostrocaudal extension of the PVN. The red dots correspond to animals treated with the FR tracer. D, The WDR firing was not statistically different when the electrode was outside the PVN region.
Effect of PVN electrical stimulation and L-368,899 (OTR antagonist) on trigeminal WDR cell activity evoked by meningeal electrical stimulation
Since PVN electrical stimulation triggered inhibition of meningeal nociceptive WDR responses, the role of OTR was tested using 10−5 M L-368,899 (Fig. 5). It is worth noting that L-368,899 was used in these experiments considering that García-Boll et al. (2020) have shown this antagonist blocked the exogenous oxytocin-induced trigeminovascular WDR firing inhibition. Figure 5A displays the firing rate of a WDR cell before and after PVN electrical stimulation. Note that although PVN stimulation decreases the ongoing and meningeal-evoked neuronal activity, this effect seemed less noticeable when L-368,899 was delivered at the TCC.
The PVN-induced inhibition of the trigeminovascular TCC responses was inhibited by an OTR antagonist. A, Effect of PVN electrical stimulation on WDR cluster neuronal activity in rats with or without an OTR antagonist. In this panel, the artifacts of meningeal electrical stimulation are illustrated. Note that after PVN stimuli, the ongoing activity of trigeminovascular TCC cells diminished. B–D, Time course changes in the percentage average of the Aδ- and C-fibers and ongoing neuronal activity in a control situation (blue circles) or before (basal) and after PVN electrical stimulation in animals treated (red circles) or untreated (green circles) with the OTR antagonist; t = 0 depicts the neuronal activity immediately after PVN stimulation. The PVN-induced inhibition of the C-fiber component was reversed when animals were pretreated with L-368,899. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 versus control. Panels E and F show that pretreatment with L-368,899 preferentially blocked the PVN-induced inhibition of the late component of the WDR cell firing (i.e., C-fibers, 26–100 ms).
Accordingly, two-way RM ANOVA revealed that PVN electrical stimulation inhibited the peripheral evoked activation of Aδ- (Fig. 5B; F(2,75) = 21.5, p < 0.0001 vs control) and C-fibers (Fig. 5C; F(2,75) = 16.3, p < 0.0001 vs control), and a similar effect was observed for the ongoing activity (Fig. 5D; F(2,75) = 7.3, p = 0.001 vs control). In all cases, strong inhibition was observed immediately after PVN stimulation (i.e., t = 0). The PVN effect was long-lasting for C-fibers rather than for Aδ-fibers or ongoing activity. Moreover, although Tukey’s post hoc test failed to reveal an effect of L-368,899 immediately after PVN stimulation, the long-lasting inhibition of C-fiber activity was abolished (Fig. 5C). This effect on C-fibers is illustrated using the PSTHs (Fig. 5E,F), where L-368,899 prevented the long-lasting PVN-induced inhibition of WDR firing associated with the activation of C-fibers.
OTRs are present in CGRPergic and GABAergic fibers near the Neurobiotin-filled cells
In the electrophysiological experiments described above, four animals were previously injected with FR into the PVN (Figs. 6A; see Materials and Methods, PVN stereotaxic microinjection of the anterograde tracer FR). At the end of the electrophysiological recordings, these trigeminal second-order WDR cells were filled with Neurobiotin. After the immunofluorescence analysis, we found that projecting neurons from the PVN were near the Neurobiotin-filled cells (Fig. 6B), and neither Neurobiotin nor FR colocalized with fibers containing CGRP, supporting the electrophysiological nature of the WDR cell recorded (i.e., a second-order neuron). Although Figure 6C suggests colocalization between oxytocin and Neurobiotin-positive cells, the 3D reconstruction data (Video 1) showed that neither oxytocin nor OTR colocalized with the Neurobiotin-positive cells. Hence, we could propose that the trigeminovascular cell recorded does not possess OTR or oxytocin and receives neuronal projections from the PVN.
TCC WDR cells receive oxytocinergic innervation from the PVN. A, Experimental protocol used to identify the nature of the trigeminal WDR cell recorded. In n = 4 rats, FR was microinjected into the PVN 10 d before the electrophysiological recording of WDR cells. At the end of the electrophysiological experiment, the WDR cell recorded was filled with Neurobiotin, and the tissue was processed for the immunofluorescence analysis against Neurobiotin, FR, CGRP, oxytocin, and OTR. B, In the photomicrographs of these panels, note that the Neurobiotin-filled cell (B1; white arrow) is near a positive FR fiber (B2). Furthermore, Neurobiotin and FR signals did not colocalize with CGRP (B3 and B4). C, Photomicrographs of these panels suggest that nearly to Neurobiotin-filled cells (white arrow; C1), an oxytocinergic signal was present. Although a colocalization between Neurobiotin and oxytocin was suggested, the 3D reconstruction failed to confirm this observation (see the 3D-rendered reconstruction of the confocal figure in Video 1).
Video 1.
3D projection reconstruction from the z-stack confocal file of a neurobiotin-filled cell (green). Immunosignal against oxytocin (red) and OTR (blue) is detected. Interestingly, although in the 2D photomicrographs (see Fig. 6C), a colocalization between neurobiotin® and oxytocin was suggested, the 3D reconstruction failed to confirm this observation. Download Video 1, MP4 file.
These data fit with the contention that endogenous oxytocinergic transmission may play a role in the modulation of trigeminal nociception. Nevertheless, at the trigeminal level, apart from the CGRPergic nociceptive afferents, GABAergic interneurons also play a preponderant role in trigeminal nociception (Goadsby et al., 2017; Fan et al., 2018). Thus, subsequent immunofluorescence experiments were designed to explore the localization of CGRP, GABA, and OTR immunofluorescence signals in relation to trigeminal WDR cells filled with Neurobiotin.
Figure 7 shows optical sections (z-stack positions; Fig. 7A–D) of a successful TCC cell filled with Neurobiotin surrounded by immunopositive CGRP and OTR signals. In this case, it is interesting to note that the CGRP signal seems to colocalize with the OTR. Besides, as shown in Figure 8, the immunofluorescence experiments showed that GABA signals colocalized with OTR and were found near the Neurobiotin-filled cells. We must keep in mind that this Neurobiotin-filled cell with sensory input from the dural and periorbital receptive fields was electrophysiologically tested.
Confocal image of a spinal trigeminal WDR cell filled with Neurobiotin and immunofluorescence against OTR and CGRP. Panels A–D showed merged confocal photomicrographs at different optical sections (focal plane or z-position at +16 µm, +12 µm, +8 µm, and +6 µm) of the same cell tissue illustrating how the immunofluorescence against Neurobiotin (*), CGRP, and OTR can be found. Interestingly, Neurobiotin was located inside the cell body, but CGRP and OTR immunosignals were found in the peripheral cell tissue. Panels A1, A2, and A3 and D1, D2, and D3 show the individual channels for Neurobiotin, CGRP, and OTR signals of merged photomicrographs of A and D panels. Note that OTR appears in CGRPergic cells but not in Neurobiotin-filled cells.
Confocal image of spinal trigeminal cells filled with Neurobiotin and immunofluorescence against GABA and OTR. Panels A–C showed merged confocal photomicrographs at different optical sections (focal plane or z-position at +16 µm, +14 µm, and +12 µm) of the same cell tissue, illustrating how the immunofluorescence against Neurobiotin (*), GABA, and OTR can be found. Note that GABA and OTR signals can be detected at sites different from the main Neurobiotin-filled cells (i.e., the electrophysiologically recorded cell). To gain more confidence about the localization of the different immunosignals, panels A1, A2, and A3 and C1, C2, and C3 show the individual channels for Neurobiotin, GABA, and OTR signals of the merged photomicrographs of panels A and C (inset). Note that OTR seems to be present in GABA-positive cells (white arrows). Furthermore, GABA and OTR immunosignals were also detected in the peripheral cell tissue (panel C).
Histological correlates of peripheral convergent neurons from V1 and V2 trigeminal branches
In the electrophysiological migraine model used, it is assumed that a single WDR cell receives input from the two different receptive fields innervated by the ophthalmic branch of the trigeminal nerve (V1); however, some data suggest that the infraorbital nerve (from the maxillary nerve branch, V2) may play a role in migraine headaches (Alp and Alp, 2013; Toyama et al., 2017). Hence, neuronal tracing was performed to search for bifurcated TG cells that send projections to the supratentorial dura mater (mainly innervated by V1; Mayberg et al., 1984) and infraorbital nerve (V2; Fig. 9A). The data showed that TB injected at the meningeal level and FG injected at the infraorbital nerve colocalized in some TG cells (Fig. 9B).
Convergence in TG cells from the meningeal tissue and infraorbital receptive fields. A, The sites of TB and FG injections are illustrated. B, Photomicrographs obtained from the TG illustrate TG cells stained with FG and TB, and the merged image is plotted. Note the colocalization of FG and TB.
Discussion
General
Electrical stimulation of the PVN inhibits trigeminovascular nociception. Since OTR antagonism partially reverses this effect, the involvement of oxytocinergic transmission is supported. Immunofluorescence and neuronal tracing experiments revealed a direct oxytocinergic projection from the PVN to the TCC (near the recorded WDR cells). In addition, OTR-CGRP and OTR-GABA immunopositive signals were found near the WDR cells. These findings support the hypothesis that oxytocinergic transmission blocks nociception at the TCC level by indirectly inhibiting the activity of second-order WDR cells (Fig. 10).
Simplified schematic about how oxytocinergic transmission could inhibit trigeminal nociception. When the first branch (V1) of the TG nerve is electrically stimulated at the meningeal or periorbital RF, it triggers the second-order WDR cells in the TCC. However, when the hypothalamic PVN is stimulated, oxytocin is released at TCC and inhibits the neuronal firing of WDR cells via the OTR activation. Certainly, a direct descending neuronal projection from the PVN to the TCC exists, and at this level, OTR can be located in GABAergic and CGRPergic neuronal tissues, supporting the idea that presynaptic inhibition occurs. At this point, it is interesting to note that OTRs, a seven-transmembrane receptor, are canonically coupled to Gq proteins, but they can also activate the intracellular Gi-dependent pathways (Busnelli and Chini, 2018; Liu et al., 2024). Coupled with evidence that PVN stimulation induces oxytocin release in the cerebrospinal fluid (Martínez-Lorenzana et al., 2008) and that spinal OTR activation inhibits nociception via Gq or Gi pathways (Espinosa de los Monteros-Zúñiga et al., 2021), we hypothesized that (1) in GABAergic interneurons, the canonical pathway is activated, favoring an increase in GABA release, thus eliciting the inhibition of nociceptive primary afferents, whereas (2) in C-fibers, the OTR activation directly inhibits the nociceptive transmission probably by recruitment of Gi proteins. See text for details.
The role of endogenous oxytocinergic transmission in the PVN-induced inhibition of trigeminovascular nociception
Considering that TCC WDR cell firing elicited by periorbital or dural electrical stimulation is inhibited by exogenous oxytocin via OTR (García-Boll et al., 2018, 2020), we hypothesized that enhancing PVN activity may induce a similar effect. Indeed, PVN electrical stimuli blocked the WDR firing associated with the activation of periorbital Aδ- and C-fibers (Fig. 2). Since dOVT partially blocked this effect, the role of OTR is suggested.
We must keep in mind that some of these second-order cells receive concurrent nociceptive input from the periorbital and durovascular regions (i.e., trigeminovascular WDR neurons), and electrophysiological recordings of these cells have been used as surrogate migraine pain model (Bergerot et al., 2006; Akerman et al., 2013). Hence, the effect of PVN stimulation on trigeminovascular WDR cells was tested.
As expected, electrical stimulation of the PVN inhibited neuronal firing of convergent trigeminovascular WDR cells (Fig. 3A–C). These cells, with inputs from the periorbital and dural regions, were located at the caudal brainstem/cervicomedullary junction, as previously reported (Burstein et al., 1998; Robert et al., 2013). Notably, this inhibition resembled an indirect action (e.g., presynaptic inhibition) as OTR did not colocalize with Neurobiotin-positive cells (Fig. 3D,E). Besides, this inhibitory effect seemed to be specific, as no reliable effect on the activity of TCC cells was induced when the tip electrode was outside the PVN (Fig. 4).
According to Figure 5, it is proposed that OTR activation partially underlies the PVN effect upon neuronal trigeminovascular responses. Specifically, the raw electrophysiological data suggest that under the OTR blockade (using L-368,899), the inhibitory effect of PVN on meningeal-evoked and ongoing neuronal activity is diminished (Fig. 5A). However, a formal analysis (Fig. 5B–D) showed that L-368,899 does not impact the immediate (t = 0) inhibitory PVN effect, but, admittedly, a slightly nonsignificative effect can be observed in the case of C-fibers (p = 0.08). Remarkably, PVN stimulation produced a long-lasting suppression of C-fibers, but not of Aδ-fibers or ongoing activity. Since the long-lasting C-fiber inhibition was prevented when the OTR was blocked (Fig. 5C), the role of this receptor is supported. Additionally, the fact that no clear effect was achieved upon ongoing activity, the contention that presynaptic inhibition occurs is supported. However, while the ongoing neuronal activity has mostly been linked to intrinsic postsynaptic processing, the afferent input characteristics also play a role (Arieli et al., 1996). Therefore, evaluating the impact of oxytocinergic transmission during the increased ongoing activity (e.g., central sensitization induced by nitric oxide donors) could be useful to unmask the impact of OTR in second-order neuronal processing.
The role of OTR gains weight considering that this receptor has been found in TG CGRPergic cells (Tzabazis et al., 2016; Maddahi et al., 2022). Certainly, in the trigeminal nucleus caudalis and small- and medium-sized TG cells, the Oxtr gene is expressed (Warfvinge et al., 2020b). Thus, Krause et al. (2021) hint that oxytocin and OTR can be found in several parts of the trigeminal system. To gain certainty in this hypothesis, the iontophoretic labeling of WDR cells recorded, coupled with neuronal tracing and immunofluorescence approach, was followed (Fig. 6).
Oxytocinergic PVN projections to TCC are involved in the inhibition of trigeminovascular WDR cells
As shown in Figure 6, projections from the PVN to TCC arrive near the recorded WDR cell filled with Neurobiotin. In conjunction with the immunofluorescence data against FR and CGRP, the results suggest that the recorded WDR cells were near the descending PVN projection and primary CGRPergic fibers (Fig. 6B). We must consider that WDR cells recorded were found between laminae III and V (Fig. 1B), and although some data suggest that CGRPergic fibers are in laminae I–II, they also can be found in laminae III–V (Eftekhari and Edvinsson 2011; Martínez-Lorenzana et al., 2021a). Indeed, seminal reports consistently showed that CGRPergic fibers are concentrated in Lissauer's tract, laminae I–III, V, and X (Gibson et al., 1984; Lennerz et al., 2008).
To indirectly test the nature of the descending projection, we performed immunofluorescence against oxytocin and OTR (Fig. 6C). In this case, signals against OTR and oxytocin were found near the WDR cell filled with Neurobiotin (Video 1 for 3D reconstruction of Fig. 6C). Earlier, Warfvinge et al. (2020a,b) showed that at the spinal trigeminal nucleus, oxytocin and OTR could be immunodetected. In conjunction with the electrophysiological and pharmacological data discussed above, the notion that a presynaptic PVN action upon WDR cells occurred is supported. Hence, when the PVN is electrically stimulated, oxytocin is released at the TCC, and upon OTR activation (not located on Neurobiotin-positive cells), the neuronal activity of WDR cells is inhibited.
It is relevant to mention that these cells juxtacellularly injected with Neurobiotin were identified as WDR cells responding to electrical stimulation of meningeal nociceptors. A refined analysis of the data obtained by confocal microscopy suggested that OTR seems to be present in the CGRPergic fibers (Fig. 7) and GABAergic cells near the WDR Neurobiotin-filled cells (Fig. 8). Incidentally, at the lumbar dorsal horn level, ultrastructural analysis showed that OTRs are present in GABAergic and in primary nociceptive cells (Martínez-Lorenzana et al., 2021b). Thus, mechanistically it has been proposed that oxytocin via the canonical OTR pathway (i.e., seven-transmembrane receptor coupled to Gq proteins) enhances GABAergic activity, resulting in an indirect diminution of the Aδ- and C-fibers discharge (i.e., presynaptic WDR inhibition; Rojas-Piloni et al., 2007; Condés-Lara et al., 2009; González-Hernández and Charlet, 2018; Nishimura et al., 2022). However, current evidence also indicates that OTR activation can result in the inhibition of nociception through a noncanonical pathway (i.e., Gi/o proteins) that directly inhibits nociceptors (González-Hernández et al., 2017; Espinosa de los Monteros-Zúñiga et al., 2021).
Together, these data suggest that oxytocinergic projections from the PVN arrive at the TCC on CGRPergic fibers or GABAergic neurons containing OTR, which in turn inhibits the activity of second-order WDR cells receiving concomitant input from the dural and periorbital receptive fields (Fig. 10).
Finally, since the colocalization of TB (deposited in the meningeal region) and FG (injected into the infraorbital nerve) was found to be colocalized in some TG cells (Fig. 9), the most parsimonious interpretation is that some TG cells simultaneously project to craniofacial structures via the V1 and V2 trigeminal branches. The physiological relevance of this finding is uncertain and deserves further attention. Nevertheless, these data suggest that some TCC WDR cells receive projections from these bifurcated TG cells.
Final comments and limitations of our study
Endogenous oxytocin released by PVN stimulation could inhibit trigeminovascular nociception via OTR activation. This suggestion is reinforced when considering that exogenous oxytocin also inhibits nociception at the trigeminal level via the OTR (Kubo et al., 2017; Tzabazis et al., 2017; García-Boll et al., 2018, 2020). In light of the findings reported by Warfvinge et al. (2020a,b), who demonstrated the expression of OTR in numerous brain regions relevant to migraine pathophysiology, it seems reasonable to consider the oxytocinergic pathway as a potential avenue for reducing pain associated with migraines. Indeed, at the meningeal level, OTR activation upon trigeminal nerves inhibits capsaicin-induced CGRP release (Tzabazis et al., 2017).
Regarding the role of OTR, the fact that dOVT or L-368,899 partially reversed the PVN inhibitory effect is not surprising if we consider that, in addition to direct oxytocinergic transmission, other mechanisms can be involved when the PVN is electrically activated (Pittman et al., 1981; Eliava et al., 2016; Motojima et al., 2017; Gamal-Eltrabily et al., 2020; Nishimura et al., 2022; Iwasaki et al., 2023). For example, PVN electrical stimulation elicited a c-Fos increase at the spinal cord level and in structures related to the descending inhibition of nociception (i.e., nucleus raphe magnus, locus ceruleus, and the periaqueductal gray area; Condés-Lara et al., 2015). Certainly, one of the drawbacks of employing electrical stimulation is its lack of specificity. Therefore, opto- or chemogenetic approaches will be useful in dissecting the oxytocinergic role at the trigeminal level. In our experiments, apart from oxytocin, vasopressin may also influence trigeminovascular nociception (Maddahi et al., 2022). Notwithstanding, we must keep in mind that although V1A receptor activation upon peripheral nociceptors elicits antinociception, its cardiovascular effects cannot be neglected (Manzano-García et al., 2018).
Finally, while the current experiments were conducted in male subjects, it is important to acknowledge the role of sexual dimorphism in migraine pathophysiology (Vetvik and MacGregor, 2017), which is also relevant to oxytocinergic transmission. In this regard, at the spinal cord level, oxytocin exerts better antinociception in male than in female rats (Chow et al., 2016; Salinas-Abarca et al., 2022), a phenomenon linked to estrogen levels (Chow et al., 2022) or variations in OTR expression (Uhl-Bronner et al., 2005). Remarkably, in humans, oxytocin levels in plasma and cerebrospinal fluid do not differ between sexes (in nonpregnant women; McCullough et al., 2013; Martin et al., 2018), but fluctuations in oxytocin plasma levels throughout the menstrual cycle should not be overlooked (Engel et al., 2019; Krause et al., 2021). Hence, the impact of sex on oxytocinergic transmission modulating trigeminovascular nociception is a topic of significant interest that requires a closer examination.
Conclusion
To our knowledge, this is the first study showing that PVN electrical stimulation inhibits trigeminovascular second-order WDR cells receiving input from the periorbital and meningeal receptive fields. Furthermore, our data showed that OTR can be found in CGRPergic and GABAergic fibers, thus promoting a diminution of nociceptive input (Fig. 10). Together, these data support an emerging theory about the role of oxytocinergic transmission modulating trigeminal nociception and further highlight the potential role of OTR as a target in migraine pain and perhaps other primary headaches.
Footnotes
This study was financially supported by the PAPIIT-Universidad Nacional Autónoma de Mexico (Grant IN218122 to A.G.-H. and IN202222 to M.C.-L.) and Fondo Sectorial de Investigación para la Educación [Consejo Nacional de Ciencia y Tecnologia (CONACyT)-Mexico Grant No. A1-S-23631 to A.G.-H.]. We acknowledge Elsa N. Hernández-Ríos for her technical assistance in the microscopy and image analysis units. A.E-Z. is a Postdoctoral Researcher who received a fellowship from CONACyT Estancias Posdoctorales por México (2022–2024) CVU Number 612715. A.C.-Q. and S.A.F.-B. are Bachelor of Science students from UNAM (Biology and Neurosciences, respectively).
The authors declare no competing financial interests.
- Correspondence should be addressed to Abimael González-Hernández at abimaelgh{at}comunidad.unam.mx.
