Gαs-Coupled CGRP Receptor Signaling Axis from the Trigeminal Ganglion Neuron to Odontoblast Negatively Regulates Dentin Mineralization

An inflammatory response following dental pulp injury and/or infection often leads to neurogenic inflammation via the axon reflex. However, the detailed mechanism underlying the occurrence of the axon reflex in the dental pulp remains unclear. We sought to examine the intracellular cyclic adenosine monophosphate (cAMP) signaling pathway in odontoblasts via the activation of Gs protein-coupled receptors and intercellular trigeminal ganglion (TG) neuron–odontoblast communication following direct mechanical stimulation of TG neurons. Odontoblasts express heterotrimeric G-protein α-subunit Gαs and calcitonin receptor-like receptors. The application of an adenylyl cyclase (AC) activator and a calcitonin gene-related peptide (CGRP) receptor agonist increased the intracellular cAMP levels ([cAMP]i) in odontoblasts, which were significantly inhibited by the selective CGRP receptor antagonist and AC inhibitor. Mechanical stimulation of the small-sized CGRP-positive but neurofilament heavy chain-negative TG neurons increased [cAMP]i in odontoblasts localized near the stimulated neuron. This increase was inhibited by the CGRP receptor antagonist. In the mineralization assay, CGRP impaired the mineralization ability of the odontoblasts, which was reversed by treatment with a CGRP receptor antagonist and AC inhibitor. CGRP establishes an axon reflex in the dental pulp via intercellular communication between TG neurons and odontoblasts. Overall, CGRP and cAMP signaling negatively regulate dentinogenesis as defensive mechanisms.


Introduction
The G-protein-coupled receptor (GPCR)-regulated adenylyl cyclase (AC) signal transduction pathways induce intracellular cyclic adenosine monophosphate (cAMP) signaling in various cells [1][2][3]. G-proteins are composed of α, β, and γ subunits, which are mainly divided into Gα s , Gα i , Gα q , and Gα 12/13 . The many conformations of the receptor lead to a variety of highly specialized downstream signaling cascades [4]. GPCRs induce two main signaling pathways: cAMP and phosphatidylinositol signaling pathways. Gα s and Gα i regulate the cAMP-generating enzyme AC. In fact, Gα s activates AC, while Gα i inhibits AC, leading to increases or decreases in intracellular cAMP ([cAMP] i ) levels. The diffusible intracellular second messenger system stimulates or inhibits downstream signaling, thereby enabling further biological effects [4].
Calcitonin gene-related peptide (CGRP), which is expressed in various organs such as the brain, heart, liver, and spleen as well as skeletal muscle and dental pulp tissue, controls general circulation by regulating blood flow as a vasodilator [5]. CGRP has a close reciprocal interaction with the sympathetic nervous system in the periphery [5][6][7]. CGRP receptors comprise the calcitonin receptor-like receptor (CALCRL) and receptor activity concentration before use. ECS or ECS-containing drugs were administered using a gravityfed perfusion system (Warner Instruments, Holliston, MA, USA). All other reagents were purchased from Sigma Chemical Co., (St. Louis, MO, USA), unless otherwise indicated.

Dental Pulp Slice Preparation
Dental pulp slice preparations were obtained from newborn Wistar rats (aged 4-8 days) [12,[20][21][22]. Decapitation was performed under isoflurane (3%) and pentobarbital sodium anesthesia (25 mg/kg, administered via intraperitoneal injection). To reduce the number of rats, both male and female rats were used in this study. The mandibles were dissected, and the hemimandibles, embedded in alginate impression material, were sectioned transversely through the incisor at a thickness of 500 µm using a standard vibration tissue slicer (ZERO-1; Dosaka EM, Kyoto, Japan). The mandibular sections were sliced to ensure direct visibility of the dentin and enamel between the bone tissue and dental pulp. We selected mandible sections with thin dentin (but with enamel and dentin distinguishable under a microscope) to avoid cellular damage in odontoblasts. The surrounding impression material, bone tissue, enamel, and dentin were carefully removed, and the remaining dental pulp slices were obtained. Pulp slices were treated with ECS containing 0.17% collagenase and 0.03% trypsin at 37 • C for 30 min. To measure [cAMP] i , enzymatically treated dental pulp slices were plated in culture dishes; cultured in α-MEM containing 10% fetal bovine serum (FBS), 5% horse serum, 1% amphotericin B, and 1% penicillin-streptomycin (Life Technologies Co., Grand Island, NY, USA); and maintained at 37 • C in a 5% CO 2 incubator for 24-40 h.

Measurements of Intracellular cAMP-and/or Ca 2+ -Sensitive Dye Fluorescence
A dish containing cADDis-transfected dental pulp slices or a co-culture of fura-2loaded TG neurons with cADDis-transfected dental pulp slices was mounted on the stage of a microscope (Olympus, Tokyo, Japan; IX73); this microscope had an HCImage system (Hamamatsu Photonic, Shizuoka, Japan), an excitation wavelength selector, and an intensified charge-coupled device camera system. The cADDis fluorescence emission was recorded at 560 nm in response to an excitation wavelength of 605 nm. Fura-2 fluorescence emission was recorded at 510 nm in response to alternating excitation wavelengths of 340 (F340) and 380 nm (F380). [Ca 2+ ] i was defined as the fluorescence ratio (R F340/F380 ) at the two excitation wavelengths, and the cADDis fluorescence and R F340/F380 of fura-2 were defined as the value of F, normalized to the resting value (F 0 ), and as F/F 0 units, respectively. All experiments were performed at 28 • C.

Mechanical Stimulation of the Single TG Neurons
Mechanical stimulation [12,13] was applied using a fire-polished glass micropipette with a tip diameter of 2-3 µm filled with ECS. The micropipette was pulled from the capillary tubes using a DMZ universal puller (Zeitz Instruments, Martinsried, Germany). The tip was positioned just above the target TG neuron, and the micropipette was moved vertically downward by 8.0 µm at a velocity of 2.0 µm/s [12,13] to generate a focal mechanical stimulation by the micromanipulator (µMp micromanipulator, Sensapex, Oulu, Finland) and software (Sensapex). The stimulation was applied for 22 s. Thereafter, the pipette was retracted at the same velocity. It is difficult to distinguish neurons from glial cells in TG cell cultures. To specifically reveal neuronal intracellular Ca 2+ responses, responses in primary cultured TG cells were measured by the application of a solution containing a high concentration of K + (50 mM K + ), which induced membrane depolarization.

Measurement of the Intercellular Distance and Size of the Stimulated TG Neurons
Co-cultured odontoblasts in the dental pulp slice preparation and TG neurons were imaged using an intensified charge-coupled device camera (Hamamatsu Photonic) and microscope (Olympus). The distance from a mechanically stimulated TG neuron to each neighboring odontoblast and the size of the stimulated TG neuron were determined using the images (HCImage) by measuring the shortest distance of each pair of cells or the diameter of the stimulated TG neuron.

Mineralization Assay
Isolated odontoblasts were grown for 20-40 h in a basal medium and transferred to a mineralization medium (10 mM β-glycerophosphate and 50 µg/mL ascorbic acid in basal medium) for growth at 37 • C in 5% CO 2 . To determine the effects of CGRP activity on mineralization, odontoblasts were cultured in the mineralization medium without (as control) or with CGRP (rat) (50 nM) and with the CGRP inhibitor BIBN 4096 (0.1 nM) or the AC inhibitor SQ22536 (0.1 µM) for 7 days. During the 7-day culture period, the mineralization medium was changed twice per week. To detect the deposition of calcium and calcium phosphate, cells were subjected to alizarin red staining, and the mineralization efficiencies were measured using a microscope (Keyence, Osaka, Japan; X710). The regions of interest (ROIs) were determined for each odontoblast to measure the mean luminance intensity of the mineralized area in the total area (I) of the ROI. The mineralizing efficiencies were normalized and represented as I/I 0 units, and the intensities (I) of alizarin red staining were normalized to the mean intensity area in the dental pulp (I 0 ).

Statistical Analysis
Data are expressed as mean ± SE or SD of the mean of N observations, where N represents the number of experiments or cells tested. Non-parametric statistical significance was determined using the Friedman test and Mann-Whitney test with Dunn's post-hoc test to assess the [cAMP] i levels and mineralized areas in rat odontoblasts (Figures 2-6). Parametric statistical significance was determined using an unpaired t-test to analyze [Ca 2+ ] i in rat TG neurons ( Figure 5). Statistical significance was set at p < 0.05. Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA, USA). The data were also analyzed using Origin 8.5 (OriginLab Corporation, Northampton, MA, USA).

Expression of Heterotrimeric Gnas and CALCRL in Rat Odontoblasts
Primary cultured rat odontoblasts from the dental pulp slice preparations displayed immunoreactivity for DSPP (green in Figure 1A,C,D,F), heterotrimeric Gnas (red in Figure 1B,C), and CALCRL (as a component of CGRP receptor; red in Figure 1E,F).

Forskolin Dose-Dependently Increases the Intracellular cAMP Level
FSK was used as an AC activator to pharmacologically activate AC in rat odontoblasts. In the presence of external Ca 2+ (2.5 mM), the application of six different concentrations of FSK (0.0001, 0.001, 0.01, 0.1, 2, and 13 μM) elicited rapid, transient, and concentration-dependent increases in [cAMP]i in rat odontoblasts ( where K is the half-maximal concentration (50% effective concentration (EC50)) of FSK; x indicates the applied concentration of FSK; and F/F0max and F/F0min are the maximal and minimal F/F0 responses in [cAMP]i, respectively. The EC50 of FSK was 0.14 μM for the increases in [cAMP]i.

Repeated Application of Forskolin Has a Desensitizing Effect on the Increase in Intracellular cAMP Level
In the presence of extracellular Ca 2+ (2.5 mM), the application of 0.1 μM FSK to odontoblasts led to a rapid and transient increase in [cAMP]i ( Figure 3A), with F/F0 units of [cAMP]i reaching a peak value of 1.51 ± 0.04; this was followed by a rapid decay to near-

Repeated Application of Forskolin Has a Desensitizing Effect on the Increase in Intracellular cAMP Level
In the presence of extracellular Ca 2+ (2.5 mM), the application of 0.1 µM FSK to odontoblasts led to a rapid and transient increase in [cAMP] i ( Figure 3A), with F/F 0 units of [cAMP] i reaching a peak value of 1.51 ± 0.04; this was followed by a rapid decay to nearbaseline levels (F/F 0 = 1). Repeated applications of 0.1 µM FSK decreased the amplitudes of FSK-evoked increases in [cAMP] i level, and by the third application, a peak value of 1.22 ± 0.02 (N = 8; Figure 3B) was achieved for F/F 0 units of [cAMP] i . i as a function of the applied FSK concentration. Each point represents the mean ± SE. The curve on the semi-logarithmic scale was fitted according to Equation (1), which is described in the text.

Repeated Application of Forskolin Has a Desensitizing Effect on the Increase in Intracellular cAMP Level
In the presence of extracellular Ca 2+ (2.5 mM), the application of 0.1 μM FSK to odontoblasts led to a rapid and transient increase in [cAMP]i ( Figure 3A), with F/F0 units of [cAMP]i reaching a peak value of 1.51 ± 0.04; this was followed by a rapid decay to nearbaseline levels (F/F0 = 1). Repeated applications of 0.1 μM FSK decreased the amplitudes of FSK-evoked increases in [cAMP]i level, and by the third application, a peak value of 1.22 ± 0.02 (N = 8; Figure 3B) was achieved for F/F0 units of [cAMP]i.

CGRP Increases the Intracellular cAMP Level
In the presence of extracellular Ca 2+ (2.5 mM), the application of 50 nM of CGRP (rat) as a G s protein-coupled CGRP receptor activator to odontoblasts induced a rapid and transient increase in [cAMP] i , with the F/F 0 units reaching a peak value of 1.94 ± 0.1 (N = 9; Figure 4A) or 2.11 ± 0.11 (N = 11; Figure 4C); this was followed by a rapid decay to near-baseline levels (F/F 0 = 1). CGRP-induced [cAMP] i increases were significantly and reversibly inhibited by 0.1 nM BIBN 4096, a non-selective CGRP receptor inhibitor, to 1.21 ± 0.04 (N = 9; Figure 4B), and by 0.1 µM SQ22536, a pharmacological AC inhibitor, to 1.44 ± 0.04 (N = 11; Figure 4D) F/F 0 units.

Direct Mechanical Stimulation of the TG Neurons Simultaneously Increases the [Ca 2+ ] i in the Neurons and the [cAMP] i in the Odontoblasts Approximating the Stimulated Neurons
Based on simultaneous measurements of [cAMP] i in the odontoblasts and [Ca 2+ ] i in the neurons, in the co-culture system with external 2.5 mM Ca 2+ , the application of focal and direct mechanical stimulation to single TG neurons (8.0 µm in depth) induced transient increases in [Ca 2+ ] i (green lines, Figure 5A,B) to a peak F/F 0 value of 3.23 ± 2.45 (N = 3; Figure 5A) in the control condition. Transient increases in [cAMP] i (red lines, Figure 5A) were also observed in neighboring odontoblasts during the direct mechanical stimulation of TG neurons (N = 27; Figure 5A). The amplitude of the [cAMP] i increase in neighboring odontoblasts reduced with increase in their distance from the mechanically stimulated TG neurons (N = 27; Figure 5A and open columns of Figure 5C). Furthermore, a time delay in the increase in [cAMP] i was observed in the nearby odontoblasts during the mechanical stimulation of single TG neurons (time when the mechanical stimulation applied to the stimulated TG neuron was set to 0 s; vertical dotted line, Figure 5A,B). In the presence of 0.1 nM BIBN 4096, a non-selective CGRP receptor inhibitor, the mechanical stimulation of TG neurons induced [Ca 2+ ] i increases, reaching a peak F/F 0 value of 2.40 ± 0.87 (N = 3; Figure 5B); however, the [cAMP] i increases in the neighboring odontoblasts were significantly inhibited (N = 42; Figure 5B) compared to the increases in those without BIBN 4096. There were no significant differences in the F/F 0 values of the mechanically stimulated [Ca 2+ ] i increases in the presence and absence of external BIBN 4096. After measuring [Ca 2+ ] i and [cAMP] i in the co-cultured cells, immunofluorescence analyses were performed to identify the nature of the stimulated TG neurons. TG cells were identified by the application of a solution containing a high concentration of K + (50 mM), which induced membrane depolarization to specifically reveal neuronal intracellular Ca 2+ responses in primary culture.
The mechanically stimulated TG neurons (arrowheads) displayed immunopositivity to CGRP (red) but not to NF-H (blue) (Figure 5D-F). The size of the stimulated TG neurons, which was measured before mechanical stimulation, was 15.34 ± 2.59 µm (N = 6).

Direct Mechanical Stimulation of the TG Neurons Simultaneously Increases the [Ca 2+ ]i in the Neurons and the [cAMP]i in the Odontoblasts Approximating the Stimulated Neurons
Based on simultaneous measurements of [cAMP]i in the odontoblasts and [Ca 2+ ]i in the neurons, in the co-culture system with external 2.5 mM Ca 2+ , the application of focal and direct mechanical stimulation to single TG neurons (8.0 μm in depth) induced transient increases in [Ca 2+ ]i (green lines, Figure 5A,B) to a peak F/F0 value of 3.23 ± 2.45 (N = 3; Figure 5A) in the control condition. Transient increases in [cAMP]i (red lines, Figure 5A) were also observed in neighboring odontoblasts during the direct mechanical stimulation of TG neurons (N = 27; Figure 5A). The amplitude of the [cAMP]i increase in neighboring odontoblasts reduced with increase in their distance from the mechanically stimulated TG neurons (N = 27; Figure 5A and open columns of Figure 5C). Furthermore, a time delay in the increase in [cAMP]i was observed in the nearby odontoblasts during the mechanical stimulation of single TG neurons (time when the mechanical stimulation applied to the stimulated TG neuron was set to 0 s; vertical dotted line, Figure 5A,B). In the presence of The mechanically stimulated TG neurons (arrowheads) displayed immunopositivity to CGRP (red) but not to NF-H (blue) (Figure 5D-F). The size of the stimulated TG neurons, which was measured before mechanical stimulation, was 15.34 ± 2.59 μm (N = 6).

CGRP-CGRP Receptor Signaling Regulates Defensive Dentin Demineralization
We investigated the effects of CGRP activity on mineralization in isolated odontoblasts. Alizarin red staining ( Figure 6A-D) indicates the mineralization levels based on the staining intensity (see Materials and Methods), represented as I/I 0 units; the intensities (I) of the stains were normalized to the mean intensity area in the dental pulp (I 0 ; Figure 6E). The application of 50 nM CGRP (rat) significantly reduced the mineralization levels compared to the control level and that with the CGRP receptor antagonist (0.1 nM), BIBN 4096, or the AC inhibitor (0.1 µM) SQ22536.
blasts. Alizarin red staining ( Figure 6A-D) indicates the mineralization levels based on the staining intensity (see Materials and Methods), represented as I/I0 units; the intensitie (I) of the stains were normalized to the mean intensity area in the dental pulp (I0; Figure  6E). The application of 50 nM CGRP (rat) significantly reduced the mineralization level compared to the control level and that with the CGRP receptor antagonist (0.1 nM), BIBN 4096, or the AC inhibitor (0.1 μM) SQ22536.

Discussion
According to the findings of the present study, DSPP-positive rat odontoblasts express Gα s -protein-coupled CGRP receptors. CGRP receptor activation was found to regulate the AC signal transduction pathway to produce [cAMP] i . The increases in [cAMP] i levels were decreased by repeated stimulation of AC. AC produces cAMP from adenosine triphosphate (ATP), and the depletion of ATP may cause a decrease in [cAMP] i levels through repeated AC activation. We identified odontoblasts based on their expression of DSPP, a mature odontoblast marker [27,28]. To record cAMP appropriately at the single-cell level, gap junctions comprising members of the connexin family between odontoblasts should be disconnected, because gap junctional communication allows intercellular communication via intracellular cAMP/Ca 2+ . Therefore, we prepared a single odontoblast level. We confirmed that cells retained the high protein expression level of DSPP and preserved their function even when the cells were growing out. The mechanical stimulation of TG neurons induced not only [Ca 2+ ] i increases in the stimulated neurons but also [cAMP] i increases in the odontoblasts near the stimulated TG neurons. The increase in [cAMP] i in the neighboring odontoblasts was inversely proportional to their distance from the mechanically stimulated TG neurons, suggesting that diffusible substances are released from mechanically stimulated TG neurons. Moreover, the increase in [cAMP] i in nearby odontoblasts during neuronal mechanical stimulation was significantly suppressed by the application of BIBN 4096, a CGRP receptor antagonist. These results functionally align with previous results that revealed the expression of CGRP receptors by odontoblasts [29] and the function of released CGRP from mechanically stimulated TG neurons as a mediator that contributes to the development of the axon reflex in dental pulp tissue [11].
After [Ca 2+ ] i and [cAMP] i measurements in TG neurons and odontoblasts in the coculture, mechanically stimulated small-sized TG neurons (15.34 ± 2.59 µm in diameter) were observed via immunofluorescence and identified to exhibit CGRP positivity and NF-H negativity. TG neurons are subclassified into nine cell types, and these cells are divided into two groups: small-sized (15-24 µm) and medium-sized (25-38 µm) [30]. NF-H is a marker for medium-to large-sized A-fiber neurons, whereas peptidergic C-fiber neurons express CGRP [31,32]. Therefore, the cells mechanically stimulated in the present study were peptidergic C-fiber neurons. CGRP is released from the terminals of pulpal nociceptors, which comprise unmyelinated C fibers [7]. Furthermore, CGRP localizes in TG neurons and nerve terminals in the dental pulp [31] and contributes to vasodilation via the axon reflex in pulpitis [7]. Nerve fibers in dental pulp express CGRP [33] and dental pulp cells express CGRP receptors increasingly during inflammatory phenomena such as acute irreversible pulpitis [34]. In this study, endogenous CGRP expression could not be observed in physiological conditions of odontoblasts using immunostaining (personal communication with TO); therefore, further study will be needed to reveal whether the inflammatory response induces endogenous CGRP expression in odontoblasts. Thus, CGRP released from peptidergic C neurons is an intercellular mediator among the cells in dental pulp, including odontoblasts, and this may imply that CGRP released from C neurons mediates neurogenic inflammation in dental pulp.
The mechanical stimulation of nerve fibers by pulsating vessels activates mechanosensitive Piezo channels and stimulates neuronal CGRP release, which contributes to neurogenic inflammation. The activation of Piezo2 channels, which are mechanosensitive ion channels, promotes CGRP release from TG neurons during migraine attacks [35]. Moreover, medium-sized primary afferent neurons innervating the dental pulp express Piezo2 channels, as they are low-threshold mechanoreceptors [36]. Although a further study is required, previous results suggest that the increases in [Ca 2+ ] i induced by mechanical stimulation are mediated by mechanosensitive Piezo channel activation in TG neurons. However, the functional roles of CGRP in the cellular functions of odontoblasts, the essential player in dentinogenesis, are unclear. A previous study demonstrated that CGRP produced by stimulated nociceptive neurons plays a role in murine molar dental pulp stem cell proliferation, differentiation, and inflammatory gene expression [37]. By contrast, it has been reported to inhibit bone mineralization in osteoblasts [38,39]. Our findings revealed that CGRP significantly decreased mineralization levels. Mineralization in odontoblasts which indicated positivity for DSPP (specific late odontoblast differentiation marker) [27] cultured in a mineralization medium without CGRP (as a control experiment) mimicked the physiological or developmental conditions of dentinogenesis. Furthermore, the application of CGRP with a CGRP receptor antagonist or an AC inhibitor led to the recovery of mineralization levels compared to single CGRP application. Our data suggest that the neuropeptide CGRP, which is released from TG neurons, might function to prevent excessive mineralization and avoid increasing pressure. Based on these results, CGRP released from peptidergic C neurons in the axon reflex suppresses dentin regeneration to prevent internal tissue pressure increases. Of note, a reduction in the volume of the dental pulp chamber by excessive pathological dentin formation results in an increase in tissue pressure caused by pulpal inflammatory responses, which may accelerate the responses to dental pulp inflammation. Thus, a CGRP-induced reduction in mineralization efficiency by odontoblasts may be essential for defensive reactions during dental pulp inflammation.
In conclusion, we revealed the functional expression of Gα s protein-coupled CGRP receptors in DSPP-immunopositive odontoblasts. CGRP receptor activation increases [cAMP] i by activating AC. The mechanical stimulation of peptidergic C neurons of the TG, which mimics mechanical stimulation due to dental pulp inflammation, induces CGRP release following mechanosensitive ion channel activation. CGRP released from peptidergic C neurons increases [cAMP] i in odontoblasts via CGRP receptor activation and reduces dentin mineralization. Thus, the CGRP-CGRP receptor axis plays a critical role in the regulation of dentinogenesis via intercellular communication. These results may provide functional evidence regarding the axon reflex mediated via the CGRP-CGRP receptor axis in dental pulp.
Author Contributions: N.S. and Y.S. contributed to the conception, design, data acquisition, analysis, and interpretation; performed the statistical analyses; and drafted and critically revised the manuscript. T.O. contributed to the conception, data acquisition, analysis, and interpretation and drafted and critically revised the manuscript. M.K. contributed to the conception, design, data acquisition, analysis, and interpretation; performed the statistical analyses; and drafted and critically revised the manuscript. T.I. contributed to the conception and design and drafted and critically revised the manuscript. All authors have read and agreed to the published version of the manuscript.