- Email: [email protected]
The expression of hyperpolarization-activated cyclic nucleotide-gated channel 1 (HCN1) and HCN2 in the rat trigeminal ganglion, sensory root, and dental pulp
Neuroscience 291 (2015) 15–25
THE EXPRESSION OF HYPERPOLARIZATION-ACTIVATED CYCLIC NUCLEOTIDE-GATED CHANNEL 1 (HCN1) AND HCN2 IN THE RAT TRIGEMINAL GANGLION, SENSORY ROOT, AND DENTAL PULP Y. S. CHO, Y. S. KIM, S. J. MOOZHAYIL, E. S. YANG AND Y. C. BAE *
INTRODUCTION
Department of Anatomy and Neurobiology, School of Dentistry, Kyungpook National University, Daegu 700-412, Republic of Korea
Hyperpolarization-activated cyclic nucleotide-gated channel 1 (HCN1) and 2 (HCN2), of the discovered four isoforms of HCN channels, are most abundantly expressed in the primary sensory neurons in the dorsal root ganglion (DRG) and trigeminal ganglion (TG, Chaplan et al., 2003; Yao et al., 2003). They contribute to hyperpolarization-activated currents (Ih) and neuronal excitability, and are implicated in pathological pain (Chaplan et al., 2003; Tu et al., 2004; Emery et al., 2011; Weng et al., 2012). What type of neuron in the DRG and TG expresses HCN1 and HCN2 is still uncertain: the expression in large-sized soma (Chaplan et al., 2003; Hogan and Poroli, 2008; Cho et al., 2009; Acosta et al., 2012; Gao et al., 2012; Schnorr et al., 2014) suggests that HCN1- and HCN2-positive neurons are predominantly low threshold mechanoreceptive (LTM), whereas studies that implicate HCN in allodynia and hyperalgesia following inflammation and nerve injury (Chaplan et al., 2003; Cho et al., 2009; Emery et al., 2012) suggest that HCN may be expressed in small, nociceptive neurons. Peripheral application of a specific HCN blocker attenuated mechanical hypersensitivity in a model of inflammation (Weng et al., 2012; Hatch et al., 2013), suggesting that HCN expressed in peripheral axons plays a role in the peripheral sensitization and inflammatory pain. However, little is known about the HCN expression in peripheral axons, particularly in the dental pulp. Dental pulp is innervated densely by nociceptive afferents that signal pain in response to a variety of stimuli (Byers, 1984; Hildebrand et al., 1995; Paik et al., 2009). We showed that glutamate receptors and vesicular glutamate transporters (VGLUTs) are expressed in axons of the dental pulp (Kim et al., 2009; Paik et al., 2012), suggesting peripheral pain processing in the pulp. In addition, even though HCN1 and HCN2 are implicated in inflammatory pain (Chaplan et al., 2003; Emery et al., 2012; Schnorr et al., 2014), it remains uncertain whether their expression in TG neurons is associated with pulpal inflammatory pain (Wells et al., 2007). To address these issues, we studied the expression of HCN1 and HCN2 in the soma and axons in the rat TG neurons and axons in the dental pulp, and the changes in that expression following inflammation, using lightand electron-microscopic immunocytochemistry and quantitative analysis.
Abstract—Hyperpolarization-activated cyclic nucleotidegated channel 1 (HCN1) and 2 (HCN2) are abundantly expressed in primary sensory neurons and contribute to neuronal excitability and pathological pain. We studied the expression of HCN1 and HCN2 in the rat trigeminal ganglion (TG) neurons and axons in the dental pulp, and the changes in their expression following inflammation, using light- and electron-microscopic immunocytochemistry and quantitative analysis. HCN1 and HCN2 were expressed predominantly in large-sized, neurofilament 200-immunopositive (+) or parvalbumin+ soma in the TG whereas they were expressed mostly in unmyelinated and small myelinated axons in the sensory root. The expression was particularly strong along the plasma membrane in the soma. In the dental pulp, majority of HCN1+ and HCN2+ axons coexpressed calcitonin gene-related peptide. They were expressed mainly in the peripheral pulp and pulp horn where the axons branch extensively in the dental pulp. The expression of HCN1 and HCN2 in TG neurons increased significantly in rats with experimentally induced inflammation of the dental pulp. Our findings support the notion that HCN1 and HCN2 are expressed mainly by both the soma of mechanosensitive neurons in the TG and peripheral axons of nociceptive neurons in the sensory root, and may play a role in the mechanisms of inflammatory pain from the dental pulp. Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved.
Key words: HCN, axon type, trigeminal ganglion, dental pulp, inflammation, ultrastructure.
*Corresponding author. Address: Department of Anatomy and Neurobiology, School of Dentistry, Kyungpook National University, 188-1, 2-Ga, Samdeok-Dong, Jung-Gu, Daegu 700-412, Republic of Korea. Tel: +82-53-660-6860; fax: +82-53-426-7731. E-mail address: [email protected] (Y. C. Bae). Abbreviations: CGRP, calcitonin gene-related peptide; CFA, Complete Freund’s Adjuvant; DRG, dorsal root ganglion; EDTA, ethylenediaminetetraacetic acid; EM, electron microscopy; HCN, hyperpolarization-activated cyclic nucleotide-gated channel; IgG, immunoglobulin G; Ih, hyperpolarization-activated currents; LM, light microscopy; LTM, low threshold mechanoreceptive; NDS, normal donkey serum; NF200, neurofilament 200; PB, phosphate buffer; PBS, phosphate-buffered saline; PGP9.5, protein gene product 9.5; TG, trigeminal ganglion; VGLUT, vesicular glutamate transporter. http://dx.doi.org/10.1016/j.neuroscience.2015.01.066 0306-4522/Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved. 15
16
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
EXPERIMENTAL PROCEDURES All animal procedures were performed according to the National Institutes of Health guidelines and were approved by the Kyungpook National University Intramural Animal Care and Use Committee. Experiments were designed to minimize the number of animals used and their suffering. Thirty male Sprague– Dawley rats weighing 300–320 g were used for this study: Fifteen rats were for light microscopic (LM) immunohistochemistry including three for normal group, and each three for Complete Freund’s Adjuvant (CFA) treatment with 1-day, 3-day survival and their controls, and three rats were for electron microscopic (EM) immunohistochemistry. Twelve rats were for Western blot analysis including each three for CFA treatment with 1-day, 3-days survival and their controls. Tooth pulp inflammation model Rats were anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and the occlusal enamel and dentin of the right maxillary 1st (M1) and 2nd (M2) molars were filed off to just before exposing the pulp using a lowspeed dental drill with a round bur under water-cooling. A small piece of tissue paper soaked in 50% solution of CFA in saline or saline for control was applied to the exposed dentinal surfaces for 5 min, and after that the dentinal surfaces were sealed with dental cement. One or 3 days later, rats were re-anesthetized and sacrificed with intravascular perfusion with fixatives. LM immunohistochemistry For immunofluorescence, rats were deeply anesthetized with sodium pentobarbital (80 mg/kg, i.p.), transcardially perfused with heparinized normal saline, followed by freshly prepared fixative containing 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4). The dental pulps of the right maxillary 1st (M1) and 2nd (M2) molars and the right TGs in normal rats, and the right TGs of CFA-treated rats (experimental) and of saline-treated rats (control) were carefully removed, postfixed in the same fixative for 2 h, and cryoprotected in 30% sucrose in PB overnight at 4 °C. The next day, 30lm-thick sections were cut on a freezing microtome and treated with 50% ethanol for 30 min and with 10% normal donkey serum (NDS, Jackson ImmunoResearch, West Groove, PA, USA) for 30 min. Sections were incubated overnight in rabbit anti-HCN1 (1:800, AB5884, Chemicon, Billerica, MA, USA) or rabbit anti-HCN2 (1:800, APC-030,
Alomone, Jerusalem, Israel) antibodies alone or in combination with mouse anti-protein gene product 9.5 (PGP 9.5, 1:5,000, YM8104, Accurate chemical and scientific Corp., Westbury, NY, USA) antibodies. To verify the identity of HCN1- and HCN2-immunopositive soma and pulpal axons, sections were also incubated overnight in rabbit anti-HCN1 or rabbit anti-HCN2 antibodies in combination with mouse anti-neurofilament 200 (NF200, 1:100,000, N0142, Sigma–Aldrich, St.Louis, MO, USA), mouse anti-parvalbumin (1:1,000, MAB1572, Chemicon), mouse anti-calcitonin gene-related peptide (CGRP, 1:1,000, ab81887, Abcam, Cambridge, MA, USA) or biotinylated-isolectin B4 (biotinylated-IB4, 5 lg/ml, L-3759, Sigma–Aldrich). On the next day, the sections were washed with phosphate-buffered saline (PBS; 0.01 M, pH7.4) and incubated for 3 h with Cy3- or fluorescein isothiocyanate-conjugated secondary antibodies (raised in donkey, 1:200, in PB, Jackson ImmunoResearch). For biotinylated-IB4 staining, sections were incubated for 3 h with streptavidin-Alexa488 (1:600, S-11223, Molecular probes, Grand Island, NY, USA). Finally, sections were mounted on slides and coverslipped with Vectashield (Vector Laboratories, Burlingame, CA, USA), and micrographs were obtained with an Exi digital camera (Q-Imaging Inc., Surrey, CA, USA), attached to a Zeiss Axioplan 2 fluorescence microscope (Carl Zeiss Inc., Jena, Germany). Quantitative analysis To assess the size distribution of HCN1-immunopositive (+) and HCN2+ neurons in the TG, cross-sectional area of all neuronal somata with clearly visible nucleoli was measured using Image J software (http://imagej. nih.gov/ij/, NIH, Bethesda, MD, USA) and graphs were built in KaleidaGraph (v 3.5; Synergy Software, Reading, PA, USA). HCN1+ and HCN2+ soma were divided into three groups, according to their crosssectional area: small (<600 lm2 in cross-sectional area), medium (600–1200 lm2) and large (>1200 lm2) and their proportions were analyzed. To quantify the immunoreactivity in HCN1+ and HCN2+ TG perikarya in the CFA-treated rats, a total of 12–16 images were collected from 3 to 4 sections of each TGs in each saline- and CFA-treated rat (1-day or 3-days). The images were captured with a 20 objective (857.14 652.94 lm, 1360 1036 pixels) from the maxillary region of the TG. The images were converted to grayscale and all images for each antibody were identically enhanced for brightness and contrast. The
Fig. 1. Immunofluorescent staining for HCN1 (A) and HCN2 (B) in trigeminal ganglion neurons is completely abolished by preadsorption with the corresponding blocking peptides (HCN1: 1 lg/ml, HCN2: 1 lg/ml). Scale bar = 50 lm.
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
17
threshold level for defining a neuron as immunopositive was set at 100–120 (out of 255) gray levels using Image J software. The fraction of HCN1+ or HCN2+ somata of all TG somata and the differences in the mean gray value of immunoreactivity in HCN1+ and HCN2+ somata between CFA- and control groups were analyzed with Student’s t-test; statistical significance was set at p < 0.05. EM immunohistochemistry For EM, rats were deeply anesthetized with sodium pentobarbital (80 mg/kg, i.p.), transcardially perfused with heparinized normal saline, followed by freshly prepared fixative containing 4% paraformaldehyde and 0.01% glutaraldehyde in PB. The peripheral sensory roots (maxillary branch of the trigeminal nerve) of the TG were dissected out under a microscope and postfixed in the same fixative for 2–4 h at 4 °C. Sections of the sensory roots were cut transversely on a Vibratome at 60-lm thickness and cryoprotected in 30% sucrose in PB overnight at 4 °C. The next day, sections of the sensory root were frozen on the dry ice for 20 min, thawed in PBS to enhance penetration, pretreated with 1% sodium borohydride for 30 min to quench glutaraldehyde and then blocked with 3% H2O2 for 10 min and with 10% NDS for 30 min and incubated in rabbit anti-HCN1 (1:500; Chemicon) or rabbit antiHCN2 (1:500; Alomone) antibodies overnight at room temperature. The sections were then rinsed in PBS for 15 min, incubated with 2% NDS for 10 min and incubated for 2 h in the biotinylated donkey anti-rabbit antibody (1:200; Jackson ImmunoResearch). After rinsing, the sections were incubated with ExtrAvidin peroxidase (1:5,000; Sigma–Aldrich) for 1 h. The immunoperoxidase was visualized by diaminobenzidine (DAB). Sections were further rinsed in PB, osmicated (1% osmium tetroxide in PB) for 1 h, dehydrated in graded alcohols, flat-embedded in Durcupan ACM (Fluka, Buchs, Switzerland) between strips of Aclar plastic film (EMS, Hatfield, PA, USA). Then, sections were cured for 48 h at 60 °C. Small chips were cut out of the wafers and glued onto blank resin blocks with cyanoacrylate. Thin sections were cut with a diamond knife, mounted on formvar-coated single slot nickel grids, and stained with uranyl acetate and lead citrate. Grids were examined on a Hitachi H 7500 electron microscope (Hitachi, Tokyo, Japan) at 80-kV accelerating voltage. Images were captured with Digital Micrograph software driving a cooled CCD camera (SC1000; Gatan, Pleasanton, CA, USA) attached to the microscope, and saved as TIFF files. The cross-sectional area of 322 HCN1+ and 304 HCN2+ fibers in each of 12–14 sections of the sensory roots of three TGs was measured in micrographs at 12,000 or 25,000 original magnification, using Image J software. Immunopositive fibers were classified into unmyelinated, small myelinated (<20 lm2 in crosssectional area, equivalent to <5 lm in diameter) and large myelinated (>20 lm2 in cross-sectional area, equivalent to >5 lm in diameter), corresponding to C,
Fig. 2. HCN1+ and HCN2+ soma in the trigeminal ganglion. (A, B) Light micrographs showing HCN1+(A) and HCN2+(B) soma; the immunoreactivity appears particularly strong along the plasma membrane. Inset in A shows schematic diagram of rat trigeminal ganglion (shaded region indicates sampled area). (C, D) Size distribution of HCN1+(C) and HCN2+(D) soma. (E) Fractions (%) of small (<600 lm2 in cross-sectional area), medium (600–1200 lm2) and large (>1200 lm2) HCN1+ and HCN2+ soma. HCN1 and HCN2 are most frequently expressed in large neurons. Scale bar = 50 lm.
18
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
Ad and Ab fibers, respectively (Debanne et al., 2011; Boron and Boulpaep, 2012). Inter-animal variability in the proportion of each fiber type for each immunostaining was insignificant (one-way ANOVA), and the data could be pooled per group for the analysis. Immunohistochemical controls To control for specificity of primary antibodies, we processed tissues according to the above protocols, except that primary or secondary antibodies were omitted or blocking peptides were added at various concentrations. Specific immunostaining with the HCN1 and HCN2 antibodies were completely abolished by
preadsorption with the respective blocking peptide (HCN1: AB5884, Chemicon; HCN2: APC-030, Alomone) at a final concentration of 1 lg/ml (Fig. 1). Omission of primary or secondary antibodies also completely abolished the specific staining. At EM, specificity of the immunoreaction was also confirmed by the consistency of immunostaining in adjacent serial thin sections of the same fiber. Western blot Perfusion of rats with saline and removal of TGs were done in the same way as for immunohistochemistry (perfusion with fixative was excluded). All chemicals, unless stated
Fig. 3. Double immunofluorescent staining for HCN1 (A–C, red) and NF200 (A), parvalbumin (B), CGRP (C, green), or for HCN2 (D–F, red) and NF200 (D), parvalbumin (E), or IB4 (F, green) in the trigeminal ganglion. Most of the HCN1+ or HCN2+ soma costain NF200 or parvalbumin. However, HCN1+ and HCN2+ soma rarely costain CGRP or IB4. Scale bar = 50 lm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
otherwise, were purchased from Sigma–Aldrich. The samples including maxillary area of TG were homogenized in extraction buffer (20 mM Tris–HCl pH 7.4, 5 mM EDTA, 140 mM NaCl, 1% Triton X-100, 1 mM Na3VO4, 1 mM PMSF, 10 mM NaF, and 1 lg/ml aprotinin) at 4 °C. The extracts were centrifuged at 12,000g for 20 min at 4 °C. Proteins in supernatant were measured with Bio-Rad Protein Assay kit (Bio-Rad, Hercules, CA, USA), and denatured at 95 °C for 5 min with 5 SDS-loading buffer. Proteins were separated by electrophoresis on SDS–PAGE gel, and transferred to Immobilon-P membranes (EMD Millipore). The membranes were blocked with blocking solution (1
19
TBS, 5% nonfat milk, 0.02% NaN3) for 2 h and incubated overnight at 4 °C with primary antibodies: Mouse anti-bactin (1:2,000, sc81178, Santa Cruz Biotechnology, Dallas, TX, USA), rabbit anti-HCN1 (1:400, AB5884, Chemicon) and rabbit anti-HCN2 (1:400, APC-030, Alomone) antisera. After incubation, the membranes were washed with 1 TBS and incubated with goat antimouse immunoglobulin G (IgG) (1:2,000, sc2005, Santa Cruz Biotechnology) or goat anti-rabbit IgG (1:2,000, sc2004, Santa Cruz Biotechnology) for 1 h at room temperature. For visualization, the membranes were treated with ECL solution (EMD Millipore), according to the manufacturer’s instructions, and exposed on
Fig. 4. HCN1+ and HCN2+ axons in the sensory root of the trigeminal ganglion. (A, B) Electron micrographs showing HCN1+ unmyelinated (arrowheads), small myelinated (asterisks), and large myelinated axons (double asterisks). The electron-dense immunoreaction product is indicated by arrows. Inset in A shows schematic diagram of sensory root of rat trigeminal ganglion (double arrows indicate sampled area in the sensory root). (C, D) Electron micrographs showing HCN2+ unmyelinated (arrowheads) and small myelinated axons (asterisk). (E, F) Proportion of HCN1+(E) and HCN2+(F) axon types. Most HCN1+ and HCN2+ axons are unmyelinated and small myelinated axons (<20 lm2 in cross-sectional area, equivalent to <5 lm in diameter). Scale bar = 500 nm.
20
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
autoradiography film (Agfa, Mortsel, Belgium). Paired Student’s t-test was used to compare mean densities, significance was set at p < 0.05.
RESULTS At LM, the immunostaining for HCN1 and HCN2 in the soma of the TG neurons was concentrated along the cell membrane, with a much weaker staining in the cytoplasm (Fig. 2A, B). The size of the immunostained neurons was 307–2381 lm2 in cross-sectional area for the HCN1+ and 309–2352 lm2 for the HCN2+ neurons but immunostaining was most frequently observed in large neurons (60.7% of the HCN1+ and 51.2% of the HCN2+ neurons), then in medium neurons (36.4% of the HCN1+ and 42.5% of the HCN2+ neurons), and only occasionally in small neurons (2.9% of the HCN1+ and 6.3% of the HCN2+ neurons, Fig. 2C-E). The differences in size distribution between the HCN1-immunopositive (+) and HCN2+ neurons were not significant. Most (70–90%) of the HCN1+ or HCN2+ soma, including almost all of the large soma, in the TG coexpressed NF200 or parvalbumin which are markers for A myelinated fiber neurons or mechanosensitive neurons. However, almost all (96% or more) the HCN1+ and HCN2+ soma did not coexpress CGRP or IB4 which are markers for nociceptive neurons (Fig. 3). At EM, the immunostaining for HCN1 and HCN2 in the sensory root of the TG was confined to axons (Fig. 4A-D); these were mostly unmyelinated and small myelinated (HCN1: 91.9%, HCN2: 95.1%) and only occasionally large myelinated (HCN1: 8.1%, HCN2: 4.9%, Fig. 4E, F). In the dental pulp, at LM, the immunostaining for HCN1 and HCN2 co-localized with that for PGP 9.5+, indicating expression in axons (Fig. 5), particularly at axonal varicosities. Most of the HCN1+ and HCN2+ axons coexpressed CGRP while some of the HCN1+ and HCN2+ axons that coexpressed NF200 or parvalbumin were also observed (Fig. 6). Immunostained axons were few within the radicular pulp and the core of the coronal
pulp (Fig. 7C, D, F), and much more numerous in the peripheral pulp and pulpal horn (Fig. 7A, B, E). HCN1+ and HCN2+ axons ascending toward the dentin between odontoblasts were also frequently observed. There were no obvious differences in the distribution and density of HCN1+ vs. HCN2+ axons. The number of HCN1+ and HCN2+ TG neurons increased significantly following pulpal inflammation (HCN1: CFA 1-day vs. control: 64.7% vs. 58.6%, CFA 3-day vs. control: 67.6% vs. 57.2%, HCN2: CFA 1-day vs. control: 68.9% vs. 58.2%, CFA 3-day vs. control: 67.8% vs. 58.9%, Fig. 8). The intensity of immunostaining also increased significantly following inflammation (HCN1: CFA 1-day vs. control: 73.7% vs. 63.7%, CFA 3-day vs. control: 72.4% vs. 65.3%, HCN2: CFA 1-day vs. control: 74.3% vs. 64.4%, CFA 3-day vs. control: 72.9% vs. 65.8%, Fig. 8D). The protein level of HCN1 and HCN2 in the TG also increased significantly in the CFA 1-day and CFA 3-day groups, compared to the control group (Fig. 9).
DISCUSSION The main findings of the present study are as follows: (1) HCN1 and HCN2 are expressed predominantly in largesized, NF200+ or parvalbumin+ soma in the TG, but mostly in unmyelinated and small myelinated axons in the sensory root. (2) They are expressed in many axons which are CGRP+, especially in the peripheral pulp and pulp horn in the dental pulp. (3) The expression of HCN1 and HCN2 in the TG neurons increased significantly following inflammation of the dental pulp. Neuronal types that express HCN1 and HCN2 mainly in the soma and peripheral axons, respectively, are different By the current convention, large DRG or TG neurons are associated with large myelinated fibers that have fast conduction velocity (Ab) and carry innocuous mechanoreceptive signals, whereas medium-size
Fig. 5. Double immunofluorescent staining for HCN1 (A) or HCN2 (B, red) and PGP9.5 (green) in the coronal portion of the rat dental pulp. Co-staining of HCN1 and HCN2 and PGP9.5 (arrowheads) indicates expression of HCN1 and HCN2 in pulpal axons. Scale bar = 20 lm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
21
Fig. 6. Double immunofluorescent staining for HCN1 (A–C, red) and CGRP (A), NF200 (B), parvalbumin (C, green) or for HCN2 (D–F, red) and CGRP (D), NF200 (E), parvalbumin (F) in the coronal portion of the rat dental pulp. Most of the HCN1+ or HCN2+ axons costain CGRP. Some HCN1+ or HCN2+ axons also costain NF200 or parvalbumin. Arrowheads indicate HCN1+ or HCN2+ axons which costain CGRP, NF200 or parvalbumin. Scale bar = 20 lm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
neurons, associated with small myelinated fibers (Ad), and small-size neurons, associated with unmyelinated fibers (C), carry nociceptive signals (Harper and Lawson, 1985; Debanne et al., 2011; Boron and Boulpaep, 2012). HCN1 and HCN2 were expressed mainly in largesized soma but only occasionally in large myelinated axons. Conversely, they were expressed mostly in small myelinated and unmyelinated axons but only occasionally in small-sized soma. Most of the large HCN1+ or HCN2+ soma in the TG coexpressed NF200 or parvalbumin, but rarely coexpressed CGRP or IB4, suggesting that most of them may belong to mechanosensitive neurons rather than to nociceptive
neurons. That HCN1 and HCN2 were expressed mainly in large-sized, NF200+ or parvalbumin+ soma is consistent with previous studies in the DRG and TG (Kouranova et al., 2008; Hatch et al., 2013; Schnorr et al., 2014) and lends morphological support to existing evidence that Ih attributed to HCN is most prominent in Ab, LTM neurons, but minimal in most C neurons (Hogan and Poroli, 2008: Acosta et al., 2012; Gao et al., 2012). The increase in Ih is associated with neuronal hyperexcitability in the large-sized DRG and TG neurons following nerve injury (Chaplan et al., 2003; Tsuboi et al., 2004). Recent studies also report release
22
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
Fig. 7. Immunofluorescent staining for HCN1 and HCN2 in the rat dental pulp. (A–D) HCN1+ axons in the peripheral portion (A), pulp horn (B), core of coronal portion (C), and radicular portion (D) of the pulp. HCN1 is expressed in a large number of axons that branch extensively in the peripheral pulp and pulp horn (A, B) and in a few axon collaterals in the core of the coronal pulp (C) and radicular pulp (D). Note that HCN1 is densely expressed in the axonal varicosities (arrowheads in B1, enlargement of boxed area in B). Inset in A is a schematic diagram showing distribution of HCN1+ axons in the rat maxillary molar pulp. (E, F) HCN2+ axons in the peripheral portion (E) and the core of coronal portion (F) of the dental pulp. HCN2 is expressed in axons that branch extensively in the peripheral pulp and in few axon collaterals in the core of the coronal pulp. Inset in E is a schematic diagram showing distribution of HCN2+ axons in the rat maxillary molar pulp. Scale bar = 20 lm.
of various molecules such as nerve growth factor and CGRP from the neuronal cell body following orofacial inflammation (Takeda et al., 2005; Shinoda et al., 2011; Yasuda et al., 2012; Matsuura et al., 2013). That the immunoreactivity for HCN1 and HCN2 in the soma of the large-sized TG neurons appeared to be concentrated at the periphery of the soma suggests that these receptors may be specifically targeted to the plasma membrane of the soma in the large-sized neurons, rather than the peripheral axon, where they may play a role in the somatic hyperexcitability under pathologic conditions. Further, the hyperexcitability of the soma may be involved in the modulation of axonal firing (Moosmang et al., 2001; Saito et al., 2006) or directly contribute to the mechanical allodynia through further excitation of adjacent nociceptive neurons by release of molecules including glutamate. On the other hand, the enhanced neuronal excitability and mechanical allodynia following nerve injury or inflammation are suppressed by peripheral application of ZD7288, a nonselective HCN channel blocker (Luo et al., 2007; Weng et al., 2012; Hatch et al., 2013; Schnorr et al., 2014). This suggests that HCN expressed mainly in the unmyelinated and small myelinated axons
may play a role in the peripheral mechanisms of pathological pain. Assuming that they are not involved in acute nociceptive pain but in pathological pain (for review, see Chaplan et al., 2003; Momin et al., 2008; Emery et al., 2011, 2012; Schnorr et al., 2014), HCN1 and HCN2 in peripheral axons may be silent under normal conditions and activated under pathologic conditions.
HCN1 and HCN2 are densely expressed in many axons in the dental pulp Most of the HCN1+ and HCN2+ axons coexpressed SP and CGRP in the dental pulp, suggesting that HCN1 and HCN2 are expressed in nociceptive afferents. However, some HCN1+ and HCN2+ axons coexpressing NF200 or parvalbumin were also observed, suggesting that HCN1 and HCN2 may also be expressed in some large myelinated, LTM axons in the dental pulp. Considering many studies showing existence of Ab, LTM axons that may be involved in nociception in the dental pulp (Fried et al., 1988, 2011; Paik et al., 2009), expression of HCN1 and HCN2 in NF200+ or parvalbumin+ pulpal
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
23
Fig. 8. Expression of HCN1 and HCN2 in TG neurons in the control, CFA 1-day and CFA 3-day rats. (A, B) Immunofluorescent staining for HCN1 (A) and HCN2 (B) in the soma of the TG neurons. (C) Number of HCN1+ and HCN2+ soma. (D) Intensity of immunostaining for HCN1 and HCN2. The number of HCN1+ and HCN2+ soma and the intensity of immunostaining for HCN1 and HCN2 are significantly higher in the CFA 1-day and CFA 3-day groups than in control. ⁄p < 0.05. Scale bar = 50 lm.
Fig. 9. Protein levels of HCN1 and HCN2 in the TG in the control, CFA 1-day, CFA 3-day rats. (A) Representative images of the Western blot assay. (B) Quantitative analysis of HCN1 and HCN2 protein in the TG. Protein levels of HCN1 and HCN2 in the TG are significantly higher in the CFA 1-day, CFA 3-day groups than in control. N = 3 animals in each group. ⁄p < 0.05.
axons suggests possible existence of mechanism for HCN-mediated nociception in the pulpal Ab, LTM axons. HCN1+ and HCN2+ axons were dense in the peripheral pulp and the pulp horn, and in axons ascending toward dentinal tubule between odontoblasts,
but scant in the core of the coronal pulp and in the radicular pulp. Most myelinated pulpal axons lose their myelin in the peripheral pulp (Hildebrand et al., 1995; Paik et al., 2009), where receptors on the axonal plasma membrane become accessible to their ligands. Because
24
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
of this, HCN1 and HCN2 may be primarily activated by inflammatory mediators at the level of the peripheral pulp rather than the core of coronal and radicular pulp. This implies that the HCN channel blocker ZD7288, applied to the peripheral pulp, can be effective for relieving inflammatory pain while avoiding the side effects of systemic application (Dalle and Eisenach, 2005; Hatch et al., 2013). The pattern of immunostaining along the pulpal axon suggests that HCN1 and HCN2 may function primarily at axonal varicosities. Since this is where the vesicular glutamate transporter 1 (VGLUT1) and VGLUT2 are also densely expressed (Paik et al., 2012), activation of HCN1 and HCN2 and VGLUT-mediated glutamate release can co-contribute to amplification of pain signaling. Expression of HCN1 and HCN2 in TG neurons increases following pulpal inflammation The density of immunostaining and protein level for HCN1 and HCN2 in TG neurons increased significantly following pulpal inflammation, confirming the results of previous studies employing models of inflammation (Cho et al., 2009; Acosta et al., 2012; Schnorr et al., 2014), except for a study that found no effect on HCN1 and HCN2 expression in the TG following temporomandibular joint inflammation (Hatch et al., 2013), possibly due to differences in the inflammation model and/or the antibodies used. Our finding is also consistent with the study by Wells et al. (2007) that shows increase in HCN1 and HCN2 immunoreactivity in the TG neurons after 6 h following tooth pulp exposure. This suggests that HCN1 and HCN2 may play an important role in pulpal pain following pulpal inflammation as well as acute pulp injury. This increase may reflect the enhanced expression of HCN1 and HCN2 channels on the cell membrane of the soma, which would increase its excitability (Chaplan et al., 2003; Weng et al., 2012; Acosta et al., 2012; Schnorr et al., 2014). In the present study, significant increase in HCN1+ and HCN2+ neurons throughout maxillary region of the TG was observed following inflammation in the maxillary molar pulps, that may be innervated by only a limited number of TG neurons. This response can involve neurons innervating inflamed pulps as well as the adjacent neurons innervating the non-inflamed tissue: Recent studies reported excitation of adjacent neurons innervating non-inflamed tissue by neurons innervating inflamed tissue through release of various diffusible molecules including nerve growth factor (Takeda et al., 2005; Shinoda et al., 2011; Yasuda et al., 2012). Whether the expression of HCN1 and HCN2 in pulpal axons is enhanced following pulpal inflammation and it contributes to peripheral sensitization is unknown. Because of the dense expression of HCN1and HCN2 in the pulpal axons, most of which are nociceptive, and because peripheral application of ZD7288 attenuates inflammation-induced mechanical hypersensitivity (Weng et al., 2012; Hatch et al., 2013; Schnorr et al., 2014), we consider this possibility likely. Acknowledgments—This work was supported by National Research Foundation of Korea (NRF) grant funded by the Korea
government (MSIP, 2008-0062282). The authors sincerely thank Dr. Juli Valtschanoff for helpful discussion and careful reading of the manuscript. The authors declare that they have no conflict of interest.
REFERENCES Acosta C, McMullan S, Djouhri L, Gao L, Watkins R, Berry C, Dempsey K, Lawson SN (2012) HCN1 and HCN2 in Rat DRG neurons: levels in nociceptors and non-nociceptors, NT3dependence and influence of CFA-induced skin inflammation on HCN2 and NT3 expression. PLoS One 7:e50442. Boron WF, Boulpaep EL (2012) Medical physiology: a cellular and molecular approach. Philadelphia, PA: Saunders/Elsevier. Byers MR (1984) Dental sensory receptors. Int Rev Neurobiol 25:39–94. Chaplan SR, Guo HQ, Lee DH, Luo L, Liu C, Kuei C, Velumian AA, Butler MP, Brown SM, Dubin AE (2003) Neuronal hyperpolarization-activated pacemaker channels drive neuropathic pain. J Neurosci 23:1169–1178. Cho HJ, Staikopoulos V, Furness JB, Jennings EA (2009) Inflammation-induced increase in hyperpolarization-activated, cyclic nucleotide-gated channel protein in trigeminal ganglion neurons and the effect of buprenorphine. Neuroscience 162:453–461. Dalle C, Eisenach JC (2005) Peripheral block of the hyperpolarization-activated cation current (Ih) reduces mechanical allodynia in animal models of postoperative and neuropathic pain. Reg Anesth Pain Med 30:243–248. Debanne D, Campanac E, Bialowas A, Carlier E, Alcaraz G (2011) Axon physiology. Physiol Rev 91:555–602. Emery EC, Young GT, Berrocoso EM, Chen L, McNaughton PA (2011) HCN2 ion channels play a central role in inflammatory and neuropathic pain. Science 333:1462–1466. Emery EC, Young GT, McNaughton PA (2012) HCN2 ion channels: an emerging role as the pacemakers of pain. Trends Pharmacol Sci 33:456–463. Fried K, Aldskogius H, Hildebrand C (1988) Proportion of unmyelinated axons in rat molar and incisor tooth pulps following neonatal capsaicin treatment and/or sympathectomy. Brain Res 463:118–123. Fried K, Sessle BJ, Devor M (2011) The paradox of pain from tooth pulp: low-threshold ‘‘algoneurons’’? Pain 152:2685–2689. Gao LL, McMullan S, Djouhri L, Acosta C, Harper AA, Lawson SN (2012) Expression and properties of hyperpolarization-activated current in rat dorsal root ganglion neurons with known sensory function. J Physiol 590:4691–4705. Harper AA, Lawson SN (1985) Conduction velocity is related to morphological cell type in rat dorsal root ganglion neurones. J Physiol 359:31–46. Hatch RJ, Jennings EA, Ivanusic JJ (2013) Peripheral hyperpolarization-activated cyclic nucleotide-gated channels contribute to inflammation-induced hypersensitivity of the rat temporomandibular joint. Eur J Pain 17:972–982. Hildebrand C, Fried K, Tuisku F, Johansson CS (1995) Teeth and tooth nerves. Prog Neurobiol 45:165–222. Hogan QH, Poroli M (2008) Hyperpolarization-activated current (I(h)) contributes to excitability of primary sensory neurons in rats. Brain Res 1207:102–110. Kim YS, Kim YJ, Paik SK, Cho YS, Kwon TG, Ahn DK, Kim SK, Yoshida A, Bae YC (2009) Expression of metabotropic glutamate receptor mGluR5 in human dental pulp. J Endod 35:690–694. Kouranova EV, Strassle BW, Ring RH, Bowlby MR, Vasilyev DV (2008) Hyperpolarization-activated cyclic nucleotide-gated channel mRNA and protein expression in large versus small diameter dorsal root ganglion neurons: correlation with hyperpolarization-activated current gating. Neuroscience 153:1008–1019. Luo L, Chang L, Brown SM, Ao H, Lee DH, Hiquera ES, Dubin AE, Chaplan SR (2007) Role of peripheral hyperpolarization-activated cyclic nucleotide-modulated channel pacemaker channels in
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25 acute and chronic pain models in the rat. Neuroscience 144:1477–1485. Matsuura S, Shimizu K, Shinoda M, Ohara K, Ogiso B, Honda K, Kataqiri A, Sessle BJ, Urata K, Iwata K (2013) Mechanisms underlying ectopic persistent tooth-pulp pain following pulpal inflammation. PLoS One 8:e52840. Momin A, Cadiou H, Mason A, McNaughton PA (2008) Role of the hyperpolarization-activated current Ih in somatosensory neurons. J Physiol 586:5911–5929. Moosmang S, Stieber J, Zong X, Biel M, Hofmann F, Ludwig A (2001) Cellular expression and functional characterization of four hyperpolarization-activated pacemaker channels in cardiac and neuronal tissues. Eur J Biochem 268:1646–1652. Paik SK, Park KP, Lee SK, Ma SK, Cho YS, Kim YK, Rhyu IJ, Ahn DK, Yoshida A, Bae YC (2009) Light and electron microscopic analysis of the somata and parent axons innervating the rat upper molar and lower incisor pulp. Neuroscience 162: 1279–1286. Paik SK, Kim SK, Choi SJ, Yang ES, Ahn SH, Bae YC (2012) Vesicular glutamate transporters in axons that innervate the human dental pulp. J Endod 38:470–474. Saito M, Murai Y, Sato H, Bae YC, Akaike T, Takada M, Kang Y (2006) Two opposing roles of 4-AP-sensitive K+ current in initiation and invasion of spikes in rat mesencephalic trigeminal neurons. J Neurophysiol 96:1887–1901. Schnorr S, Eberhardt M, Kistner K, Rajab H, Kasser J, Hess A, Reeh P, Ludwig A, Herrmann S (2014) HCN2 channels account for mechanical (but not heat) hyperalgesia during long-standing inflammation. Pain 155:1079–1090. Shinoda M, Asano M, Omagari D, Honda K, Hitomi S, Katagiri A, Iwata K (2011) Nerve growth factor contribution via transient
25
receptor potential vanilloid 1 to ectopic orofacial pain. J Neurosci 31:7145–7155. Takeda M, Tanimoto T, Nasu M, Ikeda M, Kadoi J, Matsumoto S (2005) Activation of NK1 receptor of trigeminal root ganglion via substance P paracrine mechanism contributes to the mechanical allodynia in the temporomandibular joint inflammation in rats. Pain 116:375–385. Tsuboi Y, Takeda M, Tanimoto T, Ikeda M, Matsumoto S, Kitagawa J, Teramoto K, Simizu K, Yamazaki Y, Shima A, Ren K, Iwata K (2004) Alteration of the second branch of the trigeminal nerve activity following inferior alveolar nerve transection in rats. Pain 111:323–334. Tu H, Deng L, Sun Q, Yao L, Han JS, Wan Y (2004) Hyperpolarization-activated, cyclic nucleotide-gated cation channels: roles in the differential electrophysiological properties of rat primary afferent neurons. J Neurosci Res 76:713–722. Wells JE, Rowland KC, Proctor EK (2007) Hyperpolarizationactivated channels in trigeminal ganglia innervating healthy and pulp-exposed teeth. Int Endod J 40:715–721. Weng X, Smith T, Sathish J, Djouhri L (2012) Chronic inflammatory pain is associated with increased excitability and hyperpolarization-activated current (Ih) in C- but not Adeltanociceptors. Pain 153:900–914. Yao H, Donnelly DF, Ma C, LaMotte RH (2003) Upregulation of the hyperpolarization-activated cation current after chronic compression of the dorsal root ganglion. J Neurosci 23:2069–2074. Yasuda M, Shinoda M, Kiyomoto M, Honda K, Suzuki A, Tamagawa T, Kaji K, Kimoto S, Iwata K (2012) P2X3 receptor mediates ectopic mechanical allodynia with inflamed lower lip in mice. Neurosci Lett 528:67–72.
(Accepted 28 January 2015) (Available online 7 February 2015)
THE EXPRESSION OF HYPERPOLARIZATION-ACTIVATED CYCLIC NUCLEOTIDE-GATED CHANNEL 1 (HCN1) AND HCN2 IN THE RAT TRIGEMINAL GANGLION, SENSORY ROOT, AND DENTAL PULP Y. S. CHO, Y. S. KIM, S. J. MOOZHAYIL, E. S. YANG AND Y. C. BAE *
INTRODUCTION
Department of Anatomy and Neurobiology, School of Dentistry, Kyungpook National University, Daegu 700-412, Republic of Korea
Hyperpolarization-activated cyclic nucleotide-gated channel 1 (HCN1) and 2 (HCN2), of the discovered four isoforms of HCN channels, are most abundantly expressed in the primary sensory neurons in the dorsal root ganglion (DRG) and trigeminal ganglion (TG, Chaplan et al., 2003; Yao et al., 2003). They contribute to hyperpolarization-activated currents (Ih) and neuronal excitability, and are implicated in pathological pain (Chaplan et al., 2003; Tu et al., 2004; Emery et al., 2011; Weng et al., 2012). What type of neuron in the DRG and TG expresses HCN1 and HCN2 is still uncertain: the expression in large-sized soma (Chaplan et al., 2003; Hogan and Poroli, 2008; Cho et al., 2009; Acosta et al., 2012; Gao et al., 2012; Schnorr et al., 2014) suggests that HCN1- and HCN2-positive neurons are predominantly low threshold mechanoreceptive (LTM), whereas studies that implicate HCN in allodynia and hyperalgesia following inflammation and nerve injury (Chaplan et al., 2003; Cho et al., 2009; Emery et al., 2012) suggest that HCN may be expressed in small, nociceptive neurons. Peripheral application of a specific HCN blocker attenuated mechanical hypersensitivity in a model of inflammation (Weng et al., 2012; Hatch et al., 2013), suggesting that HCN expressed in peripheral axons plays a role in the peripheral sensitization and inflammatory pain. However, little is known about the HCN expression in peripheral axons, particularly in the dental pulp. Dental pulp is innervated densely by nociceptive afferents that signal pain in response to a variety of stimuli (Byers, 1984; Hildebrand et al., 1995; Paik et al., 2009). We showed that glutamate receptors and vesicular glutamate transporters (VGLUTs) are expressed in axons of the dental pulp (Kim et al., 2009; Paik et al., 2012), suggesting peripheral pain processing in the pulp. In addition, even though HCN1 and HCN2 are implicated in inflammatory pain (Chaplan et al., 2003; Emery et al., 2012; Schnorr et al., 2014), it remains uncertain whether their expression in TG neurons is associated with pulpal inflammatory pain (Wells et al., 2007). To address these issues, we studied the expression of HCN1 and HCN2 in the soma and axons in the rat TG neurons and axons in the dental pulp, and the changes in that expression following inflammation, using lightand electron-microscopic immunocytochemistry and quantitative analysis.
Abstract—Hyperpolarization-activated cyclic nucleotidegated channel 1 (HCN1) and 2 (HCN2) are abundantly expressed in primary sensory neurons and contribute to neuronal excitability and pathological pain. We studied the expression of HCN1 and HCN2 in the rat trigeminal ganglion (TG) neurons and axons in the dental pulp, and the changes in their expression following inflammation, using light- and electron-microscopic immunocytochemistry and quantitative analysis. HCN1 and HCN2 were expressed predominantly in large-sized, neurofilament 200-immunopositive (+) or parvalbumin+ soma in the TG whereas they were expressed mostly in unmyelinated and small myelinated axons in the sensory root. The expression was particularly strong along the plasma membrane in the soma. In the dental pulp, majority of HCN1+ and HCN2+ axons coexpressed calcitonin gene-related peptide. They were expressed mainly in the peripheral pulp and pulp horn where the axons branch extensively in the dental pulp. The expression of HCN1 and HCN2 in TG neurons increased significantly in rats with experimentally induced inflammation of the dental pulp. Our findings support the notion that HCN1 and HCN2 are expressed mainly by both the soma of mechanosensitive neurons in the TG and peripheral axons of nociceptive neurons in the sensory root, and may play a role in the mechanisms of inflammatory pain from the dental pulp. Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved.
Key words: HCN, axon type, trigeminal ganglion, dental pulp, inflammation, ultrastructure.
*Corresponding author. Address: Department of Anatomy and Neurobiology, School of Dentistry, Kyungpook National University, 188-1, 2-Ga, Samdeok-Dong, Jung-Gu, Daegu 700-412, Republic of Korea. Tel: +82-53-660-6860; fax: +82-53-426-7731. E-mail address: [email protected] (Y. C. Bae). Abbreviations: CGRP, calcitonin gene-related peptide; CFA, Complete Freund’s Adjuvant; DRG, dorsal root ganglion; EDTA, ethylenediaminetetraacetic acid; EM, electron microscopy; HCN, hyperpolarization-activated cyclic nucleotide-gated channel; IgG, immunoglobulin G; Ih, hyperpolarization-activated currents; LM, light microscopy; LTM, low threshold mechanoreceptive; NDS, normal donkey serum; NF200, neurofilament 200; PB, phosphate buffer; PBS, phosphate-buffered saline; PGP9.5, protein gene product 9.5; TG, trigeminal ganglion; VGLUT, vesicular glutamate transporter. http://dx.doi.org/10.1016/j.neuroscience.2015.01.066 0306-4522/Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved. 15
16
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
EXPERIMENTAL PROCEDURES All animal procedures were performed according to the National Institutes of Health guidelines and were approved by the Kyungpook National University Intramural Animal Care and Use Committee. Experiments were designed to minimize the number of animals used and their suffering. Thirty male Sprague– Dawley rats weighing 300–320 g were used for this study: Fifteen rats were for light microscopic (LM) immunohistochemistry including three for normal group, and each three for Complete Freund’s Adjuvant (CFA) treatment with 1-day, 3-day survival and their controls, and three rats were for electron microscopic (EM) immunohistochemistry. Twelve rats were for Western blot analysis including each three for CFA treatment with 1-day, 3-days survival and their controls. Tooth pulp inflammation model Rats were anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and the occlusal enamel and dentin of the right maxillary 1st (M1) and 2nd (M2) molars were filed off to just before exposing the pulp using a lowspeed dental drill with a round bur under water-cooling. A small piece of tissue paper soaked in 50% solution of CFA in saline or saline for control was applied to the exposed dentinal surfaces for 5 min, and after that the dentinal surfaces were sealed with dental cement. One or 3 days later, rats were re-anesthetized and sacrificed with intravascular perfusion with fixatives. LM immunohistochemistry For immunofluorescence, rats were deeply anesthetized with sodium pentobarbital (80 mg/kg, i.p.), transcardially perfused with heparinized normal saline, followed by freshly prepared fixative containing 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4). The dental pulps of the right maxillary 1st (M1) and 2nd (M2) molars and the right TGs in normal rats, and the right TGs of CFA-treated rats (experimental) and of saline-treated rats (control) were carefully removed, postfixed in the same fixative for 2 h, and cryoprotected in 30% sucrose in PB overnight at 4 °C. The next day, 30lm-thick sections were cut on a freezing microtome and treated with 50% ethanol for 30 min and with 10% normal donkey serum (NDS, Jackson ImmunoResearch, West Groove, PA, USA) for 30 min. Sections were incubated overnight in rabbit anti-HCN1 (1:800, AB5884, Chemicon, Billerica, MA, USA) or rabbit anti-HCN2 (1:800, APC-030,
Alomone, Jerusalem, Israel) antibodies alone or in combination with mouse anti-protein gene product 9.5 (PGP 9.5, 1:5,000, YM8104, Accurate chemical and scientific Corp., Westbury, NY, USA) antibodies. To verify the identity of HCN1- and HCN2-immunopositive soma and pulpal axons, sections were also incubated overnight in rabbit anti-HCN1 or rabbit anti-HCN2 antibodies in combination with mouse anti-neurofilament 200 (NF200, 1:100,000, N0142, Sigma–Aldrich, St.Louis, MO, USA), mouse anti-parvalbumin (1:1,000, MAB1572, Chemicon), mouse anti-calcitonin gene-related peptide (CGRP, 1:1,000, ab81887, Abcam, Cambridge, MA, USA) or biotinylated-isolectin B4 (biotinylated-IB4, 5 lg/ml, L-3759, Sigma–Aldrich). On the next day, the sections were washed with phosphate-buffered saline (PBS; 0.01 M, pH7.4) and incubated for 3 h with Cy3- or fluorescein isothiocyanate-conjugated secondary antibodies (raised in donkey, 1:200, in PB, Jackson ImmunoResearch). For biotinylated-IB4 staining, sections were incubated for 3 h with streptavidin-Alexa488 (1:600, S-11223, Molecular probes, Grand Island, NY, USA). Finally, sections were mounted on slides and coverslipped with Vectashield (Vector Laboratories, Burlingame, CA, USA), and micrographs were obtained with an Exi digital camera (Q-Imaging Inc., Surrey, CA, USA), attached to a Zeiss Axioplan 2 fluorescence microscope (Carl Zeiss Inc., Jena, Germany). Quantitative analysis To assess the size distribution of HCN1-immunopositive (+) and HCN2+ neurons in the TG, cross-sectional area of all neuronal somata with clearly visible nucleoli was measured using Image J software (http://imagej. nih.gov/ij/, NIH, Bethesda, MD, USA) and graphs were built in KaleidaGraph (v 3.5; Synergy Software, Reading, PA, USA). HCN1+ and HCN2+ soma were divided into three groups, according to their crosssectional area: small (<600 lm2 in cross-sectional area), medium (600–1200 lm2) and large (>1200 lm2) and their proportions were analyzed. To quantify the immunoreactivity in HCN1+ and HCN2+ TG perikarya in the CFA-treated rats, a total of 12–16 images were collected from 3 to 4 sections of each TGs in each saline- and CFA-treated rat (1-day or 3-days). The images were captured with a 20 objective (857.14 652.94 lm, 1360 1036 pixels) from the maxillary region of the TG. The images were converted to grayscale and all images for each antibody were identically enhanced for brightness and contrast. The
Fig. 1. Immunofluorescent staining for HCN1 (A) and HCN2 (B) in trigeminal ganglion neurons is completely abolished by preadsorption with the corresponding blocking peptides (HCN1: 1 lg/ml, HCN2: 1 lg/ml). Scale bar = 50 lm.
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
17
threshold level for defining a neuron as immunopositive was set at 100–120 (out of 255) gray levels using Image J software. The fraction of HCN1+ or HCN2+ somata of all TG somata and the differences in the mean gray value of immunoreactivity in HCN1+ and HCN2+ somata between CFA- and control groups were analyzed with Student’s t-test; statistical significance was set at p < 0.05. EM immunohistochemistry For EM, rats were deeply anesthetized with sodium pentobarbital (80 mg/kg, i.p.), transcardially perfused with heparinized normal saline, followed by freshly prepared fixative containing 4% paraformaldehyde and 0.01% glutaraldehyde in PB. The peripheral sensory roots (maxillary branch of the trigeminal nerve) of the TG were dissected out under a microscope and postfixed in the same fixative for 2–4 h at 4 °C. Sections of the sensory roots were cut transversely on a Vibratome at 60-lm thickness and cryoprotected in 30% sucrose in PB overnight at 4 °C. The next day, sections of the sensory root were frozen on the dry ice for 20 min, thawed in PBS to enhance penetration, pretreated with 1% sodium borohydride for 30 min to quench glutaraldehyde and then blocked with 3% H2O2 for 10 min and with 10% NDS for 30 min and incubated in rabbit anti-HCN1 (1:500; Chemicon) or rabbit antiHCN2 (1:500; Alomone) antibodies overnight at room temperature. The sections were then rinsed in PBS for 15 min, incubated with 2% NDS for 10 min and incubated for 2 h in the biotinylated donkey anti-rabbit antibody (1:200; Jackson ImmunoResearch). After rinsing, the sections were incubated with ExtrAvidin peroxidase (1:5,000; Sigma–Aldrich) for 1 h. The immunoperoxidase was visualized by diaminobenzidine (DAB). Sections were further rinsed in PB, osmicated (1% osmium tetroxide in PB) for 1 h, dehydrated in graded alcohols, flat-embedded in Durcupan ACM (Fluka, Buchs, Switzerland) between strips of Aclar plastic film (EMS, Hatfield, PA, USA). Then, sections were cured for 48 h at 60 °C. Small chips were cut out of the wafers and glued onto blank resin blocks with cyanoacrylate. Thin sections were cut with a diamond knife, mounted on formvar-coated single slot nickel grids, and stained with uranyl acetate and lead citrate. Grids were examined on a Hitachi H 7500 electron microscope (Hitachi, Tokyo, Japan) at 80-kV accelerating voltage. Images were captured with Digital Micrograph software driving a cooled CCD camera (SC1000; Gatan, Pleasanton, CA, USA) attached to the microscope, and saved as TIFF files. The cross-sectional area of 322 HCN1+ and 304 HCN2+ fibers in each of 12–14 sections of the sensory roots of three TGs was measured in micrographs at 12,000 or 25,000 original magnification, using Image J software. Immunopositive fibers were classified into unmyelinated, small myelinated (<20 lm2 in crosssectional area, equivalent to <5 lm in diameter) and large myelinated (>20 lm2 in cross-sectional area, equivalent to >5 lm in diameter), corresponding to C,
Fig. 2. HCN1+ and HCN2+ soma in the trigeminal ganglion. (A, B) Light micrographs showing HCN1+(A) and HCN2+(B) soma; the immunoreactivity appears particularly strong along the plasma membrane. Inset in A shows schematic diagram of rat trigeminal ganglion (shaded region indicates sampled area). (C, D) Size distribution of HCN1+(C) and HCN2+(D) soma. (E) Fractions (%) of small (<600 lm2 in cross-sectional area), medium (600–1200 lm2) and large (>1200 lm2) HCN1+ and HCN2+ soma. HCN1 and HCN2 are most frequently expressed in large neurons. Scale bar = 50 lm.
18
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
Ad and Ab fibers, respectively (Debanne et al., 2011; Boron and Boulpaep, 2012). Inter-animal variability in the proportion of each fiber type for each immunostaining was insignificant (one-way ANOVA), and the data could be pooled per group for the analysis. Immunohistochemical controls To control for specificity of primary antibodies, we processed tissues according to the above protocols, except that primary or secondary antibodies were omitted or blocking peptides were added at various concentrations. Specific immunostaining with the HCN1 and HCN2 antibodies were completely abolished by
preadsorption with the respective blocking peptide (HCN1: AB5884, Chemicon; HCN2: APC-030, Alomone) at a final concentration of 1 lg/ml (Fig. 1). Omission of primary or secondary antibodies also completely abolished the specific staining. At EM, specificity of the immunoreaction was also confirmed by the consistency of immunostaining in adjacent serial thin sections of the same fiber. Western blot Perfusion of rats with saline and removal of TGs were done in the same way as for immunohistochemistry (perfusion with fixative was excluded). All chemicals, unless stated
Fig. 3. Double immunofluorescent staining for HCN1 (A–C, red) and NF200 (A), parvalbumin (B), CGRP (C, green), or for HCN2 (D–F, red) and NF200 (D), parvalbumin (E), or IB4 (F, green) in the trigeminal ganglion. Most of the HCN1+ or HCN2+ soma costain NF200 or parvalbumin. However, HCN1+ and HCN2+ soma rarely costain CGRP or IB4. Scale bar = 50 lm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
otherwise, were purchased from Sigma–Aldrich. The samples including maxillary area of TG were homogenized in extraction buffer (20 mM Tris–HCl pH 7.4, 5 mM EDTA, 140 mM NaCl, 1% Triton X-100, 1 mM Na3VO4, 1 mM PMSF, 10 mM NaF, and 1 lg/ml aprotinin) at 4 °C. The extracts were centrifuged at 12,000g for 20 min at 4 °C. Proteins in supernatant were measured with Bio-Rad Protein Assay kit (Bio-Rad, Hercules, CA, USA), and denatured at 95 °C for 5 min with 5 SDS-loading buffer. Proteins were separated by electrophoresis on SDS–PAGE gel, and transferred to Immobilon-P membranes (EMD Millipore). The membranes were blocked with blocking solution (1
19
TBS, 5% nonfat milk, 0.02% NaN3) for 2 h and incubated overnight at 4 °C with primary antibodies: Mouse anti-bactin (1:2,000, sc81178, Santa Cruz Biotechnology, Dallas, TX, USA), rabbit anti-HCN1 (1:400, AB5884, Chemicon) and rabbit anti-HCN2 (1:400, APC-030, Alomone) antisera. After incubation, the membranes were washed with 1 TBS and incubated with goat antimouse immunoglobulin G (IgG) (1:2,000, sc2005, Santa Cruz Biotechnology) or goat anti-rabbit IgG (1:2,000, sc2004, Santa Cruz Biotechnology) for 1 h at room temperature. For visualization, the membranes were treated with ECL solution (EMD Millipore), according to the manufacturer’s instructions, and exposed on
Fig. 4. HCN1+ and HCN2+ axons in the sensory root of the trigeminal ganglion. (A, B) Electron micrographs showing HCN1+ unmyelinated (arrowheads), small myelinated (asterisks), and large myelinated axons (double asterisks). The electron-dense immunoreaction product is indicated by arrows. Inset in A shows schematic diagram of sensory root of rat trigeminal ganglion (double arrows indicate sampled area in the sensory root). (C, D) Electron micrographs showing HCN2+ unmyelinated (arrowheads) and small myelinated axons (asterisk). (E, F) Proportion of HCN1+(E) and HCN2+(F) axon types. Most HCN1+ and HCN2+ axons are unmyelinated and small myelinated axons (<20 lm2 in cross-sectional area, equivalent to <5 lm in diameter). Scale bar = 500 nm.
20
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
autoradiography film (Agfa, Mortsel, Belgium). Paired Student’s t-test was used to compare mean densities, significance was set at p < 0.05.
RESULTS At LM, the immunostaining for HCN1 and HCN2 in the soma of the TG neurons was concentrated along the cell membrane, with a much weaker staining in the cytoplasm (Fig. 2A, B). The size of the immunostained neurons was 307–2381 lm2 in cross-sectional area for the HCN1+ and 309–2352 lm2 for the HCN2+ neurons but immunostaining was most frequently observed in large neurons (60.7% of the HCN1+ and 51.2% of the HCN2+ neurons), then in medium neurons (36.4% of the HCN1+ and 42.5% of the HCN2+ neurons), and only occasionally in small neurons (2.9% of the HCN1+ and 6.3% of the HCN2+ neurons, Fig. 2C-E). The differences in size distribution between the HCN1-immunopositive (+) and HCN2+ neurons were not significant. Most (70–90%) of the HCN1+ or HCN2+ soma, including almost all of the large soma, in the TG coexpressed NF200 or parvalbumin which are markers for A myelinated fiber neurons or mechanosensitive neurons. However, almost all (96% or more) the HCN1+ and HCN2+ soma did not coexpress CGRP or IB4 which are markers for nociceptive neurons (Fig. 3). At EM, the immunostaining for HCN1 and HCN2 in the sensory root of the TG was confined to axons (Fig. 4A-D); these were mostly unmyelinated and small myelinated (HCN1: 91.9%, HCN2: 95.1%) and only occasionally large myelinated (HCN1: 8.1%, HCN2: 4.9%, Fig. 4E, F). In the dental pulp, at LM, the immunostaining for HCN1 and HCN2 co-localized with that for PGP 9.5+, indicating expression in axons (Fig. 5), particularly at axonal varicosities. Most of the HCN1+ and HCN2+ axons coexpressed CGRP while some of the HCN1+ and HCN2+ axons that coexpressed NF200 or parvalbumin were also observed (Fig. 6). Immunostained axons were few within the radicular pulp and the core of the coronal
pulp (Fig. 7C, D, F), and much more numerous in the peripheral pulp and pulpal horn (Fig. 7A, B, E). HCN1+ and HCN2+ axons ascending toward the dentin between odontoblasts were also frequently observed. There were no obvious differences in the distribution and density of HCN1+ vs. HCN2+ axons. The number of HCN1+ and HCN2+ TG neurons increased significantly following pulpal inflammation (HCN1: CFA 1-day vs. control: 64.7% vs. 58.6%, CFA 3-day vs. control: 67.6% vs. 57.2%, HCN2: CFA 1-day vs. control: 68.9% vs. 58.2%, CFA 3-day vs. control: 67.8% vs. 58.9%, Fig. 8). The intensity of immunostaining also increased significantly following inflammation (HCN1: CFA 1-day vs. control: 73.7% vs. 63.7%, CFA 3-day vs. control: 72.4% vs. 65.3%, HCN2: CFA 1-day vs. control: 74.3% vs. 64.4%, CFA 3-day vs. control: 72.9% vs. 65.8%, Fig. 8D). The protein level of HCN1 and HCN2 in the TG also increased significantly in the CFA 1-day and CFA 3-day groups, compared to the control group (Fig. 9).
DISCUSSION The main findings of the present study are as follows: (1) HCN1 and HCN2 are expressed predominantly in largesized, NF200+ or parvalbumin+ soma in the TG, but mostly in unmyelinated and small myelinated axons in the sensory root. (2) They are expressed in many axons which are CGRP+, especially in the peripheral pulp and pulp horn in the dental pulp. (3) The expression of HCN1 and HCN2 in the TG neurons increased significantly following inflammation of the dental pulp. Neuronal types that express HCN1 and HCN2 mainly in the soma and peripheral axons, respectively, are different By the current convention, large DRG or TG neurons are associated with large myelinated fibers that have fast conduction velocity (Ab) and carry innocuous mechanoreceptive signals, whereas medium-size
Fig. 5. Double immunofluorescent staining for HCN1 (A) or HCN2 (B, red) and PGP9.5 (green) in the coronal portion of the rat dental pulp. Co-staining of HCN1 and HCN2 and PGP9.5 (arrowheads) indicates expression of HCN1 and HCN2 in pulpal axons. Scale bar = 20 lm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
21
Fig. 6. Double immunofluorescent staining for HCN1 (A–C, red) and CGRP (A), NF200 (B), parvalbumin (C, green) or for HCN2 (D–F, red) and CGRP (D), NF200 (E), parvalbumin (F) in the coronal portion of the rat dental pulp. Most of the HCN1+ or HCN2+ axons costain CGRP. Some HCN1+ or HCN2+ axons also costain NF200 or parvalbumin. Arrowheads indicate HCN1+ or HCN2+ axons which costain CGRP, NF200 or parvalbumin. Scale bar = 20 lm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
neurons, associated with small myelinated fibers (Ad), and small-size neurons, associated with unmyelinated fibers (C), carry nociceptive signals (Harper and Lawson, 1985; Debanne et al., 2011; Boron and Boulpaep, 2012). HCN1 and HCN2 were expressed mainly in largesized soma but only occasionally in large myelinated axons. Conversely, they were expressed mostly in small myelinated and unmyelinated axons but only occasionally in small-sized soma. Most of the large HCN1+ or HCN2+ soma in the TG coexpressed NF200 or parvalbumin, but rarely coexpressed CGRP or IB4, suggesting that most of them may belong to mechanosensitive neurons rather than to nociceptive
neurons. That HCN1 and HCN2 were expressed mainly in large-sized, NF200+ or parvalbumin+ soma is consistent with previous studies in the DRG and TG (Kouranova et al., 2008; Hatch et al., 2013; Schnorr et al., 2014) and lends morphological support to existing evidence that Ih attributed to HCN is most prominent in Ab, LTM neurons, but minimal in most C neurons (Hogan and Poroli, 2008: Acosta et al., 2012; Gao et al., 2012). The increase in Ih is associated with neuronal hyperexcitability in the large-sized DRG and TG neurons following nerve injury (Chaplan et al., 2003; Tsuboi et al., 2004). Recent studies also report release
22
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
Fig. 7. Immunofluorescent staining for HCN1 and HCN2 in the rat dental pulp. (A–D) HCN1+ axons in the peripheral portion (A), pulp horn (B), core of coronal portion (C), and radicular portion (D) of the pulp. HCN1 is expressed in a large number of axons that branch extensively in the peripheral pulp and pulp horn (A, B) and in a few axon collaterals in the core of the coronal pulp (C) and radicular pulp (D). Note that HCN1 is densely expressed in the axonal varicosities (arrowheads in B1, enlargement of boxed area in B). Inset in A is a schematic diagram showing distribution of HCN1+ axons in the rat maxillary molar pulp. (E, F) HCN2+ axons in the peripheral portion (E) and the core of coronal portion (F) of the dental pulp. HCN2 is expressed in axons that branch extensively in the peripheral pulp and in few axon collaterals in the core of the coronal pulp. Inset in E is a schematic diagram showing distribution of HCN2+ axons in the rat maxillary molar pulp. Scale bar = 20 lm.
of various molecules such as nerve growth factor and CGRP from the neuronal cell body following orofacial inflammation (Takeda et al., 2005; Shinoda et al., 2011; Yasuda et al., 2012; Matsuura et al., 2013). That the immunoreactivity for HCN1 and HCN2 in the soma of the large-sized TG neurons appeared to be concentrated at the periphery of the soma suggests that these receptors may be specifically targeted to the plasma membrane of the soma in the large-sized neurons, rather than the peripheral axon, where they may play a role in the somatic hyperexcitability under pathologic conditions. Further, the hyperexcitability of the soma may be involved in the modulation of axonal firing (Moosmang et al., 2001; Saito et al., 2006) or directly contribute to the mechanical allodynia through further excitation of adjacent nociceptive neurons by release of molecules including glutamate. On the other hand, the enhanced neuronal excitability and mechanical allodynia following nerve injury or inflammation are suppressed by peripheral application of ZD7288, a nonselective HCN channel blocker (Luo et al., 2007; Weng et al., 2012; Hatch et al., 2013; Schnorr et al., 2014). This suggests that HCN expressed mainly in the unmyelinated and small myelinated axons
may play a role in the peripheral mechanisms of pathological pain. Assuming that they are not involved in acute nociceptive pain but in pathological pain (for review, see Chaplan et al., 2003; Momin et al., 2008; Emery et al., 2011, 2012; Schnorr et al., 2014), HCN1 and HCN2 in peripheral axons may be silent under normal conditions and activated under pathologic conditions.
HCN1 and HCN2 are densely expressed in many axons in the dental pulp Most of the HCN1+ and HCN2+ axons coexpressed SP and CGRP in the dental pulp, suggesting that HCN1 and HCN2 are expressed in nociceptive afferents. However, some HCN1+ and HCN2+ axons coexpressing NF200 or parvalbumin were also observed, suggesting that HCN1 and HCN2 may also be expressed in some large myelinated, LTM axons in the dental pulp. Considering many studies showing existence of Ab, LTM axons that may be involved in nociception in the dental pulp (Fried et al., 1988, 2011; Paik et al., 2009), expression of HCN1 and HCN2 in NF200+ or parvalbumin+ pulpal
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
23
Fig. 8. Expression of HCN1 and HCN2 in TG neurons in the control, CFA 1-day and CFA 3-day rats. (A, B) Immunofluorescent staining for HCN1 (A) and HCN2 (B) in the soma of the TG neurons. (C) Number of HCN1+ and HCN2+ soma. (D) Intensity of immunostaining for HCN1 and HCN2. The number of HCN1+ and HCN2+ soma and the intensity of immunostaining for HCN1 and HCN2 are significantly higher in the CFA 1-day and CFA 3-day groups than in control. ⁄p < 0.05. Scale bar = 50 lm.
Fig. 9. Protein levels of HCN1 and HCN2 in the TG in the control, CFA 1-day, CFA 3-day rats. (A) Representative images of the Western blot assay. (B) Quantitative analysis of HCN1 and HCN2 protein in the TG. Protein levels of HCN1 and HCN2 in the TG are significantly higher in the CFA 1-day, CFA 3-day groups than in control. N = 3 animals in each group. ⁄p < 0.05.
axons suggests possible existence of mechanism for HCN-mediated nociception in the pulpal Ab, LTM axons. HCN1+ and HCN2+ axons were dense in the peripheral pulp and the pulp horn, and in axons ascending toward dentinal tubule between odontoblasts,
but scant in the core of the coronal pulp and in the radicular pulp. Most myelinated pulpal axons lose their myelin in the peripheral pulp (Hildebrand et al., 1995; Paik et al., 2009), where receptors on the axonal plasma membrane become accessible to their ligands. Because
24
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25
of this, HCN1 and HCN2 may be primarily activated by inflammatory mediators at the level of the peripheral pulp rather than the core of coronal and radicular pulp. This implies that the HCN channel blocker ZD7288, applied to the peripheral pulp, can be effective for relieving inflammatory pain while avoiding the side effects of systemic application (Dalle and Eisenach, 2005; Hatch et al., 2013). The pattern of immunostaining along the pulpal axon suggests that HCN1 and HCN2 may function primarily at axonal varicosities. Since this is where the vesicular glutamate transporter 1 (VGLUT1) and VGLUT2 are also densely expressed (Paik et al., 2012), activation of HCN1 and HCN2 and VGLUT-mediated glutamate release can co-contribute to amplification of pain signaling. Expression of HCN1 and HCN2 in TG neurons increases following pulpal inflammation The density of immunostaining and protein level for HCN1 and HCN2 in TG neurons increased significantly following pulpal inflammation, confirming the results of previous studies employing models of inflammation (Cho et al., 2009; Acosta et al., 2012; Schnorr et al., 2014), except for a study that found no effect on HCN1 and HCN2 expression in the TG following temporomandibular joint inflammation (Hatch et al., 2013), possibly due to differences in the inflammation model and/or the antibodies used. Our finding is also consistent with the study by Wells et al. (2007) that shows increase in HCN1 and HCN2 immunoreactivity in the TG neurons after 6 h following tooth pulp exposure. This suggests that HCN1 and HCN2 may play an important role in pulpal pain following pulpal inflammation as well as acute pulp injury. This increase may reflect the enhanced expression of HCN1 and HCN2 channels on the cell membrane of the soma, which would increase its excitability (Chaplan et al., 2003; Weng et al., 2012; Acosta et al., 2012; Schnorr et al., 2014). In the present study, significant increase in HCN1+ and HCN2+ neurons throughout maxillary region of the TG was observed following inflammation in the maxillary molar pulps, that may be innervated by only a limited number of TG neurons. This response can involve neurons innervating inflamed pulps as well as the adjacent neurons innervating the non-inflamed tissue: Recent studies reported excitation of adjacent neurons innervating non-inflamed tissue by neurons innervating inflamed tissue through release of various diffusible molecules including nerve growth factor (Takeda et al., 2005; Shinoda et al., 2011; Yasuda et al., 2012). Whether the expression of HCN1 and HCN2 in pulpal axons is enhanced following pulpal inflammation and it contributes to peripheral sensitization is unknown. Because of the dense expression of HCN1and HCN2 in the pulpal axons, most of which are nociceptive, and because peripheral application of ZD7288 attenuates inflammation-induced mechanical hypersensitivity (Weng et al., 2012; Hatch et al., 2013; Schnorr et al., 2014), we consider this possibility likely. Acknowledgments—This work was supported by National Research Foundation of Korea (NRF) grant funded by the Korea
government (MSIP, 2008-0062282). The authors sincerely thank Dr. Juli Valtschanoff for helpful discussion and careful reading of the manuscript. The authors declare that they have no conflict of interest.
REFERENCES Acosta C, McMullan S, Djouhri L, Gao L, Watkins R, Berry C, Dempsey K, Lawson SN (2012) HCN1 and HCN2 in Rat DRG neurons: levels in nociceptors and non-nociceptors, NT3dependence and influence of CFA-induced skin inflammation on HCN2 and NT3 expression. PLoS One 7:e50442. Boron WF, Boulpaep EL (2012) Medical physiology: a cellular and molecular approach. Philadelphia, PA: Saunders/Elsevier. Byers MR (1984) Dental sensory receptors. Int Rev Neurobiol 25:39–94. Chaplan SR, Guo HQ, Lee DH, Luo L, Liu C, Kuei C, Velumian AA, Butler MP, Brown SM, Dubin AE (2003) Neuronal hyperpolarization-activated pacemaker channels drive neuropathic pain. J Neurosci 23:1169–1178. Cho HJ, Staikopoulos V, Furness JB, Jennings EA (2009) Inflammation-induced increase in hyperpolarization-activated, cyclic nucleotide-gated channel protein in trigeminal ganglion neurons and the effect of buprenorphine. Neuroscience 162:453–461. Dalle C, Eisenach JC (2005) Peripheral block of the hyperpolarization-activated cation current (Ih) reduces mechanical allodynia in animal models of postoperative and neuropathic pain. Reg Anesth Pain Med 30:243–248. Debanne D, Campanac E, Bialowas A, Carlier E, Alcaraz G (2011) Axon physiology. Physiol Rev 91:555–602. Emery EC, Young GT, Berrocoso EM, Chen L, McNaughton PA (2011) HCN2 ion channels play a central role in inflammatory and neuropathic pain. Science 333:1462–1466. Emery EC, Young GT, McNaughton PA (2012) HCN2 ion channels: an emerging role as the pacemakers of pain. Trends Pharmacol Sci 33:456–463. Fried K, Aldskogius H, Hildebrand C (1988) Proportion of unmyelinated axons in rat molar and incisor tooth pulps following neonatal capsaicin treatment and/or sympathectomy. Brain Res 463:118–123. Fried K, Sessle BJ, Devor M (2011) The paradox of pain from tooth pulp: low-threshold ‘‘algoneurons’’? Pain 152:2685–2689. Gao LL, McMullan S, Djouhri L, Acosta C, Harper AA, Lawson SN (2012) Expression and properties of hyperpolarization-activated current in rat dorsal root ganglion neurons with known sensory function. J Physiol 590:4691–4705. Harper AA, Lawson SN (1985) Conduction velocity is related to morphological cell type in rat dorsal root ganglion neurones. J Physiol 359:31–46. Hatch RJ, Jennings EA, Ivanusic JJ (2013) Peripheral hyperpolarization-activated cyclic nucleotide-gated channels contribute to inflammation-induced hypersensitivity of the rat temporomandibular joint. Eur J Pain 17:972–982. Hildebrand C, Fried K, Tuisku F, Johansson CS (1995) Teeth and tooth nerves. Prog Neurobiol 45:165–222. Hogan QH, Poroli M (2008) Hyperpolarization-activated current (I(h)) contributes to excitability of primary sensory neurons in rats. Brain Res 1207:102–110. Kim YS, Kim YJ, Paik SK, Cho YS, Kwon TG, Ahn DK, Kim SK, Yoshida A, Bae YC (2009) Expression of metabotropic glutamate receptor mGluR5 in human dental pulp. J Endod 35:690–694. Kouranova EV, Strassle BW, Ring RH, Bowlby MR, Vasilyev DV (2008) Hyperpolarization-activated cyclic nucleotide-gated channel mRNA and protein expression in large versus small diameter dorsal root ganglion neurons: correlation with hyperpolarization-activated current gating. Neuroscience 153:1008–1019. Luo L, Chang L, Brown SM, Ao H, Lee DH, Hiquera ES, Dubin AE, Chaplan SR (2007) Role of peripheral hyperpolarization-activated cyclic nucleotide-modulated channel pacemaker channels in
Y. S. Cho et al. / Neuroscience 291 (2015) 15–25 acute and chronic pain models in the rat. Neuroscience 144:1477–1485. Matsuura S, Shimizu K, Shinoda M, Ohara K, Ogiso B, Honda K, Kataqiri A, Sessle BJ, Urata K, Iwata K (2013) Mechanisms underlying ectopic persistent tooth-pulp pain following pulpal inflammation. PLoS One 8:e52840. Momin A, Cadiou H, Mason A, McNaughton PA (2008) Role of the hyperpolarization-activated current Ih in somatosensory neurons. J Physiol 586:5911–5929. Moosmang S, Stieber J, Zong X, Biel M, Hofmann F, Ludwig A (2001) Cellular expression and functional characterization of four hyperpolarization-activated pacemaker channels in cardiac and neuronal tissues. Eur J Biochem 268:1646–1652. Paik SK, Park KP, Lee SK, Ma SK, Cho YS, Kim YK, Rhyu IJ, Ahn DK, Yoshida A, Bae YC (2009) Light and electron microscopic analysis of the somata and parent axons innervating the rat upper molar and lower incisor pulp. Neuroscience 162: 1279–1286. Paik SK, Kim SK, Choi SJ, Yang ES, Ahn SH, Bae YC (2012) Vesicular glutamate transporters in axons that innervate the human dental pulp. J Endod 38:470–474. Saito M, Murai Y, Sato H, Bae YC, Akaike T, Takada M, Kang Y (2006) Two opposing roles of 4-AP-sensitive K+ current in initiation and invasion of spikes in rat mesencephalic trigeminal neurons. J Neurophysiol 96:1887–1901. Schnorr S, Eberhardt M, Kistner K, Rajab H, Kasser J, Hess A, Reeh P, Ludwig A, Herrmann S (2014) HCN2 channels account for mechanical (but not heat) hyperalgesia during long-standing inflammation. Pain 155:1079–1090. Shinoda M, Asano M, Omagari D, Honda K, Hitomi S, Katagiri A, Iwata K (2011) Nerve growth factor contribution via transient
25
receptor potential vanilloid 1 to ectopic orofacial pain. J Neurosci 31:7145–7155. Takeda M, Tanimoto T, Nasu M, Ikeda M, Kadoi J, Matsumoto S (2005) Activation of NK1 receptor of trigeminal root ganglion via substance P paracrine mechanism contributes to the mechanical allodynia in the temporomandibular joint inflammation in rats. Pain 116:375–385. Tsuboi Y, Takeda M, Tanimoto T, Ikeda M, Matsumoto S, Kitagawa J, Teramoto K, Simizu K, Yamazaki Y, Shima A, Ren K, Iwata K (2004) Alteration of the second branch of the trigeminal nerve activity following inferior alveolar nerve transection in rats. Pain 111:323–334. Tu H, Deng L, Sun Q, Yao L, Han JS, Wan Y (2004) Hyperpolarization-activated, cyclic nucleotide-gated cation channels: roles in the differential electrophysiological properties of rat primary afferent neurons. J Neurosci Res 76:713–722. Wells JE, Rowland KC, Proctor EK (2007) Hyperpolarizationactivated channels in trigeminal ganglia innervating healthy and pulp-exposed teeth. Int Endod J 40:715–721. Weng X, Smith T, Sathish J, Djouhri L (2012) Chronic inflammatory pain is associated with increased excitability and hyperpolarization-activated current (Ih) in C- but not Adeltanociceptors. Pain 153:900–914. Yao H, Donnelly DF, Ma C, LaMotte RH (2003) Upregulation of the hyperpolarization-activated cation current after chronic compression of the dorsal root ganglion. J Neurosci 23:2069–2074. Yasuda M, Shinoda M, Kiyomoto M, Honda K, Suzuki A, Tamagawa T, Kaji K, Kimoto S, Iwata K (2012) P2X3 receptor mediates ectopic mechanical allodynia with inflamed lower lip in mice. Neurosci Lett 528:67–72.
(Accepted 28 January 2015) (Available online 7 February 2015)