The ADP receptor P2Y₁ is necessary for normal thermal sensitivity in cutaneous polymodal nociceptors

Several previous studies have reported that P2Y₁ is preferentially expressed in the IB4-binding population of small-diameter sensory neurons [24, 28]. These neurons represent the majority of cutaneous C-fiber afferents. To confirm these previous histological results with functional data, we tested the ability of IB4-binding neurons to show functional responses to the P2Y₁ agonist ADP (100 μM), as well as to the P2X₃, P2X_2/3 agonist α,β-me ATP (Figure 1). 73.5% (86/117) of IB4-binding neurons showed a transient increase in intracellular Ca²⁺ in response to ADP. The large majority of ADP-responsive neurons (77%, 47/61) also responded to α,β-me ATP, indicating expression of P2X₃. These results indicated widespread expression of functional P2Y₁ receptors in IB4-binding, P2X₃-expressing DRG neurons in acute cell culture.

Neurons recorded in the ex vivo preparation were sorted into subgroups depending upon their conduction velocities (CV) and responses to mechanical and thermal stimuli. Neurons with a conduction velocity of <1.2 m/s were classified as C-fibers, and all others were classified as A-fibers [29, 30].

Although we recorded from both A- and C-fibers, we focused our recording efforts primarily on C-fibers that were characterized as C-polymodal (CPM) in function, responding to mechanical and heat stimuli (CMH), and sometimes to cold stimuli as well (CMHC). We only examined enough A-fibers and other classes of C-fiber to verify that there were no significant differences in their biophysical characteristics or response properties to cutaneous stimuli (data not shown). Furthermore, while we occasionally observed mechanically and thermally unresponsive cells that were driven by the peripheral stimulating electrode, we only analyzed cells that had an identifiable cutaneous receptive field.

A total of 135 CPM fibers innervating hindlimb hairy skin via the saphenous nerve were intracellularly recorded and physiologically characterized from 18 WT (n = 69 cells) and 12 P2Y₁-/- (n = 66 cells) adult male mice. Apart from the cold response in CMHC cells, there was no statistical difference between the CMH and CMHC groups for either strain, and therefore these populations were pooled together as CPM cells. It should be noted that not all of these cells were stained with Neurobiotin, either due to logistical reasons or loss of cell.

Biophysical data including CV (calculated as the distance between the stimulating and recording electrodes, divided by the spike latency between the stimulus pulse and the triggered action potential (AP)), AP amplitude and duration at half-amplitude were collected for each recorded cell (data not shown). No significant differences were observed in AP amplitude or half-amplitude width for either WT (2.23 ± 0.07 ms) or P2Y₁-/- mice (2.13 ± 0.10 ms). However, the CV of CPMs in P2Y₁-/- mice (0.59 ± 0.01 m/s) was slightly faster than the CV in WT mice (0.53 ± 0.01 ms; p < 0.05). This slight difference in conduction velocity would not be expected to be functionally significant.

CPM fibers were tested for their response to mechanical stimuli (Figure 2A). No significant differences were observed in the mechanical response properties between WT and P2Y₁-/- mice. These properties included average mechanical threshold (WT: 15.76 ± 2.62 mN vs. P2Y₁-/-: 13.61 ± 2.48 mN; Figure 2B), peak instantaneous frequency (WT: 39.81 ± 1.78 Hz vs. P2Y₁-/-: 39.60 ± 2.00 Hz; Figure 2D), and peak mean firing rate (WT: 5.52 ± 0.44 spikes/sec vs. P2Y₁-/-: 5.04 ± 0.28 spikes/sec; data not shown). Similarly, the overall mean firing rates in response to 5 mN, 10 mN, 25 mN, 50 mN and 100 mN stimuli remained unchanged between mouse strains (Figure 2C).

CPM cells were tested for their response to cooling and warming stimuli. In CPM fibers that responded to cooling stimuli (Figure 3A), the average cold threshold in WT strains was significantly higher than those found in P2Y₁-/- strains (Figure 3B). From a baseline bath temperature that was maintained at 31.0°C, a rapid decrease in temperature to approximately 4°C resulted in APs that were evoked at relatively higher temperatures in WT (16.14 ± 1.03°C) than in P2Y₁-/- (10.62 ± 0.81°C; p < 0.05; Figure 3B). Maximum instantaneous frequency during cooling, however, was not significantly different between the strains (WT: 1.39 ± 0.29 Hz; P2Y₁-/-; 1.44 ± 0.29 Hz; and P2Y₁-/-: 1.44 ± 0.29 Hz; p < 0.05; Figure 3C).

During heating ramps from 31.0°C to 52.0°C (Figure 4A), CPMs in WT mice (42.31 ± 0.62°C) exhibited lower heat thresholds than those in P2Y₁-/- mice (45.97 ± 0.59°C; p < 0.05; Figure 4B). Similarly, maximum instantaneous frequency during heating was dramatically lower in P2Y₁-/- mice (2.46 ± 0.35 Hz) versus WT mice (17.77 ± 2.62 Hz; p < 0.05; Figure 4C). The maximal firing rate per degree was also notably higher in WT (6.17 ± 0.53 spikes/deg) than in P2Y₁-/- mice (1.31 ± 0.14 spikes/deg; p < 0.05). In addition, mean firing rates per degree (44-52°C, p < 0.05) were significantly higher in WT than in P2Y₁-/- mice (Figure 4D). The heat thresholds of mechanically insensitive CH fibers were unchanged in the P2Y₁-/- mice (41.8 vs 41.7°C) however, there were only 3 CH fibers characterized in the knockout mice, therefore data was not considered to be statistically valid.

A total of 39 CPM cells (WT: n = 25 cells; and P2Y₁-/-: n = 14 cells) were labeled with Neurobiotin and subsequently processed for immunohistochemistry. Our findings indicate that the CPMs of P2Y₁-/- mice express similar patterns of immunohistochemical markers as those contained in WT mice (Figure 5). We found that most Neurobiotin-filled WT CPMs bound IB4 (18/23; cells positive/total cells examined), but did not express either TRPV1 (0/19) or CGRP (0/6). In P2Y₁-/- mice, Neurobiotin-filled CPMs bound IB4 (11/14), but did not express TRPV1 (0/2) or CGRP (0/12; Figure 5).

A recent study has suggested that P2Y1 and the Gi-coupled P2Y receptors P2Y12-14 interact functionally to modulate nociceptor excitability [21]. To address the possibility that secondary changes in the expression of other receptors may be contribute to the phenotype of P2Y₁-/- mice, we examined mRNA levels for several genes that could contribute to transduction in cutaneous afferents, including TRPV1, P2X₃ and the Gi-coupled P2Y receptors, which are likely to be highly-coexpressed with P2Y₁ and may antagonize excitatory signaling in nociceptors [21]. As expected, expression of TRPV1 and P2X₃ were unaltered; of the P2Y receptors, P2Y₁₃ mRNA levels were significantly increased (Figure 6A, 2.7-fold increase). However, quantitative analysis of protein levels by Western blot indicated no significant difference in the amount of P2Y13 protein between WT and P2Y1-/- DRG (Figure 6B) Therefore, translation of P2Y13 was not altered in the mutant mice.

In initial studies, Tominaga et al. [40] suggested that P2Y₁ might modulate TRPV1 function. However, later studies revealed that in sensory neurons P2Y₂ and not P2Y₁ is co-expressed with TRPV1 and required for the modulation of TRPV1 by ATP [22, 24]. We have previously demonstrated the presence of TRPV1 immunoreactivity in mouse to be exclusively in mechanically-insensitive cutaneous CH afferents [32]. Here as previously, all CPM fibers examined lacked TRPV1 immunoreactivity. In addition, although we recorded from only a few CH fibers (3) in P2Y₁ deficient mice, there were no apparent effects on their heat sensitivity. Finally, there was no difference in TRPV1 mRNA levels between WT and P2Y₁-/- DRG.

Our results indicate that P2Y₁ also contributes to the response of CPM fibers to cold stimuli. Two TRP channels have been implicated in the transduction of cold stimuli: TRPA1 and TRPM8. Although we did not perform immunostaining for TRPA1 or TRPM8, results from previous studies suggest that these channels are not localized in the cutaneous CPM population. For example, TRPM8 is not expressed in IB4-positive neurons [41, 42]. Furthermore, G_q-coupled receptor signaling has been reported to reduce TRPM8 currents [43], whereas in our studies both heat and cold sensitivity were reduced in the absence of P2Y₁. TRPA1 has also been reported to co-localize almost exclusively with TRPV1 and not with IB4 binding [44, 45]. Thus, the impact of P2Y₁ deletion on the thermal transduction process in cutaneous CPM fibers does not appear to involve any of the known thermosensitive channels responsive to the temperature range examined here.

While the results presented here demonstrate a clear deficit in thermal transduction properties of CPM fibers in mice lacking P2Y₁, Malin and Molliver [21] reported that these mice had behavioral heat withdrawal latencies that were not different than those of wildtype mice. However, they did find that there was a significant difference in the level of inflammation-induced heat hyperalgesia. While not as striking, this finding is similar to that seen in mice lacking TRPV1, which had relatively normal acute heat responses but did not develop heat hyperalgesia following inflammation [36, 37].

It is interesting to note that these two fiber types, CPM and CH fibers, constitute virtually all cutaneous C-fibers that respond to heat. Based on the behavioral studies of these knockout mice it appears that while heat-evoked behaviors can be mediated by both TRPV1-containing CH fibers [46] and TRPV1-deficient CPM fibers [36, 37], the relatively small population of CH fibers are more efficient at evoking behavioral responses in mice. This seems to be most clearly evident in the response to heat following inflammation. However, this interpretation may be biased by the fact that while CPM fibers in P2Y1-/- mice show a reduced ability to transduce heat stimuli, TRPV1-/- mice are entirely lacking in CH fibers [32, 47]. Thus it is most likely that both afferent populations are necessary for acute heat hyperalgesia following injury.

Mice with a null mutation in the P2Y₁ gene have been previously described [49], and were generously provided by Beverly Koller, University of North Carolina, Chapel Hill. These mice thrive and breed normally. The mice were maintained on the C57BL6 background and were genotyped by PCR. Mice were housed in group cages, maintained on a 12:12 hour light-dark cycle in a temperature controlled environment (20.5°C) and given food and water ad libitum.

Real-time PCR analysis was carried out as previously described [50]. Mice were perfused with ice cold sterile saline. L2-5 dorsal root ganglia were dissected bilaterally and collected in RNAlater solution (Invitrogen). mRNA was isolated using the Qiagen RNeasy Mini kit according to the manufacturer's instructions and quantified by spectrophotometer. Extracted RNA was treated with DNase (Invitrogen) to remove genomic DNA (1 μl DNase, 2 μl 10× DNase buffer, 0.25 μl RNasin/5 μg RNA in H2O, 20 μl total/reaction) and reverse-transcribed using Invitrogen Superscript II reverse transcriptase according to the manufacturer's instructions. Negative control reactions were run without RNA to test for contamination. SYBR Green PCR amplification was performed using an Eppendorf Mastercycler Realplex real-time thermal cycler. All samples were run in triplicate; negative control reactions were run without template and with the reverse-transcriptase negative control reaction products in every amplification run. After amplification, a dissociation curve was plotted against melting temperature to verify selective amplification of a single product. Threshold cycle (Ct) values, the cycle number in which SYBR Green fluorescence rises above background, were recorded as a measure of initial template concentration. Relative changes in RNA levels were calculated by the ΔΔCt method using p53-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a reference standard: mean Ct values from each triplicate sample (n = 5 mice/data point, run individually) were subtracted from the mean reference standard Ct, yielding ΔCt. The difference between the ΔCt of the mutant and wildtype groups was then calculated for each gene of interest (ΔΔCt). The relative fold change was determined as 2^-DDCt. Statistical significance was determined by Student's t-test. Data are presented as the fold change in mRNA levels compared to baseline.

Dissociation of primary sensory neurons has been described in detail [51]. DRGs from all segmental levels were dissected from adult male WT mice, digested enzymatically and mechanically dissociated by trituration. Ca²⁺ imaging was performed within 18-24 hours as described previously [22]. Cells were incubated in 2 mM fura-2-AM in HBSS with 5 mg/ml bovine serum albumin and 10 μg/ml IB4 conjugated to Alexa-488 for 30 minutes at 37°C, then mounted on a microscope stage with constantly flowing HBSS at 5 ml/minute. Perfusion rate was controlled with a gravity flow system and perfusate temperature was maintained at 30°C using a heated stage and in-line heating system (Warner Instruments). Drugs were delivered with a rapid-switching local perfusion system. Firmly-attached, refractile cells were identified as regions of interest in the software (Simple PCI, C-Imaging). Absorbance data at 340 and 380 nm were collected once per second and the relative fluorescence (ratio 340/380) was plotted against time. An initial stimulus of buffer with 50 mM K⁺ was used to confirm the viability and neuronal identity of the cells. Neurons were then tested for responsiveness to the P2X agonist α,β-methylene ATP (a,b-me ATP) and ADP. Ca²⁺ transients were examined in response to application of agonists as noted in the figure legend.

The ex vivo somatosensory system preparation has been previously described in detail [52, 53]. Briefly, adult C57BL6 mice (Jackson Laboratory, Bar Harbor, ME) and P2Y₁ transgenic mice were anesthetized via an intramuscular injection of ketamine and xylazine (90 and 10 mg/kg, respectively) and perfused transcardially with chilled (6°C), oxygenated (95% O₂-5% CO₂) artificial CSF (aCSF; in mmol/l: 1.9 KCl, 1.2 KH₂PO₄, 1.3 MgSO₄, 2.4 CaCl₂, 26.0 NaHCO₃, and 10.0 D-glucose), with 253.9 mmol/l sucrose. Spinal cord, L1-L4 DRGs, saphenous nerve, and innervated skin were dissected free in continuity. Following dissection, the preparation was transferred to a separate recording chamber containing chilled oxygenated aCSF in which the sucrose was replaced with 127.0 mmol/l NaCl. The skin was pinned out on a stainless steel grid located at the bath/air interface, so that the dermal surface remained perfused with the aCSF while the epidermis was exposed to the air. The platform provided stability during the application of thermal and mechanical stimuli. The bath was then slowly warmed to 31°C prior to recording.

Individual DRG cells were impaled with quartz filament microelectrodes (impedance >100 MΩ) containing 5% Neurobiotin (Vector Laboratories, Burlingame, CA) in 1 mol/l potassium acetate. Electrical search stimuli were delivered through a suction electrode on the saphenous nerve to locate sensory neuron somata with a peripheral axon innervating the skin. Peripheral receptive fields (RF) were localized with a fine paint brush, blunt glass probe and von Frey hairs. When cells were driven by the nerve, but had no mechanical RF, a thermal search was performed by applying hot (~52°C) or cold (~0°C) physiological saline to the skin using a 10 ml syringe with a 20-gauge needle. If a thermal RF was located, the absence of mechanical sensitivity was confirmed by searching the identified RF using a glass probe. The response characteristics of the sensory neuron were determined by applying computer controlled mechanical and thermal stimuli. The mechanical stimulator consisted of a constant force controller (Aurora Scientific Aurora, Ontario, Canada) attached to a 1 mm diameter plastic disc. Computer controlled 5 s square waves of 5, 10, 25, 50, and 100 mN were applied to the cell's RF. Mechanical threshold was the lowest stimulus intensity of this ascending series to elicit at least one action potential (AP) within the first second of stimulus application. After mechanical stimulation, thermal stimuli were applied using a 3 mm² contact area Peltier element (Yale University Machine Shop). The cooling stimulus was rapidly applied by the Peltier element through the thermal conduction of circulating ice-chilled water that resulted in a drop in temperature from 31°C to approximately 4-6°C. The temperature was then brought back up to 31°C, and after a 5 s pause the heating stimulus was applied, consisting of a 12 s heat ramp from 31°C to 52°C followed by a 5 s plateau at 52°C. The stimulus then ramped back down to 31°C in 12 seconds. The cooling and heating thermal thresholds were determined to be the temperatures at which the first AP was evoked. All responses were recorded digitally for off-line analysis (Spike2 software; Cambridge Electronic Design, Cambridge, UK). After physiological characterization, the cell was labeled by iontophoretically injecting Neurobiotin (1-3 cells per DRG). Peripheral conduction velocity was calculated from spike latency and the distance between the stimulating and recording electrodes.

Once a sensory neuron was characterized and filled with Neurobiotin, the DRG containing the injected cell was removed and immersion fixed (4% paraformaldehyde in 0.1 M phosphate buffer for 30 minutes at 4°C). Ganglia were then embedded and blocked in 10% gelatin, postfixed overnight, and cryoprotected in 20% sucrose. Frozen sections (60 μm) were collected in phosphate buffer and reacted with antiserum for either TRPV1 (rabbit anti-TRPV1; Calbiochem, San Diego, CA) or CGRP (rabbit anti-CGRP; Chemicon, Temecula, CA). The binding of isolectin B₄ from Griffonia simplicifolia was examined using IB4-647 (Molecular Probes, Eugene, OR). After incubation in primary antiserum, tissue was washed and incubated in donkey anti-rabbit secondary antiserum conjugated to Cy3 (Jackson Immunoresearch, West Grove, PA), and reacted with FITC-conjugated avidin to label Neurobiotin-filled cells (Vector Laboratories). Distribution of fluorescent staining was determined using an Olympus confocal microscope and software (Fluoview; Olympus, La Jolla, CA). Sequential scanning was done to prevent bleed-through of the different fluorophores.

Lumbar DRG (L2 - L5) were homogenized in lysis buffer containing 25 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% NP-40 and 5% glycerol. Lysates containing equal amounts of protein per 30 μl for each sample were heated at 60°C for 5 min in 9% SDS, 60% glycerol, 375 mM Tris-HCl pH and bromophenol blue 0.015%, centrifuged and resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by transfer onto nitrocellulose membranes. Membranes were blocked in 5% bovine serum albumin in PBS with 0.05% Tween 20 (PBS-T) and incubated overnight at 4°C with both rabbit anti-P2Y₁₃ antibody (1:500) and mouse anti-actin (1:500). The actin monoclonal antibody, developed by Dr. Jim Jung-Chin Lin, was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by The University of Iowa, Department of Biology, Iowa City, IA 52242. Primary antibodies were followed by detection with SpectraPlex™ Fluorescent Western Blot secondary antibodies: APC-goat anti-rabbit IgG (for P2Y13) and donkey anti-mouse conjugated to CY3 at 1: 2500 dilution for both. Blots were imaged using a FluorChem Q workstation (Cell Biosciences, Santa Clara, CA). P2Y₁₃ band density normalized to actin (as a loading control) was quantified using the manufacturer's software (n = 4/genotype).

Data are expressed as means ± SE. Unpaired two-tailed Student's t-tests were used to analyze different aspects of the ex vivo responses of neurons to electrical, mechanical and heat stimuli. Heat data was normalized by multiplying the average AP spikes per degree by the percentage of cells responding at that temperature. In the analysis of mean firing rate/°C in response to the application of the heat ramp a 2-way ANOVA analysis was completed and followed with Bonferroni post hoc analysis. Statistical analysis for the real-time PCR results was performed on the raw Ct data and presented as fold changes in mRNA levels for clarity.