The chemokine CXCL1/growth related oncogene increases sodium currents and neuronal excitability in small diameter sensory neurons

Young female Sprague-Dawley rats (body weight 100–150 g) were housed one or two per cage under a controlled diurnal cycle of 12 h light and 12 h dark with free access to water and food. The ambient environment was maintained at constant temperature (22 ± 0.5°C) and relative humidity (60–70%). All the surgical procedures and the experimental protocol were approved by the institutional animal care and use committee of the University of Cincinnati (Cincinnati, OH, USA).

Rats were anesthetized by intraperitoneal injection of pentobarbital sodium (50 mg/kg). The bilateral L4 and L5 DRGs were isolated and the sheath was carefully removed in ice-cold normal Ringer solution. The connective tissue was digested by exposure to Ca²⁺-free solution containing 1.0% collagenase II (Fisher Scientific) for 30 min at 37°C followed by washout in normal Ringer solution for another 10 min. DRGs were then dissociated by trituration with fire-polished Pasteur pipettes. DRG cells were plated onto poly-D-lysine coated glass coverslips in Medium199 (Sigma, St. Louis, MO, USA) containing 10% heat-inactivated FBS and 1000 U/ml each of penicillin and streptomycin. GRO/KC (rat recombinant, from Cell Sciences, Canton, MA, USA) and/or cycloheximide (Sigma) were added to the medium immediately after plating, as indicated, and the DRG cells were incubated at 37°C (5% CO₂ balance air) before recording. Recordings were made within 30 hours of plating, and cells were selected for recording based on diameter (< 25 μm) and the absence of processes. Isolectin B4 (IB4) conjugated to fluorescein isothiocyante (FITC) (1 μg/μl; Sigma) was added to the culture medium for 30–60 min prior to start of recording sessions to identify non-peptidergic IB4-postive and peptidergic IB4-negative neurons [22, 23]. This procedure has been previously shown not to affect the currents measured [24]. Neurons showing a robust fluorescence signal were classified as IB₄-positive and those displaying no signal at all, IB₄-negative. In general, the fluorescence was not examined until after completion of current recordings from each cell; i.e., IB4 binding was not used as a criterion in selecting which cells to record. In each individual experiment, we alternated between recording from control and experimental coverslips throughout the recording period. Every attempt was made to compare control and experimental cells taken from the same cultures, though in some cases it was not possible to obtain data from all four types of cells (experimental and control for both IB4-positive and IB4-negative) in a given experiment, particularly when IB4 status was not determined until the end of the recording. Additional controls for inter-animal variability are described in the Results section. Once it had been well established that the proportion of IB4-positive cells did not change with GRO/KC exposure, IB4 binding was sometimes examined prior to recording in order to ensure that approximately equal numbers of each cell type were recorded in each experiment.

After over-night culture (16–30 hours), coverslips were transferred to a recording chamber. Whole cell voltage-clamp recordings of small DRG neurons (diameter 15–25 μm) were conducted at room temperature with an AxoPatch-200B amplifier (Molecular Devices Corp, Union City, CA, USA). Patch pipettes (2.5–4.0 MΩ) were fabricated from borosilicate glass. Data were acquired on a Pentium IV computer with the Clampex 8 program (Molecular Devices Corp). The cell capacitance artifact was cancelled by the nulling circuit of the recording amplifier. Voltage errors were minimized by using ≥ 80% series resistance compensation. The current was filtered at 5 kHz and sampled at 50 kHz. The recording chamber was continuously perfused at room temperature with oxygenated bath solution at a flow rate of 1–2 ml/min.

For the current-clamp study of excitability, we measured the threshold current (rheobase), action potential (AP) threshold, resting membrane potential, AP rising rate, AP falling rate, afterhyperpolarization, and input resistance. Any cells without AP were excluded from the study. Only neurons that had stable membrane potential more negative than -45 mV after obtaining the whole cell configuration were used for further study.

For the voltage-clamp study, total Na⁺ current was divided into tetrodotoxin-resistant (TTX-R) and TTX-sensitive (TTX-S) currents, using the method previously described [7]. Briefly, in most experiments, the TTX-R currents were isolated from the TTX-S currents by using a holding potential of -50 mV. The peak TTX-R current was measured at each potential. Both TTX-R and TTX-S current were evoked by a series of depolarizations from a holding potential of -80 mV. The amplitude of the TTX-S current in each cell was measured during depolarization to -10 mV from -80 mV, based on the distinct kinetic properties of the two currents. At this voltage, TTX-S current was measured as the difference between the early peak current and the current 5 msec after the beginning of the depolarization, a time window in which the TTX-S current essentially completely inactivates but TTX-R current changes little. In experiments in which TTX-S current evoked from -80 mV was measured at additional voltages, a time window of 11 msec (at -20 mV) or 27 msec (at -30 mV) was used, as determined in our previous studies using TTX to verify the separation of the currents.

The normal bath solution for recording AP contained (in mM) 130 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, 10 HEPES, and 10 glucose. The pH was adjusted to 7.4 with NaOH, and the osmolarity adjusted to ~300–310 mOsm with sucrose. The pipette solution contained (in mM) 140 KCl, 1 CaCl₂, 2 MgCl₂, 11 EGTA, 10 HEPES, 10 NaCl and 2 Mg ATP. The pH was adjusted to 7.3 with KOH, and osmolarity to ~290–300 mOsm with sucrose.

Sodium currents were measured under a reduced sodium gradient to reduce space clamp problems. The Na⁺ bath solution contained (in mM) 30 NaOH, 110 TMA-Cl, 5 TEA-Cl, 2 MgCl₂, 1 CaCl₂, 0.1 CdCl₂, 10 HEPES, and 10 glucose. The pH was adjusted to 7.4 with HCl, and the osmolarity was adjusted to ~300–310 mOsm with sucrose. The pipette solution contained (in mM) 140 CsCl, 5 NaOH, 2 MgCl₂, 1 CaCl₂, 11 EGTA, 10 HEPES, 2 Mg ATP, and 1 Li GTP. The pH was adjusted to 7.2 with CsOH, and the osmolarity to ~290 – 300 mOsm with sucrose. Voltages were not corrected for liquid junction potentials, which were calculated to be < 10 mV in all cases.

Rats were anesthetized and fixed by perfusing 200–300 ml of Zamboni's fixative (4% paraformaldehyde in 0.1 M phosphate buffer, pH = 7.4) through the left ventricle of the heart. The bilateral L4/L5 DRGs were removed and post-fixed in the perfusion fixative for 2 hours at room temperature. The ganglia were horizontally sectioned with a cryostat at a thickness of 8 μm. Tissue sections were incubated in polyclonal rabbit anti-rat antibody to CXCR2 (Santa Cruz Biotechnology, Santa Cruz, CA) at a dilution of 1:200 overnight at 4°C, followed by reaction with secondary antibody Alexa Fluor 594 goat anti-rabbit IgG (H+L) (1:1000, Invitrogen, Carlsbad, CA) for 1 hour.

DRG cells from L4/L5 spinal levels were removed from rats, and neurons were isolated and cultured overnight. The cultured cells were fixed with 4% paraformaldeyhde for 30 min at room temperature, washed in ice cold acetone for 1 min, and permeablized with 1% Triton for 10 min. The cells were then incubated with antibody to CXCR2 (1:100, Santa Cruz Biotechnology) and neuronal marker anti-NeuN (1:100, Milipore, Billerica, MA) for 2 hours followed by incubation in secondary antibody conjugated to Alexa Fluor 594(1:1000, Milipore, Billerica, MA) for 1 hour. After drying, the cells were mounted on coverslips with Vector Hard Set mounting medium (Vector Laboratories Inc., Burlingame, CA, USA). In some experiments, IB4 conjugated to FITC (1 μg/μl), instead of anti-NeuN antibody, was added to the culture medium for 30 min before switching to the primary antibody, to identify IB4-positive and IB4 negative neurons. The immunoreactive products were visualized using a confocal microscope and Slidebook 4.1 imaging acquisition software (Intelligent Imaging Innovations, Denver, CO, USA).

Western blot analysis was employed to confirm that the CXCR2 antibody used for immunohistochemistry recognized protein in the DRG of the size expected for CXCR2. Bilateral L4/L5 DRGs were isolated and homogenized in RIPA lysis buffer (0.5 M Tris-HCl, pH 7.4, 1.5 M NaCl, 2.5% deoxycholic acid, 10% NP-40, and 10 mM EDTA) (Millipore, Billerica, MA) supplemented just before running the assay with a protease and phosphatase inhibitor cocktail (100 μl of aprotinin, leupeptin, and pepstatin; 500 μl of PMSF, Na₃VO4, and NaF) (Pierce, Rockford, IL, USA) followed by centrifugation at 4 °C for 40 min. Samples (15 μg of total protein per lane) were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) running in Tris-HEPES-SDS buffer (Pierce, Rockford, IL) followed by electrophoretic transfer to nitrocellulose membrane (Bio-Rad, Hercules, CA, USA) in tris-glycine-SDS transfer buffer. Nonspecific binding sites were blocked with 3% nonfat dry milk (Bio-Rad) in Tween (0.1%)-phosphate-buffered saline (PBST) at room temperature for 1 hour. The CXCR2 antibody (1:300, Santa Cruz Biotechnology) was applied to the blot followed by incubation with immunopure peroxidase conjugated goat anti-rabbit IgG (H+L) (1:20,000, Pierce, Rockford, IL). The blot was developed by exposure to blue-light-sensitive Hyperfilm ECL (GE Healthcare Bio-Sciences Corp) using the enhanced Chemiluninescent (ECL) Western Blotting Detection System (GE Healthcare Bio-Sciences Corp., Piscataway, NJ).

Total RNA was isolated from DRG cultures under conditions similar to those used in electrophysiological recordings, using a Stratagene Absolutely RNA Miniprep Kit (Stratagene, La Jolla, CA, USA). Immediately after isolation, RNA was transcribed into cDNA using an iScript cDNA kit (Bio-Rad). Cells treated overnight with 1.5 nM GRO/KC were compared with control cells from a sister culture. Relative channel expression was determined by quantitative PCR using a Stratagene MX-Pro 3005P. cDNA samples were amplified in triplicate (10 min at 95°C, followed by 40 cycles of 30 sec 95°C, 60 sec 60°C, then 60 sec 72°C with fluorescence measured at the end) in a 25 μl reaction volume (400 nM each forward and reverse primers, SYBR green reaction mix (Roche Applied Science, Indianapolis, IN)). Fluorescence was normalized with the reference dye ROX, and gene expression was normalized to the housekeeping gene hypoxanthine ribosyltransferase (HPRT; [25]), run on the same plate. Primers (Table 1) were selected using the Primer3 program [26] and were designed to include exon-intron boundaries to minimize contamination by genomic DNA. Care was also taken to avoid regions of homology with other Na channels, and primers were examined with an in silico virtual PCR program to ensure that no other products were predicted from rat cDNA or genomic DNA http://​genome.​ucsc.​edu. PCR products were verified by melting point analysis at the end of each experiment and, during protocol development, by gel electrophoresis. Template from control cells was run side-by-side with template from GRO/KC-treated cells from the same culture. The efficiencies of the amplification reactions were estimated directly from the amplification plots using the DART program and a correction for efficiencies was included in the relative expression calculations [27]. Efficiencies were generally in the range of 95 to 103% and were similar for control and GRO/KC treated samples, and across experiments. Hence a single efficiency value was used for control and GRO/KC samples for each gene in any given experiment. Efficiencies were set to 100% for genes with low expression that amplified only during the last few cycles and did not have enough data points for the DART efficiency measurement; low-expressing genes were not studied in quantitative detail. With the exception of the efficiency estimate, amplification data were analyzed using MxPRO software (Stratagene) to determine values for amplification threshold and relative expression.

Data were analyzed using Clampfit 9 (Molecular Devices Corp), Graphpad Prism (GraphPad Software, Inc., San Diego, CA, USA), and Origin 7 (Origin Lab Corp., Northampton, MA, USA). Currents were normalized by cell capacitance. The ohmic portion of the observed peak Na⁺ currents was measured from current steps to -100 through -60 mV and subtracted. Data are expressed as mean ± standard error of the mean (SEM). Differences in proportion of cells binding IB4 were examined with Fisher's exact test. Statistical significance of differences between average values in experimental and control neurons were analyzed by Student's t-test or, for data that were not normally distributed, the Mann-Whitney rank sum test, using SigmaStat Software (Systat Software, Inc., San Jose, CA, USA). In the case of multiple comparisons over a voltage range for activation data, the data were analyzed by two-way repeated measures ANOVA (RM ANOVA), with pairwise multiple comparison (Holm-Sidak method) to determine at which voltages the differences between experimental and control cells were significant if an overall effect of GRO/KC treatment was observed. ANOVA (or Kruskal-Wallis test on ranks, for data not showing normal distribution) was used to analyze experiments with more than two experimental groups. Activation data was fitted to the Boltzmann equation (Figure 1) was done using Graphpad Prism software. Significance of fold-changes in relative Na channel expression in qPCR experiments was done using a ratio t-test. Significance was ascribed for p < 0.05.

Small diameter DRG neurons were acutely cultured and incubated overnight with 1.5 nM rat (recombinant) GRO/KC or vehicle. The Na⁺ currents were measured in small diameter neurons with whole cell patch clamp techniques. After incubation in GRO/KC, voltage-activated Na⁺ currents had overall properties (except for amplitude) similar to those we have previously described in cells cultured from control and locally inflamed DRG [7]. The Na⁺ currents in both control and GRO/KC treated cells activated at -30 to -20 mV (Figure 1). Persistent Na⁺ currents attributed in previous studies to the Nav 1.9 isoform, which activate at potentials near -60 mV under recording conditions similar to ours [28, 29], were observed in neither control nor GRO/KC treated cells. We observed small sustained inward currents -40 mV in less than 10% of cells, and never observed inward currents at potentials of -50 or -60 mV (Figure 1; compare to [28, 29]). Na⁺ currents from GRO/KC treated cells (like those previously observed in control cells and inflamed cells) had a faster component that was blocked by 50 nM tetrodotoxin (TTX; n = 6 GRO/KC treated cells; data not shown) or by using a holding potential of -50 mV (Figure 1 and [7]). TTX-resistant currents, which have slower kinetics, could be isolated by using a holding potential of -50 mV which inactivated TTX-sensitive currents.

As shown in Figure 2, GRO/KC incubation (16 to 30 hours) significantly increased the density both TTX-R and TTX-S currents. These effects were seen in both IB4-positive and IB4-negative neurons. These results are from 53 control cells (26 were IB4-positive) and 54 GRO/KC treated cells (28 were IB4-positive). In order to ensure that this primary finding was not confounded by inter-animal variability, we also analyzed the subset of data in which currents measured in control cells could be compared with that measured in GRO/KC treated cells from the same batch (animal), in side-by-side experiments. Statistical analysis similar to that shown in Figure 1 was then conducted using the batch averages (n = 4) instead of the individual values for each cell. Despite the much smaller N, the increase in TTX-R current was still statistically significant in both IB4-positive (overall p = 0.019) and IB4-negative (p = 0.002) cells. The GRO/KC induced increases in TTX-S currents showed the same trends as shown in Figure 2, though these did not reach significance with the smaller N value when batch averages were analyzed (p = 0.1 for IB4-negative cells and p = 0.2 for IB4-positive cells). The proportion of IB4-positive cells was not significantly altered by GRO/KC treatment (p = 0.8, Fisher's exact test).

The higher (2 – 3 fold) TTX-S current density observed in IB4-negative compared to IB4-positive cells was significant in both control cells (p = 0.03, Mann-Whitney test) and GRO/KC treated cells (p = 0.02, Mann-Whitney test). There was a tendency for the peak TTX-R current (at 0 mV) to be higher in IB4-negative cells which however did not reach significance in either control cells (28% higher, p = 0.16, Mann-Whitney test) or in GRO/KC treated cells (32% higher, p = 0.19, t-test).

The voltage dependence of activation for the TTX-resistant current was not significantly affected by GRO/KC incubation (Figure 3A) and was not different between IB4-positive and IB4-negative cells. The activation curve for the combined data was well fit by a Boltzmann equation with V_1/2 = -7.3 mV and a slope factor of 5.7. The time course of decay for the TTX-R current during depolarizing pulses was also unaffected by GRO/KC (Figure 3B). Similarly, the time constant for decay of the TTX-S current at -10 mV was not affected by GRO/KC incubation (Figure 3C). The TTX-sensitive current became too fast to accurately voltage clamp at more positive potentials, so its activation curve was not determined in these experiments. However, the TTX-sensitive current observed during steps from -80 to -10 mV or below, where it is readily kinetically separated from the TTX-resistant current (see Methods), showed a qualitatively similar threshold for activation after GRO/KC incubation (see Figure 2C and 2D).

Lower doses of GRO/KC were also tested in similar experiments. Figure 4 shows the Na⁺ current density measured at two lower doses of GRO/KC, 0.28 and 0.06 nM. As shown, only the highest dose tested, 1.5 nM, gave a significant increase over control values, In Figure 4, data from control cells were obtained for the experiments at each separate dose. However, similar results were obtained if the data were analyzed by comparing current densities at each individual dose to the combined control values.

We were unable to observe increases in Na⁺ currents following acute (5 minute) application of 1.5 nM GRO/KC (n = 5 IB4-positive and 7 IB4-negative cells). The finding that GRO/KC enhanced the magnitude of Na⁺ currents without marked shifts in voltage dependence activation or kinetics (Figure 2, 3), and was observed only after a period of incubation, suggested that the primary effect was an increased number of Na⁺ channels, and hence that protein synthesis might be required. To test this idea, we conducted experiments using the protein synthesis inhibitor cycloheximide. The results are shown in Figure 5. In general, cycloheximide prevented the GRO/KC-induced increases in Na⁺ current density without significantly affecting baseline current densities. In both IB4-positive and IB4-negative cells, only the GRO/KC without CHX group differed significantly, from all three other groups.

Nine different genes for voltage-gated Na channels have been described, with Na_v1.1 – Na_v1.7 mediating TTX-sensitive current and Na_v 1.8 and 1.9 mediating TTX-resistant currents [30]. A simple explanation for the observed effects of GRO/KC on Na channel currents would be enhanced expression of channels with properties similar to those seen in control cells. As an initial test of this possibility, we used quantitative PCR methods to quantify relative changes in the amounts of Na channel mRNA present in DRG neurons after overnight incubation in 1.5 nM GRO/KC. Culture conditions were kept as close as possible to those used for the electrophysiological experiments. Expression data were normalized to that of the housekeeping gene HPRT. Of the nine Na channels, 3 showed much higher expression relative to HPRT than the others: Na_v 1.1, 1.7, and 1.8 (Figure 6A). Na_v 1.4 (a form found primarily in skeletal muscle), and 1.5 (a form found primarily in cardiac muscle), were undetectable or showed product only during the final few cycles. Na_v 1.9 (which is thought to mediate persistent Na currents in nociceptors) was also very low abundance, with relative expression on the order of 10^-6 times that of Nav 1.8. In contrast, in mRNA isolated from a whole DRG, we observed Na_v1.9 mRNA with an abundance similar to that of Na_v 1.8. This profile of expressed channels was grossly similar both before and after GRO/KC treatment; no previously low abundance channels were markedly upregulated by GRO/KC. The fold- increases in Na_v 1.1, 1.7, and 1.8 were between 2 and 3 fold, roughly similar to the observed increase in current densities (Figure 6B); this measurement is less sensitive to the efficiency correction. These fold-increases were significant except in the case of Na_v 1.8, which showed a nonsignificant trend to similarly increase (p = 0.08).

In order to determine the likely physiological effects of GRO/KC, excitability parameters were measured after a 1.5 nM GRO/KC incubation similar to that used to study Na⁺ currents. As shown in Table 2, the overall effect of GRO/KC treatment was a marked increase in excitability, including a reduced action potential threshold, depolarized resting potential, and decreased rheobase which were observed in both IB4-negative and IB4-positive cells.

Initially, the experiments shown in Table 2 were conducted by injecting short-duration (80 msec) current pulses, and it was evident that after GRO/KC incubation many cells fired action potentials for the entire duration of these short depolarizing pulses. In order to examine this effect in more detail, subsequent experiments were conducted using much longer (1 s) current pulses. As shown in Figure 7, such long pulses demonstrated a striking ability for repetitive firing that developed after GRO/KC incubation. In IB4-negative cells, multiple action potentials (up to 27) could be observed during the 1 second current injection, and these lasted throughout the entire pulse. IB4-positive cells also acquired the ability to fire throughout the 1 second current injection after GRO/KC incubation. In this case the number of action potentials per pulse did not increase as dramatically; rather, the action potentials tended to become very broad (Figure 7B and Table 2), though still continuing until the end of the 1 second current injection. The significant increase in action potential width (68%) seen in IB4-positive cells in Table 2 may be a conservative estimate, because those measurements are from the first evoked action potential and action potentials tended to broaden even more during the course of a long depolarizing pulse (Figure 7B).

Almost all dissociated DRG neurons including IB4-positive and IB4-negative neurons displayed CXCR2 staining (Figure 8A, B). In addition, CXCR2 staining was extremely intense in a subset of DRG cells. The intensely stained cells had an average diameter of 21.9 ± 0.9 μm (n = 3 rats) and were all IB4-negative. However, the number of high CXCR2 expressing cells was very small, approximately 2.4% of all DRG cells, making it likely that the other data reported in this study have a very small contribution from this subset of neurons with extremely high expression levels.

In DRG sections, CXCR2 was expressed in most neurons, whether small, medium, or large diameter (Figure 8C, D). An antibody directed to the C terminus of CXCR2 (supplier: Santa Cruz) displayed intense staining of plasma membrane and, in some cells, the nuclear membrane, with no immunoreactivity in the nucleoli. Although there was no apparent difference in CXCR2 staining of the plasma membrane between most small and large neurons, a subset of small DRG neurons displayed very high density staining, similar to observations made in the dissociated cells. Sections or dissociated cells incubated with primary antibody preabsorbed with 30 fold excess of antigen did not show any immunoreactivity (Figure 8E). In order to further validate the antibody used for the CXCR2 staining, we conducted western blot analysis of protein extracted from whole DRGs. Using the same antibody from Santa Cruz at 1:300 dilution, the western blot showed a strong doublet of 58–66 kDa and a lighter staining band of 35 kDa. An example is shown in Figure 8F; similar results were obtained in 3 replicate experiments.