TNF-α acutely enhanced ASIC-mediated currents in rat DRG neurons
In the present study, AMG9810 (5 μM) was added to external solution to block proton-induced TRPV1 activation. As shown in Fig.1A, a sudden drop in extracellular pH from 7.4 to 6.0 produced a rapid inward current (IpH6.0) in DRG neurons. The IpH6.0 could be completely blocked by 10 μM of amiloride, a broad-spectrum ASIC channel blocker, and also by 2 μM APETx2, an ASIC3 blocker. In contrast, capsaicin (100 nM) failed to evoke any membrane currents in the presence of AMG9810. Thus, these acid-induced currents were considered to be ASIC currents or ASIC3-mediated currents after TRPV1 activation was blocked by AMG9810.
In some DRG neurons sensitive to acid stimuli, we first evaluated the effects of brief application of TNF-α on the ASIC currents. TNF-α was pre-incubated to DRG neurons for 5min prior to application of pH 6.0 acidic solution. As shown in Fig.1B and C, a brief (5min) application of TNF-α acutely increased the peak amplitude of IpH6.0. The enhancement of IpH6.0 occurred 5min after the onset of TNF-α application. This enhancement of IpH6.0 was dependent upon the doses of TNF-α treatment. In a representative DRG neuron, the peak amplitude of IpH6.0 progressively increased as doses of pre-treated TNF-α increased from 0.3 ng/ml to 30 ng/ml (Fig.1B). Fig.1C showed the dose–response curve for TNF-α with an EC50 (half-maximal effective dose) value of 1.96 ± 0.15 ng/ml. The results indicated that TNF-α rapidly enhanced ASIC currents in rat DRG neurons in dose-dependent manner.
We then investigated the effects of TNF-α on concentration-response curve for protons. ASIC currents were measured by applying a range of different pH acidifications in the absence and presence of TNF-α. Fig.2A showed that peak amplitudes of IpH6.5, IpH5.5 and IpH4.5 increased after pre-application of 10 ng/ml TNF-α for 5 min. Fig.2B showed concentration-response curve for protons shifted upwards by TNF-α treatment. First, TNF-α caused an increase of 42.34 ± 7.89% in the maximal current response to pH 4.5. Second, the Hill coefficient or slope of two curves had not significant difference in the absence and presence of TNF-α (pH: n = 1.30 ± 0.19; TNF-α + pH: n = 1.32 ± 0.21; P > 0.1, post hoc Bonferroni’s test). Third, the pH0.5 (pH for half-maximal activation) values of two curves had also no statistical difference (pH: pH0.5 = 5.94 ± 0.12; TNF-α + pH: pH0.5 = 6.03 ± 0.16; P > 0.1, post hoc Bonferroni’s test). We therefore concluded that sensitization of ASICs by TNF-α was not due to a change in the apparent affinity of ASICs for protons.
TNF-α-induced enhancement of ASIC currents was mediated by p38 MAPK, but not cyclooxygenase
We further explored the pathway linking TNF-α to its effect on ASIC currents. It has been demonstrated that TNF-α can signal through activation of p38 MAPK in DRG neurons as well as many other cell types [10, 34, 35]. In addition, TNF-α can induce functional expression of cyclooxygenase (COX)-2 in cultured DRG neurons [36]. We therefore investigated the roles of p38 MAPK and COX in the enhancement of ASIC currents by TNF-α. As shown in Fig.3A and B, the amplitude of IpH6.0 increased 44.14 ± 4.26% by TNF-α (10 ng/ml) pre-treatment alone. SB202190, a fast-acting p38 MAPK inhibitor, was applied to DRG neurons for 3 min followed by mixture of SB202190 and TNF-α for additional 5 min. The pretreatment of SB202190 (10 μM) substantially prevented the TNF-α-mediated increase in ASIC currents, and the amplitude of IpH6.0 increased only 2.67 ± 3.71% (P < 0.01, compared with TNF-α pretreatment alone, one-way ANOVA followed by post hoc Bonferroni’s test, n = 6; Fig.3A and B). Indomethacin, a potent inhibitor for both COX-1 and COX-2, was also applied to DRG neurons similar to SB202190 treatment. In contrast, indomethacin failed to change TNF-α-mediated increase in ASIC currents (Fig.3A and B). In addition, SB202190 or indomethacin alone had no effect on IpH6.0 (data not shown). These results indicated that p38 MAPK, but not COX, is necessary for TNF-α-induced enhancement of ASIC currents.
TNF-α increased acid-evoked action potentials in rat DRG neurons
ASICs are non-selective cation channels, once activation by acidification, which leads to membrane potential depolarization and neuronal excitation. We further observed whether TNF-α had effects on acid-evoked action potentials of rat DRG neurons. Although proton-induced TRPV1 activation was blocked in the presence of 5 μM AMG9810, we observed that an acid stimulus of pH 6.0 induced not only a rapid inward current with voltage-clamp recording, but also bursts of action potentials (APs) under current-clamp condition in the same DRG neuron (Fig.4A and C). Consistent with that observed under voltage-clamp conditions, TNF-α pre-treatment also acutely increased the number of APs evoked by acidic stimuli of pH 6.0 in DRG neurons (Fig.4A and B). In six DRG neurons treated with TNF-α (10 ng/ml for 5 min), the number of APs evoked by acidic stimuli of pH 6.0 significantly increased (P < 0.01, paired t-test, n = 6, Fig.4B). In other six DRG neurons pre-treated with SB202190 (10 μM), TNF-α failed to increase the number of APs (P ˃ 0. 1, paired t-test; n = 6, Fig.4C and D). These results indicated that TNF-α also rapidly enhanced acid-evoked APs via a p38 MAPK-dependent pathway.
TNF-α exacerbated acid-induced nociceptive behaviors in rats
Our above electrophysiological studies showed that TNF-α acutely sensitized ASICs via activation of p38 MAPK in vitro. We further ascertained whether TNF-α had effects on ASIC-mediated nociceptive behaviors through interacting with ASICs in vivo. Our previous studies observe that intraplantar injection of acetic acid into rats elicits ASIC-mediated nociceptive behaviors even if AMG 9810 (10 μM) blocked the activation of TRPV1 [37, 38]. We found that pretreatment with TNF-α (1, 3 and 10 ng in 50μl) dose-dependently exacerbated the acid-induced nociceptive behaviors (p < 0.05 and 0.01, one-way ANOVA followed by post hoc Bonferroni’s test, n = 10; Fig.5). However, the aggravating effect of TNF-α on acid-induced nociceptive behaviors was prevented in rats, which were co-treated with the p38 MAPK inhibitor SB202190 (25μM in 50μl). The mean number of flinches in these rats significantly decreased, compared with that observed in rats pretreated with TNF-α (10 ng in 50μl) alone (p < 0.01, one-way ANOVA followed by post hoc Bonferroni’s test, n = 10; Fig.5). In addition, injection of TNF-α (10 ng in 50μl) into the contralateral paws did not change acid-induced nociceptive behaviors. These results indicated that TNF-α exacerbated acid-induced nociceptive behaviors in rats via activation of local p38 MAPK pathway.