Materials and methods

Culture of DRG neurons

For each primary culture, 10 SD rats borned within 24 hours were selected and decapitated. T6-12 DRGs were cut and quickly moved into cold HBSS solution. Blood and debris on the surface of DRG organization were rinsed out. After shredded, DRGs were digested in 0.125% trypsin for 25 minutes. Then the DRGs were rinsed 3 times in 3ml DMEM containing 10% fetal bovine serum. Afterwards, DRGs were dissociated by gentle triturating with a pipette and the cell suspension was dropped onto the cover slips coated with Poly-D-lysine. The DRG cells were cultured in DMEM containing 10% fetal bovine serum at 37°C in a humidified atmosphere containing 5% CO₂ for 8 hours. After the cells firmly adhered, the culture medium was changed into neuralbasal medium containing 2% B27, 100ng/ml NGF. The medium was replaced every 3 days. From the 3rd day of culturing, 10mol/L cytarabine was added to inhibit glia cells growth.

Electrophysiological recording

Whole-cell voltage-clamp recordings of DRG neurons were performed at room temperature (20–22°C) with a MultiClamp 700A amplifier (Axon Instruments, Foster City, California, USA). Neurons were prepared as above, and all recordings were performed when neurons were cultured for 7-10 days. All of the recordings were made from small-diameter (15–25 μm) DRG neurons. Microelectrodes were fabricated from 1.5 mm out diameter borosilicate capillary glass (Sutter Instruments, Novato, CA) by using a P-97 puller (Sutter Instruments, Novato, CA), and had a resistance of 3–5 MΩ. Electrodes were filled with pipette solution (150mM K gluconate, 4mM MgCl₂·6H2O, 1mM CaCl₂, 10mM Hepes, pH 7.2 adjusted with KOH). During experiments, cells were superfused with an external solution (150 mM NaCl, 5 mM KCl, 1mM MgCl₂, 2mM CaCl₂, 10mM Hepes, 10mM glucose, pH 7.2 adjusted with NaOH). Membrane potential was held at −70 mV. Analogue signals were filtered at 2 kHz, sampled at 10 kHz. Drug solutions were delivered by OCTAFLOW system (ALA scientific Instruments Inc, Westbury, New York, USA). One barrel was used to apply a drug-free solution to enable rapid termination of the drug application.

We performed voltage-clamp experiments in rat DRG neurons cultured without any protease treatment. The primary cultured DRG neurons were divided into several groups according to experimental needs. Each group has eight cover-slips of cells. If a cover-slip of cells were treated by drugs once in patch clamp experiment, the remaining cells on this cover slip are no longer used for experiments, in order to avoid abnormal reaction caused by duplication of drugs. The data were obtained from more than 10 batches of primary cultured cells.

Statistical analysis

Data analysis was performed using Clampfit 8.02 (Axon Instruments, USA) and Origin 7.0 (OriginLab Corporation, Northampton, MA, USA). All results are expressed as mean ± SEM. An LSD-t test was used to compare electrophysiological data between each of the two groups. A difference was accepted as significant if the probability was < 5% (p < 0.05).

Results

Thirty micromole per litre α, β-MeATP induced inward currents in DRG neurons which decayed rapidly. We recorded α, β-MeATP sensitive inward currents in 78.4% of the small and medium neurons (127 of 162). The amplitude of the response to a second application of α, β-MeATP 3 minutes after external solution application was 94 ± 9% of the initial amplitude, indicating that the receptor had recovered from any desensitization over this period (n=40, P>0.05 vs control) [Fig 1A, C]. These currents can be blocked by 0.5 μM A317491, a P2X3-specific antagonist ^[13] (n=16, P<0.001 vs α, β-MeATP) [Fig 1B, C]. This proves that the currents recorded were generated through the P2X3 receptors.

To evaluate the characteristics of P2X3 receptors opening up, we tried to use a exponential equation [f(t)=Σ_i=1~n(A_i-A_ie^-t/τi)+C] to fit the rising phase of α, β-MeATP current curve, where A_i is the fraction of the respective component and τi is the time constant. The current curve between the point where the curve began to go away from the baseline and the point where the curve reached the peak point was regarded as the rising phase [Fig 2A]. In most of the previous works, the opening rate used to be revealed by the single-exponential fitting equations^[16] or 10-90% rising time^[17], but these methods are not accurate enough. We found that in both ends of the rising phase the current changed relatively slowly, while in the middle phase the current changed rapidly. These cannot be well described by an equation of one time constant. Recently, Moffatt L et.al explored the properties of outside-out patches containing many of P2X2 channels and used the sum of three exponential functions to present the opening characteristics of single channel^[18]. We tried to fit the rising phase curve by the sum of three exponential functions, but the fit of the data to this method was just in part excellent. Therefore, we used an equation of four exponential functions which can fit all the data well. Four time constants were obtained after fitting. In both ends of the rising phase, the current changes relatively slowly, while in the middle part of the branch the current changes rapidly. This is consistent with the results we have fitted that the values of τ1 and τ4 are large which representing the relatively slow current component, while the values of τ2 and τ3 are similar and small which representing the rapid component. We found that, after 3 minutes of external solution perfusion, P2X3 currents opened more slowly. All four τ values increased [Fig 2A, Table 1].

In order to observe how trypsin affects the P2X3 currents in the DRG neurons, we applied cells with 100nM trypsin for 3 minutes before the second α, β-MeATP application. We found that the same doses of α, β-MeATP produced smaller, not larger, current responses (67%±11%, n=18, p< 0.05 vs ES) [Fig 2B, G]. In the meantime, four τ values of the rising phase of the P2X3 currents were significantly reduced by trypsin treatment[Table 1].

PAR2 is the endogenous receptor of trypsin. In order to determine whether the inhibition of P2X3 currents by trypsin is caused by PAR2 activation, we pretreated cells with 100μM PAR2-activating peptide SL-NH₂ for 3 minutes. The second α, β-MeATP application also produced smaller current responses (65%±7%, n=18, p< 0.05 vs ES) [Fig. 2C, G], and no attenuation was detected in cells pretreated with the reversed control peptide, 100μM LR-NH₂ (96%±9%; n=16, p>0.05 vs ES) [Fig2D, G]. On the other hand, SL-NH₂ pretreatment decreased the four τ values of the rising phase[Table 1].

As PAR2 combines with its ligands, intracellular PKA will be activated through G protein signaling pathway. Therefore, we further investigated whether the effects of SL-NH₂ on P2X3 currents could be blocked by PKA inhibitor H89. After 3 minutes pretreatment of 100 nM H89, the P2X3 peak currents reduced to 85% ± 5% of the control (n=8, p< 0.05 vs ES) [Fig 2E, G] and four τ values of the rising phase reduced a little too [Table 1]. As the cells were applied with H89 and SL-NH₂ together, the second α, β-MeATP application produced much smaller current responses (28%±2%, n=8, p< 0.05 vs ES) [Fig 2F, G]. Four τ values of the rising phasees decreased significantly compared with those of ES, but not with SL-NH₂ [Table 1].

Discussion

In our study we found that the short term (3 min) trypsin treatment did not increase P2X3 currents in DRG neurons significantly. on the contrary, there was a certain degree of inhibition. Moreover, the PAR2 receptor agonist, SL-NH₂ played a similar effect as trypsin. This shows that PAR2 receptors play a role in the inhibition of trypsin, and it also rules out the possibility that PAR2 activation increases the response of pain through the rapid increase of P2X3 receptor currents. Our experiments were performed at room temperature as usual. The temperature is lower than normal of 37°C body temperature. This may affect the signal coupling process downstream of certain G protein-coupled receptor. However, many articles have reported that PAR2 activation causes its downstream PLC ^[4], PKC ^[4,19] and PKA ^[19-20] activation, which were found by patch-clamp experiments performed at room temperature. Therefore, we believe that low temperature is unlikely to be the main reason for that acute PAR2 activation failed to increase the P2X3 currents.

Zhu WJ et.al reported that the nocifensive behavior of the rat increased 4 min after PAR2 agonist injection, which indicated that the nociceptive threshold was decreased and pain response induced by α,β-me-ATP was potentiated^[14]  However, we found that P2X3 current amplitude of the single neuron failed to increase after incubated with the PAR2 agonist SL-NH2 for 3 minutes. These results may suggest that the acute decrease in nociceptive threshold after PAR2 activation was not directly mediated by the P2X3 receptor. In addition, Dai Y showed that activation of PAR2 can sensitized TRPA1^[4] and TRPV1^[6] channel in rats with acute inflammatory pain, which could be the reason for the acute change in nociceptive threshold. Therefore, α,β-me-ATP may only act to trigger off the nociceptive responses. However, we cannot exclude the possibility that P2X3 contributed to the potentiation of nociceptive responses in rats during long-term activation of PAR2. As a G protein-coupled receptor, PAR2 activation may cause a lot of intracellular signaling molecules to be activated, and may influence some protein expression^[21]. The longer the trypsin exists, the more possible for it to enhance the excitability of DRG neurons by increasing the expression of P2X3. As a matter of fact, it has been reported that in the rat model of acute pancreatitis, P2X3 expression was up-regulated one day after the model set-up, and reached its peak in the first 7 days^[15]. And the increased P2X3 expression induced by trypsin may be the reasonable explanation for enhanced pain response. But the specific molecular mechanism remains to be further studied.

In our study system, 3 minutes after the whole-cell mode formation, the opening of P2X3 receptor slowed down. The reason might be that the necessary factors for maintaining or promoting P2X3 open were diluted as the cytoplasm and electrode solution exchanged. Nevertheless, we found that both trypsin and SL-NH₂ are able to reduce the τ values of P2X3 rising phase, that is, speed up the P2X3 opening. In the case that P2X3 expression has been increased in pancreatitis, the trypsin may further increase the excitability of DRG neurons by speeding up the opening of P2X3.

It is well accepted that there are PKC phosphorylation sites in P2X3 receptors, and when PKC activity increases, P2X3 current will enhanced because of the phosphorylation of these sites^[22]. But the currents observed in this study reduced. So we investigated whether another common protein kinase PKA was involved. PKA inhibitor H89 itself inhibited the basis currents of P2X3 a little. This might be due to that H89 reduced the basis phosphorylation of P2X3 itself or some other intracellular factors which play a key role in P2X3 opening, and thus inhibiting P2X3 current amplitude. On the other hand, H89 did not reverse the SL-NH₂ induced current amplitude decrease. This suggested that PKA is not involved in the process of SL-NH₂ action. In contrast, when H89 and SL-NH₂ were applied together, the amplitude of P2X3 currents was reduced greatly. It seems that there was a synergy effect. But the specific mechanism also needs further study. In the current dynamics, H89 itself reduced the τ values of P2X3 rising phase, indicating that the opening of P2X3 was accelerated. When H89 and SL-NH2 were applied together, τ values did not further decrease compared with those of SL-NH2 treatment. This implied that the reduction of currents amplitude and the acceleration of channel opening might caused by different ways.

Conclusion

In short, according to the above results, we conclude that, by activating PAR2 receptors on DRG neurons, trypsin produces inhibition of P2X3 current amplitude within a short time, and this effect is not through the PKA pathway. Meanwhile, the opening of P2X3 is speeded up by PAR2 activation. Inhibition of PKA plays a synergistic effect on the inhibition of P2X3 current amplitude.

Acknowledgments

This research was supported by the National Natural Science Foundation of China (No. 30801071) and the Shanghai Municipal Science and Technology Commission (No. 074119609).