Modulatory effect of CCK-8S on GABA-induced depolarization from rat dorsal root ganglion
Introduction
Cholecystokinin (CCK) is a bioactive peptide that functions as a gastrointestinal hormone and as a neurotransmitter in both the central and peripheral nervous systems (Crawley and Corwin, 1994). Its C-terminal sulfated octapeptide (CCK-8S) exists abundantly in the neuron system (Dockray 1976; Vanderhaeghen et al. 1975), particularly, in hypothalamus, amygdala, cerebral cortex and hippocampus (De Belleroche et al. 1990; Innis et al. 1979). CCK-8S has been shown to be involved in behavior (Stanfa et al., 1994), anxiety (Hughes et al., 1990), learning/memory processes (Katsuura, 1986) and neuropathic pain (Cahill et al. 2003; Cesselin 1995). Two CCK receptor subtypes have been identified and cloned, CCK-A receptor (CCK-AR) and CCK-B receptor (CCK-BR) (Wank, 1998). The CCK-AR, located primarily in the gastrointestinal tract, has a high affinity for CCK-8S and a relatively low affinity for the non-sulfated form of CCK, CCK-4 and gastrin. CCK-BR has high and nearly equal affinities for CCK-8S, CCK-4 and gastrin. CCK-BR is the principal receptor for the neurotransmitter CCK-8S in nervous system. Evidence for the existence of neuronal CCK-AR has also been reported (Boden and Woodruff 1994; Day et al. 1994; Mercer and Beart 1997; Wang et al. 1998). CCK-AR is reported to activate neurons through suppressing potassium conductance in rat dorsal motor nucleus (Zheng et al., 2005) or activating L-type calcium channel in cultured myenteric neurons (Zhang et al., 2002) or nonselective cation channels in nodose ganglion (Dun et al., 1991). CCK-BR belongs to a superfamily of G protein-coupled receptors. Direct coupling of CCK-BR to Gq that activates phospholipase C (PLC) and subsequently mobilizes intracellular calcium stores has been demonstrated in pancreatic acinar cells (Yule et al., 1999) and in several cell lines (Detjen et al. 1997; Yule et al. 1993). CCK-8S has also been reported to alter [Ca2+]i level by increasing extracellular calcium influx through N-type voltage-sensitive calcium channels in cultured striatal neurons (Miyoshi et al., 1991).#
CCK-8S is one of the strongest endogenous anti-opioid substances found in the primary sensory pathway. The CCK receptors modulate the opioid system in physiological processes, such as pain control, mood alteration including emotional and/or motivational responses (Cesselin 1995; Wang and Randic 1994). CCK-BR antagonists have been used to potentiate analgesia and antidepressant-like effects induced by endogenous or exogenous opioid agonists (Noble et al., 1999). The possible mechanism of the anti-opioid effect of CCK-8S is that binding of CCK-8S to the CCK receptor reduces the binding affinity of morphine ligands (Wang et al. 1989; Wang and Han 1990), implying that direct ligand–receptor interactions between CCK-8S and opioids might exist. CCK-8S also counteract the effect of opioid receptor by activation on Ca2+ influx and uptake (Liu et al., 1995).#
GABA is the primary inhibitory transmitter in the spinal cord, which reduces release of excitatory transmitter from primary afferent nerve terminals, termed ‘pre-synaptic inhibition’ (Eccles, 1964). It has been suggested that phosphorylation and dephosphorylation of the GABA-A receptor (GABA-AR)-chloride channel complex are involved in modulation of the GABA response (Chen et al., 1990). GABA-AR is downregulated by direct phosphorylation via protein kinase C (PKC) (McMahon and Koltzenburg 1990; Si et al. 2004; Yamada and Akasu 1996). Biochemical studies have also suggested that the phosphorylation of GABA-AR by various protein kinases inhibits GABA-AR function (Gyenes et al., 1994). Opioid peptides are established regulators of primary afferent neurotransmission, including nociception (Kolaj and Randic 1996; Wang and Randic 1994). A bicuculline-sensitive GABA-induced current is potentiated by opioid peptides. Since these receptors (GABA-AR, CCK-BR, CCK-AR and opioid receptors) are co-expressed in the same rat DRG neuron (Ghilardi et al. 1992; Verge et al. 1993; Xu et al. 1993; Yamada and Akasu 1996; Zhang et al. 2002), an interaction might occur among these receptors if they are activated simultaneously. In the present study, intracellular recording technique is performed to explore the effects of CCK-8S on the GABA-AR-mediated responses in neurons freshly isolated from rat DRG. A preliminary report was published in abstract form (Ke-Tao Ma, Jun-Qiang Si et al., Neuroscience Bulletin. 2005).#
Results
CCK-8S-evoked membrane depolarization in rat DRG neurons
Experiments were carried out in 247 DRG neurons with a mean resting membrane potential of −58.73±6.20 mV. In 36 out of 143 neurons examined (25.2%), CCK-8S (3×10−9–10−6 mol/L) evoked a depolarizing response. The membrane depolarization induced by CCK-8S was concentration-dependent and abolished by prolumide (10−4 mol/L), a selective antagonist of CCK-AR (n=8). LY225910 (10−4 mol/L), a selective antagonist of CCK-BR (n=9), had no effect on CCK-8S-evoked membrane depolarization (Fig. 1).#
GABA-induced depolarization
The majority of cells (89.5%, 221/247) examined in the present experiment were sensitive to bath application of GABA (10−6–10−3 mol/L) with a depolarizing response. The GABA (10−6–10−3 mol/L)-evoked response was in a concentration-dependent manner. The threshold was about 10−6 mol/L and the maximal response was achieved by 3×10−4 mol/L GABA. The Kd value was about 3×10−5 mol/L deduced from the concentration–response curve (Fig. 3). The selective GABA-AR agonist muscimol (10−4 mol/L) mimicked GABA-evoked response (n=10). A selective GABA-AR antagonist bicuculline (10−4 mol/L, n=7) suppressed both GABA- and muscimol-evoked responses (10−4 mol/L, n=10) (Fig. 2).#
Inhibition of GABA-induced depolarization by CCK-8S
When CCK-8S was pre-incubated for 2 min prior to application of GABA, the GABA-induced depolarization was attenuated markedly in most of the neurons examined (86.7%, 124/143). Pre-application of CCK-8S (10−7 mol/L) shifted the concentration–response curve of GABA-induced depolarization downward. The estimated Kd value was about 3×10−5 mol/L in the presence of CCK-8S (10−7 mol/L). There was no statistical difference between the thresholds, the GABA concentrations that evoked the maximal response and the Kd values for GABA in the absence and presence of CCK-8S (Fig. 3). Inhibition of GABA-induced response by CCK-8S was concentration-dependent and increased gradually with the increase in CCK-8S concentration. The depolarization induced by GABA (10−4 mol/L) was suppressed by 10.7%±4.2%, 27.8%±5.0%, 39.4%±4.4%, 47.8%±4.2%, 56.3%±4.6%, 58.8%±6.2% by CCK-8S 3×10−9, 10−8, 3×10−8, 10−7, 3×10−7 and 10−6 mol/L (n=11∼21), respectively (Fig. 4). In our experiments, the inhibition of CCK-8S on GABA and muscimol-induced depolarization was associated with a decrease in conductance. The conductance was increased to 20.2 pS±5.3 pS (n=6) and 19.8 pS±5.9 pS (n=6) by GABA (10−4 mol/L) and muscimol (10−4 mol/L), respectively, from the control of 12.7 pS±3.7 pS (n=7), then decreased to 16.0 pS±6.3 pS (n=6) and 15.8 pS±3.1 pS (n=6), respectively (Fig. 5). These results indicated that CCK-8S inhibited GABA-induced depolarization in rat DRG neurons mainly through suppressing the efficacy of GABA-AR.#
Time-course of CCK-8S inhibition of GABA-induced depolarization
The inhibitory effect of CCK-8S increased with the crease of pre-incubation duration and the inhibition reached the peak at about 2 min pre-incubation of CCK-8S. The inhibitory effect of CCK-8S on GABA-induced depolarization decreased when the interval between pre-incubation with CCK-8S and application of GABA was increased and it took about 16 min to get a full recovery from CCK-8S-induced inhibition (Fig. 6).#
CCK-8S inhibited GABA-induced depolarization through CCK-BR
The CCK-8S-induced inhibition on GABA-induced depolarization was abolished by the selective CCK-BR antagonist LY225910 (10−4 mol/L, n=6) but not by the selective CCK-AR antagonist prolumide (10−4 mol/L, n=7). In 30 out of 143 cells, we found that CCK-8S induced both membrane depolarization and inhibition of GABA-induced depolarization in the same cells. It was speculated that CCK-AR and CCK-BR might coexist in some DRG neurons. With pre-incubating different-concentration LY225910 and prolumide, the inhibiting ratios of CCK-8S-induced inhibition on GABA-evoked membrane depolarization were 32.4%±4.5%, 7.7%±3.2%, −3.2%±6.4% (LY225910, 10−6 mol/L–10−4 mol/L, n=6–8), 47.2%±3.9%, 45.8%±5.4% and 42.5%±7.5% (prolumide, 10−6 mol/L–10−4 mol/L, n=6–7), respectively (Fig. 7).#
Analysis of intracellular signal transduction mechanisms underlying the CCK-8S-induced inhibition on GABA-induced depolarization
To explore intracellular signal transduction mechanisms underlying CCK-8S inhibition of GABA-induced depolarization, we used U73122 (a selective PLC inhibitor), BAPTA-AM (a highly selective calcium chelating reagent), chelerythrine (a selective PKC inhibitor) and H-89 (a selective PKA inhibitor). The CCK-8S-induced inhibition on GABA-evoked membrane depolarization was strongly suppressed by U73122 (10−6 mol/L, n=6), chelerythrine (10−6 mol/L, n=8) and BAPTA-AM (10−4 mol/L, n=7), respectively. However, H-89 (10−6 mol/L, n=9) had no effect on CCK-8S inhibition of GABA-evoked response. With pre-incubating different-concentration chelerythrine, U73122, H-89 and BAPTA-AM, the inhibiting ratios of CCK-8S-induced inhibition on GABA-evoked membrane depolarization were 41.4%±3.5%, 32.5%±4.6%, 1.8%±5.6% (chelerythrine, 10−6 mol/L–10−4 mol/L, n=6–9), 45.3%±2.1%, 21.8%±4.3%, 3.8%±7.0% (U73122, 10−8 mol/L–10−6 mol/L, n=6–9), 46.6%±4.9%, 48.8%±4.8%, 44.6%±3.2% (H-89, 10−6 mol/L–10−4 mol/L, n=6–8), 36.3%±5.3%, 23.5%±4.9% and 2.2%±3.2% (BAPTA-AM, 10−6 mol/L–10−4 mol/L, n=6–7), respectively (Fig. 8).#
Discussion
Both CCK and GABA are important neurotransmitters and/or neuromodulators in both the central and peripheral nervous systems. In DRG neurons, we recorded membrane depolarizing responses evoked by CCK and the selective CCK-AR antagonist prolumide but not the selective CCK-BR antagonist LY225910 suppressed the CCK-evoked membrane depolarizing response, indicating that the CCK-AR mediates the excitatory response of CCK in rat DRG neurons. CCK also evokes excitatory responses in neurons in the brain and gastrointestinal tract by activating the CCK-AR (Gokin et al. 1997; Tsujino et al. 2005). However, CCK-BR also mediates CCK-evoked membrane responses in rat hippocampus and rostral nucleus accubens (Boden and Hill 1988; Dodd and Kelly 1981; Kombian et al. 2004). Although the detailed ionic mechanisms involved in CCK-AR-mediated membrane depolarizing response in the rat DRG were not elucidated in the present study, nonselective cation channels have been shown to be involved in CCK-AR-mediated response in the brain neurons. As we have previously reported (Si et al., 1997), GABA-AR mediated GABA-evoked membrane depolarizing response in rat DRG neurons since the selective GABA-AR agonist muscimol mimicked GABA-activated response and the selective GABA-AR antagonist bicuculline blocked GABA-evoked membrane response.#
Pre-incubation of CCK-8S for 2 min suppressed the GABA-evoked membrane depolarization markedly in most of the neurons examined. Inhibitory effect of CCK-8S was concentration-dependent and increased gradually with the increase in CCK-8S concentrations. It was clear that activation of CCK-BR suppressed GABA-AR function because the inhibitory effect of CCK-8S was abolished by the selective CCK-BR antagonist LY225910 but not the selective CCK-AR antagonist prolumide. Therefore we conclude that both CCK-AR and CCK-BR are expressed in rat DRG neurons. The excitatory membrane response of CCK is mediated by CCK-AR. On the other hand, CCK-BR mediates an inhibitory action on GABA-AR in rat DRG neurons.#
Stronger inhibition with increased pre-incubation time up to 2 min for CCK-8S implies that intracellular transduction mechanisms might be involved in the modulatory effect instead of direct interaction of CCK-8S with GABA-AR. The long-lasting inhibitory effect of CCK-8S (up to 16 min or more) also suggests the involvement of intracellular signal molecules. To test this hypothesis, we used U73122 (10−6 mol/L, a selective PLC inhibitor), BAPTA-AM (10−4 mol/L, a highly selective calcium chelating reagent), chelerythrine (10−6 mol/L, a selective PKC inhibitor) and H-89 (10−6 mol/L, a selective PKA inhibitor). The CCK-8S-induced inhibition was almost removed by U73122, BAPTA-AM and Chelerythrine, but not H-89. Therefore, we speculated that the possible intracellular signal transduction mechanisms of the CCK-8S-induced inhibition on GABA-evoked depolarization might be the binding of CCK-8S to the CCK-BR to activate G-protein, which in turn activates PLC and subsequently mobilizes intracellular calcium, and activation of PKC which phosphorylates GABA-AR that inhibits its function in DRG neurons (Chen et al., 1990). How is recovery of the blockade mediated? Some researchers found that activated calcineurin (CaN) could make type-A GABA receptor (GABA-ARs) dephosphorylate via the direct binding of CaN catalytic domain to the second intracellular domain of the GABA-AR-γ2 subunits (Huang and Dillon, 1998).#
GABA is an established inhibitory neurotransmitter that acts through GABA-AR. It opens the Cl− channel and is involved in primary afferent depolarization (PAD), an effect known as “pre-synaptic inhibition”. This action of GABA results in the decrease of release of neurotransmitter, including SP and glutamate, from primary afferent terminals. Peripheral nerve injury increases the expression of CCK receptor mRNA transcripts, in particular, CCK-BR mRNA in DRG neurons (Ghilardi et al. 1992; Verge et al. 1993; Xu et al. 1993; Zhang et al. 1993). If CCK-8S suppressed the GABA response at the central terminal of primary afferent neurons by activating the CCK-BR, as it did at the soma membrane, then dis-inhibition of the “pre-synaptic inhibition” would directly result in the facilitation of nociception in the spinal cord, CCK-8S then might directly be associated with the modulation of primary sensory information (including pain) at the central terminal of primary afferent neurons.#
GABA-evoked response is potentiated by opioid peptides and opioid peptides regulate the strength of primary afferent neurotransmission and potentiate the ‘pre-synaptic inhibition’ of GABA (Kolaj and Randic 1996; Wang and Randic 1994). CCK-8S immunoreactivity is predominantly associated with fibers within the brainstem rostral ventromedial medulla (RVM) (Skinner et al., 1997), and CCK-8S-containing projections from the RVM to the spinal cord have been found (Ghilardi et al., 1992). It is well known that CCK-8S in small DRG neurons serves as a pain transmitter or as a modulator of noxious transmission in the dorsal horn of the spinal cord (Broberger et al. 2000; Nicoll et al. 1980). Since the peptide CCK-8S is a potent physiological antagonist of the antinociceptive effect of opiates (Boden and Hill 1988; Cesselin 1995), it is suggested that enhanced activity of the CCK after nerve injury might be one of the possible mechanisms for opiate insensitivity in neuropathic pain (Chen et al., 1990; Dauge et al., 1995; Stanfa et al., 1994; Wiesenfeld-Hallin and Xu, 1996; Wiesenfeld-Hallin et al., 1990). CCK-8S which suppressed GABA-induced depolarization through CCK-BR might be contributing to the anti-opioid effect of CCK-8S (Andre et al., 2005).#
Experimental procedures
DRG preparation
Two- to three-weeks-old Wistar rats, irrespective of sex, were used in this experiment (Si et al. 1997 2004). The rats were anaesthetized with ether followed with laminectomy at L4 or L5. The DRGs with attached dorsal roots and spinal nerves were dissected out, the fibrous sheath surrounding the DRG was torn off carefully under the stereoscope, then the isolated preparation was transferred into recording chamber (0.25 ml volume), and superfused with oxygenated balanced salt solution (BSS) at room temperature. BSS contains (mmol/L): NaCl 140, KCl 5, MgCl2 1, glucose 5 and Tris–HCl 5 (pH 7.4). The flow rate was 3–5 ml/min. The preparation was pinned with tiny steel pins onto a silicone gum block, which was placed on the bottom of the chamber. The sciatic nerve was placed on a pair of platinum stimulating electrodes in the neighboring compartment.#
Intracellular recording
Intracellular recordings were obtained using glass microelectrode filled with 2 mol/L KCl and 1 mol/L potassium acetate, the DC resistance of which was in the range of 25–60 MΩ. The membrane potentials were amplified with a microelectrode amplifier (MEZ-8301, Japan) and membrane depolarization was filtered at 20 Hz. The data were recorded with a pen recorder (XWTD-264, China). The values of resting membrane potentials used in the preparations were stable at least for 10–20 min.#
Drugs
The drugs used were GABA, muscimol, bicuculline, CCK-8S, prolumide, chelerythrine and H-89, purchased from Sigma (St. Louis, MO). LY225910 was from Tocris (UK), BAPTA-AM and U73122 were purchased from Merck. All drugs were dissolved in BSS.#
Statistic analysis
Values of GABA-induced depolarization were recorded as mean±SE and t test was used to test the significance.#
Acknowledgments
We wish to thank Dr. Hong-Zhen Hu for reading the manuscript and for valuable suggestions. Work supported by the National Natural Science Foundation of China, No. 30160026 and No. 39860027.#
Figures and Tables
References
6. F.CesselinOpioid and anti-opioid peptidesFundam. Clin. Pharmacol.91995409433
8. J.N.CrawleyR.L.CorwinBiological actions of cholecystokininPeptides151994731755
13. G.J.DockrayImmunochemical evidence of cholecystokinin-like peptides in brainNature2641976568570
16. J.C.EcclesPresynaptic inhibition in the spinal cordProg. Brain Res.1219646591
27. S.B.McMahonM.KoltzenburgNovel classes of nociceptors: beyond SherringtonTrends Neurosci.131990199201
30. R.A.NicollC.SchenkerS.E.LeemanSubstance P as a transmitter candidateAnnu. Rev. Neurosci.31980227268