It has been reported previously that norepinephrine, when applied to the spinal cord dorsal horn, excites a subpopulation of dorsal horn neurons, presumably inhibitory interneurons. In the current study, the authors tested whether norepinephrine could activate inhibitory interneurons, specifically those that are "GABAergic."
A transverse slice was obtained from a segment of the lumbar spinal cord isolated from adult male Sprague-Dawley rats. Whole-cell patch-clamp recordings were made from substantia gelatinosa neurons using the blind patch-clamp technique. The effects of norepinephrine on spontaneous GABAergic inhibitory postsynaptic currents were studied.
In the majority of substantia gelatinosa neurons tested, norepinephrine (10-60 microM) significantly increased both the frequency and the amplitude of GABAergic inhibitory postsynaptic currents. These increases were blocked by tetrodotoxin (1 microM). The effects of norepinephrine were mimicked by the alpha1-receptor agonist phenylephrine (10-80 microM) and inhibited by the alpha1-receptor-antagonist WB-4101 (0.5 microM). Primary-afferent-evoked polysynaptic excitatory postsynaptic potentials or excitatory postsynaptic currents in wide-dynamic-range neurons of the deep dorsal horn were also attenuated by phenylephrine (40 microM).
The observations suggest that GABAergic interneurons possess somatodendritic alpha1 receptors, and activation of these receptors excites inhibitory interneurons. The alpha1 actions reported herein may contribute to the analgesic action of intrathecally administered phenylephrine.
IN the companion article by Baba et al. , 1we demonstrated that norepinephrine increases the release of both γ-aminobutyric acid (GABA) and glycine through presynaptic mechanisms, without influencing postsynaptic GABA or glycine receptor sensitivity. To extend these observations, we performed an additional series of experiments to evaluate whether norepinephrine acts at somatic or dendritic sites to directly depolarize “GABAergic” interneurons. Here we report that norepinephrine activates α1receptors on either the soma or the dendrites of GABAergic neurons to elicit tetrodotoxin-sensitive, large-amplitude inhibitory postsynaptic currents (IPSCs) in substantia gelatinosa (SG) neurons. Norepinephrine acts through α1receptors located on presynaptic nerve terminals and on the soma or dendrites of GABAergic interneurons to suppress the activity of wide-dynamic-range (WDR) neurons in the deep dorsal horn.
Materials and Methods
Spinal Cord Slice Preparation and Patch-clamp Recording from Substantia Gelatinosa Neurons
This study was approved by the Animal Care and Use Committee at the Niigata University School of Medicine. Methods for preparing adult rat spinal cord slices and patch-clamp recording from SG neurons are described in detail in the companion article. 1In some experiments, a dorsal root (L4) was preserved to permit stimulation of primary-afferent fibers. Unless otherwise indicated, GABAergic IPSCs were recorded in the presence of D L -2-amino-5-phosphonovarelic acid (APV, 25 μM), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 μM), and strychnine (1 or 2 μM), to block N -methyl-D-aspartate (NMDA), α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA), and glycine receptors, respectively.
Intracellular and Patch-clamp Recording from Deep Dorsal Horn Neurons
Conventional intracellular “sharp” electrode recordings were made from neurons located in lamina IV–V using an Axoclamp 2A (Axon Instruments, Foster City, CA). The resistance of a typical sharp electrode was 150–200 MΩ when filled with 4 M potassium acetate. In some cases, blind patch-clamp recordings were also made from lamina IV–V neurons using an Axopatch 200A amplifier (Axon Instruments). The internal solution contained 135 mM potassium gluconate, 5 mM KCl, 0.5 mM CaCl2, 2 mM MgCl2, 5 mM EGTA, 5 mM HEPES, 5 mM adenosine triphosphate magnesium salt, and 0.5 mM guanosine triphosphate sodium salt. The resistance of a typical patch pipette was 5–10 MΩ. Neurons were voltage-clamped to −70 mV for recording excitatory postsynaptic currents (EPSCs). Dorsal roots were stimulated using a suction electrode. 2Minimum stimulus intensities necessary to activate Aα or β (10 μA, 0.05 ms), Aδ (30 μA, 0.05 ms) and C (200 μA, 0.5 ms) fibers were determined previously by extracellular recording of compound action potentials from the dorsal root near the dorsal root entry zone. 1Signals were filtered at 2 kHz and digitized at 5 kHz.
Drugs were applied by exchanging the perfusion solution with a solution that contained a known drug concentration, without altering the perfusion rate or the temperature. All drugs were from Sigma (St. Louis, MO) unless otherwise specified. The following drugs were used: norepinephrine (WAKO, Osaka, Japan), CNQX (Tocris Cookson, Ballwin, MO), APV, strychnine, bicuculline, tetrodotoxin (WAKO), 2-(2,6-Dimethoxyphenoxyethyl)aminomethyl-1,4-benzodioxane hydrochloride(WB-4101), propranolol, yohimbine (WAKO), phenylephrine, clonidine, and isoproterenol.
Numerical data are presented as the mean ± SD (unless otherwise stated). Modulation of the frequency of postsynaptic currents was analyzed using the Student paired t test. The effects of tetrodotoxin and selective antagonists were analyzed using analysis of variance, with the Scheffé test for post hoc comparison. The Kolmogorov-Smirnov test was used to compare the effect of norepinephrine on IPSC amplitude distributions. P < 0.05 was considered significant and is indicated by an asterisk in the figures.
Norepinephrine Increases the Frequency and Amplitude of Spontaneous Inhibitory Postsynaptic Currents
As shown in figure 1, in the absence of tetrodotoxin, norepinephrine markedly increased the frequency of spontaneous GABAergic IPSCs. This effect of norepinephrine is distinct from the effect on miniature IPSCs that is observed in the presence of tetrodotoxin. 1In the absence of tetrodotoxin, norepinephrine elicited repetitive large-amplitude IPSCs that were not present in the control state. The baseline frequency of IPSCs was 6.5 ± 3.1 Hz (n = 32; range 2.2–15.6 Hz). Norepinephrine (10–60 μM) increased the frequency of IPSCs in 46 of 51 SG neurons tested (587 ± 212% of control for norepinephrine 20 μM; n = 12;P < 0.0001; paired t test). For the neuron in figure 1, there was a persistent elevation in the baseline current after norepinephrine application that most likely reflects IPSC summation because it was tetrodotoxin-sensitive (data not shown). Norepinephrine-induced IPSCs were GABAAreceptor–mediated because they were eliminated by bicuculline (20 μM; n = 5;fig. 2A). Norepinephrine-induced large-amplitude IPSCs also were eliminated by tetrodotoxin (fig. 2B); however, the increase in small-amplitude IPSC frequency persisted in the presence of tetrodotoxin (fig. 2B, bottom; see Baba et al. 1).
The effects of norepinephrine and tetrodotoxin on IPSC amplitude were further analyzed by constructing amplitude histograms. As shown in figure 3A(top), most IPSCs were relatively small, clustering primarily between 0 and 20 pA. Overall, the median IPSC amplitude was 10.1 ± 1.5 pA (n = 6; range 8.5–12.5 pA) during control conditions. In marked contrast, the presence of norepinephrine shifted IPSC amplitude distribution to the right. The majority of events were greater than 20 pA (fig. 3A, middle). This shift is not caused by temporal summation of increased IPSCs because the amplitude of each IPSC was measured from the initial deflection point (not from the baseline) to the peak of the event (fig. 3B). Norepinephrine increased the median amplitude of IPSCs in all cells tested (n = 6; 337 ± 98% of control; range, 201–458%). This change in amplitude distribution is clearly shown in the cumulative histogram (fig. 3C). In the presence of norepinephrine, the relative frequency curve was significantly shifted to the right (P < 0.0001 in all six cells tested; Kolmogorov-Smirnov test), and in each instance this shift was inhibited by tetrodotoxin. It is unlikely that large-amplitude IPSCs were miniature IPSCs of quantal size (q ) following a Gaussian distribution. More likely, large-amplitude IPSCs represent coordinated synaptic release caused by norepinephrine-induced firing of GABAergic interneurons. Large-amplitude IPSCs (> 40 pA) were abolished by the addition of tetrodotoxin (1 μM) to the bathing medium (fig. 3A, bottom), indicating that norepinephrine-enhanced GABA release is via action-potential propagation. The norepinephrine-induced increase in total IPSC frequency, however, could not be completely blocked by tetrodotoxin (256 ± 57% of control frequency, fig. 3D) because the frequency of the relatively small amplitude IPSCs (< 40 pA) remained elevated even in the presence of tetrodotoxin. This is most likely caused by the facilitatory effect of norepinephrine on quantal release from presynaptic axon terminals, as illustrated in the article Baba et al. 1
Norepinephrine Increases Large-amplitude Spontaneous Inhibitory Postsynaptic Currents via α1Receptors
Next, the identity of the adrenergic receptor subtype responsible for the norepinephrine-induced increase in large-amplitude spontaneous IPSCs was evaluated. In all SG neurons tested (n = 8), the α1-receptor agonist phenylephrine (10–80 μM) increased the frequency of spontaneous IPSCs and elicited large-amplitude IPSCs, which were not present in the control state (fig. 4A). However, neither the α2-receptor agonist clonidine (10–40 μM n = 12) nor the nonselective β-receptor agonist isoproterenol (10–100 μM; n = 9) had any effect (data not shown). Together, this suggests that the norepinephrine-induced increase in IPSCs was mediated by α1receptors. To confirm these observations, we tested the effects of selective adrenergic receptor antagonists on norepinephrine-induced increases in IPSC frequency. WB-4101, an α1- receptor antagonist (0.5 μM; n = 7), reversibly blocked the norepinephrine-induced increase in IPSC frequency and the induction of large-amplitude IPSCs (figs. 4B and C). Neither the α2-receptor antagonist yohimbine (1 μM) nor the nonselective β-receptor antagonist propranolol (1 μM) blocked the effects of norepinephrine (fig. 4C), confirming the role of α1receptors.
α1-Receptor Agonist Attenuates Polysynaptic EPSPs and EPSCs in Deep Dorsal Horn Neurons
It has been reported that norepinephrine also inhibits WDR neurons located in the deep dorsal horn. 3To determine whether α1receptors also mediate this action of norepinephrine, conventional intracellular sharp electrode recordings and blind patch-clamp recordings were made in the absence of APV and CNQX (n = 15;fig. 5A). According to synaptic responses to graded intensity stimulation of the dorsal root, 11 of 15 deep dorsal horn neurons tested were classified as WDR; the remainder were low-threshold neurons. C-fiber electrical stimulation of the dorsal root usually evoked early, fast monosynaptic excitatory postsynaptic potentials and EPSCs [EPSP(C)s] and late, slow polysynaptic EPSP(C)s in WDR neurons (figs. 5B and C). In the majority of WDR neurons, phenylephrine (40 μM) reversibly attenuated polysynaptic EPSP(C)s, whereas monosynaptic EPSP(C)s appeared to be unchanged (fig. 5B, n = 9). In the remaining two WDR neurons, EPSP(C)s evoked by dorsal root stimulation (which had both fast and slow components) were unaffected by phenylephrine (data not shown). Synaptic responses in low-threshold neurons evoked by C-fiber intensity stimulation (n = 4) were essentially the same as those evoked by low-threshold afferent fiber (A-fiber) stimulation, consisting usually of fast, short-lasting (< 200 ms) monosynaptic EPSP(C)s (fig. 5D1). These monosynaptic EPSP(C)s were unaffected by phenylephrine (n = 4;fig. 5D2).
Mechanisms of adrenergic modulation of GABAergic synaptic transmission in the SG of the rat spinal dorsal horn were evaluated. In the presence of glutamate and glycine receptor antagonists, norepinephrine increased the amplitude and frequency of GABAAreceptor–mediated IPSCs in the majority of SG neurons tested. The appearance of large-amplitude IPSCs was tetrodotoxin sensitive, whereas the increase in IPSC frequency was only partially tetrodotoxin sensitive. The facilitatory effect of norepinephrine on GABAergic transmission was mimicked by phenylephrine and antagonized by WB-4101, whereas clonidine, isoproterenol, yohimbine, and propranolol each had no effect. Those observations indicate that adrenergic modulation of GABA-mediated fast synaptic transmission results from the activation of α1receptors located primarily at somatodendritic sites of GABAergic interneurons and, to a lesser degree, on the presynaptic terminals of GABAergic interneurons.
Adrenergic Excitation of GABAergic Neurons
The amplitude histogram for IPSCs recorded in the presence of norepinephrine showed an increase in large-amplitude IPSCs. The majority of these large-amplitude events appeared to result from propagation of action potentials from the soma of GABAergic interneurons to presynaptic terminals because they were blocked by tetrodotoxin. These large-amplitude IPSCs suggest the possibility that GABAergic interneurons are depolarized by norepinephrine acting via α1receptors (fig. 6), a mechanism that has been reported in other central nervous system regions. 4–6The norepinephrine-induced increase in overall IPSC frequency, however, could not be completely blocked by tetrodotoxin: the frequency of the relatively small-amplitude IPSCs remained elevated even in the presence of tetrodotoxin. This result is consistent with the observation in Baba et al. 1that norepinephrine facilitates quantal release of GABA from presynaptic terminals. Thus, norepinephrine activates GABAergic neurons at somatodendritic sites and at axon terminals.
The excitatory effects of norepinephrine appear to be selective for inhibitory neurons because the frequency of EPSCs recorded at −70 mV in the absence of glutamate receptor antagonists was unaffected (H. Baba, unpublished observations). The effects of norepinephrine reported herein seem to result from the activation of α1receptors on GABAergic interneurons because our data were obtained in the presence of glutamate receptor antagonists (therefore precluding a polysynaptic, or indirect, effect). Although the exact location of the GABAergic interneurons affected by norepinephrine is unknown, it is possible that they reside within the SG because some cell bodies and terminals of SG neurons are thought to contain GABA. 7Indeed, norepinephrine has been shown to depolarize a subset of SG neurons through α1-receptor activation, 8and norepinephrine produced inward currents in some spinal SG neurons when guanosine-5′-0-(2-thiodiphosphate) was omitted from the pipette solution (H. Baba, unpublished observation).
As discussed in the the article by Baba et al. , 1the majority of SG neurons are local interneurons and do not project to the thalamus. 9The main projections of SG neurons are to lamina I and to the deep dorsal horn (lamina IV–V), where the projection neurons to the thalamus are located. Therefore, in the current study, we tested whether α1-receptor activation inhibits nociceptive input to deep dorsal horn neurons. Our data show that α1-receptor activation inhibited polysynaptic, but not monosynaptic, input to deep dorsal horn neurons.
Based on these considerations, we propose the model circuit shown in figure 6. Nociceptive information, conveyed via Aδ and C fibers, is transmitted to neurons within the SG, the first relay for such information in the central nervous system. SG neurons in turn transmit this information to projection neurons located in lamina I or lamina IV–V and, in so doing, create a simple polysynaptic pathway. Release of norepinephrine by descending adrenergic fibers, however, suppresses the feed-forward aspects of the circuit by activating α1receptors located on both the somatodendritic sites and the axon terminals of GABAergic interneurons. Activation of α1receptors on soma or dendrites of GABAergic interneurons leads to depolarization, which in turn can promote to release of GABA onto SG relay neurons. The SG relay neuron is hyperpolarized after the opening of GABAA-receptor channels, and the resting membrane potential is shifted away from the action potential threshold, thereby decreasing the likelihood that the original nociceptive signal will be transmitted to the thalamic projection neuron. Thus, norepinephrine might, by directly stimulating inhibitory interneurons, inhibit nociceptive polysynaptic input to lamina IV–V, while leaving monosynaptic responses unchanged.
As discussed in the article by Baba et al. , 1GABA may play an important role in spinal antinociception. GABAergic interneurons may be responsible for suppressing evoked excitatory responses. Increased activity of these inhibitory interneurons may decrease the excitability of SG neurons, thereby increasing the threshold for noxious inputs. Our current data suggest that the GABAergic system is, in turn, regulated by adrenergic inputs acting through α1adrenoreceptors. Given that nearly all norepinephrine-containing synaptic terminals in the dorsal horn of the spinal cord are supraspinal in origin, 10,11these descending adrenergic fibers may represent an intrinsic system for the regulation of GABAergic dorsal horn neurons. The α1action on somatodendritic sites of GABAergic interneurons, together with the facilitatory action on axon terminals, may contribute to the analgesic action of intrathecally administered phenylephrine.
The authors thank Dr. K. A. Moore for critically reading the manuscript.