Brain Research, 526 (1990) 291-298 Elsevier
291
BRES 15831
Evidence that substance P selectively modulates C-fiber-evoked discharges of dorsal horn nociceptive neurons David E. Kellstein 1, Donald D. Price 2, Ronald L. Hayes 3 and David J. Mayer 1 Departments of 1Physiology, 2Anesthesiology, and 3Neurosurgery, MedicalCollege of Virginia, Virginia Commonwealth University, Richmond, VA (U.S.A.) (Accepted 6 March 1990)
Key words: Substance P; Primary afferent; C-fiber; Nociception; Neuromodulator; [D-Pro2,D-Trp7.9]-substanceP; Electrophysiology; Morphine
Previous studies suggest that the undecapeptide substance P (SP) functions as a primary afferent neurotransmitter or neuromodulator of nociception which may mediate the slow temporal summation ('windup') of discharges of dorsal horn nociceptive neurons elicited by repetitive stimulation of C-afferents. The present study tested this hypothesis by investigating the effects of local spinal application of SP and an SP antagonist, [D-PrO2,D-TrpT'9]-SP(DPDT), on A- and C-fiber-evoked firing of dorsal horn neurons in an intact, urethane-anesthetized rat preparation. Extracellular single unit recordings from both wide dynamic range and nociceptive specific neurons during controlled repetitive electrical stimulation of the ipsilateral hind paw indicated that SP enhanced C-evoked firing in an apparent dose-related manner (100 > 20 = 4 nmol), whereas DPDT inhibited C-evoked discharges with an apparent bell-shaped dose-response (20 > 100 = 4 nmol). Neither agent significantly altered either A-evoked or spontaneous activity. In agreement with previous investigators, morphine sulfate also selectively inhibited C-fiber-evoked firing without altering A-fiber-mediated activity, validating the selectivity of our system. These findings provide additional evidence that SP functions as a neuromodulator of primary afferent nociception, and further suggest that the effects of SP are selective to nociceptive transmission mediated by C-fibers. INTRODUCTION Since the initial suggestion by Lembeck in 195323, considerable evidence has accumulated which supports the hypothesis that the undecapeptide substance P (SP) may function as a primary afferent neurotransmitter of nociception. Anatomically, SP has been localized in the dorsal horn terminals and the dorsal root cell bodies of unmyelinated and small myelinated neurons15'16; a recent study using electron microscopy further restricts SP immunoreactivity to unmyelinated (C-afferent) fibers 25. A variety of noxious stimuli cause release of SP in the dorsal horn which is inhibited by opioid pretreatment 14' 21,22, and iontophoretic application of SP selectively excites (depolarizes) nociceptive dorsal horn neurons which are activated by noxious chemical 27'35'41, thermal 12, or mechanical 34 stimuli. If endogenous SP were playing a major role in afferent nociceptive transmission, then exogenous application at its putative site of action should elicit a behavior indicative of pain perception. Previous investigators have demonstrated that a behavioral syndrome characterized by caudally-directed scratching and biting follows intra-
thecal SP administration to rodents 1'17'24'28"43, and have interpreted this response as indicative of a perceived noxious stimulation. However, recent detailed analysis of this behavior indicates that such behavior is (1) not selective to SP but rather is induced by a wide variety of pro-convulsive compounds 9, (2) potentiated rather than inhibited by intrathecal morphine 3, (3) inhibited by the anticonvulsants valproic acid and chlordiazepoxide s, and (4) unaffected by the N-methyl-D-aspartate antagonist, 2-amino-5-phosphonovalerate 2. On the basis of these observations, the authors concluded that (1) scratching is not indicative of pain but rather a spinal convulsive state, (2) SP does not elicit scratching by acting at the primary afferent synapse, (3) SP is neither necessary nor sufficient to elicit pain at the spinal level, and (4) SP is probably not a traditional neurotransmitter of nociception at the primary afferent synapse 2'3's'9. What role then, if any, does SP have in afferent nociceptive transmission? The experiments cited above do not rule out the possibility that SP performs a neuromodulatory function in afferent pain processing; in fact, additional experimental evidence is compatible with such a role. In behavioral studies, intrathecal adminis-
Correspondence: D.E. Kellstein, Exploratory Pharmacology, The Procter and Gamble Co., Miami Valley Laboratories, P.O. Box 398707, Cincinnati, OH 45239-8707, U.S.A. 0006-8993/90/$03.50 ~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
292 tration of SP a p p a r e n t l y produces hyperalgesia as indicated by decreased nociceptive thresholds m e a s u r e d with the thermal tail-flick43 and hot-plate 1'11 tests as well as the mechanical tail-pressure assay 24. Conversely, intrathecal injection of putative SP antagonists elevates nociceptive thresholds 1'28, although the paralytic effect of these agents may confound interpretation of such resuits 37. Eiectrophysiological studies indicate that iontophoretic administration of SP onto dorsal horn neurons typically results in a slowly developing, progressive depolarization of p r o l o n g e d duration m26. In a subsequent study, U r b a n and R a n d i c 36, using a tissue slice p r e p a r a t i o n , o b s e r v e d that repetitive electrical stimulation of dorsal roots (5-20 V, 0 . 2 - 0 . 5 ms duration applied at 1-20 H z for 3 - 5 s) elicited an initial rapid firing of spinal neurons followed by a slow depolarization. The latter response, but not the former, was m i m i c k e d by SP and inhibited by SP antagonists or the (non-selective) SP neurotoxin capsaicin. This p r o l o n g e d depolarization ind u c e d by SP m a y facilitate the depolarization induced by a fast-acting excitatory agent 13'36'42, such as the excitatory a m i n o acid l-glutamate 18'2°. Consistent with these considerations, SP has also been p r o p o s e d as the agent which m e d i a t e s slow t e m p o r a l summation ( ' w i n d u p ' ) of Cafferent e v o k e d responses in dorsal horn nociceptive neurons 29"31"32"42. This p r o p o s a l is s u p p o r t e d by the d e m o n s t r a t i o n , in both spinal cat 33 and rat 4° p r e p a r a tions, that repetitive stimulation of C-afferents, but not A-fibers, at frequencies greater than 0.3 Hz, results in t e m p o r a l s u m m a t i o n which is associated with slow memb r a n e depolarization of dorsal horn nociceptive neurons. This slow m e m b r a n e depolarization closely resembles that o b s e r v e d in the a f o r e m e n t i o n e d iontophoretic 12"26 and tissue slice 36 studies. If SP m e d i a t e s slow depolarizing effects of primary C-afferent stimulation, then application of exogenous SP or an SP antagonist should respectively facilitate or inhibit C - f i b e r - m e d i a t e d responses of dorsal horn nociceptive neurons. This hypothesis was tested in the present study by using extracellular electrophysiological recording to d e t e r m i n e the effect of local spinal application of SP or an SP antagonist, [D-Pro2,D-Trp7'9]-SP ( D P D T ) 7 on the A - and C-afferent-evoked activity of dorsal horn nociceptive neurons during repetitive electrical stimulation. To validate our e x p e r i m e n t a l model, we also e x a m i n e d the effect of analgesic doses of m o r p h i n e sulfate (MS), which selectively inhibits Cf i b e r - m e d i a t e d firing-s, on the same responses.
MATERIALS AND METHODS Male Sprague-Dawley rats (Hilltop) weighing 450-550 g were individually housed in standard laboratory cages under a 12-h light
cycle (on at 06.00 h). Laboratory chow and water were freely available. Animals were anesthetized with urethane (1.3 g/kg, i.p.). Following the initial dorsal incision and retraction, the underlying tissue was infiltrated with a 2% solution of lidocaine. A laminectomy was performed and the dura was retracted to expose spinal segments L4 and L5. An agar pool was constructed around the exposed cord and filled with warm (40 °C) physiological saline. These animals passively breathed room air supplemented with oxygen while core temperature was maintained at 38 + 0.5 °C throughout the experiments. Under such conditions, we have found that blood pH, pCO2, and pO 2 levels remain within normal ranges over the usual 2.5-3 h period required for surgery and data collection. Respective mean values (+ S.E.M., n = 3) for pH, pCO 2, and pO 2 were 7.39 + 0.05, 39.5 + 1.0, and 87.9 + 2.1. Extracellular single unit recordings were made from the dorsal horn using plastic-coated tungsten electrodes (Micro Probe Inc.). Neurons had receptive fields on the ipsilateral hindpaw and were classified as nociceptive specific (NS) or wide dynamic range (WDR) according to the criteria of Price et al. 31. In brief, a unit was classified as WDR if it responded to light brushing with a cotton swab and exhibited increased firing in response to light pinching with flattened forceps and heavy pinching with toothed forceps. A neuron was denoted as NS if it responded in an increasing manner to light and heavy pinching, but exhibited no response to light brushing. Following isolation and classification of a unit, a bipolar stimulating electrode was inserted into the center of the receptive field. Stimuli consisted of constant current rectangular wave pulses either supramaximal for C-afferent activation (3 mA, 2 ms duration) or subthreshold for C-afferent activation but supramaximal for stimulation of A-afferents (0.5 mA, 2 ms duration) as determined by recording compound action potentials from dorsal rootlets. For repetitive stimulation, pulses were delivered in trains (6 pulses, 0.5 Hz); such repetitive stimulation of C-afferents, but not A-afferents, has been shown to produce temporal summation in dorsal horn nociceptive neurons33'4°. After pretreatment responses were determined, the amount of saline in the pool was reduced until only the dorsal surface of the cord was exposed. Substance P (Sigma), DPDT (Sigma), or MS (Mallinckrodt) was each applied onto the cord at doses of 4, 20 and 100 nmol in a volume of 10/d. Whereas DPDT and MS were dissolved in saline, SP required a vehicle of dilute acetic acid in saline (60 mM, pH 3.8). Therefore, responses following application of this vehicle served as control for SP experiments. For those neurons exposed to more than one drug (SP followed by DPDT or MS), the cord was flushed with saline after exposure to the first drug and the unit was permitted at least 10 min to return to initial levels of activity before the second agent was applied. Discharges evoked by each of the 6 electrical stimuli were quantified at 1, 5, and 10 min after application of each dose of drug; results obtained at the time of maximal effect for each cell were analyzed. Dot raster traces obtained using a window discriminator and a storage oscilloscope at a high (5ms/div) and low (100 ms/div) sweep speed were used to quantify A- and C-fiber-evoked activity, respectively. Based upon an approximate distance of 15 cm from stimulating to recording electrode, only discharges occurring within 50 ms after stimulation (conduction velocity ~>3 m/s) were considered to be evoked by A-afferents; only impulses appearing at or later than 100 ms after a pulse (conduction velocity ~
291
BRES 15831
Evidence that substance P selectively modulates C-fiber-evoked discharges of dorsal horn nociceptive neurons David E. Kellstein 1, Donald D. Price 2, Ronald L. Hayes 3 and David J. Mayer 1 Departments of 1Physiology, 2Anesthesiology, and 3Neurosurgery, MedicalCollege of Virginia, Virginia Commonwealth University, Richmond, VA (U.S.A.) (Accepted 6 March 1990)
Key words: Substance P; Primary afferent; C-fiber; Nociception; Neuromodulator; [D-Pro2,D-Trp7.9]-substanceP; Electrophysiology; Morphine
Previous studies suggest that the undecapeptide substance P (SP) functions as a primary afferent neurotransmitter or neuromodulator of nociception which may mediate the slow temporal summation ('windup') of discharges of dorsal horn nociceptive neurons elicited by repetitive stimulation of C-afferents. The present study tested this hypothesis by investigating the effects of local spinal application of SP and an SP antagonist, [D-PrO2,D-TrpT'9]-SP(DPDT), on A- and C-fiber-evoked firing of dorsal horn neurons in an intact, urethane-anesthetized rat preparation. Extracellular single unit recordings from both wide dynamic range and nociceptive specific neurons during controlled repetitive electrical stimulation of the ipsilateral hind paw indicated that SP enhanced C-evoked firing in an apparent dose-related manner (100 > 20 = 4 nmol), whereas DPDT inhibited C-evoked discharges with an apparent bell-shaped dose-response (20 > 100 = 4 nmol). Neither agent significantly altered either A-evoked or spontaneous activity. In agreement with previous investigators, morphine sulfate also selectively inhibited C-fiber-evoked firing without altering A-fiber-mediated activity, validating the selectivity of our system. These findings provide additional evidence that SP functions as a neuromodulator of primary afferent nociception, and further suggest that the effects of SP are selective to nociceptive transmission mediated by C-fibers. INTRODUCTION Since the initial suggestion by Lembeck in 195323, considerable evidence has accumulated which supports the hypothesis that the undecapeptide substance P (SP) may function as a primary afferent neurotransmitter of nociception. Anatomically, SP has been localized in the dorsal horn terminals and the dorsal root cell bodies of unmyelinated and small myelinated neurons15'16; a recent study using electron microscopy further restricts SP immunoreactivity to unmyelinated (C-afferent) fibers 25. A variety of noxious stimuli cause release of SP in the dorsal horn which is inhibited by opioid pretreatment 14' 21,22, and iontophoretic application of SP selectively excites (depolarizes) nociceptive dorsal horn neurons which are activated by noxious chemical 27'35'41, thermal 12, or mechanical 34 stimuli. If endogenous SP were playing a major role in afferent nociceptive transmission, then exogenous application at its putative site of action should elicit a behavior indicative of pain perception. Previous investigators have demonstrated that a behavioral syndrome characterized by caudally-directed scratching and biting follows intra-
thecal SP administration to rodents 1'17'24'28"43, and have interpreted this response as indicative of a perceived noxious stimulation. However, recent detailed analysis of this behavior indicates that such behavior is (1) not selective to SP but rather is induced by a wide variety of pro-convulsive compounds 9, (2) potentiated rather than inhibited by intrathecal morphine 3, (3) inhibited by the anticonvulsants valproic acid and chlordiazepoxide s, and (4) unaffected by the N-methyl-D-aspartate antagonist, 2-amino-5-phosphonovalerate 2. On the basis of these observations, the authors concluded that (1) scratching is not indicative of pain but rather a spinal convulsive state, (2) SP does not elicit scratching by acting at the primary afferent synapse, (3) SP is neither necessary nor sufficient to elicit pain at the spinal level, and (4) SP is probably not a traditional neurotransmitter of nociception at the primary afferent synapse 2'3's'9. What role then, if any, does SP have in afferent nociceptive transmission? The experiments cited above do not rule out the possibility that SP performs a neuromodulatory function in afferent pain processing; in fact, additional experimental evidence is compatible with such a role. In behavioral studies, intrathecal adminis-
Correspondence: D.E. Kellstein, Exploratory Pharmacology, The Procter and Gamble Co., Miami Valley Laboratories, P.O. Box 398707, Cincinnati, OH 45239-8707, U.S.A. 0006-8993/90/$03.50 ~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
292 tration of SP a p p a r e n t l y produces hyperalgesia as indicated by decreased nociceptive thresholds m e a s u r e d with the thermal tail-flick43 and hot-plate 1'11 tests as well as the mechanical tail-pressure assay 24. Conversely, intrathecal injection of putative SP antagonists elevates nociceptive thresholds 1'28, although the paralytic effect of these agents may confound interpretation of such resuits 37. Eiectrophysiological studies indicate that iontophoretic administration of SP onto dorsal horn neurons typically results in a slowly developing, progressive depolarization of p r o l o n g e d duration m26. In a subsequent study, U r b a n and R a n d i c 36, using a tissue slice p r e p a r a t i o n , o b s e r v e d that repetitive electrical stimulation of dorsal roots (5-20 V, 0 . 2 - 0 . 5 ms duration applied at 1-20 H z for 3 - 5 s) elicited an initial rapid firing of spinal neurons followed by a slow depolarization. The latter response, but not the former, was m i m i c k e d by SP and inhibited by SP antagonists or the (non-selective) SP neurotoxin capsaicin. This p r o l o n g e d depolarization ind u c e d by SP m a y facilitate the depolarization induced by a fast-acting excitatory agent 13'36'42, such as the excitatory a m i n o acid l-glutamate 18'2°. Consistent with these considerations, SP has also been p r o p o s e d as the agent which m e d i a t e s slow t e m p o r a l summation ( ' w i n d u p ' ) of Cafferent e v o k e d responses in dorsal horn nociceptive neurons 29"31"32"42. This p r o p o s a l is s u p p o r t e d by the d e m o n s t r a t i o n , in both spinal cat 33 and rat 4° p r e p a r a tions, that repetitive stimulation of C-afferents, but not A-fibers, at frequencies greater than 0.3 Hz, results in t e m p o r a l s u m m a t i o n which is associated with slow memb r a n e depolarization of dorsal horn nociceptive neurons. This slow m e m b r a n e depolarization closely resembles that o b s e r v e d in the a f o r e m e n t i o n e d iontophoretic 12"26 and tissue slice 36 studies. If SP m e d i a t e s slow depolarizing effects of primary C-afferent stimulation, then application of exogenous SP or an SP antagonist should respectively facilitate or inhibit C - f i b e r - m e d i a t e d responses of dorsal horn nociceptive neurons. This hypothesis was tested in the present study by using extracellular electrophysiological recording to d e t e r m i n e the effect of local spinal application of SP or an SP antagonist, [D-Pro2,D-Trp7'9]-SP ( D P D T ) 7 on the A - and C-afferent-evoked activity of dorsal horn nociceptive neurons during repetitive electrical stimulation. To validate our e x p e r i m e n t a l model, we also e x a m i n e d the effect of analgesic doses of m o r p h i n e sulfate (MS), which selectively inhibits Cf i b e r - m e d i a t e d firing-s, on the same responses.
MATERIALS AND METHODS Male Sprague-Dawley rats (Hilltop) weighing 450-550 g were individually housed in standard laboratory cages under a 12-h light
cycle (on at 06.00 h). Laboratory chow and water were freely available. Animals were anesthetized with urethane (1.3 g/kg, i.p.). Following the initial dorsal incision and retraction, the underlying tissue was infiltrated with a 2% solution of lidocaine. A laminectomy was performed and the dura was retracted to expose spinal segments L4 and L5. An agar pool was constructed around the exposed cord and filled with warm (40 °C) physiological saline. These animals passively breathed room air supplemented with oxygen while core temperature was maintained at 38 + 0.5 °C throughout the experiments. Under such conditions, we have found that blood pH, pCO2, and pO 2 levels remain within normal ranges over the usual 2.5-3 h period required for surgery and data collection. Respective mean values (+ S.E.M., n = 3) for pH, pCO 2, and pO 2 were 7.39 + 0.05, 39.5 + 1.0, and 87.9 + 2.1. Extracellular single unit recordings were made from the dorsal horn using plastic-coated tungsten electrodes (Micro Probe Inc.). Neurons had receptive fields on the ipsilateral hindpaw and were classified as nociceptive specific (NS) or wide dynamic range (WDR) according to the criteria of Price et al. 31. In brief, a unit was classified as WDR if it responded to light brushing with a cotton swab and exhibited increased firing in response to light pinching with flattened forceps and heavy pinching with toothed forceps. A neuron was denoted as NS if it responded in an increasing manner to light and heavy pinching, but exhibited no response to light brushing. Following isolation and classification of a unit, a bipolar stimulating electrode was inserted into the center of the receptive field. Stimuli consisted of constant current rectangular wave pulses either supramaximal for C-afferent activation (3 mA, 2 ms duration) or subthreshold for C-afferent activation but supramaximal for stimulation of A-afferents (0.5 mA, 2 ms duration) as determined by recording compound action potentials from dorsal rootlets. For repetitive stimulation, pulses were delivered in trains (6 pulses, 0.5 Hz); such repetitive stimulation of C-afferents, but not A-afferents, has been shown to produce temporal summation in dorsal horn nociceptive neurons33'4°. After pretreatment responses were determined, the amount of saline in the pool was reduced until only the dorsal surface of the cord was exposed. Substance P (Sigma), DPDT (Sigma), or MS (Mallinckrodt) was each applied onto the cord at doses of 4, 20 and 100 nmol in a volume of 10/d. Whereas DPDT and MS were dissolved in saline, SP required a vehicle of dilute acetic acid in saline (60 mM, pH 3.8). Therefore, responses following application of this vehicle served as control for SP experiments. For those neurons exposed to more than one drug (SP followed by DPDT or MS), the cord was flushed with saline after exposure to the first drug and the unit was permitted at least 10 min to return to initial levels of activity before the second agent was applied. Discharges evoked by each of the 6 electrical stimuli were quantified at 1, 5, and 10 min after application of each dose of drug; results obtained at the time of maximal effect for each cell were analyzed. Dot raster traces obtained using a window discriminator and a storage oscilloscope at a high (5ms/div) and low (100 ms/div) sweep speed were used to quantify A- and C-fiber-evoked activity, respectively. Based upon an approximate distance of 15 cm from stimulating to recording electrode, only discharges occurring within 50 ms after stimulation (conduction velocity ~>3 m/s) were considered to be evoked by A-afferents; only impulses appearing at or later than 100 ms after a pulse (conduction velocity ~
