Research ArticlePain

Identification of an Adenylyl Cyclase Inhibitor for Treating Neuropathic and Inflammatory Pain

See allHide authors and affiliations

Science Translational Medicine  12 Jan 2011:
Vol. 3, Issue 65, pp. 65ra3
DOI: 10.1126/scitranslmed.3001269

Abstract

Neuropathic pain, often caused by nerve injury, is commonly observed among patients with different diseases. Because its basic mechanisms are poorly understood, effective medications are limited. Previous investigations of basic pain mechanisms and drug discovery efforts have focused mainly on early sensory neurons such as dorsal root ganglion and spinal dorsal horn neurons, and few synaptic-level studies or new drugs are designed to target the injury-related cortical plasticity that accompanies neuropathic pain. Our previous work has demonstrated that calcium-stimulated adenylyl cyclase 1 (AC1) is critical for nerve injury–induced synaptic changes in the anterior cingulate cortex. Through rational drug design and chemical screening, we have identified a lead candidate AC1 inhibitor, NB001, which is relatively selective for AC1 over other adenylate cyclase isoforms. Using a variety of behavioral tests and toxicity studies, we have found that NB001, when administered intraperitoneally or orally, has an analgesic effect in animal models of neuropathic pain, without any apparent side effects. Our study thus shows that AC1 could be a productive therapeutic target for neuropathic pain and describes a new agent for the possible treatment of neuropathic pain.

Introduction

Acute or physiological pain is necessary for animals and humans to survive in a changing environment. Persistent neuropathic pain characterized by chronic shooting pain and burning sensations, however, is usually caused by injury or diseases such as cancer or AIDS (or even AIDS medications) and has no physiological benefit (15). Unfortunately, many conventional analgesics are not selective and affect both physiological and pathological pain. Treatment of pathological pain requires selective drug targets and compounds that preferentially inhibit pathological pain without or with minimal effect on protective physiological pain (3, 4, 6, 7).

Recent molecular studies of different levels of the central nervous system (CNS) have underscored the fact that the mechanisms leading to chronic neuropathic pain are far more complicated than previously assumed (2, 4, 811). At peripheral sites, injuries trigger sensitization and induce prolonged abnormal neural activity along primary afferent fibers. In the spinal cord, long-term potentiation (LTP) of sensory synaptic transmission in the spinal dorsal horn projecting neurons often occurs (10). Recent work demonstrates that rapid synaptic plasticity also takes place in cortical areas that participate in pain perception, such as the anterior cingulate cortex (ACC) (4, 12, 13). Peripheral inflammation or nerve injury produces LTP of glutamate release and glutamate receptor–mediated synaptic responses (1416). Protein and ion channels that are involved in the induction and expression of plastic changes in somatosensory pathways are potential drug targets for treating chronic pain (1719). Targeting the signaling mechanisms involved in injury-related plasticity may be particularly helpful in discovering more effective drug treatments for neuropathic pain. Among several candidates, the N-methyl-d-aspartate (NMDA) receptor, a major glutamate receptor that is important for synaptic LTP, has been intensively investigated as an analgesic target by academic laboratories and pharmaceutical companies. However, the use of a general NMDA receptor antagonist is limited by side effects such as ataxia and sedation (20). Still, signaling proteins downstream from the NMDA receptor remain viable targets for neuropathic pain treatment. Our previous studies have found that one of these targets, calcium-stimulated adenylyl cyclase 1 (AC1), plays a critical role in pain-related LTP in both the spinal cord dorsal horn (21) and the ACC (22). Behaviorally, AC1 and AC8 knockout mice show reduced inflammatory, deep muscle pain and neuropathic pain (15, 23, 24), whereas other physiological functions remain intact in AC1 knockout mice, including acute pain (24), hippocampal LTP, and related Morris water maze performance, as well as anxiety-like behaviors and motor functions (25). Considering that AC1 is mainly expressed in the CNS (2628), we propose that AC1 may be a suitable neuron-specific drug target for treating neuropathic pain.

Here, we carried out integrative chemical screening and biological experiments to identify selective inhibitors for AC1 and examined the potential analgesic effects of the AC1 inhibitor in neuropathic and inflammatory pain models. By chemical screening, we have successfully identified a lead AC1 inhibitor, NB001, and explored its effects in vitro and in vivo.

Results

General consideration of AC1 as a therapeutic pain target

One major challenge to developing a clinically useful analgesic drug is to avoid side effects in non-neuronal tissues. Many clinically proven effective drugs have been removed from the market as a result of their side effects in non-neuronal organs such as the cardiovascular system and liver. Ways to avoid side effects include (i) identifying drug targets that are mainly expressed in neurons; (ii) choosing target proteins or channels that are activity-dependent, with an almost “silent” status in normal physiological status; and (iii) choosing target proteins that are critical for chronic pain–related neuronal plasticity, but not for other cognitive functions (4). Few target proteins meet all three criteria. On the basis of previous studies and our recent studies with gene knockout mice, we proposed that AC1 would be a good target for treating chronic pain. First, AC1 is primarily expressed in neurons, and no AC1 gene expression was found in heart, liver, or kidney cells (Table 1). Second, AC1 is activated in a calcium-calmodulin (CaM)–dependent manner (24) (Fig. 1A). Third, it acts downstream from the glutamate NMDA receptors and contributes to chronic pain–related neuronal plasticity in the cortex and spinal cord (3, 4, 15). Finally, gene knockout mice lacking AC1 showed reduced or blocked behavioral sensitization to non-noxious mechanical stimuli in animal models of inflammatory and neuropathic pain.

Table 1

Distribution patterns of target proteins for treatment of neuropathic pain. Blank, no data found; +++, strongly expressed or very effluent; +, less effluent; 0, no expression. The distribution patterns of some proteins are based on in situ hybridization data. DH, dorsal horn; DRG, dorsal root ganglion; GABAR, GABA receptor; NMDAR, NMDA receptor; COX-2, cyclooxygenase 2; NOS, nitric oxide synthase; 5-HT, 5-hydroxytryptamine (serotonin); NA, noradrenaline; VR1, vanilloid receptor 1.

View this table:
Fig. 1

AC1 stably expressed in HEK293 cells. (A) A model shows that AC1 acts downstream from the glutamate NMDA receptors (NMDAR) and is activated in a calcium-dependent manner. (B) Upper panel: AC1 expression was verified by RT-PCR. 1, 100–base pair (bp) DNA ladder marker; 2, HEK293 cells transfected with AC1 plasmid; 3, HEK293 cell control; 4, mouse brain cortex sample. The size of PCR products is 384 bp. Lower panel: stimulatory effect of forskolin and calcium ionophore A23187 in HEK293 cells stably transfected with AC1. HEK293 cells were stimulated by forskolin (10 μM), A23187 (10 μM), or forskolin (10 μM) plus A23187 (10 μM) for 30 min before cyclic AMP was measured. n = 4 wells. (C) Inhibitory effects of NB001 on AC1-expressing HEK293 cells. n = 4 wells. (D) Effect of NB001 on AC5, 6, 7, or 8 activities. HEK293 cells were pretreated with NB001 for 15 min and then stimulated by 10 μM forskolin (AC5, AC6, and AC7) or 10 μM forskolin plus 10 μM A23187 (AC1 and AC8) for 30 min in the presence of NB001. n = 4 wells. AC activity was detected by cyclic AMP assay. (E) Chemical structure of NB001.

Biochemical screening and identification of an AC1 inhibitor

To our knowledge, there is no selective inhibitor for AC1. Thus, we decided to first perform chemical screening experiments to identify AC1 inhibitors, using a heterologous expression system, in which AC1 was stably expressed in human embryonic kidney (HEK) 293 cells, which do not express AC1 (29, 30). HEK293 cells were stably transfected with the AC1 expression vectors, and AC1 expression was confirmed by reverse transcription polymerase chain reaction (RT-PCR). AC1 expression was not observed in the control cells (Fig. 1B). We then tested the stimulatory effects of forskolin and the calcium ionophore A23187 in these cells. We found that co-application of forskolin and A23187 induced an increase in cyclic AMP (adenosine 3′,5′-monophosphate) in AC1-expressing cells, but caused a significantly smaller effect in control HEK293 cells, indicating little or no basal AC activity (Fig. 1B). On the basis of the chemical structures of forskolin and adenosine 5′-triphosphate (ATP), we screened a family of compounds (NB001 to NB016) for potential AC1 inhibition. We tested their effects on forskolin- and A23187-stimulated cyclic AMP concentrations in HEK293 cells stably expressing AC1. At a concentration of 100 μM, we found that NB001 to NB007 significantly reduced cyclic AMP concentrations compared to other compounds, with inhibition of 69.5 ± 0.2 to 96.9 ± 0.1% (n = 4 wells) (table S1). Among them, NB001 (C12H20N6O) (Fig. 1E) was the most effective inhibitor, with inhibition of 96.9 ± 0.1% at 100 μM (n = 4 wells) (table S1).

Additionally, we measured the transcription factor cyclic AMP response element–binding (CREB) protein activity, using a dual luciferase reporter system. In this assay, we measured changes in the amount of firefly luciferase, the transcription of which is regulated by CREB binding to an upstream cyclic AMP response element (CRE) (31). We measured the effect of NB001 to NB007 on CREB activity in AC1-expressing HEK293 cells stimulated by forskolin and A23187. Consistent with the cyclic AMP assay data, NB001 to NB007 inhibited CREB activity at a concentration of 100 μM (table S1). NB001 was again the strongest inhibitor among the compounds (n = 4 wells) (table S1).

Selectivity of NB001

Like AC1, AC8 is also stimulated by CaM, although it is less sensitive to calcium (32, 33). We have performed experiments similar to those with AC1 by expressing AC8 in HEK293 cells. The inhibitory effects of NB001 were tested at different concentrations (between 0.2 and 200 μM) in AC1- or AC8-expressing HEK293 cells after the application of forskolin and A23187. By cyclic AMP assay, we found that NB001 inhibited cyclic AMP production in a concentration-dependent manner, with an estimated median inhibitory concentration (IC50) of 10 μM in AC1-expressing HEK293 cells (Fig. 1C). However, in AC8-expressing HEK293 cells, NB001 produced significantly less inhibition, with an estimated IC50 of 139 μM (Fig. 1D). We also tested the inhibitory effect of NB001 on other ACs including AC5, AC6, and AC7, which are mostly coupled to G protein (heterotrimeric guanosine 5′-triphosphate–binding protein)–coupled receptors. The effects of NB001 at various concentrations between 0.2 and 200 μM were evaluated in AC5-, AC6-, or AC7-expressing HEK293 cells. The estimated IC50s of NB001 for AC5, AC6, and AC7 were 210, 167, and 191 μM, respectively (Fig. 1D). These results indicate that NB001 is relatively selective for AC1.

Effects of NB001 on cyclic AMP production in adult mouse brain slices

To investigate the effect of the AC1 inhibitor NB001 in brain tissues, we studied the effects of NB001 using brain slices of the ACC from adult mice. We first characterized the stimulatory effects of forskolin and calcium ionophore A23187 in ACC slices. We found that co-application of forskolin (10 μM) and A23187 (10 μM) induced significant increases in the cyclic AMP concentration, whereas application of either forskolin (10 μM) or A23187 (10 μM) alone caused less cyclic AMP production (Fig. 2A). We then tested the effect of NB001 on cyclic AMP production in ACC slices stimulated by forskolin and A23187. Application of NB001 (0.1 to 200 μM) 15 min before and during the treatment with forskolin and A23187 (10 μM for each) for 30 min inhibited cyclic AMP production in a dose-dependent manner (Fig. 2B). However, the inhibitory effect of NB001 was limited, even at a high concentration (100 or 200 μM). One possible explanation is that other isoforms of ACs were also activated in ACC slices, and NB001 is ineffective for other isoforms of ACs in ACC neurons.

Fig. 2

Effect of NB001 in mouse ACC slices and human neural cell line. (A) Cyclic AMP production in ACC slices stimulated by forskolin (10 μM), A23187 (10 μM), or forskolin plus A23187. Cyclic AMP was measured at 30 min after treatments. n = 4 mice. (B) Effect of AC1 inhibitor NB001 (0.1 to 200 μM) on cyclic AMP production stimulated by forskolin (10 μM) and A23187 (10 μM) in ACC slices. NB001 was applied 15 min before and during the treatment with forskolin and A23187 for 30 min. n = 4 mice. (C) Cyclic AMP production in ACC slices stimulated by glutamate (0.25 to 1 mM) for 30 min. n = 4 mice. (D) Effect of AC1 inhibitor NB001 (0.1 to 200 μM) on cyclic AMP production stimulated by glutamate (0.5 mM). NB001 was applied 15 min before and during the treatment with glutamate for 30 min. n = 4 mice. (E) Cyclic AMP production in SH-SY5Y cells stimulated by glutamate (0.25 to 1 mM) for 30 min. n = 4 wells. (F) Application of AC1 inhibitor NB001 (1 to 200 μM) 15 min before and during the treatment with glutamate (0.5 mM) for 30 min could inhibit cyclic AMP production stimulated by glutamate in a dose-dependent manner. n = 4 wells.

The activation of NMDA receptors by glutamate can elicit a calcium-dependent increase in cyclic AMP in the hippocampal CA1 region (34, 35). Our previous study has shown that AC1, but not AC8, contributes to the elevation of cyclic AMP elicited by NMDA receptor activation in cortical neurons (32). To investigate the effect of NB001 at a more physiological condition, we tested the stimulatory effects of glutamate in mouse ACC slices. We found that application of glutamate (0.25 to 1 mM) could induce significant cyclic AMP increases in mouse ACC slices (Fig. 2C). Application of NB001 (0.1 to 200 μM) 15 min before and during the treatment with glutamate (0.5 mM) for 30 min significantly inhibited the cyclic AMP production stimulated by glutamate in a dose-dependent manner; the effect of glutamate was completely blocked by NB001 at the concentration of 100 or 200 μM, and the estimated IC50 is 5.1 μM (Fig. 2D). These results indicate that NB001 can inhibit calcium-stimulated AC1 during glutamate treatment in adult mouse ACC neurons, which suggests that it might be effective at physiological or pathological conditions in adults.

Effect of NB001 in human neural cells

To investigate whether NB001 is effective in mammalian neurons, we tested the effect of NB001 in the human neuroblastoma SH-SY5Y cells, which have been shown to express AC1 (36). In addition, SH-SY5Y cells express glutamate receptors and respond to glutamate or NMDA stimulation (37, 38). Indeed, we found that bath application of glutamate (0.25 to 1 mM) could induce significant increases at cyclic AMP levels in SH-SY5Y cells (Fig. 2E). Application of NB001 (0.1 to 200 μM) 15 min before and during the treatment with glutamate (0.5 mM) for 30 min inhibited the cyclic AMP production in a dose-dependent manner; the effect of glutamate was completely blocked by NB001 at a concentration of 100 or 200 μM, with an estimated IC50 of 8.3 μM (Fig. 2F). These results indicate that NB001 can inhibit calcium-stimulated AC during glutamate treatment in SH-SY5Y cells, suggesting that it might be effective in human neurons at physiological conditions.

Analgesic effects of NB001 in animal models of neuropathic and inflammatory pain

Previous studies with genetic knockout mice lacking AC1 showed that behavioral allodynia (pain experienced to a usually innocuous stimulus) in animal models of neuropathic pain and inflammatory pain was significantly reduced (24). However, it is difficult to rule out the possible contribution of developmental defects in AC1 knockout mice that may contribute to behavioral results. With the identification of NB001 as an inhibitor for AC1, we examined the effects of NB001 on behavioral allodynia in animal models of neuropathic pain induced by nerve ligation (39). Similar to our previous report (15, 39), we found that behavioral allodynia reached its peak at day 7 after the nerve injury. Intraperitoneal administration of NB001 (0.1 mg/kg) given 30 min before behavioral allodynia testing produced a significant analgesic effect. NB001 at a higher dose of 1 mg/kg produced a greater inhibition of behavioral allodynia (Fig. 3A).

Fig. 3

Effect of NB001 on acute pain and neuropathic pain. (A) Effect of NB001 on allodynia after nerve ligation. Intraperitoneal administration of NB001 (0.1 mg/kg body weight) given 30 min before behavioral allodynia testing produced a significant analgesic effect. NB001 at a high dose of 1 mg/kg produced a greater inhibition of behavioral allodynia. Intraperitoneal injection of gabapentin (30 mg/kg; n = 4 mice) produced significant analgesic effects in the animals with neuropathic pain. The inhibitory effects were comparable to those produced by NB001 at 0.1 mg/kg. (B) Effect of NB001 on inflammatory pain. At 1 day after the CFA injection, NB001 (1 and 3 mg/kg ip) produced significant analgesic effects. The inhibitory effect is dose-related; greater inhibition was found with a higher dose of NB001 (n = 5 mice). Similar analgesic effects were also obtained at 3 days after CFA injection. (C) Comparison of behavioral responses to non-noxious mechanical stimulus. There was no significant difference in hindpaw withdrawal to von Frey filaments before and after intraperitoneal injection of saline (n = 5) and NB001 (n = 6). (D) Effect of NB001 in tail-flick (TF) test. NB001 (10 mg/kg ip) did not affect spinal nociceptive TF reflex. (E) Effect of NB001 in hot plate test at 50°C. There was no significant difference in response latency before and after intraperitoneal injection of saline (n = 5) and NB001 (10 mg/kg) (n = 6) in hot plate test set at 50°C. (F) Effect of NB001 in hot plate test at 55°C. There was no significant difference in response latency before and after intraperitoneal injection of saline (n = 5) and NB001 (10 mg/kg) (n = 6) in hot-plate test set at 55°C. (G) Oral application of NB001 (1 mg/kg, 3 ml per injection) produced significant analgesic effects in adult rats with neuropathic pain (n = 5). Allodynia was measured at 45 min, 2 hours, and 4 hours after oral application. *P < 0.05.

To address the possible central, spinal, or peripheral action of NB001, we performed site-directed local administration of NB001 in mice with nerve injury. We found that multiple microinjections of NB001 into the dorsal and ventral ACC on both sides of the brain (a total of 75.6 nmol, 0.5 μl for each site) produced a significant inhibitory effect on behavioral allodynia (n = 5 mice, P < 0.05) (fig. S1A), whereas NB001 (189.0 nmol, 5.0 μl) intrathecally applied to the spinal cord or subcutaneous injections of NB001 (378.0 nmol, 10 μl) did not produce any significant inhibitory effect on nerve injury–induced mechanical allodynia. These results suggest that NB001 is likely acting centrally at supraspinal sites to produce analgesic effects (fig. S1, B and C). The inhibitory effects on mechanical allodynia produced by NB001 applied locally in the ACC (even with multiple injection sites) were smaller than that applied systemically, suggesting that NB001 may also act at other cortical and subcortical sites.

In general, application of NB001 at different dosages did not cause any abnormal behaviors in animals. Animals injected with NB001 were calm and less responsive to behavioral allodynic measurement than were control animals. To compare our results with NB001 with the analgesic effect of currently available drugs for neuropathic pain, we also performed experiments with gabapentin. Intraperitoneal injection of gabapentin (30 mg/kg, n = 4 mice) produced significant analgesic effects in the animals with neuropathic pain. The inhibitory effects were comparable to those produced by NB001 at 0.1 mg/kg (Fig. 3A).

Behavioral allodynia induced by the hindpaw injection of complete Freund’s adjuvant (CFA) (50%) has been commonly used for evaluating a drug’s analgesic effects in chronic inflammatory pain. In previous studies, we reported that AC1 knockout mice showed significant reduction in behavioral allodynic responses induced by hindpaw CFA injection (24). Furthermore, peripheral neurogenic plasma extravasation induced by capsaicin injection was the same in AC1 knockout mice and in littermate wild-type mice (24). Therefore, we expect that NB001 would be analgesic in the animal model of inflammatory pain induced by CFA injection. Application of a 0.4-mN von Frey fiber to the dorsum of a hindpaw elicited no response in control mice, but at 1 and 3 days after CFA injection, mice withdrew their hindpaw in response to stimulation of the ipsilateral or, to a lesser extent, the contralateral hindpaw. One day after the CFA injection, NB001 [1 and 3 mg/kg intraperitoneally (ip)] produced significant analgesic effects. The inhibitory effect was dose-related: Greater inhibition was found with a higher dose of NB001 (n = 5 mice for each group) (Fig. 3B). Similar analgesic effects were also obtained 3 days after CFA injection (Fig. 3B).

Effect of NB001 on nociceptive responses

Next, we carried out behavioral tests to determine whether NB001 affects physiological nociceptive responses in normal mice. We tested the effect of intraperitoneally injected NB001 in different nociceptive tests. For noxious thermal pain, NB001 affected behavior in neither the spinal nociceptive tail-flick reflex nor the hot-plate test (at 55°C) (Fig. 3, D to F). To determine potential effects at a lower temperature, we also tested the animals on a hot plate set at 50°C and found that behavioral response latencies were not affected (Fig. 3E). For mechanical threshold, we measured hindpaw withdrawal responses to von Frey filaments in normal mice. Again, NB001 did not produce any significant effect (Fig. 3C). These results suggest that NB001 did not affect acute nociception, in good accord with previous findings of unchanged physiological pain in AC1 knockout mice (24).

Effect of NB001 administered orally in a rat model of neuropathic pain

We next examined the effects of NB001 in a rat model of neuropathic pain (see Materials and Methods) to see whether orally delivered NB001 can produce analgesic effects, a common method of drug delivery in patients. As shown in Fig. 3G, we found that NB001 at 1 mg/kg (orally) produced a significant reduction in behavioral allodynia 45 min after oral ingestion, and that the analgesic effect lasted for at least 2 hours (n = 6 rats). The analgesic effect of NB001 is reversible, because behavioral allodynia returned 4 hours after the oral application, suggesting that the effects of NB001 are unlikely to be toxic or to induce nonselective side effects.

Effect of NB001 on anxiety, motor function, and fear memory

To test the potential side effects of NB001, we carried out a battery of behavioral anxiety and motor function tests with the same analgesic dose of NB001. First, in two tests of anxiety-like behavior, elevated plus maze (EPM) and open-field test (40, 41), we found that NB001 did not produce any significant effects (Fig. 4, A to D). The NB001-injected mice did not make a significantly different number of entrances into the arms of the EPM, nor did they spend more time in the open arms when compared to the saline-injected mice (Fig. 4, A and B). Also, the distance traveled in the center of the open-field test did not differ between NB001- and saline-injected mice (Fig. 4C), again demonstrating that NB001 had no effect on anxiety-like behavior. We also tested the motor function of the mice injected with NB001. In the open-field test, we found that NB001 only produced a significant change during the first 5 min after the injection, and there were no differences in the total distance traveled between NB001- and saline-injected mice (Fig. 4, E and F). In the RotaRod motor test, NB001 did not cause any significant changes (Fig. 4G), illustrating that NB001 had no effect on the motor function of the mice.

Fig. 4

Effect of NB001 on anxiety-like behavior, locomotor activity, and motor function. (A) There was no significant difference in the number of arm entries (open + closed) between saline-injected (n = 5) and NB001-injected (n = 7) mice. (B) There was no significant difference in the percentage of time spent in the open arms between saline- and NB001-injected mice. (C) From left to right: diagram of the EPM, where filled boxes represent closed arms and open boxes represent open arms; representative traces showing the movement of saline- and NB001-injected mice in the EPM for 5 min. (D) Representative traces showing the movement of saline-injected (n = 6) and NB001-injected (n = 6) mice in the open field for 30 min. (E) NB001-injected mice traveled significantly more within the first 5 min during the 30-min session than did saline-injected mice. *P < 0.001. (F) There was no significant difference in the total distance traveled between saline- and NB001-injected mice. (G) There was no significant difference in latency to fall between saline-injected (n = 5) and NB001-injected (n = 6) mice. In each experiment, mice were given an intraperitoneal injection of NB001 (10 mg/kg) or saline 30 min before the test.

Cyclic AMP is a key second messenger in the CNS, and calcium-stimulated cyclic AMP signaling pathways have been suggested to contribute to learning and memory (42, 43). However, in previous studies of AC1 knockout mice, several forms of learning and memory were not affected in AC1 knockout mice, including those measured by classic fear memory tests and the Morris water maze (25). One explanation for this lack of effect may be that the function of AC1 may be compensated for by other AC isoforms such as AC8 and other calcium/calmodulin-dependent protein kinases, including CaMKII and CaMKIV. We tested whether NB001 affects fear memory induced by noxious foot shock. Consistent with previous genetic studies (25), we found that pretreatment with NB001 (1 mg/kg ip) did not significantly affect fear memory (n = 4 mice). Furthermore, we also injected NB001 1 day after the fear conditioning and found that fear memory was not significantly affected (n = 4 mice), indicating that NB001 at the analgesic dosage does not affect the formation and expression of fear memory.

Effect of NB001 on glutamate receptor–mediated currents

Glutamate AMPA receptors mediate most of the basal synaptic transmission in the spinal dorsal horn and the ACC (44, 45), whereas NMDA receptors are important for producing synaptic potentiation (4, 46). It is thus important to determine whether NB001 produces analgesic effects by affecting these glutamate receptor–mediated responses in sensory neurons. We performed whole-cell patch-clamp recordings from ACC pyramidal cells, a key cortical area for processing pain information in animals and humans (4). We first measured the effects of NB001 on AMPA receptor–mediated responses. AMPA receptor–mediated excitatory postsynaptic currents (EPSCs) were recorded from cingulate pyramidal cells before and after the application of NB001. We did not see any significant change in AMPA receptor–mediated EPSCs after the perfusion with NB001 (50 μM) (n = 8 neurons, 6 mice) (Fig. 5A). Paired-pulse facilitation (PPF), a simple form of synaptic plasticity, was also measured by application of paired stimuli at different intervals. We found that PPF tested at different interpulse intervals also did not show any significant change [PPF in the presence of NB001, n = 7 neurons, 2 mice; control PPF in the absence of NB001, n = 17 neurons, 7 mice; two-way analysis of variance (ANOVA), P > 0.05] (Fig. 5B).

Fig. 5

Effect of NB001 on AMPA receptor–mediated synaptic responses. (A) Bath application of NB001 (50 μM) did not affect AMPA receptor–mediated basal synaptic responses in the ACC neurons (n = 8). Inset: representative traces and pool data of AMPA receptor–mediated EPSCs in ACC neurons with the perfusion of NB001 (50 μM) for 10 min. (B) PPF was recorded at intervals of 35, 50, 75, 100, and 150 ms. Open circles: neurons in the absence of NB001 (n = 17 neurons); filled circles: neurons in the presence of NB001 (100 μM) applied in the pipette (n = 7 neurons). Representative traces of PPF with an interval of 50 ms recorded in the ACC are shown in the insets.

NMDA receptors are important for pain-related plasticity in the spinal cord and cortex (2, 4, 10), and NMDA receptor antagonists are analgesic in different animal models of chronic pain (47). To rule out the possibility that NB001 may produce analgesic effects through inhibition of NMDA receptor–mediated responses, we performed two different types of experiments. We first evaluated the effect of NB001 on the input-output curves of NMDA receptor–mediated EPSCs. NB001 (50 μM) application did not produce any reduction of NMDA receptor–mediated input-output curves (Fig. 6A). Next, we also tested whether NB001 affected the I-V (current-voltage) curve of NMDA receptor–mediated EPSCs. Bath application of NB001 did not significantly affect the functions of NMDA receptor–mediated I-V curves (Fig. 6B). Voltage dependence of NMDA receptor–mediated EPSCs evoked by focal electrical stimulation was not affected by the treatment with NB001 (50 μM). These findings indicate that NB001 does not produce behavioral analgesic effects by inhibiting NMDA receptors. Finally, we also examined whether NB001 affects AMPA and NMDA receptor–mediated responses. We tested the effects of NB001 on the AMPA and NMDA receptor–mediated EPSC ratio. We found that NB001 (50 μM) did not affect the AMPA/NMDA receptor–mediated EPSC ratio in ACC neurons (ratio in the absence of NB001: 1.8 ± 0.3, n = 7 neurons, 5 mice; ratio in the presence of NB001: 2.5 ± 0.6, n = 7 neurons, 5 mice; P > 0.05).

Fig. 6

Effect of NB001 on NMDA receptor–mediated responses. (A) NB001 application did not affect NMDA receptor–mediated EPSCs recorded from ACC pyramidal cells. Representative traces and pooled data of synaptic NMDA receptor–mediated input-output curve in ACC neurons in the absence (open circles; n = 6 neurons, 3 mice) and presence (filled circles; n = 6 neurons, 2 mice) of NB001 (100 μM) in the pipette. (B) NB001 application did not affect the I-V response curves of NMDA receptor–mediated responses. Representative traces and pooled data of synaptic NMDA receptor–mediated I-V curve in ACC neurons in the absence (open circles; n = 6 neurons, 3 mice) and presence (filled circles; n = 6 neurons, 2 mice) of NB001 (100 μM) in the pipette.

Effect of NB001 on sensory LTP in the spinal dorsal horn and ACC

LTP of excitatory synaptic transmission in the spinal dorsal horn and ACC are proposed to be cellular mechanisms for chronic pain (4, 10). Previous studies with AC1 knockout mice found that AC1 is required for synaptic LTP in the dorsal horn neurons and ACC (21, 22). Although NMDA receptor–mediated responses are intact in AC1 knockout mice, it is difficult to completely exclude potential developmental defects in AC1 knockout mice that may contribute to diminished LTP. Thus, we tested whether NB001 inhibits sensory-related LTP in these two areas. In mouse ACC slices, LTP in ACC pyramidal neurons was not significantly different in the presence of NB001 (10 μM) in the pipette (control LTP: 164.6 ± 11.2%, n = 15 neurons, 8 mice; LTP in the presence of NB001, 154.8 ± 16.8, n = 6 neurons, 4 mice; P > 0.05). The application of NB001 (50 and 100 μM) in the pipette completely prevented the induction of LTP in ACC pyramidal neurons (control LTP: 164.6 ± 11.2%, n = 15 neurons, 8 mice; LTP in the presence of 50 μM NB001: 117.4 ± 9.1, n = 8 neurons, 3 mice, P < 0.05; LTP in the presence of 100 μM NB001, 111.8 ± 8.8%, n = 7 neurons, 3 mice, P < 0.01) (Fig. 7A). The spinal dorsal horn is also critical for pain transmission. Previous studies reported that activation of NMDA and substance P receptors are critical for spinal LTP (48). AC1 is also expressed in spinal dorsal horn neurons; LTP induced by a pairing protocol is abolished in AC1 knockout mice (21). To confirm the requirement of AC1 activity for spinal LTP, we tested the effects of NB001 on spinal LTP induced by the pairing protocol. We found that synaptic potentiation was significantly reduced (control: 157.8 ± 10.0%, n = 8 neurons, 3 mice; LTP in the presence of 100 μM NB001: 95.3 ± 10.1%, n = 6 neurons, 5 mice; P < 0.05) (Fig. 7B). These results provide strong evidence for critical roles of AC1 in both spinal and ACC LTP, one possible cellular mechanism contributing to neuropathic pain.

Fig. 7

Effect of blockade of ACC LTP and spinal LTP by postsynaptic application of NB001. (A) Representative traces and pooled data of LTP induced by pairing protocol in ACC neurons in the absence (open squares; n = 15 neurons, 8 mice) and presence (filled squares; n = 7 neurons, 3 mice) of NB001 (100 μM) in the pipette. (B) Pooled data of LTP induced by pairing protocol in spinal lamina II neurons in the absence (open squares; n = 8 neurons, 3 mice) and presence (filled squares; n = 6 neurons, 5 mice) of NB001 (100 μM) in the pipette. (C) Pooled data of two-train tetanic stimulation–induced LTP in the CA1 region of the hippocampus in control slices (open squares; n = 5 slices) and in slices with the perfusion of 50 μM NB001 (filled squares; n = 5 slices).

Effect of NB001 on memory-related hippocampal LTP

One potential side effect of NB001 is alteration of hippocampal LTP, which is important for the formation of new spatial memories. To examine the effects of NB001 on LTP in the CA1 region of the hippocampus, we tested LTP with a bath perfusion of 50 μM NB001. LTP was induced in the CA1 region of mouse hippocampal slices by strong tetanic stimulation (two trains, 100 Hz for 1 s per train, delivered at a 20-s interval) (49, 50). We found that the amount of potentiation 45 min after tetanic stimulation with NB001 perfusion was similar to the potentiation in slices of control animals (control: n = 5 slices, 169.5 ± 12.7%; NB001: n = 5 slices, 163.9 ± 20.7%; comparing 40 to 45 min after tetanic stimulation to before the stimulation; P = 0.83) (Fig. 7C), indicating that NB001 did not affect hippocampal LTP.

Discussion

The present study is built on two previous discoveries. First, peripheral injuries trigger plastic changes in cortical areas of the brain (including the ACC), induce an array of activity-dependent immediate-early genes, and enhance synaptic responses (3, 5, 12, 24, 51). In a recent study, we elucidated the synaptic mechanism for this neuronal plasticity after nerve injury in the ACC, namely, presynaptic enhancement of glutamate release and postsynaptic enhancement of AMPA receptor–mediated responses (15). Second, AC1 and AC8, two major forms of calcium-stimulated ACs, are critical for chronic pain, including neuropathic pain (4, 23, 24). Here, we have identified a candidate molecule that may be useful as a selective AC1 inhibitor. Biochemical and behavioral studies indicate that NB001 could prove to be a useful analgesic drug for neuropathic pain.

A previous study has shown that AC1 messenger RNA is highly expressed in pain-related cortical areas such as the ACC and insular cortex, and that AC1 activity contributes to behavioral nociceptive responses to nerve injury and inflammation (24). In mice lacking AC1, chronic pain, including neuropathic pain, is significantly reduced, whereas acute pain to noxious thermal or mechanism stimuli remains intact (24). These results suggested that AC1 is selectively involved in persistent, chronic pain. In neurons, it has been proposed that AC1 couples NMDA receptor–induced cytosolic Ca2+ elevation to cyclic AMP signaling pathways (25, 32, 35). Because activation of NMDA receptors contributes to persistent, chronic pain (35, 7, 51), it is likely that AC1 acts downstream from the NMDA receptors and thus contributes to chronic pain.

Systemic administration of NB001 (0.1 to 1 mg/kg), whether administered intraperitoneally or orally, inhibited mechanical allodynia. Application of gabapentin produced a similar magnitude of analgesia at a much larger dose (higher by a factor of >50 to 100). One possible reason for this difference is that NB001 targets pain-related synaptic plasticity, whereas gabapentin affects neurotransmitter release in a nonselective manner. When we tested whether microinjection of NB001 directly into the ACC could reproduce the effects of systemic NB001 administration, we found that the AC1 inhibitor did not produce complete blockade of behavioral allodynia. Furthermore, the analgesic effects were smaller than that produced by NB001 administered systemically. Several possible factors may explain this difference. First, microinjection of NB001 into the ACC may not have completely blocked AC1 activity within the ACC, although we performed multiple injections of NB001 to inhibit AC1 activity in both the dorsal and the ventral parts of the ACC. Second, AC1 activity in other cortical structures, such as the prefrontal cortex, insular cortex, and somatosensory cortices, may also be important for chronic pain (4). Finally, we also cannot rule out the possible contribution of subcortical structures to the effects produced by AC1 on neuropathic pain. Thus, it is likely that the strong analgesic effect produced by NB001 applied systemically is due to inhibition of chronic pain–related plasticity within cortical and subcortical neural networks.

Recently, we found that both presynaptic enhancement and postsynaptic changes in AMPA receptor–mediated responses were blocked in AC1 knockout mice, indicating that AC1 may contribute both presynaptically and postsynaptically to injury-induced potentiation in the ACC. In fact, stimulation of AC1 is critical for mossy fiber LTP (52, 53). Because basic properties of synapses, including paired-pulse facilitation, AMPA receptor–mediated responses, and NMDA receptor–mediated responses, are not affected in AC1 knockout mice, we believe that the involvement of AC1 is an activity-dependent rather than a nonselective side effect. In addition to cortical plasticity, AC1 is also likely involved in spinal cord plasticity. In previous studies, we have shown that genetic deletion of AC1 inhibits spinal cord LTP induced by the pairing training protocol and that synaptic facilitation is induced by co-application of forskolin and serotonin (21, 54). Considering the increasing evidence that spinal LTP and descending facilitation contribute to chronic pain, it is conceivable that AC1 in spinal cells may also contribute to chronic pain.

In addition to its neuron-specific distribution, AC1 has been proposed to be an intracellular coincidence detector acting downstream from NMDA receptors. In many neurons, AC1 is activated by increases in postsynaptic calcium levels. This provides an ideal target for treatment of chronic pain (23, 54). Unlike a single learning event, injury is likely to trigger a series of plastic changes in the CNS. The selective effects of NB001 on neuropathic, but not acute, pain support the hypothesis that LTP can be used as a useful cellular model for persistent pain in sensory synapses. Unlike other drug targets that have wide distribution in non-neuronal tissue, AC1 is mainly expressed in neurons (2628). Although one potential side effect of NB001 is a deficit in cognitive functions, AC1 gene knockout mice did not show any marked memory defects (25). AC1 knockout mice showed normal hippocampal late-phase LTP and long-term memory (25). Here, we have shown that AC1 inhibitor NB001 did not affect anxiety-like behavior or motor functions. It is possible that cognitive functions, including learning and memory, may be compensated for by other isoforms of ACs, such as AC8, and other key signaling proteins, including CaMKII and CaMKIV. Further experiments are needed to clarify the potential side effects of NB001 before it can be tested in patients with neuropathic pain.

We have identified NB001 as a relatively selective inhibitor of AC1, as demonstrated by in vitro biochemical assays. Furthermore, NB001 showed inhibitory effects similar to those seen in AC1 knockout mice. Neither genetic deletion of AC1 nor NB001 affected acute pain responses to noxious thermal and mechanical stimuli, but both selectively reduced neuropathic pain. Genetic deletion of AC1 and NB001 produced similar inhibitory effects on ACC LTP, and AC1 may also contribute to pain-related synaptic plasticity in other cortical areas. NB001 is effective in reducing neuropathic pain even after nerve injury. One possible explanation may be that the central mechanism of chronic pain is a multiple-stage learning process. After the initial injury, only some synapses may undergo potentiation, and many of these potentiated events may be reversible as a result of long-term depression (LTD). For chronic pain to develop, continuous ongoing LTP in other synapses is required; therefore, preventing LTP (through inhibition of AC1 activity) by application of NB001 can reduce allodynia seen in neuropathic pain models even after nerve injury. Allodynia is therefore likely a consequence of a long-lasting neuronal circuit abnormality, and it is unlikely to be explained by a simple form of LTP in an individual synapse. This ongoing plasticity may also explain why AC1 is preferentially involved in chronic pain. The lack of an effect of the AC1 inhibitor on memory may be due to compensation by other key cognitive protein kinases such as PKC (protein kinase C), CaMKII, and CaMKIV. Searching for target proteins with a cellular model such as ACC LTP may be useful for identifying drugs that interfere with or inhibit such network activity. Future studies to design safe compounds with greater efficacy and selectivity on AC1 will improve prospects for clinical treatment of chronic pain in patients.

We also found that NB001 is effective in inhibiting behavioral allodynia in chronic inflammatory pain. The inhibitory effects of NB001 in inflammatory pain are weaker than that in neuropathic pain, suggesting that different mechanisms may be involved. Among possible differences is a greater component of AC1-independent peripheral sensitization that contributes to behavioral allodynia in inflammatory pain than in neuropathic pain. This notion is also supported by the ineffectiveness of NB001 when applied locally to the skin (fig. S1). In summary, our study demonstrates that calcium-stimulated AC1 is critical for nerve injury–triggered synaptic changes in the brain and that the AC1 inhibitor NB001 produced substantial analgesic effects in animals with neuropathic pain. Because of the selective effect of NB001 on spinal and cortical LTP, we argue that NB001 produces its analgesic effect by affecting, at least in part, synaptic LTP in pain-related areas.

Materials and Methods

Cell culture

HEK293 cells [American Type Culture Collection (ATCC)] were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum. Human neuroblastoma SH-SY5Y cells (ATCC) were grown in DMEM supplemented with 10% fetal bovine serum. The cell cultures were maintained at 37°C in a humidified (95% air/5% CO2) incubator.

Expression of ACs in HEK293 cells

For AC1 expression vector pcDNA3-AC1 transfection, HEK293 cells were plated onto 60-mm-diameter dishes at a density of 1 × 106 per plate, grown overnight, and transfected with pcDNA3-AC1 (0.8 μg of DNA per plate) by Lipofectamine 2000 (Invitrogen). Stable transfected clones were selected in culture medium containing G418 (0.8 mg/ml; Invitrogen) and maintained in this medium. For transient expression of other AC isoforms in HEK293 cells, HEK293 cells were plated in 96-well tissue culture dishes and transfected with plasmids for AC5, 6, 7, and 8, respectively; experiments were carried out 48 hours after transfection.

RT-PCR

Total RNA was isolated from HEK293 cells with RNeasy Mini Kit (Qiagen Inc.). RT-PCR was performed in a 50-μl reaction volume with Qiagen OneStep RT-PCR Kit, with the following amplification conditions: initial denaturation at 94°C for 5 min, followed by 35 cycles of 94°C for 45 s, 58°C for 30 s, and 72°C for 45 s, and a final 7-min extension at 72°C. The PCR primers for AC1 are as follows: forward, 5′-TGCCTTATTTGGCCTTGTCTACC-3′; reverse, 5′-GACACCCGGAAAAATATGGCTAG-3′. PCR products were electrophoresed on a 1.5% agarose gel and stained by ethidium bromide.

Brain slice preparations for cyclic AMP assay

Mouse brain slices were prepared as reported (55, 56). Briefly, mice were anesthetized with 2% halothane, and brain slices (300 μm) containing the ACC were cut at 4°C with a Vibratome, in oxygenated artificial cerebrospinal fluid (ACSF) containing the following: 124 mM NaCl, 4 mM KCl, 26 mM NaHCO3, 2.0 mM CaCl2, 1.0 mM MgSO4, 1.0 mM NaH2PO4, and 10 mM d-glucose (pH 7.4). The slices were slowly brought to a final temperature of 30°C in ACSF gassed with 95% O2/5% CO2 and incubated for at least 1 hour before the experiments. Slices were then exposed to different compounds of interest for the indicated times. For cyclic AMP assay, the ACC regions were microdissected and lysed in 0.1 M HCl.

Cyclic AMP assay

Cyclic AMP assay was carried out as reported (40, 55). Briefly, HEK293 cells expressing ACs were harvested and lysed in 0.1 M HCl after different treatments. Direct cyclic AMP measurements were performed with the Direct Cyclic AMP Enzyme Immunoassay kit (Assay Designs), and optical density values were measured at 405 nm by a microplate reader. Phosphodiesterase activity was inhibited by the addition of 1 mM 3-isobutyl-1-methylxanthine (IBMX, Sigma) to cultures.

CRE luciferase reporter assay

The HEK293 cells were subcultured into 96-well plates in the absence of antibiotics and grown overnight and transfected with the pGL3–CRE–firefly luciferase and pGL3–CMV (cytomegalovirus)–Renilla luciferase constructs (0.25 μg of DNA per well) with Lipofectamine 2000 reagent. The transfected cells were incubated overnight, and media were changed to DMEM containing 10% fetal bovine serum. After 48 hours, the cells were treated with 10 μM forskolin, 10 μM A23187, or 2 mM CaCl2, or a combination of 10 μM forskolin, 10 μM A23187, and 2 mM CaCl2, in the presence or absence of each chemical tested at the concentration of 100 μM. At the end of 6-hour incubation, cells were harvested and luciferase activity was assayed by Dual-Luciferase Reporter Assay System (Promega). Relative light units were measured by a SIRIUS luminometer.

Animal models for neuropathic and inflammatory pain

Adult (6 to 8 weeks old) male mice were used. C57BL/6 mice were purchased from Charles River. Mice were maintained on a 12-hour light-dark cycle. Food and water were provided ad libitum. AC1 knockout mice were bred for several generations on a C57BL/6 background. Experiments were performed under protocols approved by the University of Toronto Animal Care Committee. A model of neuropathic pain was induced by the ligation of the common peroneal nerve (CPN) as described previously (39). Briefly, mice were anesthetized by intraperitoneal injection of a mixture of saline, ketamine (0.16 mg/kg; Bimeda-MTC), and xylazine (0.01 mg/kg; Bayer). The left CPN was slowly ligated with chromic gut suture 5-0 (Ethicon) until contraction of the dorsiflexors of the foot was visible as twitching of the digits. Allodynia was tested on postsurgical day 7, and the mice were used for electrophysiological studies on postsurgical days 7 to 14. Although initial behavioral experiments with NB001 were performed not in a blind manner, most of the subsequent experiments were performed blind, or the behavioral observations were collected by an automatic computer program (for example, the open-field test, freezing measurement) where personal factors were avoided.

Acute nociception

The spinal nociceptive tail-flick reflex was evoked by focused, radiant heat (Columbus Instruments) provided by a 50-W projector lamp focused on a 1.5-mm by 10-mm area on the underside of the tail. The latency to reflexive removal of the tail from the heat was measured by a digital photocell timer to the nearest 0.1 s. The cutoff time of 10 s was used to minimize damage to the skin of the tail. The hot plate consisted of a thermally controlled 25.4-cm by 25.4-cm metal plate (50°C and 55°C, respectively) surrounded by four Plexiglas walls (Columbus Instruments). The time between placement of the mouse on the plate and the licking or lifting of a hindpaw or jumping was measured with a digital timer. Mice were removed from the hot plate immediately after the first response. The cutoff time of 30 s was imposed to prevent tissue damage. All behavioral tests were performed at 10-min intervals. The response latency was an average of three or four measurements. The baseline latencies were measured 1 day before drug injection. The response latencies for the animals were then retaken on hot plate and tail-flick tests 30 min after intraperitoneal injection of either saline or NB001.

Mechanical allodynia

Mice were individually placed in a round, transparent container 20 cm in diameter and were allowed to acclimate for 30 min before testing. Mechanical threshold was assessed on the basis of the responsiveness of the hindpaw to the application of von Frey filaments (Stoelting) to the point of bending. The filament was applied over the dorsum of the paw while the animal was resting. Positive responses included licking, biting, and sudden withdrawal of the hindpaw. The baseline responses were taken 1 day before intraperitoneal injection. Mechanical threshold was then reassessed 30 min after intraperitoneal injection of either saline or NB001.

Elevated plus maze

The elevated plus maze (Med Associates) consisted of two open arms and two closed arms situated perpendicular to each other. The maze was situated ~70 cm from the floor. For each test, mice were individually placed in the center square and allowed to move freely for 5 min. The number of entries and time spent in each arm were recorded. The animals were given an intraperitoneal injection of NB001 (10.0 mg/kg) or saline 30 min before testing. A video camera tracking system (Ethovision) was used to generate the traces.

Open-field test

To record locomotor activity, we used an open-field activity monitor (43.2 cm by 43.2 cm by 30.5 cm; Med Associates). Briefly, this system uses paired sets of photo beams to detect movement in the open field, and movement is recorded as beam breaks. The open field is placed inside an isolation chamber with dim illumination and a fan. Each subject was given an intraperitoneal injection of NB001 (10.0 mg/kg) or saline 30 min before being placed in the center of the open field. Locomotor activity was then measured for 30 min.

RotaRod test

To test motor function, we used a RotaRod from Med Associates. The RotaRod test was performed by measuring the time each animal was able to maintain its balance while walking on a rotating drum. One hour before testing, animals were trained on the RotaRod at a constant acceleration of 16 rpm until they could stay on for 30 s. For testing, the RotaRod was set to accelerate from 4 to 40 rpm over a 5-min period. Mice were given three trials with a maximum time of 300 s and a 5-min intertrial rest interval. The latency to fall was taken as a measure of motor function. Each subject was given an intraperitoneal injection of NB001 (10.0 mg/kg) or saline 30 min before testing.

Fear memory

Fear conditioning was performed in an isolated shock chamber (Med Associates) and performed in a blind manner to the treatment. The conditioned stimulus (CS) was an 85-dB sound at 2800 Hz, and the unconditioned stimulus (US) was a continuous scrambled footshock at 0.75 mA. After 2 min of habituation, animals received the CS-US pairing [a 30-s tone (CS) and a 2-s shock (US) starting at 28 s; three shock-tone pairings were delivered at 30-s intervals], and the mice remained in the chamber for an additional 30 s for measurement of immediate freezing. At 1 day after training, each mouse was placed back into the shock chamber and the freezing response was recorded for 3 min (contextual conditioning). Two types of experiments were performed. To test the contribution of AC1 to the induction of fear memory, we pretreated mice with NB001 (at 1 mg/kg ip) or saline. Both groups of mice then received fear conditioning 30 min after the injection. One day after the conditioning, we measured the freezing responses. In the second type of experiments, we evaluated the effects of NB001 on the expression of fear memory by injecting NB001 or saline 1 day after the conditioning. NB001 or saline was injected 30 min before the measurement of freezing responses.

Microinjection of NB001 into the ACC

To determine the effects of NB001 on the ACC, we performed local microinjection of NB001 into the ACC as reported previously (14). Briefly, under ketamine and xylazine anesthesia, mice were placed in a stereotaxic instrument (Kopf Instruments). Guide cannulas (24-gauge) were implanted bilaterally into the ACC (0.7 mm anterior to bregma, 0.3 mm lateral from the midline, and 1.7 mm beneath the surface of the skull). Mice were given at least 2 weeks to recover after cannula implantation. Thirty-gauge injection cannulas that were 0.1 and 0.6 mm longer than the guide were used for intra-ACC injections. The microinjection apparatus consisted of a Hamilton glass syringe (5 μl). NB001 (10 μg/μl in saline) was infused into each side of the ACC at a rate of 0.05 μl/min; an equivalent volume of saline was used as a control. To access the dorsal and ventral ACC, we administered two injections per side, the first with the 0.1-mm-longer cannula, followed by the 0.6-mm-longer cannula. After each injection, the microinjection needle was left in place for at least 2 min.

Intrathecal injection of NB001

A 30-gauge needle attached to a 25-μl Hamilton syringe was used for the intrathecal injection of drugs. The needle was inserted into the intervertebral space between the fifth and the sixth spinal vertebrae in conscious mice, as described (57). Accurate placement of the needle was confirmed by a quick flick of the tail. NB001 (189.0 nmol) was administered slowly in a volume of 5 μl via the intrathecal route.

Whole-cell patch-clamp recordings

Coronal brain slices (300 μm) at the level of the ACC were prepared by standard methods (15, 46). Slices were transferred to a submerged recovery chamber with oxygenated (95% O2/5% CO2) ACSF (containing 124 mM NaCl, 2.5 mM KCl, 2 mM CaCl2, 1 mM MgSO4, 25 mM NaHCO3, 1 mM NaH2PO4, and 10 mM glucose) at room temperature for at least 1 hour. Experiments were performed in a recording chamber on the stage of a BX51W1 microscope equipped with infrared DIC (differential interference contrast) optics for visualization. EPSCs were recorded from layer II/III neurons with an Axon 200B amplifier (Axon Instruments), and the stimulations were delivered by a bipolar tungsten stimulating electrode placed in layer V of the ACC. AMPA receptor–mediated EPSCs were induced by repetitive stimulations at 0.05 Hz, and neurons were voltage-clamped at −70 mV in the presence of (AP5) [(2R)-amino-5-phosphonovaleric acid] (50 μM). The recording electrodes were filled with a solution containing (pipettes, 3 to 5 megohms) 145 mM K-gluconate, 5 mM NaCl, 1 mM MgCl2, 0.2 mM EGTA, 10 mM Hepes, 2 mM Mg-ATP, 0.1 mM Na3GTP, and 10 mM phosphocreatine disodium (adjusted to pH 7.2 with KOH). For miniature EPSC (mEPSC) recording, 0.5 μM tetrodotoxin was added to the perfusion solution. Picrotoxin (100 μM) was always present to block GABAA (γ-aminobutyric acid type A) receptor–mediated inhibitory synaptic currents in all experiments, and the access resistance was 15 to 30 megohms throughout the experiment. Data were discarded if access resistance changed more than 15% during an experiment. Data were filtered at 1 kHz and digitized at 10 kHz.

Hippocampal field LTP recordings

Male adult mice (7 to 10 weeks old) were anesthetized with inhaled isoflurane. Transverse slices (400 μm) of hippocampus were rapidly prepared and maintained in an interface chamber at 28°C, where they were perfused with ACSF consisting of 124 mM NaCl, 4.4 mM KCl, 1.2 mM CaCl2, 1.0 mM MgSO4, 25 mM NaHCO3, 1.0 mM NaH2PO4, and 10 mM glucose bubbled with 95% O2 and 5% CO2. The protocol of electrical stimulation and recordings has been described previously (22, 24, 58). Slices were kept in the recording chamber for at least 2 hours before the experiments. A bipolar tungsten stimulating electrode was placed in the stratum radiatum in the CA1 region, and extracellular field potentials were also recorded in the stratum radiatum with a glass microelectrode (3 to 12 megohms) filled with ACSF. Stimulus intensity was adjusted to produce a response of ~1-mV amplitude. Test responses were elicited at 0.02 Hz. LTP was induced with two tetanic train stimulations (100 Hz, 1-s trains at 20-s intervals). To study the effects of NB001 on LTP in the CA1 region of the hippocampus, we perfused 50 μM NB001 for at least 30 min before and throughout experiments in the NB001 group.

Data analysis

Results were expressed as means ± SEM. Statistical comparisons were performed with two-way ANOVA and Student’s t test. Analysis of mEPSCs was performed with cumulative probability plots. The result of hippocampal LTP in each slice was calculated as the average of the percentage of responses between 40 and 45 min after stimulation compared with baseline. Student’s t test was used for statistical analysis. The level of significance was set at P < 0.05.

Footnotes

  • * These authors contributed equally to this work.

  • Citation: H. Wang, H. Xu, L.-J. Wu, S. S. Kim, T. Chen, K. Koga, G. Descalzi, B. Gong, K. I. Vadakkan, X. Zhang, B.-K. Kaang, M. Zhuo, Identification of an Adenylyl Cyclase Inhibitor for Treating Neuropathic and Inflammatory Pain. Sci. Transl. Med. 3, 65ra3 (2011).

Supplementary Material

www.sciencetranslationalmedicine.org/cgi/content/full/3/65/65ra3/DC1

Fig. S1. The effects of local administrations of NB001 into the periphery, spinal, or the ACC on mechanical allodynia in mice after nerve injury.

Fig. S2. The effects of NB001 on classic contextual fear memory.

Table S1. Screening of AC1 inhibitors in HEK293 cells stably expressing AC1.

References and Notes

  1. Acknowledgments: We thank L. Muglia and D. R. Storm for providing AC1 knockout mice, R. Taussig and A. G. Gilman (University of Texas Southwestern Medical Center, Dallas, TX) for providing the expression plasmid for AC1 (pcDNA3-AC1), and D. M. F. Cooper (Department of Pharmacology, University of Cambridge, Cambridge, UK) for providing expression plasmids for AC5, 6, 7, and 8. Funding: M.Z. was supported by grants from the EJLB Foundation–Canadian Institutes of Health Research (CIHR) Michael Smith Chair in Neurosciences and Mental Health, Canada Research Chair, and a CIHR operating grant (CIHR81086) and proof-of-concept grant (CIHR PPP-81425). H.W. is supported by a postdoctoral fellowship from Fragile X Research Foundation of Canada. B.K. is supported by National Creative Research Initiative, Korea. Author contributions: H.W. carried out biochemical screening experiments and drafted the manuscript, H.X., L.-J.W., T.C., K.K., B.G., and X.Z. participated in electrophysiological experiments. S.S.K., G.D., H.X., X.Z., H.W., and K.I.V. participated in behavioral experiments. B.-K.K. helped to design the cyclic AMP screening experiments. M.Z. designed, drafted, and finished the final manuscript. All authors contributed to the preliminary writing of the manuscript and approved the final manuscript. Competing interests: M.Z. has filed an international patent application (USPTO Patent Application 20090233922, “Method for treating neuronal and non-neuronal pain”) based on the results reported in this paper.
View Abstract

Stay Connected to Science Translational Medicine

Navigate This Article