We have found that mutation of d-amino acid oxidase (DAO) diminished formalin-induced tonic pain. The present research further studied the analgesic effects of a series of DAO inhibitors in this model. 5-Chlorobenzo[d]isoxazol-3-ol (CBIO), 4H-thieno[3,2-b]pyrrole-5-carboxylic acid (compound 8), 5-methylpyrazole-3-carboxylic acid (AS057278), sodium benzoate, and 4-nitro-3-pyrazole carboxylic acid (NPCA) inhibited rat spinal cord-derived DAO activity in a concentration-dependent manner, with maximal inhibition of 100% and potency rank of CBIO > compound 8 > AS057278 > sodium benzoate > NPCA. In rats, intrathecal injections of CBIO, compound 8, AS057278, and sodium benzoate but not NPCA specifically prevented formalin-induced tonic pain but not acute nociception, with the same potency order as in the DAO activity assay. The highly potent analgesia of DAO inhibitors was evidenced by CBIO, which prevented 50% pain at 0.06 μg, approximately 5-fold the potency of morphine. CBIO given after formalin challenge also reversed the established pain state to the same degree as prevention. The antihyperalgesic potencies of these DAO inhibitors were highly correlated to their inhibitions of spinal DAO activity. Maximum inhibition of pain by these compounds was approximately 60%, comparable with that of the N-methyl-d-aspartic acid receptor antagonist (+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine maleate (MK-801), suggesting that a larger portion of formalin-induced tonic pain is “DAO-sensitive,” whereas the remaining 40% of tonic pain and acute nociception is “DAO-insensitive.” These findings, combined with our previous DAO gene mutation and induction results, indicate spinal DAO mediates both induction and maintenance of formalin-induced tonic pain and further validate spinal DAO as a novel and efficacious target molecule for the treatment of chronic pain.
d-Amino acid oxidase (DAO) is a peroxisomal flavoprotein that catalyzes the oxidative deamination of neutral and polar d-amino acids, with strict stereospecificity, to α-keto acids, NH3, and H2O2 (see reviews by Pollegioni et al., 2007; Williams, 2009; Verrall et al., 2010). In the central nervous system, DAO expression is restricted to the lower brainstem, cerebellum, and spinal cord, with decreasing levels in the midbrain, cortex, and hippocampus (Kapoor and Kapoor, 1997; Moreno et al., 1999; Horiike et al., 2001; Yoshikawa et al., 2004). We discovered that mutation of the DAO gene blocked formalin-induced tonic pain but not acute nociception by using ddY/DAO(+/+) mice versus ddY/DAO(−/−) mice (Zhao et al., 2008), in sharp contrast to an earlier finding (Wake et al., 2001). Systemic and intrathecal administration of the DAO inhibitor sodium benzoate also specifically attenuated L5/L6 spinal nerve tight ligation-induced neuropathic mechanical allodynia and formalin-induced tonic pain in rats and mice. In contrast, sodium benzoate administered by either route was not effective in acute pain such as the early-phase flinch response in the formalin test or thermal nociceptive response in the tail-flick test or hot-plate test (Zhao et al., 2008, 2010). Moreover, L5/L6 spinal nerve damage up-regulated spinal DAO gene expression and DAO enzymatic activity, with the same time course as peripheral nerve damage-induced mechanical allodynia (Zhao et al., 2010). Thus, we have obtained the first evidence indicating that spinal DAO specifically participates in chronic pain transmission and is a potential target molecule for the treatment of chronic pain including chronic neuropathic pain (Zhao et al., 2008, 2010). However, several issues need to be clarified before validation of the hypothesis of spinal DAO as an efficacious pain target molecule.
DAO mutation using ddY/DAO(−/−) mice entirely eliminates DAO activity (Konno and Yasumura, 1983; Xin et al., 2007, 2010; Zhao et al., 2008); however, compensation developed during growth and nonspecific effects such as changes in motor activity might account for observed pain behaviors and could lead to false positive or negative results in animal studies. DDY/DAO(−/−) mice were indeed reported to have less locomotor activity than ddY/DAO(+/+) mice (Almond et al., 2006), thus making interpretation of our previously observed behavior changes more complex. In addition, sodium benzoate (benzoic acid) is the only DAO inhibitor used so far in animal pain studies and without dose-response analysis. Because of its low potency and rapid excretion in urine (Kubota and Ishizaki, 1991; MacArthur et al., 2004; Williams and Lock, 2005; Xin et al., 2007), it was widely accepted to give sodium benzoate in high doses such as 100 to 1000 mg/kg for systemic administration (Moses et al., 1996; Williams and Lock, 2005; Xin et al., 2005, 2007; Wu et al., 2006; Zhao et al., 2008, 2010). It is reasonable to speculate that the analgesic action of sodium benzoate may be a result of its possible nonspecific effects caused by high doses. Thus, more DAO inhibitors with higher potency and chemical structures distinct from sodium benzoate are needed for further studies. A series of compounds having a carboxylic acid group or related groups have emerged as novel inhibitors of DAO with high potencies for the treatment of schizophrenia, including 5-chlorobenzo[d] isoxazol-3-ol (CBIO) (Ferraris et al., 2008; Hashimoto et al., 2009; Horio et al., 2009), compound 8 (sc-203909; 4H-thieno[3,2-b]pyrrole-5-carboxylic acid) from Merck (Darmstadt, Germany) (Sparey et al., 2008; Smith et al., 2009), 5-methylpyrazole-3-carboxylic acid (AS057278) (Adage et al., 2008), and 4-nitro-3-pyrazole carboxylic acid (NPCA) (Fang et al., 2009), as well as 3-hydroxyquinolin-2-(1H)-one (compound 2) from Pfizer (New York, NY) (Duplantier et al., 2009). The chemical structures of CBIO, compound 8, AS057278, sodium benzoate, and NPCA are presented in Fig. 1 where compound 2 is also shown for comparison.
High efficacy is critical for selection of a target molecule for establishing drug screening and development programs against chronic pain. The efficacy of spinal DAO inhibition in tonic pain is not yet established partly because of the lack of dose-response analysis of DAO inhibitors on analgesia. It is important to compare the efficacies of DAO inhibitors to a well established chronic pain drug target molecule particularly validated in humans, such as N-methyl-d-aspartic acid (NMDA) receptors. Activation of spinal NMDA receptors is believed to play an essential role for central sensitization-mediated chronic pain (see reviews by Basbaum et al., 2009 and Hulsebosch et al., 2009). Intrathecal administration of NMDA receptor antagonists including (+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine maleate (MK-801) have been reported to suppress formalin-induced tonic pain (Näsström et al., 1992; Yamamoto and Yaksh, 1992; Chaplan et al., 1997).
Therefore, the aim of the present study is to systemically study a series of potent DAO inhibitors on tonic pain in the formalin test to further elucidate the role of spinal DAO in central sensitization-mediated pain. Selection of the formalin test was based on the well accepted notion that subcutaneous injection of formalin produces biphasic behavioral effects in animals with the early phase reflecting an acute nociceptive state and the following tonic pain caused by ongoing activation of sensory afferents and tissue damage as well as inflammation reflecting a state of central sensitization, a mechanism shared by chronic or persistent pain including neuropathic pain (Cook et al., 1987; Coderre et al., 1993; Woolf et al., 1994; Jett et al., 1997; see review by Sawynok and Liu, 2003). The present study included the following protocols: 1) determination of inhibitory effects of CBIO, compound 8, AS057278, sodium benzoate, and NPCA, compared with MK-801, on rat spinal cord-derived DAO (rsDAO) and porcine kidney-derived DAO (pkDAO) enzymatic activities; 2) examination of preventive effects of CBIO, compound 8, AS057278, NPCA, and sodium benzoate administrated intrathecally by multiple doses, compared with MK-801, on formalin-induced tonic pain in rats; the blocking effect of CBIO administered after formalin challenge on the established pain state was also studied; 3) analyses of dose-response curves of DAO inhibitors on rsDAO and pkDAO enzymatic activities and formalin-induced tonic pain to yield potencies and maximum effects for further determination of the correlation between DAO inhibition and blockade of tonic pain as well as efficacy for spinal DAO pain transduction and transmission; and 4) measurement of effects of intrathecally injected CBIO on carrageenan-induced inflammatory thermal hyperalgesia to further test its specificity on analgesia.
Materials and Methods
AS057278, NPCA, γ-carrageenan, and MK-801 were purchased from Sigma-Aldrich (St Louis, MO), and sodium benzoate and formalin were from Sinopharm Group Chemical Reagent Co., Ltd. (Shanghai, China). CBIO, compound 8, and porcine kidney DAO (5 unit/mg) were from Maybridge Chemicals (Cornwall, U.K.), Santa Cruz Biotechnology Inc. (Santa Cruz, CA), and Serva (Heidelberg, Germany), respectively. All drugs and reagents were freshly dissolved in sterile normal saline solution (Sinopharm Group Chemical Reagent Co., Ltd.) with the pH adjusted by 1 M NaOH solution as needed, or 50% DMSO in normal saline in one case as indicated below.
Male Wistar rats (180–250 g) were obtained from the Shanghai Experimental Animal Institute for Biological Sciences (Shanghai, China). The animals were housed in a temperature- and humidity-controlled environment on a 12-h light/dark cycle (lights on at 6:00 AM). Food and water were freely available. The research protocol was approved by the Animal Care and Welfare Committee of Shanghai Jiao Tong University School of Pharmacy and followed the animal care guidelines of the National Institutes of Health. Animals were acclimated to the laboratory environment for 3 to 5 days before entering the study. Experimental study groups were assigned randomly, and the researcher was blind for behavior testing.
Measurement of rsDAO and pkDAO Enzymatic Activities.
The enzymatic activities of rsDAO and pkDAO were determined according to the keto acid method (D'Aniello et al., 1993). The rats were killed by decapitation, and the spinal cords were quickly removed on ice and weighed. The tissue was homogenized (10,000 rpm for 15 s) with a homogenizer (Fluko Equipment Co., Shanghai, China) in 0.1 M Tris-HCl, pH 8.2 (1 g/3 ml) and centrifuged (4000 rpm for 10 min) at 4°C. The protein content was measured by Coomassie brilliant blue staining. Fifty microliters of d-alanine (0.015 M dissolved in 0.1 M Tris-HCl buffer, pH 8.2) was added to 50 μl of supernatants with final concentration of 6 mM d-alanine and incubated (700 rpm) at 37°C for 60 min. Preliminary results showed the Km value of d-alanine in this assay was 5.8 mM. Trichloroacetic acid (25%; 50 μl) was then added to the assay mixture, mixed, and centrifuged (14,000 rpm, 5 min). The supernatant (50 μl) was mixed with 50 μl of 1 mM 2,4-dinitrophenylhydrazine (in 1 M HCl) and incubated (700 rpm) at 37°C for 10 min. Finally, 100 μl of 1.5 M sodium hydroxide was added, mixed, and incubated (700 rpm) at 37°C for 10 min. The absorbance was read at 450 nm on an ELx800 Universal Microplate Reader (BioTek Instruments, Winooski, VT) against a blank sample consisting of the same homogenates without d-alanine. The activity of DAO in the homogenates was quantified against the standard curve of pyruvic acid (from 100 to 800 μM; R2 > 0.99). The specific enzymatic activity was expressed as pyruvate production per milligram of protein per minute. The measurement of enzymatic activity of pkDAO was the same as in the spinal DAO assay except that the spinal cord supernatant solution was replaced by pure pkDAO solution (final concentration: 0.2 unit/0.04 mg/ml), final d-alanine concentration was 1 mM (Km value in this assay) instead of 6 mM, and incubation time was 5 min instead of 60 min.
A 24-cm polyethylene catheter (PE-10: 0.28 mm i.d. and 0.61 mm o.d.; Clay Adams, Parsippany, NJ) with volume of approximately 13 μl was inserted into the rat lumbar level of the spinal cord under intraperitoneal injection of pentobarbital (50 mg/kg) anesthesia as described previously (Wei et al., 2007). After recovery from anesthesia, the placing of the catheter in the spinal cord was verified by administering 4% lidocaine (10 μl followed by 15 μl of normal saline for flushing) with a 50-μl micro injector (Shanghai Anting Micro-Injector Factory, Shanghai, China). The lidocaine test was performed 5 to 7 days before the start of the drug testing sessions. Only those rats that had no motor impairment before lidocaine injection but had bilateral paralysis of hind limbs after intrathecal administration of lidocaine were selected for the study, but the exclusion rate in our laboratory was zero. For intrathecal administration of control and test articles, the drugs were microinjected with a 50-μl micro injector in a volume of 10 μl (or 30 μl in one case because of low drug solubility as indicated below) followed by a normal saline flush in a volume of 15 μl.
The Rat Formalin Test of Acute Nociceptive Pain and Tonic Pain.
Rats were acclimated individually to the observation cage for 30 min before testing. The formalin test was performed as described previously (Wang et al., 2000) with slight modifications, by injecting 50 μl of 5% (or 1% in some experiments as indicated) formalin in 0.9% saline subcutaneously on the dorsal side of the left hindpaw, and the rat was immediately placed in a 23 × 35 × 19-cm transparent polycarbonate box. Nociceptive behavior was manually quantified by counting the number of the formalin-injected paw flinches supplemented by licking duration (considered as one count per second matching the duration of the typical flinching behavior) in 1-min epochs. Measurements were taken at 12-min intervals beginning immediately after formalin injection and ending 96 min later.
The Rat Carrageenan Model of Acute Inflammatory Pain.
To induce a local inflammation, the right hindpaws of rats received subcutaneously intraplantar surface injection of 100 μl of 2% carrageenan in normal saline. Hyperalgesia was assessed by placing the hindpaw above a radiant heat source (setting to a low intensity of 45) and measuring the paw-withdrawal latency with a noxious heat stimulus, using 390 G Plantar Test (IITC Life Science Instruments, Woodland Hills, CA) as described by Hargreaves et al. (1988). Paw-withdrawal latencies were determined in both carrageenan-injured and contralateral (uninjured) hindpaws. The cutoff latency was 30 s to avoid tissue damage. The paw-withdrawal latencies were evaluated in no less than 30 min at different time points before and after control and test article administration. Each test was calculated as a mean of three repeated measurements.
The Rotarod Motor Coordination Test and Motor Activity Test.
Motor coordination performance was assessed by means of a YLS-4C Rota Rod with automatic timers and falling sensors (Yiyan Scientific Ltd., Shandong, China). The rats were trained and tested by using an accelerated speed from 5 to 25 rpm within 1 min followed by 25 rpm for 2 more minutes. The accumulated time (seconds/3 min) for animals to spend on the rotarod was recorded during the 3-min observation period after the animals were trained once a day each for 9 min for 3 days. The accumulated time spent on the rod had to be at least 120 s allowing inclusion in the study. Motor activity was also measured with a YLS-1c electromagnetic activity monitor (YLS-1C Motor) with a printer (Yiyan Scientific Ltd.) located in a quiet environment. Animals were each placed into an individual testing cage identical to their home cages with free access to diet and water. Three hours later, automatic counting of accumulated movements (counts/15 min) were continuously recorded and printed out in a 15-min interval for the scheduled period.
For dose (concentration)-response curve analysis of DAO inhibitors on DAO enzymatic activity or formalin-induced tonic pain, the parameters, i.e., minimum effect, maximum effect (Emax), half-effective dose or concentration (ED50 or EC50), and Hill coefficient (n), were calculated from individual dose-response curves. To determine the parameters of dose-response or concentration-response curves, values of response (Y) were fitted by nonlinear least-squares curves to the relation: Y = a + bx, where x = [D]n/(ED50n + [D]n) or x = [C]n/ (EC50n + [C]n), to give the value of ED50 or EC50 and b (Emax), yielding a minimum residual sum of squares of deviations from the theoretical curve (Wang and Pang, 1993). For correlationship between potencies of DAO inhibitors on DAO enzymatic activity and formalin-induced tonic pain, lineal correlation coefficient was calculated.
Results are expressed as mean ± S.E.M., and statistical significance was evaluated by a two-way repeated measures analysis of variance (ANOVA) followed by post hoc two-tailed Student's t tests or by an unpaired and two-tailed Student's t test. The statistical significance criterion P value was 0.05. All data calculations and statistics analysis were done by using Prism version 5.01 (GraphPad Software Inc., San Diego, CA).
Inhibitory Effects of DAO Inhibitors on rsDAO and pkDAO Enzymatic Activities.
Five DAO inhibitors (CBIO, compound 8, AS057278, sodium benzoate, and NPCA) were assayed (triplicates) for their effects on DAO enzymatic activity in homogenates from rat spinal cords. In the control homogenates incubated for 60 min, the α-keto acid produced by d-alanine was determined by measurement of pyruvic acid, and the specific DAO activity was approximately 1.5 nmol/mg protein/min. CBIO (10−8 to 3 × 10−6 M), compound 8 (10−8 to 10−5 M), AS057278 (3 × 10−7 to 10−4 M), sodium benzoate (3 × 10−6 to 10−3 M), and NPCA (10−4 to 3 × 10−2 M) all inhibited DAO enzymatic activity in a concentration-dependent manner (Fig. 2A), with maximum inhibition (Emax) values of approximately 100%, and half-inhibitory concentration (IC50) values of 0.09 μM, 0.17 μM, 15.4 μM, 75.4 μM and 20 mM, respectively. In addition, the NMDA receptor antagonist MK-801 was tested and found not to inhibit DAO enzymatic activity up to 0.8 mM (data not shown).
The inhibitory effects of these compounds were also tested in the pkDAO enzymatic activity assay (triplicates) where the specific DAO activity was approximately 1 μmol/mg protein/min after 5 min of incubation. Five DAO compounds inhibited pkDAO enzymatic activity in a concentration-dependent manner (Fig. 2B), with Emax values of approximately 100%. The IC50 values for these compounds were 0.15 μM, 0.52 μM, 14.6 μM, 44.6 μM, and 5.9 mM, with the same potency order as in the rsDAO assay, i.e., CBIO > compound 8 > AS057278 > sodium benzoate > NPCA. Our results with CBIO, compound 8, AS057278, and sodium benzoate were consistent with previous reports where IC50 or Ki values were 145 or 245 nM (Sparey et al., 2008; Smith et al., 2009), 188 nM (Ferraris et al., 2008), 0.91 μM (Adage et al., 2008), and 16 μM (Ki, Vanconi et al., 1997), respectively. However, NPCA less potently inhibited rsDAO and pkDAO DAO with IC50 values of 20 and 5.9 mM, respectively, in contrast to a granted patent where the IC50 value of NPCA was less than 10 μM (Fang et al., 2009).
Differentially Inhibitory Effects of Intrathecal Injections of DAO Inhibitors and the NMDA Receptor Antagonist on Formalin-Induced Tonic Pain.
The analgesic effects of DAO inhibitors (CBIO, compound 8, AS057278, sodium benzoate) as well as NPCA, compared with that of the NMDA receptor antagonist MK-801, on formalin-induced nociceptive pain and tonic pain were studied via direct spinal cord administration in rats chronically implanted with intrathecal cannulas. Seven groups of rats (n = 5 in each group) received intrathecal bolus injection of normal saline (10 μl) or CBIO (0.01, 0.03, 0.1, 0.3, 1, or 3 μg) 30 min before formalin injection. Subcutaneous injection of formalin in control rats receiving intrathecal injection of normal saline produced a characteristic bi-phasic flinching response supplemented by licking response consisting of an initial, rapidly decaying acute phase (within 12 min after formalin injection) followed by a slowly rising and long-lived (12–96 min) tonic phase as shown in Fig. 3A. Although formalin-induced flinching and licking may be independent behaviors (Abbott et al., 1995) and may have different underlying mechanisms (Wheeler-Aceto and Cowan, 1993; Sawynok and Reid, 2002), both reflect the established pain state. Our preliminary experiment demonstrated CBIO blocked 1 or 5% formalin-induced flinching and licking behaviors in the tonic phase to a similar degree; thus, both behaviors were combined for overall measurement. Compared with the normal saline control, CBIO up to 3 μg did not prevent formalin-induced flinch response (acute nociception) in the acute phase (P > 0.05 by ANOVA). However, CBIO prevented formalin-induced tonic pain in the tonic phase in a dose-dependent manner (Fig. 3A). The areas under the flinching response curve from 12 to 96 min (AUC12–96 min) were calculated. Dose-response analysis of CBIO by best fit showed that maximum inhibition (Emax) (at 1 μg or 6.2 nmol) of formalin-induced tonic pain was 67.3% and half-effective dose (ED50) was 0.06 μg (0.39 nmol) (Fig. 3B). For a comparison, 1.5 (2 nmol) and 10 μg (13.2 nmol) of morphine sulfate administered intrathecally abolished formalin-induced tonic pain by 50 and 100%, respectively (Wang et al., 2000; Zhao et al., 2010). Because the tonic phase flinches were measured subsequently after the acute phase, the differential effects of CBIO on the acute and tonic phases may be caused by its variable tissue concentrations/pharmacokinetics changes at different time points. Therefore, two groups of intrathecally cannulated rats (n = 4 in each group) received intrathecal bolus injection of saline (10 μl) or CBIO (3 μg) 90 min before 5% formalin challenge (matching the time to the peak of the tonic phase in the previous experiment). As shown in Fig. 4A, CBIO given 90 min before did not significantly prevent formalin-induced nociceptive flinches in the acute phase (P > 0.05, unpaired and two-tailed Student's t test), the same as CBIO given 30 min before formalin injection as described above; but CBIO still significantly inhibited tonic pain measured as AUC12–96 min by 52.0%, which is a moderately smaller effect than when CBIO was given 30 min before formalin injection presumably because of its longer disposition in the spinal cord. In addition, because the ineffectiveness of CBIO on the acute nociception induced by 5% formalin might be caused by its ceiling effect, CBIO was examined by using 1% formalin. Two groups of rats (n = 4 in each group) received intrathecal bolus injection of saline (10 μl) or CBIO (3 μg) 30 min before formalin challenge. Compared with 5% formalin, 1% formalin produced similar acute nociception but less and shorter-lived tonic pain. Three micrograms of CBIO blocked 1% formalin-induced tonic pain measured as AUC12–96 min by 72.7%, but not acute nociception (P > 0.05 by unpaired and two-tailed Student's t test) (Fig. 4B).
To test whether CBIO injected postformalin challenge still effectively reversed the established pain state, two groups of intrathecally cannulated rats (n = 4 in each group) received intrathecal bolus injections of 10 μl of normal saline or 1 μg of CBIO 24 min after formalin administration. CBIO injected after formalin challenge markedly reversed the established pain state in the tonic phase, with the inhibitory rate of 60.3% measured by AUC36–96 min (Fig. 4C), the same as the preventive effect of 66.6% by CBIO administered before formalin challenge (Fig. 3A).
Because it has been reported that DAO mutant mice had less locomotor activity (Almond et al., 2006), which might account for our observed pain behaviors, the rotarod motor coordination test and motor activity test were conducted to examine the locomotor functions of CBIO, although CBIO did not exhibit apparent motor side effects during the observation period of the above experiments. Two groups of rats (n = 4 in each group) received intrathecal bolus injection of normal saline (10 μl) or CBIO (3 μg), and the accumulated time for rats to spend on the rotarod (at a rate of 5 to 25 rpm within 1 min followed by 25 rpm for 2 more minutes) was recorded before (151.8 ± 28.3 s/3 min for saline versus 143.8 ± 23.8 s/3 min for CBIO), 30 min (90.8 ± 30.6 s/3 min versus 95.8 ± 29.2 s/3 min), and 90 min (116.3 ± 37.9 s/3 min versus 141.8 ± 34.7 s/3 min) after administration of control and test articles. Both groups of rats did not show significant difference at any time points measured over the observation period (P > 0.05 by two-way repeated-measures ANOVA). In addition, both groups of rats (n = 4 in each group) receiving intrathecal injection of normal saline (10 μl) or CBIO (3 μg) showed the same motor activity at all time points during observation of at least 2 h after injection (2-h total counts: 173.2 ± 84.7 for saline versus 172.1 ± 45.1 for CBIO, P > 0.05, by unpaired and two-tailed Student's t test). Thus the analgesia of CBIO is not falsely positive because of locomotor behaviors.
Two groups of intrathecally cannulated rats received intrathecal bolus injection of 30 μl of 50% DMSO in normal saline (n = 4) or 750 μg of NPCA (n = 5) 30 min before formalin injection. Formalin produced the characteristic bi-phasic flinch responses. Compared with the normal saline control, NPCA was not effective in prevention of either formalin-induced tonic pain in the late phase or the flinch response in the early phase (Fig. 4D). No apparent motor side effects of NPCA were observed during the study period.
The inhibitory effects of compound 8 (0.1, 1, 3, or 10 μg), AS057278 (0.3, 1, 3, or 10 μg), and sodium benzoate (3, 10, 30, 100, or 300 μg) on the formalin test were also tested in 16 groups of intrathecally cannulated rats (n = 3–6 in each group). Compared with the normal saline control, compound 8 (Fig. 5A), AS057278 (Fig. 6A), and sodium benzoate (Fig. 7A) all prevented formalin-induced tonic pain measured by AUC12–96 min, in a dose-dependent fashion, but not the flinch response in the acute phase. Dose-response analysis by best fit showed that Emax values for compound 8, AS047278, and sodium benzoate to prevent tonic pain were 67.0, 55.6, and 57.6%, whereas their ED50 values were 0.17 μg (1.0 nmol), 1.4 μg (11.1 nmol), and 10.2 μg (70.8 nmol), respectively (Figs. 5B, 6B, and 7B, respectively). No apparent motor side effects were observed for these compounds during the study period.
Seven groups of intrathecally cannulated rats (n = 4 in each group) received intrathecal bolus injection of normal saline (10 μl) or MK-801 (0.03, 0.1, 0.3, 1, 3, or 10 μg) 30 min before formalin challenge. Formalin produced the same characteristic bi-phasic flinch responses. Compared with the normal saline control, MK-801 produced a dose-dependent prevention of formalin-induced tonic pain in the late phase measured by AUC12–96 min in a dose-dependent manner, but not the nociceptive response in the acute phase (Fig. 8A). Dose-response analysis by best fit showed that Emax and ED50 for MK-801 to block tonic pain were 50.9% and 0.3 μg (0.9 nmol), respectively (Fig. 8B). The potency (Yamamoto and Yaksh, 1992) and efficacy (Chaplan et al., 1997) of MK-801 were in agreement with previous reports.
Correlation between Inhibition of DAO Enzymatic Activity and Blockade of Formalin-Induced Tonic Pain.
Correlation analysis among the above DAO inhibitors between their inhibition of DAO enzymatic activity and prevention of formalin-induced tonic pain was carried out. Because NPCA was not effective in the blockade of tonic pain up to a maximally allowable dose (750 μg) and no ED50 value was obtained, the maximal allowable dose of NPCA was then plotted instead (Fig. 8). Correlation analysis showed that blockade potencies (ED50 values) of formalin-induced tonic pain by CBIO, compound 8, AS057278, sodium benzoate, and NPCA were highly correlated to their IC50 values on enzymatic activities of both rsDAO (Fig. 9A; r2 = 0.9906, P = 0.0004) and pkDAO (Fig. 9B; r2 = 0.9925, P = 0.0003).
Ineffectiveness of Intrathecal Injection of CBIO on Carrageenan-Induced Acute Heat Hyperalgesia.
Two groups of intrathecally cannulated rats (n = 4 in each group) received intraplantar injection of 100 μl of 2% carrageenan in normal saline and 2.5 h later received intrathecal injection of normal saline (10 μl) or CBIO (1 μg). Normal and inflamed paw withdrawal latency to radiant heat stimuli were measured 2.5 h before subcutaneous carrageenan injection, 0.5 h before, and 0.5, 1, 2, and 3 h after intrathecal administration of control and test articles. As shown in Fig. 10, carrageenan produced inflammation observed by swelling and redness and hyperalgesia reflected by reduction of inflamed paw withdrawal latency compared with normal paws (P < 0.05 by two-way repeated measures ANOVA). Intrathecal injection of CBIO did not produce any analgesic or antihyperalgesic effects on either the normal paw or inflamed paw (P > 0.05 by two-way repeated-measures ANOVA).
Sodium benzoate (benzoic acid) is a prototypical competitive DAO inhibitor (Vanconi et al., 1997), and the crystal structure of DAO complexed with benzoic acid showed that critical hydrogen bonds were formed between the carboxylate group of benzoic acid and Arg283 and Tyr228 residues of the enzyme (Mattevi et al., 1996). A series of compounds that share the same functional groups, such as the carboxylate or isoxazole groups, have emerged as novel inhibitors of DAO with high potencies, including carboxylate grouped compounds, i.e., AS057278 (Adage et al., 2008), compound 8 (Sparey et al., 2008; Smith et al., 2009), and NPCA (Fang et al., 2009), as well as isoxazole-grouped compound CBIO (Ferraris et al., 2008) and hydroxyquinolin-grouped compound 2 (Duplantier et al., 2009). In the present study, we systemically studied inhibitory effects of these DAO inhibitors on rsDAO and pkDAO enzymatic activities. Our results confirmed that CBIO, compound 8, AS057278, and sodium benzoate all inhibited both spinal cord-derived DAO and porcine kidney DAO enzymatic activity in a concentration-dependent manner with maximal inhibition of 100% and potency rank of CBIO (0.09 μM for rsDAO versus 0.15 μM for pkDAO) > compound 8 (0.17 versus 0.52 μM) > AS057278 (15.4 versus 14.6 μM) > sodium benzoate (75.4 versus 44.6 μM) > NPCA (20 and 5.9 mM) consistent with previous results (Vanconi et al., 1997; Adage et al., 2008; Ferraris et al., 2008; Smith et al., 2009). However, NPCA less potently inhibited DAO enzymatic activity with IC50 values of 20 or 5.9 mM, in contrast to a patent where the IC50 value of NPCA was less than 10 μM (Fang et al., 2009). It is believed that AS057278, CBIO, compound 8, and compound 2 exhibit DAO inhibition via the same mechanism as benzoic acid through formation of hydrogen bonds with Arg283 and Tyr228 residues of the enzyme (Mattevi et al., 1996; Ferraris et al., 2008; Sparey et al., 2008; Duplantier et al., 2009; Smith et al., 2009). NPCA is a 4-nitro-substituted 3-pyrazole-carboxylic acid analog and thus its low potency may be caused by its nitro group-induced electron withdrawing effect, in contrast to AS57278, which is a 5-methyl-substituted 3-pyrazole-carboxylic acid analog and provides electrons to assist the formation of hydrogen bonds.
The current study has extended previous results where systemic and intrathecal administrations of sodium benzoate prevented formalin-induced tonic pain (Zhao et al., 2008, 2010), by demonstrating that the more potent DAO inhibitors CBIO, compound 8, and AS057278 administered by the intrathecal route all powerfully prevented formalin-induced tonic pain in a dose-dependent manner. The potency order determined was CBIO > compound 8 > AS057278 > sodium benzoate > NPCA (ineffective). Highly potent analgesic actions of DAO inhibitors were evidenced by the fact that as little as 0.06 μg (0.39 nmol) of CBIO blocked pain by 50%, approximately 5- and 2-fold more potent than the μ-opioid receptor agonist morphine and the NMDA receptor antagonist MK-801, respectively. In contrast, NPCA was ineffective up to 750 μg. Taken together, preventive effects of DAO inhibitors on formalin-induced tonic pain were correlated very well to their inhibitions of spinal cord-derived DAO (as well as porcine kidney-derived DAO) enzymatic activity. It is important to point out that sodium benzoate, compound 8, and CBIO are structurally unrelated to the pyrazole-carboxylic acids (i.e., AS057278 and NPCA), which rules out the possibility that the analgesic effects are caused by their structure-related but non-DAO-specific effects. This is also emphasized by the finding that AS057278 is effective but NPCA is ineffective, although both are 3-pyrazole-carboxylic acid analogs. Furthermore, we show that CBIO, whether given before or after formalin challenge, markedly prevented or reversed formalin-induced tonic pain to the same degree, suggesting that spinal DAO involves both induction and maintenance of tonic pain. In comparison, previous studies indicated that the NMDA receptor antagonist MK-801 prevented formalin-induced tonic pain but was not effective in reversion of the established pain state (Yamamoto and Yaksh, 1992; Vaccarino et al., 1993). These findings, along with previous DAO gene mutation results (Zhao et al., 2008), indicate that spinal DAO mediates both induction and maintenance of formalin-induced tonic pain and further validates spinal DAO as a target molecule for the treatment of chronic pain.
Dose-response analysis is a powerful tool for pharmacodynamic studies, which calculates and projects several parameters such as Emax and ED50 or EC50 (Wang and Pang, 1993). ED50 values are used to determine the potencies of individual DAO inhibitors and correlate to their spinal DAO inhibitory effects, whereas the values of Emax are used to calculate the efficacy of individual DAO inhibitor and putting Emax values together to assess the efficacy of spinal DAO for pain transmission and transduction. Intrathecal injection of CBIO, compound 8, AS078278, and sodium benzoate blocked chronic pain response in rats by maximum inhibition of 67, 67, 56, and 58%, respectively, whereas these compounds inhibited rat spinal cord-derived DAO enzymatic activity by 100%. By the systemic route, DAO inhibitors also exhibited the same degree of inhibition. Subcutaneous injection of CBIO blocked formalin-induced tonic pain in a dose-dependent manner with maximum effects of 60, and 65%, respectively, in rats and Swiss mice (N. Gong et al., unpublished data). Moreover, intravenous administration of a single dose of sodium benzoate at 400 mg/kg, which was sufficient to entirely block in vivo DAO activity (Xin et al., 2005, 2007, 2010), prevented formalin-induced tonic pain by 63 and 68%, respectively, in Swiss mice (Zhao et al., 2008) and BALB/c mice (N. Gong et al., unpublished data). In addition, mutation of the DAO gene, leading to entire loss of DAO enzymatic activity, blocked formalin-induced tonic pain by 60% in ddy/DAO(−/−) mice compared with ddy/DAO(+/+) (Zhao et al., 2008). These results, summarized in Table 1, indicate that DAO is largely (approximately 60%) involved in formalin-induced tonic pain. Because both DAO inhibitors and DAO mutation did not alter formalin-induced nociception in the acute phase, formalin-induced pain state could be divided into two portions: i.e., a “DAO-sensitive” (60% tonic pain) component and a “DAO-insensitive” (40% tonic pain and 100% nociception) component.
Although the NMDA receptor in the spinal cord dorsal horn is silent under normal conditions, increased release of neurotransmitters such as glutamate and substance P from nociceptors after injuries depolarizes postsynaptic neurons to activate quiescent NMDA receptors, which in return exacerbates responses to noxious stimuli to generate tonic pain. Therefore, activation of NMDA receptors is well known to be essential for central sensitization-mediated chronic pain (see reviews by Basbaum et al., 2009 and Hulsebosch et al., 2009). Intrathecal administration of NMDA receptor antagonists has been reported to suppress formalin-induced tonic pain (Näsström et al., 1992; Yamamoto and Yaksh, 1992; Chaplan et al., 1997). For example, MK-801 attenuated tonic pain in rats by 67 ± 16% (Chaplan et al., 1997), consistent with our MK-801 observation. It is important to note that the analgesia efficacy of DAO inhibitors was the same as that of MK-801, which is a potent, selective, and noncompetitive antagonist of NMDA receptors (Wong et al., 1986) without inhibiting DAO activity (from the current study), and is commonly used as a prototype in electrophysiological and behavioral pharmacology of NMDA receptor functions. Because high efficacy is critical for selection of an ideal target molecule for treating chronic pain, the comparability of DAO inhibitors to NMDA receptor antagonists strengthens the significant role of the spinal DAO system in pain transmission and transduction. Indeed, SEP-227900, a DAO inhibitor of unknown structure, was reportedly in early-stage clinical investigation for the treatment of neuropathic pain (Williams, 2009). The similarity (e.g., efficacy and specificity) and differences (e.g., before versus after formalin) of DAO inhibitors and NMDA receptor antagonists also make it difficult to link the two pain signal pathways. It has been proposed that activation of NMDA receptors leads to the up-regulation of DAO expression (Yoshikawa et al., 2004).
Sodium benzoate's analgesic effects are chronic pain-specific because it blocked formalin-induced tonic pain and peripheral nerve damage-induced neuropathic mechanical allodynia but not acute nociception in the formalin test, hot-tail immersion test, or hot-plate test (Zhao et al., 2008, 2010). Because both formalin-induced tonic pain and peripheral nerve injury-provoked neuropathic pain are mediated by ongoing activation of sensory afferents leading to central sensitization, i.e., increased responsiveness of higher-order spinal neurons to peripheral input (Cook et al., 1987; Coderre et al., 1993; Woolf et al., 1994; Jett et al., 1997), the analgesic action of sodium benzoate is suggested to be central sensitization-specific (Zhao et al., 2008, 2010). This study extended the specificity of DAO inhibitors by confirming that CBIO, compound 8, and AS057278 are also not effective in acute nociception in the formalin test and by further demonstrating that CBIO is ineffective in carrageenan-induced acute thermal hyperalgesia. Serving as an extensively used model of inflammation and inflammatory pain (as the formalin test), carrageenan produced peripheral inflammation and heat hyperalgesia and mechanical allodynia via the mechanism of peripheral sensitization in the early phase (<3–4 h after carrageenan challenge) (Kocher et al., 1987; Dirig et al., 1998; Cui et al., 1999) without generating overt firing of sensory afferents (see review by Sawynok and Liu, 2003). Thus the ineffectiveness of DAO inhibitors on carrageenan-induced thermal hyperalgesia may further suggest that DAO is specifically involved in continuous activation of sensory afferents and central sensitization-mediated persistent pain. However, beyond this peripheral role for inflammatory mediators in carrageenan-mediated hyperalgesia, a component of spinal sensitization, at least in the late phase (>3–4 h after carrageenan challenge), may also participate in the hyperalgesia (Dirig et al., 1998; Cui et al., 1999; Hedo et al., 1999; Tao et al., 2003). The exact mechanism for the differential effects of DAO inhibitors on formalin-induced analgesia and carrageenan-induced hyperalgesia warrants further investigations.
In conclusion, intrathecal injections of a series of DAO inhibitors (CBIO, compound 8, AS057278, and sodium benzoate, but not NPCA) specifically and powerfully prevented and reversed formalin-induced tonic pain but not acute nociception (or carrageenan-induced acute thermal hyperalgesia) in a dose-dependent manner. The ED50 value of CBIO for blockade of tonic pain was as little as 0.06 μg, approximately 5- and 2-fold more potent than morphine and MK-801. Dose-response analysis revealed that analgesia potencies of these DAO inhibitors correlated well to their inhibitions of spinal cord-derived DAO enzymatic activity. Moreover, maximum inhibition by these compounds of formalin-induced tonic pain was approximately 60%, superior or compatible with that of the NMDA receptor antagonist MK-801 (approximately 50%), suggesting that 60% of formalin-induced tonic pain is DAO-sensitive, whereas formalin-induced acute nociception and the remaining 40% of tonic pain are DAO-insensitive. These results, combined with previous DAO gene mutation results, indicate spinal DAO mediates both induction and maintenance of formalin-induced tonic pain and further documents that spinal DAO is an efficacious target molecule for the treatment of chronic pain, including chronic neuropathic pain.
Participated in research design: Y. X. Wang.
Conducted experiments: Gong, Gao, Y. C. Wang, Li, and Huang.
Contributed new reagents or analytic tools: Hashimoto.
Performed data analysis: Gong, Gao, and Y. X. Wang.
Wrote or contributed to the writing of the manuscript: Y. X. Wang and Gong.
Other: Y. X. Wang acquired funding for the research.
We thank Dr. George Miljanich at Airmid Inc. (Redwood City, CA) for editing the manuscript.
This study was supported by the National Natural Science Foundation of China [Grants 81072623, 30973581]; the Mega New Drug Development Program of China [Grant 2009ZX09301-007]; and a Shanghai Jiao Tong University School of Pharmacy Predoctoral Fellowship (to N.G.).
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- d-amino acid oxidase
- rat spinal cord-derived DAO
- porcine kidney-derived DAO
- N-methyl-d-aspartic acid
- compound 8 (sc-203909)
- 4H-thieno[3,2-b]pyrrole-5-carboxylic acid
- compound 2
- 5-methylpyrazole-3-carboxylic acid
- 4-nitro-3-pyrazole carboxylic acid
- (+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine maleate
- dimethyl sulfoxide
- analysis of variance
- area under the curve
- maximum effect.
- Received July 6, 2010.
- Accepted October 14, 2010.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics