1521-0103/357/2/240–247$25.00 http://dx.doi.org/10.1124/jpet.115.230193THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS J Pharmacol Exp Ther 357:240–247, May 2016Copyright ª 2016 by The American Society for Pharmacology and Experimental Therapeutics

Repeated Dosing with NCX1404, a Nitric Oxide–DonatingPregabalin, Re-establishes Normal Nociceptive Responses inMice with Streptozotocin-Induced Painful Diabetic Neuropathy s

Katia Varani, Fabrizio Vincenzi, Martina Targa, Annalisa Ravani, Elena Bastia, Laura Storoni,Stefania Brambilla, Nicoletta Almirante and Francesco ImpagnatielloDepartment of Medical Sciences, Pharmacology Section, University of Ferrara, Ferrara, Italy (K.V., F.V., M.T., A.R.); NicoxResearch Institute, Bresso, Milano, Italy (E.B., L.S., S.B., N.A., F.I.)

Received November 6, 2015; accepted February 17, 2016

ABSTRACTNCX1404 [(3S)-5-methyl-3-(((1-(4-(nitrooxy)butanoyloxy)ethoxy)carbonylamino) methyl)hexanoic acid] is a novel nitric oxide(NO)–donating pregabalin that is readily absorbed and pro-cessed in vivo to pregabalin and NO. We determined theantiallodynic response of NCX1404 after acute or after 7, 14,and 21 days of repeated daily oral dosing in mice with strepto-zotocin (STZ)-induced painful diabetic neuropathy (PDN). Pre-gabalin and its combination with the NO donor isosorbidemononitrate (ISMN) were used for comparison. The blood levelsof pregabalin and nitrites, used as surrogate marker of NOrelease, after NCX1404 or pregabalin dosing were monitored inparallel experiments using liquid chromatography with tandemmass spectrometry (LC-MS/MS). NCX1404 and pregabalinresulted in similar pregabalin levels as it was their antiallo-dynic activity after acute dosing in STZ mice. However,NCX1404 resulted in disease-modifying properties when ad-ministered daily for 21 days, as indicated by the time- and

dose-dependent reversal of STZ-induced mechanical allodynia(paw withdrawal threshold [PWT]Veh_21d 5 1.3 6 0.15 g forvehicle; PWTNCX1404_21d 5 1.4 6 0.5 g, 2.9 6 0.2 g* and 4.1 60.2 g*, respectively for 19, 63, and 190 mmol/kg, oral gavage[PO] of NCX1404; *P , 0.05 versus vehicle). This effect was notshared by pregabalin at equimolar doses (190 mmol/kg, PO,PWTPregab_21d 5 1.4 6 0.1 g*, *P , 0.05 versus equimolarNCX1404). In addition, the NO donor ISMN (52.3 mmol/kg, PO)alone or combined with pregabalin (63 mmol/kg) was active at7 days (PWTVeh_7d 5 1.7 6 0.16 g; PWTISMN_7d 5 3.9 6 0.34 g*;PWTPregab_7d 5 1.36 0.07 g; PWTISMN1pregab_7d 5 3.86 0.29 g*;*P , 0.05) but not at later time points. The long-term effect ofNCX1404 was independent of residual drug exposure and lastedfor several days after the treatment was stopped. In summary, likepregabalin, NCX1404 is an effective antiallodynic agent. Differentlyfrom pregabalin, repeated dosing of NCX1404 re-establishednormal nociceptive responses in STZ-induced PDN in mice.

IntroductionNeuropathic pain is a form of chronic pain that can be

classified as peripheral or central. Peripheral neuropathicpain is caused by injury or infection of peripheral sensorynerves; central neuropathic pain is caused by damage to thecentral nervous system or/and the spinal cord. Both periph-eral and central neuropathic pain can occur without obviousinitial nerve damage. Various metabolic diseases causeneuropathic changes that later lead to neuropathic pain.An example is painful diabetic neuropathy (PDN), whichoccurs in a large number of patients with diabetes. Thepathophysiology of PDN remains unclear, but neurogenicand vascular components have been described (Yorek, 2003).For instance, peripheral demyelination, reduction in nerve

conductance, and progressive degeneration of peripheralsensory fibers (Watkins, 1984) along with microvascularand macrovascular abnormalities dependent on dysfunc-tional endothelium, the innermost layer of the blood vessel,have long been established in experimental animal modelsand in diabetic patients (Pieper and Gross, 1988; Pitei et al.,1997) as well as in individuals whowill later develop diabetes(Cosentino and Luscher, 1998).To date, the antidepressant duloxetine (King et al., 2015)

and the anticonvulsants gabapentin and pregabalin (Schreiberet al., 2015) are considered the first-line treatment strategies forpatients with various forms of neuropathic pain, includingPDN. These drugs are generally safe, but they do not entirelyrelieve pain symptoms, and a substantial subset of patients isvirtually insensitive to treatment. Furthermore, even whenthey work, these and other available treatments ameliorate thesymptoms but do not affect the underlying pathophysiology ofthe disease,making clinicalmanagementdifficult as the diseaseprogresses to its late stages.

dx.doi.org/10.1124/jpet.115.230193.s This article has supplemental material available at jpet.aspetjournals.org.This work was partially supported by Nicox Research Institute

ABBREVIATIONS: DMSO, dimethylsulfoxide; ISMN, isosorbide 5-mononitrate; LC-MS/MS, liquid chromatography with tandem mass spectrometry;NCX1236, 2-(1-(5-methyl-11-(nitrooxy)-3,7-dioxo-4,6,9-trioxa-2-azaundecyl)cyclohexyl)acetic acid; NCX1404, (3S)-5-methyl-3-(((1-(4-(nitrooxy)butanoyloxy)ethoxy)carbonylamino) methyl)hexanoic acid; NO, nitric oxide; PDN, painful diabetic neuropathy; PO, oral gavage; PWT, paw withdrawalthreshold; STZ, streptozotocin.

240

http://jpet.aspetjournals.org/content/suppl/2016/02/23/jpet.115.230193.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on February 1, 2021

jpet.aspetjournals.orgD

ownloaded from

Depending on the circulating concentration and thespecific tissue compartment where it is released, variousfunctions of nitric oxide (NO) ultimately lead to eitherpronociceptive or antinociceptive behaviors (Miclescu andGordh, 2009; Schmidtko et al., 2009). Low physiologicamounts of NO activate soluble guanylyl cyclase (causingvascular relaxation), favor neuronal transmission, andmodulate gene transcription through nuclear factor-kBsignaling (Colasanti and Suzuki, 2000); large quantities ofNO are thought to produce abnormal neuronal activityand central sensitization (Schmidtko et al., 2009), activateproinflammatory cytokines, and promote tissue-specificoxidative and nitrosative stress, which also are featuredin diabetic conditions (Pacher et al., 2007). Differentinvestigators (Sasaki et al., 1998; Rodella et al., 2000) havedemonstrated that impaired NO generation in diabeticrats results in hyperalgesia, and that the administrationof L-arginine, known to enhance the circulating NO concen-tration, is antinociceptive in diabetic mice. Pitei et al.(1997) showed that decreased NO production contributesto reduced endoneuronal blood flow in diabetic patients andcauses pain, whereas the local application of NO donorssuch as isosorbide dinitrate (Yuen et al., 2002) or nitroglyc-erin (Francis et al., 1977; Agrawal et al., 2007) relieves painand burning sensations in patients with PDN.Early exploratory work from our laboratories documented

the ability of NO to enhance the effect of gabapentin inneuropathic pain models (Wu et al., 2004). Additional evi-dence of NO-mediated effects in neuropathic pain modelswas later obtained with compounds possessing both antioxi-dant and NO-releasing properties (Ronchetti et al., 2009).Other investigators reported the beneficial effect of NO whencombined with gabapentin despite the different experimentalconditions employed (Curros-Criado and Herrero, 2009).Similarly, Huang et al. (2015) documented the analgesiceffects of an NO-donating derivative in a rat model of PDN,and Otari and Upasani (2015) reported on the involvement ofNO-cGMP signaling in neuropathic pain processing.In the present study, we investigated the antiallodynic

effects of the novel NO-donating pregabalin derivativeNCX1404, evaluating a single treatment schedule. Further-more, we addressed whether repeated daily dosing withNCX1404 modulates the development and maintenance ofmechanical allodynia consequent to streptozotocin (STZ). Inall experiments, the NO donor isosorbide 5-mononitrate(ISMN) and its combination with pregabalin were used forcomparison.

Materials and MethodsIn all experiments, the animals were cared for and treated in

accordance with protocols approved by the institutional animalcare committees, and all efforts were made to limit the number ofanimals and to minimize animal suffering. Male CD1mice, weighing20 to 30 g, were kept under standard temperature conditions (22–24°C)and illumination (12-hour light/dark cycle). They were allowed toadjust to this environment in cages with mesh bottoms and weregiven free access to fresh water and food for at least 7 days before theexperiments began.

NCX1404 [(3S)-5-methyl-3-(((1-(4-(nitrooxy)butanoyloxy)ethoxy)carbonylamino) methyl)hexanoic acid] is a nitrooxy derivative ofpregabalin synthesized in our laboratories in eight steps, with 12%overall yield. The compound was obtained by coupling pregabalin

and 1-{[(4-nitrophenoxy)carbonyl]oxy}ethyl 4-(nitrooxy)butanoate(intermediate 9 in Fig. 1) that was prepared in seven steps from4-bromobutanoic acid (intermediate 1 in Fig.1). To obtain the corre-sponding 15N-labeled NCX1404, 15N-AgNO3 was used instead ofAgNO3 in step b reported in Fig. 1.

15N-ISMNwas synthesized in house by adding 15N-Bu4NNO3, 2,6-di-tert-butyl-4- methyl pyridine, and trifluoromethane sulfonic anhy-dride to a solution of isosorbide in dichloromethane. 15N-ISMN wasthen isolated by flash chromatography.

Other chemicals and reagents were purchased from Sigma-Aldrich(St. Louis, MO) unless otherwise specified.

STZ-Induced Diabetes Mice Model

Diabetes was induced by injecting 200 mg/kg i.p. of STZ (Kowaluket al., 2000; Vincenzi et al., 2014). Blood glucose levels were assessedfrom day 2 until the end of the experiment using an Accu-Chek bloodglucose monitoring system (Roche Applied Science, Indianapolis, IN).A single administration of STZ induced insulin-dependent diabetesmellitus within 24 to 48 hours by destruction of pancreatic islet cells(Schenkman, 1991). Plasma glucose levels above 300 mg/dl wereconsidered indicative of diabetes (Migita et al., 2009).

Mechanical Allodynia Paw Withdrawal Test

Paw withdrawal thresholds (PWT) were determined using theDynamic Plantar Aesthesiometer (Ugo Basile, Varese, Italy), anapparatus that generates a mechanical force linearly increasing withtime (Szolcsanyi et al., 2004; Wright et al., 2007). Mice were placedindividually in plastic cages with a wire-mesh bottom and allowed toacclimatize for at least 2 hours. Increasing mechanical stimulation(0.25 g/s, cutoff force: 10 g) was applied to the plantar surface of a hindpaw. The nociceptive threshold was defined as the force, in grams, atwhich the mouse withdraws its paw. When a withdrawal responseoccurred, the stimulus was terminated, and the response thresholdwas measured electronically. Data are reported as the mean6 S.E.M.of the threshold recorded at each trial.

Drug Treatment Paradigm

Acute Schedule. On day 14 after STZ challenge, mice weretreated orally with a single challenge of drugs at the indicated dosesor vehicle (Hypromellose 0.5%, dimethylsulfoxide [DMSO] 2%).Mechanical allodynia, expressed as paw withdrawal threshold (g)was evaluated at 0, 5, 30, 45, 60, 90, 120, and 180 minutes aftertreatment.

Repeated Treatment Schedule. On day 7 after STZ challenge,mice were treated orally each day with the selected drugs at theindicated doses or vehicle (Hypromellose 0.5%, DMSO 2%) for 21 days.Mechanical allodynia expressed as paw withdrawal threshold (g),was measured on days 7, 14 and 21 in animals kept drug-free for 18to 24 hours. In addition, changes in the mechanical threshold werecontinued to bemonitored in all groups at 3, 7, and 10 days after dosingwas discontinued.

Analytical Assessments

Parallel experiments were conducted to monitor the blood levelsof NCX1404 and pregabalin after acute administration of NCX1404or pregabalin at 63 mmol/kg, administered by oral gavage (PO). Forthe acute schedule experiments, blood samples were collected fromthe left ventricle before (time 0) and at 5, 15, 30, 60, 120, 240, and 360minutes after dosing in anesthetized mice (n5 3 for each time point).The blood samples were later processed for further analytical assess-ments. 15N-nitrites, used as surrogate marker of NO release, weremonitored at the same time points as indicated above in animalstreated with 15N-NCX1404 at the dose of 63 mmol/kg, PO. For therepeated dosing experiments, blood samples were withdrawn 18 to24 hours after 7, 14, or 21 days of consecutive daily treatment with

NCX1404 Reduces Mechanical Allodynia Development in STZ Mice 241

at ASPE

T Journals on February 1, 2021

jpet.aspetjournals.orgD

ownloaded from

63 mmol/kg, PO, of pregabalin or NCX1404 and later processed tomonitor the levels of pregabalin in blood.

NCX1404 and Pregabalin Quantitation. Aliquots (50 ml) ofblood were added with 10 ml of 100% DMSO, 10 ml H2O, and 500 mlmethanol and then centrifuged at 4,000 rpm for 10 minutes at 4°C.The supernatants were collected and stored at 280°C and lateranalyzed by a liquid chromatography with tandemmass spectrometry(LC-MS/MS) method validated for simultaneous determination ofpregabalin and NCX1404. The systemwas an Acquity UPLC (Waters,Milford, MA). The detector was a Quattro microTM API LC-MS/MS(Waters). The conditions used to analyze NCX1404 and pregabalinwere: column, Acquity BEH phenyl 50 � 2.1 mm (5 mm) at 40°C;mobile phases were, A, water 1 0.1% formic acid and, B, methanol 10.1% formic acid; flow rate, 0.5 ml/min; detector wavelength, 210 nm;multiple reaction monitoring transitions in electrospray ionizationpositive were m/z 400.92 → 325.07 for NCX1404 and m/z 160.2 → 97.1for pregabalin. The raw data (average peak area at each time point)were interpolated using Quanlynx version 4.1 software (Waters) toobtain the concentration of compounds at each time point.

15N-Nitrites. Blood samples were taken from the left ventriclebefore (time 0) and at 5, 15, 30, 60, 120, 240 and 360 minutes after thedose; the blood was then extracted with acetonitrile. The quantifica-tion is based on the conversion of 15N-nitrites to 15N-naphtotriazoleusing 1.2-diaminonaphtalene in acidic conditions, followed by quan-tification by LC-MS/MS.

Briefly, the system used was an Agilent 1200 HPLC (AgilentTechnologies, Santa Clara, CA). The detector was an API4000 massspectrometer (Applied Biosystems, Foster City, CA). The 15N-nitritesconversion to naphtotriazole was performed by mixing 200 ml ofdeproteinized blood sample spiked with 10 ml of 1.2-diaminonaphtalene

(1 mM) in HCl 0.5 M, which was shaken for 20 minutes at roomtemperature. At the end, to stop the conversion, 10 ml of NaOH 1 Mand 10 ml of DMSO were added to all the samples. The resultingcompound was then separated from the mixture using an ACE C18AR 50 � 2.0 mm (3 mm) column. The mobile phase were (A) water/0.1% formic acid and, (B) acetonitrile/0.1% formic acid. The multiplereaction monitoring transition in electrospray ionization positive for15N-naphtotriazole was m/z 171.1 → 115.0. The raw data (averagepeak area at each time point) were interpolated on a standard curveto obtain the concentration of 15N-naphtotriazole at each time point.

Statistics

In all experiments, comparison between treatments groups wasperformed by two-way analysis of variance followed by a Bonferronimultiple comparison test unless otherwise specified. P,0.05 wasconsidered statistically significant.

ResultsBlood Glucose Levels and Mechanical Allodynic

Threshold after STZ Treatment in CD-1 Mice. Physio-logic glucose plasma levels in CD-1mice were 1496 4mg/dl.The administration of STZ resulted in a rapid increase inplasma glucose levels starting from the next day, and thelevels remained elevated throughout the weeks, reaching515 6 17 mg/dl at 4 weeks after treatment (Table 1).The injection of STZ also significantly diminished the PWT

tomechanical stimulation starting 14 days after STZ injection

Fig. 1. Synthesis of NCX1404. (a) SOCl2, MeOH, room temperature, overnight, 85%. (b) AgNO3, CH3CN, reflux, 4 hours 93%. (c) LiOH 2 N, MeOH, 5°C,overnight, 84%. (d) Ag2O, CH3CN / H2O, room temperature, 3 hours, 70%. (e) 1-Chloroethyl chloroformate, Py, CH2Cl2, r.t., overnight, 97%.(f) NaI, CH3CN, 53%. (g) Dry toluene, room temperature, overnight, 71%. (h) Pregabalin, triethylamine, trimethylsilylchloride, CH2Cl2, roomtemperature, overnight. Numbers in brackets identify the intermediates obtained during the synthesis of NCX1404.

242 Varani et al.

at ASPE

T Journals on February 1, 2021

jpet.aspetjournals.orgD

ownloaded from

(PWT, 4.2 6 0.31 and 1.3 6 0.44 g, vehicle and STZ,respectively). The allodynic response remained stable overthe next 2 weeks (Table 1).NCX1404, Pregabalin Blood Levels after NCX1404

and Pregabalin in Mice. The blood levels of NCX1404 inmice after 63 mmol/kg, PO (23.7 mg/kg) were below the limit ofquantitation at all time points tested. NCX1404 is a nitro-oxyester derivative of pregabalin, so we compared the extent ofcirculating pregabalin as well as that of nitrites (serving as asurrogate marker of NO release) after NCX1404 (63 mmol/kg,PO) or equimolar pregabalin (63 mmol/kg, PO) single dosing.Parallel experiments were performed using NCX1404 labeledon the nitrate group with 15N to study the levels of nitritesreleased into the blood stream after NCX1404 single dosing.Pregabalin levels after NCX1404 single dosing reached aCmax

of 6.8 6 0.6 mg/ml 30 minutes after NCX1404 administrationand declined in the next hours to reach levels close to baselineafter 360 minutes (Fig. 2A). Pregabalin administered undersimilar conditions resulted in similar pregabalin exposurecompared with that found after NCX1404 (Fig. 2A). Therelative pregabalin bioavailability after NCX1404 comparedwith pregabalin was 85%.15N-Nitrites Blood Levels after NCX1404 and ISMN

in Mice. 15N-nitrites blood levels after NCX1404 oral treat-ment (63 mmol/kg, PO) reached a maximum 5 minutes afterthe animals had been dosed and rapidly declined to basallevels thereafter (Fig. 2B). Conversely, 15N-nitrites afterISMN oral dosing (52.3 mmol/kg, PO) in the blood increasedslowly to reach maximum at 60 minutes (Cmax 5 1.77 60.4mM); they slowly decayed thereafter to reach basal valuesduring the next 4 hours (Fig. 2B).Antiallodynic Effects of Acute Dosing of NCX1404 or

Pregabalin in STZ-Induced PDN Mouse Model. Asshown in Fig. 3A, acute administration of NCX1404 at 19,63, and 190 mmol/kg, PO, (7.1, 23.7, and 71 mg/kg) reducedmechanical allodynia in a time- and dose-dependent fashion(PWTNCX1404_60 min5 2.66 0.2 g, 3.96 0.3 g, and 4.66 0.2g for19, 63, and 190 mmol/kg, PO, of NCX1404, respectively) versusvehicle (PWTVehicle_60 min 5 1.3 6 0.2 g) in STZ-induced PDNmice model. The administration of pregabalin at equimolardoses (3, 10, and 30 mg/kg) also reduced mechanical allodyniain a time- and dose-dependent fashion (PWTPregab_60 min 5 2.86 0.1 g, 3.8 6 0.3 g, 4.3 6 0.4 g after 19, 63, and 190 mmol/kg,PO, of pregabalin, respectively), reaching values close tothose observed in naïve animals (Fig. 3B). Consistent with

pregabalin exposure after NCX1404 or pregabalin oral treat-ment, the antiallodynic responses of both drugs were maximalbetween 45 and 60 minutes after dosing and slowly disap-peared thereafter to reach basal values at 120 minutes (Fig. 3,A and B).Reversal of the Allodynic Response after 21-Day

Repeated Dosing with NCX1404 in STZ-Induced PDNMouse Model. These experiments were undertaken to ad-dress whether repeated daily administration of NCX1404 orpregabalin can re-establish normal nociceptive responses tomechanical stimulation in a STZ-induced PDN mice model.The compounds were administered by oral gavage starting1 week after STZ treatment whereas testing was performed at7, 14, and 21 days, 18 to 24 hours after the respective dailytreatment to ensure the absence of circulating pregabalin,which was then confirmed in parallel pharmacokinetic exper-iments by the absence of detectable levels of pregabalin in

TABLE 1Time course changes in blood glucose level and mechanical allodynicthreshold in mice injected with streptozotocin (STZ)Streptozotocin (200 mg/kg i.p., n = 10) or vehicle (distilled water, n = 10) wereadministered by intraperitoneal injection at 5 ml/kg. The data are expressed as mean6 S.E.M.

TimeBlood Glucose Levels Paw Withdrawal Threshold

STZ Treated Vehicle STZ Treated

days mg/ml g

0 149.4 6 4.3 4.1 6 0.22 4.3 6 0.212 386.5 6 20.1* — —7 476.2 6 18.2* 4.0 6 0.23 2.1 6 0.2614 496.0 6 11.1* 4.2 6 0.31 1.3 6 0.44*21 529.3 6 12.7* 4.3 6 0.22 1.2 6 0.16*28 514.6 6 17.4* 4.4 6 0.21 1.2 6 0.33*

* P , 0.05 versus time 0 and vehicle at the respective time points.

Fig. 2. (A) Pregabalin exposure after single oral administration ofNCX1404 or pregabalin to mice. NCX1404 (d, n = 3) and pregabalin (s,n = 3) were administered at 63 mmol/kg orally (23.7 and 10 mg/kg forNCX1404 and pregabalin, respectively) dissolved in vehicle (Hypromellose0.5%, DMSO 2%). Both drugs were given by oral gavage at 5 ml/kg. (B)15N-nitrites were monitored after the administration of NCX1404 (63mmol/kg,d, n = 3) labeled with 15N given at the same dose and conditionsas previously for ISMN (52.3 mmol/kg, u, n = 3) labeled with 15N.

NCX1404 Reduces Mechanical Allodynia Development in STZ Mice 243

at ASPE

T Journals on February 1, 2021

jpet.aspetjournals.orgD

ownloaded from

blood samples collected at the same time points (data notshown). Furthermore, the duration of the effect wasmonitoredby assessing the mechanical threshold at 1, 3, 7, and 10 daysafter the 21 days of repeated daily dosing had been ended.Similar treatment paradigm with pregabalin at equimolardoses was used for comparison.

The repeated oral administration of 19 mmol/kg (7.1 mg/kg),63 mmol/kg (23.7 mg/kg), and 190 mmol/kg (71 mg/kg) ofNCX1404 resulted in dose- and time-dependent enhancementof the mechanical threshold, which wasmaximal after 21 days(Fig. 4A). The effect of NCX1404 was retained for several daysafter ending the daily treatment, as indicated by the highmechanical threshold still recorded 3 and 7 days after thetreatment was completed (Fig. 4A). Regardless of the dose

Fig. 3. Dose-dependent antiallodynic effect of NCX1404 (A) andpregabalin (B) in STZ-diabetic mice. Both drugs were administereddissolved in vehicle (Hypromellose 0.5%, DMSO 2%, , n = 6). NCX1404 wasgiven by oral gavage (5 ml/kg) at 19mmol/kg (7.1 mg/kg, ♦, n = 6), 63mmol/kg(23.7 mg/kg,d, n = 6), and 190 mmol/kg (71mg/kg,j, n = 6). Pregabalin wasadministered at equimolar doses as NCX1404 (3 mg/kg, e, n = 6; 10 mg/kg,s, n = 6; 30 mg/kg, u, n = 6). Paw withdrawal threshold after mechanicalstimulation allodynia was monitored at 0 (basal), 15, 30, 45, 60, 90, 120,and 180 minutes after treatment. *P , 0.05 versus vehicle, Two-wayanalysis of variance followed by Bonferroni’s multiple comparisons test(NCX1404 19 mmol/kg at 45 minutes; NCX1404 63 mmol/kg from 30 to90 minutes; NCX1404 190 mmol/kg from 15 to 90 minutes; pregabalin19 mmol/kg from 45 to 60 minutes; pregabalin 63 mmol/kg from 30 to90 minutes; pregabalin 190 mmol/kg from 30 to 90 minutes).

Fig. 4. Dose-dependent effects of repeated daily oral dosing with of NCX1404(A) or pregabalin (B) in STZ-diabetic mice. Both drugs were administereddissolved in vehicle (Hypromellose 0.5%, DMSO 2%, gray triangle, n = 6).NCX1404was given by oral gavage (5ml/kg) at 19mmol/kg (7.1mg/kg,♦,n=6),63 mmol/kg (23.7 mg/kg, d, n = 6), and 190 mmol/kg (71 mg/kg, j, n = 6).Pregabalin was administered at equimolar doses as NCX1404 (3 mg/kg,e, n = 6; 10 mg/kg,s, n = 6; 30 mg/kg,u, n = 6). All treatments were givendaily between 9 and 10 AM for 21 days starting 7 days after the initial STZchallenge. All measurements ofmechanical allodynia were performed in drug-free conditions. The later aspect was verified by confirming the absence of adetectable concentration of pregabalin in blood sampleswithdrawn at the timethe experiment was performed. *P , 0.05 versus vehicle, two-way analysisof variance followed by Bonferroni’s multiple comparisons test (NCX140463 mmol/kg from 7 to 21 days; NCX1404 190 mmol/kg from 7 to 28 days).

244 Varani et al.

at ASPE

T Journals on February 1, 2021

jpet.aspetjournals.orgD

ownloaded from

used, this effect was not shared by pregabalin administeredat equimolar doses under the same experimental conditions(Fig. 4B), which suggests that these effects likely depend onthe release of NO.Antiallodynic Response after Acute or 21-Day Re-

peated Dosing with the NO Donor ISMN, pregabalin, orTheir Combination in STZ-Induced PDN Mouse Model.We tested whether ISMN per se or combined with pregabalinexerts any antiallodynic effect on STZ-induced PDN micemodel. In exploratory experiments, ISMN was tested at 15.7mmol/kg, PO (3 mg/kg), 52.3 mmol/kg, PO (10 mg/kg), and 157mmol/kg, PO (30 mg/kg). In these experiments, 15.7 mmol/kgdid not modify the allodynic threshold to mechanical stimu-lation in STZ-treated animals whereas the high dose of157 mmol/kg resulted in significant antiallodynic activitycombined with robust cardiovascular hypotension (data notshown). Conversely, the administration of ISMN at 52.3mmol/kg resulted in significant antiallodynic effects withonly moderate hypotension, which was transient in nature;thus, this dose was selected for further experiments. Ani-mals treated with 52.3 mmol/kg, PO, of ISMN showed a mildantiallodynic response, as indicated by the increase in PWTat selected time points after dosing (Fig. 5). Likewise,administration of 63 mmol/kg, PO (10 mg/kg), of pregabalinalso elicited mild to moderate antiallodynic effects, as inprevious experiments (Fig. 5). However, the concomitant oraladministration of ISMN (52.3 mmol/kg, PO) and pregabalin(63 mmol/kg, PO) was marginally as effective as either druggiven alone.To address whether STZ-induced mechanical allodynia

was influenced by ISMN, we then monitored the allodynicresponse after 7, 14, and 21 days of daily oral treatment withISMN (52.3 mmol/kg, PO) in animals kept treatment-free

for 18 to 24 hours before testing to ensure the absence ofcirculating NO during the recording of PWT. In these ex-perimental sessions, pregabalin (63 mmol/kg, PO) alone orcombined with ISMN (52.3 mmol/kg, PO) administered once aday for 21 consecutive days was used for comparison.Repeated daily dosing with 52.3 mmol/kg, PO (10 mg/kg),

of ISMN reduced the development of mechanical allodyniain mice treated with STZ 1 week earlier. In particular, theeffect was evident after 7 days of repeated daily treatmentbut disappeared during the ensuing days (Fig. 6). Similarlytransient was the effect recorded after the coadministrationof pregabalin (63 mmol/kg, PO) and ISMN (52.3 mmol/kg,PO) (Fig. 6).

DiscussionNCX1404 is a dual-acting NO-donating derivative of

pregabalin featuring a mononitrate alkyl chain bound to theamino function of pregabalin by an acyloxy-alkyl carbamate.This novel compound is designed to combine the NO-mediatedanti-inflammatory and antioxidant properties with those ofpregabalin, which is known to reduce the symptoms of painwhile lacking any measurable effect on the underliningpathophysiology.The administration of STZ to mice results in sustained

diabetes (Kowaluk et al., 2000) and a measurable allodynicresponse to mechanical stimulation. These effects are thoughtto be consequent to the progressive degeneration of peripheralsensory fibers (Watkins, 1984) due to dysfunctional vascularendothelium and reduced NO availability (Yorek, 2003). TheSTZ model is thus particularly interesting to address theacute antiallodynic effects as well as long-term changes ofthe nociceptive circuits associated with the administrationof NO-donating drugs such as NCX1404.In our study, mice receiving a single challenge of STZ rapidly

increased their plasma glucose level and later developed robustallodynic response to mechanical stimulation, which was max-imal within 7 days and remained stable over several weeks. Forthis reason, day 14 after STZ treatment was selected to test theacute antiallodynic response of NCX1404 versus pregabalin,and 7 days after STZ was used as starting point for repeateddosing to address the effects of NCX1404 on the development ofallodynia in diabetic mice.NCX1404 is readily absorbed after oral dosing in STZ mice

and efficiently cleaved in vivo to release pregabalin as forother molecules carrying similar approach (Cundy et al.,2004). In addition, NCX1404 also releases relevant amountsof NO as determined by the presence of 15N-nitrites inblood of animals dosed orally with 15N-labeled NCX1404. Theantiallodynic effect of NCX1404 after a single oral challengein STZ-induced diabetic neuropathic mice model was time anddose dependent. Similarly, equimolar doses of pregabalin alsoresulted in time- and dose-dependent antiallodynic effects thatwere overlapping with that elicited by the respective equimolardose of NCX1404. Furthermore, the antiallodynic effects of bothNCX1404 and pregabalin after single acute oral dosing werecoincident with the respective time course of pregabalin bloodexposure. These data indicate that the acute antiallodyniceffects of NCX1404 are mostly dependent on pregabalinrelease with minor, if any, contribution of NO.Alternatively, the absence of superiority may be due to

partially overlappingmechanisms for pregabalin andNO. In a

Fig. 5. Acute antiallodynic effect of ISMN alone or combined withpregabalin in STZ-diabetic mice. ISMN (52.3 mmol/kg, PO equal to 10mg/kg, cross, n = 6), pregabalin (63 mmol/kg, PO equal to 10 mg/kg, s,n = 6), or their combination (., n = 6) were administered orally (5 ml/kg)dissolved in vehicle (Hypromellose 0.5%, DMSO 2%, gray triangle, n = 6).Mechanical allodynia was monitored at 0 (basal), 15, 30, 45, 60, 90, and120 minutes after dosing. *P , 0.05 versus vehicle, two-way analysisof variance followed by Bonferroni’s multiple comparisons test (ISMN52.3 mmol/kg from 15 to 45 minutes; pregabalin 63 mmol/kg from 30 to45 minutes).

NCX1404 Reduces Mechanical Allodynia Development in STZ Mice 245

at ASPE

T Journals on February 1, 2021

jpet.aspetjournals.orgD

ownloaded from

different series of experiments, the single oral administrationof the NO-donor ISMN resulted in modest antiallodynic effectafter mechanical stimulation that was similar to that of asubmaximal dose (10 mg/kg) of pregabalin. Moreover, theircoadministration (concomitant pregabalin and ISMN dosing)was only as effective as the individual compounds given alone,consistent with the concept that the acute antiallodynic effectof NO and pregabalin could be due, at least partially, tooverlapping mechanisms.Pregabalin mainly alleviates neuropathic pain symptoms

by impairing the trafficking of a2d-1 receptors, subsequentlyreducing Ca21 influx and spinal sensitization (Sills, 2006).However, additional mechanisms involving the activation ofNO/cGMP signaling pathway have recently been described forgabapentin, another a2d-1 antagonist (Mixcoatl-Zecuatl et al.,2006; Takasu et al., 2006). Whether similar mechanisms alsotake place for pregabalin remain to be established.The main observation made in our present study was,

however, the significant time- and dose-dependent long-termchanges of the nociceptive circuits exhibited byNCX1404 afterrepeated daily dosing in mice with established neuropathyconsequent to the administration of STZ 1-week earlier. Theenhancement of the nociceptive threshold after NCX1404long-term dosing reflects changes in the nociceptive circuitsrather than a direct antiallodynic effect of the drug and itsactive metabolites because no detectable levels of pregabalinand NO were measured at the time (18 to 24 hours after thelast daily dose) the behavioral assessments were taken. Thiseffect appears to be unique to NO-donating drugs such asNCX1404; pregabalin, administered under the same experi-mental conditions, did not result in measurable activity.

A similar trend, albeit of different intensity, was observedwhen the NO-donating derivative of gabapentin (NCX1236)was administered under the same conditions (SupplementalFig. 1) suggesting the importance of NO release in mediatingthis effect. Moreover, ISMN administered alone or combinedwith pregabalin was as effective as NCX1404 on day 7 butnot on days 14 and 21, a time at which ISMN-mediatedtolerance occurs (Daiber and Munzel, 2015). These latterresults further indicate the importance of NO release in theallodynic responses elicited by STZ after repeated dosingwith NCX1404 and, at the same time, suggest potentialdifferences between NCX1404 and ISMN with respect totheir tolerance liability, ability to efficiently release NO overrepeated dosing and/or efficiency to activate the NO/cGMPsignaling pathway (Klemenska and Beresewicz, 2009).NCX1404 releases pregabalin and the nitrate linker via the

action of esterases (data not shown); however, we still lackinformation on how the nitrate linker, once cleaved frompregabalin, releases NO. In our experiments, the kinetics of15N-nitrites, used as a surrogate marker of NO release,differed significantly after 15N-NCX1404 and 15N-ISMNdosing. In spite of a similar Cmax between the two compounds,15N-nitrites accumulation was rapid (Tmax 5 minutes) anddecayed quickly after 15N-NCX1404. Conversely, 15N-nitritesmeasured after 15N-ISMN were largely superior comparedwith that from 15N-NCX1404 with a Tmax at 60 minutes.Furthermore, the bioactivation and NO release from ISMNoccur via the P450 enzyme that is down-regulated whenrepeatedly stimulated (Minamiyama et al., 2004). This mech-anism, along with soluble guanylyl cyclase desensitization andactivation of oxidative stress, has been described as mediatingISMN-induced tolerance (Minamiyama et al., 2004; Daiberand Munzel, 2015). However, this mechanism may not be asimportant for NCX1404 because similar NO-donating com-pounds were reported to rely on different bioactivation path-ways compared with ISMN (Govoni et al., 2006).The effect of NCX1404 is independent from any residual

drug as pregabalin levels in blood samples withdrawn at thetime of behavioral assessments were found below the limit ofquantitation (data not shown). Furthermore, the reduction ofmechanical allodynia in animals treated with NCX1404 for21 days was still evident 3 and 7 days after the treatment wasended,making it unlikely that residual compound accounts forthe observed activity. NCX1404 did not affect plasma glycemiclevels (data not shown) in these animals, explaining why, soonafter the treatment was completed, the allodynic thresholdreturned to levels similar to that of vehicle with a time courseresembling that observed after the initial injection of STZ.The molecular and cellular mechanisms underlying the

long-term effects observed after NCX1404 repeated dosing arefar from clear. Neurogenic and vascular changes are involvedin the pathophysiology of PDN (Yorek, 2003). These includethe release of various inflammatory mediators (Purwata,2011), vascular endothelial cell dysfunction, and progressivereduction of NO, which is thought to play a pivotal role in thepathophysiology of PDN (Pieper and Gross, 1988). NCX1404releases biologically relevant amounts of NO as detected bythe blood increase in 15N-nitrites. Thus, the circulating NObrought about by NCX1404 might compensate for the reducedendogenous NO production consequent to endothelial celldysfunction in STZ-treated animals. Likewise, circulatingNO from NCX1404 might inhibit inflammatory cytokine

Fig. 6. Repeated daily oral dosing with of ISMN, pregabalin, or theircombination in STZ-diabetic mice. ISMN (52.3 mmol/kg, PO, equal to 10mg/kg,3, n = 6), pregabalin (63 mmol/kg, PO, equal to 10 mg/kg, s, n = 6),or their combination (., n = 6) were administered daily dissolved invehicle (Hypromellose 0.5%, DMSO 2%, gray triangle, n = 6) between days1 and 21 starting 7 days after the STZ challenge. All measurements ofmechanical allodynia were performed in drug-free conditions immediatelybefore the treatment of the respective day. *P , 0.05 versus vehicle,two-way analysis of variance followed by Bonferroni’s multiple com-parisons test (ISMN 52.3 mmol/kg from 7 to 14 days; ISMN + pregabalinfrom 7 to 14 days).

246 Varani et al.

at ASPE

T Journals on February 1, 2021

jpet.aspetjournals.orgD

ownloaded from

release. Consistent with this interpretation, previous stud-ies have documented the ability of NO donors to ameliorateendothelial function in rodent models of diabetes (Pieperand Gross, 1988; Ambrosini et al., 2005; Huang et al., 2015)and to reduce inflammatory cytokines viamodulation of nuclearfactor-kB translocation (Colasanti and Suzuki, 2000; Ronchettiet al., 2006).In summary, acute dosing of NCX1404 is as effective as acute

dosing of pregabalin to reverse mechanical allodynia in STZ-diabetic mice. However, NCX1404, differently from pregabalin,when administered over 21 days re-establishes normal noci-ceptive responses in STZ-induced PDN in mice. This effect islikely due to a mechanism involving the release of NO.

Authorship Contributions

Participated in research design: Varani, Impagnatiello.Conducted experiments: Vincenzi, Targa, Ravani, Bastia, Brambilla.Contributed new reagents or analytic tools: Almirante, Storoni.Performed data analysis: Bastia, Impagnatiello, Varani.Wrote or contributed to the writing of the manuscript: Impagnatiello,

Bastia.

References

Agrawal RP, Choudhary R, Sharma P, Sharma S, Beniwal R, Kaswan K, and KocharDK (2007) Glyceryl trinitrate spray in the management of painful diabetic neu-ropathy: a randomized double blind placebo controlled cross-over study. DiabetesRes Clin Pract 77:161–167.

Ambrosini MV, Mariucci G, Rambotti MG, Tantucci M, Covarelli C, De Angelis L,and Del Soldato P (2005) Ultrastructural investigations on protective effects ofNCX 4016 (nitroaspirin) on macrovascular endothelium in diabetic Wistar rats.J Submicrosc Cytol Pathol 37:205–213.

Colasanti M and Suzuki H (2000) The dual personality of NO. Trends Pharmacol Sci21:249–252.

Cosentino F and Lüscher TF (1998) Endothelial dysfunction in diabetes mellitus.J Cardiovasc Pharmacol 32 (Suppl 3):S54–S61.

Cundy KC, Branch R, Chernov-Rogan T, Dias T, Estrada T, Hold K, Koller K,Liu X, Mann A, and Panuwat M, et al. (2004) XP13512 [(1/2)-1-([(alpha-isobutanoyloxyethoxy)carbonyl] aminomethyl)-1-cyclohexane acetic acid], a novelgabapentin prodrug: I. Design, synthesis, enzymatic conversion to gabapentin, andtransport by intestinal solute transporters. J Pharmacol Exp Ther 311:315–323.

Curros-Criado MM and Herrero JF (2009) Antinociceptive effects of NCX-701 (nitro-paracetamol) in neuropathic rats: enhancement of antinociception by co-administration with gabapentin. Br J Pharmacol 158:601–609.

Daiber A and Münzel T (2015) Organic nitrate therapy, nitrate tolerance, andnitrate-induced endothelial dysfunction: emphasis on redox biology and oxidativestress. Antioxid Redox Signal 23:899–942.

Francis DR, Hubbard ER, and Johnson LE (1977) Nitroglycerin ointment as avasodilator in the lower extremities. J Am Podiatry Assoc 67:874–879.

Govoni M, Casagrande S, Maucci R, Chiroli V, and Tocchetti P (2006) In vitrometabolism of (nitrooxy)butyl ester nitric oxide-releasing compounds: comparisonwith glyceryl trinitrate. J Pharmacol Exp Ther 317:752–761.

Huang LY, Tsui DY, Williams CM, Wyse BD, and Smith MT (2015) The furoxannitric oxide donor, PRG150, evokes dose-dependent analgesia in a rat model ofpainful diabetic neuropathy. Clin Exp Pharmacol Physiol 42:921–929.

King JB, Schauerhamer MB, and Bellows BK (2015) A review of the clinical utility ofduloxetine in the treatment of diabetic peripheral neuropathic pain. Ther Clin RiskManag 11:1163–1175.

Klemenska E and Beresewicz A (2009) Bioactivation of organic nitrates and themechanism of nitrate tolerance. Cardiol J 16:11–19.

Kowaluk EA, Mikusa J, Wismer CT, Zhu CZ, Schweitzer E, Lynch JJ, Lee CH,Jiang M, Bhagwat SS, and Gomtsyan A, et al. (2000) ABT-702 (4-amino-5-(3-bromophenyl)-7-(6-morpholino-pyridin- 3-yl)pyrido[2,3-d]pyrimidine), a novelorally effective adenosine kinase inhibitor with analgesic and anti-inflammatoryproperties. II. In vivo characterization in the rat. J Pharmacol Exp Ther 295:1165–1174.

Miclescu A and Gordh T (2009) Nitric oxide and pain: ‘Something old, somethingnew’. Acta Anaesthesiol Scand 53:1107–1120.

Migita K, Moriyama T, Koguchi M, Honda K, Katsuragi T, Takano Y, and Ueno S(2009) Modulation of P2X receptors in dorsal root ganglion neurons ofstreptozotocin-induced diabetic neuropathy. Neurosci Lett 452:200–203.

Minamiyama Y, Takemura S, Yamasaki K, Hai S, Hirohashi K, Funae Y, and OkadaS (2004) Continuous administration of organic nitrate decreases hepatic cyto-chrome P450. J Pharmacol Exp Ther 308:729–735.

Mixcoatl-Zecuatl T, Flores-Murrieta FJ, and Granados-Soto V (2006) The nitricoxide-cyclic GMP-protein kinase G-K1 channel pathway participates in theantiallodynic effect of spinal gabapentin. Eur J Pharmacol 531:87–95.

Otari KV and Upasani CD (2015) Involvement of NO-cGMP pathway in anti-hyperalgesic effect of PDE5 inhibitor tadalafil in experimental hyperalgesia.Inflammopharmacology 23:187–194.

Pacher P, Beckman JS, and Liaudet L (2007) Nitric oxide and peroxynitrite in healthand disease. Physiol Rev 87:315–424.

Pieper GM and Gross GJ (1988) Oxygen free radicals abolish endothelium-dependentrelaxation in diabetic rat aorta. Am J Physiol 255:H825–H833.

Pitei DL, Watkins PJ, and Edmonds ME (1997) NO-dependent smooth musclevasodilatation is reduced in NIDDM patients with peripheral sensory neuropathy.Diabet Med 14:284–290.

Purwata TE (2011) High TNF-alpha plasma levels and macrophages iNOS and TNF-alpha expression as risk factors for painful diabetic neuropathy. J Pain Res 4:169–175.

Rodella L, Rezzani R, Corsetti G, and Bianchi R (2000) Nitric oxide involvement inthe trigeminal hyperalgesia in diabetic rats. Brain Res 865:112–115.

Ronchetti D, Borghi V, Gaitan G, Herrero JF, and Impagnatiello F (2009) NCX 2057,a novel NO-releasing derivative of ferulic acid, suppresses inflammatory and no-ciceptive responses in in vitro and in vivo models. Br J Pharmacol 158:569–579.

Ronchetti D, Impagnatiello F, Guzzetta M, Gasparini L, Borgatti M, Gambari R,and Ongini E (2006) Modulation of iNOS expression by a nitric oxide-releasingderivative of the natural antioxidant ferulic acid in activated RAW 264.7 macro-phages. Eur J Pharmacol 532:162–169.

Sasaki T, Yasuda H, Maeda K, and Kikkawa R (1998) Hyperalgesia and decreasedneuronal nitric oxide synthase in diabetic rats. Neuroreport 9:243–247.

Schenkman JB (1991) Induction of diabetes and evaluation of diabetic state on P450expression. Methods Enzymol 206:325–331.

Schmidtko A, Tegeder I, and Geisslinger G (2009) No NO, no pain? The role of nitricoxide and cGMP in spinal pain processing. Trends Neurosci 32:339–346.

Schreiber AK, Nones CF, Reis RC, Chichorro JG, and Cunha JM (2015) Diabeticneuropathic pain: Physiopathology and treatment. World J Diabetes 6:432–444.

Sills GJ (2006) The mechanisms of action of gabapentin and pregabalin. Curr OpinPharmacol 6:108–113.

Szolcsányi J, Bölcskei K, Szabó A, Pintér E, Petho G, Elekes K, Börzsei R, Almási R,Szuts T, and Kéri G, et al. (2004) Analgesic effect of TT-232, a heptapeptide so-matostatin analogue, in acute pain models of the rat and the mouse and instreptozotocin-induced diabetic mechanical allodynia. Eur J Pharmacol 498:103–109.

Takasu K, Honda M, Ono H, and Tanabe M (2006) Spinal alpha(2)-adrenergic andmuscarinic receptors and the NO release cascade mediate supraspinally producedeffectiveness of gabapentin at decreasing mechanical hypersensitivity in mice afterpartial nerve injury. Br J Pharmacol 148:233–244.

Vincenzi F, Targa M, Romagnoli R, Merighi S, Gessi S, Baraldi PG, Borea PA,and Varani K (2014) TRR469, a potent A(1) adenosine receptor allosteric modu-lator, exhibits anti-nociceptive properties in acute and neuropathic pain models inmice. Neuropharmacology 81:6–14.

Watkins PJ (1984) Pain and diabetic neuropathy. Br Med J (Clin Res Ed) 288:168–169.

Wright DE, Johnson MS, Arnett MG, Smittkamp SE, and Ryals JM (2007) Selectivechanges in nocifensive behavior despite normal cutaneous axon innervation inleptin receptor-null mutant (db/db) mice. J Peripher Nerv Syst 12:250–261.

Wu WP, Hao JX, Ongini E, Impagnatiello F, Presotto C, Wiesenfeld-Hallin Z, and XuXJ (2004) A nitric oxide (NO)-releasing derivative of gabapentin, NCX 8001, alle-viates neuropathic pain-like behavior after spinal cord and peripheral nerve injury.Br J Pharmacol 141:65–74.

Yorek MA (2003) The role of oxidative stress in diabetic vascular and neural disease.Free Radic Res 37:471–480.

Yuen KC, Baker NR, and Rayman G (2002) Treatment of chronic painful diabeticneuropathy with isosorbide dinitrate spray: a double-blind placebo-controlledcross-over study. Diabetes Care 25:1699–1703.

Address correspondence to: Dr. Francesco Impagnatiello, Nicox ResearchInstitute, Via Ariosto 21, 20091, Bresso, Milano, Italy. E-mail: impagnatiello@nicox.it

NCX1404 Reduces Mechanical Allodynia Development in STZ Mice 247

at ASPE

T Journals on February 1, 2021

jpet.aspetjournals.orgD

ownloaded from