- Email: [email protected]
Metformin and phenformin block the peripheral antinociception induced by diclofenac and indomethacin on the formalin test
Life Sciences 90 (2012) 8–12
Contents lists available at SciVerse ScienceDirect
Life Sciences journal homepage: www.elsevier.com/locate/lifescie
Metformin and phenformin block the peripheral antinociception induced by diclofenac and indomethacin on the formalin test Mario I. Ortiz ⁎ Laboratorio de Farmacología, Área Académica de Medicina del Instituto de Ciencias de la Salud, Universidad Autónoma del Estado de Hidalgo, Eliseo Ramírez Ulloa 400, Col. Doctores, Pachuca, Hgo., 42090, Mexico
a r t i c l e
i n f o
Article history: Received 20 May 2011 Accepted 15 September 2011 Keywords: Diclofenac Indomethacin Glibenclamide Glipizide Metformin Phenformin Peripheral antinociception
a b s t r a c t Aims: Recent evidence has shown that systemic administration of sulfonylureas and biguanides block the diclofenac-induced antinociception, but not the effect produced by indomethacin. However, there are no reports about the peripheral interaction between analgesics and the biguanides metformin and phenformin. Therefore, this work was undertaken to determine whether glibenclamide and glipizide and the biguanides metformin and phenformin have any effect on the peripheral antinociception induced by diclofenac and indomethacin. Main methods: Diclofenac and indomethacin were administered locally in the formalin-injured rat paw, and the antinociceptive effect was evaluated using the 1% formalin test. To determine whether peripheral antinociception induced by diclofenac or indomethacin was mediated by either the ATP-sensitive K+ channels or biguanides-induced mechanisms, the effect of pretreatment with the appropriates vehicles or glibenclamide, glipizide, metformin and phenformin on the antinociceptive effect induced by local peripheral diclofenac and indomethacin was assessed. Key findings: Local peripheral injections of diclofenac (50–200 μg/paw) and indomethacin (200–800 μg/paw) produced a dose-dependent antinociception during the second phase of the test. Local pretreatment with glibenclamide, glipizide, metformin and phenformin blocked the diclofenac-induced antinociception. On the other hand, the pretreatment with glibenclamide and glipizide did not prevent the local antinociception produced by indomethacin. Nonetheless, metformin and phenformin reversed the local antinociception induced by indomethacin. Significance: Data suggest that diclofenac could activate the K+ channels and biguanides-dependent mechanisms to produce its peripheral antinociceptive effects in the formalin test. Likewise, a biguanides-dependent mechanism could be activated by indomethacin consecutively to generate its peripheral antinociceptive effect. © 2011 Elsevier Inc. All rights reserved.
Introduction Generally speaking, the treatment of chronic diabetes includes drugs to lower blood sugar and drugs for the treatment of complications (Campbell, 2009; Guastella and Mick, 2009). Molecules from the sulfonylurea group and the biguanides are widely used to lower blood sugar in the therapeutic management of type 2 diabetes. Glibenclamide has been largely used in the management of non-insulin dependent diabetes mellitus worldwide, and this drug improves glucose tolerance, mainly by augmenting the insulin secretion (Luzi and Pozza, 1997) by the inhibition of the ATP-sensitive K + channels (Edwards and Weston, 1993). Biguanides, such as metformin and phenformin, have potent antihyperglycemic properties by suppressing hepatic gluconeogenesis and increasing peripheral tissue insulin sensitivity (Zhou et al., 2001; Lochhead et al., 2000). Recently, it was proved that phenformin, but not metformin, inhibits the ATP-sensitive K + channels in vascular and
⁎ Tel./fax: + 52 77 1717 2000x4510. E-mail address: [email protected]. 0024-3205/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2011.10.009
non-vascular smooth muscle and pancreatic beta-cells (Aziz et al., 2010). Diclofenac is a non-steroidal anti-inflammatory drug (NSAID) that has strong anti-inflammatory, antipyretic and analgesic activities (Todd and Sorkin, 1988). This drug has been proven to be effective in the treatment of rheumatic and non-rheumatic conditions (Todd and Sorkin, 1988). It is known that diclofenac, like other nonselective NSAIDs, is able to impair prostaglandin synthesis by the inhibition of the cyclooxygenase isozymes COX-1 and COX-2 (Vane and Botting, 1996; Warner et al., 1999). However, there is evidence that additional prostaglandin-independent mechanisms are involved in the antinociceptive action of diclofenac. In this sense, it has been established that the locally administered glibenclamide is able to block the peripheral antinociceptive effect of diclofenac in rats (Ortiz et al., 2002, 2003; Alves et al., 2004), suggesting the participation of ATP-sensitive K + channels in its antinociceptive effect at the peripheral level. Likewise, the antinociceptive effect of oral diclofenac was abolished by local or spinal administration of either L-NAME or glibenclamide (Ortiz et al., 2008). These results suggest that oral diclofenac achieves effective concentrations, producing an antinociceptive effect involving the
M.I. Ortiz / Life Sciences 90 (2012) 8–12
participation of the nitric oxide-ATP-sensitive K + channel pathway at both the local and spinal levels. Recently, our laboratory showed that the systemic pretreatment with two biguanides, metformin (100 and 180 mg/kg) and phenformin (100 and 180 mg/kg), blockades the diclofenac-induced systemic antinociception in phase two of the rat 1% formalin test (Ortiz, 2011). However, the pretreatment with metformin and phenformin did not prevent the systemic antinociception produced by indomethacin. Data suggest that diclofenac, but not indomethacin, could activate biguanides-dependent mechanisms to produce its systemic antinociceptive effects in the rat formalin test. Currently, there are no reports about the peripheral interaction between diclofenac and other hypoglycemic drugs, such as biguanides, thiazolidinediones or insulin. Therefore, this work was undertaken to determine whether the sulfonylureas, glibenclamide and glipizide, and the biguanides, metformin and phenformin, have any effect on the peripheral antinociception induced by diclofenac and indomethacin. Materials and methods Animals Male Wistar rats, aged 7–9 weeks (weight range, 180–220 g), from our own breeding facilities were used in this study. Animals had free access to food and drinking water before experiments. Efforts were made to minimize animal suffering and to reduce the number of animals used. The rats were used only once. At the end of the experiments, the rats were sacrificed in a CO2 chamber. All experiments followed the Guidelines on Ethical Standards for Investigation of Experimental Pain in Animals (Zimmermann, 1983), and the protocol was approved by the Institutional Animal Care and Use Committee (UAEH). Drugs Diclofenac, glibenclamide, glipizide, indomethacin, metformin and phenformin were purchased from Sigma (St. Louis, MO, USA). Glibenclamide and glipizide were dissolved in a 20% DMSO solution. Indomethacin was dissolved in a Tween 20 and buffer solution (sodium hydroxide and monobasic potassium phosphate). Diclofenac, metformin and phenformin were dissolved in saline. Measurement of antinociceptive activity Pain and antinociception were assessed by the formalin test, as previously described (Ortiz et al., 2002, 2003; Ortiz and CastañedaHernández, 2008). Briefly, 50 μl of diluted formalin (1%) were injected subcutaneously (s.c.) to the dorsal surface of the right hind paw, and the resulting flinching behavior was considered to be an expression of nociception. Curves of the number of flinches verses time were constructed, and these curves were biphasic. The initial acute phase (0–10 min) was followed by a short quiescent period, which was followed by a prolonged tonic response (15–60 min). The area under the curve for both phases was estimated, and a significant area reduction was interpreted as antinociception. Effect of hypoglycemic drugs on the antinociception induced by the NSAIDs Rats received a s.c. injection (50 μl) into the dorsal surface of the right hind paw of either vehicles or increasing doses of diclofenac (50–200 μg/paw) or indomethacin (200–800 μg/paw) 20 min before a formalin injection into the ipsilateral paw. To determine whether diclofenac and indomethacin acted locally, they were administered to the left (contralateral) paw 20 min before the formalin was injected into the right paw, and the corresponding response on nociceptive behavior was assessed. To determine whether peripheral antinociception induced by diclofenac or indomethacin was mediated by either the ATP-sensitive K + channels or biguanides-induced
9
mechanisms, the effect of pretreatment was assessed (10 min before formalin injection) with the appropriate vehicles or glibenclamide (10–100 μg/paw), glipizide (10–100 μg/paw), metformin (100–400 μg/ paw) and phenformin (100–400 μg/paw) on the antinociceptive effect induced by local peripheral diclofenac (200 μg/paw) and indomethacin (800 μg/paw). The drugs were injected in a volume of 50 μl. Each rat received 3 injections, and the appropriate controls for multiple injections and vehicles were performed before starting the formal study. The doses and drug administration schedules of hypoglycemics and analgesics for peripheral administration were selected based on previous reports and on pilot experiments in our laboratory. Rats in all groups were tested for possible side effects, which were observed as including a reduction of righting, stepping, corneal and pinna reflexes. Data analysis and statistics All experimental results are given as the means ± S.E.M. for 6–8 animals per group. Curves were constructed by plotting the number of flinches as a function of time. The area under the number of flinches verses time curves (AUC), an expression of the duration and intensity of the effect, was calculated by the trapezoidal rule. Reduction in the number of flinches or in the AUC of phase two is reported, because we were not able to observe an effect on phase one. One-way analysis of variance (ANOVA), followed by Dunnett's test were used to compare differences between treatments. Differences were considered to reach statistical significance when P b 0.05. Results Peripheral antinociceptive effects of diclofenac and indomethacin Formalin administration produced a typical pattern of flinching behavior. The first phase started immediately after the administration of formalin and then diminished gradually in about 10 min. The second phase started at about 15 min and lasted until 1 h. Ipsilateral, but not contralateral, local peripheral administration of diclofenac and indomethacin produced a dose-dependent reduction in the flinching behavior otherwise observed after a formalin injection (Fig. 1). The number of flinches during phase two was significantly reduced by diclofenac (P = 0.009 and F statistic of 5.154) and indomethacin (P = 0.023 and F statistic of 3.985), but it was not reduced during phase one (P > 0.05 and F statistic b0.95, data not shown). No reductions in the assessed reflexes were observed in either group, control or treated. Effect of sulfonylureas on the antinociceptive effects of diclofenac and indomethacin Local pretreatment with the ATP-sensitive K + channel inhibitors glibenclamide (P = 0.001 and F statistic of 32.525) or glipizide (P = 0.001 and F statistic of 24.321) were able to prevent the antinociception produced by diclofenac, but not by indomethacin (P > 0.05 and F statistic b0.5) (Figs. 2 and 3). Given alone, peripheral ATPsensitive K + channel inhibitors did not modify formalin-induced nociceptive behavior (glibenclamide: P = 0.777 and F statistic of 0.084, and glipizide: P = 0.819 and F statistic of 0.0549) (Figs. 2 and 3). Effect of biguanides on the antinociceptive effects of diclofenac and indomethacin Local pretreatment with the biguanides metformin (P = 0.001 and F statistic of 44.358) and phenformin (P = 0.005 and F statistic of 14.431) was able to prevent the antinociception produced by diclofenac (Fig. 4). Likewise, local peripheral pretreatment metformin (P= 0.001 and F statistic of 55.688) and phenformin (P = 0.005 and F statistic of 18.785) prevented the antinociception induced by indomethacin (Fig. 5). Given alone, peripheral biguanides did not modify formalin-induced
10
M.I. Ortiz / Life Sciences 90 (2012) 8–12
800
A
Phase 2
AUC (Flinches-min)
AUC (Flinches-min)
A
600
*
400
200
0 VEH
50
100
200
IL
800
** 600
*
400
200
0
200
VEH
AUC (Flinches-min)
AUC (Flinches-min)
B
600
*
200 0
VEH
200
400
IL
800
100
100
Diclofenac (200 µg/paw)
800
400
50
Glibenclamide (µg/paw)
Diclofenac (µg/paw)
B
10
CL
800 CL
Indomethacin (µg/paw)
800
**
600
*
400
200
0 VEH
10
50
100
100
Glipizide (µg/paw)
Diclofenac (200 µg/paw) Fig. 1. Local antinociceptive effects of diclofenac and indomethacin on the 1% formalin test. Rats were pretreated with a local injection of vehicle (VEH), diclofenac (A) or indomethacin (B) into the right paw before the formalin injection. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Each point corresponds to the mean ± S.E.M. of 6–8 animals. *Significantly different from the vehicle group (P b 0.05 and F statistic >3.5) as determined by a one way analysis of variance followed by a Dunnett's test.
nociceptive behavior (metformin: P = 0.547 and F statistic of 0.392, and phenformin: P = 0.839 and F statistic of 0.0438) (Figs. 4 and 5). Discussion Effect of sulfonylureas on the antinociceptive effects of diclofenac and indomethacin Sulfonylureas are the mainstay of glucose control in type 2 diabetes. Their mechanism of action at the cellular level consists of an inhibition of the ATP-sensitive K + channels (Edwards and Weston, 1993). In the current work, local administrations of glibenclamide and glipizide decreased the antinociceptive effect produced by diclofenac in the rats. For this reason, the blockade of the diclofenac-induced antinociception by sulfonylureas was due to the blockade of the ATP-sensitive K + channels. These channels, however, do not appear to be involved in the mechanism of the peripheral effects of indomethacin because glibenclamide and glipizide were not able to significantly inhibit the local antinociceptive responses of indomethacin. Overall, these results are in agreement with previous results where glibenclamide and the first generation sulfonylurea tolbutamide reverted the antinociception produced by diclofenac, but not by indomethacin (Ortiz et al., 2003; Ortiz, 2011). In the present study, we demonstrated for the first time that the antinociceptive effects of diclofenac were blocked by glipizide, a second generation sulfonylurea (Lebovitz, 1985). It is one of the most regularly prescribed drugs for the treatment of patients with noninsulin-dependent diabetes (Wysowski et al., 2003). For this reason, the use of the combination of diclofenac and glipizide at the clinical level should be considered carefully.
Fig. 2. Effect of the ATP-sensitive K+ channel inhibitors glibenclamide and glipizide on the peripheral antinociception produced by diclofenac during the second phase of the formalin test. Rats were pretreated with a local injection of diclofenac (−20 min) and then glibenclamide or glipizide (−10 min) into the right paw. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Bars are the mean± S.E.M. for 6–8 animals. *Significantly different from the vehicle group (P=0.009 and F statistic of 5.154) and **significantly different from the diclofenac group (P=0.001 and F statistic >24), as determined by a one way analysis of variance followed by a Dunnett's test.
Effect of metformin and phenformin on the antinociceptive effects of diclofenac In the case of the systemic administration of an analgesic (oral, subcutaneous or intraperitoneal), the drug must suffer absorption and distribution to produce its effect. However, we do not know where the drug is acting to produce the analgesia. Recently, it was shown that the diclofenac-induced systemic antinociception was blocked with the systemic administration of metformin and phenformin (Ortiz, 2011). In this last case, the site of the interaction between diclofenac and biguanides is unknown. In the present study, the local administration of the biguanide metformin was able to reverse the diclofenacinduced antinociception. For this reason, these findings suggest that diclofenac is activating metformin-dependent mechanisms at the peripheral level. It has been demonstrated that biguanides produce the activation of α1-adrenoceptors and the release of noradrenaline (Lee and Peuler, 2001; Peuler, 1999; Cheng et al., 2006). It is possible that metformin is causing a peripheral activation of these noradrenergic pathways to revert the antinociceptive effects induced by diclofenac. However, one argument against this is that these noradrenergic pathways were not activated by metformin in the absence of diclofenac to increase the nociception induced by formalin (see Fig. 4). The exact mechanisms by which metformin reverted the antinociceptive action of diclofenac needs future elucidation. On the other hand, phenformin was able to block the diclofenacinduced peripheral antinociception. Recently, it was elucidated that phenformin, but not metformin, is able to block ATP-sensitive K +
M.I. Ortiz / Life Sciences 90 (2012) 8–12
A 800 AUC (Flinches-min)
A 800 AUC (Flinches-min)
11
600
*
400
200
** 600
*
400
200
0
0 VEH
100
VEH
100
100
B AUC (Flinches-min)
AUC (Flinches-min)
400
Diclofenac (200 µg/paw)
B 800 600
*
200
0
400
Metformin (µg/paw)
Gliben (µg/paw)
Indomethacin (800 µg/paw)
400
200
800
** 600
*
400
200
0 VEH
100
VEH
100
Effect of metformin and phenformin on the antinociceptive effects of indomethacin It is known that indomethacin, like other nonselective NSAIDs, is able to impair prostaglandin synthesis by inhibiting the cyclooxygenase isozymes COX-1 and COX-2 in both the injured tissue and the central nervous system (Vane and Botting, 1996; Vanegas, 2002; Warner et al., 1999). Our group previously named indomethacin as a “pure” prostaglandin synthesis inhibitor (Ortiz et al., 2003), due to the fact that the indomethacin-induced antinociception does not participate in the activation of the nitric oxide-Cyclic-GMP-K+ channels pathway, of opioid systems or in the inhibition of H+ channels (Ortiz et al., 2003; Voilley et al., 2001). Furthermore, the systemic pretreatment with metformin and phenformin did not prevent the systemic antinociception produced by indomethacin in the rat formalin test (Ortiz, 2011). However, there are reports on other action mechanisms different than the inhibition of prostaglandin synthesis by indomethacin. For example, indomethacin is able to inhibit the functions of the human multidrug resistance protein MRP4 and, therefore, block the prostaglandin efflux transporter
400
400
Diclofenac (200 µg/paw)
Indomethacin (800 µg/paw)
channels (Aziz et al., 2010). Based on this latest evidence, it is now possible to suggest that phenformin blocked the diclofenac-induced antinociception by blocking potassium channels, as it was made by the sulfonylureas. However, it is necessary to investigate this issue to determine the exact mechanism used by phenformin to reverse the antinociceptive effect produced by diclofenac in the rats.
200
Phenformin (µg/paw)
Glipizide (µg/paw)
Fig. 3. Effect of the ATP-sensitive K+ channel inhibitors glibenclamide and glipizide on the peripheral antinociception produced by indomethacin during the second phase of the formalin test. Rats were pretreated with a local injection of indomethacin (−20 min) and then glibenclamide or glipizide (−10 min) into the right paw. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Bars are the mean± S.E.M. for 6–8 animals. *Significantly different from the vehicle group (P = 0.023 and F statistic of 3.985), as determined by a one way analysis of variance followed by a Dunnett's test.
100
Fig. 4. Effect of the biguanides metformin or phenformin on the peripheral antinociception produced by diclofenac during the second phase of the formalin test. Rats were pretreated with a local injection of diclofenac (−20 min) and then metformin or phenformin (−10 min) into the right paw. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Bars are the mean ± S.E.M. for 6–8 animals. *Significantly different from the vehicle group (P = 0.009 and F statistic of 5.154) and **significantly different from the diclofenac group (P b 0.006 and F statistic >14), as determined by a one way analysis of variance followed by a Dunnett's test.
(Reid et al., 2003). Likewise, indomethacin decreased the expression of L-selectin on the neutrophil surface (Diaz-Gonzalez et al., 1995). In the present study, metformin and phenformin were able to block the indomethacin-induced peripheral antinociceptive effects, suggesting the possible activation of biguanides-dependent mechanisms at the peripheral level by indomethacin. This last result is in contrast with the incapacity of metformin and phenformin to block the indomethacininduced systemic antinociception (Ortiz, 2011). A probable explanation is that biguanides were not able to block the systemic mechanisms activated by indomethacin (possibly in the central nervous system). The real participation of the biguanides-dependent mechanism in the antinociceptive actions produced by indomethacin requires further elucidation. When applied alone, neither the sulfonylureas nor the biguanides affected formalin behavior, thus excluding the possibility that the inhibition of the antinociception produced by diclofenac or indomethacin could be due to a hyperalgesic or nociceptive effect of the hypoglycemic drugs used. The lack of modification of the flinching behavior by the different modulators (sulfonylureas and biguanides) at concentrations able to prevent antinociception might also indicate that the K + channels and biguanides-dependent mechanisms at the peripheral level involved in the modulation of pain are not tonically activated. Conclusion The results shown here displayed that the interactions between sulfonylureas or biguanides with diclofenac or indomethacin result
12
M.I. Ortiz / Life Sciences 90 (2012) 8–12
AUC (Flinches-min)
A 800
**
600
*
400
200
0 VEH
100
200
400
400
Metformin (µg/paw)
Indomethacin (800 µg/paw)
AUC (Flinches-min)
B 800
** 600
*
400
200
0
VEH
100
200
400
400
Phenformin (µg/paw)
Indomethacin (800 µg/paw) Fig. 5. Effect of the biguanides metformin or phenformin on the peripheral antinociception produced by indomethacin during the second phase of the formalin test. Rats were pretreated with a local injection of indomethacin (−20 min) and then metformin or phenformin (−10 min) into the right paw. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Bars are the mean ± S.E.M. for 6–8 animals. *Significantly different from the vehicle group (P = 0.023 and F statistic of 3.985) and **significantly different from the Indomethacin group (P b 0.006 and F statistic >18), as determined by a one way analysis of variance followed by a Dunnett's test.
in a reduced analgesic efficacy. Finally, clinical studies are warranted to establish the relevance of these interactions. Conflict of interest statement The author declares that there are no conflicts of interest.
References Alves DP, Tatsuo MA, Leite R, Duarte ID. Diclofenac-induced peripheral antinociception is associated with ATP-sensitive K+ channels activation. Life Sci 2004;74:2577–91. Aziz Q, Thomas A, Khambra T, Tinker A. Phenformin has a direct inhibitory effect on the ATP-sensitive potassium channel. Eur J Pharmacol 2010;634:26–32.
Campbell RK. Type 2 diabetes: where we are today: an overview of disease burden, current treatments, and treatment strategies. J Am Pharm Assoc 2009;49:S3–9. Cheng JT, Huang CC, Liu IM, Tzeng TF, Chang CJ. Novel mechanism for plasma glucoselowering action of metformin in streptozotocin-induced diabetic rats. Diabetes 2006;55:819–25. Diaz-Gonzalez F, Gonzalez-Alvaro I, Campanero MR, Mollinedo F, del Pozo MA, Munoz C, et al. Prevention of in vitro neutrophil-endothelial attachment through shedding of L-selectin by nonsteroidal antiinflammatory drugs. J Clin Invest 1995;95:1756–65. Edwards G, Weston AH. The pharmacology of ATP-sensitive K+ channels. Annu Rev Pharmacol Toxicol 1993;33:597–637. Guastella V, Mick G. Strategies for the diagnosis and treatment of neuropathic pain secondary to diabetic peripheral sensory polyneuropathy. Diabetes Metab 2009;35: 12–9. Lebovitz HE. Glipizide: a second-generation sulfonylurea hypoglycemic agent. Pharmacology, pharmacokinetics and clinical use. Pharmacotherapy 1985;5:63–77. Lee JM, Peuler JD. A possible indirect sympathomimetic action of metformin in the arterial vessel wall of spontaneously hypertensive rats. Life Sci 2001;69:1085–92. Lochhead PA, Salt IP, Walker KS, Hardie DG, Sutherland C. 5-aminoimidazole-4-carboxamide riboside mimics the effects of insulin on the expression of the 2 key gluconeogenic genes PEPCK and glucose-6-phosphatase. Diabetes 2000;49:896–903. Luzi L, Pozza G. Glibenclamide: an old drug with a novel mechanism of action? Acta Diabetol 1997;34:239–44. Ortiz MI. Blockade of the antinociception induced by diclofenac, but not of indomethacin, by sulfonylureas and biguanides. Pharmacol Biochem Behav 2011;99:1–6. Ortiz MI, Castañeda-Hernández G. Examination of the interaction between peripheral lumiracoxib and opioids on the 1% formalin test in rats. Eur J Pain 2008;12:233–41. Ortiz MI, Torres-López JE, Castañeda-Hernández G, Rosas R, Vidal-Cantú GC, GranadosSoto V. Pharmacological evidence for the activation of K+ channels by diclofenac. Eur J Pharmacol 2002;438:85–91. Ortiz MI, Granados-Soto V, Castañeda-Hernández G. The NOcGMP-K(+) channel pathway participates in the antinociceptive effect of diclofenac, but not of indomethacin. Pharmacol Biochem Behav 2003;76:187–95. Ortiz MI, Lozano-Cuenca J, Granados-Soto V, Castañeda-Hernández G. Additive interaction between peripheral and central mechanisms involved in the antinociceptive effect of diclofenac in the formalin test in the rat. Pharmacol Biochem Behav 2008;91:32–7. Peuler JD. Opposing adrenergic actions of intravenous metformin on arterial pressure in female spontaneously hypertensive rats. Cardiovasc Res 1999;43:237–47. Reid G, Wielinga P, Zelcer N, van der Heijden I, Kuil A, de Haas M, et al. The human multidrug resistance protein MRP4 functions as a prostaglandin efflux transporter and is inhibited by nonsteroidal antiinflammatory drugs. Proc Natl Acad Sci U S A 2003;100:9244–9. Todd PA, Sorkin EM. Diclofenac sodium. A reappraisal of its pharmacodynamic and pharmacokinetic properties, and therapeutic efficacy. Drugs 1988;35:244–85. Vane JR, Botting RM. Mechanism of action of anti-inflammatory drugs. Scand J Rheumatol 1996;102:9-21. Vanegas H. Bases for a spinal analgesic action of cyclooxygenase inhibitors. Proc West Pharmacol Soc 2002;45:225–7. Voilley N, De Weille J, Mamet J, Lazdunski M. Nonsteroid anti-inflammatory drugs inhibit both the activity and the inflammation-induced expression of acid-sensing ion channels in nociceptors. J Neurosci 2001;21:8026–33. Warner TD, Giuliano F, Vojnovic I, Bukasa A, Mitchell JA, Vane JR. Nonsteroid drug selectivities for cyclo-oxygenase-1 rather than cyclo-oxygenase-2 are associated with human gastrointestinal toxicity: a full in vitro analysis. Proc Natl Acad Sci U S A 1999;96:7563–8. Wysowski DK, Armstrong G, Governale L. Rapid increase in the use of oral antidiabetic drugs in the United States, 1990–2001. Diabetes Care 2003;26:1852–5. Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, et al. Role of AMP-activated protein kinase in mechanism of metformin action. J Clin Invest 2001;108:1167–74. Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;16:109–10.
Contents lists available at SciVerse ScienceDirect
Life Sciences journal homepage: www.elsevier.com/locate/lifescie
Metformin and phenformin block the peripheral antinociception induced by diclofenac and indomethacin on the formalin test Mario I. Ortiz ⁎ Laboratorio de Farmacología, Área Académica de Medicina del Instituto de Ciencias de la Salud, Universidad Autónoma del Estado de Hidalgo, Eliseo Ramírez Ulloa 400, Col. Doctores, Pachuca, Hgo., 42090, Mexico
a r t i c l e
i n f o
Article history: Received 20 May 2011 Accepted 15 September 2011 Keywords: Diclofenac Indomethacin Glibenclamide Glipizide Metformin Phenformin Peripheral antinociception
a b s t r a c t Aims: Recent evidence has shown that systemic administration of sulfonylureas and biguanides block the diclofenac-induced antinociception, but not the effect produced by indomethacin. However, there are no reports about the peripheral interaction between analgesics and the biguanides metformin and phenformin. Therefore, this work was undertaken to determine whether glibenclamide and glipizide and the biguanides metformin and phenformin have any effect on the peripheral antinociception induced by diclofenac and indomethacin. Main methods: Diclofenac and indomethacin were administered locally in the formalin-injured rat paw, and the antinociceptive effect was evaluated using the 1% formalin test. To determine whether peripheral antinociception induced by diclofenac or indomethacin was mediated by either the ATP-sensitive K+ channels or biguanides-induced mechanisms, the effect of pretreatment with the appropriates vehicles or glibenclamide, glipizide, metformin and phenformin on the antinociceptive effect induced by local peripheral diclofenac and indomethacin was assessed. Key findings: Local peripheral injections of diclofenac (50–200 μg/paw) and indomethacin (200–800 μg/paw) produced a dose-dependent antinociception during the second phase of the test. Local pretreatment with glibenclamide, glipizide, metformin and phenformin blocked the diclofenac-induced antinociception. On the other hand, the pretreatment with glibenclamide and glipizide did not prevent the local antinociception produced by indomethacin. Nonetheless, metformin and phenformin reversed the local antinociception induced by indomethacin. Significance: Data suggest that diclofenac could activate the K+ channels and biguanides-dependent mechanisms to produce its peripheral antinociceptive effects in the formalin test. Likewise, a biguanides-dependent mechanism could be activated by indomethacin consecutively to generate its peripheral antinociceptive effect. © 2011 Elsevier Inc. All rights reserved.
Introduction Generally speaking, the treatment of chronic diabetes includes drugs to lower blood sugar and drugs for the treatment of complications (Campbell, 2009; Guastella and Mick, 2009). Molecules from the sulfonylurea group and the biguanides are widely used to lower blood sugar in the therapeutic management of type 2 diabetes. Glibenclamide has been largely used in the management of non-insulin dependent diabetes mellitus worldwide, and this drug improves glucose tolerance, mainly by augmenting the insulin secretion (Luzi and Pozza, 1997) by the inhibition of the ATP-sensitive K + channels (Edwards and Weston, 1993). Biguanides, such as metformin and phenformin, have potent antihyperglycemic properties by suppressing hepatic gluconeogenesis and increasing peripheral tissue insulin sensitivity (Zhou et al., 2001; Lochhead et al., 2000). Recently, it was proved that phenformin, but not metformin, inhibits the ATP-sensitive K + channels in vascular and
⁎ Tel./fax: + 52 77 1717 2000x4510. E-mail address: [email protected]. 0024-3205/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2011.10.009
non-vascular smooth muscle and pancreatic beta-cells (Aziz et al., 2010). Diclofenac is a non-steroidal anti-inflammatory drug (NSAID) that has strong anti-inflammatory, antipyretic and analgesic activities (Todd and Sorkin, 1988). This drug has been proven to be effective in the treatment of rheumatic and non-rheumatic conditions (Todd and Sorkin, 1988). It is known that diclofenac, like other nonselective NSAIDs, is able to impair prostaglandin synthesis by the inhibition of the cyclooxygenase isozymes COX-1 and COX-2 (Vane and Botting, 1996; Warner et al., 1999). However, there is evidence that additional prostaglandin-independent mechanisms are involved in the antinociceptive action of diclofenac. In this sense, it has been established that the locally administered glibenclamide is able to block the peripheral antinociceptive effect of diclofenac in rats (Ortiz et al., 2002, 2003; Alves et al., 2004), suggesting the participation of ATP-sensitive K + channels in its antinociceptive effect at the peripheral level. Likewise, the antinociceptive effect of oral diclofenac was abolished by local or spinal administration of either L-NAME or glibenclamide (Ortiz et al., 2008). These results suggest that oral diclofenac achieves effective concentrations, producing an antinociceptive effect involving the
M.I. Ortiz / Life Sciences 90 (2012) 8–12
participation of the nitric oxide-ATP-sensitive K + channel pathway at both the local and spinal levels. Recently, our laboratory showed that the systemic pretreatment with two biguanides, metformin (100 and 180 mg/kg) and phenformin (100 and 180 mg/kg), blockades the diclofenac-induced systemic antinociception in phase two of the rat 1% formalin test (Ortiz, 2011). However, the pretreatment with metformin and phenformin did not prevent the systemic antinociception produced by indomethacin. Data suggest that diclofenac, but not indomethacin, could activate biguanides-dependent mechanisms to produce its systemic antinociceptive effects in the rat formalin test. Currently, there are no reports about the peripheral interaction between diclofenac and other hypoglycemic drugs, such as biguanides, thiazolidinediones or insulin. Therefore, this work was undertaken to determine whether the sulfonylureas, glibenclamide and glipizide, and the biguanides, metformin and phenformin, have any effect on the peripheral antinociception induced by diclofenac and indomethacin. Materials and methods Animals Male Wistar rats, aged 7–9 weeks (weight range, 180–220 g), from our own breeding facilities were used in this study. Animals had free access to food and drinking water before experiments. Efforts were made to minimize animal suffering and to reduce the number of animals used. The rats were used only once. At the end of the experiments, the rats were sacrificed in a CO2 chamber. All experiments followed the Guidelines on Ethical Standards for Investigation of Experimental Pain in Animals (Zimmermann, 1983), and the protocol was approved by the Institutional Animal Care and Use Committee (UAEH). Drugs Diclofenac, glibenclamide, glipizide, indomethacin, metformin and phenformin were purchased from Sigma (St. Louis, MO, USA). Glibenclamide and glipizide were dissolved in a 20% DMSO solution. Indomethacin was dissolved in a Tween 20 and buffer solution (sodium hydroxide and monobasic potassium phosphate). Diclofenac, metformin and phenformin were dissolved in saline. Measurement of antinociceptive activity Pain and antinociception were assessed by the formalin test, as previously described (Ortiz et al., 2002, 2003; Ortiz and CastañedaHernández, 2008). Briefly, 50 μl of diluted formalin (1%) were injected subcutaneously (s.c.) to the dorsal surface of the right hind paw, and the resulting flinching behavior was considered to be an expression of nociception. Curves of the number of flinches verses time were constructed, and these curves were biphasic. The initial acute phase (0–10 min) was followed by a short quiescent period, which was followed by a prolonged tonic response (15–60 min). The area under the curve for both phases was estimated, and a significant area reduction was interpreted as antinociception. Effect of hypoglycemic drugs on the antinociception induced by the NSAIDs Rats received a s.c. injection (50 μl) into the dorsal surface of the right hind paw of either vehicles or increasing doses of diclofenac (50–200 μg/paw) or indomethacin (200–800 μg/paw) 20 min before a formalin injection into the ipsilateral paw. To determine whether diclofenac and indomethacin acted locally, they were administered to the left (contralateral) paw 20 min before the formalin was injected into the right paw, and the corresponding response on nociceptive behavior was assessed. To determine whether peripheral antinociception induced by diclofenac or indomethacin was mediated by either the ATP-sensitive K + channels or biguanides-induced
9
mechanisms, the effect of pretreatment was assessed (10 min before formalin injection) with the appropriate vehicles or glibenclamide (10–100 μg/paw), glipizide (10–100 μg/paw), metformin (100–400 μg/ paw) and phenformin (100–400 μg/paw) on the antinociceptive effect induced by local peripheral diclofenac (200 μg/paw) and indomethacin (800 μg/paw). The drugs were injected in a volume of 50 μl. Each rat received 3 injections, and the appropriate controls for multiple injections and vehicles were performed before starting the formal study. The doses and drug administration schedules of hypoglycemics and analgesics for peripheral administration were selected based on previous reports and on pilot experiments in our laboratory. Rats in all groups were tested for possible side effects, which were observed as including a reduction of righting, stepping, corneal and pinna reflexes. Data analysis and statistics All experimental results are given as the means ± S.E.M. for 6–8 animals per group. Curves were constructed by plotting the number of flinches as a function of time. The area under the number of flinches verses time curves (AUC), an expression of the duration and intensity of the effect, was calculated by the trapezoidal rule. Reduction in the number of flinches or in the AUC of phase two is reported, because we were not able to observe an effect on phase one. One-way analysis of variance (ANOVA), followed by Dunnett's test were used to compare differences between treatments. Differences were considered to reach statistical significance when P b 0.05. Results Peripheral antinociceptive effects of diclofenac and indomethacin Formalin administration produced a typical pattern of flinching behavior. The first phase started immediately after the administration of formalin and then diminished gradually in about 10 min. The second phase started at about 15 min and lasted until 1 h. Ipsilateral, but not contralateral, local peripheral administration of diclofenac and indomethacin produced a dose-dependent reduction in the flinching behavior otherwise observed after a formalin injection (Fig. 1). The number of flinches during phase two was significantly reduced by diclofenac (P = 0.009 and F statistic of 5.154) and indomethacin (P = 0.023 and F statistic of 3.985), but it was not reduced during phase one (P > 0.05 and F statistic b0.95, data not shown). No reductions in the assessed reflexes were observed in either group, control or treated. Effect of sulfonylureas on the antinociceptive effects of diclofenac and indomethacin Local pretreatment with the ATP-sensitive K + channel inhibitors glibenclamide (P = 0.001 and F statistic of 32.525) or glipizide (P = 0.001 and F statistic of 24.321) were able to prevent the antinociception produced by diclofenac, but not by indomethacin (P > 0.05 and F statistic b0.5) (Figs. 2 and 3). Given alone, peripheral ATPsensitive K + channel inhibitors did not modify formalin-induced nociceptive behavior (glibenclamide: P = 0.777 and F statistic of 0.084, and glipizide: P = 0.819 and F statistic of 0.0549) (Figs. 2 and 3). Effect of biguanides on the antinociceptive effects of diclofenac and indomethacin Local pretreatment with the biguanides metformin (P = 0.001 and F statistic of 44.358) and phenformin (P = 0.005 and F statistic of 14.431) was able to prevent the antinociception produced by diclofenac (Fig. 4). Likewise, local peripheral pretreatment metformin (P= 0.001 and F statistic of 55.688) and phenformin (P = 0.005 and F statistic of 18.785) prevented the antinociception induced by indomethacin (Fig. 5). Given alone, peripheral biguanides did not modify formalin-induced
10
M.I. Ortiz / Life Sciences 90 (2012) 8–12
800
A
Phase 2
AUC (Flinches-min)
AUC (Flinches-min)
A
600
*
400
200
0 VEH
50
100
200
IL
800
** 600
*
400
200
0
200
VEH
AUC (Flinches-min)
AUC (Flinches-min)
B
600
*
200 0
VEH
200
400
IL
800
100
100
Diclofenac (200 µg/paw)
800
400
50
Glibenclamide (µg/paw)
Diclofenac (µg/paw)
B
10
CL
800 CL
Indomethacin (µg/paw)
800
**
600
*
400
200
0 VEH
10
50
100
100
Glipizide (µg/paw)
Diclofenac (200 µg/paw) Fig. 1. Local antinociceptive effects of diclofenac and indomethacin on the 1% formalin test. Rats were pretreated with a local injection of vehicle (VEH), diclofenac (A) or indomethacin (B) into the right paw before the formalin injection. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Each point corresponds to the mean ± S.E.M. of 6–8 animals. *Significantly different from the vehicle group (P b 0.05 and F statistic >3.5) as determined by a one way analysis of variance followed by a Dunnett's test.
nociceptive behavior (metformin: P = 0.547 and F statistic of 0.392, and phenformin: P = 0.839 and F statistic of 0.0438) (Figs. 4 and 5). Discussion Effect of sulfonylureas on the antinociceptive effects of diclofenac and indomethacin Sulfonylureas are the mainstay of glucose control in type 2 diabetes. Their mechanism of action at the cellular level consists of an inhibition of the ATP-sensitive K + channels (Edwards and Weston, 1993). In the current work, local administrations of glibenclamide and glipizide decreased the antinociceptive effect produced by diclofenac in the rats. For this reason, the blockade of the diclofenac-induced antinociception by sulfonylureas was due to the blockade of the ATP-sensitive K + channels. These channels, however, do not appear to be involved in the mechanism of the peripheral effects of indomethacin because glibenclamide and glipizide were not able to significantly inhibit the local antinociceptive responses of indomethacin. Overall, these results are in agreement with previous results where glibenclamide and the first generation sulfonylurea tolbutamide reverted the antinociception produced by diclofenac, but not by indomethacin (Ortiz et al., 2003; Ortiz, 2011). In the present study, we demonstrated for the first time that the antinociceptive effects of diclofenac were blocked by glipizide, a second generation sulfonylurea (Lebovitz, 1985). It is one of the most regularly prescribed drugs for the treatment of patients with noninsulin-dependent diabetes (Wysowski et al., 2003). For this reason, the use of the combination of diclofenac and glipizide at the clinical level should be considered carefully.
Fig. 2. Effect of the ATP-sensitive K+ channel inhibitors glibenclamide and glipizide on the peripheral antinociception produced by diclofenac during the second phase of the formalin test. Rats were pretreated with a local injection of diclofenac (−20 min) and then glibenclamide or glipizide (−10 min) into the right paw. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Bars are the mean± S.E.M. for 6–8 animals. *Significantly different from the vehicle group (P=0.009 and F statistic of 5.154) and **significantly different from the diclofenac group (P=0.001 and F statistic >24), as determined by a one way analysis of variance followed by a Dunnett's test.
Effect of metformin and phenformin on the antinociceptive effects of diclofenac In the case of the systemic administration of an analgesic (oral, subcutaneous or intraperitoneal), the drug must suffer absorption and distribution to produce its effect. However, we do not know where the drug is acting to produce the analgesia. Recently, it was shown that the diclofenac-induced systemic antinociception was blocked with the systemic administration of metformin and phenformin (Ortiz, 2011). In this last case, the site of the interaction between diclofenac and biguanides is unknown. In the present study, the local administration of the biguanide metformin was able to reverse the diclofenacinduced antinociception. For this reason, these findings suggest that diclofenac is activating metformin-dependent mechanisms at the peripheral level. It has been demonstrated that biguanides produce the activation of α1-adrenoceptors and the release of noradrenaline (Lee and Peuler, 2001; Peuler, 1999; Cheng et al., 2006). It is possible that metformin is causing a peripheral activation of these noradrenergic pathways to revert the antinociceptive effects induced by diclofenac. However, one argument against this is that these noradrenergic pathways were not activated by metformin in the absence of diclofenac to increase the nociception induced by formalin (see Fig. 4). The exact mechanisms by which metformin reverted the antinociceptive action of diclofenac needs future elucidation. On the other hand, phenformin was able to block the diclofenacinduced peripheral antinociception. Recently, it was elucidated that phenformin, but not metformin, is able to block ATP-sensitive K +
M.I. Ortiz / Life Sciences 90 (2012) 8–12
A 800 AUC (Flinches-min)
A 800 AUC (Flinches-min)
11
600
*
400
200
** 600
*
400
200
0
0 VEH
100
VEH
100
100
B AUC (Flinches-min)
AUC (Flinches-min)
400
Diclofenac (200 µg/paw)
B 800 600
*
200
0
400
Metformin (µg/paw)
Gliben (µg/paw)
Indomethacin (800 µg/paw)
400
200
800
** 600
*
400
200
0 VEH
100
VEH
100
Effect of metformin and phenformin on the antinociceptive effects of indomethacin It is known that indomethacin, like other nonselective NSAIDs, is able to impair prostaglandin synthesis by inhibiting the cyclooxygenase isozymes COX-1 and COX-2 in both the injured tissue and the central nervous system (Vane and Botting, 1996; Vanegas, 2002; Warner et al., 1999). Our group previously named indomethacin as a “pure” prostaglandin synthesis inhibitor (Ortiz et al., 2003), due to the fact that the indomethacin-induced antinociception does not participate in the activation of the nitric oxide-Cyclic-GMP-K+ channels pathway, of opioid systems or in the inhibition of H+ channels (Ortiz et al., 2003; Voilley et al., 2001). Furthermore, the systemic pretreatment with metformin and phenformin did not prevent the systemic antinociception produced by indomethacin in the rat formalin test (Ortiz, 2011). However, there are reports on other action mechanisms different than the inhibition of prostaglandin synthesis by indomethacin. For example, indomethacin is able to inhibit the functions of the human multidrug resistance protein MRP4 and, therefore, block the prostaglandin efflux transporter
400
400
Diclofenac (200 µg/paw)
Indomethacin (800 µg/paw)
channels (Aziz et al., 2010). Based on this latest evidence, it is now possible to suggest that phenformin blocked the diclofenac-induced antinociception by blocking potassium channels, as it was made by the sulfonylureas. However, it is necessary to investigate this issue to determine the exact mechanism used by phenformin to reverse the antinociceptive effect produced by diclofenac in the rats.
200
Phenformin (µg/paw)
Glipizide (µg/paw)
Fig. 3. Effect of the ATP-sensitive K+ channel inhibitors glibenclamide and glipizide on the peripheral antinociception produced by indomethacin during the second phase of the formalin test. Rats were pretreated with a local injection of indomethacin (−20 min) and then glibenclamide or glipizide (−10 min) into the right paw. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Bars are the mean± S.E.M. for 6–8 animals. *Significantly different from the vehicle group (P = 0.023 and F statistic of 3.985), as determined by a one way analysis of variance followed by a Dunnett's test.
100
Fig. 4. Effect of the biguanides metformin or phenformin on the peripheral antinociception produced by diclofenac during the second phase of the formalin test. Rats were pretreated with a local injection of diclofenac (−20 min) and then metformin or phenformin (−10 min) into the right paw. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Bars are the mean ± S.E.M. for 6–8 animals. *Significantly different from the vehicle group (P = 0.009 and F statistic of 5.154) and **significantly different from the diclofenac group (P b 0.006 and F statistic >14), as determined by a one way analysis of variance followed by a Dunnett's test.
(Reid et al., 2003). Likewise, indomethacin decreased the expression of L-selectin on the neutrophil surface (Diaz-Gonzalez et al., 1995). In the present study, metformin and phenformin were able to block the indomethacin-induced peripheral antinociceptive effects, suggesting the possible activation of biguanides-dependent mechanisms at the peripheral level by indomethacin. This last result is in contrast with the incapacity of metformin and phenformin to block the indomethacininduced systemic antinociception (Ortiz, 2011). A probable explanation is that biguanides were not able to block the systemic mechanisms activated by indomethacin (possibly in the central nervous system). The real participation of the biguanides-dependent mechanism in the antinociceptive actions produced by indomethacin requires further elucidation. When applied alone, neither the sulfonylureas nor the biguanides affected formalin behavior, thus excluding the possibility that the inhibition of the antinociception produced by diclofenac or indomethacin could be due to a hyperalgesic or nociceptive effect of the hypoglycemic drugs used. The lack of modification of the flinching behavior by the different modulators (sulfonylureas and biguanides) at concentrations able to prevent antinociception might also indicate that the K + channels and biguanides-dependent mechanisms at the peripheral level involved in the modulation of pain are not tonically activated. Conclusion The results shown here displayed that the interactions between sulfonylureas or biguanides with diclofenac or indomethacin result
12
M.I. Ortiz / Life Sciences 90 (2012) 8–12
AUC (Flinches-min)
A 800
**
600
*
400
200
0 VEH
100
200
400
400
Metformin (µg/paw)
Indomethacin (800 µg/paw)
AUC (Flinches-min)
B 800
** 600
*
400
200
0
VEH
100
200
400
400
Phenformin (µg/paw)
Indomethacin (800 µg/paw) Fig. 5. Effect of the biguanides metformin or phenformin on the peripheral antinociception produced by indomethacin during the second phase of the formalin test. Rats were pretreated with a local injection of indomethacin (−20 min) and then metformin or phenformin (−10 min) into the right paw. Data are expressed as the area under the number of flinches verses time curve (AUC) on the second phase. Bars are the mean ± S.E.M. for 6–8 animals. *Significantly different from the vehicle group (P = 0.023 and F statistic of 3.985) and **significantly different from the Indomethacin group (P b 0.006 and F statistic >18), as determined by a one way analysis of variance followed by a Dunnett's test.
in a reduced analgesic efficacy. Finally, clinical studies are warranted to establish the relevance of these interactions. Conflict of interest statement The author declares that there are no conflicts of interest.
References Alves DP, Tatsuo MA, Leite R, Duarte ID. Diclofenac-induced peripheral antinociception is associated with ATP-sensitive K+ channels activation. Life Sci 2004;74:2577–91. Aziz Q, Thomas A, Khambra T, Tinker A. Phenformin has a direct inhibitory effect on the ATP-sensitive potassium channel. Eur J Pharmacol 2010;634:26–32.
Campbell RK. Type 2 diabetes: where we are today: an overview of disease burden, current treatments, and treatment strategies. J Am Pharm Assoc 2009;49:S3–9. Cheng JT, Huang CC, Liu IM, Tzeng TF, Chang CJ. Novel mechanism for plasma glucoselowering action of metformin in streptozotocin-induced diabetic rats. Diabetes 2006;55:819–25. Diaz-Gonzalez F, Gonzalez-Alvaro I, Campanero MR, Mollinedo F, del Pozo MA, Munoz C, et al. Prevention of in vitro neutrophil-endothelial attachment through shedding of L-selectin by nonsteroidal antiinflammatory drugs. J Clin Invest 1995;95:1756–65. Edwards G, Weston AH. The pharmacology of ATP-sensitive K+ channels. Annu Rev Pharmacol Toxicol 1993;33:597–637. Guastella V, Mick G. Strategies for the diagnosis and treatment of neuropathic pain secondary to diabetic peripheral sensory polyneuropathy. Diabetes Metab 2009;35: 12–9. Lebovitz HE. Glipizide: a second-generation sulfonylurea hypoglycemic agent. Pharmacology, pharmacokinetics and clinical use. Pharmacotherapy 1985;5:63–77. Lee JM, Peuler JD. A possible indirect sympathomimetic action of metformin in the arterial vessel wall of spontaneously hypertensive rats. Life Sci 2001;69:1085–92. Lochhead PA, Salt IP, Walker KS, Hardie DG, Sutherland C. 5-aminoimidazole-4-carboxamide riboside mimics the effects of insulin on the expression of the 2 key gluconeogenic genes PEPCK and glucose-6-phosphatase. Diabetes 2000;49:896–903. Luzi L, Pozza G. Glibenclamide: an old drug with a novel mechanism of action? Acta Diabetol 1997;34:239–44. Ortiz MI. Blockade of the antinociception induced by diclofenac, but not of indomethacin, by sulfonylureas and biguanides. Pharmacol Biochem Behav 2011;99:1–6. Ortiz MI, Castañeda-Hernández G. Examination of the interaction between peripheral lumiracoxib and opioids on the 1% formalin test in rats. Eur J Pain 2008;12:233–41. Ortiz MI, Torres-López JE, Castañeda-Hernández G, Rosas R, Vidal-Cantú GC, GranadosSoto V. Pharmacological evidence for the activation of K+ channels by diclofenac. Eur J Pharmacol 2002;438:85–91. Ortiz MI, Granados-Soto V, Castañeda-Hernández G. The NOcGMP-K(+) channel pathway participates in the antinociceptive effect of diclofenac, but not of indomethacin. Pharmacol Biochem Behav 2003;76:187–95. Ortiz MI, Lozano-Cuenca J, Granados-Soto V, Castañeda-Hernández G. Additive interaction between peripheral and central mechanisms involved in the antinociceptive effect of diclofenac in the formalin test in the rat. Pharmacol Biochem Behav 2008;91:32–7. Peuler JD. Opposing adrenergic actions of intravenous metformin on arterial pressure in female spontaneously hypertensive rats. Cardiovasc Res 1999;43:237–47. Reid G, Wielinga P, Zelcer N, van der Heijden I, Kuil A, de Haas M, et al. The human multidrug resistance protein MRP4 functions as a prostaglandin efflux transporter and is inhibited by nonsteroidal antiinflammatory drugs. Proc Natl Acad Sci U S A 2003;100:9244–9. Todd PA, Sorkin EM. Diclofenac sodium. A reappraisal of its pharmacodynamic and pharmacokinetic properties, and therapeutic efficacy. Drugs 1988;35:244–85. Vane JR, Botting RM. Mechanism of action of anti-inflammatory drugs. Scand J Rheumatol 1996;102:9-21. Vanegas H. Bases for a spinal analgesic action of cyclooxygenase inhibitors. Proc West Pharmacol Soc 2002;45:225–7. Voilley N, De Weille J, Mamet J, Lazdunski M. Nonsteroid anti-inflammatory drugs inhibit both the activity and the inflammation-induced expression of acid-sensing ion channels in nociceptors. J Neurosci 2001;21:8026–33. Warner TD, Giuliano F, Vojnovic I, Bukasa A, Mitchell JA, Vane JR. Nonsteroid drug selectivities for cyclo-oxygenase-1 rather than cyclo-oxygenase-2 are associated with human gastrointestinal toxicity: a full in vitro analysis. Proc Natl Acad Sci U S A 1999;96:7563–8. Wysowski DK, Armstrong G, Governale L. Rapid increase in the use of oral antidiabetic drugs in the United States, 1990–2001. Diabetes Care 2003;26:1852–5. Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, et al. Role of AMP-activated protein kinase in mechanism of metformin action. J Clin Invest 2001;108:1167–74. Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;16:109–10.