Decreased Expression of Intestinal P-Glycoprotein Increases the Analgesic Effects of Oral Morphine in a Streptozotocin-Induced Diabetic Mouse Model

Home , Tetrahydrobiopterin

Drug Metab. Pharmacokinet. 26 (6): 584591 (2011). Copyright © 2011 by the Japanese Society for the Study of Xenobiotics (JSSX) Regular Article Decreased Expression of Intestinal P-glycoprotein Increases the Analgesic Effects of Oral Morphine in a Streptozotocin-induced Diabetic Mouse Model

Ayaka NAWA1,WakakoFUJITA-HAMABE1,ShirohKISHIOKA2 and Shogo TOKUYAMA1,* 1Department of Clinical Pharmacy, School of Pharmaceutical Sciences, Kobe Gakuin University, Kobe, Japan 2Department of Pharmacology, Wakayama Medical University, Wakayama, Japan

Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk

Summary: Morphine is one of the strongest analgesics and is commonly used for the treatment of chronic pain. The pharmacokinetic properties of morphine are, in part, modulated by P-glycoprotein (P-gp). We previously reported that intestinal P-gp expression levels are influenced via the activation of inducible nitric oxide synthase (iNOS) in streptozotocin (STZ)-induced diabetic mice. Herein, we examine the analgesic effects of orally administered morphine and its pharmacokinetic properties under diabetic conditions, specifically we focusing on the involvement of intestinal P-gp in a type 1 diabetic mouse model. We assessed the analgesic effect of morphine using the tail-flick test. Serum and brain morphine levels were determined using a HPLC-ECD system. Oral morphine analgesic effects and serum and brain morphine content were significantly increased 9 days after STZ administration. The increase in the analgesic effects of morphine, as well as serum and brain morphine content, was suppressed by aminoguanidine, a specificiNOSinhibitor. Conversely, there were no changes in the analgesic effects obtained with subcutaneous morphine in STZ- treated mice. Our findings suggest that the analgesic effects of oral morphine are dependent on intestinal P- gp expression, and this may be one of the reasons that it is difficult to obtain stable pharmacological effects of morphine in diabetic patients.

Keywords: P-glycoprotein; diabetes; intestine; inducible nitric oxide synthase; morphine

intestines and liver.3®5¥ It has been reported that opioid Introduction disposition and analgesic effects were significantly enhanced Opioids are the strongest analgesics available and are in a P-gp knockout mouse.6¥ Furthermore, many reports commonly used for the treatment of chronic pain in the have demonstrated that functional differences in P-gp may palliative care of cancer patients.1¥ However, the severe result in differences in the analgesic effects produced by side effects associated with opioid use, such as vomiting, opioids.7¥ These findings indicate that the analgesic effects of respiratory depression, constipation and drowsiness, may opioids are highly dependent on the functional differences in outweigh their therapeutic benefits.1¥ Thus, it is critical that P-gp. Interestingly, the majority of research in the area has the pharmacokinetics and/or pharmacodynamics of opioids focused on P-gp function within the BBB, and only a few are appropriately regulated. It was previously reported that reports have focused on intestinal P-gp. Given that the the pharmacokinetics and pharmacodynamics of opioids may current guidelines of the World Health Organization be regulated not only by opioid receptor sensitivity, but also ¤WHO¥ recommend the oral administration of opioids in by other factors, such as metabolizing enzymes ¤e.g. UDP- the palliative care setting,1¥ and that the absorptive processes glucuronyltransferase¥ and drug efflux transporters.2¥ of the small intestine influence the pharmacological effects P-glycoprotein ¤P-gp¥, a drug efflux pump that transports of orally administered drugs, the present study focused on a large number of drugs, including opioids, is expressed in the influence of intestinal P-gp on the analgesic effects of numerous tissues, such as the blood-brain barrier ¤BBB¥, opioids.

Received; June 2, 2011, Accepted; August 12, 2011 J-STAGE Advance Published Date: August 30, 2011, doi:10.2133/dmpk.DMPK-11-RG-051 *To whom correspondence should be addressed: Shogo TOKUYAMA Ph.D., Department of Clinical Pharmacy, School of Pharmaceutical Sciences Kobe Gakuin University, 1-1-3 Minatojima, Chuo-ku, Kobe 650-8586, Japan. Tel. +81-78-974-4780, Fax. +81-78-974-4780, E-mail: stoku@ pharm.kobegakuin.ac.jp

584 Decrease in P-gp Increases Analgesic Effect of Oral Morphine 585

P-gp expression and activity levels are affected by certain Morphine content: Experiments were performed as pathophysiological conditions, such as cancer, diabetes or previously described by Shimizu et al.10¥ Briefly, blood and ischemic injury.5,6,8¥ We previously reported that ileal P-gp brain samples were collected 15 and 30 min after morphine expression levels are transiently decreased via the activation administration. Serum was separated by centrifugation of inducible nitric oxide synthase ¤iNOS¥ in a streptozotocin ¤3000 rpm for 10 min at 4ôC¥. The brain was homogenized ¤STZ¥-induced type 1 diabetic mouse model. These changes in 1 ml of pure water by sonication ¤20 s¥. The mixture of may affect the pharmacokinetics and pharmacodynamics of sample ¤100 µl of serum or brain homogenate¥ and 40% ¤ ¥ orally administered opioids. Additionally, diabetic patients K2HPO4 solution 1ml was shaken with 5 ml of ethyl are susceptible to developing cancer,9¥ and thus it is acetate for 20 min and then centrifuged at 2000 g for 5 min at important to investigate alterations in the pharmacokinetic 4ôC. The organic layer was collected, and the aqueous layer and pharmacodynamic properties of opioids in the diabetic was re-extracted with 5 ml of ethyl acetate. Then, the condition. morphine in the organic layer was extracted with 1 ml of Herein, we examine the analgesic effects of orally 1 M acetic acid, and a 0.9-ml portion of the aqueous layer administrated morphine and its altered pharmacokinetics in was lyophilized. The samples were dissolved in 200 µl of the diabetic condition. Specifically, we focus on the involve- 0.01 M HCl, and 20 µl was analyzed by the HPLC-ECD ment of intestinal P-gp in a type 1 diabetic mouse model. system. The HPLC-ECD system conditions were as follows: column, Eicompak MA-ODS ¤Eicom, Kyoto, Japan¥; mobile Methods phase, 0.1 M citric acetate buffer ¤pH 3.9¥/methanol Animals: Male ddY mice ¤4®5 weeks old, Japan SLC ¤82/18¥ containing 3 mg/l of EDTA and 150 mg/l of Inc., Shizuoka, Japan¥ were provided with food and water ad sodium octane sulfonate; flow rate, 1 ml/min; detector, libitum and housed in an animal room that was maintained at ECD-100 ¤Eicom¥ 750 mV Ag/AgCl; temperature, 25ôC. 24ôC and 55 + 5% humidity with a 12-h light/dark cycle Preparation of membrane fractions from intesti- ¤light phase 8:00®20:00¥. All procedures were conducted in nal mucosa: Experiments were performed as previously accordance with the Guiding Principles for the Care and Use described by Kageyama et al.11¥ with some modifications. of Laboratory Animals, adopted by the Japanese Pharmaco- Briefly, the ileal mucosa membrane was obtained from STZ- logical Society. Additionally, all experiments were approved or citrate buffer-treated ¤control¥ mice. After homogeniza- by the ethical committee for animals of Kobe Gakuin tion ¤400 rpm, 20 strokes¥ in homogenizing buffer, the University ¤approval number: A 060601-11¥. homogenate was centrifuged at 3,000 g for 10 min at 4ôC. Mice were injected with streptozotocin ¤STZ; Sigma, The supernatant was then further centrifuged at 15,000 g MO., USA¥¤230 mg/kg, i.p.¥ dissolved in citrate buffer for 15 min at 4ôC. Residual membrane fractions were ¤pH 4.28¥. On the 9th and 15th day after STZ admin- resuspended with lysis buffer. Protein concentrations were istration, mice with non-fasting blood glucose levels above measured using the Lowry method ¤DC Protein Assay kit II, 400 mg/dL were used in the study. Control mice were Bio-Rad, CA, USA¥. The membrane fraction was used for P- injected with citrate buffer ¤0.1 mL/10 g¥. gp expression quantification. Drug administration: Morphine ¤Takeda, Co., Ltd., Western blot analysis for intestinal P-gp expres- Osaka, Japan; 30 mg/kg¥ or vehicle ¤water¥ was provided sion: Western blot analysis was carried out as previously orally to mice. Additionally, morphine was administered described.12¥ Briefly, proteins extracted from the ileal subcutaneously ¤1®10 mg/kg¥ or intravenously ¤0.5®5 mucosal membrane fraction ¤50 µg/lane¥ were separated mg/kg¥. Aminoguanidine 1 mg/mL ¤AG; Sigma, MO., by electrophoresis on a 7.5% sodium dodecyl sulfate- USA¥, a specific iNOS inhibitor, was added to drinking polyacrylamide gel and then electrophoretically transferred water and provided to mice ad libitum for 9 days immediately onto a nitrocellulose membrane. After blocking in blocking after STZ administration. buffer containing Tris-buffered saline ¤TBS, pH 7.6¥, 0.1% Tail-flick test: Morphine analgesia against noxious Tween20 and 5% blocking agent ¤GE Healthcare UK Ltd, thermal stimuli was assessed with the tail-flick test. Mice Little Chalfont, England¥, the membrane was incubated with were gently held with the tail positioned in the tail-flick primary antibodies for P-gp ¤mAb C219, 1:200 dilution; apparatus ¤MK-330B; Muromachi Kikai Co., Ltd., Tokyo, Calbiochem, CA, USA¥ and glyceraldehyde-3-phosphate Japan¥ for radiant thermal stimulation of the dorsal surface of dehydrogenase ¤GAPDH¥¤clone 6C5, 1:20,000 dilution; the tail. The intensity of the thermal stimulus was adjusted Chemicon, CA, USA¥. The membrane was then incubated to cause the animal to flick its tail within 3®4 s as the base with a horseradish peroxidase-conjugated anti-mouse secondary line of the tail-flick latency. The tail-flick latency was antibody ¤1:2,000 dilution; Kierkegaard & Perry Laborato- measured before and 30, 60 and 90 min after morphine ries, MD, USA¥. Immunoreactive bands were visualized administration. The cutoff time was set at 10 s to minimize using Light Capture ¤ATTO, Tokyo, Japan¥ with an tissue damage. The area under the curve ¤AUC¥ value for the enhanced chemiluminescent substrate for horseradish perox- morphine analgesia on each mouse was calculated for the idase detection ¤ECL Western Blotting system, GE Health- relevant experiments. care UK Ltd¥. Signal intensity of the immunoreactive bands

A B C

Fig. 1. Intestinal P-gp expression levels and the analgesic effect of oral morphine in STZ-treated mice A typical Western blot image (A) and the effects of oral morphine analgesia (B and C) are shown. Mice were treated with morphine (30 mg/kg, p.o.). The cutoff time for the tail-flick experiments was set as 10 s to prevent injury to the tail. Data are presented as mean + SEM. Control (water) n ¦ 10, control (morphine) n ¦ 10, STZ (water) n ¦ 9, and STZ (morphine) n ¦ 10, *p g 0.05, **p g 0.01 vs. STZ (water), #p g 0.05 vs. control (morphine), according to Scheffe’stest. was determined by using specialized software ¤CS-Analyzer 30 min after subcutaneous administration of morphine version 3.0, ATTO¥. ¤5 mg/kg¥¤Figs. 3B and 3C¥. Statistical analysis: Data are expressed as means + Dose-dependent changes in serum and brain SEM. Statistical significance was assessed using an unpaired morphine levels in STZ-treated mice following Studentös t test or one-way analysis of variance ¤ANOVA¥ intravenous morphine administration: There were followed by Scheffeös test. Differences were regarded as dose-dependent changes in both serum and brain morphine statistically significant at p g 0.05. levels following intravenous administration of morphine in both control and STZ-treated mice ¤Fig. 4¥. Furthermore, Results there were no differences in the slopes of the dose-response Intestinal P-gp expression levels and analgesic curves between control and STZ-treated mice. effects of oral morphine in STZ-treated mice: Ileal Effects of an iNOS-specific inhibitor on ileal P-gp P-gp expression levels were significantly lower in STZ- expression levels and oral morphine analgesia in treated mice compared to control mice ¤Fig. 1A¥, and tail- STZ-treated mice: The significant decrease in P-gp flick latency following oral morphine administration was expression levels, which is apparent 9 days after STZ significantly greater in STZ-treated versus control mice administration, was suppressed by AG, a selective iNOS ¤Fig. 1B¥. Additionally, the analgesic effects of morphine, as inhibitor. However, AG had no effects on P-gp expression determined by the area under the tail-flick latency curve levels in control mice ¤Fig. 5A¥. ¤ ¥ fi fi AUC0®90 , were signi cantly greater in STZ-treated mice Furthermore, the signi cant increase in the analgesic than in control mice ¤Fig. 1C¥. effects of oral morphine were suppressed by AG in STZ- Serum morphine concentrations and brain mor- treated mice. However, AG had no effects on analgesia phine content in STZ-treated mice following oral induced by oral morphine in control mice ¤Fig. 5B¥. morphine administration: Serum morphine concen- Interestingly, AG treatment did not alter the analgesic trations 30 min after oral morphine administration were effects of subcutaneous morphine in STZ-treated mice significantly greater in STZ-treated mice than in control ¤Fig. 5C¥. mice ¤Fig. 2A¥. Similarly, brain morphine content was also Effects of an iNOS-specific inhibitor on serum significantly higher in STZ-treated mice ¤Fig. 2B¥. How- and brain morphine levels in STZ-treated mice ever, the brain/plasma ratio was not different between after oral morphine administration: The increase in control and STZ-treated mice ¤Fig. 2C¥. serum morphine concentrations and brain morphine content Analgesic effects of subcutaneous morphine and 30 min after oral morphine administration was suppressed by the serum morphine concentrations and brain AG in STZ-treated mice ¤Fig. 6¥. morphine content in STZ-treated mice: The an- Analgesic effects of oral morphine and serum and algesic effects of subcutaneous morphine ¤1®10 mg/kg¥ in brain morphine levels 15 days after STZ treatment: STZ-treated mice were not significantly different from On the 15th day after STZ treatment, ileal P-gp expression control mice ¤Fig. 3A¥. Additionally, serum morphine levels were not significantly different from control mice concentrations and brain morphine content were not ¤Fig. 7A¥, and the analgesic effects of oral morphine were significantly different between control and STZ-treated mice similar in both control and STZ-treated mice ¤Fig.7B¥.

A B C * ** 400 120 0.4 350 100 300 0.3 80 250 200 60 0.2 150 40 100 0.1 20 Brain/plasma ratio 50 Serum concentration (ng/mL)

0 Brain contents (ng/g wet tissue) 0 0 Control STZ Control STZ Control STZ Morphine (30 mg/kg, p.o.) Morphine (30 mg/kg, p.o.) Morphine (30 mg/kg, p.o.)

Fig. 2. Serum morphine concentrations and brain morphine content in STZ-treated mice following oral morphine administration Serum morphine concentrations (A), brain morphine contents (B) and the brain/plasma ratio (C) are shown. Blood and brains were harvested from control and STZ-treated mice 30 min after morphine administration (30 mg/kg, p.o.). Data are presented as mean + SEM. Control n ¦ 15, STZ n ¦ 21, *p g 0.05, **p g 0.01 vs. control, unpaired Student’s t test.

A B C

Fig. 3. The analgesic effects of subcutaneous morphine, serum morphine concentrations and brain morphine content in STZ-treated mice Effects of subcutaneous morphine analgesia (A), serum morphine concentration (B) and brain morphine contents (C) are shown. Mice were treated with morphine subcutaneously (1–10 mg/kg, s.c.). The cutoff time for the tail-ﬂickexperimentswassetas10stopreventinjurytothe tail. Blood and brain samples were harvested from mice 30 min after morphine administration (5 mg/kg, s.c.). Data are presented as mean + SEM. (A): Control (1 mg/kg) n ¦ 7; control (3 mg/kg) n ¦ 7; control (5 mg/kg) n ¦ 9; control (10 mg/kg) n ¦ 8; STZ (1 mg/kg) n ¦ 6; STZ (3 mg/kg) n ¦ 7; STZ (5 mg/kg) n ¦ 11; and STZ (10 mg/kg) n ¦ 9. (B and C): control n ¦ 6; and STZ n ¦ 9.

Additionally, serum morphine concentrations and brain decreased. These findings suggest that intestinal absorption morphine content 30 min after oral morphine administration of morphine is enhanced in the diabetic condition because were not significantly different between control and STZ- the drug efflux activity of P-gp is down-regulated. Although treated mice ¤Figs. 7C and 7D¥. there are other types of drug-sensitive transporters,2,14¥ only a few reports indicate morphine as their substrate. Thus, Discussion P-gp appears to be a key regulator of the pharmacological Although numerous drugs, including morphine, are effects of morphine. available in an oral dosage form, the bioavailability of orally Nine days after STZ treatment, the brain content and administered drugs is lower than that of subcutaneous or analgesic effects of morphine were significantly increased intravenous forms due to the first-pass effect within the following oral morphine administration. Since the analgesic intestines and liver.13¥ In the present study, it was found that effects of morphine are dependent on the binding of the analgesic effects of oral morphine were significantly morphine to its receptor within the central nervous system, altered in STZ-treated mice. Given that the analgesic effects it may be that the enhancement of morphine analgesia in of oral morphine appear to be coupled to intestinal P-gp STZ-treated mice was a result of the breakdown of the BBB expression levels, then alterations in the analgesic effects of in the diabetic condition.15¥ Indeed, it has been reported that oral morphine may be due to changes in intestinal morphine- P-gp expression and activity levels within the BBB were absorptive processes. Indeed, serum morphine concentra- decreased in the diabetic condition.16¥ However, our results tions were significantly increased 9 days after STZ clearly demonstrate that the analgesic effect, serum treatment, at which time intestinal P-gp levels had concentration and brain content of morphine following its

intestinal absorption of morphine may increase morphine brain content and serum morphine concentrations, and thereby enhance its analgesic effects in STZ-treated mice. Contrary to our present findings, it was previously reported that the analgesic effects of subcutaneous and oral morphine were reduced in animals with diabetes-induced hyperalgesia.17,18¥ As observed in these previous studies,17,18¥ diabetes-induced hyperalgesia may also be present in our diabetic mouse model. In the present study, however, the enhancement of oral morphine analgesia associated with the decrease in intestinal P-gp expression levels 9 days after STZ treatment may mask the diabetes-induced hyperalgesia. On the other hand, since morphine levels following subcutaneous administration did not differ between the STZ-treated and control group, then the diabetes-induced decrease in Fig. 4. Dose-dependent changes in serum and brain morphine subcutaneous morphine analgesia could be observed when levels in STZ-treated mice following intravenous morphine administration estimated using more appropriate experimental methods for Mice were treated intravenously with morphine (0.5–5 mg/kg, i.v.), estimating hyperalgesia, such as tactile mechanical or weak and after 15 min, blood and brain samples were harvested. Data thermal stimuli. are presented as mean + SEM. Control (0.5 mg/kg) n ¦ 15; control With respect to the mechanism of the altered intestinal (1 mg/kg) n ¦ 14; control (2.5 mg/kg) n ¦ 12; control (5 mg/kg) n ¦ ¦ ¦ P-gp levels in the diabetic condition, we previously reported 13; STZ (0.5 mg/kg) n 9; STZ (1 mg/kg) n 4; STZ (2.5 mg/kg) 12,19¥ n ¦ 5; and STZ (5 mg/kg) n ¦ 3. the involvement of NOS. In particular, intestinal NOS activity gradually and continuously increased after STZ administration accompanied with an increase of blood subcutaneous administration ¤i.e., bypassing the intestinal glucose levels.12¥ Additionally, when mice were administered absorption process¥ were not different in STZ-treated versus NOR5, a known NO donor, to reproduce the increment of control mice. Moreover, the dose-dependent increase in NO levels in intestine, intestinal P-gp expression levels were serum concentrations and brain content of morphine significantly decreased, clearly suggesting that NOS activa- following intravenous administration of morphine in STZ- tion is essential for the significant decrease of intestinal P-gp treated mice did not differ from those of control mice. Thus, expression levels. Importantly, it has been reported that it appears that the BBB had not broken down 9 days after úhigh glucose stressû per se can increase NOS activity STZ administration. In fact, after oral morphine admin- in vitro.20¥ istration, the brain/plasma ratio of morphine contents was As is well known, NOS is classified into three isoform not different between control and STZ-treated mice. This types as neuronal NOS ¤nNOS¥, endothelial NOS ¤eNOS¥ finding strongly support the hypothesis that the altered and iNOS.21¥ In our previous study, we reported that

A B C

Fig. 5. Effects of an iNOS-speciﬁc inhibitor on ileal P-gp expression levels and analgesic effects of oral morphine analgesia in STZ- treated mice A typical Western blot image (A), effects of oral morphine analgesia (B) and subcutaneous morphine analgesia (C) are shown. Mice were treated with morphine (30 mg/kg, p.o.). The cutoff time for the tail-ﬂick experiments was set as 10 s to prevent injury to the tail. Aminoguanidine (AG; 1 mg/mL) was provided in the drinking water for 9 days. Data are presented as the mean + SEM. (B): Control (water/water) n ¦ 8; control (morphine/water) n ¦ 10; control (morphine/AG) n ¦ 13; STZ (water/water) n ¦ 6; STZ (morphine/water) n ¦ 11; STZ (morphine/AG) n ¦ 12. **p g 0.01 vs. Control (water/water), ##p g 0.01 vs. STZ (water/water), ††p g 0.01 vs. STZ (morphine/water), Scheffe’s test. (C): saline (SAL) (water) n ¦ 5; SAL (AG) n ¦ 4; morphine (water) n ¦ 5; and morphine (AG) n ¦ 4. **p g 0.01 vs. SAL, Scheffe’stest.

A B

Fig. 6. Effects of an iNOS-speciﬁc inhibitor on serum and brain morphine levels in STZ-treated mice following oral morphine administration Serum morphine concentration (A), brain morphine contents (B). Blood and brain samples were harvested 30 min after morphine administration (30 mg/kg, p.o.). Aminoguanidine (AG; 1 mg/mL) was provided in the drinking water for 9 days. Data are presented as mean + SEM. Control n ¦ 14; STZ (Water) n ¦ 14; and STZ (AG) n ¦ 19, *p g 0.05, **p g 0.01 vs. Control, #p g 0.05, ##p g 0.01 vs. STZ (Water), Scheffe’stest.

A B C D

Fig. 7. The analgesic effects of morphine, serum morphine concentrations and brain morphine content after oral morphine administration 15 days after STZ administration A typical Western blot image (A), effects of oral morphine analgesia (B) serum morphine concentration (C), and brain morphine contents (D) are shown. Mice were treated with morphine (30 mg/kg, p.o.). The cutoff time for the tail-ﬂickexperimentswassetat10sinthetesttoavoid injury to the tail. Blood and brain were harvested 30 min after morphine administration (30 mg/kg, p.o.). Data are presented as mean + SEM. (B): Control (water) n ¦ 5; control (morphine) n ¦ 7; STZ (water) n ¦ 7; and STZ (morphine) n ¦ 14.(CandD):controlmice,n¦ 11; and STZ- treated mice, n ¦ 14. activation of iNOS plays an important role in decreasing on alterations in intestinal P-gp expression. Although the intestinal P-gp levels in STZ-treated mice.12¥ Thus, in the precise mechanism for alteration of intestinal P-gp present study, we used AG to inhibit decreases in intestinal expression via iNOS activation under diabetic conditions P-gp expression levels. Since AG has anti-glycation effects,22¥ was not fully elucidated in this study, it is expected that it may also affect the diabetic state; however, in our previous both transcriptional and post-translational regulation may study, AG was found not to affect blood glucose levels.12¥ be involved in the mechanisms. For example, several Despite the fact that AG was previously reported to transcription factors, such as NF-ÏB, YB-1 and PXR attenuate the development of morphine tolerance and that could bind elements of mdr1 promoter, might be dependence,23¥ it did not affect the analgesic effects of repressed under NOS activation. With respect to post- morphine,23¥ as was shown in the present study. Thus, AG translational regulation, we now hypothesize the involve- treatment appears to completely suppress the enhancement ment of the ubiquitin-proteasome system that could be of morphine analgesia in the diabetic state. It is hypothesized activated by the NOS-mediated pathway24¥ in the degrada- that the analgesic effects of oral morphine are dependent tion of P-gp.

Furthermore, 15 days after STZ treatment, ileal P-gp 3¥ Maeng, H. J., Kim, M. H., Jin, H. E., Shin, S. M., Tsuruo, T., expression levels, oral morphine analgesic effects and serum Kim, S. G., Kim, D. D., Shim, C. K. and Chung, S. J.: Functional induction of P-glycoprotein in the blood-brain barrier of and brain morphine levels had recovered back to control streptozotocin-induced diabetic rats: evidence for the involvement levels. Interestingly, our previous study clearly demon- of nuclear factor-kappaB, a nitrosative stress-sensitive transcription strated that this recovery of ileal P-gp expression levels was factor, in the regulation. Drug Metab. Dispos., 35: 1996®2005 also mediated by NOS activation.19¥ In particular, when mice ¤2007¥. ¥ were administered L-NAME, a non-selective NOS inhibitor, 4 Tomita, M., Takizawa, Y., Kishimoto, H. and Hayashi, M.: Assessment of ileal epithelial P-glycoprotein dysfunction induced by during the 9th to 15th day after STZ administration only, the ischemia/reperfusion using in vivo animal model. Drug Metab. recovery of P-gp expression levels observed on the 15th day Pharmacokinet. ® ¤ ¥ ¥ , 23: 356 363 2008 . after STZ administration was completely suppressed.19 5¥ Kameyama, N., Arisawa, S., Ueyama, J., Kagota, S., Shinozuka, These findings suggest that NOS is involved not only in K., Hattori, A., Tatsumi, Y., Hayashi, H., Takagi, K. and the decrease but also in the recovery of intestinal P-gp Wakusawa, S.: Increase in P-glycoprotein accompanied by activation of protein kinase Calpha and NF-kappaB p65 in the expression levels observed on the 9th and 15th days after livers of rats with streptozotocin-induced diabetes. Biochim. Biophys. STZ administration, respectively. Interestingly, it has been Acta, 1782: 355®360 ¤2008¥. reported that the effect of NO on P-gp function and 6¥ Zong, J. and Pollack, G. M.: Morphine antinociception is enhanced expression levels was different between short-term and in mdr1a gene-deficient mice. Pharm. Res., 17: 749®753 ¤2000¥. ¥ long-term exposure to NO in vitro. Specifically, short-term 7 Hamabe, W., Maeda, T., Kiguchi, N., Yamamoto, C., Tokuyama, S. and Kishioka, S.: Negative relationship between morphine exposure to NO in rat brain capillary endothelial cells analgesia and P-glycoprotein expression levels in the brain. J. decreased the P-gp function through PKC activation, while Pharmacol. Sci., 105: 353®360 ¤2007¥. long-term exposure to NO increased the function and 8¥ Tada, Y., Wada, M., Kuroiwa, K., Kinugawa, N., Harada, T., expression levels of P-gp.25¥ Nagayama, J., Nakagawa, M., Naito, S. and Kuwano, M.: MDR1 In conclusion, our results clearly indicate that the gene overexpression and altered degree of methylation at the fi promoter region in bladder cancer during chemotherapeutic analgesic effects of oral morphine are signi cantly enhanced treatment. Clin. Cancer Res., 6: 4618®4627 ¤2000¥. by the iNOS-mediated transient decrease in intestinal P-gp 9¥ Komura, T., Sakai, Y., Honda, M., Takamura, T., Matsushima, K. expression. Furthermore, expression levels of P-gp appear and Kaneko, S.: CD14¦ monocytes are vulnerable and functionally to fluctuate throughout the developmental course of impaired under endoplasmic reticulum stress in patients with type 2 ® ¤ ¥ diabetes. Unfortunately, there are few descriptions of diabetes. Diabetes, 59: 634 643 2010 . 10¥ Shimizu, N., Kishioka, S., Maeda, T., Fukazawa, Y., Yamamoto, clinical evidence for the changes of pharmacokinetics of C., Ozaki, M. and Yamamoto, H.: Role of pharmacokinetic effects P-gp-substrate drugs in diabetic patients. As we have in the potentiation of morphine analgesia by L-type calcium channel shown here, the significant changes in pharmacokinetics blockers in mice. J. Pharmacol. Sci., 94: 240®245 ¤2004¥. and pharmacodynamics of oral morphine in the diabetic 11¥ Kageyama, M., Fukushima, K., Togawa, T., Fujimoto, K., Taki, condition appear to be period- and severity-dependent. This M., Nishimura, A., Ito, Y., Sugioka, N., Shibata, N. and Takada, K.: Relationship between excretion clearance of rhodamine 123 and may make it hard to obtain clear and critical evidence in the P-glycoprotein ¤Pgp¥ expression induced by representative Pgp clinical setting. This study underlines the fact that it is inducers. Biol. Pharm. Bull., 29: 779®784 ¤2006¥. important for physicians to be aware of the hidden 12¥ Nawa, A., Fujita-Hamabe, W. and Tokuyama, S.: Inducible nitric pharmacological changes of P-gp-substrate drugs in diabetic oxide synthase-mediated decrease of intestinal P-glycoprotein patients during advancement of the clinical stage. In expression under streptozotocin-induced diabetic conditions. Life Sci., 86: 402®409 ¤2010¥. summary, alteration of intestinal P-gp expression levels in 13¥ Han, M. and Fang, X. L.: Difference in oral absorption of the diabetic condition may be one of the reasons that it is ginsenoside Rg1 between in vitro and in vivo models. Acta difficult to obtain stable pharmacological effects of morphine Pharmacol. Sin., 27: 499®505 ¤2006¥. in diabetic patients. 14¥ Ganapathy, V. and Miyauchi, S.: Transport systems for opioid peptides in mammalian tissues. AAPS J., 7: E852®E856 ¤2005¥. 15¥ Mogi, M. and Horiuchi, M.: Neurovascular coupling in cognitive Acknowledgments: This study was, in part, supported impairment associated with diabetes mellitus. Circ. J., 75: 1042® by grants-in-aid and special coordination funds from the 1048 ¤2011¥. Academic Frontier Project of the Cooperative Research 16¥ Liu, H., Xu, X., Yang, Z., Deng, Y., Liu, X. and Xie, L.: Impaired Center of Life Sciences. function and expression of P-glycoprotein in blood-brain barrier of streptozotocin-induced diabetic rats. Brain Res., 1123: 245®252 References ¤2006¥. 17¥ Ibironke, G. F. and Sada, O. J.: Effect of hyperglycemia on the 1¥ Glare, P.: Choice of opioids and the WHO Ladder. J. Pediatr. efficacy of morphine analgesia in rats. Afr. J. Med. Med. Sci., 35: Hematol. Oncol., 33:S6®S11 ¤2011¥. 443®445 ¤2006¥. 2¥ Zelcer, N., van de Wetering, K., Hillebrand, M., Sarton, E., Kuil, 18¥ Courteix, C., Bourget, P., Caussade, F., Bardin, M., Coudore, F., A., Wielinga, P. R., Tephly, T., Dahan, A., Beijnen, J. H. and Fialip, J. and Eschalier, A.: Is the reduced efficacy of morphine in Borst, P.: Mice lacking multidrug resistance protein 3 show altered diabetic rats caused by alterations of opiate receptors or of morphine pharmacokinetics and morphine-6-glucuronide antinoci- morphine pharmacokinetics? J. Pharmacol. Exp. Ther., 285:63®70 ception. Proc. Natl. Acad. Sci. USA, 102: 7274®7279 ¤2005¥. ¤1998¥.

19¥ Nawa, A., Fujita-Hamabe, W. and Tokuyama, S.: Regulatory the formation of advanced glycation endproducts. Arch. Biochem. action of nitric oxide synthase on ileal P-glycoprotein expression Biophys., 419:31®40 ¤2003¥. under streptozotocin-induced diabetic condition. Biol. Pharm. Bull., 23¥ Abdel-Zaher, A. O., Hamdy, M. M., Aly, S. A., Abdel-Hady, 34: 436®438 ¤2011¥. R. H. and Abdel-Rahman, S.: Attenuation of morphine tolerance 20¥ Carmo, A., Cunha-Vaz, J. G., Carvalho, A. P. and Lopes, M. C.: and dependence by aminoguanidine in mice. Eur. J. Pharmacol., 540: Effect of cyclosporin-A on the blood®retinal barrier permeability in 60®66 ¤2006¥. streptozotocin-induced diabetes. Mediators Inﬂamm., 9: 243®248 24¥ Xu, J., Wu, Y., Song, P., Zhang, M., Wang, S. and Zou, M. H.: ¤2000¥. Proteasome-dependent degradation of guanosine 5$-triphosphate 21¥ Leal, E. C., Manivannan, A., Hosoya, K., Terasaki, T., Cunha- cyclohydrolase I causes tetrahydrobiopterin deﬁciency in diabetes Vaz, J., Ambrósio, A. F. and Forrester, J. V.: Inducible nitric oxide mellitus. Circulation, 116: 944®953 ¤2007¥. synthase isoform is a key mediator of leukostasis and blood-retinal 25¥ Bauer, B., Hartz, A. M. and Miller, D. S.: Tumor necrosis factor barrier breakdown in diabetic retinopathy. Invest. Ophthalmol. Vis. alpha and endothelin-1 increase P-glycoprotein expression and Sci., 48: 5257®5265 ¤2007¥. transport activity at the blood-brain barrier. Mol. Pharmacol., 71: 22¥ Thornalley, P. J.: Use of aminoguanidine ¤Pimagedine¥ to prevent 667®675 ¤2007¥.