Effects of a Water-Soluble Curcumin Protein Conjugate vs. PureCurcumin in a Diabetic Model of Erectile Dysfunctionjsm_2741 1815..1833

Mohamed Talaat Abdel Aziz, MD,* Tarek Motawi, PhD,† Ameen Rezq, PhD,* Taymour Mostafa, MD,‡

Hanan H. Fouad, MD,* Hanan H. Ahmed, MD,* Laila Rashed, MD,* Dina Sabry, MD,*Amira Senbel, PhD,§ A Al-Malki, PhD,¶ and Raghda El-Shafiey, MSc**

*Medical Biochemistry Department, Unit of Biochemistry and Molecular Biology, Faculty of Medicine, Cairo University,Cairo, Egypt; †Biochemistry Department, Faculty of Pharmacy, Cairo University, Cairo, Egypt; ‡Andrology & SexologyDepartment, Faculty of Medicine, Cairo University, Cairo, Egypt; §Pharmacology and Toxicology Department, Faculty ofPharmacy, Alexandria University, Alexandria, Egypt; ¶Biochemistry Department, Faculty of Science, King AbdualzizUniversity, Jeddah, Kingdom of Saudi Arabia; **Medical Biochemistry Department, Faculty of Pharmacy, Beni SuefUniversity, Beni Suef, Egypt

DOI: 10.1111/j.1743-6109.2012.02741.x

A B S T R A C T

Introduction. Curcumin is involved in erectile signaling via elevation of cyclic guanosine monophosphate (cGMP).Aim. Assessment of the effects of water-soluble curcumin in erectile dysfunction (ED).Methods. One hundred twenty male white albino rats were divided into: 1st and 2nd control groups with or withoutadministration of Zinc protoporphyrin (ZnPP), 3rd and 4th diabetic groups with or without ZnPP, 5th diabeticgroup on single oral dose of pure curcumin, 6th diabetic group on pure curcumin administered daily for 12 weeks,7th and 8th diabetic groups on single dose of water-soluble curcumin administered with or without ZnPP, 9th and10th diabetic groups on water-soluble curcumin administered daily for 12 weeks with or without ZnPP. All curcumindosage schedules were administered after induction of diabetes.Main Outcome Measures. Quantitative gene expression of endothelial nitric oxide synthase (eNOS), neuronal NOS(nNOS), inducible NOS (iNOS), heme oxygenase-1 (HO-1), nuclear transcription factor-erythroid2 (Nrf2),NF-Kb, and p38. Cavernous tissue levels of HO and NOS enzyme activities, cGMP and intracavernosal pressure(ICP).Results. Twelve weeks after induction of diabetes, ED was confirmed by the significant decrease in ICP. There wasa significant decrease in cGMP, NOS, HO enzymes, a significant decrease in eNOS, nNOS, HO-1 genes and asignificant elevation of NF-Kb, p38, iNOS genes. Administration of pure curcumin or its water-soluble conjugate ledto a significant elevation in ICP, cGMP levels, a significant increase in HO-1 and NOS enzymes, a significantincrease in eNOS, nNOS, HO-1, and Nrf2 genes, and a significant decrease in NF-Kb, p38, and iNOS genes.Water-soluble curcumin showed significant superiority and more prolonged duration of action. Repeated dosesregimens were superior to single dose regimen. Administration of ZnPP significantly reduced HO enzyme, cGMP,ICP/ mean arterial pressure (MAP), HO-1 genes in diabetic groups.Conclusion. Water-soluble curcumin could enhance erectile function with more effectiveness and with more pro-longed duration of action. Abdel Aziz MT, Motawi T, Rezq A, Mostafa T, Fouad HH, Ahmed HH, Rashed L,Sabry D, Senbel A, Al-Malki A, and El-Shafiey R. Effects of a water-soluble curcumin protein conjugate vs.pure curcumin in a diabetic model of erectile dysfunction. J Sex Med 2012;9:1815–1833.

Key Words. Diferuloylmethane; cGMP; Heme Oxygenase; Erectile Dysfunction; Diabetes Mellitus; IntracavernosalPressure

1815

© 2012 International Society for Sexual Medicine J Sex Med 2012;9:1815–1833

Introduction

E ndothelial dysfunction plays a preponderantrole in a considerable number of vasculogenic

erectile dysfunction (ED) cases. The endotheliumis considered a primary mediator of vascular bloodflow and muscle tone [1–4]. In diabetes, endothe-lial dysfunction is regarded as important factor inthe pathogenesis of diabetic micro-and macroan-giopathy. There are three main sources contribut-ing to endothelial dysfunction in diabetes:hyperglycemia which influences endothelial cellfunctions indirectly by the synthesis of growthfactors, vasoactive agents, and the components ofthe metabolic syndrome [5].

Numerous studies showed that high glucoselevels in diabetes activate protein kinase-C (PKC)in vascular endothelial cells [6]. Hyperglycemiacauses de novo synthesis of diacylglycerol, leadingto the activation of PKC, a pathway now demon-strated in all vascular tissues involved in diabeticcomplications [7]. The reported deleterious effectof PKC on endothelial function is partially medi-ated by inducing phosphorylation of p115RhoGEF[8], a guanine nucleotide exchange factor for RhoGTPase [9]. Because active RhoA is implicated inarginase induction [10], it was suggested that PKCmight also be involved in regulation of arginaseactivity leading to decrease in nitric oxide (NO)bioavailability because arginase compete with NOsynthase (NOS) for a common substrate: arginine.

In addition, activated PKC results in sustainedincreases in the production of superoxide anion(O2

•-) and induces oxidative damage to diabeticblood vessels. It also induces a number of patho-genic consequences by activating NF-Kb andaffecting the expression of endothelial NOS(eNOS), endothelin-1 (ET-1), vascular endothelialgrowth factor, transforming growth factor-b andplasminogen activator inhibitor-1 [11]. Rungsee-santivanon et al. [12] stated that curcumin supple-mentation could improve diabetes-inducedendothelial dysfunction significantly in relation toits potential to decrease superoxide productionand PKC inhibition. This finding was alsoreported by Balasubramanyam et al. [13], whoconfirmed that the dose-dependent ROS inhibi-tory effect of curcumin interfered mechanisticallywith PKC activity. Okamoto et al. [14] stated thatcurcumin completely prevents the advanced glyca-tion end products (AGE) and AGE-receptors-induced increase in NF-Kb and activator protein 1(AP-1) activity and blocks microvascular compli-cations in diabetes.

Different mechanisms of endothelial dysfunc-tion in diabetes and effects of curcumin on differ-ent signaling molecules affecting endothelialfunctions are illustrated in Figures 1 and 2.

Abdel Aziz et al. [2] reported that a novel water-soluble derivative of curcumin could mediate erec-tile function via induction of heme oxygenase-1(HO-1) enzyme with subsequent upregulation ofcyclic guanosine monophosphate (cGMP) levels incavernous tissue. The inducible enzyme HO-1metabolizes heme to biliverdin and carbon monox-ide (CO). Several studies proved that CO exhibitsphysiological properties similar to NO mediated inpart by the ability of CO to act as an activator ofsoluble guanylate cyclase (sGC) [10]. Abdel Azizet al. [15–20] proved that HO and its product COdominates NO as a signaling molecule in erectilefunction and could partially mediate the actions ofphosphodiesterase inhibitors.

Curcumin is one of the nuclear transcriptionfactor-erythroid2-related factor 2-Kelch-likeECH-associated Protein1 (Keap1)-antioxidant-response element (Nrf2-Keap1-ARE) signalingpathway activators inducing de novo synthesis ofphase II detoxifying genes including HO-1 gene[21–24]. Curcumin effect on erectile function couldbe mediated via upregulation of HO-1. Andreadiet al. [22] showed the involvement of Nrf2, AP-1(via extracellular signal-regulated kinases; ERK),protein 38 mitogen-activated protein kinases; p38,and NF-Kb in induction of HO-1 by dietarypolyphenols including curcumin [25–27]. Xu et al.[28], stated that curcumin induces relaxation ofisolated precontracted porcine coronary arteriesand this effect is partially dependent on the actionof NO, cGMP, and adrenergic b-receptor. Hence,it is reasonable to assume that the relaxing effect ofcurcumin on porcine coronary arteries might berelated to its antioxidant effect which can protectthe coronary arteries from oxidative stress byblocking the production of superoxide anion andpromote NO release. Moreover, Schini-Kerthet al. [29], stated that several polyphenols as cur-cumin are able to induce NO-mediatedendothelium-dependent relaxations in a largenumber of arteries including the coronary artery;they can also induce endothelium-derived hyper-polarizing factor (EDHF)-mediated relaxations insome of these arteries.

Clinical trials in humans indicated that the sys-temic bioavailability of orally administered cur-cumin is relatively low [30]. Curcumin and itsmetabolites could not be detected in plasma atdoses lower than 3.6 g/day. Several water-soluble

1816 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

curcumin derivatives were prepared to achieveclinically efficient systemic bioavailability.

The present study was conducted to compare themolecular and physiological effects of a novelmember of a series of water-soluble curcuminprotein conjugates vs. pure curcumin in an experi-mental diabetic model of ED, with assessment of themolecular signaling targets of pure curcumin and itswater-soluble protein conjugate was also conducted.

Methods

The study was conducted in the Unit of Biochem-istry and Molecular Biology (UBMB), Faculty ofMedicine, Cairo University. Curcumin proteinconjugate was presented free of charge to the par-

ticipating researchers as a personal nonprofit scien-tific donation to help advancement of cooperationin national medical research. The novel curcuminconjugate is registered as international patent pro-tected by the rights of “The Patent CooperationTreaty” under: (PCT/EG2010/000008, PublishedPatent Pending, WO 2011/100984) and is the per-sonal property of its inventors.

One hundred and twenty adult male whitealbino rats (Cux1: HEL1) of matched age andweight (180–200 g) were included in the studyafter the approval of the Institutional Animal Careand Use Committee.

Experimental rat groups were preferably classi-fied into: single and repeated doses regimengroups and subgroups.

Figure 1 Molecular signaling of cur-cumin in diabetes-induced endothelialdysfunction.

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1817

J Sex Med 2012;9:1815–1833

Animals were equally divided into 10 groups:1st normal control group, 2nd streptozotocin(STZ)-induced diabetic group, 3rd STZ-induceddiabetic rat group on 10 mg/kg body weight (BW)single oral dose of pure curcumin administered 12weeks after induction of diabetes, 4th STZ-induced diabetic rat group on the same oral doseof pure curcumin administered daily for 12 weeksfollowing induction of diabetes, 5th and 6th STZ-induced diabetic rat groups on 2 mg/kg BW singleoral dose of curcumin protein conjugate adminis-tered 12 weeks after induction of diabetes with orwithout administration of 50 mg/kg BW intrap-eritoneal (IP) Zinc protoporphyrin (ZnPP) as HOinhibitor, 7th and 8th STZ-induced diabetic ratgroups on the same dose of curcumin protein con-jugate (water-soluble curcumin) administereddaily for 12 weeks following induction of diabeteswith or without administration of ZnPP, 9thgroup involved control rats that received ZnPP,10th group involved STZ-induced diabetic ratsthat received ZnPP. Animals were euthanized 24hours, 48 hours, and 1 week after the last admin-istered dose of pure curcumin or its conjugate.Diabetes was induced by a single IP injection of

65 mg/kg BW STZ dissolved in 0.1 mM sodiumcitrate buffer, pH 4.8 and left untreated [31]. Dia-betes induction was confirmed by assessment ofblood glucose after 1 week and followed by assess-ment of glycated hemoglobin 12 weeks afterinduction of diabetes. The treatment phase of thestudy lasted 12 weeks according to the findingsreported by Li et al. [32], who stated that ED isestablished 12 weeks after diabetic induction inrats. Curcumin and ZnPP doses were chosenaccording to previous studies [1,2]. Wash-offperiod of curcumin was 1 week as assessed in aprevious study conducted by Abdel Aziz et al. [2].ZnPP was administered as a single dose for thespecified rat groups.

Animals were euthanized after 24 hours, 48hours, and 1 week of the starting treatment. Thefollowing parameters were assessed: intracavern-osal HO-1 gene expression [20], HO-1 enzymeactivity [20], NOS activity [33], cGMP [2], intra-cavernosal pressure (ICP) [3] as physiologic assess-ment of erectile function, and gene expression ofinducible NOS (iNOS), eNOS, neuronal NOS(nNOS), Nrf2, NF-Kb, and p38 gene by quantita-tive real-time PCR.

Figure 2 Diagrammatic illustration of the studied rat groups.

1818 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

Real-Time Quantitative Assessment of HO-1,iNOS, eNOS, nNOS, Nrf2, NF-Kb, andp38 Genes ExpressionTotal RNA was extracted from cavernous tissuehomogenate using RNeasy purification reagent(Qiagen, Valencia, CA, USA). cDNA was generatedfrom 5 mg of total RNA extracted with 1 mL(20 pmol) antisense primer and 0.8 mL superscriptavian myeloblastosis virus (AMV) reverse tran-scriptase for 60 minutes at 37°C. Quantitation ofgene expression was conducted using universalprobe library sets based real-time polymerase chainreaction (PCR) (Roche Diagnostics, Indianapolis,IN, USA). Selection of genes-specific probes andprimers was done using the online ProbeFinder soft-ware (Roche Diagnostics) and the real-time PCRdesign assay of Roche Diagnostics found in theirwebsite: http://www.universalprobelibrary.com.Primers used for quantitative real time PCR areshown in Table 1. Hypoxanthine phosphoribosyl-transferase 1 was used as a positive control house-keeping gene. FastStart Universal Probe Master mixwas used in LightCycler® 480 Instrument (RocheApplied Science, Indianapolis, IN, USA). Briefly, inthe LightCycler® 480, a total reaction volume of20 mL was prepared, of which 2 mL (equivalent to5 mg RNA) of starting RNA material was includedfor RT-PCR, a final concentration of 0.5 mM of eachforward and reverse primer and 0.2 mM of theTaqMan probe was used. Cycling conditions involve

reverse transcription at 50°C for 30 minutes; enzymeactivation at 95°C for 15 minutes, followed by 50cycles of 95°C for 10 seconds and 60°C for 60seconds. LightCycler® 480 RT-PCR data were ana-lyzed using LightCycler1.2 version 3.5 software(Roche Diagnostics) using the second derivativemaximum method. Successfully amplified targetsare expressed in Ct values, or the cycle at which thetarget amplicon is initially detected above back-ground fluorescence levels as determined by theinstrument software. Each sample RT-PCR was per-formed minimally in duplicate, and the mean Ctvalue with standard deviation was reported.

NOS Enzyme Activity AssayNOS enzyme activity was assayed in cavernoustissue by kit supplied by OXIS, International, Inc.(Portland, OR, USA) according to manufacturer’srecommendations. In brief, nitrites and nitrates wereused as markers for assay of NOS enzyme activity.Nitrate reductase is used for enzymatic reduction ofnitrate to nitrite. Spectrophotometric quantitationat 540 nm of nitrite using the Griess Reagents wasconducted as described by Moshag et al. [33].

cGMP AssaysFrozen cavernous tissue (CC) samples (50 mg)stored in a specific lysis buffer were grounded witha stainless steel mortar then homogenized andcentrifuged at 600 g at 4°C for 10 minutes. Thesupernatant was used for cGMP assay using theELISA kit supplied by R&D Systems, Inc. (Min-neapolis, MN, USA).

HO Enzyme Activity AssayCC (50 mg) homogenized samples were incubatedwith heme (50 mmol/L), rat liver cytosol (5 mg/mL), MgCl2 (2 mM/L), glucose-6-phosphatedehydrogenase (1U), glucose-6-phosphate(2 mM/L), and reduced nicotinamide adeninedinucleotide phosphate (NADPH) (0.8 mM/L) in0.5 mL of 0.1 M/L phosphate buffer saline(pH 7.4) for 60 minutes at 37°C. The reaction wasstopped by putting the tubes on ice, and the reac-tion solution was extracted with chloroform. Therate of bilirubin formation was monitored at464 nm and 520 nm by a spectrophotometer [34].

ICP AssessmentMeasuring ICP changes elicited by electricalstimulation of the cavernous nerve (CN) was per-formed in rats according to the method previouslydescribed [35,36]. Briefly, rats were anesthetized byIP injection of thiopental (50 mg/kg). The skin

Table 1 Primer sequences used for quantitative RT-PCR

Beta actin Forward primer ACCAGTTCGCCATGGATGACGATReverse primer CCTAGGGCGGCCCACGATGGGenBank Accession Number: NM_031144.2

HO-1 Forward primer CTGTTGGCGACCGTGGCAGTReverse primer CTGGGCTCAGAACAGCCGCCGenBank Accession Number NM_012580.2

Nrf2 Forward primer TAGCAGAGCCCAGTGGCGGTReverse primer TGCTCTGGGGATGCTCGGCTGenBank Accession Number: NM_031789.1

P38 Forward primer GGCCGATGGGAACTGCCGTCReverse primer CAGCCCTGTGCCTTGCCTGGGenBank Accession Number: NM_031020.2

NF-Kb Forward primer AACGGCCTTCTGCACAGCGGReverse primer CCAGGTAACAGGGCGTGGCCGenBank Accession Number: NM_001024872.1

eNOS Forward primer GAAGGCGTTTGACCCCCGGGReverse primer CCACCGCTCGAGCAAAGGCAGenBank Accession Number: NM_021838.2

nNOS Forward primer GGCACTGGCATCGCACCCTTReverse primer GAGGACATCGGCGGCCATGGGenBank Accession Number: NM_052799.1

iNOS Forward primer GTTCCCCCAGCGGAGCGATGReverse primer ACTCGAGGCCACCCACCTCCGenBank Accession Number: NM_012611.3

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1819

J Sex Med 2012;9:1815–1833

overlying the penis was incised and the prepuce wasdegloved to expose the copora cavernosa. A26-gauge needle filled with heparinized saline wascarefully inserted into the corpus cavernosum onone side to measure the ICP. The needle wasinserted at mid-length of the penile shaft pointingtoward its base. The needle was sealed to polyeth-ylene 50 tubing and connected to Gould-Stathampressure transducer (Oxnard, CA, USA) and ICPwas displayed on a Grass polygraph (Model 7D,Grass Inst. Co., MA, USA). Through a lowersuprapubic midline incision, the lateral prostatewas dissected and the major pelvic ganglion identi-

fied [37,38]. The CN was unilaterally freed from itsfacial attachments and stimulated electrically usinga bipolar platinum electrode placed 3 to 4 mm distalto the major pelvic ganglion [39]. The two poles ofthe electrode were separated by 2 mm and theelectrode was connected to an electronic stimulator(Letica, Panlab, Model 12106/150, Barcelona,Spain). The CN was stimulated for 1 minute; thepulse parameters were at 5 V at 1 millisecond dura-tion, and 0.3–5 Hz frequency. Each period ofstimulation was separated by a resting period of 5minutes. A frequency response curve was obtainedwhere the maximum rise in ICP during nerve

Figure 3 Heme oxygenase (HO)enzyme activity (pmol bilirubin/mgprotein) in rat cavernous tissue. *Sig-nificant difference in comparison withcontrol rats, #significant difference incomparison with diabetic rats.

1820 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

stimulation was measured and compared statisti-cally with healthy control at each frequency.

Statistical AnalysisStatistical Package for Social Studies (SPSS)program version 16.0.1 (SPSS Inc., Chicago, IL,USA) was used. Numerical data were expressed asmean � standard deviation. For comparisonsbetween treatment groups, the null hypothesiswas tested by a single-factor anova for multiplegroups or unpaired t-test for two groups. Com-parisons were considered statistically significant ifP < 0.05.

Results

Twelve weeks after experimentally induced diabe-tes in rat, ED was confirmed by assessment ofICP which was found to be significantlydecreased in comparison with control rats. Thiswas accompanied with a significant decrease incavernous tissue-cGMP levels, NOS, HO enzymeactivities (Figures 3–5), a significant decrease ingene expression of eNOS, nNOS, and HO-1 anda significant increase in gene expression ofNF-Kb, p38, and iNOS in STZ-induced diabeticrats in comparison with the control group(Figures 6–14).

Figure 4 Cyclic guanosine mono-phosphate (cGMP) cavernous tissuelevels (pmol/mg protein). *Significantdifference in comparison with controlrats, #significant difference in compari-son with diabetic rats.

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1821

J Sex Med 2012;9:1815–1833

Administration of either pure curcumin or itswater-soluble protein conjugate led to a significantelevation in ICP, cGMP levels at all time intervals, asignificant increase in enzymatic activities of HO-1(up to 48 hours with pure curcumin and up to 1week with curcumin conjugate in the single doseregimen and up to 1 week with both curcumin for-mulae in the repeated dose regimen) and NOS (upto 24 hours with pure curcumin and up to 48 hourswith curcumin conjugate in the single doseregimen, up to 48 hours with pure curcumin and up

to 1 week with curcumin conjugate in the repeateddose regimen) (Figures 3–5, 10, 11), a significantincrease in genes expression of eNOS, nNOS,HO-1, and Nrf2 (Figures 6–8, 13) and a significantdecrease in iNOS, NF-Kb, and p38 genes expres-sion (Figures 9, 12, 14) in comparison withuntreated induced diabetic rat group but not at alltime intervals, for example; p38 gene expression wassignificantly repressed up to 48 hours in the singledose regimen and for up to 1 week in the repeateddose regimen with both curcumin formulae.

Figure 5 Nitric oxide synthase (NOS)enzyme activity (pmol/mg protein/min)in rat cavernous tissue. *Significantdifference in comparison with controlrats, #significant difference in compari-son with diabetic rats.

1822 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

Water-soluble protein conjugate showed sig-nificant superiority to the pure curcumin regard-ing most of the studied parameters with moreprolonged duration of action up to 1 week aftercurcumin conjugate administration. Repeateddoses regimens were superior to single doseregimen regarding most of the studied param-eters. For example, the effect on iNOS andnNOS genes expression is prominent up to 48hours in the single dose regimen and up to 1week in the repeated dose regimen.

Administration of ZnPP to control rats signifi-cantly reduced the HO-1 gene expression and HOenzyme activity in comparison with untreated

control rats. Moreover, ZnPP significantlyreduced HO enzyme activity, cGMP, ICP/MAP,HO-1 relative gene expression in STZ-induceddiabetic rat groups that received curcumin conju-gates either single or repeated doses in comparisonwith their corresponding STZ-induced diabeticrat groups that did not receive ZnPP. ZnPP has noeffect on the other parameters.

Discussion

In the present study, comparative assessment of themolecular and physiological effects of a water-soluble curcumin protein conjugate vs. pure cur-

Figure 6 Real-time PCR of hemeoxygenase-1 (HO-1) gene expressedin Ct values relative to housekeepinggene. *Significant difference in com-parison with control rats, #significantdifference in comparison with diabeticrats.

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1823

J Sex Med 2012;9:1815–1833

cumin in an experimental diabetic model of ED wasconducted. Results revealed that in diabetic rats,there were significant decreases in cavernous tissuecGMP, NOS, and HO enzymatic activities. Therewas a significant decrease in intracavernous pres-sure as an index of erectile function in STZ-induced diabetic rats as compared with control rats.These results coincided with previous findingsreported by Abdel Aziz et al. [40] who reported thatthe decline in erectile function in diabetes could beattributed to the decrease in HO-1 gene expression,HO enzyme activity with subsequent decline in

cGMP cavernous tissue levels and decline in ICP.Diabetes-associated decline in erectile function isattributed mainly to endothelial dysfunction.Recent studies on conditions associated withendothelial dysfunction had demonstrated thatendothelial dysfunction is a major factor contribut-ing to vasculogenic ED cases [41,42].

Factors contributing to diabetic endothelialdysfunction include the effects of oxidative stressand decreased NO bioavailability. Yang et al. [5]stated that vascular arginase activity is increased indiabetic rats and that both arginase and eNOS

Figure 7 Real-time PCR of endothe-lial nitric oxide synthase (eNOS) geneexpressed in Ct values relative tohousekeeping gene. *Significant differ-ence in comparison with control rats,#significant difference in comparisonwith diabetic rats.

1824 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

compete for their common substrate, L-arginine,with subsequent decrease in NO bioavailability inthe cavernous tissue [10,43]. Moreover, glucose-induced alterations in cellular metabolism thatmay account for endothelial dysfunction involveactivation of PKC which may also be involved inregulation of arginase activity. Furthermore, thegeneration of tumor necrosis factor-a (TNF-a) isincreased during diabetes and this cytokine hasbeen shown to upregulate the expression of argi-nase in endothelial cells [44]. Moreover, TNF canaffect NO production by decreasing eNOS expres-

sion [45] and increase the production of ROSin neutrophils and endothelial cells throughNADPH oxidase [46], xanthine oxidase, anduncoupled NOS [5]. These findings coincide withour results that demonstrated decreased NOSactivity and decreased gene expression of eNOSand nNOS in the cavernous tissue of diabetic ratsas compared with the control group. On the con-trary, iNOS was significantly increased. Wan et al.[47], reported similar findings.

Musicki et al. [48] stated that diabetes-associated ED has been attributed to a reduction in

Figure 8 Real-time PCR of neuronalnitric oxide synthase (nNOS) geneexpressed in Ct values relative tohousekeeping gene. *Significant differ-ence in comparison with control rats,#significant difference in comparisonwith diabetic rats.

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1825

J Sex Med 2012;9:1815–1833

the number of NOS-containing nerves, theimpairment of NOS activity, and both neurogenic-and endothelium-mediated smooth muscle relax-ation [49–51], and also to downregulation of themediators downstream from NO, such as cGMP[52] and cGMP-dependent protein kinase-1 [53],in the corpus cavernosum. Several mechanisms ofimpaired endothelial function in the diabetic penishave been described that contribute to vasculo-genic ED. They include reduced eNOS expres-sion, decreased eNOS activity and impaired eNOS

phosphorylation, increased oxidative stress, andincreased activity of the RhoA/Rho-kinase signal-ing pathway.

Furthermore, Hurdag et al. [54] stated that inerectile function, eNOS and nNOS may have asignificant role whereas an increase in iNOS mayhave a deleterious effect via increase in peroxyni-trite formation with decrease in nitrate bioavail-ability. Reactive oxygen species (ROS) play animportant role in vascular dysfunction in diabetes.Overproduction of superoxide can lead to scaveng-

Figure 9 Real-time PCR of inducibleNOS (iNOS) gene expressed in Ctvalues relative to housekeeping gene.*Significant difference in comparisonwith control rats, #significant differencein comparison with diabetic rats.

1826 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

ing of NO and to its reduced bioavailability [55].ROS have been implicated in increased arginaseactivity and gene expression. Arginase activationcan cause uncoupling of eNOS by reducing thesupply of L-arginine. The uncoupled eNOS usesmolecular oxygen to produce superoxide, therebyfurther reducing NO and increasing ROS forma-tion. These facts explain our results that demon-strated decreased erectile function as indicated bydecreased NOS enzyme activity, eNOS and nNOSgene expression as well as decreased intracavern-ous pressure in STZ-induced diabetic rat group.

Administration of either pure curcumin or itswater-soluble protein conjugate led to significantelevations in cGMP levels, NOS and HO-1 cav-ernous tissue enzyme levels with significantlyhigher efficacy in favor of the water-soluble cur-cumin protein conjugate.

These findings could be explained by improve-ment in diabetes-induced endothelial dysfunction.Rungseesantivanon et al. [12] stated that cur-cumin supplementation could improve diabetes-induced endothelial dysfunction via decrease invascular superoxide production and PKC inhibi-

tion. Moreover, a clinical study by Usharani et al.[56] showed that curcumin administration signifi-cantly reduced the levels of malondialdehyde,ET-1, interleukin-6, and TNF-a in type 2 diabe-tes patients. Therefore, the molecular mecha-nisms of curcumin-mediated increases in vascularNO bioavailability might be enhanced by its anti-oxidant properties and by its anti-inflammatoryeffects. Moreover, Ahmad et al. [57] and Ryteret al. [58] reported that HO protects NO throughscavenging of ROS, preventing the formation ofperoxynitrite and its subsequent degradation andconfirming that CO tissue levels parallel NOlevels. Ahmad et al. [57] stated that overexpres-sion of HO-1 may mediate an increase in eNOSand a decrease in iNOS, potentially contributingto restoration of vascular responses in diabeticrats. Curcumin as an inducer of HO-1 could indi-rectly potentiate eNOS effects on vascular endot-helium. More recently, Abdel Aziz et al. [16]stated that CO-derived HO-1 could mediateerectile signaling via upregulation of cGMP, thesignaling molecule of erectile function. COrelaxes the vascular smooth muscles and promotes

Figure 10 Intracavernosal pressure/mean arterial pressure in response to 0.3, 0.5, 0.8, 1.0 Hertz. *Significant difference incomparison with control rats, #significant difference in comparison with diabetic rats.

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1827

J Sex Med 2012;9:1815–1833

relaxation linked to activation of sGC [59,60],stimulation of calcium-activated potassium chan-nels that is involved in the regulation of mem-brane potential and tone in small myogenicallyactive arterioles [61], or inhibition of endothelinrelease from endothelial cells [62]. CO of vascularorigin is thought to serve a vasodilatory functionbased on reports that metalloporphyrins thatinhibit HO bring about vasoconstriction orincreased vascular reactivity to constrictor ago-nists [63,64]. Ryter et al. [58] stated that CO likeNO acts as a vasorelaxant and may regulate othervascular functions such as platelet aggregationand smooth muscle proliferation. The inhibitionof NOS by L-Nitro-arginine methyl ester(L-NAME) had no effect on HO enzyme activityand cGMP. This finding indicates that theHO/CO system might supervene, complement,or substitute the NOS/NO system in cavernoustissue. The role of CO as a NO-like signalingmolecule has been supported from studies on HOand NOS knockouts. HO-1-derived CO has apositive effect on both sGC and cGMP levels invascular endothelial cells [1,63,65].

Administration of ZnPP as HO inhibitor tocontrol rats significantly reduced the HO-1 geneexpression and HO enzyme activity in compari-son with untreated control rats. Moreover, ZnPPsignificantly reduced HO enzyme activity, cGMP,ICP/MAP, HO-1 relative gene expression indiabetic rat groups that received curcumin conju-gates either single or repeated doses in compari-son with their corresponding diabetic rat groupsthat did not receive ZnPP. These findings supportour hypothesis that HO/CO system might super-vene, complement, or substitute the NOS/NOsystem.

In the present study, gene expression profile ofNF-Kb and p38 were significantly increased inSTZ-induced diabetic rat. Administration ofeither pure curcumin or water-soluble protein cur-cumin conjugate led to a significant lowering effecton their gene expression with more significanteffects with water-soluble curcumin protein con-jugate than with pure curcumin treated animalsand more significant effect with repeated dosesthan single dose regimens. Moreover, Nrf2 geneexpression was unchanged in STZ-induced dia-

Figure 11 Intracavernosal pressure/mean arterial pressure in response to 2, 3, 4, 5 Hertz. *Significant difference incomparison with control rats, #significant difference in comparison with diabetic rats.

1828 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

betic rats, whereas its levels were significantlyelevated with pure curcumin and its water-solubleconjugate. Aggarwal and Sung [66] stated thatcurcumin suppresses the proinflammatory tran-scription factors nuclear factor-kappa B and signaltransducer and activators of transcription-3, andit activates peroxisome proliferator-activatedreceptor-gamma and Nrf2 cell-signaling path-ways, thus leading to downregulation of adipok-ines in several tissues and upregulation ofadiponectin and other gene products.

Several studies showed the upstream signalingnetworks involved in HO-1 induction by cur-

cumin involves Nrf2, AP-1, p38, and NF-Kb[22–26]. Moreover, the pharmacokinetics of cur-cumin have recently been found to be mediated viadownstream signaling networks such as theNF-Kb and MAPK/ERK pathways [39].

In conclusion, the water-soluble curcuminprotein conjugate could enhance erectile functionin a diabetic model via upregulation of cavernoustissue levels of HO-1 gene and cGMP. Curcuminprotein conjugate is superior to pure curcumin withmore prolonged duration of action. Molecular sig-naling of curcumin protein conjugate involveupregulation of gene expression of eNOS, nNOS,

Figure 12 Real-time PCR of NF-Kbgene expressed in Ct values relative tohousekeeping gene. *Significant differ-ence in comparison with control rats,#significant difference in comparisonwith diabetic rats.

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1829

J Sex Med 2012;9:1815–1833

HO-1, and Nrf2 genes and downregulation of geneexpression of NF-Kb, p38, and iNOS genes.

Acknowledgment

This study was supported in part by a scientific gift inthe form of the novel water-soluble curcumin deriva-tive which is registered as international patent pro-tected by the rights of “The Patent CooperationTreaty” under: (PCT/EG2010/000008, PublishedPatent Pending, WO 2011/100984) and is the personalproperty of its inventors: Rezq El-Sayed Ameen,Mohammed Talaat Abdel Aziz, and Al-Malki Abdul-rahman. The study was technically and financially

supported in part by UBMB, Faculty of Medicine,Cairo University. The patent cooperation treaty isfound at the following website: http://www.wipo.int/patentscope/search/en/detail.jsf;jsessionid+90C05C7570FE02B89EBE9013FC25D794.wapp1?docId=WO2011100984&recNum=72&office=&queryString=&prevFilter=&sortOption=Pub+Date+Desc&maxRec=8021524.

Corresponding Author: Hanan H. Fouad, MD,Medical Biochemistry Department, Unit of Biochemis-try and Molecular Biology, Faculty of Medicine, CairoUniversity, POB 11562, Cairo 11562, Egypt. Tel:(202) 23632297; Fax: (202) 23632297; E-mail:hananfouadbostamy24@gmail.com

Figure 13 Real-time PCR of nucleartranscription factor-erythroid2 (Nrf2)gene expressed in Ct values relative tohousekeeping gene. *Significant differ-ence in comparison with control rats,#significant difference in comparisonwith diabetic rats.

1830 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

Conflict of Interest: None.

Statement of Authorship

Category 1(a) Conception and Design

Mohamed Talaat Abdel Aziz; Tarek Motawi; AmeenRezq; Taymour Mostafa; Hanan H. Fouad

(b) Acquisition of DataHanan H. Ahmed; Laila Rashed; Dina Sabry; AmiraSenbel; Raghda El-Shafiey

(c) Analysis and Interpretation of DataMohamed Talaat Abdel Aziz; Tarek Motawi; AmeenRezq; Taymour Mostafa; Hanan H. Fouad; AAl-Malki

Category 2(a) Drafting the Article

Hanan H. Fouad; Hanan H. Ahmed; RaghdaEl-Shafiey

(b) Revising It for Intellectual ContentMohamed Talaat Abdel Aziz; Tarek Motawi; AmeenRezq; Taymour Mostafa; A Al-Malki

Figure 14 Real-time PCR of p38gene expressed in Ct values relative tohousekeeping gene. *Significant differ-ence in comparison with control rats,#significant difference in comparisonwith diabetic rats.

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1831

J Sex Med 2012;9:1815–1833

Category 3(a) Final Approval of the Completed Article

Mohamed Talaat Abdel Aziz; Tarek Motawi; AmeenRezq; Taymour Mostafa; Hanan H. Fouad; HananH. Ahmed; Laila Rashed; Dina Sabry; AmiraSenbel; A Al-Malki; Raghda El-Shafiey

References

1 Abdel Aziz MT, El-Asmer MF, Mostafa T, Mostafa S, Atta H,Wassef MA, Fouad H, Rashed L, Sabry D, Mahfouz S. Hemeoxygenase versus nitric oxide synthase in signaling mediatingsildenafil citrate action. J Sex Med 2007;4:1098–107.

2 Abdel Aziz MT, El Asmer MF, Rezq A, Kumosani TA,Mostafa S, Mostafa T, Atta H, Wassef MA, Fouad HH, RashedL, Sabry D, Hassouna AA, Senbel A, Abdel Aziz A. Novelwater soluble curcumin derivative mediates erectile signaling. JSex Med 2010;7:2714–22.

3 Abdel Aziz MT, El-Haggar S, Mostafa T, Atta H, Fouad H,Mahfouz S, Rashed L, Sabry D, Senbel A, Ali GA. Effect ofmesenchymal stem cell penile transplantation on erectile sig-nalling of aged rats. Andrologia 2010;42:187–92.

4 Strong TD, Gebska MA, Burnett AL, Champion HC, Bivalac-qua TJ. Endothelium-specific gene and stem cell-based therapyfor erectile dysfunction. Asian J Androl 2008;10:14–22.

5 Yang G, Lucas R, Caldwell R, Yao L, Romero MJ, CaldwellRW. Novel mechanisms of endothelial dysfunction in diabetes.J Cardiovasc Dis Res 2010;1:59–63.

6 Xu J, Zou MH. Molecular insights and therapeutic targets fordiabetic endothelial dysfunction. Circulation 2009;120:1266–86.

7 Craven PA, Studer RK, Negrete H, DeRubertis FR. Proteinkinase C in diabetic nephropathy. J Diabetes Complications1995;4:241–5.

8 Holinstat M, Mehta D, Kozasa T, Minshall RD, Malik AB.Protein kinase Calpha-induced p115RhoGEF phosphoryla-tion signals endothelial cytoskeletal rearrangement. J BiolChem 2003;31:28793–8.

9 Chen Z, Singer WD, Sternweis PC, Sprang SR. Structure ofthe p115RhoGEF rgRGS domain-Galpha13/i1 chimeracomplex suggests convergent evolution of a GTPase activator.Nat Struct Mol Biol 2005;2:191–7.

10 Romero MJ, Platt DH, Tawfik HE, Labazi M, El-Remessy AB,Bartoli M, Caldwell RB, Caldwell RW. Diabetes-inducedcoronary vascular dysfunction involves increased arginaseactivity. Circ Res 2008;102:95–102.

11 Zhang C. The role of inflammatory cytokines in endothelialdysfunction. Basic Res Cardiol 2008;103:398–406.

12 Rungseesantivanon S, Thenchaisri N, Ruangvejvorachai P,Patumraj S. Curcumin supplementation could improvediabetes-induced endothelial dysfunction associated withdecreased vascular superoxide production and PKC inhibition.BMC Complement Altern Med 2010;10:57–65.

13 Balasubramanyam M, Adaikala Koteswari A, Sampath KumarR, Finny Monickaraj S, Uma Maheswari J, Mohan V.Curcumin-induced inhibition of cellular reactive oxygenspecies generation: Novel therapeutic implications. J Biosci2003;28:715–21.

14 Okamoto T, Yamagishi S, Inagaki Y, Amano S, Koga K, Abe R,Takeuchi M, Ohno S, Yoshimura A, Makita Z. Angiogenesisinduced by advanced glycation end products and its preventionby cerivastatin. FASEB J 2002;14:1928–30.

15 Abdel Aziz MT, Mostafa T, Atta H, Rashed L, Marzouk SA,Obaia EM, Sabry D, Hassouna AA, El-Shehaby AM, Aziz AT.The Role of PDE5 inhibitors in heme oxygenase-cGMP rela-tionship in rat cavernous tissues. J Sex Med 2008;5:1636–45.

16 Abdel Aziz MT, Mostafa T, Atta H, Wassef MA, Fouad HH,Rashed LA, Sabry D. Putative role of carbon monoxide signal-ing pathway in penile erectile function. J Sex Med 2009;6:49–60.

17 Abdel-Aziz MT, Al-Asmar MF, Atta H, Shaker O, Fouad H,Hosni H, Kamal N, Rashed L. Effect of inducers of hemeoxygenase-1 on eicosanoids, cAMP, cGMP levels in rat endot-helial cells. Biochim Clin 2002;26:352–7.

18 Abdel-Aziz MT, El-Asmar MF, Mostafa T, Atta H, Abdel AzizM, Fouad H, Kamal N, Rashed L, Sabry D. Effects of nitricoxide synthase and heme oxygenase inducers and inhibitors onmolecular signaling of erectile function. J Clin Biochem Nutr2005;37:103–11.

19 Abdel Aziz MT, Mostafa T, Atta H, Rashed L, Marzouk SA,Obaia EM, Sabry D, Hassouna A, El-Shehaby A, Abdel AzizAT. Oral phosphodiesterase-5 inhibitors: Effect of heme oxy-genase inhibition on cGMP signaling in rat cavernous tissue.Andrologia 2007;39:66–70.

20 Abdel Aziz MT, Abraham N, Mostafa T, Mahfouz S, Atta H,Wassef MA, Fouad H, Rashed L, Sabry S, AbdEl-Azim O.Heme oxygenase-1 (HO-1) gene transfer and CORM-3 astherapeutic modalities in erectile dysfunction. Andrologia2008;5:363–43.

21 Yang L, Quan S, Abraham NG. Retrovirus mediated HO genetransfer into endothelial cells protects against oxidant inducedinjury. Am J Physiol 1999;25:L127–33.

22 Andreadi CK, Howells LM, Atherfold PA, Manson MM.Involvement of Nrf2, p38, B-Raf, and nuclear factor-kappaB,but not phosphatidylinositol 3-kinase, in induction of hemeoxygenase-1 by dietary polyphenols. Mol Pharmacol 2006;69:1033–40.

23 Kang KA, Lee KH, Park JW, Lee NH, Na HK, Surh YJ, YouHJ, Chung MH, Hyun JW. Triphlorethol-A induces hemeoxygenase-1 via activation of ERK and NF-E2 related factor 2transcription factor. FEBS Lett 2007;58:2000–8.

24 Jeong GS, Pae HO, Jeong SO, Kim YC, Kwon TO, Lee HS,Kim NS, Park SD, Chung HT. The alpha-methylene-gamma-butyrolactone moiety in dehydro costus lactone is responsiblefor cytoprotective heme oxygenase-1 expression through acti-vation of the nuclear factor E2-related factor 2 in HepG2 cells.Eur J Pharmacol 2007;565:37–44.

25 Hill-Kapturczak N, Thamilselvan V, Liu F, Nick HS, AgarwalA. Mechanism of heme oxygenase-1 gene induction by cur-cumin in human renal proximal tubule cells. Am J PhysiolRenal Physiol 2001;281:F851–9.

26 Balogun E, Hoque M, Gong P, Killeen E, Green CJ, ForestiR, Alam J, Motterlini R. Curcumin activates the haemoxygenase-1 gene via regulation of Nrf2 and the antioxidant-responsive element. Biochem J 2003;371:887–95.

27 Durante W, Johnson FK, Johnson RA. Role of carbon monox-ide in cardiovascular function. J Cell Mol Med 2006;10:672–86.

28 Xu PH, Long Y, Dai F, Liu ZL. The relaxant effect of cur-cumin on porcine coronary arterial ring segments. VasculPharmacol 2007;47:25–30.

29 Schini-Kerth VB, Auger C, Étienne-Selloum N, ChataigneauT. Polyphenol-induced endothelium-dependent relaxations.Role of NO and EDHF. Adv Pharmacol 2010;60:136–75.

30 Anand P, Kunnumakkara AB, Newman RA, Aggarwal BB.Bioavailability of curcumin: Problems and promises. MolPharm 2007;4:807–18.

31 Li WJ, Zhou J, Li B, Wang H, Peng YB, Wang Z. Correlationbetween nNOS expression and erectile dysfunction in experi-mental diabetes mellitus rats of different ages. NeuroembryolAging 2008;5:144–50.

32 Li WJ, Zhou J, Li B, Wang H, Peng YB, Wang Z. PARPinhibition restores erectile function by suppressing corporalsmooth muscle apoptosis in diabetic rats. J Sex Med2011;8:1072–82.

1832 Abdel Aziz et al.

J Sex Med 2012;9:1815–1833

33 Moshag H, Kok B, Huizenga JR, Jansen LK. Nitrite andnitrate determination in plasma: A critical evaluation. ClinChem 1995;41:892–96.

34 Xia ZW, Zhou WP, Cui WJ, Zhang XH, Shen QX, Li YZ, YuSC. Structure prediction and activity analysis of human hemeoxygenase-1 and its mutant. World J Gastroenterol 2004;10:2352–6.

35 Giuliano F, Bernabe J, Jardin A, Rousseau JP. Antierectile roleof the sympathetic nervous system in rats. J Urol 1993;150:519–24.

36 Ushiyama M, Morita T, Kuramochi T, Yagi S, Katayama S.Erectile dysfunction in hypertensive rats results from impair-ment of the relaxation evoked by neurogenic carbon monoxideand nitric oxide. Hypertens Res 2004;27:253–61.

37 Zvara P, Spiess PE, Merlin SL, Begin LR, Brock GB. Neuro-genic erectile dysfunction: The course of nicotinamide adeninedinucleotide phosphate diaphorase-positive nerve fibers on thesurface of the prostate. Urology 1996;47:146–51.

38 Martinez L, Brock G, Rocha-Trigo F, Hsu G, Lue TF,Tanagho EA. Rat model for the study of penile erection: Phar-macologic and electrical-stimulation parameters. Eur Urol1994;25:62–70.

39 Hedlund P, Ny L, Alm P, Andersson KE. Cholinergic nervesin human corpus cavernosum and spongiosum contain nitricoxide synthase and heme oxygenase. J Urol 2000;164:868–75.

40 Abdel Aziz MT, El Asmer MF, Mostafa T, Atta H, Mahfouz S,Fouad H, Rashed L, Sabry D, Hassouna A, Abdel Aziz AT,Senbel A, Demery A. Effects of losartan, HO-1 inducers orHO-1 inhibitors on erectile signaling in diabetic rats. J SexMed 2009;6:3254–64.

41 Bivalacqua TJ, Usta MF, Champion HC, Kadowitz PJ, Hell-strom WJ. Endothelial dysfunction in erectile dysfunction:Role of the endothelium in erectile physiology and disease. JAndrol 2003;24:S17–37.

42 Hamed S, Brenner B, Roguin A. Nitric oxide: A key factorbehind the dysfunctionality of endothelial progenitor cells indiabetes mellitus type-2. Cardiovasc Res 2011;91:9–15.

43 Romero MJ, Caldwell RB, Caldwell RW. Does elevated argi-nase activity contribute to diabetes-induced endothelial dys-function. FASEB J 2006;20:A1125.

44 Gao X, Xu X, Belmadani S, Park Y, Tang Z, Feldman AM,Chilian WM, Zhang C. TNF-alpha contributes to endothelialdysfunction by upregulating arginase in ischemia/reperfusioninjury. Arterioscler Thromb Vasc Biol 2007;27:1269–75.

45 MacNaul KL, Hutchinson NI. Differential expression ofiNOS and cNOS mRNA in human vascular smooth musclecells and endothelial cells under normal and inflammatoryconditions. Biochem Biophys Res Commun 1993;196:1330–4.

46 Sorescu D, Griendling KK. Reactive oxygen species, mito-chondria, and NAD(P)H oxidases in the development andprogression of heart failure. Congest Heart Fail 2002;8:132–40.

47 Wan ZH, Li WZ, Li YZ, Chen L, Li GH, Hu WF, Peng S, YuJJ, Guo F. Poly(ADP-Ribose) polymerase inhibition improveserectile function in diabetic rats. J Sex Med 2011;8:1002–14.

48 Musicki B, Burnett AL. eNOS function and dysfunction in thepenis. Exp Biol Med 2006;2:154–65.

49 Saenz de Tejada I, Goldstein I, Azadzoi K, Krane RJ, CohenRA. Impaired neurogenic and endothelium-mediated relax-ation of penile smooth muscle from diabetic men with impo-tence. N Engl J Med 1989;320:1025–30.

50 Cellek S, Foxwell NA, Moncada S. Two phases of nitrergicneuropathy in streptozotocin-induced diabetic rats. Diabetes2003;52:2353–62.

51 Podlasek CA, Zelner DJ, Bervig TR, Gonzalez CM, McKennaKE, McVary KT. Characterization and localization of nitricoxide synthase isoforms in the BB/WOR diabetic rat. J Urol2001;166:746–55.

52 Bivalacqua TJ, Usta MF, Champion HC, Leungwattanakij S,Dabisch PA, McNamara DB, Kadowitz PJ, Hellstrom WJ.Effect of combination endothelial nitric oxide synthase genetherapy and sildenafil on erectile function in diabetic rats. Int JImpot Res 2004;16:21–9.

53 Chang S, Hypolite JA, Velez M, Changolkar A, Wein AJ,Chacko S, DiSanto ME. Downregulation of cGMP-dependentprotein kinase-1 activity in the corpus cavernosum smoothmuscle of diabetic rabbits. Am J Physiol Regul Integr CompPhysiol 2004;287:R950–60.

54 Hurdag C, Ozkara H, Citci S, Uyaner I, Demirci C. Theeffects of alpha-lipoic acid on nitric oxide synthetase dispersionin penile function in streptozotocin-induced diabetic rats. Int JTissue React 2005;27:145–50.

55 Tsubouchi H, Inoguchi T, Sonta T, Sato N, Sekiguchi N,Kobayashi K, Sumimoto H, Utsumi H, Nawata H. Statinattenuates high glucose-induced and diabetes-induced oxida-tive stress in vitro and in vivo evaluated by electron spin reso-nance measurement. Free Radic Biol Med 2005;39:444–52.

56 Usharani P, Mateen A, Naidu M, Raju Y, Chandra N. Effect ofNCB-02, Atorvastatin and placebo on endothelial function,oxidative stress and inflammatory markers in patients with T2diabetes mellitus: A randomized, parallel-group, placebo-controlled, 8-week study. Drugs R D 2008;9:243–50.

57 Ahmad M, Turkseven S, Mingone CJ, Gupte SA, Wolin MS,Abraham NG. Heme oxygenase-1 gene expression increasesvascular relaxation and decreases inducible nitric oxide syn-thase in diabetic rats. Cell Mol Biol 2005;51:371–6.

58 Ryter SW, Otterbein LE, Morse D, Choi AM. Hemeoxygenase/carbon monoxide signaling pathways: Regulationand functional significance. Mol Cell Biochem 2002;234–235:249–63.

59 Sammut IA, Foresti R, Clark JE, Exon DJ, Vesely MJ, Sarath-chandra P, Green CJ, Motterlini R. Carbon monoxide is amajor contributor to the regulation of vascular tone in aortasexpressing high levels of haeme oxygenase-1. Br J Pharmacol1998;125:1437–44.

60 Shamloul R. The potential role of the heme oxygenase/carbonmonoxide system in male sexual dysfunctions. J Sex Med2009;6:324–33.

61 Kaide J, Zhang F, Wei Y, Jiang H, Yu C, Wang W, Balazy M,Abraham NG, Nasjletti A. Carbon monoxide of vascular originattenuates the sensitivity of renal arterial vessels to vasocon-strictors. J Clin Invest 2001;107:1163–71.

62 Wang R, Wu L, Wang Z. The direct effect of carbon monoxideon KCa channels in vascular smooth muscle cells. PflugersArch 1997;434:285–91.

63 Morita T, Kourembanas S. Endothelial cell expression of vaso-constrictors and growth factors is regulated by smooth musclecell-derived carbon monoxide. J Clin Invest 1995;96:2676–82.

64 Naito Y. Introduction to serial review: Heme oxygenase andcarbon monoxide: Medicinal chemistry and biological effects. JClin Biochem Nutr 2008;42:76–7.

65 Jin YC, Gam SC, Jung JH, Hyun JS, Chang KC, Hyun JS.Expression and activity of heme oxygenase-1 in artificiallyinduced low-flow priapism in rat penile tissues. J Sex Med2008;5:1876–82.

66 Aggarwal BB, Sung B. Pharmacological basis for the role ofcurcumin in chronic diseases: An age-old spice with moderntargets. Trends Pharmacol Sci 2009;30:85–94.

Curcumin Use in Diabetes-Induced Erectile Dysfunction 1833

J Sex Med 2012;9:1815–1833