POTENTIAL CLINICAL RELEVANCE

Nanomedicine: Nanotechnology, Biology, and Medicine9 (2013) 776–785

Research Article

SNEDDS curcumin formulation leads to enhanced protection frompain and functional deficits associated with diabetic neuropathy:

An insight into its mechanism for neuroprotectionRayanta P. Joshi, MS, Pharma, Geeta Negi, MS, Pharma, Ashutosh Kumar, PhDa,

Yogesh B. Pawar, MPharmb, Bhushan Munjal, MPharmb,Arvind K. Bansal, PhDb, Shyam S. Sharma, PhDa,⁎

aMolecular Neuropharmacology Laboratory, Department of Pharmacology and Toxicology,National Institute of Pharmaceutical Education and Research (NIPER), Punjab, India

bDepartment of Pharmaceutics, National Institute of Pharmaceutical Education and Research (NIPER), Punjab, India

Received 26 March 2012; accepted 8 January 2013

nanomedjournal.com

Abstract

Curcumin has shown to be effective against various diabetes related complications. However major limitation with curcumin is its lowbioavailability. In this study we formulated and characterized self nano emulsifying drug delivery system (SNEDDS) curcumin formulationto enhance its bioavailability and then evaluated its efficacy in experimental diabetic neuropathy. Bioavailability studies were performed inmale Sprague Dawley rats. Further to evaluate the efficacy of formulation in diabetic neuropathy various parameters like nerve function andsensorimotor perception were assessed along with study of inflammatory proteins (NF-κB, IKK-β, COX-2, iNOS, TNF-α and IL-6).Nanotechnology based formulation resulted in prolonged plasma exposure and bioavailability. SNEDDS curcumin provided better resultsagainst functional, behavioural and biochemical deficits in experimental diabetic neuropathy, when compared with naive curcumin. Furtherwestern blot analysis confirmed the greater neuroprotective action of SNEDDS curcumin. SNEDDS curcumin formulation due to higherbioavailability was found to afford enhanced protection in diabetic neuropathy.

From the Clinical Editor: In this study the authors formulated and characterized a self-emulsifying drug delivery system for formulation toenhance curcumin bioavailability in experimental diabetic neuropathy. Enhanced efficacy was demonstrated in a rat model.© 2013 Elsevier Inc. All rights reserved.

Key words: Diabetic neuropathy; Curcumin; SNEDDS formulation; NF-κB

Traditional medicine has mentioned the beneficial effects ofcurcumin owing to its anti-inflammatory and antioxidant action.These effects are mediated in part through the downregulation ofvarious transcription factors such as nuclear factor-κB (NF-κB)and activator protein-1 (AP-1). This leads to reduced expressionof genes regulated by these transcription factors such ascycloxygenase-2 (COX-2), inducible nitric oxide synthase(iNOS) and proinflammatory cytokines.1 Despite being a troveof biological activities the major problem with curcumin is its

Conflict of Interest: All the authors have no competing interests.⁎Corresponding author: Molecular Neuropharmacology Laboratory,

Department of Pharmacology and Toxicology, National Institute ofPharmaceutical Education and Research (NIPER), Sec-67, S.A.S.Nagar,Mohali, Punjab-160062, India.

E-mail address: sssharma@niper.ac.in (S.S. Sharma).

1549-9634/$ – see front matter © 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.nano.2013.01.001

Please cite this article as: Joshi RP, et al, SNEDDS curcumin formulation leaddiabetic neuropathy: An insight into its mechanism for neuroprotection. Nanom

poor bioavailability. In clinical trial quantifiable serum levelswere achieved only above the dose of 3.6 g of curcumin. Poororal bioavailability of curcumin is due to factors such as pooraqueous solubility, degradation in gastrointestinal tract at neutraland alkaline pH and high pre-systemic metabolism in theintestinal wall.2 Membrane permeability of curcumin usingCaco-2 cell model was found to be poor.3 Previously, we havecompared various formulation strategies like micronization,nanonization, amorphous solid dispersion, combination withpiperine, complexation with hydroxy propyl-β cyclodextrincomplex (HP-β-CD) and spray dried curcumin-milk compositefor improvement of oral bioavailability of curcumin.4

Self-emulsifying drug delivery system (SEDDS) has recentlyemerged as an approach to improve the solubility, dissolutionand oral absorption for poorly water-soluble drugs.5,6 Self nano-emulsifying drug delivery system (SNEDDS) is a type of

s to enhanced protection from pain and functional deficits associated withedicine: NBM 2013;9:776-785, http://dx.doi.org/10.1016/j.nano.2013.01.001

777R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

SEDDS containing isotropic mixture of oil, surfactant, co-surfactant and drug substance, which can form a nanoemulsionin gastrointestinal tract after oral administration. The resultantemulsion with a particle size less than 100 nm increases thesolubility of hydrophobic drug and enhances its absorption.7

Thus, considering its potential benefits, we have developed anovel SNEDDS of curcumin (PCT application) and evaluated itspermeation in Caco-2 cell model, wherein compared to naïvecurcumin, SNEDDS improved permeation by 6.35 times(unpublished data). Although oral bioavailability of curcuminhas been improved by several nanotechnology based formula-tions including phospholipid complexes, magnetic nanoparticles,polymeric nanoparticles, nanosuspensions and nanocrystals,8 SNEDDS offers unique advantage as it can enhancedrug absorption by a number of ancillary mechanisms,namely; (i) drug solubilization, (ii) inhibition of P-glycopro-tein-mediated drug efflux and preabsorptive metabolism by gutmembrane-bound cytochrome enzymes, (iii) promotion oflymphatic transport, that delivers drug directly to the systemiccirculation while avoiding hepatic first-pass metabolism, and(iv) increasing gastrointestinal membrane permeability. Addi-tional benefits offered by SNEDDS are dosing flexibility andease of manufacturing.9

Diabetic neuropathy (DN) is one of the common complica-tions of diabetes, affecting more than 60% of diabetes patientsworldwide. Inflammation is one of the major reasons behindvarious deficits seen in DN. NF-κB, a key transcription factorregulating the expression of various inflammatory proteins, hasalso been identified to be associated with the sensorimotoralterations and functional deficits in DN.10 Curcumin has beenshown to attenuate thermal hyperalgesia in mouse model ofdiabetic neuropathic pain mediated via inhibition of tissuenecrosis factor- α (TNF-α) and nitric oxide (NO) release.11,12

However in all the prior studies naïve curcumin was used whichhas very low bioavailability. In addition, these studies did notelaborate the multiple targets of curcumin through which itaffords beneficial effects in DN. Thus in this study we reportthe physical characterization of SNEDDS, its comparative oralbioavailability study with naïve curcumin and molecularmechanisms responsible for the neuroprotective effects ofcurcumin in DN.

Methods

Drugs and chemicals

Curcumin was purchased from Himedia Laboratories, India.Streptozotocin (STZ) was procured from Sigma-Aldrich, U.S.A.Glucose oxidase-peroxidase (GOD-POD) kit was purchasedfrom Accurex, India. Acetonitrile and tetrahydrofuran werepurchased from J.T. Baker, USA. Polyethylene glycol (PEG400) and citric acid were purchased from Merck Limited,Mumbai, India. Gelucire® 44/14 and Labrasol were gift sampleby Gattafosse Pvt. Ltd, Saint Priest, France. Vit. E TPGS waspurchased from Eurochem Asia Pvt. Ltd, Shanghai, China.Hydroxy propyl methyl cellulose E5 (HPMC E5) was giftsample by Colorcon Asia Pvt. Ltd, Verna, India. Rhodamine 6 G(R6G) was purchased from Acros Labs, Mumbai, India.

SNEDDS formulation of curcumin

Qualitative formulaSNEDDS consisted of Gelucire 44/14, Labrasol, Vit. E

TPGS, PEG 400, citric acid anhydrous, ethanol and HPMC E5.Each ml of SNEDDS contained 66.7 mg of curcumin.

Preparation of SNEDDSRequired quantities of Gelucire 44/14, Vit. E TPGS,

Labrasol and PEG 400 were added in a vial, heated at 60°Cand uniformly mixed. Citric acid predissolved in ethanol wasadded to this mixture. Curcumin was dissolved in this mixtureby vortexing for 30 min and finally HPMC E5 was dispersed inthis mixture.

Characterization of SNEDDS curcumin formulation

Globule size analysisSNEDDSwas diluted 250 times (40 μl of formulation diluted

to 10 ml) in distilled water that was previously filtered through0.2μMfilter. The globule size and size distribution of the dilutedformulation (n=3) were determined by dynamic light scattering(DLS) (Nano ZS, Malvern, UK) using Cumulants analysismethod, taking the average of five measurements. Further, massaverage globule size was calculated as per the method describedby Zhu et al.13

Transmission electron microscopy (TEM)SNEDDS was diluted 250 times with distilled water

(previously filtered through 0.45 μM filter) and mixed by slightshaking. One drop of diluted sample was deposited on a carbon-coated copper grid and morphology of globules was observed atroom temperature using a transmission electron microscope(Hitachi H-7500, Japan). No dye was used to stain the samples.Samples were analyzed at a lower accelerating voltage of 80 kVwith smaller objective aperture to achieve the contrast. Imageswere captured at 15000× magnification.

Bioavailability studies

Animals and dosingAll animal experiments were performed in accordance with

Committee for the Purpose of Control and Supervision onExperimentation on Animals (CPCSEA) guidelines. Experi-mental protocols were approved by the Institute Animal EthicsCommittee (IAEC) of NIPER S.A.S. Nagar, Punjab, India.Male Sprague-Dawley rats (250–270 g) were kept on fastingwith free access to water, for 12 h before experimentation.Naïve curcumin in the form of aqueous suspension (D90=63 μM) and SNEDDS were administered to rats (p.o.) at a doseof 250 mg/kg. Blood samples were collected after 0.5, 1, 1.5, 2,3, 5, 8, 12 and 24 h and plasma was separated by centrifugationat 10,000 RPM for 5 min at 4°C and stored at −20°C untilprocessed and analyzed.

Bioanalytical method for curcumin quantificationCurcumin in plasma samples was quantified by a validated

HPLC-RF method with a range of 2–600 ng/ml. The chromato-graphic system consisted of LC-10AT VP pump fitted with aDegasser unit DGU-20A5, a SIL-20 AC auto sampler withrefrigeration unit, RF-10AXL fluorescence detector (version

778 R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

3.20, Kyoto, Japan) and a CTO-10A VP Column oven(Shimadzu, Japan). The system was controlled by a CBM-20Asystem controller with Class VP data acquisition software(version 6.12 SP1). The analytical column was a LiChrospher®C18 column (particle size: 5 μm, 200 mm×4.6 mm, Merck).The mobile phase consisted of a mixture of 1% w/v citric acidmonohydrate (pH 3.0), acetonitrile and tetrahydrofuran(48:32:20) at a flow rate of 1 ml/min. Fluorescence detectorwas set at λex/λem of 420 nm/530 nm. The column temperatureand injection volume were 30 °C and 40 μl respectively.Curcumin was quantified by ratio of the peak area of curcuminto that of the R6G (Internal standard) using weighted (1/x2)linear regression.

Induction of diabetes and experimental design

Male Sprague–Dawley rats (250–270 g) were maintained ata constant temperature (23±1 °C) on a 12 h light/dark cycle withfree access to food and water. Diabetes was induced by a singleintraperitoneal injection of 55 mg/kg streptozotocin (STZ), in0.1 M sodium-citrate buffer (pH 4.4). Age-matched control ratsreceived an equivalent amount of the sodium-citrate buffer.Blood samples were collected from tail vein 48 h after STZadministration. The rats with blood glucose more than 250 mg/dlwere considered as diabetic and were further considered forstudy. The experimental groups comprised of non-diabeticcontrol rats, diabetic control rats and diabetic rats treated with(i) SNEDDS curcumin formulation (30, 100 and 300 mg/kg), (ii)naïve curcumin suspended in 0.5% carboxy methyl cellulose(CMC) (30, 100 and 300 mg/kg) and (iii) placebo of SNEDDScurcumin formulation. The treatment was started 6 weeks afterdiabetes induction and was continued daily for period of2 weeks. The functional, behavioural and biochemical experi-ments were performed 24 h after administration of last dose. Theanimals were euthanized and sciatic nerve was isolated forfurther biochemical estimations and mechanistic studies. Num-ber of rats used for functional and behavioural parameters were6–8 and for biochemical estimations, western blotting andimmunohistochemical studies were 4–6.

Nerve function parameters

Motor nerve conduction velocity (MNCV)MNCV was measured using PowerLab 8sp system as

previously reported.14 Animals were anesthetized by 4%halothane and anesthesia was maintained by 1% halothane in amixture of nitrous oxide and oxygen (70:30) using gaseousanesthesia system (Harvard apparatus, UK). Body temperature ofrats was maintained using homeothermic blanket. Sciatic nervewas stimulated with 3 volt, proximally at sciatic notch anddistally at ankle using bipolar needle (26½ gauge) electrodes.Receiving surface electrodes were placed on the foot muscle.The latencies of the compound muscle action potentials wererecorded via bipolar electrodes from the first interosseous muscleof the hind paw and measured from the stimulus artifact to theonset of the negative M-wave deflection. MNCV was calculatedby dividing the distance between the stimulating and recordingelectrode by difference in distal and proximal latency. MNCVwas expressed in m/s.

Sciatic nerve blood flow (NBF)NBF was measured using Laser Doppler system (Perimed,

Jarfalla, Sweden). Briefly, anesthetized rats were placed onstereotaxic apparatus to locate the uniform position of probe.Sciatic nerve was exposed by giving incision on the left flank andLaser Doppler probe (tip diameter 0.85 mm) was applied just incontact with an area of sciatic trunk free from epi or perineurialblood vessels. The exposed nerve was covered with normalsaline to avoid tissue dehydration and some time was spent tillthe stabilization of recordings (over 10–15 min). Flux measure-ment was obtained from the same part of nerve and for thesame time period (over a 10 min period). The blood flow wasreported in arbitrary perfusion units (PU), calculated as meanover the time.15

Behavioral studies

Thermal hyperalgesiaThermal hyperalgesia of the tail to both hot (45oC) and cold

(10oC) immersion test was studied. Rats were acclimatised threedays prior to the experiment. The tail flick response latency orany signs of struggle was observed as the end point response.The cut-off time was kept as 15 s in both tests. Three consecutivereadings were taken at an interval of 10 min.14

Mechanical hyperalgesiaSensitivity to noxious mechanical stimuli was determined

by quantifying the withdrawal threshold of the hind paw inresponse to mechanical stimulation using a Vonfrey anesthesi-ometer and Randall Sellitto callipers (IITC Life Sciences, USA)as described previously.14

Biochemical parameters

Plasma glucose levelsBlood was collected from tail vein in microcentrifuge

tubes containing heparin. Whole blood was centrifuged andplasma was separated. Plasma glucose levels were estimatedusing GOD–POD kit from Accurex, India as per manufacturer'sinstructions.

Lipid peroxidationFor estimation of lipid peroxidation sciatic nerve was

homogenized in phosphate buffer saline (PBS, pH 7.4).Malondialdehyde (MDA) levels were measured as per methoddescribed earlier.16

TNF-α and IL-6 levelsNerve was homogenized in PBS buffer containing phenyl

methane sulphonyl fluoride (PMSF) and protease inhibitorcocktail. After homogenization it was sonicated and detergentwas added to the homogenate. Homogenate was kept in ice coldwater for 30 min and then sonicated. Homogenate wascentrifuged at 10,000 rpm for 10 min at 4°C and supernatantwas used for estimations. Commercially available ELISA kitsfrom eBiosciences (USA) for assaying TNF-α and IL-6 proteinswere used and levels were expressed as pg/mg of protein.17

Figure 1. (A) Globule size analysis of SNEDDS after dilution. The curverepresents the cumulative globule size distribution. (B) TEM image ofdiluted SNEDDS of curcumin.

779R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

Western blotting

Protein lysates were obtained by homogenizing sciatic nerveswith lysis buffer containing 1% Triton X-100, 150 mM NaCl,1 mM EDTA, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM sodium orthovanadate, 1 μg/ml leu-peptin, 1 μg/ml aprotinin, and 20 mM Tris (pH 7.5). Equalamounts of proteins were separated by SDS–PAGE (10%) andtransferred to a nitrocellulose membrane (Pall Life Sciences, FL,USA). After blockingwith 3%bovine serum albumin,membraneswere incubated with primary rabbit polyclonal IgG for NF-κB(p65 subunit), IKK-β, phospho-IKK-β and iNOS (Cell SignalingTechnology, MA, USA) (1:1000); COX-2 (Santa Cruz Bio-technologies, CA, USA) (1:400) for 12 h at 4°C. After washing,membranes were incubated with a horse radish peroxidase/alkaline phosphatase-conjugated secondary antibody (1:2000),and bound antibody was visualized by enhanced chemilumines-cence or by using a colored reaction with BCIP-NBT. The relativeband densities were quantified by densitometry. Equal loading ofprotein was confirmed by measuring β-actin expression.18

Immunohistochemistry

Sciatic nerves were fixed in 10% buffered formaldehyde,dehydrated through graded concentrations of ethanol, embeddedin paraffin, and sectioned. Sections (5 μm thick) were mounted onslides, cleared, and hydrated. Sections were treated with a bufferedblocking solution (5%goat serum) for 15 min and then co-incubatedwith antibodies for NF-κB p65 subunit (rabbit polyclonal; CellSignaling Technology) (1:400) at 4°C for 12 h. Sections werewashedwithTris–HCl 0.5 M, pH 7.6 and incubatedwith secondaryantibody, at room temperature for 1 h. Thereafter, sections werewashed as before, and then co-incubated with a 3, 3-o-diamino-benzidine solution in the dark; at room temperature for 10 min.Sections were counterstained with hematoxylin and observed underlight microscope (Leica, Solms, Germany). Nerves were collectedfrom 4-6 rats from each group. For analyzing the nerve sections, fivedifferent regions in a section were studied. For eliminating any biasduring scoring/counting, the samples were single blinded.19

Statistical analysis

The various pharmacokinetic parameters were calculated fromthe plasma curcumin concentration-time profiles using theThermo Kinetica Version 5.0 software (Thermo Fischer Scien-tific). Data related to DN studies were expressed as mean±SEM.For comparing the differences between the two groups Student t-test was used. For multiple comparisons, analysis of variance(ANOVA) was used. If the ANOVA showed significantdifference, further post hoc Tukey or Dunnet test was applied.Significance was defined as Pb0.05. All statistical analysis wasperformed using Jandel Sigma Stat 2, statistical software.

Results

Globule size and shape analysis

The globule size of diluted SNEDDSwas in the range of 43.82–4801 nm (mass average globule size=213.7 nm) with a poly-

dispersibility index of 0.389±0.044 (Figure 1, A) (Also refer toSupplemental Figure 1). TEM analysis revealed that the globules ofdiluted SNEDDS were almost spherical in shape with smoothsurface and size below 500 nm (Mean globule size (d50) from TEMimages was 170 nm) (Figure 1, B) (Additional physicochemicalstability data is presented in Supplemental Table 1).

Bioavailability studies

Aqueous suspension of curcumin showed Cmax and AUC(0-∞)of 32.29 ng/ml and 38.07 ng.h/ml respectively. Followingadministration of naïve curcumin, plasma levels declinedbelow the limit of quantitation (LOQ) (2 ng/ml) of bioanalyticalmethod after 2 h, indicating the rapid systemic elimination ofcurcumin. This was also evident through short biological half life(t1/2) of 0.47 h (Table 1; Figure 2).

Compared to aqueous suspension of naïve curcumin, Cmax

and AUC(0-t) improved by 1632.1% (527.01 ng/ml) and7411.1% (2393.03 ng.h/ml) respectively for SNEDDS formula-tion. Half life of SNEDDS formulation (10.1 h) was alsosignificantly higher than naïve curcumin (Table 1; Figure 2). Allanimals appeared normal throughout the duration of the study.

Effect of SNEDDS curcumin formulation in diabetic neuropathy

Body weight and plasma glucose levelsEight weeks of diabetes resulted in significant (Pb0.001)

reduction in body weight and increase in plasma glucose level

Table 1Pharmacokinetic parameters (mean±S.E.M.) after oral administration of naïve curcumin and SNEDDS curcumin formulation.

Formulation Pharmacokinetic parameters

Cmax (ng/ml) Tmax (h) t 1/2 (h) AUC o-∞ (ng.h/ml)

Aqueous suspension of curcumin 32.29±14.93 0.58±0.20 0.47±0.14 38.07±7.79SNEDDS 527.01±86.93 1.50±0.18 10.10±1.71 2393.03±172.61

Figure 2. Mean plasma conc.-time profiles after oral administration of curcumin aqueous suspension (O) and SNEDDS (Δ) at a dose of 250 mg/kg (n=6).

780 R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

(Pb0.001). Two weeks treatment with naïve and SNEDDScurcumin (30, 100 and 300 mg/kg) did not produce anysignificant effect on body weight and plasma glucose level oftreated rats (Table 2).

Nociception

Thermal hyperalgesia. Tail flick latency in eighth week wassignificantly decreased in diabetic rats in both hot and coldimmersion test as compared to normal control rats. Two weektreatment with naïve curcumin (30 mg/kg) and placebo ofSNEDDS formulation did not significantly reversed hyperalge-sia (both hot and cold) in diabetic rats. SNEDDS curcuminformulation at 30 mg/kg did not significantly change tail flicklatency in cold immersion test. Higher doses of both SNEDDScurcumin formulation (100 and 300 mg/kg) and naïve curcumin(100 and 300 mg/kg), however produced a significant improve-ment in tail flick latency in both hot and cold immersion andmore improvement was seen in SNEDDS formulation groupwhen compared to corresponding naïve curcumin groups(Table 2).

Mechanical hyperalgesia. Diabetic rats also developed me-chanical hyperalgesia as measured with the Vonfrey anaesthe-siometer and Randall Sellitto callipers. The paw withdrawalthreshold in Vonfrey and Randall Sellitto test was significantly

(Pb0.001) reduced in diabetic rats when compared with normalcontrol rats. Significant dose dependant increase in pawwithdrawal pressure with both Vonfrey and Randall Sellittotests was observed with SNEDDS curcumin formulation andnaïve curcumin at dose of 30, 100 and 300 mg/kg (Table 2).Placebo of SNEDDS formulation did not show any effect inimproving the mechanical hyperalgesia. A greater improvementwas seen in SNEDDS treated groups when compared tocorresponding naïve curcumin groups.

Lipid peroxidation levelsNerve MDA level was increased significantly (Pb0.001) in

diabetic rats when compared to normal control rats. Two weektreatment showed no significant improvement with naïvecurcumin at dose of 30 mg/kg. SNEDDS curcumin formulationshowed significant correction in MDA levels at all the doses.SNEDDS formulation significantly ameliorated the increase inthe lipid peroxidation when compared with corresponding naïvecurcumin group. Placebo of SNEDDS formulation did not showany significant changes when compared with the diabetic group(Table 2).

Nerve function parametersMNCV and NBF were significantly (Pb0.001) reduced after

the eight weeks of induction of diabetes. Two week treatmentwith naïve curcumin at dose of 100 and 300 mg/kg and SNEDDScurcumin at all doses resulted in a significant improvement in

Figure 3. Effect of 2-week treatment of SNEDDS curcumin formulation and naïve curcumin on (A) Motor nerve conduction velocity (MNCV) and (B) nerveblood flow (NBF) in diabetic rats. Results are mean±S.E.M. ND: non diabetic; STZ-D: diabetic; CUR 30, CUR 100, CUR 300: Naïve curcumin (Curcumin in0.5% CMC) at doses 30,100 and 300 mg/kg; Nano CUR 30, Nano CUR 100, Nano CUR 300: SNEDDS curcumin formulation at doses 30, 100 and 300 mg/kg.###Pb0.001 vs age matched normal control rats. *Pb0.05, **Pb0.01 and ***Pb0.001 versus STZ-D; @Pb0.05, @@Pb0.01 versus corresponding naïvecurcumin groups (n=6).

Table 2Comparison of various behavioural and biochemical parameters.

ND STZ-D Placebo CUR 30 Nano CUR 30 CUR 100 Nano CUR 100 CUR 300 Nano CUR 300

Plasma glucose(mg/dl)

103±2 403±3### 412±5 427±10 406±7 418±6 415±8 418±7 401±4

Nociceptive parametersTail flicklatency (Hot)(s)

13.7±2.7 4.9±1.2### 4.5±1.5 5.8±1.9 6.5±2.1* 7.6±2.5* 10.2±1.6**, @ 8.5±1.4** 9.8±1.8**, @

Tail flicklatency (Cold) (s)

13.8±2.3 4.2±2.6### 4.6±2.0 5.3±2.4 6.2±2.2 7.1±1.3* 9.8±1.5**, @ 8.2±1.9* 10.4±1.7**, @

Mechanical hyperalgesiaVonfreyapparatus (g)

79.5±3.2 34.9±3.7### 38.2±2.8 42.7±4.1* 49.3±2.5*, @ 53.8±3.3* 64.4±2.9**, @ 60.4±3.5** 67.1±3.1**, @

Randall sellittoapparatus (g)

191.9±5.3 99.5±4.5### 104.2±4.9 128.2±3.6* 139.3±4.3*, @ 146.0±3.4** 165.3±4.7**, @ 161.9±5.2** 170.6±3.8**,@

Biochemical parametersMDA levels(μM/mg)

13.4±1.9 38.1±2.7### 35.4±1.6 34.4±2.4 31.3±2.2* 27.8±1.2* 21.7±2.1**,@ 22.6±2.4** 19.5±1.5**, @

Results are expressed as mean±S.E.M. ND: non diabetic; STZ-D: diabetic; Nano CUR 30, Nano CUR 100, Nano CUR 300: SNEDDS curcumin formulation atdoses 30, 100 and 300 mg/kg; CUR 30, CUR 100, CUR 300: Naïve curcumin (Curcumin in 0.5% CMC) at doses 30,100 and 300 mg/kg. ### Pb0.001 vs agematched ND group. *Pb0.05 and **Pb0.01 versus STZ-D. @ Pb0.05 versus corresponding naïve curcumin groups (n=6).

781R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

MNCV as compared to untreated diabetic rats. Placebo ofSNEDDS formulation did not show any significant improvementin MNCV (Figure 3, A). Deficit in NBF was significantlyameliorated with naïve curcumin and SNEDDS curcumin at alldose levels. Placebo of SNEDDS formulation did not show anyeffect. The effect seen with SNEDDS curcumin formulation washigher as compared to naïve curcumin (Figure 3, B).

NF-κB expression and IKK-β phosphorylationWestern blot studies revealed that NF-κB expression in

sciatic nerve of diabetic rats was significantly (Pb0.001)increased when compared with control rats. Curcumin treatmentwas able to reduce the expression of NF-κB. The reduction seenin the SNEDDS curcumin group was more significant thancorresponding naïve curcumin group (Figure 4, A). The

phosphorylation of IKK-β is a critical step in NF-κB activation.The increased phosphorylation in diabetic rats accounted forincrease in activation of NF-κB cascade. Curcumin was able toinhibit the phosphorylation of IKK-β thereby inhibiting theactivation of NF-κB cascade. SNEDDS curcumin formulationwas more effective than corresponding naïve curcumin group ininhibiting the phosphorylation of IKK-β (Figure 4, B).

NF-κB immunohistochemistryImmunohistochemical analysis of sciatic nerve sections

confirmed the western blot studies where NF-κB (p65 subunit)immunoreactive cells was more pronounced in the sciatic nervemicrosections of diabetic rats than in control rats (Pb0.001).Two week treatment with SNEDDS curcumin formulation andnaïve curcumin at dose of 100 mg/kg decreased NF-κB positive

Figure 4. Effect of 2-week treatment of naïve curcumin and SNEDDS curcumin formulation (100 mg/kg) on expression of NF-κB, IKK-β and phospho IKK-βin diabetic rats. Changes in the expression of proteins of (A) NF-κB and (B) IKK-β and phospho IKK-β after curcumin treatment in experimental DN weremeasured by Western blot. Equal loading was confirmed by β-Actin. ND, non diabetic; STZ-D, diabetic; CUR 100 and Nano CUR 100, diabetic group treatedwith naïve curcumin and SNEDDS curcumin formulation (100 mg/kg), respectively. Results are mean±S.E.M. of three independent experiments. ###Pb0.001versus ND; *Pb0.05, **Pb0.01 and ⁎⁎⁎Pb0.001, versus STZ-D; @@Pb0.01 versus naïve curcumin group.

Figure 5. Effect of 2-week treatment of SNEDDS curcumin formulation and naïve curcumin (100 mg/kg) on NF-κB immunohistochemistry in diabetic rats. Thephotographs were taken at 20X. ND, non diabetic; STZ-D, diabetic; CUR 100 and Nano CUR 100, diabetic group treated with naïve curcumin and SNEDDScurcumin formulation (100 mg/kg), respectively. Results are mean±S.E.M. of three independent experiments. ###Pb0.001 versus ND; *Pb0.05 and **Pb0.01versus STZ-D; @Pb0.05 versus naïve curcumin group.

782 R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

cells in nerves of treated rats. Similar to the Western blot results,SNEDDS curcumin formulation reduced the number of NF-κB(p65 subunit) immunopositive cells to a greater extent ascompared to naïve curcumin group (Figure 5).

COX-2 and iNOS expressionNF-κB regulates the gene expression of various pro-

inflammatory enzymes such as COX-2 and iNOS. The proteinlevels of these inflammatory markers were increased in the

Figure 6. Effect of 2-week treatment of SNEDDS curcumin formulation and naïve curcumin (100 mg/kg) on inducible proinflammatory proteins COX-2, iNOS,TNF-α and IL-6 in diabetic rats. Changes in the expression of proteins of (A) COX-2 and (B) iNOS after curcumin treatment in experimental DN weremeasured by Western blot. Equal loading was confirmed by β-Actin. (C) TNF-α and (D) IL-6 levels in sciatic nerve were measured with ELISA. ND,nondiabetic; STZ-D, diabetic; CUR 100 and Nano CUR 100, diabetic group treated with naïve curcumin and SNEDDS curcumin formulation (100 mg/kg),respectively. Results are mean±S.E.M. of three independent experiments. ###Pb0.001 versus ND; *Pb0.05 and **Pb0.01 versus STZ-D; @Pb0.05,@@Pb0.01 versus naïve curcumin group.

783R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

diabetic group (Pb0.001). SNEDDS curcumin treatmentreduced the expression of COX-2 (Figure 6, A) and iNOS(Figure 6, B) more effectively than naïve curcumin, thus ablatingthe inflammatory cascade.

TNF-a and IL-6 levelsInterleukins and TNF-α which are the mediators of

inflammation, also serve as biological markers for inflammation.There was an elevation in TNF-α and IL-6 levels in sciatic nerveof diabetic rats as compared to normal control (Pb0.001). Asignificant reduction in both these parameters was observed withnaïve curcumin and SNEDDS curcumin formulation. SNEDDScurcumin formulation reduced TNF-α level (Figure 6, C)more significantly than naïve curcumin. However, reduction inIL-6 levels (Figure 6, D) was comparable and there was nosignificant difference between SNEDDS curcumin formulationand naïve curcumin.

Discussion

Diabetic neuropathy is associated with various sensorimotor,functional and biochemical alterations. Sensory alteration wasevidenced by thermal and mechanical hyperalgesia. Theseresults are consistent with previous reports.20,21 Curcumintreatment reversed these sensorimotor deficits in experimentalDN in a dose-dependent manner and more significantly withSNEDDS formulation.

The therapeutic efficiency of curcumin is limited by its pooraqueous solubility and low oral bioavailability. Many studieshave been carried out to improve the bioavailability of curcumin.Curcumin-loaded PLGA nanoparticles formulation showedenhanced cellular uptake and increased bioactivity in vitro andbioavailability in vivo in inducing apoptosis of leukemic cells.22

Cyclodextrin-complexed curcumin exhibited improved anti-inflammatory and antiproliferative activities in comparison to

784 R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

naïve curcumin.23 The increase in bioavailability using SNEDDSapproach can be attributed to its fine globule size, enhanced drugdissolution (Supplemental Figure 2) and enhanced lymphaticabsorption. Gelucire 44/14 has been reported to inhibit P-glycoprotein (P-gp) efflux and open the tight junctions in Caco-2cells thereby improving the drug permeation.24 Vitamin E TPGSincreases the absorption flux of drugs by micelle solubilisation,inhibition of the efflux system and reducing the intestinalmetabolism.25 Hence the presence of these “functional excipients”having bio-enhancement properties, in SNEDDS formulationaugmented the oral bioavailability. Various experiments per-formed during our study revealed that SNEDDS curcuminformulation showed better efficacy at various dose levels inimproving various deficits ofDN in comparison to naïve curcumin.

Reduction in MNCV and NBF is an important feature ofneuropathy. Reduced NBF leads to decreased nutritional supportto nerves and failure of ATP sensitive ion exchange pumps.Demyelination, nerve ischemia and endothelial dysfunction seenin diabetes may lead to nerve conduction and blood flow deficitsin experimental DN.26 SNEDDS curcumin formulation signif-icantly improved these functional deficits. The favourable effectsof curcumin on nerve perfusion are consistent with a key role ofoxidative stress in endothelial dysfunction as curcumin also hasantioxidant effects.

Apart from functional and behavioural alterations, there existmany biochemical changes like depletion of antioxidantenzymes, enhanced lipid peroxidation etc. that accompanyDN.27–29 We studied the effect of curcumin on nerve MDAlevels, a measure of lipid peroxidation. Curcumin decreased theexcessive lipid peroxidation seen in diabetic rats which may beexplained by its free radical scavenging activity. Although theantioxidant activity of curcumin is well established30,31 the mainobjective of our study was to test the efficacy of SNEDDScurcumin formulation. It was seen that SNEDDS formulationmore significantly decreased level of nerve MDA as compared tonaïve curcumin.

NF-κB is one of the primary transcription factors initiatinginflammatory response and contributing to exaggeration ofinflammatory damage. Hyperglycaemia induced saturation ofGAPDH leads to shunting of excess glucose through alternativemetabolic pathways such as the fructose-6 phosphate ordiacylglycerol. The signalling intermediates and modifiedtranscription factors produced via these pathways lead toincrease in TGF-β and NF-κB.32 Protective effects seen instudy could be attributed to curcumin's ability to inhibittranscriptionary activity of NF-κB, which plays a central rolein the inflammation linked with diabetes. NF-κB is activated viaIKK, which gets activated after various triggering stimuli.Among various mechanisms of activation of IKK complex, theimportant one is its phosphorylation.33 SNEDDS curcuminformulation more effectively inhibited this phosphorylation thannaïve curcumin and thus progression of NF-κB mediatedoverexpression of various inflammatory genes. The immunohis-tochemical analysis of NF-κB p65 subunit also replicated theresults seen with western blot analysis and SNEDDS curcuminformulation was more effective in reducing the score of NF-κBimmuno-positive cells in the nerve microsections as compared tonaïve curcumin.

NF-κB is directly involved in the regulation of COX-2activity as it binds to the cis-acting elements in the promoter ofCOX-2. Pharmacological blockade or gene ablation of COX-2was shown to prevent diabetes-induced changes in peripheralnerves including depletion of GSH, increased TNF-α, and bloodflow and nerve conduction deficits.34 iNOS, another inducibleenzyme affected by NF-κB activation, is a source for aberrantNO generation and can cause nitrosative damage along withendothelial dysfunction. NO generated by iNOS directlymodulates the blood supply to nerves and participates inmicrovascular changes following injury.35,36 SNEDDS curcu-min formulation decreased the expression of these inducibleenzymes more effectively than naïve curcumin and affordedbetter protection in DN.

NF-κB inhibition further reduced the levels of inflammatorymediators including pro-inflammatory cytokines (TNF-α and IL-6). Anti-TNF-neutralizing antibodies were shown to reduce pain-related behavior in mouse models of painful mononeuropathy.37

TNF-α binds to its receptors on cell surface and activate the IKKtriad thereby activating NF-κB cascade. IL-6 is anotherpleiotropic cytokine that contributes to chronic inflammationthat underlies insulin resistance and diabetes.38 SNEDDScurcumin formulation abrogated the increased levels of TNF-αand IL-6 more effectively than naïve curcumin thus furtherdecreasing changes associated with DN.

Thus we conclude that curcumin reversed the functional,sensorimotor and biochemical deficits by decreasing neuroin-flammation and improving antioxidant defence in DN. Further-more, we showed that nano formulation of curcumin is moreeffective in showing these effects, which can be extrapolated tothe clinical conditions with greater benefits for treatment ofvarious diabetic complications.

Acknowledgments

The authors would like to acknowledge Department ofPharmaceuticals, Ministry of Chemical and Fertilizers, Govern-ment of India for providing partial financial support for thisresearch work. Authors also acknowledge Gattefosse Pvt. Ltd.,France for gift samples of Gelucire® 44/14 and Labrasol.

Appendix A. Supplementary data

Supplementary data to this article can be found online athttp://dx.doi.org/10.1016/j.nano.2013.01.001.

References

1. Anand P, Thomas SG, Kunnumakkara AB, Sundaram C, Harikumar KB,Sung B, et al. Biological activities of curcumin and its analogues(Congeners) made by man and Mother Nature. Biochem Pharmacol2008;76:1590-611.

2. Anand P, Kunnumakkara AB, Newman RA, Aggarwal BB. Bioavail-ability of curcumin: problems and promises.Mol Pharmaceutics 2007;4:807-18.

3. Wahlang B, Pawar YB, Bansal AK. Identification of permeability-related hurdles in oral delivery of curcumin using the Caco-2 cell model.Eur J Pharm Biopharm 2011;77:275-82.

785R.P. Joshi et al / Nanomedicine: Nanotechnology, Biology, and Medicine 9 (2013) 776–785

4. Munjal B, Pawar Y, Patel S, Bansal A. Comparative oral bioavailabilityadvantage from curcumin formulations. Drug Deliv Transl Res 2011;1:322-31.

5. Pouton CW, Porter CJ. Formulation of lipid-based delivery systems fororal administration: materials, methods and strategies. Adv Drug DelivRev 2008;60:625-37.

6. Mohsin K, Long MA, Pouton CW. Design of lipid-based formulationsfor oral administration of poorly water-soluble drugs: precipitation ofdrug after dispersion of formulations in aqueous solution. J Pharm Sci2009;98:3582-95.

7. Dahan A, Hoffman A. Rationalizing the selection of oral lipid based drugdelivery systems by an in vitro dynamic lipolysis model for improvedoral bioavailability of poorly water soluble drugs. J Control Release2008;129:1-10.

8. Yallapu MM, Jaggi M, Chauhan SC. Curcumin nanoformulations: afuture nanomedicine for cancer. Drug Discov Today 2012;17:71-80.

9. Hauss DJ. Oral lipid-based formulations. Adv Drug Deliv Rev 2007;59:667-76.

10. Cameron NE, Cotter MA. Pro-inflammatory mechanisms in diabeticneuropathy: focus on the nuclear factor kappa B pathway. Curr DrugTargets 2008;9:60-7.

11. Sharma S, Kulkarni SK, Agrewala JN, Chopra K. Curcumin attenuatesthermal hyperalgesia in a diabetic mouse model of neuropathic pain. EurJ Pharmacol 2006;536:256-61.

12. Sharma S, Chopra K, Kulkarni SK. Effect of insulin and its combinationwith resveratrol or curcumin in attenuation of diabetic neuropathic pain:participation of nitric oxide and TNF-alpha. Phytother Res 2007;21:278-83.

13. Zhu Z, Anacker JL, Ji S, Hoye TR, Macosko CW, Prud'homme RK.Formation of block copolymer-protected nanoparticles via reactiveimpingement mixing. Langmuir 2007;23:10499-504.

14. Negi G, Kumar A, Kaundal RK, Gulati A, Sharma SS. Functional andbiochemical evidence indicating beneficial effect of melatonin andnicotinamide alone and in combination in experimental diabeticneuropathy. Neuropharmacology 2009;58:585-92.

15. Negi G, Kumar A, Sharma SS. Nrf2 and NF-kappaB modulation bysulforaphane counteracts multiple manifestations of diabetic neuropathyin rats and high glucose-induced changes. Curr Neurovasc Res 2011;8:294-304.

16. Ohkawa H, Ohishi N, Yagi K. Assay for lipid peroxides in animal tissuesby thiobarbituric acid reaction. Anal Biochem 1979;95:351-8.

17. Negi G, Kumar A, Sharma SS. Melatonin modulates neuroinflammationand oxidative stress in experimental diabetic neuropathy: effects on NF-kappaB and Nrf2 cascades. J Pineal Res 2011;50:124-31.

18. Kumar A, Negi G, Sharma SS. JSH-23 targets nuclear factor-kappa Band reverses various deficits in experimental diabetic neuropathy: effecton neuroinflammation and antioxidant defence. Diabetes Obes Metab2011;13:750-8.

19. Negi G, Kumar A, Sharma SS. Concurrent targeting of nitrosative stress-PARP pathway corrects functional, behavioral and biochemical deficitsin experimental diabetic neuropathy. Biochem Biophys Res Commun2010;391:102-6.

20. Calcutt NA, Freshwater JD, Mizisin AP. Prevention of sensory disordersin diabetic Sprague-Dawley rats by aldose reductase inhibition ortreatment with ciliary neurotrophic factor.Diabetologia 2004;47:718-24.

21. Cameron NE, Jack AM, Cotter MA. Effect of alpha-lipoic acid onvascular responses and nociception in diabetic rats. Free Radic Biol Med2001;31:125-35.

22. Anand P, Nair HB, Sung B, Kunnumakkara AB, Yadav VR, Tekmal RR,et al. Design of curcumin-loaded PLGA nanoparticles formulation withenhanced cellular uptake, and increased bioactivity in vitro and superiorbioavailability in vivo. Biochem Pharmacol 2010;79:330-8.

23. Yadav VR, Prasad S, Kannappan R, Ravindran J, Chaturvedi MM,Vaahtera L, et al. Cyclodextrin-complexed curcumin exhibits anti-inflammatory and antiproliferative activities superior to those ofcurcumin through higher cellular uptake. Biochem Pharmacol 2010;80:1021-32.

24. Sachs-Barrable K, Lee S, Thamboo A, Wasan K. Influence of the lipidexcipient, gelucire 44/14, on P-glycoprotein (P-gp) activity in anintestinal endothelial cell line (Caco-2). AAPS J 2006;8(S2).

25. Varma MV, Panchagnula R. Enhanced oral paclitaxel absorption withvitamin E-TPGS: effect on solubility and permeability in vitro, in situand in vivo. Eur J Pharm Sci 2005;25:445-53.

26. Zochodne DW. Diabetic neuropathies: features and mechanisms. BrainPathol 1999;9:369-91.

27. Feldman EL. Oxidative stress and diabetic neuropathy: a newunderstanding of an old problem. J Clin Invest 2003;111:431-3.

28. Negi G, Kumar A, Joshi RP, Sharma SS. Oxidative stress and diabeticneuropathy: current status of antioxidants. IIOABJ 2011;2:71-8.

29. Negi G, Kumar A, Joshi RP, Sharma SS. Oxidative stress and Nrf2 in thepathophysiology of diabetic neuropathy: old perspective with a newangle. Biochem Biophys Res Commun 2011;408:1-5.

30. Meghana K, Sanjeev G, Ramesh B. Curcumin prevents streptozotocin-induced islet damage by scavenging free radicals: a prophylactic andprotective role. Eur J Pharmacol 2007;577:183-91.

31. Suryanarayana P, Satyanarayana A, Balakrishna N, Kumar PU, ReddyGB. Effect of turmeric and curcumin on oxidative stress and antioxidantenzymes in streptozotocin-induced diabetic rat. Med Sci Monit 2007;13:BR286-92.

32. Brownlee M. Biochemistry and molecular cell biology of diabeticcomplications. Nature 2001;414:813-20.

33. Aggarwal BB, Harikumar KB. Potential therapeutic effects of curcumin,the anti-inflammatory agent, against neurodegenerative, cardiovascular,pulmonary, metabolic, autoimmune and neoplastic diseases. Int JBiochem Cell Biol 2009;41:40-59.

34. Kellogg AP, Wiggin TD, Larkin DD, Hayes JM, Stevens MJ, Pop-BusuiR. Protective effects of cyclooxygenase-2 gene inactivation againstperipheral nerve dysfunction and intraepidermal nerve fiber loss inexperimental diabetes. Diabetes 2007;56:2997-3005.

35. Zochodne DW, Levy D. Nitric oxide in damage, disease and repair of theperipheral nervous system. Cell Mol Biol (Noisy-le-Grand) 2005;51:255-67.

36. Levy D, Zochodne DW. NO pain: potential roles of nitric oxide inneuropathic pain. Pain Pract 2004;4:11-8.

37. Sommer C, Lindenlaub T, Teuteberg P, Schafers M, Hartung T, ToykaKV. Anti-TNF-neutralizing antibodies reduce pain-related behavior intwo different mouse models of painful mononeuropathy. Brain Res2001;913:86-9.

38. Hayden MS, Ghosh S. Signaling to NF-kappaB. Genes Dev 2004;18:2195-224.