I1

PD

a

ARRAA

KBOCIPG

1

A(hieeartSbtlei

f

0h

International Journal of Pharmaceutics 464 (2014) 104–110

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

j ourna l h om epa ge: www.elsev ier .com/ locate / i jpharm

nvestigating permeability related hurdles in oral delivery of1-keto-�-boswellic acid

ravin Bagul, Kailas S. Khomane, Arvind K. Bansal ∗

epartment of Pharmaceutics, National Institute of Pharmaceutical Education and Research (NIPER), Sector-67, S.A.S. Nagar, Mohali, Punjab 160 062, India

r t i c l e i n f o

rticle history:eceived 9 October 2013eceived in revised form 9 January 2014ccepted 10 January 2014vailable online 22 January 2014

eywords:oswellic acidsral bioavailabilityaco-2 cells

ntestinal absorptionermeability hurdles

a b s t r a c t

11-Keto-�-boswellic acid (KBA) is an important and potent boswellic acids responsible for anti-inflammatory action of Boswellia extract. However, its pharmaceutical development has been severelylimited by its poor oral bioavailability. The present work aims to investigate the permeabil-ity related hurdles in oral delivery of KBA. Gastrointestinal stability, gastrointestinal metabolism,adsorption–desorption kinetics and Caco-2 permeability studies have been carried out. KBA was foundpoorly permeable with Papp value of 2.85 ± 0.14 × 10−6 cm/s. Higher absorptive transport indicated roleof carrier mediated transport. Moreover, KBA transport across monolayer showed saturation kineticsat higher concentrations. KBA exposed to 1�,25-(OH)2 vitamin D3 treated cell monolayer showed thelowest Papp value of 2.01 × 10−6 ± 0.02 × 10−6 cm/s indicating role of CYP3A4 mediated metabolism dur-ing KBA transport. Metabolic stability experiments in jejunum S9 fractions further confirmed this. KBAwas found unstable in simulated gastrointestinal fluids and also got accumulated in the enterocytes.

astrointestinal metabolism Sorption and desorption kinetic studies using Caco-2 cells further confirmed accumulation of KBA insidethe enterocytes. KBA also showed pH dependent permeability with higher flux at gradient pH conditionof pH 6.5 at apical and 7.4 at basolateral. Taken as whole, the major permeability related hurdles thathampered oral bioavailability of KBA included its gastrointestinal instability, CYP3A4 mediated intestinalmetabolism, accumulation within the enterocytes and saturable kinetics. The present investigation may

rug d

help in designing novel d

. Introduction

Traditionally Boswellia serrata extract is used in the Indianyurvedic medicine for the treatment of inflammatory diseases

Culioli et al., 2003; Gupta et al., 1998). It is among the fewerbal remedies designated with an orphan drug status by EMEA

n 2002 for the treatment of peritumoral brain edema (Winkingt al., 2000). In addition, boswellic acids, the main active ingredi-nts of B. serrata extracts has potent anti-inflammatory propertiesnd represent promising therapeutic agents for the treatment ofheumatoid arthritis, osteoarthritis, chronic colitis, ulcerative coli-is, crohn’s disease and bronchial asthma (Poeckel and Werz, 2006;harma et al., 1989). One of the potent boswellic acids is 11-keto-�-oswellic acid (KBA), a pentacyclic triterpene acid. KBA is reportedo be non-competitive inhibitor of 5-lipoxygenase and humaneukocyte elastase (HLE) enzymes (Safayhi et al., 1992; Sharmat al., 1988). Dual inhibition of HLE and 5- lipoxygenase enzyme

s unique property of KBA (Ammon, 1996; Ammon et al., 1991).

KBA has been reported to be effective in multiple animal modelsor a variety of human diseases and is a very safe molecule (Sharma

∗ Corresponding author. Tel.: +91 172 2214682x2126; fax: +91 172 2214692.E-mail addresses: akbansal@niper.ac.in, bansalarvind@yahoo.com (A.K. Bansal).

378-5173/$ – see front matter © 2014 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.ijpharm.2014.01.019

elivery system for KBA.© 2014 Elsevier B.V. All rights reserved.

et al., 2004). However, its development as drug is severely limitedby poor oral bioavailability (Sharma et al., 2004). IC50 value of KBAfor 5-lipoxygenase inhibition is 2.8 mM (Sailer et al., 1996). How-ever, the plasma level of KBA did not exceed 2 �M, after the intakeof 1.6 g Boswellia extract (Abdel Tawab et al., 2001). Therefore, theknowledge about intestinal absorption of boswellic acids is crucialto design strategy for enhancing its oral bioavailability.

Although permeability study of KBA has been reported (Gerbethet al., 2013; Krüger et al., 2009), identification of permeability-related hurdles in its oral delivery is still elusive. In the presentwork, gastrointestinal stability studies (simulated gastric fluid,simulated intestinal fluid), gastrointestinal metabolism studies,adsorption–desorption kinetics and permeability studies (timedependent, pH-dependent and concentration dependent) havebeen carried out to identify the permeability related hurdles in oraldelivery of KBA.

2. Materials and methods

2.1. Chemicals

KBA was obtained from Natural Remedies Pvt. Ltd. (Bangalore,India). Dulbecco’s modified Eagle’s medium (DMEM), Hank’s bal-anced salts solutions (HBSS), fetal bovine serum-heat inactivated

al of Ph

(tSd(spyHrICU

2

livPsotwtaaisUtasu

2

dAzDce

2

LccCmo(fl2

wq

2

(

P. Bagul et al. / International Journ

FBS), non-essential amino acids (NEAA), trypsin–ethylenediamineetra acetic acid (Trypsin–EDTA) solution, Penicillin–treptomycin–Amphotericin solution, lucifer yellow (LY) andimethyl sulfoxide (DMSO) were obtained from Sigma–AldrichSt. Louis, MO, USA). 2-[4-(2-hydroxyethyl)-1-piperazinyl] ethane-ulphonic acid (HEPES), 2-(N-morpholino) ethanesulphonic acid,hosphate-buffered saline (PBS) and 3-[4,5-dimethylthiazol-2-l]-2, 5-diphenyl tetrazolium bromide (MTT) were acquired fromimedia Laboratories Pvt. Ltd. (Mumbai, India). Verapamil was

eceived as a gift sample from Nicholas Piramal India Ltd. (Mumbai,ndia). Absolute ethanol was procured from Hong Yang Chemicalo. Ltd. (Jiangsu, China). Milli-Q grade water purified by a Milli-QV Purification System (Millipore, Bedford, MA, USA) was used.

.2. Cell culture

Caco-2 cells, originated from American Type Culture Col-ection (ATCC), Manassas, VA at passage no.37, were grownn DMEM with 4500 mg/L d-glucose, 110 mg/L sodium pyru-ate and l-glutamine and supplemented with 15% of FBS, 1%enicillin–Streptomycin–Amphotericin solution and 1% NEAAolution. Cells were cultured in T-75 cm2 tissue culture flasksbtained from Cellstar®, Greiner Bio-One (Germany). The cell cul-ures were maintained at 37 ◦C in CO2 incubator water jacketedith HEPA Class 100 (Forma Series II, Thermo electron Corpora-

ion, USA) with the atmospheric air kept at 95% air and 5% CO2t 95% humidity. The cells became 80–85% confluent in 4–7 daysfter which they were harvested with Trypsin–EDTA prior to seed-ng. The cells were grown on polycarbonate filters of 0.4 �m poreize (Millicell® 24-well Cell culture plate, Millipore, Billerica, MA,SA) at a seeding density of 75,000 cells per well for 21–22 days

o achieve a consistent monolayer. The growth media was changednd the transepithelial electrical resistance (TEER) value was mea-ured every alternate day. Cells from passage number 37–50 weresed for the experiments.

.3. 1˛,25-(OH)2 vitamin D3 treatment of Caco-2 cell monolayer

The Caco-2 cells were seeded on 24-well plates and grown for 10ays with supplemented DMEM mentioned in the previous section.fter 10 days, the medium was additionally supplemented withinc sulfate (3 nM), ferrous sulfate (5 nM) and 1�,25-(OH) 2 vitamin3 (250 nM). The concentration of FBS was reduced to 5%, and theells were maintained for another 3 weeks prior to the transportxperiments (Hou et al., 2007; Schmiedlin-Ren et al., 1997).

.4. HPLC analysis

HPLC analysis of KBA was carried out using a reversed phaseiChrospher C18 analytical column (4.6 mm × 250 mm, 5 �m parti-les; Merck, KgaA, Darmstadt, Germany), maintained at 35 ◦C. Theolumn was preceded by a 3 mm × 4 mm C18 guard column (Lichro-ART, Merck). The method was adapted from Sharma et al., andodified suitably (Sharma et al., 2004). The mobile phase consisted

f a mixture of organic phase (acetonitrile) and aqueous phase0.03% TFA buffer, pH 2.2) in the ratio of 90:10, respectively. Theow rate and detection wavelength (�max) were 1 mL min−1 and54 nm, respectively.

The method was validated according to the ICH guideline Q2(R1)ith respect to linearity, precision, accuracy, specificity, limit of

uantification, limit of detection and system suitability.

.5. Stability studies

Stability studies of KBA (50 �M) were carried out in HBSS bufferpH 6.5 and 7.4), Ringer’s solution, fasted state simulated gastric

armaceutics 464 (2014) 104–110 105

fluid (FaSSGF) and fasted state simulated intestinal fluid (FASSIF)in a shaker bath at 37 ◦C and 60 rpm. Samples were analyzed usingHPLC assay method at time intervals of 0, 30, 60, 90, 120 and180 min.

2.6. Metabolic stability in S9 fractions

The stomach, jejunum and colon were taken out from maleSprague Dawley rats weighing 280–300 g. All three parts (n = 3)were homogenized in 0.05 M PBS at 2–5 ◦C. The mixture was cen-trifuged at 9000 rpm, 4 ◦C for 20 min. The supernatant was removedand drug solution (5 �M) was incubated to it. Incubations of drugin S9 fraction were done at three different systems/reaction mix-tures. First reaction mixture was S9 fraction with 1 mM NADPH(cytochrome enzyme cofactor) and protease inhibitors (Leupeptin,EDTA and PMSF), second was S9 fraction with 1 mM NADPH,ritonavir (cytochrome enzyme inhibitor) and protease inhibitorsand third was S9 fraction with protease inhibitors but withoutNADPH. Hundred microliters of sample was withdrawn at specifictime intervals and double quantity of acetonitrile was added to it(Sohlenius-Sternbeck and Orzechowski, 2004; van de Kerkhof et al.,2007). The precipitated protein was removed by centrifugation at13,000 rpm for 10 min. The clear supernatant was concentrated andanalyzed using HPLC.

2.7. Adsorption and desorption study

The cell suspension containing Caco-2 cells was taken at acell density of 75,000 cells/400 �L. KBA (50 �M) was added toit and incubated for different time intervals. The cell suspen-sion containing KBA was then centrifuged at 10,000 rpm for 6 minand supernatant was removed. The pellet was washed with PBSand lysis solution was added. KBA accumulated in the cells wasextracted with methanol and analyzed by HPLC. KBA-loaded cellsafter an incubation period of 120 min were centrifuged for desorp-tion studies. The pellet was washed and re-suspended in PBS. Atspecific time intervals, the cell suspension was again centrifugedand supernatant was withdrawn and analyzed by HPLC (Wahlanget al., 2011).

2.8. MTT cytotoxicity assay

The cells were harvested and seeded in 96-well plates at aseeding density of 2 × 104 cells per well. KBA solutions were pre-pared at a concentration range of 0.01–2 mM and incubated for2 h. Ten microliters of MTT solution (5 mg/mL in PBS) was addedto each well and incubated for 4–5 h (37 ◦C, 5% CO2) to allowMTT to be metabolized. The media was dumped off and formazan(metabolic product of MTT) was re-suspended in 100 �l of DMSOand incubated for 1 h to enable thorough mixing of formazan intothe solvent. Optical density was read at 560 nm using ELISA PlateReader (Bio-Tek Instruments, Inc., USA) and background was sub-tracted at 670 nm. The percent cell viability was measured from Eq.(1):

Cell viability(%) = signal − backgroundblank − background

× 100 (1)

2.9. Transport studies

Permeability experiments were performed in a shaker incubatormaintained at 37 ◦C and 60 rpm. The transepithelial electrical resis-tance (TEER) value was measured with a Millipore ERS voltameter(Millipore, Billerica, MA, USA) in order to evaluate and determine

106 P. Bagul et al. / International Journal of Pharmaceutics 464 (2014) 104–110

Table 1Solution stability of KBA in different buffers.

Time (min) FaSSGF FaSSIF HBSS Ringer’s solution

pH 6.5 pH 7.4 pH 6.5 pH 7.4

0 100.0 100.0 100.0 100.0 100.0 100.0120 94.0 ± 0.35 85.1 ± 0.74 95.2 ± 0.24 93.7 ± 0.16 89.5 ± 0.32 87.6 ± 0.26

± 0.2

E devia

tm

T

wtap

stnarm(pData

stu4wpsaB

P

wltC(1

E

2

daLTCwt3wt

180 91.8 ± 0.20 84.7 ± 0.35 94.4

ach value represents an average of three measurements (n = 3) with their standard

he monolayer integrity (Khomane et al., 2012). The TEER value waseasured from Eq. (2):

EER = (Rmonolayer − Rblank) × A (2)

here Rmonolayer is the resistance of the cell monolayer along withhe filter membrane, Rblank is the resistance of the filter membranend A is the surface area of the membrane (0.7 cm2 in 24-welllates).

An initial stock solution of KBA in DMSO was prepared. The stockolution was then diluted with HBSS to achieve a working concen-ration of 0.25,1 and 1.8 mM. The final concentration of DMSO didot exceed 5%. The apical to basolateral transport studies (A → B)nd basolateral to apical transport studies (B → A) of KBA were car-ied out by classic method (HBSS as buffer) as well as modifiedethod (FaSSIF in apical side and 4% albumin in basolateral side)

Fossati et al., 2008; Ingels et al., 2002; Krishna et al., 2001). Trans-ort studies (A → B) were also conducted on 1�,25-(OH)2 Vitamin3-treated cell monolayers with or without itraconazole (100 �M)s a CYP3A4 inhibitor (Wahlang et al., 2011), at different pH condi-ions (pH 5, 6.5 and 7.4 of apical side and pH 7.4 of basolateral side)nd at different concentration of KBA (0.25, 1 and 1.8 mM).

For the apical to basolateral transport studies, 400 �L of the drugolution in HBSS was added to apical side and 800 �L of the blankransport buffer was added to the basolateral side. A sample vol-me of 600 �L was withdrawn from the basolateral side at 15, 30,5, 60, 90, and 120 min and the volume withdrawn was replacedith blank transport buffer each time. In basolateral to apical trans-ort studies, the initial KBA solution was added to the basolateralide and the concentration in the apical side was measured. Thepparent permeability coefficients, Papp (cm/s), for both A → B and

→ A studies were calculated from Eq. (3):

app = dQ/dt

C0 · A(3)

here dQ/dt is the cumulative transport rate (�M/min) obtained byinear regression of cumulative transported amount as a function ofime (min), A is the surface area of the inserts (0.7 cm2 in 24-wells),0 is the initial concentration of the compounds on the donor side�M). The efflux ratio (ER) was calculated from Eq. (4) (Ammon,996):

R = Papp(AB)Papp(BA)

(4)

.10. Monolayer integrity test

At the end of the experiment, the monolayer integrity test wasone by analyzing the concentration of Lucifer yellow (LY) in thepical and basolateral compartments. An initial stock solution ofY (50 mM) was prepared and diluted to 100 �M working solution.he working solution of LY (400 �L) was added to the apical side ofaco-2 cell monolayer (in the wells in which drug transport studyas performed) and 800 �L of HBSS buffer (pH 7.4) was added to

he basolateral side. The plate was then kept in a shaker incubator at7 ◦C and 60 rpm. After 120 min, 700 �L and 300 �L of the samplesere withdrawn from the basolateral side and apical side, respec-

ively. The samples were analyzed by fluorescence spectroscopy at

5 92.0 ± 0.49 88.7 ± 0.17 85.6 ± 0.20

tions in parentheses.

an excitation wavelength of 485 nm and emission wavelength of535 nm using a Spectro fluorophotometer (RF-5301-PC, Shimadzu,Japan).

2.11. Everted gut sac experiment of KBA and propranolol

The experiment was performed using everted rat jejunal seg-ments prepared from male SD rats. Jejunal portion of the intestinewas identified as the portion 15 cm after the pyloric sphincter andfor next 15 cm approximately and it was excised immediately aftersacrificing the rat. It was washed thoroughly three times withphosphate buffer saline at room temperature. The intestine wasimmediately placed in oxygenated (O2/CO2, 95%:5%) Ringer’s solu-tion at 37 ◦C. The everted jejunum, as a sac of approximately 3–4 cmlength (n = 3), was taken with one end mounted on to a fine glasscannula to allow sampling from the inside (serosal side), and theother end was secured with a thread. A weight of 1 g was hung atthe closed end to stretch the tissue (Barthe et al., 1998; Wilson andWiseman, 1954).

Samples were withdrawn at different intervals for 2 h, afteraddition of drug to the medium in outer side (mucosal side) ofthe mounted tissue. Samples were immediately processed and ana-lyzed by HPLC. The experiment was performed in Ringer’s solutionat pH value of 6.5. Permeability coefficients were also calculatedwithout everting the sac, i.e. from serosal to mucosal side, to findout the existence of efflux mechanism.

3. Results

3.1. HPLC method validation and stability study of KBA indifferent buffers

HPLC method was found linear, precise, specific and accurate.Peak purity of KBA peak was found greater than 0.999 in all theexperiments, indicating no degradant and or metabolic productwas eluting with KBA.

As shown in Table 1, KBA is expected to remain stable in HBSSbuffer during Caco-2 transport studies. However, it degraded up to10% in Ringer’s solution during everted gut sac transport studies.KBA degraded up to 9% in FaSSGF within 3 h (gastric retention time)and 15% in FaSSIF within 2 h (intestinal retention time).

3.2. Metabolic stability in S9 fractions

The metabolism of KBA was studied using stomach, jejunum,and colon of rat intestine. The results obtained are shown inFig. 1a–c. It was found that KBA metabolized differently in differentparts of gastrointestinal tract.

3.3. Adsorption and desorption studies

Adsorption is the amount of KBA that accumulated inside thecells and desorption is the amount that came out from the cellsafter suspending them into fresh buffer solution (pH 6.5). More than23.8% of KBA accumulated inside the cells after 90 min incubation

P. Bagul et al. / International Journal of Ph

0

20

40

60

80

100

120

0 20 40 60 80 100

% K

BA

rem

aine

d

Time (min)

a) St omach

N N+I D

020406080

100120

0 20 40 60 80 100

% K

BA

rem

aine

d

Time(min)

b) Jejunum

N N+I D

0

20

40

60

80

100

120

0 20 40 60 80 100

% K

BA

rem

aine

d

Time(min)

c) Colon

N N+I

Fig. 1. Metabolic stability of KBA in S9 fractions of rat stomach (a); jejunum (b); andcolon (c). The metabolic stability curves result from S9 fraction + NADPH + proteaseinhibitors (♦N), S9 fraction + NADPH + CYP3A4 inhibitor + protease inhibitors (�N + I) and S9 fraction + without NADPH+ protease inhibitors (�D).

0

10

20

30

0 30 60

% A

bsor

ptio

n/de

sorp

tion

Tim

Fig. 2. Percent absorption (♦) and desorption (�) of KBA w

armaceutics 464 (2014) 104–110 107

and then remained almost similar at 23.3%. The amount desorbedafter 2 h was 6% (Fig. 2).

3.4. Permeability experiments

3.4.1. MTT cytotoxicity assayViability of cells was directly measured using the MTT cyto-

toxicity assay. KBA, when incubated with Caco-2 cell over theconcentration range of 0.01–2 mM, showed no reduction in cellviability indicating its non-toxic nature.

3.4.2. Evaluation of Caco-2 cell monolayerThe TEER values for all wells were above 300 � cm2.

The apparent permeability coefficient value was determinedfor the permeability markers furosemide and propranolol.Furosemide is a poorly permeable drug and showed a Papp

value of 4.21 × 10−6 ± 0.07 × 10−6 cm/s. On the other hand, pro-pranolol, a highly permeable drug, showed a Papp value of33.4 × 10−6 ± 0.08 × 10−6 cm/s. These Papp values were comparableto the values reported in literature (Artursson and Karlsson, 1991;Hidalgo et al., 1989; Yee, 1997).

3.4.3. Caco-2 cell monolayer transport studiesThe apparent permeability values by classic method at A → B

and B → A transports were 2.85 × 10−6 ± 0.14 × 10−6 cm/s and0.92 × 10−6 ± 0.08 × 10−6 cm/s, respectively. Papp at A → B trans-port was significantly higher than Papp at B → A transport (P < 0.05).

The Papp at A → B transport using modified conditions was4.63 × 10−6 ± 0.06 × 10−6 cm/s. The mass balances (% recovery)were determined for classic and modified method with and with-out Caco-2 monolayer as shown in Fig. 3. The recoveries of classicand modified method using wells without Caco-2 monolayer werenearly same (29.46 ± 0.94% and 30.47 ± 1.53%, respectively). In con-trast, the recoveries with Caco-2 monolayer were significantlydifferent. The recoveries of classic and modified method with Caco-2 monolayer were 15.40 ± 2.17% and 27.24 ± 0.94%, respectively.

3.4.4. Role of CYP3A4Transport studies on 1�,25-(OH)2 vitamin D3-treated cell

monolayers were performed with or without itraconazole (CYP3A4

inhibitor) and results were compared with that of untreatedcell monolayers. KBA exposed to the treated cell monolayershowed the lowest Papp value of 2.01 × 10−6 ± 0.02 × 10−6 cm/s.In presence of itraconazole, the Papp value for KBA increased to

90 120e(min)

ith time (both studies were performed at pH 6.5).

108 P. Bagul et al. / International Journal of Pharmaceutics 464 (2014) 104–110

Fig. 3. The % recoveries of classic and modified method using wells with and withoutCaco-2 monolayer.

Fig. 4. P values for KBA using untreated, 1�,25-(OH) vitamin D -treated cellml

2m(

3

1twa

3

ows

Table 2Papp values at different concentrations of KBA.

Concentration (mM) Papp (cm/s)a Flux (mM/cm2 min)a

0.25 5.22 ± 0.16 × 10−6 0.05 ± 0.0031 4.60 ± 0.01 × 10−6 0.19 ± 0.0042 2.85 ± 0.14 × 10−6 0.22 ± 0.01

a Mean (±SD); n = 3.

app 2 3

onolayer and with itraconazole on 1�,25-(OH)2 vitamin D3-treated cell mono-ayer.

.83 × 10−6 ± 0.28 × 10−6 cm/s, thus confirming CYP3A4 mediatedetabolism as a permeability hurdle in poor permeability of KBA

Fig. 4).

.4.5. Saturable kineticsThe permeability was also performed at concentrations of 0.25,

and 1.8 mM of KBA to investigate the role of increasing concen-rations on Papp of KBA. The Papp values were found to decreaseith increasing concentration indicating saturable transport of KBA

cross cell monolayer (Fig. 5 and Table 2).

.4.6. pH-dependent transport studiesThese studies were performed to investigate the effect of pH

ver transport of KBA. The three different pH values (5, 6.5, and 7.4)ere kept at apical side while pH 7.4 was employed at basolateral

ide. The pH 6.5 was found optimum for transport as maximum flux

0

1

2

3

0 50 100 150

% C

umul

ativ

e ac

cum

ulat

ion

Time (min)

0.25 mM 1 mM 1. 8 mM

Fig. 5. Permeability profile of KBA at different concentrations.

Fig. 6. Papp values of KBA with respect to different pH employed at apical side.

rate was observed. The Papp values were found to decrease whenpH values at apical side were 5 and 7.4 (Fig. 6).

3.4.7. Lucifer yellow permeationLY transport across the Caco-2 cell monolayer is predominantly

by paracellular mechanism and is highly influenced by the loos-ening of the tight junctions. Therefore, LY permeation was used asindicator of monolayer integrity. The permeation of LY after trans-port studies was found less than 2%.

3.5. Permeability studies using everted gut sac method

The Papp of KBA at absorptive transport (A → B) and secre-tive transport (B → A) were 1.55 × 10−6 cm/s and 1.20 × 10−6cm/s,respectively. The mass balance studies during absorptive as wellas secretive transport gave important information regarding drugaccumulation. The % recoveries for absorptive and secretive trans-port were 16.7 and 57.7, respectively. The details of distribution

of KBA after 120 min during absorptive and secretive transport ineverted gut sacs are shown in Fig. 7.

Fig. 7. Distribution of KBA after 120 min (end of experiments) during absorptiveand secretive transport of KBA in everted gut sac model.

al of Ph

4

pdtfl

hjspmi(lt(

raodmK

memee(

cwwhianct

itsiw

wmasast

tCawmamoi

P. Bagul et al. / International Journ

. Discussion

As per USP, significant degradation (>5%) of a drug, suggestsotential instability (FDA, 2000). In case of KBA, it showed degra-ation greater than 5% in FaSSGF and FaSSIF. It was also revealedhat KBA is more liable to degradation in intestinal fluid than gastricuid (Table 1).

In case of stomach, NADPH-activated system showed slightlyigher metabolism of KBA than non-activated system. However, in

ejunum, a major part of small intestine, NADPH-activated systemhowed significantly higher metabolism (>60%) of KBA was as com-ared to other two systems. It clearly suggests CYP3A4 mediatedetabolism of KBA in jejunum. The same study was also performed

n colon wherein KBA did not show CYP3A4 mediated metabolismFig. 1). Transport studies on CYP3A4 expressed Caco-2 cell mono-ayer with or without presence of itraconazole further confirmedhe involvement of CYP3A4 enzyme in intestinal metabolism of KBAFig. 4).

In the adsorption studies, the KBA accumulated into the cellsapidly, and after the initial accumulation phase, the amountdsorbed in cell remained constant. KBA was found to be des-rbs suddenly in the desorption study and after the burst release,esorption become insignificant as shown in Fig. 2. The higher accu-ulation than release may be attributed to lipophilic nature of

BA.The transport studies by using classic method suggest poor per-

eability of KBA when compared with permeability markers. Thefflux ratio was 0.32. The lower B → A transport suggests that KBAight be partially subjected to a carrier-mediated transport. The

fflux ratio was less than 2, therefore it can be concluded that P-gpfflux mechanism was not involved in the net absorption of KBAZhang et al., 2003).

The apparent permeability using modified method was signifi-antly higher than Papp obtained by classic method. The recoveriesith Caco-2 cell monolayer obtained using these two methodsere significantly different (Fig. 3). In case of modified method,igher flux rate and % recovery of KBA were observed. This

mproved permeability of KBA could be attributed to reduction indsorption in the cell monolayer, increased solubility, reduction inon-specific absorption and providing sink conditions in receiverompartment. Therefore, modified method of permeability provedo be the most reliable method for lipophilic compounds like KBA.

It was revealed that the permeability decreases as concentrationncreases from 0.25,1 and 2 mM. The decrease in flux rate suggestedhat the transport of KBA across Caco-2 monolayer might involveaturation kinetic. The pH 6.5 at apical side and 7.4 at basolaterals optimum pH conditions of permeability of KBA. The Papp values

ere decreased when pH values at apical side are 5 and 7.4.The permeability values obtained by everted gut sac technique

ere supported the finding of permeability studies on Caco-2 cellonolayer. It was found that the % recovery in absorptive (16.70%)

nd secretive transports (57.70%) have significant difference. Thisuggests that mucus on absorptive side might play vital role in drugccumulation in absorptive transport (A → B). KBA has direct expo-ure to mucus present on the mucosal membrane in absorptiveransport that may lead in higher loss of KBA.

Thus, the present study provides insights into the oral absorp-ion related hurdles of KBA. Permeability experiments usingaco-2 cell monolayer and rat everted gut sac classified KBAs low permeability drug. Stability studies revealed that KBAas significantly unstable in gastro-intestinal fluids. The CYP3A4-ediated metabolism was contributed to its ‘unstable’ nature. The

dsorption-desorption studies showed ‘drug accumulation’ as aajor hurdle in poor permeability of KBA. The greater accumulation

f KBA within the cells could be attributed to its extreme lipophilic-ty. The lower secretive transport and saturable kinetics suggested

armaceutics 464 (2014) 104–110 109

that KBA might be subjected to a partially carrier-mediated trans-port. Thus increasing dose may not result into higher flux rate ofKBA. The lower efflux ratio ruled out the possibility of P-gp medi-ated efflux in transport of KBA. The KBA shows higher permeabilityat gradient pH condition than isocratic pH condition. The modifiedmethod employed in the present study showed improve perme-ability coefficient of KBA due to reduction in adsorption in the cellsand increased solubility. This suggests lipid and nano-formulationsmay helps to improve oral absorption of KBA, as these formulationsenhance permeability as well as solubility.

5. Conclusions

Major hurdles that hampered KBA transport across the intestineincluded instability in gastro-intestinal tract, intestinal metabolismby CYP3A4 and accumulation within the cells and mucus. More-over, the saturation kinetics and pH-dependence of transport ofKBA would critical issues to be considered. Identification of thesehurdles during the transport through the cell monolayer providesuseful inputs for designing of oral dosage forms for KBA.

References

Ammon, H., 1996. Salai Guggal—Boswellia serrata: from a herbal medicine to a spe-cific inhibitor of leukotriene biosynthesis. Phytomedicine 3, 67–70.

Ammon, H., Mack, T., Singh, G., Safayhi, H., 1991. Inhibition of leukotriene B4 for-mation in rat peritoneal neutrophils by an ethanolic extract of the gum resinexudate of Boswellia serrata. Planta Med. 57, 203–207.

Artursson, P., Karlsson, J., 1991. Correlation between oral drug absorption in humansand apparent drug permeability coefficients in human intestinal epithelial(Caco-2) cells. Biochem. Biophys. Res. Commun. 175, 880–885.

Barthe, L., Woodley, J., Kenworthy, S., Houin, G., 1998. An improved everted gut sacas a simple and accurate technique to measure paracellular transport across thesmall intestine. Eur. J. Drug Metab. Pharmacokinet. 23, 313–323.

Culioli, G., Mathe, C., Archier, P., Vieillescazes, C., 2003. A lupane triterpene fromfrankincense. Phytochemistry 62, 537–541.

FDA, 2000. Guidance for Industry Waiver of In-vivo Bioavailability and Bio-equivalence Studies for Immediate-Release Solid Oral Dosage Forms Basedon a Biopharmaceutics Classification System. Center for Drug Evaluation andResearch, Rockville.

Fossati, L., Dechaume, R., Hardillier, E., Chevillon, D., Prevost, C., Bolze, S., Maubon, N.,2008. Use of simulated intestinal fluid for Caco-2 permeability assay of lipophilicdrugs. Int. J. Pharm. 360, 148–155.

Gerbeth, K., Hüsch, J., Fricker, G., Werz, O., Schubert-Zsilavecz, M., Abdel-Tawab,M., 2013. In vitro metabolism, permeation, and brain availability of six majorboswellic acids from Boswellia serrata gum resins. Fitoterapia 84, 99–106.

Gupta, I., Gupta, V., Parihar, A., Gupta, S., Lüdtke, R., Safayhi, H., Ammon, H., 1998.Effects of Boswellia serrata gum resin in patients with bronchial asthma: resultsof a double-blind, placebo-controlled, 6-week clinical study. Eur. J. Med. Res. 3,511.

Hidalgo, I.J., Raub, T.J., Borchardt, R.T., 1989. Characterization of the human colon car-cinoma cell line (Caco-2) as a model system for intestinal epithelial permeability.Gastroenterology 96, 736–749.

Hou, X.-L., Takahashi, K., Kinoshita, N., Qiu, F., Tanaka, K., Komatsu, K., Takahashi, K.,Azuma, J., 2007. Possible inhibitory mechanism of Curcuma drugs on CYP3A4 in1�, 25 dihydroxyvitamin D3 treated Caco-2 cells. Int. J. Pharm. 337, 169–177.

Ingels, F., Deferme, S., Destexhe, E., Oth, M., Van den Mooter, G., Augustijns, P., 2002.Simulated intestinal fluid as transport medium in the Caco-2 cell culture model.Int. J. Pharm. 232, 183–192.

Khomane, K.S., Nandekar, P.P., Wahlang, B., Bagul, P., Shaikh, N., Pawar, Y.B., Meena,C.L., Sangamwar, A.T., Jain, R., Tikoo, K., 2012. Mechanistic insights into PEPT1-mediated transport of a novel antiepileptic, NP-647. Mol. Pharm. 9, 2458–2468.

Krishna, G., Chen, K.-J., Lin, C.-C., Nomeir, A.A., 2001. Permeability of lipophilic com-pounds in drug discovery using in-vitro human absorption model, Caco-2. Int. J.Pharm. 222, 77–89.

Krüger, P., Kanzer, J., Hummel, J., Fricker, G., Schubert-Zsilavecz, M., Abdel-Tawab,M., 2009. Permeation of Boswellia extract in the Caco-2 model and possible inter-actions of its constituents KBA and AKBA with OATP1B3 and MRP2. Eur. J. Pharm.Sci. 36, 275–284.

Tawab, M.A., Kaunzinger, A., Bahr, U., Karas, M., Wurglics, M., Schubert-Zsilavecz,M., 2001. Development of a high-performance liquid chromatographic methodfor the determination of 11-keto-beta-boswellic acid in human plasma. J. Chro-matogr. B: Biomed. Appl. 761, 221–227.

Poeckel, D., Werz, O., 2006. Boswellic acids: biological actions and molecular targets.Curr. Med. Chem. 13, 3359–3369.

Safayhi, H., Mack, T., Sabieraj, J., Anazodo, M.I., Subramanian, L.R., Ammon, H.,1992. Boswellic acids: novel, specific, nonredox inhibitors of 5-lipoxygenase.J. Pharmacol. Exp. Ther. 261, 1143–1146.

1 al of Ph

S

S

S

S

S

S

10 P. Bagul et al. / International Journ

ailer, E.R., Subramanian, L.R., Rall, B., Hoernlein, R.F., Ammon, H.P., Safayhi, H., 1996.Acetyl-11-keto—boswellic acid (AKBA): structure requirements for binding and5-lipoxygenase inhibitory activity. Br. J. Pharmacol. 117, 615–618.

chmiedlin-Ren, P., Thummel, K.E., Fisher, J.M., Paine, M.F., Lown, K.S., Watkins,P.B., 1997. Expression of enzymatically active CYP3A4 by Caco-2 cells grownon extracellular matrix-coated permeable supports in the presence of 1�, 25-dihydroxyvitamin D3. Mol. Pharmacol. 51, 741–754.

harma, M., Bani, S., Singh, G., 1989. Anti-arthritic activity of boswellic acids inbovine serum albumin (BSA)-induced arthritis. Int. J. Immunopharmacol. 11,647–652.

harma, M., Khajuria, A., Kaul, A., Singh, S., Singh, G., Atal, C., 1988. Effect of salaiguggal ex-Boswellia serrata on cellular and humoral immune responses andleucocyte migration. Agents Actions 24, 161–164.

harma, S., Thawani, V., Hingorani, L., Shrivastava, M., Bhate, V.R., Khiyani, R., 2004.Pharmacokinetic study of 11-keto [beta]-boswellic acid. Phytomedicine 11,255–260.

ohlenius-Sternbeck, A.-K., Orzechowski, A., 2004. Characterization of the rates oftestosterone metabolism to various products and of glutathione transferase and

armaceutics 464 (2014) 104–110

sulfotransferase activities in rat intestine and comparison to the correspondinghepatic and renal drug-metabolizing enzymes. Chem. Biol. Interact. 148, 49–56.

van de Kerkhof, E.G., de Graaf, I.A., de Jager, M.H., Groothuis, G.M., 2007. Inductionof phase I and II drug metabolism in rat small intestine and colon in vitro. DrugMetab. Dispos. 35, 898–907.

Wahlang, B., Pawar, Y.B., Bansal, A.K., 2011. Identification of permeability-relatedhurdles in oral delivery of curcumin using the Caco-2 cell model. Eur. J. Pharm.Biopharm. 77, 275–282.

Wilson, T.H., Wiseman, G., 1954. The use of sacs of everted small intestine for thestudy of the transference of substances from the mucosal to the serosal surface.J. Physiol. 123, 116–125.

Winking, M., Sarikaya, S., Rahmanian, A., Jödicke, A., Böker, D.-K., 2000. Boswellicacids inhibit glioma growth: a new treatment option? J. Neurooncol. 46,

97–103.

Yee, S., 1997. In vitro permeability across Caco-2 cells (colonic) can predict in vivo(small intestinal) absorption in man—fact or myth. Pharm. Res. 14, 763–766.

Zhang, Y., Bachmeier, C., Miller, D.W., 2003. In vitro and in vivo models for assessingdrug efflux transporter activity. Adv. Drug Deliv. Rev. 55, 31–51.