metabolism and permeability of curcumin in cultured caco-2 cells

Mol. Nutr. Food Res. 2012, 56, 1–7 1DOI 10.1002/mnfr.201200113

RESEARCH ARTICLE

Metabolism and permeability of curcumin in cultured

Caco-2 cells

Julia S. Dempe, Romy K. Scheerle, Erika Pfeiffer and Manfred Metzler

Institute of Applied Biosciences, Chair of Food Chemistry, Karlsruhe Institute of Technology (KIT), Karlsruhe,Germany

Scope: Curcumin (CUR) and its major metabolite hexahydro-CUR were studied in Caco-2 cellsand in the Caco-2 Millicell R© system in vitro to simulate their in vivo intestinal metabolism andabsorption in humans.Methods and results: Analysis of the incubation medium and cell lysate showed that Caco-2cells reduce CUR to hexahydro-CUR and octahydro-CUR, and conjugate CUR and its reductivemetabolites with glucuronic acid and sulfate. Using the Caco-2 Millicell R© system, an efficienttransfer of the conjugates into the basolateral, but not the apical, compartment was observedafter apical administration. Likewise, hexahydro-CUR was reduced to octahydro-CUR, andglucuronide and sulfate conjugates almost exclusively permeated to the basolateral side. Theapparent permeability coefficients (Papp values) of CUR, hexahydro-CUR and their metaboliteswere determined and found to be extremely low for unchanged CUR, but somewhat higher forhexahydro-CUR and the conjugated metabolites.Conclusion: The results of this study clearly show that the systemic bioavailability of CURfrom the intestine after oral intake must be expected to be virtually zero. Reductive and conju-gated metabolites, formed from CUR in the intestine, exhibit moderate absorption. Thus, anybiological effects elicited by CUR in tissues other than the gastrointestinal tract are likely dueto CUR metabolites.

Keywords:

Caco-2 cells / Curcumin / Hexahydrocurcumin / Metabolism / Permeability

Received: February 17, 2012Revised: June 5, 2012

Accepted: June 12, 2012

1 Introduction

Curcumin (CUR, Fig. 1) is the major yellow-orange pigmentof turmeric, which is obtained from the rhizomes of the Asianplant Curcuma longa and extensively used in the Ayurvedicmedicine and also in the traditional Indian cooking [1, 2].CUR exhibits a broad spectrum of biological activities, amongwhich are anti-inflammatory, antimicrobial, antioxidant, an-ticarcinogenic, and cancer chemopreventive effects [2, 3]. Inanimal models and in several human studies, no serious tox-icity has been observed even at very high doses. For exam-ple, repeated ingestion of as much as 12 g of CUR per dayhas been well tolerated in three clinical phase I studies [4].Its pharmacological safety and beneficial biological activities

Correspondence: Professor Manfred Metzler, Institute of AppliedBiosciences, Chair of Food Chemistry, Karlsruhe Institute of Tech-nology (KIT), Adenauerring 20a, 76131 Karlsruhe, GermanyE-mail: [email protected]: +49-721-608-47255

Abbreviations: CUR, curcumin; HBSS, Hank’s buffered salt solu-tion; HHC, hexahydro-CUR; OHC, octahydro-CUR

make CUR a potential therapeutic agent for the treatment ofa wide variety of human diseases [2–4].

However, there are several problems that have preventedthe marketing of CUR as a drug so far. Major obstacles are thepoor aqueous solubility, intense staining color, and extremelylow oral bioavailability [4, 5]. The latter is demonstrated bythe fact that even after oral doses of up to 12 g, blood serumlevels of CUR did not exceed the low micromolar level. Inone clinical study with 4–8 g of CUR, the maximum serumconcentration observed was 1.3 �g/mL [5], while several otherclinical and animal studies reported serum levels in the lowernanogram per milliliter range [4]. Several reasons have beenproposed for the very low gastrointestinal bioavailability ofCUR, such as poor absorption, rapid metabolism, chemicalinstability, and accumulation within the cells of the intestinalepithelium [4, 6].

It is known from in vivo studies with rats and from clin-ical studies with cancer patients, as well as from several invitro investigations that CUR undergoes both phase I andphase II metabolism [7,8 and literature cited therein]. PhaseI biotransformation of CUR comprises stepwise reductionof the olefinic double bonds, leading to tetrahydro-CUR,

C© 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mnf-journal.com

2 J. S. Dempe et al. Mol. Nutr. Food Res. 2012, 56, 1–7

Figure 1. Chemical structures of CUR and its major reductivemetabolites.

hexahydro-CUR, and octahydro-CUR (Fig. 1). Both CUR andits reductive metabolites are readily conjugated with glu-curonic acid and sulfate. Reduction and conjugation appear tobe general metabolic pathways of CUR, taking place in hepaticand intestinal tissues of rats and humans, as demonstratedby several in vitro studies [8–11]. More recently, conjugationof CUR with glutathione has been demonstrated under cell-free conditions and also in human Caco-2 cells [12]. Caco-2cells are derived from a human colon tumor. However, whenkept in cell culture for about 3 weeks, Caco-2 cells featuremany characteristics of intestinal epithelial cells: a polarizedmonolayer with tight junctions and microvilli at the apicalside is formed, and various enzymes for phase I and phaseII metabolism and transport proteins of the ABC super fam-ily are expressed [13–15]. Therefore, Caco-2 cells representa widely accepted in vitro system for the human intestinalmetabolism and also for the intestinal absorption of organiccompounds, and have been recommended by the US Foodand Drug Administration (FDA) for that purpose [16].

In the present study, the Caco-2 cell system was used toaddress the crucial question of the fate of CUR during intesti-nal absorption and to clarify whether significant proportionsof parent CUR or its intestinal metabolites are bioavailablefrom the intestine after oral intake.

2 Materials and methods

2.1 Chemicals and reagents

CUR was chemically synthesized in our laboratory fromvanillin and acetylacetone according to the method of Pabon

[17]. Hexahydro-CUR was prepared by Pd-catalyzed hy-drogenation of CUR according to the method of Ueharaet al. [18], and octahydro-CUR was obtained from hexahydro-CUR by reduction with sodium borohydride. All synthesesand the characterization of the compounds, which had a pu-rity of more than 99%, are described in detail by Hoehleet al. [8]. Tetrahydro-CUR was chemically synthesized bySabinsa Corp. (Payson, UT, USA) and kindly provided bythe Arizona Center for Phytomedicine Research (Tucson,AZ, USA). �-Glucuronidase type B-1 from bovine liver withan activity of 1.24 × 106 U/g and aryl sulfatase type VIfrom Acetobacter aerogenes with an activity of 12.25 U/mLwere purchased from Sigma/Aldrich/Fluka (Taufkirchen,Germany). Fetal calf serum was purchased from Invitrogen(Karlsruhe, Germany). Dulbecco’s modified eagle medium(DMEM/F12), penicillin, streptomycin, and all other chem-icals and reagents were obtained from Sigma/Aldrich/Fluka.

2.2 Caco-2 cell culture and subculture

Caco-2 cells (DSMZ No. ACC169) were from DeutscheSammlung von Mikroorganismen und Zellkulturen (Braun-schweig, Germany). Cells were cultured at 37�C and 5% CO2

in dishes with 15 cm diameter containing 20 mL mediumand initially 0.4 × 106 cells. The culture medium consistedof DMEM/F12 with 10% fetal calf serum, 100 U/mL peni-cillin, and 100 �g/mL streptomycin, and was replaced ev-ery other day. After seven days, the medium was removedand the attached cells were washed twice with 10 mL PBSeach, followed by treatment with 2.5 mL of a 0.625% aque-ous trypsin solution for 40 s. After removal of the trypsin,cells were incubated at 37�C for 5 min and subsequentlysuspended in 10 mL medium. Cell number was determinedin a 1:10 dilution using a Neubauer chamber, and the vol-ume containing 0.4 × 106 cells was transferred to a newdish.

2.3 Metabolic study in Caco-2 cells

For studying the metabolism of CUR and hexahydro-CURin differentiated Caco-2 cells, 106 cells per well in 4 mLDMEM/F12 medium were seeded into six-well Millicell R©

plates (Millipore, Billerica, MA, USA) without inserts andgrown for 21 days with the renewal of the medium after2 or 3 days. On day 21, medium was removed, cells werewashed twice with Hank’s buffered salt solution (HBSS),and 4 mL HBSS containing the substance at 50 �M con-centration with 1% DMSO was added. After a 3-h incubation,100 �L of a 5 mM solution of bisphenol A in DMSO wereadded to each well as internal standard for HPLC-DAD anal-ysis. Subsequently, three equal aliquots of 1.2 mL each wereremoved from each well. Aliquot-1 was immediately extractedtwice with a double volume of ethyl acetate. Aliquot-2 and


Mol. Nutr. Food Res. 2012, 56, 1–7 3

aliquot-3 were mixed with equal volumes of 0.1 M potassiumacetate buffer pH 7.1 containing 100 U of �-glucuronidaseor 0.1 U of sulfatase, and incubated for 3 h at 37�C, followedby the extraction with ethyl acetate. The extracts were evap-orated to dryness under reduced pressure and the residuesdissolved in a small volume of methanol and analyzed byHPLC-DAD. The amounts of CUR and its reductive metabo-lites determined in aliquot-1 corresponded to unconjugatedmaterial, the difference between aliquot-2 and aliquot-1 rep-resented glucuronides, and the difference between aliquot-3and aliquot-1 sulfates. The conditions for the enzymatic hy-drolysis of glucuronides and sulfates were optimized in pilotstudies.

The attached cells of each well were washed twice with2 mL PBS each, scraped off the support with a plastic spatula,suspended in 1 mL HBSS and transferred into a 2-mL cen-trifuge vial. The pellet obtained after centrifugation at 1000 gfor 5 min was resuspended in 0.2 mL HBSS and lysed at–80�C for 24 h. After ultrasonic treatment, 20 �L of a 5 mMsolution of bisphenol A in DMSO was added and the lysateanalyzed by HPLC-DAD for unconjugated material, glu-curonides, and sulfates as described above for the culturemedium.

2.4 Permeability study using the Caco-2 Millicell R©

system

Each insert (apical compartment) of the six-well Millicell R©

plates was filled with 2 mL DMEM/F12 medium contain-ing 35 × 104 cells, and each well (basolateral compartment)received 4 mL medium alone. Cells were grown into a differ-entiated monolayer for 21 days, with renewal of the mediumin both compartments every other day. Then the mediumwas removed and the apical and basolateral compartmentswere washed twice with HBSS. Subsequently, 2 mL HBSScontaining 50 �M CUR or hexahydro-CUR and 1% DMSOwas filled into the apical compartment, whereas the baso-lateral compartment received 4 mL HBSS alone. During a6-h incubation, the apical and basolateral compartments ofthree wells were processed separately every hour in order todetermine the amount of unconjugated CUR and its reduc-tive metabolites, as well as their glucuronides and sulfates.The basolateral compartments containing 4 mL HBSS wereworked up as described for the metabolism study in Section2.3, whereas the work-up of the apical compartment contain-ing only 2 mL HBSS aliquots was scaled down by a factorof 2.

In order to ensure an intact monolayer of the Caco-2cells, both compartments of each well were washed once withHBSS after each experiment, and 2 mL of HBSS containing100 �g lucifer yellow per milliliter were filled into the apicalwell, whereas the basolateral compartment contained 4 mLof dye-free HBSS. After 1 h at 37�C, 200 �L of the basolateralHBSS were analyzed for lucifer yellow fluorimetrically (exci-tation at 485 nm and emission at 535 nm). Only cell layers

with less than 1% transfer of the dye from the apical to thebasolateral side were considered intact.

2.5 HPLC analysis with DAD

The quantification of CUR and its reductive metabolites wascarried out by the analytical HPLC, using an HP 1100 systemequipped with two pumps, autosampler, diode-array detec-tor, and HP ChemStation version Rev.A.07.01 software fordata collection and analysis. The wavelength was 420 nmfor the detection of CUR and 280 nm for tetrahydro-CUR,hexahydro-CUR, and octahydro-CUR. Separation was carriedout on a 250 × 4.6 mm id, 5 �m, reversed-phase Luna C8 col-umn (Phenomenex, Torrance, CA, USA) protected by a 3 ×4.0 mm id SecurityGuard C18 (ODS) column (Phenomenex).Solvent A was deionized water adjusted to pH 3.0 with formicacid, and solvent B was acetonitrile. A linear solvent gradi-ent was started immediately after injection, changing from30 B to 70% B in 35 min. Flow rate was 1.0 mL/min. Un-der these conditions, CUR, tetrahydro-CUR, hexahydro-CUR,and octahydro-CUR had retention times of 25.3, 22.8, 11.5,and 9.2 min, respectively.

2.6 Calculation of Papp

Papp values for the apical to basolateral transition were cal-culated according to Artursson and Karlsson [19] using theformula

Papp[cm/s] = (Vapi/A · t)(Cbaso/Capi) (1)

where Vapi is the volume of the apical compartment(2 mL), A is the surface area of the monolayer (4.52 cm2),t is the time (s), Cbaso is the concentration (�M) of the com-pound in the basolateral compartment (either parent com-pound or sum of parent compound and metabolites), andCapi is the initial concentration (�M) of CUR in the apicalcompartment.

3 Results

3.1 Metabolism of CUR and hexahydro-CUR

in Caco-2 cells

Differentiated Caco-2 cells were grown in a six-well cell cul-ture dish without insert and incubated with 50 �M (80 nmoltotal amount) of CUR for 3 h. Aliquots of the incubationbuffer and the cell lysate were then extracted with ethyl ac-etate directly or after hydrolysis with �-glucuronidase or sul-fatase, and the extracts analyzed by HPLC with DAD detec-tion. In the fraction of unconjugated material, obtained bydirect extraction without hydrolysis, both CUR and its reduc-tive metabolites hexahydro-CUR and octahydro-CUR, but nottetrahydro-CUR, were identified by comparison of retention



Table 1. Pattern of reductive and conjugated metabolites in the incubation medium and lysate of differentiated Caco-2 cells after incubationwith 50 �M CUR (left column) or hexahydro-CUR (right column) for 3 h.

Compartment Compound analyzed Compound incubated

CUR Hexahydro-CUR

Medium Unconjugated CUR 1.5 ± 0.9 Not applicableUnconjugated hexahydro-CUR 2.7 ± 0.2 7.4 ± 5.5Unconjugated octahydro-CUR 1.0 ± 0.5 7.0 ± 6.7CUR glucuronide 0.2 ± 0.2 Not applicableHexahydro-CUR glucuronide 1.3 ± 0.1 1.0 ± 1.5Octahydro-CUR glucuronide 0.1 ± 0.1 0.1 ± 0.2CUR sulfate n.d. (< 0.2) Not applicableHexahydro-CUR sulfate 6.0 ± 0.1 13.1 ± 6.1Octahydro-CUR sulfate 5.0 ± 3.3 17.5 ± 2.9Total compounds in medium 17.8 46.1

Cell lysate Unconjugated CUR 0.04 ± 0.01 Not applicableUnconjugated hexahydro-CUR 0.03 ± 0.03 0.04 ± 0.05Unconjugated octahydro-CUR 0.20 ± 0.03 0.38 ± 0.26Sulfate and glucuronide of CUR, Not detectable Not detectable

hexahydro-CUR, and octahydro-CURTotal compounds in cell lysate 0.27 0.42

Total recovery 18.1 46.5

Data are the percentage of original CUR concentration (mean of three or two independent experiments).

times and UV spectra with authentic reference compounds.The same two metabolites were present in the fraction ofglucuronides and sulfates. The amounts of hexahydro-CURexceeded that of CUR and octahydro-CUR in all three frac-tions, and sulfates of the two reductive CUR metaboliteswere the major products found in the incubation medium(Table 1).

Analysis of the cell lysate after incubation with CURshowed only minute amounts of unconjugated CUR,hexahydro-CUR, and octahydro-CUR. The total recovery ofCUR-related compounds from the incubation medium andcell lysate was merely 18.1% (Table 1).

When Caco-2 cells were incubated with hexahydro-CUR,reduction to octahydro-CUR occurred, and sulfates wereagain the major conjugates found in the medium, whereasthe cell lysate contained only very little unconjugated mate-rial. Total recovery accounted for 46.5% (Table 1).

3.2 Permeation of CUR in the Caco-2 Millicell R©

system

Monolayers of differentiated Caco-2 cells were used to deter-mine the in vitro absorption of CUR. First, cells were exposedfrom the apical side to 50 �M CUR and the reductive metabo-lites and conjugates determined in both compartments byHPLC-DAD at 1-h intervals for 6 h (Fig. 2). UnconjugatedCUR had virtually disappeared from the apical compartmentafter 6 h and appeared at low concentrations on the basolateralside. Large amounts of unconjugated hexahydro-CUR andsmaller amounts of unconjugated octahydro-CUR, but virtu-ally no conjugated material increased in the apical compart-ment over time. Conversely, sulfates of hexahydro-CUR and

octahydro-CUR were the major products accumulating on thebasolateral side (Fig. 2). The pattern of reductive metabolitesand conjugates was consistent with the results of the previousmetabolism study listed in Table 1, except that the amount ofhexahydro-CUR exceeded that of octahydro-CUR.

Because significant amounts of unconjugated hexahydro-CUR were detected in the apical compartment after apicalexposure of Caco-2 cells to CUR (Fig. 2), the permeability ofthis reductive metabolite was studied separately. The sameprotocol was used as for CUR, and the results are depicted inFig. 3. Disappearance of hexahydro-CUR from the apical com-partment was slow, because 50% of the administered concen-tration was still present after 5 h, together with about 25% ofunconjugated octahydro-CUR and small amounts of the sul-fates of both compounds. On the basolateral side, about 5%of hexahydro-CUR and 2% of octahydro-CUR were presentduring the sampling time, and sulfates of both compoundsincreased to about 6% during incubation.

3.3 Apparent permeability coefficient Papp of CUR

and hexahydro-CUR

From the data depicted in Figs. 2 and 3, the Papp value for theapical to basolateral transition of CUR and hexahydro-CURwas calculated both for the parent compound and for the sumof parent compound and all the metabolites (Table 2). Forthe permeability of unconjugated CUR alone, a Papp value of0.07 × 10−6 cm/s or less was obtained, whereas the Papp

value had a maximum of 1.3 × 10−6 cm/s for all CUR-relatedmaterial. The maximum Papp values for parent hexahydro-CUR and for the sum of hexahydro-CUR metabolites were3.3 × 10−6 cm/s and 5.5 × 10−6 cm/s, respectively (Table 2).


Mol. Nutr. Food Res. 2012, 56, 1–7 5

Figure 2. Time course of CUR and its metabolites in the apical and basolateral compartments of Millicell R© plates containing Caco-2 cellmonolayers during 6 h incubation with 50 �M CUR in the apical compartment (corresponding to 80 nmol CUR). Data are the mean ± SDof three independent incubations. HHC, hexahydro-CUR; OHC, octahydro-CUR.

4 Discussion

Caco-2 cells represent a widely accepted in vitro model tostudy human intestinal metabolism and absorption [13–15,19]. However, the metabolism of CUR, which is a potentialanticancer agent and has been administered orally in heroicdoses of up to 12 g per day in clinical trials, has not yetbeen determined in Caco-2 cells. The present study showsfor the first time that Caco-2 cells are capable of reductiveand conjugative metabolism, because hexahydro-CUR andoctahydro-CUR as well as their glucuronide and sulfate con-jugates could be identified as major metabolites in the incu-bation medium, whereas the cell lysate was virtually devoidof conjugated metabolites and contained only trace amountsof the unconjugated parent compound and reduction prod-

ucts. A mass balance after 3 h showed that the total amountof recovered CUR-related material was only 18%, most ofwhich consisted of sulfates and glucuronides of the reductivemetabolites, whereas unconjugated CUR and its glucuronideaccounted for less than 2%. The low recovery may have differ-ent reasons, e.g. the well-established chemical instability ofCUR at neutral or alkaline pH [20, 21], the tendency of CURto accumulate in cell membranes [6, 11], or the formationof other metabolites. The latter is considered unlikely, as wewere unable to detect the formation of significant amountsof the hepatic CUR metabolite tetrahydro-CUR [8] or of glu-tathione adducts, which have recently been claimed by Ustaet al. [12]. We believe that the main reason for the poor massbalance for CUR is the chemical instability. This assumptionis supported by the observation that a much higher recovery of



Figure 3. Time course ofhexahydro-CUR and itsmetabolites in the apical andbasolateral compartments ofMillicell R© plates containingCaco-2 cell monolayers during6 h incubation with 50 �Mhexahydro-CUR in the apicalcompartment (correspondingto 80 nmol hexahydro-CUR).Data are the mean ± SD ofthree independent incubations.HHC, hexahydro-CUR; OHC,octahydro-CUR.

more than 46% was achieved in Caco-2 cells with hexahydro-CUR (Table 1), which is much more chemically stable thanCUR (Dempe, unpublished observation).

Our data on the metabolism of CUR in Caco-2 cells areconsistent with our results on the in vitro absorption of CURin the Caco-2 cell Millicell R© system. An extremely low Papp

value of less than 0.1 × 10−6 cm/s was obtained for the api-cal to basolateral permeation of free CUR, whereas a some-what higher value of around 1 × 10−6 cm/s was determinedfor its intestinal metabolites (Table 2). According to the cor-relation of Papp values determined in Caco-2 cells in vitrowith human absorption in vivo, Papp values greater than the10 × 10−6 cm/s predict high (80–100%) absorption, whereasPapp values between 1 and 10 × 10−6 cm/s predict moderate(20–80%) and Papp less than 1 × 10−6 cm/s predict low (0–20%) absorption [19]. Therefore, our results strongly implythat CUR is almost completely lost (by first-pass metabolismand chemical degradation) during intestinal absorption, and

only minute amounts of unconjugated CUR and very smallamounts of its glucuronide and sulfate conjugates reach theportal blood (Fig. 2). This is in agreement with a report thatonly low nanomolar levels of the parent compound and itsglucuronide and sulfate were detectable in the portal circu-lation of 12 patients with hepatic metastases from colorectalcancer after daily oral doses of 3.6 g of CUR for one week [22].No free or conjugated CUR was found in the portal bloodwhen lower doses of 1.8 or 0.45 g were given. Regrettably, noanalysis of the reductive metabolites of CUR was conductedwith the portal blood, but trace levels of hexahydro-CUR andoctahydro-CUR were detected in liver tissue [22].

In a recent study on the permeation of CUR in Caco-2cells, a somewhat higher though still poor permeability ofCUR has been determined by Wahlang et al. [6]. These au-thors calculated a Papp value of 2.9 × 10−6 cm/s for the apicalto basolateral permeation of parent CUR, without measur-ing the transition of CUR metabolites. The reasons for this

Table 2. Papp values (expressed as 10−6 cm/s) calculated for different time intervals after apical exposure of Caco-2 cells to 50 �M CUR orhexahydro-CUR (HHC).

Time point (h) CUR CUR plus HHC HHC plusmetabolites metabolites

1 0.05 ± 0.05 0.50 ± 0.24 3.30 ± 1.80 5.50 ± 4.002 0.07 ± 0.07 1.01 ± 0.17 1.90 ± 0.73 4.11 ± 1.493 0.04 ± 0.06 1.21 ± 0.49 0.84 ± 0.21 2.20 ± 0.424 0.03 ± 0.03 1.12 ± 0.49 0.78 ± 0.20 2.72 ± 0.645 0.03 ± 0.02 1.05 ± 0.30 0.61 ± 0.23 2.17 ± 0.136 0.03 ± 0.03 1.33 ± 0.55 0.54 ± 0.26 2.19 ± 0.63

Data are the mean ± SD of three independent incubations.


Mol. Nutr. Food Res. 2012, 56, 1–7 7

discrepancy with our study may reside in experimental dif-ferences, as Wahlang et al. [6] used an initial concentration of170 �M CUR, a transition time of only 120 min, and a pH of6.5 in the apical compartment, at which CUR is more stablethan at pH 7.4 used in our experiments. Wahlang et al. [6]showed that basolateral to apical transition of CUR in Caco-2cells is also very low, that efflux transporters are not involved,and that some accumulation of CUR in cells occurs.

Taken together, an increasing number of experimentaland clinical studies supports the notion stated by Sharmaet al. [23] years ago, that “the low systemic bioavailability ofCUR following oral dosing seems to limit the tissues thatit can reach at efficacious concentration to exert beneficialeffects.” Thus, if CUR has indeed effects outside the gas-trointestinal tract, such effects may be due to the reductivemetabolites of CUR formed in the intestinal epithelium, i.e.hexahydro-CUR and octahydro-CUR. Interestingly, there arequite a number of studies on the pharmacological activity oftetrahydro-CUR, but very few, if any, on hexahydro-CUR andoctahydro-CUR [3]. If hexahydro-CUR turns out to have an-ticarcinogenic and antioxidant activity, its development as atherapeutic drug may be more promising than that of CUR,because hexahydro-CUR exhibits a much higher chemicalstability than CUR (Dempe, unpublished observation) andalso a higher bioavailability, according to its permeability inCaco-2 cells (Table 2).

The authors have declared no conflict of interest.

5 References

[1] Scartezzini, P., Speroni, E., Review on some plants of Indiantraditional medicine with antioxidant activity. J. Ethnophar-macol. 2000, 71, 23–43.

[2] Aggarwal, B. B., Sundaram, C., Malani, N., Ichikawa, H., Cur-cumin: the Indian solid gold. Adv. Exp. Med. Biol. 2007, 595,1–75.

[3] Anand, P., Thomas, S. G., Kunnumakkara, A. B., Sundaram,C. et al., Biological activities of curcumin and its analogues(congeners) made by man and mother nature. Biochem.Pharmacol. 2008, 76, 1590–1611.

[4] Anand, P., Kunnumakkara, A. B., Newman, R. A., Aggarwal,B. B., Bioavailability of curcumin: problems and promises.Mol. Pharm. 2008, 4, 807–818.

[5] Cheng, A. L., Hsu, C. H., Lin, J. K., Hsu, M. M. et al., Phase Iclinical trial of curcumin, a chemopreventive agent, in pa-tients with high-risk or pre-malignant lesions. AnticancerRes. 2001, 21, 2895–2900.

[6] Wahlang, B., Pawar, Y. B., Bansal, A. K., Identification ofpermeability-related hurdles in oral delivery of curcumin us-ing the Caco-2 cell model. Eur. J. Pharm. Biopharm. 2011,77, 275–282.

[7] Sharma, R. A., Steward, W. P., Gescher, A. J., Pharmacoki-netics and pharmacodynamics of curcumin. Adv. Exp. Med.Biol. 2007, 595, 453–470.

[8] Hoehle, S. I., Pfeiffer, E., Solyom, A. M., Metzler, M.,Metabolism of curcuminoids in tissue slices and subcellu-lar fractions from rat liver. J. Agric. Food Chem. 2006, 54,756–764.

[9] Ireson, C. R., Jones, D. L. J., Orr, S., Coughtrie, M. W. H. et al.,Metabolism of the cancer chemopreventive agent curcuminin human and rat intestine. Cancer Epidemiol. BiomarkersPrev. 2002, 11, 105–111.

[10] Pfeiffer, E., Hoehle, S. I., Walch, S. G., Riess, A. et al., Cur-cuminoids form reactive glucuronides in vitro. J. Agric. FoodChem. 2007, 55, 538–544.

[11] Dempe, J. S., Pfeiffer, E., Grimm, A. S., Metzler, M.,Metabolism of curcumin and induction of mitotic catastro-phe in human cancer cells. Mol. Nutr. Food Res. 2008, 52,1074–1081.

[12] Usta, M., Wortelboer, H. M., Vervoort, J., Boersma, M. G. etal., Human glutathione S-transferase-mediated glutathioneconjugation of curcumin and efflux of these conjugates inCaco-2 cells. Chem. Res. Toxicol. 2007, 20, 1895–1902.

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

[14] Press, B., Di Grandi, D., Permeability for intestinal absorp-tion: Caco-2 assay and related issues. Curr. Drug. Metab.2008, 9, 893–900.

[15] Naruhashi, K., Kurahashi, Y., Fujita, Y., Kawakita, E. et al.,Comparison of the expression and function of ATP bindingcassette transporters in Caco-2 and T84 cells on stimulationby selected endogenous compounds and xenobiotics. DrugMetab. Pharmacokinet. 2011, 26, 145–153.

[16] Center for Drug Evaluation and Research, Food and Drug Ad-ministration, Rockville, MD, Guidance for Industry: Waiverof in vivo bioavailability and bioequivalence studies forimmediate-release solid oral dosage forms based on a Bio-pharmaceutics Classification System, 2000.

[17] Pabon, H. J. J., A synthesis of curcumin and related com-pounds. Recl. Trav. Chim. Pays Bas 1964, 83, 379–386.

[18] Uehara, S., Yasuda, I., Akiyama, K., Morita, H. et al., Diaryl-heptanoids from the rhizomes of Curcuma xanthorrhiza andAlpinia officinarum. Chem. Pharm. Bull. 1987, 35, 3298–3304.

[19] Artursson, P., Karlsson, J., Correlation between oral drug ab-sorption in humans and apparent drug permeability coeffi-cients in human intestinal (Caco-2) cells. Biochem. Biophys.Res. Commun. 1991, 175, 880–885.

[20] Wang, Y.-J., Pan, M.-H., Cheng, A.-L., Lin, L.-I. et al., Stabil-ity of curcumin in buffer solutions and characterization ofits degradation products. J. Pharm. Biomed. Anal. 1997, 15,1867–1876.

[21] Pfeiffer, E., Hohle, S., Solyom, A. M., Metzler, M. Studies onthe stability of turmeric constituents. J. Food Eng. 2003, 56,257–259.

[22] Garcea, G., Jones, D. J. L., Singh, R., Dennison, A. R. et al.,Detection of curcumin and its metabolites in hepatic tissueand portal blood of patients following oral administration.Br. J. Cancer 2004, 90, 1011–1015.

[23] Sharma, R.A., Gescher, A. J., Steward, W. P., Curcumin: thestory so far. Eur. J. Cancer 2005, 41, 1955–1968.


metabolism and permeability of curcumin in cultured caco-2 cells

Documents