Bioorganic & Medicinal Chemistry Letters 23 (2013) 4248–4252

Contents lists available at SciVerse ScienceDirect

Bioorganic & Medicinal Chemistry Letters

journal homepage: www.elsevier .com/ locate/bmcl

New bioactive dihydrofuranocoumarins from the rootsof the Tunisian Ferula lutea (Poir.) Maire

0960-894X/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.bmcl.2013.04.081

⇑ Corresponding author. Tel.: +216 73500279; fax: +216 73500278.E-mail address: hichem.benjannet@yahoo.fr (H. Ben Jannet).

Saoussen Ben Salem a, Aymen Jabrane a,b, Fethia Harzallah-Skhiri b, Hichem Ben Jannet a,⇑a Laboratory of Heterocyclic Chemistry, Natural Products and Reactivity, Team: Medicinal Chemistry and Natural Products, Faculty of Sciences, University of Monastir,Monastir, Tunisiab Laboratory of Genetics, Biodiversity and Valorisation of Bioresources (LR11ES41), Higher Institute of Biotechnology of Monastir, University of Monastir, Monastir, Tunisia

a r t i c l e i n f o

Article history:Received 22 March 2013Revised 23 April 2013Accepted 29 April 2013Available online 6 May 2013

Keywords:Ferula luteaMedicinal plantDihydrofuranocoumarinsCytotoxicAnti-acetylcholinesteraseAntioxidant

a b s t r a c t

A phytochemical investigation of the roots of Ferula lutea (Poir.) Maire led to the isolation of newdihydrofuranocoumarins as two inseparable isomers, (�)-5-hydroxyprantschimgin 1 and (�)-5-hydroxy-deltoin 2, together with eight known compounds, (�)-prantschimgin 3, (�)-deltoin 4, psoralen 5, xantho-toxin 6, umbelliferone 7, caffeic acid 8, b-sitosterol 9 and stigmasterol 10. Their structures wereelucidated on the basis of extensive spectroscopic methods, including 1D and 2D NMR experimentsand mass spectroscopy analysis, as well as by comparison with literature data. The anti-acetylcholines-terase and cytotoxic effects of the isolates and antioxidant activities of the mixture (1+2) were also eval-uated in this work. Results showed that the mixture (1+2) has the most cytotoxic activity with IC50 values0.29 ± 0.05 and 1.61 ± 0.57 lM against the cell lines HT-29 and HCT 116, respectively. The greatest ace-tylcholinesterase inhibitory activity (IC50 = 0.76 ± 0.03) was exhibited by the xanthotoxin 6. In addition,the mixture (1+2) was investigated for its antioxidant activity and showed IC50 values 18.56, 13.06,7.59, and 4.81 lM towards DPPH free radical scavenging, ABTS radical monocation, singlet oxygen andhydrogen peroxide, respectively.

� 2013 Elsevier Ltd. All rights reserved.

The genus Ferula, belonging to the family Apiaceae, includesabout 170 species, among which 133 species occurring from cen-tral Asia westward throughout the Mediterranean region to north-ern Africa and 30 species from Iran.1–3 The Tunisian flora comprisesfour species: Ferula communis, Ferula tingitana, Ferula tunetana andFerula lutea.4 Its chemistry was largely investigated by many re-search groups.5–9 Several species of this genus are used as spicesand are well-known medicinal plants since ancient times.10 Thegum resins of the roots from several Ferula species are reportedto be used for stomach disorders, rheumatism, headache, arthritis,and dizziness.11,12 Some species are used in traditional foods aswell as in folk medicine as treatment for skin infections13 anddiabetes, as well as to prevent convulsion and hysteria.10 Ferulaspp. are also known for their toxicity. A chemotype of F. communiscontaining a prenylated coumarin known as ferulenol and relatedanalogues were responsible for ferulosis, a lethal haemorrhagicdisease which affect domestic animals in Sardinia.14 The Ferulagenus is well documented as a good source of biologically activecompounds such as sesquiterpene derivatives15–19 daucanes,humulanes, himachalanes, germacranes, eudesmanes, and guain-anes.15,20–25 Sesquiterpene derivatives, especially sesquiterpenecoumarins, were stored in the roots of the plants; therefore the

roots are a better source for isolating sesquiterpene coumarinsthan the aerial parts.9,26

Daucane esters from F. communis and Ferula arrigonii showedantiproliferative activity on human colon cancer lines and calciumionophoretic and apoptotic effects in the human jurkat T-cellline.27,28 A recent study carried out on the roots of F. tunetanahas led to the isolation of two new sesquiterpenes, tunetanin Aand tunetacoumarin A, in addition to eight known compounds,13-hydroxyfeselol, 3-angeloxycoladin coladin, coladonin,isosmarcandin, umbelliprenin, propiophenone, b-sitosterol andstigmasterol.9 Currently, there is a considerable interest in thechemistry and pharmacology of Ferula species that have not beenstudied so far. The biological importance of some species of thisgenus prompted us to investigate the roots of the Tunisian F. lutea(Poir.) Maire previously not chemically studied. In this context, wereport here the investigation of the roots of F. lutea which allowedthe isolation and structure elucidation of two new furanocouma-rins in mixture, 1 and 2, together with eight known compounds,3–10 (see Fig. 1). Furthermore, the cytotoxic, anti-acetylcholines-terase activities of the isolates and the antioxidant activity of thetwo new inseparable isomers 1 and 2 were evaluated.

The CH2Cl2-soluble portion obtained from the MeOH extract ofthe roots of F. lutea was further fractionated by successive columnchromatography to afford two new dihydrofuranocoumarins 1 and2 in mixture, together with eight known compounds, 3–10.29

2

3

456

78

9

103'

2'4'

1''2''

3''4''

5''

6' 5'

R2

1''2''

3''

4''

5''

R1

1

2

O-sen

O-ang

2

OH

OH

3

4

''5

1''2''

3''4''

O-sen

O-ang

1''2''

3''

4''

''5

H

H

R1

5O OO 2

3

456

78

9

103'

2'

6

O OO 2

3

456

78

9

103'

2'

OCH3

7O O2

3

456

78

9

10

HO 8HO

HO

O

OH

1

2

3

2'

6'

3'

5'

4'

HO

36

5

20

17

24

9

HO

3 56

17

2024

22

23

10

O O

Figure 1. Structures of compounds 1–10.

S. Ben Salem et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4248–4252 4249

The mixture of compounds 1 and 2 was isolated as yellow oiland showed on TLC a spot featuring a characteristic blue fluores-cence under UV light. The common molecular formula of 1 and 2was found to be C19H20O6, on the basis of the pseudomolecularion peak at m/z 345 for the [MH]+ in the ES-MS. The IR spectrumshowed the presence of OH (3350 cm�1) and COO (1730 and1700 cm�1) groups. The structure of 1 and 2 were elucidated onthe basis of the 1H and 13C NMR spectral data (Table 1).

The 1H NMR spectrum showed the presence of all proton signalsof prantschimgin 330 and deltoin, 4,31 except those of H-5 of thetwo structures. This spectrum showed two signals at d 5.96 (1H,d, J = 9.6 Hz) and 7.96 (1H, d, J = 9.9 Hz) as AB-type signals, attrib-utable to H-3 and H-4 of the two structures, respectively. The same1H NMR spectrum displayed an ABX system at d 4.96 (1H, dd,J = 9.6, 7.8 Hz, H-20), d 3.07 (1H, dd, J = 15.9, 8.7 Hz, H-30a) and d2.96 (1H, dd, J = 15.6, 7.8, H-30b) relative to compound 3, analogousof prantschimgin and the same system at d 4.9 (1H, dd, J = 10.2,8.1 Hz, H-20) and d 3.24 (2H, m, H-30) corresponding to compound4, analogous of deltoin. These data suggested the presence of the

dihydrofuranocoumarin skeleton in 1 and 2. The presence of twosinglets very close to dH 6.20 and 6.21, attributable in prantschim-gin 3 and deltoin 4 to the aromatic proton H-8, respectively and thedisappearance of that of H-5 at around dH (7.17, s) suggest the sub-station of C-5 in both 1 and 2. The NMR spectroscopic data togetherwith the molecular formula, suggested that the proton H-5 issubstituted by a hydroxyl group in 1 and 2.

The other signals on the same spectrum coincide with the spec-tral data of a senecioyl and angeloyl moieties in both 1 and 2,respectively.

The 13C NMR spectrum shows 24 signals including 11 attributedto the common dihydrofuranocoumarin system. The disappear-ance of the signal of C-5 at dC 123.2 ppm in prantschimgin 3 andat 127.3 ppm deltoin 4 and the appearance in of a signal at dC

168.1 ppm reinforces the presence of a hydroxyl group attachedto C-5.

The 13C NMR spectrum shows perfectly that this compound is amixture of two isomeric forms, and that it is just out of the twosubstituents that are senecioyle and angeloyl. Moreover, the

Table 11H (300 MHz) and 13C (75 MHz) NMR spectral data for compounds 1, 2, 3 and 4 in CDCl3

Position 1 2 3 4

d (C) d (H) d (C) d (H) d (C) d (H) d (C) d (H)

2 160.4 — 160.4 — 161.3 — 161.3 —3 112.4 5.96 (d, J = 9.9) 112.4 5.96 (d, J = 9.9) 112.2 6.19 d (9.0) 112.8 6.20 d (9.6)4 143.9 7.96 (d, J = 9.6) 143.9 7.96 (d, J = 9.6) 143.6 7.57 d (9.6) 143.2 7.58 d (9.6)5 168.1 — 168.1 — 123.2 7.18 (s) 127.3 7.16 (s)6 107.8 — 107.8 — 124.5 — 128.3 —7 166.8 — 166.8 — 163.4 — 161.4 —8 93.9 6.21 (s) 93.9 6.20 (s) 97.9 6.72 (s) 104.5 6.77 (s)9 153.9 — 153.9 — 155.8 — 156.4 —10 112.0 — 112.0 — 112.7 — 113.1 —2’ 92.5 4.96 (dd, J = 9.6, 7.8) 92.7 4.90 (dd, J = 10.2, 8.1) 88.8 5.11 dd (9.6; 8.4) 88.7 5.12 t (5.1)30a 31.0 2.96 (dd, J = 15.6, 7.8, Ha) 31.0 3.24 (m, Ha,b) 29.6 3.18 dd (15.6, 7.8) 29.9 2.90 dd (15.8, 5.4)3’b — 3.07 (dd, J = 15.9, 8.7, Hb) — — — 3.23 dd (16.8, 9.6) — 3.22 dd (16.8, 4.8)40 85.4 86.1 81.3 77.750 24.8 1.42 (s) 24.9 1.48 (s) 21.8 1.51 (s) 20.2 1.38 (s)6’ 25.4 1.45 (s) 25.5 1.48 (s) 22.3 1.57 (s) 22.4 1.39 (s)100 170.0 171.3 165.8 166.9 —200 120.6 5.45 (m) 132.3 116.9 5.53 (s) 128.7 —300 160.3 — 141.3 5.87 (qq, J = 7.2, 1.5) 156.5 — 139.2 6.10 qq (7.2, 1.5)400 30.9 1.73 (d, J = 1.5) 19.1 1.77 (m) 27.3 1.83 (s) 16.5 1.89 (m)500 23.6 1.95 (d, J = 0.9) 23.9 1.57 (m) 20.0 2.08 (s) 15.6 1.83 (m)

4250 S. Ben Salem et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4248–4252

analysis of the spectra HSQC, 1H–1H COSY and HMBC of thiscompound allowed us to determine all the shifts and twocorresponding structures without ambiguity as being a mixtureof (�)-5-hydroxyprantschimgin and (�)-5-hydroxydeltoin with aproportion of 58% and 42%, respectively. The stereochemical pro-posal for carbon C-20 1 has been attributed based on comparisonof the coupling constants of the protons H-2 and H-3’ in compari-son to those of (�)-prantschimgin 330 and (�)-deltoin 4.31

Compounds 3–10 were identified as psoralen 5,32 xanthotoxin6,33 umbelliferone 7,30 caffeic acid 8,34 b-sitosterol 935 and stig-masterol 1035 by spectroscopic analyses and comparisons withliterature data.

Compounds 1–8 were tested for their cytotoxic activities to-wards the human colorectal cancer cell lines HCT 116 and HT-29,according to the method of Carmichael et al.36a The results areshown in Table 2. As determined by the MTT assay,36b the strongactive compounds were the mixture (�)-5(OH)-prantschimgin/deltoïn (1+2) and (�)-deltoin 4 with IC50 values of 0.29 ± 0.05and 0.93 ± 0.11 lM against HCT 116, respectively and 1.61 ± 0.57and 7.6 ± 1.05 lM against HT-29, respectively. The relative highactivity of the mixture (1+2) by comparison to that of (�)-prants-chimgin 3 and (�)-deltoin 4 towards the two used cell linesshowed the clear contribution of the hydroxyl group in C-5 in both1 and 2 for this activity.

Xanthotoxin 6 was considered inactive on both cell lines withIC50 > 100 lM and, Psoralen 5 and caffeic acid 8 showed moderatecytotoxicity on both used cell lines.

Table 2Cytotoxicities of compounds 1–8

Compound IC50 (lM) ± SD

HCT 116 HT-29

1+2 1.61 ± 0.57 0.29 ± 0.053 14. 95 ± 4.9 5.20 ± 1.024 7.60 ± 1.05 0.93 ± 0.115 45.60 ± 11.93 41.70 ± 9.826 >100 >1007 8.05 ± 2.04 4.35 ± 1.098 29.73 ± 5.18 27.16 ± 3.09Paclitaxel (6.13 ± 3.90) � 10�3 (1.98 ± 0.50) � 10�3

IC50 values are means ± standard deviations (n = 2).

The compounds 3–8 did not previously tested against HCT 116and HT-29 cell lines. Umbelliferone 7, psoralen 5, caffeic acid 8 and(�)-deltoin 4 isolated from other sources37–40 were tested againstMT-4 (IC50 = 2341 lM); K562 (IC50 = 131 lM); AGS (IC50 = 129 lM)and ATCC6538P (IC50 = 101 lM) cancer cell lines, respectively.These compounds were shown to be less cytotoxic against theabove mentioned cell lines than towards HCT 116 and HT-29.

Furthermore, xanthotoxin 6 and (�)-prantschimgin 3 isolatedby Sigurdsson et al.41 and Yan et al.42 and tested against PANC-1(IC50 = 83.3 lM) and MES-SA/DX5 (IC50 = 3.1 lM) cell lines, wereshown to be more active than towards HCT 116 and HT-29.

The results are consistent with those cited in the literatureshowing that compounds belonging to the class of furanocouma-rins and coumarins are endowed by the cytotoxic activity.43,44

The acetylcholinesterase (AChE) inhibition of the isolated com-pounds was performed according to the methods developed byFalé et al.45a,b using commercially available Eserin as a referencestandard.

Only xanthoxin 6 showed a significant inhibitory effect with anIC50 of 0.76 ± 0.31 lM. Psoralen 5 has a moderate activity (IC50 of6.48 ± 1.05 lM). It is clear that the methoxy group in position 8in xanthoxin 6 considerably increased (8.5 times) its anti-acetylch-olineterase activity compared to that of psoralen 5 whose position8 is not substituted. While the (�)-prantschimgin 3, (�)-deltoin 4,(�)-5(OH)-prantschimgin/deltoïn (1+2), umbelliferone 7 and caf-feic acid 8 were inactive with an IC50 > 100 lg/mL. The resultsare shown in Table 3.

The mixture of the two new isomers (�)-5(OH)-prantschimgin/deltoin (1+2) was tested for its antioxidant activity towards DPPH

Table 3AChE inhibitory activity of compounds 1–8

Compound IC50 (lM) ± SD

1+2 >1003 >1004 >1005 6.48 ± 1.056 0.76 ± 0.317 >1008 >100Eserin 0.0029 ± 0.0004

IC50 values are means ± standard deviations (n = 3).

Table 4Antioxidant activity of compounds 1+2

Compound IC50 (lM) ± SD

DPPH ABTS O�2 H2O2

1+2 18.56 ± 1.22 13.06 ± 2.10 7.59 ± 2.10 4.81 ± 1.29BHT 9.02 ± 0.49 6,85 ± 0.11 7.26 ± 0.13 6.38 ± 0.04

IC50 values are means ± standard deviations (n = 3).

S. Ben Salem et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4248–4252 4251

free radical, ABTS radical monocation, singlet oxygen and hydrogenperoxide. Lower IC50 value indicated higher antioxidant activity.The results showed an IC50 values 18.56, 13.06, 7.59 and 4.81 lg/mL for DPPH free radical scavenging,46a,b ABTS radical monoca-tion,47a,b singlet oxygen48a,b and hydrogen peroxide,49a,b respec-tively. The mixture (1+2) has lower scavenging ability on DPPHradicals and ABTS radical cation when compared to that of BHTused as a positive control (IC50 = 9.02 ± 0.49 lg/mL and6.85 ± 0.11 lg/mL, respectively). In case of superoxide radical, thetwo new isomers (�)-5(OH)-prantschimgin/deltoin (1+2) showedalmost the same activity (7.59 ± 2.10 lg/mL) than given by BHT(7.26 ± 0.13 lg/mL). Towards hydrogen peroxide, the mixture(1+2) (IC50 = 4.81 ± 1.29 lg/mL) was more active than BHT(IC50 = 6.38 ± 0.04 lg/mL). These activities can be due to the pres-ence of the phenolic group at C-5 in both 1 and 2. The results areshown in Table 4.

The literature reports that umbelliferone 7 and caffeic acid 8showed an interesting antioxidant activity.50,51 Umbelliferone 7exhibited an excellent DPPH free radical scavenging activity withan IC50 = 36.1 lg/mL.50 Caffeic acid 8 tested towards DPPH freeradical scavenging, ABTS radical monocation, singlet oxygen andhydroxyl radical scavenging, showed a higher antioxidant activitywith IC50 values 0.39, 0.57, 0.38 and 0.37 lg/mL, respectively.51

The interesting cytotyoxic activity of the mixture (�)-5(OH)-prantschimgin/deltoïn (1+2) will prompt us to evaluate theirin vivo cytotoxic effect towards HCT 116, HT-29 and other cancercell lines. The synthesis of some structural analogues by the exploi-tation of the phenolic function in both 1 and 2 is desirable in orderto try to check the contribution of this function to this cytotoxicactivity and to improve it if possible.

References and notes

1. Pimenov, M. G.; Leonov, M. J. Roy. Bot. Gard. 1993, 156.2. Ghasemi, Y.; Faridi, P.; Mehregan, I.; Mohagheghzadeh, A. Chem. Nat. Compd.

2005, 41, 311.3. Asili, J.; Sahebkar, A.; Fazly Bazzaz, B. S.; Sharifi, S.; Iranshahi, M. J. Essent. Oil

Bear. Plants 2009, 12, 7.4. Alapetite, G. P. Tunis 1979, 607.5. Murray, R. D. H.; Mendez, J.; Brown, S. A. The Natural Coumarins: Occurrence,

Chemistry and Biochemistry, 1st ed.; J. Wiley & Sons Inc.: New York, 1982.6. Iranshahi, M.; Hassanzadeh-Khayyat, M.; Sahebkar, A.; Famili, A. J. Essent. Oil

Bear. Plants 2008, 11, 143.7. Iranshahi, M.; Kalategi, F.; Sahebkar, A.; Sardashti, A.; Schneider, B. Pharm. Biol.

2010, 60, 657.8. Sahebkar, A.; Iranshahi, M. Asian Biomed. 2010, 4, 835.9. Jabrane, A.; Ben Jannet, H.; Mighri, Z.; Mirjoletd, J. F.; Duchamp, O.; Harzallah-

Skhiri, F.; Lacaille-Dubois, M. A. Chem. Biodivers. 2010, 7, 392.10. Boulos, L. Algonac, Medicinal Plants of North Africa; ML. Reference Publications

Inc.: Alyonae, 1983.11. Tamemoto, K.; Takaishi, Y.; Chen, B.; Kawazoe, K.; Shibata, H.; Higuti, T.;

Honda, G.; Ito, M.; Takeda, Y.; Kodzhimatov, O. K.; Ashurmetov, O.Phytochemistry 2001, 58, 763.

12. Gonzalez, A. G.; Barrera, J. B. Prog. Chem. Org. Nat. Prod. 1995, 64, 1.13. Appendino, G.; Spagliardi, P.; Cravotto, G.; Pocock, V.; Milligan, S. J. Nat. Prod.

2002, 65, 1612.14. Arnoldi, L.; Ballero, M.; Fuzzati, N.; Maxia, A.; Mercalli, E.; Pagni, L. Fitoterapia

2004, 75, 342.15. Ahmed, A. A. Phytochemistry 1999, 50, 109.16. Ahmed, A. A.; Abd El-Razek, M. H.; Nassar, M. I.; Izuma, S.; Ohta, S.; Hirata, T.

Phytochemistry 2001, 58, 1289.17. Iranshahi, M.; Shahverdi, A. R.; Mirjani, R.; Amin, G. R.; Shafiee, A. Z.

Naturforsch. C 2004, 59, 506.

18. Iranshahi, M.; Shaki, F.; Mashlab, A.; Wessjohan, L. A. J. Nat. Prod. 2007, 70,1240.

19. Iranshahi, M.; Kalategi, F.; Rezaee, R.; Shahverdi, A. R.; Ito, C.; Furukawa, H.;Tokuda, H.; Itoigawa, M. Planta Med. 2008, 74, 147.

20. Appendino, G.; Jakupovic, J.; Alloatti, S.; Ballero, M. Phytochemistry 1997, 45,1639.

21. Kojima, K.; Isaka, K.; Ondognii, P.; Zevgeegiino, O.; Davgiin, K.; Mizukami, H.;Ogihara, Y. Chem. Pharm. Bull. 1999, 47, 1145.

22. Kojima, K.; Isaka, K.; Ondognii, P.; Zevgeegiino, O.; Gombosurengyin, P.;Davgiin, K.; Mizukami, H.; Ogihara, Y. Chem. Pharm. Bull. 2000, 48, 353.

23. Chen, B.; Teranishi, R.; Kawazoe, K.; Takaishi, Y.; Honda, G.; Itoh, M.; Takeda, Y.;Kodzhimatov, O. K. Phytochemistry 2000, 54, 717.

24. Su, B. N.; Takaishi, Y.; Honda, G.; Itoh, M.; Takeda, Y.; Kodzhimatov, O. K.;Ashurmetov, O. J. Nat. Prod. 2000, 63, 436.

25. Su, B. N.; Takaishi, Y.; Honda, G.; Itoh, M.; Takeda, Y.; Kodzhimatov, O. K.;Ashurmetov, O. J. Nat. Prod. 2000, 63, 520.

26. Iranshahi, M.; Amanolahi, F.; Schneider, B. Avicenna J. Phytomed. 2012, 2, 133.27. Poli, F.; Appendino, G.; Sachetti, G.; Ballero, M.; Maggiano, N.; Ranaletti, F. O. J.

Phytother. Res. 2005, 19, 152.28. Macho, A.; Blanco-Molina, M.; Spagliardi, P.; Appendino, G.; Bremner, P.;

Heinrich, M.; Fiebich, B. L.; Munoz, E. J. Biochem. Pharmacol. 2004, 68, 875.29. Roots of Ferula lutea were collected from the region of Béja (Tunisia) and

identified by Professor Fethia Harzallah-Skhiri, in the Laboratory of Vegetalbiology and Botanic, High Institute of Biotechnology of Monastir, Tunisia and avoucher specimen (F.L.F-10) was deposited in the same laboratory. The roots ofF. lutea (fresh weight 3.5 kg) were crushed and extracted three times withMeOH (15 L) at rt. After evaporation of the solvent under vacuum, a MeOHextract (190 g) was obtained, dissolved in H2O and partitioned against CH2Cl2

(5:l), EtOAc (5:l), and BuOH (3:l). The CH2Cl2 soluble fraction was concentratedto give a residue (25 g), from which 23 g were prefractionated by CC (SiO2;6 � 110 cm) eluted with increasing amounts of EtOAc in petroleum ether (PE)and finally with MeOH to give 20 fractions (Fr. 1-Fr. 20). Fr. 18 (178 mg) wassubjected to CC (SiO2; 2 � 45 cm; PE/EtOAc with increasing polarity) to affordthe mixture 1+2 (21 mg). The precipitation of Fr. 3 (690 mg) and Fr. 4 (972 mg)in MeOH yielded 9 (47 mg) and 10 (55 mg), resp. Fr. 7 (3.94 g) was precipitatedin hexane and EtOAc to give 3 (594 mg) and 4 (60 mg), resp. Fr. 17 (2.3 g) wassubjected to CC (SiO2; 2 � 70 cm; PE/ EtOAc with increasing polarity) to givefive subfractions. Subfraction Fr. 17.4 (1.6 g) was precipitated in hexane to yield8 (840 mg). Fr. 13 (50 mg) was dissolved in EtOAc and precipitated by additionof increasing amounts of hexane to obtain, after filtration, 7 (21 mg). Fr. 6(221 mg) was subjected to CC (SiO2; 2 � 70 cm; PE/EtOAc with increasingpolarity) to give seven subfractions. Subfraction Fr. 6.5 (43 mg) was furtherpurified by prep. TLC eluted twice with PE/EtOAc (8:2) to yield 6 (15 mg). Fr. 9(216 mg) was precipitated from the mixture hexane/EtOAc (9:1) to give, afterfiltration, 5 (18 mg).

30. Khalighi-Sigaroodi, F.; Hadjiakhoondi, A.; Shafiee, A.; Mozaffarian, V. A.;Shahverdi, A. R.; Alavi, S. H. R. J. DARU 2006, 14, 214.

31. Razavi, S. M.; Nazemiyeh, H.; Hajiboland, R.; Kumarasamy, Y.; Delazar, A.;Nahar, L.; Sarker, S. D. Braz. J. Phcog. 2008, 18, 1.

32. Liu, R.; Li, A.; Sun, A.; Kong, L. J. Chromatogr. A 2004, 1057, 225.33. Sajjadi, S. E.; Noroozi, P. J. Res. Pharm. Sci. 2007, 2.34. Park, H. Y.; Nam, M. H.; Lee, H. S.; Jun, W.; Hendrich, S.; Lee, K. W. J. Food Chem.

2010, 119, 724.35. Rubinstein, I.; Goad, J.; Clague, A. D. H.; Mulheirn, L. J. Phytochemistry 1976, 15,

195.36. (a) Carmichael, J.; Degraff, W. G.; Gazdar, A. F.; Minna, J. D.; Mitchell, J. B. Cancer

Res. 1987, 47, 936; (b) Cytotoxicity was towards two human colorectal celllines, HTC 116 and HT-29. Paclitaxel was used as positive control. Theantiproliferative effect of the mixture (1+2) was monitored using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). HT-29 andHCT 116 Cells were seeded at an initial density of 5000 or 10,000 cells/wellin 96-well plates and treated with medium having various concentrations ofthe tested sample. DMSO controls (0.1%) did not affect cell proliferation. After96 h 20 lL of MTT solution (5 mg/mL in PBS (phosphate-buffered saline) wereadded to the culture medium and the reaction mixture was incubated at 37 �Cin a 5% CO2 atmosphere for 4 h. The MTT solution was aspired and 200 lL ofdimethylsulfoxide (DMSO) were added. The optical density (OD) wasmeasured spectrophotometrically at 570 nm by an ELISA plate reader. Theresults were expressed as concentrations of compound producing 50% toxicity(IC50 value). Tests were carried out in duplicate.

37. Mert, T.; Fafal, T.; Kivçak, B.; Öztürk, H. T. J. Fac. Pharm. Ankara 2008, 37, 199.38. Wang, Y.; Hong, Ch.; Zhou, Ch.; Xu, D.; Qu, H-b. Evid. Based Complement Altern

2008, 2011, 7.39. Mijeong, K.; In Young, H.; Je. Hyuk, L.; Kun Ho, S.; Choon-Sik, J.; Joohee, J. J.

Health Sci. 2011, 57, 292.40. Ganbaatar, Zh.; Shul’ts, E. E.; Petrova, T. N.; Shakirov, M. M.; Otgonsuren, D.;

Pokrovskii, A. G.; Tolstikov, G. A. Chem. Nat. Compd. 2013, 49, 102.41. Sigurdsson, S.; Ögmundsdottir, H. M.; Gudbjarnason, S. Z. Naturforsch. C. 2004,

59, 527.42. Yan, X.; Kyung, H. M.; Kyeong, L. Bull. Korean Chem. Soc. 2009, 30, 48.43. Razavi, S. M.; Zahri, S.; Motamed, Z.; Ghasemi, G. Iran J. Basic Med. Sci. 2010, 13,

138.44. Lee, C.; Chiang, L.; Cheng, L.; Liaw, C.; Abd El-Razek, M. H.; Chang, F.; Wu, Y. J.

Nat. Prod. 2009, 72, 1568.45. (a) Falé, P. L.; Borges, C.; Madeira, P. J. A.; Ascensão, L.; Araujo, M. E. M.;

Florêncio, M. H. J. Food Chem. 2009, 114, 798; (b) Briefly, 100 lL of 50 mM Tris–

4252 S. Ben Salem et al. / Bioorg. Med. Chem. Lett. 23 (2013) 4248–4252

HCl buffer, pH 8, 50 lL of sample (1+2) and 10 lL of acetylcholinesterasesolution containing 0.26 U/mL were mixed in a microwell plate and left toincubate for 15 min. Subsequently, 50 lL of a solution of AChI (0.023 mg/mL)and 140 lL of 3 mM DTNB were added. The absorbance was read at 405 nmwhen the reaction reached equilibrium. A control reaction was carried outusing water instead of essential oil and it was considered 100% activity. Thepercentage inhibition ((%) IP) is given as follow: (%) IP = 100 � (Asample/Acontrol) ⁄ 100 where Acontrol is the absorbance of the control reactioncontaining all reagents except the tested sample, and Asample is theabsorbance of the test compounds. Tests were carried out in triplicate and ablank with Tris–HCl buffer instead of enzyme solution was used.

46. (a) Ebrahimabadi, A. H.; Ebrahimabadi, E. H.; Djafari-Bidgoli, Z.; Kashi, F. J.;Mazoochi, A.; Batooli, H. J. Food Chem. 2010, 119, 452; (b) A methanolic solutionof DPPH (80 lL) was prepared daily to obtain a solution with absorbance at517 nm equal to 1.1. Then, to 1 mL of this solution were added methanolicsolutions of (1+2) at different concentrations. The resulted solution wasincubated at 37 �C for 30 min after what the absorbance of the solution wasmeasured at 517 nm. Lower absorbance of the reaction mixture indicateshigher free radical scavenging activity. Tests were carried out in triplicate.Decrease in absorption induced by the tested sample was compared to that ofthe positive control BHT. The capability to scavenge the DPPH� radical wascalculated using the following equation: inhibition ratio (DPPH� scavengingeffect) (%) = [(Acontrol � Asample)/Acontrol] � 100.

47. (a) Re, R.; Pellegrini, N.; Proteggente, A.; Pannala, A.; Yang, M.; Rice-Evans, C. J.Free Radic. Biol. Med. 1999, 26, 1231; (b) ABTS was dissolved in water to a 7 mMconcentration and the ABTS radical cation was produced by adding potassiumpersulfate to a final concentration of 2.45 mM. The completion of radicalgeneration was obtained in the dark at room temperature for 12 h. Thissolution was then diluted with methanol to adjust its absorbance at 734 nm to0.706 ± 0.009. To determine the scavenging activity, 1 mL of diluted ABTS+

solution was added to 1 mL of methanolic solutions of (1+2) at different

concentrations and the absorbance at 734 nm was measured 10 min after theinitial mixing, using methanol as the blank. The percentage of inhibition wascalculated by the equation: [(Acontrol � Asample)/Acontrol] � 100, where Acontrol isthe absorbance of the control reaction containing all reagents except the testedsample, and Asample is the absorbance of the test compound. Tests were carriedout in triplicate. BHT was used as the positive control. The IC50 was calculatedusing linear relation between the compound concentration and probability ofthe percentage of ABTS inhibition.

48. (a) Fontana, L.; Mosca, M.; Rosei, M. A. J. Biochem. Pharmacol. 2001, 61, 1253;(b) Superoxide radical is generated in phenazine methosulfate–nicotinamideadenine dinucleotide (PMS–NADH) systems by oxidation of NADH and assayedby the reduction of nitroblue tetrazolium (NBT) to a purple formazan. The 1 mLreaction mixture contained phosphate buffer (20 mM, pH 7.4), NADH (73 lM),NBT (50 lM), PMS (15 lM) and various concentrations of sample solution.After incubation for 2 min at ambient temperature, the absorbance at 562 nmwas measured against an appropriate blank to determine the quantity offormazan generated. The experiment was repeated thrice. The results werecompared with that of BHT. The % inhibition of superoxide anion generationwas calculated using the following formula: % Scavenging = [(Acontrol � Asample)/Acontrol] � 100.

49. (a) Ruch, R. J.; Cheng, S. J.; Klaunig, J. E. Carcinogenesis 1989, 10, 1003; (b) Asolution of hydrogen peroxide (43 mM) was prepared in buffer phosphate(0.1 M, pH 7.4). One hundred microliters of the solution to be tested are addedto a hydrogen peroxide. This solution contains 500 lL of NADH (73 lM), NBT(50 lM) 500 lL and 500 lL of PMS (15 lM). The free radical scavengingactivity is estimated according to the equation: percent scavenged[H2O2] = [(Acontrol � Asample)/Acontrol] � 100. The results were compared withthat of BHT. The experiment was repeated thrice.

50. Singh, R.; Singh, B.; Singh, S.; Kumar, N.; Kumar, S.; Arora, S. Food Chem. 2010,120, 830.

51. Li, X.; Lin, J.; Gao, Y.; Han, W.; Chen, D. Chem. Cent. J. 2012, 6, 140.