Bioguided fractionation and isolation of phytotoxic

0 downloads 0 Views 523KB Size Report
Sep 21, 2016 - The phytotoxic activity of celery (Apium graveolens L.) was tested on ...... by releasing water soluble phytoxins from leaves, stem, roots fruits.
South African Journal of Botany 108 (2017) 423–430

Contents lists available at ScienceDirect

South African Journal of Botany journal homepage: www.elsevier.com/locate/sajb

Bioguided fractionation and isolation of phytotoxic compounds from Apium graveolens L. aerial parts (Apiaceae) H. Sbai a,⁎, I. Zribi b, M. DellaGreca c, R. Haouala a a b c

Department of Biological Sciences and Plant Protection, Higher Institute of Agronomy of ChottMeriem, University of Sousse, 4042, UR13AGR05, Tunisia Higher Institute of Biotechnology of Monastir, University of Monastir, Tunisia Dipartimento di Scienze Chimiche, Università Federico II, Complesso Universitario Monte S. Angelo, via Cintia, 4, 80126 Napoli, Italy

a r t i c l e

i n f o

Article history: Received 24 December 2015 Received in revised form 11 July 2016 Accepted 13 September 2016 Available online 21 September 2016 Edited by Z-H Chen Keywords: Bioactive compounds Bioassay-directed fractionation Apium graveolens L. Phytotoxic activity

a b s t r a c t The phytotoxic activity of celery (Apium graveolens L.) was tested on germination and seedling growth of Lactuca sativa (lettuce). Bioactive compounds were isolated and identified by spectral data. Methanol extract of celery aerial part showed a significant phytotoxic effect on lettuce. Thus, this extract was divided into hexane and methanol fractions. Bioassay-directed of these two fractions led to the isolation and characterization of six secondary metabolites which included three phthalides (senkyunolide A, (3S)-butylphthalide and sedanolide), two furanocoumarins (bergapten and scopoletin) and one phenylpropanoid (p-hydroxyphenethyltransferulate). Among these compounds, senkyunolide A was the most toxic on lettuce germination and shoot growth. However, p-hydroxyphenethyl trans-ferulate was the most toxic on root growth. © 2016 SAAB. Published by Elsevier B.V. All rights reserved.

1. Introduction Risks and problems associated with the use of chemicals lead to increasingly stringent environmental regulation of pesticides (Pavela et al., 2010). There is therefore an urgent need to develop safer, more environmentally friendly and efficient alternatives that have the potential to replace synthetic pesticides and are convenient to use (Tapondjou et al., 2005). In this context, screening of natural products has received the attention of researchers around the world (Kebede et al., 2010). Numerous plants are reported to possess allelopathic potential and efforts have been made to apply them for weed control. A variety of allelochemicals are plant secondary metabolites for medicinal and aromatic plants (Delabays et al., 1998a, 1998b); have been identified, including the phenolic acids, coumarins, terpenoids, flavonoids, alkaloids, glycosides and glucosinolates. These substances are known to be exuded by plants to suppress emergence or growth of the other plants; they are submitted to biological and toxicological screens to identify their potential use as natural herbicide (Cespedes et al., 2006).

⁎ Corresponding author at: Higher Agronomic Institute of Chott-Mariem, BP 47, 4042 Chott-Mariem (Sousse), Tunisia. E-mail address: [email protected] (H. Sbai).

http://dx.doi.org/10.1016/j.sajb.2016.09.011 0254-6299/© 2016 SAAB. Published by Elsevier B.V. All rights reserved.

Celery (Apium graveolens L.; Apiaceae) has been cultivated for the last 3000 years, notably in pharaonic Egypt, and was known in China in the fifth century BC (Chevallier, 1998). It has been used as a popular aromatic herb and spice. This plant is grown for its stems and leaves which have culinary uses particularly in salads and soups (Kurobayashi et al., 2006). Numerous reports are describing the use in various forms of A. graveolens in traditional medicine of Morelos state to avoid the toothache and treat diarrhea, hypertension, broncho pulmonary, liver, asthma disease and used as a diuretic for bladder and kidney complaints(Castillo-España and Monroy-Ortiz, 2007). Also, it was implicated in arthritic pain relief, for treating rheumatic conditions and gout (Chevallier, 1998). Earlier studies of A. graveolens led to the isolation of phthalides (Tang et al., 1990; Momin and Nair, 2001) and furocoumarins (Garg et al., 1979). These compounds are reported for their insecticidal, nematicidal, antifungal and phytotoxic activities (Kato et al., 1977; Momin and Nair, 2001; Pavela and Vrchotová, 2013). Additionally, they have various physiological activities such as anti-inflammatory, anticarcinogenic and insecticidal effects which have recently been attracting interest (Zheng et al., 1993; Woods et al., 2001). Despite its interesting use as an aromatic and medicinal plant, few studies investigated the allelopathic effects of A. graveolens. The present work aimed to study the phytotoxicity of its aerial part (stems and leaves) on the sensitive plant Lactuca sativa L. under laboratory conditions and the identification of its bioactive substances through bioguided assays.

424

H. Sbai et al. / South African Journal of Botany 108 (2017) 423–430

2. Materiel and methods 2.1. Plant material

where extract: parameter measured in the presence of A. graveolens extract. control: parameter measured in the presence of distilled water.

Plants of celery (A. graveolens L.) were purchased from a local market in Sousse, Tunisia. Leaves and stems were dried for 4 days at room temperature, then, in hot-air oven at 70 °C for 72 h, powdered and used for extraction.

Values were expressed in percentage of control, this quantification method was calculated using the following formula reported by Wardle et al. (1996):

2.2. Aqueous extract

ðTreatment value=control valueÞ  100

50 g of dried materials were soaked in 1000 mL of distilled water at room temperature for 24 h. The extract was filtered through a filter paper several times and kept at 4 °C in the dark until use.

2.5. Bioassays with organic extracts

2.3. Organic extracts Sequential extraction was carried out with organic solvents with rising polarity: petroleum ether, chloroform and methanol. Dried powder (100 g) was immersed in organic solvent for 7 days at room temperature. Organic extracts were evaporated to dryness under reduced pressure at 45–50 °C using Rotavapor R-114 (Buchi, France). The residue was weighed and yield was determined. Residues were stored at 4 °C until use.

The residues concentrated from petroleum ether, chloroform and methanol were dissolved in methanol and three concentrations (1, 3 and 6 mg/mL) were prepared. To eliminate the effect of organic solvent, two controls (distilled water and methanol) were used. Filter paper placed in Petri dish was wetted with distilled water, methanol or various organic extracts. Solvents were evaporated at 24 °C for 24 h, then, 5 mL distilled water was added and 20 soaked seeds were placed to germinate for seven days. Germination, shoot and root length of lettuce were estimated as per aqueous extracts above. Treatments were arranged in a completely randomized design with three replications and data were transformed to percent of control for analysis.

2.4. Bioassays with aqueous extracts 2.6. Isolation and identification of bioactive compounds Aqueous extract was diluted with distilled-water to give final concentrations between 10 and 50 g/L. They were tested on lettuce. Seeds were surface sterilized with 0.525 g/L sodium hypochlorite for 15 min and then rinsed four times with deionized water and carefully blotted using a folded paper towel (Chon et al., 2005). For germination assay, twenty imbibed seeds of lettuce were placed in 9 cm plastic Petri dishes, lined with filter paper and 5 mL of each extract were added. Seeds watered with distilled water were used as control. In the growth assay, pre-germinated seeds with 1 mm root length were used. The Petri dishes were then placed in a growth chamber (24 °C/22 °C Day/Night and 14 h/10 h light/dark period). The dishes were illuminated with 400 μmol photons m2 s1 photosynthetically active radiation (PAR) (Haouala et al., 2008). Treatments were arranged in a completely randomized design with three replications. Germinated seeds were counted daily during seven days. Shoot and root length of lettuce seedlings was measured at the 7th day after sowing. Data were transformed to percent of control for analysis. The index of germination (GI) and total germination (GT) were calculated using the following formula reported by Chiapuso et al. (1997): GI ¼ ðN1 Þ  1 þ ðN2 ‐N1 Þ  1=2 þ ðN3 ‐N2 Þ  1=3 þ :::: þ ðNn ‐Nn‐1 Þ  1=n Where, N1, N2, N3, …., Nn: proportion of germinated seeds observed afterwards 1, 2, 3, …., n-1, n days. This index represents the delay in germination induced by extract (Delabays et al., 1998a, 1998b). GT ¼ ðNT 100Þ=N NT: proportion of germinated seeds at each treatment for the last time measurement. N: Number of seeds used in the bioassay. The inhibitory/stimulatory percentage was calculated using the equation given by Chung et al. (2001) (modified): Inhibitionð−Þ=stimulationðþÞ% ¼ ½ðextract–controlÞ=control  100

2.6.1. General experimental procedures Nuclear magnetic resonance (NMR) spectra have been recorded at 500 MHz for 1H and 125 MHz for 13C in CDCl3 at 25 °C. Electronic Impact Mass Spectra (EI-MS) have been obtained with a GC–MS QP5050A (Shimadzu) equipped with a 70 eV EI detector. Flash column chromatography was conducted on Kieselgel 60, 230–400 mesh (Merck), at medium pressure. Column chromatography was performed on Kieselgel 60, 70–240 mesh (Merck), or on Sephadex LH-20 (Pharmacia). Analytical TLC was made on Kieselgel 60 F254 or RP-18 F254 plates with 0.2 mm layer thickness (Merck). Spots were visualized by UV light or by spraying with EtOH:H2SO4 (93:7) and heated for 5 min at 110 °C. Preparative TLC was performed on Merck Kieselgel 60 F254 plates with 0.5 or 1 mm film thickness (Merck). 2.6.2. Extraction and isolation Dried and powdered leaves and stems of A. graveolens were extracted twice with methanol at room temperature. Methanol extract (233 g) was dissolved in hexane. The hexane soluble part (40 g) was subjected to normal phase silica gel column chromatography (cc) using a gradient of petroleum ether (EP) and acetone (1:0, 95:5, 9:1, 8:2, 6:4, v/v). According to TLC analysis, collected fractions were combined in six homogenous fractions (CH1-CH6) which were tested for their phytotoxic activity. The most bioactive fraction (CH1) was separated by Sephadex LH-20 and eluted with EP, CH2Cl2, MeOH (5:2:1) to obtain 11 subfractions which were combined according to TLC analysis. Some fractions were selected according to their weight and their chromatogram profile then purified by silica gel column chromatography using a mixture of EP, EtOAc (1:0; 95:5; 9:1; 8:2; 7:3; v/v) as mobile phase to yield sedanolide (3), senkyunolide A (2) and (3S)-butylphthalide (1). The insoluble hexane part was dissolved in methanol. Then, the soluble methanol extract (100 g) was subjected to normal phase silica gel column chromatography (cc) using a gradient of CH2Cl2, EtOAc (1:0; 9:1; 7:3,v/v); CH2Cl2, MeOH (9:1; 8:2; 6:4; 0:1) to afford several fractions which, based on TLC analysis, were combined to give three fractions. According to phytotoxic test, only two fractions were the most toxic. Both of them were separated by Sephadex LH-20 using hexane, CH2Cl2, MeOH (3:1:1). Combined fractions, according to TLC analysis, were purified by preparative TLC using a mixture of solvents

H. Sbai et al. / South African Journal of Botany 108 (2017) 423–430

425

CH2Cl2, EtOAc (7:3); EP, EtOAc (7: 3) and (9:1) to give bergapten(5), scopoletin(6) and p-hydroxyphenethyltrans-ferulate(4). The compounds were identified by comparing their alpha-D and NMR data with those of published data (Macías et al., 1990; Momin et al., 2000; Darwish and Reinecke, 2003; Begum et al., 2006; Chae et al., 2011; Oguro and Watanabe, 2011). 2.6.3. Phytotoxic assay Lettuce seeds were used to test germination and growth response. The stock solution of the purified compound was prepared by dissolving it in methanol to obtain one concentration of each compound (0.06 g L−1). Treatments were arranged in a completely randomized design with three replications and germination and growth parameters were calculated as described above. 2.7. Statistical analysis The laboratory bioassays were done in complete randomized design with three replications. In all assays, SPSS software version 20 was used for statistical analysis. Analysis of variance (ANOVA) followed by Duncan test was used to evaluate differences among means. The significance level was set at P b 0.05. 3. Results 3.1. Effect of A. graveolens aqueous extract 3.1.1. Germination The A. graveolens aqueous extract exhibited strong inhibitory effect on lettuce germination and the magnitude of inhibition increased with the concentration. Till 20 g/L, the aqueous extract didn't affect the final germination percentage but it was strongly delayed the germination and GI reached to 34.67%. However, from 30 g/L, a spectacular inhibitory effect was recorded on lettuce germination indicated by a total inhibition (Table 1). 3.1.2. Seedling growth The aqueous extracts of A. graveolens were toxic at all concentrations, especially to roots which were more sensitive than shoots. Root growth was inhibited in presence of all concentrations and the inhibition percentage varied from 79.67% at the lowest concentration to 97.26% at the highest one. The extract improved the shoot growth by inducing an average stimulation of 43.65% at the two lowest concentrations. However, for extract above 20 g/L, aqueous extract showed a toxicity provoking an average inhibition of 67% for shoot growth (Fig. 1).

Fig. 1. Inhibition/stimulation (% of control) of lettuce roots and shoots length seven days after germination in the presence of different concentrations of A. graveolens aerial parts aqueous extracts at different concentrations. The bars on each column show standard error (S.E.). Values (N = 3 ± S.E.). Different letters in columns indicate significant differences among treatments at P b 0.05.

3.2.2. Germination Results showed that methanol, already used to dissolve residues, had no effect on germination, as a control. This allows us to say that the effects, which would be observed, could be attributed to substances dissolved in this solvent. Petroleum ether and chloroform extracts generally had no significant effect on seed germination. Only a slight delay of GI was recorded in presence of the first one. Nevertheless, methanol extract was very toxic, it induced a remarkable reduction of GI at all concentrations. The GI varied from 60.55% at 1000 ppm to 0.72% at 6000 ppm. For the germination percentage, it reached 1.75% (Table 3).

3.2.3. Seedling growth An inhibition of root growth was recorded in presence of all extracts and roots were more sensitive than shoots. Petroleum ether showed toxicity for both root and shoot length but it was concentration independent. Also, chloroform extract inhibited significantly root growth and registered an average inhibition of 47.56%. However, in the presence of the three concentrations of chloroform, an average stimulation of 34.82% was recorded for shoot length. Moreover, the methanol extract was proved to be the most toxic for shoot and root growth. At the highest concentration, the inhibition of root length reached 81.21% (Fig. 2).

3.2. Effect of A. graveolens organic extracts

3.3. Isolation and identification of bioactive compounds

3.2.1. Yield of organic extracts Yields of organic extracts are given in Table 2. The methanol extract gave the highest yield (11.19%) followed by chloroform extract (4.9%) and petroleum ether extract (1%).

Bioassays of A. graveolens aqueous and organic extracts revealed that methanol extract was the most toxic. Thus, it was chosen for further chemical studies. Methanol extract was dissolved in hexane and the hexane and methanol soluble part were chromatographed on silica gel column chromatography and the fractions were purified by preparative thin-layer chromatography yielding pure compounds.

Table 1 Germination index (GI) and total germination (GT), expressed in percent of control, of lettuce in presence of A. graveolens L. aqueous extracts at different concentrations. Aqueous extracts concentrations (g/L)

GI (% of control)

GT (% of control)

10 20 30 40 50

49.48b ± 3.45 34.67b ± 4.43 0a ± 0.0 0a ± 0.0 0a ± 0.0

96.66c ± 2.88 98.33c ± 2.88 0a ± 0.0 0a ± 0.0 0a ± 0.0

Values are given as means of three replicates ± standard error. Means with the same letters in a column are not significantly different at P b 0.05.

Table 2 Yields, in percent of dry matter, of organic extracts of A. graveolens L. aerial parts. Organic extracts

Aerial parts (% dm)

Petroleum ether Chloroform Methanol

1 4.9 11.19

426

H. Sbai et al. / South African Journal of Botany 108 (2017) 423–430

Table 3 Germination index (GI) and total germination (GT) (expressed in % of control) of lettuce in the presence of organic extracts of A. graveolens L. aerial parts. Organic extracts (g/L)

GI (% of control)

GT (% of control)

Petroleum ether 1 3 6

110.01a ± 15.5 102.33a ± 12.21 93.37a ± 8.69

103.50a ± 3.03 103.50a ± 3.03 103.50a ± 3.03

Chloroform 1 3 6

134.72a ± 13.67 122.86a ± 14.54 109.16a ± 14.81

103.50a ± 3.03 100.00a ± 5.26 103.50a ± 3.03

Methanol 1 3 6

60.55c ± 9.76 20.05b ± 7.60 0.72a ± 1.25

86.40c ± 12.64 34.38b ± 16.52 1.75a ± 3.03

Values are given as means of three replicates ± standard error. Means with the same letters in a column are not significantly different at P b 0.05.

3.3.1. Phytotoxicity of hexane fractions The hexane fractions induced a significant effect on lettuce germination. All fractions delayed germination index, it varied from 18.44% to 90.77% and GT varied between 47.63% and 100.08%. The first fraction was the most toxic among the tested ones. It delayed significantly germination index (Table 4). In growth assay, shoots were more sensitive than roots. CH1 was the most toxic, inducing an inhibition of 22.92% and 31.99% in root and shoot growth respectively. However, fractions CH4 and CH5 induced a slight stimulation in root growth. An average inhibition of 11.62% for roots was registered in the presence of the other fractions. Shoot growth inhibition was recorded in the presence of all tested fractions except the fifth one (Fig. 3).

3.3.2. Biological activity of methanol fractions Germination assay showed that CH8 and CH9 were the most phytotoxic. They delayed significantly the germination and GI reached 2.42% and 5.88% with CH9 and CH8 respectively (Table 5). Roots were more sensitive than shoots. CH7 and CH9 induced an important phytotoxic effect with respective average inhibition of 57.63 and

Fig. 2. Inhibition (% of control) of lettuce roots and shoots length seven days after germination in the presence of different concentrations (1, 3 and 6 g/L) of A. graveolens aerial parts organic extracts. The bars on each column show standard error (S.E.). Values (N = 3 ± S.E.). Different letters in columns indicate significant differences among treatments at P b 0.05.

H. Sbai et al. / South African Journal of Botany 108 (2017) 423–430 Table 4 Germination index (GI) and total germination (GT) (expressed in % of control) of lettuce in the presence of A. graveolens hexane fractions at 0.6 g/L.

427

Table 5 Germination index (GI) and total germination (GT) (expressed in % of control) of lettuce in the presence of A. graveolens methanol fractions at 0.6 g/L.

Fractions

GI (% of control)

GT (% of control)

Fractions

GI (% of control)

GT (% of control)

CH1 CH2 CH3 CH4 CH5 CH6

18.44a ± 4.59 45.31b ± 5.63 50.93b ± 2.39 90.77e ± 8.73 79.41d ± 3.54 68.62c ± 4.36

47.63a ± 9.23 84.91b ± 9.87 89.82bc ± 0.30 98.33c ± 2.88 100.08c ± 5.13 95.00bc ± 5.00

CH7 CH8 CH9

85.08b ± 6.73 5.88a ± 4.62 2.42a ± 1.41

101.75b ± 3.03 16.93a ± 12.53 6.75a ± 2.81

Values are given as means of three replicates ± standard error. Means with the same letters in a column are not significantly different at P b 0.05.

56.06% for root growth. However, CH7 and CH8 are shown to be more phytotoxic for shoot growth. They induced respectively 24.14% and 45.34% of inhibition (Fig. 4). 3.4. Identification of inhibitory compounds According to the obtained results, Fractions CH1, CH8 and CH9 were selected for further analysis to purify the active substances. 3.4.1. Chemical structure of identified compounds Compound 1: EI-MS: m/z190 [M]+, 133 [M-C4H9]+. The 1H NMR (400 MHz, CDCl3) δ 7.92 (1H, d, J = 7.6 Hz, H-7), 7.69 (1H, t, J = 7.5 Hz, H-5), 7.54 (1H, t, J = 7.5 Hz, H-6), 7.46 (1H, d, J = 7.6 Hz, H4), 5.50 (1H, dd, J = 7.8, 4.1 Hz, H-3), 2.14–2.01 (1H, m), 1.84–1.73 (m, 1H), 1.54–1.33 (m, 4H), 0.94 (3H, t, J = 6.3 Hz, H-4′). The 13C NMR (101 MHz, CDCl3) δ 170.0 (C-1), 150.1 (C-3a), 135.1 (C-8), 133.5 (C-5), 128.8 (C-7), 126.0 (C-4), 121.9 (C-6), 81.4 (C-3), 34.6 (C-1′), 26.6 (C-2′), 22.4 C-3′), 14.5 (C-4′).[α]D − 54 (c = 0.6, CH2Cl2). From the comparison of these data with those reported by Chae et al. (2011), the compound was identified as (3S)-butylphthalide, IUPAC name: (S)-3-butylisobenzofuran-1(3H)-one (Fig. 5). Compound 2: EI-MS: m/z192 [M]+, 135[M-OC4H9]+. The 1H NMR (400 MHz, CDCl3) δ6.09 (1H, d, J = 9.7 Hz, H-7),5.84 (1H, dd, J = 8.8, 4.9 Hz, H-6), 4.87 (1H, dd, J = 7.1, 3.2 Hz, H-3), 2.52–2.32 (4H, m), 1.87–1.76 (1H, m), 1.46 (1H, td, J = 14.2, 7.5 Hz), 1.37–1.23 (4H, m), (3H, s, H-4′). The 13C NMR (125 MHz, CD3OD), (C-2), 158.4 (C-7), 152.8 (C-9), 150.2 (C-5), 144.8 (C-2′), 139.1(C-4), 112.7 (C-6), 112.5 (C-3), 106.4 (C-10), 105.0 (C-3′), 93.4 (C-8), 60.1 (OCH3). [α]D − 70 (c = 1.2, CH2Cl2). From the comparison of these data with those reported in the literature (Momin et al., 2000), the compound was identified as senkyunolide A, IUPAC name: (S)-3-butyl-4, 5 dihydroisobenzofuran-1(3H)-one (Fig. 5).

Fig. 3. Inhibition (% of control) of lettuce roots and shoots length seven days after germination in the presence of A. graveolens aerial parts hexane fractions at 0.6 g/L. The bars on each column show standard error (S.E.). Values (N = 3 ± S.E.). Different letters on columns indicate significant differences among dichloromethane fractions at P b 0.05.

Values are given as means of three replicates ± standard error. Means with the same letters in a column are not significantly different at P b 0.05.

Compound 3: EI-MS: m/z194 [M]+, 137 [M-C4H9]+. The1H NMR (400 MHz, CDCl3) δ 6.75 (1H, m, H-7), 3.94 (1H, m, H-3), 2.47(1H, m, H-3a), 2.31 (1H, m), 2.18 (1H, m),2.02 (1H, m), 1.91 (2H, m), 1.72 (3H, m) 1.57–1.28 (4H, m), 0.90 (3H, t, J = 7.0 Hz, H-4′). The 13C NMR (101 MHz, CDCl3) δ 170.2 (C-1), 135.2 (C-7), 131.1 (C8), 85.4 (C-3), 43.0 (C-3a), 34.3 (C-6), 27.5 (C-1′), 25.3 (C-4), 24.9 (C-5), 22.4 (C-2′), 20.7 (C-3′), 13.8 (C-4′). [α]D − 66 (c = 1.1, CH2Cl2). From the comparison of these data with those reported in the literature (Oguro and Watanabe, 2011) the compound was identified as sedanolide, IUPAC name: (3S,3aR)-3-butyl-3a,4,5,6-tetrahydroisobenzofuran1(3H)-one (Fig. 5). Compound 4: EI-MS: m/z 314 [M]+, 297 [M-OH]+. The 1H NMR (500 MHz, CDCl3) δ7.62 (1H, d, J = 16.0 Hz, H-7), 7.15 (2H, d, J = 8.4 Hz, H-2′ and H-6′), 7.09 (2H, dd, J = 8.2, 0.8,Hz, H-6), 7.05 (1H, d, J = 1.0 Hz, H-2), 6.94 (1H, d, J = 8.2 Hz, H-5), 6.81 (2H, d, J = 8.4 Hz, H-3′ and H-5′), 3.93 (3H, s, OCH3). The 13C NMR (125 MHz, CDCl3), (C-9), 154.3 (C-4′), 147.2 (C-3), 145.3 (C-7), 144.2 (C-4), 131.0 (C-1′), 130.5 (C-2′ and C-6′), 127.5 (C-1), 123.5 (C-6), 115.8 (C-8), 115.3 (C3′ and C-5′), 114.6 (C-5), 109.6 (C-2), 65.2 (C-8′), 56.4 (OCH3), 34.7 (C7′). From the comparison of these data with those reported in the literature (Macías et al., 1990), the compound was identified as bergapten (Fig. 5). Compound 5: EI-MS: m/z 216 [M]+, 185 [M-OCH3]+. The 1H NMR (500 MHz, CDCl3) δ 8.15 (1H, d, J = 9.8 Hz, H-4), 7.59 (1H, d, J = 2.3 Hz, H-2′), 7.13 (1H, s, H-8), 7.02 (1H, d, J = 2.3 Hz, H-3′), 6.27 (1H, d, J = 9.8 Hz, H-3), 4.17 (3H, s, OCH3). The 13C NMR (125 MHz, CDCl3), (C-2), 158.4 (C-7), 152.8 (C-9), 150.2 (C-5), 144.8 (C-2′), 139.1 (C-4), 112.7 (C-6), 112.5 (C-3), 106.4 (C-10), 105.0 (C-3′), 93.4 (C-8), 60.1 (OCH3). From the comparison of these data with those reported in the literature (Begum et al., 2006), the compound was identified as scopoletin (Fig. 5). Compound 6: EI-MS: m/z192 [M]+, 161 [M-OCH3]+. The 1H NMR (500 MHz, CDCl3) δ7.62 (1H, d, J = 9.5 Hz, H-4), 6.94 (1H, s, H-5), 6.87 (1H, s, H-8), 6.29 (1H, d, J = 9.5 Hz, H-3), 3.98 (3H, s, OCH3). The

Fig. 4. Inhibition (% of control) of lettuce roots and shoots length seven days after germination in the presence of A. graveolens aerial parts methanol fractions at 0.6 g/L. The bars on each column show standard error (S.E.). Values (N = 3 ± S.E.). Different letters on columns indicate significant differences among dichloromethane fractions at P b 0.05.

428

H. Sbai et al. / South African Journal of Botany 108 (2017) 423–430

O O

O

O

CH2CH2CH2CH3

CH2CH2CH2CH3

(S)-3-butyl-4,5-dihydroisobenzofuran-1(3H)-one

(S)-3-butylisobenzofuran-1(3H)-one

Compound 1: (3S)-Butylphthalide

Compound 2: Senkyunolide A

O 7

3'

1

O

O 3a 5

H3CO 2

3

1

9 8

1'

O

7'

CH2CH2CH2CH3

(3S,3aR)-3-butyl-3a,4,5,6-tetrahydroisobenzofuran1(3H)-one

Compound 3: Sedanolide

HO

5

Compound 4: p-hydroxyphenethyltrans-ferulate

5

4

H3CO

OCH3

O

4' 8'

4'

H

OH

3

HO O

O

Compound 5: Bergapten

O

O

8 Compound 6: Scopoletin

Fig. 5. Chemical structures of identified A. graveolens L. aerial parts' compounds.

13 C NMR (125 MHz, CDCl3), (C-2), 150.8(C-9), 144.3 (C-6), 143.6(C-7), 143.0(C-4), 113.4 (C-3), 111.5 (C-10), 107.5 (C-5), 103.2 (C-8), 53.4 (OCH3). From the comparison of these data with those reported by Darwish and Reinecke (2003), the compound was identified as phydroxyphenethyl trans-ferulate (Fig. 5).

3.4.2. Phytotoxicity of purified compounds Isolated compounds were tested for their phytotoxicity on lettuce at 0.06 g/L (Table 6). All molecules delayed significantly germination index. Senkyunolide A was the most toxic among the phthalides; it stopped completely the lettuce germination. Germination was more

Table 6 Germination index (GI), total germination (GT), root and shoot length of lettuce in the presence of purified compounds from A. graveolens aerial parts at 0.06 g/L. Compounds

GI (% of control)

GT (% of control)

Root length (% of control)

Shoot length (% of control)

(3S)-Butylphthalide (1) SenkyunolideA(2) Sedanolide (3) p-hydroxyphenethyltrans-ferulate(4) Bergapten (5) Scopoletin(6)

9.52ab ± 0,90 0a ± 0.00 42.31bc ± 12.38 61.55c ± 34.57 78.13c ± 18.20 60.95c ± 21.84

21.79b ± 6,60 0a ± 0.00 78.20d ± 6.6 56.41c ± 13.20 63.55cd ± 11.78 58.79c ± 14.01

6.83a ± 20.67 31.99a ± 32,07 28.70a ± 33.82 −10.09a ± 30.96 −5.84a ± 3.95 22.84a ± 22.35

−43.73ab ± 6.06 −49.01a ± 2,90 −33.05b ± 6.51 −10.30c ± 7.16 −5.92c ± 4.33 −2.71c ± 10.35

Values are given as means of three replicates ± standard error. Means with the same letters in a column are not significantly different at P b 0.05.

H. Sbai et al. / South African Journal of Botany 108 (2017) 423–430

sensitive to pthalides than furanoucoumarins and phenylpropanoids (Table 6). For growth, roots were less sensitive than shoots. Root inhibition was recorded only in the presence of bergapten and phydroxyphenethyl trans-ferulate. Shoots growth was more sensitive in presence of phthalides, the most toxic compound (senkyunolide A) induced 49% of shoot length reduction, followed by (3S)butylphthalide (43.73%) and sedanolide (33%). However, about only 6.31%, an average of shoot inhibition was registered for the other molecules (Table 6). 4. Discussion The current study revealed a potent phytotoxic effect for aerial part of A. graveolens on lettuce germination and growth. Our results suggested that seed germination was completely inhibited by extract above 20 g/L, also shoot inhibition was recorded. However, bellow this concentration, an increasing of shoot length was registered. Additionally, aqueous extract, at all concentrations, exhibited inhibitory effect on radicle length. Roots appeared more sensitive to extracts than shoots. The plant aqueous extracts effect on germination and growth and the difference between organs sensitivity are reported in the literature. Hence, Kil et al. (2002) recorded an inhibitory effect of Tagetes minuta extracts on germination and growth of Lactuca sativa. Al-Saadawi et al. (1986) reported that water extracts of different cultivars of sorghum significantly reduced the germination of redroot pigweed (Amaranthus retroflexus). Also, El-Darier and Youssef (2000) stated that there was an increase in plumule growth rate of Lepidium sativum till 50% alfalfa extract after five days of experiment. These results are also in agreement with the finding that water extract of allelopathic plants generally have more pronounced effect on radicle rather than hypocotyl growth (Inderjit and Dakshini, 1995; Muhammad et al., 2011). This may be attributable to the fact that radical have direct contact with soil and can absorb many allelochemicals. Plants are known to exhibit allelopathy by releasing water soluble phytoxins from leaves, stem, roots fruits and seeds and such metabolites play an inhibitory role in delay or complete inhibition of seed germination, stunted growth and injury to root systems of plants (Rice, 1984). Thus, the considerable inhibition of seed germination may be due to the inhibitory effect of allelochemicals such as water soluble saponins, hormones or enzyme which could affect growth directly or by altering the mobilization of storage compounds during germination (Chaves and Escudero, 1997; El-Khatib, 1997). Also, they can affect different physiological processes through their effects on enzymes responsible for plant hormone synthesis and were found to associate with inhibition of nutrients and ion absorption by affecting plasma membrane permeability (Qasem and Foy, 2001; Qasem and Rand Hassan, 2003). A fractional extraction in three organic solvents with increasing polarity aimed to identify the chemical groups of bioactive substances of A. graveloens. This study showed that the degree of seed germination was enhanced in presence of petroleum ether and chloroform extracts, however, it was affected in the presence of methanolic extract, especially at higher concentration. Lettuce radicle length was inhibited in the presence of all organic extracts. However, plumule length was markedly stimulated in presence of chloroform extract but more sensitive to petroleum ether and methanol extracts. This variant response to the allelopathic substance could be explained through that allelopathic effect is concentration and organic dependent. Growth bioassays are often more sensitive than germination bioassays (Bhowmik and Doll, 1984). Although, Fuentes et al. (2004) observed that seed germination has been regarded as a less sensitive method than plumule and radical length when they used as a bioassay for phytotoxicity evaluation. This variability in phytotoxicity could be attributed to the change in the chemical composition of different organic extracts and/or to the combined action of their chemicals, indeed, several combinations of allelochemicals have either synergistic or antagonist effects (Einhellig, 2004).

429

The investigation of allelopathic compounds is just a small part of the studies of natural compounds. Nature still represents an infinite source of unknown molecules having new structures and functionalities that evolution refined for high specific biological roles. Hence, the present study aimed to isolate different chemical constituents from A. graveolens methanol extract. The high toxicity of this extract may be attributed to the presence of a mixture of chemical substances acting together. The hexane and methanol fractions exhibited an important inhibition on lettuce germination and seedling growth. The phytotoxic effect might be related to specific compounds being produced in larger quantities in certain fraction imparting a higher level of allelopathy. Compounds responsible for this toxicity were identified as phthalides in particular sedanolide, senkyunolide A and (3S)-butylphthalide. This group of substances was previously isolated from celery (Tang et al., 1990; Momin et al., 2000; Sowbhagya et al., 2010). Phthalides have been found to exhibit interesting biological activities such as inhibition of the growth and toxin production of mycotoxin producing fungi (Morozumi et al., 1984), chemoprevention of cancer (Zheng et al., 1993), also they showed nematicidal (Momin et al., 2000), antifungal (Momin and Nair, 2001), antimicrobial (Mišić et al., 2008), mosquitocidal (Momin and Nair, 2001), insecticidal (Chae et al., 2011), acaricidal (Kwon and Ahn, 2002) and herbicidal potentialities (Purohit, 2001). Analogous works reported linear coumarins (bergapten and scopoletin) from celery (Beier et al., 1983; Heath-Pagliuso et al., 1992). Scopoletin was shown to inhibit Chinese cabbage root growth at 10−5 and 10 −7 M (Shimomura et al., 1982). Also, Bergapten, isolated from the leaves of Pilocarpus goudotianus, inhibited lettuce germination and seedling growth (Macías et al., 1993). The above coumarins have also been found to inhibit multiplication of bacteria, fungi and viruses (Kofinas et al., 1998), also demonstrated anti-allergy (Kimura and Okuda, 1997), anti-inflammation (Chen et al., 1995) and immune suppression activities (Kuzel et al., 1992) and herbicidal activity (Del corral et al., 2012). ρ-hydroxyphenethyl transferulate was previously isolated from Oenanthe javanica Blume (Fujita et al., 1995). 5. Conclusion The present study has led to the identification of some bioactive molecules in the aerial part of A. graveolens showing a phytotoxic activity. These natural products showed enhancement and inhibition effects on different pre-emergence properties (e.g., germination root and shoot length) at low and higher concentrations, respectively. The most active compound on seed germination and shoot growth was elucidated as senkyunolide A while ρ-hydroxyphenethyl trans-ferulate was the most toxic on root growth. These compounds might contribute to the complicated process of discovery and definition of natural substances which attract much interest due to their broad range of biological activities. Acknowledgments The authors are grateful to the Tunisian Ministry of Higher Education and Scientific Research for disbursing a research grant to Dr. Haifa Sbai for a period of training in Prof. Marina Della Greca's laboratory (Department of Chemical Sciences, University of degli Studi di Napoli Federico II, Italy). References Al-Saadawi, I.S., Al-Uqaili, J.K., Alrubeaa, A.J., Al-Hadithy, S.M., 1986. Allelopathic suppression of weeds and nitrification by selected cultivars of Sorghum bicolor (L.) Moench. Journal of Chemical Ecology 12, 209–219. Begum, T., Rahman, M.S., Rashid, M.A., 2006. Phytochemical and biological investigations of Phyllanthus reticulates. Dhaka University Journal Pharmaceutical Sciences 5, 21–23. Beier, R.C., Ivie, G.W., Oertli, E.H., Holt, D.L., 1983. Food and Chemical Toxicology 21, 163. Bhowmik, P.C., Doll, J.D., 1984. Allelopathic effects of annual weed residues on growth and nutrient uptake of corn and soybeans. Agronomy Journal 76, 383–388.

430

H. Sbai et al. / South African Journal of Botany 108 (2017) 423–430

Castillo-España, P., Monroy-Ortiz, C., 2007. Medicinal Plants Used in Morelos, Mexico. Morelos State University, Cuernavaca, p. 58. Cespedes, C.L., Marin, J.C., Dominguez, M., Avila, J.G., Serrato, B., 2006. Plant growth inhibitory activities by secondary metabolites isolated from Latin American flora. Advances in Phytomedicine 2, 373–410. Chae, S.H., Kim, S., Yeon, S.H., Lee, S.W., Ahn, Y.J., 2011. Adulticidal activity of phthalides identified in Cnidium officinale rhizome to B-and Q-biotypes of Bemisiatabaci. Journal of Agricultural and Food Chemistry 59, 8193–8198. Chaves, N., Escudero, J.C., 1997. Allelopathic effect of Cistrus ladanifer on seed germination. Functional Ecology 11, 432–440. Chen, Y.F., Tsai, H.Y., Wu, T.S., 1995. Anti-inflammatory and analgesic activities from roots of Angelica pubescens. Planta Medica 61, 2–8. Chevallier, A., 1998. The Encyclopedia of Medicinal Plants. DK Publishing Inc., 95 Madison Ave., New York, p. 61. Chiapuso, G., Sanchez, A.M., Reigosa, M.J., Gonzalez, L., Pellissier, F., 1997. Do germination indices adequately reflect allelochemical effects on the germination process? Journal of Chemical Ecology 23, 2445–24530. Chon, S.U., Jang, H.G., Kim, D.K., Kim, Y.M., Boo, H.O., Kim, Y.J., 2005. Allelopathic potential in lettuce (Lactuca sativa L.) plants. Scientia Horticulturae 106, 309–317. Chung, I.M., Ahn, J.K., Yun, S.J., 2001. Assessment of allelopathic potential of barnyard grass (Echinochloa crus-galli) on rice (Oryza sativa L.) cultivars. Crop Protection 20, 921–928. Darwish, F.M.M., Reinecke, M.G., 2003. Ecdysteroids and other constituents from Sida spinosa L. Phytochemistry 62, 1179–1184. Del corral, S., Cuffini, S.L., Cardoso, S.G., Bortoluzzid, A.J., Palacios, S.M., 2012. Phytotoxic halimanes isolated from Baccharis salicifolia (Ruiz & Pad.) Pers. Phytochemistry Letters 5, 280–283. Delabays, A., AnCay, A., Mermillod, G., 1998a. Recherche d`espèces végétales à propriétés allélopathiques. Annales de la 17e Conférence du COLUMA, 9, 10, 11, Dijon (sous presse). Delabays, N., Ancay, A., Mermillod, G., 1998b. Recherche d'espèces végétales à propriétés allélopathiques. Revue Suisse de Viticulture, d'Arboriculture et d'Horticulture 30, 383–387. Einhellig, F.A., 2004. Mode of allelochemical action of phenolic compounds. In: Macias, F.A. (Ed.), Allelopathy. CRC, Boca Raton, FL, pp. 217–238. El-Darier, S.M., Youssef, R.S., 2000. Effect of soil type, salinity, and allelochemicals on germination and seedling growth of a medicinal plant Lepidium sativum L. Annals of Applied Biology 136, 3179–3305. El-Khatib, A.A., 1997. Does allelopathy involve in the association pattern of Trifolium resupinatum. Biologia Plantarum 40, 425–431. Fuentes, A., LLorens, M., Saez, J., Aguilar, M.I., Otuno, I.F., Meseguer, V.F., 2004. Phytotoxicity and heavy metals speciation of stabilized sewage sludges. Journal of Hazardous Materials 108, 161–169. Fujita, T., Kadoya, Y., Aota, H., Nakayama, M., 1995. A new phenylpropanoid glucoside and other constituents of Oenanthe javanica. Bioscience Biotechnology & Biochemistry 59, 526–528. Garg, S.K., Gupta, S.R., Sharma, N.D., 1979. Coumarins from Apium graveolens seed. Phytochemistry 18, 1580–1581. Haouala, R., Khanfir, R., Tarchoune, A., Hawala, S., Beji, M., 2008. Allelopathic potential of Trigonella foenum-graecum L. Allelopathy Journal 21, 307–316. Heath-Pagliuso, S., Matlin, S.A., Fang, N., Thompson, R.H., 1992. Stimulation of furanocoumarin accumulation in celery and celeriac tissues by Fusarium oxysporum F.SP.APII. Phytochemistry 31, 2683–2688. Inderjit, Dakshini, K.M., 1995. Allelopathic potential of an annual weed, Polygonummons peliensis, in crops in India. Plant and Soil 173, 251–257. Kato, T., Kobayashi, M., Sasaki, N., Kitahara, Y., Takahashi, N., 1977. The coumarin heraclenol as a growth inhibitor in parsley seeds. Phytochemistry 17, 158–159. Kebede, Y., Gebre-Michael, T., Balkew, M., 2010. Laboratory and field evaluation of neem (Azadirachta indica A. Juss) and Chinaberry (Melia azedarach L.) oils as repellents against Phlebotomus orientalis and P. bergeroti (Diptera: Psychodidae) in Ethiopia. Acta Tropica 113, 145–150. Kil, J.H., Shim, K.C., Lee, K.J., 2002. Allelopathy of Tagetes minuta L. aqueous extracts on seed germination and root hair growth. Korean Journal of Ecology and Science 1, 171–174. Kimura, Y., Okuda, H., 1997. Histamine-release effectors from Angelica dahurica var. dahurica root. Journal of Natural Products 60, 249–251. Kofinas, C., Chinou, I., Loukis, A., Harvala, C., Roussakis, C., Maillard, M., 1998. Cytotoxic coumarins from the aerial parts of Tordylium apulum and their effects on a nonsmall-cell bronchial carcinoma line. Planta Medica 64, 174–176.

Kurobayashi, Y., Kouno, E., Fujita, A., Morimitsu, Y., Kubota, K., 2006. Potent odorants characterize the aroma quality of leaves and stalks in row and boiled celery. Bioscience, Biotechnology, and Biochemistry 70, 958–965. Kuzel, T.M., Roenigk, H.H., Samuelson, E., Rosen, S.T., 1992. Suppression of anti-interferon alpha-2a antibody formation in patients with mycosis fungoides by exposure to longwave UV radiation in the A range and methosalen ingestion. Journal of the National Cancer Institute 84, 119–121. Kwon, J.H., Ahn, Y.J., 2002. Acaricidal activity of Cnidium officinale rhizome-derived butylidenephthalide against Tyrophagus putrescentiae (Acari: Acaridae). Pest Management Science 59, 119–123. Macías, F.A., Massanet, G.M., Rodriguez-Luis, F., Salvá, J., 1990. 13C NMR of Coumarins. Magnetic Resonance in Chemistry 28, 219–222. Macías, F.A., Galindo, J.C.G., Massanet, G.M., Rodriguez-Luis, F., Zubia, E., 1993. Allelochemicals from Pilocarpus goudotianus leaves. Journal of Chemical Ecology 19, 1371–1379. Mišić, D., Zizovic, I., Stamenić, M., Ašanin, R., Ristić, M., Petrović, S.D., Skala, D., 2008. Antimicrobial activity of celery fruit isolates and SFE process modeling. Biochemical Engineering Journal 42, 148–152. Momin, R.A., Nair, M.G., 2001. Mosquitocidal, nematicidal and antifungal compounds from Apium graveolens L. seeds. Journal of Agricultural and Food Chemistry 49, 142–145. Momin, R.A., Ramsewak, R.S., Nair, M.G., 2000. Bioactive compounds and 1,3-di[(cis)-9octadecenoyl]-2-[(cis,cis)-9,12-octadecadienoyl]glycerol from Apium graveolens L. seeds. Journal of Agricultural and Food Chemistry 48, 3785–3788. Morozumi, S., Wauke, T., Hitokoto, H., 1984. Ann. Rep. Tokyo Metr. Res. Lab. P.H. 35 pp. 31–37. Muhammad, A.K., Umm, K., Muhammad, I.K., Rahamdad, K., Sher, A.K., 2011. Screening the allelopathic potential of various weeds. Pakistan Journal Weed Sciences Researsh 17, 73–81. Oguro, D., Watanabe, H., 2011. Synthesis and sensory evaluation of all stereoisomers of sedanolide. Tetrahedron 67, 777–781. Pavela, R., Vrchotová, N., 2013. Insecticidal effect of furanocoumarins from fruits of Angelica archangelic L. against larvae Spodoptera littoralis Boisd. Industrial Crops and Products 43, 33–39. Pavela, R., Sajfrtova, M., Sovova, H., Barnet, M., Karban, J., 2010. The insecticidal activity of Tanacetum parthenium (L.) Schultz Bip. extracts obtained by supercritical fluid extraction and hydrodistillation. Industrial Crops and Products 31, 449–454. Purohit, N.V., 2001. Synthesis and studies on biological activities of some substituted 2benzopyran-1H-one, 1H-2-oxo-benzopyran- 3-carboxylic acids, and 2-benzofuran1(H)-one. Indian Journal of Chemistry, Section B: Organic Chemistry Including Medicinal Chemistry 40, 222–227. Qasem, J.R., Foy, C.L., 2001. Weed allelopathy, its ecological impacts and future prospects: a review. Journal of Crop Production 4, 43–92. Qasem, J., Rand Hassan, A.A., 2003. Herbicidal properties of some medicinal plants against Malvasyl vestris and Portulaca oleracea. Agricultural Sciences Journal 30, 84–99. Rice, E.L., 1984. Allelopathy. Second edition. Academic Press, Orlando, FL, New York, p. 422. Shimomura, N., Sashida, Y., Nakata, N., Kawasaki, J., Ito, Y., 1982. Plant growth regulators from Heracleum lanatum. Phytochemistry 21, 2213–2215. Sowbhagya, H.B., Srinivas, P., Krishnamurthy, N., 2010. Effect of enzymes on extraction of volatiles from celery seeds. Food Chemistry 120, 230–234. Tang, J., Zhang, Y., Hartman, T.G., Rosen, R.T., Ho, C.T., 1990. Free and glycosidically bound volatile compounds in fresh celery (Apium graveolens L.). Journal of Agricultural and Food Chemistry 38, 1937–1940. Tapondjou, A.L., Adler, C., Fontem, D.A., Bouda, H., Reichmuth, C., 2005. Bioactivities of cymol and essential oils of Cupressus sempervirens and Eucalyptus saligna against Sitophilus zeamais Motschulsky and Tribolium confusum du Val. Journal of Stored Products Research 41, 91–102. Wardle, D.A., Nicholson, K.S., Rahman, A., 1996. Use of a comparative approach to identify allelopathic potential and relationship between allelopathy bioasays and competition experiments for ten grassland and plant species. Journal of Chemical Ecology 22, 933–948. Woods, J.A., Jewell, C., O'Brien, N.M., 2001. Sedanolide, a natural phtalide from celery seed oil: effect on hydrogen peroxide and tert-butyl hydroxyperoxide-induced toxicity in HepG2 and CaCo-2 human cell lines. In Vitro & Molecular Toxicology 14, 233–240. Zheng, G.Q., Kenny, P.M., Zhang, J., Lam, L.K.T., 1993. Chemoprevention of benzo [a] pyrene induced forestomach cancer in mice by natural phthalides from celery seed oil. Nutrition and Cancer 19, 77–86.