Bioactivity of fractionated indigenous medicinal plant extracts of Phlomis damascena Born . and Ranunculus myosuroides against the cotton whitefly , Bemisia tabaci ( Hemiptera : Aleyrodidae )

1 Department of Agricultural Sciences, Faculty of Agricultural & Food Sciences (FAFS), American University of Beirut (AUB), Lebanon. 2 Al Ryum Contracting Company Doha – Qatar, P.O. Box: 207435, Qatar. 3 Department of Chemistry, Faculty of Arts & Sciences, AUB, Lebanon. 4 Department of Agricultural Sciences, FAFS, AUB, Lebanon. 5 Department of Landscape Design & Ecosystem Management, FAFS, AUB, Lebanon.


INTRODUCTION
The cotton whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) is polyphagous in nature (Greathead, 1986;Cock, 1986).Repeated spray applications against B. tabaci on different crops have been necessary and often resulted in overuse of these chemicals.Consequently, this pest has developed resistance to numerous conventional insecticides worldwide (Dittrich and Ernst, 1990;Gerling and Kravchenko, 1995;Palumbo et al., 2001).The severity of this pest and other economic pests have increased the need for effective, biodegradable pesticides with greater selectivity, and alternative strategies that include the search for new types of insecticides and the use of traditional botanical pest control agents or their fractionated bioactive components.
Many plants have developed chemical defenses to withstand attacks of herbivores; some plants may be cultivated to provide sources of biodegradable pesticides (Farombi, 2003).The structurally diverse natural compounds of bioactive plant extracts might act as repellents, deterrents, antifeedants, growth inhibitors, and toxins.Lately, consumer awareness for natural over synthetic pesticides is growing.Many indigenous medicinal plants are known in the world as sources of drugs or herbal extracts for various chemotherapeutic purposes (Purbrick, 1998;Farombi, 2004).These plants can be used for other purposes as acaricides, fungicides, rodenticides or insecticides (Mishra et al., 2013).However, there is a need to organize the natural resources of these endemic plants, develop quality control, adopt standardi-zation strategies and modify regulatory mechanisms for use as botanical pesticides (Koul and Walia, 2009).
A few studies have dealt with the use of fractionated medicinal plant extracts or their components as potential pesticides against whiteflies.Two bioactive compounds from the Chilean plant Calceolaria andina (Scrophulariaceae), related to the familiar garden 'slipper' plant, have been identified as hydroxynapthoquinone and its acetate, which are effective against a range of commercially important pests including the tobacco whitefly, Bemisia tabaci, aphids and the two-spotted spider mite, Tetranychus urticae (Khambay et al., 1999).Certain compounds in the plant genus Agalia were found to be effective against a range of resistant insect strains including the B-biotype of the tobacco whitefly, B. tabaci (Koul and Walia, 2009).Yadegari et al. (2013) studied the effect of four plant extracts and essential oils of thyme Thymus vulgaris L., yarrow Achillea millefoliun L., lavender Lavandula angustifolia Mill and fennel Foeniculum vulgare Mill against eggs and nymphs of the whitefly Trialeurodes vaporariorum Westwood.Their results show that there were significant differences between extracts and essential oils of these plants.The best ovicidal extracts were obtained from thyme and the most nymphicidal effectiveness was with essential oils from fennel.They also found that the essential oils of fennel and lavender were strongest and weakest, respectively than other tested essential oils of thyme and yarrow, against the whitefly eggs.
The main objective of our study was to determine the bioactivity of chemical fractions of 2 botanical raw extracts that were found to have significant repellent effect against the adult whitefly B. tabaci in a previous study (Hammad et al., 2014).Another objective was the initial determination of the chemical characterization of the bioactive fractions in only one of these bioactive plant extracts.

MATERIALS AND METHODS
All plant extraction and phytochemical procedures were performed at the Department of Chemistry in the Faculty of Arts and Sciences, American University of Beirut (AUB), Lebanon, at room temperature of 18-23°C.Insect rearing and all bioassays with whiteflies were performed under controlled conditions inside a glasshouse at the Faculty of Agricultural and Food Sciences, AUB, at 25 ± 2°C, R.H. of 80 ± 10 % and photoperiod of 16:8 (L:D).

Plant extract preparation
Plant selection, collection and extraction were performed according to a procedure set for testing different bioactivities of plant extracts at the Center of Initiative for Biodiversity Studies in Arid Regions (IBSAR) located at the premises of the American University of Beirut.

Plant material
In our previous study (Hammad et al., 2014), 41 extracts of 28 medicinal plant species belonging to 10 botanical families, endemic to Lebanon, were selected and tested against the whitefly B. tabaci; five plant extracts out of the 41 extracts showed significant repellency against the adult B. tabaci and out of these five extracts only two extracts belonging to the plant species, Phlomis damascena Born.(Family Lamiaceae) and Ranunculus myosuroides Boiss.& Kotschy (Family Ranunculaceae), are used in the current study.The latter plant species were collected locally from 2 locations Mahmeit Baalabeck and Hasroun, respectively.

Extraction method of plant material
Harvested plants were washed with distilled water to remove any contaminants on the surface of plant parts and were dried in the shade at a temperature of 25-32ºC and R.H. of 50-60% with adequate ventilation for 2 weeks.Due to low availability of the two plant species P. damascena and R. myosuroides, the whole plant (wp) sample (including leaves, flowers and stems combined), were ground into very small particles (0.3 mm in diam.) by using a grinder (SM 100 Cutting Mill, Brinkman, Germany) at a speed of 1600 rpm at 60 HZ.
Powdered ground material (100 g) of each plant sp.were soaked in methanol for 16 h and placed in a shaker-incubator for the first 2 h at (Harborne, 1998).The methanol solvent was used to extract *Corresponding author.E-mail: ima27@mail.aub.edu.most of the semi-polar and polar constituents.This pure MeOH extraction was performed using the standard plant material / solvent concentration of 1:10 (w: v).The extract was filtered by vacuum pressure through several layers of sterile cheesecloth.Filtrate of each crude extract was stored in 5-10 ml glass vials wrapped with aluminum foil at -20ºC for use in the bioassays and for further phytochemical analysis.

Fractionation of the bioactive plant extracts
Fractions of the two (wp) extracts, P. damascena (Pd) and R. myosuroides (Rm) that were selected for their repellent bioactivity against the adult whitefly (Hammad et al., 2014), were collected and isolated according to the schematic procedure (Figure 1).

Chemical and chromatographic materials.
Silica gel (0.035-0.07 mm, 6 nm pore diameter; Acros organics, USA) was used as the stationary phase in the Column Chromatography (CC) analysis.The column used for CC analysis was 2.5 cm ID x 60 cm L (Chromaflex ® , Kontes, USA).Thin Layer Chromatography (TLC) glass plates (20×20 cm) precoated with 250 μm layer of Silica gel (Analtech, Uniplate No. 02521, Alltech, Lebanon) were used.The eluting pure solvents in TLC and CC separations were chloroform, methanol and ethyl acetate (CMC, Germany).

Organic solvent extraction of the bioactive crude extracts
Each crude extract was fractionated with different organic solvents (Figure 1).The crude methanol extract (I) was kept at −20°C for 24 h in order to remove the fats from the extract, followed by filtration using a cloth sheet filter and a vacuum pump (Erickson, 1991).
The residue (IR) was soaked in ethyl acetate (EtOAc) and then separated by filtration into a residue of fiber (polysaccharide) and a filtrate containing mainly fats and waxes, the filtrate is EtOAc extract (I.1).The filtrate (IF) was acidified to pH 2 using 5M sulfuric acid (H2SO4) and then extracted with a mixture of chloroform (CHCl3) and water (H2O) (2:1) to remove compounds of high polarity.The semi-polar compounds dissolve in CHCl3 whereas polar compounds will remain in the aqueous layer.Both the aqueous (IA) and organic (IO) phases were separated using a separatory funnel.
The IA layer is basified to pH 10, by adding ammonium hydroxide (NH4OH) stepwise to increasingly higher pH, and then extracted with a CHCl3 and MeOH mixture (3:1) to extract alkaloids.Two layers will develop, the CHCl3-MeOH layer (I.3) and the aqueous layer which is again extracted with MeOH to get (I.4).All four fractions of each extract were tested for their bioactivity and the most bioactive fraction was further analyzed chromatographically.

Phytochemical analysis of the bioactive fraction
The CHCl3-MeOH fraction (I.3) of P. damascena was found to be the most bioactive compared to the different fractions (I.1, I.2 and I.4) of both plant extracts.The (I.3) fraction was collected in 10 ml vials; each vial was concentrated into 1 ml volume for further use in CC and TLC analysis.The solvents chosen during these analyses were chloroform and methanol.The mobile phase, chloroformmethanol (9:1) was selected after trying several types and combination of different solvents; about 32 different combinations of organic solvents, starting from pure solvents of medium elution power to several combinations between solvents of different polarity (Hahn-Deinstrop, 2000); the choice of technique of separation depends largely on the solubility properties of compounds to be separated.Three TLC plates were tested for each solvent system for precision purposes.The mobile phase is optimized using different solvent combinations tested on silica-TLC plates.The best mobile phase was found to be CHCl3: MeOH (9:1).
CC is a valuable technique for purification of synthetic or natural products.The CHCL3-MeOH fraction of P. damascena was separated by CC as well as by TLC, through the same mechanism, by distribution between two phases.One hundred grams of silica gel was used as the stationary phase.A glass column was uniformly packed with slurry of silica gel (45 cm) in the mobile phase solvent system (chloroform-methanol 9:1) and other solvent systems of higher methanol concentration at advanced phases of CC.Thus, the most bioactive fraction material was purified using this silica gel column successively eluted with a stepwise gradient of chloroform: methanol (9:1) with increasing volume of methanol till total elution; a total volume of 4L of CHCl3 and 2L of MeOH were used.The CHCL3-MeOH fraction (0.5 g), originating from crude extract of P. damascena (100 g) was dissolved in 10 ml mobile phase (chloroform-methanol 9:1) and was gently applied with a pipette as concentrated band to the top of the column after the mobile phase has been drained until 1 cm above the bed surface.
The mobile phase was applied on the top of the column and continuous elution started; volume of each solvent used was 4 and 2 L for chloroform and methanol, respectively.The column was developed by allowing the mobile phase to pass through the silica at a flow rate of ~ 1-3 ml / min.As different organic compounds elute through the column they were collected in test tubes (20 ml) for further TLC analysis.

TLC analysis
TLC was employed in this study to isolate the compounds present in the most bioactive CHCl3-MeOH fraction (I.3) of P. damascena.Silica gel was chosen as stationary phase, as it is an efficient adsorbent for the TLC separation of most plant extracts (Barbetti et al., 1987;Houghton et al., 1996;Wagner and Bladt, 1996).Aliquots of 15-20 μl of the isolated fractions collected by CC were applied as separate spots on a TLC plate about 1.5 cm from the edge (spotting line), using 25 µl Hamilton precision syringes (Hague-Holland).
After sample application, the plate was placed vertically into a solvent vapor saturated TLC chamber (12 x 15 x 10 cm).The mobile phase used was chloroform-methanol (9:1) and the spotting line was about 0.5 cm from the developing solution.After the mobile phase had moved about 80% from the spotting line, the plate was removed from the developing chamber and air dried in a fume hood to visualize spots within 1-2 min.on the plate.The migrated spots on TLC plates representing various fractions / compounds were visualized with UV lamp at UV-254 nm.The collected CC eluted material that showed at TLC plates a common pattern for the fractionated compounds were combined, concentrated and tested for their bioactivity against adult whiteflies on leaves of plants.Thus, similar TLC pattern allowed pooling of eluted material into several potent fractions that were isolated and corresponded to bands detected on the bioactive fraction TLC plate.

Whitefly colony
B. tabaci colony was raised in a glasshouse compartment under controlled conditions as mentioned above.The colony, originally from a field population was reared on cucumber plants of the variety Beit Alpha (F1 parthenocarpic, dust free and thiram treated seeds; Edena Seeds, USA) in a whitefly proof cage (140 x 85 x 130 cm) covered completely with a fine mesh (270 x 770 µm).Two true leaf seedlings were grown in 12 cm plastic pots to provide a continuous supply of healthy young plants to the whitefly colony Hammad et al. 49 and bioassays.Fertilization with Floral ® (20-20-20+ microelements; Cifo S.p.A., Bologna, Italy) was applied at a rate of 5g per 10 L through irrigation of the seedlings, about 2 times a week.

Experimental setup with whitefly adults
Cucumber seedlings having two true leaves with detached cotyledons were used in the bioassays.In one bioassay, the treatments included the two crude whole plant extracts of the two plant species P. damascena and R. myosuroides with 4 fractions for each extract as: EtOAc (I.1), CHCL3 -H2O (I.2), CHCL3-MeOH (I.3) and MeOH-MeOH fraction (I.4) plus 2 controls: distilled water and 10% methanol.In another bioassay, treatments included the most previously detected bioactive fraction of the two extracts, the CHCL3-MeOH fraction of P. damascena with its five fractions and the two controls: distilled water and 10% Methanol.
Each treatment was replicated 9 times during the study.Ten millimeters of each crude extract or its fraction was rotovaped up to 1 ml, after which 9 ml of distilled water was added to homogenize the solution before application to the plant.Each seedling received an average of 9 ml of the extract/fraction or the control by spraying them on the upper and lower sides of the leaves using 10 ml glassbottle sprayers.Each treated seedling was allowed to air dry and consequently was placed in one plastic cage (28 cm high x 21 cm diam.manufactured locally) having an aeration opening of (15 cm diam.) at the top and two circular openings (5 cm diam.) in body of cage, covered by a mesh net (270 x 770 μm).
Ten adult whiteflies (of about 3 days old) were collected by a hand aspirator (Hausherr's machine, N.J., USA) and introduced into each treated seedling in one plastic cage.Numbers of adult whiteflies dead or alive in each cage were recorded at 72 h after treatment with specification of the location of the insect in the cage as on the plant, walls or top cover of the cage or on soil surface in pot; dead whiteflies were counted to relate to presence of toxic effect, if any, that might be attributed to residues of plant extracts after spraying on plants.Selection of repellency assessment at 72 h after treatment was based on the following observations related to the extracts.A few hours after treatment, adult whiteflies were found landing at the top of the cage, this could be attributed to the fact that the bioactive extracts contained some volatile compounds that repelled the insects from approaching the plant.However, observations at 72 h confirmed the repellent efficacy (Coudriet et al, 1985;Ateyyat et al., 2009;Martini et al., 2012;Yadav and Mendhulkar, 2015) of some extracts to whiteflies, knowing that some botanicals are characterized by reduced stability which indicates that the presence of a pest on treated leaves was more obvious with time (Sundaram, 1996), mostly with non-repellent extracts.

Statistical analysis
All experiments were laid out in a completely randomized block design with treatment as the only factor.Each treatment was replicated 9 times.Data were transformed by sqrt (x+1), x being the number of live adult whiteflies after treatment, to normalize the data.Analysis of variance (ANOVA) was performed over the treatment factor, using the SPSS statistical package (Anonymous 2010).All means were separated by Fisher's LSD test (1949), at a significance level of  = 0.05.

Determination of bioactive fractions in two whitefly repellent plant extracts
Results of our study have shown that some fractions of the two crude bioactive extracts of P. damascena and R. myosuroides were significantly different in number of live adult whiteflies encountered per plant compared to the two controls (Table 1), knowing that the non-encountered whiteflies on the plant were not necessarily dead at the end of the experiment.In this bioassay, the two CHCl 3 -MeOH fractions (Pd wp I.3 and Rm wp I.3) and the EtOAc fraction (Rm wp I.1) were not significantly different in their effect from the two crude extracts Pd wp I and Rm wp I, but all these treatments were significantly different in their effect from the two controls.However, the other fractions were not significantly different in their effect from the two controls.The CHCl 3 -MeOH fraction of P. damascena (Pd wp I.3) showed the most significant bioactivity numerically as only 2.66 of adult whiteflies remained alive per plant compared to 5.56 in both controls; the former fraction was further not significantly different from the original raw extract of P. damascena allowing survival of only 3.22 adult whiteflies per plant.The latter result was further verified and comparable to the previously determined high repellency of this raw extract to the adult insect that allowed encountering of only 2.22 live adult whiteflies per plant at 72 h after treatment (Hammad et al., 2014).However, it seems that the whole plant extract of R. myosuroides have two bioactive fractions, the CHCl 3 -MeOH fraction (Rm wp I.3) and the ethyl acetate fraction (Rm wp I.1) against adult B. tabaci (Table 1); the two fractions were comparable to the original raw extract of R. myosuroides allowing encountering of only 2.77 live adult whiteflies per plant.The latter result was further verified and comparable to the previously determined high repellency of this raw extract to the adult insect that caused encountering of only 1.22 live adult whiteflies per plant at 72 h after treatment (Hammad et al., 2014).Some studies have reported repellency to insects by fractionated plant extracts (Harwood et al., 1990;Tunon et al., 1994;Gkinis et al., 2003;Singh and Metha, 2003).In our study, only the organic fractions Pd wp I.3, Rm wp I.3 and Rm wp I.1 have insecticidal activity against adult whiteflies as very low number of live whiteflies were encountered per plant after treatment in comparison to the control, but the aqueous fractions of both plant extracts further partitioned with CHCl 3 -H 2 O or MeOH were not bioactive.Thus, the former organic bioactive fractions might be enriched with alkaloids as it was found that the combination between two solvents, chloroform and methanol of medium and high polarity, respectively would result in the accumulation of alkaloid-type active compounds (Fried and Sherma, 1994).Similar to our study, Ahn et al. (1997) found in a stepwise extraction procedure with MeOH and EtOAc that the organic phase of the Ginkgo biloba L. extract was the only bioactive material when isolating this most potent extract which was tested within 119 methanol extracts of 52 plant sp.against the brown planthopper Nilaparvata lugens Stal.(Order: Hemiptera).
Furthermore, in our study, the EtOAc fraction of R. myosuroides (Rm wp I.1) was of high bioactivity against the adult whitefly allowing insect survival of only 4.0 live adult whiteflies per plant (Table 1).Similar to our study, Neoliya et al. (2003) have found high bioactivity with EtOAc fraction isolated from aqueous-methanolic and acetone extracts of air-dried leaves of Catharanthus roseus (Linn) that were fractionated successively with nhexane, CHCl 3 , EtOAc and n-BuOH.Their results showed maximum insect growth regulator (IGR) activity  (Hammad et al., 2014), it has been found to have significant antimicrobial activity (Barbour et al., 2004) within a bio-prospection context study; in form of utilizing local plant species as an alternative in sustainable agriculture development.

Effect of isolated sub-fractions of CHCl 3 -MeOH fraction of P. damascena whole plant extract (Pd wp I.3) against whitefly adults
The bioactive CHCl 3 -MeOH fraction of P. damascena whole plant extract (Pd wp I.3) was selected for further fractionation as it caused the lowest significant numerical survival to adult whiteflies among all treatments (Table 1).Analysis of the eluted material of this fraction on TLC plates yielded several sub-fractions.Similar TLC pattern allowed pooling of eluted material into five potent subfractions that were isolated and corresponded to bands detected on Pd wp I.3 TLC plate.R f values of the isolated sub-fractions were 0.72, 0.62 and 0.55 for B1 (35 mg), B2 (42 mg), B3 (87 mg) eluted with chloroform-methanol (9: 1); B4 (76 mg) having R f value of 0.41 was eluted with chloroform-methanol (9: 2) and B5 (12 mg) having 3 bands with R f values 0.72, 0.62 and 0.55 were eluted with chloroform-methanol 9:1 (Table 2).Continued elution with chloroform-methanol (9:3) till (9:9) and with 100 % methanol revealed no fractions on TLC for the collected eluted materials.This suggests the necessity of using other solvent types for eluting the materials or using other chemicals for detecting other compounds on TLC plates.A significant insecticidal activity of the five isolated fractions B1, B2, B3, B4 and B5 was detected against the adult whiteflies (Table 3).There were significant differences in number of live adult whiteflies among all isolated fractions and the controls.However, only the isolated fraction B4 was not significantly different in its effect from the bioactive chloroform-methanol fraction (Pd wp I.3).This indicates that the similar bioactivity detected in the raw extract of P. damascena methanol extract and its bioactive CHCl 3 -MeOH fraction (Pd wp I.3) is due mainly to active components found in the isolated fraction B4.
A few studies dealt with bioassay-guided fractionation for certain bioactive plant extracts of Ginkgo bilboa (Ahn et al., 1997), Heliotropium floridum (Reina et al., 1997), and Serratula coronata L. (Odinokov et al., 2002).Different classes of glycosides comprising diterpenoids, phenylpropanoids, iridoids, and flavonoids had been identified from genus Phlomis L.; many of the phenylpropanoids showed significant biological activities.The air-dried powdered leaves of Phlomis aurea Decne.were extracted with 70% EtOH.Similar to our methodology applied to P. damascena and yielding 5 fractions, the dried ethanolic extract of P. aurea was suspended in H 2 O and deffated with n-hexane, but the MeOH extracts were chromategraphed by silica gel CC using CH 2 Cl 2 -MeOH-H 2 O (70:30:3) and (80:20:2) and yielding four and three fractions, respectively (Kamel et al., 2000).Iridoid glucosides (IG) were also isolated from the 1-butanolfraction of a methanolic extract of the aerial parts of another Phlomis sp., Phlomis rigida Labill.These IG fractions were collected after being chromatographed over Diaion HP-20 column with a stepwise increase of MeOH in water (Takeda et al., 2000).On the other hand, ether extracts of R. sceleratus caused significant reduction in larval activity, pupal weight and pupal emergence of D. melanogaster at all concentrations tested.Similarly, there were weight reductions and high mortality in treated red flour beetles; no insect could survive at 1 and 5% concentrations of the extracts beyond 15 and 10 d, respectively.These insecticidal properties exhibited by the ether extract were attributed

Figure 1 .
Figure 1.Schematic diagram of organic solvent extraction procedure for a bioactive crude extract.*: (I) = initial extract; IR = initial extract residue; IF = initial extract filtrate; IO = initial extract organic layer; IA = initial extract aqueous layer.

Table 1 .
Effect of two bioactive whole plant methanol extracts and their fractions against Bemisia tabaci adults under glasshouse conditions.Ranunculus myosuroudes.**EtOAc = Ethyl acetate; MeOH = Methanol; CHCl3 = Chloroform; MeOH-MeOH = out of aqueous basic layer; CHCl3-MeOH = out of aqueous acid layer.***Means followed by the same letter within a column are not significantly different (Fisher Lsd test; P<0.05).

Table 2 .
Isolated fractions of Pd wp I.3 eluted through CC and analyzed by TLC.