Indian Journal of Experimental Biology Vol. 38, January 2000, pp. 63-68

Growth inhibitory and antifeedant activity of extracts from Melia dubia to Spodoptera litura and Helicoverpa armigera larvae.

Opender Koul, M. P. Jain* & V. K. Sharma*

Insect Biopesticide Research Centre, 30 Parkash Nagar, Jalandhar, 144 003 , India

and

*Regional Research Laboratory, Jammu, 180001, India

Received 21 April 1999; revised 14 September 1999

The growth inhibitory activity and deterrency of Melia dubia (Meliaceae) extracts to Spodoptera lilllra and Heficovl'11J(/ armigera were investigated. Artiticial diet bioassays using neonate larvae of both S. litura and H. armigera indicated that dichloroethane (DCE) and methanol (Me) extracts of M. dubia inhibited growth in a dose dependent manner. DCE and Me-SII fractions also resulted in SO% deterrency at concentrations of 22.S and 16.8 Ilglcm2 respectively against S. lilura larvae in a leaf disc-choice test. The DCE-S fraction was found to be more toxic to larvae (LC~o of 0.6S%) than the Me-SII (LCso of 0.8%), 72hr after topical application. Both fractions lack contact toxicity, but the deterrent effect persisted for at least 60hr under laboratory conditions. Although salannin was isolated from the DCE fraction to show antifeedant activity, the physico-chemical characteristics of the active fractions DCE-S and Me-SII were not identical with either salannin or azadirachtin.

Interest in the possible application of natural products in integrated pest management (IPM) remains high; during the last decade ample emphasis has been placed on the reduced use of synthetic pesticides and the use of biopesticides less disruptive to the environment. Melia dubia plant has been used in Indian folk medicine to control insect pests, however, its taxonomic placement has remained controversial. Some authors consider M. dubia synonymous with M. composital ,2 and others synonymous to M. azedarach3

.4 although quite a few features remain different in this species and chemical profiles of these taxa differ. Accordingly we have prefered to retain the name M. dubia for the species.

The bioactivity is apparently similar to that observed in related species like the neem tree, Azadirachta indica, the chinaberry tree, Melia azedarach and Melia volkensii. Several studies have already shown the potential of A. indica for insect controI5

.6

. M. azedarach7-IO and M. volkensii also

possess insect growth inhibitory properties" ·'2 but their bioactive compounds are not identical with azadirachtin, the principal bioactive compound in neem. Toosendanin, a limonoid occurring in the bark of Melia toosendan and M. azedarach, also possesses insect growth inhibitory activity mainly against lepidopteran pests lJ

, but it is far less active than

azadirachtin . As little information exists on the bioactivity of M. dubia, we separated and tested the most active fractions against two species of lepidopterous pests in order to establish the possibility of developing a biopesticidal material from this source.

Materials and Methods Plant material-Leaves and twigs of Melia dubia

were collected from the Ganga Lake area of Arunachal Pradesh in the Eastern part of India during May and June Months for 3 subsequent years. The plant material was shade-dried and subsequently pulverised to a fine powder. The powder (2.23 kg) was extracted with petroleum ether (PE; b.p. 60°-80°C) followed by dichloroethane (DeE) and methanol (Me) extractions respectively . Extract yields 'were 2% (PE), 1.1% (DeE) and 5.3% (Me). Each extract was subjected to bioassay directed fractionation using column chromatography over silica gel following the scheme Fig. I. As the PE extract was the least active, it was not subjected to further fractionation. Salannin was isolated from one of the DeE extract fractions by the method described earlier2.

Insects-Tobacco armyworms (Spodoptera filura) and grampod borers (Helicoverpa armigera) were

64 INDIAN J EXP BIOL, JANUARY 2000

I PE-Extnct

(2.0%)

lIell. duble (2.3 Kg)

I ~ction

DCE-ExtnIct (1.1%)

Pooled Refnlctioned

.1

Me-SI

I Me-Extnlct

(5.3°,,)

e1;1m

1m

1m

Me-511

Fig. I-Extraction scheme for Melia dubia . PE and p=petroleum ether, DCE=dichloroethane, Me and

m=methanol and e=ethyl acetate.

obtained from laboratory colonies reared on artificial diet at 26°±1°C, 67-75% RH and a 16:8 LD photoperiod. Insects for bioassays were neonate (first instar) or 2-4hr-old 3rd

_ and 4th_ instar larvae '4 .

ChroniG larval growth-Test diets (lOg) containing the extracts or various fractions thereof were prepared by the method of Isman et'al. 15

• After the preparation of the diet, it was cooled and cut into 20 pieces that were placed individually into 30 ml plastic cups into which two neonate « lOhr-old) larvae were released. The cups were placed in plastic boxes lined with moistened filter papers to maintain high humidity, and boxes were held in an environment chamber at 26°±loC and a 16:8 LD photoperiod. After 9 days, larvae were weighed, and the mean weights for each treatment group expressed as a percentage of mean control weight. For each treatment, five dietary concentrations were tested along with controls. .

Deterrent activity-Deterrent activity of various preparations was assayed using a leaf disc choice test l6 with S. litura larvae. Discs (3 .5 Cm,2, average weight='110 mg) were punched from castor leaves (Ricinus communis) and treated on each side with 10 fll of aqueous treatment solutions emulsified with 0.1 % Triton~X-lOO. Untreated discs were treated with

10 fll carrier alone. After the solutions had dried, two 4th instar larvae were introduced into each arena (a 9 cm diam. plastic petridish) containing one treated and one control disc. The petridishes were held at ambient (25°C) temperature for 5 hr. Remaining leaf material was measured with a digitalized leaf area meter (Model GAM 200, Blue Line Marketing Pvt. Ltd., New Delhi). The index of feeding deterrency for

. 17 each treatment was calculated as per Koul .

Toxic action-Direct contact toxicity of extracts and various fractions was determined by topical application using acetone solutions and by spraying aqueous emulsions of the test materials . For topical application the test materials in 1 JlI of acetone were applied to the dorsum of 3"1 instar S. lilura and H. armigera larvae using a fine 5 fll syringe (7105 series syringe, Hamilton Co.~ Reno, Nevada, USA) attached to a repeating dispenser (PB-600, Hamilton Co.) . Five concentrations ranging from 0-50 flg/fll were tested (acetone alone as control) with 20 larvae treated at each concentration individually/plastic cup. Treated larvae were reared on artificial diet for 72hr. Mortality was recorded daily .

For spray application, the micro-spray plexiglass system with the nozzle from a hair spray bottle fastened at the top was used. S. litura (n=20) 4th instar larvae were placed on a section of castor leaf in a petridish (15 cm). The test material emulsion (I ml) was delivered as a fine mist to 100 cm" of leaf surface area. Five coricentrations were prepared. After spraying, the treated larvae were transferred to fresh leaf material. Other conditions were maintained as described fot topical application .

. Residual activity-Persistence of the deterrent activity of M dubia extracts and various fractions fo 4th instar S. fitura Jarvae were investigated under laboratory conditions. Five concentrations (0.1-2.0%) of aqueous emulsions were prepared and sprayed on castor leaves to runoff. Thirty minutes after application of the treatment (=day 0), leaf discs (3 .5 cm2

) were punched from treated leaves and leaf disc bioassays using ~th instar larvae were conducted as described above (n=lO/treatment) . Bioassays were conducted daily for 4 days using leaves treated on day' O. Effective concentrations resulting in 50% deter­rency were calculated . .

Statistical analysis-Data were subjected to analysis of variance (ANOVA), linear regression ' 8

and probit analysis I') to determine LD5(h LC50 and EC50 values.

KOUL el af.: INSECT GROWTH INHIBITORY & DETERRENT ACTIVITY OF MELIA DUBIA 65

Results From 2.23 kg of dry plant material, only 45.1 g of

crude extract with PE, 23.8 g with DCE and 118.9 g with Me, equivalent to 2.0, 1.1 , and 5.3% of dry weight respectively was extracted. However, these extracts were found to be active at threshold levels of 1000 ppm (PE extract) and 500 ppm (DCE and Me extracts) against neonate larvae of S. litura and H. armigera (Table 1).

and eluted with ethyl acetate or ethyl acetate: methanol (1: 1). The two active fractions obtained, DCE-4 and DCE-5, retarded the growth of both S. lituraand' H. armigera neonate larvae (Table 2), at the lower concentration of 250 ppm. Mortality in the DCE-5 treatment group was high and almost double that of the parent DCE extract. When mortality was

Leaf disc tests with 4th -instar S. litura larvae indicated that PE extracts were ineffective, but DCE and Me extracts deterred larvae in a dose-dependent manner. The deterrency index (%) was significantly (P<0.05) correlated with concentration of DCE extract (r2=0.86) and Me extract (r2=0.92) (Fig. 2). Effective concentrations resulting in 50% deterrency (ECso) for the DCE extract was 32.0 llg/cm2 (29.4-35.0, 95% CI) and for the Me extract, 28.5 Ilg/cm2 (26.6-32.2, 95% CI).

100~------------------------~

l 80

j 60

i : o

10 20 30 40 50 Concentration (ltg/em')

Most of the activity from the DCE extract was recovered when extract was absor~ed on silica gel

Fig. i-Deterrence ofDCE and Me extracts to S. litura 41h instar larvae in a leaf disc choice bioassay.

Table I-Toxicity (after 9 days) and growth inhibitory effects of three major M. dubia extracts in an artiticial diet test against S. litura (Sl) and H. armigera (Ha) larvae

Concentration PE-Extract DCE-Extract Me-Extract (ppm) SI Ha SI Ha SI Ha

Mortality (%)

0 O.Ob O.Ob O.Od O.Od O.Od D.Od

250 O.Ob O.Ob 10.Oc 10.Oc 6.7bc 3.3c

500 O.Ob O.Ob SO.Ob S6.7b 16.7b 13.3b

1000 3.3a 6.8a 73.3a 76.7a 26.7a 30.0a

Growth

0 lOO.Oa lOO.Oa 100.Oa 100.Oa 100.Oa 100.Oa

250 80.Sb 82.4b 48.Sb 57.6b 68 .8b 59.8b

500 43.lc 48.lc 7.2c 9 .5c 24.0c 22 .5c

1000 Il.2d 8.8d 2.4d 3.Sd 13.2d IO.4d

For each parameter, means within a column followed by same letter are not significantly different (P < 0.05)

Table 2-Toxicity (after 9 days) and effect on growth of two subfractions of DCE-extract in an artiticial diet test against neonate S. filura (SI) and H. armigera (Ha) larvae

[Values are mean±SE]

Concentrat DCE-4 DCE-5 ion SI Ha SI Ha

(ppm) Gr Mor Gr Mor Gr Mor Gr Mor

0 100 0 IDq 0 100 0 100 0

250 14.6+0.20 6.0+0.11 18.5+0.22 2.0+0.04 12.9+0.28 50.0+0.91 18.5+0.18 45 .0+0.76

500 7.8+0.28 15.0+0.21 9.0+0.30 5.0+0;]9 · 2.8+0.08 70.0+0.83 3.5+0.09 65.0+0.98

Gr-=growth and Mor=Mortality

66 INDIAN J EXP BIOL, JANUARY 2000

used as a criterion for the effect of the two DCE fractions on the larvae, larval survival did not exceed 72hr in DCE-S treatments. A majority of the larvae died prior to ecdysis which was delayed. Antifeedant activity of DCE fractions 4 and S revealed that fraction S was more deterrent than fraction 4. The EC50 for fraction S was 22.S llg/cm2 (r2=D.88) whereas the value for fraction 4 was 39.S Ilg/cm2 (r2=D.9D). Salannin already known to possess antifeedant activity J3 was isolated from fraction 4.

At SDD ppm, the methanol fraction Me-3 (Fig. 1) limited larval growth to 9.8% of controls in S. litura and 1O.S% in H. armigera larvae (Table 3) . Mortality ranged from 6D-7S%. Pooled fractions Me-S and Me-6 (Fig. 1) were also deterrent to both 'species (Table 3), however, at the threshold treatment level of 200 ppm. However, a further fraction Me-SII, purified from the above-pooled fraction (Fig. I) indicated the presence of the active ingredient in this fraction due to its high activity at 100 ppm level (Table 3). This fraction from the Me-extract was the most active material and showed a significant deterrent activity (P<D.DS) against S. fitura larve in a choice test with only 16.8 Ilg/cm2 required to produce 50.% deterrence (r2=0..92).

Toxic action of DCE-5 and Me-511-When an acetone solution of DCE-5 and an ethanol solution of Me-511 were applied topically to 3rd instar S. litura and H. armigera larvae, the LD50 values were very high (Table 4) . However, the LC50 values for emulsified preparations of both fractions against the

same stage of larvae were highly significant after 72hr of treatment (Table 4) .

Residual activity of active fra ctions-Regression analysis revealed that deterrence of DCE-5 and Me­Sll remained significantly dose dependent (p<D.DS) for at least 6Dh after application to foliage (Table 5) . For bioassays conducted 6Dhr after treatment, a correlation coefficient of 0. .80. gives a coefficient of determination of 0..69 for DCE-S (Table 5) indicating that 69% of the deterrent activity could be attributed to this fraction. A correlation coefficient of 0..82 gives a coefficient of determination of 0..71 for Me-5II fraction (Table 5) indicating 71 % of deterrent activity could be attributed to this fraction. These results suggest that the activity of active fractions of M. dubia persist for at least 6Dhr when sprayed on castor leaves.

Discussion The growth inhibitory and antifeedant compounds

in M. dubia have yet to be identified. However, salannin was isolated from one of the DCE fractions . Salannin, previously isolated from Azadirachta indica, is known to deter feeding in insects5

.14 and has

been recently isolated from M. elubia l. Bioactivity of

salannin was not studied further due to its well known • J: d . . 14 20 21 I d h . anthee ant actIvity " . nstea ot er active

fractions having apparently more pest control potential than salannin were studied. The isolation procedure adopted in the present study was necessiated because procedures previously used for

Table 3-Growth inhibitory effect of Me-fraction (% of control) on neonate S. fitura (SI) and H. armigera (Ha) larvae

[Values are mean±SE]

Concentration Me-3 Me-5/6 Me-51/ (ppm) SI Ha SI Ha SI Ha

100 65.0+5.3" 69.5+5.4" 70.1+6.7" 69.8+5.8" 17.6+I.W 14.g+ l.ri'

200 29.5+3 .7" 31 .2+2.8" 11.4+ 1.3" 9.5+1.5" 4.8+0.8h 5.0+0.2"

500 9.8+ 1.2" 10.5+1.8" 6.3+0.5"" 4.8+0.3"" 2.2+0.2" 1.5+0.1 "

Means within a column followed by same letter are not significantly different (P < 0.05)

Table 4-Mortality of 3ru instar S. litura larvae up to 72hr post-treatment following topical application or consumption of ti)liage treated with active fractions DCE-5 and Me-5I1

[Values are mean±SE]

LD~() (ltg/insect) LC~() (% )

24 hr 48 hr 72 hr 24 hr 48 hr 72 hr

DCE-5 40.0±4.6 38.0±3.5 29.5±3 .0 1.5±O.7 2.0±0.6 0.65±O. 1

Me-51I 45 .0±5.1 41.0±3.8 32.0±2.9 1.4±O.8 1.8±O.7 O.g±O.1

KOUL et al.: INSECT GROWTH INHIBITORY & DETERRENT ACfIYITY OF MELlA DUBIA 67

Table 5-Residual activity of DCE-5 and Me-51I fractions on 3,d instar S. fitura larvae

Time after EC5u (%)" r2

application (hr) DCE-5 Me-511 DCE-5 Me-511

0 0.50 0.61 0.94 0.92

24 0.65 0.85 0.94 0.91

36 1.28 1.30 0.92 0.90

48 1.60 1.58 0.85 0.87

60 1.85 1.90 0.69 0.71

72 >2.0 >2.0 0.32 0.28

a Effective concentration resulting in a deterrency index of 50%

neem kernel extraction did not yield consistent and satisfactory fractionation, when adopted for M. dubia . The present results show that the active principle(s) in M. dubia is(are) not necessarily related to azadirachtin. A similar situation exists in M. volkensii l2

. However, unlike A. indica and M. volkensii extracts, growth inhibitory and antifeedant bioactivities increase with increasing polarity of extracting solvents. The most active fraction is eluted with ethyl acetate:methanol or methanol alone. The amount of material extractable from M. dubia in methanol is greater than that obtained with other solvents and is of higher activity. Therefore, methanol extraction of M. dubia plant material seems to have substantial potential and should be tested against other insect species in order to determine its spectrum of action.

The present study demonstrated that DCE-S and Me-SIT fractions possess significant deterrent activity against S. litura and H. armigera larvae. In leaf disc choice bioassays, which are more sensitive than no­choice tests22 both fractions rendered foliage relatively unacceptable to these lepidopteran larvae. In terms of feeding deterrence these fractions were equipotent.

The active fractions have low contact toxicity. However, increasing toxicity over time suggests a physiological effect. The high mortality observed during ecdysis leads to speculate that these fractions may act via an endocrinal pathway. It is also obvious that antifeedant activity contributes in part to poor larval growth and not to a growth regulatory effect per se. At this time it is difficult to predict the mechanism of action of M. dubia active fractions, although the present study demonstrates that both active fractions lack contact toxicity but are significantly deterrent and to a greater extent,

physiologically active. However, more experiments are needed to determine the contribution of each of these two modes of action of the active fractions. The two activities offer a novel approach for pest management by rendering plants unacceptable to pest insects. In this respect M. dubia fractions show a relationship to various neem preparations which also possess both antifeedant and growth inhibitory properties6 and have exhibited potential in pest

. 'i management In recent years' .

It is often desirable for an insect control agent to persist in the field for at least several days. The present residual activity study indicates that in the laboratory deterrent activity persists for at least 60hr. However, this activity may be very short-lived under field conditions.

Phytotoxicity limits the use of many products as crop protectants . In spray experiments on castor foliage no phytotoxicity was observed after various applications of emulsified preparations. Although it is too early to envisage the potential of M. duhia as a botanical biopesticide, the present data have shown that this plant specIes IS worthy of further evaluations.

Acknowledgement The authors thank Dr. S.N. Sharma, Regional

Research Laboratory, Jammu for collecting the plant material.

References I Purushothaman K K, Durai swamy K & Connoll y J D.

Phytochemistry, 23 (1984) 135 . 2 Srivastava S D & Srivastava S K, Fitoterapill. LXYII ( 1996)

113. 3 Mabberley A, Gardens Bulletill. 37 (1984) 49. 4 Ascher K R S, Schmutterer H. Zebitz C P M & Naq vi S N H.

in The neem tree, edited by H Schmutterer (YCH . Weinheim. FRG) 1995 , 605.

5 Scbmutterer H, The neem tree Azadirachta indica A .lIISS alld other mefiaceous plallts: Sources (If" unique IIl1tura! products for integrated pest mGlWRemelll. medicine. industry a/ld other purposes (YCH, Wcinheim. FRG) 1995. 608.

6 Koul O. in Supplement to cultivatio/l and utilizatioll of medicinal plall/s. edited by S S Handa & M K Koul (PID. CSIR, New Delhi) 1996, 583 .

7 Mikolajczak K L & Reed D K, .I Chem £Col. 13 (1987) 99. 8 Breuer M & Devkota B, J Appl Ent. 110 (1990) 12X. 9 Dimetry N Z & Schmidt G H. Scop Boll Zoot (/gr Bachic Ser

11,23 (1991) 143 . 10 Champagne D E, Isman M B. Downum K R & Towers G H

N. Chemoeco[ogy. 4 (1993) 165. II Mwangi R W, Entomo[ Exp Appl. 32 (1982) 277. 12 Mwangi R W & Rembold H. Entolllo/ Exp App/. 46 (1988)

103.

68 INDIAN J EXP BIOL, JANUARY 2000

13 Chiu S -F, J Appl Ent, 107 (1989) 185. 14 Koul 0, Shankar J S & Kapil R S, Entornol exp appl, 79

(1996) 43. 15 Isman M B, Koul 0, Luczynski A & Kaminski J, J Agric

Food Chern, 38 (1990) 1406. 16 Koul 0 & Isman M B, Insect Sci Applic, 1\ (1990) 47. 17 Koul 0, in Chemical ecology of phytophagous insects, edited

by T N Ananthakrishnan & A Raman (Oxford & IBH Pub. Co. Pvt. Ltd. , New Delhi), 1993, 51 .

18 Wilkinson L, Systat: The system f or statistics (Systat Inc .. Illinois), 1990,677 pp.

19 Finney D J, Probit analysis (Cambridge University Press). 1971.

20 Yamasaki R B & Klocke J A • .1 Agric Food Cllem. 37 (1989) 1118.

21 Meisner J. Ascher K R S. Aly R & Warthen J D Jr. Phytoparasitica , 9 ( 1981 ) 27.

22 Schoonhoven L M. Elltolllol Exp Appl. 31 (19H2) 57.