Journal of Chemical Ecology, Vol. 11, No. 8, 1985

ISOLATION A N D PARTIAL CHARACTERIZATION PHYTOTOXIC C O M P O U N D S FROM LANTANA

(Lantana camara L.) 1

OF

N . R . A C H H I R E D D Y , M. S I N G H , L . L . A C H H I R E D D Y ,

H . N . N I G G , and S. N A G Y

University of Florida, Institute of Food and Agricultural Science Citrus Research and Education Center

700 Experiment Station Road Lake Alfred, Florida 33850

(Received August 21, 1984; accepted October 27, 1984)

Abstract--Phytotoxic compounds obtained from methanol extracts of field- grown lantana (Lantana camara L.) were quantified in terms of their inhibi- tion of ryegrass (Lolium multiflorum Lain.) seed germination and/or seedling growth. Subsequent partition of the aqueous fraction (derived from drying the MeOH extract in vacuo and redissolving in distilled water) at various pHs with solvents of differing polarity indicated the phytotoxic compounds were both polar and slightly acidic. Thin-layer chromatographic separation of the acidic butanol fraction in a solvent mixture of butanol, acetic acid, water (4:1:5, by volume) yielded an active fraction with an Ryvalue of 0.89-1.0 that was inhibitory to both root and shoot growth of ryegrass. Two other fractions with Rf values of 0.04-0.23 and 0.41-0.57 were inhibitory only to root but not shoot growth.

Key Words--Allelopathy, allelochemics, lantana, Lantana camara, ryegrass, Lolium multiflorum, TLC, osmotic potential, inhibition.

INTRODUCTION

Desp i te significant advances in weed control technology, crop losses caused by

weeds and cost o f weed control in the Uni ted States a lone are es t imated at $14

bil l ion annual ly (Worsham, 1982). The repea ted use o f herbic ides can result in

the select ion o f herbic ide-res is tant b io types (Bandeen et a l . , 1982; Gressel et

Presented at International Society of Chemical Ecology Annual Meeting held in Austin, Texas, June 10-13, 1984. Florida Agricultural Experiment Stations Journal Series No. 5812.

979

0098-0331/85/0800-0979504.50/0 �9 1985 Plenum Publishing Corporation

980 ACHHIREDDY ET AL.

al., 1982). In addition, increased public awareness of the possible adverse effects of synthetic pesticides on the environment necessitates research on alternative weed control methods.

One area of innovative research is allelopathy and its application to agri- culture (Putnam and Duke, 1978; Putnam and DeFrank, 1983). Allelopathy, the regulation of growth of one plant species by chemicals released from another, occurs widely in many plant communities and is believed to regulate species density and distribution (Rice, 1984). Isolation and identification of natural compounds that are implicated in allelopathy are receiving greater research ef- fort because it is possible that related compounds may be synthesized that can produce similar desirable effects as pesticides. Several plant products have been proven useful in control of insects (Jacobson, 1975), and the potential exists for the development of many other natural insecticides, herbicides, and growth reg- ulators.

Lantana, rapid-growing perennial woody shrub, is a serious weed in Flor- ida citrus groves. Its proliferation and spread are attributed to a number of fac- tors, including natural propagation, reduced use of hand-weeding, high toler- ance to currently available herbicides, and decreased competition from other weeds (Phillips and Tucker, 1976). In an earlier paper, we reported phytotoxicity of lantana (Achhireddy and Singh, 1984) residues to the test species milkweed vine (Morrenia odorata Lindl.) and suggested the involvement of allelopathy. The ways in which the chemical inhibitors were released into the soil from lan- tana residues were also determined. Characteristics of the inhibitory com- pound(s) reponsible for inhibition of milkweed vine were not determined. Milk- weed vine, also a troublesome weed, is spreading in Florida citrus groves (Phillips and Tucker, 1970). In our earlier studies involving the examination of allelopathic effects of lantana, we used milkweed vine as an indicator species because we did not observe milkweed vine infestation near lantana bushes, sug- gesting an allelopathic mechanism. Since phenolic compounds have been impli- cated in most allelopathy studies (Guenzi and McCalla, 1966; Liebl and Wor- sham, 1983; Putnam and Duke, 1978; Rice, 1984; Tang and Young, 1982), we examined for their presence in lantana and for their phytotoxicity against rye- grass as an indicator species.

METHODS AND MATERIALS

Methanol Extracts and their Inhibitory Potential. Lantana shoots were col- lected from field-grown plants near the Citrus Research and Education Center, Lake Alfred, Florida. Leaves and stems were separated and 100 g fresh weight of each was blended with 500 ml MeOH in a Waring blender for 2 min. The extract was vacuum-filtered through a Buchner funnel using Whatman No. 1 filter paper. The final volume of the filtrate was adjusted to 1000 ml with MeOH.

PHYTOTOXIC COMPOUNDS 981

Preliminary experiments were conducted to detect any differences in the inhibitory potential between lantana stem and leaf extracts on ryegrass germi- nation and growth. Ryegrass was chosen for the present studies because of its fast growth, uniform germination, and sensitivity to lantana phytotoxic com- pounds. Furthermore, ryegrass has been widely used as an indicator species in herbicide bioassays (Santelmann, 1977). One- or 5-ml aliquots of the extracts were placed on a filter paper in a 9-cm Petri dish. Filter papers were dried in a hood at lab temperature. Ten ryegrass seeds were placed in each Petri dish, 5 ml deionized water was added, and the Petri dishes were placed in a growth chamber maintained at 25/20~ (day/night) with an 8-hr photoperiod of 100 /xE/m2/sec. There were three replicates for each concentration, deionized water, and MeOH (1 or 5 ml). Ryegrass (cultivar Annual) seeds used in these studies were obtained from Florida Seed and Feed, Ocala, Florida. The germination of ryegrass used in these studies was over 95 % and maximum germination oc- curred within three days. A seed was considered germinated if the radical visibly protruded through the seed coat. Final germination percentage and the lengths of roots and shoots were measured on the fifth day.

Effect of Lantana Residues (MeOH-Insoluble Fraction) on Ryegrass Seed- ling Growth. Methanol-insoluble fractions of lantana stem and leaf were air- dried, and the dried samples were added to sterilized sand in various proportions (0.05, 1.0, 2.0 and 4.0%, w/w). One hundred grams of this residue-sand mix- ture was added to Petri dishes and tested for biological activity using ryegrass as the indicator species. Fifteen ryegrass seeds were placed in each dish, watered thoroughly, and incubated at 25~ (_+2~ photoperiod with a light in- tensity of 100/xE/m2/sec. The germination percentage and root and shoot lengths were measured after eight days.

Preliminary Partitioning of MeOH Extracts with Various Solvents'. The MeOH extracts (10%, w/v) were dried in a rotary evaporator at 40~ The resultant residue was dissolved in 100 ml distilled water. The pH of the solution was 6.5. Aqueous fraction was partitioned successively with hexane, methylene chloride, ethyl acetate, and butanol at pH 2 and 11. In these preliminary exper- iments, the aqueous fraction was not partitioned at neutral pH because, if the phytotoxins were neutral, the activity would be detected in both acidic and basic fractions. The separation scheme is summarized in Figure 1.

All fractions (acidic, basic, and the aqueous remainder obtained for each solvent used) were tested for inhibitory potential. Three milliliters of each frac- tion were placed on a filter paper held in a Petri dish. Solutions were dried under a hood at laboratory temperature. Ten ryegrass seeds were plced in each Petri dish, and 5 ml of deionized water was added. Petri dishes were placed in a growth chamber. There were three replicates for each fraction. Final germina- tion percentages and seedling growth measurements were made on the fifth day.

Based on the bioassays (see Table 3), further experiments involving solvent

982 ACHHIREDDY ET AL.

MeOH extract of lantana

I Filter, dry i_n ~6cup at 40~ Redissolve in lOOml distilled water

Aqueous fraction pH to ii with iN NaOH Dartition, hexane (3 x 50 ml)

Hexane fraction Aqueous fraction (basic) pH to 2 with IN HCi

partition, hexane (3x50 ml)

Hexane fraction Aqueous fraction (acidic) I pH 6,5 with IN NoOH

Aqueous fraction

I Subjected to the same treatment

using CH2CI 2, ethylacetote or n-butanol

FIG. 1. Summa~ of procedu~s used to fractionate methan~ extract ~ lantana ~liage.

partition and TLC were carried out using the butanol fraction with some 'mod- ifications.

Butanol Partition and Thin-Layer Chromatography. One hundred grams of fresh lantana leaves were ground in 1000 ml MeOH and filtered as previously described. The methanol extract was partitioned with petroleum ether to remove chlorophylls and the MeOH fraction was evaporated to dryness in vacuo at 40~ The residue was taken up in 100 ml of distilled water. A 50-ml portion of this fraction was extracted at pH 2, 7, and 11 as outlined in Figure 2. Acidic and basic fractions were dried over anhydrous MgSO4, filtered through a Buchner funnel with fritted disk (medium porosity, 0.t5 #m), and the butanol fractions saved. All fractions [original aqueous, butanol (acidic, basic, and neutral), and the final aqueous remainder] were tested for inhibitory activity as described earlier.

The maximum inhibitor activity was present in the acidic fraction (see Table 5), and further separation using TLC methods were carried out on this fraction. One milliliter of the acidic fraction was separately streaked on a silica plate (Eastman chromatogram sheet; Eastman Kodak, Rochester, New York), and the plate was developed three times in a solvent rnixture of butanol, acetic acid, and water (4 : 1 : 5, by volume, upper layer) known to separate phenolic compounds (Harborne, 1961). The plates were dried in the hood at lab temperature and examined under UV at 254 and 320 nm for absorption and fluorescence when sprayed with the Folin-Ciocalteau reagent (Krebs et al., 1969). All fluorescent

PHYTOTOXIC COMPOUNDS 983

Crude MeOH e x t r a c t of l an tgoo

, Vacuum f i l t e r , Wash with pe t ro leum e t h e r

MeOH e x t r a c t

i Dry in vacua at 40~ I

[Redissolve in lO0 ml water

Aqueous fraction Aqueous froctlon (pH 6.5) (pH 6.5) !pH to 7 with IN NaOH

IPortition, n-butanol (3x15 ml) Butanol fraction Aqueous fraction ( n e u t r a l ) IDH to 2 with 1N HC1

I

Iportition, n-butanol (3x15 ml) Butanol fraction (ocidic) Aqueous f r o c t l o n

pH 11 with IN NoOH ipartltlon, n-butanol (3x15 ml)

Butanol fraction Aqueous fractibn (basic) (DH to 6.5 with iN HCI)

FIG. 2. Summa~ of procedures used to fractionate methanot~ ext~ct of lantana her- bageusing n-butane.

or absorbent- and reagent-reactive bands were removed by scraping. Scrapings were eluted with 5 ml of distilled water. The silica coating and 5 ml of water were thoroughly mixed in a test tube, centrifuged in a clinical centrifuge and the supernatant used for bioassays. Thin-layer chromatographic plates developed in butanol-acetic acid-water mixture without the application of the extract were subjected to the same procedure and used as controls.

Osmotic Potential Measurements. Osmotic potentials of assay media were determined on a vapor-pressure osmometer (vapor-pressure osmometer model 5100B, Wescor Inc., Logan, Utah). In separate experiments, NaC1 was added to distilled water to produce a water potential up to - 130 kPa (comparable to those of lantana extracts), and the effect of this solution was also tested on the germination and growth of ryegrass.

Analysis of Data. All experiments were repeated, and the data were pooled for presentation. The results were analyzed by analysis of variance and the means were compared using least significant different (LSD) at the 5 % level.

RESULTS AND DISCUSSION

Crude MeOH extracts of lantana stem or leaf were inhibitory to ryegrass seedling growth, and the degree of inhibition was dependent on the concentra-

984 ACHHIREDDY ET AL,

TABLE 1. EFFECT OF MeOH EXTRACTS OF LANTANA LEAF OR STEM ON RYEGRASS SEEDLING GROWTH AFTER 5 DAYS

Growth (mm) Concentration

Treatments (mg/ml) a Root Shoot

Control, water 0 52 34 Control, MeOH 0 44 33 Leaf extract 20 25 b 26 b Leaf extract 100 2 b 10 b Stem extract 20 35 b 32 Stem extract 100 l0 b 19 b

abased on fresh weight. bSignificantly different from control at 95% probability level as determined by LSD tests.

tion of the extract used (Table 1). The MeOH control had no significant effects on ryegrass seedling growth. Lantana extracts had greater inhibitory activity

on ryegrass root growth than on shoot growth. Germination, however, was not affected by the MeOH extracts (data not shown).

Lantana leaf or stem residues (MeOH-insoluble fraction) were inhibitory at

high concentrations (20 mg and 40 mg/g) to root but not to shoot growth (Table 2). There was, in fact, some enhancement in shoot growth by stem and leaf

residues at 10-40 mg/g and 10-20 mg/g, respectively.

Butanol was the most effective partitioning solvent tested with regard to

overall growth inhibition (Table 3). Although the ethyl acetate (acidic fraction)

equally inhibited root growth, little or no growth inhibitory effects were ob-

served on shoot growth.

TABLE 2. INFLUENCE OF LANTANA RESIDUES (MeOH INSOLUBLE) ON RYEGRASS SEEDLING GROWTH

Residue concentration

(mg/g)

Growth (mm)

Leaf residues Stem residues

Root Shoot Root Shoot

0 (contro~ 54 54 54 55 5 48 64 60 63

10 51 76 a 58 77 a 20 41 a 75 a 48 83 a 40 30 a 64 30 a 90 a

aSignificantly different from control at 95% probability level as determined by LSD tests.

PHYTOTOXIC COMPOUNDS

TABLE 3. SOLVENT PARTITION OF PHYTOTOXINS PRESENT IN MeOH EXTRACTS a

985

Growth (mm) Germination

Solvent Fraction (%) Root Shoot

Control, water 98 39 30

Hexane Control, hexane 93 38 26 Acidic 93 22 b 18 b Basic 100 34 18 b Aq. remainder 0 b b b

CH2C12 Control, CH2C12 80 37 31 Acidic 100 10 b 26 Basic 100 38 33 Aq. remainder 90 3 b 8 b

Ethyl acetate Control, ethyl acetate 100 40 30 Acidic 100 7 b 26 Basic 100 51 b 37 Aq. remainder 0 b b b

]Butanol Control, butanol 100 37 29 Acidic 95 8 b 16 b Basic 95 35 30 Aq. remainder 80 9 b ! 3 b

aThe extracts at pH 2 and 11 were fractionated successively with hexane, methylene chloride, ethyl acetate, and butanol.

bSignificantly different from controls at 95% probability level as determined by LSD tests.

The butanol (acidic) fraction inhibited root and shoot growth by about 80 and 60% of the water control, respectively (Table 3). Therefore, the phytotoxins

present in lantana appear to be pr imari ly polar and slightly acidic. It was also interesting to note that even af ter parti t ioning with various solvents, significant inhibitory activity was still present in the aqueous remainder. These results sug- gested that lantana contain water-soluble phytotoxins. This is consistent with our earl ier observation that water leachates of lantana incorporated into the pot- ting soil were extremely inhibitory to milkweed vine seedling growth (Achhi- reddy and Singh, 1984).

In studies concerning the effects of the plant extracts on germination and growth of sensitive indicator species, osmotic potential of the solution can be a problem. Therefore, it is important to test for osmotic effects.

Osmotic potential of aqueous fractions, obtained by evaporation of solvent fractions and then redissolving in distilled water for biossays, ranged between

- 2 3 and - 1 3 8 kPa (Tables 4 and 5). These values were too high to affect

986 ACHHIREDDY ET AL.

TABLE 4. OSMOTIC POTENTIAL OF MeOH EXTRACTS VS. GROWTH

INHIBITION OF RYEGRASS

Extract Final Solution Growth (mm) volume germination osmotic potential

(ml) (%) (-kPa) Root Shoot

0 (control) 100 23 48 30 0.025 100 23 39 a 30 0.05 95 23 37 a 31 0.1 100 23 34 a 32 0.5 90 92 a 15 a 27 1.0 100 92 a 8 a 21 a 1.5 95 92 a 6 a 12 a 2.0 80 a 92 a 2 a 5 a

asignificantly different from control a 95% probability level as determined by LSD tests.

growth of ryegrass seedlings. Methanol extracts of 0.5-2.0 ml produced the same osmotic potential of - 9 2 kPa of the bioassay medium (Table 4). However, less germination and greater growth reduction of ryegrass was observed in 2.0 ml than in 0.5 mt treatment. There were no significant differences in osmotic potential among the various volumes within each of the basic or neutral butanol fractions. Significant differences in percent germination and seedling growth were evident, however, between these volumes (Table 5). This suggests that inhibition resulted from the phytotoxins present in the extracts and not from osmotic stress. In addition, ryegrass seeds germinated and grown in a NaC1 solution with osmotic potential equivalent to that of the extracts showed no sig- nificant reduction in percent germination or seedling growth (data not included).

A large portion of the phytotoxic substances present in the aqueous extract (following MeOH evaporation) partititioned into the butanol acidic and neutral fractions (Table 5). A 1-ml acidic fraction inhibited growth in length of ryegrass root and shoot by about 95 and 97 %, respectively. A 1-ml neutral fraction in- hibited root and shoot lengths by about 93 and 84%, respectively. The basic fraction reduced growth by about 50 %. Although percent germination was not affected by the acidic and neutral fractions, seedling growth was completely inhibited. One milliliter of the aqueous remainder was also inhibitory, reducing root and shoot lengths by 65 and 40 %, respectively.

Chromatographic analysis of the acidic butanol fraction revealed several bands (by UV and color reaction) (Table 6). The eluate from one band with an Rf value of 0.89-1.0 inhibited growth of roots and shoots by 66 and 27 %, re- spectively. Two other bands with Rfvalues of 0.04-0.23 and 0.41-0.57 were also significantly inhibitory only to root but not shoot growth. Greater inhibitory action on :oot growth than on shoot growth in these chromatographically sepa-

PHYTOTOX1C COMPOUNDS

TABLE 5. OSMOTIC POTENTIAL OF VARIOUS BUTANOL FRACTIONS

OF LANTANA FOLIAGE VS. GROWTH INHIBITION OF RYEGRASS

987

Osmotic Growth (ram) Volume potential Germination

Fraction (ml) (-kPa) (%) Root Shoot

Control, water - - 95 43 31

Acidic 0.1 46 100 33 27 1.0 46 70 2 1 2.0 46 65 0 0

Basic 0.1 23 100 41 31 1.0 23 100 19 14 2.0 23 100 1 t 7

Neutral 0.1 23 95 29 28 1.0 23 95 3 5 2.0 23 35 0 0

Aq. remainder 0.1 46 100 43 28 1.0 115 100 16 18 2.0 138 100 13 17

LSD (0.05) 14 25 5 5

TABLE 6. THIN-LAYER-CHROMATOGRAPHIC SEPARATION OF

PHYTOTOXINS PRESENT IN BUTANOL (ACIDIC) FRACTION

UV absorption (ab) or fluorescence (fl) Growth (mm)

Rf Folin test 254 nm 320 nm Root Shoot

0.0-0.04 36 27 0.04-0.23 29 a 27 0.23-0.26 42 31 0.26-0.33 + ab fl 38 27 0.33-0.41 37 30 0.4 t -0 .57 + ab 24 a 25 0.57-0.63 35 26 0.63-0.71 + ab fl 34 31 0.71-0.76 37 29 0.76-0.82 fl 42 30 0.82-0.86 ab 36 31 0.86-0.89 ab fl 42 29 0.89-1.00 + ab fl 13 a 19 a Blank 38 26

aSignificantly different from control at 95% probability level as determined by LSD tests.

988 ACHHIREDDY ET AL.

r a t ed phy to tox in s is c o n s i s t e n t w i th the e x p e r i m e n t s invo lv ing c r u d e ex t rac t s

(w i thou t T L C m o n i t o r i n g ) (Tables 1, 3 - 5 ) . O n the bas i s of U V and color reac-

t ion da ta , t he phy to tox ic c o m p o u n d s p r e s e n t in l a n t a n a may be phenol ic in na-

ture . W e are c u r r e n t l y i d e n t i f y i n g t he se c h r o m a t o g r a p h i c a l l y sepa ra t ed phy to-

toxic c o m p o u n d ( s ) u s ing H P L C and G C - M S m e t h o d s .

Acknowledgments--The authors thank Dr. J. M. Gardner for his helpful suggestions on TLC. Ms. Heidi J. Pieringer and Mr. H. Kevin Hostler are gratefully acknowledged for their technical assistance.

REFERENCES

ACHHIREDDY, N.R., and SINGH, M. 1984. Allelopathic effects of lantana (Lantana camara) on milkweed vine. Weed Sci. 32:757-761.

BANDEEN, J.D., STEPHENSON, G.R., and COWETT, E.R., 1982. Discovery and distribution of her- bicide-resistant weeds in North America, pp. 9-30, in H.M. LeBaron and J. Gressel (eds.). Herbicide Resistance in Plants. Wiley-Interscience, New York.

GRESSEL, J., AMMON, H.U., FOGELFORS, H., GASQUEZ, J., KAY, Q.O.N., and KEES, H. 1982. Discovery and distribution of herbicide-resistant weeds outside North America, pp. 31-55, in H.M. LeBaron and J. Gressel (eds.). Herbicide Resistance in Plants. Wiley-lnterscience, New York.

GUENZI, W.D. and MCCALLA, T.M. 1966. Phenolic acids in oats, wheat, sorghum, and corn resi- dues and their phytotoxicity. Agron. J. 58:303-304.

HARBORNE, J.B. 1961. Chromatography of phenols, pp. 607-622, in E. Heftmann (ed.). Chroma- tography. Reinhold Publishing Corporation, New York.

JACOBSON, M. 1975. Insecticides from plants. A review of the literature, 1954-1971. ARS Hand- book, p. 461. GPO, Washington, D.C.

KREBS, K.G., HEUSSER, D., and WIMMER, H. 1969. Spray reagents, pp. 854-905, in S. Stahl (ed.). Thin-Layer Chromatography. Springer-Verlag, New York.

LIEBL, R.A., and WORSHAM, A.D. 1983. Inhibition of pitted morning glory (Ipomoea lacunosa L.) and certain other weed species by phytotoxic components of wheat (Triticum aestivum L.) straw. J. Chem. Ecol. 9:1027-1043.

PHILLIPS, R.L., and TUCKER, D.P.H. 1970. Milkweed vine control in Florida citrus groves. Proc. Fla. State Hortic. Soc. 83:26-29.

PHILLIPS, R.L., and TUCKER, D.P.H. 1976. Evaluation of herbicides for lantana control in citrus groves. Proc. Fla. State Hortic. Soc. 89:19-20.

PUTNAM, A.R., and DEFRANK, J. 1983. Use of phytotoxic plant residues for selective weed control. Crop Prot. 2:173-181.

PUTNAM, A.R., and DUKE, W.B. 1978. Allelopathy in agroecosystems. Annu. Rev. Phytopathol. 16:431-451.

RICE, E.L. 1984. Allelopathy, 2rid ed. Academic Press, New York. SANTELMANN, P.W. 1977. Herbicide bioassay, pp. 79-96, in B. Truelove (ed.). Research Methods

in Weed Science. Southern Weed Science Society Publication. Auburn Printing, Inc., Auburn, Alabama.

TANG, C.S., and YOUNG, C.C. 1982. Collection and identification of allelopathic compounds from the undisturbed root system of bigalta limpograss (Hemarthria altissima). Plant Physiol. 69:155-160.

WORSHAM, A.D. 1982. Discussion of topics, pp. 219-236, in R. Charudattan and H.L. Walker (eds.). Biological Control of Weeds with Plant Pathogens. Wiley-Interscience, New York.