Journal of Chemical Ecology, Vol. 17, No. 11, 1991

ANTHRAQUINONES IN DIFFERENT DEVELOPMENTAL STAGES OF Galeruca tanaceti (COLEOPTERA, CHRYSOMELIDAE)

M . H I L K E R 1'* and S. S C H U L Z 2

ILehrstuhlfar Tieroekologie H Universiti~t Bayreuth Postfach 101251, D-8580 Bayreuth, Germany 21nstitutflir Organische Chemie Universitdt Hamburg Martin-Luther-King Platz 6 D-2000 Hamburg 13, Germany (Received May 23, 1991; accepted July 23, 1991) Abstract--The overwintering eggs and the larvae of the leaf beetle Galeruca tanaceti (L.) contain hydroxylated anthraquinones. In both developmental stages, 1,8-dihydroxy-3-methylanthraquinone (= chrysophanol) and 1,8-dihydroxyanthraquinone (= chrysazin) were detected by GC-MS and GC-FTIR analyses. In the eggs, chrysazin was found only in traces. Anthraquinones were also present in ovaries and hemolyrnph of gravid females, which were investigated in order to examine the incorporation of these substances into the eggs. Neither in acidified nor in nonacidified extracts of the host plants Tanacetum vulgare L. and Achillea millefoliumL. were anthraquinones found. The activity of these anthraquinones as chemical defense substances was proved in bioassays with the ant Myrmica ruginodis NYL. Further possible biological significances of anthraquinones are discussed.

Key Words--Galeruca tanaceti, Coleoptera, Chrysomelidae, anthraquinones, eggs, larvae, hemolymph, ovaries, Tanacetum vulgare, Achillea millefolium, feeding deterrence.

INTRODUCTION C h r y s o m e l i d s protect and d e f e n d their e g g s against e n e m i e s by different strategies (Hinton, 1981; L e n g e r k e n , 1954). F e m a l e s o f Phaedon cochleariae, for * To whom correspondence should be addressed. 2323 0098-0331/91/1100-2323506.50/09 1991PlenumPublishingCorporation

2324

HILKER AND SCHULZ

example, hide their eggs in small cavities gnawed into the undersurface of the host-plant leaf. This oviposition behavior possibly serves not only as egg camouflage, but also as protection against desiccation of the eggs. Other chrysomelid species make their eggs almost inaccessible for predators and parasites by covering the eggs with feces and/or secretion. Eggs of Timarcha species, for example, are so firmly embedded in a "nest" consisting of plant material and feces that a single egg can hardly be loosened from the site. Eggs of Clytrinae are coated by a hardened scatoshell consisting of fecal material and gland secretions. Several chrysomelid species provide their eggs with chemicals that act as feeding deterrents against predators. Other chemicals are even toxic for the predators (overview: Pasteels et al., 1988a,b). The eggs of several Chrysomela species contain salicin in amounts that are toxic for ants (Pasteels et al., 1986). Other defensive egg compounds that cause feeding deterrence or rejection in predators are cardenolides in Chrysolina species (Pasteels and Daloze, 1977; Daloze and Pasteels, 1979), isoxazolinone derivatives in several Chrysomela and Phratora species, as well as in Plagiodera versicolora and Gastrophysa viridula (Pasteels et al., 1986), high amounts of oleic acid in Gastrophysa cyanea (Howard et al., 1982a), and cucurbitacins in the eggs of several Galerucinae that feed upon Cucurbitaceae (Ferguson and Metcalf, 1985). Galeruca tanaceti L. (subfamily Galerucinae) is usually found on the Compositae Tanacetum vulgare L. and Achillea millefolium L. (Prevett, 1953). Eggs are deposited in September/October preferably at the tip of grass stalks; they overwinter until next April or May when the larvae hatch. The eggs are laid in batches that vary in size: usually about 20-30 eggs are deposited per batch. Each egg batch is cemented together and covered by a~bright yellow secretion, which is produced by glandular cells of the oviduct (Scherf, 1956, 1966). Within half an hour after egg deposition, the liquid secretion is tanned by a dopa-oxidase and colored black peripherally (Messner, 1983). The hardened secretion, which is unsoluble in organic solvents, may serve as a mechanically protective device. The aims of this study were to examine whether the eggs of G. tanaceti are chemically defended and protected during the long period of overwintering, and to determine the origin of protective compounds. Furthermore, we investigated whether biologically active substances of the eggs are also present in the larvae of G. tanaceti.

METHODS AND MATERIALS

Larvae, eggs, and adults of Galeruca tanaceti (L.) were collected in 1989 and 1990 in the environs of Bayreuth and stored at - 40~ until preparation of extracts. Chemical Analysis. The hardened secretion from 15 egg batches was

ANTHRAQUINONES IN G. tanaceti

2325

removed and the eggs were crushed in acetone, treated by ultrasound, centrifuged, and the clear supernatant fractionated by thin-layer chromatography (TLC) on silical gel 60 F254 (0.2 mm) with hexane-isopropyl acetate-l-butanol (6 : 2 : 1; v/v/v) as eluent. A parallel TLC run of an acetone extract, which was prepared from a single egg batch, showed, after spraying with 6 % methanolic KOH, a bathochromic shift from yellow to purple in the fraction of the Rf value 0.86. The remaining nonsprayed fraction was scraped off the plate, dissolved in acetone, and repeatedly centrifuged for complete removal of the silicagel. The purified TLC fraction was analyzed by GC-MS and GC-FTIR. EI mass spectra (70 eV) were obtained using a Carlo Erba Vega Series 2 gas chromatograph with split injection (split 1 : 5; injector temperature: 220~ coupled to a Finnigan Iontrap ITD 800. A 12 m • 0.32 mm FS-OV-1701 column was used, which was programmed from 100~ to 220~ at 20~ from 220~ to 260~ at 5~ and from 260~ to 280~ at 20~ (carrier: 50 kPa helium). Infrared spectra were obtained by GC-FTIR on a HP 5890 Series II gas chromatograph coupled to a HP 5965 infrared detector (temperature: 250~ with splitless injection at 220~ Samples were separated on a HP1 colunm (25 m x 0.31 mm) with an oven temperature program from 100~ to 240~ at 50~ (carrier: 65 kPa helium). Transmission spectra were recorded from 750 cm-1 to 4000 cm -~. A TLC fractionation of an acetone extract of 20 G. tanaceti larvae (third instar), conducted with the same method as used for the egg extract, showed, after spraying with 6 % methanolic KOH, a bathochromic shift from yellow to purple in the fraction of the Rf value 0.83. The remaining nonsprayed TLC fraction of an acetone extract prepared from 20 G. tanaceti larvae (L3) was purified and analyzed as described before. Hemolymph was obtained from 20 gravid G. tanaceti females by cutting the forelegs at the articulation between coxa and femur and soaking up the emerging hemolymph with small pieces of filter paper, which were extracted in acetone. Then, ovaries were dissected from these females (oviducts were cut off) and also dissolved in acetone. Detailed descriptions and drawings of galerucine genitalia have been presented by Silfverberg (1976) and Suzuki and Yamada (1976). Both hemolymph and ovary extracts were treated by ultrasound, repeatedly centrifuged, and the clear supernatants analyzed by GC-MS. Extracts of host-plant leaves also were investigated. Three leaves each of Tanacetum vulgare L. and Achillea millefolium L. were ground, extracted in acetone, and examined for hydroxylated anthraquinones using the TLC method described above. These are hereafter called nonacidified extracts. The method of Wouters (1985) was modified for preparation of acidified extracts. Fragments of three leaves from each investigated host plant were hydrolyzed in unstoppered reaction tubes with 2 N HCI for 3 hr in a boiling water bath. The hydrolyzed solutions were cleared in a centrifuge and repeatedly extracted with

2326

HILKER AND SCHULZ

diethylether. The combined ether phases were evaporated under N2 to a few milliliters and analyzed by GC-MS. The conditions for the GC-MS analyses of hemolymph, ovary, and acidified host-plant extracts were the same as for the investigation of the egg extracts. Bioassay. A feeding bioassay was conducted with the ant Myrmica ruginodis NYL., in order to examine whether eggs of G. tanaceti were avoided by predators. Twenty ants were placed in a Petri dish (14 cm diam.) and starved for two days. After this starvation period, an aqueous test suspension was prepared by crushing two egg masses in 500 #1 H20. A mealworm (last instar) was ground in 500 #1 H20 for the control suspension. Five microliters of each suspension was offered simultaneously to the ants. Every minute the number of feeding ants at the test and control suspension was recorded for a period of 10 min. After 20 replications, the bioassay was statistically evaluated by the Wilcoxon signed-rank test for paired differences (Sachs, 1984). The same method was used for testing an aqueous mealworm suspension plus synthetic chrysophanol (test) against a "pure" aqueous mealworm suspension without chrysophanol (control) to examine the effect of chrysophanol on the feeding behavior of M. ruginodis. The chrysophanol concentration of the test suspension was 1.4 x 10 -~ M and 0.6% of the mealworm weight (0.3 g/500 ~1 H20).

RESULTS

The TLC of an acetone extract of G. tanaceti eggs revealed a fraction at Ry 0.86 with a bathochromic shift from yellow to purple when sprayed with 6 % methanolic KOH. A bathochromic shift in alkaline solution is reported for a great many hydroxyquinones (Thomson, 1976). The total ion current chromatogram of this (nonsprayed) fraction revealed two peaks with mass spectra that show the typical fragmentation patterns of anthraquinones (Figure 1A). The EI mass spectrum of compound 1 had characteristic ions at m/z 240 (100%, M+), 212 (36%, M - CO), 184 (47%, M - 2CO), 155 (9%), 138 (21%), 128 (30 %), 92 (24 %), indicating 1,8-dihydroxyanthraquinone (= chrysazin) (Howard et al., 1982b; McLafferty and Stauffer, 1989). A synthetic reference sample (Roth AG, Basel, Switzerland) showed an identical mass spectrum and the same retention time as compound 1. The EI mass spectrum of compound 2 reVealed fragments at m/z 254 (100%, M+), 226 (26%, M - CO), 198 (18%, M 2CO), 169 (7%), 152 (20%), 141 (10%), 115 (19%), suggesting 1,8-dihydroxy-3-methylanthraquinone (= chrysophanol), which was confirmed upon comparison with a synthetic reference sample (Aldrich Chemical Co., Steinheim, Germany) (Howard et al., 1982b; McLafferty and Stauffer, 1989). In addition, the FTIR spectrum of compound 2 conformed with that of synthetic chrysophanol (Figure 2).

ANTHRAQUINONES

2327

G. tanaceti

IN

100-90-

f

2

8070. 60

-

50 40

~o

A/EGGS 1

2O 10

400

,I,i,IFI,I'I*

i,l,l,l,l,l,l,t,l,l,l'l 450

550

,500

| 600

SCAN NO.

100 - 90

--

80

--

70

--

f

1

f

2

60-50-40-30--

BL /ARVAE ~

~

20--

lo - _ 400

~

~ 450

500

550

600

SCAN NO.

FIG. 1. Total ion current chromatogram of TLC fraction of 15 Galeruca tanaceti egg masses (A) and 20 larvae (B) from scan number 400 to 600. Characteristic fragments of the mass spectra of peak 1 and 2 are given in the text.

The site of production of the anthraquinones may be microorganisms living on deposited eggs, the embryo itself, or the female, which incorporates anthraquinones into the eggs. The GC-MS analysis of the hemolymph of G. tanaceti females revealed the presence of chrysophanol and traces of chrysazin. Chrysophanol was also detected in the ovaries; chrysazin, however, was not found. In the TLC of the nonacidified acetone extracts of T. vulgare and A. millefolium, no bathochrornic shift occurred. GC-MS analyses of acidified extracts, which were examined in order to check for possible glycosidated chrysophanol and chrysazin derivatives, also showed no anthraquinones in the plant material.

2328

HILKER AND SCHULZ 100.0

99.9 Z O

99.8

ffl Z < re p-

99.7

99.6

OH O OH

Z i.i

~

99.5 99.4

0 ' 2000

'

'

l i t

i

~

i

I

WAVENUMBER

I I 1400

I . (cm -~)

I

I'

~ I

j

'

'

I 800

FIG. 2. FTIR spectrum of compound 2 from Figure 1A, in all respects identical with the spectrum of chrysophanol.

Therefore, the females do not acquire these substances from their host plants, since no hydroxylated anthraquinones were detected in the leaf extracts. The TLC of an acetone extract of G. tanaceti larvae showed, after spraying, a bathochromic shift from bright yellow to purple in the fraction ofRy0.83. A GC-MS analysis of the respective nonsprayed fraction revealed the presence of the same anthraquinones as in the eggs. The total ion current chromatogram of this TLC fraction is shown in Figure lB. The mass spectra of these peaks were the same as those described above for the egg extract. A comparison of the FTIR spectrum of synthetic chrysazin (Roth AG) and of compound 1 from the total ion current chromatogram of the larval extract showed no differences (Figure 3). The feeding bioassays with M. ruginodis revealed that the ants significantly preferred feeding upon mealworms compared to feeding upon the eggs of G. tanaceti (Figure 4). The highest difference between the number of feeding ants upon eggs and mealworms was reached after a test period of 10 min: only 27.5 % of the feeding ants were recorded at the eggs of G. tanaceti. In order to examine whether this deterring activity of G. tanaceti eggs is due to the detected anthraquinones, a second feeding bioassay was conducted which tested feeding of M. ruginodis upon mealworms plus chrysophanol (test) compared to feeding upon " p u r e " mealworms (control). Figure 5 shows that after a test period of 8, 9, and 10 min, significantly fewer ants fed upon the test than upon the control suspension. The highest significant difference between the number of feeding

ANTHRAQUINONESIN G. tanaceti

2329

100.0

99.6

-

Z 0 99.2 < t-

98.8--

OH O OH 98.4 -m

98,0

0

- -

I

'

'

i'i

'

~ ~

2OOO WAVENUMBER

I ' I I 1400 (cm -1)

'

'

i

I

'

i

]

r 800

FIG. 3. FTIR spectrum of compound 1 from Figure 1B, in all respects identical with the spectrum of chrysazin. ants upon test and control suspension was attained after 8 rain: 35.8% of the feeding ants were recorded at the mealworrn suspension plus chrysophanol. DISCUSSION Up to now, anthraquinones were unknown in eggs of Coleoptera; however, they have been detected in chrysomelid larvae. Howard et al. (1982b) found chrysophanol, chrysazin, and two anthrones in larvae of the elm leaf beetle, Pyrrhalta luteola, which, like G. tanaceti, belongs to the subfamily Galerucinae. Both chrysomelid species obviously do not acquire the anthraquinones from their host plants. As suggested for the production of anthraquinones in scale insects (Kayser, 1985), symbiontic microorganisms may also produce these compounds in both chrysomelid species. In G. tanaceti, the presence of anthraquinones in the hemolymph of females and in the ovaries indicates that females incorporate these compounds into the eggs, thus providing the eggs with protective compounds. Eisner et al. (1980) reported on the feeding deterrence of the anthraquinone glucoside carminic acid against the ant Monomoriurn destructor, whereas the carnivorous caterpillar of a pyralid moth, Laetilia coccidivora, is not deterred by carminic acid; rather, it utilizes the ingested carminic acid from prey cochineals for defensive purposes of its own. As demonstrated by the bioassay described above, feeding of ants upon G. tanaceti eggs is significantly reduced compared to feeding upon normally accepted mealworms. An additional bioas-

2330

HILKER AND SCHULZ 100 - -

137

n = 99

130

111

124

123

11,~

N,S.

112

***

**

98

91

*

***

0

10

Z < Z

50

~3

LU LU it

10

i

2

3

4

II

5

8

M INUTES

FIG. 4. Feeding bioassay with the ant Myrmica ruginodis. Black bars: feeding activity on an aqueous test suspension of Galeruca tanaceti eggs (2 egg masses/500/zl H20). White bars: feeding activity on simultaneously offered aqueous control suspension of mealworms (1 last-instar mealworm/500/xl H20). Registration of feeding ants during a test period of 10 min. Asterisks indicate level of significance: *, P < 0.05; **, P < 0.01; ***, P < 0.001; n.s. = not significant; Wilcoxon signed-rank test for paired differences. 100

"1 "1 n = 91 -i

128

141

147

137

129

123

117

112

n=s.

105 n=s.

n.s.

n.s.

n.s,

n.s.

n.s.

**

*

*

1

2

3

4

5

6

7

8

9

10

zo

~ MINUTES

FIG. 5. Feeding bioassay with the ant Myrmica ruginodis. Black bars: feeding activity on an aqueous test suspension of mealworms plus added synthetic chrysophanol (1.4 • 10 -2 M). White bars: feeding activity on a simultaneously offered aqueous control suspension of mealworms without chrysophanol. Registration of feeding ants during a test period of 10 min. Asterisks indicate level of significance: *, P < 0.05; **, P _< 0.01; n.s. = not significant; Wilcoxon signed-rank test for paired differences.

say revealed that synthetic chrysophanol also causes a diminished feeding activity o f the ants; however, the feeding reduction in the bioassay with eggs was stronger (see Figures 4 and 5). These results indicate that the main anthraquinone in the eggs o f G. tanaceti contributes to the feeding deterrence o f the eggs

ANTHRAQUINONES IN G. tanaceti

2331

against ants. Detailed knowledge on the anthraquinone concentrations in the eggs is essential to assess exactly the role of anthraquinones in the feedingdeterring activity of the eggs. In addition to the dosage, the mixture of the detected anthraquinones may also be relevant when bioassaying for feeding deterrence. Further quantitative investigations of the anthraquinones in G. tana c e t i will be necessary to elucidate their biological significance. In this qualitative study, only a rough estimation of the chrysophanol concentrations was possible: the chrysophanol content is lowest in the hemolymph and ovaries of a single female, increases in the larva, and is highest in one egg mass. When tasting an insect egg, a predator destroys the tiny prey. Therefore, the deposition of distasteful eggs in batches seems to be advantageous, since this oviposition behavior enhances the chance that only a single egg of an egg mass is damaged while the others remain untouched (Stamp, 1980). Anthraquinones from the above-mentioned chrysomelid larvae can only be released and display their activity against enemies by the emerging hemolymph from wounds. It remains to be proved whether the chrysomelid larvae are able to survive slight wounding. Wiklund and Jiirvi (1982, p. 1001) demonstrated that naive bird predators "often release distasteful prey insects unharmed after tasting them." In addition to the feeding-deterring activity against ants, anthraquinones are also known as repellents against birds. For the protection of seeds, 9,10anthraquinone is sold commercially as an avian repellent. Furthermore, hydroxylated anthraquinones have been demonstrated to act as antimicrobial agents (Cudlin et al., 1976) and might protect the eggs against microbial diseases. Adewunmi and Adesogan (1984) report on the molluscicidal activity of anthraquinones against snails. The ingestion of trematod eggs by herbivorous snails is a well-known phenomenon, whereas the coincident uptake of insect eggs and plant material, to our knowledge, has not been examined. Acknowledgments--Many thanks are due to Konrad Dettner, Uwe Noldt and Regina FettkOther for valuable contributions and critical review of the manuscript.

REFERENCES ADEWUNMI,C.O., and ADESOGAN,E.K. 1984. Anthraquinonesand oruwacin fromMorinda lucida as possible agents in fascioliasisand schistosomiasiscontrol. FitoterapM 55:259-263.

CUDLIN,J., BLUMAUEROVA,i . , STEINEROVA,N., MATEIU,J., and ZALABAK,V. 1976. Biological activity of hydroxyanthraquinonesand their glucosides toward microorganisms.Folia Microbiol. 21:54-57. DALOZE,D., and PASTEELS,J.M. 1979. Production of cardiac glycosidesby chrysomelidbeetles and larvae. J. Chem. Ecol. 5:63-77. EISNER,T., Nowlcrd, S., GOETZ,M., and MEINWALD,J. 1980. Red cochinealdye (carminicacid): Its role in nature. Science 208:1039-1042.

2332

HILKER AND SCHULZ

FERGUSON, J.E., and METCALF, R.L. 1985. Cucurbitacins. Plant-derived defense compounds for diabroticites (Coleoptera: Chrysomelidae). J. Chem. Ecol. 11:311-317. HINTON, H.E. 1981. Biology of Insect Eggs, Vol. II. Pergamon Press, Oxford. HOWARD, D.F., BLUM, M.S., JONES, T.H., and PHILLIPS, D.W. 1982a. Defensive adaptations of eggs and adults of Gastrophysa cyanea (Coleoptera: Chrysomelidae). J. Chem. Ecol. 8:453462. HOWARD, D.F., PHILLIPS, D.W., JONES, T.H., and BLUM, M.S. 1982b. Anthraquinones and anthrones: Occurrence and defensive function in a chrysomelid beetle. Naturwissenschaften 69:91-92. KAYSER,n . 1985. Pigments, pp. 367-415, in G.A. Kerkut and L.I. Gilbert (eds.). Comprehensive Insect Physiology, Biochemistry and Pharmacology, Vol. 10. Pergamon Press, Oxford. LENGERKEN, H. VON. 1954. Die Brutfiirsorge- und Brutpflegeinstinkte der Kfifer. Geest & PortiA K.-G., Leipzig. MCLAFFERTY, F.W., and STAUFFER, D.B. 1989. The Wiley/NBS Registry of Mass Spectral Data. Wiley, New York. MESSNER, B. 1983. Dopa-Oxidase-geh~irtete Sekrete schiitzen das Eigelege von Galeruca tanaceti L. (Coleoptera, Chrysomelidae). Entomol. Nachr. Berichte 27:221-224. PASTEELS, J.M., and DALOZE, O. 1977. Cardiac glycosides in the defensive secretion of chrysomelid beetles: Evidence for their production by the insects. Science 197:70-72. PASTEELS, J.M., DALOZE, D., and ROWELL-RAHIER, M. 1986. Chemical defence in chrysomelid eggs and neonate larvae. Physiol. Entomol. 11:29-37. PASTEELS, J.M., BRAEKMAN,J.-C., and DALOZE, D. 1988a. Chemical defense in the Chrysomelidae, pp. 233-252, in P. Jolivet, E. Petipierre, and T.H. Hsiao (eds.). Biology of Chrysomelidae. Kluwer Academic Publishers, Dordrecht. PASTEELS, J.M., ROWELL-RAHIER, M., and RAUPP, M.J. 1988b. Plant-derived defense in chrysomelid beetles, pp. 235-272, in P. Barbosa and D. Letourneau (eds.). Novel Aspects of InsectPlant Interactions. Wiley, New York. PREVETT, P.F. 1953. Notes on the feeding habit and life-history of Galeruca tanaceti L. (Co1., Chrysomelidae). Entomol. Mon. Mag. 89:292-293. SACHS, L. 1984. Angewandte Statistik. Springer-Verlag, Berlin. SCHERF,H. 1956. Zum feineren Bau der Eigelege von Galeruca tanaceti L. (Coleopt., Chrysom.). Zool. Anz. 157:124-130. SCHERF,H. 1966. Beobachtungen an Ei und Gelege von Galeruca tanaceti L. (Coleoptera, Chrysomelidae). Biol. Zentralbl. 85:7-17. SILFVERBERG,H. 1976. Studies on gaierucine genitalia I (Coleoptem, Chrysomelidae). Not. Entotool. 56:1-9. STAMP,N.E. 1980. Egg deposition patterns in butterflies: Why do some species cluster their eggs rather than deposit them singly? Am. Nat. 115:367-380. SUZUKI, K., and YAMADA,K. 1976. Intraspecific variation of ovadole number in some chrysomelid species (Coleoptera, Chrysomelidae). Kontyu 44:77-84. THOMSON, R.H. 1976. Isolation and identification of quinones, pp. 207-232, in T.W. Goodwin (e~.). Chemistry and Biochemistry of Plant Pigments, 2rid ed. Academic Press, London. WIKLUND,C., and JARVI, T. 1982. Survival of distasteful insects after being attacked by naive birds: A reappraisal of the theory of aposematic coloration evolving through individual selection. Evolution 36:998-1002. WOUTERS, J. 1985. High performance liquid chromatography of anthmquinones: Analysis of plant and insect extracts and dyed textiles. Stud. Conserv. 30:119-128.