Comp. Biochem. Physiol., 1975, Vol. 50B, pp. 77 to 82. PergamonPress. Printed in Great Britain

PHEROMONE METABOLISM IN MALE PSEUDALETIA SEPARATA (WALK.) AND MAMESTRA CONFIGURATA (WALK.) (LEPIDOPTERA: NOCTUIDAE) J. R. CLEARWATER Department of Entomology, University of Alberta, Edmonton, Alberta T6G 2E3, Canada (Received 10 September 1973)

Abstract--1. Benzyl and phenylethyl glycosides were isolated from Stobbe's glands of Pseudaletia separata and Mamestra eonfigurata.

2. These compounds appear to be storage materials releasing an aromatic pheromone upon hydrolysis. 3. Phenyl ethanol was extracted from the hairpencils of M. configurata, and benzaldehyde has been previously reported from P. separata.

INTRODUCTION

Preliminary work on the extraction of the glycoside from M. configurata was done using diapause pupae that had

MANY compounds of low molecular weight have been isolated from the abdominal hairpencils of various male noctuids (Aplin & Birch, 1968, 1970; Clearwater, 1971, 1972; Grant et al., 1972). The fine elongated scales of these hairpencils (Fig. 1), when spread, can effectively disperse these odoriferous substances (Mtiller, 1877). The intricate diamond-shaped microsculpture of these scales in Phlogophora meticulosa (L.) probably increases their evaporating surface (Birch, 1970). Each of a pair of large abdominal glands (Stobbe, 1912) (S.G.) releases its product into a long abdominal pouch (P) via a duct (D1) consisting of several elongated scales enclosed within a fold of abdominal cuticle (Fig. 1). The hairpencil (D2) is found near the posterior edge of sternite II (S II) in the pharate adult and only comes in contact with the gland product when it is retracted into the pouch several hours after eclosion. Excision of Stobbe's glands from living moths showed that they secrete the species specific scent in Phlogophora meticulosa (Aplin & Birch, 1968). The nature of the precursors of these scents remain unknown, although Birch (1970) has suggested that a fl-glycoside is produced preceding the formation of benzaldehyde in Leucania impura (L.). (Benzaldehyde is also the male pheromone of Pseudaletia separata (Clearwater, 1972).)

recommenced development following injection of 0.51.0/~g of/3-ecdysone.

N.

"..~.G. M

C. S.~ "

"~.. t,

:'".,.

~

,;

B.S.

:-.".:"

x, ...... .

Fig. 1. P, separata. Left half of male pheromone system, soon after adult emergence. B.S., Basal plate into which the hair scales insert; C, core of Stobbe's gland from which fine elongated scales emerge to form the gland duct (D1); Ds, area of highly dissected microsculpture; the evaporation surface of the halrpencil; L, lever, the hairpencil support; M, overlying muscle layer which spreads the scales of the hairpencil; N, large platelike nucleus of Stobbe's gland cell; S, line of central incision through which gland was removed; S H, abdominal sternite II; T, darkened transition area of hairpencil; V, lumen of gland cell filled with vesicles.

MATERIALS AND METHODS

Experimental animals

Pupae of Mamestra configurata were collected from field infestations of adventitious rape near Sturgeon, Alberta, Canada. Pupae of Pseudaletia separata were collected from Napier, New Zealand, by Mr. F. Agnew. 77

78

J . R . CLEARWATER

Chemistry of the pheromone from M. configurata

Structure of the hairpencil cap cells

Extraction of the hairpencils with methylene chloride was carried out between 2.00 and 4.00 a.m. ; about 3040 hr after eclosion. The time chosen for extraction was less than 1 hr before the beginning of mating activity, when more than 90 per cent of the males had completed retraction. Less than 5 per cent of the male moths had retracted their hair-pencils into their abdominal pouches when examined 10 hr after eclosion. One hundred and twelve males were used to prepare three solutions containing sixty to seventy hairpencils/ml. All gas chromatographic analyses of these crude extracts were carried out on glass columns since stainless steel appeared to cause degradation of the major component of the extracts. Two 6 ft glass columns were used; 3% OV-17 on Chromosorb G and 3% XE60 on Chromosorb W. The instability of this component also precluded the use of coupled gas-liquid chromatographymass spectrometry (G.C.M.S.) so that crude extract was evaporated from the tip of a direct probe. Great care was taken to prevent contamination. All syringes were thoroughly rinsed with solvent, and checked with the gas chromatograph at the highest sensitivity used.

The morphology of the base of the hairpencil, an area thought likely to produce a fl-glycosidase, was investigated. The basal plate of the hairpencil was dissected from both species, fixed in either 3% glutaraldehyde (Millonigs, 1961) or 3 ~ formaldehyde and 2% glutaraldehyde (0.001 M Mg 2+, 0-1 M PO4 a- pH 7.2) post fixed in buffered OsO4, dehydrated in ethanol and embedded in either Epon or Araldite. Sections were double stained in aqueous uranyl acetate and Reynolds (1963) lead citrate before examination with a Philips EM 200 electron microscope at 60 kV.

Chemistry of the precursors from the Stobbe gland Small numbers of whole clean glands from advanced pharate adults were used for the isolation procedure. A ventral cut (S) was made in the second sternite (S II) and the cut edges reflected laterally so that the duct scales (DO could be grasped with microforceps where the duct emerges from the gland (Fig. 1). Slight tension freed the gland from the enveloping fat body. Extraction was carried out in micro-separating flasks made from Pasteur pipettes. For each species two to ten glands were placed in 10/zl of distilled water in a flask and crushed with the beaded tip of a glass thread. Ten/zl of chloroform were added and the extract stirred to complete partition of the components between the two layers. Strips of silica gel sheet with fluorescent indicator (Eastman Kodak No. 6060) were spotted with 1 t~l of either crude extract, aqueous phase, non-polar phase or standard solutions, and run at room temperatures in nbutanol-acetic acid-water ( 4 : 1 : 1 ) . The strips were examined under u.v. illumination. Crude extracts and the aqueous layers were analyzed on an AEI M.S. 12 mass spectrometer obtaining chemical ionization with NH3 reagent gas. In vitro hydrolysis of the precursors Glands dissected from both species were incubated in a solution of fl-glucosidase. One small crystal of emulsin (Nutritional Biochemicals Corp.) was dissolved in 20 td of 0.08M acetate buffer (pH5"0) in which 6-15 glands had been crushed, and left for 24 hr at room temperature. Controls, lacking only the enzyme, were treated in an identical manner. The reaction was stopped by the addition of 20 or 30/~1 of methylene chloride and the compounds partitioning into the non-polar phase were chromatographed on 3% OV-17 and 5% E.F.A.P. columns. Coupled G.C. mass spectrometry (AEI M.S 12 with a Varian Aerograph 1400 G.C.), was used to identify a major component of the extract from P. separata, while a direct probe was used for that of M. configurata.

RESULTS

Chemistry o f the male pheromone of M. configurata Analysis of the hairpencil extract showed the presence of three peaks, the first of which was very m u c h larger than the two following ones (Fig. 2A).

Peak I ( 1'4 rain)

(4~'min fo 200 °) P

e

(i) Phenyl ethanol stondord

o

Peok /11"(12rnin)~ / k ~

(ii)Glonds + emulsin (0.5/.~1 of non-polor layer)

(iii) Glands only {0.5~.1 of non polar layer)

m,n,

Fig. 2. M. configurata GLC evidence for identification of pheromone and glycoside hydrolysis product with phenyl ethanol. A. Hairpencil extract 60hp/ml OV-17 glass 6 ft x ~ in. T = 125°C. A = 10 × 16 Perkin-Elmer 990 GC. B. OV-17 glass 6 f t × ~ i n . T = 1 1 0 ° C . A = l x l 0 a Beckman GC 5. (i) Standard phenyl ethanol; (ii) fifteen glands in 20 tzl buffered emulsin solution 30/~1 MeC12 extraction; (iii) four glands in 20M buffer. 20 tzl MeCI2 extraction. The major c o m p o n e n t of this extract had retention times identical to phenyl ethanol on two columns (Table 1). W h e n standard phenyl ethanol was added to extracts of the hairpencil complete superposition of the peaks occurred, confirming the fact that standard and u n k n o w n had the same retention time. Complete loss o f the m a j o r peak on stainless steel columns could also be duplicated with standard phenyl ethanol at dilutions comparable to that of the extract.

Pheromone metabolism in Noctuidae Table 1. Gas-liquid chromatography of the pheromone of M. configurata Retention Temperature (rain) Column (°C)

Material Hairpencil extract Standard phenyl ethanol Hairpencil extract Standard phenyl ethanol

1.9 1'9

OV-17 OV-17

110 110

4.4 4.4

XE60 XE60

170 170

The mass spectrum of the extract (Fig. 3A) contained peaks of mass 91, 92 and 122. The heights of these peaks were in the same ratio as those of standard phenyl ethanol (Fig. 3C). Peaks of mass 104, 105, 128, 130, 132, 133 and 149 were often present, but peak heights held no constant relationship to the peaks of mass 91, 92, and 122. Peaks of mass 84 and 86 were from methylene chloride. N o attempt was made to identify the two remaining materials found in trace amounts.

79

in both species. The RI of the spot from both steps of the extraction procedure in P. separata was 0.70, identical with authentic benzyl/~-v-glucoside. The compound from M. configurata had an R! = 0.71, close to P. separata and standard benzyl/~-D-glUCOside, while other standards run had greater values, e.g. isopropyl /3-I)-glucoside RI = 0.79. The chloroform layer from P. separata also showed a spot (Ry = 0.71) that turned pale lemon yellow on exposure to light for 24 hr. Mass spectra from both crude and aqueous layers from both species were free of significant contamination (Fig. 4). The base peak of the compound from M. configurata had a mass of 3 0 2 ( M + N H 4 +) corresponding to a molecular weight of 284. The molecular weight of the compound from P. separata was 270. Both spectra show a significant peak at 180, which corresponds to the unsubstituted hexose fragment of the molecule (M + NH~+-ROH) (Hogg & Nagabhushan, 1972). The spectra of standard benzyl /3-D-glucoside showed three major peaks at 180, 270 and 288, corresponding to those from the spectrum of P. separata. Neither benzaldehyde nor phenyl ethanol could be extracted from these glands.

IOO

I(X1_

-a

302.

Momesfro configurofo

ST=180%

Ii

o

IOC

91

Ill

Incubotion mixture

h,I

?

I

:>.

t0o!

288 Pseudolefio sepofofo

92

122

_o d Ih

., ,I

h.

ST=215°C 180 270

I

I00 --

o

Phenyl ethonol

:>.

g~

g >=

~oo

rY

=>

I

I !

288 Benzy1,8 D Glucoside

ST= 150°C 70

Jl,

80

h

90

t 'h

I00

t

IlO

I[ i I 120 130 140 150

xO-I

180

I I 1

f

o

m/e

Fig. 3. M. configurata. A. Crude extract of hairpencil 60 hp/ml. B. Hydrolysis product from fifteen glands/ 20 p,l buffered emulsin. C. Standard phenyl ethanol. AEI M.S. 9.

Chemistry of the precursors from the Stobbe gland Crude extracts and aqueous layers from the partition step showed the presence of a single large spot

1

160

II

I

I

200

240

28O

520

role Fig. 4. Mass spectra of the glycosides. A.M. conf~urata aqueous layer, two glands in 10 pl distilled water. B. P. separata aqueous layer, ten glands in 10/d distilled water. C. Standard benzyl ~-v-glucoside. (Note differences in source temperature affecting ratio of fragments to parent peaks.) AEI M.S. 12 with NH a reagent gas.

80

J. R, CLEARWATER

In vitro hydrolysis o f the precursors At least two compounds could be extracted by methylene chloride from the incubation medium containing glands from P. separata (Fig. 5A). The first peak had a comparable retention time to benzaldehyde, but they were not identical. Both peak I and benzaldehyde appeared as separate peaks when the standard solution was mixed with the test. In contrast peak II showed complete superimposition with benzyl alcohol. The mass spectrum of peak II corresponded with benzyl alcohol (Fig. 5B).

Standard

Incubation mixture

BenzaIdehyde(I-Omird

heavily sclerotized lever (L) (Fig. 1). At the base of each socket is a large cell (Fig. 6). A three- to four-chambered end apparatus (E) separates the hollow scale lumen from a microvilli lined cell lumen (L). The tips of the microvilli cluster around a ductule extending proximally from the end apparatus. Table 2. Accurate masses of the hydrolysis product from Stobbe's gland in M. configurata Nominal mass

Accurate mass

Calculated mass

122 92 91

122.0737 92.0617 91.0548

122.0732 92-0626 91.0548

Direct probe on A.E.I.M.S. 9.

okl (0.7rain)

In 3/1. configurata only, a granular zone of cytoplasm immediately adjacent to the lumen is filled with radially orientated microtubules. The nucleus (N) is large and heterochromatic.

~ Benzyl alcohol M~3.7m)

Mass spectrum of PeokIT

+. M.

I00 r

'i ......i~i~'::!i~"i'77~' ii ,

HO],~'~ 1 0 8 ,.~,~.~... ,.

79

©co;',, il

" ,,f " 4 ' . '

°" . .--"I '

-Jo.5 ,',

~..~~,~ : . : :

' .'

•

70

80

90

IO0

llO

120

L.

,,

#~

.~. • "~" • :~ ~; -, ~

...:~ E.

m/e Fig. 5. P. separata. Hydrolysis of glycoside. A. (i) Standard benzaldehyde and benzyl alcohol. (ii) Methylene chloride layer of six glands in 10 tL1 buffered emulsin. FFAP (5%) 6 f t × { i n . stainless steel. T=134°C. A = 5 × 108. B. Mass spectrum of peak II AEI M.S. 12 with coupled Varian Aerograph. 1400 G.C. The single detectable peak obtained from the same experiment with M. configurata had the same retention time as phenyl ethanol (Fig. 2B, ii). Barely discernible amounts could be extracted from the control. The mass spectrum was relatively free from contaminants and provided a useful comparison with the hairpencil extracts. This preparation was used to establish the accurate masses of the main peaks (Table 2). Structure o f the hairpencil cap cells The scales of the hairpencil insert into sockets in a small plate (B.S.) attached to the end of a long

St Fig. 6. P. separata. Section through basal plate of hairpencil. ~ , Ten/~; C, central core of microvilli around a cuticular extrusion; E, three-chambered end apparatus; L, lumen packed with microvilli with granular contents; M, overlying muscle layer; N, large heterochromatic nucleus; S, socket of lamellated cuticle.

DISCUSSION In common with the defensive secretions of Eleodes longicollis and Tribolium castaneum (Coleaptera: Tenebrionidae) (Happ, 1968) the male pheromones of M. configurata and P. separata appear to be stored as glycosides. Like the p-benzoquinones of these two beetles, benzaldehyde and phenyl ethanol are toxic to cell metabolism. Storage as a glycoside appears to have circumvented the problems of insolubility and toxicity.

Pheromone metabolism in Noctuidae The two glycosides from these two noctuids are very similar possibly differing only by a single C1 unit in the side chain linking the phenol and the sugar groups. (The sugars of these glycosides were not identified, though both are unsubstituted hexoses.) Alpin & Birch (1970) identified both benzaldehyde and phenyl ethanol from Polia nebulosa and Mamestra persicariae. The presence of both compounds in a single species suggests that only slight modifications of the metabolic pathway or slightly different precursors may be necessary for their production. Benzaldehyde is an extremely common pheromone, being found in Leucania impura, L. conigera, L. pallens, P. nebulosa, M. persicariae from Britain (Aplin & Birch, 1970), Pseudaletia unipuncta from North America (Grant et al., 1972) and Melanchra lignana, M. omoplaca, M. ustistriga, P. separata, Persectania aversa and P. steropastis from New Zealand (Clearwater, 1971). Selection may be favouring alteration of the pheromone chemistry toward other aromatics such as 2phenyl ethanol, or 3-methoxy 4 hydroxy benzaldehyde tentatively identified from Erana graminosa (Clearwater, 1971). Glycosides are stable compounds, an important property for a storage material. In P. separata, Stobbe's gland seems to begin releasing precursor 1 day after eclosion (Clearwater & Sarafis, 1973), while the amount of free aromatic on the hairpencil begins to rapidly increase on the following day (Clearwater, 1972). Cold standard solutions of benzyl fl-D-glucoside remained stable indefinitely suggesting that the conversion of glycoside to free aromatic is enzymatically mediated. It is expected CHa0H

M. configurata

eosMase(frombasal cells~)

CHtCH~OH + sugar

O

CH,OH

Oc .o P. separafa J). unipuncta

[

N~lycos;dase (frombasal ceils?)

Q

CHsOH + sugar

(7)

C

phenyl ethanol

D~¢hyd~gcau¢ ? be~zyl alcohol CHO

benzaldehyde

Fig. 7. Proposed schemefor degradation of the glycosides.

81

that the enzyme from M. configurata will possess properties similar to emulsin as emulsin is able to release the definitive pheromone from the glycoside. Is it possible that the enzyme of P. separata is also comparable to emulsin7 Aplin & Birch (1970) and Grant et al. (1972) have both identified benzyl alcohol in male noctuids. Grant et al. (1972) showed that male P. unipuncta carry large (11.31 pg/brush) amounts of benzyl alcohol and little (3.40/~g/brush) benzaldehyde immediately following eclosion, while on the second day the proportions of the two compounds are reversed. The following pathway is proposed for the two species of Pseudaletia (Fig. 7). Alternatively, a direct conversion to benzaldehyde may be possible. Birch (personal communication) has used emulsin to produce benzaldehyde from a preparation of glands from L. impura. It is suggested that the cap cells at the base of the hairpencil produce the enzyme responsible for the degradation of the glycoside. The large Stobbe's gland is unlikely to manufacture this enzyme since production of enzyme and glycoside appear to occur too close together in time to preclude interaction within the single duct system (Fig. 1, D0. The Stobbe gland cells contain large numbers of clear vesicles (Clearwater & Sarafis, 1973) while the microvilli of the lumen in the cap cell are packed with a granular material, observations that are consistent with the claim that the former contain a glycoside while the latter may contain a protein. Cells with large nuclei and large microvilli--lined lumina have been described often and suggested as the site of pheromone production in many families of Lepidoptera (e.g. Danaus gilippus benenice 0Nymphalidae) (Pliske & Salpeter, 1971), Manduca sexta (Sphingidae) (Grant & Eaton, 1973) and Trichoplusia ni (Noctuidae) (Grant, 1971)). Interaction of the enzyme and glycoside may take place on the transition area (T) of the hairpencil, a distinctly darker portion of the golden coloured hairpencil (Fig. 1). In all dissections the tip of the duct (DO makes contact with the deep brown area of the hairpencil. This suggestion has precedent-a cuticular reaction site was demonstrated in a system producing p-benzoquinones (Happ, 1968). Acknowledgements--I wish to thank the many staff members of the University of Alberta who assisted with this investigation especially Mr. T. Budd and J. Olekszyk (Chemistry) who produced the mass spectra, and Mr. D. Odynski for assistance with use of gas chromatographs. Professors A. M. Hogg (Chemistry), R. A. Locock (Pharmacy) and M. C. Birch (University of California, Davis) have contributed much helpful discussion and advice, while ProfessorsB. S. Heming and R. H. Gooding read and criticized a preliminary draft of this report. Gifts of fl-ecdysone from Dr. G. Russell (D.S.I.R., Palmerston North, New Zealand) and a collection of P. separata by Mr. F. Agneware gratefullyacknowledged. This work was supported by the National Research Council of Canada (Grant No. A5756 to B. S. Heming).

82

J . R . CLEARWATER REFERENCES

APUN R. T. & BIRCHM. C. (1968) Pheromones from the abdominal brushes of male noctuid Lepidoptera. Nature, Lond. 217, 1167-1168. APLIN R. T. & BIRCH M. C. (1970) Identification of odourous compounds from male Lepidoptera. Experientia 26, 1193-1194. BIRCH M. C. (1970) Structure and function of the pheromone-producing brush-organs in males of Phlogoplora meticulosa (L.) (Lepidoptera: Noctuidae). Trans. R. ent. Soc. Lond. 122, 277-292. CLEARWATERJ. R. (1971) The role of the male-produced pheromone in the reproduction behavior of the southern armyworm Pseudaletia separata (Walk.). Section D. Comparative biochemistry and morphology of the male pheromone system in the Noctuidae. M.Sc. thesis, Massey University, Palmerston North, New Zealand. CLEARWATERJ. R. (1972) Chemistry and function of a pheromone produced by the male of the southern armyworm Pseudaletia separata. J. Insect Physiol. 18, 781789. CLEARWATERJ. R. & SARArTSV. (1973) The secretory cycle of a gland involved in pheromone production in the noctuid moth Pseudaletia separata. J. Insect PhysioL 19, 19-28. GRAYr G. G. (1971) Scent apparatus of the male cabbage looper Trichoplusia ni. Ann. ent. Soc. Am. 64, 347-352. GRANT G. G., BRADYU. E. & BRANDJ. M. (1972) Male armyworm scent brush secretion identification and electroantermogram study of major components. Ann. ent. Soc. Am. 65, 1224-1227.

GRANT G. G. & EATONJ. L. (1973) Scent brushes of the male tobacco hornworm Manduca sexta (Lepidoptera: Sphingidae). Ann. ent. Soc. Am. 66, 901-904. HAPP G. i . (1968) Quinone and hydrocarbon production in the defensive glands of Eleodes longieollis and Tribolium castaneaum. J. Insect Physiol. 14, 1821-1837. HOGG A. M. & NAGABHUSHANT. L. (1972) Chemical ionization mass spectra of sugars. Tetrahedron Lett. 47, 48274830. MILLONIG G. (1961) Advantages of a phosphate buffer for OsO4 solutions in fixation. J. appl. Phys. 32, 1637. MiiLLER F. (1877) Crber Haarpinsel, Filzflecke und ii.hnliche Gebilde auf den Fli~geln m5nnlicher SchmetterlingeJena. Z.Naturwiss.21,99-104(fromBirch1970). PLISKE T. E. & SALPETER M. M. Tile structure and development of the hairpencil glands in males of the Queen butterfly Danaus gilippus berenice. J. Morph. 134, 215-241. REYNOLDSE. S. (1963) The use of lead citrate at high pH as an electron opaque stain in electron microscopy. J. Cell Biol. 17, 208-212. STOaBER. (1912) Die abdominalen Duftorgane der m~,nnlichen Sphingiden und Noctuiden. Zool. Jb. 32, 493-532. Key Word Index--Noctuidae; pheromones-phenyl ethanol; benzaldehyde; precursors; phenylethyl; benzylglycosides.