Cuticular hydrocarbons and novel alkenediol diacetates from wheat stem sawfly (Cephus cinctus): Natural oxidation to pheromone components

Article abstract of DOI:10.1023/A:1017994410538

The cuticular lipids of the wheat stem sawfly Cephus cinctus (Hymenoptera: Cephidae) were investigated as part of a chemical ecology project with this species. The major cuticular lipids were n-alkenes and n-alkanes. Alkenes were the most abundant and exhibited dramatic sexual dimorphism. (Z)-9-Tricosene accounted for about half of the total hydrocarbon in males but was nearly absent from females. The dominant alkenes in females were (Z)-9-pentacosene and (Z)-9-heptacosene. The alkane profiles were similar in both sexes, with n-tricosane being the most abundant, followed by n-pentacosane and n-heptacosane. In both sexes, there were minor amounts of alkanes and alkenes with other chain lengths and n-alkadienes of 29 and 31 carbons. In males, about one tenth of the surface lipids consisted of (Z)-9-alkene-1, ω-diol diacetates with 22-, 24-, and 26-carbon chains. The same compounds were also detected from females but in much smaller amounts. The structures of these novel diacetates were proven by synthesis. By analogy to methyl oleate, a well-studied food lipid, the alkenes and diacetates were expected to undergo slow oxidation in air to release specific aldehydes and other volatile products, and these were generally detected in volatiles collected from living sawflies. Atmospheric oxidation of the diacetates was also demonstrated in the absence of sawflies. One product from the diacetates, 9-acetyloxynonanal, was shown in other research to be particularly active electrophysiologically and was also attractive in the field. Aldehydes from the alkenes also showed strong electrophysiological activity. The concept of volatile pheromones originating from heavy, unsaturated cuticular lipids is discussed.

Full text of DOI:10.1023/A:1017994410538

Journal of Chemical Ecology, Vol. 28, No. 2, February 2002 ( 2002)
C
CUTICULAR HYDROCARBONS AND NOVEL ALKENEDIOL
DIACETATES FROM WHEAT STEM SAWFLY (Cephus cinctus):
NATURAL OXIDATION TO PHEROMONE COMPONENTS1
2,
2
2
´
ROBERT J. BARTELT, * ALLARD A. COSSE, RICHARD J. PETROSKI,
and DAVID K. WEAVER3
2USDA Agricultural Research Service
National Center for Agricultural Utilization Research
Crop Bioprotection Research Unit
1815 N. University Street
Peoria, Illinois 61604
3Montana State University
Department of Entomology
P.O. Box 173020
Bozeman, Montana 59717-3020
(Received August 31, 2000; accepted October 5, 2001)
Abstract—The cuticular lipids of the wheat stem sawfly Cephus cinctus
(Hymenoptera:Cephidae)wereinvestigatedaspartofachemicalecologyproject
with this species. The major cuticular lipids were n-alkenes and
n-alkanes. Alkenes were the most abundant and exhibited dramatic sexual di-
morphism. (Z)-9-Tricosene accounted for about half of the total hydrocarbon
in males but was nearly absent from females. The dominant alkenes in fe-
males were (Z)-9-pentacosene and (Z)-9-heptacosene. The alkane profiles were
similar in both sexes, with n-tricosane being the most abundant, followed by
n-pentacosane and n-heptacosane. In both sexes, there were minor amounts
of alkanes and alkenes with other chain lengths and n-alkadienes of 29 and 31
carbons. In males, about one tenth of the surface lipids consisted of (Z)-9-alkene-
1, ω-diol diacetates with 22-, 24-, and 26-carbon chains. The same compounds
were also detected from females but in much smaller amounts. The structures of
these novel diacetates were proven by synthesis. By analogy to methyl oleate,
a well-studied food lipid, the alkenes and diacetates were expected to undergo
slow oxidation in air to release specific aldehydes and other volatile products,
and these were generally detected in volatiles collected from living sawflies.
* To whom correspondence should be addressed. E-mail: BartelRJ@NCAUR.USDA.gov
1 USDA Disclaimer: Mention of trade names or commercial products in this article is solely for the
purpose of providing specific information and does not imply recommendation or endorsement by
the US Department of Agriculture.
385
C
0098-0331/02/0200-0385/0 2002 Plenum Publishing Corporation

´
386
BARTELT, COSSE, PETROSKI, AND WEAVER
Atmospheric oxidation of the diacetates was also demonstrated in the absence
of sawflies. One product from the diacetates, 9-acetyloxynonanal, was shown
in other research to be particularly active electrophysiologically and was also
attractive in the field. Aldehydes from the alkenes also showed strong elec-
trophysiological activity. The concept of volatile pheromones originating from
heavy, unsaturated cuticular lipids is discussed.
Key Words—Cephus cinctus, wheat stem sawfly, pheromone, cuticular lipid,
hydrocarbon, alkene, alkane, (Z)-9-docosene-1,22-diol diacetate, (Z)-
9-tetracosene-1,24-diol diacetate, (Z)-9-hexacosene-1,26-diol diacetate, alde-
hyde, oxidation.
INTRODUCTION
The wheat stem sawfly Cephus cinctus Norton (Hymenoptera: Cephidae) re-
mains an important pest of wheat in the northern Great Plains, and the chemical
ecology of this species is now being investigated with the goal of finding new
management tools. Aspects of this research have included the chemical identi-
fication, electrophysiological activity, and attractiveness of volatile compounds
from the sawfly (Cosse´ et al., 2002). Compounds from sawfly volatiles collections
that elicited strong electrophysiological responses included 9-acetyloxynonanal,
nonanal, tetradecanal, phenylacetic acid, and other aldehydes and carboxylic acids.
9-Acetyloxynonanal was attractive to both sexes in preliminary field tests (Cosse´
et al., 2002).
In another sawfly species, Pikonema alaskensis (Rohwer) (Hymenoptera:
Tenthredinidae), the volatile long-range pheromone also includes aldehyde com-
ponents, and these originate in an unusual way, through the oxidation of essen-
tially nonvolatile cuticular lipids (Bartelt and Jones, 1983). The cuticular lipids of
C. cinctus were, therefore, studied to explore whether such a mechanism might
operate in this species. The results suggest it does. This report includes a survey of
the cuticular lipids of C. cinctus, the synthesis of three novel constituents to prove
structure, and a discussion of the production of volatile pheromone components
from unsaturated lipids by oxidation.
METHODS AND MATERIALS
Extracts and Collections of Volatile Compounds. Hexane extracts of whole,
individual sawflies were prepared in Montana by one of us (D.K.W.) and sent
to Peoria for analysis. The insects were field collected as prepupae and handled
as described by Cosse´ et al. (2002). Individual adult sawflies were extracted for
either 20 min or 24 hr in ca. 0.5 ml hexane, and the sex and age of each insect was
recorded. Because mating occurs soon after emergence in the field, the chosen age

SAWFLY CUTICULAR LIPIDS
387
categories emphasized the younger insects: 1, 2, 3, 4, and 5 hr after emergence, 1–2
days old, and 5–6 days old. Some sawflies were also extracted before emergence;
these were removed from the wheat stem stubs after they had eclosed from their
pupae but before they had chewed through the plug in the stem. A total of 23 ex-
tractsofindividualswereanalyzedbyGC-MSandcompared. Later, additionalhex-
ane extracts were prepared in Peoria from insects received from Montana. These
included 1 sec dips and 24 hr soaks of groups of 10–25 male or female sawflies.
The collection of volatiles from living sawflies was described by Cosse´ et al.
(2002). The GC-MS data files for these previous volatiles collections were studied
further for additional information.
Instrumentation. The details of the GC-MS system were given by Cosse´ et al.
(2002). Briefly, the HP 5973 Mass Selective Detector was interfaced to an HP 6890
GC. Cuticular lipid samples were analyzed on a thin-film DB-1 GC column (15 m
length, 0.25 mm ID, 0.1 m film thickness; J & W Scientific, Folsom, California).
The oven program started at 50 C for 1 min, increased at 10 C/min to 300 C, and
remained at 300 C for 5 min. The inlet and transfer line were held at 300 C and
250 C, respectively. The usual MS scanning range was 40–550 amu (EI, 70 eV).
The Wiley mass spectral library with 275,821 spectra (Wiley, 1995) was available
on the MS data system. Quantitation of cuticular components was done by the ex-
ternal standard method on an HP 5890 Series II gas chromatograph equipped with
flame ionization detector. High-performance liquid chromatography (HPLC) was
used to separate geometrical isomers of alkenes and of one alkenediol diacetate.
The system consisted of a Waters M6000 pump, Waters R401 differential refrac-
tometer detector, and AgNO3-treated column. The AgNO3 column was prepared
from a 25-cm 4.6-mm-ID silica column (Alltech, Adsorbosphere, 5 m) by the
method of Heath and Sonnet (1980). The elution solvent was 0.25% 1-hexene in
hexane for alkene hydrocarbons and benzene for the alkenediol diacetate. HPLC
on a silica column (Alltech, Econosphere Silica 5 ) was also used to separate the
sawfly-derived alkenediol diacetates from the other cuticular lipids (10% ether in
hexane). Infrared spectra of purified synthetic alkenediol diacetates were obtained
as thin films on KBr plates using a Nicolet Impact 410 FT-IR spectrometer. The
proton NMR spectrum was obtained at 400 MHz for a synthetic alkenediol diac-
etate on a Bruker Avance 400 instrument. The solvent was deuterobenzene, and
the shifts were measured relative to TMS.
Dimethyl Disulfide Derivatives. Dimethyl disulfide (DMDS) adducts were
prepared from the sawfly-derived olefinic lipids and the synthetic alkenediol diac-
etates so that locations of double bonds could be determined (Carlson et al., 1989).
Briefly, the samples were evaporated to dryness under a stream of nitrogen, and
DMDS and 5% iodine in ether were added (equal volumes, about 25 l each). The
samples were heated at 45 C for 1–2 hr, then diluted with hexane and worked up
with aqueous sodium thiosulfate to destroy the iodine catalyst. The hexane layer,
which contained the products, was dried over sodium sulfate, evaporated under

´
BARTELT, COSSE, PETROSKI, AND WEAVER
388
nitrogen, resuspended in about 10 l of hexane, and analyzed by GC-MS. The MS
scanning range for these was 40–650 amu.
Synthesis of Alkene-1, ω-Diol Diacetates. Three alkenediol diacetates (8, 9,
and 10) were synthesized by one of us (R.J.P.), using standard methods, to confirm
chemical identifications (Figure 1). The Wittig reaction was the key synthetic
step, and most of the reactions in Figure 1 involved the creation of the required
Wittig salts and aldehydes from commercially available starting materials and the
additionorremovalofprotectinggroups. TheWittigsalts(1–4)werepreparedfrom
9- or 11-carbon bromohydrins (Aldrich, Milwaukee, Wisconsin) by refluxing with
triphenylphosphine in acetonitrile, followed by solvent removal and trituration
with dry ether to remove impurities. For two of these (1 and 2), the hydroxyl group
FIG. 1. Synthesis of the 9-alkene-1,ω-diol diacetates by Wittig routes (see text for abbre-
viations).

SAWFLY CUTICULAR LIPIDS
389
was protected as a tetrahydropyranyl (THP) ether before salt formation by reaction
with 2,3-dihydropyran (DHP) in acetonitrile, using pyridinium p-toluene sulfonate
(PPTS) as catalyst (Miyashita et al., 1977).
The carbon chains of the aldehydes (5–7) were all built up from smaller frag-
ments with Wittig reactions (Sonnet, 1974). For 5 and 6, two- and four-carbon
benzyl-ether aldehydes were condensed with the 11-carbon THP Wittig salt. The
four-carbon aldehyde was not commercially available but was made from its alco-
hol by oxidation with pyridinium dichromate (PDC) (Corey and Schmidt, 1979).
Hydrogenation of these Wittig products over 10% palladium on carbon both re-
moved the benzyl protecting group (but not the THP group) and reduced the double
bond (Greene and Wuts, 1991). The free hydroxyls were subsequently oxidized
to aldehydes with PDC. To make aldehyde 7, the six-carbon diol was first con-
verted to a mono-THP ether in a biphasic mixture of acetonitrile and hexane; the
reactants were primarily in the more polar (acetonitrile) phase, but the mono ether
migrated to the hexane phase, reducing the chance of further derivatization to the
diether. The free hydroxyl was then oxidized to an aldehyde, and the aldehyde was
condensed with hydroxy Wittig salt 3. The double bond of the resulting 17-carbon
product was hydrogenated over palladium on carbon, and the final aldehyde was
formed by PDC oxidation of the alcohol.
The final Wittig condensations with the prepared aldehydes involved either
the protected (THP) or hydroxy salts; both gave the expected products, although the
yields in either case were only fair. The THP groups were removed in MeOH under
acid catalysis (provided by Dowex 50W-X4 ion-exchange resin, which was later
easily removed by filtration) (Beier and Mundy, 1979). Finally, the free alkenediols
were acetylated with acetic anhydride, using lithium chloride as the Lewis acid
catalyst (Sabitha et al., 1999).
The synthetic requirement was only to obtain enough material for chromato-
graphic and spectral comparison with the natural material. Reactions were not
optimized for yield, and in most cases, intermediates were used in subsequent
reactions without purification. Overall yields were less than 1% from the starting
materials, but the desired final products were distinct and readily recognizable in
the reaction mixtures.
Reactions were monitored by mass spectrometry. Mass spectral and GC re-
tention data for the syntheses in Figure 1 are summarized in Table 1. Mass spectra
were not obtained for the Wittig salts 1–4. The THP ether and diol intermediates
for 9 and 10 were not analyzed by GC-MS but were converted directly to the di-
acetates before analysis. Synthetic 8–10 consisted of both E and Z isomers, which
were separable by GC. Synthetic 10 was investigated further so that the proper
configuration could be assigned to each GC peak. HPLC on the AgNO3/silica
column was used to separate the geometrical isomers, and infrared spectra were
then obtained for these. A proton NMR spectrum was also acquired for the more
abundant isomer.

´
BARTELT, COSSE, PETROSKI, AND WEAVER
390
TABLE 1. GC RETENTION TIMES AND MASS SPECTRAL DATA FOR INTERMEDIATES AND
FINAL PRODUCTS IN SYNTHESIS OF 1,ω-ALKENEDIOL DIACETATES (SEE FIGURE 1)
Compound and
(GC retention in min.)
Mass spectral fragments [ m/z] and (relative abundance)
Intermediate for compound 1
Br(CH2)11OTHP
85 (100), 101 (8), 333 (2, M), 335 (2, M)
(15.92)
Intermediate for compound 2
Br(CH2)9OTHP
85 (100), 101 (10), 305 (3, M), 307 (3, M)
(13.98)
Intermediates for compound 5
PhCH2OCH2CH CH(CH2)10OTHP 85 (100), 91 (77), 101 (13), 107 (9), 303 (4, M-85), 387
(21.57, 21.73)a
HO(CH2)13OTHP
(16.92)
(0.1), 388 (0.1, M)
41 (16), 43 (11), 55 (24), 56 (16), 57 (13), 67 (9), 69 (18),
81 (4), 83 (13), 84 (10), 85 (100), 95 (3), 97 (9),
101 (31), 111 (4), 125 (1), 215 (0.7, M-85), 299 (0.5),
300 (0.1, M)
OHC(CH2)12OTHP (5)
(16.30)
41 (17), 43 (12), 55 (21), 56 (18), 57 (14), 67 (10), 69 (9),
81 (7), 83 (8), 84 (12), 85 (100), 95 (10), 97 (6),
101 (22), 109 (6), 123 (2), 213 (0.7, M-85),
297 (0.4, M-1)
Intermediates for compound 6
PhCH2O(CH2)3CHO
(7.75)
91 (100), 107 (33), 178 (0.6, M)
PhCH2O(CH2)3CH CH(CH2)10OTHP 85 (100), 91 (62), 101 (5), 107 (7), 331 (6, M-85), 415
(22.83, 23.00)a
HO(CH2)15OTHP
(18.64)
(0.1), 416 (0.4, M)
85 (100), 101 (34), 243 (0.9, M-85), 327 (0.4), 328
(0.1, M)
OHC(CH2)14OTHP (6)
(18.10)
85 (100), 101 (27), 241 (1, M-85), 325 (0.4), 326 (0.2, M)
Intermediates for compound 7
HO(CH2)6OTHP
(9.57)
OHC(CH2)5OTHP
(8.60)
HO(CH2)10CH CH(CH2)5OTHP
(19.94, 20.02)a
HO(CH2)17OTHP
(20.22)
85 (100), 101 (32), 117 (3, M-85), 201 (0.3), 202 (0.1, M)
85 (100), 101 (25), 115 (1, M-85), 199 (0.4), 200 (0.1, M)
85 (100), 101 (2), 252 (1, M-84-18), 270 (0.8, M-84)
85 (100), 101 (36), 271 (1, M-85), 355 (0.3), 356 (0.1, M)
85 (100), 101 (27), 269 (1, M-85)
OHC(CH2)16OTHP (7)
(19.73)
Intermediates for compound 8
THPO(CH2)8CH CH(CH2)12OTHP 85 (100), 101 (11), 340 (1, M-84-84), 423 (1, M-85)
(26.61, 26.73)a
HO(CH2)8CH CH(CH2)12OH
(20.42, 20.50)a
41 (51), 55 (100), 57 (25), 67 (69), 69 (60), 81 (71), 82
(94), 95 (63), 96 (69), 109 (32), 123 (16), 135 (10),
149 (4), 304 (1, M-18-18), 322 (3, M-18), 340 (1, M)

SAWFLY CUTICULAR LIPIDS
391
TABLE 1. CONTINUED
Compound and
(GC retention in min.)
Mass spectral fragments [m/z] and (relative abundance)
AcO(CH2)8CH CH(CH2)12OAc (8)
(22.08, 22.15)a
41 (34), 43 (100), 55 (81), 57 (15), 61 (21), 67 (71), 69
(52), 81 (75), 82 (74), 95 (67), 96 (66), 109 (37),
110 (30), 121 (25), 123 (20), 135 (19), 137 (11),
149 (8), 151 (5), 152 (5), 304 (4, M-60-60),
321 (3, M-60-43), 364 (3, M-60) 424 (4, M)
Compound 9
AcO(CH2)8CH CH(CH2)14OAc
(23.43, 23.51)a
41 (34), 43 (100), 55 (85), 57 (22), 61 (24), 67 (65),
69 (59), 81 (74), 82 (70), 95 (64), 96 (63), 109 (36),
110 (29), 121 (22), 123 (21), 135 (17), 137 (12),
149 (7), 151 (5), 152 (6), 332 (4, M-60-60),
349 (3, M-60-43), 392 (4, M-60) 452 (5, M)
Compound 10
AcO(CH2)8CH CH(CH2)16OAc
(24.75, 24.81)a
41 (33), 43 (100), 55 (85), 57 (22), 61 (25), 67 (65),
69 (61), 81 (74), 82 (72), 95 (67), 96 (64), 109 (36),
110 (30), 121 (20), 123 (22), 135 (16), 137 (11),
149 (7), 151 (5), 152 (5), 360 (4, M-60-60),
377 (2, M-60-43), 420 (4, M-60) 480 (6, M)
aZ and E isomers with nearly identical mass spectra. The spectral data correspond to the earlier, larger
GC peak (Z isomer) in each case.
Other Standards. Samples of 8-acetyloxyoctanal and 10-acetyloxydecanal
were prepared for comparison of GC retention times and mass spectra with com-
ponents in sawfly volatile collections. The commercially obtained 8- and 10-carbon
diols were converted to mono-THP ethers as described above, oxidized with PDC
to mono-THP aldehydes, then deprotected and acetylated. The 13-, 15-, and 17-
carbon aldehyde acetates were obtained by ozonolysis (see Cosse´ et al., 2002)
of alkenediol diacetate samples. The saturated aldehydes of 6–14 and 16 carbons
were either commercially available (6–14 carbons, Aldrich) or were prepared by
ozonolysis of appropriate alkene hydrocarbons. Octanoic, nonanoic, decanoic,
and phenylacetic acids were obtained commercially (Aldrich). Synthetic E and Z
alkenes and various other alcohols, hydrocarbons, and 2E-alkenals were on hand
from other research.
Laboratory Oxidation of Alkenediol Diacetates. Preliminary study of light-
induced, abiotic oxidation of alkenediol diacetates was performed using a Suntest
CPS solar simulator (Atlas, Gainesville, Florida), with the system set up as de-
scribedbyMcGuireetal. (2000). Purified(HPLC)samplescontainingabout500ng
of synthetic 10 or natural male alkenediol diacetates were evaporated to dryness
on the inside of quartz cuvettes. Cuvettes were capped and sealed with Teflon tape
and placed in the solar simulator. The outside of the treated cuvette was in direct
contact with the cooled (15 C) instrument surface. Samples were exposed for 2 hr

´
BARTELT, COSSE, PETROSKI, AND WEAVER
392
to light irradiation similar in intensity and wavelength composition (300–1100 nm)
to sunlight. As a control, a sample of synthetic 10 was set up identically but without
the light. All samples were resuspended in hexane after 2 hr and then analyzed by
GC-MS.
RESULTS AND DISCUSSION
Cuticular Hydrocarbons. The cuticular hydrocarbons were the major com-
pounds in the sawfly hexane extracts (Figure 2, Table 2). Both sexes had large
amounts of n-alkanes and n-alkenes, primarily with odd numbers of carbons. The
n-alkane profiles were similar between sexes, with tricosane being most abundant
and with smaller amounts of 25-, 27-, 29-, and 31-carbon alkanes. There were
traces of the intervening even-chain compounds and also the 21- and 33-carbon
alkanes. No branched alkanes were detected.
The n-alkenes were the most abundant class of hydrocarbons, and their pro-
files differed dramatically between sexes. In males, the 23-carbon alkene ac-
counted for about half of the total hydrocarbon. In females, this alkene was far less
FIG. 2. Gas chromatograms of extracts from female and male C. cinctus. Selected hydrocar-
bons (:Hyd) and diacetates (:diAc) are labeled. Relative areas of off-scale peaks are shown
in parentheses.

SAWFLY CUTICULAR LIPIDS
393
TABLE 2. HYDROCARBONS EXTRACTED FROM 2-TO 5-DAY-OLD MALE AND FEMALE C.
cinctus CUTICLE
Relative amount
MS data: molecular ion of
(% of total)
hydrocarbons (and key ions of
Hydrocarbon
Females
Males
DMDS adducts)
n-Alkanes (by numbers of carbons)
21
22
23
24
25
26
27
28
29
30
31
33
0.31
0.13
13.9
0.24
8.6
0.32
11.1
0.20
3.5
0.57
0.29
17.5
0.23
6.1
0.20
5.9
0.14
3.4
296
310
324
338
352
366
380
394
408
422
436
464
0.038
0.52
0.10
0.074
1.2
0.36
n-Alkenes (by double bond descriptions and numbers of carbons)
(Z)-9-21
(Z)-9-22
(Z)-9-23, [(Z)-7-23]a
(Z)-9-24
(Z)-9-25
(Z)-9-26
(Z)-9-27
(Z)-9-28
0.016
0.027
4.4
0.24
0.48
48.0
0.68
13.4
0.072
0.92
NDb
0.17
ND
294, (173, 215, 388)
308, (173, 229, 402)
322, (173, 243, 416, [145, 271]a)
336, (173, 257, 430)
350, (173, 271, 444)
364, (173, 285, 458)
378, (173, 299, 472)
392, (173, 313, 486)
406, (173, 327, 500)
420, (173, 341, 514)
434, (173, 355, 528)
0.53
34.9
0.50
16.6
0.15
3.1
(Z)-9-29
(Z)-9-30
(Z)-9-31
0.049
0.67
0.071
n-Alkadienes (by double bond descriptions and numbers of carbons)
(Z,Z)-8,20-29
(Z,Z)-9,21-31
0.034
0.042
0.020
0.015
404, (159, 173, 325, 339, 498, 592)
432, (173, 187, 339, 353, 526, 620)
aPresent in trace amount (<1% of major isomer).
bND denotes not detected.
abundant, and the 25- and 27-carbon alkenes were dominant instead. There were
smaller amounts of alkenes with 21, 29, and 31 carbons in both sexes and also
the intervening even-chain compounds. In both sexes, the double bonds were al-
most exclusively in the nine position for all chain lengths, based on mass spectra
of the DMDS derivatives. The 23-carbon alkenes from both sexes did have trace
amounts of 7-alkene (<1% as much as the 9-isomer). The observed molecular
ions (alkene molecular weights plus 94) and the key fragment ions (from cleav-
age between the carbons with the thiomethyl groups) are given in Table 2. By
comparison to E and Z synthetic standards (underivatized) on the AgNO3 HPLC

´
BARTELT, COSSE, PETROSKI, AND WEAVER
394
column, allofthesawflyalkeneshadtheZconfiguration. Nobranchedalkeneswere
detected.
In females, hydrocarbon mass spectra indicated the presence of small amounts
of alkadienes of 29 and 31 carbons (molecular ions at m/z = 404 and 432, respec-
tively). The mass spectra of the DMDS derivatives were interpreted according
to Carlson et al. (1989). Fragments listed in Table 2 represent the molecular ion
(addition of two equivalents of DMDS), the loss of 94 (one DMDS) as a neutral
molecule, and subsequent cleavage at the remaining DMDS site as with the alkenes
(yielding four ions, representing the saturated and unsaturated side of each double
bond). These spectra determined the double bonds to be in the 9 and 21 positions
from one end of the chain for both dienes. (However, rules of nomenclature require
that the 29-carbon compound be named as 8,20-nonacosadiene so that the locant
numbers are as small as possible). GC retention times were consistent with straight
chains and Z configurations at the double bonds (Bartelt et al., 1986). These dienes
were also detected from males, but in smaller amounts.
The total amount of hydrocarbon per sawfly was about 15 g, with males
and females having very similar quantities. Extraction for 20 min or 24 hr gave
comparable profiles and amounts, although the longer soak also extracted fatty
acids (primarily palmitic and oleic) and sterols from the insects. Even a 1 sec rinse
gave essentially the same profiles, indicating that the compounds were present on
the immediate surface. The lipid profiles and amounts were consistent in major
features for all ages between 1 hr and 5 days. In contrast, adults removed from
wheat stubs before natural emergence had the typical amount of alkane but only
about 10% of the usual amount of alkene. Apparently, biosynthesis of alkenes is
intense around the time of emergence. Finally, younger females (several hours to
1 day old) had larger amounts of the alkadienes than older ones, with the 29-carbon
diene amounting to as much as 0.5% of the total hydrocarbon rather than less than
0.05%. This difference was not seen in males. The female in Figure 2 was less
than 1 day old at the time of extraction.
Diacetate Lipids. Initial comparison of chromatograms revealed two com-
pounds in male extracts (approximately 1 g of each per male) that were far less
abundant in females (Figure 2, retention times of 21.9 and 23.2 min). The mass
spectrum of the first of these indicated a molecular weight of 424 (Figure 3), which
was not consistent with a hydrocarbon. (A fully saturated hydrocarbon of 30 car-
bons would have a molecular weight of 422; thus oxygen was probably present in
the sawfly compound to account for the higher molecular weight). Several subtle
features in the mass spectrum gave important information about the structure. The
first was a fragment at m/z 61 (19% of base peak), which is characteristic for an
acetate group (McLafferty, 1980). Examination of library spectra of acetates indi-
cated that the m/z 61 peak is most intense when the acetate group is at or near the
end of the carbon chain and is distant from other functional groups or branches.
Second, the spectrum of the sawfly compound lacked the characteristic peaks that

SAWFLY CUTICULAR LIPIDS
395
FIG. 3. EI-mass spectrum of 22-carbon diacetate lipid from male C. cinctus (top) and EI-
mass spectrum of the DMDS derivative of this lipid, showing origins of major fragments
(bottom).
would be expected if the acetates were secondary; for example, an acetate on the
second carbon of a chain leads to a prominant spectral peak at m/z 87 due to
chain cleavage adjacent to the acetate. Finally, there were fragments at m/z 364
and m/z 304 that corresponded to two consecutive losses of 60 from the molecular
ion (believed to represent the loss of two acetic acid molecules). Taken together,
this evidence suggested a lipid chain of 22 carbons with an acetate group on each
end. However, the molecular weight for a saturated chain would be 426 instead of
424; thus a double bond was probably present in the chain. The DMDS derivative
formed readily (molecular ion at m/z 518 = 424 + 94), and the fragment ions (m/z
231 and 287) were consistent with a double bond being present in the 9 position
of the original lipid (Figure 3).
The data for the second lipid were analogous, with molecular weight of 452
and key fragments at m/z 61, 332, and 392. The DMDS adduct gave key MS ions
at m/z 231, 315, and 546, consistent with a 24-carbon diacetate, again with the
double bond in the 9 position. Closer inspection of the gas chromatograms revealed
a third, still later-eluting member of this series, about 10% as abundant as the first
two. Its mass spectrum included ions at m/z 61, 360, 420, and 480, and the MS

´
BARTELT, COSSE, PETROSKI, AND WEAVER
396
ions of its DMDS derivative included 231, 343, and 574. These data indicated
a 26-carbon diacetate, once more with the double bond in the 9 position. There
was no evidence for other double-bond positional isomers. Close examination of
female-derived extracts and their DMDS derivatives by GC-MS revealed that the
same three compounds were present but at much lower levels (about 10% of the
amounts in males, see Figure 2).
Synthetic Alkenediol Diacetates. Synthetic alkenediol diacetates were pre-
pared to confirm the proposed structures and to establish double-bond configu-
ration of the natural material. The mass spectral data (Table 1) supported that
the synthetic reactions proceeded as shown in Figure 1 because the synthetic
intermediates and products had the expected molecular weights and reasonable
fragmentation patterns. For example, THP ethers were readily recognized by their
mass spectra (Table 1) because of the base peak at m/z 85 and a moderately intense
ion at m/z 101, due to cleavage on either side of the ether oxygen. The molecular
ion and an M-1 ion were usually seen for the THP ethers, but the intensity for
both was low (ordinarily <1%). The M-85 peak was similarly small but was still
reliably present and served as further confirmation of molecular weight. Example
details for one THP ether alcohol and one THP ether aldehyde are given under
the intermediates for compound 5 in Table 1; for other THP ethers, only key ions
are listed. Analogously to THP ethers, benzyl ethers had spectral peaks at m/z 91
and 107. The THP ether bromides (intermediates for 1 and 2) had characteristic
spectra because of the bromine isotopes. For the di-THP diether precursor of 8,
the molecular ion was not seen, but the M-85 fragment was present, and there
was another ion corresponding to two losses of a deprotonated THP group (m/z
340). After deprotection, the free diol exhibited a small molecular ion and ions
corresponding to one and two neutral losses of water. Features of the mass spectra
of alkenediol diacetates were discussed above.
Patterns in GC retention times supported the mass spectral interpretations and
synthetic expectations. For example, the Wittig reactions always gave two products
(the E and Z isomers) that were separated by about 0.1 min but had nearly identical
mass spectra. Catalytic hydrogenation always collapsed the E/Z peak doublet into
a single peak of still higher GC retention time. Also, oxidation of a THP ether
alcohol to the corresponding THP ether aldehyde consistently decreased the GC
retention time (Table 1).
By GC of synthetic 8–10, the earlier-eluting isomer was always the more
abundant one; this was expected to be the Z isomer because the Wittig reaction
conditions favor production of Z double bonds (Sonnet, 1974). Compound 10 was
analyzed further by AgNO3 HPLC to confirm this. The more abundant isomer
eluted later than the less abundant one (24–26 ml vs. 20–21 ml, based on GC
analysis of HPLC fractions), which also supported that the more abundant isomer
was Z (Heath and Sonnet, 1980). The IR spectra further corroborated this conclu-
sion. The more abundant isomer was a liquid at room temperature and had key IR

SAWFLY CUTICULAR LIPIDS
397
absorbances (cm 1) indicative of an olefin (3003, w, C H), hydrocarbon chain
(2925 and 2854, both s, CH2 ), and acetate group (1741, s, C O; 1365, m,
1240, s, and 1037, m, other acetate-related peaks), but the spectrum did not indi-
cate presence of an E double bond. On the other hand, the less abundant isomer
was a solid at room temperature and had corresponding absorbances at 2918, 2850,
1732, 1369, 1247, and 1041; it also had the absorbance (962, m) that confirmed the
E double bond. The proton NMR spectrum of the more abundant isomer supported
the structure for synthetic 10 as drawn in Figure 1, but it gave no further informa-
tion about double-bond configuration because the olefinic protons had the same
shift and the coupling constant could not be observed: 5.56 (2 H, m, CH ),
4.04 (4 H, m, O CH2 ), 2.17 (4 H, m, CH CH2), 1.74 (6 H, s, CH3(C O)O),
1.2–1.6 (40 H, other CH2 , m).
Comparison of Natural and Synthetic Alkenediol Diacetates. Based on same-
day comparisons of GC retention times, each natural alkenediol diacetate matched
exactly the Z isomer of the corresponding synthetic compound for all three chain
lengths. They also matched with respect to mass spectra and with respect to GC
rentention times and mass spectra of the DMDS derivatives.
Compounds in Nature Related to Diacetates. This is believed to be the first
report of natural alkenediol diacetates, but related compounds are known. Beeswax
contains saturated 1,ω-diols of 24–28 carbons and esters of these, but the acid moi-
eties have 16–24 carbons instead of just two (Tulloch, 1971). Free 1,ω-alkanediols
of 22, 24, and 26 carbons have been isolated from various plants (Chatterjee
and Chakraborty, 1967; Austin et al., 1969; Rizvi and Sultana, 1972; Amparo-
Lopez et al., 1976; Hatam et al., 1990). 1,16-Hexadecanediol diacetate has been
reported in the defensive secretion of the gulf fritillary butterfly (M. Blum, personal
communication). Recently Oliver et al. (2000) and Doss et al. (2000) isolated 3-
hydroxypropanoylestersof(Z)-9-docosene-1,22-dioland(Z)-9-tetracosene-1,24-
diol from bruchid weevils. The compounds are potent plant growth regulators, and
callus tissue develops on legume pods exposed to these compounds during ovipo-
sition by female beetles.
Oxidation of Cuticular Alkenes. “Spontaneous” oxidation of unsaturated
lipids has been studied extensively, especially in relation to rancidity of food oils,
and much is known about the chemical mechanisms involved (reviewed by Frankel,
1998). Oxidation is complicated and involves cascades of free-radical reactions.
It can be light induced (photooxidation) or simply occur by exposure to molecu-
lar oxygen under ambient conditions (autoxidation) if triggered by trace catalysts
or reactive species such as free radicals. Once initiated, the reactions tend to be
self-sustaining. Both mechanisms generally lead to the same volatile products, but
proportions can differ. The major volatile products are usually aldehydes.
One food-related lipid, methyl oleate, provides a reasonable predictive model
for the nonenzymatic oxidation of the alkenes and cuticular alkenediol diacetates of
C. cinctus. Methyl oleate is like the sawfly compounds in having just one Z double

´
BARTELT, COSSE, PETROSKI, AND WEAVER
398
bond that is distant from other functional groups or chain ends. The mechanism for
oxidation of methyl oleate to volatile products (Frankel, 1998) suggests a specific
series of compounds that would result from the oxidation of particular alkenes,
and occurrence of these compounds in the volatile collections from sawflies would
support that such a mechanism is operating.
The oxidation scheme and expected products from (Z)-9-tricosene, the only
abundant alkene in C. cinctus males, are summarized in Figure 4. The initial re-
action is the formation of allylic hydroperoxides (four positional isomers). The
hydroperoxides decompose either thermally or catalytically (e.g., by metal ions)
into alkoxyl radicals, and these can then react in a number of ways, including cleav-
age into volatile products. Scission on either side of the radical group (indicated
by “A” or “B” in Figure 4), yields a specific free aldehyde plus an intermediate
that reacts further to give a characteristic hydrocarbon, alcohol, or other aldehyde.
This general scheme is shown at the bottom of Figure 4, and the expected products
from each of the hydroperoxide isomers are summarized at the upper right. By
analogy to methyl oleate (Frankel, 1998), nonanal and tetradecanal should be the
most abundant products.
The compounds actually detected in volatile collections are shown in bold
print in Figure 4; relatively large and intermediate quantities are indicated by
and , respectively. The compounds were identified by mass spectrum and GC
retention time, relative to authentic standards. The more abundant products are also
indicated in the total ion chromatogram in Figure 5, which represents a volatile
collection from male sawflies.
Nonanal and tetradecanal were, in fact, the most abundant of the expected
oxidation products from males. Octanal and decanal were clearly present but were
less abundant. Amounts were still smaller for the other compounds noted in bold
in Figure 4. Some of the products expected but not found (heptanal, heptane,
heptanol, octane) are relatively volatile and would have occurred too near the
solvent front of the GC run to yield sharp peaks and be easily detected. Others,
such as 2-undecenal, were probably present, but small amounts and coelution
with more abundant compounds prevented positive identification. The heavier
(15- and 16-carbon) 2-alkenals were not detected. Their lower volatilities would
have reduced the amounts collected. Furthermore, these compounds are especially
prone to further oxidation, and remaining in the oxidizing lipid mixture instead of
evaporating out would have enhanced their degradation.
Female sawflies have very little 23-carbon alkene but have large amounts of
the 25- and 27-carbon alkenes. Oxidation of these would be analogous to tricosene.
All of the scission products with 11 or fewer carbons (Figure 4) would be expected
from any of the 9-alkenes, but those of 12 or more carbons expected from 9-
tricosene would be replaced by homologs with two or four additional carbons if
the parent alkene is 9-pentacosene or 9-heptacosene. Hence, there was relatively
little tetradecanal from females, but hexadecanal was abundant instead (Cosse´

SAWFLY CUTICULAR LIPIDS
399
FIG. 4. Expected products from the autoxidation of (Z)-9-tricosene, based on the process
for methyl oleate (Frankel, 1998). Compounds in bold print were detected in volatile col-
lections from male sawflies. High and intermediate relative abundance denoted by and ,
respectively.

´
BARTELT, COSSE, PETROSKI, AND WEAVER
400
FIG. 5. GC-MS chromatograms of a volatile collection from male C. cinctus. The upper
trace is total ion chromatogram. The lower trace is for mass 61 only, which shows the acety-
loxyaldehydes (and also the carboxylic acids) very clearly. Peaks are labeled as aldehydes
(:Ald), hydrocarbons (:Hyd), carboxylic acids (:Acid), or acetyloxyaldehydes (:Ac-Ald).
Nonbiological impurities are denoted by “Imp.”
et al., 2002). The 18-carbon aldehyde was observed in the collections of volatiles
but in smaller amounts. Its lower volatility probably reduced its abundance in the
collections.
The observed product patterns support the formation of the volatiles by a
mechanism like that described for methyl oleate. Further oxidation of these alde-
hydes would yield carboxylic acids, and those of 8, 9, and 10 carbons were found
consistently in the volatile collections from sawflies (Cosse´ et al., 2002) (Figure 5),
although these may have other, as yet unexplained sources as well. The emission
of phenylacetic acid from sawflies is not, however, accounted for by the oxidation
mechanism.
The amounts of oxidation products collected from the sawflies were minute
relative to the amounts of alkene present. The most abundant aldehydes from
males were nonanal and tetradecanal, but typical collected amounts were only
50 and 15 ng/day per male, respectively, while males have about 7 g of the
parent 9-tricosene. Monounsaturated compounds such as methyl oleate oxidize

SAWFLY CUTICULAR LIPIDS
401
only 1–2% as rapidly as the polyunsaturated analogs (Frankel, 1998); thus, the
minor yields of volatile products from alkenes is not surprising. Nevertheless, the
levels of volatiles observed are still sufficiently high to be detectable to sawfly
antennae. Intriguingly, even though the intact alkenes were not found to elicit
electrophysiological responses (Cosse´ et al., 2002), the male–female dimorphism
in alkenes could still lead to behavioral recognition of sexes at a distance by way
of the aldehyde (or other) oxidation products.
Dienes. The alkadienes, which are more abundant in females, would release
additional, unique oxidation products. These would include dodecanedial and un-
saturated aldehydes of 20, 21, and 22 carbons with a double bond in the 12 position.
None of these compounds were detected, but the amount of diene starting material
that was present was extremely small.
Oxidation of Alkene-1,ω-diol Diacetates. By analogy to the alkenes, the
alkenediol diacetates of 22, 24, and 26 carbons would all be expected to pro-
duce 9-acetyloxynonanal, and each diester would also generate a unique, heavier
homolog from the other end of the lipid chain (Figure 6).
The mass spectral data files for collections of sawfly volatiles were examined
for these compounds by extracting single-mass chromatograms for m/z 61 (for
FIG. 6. Acetyloxyaldehydes expected from the oxidative cleavage of the alkenediol diac-
etates of C. cinctus. The major product would be 9-acetyloxynonanal because it can be
derived from all three parent lipids, and acetyloxyaldehydes with 13-, 15-, and 17-carbon
chains would result from the other ends of the parent lipid chains. Analogs of these acety-
loxyaldehydes having one additional or one fewer carbon would also be expected (see text),
but only those related to 9-acetyloxynonanal were likely to be detectable.

´
BARTELT, COSSE, PETROSKI, AND WEAVER
402
example, Figure 5). This approach revealed compounds containing the primary
acetate group without interference from many other functional classes. (Simple
aldehydes, alcohols, and hydrocarbons do not exhibit the m/z 61 fragment. Organic
acids do show up, however, because there is always an isotope peak or protonated
fragment corresponding to the major ion at m/z 60). 13-Acetyloxytridecanal was
readily found in samples in which 9-acetyloxynonanal was prominent (about 20%
as abundant as the major product, Figure 5). The 15-carbon homolog was some-
times detectable but was even less abundant, probably because of lower volatility.
17-Acetyloxyheptadecanal was never found, presumably because of still lower
volatility and the much smaller amount of precursor. All four of these aldehyde
acetates (of 9, 13, 15, and 17 carbons) were detected when the crude hexane extract
of sawflies was ozonized, and the 13-carbon homolog clearly stimulated antennae
(Cosse´ et al., 2002).
8-Acetyloxyoctanal and 10-acetyloxydecanal were also detected in volatiles
collections but not from the lipid ozonolysis experiment. These compounds would
be expected if the alkenediol diacetates oxidized according to the scheme for the
alkenes and methyl oleate (but not from ozonolysis). Thus the pattern of aldehyde
acetates observed in volatiles collections further supported the idea that they were
formed by “weathering” of cuticular lipids. Interestingly, in tests with the synthetic
standards, 8-acetyloxyoctanal and 10-acetyloxydecanal were clearly active by GC-
EAD. Analogs with one additional and one fewer carbon would also be expected
for the acetyloxyaldehydes with 13-, 15-, and 17-carbon chains, but these would
have been present at extremely low levels and were not detected.
Under laboratory conditions, the amount of 9-acetyloxynonanal emitted from
male sawflies was small (1–10 ng per sawfly per day) (Cosse´ et al., 2002) relative
to the amounts of precursor alkenediol diacetates present in males (about 2 g),
but it was roughly proportional to the amounts of aldehydes apparently derived
from 9-tricosene, further supporting a similar origin. In addition, the relatively
greater production of 9-acetyloxynonanal from male sawflies than females (Cosse´
et al., 2002) was consistent with the greater amounts of the alkenediol diacetates
present in males.
These qualitative and quantitative relationships strongly suggest that the
alkenediol diacetates are the precursors to 9-acetyloxynonanal, a key compound
in the sawfly volatiles collections that elicited intense antennal activity and had
demonstrable activity in field tests (Cosse´ et al., 2002), and furthermore, that pas-
sive oxidation of the cuticular lipids could account for the amounts obtained in our
volatiles collections from living sawflies.
Laboratory Oxidation of Alkenediol Diacetates. The synthetic alkenediol di-
acetate 10 was free of detectable acetyloxyaldehydes prior to artificial sunlight ex-
posure. However, after 2 hr of irradiation, and as expected from our chemical model
(Figure 6), 8-acetyloxyoctanal, 9-acetyloxynonanal, 16-acetyloxyhexadecanal,
and 17-acetyloxyheptadecanal could be readily detected and were 0.09%, 0.4%,

SAWFLY CUTICULAR LIPIDS
403
0.3%, and 0.9% as abundant as the recovered starting material 10 (calculated
from peak areas in the total ion chromatogram). 10-Acetyloxydecanal and 18-
acetyloxyoctadecanal were not detected. An HPLC preparation of natural alkene-
diol diacetates, which contained 8 and 9 in an 8:1 ratio, was also free of detectable
acetyloxyaldehydes before the light treatment. After 2 hr of light exposure, the re-
sults were analogous to synthetic 10: 8-acetyloxyoctanal, 9-acetyloxynonanal, 12-
acetyloxydodecanal, 13-acetyloxytridecanal, and 15-acetyloxypentadecanal were
detected and were, respectively, 0.1%, 0.6%, 0.2%, 0.8%, and 0.2% as abundant
as starting compound 8. The experiment with synthetic 10 without exposure to
light gave results similar to when the light was used, but the amounts of the acety-
loxyaldehydes were lower (the acetyloxyaldehydes with 8-, 9-, 16-, and 17-carbon
chains were 0.03%, 0.2%, 0.1%, and 0.4% as abundant as the starting material,
respectively).
Although this experiment was preliminary, it clearly demonstrated that the
biologically active acetyloxyaldehydes can be produced from the alkenediol diac-
etates in the absence of living sawflies. Furthermore, the amounts of acetyloxyalde-
hydes produced, relative to the amounts of parent alkenediol diacetates, were re-
markably similar to those encountered with living sawflies (e.g., low nanogram
quantities of acetyloxyaldehydes emitted per sawfly per day versus low microgram
quantities of alkenediol diacetate present on the cuticle). Based on the GC-EAD
studies of Cosse´ et al. (2002), the amounts of acetyloxyaldehydes produced by this
process are large enough to be biologically relevant.
Behavioral and Environmental Enhancement of Oxidation Reactions. Cer-
tain features of sawfly biology would favor the oxidation of cuticular lipids. The
insects are most active when the weather is warm (microclimate as hot as 35 C)
and sunny (D. K. Weaver, unpublished). The oxidation reactions are accelerated
by both heat and sunlight (Frankel, 1998, and our preliminary experiment). Sawfly
grooming behavior (D. K. Weaver, unpublished) could distribute pheromonal pre-
cursors over the cuticle in a way that enhances oxidation rates. Thus it is possible
that the reactions may be biologically (behaviorally) controlled even if there is no
direct enzymatic mediation. The abiotic oxidation of unsaturates is complicated
to study, especially in the early stages of the process, because it can be triggered
or influenced by so many factors. These include the presence of trace amounts
of metal ions, ketones, hydroperoxides, and other reaction-initiating substances,
the action of ultraviolet light, and the effect of temperature (Frankel, 1998). On
the other hand, involvement of enzymes has not been ruled out, and resolving this
issue will be essential to understanding this pheromone system.
Pheromones Produced by Oxidation. The oxidation of relatively nonvolatile
cuticular lipids is known to release volatiles with pheromonal activity in other
hymenopterans. For example, females of the yellowheaded spruce sawfly,
P. alaskensis, possess a series of (Z,Z)-9,19-alkadienes of 28–37 carbons that is
absent in males (Bartelt et al., 1982). All of these dienes have a common oxidation

´
BARTELT, COSSE, PETROSKI, AND WEAVER
404
product, (Z)-10-nonadecenal, that is highly attractive to males in the field (Bartelt
and Jones, 1983). Similarly, in the parasitic wasp, Macrocentrus grandii Goidanich
(Braconidae), the females have a series of (Z,Z)-9,13-alkadienes of 27–41 carbons
that the males lack (Swedenborg and Jones, 1992). With these dienes, the common
oxidation product is (Z)-4-tridecanal, and as with P. alaskensis, the aldehyde is
clearly attractive to males in the field. In laboratory wind-tunnel bioassays, as little
as 17 pg of this aldehyde elicited dramatic upwind orientation.
C. cinctus differs in two important ways from the above examples. First and
most obviously, the males are the major source of the key compounds (e.g., the
diacetate lipids and the derived aldehyde acetates) instead of the females. Second,
both sexes do have measurable amounts of the compounds, unlike the usual pattern
with insect sex pheromones. Although the pheromone research on C. cinctus is in
its early stages, the pheromone biology of the species already appears to be quite
atypical and may provide a useful model for comparison in the future.
The existence of oxidation-derived pheromones in insects has not been fre-
quently explored. The distinctive alkenediol diacetates of C. cinctus and their
chemical relationship to the acetyloxyaldehydes through oxidation made this con-
nection between nonvolatile precursors and volatile pheromones very clear, but as
demonstrated with the alkenes, very typical compounds can also release volatiles
that could be every bit as important biologically as the more esoteric ones. The
reported examples of biologically active oxidation products are from the
Hymenoptera, butitseemslikelythatthephenomenoncouldoccurfarmorewidely.
Acknowledgments—Useful discussions and helpful advice concerning lipid oxidation were gen-
erously provided by Dr. Hal Gardner, USDA-ARS-NCAUR. We are grateful to Dr. Dave Weisleder
of NCAUR for obtaining the NMR spectrum and to Mr. John Salch of NCAUR for acquiring the IR
spectra. Dr. Robert Behle of NCAUR allowed us to use the solar simulator in his laboratory.
REFERENCES
AMPARO-LOPEZ, M., RODRIGUEZ, B., and SERAFIN, V. 1976. Studies on terpenes of the Sideritis genus,
XXVIII: Components of Sideritis ochroleuca. Inst. Quim. Org. Gen. 72:578–581.
AUSTIN, P. W., SESHADRI, T. R., and SOOD, M. S. 1969. Chemical study of Prunus puddum (stem bark)
and Prunus cornuta (stem bark and wood). Indian J. Chem. 7:43–48.
BARTELT, R. J. and JONES, R. L. 1983. (Z)-10-Nonadecenal: A pheromonally active air oxida-
tion product of (Z,Z)-9,19 dienes in yellow-headed spruce sawfly. J. Chem. Ecol. 9:1333–
1342.
BARTELT, R. J., JONES, R. L., and KULMAN, H. M. 1982. Hydrocarbon components of the yellow-
headed spruce sawfly sex pheromone: A series of (Z,Z)-9,19 dienes. J. Chem. Ecol. 8:95–
114.
BARTELT, R. J., MOATS, M. L., and JACKSON, L. L. 1986. Relationships between structure and chro-
matographic retention for n-alkadienes. J. Chromatogr. 366:27–37.
BEIER, R. and MUNDY, B. P. 1979. A facile removal of the tetrahydropyranyl protection group from
alcohol derivatives. Synth. Comm. 9:271–273.

SAWFLY CUTICULAR LIPIDS
405
CARLSON, D. A., ROAN, C.-S., YOST, R. A., and HECTOR J. 1989. Dimethyl disulfide derivatives of
long chain alkenes, alkadienes, and alkatrienes for gas chromatography/mass spectrometry. Anal.
Chem. 61:1564–1571.
CHATTERJEE, A. and CHAKRABORTY, A. K. 1997. Chemical investigatio of Bourcerosia aucheriana:
Isolation of dibehenate of 1,26-hexacosanedial. J. Indian Chem. Soc. 44:87–90.
´
COSSE, A. A., BARTELT, R. J., WEAVER, D. K., and ZILKOWSKI, B. W. 2002. Electrophysiological
discovery of a pheromone that relates to lek-like aggregations in the wheat stem sawfly. J. Chem.
Ecol. 28:407–423.
COREY, E. J. and SCHMIDT, G. 1979. Useful procedures for the oxidation of alcohols involving pyri-
dinium dichromate in aprotic media. Tetrahedron Lett. 5:399–402.
DOSS, R. P., OLIVER, J. E., PROEBSTING, W. M., POTTER, S. W., KUY, S.-R., CLEMENT, S. L.,
WILLIAMSON, R. T., CARNEY, J. R., and DEVILBISS, E. D. 2000. Bruchins: Insect-derived plant
regulators that stimulate neoplasm formation. Proc. Natl. Acad. Sci. USA 97:6218–6223.
FRANKEL, E. N. 1998. Lipid Oxidation. Oily Press, Dundee, Scotland.
GREENE, T. W. and WUTS, P. G. M. 1991. Protective Groups in Organic Synthesis, 2nd ed. Wiley, New
York.
HATAM, N. A. R., WHITING , D. A., and YOUSIF, N. J. 1990s. Lipids and sterols of Citrullus colocynthis.
Int. J. Crude Drug Res. 28:183–184.
HEATH, R. R. and SONNET, P. E. 1980. Technique for in situ coating of Ag+ onto silica gel in HPLC
columns for the separation of geometrical isomers. J. Liq. Chromatogr. 3:1129–1135.
MCGUIRE, M. R., BEHLE, R. W., GOEBEL, H. N., and FRY, T. C. 2000. Calibration of a sunlight
simulator for determining solar stability of Bacillus thuringiensis and Anagrapha falcifera nucler
polyhedrovirus. Environ. Entomol. 29:1070–1074.
MCLAFFERTY, F. W. 1980. Interpretation of Mass Spectra, 3rd ed. University Science Books, Mill
Valley, California.
MIYASHITA, N., YOSHIKOSHI, A., and GRIECO, P. A. 1977. Pyridinium p-toluenesulfonate: A mild and
efficient catalyst for the tetrahydropyranylation of alcohols. J. Org. Chem. 42:3772–3774.
OLIVER J. E., DOSS, R. P., WILLIAMSON, R. T., CARNEY, J. R., and DEVILBISS, D. E. 2000. Bruchins:
Mitogenic 3-(hydroxypropanoyl) esters of long chain diols from weevils of the Bruchidae. Tetra-
hedron 56:7633–7641.
RIZVI, S. A. I. and SULTANA, T. 1972. Phytochemical studies of the flowers of Pterospermum aceri-
folium. Phytochemistry 11:856–858.
SABITHA, G., SUBBA, B. V., SRIVIDYA, R., and YADAV, J. S. 1999. Lithium chloride catalysed acety-
lation of alcohols, thiols, phenols and amines. Synth. Commun. 29:2311–2315.
SONNET, P. E. 1974. cis-Olefins from the Wittig reactions. Org. Prep. Proc. Int. 6:269–273.
SWEDENBORG, P. D. and JONES, R. L. 1992. (Z)-4-Tridecanal, a pheromonally active air oxidation
product from a series of (Z,Z)-9,13 dienes in Macrocentrus grandii Goidanich (Hymenoptera:
Braconidae). J. Chem. Ecol. 18:1913–1931.
TULLOCH, A. P. 1971. Beeswax: Structure of the esters and their component hydroxy acids and diols.
Chem. Phys. Lipids 6:235–265.
WILEY. 1995. Spectral Library, 6th ed. (CD-ROM). John Wiley & Sons, New York.