Behaviorally active green leaf volatiles for monitoring the leaf beetle, Diorhabda elongata, a biocontrol agent of saltcedar, Tamarix spp.

Article abstract of DOI:10.1007/s10886-006-9193-x

Biological activity and chemistry of host plant volatiles were investigated for Diorhabda elongata, Brulle (Coleoptera: Chrysomelidae), a biological control agent for the invasive tree, saltcedar (Tamarix spp., Tamaricaceae). Gas chromatographic-electroantennographic detection (GC-EAD) analysis of volatiles collected from adult D. elongata feeding on saltcedar foliage or from saltcedar foliage alone showed 15 antennally active compounds. These compounds were more abundant in collections from beetle-infested foliage. Antennally active compounds were identified by GC-mass spectrometry (MS) and confirmed with authentic standards. The emissions of the most abundant GC-EAD-active compounds, green leaf volatiles (GLV), were quantitated by GC-MS. A blend of four GLV compounds, mimicking the natural blend ratio, was highly attractive to male and female D. elongata in the field, and a combination of GLV and male-produced aggregation pheromone attracted significantly greater numbers of D. elongata than did either bait alone. A preliminary experiment with a blend of seven additional GC-EAD-active saltcedar volatiles did not show any behavioral activity. The combination of the pheromone and the green leaf odor blend could be a useful attractant in detecting the presence of the biocontrol agent, D. elongata, in stands of saltcedar newly colonized by the beetle.

Full text of DOI:10.1007/s10886-006-9193-x

J Chem Ecol (2006) 32:2695–2708
DOI 10.1007/s10886-006-9193-x
Behaviorally Active Green Leaf Volatiles for Monitoring
the Leaf Beetle, Diorhabda elongata, a Biocontrol Agent
of Saltcedar, Tamarix spp.
Allard A. Cossé & Robert J. Bartelt & Bruce W. Zilkowski &
Daniel W. Bean & Earl R. Andress
Received: 21 March 2006 /Revised: 9 June 2006 /Accepted: 18 August 2006 /
Published online: 28 November 2006
#
Springer Science + Business Media, Inc. 2006
Abstract Biological activity and chemistry of host plant volatiles were investigated for
Diorhabda elongata, Brullé (Coleoptera: Chrysomelidae), a biological control agent for the
invasive tree, saltcedar (Tamarix spp., Tamaricaceae). Gas chromatographic–electro-
antennographic detection (GC-EAD) analysis of volatiles collected from adult D. elongata
feeding on saltcedar foliage or from saltcedar foliage alone showed 15 antennally active
compounds. These compounds were more abundant in collections from beetle-infested
foliage. Antennally active compounds were identified by GC–mass spectrometry (MS) and
confirmed with authentic standards. The emissions of the most abundant GC-EAD-active
compounds, green leaf volatiles (GLV), were quantitated by GC-MS. A blend of four GLV
compounds, mimicking the natural blend ratio, was highly attractive to male and female D.
elongata in the field, and a combination of GLVand male-produced aggregation pheromone
attracted significantly greater numbers of D. elongata than did either bait alone. A
preliminary experiment with a blend of seven additional GC-EAD-active saltcedar volatiles
did not show any behavioral activity. The combination of the pheromone and the green leaf
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 U.S. Department of Agriculture.
A. A. Cossé (*)
:R. J. Bartelt :B. W. Zilkowski
Crop Bioprotection Research Unit, National Center for Agricultural Utilization Research,
USDA Agricultural Research Service, 1815 N. University Street, Peoria, IL 61604, USA
e-mail: cosseaa@ncaur.usda.gov
D. W. Bean
Department of Vegetable Crops, 133 Asmundson Hall, University of California-Davis,
Davis, CA 95616, USA
E. R. Andress
USDA APHIS, 4151 Hwy 86 Bldg 10, Brawley, CA 92227, USA
Present address:
D. W. Bean
Colorado Department of Agriculture Biological Pest Control, Palisade Insectary,
750 37.8 Road, Palisade, CO 81526, USA

2696
J Chem Ecol (2006) 32:2695–2708
odor blend could be a useful attractant in detecting the presence of the biocontrol agent, D.
elongata, in stands of saltcedar newly colonized by the beetle.
.
.
.
.
Keywords Biological control Chrysomelidae Coleoptera Diorhabda elongata
.
.
.
.
Electrophysiology Field evaluation Green leaf volatiles Host-odor attractants
Saltcedar Tamarix ramosissima
.
Introduction
The leaf beetle Diorhabda elongata Brullé (Coleoptera: Chrysomelidae) has been
successfully introduced into the United States from China as a biocontrol agent for
saltcedar (Tamarix spp., Tamaricaceae) (DeLoach et al., 2003, 2004). Saltcedars were
originally imported from Eurasia as ornamentals and for erosion control of streambanks and
river channels. However, these fast-growing shrubs or small trees are invasive and have
become a major component of western riparian ecosystems (Friedman et al., 2005), causing
serious environmental and economic damages (Zavaleta, 2000; Shafroth et al., 2005).
In an earlier study (Cossé et al., 2005), we identified the male-produced aggregation
pheromone of D. elongata as a 1:1 blend of two seven-carbon compounds, (2E,4Z)-2,4-
heptadienal (2E, 4Z-7:Ald) and (2E,4Z)-2,4-heptadien-1-ol (2E, 4Z-7:OH) and demonstrat-
ed attraction to the synthetic compounds in the field. The pheromone was identified in
collections of volatiles from beetle-infested saltcedar by using electrophysiological analysis
of these collections as a key tool. We noted that the antennae of male and female beetles
responded to a number of compounds from the host foliage in addition to the pheromone
components. Here, we report the identification and quantification of several saltcedar-
emitted compounds and their electrophysiological and field activity for D. elongata adults.
Methods and Materials
Insects
Adult D. elongata used in this study were collected from a field population at Lovelock,
NV, USA, or obtained from a colony maintained by one of us (DWB) at the United States
Department of Agriculture Agricultural Research Service (USDA-ARS) Exotic and
Invasive Weed Research Unit, Albany, CA, USA. The beetles originated from sites near
Fukang, in the Xinjiang province of northwestern China (DeLoach et al., 2003; Lewis et al.,
2
003). Upon arrival at the USDA-ARS National Center for Agriculture Utilization
Research (NCAUR) in Peoria, IL, USA, adult beetles were kept at 25°C under a 17-hr
light/7-hr dark photoperiod. Tamarix ramosissima Ledebour was grown year-round in a
greenhouse facility at NCAUR to provide food for the beetles. The beetles were used for
electrophysiological tests and also to create feeding damage on saltcedar foliage.
Collections of Volatiles
Two types of collection techniques for volatiles were used: solid-phase microextraction
(SPME) and Super-Q/activated charcoal filters. The SPME collections allowed for rapid

J Chem Ecol (2006) 32:2695–2708
2697
detection of changes in saltcedar volatile profiles, whereas the Super-Q/charcoal collections
served as the source of saltcedar volatiles for compound quantification and gas
chromatographic–electroantennogram detection (GC-EAD) analysis.
Volatile saltcedar compounds, emitted during the first 30 min of experiments and
directly after beetles were placed on the saltcedar foliage, were collected simultaneously on
polydimethylsiloxane/divinylbenzene (65 μm) fibers of SPME needles (Supelco, Belle-
fonte, PA, USA) and on collection filters (Cossé et al., 2002) containing a mixture (40 mg,
1
:1) of Super-Q porous polymer (80–100 mesh, Alltech, Deerfield, IL, USA) and activated
charcoal (Darco G-60, EM Science, Cherry Hill, NJ, USA).
Collections of volatiles were made from saltcedar foliage (branch approximately 75 cm
long), either infested with beetles (about 200, both sexes but predominately males) or
without beetles. In either case, the foliage was placed in a vertical glass tube (5 cm inside
diameter [i.d.]×65 cm) with the freshly cut end protruding from the lower opening into a
beaker of water. A piece of fine mesh nylon cloth with a small central hole for the saltcedar
stem was secured around the lower end of the tube to keep the beetles contained. In some
cases, collections of volatiles were made specifically from the flowering parts on the
foliage. The collections were made at 25°C under a 17-hr light/7-hr dark photoperiod.
Using several reducing glass adapters, the top of the tube (female 55/50 ground glass
joint) was fitted with an adapter with a male 24/40 joint on one end and a threaded fitting
with an air-tight “O” ring seal on the other (Ace Glass, Vineland, NJ, USA, #5028-30). The
threaded airtight fitting held one leg of a T-shaped glass tube (0.5 cm outside diameter). A
second leg received the SPME needle through a rubber septum, and the last leg was
connected to the Super-Q/charcoal filter assembly, which was, in turn, connected to a
vacuum line (for diagram, see Bartelt and Zilkowski, 1999).
With the vacuum line, air was drawn through the collection tube and past the SPME
fiber at 25 ml/min during the SPME sampling. After 30 min, the SPME needle was
removed and the flow rate adjusted to 1 l/min for the remainder of the 24-hr collection
period. Foliage was replaced on a daily basis and any dead beetles in the collector were
replaced with fresh ones. Collected volatiles were rinsed from the filters into vials using
methylene chloride (400 μl) after every 24 hr. 1-Octanol (10 μl of a methylene chloride
solution containing 250 ng/μl) was added to each vial as a quantitative internal standard.
It was confirmed that the 6-carbon and larger compounds did not break through the
Super-Q/charcoal filters by placing a second filter in series and analyzing both separately.
However, collection filters of Super-Q or 60–80 mesh-activated charcoal (GraphTrap-GB,
Alltech) alone did not trap 6-carbon compounds quantitatively, and the very fine powdered
Darco G-60 charcoal created excessive airflow resistance in the filters. The mixture of
Super-Q and Darco G-60 charcoal resolved both problems.
Quantification
The amounts of (E)-2-hexenal (2E-6:Ald), (Z)-3-hexenal (3Z-6:Ald), (Z)-3-hexen-1-ol (3Z-
6
:OH), and (Z)-3-hexenyl acetate (3Z-6:OAc) in the collections of volatiles were quantified
by coupled GC–mass spectrometry (GC-MS) using selected ion monitoring (SIM) mode.
Two ions were monitored for each compound, m/z 83 and 98 for 2E-6:Ald, m/z 82 and 100
for 3Z-6:OH, m/z 69 and 98 for 3Z-6:Ald, and m/z 67 and 82 for 3Z-6:OAc. The selected
ions for the internal standard, 1-octanol, were m/z 70 and 84. Quantification was based on
the italicized ions of the above-mentioned ion pairs, whereas the second ions served as
qualifiers (proper ion ratios would support compound identity and purity in the GC peaks).

2698
J Chem Ecol (2006) 32:2695–2708
Serially diluted solutions of all the synthetic compounds and internal standard (0.01–30 ng/
μl) were first analyzed by GC with flame-ionization detection (GC-FID) to verify proper
concentrations and further analyzed by GC-MS in SIM mode to obtain linear calibration
curves (log dose vs. log abundance). The linear calibration curves (average of three
replicate injections for each concentration) served as the basis of the quantification. In SIM
mode, compounds were still easily detectable at concentrations as low as 10 pg/μl.
Electrophysiology
GC-EAD analyses were made by methods and equipment generally described by Cossé and
Bartelt (2000). GC-EAD connections were made by inserting a glass-pipette silver-
grounding electrode into the back of an excised beetle head. A second glass-pipette silver-
recording probe was placed in contact with the distal end of one antenna. Both pipettes
were filled with Beadle–Ephrussi (Ephrussi and Beadle, 1936) saline.
Instrumentation
Volatile collections were analyzed by GC-FID and GC-MS. Samples were injected in
splitless mode using Hewlett Packard 6890 (Palo Alto, CA, USA) instruments fitted with
3
0-m DB-1, DB-5, or DB-Wax capillary columns (0.25 mm i.d., with 1.0-, 0.25-, and 0.5-μm
film thickness, respectively, J&W Scientific, Folsom, CA, USA). Temperature programs
were from 40 to 280°C at 10°C per minute. Inlet temperatures were maintained at 280°C, and
GC-EAD effluent interface from postcolumn splitter was kept at 280°C. For SPME analysis,
the GC inlet port was fitted with an SPME-optimized glass liner (Supelco). Mass
spectrometry was performed using a Hewlett Packard 5973 instrument (electron impact,
7
0 eV). The Wiley mass spectral library, with 275,821 spectra, was available on the MS data
system (Wiley, 1995).
Chemicals
The two pheromone components, 2E,4Z-7:Ald and 2E,4Z-7:OH, were synthesized
according to Petroski (2003). The purities of the compounds were >95% by GC-FID; by
GC-MS, impurities were other geometrical isomers. 3Z-6:Ald (50% in triacetin, Bedoukian,
Danbury, CT, USA) was obtained in pure form by Kugelrohr distillation (room temperature,
4
0 mm Hg) and stored (−70°C) as a dilute solution in methylene chloride. E,E-Farnesene
was obtained by gentle reflux (4 hr) of E,E-farnesol (1 g) with Dowex 50W-X4 cation
exchange resin (2 g) in hexane (50 ml), followed by open column chromatography on silver
nitrate-coated silica (25% AgNO , elution solvent 15% 1-hexene in hexane). All other
3
compounds were purchased (Aldrich, Milwaukee, WI, USA) with purities >97%.
Field Lures
Two types of host odor lures were prepared for field testing. The first lure type consisted of
four green leaf volatiles (GLV), 3Z-6:Ald (4 mg), 2E-6:Ald (4 mg), 3Z-6:OH (0.8 mg), and
3
Z-6:OAc (0.6 mg), in mineral oil. The second lure type consisted of seven compounds,
heptanal (0.1 mg), octanal (0.25 mg), nonanal (1 mg), E,E-2,4-heptadienal (0.2 mg), E,E-
,6-nonadienal, indole (2 mg), and E,E-farnesene (2 mg), in mineral oil. A bulk solution of
the compounds needed for each lure type was prepared in mineral oil (enough of each for
2

J Chem Ecol (2006) 32:2695–2708
2699
5
1
0 lures, in a total volume of 25 ml). Then, aliquots (0.5 ml) of this mixture were placed in
.5 ml polypropylene micro centrifuge tubes (Bio Plas Inc., San Rafael, CA, USA) together
with a piece (1 cm long) of braided cotton roll (Richmond Dental, Charlotte, NC, USA).
The emission rates of freshly prepared GLV lures were measured in an incubator, using the
general reported methods and equipment reported earlier (Cossé et al., 2005). Specifically,
volatiles were collected from lures at 25°C with a 150-ml/min airflow. Volatiles were
collected periodically over 2 d, and the release rates of individual compounds were
measured by GC-FID using 1-octanol as the quantitative internal standard.
Lures were stored at −20°C until needed. Prior to field deployment, the caps of the tubes
were pierced with a single pinhole (1 mm) to allow compound release from the headspace.
Pheromone lures (Cossé et al., 2005) consisted of a 1:1 mixture of 2E,4Z-7:Ald and
2
E,4Z-7:OH (500 μg of each component), released in a nearly 1:1 ratio at 7 μg
−1
−1
component d .
Field Study
Experiments were carried out between May and September 2004 in saltcedar stands along
the Humboldt River (40°01′N, 118°31′W) 17 km southwest of Lovelock, NV, USA, where
a population of D. elongata had been established (DeLoach et al., 2004).
Yellow sticky traps (15.5×30.5 cm, AgriSense, Pontypridd, UK) were attached to
branches (oriented vertically) in the upper half of the trees, typically at a height of 2–4 m,
and were placed in trees that were at least 10 m apart and were similar in size, foliage
density, and in accessibility to beetles flying from downwind. There were always other
saltcedar trees near the treatment trees and usually between them as well. The protective
paper was removed only from one side of the trap, so that the sticky side was oriented
downwind. The traps were set out in midafternoon, when flight activity usually began to
intensify, and trap counts were made the following morning. Thus, each replicate lasted 1 d.
A wire, wrapped around the bait tube, was used to secure the lure to a trap.
The field experiments were carried out in two types of saltcedar stands. Defoliated
saltcedar consisted of saltcedar that had previously experienced complete defoliation and
had little or no foliage present during the experiment. For the experiment beginning on May
6
, 2004, the stand had experienced defoliation the previous season, and adult beetles had
overwintered in the leaf litter beneath the stand. They were newly emerged from
overwintering when the experiment was performed. Later in the season, newly defoliated
saltcedar stands were the result primarily of heavy infestation with D. elongata larvae. The
second stand type had little or no previous defoliation; adults were just beginning to move
into these stands, whose trees had relatively lush green foliage during the tests.
Three sets of experiments were deployed. The first set was a preliminary assessment of
the host odor lures. The two types of host odor lures (four-component and seven-
component) were first combined on traps and compared to unbaited control traps in a paired
design (N=12), followed by the pairwise comparison of only the four-component GLV
lures and unbaited controls (N=6). In addition, there was one comparison of the seven-
component lure vs. unbaited control.
The second experiment, deployed in both saltcedar stand types, was a more systematic
pairwise comparison of the GLV lures and unbaited controls, throughout the 2004 season,
starting on May 6, when the overwintered beetles were becoming active, and finishing on
September 14, when all of the beetles had entered reproductive diapause (DeLoach et al.,
2
004). The first trapping day in both types of saltcedar stands had 18 replications per

2700
J Chem Ecol (2006) 32:2695–2708
treatment followed by six replications per treatment for all other days. Trapped beetles were
sexed periodically (Cossé et al., 2005). Sweep net samples were taken on June 28 in both
types of saltcedar stands and included both flying beetles and those present on foliage.
Captured beetles were counted and sexed.
The third experiment measured the effects of combining pheromone and GLV lures and
was set up on August 11, 2004, at least 200 m away from the nearest saltcedar trees. Traps
used in this experiment were two-sided yellow sticky panel traps (15.2×30.5 cm, both sides
exposed, Seabright Laboratories, Emeryville, CA, USA). Treatments were pheromone lure,
GLV lure, pheromone and GLV lures combined, and unbaited control. A randomized
complete block design was used (four traps per block), and there were four replicates of
each treatment.
Statistical Analyses
Counts of D. elongata adults on traps were transformed using log(X+1) to stabilize
variance. The data from the pairwise experiments were analyzed by paired t-tests, whereas
data from the randomized complete block design were submitted to analysis of variance and
means were compared using the least significant difference test.
Results
GC-EAD Analysis of Saltcedar Volatiles
GC-EAD comparisons of volatiles collected by Super-Q/charcoal from beetle-infested
saltcedar foliage and foliage alone showed that male and female antennae responded to a
variety of compounds (Fig. 1). Compounds that triggered male and female antennal
responses were identified as 3Z-6:Ald, 2E-6:Ald, 3Z-6:OH, 2E-6:OH, 1-hexanol, heptanal,
3Z-6:OAc, octanal, (2E,4E)-2,4-heptadienal, benzyl alcohol, nonanal, (2E,6Z)-2,6-non-
adienal, indole, and (E,E)-farnesene. By GC-MS, only trace amounts of the same
compounds could be detected in collections of volatiles from uninfested saltcedar, and
the antennal activity to these was in most cases inconclusive. More consistent GC-EAD
activity from uninfested saltcedar could be obtained from collections of volatiles of crushed
saltcedar foliage followed by sample concentration. Additional collections of volatiles from
the flowering parts of saltcedar showed a relative abundance of benzyl alcohol, which was
easily detected by the antennae of both sexes. Male and female antennae responded also to
the two pheromone compounds, 2E,4Z-7:Ald and 2E,4Z-7:OH, which were present in
beetle-infested foliage collections. Identities of GC-EAD-active compounds were verified
by GC-MS and GC-EAD analysis of authentic standards.
Collections of Volatiles by SPME
Thirty minutes after placing the beetles on saltcedar foliage, several GLV compounds were
already easily collected by SPME and detected by GC-MS (Fig. 2A), whereas these
compounds were almost absent in the SPME collections from uninfested saltcedar foliage
(Fig. 2B). Compounds that were consistently more abundant in the beetle-infested SPME
collections were 3Z-6:Ald, 2E-6:Ald, 3Z-6:OH, 3Z-6:OAc (coeluting with compound d in
Fig. 2A), nonanal, and decanal. SPME collections from uninfested saltcedar showed only

J Chem Ecol (2006) 32:2695–2708
2701
Fig. 1 Simultaneously recorded
gas chromatogram with FID and
EAD of female D. elongata an-
tenna to volatiles collected from
male and female D. elongata
adults feeding on saltcedar. Let-
ters denote antennal activity by:
(
(
Z)-3-hexenal (a), (E)-2-hexenal
b), (Z)-3-hexen-1-ol (c), (E)-2-
hexen-1-ol (d), 1-hexanol (e), (Z)-
-hexenyl acetate (f), octanal (g),
2E,4E)-2,4-heptadienal (h), ben-
zyl alcohol (i), nonanal (j),
2E,6Z)-2,6-nonadienal (k), in-
dole (l), and (E,E)-farnesene (m)
3
(
(
m
e
g
h
k
i
a
d
b
j
l
c
f
2:3 3:3 4:3 5:3 6:3 7:3 8:3 9:3 10:3 11:3 12:3
trace amounts of 3Z-6:Ald, 2E-6:Ald, nonanal, and decanal. In addition, SPME collections
from beetle-infested saltcedar also showed the presence of the two male-produced
pheromone components, 2E,4Z-7:Ald and 2E,4Z-7:OH. The most abundant and earliest
detectable GC-EAD-active compounds in collections of volatiles from beetle-infested
saltcedar were selected for further analysis.
Quantification of GLVs from Saltcedar
Quantification of four GLVs, 3Z-6:Ald, 2E-6:Ald, 3Z-6:OH, and 3Z-6:OAc, collected from
saltcedar, with and without the feeding beetles, revealed that beetle-infested saltcedar
foliage emitted about 50 times more of the four compounds than did uninfested foliage
(Table 1). Nearly equivalent amounts of each of the four compounds were released by
uninfested foliage, whereas proportionally greater amounts of 2E-6:Ald and lesser amounts

2702
J Chem Ecol (2006) 32:2695–2708
9
6
.E+06
.E+06
A
b
a
9
3
.E+06
.E+06
d
e
6
0
.E+06
.E+00
f
g
c
6
8
10
12
14
16
18
20
3
0
.E+06
.E+00
B
6
8
10
12
14
16
18
20
Time (min)
Fig. 2 GC-MS profiles of SPME-collected volatiles during the first 30 min of a 24-hr collection of volatiles
from (A) saltcedar foliage infested with feeding adult D. elongata and (B) saltcedar foliage alone. Letters
above the GC-MS peaks denote the presence of (Z)-3-hexenal (a), (E)-2-hexenal (b), (Z)-3-hexen-1-ol (c),
(
2E,4Z)-2,4-heptadienal (d), (2E,4Z)-2,4-heptadien-1-ol (e), nonanal (f), and decanal (g)
of 3Z-6:Ald and 3Z-6:OAc were released by infested foliage. Preliminary analysis of Super
Q/charcoal collected saltcedar emissions (Cossé et al., 2005) showed that some of the less
volatile compounds (2,6-nonadienal, indole, and E,E-farnesene) could only be detected
towards the end of the 24-hr collection period.
Field Lures and Field Study
After 24 hr, the GLV field lures released approximately 30 μg/hr and in a ratio of nearly
3
2:3:1:1 (2E-6:Ald/3Z-6:OH/3Z-6:Ald/3Z-6:OAc; Table 2). Preliminary field trapping
experiments captured 314±152 (Mean ± SD, N=12) beetles per trap per day in response
to the two host odor baits combined, whereas the unbaited traps caught 32±60 (Mean ± SD,
N=12) beetles per trap per day. The GLV lure alone caught 254±94 (Mean ± SD, N=6)

J Chem Ecol (2006) 32:2695–2708
2703
Table 1 Quantities of four GLVs released from saltcedar, Tamarix ramosissima, foliage infested with and
without adult leaf beetles, Diorhabda elongata
a
Compound
Foliage+200 Beetles
Foliage
Mean ± SD (μg/d) (N=8)
%
Mean ± SD (μg/d) (N=5)
%
3
2
3
3
Z-6:Ald
E-6:Ald
Z-6:OH
Z-6:OAc
1.26±0.57
16.97±4.90
6.17±3.71
2.48±1.02
26.87±8.57
4.7
0.13±0.02
0.11±0.06
0.13±0.07
0.15±0.08
0.51±0.17
24.3
21.3
24.6
29.8
63.1
23.0
9.2
Total
a
For compound abbreviations, see text.
beetles per trap per day, whereas the unbaited traps caught 12±24 (Mean ± SD, N=6)
beetles per trap per day. A single pair of traps comparing the trap catch of the seven-
compound host odor lure (16 beetles per trap per day) with that of the unbaited control (five
beetles per trap per day) elicited a much lower response. Therefore, the GLV lure was
selected for further field testing.
The second set of field trapping experiments demonstrated the attractiveness of the GLV
lures to male and female D. elongata throughout the season and in the two types of
saltcedar stands (Table 3). Significantly more adults were attracted to traps baited with GLV
compared to the control traps during seven different periods in defoliated saltcedar stands.
Similar results were obtained in saltcedar stands with abundant foliage during four trapping
periods. In both types of saltcedar stands, the response ratio between GLV and control
treatments was higher early in the season and slowly declined throughout the season
(Table 3). Relatively high numbers of beetles were trapped on May 6 and August 12 in the
defoliated saltcedar stands.
Trap counts showed a strong bias towards males in both baited and unbaited traps
(Table 3). Similar male bias was found in aerial sweep net samples, collecting 38 males and
nine females in areas with foliage and seven males and three females in defoliated areas.
The observed sex ratio was consistent between the two saltcedar foliage categories (2×2
2
contingency table, X =0.58, 1 df, P=0.44; overall, 79% males). Sweep net samples of
beetles from saltcedar trees showed a considerably more balanced sex ratio, with 43 males
and 26 females from defoliated saltcedars (from 30 plants) and 29 males and 24 females
from saltcedars with foliage (from 15 plants). Again, the sex ratio was not significantly
2
different between the two saltcedar stand types (2×2 contingency table, X =0.72, 1 df, P=
0
.40; overall, 59% males). However, the proportion of males in the aerial sample was
significantly higher than in the sample from the trees (after summing over stand type in the
Table 2 Quantities of four GLVs emitted from field lures for Diorhabda elongata
Hour
Total μg/hr
Mean ± SD (μg/hr) (N=3)
3Z-6:Ald
%
2E-6:Ald
%
3Z-6:OH
%
3Z-6:OAc
%
1
5
2
4
32.1
38.1
29.4
22.3
0.8±0.2
1.6±0.5
1.0±0.7
0.6±0.2
2.5
4.2
3.4
2.7
27.1±3.9
31.6±3.6
25.1±4.5
20.9±1.8
84.4
82.9
85.4
93.7
3.6±0.5
4.1±0.4
2.6±0.4
2.1±0.3
11.2
10.8
8.8
0.6±0.1
0.8±0.01
0.8±0.2
0.6±0.1
1.9
2.1
2.7
2.7
4
8
9.4
For compound abbreviations and applied dosages, see text.

2704
J Chem Ecol (2006) 32:2695–2708
Table 3 Response of Diorhabda elongata to sticky traps baited with four GLVs in defoliated and foliated
saltcedar, Tamarix ramosissima, stands in Lovelock, NV, May–September 2004
a
Date
Response
GLV/Ctrl
% Male Catch
Mean ± SD (Beetles per Trap per Day)
GLV
Defoliated Saltcedar
Ctrl
t
GLV
Ctrl
6
1
2
2
2
1
1
-May
293.9±135.4
8.7±6.8
22.1±50.8
0.8±1.2
1.0±1.3
5.2±5.0
2.7±2.8
30.4±15.9
1.0±0.9
12.14**
3.46*
4.87*
11.01**
7.59**
8.76**
7.99*
42.4
11.2
12.9
12.1
8.5
–
–
8-June
8-June
1-July
3-July
2-Aug.
4-Sept.
–
–
13.7±13.9
46.6±20.4
20.8±12.8
156.6±45.3
60.8±75.5
96.5
95.8
89.0
–
100
73.7
90.5
–
6.0
43.9
77.3
100
Saltcedar with Foliage
1
2
6
2
8-June
8-June
-July
34.7±43.2
11.5±8.8
19.4±20.0
51.3±59.9
1.0±2.4
1.2±1.2
4.1±5.2
20.2±34.2
3.54*
7.24**
3.81*
4.18*
7.2
10.1
4.2
4.0
–
–
96.6
–
92.1
100
–
65.4
5-July
For bait composition and emission rates, see Table 2.
a
From the statistical analysis, ratios are based on log-transformed means.
*P<0.01.
**P<0.001.
2
above data sets, a 2×2 contingency table for sex vs. aerial/tree sampling gave X =6.82, 1
df, P<0.01).
In the third field experiment, adult D. elongata were attracted to traps baited with
pheromone, GLV, or both, in numbers that were significantly higher than those found on the
unbaited control traps (Table 4). There was no significant difference between the numbers
of D. elongata attracted to traps baited with GLV lures compared to those attracted to traps
baited with pheromone lures. The combination of GLV and pheromone lures attracted the
highest number of beetles, with approximately six times more D. elongata than to GLV
lures alone, and approximately four times more D. elongata than to pheromone lures alone
(Table 4).
Table 4 Response of Diorhabda elongata to sticky traps baited with four GLVs (for bait composition and
emission rates, see Table 2), a 1:1 mixture of 2E,4Z-7:ALD and 2E,4Z-7:OH (pheromone) (for bait
composition and emission rates, see text), and a combination of GLV and pheromone in saltcedar, Tamarix
ramosissima, stands in Lovelock, NV, August 12, 2004
Lure
Response
Mean ± SD (beetles per trap per day)
a
Pheromone+GLV
Pheromone
GLV
90.4a
23.7b
15.9b
Control
3.1c
F statistic
P
24.28 (df=3,12)
<0.001
a
Means followed by the same letter do not differ by least significant difference test, P>0.01.

J Chem Ecol (2006) 32:2695–2708
2705
Discussion
In this study, we focused on possible biologically active saltcedar volatiles released during
feeding by D. elongata. The experimental results demonstrated that at least 15 volatile
compounds were antennally active to male and female D. elongata. Collections of volatiles
from uninfested saltcedar showed relatively small amounts of GC-EAD-active material, but
collections from beetle-infested foliage showed that the feeding activity of the beetles
significantly increases the amounts of GC-EAD-active saltcedar volatiles released into the air.
GLVs (Visser and Avé, 1978; Visser, 1986) especially were released in relative abundance
and could be easily detected during the first 30 min after feeding began, whereas relatively
lesser amounts could only be detected in the samples collected over a 24-hr period.
Numerous studies have demonstrated the quantitative and qualitative differences between
volatiles emitted by herbivore-infested plants compared to the volatiles emitted by uninfested
plants (for reviews, see Dicke et al., 1998; Dicke and Vet, 1999; Dicke and van Loon, 2000),
and several of these studies have shown the role of the identified GC-EAD-active compounds
in tritrophic interactions (for reviews, see above). In this study, we focused only on GC-EAD-
active material. Qualitative differences in these compounds were not noted between beetle-
infested and noninfested plants, but the quantitative differences were dramatic.
The behavioral significance of the identified GLVs was demonstrated in the field. A
blend of four of the most abundant, antennally active GLVs, released at compound ratios
mimicking the beetle-infested plant release ratios, was highly attractive to male and female
D. elongata in areas where most of the saltcedar stands were completely defoliated.
Furthermore, this synthetic blend was still highly attractive to male (and, to a lesser extent,
female) D. elongata in areas with an abundance of beetles and saltcedar foliage.
Further evidence for the importance of the GLV blend was demonstrated with the
observed synergistic effect when pheromone and GLVs were combined (Table 4). It should
be noted that this synergistic effect was demonstrated in previously uninfested saltcedar
stands. As such, the combination of the pheromone and the green leaf odor blend could be a
useful attractant in detecting the presence of D. elongata in newly colonized areas. This
would be a useful tool in the saltcedar biocontrol program.
In beetle taxa with aggregation pheromones, pheromone emission occurs at the feeding/
breeding site, and host-derived volatiles are potent synergists of the pheromones for some taxa
(Borden, 1985; Bartelt, 1999). Only recently has the synergistic interaction of host compounds
and pheromones been shown for chrysomelid beetles (Soroka et al., 2005; Tóth et al., 2005).
Field observation of D. elongata showed that the population density is not uniformly
distributed among the saltcedar bushes. Especially in newly colonized stands of saltcedar,
the presence of adult beetles is highly aggregated with, for example, a single plant
containing hundreds of feeding beetles being surrounded by many other similar-looking
plants without any beetles. In those instances where a small number of beetles landed and
started feeding on a previously uninfested saltcedar bush, it was quickly (<30 min)
followed by the arrival of numerous other beetles on the same saltcedar branch (personal
observations by authors Cossé, Bartelt, and Bean). If the original group of pioneer beetles
contained males, one could envision that the arrival of the additional beetles was caused by
the release of the male-specific aggregation pheromone, the release of attractive host
volatiles as a result of the feeding action, or both stimuli.
Our earlier study dealing with the identification of the male-specific aggregation
pheromone of D. elongata showed that males feeding on saltcedar foliage released a 1:1
mixture of 2E,4Z-7:Ald and 2E,4Z-7:OH, and that this mixture attracted significant
numbers of males and females to traps placed in saltcedar stands in various stages of

2706
J Chem Ecol (2006) 32:2695–2708
defoliation (Cossé et al., 2005). In this study, the pheromone could be detected within the
first 30 min after placing the beetles on the saltcedar foliage in the collector.
Certain chrysomelid beetles, such as Diabrotica spp. (for review, see Metcalf and
Metcalf, 1992) and Leptinotarsa decemlineata (Dickens, 1999, 2000, 2006), are attracted to
blends of some of the same GC-EAD-active compounds found in this study. For example,
benzyl alcohol and indole, two volatile components associated with the blossoms of
Cucurbita spp., were both antennally active and attractive to adult Diabrotica undecim-
punctata howardi and Diabrotica virgifera virgifera in field studies (Andersen and Metcalf,
1
986; Lampman et al., 1987). Antennae of L. decemlineata responded to several volatile
potato foliage compounds, including 3Z-6:OH, 2E-6:OH, 3Z-6:OAc, nonanal, decanal, and
indole (Dickens, 1999). However, Dickens (2000) showed that some blends containing the
GLVs 3Z-6:OH and 2E-6:OH can also be unattractive or repellent for L. decemlineata.
Also, cereal leaf beetles, Oulema melanopus, did not respond to a GC-EAD-active oat
volatile, 3Z-6:OAc, in field tests (Rao et al., 2003). Behavioral activity to GLVs has also
been demonstrated with other Coleoptera; for example, scarab beetles, Melolontha
melolontha (Reinecke et al., 2005) and Phyllopertha horticola (Ruther and Mayer, 2005),
are strongly attracted to single or blends of specific GLV compounds. In addition, GLVs
significantly enhanced the responses of the smaller European elm bark beetle, Scolytus
multistriatus, and boll weevil, Anthonomus grandis, to their respective pheromones
(Dickens et al., 1990).
Future field tests with D. elongata will be necessary, using single components and
different blends and blend ratios of saltcedar volatiles, to fully investigate the behavioral
significance of all of the EAD-active compounds.
The field bioassays showed a male-biased sex ratio that was surprisingly strong on some
of the trapping dates. It is possible that males were more attracted to the host plant odors,
but with similar male sex ratios in unbaited control traps and aerial sweeps, a more likely
explanation is that males are simply the more active fliers, especially because the beetle sex
ratio on the foliage was far less male-biased. In previous field tests of the aggregation
pheromone blend, only a slight male bias was noted in the captured beetles (Cossé et al.,
2
005). Again, it is possible that males are more strongly attracted to host plant odors than
females and that this differential attraction is not as pronounced with aggregation
pheromones. A more likely explanation is that the behavioral, physiological, and ecological
context of these experiments is critical in the outcome, both in terms of attraction to the
baits and the sex ratio of captured beetles. For instance, placement of traps outside of the
areas where beetles are mating and ovipositing might bias the sex ratio toward males, if
they are the primary colonizers. The physiological status of the beetles, whether it be
reproductive, diapause, hungry, or fed, will also undoubtedly have a strong influence on
response to behaviorally active compounds.
Future field studies might further investigate the hypothesis that male D. elongata are
pioneers in the colonization of saltcedar and that successful establishment of new
populations is being guided by both the release of the male-produced aggregation
pheromone and some of the volatile compounds released by the host plants.
References
ANDERSEN, J. F. and METCALF, R. L. 1986. Identification of a volatile attractant for Diabrotica and
Acalymma species from the blossoms of Cucurbita maxima Duchesne. J. Chem. Ecol. 12:687–699.
BARTELT, R. J. 1999. Sap beetles, pp. 69–89, in J. Hardie and A. K. Minks (eds.). Pheromones of Non-
Lepidopteran Insects Associated with Agricultural Plants. CABI Publishing, Wallingford.

J Chem Ecol (2006) 32:2695–2708
2707
BARTELT, R. J. and ZILKOWSKI, B. W. 1999. Nonequilibrium quantitation of volatiles in air streams by solid-
phase microextraction. Anal. Chem. 71:92–101.
BORDEN, J. H. 1985. Aggregation pheromones, pp. 257–285, in G. A. Kerkut and L. I. Gilbert (eds.).
Comprehensive Insect Physiology and Pharmacology, Vol. 9. Pergamon Press, Oxford.
COSSÉ, A. A. and BARTELT, R. J. 2000. Male-produced aggregation pheromone of Colopterus truncatus:
structure, electrophysiological and behavioral activity. J. Chem. Ecol. 26:1735–1748.
COSSÉ, A. A., BARTELT, R. J., and ZILKOWSKI, B. W. 2002. Identification and electrophysiological activity of
a novel hydroxy ketone emitted by male cereal leaf beetle. J. Nat. Prod. 65:1156–1160.
COSSÉ, A. A., BARTELT, R. J., ZILKOWSKI, B. W., BEAN, D. W., and PETROSKI, R. J. 2005. The aggregation
pheromone of Diorhabda elongata, a biological control agent of saltcedar (Tamarix spp.): identification
of two behaviorally active components. J. Chem. Ecol. 31:657–670.
DELOACH, C. J., LEWIS, P. A., CARRUTHERS, R. I., HERR, J. C., TRACY, J. L., and JOHNSON, J. 2003. Host
specificity of a leaf beetle, Diorhabda elongata (Coleoptera: Chrysomelidae) from Asia, for biological
control of saltcedars (Tamarix: Tamaricaceae) in the western United States. Biol. Control 27:117–147.
DELOACH, C. J., CARRUTHERS, R., DUDLEY, T., EBERTS, D., KAZMER, D., KNUTSON, A., BEAN, D., KNIGHT,
J., LEWIS, P., TRACY, J., HERR, J., ABBOT, G., PRESTWICH, S., ADAMS, G., MITYAEV, I., JASHENKO, R.,
Li, B., SOBHIAN, R., KIRK, A., ROBBINS, T., and DELFOSSE, E. 2004. First results for control of saltcedar
(Tamarix spp.) in the open field in the western United States. Eleventh International Symposium on
Biological Control of Weeds, Canberra, Australia.
DICKE, M. and VAN LOON, J. A. 2000. Multitrophic effects of herbivore-induced plant volatiles in an
evolutionary context. Entomol. Exp. Appl. 97:237–249.
DICKE, M. and VET, L. E. M. 1999. Plant–carnivore interactions: evolutionary and ecological consequences
for plant, herbivore, and carnivore, pp. 483–520, in H. Olff, V. K. Brown, and R. H. Drent (eds.).
Herbivores: Between Plants and Predators. Blackwell Science, Oxford.
DICKE, M., TAKABAYASHI, J., POSTHUMUS, M. A., SCHÜTTE, C., and KRIPS, O. E. 1998. Plant–phytoseiid
interactions mediated by herbivore-induced plant volatiles: variation in production of cues and in
responses of predatory mites. Exp. Appl. Acarol. 22:311–333.
DICKENS, J. C. 1999. Predator–prey interactions: olfactory adaptations of generalist and specialist predators.
Agric. For. Entomol. 1:47–54.
DICKENS, J. C. 2000. Orientation of Colorado potato beetle to natural and synthetic blends of volatiles
emitted by potato plants. Agric. For. Entomol. 2:167–172.
DICKENS, J. C. 2006. Plant volatiles moderate response to aggregation pheromone in Colorado potato beetle.
J. Appl. Entomol. 130:26–31.
DICKENS, J. C., JANG, E. B., LIGHT, D. M., and ALFORD, A. R. 1990. Enhancement of insect pheromone
response by green leaf volatiles. Naturwissenschaften 77:29–31.
EPHRUSSI, B. and BEADLE, G. W. 1936. A technique of transplantation for Drosophila. Am. Nat. 70:218–
225.
FRIEDMAN, J. M., AUBLE, G. T., SHAFROTH, P. B., SCOTT, M. L., MERIGLIANO, M. F., FREEHLING, M. D.,
and GRIFFIN, E. R. 2005. Dominance of non-native riparian trees in western USA. Biol. Invasions
7:747–751.
LAMPMAN, R. L., METCALF, R. L., and ANDERSEN, J. F. 1987. Semiochemical attractants of Diabrotica
undecimpunctata howardi Barber, southern corn rootworm, and Diabrotica virgifera virgifera LeConte,
the western corn rootworm (Coleoptera: Chrysomelidae). J. Chem. Ecol. 13:959–975.
LEWIS, P. A., DELOACH, C. J., KNUTSON, A. E., and TRACY, J. L. 2003. Biology of Diorhabda elongata
(
Coleoptera: Chrysomelidae), an Asian leaf beetle for biological control of saltcedars (Tamarix spp.) in
the United States. Biol. Control 27:101–116.
METCALF, R. L. and METCALF, E. R. 1992. Plant Kairomones in Insect Ecology and Control. Chapman and
Hall, New York, NY.
PETROSKI, R. J. 2003. Straightforward preparation of (2E,4Z)-2,4-heptadien-1-ol and (2E,4Z)-2,4-
heptadienal. Synth. Commun. 33:3233–3241.
RAO, S., COSSÉ, A. A., ZILKOWSKI, B. W., and BARTELT, R. J. 2003. Aggregation pheromone of the
cereal leaf beetle: field evaluation and emission from males in the laboratory. J. Chem. Ecol.
2
9:2165–2175.
REINECKE, A., RUTHER, J., and HILKER, M. 2005. Electrophysiological and behavioral responses of
Melolontha melolontha to saturated and unsaturated aliphatic alcohols. Entomol. Exp. Appl. 115:33–40.
RUTHER, J. and MAYER, C. J. 2005. Response of garden chafer, Phyllopertha horticola, to plant volatiles:
from screening to application. Entomol. Exp. Appl. 115:51–59.
SHAFROTH, P. B., CLEVERLY, J. R., DUDLEY, T. L., TAYLOR, J. P., VAN RIPER, C., WEEKS, E. P., and STUART,
J. N. 2005. Control of Tamarix in the western United States: implications for water salvage, wildlife use,
and riparian restoration. Environ. Manage. 35:231–246.

2708
J Chem Ecol (2006) 32:2695–2708
SOROKA, J. J., BARTELT, R. J., ZILKOWSKI, B. W., and COSSÉ, A. A. 2005. Responses of flea beetle Phyllotreta
cruciferae to synthetic aggregation pheromone components and host plant volatiles in field trials.
J. Chem. Ecol. 31:1829–1843.
TÓTH, M., CSONKA, E., BARTELT, R. J., COSSÉ, A. A., ZILKOWSKI, B. W., MUTO, S., and MORI, K. 2005.
Pheromonal activity of compounds identified from male Phyllotreta cruciferae: field tests of racemic
mixtures, pure enantiomers, and combinations with allyl isothiocyanate. J. Chem. Ecol. 31:2705–2720.
VISSER, J. H. 1986. Host odor perception in phytophagous insects. Annu. Rev. Entomol. 31:121–144.
VISSER, J. H. and AVÉ, D. A. 1978. General green leaf volatiles in the olfactory orientation of Colorado
beetle, Leptinotarsa decemlineata. Entomol. Exp. Appl. 24:738–749.
WILEY SPECTRAL LIBRARY. 1995. 6th edition, CD-ROM. John Wiley and Sons, New York, NY.
ZAVALETA, E. 2000. The economic value of controlling an invasive shrub. Ambio 29:462–467.