L.I. Terry et al.
Phytochemistry 186 (2021) 112715
receptive cones. In 2018, treatments involved either
α
-pinene (10
μ
L per
5 column or ZB-5 equivalent was programmed from 40 to 260 ◦C at
10 ◦C minꢀ 1, with a 2 min initial delay. The oven temperature for the
septum) or prenyl acetate (10 μL per septum), and a solvent control. In
2019 the treatments were the standard mixture without prenyl croto-
nate, the standard mixture plus prenyl crotonate, and a solvent control.
Carbowax analyses was programmed from 40 to 67 ◦C at 10 ◦C minꢀ 1
,
with an initial hold for 3 min and then at 5 ◦C minꢀ 1 to 190 ◦C. A third
instrument, a Thermo Scientific ISQ 7000, fitted with a proprietary
column (TG-XLBMS) of identical dimensions to the others, was used for
analysis of synthesized standards and selected samples. The mass spectra
used for confirmation of the structure of the major unknown compound
were obtained on this instrument. Each instrument was operated at
70eV in EI mode and scanned from 35 to 350 mass units, following a 2
min solvent delay. A homologous series of alkanes was also analyzed to
enable the calculation of Kovat’s retention index (RI) for each
compound.
On the first test day, 8 μL of prenyl crotonate was added to the standard
mixture for each septum. For the second day, 40
was added per septum.
μL of prenyl crotonate
Each field bioassay was arranged as a randomized complete block of
four blocks. Each block was associated with a plant that had a dehiscing
pollen cone, ~30–50% dehisced, to provide a source of thrips. Each
treatment and control was placed 1.5 m from the source cone, and each
treatment was separated from other treatments and the control by at
least 1.5 m. To calibrate emission rates of chemical standards from the
test device we performed separate experiments that used the dynamic
head space technique with a volatile collecting bag (same size as used for
cones) placed over the bottle device. Air was supplied to the bottle de-
vice from the fish tank pump, and withdrawn from the bag into a Por-
apaq tube by a volatile sampling pump for 30 min. We also sampled the
device without any chemical standards as a control.
Peak identities were tentatively determined using either Shimadzu
software GCMS LabSolutions GCMSSolution 4.20 Shimadzu Corpora-
tion, with GCMS solutions library database NIST11 and NIST 11s or
Agilent’s MassHunter Workstation Software Qualitative Analysis vers.
B.03.01 Agilent Technologies, Inc. 2009, with mass spectrum search
NIST MS Program. Confirmation of a compound was based on its mass
spectrum matching that of a compound identified in a database library
and its calculated RI matching published data (Babushok et al., 2011,
the NIST or PUBCHEM websites, or published work, as specified in the
results). Further full confirmation of the identity of most peaks was
based upon comparison of retention times and mass spectra with pur-
chased standards or with compounds synthesized and characterized as
part of this study. Nonane and in some samples, dodecane, were added
to samples, each at 40 ng uLꢀ 1 for use as internal standards. Integrated
peaks were quantified based on equivalence to ng of nonane or dodecane
internal standard and then converted to ng minꢀ 1 emission rate based on
the sample time. The emission rate of each compound and its percent
contribution to the total volatile emissions in each sample were
calculated.
During field bioassays, each of which lasted one day, we measured
the temperature of each source pollen cone and the adjacent shaded
ambient temperature hourly, beginning around 09:00 h until ~17:00 h,
to determine the timing of peak thermogenesis. A type K thermocouple
was inserted into each cone’s core about mid-way along its length, and
temperature was measured with a Fluke 500 data logger. We also
visually observed each source cone, every 30 min, for the times when
thrips first left pollen cones and when they were no longer flying around
cones. In addition, at each treatment’s sticky trap, we counted thrips to
determine when thrips were first trapped and when no more thrips were
trapped for the day.
To confirm the timing of thermogenesis under controlled conditions,
we used the same thermocouple equipment as in the field to measure
temperatures of excised pollen cones in an environmental chamber.
(Excised cones continue to undergo daily thermogenic events in other
species (Tang, 1987; Roemer et al., 2008)). The cones were kept at a
regimen of 06:00/18:00 h, 25/15 ◦C, light/dark. Temperatures were
measured continuously over two days from 08:00 to 17:00 h.
5.5. Chemical synthesis
The four prenyl esters were synthesized at the Synthetic and Me-
dicinal Chemical Core at the University of Utah.
In the laboratory, we used a Y-tube olfactometer to test for Cycado-
thrips attraction to chemicals, as described previously (Terry et al.,
Prenyl crotonate (3-methylbut-2-enyl (E)-but-2-enoate; Fig. 4; Sup-
plementary Fig. S1a) was prepared as follows. 3-methyl-2-buten-1-ol
(1.35 mL, 13.30 mmol) was added dropwise to a solution of (E)-cro-
tonic acid (1.021 g, 11.86 mmol), N,N′-Dicyclohexylcarbodiimide
(2.506 g, 12.15 mmol), and N,N-dimethylaminopyridine (72 mg, 0.59
mmol) in dry dichloromethane (DCM) (60 mL) at room temperature
under N2, and then stirred at room temperature for 24 h. The suspension
was then filtered through Celite, and the filtrate was dry-loaded onto
Celite (8 g) and purified by flash chromatography on silica gel eluting
with DCM/hexanes to give the target compound as a colorless oil (910
mg, 5.90 mmol). RI (DB-5/Carbowax) 1161/1563, EI-MS, 70eV, m/z
(rel. int.): 154 (1.2, M+), 139 (2.6), 111(6.7), 109 (4.1), 87 (3.5), 85
(6.2), 70 (4.8), 69 (100), 68 (39.5), 67 (33.2), 53 (10.9), 41 (42), 39
(17.2) (Supplementary Fig. S2b; HRESIMS: [M + Na]+ m/z 177.0892,
calcd. for [C9H14O2Na]+ 177.0892. (Δ = + 0.6 ppm) (Supplementary
Fig. S6; 1H NMR (500 MHz, CDCl3): δ (ppm): 6.97 (dq, J = 15.4, 6.9 Hz,
1 H), 5.81–5.88 (m, 1 H), 5.33–5.40 (m, 1 H), 4.62 (d, J = 7.3 Hz, 2 H),
1.87 (dd, J = 6.8, 1.5 Hz, 3 H), 1.76 (d, J = 1.0 Hz, 3 H), 1.72 (s, 3 H)
(Supplementary Fig. S7a). 13C NMR (126 MHz, CDCl3): δ (ppm): 18.05;
18.13; 25.88; 61.20; 111.88; 122.88; 139.01; 144.60; 166.71 (Supple-
mentary Fig. S7b). The high resolution mass spectrum was obtained on a
Waters Xevo G2-S Q-TOF with Acquity UPLC. A 0.05% methanolic so-
◦
2014). An excised pollen cone was kept in a mesh cage at 15 C as a
source of thrips until testing, when the cage was moved to a room at
22 ◦C. Thrips were captured and held separately by sex. They were
tested in a dark windowless room, in nine replicates (for the high rate)
and 14 replicates (at the low rate) of 10–14 males or females, for
movement into arms of the Y-tube. The apparatus was illuminated
evenly from above. Groups were given up to 5 min to respond. We tested
a control (1 mL of castor oil) against prenyl crotonate (1 μL or 5 μL in
castor oil) and reversed their positions after three experiments. Equip-
ment was rinsed with acetone and dried when switching the Y-tube
arms, or when changing a treatment.
5.4. Cone volatile analysis
Over the five years of this study, different GC-MS instruments were
available for use. During the 2016 and 2017 coning seasons, a Shimadzu
GCMS-QP2010 Ultra Gas Chromatograph Mass Spectrometer at the
University of Queensland was used, equipped with a Zebron ZB-5 col-
umn from Phenomenex, Inc., Canada, with dimensions 30 m × 0.25 mm
ID by 0.25 μm film thickness. Some samples from 2015, 2016 and 2017,
as well as all samples from the 2018 and 2019 coning seasons, were
analyzed on an Agilent QQQ Model 7000 GCMS at the University of
Utah, Department of Chemistry Mass Spectrometry Laboratory fitted
with an Agilent DB-5 column, and some samples were also analyzed
using a Carbowax DB column, and both were of identical dimensions to
that described above. All analyses were performed by splitless injections
lution was infused into the instrument at 5 μL/min, and spectra acquired
over a 1 min period. 1H NMR spectra were acquired on a Varian VXR 500
and 13C NMR spectra on Bruker NEO500 MHz.
Prenyl methylacrylate (3-methylbut-2-enyl 2-methylprop-2-enoate;
Supplementary Fig. S1b) was synthesized as follows. Methacrylic an-
hydride (2.0 mL, 13.42 mmol) was added dropwise to a mixture of 3-
methyl-2-buten-1-ol (1.36 mL, 13.39 mmol), triethylamine (2.25 mL,
of 1 μL, with a helium flow rate of 1.5 mL. Oven temperature for the DB-
16
Terry, L. Irene
