Experimental spiders
Experimental spiders were maintained as previously reported37. Briefly, spiders were the F1 to F4 offspring of mated females collected from hallways of the Burnaby campus of Simon Fraser University (Burnaby, BC, CA). Upon hatching, juvenile spiders were housed individually in petri dishes (100 mm × 20 mm) and provisioned with the vinegar flies Drosophila melanogaster. Subadult spiders were fed with larvae of the mealworm beetle Tenebrio molitor. Each adult female spider was kept in a separate translucent 300-mL plastic cup (Western Family, CA) maintained at 22 °C under a reversed light cycle (12:12 h). Adult males and females were fed with black blow flies, Phormia regina. All spiders had access to water in cotton wicks. Water and food were provided once per week. Laboratory experiments were run during a reversed scotophase (0900 to 1700).
Identification of contact pheromone components: Preparation of web extracts (summer 2017; spring and summer 2018)
Each of the 100 spiders was allowed to build her web for three days on a wooden triangular prism scaffold (30 cm × 25 cm × 22 cm)44 of bamboo skewers (GoodCook, CA, USA) (Fig. 1b). After the spiders were removed from the scaffold, their webs were reeled up with a glass rod (10 cm × 0.5 cm) and deposited in a 1.5-mL glass vial. Per web, 50 µL of methanol (99.9% HPLC grade, Fisher Chemical, ON, Canada) were added and the silk was extracted for 24 h at room temperature. Prior to analysis, the silk was removed and the sample was concentrated under a steady nitrogen stream to the desired concentration.
Identification of contact pheromone components: analyses of web extracts by gas chromatography–mass spectrometry (GC-MS)
Aliquots (2 µL) of pooled and concentrated web extract (100 webs in 400 µL of solvent) were analysed by GC–MS, using a Varian Saturn Ion trap 2000 (Varian Inc., now Agilent Technologies Inc., Santa Clara, CA 95051, USA) and an Agilent 7890B GC coupled to a 5977 A MSD, both fitted with a DB-5 GC-MS column (30 m × 0.25 mm ID, film thickness 0.25 µm). The injector port was set to 250 °C, the MS source to 230 °C, and the MS quadrupole to 150 °C. Helium was used as a carrier gas at a flow rate of 35 cm s−1, with the following temperature programme: 50 °C held for 5 min, 10 °C min−1 to 280 °C (held for 10 min). Compounds were identified by comparing their mass spectra and retention indices (relative to aliphatic alkanes67) with those of authentic standards that were purchased or synthesised in our laboratory (Supplementary Table 1).
Identification of contact pheromone components: high-performance liquid chromatography (HPLC) of web extracts
Web extract of virgin adult female S. grossa was fractionated by high-performance liquid chromatography (HPLC), using a Waters HPLC system (Waters Corporation, Milford, MA, USA; 600 Controller, 2487 Dual Absorbance Detector, Delta 600 pump) fitted with a Synergy Hydro Reverse Phase C18 column (250 mm × 4.6 mm, 4 µ; Phenomenex, Torrance, CA, USA). The column was eluted with a 1-mL/min flow of a solvent gradient, starting with 80% water (HPLC grade, EMD Millipore Corp., Burlington, MA, USA) and 20% acetonitrile (99.9% HPLC grade, Fisher Chemical, Ottawa, CA) and ending with acetonitrile after 10 min. A 60-web-equivalent extract was injected and 20 1-min fractions were collected. Each HPLC fraction (containing 20 web-equivalents) was tested in T-rod bioassays (Fig. 1c) for the presence of contact pheromone components. All eight fractions that elicited courtship responses by males (Supplementary Fig. 1) were analysed by HPLC-tandem MS/MS.
Identification of contact pheromone components: HPLC-tandem MS/MS of bioactive HPLC fractions
The bioactive HPLC fractions were analysed on a Bruker maXis Impact Quadrupole Time-of-Flight HPLC/MS System. The system consists of an Agilent 1200 HPLC fitted with a spursil C18 column (30 mm × 3.0 mm, 3 µ; Dikma Technologies, Foothill Ranch, CA, USA) and a Bruker maXis Impact Ultra-High Resolution tandem TOF (UHR-Qq-TOF) mass spectrometer. The LC-MS conditions were as follows: The mass spectrometer was set to positive electrospray ionisation (+ESI) with a gas temperature of 200 °C and a gas flow of 9 L/min. The nebuliser was set to 4 bar and the capillary voltage to 4200 V. The column was eluted with a 0.4-mL/min flow of a solvent gradient, starting with 80% water and 20% acetonitrile and ending with 100% acetonitrile after 4 min. The solvent contained 0.1% formic acid to improve peak shape.
Identification of contact pheromone components: 1H NMR analyses of a bioactive fraction
In HPLC-MS analyses, a single bioactive fraction (9–10 min) appeared to contain only a single compound. This fraction was then further investigated using 1H NMR spectroscopy. The 1H NMR spectrum was recorded on a Bruker Advance 600 equipped with a QNP (600 MHz) using CDCl3. Signal positions (δ) are given in parts per million from tetramethylsilane (δ 0) and were measured relative to the signal of the solvent (1H NMR: CDCl3: δ 7.26).
Identification of contact pheromone components: syntheses of candidate pheromone components
The syntheses of candidate pheromone components and synthetic intermediates are reported in the SI.
Identification of contact pheromone components: T-rod bioassays (general procedures)
The T-rod apparatus37 (Fig. 1c) consisted of a horizontal beam (25 cm × 0.4 cm) and a vertical beam (30 cm × 0.4 cm) held together by labelling tape (3 cm × 1.9 cm, Fisher Scientific, Ottawa, ON, CA). A piece of filter paper (2 cm2) was attached to each distal end of the horizontal beam. For each bioassay, an aliquot of web extract (in methanol), or a blend of synthetic candidate pheromone components, was applied to the randomly assigned treatment filter paper, whereas methanol was applied to the control filter paper. The solvent was allowed to evaporate for 1 min before the onset of a 15-min bioassay. A randomly selected naïve male spider was placed at the base of the vertical beam and the time he spent courting on each filter paper was recorded. In response to the presence of female-produced or synthetic pheromone on a filter paper, the male engaged in courtship, pulling silk with his hindlegs from his spinnerets and adding it to the paper. Sensing contact pheromone, the male essentially behaves as if he were courting on the web of a female. On a web, the male engages in web reduction prior to copulation, a behaviour that entails cutting sections of the female’s web with his chelicerae and wrapping the dismantled web bundle with his own silk pulled from his spinnerets41,56. Each T-rod apparatus was used only once. Replicates of experiments as part of specific research objectives were run in parallel to eliminate day effects on the responses of spiders. The sample size for each experiment was set to 20 unless otherwise stated.
Identification of contact pheromone components: T-rod bioassays (specific experiments) (fall 2017; spring and summer 2018)
Experiment 1 (fall 2017) tested a synthetic blend of volatile compounds 5–11 unique to mature S. grossa females (Fig. 1c and Supplementary Table 1) vs a solvent control. Parallel experiment 2 tested one web equivalent of virgin female web extract, followed by testing each of the 20 HPLC fractions in six replicates for the occurrence of courtship (spring 2018).
Parallel experiments 3–6 (summer 2018) tested web extract at one female web equivalent (1 FWE) (Exp. 3), a ternary blend of the candidate contact pheromone components 12, 16 and 17 (Fig. 2d, Exp. 4), the same ternary blend (12, 16 and 17) in combination with the volatile compounds 5–11 (Exp. 5), and 5–11 on their own (Exp. 6).
Parallel dose-response experiments 7–11 (summer 2018) tested the ternary blend of 12, 16 and 17 at five FWEs: 0.001 (Exp. 7); 0.01 (Exp. 8); 0.1 (Exp. 9); 1.0 (Exp. 10); and 10 (Exp. 11).
Parallel experiments 12–15 tested the ternary blend, and all possible binary blends, of 12, 16 and 17. Parallel experiments 16–18 tested 12 and 16 in binary combination (Exp. 16) and singly (Exps. 17, 18).
Origin of contact pheromone components (fall 2020)
To trace the origin of contact pheromone component 12 (and coeluting 16), cold-euthanized female spiders were dissected in saline solution55 (25 mL of water and 25 mL of methanol, 160 mM NaCl, 7.5 mM KCl, 1 mM MgCl2, 4 mM NaHCO3, 4 mM CaCl2, 20 mM glucose, pH 7.4). Samples were homogenised (Kimble Pellet Pestle Motor, Kimble Kontes, USA) in methanol for 1 min, kept 24 h at room temperature for pheromone extraction, and then centrifuged (12,500 rpm, 3 °C for 20 min; Hermle Z 360 K refrigerated centrifuge; B. Hermle AG, Wehingen, DE) to obtain the supernatant for HPLC-MS analyses (see above) for the presence of 12 and 16. Three sequential sets of dissections aimed to determine (1) the pheromone-containing body tagma, (2) the pheromone-containing tissues or glands in that tagma and (3) the specific gland or tissue producing 12 & 16.
To identify the pheromone-containing tagma, 11 spiders were severed at the pedicel, generating two tagmata: the cephalothorax with four pairs of legs and the abdomen. Each tagma was then extracted separately in 100 µL of methanol. Eight of 11 abdomen samples contained 12 and 16, whereas only one of 11 thorax samples contained 12 and 16 (Exp. 19), albeit at only trace amounts. With 12 and 16 being present in the abdomen, 20 additional abdomens were dissected68 to obtain separate samples of (i) haemolymph (25 µL), (ii) ventral cuticle (~0.5 cm2 near the pedicel, (iii) the ovaries, (iv) all silk glands combined, and (v) the gut (with anus, cloaca and Malpighian tubules). The remaining spider tissues (vi) were pooled as one sample, and 20 µL of the dissection buffer solution (vii) was obtained to detect potential pheromone bleeding. To each tissue sample, 50 µL of methanol were added. Only silk gland samples contained 12 and 16 (Exp. 20). Having established that only silk gland samples contained 12 and 16, the silk glands of 30 additional spiders were excised in the following order: (i) major ampullate gland, (ii) minor ampullate gland, (iii) anterior aggregate gland, (iv) posterior aggregate gland, (v) tubuliform, (vi) aciniform and flagelliform glands combined and (vii) pyriform gland. The glands from three spiders were combined in each sample and extracted in 30 µL methanol. Seven of ten posterior aggregate gland samples contained 12 and 16, with other silk gland samples not containing 12 and 16 or in only trace amounts (Exp. 21).
Transition of contact pheromone components to mate attractant pheromone components: evidence for hydrolysis of contact pheromone components (12, 16 and 17) (spring 2021)
To test for the hydrolysis of the contact pheromone components 12, 16 and 17, we compared their breakdown ratio (18/(12 + 16 + 17 + 18) on independent webs aged 0 days and 14 days (Exp. 24). Each of 140 spiders was allowed to spin a web on bamboo scaffolds for three days. Then, the spiders were removed and webs—by random assignment—were extracted immediately (0-day-old webs) or after 14 days of aging (14-day-old webs). On each web, the amount of contact pheromone components 12, 16 and 17, and of amide 18 as a breakdown product, was quantified using HPLC–MS, with 12 and 18 at 25 and 50 ng/µL as external standards.
Transition of contact pheromone components to mate attractant pheromone components: Y-tube olfactometer bioassays (general procedures)
The attraction of male spiders to web extracts and to candidate mate attractant pheromone components was tested in Y-tube olfactometers56 (Fig. 4a) lined with bamboo sticks to provide grip for the bioassay spider. Test stimuli were presented in translucent oven bags (30 cm × 31 cm; Toppits, Mengen, DE) secured to the orifice of side-arms. Test stimuli consisted of a triangular bamboo prism scaffold (each side 8.5 cm long) bearing a spider’s web, or bearing artificial webbing30 (40 ± 2 mg; Bling Star, CN) that was treated with web extract or synthetic chemicals in methanol (100 µL) as the treatment stimulus or with methanol (100 µL) as the control stimulus. For each experimental replicate, a male spider was introduced into a glass holding tube and allowed 2 min to acclimatise. Then, the holding tube was attached via a glass joint to the Y-tube olfactometer and an air pump was connected to the holding tube, drawing air at 100 mL/min through the olfactometer. Air entered the olfactometer through a glass tube secured to the oven bags’ second opening. A male that entered the olfactometer within the 5-min bioassay period was classed a responder and his first choice of oven bag (the oven bag he reached first) was recorded. Whenever a set of 30 replicates was completed by the same observer, using 30 separate Y-tubes, the Y-tubes were cleaned with hot water and soap (Sparkleen, Thermo Fisher Scientific, MA, United States) and dried in an oven at 100 °C for 3 h, whereas the bamboo sticks and the oven bags were discarded.
Transition of contact pheromone components to mate attractant pheromone components: Y-tube olfactometer bioassays (specific experiments) (summer 2018)
In experiments 22, 23 and 25–27, males were offered a choice between a solvent control stimulus and a treatment stimulus. The treatment stimulus consisted of (i) virgin female web-extract (1 web-equivalent) (Exp. 22, N = 24), (ii) the volatile compounds 5–11 unique to sexually mature females (Fig. 1d) (Exp. 23, N = 24), (iii) all breakdown products of the contact pheromone components 12, 16 and 17, consisting of the amide N-4-methylvaleroyl-l-serine (18) and the corresponding carboxylic acids 19, 20 and 21 (Exp. 25, N = 30), (iv) a blend of the acids 19, 20 and 21 (Exp. 26, N = 30) and (v) the amide 18 as a single compound (Exp. 27, N = 30). Compounds were tested at quantities as determined in virgin female web extract (50 webs in 150 μL of dichloromethane), following silyl-ester derivatization69 of acids in the extract, with valeric acid (200 ng; ≥99%, Sigma Aldrich, St. Louis, USA) added as an internal standard. Per web equivalent, there were 103 ng of 19, 3 ng of 20 and 54 ng of 21. The amide 18 was present at 200 ng per web equivalent, as determined using N-3-methylbutnaoyl-l-serine methyl ester as an external standard.
Transition of contact pheromone components to mate attractant pheromone components: hallway of buildings experiment (fall 2018)
As the ternary blend of the carboxylic acids 19, 20 and 21 attracted male spiders in Y-tube olfactometers (see Results), we aimed to confirm their functional role as mate attractant pheromone components also in ‘field’ settings (Exp. 28). To this end, we set up ten replicates of paired traps in building hallways on the Burnaby campus of Simon Fraser University. Adhesive-coated traps (Bell Laboratories Inc., Madison, WI, USA) were spaced 0.5 m within pairs and 20 m between pairs. By random assignment, one trap in each pair was baited with the carboxylic acids 19, 20 and 21 formulated in 200 µL of mineral oil (Anachemia, Montreal, CA; 2.8 mg of 19, 0.112 mg of 20 and 1.52 mg of 21), whereas the control trap received mineral oil only. Test stimuli were disseminated from a 400-μL microcentrifuge tube (Evergreen Scientific, Ranco Dominguez, CA, USA) with a hole in its lid punctured by a No. 3 insect pin (Hamilton Bell, Montvale, NJ, USA). Every week for 4 months (September to December 2018), traps were checked, lures were replaced, and the position of the treatment and the control trap within each trap pair was re-randomised.
Communication function of amide breakdown product 18 (fall 2018)
As the amide 18 did not attract males in Y-tube olfactometer experiments (see Results), we tested its alternate potential function as a contact pheromone component which, if active, would induce courtship by males. Using the T-rod apparatus (Fig. 1c), we treated one piece of filter paper with a solvent control and the other with a blend comprising both the contact pheromone components 12, 16 and 17 and the amide 18 (Exp. 29), a blend of 12, 16 and 17 (Exp. 30), and 18 alone (Exp. 31).
Mechanisms underlying the transition of contact pheromone components to mate attractant pheromone components: relationship between web pH and breakdown rates of contact pheromone components (summer 2020)
We allowed each of the 70 spiders to spin two webs, using one web to quantify the amide breakdown product (18) of the contact pheromone components (see above), and the other web to determine its pH according to the slurry method57 (Exp. 32). To this end, we first measured the pH of 50 µL water (HPLC grade, EMD Millipore Corp., Burlington, MA, USA) and then of a web with the water functioning as a conductor for the pH metre (LAQUAtwin pH 22 (Horiba, Kyoto, JP). Between web measurements, the pH metre was rinsed with water and regularly re-calibrated using a pH 7 and a pH 4 buffer (Horiba, Kyoto, JP).
Mechanisms underlying the transition of contact pheromone components to mate attractant pheromone components: testing for pH-dependent saponification of contact pheromone components (12, 16 and 17) (summer 2021)
To test whether pH alone catalyses saponification of the ester bond of contact pheromone components (12, 16 and 17), synthetic 12 was added to a 40% aqueous pH 7 buffer solution (Exp. 34), a pH 4 buffer solution (Exp. 34), and to acetonitrile (Exp. 35) as a polar aprotic solvent control (N = 12; 100 ng/µL each). pH-Dependent breakdown of 12 over time was assessed by analysing (HPLC-MS) diluted aqueous aliquots (2.5 ng/µl) of each sample at day 0 and after 14 days of storage at room temperature.
Mechanisms underlying the transition of contact pheromone components to mate attractant pheromone components: testing for the presence of a carboxylesterhydrolase (CEH) (summer 2021)
To test for the presence of a carboxylesterhydrolase (CEH), for each of three replicates we extracted (i) five webs of adult virgin female L. hesperus (positive control, known to have a CEH45), (ii) 20 webs of subadult S. grossa (deemed to have not yet produced a CEH) and (iii) ten webs of adult virgin female S. grossa, accounting for the different amounts of silk produced by these three groups of spiders. For each replicate, webs were extracted in 200 µL 0.05 M Sørensen buffer58 and analysed by Bioinformatics Solutions (Waterloo, ON, CA). After web samples were incubated for 20 min at 60 °C in 2× sample volumes of 10% SDS (lauryl sulfate; protein-denaturing anionic detergent), they were sonicated for 20 min. Then, the supernatant was withdrawn, reduced with dithiothreitol (DTT), and alkylated with iodoacetamide (IAA). Alkylated samples were treated further with a protein solvent (S-Trap kit; Protifi, Farmingdale, NY, USA). Briefly, samples were acidified by phosphoric acid to pH ≤1. Then 6× of sample volume S-trap buffer was mixed in. The mixture was loaded by centrifugation onto an S-Trap Micro Spin Column and washed 3× with S-trap buffer. Using the serine protease trypsin, protein digestions were carried out at 47 °C for 1 h in 50 mM triethylamonium bicarbonate (TEAB) buffer within the S-Trap Micro Spin column. Digestion products were eluted sequentially with 40 µL 50 mM TEAB and 0.2% formic acid. Eluates were dried and re-suspended in 0.1% formic acid.
Eluates were analysed by HPLC-MS/MS in positive ion mode on a Thermo Scientific Orbitrap Fusion Lumos Tribrid mass spectrometer (Thermo Fisher, San Jose, CA, USA), equipped with a nanospray ionisation source and a Thermo Fisher Ultimate 3000 RSLCnano HPLC System (Thermo Fisher). Peptide mixtures were loaded onto a PEPMAP100 C18 trap column (75 µm × 20 mm, 5 µm particle size; Thermo Fisher) at a constant flow of 30 μL/min and 60 °C isothermal. Peptides were eluted at a rate of 0.2 μL/min and separated using a Reprosil C18 analytical column (75 μm × 15 mm, 1.9 μm particle size; PepSep, DK) with a 60-min solvent gradient: 0–45 min: 4–35% acetonitrile + 0.1% formic acid; 45–55 min: 90% acetonitrile + 0.1% formic acid; 55–60 min: 4% acetonitrile + 0.1% formic acid.
MS data were acquired in data-dependent mode with a cycle time of 3 s. MS1 scan data were acquired with the Orbitrap mass analyser, using a mass range of 400–1600 m/z, with the resolution set to 120,000. The automatic gain control (AGC) was set to 4e5, with a maximum ion injection time of 50 ms, and the radio frequency (RF) lens was set to 30%. Isolations for MS2 scans were run using a quadrupole mass analyser, with an isolation window of 0.7. MS2 scan data were acquired with the Orbitrap mass analyser at a resolution of 15,000 m/z, with a maximum ion injection time of 22 ms, and the AGC target set to 5e4. Higher energy collisional dissociation (HCD; fixed normalised collision energy: 30%) was used for generating MS2 spectra, with the number of microscans set to 1.
Statistics and reproducibility
Data (Supplementary Table 2) were analysed statistically using R70. Data of experiments 1–18 and 29–31 (testing courtship by male spiders in response to contact pheromone components) were analysed with a Wilcoxon test or Kruskal–Wallis two-tailed rank-sum test with Benjamini–Hochberg correction to adjust for multiple comparison. Data of experiments 19–21 (revealing the presence of contact pheromone components in the abdomen, silk glands, and posterior aggregate silk gland) were analysed with two-tailed, rather than one-tailed, Wilcoxon test or Kruskal–Wallis rank tests because we had no strong assumption as to whether or not pheromone would be present in any of these potential pheromone sources. The p values were adjusted for multiple comparison using the Benjamini–Hochberg method. Y-tube olfactometer data of experiments 22, 23 and 25–27, as well as the hallway experiment 28 (revealing attraction of male spiders to volatile pheromone components) were analysed using an one-tailed71 binomial test, anticipating attraction of spiders to volatile mate attractant pheromone components rather than to solvent control stimuli. Data of experiment 32 (revealing a correlation between web pH and breakdown of web-borne contact pheromone components) were analysed using generalised linear models. Data of experiments 33–35 (showing pH-dependent breakdown of synthetic contact pheromone) were compared using a two-tailed Kruskal–Wallis test with Benjamini–Hochberg correction.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Source: Ecology - nature.com
