Non-targeted metabolomics aids in sex pheromone identification: a proof-of-concept study with the triangulate cobweb spider, Steatoda triangulosa

Abstract

Targeted metabolomics has been widely used in pheromone research but may miss pheromone components in study organisms that produce pheromones in trace amount and/or lack bio-detectors (e.g., antennae) to readily locate them in complex samples. Here, we used non-targeted metabolomics-together with high-performance liquid chromatography-mass spectrometry (HPLC-MS), gas chromatography-MS, and behavioral bioassays-to unravel the sex pheromone of the triangulate cobweb spider, Steatoda triangulosa. A ternary blend of three contact pheromone components [N-4-methylvaleroyl-O-isobutyroyl-L-serine (5), N-3-methylbutyryl-O-isobutyroyl-L-serine (11), and N-3-methylbutyryl-O-butyroyl-L-serine (12)] elicited courtship by S. triangulosa males as effectively as female web extract. Hydrolysis of 5, 11 and 12 at the ester bond gave rise to two mate-attractant pheromone components [butyric acid (7) and isobutyric acid (8)] which attracted S. triangulosa males as effectively as female webs. Pheromone components 11 and 12 are reported in spiders for the first time, and were discovered only through the use of non-targeted metabolomics and GC-MS. All compounds resemble pheromone components previously identified in widow spiders. Our study provides impetus to apply non-targeted metabolomics for pheromone research in a wide range of animal taxa.

Figure 1

Graphical comparison of total ion chromatograms of a hypothetical case sample (upper trace) and a control sample (lower trace). (a) A unique compound (green) in the case sample is absent in the control sample. (b) A novel compound (blue) in the case sample is masked—and thus easily overlooked—by a compound (brown) present in both samples. (c) A unique trace compound (red) in the case sample might not be visually detected.

Figure 2

Phylogeny and comparison of pheromone components (contact & airborne) in widow spiders (Latrodectinae). (a) Previously known pheromone components of Latrodectus hasselti, L. hesperus, L. geometricus, and Steatoda grossa: N-3-methylbutyroyl-O-(S)-2-methylbutyroyl-l-serine methyl ester (1), N-3-methylbutanoyl-O-isobutyroyl-l-serine methyl ester (2), N-3-methylbutyroyl-O-isobutyroyl-l-serine-methyl ester (3), N-4-methylvaleroyl-O-butyroyl-l-serine (4), N-4-methylvaleroyl-O-isobutyroyl-l-serine (5), and N-4-methylvaleroyl-O-hexanoyl-l-serine (6). The contact pheromone components 4–6 of S. grossa hydrolyse at the ester bond and give to three airborne mate-attractant pheromone components [butyric acid (7), isobutyric acid (8), and hexanoic acid (9)], whereas the amide N-4-methylvaleroyl-l-serine (10), as another hydrolysis breakdown product, remains on webs and has no behavioural activity. (b) Pheromone components of Steatoda triangulosa identified in this study. The contact pheromone components N-4-methylvaleroyl-O-isobutyroyl-l-serine (5), N-3-methylbutyroyl-O-isobutyroyl-l-serine (11), and N-3-methylbutyroyl-O-butyroyl-l-serine (12) hydrolyse at the ester bond and give rise to two airborne mate-attractant pheromone components [butyric acid (7) and isobutyric acid (8)], whereas N-4-methylvaleroyl-l-serine (10) and N-3-methyl-butyroyl-l-serine (13) accumulate on webs. Blue-coloured parts of molecules are phylogenetically conserved, whereas green-coloured parts are unique to Steatoda. Orange parts are shared between Latrodectus spp. and S. triangulosa.

Figure 3

Chromatograms, experimental designs, and behavioural bioassay results. (a) Total ion chromatogram (TIC) of web extract of female Steatoda triangulosa analysed by high-performance liquid chromatography—mass spectrometry. (b) TIC of silyl ester-derivatized web extract of female S. triangulosa analysed by gas chromatography—mass spectrometry. (c) Comparative XCMS online Cloud Plots of web extracts of mature and immature female S. triangulosa (depicted by solid and dotted lines, respectively), with circles denoting a > 35-fold abundance increase of fragment ions in compounds; the larger the circle, the greater the fold-change of a particular ion. (d) T-rod bioassay apparatus. (e) Effects of female S. triangulosa web extract (Exp. 1) and contact pheromone component 5 (N-4-methylvaleroyl-O-isobutyroyl-l-serine) (Exp. 2) on courtship by S. triangulosa males. (f) Effects of female S. triangulosa web extract (Exp. 3), and a ternary blend of contact pheromone components 5, 11 (N-3-methylbutyoyl-O-isobutyroyl-l-serine, and 12 (N-3-methylbutyroyl-O-butyroyl-l-serine) (Exp. 4), on courtship by S. triangulosa males. (g) Effects of contact pheromone components 5, 11 and 12 presented in ternary combination (Exp. 5), and singly (Exps. 6–8), on courtship by S. triangulosa males. (h) Arena olfactometer with prisms carrying test stimuli. (i) Attraction of male S. triangulosa to webs of female S. triangulosa (Exp. 9), and to synthetic mate-attractant pheromone components 7 (butyric acid) and 8 (isobutyric acid) in arena olfactometers. In each of subpanels e–g, different letters indicate statistical differences between test stimuli across experiments (rank sum test; p < 0.05). In experiments 9 and 10 (subpanel i), the asterisk (*) indicates a significant preference for the test stimulus (binomial test; p < 0.05).