Plant signals during beetle (Scolytus multistriatus) feeding in American elm (Ulmus americana Planch)

ABSTRACT

American Elms were devastated by an outbreak of Dutch Elm Disease is caused by the fungus Ophiostoma novo-ulmi Brasier that originated in Asia and arrived in the early 1900s. In spite of decades of study, the specific mechanisms and disease resistance in some trees is not well understood. the fungus is spread by several species of bark beetles in the genus Scolytus, during their dispersal and feeding. Our objective was to understand elm responses to beetle feeding in the absence of the fungus to identify potential resistance mechanisms. A colony of Scolytus multistriatus was established from wild-caught beetles and beetles were co-incubated with susceptible or resistant American elm varieties in a controlled environment chamber. Beetles burrowed into the auxillary meristems of the young elm shoots. The trees responded to the beetle damage by a series of spikes in the concentration of plant growth regulating compounds, melatonin, serotonin, and jasmonic acid. Spikes in melatonin and serotonin represented a 7,000-fold increase over resting levels. Spikes in jasmonic acid were about 10-fold higher than resting levels with one very large spike observed. Differences were noted between susceptible and resistant elms that provide new understanding of plant defenses.

Introduction

American elms (Ulmus americana Planch) were once the iconic trees that shaded North American towns and cities. The spread of Dutch elm disease (DED) destroyed nearly 60% of the 77 million American elm trees in the USA between its introduction in the 1930s and the 1980s.¹⁶ The disease is caused by 3 fungal species but the most aggressive is Ophiostoma novo-ulmi Brasier, which is thought to have originated in Asia and arrived in North America in shipping crates made from elm wood,^3,36 DED causing fungi are spread by several species of bark beetles in the genus Scolytus. (Scolytus multistriatus) that carry the fungal spores (conidia) on the surface of their bodies and in the gut.^21,50 The fungal spores are injected into the elm xylem tissues during beetle tunneling and feeding. The elms produce tyloses and gels as defense mechanisms but these, along with the insects and growing fungi obstruct the tree vasculature and reduce water uptake. The host tree wilts and eventually dies as a result of both the infestation and the infection.³⁹ In spite of decades of study, the mechanisms of susceptibility or resistance to the disease are not completely understood.

Plants have many pathogen and herbivore-induced defense signaling pathways that are primarily controlled by SA and JA and these 2 small molecules represent only the backbone of the plant immunity.^32,35 The cellular, biochemical and molecular changes downstream of these 2 defense lines, are orchestrated by a complex network of secondary messengers (particularly, calcium ion (Ca²⁺) and reactive oxygen species (ROS)), plant hormones (particularly, ethylene (ET) and (ABA)) and transcription factors (e.g. ethylene response factors, b-Zip proteins).^32,44,45 In a previous study, we hypothesized that the elm tree responds to the beetles and fungus with a series of signaling molecules including jasmonic acid (JA), (ABA), serotonin (5HT; 5-hydroxytryptamine) and melatonin (MEL; N-acetyl-5-methoxytryptamine).³⁹ Artificial inoculation of American elm trees demonstrated that salicylic acid (SA) and JA are involved in the differential expression of disease-responsive genes between the tolerant ‘Valley Forge’ and the susceptible elm genotypes.³⁹ Spikes in SA and JA in response to the fungus indicated that the higher expression of disease-responsive genes noticed in ‘Valley Forge’ might be regulated mostly by JA, which was detected at high levels during this time.³⁹

Interestingly, among the phytochemical analysis of the samples from the beetle mimic study was one sample with very high concentrations of MEL and 5HT. Repeated analysis of the tissue demonstrated that this anomalous result was not an artifact or technical error but rather one piece of elm wood contained MEL (364 ± 21 ng/g) and 5HT (60 ± 9 ng/g) at much higher levels than all other samples (Fig. 1). Since there was no clear explanation for the result and the information point was found to be a statistical outlier that was not included in the publication.³⁹ For the current study, we hypothesized that the elm trees responded to the wounding mimic of beetle feeding with transient signals of high concentrations of 5HT and MEL.

Figure 1.

Chromatogram of melatonin detection in elm meristems. (A) Chromatogram of melatonin peak in a resting sample. (B) Chromatogram of a melatonin peak in a spiking signal.

MEL and 5HT are indoleamine neurotransmitters highly conserved through evolution and found in microbes, insects, animals, plants and humans (reviewed in Erland et al.,^12,13). 5HT was first discovered in the plant Macuna puriens in 1954.² MEL was detected in a 1995 survey of fruit and vegetable samples bought at market^11,18 and discovered in growing medicinal plants in 1997.⁵⁶ Several recent articles have highlighted the roles of MEL and 5HT in plant physiology,^12,13 including tolerance to abiotic stress,^1,23 and resistance to pathogens.^22,54 Several roles have been hypothesized for MEL and 5HT in plant physiology but the specific biochemical mechanisms are only partially understood. Recently, hormonal crosstalk between JA, SA, MEL and 5HT has been described.^{12,13,24,51,55} The transcriptional induction of genes involved in stress-related hormone signaling pathways (i.e., ABA, ET, SA and JA pathways) by MEL has also been demonstrated in Arabidopsis, tobacco and Bermuda grass.^22,51 Although the regulation of plant defenses against insect attacks by MEL and 5HT has not yet been reported, the anti-herbivore activity of similar compounds has been demonstrated in the temperate green alga Ulvari aobscura⁴⁶ and 5HT is an important modulator of insect appetite and feeding behavior.^8,9,14,30,31 From this perspective, the spikes in 5HT, MEL and/or other indolamines in response to fungal infection or herbivore feeding could be explained from the perspective of the roles of these molecules in the cell; as antioxidants, primary or secondary messengers and/or plant growth regulators.

The current work was designed to investigate the anomaly of the previous study by investigating the response of the trees to the normal activity of the beetles. We hypothesized that transient spikes SA, JA, 5HT and MEL are induced by beetle herbivory in American elm to provide defense against disease. To investigate this hypothesis, we established a colony of elm beetles (Scolytus multistriatus) initiated by wild-caught beetles and co-incubated the beetles with a susceptible clone of American elm and the resistant variety ‘Valley Forge’ and analyzed the responses. Our data demonstrate that the beetles burrowed into the auxillary meristems of young trees and that the trees responded with large spikes in the concentration of signaling and defense molecules.

Results

S. multistriatus beetles were collected over the summer of 2015 and established in a colony at UBC. Rates of beetle capture fluctuated over the season with the highest number of beetles collected during the warmest periods as indicated by the daily average temperature for that period (Fig. 2). The adult beetles burrowed into local elm logs and produced eggs (Fig. 3C and D). Young beetles hatched from the logs and were collected into glass mason jars for experiments (Fig. 3E). The overall objective of the study was to determine the responses of American elm shoots exposed to the S. multistriatus beetle. Both male and female beetles burrowed into the auxillary meristems of the elm shoots (Fig. 5A) and there was evidence of beetle chewing on auxillary meristems throughout (Fig. 5B) (see video). Both male and female beetles burrowed into auxillary meristems. The experiments compared resistant and susceptible elm varieties with or without beetles (Valley Forge n = 24, susceptible n = 15; Table 1).

Figure 2.

Collection and rearing of Scolytus multistriatus. (A) Field collection with a 3-vane panel trap. (B) Adult S. multistriatus trapped in Kelowna, BC. C. S. multistriatus egg-laying gallery. D. S. multistriatus eggs in the colony. E. Colony rearing chamber. Collection jars were fitted on to a bell ended PVC pipe that extrudes from the hole in the side of the chamber.

Figure 3.

Elm meristems were exposed to beetles for a 24 hour period. (A) Meristems were numbered before analysis to track changes in phytochemistry. (B) Shoots were incubated in cotton spice bags with and without beetles for 24 hours.

Figure 5.

Beetles chewed elm meristems during the 24 hour exposure. (A) Elm meristem with female beetle burrowing, male beetle observing. (B) Histology of the beetle burrowing in the meristematic zone.

Table 1.

Meristems analyzed for phytochemical contents were sorted by proximal distance on the shoot.

Treatment	Meristem	Number
Susceptible	apical	5
	2nd	3
	3rd	7
Susceptible with Beetle	apical	5
	2nd	5
	3rd	5
Valley Forge	apical	8
	2nd	8
	3rd	8
Valley Forge with Beetle	apical	8
	2nd	8
	3rd	8

Our data show clear patterns or significant differences in 5HT, MEL, ABA and JA but not auxin in response to clone type and insect interaction. Clustering analysis by Principle Component Analysis showed 2 significantly differences in clustering of 5HT, MEL and JA for the susceptible and Valley Forge clones with and without beetle exposures but there was no significant correlation between the resting and spiking values with the location of specific meristems (data no shown).

Serotonin

Serotonin was present at a resting level of less than 0.1 pg/g in 60% of the susceptible and 80% of the Valley Forge meristems (Fig. 4A). When the shoot tissues were exposed to the beetles, 100% of both Valley Forge and susceptible meristems had less than 0.1 pg/g 5HT (Fig. 4A). However, large spikes in 5HT were observed in 40% of the susceptible meristems and small spikes in 5HT were observed in 20% of the Valley Forge meristems in the absence of beetle exposure (723 pg/g and 10 pg/g of 5HT, respectively; Fig. 4B). Interestingly, large increases in the concentration of 5HT was not detected in either susceptible or resistant elms after beetle feeding and the control fluctuations in 5HT may have been suppressed (Fig. 4B).

Figure 4.

Rate of beetle (Scolytus multistriatus) collection in the summer of 2015. The blue bars represent the number of individual beetles caught on a given day. The orange points correspond to the daily mean temperature (http://kelowna.weatherstats.ca/).

Melatonin

Melatonin was present at a resting level of less than 0.1 pg/g in 80% of susceptible and 58% of resistant Valley Forge meristems (Fig. 5A). When shoots were exposed to beetles, 87% of susceptible meristems and 67% of Valley Forge meristems had the resting levels of less than 0.1 pg/g of MEL (Fig. 5A). However, in 20% of the susceptible and 42% of the Valley Forge meristems without beetles, significant spikes of MEL were observed (Fig. 5B). The susceptible clones without beetle exposure spiked to an average of 16.4 ± 4.1 µg/g. The Valley Forge meristems without beetle exposure spiked to an average of 1.2 ± 0.9 mg/g. When the susceptible shoot tissues were exposed to the beetles, the average spike was 774 ± 127 pg/g which is a 2.1 × 10⁴ fold decrease in the amount of MEL spike as a result of beetle contact. Likewise, an average spike of 462 ± 0.8 ng/g was observed in the Valley Forge meristems exposed to beetles which is a 2.7 × 10³ fold decrease in the size of the MEL spike as a result of beetle contact. Similar to 5HT, the beetle contact seems to have supressed the control fluctuations in MEL concentrations in elms.

Abscisic acid

Sixty percent of the susceptible and 16.6% of the Valley Forge elms had resting levels of ABA of 400 ± 20 pg/g and 1,252 ± 59 pg/g respectively without exposure to beetles. When exposed to beetles, the resting levels of ABA in the susceptible and Valley Forge meristems were not significantly different at 530 ± 26 pg/g and 1,212 ± 59 pg/g respectively (Fig. 6A). Significant spikes in ABA were observed in 40% of susceptible clones without beetles and 53% of susceptible clones exposed to beetles at 638 ± 410 µg/g and 623 ± 123 µg/g respectively (Fig. 6B). Measurements of the Valley Forge meristems found spikes in 83% of meristems without beetles and 75% of meristems exposed to beetles of 80 ± 53 µg/g and 364 ± 148 µg/g respectively (Fig. 6B).

Figure 6.

Responses of elm meristems to beetle herbivory through spikes in serotonin. (A) Resting levels of serotonin. (B) Spiking levels of serotonin.

Jasmonic acid

More than 93% of the susceptible clone meristems without beetles and 80% of the susceptible meristems exposed to beetles had an average of 200 ± 11 pg/g and 217 ± 15 pg/g respectively. The resistant variety Valley Forge had 38 ± 3 pg/g in 83% of meristems without beetle exposure and 1028 ± 40 pg/g in 67% of meristems exposed to beetles (Fig. 7A). Spiking levels were observed in 6.6% of susceptible clones without beetles and 20% of susceptible clones with beetles. The spikes in JA in the meristems averaged 15 ± 6 ng/g when beetles were present and in the absence of beetles, the JA spike was observed at 25 µg/g (Fig. 7B). In the absence of beetles, the resistant variety Valley Forge had JA spikes in 16.6% of the meristems with an average level of 197 ± 9.6 ng/g (Fig. 8B). By far the largest spikes in JA were quantified in 33% of Valley Forge meristems with average JA levels of 2,851 ± 230 ng/g and one significant outlier quantified at 20.6 mg/g (Fig. 7B).

Figure 7.

Responses of elm meristems to beetle herbivory through spikes in melatonin. (A) Resting levels of melatonin. (B) Spiking levels of melatonin.

Figure 8.

Responses of elm meristems to beetle herbivory through spikes in abscisic acid. (A) Resting levels of abscisic acid. (B) Spiking levels of abscisic acid.

Discussion

Beetle feeding

One of the most interesting and unexpected observations of this study was the manner of beetle feeding on elm meristems. Beetles were collected in Okanagan, a region that does not currently have an outbreak of the fungal infection. A previous study described the courtship of elm beetles involving chewing and burrowing in the meristems of elm twigs.⁴² Both males and females feed on the meristems and the beetles can be engaged for 6–10 hours of feeding.⁴² In the absence of the fungus, this natural behavior of beetles damages elm meristems and could alter the relative ratio of auxin to cytokinin resulting in bud break of auxillary meristems. Viewed in this way, the beetles may be helping the elm trees to develop a stable architecture resistant to wind damage but the activity exposes the elms fungal infections. The full spectrum of elm responses to fungal infection has been recently reviewed⁵ including development of physical barriers at the site of beetle chewing, chemical defenses and induction of defense genes.

Melatonin and serotonin in plant signaling

The activity of MEL and 5HT in plant signaling has been extensively investigated (reviewed in Erland et al.,^12,13). The current data indicated that there is a system of spiking and resting concentrations of MEL and 5HT in elm meristems that could be linked to developmental and /or metabolic changes in elm meristems. In previous studies, the transcriptome profiling of Arabidopsis after treatments with low (100 pM) and high (1 mM) MEL concentrations showed that not all genes altered by low MEL concentrations were affected by the high concentration, suggesting that MEL regulation of downstream functions were dose-dependent.⁵¹ Furthermore, the genes upregulated by the high MEL concentration were mostly associated with stress-related hormones (i.e., ABA, ET, SA and JA); whereas genes downregulated by the same concentration were involved in growth-related pathways; i.e., auxin responses and signaling; cell wall synthesis and modifications.⁵¹ Since both MEL and 5HT fluctuated in both susceptible and resistant varieties, it is difficult to determine the exact mechanisms of signaling in this system. However, the present findings provide evidence that the spiking levels of MEL observed after inoculation of American elm saplings with the DED-causing fungus³⁹ were not an artifact and future studies may provide interesting insights connecting plant defense pathways in tolerant and susceptible genotypes. Indeed, the levels of both JA and MEL were significantly higher in Valley Forge than the susceptible genotype in the challenged and unchallenged tissues. Since the link between JA and the tolerance of VF to DED has already been demonstrated,³⁹ the exceptionally high levels of MEL in this variety might also be associated with its tolerant phenotype.

Signaling of defense responses

It has become evident in many plant species that JA and its derivative compounds are important regulators of defense responses against insect herbivores, including bark beetles.^4,19 The JA-mediated defense responses are usually characterized by the induction of JA-biosynthesis enzymes, stimulation of many JA-signaling proteins and transcriptional activation of a variety of wounding/herbivore-responsive genes such as protease inhibitors, thionins and enzymes involved in secondary metabolism.^5,37,48 However, it should also be noted that the timing and magnitude of these responses can vary greatly based on the infestation stage; the genetic makeup of the plant (tolerant vs. sensitive); the nature of the attackers (generalists vs. specialists) and the cross-talks between JA and other plant defense pathways associated with JA (e.g., salicylic acid (SA), abscisic acid (ABA) and ethylene (ET)).^4,5,19,43 Although the present study was not meant to investigate all these factors, it provides an experimental model that can be applied under controlled environmental conditions (e.g., greenhouses, growth chambers) and that can be optimized to examine the responses of elm genotypes to bark beetles and its fungal associates (e.g., O. ulmi, O. novo-ulmi). The quantification of JA, ABA, 5HT and MEL under these conditions has led to remarkable findings about the levels of these molecules in challenged and unchallenged elm meristems; the implications of which will form the bases of the following discussions.

Some or even all the above-mentioned factors can explain the varying levels of JA in the resting and spiking phases in different meristems. For instance, wounded pea leaves show 2 peaks of endogenous JA levels within 48h of wounding, whereas the systemic leaves show only one peak of JA during the same period. The levels of JA in the resting and spiking phases after wounding range between 40 and 190 ng/g FW, respectively.⁵³ Similarly, leaves of tobacco (Nicotiana attenuata) wounded and treated with the oral secretions of its natural herbivore, Manduca sexta, show a peak of JA (> 2 µg/g FW) at 1h of treatment. This peak shrinks to only 0.1 µg /g of JA within 2 hours.²⁶ These fluctuating levels of JA in response to herbivore feeding or even other environmental or developmental cues have been linked to the molecular changes of JA-biosynthesis and signaling pathways, where the induction and repression of JA-biosynthesis genes (i.e., lipoxygenase, allene-oxide cyclase and allene-oxide synthase) is coordinated by positive and negative feedback loops, respectively.^7,32,49 These loops are mediated through the degradation and stabilization, respectively, of JA repressor proteins known as JA ZIM-domain (JAZ) proteins, which in normal (resting) conditions suppress the activity of MYC transcriptional activators.^27,29 Interestingly, ABA activation of wounding responses and wound-inducible JA production is mediated, at least in part, through the transcriptional regulation of MYC2,^34,47 which when activated induce the expression of downstream JA-biosynthesis genes.^20,49 Such tight control of cellular JA levels through the transcriptional activation or repression of the JA-biosynthesis genes is thought to minimize the cost of allocating resources toward defense/secondary metabolism when the plant is mediating trade-offs between growth and defense; or between reproduction and defense.^10,17,38,41 In fast growing organs such as meristems, the constitutive high levels of JA might lead to stunted growth through the suppression of mitosis.⁵⁷ Therefore, the resting and spiking levels of JA and other associated hormones (i.e., ABA) might represent various stages of plant responses to different internal or external cues. However, owing to the nature of the present experiment the precise developmental stages of meristems could not be determined because it would have destroyed to the tissues before beetle exposure, JA and ABA showed large variations even in un-challenged tissues of American elm indicating a potential natural rhythm or response to another stimulant.

Antioxidant activities

Similarly, the induction of reactive oxygen species (ROS) after wounding has been demonstrated in many systems and is thought to play an integral role in defense against insect herbivores.^28,52 For example, tomato mutants which were unable to produce hydrogen peroxide were more susceptible to larvae of M. sexta.³³ ROS are also involved in plant development, and there are many lines of evidence showing that spatial regulation of ROS may regulate plant morphology, through their role in regulating cell growth.^15,25 However, it is also well-known that ROS, especially hydrogen peroxide is very toxic to the cell and hence the plant has evolved complex mechanisms involving calcium, ROS-scavenging enzymes and protein phosphorylation to control the wound-induced oxidative burst.³⁷ In light of these data, the induction of potent antioxidants such as MEL and 5HT in challenged and un-challenged meristems may be a plant strategy to abolish the damaging effects of ROS. MEL has been shown to have significant activity as a radical scavenger,⁶ enhance the activity of antioxidative enzymes (i.e., superoxide dismutases, catalase),²³ or even control the expression of genes involved in redox (i.e., peroxidase and glutathione S-transferase).^40,51

Conclusions

Overall, these data provide an interesting new approach to understanding the interactions between insects. Fluctuations in the resting and spiking levels of plant growth regulators and defense compounds have been previously reported but are not well understood. The study of plant communication between cells and tissues is in its infancy. More research is needed to fully understand how MEL and 5HT may coordinate plant responses to insect attack through interactions with JA and SA. Further, these natural plant responses to insect feeding may be interesting new targets for pesticide development.

Materials and methods

Establishment of a scolytus multistraitus colony

Beetle collections

Adult Scolytus multistriatus were collected from 16 traps at 3 sites in Kelowna, British Columbia in the summer of 2015. Global Positioning System coordinates for all individual traps are provided in Table 2. Individuals were collected using white 3-vane panel traps (Fig. 8A). Elm bark beetle lures paired with cubeb oil (Synergy Semiochemical, Burnaby, BC, Canada) were used as attractants and replaced after a period of 8 weeks. Traps were hung so that the bottom of the collecting cup was at 1.5m above ground level. All insects captured in the traps were collected every morning. Insects were sorted and S. multistriatus were identified on the basis of standard morphological criteria including; a black pronotum, 1.9 – 3.1mm in length, clubbed antennae, a concave posterior underside, and the presence of a spine on the second sternite (Fig. 8B). After identification, live and dead counts were recorded and live adults were placed in rearing chambers.

Table 2.

Traps locations, numbers, and coordinates. Coordinates obtained from Garmin Oregon 550 handheld GPS.

Location	Trap Numbers	Coordinates (degrees, minutes, seconds)
UBCO	1, 2	W119 23′ 28.9″ N49 56′ 18″
UBCO	3, 4	W119 23′ 30.1″ N49 56′ 12.2″
UBCO	5, 6	W119 23′ 33.2″ N49 56′ 18″
UBCO	7, 8	W119 23′ 28.5″ N49 56′ 16.6″
UBCO	9, 10	W119 23′ 27.6″ N49 56′ 18.5″
Bankhead Elementary	E1, E2	W119 28′ 17″ N49 53′ 30″
Knox Mountain Park	K1, K4	W119 29′ 38″ N49 54′ 23.4″
Knox Mountain Park	K2	W119 29′ 37.8″ N49 54′ 21.8″
Knox Mountain Park	K3	W119 29′ 34.8″ N49 54′ 21.1″

Beetle rearing

Rearing chambers were constructed from standard 5 gallon plastic buckets purchased at a home improvement store. A 6 cm diameter hole was cut in the side of each bucket and plugged with a rubber stopper (Fig. 8E). Bolts of elm wood (30–40 cm) were harvested from local trees (Kelowna, BC, Canada), the ends of each bolt were sealed with melted paraffin wax and bolts were placed in the rearing chambers to create a habitat for the collected beetles. Between 200 – 400 collected beetles (S. multistriatus) were placed on the bolts of Elm in each of the 5 rearing chambers. The chambers were left in complete darkness for 50 d to promote burrowing and egg laying (Fig. 8C and D). After incubation, the plug was removed from the side of the rearing chamber and a clear collection jar was fitted into the opening (Fig. 8E). During daylight hours the collection jars were exposed to light to attract newly emerging beetles. Newly collected adults were identified as S. multistriatus then either added to a new generation of rearing chambers or selected for use in experiments.

Exposing Elm trees to beetles

Saplings of American elm ‘Valley Forge’ were purchased from a commercial nursery (Connon Nurseries, West Flamborough, ON). The susceptible elm plant material was selected from clones of a susceptible elm tree in the in vitro germplasm bank at the Gosling Research Institute for Plant Preservation (GRIPP), University of Guelph. Dormant trees were transferred to UBC and grown for 6 months in a controlled environment growth chamber before experiments (Conviron, Winnipeg, MN). Dormant saplings were initiated at 16 h, 21°C full spectrum cool white light (65 µmol/m²/s) and 8 h 18°C dark cycle for 6 weeks and buds were induced to break with a 16 h, 27°C light and 8 h 18°C dark cycle. Newly emerged shoots with 3–4 nodes (Fig. 9A) were incubated with beetles in a small cotton bag (Les Herbes, 2′x3′) for 24 hours (Fig. 9B). For each exposure, a pair matched control was prepared with second shoot on the same plant incubated in the cotton bag without beetle for the same time interval. At the end of the incubation period, shoots were cut from the plants leaving the cotton bag intact and dropped immediately into liquid N. Samples were stored at −80°C until analysis.

Figure 9.

Responses of elm meristems to beetle herbivory through spikes in jasmonic acid. (A) Resting levels of jasmonic acid. (B) Spiking levels of jasmonic acid.

UPLC-MS/MS methods for quantification of plant signaling molecules

Meristems were dissected from the frozen samples and numbered according to position on the shoot (Fig. 9A). All sample preparation was performed in a dark room with a red light to avoid light degradation of the indoleamines during extraction using described previously methods.⁵⁸ In brief, meristems were placed in pre-weighed 1.5 mL Eppendorf tubes, weighed for accurate weight and prepared individually. Samples were homogenized in a solution of 80% methanol (Fisher Optima Grade, Fisher Scientific, Mississauga, ON) and 20% 0.1 tricholoracetic acid (TCA; Sigma, Mississauga, ON) in E-Pure water™ (Fisher). Homogenizing solution was added to each sample in a 1:4 (w/v) ratio and samples were homogenized (Kontes Pellet Pestle disposable tissue grinder; Fisher Scientific). Samples were centrifuged (13,000 × g) for 3 minutes and the supernatant was filtered (0.2 mm, Ultrafree-MC filtered centrifuge tubes; Millipore) before chromatography. Serotonin (5HT; RT 0.77), melatonin (Mel; RT 2.49), indole-3-acetic acid (IAA; RT 2.60), Abscisic acid (ABA; RT 2.77) and Jasmonic Acid (JA; RT 3.16) were separated on a reverse phase column (30×3 mm, 2.6 μm C18 100 Å, Phenomenex, Torrance, CA) using a Waters Acquity I-class UPLC (Waters, Mississauga, ON) over a gradient of 0.1% formic acid (Eluent A) and acetonitrile (Eluent B) (A%:B%): 0.0–0.5 min, 90:10; 0.5–3.5 min, 40:60; 3.5–4.2 min, 5:95; 4.2–6.5min, 5:95; 6.5–7.0 min, 90:10] with a flow rate of 0.3 mL/min. Analytes were quantified with a tandem mass spectrometer (Xevo TQ-S triple quadrupole Mass Spectrometer, Waters). The capillary voltage was 3500, desolvation gas rate was 800 L/hr, cone gas rate was 150 L/hr, desolvation temperature was 550 °C, and the source temperature was 150 °C for all analyses with a dwell time of 0.02s. Parent and daughter ions were detected using the appropriate optimized Multiple Reaction Monitoring (MRM) transitions (Table 3). Standards were prepared in a dilution series at the following concentrations for each compound: 0, 0.001, 0.025, 0.5, 10, 200, 400, 4000 ng/mL. Since the dynamic range of the data spread over 4 orders of magnitude, data for the external standard curve underwent a LOG transformation for linearity. The Limit of Detection (LOD) was < 1.0 pg/mL for all compounds and extracts were quantified within the linear range of 0.025–400 ng/mL.

Table 3.

MRM transitions and optimized mass spectrometer parameters for detection and quantification of plant signaling compounds in American elm.

Compound	Parent Ion (m/z)	Daughter Ion (m/z)	Dwell Time (s)	Cone (V)	Collision Cell (eV)	Delay Time (s)
Indole-3-Acetic Acid	176.00	103.00	0.020	30.00	25.00	−1.000
Indole-3-Acetic Acid	176.00	130.10	0.020	30.00	13.00	−1.000
Serotonin	177.00	115.00	0.020	45.00	27.00	−1.000
Serotonin	177.00	160.00	0.020	45.00	10.00	−1.000
Jasmonic Acid	211.00	133.00	0.020	22.00	14.00	−1.000
Jasmonic Acid	211.00	151.00	0.020	22.00	10.00	−1.000
Melatonin	233.00	159.00	0.020	30.00	23.00	−1.000
Melatonin	233.00	174.00	0.020	30.00	15.00	−1.000
Abscisic Acid	265.00	135.00	0.020	20.00	22.00	−1.000
Abscisic Acid	265.00	247.00	0.020	20.00	6.00	−1.000

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

The authors gratefully acknowledge the financial support of the Gosling Foundation, Ontario, Canada through The Gosling Research Institute for Plant Preservation at the University of Guelph (www.gripp.ca).