Article Sense, and Follow, Trail Pheromones of Community Members

Jaime M. Chalissery *, Asim Renyard, Regine Gries, Danielle Hoefele, Santosh Kumar Alamsetti and Gerhard Gries

Department of Biological Sciences, Simon Fraser University, Burnaby, BC V5A 1S6, Canada; [email protected] (A.R.); [email protected] (R.G.); [email protected] (D.H.); [email protected] (S.K.A.); [email protected] (G.G.) * Correspondence: [email protected]

 Received: 28 September 2019; Accepted: 29 October 2019; Published: 1 November 2019 

Abstract: Ants deposit trail pheromones that guide nestmates to food sources. We tested the hypotheses that ant community members (Western carpenter ants, Camponotus modoc; black garden ants, niger; European fire ants, Myrmica rubra) (1) sense, and follow, each other’s trail pheromones, and (2) fail to recognize trail pheromones of allopatric ants (pavement ants, Tetramorium caespitum; desert harvester ants, Novomessor albisetosus; Argentine ants, Linepithema humilis). In gas chromatographic-electroantennographic detection analyses of a six-species synthetic trail pheromone blend (6-TPB), La. niger, Ca. modoc, and M. rubra sensed the trail pheromones of all community members and unexpectedly that of T. caespitum. Except for La. niger, all species did not recognize the trail pheromones of N. albisetosus and Li. humilis. In bioassays, La. niger workers followed the 6-TPB trail for longer distances than their own trail pheromone, indicating an additive effect of con- and hetero-specific pheromones on trail-following. Moreover, Ca. modoc workers followed the 6-TPB and their own trail pheromones for similar distances, indicating no adverse effects of heterospecific pheromones on trail-following. Our data show that ant community members eavesdrop on each other’s trail pheromones, and that multiple pheromones can be combined in a lure that guides multiple species of pest ants to lethal food baits.

Keywords: Lasius niger; black garden ant; Camponotus modoc; Western carpenter ant; Myrmica rubra; European fire ant; trail pheromone; eavesdropping; pheromonal communication; gas chromatographic- electroantennographic detection

1. Introduction Ant colonies use multimodal communication signals to coordinate specific tasks such as , nest defense, and cooperative brood care [1]. Trail pheromone signals are particularly important in the context of foraging [2]. When a forager has located a profitable food source and then returns to her nest, she deposits trail pheromones that guide nest mates to the same resource [2]. Additional foragers recruited to this resource may also deposit trail pheromones and thus reinforce the original trail [2,3], effectively resulting in collective decisions by nestmates as to which resource to exploit [3,4]. Pheromone trails leading to persistent food sources are generally well maintained by foragers [2] and thus are readily exploited by (heterospecific) non-nestmates [1,5] that learn about the location of profitable food sources through eavesdropping [6–9]. We use the term “eavesdropping” here to describe the behavior of ants gleaning trail pheromone information from community members but not to imply inevitably adverse effects for any community member involved. Indeed, aggressive encounters of ants with non-nest mates on shared (eavesdropped) trails [1,8,9] are kept to a minimum, in part, by using dissimilar foraging schedules. Temporal partitioning of activity schedules has been

Insects 2019, 10, 383; doi:10.3390/insects10110383 www.mdpi.com/journal/insects Insects 2019, 10, 383 2 of 11 reported for workers of Ca. pennsylvanicus and Formica subsericea that forage on the same -infested trees but at different times of the day [10], and for workers of Ca. beebei that follow trails of Az. charifex when Azteca ants are resting [1,6,8,9]. We anticipate that mutual recognition of pheromone trails is more likely for co-evolved ant species than for native and invasive species. However, two exceptions are conceivable. First, the invasion event took place a long time ago and, over time, the invading species has become a well-established and integrated community member. Second, the invading species is closely related to native species and thus produces a similar trail pheromone. Ant communities in the Lower Mainland of British Columbia (BC), Canada are complex and comprise both native and invasive species. For the purpose of this study, we have selected three species that co-exist in the same community: (1) the Western carpenter ant, Camponotus modoc (), which is considered native to the Pacific Northwest and has been recorded in BC as early as 1919 [11]; (2) the black garden ant, Lasius niger (Formicinae), which is native to Europe, and possibly to North America, having been recorded in the New World as early as 1979 [11–13]; and (3) the European fire ant, Myrmica rubra (Myrmicinae), which is native to Europe but has invaded the Pacific Northwest and other parts of North America, likely in the first decade of the 20th century [14,15]. While Ca. modoc and La. niger have co-existed for at least 39 years [13], M. rubra as a more recent adventive is known to have occurred in ant communities of BC’s Lower Mainland for nearly 20 years [15] and has already become a well-established and integrated community member. During 20 years of co-existence, all three species might have “learned” to sense each other’s trail pheromones. We prepared the trail pheromone components currently known for these three species (Ca. modoc: (2S,4R,5S)-2,4-dimethyl-5-hexanolide (henceforth “hexanolide”) [16]; La. niger: 3,4-dihydro-8-hydroxy-3,5,7-trimethylisocoumarin (henceforth “isocoumarin”) [17]; M. rubra: 3-ethyl-2,5-dimethylpyrazine [18]) in a synthetic blend (Table1). This blend also contained 3-ethyl-2,6-dimethyl pyrazine (a non-natural isomer in the commercial source of the M. rubra trail pheromone). To determine whether Ca. modoc, La. niger, and M. rubra sense the trail pheromones not only of community members but also of allopatric ant species, we expanded the synthetic blend to include the trail pheromone of the pavement ant, Tetramorium caespitum (2,5-dimethylpyrazine), the desert harvester ant, Novomessor albisetosus (4-methyl-3-heptanone), and the Argentine ant, Linepithema humilis ((Z)-9-hexadecenal) [19–21]. InsectsInsects 2019 2019, 10, 10, x, x 3 3of of 12 12

Table 1. List of trail pheromones (and select species producing them) comprising the six-trail TableTable 1.1. ListList ofof trailtrail pheromonespheromones (and(and selectselect specspeciesies producingproducing them)them) comprisingcomprising thethe six-trailsix-trail pheromone blend (6-TPB) tested in circular trail bioassays (Figure1) and in electrophysiological pheromonepheromone blend blend (6-TPB) (6-TPB) tested tested in in circular circular trail trail bioassays bioassays (Figure (Figure 1) 1) and and in in electrophysiological electrophysiological recordings (Figure2, Table2 ). In trail bioassays (Figures3–5), the trail-following of Camponotus modoc, recordingsrecordings (Figure (Figure 2, 2, Table Table 2). 2). In In trail trail bi bioassaysoassays (Figures (Figures 3–5), 3–5), the the trail-following trail-following of of Camponotus Camponotus modoc modoc, , Lasius niger Myrmica rubra i LasiusLasius niger niger, and, and and Myrmica Myrmica rubra rubra was waswas each each each tested tested tested in in response inresponse response to to ( i()i to)the the ( 6-TPB) 6-TPB the 6-TPB formulated formulated formulated in in pentane, pentane, in pentane, ( ii(ii) ) (iitheirtheir) their own own own trail trail trail pheromone pheromone pheromone form form formulatedulatedulated in in pentane, pentane, in pentane, and and ( andiii(iii) )a (a iiipentane pentane) a pentane control. control. control. Stimuli Stimuli Stimuli were were tested weretested testedat at at1–21–2 1–2 ant ant ant equivalents equivalents equivalents (AE)/58 (AE)/58 (AE) /µL58 µL (µ Ca(L(Ca. .Ca.modoc modoc modoc) )and and) 1–2 and 1–2 AEs/25 1–2AEs/25 AEs µL µL/25 ( La.(La.µL( niger nigerLa. and nigerand M Mand. .rubra rubraM.) )to rubrato account account) to account for for forthethe the length length length differential differential differential of of stimulus stimulus of stimulus trai trailsls trails that that were thatwere weretested tested tested for for large large for ants largeants ( Ca( antsCa. .modoc modoc (Ca.) )and modoc and small small) and ants ants small ( La.(La. ants (La.nigerniger niger and andand M M. .rubra M.rubra rubra) )(see (see) (seeMethods Methods Methods for for detail). detail). for detail).

NameNameName of of Pheromone ofPheromone Pheromone (Amount (Amount (Amount Tested; Tested; Tested; Ant Ant PheromonePheromonePheromone Structures StructuresStructures StudyStudyStudy Species Species Species AntEquivalentsEquivalents Equivalents (AEs)) (AEs)) (AEs)) (Synthetic(Synthetic(Synthetic Sources SourcesSources a–e) a–e)a–e)

(2(2(2SSS,4,4,4RRR,5,5,5SSS)-2,4-Dimethyl-5-hexanolide)-2,4-Dimethyl-5-hexanolide)-2,4-Dimethyl-5-hexanolide Ca.Ca.Ca. modoc modoc modoc (7.5(7.5(7.5 ng; ng;ng; 2 22 AEs) AEs)AEs) (“hexanolide”) (“hexanolide”)(“hexanolide”) [16] [[16]16] (a) (a) (a)

3,4-Dihydro-8-hydroxy-3-7-3,4-Dihydro-8-hydroxy-3-7- 3,4-Dihydro-8-hydroxy-3-7-trimethylisocoumarin La.La. niger niger trimethylisocoumarin (0.5 ng; 1 AE) La. niger trimethylisocoumarin(0.5 ng; 1 AE) (“isocoumarin”) (0.5 ng; 1 [ 17AE)] (“isocoumarin”)(“isocoumarin”) [17] [17] (a)(a)

M.M. rubra rubra 3-Ethyl-2,5-dimethylpyrazine 3-Ethyl-2,5-dimethylpyrazine (5 (5 ng; ng; 1 1 AE) AE) [18] [18] (b) (b)

T.T. caespitum caespitum 2,5-Dimethylpyrazine 2,5-Dimethylpyrazine (1 (1 ng; ng; 1 1 AE) AE) [19] [19] (c) (c)

N.N. cockerelli cockerelli 4-Methyl-3-heptanone 4-Methyl-3-heptanone (10 (10 ng;1 ng;1 AE) AE) [20] [20] (d) (d) Li.Li. humilis humilis (Z(Z)-9-Hexadecenal)-9-Hexadecenal (10 (10 ng; ng; 1 1 AE) AE) [21] [21] CHCH3-(CH3-(CH2)24)-CH=CH-(CH4-CH=CH-(CH2)27)-CHO7-CHO (e) (e) (a)(a) Synthesized Synthesized as as described described by by Re Renyardnyard et et al. al. [16]; [16]; (b) (b) Acros Acros Organics, Organics, New New Jersey, Jersey, USA USA (contains (contains 3- 3- ethyl-2,5-dimethylpyrazineethyl-2,5-dimethylpyrazine at at 50%); 50%); (c) (c) Aldrich Aldrich Chem Chem Co. Co. Milwaukee, Milwaukee, WI, WI, USA; USA; (d) (d) oxidized oxidized from from 4- 4- methyl-3-heptanolmethyl-3-heptanol (Sigma-Aldrich, (Sigma-Aldrich, St. St. Lo Louis,uis, MO, MO, USA); USA); (e) (e) oxidized oxidized from from ( Z(Z)-9-hexadecenol)-9-hexadecenol (Sigma- (Sigma- Aldrich).Aldrich).

2.2. Materials Materials and and Methods Methods

2.1.2.1. Experimental Experimental Insects Insects

2.1.1.2.1.1. Lasius Lasius niger niger BetweenBetween 13 13 and and 31 31 August August (2018), (2018), 5–10 5–10 ants ants were were co collectedllected in in various various containers containers from from each each of of 20 20 sitessites throughout throughout Vancouver Vancouver and and Burnaby, Burnaby, BC. BC. Ants Ants were were bioassayed bioassayed within within 24 24 h h of of collection, collection, and and thenthen cold-euthanized cold-euthanized for for taxonomic taxonomic confirmation confirmation using using multiple multiple keys keys [13,22–24]. [13,22–24].

2.1.2.2.1.2. Camponotus Camponotus modoc modoc CollectionCollection and and maintenance maintenance of of Ca. Ca. modoc modoc nests nests have have recently recently been been described described in in detail detail [16]. [16]. Briefly, Briefly, infestedinfested log log sections sections were were kept kept in in large large plastic plastic bins bins (64 (64 cm cm × × 79 79 cm cm × × 117 117 cm) cm) in in an an outdoor outdoor undercover undercover areaarea exposed exposed to to natural natural light light and and temperature temperature cycles cycles throughout throughout the the year. year. Each Each plastic plastic bin bin housing housing a a nestnest was was connected connected via via clear clear PVC PVC tu tubingbing (2.54 (2.54 cm cm I.D., I.D., Nalgene™ Nalgene™ 180; 180; Sigma-Aldrich, Sigma-Aldrich, St. St. Louis, Louis, MO, MO, USA)USA) to to a a glass glass aquarium aquarium (51 (51 × × 28 28 × × 30 30 cm), cm), which which served served as as the the ants’ ants’ foraging foraging area area provisioned provisioned with with blowblow flies, flies, live live mealworms, mealworms, honey, honey, apples, apples, canned canned chicken, chicken, and and 20% 20% sugar sugar water, water, all all ad ad libitum libitum. .

InsectsInsects 20192019,, 1010,, xx 33 ofof 1212

TableTable 1.1. ListList ofof trailtrail pheromonespheromones (and(and selectselect specspeciesies producingproducing them)them) comprisingcomprising thethe six-trailsix-trail pheromonepheromone blendblend (6-TPB)(6-TPB) testedtested inin circularcircular trailtrail bioassaysbioassays (Figure(Figure 1)1) andand inin electrophysiologicalelectrophysiological recordingsrecordings (Figure(Figure 2,2, TableTable 2).2). InIn trailtrail bibioassaysoassays (Figures(Figures 3–5),3–5), thethe trail-followingtrail-following ofof CamponotusCamponotus modocmodoc,, LasiusLasius nigerniger,, andand MyrmicaMyrmica rubrarubra waswas eacheach testedtested inin responseresponse toto ((ii)) thethe 6-TPB6-TPB formulatedformulated inin pentane,pentane, ((iiii)) theirtheir ownown trailtrail pheromonepheromone formformulatedulated inin pentane,pentane, andand ((iiiiii)) aa pentanepentane control.control. StimuliStimuli werewere testedtested atat 1–21–2 antant equivalentsequivalents (AE)/58(AE)/58 µLµL ((CaCa.. modocmodoc)) andand 1–21–2 AEs/25AEs/25 µLµL ((La.La. nigerniger andand MM.. rubrarubra)) toto accountaccount forfor thethe lengthlength differentialdifferential ofof stimulusstimulus traitrailsls thatthat werewere testedtested forfor largelarge antsants ((CaCa.. modocmodoc)) andand smallsmall antsants ((La.La. nigerniger andand MM.. rubrarubra)) (see(see MethodsMethods forfor detail).detail).

NameName ofof PheromonePheromone (Amount(Amount Tested;Tested; AntAnt PheromonePheromone StructuresStructures StudyStudy SpeciesSpecies EquivalentsEquivalents (AEs))(AEs)) (Synthetic(Synthetic SourcesSources a–e)a–e)

(2(2SS,4,4RR,5,5SS)-2,4-Dimethyl-5-hexanolide)-2,4-Dimethyl-5-hexanolide Ca.Ca. modocmodoc (7.5(7.5 ng;ng; 22 AEs)AEs) (“hexanolide”)(“hexanolide”) [16][16] Insects 2019, 10, 383 3 of 11 (a)(a)

3,4-Dihydro-8-hydroxy-3-7-3,4-Dihydro-8-hydroxy-3-7- 3,4-Dihydro-8-hydroxy-3-7-Table 1. Cont. La.La. nigerniger trimethylisocoumarintrimethylisocoumarin (0.5(0.5 ng;ng; 11 AE)AE) Name of(“isocoumarin”)(“isocoumarin”) Pheromone (Amount [17][17] Tested; Pheromone Structures Study Species Ant Equivalents (AEs)) (Synthetic Sources a–e)(a)(a)

M.M.M. rubrarubra rubra 3-Ethyl-2,5-dimethylpyrazine 3-Ethyl-2,5-dimethylpyrazine3-Ethyl-2,5-dimethylpyrazine (5(5 ng;ng; 111 AE)AE)AE) [ [18][18]18] (b)(b)(b)

T.T.T. caespitumcaespitum caespitum 2,5-Dimethylpyrazine 2,5-Dimethylpyrazine (1 (1 ng; ng; 1 1 AE) AE) [19] [[19]19] (c) (c)(c)

N. cockerelli 4-Methyl-3-heptanone (10 ng;1 AE) [20] N.N. cockerellicockerelli 4-Methyl-3-heptanone 4-Methyl-3-heptanone (10(10 ng;1ng;1 AE)AE) [20][20] (d) (d)(d) Li.Li.Li. humilishumilis humilis ((ZZ)-9-Hexadecenal)-9-Hexadecenal (10 (10 ng; ng; 1 1 AE) AE) [21] [[21]21] CHCHCH3-(CH33-(CH-(CH222)))444-CH-CH=CH-(CH-CH=CH-(CH=CH-(CH22)2)7)7-CHO7-CHO-CHO (e)(e) (e) Li. humilis (Z)-9-Hexadecenal (10 ng; 1 AE) [21] CH3-(CH2)4-CH=CH-(CH2)7-CHO (e) (a) Synthesized as described by Renyard et al. [16]; (b) Acros Organics, New Jersey, USA (contains 3-ethyl-2,5- (a)(a) SynthesizedSynthesized asas describeddescribed byby ReRenyardnyard etet al.al. [16];[16]; (b)(b) AcrosAcros Organics,Organics, NewNew Jersey,Jersey, USAUSA (contains(contains 3-3- dimethylpyrazineethyl-2,5-dimethylpyrazineethyl-2,5-dimethylpyrazine at 50%); (c) Aldrich atat 50%);50%); Chem (c)(c) AldrichAldrich Co. Milwaukee, ChemChem Co.Co. WI,Milwaukee,Milwaukee, USA; (d) WI,WI, oxidized USA;USA; from (d)(d) oxidizedoxidized 4-methyl-3-heptanol fromfrom 4-4- (Sigma-Aldrich,ethyl-2,5-dimethylpyrazine St. Louis, MO, USA); at 50%); (e) oxidized (c) Aldrich from Chem (Z)-9-hexadecenol Co. Milwaukee, (Sigma-Aldrich). WI, USA; (d) oxidized from 4- methyl-3-heptanolmethyl-3-heptanol (Sigma-Aldrich,(Sigma-Aldrich, St.St. LoLouis,uis, MO,MO, USA);USA); (e)(e) oxidizedoxidized fromfrom ((ZZ)-9-hexadecenol)-9-hexadecenol (Sigma-(Sigma- Aldrich).Aldrich). TableAldrich). 2. List of trail pheromone components produced by sympatric ant species inhabiting ant communities in the Pacific Northwest (Western carpenter ants, Camponotus modoc; black garden ants, 2.2. MaterialsMaterials andand MethodsMethods 2. MaterialsLasius niger and; European Methods fire ants, Myrmica rubra) and by allopatric ant species (pavement ants, Tetramorium caespitum; desert harvester ants, Novomessor albisetosus; Argentine ants, Linepithema humilis), as well 2.1.2.1. ExperimentalExperimental InsectsInsects as information as to whether community members (La. niger, Ca. modoc, and M. rubra) antennally respond to these components in electrophysiological recordings (summary of gas chromatographic- 2.1.1.2.1.1. LasiusLasius nigerniger electroantennographic detection (GC-EAD) results; see Figure2). BetweenBetween 1313 andand 3131 AugustAugust (2018),(2018), 5–105–10 antsants werewere cocollectedllected inin variousvarious containerscontainers fromfrom eacheach ofof 2020 sitessites throughoutthroughout VancouverVancouver andand Burnaby,Burnaby, BC.BC. AntsAnts werewere bioassayedbioassayedProduced withinwithin by 24/24Antennal hh ofof collection,collection, Response andand sitesDistribution throughout VancouverSpecies and Burnaby,Trail PheromoneBC. Ants were bioassayed within 24 h of collection, and thenthen cold-euthanizedcold-euthanized forfor taxonomictaxonomic confirmationconfirmation usingusing multiplemultipleLa. keyskeys niger [13,22–24].[13,22–24]. Ca. modoc M. rubra La. niger Isocoumarin * yes/yes yes/yes no/yes 2.1.2.2.1.2. CamponotusCamponotus modocmodoc 2.1.2.Sympatric Camponotus Ca.modoc modoc Hexanolide ** no/yes yes/yes no/yes M. rubra 3-Ethyl-2,5-dimethylpyrazine no/yes no/yes yes/yes CollectionCollection andand maintenancemaintenance ofof Ca.Ca. modocmodoc nestsnests havehave recentlyrecently beenbeen describeddescribed inin detaildetail [16].[16]. Briefly,Briefly, infestedinfested loglog sectionssectionsT. caespitum werewere keptkept inin largelarge2,5-Dimethylpyrazine plasticplastic binsbins (64(64 cmcm ×× 7979 cmcm ×× no 117117/yes cm)cm) inin anan no outdooroutdoor/yes undercoverundercover no/yes areaareaAllopatric exposedexposed toto naturalnaturalN. albisetosus lightlight andand temperaturetemperature4-Methyl-3-heptanone cyclescycles throughoutthroughout thethe no/ yesyear.year. EachEach no plasticplastic/no binbin housinghousing no/no aa Li. humilis (Z)-9-Hexadecenal no/yes no/no no/no nestnest waswas connectedconnected viavia clearclear PVCPVC tutubingbing (2.54(2.54 cmcm I.D.,I.D., Nalgene™Nalgene™ 180;180; Sigma-Aldrich,Sigma-Aldrich, St.St. Louis,Louis, MO,MO, USA)USA) toto aa glassglass aquarium*aquarium 3,4-Dihydro-8-hydroxy-3,5,7-trimethylisocoumarin; (51(51 ×× 2828 ×× 3030 cm),cm), whichwhich servedserved asas thethe ** 2,4-Dimethyl-5-hexanolide. ants’ants’ foragingforaging areaarea provisionedprovisioned withwith blowblow flies,flies, livelive mealworms,mealworms, honey,honey, apples,apples, cannedcanned chicken,chicken, andand 20%20% sugarsugar water,water, allall adad libitumlibitum.. blowHere, flies, we live tested mealworms, the hypotheses honey, apples, that canned sympatric chicken,Ca. and modoc 20%, La. sugar niger water,, and allM. ad rubralibitum(1). sense, and behaviorally respond to, each other’s trail pheromones, and (2) fail to recognize the trail pheromones of allopatric ant species (T. caespitum, N. albisetosus, Li. humilis).

2. Materials and Methods

2.1. Experimental Insects

2.1.1. Lasius Niger Between 13 and 31 August (2018), 5–10 ants were collected in various containers from each of 20 sites throughout Vancouver and Burnaby, BC. Ants were bioassayed within 24 h of collection, and then cold-euthanized for taxonomic confirmation using multiple keys [13,22–24].

2.1.2. Camponotus Modoc Collection and maintenance of Ca. modoc nests have recently been described in detail [16]. Briefly, infested log sections were kept in large plastic bins (64 cm 79 cm 117 cm) in an outdoor undercover × × area exposed to natural light and temperature cycles throughout the year. Each plastic bin housing a nest Insects 2019, 10, 383 4 of 11 was connected via clear PVC tubing (2.54 cm I.D., Nalgene™ 180; Sigma-Aldrich, St. Louis, MO, USA) to a glass aquarium (51 28 30 cm), which served as the ants’ foraging area provisioned with blow flies, × × live mealworms, honey, apples, canned chicken, and 20% sugar water, all ad libitum.

2.1.3. Myrmica Rubra In the spring and summer of 2017 and 2018, 20 nests of M. rubra were dug out of the ground at Inter River Park (North Vancouver, BC, Canada), the Regional Allotment Garden (Burnaby, BC, Canada), and the VanDusen Botanical Garden (Vancouver, BC, Canada). Nests were kept indoors in the Science Research Annex of Simon Fraser University (49◦16033” N, 122◦54055” W) at 25 ◦C and a photoperiod of 12 h L to 12 h D. Nests were housed in small Tupperware dishes (15 15 9 cm) × × (Rubbermaid®, Newell Brands, Atlanta, GA, USA & Royal Sponge Manufacturing Ltd., Toronto, ON, Canada), which were fitted with sterilized potting soil as nesting material and placed inside a small or large tote (41 29 24 cm; 58 43 31 cm) that served as the ants’ foraging area. Twice a week, × × × × the nests were sprayed with water and provisioned with food (fruits, nuts, mealworms, and processed meat). Test tube water reservoirs were replaced when low.

2.2. Gas Chromatographic-Electroantennographic Detection (GC-EAD) Analyses of Synthetic Ant Trail Pheromones For GC-EAD analyses and behavioral bioassays, a synthetic blend of six ant trail pheromones (see above), henceforth six-trail pheromone blend (6-TPB; Table1), was prepared. The blend was analyzed by gas chromatographic-electroantennographic detection (GC-EAD), with procedures and equipment previously described in detail [25,26]. Briefly, the GC-EAD setup employed a Hewlett-Packard 5890 gas chromatograph (GC) fitted with a DB-5 GC column (30 m 0.32 mm I.D.; J&W Scientific, × Folsom, CA, USA). Helium served as the carrier gas (35 cm s 1) with the following temperature program: · − 50 C for 1 min, 20 C min 1 to 280 C. The injector port and flame ionization detector (FID) were set ◦ ◦ · − ◦ to 260 ◦C and 280 ◦C, respectively. For GC-EAD recordings (three antennae each for Ca. modoc, La. niger, and M. rubra), an antenna was carefully dislodged from a worker ant and suspended between two glass capillary electrodes (1.0 0.58 100 mm; A-M Systems, Carlsborg, WA, USA) prepared to accommodate × × the antenna and filled with a saline solution [27].

2.3. General Design of Trail-Following Bioassays All bioassays were run within a metal scaffold (123 57 36 cm) encased in black fabric to eliminate × × external visual stimuli, lit from above with two fluorescent lights (48” 32 W F32T8, one plant and aquarium bulb, and one daylight bulb, Phillips, Amsterdam, The Netherlands), and fitted with a video camera (Sony HDR CX210, Sony, Tokyo, Japan or Canon FS100 A, Canon, Tokyo, Japan) mounted above the bioassay arena (Figure1). The edges of the bioassay arenas (see below) were coated with a mixture of petroleum jelly and mineral oil to prevent the escape of bioassay ants. The specific experimental design to test trail-following responses accounted for body size differentials of large ants (Ca. modoc) and small ants (La. niger and M. rubra). The design for testing Ca. modoc was previously described [16] and is outlined here. Camponotus modoc was tested in a large plexiglass arena (64 44 10 cm) fitted with a filter paper (18.5 cm diam; Sigma-Aldrich, St. Louis, × × MO, USA), with its circular circumference marked with pencil in 1 cm intervals (58 marks total) and treated with one of three test stimuli (see below) at 1–2 ant equivalents (AE)/58 µL. Each bioassay ant (n = 60) entered the arena by exiting a 15 mL Falcon™ “holding” tube (Thermo Fisher Scientific, Waltham, MA, USA) through a hole cut in its tapered tip. Lasius niger and M. rubra were tested in a Pyrex petri dish (15 cm diam) fitted with a small circular filter paper (9.0 cm diam), with its circumference marked with pencil in 1 cm intervals (25 total) and treated with one of three test stimuli (see below) at 1–2 AEs/25 µL. Each worker ant of La. niger (n = 60) and M. rubra (n = 60) entered the Petri dish by exiting a 1.5 mL Axygen™ MaxyClear Snaplock “holding” microtube (Thermo Fisher Scientific, Waltham, MA, USA) through a hole cut in its tapered Insects 2019, 10, 383 5 of 11

tip. Bioassays of large and small ants were initiated by removing the cotton plug from the exit hole

Insectsof a2019 holding, 10, x tube and were terminated after 5 min (Ca. modoc) and 10 min (La. niger and M.5 of rubra 12 ). Filter papers were prepared for bioassays by applying a continuous trail of test stimulus [(i) synthetic 6-TPB;The number (ii) synthetic of 1 trailcm intervals pheromone an ofant the had bioassay followed ant; during or (iii) a solventbioassay control; served Table as the1]. response criterionThe and number was analyzed of 1 cm intervals by viewing an ant the had video followed footage. during Ants a bioassaynot leaving served their as theholding response tube criterion after 10 andmin was were analyzed considered by viewing non-responders the video footage. and excl Antsuded not leavingfrom analyses. their holding Between tube afterbioassays, 10 min all were preparativeconsidered surfaces non-responders and bioassay and excluded arenas fromwere analyses.cleaned Betweenwith 70% bioassays, EtOH and all preparativehexane, and surfaces the experimentand bioassay room arenas was aired were cleanedout for 5 with to 10 70% min EtOH by opening and hexane, an exterior and the door. experiment A new ant room was was tested aired for out eachfor treatment. 5 to 10 min All by openingLa. niger an and exterior M. rubra door. ants A new were ant collected was tested from for eachdifferent treatment. laboratory All La. or niger fieldand colonies.M. rubra Workerants were ants collected of Ca. frommodoc di ffwereerent collected laboratory from or field six colonies.colonies Workermaintained ants ofin Ca.an modocoutdoorwere enclosure.collected from six colonies maintained in an outdoor enclosure.

FigureFigure 1. Graphical 1. Graphical illustration illustration of the of theexperimental experimental design design used used for fortesting testing trail-following trail-following of Western of Western carpentercarpenter ants, ants, CamponotusCamponotus modoc modoc, ,black black garden garden ants, LasiusLasius nigerniger,, and and European European fire fire ants, ants,Myrmica Myrmica rubra , rubrain, response in response to theirto their own own trail trail pheromone pheromone or or a a complex complex blend blend of of six six trailtrail pheromonespheromones (see(see Table1 ). To account for body size differentials of large ants (Ca. modoc) and small ants (La. niger; M. rubra), 1). To account for body size differentials of large ants (Ca. modoc) and small ants (La. niger; M. rubra), bioassay arenas were large (64 cm wide 44 cm long 10 cm high) or small (circular, 15 cm diam 1 cm bioassay arenas were large (64 cm wide ×× 44 cm long × 10 cm high) or small (circular, 15 cm diam× × 1 high), and the diameter of the filter paper was 18.5 and 9 cm, respectively. Pheromone trails were cm high), and the diameter of the filter paper was 18.5 and 9 cm, respectively. Pheromone trails were applied to the filter paper along the red dotted line (which was absent in bioassays). applied to the filter paper along the red dotted line (which was absent in bioassays). 2.4. Statistics 2.4. Statistics R (Version 3.5.0; multicomp, plotrix, & plyr packages) was used to analyze the data and produce graphicsR (Version [28–30 3.5.0;]. A multicomp, generalized plot linearrix, model & plyr (GLM; packages) quasi-Poisson was used distribution) to analyze the was data used and to analyzeproduce the graphicsdistances [28–30]. ants travelled A generalized following linear trails model in response (GLM; quasi-Poisson to the various typesdistribution) of trails was presented. used to An analyze analysis theof distances variance ants (ANOVA) travelled and following Tukey’s honesttrails in significant response dito fftheerence various (HSD) types test of were trails used presented. to determine An analysissignificant of variance differences (ANOVA) in mean and distance Tukey’s travelled honest in significant response todifference trail type. (HSD) test were used to determine significant differences in mean distance travelled in response to trail type.

3. Results

3.1. Gas Chromatographic-Electroantennographic Detection (GC-EAD) Analyses of Synthetic Ant Trail Pheromones In GC-EAD analyses, La. niger antennae responded to (in the order of elution) synthetic 2,5- dimethylpyrazine, 4-methyl-3-heptanone, 3-ethyl-2,5-dimethylpyrazine, 3-ethyl-2,6-dimethyl pyrazine, hexanolide, isocoumarin (its own trail pheromone), and (Z)-9-hexadecenal (Figure 2; Table

Insects 2019, 10, 383 6 of 11

3. Results

3.1. Gas Chromatographic-Electroantennographic Detection (GC-EAD) Analyses of Synthetic Ant Trail Pheromones In GC-EAD analyses, La. niger antennae responded to (in the order of elution) synthetic 2,5-dimethylpyrazine, 4-methyl-3-heptanone, 3-ethyl-2,5-dimethylpyrazine, 3-ethyl-2,6-dimethyl pyrazine, Insectshexanolide, 2019, 10, xisocoumarin (its own trail pheromone), and (Z)-9-hexadecenal (Figure2; Table2 ). Antennae6 of 12 of Ca. modoc responded to 2,5-dimethylpyrazine, 3-ethyl-2,5-dimethylpyrazine, 3-ethyl-2,6-dimethyl 2).pyrazine, Antennae hexanolide of Ca. modoc (its own responded trail pheromone), to 2,5-dimethylpyra and isocoumarinzine, 3-ethyl-2,5-dimethylpyrazine, (Figure2; Table2 ). Antennae 3- of ethyl-2,6-dimethylM. rubra responded pyrazine, to 2,5-dimethylpyrazine, hexanolide (its own 3-ethyl-2,5-dimethylpyrazine trail pheromone), and isocoumarin (its own (Figure trail pheromone), 2; Table 2).3-ethyl-2,6-dimethyl Antennae of M. rubra pyrazine, responded hexanolide, to 2,5-dimethylpyrazine, and isocoumarin (Figure 3-ethyl-2,5-dimethylpyrazine2; Table2). (its own trail pheromone), 3-ethyl-2,6-dimethyl pyrazine, hexanolide, and isocoumarin (Figure 2; Table 2).

FigureFigure 2. 2.RepresentativeRepresentative recordings (n(n= =3 3 each) each) of theof the responses responses of a gasof chromatographica gas chromatographic flame ionization flame ionizationdetector (FID)detector and (FID) an electroantennographic and an electroantennographic detector (EAD:detector Antenna (EAD: ofAntenna a Lasius of niger a Lasius, Camponotus niger, Camponotusmodoc, or Myrmicamodoc, orrubra Myrmicaworker rubra ant) worker to pheromone ant) to pheromone components comp presentonents in present the six-trail in the pheromone six-trail pheromoneblend (see blend Table1 ).(see Numbers Table 1). in theNumbers FID trace in referthe toFID the trace following refer pheromoneto the following components: pheromone (1) 2,5- components:dimethylpyrazine; (1) 2,5-dimethylpyrazine; (2) 4-methyl-3-heptanone; (2) 4-methyl (3) 3-ethyl-2,5-dimethylpyrazine;-3-heptanone; (3) 3-ethyl-2,5-dimethylpyrazine; * = 3-ethyl-2,6-dimethyl * = 3-ethyl-2,6-dimethylpyrazine (non-natural pyrazine isomer present (non-natural in synthetic isomer source); present (4) (2 inS,4 syntheticR,5S)-2,4-dimethyl-5-hexanolide; source); (4) (2S,4R,5S)-2,4- (5) 3,4- dimethyl-5-hexanolide;dihydro-8-hydroxy-3,5,7-trimethylisocoumarin; (5) 3,4-dihydro-8-hydroxy-3,5,7-trimethylisocoumarin; and (6) (Z)-9-hexadecanal. and (6) (Z)-9- hexadecanal. 3.2. Trail-Following Bioassays TableThere 2. wereList of significant trail pheromone differences components in the produced distances by (mean sympatricSE) ant that species worker inhabiting ants of antLa. niger ± travelledcommunities following in the trails Pacific of Nort thehwest 6-TPB (Western (495.5 carpenter57.5 cm), ants, the Camponotus isocoumarin modoc (199.3; black garden43.1 cm), ants, and the ± ± solventLasius controlniger; European (32.2 14.0fire ants, cm) (ANOVA,Myrmica rubra F =) and34.028, by allopatric degrees ofant freedom species (pavement (df) = 2, residualants, df ± = 57,Tetramoriump < 0.001; caespitum Figure; 3desert). Based harvester on Tukey’s ants, Novomessor HSD tests, albisetosus all distances; Argentine di ffered ants, from Linepithema one another (pairwisehumilis), comparisons: as well as information Isocoumarin as to whether vs. solvent community control: membersp = 0.001; (La. 6-TPB niger, vs.Ca. modoc, solvent and control: M. rubrap )< 0.001; 6-TPBantennally vs. isocoumarin: respond top these< 0.001). components in electrophysiological recordings (summary of gas chromatographic-electroantennographic detection (GC-EAD) results; see Figure 2).

Produced by/Antennal Response Distribution Species Trail Pheromone La. niger Ca. modoc M. rubra La. niger Isocoumarin * yes/yes yes/yes no/yes Ca. modoc Hexanolide ** no/yes yes/yes no/yes Sympatric 3-Ethyl-2,5- M. rubra no/yes no/yes yes/yes dimethylpyrazine T. caespitum 2,5-Dimethylpyrazine no/yes no/yes no/yes Allopatric N. albisetosus 4-Methyl-3-heptanone no/yes no/no no/no Li. humilis (Z)-9-Hexadecenal no/yes no/no no/no * 3,4-Dihydro-8-hydroxy-3,5,7-trimethylisocoumarin; ** 2,4-Dimethyl-5-hexanolide.

Insects 2019, 10, x 7 of 12

3.2. Trail-Following Bioassays There were significant differences in the distances (mean ± SE) that worker ants of La. niger travelled following trails of the 6-TPB (495.5 ± 57.5 cm), the isocoumarin (199.3 ± 43.1 cm), and the solvent control (32.2 ± 14.0 cm) (ANOVA, F = 34.028, degrees of freedom (df) = 2, residual df = 57, p <

Insects0.001;2019 Figure, 10, 383 3). Based on Tukey’s HSD tests, all distances differed from one another (pairwise7 of 11 comparisons: Isocoumarin vs. solvent control: p = 0.001; 6-TPB vs. solvent control: p < 0.001; 6-TPB vs. isocoumarin: p < 0.001).

FigureFigure 3. 3.Distances Distances workerworker antsants of Lasius niger (n(n == 60)60) travelled travelled following following trails trails of ofsynthetic synthetic 3,4- 3,4-dihydro-8-hydroxy-3,5,7-trimethylisocoumarindihydro-8-hydroxy-3,5,7-trimethylisocoumarin (the (the known known trail trail pheromone pheromone of La. of La.niger niger [17]),[17 a ]),six- a six-trailtrail pheromone pheromone blend blend (Table (Table 1),1 and), and a solvent a solvent control control applied applied to tothe the circumferenc circumferencee of a of circular a circular filter filterpaper paper (diam: (diam: 90 mm) 90 mm) marked marked in 1-cm in 1-cm intervals intervals (Figure (Figure 1). Grey1). Greyand black and blacksymbols symbols show showthe distance the distancethat each that ant each and ant 20 and ants 20 on ants aver onage average (mean (mean± whiskers),whiskers), respectively respectively,, travelled travelled following following trails.trails. Means ± Meansassociated associated with withdifferent different letters letters are statistically are statistically different different (Tukey’s (Tukey’s honest honest significant significant difference difference (HSD) (HSD)test, test,p < 0.01);p < 0.01); six out six of out 66 of ants 66ants tested tested did didnot notenter enter the thebioassay bioassay arena arena and and were were excluded excluded from from this thisdata data set. set.

ThereThere were were also also significant significant di ffdifferenceserences in in the the distances distances that that worker worker ants ants of ofCa. Ca. modoc modoctravelled travelled following trails of the 6-TPB (213.1 48.7 cm), the hexanolide (206.0 32.7 cm), and the solvent control following trails of the 6-TPB (213.1± ± 48.7 cm), the hexanolide (206.0± ± 32.7 cm), and the solvent control (69.1 12.6 cm) (ANOVA, F = 7.5583, df = 2, residual df = 57, p < 0.01; Figure4). Based on Tukey’s (69.1± ± 12.6 cm) (ANOVA, F = 7.5583, df = 2, residual df = 57, p < 0.01; Figure 4). Based on Tukey’s HSDHSD tests, tests, distances distances were were di ffdifferenterent between between the the 6-TPB 6-TPB and and the the solvent solvent control control (p <(p 0.01),< 0.01), and and between between thethe hexanolide hexanolide and and the the solvent solvent control control (p (

Figure 4. Distances worker ants of Camponotus modoc (n = 60) travelled following trails of synthetic (2SFigure,4R,5S)-2,4-dimethyl-5-hexanolide 4. Distances worker ants of (the Camponotus known trail modoc pheromone (n = 60) of Ca.travelled modoc [following16]), a six-trail-pheromone trails of synthetic (2S,4R,5S)-2,4-dimethyl-5-hexanolide (the known trail pheromone of Ca. modoc [16]), a six-trail- pheromone blend (Table 1), and a solvent control applied to the circumference of a circular filter paper (diam: 185 mm) marked in 1 cm intervals (Figure 1). Grey and black symbols show the distance that each ant and 20 ants on average (mean ± whiskers) travelled, respectively, following trails. Means associated with different letters are statistically different (Tukey’s honest significant difference (HSD) test, p < 0.01); two out of 62 ants tested did not enter the bioassay arena and were excluded from this data set.

There were no significant differences in the distances that worker ants of M. rubra travelled following trails of the 6-TPB (22.9 ± 8.7 cm), the 3-ethyl-2,5-dimethylpyrazine (41.2 ± 8.3 cm), and the solvent control (32.9 ± 8.1 cm) (ANOVA, F = 1.1555, df = 2, residual df = 57, p = 0.3222; Figure 5).

Figure 5. Distances worker ants of Myrmica rubra (n = 60) travelled following trails of synthetic 3- ethyl-2,5-dimethylpyrazine (the known trail pheromone of M. rubra [19]), a six-trail-pheromone blend (Table 1), and a solvent control applied to the circumference of a circular filter paper (diam: 90 mm) marked in 1-cm intervals (Figure 1). Grey and black symbols show the distance that each ant and 20 ants on average (mean ± whiskers) travelled, respectively, following trails. Means associated with different letters are statistically different (Tukey’s honest significant difference (HSD) test, p < 0.01).

Insects 2019, 10, x 8 of 12

Figure 4. Distances worker ants of Camponotus modoc (n = 60) travelled following trails of synthetic Insects 2019, 10, 383 8 of 11 (2S,4R,5S)-2,4-dimethyl-5-hexanolide (the known trail pheromone of Ca. modoc [16]), a six-trail- pheromone blend (Table 1), and a solvent control applied to the circumference of a circular filter paper blend(diam: (Table 1851 ),mm) and marked a solvent in control 1 cm appliedintervals to (Figure the circumference 1). Grey and of ablack circular symbols filter paper show (diam: the distance 185 mm) that markedeach ant in 1and cm intervals20 ants on (Figure average1). Grey (mean and ± blackwhiskers symbols) travelled, show the respectively distance that, following each ant andtrails. 20 antsMeans onassociated average (mean with differentwhiskers) letters travelled, are statistically respectively, different following (Tukey’s trails. honest Means significant associated difference with different (HSD) ± letterstest, arep < 0.01); statistically two out different of 62 ants (Tukey’s tested honest did not significant enter the differencebioassay arena (HSD) and test, werep < excluded0.01); two from out ofthis 62data ants set. tested did not enter the bioassay arena and were excluded from this data set. There were no significant differences in the distances that worker ants of M. rubra travelled There were no significant differences in the distances that worker ants of M. rubra travelled following trails of the 6-TPB (22.9 8.7 cm), the 3-ethyl-2,5-dimethylpyrazine (41.2 8.3 cm), and the following trails of the 6-TPB (22.9± ± 8.7 cm), the 3-ethyl-2,5-dimethylpyrazine (41.2± ± 8.3 cm), and the solvent control (32.9 8.1 cm) (ANOVA, F = 1.1555, df = 2, residual df = 57, p = 0.3222; Figure5). solvent control (32.9± ± 8.1 cm) (ANOVA, F = 1.1555, df = 2, residual df = 57, p = 0.3222; Figure 5).

Figure 5. Distances worker ants of Myrmica rubra (n = 60) travelled following trails of synthetic Figure 5. Distances worker ants of Myrmica rubra (n = 60) travelled following trails of synthetic 3- 3-ethyl-2,5-dimethylpyrazine (the known trail pheromone of M. rubra [19]), a six-trail-pheromone blend ethyl-2,5-dimethylpyrazine (the known trail pheromone of M. rubra [19]), a six-trail-pheromone blend (Table1), and a solvent control applied to the circumference of a circular filter paper (diam: 90 mm) (Table 1), and a solvent control applied to the circumference of a circular filter paper (diam: 90 mm) marked in 1-cm intervals (Figure1). Grey and black symbols show the distance that each ant and marked in 1-cm intervals (Figure 1). Grey and black symbols show the distance that each ant and 20 20 ants on average (mean whiskers) travelled, respectively, following trails. Means associated with ants on average (mean ±± whiskers) travelled, respectively, following trails. Means associated with different letters are statistically different (Tukey’s honest significant difference (HSD) test, p < 0.01). different letters are statistically different (Tukey’s honest significant difference (HSD) test, p < 0.01). 4. Discussion

As predicted, La. niger, Ca. modoc, and M. rubra did sense (antennally respond to) the trail pheromone of all community members (La. niger, Ca. modoc, M. rubra; Figure2) and, except for La. niger, did not recognize the trail pheromones of two allopatric ant species (N. cockerelli and Li. humilis; Table2). That all three ant species sensed the trail pheromone of allopatric T. caespitum could be due its molecular structure (2,5-dimethylpyrazine) resembling that of the M. rubra trail pheromone (3-ethyl-2,5-dimethylpyrazine). In light of our behavioral data that the 6-TPB (which contains trail pheromones of con- and hetero-specifics) readily induced trail-following behavior of Ca. modoc and La. niger, it seems that these ants either simply ignore (Ca. modoc), or indeed eavesdrop on (La. niger), each other’s trail pheromone communication. In general, eavesdropping ants can face aggression, increased competition, or even displacement [1,6,8–10], but in the ant community we studied here, eavesdropping may accrue more benefits than harm, or at least, no harm. This inference is based on our bioassay data showing that (i) La. niger workers followed trails of the 6-TPB for a longer distance than they followed their own trail pheromone (isocoumarin), and (ii) Ca. modoc workers followed trails of the 6-TPB and their own pheromone (hexanolide) for similar distances. Unexpectedly, workers of M. rubra followed trails of the 6-TPB and their own trail pheromone (3-ethyl-2,5-dimethylpyrazine) only as much as a solvent control trail, demonstrating no effect of the trail pheromone in this type of bioassay. There are at least two explanations why M. rubra did not Insects 2019, 10, 383 9 of 11 follow a trail of synthetic 3-ethyl-2,5-dimethylpyrazine. Prior studies that demonstrated distinct trail following by M. rubra, either in “no-choice bioassays” comparable to our experimental design [18] or in “binary-choice arena bioassays” [31], tested the responses of multiple workers (the entire nest), whereas we tested the responses of individual ants. Given the small foraging range and high nest density of M. rubra in North America [31,32], it is conceivable that nest mates do not forage on their own but engage in group foraging, as shown in many Myrmica species [33]. Group foraging entails cooperative interactions, where, for example, a successful forager recruits nestmates and physically guides them to the food source [33]. Group foraging may improve the overall foraging effort of a nest and facilitate transport of food particles that are too heavy for single ants to carry [34]. Alternatively, the trail pheromone blend of M. rubra comprises not only 3-ethyl-2,5-dimethylpyrazine but additional pheromone components, which, thus far, have eluded identification. The evidence presented here that some ant community members eavesdrop on and exploit each other’s trail pheromone has major implications for ant control. Food baits laced with lethal agents show promise as an ant control tactic because many ants share food through trophallaxis and thus may spread the poison together with the food throughout their entire nest. The effect of lethal food baits can be enhanced by adding attractants. For example, the admixture of trail pheromone to food baits increased bait consumption by the invasive Argentine ant, L. humile [35]. In M. rubra, a path of synthetic trail pheromone leading from a nest to a food bait is more effective in recruiting foragers than applying the trail pheromone around a food bait [31]. Commercial development of trail pheromones for ant control is contingent upon economic feasibility. With so many important ant species in need of control, and with each species producing its own trail pheromone, manufacturing species-specific (single target) trail pheromone lures (ropes, strings) does not seem economically viable. However, if trail pheromones of multiple ant species were to be combined in a single lure (multiple targets), with potential synergism and no antagonism between components, as shown in our study, then an ant control tactic that couples a lethal food bait with a trail pheromone lure seems commercially feasible. As an added advantage, the ant species targeted for control would not even need to be identified by a pest control professional or the lay person buying the control technology in a retail store. Future studies should aim to strengthen the proof of concept presented in our study. Trail pheromones of major ant pests such as the red imported fire ant, Solenopsis invicta, should be added to the multiple-species trail pheromone lure and tested for the response of S. invicta and other species. Moreover, research needs to be initiated on dispensers capable of sustained release of trail pheromones in field experiments and, eventually, operational applications.

5. Conclusions All three select members of ant communities in the Lower Mainland of BC (La. niger, Ca. modoc, M. rubra) sensed each other’s trail pheromone and, except for La. niger, did not recognize the trail pheromones of two allopatric ant species (N. cockerelli and Li. humilis). Workers of La. niger followed a synthetic trail pheromone blend (containing the trail pheromone of all three community members and those of three allopatric ant species) for a longer distance than they followed their own trail pheromone, and Ca. modoc workers followed this blend and their own trail pheromone for similar distances. Apparently, these ants either ignore (Ca. modoc), or indeed eavesdrop on (La. niger), each other’s trail pheromone. Eavesdropping ants may accrue benefits by learning about the location of profitable food sources. If synthetic trail pheromones of multiple pest ant species were to be combined in a single (rope-type) lure, with potential synergism and no antagonism between components (as shown in our study), an ant control tactic that presents a lethal food bait together with a trail pheromone lure seems commercially viable.

Author Contributions: Conceptualization, J.M.C., A.R., R.G., D.H. and G.G.; methodology, J.M.C., A.R., R.G., D.H. and G.G.; validation, J.M.C., A.R., R.G., D.H. and G.G.; formal analysis, J.M.C., A.R. and R.G.; investigation, J.M.C., A.R. and R.G.; resources, J.M.C., A.R., R.G., D.H., S.K.A. and G.G.; data curation, J.M.C. and A.R.; Insects 2019, 10, 383 10 of 11

Writing—Original draft preparation, J.M.C., R.G. and G.G.; Writing—Review and editing, J.M.C., A.R., R.G., D.H. and G.G.; visualization, J.M.C. and S.K.A.; supervision, J.M.C., A.R., R.G., D.H. and G.G.; project administration, J.M.C., R.G. and G.G.; funding acquisition, J.M.C., D.H., A.R. and G.G. Funding: This research was supported by a Vice President Research-Undergraduate Student Research Award from Simon Fraser University (SFU) and a Graduate Entrance Scholarship to J.M.C.; a Graduate Fellowship from SFU and a Natural Sciences and Engineering Research Council of Canada (NSERC)-CGSM to A.R.; a Graduate Fellowship from SFU and a Thelma Finlayson Graduate Fellowship to D.H., and by an NSERC-Industrial Research Chair to G.G., with Scotts Canada Ltd. as an industrial sponsor. Acknowledgments: We thank Robert Higgins and Laurel Hansen for identification of ant species, Michael Gudmundson for assisting with field collections of Ca. modoc, Grady Ott for providing nest bins, Jan Lee, Ashley Munoz, Shelby Kwok, Sebastian Damin, Stephanie Fan, Kris Cu, Hanna Jackson, Jessica Chalissery, Rosemary Vayalikunnel, and Zhanata Almazbekova for assistance with colony maintenance (all species), field collections of M. rubra, and experiments. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

References

1. Hölldobler, B.; Wilson, E.O. The Ants, 1st ed.; Belknap Press of Harvard University Press: Cambridge, MA, USA, 1990. 2. Morgan, E.D. Trail pheromone of ants. Physiol. Entomol. 2009, 34, 1–17. [CrossRef] 3. Detrain, C.; Deneubourg, J.L. Collective decision-making and foraging patterns in ants and honeybees. Adv. Physiol. 2008, 35, 123–173. [CrossRef] 4. Beckers, R.; Deneubourg, J.L.; Goss, S.; Pasteels, J.M. Collective decision making through food recruitment. Insectes Soc. 1990, 37, 258–267. [CrossRef] 5. Vander Meer, R.K.; Alonso, L.E. Pheromone directed behaviour in ants. In Pheromone Communication in Social Insects; Ants, , Bees, and Termites, 1st ed.; Vander Meer, R.K., Breed, M.D., Espelie, K.E., Winston, M.L., Eds.; Westview Press: Boulder, CO, USA, 1998. 6. Wilson, E.O. Trail sharing in ants. Psyche 1965, 72, 2–7. [CrossRef] 7. Adams, E.S. Interaction between the ants Zacryptocerus maculatus and Azteca trigona: Interspecific parasitization of information. Biotropica 1990, 22, 200–206. [CrossRef] 8. Gobin, B.; Peeters, C.; Billen, J.; Morgan, E.D. Interspecific trail following and commensalisms between the ponerine ant Gnamptogenys menadensis and the formicine ant Polyrhachis rufipes. J. Insect. Behav. 1998, 11, 361–369. [CrossRef] 9. Menzel, F.; Pokorny, T.; Blüthgen, N.; Schmitt, T. Trail-sharing among tropical ants: Interspecific use of trail pheromones? Ecol. Entomol. 2010, 35, 495–503. [CrossRef] 10. Klotz, J.H. Diel differences in foraging in two ant species (: Formicidae). J. Kansas Entomol. Soc. 1984, 57, 111–118. 11. Buckell, E.R. A list of the ants of British Columbia. J. Entomol. Soc. B. C. 1932, 29, 22–25. 12. Higgins, R.J.; Lindgren, B.S. An evaluation of methods for sampling ants (Hymenoptera: Formicidae) in British Columbia, Canada. Can. Entomol. 2012, 144, 491–507. [CrossRef] 13. Naumann, K.; Preston, W.B.; Ayre, G.L. An annotated checklist of the ants (Hymenoptera: Formicidae) of British Columbia. J. Entomol. Soc. B. C. 1999, 96, 29–68. 14. Groden, E.; Drummond, F.A.; Garnas, J.; Franceour, A. Distribution of an invasive ant, Myrmica rubra (Hymenoptera: Formicidae), in Maine. J. Econ. Entomol. 2005, 98, 1774–1784. [CrossRef][PubMed] 15. Naumann, K.; Higgins, R.J. The European fire ant (Hymenoptera: Formicidae) as an invasive species: Impact on local ant species and other epigaeic . Can. Entomol. 2015, 147, 592–601. [CrossRef] 16. Renyard, A.; Alamsetti, S.; Gries, R.; Munoz, A.; Gries, G. Identification of the trail pheromone of the carpenter ant, Camponotus modoc. J. Chem. Ecol. 2019. accepted for publication. 17. Bestmann, H.J.; Kern, F.; Schäfer, D.; Witschel, M.C. 3,4-Dihydroisocoumarins, a new class of ant trail pheromones. Angew. Chem. Int. Ed. 1992, 31, 795–796. [CrossRef] 18. Evershed, R.P.; Morgan, E.D.; Cammaerts, M.C. 3-Ethyl-2,5-dimethylpyrazine, the trail pheromone from the venom gland of eight species of Myrmica ants. Insect Biochem. 1982, 12, 383–391. [CrossRef] Insects 2019, 10, 383 11 of 11

19. Attygalle, A.B.; Morgan, E.D. Identification of trail pheromone of the ant Tetramorium caespitum L. (Hymenoptera: Myrmicinae). Naturwissenschaften 1983, 70, 364–365. [CrossRef] 20. Hölldobler, B.; Oldham, N.J.; Morgan, E.D.; König, W.A. Recruitment pheromones in the ants Aphaenogaster albisetosus and A. cockerelli (Hymenoptera: Formicidae). J. Insect Physiol. 1995, 41, 739–744. [CrossRef] 21. Van Vorhis Key, S.E.; Baker, T.C. Trail-following responses of the Argentine ant, Iridomyrmex humilis (Mayr), to a synthetic trail pheromone component and analogs. J. Chem. Ecol. 1982, 8, 3–14. [CrossRef] 22. Mackay, W.P.; Mackay, E. The Ants of New Mexico (Hymenoptera: Formicidae); Edwin Mellend Press: Lewiston, NY, USA, 2002. 23. Wilson, E.O. A monographic revision of the ant genus Lasius. Bull. Mus. Comp. Zool. 1955, 113, 1–201. 24. Wing, M.W. Taxonomic Revision of the Nearctic Genus Acanthomyops (Hymenoptera: Formicidae); Cornell University, New York State College of Agriculture, Agricultural Experimentation Station: Ithica, NY, USA, 1968. 25. Arn, H.; Städler, E.; Rauscher, S. The electroantennographic detector—A selective and sensitive tool in the gas chromatographic analysis of insect pheromones. Z. Naturforsch. 1975, 30c, 11–12. [CrossRef] 26. Gries, R.; Khaskin, G.; Gries, G.; Bennett, R.G.; King, G.G.S.; Morewood, P.; Slessor, K.N.; Morewood, W.D. (Z,Z)-4,7-Tridecadien-(S)-2-yl acetate: Sex pheromone of Douglas-fir cone gall midge, Contarinia oregonensis. J. Chem. Ecol. 2002, 28, 2283–2297. [CrossRef][PubMed] 27. Staddon, B.W.; Everton, I.J. Haemolymph of the milkweed bug Oncopeltus fasciatus (Heteroptera: Lygaeidae): Inorganic constituents and amino acids. Comp. Biochem. Physiol. A 1980, 65, 371–374. [CrossRef] 28. Hothorn, T.; Betz, F.; Westfall, P. Simultaneous inference in general parametric models. Biom. J. 2008, 50, 346–363. [CrossRef][PubMed] 29. Lemon, J. Plotrix: A package in the red light district of R. R-News 2006, 6, 8–12. 30. Wickham, H. The split-apply-combine strategy for data analysis. J. Stat. Softw. 2011, 40, 1–29. [CrossRef] 31. Hoefele, D.; Chalissery, J.M.; Gries, R.; Gries, G. Effects of trail pheromone purity, dose and type of placement on recruiting European fire ants, Myrmica rubra, to food baits. J. Entomol. Soc. B. C. 2019. under review. 32. Higgins, R.J.; (Thompson Rivers University, Kamloops, BC, Canada); Hoefele, D.; (Simon Fraser University, Burnaby, BC, Canada). Personal communication, 2017. 33. Fedoseeva, E.B. A technological approach to the description of group foraging in the ant Myrmica rubra. Entomol. Rev. 2015, 95, 984–999. [CrossRef] 34. Carroll, C.R.; Janzen, D.H. Ecology of foraging by ants. Annu. Rev. Ecol. Syst. 1973, 4, 231–257. [CrossRef] 35. Welzel, K.F.; Choe, D.H. Development of a pheromone-assisted baiting technique for Argentine ants (Hymenoptera: Formicidae). J. Econ. Entomol. 2016, 109, 1303–1309. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).