Open Life Sci. 2015; 10: 1–6

Research Article Open Access

Peter Kaňuch, Anna Sliacka, Anton Krištín* Habitat-conditioned feeding behaviour in constrictus (: )

Abstract: Some herbivores can regulate their their host plants [1,2]. While strictly monophagous species nourishment intake by different feeding behaviour. suffer from a deficiency of eating their preferred food and This mechanism allows them to persist with utilising are forced to migrate into new areas [e.g. 3,4], other species different food resources according to the composition of are able to persist with utilising different food resources the vegetation within their habitats. Using a two-choice according to the composition of the vegetation within experiment, we analysed foraging behaviour in females their habitats [5,6]. A balanced intake of nutrients is the of the tree-dwelling bush-cricket most limiting factor influencing the fitness of herbivores (Orthoptera), which originated from two different [e.g. 7,8]. However, insect herbivores can actively regulate forest habitats, spruce and beech forest. We found that protein-carbohydrate intake using both physiological and individuals from the spruce forest mainly foraged on behavioural mechanisms [reviewed by 9,10]. needle tips, and thus they nibbled more needles per day One such behavioural mechanism can be an insect’s than individuals from the beech forest (medians 106.0 vs. ability to adapt feeding behaviour in different habitats, 42.5; p < 0.0001). However, when the contents of droppings and is defined as a change in behaviour that comes with were dissected, the volume of consumed spruce was experience [11]. It can facilitate host selection and feeding similar in both groups of bush-crickets (median > 90%), strategy in herbivorous species and is often interpreted which is explained by the different feeding techniques of as an adaptive consequence of natural selection [12-14]. bush-crickets from different habitats. We propose possible This insect’s ability has been clearly documented scenarios for bush-cricket feeding adaptations to the through artificial selection and conditioning over several deleterious effects of the host plant chemical compounds generations that produced populations with improved serving as a plant defence against herbivores. behavioural responses [e.g. 15,16]. Therefore, some species that are successfully introduced into new areas or habitats Keywords: Diet, Bush-crickets, Phytophagy, Insect, may also attack plants that were not previously known to Choice, Woodland be acceptable for them, which is very important for the control of pests [17-19]. In order to investigate how a herbivorous insect can DOI 10.1515/biol-2015-0001 alter its foraging behaviour, we used the tree-dwelling and Received March 4, 2014; accepted August 12, 2014 flightless bush-cricket, Barbitistes constrictus (Orthoptera: Tettigoniidae). The range of the species closely matches the distribution of coniferous and beech forests in Central 1 Introduction and Eastern Europe [e.g. 20,21,28]. However, besides its primary habitat, it seldom also breeds (most likely Many herbivorous are consistent in their feeding thanks to human-mediated introduction events, e.g. [2]) habits during their evolution, but some species may alter in non-characteristic areas including broadleaved forests. their foraging behaviour as a response to changes within Thus the availability of the most preferred coniferous food [20,22,23] strongly differs in these habitats as it is either superabundant (in spruce forests) or very limited (in beech forests). In addition, such populations are *Corresponding author: Anton Krištín: Institute of Forest Ecology, potentially long-term conditioned to overcome deleterious Slovak Academy of Sciences, 960 53 Zvolen, Slovakia, E-mail: effects of different secondary carbon-based compounds, [email protected] which are synthesised by available host plants to defend Peter Kaňuch, Anna Sliacka: Institute of Forest Ecology, Slovak against herbivores (monoterpenes in spruce versus Academy of Sciences, 960 53 Zvolen, Slovakia

© 2015 Peter Kaňuch, et al., licensee De Gruyter Open. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License. 2 P. Kaňuch et al. phenols in beech) [24,25]. Therefore, one can hypothesise (dominant tree species – P. abies) were collected at two that populations are subjected to natural selection on sites (1068–1658 m a.s.l.) while individuals from a ‘beech account of differences in food plants and their different forest’ (dominant tree species – European beech, Fagus parts [cf. 11,26]. In this experimental study, we analysed sylvatica) were collected at five sites (497–868 m a.s.l.). the foraging behaviour of B. constrictus in relation to After transporting the specimens into the lab, they the source environment from where these individuals were reared (7–10 days) according to standard conditions originated. (air temperature 22–26 °C, relative humidity ~50%, natural daylight) and fed leaves of European dewberry, Rubus caesius, ad libitum. All individuals, before entering the 2 Experimental Procedures experiment (June 24 – July 10), were kept acclimatised for at least one week or until they reached adulthood. 2.1 Study species Finally, 22 adult females (2–5 days of adulthood) from the spruce forest (91% of them were classified as dark-green The eastern saw-tailed bush-cricket, B. constrictus, is a tree- phenotype) and 23 (2–6 days of adulthood) from the beech dwelling species occurring mainly in coniferous forests, forest (74% light-green phenotype) were used for 10 days but some isolated populations can also be found in forests in a parallel experiment (from both populations) that where broadleaved trees (beech, oak, hornbeam) dominate involved a choice of two different diets. (No males were [22,27]. Although the species was found to be a pest among involved in the experiment because of the low sample size.) the seedlings of conifers in the past, nowadays it is recorded The standard daily fresh-food dose (twigs stored in a glass in very low abundance, probably because of the change in of water) comprised one twig of spruce where one-year-old forest management towards more nature-friendly practices needles had grown on 25 cm of its total length and one twig excluding monoculture plantations [20,21]. As a herbivorous of young beech (<5 years old) wearing three fresh leaves, species, it can consume a wide variety of plants, but it each with an average length of 6 cm. Both the offered host mainly feeds upon the needles of the Norway spruce (Picea plant materials were collected from plant individuals and abies) and Scots pine (Pinus sylvestris) [20,21,29]. Eggs are twigs of the same age at one site daily during the time of hatched after 1–3 years (winter diapauses) depending on the experiment. Each female was housed separately in a the local climatic conditions and five or six nymphal instars glass container (volume 4 l) with air circulating through the precede adult moulting [21,23]). Adults that reach a body netting at the top (air temperature 22–26°C, relative humidity length of 14–25 mm are short-winged, hence flightless, ~50%, natural daylight). Along with the replacement of with an average adult lifetime of about two months [29]. food and fresh water, the containers that they lived in were This species has two colour phenotypes – dark or light cleaned each day from 06:00 to 06:30 CEST. We recorded green – with possible intermediate forms [20]. Though two parameters of foraging behaviour each day of the dark-green are the most typical, light-green individuals experiment: 1) the number of needles that had signs of prevail in broadleaved forests (see below) where specific being nibbled by bush-crickets according to the description environmental conditions can determine their body colour [29], and 2) the relative proportion of spruce droppings from [cf. 30]. all of the droppings that were defecated. The consumed needle area was noted (measured as % of needle length) 2.2 Experimental design in both bush-cricket populations. A needle was considered to be fully consumed when >90% of its length was missing For the lab experiment, we used bush-crickets hatched in and as nibbled when only tip of the needle was consumed natural conditions; thus the sample size was limited by (<10%). Less than 5% of beech leaves had traces of nibbling the rare occurrence (especially in broadleaved forests) (in positive cases a maximum 20% of the leaf surface), so and difficult field collection of the species [21]. Third we did not compare the damage to beech leaves between to fifth nymphal instars were collected from woody the two tested groups. All the droppings were dissected vegetation by sweeping hand nets from May 25 until under 30× magnification using a Motic SMZ-168 stereo June 8 (2010 and 2011) in Central and Northern Slovakia zoom microscope and the determination of their contents (48°29´–49°09´N, 19°01´–19°55´E). We sampled bush- was based on the structure of the plant tissue remains crickets in different source forest habitats regarding a [31]. These include observations of spruce having a light dominant tree species (i.e. >90% of cover in the canopy fibrous structure and beech showing green fragments of layer) and pooled individuals from several independent leaf nervation (Fig. 1). Before the experiment, the structure sites into two groups. Individuals from a ‘spruce forest’ of defecated remains was verified using independent Conditioned foraging of bush-crickets 3 individuals when only single plant species were provided; thus, it was possible to easily classify each dropping as either a spruce or beech dropping. This method allowed the unambiguous classification of droppings into these two categories and the proportion of some intermediate forms (mixed spruce and beech plant material in one dropping) was negligible.

2.3 Data analysis

The experiment was designed so that each individual was tested repeatedly for ten days. Therefore, we analysed data using a repeated-measures analysis of variance. We tested for the source habitat as a between-groups effect (spruce or beech forest) and for the day (day in the experiment) and source habitat × day (interaction term) as within- Figure 1: Determination of food consumed by the Barbitistes groups effects. The number of needles that had signs of constrictus females based on the structure of the plant tissue being nibbled and the proportion of spruce droppings remains; dissected droppings containing spruce needles (a) and beech leaf fragments (b). from all of the droppings that were defecated did not have normal distributions, thus we log-transformed data prior to analysis. We used Mauchly’s test to control for the sphericity assumption of univariate repeated-measures analysis. When the data did not meet that assumption, degrees of freedom and p-values were adjusted by using the Greenhouse-Geisser correction for effects with more than two levels. The total number of droppings per individual was compared between the two groups by non-parametric Mann-Whitney U test. Computations were performed in R 2.11.1 [32]. 3 Results

In the experiment we found that the source habitat had a highly significant effect on the number of needles nibbled by adult bush-cricket females (p < 0.001; Table 1, Fig. 3a). Individuals from the spruce forest mainly foraged only on needle tips (<10% of the needle length) and Figure 2: Spruce needles with tips nibbled by bush-crickets often tore off needles from the twig while eating (Fig. 2). originating from the spruce forest.

Table 1: Effects of the source habitat and the day with interaction term on the number of needles that had signs of being nibbled during the 10-day experiment (repeated-measures ANOVA) Effect df dfa MS F p pa

Between groups source habitat 1 15.8 16.25 <0.001 Error 43 0.9 Within groups day 9 5.7 0.1 0.53 0.850 0.698 source habitat × day 9 5.7 0.2 0.96 0.477 0.429 Error 387 245.4 0.3 aSince sphericity could not be assumed (Mauchly’s test, ε = 0.63, p < 0.05), degrees of freedom and p-values were adjusted by Greenhouse- Geisser correction. 4 P. Kaňuch et al.

In contrast to this, individuals from the beech forest had a effects of their hosts [24,33]. When a population is exposed different strategy, since they almost consumed the entire to such an effect on individuals’ fitness over generations, needle (>90% of the needle length). Hence, individuals a selection process can alter foraging behaviour [16,26]. from the spruce forest nibbled more needles per day In our case, the bush-crickets from the spruce forest than those from the beech forest (medians 106.0 vs. 42.5; adapted to consume only high-quality food, the needle Fig. 3a). The day of the experiment did not have an effect tips [24]. The observed variation in foraging behaviour on this pattern. Though they had different strategies when foraging on needles, both groups of bush-crickets avoided beech leaves as spruce droppings represented 93.6% (median) in individuals from the spruce forest and 90.2% in individuals from the beech forest (Fig. 3b). This composition of droppings did not change significantly during the experiment (Table 2). Nevertheless, the overall appetite of the individuals was very similar regardless of the source habitat. Individuals from both the spruce and beech forest produced almost the same number of droppings, 7644 and 7621 respectively, i.e. 34 (33–35) droppings a day (as a median and 95% CI; Z = 0.2, p = 0.842).

4 Discussion

Our experimental study showed that the foraging behaviour of B. constrictus individuals may differ with regard to the source habitat they come from [cf. 9,23]. Besides equal consumption of spruce mass (expressed by the production of droppings) in both tested groups, individuals from the spruce forest nibbled significantly more needles which was the result of a different foraging strategy. To our knowledge, this is the first report on conditioned feeding strategy coming with experience in needle eating insect. Specialised foraging on conifer Figure 3: (a) The number of nibbled needles per day in females that needles seems to be typical for all developmental stages originated from different source habitats and (b) relative proportion in our studied species and it is a very special diet among of spruce droppings from all droppings defecated per day (n = 23 orthopterans [22,23]. The consumption of both tested females from beech forest, 22 from spruce forest). Box-plots repre- plants is generally possible for specialised herbivores sent medians (lines), 95% confidence intervals (notches), quartiles only, which possess mechanisms to overcome the harmful (boxes) and non-outlier ranges (whiskers).

Table 2: Effects of the source habitat and the day with interaction term on the relative proportion of spruce droppings from all of the drop- pings that were defecated during the 10-day experiment (repeated-measures ANOVA) Effect df dfa MS F p pa

Between groups source habitat 1 1.1 1.21 0.277 Error 43 0.9 Within groups; day 9 6.3 0.1 0.22 0.992 0.942 source habitat × day 9 6.3 0.2 1.23 0.277 0.300 Error 387 272.4 0.3

aSince sphericity could not be assumed (Mauchly’s test, ε = 0.70, p < 0.05), degrees of freedom and p-values were adjusted by Greenhouse- Geisser correction. Conditioned foraging of bush-crickets 5

[5] Berner D., Blanckenhorn W.U., Korner C., Grasshoppers cope when the bush-crickets consumed the generally preferred with low host plant quality by compensatory feeding and food spruce [20,22,23] supports this possibility. selection: N limitation challenged, Oikos, 2005, 111, 525-533 Although our data set is limited due to methodological [6] Franzke A., Unsicker S.B., Specht J., Köhler G., Weisser W.W., constraints, we may suggest that differences in obtaining Being a generalist herbivore in a diverse world: how do diets nutrition from spruce needles have most likely evolved from different grasslands influence food plant selection and through natural selection [15,16]. Here we briefly fitness of the grasshopper Chorthippus parallelus? Ecol. Entomol., 2010, 35, 126-138 propose possible scenarios of adaptations that are rather [7] Behmer S.T., Joern A., Coexisting generalist herbivores occupy remarkable in insect evolution. unique nutritional feeding niches, P. Natl. Acad. Sci. USA, What needs to be considered first is that coniferous 2008, 105, 1977-1982 foliage has a characteristic increase in concentration [8] Pearson R.E.G., Behmer S.T., Gruner D.S., Denno R.F., Effects and amount of monoterpenes along the needle from tip of diet quality on performance and nutrient regulation in an omnivorous katydid, Ecol. Entomol., 2011, 36, 471-479 to base, which deters herbivore insects from feeding on [9] Bernays E.A., Bright K.L., Mechanisms of dietary mixing in entire needles [24]. Preference for needle tips may also grasshoppers: A review, Comp. Biochem. Phys. A, 1993, 104, be related to softer tissue or higher nutritional value 125-131 and can be demonstrated by individuals that originated [10] Behmer S.T., Insect herbivore nutrient regulation, Annu. Rev. from spruce forests and preferred needle tips. Thanks Entomol., 2009, 54, 165-187 to this adaptation, bush-crickets from spruce forests [11] Papaj D.R., Lewis A.C., Insect learning: Ecological and evolutionary perspectives, Chapman & Hall, New York, 1993 can effectively avoid harmful effects and maintain their [12] Papaj D.R., Prokopy R.J., Ecological and evolutionary aspects of fitness or breeding success better than individuals from learning in phytophagous insects, Annu. Rev. Entomol., 1989, beech forests [2,14,33]. This hypothesis is supported by 34, 315-350 a significantly higher abundance of studied species in [13] Cunningham J.P., Jallow M.F.A., Wright D.J., Zalucki M.P., coniferous than in broadleaved forests [20,21]. The other Learning in host selection in Helicoerpa armigera (Hübner) (Lepidoptera: Noctuidae), Anim. Behav., 1998, 55, 227-234 consideration involves individuals that hatched in a beech [14] Dukas R., Bernays E.A., Learning improves growth rate in forest and artificial surplus of spruce needles as a valuable grasshoppers, P. Natl. Acad. Sci. USA, 2000, 97, 2637-2640 food source could alter their behaviour [10]. Either that, [15] McGuire T.R., Hirsch J., Behavior-genetic analysis of Phormia or bush-crickets are possibly adapted to worse beech regina: conditioning, reliable individual differences, and (phenolic) conditions, this encourages the feeding of selection, P. Natl. Acad. Sci. USA, 1977, 74, 5193-5197 whole needles when available [23,25]. Both considerations [16] Mery F., Kawecki T.J., Experimental evolution of learning ability in fruit flies, P. Natl. Acad. Sci. USA, 2002, 99, 14274-14279 are quite plausible as conditioned behaviour appears to be [17] Cox G.W., Alien species and evolution: The evolutionary an adaptation mechanism when it comes to the foraging of ecology of exotic plants, , microbes, and interacting B. constrictus. However, multi-choice feeding experiments native species, Island Press, Washington DC, 2004 with different host plants may shed more light on the diet [18] Scriber J.M., Integrating ancient patterns and current dynamics selection process of this herbivorous insect. of insect-plant interactions: Taxonomic and geographic variation in herbivore specialization, Insect Sci., 2010, 17, 471-507 Acknowledgements: We would like to thank V. Badinková [19] Cunningham J.P., Zalucki M.P., West S.A., Learning in for her assistance during the experiments and P. Tuček and Helicoverpa armigera (Lepidoptera: Noctuidae): a new look at M. Mikuš for help in the field. This work was funded by the the behaviour and control of a polyphagous pest, B. Entomol. Scientific Grant Agency (VEGA 2/0157/11, 2/0035/13) and the Res., 1999, 89, 201-207 Slovak Research and Development Agency (APVV-0497-10). [20] Detzel P., Die Heuschrecken Baden-Württembergs, Ulmer, Stuttgart, 1998, (in German) [21] Holuša J., Heralt P., Drápela K., Occurrence and bionomy of Barbitistes constrictus (Orthoptera: Tettigoniidae) in the References eastern part of the Czech Republic, J. Forest Sci., 2006, 52, 61-73 [1] Dethier V.G., Evolution of feeding preferences in phytophagous [22] Ingrisch S., Köhler G., Die Heuschrecken Mitteleuropas, insects, Evolution, 1954, 8, 33-54 [2] Bernays E.A., Chapman R.F., Host-plant selection by Westarp Wissenschaften, Magdeburg, 1998, (in German) phytophagous insects, Chapman & Hall, New York, 1994 [23] Gottwald J., Richter C., Wörner M., Habitatwahl, Nahrungswahl [3] Singer M.C., Thomas C.D., Evolutionary responses of a butterfly und Entwicklung von B. serricauda (Fabricius, 1787) und B. metapopulation to human- and climate-caused environmental constrictus Brunner von Wattenwyl, 1878 (), variation, Am. Nat., 1996, 148, 9-39 Articulata, 2002, 17, 51-78, (in German) [4] Hanski I., Singer M.S., Extinction-colonization dynamics and [24] Litvak M.E., Monson R.K., Patterns of induced and constitutive host-plant choice in butterfly metapopulation, Am. Nat., 2001, monoterpene production in conifer needles in relation to insect 158, 341-353 herbivory, Oecologia, 1998, 114, 531-540 6 P. Kaňuch et al.

[25] Petrakis P.V., Spanos K., Feest A., Daskalakou E., Phenols in [30] Saglam I.K., Caglar S.S., Distribution and habitat characte- leaves and bark of Fagus sylvatica as determinants of insect ristics of the color polymorphic bush-cricket Isophya rizeensis occurrences, Int. J. Mol. Sci., 2011, 12, 2769-2782 Sevgili (Orthoptera: Tettigoniidae: Phaneropterinae) in Turkey, [26] Dukas R., Evolutionary biology of insect learning, Annu. Rev. Entomol. News, 2005, 116, 309-324 Entomol., 2011, 53, 145-160 [31] Gangwere S.K., Monograph on food selection in Orthoptera, T. [27] Zuna-Kratky T., Karner-Ranner E., Lederer E., Braun B., Berg Am. Entomol. Soc., 1961, 87: 67-230 H-M., Denner M., et al., Verbreitungsatlas der Heuschrecken [32] R Development Core Team, R: A language and environment for und Fangschrecken Ostösterreichs, Verlag Naturhistorisches Museum, Wien, 2009, (in German) statistical computing, R Foundation for Statistical Computing, [28] Krištín A., Kaňuch P. (eds), Distribution of Barbitistes Vienna, 2010 constrictus in Slovakia, http://www.orthoptera.sk, 2013 [33] Nykänen H., Koricheva J., Damage-induced changes in woody [29] Haber A., Opaślik sosnowiec Barbitistes constrictus Br. Watt. plants and their effects on insect herbivore performance: a (Locustidae Orth.), Państwowe wydawnictwo rolnicze i leśne, meta-analysis, Oikos, 2004, 104, 247-268 Warszawa, 1953, (in Polish)