Figures
Abstract
Herbivory activates the synthesis of allelochemicals that can mediate plant-plant interactions. There is an inverse relationship between the activity of xylophages and the abundance of epiphytes on Ipomoea murucoides. Xylophagy may modify the branch chemical constitution, which also affects the liberation of allelochemicals with defense and allelopathic properties. We evaluated the bark chemical content and the effect of extracts from branches subjected to treatments of exclusion, mechanical damage and the presence/absence of epiphytes, on the seed germination of the epiphyte Tillandsia recurvata. Principal component analysis showed that branches without any treatment separate from branches subjected to treatments; damaged and excluded branches had similar chemical content but we found no evidence to relate intentional damage with allelopathy; however 1-hexadecanol, a defense volatile compound correlated positively with principal component (PC) 1. The chemical constitution of branches subject to exclusion plus damage or plus epiphytes was similar among them. PC2 indicated that palmitic acid (allelopathic compound) and squalene, a triterpene that attracts herbivore enemies, correlated positively with the inhibition of seed germination of T. recurvata. Inhibition of seed germination of T. recurvata was mainly correlated with the increment of palmitic acid and this compound reached higher concentrations in excluded branches treatments. Then, it is likely that the allelopathic response of I. murucoides would increase to the damage (shade, load) that may be caused by a high load of epiphytes than to damage caused by the xylophages.
Citation: Flores-Palacios A, Corona-López AM, Rios MY, Aguilar-Guadarrama B, Toledo-Hernández VH, Rodríguez-López V, et al. (2015) Is Allelopathic Activity of Ipomoea murucoides Induced by Xylophage Damage? PLoS ONE 10(12): e0143529. https://doi.org/10.1371/journal.pone.0143529
Editor: Paul Adam, University of New South Wales, AUSTRALIA
Received: September 4, 2015; Accepted: November 5, 2015; Published: December 1, 2015
Copyright: © 2015 Flores-Palacios et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper.
Funding: This research was funded by a Postdoctoral PROMEP (Programa de Mejoramiento del Profesorado) grant to SVD (2011-2013) and by funding granted to the Cuerpo Académico Biología del Dosel. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Herbivory is a selection pressure that releases defense mechanisms in plants, such as the synthesis of allelochemicals in response to damage [1]. As a response to herbivory, plants can modify their phenotype to increase the concentrations of constitutive compounds (e.g. lignin, tannins) or synthesize new compounds that act directly or indirectly on the herbivore [2, 3, 4, 5]. Moreover, certain allelochemicals, or combinations of these synthesized in response to herbivory have allelopathic properties that inhibit seed germination in their competitors [6, 7]. This increases the complexity of the biological interactions and could contribute to the explanation of patterns of assemblage of species [8, 9].
Trees provide resources for a wide variety of species, hosting and facilitating the interaction of different guilds (e.g. xylophages and epiphytic plants) [10, 11]. The chemical constituents of the host species influence the organisms they support [12, 13, 14] and may vary intra-specifically and as a function of herbivory [1]. It is known that certain xylophages (Coleoptera: Cerambycidae) prefer species that are chemically similar [12] and that concentrations of tannins increase in the affected area of plants damaged by curculionid xylophages [15]. It is kwon also that bark of trees and epiphytic lichens may exert an allelopathic effect on epiphytes [16, 17]. For example, the seed germination of the epiphyte orchid Dendrobium aphyllum (Roxb.) C.E.C. Fisch is reduced by bark extracts of those trees in which the species rarely occur [17], and the host Ipomoea murucoides Roem. & Schult (Convolvulaceae) has an allelopathic effect on seed germination in the epiphytic bromeliad Tillandsia recurvata (L) L. [14].
There is evidence of a negative association between the presence of xylophages and species of epiphytic bromeliads on different host species [18]. This is most clearly seen in the Ipomoea murucoides-Tillandsia recurvata interaction, where this epiphyte is not abundant but the number and activity of xylophages is greater than in other host species [18]. The guild of xylophages is characterized by boring holes in their host tree while feeding on the conductive tissues [19]. It is likely that when the xylophages bore tunnels they promote the liberation of allelochemicals that act as allelopathics on the germination of T. recurvata seeds. It is also known that epiphytes can mechanically weaken their hosts [20] and it is likely that by themselves they trigger the synthesis of allelopathics in the hosts. The objective of this study was to determine the effect of intentional damage (simulating damage by xylophages) in branches of I. murucoides on the germination of T. recurvata seeds and distinguish this from the effect of epiphytes on the host. It is hypothesized that the activity of xylophages on the host, simulated by the intentional damage, induces the liberation of allelochemicals with allelopathic properties on the germination of seeds of this epiphyte.
Materials and Methods
Species and study area
The study area was located in San Andrés de la Cal, in Tepoztlan, in Morelos state, Central Mexico (18°57’22.2”W, 99°06’50.2”N, 1495 m asl). This area belongs to the people of San Andrés de la Cal, they are represented by the President of communal lands who gave us permission to carry out this survey. Mean annual precipitation is 1098 mm and mean temperature ranges from 12°C to 18°C [21]. The vegetation is tropical dry forest, characterized by an open canopy, with trees < 16 m height [22]. At the study area, the dominant tree species include Sapium macrocarpum Müll. Arg. (Euphorbiaceae), Bursera copallifera (DC.) Bullock, B. fagaroides (Kunth) Engl., B. glabrifolia (Kunth) Engl. (Burseraceae), Ipomoea murucoides (Convolvulaceae), Conzattia multiflora (B.L. Rob.) Standl. and Lysiloma acapulcense (Kunth) Benth (Fabaceae) [23].
Ipomoea murucoides is a deciduous tree that flowers from October to March and reaches approximately 8 m in height in the tropical dry forests of central and southern Mexico [24]. It has been reported that plants of this genus (e.g. I. tricolor) produce allelopathically active alkaloids and resin glycosides [25, 26]. In the case of I. murucoides, it contains allelopathics that inhibit the germination of the epiphyte T. recurvata through dichloromethane bark extract [14, 27]. On the other hand, it is known that extracts of diverse organs from different Ipomoea species have insecticidal effects on herbivores [28, 29].
At the study site, Tillandsia recurvata comprises 72% of the epiphytic individuals and is mainly distributed on hosts from the family Burseraceae, such as Bursera bipinnata (DC.) Engl., B. copallifera and B. glabrifolia. However, in others such as B. fagaroides, I. murucoides and I. pauciflora, this epiphyte is practically absent [23]. It has been reported that these host species may harbor 63.6% of the xylophagous insects present in the study area; these insects were mainly Coleoptera (41.8%) or those that cause boring damage to the bark such as Hymenoptera (Formicidae) [18]. Within the order Coleoptera, the family Cerambycidae was the most represented (67.4% of coleopterans), followed by Tenebrionidae (19.6%), Buprestidae (4.4%), Bostrichidae and Elateridae (both 2.2%) [18]. We state that the field studies did not involve endangered or protected species
Field experimentation and processing of plant material
In Nov. 2012, along a 1 km path in the “Cerro de la Cruz”, in San Andrés de la Cal, ten individuals of I. murucoides (height > 4m; mean DBH at 1.3 m = 21.7 cm ± 5.1 cm) were randomly selected. Using a 6 m ladder, eight branches were chosen at random from the exterior stratum of the tree (hereafter, all values are mean ± 1 SE, unless otherwise stated, diameter = 1.7 cm ± 0.05 cm, and length = 1.2 m ± 0.5 m). The branches chosen all appeared healthy (with leaves or flowers, no dry areas, and green bark) and had no epiphytes or apparent damage in the bark.
In order to determine whether the association between the xylophages and epiphytes on I. murucoides is mediated by allelochemicals liberated when the xylophages pierce the branches of the host, each treatment was assigned at random to each of the eight selected branches on each trees of I. murucoides. The factors combined to obtain the eight branch treatments were: exclusion, intentional mechanical damage and the presence of T. recurvata. Exclusion (excluded vs. non-excluded branches) consisted of covering the branch with a mesh bag in order to avoid oviposition and decrease the activity of xylophages over the course of the experiment. Intentional damage to the branches (damaged vs. non-damaged branches) attempted to mechanically simulate the damage caused by xylophages and thereby weaken the branches in order to promote a higher xylophage activity. The intentional damage consisted of making 2–3 mm wide groove, lifting the bark along 25% of the branch length. The groove was made with the hook of a diametric measuring tape (Forestry Suppliers Metric Fabric Diameter Tape). The third factor was the presence/absence of T. recurvata. This was to determine whether epiphytes could independently activate allelopathic mechanisms in the host. In the treatment of branches with epiphytes, four individuals of T. recurvata, collected from hosts of the same 1 km transect, were transplanted. The bromeliads were adults of more than 10 cm in diameter and were fixed at equal distances apart on the branch with plastic straps. Combination of these factors produced the following eight treatments: 1) excluded branches, with T. recurvata and with damage, 2) excluded branches, with T. recurvata and without damage, 3) excluded branches, without T. recurvata and with damage, 4) excluded branches, without T. recurvata and without damage, 5) non-excluded branches, with T. recurvata and with damage, 6) non-excluded branches, with T. recurvata and without damage, 7) non-excluded branches, without T. recurvata and without damage, and finally 8) non-excluded branches, without T. recurvata and without damage.
In order to avoid leaching of allelochemicals due to the rainy season (Oct-May), the branches were cut from the trees in May 2013 and transported to the ecology laboratory of Centro de Investigación en Biodiversidad y Conservación-Universidad Autónoma del Estado de Morelos (CIByC-UAEM) for processing. The branches were cleaned of all external agents (e.g. lichens) and before stripping the bark, the following data were recorded: number of orifices evidently caused by xylophages (hereafter orifices), total branch length (cm, length of main branch plus length of secondary branches), diameter of main branch base (mm) and non-intentional damage (cm) (i.e. pooled length measurements of bark scars, girdled branches or orifices). The percentage of non-intentional damage was obtained by dividing the non-intentional damage value by the total length of the branch and multiplying by 100.
Also recorded were the number and identity of the larvae and imagoes of insect xylophages found in the branches of the eight treatments. These were preserved in 60% ethanol and identified to taxonomic group or, when possible, to species. Larval identification was conducted at the Insect Collection of the Universidad de Morelos (CIUM) at the CIByC-UAEM. Keys used for identification were those used by Costa and collaborators [30] and Chu [31].
The bark was removed from the branches and dried in a fan oven (FD 115-UL, Binder) at 30°C until reaching a constant dry weight. The dry bark was ground to < 3 mm in an electric mill (PULVEX S. A. de C.V. model Mini-100). Disregarding the identity of the tree, the resulting dry bark powder samples were pooled according to branch treatment and stored in darkness at -15° C until use. The dry bark material from the eight treatments was subjected to extraction by maceration at room temperature with dichloromethane. The dichloromethane extract from I. murucoides was chosen because it has been shown to produce the highest inhibition (58%) of seed germination in T. recurvata [14, 27]. Following the protocol used by Valencia-Díaz and collaborators [14] to obtain dichloromethane extracts, fats were first removed from the sample by maceration with hexane (three periods of 72 h each) at ambient temperature and then the dichloromethane extracts were obtained by repeating the protocol used with the hexane. Hexane and dichloromethane solubilized compounds that are water insoluble or that have low water solubility (i.e. terpenoids) [32]. Following filtration, extracts were concentrated under vacuum (Rota-evaporator Buchi R-200) at 39°C for hexane and dichloromethane. Dry extracts were stored at -15°C until use; however, only the dichloromethane extracts were used in the germination experiments.
Seed collection and germination tests
The germination experiments were carried out with T. recurvata seeds collected in April 2013. We collected T. recurvata seed capsules from 18 individual plants. Seeds were removed from the capsules once dehiscence had taken place. We removed the coma hairs from the seeds to facilitate manipulation and reduce the risk of contamination during the experiments. Tillandsia recurvata seeds were joined in a seed batch; then they were randomized and sterilized prior to sowing using 0.25% sodium hypochlorite solution, with subsequent rinsing three to four times in sterile distilled water to completely remove remnants of the sodium hypochlorite.
We hypothesized that extracts obtained from the artificially damaged branches of I. murucoides would have allelopathic properties on T. recurvata seed germination. To test this hypothesis, we applied dichloromethane extract obtained from each of the eight treatments at its most inhibitory concentration [1 μg/mL of distilled water; 14]. Ten seeds of T. recurvata were placed in a Petri dish using filter paper as substrate. Twenty-four Petri dishes were used for each treatment along with four Petri dishes with distilled water but without extract (control). In order to exclude the effects of factors other than allelopathy (e.g. non-viable seeds or dormancy) from the calculation, percentage of inhibition of seed germination was calculated relative to that recorded in the control. Percentage inhibition of seed germination was therefore calculated according to the following equation:
Where,
is the mean number of germinated seeds in the control, Xi is the number of germinated seeds under a specific concentration and n is the number of seeds (n = 10) in the Petri dish. Positive and negative values indicate inhibition and promotion of germination, respectively, while zero indicates a complete absence of bark host effect.
As with the dichloromethane extracts, dimethyl sulfoxide (DMSO: (CH3)2SO, Fermont) was used, at a maximum concentration of 1% (v/v), in order to prepare the solutions [14]. In all germination tests, petri dishes were placed within an environmental growth chamber (Scorpion Scientific A 50624, Mexico) for 10 days under a photoperiod regime of 12 h light/ 12 h dark and a constant temperature of 30°C. Since bromeliad seeds do not produce a radicle during the early germination process, seeds were considered to have germinated upon rupture of the testa [33].
Chemical analysis
In order to know the chemical constituents among extracts obtained from the branch treatments 21 mg of each of the eight samples (one sample per branch treatment), were solubilized in ethyl acetate (1 mL). Solid phase extraction was done using 1 mL cartridges SUPELCO® with a normal phase (LC-SI). 1 mL of ethyl acetate was recovered from the sample loaded in the cartridge. Just 400 μL were evaporated. Six mL of ethyl acetate was recovered from washing cartridge. Solvent was evaporated. The eight samples were reconstituted in dichloromethane (1 mL) and injected in GC-MS. Aliquots (150 μL) from these solutions were injected into a gas chromatograph (Agilent 6890) coupled to a mass detector [Agilent 5379 series N, 30–550 atomic mass units, electron impact ionization (70 eV)], equipped with an HP5-MS capillary column (25 m × 0.20 mm × 0.33 M film thickness, 5% phenylmethyl siloxane stationary phase). Helium was the carrier gas at a flow rate of 1 mL min−1. All extracts were analyzed using the following temperature program: 1–45°C 1 min, with an increase of 10°C min−1 until reaching 250°C and then 260°C was maintained for 30 min. The database used for comparison of the results was NIST-MS version 1.7a.
Data analyses
The variables total length of branch, diameter, percentage of non-intentional damage, number of orifices and number of xylophages were analyzed using an analysis of variance with three factors [34], where the main factors and their levels were: exclusion (excluded and non-excluded branches), intentional damage (damaged and undamaged branches) and epiphytes (presence and absence). Differences among factors and interactions were analyzed with the Tukey multiple comparisons test [34]. Due to the nature of the data, and to comply with the assumptions of normality and homogeneity of variance, the variable percentage of non-intentional damage was transformed using an arcsin function, and the variables number of orifices and number of xylophages were square root transformed [34].
The percentage of inhibition of germination of T. recurvata seeds caused by the dichloromethane extracts was analyzed with an analysis of variance of three factors with two levels per factor. To normalize the data and make the variances uniform, the data were transformed using the arcsin function [34]. The factors were exclusion (excluded and non-excluded branches), intentional damage (branches with and without damage) and epiphyte presence/absence. Differences among factors and interactions were analyzed with the Tukey multiple comparisons test [34]. All analysis was performed using the program Stata 13 [35].
We hypothesized that bark extracts would form associations according to their chemical constituents and that those extracts obtained from branches under intentional damage would have allelopathic properties on T. recurvata seed germination. In order to test this hypothesis we performed a principal components analysis (PCA) [36] using as variables the chemical compound concentration (%) within the extracts of each treatment and the mean percentage of inhibition of seed germination of T. recurvata exerted by these extracts. PCA analysis was based on the correlation matrix, for each eigenvector (principal component), we considered as significant, those variables that contribute with at least 10% to the eigenvector normalized length and that at the same time were highly correlated with it (r ≥ 0.6) [36].
Results
In three of the ten trees in which the experiment was conducted, mesh bags used to achieve exclusion were lost, for this reason only the data derived from the branches (n = 56) of seven trees were considered. General measurements of branches were: length of 188.62 cm ± 13.60 cm, diameter of 17.24 mm ± 0.53 mm, 0.82 ± 0.23 orifices, 1.28 ± 0.30 xylophages and non-intentional damage of 8.30 cm ± 1.21 cm. There were no significant differences of length of branches, diameter, number of xylophages, number of orifices and natural damage between levels of the main factors (Table 1). Furthermore, the treatment interactions did not reveal statistical differences for any measured variable (all F1,48 < 2.3, all P > 0.05).
No single variable showed statistical differences between factor levels (P > 0.05). F 1,48 values of the variables within each factor were in italics and in parenthesis.
A total of 85 arthropods were found, of which 83 were larvae and two were adults one belonged to Brachymyrmex sp. (Hymenoptera: Formicidae) and the other was a bark-bettle (Curculionidae: Scolytinae). Due to the fact that the majority of individuals found were larvae, identification was limited. Of the total number of arthropods, the majority (83.52%) belonged to the order Coleoptera, followed by Lepidoptera (12.94%), Hemiptera (1.17%), Hymenoptera (1.17%) and by the class Arachnida (1.17%). Four families of the order Coleoptera were identified, with Cerambicidae accounting for most individuals (92.95%), followed by Buprestidae (4.22%), Curculionidae (1.40%) and Tenebrionidae (1.40%). The analysis only included the insect xylophages; in this case, Coleoptera.
Germination tests
The percentage of inhibition of T. recurvata seeds was different in the exclusion treatments (F1,184 = 86.30, P < 0.001), with lower germination recorded on application of the extract derived from the excluded (51.77% ± 1.68% inhibition) than non-excluded (30.31% ± 1.59% inhibition; Tukey < 0.05) branches. There were no differences (F1,184 = 1.60, P = 0.20) in the inhibition of T. recurvata seed germination on application of extract derived from branches with intentional damage (39.58% ± 1.98% inhibition) and from those without intentional damage (42.50% ± 1.95% inhibition; Tukey > 0.05). The percentage of inhibition of T. recurvata seed germination was equal (F1,184 = 0.60, P = 0.43) when tested with extracts derived from branches with (40.31% ± 2.09%) and without (41.80% ± 1.83%; Tukey > 0.05) epiphytes.
Regarding the interaction between each pair of main factors, the T. recurvata seeds were equally inhibited by the dichloromethane extract obtained from excluded branches and from those with damage (F1,184 = 1. 47, P = 0.0006; Table 2). There was an effect of the interaction between exclusion and presence/absence of epiphytes (F1,184 = 5.03, P = 0.02) on the germination of T. recurvata seeds, with greater inhibition found in the seeds treated with extracts derived from excluded branches, regardless of the presence/absence of epiphytes (Tukey < 0.05; Table 2). The interaction between intentional damage and the presence/absence of epiphytes also affected the percentage of inhibition of germination in T. recurvata seeds (F1,184 = 11.9, P = 0.0006); however, multiple comparisons testing did not establish significant differences between the percentages of inhibition of germination, for which reason this result may be considered spurious. None of the interactions among the three main factors on inhibition of T. recurvata seed germination were statistically significant (F1,184 = 0.26, P = 0.61; Table 3).
The data refer the interaction between pairs of the main factors of the experiment: exclusion (excluded vs. non-excluded branches), intentional damage (branches with intentional damage vs. branches without intentional damage) and with the presence/absence of T. recurvata (N = 48). Within each section of the table, different letters indicate statistical differences between the interaction of the pairs of treatments (Tukey test P < 0.05).
The data shown correspond to the triple interaction of the three main factors of the experiment: exclusion (excluded vs. non-excluded branches), intentional damage (branches with intentional damage vs. branches without intentional damage) and with the presence/absence of T. recurvata (N = 24).
Chemical content
GC-MS of the eight extracts identified 17 different chemical compounds (Table 4, Fig 1), one of them, araucarolone, reported here for second time [37]. The first two principal components (PC) explained 65.7% of the variance. PC1 explained 43.1% of the variance, within this component three chemical compounds contributed with 33.6% of the total length of the eigenvector and were highly correlated with it (r ≥ |0.9|) (Table 4). PC2 explained 22.6% of the variance, and in this component five chemical compound along with inhibition of seed germination of T. recurvata (r ≥ |0.65|) contributed with 82% of the total length of the eigenvector (Table 4). Based on their chemical composition, first component separated extracts obtained from branches subject to exclusion or damage respect to control branches; while branches whose treatments were to have only epiphytes were in the middle of the previous groups (Fig 1). Second component separated extracts obtained from branches with a unique treatment (excluded or damaged or with epiphytes, at bottom of the y-axis in Fig 1) and extracts obtained from branches without exclusion but subject to damage and epiphytes, respect to the control and excluded branches with damage and/or epiphytes (Fig 1).
Branches excluded (Ex), damaged branches (Da), Presence of epiphytes (Ep), control means those branches without any treatment (000).
The analyzed variables were the inhibition of seed germination of T. recurvata and 17 chemical compounds identified with GC-MS in the bark of Ipomoea murucoides. In bold those chemical compounds whose percentage with contribution ≥ 10% to the eigenvalue length.
Discussion
Previous evidence shows that dichloromethane bark extract of I. murucoides exerts an allelopathic effect on the epiphyte T. recurvata seed germination by reducing about 58% [14]. We found that the dichloromethane extracts obtained from the eight treatments always inhibited the seed germination of this epiphyte. However, the highest percentages of inhibition occurred when extracts obtained from excluded branches were applied, reaching 60% inhibition (Table 3). In the study of Valencia-Díaz and collaborators [14], the extracts were obtained from the bark of branches that had not been subjected to any treatment of exclusion or damage. If we compare exclusively the percentages of inhibition (30.41%; Table 3) produced with extracts from branches with similar conditions (non-excluded branches, not intentionally damaged and with or without epiphytes) to the percentage of inhibition produced by the dichloromethane extract of I. murucoides reported in the previous study, we find a significant lower inhibition (Mann-Whitney; U = 7, P = 0.002). While the extracts obtained are from individuals of I. murucoides from the same study zone, it is probable that the differences found are due to the temporal variability of the climate, to the availability of resources that affect the composition and concentration of secondary metabolites [1] in branches, or to intrinsic differences (v. gr. seed viability) in the sets of T. recurvata seeds from different years.
As stated previously, herbivory induces the synthesis of allelochemicals that can perform multiple biological functions [6, 7] and then mediate multispecies interactions. Specifically, in an epiphytic system, xylophagy allows the liberation of defense chemical compounds [38] that may indirectly influence the epiphytes that the host may support. At the other hand, since xylophages can often be found in weakened or diseased tissues in the host [39, 40], epiphytic plants weaken the host and provoke the arrival of xylophages. We found that the Cerambicidae represented the most numerous xylophagous group, but did not find direct evidence that epiphytes or intentional damage to the branches would provoke the arrival of xylophages to the host since, statistically, all of the branches had the same number of xylophages, the same percentage of non-intentional damage and the same epiphyte cover. It is important to mention that our results are based on the data obtained at the end of the experiment, since in order to determine the initial percentage of non-intentional damage and number of xylophages in the branches, it is necessary to destroy the branches prior the execution of the field experiment. However in order to have a reference, we found statistical differences (Mann-Whitney; U = 6.00, P = 0.018) between the weighted means of a) xylophages/length (cm) of branches of this study (1.75 ± 0.56 xilophages/cm) and b) data of Valencia-Díaz et al. [18] (0.006 ± 0.002 xilophages/cm), where branches were not subject to any treatment (exclusion, damage or epiphyes). This may suggest that xylophages are more abundant in branches subject to any of the treatments performed here. Although both studies were performed in the same population of I. murucoides, this comparison must be considered with caution because of the temporal variability of xylophages populations.
We did not find evidence that extracts obtained from intentionally damaged branches provoked a greater allelopathic effect in T. recurvata seed germination, but according to the PCA chemical content analysis, those extracts with intentional damage contain more trans-9-hexadecen-1-ol, cis-9-eicosen-1-ol and 1-hexadecanol; of these it has been shown that 1-hexadecanol may increase their concentration due to plant injury [41]. When herbivores feed they liberate specific enzymes that activate the host defense mechanisms [1, 2, 42, 43, 44]. The fact that there is no direct relationship between the damage and increased inhibition of germination in T. recurvata does not preclude the activation of direct defenses against the herbivore. It is possible that the chemicals constituents of the branches that suffered damage can have a direct detrimental effect on the xylophages, but confirmation of this would require studying the insecticidal activity of extracts of I. murucoides on this guild.
Mesh bag exclusion was done in order to prevent subsequent arrival of xylophages to the experimental branches, but contrary to expectation, the extracts from excluded branches had a greater inhibitory effect on the germination of T. recurvata seeds. Two of the five chemical compounds correlated positively with the inhibition of seed germination of T. recurvata were hexadecanoic acid (palmitic acid) and squalene. Palmitic acid has been reported as a constituent of I. murucoides and as an inhibitor of seed germination of T. recurvata, also it has been confirmed that commercial compound inhibited 65% the seed germination of this epiphyte [14]. This compound also exerts allelopathic negative effects in parasitic weeds [45]. In the case of squalene, it has been reported that herbivory increases its content in leaves of Sebastiana adenophora (Euphorbiaceae) [46], and may act as a sinomone attracting herbivore enemies [47]. It is possible that the weight or the shadow created by the mesh used for exclusion treatments increase the squalene and palmitic acid content.
On the other hand, the treatment of presence/absence of epiphytes did not show differences in terms of the inhibitory effect of the extracts obtained from these two treatments. It is reasonable to consider that the branches with epiphytes would be more shaded than those without epiphytes and thus the chemical constitution could change as occurred in the exclusion treatment; however, it is possible that the coverage provided by the four epiphytes that were transplanted on the branches of this treatment was insufficient to bring about a change in the chemical content of the bark of these branches; in fact PC1 separated control branches along with branches with only epiphytes.
The only interaction of the treatments in which there were differences in the inhibition of seed germination in T. recurvata was the interaction between exclusion and presence of epiphytes. However, as can be seen from the results, this was determined basically by the treatment of exclusion; i.e., regardless of the presence of T. recurvata, the greatest inhibition of germination was produced when exclusion of the branches was included in the interaction. In fact, no interaction presented among the eight treatments of this experimental design exercised an inhibitory effect that was distinctively greater than the others. The fact that no statistical differences were found in the inhibition of the germination of T. recurvata seeds as a function of the treatments does not mean that there may not have been inhibitory effects, but rather that there is no relationship between this effect and the damage while there is a relationship with the exclusion and presence of epiphytes. As stated previously, all of the treatments inhibited germination, a result that corroborates the allelopathic effect of I. murucoides [14, 27].
At the interspecific level, there is evidence that in host species with a low abundance of epiphytes (Ipomoea murucoides, I. pauciflora and Bursera fagaroides) there is a negative association between the xylophages (damage and abundance) and the epiphytes (coverage and abundance) they host. This pattern is most evident in I. murucoides [18]. We found no evidence to suggest that this multispecies association is mediated chemically; however, we did find evidence that showed a different chemical content in extracts obtained from excluded branches and this may be responsible for the increment of allelopatic activity. If the mesh of excluded branches would seem as a high load of epiphytes, then it is possible that the shade or load generated by epiphytes change also the chemical content and consequently causes more inhibition of the germination of seeds of T. recurvata.
Acknowledgments
We acknowledge the assistance of A. Mariel, D. López-Cobos, E. Sanvicente and K. Columba in laboratory and fieldwork. We thank the Departamento de Ecología Evolutiva (CIByC-UAEM) and Laboratorio de Productos Naturales (Facultad de Farmacia) for their support. We thank C.A Vergara-Torres for the ant identification. We thank K. MacMillan for translation and comments. We appreciate the comments of Dr. Mick Hanley of a previous version of this manuscript. This research was funded by a Postdoctoral PROMEP grant to S.V.D. and by funding granted to the Cuerpo Académico Biología del Dosel.
Author Contributions
Conceived and designed the experiments: SVD AFP. Performed the experiments: AFP SVD MYR BAG. Analyzed the data: AFP SVD AMCL VHTH. Contributed reagents/materials/analysis tools: AFP SVD VRL MYR. Wrote the paper: AFP SVD.
References
- 1.
Schoonhoven LM, von Loon JJA, Dicke M. Insect-Plant Biology. 2nd ed. Great Britain: Oxford University Press; 2005.
- 2. Kessler A, Baldwin IT. Defensive function of herbivore-induced plant volatile emissions in nature. Science. 2001; 291 (5511):2141–43. pmid:11251117.
- 3. Vivanco JM, Cosío E, Loyola-Vargas VM, Flores HE. Mecanismos químicos de defensa de las plantas. Investigación y Ciencia. 2005; 341: 68–75.
- 4. Turlings TCJ, Ton J. Exploiting scents of distress: the prospect of manipulating herbivore-induced plant odors to enhance the control of agricultural pests. Curr Opin Plant Biol. 2006; 9(4):421–27. pmid:16723271.
- 5. Heil M. Indirect defence via tritrophic interactions. New Phytol. 2008; 178(1): 41–61. pmid:18086230.
- 6. Thelen GC, Vivanco JM, Newingham B, Good W, Bais HP, Landres P. et al. Insect herbivory stimulates allelopathic exudation by an invasive plant and the suppression of natives. Ecol Lett. 2005; 8(2):209–17.
- 7. Karban R. Experimental clipping of sagebrush inhibits seed germination of neighbours. Ecol Lett. 2007; 10(9):791–97. pmid:17663712.
- 8. Strauss SY, Irwin RE. Ecological and evolutionary consequences of multispecies plant-animal interactions. Ann Rev Ecol Evol Syst. 2004; 35:435–466.
- 9. Inderjit , Wardle DA, Karban R, Callaway RM. The ecosystem and evolutionary contexts of allelopathy. Trends Ecol Evol. 2011; 26(12):655‒62. pmid:21920626.
- 10. Nadkarni NM. Diversity of species and interactions in the upper tree canopy of forest ecosystems. Am Zool. 1994; 34(1):70–78.
- 11.
Stunt S. The influence of epiphytes on arthropods in the tropical forest canopy. Ph. D. Theses. Bayerischen Julius-Maximilians-Universität Würzburg. 2001.
- 12. Meurer-Grimes B, Tavakilian G. Chemistry of Cerambycid host plants. Part I.: Survey of Leguminosae- A study in adaptive radiation. Bot Rev. 1997; 63(4):356–94.
- 13. Ode PJ, Berenbaum MR, Zangerl AR, Hardy ICW. Host plant, host chemistry and the polyembryonic parasitoid Copidosoma sosares: indirect effects in a tritrophic interaction. Oikos. 2004; 104(2):388–400.
- 14. Valencia-Díaz S, Flores-Palacios A, Rodríguez-López V, Ventura-Zapata E, Jiménez-Aparicio AR. Effect of host-bark extracts on seed germination in Tillandsia recurvata, an epiphytic bromeliad. J Trop Ecol. 2010; 26(6):571–81.
- 15. Burgos-Solorio A, Terrazas T. El daño de Chaetophloeus mexicanus (Curculionidae: Scolytinae) a tallos de Eysenhardtia polystachya (Fabaceae). Madera y Bosques. 2010; 16 (4):67–79.
- 16. Callaway RM, Reihnart KO, Moore GW, Moore DJ, Pennings SC. Epiphyte host preferences and host traits: mechanisms for species-specific interactions. Oecologia. 2002; 132: 221–230. http://dx.doi.org/10.1007%2Fs00442-002-0943-3.
- 17. Harshani HBC, Senanayake SP, Sandamali H. Host tree specificity and seed germination of Dendrobium aphyllum (Roxb.) C.E.C. Fisch. in Sri Lanka. J. Nat. Sci. Foundation Sri Lanka. 2013; 41(3): 71–86.
- 18. Valencia-Díaz S, Corona-López AM, Toledo-Hernández VH, and Flores-Palacios A. Is branch damage by xylophages related to the presence of epiphytes? Arthropod-Plant Inte. 2014; 8(8):25‒32.
- 19. Suárez LH, Aguilar AV, González WL. Patrones de infestación por insectos xilofágos en renovales de Nothofagus obliqua Mirb. y Notofagus dombeyi (Mirb.) Oerst. (Fagales: Nothofagaceae). Rev Chil Hist Nat. 2005; 78(1):23–32.
- 20. Montaña C, Dirzo R, Flores A. Structural parasitism of an epiphytic bromeliad upon Cercidium praecox in an intertropical semiarid ecosystem. Biotropica. 1997; 29(4):517–21.
- 21.
Ruiz-Rivera CA. San Andrés de la Cal: Culto a los señores del tiempo en rituales agrarios. 1st ed. México: Universidad Autónoma del Estado de Morelos; 2001.
- 22.
Rzedowski J. La vegetación de México. 1st ed. México: Limusa; 1978.
- 23. Vergara-Torres CA, Pacheco-Álvarez MC, Flores-Palacios A. Host preference and host limitation of vascular epiphytes in a tropical dry forest of central Mexico. J Trop Ecol. 2010; 26(6):563–70.
- 24. Carranza E. Familia Convolvulaceae I. Flora del Bajío y de Regiones Adyacentes. 2007; Fascículo 151:1‒129.
- 25. Anaya LAL, Calera MR, Mata R, Pereda-Miranda R. Allelopathic potential of compounds isolated from Ipomoea tricolor Cav. (Convolvulaceae). J Chem Ecol. 1990; 16 (7):2145‒52. pmid:24264082
- 26. Pereda-Miranda R, Mata R, Anaya LAL, Wickramaratne DBM, Pezzuto JM, Kinghorn AD. Tricolorin A, major phytogrowth inhibitor from Ipomoea tricolor. J Nat Prod. 1993; 56(4):571–82. pmid:8496705.
- 27. Valencia-Díaz S, Flores-Palacios A, Rodríguez-López V, Jiménez-Aparicio AR. Inhibitory effects of bark chemicals of host Ipomoea murucoides on seed germination of epiphyte Tillandsia recurvata. Allelopathy J. 2013; 32(1):91‒100.
- 28. Hebling M, Bueno O, Maroti P, Pagnocca F, Silva O. Effects of leaves of Ipomoea batatas (Convolvulaceae) on nest development and on respiratory metabolism of leaf cutting ants Atta sexdens L. (Hym. Formycidae). J Appl Entomol. 2000; 124:249‒52.
- 29. Vera CLG, Hernández VVM, León RI, Guevara FP, Aranda EE. Biological activity of methanolic extracts of Ipomoea murucoides Roem et Schult of Spodoptera frugiperda J.E. Smith. J Entomol. 2009; 6(2):109–116.
- 30.
Costa CSA, Vanin SA, Casari-Chen S. Larvas de Coleoptera do Brasil. 1st ed. Brasil: Museo de Zoología, Universidade de Sao Paulo; 1988.
- 31.
Chu HF. How to know the immature insects. 1st ed. USA: Pictured key nature series; 1949.
- 32.
Pavia DL, Lampman GM, Kriz GS. Introduction to organic laboratory techniques. A small scale approach. Belmont, USA: Thomson Bookscole; 2005.
- 33. García-Suárez MD, Rico-Gray V, Molina-Aceves N, Serrano H. In-vitro germination and clonal propagation of endemic Tillandsia califanii Rauch (Bromeliaceae) from Mexico. Selbyana. 2006; 27(1):54‒9.
- 34.
Zar JH. Biostatistical Analysis. 5ft ed. New Jersey, USA: Prentice Hall, Upper Saddle River; 2010.
- 35.
StataCorp. 2013. Stata Statistical Software: Release 13. College Station, TX: StataCorp LP; 2013.
- 36.
Johnson RA, Wichern DW. Applied Multivariate Statistical Analysis. 5th ed. New Jersey, USA: Prentice Hall, Upper Saddle River; 2002.
- 37. Enzell CR, Thomas BR. The chemistry of the order Araucariales. 3. Structure and configuration of araucarolone and some related compounds from Agathis australis. Acta Chem Scand. 1965; 19 (8):1875‒96. 1965.
- 38. Becerra JX, Venable DL, Evans PH, Bowser WS. Interactions between chemical and mechanical defenses in the plant genus Bursera and their implications for herbivores. Am Zool. 2001; 41(4):865‒76.
- 39.
Corona-López MA, Toledo-Hernández VH. Patrones de distribución de la familia Bupestridae (Coleoptera). In: Morrone JJ, Llorente Bousquets J. Editors. Componentes Bióticos Principales de la Entomofauna Mexicana. México D.F.: Las Prensas de Ciencias UNAM; 2006.
- 40.
Toledo-Hernández VH, Corona-López AM. Patrones de distribución de la familia Cerambycidae (Coleoptera). In: Morrone JJ, Llorente Bousquets J. Editors. Componentes Bióticos Principales de la Entomofauna Mexicana. México D.F.: Las Prensas de Ciencias UNAM; 2006.
- 41. Sarkar N, Mukherjee A, Barik A. Attraction of Epilachnia dodecastigma (Coleoptera: Coccinellidae) to Momordica charantia (Cucurbitaceae) leaf volatiles. Can Entomol. 2015; 147(2):1–12.
- 42. Bennett RN, Wallsgrove RM. Secondary metabolites in plant defence mechanisms. New Phytol. 1994; 127(4):617‒33.
- 43. Kessler A, Halitschke R. Specificity and complexity: the impact of herbivore-induced plant responses on arthropod community structure. Curr Opin Plant Biol. 2007; 10(4):409–14. pmid:17652009.
- 44. Sepúlveda-Jiménez G, Porta DH, Rocha SM. La participación de los metabolitos secundarios en la defensa de las plantas. Rev Mex Fitopatol. 2003; 21(3):355‒363.
- 45.
Qasem RJ. Parasitic weeds and allelopathy: From the hypothesis to the proof. In: Reigosa MJ, Pedrol N, González L. Editors. Allelopathy: A Physiological Process with Ecological Implications. Dordrecht: Springer; 2006.
- 46. Macias-Rubalcava ML, Hernández-Bautista BE, Jiménez-Estrada M, Cruz-Ortega FR, Anaya AL. Pentacyclic triterpenes with selective bioactivity from Sebastiana adenophora leaves, Euphorbiaceae. J Chem Ecol. 2007; 33(1):147–56. pmid:17151909.
- 47. Dutton A, Mattiacci L, Amado R, Dorn S. A novel function of the triterpene squalene in a tritrophic system. J Chem Ecol. 2002; 28(1):103‒16. pmid:11868668.