Introduction

A number of factors, both in the course of initial acceptance by the pest and after colonization of the plant, determine the plant’s resistance to pests. This resistance is influenced by both the features of the plant’s morphological structure and the chemical composition of the penetrated plant tissues. Secondary plant metabolites, which act both as the plant’s choice of resistance and during the physiological relationship between pest and plant, are considered to be the most important source of cultivar resistance to pathogenic factors and pests. These factors regulate the behaviour of herbivores and affect the phenomenon of natural resistance (Harborne 1997; Gatehouse 2002; Van Den Boom et al. 2004). Essential oils (EO) are an important group of secondary metabolites which have an allelopatic influence on phytophagous species (Chamberlain et al. 2000; Holopainen 2004; Ibrahim et al. 2001). They occur commonly in the plant world where they perform a significant function in the interaction between plants and herbivores. Mono- and sesquiterpenes are aromatic stimuli that can be detected by insects even at considerable distances from the host plants. One of these, limonene, a hydrocarbon oil released by conifers, is largely considered to function as an attractant for herbivorous insects of conifers (Ibrahim et al. 2001). Chapman et al. (1981) showed that the menoterpene carvone is an attractant for the aphid Cavariella aegopodii which feeds on parsnip, whereas limonene and α-pinen attract the beetle Ips pini (Wallin and Raffa 2000). Certain monoterpenes are known as strong toxins against beetles feeding in the wood of coniferous trees (Cook and Hain 1988) and as deterrents of feeding and egg-laying (Leal et al. 1997). Sesquiterpenes are mainly deterrents of insect feeding, especially feeding by butterfly species (Qiu et al. 1998). Some of these are toxic; for example, zingiberen exudated by Lycopersicon hirsutum is toxic for the larvae of the Colorado potato beetle (Carter et al. 1989).

The aim of our study was to determine the effect of the aromatic components of EO occurring in the buds and leaves of two cultivars of hazel (Corylus L.) on the degree of acceptance during the initial choice by Phytoptus avellanae Nal., an acarine gall-mite species (filbert big bud mite), and Myzocallis coryli Goetze, commonly known as the filbert aphid.

Materials and methods

Plant and insect material

Determination of the degree of colonization of leaf buds by P. avellanae and of leaves by M. coryli was conducted between 2004 and 2006 under field conditions on a production plantation (chemically protected against Monilia coryli Schellenb. and Curculio nucum L.) located in Końskowola (eastern Poland, 51°40′ N, 22°06′ E). An additional field location for the filbert aphid was the experimental plantation in Motycz near Lublin (eastern Poland, 51° 14′ N 22° 34′ E), which was not chemically protected. The studies were conducted on five shrubs of two hazel cultivars, ‘Barra’ and ‘Mogulus”, characterized by their different resistance to those two pests.

The degree to which the cultivars were injured by the filbert big bud mite was directly established in the field through determining the percentage of buds with clear signs of deformation. These observations were made in the no-leaf period of the plants. The assessment consisted of determining the percentage of infected buds on ten branches of each shrub, together with the ramifications. In addition, the eggs, larvae and adult forms of Eriophyoidea were counted under a microscope: 20 injured and 20 uninjured buds from each selected shrub of each cultivar were analysed, and the final number of overwintering stages was determined, converted to a value per 10 ml of water, according to the method used by Maziarz (1984). In the case of the filbert aphid, the population of the pest was monitored from early spring until the occurrence of the first larvae on hazel leaves in 10-day intervals. The method used was to count the aphids on 100 leaves of each plant (Wilkaniec 1994). The aphid population was checked at 10-day intervals until it had completely disappeared. The number of larvae and winged females as well as the number of colonized leaves were noted. The level of the pest population on the tested cultivars was expressed as a cumulative aphid index (the sum of aphids on 100 leaves), and the mean percentage of colonized leaves was established.

Laboratory studies

The material for studies were the buds and leaves of two hazel cultivars: ‘Barra’ and ‘Mogulus’. The buds (undamaged by mites) were collected from the plants of these cultivars in April, while the plant was still in the no-leaf period, whereas the leaves (previously covered, undamaged by aphids) were collected in June, which is when the aphid can be found at peak concentrations on hazel (Gantner 2007). The plant material was sampled before noon, on sunny and dry days. Directly after harvesting, the study material was dried at temperatures of up to 35 °C in the shade and under air flow conditions.

Essential oils

Essential oils were obtained using the distillation method with water vapour with the addition of meta-xylene in 20 g of dry air. The EO were analysed using gas chromatography (GC) and GC/mass spectrometry (MS) methodology (Polish Pharmacopoeia VI 2002).

GC and GC/MS determinations

The GC/MS system consisted of an ITMS Varian 4,000 GC–MS/MS instrument (Varian, Palo Alto, CA) equipped with a CP-8410 auto-injector and a 30 m × 0.25-mm VF-5 ms column (Varian); film thickness was 0.25 μm, the carrier gas and flow rate was He at 0.5 ml/min, the injector and detector temperatures were 220 and 200 °C, respectively, the split ratio was 1:20 and the inject volume was 1 μl. A temperature gradient was applied (60 °C for 0.5 min, then increments of 3 °C/min up to 246 °C, then holding at 246 °C for 10 min). The mass ranged from 40 to 1,000 Da, the GC system consisted of a Varian 3800 Series instrument (Varian) equipped with a JW Scientific DB-5 column (Agilent Technologies, Santa Clara, CA) operated under the same conditions as the GC/MS (flame ionization detector 256 °C; split ratio 1:50). The qualitative analysis was carried out on the basis of MS spectra which were compared with the spectra of the NIST library (NIST/EPA/NIH 2002) and from data available in the literature (Joulain and König 1998; Adams 2004). The identification of volatiles was verified with authentic standards purchased from commercial sources (Sigma-Aldrich, St. Louis, MO) that had the same GC retention times and mass spectra. The identify of compounds was confirmed by their retention indices taken from the literature and own data (Joulain and König 1998; Adams 2004). The quantitative composition of oils was determined by GC (flame ionization detector), by assuming the total of all the different oils to be 100 %.

Statistical analysis

For all statistical laboratory experiments and analysis of the results, we used the software package Statistica 9.1 (StatSoft, Tulsa, OK). Analysis of variance with the Tukey simultaneous test was applied, with significance set at p ≤ 0.05. The average results on the figures were determined with standard deviation. Correlation coefficients between the total aphid population and the number of colonized leaves were calculated. All determinations were performed with five replicates (field experiments) and in triplicate (laboratory experiments).

Results

Field observations showed that the plants of the studied hazel cultivars are characterized by varying resistance to colonization by the filbert big bud mite. The percentage of infected buds of cv. ‘Barra’ varied during the study years, ranging from 5.4 % in 2004 to 8.4 % in 2005 and up to 13.2 % in 2006. In contrast, the infection index for cv. ‘Mogulus’, based on visual estimation, remained at the zero level during the same period. Winter analyses of buds deformed by the filbert big bud mite and those without any external signs of its feeding indicated the occurrence of all developmental stages during the winter. Eggs, larvae and adult forms of Eriophyoidea were present a high numbers, ranging from 786.6 in 2004 to 1,533.8 in 2006, with, on average, 1,293.8 individuals. In both cultivars, signs of feeding by the filbert big bud mite were found in buds showing no external signs of feeding, with the population of the wintering stages being more than eightfold higher in the buds of cv. ‘Barra’ (Table 1).

Table 1 The average number of Phytoptus avellanae development stages in damaged and undamaged buds during the study period of 2004–2006
Full size table

Parallel field studies on the resistance of the selected hazel cultivars to filbert aphid showed, similar to our observations on the filbert big bud mite, that cv. ‘Barra’ was more susceptible to the occurrence of aphids than cv. Mogulnus. This pest was more numerous on the chemically protected plantation, and it clearly preferred the leaves of cv. ‘Barra’ to those of cv. ‘Mogulnus’. This is reflected both by the higher value of the cumulative aphid index and by the higher percentage of colonized leaves on the plants of cv. ‘Barra’ (Table 2). We also observed a directly proportional relationship between the total aphid population and the number of colonized leaves, both in the protected plantation (r = 0.93) and in the plantation that was not chemically protected (r = 0.95). The degree of colonization of the cultivars growing in a protected and unprotected orchard was highly convergent (r = 0.70).

Table 2 Total mean number of Myzocallis coryli individuals on 100 leaves and mean number of colonized leaves of two hazelnut cultivars under field conditions during the study period of 2003–2006
Full size table

Analysis of the results of the laboratory analyses revealed that the content of EO in the study material ranged from 0.32 to 0.46 % in buds and from 0.75 to 0.95 % in leaves. Comparing the studied cultivars, we observed that both the buds and leaves of cv. ‘Mogulus’ contained more EO than those of cv. ‘Barra’. Irrespective of the cultivar, the leaves contained more than twice as many EO than the buds.

The analysis of EO by GC and GC/MS showed the presence of 45 compounds of various chemical groups in the raw materials examined. The chemical composition of the EO changed depending on the kind of raw material, the cultivar and the date of harvest.

Monoterpenes represented >70 % of the EO and therefore constituted the dominating fraction of compounds identified in hazel buds (Fig. 1). The second highest fraction of EO consisted of sesquiterpenes, whose quantity—depending on the cultivar—ranged from 0.3 % (cv. ‘Barra’) to 2.6 % (cv. ‘Mogulus’). The EO obtained from hazel leaves was also a mixture of mono- and sesquiterpenes, whose percentage is graphically presented in Fig. 2.

Fig. 1
figure 1

Percentage of identified fractions present in the mixture of essential oils (EO) isolated from the buds of two hazelnut (Corylus L.) cultivars ‘Barra’ and ‘Mogulnus’. For each fraction, different lower-case letters above bars indicate a significant difference at p ≤ 0.05 (Tukey test) between cultivars

Full size image
Fig. 2
figure 2

Percentage of identified fractions present in the mixture of EO isolated from the leaves of two hazelnut (Corylus L.) cultivars ‘Barra’ and ‘Mogulnus’. For each fraction, different lower-case letters above bars indicate a significant difference at p ≤ 0.05 (Tukey test) between cultivers

Full size image

Our analyses revealed that monoterpene hydrocarbons constituted the main group among the monoterpenes present in the hazel buds of cvs. ‘Barra’ (42.0 %) and ‘Mogulus’ (41.4 %). In comparison, in the EO obtained from the leaves of these cultivars, monoterpene hydrocarbons and monoterpenes soluble in water constituted 16.2 and 24.2 % (cv. ‘Barra’) and 23.9 and 19.9 % (cv. ‘Mogulus’) of the EO (Figs. 3, 4).

Fig. 3
figure 3

Percentage of identified groups of monoterpenes present in the mixture of EO isolated from the buds of two hazelnut cultivars. For each type of monoterpene, different lower-case letters above bars indicate a significant difference at p ≤ 0.05 (Tukey test) between cultivers. moh Monoterpene hydrocarbons, akmo aldehydes and ketones monoterpenes, mow monoterpenes soluble in water, amo alcohol monoterpenes

Full size image
Fig. 4
figure 4

Percentage of identified groups of monoterpenes present in the mixture of EO isolated from the leaves of two hazelnut cultivars. For each type of monoterpene, different lower-case letters above bars indicate a significant difference at p ≤ 0.05 (Tukey test) between cultivers

Full size image

The mixture of EO present in the buds of cv. ‘Mogulus’, which is resistant to the filbert big bud mite, was characterized by a high content of such compounds as nerol, α-campholenol, methyl salicylate, spathulenol, β-caryophyllene, δ-cadinene (Fig. 5). In comparison, the mixture of EO present in the leaves of this cultivar, which were colonized by the filbert aphid to only a small extent, contained greater quantities of nerol, α-campholenol, p-cymene, α-terpineol and germacrene D than those of cv. ‘Barra’, which showed a higher acceptance by aphids (Fig. 6). In turn, the leaves of cv. ‘Barra’ were characterized by a considerably high content of menthol, limonene, isomenthone, methyl salicylate and L-menthone (Fig. 7).

Fig. 5
figure 5

Average percentage of repellent components in the mixture of EO isolated from the buds of two hazelnut cultivars. For each repellent compound, different lower-case letters above bars indicate a significant difference at p ≤ 0.05 (Tukey test) between cultivers

Full size image
Fig. 6
figure 6

Percentage of EO components found to be a repellent for aphids isolated from the leaves of two hazelnut cultivars. For each repellent compound, different lower-case letters above bars indicate a significant difference at p ≤ 0.05 (Tukey test) between cultivers

Full size image
Fig. 7
figure 7

Percentage of EO components found to be an attractants for aphids isolated from the leaves of two hazelnut cultivars. For each attractant, different lower-case letters above bars indicate a significant difference at p ≤ 0.05 (Tukey test) between cultivers

Full size image

Discussion

Among the many possibilities provided by contemporary methods aimed at reducing the population of agrophagous species, alternative methods taking cultivar resistance of the species into account are finding an increasingly greater application in integrated programmes of plant protection (Chamberlain et al. 2000; Dudareva et al. 2004; Holopainen 2004; Zhao et al. 2005). The cultivation of less susceptible cultivars or those which are more resistant is an important element of plant protection methods (Kogan 2000). The relationship between plants and insects or mites is one of the frontier topics in chemoecology, and a clear understanding of the selectivity of mites for the host plant is important for developing control measures (Xugen and Luqin 2006). Secondary metabolites are considered to be the most important source of a plant’s resistance to pests (Gatehouse 2002). Some of the secondary metabolites consist of constitutively expressed volatile organic compounds (VOCs), which can be quantitatively or qualitatively induced after herbivory (Dicke et al. 2009; Turlings and Ton 2006; Van Den Boom et al. 2004). However, many terpene VOCs, such as indole and methyl salicylate, are commonly produced as a result of injury to the plant following insect herbivory (De Moraes et al. 2001). VOCs are formed and released even by healthy plants or undamaged parts of the plant. For example, Röse and Tumlinson (2005) observed that herbivorous cotton plants which were attacked released VOCs even from those leaves which did not suffer injuries. VOCs are directly involved in plant defense system. The same compounds may attract natural enemies of insects and carnivorous insects as well as parasitoids (Arimura et al. 2005; Leither et al. 2005).

Our results suggest that the degree of resistance to P. avellanae is a fairly stable cultivar property. This notion is confirmed by the results in the successive years of studies, when the buds of plants of cv. ‘Mogulnus’ did not show any external signs of feeding by the filbert big bud mite. In comparison, infestation of cv. ‘Barra’ plants showed an increasing trend, and the buds of this cultivar were colonized to a considerable degree by mites. Bud infestation by the filbert big bud mite in the regions of hazel cultivation ranged from 18 to 70 % and is first of all related to the cultivar and the manner of its cultivation (Özman and Toros 1996; Stamenković et al. 1997; Viggiani and Bianco 1973). This problem appears to be different in comparison to the infection of black currant by filbert big bud mite (Cecidophyopsis ribis), which—according to Gajek et al. (1996)—is mainly dependent on the environment.

In our experiment, we confirmed that resistance of hazel cultivars to filbert aphid is a stable feature, independent of the year and intensity of the applied chemical protection. Filbert aphid is a serious pest of hazel plantations throughout the world and represents both a direct and indirect threat to the cultivated plants. By the 1980s it had become the most dangerous pest of hazel in the USA and Canada as it was resistant to most insecticides (AliNiazee 1997; Katundu and AliNiazee 1990). It was only after applying an integrated programme of controlling this aphid, based on establishing the threshold of harmfulness and introducing a parasitic hymenoptera Trioxys pallidus, was it possible to control their occurrence (Olsen, 2000). Controlling M. coryli on commercial hazel plantations is difficult because changes in the intensity of its occurrence are related to weather conditions (Gantner 2000).

The available literature lacks data on the content of EO occurring in hazel buds. Based on previous research (Najda and Gantner 2012) we conclude that nerol, myrthenol, α-campholenol and p-cymene were the predominant constituents of EO in hazel buds, with their contents ranging up to 20.6, 19.5, 4.7 and 3.7 %, respectively. The main constituents of the EO of hazel leaves were nerol (13.0 %), myrthenol (9.4 %), α-campholenol (9.0 %), menthol (6.7 %), geraniol (4.8 %) and limonene (3.9 %). The organic compounds constituting the EO are secondary metabolites of plants that show multi-directional pharmacological action towards humans and animals. It is currently recognized that monoterpenes play many roles in plant–insect interactions (Piesik et all. 2013b).

The chemical composition of EO can determine the acceptance of the host plant, or its lack, by herbivores. EO perform an important function as different types of aromatic stimuli: attractants, repellents and toxins. In the case of the filbert big bud mite, which moves in a passive manner, EO affect the olfactory receptors of the pest at the moment of its direct contact with the plant (Johnson and Schaal 1957). Therefore, it would appear that compounds of a repellent nature will be of decisive importance in the resistance of a cultivar to Eriophyoidea. As was shown in our study, the EO present in the buds of cv. ‘Mogulus’ is characterized by a greater content of nerol, α-campholenol, methyl salicylate, spathulenol and β-caryophylene. Hence, these compounds can be designated as substances which deter mites from feeding, or which make it impossible for them to feed. Of note is the high content of methyl salicylate in the buds of cv. ‘Mogulus’, which results from the indirect induction of the plant’s resistance in a response to the feeding of mites. As reported by Dicke et al. (1990) and Bouwmeester et al. (1999), the level of this substance increased following feeding of Tetranychus urticae on Phaseolus lunatus plants, which increased the attractiveness of the plants to the species of predatory spider mite Phytoseiulus persimilis. Gajek et al. (1996) showed that the increased content of 3-carene and α-caryophylene in plants of a black currant cultivar resistant to the mite Cecidophyopsis ribis had a repellent effect and caused signs of vigour loss in the mite. A similar phenomenon in which specific concentrations of selected components of EO had a repellent effect on spider mites feeding on strawberry plants was also observed by Dąbrowski and Rodriguez (1972). According to Herr (1988), the negative effect of factors on the olfactory organs affecting cultivar resistance to the filbert big bud mite can be overcome by the latter. This is reflected in the results obtained by the author since individuals of P. avellanae were also observed on the buds of all studied cultivars without any external signs of deformation.

EO content can also be decisive in determining the acceptance (or its lack thereof) of the host plant by aphids colonizing new plants in an active manner. Our study demonstrated that the leaves of cv. ‘Mogulus’ plants, a cultivar that is colonized to only a minor extent by the filbert aphid, contains greater quantities of nerol, α-campholenol, p-cymene, geraniol, α-terpineol, germacrene D, n-nonanol and α-copaene than those of cv. ‘Barra’, a cultivar with a higher acceptance by aphids and which contains a considerably higher content of menthol, limonene, isomenthol and methyl salicylate. Hence, these latter compounds can be included within the category of attractants which help the aphid to localize the host plant. In contrast the compounds of the EO whose presence was significantly higher in the leaves of cv. ‘Mogulus’ than in those of cv. ‘Barra’ can be considered to affect the winged individuals of the filbert aphid as anti-nutritional compounds. According to Francis et al. (2005), aphids, especially two-host species, use VOCs to localize the host plant, and the fact that they make spiral circles of a decreasing radius around the plant testifies to their using the aromatic stimuli to localize the host (Chapman et al. 1981; Kennedy 1974). According to Cheng et al. (2007) and Piesik et al. (2011a, b, 2013a, b), plant terpenoids take part in the direct and indirect responses of the plant to injuries, and the majority of mono- and sesquiterpenes are produced by plants after an attack by plant herbivores with the aim of attracting parasitoids. In a study by Ozawa et al. (2000), plants of Lotus japonicus attacked by spider mites Tetranychus urticea produced VOCs that attracted the predator Phytoseiulus persimilis. Scots pine has been shown to produce a volatile bouquet that attracts egg parasitoids in response to oviposition of the herbivorous sawfly Diprion pini (Mumm and Hilker 2005).

These results indicate that aromatic VOCs play a major role in plant–pest interactions. Future studies should determine the exact concentrations of selected compounds of the EO and their application as defensive substances against harmful insects and mites.