Introduction
Galls are the result of mutual interactions between gall-inducing herbivores and their host plants (Patra et al., Reference Patra, Bera and Mehltreter2010; Samsone et al., Reference Samsone, Andersone and Ievinsh2012; Jiang et al., Reference Jiang, Veromann-Jürgenson, Ye and Niinemets2018). They are formed entirely from the plant tissues and this process involves the enlargement and proliferation of plant cells (Oliveira et al., Reference Oliveira, Isaias, Fernandes, Ferreira, Carneiro and Fuzaro2016). Their structures are determined by the gall-inducing species, even on the same host plant (Hartley, Reference Hartley1998). Oak gall wasps (Hymenoptera: Cynipidae) appear to be most specialized among gall-inducers. They have complex cyclic parthenogenetic life cycles and the ability to induce not only species-dependent, but even generation-dependent galls on oaks and other Fagaceae (Stone et al., Reference Stone, Schönrogge, Atkinson, Bellido and Pujade-Villar2002). An example of such a model is Neuroterus quercusbaccarum (L.), which inhabits oak trees – Quercus robur (L.), Q. pubescens Willd. and Q. petrea (Matt.) Liebl. – throughout Europe, North Africa and Asia Minor (Kierych, Reference Kierych1979). N. quercusbaccarum is heterogonic, with sexual (♀♂) and agamic (ȣ) generation per year. These two generations differ from each other in galls structure and morphology of adult wasps. Females of agamic generation are flying from January to May. They lay eggs on the developing leaves or into flower buds of male catkins (Kovácsné-Koncz et al., Reference Kovácsné-Koncz, Szabó, Máthe, Jámbrik and M-Hamvas2011). A spherical and smooth-surfaced currant gall forms around the developing larva. Galls are approximately 4 mm in diameter, one-chambered and single in catkins or below the leaves, green at first, then with a red mark above when mature in May and June. The larvae pupate in galls on a tree; females and males of sexual generation emerge from May to July. Fertilized females are laying eggs on oak leaves. A spangle gall develops around the hatched larvae. Galls are single-chambered, in the shape of a disc with a rising center, yellow with red hairs. They have a diameter of about 5 mm; they occur from July to October beneath the leaves and mature in August. The larva pupates in the gall on the ground in winter (Kampichler & Teschner, Reference Kampichler and Teschner2002).
Gall formation is associated with multiple changes in plant cells, including morphological features, nutrient concentration, metabolic signaling, photosynthetic capacity or oxidative enzyme activities (Oliveira et al., Reference Oliveira, Isaias, Moreira, Magalhães and Lemos-Filho2011; Giron et al., Reference Giron, Huguet, Stone and Body2016; Kmieć et al., Reference Kmieć, Rubinowska and Golan2018; Kot & Rubinowska, Reference Kot and Rubinowska2018). The earliest response of plants to galling stimuli have been often associated with increased production of reactive oxygen species (ROS) (Isaias & Oliveira, Reference Isaias, Oliveira, Mérillon and Ramawat2012), which may induce detrimental oxidation of macromolecules including nucleic acids, proteins and lipids, leading to cellular toxicity (Bela et al., Reference Bela, Horvátha, Galléa, Szabadosb, Tari and Csiszára2015). On the other hand, ROS are also produced as a natural by-product of the plant metabolism and have a crucial role in cell signaling, which may determine the extent of gall tissue alterations (Isaias & Oliveira, Reference Isaias, Oliveira, Mérillon and Ramawat2012; Carneiro et al., Reference Carneiro, Castro and Isaias2014). ROS include hydroxyl radical (OH•), superoxide anion radical (O2•–), perhydroxyl radical (HO2•), alkoxyl radical (RO•), singlet oxygen (1O2) and hydrogen peroxide (H2O2). H2O2 has an important role in plants as a signaling molecule in defense response due to its relatively stable form of oxygen and the ability to diffuse freely (Maffei et al., Reference Maffei, Mithöfer and Boland2007). Nevertheless, excessive production of H2O2 in plant cells leads to oxidative stress and programed cell death (Quan et al., Reference Quan, Shang, Shi and Li2008). The accumulation of ROS increases lipid peroxidation (LPO) in biological membranes affecting their structure, which decreases membrane fluidity, increases permeability and loss of enzymatic activity (Gill & Tuteja, Reference Gill and Tuteja2010). High ROS production is buffered by several non-enzymatic (e.g., ascorbic acid, glutathione, tocopherol and carotenoids) and enzymatic (e.g., peroxidases, dismutase, catalase, polyphenol oxidase) mechanisms. This cellular detoxification system protects plant cells from oxidative damage by converting ROS into less toxic products in the cell (Gill & Tuteja, Reference Gill and Tuteja2010; Pandey et al., Reference Pandey, Fartyal, Agarwal, Shukla, James, Kaul, Negi, Arora and Reddy2017). Among enzymatic components, peroxidases play a key role in the suppression of toxic H2O2 (Caverzan et al., Reference Caverzan, Passaia, Rosa, Ribeiro, Lazzarotto and Margis-Pinheiro2012).
It is known that galling insects induce local accumulation of nutrients in plant tissues (Giron et al., Reference Giron, Huguet, Stone and Body2016; Kot et al., Reference Kot, Jakubczyk, Karaś and Złotek2018a). This also applies to amino acids, thus changes in the metabolism of these compounds may be an important part of biochemical plant responses to signals associated with the presence of gall-inducing species. Few published studies (Sempruch et al., Reference Sempruch, Marczuk, Leszczyński, Kozak, Zawadzka, Klewek and Jankowska2013; Reference Sempruch, Golan, Górska-Drabik, Kmieć, Kot and Łagowska2014; Kmieć et al., Reference Kmieć, Rubinowska and Golan2018), regarding insect–plant interactions have suggested that the degradation of amino acids as nutrients for herbivores may be connected with the state of amino acid decarboxylases. They are key enzymes of plant amines biosynthesis and ornithine decarboxylase (ODC; EC 4.1.1.17) and lysine decarboxylase (LDC; EC 4.1.1.18) play a crucial role in this process. However, their activity depends on both plant and herbivore species, their density and duration of infestation (Sempruch et al., Reference Sempruch, Leszczyński, Protasiuk and Zarzecka2012b). In turn, tyrosine decarboxylase (TyDC; EC 4.1.1.25) participate in the biosynthesis of aromatic monoamines and other classes of defensive plant compounds (Miller-Fleming et al., Reference Miller-Fleming, Olin-Sandoval, Campbell and Ralser2015).
Changes in the content of assimilation pigments and photosynthetic activity are a common mechanism to assess the impact of insects on plants. Insect feeding, as a biotic stress, causes a loss of chlorophyll content, because chloroplasts during stress are able to produce strong oxidants responsible for the oxidation of pigments (Guidi & Degl'Innocenti, Reference Guidi, Degl'Innocenti, Venkateswarlu, Shanker, Shanker and Maheswari2012). In turn, galls show a high demand for assimilates; however, certain gall-inducing species have the ability to stimulate the rate of photosynthesis, while others reduce it, which probably depends on the tolerance of plant species to insect feeding (Nabity et al., Reference Nabity, Zavala and DeLucia2009; Haiden et al., Reference Haiden, Hoffmann and Cramer2012). The analysis of chlorophyll a fluorescence is one of the most widely used methods of investigating changes in the photosynthetic apparatus and photosynthetic efficiency in higher plants (Kalaji et al., Reference Kalaji, Carpentier, Allakherdiev and Bosa2012).
Interactions between gall-inducing Cynipidae and oaks as their host plants seem to be a model system for studying gall insect–host plant interactions due to their high specialization. Cynipidae gall wasps are highly monophagous and their galls are the archetype of highly specialized nutritive tissues that form discrete microhabitats. It will be useful to carry out physiological and biochemical studies, which will allow clarifying some key events of oak–Cynipidae interactions to better understand host plant responses after the attack of gall-inducing Cynipidae. The present study used galls caused by sexual generation (♀♂) of N. quercusbaccarum on pedunculate oak trees (Q. robur L.) as a model. We analyzed the enhancement of selected physiological and biochemical parameters in foliar tissues with galls and gall tissues themselves and compared it to non-infested tissues. We aimed to answer the following questions: can galls alter the fluidity of cell membranes? Can galls induce oxidative stress and modulate plant's reaction to it? How galling process affects photosynthesis and pigment contents? We determined cytoplasmic membrane condition, H2O2 content, the activity of antioxidant enzymes and amino acid decarboxylase as well as chlorophyll fluorescence parameters to answer these questions.
Methods and materials
Study site and sampling
The study site was located in Lublin (Poland) (22°34′E, 51°14′N), where the climate is typical for a humid continental climate with cold, damp winters and warm summers. Plant material was collected in May from pedunculate oak trees (Q. robur L.), which were part of urban green areas. The trees were about 15–20 years old and 5–8 m tall. Individual trees with fully developed galls induced by sexual generation (♀♂) of N. quercusbaccarum (L.) were marked (n = 10). Samples of leaves with visible, fully developed galls and accessible at arm's length were randomly collected, however leaves with only one gall were included. Leaves located at a similar site on the shoots and a similar canopy position just like leaves with galls were used as a control. Ten leaves with galls and ten non-galled leaves were collected from each tree, so each sample consisted of 100 leaves. Leaves were cut off with scissors and brought in plastic bags to the laboratory within 1 h after collection. In the laboratory, galls were cut off with a scalpel and sectioned to remove larvae, and the plant material was categorized as follows: control leaves (leaves without galls), leaves with removed galls and galls. Leaves and galls in the sample were cut into small pieces and mixed to be more representative. Such plant material was weighed and used directly for physiological analysis (H2O2 assay, E L assay, LPO, GPX and APX activity, photosynthetic and photoprotective pigment contents). The material used to assess amino acid decarboxylase activities was frozen and stored at −80 °C until analysis. Physiological and biochemical assays were made in three biological replicates (n = 3).
The effect of N. quercusbaccarum galls on photosynthesis was evaluated by chlorophyll a fluorescence measurements. It was determined in field conditions on 40 leaves-20 intact leaves as control and 20 leaves with fully developed galls. The same leaves were subsequently detached for pigment analysis after chlorophyll a fluorescence measurements.
Fluorescence measurement
Fluorescence measurements were conducted using the saturation pulse method with a PAM-2000 fluorometer (Walz GmbH, Germany) (Schreiber, Reference Schreiber and Papageorgiou2004). They were taken before 11 am, after adaptation in the dark for about 20 min. The minimum (F 0) fluorescence was measured after dark adaptation. Subsequently, a flash of light sufficient to drive photosynthesis was applied and the maximum fluorescence (F m) was measured. After 10 min the substrate fluorescence (F s) was determined under steady-state conditions. The maximum chlorophyll fluorescence (F m′) was measured by applying pulses of the saturated white light every 60 s when actinic light was on.
The maximum quantum yield of photosystem II (PSII) was determined using the following formula:
$$F_{\rm v}/F_{\rm m} = {\rm (}F_{\rm m} - F_0{\rm )}/F_{\rm m},$$where F v (variable fluorescence) is equal to the increase in fluorescence induced by the saturation pulse. The effective quantum yield (Y) of PSII photochemistry was determined as
$$\Delta F/F^{\prime}_{\rm m} = {\rm} F^{\prime}_{\rm m} - {\rm (}F_{\rm s}/F^{\prime}_{\rm m}{\rm )}$$Fluorescence quenching parameters, such as q P (photochemical quenching) and q N (non-photochemical quenching) were calculated according to the equations:
$$q_{\rm P} = [({F}^{\prime}_{\rm m}-{\rm} F_{\rm s})/({F}^{\prime}_{\rm m}-{\rm} {F}^{\prime}_0)]$$and
$$q_{\rm N} = [(F_{\rm m} - {\rm} {F}^{\prime}_{\rm m})/({F}^{\prime}_{\rm m}-{\rm} {F}^{\prime}_0)]$$q P and q N requires the F 0′ parameter, which was obtained after a dark red light pulse applied to previously light-adapted leaves
Laboratory assay
H2O2 content
The procedure of Jena & Choudhuri (Reference Jena and Choudhuri1981) was followed to determine H2O2 content. Briefly, 0.5 g of plant material was ground in 3 ml of phosphorus buffer (50 mM, pH 6.5) at 4 °C. Then, the mixture was centrifuged at 6000 × g for 25 min. Next, 1.5 ml of the supernatant was added to 0.5 ml TiO2 in 20% (v/v) H2SO4 and centrifuged again at 6000 × g for 15 min at room temperature. The absorbance of the supernatant was measured at 410 nm using a spectrophotometer (Cecil CE 9500, UK). H2O2 content was calculated using the molar absorbance coefficient (0.28 µM−1 cm−1) and expressed as nanomoles per 1 g fresh weight.
Electrolyte leakage (E L) assay
It was measured according to the method of Kościelniak (Reference Kościelniak1993) using an Elmetron CC-317 microcomputer conductometer. Ten leaf rings of 0.9 cm diameter were cut with a cork borer from each sample, and subsequently transferred to 20 cm3 of deionized water and incubated at room temperature on a rotary shaker for 24 h. The initial electrical conductivity (K1) then was measured. The samples were autoclaved at 100 °C for 15 min. Final conductivity of the solution was measured (K2) after 24 h of shaking. Electrolyte leakage was calculated using the following formula:
$$E_{\rm L}\;(\% ) = (K1/K2) \times 100.$$Membrane lipid peroxidation
Its level was measured as the amount of thiobarbituric acid reactive substances (TBARS), according to Heath & Packer (Reference Heath and Packer1968). Plant tissues (0.2 g) were homogenized in 0.1 M potassium phosphate buffer (pH 7.0) and then the homogenate was centrifuged at 12,000 × g for 20 min at room temperature. The supernatant (0.5 cm3) was mixed with 2 cm3 of 20% trichloroacetic acid (TCA) containing 0.5% thiobarbituric acid (TBA). The mixture was incubated at 95 °C for 30 min, and then the samples were quickly quenched. Another centrifugation was carried out at 10,000 × g for 10 min. The absorbance was measured at 532 and 600 nm using a spectrophotometer (Cecil CE 9500, UK). The TBARS content was calculated using the molar absorbance coefficient (155 nM−1 cm−1) and expressed as nanomoles per 1 g fresh weight.
Peroxidase activity
For guaiacol peroxidase (GPX, EC 1.11.1.7) and ascorbate peroxidase (APX, EC 1.11.1.11) activities, 0.2 g of each sample was homogenized in 0.05 mol dm−3 phosphate buffer (pH 7.0) containing 0.2 mol dm−3 EDTA and 2% PVP at 4 °C. Then, the homogenates were centrifuged for 10 min (10,000 × g, 4 °C) and immediately used for analyses. GPX activity was measured as described by Małolepsza et al. (Reference Małolepsza, Urbanek and Polit1994). The reaction mixture contained 0.5 cm3 of 0.05 mol dm−3 phosphate buffer (pH 5.6), 0.5 cm3 of 0.02 mol dm−3 guaiacol, 0.5 cm3 of 0.06 mol dm−3 H2O2 and 0.5 cm3 of enzyme extract. The variation in absorbance was measured at 480 nm for 4 min, at 1 min intervals using a spectrophotometer (Cecil CE 9500, UK). GPX activity was calculated using the absorbance coefficient for this enzyme (26.6 mM cm−1) and expressed as the change in peroxidase activity per fresh weight (U mg−1 FW).
APX activity was determined according to the method of Nakano & Asada (Reference Nakano and Asada1981). The reaction mixture contained 1.8 ml 0.1 M phosphorus buffer (pH 6.0), 20 µl of 5 mM sodium ascorbate, 100 µl of 1 mM H2O2 and 100 µl of enzymatic extract. The absorbance was measured at 290 nm for 5 min, at 1 min intervals with a spectrophotometer (Cecil CE 9500, UK). APX activity was calculated using the absorbance coefficient for this enzyme (2800 M−1 cm−1) and expressed as the change of peroxidase activity per fresh weight (U mg−1 FW).
Amino acid decarboxylases assays
Plant material was homogenized with 0.2 M phosphate buffer (pH 8.2) containing β-mercaptoethanol and ethylenediaminetetraacetic acid (EDTA) for ODC, while Tris-HCl buffer pH 5.6 (0.2 M) was used for the extraction of LDC and 0.5 M acetate buffer (pH 5.6) for TyDC. Subsequently, the obtained extracts were filtered through two layers of cheese-cloth and centrifuged at 18,000 × g at 5 °C. The activity of ODC, LDC and TyDC was assayed using a UV-Vis spectrophotometer (Hewlett Packard 8453) according to Ngo et al. (Reference Ngo, Brillhart, Davis, Wong, Bovaird, Digangi, Risov, Marsh, Phan and Lenhoff1987) and Phan et al. (Reference Phan, Ngo and Lenhoff1982, Reference Phan, Ngo and Lenhoff1983), respectively. Enzyme activities were expressed as μmol of appropriate amine, generated during 1 h of enzymatic reaction by 1 mg of enzymatic protein. The protein content in enzymatic extracts was determined with the method of Lowry et al. (Reference Lowry, Rosebrough, Farr and Randal1951).
Photosynthetic and photoprotective pigment content
In total, 0.5 g of each sample was extracted in 80% acetone. The procedure of Lichtenthaler & Wellburn (Reference Lichtenthaler and Wellburn1983) was followed to determine the content of chlorophyll a, b and carotenoids. The absorbance was determined at 470 nm (for Car), 646 nm (for chlorophyll b) and 663 nm (for chlorophyll a) using a spectrophotometer (Cecil CE 9500, UK). The following formulas were used to estimate pigment contents:
$$\eqalign{&C_{{\rm chl}\,\,{\rm a}} = 12.21 \times A_{663} - 2.81 \times A_{646} \cr &C_{{\rm chl}\;{\rm b}} = 20.13 \times A_{646} - 5.03 \times A_{663} \cr &C_{{\rm car}} = \left( {1000 \times A_{470} - 3.27 \times C_{{\rm chl}\;{\rm a}} - 104 \times C_{{\rm chl}\;{\rm b}}} \right)/227,} $$where A λ – absorbance value for wale length λ.
The values of pigment contents were expressed as mg/g fresh weight.
To determine anthocyanin contents, 1 g of each sample was taken and extracted for 4 h in 10 ml of 0.1% HCl–MeOH at room temperature. Then, the extracts were measured using a spectrophotometer (Cecil CE 9500, UK) at 530 and 657 nm. The formula: A = (A 530 − 0.25A 657) was used to compensate for the contribution of chlorophyll and its degraded products to the absorption at 530 nm. Anthocyanidin content was calculated following the formula of Rabino & Mancinelli (Reference Rabino and Mancinelli1986): [Absorbance × 449.2 × dilution factor]/[29.600 × Sample Weight (g)],
where 449.2 = molecular weight of Cyanidin-3-glucoside; dilution factor = final volume/initial volume; 29.600 = molar extinction coefficient.
The values of anthocyanin contents were expressed in milligrams of Cyanidin-3-glucoside equivalent per 1 g of fresh weight.
Statistical analysis
All data were presented as means (x̄) with standard error values (±SE) and statistical analyses were calculated using Statistica for Windows v. 13.1 (Statistica StatSoft Inc., 2016). The Shapiro–Wilk test was used to verify the normality of the obtained results. The homogeneity of variance was analyzed using test of Levene. One-way ANOVA with the Tukey's simultaneous test (HSD) was used to compare differences in physiological and biochemical parameter contents/activities, as well as photosynthetic and photoprotective pigments concentrations in the samples. Changes in chlorophyll a fluorescence parameters between tissues were examined using the Student t-test or the non-parametric Mann–Whitney U test. The significance threshold was set at p < 0.05. Physiological and biochemical assays were performed in three independent biological replicates (n = 3), while chlorophyll a fluorescence in twenty replicates (n = 20).
Results
H2O2 content
The presence of N. quercubaccarum galls on the oak leaves caused a twofold reduction in the H2O2 content compared to control leaves (fig. 1a). The level of this free radical in the galls was the lowest, as a 20-fold difference between gall tissues and control was measured.
Changes in the content/activity of hydrogen peroxide (H2O2) (A), electrolyte leakage (E L) (B), thiobarbituric acid reactive substances (TBARS) (C), guaiacol peroxidase (GPX) (D) and ascorbate peroxidase (APX) (E) in the leaves with galls and in galls of N. quercusbaccarum (L.) compared to control leaves; values followed by the same letter do not differ significantly at p < 0.05.
The state of cell membranes
E L measurements from cells were used to estimate the degree of cell membrane damage during the gall formation process of N. quercusbaccarum on oak leaves. E L levels did not differ significantly between control and leaves with galls (fig. 1b). E L in galls was higher by almost 60% compared to control and leaves with galls.
Lipid peroxidation was expressed by the amount of malondialdehyde (MDA) determined with TBARS measurements. It was considered an indicator of oxidative stress. A significant increase in TBARS content was observed in leaves with N. quercusbaccarum galls in comparison to control samples (fig. 1c), as a 21% difference was noted. In turn, a high decrease in TBARS content was recorded in gall tissues; it was reduced by 57.91 and 62.42% when compared to control and leaves with galls, respectively.
Activity of antioxidants
The presence of N. quercusbaccarum galls on oak leaves had no effect on GPX activity (fig. 1d). However, this enzyme showed low activity levels in the galls. It was nearly 70% lower compared with control and leaves with galls.
Leaves with galls and galls of N. quercusbaccarum were characterized by a lower APX activity compared with control samples, but changes between leaves with galls and control were not significant (fig. 1e). The activity of this enzyme in galls was almost threefold and 2.5-fold lower when compared to control and leaves with galls, respectively.
Changes in the activity of amino acid decarboxylases
Statistical analysis showed significant differences in the activity of amino acid decarboxylases in leaves with galls of N. quercubaccarum and galls themselves, with the exception of TyDC (Table 1). Leaves with galls were characterized by an extremely low activity of LDC, which was 4.5-fold and 1.5-fold lower than in control and gall tissues, respectively. Leaves with galls also exhibited lower ODC activity of this enzyme compared to control samples. Whereas, the sharpest enhancement of its activity, 25-fold and 29-fold, was observed in galls of this cynipid species compared to control tissues and leaves with galls, respectively. The pattern of TyDC activity in N. quercusbaccarum galls and leaves with galls was similar and was characterized by slightly higher values than in control leaves.
Changes in the activity of amino acid decarboxylase in oak leaves with galls and gall tissues of N. quercusbaccarum (L.); values followed by the same letter do not differ significantly at p < 0.05 (n = 3).
Changes in the content of photosynthetic and photoprotective pigments
The average contents of chlorophyll a and b and total chlorophyll in leaves with galls were slightly higher, but not statistically confirmed, when compared to control leaves. In turn, galls of N. quercusbaccarum were characterized by extremely low levels of these pigments, which were more than 97% lower than in both, control and leaves with galls (Table 2). The content of carotenoids was lower in leaves with galls as well as in gall tissues, as compared to control samples. However, these changes were significant only between control and galls of N. quercusbaccarum (133-fold difference). The chlorophyll a/b ratio was significantly lower in both, leaves with galls and galls of N. quercubaccarum when compared to the average control value. Nevertheless, a higher reduction (more than 2.6-fold) was measured in gall tissues. A similar trend to the chlorophyll a/b ratio was found for the carotenoids/total chlorophyll ratio. The carotenoids/total chlorophyll ratio was lower by 43.48 and 38.09% in control and leaves with galls, respectively. The presence of N. quercubaccraum galls did not cause any significant changes in anthocyanin concentrations when compared to control leaves. The sharpest, almost 7.7-fold decrease was recorded in galls compared to control as well as leaves with galls.
Content of photosynthetic and photoprotective pigments in oak leaves with galls and gall tissues of N. quercusbaccarum (L.); values followed by the same letter do not differ significantly at p < 0.05 (n = 3).
Fluorescence measurement
The presence of N. quercusbaccarum galls resulted in a significant decrease in the values of all photosynthetic activity indices, except for the q N parameter (Table 3). The initial fluorescence intensity (F 0) and the maximum intensity (F m) showed significant decreases (1.2 and 1.3-fold, respectively) when compared to control leaves. In consequence, the reduction of the maximum fluorescence (F m) values and variable fluorescence (F v), determined by the F v = F m – F 0 equation, led to a decrease in the maximum quantum yield of photosystem II (F v/F m). In our study, F v/F m ratios in leaves with N. quercubaccarum galls were significantly reduced, by almost 20%. A subsequent reduction was observed for the maximum quantum yield of PSII values (Y), as a 1.4-fold difference between leaves with cynipid galls and control was measured. The values of fluorescence quenching parameters, such as photochemical fluorescence quenching (q P) and non-photochemical quenching (q N) showed a varied pattern of changes. We found that the presence of galls significantly reduced the q P coefficient and stimulated q N in oak leaves (Table 3). A decrease of 27.75% in q P values, and an increase of 37.44% of the q N value was detected in leaves with galls compared to control tissues.
Changes in chlorophyll a fluorescence parameters along a control leaves of Q. robur L. and leaves with N. quercusbaccarum (L.) galls (n = 20).
Discussion
Oxidative stress is an important part of plant response following the attack of herbivores. Feeding of Helicoverpa zea (Boddie) (Lepidoptera: Noctuidae) increased the rate of lipid peroxidation and •OH radical formation within soybean tissues (Bi & Felton, Reference Bi and Felton1995). These results are consistent with our data, proving an increase in the rate of lipid peroxidation in oak leaves, on which N. quercusbaccarum induced galls. Such a reaction can lead to damage of membrane phospholipids, loss of cell membrane integrity and uncontrolled leakage of electrolytes. Moreover, according to Bi et al. (Reference Bi, Murphy and Felton1997), oxidative stress caused by H. zea in cotton tissues was accompanied by the induction of the activity of certain oxidative enzymes, such as lipoxygenases, peroxidase, diamine oxidase, ascorbate oxidase, and NADH oxidase I, with a simultaneous decrease of such foliar antioxidants as ascorbic acid, total carotenoids and non-protein thiols. On the other hand, the attack of B. brassicae increased ascorbic acid level and the rate of lipid peroxidation and reduced the activity of superoxide dismutase (SOD), APX and ascorbate oxidase in cabbage leaves (Khattab, Reference Khattab2007). Ascorbic acid accumulation was also induced in triticale tissues under Sitobion avenae (F.) and Rhopalosiphum padi (L.) (Hemiptera: Aphididae) attack in the initial stage of infestation (Łukasik et al., Reference Łukasik, Goławska and Wójcicka2012). Prolonged aphid feeding resulted in losses of ascorbate. Moreover, this study showed an increase in APX activity in triticale tissues throughout the test period. These data strongly suggested that the mode of oxidative changes caused by herbivorous insects in plant tissues is strictly dependent on the genotype of herbivore and/or host plant and the time of infestation. In our study, a decrease in H2O2 levels in leaves with galls and in galls of N. quercusbaccarum on pedunculate oak leaves was observed, which was similar to changes induced by H. zea in soybean tissues (Bi & Felton, Reference Bi and Felton1995). These results may suggest that this process involves reducing the antioxidant potential of plant tissues by decreasing the level of some antioxidants with low molecular weight, such as carotenoids, and reducing peroxidases activities. Furthermore, it is possible that SOD activity is also inhibited during this phenomenon. The latter enzyme catalyzes the dismutation reaction of superoxide anion radical to oxygen and hydrogen peroxide constituting the primary line of defense (Sytykiewicz, Reference Sytykiewicz2014). Thus, a decrease in its activity may at least partly result in lower H2O2 level. However, this problem requires further research.
It can be assumed that the direction and intensity of oxidative changes are regulated at the transcriptional level by up-regulation and down-regulation of specific genes. This mechanism has been demonstrated for maize response to the feeding of S. avenae and R. padi (Sytykiewicz, Reference Sytykiewicz2014; Reference Sytykiewicz2016a; Reference Sytykiewicz2016b; Sytykiewicz et al., Reference Sytykiewicz, Chrzanowski, Czerniewicz, Sprawka, Łukasik, Goławska and Sempruch2014) as well as in other host–herbivore systems (Collins et al., Reference Collins, Afzal, Ward, Prescott, Sait, Rees and Tomsett2010; Munner et al., Reference Muneer, Jeong, Park and Jeong2018). Transcriptional responses were dependent on plant genotype, insect species and duration of infestation. These conclusions are also consistent with the results of Khattab (Reference Khattab2007), who found a decrease in soluble protein content in cabbage tissues during the oxidative response to B. brassicae foraging.
The herbivorous insects are often able to induce beneficial changes in the physiology and biochemistry of host plants. For example, aphid feeding often resulted in a local increase in the content of free amino acids, thereby limiting the nutritional value of host plant tissues for these insects (Florencio-Ortiz et al., Reference Florencio-Ortiz, Sellés-Marchart, Zubcoff-Vallejo, Jander and Casas2018). Similarly, some gallers induce the over accumulation of amino acids, sugars and other plant nutrients in gall tissues (Giron et al., Reference Giron, Huguet, Stone and Body2016). Therefore, amino acid transformations can be an important component of biochemical mechanisms regulating galls development. Our results showed that the decarboxylation rate of some amino acids was affected by the development of N. quercusbaccarum galls on pedunculate oaks leaves. The response included an increase in ODC activity in galls and inhibition of LDC activity in galls and leaves with galls. Since ODC participates in the degradation of non-protein ornithine, and LDC breaks down essential lysine, both reactions can be beneficial to insects. On the other hand, ODC and LDC are involved in the biosynthesis of plant polyamines and its hydroxycinnamic acid amide derivatives (HCAAs). Kmieć et al. (Reference Kmieć, Rubinowska and Golan2018) conducted research on galls induced by Tetraneura ulmi (L.) (Hemiptera: Eriosomatinae) on Siberian elm leaves, detecting an increasing ODC activity during the initial period of galls formation. The activity of ODC, LDC and TyDC decreased at later stages of this process. Such responses were connected with an increase in the content of some plant amines during the initial period of galls formation and in the fully developed galls. However, these compounds were not found in mature galls. In turn, Subramanyan et al. (Reference Subramanyam, Sardesai, Minocha, Zheng, Shulke and Williams2015) studied the response of wheat (Triticum aestivum) to virulent larvae of Mayetiola destructor (Say) (Diptera: Cecidomyiidae). In that study, M. destructor induced a strong increase of polyamines (PAs) in wheat tissues and the response was connected with a higher abundance of Ta-odc, Ta-sams and Hfr-samdc transcripts encoding key enzymes of polyamine biosynthesis. The authors concluded that polyamines did not participate in wheat defense mechanisms induced by M. destructor larvae, because the response was typical for the susceptible wheat genotype and not the resistant one. However, our earlier studies showed that the excessive accumulation of these compounds under R. padi infestation was characteristic of more resistant triticale cultivar (Sempruch et al., Reference Sempruch, Horbowicz, Kosson and Leszczyński2012a). In addition, it was proved that these biomolecules at 1 and 10 mM concentrations disturbed the feeding behavior of bird cherry-oat aphid on triticale seedlings (Sempruch et al., Reference Sempruch, Goławska, Osiński, Leszczyński, Czerniewicz, Sytykiewicz and Matok2016). Therefore, the role of plant amines in their interactions with herbivorous insects is not clear and requires further research.
Recent studies indicated rather small and insignificant differences in the photosynthetic pigment contents (chlorophylls and carotenoids), the ratios of chlorophyll a/b and carotenoids/chlorophylls as well as anthocyanin contents in both the control leaves and in the leaves with galls. In turn, the content of pigments and their ratios in gall tissues was extremely low. Several studies documented a similar pattern in galls induced by Nothotrioza myrtoidis Burck (Hemiptera: Psylloidea) on Psidum myrtoides (Carneiro et al., Reference Carneiro, Castro and Isaias2014) and cecidomyiid galls on Aspidosperma australe and A. spruceanum (Oliveira et al., Reference Oliveira, Isaias, Moreira, Magalhães and Lemos-Filho2011). The decrease of assimilatory pigments in galls of Cynipidae species can be explained by the fact that chloroplast location in tissues is usually restricted to external cortical layers (Patra et al., Reference Patra, Bera and Mehltreter2010). Nevertheless, chlorophyll loss is also associated with medium- and long-term abiotic and biotic stresses (Barry & Newnham, Reference Barry and Newnham2012) and gall formation can be considered to the stress agent. On the other hand, the accumulation of anthocyanins is induced under environmental stresses in plants (Ramakrishna & Ravishankar, Reference Ramakrishna and Ravishankar2011). In our experiment, the level of this pigment in gall tissues was low, which did not match the results of Yang et al. (Reference Yang, Yang, Hsu and Jane2003), who did not detect anthocyanins in infected leaves, while galls contained significant amounts of these pigments.
According to Dorchin et al. (Reference Dorchin, Cramer and Hoffman2006), gall-inducing species that feed on specialized nutritive tissues (e.g., cynipid wasps) cause less damage to surrounding tissues, thus it is more likely that assimilation rates will increase. However, other findings (Aldea et al., Reference Aldea, Hamilton, Resti, Zangerl, Berenbaum, Frank and DeLucia2006; Kot et al., Reference Kot, Rubinowska and Michałek2018b) and the results presented here showed no evidence to support this hypothesis. Galling process of N. quercusbaccarum on oaks leaves caused a decrease of the maximum quantum yield of photosystem II (F v/F m), which characterizes the functional state of PSII in dark-adapted leaves. This probably could be caused by physical damages to the reaction centers in the photosystem (Huang et al., Reference Huang, Chou, Chang and Yang2014a). Moreover, chlorophyll fluorescence depends on the number of galls, and it declines with the increasing gall number (Huang et al., Reference Huang, Huang, Chou, Lin, Chen, Chen, Chang and Yang2014b; Kmieć et al., Reference Kmieć, Rubinowska and Golan2018). Changes in F v/F m values can sometimes be misinterpreted. Vassilev & Manolov (Reference Vassilev and Manolov1999) recorded slight changes of F v/F m together with a stronger decrease of maximal fluorescence (F m) values and variable fluorescence (F v = F m – F 0) values. Therefore, all fluorescence parameters mentioned above should also be analyzed (Huang et al., Reference Huang, Zhang, Zhang, Lu, Huang and Li2013). In this study, a significant decrease of F m and F 0 values was observed, which proved, together with our previous study (Kot et al., Reference Kot, Rubinowska and Michałek2018b), that Cynipidae species that induce galls on leaves exerted a negative effect on photosynthetic rates. The occurrence of N. quercusbaccarum galls also resulted in a strong down-regulation of the effective quantum yield of photosystem II photochemistry (Y), i.e., the measure of the actual photochemical efficiency of PSII in illuminated leaves (Vassilev & Manolov, Reference Vassilev and Manolov1999). The results of previous studies demonstrated that insect feeding had various effects on this parameter, depending on insect feeding modes or duration of infestation. For example, the presence of aphid galls (Kmieć et al., Reference Kmieć, Rubinowska and Golan2018), midge galls (Nabity et al., Reference Nabity, Hillstrom, Lindroth and DeLucia2012), cynipid wasp galls (Kot et al., Reference Kot, Rubinowska and Michałek2018b) as well as aphids on barley (Gutsche et al., Reference Gutsche, Heng-Moss, Higley, Sarath and Mornhinweg2009) caused a significant reduction of Y values. On the other hand, plants infested by scale insects showed increased (Retuerto et al., Reference Retuerto, Fernandez-Lema, Rodriguez and Obeso2004) or decreased (Kmieć et al., Reference Kmieć, Kot, Golan, Górska-Drabik, Łagowska, Rubinowska and Michałek2016) values of this parameter.
Fluorescence quenching parameters, such as photochemical (q P) and non-photochemical quenching (q N) coefficients are commonly used as indicators of plant stress exposed to insect feeding (Gutsche et al., Reference Gutsche, Heng-Moss, Higley, Sarath and Mornhinweg2009; Golan et al., Reference Golan, Rubinowska, Kmieć, Kot, Górska-Drabik, Łagowska and Michałek2015; Kmieć et al., Reference Kmieć, Kot, Golan, Górska-Drabik, Łagowska, Rubinowska and Michałek2016). However, it seems that q N is a much more sensitive indicator of a stress response than q P (Juneau et al., Reference Juneau, Green and Harrison2005). In our study, the presence of N. quercusbaccarum galls on oak leaves reduced q P and stimulated q N. This result was consistent with the findings of Kmieć et al. (Reference Kmieć, Rubinowska and Golan2018). The decrease of ΦPSII and q P indicates that less of the absorbed photon-energy captured by open PSII reaction centers is used in the photochemical reaction (Yang et al., Reference Yang, Wang, Wei, Hikosaka and Goto2009).
In conclusion, the development of N. quercusbaccarum galls on pedunculate oak leaves has a negative effect on host physiology, related to the increase in lipid peroxidation, disruption of cell membrane integrity and disturbance of photosynthesis. During this process, the antioxidant potential of the host plant is also reduced, as a result of a decrease in the content of hydrogen peroxide and low-molecular and enzymatic antioxidants. The activity of ODC and LDC changes in gall tissues, which may affect the content of amino acids and their decarboxylation products – amines.
Acknowledgements
The study was financed by the University of Life Sciences in Lublin (Project no. OKE/DS/2 in 2013–2017) and Siedlce University of Natural Sciences and Humanities (Project no. 245/08/S).
