The in fluence of Tetranychus cinnabarinus-induced plant defense responses on Aphis gossypii development

MA Guang-min, SHl Xue-yan KANG Zhi-jiao GAO Xi-wu

1 Department of Entomology, China Agricultural University, Beijing 100193, P.R.China

2 College of Agriculture, Liaocheng University, Liaocheng 252000, P.R.China

RESEARCH ARTICLE

The in fluence of Tetranychus cinnabarinus-induced plant defense responses on Aphis gossypii development

MA Guang-min1,2, SHl Xue-yan1, KANG Zhi-jiao1, GAO Xi-wu1

1 Department of Entomology, China Agricultural University, Beijing 100193, P.R.China

2 College of Agriculture, Liaocheng University, Liaocheng 252000, P.R.China

Carmine spider mites (Tetranychus cinnabarinus) and cotton aphids (Aphis gossypii) are both serious pests of cotton, and cause reductions in yields of this key agricultural crop. In order to gain insights into how plant defense responses induced by one herbivore species affect the behavior and performance of another, we examined how infestation with T. cinnabarinus in fluences the development of A. gossypii using cotton as a model. In this study, we measured the activities of several important biochemical markers and secondary metabolites in the leaves of cotton seedlings responding to infestation by T. cinnabarinus. Furthermore, the in fluences of T. cinnabarinus infestation on the development of A. gossypii in cotton were also examined. Our data showed that the activities of several key defense enzymes, including phenylalanine ammonia-lyase(PAL), peroxidase (POD), lipoxygenase (LOX), and polyphenol oxidase (PPO), were substantially increased in cotton seedlings responding to spider mite infestation. Further, the contents of gossypol and condensed tannins, key defensive compounds, were signi ficantly enhanced in leaves of cotton seedlings following T. cinnabarinus infestation. Moreover, the T. cinnabarinus-induced production of defense enzymes and secondary metabolites was correlated with infestation density.The developmental periods of A. gossypii on cotton seedling leaves infested with T. cinnabarinus at densities of 10 and 15 individuals cm–2were 1.16 and 1.18 times that of control, respectively. Meanwhile, the mean relative growth rates of A. gossypii on cotton leaves infested with T. cinnabarinus at densities of 8, 10 and 15 individuals cm–2were signi ficantly reduced. Therefore, these data suggested that the developmental periods of A. gossypii were signi ficantly lengthened and the mean relative growth rates were markedly reduced when cotton aphids were reared on plants infested with high densities of spider mites. This research sheds light on the role that inducible defense responses played in plant-mediated interspeci fic interactions between T. cinnabarinus and A. gossypii.

inducible defense, cotton, Tetranychus cinnabarinus, Aphis gossypii, interaction

1. lntroduction

Inducible defense mechanism is an important strategy used by plants to counter the damaging effects of herbivorous invaders. The biochemical and molecular mechanisms used to defend against the herbivores are highly dynamic,and are classi fied as either direct or indirect, based on how they affect the herbivores. Direct defense responses in flu-ence the herbivore by directly interacting with its biological processes. For example, morphological structures, such as plant cell walls, trichomes, and waxes form physical obstacles that directly affect the abilities of insects to feed (War et al. 2012). In addition to physical obstructions, plants also produce secondary metabolites and toxic, defensive proteins that act to reduce the palatability and digestibility of plants to their invaders. In contrast, indirect defense mechanisms enact in fluence over herbivores by promotion and/or attraction of their natural enemies (Dicke et al. 1999). Inducible direct defense responses are often initiated by damage to the plant, resulting in the induction and production of a range of secondary metabolites and defense-related enzymes that disrupt nutrient uptake by herbivores. These include inhibitors of insect digestive enzymes, proteases, lectins, amino acid deaminases and oxidases, as well as protective proteins that adversely affect herbivore physiology and metabolism (Chen et al. 2009). Inducible indirect defensive responses usually involve the release of plant volatiles that function to attract natural enemies of herbivores (Sabelis et al. 1999).

Phenylalanine ammonia-lyases (PALs), lipoxygenases(LOXs), polyphenol oxidases (PPOs), and peroxidases(PODs) are important biochemical markers for plant defense responses to herbivore infestation (Han et al. 2009). PAL is a key enzyme in the phenylpropanoid pathway, which is an important secondary metabolic pathway in higher plants(Koukol and Conn 1961; Karban and Myers 1989; MacDonald and D’Cunha 2007). LOXs, PPOs, and PODs also play important roles in plant defense against pests (Heitz et al.1997; Chen et al. 2000; Chaman et al. 2001). LOXs are dioxygenase that catalyze the hydroperoxidation of polyunsaturated fatty acids forming fatty acid hydroperoxides that are degraded into unstable, highly reactive, toxic products.Further, the products catalyzed by LOX activity themselves,such as trauma hormones including the traumatin and jasmonic acid, may act as direct deterrents in response to mechanical trauma, pathogenesis and pests (Qin and Gao 2005; Sha et al. 2015). PPOs are plant metallo-enzymes that catalyze the formation of quinones from various phenolic precursors, leading to the formation of black or brown pigments that affect the quality and nutritional value of the plants under attack (Mayer 1987; Sha et al. 2015). PODs are monomeric hemoproteins involved in the synthesis of lignin and suberin, which form the main physical barriers against herbivore invasion (Ralph et al. 2004). Several studies have shown that plants increased the production of these enzymes in response to infestation with spider mites and other insects (Qin et al. 2005; Spence et al. 2007; Duan et al. 2012; ?wi?tek et al. 2014; Sha et al. 2015).

Gossypol, a dimeric sesquiterpenoid aldehyde, and condensed tannins are the two key secondary metabolites that function as defensive compounds against herbivores due to their cytotoxicity (Williams et al. 2011; Zhou et al. 2013).The gossypol gland is responsible for gossypol biosynthesis and secretion, and both gland density and terpenoid content have been shown to be increased in young and developing leaves by herbivore attacking (Hagenbucher et al. 2013a).Condensed tannin content was dramatically increased in Lygus lucorum-infested cotton, thereby enhancing L. lucorum resistance (Liu et al. 2013).

Tetranychus cinnabarinus (Boisduval) (Acarina:Tetranychidae) and Aphis gossypii (Glover) (Homoptera:Aphididae) are both economically important pests that attack cotton seedlings, sharing the same ecological niche. Interestingly, in cases where herbivores share the same niche,automatic competition often occurs (Inbar et al. 1999b).Further, many studies have shown that plant-mediated interactions among herbivores plays an important role in interspecies competition. Changes in expression of defense enzymes and secondary metabolites induced by herbivory not only affect the herbivores that induce the original response, but also those in proximity sharing the same ecological niche (Inbar et al. 1999a). For example,jasmonate and other defensive metabolites were increased in caterpillar-damaged Asclepias syriaca, leading to a 50%decrease in milkweed aphid (Aphis nerii) population (Ali et al.2014). Further, the population density and larval survival of Liriomyza trifolii was reduced by 41 and 26.5% in Bemisia argentifolii-infested tomato plants, respectively (Inbar et al.1999b). During the course of our work, we observed that populations of A. gossypii were signi ficantly reduced in T. cinnabarinus-infested cotton seedlings when compared to the uninfested controls. Further, we propose that this phenomenon may be attributed to the T. cinnabarinus-induced production of defensive enzymes and compounds in cotton seedlings.

2. Materials and methods

2.1. Plant and insect growth conditions

Cotton plants were grown in plastic pots filled with vermiculite and nutrient soil mix in a climate controlled growth chamber (16 h L:8 h D, (70–80)% RH, (28±2)°C). Thirty-dayold cotton plants with 4 true leaves were used for all experiments. T. cinnabarinus and A. gossypii were collected from cotton fields maintained at China Agricultural University and reared separately in a climate controlled growth chambers(16 h L:8 h D, (50–60)% RH, (28±2)°C) on mung bean and cotton seedlings, respectively.

2.2. Reagents

L-Phenylalanine, catechol, and guaiacol were purchased from Shanghai Chemical Company, China. Linoleic acid,gossypol, β-mercaptoethanol, and phenylmethylsulfonyl fluoride (PMSF) were purchased from Sigma-Aldrich, USA.Bovine serum albumin (BSA) was purchased from Beijing Tongzheng Biological Company, China. Catechin and Coomassie brilliant blue G-250 were acquired from Fluka,USA. Analytical grade polyvinyl pyrrolidone (PVP-40) was obtained from Sinopharm Chemical Reagent Co., China.HPLC grade acetonitrile and methanol were purchased from Fisher Scienti fic, USA. Ultrapure water was prepared using a Milli-Q Academic Water Puri fication System (Millipore, USA).All other reagents used in this study were analytical grade and purchased from Sinopharm Chemical Reagent Co., China.

2.3. Treatment of cotton seedlings with T. cinnabarinus

Morphologically uniform cotton seedlings were selected,and leaf areas of each third true leaf were determined. The third true leaf of individual cotton seedlings was inoculated with female adult mites at different densities (5, 8, 10, and 15 individuals cm–2), then enclosed in modi fied Petri dishes to prevent escape of mites. The fourth true leaves were detached 24, 48, and 72 h after inoculation to determine enzymatic activities and secondary metabolite contents.Uninoculated cotton seedlings were used as controls, and each treatment was replicated three times.

2.4. ln fluence of T. cinnabarinus infestation herbivory on A. gossypii development

The third true leaves of several individual cotton seedlings were innoculated with female adult mites of T. cinnabarinus at increasing population densities (5, 8, 10, and 15 individuals cm–2). After 12 h, 10 first instar nymphs of A. gossypii were placed on the fourth true leaf, and the inoculated leaves were enclosed in special Petri dishes to prevent escape.The time of adult aphis emergence and the mean number of offspring produced per surviving aphid were recorded.Cotton seedlings inoculated with aphids alone were used as controls. Each treatment was replicated six times.

To examine the in fluence of T. cinnabarinus infestation on the mean relative growth rate (MRGR) of A. gossypii, the following experiment was used. First, 100 first instar nymphs of A. gossypii were weighed (initial weight). Next, female adult mites were inoculated onto the third true leaf of each cotton seedling at different densities as described above. At 12 h after T. cinnabarinus infestation, the fourth true leaf of each cotton seedling was inoculated with 30 first instar nymphs of A. gossypii. After 6 days, the aphids inoculated on cotton seedlings were weighed again ( final weight). The MRGR of cotton aphids was calculated for each treatment according to Adams and van Emden (1972) and Hodge et al. (2005):

MRGR=[loge(Final weight)-loge(Initial weight)]/6

Cotton aphids inoculated on cotton seedling without T. cinnabarinus infestation served as controls. Each treatment was replicated six times.

2.5. PAL, LOX, PPO, and POD activity assays

PAL activity was measured using the method described by Dubery and Smit (1994) with minor modi fications.Brie fly, cotton leaves (1 g) were ground in liquid nitrogen and homogenized in 3 mL of extraction buffer containing 50 mmol L–1sodium borate buffer (pH 8.8), 20 mmol L–1β-mercaptoethanol, 1 mmol L–1PMSF and 0.1 g PVP-40.The homogenates were filtered through four layers of cheesecloth and centrifuged at 12 000×g for 20 min at 4°C,and the supernatants were collected and used for assays.PAL reactions contained 0.4 mL enzyme extract (0.36 mg of protein), 2.4 mL of 0.05 mmol L–1sodium borate buffer(pH 8.8), and 0.2 mL of 0.2 mmol L–1L-phenylalanine. For controls, the L-phenylalanine was replaced by 200 μL of sodium borate buffer (pH 8.8). The reactions were incubated at 37°C for 1 h, then stopped by the addition of 0.2 mL of 6 mol L–1hydrochloric acid. The absorbance of the reactions was measured at 290 nm. PAL activity is expressed as μmol h-1mg-1protein.

LOX activity was determined according to the method described by Axelrod et al. (1981) with some modi fications.Brie fly, ground leaf tissue was homogenized in 3 mL of 50 mmol L–1phosphate buffer (pH 7.5), containing 0.1% (v/v)Triton X-100 and 0.1 g PVP. Samples were centrifuged at 12 000×g for 20 min at 4°C, and the supernatants were collected and used for downstream assays. The assay mixtures contained 0.1 mL of enzyme extract (0.12 mg of protein), 2.875 mL of 50 mmol L–1glycine-hydrochloric acid buffer (pH 3.5) and 25 μL of 10 mmol L–1linoleic acid sodium salt. LOX activity was determined by monitoring the increase in absorbance at 234 nm for 2 min at room temperature and is expressed as μmol min-1mg-1protein.

PPO activity was determined according to the method described by Chen et al. (2000) with some modi fications.Leaf samples were homogenized in 3 mL of 50 mmol L–1phosphate buffer solution at pH 7.0, containing 1 mmol L–1ethylenediaminetetraacetic acid (EDTA) and 0.1 g PVP.Then, samples were centrifuged at 12 000×g for 20 min at 4°C, and the supernatants were collected and used for analysis. PPO assays consisted of 0.6 mL PBS, 0.4 mL of 0.1 mol L–1catechol, and 0.15 mL of the enzyme extract(0.18 mg of protein). Absorbance was measured at 420 nm.PPO activity was determined by monitoring the increase in absorbance at 420 nm for 2 min at room temperature.PPO activity was presented as ΔOD420nmmin-1mg-1protein.

POD activity assays were carried out as previously described by Patykowski et al. (1988) with some modi fications. Brie fly, the reactions were initiated by the addition of 20 μL of 50 mmol L–1hydrogen peroxide into a reaction comprised of 1.345 mL of 50 mmol L–1phosphate buffer solution (pH 7.5) containing 20 μL of 20 mmol L–1guaiacol,and 0.1 mL of enzyme extract (0.025 mg of protein) prepared as described above. POD activity was determined by monitoring the increase in absorbance at 470 nm for 2 min at room temperature. The results are presented as ΔOD470nmmin-1mg-1protein.

2.6. Determination of total gossypol and condensed tannin content

Leaf samples were flash frozen in liquid nitrogen and lyophilized for 24 h. Dry leaf samples were ground to powder,sieved through 80-muse sieves, and stored at –80°C.

Gossypol content was determined according to the method described by Stipanovic et al. (1988) with minor modi fications. A total of 50 mg of leaf powder and 7 mL of a solution of hexane:ethyl acetate (3:1,v/v) were combined in a 10-mL centrifuge tube. The mixture was sonicated for 40 min, then centrifuged at 5 369×g for 10 min. The supernatant was collected and dried under N2gas. The resulting residue was re-dissolved in 2 mL of mixture of isopropanol:acetonitrile:water:ethyl acetate (35:21:39:5, v/v) and passed through a 0.45-μm filter.

Filtered samples were analyzed on an Agilent 1100 liquid chromatograph combined with a quaternary pump, online degasser, and a diode array detector (DAD) (Agilent Co., USA).The filtered sample of 20 μL was injected onto a ZORBAX SB-C18column (4.6 mm×250 mm, 5 μm) (Agilent Co., USA)and run isocratically at a flow rate of 1 mL min–1at a constant temperature of 55°C in mixture of ethanol:methanol:isopropanol:acetonitrile:water:ethyl acetate:dimethylformamide:phosphoric acid (16.7:4.6:12.1:20.2:37.4:3.8:5.1:0.1,v/v). Gossypol was monitored at UV wavelength of 272 nm.Gossypol concentration was quanti fied using external standards. Calibration curves were produced by analysis of solutions containing 12.5, 6.25, 3.12, 1.56, and 0.78 μg mL–1gossypol, respectively.

Condensed tannin content was determined according to the methods described by Hagerman and Butler (1989) and Wu and Guo (2000) with some modi fications. Brie fly, 50 mg of leaf powder was homogenized in 8 mL of 80% (v/v)aqueous methanol in a 10-mL centrifuge tube. The mixture was sonicated in a water bath sonicator for 40 min, then centrifuged at 5 369×g for 15 min. The supernatant was collected and brought to a constant volume of 10 mL with 80%(v/v) aqueous methanol. For reactions, 0.5 mL of extract,3 mL of 4% vanillin and 1.5 mL of 12 mol L–1hydrochloric acid were combined in a test tube, covered with aluminum foil, and incubated for 20 min at 25°C. The absorbance of the reaction mixture was measured at 510 nm. Catechin was used as an external standard. Calibration curves were produced by analysis of solutions containing 200, 160, 128,64, 32, 16, and 8 μg mL–1of catechin, respectively.

2.7. Protein determination

Protein content of enzyme preparation was determined by the dye binding method of Bradford (1976) with bovine serum albumin as the standard.

2.8. Statistical analysis

ANOVA for analysis of variance was run using SPSS 19.0 statistical software (IBM Co., USA). Means were separated by LSD multiple range tests at P<0.05.

3. Results

3.1. The effects of T. cinnabarinus infestation on cotton defense enzymes

Fig. 1 Phenylalanine ammonia-lyase (PAL) activity in Tetranychus cinnabarinus-infested cotton. Data are means±SD.Different letters above error bars indicate a signi ficant difference at the 95% con fidence interval (ANOVA, P<0.05).

PAL, LOX, PPO, and POD are important cotton defense enzymes and show changes in response to T. cinnabarinus infestation. The activities of PAL and PPO in cotton leaves at 24 h post infestation (HPI) by T. cinnabarinus at densities of 10 and 15 individuals cm–2were found to be signi ficantly higher than those of controls (P<0.05) (Figs.1 and 2). However, no signi ficant differences were observed between the control and infested cotton leaves when the inoculation density of T. cinnabarinus was less than 10 individuals cm–2(P>0.05) (Figs. 1 and 2). The activity of LOX in T. cinnabarinus infested cotton leaves 24 HPI at a density of 15 individuals cm–2was signi ficantly higher than the controls(P<0.05), though the activities of LOX in T. cinnabarinusinfested cotton leaves at densities of 5, 8, and 10 individuals cm–2were not signi ficantly different than that of the control plant (P>0.05) (Fig. 3). The activities of POD in cotton leaves infested with T. cinnabarinus at densities of 5, 8, 10,and 15 individuals cm–2were signi ficantly higher than that in control (P<0.05). Furthermore, POD activity was also shown to be signi ficantly higher when plants were innoculated with at least 15 individuals cm–2(P<0.05) when compared to plants infested with lower population densities (Fig. 4).

Fig. 2 Polyphenol oxidase (PPO) activity in Tetranychus cinnabarinus-infested cotton. Data are means±SD. Letters above error bars indicate signi ficant differences at the 95%con fidence interval (ANOVA, P<0.05).

Fig. 3 Lipoxygenase (LOX) activity in Tetranychus cinnabarinusinfested cotton. Data are means±SD. Letters above error bars indicate signi ficant differences at the 95% con fidence interval(ANOVA, P<0.05).

At 48 and 72 h after T. cinnabarinus infestation, with all four population densities, the activities of PAL, LOX, PPO,and POD were found to be signi ficantly higher than that of the controls (P<0.05) (Figs. 1–4). After infestation with T. cinnabarinus at a density of 8 individuals cm–2for 72 h,the activity of PAL reached its highest level, which was 5.45-fold higher than that of the controls (Fig. 1). The activity of PPO reached its highest level at 48 HPI for the treatment of T. cinnabarinus at a density of 10 individuals cm–2(Fig. 2).The activities of LOX and POD reached their highest levels at 72 HPI with T. cinnabarinus at a density of 5 individuals cm–2, 2.15- and 2.16-fold higher than that of the controls,respectively (Figs. 3 and 4).

Our data showed that the inducible defense responses of cotton triggered by infestation with high densities of T.cinnabarinus (10 and 15 individuals cm–2) occurs both rapidly and intensely. When plants were infested at densities of 10 and 15 individuals cm–2, activities of the four defense enzymes (PAL, PPO, LOX, and POD) were found to be signi ficantly higher than those of controls (P<0.05) at 24 HPI,with the exception of LOX at 10 individuals cm–2. Moreover,the four defense enzyme activities reached the highest levels at 48 or 72 HPI. However, the activities of PAL, PPO,and LOX in cotton seedlings infested with low densities of T. cinnabarinus (5 and 8 individual cm–2) were not signi ficantly different than that of the controls (P>0.05) at 24 HPI. The observed changes in activities of the four defense enzymes may indicate that cotton has the ability to self-adjust the tradeoffs between the cost and bene fits based on the degree of T. cinnabarinus infestation.

3.2. The effects of T. cinnabarinus infestation on the production of gossypol and condensed tannins in cotton

Fig. 4 Peroxidase (POD) activity in Tetranychus cinnabarinusinfested cotton. Data are means±SD. Letters above error bars indicate signi ficant differences at the 95% con fidence interval(ANOVA, P<0.05).

Gossypol and condensed tannins are key defensive sec-ondary metabolites produced by cotton. Our study showed that at 24, 48, and 72 HPI, the contents of gossypol and condensed tannins in the infested cotton leaves were signi ficantly higher than that in controls (P<0.05) for plants infested with T. cinnabarinus population densities of 8, 10 and 15 individuals cm–2. However, at 24 HPI, no signi ficant differences were observed between gossypol and condensed tannin contents in controls and plants treated at a density of 5 individuals cm–2(P>0.05) (Figs. 5 and 6).

Gossypol content in cotton leaves reached its highest level at 24 HPI with T. cinnabarinus at a density of 15 individuals cm–2, which was 1.44 times higher than the controls(Fig. 5). Condensed tannin content reached the highest level, 2.7 times higher relative to the controls (Fig. 6), at 48 HPI at the highest treatment density of 15 individuals cm–2.

3.3. ln fluence of T. cinnabarinus-infested cotton on A. gossypii development

Plant-mediated interactions among herbivores is an important part of interspeci fic competition. In our work, we have shown that the T. cinnabarinus infestation likely exerts a delayed developmental effect on A. gossypii with increasing intensities of T. cinnabarinus infestation (Fig. 7). The developmental periods were signi ficantly longer than those of controls (P<0.05) for A. gossypii growing on cotton infested with T. cinnabarinus at densities of 10 and 15 individuals cm–2. However, no signi ficant difference was observed between control and infested plants at infestation densities of 5 and 8 individuals cm–2(P>0.05) (Fig. 7).

Fig. 5 The gossypol content in Tetranychus cinnabarinusinfested cotton. Data are means±SD. Letters above error bars indicate signi ficant differences at the 95% con fidence interval(ANOVA, P<0.05).

The mean relative growth rates of A. gossypii innoculated onto cotton leaves infested by T. cinnabarinus at 8, 10, and 15 individual cm–2was found to be signi ficantly decreased in comparison with controls (P<0.05). Interestingly, no differences were observed when the infestation density was reduced to 5 individuals cm–2(P>0.05) (Fig. 8). In addition,no signi ficant changes were observed for the mean number of offspring produced per surviving aphid living on cotton leaves infested with T. cinnabarinus at all four population densities (data not shown).

4. Discussion

Fig. 6 The condensed tannins content in Tetranychus cinnabarinus-infested cotton. Data are means±SD. Letters above error bars indicate signi ficant differences at the 95%con fidence interval (ANOVA, P<0.05).

Fig. 7 Development periods of Aphis gossypii in Tetranychus cinnabarinus-infesed cotton. Data are means±SD. Letters above error bars indicate signi ficant differences at the 95%con fidence interval (ANOVA, P<0.05).

Fig. 8 Mean relative growth rate of Aphis gossypii in Tetranychus cinnabarinus-infested cotton. Data are means±SD. Letters above error bars indicate signi ficant differences at the 95%con fidence interval (ANOVA, P<0.05).

When plants initiated the induced defense responses,several biochemistry pathways were activated and that triggered increases in both defense enzymatic activities and the production of secondary metabolites (Karban and Myers 1989). The main plant defense response enzymes, PAL,LOX, PPO, and POD, have been shown to exhibit defensive roles in response to herbivores. PAL and LOX are involved in the salicylic acid/jasmonic acid signaling pathways, and play important roles in plant inducible defense responses(Kessleret al. 2004; Qin and Gao 2005). PPO participates in the biosynthesis of quinone, which can bind plant proteins, resulting in a reduction of the nutrients available to herbivores (Mayer 1987). POD is involved in the synthesis of lignin and suberin, which form the main physical barriers against herbivore invasion (Ralph et al. 2004).

Previous studies have shown that mite infestation induced plant defense responses. The activities of PPO and POD in tomato leaves increased after infestation with the tomato russet mite, Aculops lycopersici (Wang et al. 2008).LOX activity has been shown to be increased in soybean leaves infested by the two-spotted spider mite, Tetranychus urticae (Hildebrand et al. 1989). Further, Roseleen and Ramaraju (2010) reported that herbivory by the two-spotted spider mite, T. urticae, led to increased POD, PPO, and PAL activities in okra. Activities of the oxidative enzymes,POD and PPO, also increased in tomato plants during mite infestation (Kielkiewicz 2001).

Our results showed that the activities of four defense enzymes, PAL, LOX, PPO and POD, were increased signi ficantly in cotton leaves during T. cinnabarinus infestation.Despite the observation that all the four defensive enzymes responded to T. cinnabarinus herbivory and played their corresponding defensive roles, PAL activity changed both rapidly and intensely. PAL activity in cotton leaves was signi ficantly upregulated at 24 HPI, and kept increasing from 48 to 72 HPI. Moreover, at 72 HPI of T. cinnabarinus at 8 individuals cm–2, the highest level of PAL activity was observed, showing a 5.45-fold higher than that in controls(Fig. 1). These data suggested that PAL plays an important role in T. cinnabarinus defense. These changes in PAL activity may be related to the sensitivity of PAL to insect herbivory, because PAL is a key enzyme in pathway used by plants to produce phenylpropanoids and phenols, which are known to be toxic secondary metabolites for defending against environmental stresses (Sha et al. 2015).

Gossypol and condensed tannins are two of the major defensive secondary metabolites of cotton. The binding of insect digestive enzymes to gossypol or tannins can affect insect feeding, digestion, growth, and development (Wang et al. 1993; Wang 1997). Our results showed that the gossypol and condensed tannin contents were increased signi ficantly in cotton leaves during T. cinnabarinus infestation.

These combined physiological changes, including the activities of four defense enzymes, as well as the contents of gossypol and condensed tannins induced by T. cinnabarinus infestation, are characteristic of a successful, induced defense response in cotton.

The competitive advantage of one herbivore over another is re flected not only in its successful acquisition of more nutrients and space, but also in its ability to induce biochemical changes (such as activation of defense enzymes and biosynthesis of secondary metabolites) during plant colonization that results in a restraining effect on other herbivores. Hagenbucher et al. (2013b) reported that terpenoids increased signi ficantly in caterpillar-induced cotton,leading to a 40% reduction in A. gossypii abundance. On A. lineatus-infested cotton, terpenoid levels were enhanced and the growth rate of Spodoptera exigua larvae was shown to be reduced (Bezemer et al. 2003).

Similarly, our results suggested that T. cinnabarinus infestation to cotton affected A. gossypii development on cotton. Our results indicated that the developmental periods of A. gossypii signi ficantly increased, while the MRGR was signi ficantly reduced in cotton infested with T. cinnabarinus at population densities of 10 and 15 individuals cm–2. After T. cinnabarinus infestation, the rapid enhancement of gossypol and condensed tannin content, combined with increased activities of several defense enzymes in cotton, may cause a competitive disadvantage for A. gossypii. Notably, this phenomenon was not observed in cotton colonized by T. cinnabarinus at low population density (5 individuals cm–2). Moreover, no signi ficant changes were observed for A. gossypii reproduction on cotton leaves infested with T. cinnabarinus at all four population densities, when compared with that of control cotton plants (data not shown).

According to the optimal defense theory, plants are able to self-adjust the trade-offs between costs and bene fits of defense. Defensive resources are allocated, based on priority, to the most valuable plant tissues because the costs of plant defense (synthesis, storage, and maintenance of defensive secondary metabolites) are high compared to basic metabolism (Zangerl and Bazzaz 1992; Gershenzon 1994; Wang et al. 1994; Gianoli and Niemeyer 1998). Our study showed that the inducible defense responses of cotton to low-density (5 individual cm–2) T. cinnabarinus infestation was low during early infestation. We hypothesized that cotton can adjust its defense levels relative to the severity of T. cinnabarinus infestation. Thus, cotton plants showed no signi ficant changes in their inducible defense responses when low densities (5 individuals cm–2) of T. cinnabarinus were used to inoculate plants, which also exhibited no observable effect on the development of A. gossypii.

5. Conclusion

The increased activities of the four defense enzymes,combined with elevated gossypol and condensed tannin contents in T. cinnabarinus-infested cotton leaves demonstrated an induced defense response to T. cinnabarinus herbivory. Moreover, the data indicated cotton responded differently based on the severity of infestation. Further,T. cinnabarinus infestation affected A. gossypii development on cotton. These data indicated that inducible plant defense responses mediated competitive interactions between herbivores sharing the same ecological niche.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (31672045).

Adams J B, van Emden H F. 1972. The biological properties of aphids and their host plant relationships. In: van Emden H F, ed., Aphid Technology. Academic Press, London. pp.47–104.

Ali J G, Agrawal A A. 2014. Asymmetry of plant-mediated interactions between specialist aphids and caterpillars on two milkweeds. Functional Ecology,28, 1404–1412.

Axelrod B, Cheesbrough T M, Laakso S. 1981. Lipoxygenase from soybeans. Methods in Enzymology,7, 411–451.

Bezemer T M, Wagenaar R, Van Dam N M, W?ckers F L.2003. Interactions between above- and belowground insect herbivores as mediated by the plant defense system.OIKOS,101, 555–562.

Bradford M M. 1976. A rapid and sensitive method for the quanti fication of microgram quantities of protein using the principle of protein dye binding. Analytical Biochemistry,72, 248–254.

Chaman M E, Corcuera L J, Zuniga G E, Cardemil L, Argandona V H. 2001. Induction of soluble and cell wall peroxidases by aphid infestation in barley. Journal of Agricultural and Food Chemistry,49, 2249–2253.

Chen C, Bélanger R, Benhamou N, Paulitz T C. 2000. Defense enzymes induced in cucumber roots by treatment with plant growth-promoting rhizobacteria (PGPR) and Pythium aphanidermatum. Physiological and Molecular Plant Pathology,56, 13–23.

Chen M S, Wu J X, Zhang G H. 2009. Inducible direct plant defense against insect herbivores. Chinese Bulletin of Entomology,46, 175–186. (in Chinese)

Dicke M, Gols R, Ludeking D, Posthumus M A. 1999. Jasmonic acid and herbivory differentially induce carnivore-attracting plant volatiles in lima bean plants. Journal of Chemical Ecology,25, 1907–1922.

Duan W C, Duan L Q, Li H P, Feng S J, Zhang Y, Zhang L N.2012. Defense responses in wolfberry (Lycium barbarum)induced by exogenous jasmonic acid and gall mite Aceria pallida (Acari: Eriophyidae). Acta Entomologica Sinica,55,804–809. (in Chinese)

Dubery I A, Smit F. 1994. Phenylalanine ammonia-lyase from cotton (Gossypium hirsutum) hypocotyls: Properties of the enzyme induced by a Verticillium dahliae phytotoxin.Biochimica et Biophysica Acta,1207, 24–30.

Gershenzon J. 1994. Metabolic costs of terpenoid accumulation in higher plants. Journal of Chemical Ecology,20, 1281–1328.

Gianoli E, Niemeyer H M. 1998. Allocation of herbivory-induced hydroxamic acids in the wild wheat Triticum uniaristatum.Chemoecology,8, 19–23.

Hagenbucher S, Olson D M, Ruberson J R, Wackers F L, Romeis J. 2013a. Resistance mechanisms against arthropod herbivores in cotton and their interactions with natural enemies. Critical Reviews in Plant Sciences,32,458–482.

Hagenbucher S, W?ckers F L, Wettestein F E, Olson D M, Ruberson J R, Romeis J. 2013b. Pest tradeoffs in technology: Reduced damage by caterpillars in Bt cotton bene fits aphids. Proceedings of the Royal Society (B:Biological Sciences),280, 20130042.

Hagerman A E, Butler L G. 1989. Choosing appropriate methods and standards for assaying tannin. Journal of Chemical Ecology,15, 1795–1810.

Han Y, Wang Y, Bi J, Yang X, Huang Y, Zhao X, Hu Y, Cai Q N.2009. Constitutive and induced activities of defense-related enzymes in aphid-resistant and aphid-susceptible cultivars of wheat. Journal Chemical Ecology,35, 176–182.

Heitz T, Bergey D R, Ryan C A. 1997. A gene encoding a chloroplast targeted lipoxygenase in tomato leaves is transiently induced by wounding, systemin, and methyl jasmonate. Plant Physiology,114, 1085–1093.

Hildebrand D F, Rodriguez J G, Legg C S, Brown G C, Bookjans G. 1989. The effects of wounding and mite infestation on soybean leaf lipoxygenase levels. Zeitschrift für Naturforschung (C: Biosciences),44, 655–659.

Hodge S, Thompson G A, Powell G. 2005. Application of DL-β-aminobutyric acid (BABA) as a root drench to legumes inhibits the growth and reproduction of the pea aphid Acyrthosiphon pisum (Hemiptera: Aphididae). Bulletin of Entomological Research,95, 449–455.

Kielkiewicz M. 2001. In fluence of carmine spider mite Tetranychus cinnabarinus Boisd. (Acarida: Tetranychidae)feeding on ethylene production and the activity of oxidative enzymes in damaged tomato plants. In: Bernini F,Nannelli R, Nuzzaci G, Lillo E E, eds., Acarid Phylogeny and Evolution: Adaptation in Mites and Ticks. Springer Netherlands, Berlin. pp. 389–392.

Inbar M, Doostdar H, Leibee G L, Mayer R T. 1999a. The role of plant rapidly induced responses in asymmetric interspeci fic interactions among insect herbivores. Journal of Chemical Ecology,25, 1961–1979.

Inbar M, Doostdar H, Mayer R T. 1999b. Effects of sessile white fly nymphs (Homoptera: Aleyrodidae) on leafchewing larvae (Lepidoptera: Noctuidae). Environmental Entomology,28, 353–357.

Karban R, Myers J H. 1989. Induced plant responses to herbivory. Annual Review of Ecology and Systematics,20, 331–348.

Kessler A, Halitschke R, Baldwin I T. 2004. Silencing the jasmonate cascade: Induced plant defenses and insect populations. Science,305, 665–668.

Koukol J, Conn E E. 1961. Metabolism of aromatic compounds in higher plants. IV. Puri fication and properties of phenylalanine deaminase of Horden vulgare. Journal of Biological Chemistry,236, 2692–2698.

Liu Y, Zhang P T, Chen B L, Guo W Q, Zhang H Y. 2013.Analysis on changes of physiological indices of young leaves infected by Lygus lucorum of transgenic Bt cotton.Cotton Science,25, 51–56. (in Chinese)

MacDonald M J, D’Cunha G B. 2007. A modern view of phenylalanine ammonia lyase. Biochemistry and Cell Biology,85, 273–282.

Mayer A M. 1987. Polyphenol oxidase in plants-recent progress.Phytochemistry,26, 11–20.

Patykowski J, Urbanek H, Kaczorowska T. 1988. Peroxidase activity in leaves of wheat cultivars differing in resistance to Erysiphe graminis DC. Journal of Phytopathology,122,126–134.

Qin Q J, Gao X W. 2005. Plant defense responses induced by insect herbivory. Acta Entomologica Sinica,48, 125–134.(in Chinese)

Qin Q J, Shi X Y, Liang P, Gao X W. 2005. Induction of phenylalanine ammonia-lyase and lipoxygenase in cotton seedlings by mechanical wounding and aphid infestation.Progress in Natural Science,15, 419–423.

Ralph J, Bunzel M, Marita J M, Hat field R D, Lu F C, Kim H,Schatz P F, Grabber J H, Steinhart H. 2004. Peroxidasedependent cross-linking reactions of p-hydroxycinnamates in plant cell walls. Phytochemistry Reviews,3, 79–96.

Roseleen S S J, Ramaraju K. 2010. Host-plant resistance in okra against the two-spotted spider mite, Tetranychus urticae Koch. Pest Management and Economic Zoology,18, 179–187.

Sabelis M W, van Baalen M, Bakker F M, Bruin J, Drukker B,Egas M, Janssen A R M, Lesna I K, Pels B, van Rijn P,Scutareanu P. 1999. The evolution of direct and indirect plant defence against herbivorous arthropods. In: Olff H,Brown V K, Drent R H, eds., Herbivores: Between Plants and Predators. Blackwell Science, Oxford. pp. 109–166.

Sha P J, Fan Y J, Wang Z C, Shi X Y. 2015. Response dynamics of three defense related enzymes in cotton leaves to the interactive stress of Helicoverpa armigera (Hübner)herbivory and omethoate application. Journal of Integrative Agriculture,14, 355–364.

Spence K O, Bicocca V T, Rosenheim J A. 2007. Friend or foe?:A plant’s induced response to an omnivore. Environmental Entomology,36, 623–630.

Stipanovic R D，Altman D W，Begin D L, Greenblatt G A, Benedict J H. 1988. Terpenoid aldehydes in upland cottons: Analysis by aniline and HPLC methods. Journal of Agricultural and Food Chemistry,36, 509–515.

?wi?tek M, Kielkiewicz M, Zagdanska B. 2014. Insect-resistant Bt-maize response to the short-term non-target mite-pest infestation and soil drought. Acta Physiologiae Plantarum,36, 2705–2715.

Wang H B, Tao Y, Jin S. 1994. Chitinase in Vicia faba leaves:induction by Aphis craccivora - A convergent plant physiological stress reaction. Chinese Journal of Applied Ecology,5, 68–71. (in Chinese)

Wang H B, Wu Q H, Gao W D. 1993. Studies on Tetranychus cinnabarinus-Solanum melongena interaction system 1.Relationship between population dynamics of red spider and tannic acid fluctuation in plant leaves. Chinese Journal of Applied Ecology,4, 174–177. (in Chinese)

Wang M Y, Wang D S, Hong X Y, Yuan Y D, Gan J H. 2008.Relationship between physical and chemical differences of tomato leaf and the resistance to Aculops lycopersici.Chinese Bulletin of Entomology,45, 904–908. (in Chinese)

Wang Z S. 1997. Effects of gossypol and tannic acid on the growth and digestion physiology of cotton bollworm larvae.Acta Phytophylacica Sinica,24, 13–18. (in Chinese)

War A R, Paulraj M G, Ahmad T, Buhroo A A, Hussain B,Ignacimuthu S, Sharma H C. 2012. Mechanisms of plant defense against insect herbivores. Plant Signaling &Behavior,7, 1306–1320.

Williams J L, Ellers-Kirk C, Orth R G, Gassmann A J, Head G,Tabashnik B E, Carrière Y. 2011. Fitness cost of resistance to Bt cotton linked with increased gossypol content in pink bollworm larvae. PLoS One,6, e21863.

Wu Y Q, Guo Y Y. 2000. Determination of tannin in cotton plant. Chinese Journal of Applied Ecology,11, 243–245.(in Chinese)

Zangerl A R, Bazzaz F A. 1992. Theory and pattern in plant defense allocation. In: Fritz R, Simms E L, eds., Plant Resistance to Herbivores and Pathogens. University of Chicago Press, Chicago. pp. 363–392.

Zhou M L, Zhang C C, Wu Y M, Tang Y X. 2013. Metabolic engineering of gossypol in cotton. Applied Microbiology and Biotechnology,97, 6159–6165.

3 March, 2017 Accepted 24 April, 2017

MA Guang-min, Tel: +86-10-62732974, E-mail: maguangmin@lcu.edu.cn; Correspondence SHI Xue-yan, Tel: +86-10-62731306, E-mail: shixueyan@cau.edu.cn; GAO Xi-wu, Tel: +86-10-62732974, E-mail: gaoxiwu@263.net.cn

? 2018 CAAS. Publishing services by Elsevier B.V. All rights reserved.

10.1016/S2095-3119(17)61666-6

Section editor WAN Fang-hao

Managing editor SUN Lu-juan