ZHANG Xiang , RUI Qiu-zhi LIANG Pan-pan WEI Chen-hua DENG Guo-qiang CHEN YuanCHEN Yuan DONG Zhao-di CHEN De-hua

1 Key Laboratory of Crop Genetics and Physiology of Jiangsu Province, Yangzhou University, Yangzhou 225009, P.R.China

2 Plant Genome Mapping Laboratory, University of Georgia, Athens 30605, USA

Abstract This study was conducted to investigate the effects of alternating high temperature on Cry1Ac protein content on Bt cotton cultivars Sikang 1 (SK-1, a conventional cultivar) and Sikang 3 (SK-3, a hybrid cultivar). In 2011 and 2012, cotton plants were subjected to high temperature treatments ranging from 32 to 40°C in climate chambers to investigate the effects of high temperature on boll shell insecticidal protein expression. The experiments showed that significant decline of the boll shell insecticidal protein was detected at temperatures higher than 38°C after 24 h. Based on the results, the cotton plants were treated with the threshold temperature of 38°C from 6:00 a.m. to 6:00 p.m. followed by a normal temperature of 27°C during the remaining night hours (DH/NN) in 2012 and 2013. These treatments were conducted at peak boll growth stage for both cultivars in study periods of 0, 4, 7, and 10 d. Temperature treatment of 32°C from 6:00 a.m. to 6:00 p.m. and 27°C in the remaining hours was set as control. The results showed that, compared with the control, after the DH/NN stress treatment applied for 7 d, the boll shell Cry1Ac protein content level was significantly decreased by 19.1 and 17.5% for SK-1 and by 15.3 and 13.7% for SK-3 in 2012 and 2013, respectively. Further analysis of nitrogen metabolic physiology under DH/NN showed that the soluble protein content and the glutamic pyruvic transaminase (GPT) activities decreased slightly after 4 d, and then decreased sharply after 7 d. The free amino acid content and the protease content increased sharply after 7 d. The changes in SK-1 were greater than those in SK-3. These results suggest that under DH/NN stress, boll shell Cry1Ac protein content decline was delayed. Reduced protein synthesis and increased protein degradation in the boll shell decreased protein content, including Bt protein, which may reduce resistance to the cotton bollworm.

Keywords: Bt cotton, alternating temperature, Cry1Ac protein, nitrogen metabolism

1. Introduction

Cotton bollworm, Heliothis armigera, a lepidopteran, is a major pest in cotton. It damages reproductive organs,mainly the square and boll (Gujar et al. 2010). Transgenic Bt (Bacillus thuringiensis) cotton can produce insecticide Cry1Ac protein that kills the cotton bollworm (Russelland Deguine 2006). The use of transgenic Bt cotton has improved farmers’ profitability, reduced insecticide use,and protected the environment without additional pollution(Zhang 2013). By 2012, the planting area for Bt cotton had reached 18.8 million hectares, which accounted for 62.7%of the total cotton planting area in the world.

Previous researchers have found that there are not only temporaland spatial variations in the expression of the Bt cotton insecticidal Cry1Ac protein (Rui et al. 2002;Bakhsh et al. 2011), but also environmental condition has an effect on its expression (Chen et al. 2005b, 2012b;Jiang et al. 2006, 2012; Luo et al. 2008; Martins et al.2008; Blaise and Kranthi 2011; Hallikeri et al. 2011).Among the environmental factors, high temperature stress is one of the most important. Many studies have shown that the temperature of the earth has been increasing,and consequently cotton-growing regions have regularly recorded extremely high temperatures (Wang et al. 2008),which directly affects the expression of the insecticidal Cry1Ac protein and the levelof Bt cotton plants’ resistance to bollworm (Chen et al. 2012a, 2014). Chen et al.(2015b) showed that the content of Cry1Ac protein in bolls decreases by 63-73% when exposed to the stress of 37°C for 48 h. A previous study also showed that a high threshold temperature from 38 to 40°C affects the expression of Cry1Ac protein in Bt cotton (Lü 2013).

In previous studies, researchers have kept the same temperature for the entire day. However, in reality, the temperature in natural environments varies during day and night, with temperatures at night usually lower than that during the day. Some researchers believe that when the external stresses on plants terminate, their physiological metabolism can recover from the effects of those stresses.Ian (2006) found that short-term abiotic stresses had no significant effects on the expression of Cry1Ac protein. In addition, the experiments reported by Adamczyk et al. (2001)also showed that there were no significant correlations between the environment and the levelof resistance to pests.However, both of these studies were conducted in a natural environment with changing environmental conditions, and the stress time during the day was less than 12 h. Therefore,we believe that the levelof Cry1Ac expression might recover during the night after the environmental stress during the day terminate. This hypothesis is supported by the observation of Chen et al. (2013), who detected the recovery of Cry1Ac protein content after temperature stress ended in transgenic Bt cotton. Therefore, investigating the effects of periodic variation of temperature on Bt cotton plants is very important to guide future Bt cotton planting practices.

The synthesis of the Bt protein and its cycle in cotton plants are also the physiological process of nitrogen metabolism.Several key enzymes affecting nitrogen metabolism such as soluble protein, glutamic-pyruvic transaminase (GPT),protease, and peptidase may also affect Bt protein content(Steward et al. 1965; Dong et al. 2007). Therefore, it is important to investigate the relationships between boll toxin levels and nitrogen metabolism with the stress of periodic variation of temperature.

The primary aim of this study was to determine the effects of alternating temperature, 38°C (threshold high temperature for Bt cotton) during the day and 27°C (normal temperature)at night, on the expression of Cry1Ac protein. A secondary objective was to examine the relationship between the Cry1Ac protein content under alternating temperature stress and the activities of several key enzymes affecting nitrogen metabolism.

2. Materials and methods

2.1. Plant materials and experimental design

In the summers of 2011, 2012, and 2013, experiments were conducted in the greenhouses at Yangzhou University,Yangzhou, China (32°30′N, 119°25′E). Two transgenic Bt cotton cultivars (medium in maturity, Gossypium hirsutum L.)Sikang 1 (conventional cultivar, SK-1) and Sikang 3 (hybrid cultivar, SK-3) were studied. In April, seeds of the two cultivars were planted in a warm room and covered with a plasticfilm.After 43 d, the seedlings were transplanted to pots. Each pot (50 cm height and 40 cm diameter) wasfilled with sandy loam soil (Typic Fluvaquents, Entisols (U.S. taxonomy)) up to 2 cm from the upper edge and watered as required. One seedling was transplanted to each pot.

In 2011 and 2012, temperatures of 32, 34, 36, 38, and 40°C were imposed for 24 h at peak boll stage (104 d after transplanting). The same air humidity (65-70%) was maintained forfive temperature treatments. The study was conducted using a completely randomized design with three plants per temperature treatment for each cultivar. The temperature treatment of 32°C was treated as the control.

Based on the results in 2011 and 2012, the experiment was continued during the 2012 and 2013 cotton-growing season. At the peak boll developing stage (104 d after transplanting), potted plants of the two cultivars were treated with the threshold temperature of 38°C from 6:00 a.m. to 6:00 p.m. followed by a normal temperature of 27°C during the remaining night hours (DH/NN) for 0, 4, 7, and 10 d.The temperature was regulated by air conditioning, and the air humidity (65-70%) of the greenhouse was maintained.Three replicates were used for both cultivars. Control pots were grown in another greenhouse under a 32/27°C day/night temperature regime.

2.2. Sampling and measurement

Preparation of samples Boll samples were collected from

the first or second node on the middle fruit branches. In July, white flowers from the first or second position in the middle of plants were labeled. On 30 July, 10 d after anthesis, the potted plants with labeled flowers were exposed to the stress in the greenhouse. In 2011 and 2012, boll samples were harvested after exposing them to 32, 34, 36, 38, and 40°C for 24 h. In 2012 and 2013, bolls were collected after exposing them to DH/NN treatment for 0, 4, 7, and 10 d, respectively. The controls (32/27°C day/night temperature regime) were sampled at the same time. Five bolls per treatment were collected, placed on ice, and brought to the laboratory where the boll shell,fiber,and seed were immediately separated. The boll shells were cleaned with distilled water and dried with paper towels.All samples were frozen in liquid nitrogen and stored at-80°C. These frozen tissues were used for assays of the Cry1Ac protein, free amino acid, soluble protein content,protease, and GPT activity.

Determination of Cry1Ac protein concentration The concentrations of Cry1Ac protein in the boll shellandfiber extracts were determined by immunologicalanalysis by means of ELISA (Chen et al. 1999). The tissue extracts were harvested by homogenizing the frozen tissue (1.5 g)in 2 mlof extraction buffer (Na2CO31.33 g, DTT 0.192 g,NaCl 1.461 g, and Vc 0.5 g dissolved in 250 mlof distilled water). Then, the contents were transferred to a 10-mL centrifuge tube. The residue was washed with 3 mlof the buffer, and added to a centrifuge tube. The tubes were shaken by hand and stored at 4°C for 4 h. After centrifugation at 10 000×g, samples were held at 4°C for 20 min, the extracts were collected and thenfiltered through a C18Sep-Pak Cartridge (Waters, Milford, MA), and thefiltered supernatants were collected for determination.Microtitration plates were coated with the standard Cry1Ac insecticidal proteins and samples and then incubated at 37°C for 4 h. The antibodies against the Cry1Ac insecticidal protein were added to each welland incubated for another 30 min at 37°C. After that, horseradish peroxidase-labelled goat anti-rabbit immunoglobulin was added to each welland samples were incubated for 30 min at 37°C. Finally,the buffered enzyme substrate (1,2-phenylenediamine)was added (Chen et al. 1999). The enzyme reaction was carried out in the dark at 37°C. After 15 min, the reaction was terminated using 50 μlof H2SO4(3 mol L-1). The results from absorbance measurements at 490 nm were recorded. ELISA data were calculated as described by Weiler et al. (1981).

GPT activity assay GPT of boll shells was extracted and GPT activity was assayed according to Chen et al. (2005a) with slight modifications. The samples(1.0 g) were homogenized in 5 mlof 0.05 mmol L-1Tris-HCl (pH 7.2), and the mixture was centrifuged at 26 100×g for 10 min. During the extraction process, the temperature was kept at 0°C. The supernatant solution was analyzed to detect GPT activity. First, 0.2 mlof the supernatant solution was added to a mixture containing 0.5 mlof 0.8 mol L-1alanine in a 0.1 mol L-1Tris-HCl(pH 7.5), 0.1 mlof 2 mmol L-1pyriodoxal phosphate solution,and 0.2 mlof 0.1 mol L-12-oxoglutarate solution. Then,the reaction mixture was incubated at 37°C for 10 min, and 0.1 mlof trichloroacetic acid solution (0.2 mol L-1) was added to stop the reaction. Then, the pyruvate was converted to pyruvate hydrazine with chromogen. The color intensity of the hydrazine in waster saturated with toluene was red at 520 nm. The GPT activity, in terms of pyruvate production, was calculated from authentic pyruvate standards run simultaneously (Thomas 1975).Assay offree amino acid and soluble protein content The boll shells (0.8 g) were used for the extraction of amino acid and soluble protein according to Wei et al. (2016).Samples were stirred at 4°C in 3 mlof cold water (MilliQ reagent grade) and centrifuged at 800×g for 5 min. The supernatant was stored on ice. Then, the pellets were extracted twice, and the resulting supernatants were pooled for analysis. The absorbance was detected at 570 nm and the free amino acid content was expressed as g g?1fresh weight (FW).

The total soluble protein content was determined by the Coomassie blue dye-binding assay of Bradford (1976). First,0.1 mlof extraction was pipetted into a test tube. Then 5 mL Coomassie brilliant blue G-250 solution was added to the test tubes before vortexing. The absorbance at 595 nm was measured. Bovine serum albumin was used to make a standard curve.

Protease activity assay Protease was extracted and its activity was assayed according to Jessen et al. (1988). The boll samples (0.8 g) were homogenized at 4°C in 1 mlof β-mercaptoethanol extraction buffer (a mixture of ethylene glycol, sucrose, and phenylmethyl sulfonyl fluoride,pH=6.8). Cell debris was removed by centrifugation,and the supernatant was placed on ice and immediately used to estimate protease activity. Protease activity was determined spectrometrically at 400 nm using azocasein as a substrate (Vance et al. 1979) and expressed in mg pro g?1FW h?1.

2.3. Statistics analysis

All samples were analyzed based on three replicates, and the data shown in Tables are their means. The statistical significance of differences between means was analyzed by analysis of variance using Proc ANOVA in SAS 6 (SAS institute, NC). Multiple mean comparisons were evaluated using the LSA test at P＜0.05.

3. Results

3.1. Boll Cry1Ac protein concentration in response to high temperatures at peak boll development stage

The boll insecticidal protein contents decreased with increased temperature after 24 h treatment for SK-1 and SK-3 both in 2011 and 2012 (Fig. 1). However, the effects varied under different high temperature treatments. Greater decreases were detected when the temperature was above 36°C. In 2011, the insecticidal protein contents decreased by 46.2% for SK-1 and by 6.7% for SK-3 when the temperature increased from 36 to 38°C, and decreased by 35.0% for SK-1 and by 58.9% for SK-3 when temperature increased from 38 to 40°C, respectively. Similar results were found in 2012.

3.2. Boll shell Cry1Ac protein levels in response to DH/NN stress

In contrast to the results in 2011 and 2012 (Fig. 1), there were no significant temperature effects on Cry1Ac protein concentration after exposure to the DH/NN stress for 4 d(96 h) in 2012 and 2013. After the DH/NN stress treatment applied for 7 d, significant temperature effects on the boll shell Cry1Ac protein content level were detected, with DH/NN treatment decreasing Cry1Ac protein content by 19.1 and 17.5% for SK-1 and by 15.3 and 13.7% for SK-3 in 2012 and 2013, respectively, compared to the control(Table 1). After 10 d under the stress treatment, Cry1Ac protein concentration levels declined by 44.5 and 44.7% for SK-1 and by 33.2 and 32.7% for SK-3 in 2012 and 2013,respectively. These results indicated that Cry1Ac protein concentrations in Bt cotton boll shells began to decline markedly after only 4 d of exposure to the stress treatment.As the stress time increased, we observed a greater decline in Cry1Ac protein content. We also observed a significant difference in Cry1Ac protein content levels in response to the DH/NN stress between the two cultivars. The decline in protein concentration levels was greater for SK-1 than that for SK-3.

Fig. 1 The boll shell insecticidal protein contents of cotton cultivars exposed to 32, 34, 36, 38, and 40°C for 24 h in 2011 and 2012. Sikang 1 (SK-1, a conventional cultivar) and Sikang 3 (SK-3, a hybrid cultivar) represent the names of the two Bt cultivars.Vertical bars represent SE of the mean (n=3). The same letter within each cultivar represents non-significant difference according to Duncan multiple range tests (P=0.05).

Table 1 Effect of the 38/27°C day/night temperature regime stress on the expression of Cry1Ac protein in the boll shellof Bt cotton at boll stage (ng g-1 fresh weight (FW))

3.3. GPT activity in the cotton boll in response to DH/NN stress treatments

There was no significant difference in GPT activity after exposure to the temperature stress for 4 d. Compared with the control, the DH/NN stress treatment had a significant effect on the GPT activity in the bollafter 7 d of exposure, causing a reduction in activity of 18.06 and 15.11% for SK-1 and SK-3,respectively (Table 2). After 10 d of exposure, the activity for the two cultivars was decreased by 42.4 and 36.8% for SK-1 and SK-3, respectively. The correlation analysis further showed that there was a positive correlation between GPT activity and Cry1Ac protein content in boll shell (r=0.989＊＊).

3.4. The boll shell soluble protein content in response to DH/NN stress treatment

The DH/NN stress treatment decreased the soluble protein in the boll shell (Table 3). There was no significant difference in soluble protein content on the 4th d under the stress.Compared with their corresponding controls, on the 7th d of treatment, the content of soluble protein was reduced by 24.1 and 17.7% for SK-1 and SK-3, respectively. On the 10th d under stress, the content in the two cultivars was decreased by 50.4 and 38.5%, respectively. The correlationanalysis further showed that there was a positive correlation between the content of soluble protein and the Cry1Ac protein content in the boll shell (r=0.996＊＊).

Table 2 Effect of the 38/27°C day/night temperature regime stress on the glutamic pyruvic transaminase (GPT) activity in the boll shellof Bt cotton at boll stage (μmol g-1 FW h-1) in 2013

Table 3 Effect of the 38/27°C day/night temperature regime stress on the content of soluble protein in the boll shellof Bt cotton at boll stage (mg g-1 FW) in 2013

3.5. Amino acid activity in the boll shell in response to DH/NN stress treatments

The results showed that the stress of the DH/NN treatment significantly affected the contents of amino acids for the two cultivars (Table 4). Compared with the respective control,the DH/NN stress treatment increased the amino acid content in the boll shellon the 7th d by 36.3 and 18.6% for SK-1 and SK-3, respectively. After 10 d under the stress,the contents for the two cultivars increased by 90.3 and 59.3%, respectively.

The correlation analysis further showed that there was a negative correlation between amino acid content and Cry1Ac protein content in the boll shell (r=-0.924＊＊).

3.6. The boll shell protease activity in response to DH/NN stress treatments

The stress of DH/NN significantly increased the protease activities in the boll shell (Table 5). On the 7th and 10th d,the activities for the treatments of SK-1 significantlyincreased by 1.30 and 2.79 mg pro g-1FW h-1, respectively,compared to the control. On the corresponding days, those activities for SK-3 increased by 0.98 and 2.80 mg Pro g-1FW h-1, respectively. We also observed a greater increase of protease activity in SK-1 than that in SK-3.

Table 4 Effect of the 38/27°C day/night temperature regime stress on the content offree amino acids in the boll shellof Bt cotton at boll stage (mg g-1 FW) in 2013

Table 5 Effect of the 38/27°C day/night temperature regime stress on the protease activity in the boll shellof Bt cotton at boll stage (mg pro g-1 FW h-1) in 2013

4. Discussion

4.1. Delayed decrease of Cry1Ac protein content under high day/normal night temperature stress in comparison with constant high temperature stress

Studies have shown that several environmental factors,such as temperature, water, and fertilizers, affect the levelof resistance to pests and the expression of the Bt cotton Cry1Ac protein. High temperature is a main factor confirmed by many researchers. However, the stress regime used in previous studies was conducted at a constant temperature over 24 h for several consecutive days (Chen et al. 2005b,2012; Jiang et al. 2012). In our study, the significant decrement of Cry1Ac protein content was found after the Bt cotton was exposed to 38°C continuously for 24 h in 2011 and 2012 (Fig. 1). However, this temperature regime does not accurately mimic nature where there are periodic(hourly) variations of temperature. In nature, the duration of high temperature stress is usually less than 12 h. Therefore,the effects of a constant high temperature on toxic protein expression of Bt cotton are likely different from the effects of a periodic variation of temperature. Moreover, Chen et al. (2013) found that the expression of the insecticide Cry1Ac protein in the leaf could recover completely after short periods (24 h) of extreme temperature. In summer, it is possible that the exposure of Bt cotton to high temperatures will terminate at night. Thus, the results of previous studies might not be truly representative of the condition of heat stress, meaning that results of these studies are not practically applicable in the field. It is more important to study the effects of periodic variations of high temperature on the levelof pest resistance for Bt cotton due to its significance in guiding agricultural practices.

The studies in 2011 and 2012 indicated that 38°C might be the temperature threshold for significant effects on boll shell insecticidal protein content. Thus, the temperature regime of 38°C from 6:00 a.m. to 6:00 p.m. followed by a normal temperature of 27°C during the night was imposed in 2012 and 2013 to study the effects of alternating temperature during day and night on boll shell Bt protein content. Our results showed that under the stress of the 38/27°C day/night temperature regime, Cry1Ac protein contents in the boll shell decreased with prolonged duration of the stress. Compared with their corresponding controls, the difference of Cry1Ac protein content for both cultivars was significant after 7 d and longer. The study also showed that the reduction in content was not significant in Bt cotton subjected to 4 d of stress. Therefore, the reduction in Cry1Ac protein content was in fluenced by the duration of the stress; the longer the stress period, the greater decline in the content. Thus,when compared to constant high temperature stress, we see a difference in the actual stress time required to make a significant reduction of the Bt protein content. This indicates that a periodic, cyclical change of temperature (38/27°C day/night temperature regime) delayed the reduction of Cry1Ac protein content. The possible explanation for the delayed insecticidal protein reduction might be that under the day/night temperature regime, the effects of high temperature on the plants during daytime would be counteracted at night. Hence, the weather should be closely monitored in crops at the flowering stage. When the temperature is above 38°C during the day and lasts for 4 d or more, a significant loss of Cry1Ac protein should be expected in Bt cotton with little possibility of recovery. This means that the levelof insect resistance of these plants willalso decrease significantly. These results will help farmers implement efficient agricultural techniques to control the bollworm and increase farmers’ profitability.

This study also provides a comparison of the stress effects on the conventional SK-1 versus the hybrid SK-3. The extent and rate of decrease of Cry1Ac protein in the boll shell was higher in SK-1 than SK-3(Table 1). These conclusions are consistent with previous research and could be explained by two possible reasons.First, tolerance for high temperature was greater for the hybrid cultivar SK-3 than for the conventional cultivar SK-1.The hybrid plants seemed more resistant to protein loss under high temperature. This supports the work of Chen et al. (2005b). Second, the recovery rate of insecticidal protein in the conventional cultivars was weaker than their hybrid counterparts. This is consistent with the observations of Chen et al. (2013).

4.2. Decreased protein synthesis and increases protein degradation under high day/normal night temperature reduced Bt protein content

Nitrogen metabolism is associated with the levelof insecticidal protein in Bt cotton plants. Research has shown that nitrogen is absorbed, transmitted, reduced, and assimilated throughout the expression of Cry1Ac insecticidal protein (Saini and Dhawan 2014). Therefore, enzymes taking part in nitrogen metabolism, such as GPT and GOT,can be used to determine the content of insecticidal protein(Hallikeri et al. 2011; Dadgale et al. 2014).

The specific measurements showed that, under the high day temperature (38°C)/normal night temperature (27°C)regime, the soluble protein and the GPT activity in the reproductive organs declined. The declining trends were similar to that of the Cry1Ac protein content. In contrast,the concentration offree amino acids and protease activity increased. Thus, the synthesis of soluble protein decreased in bolls, and degradation increased markedly under the high temperature regime.

The fact that the process of nitrogen metabolism is associated with the levelof insecticidal protein was ascertained through the correlation analyses, which showed significant positive correlations between Cry1Ac protein in bolls and the soluble protein contents and between GPT activity and Cry1Ac protein. Additionally, significant negative correlations between Cry1Ac protein and free amino acid contents and between Cry1Ac protein content and protease activity were also detected. These results supported the observations of Chen et al. (2005a).

Compared with previous research (Chen et al. 2005b),we found that the rate of GPT activity and soluble protein content decreased more slowly under high temperature day/night regime stress than that under constant high temperature, while the extent offree amino acid and protease activity increased faster than that under constant high temperature.

Compared with the hybrid cultivar, the range of variation in substances and enzymes related to nitrogen metabolism is larger in the conventional cultivar. These results indicate that the recovery of nitrogen metabolic rates in the bolls for the hybrid cultivar is faster than that for the conventionalone. Consequently, based on the results of this study, it is believed that the hybrid cultivar has advantages over its conventional counterpart by exhibiting higher insecticidal resistance under high temperature. This study can provide some valuable considerations towards future strategies in cotton growing for rational pesticide use.

5. Conclusion

A significant decrease in Cry1Ac protein content was found after Bt cotton was exposed to 38°C continuously for 24 h.The temperature of 38°C might be threshold for significant effects on boll shell insecticidal protein content. However,under the stress of alternating high day temperature(38°C)/normal night temperature (27°C), decline of Cry1Ac protein content was delayed. Bt protein content decreased significantly at 7 d after the alternating high temperature for SK-1 and SK-3. The analysis of nitrogen metabolic physiology further showed that reduced protein synthesis and increased protein degradation in the boll shellunder the alternating high temperature regimes reduced protein content, including Bt protein content, which may result in reduced resistance to cotton bollworm.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (31471435, 31671613, and 31301263),the China Postdoctoral Science Foundation Grant(2016M591934), the Postdoctoral Science Foundation Grant in Jiangsu Province, China (1601116C), the Key Projects of Natural Science Research in Colleges and Universities of Jiangsu, China (17KJA210003), a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions, China (PAPD), and the Practice Innovation Training Project for College Students in Jiangsu Province, China.