The population of O. excavata was commercial, and acquired from Yantai Bifeng Agricultural Technology Co. Ltd., China. It has been continuously mass-reared for more than 10 generations in fruit orchards in Shandong Province, China. No pesticides were sprayed 20 days before flowering and throughout the flowering period.

The effects of high-temperature stress on O. excavata refer to previous research methods with minor modifications (Wang et al., 2017; Zhang et al., 2020). The tubes containing O. excavata were placed in growth climatic chambers (RXZ-600C, Ningbo Jiangnan, China) at different temperatures (30°C, 35°C, 40°C, and 45°C) during the young (2nd instar) larval, mature (5th instar) larval, and adult stages of O. excavata. After a certain period of temperature stress treatment, the larval and adult of O. excavata were taken out for life index detection.

In the acute response test, the young and mature larvae and artificial cocoon-break adults of O. excavata were measured. Bee tubes were opened by hand and cocoons were artificially and carefully dissected with scissors after different periods of high-temperature stress and the mortality of O. excavata was observed according to the method of Song et al. (2021). There were three tubes per replicate (i.e., 18–24 larvae or cocoons per replicate) and three repeats (i.e., 54–72 larvae or cocoons per treatment) every sample date. The number of cocoons in each tube (6–8 cocoons per tube) was determined according to the egg-laying situation of maternal O. excavata. Mortality was observed at 0.25 h intervals, the longest observation period was 2.5 h for larvae and 24 h for adults.

To observe the long-term effects of high-temperature stress on the adults, the chronic response for natural cocoon-break adults was tested. The cocoons containing O. excavata that have been exposed to heat stress for a certain length of time were transferred to their optimal growing environment, that is a darkroom with 65%–75% relative humidity and 25°C ± 2°C temperature (Song et al., 2021). Then the mortality of O. excavata was observed after a natural cocoon break.

The young and mature larvae, natural cocoon-break, and artificial cocoon-break adults of O. excavata were both used to measure the physiological and biochemical indexes after high-temperature stress exposure. There were three tubes per replicate and three repeats.

The trehalose content in each O. excavata was assayed using a trehalose quantification kit (Suzhou Keming Biotechnology Co., Ltd., China). In brief, each O. excavata was weighted using an electronic balance (AL104; Mettler-Toledo, China). The lapping liquid of O. excavata and extracting solution was left standing at room temperature for 45 min and oscillated 3–5 times. After cooling, the sample was centrifuged for 10 min at 8,000 rpm and 25°C to obtain the supernatant. The value was recorded spectrophotometrically at 620 nm.

Acetylcholinesterase (AchE) activity was measured using a kit provided by Suzhou Keming Biotechnology Co., Ltd. The O. excavata and extracting solution were ground to homogenate under ice bath conditions. The abrasive liquid was centrifuged for 10 min at 8,000 rpm and 4°C. Then, the activity of AchE was evaluated spectrophotometrically at 412 nm.

The fat and free water were also assayed. First, the fresh weight of O. excavata was determined using an electronic balance. Then the body was dried for 48 h at 60°C and the dry weight was determined. The free water was obtained by calculating the difference between the wet and dry weight of O. excavata. Then dried O. excavata was added to 2 ml of chloroform:methanol (2:1), and fully ground until homogenized. Afterward, the homogenate was centrifuged for 10 min at 2,600 rpm and the supernatant was discarded. Another 2 mL of the mixture was added to the residue, the centrifugation was repeated once, and the supernatant was discarded. The remaining residue was dried for 72 h at 60°C to constant weight. Finally, the fat was calculated according to a method described previously (Colinet et al., 2007).

Before statistical analysis, data were transformed to log₁₀, arcsine, or square-root when necessary to evaluate data normality and homogeneity. The median lethal time (LT₅₀) and analysis of variance were determined using SPSS v.21.0 (SPSS Inc., Chicago, IL; Abbott, 1925). One-way analysis of variance (ANOVA) was used to estimate the impact of different duration of high-temperature stress on the mortality and the physiological and biochemical indexes of O. excavata. Two-way ANOVAs were performed to determine the impact of insect stages (or response mode) and high-temperature stress on the physiological and biochemical indexes of O. excavata. Means were compared by using Tukey’s LSD test at p < 0.05. Pearson’s correlation was used to analyze the relationships between the mortality and the physiological and biochemical indexes of O. excavata at different insect stages under high-temperature stress.

The mortality of O. excavata significantly increased with an increase in the duration of high-temperature stress (p < 0.05; Figure 1) and increased with an increase in temperature (R² = 0.92–0.99; Table 1).

For the young larvae of O. excavata, the mortality at 30°C for 2.5 h was about 15%, and that at 35°C for 2.5 h was about 30%, but those were significantly higher than those at 30°C and 35°C for less than 1.75 h (+66.7% and +79.96%), respectively (p < 0.05). The mortality of young larvae at 40°C for 1.75–2.0 h was more than 40%, and both were markedly higher than those at 40°C for less than 1.25 h (p < 0.05). The mortality of the young larvae at 45°C for 1.0 h reached 96.67%, and significantly higher than that at 45°C for 0.75 h (+20.84%; p < 0.05). The young larvae did not die when the temperature was lower than 40°C for 0.5 h, but the mortality reached 20% after 0.25 h at 45°C (Figure 1). Additionally, the LT₅₀ of the young larvae decreased with an increase in temperature (Table 1). The LT₅₀ of the young larvae at 30°C and 35°C were over 3 h, and at 40°C and 45°C, they were 2.01 and 0.53 h, respectively (R² = 0.98–0.99; Table 1; Figure 1).

For the mature larvae of O. excavata, the mortality at 30°C for 2.25 h was about 13% and significantly higher than that at 30°C for less than 1.25 h (+100%; p < 0.05). The mortalities of the mature larvae at 35°C and 40°C for 2.0 h were markedly higher than those for 1.5 h, respectively (+66.65% and + 30%; p < 0.05). The mortality of the mature larvae at 45°C for 1.0 h reached 100%, and significantly higher than that at 45°C for 0.75 h (+20%; p < 0.05; Figure 1). The LT₅₀ of the mature larvae at 40°C and 45°C were 2.35 h and 0.47 h, respectively (R² = 0.95–0.99; Table 1; Figure 1).

For the artificial cocoon-break adults, the mortality increased slowly with time at 35°C and 40°C, but rapidly at 45°C (Figure 2). The mortality of the artificial cocoon-break adults at 45°C for 0.5 h was nearly 50% and markedly increased to 90% after 2 h (+107.14%; p < 0.05). The LT₅₀ of the artificial cocoon-break adults at 40°C and 45°C were more than 25 and 0.6 h, respectively (R² = 0.92–0.99; Table 1; Figure 2).

In the natural cocoon-break adults of O. excavata, no death occurred under 35°C stress after 17 h, but the mortality increased significantly to 30% after 24 h (p < 0.05; Figure 2). The mortality of natural cocoon-break adults increased faster with time at 40°C than that at 35°C, but slower than that at 45°C (p < 0.05). The mortality of natural cocoon-break adults at 45°C for 1 h was nearly 23.33% and markedly increased to 90% after 3.5 h at 45°C (+285.77%; p < 0.05). The LT₅₀ of natural cocoon-break adults at 40°C and 45°C were 15.55 and 1.51 h, respectively (R² = 0.88–0.99; Table 1; Figure 2).

Although the acute response of the physiological and biochemical indexes of O. excavata were significantly different between young larvae, mature larvae, and artificial cocoon-break adult stages (F = 5.52–653.19; p < 0.01), the high-temperature stress had no significant effect on the acute response of the physiological and biochemical indexes (F = 0.36–0.88; p > 0.05; Table 2). There were significant interactions of the fat content between temperature stress and insect stage (F = 3.38; p < 0.01). Additionally, the trehalose content of the mature larvae at 35°C was significantly increased compared with that at 30°C (+126.20%; p < 0.05; Figure 3). The fat content of the artificial cocoon-break adults at 45°C was markedly higher than that at 35°C (+32.12%; p < 0.05). The free water content was significantly decreased compared with that at 35°C (−9.21%; p < 0.05; Figure 3).

There was no significant difference between the acute and chronic responses on the physiological and biochemical indexes of adult Osmia obtained by either artificial or natural cocoon-break methods (F = 1.42–2.76; p > 0.05), except the fat contents (F = 5.03; p < 0.05; Table 3). The high-temperature stress had a significant effect on the contents of trehalose and fat (F = 4.38–6.86; p < 0.05). The trehalose content of adult Osmia obtained by the natural cocoon-break method at 45°C was significantly higher than that at 30°C (+359.31%; p < 0.05), 35°C (+177.17%; p < 0.05), and 40°C (+342.15%; p < 0.05). The fat content of adults by the natural cocoon-break method at 45°C was markedly higher than that at 35°C (+29.59%; p < 0.05; Figure 3).

The mortality of the young larvae was significantly and positively correlated with the trehalose content (r = 0.78, p < 0.05) and AchE activity (r = 0.67, p < 0.05; Figure 4). The mortality of the artificial cocoon-break adults of O. excavata was also significantly and positively related to the trehalose content (r = 0.82, p < 0.01). However, the mortality of the natural cocoon-break adults was negatively related to the AchE activity (r = −0.93, p < 0.001). The mortality of the mature larvae was not significantly correlated with physiological and biochemical indexes [r = (−0.10) − 0.11, p > 0.05; Figure 4].

Temperature is one of the most important external conditions affecting the life activities of insects (Dongmo et al., 2021; Gaytán et al., 2022), but an abnormal high-temperature environment has a serious impact on growth and development, and even causes the risk of extinction in some insects (García-Robledo et al., 2016; Abou-Shaara et al., 2017; Wang et al., 2017). Populations of O. excavata are more easily affected by adverse factors because they only have one generation a year (Men et al., 2018). Our results demonstrated that the higher the temperature, the faster the death rate of O. excavata, which was consistent with our hypothesis (i). Similar results have been obtained in Carposina nipponensis (Zhang et al., 2020), Ophraella communa (Chen et al., 2018), and Corythucha ciliata (Ju et al., 2013). In addition, this study found that under a slightly higher temperature (35°C and 40°C), the LT₅₀ value of adults was higher than that of larvae. Thus, we speculated that the harmful effects of high temperatures on O. excavata depend not only on the intensity and duration of stress but also on the developmental stage (Enriquez and Colinet, 2017). As the adults of O. excavata will naturally emerge from the cocoon with the rising temperature in the spring (Men et al., 2018), this may be the reason for a higher temperature tolerance in adults than in larvae. By comparing the results of acute and chronic responses of O. excavata to high-temperature stress, it was found that the LT₅₀ values of natural cocoon-break adults at 35°C and 40°C were smaller than those of artificial cocoon-break adults, but similar LT₅₀ values were observed at 45°C. This may occur for three reasons: (a) the physiological and biochemical substances that maintain the normal growth and development of O. excavata were destroyed after slightly higher temperature stress, even if they did not die immediately, but it was not enough to support their survival for long; (b) the functions related to cocoon breaking were severely damaged and failed to successfully break the cocoon; (c) extreme high-temperature stress (45°C) may cause O. excavata to die instantly. Considering the high mortality we found in larvae and adult Osmia, apiarists should store bee tubes in a cool ventilated place out of direct sunlight to maintain an adequate population of O. excavata.

Trehalose is an important blood sugar in insects, not only can it be stored as an energy source and carbohydrate reserve but also as a compatible solute adapted to various stresses, such as heat, cold, osmotic stress, and drought (Xin et al., 2013; Qin et al., 2015). There was no significant difference in the trehalose content in larvae and artificial cocoon-break adults at 30°C, but the trehalose content in the former increased faster than the latter with an increase in temperature, especially when the temperature reached 40°C and 45°C, the content and growth rate of trehalose in the young larvae were both higher than that of the mature larvae. These results showed that the smaller the insect stage under high-temperature stress, the more drastic the change of trehalose content in O. excavata. This may be due to the poor ability of the younger larvae to resist an adverse environment. Jiang et al. (2016) also believed that the body surface of younger larvae was soft and its cuticle was thin, but the body surface of adult insects was relatively hard and the cuticle layer was thicker, which can reduce stress and retain normal life activities of adults.

Fat plays an important role in the energy storage and metabolism of insects (Wu et al., 2019). At the same time, insects cannot live without the participation of free water, which is a good solvent in the cell, can participate in the cellular biochemical reaction, and transport nutrients and waste produced by metabolism. Adults of O. excavata were found to have less fat and more free water than larvae. The temperature in the larval stage of O. excavata is relatively low under natural conditions, so the larvae would resist the cold climate by storing more fat and reducing free water content; as temperatures rise in the adult stage, the O. excavata would break diapause, cocoon breaking, flight, etc., therefore, they need to use more energy and free water to speed up metabolism (Sgolastra et al., 2015; Men et al., 2018). It was also found that the contents of fat and free water in larvae were not significantly affected after extreme heat stress, whereas the contents in adults were more affected, indicating that the fat and free water contents in larvae of O. excavata were less sensitive to high-temperature stress than that in adults. This may be because the larvae of O. excavata have fewer activities and slower metabolism than adults in the cocoon, but further research is needed.

In the chronic response test, the trehalose content of adult Osmia was significantly increased after extreme high-temperature stress compared to slightly higher temperature stress. We speculated that there may be two reasons: (a) trehalase was inactivated in insects under extreme high-temperature stress, so trehalose could not be hydrolyzed, and more trehalose content was accumulated (Qin et al., 2015); (b) a special protective membrane is formed on the cell surface by increasing trehalose content under harsh conditions, such as extreme high temperature, to prevent the structure of biomolecules from being disrupted (Ma et al., 2018). These results were similar to previous studies on Monolepta hieroglyphica and Gomphocerus sibiricus and suggested that this was a manifestation of insect adaptation to high-temperature stress (Li et al., 2014; Ma et al., 2018). However, in the acute response test, the trehalose contents of larvae or artificial cocoon-break adults were not markedly changed after being subjected to high temperatures. We believed that although the trehalose content in O. excavata increased with an increase in temperature stress at each insect stage, it did not increase to a significant level in a short time.

The fat contents in natural and artificial cocoon-break adults of O. excavata after extreme high-temperature stress were both significantly higher than those after slightly high-temperature stress, but the change in the trend of free water was opposite to that of the fat content. Zhao et al. (2010) found that low-temperature stress can also lead to an increase in insect fat content. We believe that the reasons for the increase in fat content are different between high and low-temperature stress: the former may be due to the serious loss of free water in the adult body of O. excavata under extreme temperature stress, which hinders normal metabolism and inhibits the decomposition of fat, even causing adults to die faster; the latter may be because insects would reduce their supercooling point in a low-temperature environment by increasing their fat content, which can increase their cold tolerance and resist cold damage (Zhao et al., 2010). Interestingly, Liu et al. (2010) reported that heat-shock stress decreased lipid storage in planarian Dugesia japonica to suppress its development. We also found that the fat content of adults decreased at 35°C compared with 30°C, but did not reach a significant level. This may be because the life activities of O. excavata have not been completely disordered after they endured tolerable temperature stress, and the metabolic ability was enhanced to maintain normal life activities by consuming more energy substances such as fat.