Impacts of Drought and Rehydration Cycles on Isoprene Emissions in Populus nigra Seedlings

The volatile organic compounds emitted by plants significantly impact the atmospheric environment. The impacts of drought stress on the biogenic volatile organic compound (BVOC) emissions of plants are still under debate. In this study, the effects of two drought–rehydration cycle groups with different durations on isoprene emissions from Populus nigra (black poplar) seedlings were studied. The P. nigra seedlings were placed in a chamber that controlled the soil water content, radiation, and temperature. The daily emissions of isoprene and physiological parameters were measured. The emission rates of isoprene (Fiso) reached the maximum on the third day (D3), increasing by 58.0% and 64.2% compared with the controlled groups, respectively, and then Fiso significantly decreased. Photosynthesis decreased by 34.2% and 21.6% in D3 in the first and second groups, respectively. After rehydration, Fiso and photosynthesis recovered fully in two groups. However, Fiso showed distinct inconsistencies in two groups, and the recovery rates of Fiso in the second drought group were slower than the recovery rates of Fiso in the first groups. The response of BVOC emissions during the drought-rehydration cycle was classified into three phases, including stimulated, inhibited, and restored after rehydration. The emission pattern of isoprene indicated that isoprene played an important role in the response of plants to drought stress. A drought–rehydration model was constructed, which indicated the regularity of BVOC emissions in the drought–rehydration cycle. BVOC emissions were extremely sensitive to drought, especially during droughts of short duration. Parameters in computational models related to BVOC emissions of plants under drought stress should be continuously improved.


Introduction
A considerable amount and types of biogenic volatile organic compounds (BVOCs) are exchanged between the vegetation and the surrounding air [1,2]. The global BVOC emissions reached 1150Tg in 1990, accounting for 90% of the global annual discharge of VOCs [3]. It plays a significant role in tropospheric chemical reactions, ozone generation, as of the most important models for studying isoprene emission, especially under stress [48]. P. nigra is therefore meaningful for the study of isoprene emissions under different drought or rehydration conditions. P. nigra is one of the most widely distributed species in China and has been selected and planted in many arid and barren places, such as the Three North Shelterbelt [49,50]. P. nigra is therefore meaningful for the study of isoprene emissions under different drought or rehydration conditions. The daily isoprene emissions from P. nigra under drought and rehydration treatment in a chamber were measured. The objectives of this study were: (1) to investigate the impact of short drought durations (~10 days) and long drought durations (~20 days) drought stress on the emissions of isoprene; (2) to explore the relationships between isoprene emission rates and photosynthetic parameters; (3) to compare the difference in isoprene emission rates under rehydration after different durations of drought; and (4) to hypothesize a theory to explain the mechanism of BVOC emissions in drought rehydration cycles of different durations.

Plant Material and Chamber Design
In September 2020, a total of 8 3-year-old P. nigra seedlings were obtained from a nursery in Qingdao, Northern China. The trees were transplanted to 50-L plastic pots (diameter 40 cm, height 50 cm) that contained 40% commercial potting soil and 60% clay soil. The soil mixture allowed fast drought application. These 8 pots were placed in a chamber without sun radiation, and the chamber size was 2.5 m × 3.5 m × 4 m (length × width × height). To provide consistent radiation, five 180 W light emitting diode (LED) lights (WEN-180, Guixiang Inc., Weifang, China) were installed. The LED lights were turned on at 7:00 a.m. and turned off at 6:00 p.m. during the whole experiment. The air circulation was controlled through an air intake and outlet pump, with a flow rate of 210 m 3 /h. The chamber temperature was maintained at 25 • C by an air conditioner. The chamber temperature and humidity were both recorded every minute (Supplementary Material Figure S1).

Soil Water Content
Soil water content (SWC) was monitored by a soil moisture sensor (EC-5, METER Group, Inc., Pullman, WA, USA), which was placed under 10 cm of soil. SWC was recorded every 30 s by a data collector (ZL6, METER Group, Inc., Pullman, WA, USA). To explore the wilting point of P. nigra in pots, a drought pre-experiment was set up (Supplementary Material Figure S2). Four trees were randomly selected without watering until the leaves and trees were completely wilted, and the duration was recorded. On D17, one piece of leaf was completely wilted, and on D22~D25, the whole trees were completely wilted. Therefore, the drought of the first group lasted 8~9 days, and the drought of the second group lasted 17 days. Two drought-rehydration cycle groups were performed in two independent experiments, each consisting of 2 replicates. In the first drought-rehydration cycle group (short drought durations), there were two drought-rehydration cycles. The first lasted for 8 days, and 2 L of water was rehydrated on the night of D8. The second drought stress lasted 9 days, and 2 L of water was rehydrated on the night of D18. In the second drought-rehydration cycle group (long drought durations), the drought-rehydration cycle lasted for 17 days, and 2 L of water was rehydrated on the night of D19.

Isoprene Sampling and Gas Exchange
The emissions of isoprene were sampled by a portable photosynthetic apparatus (CIRAS-3, PP Systems Inc., Hitchin, UK) in the chamber. Fresh air was filtered into the leaf chamber (18 mm × 25 mm) of a portable photosynthetic apparatus with a flow rate of 100 mL/min. The sampling temperature and photosynthetic active radiation (PAR) in the leaf chamber were 30 • C and 1000 µmol m −2 s −1 , respectively. The sample was pumped into the adsorption tube (Markes International Ltd., UK) through the pump (GilAir plus, Sensidyne Gilian Inc., Petersburg, FL, USA). The flow rate was set at 90 mL/min and lasted for 30 min. Net photosynthetic rate (A net ), intracellular carbon dioxide concentration (C i ), stomatal conductance (G s ), transpiration rate (E), water use efficiency (WUE), and vapor pressure deficit (VPD) measurements were performed using the portable photosynthetic apparatus CIRAS-3.

Quantification of Isoprene
The gas chromatography/mass spectrometry (GC/MS) system used an Agilent 5977B GC/MSD coupled to an Agilent 7890B GC. The desorber was UNIT-xr (Markes International Ltd., Bridgend, UK) with an automatic sample processor (ULTRA-xr, Markes International Ltd., Bridgend, UK). The initial oven temperature was increased at 20 • C min −1 from 40 to 280 • C. The desorbed isoprenoids were cryofocused at −3 • C for 2 min, after which the cryotrap was heated rapidly to 280 • C and placed into a 30 m × 0.32 mm × 0.18 µm column (DB-624, Agilent (J&W), Santa Clara, CA, USA). The flow of helium was 1 mL/min, and the total run time was 38 min, including a solvent delay of approximately 2 min. The initial oven temperature was increased on the capillary column at 5 • C min −1 to 200 • C, then increased with a 20 • C min −1 ramp to 260 • C and maintained at 260 • C for 2 min.

Quality Assurance/Quality Control
All P. nigra seedlings were exposed to the same temperature and radiation. The isoprene emission and photosynthetic parameters on D1 were used as controls. There were two trees in each of the two groups. One leaf was selected from each tree that had a similar size, degree of development and height. Increases or decreases in isoprene emissions and photosynthesis during drought and rehydration were calculated as the treatment effect, which equaled the treatment minus the control and was divided by the control. A pair comparison test (mean comparison) was applied to compare isoprene emissions over the drought rehydration cycle (from D2 to D24) with those from D1 (control). All statistical tests were considered significant at p < 0.05. Error bars represent the standard deviation of the two independent experiments.

SWC and Physiological Parameters
The F iso and SWC are shown in Figure 1. The SWC of the first group was 0.152 ± 0.002 m 3 /m 3 on D1. With increasing drought stress, the SWC value decreased to 0.117 ± 0.013 m 3 /m 3 on D8, and the lowest was 0.096 ± 0.007 m 3 /m 3 on D17 ( Figure 1A). The SWC in the second group was 0.143 ± 0.002 m 3 /m 3 on D1, and with the deepening of drought, the SWC gradually decreased to 0.093 ± 0.003 m 3 /m 3 on D16 ( Figure 1B). The change in the gas exchange parameters is shown in Figure 2. Drought stress obviously limited the Anet (Figure 2A), gs ( Figure 2B), E ( Figure 2C), and WUE ( Figure 2D), which gradually decreased with the deepening of the drought, but the VPD ( Figure 2E) showed the opposite trend in both groups. In the first and second groups, the decreasing trend and size of the five physiological parameters showed obvious consistency ( Figure  2). The Anet and gs dropped rapidly by more than 90% in the first five days of the drought. Drought stress is the largest limiting factor for poplar growth, especially for photosynthesis [16]. On the one hand, drought stress is the direct reduction of photosynthetic raw The change in the gas exchange parameters is shown in Figure 2. Drought stress obviously limited the A net (Figure 2A), g s ( Figure 2B), E ( Figure 2C), and WUE ( Figure 2D), which gradually decreased with the deepening of the drought, but the VPD ( Figure 2E) showed the opposite trend in both groups. In the first and second groups, the decreasing trend and size of the five physiological parameters showed obvious consistency ( Figure 2). The A net and g s dropped rapidly by more than 90% in the first five days of the drought. Drought stress is the largest limiting factor for poplar growth, especially for photosynthesis [16]. On the one hand, drought stress is the direct reduction of photosynthetic raw materials, and on the other hand, drought stress indirectly limits g s and enzyme activity reduction [41,51].

Isoprene Emission Rates under Drought Stress
The Fiso of the control was 15.7 ± 2.3 nmol m −2 s −1 . In the first and second groups, Fiso had an initial stimulation followed by a dramatic decrease from D4 to D6 (Figure 1). There is a high consistency of Fiso between different drought durations, such as the time of changes in the peak of Fiso. On D3, the maximum Fiso values in the first and second groups were 24.9 ± 1.05 nmol m −2 s −1 and 25.8 ± 2.50 nmol m −2 s −1 , respectively. On D4, Fiso rapidly decreased, and on D6, Fiso decreased to almost zero.

Isoprene Emission Rates under Drought Stress
The F iso of the control was 15.7 ± 2.3 nmol m −2 s −1 . In the first and second groups, F iso had an initial stimulation followed by a dramatic decrease from D4 to D6 (Figure 1). There is a high consistency of F iso between different drought durations, such as the time of changes in the peak of F iso . On D3, the maximum F iso values in the first and second groups were 24.9 ± 1.05 nmol m −2 s −1 and 25.8 ± 2.50 nmol m −2 s −1 , respectively. On D4, F iso rapidly decreased, and on D6, F iso decreased to almost zero. Isoprene emission was stimulated by drought stress, regardless of exposure to droughtrehydration cycles of different durations (Figure 1). On D3 in the first and second groups, compared with the control, F iso increased by 58.0% and 64.2%, respectively. These results were consistent with the results obtained from studies on the effects of water stress on isoprene emission [27,52]. After 6 days of drought stress, the F iso in Alnus glutinosa on D6 was higher than the F iso in Alnus glutinosa on D1 [29]. Limited studies clearly identified that under drought stress, isoprene emissions were briefly stimulated, increasing by 33.7% to 300% (Table 1). Our results fall within the range, and that peak occurred after D3 in both groups. The stimulated F iso in our study was relatively lower than previous results (33.7-300%). However, other studies showed either almost no change in isoprene emission or slightly lower than the control [31,37,53,54], possibly attributed to the low time resolution (not daily measurement) of BVOC emissions, which might miss the peaks of BVOC emissions. Pegoraro et al. [55] studied Quercus virginiana Mill, and F iso remained essentially constant for 8 days of treatment. Whether this threshold is common to all species remains uncertain. The rare isoprene emitter Hakonechloa macra, compared with the stronger isoprene emitter, was observed to impair chloroplast ultrastructure, indicating damage to photosynthetic machinery under drought conditions [54,60]. There are three possible explanations: (1) isoprene is an effective antioxidant; drought can promote oxidation by increasing the oxidative pressure of plant cells, and isoprene protects plant cells from oxidative damage [52,57]; (2) membrane stabilizers protect cells of plants such as thylakoids and chloroplasts during drought [60,61]; and (3) membrane stabilizers reduce the damage of reactive oxygen species (ROS) to plants and suppress the generation of ROS [62][63][64].
However, as SWC decreased, progressive and steady declines in F iso under stress conditions were observed; on D7 in the first groups, the decline in F iso decreased to 1.73 ± 0.018 nmol m −2 s −1 , and on D10 in the second drought, the decline in F iso decreased to zero. The results were consistent with previous studies [31,34,65,66]. Beckett et al. [27] studied Xerophyta humilis subjected to severe drought treatment. When the relative soil water content (RWC) decreased to 80%, the F iso peaked, and when the RWC decreased to 53%, the F iso decreased by zero. The second group had a longer drought, and the plants needed more effective protection; isoprene is increased when desiccation is moderate, while nonvolatile isoprenoids operate when drought stress is more extreme [27,57].

Isoprene Emissions and Physiological Parameters during Rehydration
The difference in the isoprene emission rates of P. nigra between the first and second groups during rehydration is shown in Figure 3. After rehydration, F iso in the first and second groups showed distinct inconsistencies (Figure 3), but physiological parameters showed clear consistency (Figure 2). After rehydration, emissions of isoprene recovered at a slower rate than photosynthesis. F iso was slightly different in the two groups after rehydration. After rehydration in the first and second groups, the maximum F iso was 18.3 ± 0.238 nmol m −2 s −1 and 17.8 ± 0.475 nmol m −2 s −1 , respectively, which increased by almost 10% compared with the control.

Isoprene Emissions and Physiological Parameters during Rehydration
The difference in the isoprene emission rates of P. nigra between the first and second groups during rehydration is shown in Figure 3. After rehydration, Fiso in the first and second groups showed distinct inconsistencies (Figure 3), but physiological parameters showed clear consistency (Figure 2). After rehydration, emissions of isoprene recovered at a slower rate than photosynthesis. Fiso was slightly different in the two groups after rehydration. After rehydration in the first and second groups, the maximum Fiso was 18.3 ± 0.238 nmol m −2 s −1 and 17.8 ± 0.475 nmol m −2 s −1 , respectively, which increased by almost 10% compared with the control. Full recovery of Anet occurred after rehydration, confirming that photosynthetic limitations were fully reversible and that no permanent damage occurred. On D3 of rehydration, the Anet, E and WUE of the first and second groups were fully recovered. However, after the second rehydration of the first group, the gs and VPD recovered less than 50% compared to the control. After the first rehydration of the first groups, the gs and VPD increased rapidly and almost recovered to the control level ( Figure 2B,E). This result indicates a complete recovery of photosynthesis and no permanent limitations caused by drought. This pattern was in good agreement with the characteristic adaptation strategy of this species to withstand summer drought [67].
After rehydration, the Fiso completely recovered, which is in line with others reporting on isoprene emitters (Table 2), such as Arundo donax, Populus alba and Quercus virginiana Mill [53,55,60]. Limited studies have investigated the effects of rehydration after drought stress on isoprene emissions for P. nigra. Fortunati et al. [37] found that after a 3day rehydration phase, the Fiso of P. nigra recovered in plants grown at both 25 and 35 °C. However, isoprene emissions did not reach the prestress levels even 15 days after rehydration; when photosynthesis had completely recovered, the reduction of isoprene emissions after recovering from drought stress was particularly strong in leaves grown at 35 °C . Previous studies of rehydration showed that isoprene emitters could still recover and increase within a short period even under severe drought conditions. Brilli et al. [53] found that after rehydration of Populus alba under severe drought treatment, Fiso was restored to Full recovery of A net occurred after rehydration, confirming that photosynthetic limitations were fully reversible and that no permanent damage occurred. On D3 of rehydration, the A net , E and WUE of the first and second groups were fully recovered. However, after the second rehydration of the first group, the g s and VPD recovered less than 50% compared to the control. After the first rehydration of the first groups, the g s and VPD increased rapidly and almost recovered to the control level ( Figure 2B,E). This result indicates a complete recovery of photosynthesis and no permanent limitations caused by drought. This pattern was in good agreement with the characteristic adaptation strategy of this species to withstand summer drought [67].
After rehydration, the F iso completely recovered, which is in line with others reporting on isoprene emitters (Table 2), such as Arundo donax, Populus alba and Quercus virginiana Mill [53,55,60]. Limited studies have investigated the effects of rehydration after drought stress on isoprene emissions for P. nigra. Fortunati et al. [37] found that after a 3-day rehydration phase, the F iso of P. nigra recovered in plants grown at both 25 and 35 • C. However, isoprene emissions did not reach the prestress levels even 15 days after rehydration; when photosynthesis had completely recovered, the reduction of isoprene emissions after recovering from drought stress was particularly strong in leaves grown at 35 • C. Previous studies of rehydration showed that isoprene emitters could still recover and increase within a short period even under severe drought conditions. Brilli et al. [53] found that after rehydration of Populus alba under severe drought treatment, F iso was restored to 57%, 160% and 120% of the control on D2, D7 and D14, respectively. Upon full rehydration to 100% of soil water content, isoprene emission reached levels of 3.6-5.2 nmol m −2 s −1 , which was equivalent to the maximum emission rate upon dehydration and much higher than prior to dehydration in X. humilis [27], demonstrating that drought stress, even severe drought, did not affect P. nigra production or the emission of isoprene. Notes: RWC is relative water content; FTSW5: fraction of transpirable soil water is 5%.

The Impact of Physiological Parameters on Isoprene
In the second group, F iso and the physiological parameters were significantly positively correlated. Among these compounds, g s had the highest correlation. Generally, the first event characterizing the plant response to water shortage is progressive stomatal closure triggered by increased stomatal closure [68], a significant effect of F iso [59,69]. In the drought rehydration cycle, A net and g s, the main photosynthetic factors affecting isoprene emissions, have previously been reported [16,55,59]. However, a lower drought sensitivity of F iso compared with A net and g s was found; for example, on D3, F iso increased by 42.5% in the second group, but A net and g s decreased by 47.1% and 37.6%, respectively, consistent with other results demonstrating the response to drought stress [16,55,59]. Pegoraro et al. [55] studied Quercus virginiana Mill saplings. A net and g s decreased by 92% and 91%, respectively, while F iso remained essentially constant for 8 days of treatment and for 12 days under severe drought conditions, and F iso was reduced by 64%. Seco et al. [34] found that in the Ozark area where extreme drought occurred, the net flux of CO 2 reached its seasonal maximum approximately a month earlier than isoprenoid fluxes, highlighting the different responses between isoprene emissions and physiological parameters to drought stress, and previous studies confirmed that the different responses progress under drought conditions [37,67]. This addition of isoprene emissions seemed to be relatively independent from photosynthesis, indicating a more complex regulation of isoprene emissions [31,57,60,69]. In these cases, an uncoupling between isoprene emissions (that remains stable or decreases slightly) and photosynthesis (that decreases dramatically) has been observed for different plant species [27,57].
Furthermore, as pointed out in our previous investigations, isoprene is important for plants to withstand drought stress. Plants produced high isoprene concentrations under environmental stress conditions because of a low allocation of carbon to growth, suggesting a trade-off between growth and defense [70]. Previous findings confirmed that isoprene emission was not inhibited by drought stress-induced stomatal closure and that isoprene emission was uncoupled from A net under drought stress conditions [27,60]. The ability to use stored carbon (alternative carbon sources), as opposed to assimilated photosynthate, for isoprene production might be important as plants routinely experience photosynthetic depression in response to environmental stress [16,33,53,59]. Isoprene emission has been proposed to possibly be regulated by substrate availability. Studies using 13 C isotopes in plants have confirmed that mainly carbon sources perform photosynthesis in well-watered plants, and the percentage contribution of these additional carbon sources greatly increases under stress conditions. In well-watered plants, 75~88% of the carbon in isoprene was derived from photosynthate. Under moderate stress and drought stress it dropped to 60%, but this percentage dropped significantly under severe stress to only 10~20% [16,53,59]. Thus, increasing the duration of drought led to an increased contribution of alternative carbon sources to the 2-C-Methyl-D-erythritol 4-phosphate (MEP) pathway, rather than newly made products of photosynthesis [60]. In the early stages of drought, isoprene was used as a protection against drought. As the duration of drought increased, F iso experienced a very significant reduction, and nonvolatile isoprenoids in the MEP pathway, such as zeaxanthin and lutein, were found to be produced in large quantities using alternative carbon sources to cope with drought stress [27,33].
Based on our results and a review of previous research on the leaf-level response of isoprene emissions to the drought-rehydration cycle, we hypothesize that the response has three phases ( Figure 4). The drought-rehydration cycle is divided into phases I, II and III. Figure 4A shows the changes in BVOC emissions in phases I, II and III of the drought rewatering cycle. Figure 4B shows that in phase I, the stimulation of BVOCs by drought stress is the focal point of the debate. Figure 4C shows the effect of short and long drought durations on the recovery rate of BVOCs in phase III. In phase I of mild drought stress, emissions were stimulated using alternative carbon sources and used to protect plants against the stress, even though reduced values of stomatal conductance were associated with the physiological response to drought stress ( Figure 4A,B). In phase II of more severe drought stress, emissions were suppressed by reductions in substrate availability and/or isoprene synthase transcription ( Figure 4A). In the III phase of rehydration, emissions recovered fully ( Figure 4A), but the recovery rate of emissions under long-duration drought conditions was slower than the recovery rate of emissions under short-duration drought conditions ( Figure 4C). The hypothesis for phase I of Figure 4A is based on the observation that drought stress enhances BVOC emissions ( Table 1). The second part of the hypothesis ( Figure 4B) is based on numerous observations [16,52,53,57,[59][60][61][62][63][64], while the last part of the hypothesis ( Figure 4C) is based on observations and theoretical considerations [35].

Conclusions
Anet, gs, E, and WUE gradually decreased with increasing drought, but VPD showed the opposite trend. The Fiso of the control was 15.7 ± 2.3 nmol m −2 s −1 . Fiso showed an initial stimulation followed by a dramatic decrease when the stress was severe. On the third day of the first and second groups, Fiso increased by 58.0% and 64.2% compared to the control,

Conclusions
A net , g s , E, and WUE gradually decreased with increasing drought, but VPD showed the opposite trend. The F iso of the control was 15.7 ± 2.3 nmol m −2 s −1 . F iso showed an initial stimulation followed by a dramatic decrease when the stress was severe. On the third day of the first and second groups, F iso increased by 58.0% and 64.2% compared to the control, respectively. After rehydration, F iso under drought stress showed distinct inconsistencies in the first and second groups. Isoprene emissions recovered at a slower rate than photosynthesis.
Further research is necessary to determine the change in isoprene emission rates in the drought rehydration cycle. As the world's climate changes, such knowledge may be especially valuable for boreal tree species BVOC emission inventories, such as those in China.
Emission rates may be underestimated for isoprene, which show fast reactions with ozone. To further unravel the effect of long-term drought-rehydration on isoprene emissions, more studies characterizing emission patterns in predrought and rehydration periods are needed. These results highlight that direct plant stress sensing creates opportunities to understand the overall complexity of stress-related BVOC emissions.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/ijerph192114528/s1, Figure S1: Chamber design. Figure S2: Pre-experiment was set up, with Populus nigra leaf and seedling wilting. (A) In the long-term drought a leaf is completely wilted on D16. (B) After 22~23 days of drought, the Populus nigra seedling were completely wilted. Figure S3: Under short and long drought durations, the value for chlorophyll SPAD.