Combined Cold and Drought Stress-Induced Response of Photosynthesis and Osmotic Adjustment in Elymus nutans Griseb.

Abstract

Elymus nutans Griseb. is a dominant forage in the Qinghai–Tibetan Plateau. However, the combined cold and drought (CD) stress is a major problem inhibiting its growth, development, and yield. Here, the responses of morphological, photosynthetic, osmoregulation levels, and signal transduction under cold, drought, and CD stress were explored. Both cold- and drought-stressed plants showed varying degrees of damage. In addition, CD stress led to more severe damage than single stress, especially in total biomass, photosynthetic capacity, and electron transfer efficiency. The total biomass, net photosynthetic rate, and maximal quantum yield of photosystem II (PSII) photochemistry reduced by 61.47%, 95.80%, and 16.06% in comparison with the control, respectively. Meanwhile, CD stress was accompanied by lower chlorophyll contents, down-regulated expression level of key photosynthetic enzymes (EnRbcS, EnRbcL, and EnRCA), stomatal closure, disrupted chloroplast ultrastructure, and reduced starch content. Furthermore, CD stress induced some adaptability responses in cold- and drought-tolerant E. nutans seedlings. The combined stress provoked alterations in both cold- and drought-related transcription factors and responsive genes. EnCBF12, EnCBF9, EnCBF14, and EnCOR14α were significantly up-regulated under cold or drought stress, and the transcript level of EnCBF3 and EnCBF12 was even 2.94 and 12.59 times higher than control under CD treatment, which indicated the key role of transcription factors activation in coping with CD stress. In addition, the content of soluble sugar, reducing sugar, proline, glycine betaine, and other osmolytes was significantly improved under CD stress. Therefore, we demonstrated that exposure to CD stress led to severe morphological and photosynthetic damage and revealed the acclimation to the cold and drought stress combination via osmotic adjustment and transcription factors activation in the Tibetan wild E. nutans.

1. Introduction

Plants face various adverse environmental stresses that significantly inhibit the growth and development of plants at different levels, resulting in huge losses in agricultural and animal husbandry production. It was widely reported that the proper management practices, including irrigation, fertilization, and the application of exogenous protective substances, mitigate the adverse effects of cold or drought stress on plant growth [1,2,3]. Moreover, higher plants have evolved the acclimation mechanism to cope with these stress conditions during long-term subsistence in the natural environment [4,5]. However, adverse environmental stresses rarely occur individually, and the physiological effects and adaptation mechanisms of plants under combined stress are less researched. Furthermore, the distribution of forage production region is mainly in the northwest and southwest marginal land of China, which coincides with the semi-arid continent and high-altitude frigid climate zones. Therefore, combined cold and drought (CD) stress is one of the primary environmental stressors that inhibit forage production [6].

Cold stress induces the formation of ice crystals in intercellular spaces and causes cell dehydration [7], which leads to the disintegration of the lipid bilayer and subsequent damage to the whole plant. Similarly, drought stress also causes some common features, such as the loss of membrane integrity and plasmolysis [8,9]. Both cold- and drought-stressed plants exhibit stress symptoms of chlorosis, wilt, and severely reduced plant growth [10,11]. In addition, some studies proposed the priming effect of plant exposure to combined stress, indicating that the simultaneous stress resistance was increased by the same or other stresses [12]. Nevertheless, more studies suggested that the combination of cold and drought stresses elicited more serious damage and different responses in plants compared to a single stress [13]. For instance, morpho-physiological alterations were observed under combined cold and water-deficit stresses from the growth to the reproduction stage in Arabidopsis [14]. Moreover, the maximal quantum yield of photosystem II (PSII) photochemistry (F_v/F_m) and electrolyte leakage (EL) could be used to assess the plant’s tolerance to cold–drought stress because combined stress tends to have a negative effect on photosynthesis [15,16]. In addition, transcriptome analysis identified more differential expressed genes (DEGs) in Camellia sinensis under combined stress; meanwhile, DREB1s (dehydration-responsive element-binding proteins)/CBFs (C-repeat binding factors) alleviated drought-induced leaf senescence [17]. However, more physiological indicators associated with photosynthetic capacity, osmotic adjustment, and chloroplast structure were still of limited concern. Therefore, it is worthwhile to investigate the comprehensive responses under CD stress.

Elymus nutans Griseb. is a native and dominant forage grass grown in the alpine grasslands of the Qinghai–Tibetan Plateau, China [18]. It has been widely used for livestock husbandry and environmental conservation due to its good nutrition and adaptability to cold or drought stress [19]. Therefore, analysis of the physiological modifications in E. nutans attaches more importance to expanding the knowledge of stress response than in the model plants. In our previous study, E. nutans was exposed to single cold stress, and then we determined the metabolic parameters and diffusion limitations to elucidate the critical factors for recovery from cold-stress-induced damage [20]. However, the effects on the photosynthetic mechanism and osmotic adjustment in E. nutans seedlings were poorly clarified, especially under CD stress. Here, analyses of photosynthesis, stomatal properties, chloroplast ultrastructure, osmotic regulatory substances, and gene expression of transcription factors were conducted under drought, cold, and CD stresses. Therefore, we aimed to investigate the mechanisms of CD stress response in E. nutans seedlings. Meanwhile, we hypothesized that the simultaneous stress of cold and drought triggered more severe damage than single stress and that the accumulation of osmotic adjustment substances and the activation of transcription factors enhanced the ability to cope with CD stress.

2. Materials and Methods

2.1. Plant Materials and Growth Conditions

Elymus nutans Griseb. cv. Damxung (DX) seeds were obtained from the wild plants in Damxung County, Tibet, China (30°28.535′ N, 91°06.246′ E, altitude 4678 m). The seeds were surface sterilized in 0.1% (w/v) sodium hypochlorite, washed with sterile water, and then germinated on moistened filter paper at 25 °C. Four morphologically uniform 7-day-old seedlings were selected and transferred into each lightproof plastic pot (9 cm in diameter and 15 cm in depth) containing quartz sand, which was washed and sterilized at 180 °C for 2 h before use. All plants were cultured in quartz sand by irrigating with water every day and Hoagland’s nutrient solution every three days, which was composed of 1 mM NH₄H₂PO₄, 6 mM KNO₃, 4 mM Ca(NO₃)₂, 2 mM MgSO₄, 46.25 μM H₃BO₃, 9.15 μM MnCl₄·4H₂O, 0.77 μM ZnSO₄·7H₂O, 0.32 μM CuSO₄·5H₂O, 0.45 μM H₂MoO₄·H₂O, and 50 μM EDTANa-Fe. The seedlings were placed in an artificial incubator for 28 d at a day/night temperature of 25/20 °C, relative humidity of 60%, and a 16/8 h photoperiod (day/night). The suitable photosynthetic photon flux density (PPFD) was set at 650 μmol m^–2 s^–1 according to our previous research [20].

2.2. Treatments and Experimental Design

After 4 weeks of culture, the 35-day-old seedlings were randomly grouped into four different treatment sets: control (CK), drought stress (D), cold stress (C), and combined cold and drought (CD) stress. The control plants were still cultured under the same conditions as above. The cold stress was arranged by transferring seedlings to another artificial incubator set to a day/night temperature of 4 ${}^{\circ }$C [21], with a 16/8 h photoperiod and 650 $\mathsf{\mu }$mol m^–2 s^–1 PPFD conditions described above. For drought stress, plants were treated by withholding irrigation, and each pot was weighed to maintain the field capacity at 30%. The CD-stress-treated plants were exposed to simultaneous 4 ${}^{\circ }$C and 30% of field capacity. In this experiment, a completely randomized design was arranged, and three biological replications were included in each treatment. All stress conditions were continued for 5 d to observe the stress acclimation-induced changes. Afterward, the second leaf from the top of the plants in each treatment was sampled, quickly frozen in liquid nitrogen, and immediately stored at −80 ${}^{\circ }$C for further measurements.

2.3. Measurement of Plant Growth Characteristics

The plant height and tiller number of each plant were first recorded. The leaf width of the second leaf from the top was measured by using a Vernier caliper. Then, the whole plant of each treatment was taken for the determination of root length and dried to constant weight to determine total biomass. Moreover, the fresh weight, dry weight, and turgid weight (by immersing in distilled water) were determined to calculate leaf relative water content [22].

2.4. Determination of Photosynthetic Pigments, Enzyme Activities, and Products

The chlorophyll (Chl) and carotenoids (Car) were determined using the method of Lichtenthaler [23] with slight modifications. The photosynthetic pigments were extracted using a mixed solution of 80% acetone and 95% ethyl alcohol, and then the extract absorbance was measured at 440 nm, 645 nm, and 663 nm to calculate Chl a, Chl b, total Chl, Chl a/b, and Car concentrations.

The measurement of Rubisco activity was performed according to the method of Cheng and Fuchigami [24]. Frozen leaves were homogenized in 50 mM Hepes buffer (pH 7.5, containing 10 mM MgCl₂, 2.0 mM EDTA, 1% Triton X-100, 10 mM dithiothreitol, and 0.75% polyvinyl pyrrolidone) and centrifuged at 12,000×g for 20 min at 4 ${}^{\circ }$C for extraction. The Rubisco activity was assayed by the NADH oxidation at 340 nm in the presence of ribulose 1,5-bisphosphate (RuBP).

Starch content was quantified by using the ethanol-insoluble residue method [25]. The distilled water was added to the residue and incubated at 100 ${}^{\circ }$C for 20 min. Afterward, starch was hydrolyzed by separately adding 9.2 M and 4.6 M HClO₄, mixed with anthrone reagent, and boiled at 100 ${}^{\circ }$C. The absorbance of the mixture at 620 nm was recorded to calculate starch content. The soluble protein content of the sample extract was determined using Bradford’s method [26].

2.5. Analysis of Gas Exchange and Fluorescence Parameters

The second leaf from the top of each plant was chosen to measure the Pn, gs, Ci, and Tr using the Li-6800 photosynthesis apparatus (LICOR, Inc., Lincoln, NE, USA) at 9:00 a.m. Inside the leaf chamber, the illuminance was set at 800 $\mathsf{\mu }$mol m^–2 s^–1 by the 6400-02B LED light source, the CO₂ concentration was maintained at 400 $\mathsf{\mu }$mol mol^–1 using a CO₂ cylinder, and the relative humidity was about 65%.

The F_v/F_m on the same leaf was detected using the portable pulse-modulated fluorometer (PAM 2500, Walz, Effeltrich, Germany) after 20 min of dark adaptation. Then, the leaves were respectively applied to actinic light (371 $\mathsf{\mu }$mol m^–2 s^–1) and saturation pulse to obtain the maximum and minimum fluorescence in the light. Finally, the following Chl fluorescence parameters, including F_v’/F_m’, Φ_PSII, qP, NPQ, and ETR, were calculated according to the formulas described by Oliveira and Penuelas [27].

2.6. Leaf Stomata and Chloroplast Ultrastructure Assay

The leaves of each treatment were trimmed into small pieces (3 mm²) and fixed in 4% glutaraldehyde at 4 ${}^{\circ }$C overnight. After washing in 0.1 M phosphate buffer (pH 7.2) thrice, the samples were post-fixed in 1% osmium acid at 4 ${}^{\circ }$C for 2 h. Next, they were dehydrated with an ascending series of ethanol solutions (30%, 50%, 70%, 80%, 90%, and 100% ethanol) for 15 min, respectively.

For scanning electron microscope (SEM) observation, the leaf samples were treated with critical point drying, attached to metallic stubs using carbon stickers, and sputter-coated with gold for 30 s. Finally, the stomatal status was observed and photographed by SEM (Nova NanoSEM 450, FEI, Hillsboro, OR, USA), and the stomatal characteristics (width and length) were calculated by using Image J software. For transmission electron microscope (TEM) observation, dehydrated samples were embedded in epoxy resin at 55 ${}^{\circ }$C for 48 h and cut into ultrathin sections using an ultramicrotome (EM UC7, Leica, Sydney, Austria). Then, the ultrathin sections were double-stained with uranyl acetate and lead citrate solution. Lastly, the chloroplast ultrastructure of leaf samples was observed with a TEM (HT-7800, Hitachi, Tokyo, Japan).

2.7. Measurement of Osmotic Substances

The soluble sugar content was determined using the anthrone method [28]. In brief, E. nutans leaves (0.2 g) were immersed in 10 mL of distilled water and then boiled for 20 min to extract soluble sugar. Then, the supernatant was filtered, mixed with anthrone, and incubated in boiling water for 10 min. The absorbance of the reaction solution was recorded at 620 nm after cooling to room temperature.

The sample extraction of reducing sugar was performed following the same method as soluble sugar. Afterward, the 3,5-dinitrosalicylic acid (DNS) method was utilized to determine reducing sugar content [29]. The sample extraction (1.0 mL) and an equal volume of DNS reagent were mixed well in the tubes and heated at 100 ${}^{\circ }$C for 15 min. The reaction mixture was cooled to 25 ${}^{\circ }$C, made up to 25 mL with distilled water, and the absorption at 540 nm was measured using a UV–Vis spectrophotometer (UV-1800, MAPADA, Shanghai, China).

The sucrose was extracted from frozen leaf samples in 80% ethanol (v/v) at 80 ${}^{\circ }$C, and the supernatants were collected by filtering for sucrose content analysis. Then, 1.0 mL of extract solution, 2.0 mL of resorcinol solution, and 7.0 mL of hydrochloric acid were mixed for 10 min of reaction at 80 ${}^{\circ }$C. Finally, the absorbance was determined at 480 nm to calculate sucrose content according to Buysse and Merckx [30].

The measurements of free amino acid and proline contents were conducted using the acidic ninhydrin method [31]. Leaf samples were ground with 10 mL of 3% aqueous sulfosalicylic acid, and the homogenate was centrifugated at 4000×g for 10 min. Next, two milliliters of extract were reacted with the same volume of acidic ninhydrin and glacial acetic acid for 1 h at 100 ${}^{\circ }$C. After extracting with toluene, standing, and layering, the absorbance of the organic layer was read at 520 nm and 570 nm for proline and free amino acid contents, respectively.

The determination of glycine betaine was performed by using the Grieve and Grattan [32] method with slight modifications. Plant samples (0.5 g) were homogenized with 5 mL of toluene–water mixture, shaken for 24 h at room temperature, and filtered to collect the supernatant. The reaction mixture was composed of 0.5 mL of sample extract, 1.0 mL of 2 M HCl, 0.1 mL of potassium triiodide, 2.0 mL of ice-cooled water, and 10 mL of dichloromethane. Afterward, the optical density was recorded by using the organic layer at 365 nm.

2.8. Gene Expression Analysis

Total RNA was isolated with the Plant RNA Extraction Kit (TaKaRa, Kusatsu, Japan) and exposed to absorbance analysis and gel electrophoresis to ensure its integrity. Then, high-quality RNA was reverse-transcribed to first-strand cDNA using HiScript^®III 1st Strand cDNA Synthesis Kit with gDNA wiper (R312, Vazyme, Nanjing, China) according to the manufacturer’s protocol. The primers of gene sequences were designed and listed in Table S1. Quantitative real-time PCR (qRT-PCR) was performed following the manufacturer’s instructions on QuantStudio^®7 real-time PCR system (Life Technologies, Waltham, MA, USA). The relative gene expression levels were calculated by the 2^{−$\Delta$$\Delta$Ct} method and standardized using the geometric mean of the expression of EnTublin and En18s rRNA as the internal reference [33]. Each qRT-PCR reaction included three technical replicates times, and each treatment was repeated three times using three independent RNA samples.

2.9. Statistical Analysis

Differences in each index among treatments were examined using the one-way analysis of variance (ANOVA). Means were separated by Duncan’s multiple comparison tests at the p < 0.05 threshold of significance. All statistical analysis was performed using SPSS 22 software (SPSS Inc., Chicago, IL, USA).

3. Results

3.1. Plant Growth Performance

Compared to the control, the plant height, leaf width, root length, tiller number, leaf relative water content, and total biomass significantly decreased under drought or cold stress (Figure 1). Meanwhile, the growth suppression of E. nutans seedlings exposed to cold stress was more severe than that under drought stress, indicated by the lower leaf width and total biomass. Moreover, CD stress inhibited all of these parameters more remarkably compared with those under single stressors.

3.2. Pigment Contents, Enzyme Activities, Gas Exchange Parameters, and Products in Photosynthesis

Plant photosynthetic pigment contents were remarkably decreased after drought, cold, and CD treatments for 5 d. Compared with the control conditions, chlorophyll a (Chl a), Chl b, and carotenoids (Car) contents were respectively decreased by 12.97%, 13.40%, and 13.72% under drought stress, and by 36.85%, 40.55%, and 38.45% under cold stress (Figure 2A–D). In addition, combined stress caused a more significant diminution of photosynthetic pigment contents compared with single stress. However, the application of adverse conditions observably increased the ratio of Chl a/b, especially under cold and CD stresses.

The activity of Rubisco was obviously reduced under drought, cold, and combined stresses, with values of 37.34%, 52.83%, and 69.85%, respectively (Figure 2F). Similarly, the stress treatments induced the down-regulated expression of genes encoding photosynthetic enzymes. Although drought stress slightly up-regulated the expression of EnRbcL (Rubisco’s large subunit), the transcript levels of EnRbcS (Rubisco’s small subunit), EnRCA (Rubisco activase), and EnGAPDH (Glyceraldehyde-3-phosphate dehydrogenase) dramatically decreased in drought- and cold-stress-treated plants (Figure 3). Apparently, CD stress caused a greater decline than that under single stress, except EnGAPDH.

In comparison with the control, plants exposed to drought and cold conditions resulted in a remarkable lessening of net photosynthetic rate (Pn) by 28.02% and 76.77%, stomatal conductance (gs) by 30.88% and 68.97%, and transpiration rate (Tr) by 35.38% and 82.30%, respectively. Furthermore, the maximal decrease in these traits was observed in the CD treatment, while the variation in intercellular CO₂ concentration (Ci) was the opposite among different treatments (Figure 2G–J).

The accumulation of photosynthetic products was significantly inhibited under stress conditions, including starch and soluble protein. With drought or cold treatments, starch content was greatly decreased by 12.43% and 30.68%, in comparison with control plants, and it was further reduced by 48.77% in CD-stress-treated plants. The same tendency was observed in the content of soluble protein (Figure 2K,L).

3.3. Chlorophyll Fluorescence Parameters

The maximal quantum yield of photosystem II (PSII) photochemistry (F_v/F_m), intrinsic PSII efficiency (F_v’/F_m’), actual quantum yield of PSII photochemistry (Φ_PSII), photochemical quenching coefficient (qP), non-photochemical quenching coefficient (NPQ), and relative electron transport rate (ETR) were remarkably decreased under single cold and simultaneous CD stresses, whereas drought treatment had no significant on F_v/F_m and Φ_PSII. Under CD stress, the value of ETR was reduced most, by 44.08%, while the F_v/F_m decreased least, by 16.06%, compared with the control treatment (Figure 4).

3.4. Stomatal Characteristics and Chloroplast Ultrastructure

SEM results of stomatal properties first showed that the stomatal opening was significantly decreased under drought, cold, and combined stresses. The stomatal width was greatly reduced no matter the stress conditions; even the CD-stress-treated stoma was almost closed compared to the control plants. Meanwhile, the stomatal length significantly increased under single drought or cold stress compared with the control; the largest increase was observed under CD treatment (Figure 5).

The chloroplast ultrastructure, mainly including the membrane, starch grains, and thylakoid, was observed using TEM. Under the control treatment, the chloroplast was well organized in the cell, and the stacked grana thylakoids existed. However, both deformed chloroplasts and loose grana lamellae were observed in single-stress-treated leaves, and the number of starch grains and chloroplasts significantly decreased. Meanwhile, grana lamellae were distorted and even disappeared, especially under cold stress. Compared to single-stress treatment, combined stress caused more severe damage to the whole subcellular components in E. nutans seedlings. Except for smaller chloroplasts and the disappearance of starch grains, plasmolysis occurred, and the chloroplast envelope membrane structure was destroyed under CD stress (Figure 6).

3.5. Transcript Levels of Transcription Factors

To further elucidate the molecular response to CD stress in E. nutans, some cold- and dehydration-responsive transcription factors were chosen and assayed. Drought stress significantly up-regulated the expression of EnCBF12 and EnCBF9 and down-regulated the expression of EnCBF3 and EnDREB1H while having no apparent effect on EnCBF14 and EnCOR14α (cold-responsive 14$\alpha$) expression (Figure 7). Meanwhile, cold stress induced the up-regulation of EnCBF12, EnCBF14, and EnCOR14α and the down-regulation of EnCBF3, EnDREB1H, and EnCBF9. Unexpectedly, almost all of the genes exhibited the highest transcript levels in the simultaneous presence of cold and drought stresses; the increases in EnCBF3 and EnCBF12 were 2.94 times and 12.59 times, respectively, relative to the control plants (Figure 7A,B).

3.6. Osmotic Substances

The application of drought or cold stresses caused the remarkable accumulation of sugars. Treatment of plants with drought stress increased the soluble sugar, reducing sugar, and sucrose contents by 126.23%, 78.12%, and 34.61%, compared with the control treatment, respectively. Meanwhile, a significant difference was only observed in reducing sugar between cold and drought stresses, whereas CD-treated plants showed the significantly highest level on these three parameters (Figure 8A–C).

Additionally, organic solutes varied greatly among these four treatment groups. Compared to the control, the proline content increased by 84.15% and 225.60% in drought- and cold-treated plants, respectively (Figure 8D), whereas no significant difference was observed in the content of free amino acid and glycine betaine under the application of cold or drought stress. The free amino acid and glycine betaine slightly increased by 4.22% and 16.70% in drought-treated plants and 23.29% and 13.03% in cold-stressed plants (Figure 8E,F). In addition, the maximal accumulation of these osmotic substances was presented in the combined application of cold and drought.

4. Discussion

Plants are frequently exposed to various wide-ranging and frequent existence environmental stressors, such as cold and drought, that severely inhibit their growth and yield [34,35]. In dry winters or high-altitude regions, the extremely low temperature often coincides with soil water deficit, creating combined stress conditions [36]. Elymus nutans Griseb. is a forage grass geographically distributed in the Qinghai–Tibetan Plateau and the northwestern continent of China, where it faces CD stress [37]. In this study, morphological and physiological responses of E. nutans could be observed under CD stress. We found that drought stress apparently reduced leaf relative water content, while cold stress alone mainly decreased leaf width, which might be attributed to the rapid cell dehydration caused by the drought condition. Compared to single-stress treatments, CD stress aggravated the morphological damage, as indicated by the lower plant height, root length, and total biomass (Figure 1). Similarly, drought and low-temperature stress also affected leaf area and yield in rice (Oryza sativa L.) during some growth stages [38,39]. Furthermore, cold stress caused more obvious injury relative to drought stress, especially in biomass, indicating that it was difficult to survive and recover after cold treatment in plants [40].

Photosynthesis involves photosynthetic pigments, photosystems, and photosynthetic key enzymes, which are responsible for light harvesting, electron transfer, and carbon assimilation, respectively [41,42]. However, both cold and drought stress progressively inhibited many aspects of photosynthesis, such as CO₂ availability, stomatal aperture, and chloroplast structure [21,43]. In this study, we found that cold stress alone reduced chlorophyll and carotenoid contents more than drought and also decreased Rubisco activity (Figure 2). These results indicated that cold-treated E. nutans exhibited worse tolerance than plants under water-deficit conditions. A similar observation was reported in different Brassica napus L. varieties, where water potential had a significant effect on chlorophyll, while cold-stress factors remarkably decreased pigment contents [44]. On the other hand, the Chl a/b ratio significantly increased under cold and CD stresses, which indicated a lower rate of decrease in Chl a than Chl b. This implied that the Chl b might convert to Chl a during the Chl degradation process. Considering that only Chl a has the function of light energy conversion, it might be the acclimation performance of E. nutans under CD stress. Moreover, plants subjected to cold or drought stress exhibited impaired photosynthetic characteristics, such as photosynthetic enzyme activities and expression levels. It seems that both drought and cold weakened the carbon assimilation of photosynthesis, but they affected different stages of the Calvin cycle. Drought stress mainly inhibited the reduction stage, while cold stress mainly inhibited the fixation stage, as indicated by the distinct expression levels of EnRbcL, EnRbcS, and EnGAPDH. Under CD treatment, the lowest transcript level of EnRbcL and EnRbcS was recorded, and the starch content lessened more than under single stress (Figure 3). It is well-known that both drought and cold stress inhibit this process in which the inactivation of Rubisco could disrupt photosynthesis, thereby limiting the production and output of carbohydrates in leaves [45,46].

Moreover, E. nutans seedlings exhibited reduced net assimilation rate and PSII activity, especially under combined stress (Figure 2 and Figure 4). The net photosynthetic rate under stresses can be limited by different factors. One is stomatal conductance, which controls the diffusion of ambient CO₂ into the intercellular space. Another is the CO₂ assimilation in the chloroplast stroma, which also occupies a large part [47]. In this study, the variation in Pn is consistent with gs and opposite to Ci under CD stress, indicating both non-stomatal and stomatal limitations in E. nutans (Figure 2). This is in accordance with previous research, where neither F_v/F_m nor NPQ significantly decreased under drought stress in tomato (Solanum lycopersicum L.), but individual cold and CD stress adversely affected Chl fluorescence parameters [48]. NPQ is the part of light energy absorbed by antenna pigment, which is not used for electron transfer. Higher NPQ is helpful for the dissipation of excess energy [49]. As expected, CD stress displayed the lowest Φ_PSII, ETR, and NPQ among all treatments. It reflected the reduction in dissipation proportion induced by CD stress, which in turn increased the risk of damage to the photosystem, light energy conversion efficiency, photosynthetic structures, and other physiological characteristics [48].

Given the regulatory effect of stomata in the influx of ambient air and transportation, SEM observation was conducted to further analyze the photosynthesis in E. nutans. Theoretically, the stomatal properties are critical indicators to evaluate the plant response to stress environments [50]. In this study, we found that single drought or cold stress dramatically decreased the stomatal aperture (the ratio of width to length), which is consistent with the results in Zea mays L. [51]. Drought stress induced more direct and rapid injury on stomatal closure than cold stress, which was manifested as lower stomatal width in drought-treated E. nutans. Although the stomatal width was not significantly affected by different stresses, the stomatal length was greatly increased, especially under CD stress, reflecting more serious injury on the stomatal aperture (Figure 5). Moreover, the stomatal width had a positive relationship with gs and Tr, which further implies that the decline in gas exchange was partially due to stomatal closure [52]. However, cold stress impaired most gas exchange parameters more severely than drought stress, which implied that drought might induce the early stomatal response and cold stress might cause more comprehensive and serious damage in photosynthesis.

Additionally, the chloroplast is an important organelle for energy production, and the ability and activity of photosynthesis depend on the integrity of the chloroplast ultrastructure [53]. Furthermore, abundant starch grain provides the carbon source by its decomposition and promotes stress tolerance in plants [54]. In this study, we observed that cold stress severely damaged chloroplast shapes and structures, such as irregular grana lamellae and decreased starch grains, while drought stress caused less serious changes (Figure 6). This is consistent with previous studies [55]. In addition, the Chl fluorescence characteristics were thought to be closely associated with these phenomena, indicating that cold stress resulted in more severe photosynthetic efficiency injuries via destroying chloroplast ultrastructure than drought stress. However, when it comes to CD stress, we found that the grana lamellae were disordered, and the starch grains disappeared, which was not in agreement with some other research. Guo et al. [56] emphasized that drought before cold stress ameliorated chloroplast damage in maize, while Lee and Back [57] confirmed the negative impact of CD stress. We attribute the difference to the distinct tolerance between crops and high-altitude native E. nutans; that is, this tendency of CD stress aggravating chloroplast ultrastructure damage more than single stress in tolerant E. nutans might be unique relative to sensitive crops. Based on the photosynthesis parameters, we proposed that CD stress first induced stomatal closing and distorted chloroplasts, subsequently decreasing the CO₂ acquisition, carbon assimilation ability, and photosynthetic capacity in mesophyll cells, which eventually stunted plant growth and biomass accumulation (Figure 9).

Cold and drought stresses can cause primary or secondary damage and induce some adaptability responses, which involve biochemical and signal transduction processes [58]. The perception, transduction, and response of stress signals were the foundation of stress adaptability. Some transcription factors, such as the DREB and CBF family members, have been shown to activate downstream responsive pathways and enhance drought and cold tolerance in plants [59,60]. In this study, single drought and cold stresses induced the expression of some key transcription factors with different degrees of increase. Moreover, most cold regulator factors showed the highest transcript level under CD stress (Figure 7). It has been reported that as the core network, CBFs were able to impart cold and drought tolerance by activating the target proteins (such as COR genes) [61]. For example, overexpressing the CBF3 gene conferred cold or drought resistance, which was associated with downstream genes such as COR15 [62]. In addition, AtCBF3 transgenic tall fescue drove the proline biosynthesis and thus improved the drought and cold tolerance, which was consistent with our findings [63]. It was apparently found that plants activated different transcription factors in the same family under cold or drought stress, such as EnCBF12, EnCBF9, and EnCBF14 (Figure 7). Therefore, the adaptation mechanism to cold, drought, and combined stresses in E. nutans might be attributed to the signal transduction of different transcription factors (Figure 9).

Unlike the model plants or crops, E. nutans has good acclimation to the stressors due to the long-term drought and fluctuating temperature environment, which leads to the hardening process, such as osmolytes accumulation and signal transduction [36,64]. It is widely accepted that some compatible solutes can protect the cell membrane from dehydration and disintegration under cold and drought stresses. On the one hand, soluble carbohydrates are helpful to maintain osmotic homeostasis as protectants and thus enhance plant growth performance [65]. In this study, the content of soluble sugar (including sucrose and reducing sugar) was significantly increased under single stress, and the highest content was observed in the CD treatment, while the starch content declined (Figure 2 and Figure 8). The transition of carbohydrate metabolism from starch to sucrose was crucial for cryoprotectant enhancement to cope with CD stress. On the other hand, glycine betaine and proline are nonprotein amino acids in the cytosol that are related to the improvement in osmotic adjustment under both drought and cold stress [66]. In comparison with the control, single drought or cold stress did not significantly affect the content of free amino acid and glycine betaine, whereas CD stress obviously improved all the osmolyte contents (Figure 8). Similar results were also reported by Moghimi et al. [67] and Patel et al. [16], who indicated that proline was involved in the defense mechanisms to ensure the stability of cell membranes, nucleic acids, and proteins. Thus, we inferred that the CD stress acclimation might also be partially reflected by the downstream osmotic adjustment (Figure 9).

5. Conclusions

In summary, we demonstrated that CD stress caused more serious damage to plant growth and photosynthetic performance than drought or cold stress alone. Drought stress induced a more direct response in stomata, and cold stress destroyed chloroplast ultrastructure more severely. Therefore, CD stress resulted in stomatal closure and chloroplast ultrastructure disruption, thereby reducing photosynthetic ability and efficiency and ultimately decreasing total biomass. Furthermore, we detected the up-regulation of stress-related transcription factors and the osmolyte accumulation under CD stress, which explained the adaptation of E. nutans to the adverse environment in the Qinghai–Tibetan Plateau, China.