The Effect of Combined Drought and Temperature Stress on the Physiological Status of Calcareous Grassland Species as Potential Candidates for Urban Green Infrastructure

Nature-based solutions are promising for climate adaptation and environmental management in urban areas, but urban conditions are stressful for vegetation. In particular, the interaction of drought and high temperatures may be detrimental. Guiding plant selection for urban greening with native species requires a far better knowledge of plant adaptations and stress acclimation. We tested the physiological responses of four candidate calcareous grassland species for green roofs and walls to the combined effects of drought and high temperatures under controlled conditions. The tested species proved relatively resistant to stress despite different strategies to protect the photosynthetic apparatus, maintain water balance, and repair damages. Based on the physiological responses, we rank the species in descending order of resistance to the stress factors tested: Trifolium medium > Festuca ovina > Carex flacca > Potentilla reptans, but all four can serve as potential candidates for green walls and roofs. Physiological stress screening of plant species for use on green roofs and walls supplements the habitat template approach to provide a stronger and wider base for prioritizations.


Introduction
Urban nature-based systems such as green roofs and walls, rain gardens, bioswales, and parks have the potential to support stormwater management, reduce the urban heat island effect, and promote biodiversity and esthetic qualities [1][2][3]. However, such uses of nature-based solutions to support urban climate change adaptation place high demands on vegetation's adaptive capacity and ability to acclimatize to fluctuating and combined stressors. Heat, air and soil pollution, fluctuating soil moisture, and limited rooting volumes combine to create complex urban stresses with negative impacts on vegetation [4]. In particular, extreme weather events during the growing season, such as the combination of drought and high temperatures, can have a detrimental effect [5]. This requires species that are resilient enough to withstand stressful conditions over multiple growing seasons, especially for the limited soil volumes and harsh environmental conditions of green roofs and walls [6,7], and all this needs to be in a low-input, low-maintenance environment where urban greening contributes to carbon-efficient and sustainable urban solutions [8].
To overcome these limitations and provide ecosystem services, plant selection should meet several requirements [9]. The selected plants should have minimal fertilizer and water requirements and contribute to native urban biodiversity. Species must also have

The Effect of Drought and Temperature Stress on Growth, Gas Exchange, and Pigment Content
Simulated drought and heat did not cause any significant decrease in biomass or protein content in the shoots of the investigated plant species, with the exception of Trifolium medium (Tm) ( Table 1). In Festuca ovina (Fo), an increase in biomass was measured in the drier variants compared to the control. On the other hand, a significant decrease in the content of chlorophylls and carotenoids was observed in all species under the influence of heat stress. The decrease in the content of plant pigments was accompanied by a significant decrease in the values of gas exchange parameters, such as transpiration and CO 2 assimilation. At the same time, Fo and Tm showed an increase in water use efficiency in the drought variants compared to the control (Table 1). Table 1. The effect of combined drought and heat on growth, pigment concentration, and gas exchange of investigated plant species. Data are means ± SE (n = 3 for dry weight, protein, and pigments; n = 5 for gas exchange parameters). Means followed by the same letter for each parameter are not significantly different from each other using the Fisher LSD test (p < 0.05). Abbreviations: Cf-Carex flacca, Fo-Festuca ovina, Pr-Potentilla reptans, Tm-Trifolium medium.

The Impact of Drought and Temperature Stress on Photosynthetic Apparatus
The strongest negative impact of drought on the efficiency of photosystems was measured in Fo, where the significant decrease in chlorophyll fluorescence parameters was observed compared to the control (Figure 1). At the same time, in both tested angiosperm species, Potentilla reptans (Pr) and Tm, a significant increase in dissipated energy (ϕDo) was measured with increasing drought. Drought and heat had the least impact on the Carex flacca (Cf) photosynthetic apparatus among the tested species (Figure 1). Tm-Trifolium medium, RWC-relative water content, F0-minimal fluorescence, Fm-maximal fluorescence, Fv-variable fluorescence, φP0-maximum quantum yield of primary photochemical reactions, φD0-quantum efficiency of energy dissipation, ψE0-values of probability that a trapped exciton moves electron into the electron transport chain beyond QA − , φE0-quantum efficiency of electron transfer from QAto electron transport chain beyond QA − , δR0-probability that an electron is transferred to reduce end electron acceptors at the PSI acceptor side, φR0-quantum yield for the reduction of terminal electron acceptors on the acceptor side of PSI.
Models of energy flow through the photosystem II (PSII) of excited cross sections (CS) of leaves enable a quantitative and qualitative comparison of the efficiency of the light phase of photosynthesis between the studied species and their response to drought and heat stress ( Figure 2). Under the control conditions, the Tm and Fo photosystems showed similar efficiency, while the Pr showed the worst. The analysis using the model confirmed a lack of significant differences in most parameters of Cf under the influence of drought. At the same time, it was shown that the best quantum yield in the conditions of the strongest drought compared to the other species tested was characterized by PSII in Tm ( Figure 2). Tm-Trifolium medium, RWC-relative water content, F 0 -minimal fluorescence, Fm-maximal fluorescence, Fv-variable fluorescence, ϕP 0 -maximum quantum yield of primary photochemical reactions, ϕD 0 -quantum efficiency of energy dissipation, ψE 0 -values of probability that a trapped exciton moves electron into the electron transport chain beyond Q A − , ϕE 0 -quantum efficiency of electron transfer from Q A to electron transport chain beyond Q A − , δR 0 -probability that an electron is transferred to reduce end electron acceptors at the PSI acceptor side, ϕR 0 -quantum yield for the reduction of terminal electron acceptors on the acceptor side of PSI.
Models of energy flow through the photosystem II (PSII) of excited cross sections (CS) of leaves enable a quantitative and qualitative comparison of the efficiency of the light phase of photosynthesis between the studied species and their response to drought and heat stress ( Figure 2). Under the control conditions, the Tm and Fo photosystems showed similar efficiency, while the Pr showed the worst. The analysis using the model confirmed a lack of significant differences in most parameters of Cf under the influence of drought. At the same time, it was shown that the best quantum yield in the conditions of the strongest drought compared to the other species tested was characterized by PSII in Tm ( Figure 2). White circles inscribed in squares represent reduced QA reaction centers (active), black (or orange) circles represent nonreduced QA reaction centers (inactive), 100% of the active reaction centers responded with the highest mean value observed in the control conditions. Means followed by the same letter for each parameter are not significantly different from each other using the Fisher LSD test (p < 0.05). Letters are inscribed into arrows, except for RC/CS, where they are placed in a box in the bottom right corner of the square with circles.

The Effect of Drought and Temperature Stress on Enzyme Activity and MDA Content
The level of oxidative stress and associated enzymatic activity in response to drought and heat were definitely species-specific ( Figure 3). Sedge under dry conditions did not show an increase in APX activity, with a significant increase in CAT and GR activity and the highest MDA concentrations among all the tested species under the influence of the tested stresses. Qualitatively similar responses to drought were noted for Fo and Pr, although the highest activity of APX and GR was measured in Pr, and CAT activity was highest in Fo, in response to drought. In the case of clover, there was no significant increase in APX and CAT activity under the influence of increasing drought, with a simultaneous decrease in GR activity; however, the effect of drought was a significant increase in the content of MDA in Tm leaves at the end of the experiment compared to the control ( Figure 3). White circles inscribed in squares represent reduced Q A reaction centers (active), black (or orange) circles represent nonreduced Q A reaction centers (inactive), 100% of the active reaction centers responded with the highest mean value observed in the control conditions. Means followed by the same letter for each parameter are not significantly different from each other using the Fisher LSD test (p < 0.05). Letters are inscribed into arrows, except for RC/CS, where they are placed in a box in the bottom right corner of the square with circles.

The Effect of Drought and Temperature Stress on Enzyme Activity and MDA Content
The level of oxidative stress and associated enzymatic activity in response to drought and heat were definitely species-specific ( Figure 3). Sedge under dry conditions did not show an increase in APX activity, with a significant increase in CAT and GR activity and the highest MDA concentrations among all the tested species under the influence of the tested stresses. Qualitatively similar responses to drought were noted for Fo and Pr, although the highest activity of APX and GR was measured in Pr, and CAT activity was highest in Fo, in response to drought. In the case of clover, there was no significant increase in APX and CAT activity under the influence of increasing drought, with a simultaneous decrease in GR activity; however, the effect of drought was a significant increase in the content of MDA in Tm leaves at the end of the experiment compared to the control (Figure 3).

Principal Component Analysis and Corelations
The PCA analysis confirmed that the tested species were characterized by different mechanisms of response to drought stress, because they formed separate groups arranged in a clear gradient consistent with the stress intensity ( Figure 4a). Moreover, the correla-tions between the examined physiological parameters revealed species-specific patterns of response to drought and heat stress (Figure 4b-e).

Principal Component Analysis and Corelations
The PCA analysis confirmed that the tested species were characterized by different mechanisms of response to drought stress, because they formed separate groups arranged in a clear gradient consistent with the stress intensity ( Figure 4a). Moreover, the correlations between the examined physiological parameters revealed species-specific patterns of response to drought and heat stress (Figure 4b-e).

Discussion
All species studied were affected by the drought and temperature treatments, and in most cases, the symptoms were more pronounced as the water deficit increased. Nevertheless, the response to the combined drought and heat stress was very species-specific, although our hypothesis of better adaptation to the combined stress was not supported in the monocotyledonous species [25]. Rayner et al. [26] reported that in a hot climate environment where summers are hot and dry, only plants with leaf succulence survived after 42 weeks of cultivation on the green roof without irrigation. Given extreme weather events, one might assume that alien species could be the solution. However, when creating plant communities for green infrastructure, the risks of biological plant invasions should be taken into account, including the selection of initial planting, maintaining a community that is not dominated by invaders, and preventing the spread of invasive plants [27]. In light of the above, it is important to note that urban ecosystems in general are not only entry points for many non-native species, but can also become focal points for dispersal into surrounding landscapes [28]. Consideration of potential risks in habitat design, together with conservation of local biodiversity, could be the solution. Species that could fit green infrastructure habitat templates are often found in rocky habitats, dunes, or other open areas where harsh conditions prevent forest formation [16].
In this work, all plants were possessed according to the habitat template, but they belonged to different functional groups. A comprehensive study of the physiological response of plants to drought and high temperatures is important for plant selection and/or irrigation schedules. It is not easy to directly determine which plant species is most sensitive to a combined drought and heat treatment based on individual parameters.
Carex flacca (Cf) was characterized by a significant decline in photosynthetic parameters, even when drought was not the least limiting factor. There was a huge decrease in gas exchange parameters, especially when comparing the control and the 30% RWC treatment. In drought conditions, the plant produces ROS, while as a response, antioxidants, flavonoids and secondary metabolites play a protective role in detoxifying the plant ROS and protect the plant against stress conditions by stabilizing the proteins and amino acids [29]. It seems that the combination of heat and drought stress increases the activity of glutathione reductase (GR) in the plant, while ascorbate peroxidase (APX) is not involved at all in the mitigation of stress. Volk et al. [30] reported for Cf that even small amounts of water in the soil can increase biomass production and gas exchange parameters. There is no further information in the literature on the response of this species to drought and temperature, but Gretter et al. [31] listed this plant as a potential species for extensive green roofs.
Festuca ovina (Fo) showed an interesting pattern, mainly associated with significant damage to the photosynthetic apparatus, which was mainly visible in the phenomenological energy fluxes of the leaves in the pipeline models. It seems that this plant could not maintain the efficiency of the photosynthetic apparatus at the same level under higher water deficit. Despite this fact, the reduction in efficiency at the lowest RWC was not visible in biomass production, as it remained unchanged compared to the control. In contrast to Cf, catalase (CAT) was the leading enzyme in the antioxidant apparatus, next to GR, which was much lower. Interestingly, the MDA content in Fo was lower regardless of the water treatment, especially considering the similar values in the control. Wang et al. [32] reported for different fescue species that F. ovina is characterized by high drought and heat tolerance regardless of the genotypic variation studied. Moreover, Khoshkholghsima and Rohollahi [33], comparing F. ovina with other grass species in Iran under drought stress, found a significant reduction in biomass only at higher water deficits. Using F. ovina on extensive rooftops, Vahdati et al. [34] noted that F. ovina might have problems establishing itself due to spring and autumn cold, which in turn could lead to midseason drought problems.
For Potentila reptans (Pr), it was found that drought treatments did not significantly affect its biomass production. There were visible symptoms in photosynthetic apparatus activity and efficiency; however, those are minor and related to the highest water deficit, particularly visible in gas exchange parameters. Considering antioxidative enzyme activities, all of those investigated were increased (APX, GR, CAT), while malone di-aldehyde (MDA) content was the lowest, and at the same level as for Trifolium medium (Tm) compared to the rest of the plants studied. There is not much information about this species in terms of physiology and cooperation with drought and/or heat stress, but it was directly suggested as a green roof species by Seyedabadi [35]. They reported that although Potentilla reptans (Pr) has excellent cold and drought resistance, its CO 2 uptake performance is poor, which is also true for this study compared to other species studied. This makes it an unfavorable plant for use cases where the reduction of CO 2 emissions is the main objective.
Trifolium medium (and other Trifolium species) are probably the most commonly used plant for green roofs in European climates [36,37]. T. medium was characterized by a significant decrease in photosynthetic parameters, but only at the highest water deficit, which was also true for the biomass produced by these plants. There is an interesting pattern for the antioxidant mechanism based mainly on the contribution of CAT; however, these values are much lower compared to other species studied, while APX reaches the lowest value among all plants studied, even for the control. The lowest value of antioxidant enzyme activity could be explained by the data of Li et al. [38], who showed that in Trifolum repens during prolonged drought, antioxidant enzyme activities increase in the first days of drought, then decrease and remain unchanged during the recovery period. Reynolds-Henne et al. [39] showed that the stomatal response of Trifolium to heat and drought is complex: legumes respond to moderate stress by closing stomata or irregular conductance and to high temperature stress by increasing stomatal conductance, even under drought. While the response to moderate overall stress indicates a tendency to favor water balance, the response to higher stress suggests that the metabolic effects involving protection of the photosynthetic apparatus from heat are limited. This observation was also confirmed in this work by the increase in water use efficiency (WUE), with a simultaneous decrease in stomatal conductance and transpiration. All the plants studied showed different mechanisms of coping with drought and heat stress, as well as different levels of resistance.

Experiment Design
The soil substrate for the pot experiment consisted of 70% sand (w/w), 5% garden compost (w/w), and 25% soil (w/w), collected from the calcareous grassland near Bytom, Poland. The collected soil was air-dried and passed through a 4 mm sieve to remove stones and plant debris. The soil substrate was manually mixed. The physicochemical characteristics of soil substrate are presented in Table 2 In total, 48 pots were planted (4 species × 3 treatments × 4 repetitions). In this paper, drought was defined as a decrease in relative water content from 70% in the control to 50% and 30% in experimental drought treatments. Plant material was collected during the field visit to the Upper Silesia and Cracow Jura calcareous grasslands (17 sites) and propagated from wild-collected material in the nursery established in the Institute for Ecology of Industrial Areas backyard. After plant acclimation during one growing season, seedlings were selected and transfer to the pot experiment. Plant size was uniform within species. Plants were grown in a phytotron under controlled conditions: temperature 22/16 • C (day/night), light intensity PAR = 300 µmol (photons) m −2 s −1 , photoperiod 16/8 h, and relative humidity around 40%. Pots were randomized on one of three levels of soil moisture control with relative water content (RWC) of 70%, 50%, 30%, with four replicates per species and treatment. During the first 6-week growth period, a relative water content (RWC) of 70% was applied in all pots. In the next phase of the experiment, irrigation was differentiated to achieve 70%, 50%, and 30% of RWC. Soil moisture was monitored daily with a WET sensor (Delta-T Devices, UK). Data from the device (volumetric water content) were calculated to RWC using calibration curve obtained for the tested soil substrate. The amount of irrigation water provided daily was based on the empirically measured full water capacity and the daily RWC measurement ( Figure S1). After cultivating the plants under differentiated conditions for three weeks, an additional stress factor was applied to all treatments, including control, by increasing the air temperature to 30 • C for three days, which was defined as high temperature (heat) stress. Subsequently, the photosynthetic rate, intercellular CO 2 , transpiration, stomatal conductance, chlorophyll a fluorescence, and pigment content of the plants were measured. At day 3 of heat treatment, samples for enzymatic activities were collected and plant shoot biomass was determined.

Soil Substrate Physicochemical Parameters
The pH of the soil substrate was measured in H 2 O (ratio 1:2.5 m/v) and KCl using a combined glass/calomel electrode (OSH 10-10, METRON, Gliwice, Poland) and a pH-meter (CPC-551, Elmetron, Gliwice, Poland) at 20 • C. Electrical conductivity was determined with an ESP 2ZM electrode (EUROSENSOR, Gliwice, Poland) according to the Polish standard [40]. Soil texture was determined using the hydrometric method according to the Polish standard [41]. The total nitrogen concentration in the soil was determined using the dry combustion method [42]. The concentrations of available phosphorus and available potassium were determined according to the method described by Egnér et al. [43]. Total nitrogen (N) concentration in plants was measured by the titration method [44], while total phosphorus (P), potassium (K), calcium (Ca), and magnesium (Mg) concentrations in plants were determined by ICP (Liberty 220, Varian, Palo Alto, CA, USA) in previously mineralized samples (microwave sample digestion, ETHOS 1, Milestone, Sorisole, Italy) according to the procedure specified by the manufacturer (concentrated HNO 3 and H 2 O 2 , 4:1 v/v). The sulfur content (S) was determined according to the method [45].

Plant In Vivo Measurements
Gas exchange measurements were performed using an infrared gas analyzer (LCpro SD, ADC Bioscientific, Hoddesdon, UK). Saturated photosynthetic rate (A), transpiration rate (E), and stomatal conductance (gs) were measured as previously described by Rusinowski et al. [46]. The LCpro SD was equipped with a narrow chamber (580 mm 2 ) set at 22 • C, 1500 µmol m −2 s −1 for photosynthetically active radiation (PAR), and a CO 2 concentration of about 400 ppm. Chlorophyll a fluorescence was measured using Handy Plant Efficiency Analyser (Hansatech Instruments Ltd., Norfolk, UK). Before analysis, leaves were dark-adapted for 25 min using specially designed clips (LC, Hansatech Instruments, Ltd., Norfolk, UK). Subsequently, a saturated pulse of 3500 µmol m −2 s −1 was applied for 1s to obtain readings. Five measurements were performed for single treatment on fully developed leaves.

Plant Enzymatic Activity
The activities of catalase (CAT), glutathione reductase (GR), and ascorbate peroxidase (APX) were determined according to the method described in Noctor et al. [47]. The enzyme activity was expressed as micro-(CAT) and nanomoles (APX and GR) per milligram of protein analyzed in the plant extract per minute. The protein content of the desalted plant extracts was determined according to the Bradford [48] method.
The content of chlorophyll a and b was determined colorimetrically [49], as was the content of carotenoids [50] using DMSO as the extraction agent. The amount of malondialdehyde (MDA) in the plants was determined spectrophotometrically according to the method of Hodges et al. [51]. Here, as well as in the case of plant pigments, the concentration in the homogenate was calculated to one gram fresh weight. Three measurements were performed for single treatment on leaf samples.

Data Analysis
The statistically significant differences among mean values were determined using one-way ANOVA and post hoc Fisher LSD test (p < 0.05). The statistical analysis was performed using the computer software Statistica v.13.1 (Dell Inc., Round Rock, TX, USA) and R v.4.2.0. (R Foundation for Statistical Computing, Vienna, Austria). The pipeline model of energy fluxes through leaves' cross sections was made using CorelDRAW X6 (Corel Corp., Ottawa, ON, Canada). PCA was used to identify the dominant groups of the factors that determine and describe the variety differences. Relative coordinate spatial distribution maps of assimilable elements were constructed based on plot elemental analysis data and inverse distance weighted (IDW) interpolation.

Conclusions
The results obtained did not allow for a clear verification of the research hypothesis, which assumed a higher resistance of monocotyledonous species compared to dicotyledonous ones to combined drought and high temperature stress. The selected plant species are well-but variably-adapted to the environmental stresses of urban environments such as drought and high temperatures. Trifolium medium and Festuca ovina were found to be the most resistant species to drought and high temperature stress, based on principal component 1 (PC1), which identifies these species as the most photosynthetically efficient and productive in terms of biomass (right side of PCA coordinate system). In contrast, the species that responded worst to stress were Potentilla reptans and Carex flacca. Interestingly, monocotyledons and dicotyledons differ in oxidative stress parameters (bottom versus top of the PCA coordinate system). Nevertheless, all tested species proved to be relatively resistant to stress despite different strategies, which predestines them as species that can be used in nature-based solutions such as green walls and roofs.
Supplementary Materials: The following supporting information can be downloaded at https:// www.mdpi.com/article/10.3390/plants12102003/s1, Figure S1: Irrigation curve based on sensor measurements between full water capacity and fully dry soil.