Effects of Elevated Atmospheric CO 2 Concentration on Phragmites australis and Wastewater Treatment Efﬁciency in Constructed Wetlands

: Elevated atmospheric CO 2 concentration ( e [CO 2 ]) has been predicted to rise to more than 400 ppm by the end of this century. It has received extensive attention with regard to the pros and cons of e [CO 2 ] effects in terrestrial and marine ecosystems, while the effects of e [CO 2 ] on wastewater treatment efﬁciency in constructed wetlands (CWs) are rarely known. In this study, the atmospheric CO 2 concentration was set as 400 ppm (that is, ambient [CO 2 ]) and 800 ppm (that is, e [CO 2 ]). The physiological performance of Phragmites australis and microbial enzyme activities in constructed wetlands in response to e [CO 2 ] were tested. Signiﬁcantly higher net photosynthetic rate and plant growth were found under e [CO 2 ]. The concentrations of nitrate, total anions, and total ions in the xylem sap of Phragmites australis were reduced, while the uptake of N and P in plants were not affected under e [CO 2 ] condition. In addition, the ammonia monooxygenase activity was reduced, while the phosphatase activity was enhanced by e [CO 2 ]. The increased removal efﬁciency of chemical oxygen demand and total nitrogen in CWs could be ascribed to the changes in physiological performance of Phragmites australis and activities of microbial enzymes under e [CO 2 ]. These results suggested that the future atmospheric CO 2 concentration could affect the wastewater treatment efﬁciency in CWs, due to the direct effects on plants and microorganisms.


Introduction
Since the Industrial Revolution, the atmospheric concentration of carbon dioxide ([CO 2 ]) has risen from 280 ppm to more than 400 ppm, and it has been predicted to reach levels of 420 ppm (RCP2.6) to 1300 ppm (RCP8.5) by the end of this century (IPCC, 2013). The effects of elevated [CO 2 ] (e[CO 2 ]) on terrestrial and marine ecosystems have attracted extensive attention, because it directly affects the photosynthesis and growth of plants, soil microorganisms, and ocean acidification [1][2][3].
Net photosynthesis (A n ) has a tight relationship with e[CO 2 ] stimulation because e[CO 2 ] enhances photosynthetic carbon assimilation, whereas it inhibits photorespiration in plants [4]. The stimulations of photosynthesis in plants grown under e[CO 2 ] are mostly

Experimental Setup and Synthetic Wastewater
Six subsurface-flow CW microcosms were setup in CO 2 -controlled chambers (Kesheng, Ningbo, China) in Northlake Science Park, Changchun, China. The CWs (45 cm × 30 cm × 35 cm) were filled with artificial gravel (8-10 mm in diameter) to an even depth of 35 cm (Figure 1). The water level was 30 cm. Six Phragmites australis seedlings with uniform size (approximately 55 cm in plant height) and vigor, were transplanted from the natural environment into each CW on 15 August 2019. The Phragmites australis plants included six leaves per tiller. The Phragmites australis plants were fed with 1/10 Hoagland solution for 30 days before the formal experiment, during this period the plants developed new leaves and roots. When the formal experiment started, three CWs were moved into the chamber at elevated CO 2 concentration (e[CO 2 ], 800 ppm), while another three CWs were kept at ambient CO 2 concentration (a[CO 2 ], 400 ppm). The actual CO 2 concentration in the e[CO 2 ] and a[CO 2 ] chambers during the experimental period is shown in Figure 2. The climate conditions in the chambers were set to a day/night temperature of 26/20 • C, a relative humidity of approximately 70%, and a 16 h photoperiod with a photosynthetic active radiation of 400 µmol m −2 s −1 . Synthetic wastewater was prepared following Wiessner et al. [18], containing 61.15 mg L −1 total organic carbon (TOC), 30.8 mg L −1 total nitrogen (TN), and 2.5 mg L −1 total phosphorus (TP) with a chemical oxygen demand (COD) of 163 mg L −1 . During the experiment, 40 L of synthetic wastewater was added to each CW microcosm, and the sequencing fill and draw batch mode was applied, with each batch maintained for 6 days. After 10 batches (60 days), the samplings and measurements were initiated.

Experimental Setup and Synthetic Wastewater
Six subsurface-flow CW microcosms were setup in CO (Kesheng, Ningbo, China) in Northlake Science Park, Changchun 30 × 35 cm) were filled with artificial gravel (8-10 mm in diameter cm ( Figure 1). The water level was 30 cm. Six Phragmites australis size (approximately 55 cm in plant height) and vigor, were transp environment into each CW on 15th August 2019. The Phragmites a six leaves per tiller. The Phragmites australis plants were fed with for 30 days before the formal experiment, during this period the leaves and roots. When the formal experiment started, three CW chamber at elevated CO2 concentration (e[CO2], 800 ppm), while a kept at ambient CO2 concentration (a[CO2], 400 ppm). The actual C e[CO2] and a[CO2] chambers during the experimental period is climate conditions in the chambers were set to a day/night tempe ative humidity of approximately 70%, and a 16 h photoperiod with radiation of 400 μmol m −2 s −1 . Synthetic wastewater was prepared al. [18], containing 61.15 mg L −1 total organic carbon (TOC), 30.8 mg and 2.5 mg L −1 total phosphorus (TP) with a chemical oxygen de L −1 . During the experiment, 40 L of synthetic wastewater was ad cosm, and the sequencing fill and draw batch mode was applied tained for 6 days. After 10 batches (60 days), the samplings and m ated.

Water Samplings and Analysis
When the CWs' operation was stable, a 250 mL water sample w drain tap at 1 cm from the bottom for each CW microcosm every da day) in the typical target experimental batch. The concentrations of and TP in water samples were analyzed according to the Water and W ing and Analysis Methods [23]. The above measurements were obtai tive days. The removal efficiencies (REs) of COD, TN, NH4 + -N, and T follows:

Water Samplings and Analysis
When the CWs' operation was stable, a 250 mL water sample was collected by the drain tap at 1 cm from the bottom for each CW microcosm every day (from the influent day) in the typical target experimental batch. The concentrations of COD, TN, NH 4 + -N, and TP in water samples were analyzed according to the Water and Wastewater Monitoring and Analysis Methods [23]. The above measurements were obtained on six consecutive days. The removal efficiencies (REs) of COD, TN, NH 4 + -N, and TP were calculated as follows: where C e and C i indicate effluent and influent concentrations (mg L −1 ) of a given pa- At the end of the experiment, total leaf area per plant was measured with a leaf area meter (LI-3100, Li-Cor, Inc., Lincoln, NE, USA). The roots were cleaned with ultrapure water, and the total root length was analyzed with a WinRHIZO root analysis system (WinRHIZO 2012a, Regent Instruments Canada Inc., Montreal, Canada). Shoot and root samples were oven-dried at 70 • C for 72 h and weighed to get dry weight (DW). Then, the dry samples were ground to a fine powder for C and N analysis by the Dumas dry combustion method [24]. Herein, the dry sample (200 mg) was transferred onto tinfoil placed in an automated sample loader. Then, the sample was burned, and the N 2 was quantified when it passed a conductivity cell. The total P concentration was measured after ashing at 450 • C and solubilizing in hydrochloric acid and nitric acid by spectrophotometric analysis using a spectrophotometer at 410 nm after the addition of ammonium vanadomolybdate reagent [25]. Ammonium molybdate (25 g) was dissolved in boiling water (that is, Solution A). Ammonium vanadate (1.25 g) was dissolved in boiling water and diluted with nitric acid (that is, Solution B). Subsequently, Solution A was added into Solution B followed by dilution with deionized water, namely ammonium vanadomolybdate reagent.

Analysis of Xylem Sap Constituents and Leaf Water Potential
At the end of the experiment, midday leaf water potential was measured with a pressure chamber (Soil Moisture Equipment Corp., Santa Barbara, CA, USA) on fully expanded upper canopy leaves. Xylem sap was collected by pressurizing the roots of the Phragmites australis plants in a Scholander-type pressure chamber. The underground part of the plant was sealed in the pressure chamber, and the shoot was detopped at 5 cm from the stem base. Xylem sap (1 mL) was stored in an Eppendorf-vial wrapped with aluminum foil. All sap samples were frozen immediately after sampling and stored at −80 • C until analysis. The anions and cations in the xylem sap were measured by ion chromatography (Metrohm, Herisau, Switzerland). The anions were determined on a Metrosep A Supp 4 analytical column, and the cations were determined on a Metrosep C4-100 analytical column. For the above measurements, the anions included F − , Cl − , SO 4 2− , NO 3 − , NO 2 − , PO 4 3− , and Br − , and the cations included Li + , Na + , K + , Ca 2+ , Mg 2+ , and NH 4 + .

Gas Exchange
Photosynthetic rate (A n ) and stomatal conductance (g s ) were measured every day before water samplings for the target experimental batch. A portable photosynthesis system equipped with a 6400-02 LED leaf chamber (LI-Cor, Lincoln, USA) was used to measure the above indicators of the fully expanded upper canopy leaf (the marked leaves in each treatment). Before measurement, the plants were stimulated with environmental light for 2 h. The leaf chamber provided a controlled environment, maintaining a photosynthetic photon flux density (PPFD) of 1200 µmol m −2 s −1 and a CO 2 concentration of 400 µmol mol −1 . The above measurement was carried out on six consecutive days. On each measurement occasion, three leaves were taken for each treatment.

Microbial Enzyme Activities
The dehydrogenase (DHA) activity was determined by the methods of Wang et al. [27] and Hu et al. [28]. The urease (URE) activity was determined according to Klose and Tabatabai [29]. The ammonia monooxygenase (AMO) activity was measured following the method of Zheng et al. [30]. Phosphatase (PHO) activity was analyzed according to the method of Schinner and Mersi [31] and Hu et al. [28].

Statistical Analysis
All data were subjected to the one-way ANOVA using the SigmaSATA (V3.5, Systat Software, San Jose, USA). Duncan's multiple range test was applied to assess the differences between [CO 2 ] treatments at a significance level of p < 0.05.

Plant Growth and Photosynthesis as Affected by e[CO 2 ]
After a 66-day incubation under different [CO 2 ] conditions, the e[CO 2 ] plants had significantly higher leaf area, shoot DW, and root DW than that under a[CO 2 ], while total root length did not appear to be remarkably different in e[CO 2 ] plants ( Table 1). The e[CO 2 ] induced the enhancement of C content in plants, while no significant difference was found in contents of N and P between the two [CO 2 ] treatments. The gas exchange parameters of Phragmites australis plants were measured every day for the target experimental batch. The A n in the upper canopy leaves was markedly increased by 51%-97% under e[CO 2 ] treatment ( Figure 3). However, the g s was markedly decreased by 37%-51% under e[CO 2 ] treatment. Table 1. Leaf area; total root length; dry weight (DW); and C, N, P content of Phragmites australis grown under ambient (a[CO 2 ], 400 ppm) and elevated (e[CO 2 ], 800 ppm) atmospheric CO 2 concentrations. Different small letters mean significant difference at p < 0.05 level. Data are expressed as means ± SEM (n = 3).

Treatment
Leaf Area (cm 2 Plant −1 )  (Table  induced the enhancement of C content in plants, while no significant differe in contents of N and P between the two [CO2] treatments. The gas exchan of Phragmites australis plants were measured every day for the target exper The An in the upper canopy leaves was markedly increased by 51%-97% treatment ( Figure 3). However, the gs was markedly decreased by 37%-51% treatment.

Xylem Sap Constituents as Affected by e[CO2]
In the e[CO2] condition, Phragmites australis plants had a decline in the tration in xylem sap (

Microbial Enzymes as Affected by e[CO 2 ]
Four microbial enzyme activities showed different expression levels, with the significant decrease of AMO activity, the significant increase of PHO activity, and no significant difference in the activities of DHA and URE being seen in e[CO 2 ] plants ( Figure 5).

Microbial Enzymes as Affected by e[CO2]
Four microbial enzyme activities showed different expression levels, with the sign icant decrease of AMO activity, the significant increase of PHO activity, and no significa difference in the activities of DHA and URE being seen in e[CO2] plants ( Figure 5).

Treatment Performance of CWs as Affected by e[CO2]
A sharp decrease was found in the effluent concentrations of COD on Day

Treatment Performance of CWs as Affected by e[CO 2 ]
A sharp decrease was found in the effluent concentrations of COD

Effects of e[CO2] on Plant Performance in CWs
As a major part of CWs, the plants play central roles in growth and interaction with other factors in CWs directly affec [28]. The physiological processes including photosynthesis, ion u metabolism would be influenced by e[CO2] [6,7,13].
The changes in plant physiological status were expected to bial activity and contaminant removal in CWs [22]. In present s

Effects of e[CO 2 ] on Plant Performance in CWs
As a major part of CWs, the plants play central roles in pollutant removal; their growth and interaction with other factors in CWs directly affect the CWs' performance [28]. The physiological processes including photosynthesis, ion uptake, and carbohydrate metabolism would be influenced by e[CO 2 ] [6,7,13].
The changes in plant physiological status were expected to play a role in the microbial activity and contaminant removal in CWs [22]. In present study, the Phragmites australis plant growth had an increase trend in the presence of e[CO 2 ], showing a larger leaf area and a higher biomass in e[CO 2 ] plants. However, the contents of N and P were similar in the plants, regardless of atmospheric CO 2 concentration, though e[CO 2 ] had significantly higher C content, indicating that the uptake and accumulation of N and P were not changed though the biomass of Phragmites australis plants was obviously increased by e[CO 2 ]. This should be ascribed to the dilution effect. The dilution of ions by the enhanced plant growth has been documented well in various plant species including crops and annual grass under e[CO 2 ] [32][33][34].
The promoted plant growth of Phragmites australis in CWs was caused by the higher photosynthetic carbon assimilation in the presence of e[CO 2 ]. The A n in the upper canopy leaf was significantly higher by 51%-97% under e[CO 2 ], in contrast to that under a[CO 2 ]. This indicated that e[CO 2 ] enhanced the plant growth in CWs by increasing the contribution of photosynthesis to the carbohydrate accumulation in Phragmites australis. This e[CO 2 ]induced plant growth promotion has been reported in most kinds of plant species [35][36][37][38]. It should be also noted that the g s of Phragmites australis in the e[CO 2 ] condition was significantly reduced, in contrast to that in a[CO 2 ] control. This was because the high [CO 2 ] triggered the stomatal closure, thus leading to the lower g s [39]. In addition, the e[CO 2 ]induced stomatal closure also resulted in reduced transpiration rate of Phragmites australis in CWs, inducing a corresponding decrease of the ions mass flow to the root surface [12], thereby lowering ion absorption including that of N and P. The nitrate concentration in the xylem sap of Phragmites australis was decreased by e[CO 2 ], while the concentrations of phosphate and ammonium were not affected. In addition, the concentration of total ions and leaf water potential were both reduced by e[CO 2 ]. Though no clear correlation between the contents of N and P in Phragmites australis and e[CO 2 ] was found in our results, the changes in xylem sap constituents showed a trend that e[CO 2 ] might reduce the level of total ions, hence affecting the N and P uptake and accumulation of plants in CWs. In this study, the key carbohydrate metabolism enzymes exhibited different expression levels in the presence of e[CO 2 ]. Sucrolytic enzyme activity (cwInv) was reduced by e[CO 2 ], while the starch-biosynthesis-related enzyme (PGM) was enhanced by e[CO 2 ] in the leaves of Phragmites australis in CWs. This was in agreement with the finding in sugarcane that the cwInv gene appeared to be downregulated in leaves under e[CO 2 ] [40]. For the leaf PGM activity, a similar response to e[CO 2 ] was found in Triticum aestivum [7]. However, the PGM activity in Brassica napus was weakly but not significantly increased in the e[CO 2 ] condition, suggesting that the e[CO 2 ]-induced PGM activity in leaves might be plant species dependent [26]. In the sucrose biosynthesis pathway, the UGPase activity in leaves of Phragmites australis was significantly enhanced by e[CO 2 ]. UGPase is a regulator for UDP-glucose metabolism, and UDP-glucose is a precursor of sucrose, starch, and structural polysaccharides [41]. Thus, higher UGPase activity would benefit sucrose biosynthesis in Phragmites australis under e[CO 2 ]. In the glycolysis pathway, the PGI activity was the only enzyme activity enhanced by e[CO 2 ]. Sucrose is synthesized in the cytosol from fructose-6-phosphate, and PGI is a key member of the enzyme cascade for this process, which is responsible for reversible conversion of fructose-6-phosphate to glucose-6-phosphate [41]. Thus, higher PGI activity under e[CO 2 ] indicated that fructan metabolism in the leaves was promoted by e[CO 2 ], which benefited the plant growth in CWs.

Effects of e[CO 2 ] on Microbial Enzyme Activity in CWs
Several microbial enzymes, such as DHA, URE, AMO, and PHO, play key roles in the microbial physiological metabolism processes, hence affecting the pollutant removal in CWs [28]. Among the microbial enzymes, DHA is responsible for the biodegradation of organic compounds [42]. Here, a remarkable effect on DHA activity in the e[CO 2 ] condition was found, indicating that the process of organic compound biodegradation was similar in CWs, whether it was under a[CO 2 ] or e[CO 2 ]. URE and AMO are involved in organic N transformation and nitrification [43,44]. Compared with that under a[CO 2 ], AMO activity was decreased in CWs under e[CO 2 ], while URE activity was not affected. This result indicated that the changed organic N nitrification might affect the TN removal in CWs under e[CO 2 ]. PHO is the key enzyme involved in the process of P transformation [43]. Interestingly, PHO activity showed a remarkable increased trend in the e[CO 2 ] condition, which might affect the TP removal in CWs.

Effects of e[CO 2 ] on Treatment Performance of CWs
Changes in plants and microorganisms bearing diverse physiological functionalities would directly affect the wastewater treatment efficiency in CWs [28]. In the present study, the RE of COD in CWs was significantly promoted by e[CO 2 ]. This could be attributed to the e[CO 2 ]-induced promotion in plant growth and the microorganisms' activities. The biodegradation of organics is driven by aerobic and anaerobic microorganisms, which is related to the oxygen availability in CWs. It should be noted that DHA activity was less influenced by e[CO 2 ], which is not in line with the difference in RE of COD. This suggested that the changes in DHA activity could not explain all difference in COD removal under various [CO 2 ] conditions. For the N removal, e[CO 2 ] significantly increased the RE of TN, while it had no significant effect on the RE of NH 4 + -N. It has been well documented that the microbially driven processes of nitrification and denitrification are the pivotal players for N removal in CWs [45]. Under e[CO 2 ], nitrification was enhanced in CWs, as exemplified by the higher AMO activity, which benefited the N removal in CWs. This could be also related to the changes in pH of soil and water induced by e[CO 2 ] [46]. The prevailing consensus is that combined e[CO 2 ] and higher temperature increases nitrification rates and shifts the community structure of nitrifying communities [46,47]. Here, it was found that the concentration of NH 4 + -N in the e[CO 2 ] condition was slightly lower than that in the a[CO 2 ] condition. However, a current study indicated that e[CO 2 ] had no remarkable effect on the composition and activity of nitrifier communities in soil, which was not in line with the results in the present study [48]. The different responses of nitrifier communities should be ascribed to the variations in soil moisture content. The prevailing viewpoint is that the inorganic carbon does not directly act as the carbon source of denitrifying microorganisms, and it does not directly affect the denitrification. Hence, e[CO 2 ] could have an effect on denitrification via regulating physicochemical and biological environments in waterbodies, mainly including three environmental factors, namely, pH, dissolved oxygen, and NO 3 − availability. The denitrifying enzyme activities and NO 3 − availability depend on the change of pH caused by e[CO 2 ] in waterbodies, thereby dominating denitrification efficiency. In addition, the RE of TP in CWs was not significantly affected by e[CO 2 ], which agreed with the similar P uptake of Phragmites australis under both conditions. Higher acquisition of P under e[CO 2 ] was reported in some crop species, but it depended on the physiological P-acquiring mechanisms of crops and the soil P condition [49,50].

Conclusions
To capture the impacts of e[CO 2 ] (800 ppm) on N and P removal in CWs, aquatic plant performance, microbial enzyme activities, and pollutant removal were investigated in the e[CO 2 ] condition. It should be noted that e[CO 2 ] significant promoted photosynthetic carbon assimilation and plant growth, in relation to a[CO 2 ] in our study. The uptakes of N and P in plants were not affected by e[CO 2 ], while the concentrations of xylem sap constituents were changed. This indicates that plant uptake of N and P is not tightly correlated with increased plant biomass under e[CO 2 ], and e[CO 2 ]-induced stomatal closure causing reduced transpiration is responsible for the reduced absorption of N and P in plants. In addition, the ammonia monooxygenase activity declined in the e[CO 2 ] condition, whereas the phosphatase activity increased. The increased removal efficiency of COD and TN in CWs could be ascribed to the changes in physiological performance of Phragmites australis and activity of microbial enzymes under e[CO 2 ]. This study provides compelling evidence that future atmospheric CO 2 concentration could affect the wastewater treatment efficiency in CWs, due to its direct effects on plants and microorganisms.