Effect and Mechanism of Applying Myriophyllum Verticillatum for Reclaimed Water Purification in Urban Rivers

Abstract

Reclaimed water produced via the advanced treatment of domestic wastewater has broad application prospects for reuse in urban rivers, while the nutrients in reclaimed water, especially nitrogen and phosphorus, lead to eutrophication or ecological impacts. Submerged plants are preferred as an enhanced technology used at reclaimed water recharge sites for both water quality improvement purposes and ecological conservation functions. In this study, which adopted the typical submerged plant Myriophyllum verticillatum (M. verticillatum) as its experimental object, experiments were carried out in an illumination incubator without a substrate and under hydrostatic conditions to investigate the water purification effects and mechanisms of action of M. verticillatum at different planting densities. The analysis showed that the group with a wet weight of 2.5 g L⁻¹ had the best growth status and the best overall performance with respect to improvements in water quality indicators, including COD, nitrogen, and phosphorus, as well as demonstrating excellent uptake and synergistic effects in the process of removing nitrogen and phosphorus. The contributions of natural effects, the uptake and enrichment by M. verticillatum itself, and the synergistic effects during the nitrogen and phosphorus removal process were quantified. Furthermore, 16S rRNA gene sequencing was used to determine the surface-attached bacterial colonies of M. verticillatum, to analyze their population diversity, and to identify environmental functional genera. In conclusion, an appropriate density of M. verticillatum can improve water quality and provide a suitable environment for the survival and growth of relevant environmentally functional microorganisms, effectively removing nitrogen and phosphorus through its own absorption and synergistic effects.

1. Introduction

As important resources and environmental carriers, urban rivers and lakes perform flood control and water storage functions and contribute to ecological improvement, climate regulation, landscape beautification, and so forth [1,2,3]. Deteriorating aquatic conditions can severely limit the development of surrounding cities [4]. In recent years, China’s requirements for water environment management have become increasingly stringent [5,6]. In order to alleviate water pollution and augment watershed flow, it has become advantageous to reuse reclaimed water generated from the advanced treatment of municipal wastewater to replenish waterbodies [7,8,9,10]. However, considering the effects of wastewater treatment and its economic costs, it is found that reclaimed water reused in urban rivers may affect water quality, making it difficult to meet the corresponding requirements for urban rivers (which are usually class IV standards for surface water in China) [11]. In particular, the concentrations of nutrients, such as nitrogen [12] and phosphorus [13], in reclaimed water are relatively high, which may present a hidden danger for the eutrophication of water bodies. Therefore, appropriate measures must be taken to provide stable water quality in receiving rivers that are replenished with reclaimed water.

Submerged plants play important roles in improving the ecology of many urban, slow-flowing rivers. Since the whole plants are under water, they can effectively absorb nutrients, such as nitrogen and phosphorus [14], and enhance the transparency of water bodies [15,16]. They have well-developed chlorophyll in their epidermal cells and are able to provide an oxygen-rich environment underwater. Submerged plants also provide habitats and attachment sites for benthic animals and a large number of environmentally functional microorganisms [17,18,19]. They have been described as forests in rivers [20]. For water purification and ecological stabilization, the planting of submerged plants is a common method with good sustainability and low levels of risk [21,22]. Among submerged plants, M. verticillatum is widely distributed in China [23]. This species not only reduces nitrogen and phosphorus concentrations in water [24] but also exhibits better cold tolerance and fouling resistance compared to other submerged plants and can adapt to water bodies with a certain level of mobility. These qualities make it a pioneer species in river and lake ecological restoration projects [25].

It has been established that the main factors affecting the growth status and ecological effects of submerged plants are water quality [26,27] and population density [28,29]. Based on the long-term monitoring data of a water reuse pilot project, it is known that reclaimed water recharge sites are the most impacted and have the worst relative water quality in rivers, which are the main battlefields where submerged plants play significant roles. Within a certain range, as the population density increases, submerged plants grow vigorously, and the water purification effect gradually improves [30]. Beyond a certain threshold, continuing to increase the number of plants per unit area does not enhance their environmental effects [31]. In the case of high plant density, even the residues of dead plants may have a negative impact on water quality. Therefore, it is necessary to scientifically study the optimal planting density at recharge sites for the reuse of reclaimed water in urban rivers.

It has been suggested that the purification of water using submerged plant technology can be mainly attributed to the combination of three types of effects, including natural action (primarily volatilization and sedimentation) [32], nutrient uptake by M. verticillatum, and other synergistic effects due to plant growth (primarily surface microbial action, interception, and flocculation) [21]. Which of these roles contributes more has rarely been discussed in previous studies and should be clarified and quantified through research. Further, it is unclear whether the presence of submerged plants at the recharge sites of a reclaimed water replenishment project has a significant effect on bacterial colonies, and analysis of the community diversity of bacterial colonies and the identification of specific functional microorganisms are relatively lacking, aspects that are also focused on in this paper.

Ningbo, located in the southern part of China, is a typical example of a water-scarce coastal city in a subtropical region with an urgent need for reclaimed water reuse. This study was based on an ecological restoration project in urban rivers replenished with reclaimed water in Ningbo, and the representative submerged plant M. verticillatum was selected as the experimental object. The influent of the WWTP was domestic sewage, and the effluent, following a secondary biochemical treatment, was further treated with denitrification filter, high-density sedimentation tank, D-type filter, chlorine dioxide disinfection, and ultraviolet disinfection before reclaimed water was supplied. The water purification experiment was simulated in an illumination incubator at the water recharge site. The role of M. verticillatum in the water purification process of reclaimed water-receiving rivers was investigated by comparing the purification effects of different planting densities and calculating the respective shares of natural effects, absorption by M. verticillatum, and other synergistic effects during the nutrient removal process. The 16S rRNA gene sequencing and colony composition analysis were performed on the microorganisms attached to the plant surface. This study on the water purification effect and mechanism of M. verticillatum may provide a reference for the practice of water quality assurance after the reuse of reclaimed water in urban rivers.

2. Materials and Methods

2.1. Experimental Subjects and Condition Settings

M. verticillatum can carry out vegetative propagation via stolons and fragments; therefore, it was purchased from an aquatic plant planting base (Honghu lake in Hubei province, China)and stem segments with uniform shape, healthy growth conditions, and similar height were selected as the experimental subjects before the experiment. In order to calculate the proportions of nitrogen and phosphorus being removed, the experimental culture conditions were set to be substrate-free to avoid the interference of pollutants in the substrate. The reclaimed water-receiving river in Ningbo is a slow-flowing water body, and the growth of submerged plants will also further slow down the flow rate [33]. Higher flow rates also reduce the photosynthetic rate and inhibit plant growth [34], which is not conducive to the fixation of stem segments after the de-rooting of M. verticillatum. Taking the above factors into account, the experiments were conducted under hydrostatic conditions.

The experimental water was prepared according to the average concentrations of water quality indicators of the water samples collected from September 2017 to August 2019 at the N1 point of the reclaimed water recharge site of River L in Ningbo. The main focus was on COD and nutrient indicators (TP, NH₃-N, and TN). Hoagland 2 mixed salt medium was used as the base and supplemented with sucrose, KNO₃, NH₄H₂PO₄, and (NH₄)₂SO₄ to meet the plant growth requirements while maintaining consistency with the water quality of River L. Among them, sucrose has a role in maintaining the osmotic pressure of the medium and reducing microbial contamination, in addition to being the carbon source of the experimental system. The concentrations of the water samples in River L and the experimental water prepared are shown in

Table 1. Concentrations of water quality indicators at point N1 in River L replenished with reclaimed water and the experimental water prepared.

Water Quality Indicators	COD	TP	TN	NH3-H
Concentrations at point N1 in River L replenished with reclaimed water	15–29 mg L−1	0.11–1.40 mg L−1	4.56–18.8 mg L−1	0.05–3.50 mg L−1
Experimental water	19.38 ± 1.58 mg L−1	0.49 ± 0.17 mg L−1	11.31 ± 3.50mg L−1	0.93 ± 0.51 mg L−1
Artificially configured agents	C12H22O11	NH4H2PO4	KNO3	(NH4)2SO4

. The Hoagland 2 was purchased from Sigma-Aldrich Trading Co.Ltd. (Shanghai, China), and the other chemicals were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China).

2.2. Experimental Methods

The experiments were performed in 3 L glass beakers. Tailored plexiglass devices were used to fix the submerged plants, as shown in Figure 1. According to the results of the pre-experiment, the plant densities of the treatment groups were set to 1.5 g L⁻¹, 2.0 g L⁻¹, 2.5 g L⁻¹, 3.0 g L⁻¹, and 3.5 g L⁻¹ by wet weight (WW), and no submerged plants were placed in the control groups. After adding 3 L of experimental water, all beakers were placed together in the illumination incubator. The culture conditions were set to a constant temperature of 25 ± 1 °C, a light intensity of 4500 lx, and a light/dark ratio of 12 h:12 h. Sampling was performed every 3 days at 10:00 am starting from the first day. The water lost by evaporation in the beaker was replenished with distilled water the day before sampling. After measuring the pH and dissolved oxygen (DO) in each group, 10 mL of the water samples was collected at 3 cm, 9 cm, and 15 cm below the water surface and mixed as the composite samples, and the concentrations of the remaining water quality indicators were analyzed in the same day. Subsequently, the initial experimental water with the volume of 30 mL was replenished to maintain a stable nutrient concentration in the system. The indicators of the stems and leaves of M. verticillatum were analyzed before and after the experiment.

2.3. Analysis Methods

2.3.1. Analysis of Water Quality Indicators

The pH and DO were measured using a Hach HQ30D handheld meter. Turbidity was measured using a Hach 2100Q turbidimeter (Hach Company, Loveland, United States). COD, TP, TN, and NH₃-N were analyzed using a Hach HQ40d53000000 tester and Hach kits with corresponding ranges.

2.3.2. Analysis of Plant Growth and Composition

The plants were cleaned with distilled water before and after the experiment without damaging them, an absorbent paper was use to absorb the surface water, and then the wet weight was quickly weighed using an electronic balance. The average water content of other M. verticillatum with similar state to the experimental M. verticillatum was determined to calculate the dry weight (DW) of M. verticillatum before the experiment. After the experiment, the plants were dried in 500 mL beakers in an oven at 80 °C to a constant weight, and the dry weight values were recorded to calculate the relative growth rate (RGR) (Equation (1)).

$$\mathrm{Relative}\mathrm{growth}\mathrm{ratesRGR},\%=\frac{{\mathrm{lnW}}_1-{\mathrm{lnW}}_0}{19}\times 100\%$$

(1)

where W₀ is the pre-experimental plant’s dry weight, W₁ is the post-experimental plant’s dry weight, and 19 is the experimental period (d).

The malondialdehyde (MDA) content in the plants was determined as follows: The plant leaves to be tested were weighed to be 0.1 g (WW) and ground with 5 mL of 1% trichloroacetic acid (TCA). The homogenate was centrifuged at 10,000 rpm for 15 min, and 0.5 mL of the supernatant was mixed with 2 mL of 20% TCA solution containing 0.5% thiobarbituric acid. The mixture was heated at 96 °C for 30 min, immediately cooled to 0 °C, and centrifuged at 8000 rpm for 10 min. The absorbance at 450, 532, and 600 nm was then measured using a UV-Vis spectrophotometer and calculated according to Equation (2). After the experiment, the growth rate values were also calculated based on Equation (3).

$$\mathrm{MDA}\left({\mathsf{\mu }\mathrm{mol}\mathrm{g}}^{-1}\right)=\frac{\left[6.45\times \left({\mathrm{A}}_{532}-{\mathrm{A}}_{600}\right)-0.56\times {\mathrm{A}}_{450}\right]\times \mathrm{V}}{\mathrm{W}\times 1000}$$

(2)

where V (mL) is the volume of the extraction solution, and W (for 0.1 g of WW) is the wet weight of the extracted leaves.

$$\mathrm{The}\mathrm{growth}\mathrm{rate}\mathrm{of}\mathrm{MDA}\left(\%\right)=\frac{{\mathrm{E}}_1-{\mathrm{E}}_0}{{\mathrm{E}}_0}\times 100\%$$

(3)

where E₀ is the pre-experimental MDA content, and E₁ is the post-experimental content.

The dried plant samples were ground and crushed, sieved at 200 mesh, and then passed through an organic element analyzer to analyze the TN content of the plant stems and leaves. To determine the TP content of the plant stems and leaves, the dried plant samples were first weighed, and 0.2 g was placed in a ceramic crucible and calcined in a muffle furnace at 450 °C for 3 h. The residue was removed into a test tube, added with 20 mL of 3.5 mol L⁻¹ HCl, and shaken for 16 h. The shaking mixture was centrifuged at 5000 rpm for 15 min, and the supernatant was used to determine the TP with the corresponding range of the Hach kit.

2.3.3. Analysis of Microbial Communities on Plant Surfaces

At the end of the experiment, approximately 0.5 g of each plant sample was weighed and placed in a sterile centrifuge tube. Then, 10 mL of 0.1 mol L⁻¹ potassium phosphate buffer with a pH of 8.0 was added to each gram of the sample. The centrifuge tubes were placed in an ultrasonic cleaner, shaken for 1 min, and then vortexed for 10 s. This step was repeated twice. The plant samples were removed and placed again in another sterile centrifuge tube, and all previous steps were repeated after the addition of potassium phosphate buffer.

The two washes were mixed and filtered through a 0.22 μm filter membrane. The filter membranes were placed in sterile centrifuge tubes. The high-throughput gene sequencing of 16S rRNA of plant surface microorganisms was analyzed by Shanghai Meiji Biomedical Technology Co., Ltd. (Shanghai, China).

3. Results and Discussion

3.1. Water Quality Improvement Performance

M. verticillatum plays an important role in dissolved oxygen replenishment in watercourses [22]. Figure 2a shows that, compared to the control groups, dissolved oxygen content in the water of the treatment groups exhibits an increasing trend. Meanwhile, the dissolved oxygen concentration of the two treatment groups with a lower density of M. verticillatum had a brief decrease at the initial stage of the experiment, probably due to the fact that M. verticillatum was still in the process of environmental adaptation and respiration was stronger than photosynthesis, thus consuming oxygen in the water.

In natural water bodies, submerged plants can reduce disturbances caused by wind, currents, biological activities, etc., and reduce the resuspension of sediments (especially nutrients in the substrate), thus achieving the goal of maintaining water clarity [35]. The dense fronds of M. verticillatum increase the contact area with water, which can intercept insoluble solid particles and secrete coagulation aids [36] to accelerate the settling of suspended materials [37]. Figure 2b shows that the turbidity of the experimental system is effectively controlled by M. verticillatum, as the highest turbidity value in the treatment groups is less than half of that in the control groups.

As presented in Figure 2c, the COD concentrations in the water decrease significantly in all groups at the beginning of the experiment, reaching the lowest values (within the range of 4.7~17.8 mg L⁻¹) around the fourth day. The COD concentration in the treatment groups reduced more rapidly than that in the control groups. Then, the appearance of a light white film on the beaker surface and subsequent rebound of COD concentration were observed. This may be because M. verticillatum produces and releases organic substances, such as organic acids, phytonutrients, alkaloids, and phenolic and terpenoid compounds, usually within 5% to 25% of photosynthetically fixed carbon [38].

As important components of essential compounds for life such as nucleic acids, proteins, and ATP, nitrogen and phosphorus are involved in a variety of metabolic processes in submerged plants. However, too high concentrations of them in a water environment may lead to the occurrence of eutrophication. Whether or not M. verticillatum could effectively control the concentrations of nitrogen and phosphorus at the reclaimed water recharge site became the focus of the experimental process. The main form of elemental phosphorus present in the experimental water is reactive phosphate. As shown in Figure 2d, the TP concentration in the water of all treatment groups decreases rapidly after the start of the experiment and begins to stabilize at the lowest value around the sixth day, with removal rates ranging from 65.3% to 91.8% and a slightly rebounding trend at the end. The TP concentration in the control groups decreases slowly with a steady trend, and the removal rate is only 30.8%~37.4%. The analysis concluded that the initial phase of the experiment was characterized by rapid growth of M. verticillatum, which removed most of the phosphorus content from the water through direct absorption, interception, and secretion of chemicals to enhance coagulation and other processes.

As shown in Figure 2e,f, the indicators related to nitrogen focus on TN and NH₃-N. The TN concentration in the control groups did not change significantly compared to NH₃-N, and decreased smoothly in all treatment groups, with total removal rates ranging from 46.3% to 81.4%. The treatment groups with a density of 2.5 g WW L⁻¹ of M. verticillatum were still the most effective. The NH₃-N content was almost completely removed by about the fourth day of the experiment. This may be due to the higher ability of M. verticillatum to absorb NH₃-N than nitrate nitrogen, in spite of nitrate nitrogen accounting for a larger proportion of the concentration [39]. The reduction in TN concentration in water also includes the processes of ammonia volatilization, nitrification, denitrification, etc. [40]. NH₃-N removal in the control groups may be due to volatilization. Plants release oxygen through photosynthesis, creating an aerobic zone on the surface. The decomposition of enriched organic matter consumes oxygen, which, in turn, leads to alternating anaerobic and anoxic habitats in the attachment layer [41,42]. This is more conducive to microbial-led nitrification and denitrification reactions under different conditions, reducing the concentrations of NH₃-N and TN in water. The secretions of nitrogen and lignin from the roots and stems of submerged plants are lower than those of jerking plants, which can provide denitrifying bacteria with substances that can be metabolized more easily and facilitate denitrification in the process of nitrogen removal from eutrophic water bodies [43].

In summary, the planting of M. verticillatum in a reclaimed water-receiving river had a positive effect on water quality improvement. The best improvement could be seen in the treatment groups with a density of 2.5 g WW L⁻¹.

3.2. Stem and Leaf Indicators of M. verticillatum with Different Roles in Nutrient Removal

Significant increases in biomass were observed in all groups of M. verticillatum during the experiment. The highest relative growth rate of 0.68% was achieved in the treatment groups with a plant density of 2.5 g WW L⁻¹, as shown in Figure 3a. With the extension of time, the submerged plants reached a large size and started to grow slower due to the limitation of space and nutrients. Some treatment groups reached saturation and apoptosis occurred.

As an indicator of stress resistance in submerged plants subjected to environmental factors, malondialdehyde (MDA) indicates the degree of lipid peroxidation and membrane damage in the plant body, as shown in Figure 3b. During the experiment, the malondialdehyde content in the stems and leaves of the treated groups of M. verticillatum were monitored, and the results showed that the values increased after the experiment compared to those before the experiment. All growth rate related to the MDA content was greater than 17%, and a resistance response was obvious [44,45]. It suggests that the aquatic environment after being replenished with reclaimed water may have an impact on the survival and growth of submerged plants.

Table 2. N and P contents in plant stems and leaves before and after the experiment.

	Initial	Terminal
		1.5	2	2.5	3	3.5
TN (mg g−1 DW)	33.27 ± 0.04	39.75 ± 0.41	38.02 ± 1.58	40.31 ± 0.33	39.56 ± 1.50	40.45 ± 0.29
TN content growth rate	—	19.47%	14.27%	21.16%	18.89%	21.57%
TP (mg g−1 DW)	0.66 ± 0.03	0.83 ± 0.04	0.86 ± 0.04	0.80 ± 0.01	0.81 ± 0.03	0.86 ± 0.02
TP content growth rate	—	26.52%	30.30%	21.21%	22.73%	30.30%
N/P	0.0198	0.0210	0.0226	0.0198	0.0205	0.0213

shows that the growth of M. verticillatum absorbs nutrients from water, and the average contents of nitrogen and phosphorus elements in the mixed samples of M. verticillatum stems and leaves increase significantly after the experiment compared to the values measured before the experiment. The increase in phosphorus content ranges from 21.21% to 30.30%, and the increase in nitrogen content reaches 14.27% to 21.57%. Overall, the ratio of nitrogen to phosphorus content of plants is relatively stable in accordance with the plant growth status, ranging from 0.0198 to 0.0226.

In order to clearly analyze the respective share of natural effects, uptake by M. verticillatum and other synergistic effects due to M. verticillatum growth during nutrient removal, the experimental data were processed using Equations (4)–(7):

$${\mathsf{\eta }}_1=\frac{{\mathrm{C}}_1{\mathrm{Q}}_1-{\mathrm{C}}_2{\mathrm{Q}}_2}{{\mathrm{C}}_1{\mathrm{Q}}_1}\times 100\%$$

(4)

$${\mathsf{\eta }}_2=\frac{{\mathrm{m}}_2{\mathrm{x}}_2-{\mathrm{m}}_1{\mathrm{x}}_1}{{\mathrm{C}}_1{\mathrm{Q}}_1}\times 100\%$$

(5)

$${\mathsf{\eta }}_3={\mathsf{\eta }}_{\mathrm{control}}$$

(6)

$${\mathsf{\eta }}_4={\mathsf{\eta }}_1-{\mathsf{\eta }}_2-{\mathsf{\eta }}_3$$

(7)

where η₁ represents the complete removal ratio; η₂ represents the ratio of plant uptake; η₃ represents the ratio of volatilization and sedimentation; η₄ represents the ratio of plant synergistic effects, including microbial uptake and degradation, as well as interception on the surface of M. verticillatum; η_control represents the nutrient removal ratio of the control groups; C₁ and C₂ represent the water quality concentrations of nitrogen and phosphorus elements before and after the experiment (mg L⁻¹), respectively; Q₁ and Q₂ represent the total volume of artificially dispensed water added to the incubation system (L) and the volume of water in the system at the end of the experiment (L), respectively; m₁ and m₂ represent the dry weights of M. verticillatum before and after the experiment (g), respectively; and x₁ and x₂ represent the average nitrogen and phosphorus contents per gram of mixed samples of M. verticillatum stems and leaves before and after the experiment (mg g⁻¹), respectively.

The complete removal of phosphorus during the experiment was better than that of nitrogen. As shown in Figure 4, plant uptake and utilization do not necessarily play a dominant role in the removal of nutrients from water bodies [46]. Especially for phosphorus removal, natural sedimentation and plant synergistic effects are more significant and indispensable. The contribution ratio of the direct uptake of nitrogen and phosphorus, respectively, ranged from 12.6% to 21.1% and from 2.3% to 4.7%. In parallel, the contribution ratio of the submerged plants’ synergistic effect to nitrogen and phosphorus removal, respectively, ranged from 16.4 to 43.0% and from 25.5 to 57.9%. This is most likely because the complex environment on the surface of submerged plants provides favorable conditions for the growth of different species of microorganisms (mainly bacteria), which effectively promotes the degradation of organic matter, the removal of nitrogenous compounds, and the conversion of phosphorus-containing compounds in water bodies.

3.3. Analysis of Colonies Attached to the Surface of Myriophyllum verticillatum

The biofilm on the surface of submerged plants is a complex microporous bioreactor composed of multiple microbial and abiotic components [47], which can provide a suitable attachment surface for the growth of colonies as well as a stable source of organic carbon [48]. High-throughput sequencing of bacterial 16S rRNA genes was performed to investigate more deeply the role played by microorganisms in potentiation. The sequenced colonies were obtained from the surface of M. verticillatum in the experiment, including the water samples from the control groups and the water samples from the experimental groups with the best growth status of M. verticillatum at a density of 2.5 g WW L⁻¹. The analysis was performed using the DADA2 plug-in in the QIIME2 process for noise reduction of sequencing results. The steps included filtering noise and correcting sequence errors, removing chimeras and single sequences, and sequence de-duplication to obtain high-resolution ASVs for subsequent analysis. A total of 5,321 ASVs (features) were generated after the DADA2 noise reduction, with an average of 31,088 sequences per sample.

Alpha diversity is mainly used to study community diversity in samples and contains a variety of characterization indices. The typical Sobs index and Shannon index were selected as representatives. Among them, the Sobs index can reflect community richness and the Shannon index can be used to estimate community diversity. The greater the value of these two indices, the stronger the characterization of the community. The results (Figure 5a) show that the samples with the plant density of 2.5 g WW L⁻¹ and 3.0 g WW L⁻¹ were more abundantly colonized, which was consistent with their relatively better water purification effect. The treatment groups of 3.5 g WW L⁻¹ already showed colony decay with plant die-off at the end of the experiment. The diversity indexes of all samples in the treatment groups were significantly higher than those in the control groups, indicating that submerged plant growth improves the microbial community composition in water.

Principal coordinates analysis (P_COA) is used to study the similarity or difference of sample community composition and to find out the potential principal components affecting the difference in sample community composition via dimensionality reduction. The results of dispersion or aggregation between different samples (Figure 5b) show that the two treatment groups with the lowest densities of M. verticillatum (1.5 g WW L⁻¹ and 2.0 g WW L⁻¹) have close structure of attached colonies, and the three treatment groups with higher densities (2.5 g WW L⁻¹, 3.0 g WW L⁻¹, and 3.5 g WW L⁻¹) are similar. All treatment groups are significantly different from the control groups. The water sample results of the experimental groups with 2.5 g WW L⁻¹ are distributed between the control groups and the high-density plant groups. Thus, with similar decontamination efficacy, the density of 2.5 g WW L⁻¹ M. verticillatum is more economical.

FAPROTAX is a literature-based manual database that allows metabolic and ecological functional analysis of bacterial genera or species. It covers a relatively complete and continuously updated set of functions in biogeochemistry related to aqueous environments, especially the cycling of nitrogen, hydrogen, sulfur, and carbon, as well as pathogenicity analysis. Heat maps were plotted after z-score normalization of the FAPROTAX database analysis results. As shown in Figure 5c, the levels of chemoenergetic heterotrophic bacteria and the risk of parasites are higher in the control groups than in the treatment groups. Nitrogen-related metabolic functions, such as ureolysis, nitrate reduction, and nitrogen fixation, are significantly better in the treatment groups than in the control groups, which are consistent with the previously good nitrogen removal effect seen in the treatment groups.

Community structural composition was analyzed at the genus level, and species with an abundance less than 0.01 in all samples were classified as others. According to the taxonomic results, 6 phyla, 8 classes, 26 orders, 39 families, and 52 genera were found in all samples. The control groups in the phylum level revealed poor diversity, containing only Proteobacteria, Bacteroidota, and Firmicutes. The treatment group samples (including the water samples of the group of 2.5 g WW L⁻¹) additionally included Gemmatimonadota, Cyanobacteria, and Acidobacteriota. Proteobacteria were the overwhelmingly dominant species in all samples, with an abundance value range of 49.3% to 88.1%. The community histogram shows the relative abundance of various microorganisms in the samples, which can be more visually seen to be in high agreement with the results of the P_COA analysis (Figure 6). The purification functions of water corresponding to different bacterial genera were also analyzed. As shown in

Table 3. Environmental functional genus characteristics of surface-attached colonies of M. verticillatum.

Family	Genus	Environmental Functions
Hyphomonadaceae	UKL13-1	Degradation of organic matter and hydrocarbons [49] Denitrification [50] Secretion of extracellular proteins and biofilm formation [51]
Caulobacteraceae	Caulobacter	Secretion of extracellular polymers [50] Adapts to and may be able to degrade triclosan (TCS) [52]
Rhizobiaceae	g__unclassified_f__Rhizobiaceae	Denitrification [53]
Devosiaceae	Devosia	Secretion of extracellular polymers [54] Nitrification [55,56]
Beijerinckiaceae	Bosea	Removal of nitrite [57]
Rhodobacteraceae	g__unclassified_f__Rhodobacteraceae	Promotes initial biofilm formation [58]
Comamonadaceae	g__unclassified_f__Comamonadaceae	Denitrification [59]
Comamonadaceae	Ramlibacter	Anaerobic or partly anaerobic bacteria grown via the reduction of oxygen anions, such as sulfate, perchlorate, nitrate, and nitrite [60]
Methylophilaceae	Methylotenera	Methanol oxidation and denitrification [61] Nitrogen fixation and degradation of simple alkanes [62]
Methylophilaceae	Methylophilus	Promotes aerobic denitrification and PAH biodegradation [63]
Methylophilaceae	g__unclassified_f__Methylophilaceae	Methanol oxidation and denitrification [64]
Alcaligenaceae	Derxia	Nitrate reduction [65]
		Nitrogen fixation [66]
Nitrosomonadaceae	Ellin6067	Ammonia oxidation [67]
Rhodocyclaceae	Zoogloea	TN removal [68]
		COD, nitrogen, and phosphorus removal [69]
Enterobacteriaceae	Kosakonia	Ammonium nitrogen fixation [70]
Gemmatimonadaceae	Gemmatimonas	Nitrogen removal [71]
Microscillaceae	g__norank_f__Microscillaceae	Nitrification [72]
Saprospiraceae	g__norank_f__Saprospiraceae	Nutrient removal [73]

, many of the bacteria attached to the surface of M. verticillatum in the treatment groups have water purification effects and are mainly involved in reactions related to nitrogen and phosphorus removal, degradation of simple or trace amounts of complex organic matter, and secretion of extracellular polymers to form biofilms.

4. Conclusions

In this study, the water purification effect and mechanism of M. verticillatum at the recharge site of an urban river replenished with reclaimed water were investigated. In the range of plant density gradient set up in the experiment, the groups with the plant density of 2.5 g WW L⁻¹ had the best growth status and the best overall performance in the removal of basic physical indicators including COD, nitrogen, and phosphorus. Meanwhile, these groups had greater plant uptake and synergistic effect in the process of nitrogen and phosphorus removal, which provided a reference for determining the density value in engineering practice. The analysis of microbial colonies attached to plant surfaces showed that the submerged plants provided a suitable environment for the survival and growth of environmental functional microorganisms. There were a rich diversity of colony populations and correlations between the main components and plant density. Microorganisms of several genera played an important role in the removal of pollutants (especially nitrogen). The experimental results provide a scientific basis and reference path for the selection of ecological measures for water purification in rivers replenished with reclaimed water, and submerged plants represented by M. verticillatum can achieve this purpose well.

Furthermore, the major purpose of reclaimed water reuse in landscape rivers is to promote water purification and ecological conservation. Most of the previous receiving watercourses have been polluted and may not be able to achieve submerged plant regeneration on their own. If artificially grown plants have short growth cycles and are susceptible to decay, their management requires a high level of personnel and materials; otherwise, they will likely cause more serious contamination problems.