Potential Phytoremediation of Aquatic Macrophyte Species for Heavy Metals in Urban Environments in the Southern Area of Brazil

Abstract

This research investigated four different species of aquatic macrophytes with natural occurrence in an urban environment highly anthropized in Southern Brazil. The aim of the research was to compare the phytoremediation potential among the species E. anagallis, H. grumosa, H. ranunculoides, and S. montevidensis through Pearson´s correlation analysis and cluster analysis, using the heavy metal content identified through HNO₃ - HClO₄ and phytoremediation indexes. The results highlighted the bioconcentration factor (BCF) of H. ranunculoides, with outstanding results for Cu BCF = 667.09, Zn BCF = 149.93, Cd BCF = 26.85, Cr BCF = 31.77, Ni BCF = 35.47, and Pb BCF= 126.29. Additionally, H. grumosa and S. montevidensis were also highlighted, considering the potential phytoremoval (g ha⁻¹). Therefore, this study demonstrates the tolerance and potential for removal of heavy metals Cu, Cr, Cd, Pb, Ni, and Zn by the evaluated aquatic macrophyte species and elucidates the outstanding potential of application in phytoremediation purposes.

1. Introduction

The urbanization and industrialization processes within cities contribute to the release of contaminants into aquatic and terrestrial ecosystems. The contaminants reaching the aquatic environments are responsible for an increase in both sediments and water in these areas. In urban locations, the pollutants reach aquatic environments through rainfall runoffs that carry different loadings depending on the area and land use. Residential areas can contribute to nitrogen and phosphorous loadings, as well as domestic sewage, along with industrial wastewater discharges with high levels of heavy metals [1]. Additionally, means of transport are a source of heavy metals, hydrocarbons, sulfur, oil, and grease [2].

Among the contaminants, the presence of heavy metals in watercourses represents a threat considering the bioaccumulation properties in the food chain and risks to environment, animals, fish, plants, and population health. General consequences of heavy metal in excess are damage to the kidneys and bones, endocrine, cardiovascular, and neurological problems, in addition to potentiating the development of cancer [3].

The entrance of heavy metals into organisms can occur due to invertebrates which absorb levels of these elements and are a food source for fish and other aquatic species. Another route is immediate absorption via water, or even via sediment deposition—which is the first location for heavy-metal accumulation in aquatic environments [4].

In Brazil, the origin of most heavy-metal content in excess is due to mining activities, along with untreated effluent discharges from agricultural, industrial, and urban areas [5]. The Santa Bárbara Stream is an important watercourse for the municipality of Pelotas/RS, south Brazil, being responsible for the water supply of most parts of this city. The presence of aquatic macrophytes is widely noticed in the area, along with a smell and foam also present, indicating a possible eutrophication condition and highlighting the polluted condition of the area [6].

Several conventional methods can be applied to remove heavy metals from these environments, such as chemical separation, dredging, and electro-oxidation [7]. However, there are some cost limitations and environmental issues that limit the wide application of these techniques. Phytoremediation is an alternative and innovative method, since the specificity of species and genotypes allows the removal of different pollutants [7,8]. However, the efficient application of phytoremediation techniques presents a straight relationship with the understanding of mechanisms in which the plants tolerate and bioaccumulate heavy metals, with attention to the limits of the toxic level. It is also important the understand how the elements are translocated in the plants [9].

Different mechanisms enable plants to accumulate heavy metals, nutrients, and other contaminants. For heavy-metal uptake, the plant performs intracellular accumulation, alteration of gene expression responsible for uptake of these elements, efflux, sequestration, and translocation [10,11]. Among phytoremediation techniques, there is: phytoextraction, the accumulation by the shoots of the plants; phytostabilization, with the immobilization of contaminants; phytodegradation, degradation through the metabolic process; rhizodegradation, degradation in the rhizosphere; phytovolatilization, elimination of contaminants through a transpiration process; and rhizofiltration, filtration in the rhizosphere area without translocation [12].

For the efficient application of phytoremediation, the first step is the screening and identification of plant species with the ability to tolerate the high levels of the selected pollutant. For this prospection of plants, the strategy of studying plants from contaminated areas is promising [13]. In addition, it is necessary to understand the main mechanism by which the plant removes the contaminant, its transport within the plant, and its tolerance, among other aspects, in order to exploit its full potential [14].

The identification of tolerant native species is essential for the proper application of the technique, and there are essential factors such as adaptation to local condition and competitions with invasive/exotic species [15]. Worldwide, researchers have been investigating different aquatic macrophyte species’ potential for application in phytoremediation purposes, e.g., Eichhornia, Lemna, and Azolla, and in general, the requirements for the application are rapid growth and higher biomass production, along with tolerating higher levels of the selected contaminant [16].

Therefore, in this study, we propose the utilization of aquatic macrophyte species’ characteristics of South America, i.e., Enydra anagallis, Hydrocotyle ranunculoides, Hymenachne grumosa, and Sagittaria montevidensis, for phytoremediation, as well as comparison of the potential of accumulating heavy metals of these plants. We then highlight their great potential for application in phytoremediation purposes, thus representing a sustainable alternative for decontamination of water and wastewater.

2. Materials and Methods

2.1. Study Area

The study area is located in the municipality of Pelotas/RS (31°45′43″ S and 52°21′00″ W), Brazil. The Santa Bárbara Stream is the principal watercourse from Santa Bárbara Subwatershed, with 30,000 m of total length, and presents importance for the city since it serves as a water source for a great part of the population [17]. The Stream has suffered changes in its drainage bed, along with the construction of a dam for public supply and the implementation of a flood protection system [18].

The current condition of degradation in the Santa Bárbara Stream begins with the expansion of urban areas and changes in the natural characteristics of the environment. Some aspects can be highlighted as the land use and occupation occurring in a disorderly manner, the absence of a sanitary sewer system, resulting in the domestic and industrial release of contaminants, along with lack of environmental education [19].

2.2. Sampling Plants and Characterization

The collection of the emergent aquatic macrophytes (Figure 1) was performed in Santa Bárbara Stream, and the location (31^o45’24.4” S, 52^o21’22.1” W) was selected according to the presence of a large number of plants in the area, along with easy access to the riverbanks. The sampling occurred in the summer season, when presence of aquatic macrophytes is usually more visible, presenting the highest amount and diversity. The plants were sampled through the area and the biomass was placed in plastic bags and properly stored until in the laboratory. The plants (

Table 1. Studied aquatic macrophyte species with family and common name, dry mass, and biomass production (ton ha⁻¹).

Species	Family	Common Name	Dry Mass (g)	Biomass Production (ton ha−1)
Enydra anagallis	Asteraceae	–	2.58	0.56
Hydrocotyle ranunculoides	Araliaceae	Floating pennywort	0.23	0.16
Hymenachne grumosa	Poaceae	Carnivão	11.69	2.92
Sagittaria montevidensis	Alismataceae	Giant arrowhead	20.04	3.20

) were then washed and rinsed using distilled water to remove soil or sediment particles attached to plant biomass [20].

2.3. Detection of Nutrients and Heavy-Metal Content

The roots were separated from the shoots of the plants and the different portions were dried until achieving constant mass (approximately 48 h at 60 °C), and subsequently ground. The ground material was used for 3:1 nitric-perchloric acid digestion (HNO₃-HClO₄) following the methodology described by Tedesco [20], with approximately 15 hours in HNO₃ and 5 hours in HClO₄. We used a Micro Tube Kjeldahl digester block (LUCADEMA, Luca-23/02, Brazil) in a gas exhaust housing (Ideoxima ORG.10, Brazil), and the element levels were detected with Inductively Coupled Plasma-Optical Emission Spectrometry (PerkinElmer®-Optima™ 8300 ICP-OES), at the Laboratory of Soil at the Federal University of Rio Grande do Sul (UFRGS). We used internal standards [21] for quality control with limits of detection (LoD) of Cu—0.6; Zn—2.0; Cd—0.2; Cr—0.4; Ni—0.4; Pb—2.0 mg kg⁻¹.

2.4. Indexes for Phytoremediation

The bioconcentration factor (BCF) and the translocation factor (TF) phytoremediation indexes were determined by applying Equations (1) and (2), respectively [22]:

$$\mathrm{BCF}=\frac{\mathrm{level}\mathrm{in}\mathrm{roots}}{\mathrm{level}\mathrm{in}\mathrm{environment}}$$

(1)

$$\mathrm{TF}=\frac{\mathrm{level}\mathrm{in}\mathrm{shoots}}{\mathrm{level}\mathrm{in}\mathrm{roots}}$$

(2)

where levels in roots refer to the concentration determined in the roots of the plants, in mg kg⁻¹, the level in the environment refers to the concentration of the element in water (mg L⁻¹), and the level in shoots refers to the concentration of the element in the shoots of the plants (mg kg⁻¹)—aerial biomass.

In addition, we calculated the phytoremediation potential, in g ha⁻¹. For this, firstly, we estimated the biomass production (

Table 2. Reference values for general plants (in mg kg⁻¹ dry mass, according with Kabata-Pendias and Pendias [23] and Hopkins [24] and total levels found in the aquatic macrophyte species, in mg kg ⁻¹.

	Sufficient Level	Toxic Level	H. grumosa	H. ranunculoides	S. montevidensis	E. anagallis
	----------------------------------mg kg−1------------------------------
Cu	5–30 [23]	20–100 [23]	42.29	236.83	145.39	125.61
Zn	27–150 [23]	100–400 [23]	298.13	373.74	235.16	205.55
Cd	0.05–0.2 [23]	5–30 [23]	0.64	0.75	0.55	1.28
Cr	0.1–0.5 [23]	5–30 [23]	21.28	179.34	22.55	59.77
Ni	0.1–5 [23]	10–100 [23]	13.37	83.44	13.17	34.66
Pb	5–10 [23]	30–300 [23]	22.76	33.37	13.75	25.76
	-------- mmol kg−1-------		----------------------------------g kg−1-----------------------------
P	60 [24]	-	7.23	17.05	11.63	10.64
K	250 [24]	-	15.32	35.34	28.65	29.15
Ca	125 [24]	-	13.48	25.9	20.6	20.85
Mg	80 [24]	-	4.47	8.9	8.35	6.5
S	30 [24]	-	6.81	10.17	6.31	7.1

) for each species, obtained from the area occupied by one plant and its total dry weight. Secondly, the phytoremoval potential was calculated by multiplying the biomass production by the total element levels (total mg kg⁻¹).

2.5. Statistical Analysis

The experimental data were analyzed using the Statistica® software program, v. 7.0. A Pearson’s analysis of correlation was run, along with cluster analysis (CA) using Ward’s linkage method based on the Euclidean distance, which produced a dendrogram with well-defined clusters.

3. Results

3.1. Total Content of Heavy Metals

Some guidelines values established by Kabata-Pendias and Pendias [23] described the usual/sufficient values and toxic/exceeding levels for general plants and were used as a comparative (Table 2). H. grumosa exceeded the sufficient levels for Cu (42.29 mg kg⁻¹), Zn (298.13 mg kg⁻¹), Cd (0.64 mg kg⁻¹), Cr (21.28 mg kg⁻¹), Ni (13.37 mg kg⁻¹), and Pb (22.76 mg kg⁻¹), presenting then total values that were between the toxic ranges, excepting Cd and Pb. Adapted from Demarco et al. [25]; Demarco et al. [26]; Demarco et al. [6].

The H. ranunculoides species was detected with copper levels up to two-fold higher than the highest toxic limit of 100 mg kg⁻¹. Another outstanding ability detected for H. ranunculoides was the chromium total levels reaching almost six-fold higher than the highest toxic limit of 30 mg kg⁻¹.

The species S. montevidensis was highlighted for copper total levels (145.39 mg kg⁻¹), exceeding the toxic range for general plant species. This aquatic macrophyte presented Zn, Cr, and Ni levels standing among the toxic/exceeding levels for general plants. These results for S. montevidensis can be pointed out as demonstrating the plant with great capacity of tolerance for these toxic metals. S. montevidensis Cd and Pb levels were above sufficient. It might be mentioned that both heavy metals Cd and Pb are known for their lack in terms of biological function in plants and their toxic effects as oxidative stress [27]. Finally, E. anagallis presented Cu and Cr exceeding the toxic levels, with values of 125.61 mg kg⁻¹ and 59.77 mg kg⁻¹. Zn and Ni stayed among the toxic range.

3.2. Correlation Analysis

The correlation in H. grumosa regarding total levels of macronutrients and heavy metals is shown in

Table 3. Correlation analysis in H. grumosa regarding total levels of macronutrients and heavy metals.

	P	K	Ca	Mg	S	Cu	Zn	Cd	Cr	Ni	Pb
P	1.00
K	−0.04	1.00
Ca	0.21	−0.12	1.00
Mg	0.80	0.51	0.17	1.00
S	0.43	0.85	0.13	0.77	1.00
Cu	−0.12	0.97	−0.27	0.37	0.80	1.00
Zn	0.44	0.83	0.21	0.88	0.90	0.70	1.00
Cd	0.12	0.70	−0.73	0.42	0.56	0.77	0.48	1.00
Cr	0.67	0.04	−0.19	0.71	0.16	−0.08	0.42	0.34	1.00
Ni	0.15	0.96	-0.20	0.67	0.86	0.92	0.89	0.78	0.29	1.00
Pb	0.73	−0.08	−0.06	0.70	0.10	−0.21	0.37	0.19	0.99	0.17	1.00

, which the strong positive correlation between K and Cu (r = 0.97), K and Ni (r = 0.96), S and Zn (r = 0.90), Cu and Ni (r = 0.92), and Cr and Pb (r = 0.99).

In the case of H. ranunculoides (

Table 4. Correlation analysis in H. ranunculoides regarding total levels of macronutrients and heavy metals.

	P	K	Ca	Mg	S	Cu	Zn	Cd	Cr	Ni	Pb
P	1.00
K	0.12	1.00
Ca	−0.17	0.06	1.00
Mg	−0.62	−0.07	0.87	1.00
S	0.71	0.62	−0.17	−0.47	1.00
Cu	0.42	−0.42	0.19	−0.11	−0.19	1.00
Zn	0.39	−0.36	0.30	0.00	−0.18	0.99	1.00
Cd	−0.84	0.10	0.64	0.89	−0.54	−0.24	−0.14	1.00
Cr	−0.68	0.34	0.39	0.61	−0.15	−0.32	−0.22	0.82	1.00
Ni	−0.61	0.33	0.40	0.57	−0.14	−0.19	−0.09	0.78	0.99	1.00
Pb	−0.08	0.38	0.81	0.66	0.15	0.16	0.28	0.58	0.61	0.66	1.00

), the highlighted positive correlation was among Cu and Zn (r = 0.99) and Cr and Ni (r = 0.99).

The S. montevidensis correlation matrix (

Table 5. Correlation analysis in S. montevidensis regarding total levels of macronutrients and heavy metals.

	P	K	Ca	Mg	S	Cu	Zn	Cd	Cr	Ni	Pb
P	1.00
K	−0.98	1.00
Ca	0.91	−0.96	1.00
Mg	−0.65	0.78	−0.88	1.00
S	0.94	−0.98	0.99	−0.87	1.00
Cu	−1.00	0.99	−0.92	0.68	−0.95	1.00
Zn	−0.99	0.96	−0.85	0.59	−0.89	0.99	1.00
Cd	−0.99	0.96	−0.87	0.57	−0.90	0.99	0.99	1.00
Cr	0.93	−0.98	0.95	−0.86	0.99	−0.95	−0.91	−0.89	1.00
Ni	−0.29	0.16	−0.07	−0.39	−0.03	0.26	0.28	0.40	0.06	1.00
Pb	−0.30	0.37	−0.55	0.43	−0.43	0.32	0.17	0.32	−0.27	0.37	1.00

) showed a strong positive correlation of P and Ca (r = 0.91), P and S (r = 0.94), P and Cr (r = 0.93), K and Cu (r = 0.99), K and Zn (r = 0.96), K and Cd (r = 0.96), Ca and S (r = 0.99), Ca and Cr (r = 0.95), S and Cr (r = 0.99), Cu and Zn (r = 0.99), Cu and Cd (r = 0.99), and Zn and Cd (r = 0.99). Thus, it might be mentioned that, unlike the other species, S. montevidensis demonstrated that heavy-metal uptake strongly correlated with nutrient uptake.

It can be seen that E. anagallis demonstrated a strong positive correlation among P and Ca (r = 0.92), P and S (r = 0.96), P and Cd (r = 0.93), Ca and S (r = 0.90), S and Cd (r = 0.96), Cu and Zn (r = 0.93), Zn and Pb (r = 0.96), and Cr and Ni (r = 0.93) (

Table 6. Correlation analysis in E. anagallis regarding total levels of macronutrients and heavy metals.

	P	K	Ca	Mg	S	Cu	Zn	Cd	Cr	Ni	Pb
P	1.00
K	−0.40	1.00
Ca	0.92	−0.60	1.00
Mg	−0.63	−0.34	−0.41	1.00
S	0.96	−0.23	0.90	−0.75	1.00
Cu	0.37	−0.25	0.52	0.19	0.37	1.00
Zn	0.49	−0.60	0.69	0.26	0.42	0.93	1.00
Cd	0.93	−0.17	0.87	−0.64	0.96	0.58	0.57	1.00
Cr	−0.10	0.81	−0.15	−0.46	0.15	0.13	−0.19	0.22	1.00
Ni	−0.47	0.87	−0.48	−0.17	−0.23	−0.01	−0.34	−0.15	0.93	1.00
Pb	0.57	−0.78	0.79	0.22	0.48	0.79	0.96	0.56	−0.36	−0.52	1.00

).

3.3. Phytoremediation Evaluation by Cluster Analysis

The phytoremediation indexes calculated allowed comparison between the plants for removing the toxic metals from Santa Bárbara Stream. Generally, the elements were grouped by BCF in two main groups (Figure 2a), one composed of Ca, Pb, K, and Cd, and the other presenting Ni, Cr, Zn, S, Mg, P, and Cu. H. ranunculoides has been classified separately from other species regarding BCF (Figure 2b). This species presented outstanding results for the studied heavy metals: Cu BCF = 667.09, Zn BCF = 149.93, Cd BCF = 26.85, Cr BCF = 31.77, Ni BCF = 35.47, and Pb BCF = 126.29 [25].

The cluster analysis regarding TF by element demonstrated two main groups, the first composed of Mg, Cd, Ni, S, P, K, and Zn, and the other by Ca, Cr, Pb, and Cu (Figure 3a). Plants with TF > 1.0 present the ability to translocate the element to the shoot’s tissues [22]. Therefore, when BCF > 1.0 and TF > 1.0, the mechanism performed is then known as phytoextraction. S. montevidensis and H. ranunculoides were grouped in Figure 3b regarding phytoremediation index TF. This fact may be justified by the characteristic of S. montevidensis and H. ranunculoides being identified as presenting the ability of maintaining high levels of elements mainly in their roots, since TF < 1.0. Therefore, phytoremediation mechanism performed by S. montevidensis and H. ranunculoides for Zn, Cu, Cr, Cd, Ni, and Pb is rhizofiltration.

The evaluation of phytoremoval potential was performed, aiming at the identification of the most promising plant species for application in phytoremediation technique, considering biomass production for removal estimation. The cluster analysis regarding phytoremoval potential by element demonstrated the formation of two groups, the first with the elements Cd, Pb, Zn, Cr, Ni, and S, and the second with Cu, Mg, K, Ca, and P (Figure 4a).

Both H. grumosa and S. montevidensis were grouped (Figure 3b), considering the outstanding potential found. It should be emphasized that H. grumosa presented great results for all the studied heavy metals, reaching values of 871.29 g ha⁻¹ for Zn. This was quite similar to the value estimated for Zn removal by S. montevidensis, which was 754.01 g ha⁻¹. In addition, S. montevidensis copper removal was detected at 466.20 g ha⁻¹ and the results thus demonstrated the high tolerance of S. montevidensis to zinc and copper levels. Among the species, S. montevidensis was also detected with the highest values of Cr and Ni, being 72.31 g ha⁻¹ and 42.25 g ha⁻¹, respectively.

H. grumosa was also highlighted by Pb phytoremoval potential (66.52 g ha⁻¹) and Cd (1.89 g ha⁻¹). The Cd values, besides being less than the found for other heavy metals, represent potential for phytoremediation of cadmium, considering its toxicity and consequent damage to human and animal health [28,29].

4. Discussion

The Cr values detected for H. ranunculoides were impressively higher than other studies investigating the potential of different plants bioaccumulating this heavy metal from natural environments. The study of Tiwari et al. [30] investigating the biofiltration of heavy metals by Eichhornia crassipes in a contaminated watercourse in Bhopal, India, detected maximum chromium levels of 10.1 mg kg⁻¹.

Regarding copper content, Melo et al. [31] studied the phytoremediation potential of spontaneous species in vineyard soils contaminated with copper and detected values in plants that were similar to those found in this study in H. ranunculoides. It was verified that, despite the study being performed in soil, more specifically in Inceptisol, the total levels of the highlighted species were with similar magnitude, being Lolium multiflorum (198.6 mg kg⁻¹ of Cu), Cyperus compressus (276 mg kg⁻¹ of Cu), and Chrysanthemum leucanthemum (160.3 mg kg⁻¹ of Cu). The authors pointed out L. multiflorum potential for phytostabilization among the studied plant species, considering its dry matter production and ability to maintain the highest levels in its roots.

Regarding correlation analysis, Üçüncü et al. [32], studying the removal potential of Cu, Cr, and Pb by the aquatic macrophyte Lemna minor, found that Cr and Pb resembled each other in the time required for achieving the maximum removal rate (48 h), along with similar variations in contaminant levels during the experiment period (144 h). Thus, the authors obtained a high degree of correlation between these two metals.

Li et al. [33], performing a meta-analysis regarding Cu, Zn, and Cd uptake by aquatic plants, found that the ability of a given species to absorb a metal was strongly correlated with its ability to remove the other studied metals. However, the authors also identified other aspects such as the water pH, the morphology of plants (submerged or emerged), and the surface area exposed to water as highly influencing the uptake ability. The authors also identified a strong correlation between Cu and Zn, suggesting that the process of concentrating could be cooperative.

Considering the results found in BCF clustering analysis, the results found were higher than the BCF found by Afonso et al. [34] in a study aiming at the bioprospection of indigenous flora grown in copper-mining tailing areas for phytoremediation of metals. The BCF of Pb ranged from 0.4 by the species Juncus sp. to 16.4 in Solanum viarum Dunal. Regarding Cu, the species highlighted by the authors were Eryngium horridum Malme, Equisetum giganteum L., and Baccharis trimera, presenting values of BCF for Cu of 1.1, 1.5, and 1.8, respectively. Despite being somewhat lower, this species was highlighted by the authors considering its potential for Cu toleration and accumulation in its tissues, with impressive values of 440 mg kg⁻¹, being thus indicated for phytoremediation application.

Diverging from the results of TF clustering, Gomez et al. [35] found Cu and Ni being translocated to the shoots of S. montevidensis. Regarding the mechanism of phytoremediation, rhizofiltration was also detected as the main mechanism performed by aquatic macrophyte plants in other studies. One example is the research conducted by Woraharn et al. [36], which pointed out Typha angustifolia as accumulating Cd and Zn levels mainly in roots, in all tested treatments of the experiment.

In the studied area, the Santa Bárbara stream, the application of aquatic macrophytes could be performed in a controlled form, using the systems of floating islands for water decontamination.

5. Conclusions

The studied plant species showed tolerance for Cu, Cr, Cd, Pb, Ni, and Zn, demonstrating potential for application in phytoremediation. Comparison between the bioconcentration factors allowed the identification of H. ranunculoides as presenting excellent results of accumulation. In addition, the study evaluated the phytoremoval potential of each plant, and the highlighted species were H. grumosa and S. montevidensis.

This study serves as a background for the application of aquatic macrophyte species for phytoremediation purposes and for a biofilter proposal as floating-island systems for wetland treatments that combine the application of living plants with floating devices for water decontamination. In addition, it also helps other researchers to apply these aquatic macrophyte species with excellent natural potential for remediating other aquatic environments worldwide.

The correct disposal of plant biomass after remediating a contaminated area is primordial to ensuring that the metals do not re-enter the systems, affecting other sites. This study recommends the investigation of alternatives to be explored, such as heat treatment which is followed by co-product generation. Among the treatments one can mention are pyrolysis and the production of new adsorbents such as activated carbon. In addition, we recommend the bioprospection of native species in studies aiming at the remediation of other areas worldwide.