**3. Results**

The main characteristics of the analyzed soils are given in Table 1.

**Table 1.** Characteristics of urban soils used in the study.


\* Values exceeding MPC are marked in bold ([43,44]).

The KMP soil was characterized by a high content of Fe (45,750 mg/kg), exceeding the MPC for Mn (by 4.4 times) and Zn (by 41%); Tulachermet soil was characterized by a very high content of Fe (116,650 mg/kg), exceeding the MPC (APC) for V-Mn (by 40–50% for V and 10% for Mn), Ni (by 110–175%), Cu (by 127%), Zn (by 29–192%), As (by 115–220%), and oil products (by 4 times); the soil of Lenin Avenue was characterized by a high content of Fe (45,300 mg/kg), exceeding the permissible concentrations of Mn (by 6%) and Cu (by 186%). The soil of the embankment opposite the arms factory was characterized by the greatest pollution, it exceeded the permissible content of the complex of heavy metals, Mn (by 6%), Cr, Ni (3.6 times), Cu (36 times), Zn (83 times), As (by 25%), and petroleum products (by 9.5 times). In the background soils, the excess of MPC and APC for normalized elements was not noted.

In the performed study, the adaptive characteristics were studied and data on the sowing qualities of seeds and biometric parameters of *Echinochloa frumentacea* on soils contaminated with heavy metals were obtained. The results obtained are presented in Table 2. The soils of sanitary protection zones of metallurgical companies did not affect significantly the decrease in the germination of *Echinochloa frumentacea*. However, the soils of the SPZ of the highways and the soils of the embankment most contaminated with toxic elements caused a decrease in seed germination by 21–23% (Table 2).


**Table 2.** Effect of soil pollution with heavy meals on germinating ability and biometric parameters of *Echinochloa frumentacea*.

\*—*p* < 0.05 for the difference between experimental and background samples.

The biometric parameters of *Echinochloa frumentacea* grown on the soils of the experimental zones were measured one month after germination. The results of biometric measurements showed that soil contamination with trace elements had a toxic effect on plants, which was expressed in a decrease in their growth parameters. The maximum toxic effect on the experimental plants was exerted by the soils of the embankment, where there were multiple excesses of MPC (MAC) for a complex of toxic components: Cr, Ni (4 times), Cu (36 times), Zn (83 times), and oil products (more than 9 times). The shoot length of plants on the most polluted soils decreased 3.3 times compared to the background. The soils of the SPZ of metallurgical industries caused a decrease in the growth parameters of the shoot up to 35% compared to the background. The soils of the sanitary protection zone of the highway had an increased toxic effect, reducing the shoot length of the experimental plants by 43% compared to the background (Table 2).

The development of the root system, which performs the function of supplying photosynthetic organs with minerals and water, also suffered from the presence of toxic elements in soils. The length of roots on the soils of the sanitary protection zone of metallurgical industries and the arms factory (embankment) was lower than on the background soils by 19–20%. The most polluted soils of the highway had the maximum inhibitory effect on the development of the root system, reducing the length of the roots by 54% compared to the background (Table 2, Figure A1). Such pronounced impact of the soils of the SPZ of the highway on the growth and development of the root system can also be due to its salinization caused by the use of reagents during the winter.

The formation of plant biomass, which is important to take into account when carrying out phytoremediation measures, depends on the leaf surface area formed by plants, the work of photosynthetic pigments, and enzymatic activity. The quantitative characteristics of the photosynthetic apparatus are the most important indicators of plant adaptation to oxidative stress in the condition of soil contamination. The obtained results showed that the soils of the sanitary protection zone of metallurgical plants had a stimulating effect on the pigment apparatus of *Echinochloa frumentacea*.

The content of chlorophyll a in plants grown on the soils of Tulachermet was higher than in plants grown on the background soils by 36% and on the soils of KMP by 19% (Table 3). The content of chlorophyll a in plants grown on the soils of Lenin Avenue was lower than on the background soils by 7%. The greatest depressing effect on the content of pigments was exerted by the soils of the most polluted experimental zone, the embankment near the arms plant. The content of chlorophyll a in the shoots of Japanese millet was lower by 22%. At the same time, the content of chlorophyll b decreased by 34% compared with the control. On the soils of the embankment, the development of chlorosis and apical necrosis of the leaves of *Echinochloa frumentacea* was observed.


**Table 3.** Effect of the polyelemental pollution of soils on the quantitative characteristics of the photosynthetic pigments of *Echinochloa frumentacea*.

\*—*p* < 0.05 for the difference between experimental and background samples.

The content of chlorophyll b in the seedlings of *Echinochloa frumentacea* grown on the soils of metallurgical industries was higher than on the soils of the background zones by 9 and 53% for the soils of the SPZ KMP and Tulachermet, respectively. However, in plants growing on the soils of the highway, the content of chlorophyll b decreased by 10% compared to the control.

The same trends were observed for the content of carotenoids in the shoots of *Echinochloa frumentacea*: on the soils of the SPZ KMP and Tulachermet, it increased by 16 and 35%, respectively, compared to the background soils (Table 3). On the contrary, on soils with the highest pollution, it decreased significantly (by 19%). Carotenoids are one of the components of the antioxidant system of the plant, as well as a component of the pigment systems. Therefore, a decrease in their content can lead not only to a decrease in plant biomass but also to the development of severe oxidative stress [45].

The study showed that the pigment apparatus of *Echinochloa frumentacea* is well adapted to soils with a low degree of polyelement pollution (SPZ of metallurgical industries); however, on soils with strong polyelement anomalies (Embankment), the photosynthetic apparatus of the plant suffers due to a drastic decrease in the chlorophylls level (up to 2 times).

An important adaptive characteristic of plants grown on soils contaminated with HMs is their ability to produce antioxidant compounds that prevent the development of oxidative stress, which is accompanied by the oxidation of membrane lipids and disruption of transport and homeostasis processes [46]. The plant antioxidant system (AOS) includes low molecular weight antioxidants such as ascorbic acid (AA) and glutathione (GSH) [47,48]. Due to the ability to reversibly oxidize and reduce, ascorbic acid is involved in the most important plastic and energy processes of the plant cell: photosynthesis and respiration [47,49–51]; it is a recognized antioxidant [52–54], participates in growth and development processes [48,55], and forms plants' resistance to many adverse factors: UV radiation, pathogens [56], ozone [55,57], low temperatures, drought [48], soil salinity [48,58], heavy metals [59,60], and petroleum products [61].

The content of ascorbic acid and glutathione as components of AOS that prevent the development of oxidative stress in seedlings of *Echinochloa frumentacea* grown on soils with polyelement pollution was studied. The results are presented in Table 4.


**Table 4.** Effect of the polyelemental pollution of soils on the content of ascorbic acid and glutathione in *Echinochloa frumentacea* shoots.

\*—*p* < 0.05 for the difference between experimental and background samples.

The content of low-molecular antioxidants ascorbic acid and glutathione in the shoots of *Echinochloa frumentacea* was 2–2.5 times higher than in *Poa pratensis* grown on the same soils and varied within 6.9–8.2 mg/g (AA) and 1.35–1.67 mg/g (GSH) (Table 4). On the soils of the SPZ Tulachermet, KMP, and the highway, it was higher than in the control by 4, 19, and 17%, respectively. On the soils of the embankment, a slight decrease in the content of AA was observed compared to the control (by 6%). The content of glutathione in the shoots of *Echinochloa frumentacea* varied within 1.35–1.67 mg/g and was also higher on the soils of three experimental zones by 5, 15, and 9% for Tulachermet, KMP, and Lenin Avenue, respectively. The content of glutathione slightly decreased in comparison with the control on the soils of the embankment.

Study of the formation of the biomass of *Echinochloa frumentacea* showed that after 1.5 months of growth on the soils of SPZ of metallurgical plants, the plant formed more biomass than on the background soils. The biomass of Japanese millet grown on experimental soils varied in the range of 428–1024 g/m<sup>2</sup> (Table 5). The increase in shoot biomass relative to plants grown on background soils was 39% for plants on KMP soils and 48% for plants on Tulachermet soils. The biomass of plants grown on the soils of the SPZ of Lenin Ave. and the embankment was 12.6% and 20% less than on the background soils, respectively. The moisture content in the shoots on the background soils and Tulachermet was 82–84%, on the soils of KMP and highway in the experimental plants, the accumulation of dry matter increased and the water content decreased to 78–65%.


**Table 5.** Formation of *Echinochloa frumentacea* biomass on the soils with polymetal anomalies.

\*—*p* < 0.05 for the difference between experimental and background samples.

To assess the phytoremediation capacity of plants, it is important to determine the bioaccumulation of toxic elements in plant organs and to calculate the efficiency of removal of the element from contaminated soils.

Since there are no MPCs established for plants, and the data on the composition of elements in different works differ greatly depending on the applied methods of measurement and sample preparation techniques [62–65] (some studies were performed on unwashed plant material), when determining the bioaccumulative characteristics of plants, a comparison was made with the average data obtained for different plants species and published in Markert [66]—Reference Plants (RP).

The accumulative capacity of the shoots and roots of adult plants (4.5 months) grown on contaminated soils was assessed using atomic absorption spectroscopy.

The content of V in the roots of *Echinochloa frumentacea* grown on contaminated soils varied within 8.9–47.8 mg/kg of dry weight, which is 43–78 times higher than in the aboveground parts of the plant (Table 6). On the soil with a high content of V (Tulachermet), the accumulation of elements by the root system of Japanese millet increased significantly, while in the shoots the content of the element was at a low level of 0.61 mg/kg of dry weight.

**Table 6.** The average content of heavy metals in the shoots and root system of *Echinochloa frumentacea* grown on soils with polyelement anomalies (full vegetation period), mg/kg dry weight (AAS).


\*—*p* < 0.05 for the difference between experimental and background samples.

The root system of *Echinochloa frumentacea* is an active barrier that prevents the transport of V into the photosynthetic organs of the plant. The coefficient of transfer of the element from roots to shoots was 0.01–0.05. A possible way to clean V-contaminated soil using Japanese millet is rhizofiltration.

The content of Pb in the roots of *Echinochloa frumentacea* was 1.6–7.4 mg/kg, and in the shoots, 0.14–7.42 mg/kg dry weight (Table 6). An analysis of the obtained data showed that the root system of *Echinochloa frumentacea* ceases to perform barrier functions with respect to Pb when the threshold concentrations of the element in soils are exceeded. The content of Pb in Japanese millet grown on soils, in which Pb content exceeded the MPC value (71 mg/kg soil) was the highest and the same for shoots and roots, 7.4 mg/kg dry weight (Table 6). At the obtained values of the removal of the element from soils, which exceeded RP by more than 7 times, it should be expected that the content of the element in the soil during phytoremediation measures will come to acceptable values in 4–5 years.

For most plants, the critical level of copper content is 10–20 mg/kg dry weight [67]. Cu accumulation by the shoots and roots of *Echinochloa frumentacea* exceeded the threshold of normal regulation for plants on KMP-contaminated soils, where the content of the copper was quite high (52 mg/kg) and amounted to 16.3 mg/kg in shoots and 41 mg/kg in roots. In general, the root system of Japanese millet had a high affinity for copper. Therefore, one of the ways to clean soils from Cu in the case of polyelement pollution using *Echinochloa frumentacea* is rhizofiltration. However, in the shoots of KMP, highway, and embankment soils, the content of Cu varied from 16.3 to 96.6 mg/kg of dry weight (Tables 6 and 7). These values are higher than those given in Murillo et al. [68] for sorghum of 4.6–11.7 mg/kg; therefore, it can be assumed that Japanese millet can also be used for the phytoextraction of Cu from soils with polyelement contamination.


**Table 7.** Bioaccumulation of toxic elements by *Echinochloa frumentacea* grown on soils with polyelement anomalies (vegetation period 1.5 months) (average data for ICP-MS and INAA), mg/kg.

\*—*p* < 0.05 for the difference between experimental and background samples.

However, on soils with a high content of V and Cr, the absorption of the element by the plant decreased, which should be taken into account when carrying out remediation measures. According to the literature data, when the content of copper in plants exceeds the critical level, the signs of toxicity may develop: the concentration of chlorophyll decreases, and the growth of shoots and roots is suppressed [69]. The complex of HM pollutants in the experimental soils suppressed the growth parameters of Japanese millet; however, on the soils of the sanitary protection zone of metallurgical industries, a compensatory mechanism was observed and the content of photosynthetic pigments increased (Table 3).

Accumulation of Mo by shoots of *Echinochloa frumentacea* on the soils of Tulachermet up to 12.7 mg/kg dry weight, which is 3–6 times more than on the soils of other experimental zones and 12 times more than on the background soils (Table 7). High Mo-uptake can be explained by the slightly alkaline pH of analyzed soils [70].

The accumulation of Cd did not exceed 0.28–0.52 μg of dry weight in most of the experimental variants. These values are 5.5–10.4 times higher than the average data for RP. The Cd enrichment factor for *Echinochloa frumentacea* is 5.6–19 and is at its maximum on the soils of the sanitary protection zone of the highway. Low Cd uptake can be associated with the presence of zinc in soil, which inhibits Cd uptake and translocation under cadmium/zinc combined stress [71].

The accumulation of Cr by the shoots of *Echinochloa frumentacea* on soils with the highest polyelement pollution, where the content of the element was many times higher than APC (1260 mg/kg), was higher than for other soils (Cr content up to 117 mg/kg) and amounted to 4.7 mg/kg of dry weight (Table 7). These values were 3 times higher than RP; however, with the soils' contamination with Cr when its content exceeds TEC 4 times, this level of bioaccumulation of the element is not enough for effective biological treatment in terms less than 10 years. On the soils of other sampling sites, Cr accumulated in the shoots of *Echinochloa frumentacea* was up to three times higher than the upper limit of normal values (0.1–0.5 mg/kg, [62]): 0.93–1.62 mg/kg.

The critical level of Co in most plants starts from 0.4 mg/kg. Cereals are most sensitive to their excess [72] The Co content in Japanese millet ranged from 0.10 to 0.27 mg/kg dry weight and lies within the average values characteristic for vegetation, 0.02–1 mg/kg [62].

The content of Ni in *Echinochloa frumentacea* plants varied in the range of 1.4–4.7 mg/kg dry weight and its maximum accumulation was attained on the most polluted soils. These values are 2–3 times higher than the average values for vegetation grown on the soils of Tulachermet, the highway, and embankment.

The content of As in the shoots of *Echinochloa frumentacea* was 0.15–0.29 mg/kg of dry weight. Its content on the soils of the SPZ of metallurgical industries and the embankment was 2–3 times higher than RP values. However, based on the data on the bioaccumulation and removal of As, Japanese millet is not recommended for the phytoremediation of soils contaminated with arsenic.

The critical concentrations of manganese in plants vary from 220 to 5300 mg/kg dry weight [67]. Japanese millet is not an Mn bioaccumulator. The content of the element in the shoots of *Echinochloa frumentacea* falls within the limits of average values and amounts to 57–199 mg/kg of dry weight and lies below the limits of critical values (Table 7).

Visible symptoms of zinc toxicity appear with its concentrations in plants above 300 mg/kg but are sometimes possible at lower concentrations (100 mg/kg) [67]. The content of Zn in the shoots of *Echinochloa frumentacea* was 53 mg/kg on the background soils and varied in the range of 41–457 mg/kg dry weight on the soils of the experimental zones. The content of the element exceeded the critical concentrations for plants on the most polluted soils of the embankment, which could cause the observed toxic effects. The factor *Echinochloa frumentacea* enrichment with Zn was two on the soils of the KMP and Lenin Avenue and eight on the soils of the SPZ of the arms plant (embankment). These are a good indicator for the phytoextraction of the element from soils, taking into account the accumulation of biomass by plants.

The content of Fe in the shoots of *Echinochloa frumentacea* varied within 282–389 mg/kg of dry weight and was below the critical concentration limits for plants (500 mg/kg [62,67]) but higher than the average data for RP 2 or more times. Apparently, for Fe, which is a soil pollutant in the region, there is no barrier at the level of the root system of the plant.

#### *Culture-Based Assessment of Microbial Communities in the Rhizosphere of E. frumentacea*

At the end of the experiment, the total number of heterotrophic bacteria, and the number of actinomycetes and micromycetes, were determined in the rhizosphere of plants. The results of the analysis of the main groups of cultivated heterotrophic soil microorganisms in the rhizosphere of *E. frumentacea* are shown in Figure A2.

The abundance of rhizospheric microorganisms depended on the kind of soil in which the plants were grown. The number of bacteria in the plant rhizosphere varied from 0.8 to 3.3 × 10<sup>7</sup> CFU/g of soil sampled from the Embankment and Background, respectively. It was found that the soil of the Embankment near the Tula Arms Plant was the most polluted (Table 1). The high content of almost all metals analyzed clearly caused the toxicity of this soil, which led to the inhibition of the number of all groups of microorganisms studied, bacteria, actinomycetes, and micromycetes. In addition, this soil was characterized by a minimum content of humus carbon (Table 1), which is important for the development of soil microflora.

The number of cultivated heterotrophic bacteria in the rhizosphere soil from the Embankment (Figure 1a) was two times lower than in the background soil (Yasnaya Polyana). The almost two-fold excess of the number of bacteria in the Tulachermet soil may be due to the presence of an additional carbon source in this soil (oil products' content was 4.1 g/kg, and humus carbon was higher than in the Embankment soil). However, an even higher concentration of oil products (9.5 g/kg) did not support the number of bacteria in the soil of the Embankment, which was associated with the toxic effect of heavy metals on oxidative enzymes involved in the hydrocarbon degradation. The soil from Lenin Ave. also stimulated the rhizospheric bacteria; however, there was no obvious explanation of this effect from the data of analyses of this soil (Table 1). We can only assume that in terms of background pollution, the Lenin Ave. soil, just like the KMP soil, was the closest to the background one.

**Figure 1.** The number of cultivable heterotrophic bacteria, actinomycetes, and micromycetes in the rhizosphere of *E. frumentacea*. (**a**)—bacteria; (**b**)—actinomycetes; (**c**)—micromycetes. Error bars mean standard errors/\*—*p* < 0.05 for the difference between experimental and background samples.

The number of actinomycetes in the rhizosphere of *E. frumentacea* varied from 0.8 to 3.3 × 10<sup>6</sup> CFU/g of the soil in the Embankment and KMP samples, respectively. The influence of urban soils on the actinomycete population in the rhizosphere was most noticeable in the soils of the Embankment and KMP (Figure 1b). In the first case, in relation to bacteria, the inhibitory effect of the contaminated urban soil was visible: the number of actinomycetes in the rhizosphere decreased by 2.3 times compared to the control soil. However, in the KMP soil, stimulation of actinomycetes (1.7 times) was observed. Probably, this could be associated with an increase in the actinomycete taxa resistance to heavy metals

and the ability to participate in remediation of the environment [73]. This explanation was also confirmed by our data on the taxonomic structure of the rhizosphere community of *E. frumentacea* grown in the KMP soil. There were no significant differences between the abundance of actinomycetes in the Tulachermet and Lenin Ave. soils and the control soil of Yasnaya Polyana.

The number of micromycetes in the rhizosphere of *E. frumentacea* reached 5.5–13.3 × 10<sup>5</sup> CFU/g of the soil of the Embankment and KMP samples, respectively (Figure 1c). The changes in the abundance of micromycetes depending on the soil type were similar to those described for actinomycetes. The minimal number of micromycetes was observed for the Embankment soil (a 1.6-fold decrease from the control), and the maximal number was observed for the KMP soil (a 1.5-fold increase in comparison with the control), which could be associated with specific stimulation by the plant of micromycetes taxa resistant to heavy metals.

The sequencing of 16s rRNA gene from rhizosphere samples resulted in 133.776 reads. After data denoising and chimera screening, a total of 92.527 sequences were used for further identification. The average number of nucleotide sequences in the library per sample was 15.728.

To characterize biodiversity and carry out a comparative analysis of the communities, the parameters of ܤ-diversity were calculated, the taxonomic composition of the community in all samples was determined and compared, and taxa that reliably decreased or increased their number in the rhizospheres of the plants studied were identified.

The calculation of the sampling effort indicated that, on average, 73.7% of the true diversity of the rhizosphere communities of *E. frumentacea* were covered (Table 8). The ܤ-diversity was assessed using species richness indices (observed OTUs) and the Shannon and the Simpson indices in the rhizosphere of *E. frumentacea* grown in the control soil, and the average number of OTUs was minimal (2013 on average) in comparison with other soils. The species richness index of plants grown in Tulachermet was maximal (3816 observed OTUs) (Table 8).


**Table 8.** The ܤ-diversity indices for rhizospheric microbial communities of *E. frumentacea* grown on different soils.

The Shannon and Simpson indices indicated a high taxonomic diversity of bacterial communities in the rhizosphere of plants grown in contaminated urban soils, while the background soil (Yasnaya Polyana) was characterized by the lowest taxonomic diversity (Table 8).

Figure A2 shows the bacterial types that dominate in the rhizosphere of *E. frumentacea.* The Proteobacteria, Actinobacteria, Planctomycetes, Acidobacteria, as well as Bacteroides and Chloroflexi phyla occupied the dominant positions in the taxonomic structure of the rhizosphere community of *E. frumentacea*. However, the ratio of these types was different for plants grown on different soils. Thus, the share of Proteobacteria in the total taxonomic structure of the microbial community in the background soil (Yasnaya Polyana) reached 52%, but it significantly decreased when plants were grown on urban soils. The minimal amount of Proteobacteria was noted in the soils of KMP (27%) and Tulachermet (28%), followed by the soils of Lenin Ave. (30%) and the Embankment (36%). At the same time, the share of another dominant phylum, Actinobacteria, increased from 18% (in control) to 22, 26, 29, and 31% in the soils of the Embankment, Tulachermet, KMP, and Lenin Ave.,

respectively. In the rhizospheric communities of plants grown in the polluted urban soils, the share of other phyla also clearly increased. Thus, the share of Acidobacteria in the control soil was only 4% but increased to 5, 8, 9, and 9.2 in the soils of Lenin Ave., the Embankment, Tulachermet, and KMP, respectively. The Planctomycetes phylum in the rhizospheric community of *E. frumentacea* grown in the control soil was only 4%, while in contaminated urban soils it increased to 6, 9, and 10% (in the soils of the Embankment, Tulachermet, Lenin Ave., and KMP, respectively). In contaminated soils, an increase in Chloroflexi share from 3% in the control to 6–7% in urban soils was noted. In addition to Proteobacteria, under the influence of pollution, the share of another phylum, Firmicutes, decreased (albeit less significantly), which was more than 4% in clean soil and decreased to 2–3% in polluted soil. Changes in the share of other phyla of the microbial community in the rhizosphere of *E. frumentacea* under the influence of pollution were unsignificant. Thus, a comparative analysis of the taxonomic structure of the rhizosphere microbial community at the phylum level revealed a distinct influence of urban soils characterized by technogenic polyelement anomalies.

The taxonomic analysis of the dominant phylum Actinobacteria made it possible to identify 39 families, among which 16 families occupied a fundamental position, presented in Table A1 and constituting from 60% to 77% of all found families of Actinobacteria in the rhizosphere of *E. frumentacea*. The Gaiellaceae family had the maximal share in the population of rhizospheric actinomycetes in almost all soil samples; only in the soil of the Embankment was the abundance of Gaiellaceae minimal. At the same time, this soil was characterized by an increased share of the Micrococcaceae family (Table A1). Nocardioidaceae, Micromonosporaceae (0.9–3.4%), as well as representatives of the Solirubrobacterales and 0319-7L14 orders were other notable actinomycetes in the rhizosphere of *E. frumentacea*.

As part of another dominant phylum, Proteobacteria, 66 families of bacteria were identified. Thirteen families accounting for 72–86% of all OTUs assigned to Proteobacteria made the most significant contribution to the rhizosphere microbiome structure of *E. frumentacea*, (Table A1). Alphaproteobacteria is the most numerous class of the Proteobacteria phylum, accounting for 41 to 53% of all Proteobacteria. Among Alphaproteobacteria, the dominant position was occupied by the Sphingomonadaceae (2.6–12.7%) and Hyphomicrobiaceae (2.0–5.6%) families. The Betaproteobacteria class accounted for 15 to 32% of all OTUs assigned to the Proteobacteria phylum, and the Comamonadaceae (0.8–5.1%) and Oxalobacteraceae (1.3–3.5%) families were its dominant representatives. The Gammaproteobacteria class accounted for 16 to 30% of all OTUs assigned to the Proteobacteria phylum, and the Xanthomonadaceae family, whose share reached 1.6 to 13.5%, was its dominant representative. It can be assumed that, along with Sphingomonadaceae, the Xanthomonadaceae family can also form the basis of the rhizobiome of *E. frumentacea*. The share of these families decreased when plants were grown in all urban soils, except for Lenin Ave., where it distinctly increased, as did the share of the Hyphomicrobiaceae and Oxalobacteraceae families. Among other bacterial taxa the dominant position in the soil of the Embankment was occupied by representatives of Comamonadaceae, and in the soil of KMP, by Sphingomonadaceae, Hyphomicrobiaceae, and Xanthomonadaceae. The share of Oxalobacteraceae increased in all urban soils compared to the control soil. In the rhizosphere of plants grown in Tulachermet soil, it reached a maximum (3.5%).

In the course of this study, 22 morphologically different strains of bacteria were isolated from the rhizosphere of *E. frumentacea* plants. All isolates were characterized by their resistance to heavy metals in the environment and plant growth-promoting potential. The strains combining these properties were transferred to the collection of rhizospheric microorganisms of the Institute of Biopharmaceutics, Russian Academy of Sciences (http: //collection.ibppm.ru, accessed on 14 March 2022).
