Effect of Heavy-Metal-Resistant PGPR Inoculants on Growth, Rhizosphere Microbiome and Remediation Potential of Miscanthus × giganteus in Zinc-Contaminated Soil

Microbial-assisted phytoremediation is considered a more effective approach to soil rehabilitation than the sole use of plants. Mycolicibacterium sp. Pb113 and Chitinophaga sp. Zn19, heavy-metal-resistant PGPR strains originally isolated from the rhizosphere of Miscanthus × giganteus, were used as inoculants of the host plant grown in control and zinc-contaminated (1650 mg/kg) soil in a 4-month pot experiment. The diversity and taxonomic structure of the rhizosphere microbiomes, assessed with metagenomic analysis of rhizosphere samples for the 16S rRNA gene, were studied. Principal coordinate analysis showed differences in the formation of the microbiomes, which was affected by zinc rather than by the inoculants. Bacterial taxa affected by zinc and the inoculants, and the taxa potentially involved in the promotion of plant growth as well as in assisted phytoremediation, were identified. Both inoculants promoted miscanthus growth, but only Chitinophaga sp. Zn19 contributed to significant Zn accumulation in the aboveground part of the plant. In this study, the positive effect of miscanthus inoculation with Mycolicibacterium spp. and Chitinophaga spp. was demonstrated for the first time. On the basis of our data, the bacterial strains studied may be recommended to improve the efficiency of M. × giganteus phytoremediation of zinc-contaminated soil.


Introduction
The interest in nonfood perennial plant species that can be used as alternative energy sources (biofuels) continues unabated. Miscanthus × giganteus Greef et Deu. (Poaceae family), known as the giant miscanthus, a sterile hybrid of Miscanthus sinensis and Miscanthus sacchariflorus, is a promising bioenergetic species owing to its high biomass yield and low production costs [1][2][3]. Planting this crop on contaminated soils unsuitable for food production, in addition to yielding biomass for biofuel production, can simultaneously solve the soil remediation problem [3][4][5]. M. × giganteus is resistant to various levels of heavy metal pollution of soil and can be applied for soil cleanup [6,7]. At the same time, the microorganisms associated with its root zone can affect both biomass production and soil cleanup, promoting plant growth directly (through improved plant nutrition and phytohormone production) and/or indirectly (through a change in the bioavailability of metals) [8,9]. On the basis of this, the microbe-assisted phytoremediation is considered a more preferable approach than the sole use of plants [10][11][12].

Experimental Design
One-year-old rhizomes of Miscanthus × giganteus J.M. Greef, Deuter ex Hook, Renvoize were obtained from the Institute of Plant Biology and Biotechnology in Almaty, Kazakhstan, and were used in the research. For planting, experimental pots were filled by pouring expanded clay (300 mL per pot; granule diameter 1.5-2.0 cm) as a drainage on the bottom of the vessel. The drainage was closed with gauze, and river sand (300 g per pot) was poured on top and covered again with gauze. Then, the pots were filled with soil (1000 g dry weight per pot). Miscanthus rhizomes were planted in the prepared pots (2 pieces per vessel). After planting, the soil surface was covered with sand (100 g per pot). All experiments were performed in triplicate. The pots with uncontaminated soil (without Zn treatment) were used as controls. Soil moisture in the pots was maintained at 50% of the full moisture capacity by daily watering with standing tap water, the need for which was determined by weighing the pots. The plants were grown in a greenhouse at 24-28/20-22 • C and natural illumination for 4 months (from May to September 2021).
Inoculation of one-week-old seedlings with heavy-metal-resistant PGPR strains Zn19 and Pb113 was carried out as follows. Two-day-old microbial biomass was collected from R2A agar medium [40], washed with saline twice, and resuspended in the plant watering liquid. Inoculation of plants was carried out by watering of the corresponding variants with a microbial suspension to the final concentration of cells in the soil of at least 10 7 cells per gram. The plants were inoculated once.

Plant Biomass Measurement
To control the changes in plant biomass, the weight of the rhizomes planted was measured before planting. In the course of the experiment, the plant tillering and height were measured monthly. At the end of cultivation, the plants were removed from the pots, and the shoots and roots were separated from the rhizomes. Samples of young roots with the remaining attached rhizospheric soil were taken for microbiological analysis. The remaining roots were separated from the rhizomes, washed free of soil with tap water, weighed and dried to constant weight in an oven at 70 • C. The remaining rhizomes were also washed free of soil with tap water and weighed.

Measurement of Metal Content in Plant Biomass
The content of Zn in plant biomass was determined by atomic absorption spectroscopy as reported previously [41].
The translocation factor (TF) was calculated for each treatment. The TF, reflecting the plant's ability to transport and accumulate metals into aboveground biomass, was Microorganisms 2023, 11, 1516 4 of 19 calculated as the ratio of the metal concentration in the shoots to the metal concentration in the roots.

16S rRNA Gene-Based Metagenomic Analyses of Rhizosphere Soil and Bioinformatics
Extraction and purification of soil DNA, 16S rRNA sequencing library preparation, 16S rRNA amplicon sequencing, quality filtering of reads, OTU taxonomy, characterization of the richness and evenness of bacterial communities, assessment of the similarity between the microbial composition of samples were carried out as described in [42]. Raw reads are deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA973256.

Statistics
All the experimental data obtained were subjected to statistical processing, calculating the average values, for comparison of which the standard deviation and the confidence interval at p ≤ 0.05 were used. Calculations were performed in Microsoft Excel 2016 (Microsoft, Redmond, WA, USA). To compare the average values, identify the effect of soil pollution and/or microbial inoculation factors, analysis of variance and Spearman's rank correlations (r s ) were used, which were performed in Statistica 13.3.721.1 (TIBCO Software Inc. 2017, Statsoft, Moscow, Russia).

Characteristics of Bacterial Inoculants
Previously isolated strains Pb113 and Zn19 were studied for their resistance to Pb 2+ and Zn 2+ and for plant growth promoting potential, manifested through nitrogen fixation, phosphorous mobilization, and synthesis of siderophores and phytohormones. These isolates manifested combined properties of PGPR and resistance to metals, were selected (Table 1). (a) (b) Figure 1. Maximum-likelihood phylogenetic trees based on the 16S rRNA gene sequences from the strains under study-Pb113 (a) and Zn19 (b)-and their phylogenetic neighbors. The search for the best evolutionary model and the reconstruction of phylogenetic trees were conducted in MEGA7 [43]. Nonuniformity of evolutionary rates among sites was imitated by using a discrete Gamma distribution (+G) and by assuming that a certain fraction of sites was evolutionarily invariable (+I  The search for the best evolutionary model and the reconstruction of phylogenetic trees were conducted in MEGA7 [43]. Nonuniformity of evolutionary rates among sites was imitated by using a discrete Gamma distribution (+G) and by assuming that a certain fraction of sites was evolutionarily invariable (+I As shown by the phylogenetic analysis, Zn19 occupies a robust position among the bacteria of the genus Chitinophaga and is most closely related to C. polysaccharea (Figure 1b). The pairwise sequence similarity of Zn19 to C. polysaccharea MRP-15 T was 99.00%, whereas with other relatives, it was below the threshold for species demarcation. Thus, it can be concluded that Pb113 belongs to the Mycolicibacterium genus and that Zn19 belongs to the Chitinophaga genus.  Microorganisms 2023, 11, x FOR PEER REVIEW As shown by the phylogenetic analysis, Zn19 occupies a robu bacteria of the genus Chitinophaga and is most closely related to C 1b). The pairwise sequence similarity of Zn19 to C. polysaccharea M whereas with other relatives, it was below the threshold for species can be concluded that Pb113 belongs to the Mycolicibacterium genus a to the Chitinophaga genus.

Effect of Bacterial Inoculants and Zn on the Growth Performances of M
The Zn contamination of soil and inoculation affected M. × gig ering ( Figure 2).
Owing to the large variation in the height of shoots of differen significant differences between treatments. However, a clear trend o shoot height was observed (Figure 2a). The inoculation effect on pended on the microorganism used and on the soil treatment. Thu soil, there was a trend to increase the shoot height under the influen Zn19 and Mycolicibacterium sp. Pb113, and in Zn-contaminated soil the influence of Mycolicibacterium sp. Pb 113. Plant tillering was p (62%) by Chitinophaga sp. Zn19 only (Figure 2b). -uncontaminated + Chitinop contaminated, non-inoculated; -Zn-contaminated + Mycolicibacterium s taminated + Chitinophaga sp. Zn19. Values represent means, and bars repre (n ≥ 3); different letters mean significant difference between treatments at p The changes in miscanthus biomass under the influence of the lants and zinc are illustrated in Figure 3.
Despite the attempts to evenly distribute the biomass of plantin at the beginning of the experiment, there still was some differenc Therefore, when the final biomass of the rhizomes was measured weight of the rhizomes was also taken into account (Figure 3a). A obtained, no significant change in wet rhizome weight was found in control soil. There was only a tendency to decrease biomass under t bacterium sp. Pb 113. In Zn-contaminated soil, there was no significa zome weight, but there was a tendency to decrease the biomass o non-inoculated plants. Microorganisms 2023, 11, x FOR PEER REVIEW As shown by th bacteria of the genu 1b). The pairwise se whereas with other r can be concluded tha to the Chitinophaga g

Effect of Bacterial
The Zn contami ering ( Figure 2).
Owing to the la significant difference shoot height was ob pended on the micro soil, there was a tren Zn19 and Mycoliciba the influence of Myc (62%) by Chitinophag (a) Microorganisms 2023, 11, x FOR PEER REVIEW 6 As shown by the phylogenetic analysis, Zn19 occupies a robust position amon bacteria of the genus Chitinophaga and is most closely related to C. polysaccharea (Fi 1b). The pairwise sequence similarity of Zn19 to C. polysaccharea MRP-15 T was 99. whereas with other relatives, it was below the threshold for species demarcation. Th can be concluded that Pb113 belongs to the Mycolicibacterium genus and that Zn19 bel to the Chitinophaga genus.

Effect of Bacterial Inoculants and Zn on the Growth Performances of M. × giganteus
The Zn contamination of soil and inoculation affected M. × giganteus height and ering ( Figure 2).
Owing to the large variation in the height of shoots of different ages, there wer significant differences between treatments. However, a clear trend of zinc inhibition o shoot height was observed (Figure 2a). The inoculation effect on the shoot heigh pended on the microorganism used and on the soil treatment. Thus, in uncontamin soil, there was a trend to increase the shoot height under the influence of Chitinophag Zn19 and Mycolicibacterium sp. Pb113, and in Zn-contaminated soil, that was only u the influence of Mycolicibacterium sp. Pb 113. Plant tillering was promoted significa (62%) by Chitinophaga sp. Zn19 only (Figure 2b). The changes in miscanthus biomass under the influence of the rhizobacterial in lants and zinc are illustrated in Figure 3.
Despite the attempts to evenly distribute the biomass of planting material (rhizo at the beginning of the experiment, there still was some differences between the Therefore, when the final biomass of the rhizomes was measured, the initial plan weight of the rhizomes was also taken into account ( Figure 3a). According to the obtained, no significant change in wet rhizome weight was found in the uncontamin control soil. There was only a tendency to decrease biomass under the influence of M bacterium sp. Pb 113. In Zn-contaminated soil, there was no significant change in we zome weight, but there was a tendency to decrease the biomass of the rhizomes o non-inoculated plants. Owing to the large variation in the height of shoots of different ages, there were no significant differences between treatments. However, a clear trend of zinc inhibition of the shoot height was observed ( Figure 2a). The inoculation effect on the shoot height depended on the microorganism used and on the soil treatment. Thus, in uncontaminated soil, there was a trend to increase the shoot height under the influence of Chitinophaga sp. Zn19 and Mycolicibacterium sp. Pb113, and in Zn-contaminated soil, that was only under the influence of Mycolicibacterium sp. Pb 113. Plant tillering was promoted significantly (62%) by Chitinophaga sp. Zn19 only (Figure 2b).
The changes in miscanthus biomass under the influence of the rhizobacterial inoculants and zinc are illustrated in Figure 3.
Despite the attempts to evenly distribute the biomass of planting material (rhizomes) at the beginning of the experiment, there still was some differences between the pots. Therefore, when the final biomass of the rhizomes was measured, the initial planting weight of the rhizomes was also taken into account (Figure 3a). According to the data obtained, no significant change in wet rhizome weight was found in the uncontaminated control soil. There was only a tendency to decrease biomass under the influence of Mycobacterium sp. Pb 113. In Zn-contaminated soil, there was no significant change in wet rhizome weight, but there was a tendency to decrease the biomass of the rhizomes of the non-inoculated plants.
Plant inoculation with Mycolicibacterium sp. Pb113 caused a significant increase in the shoot (  As shown by the phylogenetic analysis, Zn19 occupies a robust position among the bacteria of the genus Chitinophaga and is most closely related to C. polysaccharea ( Figure  1b). The pairwise sequence similarity of Zn19 to C. polysaccharea MRP-15T was 99.00%, whereas with other relatives, it was below the threshold for species demarcation. Thus, it can be concluded that Pb113 belongs to the Mycolicibacterium genus and that Zn19 belongs to the Chitinophaga genus.

Effect of Bacterial Inoculants and Zn on the Growth Performances of M. × giganteus
The Zn contamination of soil and inoculation affected M. × giganteus height and tillering ( Figure 2).
Owing to the large variation in the height of shoots of different ages, there were no significant differences between treatments. However, a clear trend of zinc inhibition of the shoot height was observed ( -Zn-contaminated + Mycolicibacterium sp. Pb-113; -Zn-contaminated + Chitinophaga sp. Zn19. Values represent means, and bars represent confidence interval (n ≥ 3); different letters mean significant difference between treatments at p ≤ 0.05.
The changes in miscanthus biomass under the influence of the rhizobacterial inoculants and zinc are illustrated in Figure 3.
Despite the attempts to evenly distribute the biomass of planting material (rhizomes) at the beginning of the experiment, there still was some differences between the pots. Therefore, when the final biomass of the rhizomes was measured, the initial planting weight of the rhizomes was also taken into account ( Figure 3a). According to the data obtained, no significant change in wet rhizome weight was found in the uncontaminated control soil. There was only a tendency to decrease biomass under the influence of Mycobacterium sp. Pb 113. In Zn-contaminated soil, there was no significant change in wet rhizome weight, but there was a tendency to decrease the biomass of the rhizomes of the non-inoculated plants. Microorganisms 2023, 11, x FOR PEER REVIEW As shown by the phylogenetic analysis, Z bacteria of the genus Chitinophaga and is most 1b). The pairwise sequence similarity of Zn19 whereas with other relatives, it was below the t can be concluded that Pb113 belongs to the Myc to the Chitinophaga genus.

Effect of Bacterial Inoculants and Zn on the Gro
The Zn contamination of soil and inoculat ering ( Figure 2).
Owing to the large variation in the height significant differences between treatments. How shoot height was observed ( Figure 2a). The in pended on the microorganism used and on the soil, there was a trend to increase the shoot heig The changes in miscanthus biomass under lants and zinc are illustrated in Figure 3.
Despite the attempts to evenly distribute th at the beginning of the experiment, there still Therefore, when the final biomass of the rhiz weight of the rhizomes was also taken into ac obtained, no significant change in wet rhizome control soil. There was only a tendency to decre bacterium sp. Pb 113. In Zn-contaminated soil, t zome weight, but there was a tendency to dec non-inoculated plants. As shown by the phylogenetic analysis, Zn19 occupies a robust position among the bacteria of the genus Chitinophaga and is most closely related to C. polysaccharea ( Figure  1b). The pairwise sequence similarity of Zn19 to C. polysaccharea MRP-15 T was 99.00%, whereas with other relatives, it was below the threshold for species demarcation. Thus, it can be concluded that Pb113 belongs to the Mycolicibacterium genus and that Zn19 belongs to the Chitinophaga genus.

Effect of Bacterial Inoculants and Zn on the Growth Performances of M. × giganteus
The Zn contamination of soil and inoculation affected M. × giganteus height and tillering ( Figure 2).
Owing to the large variation in the height of shoots of different ages, there were no significant differences between treatments. However, a clear trend of zinc inhibition of the shoot height was observed (Figure 2a). The inoculation effect on the shoot height depended on the microorganism used and on the soil treatment. Thus, in uncontaminated soil, there was a trend to increase the shoot height under the influence of Chitinophaga sp. The changes in miscanthus biomass under the influence of the rhizobacterial inoculants and zinc are illustrated in Figure 3.
Despite the attempts to evenly distribute the biomass of planting material (rhizomes) at the beginning of the experiment, there still was some differences between the pots. Therefore, when the final biomass of the rhizomes was measured, the initial planting weight of the rhizomes was also taken into account (Figure 3a). According to the data obtained, no significant change in wet rhizome weight was found in the uncontaminated control soil. There was only a tendency to decrease biomass under the influence of Mycobacterium sp. Pb 113. In Zn-contaminated soil, there was no significant change in wet rhizome weight, but there was a tendency to decrease the biomass of the rhizomes of the non-inoculated plants. 16S rRNA sequencing from 16 rhizosphere samples generated a total of 807,695 raw reads. Data denoising and chimera screening was carried out. A total of 10,771 joined read pairs per sample were used for identification. The sequences with >97% similarity were combined by classification into operational taxonomic units (OTUs). The OTUs were assigned to 48, 126, 264, 380, and 1148 taxa at the phylum, class, order, family, and genus levels, respectively.
αand β-Diversity calculations were performed to assess the richness of the microbial communities and characterize the microbial diversity (Table 2 and Figure 4).
The α-diversity was measured by using Chao1, Simpson, and Shannon species richness indices and phylogenetic diversity (Faith's PD) ( Table 2). Shannon's, Simpson's, and Faith's PD indices of bacterial α-diversity indicate that the microbial communities of the rhizosphere of the inoculated plants were more diverse than those for the non-inoculated ones.  The influence of soils on the formation of rhizosphere communities was revealed by comparing the microbiomes of the rhizosphere of different samples. Principal coordinate analysis showed differences in the formation of the taxonomic structure of the miscanthus rhizosphere microbiomes, which was largely affected by the zinc contamination of soil.

Taxonomic Structure of Rhizosphere Microbial Communities
The results of MiSeq sequencing showed that the miscanthus rhizosphere communities included 1148 genera of bacteria belonging to 380 families of 48 phyla. Figure 5 represents the relative abundances of OTUs associated at the phylum level in the rhizosphere of M. × giganteus non-inoculated and inoculated with the heavy-metalresistant PGPR studied. In the rhizosphere communities of M. × giganteus, most OTUs were associated with the Actinobacteriota (31-41%), Proteobacteria (13-22%), Acidobacteriota (7-17%), Bacteroidota (4-13%), and Gemmatimonadota (4-10%). The number of OTUs assigned to other phyla was much smaller.
Soil contamination with zinc had a significant effect on the taxonomic profile of the rhizosphere microbiome of both inoculated and non-inoculated M. × giganteus plants. Under the influence of zinc in the rhizosphere of plants non-inoculated and inoculated with Mycolicibacterium sp. Pb113 or Chitinophaga sp. Zn19, the abundance of Proteobacteria increased by 37%, 12%, and 61%, respectively (p ≤ 0.0003), and that of Bacteroidota increased by 133%, 52%, and 170%, respectively, (p ≤ 0.0003). Yet, Zn contamination decreased the abundance of: Acidobacteriota by 14%, 41%, and 58%, respectively (p ≤ 0.0005); Methylomirabilota by 15%, 47%, and 77%, respectively (p ≤ 0.004); Myxococcota by 54%, 11%, and 35%, respectively (p ≤ 0.002); and Planctomycetota by 78%, 78%, and 56%, respectively (p The influence of soils on the formation of rhizosphere communities was revealed by comparing the microbiomes of the rhizosphere of different samples. Principal coordinate analysis showed differences in the formation of the taxonomic structure of the miscanthus rhizosphere microbiomes, which was largely affected by the zinc contamination of soil.

Taxonomic Structure of Rhizosphere Microbial Communities
The results of MiSeq sequencing showed that the miscanthus rhizosphere communities included 1148 genera of bacteria belonging to 380 families of 48 phyla. Figure 5 represents the relative abundances of OTUs associated at the phylum level in the rhizosphere of M. × giganteus non-inoculated and inoculated with the heavy-metalresistant PGPR studied. In the rhizosphere communities of M. × giganteus, most OTUs were associated with the Actinobacteriota (31-41%), Proteobacteria (13-22%), Acidobacteriota (7-17%), Bacteroidota (4-13%), and Gemmatimonadota (4-10%). The number of OTUs assigned to other phyla was much smaller.
Soil contamination with zinc had a significant effect on the taxonomic profile of the rhizosphere microbiome of both inoculated and non-inoculated M. × giganteus plants. Under the influence of zinc in the rhizosphere of plants non-inoculated and inoculated with Mycolicibacterium sp. Pb113 or Chitinophaga sp. Zn19, the abundance of Proteobacteria increased by 37%, 12%, and 61%, respectively (p ≤ 0.0003), and that of Bacteroidota increased by 133%, 52%, and 170%, respectively, (p ≤ 0.0003). Yet, Zn contamination decreased the abundance of: Acidobacteriota by 14%, 41%, and 58%, respectively (p ≤ 0.0005); Methylomirabilota by 15%, 47%, and 77%, respectively (p ≤ 0.004); Myxococcota by 54%, 11%, and 35%, respectively (p ≤ 0.002); and Planctomycetota by 78%, 78%, and 56%, respectively (p ≤ 0.04). The proportion of the dominant Actinobacteriota type under the influence of zinc decreased insignificantly (by 2-8%). Analysis of changes in the taxonomic profile of the rhizosphere communities under the influence of the inoculant strains revealed a significant (p ≤ 0.05) promoting effect of Mycolicibacterium sp. Pb113 for the Firmicutes phylum, only compared to non-inoculated plants.  The family-level taxa for which the share in the rhizosphere microbiome of miscanthus was ≥1% are listed in Table S1.
The dominant Actinobacteriota phylum included 31 families, 11 of which were the most abundant (Table S1, Figure 6), making up 82 to 90% of all detected actinobacterial families in the miscanthus rhizosphere. Within the Actinobacteriota, Rubrobacterales was the most abundant order (12%) followed by Gaiellales (10%) and Solirubrobacterales (9%) in the rhizosphere microbiome of M. ×giganteus. Rubrobacter, 67-14 genus, and the genus of uncultured bacteria of the Gaiellales order were the dominant taxa of the Actinobacteria phylum. These taxa prevailed in all treatments, regardless of Zn contamination and microbial inoculation of the rhizosphere soil. A negative correlation was observed between Solirubrobacter and metal contamination of soil (rs = −0.579; p < 0.05). The abundance of this genus decreased significantly in Zn-contaminated soil non-inoculated (by 9%) and inoculated with Mycolicibacterium sp. Pb113 (by 33%) or Chitinophaga sp. Zn19 (by 64%). The effect of the inoculants on the abundance of some taxa of the Actinobacteriota was revealed. Mycolicibacterium sp. Pb113 increased the abundance of Nocardioides but significantly (p < 0.05) decreased the abundance of Propionibacteriaceae, MB-A2-108, and Gaiella both in uncontaminated and Zn-contaminated soil. In contrast, Pb133 increased the abundance of Nocardioides in Zn-contaminated soil. Chitinophaga sp. Zn19 significantly (p < 0.05) decreased the abundance of an unknown taxon of the Gaiellales order.
There was a downward trend in the abundance of the Mycobacteriaceae family under the influence of both inoculation (including with Mycolicibacterium sp. Pb113) and zinc contamination. Moreover, representatives of this taxon were found only in non-inoculated and Pb113-treated uncontaminated soil and were not found at all in Zn-contaminated soil (Table S1).
Seventy-five bacterial families in another dominant phylum, Proteobacteria, were identified. Thirteen of them (Table S1, Figure 6) made the largest contribution to the structure of the miscanthus rhizosphere microbiome. The major class of the phylum was Alphaproteobacteria. Among this class, the dominant position was occupied by the families Sphingomonadaceae (up to 66% of all Alphaproteobacteria families), Beijerinckiaceae (up to 29%), and Xanthobacteraceae (up to 22%). A large proportion also belonged to the Azospirillaceae (up to 20%), Caulobacteraceae (up to 34%), and Rhizobiales Incertae Sedis families. The Nitrosomonadaceae, Comamonadaceae, and Oxalobacteraceae families The family-level taxa for which the share in the rhizosphere microbiome of miscanthus was ≥1% are listed in Table S1.
The dominant Actinobacteriota phylum included 31 families, 11 of which were the most abundant (Table S1, Figure 6), making up 82 to 90% of all detected actinobacterial families in the miscanthus rhizosphere. Within the Actinobacteriota, Rubrobacterales was the most abundant order (12%) followed by Gaiellales (10%) and Solirubrobacterales (9%) in the rhizosphere microbiome of M. ×giganteus. Rubrobacter, 67-14 genus, and the genus of uncultured bacteria of the Gaiellales order were the dominant taxa of the Actinobacteria phylum. These taxa prevailed in all treatments, regardless of Zn contamination and microbial inoculation of the rhizosphere soil. A negative correlation was observed between Solirubrobacter and metal contamination of soil (r s = −0.579; p < 0.05). The abundance of this genus decreased significantly in Zn-contaminated soil non-inoculated (by 9%) and inoculated with Mycolicibacterium sp. Pb113 (by 33%) or Chitinophaga sp. Zn19 (by 64%). The effect of the inoculants on the abundance of some taxa of the Actinobacteriota was revealed. Mycolicibacterium sp. Pb113 increased the abundance of Nocardioides but significantly (p < 0.05) decreased the abundance of Propionibacteriaceae, MB-A2-108, and Gaiella both in uncontaminated and Zn-contaminated soil. In contrast, Pb133 increased the abundance of Nocardioides in Zn-contaminated soil. Chitinophaga sp. Zn19 significantly (p < 0.05) decreased the abundance of an unknown taxon of the Gaiellales order.
There was a downward trend in the abundance of the Mycobacteriaceae family under the influence of both inoculation (including with Mycolicibacterium sp. Pb113) and zinc contamination. Moreover, representatives of this taxon were found only in non-inoculated and Pb113-treated uncontaminated soil and were not found at all in Zn-contaminated soil (Table S1).
Seventy-five bacterial families in another dominant phylum, Proteobacteria, were identified. Thirteen of them (Table S1, Figure 6) made the largest contribution to the structure of the miscanthus rhizosphere microbiome. The major class of the phylum was Alphaproteobacteria. Among this class, the dominant position was occupied by the families Sphingomonadaceae (up to 66% of all Alphaproteobacteria families), Beijerinckiaceae (up to 29%), and Xanthobacteraceae (up to 22%). A large proportion also belonged to the Azospirillaceae (up to 20%), Caulobacteraceae (up to 34%), and Rhizobiales Incertae Sedis families. The Nitrosomonadaceae, Comamonadaceae, and Oxalobacteraceae families were predominant members of the Betaproteobacteria class. The most abundant family of the Gammaproteobacteria was Xanthomonadaceae. The rhizosphere proteobacteria of M. × giganteus were sensitive to the presence of zinc. The abundance of the Sphingomonas, Nordella, and Lysobacter genera significantly increased in Zn-contaminated soil (by 4.9, 2.2, and 2.6-fold, respectively). The Pseudoxanthomonas genus was absent in uncontaminated rhizosphere soil but present in Zn-contaminated soil.   The close and strong positive correlations between Zn contamination and the abundance of the Sphingomonas (r s = 0.63, p < 0.05), Nordella (r s = 0.58, p < 0.05), Lysobacter (r s = 0.70, p < 0.05), and Pseudoxanthomonas (r s = 0.81, p < 0.005) genera were found ( Figure 6). In contrast, the abundance of the Skermanella, Phenylobacterium, and Microvirga genera decreased significantly under the influence of metal contamination of soil. A strong negative correlation (r s = 0.77, p < 0.005) between the abundance of the Microvirga genus and Zn contamination was observed. Inoculation of plants with Mycolicibacterium sp. Pb113 significantly (p < 0.05) increased the abundance of the Skermanella and MND1 genera in uncontaminated soil, whereas inoculation with Chitinophaga sp. Zn19 significantly (p < 0.05) increased the relative abundance of the MND1 and Phenylobacterium genera both in uncontaminated and in Zn-contaminated soil. A close correlation between inoculation and abundance of the MND1 (r s = 0.63, p < 0.05) and Phenylobacterium (r s = 0.68, p < 0.05) genera was found.
The rhizosphere microbiome of M. × giganteus was enriched with members of the Acidobacteriota phylum, among which the Vicinamibacteraceae, an uncultured family of the Vicinamibacterales order, Pyrinomonadaceae, and Subgroup_7 families predominated, making up 80% to 91% of all Acidobacteriota families detected in all treatments (Table S1, Figure 6). Vicinamibacteraceae, Subgroup_7, and RB41 were the predominant genera of the phylum. Zn contamination of soil clearly reduced the abundance of Acidobacteriota in the microbial communities of the miscanthus rhizosphere. A strong negative correlation was found between Zn contamination and the abundance of the Vicinamibacteraceae, Subgroup_7, and RB41 (r s = −0.87, p < 0.001).
A considerable part of the microbial communities of the miscanthus rhizosphere were members of the Bacteroidota phylum. Chitinophagaceae and Microscillaceae were its most abundant families. This phylum was susceptible to heavy metal contamination, which was manifested as a strong increase in the relative abundance of all major taxa. The close positive correlations between Zn contamination and the abundance of Flavisolibacter (r s = 0.66, p < 0.05), Lacibacter (r s = 0.69, p < 0.05), and an uncultured genus of the Microscillaceae family (r s = 0.87, p < 0.001) were found. The inoculation of plants with Mycolicibacterium sp. Pb113 significantly (p < 0.05) increased the abundance of Bacteroides both in uncontaminated and in Zn-contaminated soil. Both inoculants promoted the growth of Ohtaekwangia in Zn-contaminated soil.
The Gemmatimonadota phylum, mainly represented by the Gemmatimonadaceae family (78-96% of all families of the phylum), also made a significant contribution to the formation of the miscanthus rhizobiome in all treatments ( Figure 6). The abundance of the uncultured genus of the Gemmatimonadaceae reached 3.9-8.7% in the rhizosphere microbial communities. The relative abundance of this taxon doubled in the rhizosphere of non-inoculated plants under the influence of Zn; however, plant inoculation reduced this effect.
The Myxococcota phylum also occupied a prominent position in the miscanthus rhizobiome. Both inoculants were found to increase the relative abundance of its representative Haliangium both in uncontaminated and in Zn-contaminated soil.

Effect of Bacterial Inoculants on Zn Content in Plant Biomass
The results of zinc content analysis in the miscanthus biomass are given in Figure 7. It was found that the non-inoculated plants accumulated Zn mainly in the roots (329 mg/kg), more than half as much in the leaves (125 mg/kg) and even less in the rhizomes (46 mg/kg). The TF value was 0.37, indicating phytostabilization as the principal mechanism of soil phytoremediation from Zn. The treatment of miscanthus with the metalresistant PGPR revealed differences in the effect of the inoculants on Zn accumulation in plant biomass. Thus, the strain Mycolicibacterium sp. Pb113 decreased Zn content in the roots but had no significant effect on metal content in the rhizomes and leaves, significantly increasing the TF to 0.63. The strain Chitinophaga sp. Zn19 increased the metal content in the leaves (by 51%) and rhizomes (by 180%) without affecting metal accumulation in the roots. As a result, the TF increased to 0.64.
Microorganisms 2023, 11, x FOR PEER REVIEW 12 of 19 content in the leaves (by 51%) and rhizomes (by 180%) without affecting metal accumulation in the roots. As a result, the TF increased to 0.64.

Discussion
The efficacy of M. × giganteus in the rehabilitation of soils contaminated by heavy metals is well known (e.g., [5,6,46,47]). However, despite the resistance of this plant to heavy metals, its growth and biomass accumulation may be weakened under conditions of heavy pollution [48,49], which may decrease its remediation potential. Under such conditions, the role of indigenous and/or introduced rhizosphere microorganisms able to be associated with the plant is increased. Being resistant to pollutants and able to promote plant growth, such microorganisms can compensate for the stressful effects of pollutants on plants. On the basis of this, plant-assisted phytoremediation is currently a promising approach to soil remediation in situ, as compared with traditional bio-and phytoremediation [12,26,50]. In this regard, the isolation and examination of promising microbial inoculants for remediating plants is of great importance.
During a microbiological study using traditional cultural methods, we obtained isolates Pb113 and Zn19 from the rhizosphere of M. × giganteus grown in soil contaminated with heavy metals [32,34]. The isolates showed plant-growth-promoting properties and were resistant to heavy metals. In this research, the isolates were identified as Mycolicibacterium sp. Pb113 and Chitinophaga sp. Zn19. The subsequent metagenomic analysis of the rhizospheric microbial communities of M. × giganteus confirmed the presence of taxa corresponding to Mycolicibacterium sp. Pb113 (d_Bacteria; p_Actinobacteriota; c_Actinobacteria; o_Corynebacteriales; f_Mycobacteriaceae; g_Mycobacterium) and Chitinophaga sp. Zn19 (d_Bacteria; p_Bacteroidota; c_Bacteroidia; o_Chitinophagales; f_Chitinophagaceae; g_Chitinophaga) in the root zone of this plant (Table S1).
The presence of PGPR traits in a microbial candidate for association with M. × giganteus is important, because in the case of its application, biomass growth in the remediator plant is expected to increase, as is the efficacy of soil cleanup. The strains we had isolated manifested such signs of potential PGPR. On the basis of the obtained data, it can be assumed that these microorganisms would be able to promote the growth of miscanthus owing to nitrogen fixation and to the production of the phytohormone IAA and siderophores. Such properties have been described previously for Chitinophaga spp. and Mycolicibacterium spp. [51][52][53]. However, the use of such inoculants can be effective only if they survive in polluted soil, which should be due to their resistance to the toxic effect of the pollutant. Therefore, the resistance of the strains to heavy metals in the environment was an important selection criterion. In testing the microorganisms, we found that they were resistant to zinc and lead (Table 1), in agreement with the previously reported data on the The chan lants and zinc Despite t at the beginn Therefore, wh weight of the obtained, no s control soil. T bacterium sp. P zome weight, non-inoculate soil, there was a trend to increase the shoot height under the influence of Chitinophaga sp. Zn19 and Mycolicibacterium sp. Pb113, and in Zn-contaminated soil, that was only under the influence of Mycolicibacterium sp. Pb 113. Plant tillering was promoted significantly (62%) by Chitinophaga sp. Zn19 only (Figure 2b). -Zn-contaminated + Mycolicibacterium sp. Pb-113; -Zn-contaminated + Chitinophaga sp. Zn19. Values represent means, and bars represent confidence interval (n ≥ 3); different letters mean significant difference between treatments at p ≤ 0.05.
The changes in miscanthus biomass under the influence of the rhizobacterial inoculants and zinc are illustrated in Figure 3.
Despite the attempts to evenly distribute the biomass of planting material (rhizomes) at the beginning of the experiment, there still was some differences between the pots. Therefore, when the final biomass of the rhizomes was measured, the initial planting weight of the rhizomes was also taken into account (Figure 3a). According to the data obtained, no significant change in wet rhizome weight was found in the uncontaminated control soil. There was only a tendency to decrease biomass under the influence of Mycobacterium sp. Pb 113. In Zn-contaminated soil, there was no significant change in wet rhizome weight, but there was a tendency to decrease the biomass of the rhizomes of the non-inoculated plants. The changes in miscanthus biomass under the influence o lants and zinc are illustrated in Figure 3.
Despite the attempts to evenly distribute the biomass of pla at the beginning of the experiment, there still was some diffe Therefore, when the final biomass of the rhizomes was meas weight of the rhizomes was also taken into account (Figure 3 obtained, no significant change in wet rhizome weight was fou control soil. There was only a tendency to decrease biomass un bacterium sp. Pb 113. In Zn-contaminated soil, there was no sign zome weight, but there was a tendency to decrease the bioma non-inoculated plants.

Discussion
The efficacy of M. × giganteus in the rehabilitation of soils contaminated by heavy metals is well known (e.g., [5,6,46,47]). However, despite the resistance of this plant to heavy metals, its growth and biomass accumulation may be weakened under conditions of heavy pollution [48,49], which may decrease its remediation potential. Under such conditions, the role of indigenous and/or introduced rhizosphere microorganisms able to be associated with the plant is increased. Being resistant to pollutants and able to promote plant growth, such microorganisms can compensate for the stressful effects of pollutants on plants. On the basis of this, plant-assisted phytoremediation is currently a promising approach to soil remediation in situ, as compared with traditional bio-and phytoremediation [12,26,50]. In this regard, the isolation and examination of promising microbial inoculants for remediating plants is of great importance.
During a microbiological study using traditional cultural methods, we obtained isolates Pb113 and Zn19 from the rhizosphere of M. × giganteus grown in soil contaminated with heavy metals [32,34]. The isolates showed plant-growth-promoting properties and were resistant to heavy metals. In this research, the isolates were identified as Mycolicibacterium sp. Pb113 and Chitinophaga sp. Zn19. The subsequent metagenomic analysis of the rhizospheric microbial communities of M. × giganteus confirmed the presence of taxa corresponding to Mycolicibacterium sp. Pb113 (d_Bacteria; p_Actinobacteriota; c_Actinobacteria; o_Corynebacteriales; f_Mycobacteriaceae; g_Mycobacterium) and Chitinophaga sp. Zn19 (d_Bacteria; p_Bacteroidota; c_Bacteroidia; o_Chitinophagales; f_Chitinophagaceae; g_Chitinophaga) in the root zone of this plant (Table S1).
The presence of PGPR traits in a microbial candidate for association with M. × giganteus is important, because in the case of its application, biomass growth in the remediator plant is expected to increase, as is the efficacy of soil cleanup. The strains we had isolated manifested such signs of potential PGPR. On the basis of the obtained data, it can be assumed that these microorganisms would be able to promote the growth of miscanthus owing to nitrogen fixation and to the production of the phytohormone IAA and siderophores. Such properties have been described previously for Chitinophaga spp. and Mycolicibacterium spp. [51][52][53]. However, the use of such inoculants can be effective only if they survive in polluted soil, which should be due to their resistance to the toxic effect of the pollutant. Therefore, the resistance of the strains to heavy metals in the environment was an important selection criterion. In testing the microorganisms, we found that they were resistant to zinc and lead (Table 1), in agreement with the previously reported data on the resistance of bacteria of the same genera Chitinophaga [54] and Mycolicibacterium [55,56] to heavy metals.
On the basis of their properties, Mycolicibacterium sp. Pb113 and Chitinophaga sp. Zn19 have been tested as inoculants in a pot trial to promote the growth and remediation potential of M. × giganteus grown in Zn-contaminated soil.
The results of the experiment showed that the inoculation of miscanthus with Mycolicibacterium sp. Pb113 and Chitinophaga sp. Zn19 in clean soil contributed to increased growth, tillering, and biomass accumulation of M. × giganteus, and the effect of inoculation with Mycolicibacterium sp. Pb113 was greater. The phytotoxicity of the Zn-contaminated soil was manifested as a decrease in plant height, whereas no significant inhibition of M. × giganteus biomass accumulation was observed (Figures 2 and 3). Under pollution conditions, inoculation of miscanthus with Chitinophaga sp. Zn19 resulted in an increase in tillering but not in plant height, as well as a 50% increase in aboveground biomass accumulation. The effect of inoculation with Mycolicibacterium sp. Pb113 was manifested as an increase in the height and tillering of plants, as well as an increase in the accumulation of both aboveground and belowground biomass by 54 and 34%, respectively. The results obtained are comparable to and in some cases surpass the previously reported data on the inoculation of Miscanthus spp. with PGPR belonging to various taxa [26][27][28][29][30][31]. A positive effect of the PGPR strain Pseudomonas koreensis AGB-1 on the growth of M. sinensis in a metal-contaminated mining soil, which was an increase in the plant biomass of 41.6%, was reported by [26]. Fei et al. observed a significant increase in shoot weight in M. × giganteus cv. Amuri treated with Gluconacetobacter diazotrophicus PAL5T LsdB++, Gluconacetobacter johannae UAP-Cf-76, and Variovorax paradoxus JM67 by 15-24%, as compared with the untreated controls [28]. Similar positive effects were observed by Schmidt et al. [27], when M. × giganteus was treated, before planting, with RhizoPlus ® , a commercial formulation containing Bacillus amyloliquefaciens strain FZB24. According to [29], the treatment of M. × giganteus with the PGPR Bacillus altitudinis KP-14 resulted in a significant increase in total shoot and root weight by 77.7% and 55.5%, respectively. Recently, Pidlisnyuk et al. also observed a positive effect of three bacterial strains (Stenotrophomonas maltophilia KP-13, B. altitudinis KP-14, and Pseudomonas fluorescens KP-16), both singly and in combination, on leaf biomass productivity in M. × giganteus [31]. A positive effect of miscanthus inoculation with Herbaspirillum frisingense GSF30T was observed by [30].
In this study, the promotion of miscanthus growth by members of the Mycolicibacterium and Chitinophaga genera was described for the first time. The ability of mycobacteria to promote the growth of plants was reported previously for Dendrobium moschatum [57]; Brassica napus [58]; Triticum sp., Zea mays, Pisum sativum, and Gossypium hirsutum [59]; Sorghum bicolor [60]; a mixture of Festuca arundinacea, F. elata, and F. gigantean [61]; Trifolium repens [62]; and Medicago sativa [39]. The presence of signs of PGPR, including the synthesis of the phytohormone IAA, the dissolution of phosphates, the synthesis of ACC deaminase, in members of Chitinophaga spp. has also been described previously [63,64].
The beneficial effects of microbial inoculation can be driven indirectly through effects on the diversity and composition of the resident plant rhizosphere microbiome [30,65]. Therefore, in our study, we paid special attention to the study of the influence of the pollutant and inoculant strains on the structure of the rhizosphere microbial communities of M.× giganteus.
Along with the characteristics of plant species, soil environment is responsible for the formation of microbial communities in the plant rhizosphere [74,75]. The results of the study showed that the microbial communities of miscanthus were represented by the dominant Proteobacteria, Acidobacteriota, as well as the Bacteroidota and Gemmatimonadota phyla. The features of the rhizobiomes of miscanthus grown in the soil type studied was the predominance of Rubrobacter, 67-14 and uncultivated bacteria of the order Gaiellales (Actinobacteriota), uncultivated bacteria of the Gemmatimonadaceae family (Gemmatimonadota), as well as Vicinamibacteraceae and uncultivated bacteria of the Vicinamibacteraceae family (Acidobacteriota). The high enrichment of the miscanthus rhizobiomes by members of these taxa was maintained regardless of the treatment used (Zn contamination or PGPR inoculation). Of interest, the identified taxa included Rubrobacter, whose members differ in their tolerance to extreme environmental conditions such as radiation and elevated temperature [76]. Some authors have noted a correlation between the presence of Rubrobacter members in the environment and metal contamination [77,78]. However, in our study, the abundant of this genus was found both in Zn-contaminated and in uncontaminated soil.
Changes in the composition and structure of the soil microbial community of the M. × giganteus rhizosphere under the influence of metals has also been reported in a number of works [18,21,22,24,70]. Various changes were noted in the taxonomic profile of the miscanthus rhizobiome, which may be due to the different soil types used in research, the duration of plant cultivation, and other factors. According to the data obtained, soil contamination with Zn had a pronounced influence on the formation of the rhizosphere microbiome of miscanthus ( Figure 4). This was manifested mainly as a decrease in the abundance of Acidobacteriota, as well as Methylomirabilota, Myxococcota, and Planctomycetota, and a simultaneous increase in the abundance of Proteobacteria and Bacteroidota. The indicator taxa whose abundance decreased most markedly under the influence of soil contamination with Zn were RB41, Subgroup_7 and Vicinamibacteraceae (Acidobacteriota), Gemmatimonas (Gemmatimonadota), bacteriap25 (Myxococcota), WD2101_soil_group (Planctomycetota), and Microvirga (Proteobacteria). Yet, the number of other taxa clearly increased-an uncultured genus of the Microtrichales order (Actinobacteriota), Nocardioides (Actinobacteriota), Flavisolibacter, Lacibacter, Ohtaekwangia, and an uncultured genus of the Microscillaceae family (Bacteroidota), Nordella, Lysobacter, Pseudoxanthomonas, and especially Sphingomonas (Proteobacteria). Under the influence of the inoculants in polluted soil, the number of Bacillus (Firmicutes) increased. Many of the identified taxa can take part in the remediation of soil from Zn. According to the literature, such properties are possessed by members of Bacillus [79], Sphingomonas [70,80], and Pseudoxanthomonas [81].
Interestingly, the taxa corresponding to the inoculants used reacted differently to soil contamination: the abundance of the Mycobacteriaceae family decreased, and the abundance of the Chitinophagaceae family increased significantly in response to Zn treatment of soil (Table S1, Figure 6). In this connection, we assume that the effect of inoculants on the growth and phytoremediation efficiency of miscanthus could be direct (probably for Chitinophaga sp. Zn19) and indirect (mainly for Mycolicibacterium sp. Pb113) through the impact on other rhizospheric microorganisms.
Inoculation of M. ×giganteus with the PGPR strains used improved the phytoremediation ability of the plant. All the rhizobacteria tested increased Zn uptake by miscanthus, enhancing the bioaccumulation and translocation of the metal to the aboveground organs, as evidenced by a rising the TF more than 70%. Chitinophaga sp. strain Zn19 was characterized by the highest efficiency, increasing the translocation of the metal into plant leaves by 180%, as compared with the non-inoculated control. Thus, the use of bacterization of plants allowed us to increase the removal of Zn from roots to shoots. We speculate that under conditions of longer cultivation of miscanthus, this technique will contribute to improving the cleanup of soil contaminated with Zn. The bacteria used in this study, after being tested for virulence and pathogenicity, can be recommended for use to improve the efficiency of the phytoremediation of Zn-contaminated soil.

Conclusions
A positive effect of inoculation with two heavy-metal-resistant PGPR, Chitinophaga sp. Zn19 and Mycolicibacterium sp. Pb113, on the growth, rhizosphere microbiome, and remediation ability of Miscanthus × giganteus was revealed. Mycolicibacterium sp. Pb113 significantly promoted the accumulation of the aboveground and belowground biomass of miscanthus in uncontaminated as well as in Zn-contaminated soil. Chitinophaga sp. Zn19 enhanced the accumulation of aboveground biomass in Zn-contaminated soil. Mycolicibacterium sp. Pb113 reduced Zn accumulation in roots and had no effect on metal accumulation in rhizomes and in leaves. Chitinophaga sp. Zn19 increased the Zn content in leaves and rhizomes. The inoculation of plants with heavy-metal-resistant bacteria increased the translocation of Zn from roots to shoots. Chitinophaga sp. Zn19 showed the highest translocation of Zn into plant leaves, as compared with the non-inoculated control. Evaluation of the impact of Zn contamination and inoculation of plants with the bacteria studied on the structure and diversity of the rhizosphere microbial communities showed differences in the formation of the miscanthus rhizosphere microbiomes, which was affected by Zn rather than by the inoculants. Zn contamination of soil reduced the abundance of Acidobacteriota, Methylomirabilota, Myxococcota, and Planctomycetota and simultaneously increased the abundance of Proteobacteria and Bacteroidota. Bacterial taxa that are potentially involved in the promotion of plant growth and in assisted phytoremediation were identified. The effect of miscanthus inoculation with Chitinophaga spp. and Mycolicibacterium spp. was shown for the first time. On the basis of the obtained data, the bacterial strains studied may be recommended for use to improve the efficacy of M. × giganteus remediation of Zn-contaminated soil.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/microorganisms11061516/s1, Table S1: Taxonomic structure of the rhizosphere microbial communities of non-inoculated and PGPR-inoculated Miscanthus × giganteus, cultivated in uncontaminated and Zn-contaminated soil, %. Data Availability Statement: The Pb113 and Zn19 16S rRNA gene sequences as well as the raw sequence reads can be accessed via NCBI archives with IDs OQ680140, OQ680520, and PRJNA973256, respectively.