New Plant Growth-Promoting, Chromium-Detoxifying Microbacterium Species Isolated From a Tannery Wastewater: Performance and Genomic Insights

Hexavalent chromium [Cr(VI)], widely generated by tannery activities, is considered among the most toxic substances and causes a serious damage for the environment and for human health. Interestingly, some microorganisms have a potential of bioremediation of chromium-contaminated wastewaters and soils through the reduction of Cr(VI) (soluble and harmful form) into Cr(III) (stable and non-toxic form). Here, we present the full genome sequence of a novel heavy-metal-resistant, plant growth-promoting bacterium (PGPB), Microbacterium metallidurans TL13, which was isolated from a Tunisian leather industry. The strain TL13 was resistant to many heavy metals, such as chromium, copper, nickel, cobalt, and arsenic. The 50% TL13 growth inhibitory concentration (IC50) values of HgCl2, CoCl2, K2Cr2O7, CuSO4, NiCl2, FeSO4, and Na2HAsO4 are 368, 445, 676, 1,590, 1,680, 4,403, and 7,007 mg/L, respectively, with the following toxicity order: HgCl2 > CoCl2 > K2Cr2O7 > CuSO4 > NiCl2 > FeSO4 > Na2HAsO4. This new strain was also able to promote the growth of the hybrid tomato (Elika F1) under chromium metal stress. Its whole genome sequence length was estimated to be 3,587,460 bp (3,393 coding sequences) with a G + C content of 70.7%. Functional annotation of the genome of TL13 revealed the presence of open reading frames (ORFs) involved in adaptation to metal stress, such as the chromate transport protein, cobalt–zinc–cadmium resistance protein, copper resistance protein, copper responsive transcriptional regulator, multidrug resistance transporters, arsenical resistance operon repressor, arsenate reductase, arsenic resistance protein, mercuric resistance operon regulatory protein, mercuric ion reductase, and organomercurial lyase. Moreover, genes for the production of glutathione peroxidase, catalase, superoxide dismutase, and thioredoxin reductase, which confer a higher tolerance to oxidative/metal stresses, were identified in TL13 genome. In addition, genes for heat shock tolerance, cold shock tolerance, glycine-betaine production, mineral phosphate solubilization, ammonia assimilation, siderophores, exopolysaccharides, polyketides, and lytic enzymes (cellulase, chitinase, and proteases) production that enable bacteria to survive biotic/abiotic stress and to promote plant growth and health were also revealed. Based on genome analysis and experimental approaches, strain TL13 appears to have evolved from various metabolic strategies and could play a role in ensuring sustainable environmental and agricultural systems.


INTRODUCTION
The leather industry plays an important role in the world's economy. This industry uses a raw material that is originally a livestock coproduct. Leather making consists of converting putrescible hides and skins into a more resistant commercial leather (Fang et al., 2017). In fact, this process comprises three main operations: (i) beamhouse operation, which consists of several steps aimed to eliminate all components other than collagen; (ii) tanning operation, which provides resistant leather using chromium; and (iii) finishing operation, which consists of giving the treated leather its last aspect ready for commercialization. During the process of leather making, several baths containing chemicals harmful to the environment are used in order to treat hides and skins. Indeed, spilled water has a significant pollutant load (Thanikaivelan et al., 2004). Around 30 m 3 of effluents is generated for every ton of hides and skins, which have significant impacts on environments and human health due to the presence of heavy metals especially chromium (Suthanthararajan et al., 2004). The toxic form of this metal [Cr(VI)] seems to be the cause of the contamination of aquatic and terrestrial ecosystems and also the cause of several diseases such as cancer, particularly lung cancer (Pradhan et al., 2016).
In recent years, several bacterial whole genomes were sequenced in order to identify loci involved in metal resistance, particularly chromium reduction and resistance. Genomic annotations indicated that chrR and yieF are implicated in the mechanism of chromate reduction. In fact, some bacterial strains, like E. coli, use YieF reductase to reduce Cr(VI) into Cr(III). Other species, like Pseudomonas putida, are known for their chromium reduction through the ChrR reductase (Ramirez-Diaz et al., 2008;Learman et al., 2011;Ahemad and Kibret, 2014;Henson et al., 2015;Baldiris et al., 2018). Otherwise, M. laevaniformans strain OR221 was able to resist some heavy metals, but the annotation of its genome after sequencing suggested the absence of both ChrR and YieF (Brown et al., 2012). In this paper, we describe the identification of a novel heavymetal-resistant, PGPB M. metallidurans TL13, isolated from a tannery effluent. TL13 could prove to be a suitable candidate for the bioremediation of heavy-metal-contaminated soils and wastewaters and also could be used as a biofertilizer due to its traits. The complete genome sequencing and annotation of TL13 were conducted, and experimental data were collected, in order to better understand its mechanisms of chromium resistance and reduction and genes involved in stress resistance and PGP activities.

Reagents and Chemicals
All chemicals and reagents used in this study are of pure analytical grade and available commercially.

Isolation and Characterization of a Novel Heavy-Metal-Resistant Bacterium
In this study, a heavy-metal-resistant bacterium strain TL13 was isolated from tannery (TMM) wastewater using tryptic soy agar (TSA). Water content, total organic carbon (TOC), organic content, and total heavy metal content from the sludge generated in TMM wastewater treatment plants were determined by gravimetry, calcination, colorimetry, and inductively coupled plasma atomic emission spectroscopy (ICP-OES) according to the standards ISO 12880: 2000, Rodier, NF ISO 14235: 1998and ISO 1188: 2007, respectively (Zeng et al., 2016. Analysis of bacterial resistance to heavy metals was carried out on TSA medium supplemented with increasing concentrations (50-2,500 mg/L) of iron sulfate (FeSO 4 ), copper sulfate (CuSO 4 ), cobalt chloride (CoCl 2 ), disodium arsenate (Na 2 HAsO 4 ), potassium dichromate (K 2 Cr 2 O 7 ), mercury chloride (HgCl 2 ), and nickel chloride (NiCl 2 ). The spot inoculation method was used to involve in triplicate 10 µl of bacterial culture suspension (10 8 CFU/mL) (Fu et al., 2016). Bacterial cultures were incubated for 24 h at 30 • C to determine minimum inhibitory concentrations (MIC) (Davis, 2005). A positive control was used by inoculating bacterial suspension on agar plates without metal. The half maximal inhibitory concentration (IC 50 ) was determined using experiments conducted in tryptic soy broth (TSB) medium at 30 • C with shaking at 120 rpm (Fassy et al., 2017). Bacterial growth was measured after 24 h on a UV-visible (UV-Vis) spectrophotometer at 600 nm. All experiments (solid and liquid) were performed under alkaline conditions (pH 9). Metal reduction was determined by inductively coupled plasma mass spectrometry. Bacterial cell free extracts were recovered after a 10 min of centrifugation at 10,000 rpm and then filtered using 0.22 µm Whatman filter paper.

Plant Growth-Promoting Activities
In vitro Screening of TL13 for Its Plant Growth-Promoting Activities The screening of TL13 strain for various in vitro plant growthpromoting activities was performed by adopting standard methods (Hassen et al., 2018). Inorganic phosphate-solubilizing activity and growth on nitrogen free medium was estimated by the method of Pikovskaya (1948) and Jensen (1942), respectively. To detect the ability of TL13 strain to produce siderophore, chrome azurol sulfonate (CAS) agar solid medium was used (Alexander and Zuberer, 1991). Cellulase activity was tested on minimal medium amended with 1% carboxymethyl cellulose (CMC) (Souii et al., 2018). The protease production was screened by skim milk agar . The ability of TL13 to produce indole-3-acetic acid (IAA) and exopolysaccharides (EPS) was assessed as indicated by Penrose and Glick (2003) and Naseem et al. (2018), respectively.

In vivo Plant Growth Promotion Assays
In vivo plant growth promotion and tomato seed germination assays were performed as described by Passari et al. (2019a) with slight modification. Elika F1 tomato seeds were surface sterilized with 1% sodium hypochlorite solution for 20 min and subsequently rinsed three times with sterile distilled water. Disinfected Elika F1 tomato seeds were immersed for 2 h in a TL13 bacterial suspension (10 8 CFU/mL). Seeds immersed in sterile distilled water served as a control. A sterile Whatman paper was soaked in a chromium solution at different concentrations (0, 100, and 500 mg/L) and deposited in Petri dishes. Seeds were then gently deposited on Whatman filter paper under appropriate conditions, incubated in darkness at 30 • C for 8 days and irrigated every day with sterile distilled water. Germination percentages were calculated by dividing total germinated seeds to total number of seeds that were planted (Passari et al., 2019a,b). The experiment was repeated three times using 20 Elika F1 tomato seeds per germination condition. Different parameters, including shoot length, shoot fresh, and dry weights, were measured.

Genome Sequencing, Assembly, and Annotation of TL13 Strain
Genomic DNA extraction of TL13 strain was performed by sodium dodecyl sulfate-proteinase K treatment . Purified genomic DNA was sequenced on an Illumina MiSeq platform (MRDNA, USA) yielding 2 × 250 bp pairedend reads. The Prinseq-lite software was used for trimming the 4,320,030 paired reads by discarding low-quality sequences with a score lower than 23 . Genome assembly was then performed using the SPAdes algorithm (Bankevich et al., 2012). Genome annotation was done using the Integrated Microbial Genomes/Expert Review (IMG/ER) database (Markowitz et al., 2009) and the Rapid Annotations using Subsystems Technology (RAST) server (Aziz et al., 2008). Annotated protein sequences were downloaded from the National Center for Biotechnology Information (NCBI) (Galperin et al., 2015) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) (Ogata et al., 1999) databases.

Phylogenetic Analysis
The 16S ribosomal DNA (rDNA) gene sequence was annotated within the genome and compared to sequences of bacterial strains from the NCBI and EzBioCloud databases (Kim et al., 2012). The phylogenetic tree was inferred by the neighborjoining method using the Molecular Evolutionary Genetics Analysis (MEGA 6) software (Kumar et al., 2016). A 95-96% average nucleotide identity (ANI) value was used as a genomic measure for prokaryotic species delineation (Richter et al., 2016) and was measured using the JSpecies Web Server (JSpeciesWS) available at http://jspecies.ribohost.com/ jspeciesws. A circular genome map was drawn used CGView software (Grant and Stothard, 2008).

Structural Analysis
The structures of enzymes involved in chromate reduction were predicted by the online structure prediction tool, Iterative Threading ASSEmbly Refinement server (I-TASSER) (Yang and Zhang, 2015;Ouertani et al., 2018). The confidence score (C score) was used to identify the best generated model, and MolProbity was applied to evaluate the refined model coordinates  (Davis et al., 2007). Enzyme models were assessed using COACH software (Yang et al., 2013). A final analysis with CCP4mg molecular-graphics software was performed to validate the quality of the structures (McNicholas et al., 2011).

Statistical Analysis
Experimental results were expressed as the mean value of three independent replicates ± the standard deviation (SD). Data were analyzed using one-way ANOVA with post-hoc Duncan's tests (SPSS 16.0 for Windows, SPSS Inc., USA). Statistical significance was determined as p < 0.05.

Nucleotide Sequence Accession Number
The Whole Genome Shotgun project was deposited at DDBJ/ENA/GenBank under the accession SZZQ00000000. The version reported in this work is SZZQ01000000 (https:// www.ncbi.nlm.nih.gov/nuccore/SZZQ00000000).

Characterization of a Heavy-Metal-Resistant, Plant Growth-Promoting Actinobacterium Isolated From a Tannery Wastewater
The physicochemical composition of the sludge collected from TMM industry is shown in Table 1. The content of chromium in sludge exceeds the thresholds for agriculture, according to national and international standards (Jakov, 2005). Qualitative analyses indicated that TL13 strain has the capacity to grow in the presence of K 2 Cr 2 O 7 concentrations that could exceed 1,000 mg/L ( Figure 1A). The bacterial growth was followed in the presence of K 2 Cr 2 O 7 with variable concentrations ranging from 50 to 2,500 mg/L on TSB medium. As shown in Figure 1B, TL13 strain tolerated more than 1,000 mg/L of K 2 Cr 2 O 7 . The TL13 IC 50 of K 2 Cr 2 O 7 is equal to 676 mg/L ( Figure 1C). ICP-MS analysis revealed that TL13 strain was able to reduce 71.68% of chromium [Cr(VI)] with initial concentration of 500 mg/L K 2 Cr 2 O 7 . Heavy metal resistance experiments indicated also that TL13 strain has the capacity to grow under high concentrations of FeSO 4 and Na 2 HAsO 4 (up to 2,500 mg/L), NiCl 2 and CuSO 4 (up to 1,000 mg/L), and CoCl 2 and HgCl 2 (up to 500 mg/L). IC 50 values of HgCl 2 , CoCl 2 , K 2 Cr 2 O 7 , CuSO 4 , NiCl 2 , FeSO 4 , and Na 2 HAsO 4 are 368, 445, 676, 1,590, 1,680, 4,403, and 7,007 mg/L, respectively ( Table 2).
The in vitro study of plant growth-promoting activities showed that TL13 strain was able to fix atmospheric nitrogen and to solubilize phosphate. The strain TL13 has also the capacity to produce siderophores, EPS, and cellulase as materialized by the presence of transparent halo around the colony in the corresponding culture media. Furthermore, the development of a pink color following the addition of Salkowski reagent to the culture supernatant showed the capacity of TL13 to produce IAA ( Figure S1). The results of antibiograms ( Figure S2, Table S1) showed the resistance of TL13 strain to cefotaxime, ceftazidime (antibiotics of β-lactam group), and norfloxacin (antibiotic of fluoroquinolone group). M. metallidurans TL13 was sensitive against all other tested antibiotics. Tomato in vivo experiments revealed positive effect of the TL13 strain on seeds germination under chromium stress. In fact, with simple visual inspection, the toxic effect of high levels of chromium and the promoting effect of TL13 on seeds germination were well distinguishable (Figure 2A). Tolerance to higher chromium levels was observed with TL13-treated seeds compared to control seeds since a significant increase in shoot fresh and dry weights and shoot length were recorded (Figures 2B-D).

Phylogenetic Analysis and Genomic Insights Inferred From TL13 Genome Sequence
Based on the proposed rule of using 16S rDNA sequence similarity threshold value of 98.65% for species delineation , TL13 and Microbacterium proteolyticum (RZ36) should belong to the same species as their 16S rDNA sequences exhibited an identity of 99.3%. However, this taxonomic result was not corroborated by the phylogenetic tree (Figure 3)   for which the name M. metallidurans is proposed. Figure 4 shows the graphical representation map of the genome of TL13 strain compared with the genomes of M. enclense NIO-1002 and M. hominis TPW29. The complete genome of M. metallidurans TL13 measured 3,587,460 bp with 70.70% G + C content (Table 3). De novo assembly of TL13 strain genome sequencing data is represented by 5 contigs and 3,393 coding sequences.
Based on the annotation of the RAST pipeline, TL13 genes were classified into 380 subsystems revealing its high metabolic diversity. Analysis of the predicted protein sequences showed that most of the annotated genes were involved in carbohydrates (399), amino acids and their derivatives (354), cofactors, vitamins, prosthetic groups, pigments (214), protein metabolism (217), fatty acids, lipids and isoprenoids (129), cell membrane and capsule (91), DNA (88) and RNA (85) metabolism and response to stresses (74), virulence, defense (55), respiration (55), metabolism of aromatic compounds (47), and phosphorus metabolism (40). Analysis of M. metallidurans TL13 genome revealed the presence of several genes contributing directly or indirectly to PGP activities (Table 4). Indeed, the annotation study showed the presence of genes involved in the solubilization of inorganic phosphate, assimilation of ammonia, production of levan (EPS), synthesis of IAA, siderophores and polyketides, as well as genes encoding the key cell-wall-degrading enzymes ( Table 4).
The genome of M. metallidurans TL13 contains 32 genes related to heavy metal resistance. Based on genomic analyses, enzymatic and non-enzymatic mechanisms could be proposed to explain hexavalent chromium resistance and removal by this strain (Figure 5).

DISCUSSION
Bacteria of the genus Microbacterium (phylum of Actinobacteria) include more than 110 cultivable species (http://www.bacterio. net/microbacterium.html) isolated from aquatic and terrestrial ecosystems and from food and clinical samples. Figure S3 illustrates the microbial world (according to the Gold and DSMZ databases) with each point representing a type of bacterial strain, and if applicable, the associated ongoing sequencing projects (colored red). As shown in Figure S3, many taxonomic groups still need to be explored at the genomic level and, in particular, the Microbacterium group for which only few genomes are available (https://microbial-earth.namesforlife.com/). Several Microbacterium members have been isolated from extreme environments (Li et al., 2008) or contaminated with organic pollutants (Schippers et al., 2005) and metals such as uranium (Reardon et al., 2004;Gallois et al., 2018), plutonium and iron(III) (John et al., 2001), arsenic (Cai et al., 2009), and chromium (Pattanapipitpaisal et al., 2001;Humphries et al., 2005).
There are only few studies reported in the literature on the predominance and ecological importance of bacterial populations of the genus Microbacterium as PGPB (Alves et al., 2015;Gao et al., 2017). Although such bacteria directly stimulate FIGURE 3 | Phylogenetic tree highlighting the position of M. metallidurans TL13 among related taxa within the genus Microbacterium based on 16S rDNA gene sequences. Evolutionary distances were calculated using the method of maximum composite likelihood, and the topology was inferred using the neighbor-joining method using MEGA 7. Numbers on the nodes present % bootstrap values based on 500 replicates. Scale bar represents 0.02 substitutions per site. The 16S rDNA gene sequence of Actinopolyspora xinjiangensis was arbitrarily chosen as the outgroup to define the root of the tree.
FIGURE 4 | Circular representation of M. metallidurans TL13 genome generated by CG viewer. The innermost rings depict GC content (Black) and GC Skew (purple/green) followed by concentric rings of query sequences colored according to BLAST identity. The outermost rings depict genomes of the following microbes M. enclense NIO-1002 (pink) and M. hominis TPW29 (green). plant growth by increasing soil nutrient uptake, inducing and producing plant growth regulators, and activating induced resistance mechanisms in plants, special attention has recently been given to bioremediation of organic and inorganic pollutants by PGPB (Bishnoi, 2015). Therefore, the present study aimed to characterize a new heavy-metal-resistant PGPB isolated from a Tunisian tannery wastewater for potential application as a green and sustainable tool for decreasing accumulation of metals in industrial sludges and wastewaters and in plant rhizosphere and tissues.
The actinobacterium TL13 exhibited resistance to all of the tested heavy metal ions. The half inhibitory concentration values of mercury, cobalt, chromium, copper, nickel, iron, and arsenic were 368, 445, 676, 1,590, 1,680, 4,403, and 7,007 mg/L, respectively, and their toxic order was HgCl 2 > CoCl 2 > The whole genome of TL13 strain was sequenced and annotated in order to provide information on genomic elements related to heavy metal resistance and plant growth-promoting traits. The genome consists of 3,587,460 bp with an average GC content of 70.7% and a total of 52 genes encoded RNAs. Of the 3,393 genes predicted, 2,425 were identified as genes encoding proteins (genes had a specific function) and 968 genes (28.53%) were classified as hypothetical proteins (genes without function prediction). The ANI values between the strain TL13 and the  closely related strains were 76.5-82.7%, indicating that TL13 strain is a novel species of the genus Microbacterium. Functional annotation of TL13 genome revealed the presence of genes encoding proteins involved in resistance to metal stress, such as cobalt-zinc-cadmium resistance protein, copper resistance protein, copper responsive transcriptional regulator, arsenical resistance operon repressor, arsenate reductase, arsenic resistance protein, mercuric resistance operon regulatory protein, mercuric ion reductase, and organomercurial lyase. The TL13 genome analysis revealed also the presence of different potential genes involved in chromium reduction such as chromate transporter, superoxide dismutase, glutathione peroxidase, and the thioredoxin reductase. The chromate transporter protein (ChrA) has been reported to play a crucial role in efflux of cytoplasmic chromate (Alvarez et al., 1999;Ahemad and Kibret, 2014). The participation of bacterial enzymes in the defense against oxidative stress induced by chromate represents an additional mechanism of chromate resistance (McCord and Fridovich, 1988;Ackerley et al., 2006;Ramirez-Diaz et al., 2008). The TL13 strain has the potential to have a powerful enzyme system to combat oxidative stress involving catalase (EC 1.11.1.6), glutathione peroxidase, and superoxide dismutase [Mn] (EC 1.15.1.1). The theriodixin reductase, an Mg 2+dependent enzyme, is also involved in the reduction in Cr(VI) (Li and Krumholz, 2009;Collet and Messens, 2010). The two cysteines of the catalytic motif (Cys64, A65, R66, Cys67), located at the amino end of the two-helix, are the key players involved in the reduction of oxidized substrates. Chen et al. (2014) reported that the entire operon of thioredoxin of Streptomyces violaceoruber strain LZ-26-1 was upregulated under the stress of Cr(VI). This result suggests that TL13 strain can use thioredoxin pathway to reduce Cr(VI). Various oxidoreductases have been shown to reduce Cr(VI) such as ChrR from E. coli and Pseudomonas and NfsA from E. coli (Ackerley et al., 2004). Genes coding for Cr(VI)-reducing oxidorecductase were not detected on TL13 genome. This result probably shed light on novel Cr(VI) reductases within the thioredoxin pathway. M. metallidurans TL13 was able to produce levan based on genomic signature (presence of the gene encoding levan sucrase) linked to experimental evidence (mucoid phenotype on agar plate). The anionic property of the EPS enables it to trap the positively charged chromium by electrostatic interaction and hence playing a key role as biosorbents for metal remediation (Harish et al., 2012;Gupta and Diwan, 2017).
TL13 genome encodes an efficient framework for iron acquisition mediated by a siderophore-dependent pathway as similar to that of Cupriavidus metallidurans CH34 (Tables S2,  S3). Iron is essential for vital processes including DNA, RNA, and protein synthesis, electron transport, cell respiration and proliferation, as well as gene expression regulation. Iron has the ability to easily acquire and lose electrons from the iron form Fe 2+ to the iron form Fe 3+ , and vice versa. This rather unique property gives it an essential role in oxidation and reduction processes. The chemical properties of iron make it potentially toxic. When iron is in limited supply, TL13 can produce and secrete siderophores capable of chelating ferric Fe 3+ with very high affinity. The siderophore-Fe 3+ complexes are then recovered by the bacterium using specific membrane transporters. The dissociation of iron from its chelator generally requires a reduction in Fe 3+ to Fe 2+ and a structural modification of the siderophore, making it less affine for the metal (Payne et al., 2016;Li and Ma, 2017).
Besides these genes conferring resistance to metal and oxidative stresses identified in TL13 genome, the RAST analysis showed the presence of 10 genes involved in osmoadaptation, in particular in the biosynthesis of betaine and choline osmolytes, 14 proteins (operon dnaK) to fight against thermal stress, and 4 against cold (proteins from the CspA family). Analysis of the genome of M. metallidurans TL13 revealed also the presence of 118 genes involved in different pathways of central carbohydrate metabolism like the tricarboxylic acid cycle and the pentose phosphate pathway. It also indicated the presence of 43 proteins involved in anaerobic fermentation processes. The genome of M. metallidurans TL13 codes for exopolyphosphatase (EC 3.6.1.11), inorganic manganese-dependent pyrophosphatase (EC 3.6.1.1), and low affinity inorganic phosphate carriers, which play a crucial role in phosphorus cycle by solubilizing its inorganic form that is fixed and precipitated in the soil (Al-Kaisi et al., 2013;Bhattacharyya et al., 2017). Moreover, the genome of M. metallidurans TL13 encodes all the genes necessary for ammonia assimilation and plant auxin biosynthesis (Spaepen and Vanderleyden, 2011). Several Microbacterium strains isolated from rhizospheric and non-rhizospheric environments are able to fix nitrogen, as confirmed by acetylene-reducing activity and PCR amplification of nifH gene (Gtari et al., 2011). Whereas, some nitrogen-fixing bacteria are able to produce nitrogenases with iron as the only metal cofactor, other bacterial strains incorporate molybdenum or vanadium into the cofactor as well (Johnstone and Nolan, 2015). For the strain TL13, a search for nitrogenase systems merits further investigation at the genome and physiological levels.
Besides these direct PGP traits, several genes involved in the biocontrol of phytopathogens exist in the genome of M. metallidurans TL13, including genes coding for cell-walldegrading enzymes (cellulose, β-hexosaminidase, and proteases) (Dutta and Thakur, 2017) and genes coding enzymes responsible for the biosynthesis of polyketides, a class of secondary metabolites with antimicrobial roles (Khanna et al., 2019). M. metallidurans TL13 showed resistance to β-lactams and fluoroquinolones antibiotics leading to improved chance of survival in a competitive rhizospheric soil (Bhattacharyya et al., 2017). All genomic plant growth-promoting features make TL13 strain an excellent actinobacterium for utilization in sustainable agriculture (Table 5). Overall, these PGP results were in line with those reported for Microbacterium sp. strain 3J1 (Manzanera et al., 2015), Microbacterium sp. Yaish 1 (Jana et al., 2017), and Microbacterium hydrothermale BPSAC84 (Passari et al., 2019b).

CONCLUSION
This study reports the characterization of a heavy-metalresistant, plant growth-promoting actinobacterium, M. metallidurans TL13, isolated from a tannery effluent. Genome analyses, as well as experimental studies, indicated that M. metallidurans TL13 could promote plant growth, especially by solubilizing inorganic phosphate, synthesizing siderophores, and producing IAA, hydrolytic enzymes, and EPS. Besides, genome annotation revealed the presence of multiple genetic determinants related to the resistance to heavy metals, particularly chromium, arsenic, and mercury.
To the best of our knowledge, this is the first report providing new insights into the molecular mechanisms underlying the enhancement of plant growth and the resistance to chromium and other heavy metals in a new species of Microbacterium. Further comparative genomic analysis and direct redox experimentation will provide additional information about the variance and pathways involved in the multimetal resistance and rhizosphere bioremediation by the actinobacterium M. metallidurans TL13 that is expected to facilitate environmental and agricultural applications in heavy metal-polluted sites.

DATA AVAILABILITY STATEMENT
The Whole Genome Shotgun project has been deposited at DDBJ/ENA/GenBank under the accession SZZQ00000000. The version reported in this work is SZZQ01000000.