Nano-metals forming bacteria in Egypt. I. Synthesis, characterization and effect on some phytopathogenic bacteria in vitro

Bacterial metal reducers were isolated from water samples collected from harsh condition locations in Egypt. Four selected isolates were identified as Enterococcus thailandicus, Pseudomonas putida, Marinobacter hydrocarbonoclasticus, and P. geniculata for Copper (Cu), Iron (Fe), Cobalt (Co) and Zinc (Zn) Nanoparticles (NPs) production sequentially. Nitrate reductase enzyme was assayed for bacterial isolates which demonstrated that P. putida, and M. hydrocarbonoclasticus have the maximum enzyme production. The produced NPs were characterized by using XRD, TEM, UV–VIS spectroscopy. Magnetic properties for all selected metals NPs were measured using Vibrating Sample Magnetometer (VSM) and demonstrated that FeNPs recorded the highest magnetization value. The antibacterial activity of selected metals NPs was tested against some phytopathogenic bacteria causing the following diseases: soft rot (Pectobacterium carotovorum, Enterobacter cloacae), blackleg (Pectobacterium atrosepticum and Dickeya solani), brown rot (Ralstonia solanacearum), fire blight (Erwinia amylovora) and crown gall (Agrobacterium tumefaciens). All metals NPs showed an antagonistic effect against the tested isolates, particularly, FeNPs showed the highest antibacterial activity followed by CuNPs, and ZnNPs. Due to the small size, high reactivity, and large surface area of biologically synthesized NPs, they are used as a good disinfector, and can be considered as a new and alternative approach to traditional disease management methods.

www.nature.com/scientificreports/ P. geniculata KU550146. The phylogenetic trees of selected isolates were related to isolates from the gene bank (Fig. 3).
Production of metals NPs. The reduction of metals ions to metals NPs could be optically appreciated after incubation of the bacterial culture by color changes of the medium from yellow to dark brown/black as illustrated in S3.
Characterization of metals NPs. UV-VIS spectroscopic analysis. The UV-VIS spectroscopic analysis of synthesized metals NPs was shown in (Fig. 4). UV-VIS spectroscopic analysis of FeNPs and ZnNPs (Fig. 4A,D) showed the absorption maximum peak at 233 nm and around 230-250 nm respectively. The maximum absorption peaks of CuNPs and CoNPs appeared at around 569 nm (Fig. 4B), and 530 nm (Fig. 4C) consecutively.
Transmission electron microscope (TEM) analysis. The Ultrastructure study of the intracellular biosynthesis of metals NPs was examined using TEM. Particles with higher electron density appear darker in the TEM negative film. Figure 5 shows that the synthesis of FeNPs and ZnNPs is in both periplasmic space and the cytoplasm of the bacterial cell of P. putida (Fig. 5A) and P. geniculata ( Fig. 5D) respectively, while synthesis of CuNPs and CoNPs, was in the cytoplasm of cells of E. thailandicus (Fig. 5B) and M. hydrocarbonoclasticus (Fig. 5C) respectively. The spherical morphology and size of metals NPs are shown in Fig. 6. Synthesized FeNPs, CuNPs, CoNPs, and ZnNPs were in the range of 1-4, 8-14, 9-20, and 4-13 nm, respectively. X-ray powder diffraction (XRD) analysis. The crystalline nature, quality, and crystallographic identity of the examined NPs in addition to the phase purity were determined by the XRD spectrum over a wide range of Bragg angles 10° ≤ 2θ ≤ 80. XRD pattern was characterized by the interplanar d-spacing/2θ degree and the different intensities of the strongest peaks. The X-ray diffractograms of metals NPs synthesized by the selected bacterial isolates are illustrated in Fig. 7. The XRD pattern of FeNPs in Fig. 7A reveales that a peak of the highest intensity occurs at 45.50°. Tow main characteristic diffraction peaks for CuNPs were observed (Fig. 7B) at around 2θ = 42°, 50°. The XRD pattern (Fig. 7C) of CoNPs shows the diffraction peaks at 43.2°, 54.5°, and 77.9°, also XRD pattern in Fig. 7D of ZnNPs exhibits diffraction peaks at about 2θ = of 43°, 56°, and 79.83°.
Raman spectroscopy analysis. Raman spectroscopy allows the characterization of many types of samples without any specific requirement preparation. NPs Raman spectra of metallic NPs are shown in Fig. 8. Three peaks were observed around 200, 290, and 1100 cm −1 for FeNPs (Fig. 8A). CuNPs absorptions peaks were around at 500 cm −1 , 1250 cm −1 , and 1500 cm −1 (Fig. 8B). In addition, CoNPs (Fig. 8C) absorption peaks were around 500 cm −1 and 1000 cm −1 , and (Fig. 8D), around 300-500 cm −1 and 1400 cm −1 in the case of ZnNPs.  www.nature.com/scientificreports/ Particle size analysis.. The particle size distribution of metals NPs, as showed in Fig. 9, was analyzed using a particle size analyzer. The average size distribution of FeNPs ( Energy-dispersive X-ray spectroscopy (EDX). EDX analysis gives qualitative as well as quantitative status of elements that may be involved in the formation of NPs and proves the ability of bacterial cells to reduce and accumulate metals from the culture medium. The high presence of Fe, Cu, Co, and Zn in analyzed bacterial cells, as observed in EDX analysis, is reported in Table 3. Iron contents in P. putida isolate was 33%, the content of copper in E. thailandicus isolate was 39.2%, the content of cobalt in M. hydrocarbonoclasticus was 32% and zinc in P. geniculata was 39.1%. The presence of P, Cd, Al, K, Si, Na, Mg, Cl, and S as debris of bacterial cells was observed.
Vibrating sample magnetometer (VSM) analysis. Magnetic characterization of metallic NPs was performed using a VSM. Plots of magnetization versus the magnetic field at 300 K for Fe, Cu, Co, and ZnNPs are shown in Fig. 10. As shown in Fig. 10A the plots of the FeNPs measured a saturation magnetization value around 36.9 emu/g with no hysteresis, the magnetic characterization of CuNPs, as shown in Fig. 10B indicated, low magnetic behavior with saturation magnetization value around 1.3 emu/g without any hysteresis. Magnetization curve of CoNPs showed a very low value around 0.2-0.3 emu/g (Fig. 10C), and ZnNPS saturation magnetization value was` around 5 emu/g as shown in Fig. 10D.
Inductively coupled plasma-optical emission spectroscopic analysis (ICP-OES). The bacterial bio-sorption and reduction of metals capability was assayed by ICP-OES. Determination of Residual Metal was measured in cellfree supernatant compared with the control medium supplemented with metals. The ability of the selected bacterial isolates to accumulate metals intracellularly is shown in S (4). Bacterial isolate P. geniculata can accumulate all amount of Zn 2+ within 2 days (S4D), followed by P. putida that can accumulate all amount of Fe 3+ from the culture medium within 4 days (S4A), while isolates E. thailandicus and M. hydrocarbonoclasticus can accumulate Cu 2+ (S4B) and Co 2+ (S4C) within 5 and 6 days sequentially.

Discussion
Water samples were collected from different locations in Egypt, industrial wastewater, seawater, wastewater, and lake water. The chemical analysis demonstrated that samples contained inorganic pollutants such as Fe 3+ , Zn 2+ , Cu 2+ , and highly toxic heavy metals like Cd 2+ and Pb 2+ . Nano-metals forming bacteria that could be adapted to detoxify these heavy metals were isolated from collected samples. Different studies reported that microorganisms www.nature.com/scientificreports/ have developed the capabilities to protect themselves from heavy metal toxicity by various mechanisms such as adsorption, uptake, methylation, oxidation, and reduction 20 . Metal-reducing bacteria were isolated with dark brown colony or dark zone around bacterial growth. Lima de Silva et al. 20 reported that the changes observed in the colony colors were due to chemical modification of metals when interacting with the bacteria, and not to the induction of real pigmentation. Zaki et al. 21 reported that color change in Stenotrophomonas rhizophila grown at silver nitrate indicates to reduction reaction of AgNo 3 and formation of AgNPs. The present study concluded that all selected bacterial isolates could be able to produce nitrate reductase enzyme. The nitrate reductase was reported to be responsible for nanoparticles (NPs) production especially AgNPs 22,23 .
The morphological, biochemical, and physiological behavior of 4 tested isolates confirmed that two isolates belonged to Pseudomonas spp. and the other two isolates belonged to Enterococcus sp. and Marinobacter sp. Molecular approaches, i.e. 16S rRNA sequencing and phylogenetic analysis of selected isolates, revealed similarity to P. putida, E. thailandicus, M. hydrocarbonoclasticus, and P. geniculata. The literature reported the use of P. putida for bioremediation, due to its ability to degrade organic solvents such as toluene 24 , and for silver and selenium NPs synthesis 25 . Tang et al. 26 reported that P. geniculata was capable of efficiently degrading nicotine. E. thailandicus was first isolated from fermented sausage in Thailand in 2008 and can produce L-lactic acid 27 . On the other hand, M. hydrocarbonoclasticus which use nitrate (NO 3− ) or nitrite (NO 2− ) as the terminal electron acceptor to form gas product such as N 2 O, was first isolated near a petroleum refinery in the Mediterranean Sea.
UV-Vis spectroscopy used for characterization of metal NPs, had proven to be a very useful technique for monitoring the signature of colloidal particles, especially for noble metal since they exhibit strong surface plasmon resonance absorption in the visible region and are highly sensitive to the surface modification 28 . Present data showed the absorption maximum peak of FeNPs, CoNPs, CuNPs, and ZnNPs in agreement with the results of Zhang and Lan 29 , Yuvakkumar et al. 30 , Annapurna et al. 31 and Devasenan et al. 32 .
All UV absorption peaks in the range from 200 to 600 nm, characterized by the absence of long tailing on large wavelengths suggested the absence of aggregation between particles 33 . The presence of only a single band at such wavelength ranges reflected the presence of small spherical particles according to Mie's theory 34 . The position and shape of the surface plasmon absorption of noble metal nanoclusters were strongly dependent on the particle size, dielectric medium and surface adsorbed species 35 .
Data obtained from TEM concluded that the selected isolates were able to detoxify and reduce metallic ions to nanoscale particles. FeNPs in the cytoplasm and periplasmic space of P. putida was observed. Such results were in agreement with Varshney et al. 36 , who found that P. stutzeri synthesized silver NPs within its periplasmic www.nature.com/scientificreports/ space. Synthesis of CuNPs was shown in the cytoplasm space of E. thailandicus, also particles were aggregated in the center of cells by the action of proteins of cytoplasm which prevent their release to the surrounding media through the cell wall. CoNPs synthesis in the case of isolate M. hydrocarbonoclasticus appeared diffused in the cytoplasm. ZnNPs appeared in periplasmic and cytoplasmic space of P. geniculata. Iravani 37 explained that the ability of bacteria to survive and grow in stressful situations might be due to specific mechanisms of resistance which include: efflux pumps, metal efflux systems, inactivation of metals, impermeability to metals, the lack of specific metal transport systems, alteration of solubility, toxicity by changes in the redox state of the metal ions, intracellular precipitation of metals, and volatilization of toxic metals by enzymatic reactions. On the other hand, Gomathy and Sabarinathan 38 suggested that cellular mechanisms may be implicated in the resistance and tolerance of microorganisms to excessive concentrations of heavy metals in the environment. The strategy adopted include slow transport into the cell, detoxification or incorporation of specific metals into enzymes. Deepak et al. 39 reported the process that involves the reduction of metals ions. The first step involves the interaction between these metal ions with nitrate reductase present inside the cell (periplasmic and cytoplasmic membrane) and their bioreduction to metallic form. The accumulation of intracellular NPs was reported in different studies like the synthesis of silver NPs by Stenotrophomonas rhizophilia 21  Patterns of XRD were analyzed to determine peak intensity, position, and width. The diffraction peaks of all NPs appeared sharp, and clearly distinguishable, which indicates the ultra-fine nature and small crystallite size. The XRD spectrum contained no other phase, indicating the purity of the sample 49 . The XRD patterns correspond to FeNPs diffractogram, in which the characteristic peak of highest intensity occurred at 45.50° correspond to (111), crystallographic planes of face-centered cubic iron zerovalent crystals 50 . Two main characteristic diffraction peaks for CuNPs were observed at around 2θ = 42°, 50° which correspond to the (111),  This diffraction pattern has corresponded to pure Zn Nanopowder. A small peak is also observed at around 36° indicated that a small amount of zinc was oxidized and converted into zinc oxide 32 . The XRD pattern of all metallic NPs indicated that the metals NPs had a spherical structure with a small size. The obtained results illustrated that metal ions had indeed been reduced to Nano-form by reducing enzyme of bacterial isolates and prof the ability of selected bacterial isolates to produce metals NPs 9 .
Raman spectroscopy is especially useful if minerals are to be identified that are poorly defined or that cannot easily be distinguished using other methods such as XRD. In this study, Raman spectra of metal NPs showed that the 3 Raman spectra of FeNPs had mainly 3 bands observed around 200, 290, and 1100 cm −1 . Raman region of interest in this investigation was around 200-1100 cm -1 , and it was confirmed to Fe zerovalent Fe 0 NPs peaks. These results were matched with Dong et al. 50 . Besides, results of Raman spectra of CuNPs with peak absorptions around 500 cm −1 , 1250 cm −1 , and 1500 cm −1 indicated that the sample contains Cu 0 NPs 53 . Raman spectra showed absorptions peak around 500 cm −1 and 1000 cm −1 and it corresponded with the Raman shift pattern of the CoNPs 52 . In addition to Raman spectra with 2 strong peaks around 300-500 cm −1 and 1400 cm −1 , these peaks characteristics were identical to peaks of ZnNPs 54 . All Raman spectra of samples indicated the high quality of metal NPs, as well as, these results were compatible with XRD analyses.  www.nature.com/scientificreports/ Analysis using EDX indicated that the content of Fe in P. putida isolate was 46%, the content of Cu in E. thailandicus was 47%, which was coherent with the results by Eltarahony et al. 55 . Content of Co in M. hydrocarbonoclasticus was 41%, and Zn in P. geniculata was 59%, in agreement with finding by Eltarahony et al. 56 . These results fixed that all bacterial isolates could accumulate and reduce the high amount of metal for NPs intracellular synthesis. The presence of other metals might be referred to some elements in bacterial cell structure and biomolecules such as DNA, ATP, RNA, and amino acids 19 .
The magnetic properties of the metal NPs have been investigated using a VSM. The plots of the FeNPs indicate a superparamagnetic behavior at room temperature which is compatible with Eivari et al. 57 , with no hysteresis. The measured saturation magnetization value around 36.9 emu/g. The magnetic characterization of CuNPs indicated low magnetic behavior with a saturation magnetization value of around 1.3 emu/g without any hysteresis 58 . Also, magnetization curves of CoNPs were very low around 0.2-0.3 emu/g, with a soft hysteresis loop 59 . In addition to the magnetization curve of ZnNPs with saturation magnetization had a value of around 5 emu/g 60 . From these results, we concluded that Fe, Cu, and Zn were paramagnetic NPs, and Co was ferro-magnetic NPs.
Determination of residual metal concentration using ICP-OES demonstrated that selected bacterial isolates could detoxificate and accumulates metals intracellular. P. putida could accumulate all amount of Fe 3+ from the culture medium within 4 days, as well P. geniculata could accumulate all amount of Zn 2+ within 2 days, and it might be referring to the ability degree of isolates to detoxification of metals and the toxicity level of metals like Cu 2+ and Co 2+ . Different studies reported the importance of Fe 3+ and Zn 2+ to bacterial cells and their role in different biological pathways. Ahmad et al. 61 reported that Zn 2+ plays a significant role in cellular division and protein synthesis contributes to carbohydrate, lipid, and nucleic acid metabolism. Fe 3+ participates in a large number of cellular processes, the most important of which were oxygen transport, ATP generation, cell growth, proliferation, and detoxification. It was a coenzyme or enzyme activator of ribonucleotide reductase, a key enzyme for DNA synthesis, which catalyzes the conversion of ribonucleotides to deoxyribonucleotides and particularly of deoxyuridine to thymidine. ICP-OES results were compatible and confirm ability of selected isolates to reduce and accumulate metals 19 .
The antibacterial effect of metal NPs against some phytopathogenic bacteria noted that in case of low concentrations of FeNPs, CuNPs and CoNPs increase the antibacterial effect and this effect significantly decreased with www.nature.com/scientificreports/  www.nature.com/scientificreports/ high concentrations. Rout et al. 62 declared that size of the inhibition zone increased significantly with decreasing the size of the NPs. It is reasonable to state that binding of the NPs to the bacteria depends on the surface available for interaction. They added that smaller and monodispersed particles having the larger surface area available for interaction would give more of a biocidal effect than the aggregated and larger particles. Moreover, present data revealed that the antibacterial potential increased with an increase in the concentration of ZnNPs. Several studies reported the antibacterial effect of metals NPs and their antibacterial mechanisms. Lee et al. 63 reported the inactivation of E. coli by zero-valent Fe, furthermore, NPs leads to the production of reactive oxygen species (ROS), resulting in the generation of hydroxyl radicals (OH − ) from superoxide (O 2− ) and hydrogen peroxide (H 2 O 2 ) in microbial cells. These radicals promote oxidative stress and cause cell membrane damage, which contributes to the outflow of intracellular contents and, finally, cell death due to the penetration of the small NPs (sizes ranging from 10-80 nm) into E. coli membranes 64 . Other NPs such as ZnO-NPs 65 , Aerogel-MgO-NPs 66 , and TiO2-NPs 67 have been reported also to cause a loss of membrane integrity and leakage. Other studies on ZnO-NPs and MgO-NPs concluded that antibacterial activity increased with decreasing particle size 68 . Moreover, researches had demonstrated that the small size of NPs, which have characteristic dimensions < 100 nm, can contribute to the bactericidal effect. Their uniquely small size results in novel properties, like the greatest interaction with cells due to a larger surface area-to-mass ratio and multilateral and controllable application 69 . The antibacterial properties of CuO NPs have been investigated, which were found that able to cause protein oxidation, lipid peroxidation and DNA degradation in E. coli cells 70 . Khalil et al. 71 reported that synthesized cobalt oxide NPs were studied for their antibacterial potential against 3 Gve (Pseudomonas aeruginosa, Klebsiella pneumonia, and Escherichia coli) and 3 G + ve bacterial strains (Staphylococcus epidermis, Staphylococcus aureus and Bacillus subtilis). It was noted that the antibacterial potential increased with an increase in the concentration of the nanoparticles.
In this study, FeNPs, CuNPs, and CoNPs were found to be effective at 50-100 µg/mL against some phytopathogenic bacteria, likewise, 200-400 µg/mL of ZnNPs were effective. This effectiveness at lower concentrations might be referred to the small size of bacterial synthesized metal NPs. It was reported that ultrafine particle size causes its action at a lower concentration, whereas our study used NPs with a size range of 2-10 nm. A previous study by Niemirowicz et al. 72 indicated that MIC and MBC of FeNPs against Staphylococcus aureus and P. aeruginosa was 64, 182 µg/mL and 128, 265 µg/mL on the relay, also Ruparelia et al. 73 noted that MIC and MBC of CuNPs against E. coli was 140, 160 µg/mL. On the other hand, Aysa and Salman 74 demonstrated that 3.7 µg/ mL was the MIC and MBC value of ZnNPs against P. aeruginosa.

Conclusion and recommendations:
From this study, we concluded that biological synthesis of NPs using bacterial cells is eco-friendly, fast, and inexpensive. The toxicity of FeNPs, CuNPs, CoNPs, and ZnNPs to phytopathogenic bacteria in addition to their bactericidal effect, recommend future investigations on the toxicity and safety concentrations of metal NPs to animals and human cells. These future studies seem to be necessary to be performed, in order to precise for toxicity prevention and the applicability of metals NPs in bactericide industry.

Material and methods
Samples collection. Various environmental water samples showing harsh conditions for microbial growth (industrial wastewater, seawater, wastewater, and lake water), containing organic and inorganic pollutants were collected from several diverse habitats (Alexandria, Hurghada, and Damietta Governorates), during April 2015 (Table 6).
Physicochemical analysis of water samples. pH, NO 2 − , NO 3 − , salinity, Fe 3+ , Zn, Cu 2+ , Pb 2+ , and Cd 2+ were analyzed and performed according to APHA 75 . Biochemical characterization tests were performed using diagnostics ENC 8 kit and GN 24 kit identification system for Gram-positive and Gram-negative bacteria, respectively. As well DNAase test was performed by streaking each bacterial culture on the DNAase agar plate (HIMEDIA M482) and incubated at 28 ± 2 °C for 3 days, then the plate was flooded with 0.1% toluidine blue. A clearance zone around the colonies was recorded as a positive reaction 81 .

Molecular identification of NPs forming bacteria through 16S rRNA gene. For DNA extraction
protocol from bacterial cells, an aliquot of 1 mL of AMSHAGE DNA extraction kit was added and followed the steps of Abd-El-Haleem 82 . PCR amplification of the 16S rRNA gene was performed for 4 isolates on a "Biometra PCR Thermocycler" using one pair of primer 16S-F (27F) AGA GTT TGATCMTGG CTC AG and 16S-R(1492R) GAT TAC CTT GTT ACG ACT T according to Wang et al. 83 .
PCR products electrophoresis and visualization. Ten μL of PCR product was loaded per gel slot. Electrophoresis was performed at 100 Volt with (0.5 x) Tris-Acetate-EDTA buffer (TAE) [Tris base, 108 g/L; acetic acid, 55 g/L and 0.5 M EDTA with a pH of 8] as running buffer in 1.5% agarose gel cast in 0.5 × (TAE buffer). The gel was stained in (1 μL) ethidium bromide (Royal Biogene). Finally, the gel contained PCR product was visualized with a syngeneic gel documentation system.
Purification of PCR product and sequencing of 16S rRNA gene. PCR products were purified using NEPRAS DNA purification kit 84 . Partial DNA sequencing was performed for the PCR amplified 16S rRNA gene using an ABI PRISM dye terminator cycle sequencing kit with Ampli Taq DNA polymerase and an Applied Biosystems 373 DNA Sequencer (Perkin-Elmer, Foster City, Calif.).
Alignment and phylogenetic analysis. Bootstrap neighbor-joining tree was generated using MEGA version 6.1 from CLUSTALW alignment. Comparisons with sequences in the GenBank database were achieved in BLASTN searches at the National Center for Biotechnology Information (NCBI) site (http:// www. ncbi. nlm. nih. gov). 85 Production and Extraction of metals nanoparticles (NPs) were performed according to Zaki et al. 86 and Kamal et al. 87 respectively.
Application of metals NPs on some phytopathogenic bacteria in vitro. The metals NPs were evaluated for antibacterial activity against 7 molecular identified phytopathogenic bacteria (Table 7). Determination of the inhibition zone (IZ) of tested phytopathogenic bacteria was done by a well diffusion method 93 . The minimum inhibitory concentration (MIC) was determined based on batch cultures containing varying concentrations of metals NPs with serial two-fold dilution in suspension (25-800 µg/mL). All the experiments were carried out in triplicate. Also, the determination of the minimum bactericidal concentration (MBC) was performed as suggested by Avadi et al. 94 . To test for bactericidal effect, one mL of each culture flask which was used in the MIC experiment was plated in duplicates on LB agar free of NPs and incubated at 30 °C for 48 h. NPs concentration causing the bactericidal effect was selected based on the absence of colonies on the agar plate 95 .