Frenemies: Interactions between Rhizospheric Bacteria and Fungi from Metalliferous Soils

Is it possible to improve the efficiency of bioremediation technologies? The use of mixed cultures of bacteria and fungi inoculated at the rhizosphere level could promote the growth of the associated hyperaccumulating plant species and increase the absorption of metals in polluted soils, broadening new horizons on bioremediation purposes. This work investigates interactions between Ni-tolerant plant growth-promoting bacteria and fungi (BF) isolated from the rhizosphere of a hyperaccumulating plant. The aim is to select microbial consortia with synergistic activity to be used in integrated bioremediation protocols. Pseudomonas fluorescens (Pf), Streptomyces vinaceus (Sv) Penicillium ochrochloron (Po), and Trichoderma harzianum group (Th) were tested in mixes (Po-Sv, Po-Pf, Th-Pf, and Th-Sv). These strains were submitted to tests (agar overlay, agar plug, and distance growth co-growth tests), tailored for this aim, on Czapek yeast agar (CYA) and tryptic soy agar (TSA) media and incubated at 26 ± 1 °C for 10 days. BF growth, shape of colonies, area covered on plate, and inhibition capacity were evaluated. Most BF strains still exhibit their typical characters and the colonies separately persisted without inhibition (as Po-Sv) or with reciprocal confinement (as Th-Sv and Th-Pf). Even if apparently inhibited, the Po-Pf mix really merged, thus obtaining morphological traits representing a synergic co-growth, where both strains reached together the maturation phase and developed a sort of mixed biofilm. Indeed, bacterial colonies surround the mature fungal structures adhering to them without any growth inhibition. First data from in vivo experimentation with Po and Pf inocula in pot with metalliferous soils and hyperaccumulator plants showed their beneficial effect on plant growth. However, there is a lack of information regarding the effective co-growth between bacteria and fungi. Indeed, several studies, which directly apply the co-inoculum, do not consider suitable microorganisms consortia. Synergic rhizosphere BFs open new scenarios for plant growth promotion and soil bioremediation.


Introduction
Bacteria and fungi are simultaneously present in a wide variety of environments [1]. These organisms are capable not only of coexisting, but also of actively interacting and the bacteria-fungal interactions (BFIs) range from antagonism to mutualism [2,3]. BFIs in many cases form physically and metabolically interdependent consortia with peculiar characteristics different from those of their individual components [3]. St. Louis, MO, USA) spiked with Ni up to 20 mM [24]. Thanks to these traits, strains were used in the following co-growth tests.
Bacterial strains were grown in cell culture flasks containing tryptic soy broth (TSB, Sigma Aldrich) until reaching the exponential growth phase. The first strain was incubated overnight at 26 ± 1 • C, while the second was kept at 26 ± 1 • C for 72 h, then revived 1:20 in the same growth medium and finally incubated for 30 h. Bacterial cells pellets were washed twice in 1× phosphate-buffered saline (PBS) solution pH 7.4 and then concentrated [43] to obtain the final concentration of 10 8 Colonies Forming Units (CFUs) mL −1 measured spectrophotometrically.
Fungal strains of P. ochrochloron and T. harzianum group were plated on malt extract agar (MEA) and kept at 24 ± 1 • C until conidiogenesis. In addition to the pure fungal culture on MEA agar, an aliquot of well-grown fungal colony was diluted in distilled sterile water and estimated by cell counts using a Bürker chamber up to a liquid culture concentration of 10 8 CFUs mL −1 .
For each bacterium and each fungus previously selected, in addition to the controls, three methods of inoculum were adopted onto two different substrates Czapek Yeast extract Agar (CYA) and TSA at pH 6, and each method was carried out in triplicate (N = 30 each mixed culture test): (a) agar overlay method (100 µL of fungal liquid culture were spread on solid medium and 100 µL of bacterial suspension were inoculated at the center of the Petri dishes); (b) agar plug method (Ø 5 mm agar disc from well-grown lawn of the fungus was cut and transferred on agar plate previously stroked with 100 µL of the bacterial suspension); (c) distance growth method (Ø 5 mm fungal agar plug and 100 µL of the bacterial suspension were inoculated 3 cm away from each other on the same plate). Agar plates (90 mm Ø) with a single central culture of the selected bacteria and fungi were used as positive control.
P. fluorescens SERP1 complex and S. vinaceus SERP4 were grown in cell culture flasks containing tryptic soy broth (TSB, Sigma-Aldrich) as previously described to reach the exponential growth phase.
P. ochrochloron Serp03S and T. harzianum Serp05S group were plated on MEA agar and kept at 24 ± 1 • C until conidiogenesis. The different bacterium-fungus mixes and the positive controls were then incubated at 26 ± 1 • C and the growth of the isolates was daily monitored for 10 days.
Mycelial and bacterial growth was assessed by measuring colony area and the percentage inhibition (Pi%) [44], calculated according to the formula: where Pi% = percentage inhibition ratio, C = average control area (mm 2 ) of the colony and T = average treatment area (mm 2 ) of the colony.

Stereomicroscopy and Optical Microscopy
The Petri dishes were observed under a stereomicroscope and the images were processed with the Leitz(sche) Camera (LEICA, Wetzlar, Germany) application suite (LAS EZ) software to monitor the co-growth of the microbiota after 10 days of incubation and the corresponding shape of both colonies. Subsequently, the in-depth observation of microfungal structures of culturable mix was appreciated through an optical microscope magnified at 40× and 63× and the resulting images were processed using the LEICA IM50 software, while the Gram staining allowed us to underline the Gram+ and Gram-bacteria.

Scanning Electron Microscopy
In order to carefully observe the best grown mixed culture of microbiota after 14 daysincubation on CYA, the isolates mix were fixed in phosphate 10% buffered formalin and dehydrated in ethanol series [45]. The samples were then mounted on SEM stubs, sputtercoated with gold and viewed with a VEGA3 SEM (TESCAN, Libušina, Czech Republic) at HV20.0 kV, using backscattered electrons (BSE).

Soil Sampling and Processing for In Vivo Experimentation
Serpentine soil was collected from Sassello (SV, Italy N 44 • 48 63"; E 8 • 51 87") at 0-30 cm (A horizon) and 30-60 cm (C horizon) depth, mixed to obtain a homogeneous sample, air de-hydrated and sieved through a sequential 4 mm and 2 mm mesh. One soil fraction was used as is for Control, the other fractions were oven dried at 130 • for 24 h to sterilize the growing substrate, then mixed 1:1 (v/v) to obtain the testing substrates for pot experiment. Soil chemical composition is reported [46], as shown following in Table 1.

In Vivo Experimental Design
Four L of substrate prepared as previously described (approx. 4.5 kg each pot) was divided into Control, Bacteria (BI, inoculum with Pseudomonas fluorescens), Fungi (FI inoculum with Penicillium ochrocloron) and Mix (MIX co-inoculum Po + Pf ), 30 replicates each series. Soils were hydrated at 70% water holding capacity and subsequently irrigated once a week to keep them at optimum hydration level. After three months from the inocula, pots series were completely randomized, and six-months old plants were then transplanted in the pots. All the pots had their saucers to recover the irrigation water.

Bacteria and Fungi Co-Inoculum in In Vivo Experimentation
The selected P. ochrochloron strain (FI) was isolated as pure colony on MEA (Sigma-Aldrich) and incubated in the dark at 24 ± 1 • C for at least one week. Once the pure colony has been grown and the conidiogenesis was reached, mycelium and conidia were collected with sterile loop and resuspended in 100 mL of sterile distilled water up to a final concentration of about 10 8 CFUs mL −1 . To verify the final fungal concentration, aliquots of this latter solution were transferred to the Bürker chamber for conidia count.
P. fluorescens strain (BI) was repeatedly isolated to obtain a pure colony on TSA (Sigma-Aldrich) incubated at 26 ± 1 • C in the dark for at least 24-48 h. One day before the experiment, a well isolated colony is transferred to a cell culture flask containing 200 mL of TSB (Sigma-Aldrich) and incubated overnight at 26 ± 1 • C. The subsequent day, the cultural broth is transferred into 50 mL falcon tubes which are centrifuged at 4500× g for 10 min at 20 • C. Then, the supernatant is gently removed from each falcon tube and the pellet is subjected to washing with an equal volume of PBS (pH 7.4). A spectrophotometer analysis allows to determine the optical density (OD) at a wavelength of 600 nm. Finally, the pellet obtained after a second centrifugation is resuspended in 100 mL of physiological saline solution. After an overnight incubation of P. fluorescens at 26 ± 1 • C the obtained OD is 0.5, which corresponds to an average concentration of 2.5 × 10 8 CFUs mL −1 of the Gramnegative bacterium P. fluorescens. Furthermore, the bacterial suspension is concentrated twice, to obtain a final concentration of 5 × 10 8 CFUs mL −1 . The BI and FI inoculation in pots is carried out immediately after their preparation, transferring 1 mL of bacterial and fungal suspension of about 10 8 CFUs mL −1 on the soils in the pot centre.

Microbiota Assessment
The fungal and bacterial diversity and abundance were monitored at month I (without plants) and XII (with six-months old plants) after the inocula. Soil was sampled in each pot and mixed for each treatment to obtain a composite sample. Soils samples were collected for every type of treatment: fungal inoculum (FI), bacteria inoculum (BI), fungal-bacteria co-inoculum (MIX), native soil without inocula (Control). Microorganisms were extracted from 3 g of soil by using 30 mL of sterile saline solution (NaCl 0.9% w/v). Aliquots were serially diluted (1:10) up to 10 −5 and 100 µL of each dilution was spread on Rose Bengal Agar (RBA) and MEA. Petri dishes (Ø 90 mm) for fungi and TSA (Sigma-Aldrich) for bacteria were employed. Then, fungal plates (N = 24) were incubated in the dark at 24 ± 1 • C and monitored every day for a week, while bacterial plates (N = 16) were incubated in the dark at 26 ± 1 • C for 72-96 h. The fungal and bacterial colonies were counted as Colonies Forming Units per g of dry soil (CFUs g −1 ) for each morphotype grown and repeatedly isolated on agar plates to obtain a pure colony.

Data Analysis
For each treatment, the images of the bacterial and fungal colonies were processed with ImageJ software [47,48] to estimate the surface area. Data are presented as mean ± Standard Deviation. The percentage of the inhibition rate was obtained considering the average surface of the bacterial and fungal colonies.

In Vitro Tests
The screening tests of co-growth showed that CYA and TSA buffered at pH 6.0 resulted the most suitable agar substrates for the growth of both bacterial and fungal strains.
Fungal strains exhibited an increased growth on the CYA substrate compared to TSA pH 6.0, whereas bacteria thrive on both agar culture media.
As concerns fungi, the colony of P. ochrochloron (Po) on CYA appears round, grooved in a radial pattern, white to light grey in the margin once maturation is reached (Figure 2A,C). This fungus exhibits fast growth rate, and its colonies are variously branched, with asymmetrically and/or symmetrically branched conidiophores [51].
T. harzianum group (Th) isolates ( Figure 2B,D) produce typical rough and globose to subglobose conidia and their color was yellow or green [52][53][54]; conidia production is greater in the middle of the plate with respect to the margins.
The results of co-growth tests are summarized as follows: (a) P. ochrochloron (Serp03S) vs. P. fluorescens (SERP1): Po-Pf mix The agar overlay method ( Figure 3A,D) on CYA does not seem to clarify any relationship between the two organisms. The bacterium develops where it has been inoculated, while the fungus grows better around bacterium. After 10 days of incubation the fungus reaches maturation, taking on a grey-black color. . Stereo Microscope details of single colony shape of P. fluorescens (C) that forms small, low convex, colorless or white-cream colonies, and S. vinaceus (D) that forms circular, convex, and rough colonies with a macroscopic mycelium morphology including pellets and clump formation. N = 30 each test.
As concerns fungi, the colony of P. ochrochloron (Po) on CYA appears round, grooved in a radial pattern, white to light grey in the margin once maturation is reached ( Figure  2A,C). This fungus exhibits fast growth rate, and its colonies are variously branched, with asymmetrically and/or symmetrically branched conidiophores [51].
T. harzianum group (Th) isolates ( Figure 2B,D) produce typical rough and globose to subglobose conidia and their color was yellow or green [52][53][54]; conidia production is greater in the middle of the plate with respect to the margins. and on CYA (lower row of plates, A and B, respectively). Stereo Microscope details of single colony shape of P. fluorescens (C) that forms small, low convex, colorless or white-cream colonies, and S. vinaceus (D) that forms circular, convex, and rough colonies with a macroscopic mycelium morphology including pellets and clump formation. N = 30 each test.
Interestingly, in the agar plug ( Figure 3B,E) and in the distance growth methods ( Figure 3C,F) the contact between the two organisms develops a peculiar round structure with concentric rings that represents a sort of synergic growth.
(b) P. ochrochloron (Serp03S) vs. S. vinaceus (SERP4): Po-Sv mix The mixed culture covers almost the entire agar plate, taking on a grey-white color both in the overlay ( Figure 4A,D) and plug test ( Figure 4B,E). No zones of inhibition are visible in the distance growth method, specifically in the contact zone ( Figure 4C,F), but each strain maintained a distinct colony shape in terms of macroscopic characteristics.
(c) T. harzianum group (Serp05S) vs. P. fluorescens complex (SERP1): Th-Pf mix In the agar overlay method the fungus does not thrive where the bacterium was inoculated ( Figure 5A,D). Conversely in the other treatments (agar plug, Figure 5B,E and distance growth, Figure 5C,F) the fungus covers almost the entire agar plate, but its growth appears to be disturbed: in the last distance growth method, fungal strain is confined in its half of the plate. The results of co-growth tests are summarized as follows: (a) P. ochrochloron (Serp03S) vs. P. fluorescens (SERP1): Po-Pf mix The agar overlay method ( Figure 3A,D) on CYA does not seem to clarify any relationship between the two organisms. The bacterium develops where it has been inoculated, while the fungus grows better around bacterium. After 10 days of incubation the fungus reaches maturation, taking on a grey-black color.
Interestingly, in the agar plug ( Figure 3B,E) and in the distance growth methods (Figure 3C,F) the contact between the two organisms develops a peculiar round structure with concentric rings that represents a sort of synergic growth.  In the agar overlay method, Po grows around Pf reaching maturation and its color turns to grey-black (D). In the agar plug method (E) and in the distance growth method (F) the contact between the two organisms develops a peculiar round structure with concentric rings. N = 30 each test. agar overlay method, Po grows around Pf reaching maturation and its color turns to grey-black (D). In the agar plug method (E) and in the distance growth method (F) the contact between the two organisms develops a peculiar round structure with concentric rings. N = 30 each test.
(b) P. ochrochloron (Serp03S) vs. S. vinaceus (SERP4): Po-Sv mix The mixed culture covers almost the entire agar plate, taking on a grey-white color both in the overlay ( Figure 4A,D) and plug test ( Figure 4B,E). No zones of inhibition are visible in the distance growth method, specifically in the contact zone ( Figure 4C,F), but each strain maintained a distinct colony shape in terms of macroscopic characteristics.   Figure 5A,D). Conversely in the other treatments (agar plug, Figure 5B,E and distance growth, Figure 5C,F) the fungus covers almost the entire agar plate, but its growth appears to be disturbed: in the last distance growth method, fungal strain is confined in its half of the plate. The bacterial colonies appear smaller than the control and the fungus reaches maturation later than pure colonies. The bacterium grows only in the center of the plate in the overlay agar test ( Figure 6A,D), while the fungus appears to be inhibited in the agar plug test (Figure 6B,E). Bacteria and fungi tolerate each other, as shown by the contact between (d) T. harzianum (Serp05S) vs. S. vinaceus (SERP4): Th-Sv mix The bacterial colonies appear smaller than the control and the fungus reaches maturation later than pure colonies. The bacterium grows only in the center of the plate in the overlay agar test ( Figure 6A,D), while the fungus appears to be inhibited in the agar plug test (Figure 6B,E). Bacteria and fungi tolerate each other, as shown by the contact between the two microorganisms ( Figure 6C,F), but they remain distinct. In the distance growth method, Th and Sv tolerates each other without reciprocal inhibition in the contact zone but persists distinct (F). N= 30 each test.

Stereomicroscopy and Optical Microscopy
In-depth observation of the best co-grown bacterium and fungus mix (Po-Pf mix) under the optical microscope ( Figure 7A,B) shows the typical fungal conidiophore branching patterns mostly symmetrical and mono-and biverticillated [55]. The Gram staining underlines the Gram-negative bacterium P. fluorescens complex, counter-stained pink by safranin ( Figure 7C,D). However, the conidiophores of P. ochrochloron are also clearly stained with a bright red color. In the agar overlay method, Sv only grows in the center of the plate (D). In the agar plug method, Th appears to be inhibited (E). In the distance growth method, Th and Sv tolerates each other without reciprocal inhibition in the contact zone but persists distinct (F). N = 30 each test.

Stereomicroscopy and Optical Microscopy
In-depth observation of the best co-grown bacterium and fungus mix (Po-Pf mix) under the optical microscope ( Figure 7A,B) shows the typical fungal conidiophore branching patterns mostly symmetrical and mono-and biverticillated [55]. The Gram staining underlines the Gram-negative bacterium P. fluorescens complex, counter-stained pink by safranin ( Figure 7C,D). However, the conidiophores of P. ochrochloron are also clearly stained with a bright red color.

Scanning Electron Microscopy
The backscattered (BSE) SEM imaging of the best co-grown bacterium and fungus mix (Po-Pf mix) highlights clusters of fungal hyphae enveloped by groups of flagellate Pf ( Figure 8A,B). Conidiophores are often covered on the tip by bacterial colonies adhering by means of adhesive filaments that provide a compact structure to bacterial biofilm consisting of live and dead cells ( Figure 8C,D). Fungal spherical conidia are still borne ( Figure 8E) or free ( Figure 8F) on specialized stalks of the conidiophore that is characterized by typical vial-shaped cell (i.e., phialide), as shown in Figure 8D.

Scanning Electron Microscopy
The backscattered (BSE) SEM imaging of the best co-grown bacterium and fungus mix (Po-Pf mix) highlights clusters of fungal hyphae enveloped by groups of flagellate Pf ( Figure 8A,B). Conidiophores are often covered on the tip by bacterial colonies adhering by means of adhesive filaments that provide a compact structure to bacterial biofilm consisting of live and dead cells ( Figure 8C,D). Fungal spherical conidia are still borne ( Figure  8E) or free ( Figure 8F) on specialized stalks of the conidiophore that is characterized by typical vial-shaped cell (i.e., phialide), as shown in Figure 8D.

Mutual Growth Inhibition
The analysis of bacterial and fungal growth reveals that the measured area is a very critical parameter to evaluate the inhibition of the colonies' growth ( Table 2).

Mutual Growth Inhibition
The analysis of bacterial and fungal growth reveals that the measured area is a very critical parameter to evaluate the inhibition of the colonies' growth (Table 2). Fungi grow slower than bacteria, but the expansion over time of mycelium is greater than that of bacterial colonies. Therefore, fungal strains show a greater surface coverage than the bacterial one. T. harzianum group covers the entire round surface area of the agar plate (90 mm Ø) and P. ochrochloron occupies about 70% of the surface area of the circle.
Conversely, the bacterial strains show a limited growth close to the initial bacterial suspension (~25% of the plate surface). As shown in Table 3, P. fluorescens shows a marked reduction in colony area in the co-growth treatment with both P. ochrochloron (>50% for the average area) and T. harzianum group (>70% for the average area). Table 3. Mutual inhibition of bacterial and fungal growth at different distance growth treatments. N = 3 each colony. Po: P. ochrochloron; Th: T. harzianum; Pf : P. fluorescens; Sv: S. vinaceus.

Po-Pf Mix
Po-Sv Mix However, this result is difficult to explain in the distance growth test with P. ochrochloron: the development of a sort of synergic growth between these two microorganisms does not allow to distinguish the distinct shape of the two colonies that completely merged. This phenomenon was also observed by optical and scanning electron microscopy. On the contrary, S. vinaceus is less sensitive to co-growth with other fungal strains, showing a low percentage of inhibition. This is especially true for the test between S. vinaceus and P. ochrochloron (~8% inhibition). The fungus instead shows a surface area variation roughly ranging from 30 to 40%. Similar findings for all the treatments including T. harzianum group, which shows signs of suffering with respect to the control.

Assessment of Culturable Fungi
One month from the inoculation five different fungal morphotypes were recognized in the plated soils. The fungal abundance was very high in the control and in the BI samples. The main representative fungus morphotype was Penicillium cfr. ochrochloron. Other fungi were Trichoderma strain, isolated in each soil sample except the control; Neurospora sp., isolated in each sample except the FI and MIX; Mucor sp. isolated in each sample, and Aspergillus sp. isolated in control and BI samples. The abundance of each morphotype is reported in Figure 9. reported in Figure 9.
Twelve months after the inoculation, three different fungal morphotypes were identified in the plates. The fungal abundance was homogenized in the samples, but the Penicillium cfr. ochrochloron was the most represented species, followed by Mucor sp. and Trichoderma sp.. No Aspergillus sp. and Neurospora sp. strains were isolated from the plates. The abundance of each morphotype is reported in Figure 10.  Twelve months after the inoculation, three different fungal morphotypes were identified in the plates. The fungal abundance was homogenized in the samples, but the Penicillium cfr. ochrochloron was the most represented species, followed by Mucor sp. and Trichoderma sp. No Aspergillus sp. and Neurospora sp. strains were isolated from the plates. The abundance of each morphotype is reported in Figure 10.

Assessment of Culturable Bacteria
One month after inoculation a greater diversity is observed, where P. fluorescens has been added to the soil (BI and MIX) compared to control and FI treatments ( Figure 11). In general, there is a prevalence of genus Pseudomonas sp. which represents more than 75%

Assessment of Culturable Bacteria
One month after inoculation a greater diversity is observed, where P. fluorescens has been added to the soil (BI and MIX) compared to control and FI treatments ( Figure 11). In general, there is a prevalence of genus Pseudomonas sp. which represents more than 75% of total culturable bacterial strains in BI, while it stands around 34% in the MIX.  It is interesting to note that the dominant bacterial strain in FI (~85%) is Erwinia sp., while the Pseudomonadaceae are dominant in control (97.5%).
After twelve months, although there are no significant differences on the isolated morphotypes, Pseudomonas sp. it is no longer the most abundant genus in BI (−22%). In fact, Streptomyces sp. (−24%) and Erwinia sp. thrive in all types of treated soils. In fact, the pie chart in Figure 12 shows that Erwinia sp. represents half of the bacterial colonies isolated in BI, 66% and 68% in FI and MIX respectively and even represents the monospecific community in control.

Plant Growing
Six months after transplanting, plants from inoculated or co-inoculated pots are markedly different respect to control. These latter showed reduced growth with absence of secondary branches and were about 10 times smaller than inoculated plants ( Figure 13).
The overall growth was MIX > BI > FI > CONTROL. Figure 11. Average bacterial count expressed as genus diversity and colony forming units (CFUs g −1 ) in the rhizospheric soil one-month post inoculation: control, fungal inoculum; bacterial inoculum; MIX treatment. N = 16 each treatment.

Plant Growing
Six months after transplanting, plants from inoculated or co-inoculated pots are markedly different respect to control. These latter showed reduced growth with absence of secondary branches and were about 10 times smaller than inoculated plants ( Figure 13).
The overall growth was MIX > BI > FI > CONTROL.

Discussion
The co-growth tests between bacterial and fungal strains allow establishing the positive or negative interactions and the mutual behavior of the selected strains. Most of the bacterial and fungal strains used in mixes showed typical macro-and micromorphology where the two strains separately persisted without inhibition (as in Po-Sv), or with reciprocal confinement sometimes developing slower than expected (as in Th-Sv and Th-Pf).

Discussion
The co-growth tests between bacterial and fungal strains allow establishing the positive or negative interactions and the mutual behavior of the selected strains. Most of the bacterial and fungal strains used in mixes showed typical macro-and micromorphology where the two strains separately persisted without inhibition (as in Po-Sv), or with reciprocal confinement sometimes developing slower than expected (as in Th-Sv and Th-Pf ). On the contrary, this study highlighted for the first time the morphological traits that represent a sort of synergic co-growth, as in the Po-Pf mix, where both strains really merged and developed some new features that can be assumed as a mixed biofilm, while reaching the maturation phase ( Figure 7C,D and Figure 8). However, further investigations are needed to assure that the reduction of surface area in both strains is only related with the merged shape that do not allow to distinguish the distinct morphology of the two colonies. Specifically, the Pseudomonas strain belonging to the fluorescens complex is considered the most promising group of PGPR involved in the biocontrol of plant diseases for its antifungal metabolites production [42], despite this trait is almost absent in the SERP1 strain used for the present co-growth test. In addition, certain P. fluorescens strains can detoxify organic and inorganic pollutants, combating heavy metal pollution and bioremediating pesticides [56,57]. Regarding Penicillium ochrochloron here studied, the Ni tolerance and the metal uptake ability were already documented [24,40]. Moreover, P. ochrochloron is a metal-resistant fungus [58], highly tolerant to copper and other metals [59].
Although the analysis of bacteria and fungi allowed defining their individual properties, and potential implications and applications in bioremediation as previously described, the development of integrated protocols, jointly exploiting fungi and bacteria, requires and adequate knowledge of their mutual behavior in the microbial consortium. To date, few works investigated the possible BFIs and synergy between rhizosphere PGP bacteria and microfungi [60,61].
The monitoring of the fungal and bacterial abundance in the in vivo experimentation evidenced how the soils went through an intermediate stage before stabilization. After one month, in fact, the presence of typical environmental fungal strains such as Neurospora sp., Mucor sp., Trichoderma sp. and Aspergillus sp. was reported, and it was particularly high in the Control tests where no fungi were inoculated. Similarly, we can affirm that bacterial diversity traced that observed in other serpentinitic soils [24]. The high abundance of the genus Pseudomonas sp. is affected by the inoculum of 10 8 CFU mL −1 .
However, twelve months after inoculation (and six months after transplanting) data on fungi were more homogeneous. Fungi, in fact, were equally distributed in each test group, Penicillium cfr. ochrochloron was the most abundant strain in each treatment and some morphotypes disappeared (Aspergillus sp. and Neurospora sp.). This is probably due to the fungal strains' competition, which favored the inoculated strain over time, allowing its aerial dispersion and colonization of sterile near soils of control tests.
On the contrary, the bacterial community showed a greater dominance of a few genera: Erwinia sp. in all treatments (deriving from original soil microbial strains), Pseudomonas sp. (BI and MIX), and Bacillus sp. (FI and MIX). In this case, the competition between microorganisms seemed to favor a ubiquitous strain such as Erwinia sp. isolated on both serpentinite and non-serpentinite sites, in rhizospheric and adjacent bare soil [24]. Previously, PGP and Ni tolerance tests performed on bacterial strains isolated from serpentinite soil [24] belonging to Erwinia sp. showed significant PGPR traits (production of siderophores and solubilization of phosphates). Erwinia tolerated up to 5 mM of Ni compared to Pseudomonas sp. and Streptomyces sp. (up to 15 mM), although it was more competitive from an ecological point of view.
Plants seemed to have a good response to bacterial and fungal inoculum after six months from transplanting in pots. However, in controls tests, where Erwinia genus is dominant, plant growth appeared modest compared to the other treatments. This might be related with the low ability of this microbial strain to alleviate root stress linked to high metal concentrations. Further investigations are required to clarify the interaction between Erwinia and roots and among other fungal strains in co-growth to clarify its possible synergistic traits. Potential Erwinia 'rhizosphere effect', described for other microbial strains where root system could promote the bacterial and fungal proliferation [62][63][64][65][66] should be investigated.

Conclusions
In conclusion, we identified the PGP strains of P. ochrochloron and P. fluorescens isolated from the rhizosphere of A. utriculata as able to synergically co-grow in a mixed culture developing a biofilm where the two microorganisms merge, reaching the mature stage. Hence, these strains have been employed in the in vivo experimentation on metalliferous soil to directly evaluate their potential synergic role in the rhizosphere of A. utriculata for phytoremediation purposes. Inocula of MIX and Penicillium ochrochloron show the positive effect.
This synergistic fungi-bacteria co-growth opens new perspectives in plants' growth promotion and soil bioremediation at the rhizosphere level. The achieved results suggest the possible use of mixed inocula to improve the metal uptake of the associated hyperaccumulator plant species in metal-contaminated soils. Moreover, our study evaluated and underlined the relevance and effectiveness of the agar overlay, the agar plugs, and the distance co-growth screening tests for selecting microorganism consortia for biotechnological purposes.