Poaceae with PGPR Bacteria and Arbuscular Mycorrhizae Partnerships as a Model System for Plant Microbiome Manipulation for Phytoremediation of Petroleum Hydrocarbons Contaminated Agricultural Soils

: Total petroleum hydrocarbons (TPHs) are a persistent environmental organic contaminant. The possibility of obtaining synergistic effects between plants and microorganisms has further increased the possibility of alternative techniques for bioremediation. Oloptum miliaceum (L.) Röser & H.R. Hamasha and Pennisetum setaceum (Forssk.) Chiov. are, undoubtedly, good model plants for phytoremediation because they have large roots, leaf biomass, and a fast and effective renewal capacity, also, they have a great capacity to host endophytes in their roots. Gas chromatography-mass spectrometry (GC-MS) based on carbon fraction number was a basic technique used to determine the hydrocarbon degradation, and microorganism’s population was identified by high-throughput sequencing of 16s rRNA. The microbial consortium used allows the plant to increase overall biomass, adapt more in terms of redox biology (Superoxide dismutase SOD, catalase CAT, ascorbate peroxidase APX, guaiacol peroxidase GPX), and stress markers (Glutathione S-transferase GST, Phenylalanine Ammonia Lyase PAL, Proline content, and lipid peroxidation MDA). In addition, the photosynthetic efficiency and the soil dehydrogenase activity were monitored. After 240 days, the percentage of TPHs removed in Group 2 was 94%, whereas in Group 1, it was 78% in Oloptum miliaceum and Pennisetum setaceum. The removal of aliphatic hydrocarbons (C 13 –C 36 ) was observed in Oloptum miliaceum and Pennisetum setaceum inoculated with the consortium of indigenous bacteria selected from rhizosphere soil and mycorrhizae strains. Our data demonstrate that, the Poaceae , in relation to its great ecological and vegetative potential, could be a great candidate for extensive remediation of soils contaminated by TPHs. These results demonstrate a hold relationship between DHA and biodegradation of petroleum-contaminated soil [46–49]. Moreover, between 60 and 90 days, there was a positive relationship between DHA and root lengths; in other words, the root growth contributes to TPHs’ degradation by the microbial activity of rhizosphere.


Introduction
Total petroleum hydrocarbons (TPHs) are a recalcitrant environmental organic pollutant [1][2][3]. They provoke consistent detriment to ecosystems and their buildup of in animal and plant tissues can result in significant gene mutations [4][5][6][7]. Recently, uses of plants to remove pollutants have significantly increased, considering that the use of plants was also a good remediation approach with regard to cost-containment. Moreover, the possibility to obtain synergistic effects between plants and microorganisms useful for the growth and development has increased the possibility of alternative techniques [8][9][10][11][12]. Therefore, bacteria associated with plants, such as endophytic bacteria and fungi that live in cortical tissue of roots of plants and rhizosphere bacteria that live on/near the roots of plants, contribute to biodegradation of organic contamination in soil, and improve phytoremediation. These endophytes, because of intimate contact with their host plants, play vital roles in plant development, growth, and fitness, as well as decontaminating polluted soil [13,14]. Endophytic microorganisms are a highly varied group that stay within the tissues of plants for at least part of their life cycle [15]. Endophytic bacteria and fungi host a large amount of genes that degrade pollutants [16]. These benefits from endophytic microorganisms could improve the usage of endophytes in the phytoremediation of organic contaminants in soils. They produce phytohormones [17,18] such as indole-3-acetic acid [19,20], cytokinins, and gibberellins. Once a plant-endophyte association has been established, this partnership can generate new phytoremediation targets for polluted land [21]. Many species have been tested for eliminating TPHs, such as, ryegrass (Lolium multiflorum) [22], tall fescue (Festuca arundinacea) [1], alfalfa (Medicago sativa) [23], maize (Zea mays) [2,24,25], poplar (Populus ssp.) [26], and red weed (Cynodon dactylon) [27]. Many studies have focused attention to the Poaceae, cleaning up from TPHs, since they are very resistant to stress TPHs, have good root spread (colonize large areas), and do not need special care and nutrients [28][29][30]. Grasses generally have these alternatives reproductive characteristics [31].
Oloptum miliaceum and Pennisetum setaceum are bushy plants with numerous upright or ascending culms, wrapped in sheaths and with the appearance of tight tufts. They are perennial plants cultivated by means of gems placed at ground level [32]. They have strong ecological value as they resist drought and live in poor and exposed land. They live well even on disturbed soils. This species is very resistant to dryness because it does not need any special cultural practices nor is it subject to plant diseases [32].
The aims of this study are to assess soil TPHs remediation by plants and microorganisms to use this technology in the field. The mesocosms were carried out to evaluate the capacity of two species of Poaceae (Oloptum miliaceum and Pennisetum setaceum) for TPHs removal by a mechanism of phytodegradation. Poaceae are a good candidate for rhizodegradation because of their large and collated roots supplying a great root surface for the development of bacterial communities.

Selection of the Native Bacteria of Contaminated Soil
The bacterial strains were isolated from contaminated soil that has been withdrawn from a periurban cultivated area (Naples) near a fuel station, with contamination by petroleum hydrocarbons, mostly diesel. The isolation was as described previously by Guarino et al. [8]. Indolacetic acid production, siderophores release, exopolysaccharides production, and ammonia production were determined in vitro, as described by Guarino et al. [8].

Mesocosm Experiment
One thousand seeds of Italian multifloral millet (O. miliaceum (L.) Röser & H.R. Hamasha and one thousand seeds of fountain grass (P. setaceum (Forssk.) Chiov.) collected in abandoned peri-urban areas of Naples near the industrial complex of the refineries were sterilized in commercial surface and sterile distilled water for 1 h. Subsequently, twenty seeds of two species for each pot were selected, and seeded mesocosms were filled with about 29 kg of contaminated soil. The soil was sieved through a 2 mm sieve mesh. The main chemical and physical characteristics of contaminated soil are showed in Supplementary Table S1. Six of the forty-six isolated endophyte bacteria ( Table 1) that showed different activity of Plant growth-promoting rhizobacteria (PGPR) (Comamomas koreensis, Ochrobactrum anthrophi, Pseudomonas fluorescens, Pseudomonas putida, Bacillus cereus, and Paenibacillus mucilaginosus) were selected for the mesocosms experiment.
Besides, a consortium of endomycorrhize (Glomus versiforme, Funneliformis mosseae, Rizophagus irregular, Rizophagus intraradicens) obtained from the repository of the Department of Science and Technology of the University of Sannio was added to isolated endophyte bacteria.
The experimental design included three different experimental conditions for each specie: 1) The first set of mesocosms was formed by 50 pots of O. miliaceum (20 seeds per pot) in contaminated soil without the inoculum of endophytic consortium strains selected (Group 1). 2) The second set of mesocosms consisted of 50 pots of O. miliaceum (20 seeds per pot) in contaminated soil supplemented by the endophytic consortium strains selected (4.0 × 10 11 -1.0 × 10 12 CFU ml −1 g of soil) (Group 2). 3) The control set consisted of 50 pots of O. miliaceum (20 seeds per pot) with select endophytic consortium strains selected (4.0 × 10 11 -1.0 × 10 12 CFU ml −1 g of soil) in commercial noncontaminated soil (Control Group). The pots, which were watered daily, were placed in the controlled greenhouse according to the following protocol: Plants were removed from pots, and roots and leaves were divided. Fresh leaves were kept for subsequent analyses. Roots were washed with distilled water and blotted with tissue paper. Soil samples were kept at 4 °C before analysis.

Stress Marker and Antioxidant Enzyme Activity
For enzyme analysis, thirty samples of fresh leaves (250 g) of O. miliaceum and P. setaceum were homogenized under liquid nitrogen into a fine powder. The homogenates were centrifuged at 19,000g for 30 min at 4 °C. The supernatant was used to evaluate the activity of stress markers (Glutathione S-transferase, phenylalanine ammonia lyase, Proline content, and lipid peroxidation), as described by Guarino et al. [33].

Chlorophyll Content
The contents (µg mL −1 ) of Chl a, Chl b, and Chl a + b in the leaves of O. miliaceum and P. setaceum (0.5 g) were measured spectrophotometrically in accordance with the study by Lichtenthaler and Wellburn [34].

Biomass
The root system of O. miliaceum and P. setaceum was divided from the leaves. The roots were washed with deionized water and then with 10 mM Ethane-1,2-diyldinitrilo-tetraacetic acid (EDTA) to eliminate the soil particles. The fresh weight of a sample of leaves was determined, followed by flotation on water for up to 4 h. The turgid weight was then recorded, and the leaves were oven-dried to a constant weight, at about 85 °C.

Quantification of Petroleum Hydrocarbons
Thirty-five samples of O. miliaceum and P. setaceum of each set of mesocosms were divided into two sections (roots and leaves) and were carefully washed with deionized water. Two grams of leaves and roots were ground with anhydrous Na2SO4 using the method described by Guarino and Sciarrillo [9].

Quantification of Petroleum Hydrocarbons
Soil samples were exposed to a pre-treatment comprising air-drying until continuous weight followed by sieving to 2 mm. Guarino and Sciarrillo [9] described the procedure of quantification of TPH. TPH content was quantified in low (C < 12) and higher molecular weight (C > 12) according to the method used for the SimDist analysis (Guarino and Sciarrillo, [9]).

Dehydrogenase Activity
Soil dehydrogenase activity (DHA) was determinate by Hayano and the iodo nitrotetrazoliumformazan was measured spectrophotometrically at 485 nm.

Statistical Analysis
The data are the means of three replicates. The Dixon's test was used to test the presence of outliers in the dataset. Differences between groups were tested using Duncan's test.

Screening of the Isolated Hydrocarbon Degraders' Bacteria
Forty-six native bacteria were designated from contaminated soils. The isolates are Proteobacteria phylum, with more than 60% being part of the γ-subclass, about 25% being part of the b-subclass, and two strains being part of the a-subclass. Pseudomonas genus is the most represented ( Table 1). The obtained results showed that 56% of the isolates produce the phythormones Indole acetic acid (IAA), with 45% of isolates being able to produce Exopolysaccharides (EPSs), and 60% of isolates releasing siderophores. Ammonia-producing bacteria can act as biofertilizers in depleted soils, such as those which are contaminated, and more than half of the 56% of isolates (36%) showed this property (Table 1). Based on the results that established that the bacteria isolates are good applicants to support and improve a promising phytoremediation stage, we have selected the isolates strains for the mesocosms experiment. Table 1. Primary screening of the assessment of potential Plant growth-promoting rhizobacteria (PGPR) by bacteria isolates recovered from contaminated soil from a peri-urban cultivated area (Naples). The isolates were categorized into three groups according to the produced amount: + low concentrations (<1 µg/mL), ++ moderate concentrations (1-2.99 µg/mL), and +++ high concentrations (>3 µg/mL).

Bacterial isolates
Production IAA Siderophores release Exopolysaccharides production

Plant Analysis: Antioxidant Enzyme Activity and Stress Marker
Our results showed that rhizosphere microorganisms and exposure time to different treatments influence activities of the antioxidative enzymes. The activity of SOD in O. miliaceum and P. setaceum from 30 to 90 days rose significantly (p < 0.05) in plants with the rhizosphere microorganism in comparison to the plants without microorganisms (Group 1) and the control group ( Figure 1). The highest SOD activity in leaves of O. miliaceum and P. setaceum with microflora at T1 and T3 could be a better response for plants to protect themselves from induced oxidative damage from hydrocarbons exposure in the presence of microorganisms that promote the growth of the plants. In this work, the CAT, GPX, and APX activities in leaves of O. miliaceum and P. setaceum showed parallel patterns to SOD activity and increased from T1 to T3 treatments compared with the control group and mesocosms with microorganisms (Group 2). For assessment of the degree of oxidative stress, the malandialdehyde content (MDA) was tested, such as Glutathione S-Transferase, Phenylalanine Ammonia Lyase, and Proline content ( Figure 2). An increased level of MDA content was observed in the leaves of O. miliaceum and P. setaceum inoculated in response to hydrocarbons in the present study. After a longer exposure time (TF 240 days), decreased MDA content in the leaves of O. miliaceum was found, with no significant variance between the Control Group and Group 2. The increase of MDA content from T1 to T3 suggests a strong internal detoxification mechanism inside plant cells, while the decrease of MDA content during TF indicates that the plants need some time to tolerate the hydrocarbons' stress and therefore, respond to activate the antioxidative defense system against higher levels of Reactive oxygen species (ROS). The role of Glutathione S-transferase (GST) is very evident and is in line with other antioxidant enzymes, such as Phenylalanine Ammonia Lyase (PAL) (Figure 2).

Biomass
Phenological characteristics of O. miliaceum and P. setaceum (roots' and leaves' length, root biomass, leaves biomass) were determinate. In Table 2, there is the variation in plants' growth parameters under the influence of contaminated soil. Growth depression was observed for the plants in Group 1, in fact, the non-inoculated plants exposed to contaminated soil showed a marked decrease of length of roots and leaves. However, the contamination of TPHs on plant growth was more pronounced after 90-120 days. The maximum root and leaf length were achieved in plants in Group 2, with inoculum of autochthonous bacteria after 240 days in O. miliaceum (285 ± 15 cm and 118 ± 4 cm, respectively) and O. setaceum (291.6 ± 15 cm and 97.6 ± 5 cm, respectively).
This finding demonstrates that the plants inoculated with PGPR displayed greater growth response than non-inoculated control plants. No plant was dead, however, some of the plants presented signs of phytotoxicity, such as yellowing of leaves and slower growth, compared with the control, which is consistent with the findings of Wyszkowski and Ziólkowska [35] and Al-Baldawi et al. [36]. Therefore, the presence of microorganisms reduced the toxic effect produced by contaminated soil, because the rhizobacteria provide defense to the plants and at the same time, benefit from plant exudates and from the plant itself [3,37]. The rest of the phenological characteristics (root and leaf biomass) of O. miliaceum and P. setaceum inoculated with the microbial consortium were also improved ( Table 2). The morphology of the O. miliaceum and P. setaceum plants in response to different levels of TPHs was observed throughout the 240-day experimental period. The hydrocarbons in the soil can inhibit and reduce plant growth [38]. Plants in soil polluted with TPHs exhibited signs of phytotoxicity, with the death of some O. miliaceum and P. setaceum plants verified after 30 days. The roots' and leaves' biomass of O. miliaceum and P. setaceum were influenced by the TPHs treatments compared with the corresponding control, but the plants survived after 60 days of treatment. The dry biomass of O. miliaceum and P. setaceum significantly increased for all treatments in Group 2, as shown in Table 2.

Chlorophyll Content
The O. miliaceum and P. setaceum of Group 1 exhibited a significant decrease of Chl a, Chl b, and Chl a + b content in leaves (Table 3). This significant decrease is caused by the inhibition of biosynthesis of precursors of chlorophyll, loss of pigments, and disorganization of thylakoid membranes [39]. O. miliaceum and P. setaceum inoculated with bacteria (Group 2) had greater Chl a, Chl b, and Chla + b than the non-inoculated plants grown under the same stress (Table 3).

The Adsorption of TPHs by O. Miliaceum and P. Setaceum Tissue (Roots and Leaves)
The levels of TPHs in the plant tissue (roots and leaves) along with a summary of the statistical analysis are presented in Figure 3. The value of TPHs of O. miliaceum in Group 1 (plants without bacteria) ranged between 15 and 28 mg/kg in roots and 9 and 14 mg/kg in leaves, from 30 to 240 days. These results improve significantly with the inoculation of the consortium (Group 2), since there is a concentration of TPH in the roots of 48 and 28 mg/kg in the leaves referred to 240 days.
The value of TPHs in the leaves of P. setaceum in Group 1 (plants without bacteria) ranged between 19 and 31 mg/kg in roots and 8 and 12 mg/kg in leaves from 30 to 240 days. These results improve significantly with the inoculation of the consortium (Group 2) since there is a concentration of TPH in the roots of 45 and 19 mg/kg in the leaves referred to 240 days (Figure 3).
This result shows that the consortium also plays an important role in increasing the uptake of TPHs in the plant.
Mostly, the concentration of TPHs in plant samples was observed in the order: root > leaves. The content of TPHs in the plant samples shows that there is a good uptake of TPHs across root and leaf samples of O. miliaceum and P. setaceum. In rhizodegradation, contaminants will be degraded in the soil through the bioactivity exuded by plants or from soil organisms such as bacteria. The lower level of TPHs in leaves may have been caused by phytodegradation or phytotransformation of petroleum hydrocarbons, which occurs within the leaves of the plant itself [40].

Soil Analysis: Quantification of Petroleum Hydrocarbons and Dehydrogenase Activity
For the analysis of the TPHs levels in the three Groups of treatment, a specific analytical procedure was developed in order to identify five different classes on the basis of the number of carbon atoms in the hydrocarbon chain: low molecular weight hydrocarbons (C ≤ 12), aliphatic (C13-C18), aliphatic (C19-C36), aromatics (C11-C22), and high molecular weight hydrocarbons (C > 12). Low molecular weight hydrocarbons (C ≤ 12) were absent in all samples. The presence of aromatic compounds C11-C22 was never detected ( Figure 4). As is already known, long-chain hydrocarbons are less biodegradable and consequently, the residual observed contamination was reasonably made by the TPHs fraction most recalcitrant to biological actions. Moreover, a high number of carbon atoms leads to a reduction of both volatility and solubility in water, with a resulting decrease of natural mitigation of contamination. Reasonably, a fraction of branched chain compounds was also present, characterized by very slow biodegradation. All of this experimental evidence allowed for the identification of the residual TPHs fraction as particularly recalcitrant and difficult to be removed by a simple chemical-physical action. After 240 days, the plants of O. miliaceum and P. setaceum supplemented by the endophytic consortium (Group 2) had removed 94% of TPHs from soil, the plants of only O. miliaceum and P. setaceum removed 78% (Group 1). These results demonstrate that a successful strategy for overcoming the challenge of plant stress in phytoremediation of TPHs is to use plant growth-promoting rhizobacteria (PGPR) [4,8,[41][42][43][44][45]. Soil dehydrogenase activity was strongly related to removal of TPHs, indeed, the maximum value (67.8 ± 1.15 in O. miliaceum and 70.1 ± 0.05 in P. setaceum) was measured in Group 2 after 90 days (Table 4).
These results demonstrate a hold relationship between DHA and biodegradation of petroleumcontaminated soil [46][47][48][49]. Moreover, between 60 and 90 days, there was a positive relationship between DHA and root lengths; in other words, the root growth contributes to TPHs' degradation by stimulating the microbial activity of rhizosphere.

Discussion
The phytoremediation of soil contaminated by TPH, assisted by consortia, is based on the potential of the plant and on the degree of association with endophyte microorganisms able to change organic contaminants into harmless final products. Endophytes improve the efficacy of phytoremediation of organic pollutants, and they may also manifest inherent biodegradation abilities around or within host plants, increasing metabolic activities in the rhizosphere and in the endosphere. Thus, the integrated actions between host plants and endophytes are critical factors in their ability to clean up polluted soil [11,12,16]. Apart from the above-mentioned benefits, endophytes also promote plant growth by enhancing nutrient acquisition and absorption of water, bringing significant improvements to the development and growth, and causing less stress, as evidenced by the results of the analysis of the stress markers. These characteristics of increased plant growth by endophyte microorganisms facilitate adaptation to abiotic stress factors and increase the biomass of Oloptum and Pennisetum, making it suitable for phytoremediation. The absorption of TPH by two species of Poaceae is controlled by the hydrophobicity of these organic pollutants [50]. The plant roots before their degradation discreetly absorb hydrophobic complexes by microflora rhizosphere. In this context, the functional bioavailability grade of hydrophobia is important. It can become the real limiting factor of phytoremediation of recalcitrant organic contaminants in soils [51]. The biosurfactants released by microorganisms or plants can help in the removal of organic pollutants in soils by rising their bioavailability and uptake by plants or microbial degradation, and thus, improve the effectiveness of phytoremediation in situ [52,53]. In this work, we have magnified the effectiveness of phytoremediation of TPH in soil by increasing inoculation within the rhizosphere with endophytes makers of biosurfactants. For example, the root with the endophytes that belong to the genus Bacillus, secretes ecosurfactants and show a good emulsification activity, which are excellent biosurfactant properties for bioremediation of soil contaminated by TPH.
Therefore, it is realistic to assume that endophytes could produce biosurfactants in vivo, which facilitate degradation of organic pollutants in plant, leading to a better rhizodegradation of organic contaminants in soils [54]. Our results show exactly this tendency of rhizodegradation, probably acting on the bioavailability of pollutants and the related rhizosphere environment that has been formed. The choice to use high EPS-producing endophytes bacteria is aimed at the possibility of creating ideal and effective conditions for the degradation of TPH [55]. Our system is responsible for the effectiveness of phytoremediation. Oloptum and Pennisetum, in relation to the morphological composition of roots, are generally free from secondary growth with roots that have little to no lignification and the ability to quickly explore the soil, and in relation to their biochemical activity of the production of exudates, mucilage allows for the specialization of a highly effective phytoremediation rhizoplane microbial architecture. Root development of these interesting Poaceae in partnership with endophytes has almost doubled, and the functional redox biology activity demonstrates the metabolic skills of combining, detoxifying, and/or rhizodegrading pollutants. Our results demonstrate that the prospects of degrading hydrocarbons are concentrated at the rhizosphere level. In fact, the degradation of contaminants is extremely elevated at the rhizosphere level, where the production of a diversity of biomolecules produced at different levels from the endophyte-plant system plays a key role. The activated enzymatic processes and degraded genes produced are just one example of a co-metabolic perspective that trigger between plants and endophytes. The ability of the Poaceae, and especially Oloptum and Pennisetum, to develop an active and wide root and the capacity to host a high quantity of microorganisms both in the rhizosphere and in the endosphere, allow for effective activities of break down in hydrocarbons.

Conclusions
O. miliaceum and P. setaceum have never been studied for their phytoremediation characteristics of organic compounds nor is their great capacity for endophytic partnership known [54]. They were only once used in some heavy metal phytostabilization experiments in combination with biochar. These plants have a great ecological value, as they are a phytotype that naturally lives on poor and disturbed soils and stands up well, even in xerophytics environments. Also, in nature, they show a significant ecological value and an amazing ability to regenerate after the removal of leaves. They show a rapid and effective regenerative capacity (about 15 days from the cut) due to the considerable amount of meristems present at the base of the leaves, with a significant increase of biomass production. Our data also shows that O. miliaceum and P. setaceum have great adaptability to disturbed substrates and other physico-chemical composition.
The high plasticity and environmental adaptability are the foundation that will enable these plants to become an important model of bioremediation systems' development in drought environments and in poor and disturbed soils. The ability to quickly change various ecophysiological parameters, such as PAL and GST, and rebalance the photosynthetic efficiency, demonstrates the great biological plasticity of these plants. In addition, the high and diverse microbial population that is hosting the rhizosphere contributes synergistically to the species adaptation speed and performance-specific response to stresses. These features, together with minimum maintenance incidence (little water, the absence of serious diseases), are a prerequisite for O. miliaceum and P. setaceum to become new candidate species for bioremediation TPHs. The ecosystem service and the physiological response of the plants during the endophytic-assisted phytoremediation demonstrate our experimental model: Oloptum and Pennisetum show a great resolute value. Choosing bacteria that largely produce EPs, ACC, and siderophores guarantee the possibility of different strategies of attack on recalcitrant molecules like TPH. The addition of mycorrhizae fungi that naturally colonize these roots have magnified the rhizodegradation activity. In other words, the plant-endophytemicroorganism association should not be seen in a perspective of a summation of effects but a rearrangement of co-metabolism to consider, together, one big organism that reorganizes itself through signals and metabolic adjustments. In fact, more effective degradation of contaminants requires interactions to support growth and catabolic collaboration between plants and hypothetically thousands of degradative microbial taxa that certainly colonize the host. This is ultimately controlled to the concept of "Meta-organism" or Microbiome vegetable, mentioning the total of a host plant and its connected microbial population. Omics tools' accessibility has facilitated extensive surveys of biodiversity and functional characteristics of the endophytic community of Oloptum and Pennisetum. However, it still lacks studies in situ of endophyto-assisted phytoremediation at the metaorganism level.