Phytoremediation of Oil-Contaminated Soils by Combining Flowering Plant Cultivation and Inoculation with Acinetobacter junii Strain M-2

Oil contamination of the soil by petroleum products has become an enormous environmental problem. In this study, we examined whether remediation of oil-contaminated soils by cultivating three flowering plants (Mimosa, Gazania, and Zinnia) could be enhanced by inoculation with Acinetobacter junii strain M-2 at different plant growth stages (at sowing, at early growth, and at mid-growth). The growth of Zinnia cultivated in oil-contaminated soils inoculated at sowing was significantly superior to that in the non-inoculated soil. Although total petroleum hydrocarbon concentrations in soils inoculated at sowing were nominally lower than those in non-inoculated soils, especially in the case of Zinnia planting, the effect did not reach statistical significance. However, dehydrogenase activity was significantly higher in the soils inoculated with A. junii strain M-2 than in non-inoculated soils for all three plant species tested. These results demonstrate that a combination of ornamental plant cultivation (particularly Zinnia) and inoculation with A. junii strain M-2 increases the efficiency of oil-contaminated soil phytoremediation.


Introduction
Petroleum products are used for many purposes, e.g., as fuel for cars and heaters, and as lubricants of mechanical parts, so they are essential for daily life in most societies. There are many known incidents when soils have become contaminated with these oils, and it is likely that many more areas of oil-contaminated ter soil contamination by oil [6]. Contaminated soils in Japan are often found and treated at locations adjacent to urban areas, which is in contrast to the situation in the majority of oil-producing countries. In urban areas, it is desirable to use ornamental flowering plants for the phytoremediation of contaminated soils.
Ornamental plants have various advantages for remediation, as they also improve landscapes and have high aesthetic value. Various phytoremediation measures involving ornamental plants have been used to clean up oil-contaminated soils throughout the world [7] [8]. The use of the ornamental plant Mirabilis jalapa for the phytoremediation of oil-contaminated soils has been described previously [9] [10]. In particular, Peng et al. [10] grew plants for 127 days in soils contaminated with total petroleum hydrocarbons (TPH) at total concentrations of up to 10,000 mg/kg. They found that TPH concentrations decreased significantly faster in soils with plants than in soils without plants. Furthermore, growth rates of plants in contaminated soils were similar to those of plants grown in uncontaminated soils. Zhang et al. [11] grew Impatiens balsamina in oil-contaminated soil for 120 days and found that this plant species promoted the decomposition of persistent organic petroleum hydrocarbons, such as resin and asphaltene.
Bordoloi et al. [12] grew Axonopus compressus for 360 days in oil-contaminated soils and found that the addition of fertilizer enhanced both plant growth and remediation effects; the TPH level was reduced by 70% in fertilized soils compared to a 40% reduction in unfertilized-soils. Therefore, these results indicate that optimizing soil fertility by applying fertilizer also may enhance phytoremediation effects.
Studies that have used ornamental flowering plants for phytoremediation in Japan are rare. Of 33 ornamental plant species that grow well under Japanese environmental conditions, 3 species (Mimosa pudica, Gazania rigens, and Zinnia profusion) are known to grow well in oil-contaminated soil and have a remediation effect. In particular, species belonging to the genus Zinnia are the most effective for the phytoremediation of oil-contaminated soils [13]. In this study, among the four Zinnia species that we tested (Z. elegans, Z. angustifolia, Z. haageana, and Z. hybrida), we found that Z. hybrida was best suited to remediate contaminated soils. In addition, we found that "Profusion White" variety of Z. hybrida showed the highest growth and remediation rates in oil-contaminated soils [14]. Furthermore, we showed that phytoremediation of soils with higher levels of oil contamination could be enhanced by amending the soil with perlite and higher basal fertilizer levels [15]. Therefore, in such modified conditions, Z. hybrida "Profusion White" could be more effective over a wider range of areas.

T. Kai et al. Journal of Agricultural Chemistry and Environment
To further increase the phytoremediation capacity of Zinnia in oil-contaminated soils, we decided to examine the effects of inoculating rhizosphere soils with Acinetobacter junii strain M-2. Acinetobacter is a genus of gram-negative eubacteria that prefers wet environments, such as soil, and it is distributed widely in different environments. A. junii strain M-2 has been shown to be harmless to humans and plants [16] [17] isolated the A. junii strain M-2 from oil-contaminated soil in California, but their studies did not show whether this strain reduced TPH concentrations in the soil. Therefore, in this study, we investigated the effect of inoculation with A. junii M-2 on phytoremediation effects of three flowering plants, Mimosa, Gazania, and Zinnia in oil-contaminated soils.

Isolation of Acinetobacter junii M-2
One gram of rhizosphere soil around the roots of a Zinnia "Profusion White" plant was mixed with diesel oil. The soil suspension (0.  [18] and cultivated at 30˚C to obtain single colonies. Single colonies were transferred on LB slants. After checking growth of isolates, strain M-2, showing good growth, was selected for further experiments. A partial 16S rDNA fragment (about 1.5 kb) was amplified by PCR [19] using a cell suspension of strain M-2 as a template. The amplified fragment was cloned and sequenced as described previously [20] [21]. The determined sequence was searched at the DNA database of the DNA Data Bank of Japan (www.ddbj.nig.ac.jp/index-j.html) using a Blast program. As a result, strain M-2 was identified as Acinetobacter junii based on the highest identity of reference strains.

Preparation of Oil-Contaminated Soil
Kanto loam soil from the Kurokawa Experimental Field at the Meiji University in Japan was air-dried in a greenhouse until its moisture content was below 1% (w/w). Then, soil specimens were sterilized in an autoclave at 120˚C for 20 min.
Diesel oil was sprayed gradually with a pump sprayer while stirring the soil at 32 rpm with a soil mixer in a closed greenhouse to ensure even mixing. The contaminated soil was stirred once every 2 days in a closed greenhouse for 14 days, and then the remaining oil particles were volatilized [22]. The initial TPH concentration in the contaminated soil after volatilization was analyzed using a gas chromatograph hydrogen flame ionization detector (GC-17A GC-FID, Shimad-Journal of Agricultural Chemistry and Environment zu Corp. Ltd., Kyoto, Japan). A commercial solid fertilizer (N:P 2 O 5 :K 2 O = 10:9:8) was added to the oil-contaminated soil at a concentration of 0.2 g/L.

Experiment 1: Effect of the Timing of the Inoculation with Acinetobacter junii Strain M-2 on Plant Growth and Phytoremediation Effects
We

Experiment 2: Effect of the Inoculation with A. junii Strain M-2 on Zinnia Plant Growth and Phytoremediation Effects
Samples of oil-contaminated soil (200 g each, initial TPH concentration: 13,434 mg diesel/kg soil) and uncontaminated soils containing basal fertilizer were placed in vinyl pots (φ 100 × 120 mm). Only Zinnia plants were used in Experi- , not planted + not inoculated (NP-NI plot), not planted + not inoculated + not irrigated (D plot). Treatments for uncontaminated soils were as follows: planted + inoculated (NC-P-I plot) and planted + not inoculated (NC-P-NI plot). All pots except for those in the D plot were irrigated with 30 mL of water every 2 -3 days, which was sufficient to keep the soil surface moist, while avoiding any run-off from the pot. A commercial liquid fertilizer (N:P 2 O 5 :K 2 O = 6:10:5) was applied weekly, instead of water (30 mL of the 500-fold diluted fertilizer). In total, 15 pots per treatment were used. At 45, 90, and 135 DAS, five pots per treatment were selected randomly for the analysis described below. TPH concentrations and soil DHA were measured. The height of five plants in each pot was measured, giving 25 plant height measurements per treatment. All roots and shoots were harvested from each pot and roots were washed carefully. The total dry weights of shoots and roots in each pot were measured after oven drying at 80˚C for 3 days.

Analysis of Soil TPH Concentration
TPH levels were measured according to the guidelines issued by the Ministry of the Environment of Japan for preventing oil pollution in 2006 [23]. After removing the plant roots from each pot (apart from those in the unplanted group), the soil was mixed, and 30 g of it was dried at 30˚C for 4 days. A soil sample (5 g) was then placed in a 50 mL conical flask with 15 mL of carbon disulfide, and the mixture was shaken for 30 min. The supernatant was separated from the residue 2 h after shaking. Then, 15 mL of carbon disulfide was added to the residue, the mixture was shaken for 30 min, and the supernatant was separated 1 h after shaking. This process was repeated one more time, and the three supernatants were combined. The recovered supernatants were diluted to 50 mL and filtered through a 0.45-μm filter. Then, 1 μL of the filtrate was injected into a gas chromatograph hydrogen flame ionization detector (GC-FID, GC-2010, Shimadzu Corp. Ltd., Kyoto, Japan) for analysis. The gas chromatography system was equipped with an Intercap IMS capillary column (liquid phase, 5% phenylmethyl silicon, 30 m × 0.32 mm internal diameter, 0.25-μm film thickness; GL Science Inc., Japan) and a flame ionization detector. The injection and detector temperatures were both 320˚C. The heating program was set to maintain 35˚C for 5 min and then increased to 320˚C at a rate of 10˚C/min. Helium was used as the carrier gas in the splitless mode. The analysis was repeated three times for each sample.

Analysis of Soil DHA Level
The soil DHA level was determined according to the method of [24] Plant roots were removed from the pots in which plants were grown (apart from those in the unplanted group). The soil from each pot was stirred, and a 30 g soil sample was transferred to a 100 mL test tube. Next, 1.0 mL of 0.25 mmol Tris-hydrochloric buffer solution (pH 6.8), 200 μL of 0.4% 2-(4-iodophenyl)-3-(4-nitrophenyl)-5phenyl tetrazolium chloride (INT), and 50 μL of 1% glucose were added per 1 g Journal of Agricultural Chemistry and Environment of soil. The mixture was sealed tightly and incubated for 24 h in the dark at 30˚C. Methanol (10 mL) was added to stop the enzyme reaction, and the mixture was stirred for 1 min using a vortex mixer. The mixture was allowed to settle for 10 -15 min, and then, the supernatant was filtered. Approximately 4 mL of filtrate was analyzed using a spectrophotometer (UV-1700, Shimadzu Corp., Ltd., Kyoto, Japan) at a wavelength of 485 nm, and the rate of conversion of INT to iodonitrotetrazolium formazan was determined. Each filtrate was measured three times.

Statistical Analysis
Data are presented as mean±SD values per groups of five pots and analyzed using Excel Statistics 2012 for Windows (Social Survey Research Information Co., Ltd., Tokyo, Japan). All data were analyzed by using one-way analysis of variance followed by the Fisher's least significant difference test, where appropriate. All statistical analyses were conducted with a significance level of α = 0.05 (P < 0.05).

Effects of Inoculation Timing on Plant Growth
The effects of the timing of inoculation with A. junii M-2 on plant height as well as shoot and root dry weights at 90 DAS are shown in Table 1. In Mimosa, inoculation timing did not have a significant effect on plant height. However, shoot and root dry weights were lower in all inoculated groups than those in the non-inoculated group. In Gazania, there was no significant effect of treatment on plant height either. The shoot and root dry weights in inoculated groups were heavier than those in the non-inoculated group. Plant height of Zinnia was shorter in groups inoculated during early-and mid-growth than that in the non-inoculated group. However, in the group inoculated at sowing, Zinnia plant height was about 1.3 times taller than that in the non-inoculated group. Furthermore, the shoot and root dry weights in the sowing treatment group were significantly heavier than those in the non-inoculated group. Thus, inoculation

Influence of Inoculation Timing on Soil TPH Concentration
There were no significant effects of inoculation timing on soil TPH concentrations in oil-contaminated soils planted with Mimosa, Gazania, or Zinnia ( Figure   1). However, soil TPH concentrations in all treatment groups were significantly lower than the initial soil TPH concentration. In addition, soil TPH concentrations of inoculated groups were lower than those of non-inoculated groups for all three plants. TPH concentrations in the groups inoculated at sowing treatment were nominally lower than in the groups inoculated at other points. Furthermore, among the groups inoculated at sowing, the lowest TPH concentration was observed in Zinnia.

Influence of Inoculation on Zinnia Growth
Zinnia plant height in oil-contaminated soils (P-I and P-NI plots) at 45 DAS was significantly lower than that in non-contaminated soils (NC-P-I and NC-P-NI plots), but at 90 and 135 DAS there were no differences in plant height among all plots ( Table 2). Inoculation with A. junii M-2 did not affect plant height. However, the shoot and root dry weights in the inoculated NC-P-I and P-I plots were heavier than those in the non-inoculated NC-P-NI and P-NI plots throughout the 135-day growing period.

Influence of Inoculation on Soil TPH Concentration
Inoculation with A. junii M-2 did not significantly affect soil TPH concentra-

Influence of Inoculation on Soil DHA Level
Soil DHA levels in the oil-contaminated soils (P-I, P-NI, NP-I, and NP-NI plots) were higher than those in the non-contaminated soils (NC-P-I and NC-P-NI plots) at 45, 90, and 135 DAS (Figure 4). Furthermore, soil DHA levels in the plots with planted Zinnia (P-I and P-NI plots) were higher than those in the plots without plants in oil-contaminated soil (NP-I and NP-NI plots) at 90 and

Discussion
In the present study, we sought to establish if the phytoremediation activity of ornamental plants in oil-contaminated soils could be enhanced further by the inoculation of the soil with a widely distributed Acinetobacter species. We found that Zinnia growth was greatly influenced by the inoculation of oil-contaminated soil surface with A. junii strain M-2. Plant height, shoot dry weight, and root dry weight were significantly higher upon inoculation at sowing. There were no significant effects of inoculation timing on TPH concentrations in the soils planted with Mimosa, Gazania, and Zinnia plants, although there was a greater tendency of lower TPH concentrations in the inoculated groups than in the non-inoculated groups, particularly, in the Zinnia group inoculated at sowing. In addition, soil DHA levels were higher in the inoculated groups than those in the non-inoculated groups, again, particularly in the Zinnia group inoculated at sowing.
In the previous studies, it is reported that many kinds of bacteria possess some alkane degrading pathways and can therefore utilize oils and alkanes [25].
Strains of Acinetobacter were reported to be excellent alkane degraders [26] [27]. For example, Acinetobacter sp. strain DSM 17874 can grow on n-alkanes ranging from decane (C 10 H 22 ) to tetracontane (C 40 H 82 ) as a sole carbon source [26]. Although we did not analyze the chain length of the residual diesel oil in soil samples after phytoremediation treatment using A. junii strain M-2, the oil-degrading ability of A. junii strain M-2 seems to be enough to remove the diesel oil contaminant.
Kubota et al. [28] reported that the amount of root exudations varies depending on plant species. When A. junii strain M-2 was isolated from the rhizosphere of cultivated Zinnia, it was found to populate the exudates of Zinnia roots. It appears that increased contact with both A. junii strain M-2 and exudates from Zinnia roots potentiated Zinnia growth. Sun et al. [29] reported that the level of soil DHA and the number of bacteria in diesel-contaminated soil inoculated with Gordonia sp. S2RP-17 were significantly higher than that in non-inoculated contaminated soil. Jing et al. [30] reported that the level of soil DHA and the number of bacteria in the soil contaminated by polycyclic aromatic hydrocarbons were higher if it was planted with alfalfa than if it was in soil without this plant. In our experiments, soil DHA levels in the soils with planted cultivated Mimosa and Gazania were not affected by the timing of inoculation with A. junii strain M-2. Soil DHA levels in the inoculated groups were higher than those in the corresponding non-inoculated groups, but there was no significant relationship between the increase in soil DHA levels and inoculation timing. These observations suggest that A. junii M-2 might be more specific to Zinnia or that the exudates from the roots of Mimosa and Gazania do not promote substantial Root growth forms pores in the soil, improves water and air permeability [31], and promotes the contact of insoluble pollutants with soil particles and microorganisms [32]. Furthermore, the root surface is an ideal environment for the growth of microorganisms because root exudates provide abundant nutrients, such as carboxylic acids, amino acids, proteins, and sugars [33]. It has been previously shown that soil inoculation with oil-borne bacteria improves plant growth and enhances the remediation of soil polluted by oil. For example, Huang et al. [1] reported that Enterobacter cloacae strains UW4 and CAL2, which promoted plant growth, reduced TPH concentrations when they were inoculated into soils planted with Festuca arundinacea. In addition, Tang et al. [34] reported reductions of TPH concentrations in oil-contaminated soils when cotton (Gossypium hirsutum) was planted after the inoculation of plant growth promoting microorganisms, such as A. brasilence, A. radioresistens, or R. erythropolis. Furthermore, Sun et al. [29] reported that the growth of Zea mays L. and soil characteristics were improved after inoculation of oil-contaminated soil with Gordonia sp. strain S2RP-17. Therefore, these studies show that the combined use of a specific plant and a fungal microbial strain can improve the efficiency of phytoremediation. In our study, we demonstrated that using a combination of Zinnia plants and inoculation with A. junii M-2 improved remediation efficiency as indicated by lower TPH concentrations in oil-contaminated soils. A. junii strain M-2 likely promoted the stronger growth of Zinnia roots, thereby increasing the level of soil DHA and reducing TPH concentrations, compared to its effects in combination with other plants. This study shows that A. junii strain M-2 could be a promising microorganism for enhancing the phytoremediation of diesel-contaminated soils, particularly in combination with Zinnia plant cultivation.

Conclusion
This study showed that Acinetobacter junii strain M-2 enhanced the phytoremediation of oil-contaminated soils using Zinnia plants, when it was applied at sowing. This combined bacterial and phytoremediation approach is inexpensive and environmentally friendly. However, because sterilized soils were used in this study to examine phytoremediation effects, the A. junii strain M-2 could have been in an unnaturally favorable environment. Further studies should be conducted using unsterilized soils to validate the practical application of this treat-