TRANSFER OF METALS FROM CRUDE OIL IMPACTED SOILS TO SOME NATIVE WETLAND SPECIES , THE NIGER-DELTA , NIGERIA : IMPLICATIONS FOR PHYTOREMEDIATION POTENTIALS

In this study, wetland species growing naturally in the surrounding of two crude oil facilities were sampled and screened for their phytoremediation potentials for zinc, lead, nickel, chromium and cadmium in soil. Concentrations of metals in the root and shoot samples of the wetland species alongside the rhizosphere soil were determined. Metal accumulation in wetland species exceeded the permissible limits, but it was still within phytotoxic thresholds except for chromium. The use of a bioconcentration factor and a transfer factor to screen the wetland species for phytoremediation potentials identified six out of the eight studied species as multi-elemental phytostabilizers of metals in soil. In addition, five of the eight wetland species displayed potentials for phytoextraction of metal, though there was no multi-elemental phytoextractor among the wetland species. Paspalum vaginatum, Andropogon tectorum and Kyllinga squamata portend potential abilities to phytoextract nickel. In addition, Setaria longiseta and Pteridium aquilinum also showed strong potential to phytoextract lead and cadmium respectively from soil. This screening assessment is hoped to be useful in the applications of a cost-effective green technology to remediate heavy metals in contaminated soil.


Introduction
Metal-contaminated soils are known to be difficult to remediate (Lasat, 2000); and cleaning-up of such soils via conventional engineering methods can be relatively expensive (Salt et al., 1995).The need for less expensive clean-up technologies has led to the emergence of the concept of phytoremediation which Clement O. Ogunkunle et al. 182 seems to be a cost-effective alternative in moderately contaminated areas (Weis and Weis, 2004).The use of plants to extract and remove heavy metals from soil was first documented in Thlaspi caerulescens and Viola calaminaria in the late 19 th century (Baumann, 1885).In recent years, the idea of using plants to extract metals from contaminated soil has been reintroduced by Chaney (1983) and Baker et al. (1991).In assessing plants for phytoremediation, one of the important features of potential plants is the ability to accumulate high concentrations of metals in their shoot biomass even when metals are present at low concentrations in the soil (Salt et al., 2000;Baker and Whiting, 2002).In addition to metal concentration in shoots, two other important factors that need to be considered are the bioconcentration factor (BCF) and translocation factor (TF) (Ma et al., 2001).The BCF is the ratio of metal concentration in plant root to metal concentration in the soil.This approach is important for phytostabilization, a situation where plants are used to immobilize metals and store them in underground biomass (roots).TF is the ratio of metal concentration in plant shoot to the concentration in root and it determines the efficiency of a plant to translocate metals from the root to the shoot (Ma et al., 2001;Varun et al., 2012).
The TF has been used to characterize plant species as either accumulators or excluders (Baker, 1981).Furthermore, it has also been used to define the ability of plants to remove metals from the soil solution and concentrate them in aboveground tissues (phytoextraction) in plants (Baker, 1981;Yoon et al., 2006).Plant species that exhibit BCF and TF values that are greater than 1.0 are considered potential phytoextractors (Fitz and Wenzel, 2002); whereas plant species with high BCF (BCF>1.0)and low TF (TF<1.0)are referred to as potential phytostabilizers (Mendez and Maier, 2008;Garba et al., 2013).The option of using wild or ruderal species to remediate soils is an attractive option.It does not require or involve any agronomic inputs, but gives significant positive results for both site remediation and ecological restoration.A great number of native species such as Sesbania spp, Crotalaria spp, Calotropis procera, Vetiver, lemon grass, Sida acuta, Pennisetum purpureum etc. have been documented for phytoremediation purposes (Yang et al., 2003;Uraguchi et al., 2006;D'Souza et al., 2010;Varun et al., 2011;Ogunkunle et al., 2013).
There are limitations to the study of phytoremediation, and one critical aspect of the limitation is the use of ex-situ study to assess phytoremediation potentials of plants which does not always portray on-field conditions (Zhuang et al., 2007;D'Souza et al., 2010).Native or wild species are preferred for phytoremediation because they are often better than introduced species in terms of survival, growth and reproduction under environmental stress and require no agronomic inputs (Yoon et al., 2006;Antonsiewicz et al., 2008;D'Souza et al., 2010).It is necessary to consider native plants for phytoremediation because of their non-invasiveness (Jiménez et al., 2011;Concas et al., 2015).
The present study is an effort to screen wild species growing in soils surrounding two crude oil facilities for phytoremediation potential.The objectives are to (i) assess the level of accumulation of Zn, Pb, Ni, Cr and Cd in root and shoot tissues and (ii) evaluate the feasibility of usage of the species for phytoremediation based on their metal partitioning in tissues.

Study area and collection of native plants
The study locations are the surroundings of Shell Petroleum Development Company (SPDC) flow stations in Kokori and Kolo Creek, Niger Delta, Nigeria (Figure 1).The description of the two study locations has been reported in our earlier study (Fatoba et al., 2016).Systematic random sampling described in Fatoba et al. (2016) was employed to collect plant and soil samples around the crude oil facilities.Plant species encountered at every sampling point were collected and properly tagged in polyethylene bags.All plant samples were cleaned of soil debris and later identified at the Herbarium of the Department of Plant Biology, University of Ilorin, Nigeria.Four plant species that were identified to be common were utilized in this study.The plant species with their corresponding voucher numbers are listed as follows: Pteridium aquilinum (UIH005/1118), Synedrella nodiflora (UIH003/1116), Mimosa pudica (UIH007/1119), and Andropogon tectorum (UIH006/1119) were collected at the Kolo creek oil facility.In addition, Kyllinga squamulata (UIH001/1114), Setaria longiseta (UIH004/1117), Ludwigia abyssinica (UIH008/1120) and Paspalum vaginatum (UIH002/1115) were collected at the Kokori oil facility.Samples (≥8 samples for each plant species) were separated into shoots/fronds and roots/rhizomes, washed with distilled water to remove soil/dust particles and later oven-dried to constant weight.Soil samples were air-dried, pulverized into a powdery form in a ceramic mortar in the laboratory and later sieved using a 2-mm mesh.

Chemical analysis of plant and soil samples
Physico-chemical variables of the soil of the two study locations have been reported in Fatoba et al. (2016).Soil pHs at Kokori and Kolo Creek were 5.39 and 5.04 respectively; and soil organic matter contents were reported to range from 0.14% to 2.00% and from 0.03% to 1.00% respectively (Fatoba et al., 2016).One gram of powdered soil sample was digested in a mixture of HNO 3 +HCl+H 2 O 2 (4:2:1 v/v) according to USEPA 3050B method (USEPA, 1996).The solution resulting from the digestion was filtered using Whatman No. 42 filter paper, and concentrations of Zn, Pb, Ni, Cr and Cd were determined using Flame Atomic Absorption Spectrophotometer (FAAS-Perkin-Elmer A Analyst 200).In addition, plant sample was analyzed according to the dry ashing method described in D 'Souza et al. (2010) with slight modification.Briefly, 5 g of each powdered plant sample was ashed at 450 o C in a crucible for 4 h and boiled in 15 ml of HCl for 30 min on a hot plate.Concentrations of Zn, Pb, Ni, Cr and Cd in the filtrate were determined using the FAAS-Perkin-Elmer A Analyst.Quality control was performed by conducting replicate digestion of samples and measurement of concentrations was done in triplicates.Certified reference materials (IAEA 359cabbage and IAEA-SL-1-soil) were also used to confirm the precision of the analytical procedures.Percentage recovery of elements in the certified reference materials ranged from 89% to 105% and from 78% to 98% for plant and soil respectively.

Statistical analysis
Significance of means of parameters was tested at P<0.05 and figures were presented using Origin 7.0 software.Models employed in the study for assessing partitioning of metals are the bioconcentration factor (BCF) and transfer factor (TF).

Results and Discussion
Level of trace metal in the soils Concentrations of trace metals in the soils of the two locations are presented in Table 1.Average levels of soil Zn and Cd from the two locations exceeded the normal crustal value (Lindsay, 1979), whereas concentrations of Pb, Ni and Cr were below the normal crustal values.This is not surprising as Zn and Cd are reported as major constituents of crude oil (Dickson and Udoessien, 2012), and could enrich soil through oil spills.Furthermore, levels of the trace metals from the two locations, except for Cd at Kolo Creek, were below the Canadian Council of Ministers for Environment (CCME) limits (CCME, 2007).And none exceeded the maximum permissible concentrations for agricultural soils in some European countries (Kabata-Pendias and Pendias, 2001).Though the concentrations of the majority of the studied trace metals did not exceed several allowable limits, it is worthy to note that Fatoba et al. (2016) reported low pollution of soil by the index of accumulation in these locations in relation to the background metal concentrations.Level of metal in tissue biomass of the native species Samples of the eight common native flora (4 species each from Kolo Creek and Kokori) were investigated in this study for phytoremediation potentials (Table 2).Metal partitioning in root and shoot tissues of the native species from Kolo Creek and Kokori is presented in Figures 2 and 3.In Kolo Creek, all the studied species showed significantly higher Zn concentrations (P<0.05) in root tissues than in shoot tissues (Figure 2).However, Pb, Ni, Cr and Cd were equally partitioned in both the shoot and root tissues in all the studied native species in Kolo Creek except for M. pudica that had greater Cr concentration in the shoot tissue than in the root tissue (p<0.05).The allocation of equal concentrations to the two tissuetypes may suggest tolerance strategy of coping with elevated concentrations in tissue biomass.In species from Kokori, a different partitioning pattern of metals was observed (Figure 3).Zn partitioning in the root and shoot of K. squamata and L. abyssinica showed no significant difference (p>0.05).However, Zn concentration in the root tissue of S. longiseta was significantly higher than shoot concentration, whereas shoot concentration was significantly greater in P. vaginatum (Figure 3).No significant difference was observed between root and shoot Pb concentrations in all the species except for K. squamata and P. vaginatum that displayed a greater accumulation of Pb in the shoot and root tissues respectively (P<0.05).Also, Ni and Cd concentrations in the tissues of the studied species showed no significant difference (P>0.05).Contents of Cr in the tissue biomasses showed no significant difference in all the species at Kokori except for S. longiseta that accumulated a significant amount (P<0.05) in the roots.The varied patterns of accumulation/partitioning observed for some metals may be based on the varying physiological and biochemical processes that exist in different plant species.This could result in differential accumulation and partitioning patterns in different plant species.For instance, transpiration is a driving force for metal transport from root and transpiration and xylem transport have been reported as major driving forces that promote efficient translocation of Cd in plants (Uraguchi et al., 2009).Significantly, all the species exhibited higher concentrations of metals in both the root and shoot tissues as compared to the soil.Zn seems to be partitioned mainly in the roots of most of the plant species in the two locations except for P. vaginatum at Kokori that accumulated a significant amount in shoots.This trend is supported by the assertion that root system serves as the primary target in metal toxicity in plants; and most plant species make efficient translocation of metals from roots to shoots to survive metal toxicity (Baker et al., 1994;Kuper et al., 2000).So, the presence of significant amounts of Zn in the root tissues rather than in the shoot tissues may be an indication that Zn concentrations in the roots were still within tolerable thresholds and have not reached phytotoxic levels in these species.Furthermore, concentrations of Zn in the root and shoot compartments separately were higher than tissue contents of all other metals.It may be posited that the high levels of Zn in the partitioned tissues may be linked to nutrient demands/requirements as it is an indispensable element for plant growth and also required as an essential constituent of numerous proteins and enzymes (Kupper and Kroneck, 2005).The highest concentration of Zn in the above-ground biomass was recorded in L. abyssinica (Zn >200 mg/kg), while the least was observed in P. aquilinim (Zn <75 mg/kg).These concentrations were higher than the soil concentrations at the two locations and showed that there was significant translocation in the species studied.Zn binds more efficiently with transporter proteins than with other metals especially in aquatic/wetland species where they use the enzyme carbonic anhydrase for converting bicarbonate to carbon dioxide (Kupper and Kroneck, 2005).This enhances easy translocation into the above-ground biomass and greatly enhances biomass accumulation.
The observed equal allocation of Pb, Ni and Cd contents in both the root and above-ground biomasses of all the species in this study could possibly be a strategy to withstand metal toxicity as metal concentrations in individual tissue compartments were within the phytotoxic thresholds (Table 1).In addition, accumulation of potentially toxic metals such as Pb and Ni in plants could also be part of potential defense mechanisms against browsing herbivores as reported by Liphadzi and Kirkham (2005).Concentrations of Cr in both the root and aboveground biomasses of all the studied species were below the phytotoxic level in plants ( Kabata-Pendias and Pendias, 1984) with variations in pattern of accumulation between the two tissue parts in the studied species.The differential pattern of Cr accumulation in all the species could be adduced to the effect of plant species type.The ability of plant species to accumulate a particular metal is selective and it varies (Alloway et al., 1990;Secu et al., 2008;Lacatusu et al., 2009;Varun et al., 2015).The reported low pH and organic matter of the soil (Fatoba et al., 2016) in the study locations may play a role in enhancing phytoavailability of metals in the soil, resulting in high plant uptake as biomass contents of all metals were higher than the soil contents.A morphological examination of the studied species showed no symptoms of metal toxicity.This is suggestive of plant's inherent potential to naturally adapt to metals in the soil.
Moreover, many studies have shown that metal toxicity does not generally arise in native flora as they are naturally adapted to locally elevated levels of metals over a period of time (D 'Souza et al., 2012;Varun et al., 2012;Zhang et al., 2013).Native species also generally develop the ability to withstand high concentrations of heavy metals in the soil with differential accumulation patterns (Pratas et al., 2013).

Screening plants for metal phytoremediation potentials
Plants can be grouped into accumulators and excluders based on concentrations of metals in the root and above-ground biomasses (Baker, 1981); such plants with BCF less than 1.0 can be categorized as excluders while those with TF greater than 1.0 can be classified as accumulators.In this study, BCFs>1.0 were observed in P. aquilinum for Zn, Pb, Ni and Cr; S. nodiflora (Pb, Ni, Cr and Cd); M. pudica (Zn, Pb and Cd); A. tectorum (Zn, Pb and Cd); L. abyssinica; S. squamata (Cd); S. longiseta (Zn and Ni); L. abyssinica (Pb, Ni, Cr and Cd) and P. vaginatum (Pb and Cd).These native wetland species can be classified as potential excluders as they were able to conveniently accumulate metals in the root biomass, but translocation to the above-ground parts was restricted.Baker (1981) reported that excluder plants can accumulate metals in their roots, but restriction of transport from the roots to the shoots is ensured as a control mechanism.In considering a mechanism for accumulator plants, metals should be concentrated in the aboveground biomass either in low or high soil metal levels (Baker, 1981); thereby displaying TF that is greater than 1.0.In this study, P. aquilinum exhibited TF that was conveniently greater than 1.0, hence it could be regarded as a potential accumulator of Cd.In addition, obtained TFs for Pb in K. squamata and S. longiseta were above 1.0, so these two species can be considered as potential accumulators of Pb.M. pudica, A. tectorum, K. squamata and P. vaginatum displayed TFs >1.0; thereby they can be regarded as potential accumulators of Ni.Though the TFs displayed by L. abyssinica for Zn (TF=1.4),Ni (TF=1.3)and Cr (TF=1.5)were above 1.0, it cannot be succinctly classified as an accumulator for these metals because the values were relatively low and just a little above 1.0.
Plants exhibiting the high bioconcentration factor (BCF) and the low transfer factor (TF) are generally considered potential phytostabilizers (Mendez and Maier, 2008;Garba et al., 2013).Such plants are able to mobilize metals from the rhizosphere soil into the root system, but there is always reduced translocation to above-ground tissues in such plants.Though none of the wetland species showed a promising potential as metal hyper-accumulators, some interesting observations were noted.In the study, based on the average BCF values that are greater than 1.0 and TF values<1.0,it is observed that all the species seem to be multi-metal phytostabilizers (Table 3).P. aquilinum showed potential to phytostabilize Zn, Pb, Ni and Cr in the soil.This is not surprising as this wetland species has been noted to be one of the plants that are ubiquitous in contaminated soils, particularly in mining areas (Kubicka et al., 2015;Samecka-Cymerman et al., 2012).Kachenko et al. (2007) also reported that translocation of some metals was limited in P. aquilinum because of absorption and retention in roots, suggesting an exclusion mechanism as part of its tolerance to the metals which is a major factor in phytostabilization.S. nodiflora seems to phytostabilize Pb, Ni and Cr and A. tectorum possesses the potential to stabilize Zn, Pb and Cd.It is also observed that L. abyssinica and P. vaginatum can potentially phytostabilize Pb, Ni and Cd; whereas M. pudica possesses the potential to phytostabilize Zn, Pb, and Cd while K. squamata can phytostabilize only Cd (Table 3).In these species, nominated as possible phytostabilizers of metals, there is less translocation of metals to the above-ground biomass/tissues, which could be due to the immobilization of metals in roots by vacuolar sequestration or cell wall binding, thereby preventing interaction with high molecular weight compounds serving as transporters in the cell cytoplasm (Salt et al., 1995).Phytostabilizers have the capabilities to mobilize metals in the root zone at the concentrations present in the soil at all sites and at the same time restrict the translocation of the same to above-ground parts (shoot).Parent species type and soil characteristics have been reported to be major factors that control heavy metal distribution and absorption by plants in soils (Xu and Tao, 2004).The soil with high pH enhances binding of metals to organic matter and reduces metal mobility in soil (Xu and Tao, 2004).Clement O. Ogunkunle et al. 194 Plants that are considered as phytoextractors exhibit BCF value>1.0 in addition to TF>1.0 (Fitz and Wenzel, 2002); thereby allocating a greater proportion of metal absorbed into the above-ground biomass.Among the 8 species screened for phytoremediation potential, 5 species can be referred to as phytoextractors in this study based on the screening index (Fitz and Wenzel 2002).P. aquilinum had BCF Cd =3.4 and TF Cd =2.6, and it can be referred to as Cd phytoextractor in this study.Also in this study, A. tectorum (BCF Ni =1.9 and TF Ni =3.3), K. squamata (BCF Ni =2.0 and TF Ni =15.6) and P. vaginatum (BCF Ni =8.4 and TF Ni =7.6) can be described as extractors of Ni in the soil.According to Ashraf et al. (2011) that put TF>10.0 as hyper-tolerance property, K. squamata can be referred to as hypertolerant species to Ni in the soil.S. longiseta is the only species with the potential to phytoextract Pb with BCF Pb =3.5 and TF Pb =2.1.These species may have become normal and naturalized components of the contaminated soil due to their multielemental phytoextraction potentials as previously opined by Varun et al. (2012).Their physiological and metabolic activities may have been acclimatized to the level of metal in the soil and probably have evolved tolerance traits.It has been reported that metal tolerance in wild plants could possibly be dependent on internal detoxification (Nouri et al., 2009) or selective pressure from metalliferous soils (Becerra-Castro et al., 2009;Quintela-Sabarís et al., 2012).The high biomass of most of these species is an added advantage for effective metal phytoextraction.

Conclusion
The observed elevated concentrations of metals in the studied plant species that exceeded the soil levels are an indication that the plants readily take up metals but possess strategies for metal-tolerance.In this study, we have been able to identify P. aquilinum as the only accumulator of Cd while M. pudica, A. tectotum, K. squamata and P. vaginatum can potentially accumulate Ni. S. longiseta and K. squamata can also potentially accumulate Pb from soil either in low or high soil concentrations.The potential ability of these species to accumulate these specific metals in their biomasses has also put them in the category of potential phytoextractors as they can conveniently accumulate the identified metals in the above-ground biomass.The study has also been able to identify P. aquilinumas excluders of Zn, Pb, Ni and Cr; S. nodiflora, (Pb, Ni, Cr and Cd); M. pudica (Zn, Pb and Cd); A. tectorum (Zn, Pb and Cd); L. abyssinica; S. squamata (Cd); S. longiseta (Zn and Ni); L. abyssinica (Pb, Ni, Cr and Cd) and P. vaginatum (Pb and Cd).Considering the ecotoxicological aspect, harvesting of biomass of these native plants could be an appropriate strategy to reduce the transfer of these metals along the food chain via ingestion by herbivores as all the species exhibited the phytoextraction potential for a particular metal.This would reduce or perhaps avoid the potential health risk of the metals in humans.
In this study, preliminary screening of these wetland native species for phytoremediation potentials based on in situ assessment has been provided.It is suggested that further study should be carried out ex situ; for instance, pot experiment in either screen house or greenhouse to ascertain and confirm the identified phytoremediation potentials of the screened species in this study.In addition, further field study is needed, as the soil metal concentrations were mostly below the maximum allowable values.

Figure 1 .
Figure 1.Map of Niger Delta region of Nigeria showing the study locations (adapted from Fatoba et al., 2016).

Figure 2 .Figure 3 .
Figure 2. Concentrations of metals (mg/kg) in different tissues of wetland species in Kolo Creek, Niger Delta, Nigeria (bars represent the standard deviation).

Table 1 .
Level of trace metals (mg/kg±SD) in study locations and recommended limits in soil and plants.

Table 2 .
Species of native flora encountered around the two crude oil facilities.

Table 3 .
Bioconcentration factor (BCF) and transfer factor (TF) of metals in native wetland flora