Introduction

Soil petroleum contamination is hugely prevalent in and around oilfields in Nigeria. It poses a severe danger to the environment when it is not adequately degraded, hence the need for explicit knowledge of microbial communities in contaminated sites and their dynamic roles in transforming the surrounding ecosystem (Mafiana et al. 2020). The most common causes of land oil spills are operational failures in and around oilfields, rupture of transit pipelines and storage facilities, and theft (Babatunde et al. 2018). A few of the numerous challenging ordeals suffered by host communities from poorly remediated spill sites are hampered farming activities (Albert et al. 2018) and spill extension into streams and groundwater (Adesipo et al. 2020).

In an earlier investigation by the United Nations Environmental Programme (UNEP) in 2011, over 69 contaminated sites in 200 locations of Ogoniland in the Niger Delta region were observed to have toxic compounds of benzene, toluene, ethylbenzene, xylene (BTEX), methyl tert-butyl ether (MTBE), and other hydrocarbons of lower molecular weight (LMW). These toxic compounds from poorly remediated sites were leaching into surface and groundwater, vegetation, and sediments (UNEP 2011). According to reports, failed remediation sites that had adopted the Remediation by Natural Attenuation (RENA) approach relied on the degradation capabilities of naturally inhabiting microorganisms by tilling and windrow stimulation. Besides, 10 out of 15 remediated sites still had contaminant residues leaching into groundwater (UNEP 2011; Sam and Zabbey 2018). Studies on the ecological hazards from spilled areas and poorly remediated sites have recorded bioaccumulation in fish (Nematopalaemon hastatus) amounting to 2585.62 mg kg−1 (Akinola et al. 2019) and an increase in neonatal and infant mortality by 38.3 deaths per 1000 live births (Bruederle and Hodler 2019) among other numerous health risks associated with soil petroleum contamination (Chinedu and Chukwuemeka 2018; Iwegbue et al. 2020).

The microbial community structure and function are products of its surrounding ecosystem, altered by the hazardous epoxides and toxic effect of petroleum on soil microorganisms (Sam and Zabbey 2018). Several studies on spill site remediation approaches have observed the predominance of different microbial phyla such as Proteobacteria, Acidobacteria, Firmicutes, Fusobacteria, Ascomycota, and Basidiomycota that play significant roles in hydrocarbon degradation in petroleum-contaminated soils in Nigeria (Obi et al. 2016; Ezekoye et al. 2018a, b; Obieze et al. 2019a; Okafor et al. 2021). While some petroleum compounds are easily degradable, most of their other components are toxic to microorganisms and could cause sporulation or total dormancy of a non-tolerant group and significantly alter the soil’s microbial community structure (Oyetibo et al. 2017; Ventorino et al. 2018). The relative abundance of hydrocarbon-degrading microorganisms often differs from one contaminated site to another within the Niger Delta region (Mafiana et al. 2020), hence the crucial need to understand the microbial community structure specific to contaminated sites in and around oilfields, the soil physicochemical properties, and the nature of contaminant petroleum hydrocarbon.

Environmental factors such as temperature, soil moisture, pH, nutrients, organic matter, and nature of contaminants directly affect the microbial population (Brockett et al. 2012; Siles and Margesin 2018; Jiang et al. 2019; Wang et al. 2020). Significant microbial correlation to soil moisture, total organic carbon (TC), carbon/nitrogen ratio (C/N), and pH in petroleum-contaminated soils have been reported at varying degrees (Cui et al. 2020; Trejos-Delgado et al. 2020), as well as successive transitions linked to long-term soil complexity (Zegeye et al. 2019; Zhao et al. 2019) and different spatial or temporal distributions relative to soil nutrients (Kivlin and Hawkes 2016). Besides biological and particularly microbiological disturbance, petroleum contamination negatively alters soil physicochemical properties mostly at the surface and decreases with increasing soil depth (Sheng et al. 2021). Some notable soil properties altered include a decreased wettability (Błońska et al. 2016; Rodríguez-Rodríguez et al. 2016), hydraulic conductivity (Devatha et al. 2019), increased hydrophobicity, and CO2 efflux (Hewelke et al. 2018). Microbial tolerance to changes in soil physicochemical properties differ, bacteria strains are better known as effective degraders of PAHs, but fungi are better at surviving harsh environments. Hence, fungi could also be potential stimulators of bacterial growth in polluted environments where bacteria are limited (Oyetibo et al. 2017).

Microbes with hydrocarbonoclastic capabilities are not easily isolated using standard nutrient media in laboratory conditions and a culture-dependent method, which could only recover approximately 0.3% of its population (Chikere et al. 2019). Because of the limitations of biochemical techniques in microbial identification and characterization (Adzitey et al. 2013), as well as bacterial Gram variabilities (Ramamurthy et al. 2014; Franco-Duarte et al. 2019), in recent decades, a significant number of studies have suggested that 16S and 18S rRNA amplicon sequencing are more reliable for exploring soil microbiomes and the prevailing geophysical and chemical properties of soil (Mukherjee et al. 2017; Abia et al. 2018; Chikere et al. 2019; Kang et al. 2020). Thus, the objectives of this study were as follows: (i) to investigate the microbial community structure and diversities in petroleum-contaminated soils of Utorogu (UT), Ughelli East (UE), and Ughelli West (UW) oilfields in Delta State using 16S and 18S rRNA amplicon sequencing and (ii) to assay the difference in component composition of crude oil from these oilfields using a gas chromatography-mass spectrometer (GC-MS) analysis. We also assessed the soil physicochemical properties and their correlation to the microbial communities as a potential guide for bioremediation indicators.

Materials and methods

Characteristics of the study sites

The contaminated oilfield sites comprised UE, UT, and UW in Delta State and the non-contaminated soil sample from Alagomeji (AL) in Lagos State, Nigeria, a non-oil-producing area. The UE oil flow station has been in operation since 1959. Its average ambient temperature is 26.4 °C and 2.768 mm of precipitation. The UT oilfield exploration began in 1964. The average ambient temperature is 26.7 °C, and it receives 2768 mm of precipitation (Climate-Data.org 2019). The UW oilfield exploration began in 1960. The average temperature of the location is 26.6 °C, and it receives 2.638 mm of precipitation. The climate conditions are classified as tropical monsoon climate, and they have loamy wetland soils (Chibuzo 2016). The average temperature of the sample site AL is 27.0 °C, and it receives 1675 mm of precipitation. We also characterize its soil as a loamy wetland area with little stones (Obiefuna et al. 2013).

Soil sample collection

Soil samples were collected in August 2018 from four different sampling locations: UE (5° 32′ 17.8″ N, 5° 53′ 46.8″ E), UT (5° 25′ 55.6″ N, 5° 53′ 02.5″ E), UW (5° 32′ 17.8″ N, 5° 53′ 46.8″ E), and AL (6° 29′ 48.4″ N, 3° 22′ 36.5″ E) in triplicates to a total of 12 samples. The distances between sampling locations are 11.89 km (UT and UW), 5.29 km (UW and UE), 10.80 km (UE and UT), and 301.21 km (AL and UT) (Fig. 1). Each sample was randomly collected using an aseptic sterile auger and 50-mL tubes at 10 cm depth, and the collection points were > 5 m apart. A portion of the collected samples was immediately measured for moisture content and DNA extraction, and the remaining portions were air-dried for further physicochemical property analysis.

Fig. 1
figure 1

 Sampling areas in Delta State among other oil-producing states in the Niger Delta region

Full size image

Soil physicochemical analysis

The soil samples were oven-dried at 105 °C to constant weight for their moisture contents to be determined gravimetrically. The soil pH was determined using a 1:4 (w/v) soil-deionized water and analyzed with a standard calibrated pH probe. The electric conductivity (EC) value was measured using a digital conductivity instrument (Leici DDSJ-308A) in a soil-deionized water slurry of 1:5. The total carbon (TC) and total nitrogen (TN) in the soil were determined using an elemental analyzer (Elementar Vario-EL, Germany).

GC-MS analysis of crude oil samples

The crude oil samples were prepared using the dichloromethane solvent in a high-resolution Agilent GC-MS 7000 C Triple Quad machine fitted with an HP-5MS (Agilent Technologies, Inc. USA), 30 m ×250 μm, 0.25 μm chromatographic column. The oven temperature was initially kept at 30 °C for 21 min and finally increased to 310 °C at 15 °C·min−1 for 9 min. The injection volume was at 1 μL, and the detector temperature was set at 350 °C with helium carrier gas of > 99.99% purity at the flow rate of 1 mL·min−1. The detector signal was integrated and analyzed in the Mass Hunter computer program through a chromatographic server that generated corresponding gas peak heights and details of specific components. The properties of the component organic compounds detected were further characterized using the National Library of Medicine database (PubChem 2020). For sample quantification and procedural recovery, the internal and surrogate standard of 16 PAH (Ultra Scientific PM-610) solution containing anthracene (AN), acenaphthylene (ACY), acenaphthene (ACE), benzo[a]anthracene (BaA), benzo[g,h,i]perylene (BP), benzo[b]fluoranthene (BbF), benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP), fluorene (FL), fluoranthene (FLU), pyrene (PY), chrysene (CH), indeno[1,2,3-cd]pyrene (IP), dibenzo[a,h]anthracene (DA), naphthalene (NA), and phenanthrene (PH) was used as calibration standards (Mills et al. 1999).

DNA extraction and rRNA amplicon sequencing

The genomic DNA was extracted from 5 g of each soil sample using the Fast DNA® SPIN Kit for soil and MP Biomedicals Fast Prep® instrument according to the manufacturer’s protocol (MP Biomedicals 2007). The bacterial 16S rRNA genes were amplified using universal primer 338F_806R 5′-ACTCCTACGGGAGGCAGCA-3′ and 5′-GGACTACHVGGGTWTCTAAT-3′ (Bakke et al. 2011). For fungi, ITS1F_ITS2R primer 5′-CTTGGTCATTTAGAGGAAGTAA-3′ and 5′-GCTGCGTTCTTCATCGATGC-3′ were used (White et al. 1990). Finally, sequencing was performed on an Illumina MiSeq PE300 platform (Illumina, USA) in Majorbio Co., Ltd. (Shanghai, China). The processed sequences were aligned and deposited on the National Center for Biotechnology Information (NCBI) with the sequence read archive accession numbers for bacteria and fungi being PRJNA593923 and PRJNA594087.

Statistical analysis

The species annotation and genomic assessment were assigned to their specific categories of operational taxonomic units (OTUs) according to the USEARCH (version 7.0 http://drive5.com/usearch/) software platform (Fan et al. 2018). Non-repeated sequences were extracted to reduce redundancy, while single sequences without repetition and chimeras were removed during clustering to obtain the representative OTU at 97% similarity. The bacteria and fungi OTUs were classified into their exact taxonomy based on the SILVA database (Release 128 http://www.arb-silva.de) and Ribosomal Database Project (https://rdp.cme.msu.edu/) and UNITE database for environmental ITS identification (Release 8.0 http://unite.ut.ee/). The Bayesian taxonomic algorithm at 97% similarity classified the community compositions as phylum, class, order, family, genus, and species.

Based on the number and abundance of OTUs, community diversity (Shannon and Simpson) and richness (Sobs, Chao, and ACE) were estimated (Hugerth and Andersson 2017). The redundancy analysis index (RDA) was calculated, and permutational multivariate analysis of variance (PERMANOVA) and environmental correlation analysis were performed using the Majorbio I-Sanger Cloud Platform (Majorbio Co., Ltd., Shanghai, China). The sample distance matrix was clustered for hierarchical analysis to understand the similarities and differences between the sample’s OTU composition and their structural differences (Borem et al. 2014). The unweighted pair-group method with arithmetic mean (UPGMA) algorithm was used to determine the degree of microbial evolution in various samples, while a weighted UniFrac analysis further quantified the lineage variation between samples (Nye et al. 2006; Clarke et al. 2006).

Bray-Curtis distance matrix algorithm:

$$ {D}_{\mathrm{B}\mathrm{ray}-\mathrm{Curtis}}=1-2\frac{\sum \min \left({S}_{{\mathrm{A}}_{i^1}}{S}_{\mathrm{B},i}\right)}{\sum {S}_{\mathrm{A},i}+\varSigma {s}_{\mathrm{B},i}} $$

SA,i = number of sequences in the ith taxon in the A sample and SB in the ith taxon in the B sample.

$$ {D}_{12}=\mathrm{sqrt}\left(\sum {\left({x}_{11}-{x}_{12}\right)}^2\right),i=1,2,3\dots $$

where the number of ith OTU represents samples 1 and 2; sqrt = square root.

Results

Dissimilarities in the soil physicochemical properties

The most distinct soil pH range (6.8–7.7) across samples was presented in UE and AL, which was relatively narrow compared to water content, TN, TC, and EC, with significant differences (Table 1). The same EC value at 48.2 μS cm−1 was detected in UT and UW soils. The AL soil presented the highest EC value at 93.5 μS cm−1, while the UE soil presented the lowest EC at 24.2 μS cm−1. The highest TC at 10.4 mg g−1 was recorded in the UE soil. The moisture content (MO) in UT soil at 41.47% was higher than the UE and UW soils by 22% and 23.9%, respectively. The AL soil MO significantly differed from UT, UE, and UW soils, possibly due to the riverine terrain where the oilfields are located.

Table 1 Soil physicochemical properties of the 12 soil samples from four different locations
Full size table

Petroleum organic components of UT, UE, and UE oil samples

The GC-MS analysis of crude oil samples from the UE, UT, and UW oilfields showed highly complex mixtures of constituent aromatic, paraffinic, and naphthenic (cycloparaffin) hydrocarbon compounds, which were further characterized based on the National Library of Medicine database (Table S1). A total of 293 peaks (Fig. S1) detected in the crude oil samples (UT: 69, UW: 94, and UE: 130) were related to paraffin, naphthene, and aromatic hydrocarbon compounds (Table S1). Compounds of low molecular weight (LMW) (C7–C10) and complexity, such as cyclohexane, ethylbenzene, xylene, benzene, indene, octahydro-, cis-, 1-ethyl-3-methylcyclohexane, and menthane, were unique to UT. Medium molecular weight (MMW) compounds (C11–C15) were most observed in UE and UW, notably decahydro-1,1,4a,5,6-pentamethylnaphthalene highest peak areas by 80.20% and 93.80%, respectively. Among the MMW compounds, 4-(2,2,6-trimethyl-bicyclo(4.1.0) hept-1-yl)-butan-2-one (C14H24O) and 2-pentanone (C15H26O) were observed only in UE, while naphthalenone (C15H22O2) was absent in UW and 3-buten-2-one specific to UW. High molecular weight (HMW) compounds (C16–C29) were disproportionately found among the three samples but most significant in UE by 4–36% peak area. The properties of these petroleum compounds could determine their fate in the contaminated soils, which can be categorized into three principal fractions, namely the volatile (C7–C10), highly extractible (C11–C16), and low extractible (C17–C19). The LMW compounds are highly volatile, less persistent in soil, and likely to repel less soil moisture, but highly toxic due to bioavailability compared to the MMW and HMW compounds.

Variation in the number of OTUs and α-diversity of microbial communities in the soils

The clustering analysis revealed a total average of 1515 bacterial and 919 fungal OTUs from the soil samples. The highest occurring bacterial (1830) and fungal (1184) OTU numbers were found in UW soil and the lowest (1120 and 747) OTU numbers, respectively, in UE soil (Fig. S2). The bacterial richness (Chao1 index) and diversity (Shannon) exhibited the trend AL > UT > UW > UE and AL > UW > UT > UE, respectively, while the fungal richness (Chao1 index) and diversity (Shannon) was AL > UW > UT > UE and UW > UT > AL > UE, respectively (Table S2). The results suggest that the impact of petroleum toxicity reduced the richness and diversity of soil bacterial communities and the richness of fungal communities but had a lesser impact on fungal diversity.

Variation in bacterial and fungal community structure in the soils

In total, 42 bacterial phyla and 360 families were identified from the soils. The predominant bacterial phyla account for 82% of the total community abundance. These include Proteobacteria, Actinobacteria, Acidobacteria, and Chloroflexi by 44%, 17%, 12%, and 9%, respectively (Fig. 2a). Proteobacteria was the most abundant phylum (55.4–42.2%) in the contaminated soils due to its high aromatic degradative activity compared with the non-contaminated AL soil where Proteobacteria (27.2%) and Acidobacteria (8.4%) were markedly decreased, suggesting that petroleum contamination was significantly a driving factor. Bacteroidetes, Firmicutes, Fusobacteria, and Planctomycetes were also highly abundant in the UT soil and twofold higher than in UE and UW soils. The relative abundance differences suggest that these phyla could have been increased by higher moisture content and unique petroleum compounds associated with alkanes, benzene, ethylbenzene, hexane, toluene, and xylene in UT oilfields.

Fig. 2
figure 2

Relative abundance and predominant bacterial taxa. a Bacterial community composition at the phylum level. b Bacterial community composition at the family level

Full size image

The predominant bacterial families in the contaminated soils were unclassified Rhizobiales (14.4%), KCM-B-112 (12.8%), and Acidobacteriaceae (8.5%) in UT, UE, and UW soils, respectively (Fig. 2b). Other dominant petroleum-degrading families notably present across contaminated soils were Desulfurellaceae, Rhodospirillaceae, and Anaerolineaceae except KCM-B-112, a hydrocarbon-degrading family whose absence in UT is not entirely apparent but could be associated with BTEX toxicity or excess MO content in the soil. Additionally, Rhodospirillaceae and Rhodocyclaceae are medium- and long-chain hydrocarbon degraders predominant in UE and distributed across UT and UW. The above distribution suggests that the relative abundance of some predominant oil-degrading biomarker families across soils were in response to petroleum contamination and soil physicochemical properties.

In total, 8 fungal phyla, 196 families, and 419 genera were identified in the soil samples. Ascomycota (92.3%) predominated the AL non-contaminated soil but was reduced to 41.5%, 55.9%, and 54.2% in UE, UT, and UW soils, respectively (Fig. 3a). Unclassified Fungi phylum predominated UE soil but was significantly decreased in UT and UW soils, where the abundance of Basidiomycota, Rozellomycota, and Chytridiomycota was notably increased. Trichocomaceae (38%) dominated AL soil at the family level, while unclassified Fungi dominated accounted for 56.2%, 27.9%, and 40.4% in UE, UT, and UW, respectively. Chaetomiaceae, Nectriaceae, unclassified Hypocreales, and Sporormiaceae were markedly decreased in the contaminated soils compared to AL. Notable petroleum-degrading fungal families in the contaminated soils include unclassified Sporidiobolales (10.6%) and Montagnulaceae (5.2%) in UT, unclassified Chaetothyriales (7%) and Pleosporaceae (6%) in UW, and unclassified Hypocreales (5%) in UE.

Fig. 3
figure 3

Relative abundance and predominant fungal taxa. a Fungal  community composition at the family level. b Fungal community composition at the genus level. Different colors represent taxa and proportions present

Full size image

Aspergillus predominated the non-contaminated AL soil at the genus level while unclassified Fungi predominated the contaminated soils. The relative abundance of Aspergillus, Talaromyces, unclassified Sporidiobolales, and Paraconiothyrium was markedly more in UT than other contaminated soils (Fig. 3b). Among the dominant hydrocarbon-degrading fungi, Aspergillus is one of the well-reported genera known to survive a wide range of environmental conditions; hence, its relatively low abundance in other contaminated soils is unclear but could be linked to other inhibiting factors such as toxic heavy metals associated with crude oil which was not tested in this study. However, high soil MO and TN content in UT seem to have influenced most other dominant genera, except unclassified Fungi, whose abundance does not correlate with any of the above factors.

The responses of microbial communities to soil properties

Principal coordinate analysis (PCoA) at the genus level was used to visually evaluate microbial community responses to soil petroleum contamination and physicochemical properties based on the weighted normalized UniFrac distance statistics. The contaminated soils’ bacterial communities showed a tight eclipse matching distinct from the clustered non-contaminated AL soil (Fig. 4a). Among the contaminated soils, a certain degree of similarity among UE1, UE2, and UW3 on the positive axis is most notable. Overall, both coordinates explain 59.1% of the total bacterial community variations. The community similarities in response to petroleum contamination shown by the distance matrix are most significant between UE and UW soil by a cross-connection, suggesting a correlation with some MMW and HMW compound degrading bacterial genera. The fungal communities also showed similar tightly cluttered AL soils on the negative PC2 axis with 15.49% total variation distinct from the contaminated soils on the positive PC1 axis, representing 64.3% of the total variations. The contaminated soils formed a crossed eclipse formation of UE2, UE1, UT2, and UW1 on the positive PC1 axis. This clustered formation shows a stronger interconnected relationship within the communities in contrast with UT1, UW3, and UT3 (Fig. 4b). Overall, both principal components explain 79.79% of the total fungal genera community variations. Interestingly, the UW distribution pattern covers a narrowly extensive range across UT and UE, while UT was widely dispersed. Comparatively, the fungal genera coordinates are slightly more interconnected than those of the bacterial genera coordinates, notably a wider spread of UT and UW soils, suggesting a distinct response to petroleum contamination between the bacterial and fungal communities.

Fig. 4
figure 4

Principal coordinate analysis (PCoA) at the genus level. a Bacteria. b Fungi. The X-axis and Y-axis represent the two selected main coordinate axes, and the percentage value represents the main coordinate axis’ interpretation of the difference in sample composition. The closer the two sample points are, the more similar the genera composition

Full size image

Correlation between bacterial and fungal community relationship to soil physicochemical properties

To evaluate the relationship between the relative abundance of community and environmental parameters, RDA was performed. In Fig. 5a, the bacterial class RDA by 62.56% (RDA2: 20.70% and RDA1: 41.86%) explained emerging acute angles between MO, TN, and TC positive correlation with Alpha-, Delta-, Beta-, and Gamma-proteobacteria, Acidobacteria, Anaerolineae, and unclassified Saccharibacteria, indicating a significant synergetic effect of these variables  on the relative abundance of the bacterial community, while EC and pH jointly impacted on Actinobacteria, Thermomicrobia, and KD4-96 communities. The fungal family by 73.3% (RDA2: 11.52% and RDA1: 61.78%) (Fig. 5b) community variances explained the emerging acute angles between MO, TN, and TC positive correlation with unclassified Sporidiobolales, Pleosporaceae, Montagnulaceae, and unclassified Fungi, while EC most impacted on Trichomaceae, Chaetomiaceae, and Nectriaceae. These bacterial and fungal community abundance variances across soils correspond to 41.5% MO in UT and 10.4 mg·g−1 TC in UE, 93.47 μS cm−1 in AL, and 0.33 mg·g−1 in UT. Besides, the most abundant phylum Proteobacteria was positively correlated with all parameters except EC. In contrast, soil pH representing a crucial soil parameter was not strongly correlated positively with any fungal family.

Fig. 5
figure 5

Redundancy analysis (RDA) between bacterial communities at the class level (a) and fungal communities at the family level (b) and soil physicochemical parameters. Different shapes or colors represent sample groups under different environmental conditions. The red color arrow depicts the physicochemical parameters, and the green arrow depicts microbial taxa. The arrow length represents the degree of influence, while the angle of deviation between the parameters represents a positive or negative correlation. Sharp angle: positive correlation; obtuse angle: negative correlation; right angle: no correlation

Full size image

A Spearman heatmap correlation analysis was used to reveal further 30 bacterial and fungal genera responses to specific soil physicochemical parameters. The top cluster of positive color gradient revealed 47% and 43% positively correlated genera with pH and EC, respectively (Fig. 6a). The most significantly positively correlated genera were unclassified Acidobacteria and unclassified Acidimicrobiales to pH and EC, respectively, while Anaerolinea was most negatively correlated with EC. The bottom-positive color gradient cluster revealed 53%, 50%, and 43% genera positively correlated with TC, TN, and MO, respectively, which are most statistically correlated with Anaerolinea (TC and TN) and unclassified Rhizobiales (MO). In the fungal genera, a cluster of color gradients at the bottom represents positively correlated genera with MO, TN, and TC by 40%, 33%, and 40%, respectively (Fig. 6b). Paraconiothyrium and Pyrenochaetopsis were the most statistically positively correlated genera with all three parameters. The top cluster of positively correlated fungal genera was 37% pH and 57% EC. Unclassified Chaetothyriales and Aspergillus most statistically correlated with EC, but none of the parameters was statistically significant with pH. In contrast, TC and EC were the most significant parameters to bacteria, but dissimilarly, fungi showed a relatively higher positive correlation with MO content than bacteria.

Fig. 6
figure 6

Hierarchical clustering heatmap analysis of 30 genera of bacteria (a) and fungi (b) and their correlation significance with physicochemical parameters of pH, EC, TN, TC, and MO. Hierarchical clustering on the top shows a classified relationship between the parameters, while on the side depicts the relationship with microbial taxa. The color intensity depicts the genera relative abundance using the legend indicator, which ranges from negative blue color coefficients to red color ones. *0.01 < P ≤ 0.05, **< P ≤ 0.001, ***P ≤ 0.001

Full size image

Discussion

Extensive researches have focused on petroleum bioremediation using the RENA approach and various isolated consortia from oil spilled sites in Nigeria. However, no previous study has reported a comparative analysis of microbial community structure in petroleum-contaminated soils across these three different oilfields in Nigeria. This study analyzed the impact of soil petroleum contamination on bacterial and fungal communities across Ughelli East (UE), Ughelli West (UW), and Utorogu (UT) oilfields. The results characterized the similarities and differences in petroleum component compounds of these oilfields and soil physicochemical properties that jointly shaped the soil microbial communities.

Although these oilfields have existed since the 1960s, the soil samples were severely re-contaminated periodically with fresh crude oil within 1–12 months before sampling. Previous studies on freshly contaminated soils in operational oilfield sites have shown that indigenous microbes within a short time are adapted to petroleum contamination and activated for biodegradation rapidly (Kaufmann et al. 2004; Liu et al. 2020). Microbial statistical correlation to soil MO, pH, EC, TC, TN, and petroleum compounds revealed that differences in MO, TN, and the contaminant crude oil most jointly shaped the soil microbial communities. A previous study on petroleum-contaminated soils has reported that crude oil contamination decreases soil moisture content due to its water repellency effect (Liang et al. 2012). In the present study, a similar decrease in soil moisture content was observed in the UE soil with the most contaminant MMW and HMW compounds by GC-MS analysis. Compared to other contaminated soils, the higher concentration of these compounds could be responsible for the lower MO content and consequently decreased microbial richness and diversity in UE soil. This effect of petroleum concentration also agrees with earlier studies that reported a strong correlation between soil moisture and oil contamination due to water repellency, hydrophobicity, and decreased soil porosity and permeability (Wang et al. 2013; Hewelke et al. 2018; Devatha et al. 2019), although previous studies have shown that petroleum contamination increases soil pH up to 8.0 (Wang et al. 2013) and could be associated with increasing soil nitrogen from fertilizer application in and around agricultural land use areas (Wang et al. 2018; Shen et al. 2019). Considering the high level of nitrogen input from agricultural NPK and gas flaring in the Niger Delta region (Osuji et al. 2006; Giwa et al. 2019), it might be challenging to ascertain the level of petroleum contamination impact on the soil’s pH. However, in this study, the UW soil with the highest soil pH (7.6) and the lowest TN (0.08) level had higher microbial diversity and richness than other contaminated soils, which is in agreement with an earlier study by Obieze et al. (2019b) that a near-neutral pH in artisanal crude oil refining sites in the Niger Delta region is suitable to improve the diversity of microbial communities (Obieze et al. 2019b). A study on soil TN correlation with microbial diversity by Roesch et al. (2007) has also reported a higher microbial diversity in forest soil with a relatively lower pH than the counterpart agricultural soils with higher pH values. This factor was attributed to commercial NPK fertilizer use (Staley et al. 2018), increase in soil NH4+, which positively correlates with soil acidity and C:N ratio leading to a buildup of organic C (Zhalnina et al. 2014).

Our present results showed that microbial community linage across contaminated soils is slightly varied at the family and genus level. Beta- and Gamma-proteobacteria dominated the contaminated population by 30% abundance. Consequently, their predominance decreased the population of Acidobacteria, Gemmatimonadetes, Actinobacteria, Planctomycetes, and Chloroflexi phyla, and Delta-proteobacteria class in UW and UT. These lower phyla have been reported as mainly composed of slow-growing organisms and are thus dominated by Alpha-, Beta-, and Gamma-proteobacteria, which are regarded as fast-growing r-strategists (Maron et al. 2018; Kurm et al. 2019). They are also regarded as obligate anaerobes of sulfate reducers and early-stage hydrocarbon degraders (Viñas et al. 2005; Prince et al. 2018). Like most previous studies, Ascomycota was the predominant known petroleum-degrading fungi phylum. Ezekoye et al. (2018b) recorded > 90% relative abundance in Ikarama community spill site remediation in Bayelsa state, Nigeria. Liu et al. (2019) recorded 58.96–99.67% in Tianjin Dagang oilfield in eastern China. Some similarities and differences with previous studies include the absence of Nocardioidaceae, Sphingomonadaceae, Flavobacteriaceae, and Methylobacteriaceae families, which were significantly dominated in some petroleum-contaminated soils (Mukherjee et al. 2017; Feng et al. 2020). Conversely, these families were absent or not significantly present in this study; however, the reason remains unclear. Obligate anaerobe genera, such as Anaeromyxobacter, Ruminiclostridium, and Geobacter, have been reported with low survival tendencies in hydrocarbon-contaminated soils (Brzeszcz et al. 2020) but were significantly represented across all contaminated soils in this study. These differences could be attributed to a long-term adaptation to these oilfields or autochthonously site-specific. Yang et al. (2019) reported the presence of Nocardioidaceae in a petroleum-contaminated soil in Qinghai-Tibetan Plateau with a high petroleum degradation capability and was attributed to adaptation by gene transformation through crude oil contamination. Likewise, in the present study, Fusobacteriaceae, Prevotellaceae, and Bacteroidales S24-7 in UT are human and animal gastrointestinal tract or fecal bacteria families that have not been previously reported in petroleum-contaminated soil. Therefore, speculate that these intestinal microbes may have gained resistance to toxic hydrocarbons due to the presence of animals or indiscriminate sewage disposal associated with poor human hygiene (Adserias-Garriga et al. 2017). The presence of unclassified Acidimicrobiales in contaminated soils in this study has been reported earlier by Kämpfer (2010), who noted that some genera belonging to family Acidimicrobiales have not been reported to perform hydrocarbon degradation but should be strictly observed. Furthermore, Peng et al. (2015) identified this family in an oil-contaminated soil through a high-throughput sequencing and linked its abundance to soil saline alkalinity.

From studies, not all bacteria and fungi taxon can be affiliated to a specific hydrocarbon compound or soil property; however, relationships between microbial communities and pollutant load can be drawn (Spini et al. 2018). A previous study on petroleum-contaminated soil in Shanxi province oilfields has reported that moisture increase by 15% from the initial 5% increased the relative abundance of Firmicutes from 4.6 to 29.7% while decreased Proteobacteria phylum (Liu et al. 2020). Similar population dynamics in hydrocarbon-degrading phyla were observed in this present study. The UT soil with extreme MO content supported the growth of rare taxa in the absence of microbial competition and increased bacteria richness, notably the abundance of Firmicutes in UT. In contrast, the UW soil with intermediate soil MO promoted microbial diversity, which can be attributed to a competitively driven population. This result is in line with previous studies (Erlandson et al. 2016; Bickel and Or 2020). TC and TN were also essential factors limiting microbial population in petroleum-contaminated soils. In a previous study, Feng et al. (2020) noted that total petroleum hydrocarbon (TPH), TC, and TN were the most significant factors affecting the bacterial communities in petroleum-contaminated Dagang oilfield soils but varied at different soil levels. In the present study, TC most significantly correlated positively and negatively with bacterial genera compared to fungal genera (Fig. 6). This factor could be related to Kaufmann et al. (2004), which reported that the optimal C:N for bacteria cell growth in a petroleum-contaminated soil is between 20:1 and 50:1. The influence of soil pH (6.8–7.7) on the fungi community variation was not statistically significant with any genus, unlike EC, TC, and MO. The reason for this result may be that the soil pH range is relatively narrow. Nevertheless, most fungi could be well adapted to a wide range of pH values sufficient to obtain optimal growth (Padmavathi et al. 2003; Rousk et al. 2010; Borowik et al. 2017). For example, a survival test on a wide pH range showed that some zoosporic fungi (Blastocladiomycota and Chytridiomycota) isolated from soil in Australia could grow in a pH range between 2.9 and 11.2 (Gleason et al. 2010).

The present study showed that soil pH most statistically correlated strongly with unclassified Acidobacteria. Similarly, Acidobacteria has been reported to have a significant positive correlation with soil pH (Conradie and Jacobs 2020) and survive under a broad range of TC sources (De Castro et al. 2013) and depleted oxygen (Pankratov et al. 2012). This broad survival range implies the reason for its significant relative abundance across all soils. Soil EC is closely related to salinity and strongly correlates with Actinobacteria abundance (O’Brien et al. 2019), which affirms its significantly higher abundance in non-contaminated soils in this study. This result is also similar to Peng et al. (2015), which reported that Actinobacteria was most abundant in contaminated soils across the three oil-mined wells in the 1960s, 1970s, and 1990s Daqing oilfields.

The GS-MS analysis indicated that the oilfields situated within the same region produced slightly different petroleum compounds. The observed differences in hydrocarbon components support a previous study that closely situated oilfields in an area of the same stratum, or oil-bearing sands may produce dissimilar crude oils (International Agency for Research on Cancer 1989). While complex high molecular compounds are ultimately recalcitrant in contaminated soils (Feyzi et al. 2020), studies have shown that straight-chain alkanes are more easily oxidized than branched-chain and cycloalkanes, so also the degradation rates of medium-chain alkanes (C5–C15) significantly vary from those > C20 (Jia et al. 2017). Rhodocyclaceae have a high correlation with the degradation of mid-chain C14, C16, and C18 hydrocarbons, while Rhodospirillaceae most correlated with only C14 mid-chain alkanes (Liu et al. 2018). In the present study, both families are more abundant in UE, while Rhodospirillaceae in UT and Rhodocyclaceae in UW. A toxic compound similar to 4H-cyclopropa[5′,6′]benz[1′,2′:7,8]azuleno[5,6-b]oxiren-4-one,8-acetyloxy was most significant in UE and UT, accounting for 53% and 74%, respectively. This compound has been reported as a phytochemical component extracted from a plant with an antimicrobial effect on the Bacillus, Pseudomonas, Escherichia bacterial genera, and Candida, Trichoderma, and Aspergillus fungal genera (Abdul-Hafeez et al. 2016; Fierascu et al. 2017). Speculatively, it could have a similar toxic effect on these soil microbial communities and should be investigated in further studies. The highest degradation percentage of benzene, ethylbenzene, and xylene compounds have been observed with Aspergillus (Asemoloye et al. 2020; Usman et al. 2020a, b), which suggest its relative abundance by 7% more in UT than other contaminated soils. However, its relatively low abundance in other contaminated soils could only be associated with possible heavy metal contamination (Zeng et al. 2020). Penicillium was significantly abundant in contaminated soils but showed no significant correlation to any environmental factor. Hence, we suggest its strong affinity for long-chain n-alkane (C20–C29) hydrocarbon degradation by adaptation to petroleum contamination as reported by Barnes et al. (2018) and Yamada-Onodera et al. (2002).

Our study shows a shift in the relative abundance of some bacterial families and fungal genera, which might be useful indicators to assess soil petroleum contamination. Moreover, the distribution patterns of Proteobacteria, Firmicutes, Ascomycota, and Aspergillus communities might be useful microbial parameters in the evaluation of petroleum-contaminated soil in and around oilfields. Further studies on these oilfields would be required to determine specific roles of some newly identified genera (Acidimicrobiales, Fusobacteriaceae, Prevotellaceae, and Bacteroidales S24-7) in petroleum hydrocarbon degradation. Furthermore, our study implies that petroleum-contaminated soils did not support more bacterial and fungal community richness but only fungal diversity.

Conclusion

This study characterized bacterial and fungal diversity in petroleum-contaminated soils from three different oilfields in Nigeria. The impact of petroleum contamination on the microbial communities was observed to have caused a decline in microbial richness and diversity distinct from a non-contaminated site, except fungal diversity. Petroleum contamination favored fast-growing autochthonous petroleum-degrading class, such as Alpha-, Beta-, and Gamma-proteobacteria, while some obligate genera were observed to have undergone long-term adaptation to hydrocarbon toxicity. Proteobacteria and Ascomycota phyla played a significant role in the degradation of hydrocarbons in these contaminated sites. Excess soil MO content in contaminated soil enriched rare bacteria OTUs in the absence of competition, but intermediate soil MO increased OTU diversity. In contrast, bacteria and fungi respond differently to excess soil MO and petroleum compounds. The LMW compounds in UT different from UE and UW crude oil are partly responsible for the significant disproportion in some vital bacterial phylotypes. These compounds include cyclohexane, ethylbenzene, p-xylene, benzene, 1H-indene, and m-menthane with low complexity and biodegradation half-life values. Conversely, benzene, toluene, ethylbenzene, and xylene (BTEX) compounds are known for groundwater contamination due to easier penetration through soil and bioavailability. Thus, it should be considered a potential threat to human health when leached into the ground and nearby streams and rivers. This analysis of the microbial community richness and diversity impacted by oil pollution is limited to and around these oilfields. At the same time, it provides useful insight for an in-depth study on microbial isolation and biodegradation capabilities to improve spill site remediation application.