Introduction

Petroleum hydrocarbons released into the marine environment tend to accumulate in marine sediment due to their low solubility and high affinity to sediment particles, leading to a serious contamination of marine sediment and adversely affecting the health of aquatic life (Jonker et al. 2006). As a result, many efforts have been made to explore suitable strategies that can be applied to decontaminate marine sediment from petroleum hydrocarbons (Chang et al. 2008; Dell'Anno et al. 2009).

Compared with physical and chemical remediation methods, bioremediation is attracting increasing popularity, as it is considered to be non-invasive and relatively cost-effective (Manconi et al. 2007; Perelo 2010). Biostimulation is one of the bioremediation strategies which have been commonly employed. Biostimulation agents including electron acceptor, co-substrate, and nutrient have been employed in an attempt to stimulate indigenous microorganisms that are capable of degrading organic contaminants such as petroleum hydrocarbons (Boopathy 2003; Dell'Anno et al. 2009; Beolchini et al. 2010).

Electron acceptors have been widely studied as biostimulation agent to facilitate the degradation of the petroleum hydrocarbons in marine sediment (Hasinger et al. 2012). In the absence of oxygen, anaerobic microorganisms can grow with alternative electron acceptors such as nitrate, sulfate, iron, and CO2, which link to various anaerobic metabolic processes including denitrification, sulfate reduction, iron reduction, and methanogenesis (Boopathy 2003; Qian et al. 2015). Degradation of petroleum hydrocarbons coupled to these anaerobic metabolic processes has been previously demonstrated (Lovley et al. 1989; Rueter et al. 1994; Coates et al. 1997; Zengler et al. 1999). With the addition of different electron acceptors, dissimilar performances of petroleum hydrocarbon degradation have been observed. The differences can be reflected in at least three aspects, including (i) degradation activation (Aitken et al. 2013), (ii) preferential degradation with different hydrocarbon structures and carbon chains (Hasinger et al. 2012), and (iii) degradation rate (Chang et al. 2003). These findings suggest a vital role of various metabolic processes in regulating the degradation of petroleum hydrocarbons, which deserves more research attention.

Biostimulation by adding co-substrate has also been verified to be a feasible strategy to enhance the anaerobic degradation of petroleum hydrocarbons (Kim et al. 2003; Okolo 2005). Previous studies found that polycyclic aromatic hydrocarbons (PAHs), one of the important petroleum hydrocarbon components, can be more effectively removed when using co-substrate as the biostimulation agent, as compared with electron acceptors and nutrients (Taylor and Jones 2001; Ambrosoli et al. 2005). These findings suggest that co-substrate could be a better option for improving the petroleum hydrocarbon degradation in marine sediment. Acetate and methanol have been suggested as co-substrates for enhancing the degradation of recalcitrant organic pollutants in the biostimulation of contaminated sediment (Lu 2001; Dell'Anno et al. 2009). Up to date, the relevant studies mainly focused on the feasibility of using acetate or methanol to facilitate the degradation of some specific organic pollutants (Lu 2001; Ambrosoli et al. 2005) and kinetics of organic pollutant degradation with the addition of acetate or methanol (Yuan et al. 2001). Some studies demonstrate that acetate and methanol can be readily utilized by different microorganisms such as sulfate-reducing bacteria (SRB) and methanogen through various metabolic processes (Liamleam and Annachhatre 2007). However, the investigation of various metabolic processes involved in the utilization of acetate and methanol mostly focused on the treatment of sulfate-containing wastewater, in which acetate and methanol mainly act as electron donors to remove the sulfate in wastewater (Liamleam and Annachhatre 2007). Very little information is available on the metabolic processes involved using acetate and methanol as co-substrates in the biostimulation of contaminated marine sediment, and especially, the consequential influence on the degradation of petroleum hydrocarbons. But, this information is essential for obtaining a deeper understanding of the effect of co-substrate on petroleum hydrocarbon degradation.

The biogeochemical process is closely linked with the microbial community present in the environment (Kleikemper et al. 2002a, b). Thus, the information on the microbial community response to different co-substrates is necessary to elucidate and verify the involved biogeochemical process in biostimulation of contaminated marine sediment with acetate and methanol as co-substrates. And, a more comprehensive picture of the effect of co-substrate on petroleum hydrocarbon degradation could be expected if the information of microbial community was considered.

Therefore, the aims of this study were to (i) determine the anaerobic biodegradation of total petroleum hydrocarbons (TPHs) in marine sediment with the addition of acetate and methanol as co-substrates, (ii) study the involved metabolic processes in the biostimulation of contaminated marine sediment with the addition of different co-substrates, and (iii) investigate the major composition of the microbial community in sediment dosed with different co-substrates, as well as the shifts of the microbial community over the incubation period.

Materials and methods

Sediment and seawater

The marine sediment used in this study was collected from the southern part of Kowloon in Hong Kong during March. With regard to the petroleum hydrocarbon contamination problems of the sediment at this site, it is continuously receiving the discharge of oil-containing wastewater from a large old urban area in East Kowloon. The sediment was collected from 10 to 100 cm below the seawater/sediment interface using a corer hammered into the sediment up to the required depth. A plunger was inserted into the top of the corer to push the sediment out from the corer. The collected sediment was sealed in polyethylene bags to preclude the possibility of sediment oxidation, transported to the laboratory, and stored at 4 °C (without sediment freezing) before use. The sediment was further sieved to remove debris over 2 mm in size and thoroughly homogenized in an anaerobic glove box under N2 atmosphere before sediment characterization.

The selected physicochemical properties of the sediment used in the present study are shown in Table 1. The high concentration of TPH (C10–C40) in the investigated sediment indicates heavy contamination with petroleum hydrocarbons (typically >1000 mg/kg dry weight) (Mille et al. 2007). The high negative oxidation reduction potential (ORP) value indicated strongly reducing conditions in the sediment. High concentrations of SO4 2− and acid-volatile sulfide (AVS) were detected in the sediment.

Table 1 Selected physicochemical properties of sediment and seawater
Full size table

Seawater was collected from the same contaminated site. Sufficient amount of seawater sample was collected at the mid-depth of the water column overlaying the sediment using a water sampler comprising a transparent PVC cylinder with capacity of 2 L. The collected seawater was filtered through a 0.22-μm polyethersulfone membrane (Advantec MFS, CA, USA) to remove suspended solids and purged using N2 gas to reduce dissolved oxygen in the seawater. The selected physicochemical properties of the seawater are also given in Table 1.

Bioremediation experiments

Batch experiments were conducted in a series of 250-mL serum bottles. Each bottle was filled with 100 mL of wet sediment (47.84 % dry weight) and 100 mL of pre-filtered (0.22 μm) and N2-purged seawater. The treatment groups were added with (i) acetate at dosages of 5, 7.5, 10, and 15 mmol or (ii) methanol at dosages of 10, 15, 20, and 30 mmol in the sediment. It must be noted that the dosages of acetate and methanol were designed in order to provide the same carbon content. The sediments without the addition of acetate or methanol were utilized as controls. All controls and treatment groups were conducted in triplicate. After purging the headspace with N2 gas, all bottles were capped with butyl rubber stoppers and incubated without shaking at room temperature in darkness. Triplicate bottles of each treatment and control were sampled for analysis after 1, 6, 11, 20, and 30 weeks of incubation. An inhibitory test was conducted with 11 weeks of incubation, in which the sediment was dosed with acetate (10 mmol in 100 mL of wet sediment) and sodium molybdate (2 mmol in 100 mL of wet sediment). The purpose of adopting molybdate as the specific inhibitor of sulfate reduction was mainly to investigate the role of specific bacteria (SRB) in the consumption of acetate and the removal of TPH.

Chemical analysis

The gas in the headspace of each serum bottle was collected using a sterile syringe. The volume of produced gas was determined by displacement of the syringe under atmospheric pressure (Gerlach et al. 1999). In order to minimize the potential effect of friction, all the syringes were made of glass and were pre-moisturized by deionized water. The measurement was recorded after a few minutes (i.e., when the syringe plunger had reached a stable position). The CH4 gas was quantified using an Agilent 7890 series gas chromatograph (GC) equipped with a flame ionization detector (FID) (Phillips et al. 2001).

Acetate and SO4 2− in sediment pore water were analyzed using ion chromatography (HIC-20A super, Shimadzu), as described elsewhere (Kleikemper et al. 2002a). Methanol in sediment pore water was analyzed using GC-FID. The moisture content of the sediment was calculated from the sediment weight before and after drying to a constant weight at 105 °C in an oven. The pH and ORP value of the sediment were determined using a Multi 3420 meter (WTW, Germany), equipped with Sen Tix® pH and ORP electrode probes. Specifically, the calibrated pH and ORP electrode probes were inserted into the serum bottle and left to stabilize for 30 min before a measurement was recorded. The pH and ORP measurements were performed in an anaerobic glove box under N2 atmosphere. The sulfide content of the sediment was measured as acid-volatile sulfide (AVS) following the USEPA protocol (1991).

TPH was analyzed in accordance with the standard method of European Committee for Standardization (CEN 2004). Briefly, freeze-dried sediment was extracted with acetone/n-heptane mixture (1:1, v/v) in a microwave system under controlled temperature at 150 °C for 15 min. The extract was filtered and concentrated using rotary evaporation (Laborota 4000 efficient, Heidolph, Germany). The concentrated filtrate was then made up to 10 mL with n-heptane, followed by gas chromatography analysis. The concentration of TPH was analyzed by the GC-FID, equipped with an Agilent 7683 B autosampler and a low-bleed Rtx® 5MS capillary column (30 m × 250 μm i.d.) with a nominal film thickness of 0.25 μm. The injection temperature was 330 °C and injection volume was 1 μL. Helium was used as the carrier gas (3 mL/min). The column was held at 50 °C for 10 min and ramped up at 25 °C/min to 320 °C and then held for 15 min. The TPH (C10–C40) amount was determined by summing up both the unresolved and resolved components eluted from the GC capillary column between the retention times of n-C10H22 and n-C40H82 (Supplementary Fig. S1). Similarly, the amount of different TPH fractions with various carbon number ranges was obtained by integrating the eluted components between the corresponding retention times of n-alkanes (Supplementary Fig. S2). The concentrations of TPH were normalized to the sediment on the dry weight basis. The TPH removal efficiency was then calculated according to the following expression: TPH removal efficiency (%) = [(Initial TPH concentration − Residual TPH concentration after incubation) / Initial TPH concentration] × 100 %.

16S rRNA gene–based microbial analysis

DNA extraction and PCR amplification

The 16S rRNA gene clone library, which has been successfully applied to characterize the active microbial communities, indicate the predominant organism, and infer the degradation pathway of hydrocarbons (Zhang et al. 2009; Siddique et al. 2011), was employed in the present study for microbial analysis. The microbial communities in the original sediment before treatment and in the sediments dosed with acetate or methanol after 6 and 30 weeks of incubation were analyzed by the construction of 16S rRNA gene clone libraries. To enhance the DNA recovery, all sediments were pretreated using sodium dodecyl sulfate (SDS, 2 %, w/v) with freeze-thawing to improve the efficiency of cell wall breaking (Zhou et al. 1996) and using phosphate-buffered saline (PBS, pH 7.4) to wash out the inorganic salts from the sediment pore water before DNA extraction (Lee et al. 1999). Genomic DNA was extracted from the sediment using the FastDNA SPIN Kit for Soil (Q-BIOgene, USA) according to the manufacturer’s instructions.

DNA extracts were amplified by polymerase chain reaction (PCR). The bacterial primers or archaeal primers and the corresponding conditions for bacterial and archaeal 16S rRNA gene amplification are given in Supplementary Table S1. The PCR efficiency and specificity were confirmed by performing the electrophoresis at a voltage of 120 V for 25 min on an ethidium-bromide-stained agarose gel.

Cloning and sequencing

The TA Cloning Kit (Invitrogen Corporation, Carlsbad, CA) was used to clone the PCR amplicons of the 16S rRNA gene. PCR products were purified by a Qiagen II Extraction Kit (Qiagen Corp., Germany), ligated into pGM-19 T vector (Promega), transformed into Escherichia coli DH5α competent cells, and followed by inoculation on 50 mg/L ampicillin-containing LB medium and incubation at 37 °C for 16 h. White clones from the LB medium were then randomly selected, and the cloned 16S rRNA genes were sequenced by BGI Life Tech Co., Ltd. (BGI, China).

Phylogenetic analysis

Sequence data was checked for the presence of PCR-amplified chimeric sequences using the CHECK_CHIMERA program. The comparison with the 16S rRNA gene sequences was performed by submitting the correct sequences into the GenBank nucleotide library using the BLAST similarity searches. The multiple sequence alignment of the DNA sequences and their closest 16S rRNA sequences of reference microorganisms retrieved from the GenBank were conducted using the software of BioEdit. Sequences less than 3 % divergent were defined as an operational taxonomy unit (OTU), and one representative sequence per OTU was selected for the following construction of phylogenetic trees. The construction of phylogenetic trees was performed by the software of MEGA 4 (Tamura et al. 2007) using the neighbor-joining method. Bootstrap re-sampling analysis for 1000 replicates was performed to evaluate the confidence of the tree topologies. The sequences reported in the present study were submitted to GenBank and their accession numbers are KM925084–KM925132.

Results

TPH biodegradation

About 30–41 % of the TPH was removed in the sediment using acetate as co-substrate at various dosages of 5, 7.5, 10, and 15 mmol after 6 weeks of incubation (Fig. 1). Compared with 12 % of TPH removal efficiency in the untreated sediment, the addition of acetate showed a significant favorable effect on TPH degradation (assessed by the one-way ANOVA test, p < 0.05). After 30 weeks of incubation, approximately 64 % of TPH removal efficiency was achieved in the sediment dosed with 10 mmol of acetate. A higher dosage of acetate did not achieve higher TPH removal efficiency. The addition of methanol was also found to be capable of facilitating the degradation of TPH. Figure 1 illustrates that 32–39 % of TPH was degraded in the sediment dosed with methanol at various dosages of 10, 15, 20, and 30 mmol after 6 weeks of incubation, versus 12 % of TPH removal efficiency in the untreated sediment (p < 0.05). When the incubation time was prolonged to 30 weeks, approximately 53 % of TPH removal efficiency was achieved at the dosage of 20 mmol. A higher dosage of methanol did not achieve higher TPH removal efficiency. The results above show that both acetate and methanol enhanced the degradation of TPH. The removal efficiencies of TPH in sediment dosed with acetate at all dosages were generally higher, although not significantly, than that dosed with methanol at the corresponding dosages in terms of the carbon content (0.05 < p < 0.10). After 30 weeks of incubation, the highest removal efficiency of TPH was achieved in acetate-treated sediment at dosages of 10 mmol (p = 0.037), while in the methanol-treated sediment, TPH removal efficiency was relatively higher at dosages of 20 mmol compared with that at other dosages of methanol (p = 0.232). Therefore, these two groups were further selected for investigating the degradation of TPH fractions with different carbon number ranges.

Fig. 1
figure 1

TPH removal efficiency in untreated sediment and in sediment dosed with acetate or methanol after 1, 6, 11, 20, and 30 weeks of incubation. Error bars represent standard deviations

Full size image

The TPH (C10–C40) was further divided into six groups in terms of different carbon number ranges. After 30 weeks of incubation, the addition of acetate was more favorable for the degradation of TPH fractions with carbon number in the ranges of C10–C20 and C31–C40 (Fig. 2a, p = 0.006). The removal efficiencies of TPH fractions in the range of C10–C20 were as high as 70–72 %, while the removal efficiencies of TPH fractions in the range of C31–C40 were 60–64 %. In the sediment dosed with methanol, the removal efficiency of TPH fractions significantly decreased with an increase of carbon number (Fig. 2b, p = 0.001). The TPH fractions in the range of C10–C20 were degraded to a relatively higher extent (58–69 % of the removal efficiencies), while only 26–29 % of TPH fractions in the range of C31–C40 was degraded after 30 weeks of incubation.

Fig. 2
figure 2

Removal efficiency of TPH with different carbon number ranges in sediment dosed with (a) acetate at the dosage of 10 mmol and (b) methanol at the dosage of 20 mmol after 1, 6,11, 20, and 30 weeks of incubation. Error bars represent standard deviations

Full size image

The consumption of both acetate and methanol was observed from the first week of incubation (Fig. 3). Most of the acetate and methanol was found to be consumed after 6 weeks. This suggests that either acetate or methanol can be readily utilized by the indigenous microorganisms in the marine sediment. Given that TPH degradation was facilitated with the consumption of acetate and methanol during the first 6 weeks of incubation, it was speculated that both acetate and methanol could stimulate the growth of the indigenous microorganisms by effectively proving available carbon source (Handley et al. 2012), and consequently facilitate TPH degradation.

Fig. 3
figure 3

(a) Residual acetate and (b) residual methanol in sediment dosed with acetate or methanol after 1, 6, 11, 20, and 30 weeks of incubation. Error bars represent standard deviations

Full size image

Metabolic processes involved using different co-substrates

The increase of AVS and decrease of SO4 2− were observed in the sediments dosed with acetate after 1 week of incubation (Figs. 4a and 5), while almost no CH4 gas was detected (Fig. 4b). This reveals that sulfate reduction was prevalent in the acetate-treated sediment at the beginning of the incubation (Eq. 1). After 6 weeks of incubation, besides the increase of AVS and decrease of SO4 2−, the CH4 gas was also detected (Fig. 4b), indicating that both sulfate reduction and the methanogenesis process (Eq. 2) occurred during this incubation period. AVS concentration increased until the 11th week and reached a plateau afterward. The methanogenesis process predominated from the 11th week to the 30th week, as indicated by the continual generation of CH4 gas during this incubation period (Fig. 4b). As for the sediment dosed with methanol, a large amount of CH4 gas was detected within 30 weeks of incubation (Fig. 4d). Compared with the increase of AVS in the sediment dosed with acetate, relatively less AVS was generated in the sediment dosed with methanol after 11 weeks of incubation, and extending the incubation period to the 30th week did not result in further increase of AVS in this group (Fig. 4c). This suggests that the degradation of methanol was mainly coupled to CO2 as the terminal electron acceptor. Therefore, the methanogenesis was prevalent in the sediments dosed with methanol within 30 weeks of incubation (Eq. 3).

Fig. 4
figure 4

(a) Concentration of AVS and (b) production of CH4 gas in untreated sediment and in sediment dosed with acetate, and (c) concentration of AVS and (d) production of CH4 gas in untreated sediment and in sediments dosed with methanol after 1, 6, 11, 20, and 30 weeks of incubation. Error bars represent standard deviations

Full size image
Fig. 5
figure 5

Residual SO4 2− in untreated sediment and in sediments dosed with acetate or methanol after 1, 6, 11, 20, and 30 weeks of incubation. Error bars represent standard deviations

Full size image
$$ {\mathrm{CH}}_3{\mathrm{COO}}^{-}+{\mathrm{SO}}_4^{2-}\to {\mathrm{HS}}^{-}+2{\mathrm{HCO}}_3^{-} $$
(1)
$$ {\mathrm{CH}}_3{\mathrm{COO}}^{-}+{\mathrm{H}}_2\mathrm{O}\to {\mathrm{CH}}_4+{\mathrm{H}\mathrm{CO}}_3^{-} $$
(2)
$$ 4{\mathrm{CH}}_3\mathrm{O}\mathrm{H}\to 3{\mathrm{CH}}_4+{\mathrm{H}\mathrm{CO}}_3^{-}+{\mathrm{H}}_2\mathrm{O}+{\mathrm{H}}^{+} $$
(3)

Inhibitory test

Since both sulfate reduction and methanogenesis were observed in the sediment dosed with acetate in the first 11 weeks of incubation, an inhibitory test was conducted by dosing into the sediment with both acetate and sodium molybdate to investigate the contribution of SRB to the utilization of acetate and degradation of TPH. Most SO4 2− was depleted in the sediment dosed with 10 mmol of acetate within 11 weeks of incubation (Fig. 5). However, almost no decrease of SO4 2− could be detected in the sediment dosed with both acetate and molybdate (data not shown), indicating that the sulfate reduction was completely inhibited by adding molybdate (Oremland and Taylor 1978). The consumption rate of acetate substantially decreased from 1.57 mmol L−1 day−1 (without molybdate) to 0.43 mmol L−1 day−1 (with molybdate), as shown in Supplementary Table S2, indicating that the preferential degradation of acetate was coupled to SO4 2− as the terminal electron acceptor. In the meantime, the degradation efficiency of TPH decreased from 47.30 % (without molybdate) to 17.57 % (with molybdate) within 11 weeks of incubation (Supplementary Table S2), indicating an important role of sulfate reduction in the degradation of TPH in sediment using acetate as the co-substrate.

Microbial community structure

A total number of 194 clones (of 200 in the bacterial clone library) were sequenced for the bacterial 16S rRNA gene clone libraries. These clones were classified into 33 OTUs (Fig. 6 and Supplementary Table S3). These 33 OTUs revealed a diverse bacterial community belonging to Firmicutes, Synergistetes, Planctomycetes, Thermotogae, Acidobacteria, Deltaproteobacteria, Lentisphaerae, Actinobacteria, Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria, Epsilonproteobacteria, and Bacteroidetes (Fig. 6).

Fig. 6
figure 6

Phylogenetic tree retrieved from bacterial 16S rRNA gene clone libraries constructed from original sediment and sediments dosed with acetate or methanol after 6 and 30 weeks of incubation. Bootstrap values less than 50 are not shown

Full size image

The proportion of Deltaproteobacteria increased from 6 % of the sequenced clones in the original sediment to as high as 50 % in the sediment dosed with acetate after 6 weeks of incubation, and decreased to 32 % when the incubation time was prolonged to 30 weeks (Fig. 7a). The majority of Deltaproteobacteria OTUs were affiliated with sulfate-reducing Desulfobacterales, mainly represented by the family Desulfobacteraceae including the genera Desulfobacter, Algorimarina, and Desulfosarcina. These sulfate-reducing species have been widely detected in oil-contaminated marine sediment in the Cíes Islands (Acosta-Gonzalez et al. 2013), Alaska (Kendall et al. 2006), and Japan (Higashioka et al. 2011). Clones related to Bacteroidetes also increased in the sediment dosed with acetate, from 6 % in the original sediment to 20 % after 6 weeks of incubation and further increased to 30 % after 30 weeks of incubation (Fig. 7a). Within this taxon, the OTU of B-OTU-17 was relatively more abundant, with 98 % similarity to an uncultured Bacteroidetes bacterium (Supplementary Table S3). They have been detected from the oil-contaminated subtidal sediment in the Cíes Islands (Acosta-Gonzalez et al. 2013). In the acetate-treated sediment, the relatively high proportion of clones related to Firmicutes was only detected after 30 weeks of incubation (Fig. 7a). Most of these clones were affiliated with Clostridiales, particularly Romboutsia ilealis in family Peptostreptococcaceae (B-OTU-19, as shown in Supplementary Table S3).

Fig. 7
figure 7

Relative abundance of sequenced clones in (a) bacterial 16S rRNA gene clone libraries and (b) archaeal 16S rRNA gene clone libraries constructed from original sediment and sediments dosed with acetate or methanol after 6 and 30 weeks of incubation

Full size image

Distinct from the sediment dosed with acetate, no obvious increase of Deltaproteobacteria-related clones was found in the sediment dosed with methanol. Instead, the clones related to Firmicutes were the most abundant within 30 weeks of incubation (Fig. 7a). Within this taxon, clones related to Clostridiales in OTUs of B-OTU-19 and 20 were predominant after 6 weeks of incubation (Supplementary Table S3). They were closely related to the sequences previously detected in anaerobic sediment (Sanchez-Andrea et al. 2011). The second abundant Firmicutes OTU was related to Erysipelotrichales, followed by Lactobacillales. After 30 weeks of incubation, most Firmicutes-related clones were affiliated with Clostridiales (Supplementary Table S3).

A total number of 97 clones (of 120 in the archaeal clone library) were sequenced from the original sediment and the sediments dosed with acetate or methanol after 6 and 30 weeks of incubation. These clones were classified into 17 OTUs (Fig. 8 and Supplementary Table S4). Most of the archaeal OTUs were related to Methanomicrobia, Methanobacteria, Thermoplasmata, and Thaumarchaeota (Fig. 8). Among these 17 OTUs, the clones related to Methanomicrobia were found to be predominant in original sediment and the sediment dosed with either acetate or methanol (Fig. 7b). Within this taxon, the proportion of Methanosarcinales (Ar-OTU-1, 2 and 17) in the sediment dosed with acetate increased from 47 % in original sediment to 84 % after 6 weeks of incubation and slightly decreased to 65 % after 30 weeks of incubation (Supplementary Table S4). Conversely, the proportion of Methanomicrobiales in Methanomicrobia (Ar-OTU-3, 4, and 5) did not significantly increase along the incubation period (Supplementary Table S4). In the sediment dosed with methanol, the proportion of Methanosarcinales increased from 47 % in original sediment to 63 % after 6 weeks of incubation and decreased to 35 % after 30 weeks of incubation. The proportion of Methanomicrobiales increased from 16 % in original sediment to 26 % after 6 weeks of incubation and further increased to 35 % after 30 weeks of incubation (Supplementary Table S4).

Fig. 8
figure 8

Phylogenetic tree retrieved from archaeal 16S rRNA gene clone libraries constructed from original sediment and sediments dosed with acetate or methanol after 6 and 30 weeks of incubation. Bootstrap values less than 50 are not shown

Full size image

Discussion

TPH degradation linked to various metabolic processes involved using different co-substrates

In the sediment dosed with acetate, the increase of AVS and decrease of SO4 2− were observed, while almost no CH4 gas was detected in the first week of incubation (Figs. 4a, b and 5), indicating that acetate may preferentially facilitate sulfate reduction. The inhibitory test also revealed that the consumption rate of acetate significantly decreased when using molybdate as the specific inhibitor of SRB. Interestingly, besides the sulfate reduction, methanogenesis was also observed within 6 weeks of incubation. This is probably due to the substantial decrease of SO4 2−, which acts as the electron acceptor in the sulfate reduction process (Fig. 5). A low amount of sulfate ion may limit the bioactivity of SRB (Oremland and Polcin 1982). Methanogens, thus, may not be completely outcompeted by SRB for consuming the acetate. The phenomenon of the co-existence of sulfate reduction and methanogenesis has been previously reported in the sediment underlying marine fish farms which received a high input of organic matter (Holmer and Kristensen 1994). The methanogenesis process predominated after 11 weeks of incubation, which is because most SO4 2− was depleted within 11 weeks of incubation (Fig. 4). The results above indicate that the preferential utilization of acetate and the corresponding metabolic processes involved could be affected by the available amount of SO4 2− in marine sediment. This result may have some reference value for the application of biostimulation with acetate as co-substrate in marine sediment, since the concentrations of SO4 2− in sediment pore water normally decrease strongly with the depth of sediment (Beck and Brumsack 2012). Acetate is a type of short-chain volatile fatty acid which can enter the tricarboxylic acid cycle (Lv et al. 2014) and thus be directly degraded by SRB or methanogens (Elferink et al. 1998). The different outcomes of the competition for acetate between SRB and methanogens have been observed in previous studies, as reviewed by Liamleam and Annachhatre (2007). In the present study, the degradation of acetate with either SO4 2− or CO2 as the terminal electron acceptor was observed, likely affected by the amount of residual SO4 2− present in the sediment. Notably, as for the methanogenic degradation pathway of acetate, it was proposed by Siddique et al. (2011) that acetate can be directly converted to CH4 through the acetoclastic methanogenesis process, or first converted to CO2, followed by conversion from CO2 to CH4 through the hydrogenotrophic methanogenesis process. In the present study, the methanogenic degradation of acetate was speculated to mainly proceed through the acetoclastic methanogenesis process (i.e., direct conversion from acetate to CH4), since acetoclastic methanogens (order Methanosarcinales) were found to be the predominant archaeal species in the sediment dosed with acetate (Supplementary Table S4). Different from the sediment dosed with acetate, methanogenesis was found to be prevalent alone in the sediment using methanol as the co-substrate within the whole incubation period (i.e., 30 weeks). The presence of SO4 2− in sediment does not retard methane production. The result obtained here differs from that of the previous study, which found that the onset of methanogenesis in an oil sand tailings pond required the depletion of SO4 2− (Siddique et al. 2006). As for the fate of methanol, it was reported by Weijma and Stams (2001) that methanol could be directly utilized by SRB or methanogens, or indirectly used via involvement of other microorganisms such as homoacetogens with conversion to acetate or formate, followed by utilization by SRB or methanogens. In the present study, the possibility of direct or indirect utilization of methanol by SRB could be excluded, since sulfate reduction was insignificant in the sediment dosed with methanol throughout the experiment (Fig. 4).

The addition of either acetate or methanol exhibited favorable effects on the degradation of TPH in marine sediment. However, the preferential degradation patterns for different TPH fractions with various carbon number ranges were found to be different with the addition of acetate and methanol. In the sediment dosed with methanol, in which only the methanogenesis process was prevalent within 30 weeks of incubation, the degradation of TPH was more pronounced for the fractions with lower carbon number ranges (C10 to C20) (Fig. 2b). This result may indicate that the degradation of lower carbon number ranges of TPH fractions was more favored under methanogenic condition. Up to now, studies on the degradation of petroleum hydrocarbons with different carbon number fractions under methanogenic conditions are very limited. It has been found that only LMW PAHs are degraded under methanogenic condition, while the high molecular weight (HMW) PAHs cannot be effectively removed (Johnson and Ghosh 1998; Chang et al. 2008). The preferential degradation of different TPH fractions in the sediment dosed with acetate was distinct from that dosed with methanol. The addition of acetate exhibits more favorable effects on the degradation of TPH in the ranges of both C10–C20 and C31–C40 (Fig. 2a). Unlike the methanol-treated sediment dominated by the methanogenesis process, both sulfate reduction and methanogenesis processes were observed in the sediment dosed with acetate. The preferential degradation of higher carbon number fractions (C31–C40) of TPH was likely due to proceed under sulfate reducing condition. The speculation of preferential degradation for higher carbon number fractions of TPH coupled to sulfate reduction process could be supported by previous studies. Caldwell et al. (1998) evidenced that long-chain alkanes (C15–C34) can be degraded from a weathered oil under sulfate reducing condition. Hasinger et al. (2012) found that most long-chain alkanes (C32–C34 and C38–C39) were significantly more depleted than alkanes in the low- to mid-weight carbon range (C11–C25) under sulfate reducing condition.

To our knowledge, few studies have compared the preferential degradation pattern of petroleum hydrocarbons with different carbon number fractions under sulfate-reducing and methanogenic conditions. In the present study, dissimilar preferential degradation patterns of TPH were found in the acetate-treated sediment with occurrence of both sulfate reduction and methanogenesis and in the methanol-treated sediment with the predominance of methanogenesis. Sulfate reduction was speculated to be more favorable for the degradation of TPH fractions with higher carbon number ranges (C31–C40), while the degradation of TPH fractions coupled to methanogenesis was speculated to be more important with lower carbon number ranges (C10–C20). Therefore, compared with the sediment dosed with methanol, the relatively higher removal efficiency of TPH (C10–C40) achieved in the sediment with the addition of acetate should be due to the occurrence of both sulfate reduction and methanogenesis processes over the incubation period, which may lead to the degradation of TPH fractions with a broader carbon number range. The reason for the different behaviors of TPH degradation coupled to sulfate reduction and methanogenesis processes may be related to the different microorganisms involved during the sulfidogenic and methanogenic degradation of petroleum hydrocarbons. Some of the hydrocarbon-degrading sulfate-reducing species have been found to be capable of utilizing petroleum hydrocarbons with higher carbon number fractions, such as long-chain alkanes (Caldwell et al. 1998; Miralles et al. 2007) and high molecular weight (HMW) PAHs (Coates et al. 1997; Johnson and Ghosh 1998). However, the methanogenic consortium that has been reported up to date was only capable of utilizing petroleum hydrocarbons with relatively lower carbon number fractions (Genthner et al. 1997; Siddique et al. 2006, 2011).

Effect of different co-substrates on microbial community

Clone libraries of bacterial 16S rRNA genes constructed from original sediment and sediments dosed with acetate or methanol suggested that the addition of different co-substrates led to distinct structures of bacterial community. Clones related to Deltaproteobacteria were the most abundant in the sediment dosed with acetate, followed by Bacteroidetes, while Firmicutes predominated in the sediment dosed with methanol. The fundamental difference between these co-substrates (i.e., acetate and methanol) in terms of degrees of reduction could be one of the possible reasons for the distinct bacterial structures in the sediment dosed with different co-substrates (Lv et al. 2014).

In the acetate-treated sediment, the predominance of sulfate-reducing Desulfobacteraceae (order Desulfobacterales) was consistent with that obtained from the saline sewage fed with acetate as the electron donor, which revealed that one of the two predominant species in the sludge bed was related to Desulfobacteraceae (Wang et al. 2011a). Leloup et al. (2009) reported that the members in the family Desulfobacterales can completely oxidize acetate in marine sediment. The result supports our speculation that acetate can be effectively consumed by SRB in marine sediment, and thus, sulfate reduction was facilitated. Further, inferred from the significant increase of the clone affiliated with sulfate reducing Desulfobacteraceae in the sediment dosed with acetate after 6 weeks of incubation versus that in original sediment, SRB may play an important role in the anaerobic degradation of TPH in this period. This is consistent with the results obtained from the inhibitory test. The genus of Desulfobacter, a predominant sulfate-reducing species within the family Desulfobacteraceae in the acetate-treated sediment after 6 weeks of incubation, is closely related to the sequence which has been suggested to have the potential of degrading benzene in basin sediment (Phelps et al. 1998) and other aromatic hydrocarbons such as naphthalene and toluene in subtidal sediment (Acosta-Gonzalez et al. 2013) and has also been found to be capable of growing on crude oil (Higashioka et al. 2011). Also, it has been reported that long-chain alkanes ranging from C17 to C30 were significantly removed by the bacterial community with 41–54 % of sequenced clones belonging to the family Desulfobacteraceae (Miralles et al. 2007). Therefore, the predominance of Desulfobacteraceae in the sediment dosed with acetate after 6 weeks of incubation supports our speculation that sulfate reduction plays an important role in regulating the degradation of TPH with higher carbon number ranges. Clones related to Bacteroidetes also increased in the sediment dosed with acetate over the incubation period. Bacteroidetes are normally associated with fermentative metabolism of labile high molecular weight organic matter, particularly under sulfate-reducing conditions (Llobet-Brossa et al. 1998; Acosta-Gonzalez et al. 2013). In addition, most Firmicutes-related OTUs are affiliated with Clostridiales, which has been suggested to participate in the methanogenic degradation of n-alkanes (i.e., C15–C20) as a syntrophic partner (Wang et al. 2011b). Thus, the increased proportion of these clones in acetate-treated sediment after 30 weeks of incubation implied that the methanogenic degradation of TPH got more important after the complete depletion of SO4 2−. In the sediment dosed with methanol, the predominance of Clostridiales and insignificant proportion of sulfate-reducing species support our finding that TPH was mainly degraded under methanogenic condition rather than sulfate-reducing condition in sediment with methanol as co-substrate.

A type of methanogen (Methanomicrobia) was the predominant archaea species in both acetate and methanol groups, in line with our finding, that methanogenesis occurred in both the acetate-treated and methanol-treated sediment. Although the structure of the archaeal community seems not to be significantly affected by the addition of different co-substrates (Fig. 7b), it is interesting that the proportions of acetoclastic methanogens (order Methanosarcinales) and hydrogenotrophic methanogens (order Methanomicrobiales) belonging to Methanomicrobia differed in these two treatment groups (Supplementary Table S4). The majority of Methanomicrobia-related OTUs were affiliated with acetoclastic methanogens in the sediment dosed with acetate during the incubation period. This result indicates that the acetoclastic methanogens may have been the main contributor for CH4 gas emission in the acetate-treated sediment and that the methanogenic degradation of TPH was mainly coupled to acetoclastic methanogenesis. However, the role of hydrogenotrophic methanogens cannot be discounted in the sediment dosed with methanol, as both acetoclastic methanogens and hydrogenotrophic methanogens accounted for large proportions. The petroleum hydrocarbon degradation coupled to both acetoclastic and hydrogenotrophic methanogenesis has been reported previously in oil sand tailings under methanogenic conditions (Siddique et al. 2011), and the predominant methanogenic pathway is dictated by various biological factors and thermodynamic factors (Dolfing et al. 2008). The results obtained from 16S rRNA clone libraries suggest that different co-substrates could lead to dissimilar structures of microbial community, especially the bacterial community in marine sediment, and further influence the degradation of TPH.

In the present study, the promising potential of acetate and methanol as co-substrates on enhancing the biodegradation of TPH (C10–C40) was demonstrated during the marine sediment bioremediation. Currently, the conventional method for sediment bioremediation is commonly to provide microorganisms with some electron acceptors such as O2 or NO3 − (USEPA 2005). However, the reducing conditions below the sediment surface limit the usefulness of O2 (Dell'Anno et al. 2009); meanwhile, the injection of NO3 − seems more effective in oxidizing sulfide instead of removing organic pollutants in the sediment with high sulfide content (Babin et al. 2003; Zhang et al. 2009). Therefore, this study is expected to provide environmental engineers with a better alternative option for the in situ treatment of contaminated sediments. In addition, the involvement of sulfate reduction and methanogenesis processes as well as the distinct structures of microbial community in the sediment treated by adding co-substrates (i.e., acetate and methanol) were detected in this study. The results obtained here are meaningful for better understanding the role of co-substrates (i.e., acetate and methanol) in the bioremediation of marine sediments contaminated by petroleum hydrocarbons and, more broadly, may help in determining the suitable co-substrates applied in contaminated sites possessing specific environmental conditions and petroleum-related contaminants.