Biodegradation potential assessment by using autochthonous microorganisms from the sediments from Lac Mégantic (Quebec, Canada) contaminated with light residual oil

Chemosphere 239 (2020) 124796

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Biodegradation potential assessment by using autochthonous gantic (Quebec, microorganisms from the sediments from Lac Me Canada) contaminated with light residual oil Zeyu Yang a, *, Bruce P. Hollebone a, Keval Shah a, Chun Yang a, Carl E. Brown a, Sabine Dodard b, Manon Sarrazin b, Geoffrey Sunahara b a b

Emergencies Science and Technology Section, Science and Technology Branch, Environment and Climate Change Canada, Ottawa, ON, Canada Aquatic and Crop Resource Development, National Research Council Canada, Montreal, Quebec, Canada

h i g h l i g h t s
Biodegradation of petroleum hydrocarbons depends on their chemical structure.
Higher temperatures favoured the biodegradation of petroleum hydrocarbons.
Native microorganisms from LM sediments can biodegrade petroleum hydrocarbons.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 16 July 2019 Received in revised form 3 September 2019 Accepted 5 September 2019 Available online 6 September 2019

gantic (LM), In July 2013, a fatal train derailment led to an explosion and fire in the town of Lac-Me Quebec, and the crude oil contamination of regional surface water, soil, and sediment in the adjacent gantic. This study investigated the degradation potential of the spilled crude oil by using the Lake Me sediments from the incident site as the source of microorganisms. Two light crude oils (LM source oil and Alberta Sweet Mixed Blend (ASMB)) were tested at 22  C for 4 weeks and 4  C for 8 weeks, respectively. The post-incubation biological and chemical information of the samples were analysed. There was no marked difference in degradation efficacy and biological activities for both the LM and ASMB oils, although the biodegradation potential differed between the two incubations. Higher temperature favoured the growth of microorganisms, thus for the degradation of all petroleum hydrocarbons, except for some conservative biomarkers. The degradation of both oils followed the order of resolved components > total saturated hydrocarbons (TSH) > unresolved complex mixture (UCM) >total aromatic hydrocarbons (TAH). Normal alkanes were generally degraded more significantly than branched ones, and polycyclic aromatic hydrocarbons (PAHs). Degradation of polycyclic aromatic hydrocarbons (PAHs) and their alkylated congeners (APAHs) for both incubations generally decreased as the number of aromatic rings, and the degree of alkylation increased. This study showed that the LM sediments can biodegrade the petroleum hydrocarbons efficaciously if appropriate ambient temperatures are generated to favour the growth of autochthonous microorganisms. © 2019 Elsevier Ltd. All rights reserved.

Handling Editor: Keith Maruya Keywords: gantic oil spill Lac-Me Autochthonous microorganisms Petroleum hydrocarbons Biological activities

1. Introduction gantic rail disaster occurred in the town of LacThe Lac-Me gantic, in the Eastern Townships of Quebec, Canada. On July 6, Me 2013, an unattended 74-car freight train carrying Bakken Formation

* Corresponding author. Emergencies Science and Technology Section, Environment and Climate Change Canada, 335 River Road, Ottawa, K1A0H3. E-mail address: [email protected] (Z. Yang). https://doi.org/10.1016/j.chemosphere.2019.124796 0045-6535/© 2019 Elsevier Ltd. All rights reserved.

crude oil rolled down the rail line and derailed in downtown Lacgantic. The resulting fire and explosion of multiple tank cars Me killed 47 people, destroyed much of the downtown area, and caused approximately 6 million liters of crude oil to be spilled into the environment, including the town center, the adjacent Lac gantic waterbody, and the Chaudie re River (TSBC, 2014). Me The Bakken crude oil spilled and burned in the accident is an unconventional oil imported from North Dakota, USA. It is a light oil (>30 American Petroleum Institute gravity (API)) consisting of a complex combination of paraffinic and aromatic hydrocarbons

2

Z. Yang et al. / Chemosphere 239 (2020) 124796

having small amounts of N-, S-, and O-compounds, as well as toxic metalsdmainly Cr, Fe, Ni, Pb, and V (TSBC, 2014). Following the derailment, most of the crude oil burned during the fire, resulting in the loss of the lighter carbon fractions (
LM on December 4, 2013. The sampling area was chosen based on where the initial contamination was found back in July 2013 when the derailment occurred. No obvious oil was found on the shoreline during sampling in December 2013. Sediments were stored at 4  C before further use. ASMB was used as a reference oil for comparison with the LM source oil due to their similar physicochemical properties. This comparison will help us confirm whether the autochthonous microorganism can degrade oils with similar physicochemical properties, or only for the source oil itself. The LM source oil is Bakken crude oil, which was collected from the unburnt rail car at the incident site. The representative physicochemical properties of ASMB and the LM source oil are listed in Table 1. ASMB and the LM source oil have an API of 35.7 and 40.6 , respectively; their dynamic viscosities are 6 mPa s (ASMB) and 3.8 mPa s (LM source oil) at 15  C, and flash points are 4  C for ASMB and <-5  C for LM source oil. Small amounts of resins and asphaltenes were found in both oils, especially in the LM source oil. Both ASMB and LM source oil have huge portion of saturates (~68e77%), while only 20%e26% of petroleum hydrocarbons are aromatics. Sulfur content in ASMB (0.70%) is much higher than that of LM source oil (0.08%). These parameters primarily indicate that the two oils have similar physicochemical properties, while LM source oil is relatively lighter than ASMB.

2.2. Chemicals and materials All culture media used in biodegradation study were prepared using reagent grade chemicals obtained from commercial suppliers. Ultrapure water produced from a Super-QTM water purification system (Millipore, Nepean, ON, Canada) was used throughout the present study, unless otherwise stated. All solvents used are of the highest purity available and were used without further purification (Caledon Laboratory Chemicals, Georgetown, ON, Canada). Silica gel (100e200 mesh, 150 Å, pore 1.2 cm2/g, active surface 320 m2/g) was obtained from SigmaAldrich Canada (Oakville, Ontario, Canada). Normal alkane calibration standards from n-C9 to n-C36 and polycyclic aromatic hydrocarbon (PAH) calibration certified standard mixtures were purchased from Restek (Bellefonte, PA, USA) and the National Institute of Standards and Technology (NIST, Gaithersburg, MDUSA). Deuterated internal and surrogate standards including [2H14]terphenyl (terphenyl-d14), [2H8] naphthalene (naphthalened8), [2H10]acenaphthene (acenaphthene-d10), [2H10]phenanthrene (phenanthrene-d10), [2H12]benz[a]anthracene (benz[a]anthracened12), [2H12]perylene (perylene-d12), and o-terphenyl were purchased from Supelco (Bellefonte, PA, USA). Biomarker sterane and terpane standards were obtained from Chiron (Trondheim, Norway).

Table 1 Representative physicochemical properties for ASMB and LM source oil. Parameters o

2. Experimental section 2.1. Lac M egantic sediments, LM source oil and reference oil Surface sediments (0e10 cm), mainly composed of fine sands and visible organic matters, were collected from the shoreline of

API ( ) Viscosity (mPa, 15 C) Flash point (C) Sulfur (%) weight Saturates (%) Aromatics (%) Resin (%) Ashphaltene (%) a

ASMB

LM source oil

35.7 6.0 4 0.70 67.8 26.4 4.1 1.7

40.6 3.8 <-5 0.08 77.2 20.0 2.7 0.1

Data cited from reference (Yang et al., 2017a).

a

Z. Yang et al. / Chemosphere 239 (2020) 124796

3

2.3. Assessment of biodegradation and biological activity

2.4. Analytical procedures for petroleum hydrocarbons

2.3.1. Biodegradation setup Sterilization was performed by autoclaving glassware and/or samples at 121  C for 1 h and allowing them to cool at room temperature. This autoclaving procedure was repeated at the following day. Freshwater media was prepared by dissolving K2HPO4 (1 g/L), Na2SO4 (4 g/L), MgSO4$7H2O (0.4 g/L), and KCl (1.5 g/L) into distilled water. The final media had a final pH of 7.2. A 200 mL standard N, P nutrient solution was prepared by dissolving NH4Cl (10 g), KNO3 (20 g), and K2HPO4 (5 g) in water. All prepared solutions were filtered through 0.22 mm Millipore filters before use. All biodegradation steps described below were performed under sterilized conditions in a laminar flow hood. Sixteen sterilized Erlenmeyer flasks were filled with a 200 mL freshwater medium and a 4-mL nutrient solution that had been sterilized to deactivate the culture. Sediments were thoroughly mixed, and then 20 mL sediment (wet weight 33 g) was added to the sterilized solutions in the above flask. Eight flasks (of the 16 total) were then sterilized to serve as control flasks lacking microbial activity and microbial degradation. Appropriately 400 mg of ASMB was spiked into 4 sterilized and 4 non-sterilized flask mixtures. Appropriately 400 mg of the LM source oil was then spiked into each of the remaining 4 sterilized and 4 non-sterilized flask mixtures. For each oil (ASMB and LM oil), 4 sterilized and 4 non-sterilized samples were incubated at shaking speed of 150 rpm at 4  C for eight weeks. Another set of 4 sterilized and 4 non-sterilized samples were incubated at 22  C for four weeks (See Fig. S1 for the full biodegradation experimental setup). At the end of the incubation period, we sampled the slurries for biological activity. The remaining slurries were then acidified and stored in the freezer until their use for petroleum hydrocarbon analysis. The purpose of selecting these two test temperatures is trying to cover the possible temperature range in the incident site gantic region, the temover the course of the year. In the Lac Me perature typically varies from 16  C to 24  C from winter to summer. Considering the operational feasibility of the test, the low temperature was set as 4  C, and the high temperature was set as 20  C in this study.

2.4.1. Sample information Detailed sample information for the chemical analyses is listed in Table 2. Samples used for the analysis of petroleum hydrocarbons include the LM sediment, 16 incubated samples, both with and without sterilizationdlabelled as sterilized or non-sterilized ASMB or LM under the two different incubations. In addition, the LM source oil and ASMB were analysed for their original chemical composition analysis.

2.3.2. Biological activity The number of viable microorganisms in the incubated slurries was measured by serial diluting of the slurries with a sterilized phosphate buffer to 106 for non-sterlized slurries, and to 103 for sterilized slurries. For counting the number of viable microorganisms in the collected sediment before biodegradation, 33 g sediment was diluted by mixing with 20 mL freshwater medium and 4 mL nutrient solution. Then the mixture was diluted to 103 with a sterilized phosphate buffer. A 100-mL subsample of each dilution was plated in triplicate. Petri dishes were incubated at 22  C for 5e14 days, and the colony forming units (CFU) was counted for the number of aerobic viable microorganisms. The three tube most probable number (MPN) method was also used to measure the MPN of the viable microorganisms in incubated slurries. Specifically, the incubated slurries were diluted with sterilized freshwater medium serially to 103, and then continuously diluted with sterilized Bunshell-Hass medium to 104. Two drops of oil were added to each tube, and then all the samples were incubated at 22  C for 1 and 4 weeks to determine the concentration of viable aerobic microorganisms in the slurries after biodegradation. MPN concentrations were determined by visual observation of the turbidity in the sample tube compared to the control tube (medium mixed with oil, but no inoculum).

2.4.2. Analytical procedures Before sample extraction, all glassware was rinsed successively with methanol, dichloromethane (DCM), and hexane three times before use. All biodegradation samples were a mixture of sediment and water. Therefore, once weighed and spiked with appropriate surrogates (o-terphenyl and the mixture of deuterated naphthalene, acenaphthene, phenanthrene, benz [a]anthracene, and perylene) in acetone, all samples were then separated through a 0.25 mm GF/F membrane filtration system into a sediment and water phase. All water samples were then transferred into separatory funnels for liquid/liquid extraction using DCM three times (50/50/50 mL). All sediment samples were extracted for at least 16 h using Soxhlet extraction equipment. The extracts were then combined and concentrated to an appropriate volume by rotary evaporation, and solvent-exchanged to hexane to a final volume of 10.0 mL in hexane. Procedures for sample clean-up and analysis were described (Yang et al., 2017b). Briefly, an appropriate amount of extract was transferred into a 3-g silica gel chromatography column topped with approximately 1 cm of anhydrous granular sodium sulphate for clean-up and fractionation. Hexane (12 mL) and 50% DCM in hexane (v/v, 15 mL) were used to elute the saturated and aromatic hydrocarbons, respectively. The saturated fraction was spiked with 5a-androstane as an internal standard for analysis of n-alkanes, total petroleum hydrocarbons (TPH), total saturated hydrocarbons (TSH), and biomarkers; the aromatic fraction was spiked with d144-terphenyl as an internal standard for the analysis of polycyclic aromatic hydrocarbons (PAHs) and total aromatic hydrocarbons (TAH). TPH, TSH, and TAH concentrations (n-C8 through n-C50) were determined using an Agilent 6890 gas chromatograph equipped

Table 2 Sample information. Sample code

Incubation condition

Sample information

2013/12/04-2399 2013/12/04-2400 2014/01/17-2401 2014/01/17-2402 2014/01/17-2403 2014/01/17-2404 2014/01/17-2405 2014/01/17-2406 2014/01/17-2407 2014/01/17-2408 2014/02/14-2418 2014/02/14-2419 2014/02/14-2420 2014/02/14-2421 2014/02/14-2422 2014/02/14-2423 2014/02/14-2424 2014/02/14-2425 2013/08/08-2283 2005/10/31/0512

/ / 22  C for 4 weeks 22  C for 4 weeks 22  C for 4 weeks 22  C for 4 weeks 22  C for 4 weeks 22  C for 4 weeks 22  C for 4 weeks 22  C for 4 weeks 4  C for 8 weeks 4  C for 8 weeks 4  C for 8 weeks 4  C for 8 weeks 4  C for 8 weeks 4  C for 8 weeks 4  C for 8 weeks 4  C for 8 weeks / /

gantic travel blank Lac Me gantic sediment Lac Me Sterilized ASMB replication 1 Sterilized ASMB replication 2 Non-sterilized ASMB replication 1 Non-sterilized ASMB replication 2 Sterilized LM spill replication 1 Sterilized LM spill replication 2 Non-sterilized LM spill replication Non-sterilized LM spill replication Sterilized ASMB replication 1 Sterilized ASMB replication 2 Sterilized LM spill replication 1 Sterilized LM spill replication 2 Non-sterile ASMB replication 1 Non-sterilized ASMB replication 2 Non-sterilized LM spill replication Non-sterilized LM spill replication gantic source oil Lac Me ASMB

1 2

1 2

4

Z. Yang et al. / Chemosphere 239 (2020) 124796

with a flame-ionization detector (FID) and an Agilent 7683 auto sampler. Characterization of n-alkanes, PAHs, and petroleum biomarkers was conducted using an Agilent 6890 GC system interfaced with an Agilent 5973 mass spectrometer (MS). Biomarkers, C21, C22, C23, and C24 tricyclic terpanes, hopanes (e.g.18a(H),21b(H)-22,29,30-trinorhopane (C27Ts), 17a (H),21b (H)22,29,30-trinorhopane (C27 Tm), 17a(H),21b(H)-30-norhopane (C29 ab hopane), 17a (H),21b (H)-hopane (C30ab hopane),C31eC35 (S, R)-17a (H),21b (H)-homohopanes (C31eC35 (S/R) hopanes), and C27eC295a (H),14b (H), 17b (H)steranes (C27eC29abb steranes)) were identified and quantified. Sixteen EPA priority PAHs were quantified: naphthalene (N), acenaphthylene (Acl), acenaphthene (Ace), fluorene (F), phenanthrene (P), anthracene (An), fluoranthene (Fl), pyrene (Py), benz (a)anthracene (BaA), chrysene (C), benzo(b)fluoranthene (BbF), benzo(k)fluoranthene (BkF), benzo(a)pyrene (BaP), indeno(1, 2, 3-c, d)pyrene (IP), dibenz(a, h)anthracene (DA), and benzo(g, h, i)perylene (BgP). The other quantified non-alkylated PAHs were biphenyl (Bph), benzo (e) pyrene (BeP), and perylene (Pe). The quantified alkylated PAHs included five alkylated homologous families of naphthalene (Ci-N), phenanthrene/anthracene (Ci-P), dibenzothiophene (Ci-D), fluorene (Ci-F), and chrysene (Ci-C), where i ¼ 0, 1, 2, 3, or 4) to indicate the degree of alkylation. 2.5. Quality control, quality assurance, and reporting limits For quality control and statistical analysis, a source oil for the LM spill (ESTS 2013/08/08-2283) and ASMB oil were used as the reference oil; we also ran a method blank analysis (250 mL of deionized water combined with a Soxhlet extraction blank). The recovery of the surrogate for TPH analysis was determined to be 85% ± 11% of the actual studied and blank samples as well as of the reference oil. The recovery of the PAH surrogates was 57% ± 9%, 75% ± 9%, 79% ± 9%, 87% ± 9%, and 85% ± 10% for d8-naphthalene, d10-acenaphthene, d10-phenanthrene, d12-benz[a]anthracene, and d12-perylene, respectively. These recoveries are within normal quality control limits. Instrument stability was calibrated every ten injections of samples using standard solutions in the middle level of the calibration curve. Daily calibrations were <20% for all targets. The report limits in this study range 1e7 ng/g for n-alkanes, 0.2e2.2 ng/g for PAHs, and 1.0e2.3 ng/g for biomarkers of terpanes and steranes (defined as the ratio of signal to noise (S/N) >10) by assuming 5 g dry sediment was used for analysis. 2.6. Data analysis Petroleum biomarkers, such as hopanes and steranes, are more recalcitrant to weathering through biodegradation and photooxidation (Aepplli et al., 2014; Yang et al., 2016) than n-alkanes and PAHs. The relative ratio between the measured concentration of individual targets and the conservative biomarker within a sample is used to evaluate the weathering potential of oil in the environment (Yang et al., 2016). Given the evaporation loss of some light molecular weight terpanes, the sum of terpanes and hopanes (i.e., Ts, Tm, C29 ab hopane, C30 ab hopane H30, and C31eC33 (S/R) hopanes) was used as a conservative marker in the present study. The sum of the concentrations of these biomarkers within a sample served as an index for assessing biodegradation by minimizing the sample matrix bias. We calculated the %loss of hydrocarbons based on the ratios of sterilized and non-sterilized samples. Sterilized samples were the control samplesdused to avoid any bias due to evaporation and/or photo-degradation; non-sterilized samples were the samples that simulated the biodegradation process. The %-loss of hydrocarbons relative to the sterilized samples was calculated following Wang and Fingas (1997):

%loss of hydrocarbons ¼ ½ððAo=HoÞ ­ ðAs=HsÞÞ=ðAo=HoÞ  100 (1) where A0 and Ho are the concentrations of the target components and the selected conservative biomarkers in the sterilized samples, respectively; As and Hs are the respective concentrations of the target components and the selected conservative biomarkers in the non-sterilized samples. Degradation as presented in the following sections is the average of two replications for the measured hydrocarbons.

3. Results and discussion 3.1. Characterization and quantification of the ASMB and LacM egantic source oil The LM source and ASMB oil samples had similar TPH (604 and 545 mg/g, respectively), TSH (75% and 79%), and TAH (25% and 21%) values, as well as similar resolved peaks to TPH ratios (26% and 32%, for LM and ASMB, respectively) and similar carbon distribution profiles (Table 3). Specifically, the TPH breakdown of both oil samples was 10e11% in the n-C6 to n-C10 range, 32e34% in the n-C10 to n-C16 range, 46e50% in the n-C16 to n-C34 range, and 8.5e8.8% >n-C34. The TPH composition of the LM source oil is the same as reported previously, where the TPH spanned from n-C6 to n-C50 with the n-C10 to n-C34 range being predominant. A TPH distribution for the unconventional Bakken crude oil has also been reported previously (Yang et al., 2017a). The distribution profiles of n-alkanes, biomarkers, and PAHs are shown in Fig. S2 for the ASMB, LM source oil, and LM sediment samples. In both oils, the n-alkanes were in the n-C9 to n-C40 range with maxima at n-C9 to n-C15. Alkane concentrations decreased as carbon chain length increased; this pattern is opposite to that observed for the LM sediment. Similar patterns for steranes and terpanes were shared among the three samples (Fig. S2). The biomarker concentrations in the LM oil, however, were far lower than that of the ASMB, especially for hopanes. The two oils shared similar PAH and APAH characteristics; for example, Ci-N was the most abundant group, followed by Ci-P, Ci-F, and then Ci-D and Ci-C among the alkylated PAHs (TSBC, 2014). However, the ASMB had higher levels of Ci-D than the LM source oil. We also observed a typical bell shape for most APAH groups for the various degrees of alkylation. For the other non-alkylated PAHs, 2e3 ring PAHs were more abundant than those having 4e6 rings. Specifically, biphenyl was the most abundant PAH congener ; pyrene was the second most abundant congener, followed by several light congeners.

Table 3 gantic (LM) sediment, LM source oil, and ASMB. TPH Results for Lac Me ESTS code

2013/12/04-2400 2013/08/08-2283 2005/10/31/0512

Customer sample info.

LM sediment

LM source oil

ASMB

mg/kg

mg/kg

mg/kg

44.1 14.9 29.2 33.7 66.3 5.5

604200 453622 150578 75.1 24.9 25.6

544861 430180 114681 79.0 21.0 31.5

0.00 0.62 50.4 49.0

11.4 33.9 46.2 8.49

10.0 31.6 49.7 8.77

TPH TSH TAH TSH/TPH (%) TAH/TPH (%) Resolved peaks/TPH (%) TPH distribution profile n-C6 to n-C10 n-C10 to n-C16 n-C16 to n-C34 >n-C34

Z. Yang et al. / Chemosphere 239 (2020) 124796

5

Additionally, both of htem have >35 API. All above analysis confirms that both oils are light oils having similar physicochemical properties. 3.2. Characterization of the initial biological and petrochemical properties of Lac M egantic sediments The TPH value for LM sediment was 44 mg/kg. The relative percentages of TSH/TPH and TAH/TPH were 34% and 66%, respectively. The relative ratio for resolved components to TPH was 5.5%. The most abundant TPH were found in the n-C16 to n-C34 range (50% of TPH), followed by those > n-C34 (49% of TPH). The n-alkanes in sediment ranged from n-C9 to n-C40 with the maximal at n-C25 to nC33 (Fig. S2). Biomarkers of terpanes and hopanes shared similar patterns with the two oils. The distribution patterns and concentrations of PAHs and APAHs differed between the LM sediment and the two oils. In the LM sediment, the Ci-P group was most abundant, followed by Ci-C, and then the other three series. In most APAH families, the non-alkylated PAH dominated, suggesting a pyrogenic input as an important source for these PAHs (Yang et al., 2017b, 2018). For the other non-alkylated PAHs in the LM sediment, the abundance of 2e3 ring PAHs was much lower than for 4e6 ring PAHs; this pattern further confirms the predominant pyrogenic contribution. The pyrogenic index (expressed as the ratio of the sum of the other non-alkylated PAHs to total APAHs) was relatively high, reaching 1.7 in the sediment. This index varied between 0.004 and 0.01 in the two test oils. These characteristics indicate the dominance of pyrogenic PAHs, probably a combustion by-product from the fire explosion during the incident (de Santiago-Martín, 2016). Therefore, mixed pyrogenic and petrogenic inputs are the source of PAHs in the LM sediment (Wang et al., 1999; Yang et al., 2018). The number of viable aerobic microorganisms was counted in the sediment before biodegradation. The collected LM sediment had an initial concentration of 4.9  106 (standard derivation of 2.13) colony forming units (CFU)/mL after incubated for 5 days at 22  C. The incubation of the mixture of Bushnell-Hass medium and both ASMB and LM oil without the presence of sediments for 4 weeks at 22  C only produced an estimated concentration of 5  102 MPN bacteria/mL, while no obvious bacterial growth after one week of 22  C incubation. Thus, the lake sediment contains viable aerobic microorganisms. 3.3. Number of viable microorganisms after biodegradation Viable microorganisms were formed after incubating the LM sediment with ASMB for 4 weeks at 22  C and 8 weeks at 4  C. However, the numbers of viable microorganisms did not differ markedly between the two incubations (Fig. 1). Similarly, the numbers of viable microorganisms increased evidently for the LM sediment coupled with the LM source oil after the incubations. This was particularly evident after the 4-week incubation at 22  C. These observations demonstrate that bacterial growth occurred when either ASMB or the LM source oils served as the carbon source in incubations with the LM sediment. A summary of the concentrations of viable aerobic microorganisms measured by MPN method under two separate incubation conditions is shown in Table 4. Results showed at least a 10-fold increase in MPN values compared to the initial amount in the sediment (5  102) under both test conditions. There was no major difference between the two test oils. Generally, samples incubated at 22  C for four weeks have a higher number of viable aerobic microorganisms than the samples incubated at 4  C for eight weeks, which is consistent with the measured CFU results.

Fig. 1. Number of viable bacteria in sediment before and after biodegradation.

Table 4 Summary of viable aerobic microorganisms after the biodegradation experiment.

LM source oil-1 LM source oil-2 ASMB oil-1 ASMB oil-2

Biodegradation of 4 weeks at 22  C

Biodegradation of 8 weeks at 4  C

MPNa/mL (95% interval)

MPN/mL (95% interval)

1.5  104 1.1  104 4.6  104 1.1  104

1.5  103 2.1  103 1.1  104 1.2  103

(3.7e4.2) (0.2e4.1) (0.9e20.0) (0.2e4.1)

(0.4e4.2) (0.4e4.3) (0.2e4.1) (0.4e4.2)

a Data expressed as most probable number (MPN) per mL of inoculum; mean with 95% confidence intervals in parentheses (n ¼ 3 replicates per dilution).

3.4. Degradation of the various petroleum hydrocarbon groups Resolved components in both oil samples were lost markedly (>80% for all samples), especially for the incubations at 22  C. The TPH of both oils degraded 67%e71% after 4 weeks of incubation at 22  C, and 34%e36% after 8 weeks incubation at 4  C (Fig. 2). TSH was degraded 78% at 22  C, and 41%e49% at 4  C for both oils. The oils lost 37%e49% of their TAH after the incubation at 22  C; however, only 6%e15% of TAH was lost after incubation at 4  C. Although the resolved components, TSH, and TPH in both oils decreased markedly at both temperatures, this was especially true at 22  C, TAH did not degrade as much as TSH and TPH; this was particularly evident for the 4  C incubation. Analysing the relative TPH distribution profiles at different carbon range after biodegradation (Fig. 2) shows that the petroleum hydrocarbons for all carbon ranges had been degraded; however, their degree of degradation depended on the specific carbon range. The light-molecular weight hydrocarbons (n-C10 to nC16) were mostly eliminated, followed by those located in the n-C16 to n-C34 range, and then finally hydrocarbons > n-C34. For all samples, hydrocarbon losses at 22  C were higher than those at 4  C. It is noted that the TPHs in the
6

Z. Yang et al. / Chemosphere 239 (2020) 124796

lower concentrations at 22  C than at 4  C. Therefore, severe biodegradation occurred in the 4-week/22  C incubation. Similar to the TPH biodegradation results, higher temperatures heightened the biodegradation of these aliphatic alkanes. Ratios between n-C17/pristane and n-C18/phytane decreased greatly after biodegradation but not the ratios of pristane/phytane (Fig. S5). The ratios of n-C17/pristane decreased from 2.5 to 0.2/0.3 for ASMB and from 3.9 to 0.01/0.8 for the LM source oil. Similarly, the ratios of n-C18/phytane decreased from 1.6 to 0.1/0.3 for ASMB and from 2.4 to 0.2/0.6 for the LM source oil. This indicates that pristine and phytane were less biodegraded than the corresponding n-alkanes of the same carbon length. The predominance of multi-methylated alkanes following the biodegradation suggests that normal alkanes were removed more rapidly than the monoand multi-methylated alkanes through biodegradation. Normal alkanes are more vulnerable to biodegradation than branched alkanes, cyclic alkanes, or substituted/unsubstituted aromatics (Dutta and Harayama, 2000). However, the photolysis efficacy of branched alkanes is higher than that of n-alkanes (D'Auria et al., 2008; Yang et al., 2016). The ratios at 22  C were generally higher than those at 4  C. This reflects more branched alkanes being biodegraded at 22  C than at 4  C. It is reasonable because greater growth of microorganisms occurred at the higher temperature, which results in microorganisms consuming more petroleum hydrocarbons as their feeding carbon source. Fig. 3 depicts the degradation of individual and total alkanes for both incubations. For ASMB, the degradation of individual alkanes

Fig. 2. Degradation of petroleum hydrocarbons after incubation, one incubation at 4  C for 8 weeks, another at 22  C for 4 weeks.

biodegradation. Theoretically, saturated hydrocarbons degrade more easily than substituted/unsubstituted aromatics when biodegradation is the main weathering process (Dutta and Harayama, 2000). On the other hand, the depletion of aromatic hydrocarbons should be faster than aliphatic hydrocarbons when photochemical oxidation is the major aging contributor (Yang et al., 2016). Therefore, the rapid decay of saturated hydrocarbons, such as TSH and n-alkanes, suggests that biodegradation, rather than photochemical oxidation, was the major weathering process responsible for their loss in this study. 3.5. Biodegradation of alkanes Individual n-alkanes were greatly biodegraded at both temperatures. Markedly lower concentrations of alkanes were detected in all non-sterilized samples compared to sterilized samples (Fig. S3 and Fig. S4). The distribution patterns of alkanes in both sterilized oils had a typical bell shape from n-C9 to n-C40. This differed from the patterns observed for the two original oils prior to incubation. This discrepancy can be ascribed to evaporation loss during incubation. The concentrations of alkanes in all the duplicated sterilized samples are consistent with each other, suggesting good reproducibility for incubation tests. The concentrations of individual alkanes in non-sterilized samples at 22  C were generally lower than those at 4  C, especially for normal alkanes having light molecular weights. Branched alkanes, such as pristine and phytane, had much

Fig. 3. Degradation of n-alkanes after incubation, one incubation at 4  C for 8 weeks, another at 22  C for 4 weeks.

Z. Yang et al. / Chemosphere 239 (2020) 124796

7

varied from 21% to 100% after 8 weeks of incubation at 4  C, whereas they varied from 92% to 100% after 4 weeks of incubation at 22  C. For the LM source oil, degradation of all alkanes was >90% for both incubations. Some heavy molecular weight n-alkanes in the LM source oil had greater degradation than those in the ASMB after 8 weeks of incubation at 4  C. The pre-incubation abundance of high-molecular weight n-alkanes in the LM source oil was much less than that in the ASMB (Fig. S1). Therefore, the difference in degradation for these heavy molecular weight n-alkanes can be ascribed to this variance in their initial concentrations. For both oils, the increase in the carbon chain length of n-alkanes is accompanied by a decreased degradation at 4  C but not at 22  C. Light-saturated hydrocarbons, such as normal alkanes, are more biodegradable than heavier ones (Bento et al., 2005; Pontes et al., 2013). The biodegradation of total alkanes reached as high as 96% after the 8-week incubation at 4  C and 99% after the 4-week incubation at 22  C. Alkanes were easily biodegraded under both incubations. Although the degradation of some heavy n-alkanes in the ASMB was limited at 4  C, the degradation of total alkanes reached up to 96% as the contents of heavy n-alkanes can be negligible compared to the lighter homologs. Accordingly, the dominant lighter alkanes controlled the degradation of alkanes in both study oils. 3.6. Biodegradation of biomarkers Unlike n-alkanes, the biomarker distribution profiles for all biodegradation samples and the controls were similar (Fig. S6 and Fig. S7). Thus, hopanes and steranes were relatively nonbiodegradable compared to n-alkanes, phytane, and pristine, as observed previously (Wang and Stout, 2007). However, there were few noticeable changes in some samples. In both oils, congeners like C21 to C24 terpanes, experienced obvious degradation at 22  C, whereas almost no degradation of these congeners occurred at 4  C (Fig. 4). For ASMB, the loss of some light molecular biomarkers, such as C21 to C24 terpanes, was high, reaching 60%e70% at 22  C. No obvious loss was observed for all other hopanes and steranes, except for C27 abb steranes that experienced a degradation maximum of 50%. For the LM source oil, only C27 abb steranes were lost (approximately 45%) at 22  C. All other congeners were relatively stable in both 4  C and 22  C incubations. Higher temperatures appear to have also contributed to the loss of some biomarkers, although these biomarkers are less biodegradable than alkanes given that most of the alkanes were digested exhaustively at 22  C. Some light molecular biomarkers may have been evaporated or consumed by microorganisms at 22  C, but not at 4  C. The loss of C27 abb steranes for both oils highlights their preferential depletion from biodegradation. The aerobic biodegradation of steranes was previously reported based on a laboratory-based analysis of oil biodegradation over 90 days (Cai et al., 2013). (Yu et al., 2018) showed that the order of susceptibility to bacterial degradation for regular steranes in reservoirs, laboratories seeps, and tars is aaaR >abbR >aaaS >abbS with C27 > C28 > C29. 3.7. Biodegradation of alkylated PAHs and the other non-alkylated PAHs 3.7.1. APAHs All sterilized samples had similar APAH patterns as their original oils (Fig. S8 and Fig. S9). As mentioned previously, naphthalene and its alkylated derivatives were the dominant homologs among the five detected PAH series in the original test oils, same with all the sterilized oils. The most noticeable change in APAHs was the dramatic decrease in abundance of naphthalene in the non-sterilized oils in both incubations, yet especially at 22  C. With the

Fig. 4. Degradation of individual biomarkers after incubation, one incubation at 4  C for 8 weeks, another at 22  C for 4 weeks.

exception of a pronounced decrease in the concentration of nonsterilized samples incubated at 22  C, the distribution patterns of APAHs also changed markedly. Furthermore, we observed a typical biodegraded petroleum APAH distribution pattern for all other series with the exception of the naphthalene series in ASMB (Wang et al., 1999; Wang and Stout, 2007). Specifically, the abundance of APAH congeners within one family followed the order of C0- < C1 C4-) for the naphthalene, phenanthrene, and dibenzothiophene series. However, the fluorene and chrysene series increased with greater alkylation (i.e., C0- < C1< C2- < C3-). The degradation of individual APAHs is depicted in Fig. 5. Except for the chrysene series, after 4 weeks of incubation at 22  C, the other four families were depleted almost entirely. The degradation of the chrysene series decreased as the alkylated degree increased (i.e., C0 > C1 > C2 > C3). The APAHs, after being incubated for 8 weeks at 4  C, had much lower degradation than those at 22  C for most of them. At 4  C, non-alkylated PAHs, such as C0eN, C0eP, C0eF, and C0-D, were fully depleted; however, their biodegradation decreased with increased alkylation, as reported previously (Yang et al., 2018) and opposite to that of the photolysis behaviour of

8

Z. Yang et al. / Chemosphere 239 (2020) 124796

Fig. 5. Degradation of individual APAHs after incubation, one incubation at 4  C for 8 weeks, another at 22  C for 4 weeks. Fig. 6. Degradation of the grouped and total PAHs and APAHs after incubation, one incubation at 4  C for 8 weeks, another at 22  C for 4 weeks.

APAHs (D'Auria et al., 2009; Yang et al., 2016). Chrysenes were not depleted in this condition. Similarly, C4eN, C2eC4eP, C2eC3-D, and C2eC3eF were not depleted. The degradation of all the other APAH congeners ranged from 50% to 100% for both oils. Consistent with previous reports, the %loss of APAHs at 4  C was generally lower than that of alkanes as aromatics are not as vulnerable to biodegradation as n-alkanes (Dutta and Harayama, 2000; Yang et al., 2018). Degradation of APAHs was also found to decrease as aromatic ring number increased (Fig. 6). Specifically, the naphthalene series was the most degraded, followed by the phenanthrene, dibenzothiophene, fluroene, and chrysene families after 8 weeks at 4  C for both oils. Almost all Ci-N, Ci-P, Ci-F, and Ci-D were depleted, whereas the Ci-C of both oils were only somewhat degraded after incubation for 4 weeks at 22  C. The number of aromatic rings does not appear to be critical for the degradation of the 2e3 ring PAHs at 22  C. Similar to the results of the previous sections, temperature was a key parameter affecting the biodegradation of APAHs. Biodegradation at 22  C was superior to that at 4  C as higher temperatures favour the growth of microorganisms, accelerating the degradation of petroleum hydrocarbons, especially those having a high bioavailability. In addition to temperature, the chemical structure of APAHs, such as their degree of alkylation and the number of rings, are critical controls of the potential biodegradation of APAHs (Wang et al., 2018).

3.7.2. Other non-alkylated PAHs In both incubated oils, non-alkylated PAHs were not as abundant as the alkylated congeners. Their original pyogenic index (PI), defined as the ratio between the sum of the measured APAHs and the sum of other non-alkylated PAHs, was developed to identify the source of PAHs by Wang et al. (1999), was ca. 0.01. This value shifted to 0.24e0.35 after the 4-week incubation at 22  C; however, we observed no alteration after the 8-week incubation at 4  C. This suggests that more APAHs were degraded at 22  C than nonalkylated PAHs. However, the limited bacterial activity inhibited the degradation of both APAHs and PAHs at 4  C. The degradation of these non-alkylated PAHs ranged from 0% to 99% for both the ASMB and the LM source oil at 22  C (Fig. 7). Specifically, the degradation of non-alkylated PAHs having 5e6 rings was negligible, whereas those having a lower number of rings experienced greater biodegradation for ASMB. LM source oil had negligible degradation of the 6-ring PAHs; all other PAHs were markedly degraded (degradation ranging from 30% to 99%). After incubated for 8 weeks at 4  C, most PAHs (from 3 to 6 rings) were not depleted in both oils. Only certain 2-ring PAHs, such as biphenyl, acenaphthylene, and acenaphthene, were greatly depleted. The loss for these 2-ring PAHs ranged from 87 to 99% in the ASMB and from 62% to 99% in the LM source oil. The loss of

Z. Yang et al. / Chemosphere 239 (2020) 124796

9

resolved components biodegraded faster than unresolved complex mixture (UCM), especially at a higher temperature. For the various hydrocarbons, almost 100% of n-alkanes were degraded at 22  C. Degradation at 4  C, however, was much more limited, especially for pristine, phytane, and some high-molecular weight n-alkanes. Most of the biomarkers (steranes, hopanes, and terpanes) were non-biodegradable compared to n-alkanes and PAHs. Nonalkylated PAHs and particularly APAHs were biodegraded at 22  C, but not at 4  C. The chrysene series is the most stable homologs biodegraded among the five APAH series. The degradation at 4  C of each APAH series followed the order C0 > C1 > C2 > C3 > C4, except for the naphthalene series. For other non-alkylated PAHs, their degradation generally decreased with an increased number of aromatic rings at 22  C; however, degradation was minimal for the other 3e6 ring PAHs at 4  C. Accordingly, higher temperatures favoured the degradation of all petroleum hydrocarbons, and the chemical structure of the various petroleum hydrocarbons controlled their respective degradation. In conclusion, sediment gantic can degrade the contaminated from the spill site of Lac Me petroleum hydrocarbons effectively if an appropriate ambient temperature is provided to favour the growth of autochthonous microorganisms. Acknowledgments This work was funded and supported by the World Class Tanker Safety System (WCTSS) and Ocean Protection Plan programs funded by Government of Canada. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.chemosphere.2019.124796. Fig. 7. Degradation of the other non-alkylated PAHs after incubation, one incubation at 4  C for 8 weeks, another at 22  C for 4 weeks.

these 2-ring PAHs may be ascribed to microbial degradation and/or evaporative loss, although sterilized samples were used as controls to eliminate the loss through evaporation. Similar to our results obtained in the earlier sections, higher temperatures accelerated the biodegradation of both APAHs and non-alkylated PAHs. The degradation of the various non-alkylated PAHs depended on the number of aromatic rings when the bacterial activity was sufficiently high. Generally, degradation decreased with the increase in the number of aromatic rings at 22  C. At 4  C, some 2e3 ring non-alkylated PAHs also followed this rule; degradation was limited for the other 3e6 ring parent PAHs due to the inadequate level of bacterial activity. Typically, the biodegradation process is dependent on the bioavailability of organic matter, given the greater potential of microorganisms to interact with dissolved organic matter through the water phase. The ability of organic matter to dissolve depends on its chemical structure. For PAHs having fewer aromatic rings and less alkylation, their water solubility is greater than PAHs having a greater number of rings and a greater degree of alkylation. This results in these PAHs having a greater bioavailability than heavier PAHs, and as such, higher biodegradation was observed.

4. Conclusions This study showed evidence that the light, saturated hydrocarbons biodegraded faster than aromatic hydrocarbons, and the

References Abed, R.M.M., Al-Sabahi, J., Al-Maqrashi, F., Al-Habsi, A., Al-Hinai, M., 2014. Characterization of hydrocarbon-degrading bacteria isolated from oil-contaminated sediments in the Sultanate of Oman and evaluation of bioaugmentation and biostimulation approaches in microcosm experiments. Int. Biodeterior. Biodegrad. 89, 58e66. Aepplli, C., Nelson, R.K., Radovi c, J.R., Camichael, C.A., Valentine, D.L., Reddy, C.M., 2014. Recalcitrance and degradation of petroleum biomarkers upon abiotic and biotic natural weathering of Deepwater Horizon oil. Environ. Sci. Technol. 48, 6726e6734. Agarwal, A., Liu, Y., 2015. Remediation technologies for oil-contaminated sediments. Mar. Pollut. Bull. 101, 483e490. Atlas, R.M., Hazen, T.C., 2011. Oil biodegradation and bioremediation: a tale of the two worst spills in U.S. history. Environ. Sci. Technol. 45, 6709e6715. Bento, F.M., Camargo, F.A.O., Okeke, B.C., Franenberger, W.T., 2005. Comparative bioremediation of soils contaminated with diesel oil by natural attenuation, biostimulation and bioaugmentation. Bioresour. Technol. 96, 1049e1055. Cai, M., Yao, J., Yang, H., Wang, R., Masakorala, K., 2013. Aerobic biodegradation process of petroleum and pathway of main compounds in water flooding well of Dagang oil field. Bioresour. Technol. 144, 100e106. D'Auria, M., Emanuele, L., Racioppi, R., Velluzzi, V., 2009. Photochemical degradation of crude oil: comparison between direct irradiation, photocatalysis, and photocatalysis on zeolite. J. Hazard Mater. 164, 32e38. D'Auria, M., Racioppi, R., Velluzzi, V., 2008. Photodegradation of crude oil: liquid injection and headspace solid-phase microextraction for crude oil analysis by Gas Chromatography with Mass Spectrometer Detector. J. Chromatogr. Sci. 46, 339e344. gantic, Canada: environmental de Santiago-Martín, A., 2016. Oil spill in Lac-Me monitoring and remediation. Int. J. Water Wastewater Treat. 2 (1), 1e6. Duan, J., Liu, W., Zhao, X., Han, Y., O'Reilly, S.E., Zhao, D., 2018. Study of residual oil in Bay Jimmy sediment 5 years after the Deepwater Horizon oil spill: persistence of sediment retained oil hydrocarbons and effect of dispersants on desorption. Sci. Total Environ. 618, 1244e1253. Dutta, T.K., Harayama, S., 2000. Fate of crude oil by the combination of photooxidation and biodegradation. Environ. Sci. Technol. 34, 1500e1505. Ganesh Kumar, A., Vijayakumar, L., Joshi, G., Magesh Peter, D., Dharani, G., Kirubagaran, R., 2014. Biodegradation of complex hydrocarbons in spent engine oil by novel bacterial consortium isolated from deep sea sediment. Bioresour. Technol. 170, 556e564.

10

Z. Yang et al. / Chemosphere 239 (2020) 124796

Kucharzyk, K.H., Benotti, M., Darlington, R., Lalgudi, R., 2018. Enhanced biodegradation of sediment-bound heavily weathered crude oil with ligninolytic enzymes encapsulated in calcium-alginate beads. J. Hazard Mater. 357, 498e505. Pontes, J., Mucha, A.P., Santos, H., Reis, I., Bordalo, A., Basto, M.C., Bernabeu, A., Almeida, C.M.R., 2013. Potential of bioremediation of buried oil removal in beaches after an oil spill. Mar. Pollut. Bull. 76, 258e265. TSBC, 2014. Railway Investigation Report-R13d0054. Runaway and Main-Track gantic. In: Montreal, Maine & Derailment. Sherbrooke Subdivision Lac-Me Atlantic Railway Freight Train MMA-002 Mile 0, vol. 23. Transportation Safety Board of Canada, Quebec, p. 3. Usman, M., Hanna, K., Faure, P., 2018. Remediation of oil-contaminated harbor sediments by chemical oxidation. Sci. Total Environ. 634, 1100e1107. Vergeynst, L., Christensen, J.H., Kjeldsen, K.U., Meire, L., Boone, W., Malmquist, L.M.V., Rysgaard, S., 2019. In situ biodegradation, photooxidation and dissolution of petroleum compounds in Arctic seawater and sea ice. Water Res. 148, 459e468. Wang, C., Liu, X., Guo, J., Lv, Y., Li, Y., 2018. Biodegradation of marine oil spill residues using aboriginal bacterial consortium based on Penglai 19-3 oil spill accident, China. Ecotoxicol. Environ. Saf. 159, 20e27. Wang, Z., Fingas, M., 1997. Developments in the analysis of petroleum hydrocarbons in oils, petroleum products and oil-spill-related environmental samples by gas chromatography. J. Chromatogr. A 774, 51e78. Wang, Z., Fingas, M., Shu, Y.Y., Sigouin, L., Landriault, M., Lambert, P., 1999. Quantitative characterization of PAHs in burn residue and soot samples and differentiation of pyrogenic PAHs from petrogenic PAHs-The 1994 mobile burn study Environ. Sci. Technol. 33, 3100e3109.

Wang, Z., Stout, S.A., 2007. Oil Spill Environmental Forensics. Elsevier, MA, USA. Xu, R., Lau, A.N.L., Lim, Y.G., Obbard, J.P., 2005. Bioremediation of oil-contaminated sediments on an inter-tidal shoreline using a slow-release fertilizer and chitosan. Mar. Pollut. Bull. 51, 1062e1070. Yang, C., Lambert, P., Zhang, G., Yang, Z., Landriault, M., Hollebone, B., Fieldhouse, B., Mirnaghi, F., Brown, C.E., 2017a. Characterization of chemical fingerprints of unconventional Bakken crude oil. Environ. Pollut. 230, 609e620. Yang, Z., Hollebone, B.P., Brown, C.E., Yang, C., Wang, Z., Zhang, G., Lambert, P., Landriault, M., Shah, K., 2016. The photolytic behavior of diluted bitumen in simulated seawater by exposed to the natural sunlight. Fuel 186, 128e139. Yang, Z., Hollebone, B.P., Laforest, S., Lambert, P., Brown, C.E., Yang, C., Shah, K., Landriault, M., Goldthorp, M., 2017b. Occurrence, source and ecological assessment of baseline hydrocarbons in the intertidal marine sediments along the shoreline of Douglas Channel to Hecate Strait in British Columbia. Mar. Pollut. Bull. 122, 450e455. Yang, Z., Shah, K., Laforest, S., Hollebone, B.P., Lambert, P., Brown, C.E., Yang, C., Goldthorp, M., 2018. A study of the 46-year-old Arrow oil spill: persistence of oil residues and variability in oil contamination along Chedabucto Bay, Nova Scotia, Canada. J. Clean. Prod. 198, 1459e1473. Yu, X., Zhang, W., Liu, X., Lei, J., Lin, Z., Yao, Z., Yao, X., Jin, X., Yang, H., Huang, H., 2018. The distribution of and biodegradation impact on spilled oil in sediments from Dalian Bay, NE China. Mar. Pollut. Bull. 135, 1007e1015. Zhao, G., Sheng, Y., Wang, C., Yang, J., Wang, Q., Chen, L., 2018. In situ microbial remediation of crude oil-soaked marine sediments using zeolite carrier with a polymer coating. Mar. Pollut. Bull. 129, 172e178.