Marine Pollution Bulletin 72 (2013) 165–173

Contents lists available at SciVerse ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier .com/locate /marpolbul

Evaluation of autochthonous bioaugmentation and biostimulation duringmicrocosm-simulated oil spills

M. Nikolopoulou a, N. Pasadakis b, N. Kalogerakis a,⇑a Dept. of Environmental Engineering, Technical University of Crete, Chania, Greeceb Dept. of Mineral Resources Engineering, Technical University of Crete, Chania, Greece

a r t i c l e i n f o

Keywords:Oil spillsBiostimulationAutochthonous bioaugmentationBiosurfactantsLipophilic fertilizersHydrocarbon degrading bacteria

0025-326X/$ - see front matter � 2013 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.marpolbul.2013.04.007

⇑ Corresponding author. Tel.: +30 2821037794; faxE-mail address: nicolas.kalogerakis@enveng.tuc.gr

a b s t r a c t

Oil spills are treated as a widespread problem that poses a great threat to any ecosystem. Following firstresponse actions, bioremediation has emerged as the best strategy for combating oil spills and can beenhanced by the following two complementary approaches: bioaugmentation and biostimulation. Bio-augmentation is one of the most controversial issues of bioremediation. Studies that compare the relativeperformance of bioaugmentation and biostimulation suggest that nutrient addition alone has a greatereffect on oil biodegradation than the addition of microbial products because the survival and degradationability of microbes introduced to a contaminated site are highly dependent on environmental conditions.Microbial populations grown in rich media under laboratory conditions become stressed when exposedto field conditions in which nutrient concentrations are substantially lower. There is increasing evidencethat the best approach to overcoming these barriers is the use of microorganisms from the polluted area,an approach proposed as autochthonous bioaugmentation (ABA) and defined as a bioaugmentation tech-nology that exclusively uses microorganisms indigenous to the sites (soil, sand, and water) slated fordecontamination. In this work, we examined the effectiveness of strategies combining autochthonousbioaugmentation with biostimulation for successful remediation of polluted marine environments.Seawater was collected from a pristine area (Agios Onoufrios Beach, Chania) and was placed in a biore-actor with 1% v/v crude oil to facilitate the adaptation of the indigenous microorganism population. Thepre-adapted consortium and the indigenous population were tested in combination with inorganic orlipophilic nutrients in the presence (or absence) of biosurfactants (rhamnolipids) during 90-day longexperiments. Chemical analysis (gas chromatography–mass spectrometry) of petroleum hydrocarbonsconfirmed the results of previous work demonstrating that the biodegradation processes were enhancedby the addition of lipophilic fertilizers (uric acid and lecithin) in combination with biosurfactants (rhamn-olipids), resulting in increased removal of petroleum hydrocarbons as well as reduction of the lag phasewithin 15 days of treatment. Considering this outcome and examining the results, the use of biostimula-tion additives in combination with naturally pre-adapted hydrocarbon-degrading consortia (bioaugmen-tation) has proved to be an effective treatment and is a promising strategy that could be appliedspecifically when an oil spill approaches near a shore line and an immediate hydrocarbon degradationeffort is needed.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The recent Deep Horizon oil spill accident in the Gulf of Mexicoprovided an alert and a reminder that despite the stricter environ-mental regulations that have been adopted by most countries, oilspills still remain as a serious risk to marine ecosystems. In bothsocioeconomic and ecological terms, the impact of an oil spill onthe marine environment can be quite significant. Most impor-tantly, the loss of species richness, downgraded sediment qualityand further negative impacts on offshore fish and crustacean

ll rights reserved.

: +30 2821037852.(N. Kalogerakis).

fisheries represent a subset of the side effects of oil spills (Kirbyand Law, 2008).

Conventional first response actions, such as physical removalwith booms, skimmers and absorbent materials, cannot achievecomplete clean-up of oil spills and must be deployed shortly afterthe oil spill occurs. However, when applicable, the use of chemicaldispersants is only allowed when the coastline depth is at least15 m due to their potential toxic effects on marine organisms;otherwise, their overall effectiveness is questionable.

In past years, bioremediation has emerged as an effective andenvironmentally friendly treatment for shorelines contaminatedas a result of marine oil spills. The majority of compounds in crudeoil and refined products are biodegradable and will eventually be

166 M. Nikolopoulou et al. / Marine Pollution Bulletin 72 (2013) 165–173

removed from the environment through consumption by microbes.Enhanced bioremediation aims to stimulate the rate of this processwith the following two complementary approaches: bioaugmenta-tion and biostimulation. In bioaugmentation, the addition ofoil-degrading bacteria boosts biodegradation rates, whereas inbiostimulation, the growth of indigenous hydrocarbon degradersis stimulated by the addition of nutrients (mainly N and P) or othergrowth-limiting nutrients (Nikolopoulou and Kalogerakis, 2010).

Although the effectiveness of bioaugmentation in the marineenvironment is still under investigation, the addition of oil-degrading microorganisms has been proposed as a bioremediationstrategy. Crude oil is composed of a wide range of different com-pounds, which makes it difficult for the indigenous population tocope with this broad variety of substrates, and hence, oil-degradingmicroorganisms could be added to supplement the indigenous pop-ulation (Leahy and Colwell, 1990).

Although laboratory studies on bioaugmentation have shownan enhancement of oil biodegradation, the effectiveness of bioaug-mentation has not been convincingly demonstrated in the field.Most of the field studies conducted thus far suggest that bioaug-mentation is not effective in significantly enhancing oil biodegra-dation for most environments in the long run (Nikolopoulou andKalogerakis, 2011). Generally, in most environments, it appearsthat indigenous oil-degrading microorganisms can degrade oil ifthey are not limited by the prevailing environmental conditions.Case studies support the potential of bioaugmentation as a remedi-ation strategy to combat oil spills, but this promising technology isstill in the experimental stage (El Fantroussi and Agathos, 2005).

There is increasing evidence that the best approach for overcom-ing these barriers is the use of microorganisms from the pollutedarea. A new concept in bioaugmentation, known as ‘‘autochthonousbioaugmentation’’ (ABA), has been proposed by Ueno et al. (2007)and is defined as a bioaugmentation technology that exclusivelyuses microorganisms indigenous to the sites (soil, sand, and water)slated for decontamination. Isolated single strains or enriched cul-tures, which are obtained ‘‘before’’ or ‘‘after’’ the contamination ofthe target sites, are administered to the sites once contaminationoccurs. The key concept is to conduct the enrichment of contami-nant-degrading bacteria under the same or similar conditions asthose present where the bioaugmentation will be performed. TheABA approach uses autochthonous microbial consortia or isolatesthat are highly enriched and much better adapted to chronicallyor artificially contaminated environments (Hosokawa et al.,2009). The success of oil spill bioremediation depends on the estab-lishment and maintenance of physical, chemical and biological con-ditions that favor enhanced oil biodegradation rates in the marineenvironment. Through biostimulation, the growth of indigenousoil degraders is stimulated by the addition of nutrients (Nitrogenand Phosphorous) or other growth-limiting co-substrates and/orby alterations in environmental conditions (e.g., surf-washing, oxy-gen addition by plant growth, etc.). In this study, we examined thecapabilities of an acclimated indigenous microbial consortium(ABA) sampled from a pristine environment in the presence or ab-sence of other rate limiting factors (i.e., nutrients and biosurfac-tants) (biostimulation) as a potential strategy for the successfulremediation of polluted marine environments.

2. Materials and methods

2.1. Experimental design

In this study, we examined the effectiveness of autochthonousbioaugmentation together with biostimulation versus biostimula-tion-only strategies for the successful remediation of pollutedmarine environments. Seawater was collected from a pristine envi-

ronment in the Eastern Mediterranean Sea (Agios Onoufrios Beach,Chania, Crete) and was placed in a bioreactor with 1% v/v crude oilto grow and adapt the indigenous population for later use of thisconsortium for bioaugmentation purposes. Crude oil (complimentsof Hellenic Petroleum Co., Aspropyrgos, Greece) was weatheredartificially by distillation according to ASTM method D 86. Dupli-cate microcosms were established in sterile 40-ml vial bottles con-taining 20 ml of seawater and contaminated with 0.5% w/vweathered crude oil.

Three biostimulation treatments were designed: (i) seawater +oil, (ii) seawater + oil supplemented with KNO3 and K2HPO4

(NPK) and (iii) seawater + oil supplemented with uric acid, lecithinand biosurfactant (rhamnolipids) (ULR). In addition, three autoch-thonous bioaugmentation treatments were established as shownin Table 1: (iv) seawater + oil supplemented with KNO3, K2HPO4

and pre-adapted indigenous cultures (NPKM); (v) seawater + oilsupplemented with KNO3, K2HPO4, biosurfactant (rhamnolipids)and pre-adapted indigenous cultures (NPKMR); and (vi) seawa-ter + oil supplemented with uric acid, lecithin, biosurfactant(rhamnolipids) and pre-adapted indigenous cultures (ULRM).Nutrients were added in an amount that resulted in a final concen-tration equivalent to a C:N:P molar ratio of 100:10:1. The micro-cosms were incubated under aerobic conditions at 20 �C withcontinuous agitation on an orbital shaker (200 rpm). The JBR210biosurfactant of microbial origin (rhamnolipid) consisted of ablend of C26H48O9 and C32H58O13 and was composed of 10% activeingredient supplied by Jeneil Biosurfactants Co., USA. The growthof the oil degraders was measured by the most probable number(MPN) procedure, and hydrocarbons were analyzed via chromato-graphic techniques (solid-phase extraction followed by gas chro-matography–mass spectrometry) after 0, 5, 15, 30, 60 and 90 days.

We investigated the effects of autochthonous bioaugmentationwith these organisms on hydrocarbon degradation in seawater andalso compared the role of bioaugmentation with biostimulation viadifferent types of nutrients (organic and inorganic) with or withouta rhamnolipid biosurfactant amendment.

2.2. Microbiological analyses

The amount of hydrocarbon degraders in the flasks was esti-mated by the most probable number (MPN) method according toWrenn and Venosa (1996). The growth medium was a Bushnell–Hass minimal salts medium (BHS) supplemented with crude oilas the hydrocarbon substrate. The MPN plates were 96-wellmicro-titer tissue culture plates, with each well containing 180 llBHS, 5 lL crude oil and 20 lL of the appropriate dilution of sample.One milliliter of each microcosm was diluted in a 9-mL aliquot ofBushnell–Hass solution (pH 7). Tenfold serial dilutions were car-ried out to 10�10, and the plates were inoculated by adding 20 lLof each dilution to one of the 12 rows of eight wells. The inoculatedplates were incubated at 20 �C for 2 weeks. At the end of the incu-bation period, 50 lL of p-iodonitrotetrazolium violet dye (INT 3 g/L) was added to each well of the tissue culture plates and allowedto stand at room temperature for 1 h. The dye turns from colorlessto red (when reduced) in the presence of actively respiring micro-organisms. The MPN values were calculated using the ‘‘MPN Calcu-lator’’ software program by Klee (1993) of the EPA Risk ReductionEngineering Laboratory.

2.3. Chemical analysis

2.3.1. Reagents, materials and standardsTrace analysis (SupraSolv) dichloromethane (CH2Cl2) and

n-hexane (C6H14) were obtained from Merck (Darmstadt,Germany). Solid-phase cartridges of silica/cyanopropyl (SiO2/C3–CN, 1.0/0.5 g, 6 ml) were obtained from Interchim (Best Buy Analyt-

Table 1Experimental set-up.

Treatment Weathered crude oil0.5% w/v

Nutrients(KNO3, KH2PO4)

Nutrients(uric acid, lecithin)

Rhamnolipidbiosurfactant

Pre-adapted indigenouspopulation

Control +NPK + +ULR + + +NPKM + + +NPKMR + + + +ULRM + + + +

M. Nikolopoulou et al. / Marine Pollution Bulletin 72 (2013) 165–173 167

ical, Greece), and the solid-phase cartridges were composed ofpolypropylene.

The standard hydrocarbon mix (100 ppm in hexane/DCM, 9:1)for the calibration curve that contained aliphatic hydrocarbons (n-C10- n-C35, pristane, phytane) and polycyclic aromatic hydrocar-bons (PAHs) (naphthalene, phenanthrene, anthracene, fluorene,dibenzothiophene, fluoranthene, pyrene, chrysene, benzo(b)fluo-ranthene, benzo(k)fluoranthene, benzo(e)pyrene, benzo(a)pyrene,perylene, indeno(g,h,i)pyrene, dibenzo(a,h)anthracene, benzo(1,2,3-cd)perylene) was obtained from Absolute Standards Inc. Thesemi-volatile internal standard mix containing seven deuteratedcompounds, i.e., d8-naphthalene, d12-chrysene, d12-perylene, d10-acenaphthene, d10-phenanthrene and d4-1,4-dichlorobenzene(2000 ppm in DCM), was obtained from Supelco Co. The surrogatestandards (d10-anthracene and 5a-androstane 2000 ppm each inDCM) were obtained from Supelco Co.

2.3.2. Procedure and sample preparation for spilled oilQuantification of the hydrocarbon target analytes was per-

formed by gas chromatography/mass spectrometry. The tubecontents (20 mL) were Liquid–Liquid extracted by adding approx-imately 20 mL of dichloromethane spiked with 400 lL of surrogaterecovery standard (200 ppm of each d10-anthracene and 5a-andro-stane). After mixing for several minutes, the flask was set aside toallow the dichloromethane and water layers to partition. Thedichloromethane layer was drained by passing through a funnelpacked with anhydrous sodium sulfate. Subsequently, the dichlo-romethane was evaporated in a rotavapor concentrator. The recov-ered oil was fractionated using Solid Phase Extraction cartridges asdescribed by Alzaga et al. (2004). According to this method, aknown weight of 5–10 mg of the dried oil was dissolved in n-hex-ane and was transferred onto the SiO2/C3–CN SPE cartridge andeluted under positive pressure with 4 ml of n-C6 (FI-aliphatics)and 5 ml of n-C6–DCM (1:1) (FII-aromatics). Prior to sample load-ing and before SPE fractionation, the cartridges were conditionedwith 4.0 ml of hexane. The two fractions were blown down todryness with nitrogen. The weights of the FI-aliphatics and FII-aromatics were recorded and re-dissolved in 1 mL n-C6 and 1 mLn-C6–DCM (1:1), respectively, for use in the auto-sampler of theGC/MS instrument. The final concentration of the internal stan-dards added in each fraction immediately before the injectionwas 1 ppm. This solution contained four deuterated compounds:d8-naphthalene, d10-phenanthrene, d12-chrysene and d12-perylene.For quantitative analyses, an Agilent HP 7890/5975C GC–MS sys-tem with an Agilent HP-5 5% phenyl methyl siloxane column(30 m � 250 lm � 0.25 lm) was operated in full scan mode (rangeof 50–500 m/z). The initial oven temperature was set at 60 �C,followed by a temperature ramp of 6 �C/min up to 300 �C. The sam-ples (1 lL) were injected through a split–split-less injector (pulsed/split-less mode at 250 �C) by an Agilent 7693A Automatic LiquidSampler. The transfer line, MS source and quadrupole tempera-tures were set at 280 �C, 230 �C and 150 �C, respectively.

External multilevel calibrations were carried out for bothalkanes and PAH quantification ranging from 1 ppb to 20 ppm.

The major hydrocarbons in crude oil were identified based on theirretention time and by comparison with analytical standards. Therepeatability of the entire experimental procedure is demonstratedin Fig. 1, in which the analytical data for the saturated hydrocar-bons determined from all control experiments are shown in box-plot form. For clarity, the data are presented after subtraction oftheir average value.

To ensure that the observed decline in the target analytes iscaused by biodegradation rather than by physical loss from mis-handling or inefficient extraction, it is necessary to normalize theconcentrations of the target analytes via a ‘‘conserved internalmarker’’. One conserved internal marker that is useful for quantifi-cation is C3017a(H), 21b(H)-hopane. Analytes of crude oil werenormalized to the conservative biomarker 17a(H), 21b(H) C30-ho-pane, which is naturally present in crude oil (Prince et al., 1994).The percent depletion of all analytes within the oil was calculatedusing the following equation:

% Depletion ¼ A0=H0 � AS=HS½ �A0=H0

� 100% ð1Þ

where AS is the concentration of target analyte in the sample, A0 isthe concentration of the target analyte in the initial sample, HS isthe concentration of 17a(H), 21b(H) C30-hopane in the sample,and H0 is the concentration of the 17a(H), 21b(H) C30-hopane ana-lyte in the initial sample (Prince et al., 2003).

2.4. Biodegradation kinetics

The degradation rate of petroleum hydrocarbons depends onthe biomass concentration and the specific degradation rate.Therefore, it is essential to differentiate whether the enhanceddegradation of any particular hydrocarbon compound is due toan increase in biomass or due to an increase of the specific degra-dation rate, which suggests a different metabolic pathway. Thedegradation kinetics of representative compounds from the n-alkanes group and the PAH group were investigated, namely, C15,C20, C25, C30, Pristane, Phytane, Fluorene, Dibenzothiothene, Phen-anthrene and Chrysene.

In a batch culture, the cell growth rate is given by the followingexpression:

rX ¼dXdt¼ lX ð2Þ

where X(cells/mL)is the biomass concentration and l is the specificgrowth rate (1/h). The average specific growth rate can be readilyestimated as the slope of the plot of ln(X) versus time. Similarly,the rate of any particular substrate utilization (i.e., removal of a par-ticular hydrocarbon) S ((mg-compound/mg-oil)/mL) is given by thefollowing expression:

rs ¼dSdt¼ �qsX ð3Þ

where qs is the specific degradation rate [((mg-compound/mg-oil)/mL)/(cells/mL h)]. The estimation of qs can be more reliably carried

Fig. 1. Box-plot showing the distributions of n-alkane concentrations in oil samples obtained at the 0-day of each experiment.

168 M. Nikolopoulou et al. / Marine Pollution Bulletin 72 (2013) 165–173

out using the integral method (Englezos and Kalogerakis, 2001). Inthis approach, we first integrate Eq. (3) to yield:

St � S0 ¼ ��qs

Z t

0Xdt ð4Þ

The average specific degradation rate �qs can be obtained viaLeast Squares Estimation as the slope of a linear plot of St versusthe integral I(t) =

RX(t)dt. The latter is readily computed numeri-

cally from the experimental data of X versus time (Englezos andKalogerakis, 2001).

3. Results and discussion

Evaluation of the effectiveness of each treatment on the crudeoil biodegradation rate was estimated in terms of alkanes, PAHsand the compositional changes of the hydrocarbon degradersthroughout the period of the experiment. Fig. 2 represents the totaldepletion rate of the saturated fraction of n-alkanes (C14–C35) ofthe control treatment (C) as well as of the NPK, NPKM, NPKMR,ULR and ULRM treatments at different time intervals of the exper-iment. The control had no significant effect on the degradation rate

Fig. 2. Depletion rate of C14–C35n-alkanes after 0, 5, 15, 30, 60 and 90 days o

because there were no hydrocarbon degraders detected. The NPKtreatment was also less successful than the other treatments interms of time and the quantity of hydrocarbons depleted. Onlyapproximately 20% of the n-alkanes were removed during the60 days of the experimental period. In contrast, treatments usingABA and biostimulation (NPKM, NPKMR and ULRM) and treatmentusing biostimulation with the indigenous population (ULR) werethe most effective treatments, yielding a fast degradation rate ofn-alkanes greater than 80% within 30 days of the experiment(Fig. 2). Moreover, the degradation in the NPKMR, ULR and ULRMtreatments reached 99%, 97% and 88%, respectively, within 15 days.

As shown in Figs. 4–6, in the treatments with ULR, NPKMR andULRM (in which rhamnolipids were added), a high rate of decreasewas observed for medium-chain n-alkanes (C14–C30) as well as forhigh-chain alkanes (C31–C35) over approximately 15 days of theexperiment. In treatment with NPKM, however, the high-chain al-kanes (C31–C35) remained stable for the entire duration of theexperiment, as shown in Fig. 3. The average specific consumption(degradation) rate �qs was calculated over the time period in whichthe degradation occurred, which can be observed in Fig. 1 forNPKMR, and is 0–15 days for ULR, 0–30 days for ULRM and 0–

f monitoring in control, NPK, NPKM, NPKMR, ULR and ULRM treatments.

Fig. 4. Concentration of C14–C35n-alkanes after 0, 5, 15, 30, 60 and 90 days of monitoring in NPKMR treatment.

Fig. 5. Concentration of C14–C35n-alkanes after 0, 5, 15, 30, 60 and 90 days of monitoring in ULR treatment.

Fig. 3. Concentration of C14–C35n-alkanes after 0, 5, 15, 30, 60 and 90 days of monitoring in NPKM treatment.

M. Nikolopoulou et al. / Marine Pollution Bulletin 72 (2013) 165–173 169

60 days for NPKM. Although there is nearly zero growth detectedin the control treatment, we can assume that the specific degrada-tion rate in the control treatment is quite low, with a magnitude of0.0001. If we compare the specific degradation rate of the NPKM,NPKMR, ULR and ULRM treatments (Table 2 and Fig. 11), we canestimate that the specific degradation rate in those treatmentscan exceed 106 orders of magnitude over that of the control. Ifwe compare among treatments, we can estimate that the specificdegradation rate for C15 in the NPKMR, ULR and ULRM treatmentsis 18, 13 and 1.5 times higher than that of the NPKM treatment.Additionally, the specific degradation rate for C20 is approximately16, 14 and 1.5 times higher for the NPKMR, ULR and ULRM treat-

ments than the NPKM treatment. Moreover, the specific degrada-tion rate for C25 and C30 in the NPKMR and ULRM treatments isapproximately 16 and 3 times higher than the NPKM treatment,respectively, in which the specific degradation rates for C25 andC30 in the ULR treatment are 12 and 18 times higher than in theNPKM treatment. The specific degradation rate for the heavierC35 components is 21, 37 and 6.5 times higher for the NPKM treat-ment than for the NPKMR, ULR and ULRM treatments, respectively.More specifically, for the NPKMR, ULR and ULRM treatments (inwhich the biosurfactant is present), the average specific growthrate l is approximately equal to 0.02 (1/h), whereas with the ULRMtreatment, the specific degradation rate is not equal to the other

Table 2Specific growth and degradation rates of selected alkanes.

Treatment l qs

C15 C20 C25 C30 Pristane Phytane

NPKM 0.004 74.6 58.9 18.0 9.7 45.5 48.6NPKMR 0.023 1397.9 964.1 290.2 148.1 857.9 910.6ULR 0.024 993.1 865.8 223.6 175.1 524.7 631.2ULRM 0.010 120.9 96.9 54.9 28.7 61.5 67.9

Fig. 6. Concentration of C14–C35n-alkanes after 0, 5, 15, 30, 60 and 90 days of monitoring in ULRM treatment.

Table 3Specific degradation rates of selected PAHs.

Treatment qs

Fluorene Dibenzothiothene Phenanthrene Chrysene

NPKM 0.32 0.91 0.88 0.07NPKMR 0.52 1.08 0.97 0.008ULR 2.40 7.47 7.18 1.28ULRM 0.34 0.71 0.77 0.23

Fig. 8. Concentration of fluorene, dibenzothiophene, phenanthrene and chrysene

170 M. Nikolopoulou et al. / Marine Pollution Bulletin 72 (2013) 165–173

two treatments, leading to the conclusion that in the presence oforganic lipophilic nutrients, this consortium prefers to use thiscarbon source rather than the petroleum hydrocarbons. Branchedalkanes and formerly used biomarkers such as Pristane andPhytane are also degraded 18–12 times higher in the NPKMR andULR treatments and 1.5 times higher in the ULRM treatment com-pared to that in the NPKM treatment.

In terms of the specific degradation rate, the same behavior be-tween NPKM–NPKMR–ULR–ULRM treatments is also observed forselected PAHs, as shown in Figs. 7–10. Furthermore, we observe

Fig. 7. Concentration of fluorene, dibenzothiophene, phenanthrene and chryseneafter 0, 5, 15, 30, 60 and 90 days of monitoring in NPKM treatment.

a low but decreasing rate in the low-molecular-weight PAHs (fluo-rene–dibenzothiothene–phenanthrene) compared with that of thehigh-molecular-weight PAH (chrysene), which remains practicallystable. Degradation in the ULR and NPKMR treatments of lowmolecular PAHs is achieved within 15 days, whereas for the NPKMand ULRM treatments, it is achieved within 30 days of the experi-ment as seen in Table 3 (Figs. 7–10).

The PAH degradation in the NPKMR treatment reached 60%within 15 days, equally for fluorine, dibenzothiothene and phenan-threne, which remained practically stable for the next 45 days andwas reduced to 90% (fluorine) and 75% (dibenzothiothene–phenan-threne) by the end of the experiment. Accordingly, in the ULRtreatment, the degradation rate reached 60% for fluorine and great-er than 75% for dibenzothiothene and phenanthrene during thefirst two weeks of the experiment, and all rates reached 87% bythe 30th day and remained stable until the end of the experiment,

after 0, 5, 15, 30, 60 and 90 days of monitoring in NPKMR treatment.

Fig. 9. Concentration of fluorene, dibenzothiophene, phenanthrene and chryseneafter 0, 5, 15, 30, 60 and 90 days of monitoring in ULR treatment.

Fig. 10. Concentration of fluorene, dibenzothiophene, phenanthrene and chryseneafter 0, 5, 15, 30, 60 and 90 days of monitoring in ULRM treatment.

M. Nikolopoulou et al. / Marine Pollution Bulletin 72 (2013) 165–173 171

with only fluorine reaching 98% after 60 days. In contrast, in theNPKM treatment, fluorine and dibenzothiothene were depleted>50% and phenanthrene to approximately 35% only after 30 days.Moreover, the depletion for both dibenzothiothene and phenan-threne exceeded 70%, fluorine reached 86% by the 60th day, andby the end of the experiment, all of them were completely de-pleted. A similar trend was also observed in the ULRM treatment,

Fig. 11. Concentration profiles of selected n-alkanes and PAH c

in which fluorine was decreased by 85%, whereas dibenzothioth-ene–phenanthrene decreased only to 60% after 60 days of theexperiment.

In the NPKM treatment and, to lesser extent, in the ULRM treat-ment, the late response of the degradation of the PAHs and thelong-chain-length alkanes is related to the gradual growth rate ofthe hydrocarbon degraders (30 days), which remained at a consid-erably high level (102 orders of magnitude greater in cells/mL)compared with those in the ULR and NPKMR treatments at theend of the experiment. Considering the above specific growth ratel, which for both NPKM and ULRM was 4 and 2 times higher (datanot shown) than for ULR and NPKMR after 30 days of treatment,respectively, most of the hydrocarbon degraders decreased drasti-cally after most of the oil was consumed within 15–30 days. Pre-sumably, the incomplete depletion of PAHs by the end of theexperiment could be attributed to the low number of hydrocarbondegraders, which was caused by a lack of essential nutrients thatmost likely had already been used for the depletion of the satu-rated fraction. Despite the late response of the microbial commu-nity in oil degradation, especially in the NPKM treatment, thedegradation ability of the adapted consortium for both oil fractions

ompounds in NPKM, NPKMR, ULR and ULRM treatments.

172 M. Nikolopoulou et al. / Marine Pollution Bulletin 72 (2013) 165–173

was demonstrated, and in the presence of suitable biostimulants, itcould even be accelerated.

The preferred biodegradation of the more easily biodegradablesubstrates, such as the lower-molecular-weight PAHs and thesmall-chain-length aliphatic hydrocarbons that are found in con-taminated areas, could also be associated with the difference inaqueous solubility, which decreases as the carbon number in-creases. In terms of chemical composition, as expected, the satu-rated fraction of the residual oil was degraded more extensivelythan the aromatic fraction. The trend in the degradation rate fol-lows the pattern C15 > C20 > (Pristane, Phytane) > C25 > C30 > C35 >(PAHs). Although Pristane and Phytane were previously consideredto be conserved internal markers in the biodegradation index, asshown in Fig. 5 and Table 2 (especially in ULR treatment), theywere completely degraded within 15 days, and thus, they shouldbe considered unreliable as biodegradation indices.

Comparison of the removal of the saturated fraction and themicrobial growth among the NPKM (Fig. 12), NPKMR and ULRtreatments (data not shown here) suggests that the removal ofthe saturated fraction depends on the increase in the populationof hydrocarbon degraders. The application of nutrients in the solu-tions enhanced the growth of the hydrocarbon degraders, as esti-mated by the MPN method, compared with the control solution,to which no nutrients were added (data not shown). The hydrocar-bon degrader population in the NPKMR, ULR and ULRM micro-cosms reached 106 per mL within 15 days, where nearlycomplete degradation above 90% was achieved. In contrast, in theNPKM treatment, the hydrocarbon degrader population reached105 per mL, and the degradation reached 65%. The application ofbiosurfactants in the solutions enhanced the growth of the hydro-carbon degraders within 5 days, as estimated by the MPN method,in the NPKMR, ULR and ULRM treatments.

A comparison of the profiles of the n-alkanes and PAHs (Figs. 3–10) after oil application revealed that the application of fertilizerplus biosurfactant can favor the degradation of crude oil in com-munities that are already well adapted but lack essential nutrients.It was observed in a previous study by McKew et al. (2007) that theaddition of rhamnolipid biosurfactants alone had little effect onbiodegradation; however, combination with water-soluble nutri-ent additions provoked a significant increase in biodegradation.The sole addition of biosurfactant only increases the bioavailabilityof petroleum components; however, if there is a lack of essentialnutrients (N and P), microbial activity will still be limited.

In contrast, when the added fertilizer is of organic origin, itcould result in increased consumption of the organic source ratherthan the consumption of oil. In cases in which there is an excess of

Fig. 12. % Depletion of alkanes and microbial growth curve between differenttreatments (NPKM and ULRM) during 90 days of monitoring.

biosurfactant present, any further production of biosurfactant isstopped while the microbes utilize these nutrients. As was re-ported in previous studies on liquid microcosms, when these lipo-philic nutrients are applied to the indigenous population alone(Nikolopoulou et al., 2007) or in combination with rhamnolipids(Nikolopoulou and Kalogerakis, 2008), a high degradation of petro-leum hydrocarbons can be achieved, which was also confirmed inthe present study (ULR treatment). Furthermore, most of the crudeoil saturated fractions were completely utilized after 5 days oftreatment with the fertilizer plus biosurfactant (NPKMR). The bio-degradation reached almost 50% within 5 days of incubation,whereas 70% biodegradation was reached after 15 days of incuba-tion (NPKMR).

It can be concluded that, in this study, biosurfactants (particu-larly rhamnolipids) accelerated the biodegradation of crude oilby making it more available to microorganisms, as expected inthe two ABA treatments (ULRM and NPKMR) and the biostimula-tion treatment (ULR). The bioavailability of oil hydrocarbons isthe critical factor that affects the efficiency of bioremediation inoil-contaminated environments. The ability of biosurfactants toemulsify hydrocarbon–water mixtures, enhance the water solubil-ity of hydrocarbons and thus increase the uptake and assimilationof hydrocarbons by microorganisms is highly recognized (Ron andRosenberg, 2002; Banat et al., 2010). Kinetics investigation of thespecific degradation rate (qs) supports this conclusion becausethe specific degradation rate is not only growth associated but isalso enhanced by intermediate products or biosurfactant activitythat possibly affect the metabolic pathway. However, this wasnot the case for the ULRM treatment in the presence of other or-ganic sources (lipophilic nutrients–rhamnolipids), leading to theconclusion that either other intermediate products delay the deg-radation process or this certain community prefers to use organiccarbon from organic fertilizers, a possibility that requires furtherinvestigation.

Nonetheless, the combination of lipophilic nutrients with orwithout biosurfactants in conjunction with a well-adapted indige-nous community of hydrocarbon degraders can be a promising toolfor use in cases where an immediate response to oil spill incidentsis necessary, particularly in pristine environments.

Acknowledgments

This work was funded by FP-7 PROJECT No. 266473, ‘‘Unravel-ling and exploiting Mediterranean Sea microbial diversity andecology for xenobiotics’ and pollutants’ clean up’’ – ULIXES andby the European Union (European Social Fund – ESF) and Greek na-tional funds through the Operational Program ‘‘Education and Life-long Learning’’ of the National Strategic Reference Framework(NSRF) – Research Funding Program: Heracleitus II, Investing inknowledge society through the European Social Fund.

References

Alzaga, R., Montuori, P., Ortiz, L., Bayona, J.M., Albaigés, J., 2004. Fast solid-phaseextraction–gas chromatography–mass spectrometry procedure for oilfingerprinting. J. Chromatogr. A 1025, 133–138.

Banat, I.M., Franzetti, A., GandolfiI, 2010. Microbial biosurfactants production,applications and future potential. Appl. Microbiol. Biotechnol. 87, 427–444.

Englezos, P., Kalogerakis, N., 2001. Applied Parameter Estimation for ChemicalEngineers. Marcel Dekker, New York, NY.

El Fantroussi, S., Agathos, S.N., 2005. Is bioaugmentation a feasible strategy forpollutant removal and site remediation? Curr. Opin. Microbiol. 8, 268–275.

Hosokawa, R., Nagai, M., Morikawa, M., Okuyama, H., 2009. Autochthonousbioaugmentation and its possible application to oil spills. World J. Microbiol.Biotechnol. 25, 1519–1528.

Kirby, M.F., Law, R.J., 2008. Oil spill treatment products approval: the UK approachand potential application to the Gulf region. Mar. Pollut. Bull. 56, 1243–1247.

Klee, A.J., 1993. A computer program for the determination of most probablenumber and its confidence limits. J. Microbiol. Methods 18, 91–98.

M. Nikolopoulou et al. / Marine Pollution Bulletin 72 (2013) 165–173 173

Leahy, J.G., Colwell, R.R., 1990. Microbial degradation of hydrocarbons in theenvironment. Microb. Rev. 53, 305–315.

McKew, B.A., Coulon, F., Yakimov, M., Denaro, R., Genovese, M., Smith, C.J., Osborn,A.M., Timmis, K.N., McGenity, T.J., 2007. Efficacy of intervention strategies forbioremediation of crude oil in marine systems and effects on indigenoushydrocarbonoclastic bacteria. Environ. Microbiol. 9, 1562–1571.

Nikolopoulou, M., Pasadakis, N., Kalogerakis, N., 2007. Enhanced bioremediation ofcrude oil utilizing lipophilic fertilizers. Desalination 211, 286–295.

Nikolopoulou, M., Kalogerakis, N., 2008. Enhanced bioremediation of crude oilutilizing lipophilic fertilizers combined with biosurfactants and molasses. Mar.Pollut. Bull. 56, 1855–1861.

Nikolopoulou, M., Kalogerakis, N., 2010. Biostimulation strategies for enhancedbioremediation of marine oil spills including chronic pollution. In: Timmis, K.N.(Ed.), Handbook of Hydrocarbon and Lipid Microbiology, vol. 9. Springer-VerlagBerlin Heidelberg, pp. 2521–2529.

Nikolopoulou, M., Kalogerakis, N., 2011. Petroleuml spill control with biologicalmeans. In: Moo-Young, Murray (Ed.), Comprehensive Biotechnology, vol. 6.Elsevier B.V., pp. 263–274.

Prince, R.C., Elmendorf, D.L., Lute, J.R., Hsu, C.S., Haith, C.E., Senius, J.D., 1994. 17a(H),21b(H)-hopane as a conserved internal maker for estimating thebiodegradation of crude oil. Environ. Sci. Technol. 28, 142–145.

Prince, R.C., Garrett, R.M., Bare, R.E., Grossman, M.J., Townsend, T., Suflita, J.M., Lee,K., Owens, E.H., Sergy, G.A., Braddock, J.F., Lindstrom, J.E., Lessard, R.R., 2003. Theroles of photooxidation and biodegradation in long-term weathering of crudeand heavy fuel oils. Spill Sci. Technol. Bull. 8, 145–156.

Ron, E., Rosenberg, E., 2002. Biosurfactants and oil bioremediation. Curr. Opin.Biotechnol. 13, 249–252.

Ueno, A., Ito, Y., Yumoto, I., Okuyama, H., 2007. Isolation and characterization ofbacteria from soil contaminated with diesel oil and the possible use of these inautochthonous bioaugmentation. World J. Microbiol. Biotechnol. 23, 1739–1745.

Wrenn, A., Venosa, A.D., 1996. Selective enumeration of aromatic and aliphatichydrocarbon degrading bacteria by a most-probable number procedure. Can. J.Microbiol. 42, 252–258.