R T a c R E a C b 1 a A R R 2 A A K A P G R S 1 m d m h 1 Environmental Toxicology and Pharmacology 50 (2017) 119–127 Contents lists available at ScienceDirect Environmental Toxicology and Pharmacology j o ur na l ho mepage: www.elsev ier .com/ locate /e tap esearch Paper he effects of fluoride based fire-fighting foams on soil microbiota ctivity and plant growth during natural attenuation of perfluorinated ompounds enato Nallin Montagnolli a, Paulo Renato Matos Lopesb, Jaqueline Matos Cruza, lis Marina Turini Claroa, Gabriela Mercuri Quiterioa, Ederio Dino Bidoiaa,∗ Departamento de Bioquímica e Microbiologia, Instituto de Biociências, UNESP – São Paulo State University, Avenida 24 A, 1515–Bela Vista, 13506-900, Rio laro-SP, Brazil Faculdade de Ciências Agrárias e Tecnológicas, UNESP – São Paulo State University, Rodovia Comandante João Ribeiro de Barros (SP 294), Km 651, 7900-000, Dracena-SP, Brazil r t i c l e i n f o rticle history: eceived 1 November 2016 eceived in revised form 7 December 2016 ccepted 26 January 2017 vailable online 30 January 2017 eywords: FFF erluorinated compounds ermination esponse surface analysis oil toxicity a b s t r a c t The use of fluoride based foams increases the effectiveness of fire-fighting operations, but they are also accompanied by major drawbacks regarding environmental safety of perfluorinated compounds (PFCs). The main concern with PFCs release is due to their well-known persistence and bioaccumulative poten- tial, as they have been detected in many environmental samples. There is a significant knowledge gap on PFC toxicity to plants, even though such data could be useful towards bioremediation procedures. It is consensus that a realistic assessment of fire-fighting foam toxicity should cover as many test organisms as possible, however, few studies combine the performance of ecotoxicological tests with a detailed study of microbial communities in soil contaminated with firefighting foams. Our research evaluated the effects of natural attenuation of PFCs on the development of arugula and lettuce seeds. The effects of variable PFCs amounts were also observed in soil microbiota using the 2,6 dichlorophenol-indophenol redox dye as microbial metabolism indicator. We aimed to determine whether aqueous film forming foams toxicity increased or decreased over time in a simulated contamination scenario. We argued that the long-term biotransformation of fire-fighting foams should be taken in to account when evaluating toxicity, focus- ing on a time-based monitoring analysis, since potentially toxic intermediates may be formed though biodegradation. The phyto-toxicity of PFCs to lettuce and arugula was high, increasing as a function of the concentration and decreasing as a function of exposure time to the environment. However, very spe- cific concentrations throughout biodegradation result in the formation of non-inhibiting intermediates. Therefore, variable biodegradation-dependent germination rates may be misleading on non-time-based monitoring approaches. Also, the low phyto-toxicity after 240 days does not exclude the potential for PFC bioaccumulation in plants. We also proposed that the colorimetric data modelling could also establish a novel toxicity parameter to evaluate the release impacts to soil and biota. The combined assays allowed the monitoring of PFCs during long-term exposition to plants as well as their immediate effects on the same soil microbiota. © 2017 Elsevier B.V. All rights reserved. . Introduction Firefighting foams have been developed for better adhesion to aterials on fire, producing a continuous coating on it. Their low ensity allows better spreadability over the surface of burning aterials, covering and isolating them from atmospheric oxygen. ∗ Corresponding author. E-mail address: ederio@rc.unesp.br (E.D. Bidoia). ttp://dx.doi.org/10.1016/j.etap.2017.01.017 382-6689/© 2017 Elsevier B.V. All rights reserved. The suppression of oxygen and the cooling of burning materials prevent re-ignition. The use of fluorine-containing aqueous film forming foams (AFFF) has improved the firefighting effectiveness of hydrocarbon related operations (Sardqvist, 2002) aiming to ensure the promptness and safety of firefighting techniques. The AFFF advantages are clear, but they are also accompanied by major draw- backs concerning environmental safety. Firefighting foams contain various substances to achieve proper formation of foam and grant its functional properties. Most of the foam compounds are fluorinated surfactants or hydrolyzed dx.doi.org/10.1016/j.etap.2017.01.017 http://www.sciencedirect.com/science/journal/13826689 http://www.elsevier.com/locate/etap http://crossmark.crossref.org/dialog/?doi=10.1016/j.etap.2017.01.017&domain=pdf mailto:ederio@rc.unesp.br dx.doi.org/10.1016/j.etap.2017.01.017 1 xicolo p a ( a b t a t fl f p c c b r p a r d t i ( ( w b ( s 2 a b e 2 C h v 2 w r ( a 2 t u 2 a 2 i t a o 2 r o t e t i c A t t i between the inner soil and the external environment. The recipi- ent was buried 5 cm from the surface. We used the retail 6% AFFF (Sintex S1371/11) formulation, commercially available and com- monly applied to petroleum fires in Brazil. Its formulation contains Table 1 Phyto-toxicity bioassays. Sample ID Components AFFF Concentration Biodegradation time C0 Soil – 0 days C60 Soil – 60 days C120 Soil – 120 days C180 Soil – 180 days C240 Soil – 240 days F0 Soil + AFFF 0.1x, 1x and 10 x m/v 0 days 20 R.N. Montagnolli et al. / Environmental To roteins, solvents and water. Commercial formulations of AFFF re complex mixtures whose main components are a solvent typically a glycol ether), fluorinated surfactants (amphoteric or nionic partially fluorinated or perfluorinated), and surfactant- ased hydrocarbons. Fluorinated surfactants in AFFF contribute o its performance when extinguishing fires (Kissa, 1994; Alm nd Stern, 1992; Falk, 1978). The presence of fluorine contributes o the rigidity of perfluorocarbon chains (Key et al., 1997). The uoro-carbon bond is strongly polarized. Fluorination also rein- orces C C adjacent bonds (Hudlicky and Pavlath, 1995). Therefore, erfluorinated surfactants are more thermally stable than their orresponding hydrocarbon analogs. In particular, perfluorinated arboxylic acids (PFCAs) and perfluorinated sulfonic acids (PFSAs), oth found in AFFF, are among the most thermally stable perfluo- inated compound (PFC) groups. In addition to thermal stability, erfluorinated surfactants are stable to acids, bases, oxidizers nd reducers. This stability allows fluorinated compounds to emain intact in environments where hydrocarbon surfactants are egraded (Kissa, 1994). The main concern with AFFF release is due to PFCs persis- ence and their bioaccumulative potential. They have been detected n many environmental samples, including air, surface water Murakami et al., 2008; Kim and Kannan, 2007), waste waters Sinclair and Kannan, 2006), soil (Higgins et al., 2005) and ground- ater (Schultz et al., 2004). PFCs were also found accumulating in iota, including mammals (Giesy and Kannan, 2002) and humans Hölzer et al., 2008). The PFCAs, PFSAs and their potential precur- ors have attracted attention as global contaminants (Buck et al., 011). PFCAs and long chain PFSAs are described as very problem- tic because they are highly persistent (Frömel and Knepper, 2010), ioaccumulative and found scattered almost universally in abiotic nvironments (Rayne and Forest, 2009), in biota (Giesy and Kannan, 001), food (Clarke and Smith, 2011) and humans (Vestergren and ousins, 2009). As a result, many firefighting foams based on PFCs ad their production restricted and were listed as substances of ery high concern in European Regulation of Chemicals (ECHA, 013). The AFFFs are predominantly released in the form of liquid foam, hich increases the potential of PFAS to penetrate in aquatic envi- onments. The PFC inflow to the medium may occur via four routes: i) release of volatile PFCs into the atmosphere (Dinglasan-Panlilio nd Mabury, 2006), which is oxidized photochemically (Ellis et al., 004) and back to the water cycle by precipitation; (ii) discharge o wastewater treatment plants (Yu et al., 2009); (iii) discharge of rban runoff contaminated by diffuse sources (Houtz and Sedlak, 012; Zushi and Masunaga, 2009), and (iv) the infiltration of waste nd spills in groundwater (Moody and Field, 2000; Moody et al., 003). Even though the PFC abiotic routes have been continuously nvestigated, there is still a lack of knowledge about the PFCs from he toxicological standpoint in many organisms. Toxicity data can be used to better remediate AFFF contaminated reas. Toxicity tests are based on determining the potential impact f pollutants towards biota in a set environment (Hagner et al., 010). Thus, ecotoxicological datasets have long been used with elative success as an additional tool for monitoring the efficiency f soil bioremediation, making it essential to assess environmen- al hazards in contamination scenarios (Lladó et al., 2012; Sheppard t al., 2011). However, few studies combine the performance of eco- oxicological tests with a detailed study of microbial communities n soil contaminated with firefighting foams. It is consensus that a realistic assessment of AFFF toxicity should over as many test organisms as possible. Phyto-toxicity tests with FFF sources are scarce in current the literature. We also argue hat the biotransformation of AFFF compounds should be taken in o account when evaluating long-term toxicity, as potentially toxic ntermediates may be formed. Our research evaluated the effects gy and Pharmacology 50 (2017) 119–127 of natural attenuation of AFFF on the development of plants. We aimed to determine whether AFFF toxicity increases or decrease over time in a simulated soil contamination scenario. Unlike most studies found in the literature, which monitor AFFF original formu- lation or persistent final biotransformation products, we designed intermediate toxicity evaluation points. Changes that affected the development of vegetable tissues provided an overall assessment of PFCs environmental safety. Moreover, we aimed to propose a novel toxicity classification towards environmentally safe release of pollutants using a colorimetric approach. A redox indicator (2,6 dichlorophenol-indophenol) usually associated with biodegrada- tion studies was repurposed by our research group to evaluate soil microbiota response to various concentrations of AFFF. 2. Material and methods 2.1. Soil samples Soil samples were acquired from the Biosciences Institute Experimental Garden at the Sao Paulo State University in Rio Claro, SP, Brazil (22◦43′24.2′′S 47◦08′00.3′′W). The area has a petroleum contamination background that is analogous to oil industry sites affected by hydrocarbon fires. The sampling area has been exclu- sively used for experiments with gasoline, diesel, kerosene and other petroleum hydrocarbons over the past 8 years. 2.2. Toxicity assessment of AFFF dilutions Various dilutions of AFFF samples were prepared from stock AFFF solution. In this first group of experiments no soil was added to the assays. We evaluated pure AFFF effects on plants germina- tion, wherein the concentration of perfluorinated compounds was 195 g L−1. Dilutions were then made to match 3% and 1% concentra- tions, yielding 97.5 and 19.5 g L−1 of PFCs. Both dilutions are also available from firefighting foam distributors. The solutions were then directly inserted into toxicity bioassays for the germination and development of seeds. 2.3. Toxicity throughout biodegradation A set of experiments on toxicity was performed to evaluate AFFF toxicity at different concentrations through natural attenuation. The soil matrix was designed to simulate widespread PFC contam- ination scenarios with AFFF. The biodegradation environment was set up through a simulated soil contamination within a plastic bag filled with 3 kg of soil and 0.1x, 1x and 10x m/v AFFF (Table 1). The container had small holes with approximately 1 mm diame- ter, spaced 1 cm each, to promote the exchange of microorganisms F60 Soil + AFFF 0.1x, 1x and 10 x m/v 60 days F120 Soil + AFFF 0.1x, 1x and 10 x m/v 120 days F180 Soil + AFFF 0.1x, 1x and 10 x m/v 180 days F240 Soil + AFFF 0.1x, 1x and 10 x m/v 240 days xicolo p ( b c A 2 f t c a o i a c 0 t t s a 2 t a m i m t c e s s t a s G t l p b a 5 p l u t k c 2 s e m h s c l m a % R.N. Montagnolli et al. / Environmental To erfluorinated substances fluorotelomer thioether amido sulfonate FtTAoS) known by the brand name Lodyne, and diethyleneglycol utyl ether (DGBE or butyl carbitol). Together, these substances onstitute the largest part of the concentrated formulation of Sintex FFF. The treatment groups were removed at 60, 120, 180 and 40 days. Once collected, the soils were stored at 4.0 ± 0.1 ◦C before urther toxicological analysis. The AFFF added to each plastic container matched the concen- rations from recently reported spills in soil (USEPA, 2009a,b). The oncentrations were adjusted to better represent an oil refinery fter the application of firefighting foam. The typical concentration f AFFF found in soil is up to 15.0 gL−1 soil (USEPA, 2009a,b), result- ng in 4.5 mL kg−1, considering the average soil density as 1.5 g cm3 nd AFFF as 1.12 g cm−3. The amount of AFFF in the simulated soil ontamination was conducted in 3 orders of magnitude, namely: .1x (or 0.45 mL kg−1), 1 x (4.5 mL kg−1) and 10x (45 mL kg−1). Each reatment was buried back at the Experimental Garden of the Insti- ute of Biosciences – Sao Paulo State University, Rio Claro – SP. The oil samples were dug up and had their phyto-toxicity analyzed fter 60, 120, 180 and 240 days. .3.1. Phyto-toxicity bioassays From a physiological point of view, germination occurs when he embryo ceases physiological hiatus and goes into metabolic ctivity (Nassif et al., 1998), provided that internal and environ- ental conditions are met for the tissues to develop. Any changes n these parameters by toxic substances may affect the develop- ent of plant tissue (Ayers and Westcot, 1999). Based on this, our oxicity assays were performed in soils according to previous proto- ols proposed by Morales (2004) and Montagnolli et al. (2015). The ffects of AFFF were evaluated using two test organisms: Lactuca ativa (lettuce) and Eruca sativa (arugula). The pesticide-free seeds were all purchased from the same upplier (TopSeed ® – blue line). Each bioassay was performed in riplicate per soil sample. Positive controls with pure MilliQ water nd negative controls with ZnSO4 0.10 M were performed to verify eed sensitivity and viability. Seed germination and seedling tissue elongation was measured. ermination tests were conducted in petri dishes containing a fil- er paper that covered its entire base. We added 4.0 mL of soil eachate from each assay at the filter papers. In other words, we oured into each filter paper the supernatant portion of each soil ioassay proposed in Table 1. The leachate source was kept under gitation at 210 rpm in shaker table for 24 h. Each vial contained .0 g of soil and 100.0 mL of deionized water. The seeds were then laced on top of the filter paper. The bioassays were performed with ettuce and arugula separately, as 30 seeds were added to individ- al petri dishes. Experiments were conducted in triplicate for each est substance. The petri dishes were covered with plastic film to eep moisture, and incubated at 20.0 ± 1.0 ◦C for 120 h under dark onditions. .3.2. Germination and growth analysis Our analysis of seedling development started by freezing eedlings after 120 h of incubation. This procedure ceased veg- table growth and decreased the rigidity of plant tissues to facilitate easurements by avoiding possible breaches during experimental andling. Root and hypocotyl measurements were taken for each eedling. Germination count was performed after 120 h. Seeds were onsidered germinated from a minimum 2 mm radicle protrusion imit. The data obtained in the tests were used to calculate the ger- ination percentage (% G) given by Labouriau and Agudo (1987) ccording to Eq. (1). G = arcsin[(SGa/SGt).100]0.5 (1) gy and Pharmacology 50 (2017) 119–127 121 where: SGa is the total number of germinated seeds and SGt is the total number of germinated control seeds; The statistical analysis verified the difference between treat- ments by using the Tukey test at 5% probability (Sokal and Rohlf, 1995). Germination values were normalized according to legacy references by Snedecor (1962) to allow the comparison across our all of our results. 2.3.3. Data plotting and interpretation A novel approach towards germination data analysis was performed using a three-dimensional response surface relating germination time and concentration of AFFF on an axis x, y and z. This data plotting technique evaluates the influence of two fac- tors (time and concentration) simultaneously in the germination of arugula and lettuce seeds. The 3D plots were made with MICROCAL ORIGIN 8.0 software. It is important to highlight that the experi- ments have not been based on factorial experimental design and, unlike factorial design, the color gradient in our study is filled by linear extrapolation of the experimental data directly through a multidirectional method proposed by Renka (1984) and Yuan (1998). Therefore, the response surface was created from the raw data converted into a three-dimensional Cartesian coordinate plane using Eq. (2) system. r = (x2 + y2) 1/2 (a) � = tan −1(y/x)(b) z = z(c) (2) Each variable in Eq. (2) – time (x), AFFF concentration (y) and germination (z) – went through the process of triangulation and definition of response surface contours. The first step was to create a three-dimensional map and triangulation of data based on the distance between each. All points of data were connected to create triangles in a plane x and y. Each of these points had a value of z. The angles were designed to be as close as possible to equilateral geom- etry. Moreover, the triangles could not intersect. The layers were set perpendicularly through all planes of the z axis looking for the sides of the triangles of the three-dimensional map. Where the side of each triangle intersects the plane, a point was defined. Each of these points on separate planes were computed by linear interpo- lation to generate the polygons with a contour line of the response surface (xc, yc, zc). In other words, the contour line is drawn at the edge of the triangles that form the three-dimensional data map. The process of searching for intersection points of three- dimensional triangulation map was repeated in several planes for creating various contour lines. The outline was not smoothed with spline type curves since it would result in less data precision on intermediate values. The resolution of the response surface was given by the number of contour levels made on the plan of triangu- lation of data. Each of the planes was assigned to a different shade of gray. In this study, we described the germination data with 50 distinct levels, each spanning 2% units. 2.4. Redox indicators and microbial inhibition The colorimetric approach towards toxicity is fundamentally different from the described phyto-toxicity assays, as they are a short-term detection protocol. Colorimetric methods are often referred to as rapid and low cost procedures in detecting microbial metabolism occurrence. Yet, both methods can provide equiva- lent data in toxicity processes. Generally, colorimetric methods are based in the different coloration of oxidized and reduced forms. Due to this fact, such color change can be monitored with a spec- trophotometer, evaluating changes in redox indicators absorbance values in a set time or across the discoloration process. 122 R.N. Montagnolli et al. / Environmental Toxicolo Table 2 Colorimetric bioassays. Assay ID DCPIP BH Media Soil Microbiota AFFF C1 400 �L 7,5 mL – – C2 400 �L 7,5 mL 100 �L – C3 400 �L 7,5 mL – 1000 �L c ( o a B s d a i w 1 I b c m s o a A t w t s c a C t ( o i a D u t w t b o 2 b m o t e d f f r ( c B1 400 �L 7,5 mL 100 �L 1000 �L B2 400 �L 7,5 mL 100 �L 100 �L B3 400 �L 7,5 mL 100 �L 10 �L The 2,6 dichlorophenol-indophenol (DCPIP) is an enzyme- atalyzed redox electron acceptor that is blue in its oxidized form DCPIPox) and colorless in its reduced form (DCPIPred). DCPIPox loss f color is monitored at a wavelength of 600 nm since its peak bsorbance occurs at 600 nm, as reported by Yoshida et al. (2001). y incorporating an electron acceptor such as DCPIPox, it was pos- ible to ascertain the ability of a microorganism to grow on AFFF ilutions by simply observing the color change. The redox reactions pplied to AFFF allowed a brief, yet responsive analysis of toxicity n various concentrations. To study AFFF toxicity to the soil microbiota, colorimetric assays ere setup using 5 g of C0 soil (Table 1) aliquot suspended in 00 mL of Bushnell-Haas (BH) media (Montagnolli et al., 2015). nitial experiments optimized the DCPIPox color change output in ioassays by adjusting AFFF initial concentration, soil microcosm ell density and length of incubation with DCPIPox. After deter- ining ideal substrate concentration, the cells were counted using tandard Plate Count Agar and transferred from storage culture. The microbial cultures were reactivated and inoculated to 50 mL f BH medium (Montagnolli et al., 2015) at 35 ◦C for biomass growth long with AFFF. No agitation or darkness conditions were applied. fter 48 h in BH medium, the C0 soil culture was inoculated to ubes along with DCPIPox indicator and AFFF. Colorimetric assays ere conducted in test tubes with lids, to retain CO2 saturation. The ubes were transparent and suitable for spectrophotometry mea- urements. The contents of each colorimetric assay followed the oncentrations in Table 2. All assays were performed in triplicates. After assembled, each tube underwent vortex agitation for 10 s nd were then stored at 35 ◦C and 180 rpm. Control assays C1 and 2 (media control and bacterial control) evaluated interactions of he assay components and DCPIPox indicator. A third control set C3 – AFFF control) verified the sole AFFF influence in DCPIP discol- ration. The full assays included DCPIPox indicator, BH medium, soil noculum and the different AFFF concentrations. The absorbance of ll assays was measured by a Hach ® DR 2500 spectrophotometer. ata was collected three times per day, up to 35 h. Absorbance val- es were compared between different assays. Measurements were aken with Odyssey Hach DR-2500 spectrophotometer at 600 nm avelength. The tubes were left without agitation for 10 min before he start of each analysis. Tubes were not abruptly moved before eing placed into the spectrophotometer to avoid any turbulence f the liquid, which could alter the absorbance results. .4.1. Toxicity data analysis The colorimetric data from DCPIPox discoloration has usually een considered analogous biodegradation data in microbial com- unities. Hanson et al. (1993), for example, initiated the trend f DCPIP applications in the monitoring of crude oil degrada- ion by bacteria. However, we argue that experiments by Hanson t al. (1993) did not determine the biodegradability of compounds irectly, but the amount of inhibition of microbial activity in dif- erent compounds. Many other legacy experiments using DCPIP ollowed this biodegradation correlation since then, including the ecent studies by Varjani and Upasani (2016) and Souza et al. 2016). However, redox assays are in fact a direct measure of toxi- ity, instead of biodegradation. The electron transference in redox gy and Pharmacology 50 (2017) 119–127 reactions drives the DCPIP molecular conformation changes and consequently color emission. Such redox reactions are not solely associated to a unique substrate consumption, but to the whole cell metabolism. Thus, what DCPIP actually measure is the micro- bial metabolic response to a given substrate and/or environment in which it is exposed instead of biodegradation. The legacy DCPIP methodology is more properly described as a toxicity evaluation tool. Therefore, we exploited the parameters in discoloration kinetics to propose a new score system and assess the effect of AFFF to soil microbiota. Our research proposes the establishment and standardization of the DCPIP as a toxicity indica- tor. Models applied to the colorimetric dataset can lead to a better understanding of AFFF toxicity, thereby quickly sorting the level of toxicity of each concentration. The colorimetric dataset was presented as DCPIPox discoloration values as the oxidized form decreased though biodegradation. The depletion profile yielded by this process fitted to a sigmoidal non-linear model. The parameters were adjusted to describe our colorimetric data output (Eq. (3)). The model provided the toxicity level for the microbiota within each assay with the S parameter. D = Dmax + [(Dmin-Dmax)/(1 + 10(logD50−t)S)] (3) where: D = DCPIPox concentration, Dmax = initial DCPIPox con- centration, Dmin = minimum detectable DCPIPox concentration, D50 = relative microbial response, r = toxicity score for a specific substrate and environment, t = time The parameters of the sigmoidal equations were useful to obtain physiological activity data for each AFFF dilution. The D50 and S kinetic parameters corresponded to units of time, where D50 defines the time when amplitude has reached its half-way point, i.e., when half of the DCPIPred has been oxidized. The toxicity score S was defined as the time and concentration relationship for the parameter to go from maximum to minimum absorbance, which is translated as the toxicity intensity when soil microbiota is exposed to AFFF. 3. Results and discussions 3.1. Toxicity of AFFF dilutions No seeds germinated in bioassays containing pure retail AFFF. The original 6% concentration and the 3% dilution did not result in germination in any of the triplicates. In 1% dilutions, there was an average germination of 3.4% germination of lettuce seeds and 1.7% of arugula seeds. The commercially available AFFF is labeled as a non-toxic product, however the phyto-toxicity results found in this study contradicts such statement. The original formulation of AFFF is quite concentrated and dis- rupted seed germination environment. The acidity and the organic components of AFFF may have inhibited the germination of the test organisms. The fluorinated compounds are considered inert due to the number of C-F bonds in their molecule, however they must have been another inhibiting factor. Fortunately, environmental samples contaminated with AFFF rarely contain concentrations as high as the ones tested. Contaminated environmental samples are usually close to the maximum 15.0 gL−1 concentration allowed by USEPA (2009a). 3.2. Toxicity of AFFF contaminated soil After 120 h of incubation, we observed that there was no germi- nation in ZnSO4 control assays. Positive controls presented over 90% germination rates, meaning viable and healthy seeds. The germina- tion percentages though time under different AFFF concentrations is shown in Fig. 1. The darker areas correspond to the lowest ger- mination rates. R.N. Montagnolli et al. / Environmental Toxicology and Pharmacology 50 (2017) 119–127 123 90 80 70 60 50 40 30 20 40 10 70 30 0 60 120 180 240 Germination % 0 10 20 30 40 50 60 70 80 90 100 0 10 1 0.1 A B C 80 70 6050 40 30 20 40 10 0 60 120 180 240 0 10 1 0.1 D E (a) (b)90 Time / days A FF F co nc en tra tio n / d ilu tio n fa ct or A FF F co nc en tra tio n / d ilu tio n fa ct or Time / days F x, 1x c ated b c C 9 o t p B c s i F i ( a t g ig. 1. Lettuce (a) and arugula (b) germination at three AFFF concentrations (0.1 oncentration after 120 days. The regions of lower and higher germination are indic oncentration. Region E shows the low sensitivity of arugula to AFFF. The germination of seeds that were not exposed to AFFF in C0, 60, C120, C180 and C240 reached the highest values between 0 and 100%. This can be observed throughout the lower region f both response surfaces in Fig. 1. The most toxic substrate, on he other hand, is seen on the upper part of the response surface lot, corresponding to the highest concentrations of AFFF (10x). oth arugula and lettuce had their lowest germination rates at 10x oncentration even after 240 days. Lettuce was noticeably more ensitive to AFFF exposure than arugula. The lettuce sensitivity s especially noticeable in the region B (upper left corner) from ig. 1a, where germination rates were inferior to 30% during the nitial measurement time and after 60 days of natural attenuation F0–10x and F60–10x). Arugula seeds, on the other hand, showed n increased germination (and decreased toxicity) at 10x concen- rations in Fig. 1b. Even higher concentrations of AFFF allowed ermination in the 30–50% range at late attenuation time according and 10x) during 240 days. Region A is the highest germination found at the 1x y the letter B and C, respectively. Region D is the lowest germination found at 0.1x to the region E (Fig. 1b), whereas lettuce seed were still germi- nating between the 0–20% range. Therefore, arugula seeds are less sensitive indicator of AFFF toxicity. It is possible that biodegradation of non-fluorinated compo- nents of the formulation of AFFF may have contributed to the lower toxicity though time, resulting in increased germination after 120 days. The response surface indicates with the expansion of light gray tones on the right side of the chart a clear tendency to increase in germination. Thus, the simulated environment became more favorable to the growth of the seeds after 240 days. The Fig. 1 plot confirms visually that natural attenuation reduced the toxicity of AFFF especially at 1x and 0.1x concentrations. According to Fig. 1, the germination was not exclusively a func- tion of AFFF concentration (y) or time (x) independently. The entire lettuce and arugula response surface approached a nonlinear corre- lation with time and germination, wherein the AFFF concentration 1 xicolo a i t t T s i c s o a p 1 w p t i h F – 24 R.N. Montagnolli et al. / Environmental To ffected sprouting. In other words, the response surface has shown nterference between AFFF concentration (y) and germination (z) hrough time (x). Even though we observed that AFFF had a progressively low oxicity, there are some specificities that need to be highlighted. he region A in the lettuce response surface (Fig. 1a), for example, hows an interesting change in the expected pattern. An increase n germination occurred, specifically, around 120 days at 1x con- entration (assay F120 at 1x concentration). Although these soil amples have not been chemically analyzed, it is likely that variants f secondary metabolites due to the biodegradation of AFFF have ppeared throughout the process at that specific time. Those com- ounds that appeared only at the intermediate times (120 days) at x concentration are therefore less toxic and favored germination, hereas further AFFF degradation steps retained growth inhibition roperties. The natural attenuation is important to decrease toxicity of AFFF o such seeds. However, the seasonality factor must also be taken nto account. We examined for any long-term correlation that may ave disturbed germination. Despite the slightly higher germina- 0x 0.1x 1x 10x 0 5 10 15 20 25 30 35 40 45 50 0x 0.1x 1x 10x 0 5 10 15 20 25 30 35 40 45 50 55 60 Hipoco Lenght / cm T0 (a) (c) A FF F co nc en tra tio n / d ilu tio n fa ct or A FF F co nc en tra tio n / d ilu tio n fa ct or ig. 2. Growth of arugula (a, b) and lettuce (c, d) seedlings in at three AFFF concentrations b, d). gy and Pharmacology 50 (2017) 119–127 tion observed in region C of Fig. 1a, all datasets underwent Tukey test analysis to clarify whether the variation in germination rates between assays containing AFFF were significantly different from a seasonal variation. There was a significant difference between all assays compared to controls. Thus, the seasonal variation of soil conditions was not the factor responsible for increase in germina- tion, but the difference of AFFF concentration between treatments instead. Vegetable tissue growth also indicated the toxicity of the com- pounds before and after natural attenuation (Fig. 2). As expected, soil controls without any PFC promoted higher veg- etable tissue growth. Interestingly, a slightly higher average growth is found in lettuce at 1x concentrations rather than 0.1x. Once more we observed the intermediate concentration (1x) interference with toxicity pattern in lettuce assays, where a decreased growth inhi- bition takes place at 1x concentration (Fig. 2d). Even though the phyto-toxicity of AFFF to lettuce and arugula was found to be high (increasing as a function of the concentration and decreas- ing as a function of exposure time to the environment), such very specific concentrations where non-inhibiting intermediates are 0x 0.1x 1x 10x 0 5 10 15 20 25 30 35 40 45 50 55 60 0x 0.1x 1x 10x 0 5 10 15 20 25 30 35 40 45 50 55 60 tyl Root Lenght / cm T240 (b) (d) 60 (0.1x, 1x and 10x) in toxicity bioassays at the initial time (T0 – a, c) and final (T240 xicolo f i P a t g d A a r a a p c l e r s s 2 g n m fl e c S R.N. Montagnolli et al. / Environmental To ormed may mislead other studies with non-time-based monitor- ng techniques. Therefore, it is imperative to consider the toxicity of FCs in soil as a dynamic process where biodegradation and natural ttenuation are key processes. Arugula (Fig. 2a and b), however, followed a straightforward oxicity pattern. Higher AFFF concentrations yielded less arugula rowth whereas lower AFFF concentrations allowed more cells to evelop. The arugula seedlings were also much less sensitive to FFF, as they were able to outgrow to lettuce seeds by 17.90% on verage by the end of 240 days. Therefore, arugula seeds are not ecommended to evaluate long-term AFFF toxicity during natural ttenuation processes. Except for the low sensitivity after natural ttenuation in arugula, both bioassays followed a similar growth attern. The increase in germination and tissue growth over time aused AFFF toxicity was less pronounced in arugula seeds, hence ettuce is a better toxicity indicator for AFFF phyto-toxicity. Lettuce and arugula tissue elongation data confirmed the influ- nce of biodegradation time in reducing the toxicity of AFFF. The esponse surface in Fig. 1 has indicated higher germination repre- enting higher germination (from 70% and upwards) on the right ide of the chart. The seedling growth length reinforces that after 40 days a vegetable tissue growth is more likely to intensify. In eneral, the medium becomes more favorable to seeds growth after atural attenuation, but more sensitive organisms may not grow as uch as the control assays. There are many references that support low toxicity of non- uorinated components of AFFF, such as butyl carbitol (Staples t al., 1998; Johansson, 1998). The toxicity of many perfluorinated ompounds to plants, on the other hand, is not very well known. tudies report that these persistent PFCs are widely distributed in S= -0.0359 S= -0.0078 D C P IP (g .L ) ox -1 D C P IP (g .L ) o x -1 Time (hours) (a) (c) Color - less Color - less Fig. 3. DCPIPox concentration in AFF control – C2 (a), 1000 �L AF gy and Pharmacology 50 (2017) 119–127 125 animals. It is known that exposure to them results in hepatotox- icity, development problems, immunotoxicity, adverse hormonal effects and potential carcinogen in animals (Clarke et al., 2010; Hölzer et al., 2008). Based on such previous toxicity studies in other organisms, it is safe to associate the toxicity observed in our bioas- says to the presence of perfluorinated compounds. Still, there are no other reports of AFFF phyto-toxicity assays with arugula or let- tuce that allow direct comparison to the results obtained in this study. The literature lacks firefighting foams related toxicity bioas- says on plants. The hereby presented results comprehend an initial effort towards building a better comprehension of the AFFF effects in plants. Moreover, few studies on PFC toxicity analyze individ- ual components and metabolites, since both the formulation as the secondary metabolites of biotransformation of PFCs are mostly unidentified. Thus, more in depth studies are encouraged to define the influence of AFFF in the development of many other test organ- isms. The data obtained in toxicity bioassays, particularly shown in Fig. 1 indicate that, despite the initial toxicity of AFFF released into the environment either as arugula and lettuce have grown in con- taminated soil. There is a major concern about the bioaccumulation of these compounds in the food chain, including in plants (Van Asselt et al., 2011). Although the arugula and lettuce growth may intuitively appear as an indication that the soil is safe for plants development due to decreased toxicity after 240 days, that does not mean that low toxicity is associated with the biodegradation of perfluorinated compounds. The butyl carbitol portion of AFFF formulation is the most biodegradable portion, whereas molecules containing multiple C-F bonds are considered biologically inert for these organisms. According to Filipovic et al. (2015), PFCs can enter S= -0.0104 S= -0.0855 Time (hours) (b) (d) Color- less Color- less FF – B1 (b), 100 �L – B2 (c) and 10 �L AFFF – B3 (d) assays. 126 R.N. Montagnolli et al. / Environmental Toxicolo Table 3 Parameters set to discoloration model of DCPIPox. Assay ID Score (g L−1 h−1) DminT (h) R2 C0 −0.0068 – 0.9932 C1 −0.0070 – 0.9928 C2 −0.0078 – 0.9956 t f c e e a f P P c h b u c t b b c r r t t d 3 f q p l w c c a p t m A − p a I l ( a t 3 c a c B1 −0.0104 – 0.9942 B2 −0.0359 28.03 0.9939 B3 −0.0855 16.89 0.9968 he tissues of plants during growth and accumulate in food chain or up to 45 years. This is potentially harmful, as fluorinated organic ompounds are known for a decade to have bioaccumulative prop- rties as well as toxicity to organisms, including mammals (Schultz t al., 2004). We propose that AFFF germination and growth after 240 days, s observed in this study, may result in the PFCs entrance into the ood-chain. The final germination rates (Fig. 1) can be related to FC bioaccumulative potential. The main group of PFCs include FOS, and has been frequently detected in food products at high oncentrations (FSA, 2009). The route of human exposure to PFAS, owever, has not yet been characterized (Halldorsson et al., 2008), ut may as well come from the consumption of agricultural prod- cts. Although non-food sources, as even dust particles, may ontain PFCs (and taken into consideration in Figure 85), the intake hrough food is the cause of 98% of human exposure the fluorocar- on (Fromme et al., 2009; Wilhelm et al., 2009). A recent study y Lorber et al. (2015) found up to 19 ng mL1 of perfluorinated ompounds in human blood plasma. Therefore, considering all potentially harmful substances eported by other authors, combined with our phyto-toxicity esults, we encourage further studies on the bioaccumulation and oxicity of AFFF formulations to plants. For this reason, we also alert o the lack of strict legislation and control over the use, sale and isposal of AFFFs near cultivation areas. .2.1. Soil microbiota inhibition The time required to complete decolorization varied among dif- erent concentrations of AFFF. Our colorimetric assays allowed the uantification of AFFF toxicity to soil microbiota (Fig. 3). The basic rinciple behind this technique is that the faster the curve reaches ower values of DCPIPox, the faster microbial metabolism occurs. The B2 and B3 assays were the only ones whose discoloration as complete. High AFFF concentrations in a soil sample may ompromise microbial activity, thus decreasing the indicator color hanges, especially when low amounts of AFFF is present. Table 3 shows the expected parameters obtained from the djustment model to the data of both methods. The most important arameter to be considered is S since this parameter can be arbi- rarily employed towards toxicity rating in any given environment. The equation parameters were adapted to the model when easuring absorbance values from each assay with differential FFF concentration. The S parameter varies approximately between 0.001 (more toxic) and −0.099 (less toxic) and allowed us to pro- ose a quantifiable toxicity score from DCPIP color. The S parameter nalysis brings an interesting information on the toxicity of AFFF. t is expected to notice that the more diluted the sample is, the ess toxic compound to soil microbiota. Thus, the highest dilution Fig. 1d) presented the lowest level of toxicity, followed by B2, B1, nd C2 assay. The rates varied between the different substances. Even without he occurrence of full discoloration until the final measurement at 5 h, we could precisely estimate the toxicity for each tested con- entration. The parameters extrapolated by the model in adjusting llowed the prediction of toxicity even without complete dis- oloration DCPIP. Still, we observed from pilot assays that the gy and Pharmacology 50 (2017) 119–127 predictability of dataset fitting to the equation is only minimally acceptable with a regular collection of data until at least 30 h into the process. With less than 30 h of periodically collected absorbance date, it is impossible to estimate relative toxicity with precision and consistency from the data set. Overall, the addition of higher concentrations of AFFF caused less discoloration at all assays. PFCs would be responsible for inhi- bition of microbial activity in the assays as much as the observed phyto-toxicity in arugula and lettuce. Despite the positively obvi- ous result, the curve fitting analysis should aid future regulations to define limits of AFFF and may as well be applied to any other pollutants released in soil. 4. Conclusions The phyto-toxicity of AFFF to lettuce and arugula was found to be high, increasing as a function of the concentration and decreas- ing as a function of exposure time to the environment. Also, the low phyto-toxicity after 240 days and the potential for bioaccumu- lation in plants increases the persistence issues related to these compounds. The combination of two different methods of monitor- ing toxicity widened the current knowledge on the environmental release profile of AFFF. The assays allowed the monitoring of AFFF during long-term exposition to plants as well as their immediate effects on the same soil microbiota. While evaluating the com- pounds through natural attenuation, some possible biodegradation metabolites may have appeared and interfered with environmental safety at very specific concentrations. Even though short-term experiments toxicity assays, such as the colorimetric assays, are usually not indicated towards mon- itoring of toxicity in degradation by-products, they allowed the precise modelling and quantification of AFFF effects on the soil microbiota. Phyto-toxicity assays further expanded the AFFF envi- ronmental safety assessment since they are operated on a much longer period, observing the appearance and disappearance of cer- tain metabolites with an impact on toxicity. We observed that both techniques describe the AFFF environmental release scenario not only with a long-term decrease in toxicity, but also with an actual retardation of microbial metabolism in soil. Future studies are nec- essary to determine via further chemical analysis in plant tissues which bioconcentrated compounds result from AFFF metabolism in the environment. As plants were observed capable of growth in AFFF contaminated environment after 240 days, bioaccumulation of PFCs in the both test organisms (arugula and lettuce) is a likely assumption. Conflict of interest Nothing to disclose. Acknowledgments We gratefully acknowledge the thoughtful review and impor- tant insights provided by Sarah Mae Wachlin and Brent Perumal, providing comments on the manuscript in its entirety and work- ing with the authors as we made revisions to the text. We alone, of course, take responsibility for the final text and any errors that appear therein. Our research group acknowledges CAPES (Coordenaç ão de Aperfeiç oamento de Pessoal de Nível Superior), CNPq (Con- selho Nacional de Desenvolvimento Científico e Tecnológico), FUNDUNESP (Fundaç ão para o Desenvolvimento da UNESP), PRH- ANP/MCT (Programa de Formaç ão de Recursos Humanos em Geologia do Petróleo e Ciências Ambientais Aplicadas ao Setor de Petróleo e Gás) and UNESP (Universidade Estadual Paulista “Julio xicolo d i c r R A A B C C D E E F F F F F G G H H H H H H H J K K K Yuan, F., 1998. Automatic drawing of equal quantity curve. Comput. Aided Chem. 1, 1–7. R.N. Montagnolli et al. / Environmental To e Mesquita Filho”) The financial support. funding sources had no nvolvement then this should be stated in the study design; in the ollection, analysis and interpretation of data; in the writing of the eport; and in the decision to submit the article for publication. eferences lm R.R., Stern R.M., 1992. Aqueous Film-Forming Foamable Solution Useful as Fire Extinguishing Concentrate. U.S. Patent 5085786. yers, R.S., Westcot, D.W., 1999. A qualidade da água na agricultura, Thesis. UFPB, Campina Grande. uck, R.C., Franklin, J., Berger, U., Conder, J.M., Cousins, I.T., De Voogt, P., Jensen, A.A., Kannan, K., Mabury, S.A., Van Leeuwen, S.P., 2011. Perfluoroalkyl and polyfluoroalkyl substances in the environment: terminology classification, and origins. Integr. Environ. Assess. Manag. 7, 513–541. larke, B.O., Smith, S.R., 2011. Review of emerging organic contaminants in biosolids and assessment of international research priorities for the agricultural use of biosolids. Environ. Int. 37, 226–247. larke, D.B., Bailey, V.A., Routledge, A., Lloyd, A.S., Hird, S., Mortimer, D.N., 2010. Dietary intake estimate for perfluorooctanesulphonic acid (PFOS) and other perfluorocompounds (PFCs) in UK retail foods following determination using standard addition LC–MS/MS. Food Addit. Contam. Part A Chem. 27, 530–545. inglasan-Panlilio, M.J.A., Mabury, S.A., 2006. Significant residual fluorinated alcohols present in various fluorinated materials. Environ. Sci. Technol. 40, 1447–1453. ECHA (European Chemical Agency), 2013. Candidate List of Substances of Very High Concern for Authorization, Available: http://echa.europa.eu/web/guest/ candidate-list-table. Accessed: June 2016. llis, D.A., Martin, J.W., Desilva, A.O., Mabury, S.A., Hurley, M.D., Sulbaekandersen, M.P., Wallington, T.J., 2004. Degradation of fluorotelomer alcohols: a likely atmospheric source of perfluorinated carboxylic acids. Environ. Sci. Technol. 38, 3316–3321. SA (Food Standards Agency), 2009. Survey of Fluorinated Chemicals in Food. FSA, Available: www.food.gov.uk/multimedia/pdfs/fsis0509.pdf. Acessed: November 7. alk R.A., 1978. Aqueous Wetting and Film Forming Compositions. U.S. Patent 4090967. ilipovic, M., Woldegiorgis, A., Norström, K., Bibi, M., Lindberg, M., Österås, A., 2015. Historical usage of aqueous film forming foam: a case study of the widespread distribution of perfluoroalkyl acids from a military airport to groundwater, lakes, soils and fish. Chemosphere 129, 39–45. römel, T., Knepper, T.P., 2010. Biodegradation of fluorinated alkyl substances. Rev. Environ. Contam. Toxicol. 208, 161–177. romme, H., Tittlemier, S.A., Völkel, W., Wilhelm, M., Twardella, D., 2009. Perfluorinated compounds –exposure assessment for the general population in western countries. Int. J. Hyg. Environ. Health 212, 239–270. iesy, J.P., Kannan, K., 2001. Global distribution of perfluorooctane sulfonate in wildlife. Environ. Sci. Technol. 35, 1339–1342. iesy, J.P., Kannan, K., 2002. Perfluorochemical surfactants in the environment. Environ. Sci. Technol. 36, 146–152. ölzer, J., Midasch, O., Rauchfuss, K., Kraft, M., Reupert, R., Angerer, J., Kleeschulte, P., Marschall, N., Wilhelm, M., 2008. Biomonitoring of perfluorinated compounds in children and adults exposed to perfluorooctanoate contaminated drinking water. Environ. Health Persp. 116, 651–657. agner, M., Penttinen, O.P., Pasanen, T., Tiilikkala, K., Setälä, H., 2010. Acute toxicity of birch tar oil on aquatic organisms. Agric. Food Sci. 19, 24–32. alldorsson, T.I., Fei, C., Olsen, J., Lipworth, L., Mclaughlin, J.K., Olsen, S.F., 2008. Dietary predictors of perfluorinated chemicals: a study from the Danish National Birth Cohort. Environ. Sci. Technol. 42, 8971–8977. anson, K.G., Desai, J.D., Desai, A.J., 1993. A rapid and simple screening technique for potential crude oil degrading microorganisms. Biotechnol. Tech. 7, 745–748. iggins, C.P., Field, J.A., Criddle, C.S., Luthy, R.G., 2005. Quantitative determination of perfluorochemicals in sediments and domestic sludge. Environ. Sci. Technol. 39, 3946–3956. outz, E.F., Sedlak, D.L., 2012. Oxidative conversion as a means of detecting precursors to perfluoroalkyl acids in urban runoff. Environ. Sci. Technol. 46, 9342–9349. udlicky, M., Pavlath, A.E., 1995. Chemistry of Organic Fluorine Compounds II: A Critical Review. American Chemical Society, Washington. ohansson, G., 1998. Aspects of biological monitoring of exposure to glycol ethers. Toxicol. Lett. 43, 5–21. ey, B.D., Howell, R.D., Criddle, C.S., 1997. Fluorinated organics in the biosphere. Environ. Sci. Technol. 31, 2445–2454. im, S.K., Kannan, K., 2007. Perfluorinated acids in air, rain, snow, surface runoff, and lakes: relative importance of pathways to contamination of urban lakes. Environ. Sci. Technol. 41, 8328–8334. issa, E., 1994. Fluorinated Surfactants: Synthesis, Properties and Applications, first ed. Marcel Dekker, New York. gy and Pharmacology 50 (2017) 119–127 127 Labouriau, L.G., Agudo, M., 1987. On the physiology of seed germination in Salvia hispanica L.I. Temperature effects. An. Acad. Bras. Ciênc. 59, 37–56. Lladó, S., Solanas, A.M., De Lapuente, J., Borràs, M., Viñas, M., 2012. A diversified approach to evaluate biostimulation and bioaugmentation strategies for heavy-oil-contaminated soil. Sci. Total Environ. 435, 262–269. Lorber, M., Eaglesham, G.E., Hobson, P., Toms, L.M.L., J.F. Mueller, J.F., Thompson, J.S., 2015. The effect of ongoing blood loss on human serum concentrations of perfluorinated acids. Chemosphere 118, 170–177. Montagnolli, R.N., Lopes, P.R.M., Bidoia, E.D., 2015. Screening the toxicity and biodegradability of petroleum hydrocarbons by a rapid colorimetric method. Arch. Environ. Contam. Toxicol. 68, 342–353. Moody, C., Field, J., 2000. Perfluorinated surfactants and the environmental implications of their use in fire-fighting foams. Environ. Sci. Technol. 34, 3864–3870. Moody, C.A., Hebert, G.N., Strauss, S.H., Field, J.A., 2003. Occurrence and persistence of perfluorooctanesulfonate and other perfluorinated surfactants in groundwater at a fire-training area at Wurtsmith Air Force Base Michigan, USA. J. Environ. Monit. 5, 341–345. Morales, C.G., 2004. Ensayos Toxicológicos y Métodos de Evaluación de Calidad de Agua: estandarización, intercalibración, resultados y aplicaciones, first edition. IMTA, Ciudad de México. Murakami, M., Imamura, E., Shinohara, H., Kiri, K., Muramatsu, Y., Harada, A., Takada, H., 2008. Occurrence and sources of perfluorinated surfactants in rivers in Japan. Environ. Sci. Technol. 42, 6566–6572. Nassif, S.M.L., Vieira, I.G., Fernandes, G.D., 1998. Informativo Sementes do Instituto de Pesquisas e Estudos Florestais, Thesis. Escola Superior de Agricultura Luís de Queiroz (ESALQ). Universidade de São Paulo (USP) Piracicaba. Rayne, S., Forest, K., 2009. Perfluoroalkyl sulfonic and carboxylic acids: a critical review of physicochemical properties, levels and patterns in waters and wastewaters, and treatment methods. J. Environ. Sci. Health A 44, 1145–1199. Renka, R.J., 1984. Interpolation of Data on the Surface of a Sphere. ACM Trans. Math. Softw. 10, 417–436. Sardqvist, S., 2002. Water and Other Extinguishing Agents, fourth ed. Swedish Rescue Services Agency Karlstad. Schultz, M., Barofsky, D., Field, J., 2004. Fluorinated alkyl surfactants. Environ. Eng. Sci. 20, 487–501. Sheppard, P., Adetutu, E., Makadia, T., Ball, A., 2011. Microbial community and ecotoxicity analysis of bioremediated: weathered hydrocarbon-contaminated soil. Soil Res. 49, 261–269. Sinclair, E., Kannan, K., 2006. Mass Loading and fate of perfluoroalkyl surfactants in wastewater treatment plants. Environ. Sci. Technol. 40, 1408–1414. Snedecor, G.W., 1962. Statistical Methods. Iowa State University Press, Iowa. Sokal, R.R., Rohlf, F.J., 1995. Biometry: the Principles and Practice of Statistics in Biological Research, third ed. W. H Freeman and Co., New York. Souza, M.M., Colla, T.S., Bücker, F., Ferrão, M.F., Huang, C.T., Andreazza, R., Camargo, F.A.O., Bento, F.M., 2016. Biodegradation potential of Serratiamarcescens for diesel/biodiesel blends. Int. Biodeter. Biodeg. 110, 141–146. Staples, C.A., Boatman, R.J., Cano, M.L., 1998. Ethylene glycol ethers: an environmental risk assessment. Chemosphere 36, 1585–1613. USEPA (U.S. Environmental Protection Agency), 2009a. Provisional Health Advisories for Perfluorooctanoic Acid (PFOA) and Perfluorooctane Sulfonate (PFOS). U.S. Environmental Protection Agency, Washington. USEPA (U.S. Environmental Protection Agency), 2009b. Soil Screening Levels for Perfluorooctanoic Acid (PFOA) and Perfluorooctyl Sulfonate (PFOS). U.S. Environmental Protection Agency, Washington. Van Asselt, E.D., Rietra, R.P.J.J., Römkens, P.F.A.M., Van Der Fels-Klerx, H.J., 2011. Perfluorooctane sulphonate (PFOS) throughout the food production chain. Food Chem. 128, 1–6. Varjani, S.J., Upasani, V.N., 2016. Biodegradation of petroleum hydrocarbons by oleophilic strain of Pseudomonas aeruginosa NCIM 5514. Bioresour. Technol. 222, 195–201. Vestergren, R., Cousins, I.T., 2009. Tracking the pathways of human exposure to perfluorocarboxylates. Environ. Sci. Technol. 43, 5565–5575. Wilhelm, M., Holzer, J., Dobler, L., Rauchfuss, K., Midasch, O., Kraft, M., 2009. Preliminary observations on perfluorinated compounds in plasma samples (1977–2004) of young German adults from an area with perfluorooctanoate-contaminated drinking water. Int. J. Hyg. Envir. Heal. 212, 142–145. Yoshida, N., Hoashi, J., Morita, T., McNiven, S.J., Nakamura, H., Karube, I., 2001. Improvement of a mediator-type biochemical oxygen demand sensor for on-site measurement. J. Biotechnol. 88, 269–275. Yu, J., Hu, J., Tanaka, S., Fujii, S., 2009. Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) in sewage treatment plants. Water Res. 43, 2399–2408. Zushi, Y., Masunaga, S., 2009. First-flush loads of perfluorinated compounds in stormwater runoff from Hayabuchi River basin, Japan served by separated sewerage system. Chemosphere 76, 833–840. http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0010 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0015 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0020 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0025 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0030 http://echa.europa.eu/web/guest/candidate-list-table http://echa.europa.eu/web/guest/candidate-list-table http://echa.europa.eu/web/guest/candidate-list-table http://echa.europa.eu/web/guest/candidate-list-table http://echa.europa.eu/web/guest/candidate-list-table http://echa.europa.eu/web/guest/candidate-list-table http://echa.europa.eu/web/guest/candidate-list-table http://echa.europa.eu/web/guest/candidate-list-table http://echa.europa.eu/web/guest/candidate-list-table http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0040 http://www.food.gov.uk/multimedia/pdfs/fsis0509.pdf http://www.food.gov.uk/multimedia/pdfs/fsis0509.pdf http://www.food.gov.uk/multimedia/pdfs/fsis0509.pdf http://www.food.gov.uk/multimedia/pdfs/fsis0509.pdf http://www.food.gov.uk/multimedia/pdfs/fsis0509.pdf http://www.food.gov.uk/multimedia/pdfs/fsis0509.pdf http://www.food.gov.uk/multimedia/pdfs/fsis0509.pdf http://www.food.gov.uk/multimedia/pdfs/fsis0509.pdf http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0055 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0060 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0065 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0070 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0075 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0080 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0085 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0090 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0095 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0100 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0105 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0110 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0115 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0120 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0125 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0130 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0135 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0140 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0145 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0150 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0155 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0160 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0165 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0170 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0175 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0180 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0185 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0190 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0195 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0200 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0205 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0210 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0210 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0210 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0210 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0210 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0210 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0210 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0215 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.com/S1382-6689(17)30025-X/sbref0220 http://refhub.elsevier.