Synthesis of trans-dihydronaphthalene-diols and evaluation of their use as standards for PAH metabolite analysis in fish bile by GC-MS

Phenols and trans-1,2-dihydro-1,2-diols are metabolites commonly formed in vivo in fish upon exposure to polycyclic aromatic hydrocarbons (PAHs). These metabolites are excreted via the bile and gas chromatography-mass spectrometry (GC-MS) analysis of bile is becoming more frequently used for evaluating PAH exposure levels in fish. Current protocols focus on the detection and quantification of phenols formed during in vivo oxidation of PAHs, leaving out analyses and quantification of other oxidation products such as trans-1,2-dihydro-1,2-diols, potentially underestimating exposure levels. Herein, four trans-1,2-dihydro-1,2-diols, namely trans-1,2-dihydronaphthalene-1,2-diol, trans-6-methyl1,2-dihydronaphthalene-1,2-diol, trans-5,7-dimethyl-1,2-dihydronaphthalene-1,2-diol, and trans-4,6,7trimethyl-1,2-dihydronaphthalene-1,2-diol, were successfully prepared and used as standards in the GC-MS analysis, aiming to further develop this qualitative and quantitative analytical method for the determination of PAH exposures. This study shows that the currently used GC-MS analysis, including sample workup, is not suitable for determining the quantity of the corresponding diols derived from naphthalene and methylated naphthalenes. Alternative approaches are needed to provide a correct estimate of PAH exposure levels. © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

The analysis of PAH metabolites in fish bile is included in many international monitoring programs, using various species (HELCOM, 2015;Nyberg et al., 2013;Kammann et al., 2017), and is currently used as a biological marker of exposure to oil (Ariese et al., 2005;Beyer et al., 2010;Pampanin and Schlenk, 2020). Determination methods include fixed wavelength synchronous fluorescence scanning (SFS) (Ariese et al., 1993) or fluorescence (FF) (Lin et al., 1996), high performance liquid chromatography (HPLC) , and gas chromatography-mass spectrometry (GC-MS) (Jonsson et al., 2004), where the GC-MS analysis is the most commonly used for quantitative analyses (Beyer et al., 2010;Davies and Vethaak, 2012;Pampanin, 2017). Although FF and SFS are easy to carry out and not very costly (Aas et al., , 2000Beyer et al., 1998;Dissanayake and Galloway, 2004;Pathiratne et al., 2010;Sundt et al., 2012;Elcoroaristizabal et al., 2014), our previous studies have shown that results from FF and SFS should be treated with great care prior to drawing conclusions concerning exposure levels (Pampanin et al., 2016a(Pampanin et al., , 2016b. Therefore, GC-MS or various LC analyses are required in order to verify results from the former methods (Sundt et al., 2011;Pampanin et al., 2016aPampanin et al., , 2016b. In addition, the GC-MS analysis is the most suitable method for applied research in biomonitoring programs, as outlined by Davies and Vethaak (2012) and Iversen et al. (2015). More advance approaches, including double mass spectrometry analyses are currently under evaluation in the research community, and could provide an alternative in the long run. However, it is not expected that most laboratories can use this approach for routine monitoring activities at present.
The GC-MS method has its limitations and is only suitable for analysing metabolites derived from lighter PAHs found in crude oil, namely naphthalene, methylated variations of naphthalene (one, two, and three methyl groups), phenanthrene, and chrysene and some of their methylated variations (Beyer et al., 2010). The analytical method development focus has only been on the evaluation of metabolites that are easy to obtain analytical standards of, e.g. PAHs and their corresponding phenols (Fig. 2). However, it is well known that trans-1,2-dihydro-1,2-diols are commonly formed metabolites in vivo in fish upon exposure to PAHs (Pangrekar et al., 2003;Jonsson et al., 2004;Jacob, 2008;Pampanin and Sydnes, 2013). In order to evaluate the presence of these environmentally relevant metabolites in fish bile, synthetic standards need to be prepared, since they are not commercially available.

General experimental
Automated flash chromatography was performed on an Interchim PuriFlash® 215 chromatography system, detection at 254 nm. Infrared absorption spectroscopy was performed on a Cary 630 FTIR from Agilent Technologies. Proton ( 1 H) and carbon ( 13 C) NMR spectra were conducted on an AscendTM 400 NMR spectrometer from Bruker, which operated at 400 MHz and 100 MHz for proton and carbon, respectively. Chemical shifts (d) are reported relative to residual chloroform (CHCl 3 ) in deuterated chloroform (CDCl 3 ) (d 7.26 ppm, 1 H; d 77.16 ppm, 13 C) and residual methanol (CH 3 OH) in deuterated methanol (CD 3 OD) (d 3.31 ppm, 1 H; d 49.0 ppm, 13 C) as references. 1 H NMR data are reported by the following sequence: chemical shift (d) [multiplicity, coupling constant(s) J (Hz), relative integral], in which the multiplicity is reported as: s ¼ singlet; d ¼ doublet; dd ¼ doublet of doublett; t ¼ triplet; m ¼ multiplet; bs ¼ broad singlet. For 13 C NMR spectra, data are reported as chemical shift (d). Melting points (mp) were determined on a Stuart SMP20 melting point apparatus and are uncorrected. The GC-MS analysis was run on an Agilent 6890 N gas chromatograph, Gerstel MPS autosampler and an Agilent 5975 mass spectrometer. Bile samples from exposed Atlantic cod were obtained from a previously reported study (Enerstvedt et al., 2018).

Preparation of bile for GC-MS analysis
The preparation of hydrolyzed bile samples was performed as described in the standard operating procedure developed at NORCE and is based on previous work by Krahn et al. (1987), Jonsson et al. (2003), and Aas et al. (1998Aas et al. ( , 2000. Bile samples were obtained from freezer (À80 C) and thawed on ice for about 30 min before the hydrolysis.
In brief, bile samples (30 mL) and surrogate standard 2,6dibromophenol (2.24 ppm, 100 mL) were treated with 300 mL of b-glucuronidase (100 000 units/mL) with sulphathase activity (7500 units/mL) diluted by in 1:10 sodium acetate buffer (0.4 M, pH 5) for 2 h at 40 C. Hydrolyzed metabolites were extracted with EtOAc (0.5 mL x 4) and extracts were dried with sodium sulphate. Extracts were then transferred to clean scintillation vials and evaporated until approximately 500 mL EtOAc was remaining, the BSTFA (200 mL, 0.19 g, 0.75 mmol) was added. The mixture was incubated at 60 C for 2 h the TPA as a GC internal standard (4.23 ppm, 20 mL) was added prior to the GC-MS analysis.

GC-MS conditions
TMS derivatives of naphthalene trans-1,2-dihydro-1,2-diols were analysed by the GC-MS system. Helium was used as carrier gas and the applied column was HP 5MS (30 m Â 0.25 mm and 0.25 mm from Agilent Technologies). Samples and calibration standards (1 mL) were injected on a split/splitless injector with splitless mode for 1 min. Temperatures for the injector, transferline and ion source were held at 280, 300 and 250 C, respectively, and the GC oven temperature program was: 85 C for 1 min, 85e120 C at 20 C/min, 120e300 C at 8 C/min, and held at 300 C for 7 min. The quantitative determination was done in selected ion mode (SIM). Targeted mass to charge (m/z) ratios were selected on the basis of the preliminary analysis in full scan mode (SCAN) at 70 eV, in order to identify the most abundant ions.

Preparation of trans-dihydronaphthalene-diols 1(±)-4(±) for the GC-MS analysis
With the four dihydro-diols (1(±)-4(±)) in hand, the work shifted towards the evaluation of their ability to function as standards for the GC-MS analysis. In order to improve the currently used GC-MS method for PAH metabolites in bile, the diols also needed to be readily converted to the corresponding trimethylsilyl (TMS) ethers in the same efficient way as the phenols depicted in Fig. 2 are converted to the corresponding TMS-ethers in the work by Krahn et al. (1987) and Jonsson et al. (2003). By treating the trans-dihydronaphthalene-diols with BSTFA in EtOAc at 60 C for 2 h, the standard method used for conversion of phenols to the corresponding TMS-ethers, a range of products were formed from compounds 1(±), 3(±), and 4(±), as highlighted by the products detected by the GC-MS analysis (Scheme 2) and as exemplified by the chromatogram of compound 1(±) in Fig. 4b. One of the products formed was the desired di-TMS-ether 5(±), in addition to both the possible mono TMS-ethers 6(±) and 7(±). Moreover, small amounts of compounds 8 and 9, derived from the loss of water followed by the conversion of the resulting phenols to the corresponding TMS-ethers, were also detected. A similar product distribution was also detected when compounds 3(±) and 4(±) were subjected to the same reaction conditions. Products derived from trans-6-methyl-1,2-dihydronaphthalene-1,2-diol 2(±) were not detected in the GC-MS analysis, most likely due to the decomposition of the derived compound on the GC column.
Attempts to improve the results from the derivatization, by enhancing the reaction time, increasing the temperature, or increasing the amount of BSTFA, did not increase the outcome. In order to verify our reaction conditions, we utilized the same derivatization conditions on 1-naphthol, which resulted in, as expected, a clean conversion to the corresponding TMS-ether 8, as shown in the GC chromatogram (mass confirmed by the MS analysis) (Fig. 4a). This confirmed that our standard reaction conditions were providing the desired result for the phenols. The lack of conversion to single products, when compounds 1(±), 3(±), and 4(±) were treated under standard derivatization conditions (i.e. conditions used on bile samples), highlighted the fact that sample workup conditions used for the GC-MS analysis of fish bile has limitations when it comes to the quantification of lighter PAH metabolites. Small quantities of diol are converted to the two naphthol derivatives, viz compounds 8 and 9, which are Scheme 1. Synthesis of trans-dihydronaphthalene-diols 1(±)-4(±). Reaction conditions: (a) NaBH 4 , O 2 , EtOH, rt; (b) IBX, dry DMSO, rt; (c) Laccase (113 U), acetate buffer (0.1 M, pH 4.5), 0 C / rt (% ¼ chemical yield for the reaction).
Scheme 2. Products formed upon reaction with N,O-bis(trimethylsilyl-fluoro)acetamide (BSTFA) exemplified with trans-1,2-dihydronaphthalene-1,2-diol (1(±)). quantified in the method used today. However, the full overview is lost since the majority of the product mixture derived from the diols are other derivatives that are not included in the current analytical scheme.

The GC-MS analysis of fish bile samples
Although the derivatization and analysis of the three standard compounds indicated that it would not be possible to utilize the method for quantitative analysis of compounds 1(±), 3(±), and 4(±) in bile, we did investigate their use as standards for the qualitative analysis of fish bile. Therefore, 31 bile samples from Atlantic cod exposed to dispersed crude oil were analysed (Enerstvedt et al., 2018). The dispersed crude oil contained 6.7 mg/L naphthalene, 23 mg/L C1-naphthalene, 29 mg/L C2-naphthalene (containing two methyl groups), and 44 mg/L C3-naphthalene (containing three methyl groups) (Enerstvedt et al., 2018). The GC-MS analysis confirmed the presence of derivatives derived from trans-dihydronaphthalene-diols 1(±), 3(±), and 4(±) in samples from Atlantic cod exposed to medium and high concentrations of crude oil (see Fig. 5 for an example and supporting information for the full data set). The presence of TMS-ether derivatives from the different naphthalenes could be distinguished from the GC chromatogram and confirmed by their mass. However, a proper quantification of the parent compounds, viz. the trans-dihydronaphthalene-diols, could not be obtained due to the challenges outlined above.
Naturally the diols described in this work could have been detected by LC-MS techniques with plenty of examples of that being reported (Beyer et al., 2010). However, the aim of this study was to investigate the possibility to broaden the scope of the commonly used GC-MS method for analysis of fish bile for petrogenic PAH metabolites.

Conclusion
Four trans-1,2-dihydro-1,2-diols (1(±)-4(±)) were successfully prepared by synthesis, providing new standard compounds for PAH metabolite analyses. Unfortunately, the conversion of these standards to the corresponding single TMS-ethers failed, resulting in a range of products. These results show that the currently used sample workup conditions for the GC-MS analysis are not suitable for determining the quantity of the corresponding naphthalene trans-1,2-dihydro-1,2-diols metabolites in fish bile. In order to successfully be able to conduct quantification of trans-1,2-dihydro-1,2-diols (1(±)-4(±)), other analytical methods are therefore required. LC-MS techniques (e.g. atmospheric pressure chemical ionization in positive ionization mode (APCI þ ) LC/MS/MS) are recommended as preferable tools for studying PAH metabolites in fish bile, considering the limitation of the GC-MS method and the preliminary positive results obtained in recent studies (Sette et al., 2013). Our results also open up the possibility for further studies where other sample preparation methods for converting PAH metabolites in bile to suitable derivatives for the GC-MS analysis could be considered to facilitate the analysis of both phenols and diols.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.  5. The overlay total ion chromatograms (TICs) of two samples analysed by gas chromatography-mass spectrometry (GC-MS). The blue line represents a control bile sample spiked with the standards (1(±)-4(±)), and the black line shows a bile sample from fish exposed to medium concentration level of crude oil. C0, C2 and C3 represents the standards 1(±), 3(±) and 4(±), respectively (TPA ¼ triphenylamine; 2,6-DBP ¼ 2,6-dibromophenol). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)