Unique Physicochemical Properties of Perfluorinated Compounds and Their Bioconcentration in Common Carp Cyprinus carpio L.
Carp (Cyprinus carpio L.) was exposed to perfluorinated compounds (PFCs)—perfluoroalkyl carboxylic acids (number of carbon atoms, C = 8, 11, 12, 14, 16, and 18) and perfluorooctane sulfonate (PFOS)—in bioconcentration tests to compare the bioconcentration factors (BCFs) and physicochemical properties of each specific compound. Despite having the same number of carbon atoms (C = 8), the BCFs of perfulorooctanoic acid (PFOA) and PFOS differed by more than two orders of magnitude (PFOA BCF = < 5.1 to 9.4; PFOS BCF = 720 to 1300). The highest BCFs were obtained from perfluorododecanoic acid (BCF = 10,000 to 16,000) and perfluorotetradecanoic acid (BCF = 16,000 to 17,000). The longest observed depuration half-lives were for perfluorohexadecanoic acid (48 to 54 days) and PFOS (45 to 52 days). The concentrations of PFCs were highest in the viscera, followed by the head, integument, and remaining parts of the test fish. PFCs concentrations in the integument, which was in direct contact with the test substances, were relatively greater than that of other lipophilic substance (hexachlorobenzene). It is likely that Clog P would be a better parameter than log Kow for the prediction of BCFs for PFCs. Threshold values for PFCs bioaccumulation potential (molecular weight = 700, maximum diameter = 2 nm) seemed to deviate from those generally reported because of the specific steric bulk effect of molecule size.
Since the 1970s, there has been a steady increase in the commercial use of perfluorinated compounds (PFCs) as surfactants (Giesy and Kannan 2001). The perfluoroalkyl acids are a family of anionic fluorinated surfactants that consist of a carbon chain, typically 4 to 14 carbon atoms in length, and a charged functional moiety (primarily carboxylate, sulfonate, or phosphonate) (Lau et al. 2007). The most useful fluorinated surfactants are eight-carbon chemicals: perfulorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS). PFOS precursors have been used as water, oil, soil, and grease repellents for packaging, rugs and carpets, and leather; as surfactant detergents, emulsifiers, wetting agents, and dispersants; and in firefighting foam and other applications (Organisation for Economic Cooperation and Development [OECD] 2005). However, these chemicals, which do not exist naturally, are only slightly biodegradable and persist in the environment (Yakata et al. 2003). Some overviews of the monitoring and toxicological findings for PFCs (Lau et al. 2007; Buck et al. 2011; Kannan 2011) include studies of human blood (whole blood, plasma, and serum) (Olsen et al. 2003; Kannan et al. 2004; Calafat et al. 2007) and that of wildlife, including fish, birds, and marine mammals (Tolls et al. 1994; Giesy and Kannan 2001; Taniyasu et al. 2003; Lehmler 2005; Sinclair et al. 2006) as well as of their hepatotoxicity (Klaunig et al. 2003; Shipley et al. 2004), immunotoxicity (Yang et al. 2000; Wan and Badr 2006), hormonal effects (Langley and Pilcher 1985; Maras et al. 2006), reproductive toxicity (Peng et al. 2010; Stump et al. 2008), and developmental toxicity (Olsen et al. 2009).
There are many pathways for human exposure to PFCs, e.g., the atmosphere, drinking water, food, plants, soils, and fish (So et al. 2006; Stahl et al. 2009; Lau 2010; Murakami et al. 2011; Xiao et al. 2011). Fish is a primary source of protein for human consumption and would seem to be an important source of human exposure, considering PFCs pollution in the aquatic environment (Jin et al. 2009; Yoo et al. 2009; Tsuda et al. 2010).
On May 9, 2009, at the fourth meeting of the Conference of the Parties to the Stockholm Convention in Geneva (http://chm.pops.int/Convention/Pressrelease/COP4Geneva9May2009/tabid/542/language/en-US/Default.aspx), PFOS, and perfluorooctane sulfonyl fluoride were placed in Annex B of the convention. Despite worldwide interest, as evidenced by similar nominations of persistent organic pollutants (POPs), there are only a few experimental data for these chemicals using fish as test subjects, and there is a need for reliable hazard information about PFCs with longer (>7) fluorinated carbon chains (Conder et al. 2008). The development of quantitative structure–activity relationships (QSARs) has allowed the prediction of bioconcentration factors (BCFs) from the physicochemical properties of chemical substances, but these relationships are not well known for PFCs.
Our goal in this study was to clarify unique BCFs of PFCs using common carp (Cyprinus carpio L.) as a test organism and to discuss the bioaccumulation potential of PFCs based on their physicochemical properties.
Materials and Methods
Perfulorooctanoic acid (PFOA [98%]) was purchased from Daikin Industries, Ltd. (Osaka, Japan). Perfluoroundecanoic acid (PFUnA [99.0%]), perfluorotetradecanoic acid (PFTA [97.3%]), and perfluorooctadecanoic acid (PFODA [98.7%]) were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Perfluorododecanoic acid (PFDoA [98.5%]) was purchased from Wako Chemical, Ltd. (Kanagawa, Japan). Perfluorohexadecanoic acid (PFHxDA [100.67%]) was purchased from Lancaster Synthesis Ltd. (Lancashire, UK). PFOS (potassium salt [100.3%]) was purchased from Kishida Chemical Co., Ltd. (Osaka, Japan).
Measurement of BCFs
Test Fish and Feeding
Test fish were yearling carp Cyprinus carpio L. either purchased from a commercial fish farm or produced in our laboratory and were acclimatized according to the OECD (1996) test guideline (TG) 305. The lipid content of the test fish was measured as described by Bligh and Dyer (1959). The body weights, lengths, and lipid contents of the test fish at the beginning of the exposure phase were 3.15 to 7.77 g, 6.1 to 8.8 cm, and 3.10 to 3.87%, respectively. The fish were fed pellets for carp (Nippon Formula Feed Manufacturing, Kanagawa, Japan) twice a day (2% of the total body weight). Feeding was stopped 1 day before the fish were removed for analysis.
Preparation of Test Substance Stock Solution
Flow-through exposure tests require a system that continuously dispenses and dilutes a stock solution of the test substance to deliver the test concentrations to the test tanks (OECD 1996). The stock solution for the PFOS test was prepared by simply dissolving the test substance in deionized water. Stock solutions of other test substances were prepared using polyoxyethylene hydrogenated castor oil (HCO-40 or HCO-60; Nikko Chemicals Tokyo, Japan) as a dispersant at 1–50 times the amount of the test substance and then dissolving in 2-methoxyethanol (Wako Pure Chemical Industries) or N,N-dimethylformamide (Nacalai Tesque, Kyoto, Japan). For each test, a stock solution (at a flow rate of 0.02 to 2 ml min−1) was diluted by groundwater (800 to 1600 ml min−1) from the premises of the investigators’ laboratory. The solubilizing agents in the test water were within the range prescribed in OECD TG 305 (<0.1 ml l−1).
Test and Environmental Conditions
Bioconcentration test conditions
Test substance (acronym)
Nominal test concentrationa
Measured test concentrationa,b
Exposure period (d)
Depuration period (d)
66 or 23d
28 or 47e
Analysis for Test Substances
Concentrations of test substances in the test water were measured six times, and test substances in the fish were measured five times (in duplicate) during the exposure phase. At the end of the exposure phase, the distribution of test substances in tissues was determined, except for PFOA and PFODA, by measuring the concentrations in the integument (including alimentary canal and gills), head, viscera (internal organs except the alimentary canal), and remaining parts. During the depuration phase, concentrations of test substances in the test fish were measured at 4 to 5 time points (in duplicate) until the test duration was clearly past the depuration half-life of the test substances.
Test substances were analyzed by liquid chromatography-mass spectrometry (LC-MS) with a Micromass ZMD LC/MS (Micromass, Manchester, UK) and a Waters 2690 separations module (Waters, Milford, MA) or by liquid chromatography-tandem mass spectrometry (LC-MS/MS) with a Micromass Quattro-LC (Micromass) and liquid chromatograph type 1100 Series (Agilent, Waldbronn, Germany) in electrospray negative ionization mode; m/z = 413 (PFOA), 563 (PFUnA), 613 > 569 (PFDoA), 713 (PFTA), 813 > 769 (PFHxDA), 913 > 869 (PFODA), and 499 (PFOS). An L-column octadecyl-silane (ODS) (150 × 2.1 mm ID; particle size 5 μm [Chemicals Evaluation and Research Institute, Japan]) was used for all analyses. Mobile phase was mixture of acetonitrile/water or methanol/water containing 5 mmol l−1 ammonium acetate for PFUnA or 5 mmol l−1 di-n-butylammonium acetate for the other test substances.
For analysis of PFOA and PFOS in test water, the test water was directly analyzed by LC-MS after sampling from the test tank. For the other test substances, acetic acid (Kanto Chemical Tokyo, Japan) or formic acid (Wako Pure Chemical Industries) was added to the test water to create acidic conditions (pH 2 to 3), and then test substances were extracted by solid-phase extraction using a Sep-Pak C8 (Nihon Waters K.K., Tokyo, Japan) for PFUnA, PFDoA, and PFHxDA or Sep-Pak C18 (Nihon Waters K.K.) for PFTA and PFODA.
For analysis of test substances in test fish, whole fish were shredded using a polytron (Kinematica, Bohemia, NY), and a 3- to 5-g sample of each test fish was analyzed. The fish samples for PFOS analysis were homogenized in 15 ml of methanol (Wako Pure Chemical Industries) to extract the test substances from the fish sample. For the fish samples for PFOA analysis, acetonitrile (Wako Pure Chemical Industries) was used instead of methanol, and for the other test substances, acetonitrile with acetic acid or formic acid (0.5 to 2 ml) was used. The supernatant obtained by centrifugation of the homogenized sample (7000×g, 5 min) was brought to a volume of 25 ml with extracting solvent. These samples were analyzed by LC-MS or LC-MS/MS after diluting with water or methanol/water to the appropriate concentrations for analysis.
Calculation of BCFs
Depuration rate constants (k2) were determined by fitting the data to a first-order decay curve (ln concentration in fish = a + b × time [day], where a is a constant and b is k2). The depuration half-life (t1/2) was then calculated as: t1/2 = 0.693/k2.
Measurement of Physicochemical Properties
The solubility of each test substance in water was measured according to OECD TG 105 (OECD 1995); the flask method was used for PFOA and PFOS, and the column elution method was used for the other test substances. The n-octanol-water partition coefficient (log Pow) was calculated by two software programs: Kowwin v. 1.67 (United States Environmental Protection Agency [USEPA]) for log Kow and Clog P v. 4.0 (Biobyte, Claremont, CA) for Clog P, respectively. Clog P is a model for calculating the log P from chemical structure based on fragmental method, which represents log P as sum of fragment constants and correction factors (Hansch and Leo 1995). Detailed information about Clog P is available at: http://www.biobyte.com/index.html.
The log Pow values for PFOA, PFUnA, and PFTA were measured according to OECD TG 117 (OECD 2004) using the high-performance liquid chromatography method with acidic buffer for the eluent (pH 1 to 2) to measure the undissociated (free acid) form of the test substances. To estimate the molecular size of test substances, we used OASIS Software–Database Manager v. 1.4 (Laboratory of Mathematical Chemistry, Bourgas, Bulgaria) to calculate molecular weight (MW), effective cross-sectional diameter (Deff), and maximum diameter (Dmax) with the following program settings: converter mode, automated precise; conformer generation, accurate; and calculation method, AM1.
Results and Discussion
BCFss and physicochemical properties of test substances
Depuration half-life (d)
Water solubilitya (mg l−1)
The BCFs and associated parameters of test substances are listed in Table 2. There was no mortality or abnormal behavior in any treated or control fish. The BCFs of test substances did not differ substantially between the 2 test concentrations, which differed by a factor of 10. The BCF of PFOA was low (<5.1 to 9.4), whereas the BCF of PFOS (720 to 1300), which has the same number of carbon atoms (C = 8), was greater by 2 orders of magnitude. Monitoring data showing a similar trend are reviewed by Lau et al. (2007); concentrations of PFOS in serum or plasma of humans and wildlife were approximately the same or one order of magnitude greater than those of PFOA. The reason for the different BCFs for PFOS and PFOA is not clear but may be due to the different surfactant properties of the molecules. Additional study will be required to resolve this issue.
There was no apparent relationship between the number of carbon atoms in the test substances and the depuration half-life. The depuration half-lives of PFCs were between 8 and 29 days, but they were much longer for PFHxDA (48 to 54 days) and PFOS (45 to 52 days), with values similar to those for POPs, such as polychlorinated biphenyls (24 to 224 days) and hexachlorobenzene (HCB; 42 to 43 days) (Fisk et al. 1998). This result reflects the persistence of PFCs as pollutants in aquatic organisms; more laboratory and field research is necessary to fully characterize the hazards of PFCs with longer perfluoroalkyl chains as well as those of PFOS.
Concentrations of test substances and other type of chemicals as well as lipid content in each fish tissue and ratio of the concentrations to that in the head
ng g−1 (ratio)
Lipid content (μg g−1)e
Relationship Between BCFs and Physicochemical Properties
There are large differences between the curvilinear relationships for BCFs and log Kow or Clog P. The measured log Pow of PFOA (log Pow = 2.8), PFUnA (log Pow = 4.0), and PFTA (log Pow = 5.1) was a good fit to the curvilinear relationship between BCFs and Clog P. Thus, it is likely that Clog P would be a better parameter than log Kow for the prediction of BCFs for PFCs.
In this study, we emphasized that the physicochemical properties of PFCs lead to unique bioconcentration potentials in carp, and threshold values for PFCs bioaccumulation potential seemed to deviate from those generally reported because of the specific steric bulk effect of molecule size. Care should be taken when estimating the bioaccumulation potential of PFCs using existing regulatory criteria or QSARs. Our results also provide evidence to explain the persistent residual of PFCs with long perfluoroalkyl chains (e.g., PFTA and PFUnA) in aquatic organism; more reliable hazard information for PFCs is required in the future.
This work was supported by the Ministry of Economy, Trade and Industry and the New Energy and Industrial Technology Development Organization. We especially thank Y. Sakuratani (Chemical Management Center, National Institute of Technology and Evaluation) for the estimation of molecular size and for technical support.
- Arnot JA, Arnot M, Mackay D, Couillard Y, MacDonald D, Bonnell M, Doyle P (2010) Molecular size cutoff criteria for screening bioaccumulation potential: fact or fiction? Integr Environ Assess Manag 6:210–224Google Scholar
- Bertelsen SL, Hoffman AD, Gallinat CA, Elonen CM, Nichols JW (1998) Evaluation of log K ow and tissue lipid content as predictors of chemical partitioning to fish tissues. Environ Toxicol Chem 17:1447–1455Google Scholar
- European Community (2003) Second edition of the technical guidance document in support of commission directive 93/67/eec on risk assessment for new notified substances. Commission regulation (EC) 1488/94 on risk assessment for existing substances and directive 98/8/EC of the European Parliament and of the Council concerning the placing of biocidal products on the market. Office for Official Publications of the European Communities, Luxembourg (BE)Google Scholar
- Hansch C, Leo A (1995) Exploring QSAR: fundamentals and applications in chemistry and biology. American Chemical Society, WashingtonGoogle Scholar
- Kannan K (2011) Perfluoroalkyl and polyfluoroalkyl substances: current and future perspectives. Environ Chem 8:333–338Google Scholar
- Lau C (2010) Toxicology of perfluoroalkyl acids. Available at: http://water.epa.gov/scitech/swguidance/fishshellfish/fishadvisories/upload/day2b.pdf. Accessed 14 Mar 2011
- Martin JW, Mabury SA, Solomon KR, Muir DCG (2003) Bioconcentration and tissue distribution of perfluorinated acids in rainbow trout (Oncorhynchus mykiss). Environ Toxicol Chem 22:196–204Google Scholar
- National Institute of Technology and Evaluation, Japan. CHEmical Collaboration Knowledge database (J-CHECK)(1994). Available at: http://www.safe.nite.go.jp/jcheck/data/Article_PDF/2-0287_000307-70-0/2-0287_000307-70-0_Bioacc_01.pdf. Accessed 14 Mar 2011
- Organization for Economic Cooperation, Development (1995) OECD Guidelines for Testing of Chemicals, Guideline 105, Water Solubility. Organization for Economic Cooperation and Development, ParisGoogle Scholar
- Organization for Economic Cooperation, Development (1996) OECD Guidelines for Testing of Chemicals, Guideline 305, Bioconcentration: Flow-through Fish Test. Organization for Economic Cooperation and Development, ParisGoogle Scholar
- Organization for Economic Cooperation, Development (2004) OECD Guidelines for Testing of Chemicals, Guideline 117, Partition Coefficient (n-octanol/water), High Performance Liquid Chromatography (HPLC) Method. Organization for Economic Cooperation and Development, ParisGoogle Scholar
- Organization for Economic Cooperation and Development (2005) Results of survey on production and use of PFOS, PFAS and PFOA, relates substances and products/mixtures containing these substances. ENV/JM/MONO(2005)1Google Scholar
- Sakuratani Y, Noguchi Y, Kobayashi K, Yamada J, Nishihara T (2008) Molecular size as a limiting characteristic for bioconcentration in fish. J Environ Biol 29:89–92Google Scholar
- Tietge JE, Johnson RD, Jensen KM, Cook PM, Elonen GE, Fernandez JD et al (1998) Reproductive toxicity and disposition of 2,3,7,8-tetrachlorodibenzo-p-dioxin in adult brook trout (Salvelinus fontinalis) following a dietary exposure. Environ Toxicol Chem 17:2395–2407Google Scholar
- Xiao F, Simcik MF, Gulliver JS (2011) Partitioning characteristics of perfluorooctane sulfonate between water and foods. Arch Environ Contam Toxicol. doi:10.1007/s00244-011-9683-x