- Email: [email protected]
Detection of free and covalently bound microcystins in animal tissues by liquid chromatography–tandem mass spectrometry
Environmental Pollution 158 (2010) 948–952
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Short communication
Detection of free and covalently bound microcystins in animal tissues by liquid chromatography–tandem mass spectrometry Milla-Riina Neffling a, *, Emilie Lance b, Jussi Meriluoto a a b
¨ katu 6 A, Biocity 3rd floor, FI-20520, Turku, Finland Department of Biochemistry and Pharmacy, Åbo Akademi University, Tykisto UMR CNRS Ecobio 6553, University of Rennes 1, Avenue du Ge´ne´ral Leclerc, 35042, Rennes Cedex, France
The study concerns method development for the LC–MS–MS analysis of both free and protein-bound microcystin in tissue materials.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 June 2009 Received in revised form 22 August 2009 Accepted 15 October 2009
Microcystins are cyanobacterial hepatotoxins capable of accumulation into animal tissues. The toxins act by inhibiting specific protein phosphatases and both non-covalent and covalent interactions occur. The 2-methyl-3-methoxy-4-phenylbutyric acid (MMPB) method determines the total, i.e. the sum of free and protein-bound microcystin in tissues. The aim of the method development in this paper was to tackle the problems with the MMPB methodology: the rather laborious workflow and the loss of material during different steps of the method. In the optimised workflow the oxidation recovery was of acceptable level (29–40%), the extraction efficiency good (62–97%), but the signal suppression effect from the matrix remained severe in our system (16–37% signal left). The extraction efficiency for the determination of the free, extractable microcystins, was found to be good, 52–100%, depending on the sample and the toxin variant and concentration. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Microcystins MMPB method Total contamination levels
1. Introduction Cyanobacteria (blue-green algae) are encountered worldwide, especially in eutrophicated water bodies (van Apeldoorn et al., 2007). Toxic cyanobacteria cause problems in aquatic biota but also in terrestrial animals feeding on the aquatic biota (Codd et al., 2005). The main hepatotoxins produced by cyanobacteria, microcystins (MCs) and nodularins (Nods), are cyclic hepta- and pentapeptides, respectively. The cyclic structure contains some uncommon amino acids, including the b- amino acid Adda, 3-amino-9-methoxy-2,6, 8-trimethyl-10-phenyldeca-4(E),6(E)-dienoic acid (Rinehart et al., 1988), which is characteristic for these hepatotoxins and required for toxicity. MCs and Nods are known to accumulate into animal tissues and the highest toxin concentrations have been found in the liver (WHO, 1998; van Apeldoorn et al., 2007). MCs and Nods are potent and specific inhibitors of protein phosphatases (PPs) 1, 2A, 4 and 5 (Honkanen et al., 1990; MacKintosh et al., 1990; Hastie et al., 2005). After exposure MCs can form a covalent bond with PPs 1 and 2A (Robinson et al., 1991). Both covalent and non-covalent interactions with the PPs give rise to effective inhibition.
* Corresponding author. Tel.: þ358 2 215 4028; fax: þ358 2 241 0014. E-mail address: mneffl[email protected] (M.-R. Neffling). 0269-7491/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2009.10.023
There are several methods for the detection of MCs: enzymelinked immunosorbent assay (ELISA), protein phosphatase inhibition assay and high-performance liquid chromatography (HPLC) with UV or mass spectrometic (MS) detection (van Apeldoorn et al., 2007). These methods only detect the toxin available to extraction, but not the covalently bound toxin. ELISA and LC–MS are more suitable for toxin detection in difficult, complex matrices, while protein phosphatase assay and HPLC–UV often suffer from high background in biological samples. The detection responses with ELISA assays may vary significantly towards different MC variants (Young et al., 2006). The 2-methyl-3-methoxy-4-phenylbutyric acid (MMPB) method comprises the oxidation of Adda amino acid to form the carboxylic acid MMPB (Fig. 1) by Lemieux oxidation (Sano et al., 1992) or by ozonolysis (Harada et al., 1996). All MC variants containing Adda give rise to one MMPB molecule and are therefore detected with equal sensitivity. The MMPB method determines the concentration of total, i.e. free and covalently bound MCs. After the oxidation, the MMPB-containing sample is cleaned up by liquid–liquid (Williams et al., 1997a,b; Kaya and Sano, 1999) or solid-phase extraction (SPE) (Ott and Carmichael, 2006; Yuan et al., 2006), or derivatised for GC–MS analysis (Tsuji et al., 2001). The aim of this study was to further optimise the MMPB–method for the LC–MS–MS detection of the total (free and protein-bound) MC contamination in tissues.
M.-R. Neffling et al. / Environmental Pollution 158 (2010) 948–952
(6) D-Glu
CH3
N H
O
CH2
NH CH3
Oxidation: KMnO4 and NaIO4
CH3
H2N (4) L-Arg
O
N
H3C
OCH3
(7) Mdha
COOH
O
(5) Adda
NH
H 3C
O NH CH3
NH
COOH
(1) D-Ala
O NH
H N
O NH
949
O
CH3 CH3 (2) L-Leu
(3) D-methyl-Asp
OCH3 O OH MMPB
CH3
Fig. 1. The oxidation of Adda in microcystins (exemplified here by MC-LR) gives rise to the carboxylic acid MMPB (2-methyl-3-methoxy-4-phenylbutyric acid). This method detects both free and protein-bound microcystins.
2. Materials and methods 2.1. Materials The reagents were of analytical or chromatographic grade. Trypsin (10 solution, 25 g L1) and fetal bovine serum (F7524) were from Sigma–Aldrich Chemie (Steinheim, Germany). MMPB standard was a kind gift from Wako Pure Chemical Industries, Ltd (Osaka, Japan) and Prof. K.-i. Harada. The aquatic snail, Lymnaea stagnalis, tissue used for the experiments was prepared as described in Lance et al. (in press). In short, the L. stagnalis were fed with MC-producing cyanobacteria in diet or exposed to dissolved MC-LR. The fetal bovine serum was chosen as a biological test fluid in order to evaluate the current methodology for serological studies, such as those performed by Yuan and co-workers (Yuan et al., 2006). 2.2. Extraction of free microcystins The extraction efficiency for MCs in freeze-dried L. stagnalis tissue was tested with two different extraction solvents, 75% methanol and water:methanol:butanol (75:20:5) (Karlsson et al., 2003), and with two SPE cartridges, Oasis HLB 30 mg from Waters (Milford, MA, USA) and Supelclean LC-18 (1 mL) from Supelco (Bellefonte, PA, USA). The effect of the SPE washing solution (10% or 20% aqueous methanol) was also tested. The optimised extraction method for L. stagnalis freeze-dried tissue is described in Fig. 2. The extraction efficiency and matrix effect were determined by spiking the samples before and after the extraction (Table 1, Fig. 2). The spiking solution, an extract of cyanobacterial natural bloom, contained known amounts of MC-RR, MC-LR and MC-YR. The spiked fetal bovine serum sample extraction included a protein precipitation step with 80% acetonitrile containing 0.1% trifluoroacetic acid (TFA). The serum extract was evaporated to dryness in glass vials with argon at 40 C, and then reconstituted in water for SPE. The SPE wash step for serum samples was performed with 15% methanol containing 0.1 M acetic acid (HOAc) and the elution with 80% methanol containing 0.1 M HOAc. 2.3. Determination of total microcystins 10 mg of freeze-dried tissue or 1 mL of serum was trypsinated with 1 mL of 500 mg mL1 of trypsin in So¨rensen’s phosphate buffer (pH 7.5) at 37 C in shaking water bath for 2 h, followed by oxidation with 0.1 M KMnO4 and 0.1 M NaIO3, pH 9.0 for 3 h. When the colour of the sample solution turned from purple (colour of KMnO4) to brown (colour of the reduced product MnO2) during the oxidation procedure, more oxidation solution was added until the colour was purple again. The reaction was quenched by sodium bisulphite solution (40% w/v), until the sample was colourless. 10% sulphuric acid was added until the pH of the solution was acidic (w2). The optimised MMPB method workflow for both aquatic snail and serum samples is described in Fig. 2. The MMPB recovery during the drying process, argon at 40 C, was assessed with preparing 1 mL of MMPB solutions in 80% methanol, 0.05% HOAc with concentration of 2 ng mL1 (n ¼ 6) and 20 ng mL1 (n ¼ 6). Three samples of each concentration
were left to evaporate at 40 C, under argon until 20–30 mL solution remained, and three samples were completely dried down. The dried samples were kept at 40 C under argon for at least 5 min. All samples were then reconstituted to 200 mL with 35% methanol containing 0.05% HOAc before analysis. Adduct formation (Naþ, Kþ, NHþ 4 ) in LC–ESI-MS was tested by preparing MMPB samples (0.5 mg mL1) in 40% methanol containing 0.6 mM of either sodium chloride, potassium chloride or ammonium acetate. The adduct formation was tested with LC–MS runs in single ion recording (SIR), scan and multiple reactant monitoring (MRM) modes. The sample oxidation recovery is the measure of the proportion of MCs in a tissue recovered as the oxidation product MMPB. Some of the MCs in the tissue may escape oxidation and some of the formed MMPB may be further oxidised and lost. The oxidation recovery, extraction efficiency and the matrix effects were controlled by spiking of control samples (Fig. 2). The oxidation recovery controls were spiked with a known amount of MCs before any treatment and compared to controls spiked with the corresponding amount of MMPB standard after the oxidation procedure, but before the SPE (Fig. 2). To assess the MMPB extraction efficiency controls spiked with MMPB standard before and after SPE were compared (Fig. 2).
2.4. LC–MS analysis The LC–MS analysis was performed with an Agilent 1100 binary HPLC system (Waldbronn, Germany) coupled to a Quattro Micro (Waters, Manchester, UK) triple quadrupole mass spectrometer with electrospray ionisation (ESI) source. The column was a Purospher STAR RP-18e (Merck, Darmstadt, Germany) 30 mm 4 mm i.d. with 3 mm particles, kept at 40 C. The mobile phase consisted of solvent A: 0.1% formic acid and B: acetonitrile, and the gradient programme for MC detection was from 25% to 90% B over 8 min, held at 90% B for 1 min, flow rate 0.5 mL min1, followed by equilibration at 25% for 4 min, flow rate 1 mL min1. The injection volume was 10 mL. Some MS ion source parameters were: source temperature 150 C, desolvation temperature 300 C, cone gas flow 60 L h1, desolvation gas flow 640 L h1, capillary voltage 3 kV, positive ESI mode. The cone voltages (33 V-75 V) and collision energies (25 eV-65 eV) were optimised for each analyte. The monitored signals in the SIR mode were m/z [dmMC-RRþ2H]2þ 512.8, [MCRRþ2H]2þ 519.8, [dmMC-LRþH]þ 981.5, [MC-LRþH]þ 995.5, [MC-YRþH]þ 1045.5. In the MRM mode the corresponding transitions to m/z 135 (Adda fragment) were recorded. The mobile phase gradient for MMPB detection was from 40% to 70% B over 3 min, increase of B to 90% B over 0.1 min, which was held for 1 min, flow rate 0.5 mL min1, followed by equilibration at 40% B for 4 min, flow rate 1 mL min1. For MMPB MS analysis cone voltage 5 V and collision energies 7 eV-11 eV were used for transitions in positive MRM mode, [MMPB þ H]þ m/z 209.2, to m/z 91, 131 and 191.
950
M.-R. Neffling et al. / Environmental Pollution 158 (2010) 948–952
Fig. 2. Optimised workflow for the determination of free/extractable microcystins in freeze-dried tissue, as well as total (free þ bound) microcystins in freeze-dried tissue and in serum.
3. Results and discussion
better recoveries than 10% methanol. Serum and freeze-dried tissue extraction efficiencies are reported in Table 1.
3.1. Determination of free microcystins 3.2. Determination of total microcystins The extraction efficiency with the water:methanol:butanol solution was higher than with 75% methanol and the Oasis HLB cartridges gave better recoveries than Supelclean C-18. 20% methanol in the wash step reduced the signal suppression and thus gave
The earlier reported MMPB workflow (Ott and Carmichael, 2006; Yuan et al., 2006) was changed by including trypsination, by stronger concentration of oxidative solution and by sample
M.-R. Neffling et al. / Environmental Pollution 158 (2010) 948–952 Table 1 Extraction of the toxins, extraction efficiencies. Extraction efficiency/% Sample
Spiked with 100 mL of
Serum 1 mL
Natural bloom extract 100.2 Natural bloom extract 1:10 86.7 Natural bloom extract 1:100 83.9
MC-RR MC-LR MC-YR Average 86.1 78.9 73.2
87.2 80.9 84.0
91.2 82.2 80.4
Serum 0.1 mL Natural bloom extract 1:2 Natural bloom extract 1:20
90.8 80.8
86.1 74.0
82.7 86.3
86.5 80.3
Snail 10 mg
78.1 51.0 66.7
76.8 56.8 57.2
77.3 58.5 73.8
77.4 55.4 65.9
Natural bloom extract Natural bloom extract 1:10 Natural bloom extract 1:100
The cyanobacterial natural bloom extract contained 0.6 mg MC-RR, 0.6 mg MC-LR, 0.4 mg MC-YR. This extract or its dilutions (1:2, 1:20, 1:10, 1:100) were used to spike control samples (Fig. 2). The values given are averages from SIR and MRM measurements.
extraction as described in Fig. 2. The trypsination step was adapted from a forensic procedure (Pelander et al., 2007) and was expected to better expose the covalently bound MCs to oxidation. The addition of oxidant solution (to keep the solution purple) and solution of sodium bisulphite (to quench the reaction) instead of solid KMnO4 and solid sodium bisulphite (as reported in Ott and Carmichael, 2006) enabled more accurate proportioning of the reagents. Centrifugation after oxidation was found to be crucial for successful SPE treatment. No significant losses due to the centrifugation were observed. The sodium adduct of MMPB was observed, but the protonated molecular ion was dominant. The potassium and ammonium adducts were of signal intensity below limit of detection. Furthermore, there were no observable losses of the MMPB during the evaporation test. The SPE efficiency of MMPB was relatively good (considering the difficult matrix), 70–83% (Table 2). The cartridges thus tolerated the low pH and the rather large sample volume (up to 15 mL). The matrix effects were severe with the MMPB tissue analysis samples. The serum samples spiked at the final step of sample preparation, after oxidation and extraction, revealed that the recovered signal was only 16% (1 mL serum handled, spiked with 33 ng MMPB, corresponding to 160 ng of MCs) to 37% (0.1 mL serum handled; 16.6 ng of MMPB) as compared to the signal from the standards (Table 2). The same signal suppression phenomenon was also seen with the aquatic snail samples (Table 2.). Some problems might be overcome by switching to negative mode MS detection as described in Ott and Carmichael (2006), but in our MS system the signal intensity of the negative MMPB ions was not adequate.
Table 2 MMPB method: oxidation recovery, extraction efficiency and matrix effects. Sample Serum 1 mL 1 mL 0.1 mL 0.1 mL
Spiked with 100 mL containing MCs/ng
Oxidation recovery/%
Extraction efficiency/%
Matrix effects, Signal in spiked samples/%
1600 160 800 80
31.6 39.5 28.9 33.9
85.6 97.2 70.7 73.0
18.4 16.3 34.8 37.3
38.4 36.2 22.4
62.4 76.3 44.9
25.8 18.0 27.4
Aquatic Snail Tissue 10 mg 8800 10 mg 1600 10 mg 160
Oxidation recovery refers to comparison of samples spiked with MCs before oxidation to samples spiked with the corresponding amount of MMPB after the oxidation. Extraction efficiency refers to comparison of samples spiked with MMPB after oxidation to samples spiked after SPE. The matrix effect gives the percentage signal recovered from spiked samples as compared to recovered signal from MMPB standard in the absence of tissue. Please see Fig. 2 for spiking points.
951
Due to the high amounts of covalently bound MCs in the tissues of L. stagnalis found in an accompanying publication (Lance et al., in press) using the methods described here, it seems unlikely that the only target for covalent binding of MCs in the tissue would be the protein phosphatases. We found up to 37 mg of covalently bound toxins per gram of freeze-dried aquatic snail tissue. The proportion of covalently bound toxin, as compared to proportion of extractable toxin, is lower than found by Williams and co-workers (Williams et al., 1997a,b,c). No glutathione and cysteine conjugates involved in detoxification of MCs (Kondo et al., 1992) were detected. Assuming that all covalently bound toxins would be bound to protein phosphatases (one binding site per 35–36 kDa catalytic subunit) the amount of covalently inhibited protein phosphatase catalytic subunits in the freeze-dried tissue would be 1.3 mg g1, which is unrealistic. MC-LR functions also as an ATP synthase inhibitor (Mikhailov et al., 2003). Imanishi and Harada presented a study to discover other MC-binding proteins by affinity chromatography and proteomics methods (Imanishi and Harada, 2004). This kind of experimental set-up could be useful in revealing other possible targets for MC binding and thus contributing in the toxicosis by MCs in different tissues. 4. Conclusions The MMPB method from early 1990s is important when assessing the total MC contamination levels in different organisms affected by the cyanobacterial blooms. However, this method has been quite laborious and not employed in many publications. The modified method reported here worked with reasonable oxidation recoveries and extraction efficiencies for two different sample types, aquatic snail tissue and bovine serum. Acknowledgements The authors would like to thank the National Graduate school in Informational and Structural Biology, Medicinska understo¨dsfo¨reningen Liv och Ha¨lsa, Otto A. Malms donations fond (M-R N.), and the CIMO (Center for International Mobility) and the French and Finnish Ministries of Foreign Affairs (E.L.), Tekes decision number 40366/06 and Academy of Finland decision number 108947 (J.M.). Dr. Krister Karlsson is thanked for his helpful advice in the early stages of the study. References van Apeldoorn, M.E., van Egmond, H.P., Speijers, G.J., Bakker, G.J., 2007. Toxins of cyanobacteria. Mol. Nutr. Food Res. 51, 7–60. Codd, G.A., Morrison, L.F., Metcalf, J.S., 2005. Cyanobacterial toxins: risk management for health protection. Toxicol. Appl. Pharmacol. 203, 264–272. Harada, K., Murata, H., Qiang, Z., Suzuki, M., Kondo, F., 1996. Mass spectrometric screening method for microcystins in cyanobacteria. Toxicon 34, 701–710. Hastie, C.J., Borthwick, E.B., Morrison, L.F., Codd, G.A., Cohen, P.T.W., 2005. Inhibition of several protein phosphatases by a non-covalently interacting microcystin and a novel cyanobacterial peptide, nostocyclin. Biochim. Biophys. Acta 1726, 187–193. Honkanen, R.E., Zwiller, J., Moore, R.E., Daily, S.L., Khatra, B.S., Dukelow, M., Boynton, A.L., 1990. Characterization of microcystin-LR, a potent inhibitor of type 1 and type 2A protein phosphatases. J. Biol. Chem. 265, 19401–19404. Imanishi, S., Harada, K., 2004. Proteomics approach on microcystin binding proteins in mouse liver for investigation of microcystin toxicity. Toxicon 43, 651–659. Karlsson, K., Sipia, V., Kankaanpaa, H., Meriluoto, J., 2003. Mass spectrometric detection of nodularin and desmethylnodularin in mussels and flounders. J. Chromatogr. B 784, 243–253. Kaya, K., Sano, T., 1999. Total microcystin determination using erythro-2-methyl-3(methoxy-d(3))-4-phenylbutyric acid (MMPB-d(3)) as the internal standard. Anal. Chim. Acta 386, 107–112. Kondo, F., Ikai, Y., Oka, H., Okumura, M., Ishikawa, N., Harada, K., Matsuura, K., Murata, H., Suzuki, M., 1992. Formation, characterization, and toxicity of the glutathione and cysteine conjugates of toxic heptapeptide microcystins. Chem. Res. Toxicol. 5, 591–596. Lance, E., Neffling, M.-R, Ge´rard, C., Meriluoto, J., Bormans, M., Accumulation of free and covalently bound microcystins in tissues of Lymnaea stagnalis (Gastropoda)
952
M.-R. Neffling et al. / Environmental Pollution 158 (2010) 948–952
following toxic cyanobacteria or dissolved microcystin-LR exposure. Environ. Pollut., in press, doi:10.1016/j.envpol.2009.10.025. MacKintosh, C., Beattie, K.A., Klumpp, S., Cohen, P., Codd, G.A., 1990. Cyanobacterial microcystin-LR is a potent and specific inhibitor of protein phosphatases 1 and 2A from both mammals and higher plants. FEBS Lett. 264, 187–192. Mikhailov, A., Ha¨rma¨la¨-Braske´n, A.S., Hellman, J., Meriluoto, J., Eriksson, J.E., 2003. Identification of ATP-synthase as a novel intracellular target for microcystin-LR. Chem. Biol. Interact 142, 223–237. Ott, J.L., Carmichael, W.W., 2006. LC/ESI/MS method development for the analysis of hepatotoxic cyclic peptide microcystins in animal tissues. Toxicon 47, 734–741. Pelander, A., Backstro¨m, D., Ojanpera¨, I., 2007. Qualitative screening for basic drugs in autopsy liver samples by dual-plate overpressured layer chromatography. J. Chromatogr. B Analyt Technol. Biomed. Life Sci. 857, 337–340. Rinehart, K.L., Harada, K., Namikoshi, M., Chen, C., Harvis, C.A., Munro, M.H.G., Blunt, J.W., Mulligan, P.E., Beasley, V.R., Dahlem, A.M., Carmichael, W.W., 1988. Nodularin, microcystin, and the configuration of Adda. J. Am. Chem. Soc. 110, 8557–8558. Robinson, N.A., Matson, C.F., Pace, J.G., 1991. Association of microcystin-LR and its biotransformation product with a hepatic-cytosolic protein. J. Biochem. Toxicol. 6, 171–180. Sano, T., Nohara, K., Shiraishi, F., Kaya, K., 1992. A method for microdetermination of total microcystin content in waterblooms of cyanobacteria (Blue-Green-Algae). Int. J. Environ. Ch 49, 163–170.
Tsuji, K., Masui, H., Uemura, H., Mori, Y., Harada, K., 2001. Analysis of microcystins in sediments using MMPB method. Toxicon 39, 687–692. WHO, 1998. Guidelines for Drinking-water Quality. World Health Organization, Geneva. Williams, D.E., Craig, M., McCready, T.L., Dawe, S.C., Kent, M.L., Holmes, C.F., Andersen, R.J., 1997a. Evidence for a covalently bound form of microcystin-LR in salmon liver and dungeness crab larvae. Chem. Res. Toxicol. 10, 463–469. Williams, D.E., Dawe, S.C., Kent, M.L., Andersen, R.J., Craig, M., Holmes, C.F., 1997b. Bioaccumulation and clearance of microcystins from salt water mussels, Mytilus edulis, and in vivo evidence for covalently bound microcystins in mussel tissues. Toxicon 35, 1617–1625. Williams, D.E., Craig, M., Dawe, S.C., Kent, M.L., Andersen, R.J., Holmes, C.F., 1997c. 14C-labeled microcystin-LR administered to Atlantic salmon via intraperitoneal injection provides in vivo evidence for covalent binding of microcystin-LR in salmon livers. Toxicon 35, 985–989. Young, F.M., Metcalf, J.S., Meriluoto, J.A.O., Spoof, L., Morrison, L.F., Codd, G.A., 2006. Production of antibodies against microcystin-RR for the assessment of purified microcystins and cyanobacterial environmental samples. TOXICON 48, 295–306. Yuan, M., Carmichael, W.W., Hilborn, E.D., 2006. Microcystin analysis in human sera and liver from human fatalities in Caruaru, Brazil 1996. Toxicon 48, 627–640.
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Short communication
Detection of free and covalently bound microcystins in animal tissues by liquid chromatography–tandem mass spectrometry Milla-Riina Neffling a, *, Emilie Lance b, Jussi Meriluoto a a b
¨ katu 6 A, Biocity 3rd floor, FI-20520, Turku, Finland Department of Biochemistry and Pharmacy, Åbo Akademi University, Tykisto UMR CNRS Ecobio 6553, University of Rennes 1, Avenue du Ge´ne´ral Leclerc, 35042, Rennes Cedex, France
The study concerns method development for the LC–MS–MS analysis of both free and protein-bound microcystin in tissue materials.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 June 2009 Received in revised form 22 August 2009 Accepted 15 October 2009
Microcystins are cyanobacterial hepatotoxins capable of accumulation into animal tissues. The toxins act by inhibiting specific protein phosphatases and both non-covalent and covalent interactions occur. The 2-methyl-3-methoxy-4-phenylbutyric acid (MMPB) method determines the total, i.e. the sum of free and protein-bound microcystin in tissues. The aim of the method development in this paper was to tackle the problems with the MMPB methodology: the rather laborious workflow and the loss of material during different steps of the method. In the optimised workflow the oxidation recovery was of acceptable level (29–40%), the extraction efficiency good (62–97%), but the signal suppression effect from the matrix remained severe in our system (16–37% signal left). The extraction efficiency for the determination of the free, extractable microcystins, was found to be good, 52–100%, depending on the sample and the toxin variant and concentration. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Microcystins MMPB method Total contamination levels
1. Introduction Cyanobacteria (blue-green algae) are encountered worldwide, especially in eutrophicated water bodies (van Apeldoorn et al., 2007). Toxic cyanobacteria cause problems in aquatic biota but also in terrestrial animals feeding on the aquatic biota (Codd et al., 2005). The main hepatotoxins produced by cyanobacteria, microcystins (MCs) and nodularins (Nods), are cyclic hepta- and pentapeptides, respectively. The cyclic structure contains some uncommon amino acids, including the b- amino acid Adda, 3-amino-9-methoxy-2,6, 8-trimethyl-10-phenyldeca-4(E),6(E)-dienoic acid (Rinehart et al., 1988), which is characteristic for these hepatotoxins and required for toxicity. MCs and Nods are known to accumulate into animal tissues and the highest toxin concentrations have been found in the liver (WHO, 1998; van Apeldoorn et al., 2007). MCs and Nods are potent and specific inhibitors of protein phosphatases (PPs) 1, 2A, 4 and 5 (Honkanen et al., 1990; MacKintosh et al., 1990; Hastie et al., 2005). After exposure MCs can form a covalent bond with PPs 1 and 2A (Robinson et al., 1991). Both covalent and non-covalent interactions with the PPs give rise to effective inhibition.
* Corresponding author. Tel.: þ358 2 215 4028; fax: þ358 2 241 0014. E-mail address: mneffl[email protected] (M.-R. Neffling). 0269-7491/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2009.10.023
There are several methods for the detection of MCs: enzymelinked immunosorbent assay (ELISA), protein phosphatase inhibition assay and high-performance liquid chromatography (HPLC) with UV or mass spectrometic (MS) detection (van Apeldoorn et al., 2007). These methods only detect the toxin available to extraction, but not the covalently bound toxin. ELISA and LC–MS are more suitable for toxin detection in difficult, complex matrices, while protein phosphatase assay and HPLC–UV often suffer from high background in biological samples. The detection responses with ELISA assays may vary significantly towards different MC variants (Young et al., 2006). The 2-methyl-3-methoxy-4-phenylbutyric acid (MMPB) method comprises the oxidation of Adda amino acid to form the carboxylic acid MMPB (Fig. 1) by Lemieux oxidation (Sano et al., 1992) or by ozonolysis (Harada et al., 1996). All MC variants containing Adda give rise to one MMPB molecule and are therefore detected with equal sensitivity. The MMPB method determines the concentration of total, i.e. free and covalently bound MCs. After the oxidation, the MMPB-containing sample is cleaned up by liquid–liquid (Williams et al., 1997a,b; Kaya and Sano, 1999) or solid-phase extraction (SPE) (Ott and Carmichael, 2006; Yuan et al., 2006), or derivatised for GC–MS analysis (Tsuji et al., 2001). The aim of this study was to further optimise the MMPB–method for the LC–MS–MS detection of the total (free and protein-bound) MC contamination in tissues.
M.-R. Neffling et al. / Environmental Pollution 158 (2010) 948–952
(6) D-Glu
CH3
N H
O
CH2
NH CH3
Oxidation: KMnO4 and NaIO4
CH3
H2N (4) L-Arg
O
N
H3C
OCH3
(7) Mdha
COOH
O
(5) Adda
NH
H 3C
O NH CH3
NH
COOH
(1) D-Ala
O NH
H N
O NH
949
O
CH3 CH3 (2) L-Leu
(3) D-methyl-Asp
OCH3 O OH MMPB
CH3
Fig. 1. The oxidation of Adda in microcystins (exemplified here by MC-LR) gives rise to the carboxylic acid MMPB (2-methyl-3-methoxy-4-phenylbutyric acid). This method detects both free and protein-bound microcystins.
2. Materials and methods 2.1. Materials The reagents were of analytical or chromatographic grade. Trypsin (10 solution, 25 g L1) and fetal bovine serum (F7524) were from Sigma–Aldrich Chemie (Steinheim, Germany). MMPB standard was a kind gift from Wako Pure Chemical Industries, Ltd (Osaka, Japan) and Prof. K.-i. Harada. The aquatic snail, Lymnaea stagnalis, tissue used for the experiments was prepared as described in Lance et al. (in press). In short, the L. stagnalis were fed with MC-producing cyanobacteria in diet or exposed to dissolved MC-LR. The fetal bovine serum was chosen as a biological test fluid in order to evaluate the current methodology for serological studies, such as those performed by Yuan and co-workers (Yuan et al., 2006). 2.2. Extraction of free microcystins The extraction efficiency for MCs in freeze-dried L. stagnalis tissue was tested with two different extraction solvents, 75% methanol and water:methanol:butanol (75:20:5) (Karlsson et al., 2003), and with two SPE cartridges, Oasis HLB 30 mg from Waters (Milford, MA, USA) and Supelclean LC-18 (1 mL) from Supelco (Bellefonte, PA, USA). The effect of the SPE washing solution (10% or 20% aqueous methanol) was also tested. The optimised extraction method for L. stagnalis freeze-dried tissue is described in Fig. 2. The extraction efficiency and matrix effect were determined by spiking the samples before and after the extraction (Table 1, Fig. 2). The spiking solution, an extract of cyanobacterial natural bloom, contained known amounts of MC-RR, MC-LR and MC-YR. The spiked fetal bovine serum sample extraction included a protein precipitation step with 80% acetonitrile containing 0.1% trifluoroacetic acid (TFA). The serum extract was evaporated to dryness in glass vials with argon at 40 C, and then reconstituted in water for SPE. The SPE wash step for serum samples was performed with 15% methanol containing 0.1 M acetic acid (HOAc) and the elution with 80% methanol containing 0.1 M HOAc. 2.3. Determination of total microcystins 10 mg of freeze-dried tissue or 1 mL of serum was trypsinated with 1 mL of 500 mg mL1 of trypsin in So¨rensen’s phosphate buffer (pH 7.5) at 37 C in shaking water bath for 2 h, followed by oxidation with 0.1 M KMnO4 and 0.1 M NaIO3, pH 9.0 for 3 h. When the colour of the sample solution turned from purple (colour of KMnO4) to brown (colour of the reduced product MnO2) during the oxidation procedure, more oxidation solution was added until the colour was purple again. The reaction was quenched by sodium bisulphite solution (40% w/v), until the sample was colourless. 10% sulphuric acid was added until the pH of the solution was acidic (w2). The optimised MMPB method workflow for both aquatic snail and serum samples is described in Fig. 2. The MMPB recovery during the drying process, argon at 40 C, was assessed with preparing 1 mL of MMPB solutions in 80% methanol, 0.05% HOAc with concentration of 2 ng mL1 (n ¼ 6) and 20 ng mL1 (n ¼ 6). Three samples of each concentration
were left to evaporate at 40 C, under argon until 20–30 mL solution remained, and three samples were completely dried down. The dried samples were kept at 40 C under argon for at least 5 min. All samples were then reconstituted to 200 mL with 35% methanol containing 0.05% HOAc before analysis. Adduct formation (Naþ, Kþ, NHþ 4 ) in LC–ESI-MS was tested by preparing MMPB samples (0.5 mg mL1) in 40% methanol containing 0.6 mM of either sodium chloride, potassium chloride or ammonium acetate. The adduct formation was tested with LC–MS runs in single ion recording (SIR), scan and multiple reactant monitoring (MRM) modes. The sample oxidation recovery is the measure of the proportion of MCs in a tissue recovered as the oxidation product MMPB. Some of the MCs in the tissue may escape oxidation and some of the formed MMPB may be further oxidised and lost. The oxidation recovery, extraction efficiency and the matrix effects were controlled by spiking of control samples (Fig. 2). The oxidation recovery controls were spiked with a known amount of MCs before any treatment and compared to controls spiked with the corresponding amount of MMPB standard after the oxidation procedure, but before the SPE (Fig. 2). To assess the MMPB extraction efficiency controls spiked with MMPB standard before and after SPE were compared (Fig. 2).
2.4. LC–MS analysis The LC–MS analysis was performed with an Agilent 1100 binary HPLC system (Waldbronn, Germany) coupled to a Quattro Micro (Waters, Manchester, UK) triple quadrupole mass spectrometer with electrospray ionisation (ESI) source. The column was a Purospher STAR RP-18e (Merck, Darmstadt, Germany) 30 mm 4 mm i.d. with 3 mm particles, kept at 40 C. The mobile phase consisted of solvent A: 0.1% formic acid and B: acetonitrile, and the gradient programme for MC detection was from 25% to 90% B over 8 min, held at 90% B for 1 min, flow rate 0.5 mL min1, followed by equilibration at 25% for 4 min, flow rate 1 mL min1. The injection volume was 10 mL. Some MS ion source parameters were: source temperature 150 C, desolvation temperature 300 C, cone gas flow 60 L h1, desolvation gas flow 640 L h1, capillary voltage 3 kV, positive ESI mode. The cone voltages (33 V-75 V) and collision energies (25 eV-65 eV) were optimised for each analyte. The monitored signals in the SIR mode were m/z [dmMC-RRþ2H]2þ 512.8, [MCRRþ2H]2þ 519.8, [dmMC-LRþH]þ 981.5, [MC-LRþH]þ 995.5, [MC-YRþH]þ 1045.5. In the MRM mode the corresponding transitions to m/z 135 (Adda fragment) were recorded. The mobile phase gradient for MMPB detection was from 40% to 70% B over 3 min, increase of B to 90% B over 0.1 min, which was held for 1 min, flow rate 0.5 mL min1, followed by equilibration at 40% B for 4 min, flow rate 1 mL min1. For MMPB MS analysis cone voltage 5 V and collision energies 7 eV-11 eV were used for transitions in positive MRM mode, [MMPB þ H]þ m/z 209.2, to m/z 91, 131 and 191.
950
M.-R. Neffling et al. / Environmental Pollution 158 (2010) 948–952
Fig. 2. Optimised workflow for the determination of free/extractable microcystins in freeze-dried tissue, as well as total (free þ bound) microcystins in freeze-dried tissue and in serum.
3. Results and discussion
better recoveries than 10% methanol. Serum and freeze-dried tissue extraction efficiencies are reported in Table 1.
3.1. Determination of free microcystins 3.2. Determination of total microcystins The extraction efficiency with the water:methanol:butanol solution was higher than with 75% methanol and the Oasis HLB cartridges gave better recoveries than Supelclean C-18. 20% methanol in the wash step reduced the signal suppression and thus gave
The earlier reported MMPB workflow (Ott and Carmichael, 2006; Yuan et al., 2006) was changed by including trypsination, by stronger concentration of oxidative solution and by sample
M.-R. Neffling et al. / Environmental Pollution 158 (2010) 948–952 Table 1 Extraction of the toxins, extraction efficiencies. Extraction efficiency/% Sample
Spiked with 100 mL of
Serum 1 mL
Natural bloom extract 100.2 Natural bloom extract 1:10 86.7 Natural bloom extract 1:100 83.9
MC-RR MC-LR MC-YR Average 86.1 78.9 73.2
87.2 80.9 84.0
91.2 82.2 80.4
Serum 0.1 mL Natural bloom extract 1:2 Natural bloom extract 1:20
90.8 80.8
86.1 74.0
82.7 86.3
86.5 80.3
Snail 10 mg
78.1 51.0 66.7
76.8 56.8 57.2
77.3 58.5 73.8
77.4 55.4 65.9
Natural bloom extract Natural bloom extract 1:10 Natural bloom extract 1:100
The cyanobacterial natural bloom extract contained 0.6 mg MC-RR, 0.6 mg MC-LR, 0.4 mg MC-YR. This extract or its dilutions (1:2, 1:20, 1:10, 1:100) were used to spike control samples (Fig. 2). The values given are averages from SIR and MRM measurements.
extraction as described in Fig. 2. The trypsination step was adapted from a forensic procedure (Pelander et al., 2007) and was expected to better expose the covalently bound MCs to oxidation. The addition of oxidant solution (to keep the solution purple) and solution of sodium bisulphite (to quench the reaction) instead of solid KMnO4 and solid sodium bisulphite (as reported in Ott and Carmichael, 2006) enabled more accurate proportioning of the reagents. Centrifugation after oxidation was found to be crucial for successful SPE treatment. No significant losses due to the centrifugation were observed. The sodium adduct of MMPB was observed, but the protonated molecular ion was dominant. The potassium and ammonium adducts were of signal intensity below limit of detection. Furthermore, there were no observable losses of the MMPB during the evaporation test. The SPE efficiency of MMPB was relatively good (considering the difficult matrix), 70–83% (Table 2). The cartridges thus tolerated the low pH and the rather large sample volume (up to 15 mL). The matrix effects were severe with the MMPB tissue analysis samples. The serum samples spiked at the final step of sample preparation, after oxidation and extraction, revealed that the recovered signal was only 16% (1 mL serum handled, spiked with 33 ng MMPB, corresponding to 160 ng of MCs) to 37% (0.1 mL serum handled; 16.6 ng of MMPB) as compared to the signal from the standards (Table 2). The same signal suppression phenomenon was also seen with the aquatic snail samples (Table 2.). Some problems might be overcome by switching to negative mode MS detection as described in Ott and Carmichael (2006), but in our MS system the signal intensity of the negative MMPB ions was not adequate.
Table 2 MMPB method: oxidation recovery, extraction efficiency and matrix effects. Sample Serum 1 mL 1 mL 0.1 mL 0.1 mL
Spiked with 100 mL containing MCs/ng
Oxidation recovery/%
Extraction efficiency/%
Matrix effects, Signal in spiked samples/%
1600 160 800 80
31.6 39.5 28.9 33.9
85.6 97.2 70.7 73.0
18.4 16.3 34.8 37.3
38.4 36.2 22.4
62.4 76.3 44.9
25.8 18.0 27.4
Aquatic Snail Tissue 10 mg 8800 10 mg 1600 10 mg 160
Oxidation recovery refers to comparison of samples spiked with MCs before oxidation to samples spiked with the corresponding amount of MMPB after the oxidation. Extraction efficiency refers to comparison of samples spiked with MMPB after oxidation to samples spiked after SPE. The matrix effect gives the percentage signal recovered from spiked samples as compared to recovered signal from MMPB standard in the absence of tissue. Please see Fig. 2 for spiking points.
951
Due to the high amounts of covalently bound MCs in the tissues of L. stagnalis found in an accompanying publication (Lance et al., in press) using the methods described here, it seems unlikely that the only target for covalent binding of MCs in the tissue would be the protein phosphatases. We found up to 37 mg of covalently bound toxins per gram of freeze-dried aquatic snail tissue. The proportion of covalently bound toxin, as compared to proportion of extractable toxin, is lower than found by Williams and co-workers (Williams et al., 1997a,b,c). No glutathione and cysteine conjugates involved in detoxification of MCs (Kondo et al., 1992) were detected. Assuming that all covalently bound toxins would be bound to protein phosphatases (one binding site per 35–36 kDa catalytic subunit) the amount of covalently inhibited protein phosphatase catalytic subunits in the freeze-dried tissue would be 1.3 mg g1, which is unrealistic. MC-LR functions also as an ATP synthase inhibitor (Mikhailov et al., 2003). Imanishi and Harada presented a study to discover other MC-binding proteins by affinity chromatography and proteomics methods (Imanishi and Harada, 2004). This kind of experimental set-up could be useful in revealing other possible targets for MC binding and thus contributing in the toxicosis by MCs in different tissues. 4. Conclusions The MMPB method from early 1990s is important when assessing the total MC contamination levels in different organisms affected by the cyanobacterial blooms. However, this method has been quite laborious and not employed in many publications. The modified method reported here worked with reasonable oxidation recoveries and extraction efficiencies for two different sample types, aquatic snail tissue and bovine serum. Acknowledgements The authors would like to thank the National Graduate school in Informational and Structural Biology, Medicinska understo¨dsfo¨reningen Liv och Ha¨lsa, Otto A. Malms donations fond (M-R N.), and the CIMO (Center for International Mobility) and the French and Finnish Ministries of Foreign Affairs (E.L.), Tekes decision number 40366/06 and Academy of Finland decision number 108947 (J.M.). Dr. Krister Karlsson is thanked for his helpful advice in the early stages of the study. References van Apeldoorn, M.E., van Egmond, H.P., Speijers, G.J., Bakker, G.J., 2007. Toxins of cyanobacteria. Mol. Nutr. Food Res. 51, 7–60. Codd, G.A., Morrison, L.F., Metcalf, J.S., 2005. Cyanobacterial toxins: risk management for health protection. Toxicol. Appl. Pharmacol. 203, 264–272. Harada, K., Murata, H., Qiang, Z., Suzuki, M., Kondo, F., 1996. Mass spectrometric screening method for microcystins in cyanobacteria. Toxicon 34, 701–710. Hastie, C.J., Borthwick, E.B., Morrison, L.F., Codd, G.A., Cohen, P.T.W., 2005. Inhibition of several protein phosphatases by a non-covalently interacting microcystin and a novel cyanobacterial peptide, nostocyclin. Biochim. Biophys. Acta 1726, 187–193. Honkanen, R.E., Zwiller, J., Moore, R.E., Daily, S.L., Khatra, B.S., Dukelow, M., Boynton, A.L., 1990. Characterization of microcystin-LR, a potent inhibitor of type 1 and type 2A protein phosphatases. J. Biol. Chem. 265, 19401–19404. Imanishi, S., Harada, K., 2004. Proteomics approach on microcystin binding proteins in mouse liver for investigation of microcystin toxicity. Toxicon 43, 651–659. Karlsson, K., Sipia, V., Kankaanpaa, H., Meriluoto, J., 2003. Mass spectrometric detection of nodularin and desmethylnodularin in mussels and flounders. J. Chromatogr. B 784, 243–253. Kaya, K., Sano, T., 1999. Total microcystin determination using erythro-2-methyl-3(methoxy-d(3))-4-phenylbutyric acid (MMPB-d(3)) as the internal standard. Anal. Chim. Acta 386, 107–112. Kondo, F., Ikai, Y., Oka, H., Okumura, M., Ishikawa, N., Harada, K., Matsuura, K., Murata, H., Suzuki, M., 1992. Formation, characterization, and toxicity of the glutathione and cysteine conjugates of toxic heptapeptide microcystins. Chem. Res. Toxicol. 5, 591–596. Lance, E., Neffling, M.-R, Ge´rard, C., Meriluoto, J., Bormans, M., Accumulation of free and covalently bound microcystins in tissues of Lymnaea stagnalis (Gastropoda)
952
M.-R. Neffling et al. / Environmental Pollution 158 (2010) 948–952
following toxic cyanobacteria or dissolved microcystin-LR exposure. Environ. Pollut., in press, doi:10.1016/j.envpol.2009.10.025. MacKintosh, C., Beattie, K.A., Klumpp, S., Cohen, P., Codd, G.A., 1990. Cyanobacterial microcystin-LR is a potent and specific inhibitor of protein phosphatases 1 and 2A from both mammals and higher plants. FEBS Lett. 264, 187–192. Mikhailov, A., Ha¨rma¨la¨-Braske´n, A.S., Hellman, J., Meriluoto, J., Eriksson, J.E., 2003. Identification of ATP-synthase as a novel intracellular target for microcystin-LR. Chem. Biol. Interact 142, 223–237. Ott, J.L., Carmichael, W.W., 2006. LC/ESI/MS method development for the analysis of hepatotoxic cyclic peptide microcystins in animal tissues. Toxicon 47, 734–741. Pelander, A., Backstro¨m, D., Ojanpera¨, I., 2007. Qualitative screening for basic drugs in autopsy liver samples by dual-plate overpressured layer chromatography. J. Chromatogr. B Analyt Technol. Biomed. Life Sci. 857, 337–340. Rinehart, K.L., Harada, K., Namikoshi, M., Chen, C., Harvis, C.A., Munro, M.H.G., Blunt, J.W., Mulligan, P.E., Beasley, V.R., Dahlem, A.M., Carmichael, W.W., 1988. Nodularin, microcystin, and the configuration of Adda. J. Am. Chem. Soc. 110, 8557–8558. Robinson, N.A., Matson, C.F., Pace, J.G., 1991. Association of microcystin-LR and its biotransformation product with a hepatic-cytosolic protein. J. Biochem. Toxicol. 6, 171–180. Sano, T., Nohara, K., Shiraishi, F., Kaya, K., 1992. A method for microdetermination of total microcystin content in waterblooms of cyanobacteria (Blue-Green-Algae). Int. J. Environ. Ch 49, 163–170.
Tsuji, K., Masui, H., Uemura, H., Mori, Y., Harada, K., 2001. Analysis of microcystins in sediments using MMPB method. Toxicon 39, 687–692. WHO, 1998. Guidelines for Drinking-water Quality. World Health Organization, Geneva. Williams, D.E., Craig, M., McCready, T.L., Dawe, S.C., Kent, M.L., Holmes, C.F., Andersen, R.J., 1997a. Evidence for a covalently bound form of microcystin-LR in salmon liver and dungeness crab larvae. Chem. Res. Toxicol. 10, 463–469. Williams, D.E., Dawe, S.C., Kent, M.L., Andersen, R.J., Craig, M., Holmes, C.F., 1997b. Bioaccumulation and clearance of microcystins from salt water mussels, Mytilus edulis, and in vivo evidence for covalently bound microcystins in mussel tissues. Toxicon 35, 1617–1625. Williams, D.E., Craig, M., Dawe, S.C., Kent, M.L., Andersen, R.J., Holmes, C.F., 1997c. 14C-labeled microcystin-LR administered to Atlantic salmon via intraperitoneal injection provides in vivo evidence for covalent binding of microcystin-LR in salmon livers. Toxicon 35, 985–989. Young, F.M., Metcalf, J.S., Meriluoto, J.A.O., Spoof, L., Morrison, L.F., Codd, G.A., 2006. Production of antibodies against microcystin-RR for the assessment of purified microcystins and cyanobacterial environmental samples. TOXICON 48, 295–306. Yuan, M., Carmichael, W.W., Hilborn, E.D., 2006. Microcystin analysis in human sera and liver from human fatalities in Caruaru, Brazil 1996. Toxicon 48, 627–640.