Interference by plastics additives in the HPLC determination of microcystin-lr and -YR

Transcription

1 University of New Hampshire University of New Hampshire Scholars' Repository New Hampshire Agricultural Experiment Station Research Institutes, Centers and Programs Interference by plastics additives in the HPLC determination of microcystin-lr and -YR Miyoshi Ikawa University of New Hampshire Nancy Phillips University of New Hampshire - Main Campus James F. Haney University of New Hampshire, Jim.Haney@unh.edu John J. Sasner University of New Hampshire, John.Sasner@unh.edu Follow this and additional works at: Recommended Citation Miyoshi Ikawa, Nancy Phillips, James F Haney, John J Sasner. Interference by plastics additives in the HPLC determination of microcystin-lr and -YR. Toxicon, Volume 37, Issue 6, June 1999, Pages , ISSN , S (98) This Article is brought to you for free and open access by the Research Institutes, Centers and Programs at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in New Hampshire Agricultural Experiment Station by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact nicole.hentz@unh.edu.

2 Interference by plastics additives in the HPLC determination of microcystin-lr and -YR Rights Copyright 1999 Elsevier Science Ltd. All rights reserved. This article is available at University of New Hampshire Scholars' Repository:

3 PERGAMON Toxicon 37 (1999) 923±929 Interference by plastics additives in the HPLC determination of microcystin-lr and -YR Miyoshi Ikawa *, Nancy Phillips, James F. Haney, John J. Sasner Department of Zoology and Center for Freshwater Biology, University of New Hampshire, Rudman Hall, Durham, NH 03824, USA Received 18 July 1998; accepted 15 October 1998 Abstract Plastics devices used for the eld collection of water samples may contain plastics additives which will interfere with the HPLC determination of the cyanobacterial toxins microcystins. The presence of the additives resorcinol monobenzoate or 2,4- dihydroxybenzophenone can interfere with the determination of microcystin-lr. The presence of bisphenol A in plastics can interfere with the determination of microcystin-yr. # 1999 Elsevier Science Ltd. All rights reserved. 1. Introduction Reported incidences of toxin-producing freshwater cyanobacteria have increased with alarming frequency worldwide and a rapidly expanding literature has developed on this speci c topic (Carmichael, 1992; Codd et al., 1994; Ste ensen and Nicholson, 1994; Yoo et al., 1995; Watanabe et al., 1996). The most prevalent of the recognized cyanobacterial toxins are the microcystins, which are commonly found in Microcystis, but which have also been reported in other cyanobacteria, such as Anabaena, Oscillatoria and Nostoc (Rinehart et al., 1994). The microcystins produce liver damage in endothermic vertebrates (Foxall and Sasner, 1981; Runnegar and Falconer, 1981) and promote liver tumor growth (Fujiki et al., 1996). The microcystins, rst isolated and characterized in 1982 (Botes et al., 1982), consist of a class of closely related cyclic heptapeptides and HPLC is widely used * Corresponding author. Fax: /99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S (98)

4 924 M. Ikawa et al. / Toxicon 37 (1999) 923±929 for the determination of individual microcystins of this hazardous group of compounds. With the increasing sensitivity of detection methods, the use of plastics for the collection and storage of water samples can cause interference problems in the HPLC determination of naturally occurring toxins. A problem became apparent when we were obtaining extraordinarily high values for microcystin-lr in lake water samples. This was traced to the gravity-type plastic coring device used in collecting the samples. A similar problem, but of lesser magnitude, was also encountered in the determination of microcystin-yr when a small diameter exible tubing was used as an integrated water column sampler. This paper identi es the plastics additives which were found to interfere with the HPLC determination of microcystin-lr and -YR. 2. Materials and methods 2.1. Field sampling and plastics extraction The original objective of the study was to determine microcystin concentrations in lake sediments using a gravity type messenger-activated core sampler. Liner tubes (50 5 cm ID) made of clear cellulose acetate butyrate were placed inside a brass core tube. Upon retrieval, samples of water and sediment were removed from the liner within 0.5±1 h and placed in high density polyethylene (HDPE) sample containers and placed in a freezer ( 308C). Within 1±2 weeks samples were thawed and analyzed for microcystins using HPLC. Once contamination was suspected, all parts of the coring device, as well as other sampling devices and storage containers, were tested by placing them in distilled water at 20±228C for 1 h. Extracts of the HDPE containers were made by leaving the water in the containers for several days. The extracts were concentrated on C18 cartridges and run on the HPLC as described below. The results of the extraction of various plastics components are shown in Table HPLC The HPLC method described by Harada et al. (1988) was used. The system consisted of a Hewlett-Packard Series 1050 pump operated at 1 ml/min, a HP 1050 variable wavelength detector set at 238 nm, a HP 3396A integrator and a Nucleosil C18, 5 micron, mm column. The mobile phase was methanol, 0.05 M ph 3 phosphate bu er (58:42), which gave elution times of the order of 6.5 min for microcystin-rr, 9 min for microcystin-yr and 11±12 min for microcystin-lr. For sample preparation, Waters Sep-Pak Plus C18 or Waters Sep-Pak Vac 3 cc (500 mg) C18 cartridges were used. Extract (100 ml) was acidi ed with 5 ml of glacial acetic acid and adsorbed onto the cartridge. The cartridge was washed with 20 ml of water followed by 20 ml of 10% methanol in water and the microcystins eluted with 20 ml 90% methanol:water (Tsuji et al., 1994). The eluate was taken to dryness in vacuo, the residue taken up in 1 ml of

5 M. Ikawa et al. / Toxicon 37 (1999) 923± Table 1 False concentrations of microcystins in extracts of various sampling devices and materials caused by release of plastics additives Device/part `Microcystin' (mg/l) `LR' `YR' `RR' Rigid corer liner (clear cellulose acetate butyrate) 2500 <1 <0.5 Rigid corer liner (cast acrylic) 6 1 <0.5 Rigid sampler cylinder (Aquatic Research Instruments, clear polycarbonate) <1 1 <0.5 Flexible large vertical integrated sampler tubing, 15 mm ID (Tygon) Flexible small vertical integrated sampler tubing, 4 mm ID < Flexible tubing, 5 mm ID (Tygon) <1 1.5 a <0.5 a Flexible tubing, 8 mm ID (Nalgene) <1 <1 a 0.5 a Flexible tubing, 9 mm ID (Nalgon) <1 1 a 0.5 a Flexible tubing, 5 mm ID (PVC) < Specimen container (white, HDPE) <1 <1 <0.5 Storage bottle (translucent, HDPE) <1 <1 <0.5 a Large peaks preceding but separable from `microcystin' peaks. methanol:water (60:40) and the solution syringe ltered through a 0.2 mm PTFE lter before HPLC. When 2 mg each of microcystin-rr, -YR and -LR were added to 100 ml of water and acidi ed with 5 ml of acetic acid, recoveries in the range of 80±100% were obtained using these cartridges. We found that some batches of C18 cartridges of some brands gave very poor recoveries (near 0%) of the microcystins and we strongly recommend testing each cartridge lot for recovery characteristics before use with unknown samples Chemicals Bisphenol A, resorcinol monobenzoate and 2,4-dihydroxy-benzophenone were purchased from Aldrich Chemical Co., Milwaukee, WI and the microcystin-lr, -YR and -RR from Calbiochem-Novabiochem, La Jolla, CA Mass spectrometry Mass spectra were obtained with a Hewlett-Packard Model 5988A quadrupole mass spectrometer. 3. Results and discussion 3.1. Interference of plastics and additives (Fig. 1) in the determination of microcystins Table 1 shows the results from the extracts of some plastics in the determination of microcystins. Especially high values for microcystin-lr and -YR

6 926 M. Ikawa et al. / Toxicon 37 (1999) 923±929 Fig. 1. Structures of plastics additives which can interfere with the determination of microcystin-lr and -YR. were observed in some cases. The determination of microcystin-rr did not seem to be grossly a ected here. In the case of both microcystin-yr and -RR, however, large peaks sometimes preceded the `microcystin' peaks, but were separable. The HDPE containers used to store the water samples did not contribute signi cantly to the microcystin values even when the water was in contact with the container for several days `Microcystin-LR' We consistently began to notice surprisingly high values for microcystin-lr (ca. 1000±2000 mg/l of lake water) whenever a particular rigid gravity type coring device was used for collecting samples of lake bottom sediment and benthic water samples. Intraperitoneal injection of 1 ml of lake water concentrate, estimated to contain 20 mg of `microcystin-lr', into 20 g mice did not cause death or produce any hepatotoxic symptoms (LD 50 =1 mg microcystin-lr/20 g mouse), indicating that the material was probably not microcystin-lr. Extraction of the various components of the coring device with distilled water for 1 h, revealed that the rigid clear plastic corer liner was responsible for the false reading of microcystin- LR. 100 ml of the corer liner extract was acidi ed with 5 ml of acetic acid and the material concentrated by passing it through a C18 cartridge as described above. Mass spectrometry of the extract indicated a compound with a molecular weight of 214 Da. A base peak at m/z 105 and a prominent peak at m/z 77 indicated the presence of an unsubstituted benzoyl and phenyl group. A molecular formula of C 13 H 10 O 3 corresponding to a monobenzoyl ester of a dihydroxybenzene or a dihydroxybenzophenone was indicated. Ash and Ash (1991) list resorcinol monobenzoate and 2,4-dihydroxybenzophenone as industrially used UV absorbing plastics additives. When these compounds were added to water at 2 mg/100 ml and the solutions acidi ed with acetic acid and concentrated through a C18 cartridge, they were recovered in +90% yield. Neither compound was separable from microcystin-lr on HPLC (Fig. 2A±E). However, Fig. 2(C) showed a slight

7 M. Ikawa et al. / Toxicon 37 (1999) 923± Fig. 2. HPLC comparison of plastics additives with microcystins: (A) microcystin-lr (2 mg/ml); (B) resorcinol monobenzoate (1 mg/ml); (C) resorcinol monobenzoate (B) + microcystin-lr (A) (equal volumes); (D) 2,4-dihydroxy-benzophenone (2 mg/ml); (E) 2,4-dihydroxybenzophenone (D) + microcystin-lr (A) (equal volumes); (F) microcystin-yr (1 mg/ml); (G) bisphenol A (2 mg/ml); (H) bisphenol A (G) + microcystin-yr (F) (equal volumes). The plastics additives samples had all been processed through a C18 cartridge and the nal volume adjusted to 1 ml. broadening on the trailing edge, indicating that resorcinol monobenzoate may slightly precede microcystin-lr. Resorcinol monobenzoate (1 mg) gave a HPLC response equivalent to 2 mg of microcystin-lr and 1 mg of 2,4-dihydroxybenzophenone gave a response equivalent to 0.85 mg of LR. The separation of resorcinol monobenzoate from 2,4-dihydroxybenzophenone was barely detectable, the resorcinol monobenzoate preceding the 2,4-dihydroxybenzophenone by about 12

8 928 M. Ikawa et al. / Toxicon 37 (1999) 923±929 s at a 12 min retention time. Mass spectra indicated that the plastics additive involved was resorcinol monobenzoate (m/z = 214 (M), 105 (base peak), 77) rather than 2,4- dihydroxybenzophenone (m/z = 214 (M), 137 (base peak), 213 (intense), 105, 77) `Microcystin-YR' `Microcystin-YR' was associated with a small diameter exible integrated water column sampler. A 1-h water extract of the device showed a microcystin-yr content corresponding to ca mg/l. When the extract of the device was acidi ed and concentrated through a C18 cartridge, the concentrate gave a mass spectrum with a very strong peak at m/z = 213 and signi cant peaks at m/ z = 228 and 119. The spectrum indicated bisphenol A, whose most signi cant peaks were m/z = 228 (M), 213 (base peak) and 119. Bisphenol A (2,2-bis(4- hydroxyphenyl)-propane) is used commercially to produce toughness and clarity in resins (Ash and Ash, 1991). Bisphenol A was recovered in quantitative yield when dissolved in water at 2 mg/100 ml, acidi ed and concentrated on a C18 cartridge. The substance did not separate from microcystin-yr on HPLC (Fig. 2F±H). 1 mg of bisphenol A gave a response equivalent to 0.6 mg of microcystin-yr. The HPLC method of Harada et al. (1988), using ph 3 bu ered aqueous methanol in an isocratic mode, has been a simple and useful assay system in our lake water studies for the commonly occurring more hydrophilic microcystins (-LR, -YR and -RR). The problem of interference in our determinations by additives in the plastics arose as a result of the use of plastic sampling devices. The use of HDPE storage containers appeared not to be the problem. The di culty has been surmounted by testing various components and switching to those which did not contain the interfering additives. Considering the number and pervasiveness of plastics additives in use in the world today, any system containing suspect plastics would need to be tested for interfering substances under the prevailing HPLC conditions. Acknowledgements The authors wish to thank Mr. William Dotchin of the UNH Instrumentation Center for performing the mass spectrometer analyses. This research was supported by the New Hampshire Agricultural Experiment Station Hatch Project 205. Scienti c Contribution No of the New Hampshire Agricultural Experiment Station. References Ash, M., Ash, I., Handbook of Industrial Chemical Additives. VCH Publishers, New York. Botes, D.P., Kruger, H., Viljoen, C.C., Isolation and characterization of four toxins from the blue± green alga, Microcystis aeruginosa. Toxicon 20, 945±954.

9 M. Ikawa et al. / Toxicon 37 (1999) 923± Carmichael, W.W., Status Report on Planktonic Cyanobacteria (Blue±Green Algae) and their Toxins. Tech. Report EPA No. 600(R-92) 079. US Environmental Protection Agency, Cincinnati, OH. Codd, G.A., Je eries, T.M., Keevil, C.W., Potter, E. (Eds.), Detection Methods for Cyanobacterial Toxins. The Royal Society of Chemistry, Cambridge, UK. Foxall, T.L., Sasner, J.J., Jr., E ects of a hepatic toxin from the cyanophyte Microcystis aeruginosa. In: Carmichael, W.W. (Ed.), The Water Environment: Algal Toxins and Health. Plenum, New York, pp. 365±387. Fujiki, H., Sueoka, E., Suganuma, M. (1996) Carcinogenesis of microcystins. In: Watanabe, M.F., Harada, K.-I., Carmichael, W.W., Fujiki, H. (Eds.), Toxic Microcystis. CRC Press, Boca Raton, FL, pp. 203±232. Harada, K.-I., Matsuura, K., Suzuki, M., Oka, H., Watanabe, M.F., Oishi, S., Dahlem, A.M., Beasley, V.R., Carmichael, W.W., Analysis and puri cation of toxic peptides from cyanobacteria by reversed-phase high-performance liquid chromatography. Journal of Chromatography 448, 275±283. Rinehart, K.L., Namikoshi, M., Choi, B.W., Structure and biosynthesis of toxins from blue±green algae (cyanobacteria). Journal of Applied Phycology 6, 159±176. Runnegar, M.T., Falconer, I.R., Isolation, characterization, and pathology of the toxin from the blue±green alga Microcystis aeruginosa. In: Carmichael, W.W. (Ed.), The Water Environment: Algal Toxins and Health. Plenum, New York, pp. 325±342. Ste ensen, D.A., Nicholson, B.C. (Eds.), Toxic cyanobacteria: current status of research and management. Proceedings of the International Workshop, March 22±26, Adelaide, Australia. Tsuji, K., Naito, S., Kondo, F., Watanabe, M.F., Suzuki, S., Nakazawa, H., Suzuki, M., Shimada, T., Harada, K.-I., A cleanup method for analysis of trace amounts of microcystins in lake water. Toxicon 32, 1251±1259. Watanabe, M.F., Harada, K.-I., Carmichael, W.W., Fujiki, H. (Eds.), Toxic Microcystis. CRC Press, Boca Raton, FL. Yoo, R.S., Carmichael, W.W., Hoehn, R.C., Hrudey, S.E., Cyanobacterial (Blue±Green Algal) Toxins: A Resource Guide. AWWA Research Foundation and the American Water Works Association, Denver, CO.