Matrix solid phase dispersion isolation and liquid chromatographic determination of oxytetracycline in catfish (Ictalurus punctatus) muscle tissue

A method for isolation and liquid chromatographic determination of oxytetracycline in catfish (Ictalurus punctatus) muscle tissue is presented. Blank control and oxytetracycline-fortified fish muscle tissue samples (0.5 g) were blended with octadecylsllyl (C18, 40 microns, 18% load, endcapped) derivatized silica packing material (2 g) containing 0.05 g each of oxalic acid and disodium ethylenediaminetetraacetate. A column made from the C18/fish tissue matrix was first washed with hexane (8 mL), following which the oxytetracycline was eluted with acetonitrile-methanol (1 + 1, v/v) containing 0.06% w/v each of butylated hydroxyanisole and butylated hydroxytoluene. The eluate contained oxytetracycline analyte that was free from interfering compounds when analyzed by liquid chromatography with UV detection (photodiode array set at 365 nm). Standard curves for oxytetracycline isolated from fortified samples were linear (0.998 +/- 0.002) with an average absolute percentage recovery of 80.9 +/- 6.6% for the concentration range (50, 100, 200, 400, 800, 1600, and 3200 ng/g) examined. The interassay variability was 11.3 +/- 5.2% with an intra-assay variability of 1.1%.     

Matrix solid phase dispersion (MSPD) isolation and liquid chromatographic determination of furazolidone in pork muscle tissue

A method for the isolation and liquid chromatographic (LC) determination of furazolidone in pork muscle tissue is presented. Blank or furazolidone-fortified pork muscle tissue samples (0.5 g) were blended with octadecylsilyl (C18, 18% load, endcapped, 2 g) derivatized silica. A column made from C18/pork matrix was first washed with hexane (8 mL), followed by elution of furazolidone with ethyl acetate. The ethyl acetate extract was then passed through an activated alumina column. The eluate contained furazolidone that was free from interfering compounds when analyzed by LC with UV detection (photodiode array, 365 nm). Detector response with increasing concentrations of furazolidone isolated from fortified samples was linear (r = 0.998 +/- 0.002) with an average percentage recovery of 89.5 +/- 8.1% for the concentration range (7.8-250 ng/g) examined and resulted in a minimum detectable limit of 390 pg on column, and a detector response of more than 5 times baseline noise. The inter-assay variability was 9.9 +/- 5.4% with an intra-assay variability of 1.5%.     

Matrix solid phase dispersion isolation and liquid chromatographic determination of five benzimidazole anthelmintics in fortified beef liver

A multiresidue method for isolation and liquid chromatographic determination of 5 benzimidazole anthelmintics (thiabendazole, oxfendazole, mebendazole, albendazole, and fenbendazole) in beef liver tissue is presented. Blank or benzimidazole-fortified liver samples (0.5 g) were blended with octadecylsilyl derivatized silica packing material (C18, 18% load, endcapped, 2 g). A column made from the C18/liver matrix was first washed with hexane (8 mL), following which the benzimidazoles were eluted with acetonitrile. The acetonitrile extract was then passed through an activated alumina column. The eluate contained benzimidazole analytes that were free from interfering compounds as determined by UV detection (photodiode array, 290 nm). Correlation coefficients of standard curves for individual benzimidazoles isolated from fortified samples, using internal standardization, were linear (0.996 +/- 0.002 to 0.999 +/- 0.001) with average relative percentage recoveries from 62.0 +/- 6.7 to 86.8 +/- 8.6% for the concentration range (100-3200 ng/g) examined. The interassay variability was 7.0 +/- 4.1 to 12.9 +/- 10.2% with an intra-assay variability from 2.2 to 4.0%.     

Matrix solid-phase dispersion (MSPD) isolation and liquid chromatographic determination of oxytetracycline, tetracycline, and chlortetracycline in milk

A multiresidue method for the isolation and liquid chromatographic determination of oxytetracycline (OTC), tetracycline (TC), and chlortetracycline (CTC) antibiotics in milk is presented. Blank and tetracycline (OTC, TC, and CTC) fortified milk samples (0.5 mL) were blended with octadecylsilyl (C18, 40 microns, 18% load, endcapped, 2 g) derivatized silica packing material containing 0.05 g each of oxalic acid and disodium ethylenediaminetetraacetic. A column made from the C18/milk matrix was first washed with hexane (8 mL), following which the tetracyclines were eluted with ethyl acetate-acetonitrile (1 + 3; v/v). The eluate contained tetracycline analytes that were free from interfering compounds when analyzed by liquid chromatography with UV detection (photodiode array, 365 nm). Correlation coefficients of standards curves for individual tetracycline isolated from fortified samples were linear (from 0.982 +/- 0.009 to 0.996 +/- 0.004) with average percentage recoveries from 63.5 to 93.3 for the concentration range (100, 200, 400, 800, 1600, and 3200 ng/mL) examined. The inter-assay variability ranged from 8.5 +/- 2.4% to 20.7 +/- 13.0% with an intra-assay variability of 1.0-9.3%.     

Determination of niclosamide residues in rainbow trout (Oncorhynchus mykiss) and channel catfish (Ictalurus punctatus) fillet tissue by high-performance liquid chromatography

Bayluscide [the ethanolamine salt of niclosamide (NIC)] is a registered piscicide used in combination with 3-(trifluoromethyl)-4-nitrophenol (TFM) to control sea lamprey populations in streams tributary to the Great Lakes. A high-performance liquid chromatography (HPLC) method was developed for the determination of NIC residues in muscle fillet tissues of fish exposed to NIC and TFM during sea lamprey control treatments. NIC was extracted from fortified channel catfish and rainbow trout fillet tissue with a series of acetone extractions and cleaned up on C(18) solid-phase extraction cartridges. NIC concentrations were determined by HPLC with detection at 360 and 335 nm for rainbow trout and catfish, respectively. Recovery of NIC from rainbow trout (n = 7) fortified at 0.04 microg/g was 77 +/- 6.5% and from channel catfish (n = 7) fortified at 0.02 microg/g was 113 +/- 11%. NIC detection limit was 0.0107 microg/g for rainbow trout and 0.0063 microg/g for catfish. Percent recovery of incurred radioactive residues by this method from catfish exposed to [(14)C]NIC was 89.3 +/- 4.1%. Percent recoveries of NIC from fortified storage stability tissue samples for rainbow trout (n = 3) analyzed at 5 and 7.5 month periods were 78 +/- 5.1 and 68 +/- 2.4%, respectively. Percent recoveries of NIC from fortified storage stability tissue samples for channel catfish (n = 3) analyzed at 5 and 7.5 month periods were 88 +/- 13 and 76 +/- 21%, respectively.     

Multiresidue matrix solid phase dispersion (MSPD) extraction and gas chromatographic screening of nine chlorinated pesticides in catfish (Ictalurus punctatus) muscle tissue

A multiresidue technique for extraction and gas chromatographic screening of 9 insecticide (lindane, heptachlor, aldrin, heptachlor epoxide, p,p'-DDE, dieldrin, endrin, p,p'-TDE, and p,p'-DDT) residues in catfish (Ictalurus punctatus) muscle tissue is presented. The 9 insecticides, plus dibutyl chlorendate internal standard, were fortified into catfish muscle tissue (0.5 g) and blended with 2 g C18 (octadecylsilyl derivatized silica reverse-phase material). The C18/muscle tissue matrix blend was fashioned into a column by adding the blend to a 10 mL syringe barrel containing 2 g activated Florisil. The insecticides were then eluted from the column with acetonitrile (8 mL), and a portion (2 microL) of the acetonitrile eluate was then directly analyzed by gas chromatography with electron capture detection. Unfortified blank controls were treated similarly. The resultant extracts contained pesticide analytes (31.25-500 ng/g) free of interfering compounds when analyzed. Correlation coefficients for the 9 extracted pesticide standard curves (linear regression analysis, n = 5) ranged from 0.9967 (+/- 0.0018) to 0.9999 (+/- 0.0001). Average percentage recoveries (82 +/- 4.8% to 97 +/- 3.6%, n = 25 for each insecticide), interassay (5.0 +/- 2.7% to 16.9 +/- 6.5%, n = 25 for each insecticide) and intraassay (1.8 to 4.7%, n = 5 for each insecticide) variabilities were indicative of an acceptable methodology for the analysis and screening of these residues in catfish muscle tissue.     

Development of an immunochromatographic lateral-flow test strip for rapid detection of sulfonamides in eggs and chicken muscles

A rapid immunochromatographic lateral-flow test strip was developed in the competitive reaction format for the detection of sulfonamides in eggs and chicken muscle. A monoclonal antibody against the common structure of sulfonamides was conjugated to colloidal gold particles as the detection reagent and an N-sulfanilyl-4-aminobenzoic acid (SUL)-bovine serum albumin (BSA) conjugate was immobilized to a nitrocellulose membrane as the capture reagent to prepare the test strip. With this method, it required only 15 min to accomplish the semiquantitative or quantitative detection of sulfonamides. The sensitivity to sulfonamides (sulfamonomethoxine, sulfamethoxydiazine, sulfadimethoxine, and sulfadiazine) was at least 10 ng/mL, as determined with an optical density scanner. By eye measurement, the sensitivity was 20 ng/mL for sulfamonomethoxine, sulfamethoxydiazine, and sulfadimethoxine and 40 ng/mL for sulfadiazine. On the basis of a sulfamonomethoxine standard curve, recoveries were from 89.5 to 95.6% for sulfamonomethoxine, from 89.5 to 95.1% for sulfamethoxydiazine, from 85.0 to 95.6% for sulfadimethoxine, and from 44.8 to 60.9% for sulfadiazine in egg and chicken muscle samples. A parallel analysis of 27 egg samples and 28 chicken muscle samples from the animal experiment showed that the differences between test strips and high-performance liquid chromatography (HPLC) were from 0.8 to 11.2% for egg samples and from 2.2 to 34% for chicken muscle samples for the quantitative detection, and the agreement rates between test strips and HPLC were 100%, based on the maximum allowed residue level of sulfadiazine (100 ng/g) established by the European Union and China. In conclusion, the method is rapid and accurate for the detection of sulfonamides in eggs and chicken muscles.     

Multiresidue method for isolation and liquid chromatographic determination of seven benzimidazole anthelmintics in milk

A multiresidue method for the isolation and liquid chromatographic determination of 7 benzimidazole anthelmintics (thiabendazole, oxfendazole, para-hydroxyfenbendazole, fenbendazole sulfone, mebendazole, albendazole, and fenbendazole) in milk is presented. Blank or benzimidazole-spiked milk samples (0.5 mL) were blended with octadecylsilyl (C-18, 18% load, end-capped) derivatized silica packing material. A column made from the C-18/milk matrix was first washed with hexane (8 mL), and then the benzimidazoles were eluted with methylene chloride-ethyl acetate (1 + 2, v/v; 8 mL). The eluate contained benzimidazole analytes which were free from interfering compounds as determined by UV detection (photodiode array, 290 nm). Correlation coefficients of standard curves for individual benzimidazoles isolated from spiked samples were linear (0.989 +/- 0.003 to 0.998 +/- 0.001) with recoveries ranging from 70 +/- 9% to 107 +/- 2% for the concentration range (62.5-2000 ng/mL) examined. The inter-assay variabilities ranged from 4 +/- 1% to 9 +/- 7% with intra-assay variabilities of 3-6%.     

Matrix solid phase dispersion (MSPD) extraction and gas chromatographic screening of nine chlorinated pesticides in beef fat

A multiresidue technique is presented for the extraction and quantitative gas chromatographic screening of 9 insecticides (lindane, heptachlor, aldrin, heptachlor epoxide, p,p'-DDE, dieldrin, endrin, p,p'-TDE, and p,p'-DDT) as residues in beef fat. Beef fat was fortified by adding the 9 insecticides, plus dibutyl chlorendate as internal standard, to 0.5 g portions of beef fat and blending with 2 g C18 (octadecylsilyl)-derivatized silica. The C18/fat matrix blend was fashioned into a column by adding the blend to a 10 mL syringe barrel containing 2 g activated Florisil. The insecticides were then eluted from the column with 8 mL acetonitrile, and a 2 microL portion of the acetonitrile eluate was then directly analyzed by gas chromatography with electron capture detection. Unfortified blank controls were treated similarly. The acetonitrile eluate contained all of the pesticide analytes (31.25-500 ng/g) and was free of interfering co-extractants. Correlation coefficients for the 9 extracted pesticide standard curves (linear regression analysis, n = 5) ranged from 0.9969 (+/- 0.0021) to 0.9999 (+/- 0.0001). Average relative percentage recoveries (85 +/- 3.4% to 102 +/- 5.0%, n = 25 for each insecticide), inter-assay variability (6.0 +/- 1.0% to 14.0 +/- 6.7%, n = 25 for each insecticide), and intra-assay variability (2.5-5.1% n = 5 for each insecticide) indicated that the methodology is acceptable for the extraction, determination, and screening of these residues in beef fat.     

Liquid chromatographic determination of rotenone in fish, crayfish, mussels, and sediments

An analytical procedure is described for determining residues of rotenone in fish muscle, fish offal, crayfish, freshwater mussels, and bottom sediments. Tissue samples were extracted with ethyl ether and extracts were cleaned up by gel permeation chromatography and silica gel chromatography. Sediment samples were extracted with methanol, acidified, partitioned into hexane, and cleaned up on a silica gel column. Rotenone residues were quantitated by liquid chromatography, using ultraviolet (295 nm) detection. Recoveries from sediment samples fortified with rotenone at 0.3 microgram/g were 80.8%, whereas recoveries from tissue samples fortified with 0.1 microgram/g ranged from 87.7 to 96.8%. Samples fortified with 0.3 microgram/g and stored at -10 degrees C for 6 months before analysis had recoveries ranging from 83.2 to 90.5%. Limits of detection were 0.025 microgram/g for sediments and 0.005 microgram/g for tissue samples.     

Detection of giant myofibrillar proteins connectin and nebulin in fish meat by electrophoresis in 3-5 gradient sodium dodecyl sulfate polyacrylamide slab gels

An improved method was investigated for sodium dodecyl sulfate polyacrylamide slab gel electrophoresis (SDS-PAGE) to facilitate the analysis of the giant myofibrillar proteins, connectin and nebulin, in fish meat by using jack mackerel (Trachurus japonicus) as the sample fish. It was established that separation of the alpha-connectin band from the beta-connectin band by SDS-PAGE could be achieved by using 3-5% gradient gels with glycerol to facilitate the formation of a gradient with polymerization at 35 degrees C. SDS-PAGE samples of white dorsal muscle from the jack mackerel were homogenized with a 2% SDS solution containing an inhibitor mixture (1 microg/mL of phenylmethanesulfonyl fluoride, 1 microg/mL of leupeptin, and 1 microg/mL of E-64) and heated at 50 degrees C for 20 min. Heating these samples at 100 degrees C for 2 min resulted in the disintegration of connectin but did not affect nebulin. A purified myofibril sample and a whole muscle sample showed similar changes in the overall rate of degradation of whole connectin and nebulin during the postmortem storage period, but it was clear that beta-connectin was cleaved from alpha-connectin during the preparation of myofibrils at the early stage postmortem. Storage of the SDS-PAGE samples at -85 degrees C was preferable to storage at -18 degrees C for a long period.     

Determination of albendazole and its major metabolites in the muscle tissues of Atlantic salmon,tilapia, and rainbow trout by high performance liquid chromatography with fluorometric detection

A liquid chromatographic procedure for the determination of albendazole ([5-(propylthio)-1H-benzimidazol-2yl]carbamic acid methyl ester) and its major metabolites, albendazole sulfoxide, albendazole sulfone, and albendazole-2- aminosulfone in rainbow trout, tilapia, and salmon muscle with adhering skin tissue is described. The muscle tissue samples are made alkaline with potassium carbonate and extracted with ethyl acetate. The extracts are further subjected to cleanup by utilizing a number of liquid-liquid extraction steps. After solvent evaporation, the residue is reconstituted in mobile phase and chromatographed. The chromatography is carried out on a reversed phase Luna C(18) column, using acetonitrile/methanol/buffer as a mobile phase and a fluorescence detector. The average recoveries from the fortified muscle tissue of the three fish species for albendazole (25-100 ppb), albendazole sulfoxide (15.5-62 ppb), albendazole sulfone (1-10 ppb), and albendazole-2- aminosulfone (10-100 ppb) were 94, 77, 82, and 67%, respectively. The average CV for each compound was < or =10%. The procedure was validated and then applied to the determination of albendazole and its three major metabolites in the muscle tissue of the three fish species obtained after orally dosing with albendazole.     

Extraction methods for quantitation of gentamicin residues from tissues using fluorescence polarization immunoassay

Sodium hydroxide digestion of unhomogenized kidney and skeletal muscle for 20 min at 70 degrees C was a superior method for extracting gentamicin from tissues, compared with simple homogenization, trichloroacetic acid precipitation of homogenized tissue, and sodium hydroxide digestion of homogenized tissue. Fluorescence polarization immunoassay was used to quantitate gentamicin. Sodium hydroxide digestion of unhomogenized tissue allowed for the recovery of 90.0 +/- 5.9% (means +/- SD) from renal cortex and 79.9 +/- 3.5% from skeletal muscle. The limit of sensitivity was 17.4 ng/g kidney tissue, 15.8 ng/g digested muscle, and 39.0 ng/g digested heart. The within-assay coefficient of variation (CV) at 100 ng/g kidney was 9.2%; at 500 ng/g kidney, the CV was 2.5%; and at 2000 ng/g kidney, the CV was 1.5%. The between-assay coefficient of variation was less than 7.5% for all concentrations from kidney, and the 99% confidence interval at 100 ng/g kidney was 71.7-112.4 ng gentamicin/g kidney. The within-assay coefficient of variation (CV) at 100 ng/g muscle was 15%; at 500 ng/g muscle, the CV was 2.6%; and at 2000 ng/g muscle, the CV was 2.3%. The between-assay coefficient of variation was less than 15% for all concentrations from muscle, and the 99% confidence interval at 100 ng/g muscle was 72.5-136.8 ng gentamicin/g muscle. Gentamicin-free milk could be distinguished from milk containing gentamicin concentrations of 10 ng/mL milk with 95% confidence, and from milk containing concentrations of 30 ng gentamicin/mL milk with 99% confidence. Quantitative results at or below the tolerance level can be obtained within 90 min of sample acquisition using these extraction and assay methods.     

Liquid chromatographic determination of three sulfonamides in animal tissue and egg

Sulfonamides are widely used as a feed additive in animal production in Japan. The present paper is a determination of 3 sulfonamides: sulfamethazine (SMZ), sulfamonomethoxine [SMX, 4-amino-N-(3-methoxypyrazinyl)-benzenesulfonamide], and sulfadimethoxine (SDX) in animal tissue and egg by liquid chromatography (LC). Tissues were extracted with acetonitrile and fat was removed by liquid/liquid partition. The sulfonamides were purified by an ODS cartridge column; then each compound was separated by an ODS LC column and detected at 268 nm. Quantification levels were 0.02 ppm for SMZ and SMX, and 0.04 ppm for SDX; detection limits were 0.01 ppm for SMZ and SMX, and 0.02 ppm for SDX. Calibration curves were linear between 2 and 40 ng for SMZ and SMX, and between 4 and 80 ng for SDX. Recoveries from muscle and egg samples spiked with 1-2 micrograms/10 g were 81-98%.     

Simultaneous liquid chromatographic determination of residual synthetic antibacterials in cultured fish

A liquid chromatographic method is described to determine simultaneously the following 11 synthetic antibacterial agents used in a fishery: nitrofuran derivatives furazolidone, nifurpirinol, difurazone, and furamizole; sulfa drugs sulfamerazine, sulfisozole, sulfamonomethoxine, and sulfadimethoxine; and, oxolinic, nalidixic, and piromidic acids. A Nucleosil C18 column was used with tetrahydrofuran-acetonitrile-phosphoric acid-water (29 + 1 + 0.06 + 69.94) as the mobile phase. Pretreatment of the fish meat sample with acetone extraction and alumina column cleanup gave good separation of the LC peaks without interference from any other components. Recovery of the antibacterial agents was ca 80%. The lower limit of detection of the drugs was 1-2 ng for 10 microL injection.     

Residues of the lampricides 3-trifluoromethyl-4-nitrophenol and niclosamide in muscle tissue of rainbow trout

Rainbow trout (Oncorhyncus mykiss) were exposed to the (14)C-labeled lampricide 3-trifluoromethyl-4-nitrophenol (TFM) (2.1 mg/L) or niclosamide (0.055 mg/L) in an aerated static water bath for 24 h. Fish were sacrificed immediately after exposure. Subsamples of skin-on muscle tissue were analyzed for residues of the lampricides. The primary residues in muscle tissue from fish exposed to TFM were parent TFM (1.08 +/- 0.82 nmol/g) and TFM-glucuronide (0.44 +/- 0.24 nmol/g). Muscle tissue from fish exposed to niclosamide contained niclosamide (1.42 +/- 0.51 nmol/g), niclosamide-glucuronide (0.0644 +/- 0.0276 nmol/g), and a metabolite not previously reported, niclosamide sulfate ester (1.12 +/- 0.33 nmol/g).     

Determination of anthocyanidins in berries and red wine by high-performance liquid chromatography

A high-performance liquid chromatographic (HPLC) method for the determination of anthocyanidins from berries and red wine is described. Delphinidin, cyanidin, petunidin, pelargonidin, peonidin, and malvidin contents of bilberry (Vaccinium myrtillus), black currant (Ribes nigrum), strawberry (Fragaria ananassa cv. Jonsok), and a Cabernet sauvignon (Vitis vinifera) red wine were determined. The aglycon forms of the anthocyanins present in the samples were revealed by acid hydrolysis. A reversed phase analytical column was employed to separate the anthocyanidins before identification by diode array detection. The suitability of the method was tested by determining the recovery (95-102% as aglycons and 69-104% from glycosides) for each anthocyanidin. Method repeatability was tested by charting the total aglycon content of two samples over a period of 14 analyses and determining the coefficients of variation (1.41% for bilberry and 2.56% for in-house reference material). The method developed proved thus to be effective for reliable determination of anthocyanidins from freeze-dried berry samples and red wine. The total anthocyanidin content of the tested samples was as follows: in-house reference material, 447 +/- 8 mg/100 g; strawberry, 23.8 +/- 0.4 mg/100 g; black currant, 135 +/- 3 mg/100 g; bilberry, 360 +/- 3 mg/100 g; and Cabernet sauvignon red wine, 26.1 +/- 0.1 mg/100 mL.     

Liquid chromatographic determination of propranolol hydrochloride in various dosage forms

A simple and rapid liquid chromatographic method is described for the determination of propranolol hydrochloride in pharmaceutical preparations. The separation was achieved on a reverse-phase octylsilane (C8) column by using a mobile phase composed of a mixture of 0.5 g dodecyl sodium sulfate in 18 mL (0.15 M) H3PO4 plus 90 mL methanol, 90 mL acetonitrile, and 52 mL water. Detector response was linear for 0.03-3.1 mg/mL of propranolol. Recoveries from synthetic mixtures ranged from 99.6 to 101.7%. The results obtained by the proposed method were similar to those obtained by the USP XXI method.     

Optimizing angiotensin I-converting enzyme inhibitory activity of Pacific hake (Merluccius productus) fillet hydrolysate using response surface methodology and ultrafiltration

The in vitro angiotensin I-converting enyzme (ACE) inhibitory activity of Pacific hake hydrolysates was investigated as a function of hydrolysis conditions, starting material variability, and ultrafiltration. Hake fillets were hydrolyzed using Protamex protease under various conditions of pH, hydrolysis time, and enzyme-to-substrate ratio (% E/S) according to a response surface methodology (RSM) central composite design. The hydrolysate produced at pH 6.5, 125 min, and 3.0% E/S had an IC 50 of 165 +/- 9 microg of total solids/mL. ACE-inhibitory activity was not significantly different (P < 0.05) for hydrolysates produced using higher time-enzyme combinations within the model or from fish of different catches. Ultrafiltration (10 kDa molecular mass cutoff) resulted in an IC50 value of 44 +/- 7 microg of peptides/mL, 2.5 times more potent than the commercial product PeptACE Peptides (IC50 = 114 +/- 8 microg of peptides/mL). These results suggest that hydrolysates prepared with minimal fractionation from Pacific hake, an undervalued fish, may be a commercially competitive source of ACE-inhibitory peptides.     

Simple assay for analyzing five microcystins and nodularin in fish muscle tissue: hot water extraction followed by liquid chromatography-tandem mass spectrometry

A simple, specific, and sensitive procedure for determining six cyanotoxins, that is, microcystins RR, LR, YR, LA, and LW and nodularin, in fish muscle tissue is presented. This method is based on the matrix solid-phase dispersion technique with heated water as extractant followed by liquid chromatography (LC)-tandem mass spectrometry (MS) equipped with an electrospray ion source. Target compounds were extracted from tissue by 4 mL of water acidified to pH 2 and heated at 80 degrees C. After acidification and filtration, 0.2 mL of the aqueous extract was injected in the LC column. MS data acquisition was performed in the multireaction monitoring mode, with at least two precursor ion > product ion transitions selected for each target compound. Analyte recovery ranged between 61 and 82% and was not substantially affected by either the analyte concentrations or the type of fish. The nonexcellent recovery of some of the microcystins was traced to binding of these compounds to protein phosphatases in fish tissue occurring during sample treatment. The existence of covalently bound microcystins in fish has been evidenced by several studies. Compared to an older sample preparation procedure, this one extracted larger amounts of the analytes in a simpler and much more rapid way. On the basis of a signal-to-noise ratio of 10, limits of quantification were estimated to range between 1.6 and 4.0 ng/g. The effects of temperature and volume of the extractant on the analyte recovery were studied.