Determination of testosterone propionate in human plasma by gas chromatography-mass spectrometry

A specific, sensitive and accurate quantitative analysis of testosterone propionate in human plasma was developed using gas chromatography-mass spectrometry-selected-ion monitoring. For the calculation of testosterone propionate in plasma, peak height ratios were measured by selected-ion monitoring performed on the molecular ions of the trifluoroacetyl derivative of testosterone propionate (m/z 440) and testosterone propionate-19,19,19-d3 (m/z 443). The sensitivity of the method was judged from the lower limit of the detection of the mass spectrometer which was at 20 pg. The inter-assay coefficients of variation and relative error at a concentration of 1.31 ng/ml of plasma were 5.47% and -2.3%, respectively. The method described was applied to the determination of plasma concentrations of testosterone propionate-19,19,19-d3 following an intramuscular dose of testosterone propionate-19,19,19-d3 in a healthy male volunteer.     

Electrospray collision-induced dissociation of testosterone and testosterone hydroxy analogs

Complications with the gas chromatographic analysis of steroids prompted the use of alternative techniques for their identification. High-performance liquid chromatography/mass spectrometry with atmospheric pressure ionization allowed the collection of data for structural identification of these compounds. The objective of this study was to investigate the up-front collision-induced dissociation (UFCID) electrospray ionization (ESI) mass spectra of testosterone and monohydroxylated testosterones. The positive ion UFCID ESI mass spectrum of testosterone showed three significant ions at m/z 97, 109 and 123. The relative abundance of these ions in the UFCID ESI mass spectra of monohydroxylated testosterones varied with the position of the hydroxy group. Statistical data allowed the prediction of hydroxy group position on testosterone by evaluation of the relative abundance of the m/z 97, 109, 121 and 123 ions. Data from the ESI mass spectral analysis of testosterone in a deuterated solvent and from the analysis of cholestenone and 4-androstene-3 beta, 17 beta-diol indicated that the initial ionization of testosterone occurred at the 3-one position. CID parent ion monitoring analyses of the m/z 97, 109 and 123 ions indicated that each resulted from different fragmentation mechanisms and originated directly from the [M + H]+ parent ion. The elemental composition of these fragment ions is proposed based on evidence gathered from the CID analysis of the pseudo-molecular ions of [1,2-2H2]-, [2,2,4,6,6-2H5]-, [6,7-2H2]-, [7-2H]-, [19,19,19-2H3]- and [3,4-13C2]testosterone. The structure and a possible mechanism of formation of the m/z 109 and 123 ions is presented. The results of this study advance the understanding of the mechanisms of collision-induced fragmentation of ions.     

Endogenous and exogenous testosterone levels after administration of deuterium-labelled testosterone propionate in hypogonadotropic hypogonadism

A gas chromatography-mass spectrometry-selected ion monitoring technique was employed to determine simultaneously the plasma concentrations of endogenous and exogenous testosterone in three patients with hypogonadotropic hypogonadism after a single i.m. dose of testosterone propionate-19,19,19-d3. The plasma levels of testosterone-19,19,19-d3 derived from testosterone propionate-19,19,19-d3 were maintained above the normal testosterone levels (greater than 4 ng/ml) for 48 h, while the plasma levels of endogenous testosterone changed little.     

Analysis of anabolic steroids by partial filling micellar electrokinetic capillary chromatography and electrospray mass spectrometry

A partial filling micellar electrokinetic capillary chromatography (PF-MEKC) separation of six anabolic androgenic steroids (androstenedione, metandienone, fluoxymesterone, methyltestosterone, 17-epimetandienone and testosterone) is introduced. The method utilises a mixed micellar solution consisting of sodium dodecyl sulphate (SDS) and sodium taurocholate. The analytes are detected with a photodiode array detector at 247 nm wavelength. Methyltestosterone is used as internal standard. The detection limits were 39 microg/L for androstenedione, 40 microg/L for testosterone, 45 microg/L for fluoxymesterone, 45-90 microg/L for 17-epimetandienone, 59 microg/L for methyltestosterone and 90 microg/L for metandienone. Linear correlation between concentration (0.1-5.0 mg/L) and detector response was obtained with r2 of 0.994 for fluoxymesterone, 0.998 for 17-epimetandienone and 0.999 for androstenedione, metandienone and testosterone. In addition, ionisation of the investigated compounds in electrospray mass spectrometry (ESI-MS) was studied in positive ion mode. The most intense signal (100%) was the protonated molecular ion [M + H]+, except for 17-epimetandienone, which gave its strongest signal at m/z corresponding to [M - H2O + H]+. Finally, separation and identification of fluoxymesterone, androstenedione and testosterone by PF-MEKC-ESI-MS is described. This is the first use of PF-MEKC and PF-MEKC-ESI-MS assays for anabolic androgenic steroids.     

Determination of plasma testosterone by mass fragmentography using testosterone-19-d3 as an internal standard. Comparison with radioimmunoassay

Analytical procedures for the measurement of testosterone by mass fragmentography (MF) using trideuterated testosterone (testosterone-19,19,19-d3) are described. For the calculation of plasma testosterone, peak height ratios were measured by MF performed on the molecular ions of the TFA derivative of testosterone (m/e 480) and testosterone-19,19,19-d3 (m/e 483). The sensitivity of the method was judged from the lower limit of detection of the mass spectrometer which was at 10 pg. For the measurement of the precision, the inter- and intra-assay coefficients of variation (C.V.) were calculated by using a pooled plasma sample; they were 3.15% and 1.79%, respectively. The specificity was investigated by the use of 5 alpha-dihydrotestosterone and the MF method was found to afford a highly selective technique. These results obtained by MF have been compared with the results obtained by a radioimmunoassay method.     

Simultaneous determination of cortisol and cortisone in human plasma by stable-isotope dilution mass spectrometry.

A method for the simultaneous determination of cortisol and cortisone in human plasma was developed using capillary gas chromatography-mass spectrometry-selected ion monitoring. [2H5]Cortisol and [2H5]cortisone were used as internal standards. Cortisol and cortisone in plasma were determined from the peak-height ratios of the [M-31] fragment ions of the methoxime-trimethylsilyl derivatives of cortisol and [2H5]cortisol (m/z 605 and 610) and of cortisone and [2H5]cortisone (m/z 531 and 536). Sensitivity, specificity, precision, accuracy and reproducibility of the method were demonstrated to be satisfactory for measuring the circulating concentrations of cortisol and cortisone.     

Simultaneous determination of testosterone and androstadienone (sex attractant) in human plasma by gas chromatography-mass spectrometry with high-resolution selected-ion monitoring

Androsta-4,16-dien-3-one (androstadienone) and androst-4-en-3-one-17 beta-ol (testosterone) in healthy human plasma were simultaneously determined under several experimental conditions by gas chromatography-mass spectrometry with high-resolution selected-ion monitoring. Internal standards were [2,2,4,6,6-2H5]androstadienone and [2,2,4,6,6-2H5]testosterone. Samples were extracted with an Extrelut column, purified using Lipidex 5000 and converted into hydroxime-trimethylsilyl derivatives for determination. Physiological concentrations of androstadienone and testosterone found in eleven healthy men were 2.05 +/- 0.74 and 18.6 +/- 4.9 pmol/ml in plasma (mean +/- S.D.), respectively. No correlation was observed between these steroid concentrations.     

Quantitative determination of tertatolol in biological fluids by gas chromatography-mass spectrometry

Quantitative determination of tertatolol concentrations in plasma and urine was performed by gas chromatography-mass spectrometry in the chemical-ionization mode with ammonia after successive extractions of the beta-blocking drug in alkaline, acid and final alkaline medium. [2H9]Tertatolol, isotopically stable under the operating conditions employed, was used as an internal standard, thus allowing quantities of 1 ng/ml to be specifically determined. Overall analytical error was less than 10%. Prior to isothermal chromatography at 240 degrees C on a column packed with 3% SE-30, both compounds were silylated with bis(trimethylsilyl)trifluoroacetamide. Detection was performed by monitoring the quasimolecular ions of tertatolol, m/z 368 and m/z 377, for the [2H9]tertatolol in the chemical-ionization mode with ammonia. The calibration curves obtained had linear characteristics for the concentration range 1-1125 ng/ml.     

Liquid chromatography with electrospray ion-trap mass spectrometry for the determination of yessotoxins in shellfish

Yessotoxins are a group of large polyether toxins, produced by marine dinoflagellates, which cause widespread contamination of filter-feeding shellfish. A new, sensitive liquid chromatography-mass spectrometry (LC-MS) method has been developed for the determination of yessotoxin (YTX) and 45-hydroxy-yessotoxin (45-OHYTX), a major metabolite in shellfish. The LC system was coupled, via an electrospray ionisation (ESI) source, to an ion-trap MS in negative mode. The molecular related ion species at m/z 1141 [M-2Na+H]- was used as the parent ion for multiple MS experiments. MS-MS and MS3 gave major fragment ions at m/z 1061 [1141-SO3H]- and m/z 945 [1061-C9H12O]-. Predominant ions, that are due to the fragmentation of the backbone structure of YTXs, were observed at the MS4 stage. Reversed-phase LC using a C16 amide column was preferable to C18 phases for the separation of YTX and 45-OHYTX. Optimum calibration and reproducibility data were obtained for YTX using LC-MS-MS; r 2=0.9960, RSD < or = 6.3% at 0.25 microg YTX/g (n=5). The detection limit (S/N=3) was 30 pg YTX on-column which corresponded to 3 ng/g shellfish tissue.     

Thermochemistry of N-N containing ions: a threshold photoelectron photoion coincidence spectroscopy study of ionized methyl- and tetramethylhydrazine

The unimolecular dissociation reactions of the methylhydrazine (MH) and tetramethylhydrazine (TMH) radical cations have been investigated using tandem mass spectrometry and threshold photoelectron photoion coincidence spectroscopy in the photon energy ranges 9.60-31.95 eV (for the MH ion) and 7.74-29.94 eV (for the TMH ion). Methylhydrazine ions (CH3NHNH2(+*)) have three low-energy dissociation channels: hydrogen atom loss to form CH2NHNH2(+) (m/z 45), loss of a methyl radical to form NHNH2(+) (m/z 31), and loss of methane to form the fragment ion m/z 30, N2H2(+*). Tetramethylhydrazine ions only exhibit two dissociation reactions near threshold: that of methyl radical loss to form (CH3)2NNCH3(+) (m/z 73) and of methane loss to form the fragment ion m/z 72 with the empirical formula C3H8N2(+*). The experimental breakdown curves were modeled with Rice-Ramsperger-Kassel-Marcus theory, and it was found that, particularly for methyl radical loss, variational transition state theory was needed to obtain satisfactory fits to the data. The 0 K enthalpies of formation (delta(f)H0) for all fragment ions (m/z 73, m/z 72, m/z 45, m/z 31, and m/z 30) have been determined from the 0 K activation energies (E0) obtained from the fitting procedure: delta(f)H0[(CH3)2NNCH3(+)] = 833 +/- 5 kJ mol(-1), delta(f)H0 [C3H8N2(+*)] = 1064 +/- 5 kJ mol(-1), delta(f)H0[CH2NHNH2(+)] = 862 +/- 5 kJ mol(-1), delta(f)H0[NHNH2(+)] = 959 +/- 5 kJ mol(-1), and delta(f)H0[N2H2(+*)] = 1155 +/- 5 kJ mol(-1). The breakdown curves have been measured from threshold up to h nu approximately 32 eV for both hydrazine ions. As the photon energy increases, other dissociation products are observed and their appearance energies are reported.     

Forensic confirmatory analysis of ethyl sulfate—A new marker for alcohol consumption—by liquid-chromatography/electrospray ionization/tandem mass spectrometry

Ethyl sulfate (EtS)--a new direct marker for ethanol intake besides ethyl glucuronide (EtG) and others--was detected in urine samples by electrospray ionization tandem mass-spectrometry (LC-ESI-MS/MS). Ethyl sulfate sodium salt was used for method development, yielding a precursor [M - H]- m/z 125 and product ions m/z 97 [HSO4]- and m/z 80 [SO3]-. Pentadeuterated EtS (D5-EtS) was synthesized by esterification of sulfuric acid with anhydrous hexadeutero ethanol ([M - H]- m/z 130, product ions m/z 98 [DSO4]- and m/z 80 [SO3]-). After addition of D5-EtS and D5-EtG, urine samples were analyzed by direct injection into the gradient LC-MS/MS system. Analysis was performed in accordance with forensic guidelines for confirmatory analysis using one precursor and two product ions. EtS has been detected (in addition to EtG) in the urine samples of nine volunteers after drinking sparkling wine containing between 9 and 49 g of ethanol. Both EtS and EtG could be detected up to 36 h after consumption of alcohol. The excretion profile was found to be similar to that of EtG. No EtS was found in teetotalers' urine samples. Method validation parameters are presented. EtS was stable in urine upon storage up to twenty days at room temperature. In addition to EtG, EtS can be used to detect recent alcohol consumption, thus providing a second marker for the time range of up to approximately one day after elimination of ethanol from urine samples. The determination of EtS can be used in addition to EtG as proof of ethanol consumption in workplace monitoring programs.     

Determination of 20 alpha-hydroxy-9 beta,10 alpha-pregna-4,6-dien-3-one in plasma by selected ion monitoring

A selected ion monitoring (SIM) method has been devised for the determination of metabolites of dydrogesterone, 20 alpha-hydroxy-9 beta,10 alpha-pregna-4,6-dien-3-one (DHD) and DHD glucuronide, in plasma. Using testosterone as an internal standard (IS), DHD and IS were extracted with n-hexane and were purified by means of magnesium oxide column chromatography. The purified DHD and IS were converted to their diheptafluorobutyryl derivatives (DHD diHFB and testosterone diHFB) with heptafluorobutyric anhydride in acetone for analysis by SIM. SIM was carried out with a 2% OV-17 column (1 m) at 230 degrees C by monitoring the molecular ions of the derivatives (m/z 706 for DHD diHFB, m/z 680 for testosterone diHFB). DHD was determined from a calibration curve using a peak area method. The determination limit of the devised method was about 5 ng DHD per ml of plasma and the reproducibility was within +/- 6% of the coefficient of variation for 30 ng of DHD per ml of plasma or above.     

Quantitative determination of sotolon, maltol and free furaneol in wine by solid-phase extraction and gas chromatography-ion-trap mass spectrometry

A method for the analytical determination of sotolon [4,5-dimethyl-3-hydroxy-2(5H)-furanone], maltol [3-hydroxy-2-methyl-4H-pyran-4-one] and free furaneol [2,5-dimethyl-4-hydroxy-3(2H)-furanone] in wine has been developed. The analytes are extracted from 50 ml of wine in a solid-phase extraction cartridge filled with 800 mg of LiChrolut EN resins. Interferences are removed with 15 ml of a pentane-dichloromethane (20:1) solution, and analytes are recovered with 6 ml of dichloromethane. The extract is concentrated up to 0.1 ml and analyzed by GC-ion trap MS. Maltol and sotolon were determined by selected ion storage of ions in the m/z ranges 120-153 and 79-95, using the ions m/z 126 and 83 for quantitation, respectively. Furaneol was determined by non-resonant fragmentation of the m/z 128 mother ion and subsequent analysis of the m/z 81 ion. The detection limits of the method are in all cases between 0.5 and 1 microg l(-1), well below the olfactory thresholds of the compounds. The precision of the method is in the 4-5% range for levels in wine around 20 microg l(-1). Linearity holds at least up to 400 microg l(-1), and is satisfactory in all cases. The recoveries of maltol and sotolon are constant (70 and 64%, respectively) and do not depend on the type of wine. On the contrary, in the case of furaneol, red wines show constant and high recoveries (97%), while the recoveries on white wines range between 30 and 80%. Different experiments showed that this behavior is probably due to the existence of complexes formed between furaneol and sulphur dioxide or catechols. Sensory experiments confirmed that the complexed forms found in white wines are not perceived by orthonasal olfaction, and that the furaneol determined by the method can be considered as the free and odor-active fraction.     

Measurement of delta 1-tetrahydrocannabinol in plasma to the low picogram range by gas chromatography-mass spectrometry using metastable ion detection

A method for the assay of delta 1-tetrahydrocannabinol (delta 1-THC) in plasma using combined gas chromatography-mass spectrometry with metastable ion monitoring is described. delta 1-THC was extracted with hexane and the extracts were methylated with diazomethane to shift the peaks produced by endogenous plasma constituents away from the cannabinoid region. The delta 1-THC was then converted into its trimethylsilyl derivative and quantitated using the metastable ion at m/z 371 formed in the M+ leads to [M - CH3]+ transition with [1",1",2",2"-2H4]cannabinol as the internal standard. delta 1-THC could be measured to 5 pg/ml in plasma. This assay is 20-100 times more sensitive than existing assays and has the advantage of not needing the usual extensive purification step.     

Unusual collision-induced dissociation of fluorated and non-fluorated alpha-nitrotoluene analogs in a gas chromatograph triple-stage quadrupole mass spectrometer under electron-capturing negative-ion chemical ionization conditions

Unusual collision-induced dissociation (CID) of perfluorated and non-perfluorated alpha-nitrotoluene analogs in a gas chromatograph triple-stage quadrupole (TSQ) mass spectrometer (GC-QqQ-MS) under electron-capturing negative-ion chemical ionization conditions is reported. CID of [M - 1]- of alpha-nitro-2,3,4,5,6-pentafluorotoluene (C6F5CH2-NO2) and alpha-nitro-2,5-difluorotoluene (C6H3F2CH2-NO2) produced an intense ion with m/z 66. By using 15N- or 18O-labelled C6F5CH2-NO2 analogs, we found that this anion has the formula C3NO. By contrast, CID of [M - 1]- of alpha-nitrotoluene (C6H5CH2-NO2) and alpha-nitro-3,5-difluorotoluene (C6H3F2CH2-NO2) produced an anion with m/z 86 with the formula C3H4NO2. The expected CID of the C-N-bond of all alpha-nitrotoluene analogs to form the nitrite anion (NO2-, m/z 46) did not occur. We propose mechanisms for the formation of the anions C3NO and C3H4NO2 in the collision chamber of the TSQ mass spectrometer. The most likely structures for the anion C3NO are :C=C=C=N--O and N triple bond C-C triple bond C--O-. The unique CID behavior of C6F5CH2--NO2 can be utilized to unequivocally identify and accurately quantify nitrite in biological fluids by GC-tandem MS.     

A rapid method to determine plasma homocysteine concentration and enrichment by gas chromatography/mass spectrometry

Homocysteine is an independent risk factor for cardio- and/or cerebrovascular diseases. Many methods are used to measure plasma homocysteine levels in physiological fluids. Current gas chromatographic/mass spectrometric (GC/MS) methods allow determination not only of plasma homocysteine concentration, but also of its turnover. However, they have some methodological limitations due to the reduction of disulfide bonds between homocysteine and other thiols or proteins often requiring the use of several very toxic compounds or multi-step procedures that are particularly time-consuming, and/or utilize expensive instruments. Herein is described a rapid and precise GS/MS method to determine homocysteine turnover from a relatively low volume of plasma (200 microL). First disulfide bonds were reduced by 2-mercaptoethanol, which allows the maintenance of the reduced status preventing the rebuilding of the disulfide bond. Then the sample was derivatized to form the bis-tert-butyldimethylsilyl derivative. A deuterated internal standard, DL-[3,3,3',3',4,4,4',4'-2H8]-homocystine, was employed to account for losses associated with each analytical step. To evaluate the 'in vivo' homocysteine metabolic turnover, [1-13C]-methionine was infused and the derived [1-13C]-homocysteine quantitated. So a standard curve of [1-13C]-homocysteine was prepared by the decomposition of the [1-13C] methionine. The ions at m/z 325 and 326 were monitored, corresponding to the unlabeled [12C]-homocysteine and to labeled [13C]-homocysteine, respectively. The ion at m/z 325 ([M-114)]+) probably resulted from the loss of one derivatizing group to regenerate a free amino group. The intra-assay coefficient of variation (CV-intra%) was consistently less than 1.06%, the inter-assay (CV-inter%) less than 1.05%. The method described here seems to be simpler, more rapid, and less toxic than those published so far. In particular, its main strength appears to be the degree of precision obtained. We suggest applying this method to the measurement of the 'in vivo' rate of production of homocysteine (by the plasma 13C-homocysteine enrichment) from its precursor (13C-methionine).     

Simultaneous quantification of a non-nucleoside reverse transcriptase inhibitor efavirenz, a nucleoside reverse transcriptase inhibitor emtricitabine and a nucleotide reverse transcriptase inhibitor tenofovir in plasma by liquid chromatography positive ion electrospray tandem mass spectrometry

A high-performance liquid chromatography/positive ion electrospray tandem mass spectrometry method for the simultaneous quantification of efavirenz, emtricitabine and tenofovir was developed and validated with 100 microL human plasma. Following solid-phase extraction, the analytes were separated using a gradient mobile phase on a reverse-phase column and analyzed by MS/MS in the multiple reaction monitoring mode using the respective [M + H]+ ions, m/z 316 to 168 for efavirenz, m/z 248-130 for emtricitabine and m/z 288-176 for tenofovir, m/z 482-258 for rosuvastatin (IS), m/z 260-116 for propranolol (IS). The method exhibited a 100-fold linear dynamic range for all the three analytes in human plasma (20-2000, 2-200 and 20-2000 ng/mL for efavirenz, emtricitabine and tenofovir respectively). The lower limit of quantification was 2 ng/mL for emtricitabine and 20 ng/mL for both efavirenz and tenofovir with a relative standard deviation of less than 11%. Acceptable precision and accuracy were obtained for concentrations over the standard curve range. The total chromatographic run time of 4 min for each sample made it possible to analyze more than 250 human plasma samples per day. The method is precise and sensitive enough for its intended purpose. The method is also successfully applied to quantify efavirenz, emtricitabine and tenofovir concentrations in a rodent pharmacokinetic study.     

Trimethylsilyl-O-methyloxime derivatives for the measurement of [6,6-2H2]-D-glucose-enriched plasma samples by gas chromatography-mass spectrometry.

A new method for the determination of the enrichment of [6,6-2H2]-D-glucose in human plasma by gas chromatography-mass spectrometry (GC-MS) is described. (2,3,4,5,6)-Pentakis-O-trimethylsilyl-O-methyloxime-D-glucose is used as a derivative for the GC measurement. Using GC-MS with electron-impact ionization, the enrichment is measured in the single-ion monitoring mode observing the masses m/z 319 and 321. In contrast to other methods the use of this glucose derivative reduced the amount of plasma needed from 200 to 10 microliters and no chemical ionization equipment is needed for the mass spectrometer.     

Electrospray ionization mass spectrometry monitoring of indigo carmine degradation by advanced oxidative processes

The degradation of the dye indigo carmine in aqueous solution induced by two oxidative processes (H(2)O(2)/iodide and O(3)) was investigated. The reactions were monitored by electrospray ionization mass spectrometry in the negative ion mode, ESI(-)-MS, and the intermediates and oxidation products characterized by ESI(-)-MS/MS. Both oxidative systems showed to be highly efficient in removing the color of the dye aqueous solutions. In the ESI(-)-MS of the indigo carmine solution treated with H(2)O(2) and H(2)O(2)/iodide, the presence of the ions of m/z 210 (indigo carmine in its anionic form, 1), 216, 226, 235, and 244 was noticeable. The anion of m/z 235 was proposed to be the unprecedented hydroperoxide intermediate 2 formed in solution via an electrophilic attack by hydroxyl and hydroperoxyl radicals of the exocyclic C=C bond of 1. This intermediate was suggested to be rapidly converted into the anionic forms of 2,3-dioxo-1H-indole-5-sulfonic acid (3, m/z 226), 2-amino-alpha-oxo-5-sulfo-benzeneacetic acid (4, m/z 244), and 2-amino-5-sulfo-benzoic acid (5, m/z 216). In the ESI(-)-MS of the indigo carmine solution treated with O(3), two main anions were detected: m/z 216 (5) and 244 (4). Both products were proposed to be produced via an unstable ozonide intermediate. Other anions in this ESI(-) mass spectrum were attributed to be [4 - H + Na](-) of m/z 266, [4 - H](2-) of m/z 121.5, and [5 - H](2-) of m/z 107.5. ESI-MS/MS data were consistent with the proposed structures for the anionic products 2-5.     

Liquid chromatography with electrospray ion-trap mass spectrometry for the determination of anatoxins in cyanobacteria and drinking water

Anatoxin-a (AN) and homoanatoxin-a (HMAN) are potent neurotoxins produced by a number of cyanobacterial species. A new, sensitive liquid chromatography/multiple tandem mass spectrometry (LC/MS(n)) method has been developed for the determination of these neurotoxins. The LC system was coupled, via an electrospray ionisation (ESI) source, to an ion-trap mass spectrometer in positive ion mode. The [M+H](+) ions at m/z 166 (anatoxin-a) and m/z 180 (homoanatoxin-a) were used as the precursor ions for multiple MS experiments. MS(2)bond;MS(4) spectra displayed major fragment ions at m/z 149 (AN), 163 (HMAN), assigned to [Mbond;NH(3)+H](+); m/z 131 (AN), 145 (HMAN), assigned to [Mbond;NH(3)bond;H(2)O+H](+), and m/z 91 [C(7)H(7)](+). Although the chromatographic separation of these neurotoxins is problematic, reversed-phase LC, using a C(18) Luna column, proved successful. Calibration data for anatoxin-a using spiked water samples (10 mL) in LC/MS(n) modes were: LC/MS (25-1000 microg/L), r(2) = 0.998; LC/MS(2) (5-1000(microg/L), r(2) = 0.9993; LC/MS(3) (2.5-1000 microg/L), r(2) = 0.9997. Reproducibility data (% RSD, N = 3) for each LC/MS(n) mode ranged between 2.0 at 500 microg/L and 7.0 at 10 microg/L. The detection limit (S/N = 3) for AN was better than 0.03 ng (on-column) for LC/MS(3) which corresponded to 0.6 microg/L.