Erlend Hvattum

Analysis of triacylglycerols with non-aqueous reversed-phase liquid chromatography and positive ion electrospray tandem mass spectrometry

A non-aqueous reversed-phase liquid chromatographic method coupled to electrospray ionisation (ESI) tandem mass spectrometry was developed for the analysis of triacylglycerols (TGs). The synthetic TGs studied were separated according to their equivalent carbon number with a gradient of methanol (containing 0.01% (v/v) formate adjusted to pH 5.3 with ammonia) and chloroform. ESI mass spectra of TGs yielded positive ion current signals for [M + NH(4)](+) and [M + NH(4)-RCOONH(4)](+). The mass spectra also showed signals believed to arise from [M + K](+). Collision-induced dissociation (CID) of the [M + NH(4)](+) precursor ion yielded [M + NH(4) - RCOONH(4)](+), [RCO + 74](+) and [RCO](+) product ions as aids for the structural elucidation of the TGs. In addition, [RCO - 18](+) and small amounts of [RCO - 2](+) product ions were also found. The latter ions were observed only for TGs containing unsaturated fatty acids. CID of ammoniated 1-stearoyl-2-oleoyl-3-linoleoyl-glycerol (18:0/18:1/18:2) indicated that neutral loss of the sn-2 fatty acid was energetically less favourable than loss of the fatty acid from the sn-1 or sn-3 position.

Study of mechanisms involved in the collision‐induced dissociation of carboxylate anions from glycerophospholipids using negative ion electrospray tandem quadrupole mass spectrometry

The collision-induced dissociation of the carboxylate anions from human blood phosphatdilycholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), phosphatidylinositol (PI) and phosphatidic acid (PA) containing the C18:0 (sn-1) and C20:4 (sn-2) fatty acyl residues was studied using normal phase liquid chromatography coupled with negative ion electrospray tandem mass spectrometry. The product ion peak area ratio of C18:0 to C20:4 was calculated for each phospholipid species and was found to increase with increasing collision energy for all classes. For the phospholipids with a net neutral charge (PE, PC) there was a preferential loss of the sn-2 carboxylate anion (C20:4) at low collision energy, while at higher energy there was a preferential loss of the sn-1 carboxylate anion (C18:0). For the phospholipids with a net negative charge (PI, PA, PS) the intensity of the sn-1 carboxylate anion peak was equal to or higher than the sn-2 carboxylate anion peak at the energies measured. At a given collision energy the product ion peak area ratio decreased in the order PA > or = PS > PI. Studying PS and PE species at different collision energies, it was found that for both classes the increase in the abundance ratio with increasing collision energy was largely dependent on the chain length and degree of unsaturation of the sn-2 acyl chain.

Effect of soybean oil and fish oil on individual molecular species of Atlantic salmon head kidney phospholipids determined by normal-phase liquid chromatography coupled to negative ion electrospray tandem mass spectrometry

The effect of soybean oil (SO) and fish oil (FO) on the relative molecular species distribution of phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylinositol (PI) and phosphatidylserine (PS) in Atlantic salmon head kidney was studied using normal-phase liquid chromatography coupled with negative ion electrospray tandem mass spectrometry. The conformation of identity of the phospholipid species was based on retention time, the mass of the [M-H]- ([M-15]- for PC) molecular ions and the carboxylate anion fragments in the product ion spectrum. The intensity ratio of sn-1/sn-2 fragment ions increased with increasing number of double bonds in the sn-2 acyl chain but was not affected by increasing number of double bonds in the sn-1 acyl chain of the species examined. The relative distribution of the molecular species was determined by multiple reaction monitoring of the carboxylate anion fragment from the sn-1 position. A total of 68 different phospholipid species were determined in the head kidney and the largest amount was found in PE (22 species). Depending on the diet, the main species identified in the different phospholipid classes were; PC 16:0/18:1, PE 16:0/22:6, PI 18:0/20:4 and PS 16:0/22:6. The SO diet significantly increased the 18:2, 20:3 and most 20:4 containing species and significantly reduced the 14:0 and most 20:5 and 22:6 fatty acid containing species. The increase of the 20:4 and the decrease of the 20:5 and 22:6 containing species were dependent on the fatty acid combination of the species.

Detection and quantitation of gadolinium chelates in human serum and urine by high-performance liquid chromatography and post-column derivatization of gadolinium with Arsenazo III

A narrow-bore high-performance liquid chromatography method was developed for simultaneous separation of gadolinium diethylenetriaminepentaacetic acid (GdDTPA), the monomethylamide (GdDTPA-MMA) and the bis-methylamide (GdDTPA-BMA) in human serum and urine. The Gd complexes were detected at 658 nm after post-column derivatization with Arsenazo III. The serum samples were ultrafiltrated, whereas the urine samples were centrifuged and diluted before analysis. With an injection volume of 10 microliters on a 2.1 mm ID reversed-phase column, the limit of detection of GdDTPA-BMA was calculated as 0.3 microM and 1.1 microM in serum and urine, respectively. The method was validated with respect to GdDTPA-BMA with a limit of quantification set to 2 microM and 10 microM in serum and urine, respectively. The best fit of the calibration curve was obtained using non-linear regression according to the equation Y = A+BX+CX2 in the concentration ranges 2-800 microM and 10-2000 microM of GdDTPA-BMA in serum and urine, respectively. The precision of the method was found to range from 1 to 4% RSD. The recoveries of GdDTPA-BMA spiked in serum and urine were higher than 95% with an RSD equal to or less than 4%. The serum samples were stable for at least 5 months when stored at -70 degrees C, and the urine samples were stable for a least 6 months when stored at -20 degrees C.

Separation and identification of phosphatidylserine molecular species using reversed-phase high-performance liquid chromatography with evaporative light scattering and mass spectrometric detection

A reversed-phase HPLC method compatible with evaporative light scattering (ELS) and electrospray mass spectrometric (ES-MS) detection was developed for separation of phosphatidylserine (PS) molecular species. The method was optimised for separation of three disaturated synthetic species: dipalmitoyl glycerophosphoserine, palmitoyl-stearoyl glycerophosphoserine and distearoyl glycerophosphoserine using isocratic elution with a mixture of 2-propanol, tetrahydrofuran and ammonium formate. Baseline separation was obtained on three different columns: one polystyrene/divinylbenzene (PS/DVB) column and two silica based C(18) and C(30) columns. The best chromatographic resolution was achieved with the C(30) column. The limit of detection for DPPS was 5 microg/ml (S/N=3) with ELS detection and 0.1 microg/ml (S/N=3) with negative ion ES-MS in the single ion monitoring mode. Baseline separation of the five main species in a biological PS sample, bovine brain PS, was obtained with the PS/DVB column. Species identification was done by using the retention times of the intact PS species and their corresponding carboxylate anion fragments obtained by in-source fragmentation. Data have shown that individual PS species can be identified by their retention times using direct ELS detection in a mixture of disaturated PS species. However, for the bovine brain PS electrospray-MS detection was necessary for species identification due to the many possible fatty acid combinations in biological PS.

Specific detection and quantification of palmitoyl-stearoyl-phosphatidylserine in human blood using normal-phase liquid chromatography coupled with electrospray mass spectrometry

A narrow-bore normal-phase high-performance liquid chromatography (HPLC) method was developed for separation of phospholipid classes using an HPLC diol column and a gradient of chloroform and methanol with 0.2% formic acid titrated to pH 5.3 with ammonia. The HPLC system was coupled on-line with an electrospray mass spectrometry (ES-MS) or electrospray tandem mass spectrometry (ES-MS-MS) system and the separation of several major phospholipid classes was shown. The molecular species of some phospholipid classes in human blood were qualitatively determined. A method was further developed for specific determination of a molecular species from phosphatidylserine, palmitoyl-stearoyl-phosphatidylserine (PSPS), in human blood using HPLC-ES-MS. The analyses were performed by single ion monitoring of the [M-H]- molecular ions of PSPS and an internal standard, dipalmitoyl-phosphatidylserine. The limit of quantification of the method was 1.2 ng of PSPS. The calibration curve ranged from 0.12 to 5.81 microg/ml of PSPS dissolved in the mobile phase. The curve was fitted to a second-order polynomial equation and found to be highly reproducible. Analysis of control samples was found to be reproducible with a between-series precision below 9.2% R.S.D. The amount of endogenous PSPS in human blood was determined in 13 subjects and found to range from 1.73 to 3.09 microg/ml. The identity of endogenous PSPS was confirmed by HPLC-ES-MS-MS.

Determination of perfluorobutane in rat blood by automatic headspace capillary gas chromatography and selected ion monitoring mass spectrometry.

A new contrast agent (Sonazoid; NC100100) for ultrasound imaging has been developed. It is an aqueous suspension of lipid stabilised perfluorobutane (PFB) gas microbubbles. An automatic headspace capillary gas-chromatographic mass spectrometric method using electron impact ionisation was developed for analysis of Sonazoid PFB in rat blood. The calibration standards were gaseous PFB dissolved in ethanol in the range of 0.5-5000 ng PFB. Fluorotrichloromethane (CFC 11) was used as an internal standard of the method and the MS detector was set to single ion monitoring of the base fragment ions of PFB (m/z 69 and 119) and CFC 11 (m/z 101). The calibration graph, made by plotting the peak area ratios of PFB (m/z 69) to CFC 11(m/z 101) against the amount of PFB, was fitted to a second-order polynomial equation with weighting 1/y2 and found to be reproducible. The limit of quantification of the method was set to 0.4 ng PFB. The between-day variation of the method was below 9.2% relative standard deviation (RSD) and the within-day variation of the method was below 7.6% RSD. The accuracy of the method, as compared to Coulter counter, was estimated by determination of PFB in samples where Sonazoid was added to saline and found to range from 91.5% to 105.2%. PFB, added as Sonazoid, was found to be stable for at least 7 months in rat blood samples when stored at -20 degrees C.