Search

Phosphine reagents were designed and synthesized as a new type of fluorescent reagents for the determination of lipid hydroperoxides in foodstuff and biological materials. All phosphine reagents prepared had no fluorescence but their oxides, which were produced by the reaction of the phosphines with hydroperoxides, had strong fluorescence. Among the phosphine reagents prepared, diphenyl-1-pyrenylphosphine had the most suitable properties as a fluorescent reagent and was successfully applied to the determination of hydroperoxides by batch, flow injection and HPLC post-column methods.

Per-O-methylated pentopyranoses and hexopyranoses were converted to their glycosyl chlorides and coupled with cesium salt of fluorescent chiral derivatization reagent, (+)-2-methyl-2-beta-naphthyl-1,3-benzodioxole-4-carboxylic acid [(+)-MND carboxylic acid], to afford their 1-O-(+)-MNB carboxylates. The D,L enantiomers were separated by normal-phase HPLC and determined at the picomolar level, and this methodology could be extended to the highly sensitive simultaneous determination of per-O-methylated monosaccharides.

The fluorimetric determination of okadaic acid (OA) and dinophysistoxin-1 (DTX-1), the principal toxins of diarrhetic poisoning, is reported. The digestive glands of mussels or scallops were homogenized with 2-propanol. OA and DTX-1 were extracted from the homogenate, with hexane-ethyl acetate and labelled with 2,3-(anthracenedicarboximido)ethyl trifluoromethanesulfonate in dry acetonitrile. After cleaning up by passage through a short silica gel column, the fluorescent derivatives were determined by HPLC. The derivatives were at first separated on a Develosil Ph-5 column, and only the target fraction obtained was introduced into a Develosil ODS K-5 column by a valve-switching device. Both toxins were determined in the range 2.5-500 pg, and the detection limits were 0.8 pg (OA) and 1.3 pg (DTX-1) with a signal-to-noise ratio of 3.

An automatic method for the determination of hydroperoxides of phosphatidylcholine (PC) and phosphatidylethanolamine (PE) is reported. Sample plasma was deproteinized with a fourfold volume of methanol. After centrifugation, the supernatant was injected directly into an HPLC system without further treatment. The hydroperoxides of PC and PE were concentrated and washed on an ODS column followed by introduction into two analytical columns, a silica gel and an aminopropylsilica gel column, which were connected in series, by column switching. After the separation, they were detected by postcolumn detection with diphenyl-1-pyrenylphosphine. The compounds were determined at picomole levels within 30 min with good reproducibilities. By using only a silica gel column as an analytical column, PC hydroperoxides were determined within 20 min, and samples could be injected into it at 15-min intervals. Those methods made it possible to inject a sample of up to 2 ml at one time and up to 8 ml by repeated injections and to determine phospholipid hydroperoxides in human plasma at picomole levels.

A method for the simultaneous determination of hydroperoxides of phosphatidylcholines (PC), triacylglycerols (TG) and cholesterol esters (CE) has been developed. A sample was separated into a combined TG and CE hydroperoxides fraction and a PC hydroperoxides fraction on a short silica column. The fractions were introduced into an ODS column and another silica column by a valve-switching device. The PC hydroperoxides were monitored by a post-column detection system with diphenyl-1-pyrenylphosphine, and the TG and CE hydroperoxides were monitored by another switching device. With this system, the hydroperoxides were determined at the picomole level within 32 min. Their detection limits were 2-4 pmol at a signal-to-noise ratio of 3, and the relative standard deviations of the peak areas were 1.6-3.1%. This method was successfully applied to determine lipid hydroperoxides in human plasma.

Cholesterol ester (ChE) and triacylglycerol (TG) hydroperoxides in human plasma were determined by high-performance liquid chromatography with postcolumn detection with diphenyl-1-pyrenylphosphine. Human plasma was extracted once with n-hexane. 2,6-Di-tert.-butyl-4-methylphenol and N-stearylcinnamide were added to human plasma before extraction as an antioxidation agent and an internal standard, respectively. The detection limits of both ChE and TG hydroperoxides were 1 pmol. The sample size was minimized to 250 microliters for each run. The recoveries of ChE and TG hydroperoxides from fresh plasma were ca. 90 and 80%, respectively. The relative standard deviations (n = 8) of their values in frozen human plasma were 5.4% (ChE hydroperoxides, 298 nM) and 5.7% (TG hydroperoxides, 267 nM). No TG hydroperoxides and 24.5 +/- 9.6 nM (n = 15) ChE hydroperoxides were detected in fresh human plasma. The relative standard deviation (n = 8) of ChE hydroperoxides values in fresh plasma was 5.8% (27.1 nM).

Triacylglycerol monohydroperoxides (TG-mHPO) were selectively detected at the picomole levels after post-column reaction with diphenyl-1-pyrenylphosphine (DPPP). TG-mHPO were separated on two types of reserved-phase columns, an ODS column and a phenylated silica gel column, which were useful for determining TG-mHPO at their molecular species levels and their class levels, respectively. After the separation, DPPP solution was mixed with the eluent followed by reaction in a stainless-steel coil 20 m x 0.5 mm I.D. at 80 degrees C, then the fluorescence intensity of DPPP oxide was measured (lambda ex. 352 nm, lambda em. 380 nm). Using these systems, TG-mHPO were determined in the range 2-1000 pmol. The relative standard deviations were 2.3-2.8%.

We describe a sensitive CD method for determining the stereospecificity in lipase (E.C.3.1.1.3) catalysed hydrolysis of triacyl glycerols into diacyl glycerols. The diglycerols were converted to chiral tert-butyldimethylsilylated 1,2- or 2,3-di-O-benzoyl-sn-glycerol (5 or 5'), and their CD was measured. This approach showed for the first time that lipases produce optically active diacyl glycerides from achiral tripalmitin and tribenzoyl glyceride with a variable extent of enantioselectivity depending on the acyl groups and the enzymes.