Quantitative high-performance liquid chromatography analysis of plant phospholipids and glycolipids using light-scattering detection

Application of the evaporative light-scattering principle to quantitative high-performance liquid chromatography (HPLC) analyses of plant membrane lipids has received little study. Light-scattering detection response curves were generated for nine classes of plant membrane phospholipid and glycolipids. Quantitative results obtained by HPLC/light-scattering detection and conventional lipid analytical methods (thin-layer chromatography and lipid-P assay) were in close agreement, confirming the reliability of HPLC/evaporative light-scattering detection (ELSD) analyses. Only three of the nine plant lipid classes gave linear detector response functions above 10 μg injected lipid mass. This finding contradicts earlier precepts involving light-scattering detection of lipids. At a given mass, appreciable variation in ELSD signal intensity and detection limit was found to exist among the various plant membrane lipid classes. The variation in detector response among plant lipid classes is an important consideration in achieving accurate quantitative results in plant lipid analyses.     

Natural and accelerated docosahexaenoic acid accumulation in the prenatal rat brain

The quantity and distribution of docosahexaenoic acid (DHA) in major brain phospholipids (PL) was examined in the fetal rat brain before birth, using thin-layer and capillary column gas chromatography. A rapid increment of DHA content of about 187 μg/g brain/day was observed between 17 to 20 days gestation, as opposed to 39.3±2.9 μg/g brain/day prior to that. Single intraamniotic injections of 5 μL ethyl-docosahexaenoate (Et-DHA) 12 μM, 4.25 mg) administered to 17-day-old fetuses were used to examine the uptake of DHA into brain PL. Three days following injection, the amount of n-3 polyunsaturated fatty acids increased by 28% compared to ethyl-oleate (Et-Ole) injected fetuses. Compared to the n-6 fatty acid family, the relative amount of DHA increased in the phosphoatidylserine (PS), phosphatidylethanolamine and phosphatidylcholine (PC) lipids by 15 (P=0.02), 13 and 14%, respectively. A major increase in the pool size of phosphatidylinositol and PS (110 and 50.3%, respectively), and a decrease in PC (8.2%) were observed 3 d after Et-DHA as compared to Et-Ole administration. The data suggest that a single intraamniotic administration of Et-DHA can modulate membrane PL content and alter PUFA composition.     

Rapid Detection and Quantification of Triacylglycerol by HPLC–ELSD in Chlamydomonas reinhardtii and Chlorella Strains

Triacylglycerol (TAG) analysis and quantification are commonly performed by first obtaining a purified TAG fraction from a total neutral lipid extract using thin-layer chromatography (TLC), and then analyzing the fatty acid composition of the purified TAG fraction by gas chromatography (GC). This process is time-consuming, labor intensive and is not suitable for analysis of small sample sizes or large numbers. A rapid and efficient method for monitoring oil accumulation in algae using high performance liquid chromatography for separation of all lipid classes combined with detection by evaporative light scattering (HPLC–ELSD) was developed and compared to the conventional TLC/GC method. TAG accumulation in two Chlamydomonas reinhardtii (21 gr and CC503) and three Chlorella strains (UTEX 1230, CS01 and UTEX 2229) grown under conditions of nitrogen depletion was measured. The TAG levels were found to be 3–6 % DW (Chlamydomonas strains) and 7–12 % DW (Chlorella strains) respectively by both HPLC–ELSD and TLC/GC methods. HPLC–ELSD resolved the major lipid classes such as carotenoids, TAG, diacylglycerol (DAG), free fatty acids, phospholipids, and galactolipids in a 15-min run. Quantitation of TAG content was based on comparison to calibration curves of trihexadecanoin (16:0 TAG) and trioctadecadienoin (18:2 TAG) and showed linearity from 0.2 to 10 μg. Algal TAG levels >0.5 μg/g DW were detectable by this method. Furthermore TAG content in Chlorella kessleri UTEX 2229 could be detected. TAG as well as DAG and TAG content were estimated at 1.6 % DW by HPLC–ELSD, while it was undetectable by TLC/GC method.     

Analysis of nonionic surfactants in bench-scale biotreater samples

The effluents and activated sludges used in benchscale biotreater units have been analyzed for nonionic alcohol ethoxylates and their residues. Separate bench-scale units were fed linear alcohol ethoxylates (AE), highly branched and branched nonylphenol ethoxylates. Effluents and sludges were first pretreated by a foam sublation technique to provide a gross separation of surfactants from the environmental matrix. This step was followed by normal-phase high-performance liquid chromatography (HPLC) with either fluorescence detection (FD) or evaporative light-scattering detection (ESLD). The AEs were derivatized with phenylisocyanate and analyzed by normal-phase HPLC coupled with FD. At extremely low surfactant levels, pretreatment of large sample volumes resulted in interferences on derivatization. Hence, a normal-phase HPLC method with ELSD was developed. Although some interferences do appear using ELSD, this method appears to be a more viable alternative to derivatization/FD for very low levels of AE. HPLC with FD and ELSD detection methods are more quantitative and provide information on the polyoxyethylene chain than is possible with traditional methods like cobalt-thiocyanate active substance. Presented at the 82nd AOCS Annual Meeting & Expo, May 1991, Chicago, Illinois.     

Tentative identification and quantification of TAG core aldehydes as dinitrophenylhydrazones in autoxidized sunflowerseed oil using reversed-phase HPLC with electrospray lonization MS

The molecular species of TAG core aldehydes (aldehydes still esterified to parent molecules) were detected and quantified in dietary-quality sunflowerseed oil autoxidized for 0‐18 d at 60°C in the dark. The analyses were performed by reversed-phase HPLC with UV (358 nm) absorption or light scattering and electrospray ionization-MS (ESI/MS) detection following preparation of the dinitrophenylhydrazone derivatives. Aldehyde production, as estimated by UV and ESI/MS, increased gradually over the 18-d period following a rapid initial destruction of the core aldehydes accumulated during storage of the commercial oil at 10°C for 3 mon. The contents of hydroperoxides and hydroperoxide core aldehyde combinations were estimated to account for about 5% of total TAG, quantified as area in the chromatographic trace, after 18 d of autoxidation as estimated by an evaporative light scattering detector (ELSD). The major species of core aldehydes were tentatively identified as 9-oxononanoyl (70%)-, 12-oxo-9,10-epoxydodecenoyl (10%)-, and 13-oxo-9,11-tridecadienoyl (5%)-containing acylglycerols, plus smaller amounts of simple and mixed chain-length dialdehydes, and hydroxy and epoxy monoaldehyde-containing acylglycerols (15% of total). Quantitatively, the core aldehydes made up 2–12 g/kg of oil by UV detection and 2–9 g/kg of oil by EsI/MS detection, whereas the hydroperoxides measured in the unreduced state by HPLC with ELSD were estimated at 200 g/kg after 18 d of autoxidation. The major hydroperoxides of sunflowerseed oil were as previously identified.     

Effect of absorbent in solid‐phase extraction on quantification of phospholipids in palm‐pressed fiber

Palm‐pressed fiber (PPF) is a by‐product of palm oil milling and it has been found to contain a high percentage of phospholipids (PL). The aim of this work was to evaluate the best solid‐phase extraction (SPE) method to purify PL from PPF. The purified PL were analyzed using high‐performance liquid chromatography (HPLC) with an evaporative light‐scattering detector (ELSD). The oil was extracted from PPF using the Folch method and classes of PL were purified using SPE. Different solvent phases and normal‐phase SPE cartridges [silica (Si), aminopropyl‐bonded silica (NH₂) and diol‐bonded silica (2OH)] at the same ratio of oil to sorbent mass were used to study the separation of PL from the extracted oil. The recovery of each standard PL was performed in a model oil system with the same amount of standard PL being added, and the repeatability of the SPE method was investigated. The isolation of PL by SPE diol cartridge and subsequent analysis by HPLC/ELSD have shown to be an accurate and reproducible analytical method for the determination of phosphatidylethanolamine, phosphatidylglycerol, phosphatidylcholine and lysophosphatidylcholine in PPF. This method provided a high yield and rapid separation of PL in PPF with less contamination from other lipid groups.     

Separation and analysis of phospholipids in different foods with a light-scattering detector

A method for the separation of phospholipids (PL) from total lipids by solid-phase extraction (SPE) with reversephase C₈ cartridges is described. The method was validated with a standard mixture of PL and applied to natural food matrixes, such as egg, chicken meat, salami, and ripened cheese. The recovery of PL ranged between 93 and 99.7% and was evaluated by an organic phosphorus spectrophotometric determination. The egg powder PL fraction obtained by SPE contained about 20% (w/w) nonpolar PL material when 100–150 mg of lipids were loaded onto the cartridge. Higher percentages of nonphospholipid components (30–43%) were obtained when the amount of lipids loaded was below or above the 100–150 mg range. The purified PL fractions were analyzed by high-performance liquid chromatography (HPLC) with an evaporative light-scattering detector. Good HPLC performance was observed even with low-purity SPE fractions (43% nonphospholipid material).     

Dietary supplementation with arachidonic and docosahexaenoic acids has no effect on pulmonary surfactant in artificially reared infant rats

Despite the potential use of long-chain polyunsaturated fatty acid (LCPUFA) supplementation to promote growth and neural development of the infant, little is known about potential harmful effects of the supplementation. The present study determined whether supplementation with arachidonic acid (AA) and/or docosahexaenoic acid (DHA) in rat milk formula (RMF) affects saturation of pulmonary surfactant phospholipids (PL). Beginning at 7 d of age, infant rats were artificially fed for 10 d with RMF supplemented with AA at 0, 0.5, and 1.0% of total fatty acid, or supplemented with DHA at 0, 0.5, and 1.0%, or cosupplemented with AA and DHA at levels of 0∶0, 0.5∶0.3, and 1.0∶0.6% of the fat blend. Lung tissue PL contained 43 weight percent palmitate (16∶0) of total fatty acids in infant rats fed the unsupplemented RMF. The supplementation with AA at both 0.5 and 1.0% decreased the weight percentage of 16∶0 and stearate (18∶0), indicating a decrease in saturation of PL. The observed decreases were accompanied by increases in AA and linoleic acid (18∶2n−6). Surfactant phosphatidylcholine (PC) consisted of 71 weight percent 16∶0 in the unsupplemented group, and this highly saturated PC was not altered by the cosupplementation with AA and DHA although there was a slight increase in DHA. Similarly, the cosupplementation did not change fatty acid composition of surfactant PL when compared with the unsupplemented group. The cosupplementation slightly decreased the weight percentage of 16∶0 with a proportional increase in 18∶0 leading to an unchanged weight percentage of total saturated fatty acids. These results suggest that, unlike lung tissue PL, the composition of saturated fatty acids in surfactant PL, particularly PC, is resistant to change by dietary AA and DHA supplementation. This, together with the unchanged concentration of total fatty acids in surfactant PC, indicates that LCPUFA cosupplementation causes no effect on pulmonary surfactant.     

Rapid analysis of oxidized cholesterol derivatives by high-performance liquid chromatography combined with diode-array ultraviolet and evaporative laser light-scattering detection

Extensive evidence of the deleterious biological effects of oxidized 5-cholesten-3β-ol (cholesterol) derivatives has led to great interest in their detection. We observed that known oxidized cholesterol derivatives can be rapidly quantitated by combining reversed-phase high-performance liquid chromatography (HPLC) with ultraviolet (UV) absorption and evaporative laser light-scattering (ELSD) detection. Using a 20 × 0.46 cm C18 HPLC column and methanol/acetonitrile (60:40, vol/vol) as the mobile phase at 1.0 mL/min, 10 species of oxidized cholesterol derivative were measured by UV (205, 234, and 280 nm) while 5-cholestan-5α,6α-epoxy-3β-ol (5α-epoxycholesterol), 5-cholestan-5β,6β-epoxy-3β-ol (5β-epoxycholesterol), and 5-cholestan-3β,5α,6β-triol (cholestanetriol) were detected by only ELSD. The minimal limits of detection ranged from 100 to 500 ng depending on sterol and detector. The response was linear in the range 0–1000 or 0–2000 ng depending on detector. These oxidized cholesterol derivatives were also identified by HPLC/mass spectrometry analysis combined with UV detector. Heated tallow contained cholestanetriol, 5-cholesten-3β,7α-diol (7α-hydroxycholesterol), 5-cholesten-3β,7β-diol (7β-hydroxycholesterol), 5-cholesten-3β-ol-7-one (7-ketocholesterol), 5α- and 5β-epoxycholesterols under the developed analysis condition. Photooxidized cholesterol had cholestanetriol, 7α- and 7β-hydroxycholesterols and 3,5-cholestadien-7-one. On the other hand, 7α- and 7β-hydroxycholesterols, 7-ketocholesterol, 5α- and 5β-epoxycholesterols and 3,5-cholestadien-7-one were observed in copper-oxidized low-density lipoprotein. Thus, this developed HPLC analysis method could be applied to identification of oxidized cholesterol derivatives in food and biological specimen.     

Determination of partially hydrogenated terphenyls-based thermal heating fluid in vegetable oils by HPLC with fluorescence detection

An HPLC method for the determination of partially hydrogenated terphenyls-based thermal heating fluid, Therminol 66^™, in various vegetable oils is described. Direct analysis of palm olein showed that the 3- and 4-cyclohexylbiphenyl peaks of the Therminol 66^™ used in quantitative analysis co-eluted with other fluorescent peaks present naturally in the oil. However, those interfering peaks were readily removed after saponification of palm olein. The concentrations of the 3- and 4-cyclohexylbiphenyls of Therminol 66^™ were monitored by fluorescence detection at 257 (excitation) and 320 nm (emission). The calibration graph obtained by using the peak areas of the 3- and 4-cyclohexylbiphenyls against the concentrations of Therminol 66^™ was linear, with a correlation coefficient of 0.994. The limit of quantitation, using spiked palm olein, was as low as 0.2 μg/g. The coefficients of variation obtained from the intra- and interday studies obtained by using three spiked concentrations (0.2, 0.5, and 1.0 μg/g) were 1.76–6.43 and 3.77–10.4%, respectively. The mean recovery value obtained from sunflower, soybean, and canola oils was more than 88.7%.     

Determination of the phospholipid composition of trout gill by latroscan TLC/FID: Effect of thermal acclimation

The phospholipid composition of gill tissue from thermally acclimated rainbow trout,Salmo gairdneri, was determined by Iatroscan analysis following an initial development of the chromarods in a non-polar solvent to remove neutral lipids. Standard curves for all phospholipids, although linear through most of the concentration range tested (1–40 μg), extrapolated to negative intercepts on the ordinate, indicating a decline in sensitivity at low phospholipid levels. In addition, the concentration dependence of the Iatroscan response varied by nearly 6-fold among phospholipids. Of the major phospholipids, only lysophosphatidylcholine could not be quantitated accurately because of the presence of an interfering peak. Quantitation by Iatroscan yielded results which, in general, agreed well (within 5%) with results obtained by an independent phosphate analysis. Only in the case of phosphatidylinositol (PI) did the two analytical methods differ significantly; proportions of PI were 55% higher when determined by Iatroscan as opposed to phosphate analysis. Gill tissue from 5 C-acclimated trout possessed higher proportions of phosphatidylethanolamine than tissue from 20 C-acclimated trout. The Iatroscan provided a rapid and reliable means of quantitating the proportions of all the major phospholipids of trout gill, although there are some limitations to the general applicability of the technique.     

Quantitative analysis of lipid classes by liquid chromatography via a flame ionization detector

A method is described for the direct quantitative analysis of the lipid classes of mammalian tissue lipids using high performance liquid chromatography (HPLC) with a flame ionization detector (FID). The lipid is extracted from the tissue with chloroform/methanol after deactivation of hydrolytic enzymes and removal of nonlipid substances by extraction with hot dilute acetic acid (0.05N). Separation of the lipid classes is performed with a column (45 cm × 0.2 cm id) of 8 μm silica (Spherisorb, Phase Sep, Hauppague, NY) treated with concentrated ammonium hydroxide at a solvent flow rate of 0.5 ml/min, which requires a pressure of ca. 900 psi. Cholesteryl esters (CE) and triglycerides (TG) are eluted first with Skellysolve B/methylene chloride (1∶1, v/v); cholesterol (CH) is eluted with chloroform/methylene chloride (1∶2, v/v) and the phospholipids with methanol containing 6% ammonium hydroxide added to the latter solvent in a linear gradient. The neutral lipids are eluted in ca. 12 min and the phospholipids in an additional 30 min. The relative amount of each lipid class was determined from standard curves of the peak areas obtained according to response factors using erucyl alcohol as an internal standard. The method was applied to samples of kidney, liver and serum of rats. Duplicate analyses were generally within ca. 1.0% and good agreement was obtained in the analysis of the lipid classes of Azolectin and liver mitochondria lipid compared to thin layer chromatography (TLC) via photodensitometry of charred spots or phosphorus analysis of recovered phospholipids.     

The use of the Iatroscan TH-10 analyzer to quantify total lipids in a variety of sample types and lipid classes in human gallbladder bile

Two methods for the measurement of total lipid weight in biological and geological samples and the major lipid classes in human gallbladder bile using the Iatroscan TH-10 analyzer are described. Total lipid determination involves the application of small (5 μl) volumes to Chromarods, focusing of the sample into one band by partial development in chloroform-methanol (1∶1), and quantification by flame ionization detection (FID). The response variation between different sample types did not affect the linearity of response, allowing a reproducibility of ±10% of the mean or better for samples ranging from 0.5 to 32 μg. Total lipid determinations in 10 samples could be performed in 30 min. The three major components of human gallbladder bile (cholesterol, phospholipids and bile acids) also were quantified with the Iatroscan. Samples focused on Chromarods were separated using a double development scheme in two solvent systems. All three components exhibited a linear response over the range of 0.25 to 8 μg. The repeated scanning of rods required at concentrations greater than 3 μg did not affect linearity of response. Samples from 10 patients could be processed in less than one hr. Several techniques are discussed to increase reproducibility when performing quantitative lipid analysis with the Iatroscan.     

Separation and quantitation of hydroxy and epoxy fatty acid by high-performance liquid chromatography with an evaporative light-scattering detector

A new high-performance liquid chromatography technique with an evaporative light-scattering detector (ELSD) has been developed for the separation and quantitative analysis of hydroxy and epoxy fatty acids. This method employs a gradual binary gradient (hexane/isopropanol) and ELSD detection. The minimum limit of detection is about 1 μg and ratio of mass to signal is essentially linear in the range of 10 to 200 μg. This high-performance liquid chromatography (HPLC) technique is able to separate various positional isomers of mono-hydroxy and dihydroxy fatty acids and can also discriminate between monohydroxy, epoxy, epoxyhydroxy, dihydroxy and trihydroxy fatty acids.     

Phospholipid class and FA compositions of modified soybeans processed with two extraction methods

Soybean lecithin is used as an emulsifier in food, cosmetic, and pharmaceutical industries. The proportion of individual phospholipids (PL) and their FA composition may affect the functional properties of lecithin. In this research, lecithins recovered from four modified soybeans and one commodity soybean, which were processed by extrusion-expelling and conventional solvent extraction, were analyzed for proportion of PL class and FA composition. HPLC with an ELSD analysis demonstrated that the percentage of PC in extrusion-expelled lecithin was higher than in solvent-extracted lecithin, whereas the PE content was lower. GC analysis showed that FA compositions of the PL varied with soybean type. The oil extraction method did not significantly affect FA composition. Critical micelle concentration tested with a tensiometer showed differences among the lecithins.     

A micro enzymic method for determination of choline-containing phospholipids in serum and high density lipoproteins

A micro method is described for the assay of choline-containing phospholipids in serum and high density lipoproteins (HDL) using an automated microtiter plate reader. The method is adapted from the enzymic method of Takayama, Itoh, Nagasaki, and Tanimuzu (Clin. Chim. Acta 79, 93–98, 1977) using phospholipase D, choline oxidase, and peroxidase coupled with the color generating system phenol and 4-amino-antipyrine. The micro method requires 5 μL of serum or HDL sample, and 42 samples can be assayed in duplicate in one run using a 96-well flat-bottom microtiter plate. The reaction is linear up to 400 mg/dL and the lower limit of detection is 0.25 mg of choline-containing phospholipids per assay. The coefficient of variation within an assay is 0.86–0.79%, and day-to-day variation is 0.9–1.5%. Results obtained by the micro method are in excellent agreement with those obtained by the procedure of Takayamaet al. (r=0.997). The supernatant left after removal of low density lipoproteins and very low density lipoproteins from serum and precipitation with heparin/manganese chloride reagent can thus be conveniently used for the micro assay of choline phospholipids in HDL.     

Effect of pH on a gossypol complex with phospholipids: Gossypol-PE

PE, one of the major phospholipids in oilseed soapstock, may react with gossypol to form Schiff bases. PE amounts to 20–30% of the phosphorus compounds in soap-stock. In this report, the dependence on pH of the Schiff base products between PE and gossypol was investigated using a spectrophotometer, an HPLC equipped with an ELSD, and an LC-MS system. We observed that at pH 7 the Schiff reaction product and reactants were clearly detected by ELSD, absorption, and LC-MS spectra. The absorption spectra displayed the characteristic peak for the Schiff bases around 430–440 nm. The absorption spectra also indicated that the reaction was pH dependent. The reaction temperatures were 60 and 90°C. The LC-MS spectra supported the formation of Schiff bases as well as methyl ether derivatives of gossypol in alcohol at the elevated temperatures. The implications of these experimental findings are presented in this paper.     

Isolation of Pure Phospholipid Fraction from Egg Yolk

The phospholipid (PL) fraction from egg yolk was isolated and purified. In the procedure applied (method 2) the egg yolk was extracted with ethanol, precipitated using acetone chilled to −20 °C and washed using acetone. The purity of the samples was checked by HPLC analysis using a Charged Aerosol Detector (CAD). The results were compared with those obtained for the phospholipid fraction isolated and purified by deoiling yolk before extraction and the precipitation of PL with acetone chilled to 4 °C (method 1). The use of acetone chilled to −20 °C to precipitate and wash the phospholipids yielded the phospholipid fraction with 100% purity (78.7 ± 0.2 of phosphatidylcholine and 21.3 ± 0.2 of phosphatidylethanolamine). When deoiling and the 4 °C purification process was used (method 1) 0.4 ± 0.1% cholesterol and some traces of triacylglycerols remained in the PL fraction.     

Fast analysis of rapeseed glucosinolates by near infrared reflectance spectroscopy

Industry and plant breeders require fast methods to analyze glucosinolates in rapeseed. We tested the potential of near infrared reflectance spectroscopy (NIRS) for this analysis and developed calibration equations on a large population of whole seeds. Reference methods used are high performance liquid chromatography (HPLC) and gas liquid chromatography (GLC) of desulphoglucosinolates, a glucose-release method after purification on an anion exchange column and a palladium test. In the range from 2 to 107 μM/g and after transformation of the data in first derivative, a correlation coefficient of 0.99 was observed, as well as standard errors of estimated values of 2.15, 2.52, 2.67 and 4.07 for samples analyzed by HPLC, GLC, glucose and the palladium test, respectively. With different wavelengths, a limited calibration test on seeds containing from 4 to 40 μM/g gives a standard error of 1.91 μM/g (HPLC).     

Measurement of Ether Phospholipids in Human Plasma with HPLC–ELSD and LC/ESI–MS After Hydrolysis of Plasma with Phospholipase A1

Ethanolamine ether phospholipid (eEtnGpl) and choline ether phospholipid (eChoGpl) are present in human plasma or serum, but the relative concentration of the ether phospholipids in plasma is very low as compared to those in other tissues. Nowadays, measurement of ether phospholipids in plasma depends on tandem mass spectrometry (LC/MS/MS), but a system for LC/MS/MS is generally too expensive for usual clinical laboratories. Treatment of plasma with phospholipase A₁ (PLA1) causes complete hydrolysis of diacylphospholipids, but ether phospholipids remain intact. After the treatment of plasma with PLA1, both eEtnGpl and eChoGpl are detected as independent peaks by high‐performance liquid chromatography with evaporative light scattering detection (HPLC–ELSD). The same sample used for HPLC–ELSD can be applied to detect eEtnGpl and eChoGpl with electrospray ionization mass spectrometry. Presence of alkylacylphospholipids in both eChoGpl and eEtnGpl in human plasma was indicated by sequential hydrolysis of plasma with PLA1 and hydrochloric acid.