Electron Paramagnetic Resonance Investigation of Stratum Corneum Lipid Structure

Electron paramagnetic resonance (EPR) in conjunction with a slow-tumbling simulation was utilized for defining stratum corneum (SC) lipid structure. We found that ordering calculated from the simulation is an appropriate index for evaluating SC lipids structure. The SC from two sites (mid-volar forearm and lower-leg) of human volunteers was stripped consecutively from one to three times using a glass plate coated with a cyanoacrylate resin. Aliphatic spin probes, 5-doxylstearic acid (5-DSA) and 3β-doxyl-5α-cholestane (CHL), were used to monitor SC ordering. EPR spectrum of 5-DSA incorporated in the SC demonstrated a characteristic peak for the first strip. However, EPR spectra of CHL in the SC did not show a clear difference for each strip, except for the peak intensity. The results imply that CHL is not incorporated into the lipid phase as easily as 5-DSA is. A slow-tumbling simulation of the EPR spectrum was performed to analyze the detailed lipid structure. The simulation results for 5-DSA show differences in values of the SC ordering as a function of depth. Thus, these results along with the simulation analysis provide a detailed SC layer structure.     

Spin-Probe Investigations of Head Group Behavior in Aqueous Dispersions of a Nonionic Amphiphilic Compound

Head group behavior of nonionic amphiphilic compound, (poly(oxyethylene) hydrogenated castor oil, HCO), in aqueous dispersions were investigated by EPR (electron paramagnetic resonance) in conjunction with a modern slow-tumbling simulation. The aliphatic spin probes, 5-doxylstearic acid (5-DSA) and 3β-doxyl-5α-cholestane (CHL), were used to obtain fluidity of the surface region of the membrane. The order parameter (S ₀) using the simulation for 5-DSA and CHL in the region were approximately 0.4 and 0.2, respectively. The ordering results suggest that the head group region of the membrane is somewhat fluid. The rotational diffusion coefficients (R _⊥ ≈ 1/(6τ_R)) for the probes were 3.4 × 10⁷ and 7.1 × 10⁷ s⁻¹, respectively. Activation energies, calculated using the temperature dependence of diffusion coefficients, were 18 and 17 kJ/mol for the probes. The EPR results imply that the CHL probe in the HCO membrane has quite different behavior in comparison with that of PC (phosphatidylcholine) from egg. Thus, the present EPR analyses have provided quantitative insight into the surface region of the amphiphilic membrane.     

EPR Investigations of Spin-Probe Dynamics in Aqueous Dispersions of a Nonionic Amphiphilic Compound

Dynamics of various spin probes in aqueous dispersions of nonionic amphiphilic compound, [poly(oxyethylene) hydrogenated castor oil, HCO], were investigated by EPR (electron paramagnetic resonance) and saturation recovery (SR) spectroscopies. Partitioning, rotational correlation time (τ_R), rotational diffusion coefficient, and electron spin-lattice relaxation time (T _1e) in dispersions of the HCO membrane were obtained. The partitioning of water soluble spin probes, DTBN and TEMPO, in the aqueous and vesicle phases was determined by an EPR linewidth simulation as a function of temperature. The results suggest that DTBN and TEMPO have a similar partitioning in the vesicle phase throughout the temperatures studied. The longer τ_R and shorter T _1e (~0.33 μs) values of DTBN in the vesicle phase were obtained, and could be attributed to the probe environment in the membrane. The simulation results for fast tumbling probes were quite different from those of conventional intensity analysis (spectral parameter, f). Thus, the simulation and T _1e analyses have provided a quantitative understanding of the probe dynamics in both phases. Aliphatic spin probes, doxylstearic acids (DSAs) and 3β-doxyl-5α-cholestane (CHL), were used for monitor of various membrane motions. The EPR spectra were quantitatively analyzed by a slow tumbling simulation. The rotational diffusion coefficients and order parameter were obtained by the simulation. In addition, the direct observations of the behavior of the probes were measured by SR method. The results were consistent with T _1e obtained for spin probes. Thus, the quantitative results regarding EPR, SR method, various simulation analyses have provided detailed information regarding physicochemical properties of the various moieties of the probe region in the amphiphilic compound.     

Effects of dietary cholesterol and triglycerides on lipid concentrations in liver, plasma, and bile

Dietary cholesterol (CHL) and triglycerides (TG) can influence plasma, hepatic, and biliary lipid composition, but effects on lipids in these three compartments during the early stages of CHL gallstone formation have not been studied in parallel. We fed prairie dogs diets containing one of four tes oils (safflower, coconut, olive, or menhaden) at either 5 or 40% of calories, in the presence of 0 or 0.34% CHL, for 3 wk. In the absence of dietary CHL, increases in dietary TG produced 50–200% increases in the concentrations of biliary CHL and hepatic cholesteryl ester (CE), while the concentrations of hepatic free CHL (FC) as well as plasma FC and CE remained relatively unchanged. Increasing dietary CHL to 0.34% resulted in increases in hepatic FC of approximately 50% for all four fats regardless of whether they were supplied at 5 or 40% of calories. CHL supplementation caused more pronounced increases in biliary CHL (200–400%), hepatic CE (50–200%), plasma FC (up to 100%), and plasma CE (up to 150%), and these increases were exacerbated by concurrent supplementation of dietary fat and CHL (biliary CHL: 300–700%; hepatic CE: 100–250%; plasma FC: up to 165%; plasma CE: 100–350%). These results indicate that enhanced secretion of biliary CHL and, to a lesser extent, increased synthesis of hepatic CE, may be primary mechanisms for maintaining the hepatic FC pool. Furthermore, dietary CHL and high levels of fat intake are independent risk factors for increasing biliary CHL concentrations, and adverse effects on lipid concentrations in plasma and bile tend to be exacerbated by ingestion of diets rich in both fat and CHL.     

Anomalous enantioselectivity in the sharpless asymmetric dihydroxylation reaction of 24-nor-5β-cholest-23-ene-3α,7α,12α-triol: Synthesis of substrates for studies of cholesterol side-chain oxidation

Recently we described a block in bile acid synthesis in cerebrotendinous xanthomatosis (CTX), a lipid storage disease related to an inborn error of bile acid metabolism. In this disease a defect in hepatic microsomal (24S) hydroxylation blocks the transformation of 5β-cholestane-3α,7α,12α,25-tetrol into (24S) 5β-cholestane-3α,7α,12α,24,25-pentol and cholic acid. Mitochondrial cholesterol 27-hydroxylation has also been reported to be abnormal in CTX subjects, but the relative importance of the enzymatic defect in this alternative microsomal pathway (namely, the 24S hydroxylation of 5β-cholestane-3α,7α,12α,25-tetrol relative to the abnormality in mitochondrial 27-hydroxylase) has not been established in CTX. To delineate the sequence of side-chain hydroxylations and the enzymatic block in bile acid synthesis, we synthesized the (23 R and 23 S) 24-nor-5β-cholestane-3α,7α,12α,23,25-pentols utilizing a modified Sharpless asymmetric dihydroxylation reaction on 24-nor-5β-cholest-23-ene-3α,7α,12α-triol, a C₂₆ analog of the naturally occurring C₂₇ bile alcohol, 5β-cholest-24-ene-3α,7α,12α-triol. Stereospecific conversion of the unsaturated 24-nor triol to the corresponding chiral compounds (23R and 23S), 24-nor-5β-cholestane-3α,7α,12α,23,25-pentols, was quantitative. However, conversion of the unsaturated 24-nor triol to the chiral nor-pentols had absolute stereochemistry opposite to the products predicted by the Sharpless steric model. The absolute configurations and enantiomeric excess of the C₂₆ nor-pentols and the C₂₇ pentols (synthesized from 5β-cholest-24-ene-3α,7α,12α-triol for comparison) were confirmed by nuclear magnetic resonance and lanthanide-induced circular dichroism Cotton effect measurements. These results may contribute to a better understanding of the role of the 24S-hydroxylation vs. 27-hydroxylation step in cholic acid biosynthesis. Presented in part at the 216th American Chemical Society National Meeting (Medicinal Chemistry Division, Abs. 368), Boston, MA, August 21–26, 1998.     

Rate constants for quenching singlet oxygen and activities for inhibiting lipid peroxidation of carotenoids and α-tocopherol in liposomes

The ¹O₂ quenching rate constants (k _Q ) of α-tocopherol (α-Toc) and carotenoids such as β-carotene, astaxanthin, canthaxanthin, and lycopene in liposomes were determined in light of the localization of their active sites in membranes and the micropolarity of the membrane regions, and compared with those in ethanol solution. The activities of α-Toc and carotenoids in inhibiting ¹O₂-dependent lipid peroxidation (reciprocal of the concentration required for 50% inhibition of lipid peroxidation: [IC₅₀]⁻¹) were also measured in liposomes and ethanol solution and compared with their k _Q values. The k _Q and [IC₅₀]⁻¹ values were also compared in two photosensitizing systems containing Rose bengal (RB) and pyrenedodecanoic acid (PDA), respectively, which generate ¹O₂ at different sites in membranes. The k _Q values of α-Toc were 2.9×10⁸M⁻¹s⁻¹ in ethanol solution and 1.4×10⁷ M⁻¹s⁻¹ (RB system) or 2.5×10⁶ M⁻¹s⁻¹ (PDA system) in liposomes. The relative [IC₅₀]⁻¹ value of α-Toc in liposomes was also five times higher in the RB system than in the PDA-system. In consideration of the local concentration of the OH-group of α-Toc in membranes, the k _Q value of α-Toc in liposomes was recalculated as 3.3×10⁶ M⁻¹s⁻¹ in both the RB and PDA systems. The k _Q values of all the carotenoids tested in two photosensitizing systems were almost the same. The k _Q value of α-Toc in liposomes was 88 times less than in ethanol solution, but those of carotenoids in liposomes were 600–1200 times less than those in ethanol solution. The [IC₅₀]⁻¹ value of α-Toc in liposomes was 19 times less than that in ethanol solution, whereas those of carotenoids in liposomes were 60–170 times less those in ethanol solution. There were no great differences (less than twice) in the k _q and [IC₅₀]⁻¹ values of any carotenoids. The k _Q values of all carotenoids were 40–80 times higher than that of α-Toc in ethanol solution but only six times higher that of α-Toc in liposomes. The [IC₅₀]⁻¹ values of carotenoid were also higher than that of α-Toc in ethanol solution than in liposomes, and these correlated well with the k _Q values.     

The inhibitory guanine nucleotide-binding protein Gi2α induces and potentiates adipocyte differentiation

The present study further elucidates the involvement of the α-subunit of the GTP-binding protein G_i2 in the differentiation of murine 313-L1 cells. Control and vector-transfected cells attained a fully differentiated adipocyte phenotype showing ample lipid droplets. Cells expressing wild type (WT)-G_i2α or the constitutively active R179E-G_i2α, however, became enlarged, less confluent, and produced large amounts of lipids. Differentiation consistently increased the triglyceride (TAG) content in control cells. In both WT-G_i2α and R179E-G_i2α clones, a marked increase in TAG could be detected even prior to insulin/dexamethasone/isobutyl methylxanthine exposure. The activity of palmitoyl-CoA synthetase (PCS) and glycerophosphate acyltransferase (GPAT) also increased upon differentiation. WT-G_i2α and R179E-G_i2α overexpression also enhanced PCS and GPAT activities even before differentiation medium was added. The total amount of phospholipids (PL) generally increased upon differentiation; however, pre- and postdifferentiation values were insignificantly different in cells expressing WT-G_i2α and R179E-G_i2α. Differentiation altered the PL profile with a relative shift from phosphatidylcholine and phosphatidylethanolamine to phosphatidylinositol (PI) in differentiated cells. Finally, differentiation yielded a general increase in the activity of basal PI-phospholipase-C activity. Again, cells expressing WT-G_i2α and R179E-G_i2α demonstrated elevated enzyme activity and enhanced second messenger accumulation subsequent to differentiation. In summary, cells with the R179E-mutants of G_i2α exhibited stimulated lipid turnover and accumulation in both undifferentiated and differentiated cells.     

Levels of Lipid Peroxidation in Human Plasma and Erythrocytes: Comparison between Fatty Acids and Cholesterol

Lipid peroxidation has gained renewed attention with increasing evidence showing its biological role in producing toxic compounds and cellular signaling mediators. The assessment of lipid peroxidation levels in vivo is difficult partly because lipids are oxidized by different oxidants by different mechanisms to give versatile types of products, which may undergo metabolism and secondary reactions. In the present study, total hydroxyoctadecadienoic acids (tHODE) and 7α- and 7β-hydroxycholesterol (t7-OHCh) from 44 healthy human subjects were assessed as biomarkers after reduction with sodium borohydride followed by saponification with potassium hydroxide comparing with the prevailing standard 8-isoprostaglandin F_2α (t8-iso-PGF_2α). The average concentrations of tHODE, total 8-isoprostaglandin F_2α (t8-iso-PGF_2α), t7α-OHCh, and t7β-OHCh were 203, 0.727, 87.1, and 156 nmol/l plasma and 1,917, 12.8, 1,372, and 3,854 nmol/l packed erythrocytes, respectively. The ratios of tHODE and t7-OHCh to the parent substrates were 194 and 3,519 μmol tHODE/mol linoleates and 40.9 and 686 μmol t7-OHCh/mol cholesterol in plasma and erythrocytes, respectively. It was found that (1) t7-OHCh in blood was unexpectedly high, as high as or even higher than tHODE, (2) the amounts of tHODE was more than 100 fold higher than t8-iso-PGF_2α (3) the level of lipid oxidation products in erythrocytes was higher than that in plasma, and (4) lipid peroxidation products level tended to increase while antioxidant level decrease with age. These products may be used as potential biomarker for assessment of lipid peroxidation and oxidative stress in vivo.     

Role of prostanoids and lipid peroxides as mediators of the 12-O-tetradecanoylphorbol-13-acetate effect on cell growth

Confluent cultures of guinea pig smooth muscle cells (SMC) or human fibroblasts (HNF) were treated with 12-tetradecanoylphorbol-13-acetate (TPA). Prostanoid levels were measured by the radioimmunoassay of 6-keto-PGF_1α and PGE₂, and lipid peroxides were measured by the thiobarbituric acid test for malondialdehyde (MDA). Cells were seeded at low densities, and growth was calculated both from the cell count (Coulter Counter) and the colony number (image analysis). When confluent SMC and HNF were incubated in media alone, 6-keto-PGF_1α levels were a function of the TPA concentration, increasing to a maximum at 10⁻⁸ M TPA and then decreasing at higher TPC concentrations. When confluent SMC and HNF were incubated in media containing exogenous arachidonic acid, 6-keto-PGF_1α levels again increased to a maximum at 10⁻⁸M TPA but decreased at higher TPA concentrations only with SMC. The increase in 6-keto-PGF_1α levels was much greater in HNF (1310%) than SMC (680%). SMC synthesized similar amounts of 6-keto-PGF_1α and PGE₂, and the stimulatory effect of TPA was similar with 6-keto-PGF_1α and PGE₂. Indomethacin (IM) blocked prostanoid synthesis at all TPA concentrations. TPA did not have a significant effect on MDA levels in either cell line. The lipid antioxidants α-tocopherol and α-tocopherylquinone blocked lipid peroxidation without affecting the stimulation of prostanoid synthesis with TPA. Cell number decreased to a minimum at 10⁻⁸M TPA in both cell lines. The decrease in cell number was much greater in HNF (72%) than SMC (30%). SMC colony number decreased at 10⁻⁸ TPA and then increased at 10⁻⁶M TPA. IM did not block the TPA effect on cell number in either cell line. The lipid antioxidant butylated hydroxytoluene (BHT) did not block the TPA effect on SMC cell number. The IM and the BHT data show that the TPA effect on cell growth is not mediated by prostanoid or lipid peroxide products of arachidonic acid metabolism. However, the increase in prostanoid synthesis parallels the decrease in cell number, and the effects are maximal at the same TPA concentration. These correlations suggest a common cellular process affecting both prostanoid synthesis and cell growth that is initiated or enhanced by TPA. This publication is Part VIII in the series, “Fatty Acid Metabolism and Cell Proliferation”.     

Penetration and distribution of α-tocopherol, α- or γ-tocotrienols applied individually onto murine skin

To evaluate skin penetration of various vitamin E homologs, a 5% solution of either α-tocopherol, α-tocotrienol, or γ-tocotrienol in polyethylene glycol was topically applied to SKH-1 hairless mice. After 0.5, 1, 2, or 4 h (n=four per time point and four per vitamin E homolog), the skin was washed, the animals killed, the skin rapidlly removed, frozen on dry ice, and a biopsy taken and sectioned: stratum corneum (two uppermost, 5-μm sections—SC1 and SC2), epidermis (next two 10μm sections—E1 and E2), papillary dermis (next 100μ, PD), dermis (next 400 μm, D), and subcutaneous fat (next 100 μm, SF). SC1 contained the highest vitamin E concentrations per μ thickness. To compare the distribution of the various vitamin E forms into the skin layers, the percentage of each form was expressed per its respective total. Most surprising was that the largest fraction of skin vitamin E following topical application was found in the deeper subcutaneous layers—the lowest layers, PD (40±15%) and D (36±15%), contained the major portion of the applied vitamin E forms. Although PD only represents about 16% of the total skin thickness, it contains sebaceous glands—lipid secretory organs, and, thus, may account for the vitamin E affinity for this layer. Hence, applied vitamin E penetrates rapidly through the skin, but the highest concentrations are found in the uppermost 5 microns.     

The phases of saturated 1-monoglycerides C14-C22

Phase behavior of a homologous series of saturated even 1-monoglycerides, starting with monomyristin, has been reviewed and their study extended to monoarachidin and monobehenin. The occurrence of sub α, α, β' and β polymorphs was confirmed for all compounds, except in the case of β' for monomyristin. It has been firmly established that there is a reversible sub α₂ ⇄ sub α₁ transition, (indicated by Malkin for monostearin) below the reversible sub α (sub α₁)⇄α transition, for C₁₈ through C₂₂ compounds; it occurs at about 50 C and is apparently almost independent of chain length. The sub α₂ to sub α₁ transformation is particularly sensitive to impurity and disappears for 1-monobehenin at about 10% 2-monobehenin as produced by heating at 96 C. Heats of transformation are, for β' and β crystal melting, about 50 cal/g; for α melting, about 35; for sub α → α transition, about 15 and for sub α₂ to sub α₁ transition about 3, which is several times as large as typical heats of melting of mesomorphic states. Diffraction data confirm the essential identity of all long spacing values and the occurrence of tilted chains for all polymorphs of a given compound. Much similarity is encountered between sub α and β' patterns. Sub α₂ and sub α₁ are difficult to distinguish by diffraction pattern.     

Lipid peroxidation in erythrocyte membranes: Cholesterol product analysis in photosensitized and xanthine oxidase-catalyzed reactions

The effects of singlet oxygen- and oxygen radical-induced lipid peroxidation on cell membrane integrity were compared, using the human erythrocyte ghost as a model system. Resealed ghosts underwent lipid peroxidation and lysis (release of trapped glucose-6-P) when irradiated in the presence of uroporphyrin (UP) or when incubated with xanthine (X), xanthine oxidase (XO) and iron. The UP-sensitized process was inhibited by azide but not by phenolic antioxidants, consistent with singlet oxygen (nonradical) involvement. This was confirmed by showing that the predominant photoproduct of membrane cholesterol was the 5α-hydroperoxide. Total hydroperoxide (LOOH) content in UP-photooxidized ghosts increased linearly during the prelytic lag and throughout the period of rapid lysis. Unlike the photoreaction, X/XO/iron-dependent peroxidation and lysis was inhibited by catalase, superoxide dismutase and phenolic antioxidants, indicating O₂ ⁻/H₂O₂ intermediacy and a free radical mechanism. Correspondingly, only radical reactions products of cholesterol were formed, notably the 7α-, 7β-hydroperoxide pair. Membranes lysis had a distinct lag as in photooxidation; however, the LOOH profile was more complex, with an initial lag followed by a sharp increase and then slow decline. X/XO/iron-induced lysis commenced when LOOH levels were 2–3 times higher than in photosensitized lysis, suggesting that the pathways of membrane lesion formation are different in the two systems. In low concentrations, ascorbate exacerbated the damaging effects of photoperoxidation, switching the reaction from primarily singled oxygen- to oxygen radical-dependence, as indicated by cholesterol product analysis.     

Combined effects of EFA deficiency and tumor necrosis factor-α on circulating lipoproteins in rats

Both tumor necrosis factor-α (TNF-α) and EFA deficiency (EFAD) have been established as causes of marked perturbations in lipid and lipoprotein metabolism. Excessive levels of circulating TNF-α can coexist with EFAD in various clinical disorders such as cystic fibrosis and type I diabetes. The present study therefore aimed to investigate their combined effects on lipid profile and lipoprotein composition by administering TNF-α to EFAD rats. Lipoprotein lipase (LPL), the ratelimiting enzyme in TG catabolism, was also measured in epididymal adipose tissue. EFAD, after a 4-wk period, induced significant increases in plasma TG (80%, P<0.001), total cholesterol (TC, 27%, P<0.025), and HDL-cholesterol (HDL-C, 62%). Two hours after the administration of TNF-α, a further rise in TG (43%, P<0.05) was noted in controls, but not EFAD animals. TC and HDL-C were unaffected by TNF-α treatment. In addition, TNF-α modified lipoprotein-lipid composition. VLDL and HDL₂ derived from EFAD rats were depleted in apolipoprotein (apo) E and apo A-II, and enriched in apo A-12 h after TNF-α administration. Finally, TNF-α decreased adipose tissue LPL activity in both control and EFAD animals. The TNF-α-induced inhibition was more marked in EFAD rats. The present results demonstrated that TNF-α can amplify or antagonize the effects of EFAD on lipid profile, lipoprotein composition, and LPL activity. These data also suggest that the host's nutritional status is a determining factor for the modulating effect of TNF-α on lipid metabolism.     

Evaluation of Lipophilic Antioxidant Efficacy in Vivo by the Biomarkers Hydroxyoctadecadienoic Acid and Isoprostane

The evaluation of antioxidant activity in vivo is difficult. In this study, the effects of dietary natural and synthetic antioxidants on the lipid peroxidation in mice were assessed using a biomarker, total hydroxyoctadecadienoic acid (tHODE). Biological samples such as plasma, erythrocytes, and tissues were first reduced and then saponified to convert various oxidation products of linoleates to tHODE. Subsequently, the absolute concentration of tHODE and its stereoisomer ratio, [9- and 13-(Z,E)-HODE)/[9- and 13-(E,E)-HODE], which is a measure of the hydrogen donor capacity of antioxidants, were determined by gas chromatography–mass spectrometry (GC–MS) analyses. These were then compared with total 8-iso-prostaglandin F_2α (t8-iso-PGF_2α) which was also assessed after reduction and saponification. Remarkable increases in tHODE and t8-iso-PGF_2α levels were observed in the plasma, erythrocytes, liver, and brain of mice that were fed an α-tocopherol (αT)-stripped (E-free) diet for 1 month when compared with those of mice that were fed a standard diet (αT = 0.002 wt%). When mice were fed for 1 month on an E-free diet supplemented with a lipophilic antioxidant (0.04 wt%), namely, αT, α-tocotrienol (αT3), γ-tocopherol (γT), or 2,3-dihydro-5-hydroxy-4,6-di-tert-butyl-2,2-dipentylbenzofuran (BO-653), a potent synthetic antioxidant, the increases of tHODE and t8-iso-PGF_2α in the plasma, erythrocytes, liver, and brain were suppressed to the levels lower than those of mice fed a standard diet. The (Z,E/E,E) HODE ratio was decreased in the plasma and erythrocytes of mice fed the E-free diet when compared with that in mice fed the standard diet. This stereo-isomeric ratio was significantly recovered by the addition of αT and BO-653. These results show that the tHODE level and the (Z,E/E,E) HODE ratio are useful biomarkers for the assessment of antioxidant capacity in vivo and that the antioxidant capacity decreased in the order: BO-653 > αT3 ≧ αT, γT, as assessed by tHODE levels from blood, liver, and brain.     

Gamma-irradiation of individual cholesterol oxidation products

Cholesterol and seven of its oxidation products in aqueous suspensions of multilamellar vesicles or sonicated aqueous suspensions were subjected individually to γ-radiation (10 KGy) at 0–4°C in air, N₂ or N₂O. All compounds underwent some changes under the influence of radiation. β-Epoxide (cholesterol 5β,6β-epoxide) and, to a much lesser extent, α-epoxide (cholesterol 5α,6α-epoxide) were converted in low yield to 6-ketocholestanol (5α-cholestan-3β-ol-6-one). 7β-Hydroxycholesterol (cholest-5-ene-3β,7β-diol) and, to a lesser extent, 7α-hydroxycholesterol (cholest-5-ene-3β,7α-diol) gave low yields of 7-ketocholestanol (5α-cholestan-3β-ol-7-one). The latter compound also was obtained by irradiation of 7-ketocholesterol (cholest-5-ene-3β-ol-7-one). 6-Ketocholestanol and 7-ketocholestanol are potential biomarkers for irradiated meat and poultry.     

Combination of TLC Blotting and Gas Chromatography–Mass Spectrometry for Analysis of Peroxidized Cholesterol

We have established a sensitive and convenient method for analysis of cholesterol hydroperoxides (Chol-OOHs) as trimethylsilyloxyl derivatives using diphenylpyrenylphosphine (DPPP)-thin-layer chromatography (TLC) blotting and gas chromatography–electron ionization–mass spectrometry/selected-ion monitoring (GC–EI–MS/SIM). Chol-OOH standards were prepared by photosensitized oxidation and azo radical-induced peroxidation of cholesterol. Trimethylsilyloxyl derivatives of cholesterol 5α-hydroperoxide (Chol 5α-OOH), cholesterol 7α-hydroperoxide (Chol 7α-OOH), and cholesterol 7β-hydroperoxide (Chol 7β-OOH) could be separated from one another in the SIM chromatogram using a fragment ion with elimination of trimethylsilanol from the molecular ion. This method was used to characterize peroxidized cholesterol from azo radical-exposed human low-density lipoprotein and UVA-irradiated human keratinocytes in the presence of hematoporphyrin. Finally, we succeeded in the quantification of each Chol-OOH isomer present in hairless mouse skin with and without UVA irradiation by use of β-sitosterol hydroperoxide as internal standard. The accumulation of Chol 5α-OOH with Chol 7α/βOOH in the skin indicates that singlet molecular oxygen (¹O₂) participated in the peroxidation of skin cholesterol, because Chol 5α-OOH is known to be a ¹O₂ specific cholesterol peroxidation product. We concluded that the combination of DPPP-TLC blotting and GC–EI–MS/SIM is useful for quantifying peroxidized cholesterol in biological samples and confirming the participation of ¹O₂ in oxidative stress.     

The relationship between fatty acid peroxidation and α-tocopherol consumption in isolated normal and transformed hepatocytes

The response of normal and transformed rat hepatocytes to oxidative stress was investigated. Isolated normal rat hepatocytes and differentiated hepatoma cells (the Fao cell line was derived from the Reuber H 35 rat hepatoma) in suspension were incubated with the ADP/Fe³⁺ chelate for 30 min at 37°C. Membrane lipid oxidation was assessed by measuring (i) free malondialdehyde (MDA) production by a high-performance liquid chromatography (HPLC) procedure, (ii) membrane fatty acid disappearance as judged by capillary gas chromatography, and (iii) α-tocopherol oxidation as determined by HPLC and electrochemical detection. The addition of iron led to increased MDA production in normal as well as in transformed cells, and to simultaneous consumption of polyunsaturated fatty acids (PUFA) and α-tocopherol. In addition, in Fao cells more α-tocopherol was consumed during lipid peroxidation while less PUFA was oxidized. Lipid peroxidation was lower in tumoral hepatocytes than in normal cells. This could be due to a difference in membrane lipid composition because of a lower PUFA content and a higher α-tocopherol level in Fao cells. During oxidation, Fao cells produced 1.5 to 2 times less MDA than normal cells, while in the tumoral cells the amount of oxidized PUFA having 3 or more double bonds was 7 to 8 times lower. Therefore, measuring MDA alone as an index of lipid peroxidation did not allow for proper comparison of the membrane lipid oxidizability of transformed cellsvs. the membrane lipid oxidizability of normal cells.     

Ascorbate-enhanced lipid peroxidation in photooxidized cell membranes: Cholesterol product analysis as a probe of reaction mechanism

Cholesterol was used as an in situ probe for studying mechanisms of lipid peroxidation in isolated erythrocyte membranes subjected to different prooxidant conditions. The membranes were labeled with [¹⁴C]cholesterol by exchange with prelabeled unilamellar liposomes and photosensitized with hematoporphyrin derivative. Irradiation with a dose of blue light resulted in thiobarbituric acid-detectable lipid peroxidation that was increased markedly by subsequent dark incubation with 0.5–1.0 mM ascorbate (AH⁻). Ascorbate-stimulated lipid peroxidation was inhibited by EDTA, desferrioxamine (DOX) and butylated hydroxytoluene (BHT), suggesting that the process is free radical in nature and catalyzed by membrane-bound iron. Thin layer chromatography and radiometric scanning of extracted lipids from photooxidized membranes revealed that the major oxidation product of cholesterol was the 5α-hydroperoxide (5α-OOH), a singlet oxygen adduct. Post-irradiation treatment with AH⁻/Fe(III) resulted in an almost-total disappearance of 5α-OOH and the preponderance of free radical oxidation products, e.g. 7-ketocholesterol, the epimeric 7α-/7β-hydroperoxides (7α-/7β-OOH) and their respective alcohols (7α-/7β-OH). EDTA, DOX and BHT inhibited the formation of these products, while catalase and superoxide dismutase had no effect. These results are consistent with a mechanism involving 1-electron reduction of photogenerated hydroperoxides to oxyl radical, which trigger bursts of free radical lipid peroxidation. Though generated in this system, partially reduced oxygen species, viz. superoxide, hydrogen peroxide and hydroxyl radical, appear to be relatively unimportant in the autoxidation process. Presented at the symposium “Free Radicals Antioxidants, Skin Cancer and Related Diseases” at the 78th AOCS Annual Meeting in New Orleans, LA, May 1987.     

Peroxide dependent and independent lipid peroxidation: Site-specific mechanisms of initiation by chelated iron and inhibition by α-tocopherol

Peroxidation of linoleic acid (LA) was catalyzed by Fenton reagent (H₂O₂, and Fe²⁺) in positively charged tetradecyltrimethylammonium bromide (TTAB) micelles, but not in negatively charged sodium dodecylsulfate (SDS) micelles. However, more hydroxyl radicals formedvia the Fenton reaction were trapped byN-t-butyl-α-phenylnitrone (PBN) in SDS micelles than in TTAB micelles. Generation of linoleic acid alkoxy (LO) radicals by Fe²⁺ via reductive cleavage of linoleic acid hydroperoxide (LOOH) resulted in peroxidation of LA and formation of PBN-LO· adducts in SDS micelles, but not in TTAB micelles. This LOOH dependent lipid peroxidation could be catalyzed in TTAB micelles in the presence of a negatively charged iron chelator, nitrilotriacetic acid (NTA). LO radicals formed by the LOOH dependent Fenton reaction were also trapped by PBN at the surface of TTAB micelles in the presence of NTA, but not in its absence. The consumption of a spin probe, 16-(N-oxyl-4,4′-dimethyloxazolidin-2-yl)stearic acid (16-NS) during the LOOH dependent Fenton reaction in the presence of NTA was higher in TTAB micelles of LA than in those of lauric acid (LauA), although the rates and amounts of LO radicals formed in the two types of fatty acid micelles were similar. The rates of 5-NS consumption in LA and LauA micelles were almost the same, and were lower than the rate of 16-NS in LA micelles. NTA-Fe²⁺ initiated peroxidation of LA in TTAB micelles without a lag time in the presence of LOOH, but after a lag period, peroxidation occurred without LOOH. α-Tocopherol inhibited peroxidation of LA catalyzed by Fenton reagent by scavenging OH radicals in TTAB micelles. In contrast, α-tocopherol enhanced free Fe²⁺ induced LOOH dependent lipid peroxidation through the regeneration of Fe²⁺ in SDS micelles. However, it inhibited NTA-Fe²⁺ induced LOOH dependent lipid peroxidation in TTAB micelles. The rate and amount of α-tocopherol oxidized by the Fe²⁺ induced, H₂O₂ dependent Fenton reaction were almost the same in TTAB micelles of LA and LauA. The oxidation of α-tocopherol by the NTA-Fe²⁺ induced, LOOH dependent Fenton reaction was greater and faster in LA micelles than in LauA micelles, although the rates of LO radical production in the two types of micelles were the same. During NTA-Fe²⁺ induced, LOOH dependent lipid peroxidation, α-tocopherol inhibited more effectively the consumption of 16-NS than 5-NS. The results are discussed in relation to the location of iron, the unsaturated bonding region of LA, the OOH group of LOOH, the radical trapping site of PBN, the spin sites of 5-NS and 16-NS, and the phenolic hydroxyl group of α-tocopherol in micelles with different charges. Based on a paper presented at the Symposium on Metals and Lipid Oxidation, held at the AOCS Annual Meeting in Baltimore, MD, April 1990.