Zellweger syndrome

ABSTRACT

Oxiclative damage to plasmenyl-type lipids contributes to decreased membrane barrier function, loss of membrane structure and formation of nonlamellar defects in membrane bilayers. Previous results from this laboratory have shown that membrane-soluble sensitizers (e.g. zinc phthalocyanine and bacteriochlorophyll a) mediate the photooxidation of palmitoyl plasmenylcholine (1-O-alk-1'-Z-enyl-2-palmitoyl-sn-glycero-3-phosphocholine; PPlsC) vesicles with the subsequent creation of lamellar defect structures, vesicle contents leakage and membrane-membrane fusion. Because plasmalogen lipids are significant components of sarcoplasma and myelin membranes, we sought to characterize the products of their photooxidation. This study focuses on the photooxidation of PPlsC vesicles in the presence of the water-soluble sensitizer, aluminum phthalocyanine tetrasulfonate (AlPcS^sup 4-^^sub 4^). Attack of photogenerated singlet oxygen on the 1-O-alkenyl ether linkage of PPlsC lipids was expected to generate dioxetane-and ene-type photoproducts. The products formed during continuous aerobic irradiation (28 mW/cm^sup 2^, (610 nm) of PPlsC vesicles in the presence of AlPcS^sup 4-^^sub 4^ were separated via reverse-phase high-performance liquid chromatography (HPLC) with electrochemical detection (ECD) or evaporative light-scattering detection (ELSD). Photooxidized dipalmitoyl-phosphatidylcholine-cholesterol vesicles (control) were used to optimize the HPLC-ECD conditions, using 7[alpha]-hydroperoxy-cholesterol as standard. HPLC-ECD was found to be most sensitive for PPlsC hydroperoxides, whereas HPLC-ELSD was more sensitive for nonhydroperoxide photoproducts. The three major photoproducts formed during vesicle irradiation were isolated via preparative HPLC and then characterized by ^sup 1^H-nuclear magnetic resonance and mass spectrometry. 1-Formyl-2-palmitoyl-sn-glycero-3-phosphocholine and 1-hydroxy-2-palmitoyl-sn-glycero-3-phosphocholine were identified as dioxetane cleavage products that coeluted at ~3 min. The second fraction (retention time [R^sub T^] = 48 min) was identified as a PPlsC allylic hydroperoxide. The third photoproduct, elating at R^sub T^ = 64 min, is tentatively identified as an oxidation product arising from allylic hydroperoxide degradation via Hock rearrangement or free radical decomposition.

Abbreviations: AlPcS^sup 4-^^sub 4^, aluminum phthalocyanine tetrasulfonate; Chol, cholesterol; DPPC, dipalmitoylphosphatidylcholine; ECD, electrochemical detection; ELSD, evaporative light-scattering detection; HPLC, high-performance liquid chromatography; MALDI-MS, matrix-assisted laser desorption-ionization mass spectrometry; NMR, nuclear magnetic resonance; ^sup 1^O2, singlet oxygen; 7[alpha]-OOH, 7[alpha]-hydroperoxycholesterol; PBS, phosphate-buffered saline; PlsC, plasmenylcholine; PlsE, plasmenylethanolamine; PPlsC, palmitoyl plasmenylcholine (1-O-alk-1'-Z-enyl-2-palmitoyl-sn-glycero-3-phosphocholine); PUFA, polyunsaturated fatty acids; QLS, quasielastic light scattering; R^sub T^, retention time.

INTRODUCTION

Plasmalogens are widely distributed mammalian phospholipids containing a Z-1'-alkenyl ether linkage at the sn-1 position of the glycerophospholipid backbone (Scheme 1). Plasmalogens comprise as much as 18% of the total phospholipid mass in humans (1). The choline and ethanolamine derivatives of plasmalogen are known as plasmenylcholine (PlsC) and plasmenylethanolamine (PlsE), respectively. PlsC are found in high abundance in human cardiac tissues (as much as 36-41% of the total glycerophosphocholine in human heart is PlsC [2]) and are known to be an important depot for arachidonic acid (3,4). Human nervous system tissues, especially brain and myelin sheath, contain high concentrations of PlsE (PlsE constitute as much as 70-80% of the phospholipids in brain white matter and myelin [5]), which often contains high levels of polyunsaturated fatty acids (PUFA). A biosynthetic pathway separate from that responsible for diacylglycerophosphocholine production exists for plasmalogen formation (6). Genetic defects in this pathway are known to be the basis for peroxisomal disorders such as Zellweger's syndrome (7). In spite of their extensive distributions in mammalian tissues, the biological role of plasmalogens remains unclear. Three hypotheses have been proposed to rationalize their existence. The first is based on the role of plasmalogens in arachidonate storage and signal transduction (8). Gross and coworkers (9,10) have proposed, based on two-dimensional nuclear magnetic resonance (NMR) evidence showing that PlsC lipids have a different glycerol backbone conformation with respect to the membrane interface, that the Z-1-alkenyl ether bond provides a unique conformational motif at the membrane surface that is selectively recognized by arachidonic acid-liberating phospholipase A^sub 2^ enzymes. The second hypothesis, supported by mounting evidence from several laboratories (11-29), suggests that the alkenyl ether bond is present to serve as a sacrificial trap for reactive oxygen species that would otherwise attack the PUFA moieties located at the nearby sn-2 position. Unique membrane protein activities observed in the presence of plasmalogens have also led to the suggestion that they may be involved in plasmalogen-specific membrane protein interactions with these alkenyl (vinyl) ether lipids (1).

Histological studies of myelin samples from multiple sclerosis patients have found that the oligodendricytes within these tissues degenerate with time, causing severe neural damage (3033). Because oligodendricytes myelinate the nerves of the central nervous system, it is conceivable that oxidative degradation of the high PlsE concentrations found in these cells by invading leukocytes may lead to loss of membrane asymmetry and irreversible myelin damage, thus degrading the central nervous system function. As a potential model for PlsE oxidation in oligodendritic cells, we sought to characterize the oxidative degradation products of the related plasmalogen lipid, palmitoyl plasmenylcholine (PPlsC). Because aluminum phthalocyanine tetrasulfonate (AlPcS^sup 4-^^sub 4^)-sensitized formation of singlet oxygen (^sup 1^O2) (AlPcS^sup 4-^^sub 4^-sensitized formation of ^sup 1^O2 occurs with a quantum efficiency of ~0.3 [34]) is known to increase the permeability of semisynthetic PPlsC membrane vesicles (35,36), it was anticipated that photosensitized oxidation of PPlsC could provide useful insights into the degradative pathways affecting oligodendritic cells that may be important for oxidative damage mechanisms and disease progression in multiple sclerosis. PPlsC was chosen as the model lipid for our investigations because the vesicles formed by these lipids are more convenient experimental systems than the H^sub II^ phases formed by PlsE species.

Attack of PPlsC by ^sup 1^O2 is expected to produce two primary oxidation products (Scheme 2), a 1,2-dioxetane intermediate and an allylic 1-hydropcroxy intermediate from 2 + 2 (dioxetane) and 2 + 4 (ene) cycloaddition reactions, respectively (37-40). Thermal degradation of the labile dioxetane would produce fatty aldehyde, 1-formyl-2-palmitoyl-sn-glycero-3-phosphocholine (1), and its hydrolysis product-2-palmitoyl-sn-glycero-3-phosphocholine (sn-1 lysolipid, 2)-as stable products. Production of an allylic hydroperoxide product (3), resulting from the ene reaction, would also be expected. The occurrence of these processes was not known for PlsC, and the relative yield of these anticipated photoproducts was unpredictable based on the location of the reactive alkenyl ether bond at the interface of polar and nonpolar environments. This study was directed at determining the relative importance of these oxidation reactions in plasmalogen vesicle degradation. We also sought to characterize the primary degradation products of PPlsC photooxidation.

MATERIALS AND METHODS

Materials. PPlsC (1-O-alk-1'-Z-enyl-2-palmitoyl-sn-glycero-3-phosphocholine) was prepared as described elsewhere (35,41). Lyophilized dipalmitoylphosphalidylcholine (DPPC, Avanti Polar Lipids, Alabaster, AL), cholesterol (Chol, Sigma, St. Louis, MO) and chloroaluminum phthalocyanine tetrasulfonate (AlPcS^sup 4-^^sub 4^ , Porphyrin Products, Logan, UT) were purchased and used without further purification. Spectrophotometric grade chloroform was filtered through anhydrous Na^sub 2^CO^sub 3^ to remove adventitious HCl immediately before use. All solvents used for high-performance liquid chromatography (HPLC) (e.g. methanol, acetonitrile and 2-propanol) were of spectrophotometric grade and filtered through a 0.45 [mu]m membrane before use. Phosphate-buffered saline (PBS, 20 mM phosphate in 100 mM NaCl, pH 7.4) was prepared in 18 Mohm/cm H2O (obtained from a Barnstead NANOpure system) and treated with Chelex 100 to remove any polyvalent metal ion contamination. 7[alpha]-Hydroperoxycholesterol (7[alpha]-OOH), kindly supplied by W. Korytowski (Jagiellonian University, Krakow, Poland), was used as a hydroperoxide standard for HPLC with electrochemical detection (ECD) calibration.

Liposome preparation. Liposomes were prepared by hydrating lipid films, prepared by evaporation of lipid solutions in deacidified CHCl^sub 3^. After evaporating CHCl^sub 3^ under a gentle N^sub 2^ stream, the lipid residue was further dried under a 50 [mu]m Hg vacuum for at least 8 h. The powder was then hydrated in PBS via 10 freeze-thaw-vortex cycles and the resulting multilamellar vesicle solution extruded 10 times through two stacked 100 nm polycarbonate filters at 50[degrees]C, 200 psi N^sub 2^. DPPC-Chol liposomes were 5.0 mM DPPC:3.6 mM Chol; PPlsC liposomes were 5.0 mM unless specified otherwise. The AlPcS^sup 4-^^sub 4^ sensitizer concentration was 10 [mu]M after addition of an aliquot of AlPcS^sup 4-^^sub 4^ stock solution. Liposome sizes were determined by quasielastic light scattering (QLS) using a Coulter N4-Plus instrument. The mean liposome size and distribution were calculated using the manufacturer's supplied software (version 1.1). Liposomes formed in this manner were typically irradiated immediately after preparation; however, they remained stable for up to 1 week upon storage at 4[degrees]C (as monitored by QLS).

Liposome irradiation. AlPcS^sup 4-^^sub 4^-sensitized liposomes were placed in a jacketed beaker held at 25 + or - 1[degrees]C and continuously stirred during aerobic irradiation of the entire sample surface from above with a quartz-halogen lamp. Light fluency at the suspension surface was approximately 100 mW/ cm^sup 2^, measured radiometrically (Molectron Detector 5100). After irradiation, 0.1 mM ethylenediaminetetraacetic acid was added to the samples before extraction with 2:1 CHCl^sub 3^-CH^sub 3^OH. The extracts were then dried immediately under Ar and stored at -40[degrees]C before analysis.

Iodometric determination of lipid hydroperoxides. The total concentration of hydroperoxides formed during irradiation was determined by iodometric assay (42). Triiodide (I^sup -^^sub 3^) generated by the reaction of lipid hydroperoxides with excess KI was quantified spectrophotometrically at 353 nm ([epsilon]^sub 353^ = 22 500 M^sup -1^ cm^sup -1^).

High-performance liquid chromatography. Reverse-phase HPLC analysis was performed with a Spherisorb C^sub 18^ column (4.6 x 150 mm, 5 [mu]m particles) protected with Waters C^sub 18^ Guard Pack(TM) prefilters. The photooxidation products were determined using reductive-mode ECD (Model CC-5/LC-4C/DA-5, BioAnalytical Systems, West Lafayette, IN) with a gold amalgam (Au/Hg) electrode operating at -300 mV (Ag/AgCl reference). The electrode set potential was based on literature data (43), which suggested that this potential provides maximum detector response. The mobile phase, consisting of either 90:8:2 CH^sub 3^OH-CH^sub 3^CN-H2O with 0.25 nM NaClO^sub 4^ (44) or 81:10.5:8.5 CH^sub 3^OH-CH^sub 3^CN-H2O with 10 mM NH^sub 4^OAc and 1.0 mM NaClO^sub 4^ (45), was delivered isocratically at 1.7 mL/ min. Extraordinary precautions were undertaken to avoid O2 contamination in the mobile phase by sparging it continuously with he that was first passed through an O2-scrubbing catalyst tower. All gas and mobile phase delivery tubing was stainless steel. Before sample injection, the system was equilibrated with eluent for several hours at -50 mV and then adjusted to the working potential. Samples were dissolved in 2-propanol (also sparged with He) just before injection. Signals from the detector were collected and analyzed using the manufacturer's software. Chromatographic peak detection was also achieved using a Sedex 55 evaporative light-scattering detector (ELSD, Sedere, Alfortville, France). Data analysis was performed in these cases using EZChrom(TM), version 6.8 (Scientific Software, Pleasanton, CA). The mobile phase for ELSD detection was identical to that used for ECD, except that electrolytes were not added.

Matrix-assisted laser desorption-ionization, time-of-flight mass spectrometry. The matrix-assisted laser desorption-ionization mass spectrometry (MALDI-MS) results were obtained using a PerSeptive Biosystems (Framingham, MA) Voyager mass spectrometer. This instrument uses a nitrogen laser (337 nm excitation) for ionization with a time-of-flight mass analyzer. The sample and matrix ([alpha]-cyano-4-hydroxy-cinnamic acid) were mixed in a ratio of 1:1 [mu]L on the sample plate. This mixture was allowed to air dry before analysis. All samples were analyzed in positive-ion mode using an accelerating potential of 28 kV. External calibration was achieved using a standard containing insulin (bovine), m/z = 2867.80 (z = +2) and 5734.59 (z = +1). Calibration was done before the experiment.

NMR spectroscopy. ^sup 1^H-NMR spectra were recorded with a Varian INOVA spectrometer at 300 MHz using CDCl^sub 3^ as the solvent. Chemical shifts are reported relative to tetramethylsilane as the internal standard.

RESULTS

Our initial experiments focused on establishing the AIPcS^sup 4-^^sub 4^-sensitized irradiation conditions needed to detect the formation of lipid photooxidation products. These experiments initially used DPPC-Chol vesicles irradiated under aerobic conditions (46), where a variety of cholesterol hydroperoxide products arc known to form (47). Because DPPC lacks activated allylic C-H bonds, these experiments were expected to produce only well-characterized Chol photooxidation products that would allow us to standardize the HPLC separation conditions. They would also serve as a benchmark for validating our modified HPLC-ECD method, which differed from the method reported by Girotti and coworkers (44,45) in that a gold amalgam cathode was used instead of a hanging mercury drop cathode. Depending on the irradiation conditions used, three or four major peaks were detected at retention times (R^sub T^) similar to those described previously (44,47). The expected products of ^sup 1^O2 attack on Chol (Scheme 3) are 5[alpha]-OOH and 6[alpha]-OOH. The 7[alpha]-OOH and 7[beta]-OOH arise from free radical oxidation of Chol and allylic rearrangement of 5[alpha]-OOH (29). Our HPLC-ECD analysis of the cholesterol hydroperoxides formed by AlPcS^sup 4-^^sub 4^-sensitized oxidation under these conditions (Fig. 1) indicates that cholesterol hydroperoxide detection was reproducible with a sensitivity limit of approximately 10 pmol. Peak 1 was positively identified as 7[alpha]-OOH by comparison with an authentic sample of 7[alpha]-OOH as a calibration standard. Peaks 2 and 3 were assigned as 5[alpha]-OOH and 6[alpha]-OOH, respectively, based on the previously reported HPLC-ECD response of cholesterol hydroperoxides (47).

Extension of these conditions to PPlsC-Chol vesicles, however, was not successful because of overlapping peaks resulting from oxidation of both lipid components. Nevertheless, these experiments provided the first clear qualitative evidence that PPlsC is oxidized more rapidly than Chol-it was noted that the PPlsC peak was depleted very rapidly relative to the Chol peak. All subsequent experiments used pure PPlsC vesicles (diameter, 100 + or - 5 nm) to avoid the Chol-PPlsC oxidation product coelution problem. This strategy also enabled PPlsC photoproduct isolation and characterization. These vesicles were irradiated at 100 mW/cm^sup 2^ for 0.5-3 h before determining the total hydroperoxide concentration in solution by iodometric assay. These assays indicated that lipid conversions ranged between 17% and 40% of the total PPlsC concentration depending on the irradiation time.

Detection of PP;sC photooxidation products during HPLC separation was achieved using two different detectors in separate chromatographic experiments. Reducible species formed by photooxidation were detected by reductive mode HPLC-ECD. These experiments indicated the presence of one main peak at R^sub T^ = 51 min and a few minor components that were just above our detection limit (Fig. 2). HPLC-ELSD analysis was used to complement the ECD detection scheme. Three peaks were observed at R^sub T^ = 2.9-3.5 min, 48 min and 64 min (Fig. 3), in addition to a poorly detectable peak at R^sub T^ = 84 min that was assigned to unreacted PPlsC based on the R^sub T^ of nonirradiated control samples. The peak at R^sub T^ = 48 min appears to be the same species detected by HPLC-ECD at R^sub T^ = 51 min. We believe that the electrolytes required for ECD detection are responsible for the increase in observed R^sub T^ relative to ELSD detection when all other parameters are held constant.

The effects of irradiation time on the formation of PPlsC photooxidation products in PBS buffer are shown in Fig. 3. The maximum amount of lipid hydroperoxides, detected by iodometric analysis, was produced during irradiation periods up to 1.5 h. As the irradiation time increased from 30 min to 1.5 h, the lipid hydroperoxide peak eluting at R^sub T^ ~ 48 min also increased-as did the dioxetane degradation product peaks at early R^sub T^ values (R^sub T^ ~ 3 min). In contrast, little change in peak intensities was observed as a function of irradiation time for the photoproduct eluting at R^sub T^ = 64 min. Although the dioxetane degradation product yields increased with further irradiation up to 3 h, the intensity of the lipid hydroperoxide peak actually decreased with increasing irradiation time, suggesting that this photoproduct undergoes decomposition under these reaction conditions.

The effects of ^sup 1^O2 inhibitors and promoters on PPlsC photooxidation were also studied. The results of these experiments are shown in Fig. 4. PPlsC vesicle irradiations in D^sub 2^O, where the ^sup 1^O2 lifetime is longer (48), as well as in the presence of histidine (a competitive trap for ^sup 1^O2) or sodium azide (a ^sup 1^O2 quencher), were compared for their ability to generate PPlsC photooxidation products relative to control solutions in PBS. In contrast to the expected increase in ^sup 1^O2 degradation products in the presence of D^sub 2^O, we observed no significant changes in the lipid hydroperoxide (R^sub T^ ~ 48 min) or dioxetane (R^sub T^ = 3 min) photoproduct profile upon addition of D^sub 2^O. This suggests that the presence of D^sub 2^O has no impact on the efficiency of either the ene or dioxetane reaction pathways. Addition of azide or histidine to the liposome solutions before irradiation, however, did affect the PPlsC photooxidation reaction. As expected, the amounts of lipid hydroperoxide and dioxetane degradation products both decreased during the irradiation in the presence of histidine, leaving some unreacted PPlsC in the sample (R^sub T^ ~ 90 min). The most dramatic effect was observed with lipid hydroperoxide formation, where less than 10% of this photoproduct was formed in the presence of 40 mM histidine. Azide had a similar but much less pronounced effect on photoproduct formation yields.

The PPlsC photooxidation products were then isolated and characterized by MALDI-MS (49) (Fig. 5) and ^sup 1^H-NMR (Table 1, Fig. 6) after irradiation for 3 h. Three fractions were collected from preparative HPLC-ELSD separation (i.e. multiple injections fractionated at high initial concentrations of PPlsC; Fraction 1, 2-5 min; Fraction 2, 41-52 min; Fraction 3, 58-68 min) and analyzed after solvent removal.

DISCUSSION

Attack of PPlsC by ^sup 1^O2, generated via AlPcS^sup 4-^^sub 4^ sensitization (Scheme 2), is expected to produce fatty aldehyde, 1-formyl-2-palmitoyl-sn-glycero-3-phosphocholine (1), its hydrolysis product, 2-palmitoyl-sn-glycero-3-phosphocholine (sn-1 lysolipid, 2), and 1-O-1[degrees]-hydroperoxy-hexadec-2'(Z)-enyl-2-palmitoyl.sn-glycero-3-phosphocholine (3) via the dioxetane and ene reaction pathways. Hock rearrangement (16,50) of 3 would be expected to produce an orthoformate product 4, which that would give two different fatty aldehydes and the lysolipid products 1 and 2 upon subsequent hydrolysis (Scheme 4). Allylic hydroperoxide 3 can also decompose via free radical processes, as shown in Scheme 5.

Mass spectra of the photooxidation products detected by HPLC-ELSD at R^sub T^ = 2.9 and 3.5 min (i.e. Fraction 1) were assigned to the initial products of dioxetane-mediated degradation of PPlsC: 1 (observed m/z = 525, expected m/z = 524.3, [M + 1]) and 2 (observed m/z = 497, expected m/z = 496.3, [M + 1]), respectively (Fig. 5a). The appearance of three high mass peaks is attributed to homodimerization and heterodimerization of the dioxetane products (i.e. m/z = 993.6, [2-2 + 1]; m/z = 1021.6, [1-2 + 11; m/z = 1049.6, [1-1 + 1]) that presumably occur during sample preparation for MALDI analysis.[dagger] The NMR spectrum of Fraction 1 does not indicate the presence of proton resonances attributable to an alkenyl ether bond (Fig. 6a) because chemical shifts are typically observed in the 5.6-6.0 ppm range for protons [alpha] to the alkenyl ether oxygen atom. The quantity of these photoproducts appeared to increase with prolonged irradiation-from 28% at 0.5 h to 42% at 3 h excitation-as would be expected for the accumulation and thermal degradation of dioxetane intermediates; however, the experimental uncertainties in these measurements prevent rigorous interpretation of these relative reaction efficiencies.

The peak eluting near 50 min (i.e. R^sup ELSD^^sub T^ = 48 min and R^sup ECD^^sub T^ = 51 min) is assigned to allylic 1-hydroperoxide 3, generated via an ene reaction (Scheme 2). This assignment is based on three lines of evidence. First, the quantity of this photoproduct gradually decreased with increasing irradiation time (i.e. 69%, 63% and 49% of the total lipid species present at 0.5, 1.5 and 3 h irradiation, respectively; Table 2), presumably because of secondary degradation reactions such as those illustrated in Scheme 5. Second, ^sup 1^H-NMR analysis of this fraction indicates the presence of at least one set of olefinic proton signals having a chemical shift value that is consistent with an allylic hydroperoxide moeity (Table 1, Fig. 6b). Finally, the ECD response of this photoproduct was significant and similar in magnitude to that of the 7[alpha]-OOH cholesterol hydroperoxide standard. Taken together, these data strongly suggest that a PPlsC hydroperoxide was formed; however, this conclusion is only indirectly supported by the mass spectrometric data (Fig. 5b). The major ions detected are consistent with sn-1 lysolipid (m/z = 497) and an intermediate assignable to 1-allylic hemiacetal 6, 1-epoxide 5 or the Hock acetal rearrangement product (observed m/z = 734, [M + 1], expected m/z = 733.6). We conclude that the latter species arise from hydroperoxide degradation during matrix contact and ionization.[double dagger]

The third fraction, eluting at R^sup ELSD^^sub T^ = 64 + or - 4 min, is assigned to the product of allylic hydroperoxide degradation via Hock rearrangement or free radical decomposition based on the presence of alkenyl ether proton signals in the NMR spectrum and the observation of m/z = 733 and 755 peaks [733 + Na+] in MALDI-MS (Figs. 6, Fraction 3c and 5c, respectively). It is not possible to distinguish between these possibilities based on our MS data because compounds 5 and 6 and the Hock product are isomers.[sec]

Photooxidation reactions performed in the presence of ^sup 1^O2-modulating agents, such as histidine, NaN^sub 3^ and D^sub 2^O, also support our conclusions about the photoproduct identities. Although no significant changes in photoproduct yields were observed when the photooxidation reactions were performed in D^sub 2^O (Table 2), irradiation in the presence of histidine or NaN^sub 3^ produced significant changes in the photoproduct profile. Histidine greatly suppressed the formation of 1, 2 and 3 and slightly increased the recovery of unreacted PPlsC (R^sup ELSD^^sub T^ = 84 min). Histidine also appeared to increase the yield of the material eluting at 64 min, even though no evidence of hydroperoxide formation was detected in these samples by iodide assay. If this peak is indeed due to a Hock degradation product, it suggests that histidine either inhibits the dioxetane reaction pathway more effectively than the ene pathway (a selectivity pattern that would not be expected) or it very efficiently promotes the degradation of 3 via proton transfer to initiate the rearrangement. Additional studies are needed to clarify this issue. Our results also show that sodium azide affects the photooxidation process by slightly inhibiting the formation of 1, 2 and 3 and completely inhibiting the formation of 4. The presence of significant quantities of intact PPlsC (R^sub T^ = 84 min) further indicates that the azide effectively quenches the formation of ^sup 1^O2 under these conditions.

CONCLUSIONS

Our data indicate that PPlsC photooxidation produces lysolipid species (1-formyl-2-palmitoyl-sn-glycero-3-phosphocholine along with its hydrolysis product 2-palmitoyl-sn-glycero-3-phosphocholine) and an allylic hydroperoxide species assigned as 1-O-1'-hydroperoxy-hexadec-2'(Z)-enyl-2-palmitoyl-sn-glycero-3-phosphocholine as primary intermediates from dioxetane- and ene-type reactions. Decomposition of these primary photoproducts to give secondary products arising from allylic hydroperoxide degradation-either by Hock rearrangement or by alkenyl ether epoxidation-was also observed. The oxidized plasmalogen species observed in these studies may also form during the degradative processes of plasmalogen-rich tissues occurring in various pathological conditions. Further characterization of these reactions is needed to enable correlations between these model studies and their possible occurrence in vivo.

Acknowledgements-We thank Dr. Witold Korytowski for the generous sample of 7[alpha]-cholesterol hydroperoxide, Adrian Bott of Bioanalytical Systems for technical assistance with the ECD unit, Karl Wood of the Purdue Mass Spectrometry Center for collection of the mass spectral data and the National Institutes of Health (GM 55266 and FIRCA Grant TW01251) for financial support.

[para] Posted on the websile on 2 September 2003

[dagger]We base this conclusion on the expectation that any dimeric products formed during photooxidation would have eluted at longer R^sub T^ and, therefore, would not have been collected in the sample eluting at R^sub T^ = 2-5 min. It is likely that dimerization was promoted by the acidic matrix used for MALDI, a complication that may be avoidable using electrospray mass spectrometry (31).

[double dagger]The acidic MALDI matrix may catalyze allylic hydroperoxide degradation via Hock rearrangement to produce orthoformate 4, as well as hydrolysis of 4 to form 1.

[sec]Allylic rearrangement of 3 before free radical decomposition would produce a 3-allylic alcohol isomer of 6. This species, if formed, would also possess an [alpha]-alkenyl proton with a chemical shift in the 5.6-6.0 ppm range.

REFERENCES

1. Lee, T.-c. (1998) Biosynthesis and possible biological functions of plasmalogens. Biochim. Biophys. Acta 1394, 129-145.

2. Sugiura, T. and K. Waku (1987) Composition of alkyl ether-linked phospholipids in mammalian tissues. In Platelet Activating Factor and Related Lipid Mediators (Edited by F. Snyder), pp. 55-85. Plenum Press, New York.

3. Mueller, H. W., A. D. Purdon, J. B. Smith and R. L. Wykle (1983) 1-O-Alkyl-linked phosphoglycerides of human-platelets: distribution of arachidonate and other acyl residues in the ether-linked and diacyl species. Lipids 18, 814-819.

4. Gross, R. W. (1984) High plasmalogen and arachidonic acid content of canine myocardial sarcolemma: a fast atom mass spectroscopic and gas chromatography mass spectroscopic characterization. Biochemistry 23, 158-165.

5. Horrocks, L. A. and M. Sharma (1982) Plasmalogens and O-alkyl glycero-phospholipids. Vol. IV (Edited by J. N. Hawthorne and G. B. Ansell), pp. 51-93. Elsevier Biomedical Press, Amsterdam.

6. Nagan, N. and R. A. Zoeller (2001) Plasmalogens: biosynthesis and functions. Prog. Lipid Res. 40, 199-229.

7. Poulos, A., A. Bankier, K. Beckman, D. Johnson, E. F. Robertson, P. Sharp, L. Sheffield, H. Singh, S. Usher and G. Wise (1991) Glyceryl ethers in peroxisomal disease. Clin. Genet. 39, 13-25.

8. Wolf, R. A. and R. W. Gross (1985) Identification of neutral active phospholipase C which hydrolyzes choline glycerophospholipids and plasmalogen selective phospholipase A2 in canine myocardium. J. Biol. Chem. 260, 7295-7303.

9. Han, X. and R. W. Gross (1990) Plasmenylcholine and phosphatidylcholine membrane bilayers possess distinct conformational motifs. Biochemistry 29, 4992-4996.

10. Chen, X., X. Han and R. W. Gross (1993) Dynamics of binary mixtures of plasmenylcholine/arachidonic acid and phosphatidylcholine/ arachidonic acid: a study using fluorescence and NMR spectroscopy. Biochim. Biophys. Acta 1149, 241-248.

11. Zoeller, R. A., O. H. Morand and C. R. H. Raetz (1988) A possible role for plasmalogens in protecting animal cells against photosensitized killing. J. Biol. Chem. 263, 11590-11596.

12. Zoeller, R. A., A. C. Lake, N. Nagan, D. P. Gaposchkin, M. A. Legner and W. Lieberthal (1999) Plasmalogens as endogenous antioxidants: somatic cell mutants reveal the importance of the vinyl ether. Biochem. J. 338, 769-776.

13. Zoeller, R. A., T. J. Grazia, P. LaCamera, J. Park, D. P. Gaposchkin and H. W. Farber (2002) Increasing plasmalogen levels protects human endothelial cells during hypoxia. Am. J. Physiol. 283, H671-H679.

14. Khaselev, N. and R. C. Murphy (1999) Susceptibility of plasmenyl glycerophosphoethanolamine lipids containing arachidonate to oxidative degradation. Free Radic. Biol. Med. 26, 275-284.

15. Khaselev, N. and R. C. Murphy (2000) Peroxidation of arachidonate containing plasmenylglycerophosphocholine: facile oxidation of esterified arachidonate at carbon-5. Free Radic. Biol. Med. 29, 620-632.

16. Khaselev, N. and R. C. Murphy (2000) Structural characterization of oxidized phospholipid products derived from arachidonate-containing plasmenyl glycerophosphocholine. J. Lipid Res. 41, 564-572.

17. Murphy, R. C. (2001) Free radical-induced oxidation of arachidonoyl plasmalogen phospholipids: antioxidant mechanism and precursor pathway for bioactive eicosanoids. Chem. Res. Toxicol. 14, 463-472.

18. Zommara, M., N. Tachibana, K. Mitsui, N. Nakatani, M. Sakono, I. Ikeda and K. Imaizumi (1995) Inhibitory effect of ethanolamine plasmalogen on iron- and copper-dependent lipid peroxidation. Free Radic. Biol. Med. 18, 599-602.

19. Sindelar, P. J., Z. Guan, G. Dallner and L. Ernster (1999) The protective role of plasmalogens in iron-induced lipid peroxidation. Free Radic. Biol. Med. 26, 318-324.

20. Schulz, R. (1996) Plasmalogens, nitroxide free radicals, and ischemiareperfusion injury in the heart. In Advances in Lipobiology, Vol. I (Edited by R. W. Gross), pp. 193-214. JAI Press; Tokyo, Japan

21. Engelmann, B., C. Brautigam and J. Thiery (1994) Plasmalogen phospholipids as potential protectors against lipid peroxidation of low density lipoprotein. Biochem. Biophys. Res. Commun. 204, 1235-1242.

22. Jurgens, G., A. Fell, G. Ledinski, Q. Chen and F. Paltauf (1995) Delay of copper-catalyzed oxidation of low density lipoprotein by in vitro enrichment with choline or ethanolamine plasmalogens. Chem. Phys. Lipids 77, 25-31.

23. Felde, R. and G. Spiteller (1995) Plasmalogen oxidation in human serum lipoproteins. Chem. Phys. Lipids 76, 259-267.

24. Jira, W. and G. Spiteller (1996) Plasmalogens and their oxidative degradation products in low and high density lipoprotein. Chem. Phys. Lipids 79, 95-100.

25. Spiteller, G. (1996) Enzymic lipid peroxidation: a consequence of cell injury. Free Radic. Biol. Med. 21, 1003-1009.

26. Hofer, G., D. Lichtenberg, G. M. Kostner and A. Hermetter (1996) Oxidation of fluorescent glycero- and sphingophospholipids in human plasma lipoproteins: alkenylacyl subclasses are preferred targets. Clin. Biochem. 29, 445-450.

27. Brosche, T. (1997) Plasmalogen phospholipids: facts and theses to their antioxidant qualities. Arch. Gerontol. Geriatr. 25, 73-81.

28. Brosche, T. and D. Platt (1998) The biological significance of plasmalogens in defense against oxidative damage. Exp. Gerontol. 33, 363-369.

29. Brosche, T., D. Platt and B. Knopf (2002) Decreased concentrations of serum phospholipid plasmalogens indicate oxidative burden of uraemic patients undergoing haemodialysis. Nephron 90, 58-63.

30. McFarland, H. I., A. A. Lobito, M. M. Johnson, J. T. Nyswaner, J. A. Frank, G. R. Palardy, N. Tresser, C. P. Genain, J. P. Mueller, L. A. Matis and M. J. Lenardo (1999) Determinant spreading associated with demyelination in a nonhuman primate model of multiple sclerosis. J. Immunol. 162, 2384-2390.

31. von Budingen, H. C., S. L. Hauser, A. Fuhrmann, C. B. Nabavi, J. I. Lee and C. P. Genain (2002) Molecular characterization of antibody specificities against myelin/oligodendrocyte glycoprotein in autoimmune demyelination. Proc. Natl. Acad. Sci. USA 99, 8207-8212.

32. Brett, R. and M. G. Rumsby (1994) Susceptibility of myelin glycerophospholipids and sphingolipids to oxidative attack by hydroxyl free radicals as measured by the thiobarbituric acid test. Neurochem. Int. 24, 241-251.

33. LeVine, S. M. (1997) Iron deposits in multiple sclerosis and Alzheimer's disease brains. Brain Res. 760, 298-303.

34. Edrei, R., B. Gottfried, J. E. van Lier and S. Kimel (1998) Sulfonated phthalocyanines: photophysical properties, in vitro cell uptake and structure-activity relationships. J. Porph. Phthalo. 2, 191-198.

35. Anderson, V. C. and D. H. Thompson (1992) Triggered release of hydrophilic agents from plasmalogen liposomes using visible light or acid. Biochim. Biophys. Acta 1109, 33-42.

36. Thompson, D. H., O. V. Gerasimov, J. J. Wheeler and V. C. Anderson (1996) Triggerable plasmalogen liposomes: improvement of system efficiency. Biochim. Biophys. Acta 1279, 25-34.

37. Frimer, A. A. (1979) The reaction of singlet oxygen with olefins: the question of mechanism. Chem. Rev. 79, 359-387.

38. Frimer, A. A. and L. M. Stephenson (1985) The singlet oxygen ene reaction. Singlet O2, Vol. II: Reaction Modes and Products. Part I (Edited by A. A. Frimer). P67-91 CRC Press, Boca Raton, FL.

39. Maranzana, A., G. Ghigo and G. Tonachini (2000) Diradical and peroxirane pathways in the [[pi]2 + [pi]2] cycloaddition reactions of ^sup 1^[Delta]^sub g^ dioxygen with ethene, methyl vinyl ether, and butadiene: a density functional and multireference perturbation theory study. J. Am. Chem. Soc. 122, 1414-1423.

40. Gollnick, K. and K. Knutzenmies (1991) Dye-sensitized photooxygenation of 2,3-dihydrofurans: competing [2 + 2] cycloadditions and ene reactions of singlet oxygen with a rigid cyclic enol ether system. J. Org. Chem. 56, 4017-4027.

41. Shin, J. and D. H. Thompson (2003) Direct synthesis of plasmalogens from allyl substituted glycerols. J. Org. Chem. 68, 6760-6766.

42. Hicks, M. and J. M. Gembicki (1979) Spectrophotometric method for the determination of lipid hydroperoxides. Anal. Biochem. 99, 249-253.

43. Korytowski, W., G. J. Bachowski and A. W. Girotti (1993) Analysis of cholesterol and phospholipid hydroperoxides by high-performance liquid-chromatography with mercury drop electrochemical detection. Anal. Biochem. 213, 111-119.

44. Korytowski, W., G. J. Bachowski and A. W. Girotti (1992) Photoperoxidation of cholesterol in homogeneous solution, isolated membranes, and cells: comparison of the 5-[alpha]-hydroperoxides and 6-[beta]-hydroperoxides as indicators of singlet oxygen intermediacy. Photochem. Photobiol. 56, 1-8.

45. Korytowski, W., P. G. Geiger and A. W. Girotti (1995) High-performance liquid-chromatography with mercury cathode electrochemical detection: application to lipid hydroperoxide analysis. J. Chromatogr. B 670, 189-197.

46. Anderson, V. C. and D. H. Thompson (1992) Photoinduced morphology changes in plasmalogen liposomes using visible light. In Macromolecular Assemblies in Polymeric Systems, ACS Symposium, Vol. 493 (Edited by P. Stroeve and A. C. Balzacs), pp. 154-170. ACS Press, Washington, DC.

47. Korytowski, W., G. J. Bachowski and A. W. Girotti (1991) Chromatographic-separation and electrochemical determination of cholesterol hydroperoxides generated by photodynamic-action. Anal. Biochem. 197, 149-156.

48. Fu, Y., P. D. Sima and J. R. Kanofsky (1996) Singlet-oxygen generation from liposomes: a comparison of 6[beta]-cholesterol hydroperoxide formation with predictions from a one-dimensional model of singlet-oxygen diffusion and quenching. Photochem. Photobiol. 63, 468-476.

49. Wagner, F., S. Rottem, H.-D. Held, S. Uhlig and U. Zabringer (2000) Ether lipids in the cell membrane of Mycoplasma fermentans. Eur. J. Biochem. 267, 6276-6286.

50. Havrilla, C. M., D. L. Hachey and N. A. Porter (2000) Coordination (Ag+) ion spray-mass spectrometry of peroxidation products of cholesterol linoleate and cholesterol arachidonate: high-performance liquid chromatography-mass spectrometry analysis of peroxide products from polyunsaturated lipid autooxidation. J. Am. Chem. Soc. 122, 8042-8055.

51. Khaselev, N. and R. C. Murphy (2000) Electrospray ionization mass spectrometry of lysoglycerophosphocholine lipid subclasses. J. Am. Soc. Mass Spectrom. 11, 283-291.

David H. Thompson*1, Halina D. Inerowicz1,2, Jason Grove1 and Tadeusz Sarna3

1 Department of Chemistry, Purdue University, West Lafayette, IN;

2 Department of Chemistry, Technical University of Gdansk, Gdansk, Poland and

3 Department of Biophysics, Jagiellonian University, Krakow, Poland

Received 31 December 2002; accepted 23 June 2003

* To whom correspondence should be addressed at: Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, IN 47907-2084, USA. Fax: 765-496-2592; e-mail: davethom@purdue.edu

Copyright American Society of Photobiology Oct 2003
Provided by ProQuest Information and Learning Company. All rights Reserved