Original Article

Accumulation of N-acyl-ethanolamine phospholipids in rat brains during post-decapitative ischemia: a ³¹P NMR study

Correspondence to: Harald S. Hansen

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

Phosphorus-31 nuclear magnetic resonance (³¹P NMR) spectroscopy has been used to study accumulation of N-acyl-ethanolamine phospholipids in rat brains during post-decapitative ischemia. Lipids were extracted from rat brain homogenates and the extracts were thoroughly washed with aq. potassium ethylenediaminetetraacetic acid (EDTA). The lower organic phases were isolated and evaporated to dryness under a stream of nitrogen and the lipids were redissolved in CDCl₃–CH₃OH–H₂O 100.0:29.9:5.2 (v/v/v) for NMR analysis. Increasing the period of post-decapitative ischemia resulted in an accumulation of two signals in the NMR spectra at 0.18 and 0.22 ppm (relative to the chemical shift of 1,2-diacyl-sn -glycero-3-phosphocholine (PC_DIACYL) at -0.84 ppm). These signals were identified as originating from 1,2-diacyl-sn -glycero-3-phospho-(N-acyl)-ethanolamine (NAPE_DIACYL) and 1-(1'-alkenyl)-2-acyl-sn -glycero-3-phospho-(N-acyl)-ethanolamine (NAPE_PLAS), respectively, by spiking with authentic materials. Additionally, the identification was verified by thin-layer chromatography, which also showed the accumulation of N-acyl-ethanolamine phospholipids. The use of K-EDTA instead of the commonly used Cs-EDTA in the preparation of the NMR samples allowed the separation of the chemical shifts of N-acyl-ethanolamine phospholipids from those of the ethanolamine phospholipids. Moreover, the chemical shift of cardiolipin was moved from 0.15 ppm observed with Cs-EDTA to about 0.31 ppm with K-EDTA.

The present study demonstrates that it is possible to detect and quantify post-decapitative accumulation of NAPE subclasses (NAPE_DIACYL and NAPE_PLAS) in rat brains by the use of ³¹P NMR spectroscopy.—Moesgaard, B., J. W. Jaroszewski, and H. S. Hansen. Accumulation of N -acyl-ethanolamine phospholipids in rat brains during post-decapitative ischemia: a ³¹P NMR study. J. Lipid Res. 1999. 40: 515–521.

Supplementary key words: N-acyl-phosphatidylethanolamine, 1,2-diacyl-sn-glycero-3-phospho-(N-acyl)-ethanolamine, 1-(1'-alkenyl)-2-acyl-sn-glycero-3-phospho-(N-acyl)-ethanolamine, cardiolipin, rat brain, ischemia, phosphorus nuclear magnetic resonance, plasmalogen

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

The accumulation of N-acyl-ethanolamine phospholipids (NAPE) under the condition of post-decapitative ischemia has earlier been reported in rat brains using a chromatographic method (1). This rare lipid is of biological interest because it is the precursor for N -acyl-ethanolamines, some of which are ligands for the cannabinoid receptors (2), as well as because this class of phospholipids by itself may have membrane stabilizing properties and their accumulation may indicate neuronal injury (2) (3). Phosphorus-31 nuclear magnetic resonance (³¹P NMR) spectroscopy is a convenient method for the determination of phospholipids (4), and the chemical shifts of many phospholipids extracted from biological tissues, including cardiolipin (CL), 1,2-diacyl-sn -glycero-3-phosphoethanolamine (PE_DIACYL), 1-(1'-alkenyl)-2-acyl-sn-glycero-3-phosphoethanolamine (PE_PLAS), ethanolamine lysophospholipids (LPE), 1-2-diacyl-sn -glycero-3-phosphocholine (PC_DIACYL), 1-(1'-alkenyl)-2-acyl-sn -glycero-3-phosphocholine (PC_PLAS), 1-alkyl-2-acyl-sn -glycero-3-phosphocholine (PC_AA), choline lysophospholipids (LPC), sphingomyelin (SPH), serine phospholipids (PS), inositol phospholipids (PI), and phosphatidic acid (PA) have been characterized (4) (5) (6) (7) (8) (9) (10) (11) (12) (13). To our knowledge, NAPE has not previously been detected in biological samples by use of ³¹P NMR. It was our objective to identify and quantify NAPE in extracts of animal tissue by the use of ³¹P NMR.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

Animals
For the study of post-decapitative ischemia-induced accumulation of NAPE, Sprague-Dawley rats (10 days old, mixed sexes) were purchased from Møllergaard Breedings, Lille Skensved, Denmark. The rats were decapitated and the whole heads were wrapped in foil and immediately incubated at 37 °C for different periods of time. Control rats were decapitated and the brains were excised within 1 min and kept in CHCl₃–CH₃OH 2:1 (v/v) on ice, until homogenized and extracted (within 15 min) as described below.

For the identification of the ³¹P NMR signals of NAPE-subclasses and cardiolipin, Sprague-Dawley rats (8 weeks old, mixed sexes) were obtained from our own animal stable. The rats were decapitated and the brains were excised within 1 min and kept in CHCl₃–CH₃OH 2:1 (v/v) on ice, until homogenized and extracted (within 15 min) as described below.

Materials
All commercial chemicals were used as received without further purification. CH₃OH, CHCl₃, CuSO₄, KOH (all analytical grade reagents) and thin-layer chromatography (TLC) plates (Silica Gel 60) were obtained from Merck (Darmstadt, Germany). EDTA (free acid), K-EDTA (dipotassium salt), CsOH, and CL (bovine heart) were purchased from Sigma Chemical Co. (St. Louis, MO). Aqueous solutions of Cs-EDTA were obtained by titration of EDTA (free acid) with CsOH to pH 6.0. Aqueous solutions of K-EDTA were adjusted to pH 6.0 with aqueous KOH. CDCl₃ (99.8%) was from the Cambridge Isotope Laboratories (Andover, MA). Ammonia solution and orthophosphoric acid were obtained from Aldrich (Steinheim, Germany) and Riedel-de Haën AG (Seelze-Hannover, Germany), respectively. The mixture of N -decanoyl species of the NAPE subclasses [1,2-diacyl-sn -glycero-3-phospho-(N-acyl)-ethanolamine (NAPE_DIACYL), 1-(1'-alkenyl)-2-acyl-sn -glycero-3-phospho-(N - acyl)- ethanolamine (NAPE_PLAS), and 1-alkyl-2-acyl-sn -glycero-3-phospho-(N -acyl)-ethanolamine (NAPE_AA)] were synthesized from bovine brain ethanolamine phospholipids (PE) [containing a mixture of PE _DIACYL, PE_PLAS, and PE_AA (14)] and decanoyl chloride using a modification of the procedures described by Epps et al. (15) and Sugiura et al. (16). PE and decanoyl chloride were purchased from Sigma Chemical Co. (St. Louis, MO). 1,2-Dioleoyl-sn-glycero-3-phospho-(N-oleoyl)-ethanolamine (NAPE _TRIOLEOYL) was obtained from Avanti Polar Lipids Inc. (Alabaster, AL).

Lipid extraction
The brains were excised from the rat heads and lipids were extracted by a modified (6) (10) procedure of Folch, Lees, and Sloane Stanley (17). The brains (weighing ~700 mg) were homogenized on ice in 20 times amount (v/w) of CHCl₃–CH₃OH 2:1 (v/v) with Ultra Turrax T25 (Janke & Kunkel IKA-Labortechnik, Staufen, Germany). The homogenates were placed on ice for 2 h to settle and filtered through a funnel packed with glass wool. The extracts, 0.2 ml of either 0.1 M aq. K-EDTA or 0.1 M aq. Cs-EDTA (pH 6.0) per ml of CHCl₃–CH₃OH 2:1 (v/v), were added and mixed thoroughly before being transferred to long glass tubes and left at 4°C overnight in order to separate the phases. The lower organic phases were isolated and evaporated to dryness under a stream of nitrogen and the total amount of lipids was determined. The lipids were redissolved in CHCl₃–CH₃OH 2:1 (v/v) and kept at -20°C until used. Henceforward the lipids will be referred to as either the K- or the Cs-salts of the lipids depending on whether the lipid extracts have been prewashed with K-EDTA or Cs-EDTA.

³¹P NMR analysis
Appropriate volumes of the lipid extracts, containing 20–25 mg of lipids (K- or Cs-salts), were transferred to small vials with airtight screwcaps. The solvents were evaporated under a stream of nitrogen and the lipids were redissolved in 700 µl of a mixture of CDCl₃, CH₃OH, and either H₂O, 0.2 M aq. K-EDTA (pH 6.0), or 0.2 M aq. Cs-EDTA (pH 6.0) 100.0:29.9:5.2 (v/v/v). The NMR samples were stored tightly closed at -20°C until used.

The standards of phospholipids (CL, NAPE _TRIOLEOYL, and a mixture of the three NAPE-subclasses) were washed with either K-EDTA or Cs-EDTA according to Branca et al. (18). Thus about 0.5 mg of CL or NAPE _TRIOLEOYL, or about 1.0 mg of the mixture of NAPE-subclasses was dissolved in 500 µl of CHCl₃, mixed with 200 µl of CH₃OH and 50 µl of 0.2 M aq. K- or Cs-EDTA (pH 6.0) and vortexed for about 5 min. The solutions were left at 4°C overnight in order to separate the phases. The organic phases containing the phospholipids were isolated and evaporated under a stream of nitrogen. The K- or Cs-salts of the phospholipids were redissolved in CHCl₃ and kept at -20°C, until mixed with rat brain lipids and used for the preparation of NMR samples as described above.

The NMR spectrometer used was a Bruker AMX 400 WB system operating at 161.98 MHz for ³¹P. Analytical samples were placed in standard (5 mm) NMR sample tubes spun at 20 Hz during the accumulation of data. Proton broad-band decoupling was effected by the Waltz-16 decoupling sequence. The ²H NMR signal of CDCl₃ was used for field-frequency locking. Homogeneity of the magnetic field was adjusted for the ¹H signal of CHCl₃ to a typical line width of 0.6 Hz (measured at half height). Probe temperature was stabilized at 25°C.

Each NMR spectrum was the sum of 4096 free induction decays (FIDs) with total accumulation time of 2 h. The FIDs were typically collected as 16384 complex data points during an acquisition period of 0.82 s. The spectral width was 10000 Hz. The spin-flip angle used was 60° (8.5 µs), and the relaxation delay was 1.0 s. Occasionally, spectra with the relaxation delay of 12.0 s were obtained. The longitudinal relaxation times (T₁) were measured using the inversion recovery method and were found to be 1.5 s and 1.3 s for the signals of PE_DIACYL and PE_PLAS, respectively. The chemical shifts were standardized relative to the chemical shift of the signal of the naturally occurring phospholipid PC _DIACYL set to -0.84 ppm (6). Data processing was performed on Lorentz-Gauss transformed spectra with a line broadening of -1.0 Hz and a Gaussian broadening of 0.5. The area of the signals was determined using the standard Bruker software.

Thin-layer chromatography
Two samples, each containing about 0.5 mg of lipids (K-salts) extracted from a control brain and from a brain exposed to post-decapitative ischemia for 6 h at 37 °C were evaluated by TLC with CHCl₃–CH₃OH–aq. NH₃ 80:20:2 (v/v/v) as the developing solvent. CL or NAPE_TRIOLEOYL (10 µg samples) were used as standards. The spots were visualized by spraying with a 10% (w/w) solution of copper sulfate in 85% orthophosphoric acid and charred at 200 °C.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

³¹P NMR spectra of lipid extracts from rat brains exposed to post-decapitative ischemia
Examples of ³¹P NMR spectra of rat brain lipids (K-salts) are shown in Figure 1. The top (A) spectrum was obtained with the lipids from a control brain (lipids extracted immediately after the decapitation) and the bottom (B) spectrum was obtained with the lipids extracted from a rat brain exposed to 6 h of post-decapitative ischemia at 37°C. The spectra demonstrate that during the 6-h period of post-decapitative ischemia there was an accumulation of phospholipids giving the signals labeled 1 (0.43–0.52 ppm), 3 (0.22 ppm), 4 (0.18 ppm), 11 (-0.27 ppm), and 12 (-0.38 ppm), while the other signals remain at a rather constant level during the period of observation.

View larger version (17K): [in this window] [in a new window]   Figure 1. 31P NMR spectra (161.98 MHz) of phospholipids extracted from 10-day-old rat brains excised and homogenized either immediately after decapitation (A) or after the exposure to post-decapitative ischemia for 6 h at 37°C (B). The lipid extracts were prewashed with K-EDTA, the solvents were evaporated under a stream of nitrogen, and the lipids were redissolved in CDCl3–CH3OH–H2O 100:29.9:5.2 (v/v/v). The NMR spectra were recorded at 25°C (4096 transients). The region from -0.5 to 0.5 ppm is plotted with four times higher vertical scale. The chemical shifts are standardized relative to the signal of PCDIACYL set at -0.84 ppm. The assignments of the signals are: 1 (the region 0.43–0.52 ppm), LPE-subclasses and PA; 2, CL; 3, NAPEPLAS; 4, NAPEDIACYL; 5, PEAA; 6, PEPLAS; 7, PEDIACYL; 8, PS; 9, SPH; 10, PI; 11, LPC; 12, unknown; 13, PCAA; 14, PCDIACYL.

Identification of signals of NAPE-subclasses and of CL
In ³¹P NMR spectra of lipid extracts CL is very often, if not always, reported to have resonance at 0.18 ppm relative to PC_DIACYL at -0.84 ppm in the CHCl₃–CH₃OH–aq. Cs-EDTA or K-EDTA system, and independent of whether the lipid extracts have been prewashed with KCl, K- or Cs-EDTA (4) (6) (7) (8) (9) (10) (11) (12) (13). However, in the present work, ³¹P NMR spectra of the K-salts of the lipids extracted from freshly excised rat brains in the CHCl₃–CH₃OH–H₂O system displayed no signal at 0.18 ppm, but the appearance of a signal at this chemical shift was observed during post-decapitative ischemia (Figure 1). Spiking of the solutions with NAPE_TRIOLEOYL or CL and redetermination of the ³¹P NMR spectra (not shown) showed that the signal observed at 0.18 ppm originated from NAPE_DIACYL (= NAPE_TRIOLEOYL), and that CL had resonance at about 0.34 ppm. This is in contradiction to the earlier reports stating that the chemical shift of CL is 0.18 ppm irrespective of the counterion K⁺ or Cs⁺ (4) (6) (12). Therefore we decided to compare ³¹P NMR spectra of brain lipids that were prepared as K- or Cs-salts by carefully washing the lipid extracts with either K- or Cs-EDTA (pH 6.0). The results are shown in Figure 2. Spectra in (A) and (D) were obtained with lipids extracted from rat brains using Cs-EDTA and K-EDTA, respectively. The remaining spectra were obtained for the brain lipids with added authentic CL [ Figure 2 (B) and (E)] or a mixture of NAPE subclasses [ Figure 2 (C) and (F)]. It is apparent that the chemical shift value of CL is about 0.15 ppm when using Cs-EDTA (B) and about 0.31 ppm when using K-EDTA [ Figure 2 (E) and other experiments (spectra not shown)]. We cannot explain the discrepancy between these findings and the earlier reported value of the chemical shift of CL of 0.18 ppm, regardless of whether the countercation is Cs⁺ or K⁺ (4) (6) (12). There are small differences among the chemical shift values of CL in Figure 1 (0.34 ppm) and in Figure 2 (0.31 ppm), which could be due to the presence of K-EDTA in the NMR solvent system. Additionaly, there is a small difference between the reported chemical shift of CL of 0.18 ppm in the Cs-EDTA system (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) and the value of 0.15 ppm observed in the present work (Figure 2). This could be due to differences in the concentration of lipids in the NMR samples or differences in the solvent composition in the NMR samples.

View larger version (22K): [in this window] [in a new window]   Figure 2. 31P NMR spectra (161.98 MHz) of lipids extracted from 8-week-old rat brains, spiked with standard phospholipids. Either Cs-EDTA (A–C) or K-EDTA (D–F) was used in the washing procedure of the extracts. The solvents in the lipid extracts were evaporated under a stream of nitrogen and the lipids (total 21–22 mg) were redissolved in 700 µl CDCl3–CH3OH–aq.Cs-EDTA (A–C) or aq. K-EDTA (D–F) (0.2 M, pH 6.0) 100:29.9:5.2 (v/v/v). (A) and (D): rat brain lipids. (B) and (E): rat brain lipids plus 0.5 mg CL. (C) and (F): rat brain lipids plus 1 mg of N -decanoyl-ethanolamine phospholipids (a mixture of the three NAPE subclasses). The spectra were recorded at 25°C (4096 transients). The chemical shifts are standardized relative to the signal of PCDIACYL set at -0.84 ppm. The assignments of signals are as in Figure 1.

For spiking lipid extracts with NAPE [ Figure 2 (C) and (F)], a material synthesized from decanoyl chloride and bovine brain PE was used. Bovine brain PE fraction is stated by Sigma to contain about 60% of PE_PLAS. The synthesis was therefore expected to produce about 40% NAPE_DIACYL and 60% NAPE_PLAS. As seen in Figure 2 (F), NAPE_DIACYL and NAPE_PLAS gave rise to two signals, at about 0.16 and 0.19 ppm, respectively. Additionally, a minor signal at about 0.22 ppm was observed upon addition of the synthetic mixture of NAPE-subclasses. This minor signal may arise from 1-alkyl-2-acyl subclass of NAPE (NAPE_AA), if the commercial bovine brain PE contained not only PE_DIACYL and PE_PLAS as reported by the supplier, but also PE_AA. Indeed, Malewicz and Baumann (14) estimated the composition of a commercial bovine brain PE product by ³¹P NMR in CDCl₃–CH₃OD–D₂O 50:50:15 (v/v/v). Their results demonstrate that the commercial bovine brain PE consisted of 64.0 ± 1.2 mol % PE_PLAS, 31.8 ± 0.5 mol % PE_DIACYL, and 4.2 ± 0.5 mol % PE_AA. The relative areas of the three signals of NAPE_PLAS, NAPE_DIACYL, and NAPE_AA observed in a spectrum (not shown) of lipids added to the synthetic mixture of NAPE subclasses was 61:34:5. This is indeed in good accordance with the results of Malewicz and Baumann (14). When comparing spectra (C) and (F) in Figure 2, one will notice that NAPE is not seen as separate signals in the spectrum of the Cs-salts of the lipids. ³¹P NMR spectra (not shown) of mixtures of Cs-salts of the phospholipids PC, PS, SPH, CL, and NAPE and of cholesterol demonstrated that the resonances of the NAPE subclasses coincide with the corresponding resonances of PE subclasses (i.e., NAPE_DIACYL coincides with PE_DIACYL, etc.). This explains the apparent lack of separate signals from the different NAPE subclasses in Figure 2 (C). However, a change in the peaks of ethanolamine phospholipids can be seen in Figure 2 (C) as compared to (A) and (B) upon spiking with N -acyl-ethanolamine phospholipids.

Comparing the spectra shown in Figure 1 and Figure 2 one will notice minor variations in the chemical shifts and line widths of the phospholipids. These variations may, as mentioned above, be due to the use of H₂O instead of K-EDTA, minor variations in the NMR solvent compositions, and differences in the lipid concentration. It has been shown previously that the NMR solvent hydration as well as the amount of lipids affects the chemical shifts (6). However, these factors and the presence of K-EDTA in the NMR solvent system appear not to be of major importance for obtaining good quality ³¹P NMR spectra as long as the lipid extracts have been thoroughly prewashed with K-EDTA.

Tentative assignment of the other phospholipid resonances in the ³¹P NMR spectra of rat brain lipid extracts
We have made a tentative assignment of the remaining resonances in the ³¹P NMR spectra (Figure 1), based on earlier studies of phospholipids (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (19) and individual lipid classes of whole rat brains (20). The signals in the region 0.43–0.52 ppm (labeled 1) may include LPE-subclasses and PA. A ³¹P NMR spectrum (not shown) of a rat brain lipid extract spiked with PA, dissolved in CDCl₃–CH₃OH–H₂O 100.0:29.9:5.2 (v/v/v) showed a resonance of PA at 0.52 ppm. Signal 2 (0.34 ppm) is derived from CL and signals 3 and 4 (0.22 and 0.18 ppm) are derived from NAPE_PLAS and NAPE_DIACYL, respectively, according to our own data. Signals 6 (0.08 ppm) and 7 (0.03 ppm) are derived from PE_PLAS and PE_DIACYL, respectively. We propose that the signal 5 (0.12 ppm) arises from PE_AA, because the chemical shift separation of the three signals is in accordance with the spectrum of bovine brain PE (containing PE_AA, PE_PLAS, and PE_DIACYL) published by Malewicz and Baumann (14). That the signal at about 0.12 ppm corresponds to PE_AA has already been proposed by Metz and Dunphy (13) and Edzes et al. (19). A resonance at 0.12 ppm has earlier been shown to be derived from dihydro-SPH by Byrdwell et al. (21), but the identification was performed on lipid extracts from eye lenses and dihydro-SPH has not yet been reported to occur in brain tissue. Signals 8 (-0.03 ppm), 9 (-0.09 ppm), 10 (-0.25 ppm), and 11 (-0.27 ppm) are proposed to originate from PS, SPH, PI, and LPC, respectively. The compound causing signal 12 (-0.38 ppm) is unknown, and this signal is not seen in the spectra of lipids isolated from freshly excised brains. Signal 13 (-0.78 ppm) is tentatively assigned as originating from PC_AA, and the signal 14 which is assumed to originate from PC_DIACYL is used as the internal reference set at -0.84 ppm.

Accumulation of NAPE in rat brains during post-decapitative ischemia
Accumulation of NAPE subclasses during the observation period of post-decapitative ischemia was followed by ³¹P NMR spectroscopy as discussed below. The accumulation was additionally investigated by TLC, as shown in Figure 3. The figure shows an accumulation of the NAPE subclasses (the individual subclasses are not separated) during the 6-h period of post-decapitative ischemia.

View larger version (50K): [in this window] [in a new window]   Figure 3. TLC of rat brain lipids. The plate was eluted with CHCl3–CH3OH–aq. NH3 80:20:2 (v/v/v), sprayed with 10% (w/w) copper sulfate in 85% orthophosphoric acid and charred at 200°C. Lane (A): lipid extract of a brain exposed to 6 h of post-decapitative ischemia (0.5 mg); lane (B): NAPEDIACYL standard (10 µg); lane (C): CL standard (10 µg); lane (D): lipid extract prepared immediately after decapitation (0.5 mg).

The conditions used for the determination of the ³¹P NMR spectra were optimized for the maximal signal-to-noise ratio for a given experimental time. Thus, the spectra were recorded with 60° pulses and the relaxation delay of 1.0 s (the interpulse interval of 1.82 s). Consequently, signals of various phospholipids were saturated to a minor degree (T₁ relaxation times of 1.5 s and 1.3 s determined for PE_DIACYL and PE_PLAS correspond to the observation of 83% and 86% of the actual signal intensity, respectively). The possible differences in T₁ relaxation times of various phospholipids will therefore lead to slight discrepancies between the relative integrals and molar ratios. Moreover, Lorentz-Gauss transformation was applied to the spectra in order to achieve optimal resolution of the resonances. However, as the T₁ and the apparent transverse (T₂*) relaxation times of the individual phospholipids in different samples are expected to be similar (the total concentration of the phospholipids in the NMR samples was fairly constant), the ratios between the integral intensities of the resonances of a particular compound in different spectra will reflect the molar ratios satisfactorily. To assess the effect of the relaxation delay on the ratio between the integral intensity of the signals of NAPE subclasses and the total integral intensity of all lipid signals, a spectrum of lipids extracted as K-salts from the brain of a 10-day-old rat, exposed to post-decapitative ischemia for 6 h, was recorded with a relaxation delay of 12 s instead of 1 s (total accumulation time over 14.5 h). In this fully relaxed spectrum, the ratio between the integral of the NAPE_DIACYL signal and the total lipid signals was identical within experimental error to that observed in the spectrum recorded with the relaxation delay of 1 s. This shows that the effects of the possible variations in T₁ of the individual phospholipids are minor and do not affect the relative integral intensities of the NAPE_DIACYL signals in the (almost fully relaxed) spectra recorded with the relaxation delay of 1.0 s to any appreciable degree. Thus, the latter spectra are satisfactory to follow changes in the content of NAPE subclasses in the rat brain lipid extracts by ³¹P NMR.

Figure 4 shows the changes in the integral intensity of NAPE_DIACYL resonance as a function of time. The integral intensity of the resonance was expressed in % of the total area of the ³¹P resonances of all phospholipids in the sample. During the 6 h of post-decapitative ischemia, the relative integral intensity of the NAPE_DIACYL signal changed from below 0.09% (detection limit for area determination) to 1.4% (Figure 4). In addition, the ratio of NAPE_PLAS/NAPE_DIACYL was determined to be 0.53 ± 0.11 (n = 6). This ratio is similar to the ratio of PE_PLAS/PE_DIACYL, found to be 0.60 ± 0.06 (n = 3). Assuming that the resonance at 0.12 ppm (signal 5) originates from PE_AA, the ratio PE_DIACYL/PE_PLAS/PE_AA is (60.2 ± 2.3)/(35.9 ± 2.3)/(3.9 ± 0.2) (n = 3).

View larger version (12K): [in this window] [in a new window]   Figure 4. Post-decapitative ischemia accumulation of NAPEDIACYL in rat brains as measured by 31P NMR spectroscopy of lipid extracts in CDCl3–CH3OH–H2O 100:29.9:5.2 (v/v/v). After being decapitated, the heads were incubated at 37°C for various periods of time before the brains were excised, homogenized, and extracted. The areas of the NMR signals of NAPEDIACYL were determined as the percentage of the total area of the phospholipid signals. For each time point, two rat brains were analyzed and the line is drawn through the averages.

After 1 h of post-decapitative ischemia, the formation of NAPE_DIACYL amounted to about 0.5% of the total ³¹P NMR signal (integral intensity). From the observed ratio of NAPE_PLAS/NAPE_DIACYL of 0.53 ± 0.11, the production of total NAPE-subclasses amounts to 0.8% within 1 h and 2% within 6 h of post-decapitative ischemia. This result is in agreement with that reported by Natarajan, Schmid, and Schmid (1), who found a linear increase of NAPE subclasses within 60 min of post-decapitative ischemia (at 37°C) in 10-day-old rats up to 0.5% of total phospholipids determined by a chromatographic assay.

Conclusion
NAPE has for many years been known to occur in plant tissue, where its formation has been suggested to serve a cytoprotective role (for a recent paper see ref. (22)). It is now generally recognized that this phospholipid is also produced in mammalian tissue, especially in the heart and brain, under pathophysiological conditions such as ischemia (1) (2) (3) (23) (24). Under such conditions, the Ca²+ level in the cells is elevated, and consequently the enzyme N-acyl-transferase, which catalyzes the formation of NAPE by transferring an acyl group from the sn-1 position of phospholipids to the amino group of ethanolamine phospholipids, is activated leading to the accumulation of NAPE. Recently, it has been reported that NAPE may be a normal but minor constituent of rat brain lipids (23). Furthermore, NAPE formation may be induced under the condition of glutamate neurotoxicity (25) (26). As glutamate excitotoxicity may be involved in the pathogenic mechanisms of chronic neurodegenerative diseases like Parkinson's disease, Huntington's disease, and Alzheimer's disease (27), the formation of NAPE may be increased in these diseases. The present work demonstrates that ³¹P NMR spectroscopy can be used for the determination and quantification of NAPE in brain lipids, when the lipids are prepared as K-salts by careful washing of the extracts with K-EDTA. Such determination cannot be carried out with the Cs-salts of lipids, where the NAPE subclass signals coincide with those of PE subclasses. Due to the possible involvement of NAPE in glutamate neurotoxicity and chronic neurodegenerative diseases, ³¹P NMR spectroscopy may be highly useful in studies of animal models of neurodegenerative diseases. The detection limit of the ³¹P NMR analysis of NAPE subclasses is subject to substantial improvement by use of improved instrumentation and via extending the time of data accumulation.

	ACKNOWLEDGMENTS

This study has been supported by grants from the Danish Medical Research Council and Novo Nordisk Foundation. The NMR equipment used in this work was provided via grants from the Alfred Benzon Foundation and PharmaBiotec Research Center. Finnancial support from PharmaBiotec Research Center and technical assistance of Jytte Palmgren-Salomonsson are gratefully acknowledged.

Manuscript received January 27, 1998; and in revised form October 5, 1998.

Abbreviations: CL, cardiolipin; EDTA, ethylenediaminetetraacetic acid; FID, free induction decay; LPC, choline lysophospholipids; LPE, ethanolamine lysophospholipids; NAPE, N-acyl-ethanolamine phospholipids; NAPE_AA, 1-alkyl-2-acyl-sn - glycero -3 -phospho-(N -acyl)-ethanolamine; NAPE_DIACYL, 1,2-diacyl-sn -glycero-3-phospho-(N -acyl)-ethanolamine; NAPE_PLAS, 1-(1'-alkenyl)-2-acyl-sn -glycero -3 -phospho-(N -acyl)-ethanolamine; NAPE_TRIOLEOYL, 1,2-dioleoyl-sn -glycero -3 -phospho-(N -oleoyl)-ethanolamine; NMR, nuclear magnetic resonance; PA, phosphatidic acid; PC, choline phospholipids; PC_AA, 1-alkyl-2-acyl-sn-glycero-3-phosphocholine; PC_DIACYL, 1,2-diacyl-sn-glycero-3-phosphocholine; PC_PLAS, 1-(1'-alkenyl)-2-acyl-sn -glycero-3-phosphocholine; PE, ethanolamine phospholipids; PE_AA, 1-alkyl-2-acyl-sn -glycero-3-phosphoethanolamine; PE_DIACYL, 1, 2-diacyl-sn -glycero-3-phosphoethanolamine; PE_PLAS, 1-(1'-alkenyl)-2-acyl-sn -glycero-3-phosphoethanolamine; PI, inositol phospholipids; PS, serine phospholipids; SPH, sphingomyelin; T₁, longitudinal NMR relaxation time; T₂*, apparent transverse NMR relaxation time; TLC, thin-layer chromatography

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION REFERENCES

1. Natarajan, V., Schmid, P. C., Schmid, H. H. O. 1986. N-acylethanolamine phospholipid metabolism in normal and ischemic rat brain. Biochim. Biophys. Acta. 878:32-41[Medline].

2. Hansen, H. S., Lauritzen, L., Moesgaard, B., Strand, A. M., Hansen, H. H. 1998. Formation of N-acyl-phosphatidylethanolamines and N -acylethanolamines—Proposed role in neurotoxicity. Biochem. Pharmacol. 55:719-725[Medline].

3. Schmid, H. H. O., Schmid, P. C., Natarajan, V. 1990. N-acylated glycerophospholipids and their derivatives. Prog. Lipid Res. 29:1-43[Medline].

4. Meneses, P., Glonek, T. 1988. High resolution ³¹P NMR of extracted phospholipids. J. Lipid Res. 29:679-689[Abstract].

5. Driscoll, D. M., Romano, F. D., Smith, C. A., Meneses, P. 1996. Insulin inhibits changes in the phospholipid profiles in sciatic nerves from streptozocin-induced diabetic rats: a phosphorus-31 magnetic resonance study. Comp. Biochem. Physiol. 113C:11-16.

6. Glonek, T. 1994. ³¹P NMR in the analysis of extracted tissue phospholipids. In Phosphorus-31 NMR Spectral Properties in Compound Characterization and Structural Analysis. L. D. Quin and J. G. Verkade, editors. VCH Publishers, Inc., New York. 283–294.

7. Iwata, J. L., Bardygula-Nonn, L. G., Glonek, T., Greiner, J. V. 1995. Interspecies comparisons of lens phospholipids. Curr. Eye Res. 14:937-941[Medline].

8. Liang, M. T. C., Meneses, P., Glonek, T., Kopp, S. J., Paulson, D. J., Schwartz, F. N., Gierke, L. W. 1993. Effects of exercise training and anabolic steroids on plantaris and soleus phospholipids: a ³¹P nuclear magnetic resonance study. Int. J. Biochem. 25:337-347[Medline].

9. Meneses, P., Para, P. F., Glonek, T. 1989. ³¹P NMR of tissue phospholipids: a comparison of three tissue pre-treatment procedures. J. Lipid Res. 30:458-461[Abstract].

10. Meneses, P., Navarro, J. N., Glonek, T. 1993. Algal phospholipids by ³¹P NMR: comparing isopropanol pretreatment with simple chloroform/methanol extraction. Int. J. Biochem. 25:903-910[Medline].

11. Merchant, T. E., de Graaf, P. W., Minsky, B. D., Obertop, H., Glonek, T. 1993. Esophageal cancer phospholipid characterization by ³¹P NMR. NMR Biomed. 6:187-193[Medline].

12. Merchant, T. E., Glonek, T. 1992. ³¹P NMR of tissue phospholipids: competition for Mg²+, Ca²+, Na⁺ and K⁺ cations. Lipids. 27:551-559[Medline].

13. Metz, K. R., Dunphy, L. K. 1996. Absolute quantitation of tissue phospholipids using ³¹P NMR spectroscopy. J. Lipid Res. 37:2251-2265[Abstract].

14. Malewicz, B., Baumann, W. J. 1996. Alk-1-enylacyl, alkylacyl, and diacyl subclasses of native ethanolamine and choline glycerophospholipids can be quantified directly by phosphorus-31 NMR in solution. Lipids. 31:1189-1195[Medline].

15. Epps, D. E., Natarajan, V., Schmid, P. C., Schmid, H. H. O. 1980. Accumulation of N-acylethanolamine glycerophospholipids in infarcted myocardium. Biochim. Biophys. Acta. 618:420-430[Medline].

16. Sugiura, T., Kondo, S., Sukagawa, A., Tonegawa, T., Nakane, S., Yamashita, A., Ishima, Y., Waku, K. 1996. Transacylase-mediated and phosphodiesterase-mediated synthesis of N-arachidonoylethanolamine, an endogenous cannabinoid-receptor ligand, in rat brain microsomes—Comparison with synthesis from free arachidonic acid and ethanolamine. Eur. J. Biochem. 240:53-62[Medline].

17. Folch, J., Lees, M., Sloane Stanley, G. H. 1957. A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 226:497-509[Free Full Text].

18. Branca, M., Culeddu, N., Fruianu, M., Serra, M. V. 1995. ³¹P nuclear magnetic resonance analysis of phospholipids in a ternary homogeneous system. Anal. Biochem. 232:1-6[Medline].

19. Edzes, H. T., Teerlink, T., van der Knaap, M. S., Valk, J. 1992. Analysis of phospholipids in brain tissue by ³¹P NMR at different compositions of the solvent system chloroform–methanol–water. Magn. Reson. Med. 26:46-59[Medline].

20. Macala, L. J., Yu, R. K., Ando, S. 1983. Analysis of brain lipids by high performance thin-layer chromatography and densitometry. J. Lipid Res. 24:1243-1250[Abstract].

21. Byrdwell, W. C., Borchman, D., Porter, R. A., Taylor, K. G., Yappert, M. C. 1994. Separation and characterization of the unknown phospholipid in human lens membranes. Invest. Ophthalmol. Vis. Sci. 35:4333-4343[Abstract].

22. Chapman, K. D., Sprinkle, W. B. 1996. Developmental, tissue-specific, and environmental factors regulate the biosynthesis of N-acylphosphatidylethanolamine in cotton (Gossypium hirsutum L). J. Plant Physiol. 149:277-284.

23. Cadas, H., Di Tomaso, E., Piomelli, D. 1997. Occurrence and biosynthesis of endogenous cannabinoid precursor, N-arachidonoyl phosphatidylethanolamine, in rat brain. J. Neurosci. 17:1226-1242[Abstract/Free Full Text].

24. Natarajan, V., Reddy, P. V., Schmid, P. C., Schmid, H. H. O. 1981. On the biosynthesis and metabolism of N-acylethanolamine phospholipids in infarcted dog heart. Biochim. Biophys. Acta. 664:445-448[Medline].

25. Hansen, H. S., Lauritzen, L., Strand, A. M., Vinggaard, A. M., Frandsen, A., Schousboe, A. 1997. Characterization of glutamate-induced formation of N-acylphosphatidylethanolamine and N-acylethanolamine in cultured neocortical neurons. J. Neurochem. 69:753-761[Medline].

26. Hansen, H. S., Lauritzen, L., Strand, A. M., Moesgaard, B., Frandsen, A. 1995. Glutamate stimulates the formation of N-acylphosphatidylethanolamine and N-acylethanolamine in cortical neurons in culture. Biochim. Biophys. Acta. 1258:303-308[Medline].

27. Horowski, R., H. Wachtel, L. Turski, and P-A. Löschmann. 1994. Glutamate excitotoxicity as a possible pathogenetic mechanism in chronic neurodegeneration. In Neurodegenerative Diseases. D. B. Calne, editor. W. B. Saunders Company, London. 163–175.