The compounds used for this structure activity relationship study were synthesized from readily available starting materials as shown in the synthetic scheme (Figure 1) and in Table 1.

All chemicals and solvents were purchased from Sigma-Aldrich Canada Ltd., Oakville, ON., VWR Canada and Nu-Chek Prep., Elysian, MN. All solvents used were anhydrous. Analytical thin layer chromatography (TLC) was carried out on precoated silica gel TLC aluminum sheets (EM science, Kieselgel 60 F₂₅₄, 5 × 2 cm × 0.2 mm). Compounds were visualized under UV light (254/366 nm) or placed in iodine vapor tank and by dipping the plates in a 5% aqueous (w/v) phosphomolybdic acid solution containing 1% (w/v) ceric sulfate and 4% (v/v) H₂SO₄, followed by heating. Flash column chromatography was carried out using silica gel, Merck grade 60, mesh size 230-400, 60 Å. NMR spectra were recorded on Bruker Avance spectrometers; for ¹H (500 MHz), δ values were referenced to CDCl₃ (CHCl₃ at 7.24 ppm) and for ¹³C NMR (125.8 MHz) referenced to CDCl₃ (77.23 ppm). Coupling constants (J) are reported to the nearest 0.5 Hz. High resolution mass spectral data were obtained on Bruker Apex 7T Fourier transform ion cyclotron resonance mass spectrometer (FT-ICRMS) with atmospheric pressure chemical ionization in the positive mode (HRAPCI-MS). MS/MS data collected using QStar XL TOF mass spectrometer with atmospheric pressure chemical ionization (APCI) source in the positive mode and collision energy of 20 and 35 V. Fourier transform infra-red (FTIR) spectra were recorded on Bio-Rad FTS-40 spectrometer using the diffuse reflectance method on samples dispersed in KBr. Refer synthetic scheme (Figure 1) and Table 1 for details of compounds synthesized.

To sodium hydride (1.85 g, 60% dispersed in mineral oil) under argon was added anhydrous N, N-dimethylformamide (DMF, 10 ml) at room temperature (RT). Solketal (16.7 mmol) in 10 ml anhydrous DMF was then added dropwise with constant stirring. 1-Bromohexadecane, 1-bromooctadecane or octadec-9-enylmethane sulphonate (16.7 mmol) [40,41] dissolved in anhydrous DMF (10 ml) was then added to the reaction mixture dropwise and stirred for 48 hours. The reaction mixture was poured into cold ice water (100 ml) and extracted with hexane (100 ml, 3×). After drying over anhydrous Na₂SO₄ and removal of solvent, the crude product was treated with 10% HCl solution (40 ml) and refluxed at 120°C for 30 min. The reaction mixture was then kept at RT for 24 hours. The off white lumps and the mother liquor was extracted with diethylether (100 ml, 3×) washed successively with saturated aqueous NaHCO₃ (100 ml) and water (100 ml), dried over anhydrous Na₂SO₄ and the solvent removed under reduced pressure to obtain the products C11-13. Satisfactory spectral and analytical data were obtained [42].

Each of compounds C11-13, monopalmitin and monostearin (1.63 mmol) was dissolved separately in anhydrous DMF (2.00 ml) followed by the addition of imidazole (3.72 mmol) and tert-butyl dimethylsilyl chloride (1.88 mmol). The reaction mixture was stirred at RT for 24 hours, poured into water (100 ml) and extracted with diethyl ether (100 ml, 3×). After removal of solvent, the crude product was chromatographed on silica gel using CH₂Cl₂-MeOH to obtain the products [43].

To a mixture of each of C14-18 (0.578 mmol), anhydrous pyridine (0.35 ml), catalytic amount of dimethylaminopyridine (0.100 mmol) and toluene (5 ml) was added the appropriate acyl chloride like (4Z,7Z,10Z,13Z,16Z,19Z)-docosa-4,7,10,13,16,19-hexaenoyl chloride (0.576 mmol) dropwise under argon and stirred at RT for 48 hours. The reaction mixture was poured into water (100 ml), extracted with diethyl ether (100 ml, 3×), washed successively with 0.25 M H₂SO₄ solution (100 ml), saturated aqueous NaHCO₃ (100 ml) and water (100 ml). After drying over anhydrous Na₂SO₄ and removal of solvent, the crude product was subjected to chromatography on silica gel using hexane-CH₂Cl₂ mixtures to obtain the products.

To a mixture of each of C19-28 (0.272 mmol) and THF (2 ml) was added imidazole (0.942 mmol) and 0.80 ml of 1.0 M TBAF in THF dropwise at - 20°C and kept at this temperature for 24 hours with constant stirring. The reaction mixture was then passed through a plug of silica gel and eluted with cold diethylether (10 ml, - 20°C). After removal of solvent, the crude products were chromatographed on silica gel using hexane - ethyl acetate mixtures to obtain the products [44,45].

HEK 293 cells were purchased from ATCC, and cultured in DMEM, 10% FBS at 37°C, 5% CO₂. CHO and NRel-4 cells were a kind gift from Dr. R.A. Zoeller (Boston University) and were cultured in F-12 medium, 10% FBS at 37°C, 5% CO₂. NRel-4 cells are deficient in peroxisomal dihydroxyacetonephosphate acyltransferase (DHAPAT; EC 2.3.1.42; Figure 2).

The plasmalogen-deficient CHO cell line (NRel-4) was used to assay the efficacy of test compounds C1-10 in plasmalogen restoration. CHO (wild type) or NRel-4 (plasmalogen deficient) cells were seeded in DMEM/F-12 medium (10% FBS) on a 10 cm dish the day before the experiment. The following day, the media was replaced with fresh media containing the test compound or the solvent ethanol (0.1% V/V) as a control [46]. Cells were cultured for 72 hours at 37°C, 5% CO₂, after which they were harvested using Versene/TrypLE express (Gibco Life Technology, Rockville, MD). The cell pellet was washed with phosphate buffered saline (PBS, pH 7.4), and the phospholipids were extracted and analyzed using a linear ion-trap mass spectrometer coupled to a LC system as described [31].

Human embryonic kidney 293 (HEK293) cells and CHO/NRel-4 cells were seeded the day before the treatment. The following day, the cells were treated with the test compounds C1-10 or with ethanol as the control. Concentrations used for pravastatin, clofibrate, and troglitazone treatments were as described [47-49]. Cells were harvested after 48 hours using Versene: TryPLe express cocktail, washed with PBS. Lipids were extracted with chloroform containing 1%Triton X-100. The organic fraction was recovered and dried under a stream of nitrogen. The dried lipids were resuspended in cholesterol reaction buffer (Biovision, Mountain View, CA), and the total, free and esterified fractions of cholesterol were quantified using the cholesterol quantification kit (Biovision, Mountain View, CA) as per the manufacturer's recommendations. Cholesterol was reported as μg/million cells.

HEK293 cells were treated as described in the amyloid assay. The cell pellet was washed in PBS and lysed in RIPA buffer containing a protease inhibitor cocktail (Sigma, St. Louis, MI). Protein in the cell lysate was quantified using the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA). The following antibodies were used for western analyses: SOAT1 (Santa Cruz Biotechnology Inc., CA) and β-actin (Sigma, St. Louis, MI). Band intensities were calculated using ImageJ (National Institutes of Health).

Statistical Analysis of the data was performed using Microsoft Office Excel 2007 and JMP version 8. Multiple comparison Dunnett's tests were applied to analyze the differences between the treatments and the control.

Membrane plasmalogen levels of NRel-4 cells, lacking dihydroxyacetone phosphate acyl transferase (DHAPAT), an obligate enzyme in the plasmalogen biosynthesis pathway [50,51], were less than 10% of wild-type CHO cells ( Table 2 and Figure 2A) as assessed through the relative quantification of five common palmityl ether ethanolamine plasmalogens (PlsEtn). The cell membranes of NRel-4 cells also contained nearly 2-fold more free cholesterol (Figure 2B) and 4-fold less esterified cholesterol than CHO cells (Figure 2C).

Using wild CHO and NRel-4 cells, side chain-specific PlsEtn and phosphatidylethanolamine (PtdEtn) precursors (Table 1) were evaluated for their abilities to augment cellular plasmalogen levels in control (CHO) and PlsEtn-deficient (NRel-4) cells. These alkylacylglyceryl ethers bypass the requirement for peroxisomes in the synthesis of plasmalogens (Figure 3). The following observations were made:

1. Maintaining the free alcohol at sn-3 and DHA at sn-2, PlsEtn precursors C1-3 with differing long chain ether substitutions at sn-1 revealed that these precursor compounds either partially or fully restored all ethanolamine plasmalogens with the same sn-1 alkyl ether but had no effect on PlsEtn with different sn-1 compositions (Figure 4). For example, treatment with a palmityl PlsEtn precursor (C1) restored the downstream pool of 16:0 ethanolamine plasmalogens with no effect on the 18:0 and 18:1 PlsEtn pools. Such side chain-specific restoration indicates that no rearrangement of the sn-1 moiety (O-alkyl linkage) occurs, while the sn-2 moiety is able to undergo deacylation and subsequent reacylation with other fatty acid residues.

2. Similarly, compounds C6-C10 (16:0 at sn-1 but differing at the sn-2 substituent) significantly elevate the 16:0 pool, with no effect on the 18:0 and 18:1 pools of PlsEtn (Figure 4).

3. Distribution of PlsEtn within a pool (16:0, 18:0 or 18:1) depends on the fatty acid at sn-1 position. C1 and C3 showed maximum restoration of the PlsEtn directly downstream in the pathway (16:0/22:6 PlsEtn and 18:1/22:6 PlsEtn respectively). C2 on the other hand significantly augments all PlsEtns in the 18:0 pool (Figure 5).

4. Comparison of compounds C1, C6-10, revealed that whereas DHA containing precursors can partially or fully restore all other sn-2 PlsEtn, non-DHA containing precursors cannot completely restore DHA-PlsEtn (Figure 6A).

5. DHA-PtdEtn precursors (C4 and C5) cannot restore DHA-PlsEtn deficiencies (Figure 6B).

6. PlsEtn precursors with DHA at sn-2 concentration-dependently increase DHA-PlsEtn in both DHAPAT deficient cells (NRel-4) and wild-type cells (CHO) (Figure 7A). However, with respect to total plasmalogen content, only the deficient cell line showed an increase; no augmentation in total plasmalogen content was observed in wild-type CHO cells (Figure 7B).

As demonstrated above, plasmalogen deficient cells have higher content of free cholesterol and lower amounts of esterified cholesterol in their cell membranes. To determine whether this effect was due to a general decrease in membrane PlsEtn composition or to decreased levels of specific PlsEtn, membrane PlsEtn levels in PlsEtn depleted cells (NRel-4) were selectively restored as described above and the corresponding effect on membrane cholesterol composition ascertained. The key observations were:

1. PtdEtn precursors (C4 and C5) had no effect, while PlsEtn precursors with < 3 unsaturations (C7 and C8) had a mild effect on membrane cholesterol composition (Figure 8).

2. PlsEtn precursors with 3 or more unsaturations (C1, C9 and C10) had a more profound effect on reducing free cholesterol (Figure 8B) and increasing esterified cholesterol (Figure 8C).

The effect of plasmalogen precursors and other compounds on membrane cholesterol composition was further studied in PlsEtn normal human HEK293 cells (Figure 9). The key observations were:

1. PlsEtn precursor C1 exhibited a concentration-dependent decrease in free cholesterol (47% at 20 μM) (Figure 9B) and a reciprocal increase in the esterified fraction of cholesterol (31% at 20 μM) (Figure 9C)

2. PtdEtn precursors (C4 and C5) had no effect on free cholesterol and resulted in slight decreases in esterified cholesterol.

3. PlsEtn precursors with sn-2 substituents containing < 3 unsaturations (C6, C7 and C8) either elevated or had no effect on free cholesterol.

4. PlsEtn precursors with sn-2 substituents containing 3 or more unsaturations either reduced free cholesterol (C1 and C9) and/or increased esterified cholesterol (C1 and C10).

5. Free DHA had a slight impact on free cholesterol (14% reduction) compared to control, while it exhibited a 24% increase in the esterified cholesterol fraction at the 20 μM concentration.

6. Pravastatin treatments were most potent in reducing free cholesterol at 10 μM concentration (24% reduction compared to control, p = 0.01), while the 100 μM concentration did not result in a further reduction of free cholesterol. The changes observed in the esterified cholesterol were not significant.

7. Treatments with PPARα (clofibrate; 100 and 200 μM) and PPARγ (troglitazone; 1 and 10 μM) agonists had no impact on the cholesterol profile of HEK293 cells at the concentrations tested.

The effects of the potent cholesterol esterification enhancing/total cholesterol lowering PlsEtn precursor, C1,and of the potent HMG-CoA reductase inhibiting/total cholesterol lowering statin, pravastatin, on the basal levels of cholesterol processing enzyme SOAT1 was determined. The maximum cholesterol-lowering concentration of C1, resulted in a 50% elevation of SOAT1 levels (calculated using ImageJ software), whereas the maximum cholesterol-lowering concentration of pravastatin had no effect on SOAT1 levels (Figure 10).