The Synthesis of 3-(R)- and 3-(S)-Hydroxyeicosapentaenoic Acid

Monohydroxylated polyunsaturated fatty acids belonging to the oxylipin class of natural products are present in marine and terrestrial sources as well as in the human body. Due to their biological activities and role in diverse biosynthetic pathways, oxylipins biosynthesized from eicosapentaenoic acid and arachidonic acid have attracted great interest from the scientific community. One example is 3-hydroxyeicosapentaenoic acid where the absolute configuration at C-3 has only been tentatively assigned. In this paper, studies on acetate type aldol reactions that enabled the preparation of 3-(R)-hydroxyeicosapentaenoic acid (3R-HETE, 2) and its enantiomer are presented.


Introduction
Biologically active oxygenated natural products derived from the two ω-3 polyunsaturated fatty acids (PUFAs), eicosapentaenoic acid and arachidonic acid, have for a long time been the topic of biological and biosynthetic investigations [1,2], but also targets for stereoselective total synthesis [3]. Examples of classes of such natural products are prostaglandins [4], leukotrienes [5], lipoxins [6], resolvins, maresins, protectins [7], epoxy fatty acids [8] and oxylipins [9]. These natural products are biosynthesized by three distinct main pathways where cytochrome P450 oxidases or various enzymes of the cyclooxygenase, lipoxygenase and epoxygenase classes are involved ( Figure 1). Very often these biosynthetic pathways involve the formation of a mono-hydroxy substituted intermediate that is further converted to the aforementioned oxylipin classes [5,7].

Introduction
Biologically active oxygenated natural products derived from the two ω-3 polyunsaturated fatty acids (PUFAs), eicosapentaenoic acid and arachidonic acid, have for a long time been the topic of biological and biosynthetic investigations [1,2], but also targets for stereoselective total synthesis [3]. Examples of classes of such natural products are prostaglandins, [4] leukotrienes, [5], lipoxins, [6] resolvins, maresins, protectins [7], epoxy fatty acids [8] and oxylipins. [9] These natural products are biosynthesized by three distinct main pathways where cytochrome P450 oxidases or various enzymes of the cyclooxygenase, lipoxygenase and epoxygenase classes are involved ( Figure 1). Very often these biosynthetic pathways involve the formation of a mono-hydroxy substituted intermediate that is further converted to the aforementioned oxylipin classes [5,7]. To date, numerous C-6 to C-20 fatty acids with a hydroxy substituent in the C-3 position have been reported isolated from several fungi [10]. However, only two mono- To date, numerous C-6 to C-20 fatty acids with a hydroxy substituent in the C-3 position have been reported isolated from several fungi [10]. However, only two mono-hydroxy oxylipins substituted in the 3-position derived from the PUFA eicosapentaenoic acid and arachidonic acid have been reported (see Figure 2) [10,11]. The first hydroxy oxylipins substituted in the 3-position derived from the PUFA eicosapentaenoic acid and arachidonic acid have been reported (see Figure 2) [10,11]. The first one to be reported was 3-(R)-hydroxy arachidonic acid (3R-HETE, 1) [11,12], while 3-(R)hydroxyeicosapentaenoic acid (3R-HEPE, 2) was reported to have been isolated from the yeast Dipodascopsis uninucleata [13,14]. To date, only the oxylipin 1 has been synthesized [15,16] and subjected to biological investigations [11,14]. To date, limited knowledge with respect to the role of the oxylipin 2 constitutes in microbial pathogenesis is available. In order to gain new knowledge of the biological processes mediated by this oxylipin, biosynthetic and biological studies are needed. However, such endeavors require access to sufficient amounts of chemically pure EPAderived oxylipin 2, not easily available by isolation from biological sources. Moreover, the structural elucidations of these lipids have been based on GC/MS-and HPLCanalyses [13,17], due to the minute amount isolated from fungal sources. Hence, stereoselective synthesis is in demand in order to establish the tentatively assigned Rconfiguration of the PUFA derived 2. Furthermore, the need for new antifungal drugs is imminent [18].
Based on the above arguments, we aim for a convenient and stereoselective synthesis of (R)-3-hydroxyeicosapentaenoic acid (2). In order to meet this aim, the four all-Z double bonds present in 2 called for docosahexaenoic acid (3) as a convenient starting material, in particular based on our prior successful experiences [19,20]. This approach keeps the all-Z methylene interrupted double bonds present in 3, biosynthesized by highly specific desaturases, intact throughout the synthesis of 2.
The C18 aldehyde all-Z- (3,6,9,12,15)-octadeca-(3Z,6Z,9Z,12Z,15Z)-pentaenal (4) constitutes a useful starting material in order to meet these needs. Regarding the chiral center of the homoallylic alcohol present in 2, we noted that the absence of any large stereo-directing groups and the vast number of possible conformers of 2 limited as well as directed our retrosynthetic analyses towards diastereoselective acetate type aldol reactions. The Nagao-Fujita reaction [21] allows the use of different thiazolidinedione auxiliaries to participate in aldol reactions with a range of different aldehydes, including aliphatic ones. Early reports were on the tin-mediated aldol addition of various substituted N-acetyl-1,3-thiazolidine-2-thione (5a-5e) to a wide range of aldehydes [21,22].
Later, Urpí and Vilarrasa expanded this methodology to titanium enolates using different bases, where diisopropylamine (DIPEA) gave high diastereoselection [23], also studied by Crimmins and co-workers with aliphatic aldehydes [24,25]. Moreover, Hodge and Olivo investigated the combined use of (−)-sparteine and titanium enolates of the iso-propyl substituted auxiliary 5b, which afforded high diastereoinduction of aliphatic aldehydes [26]. In this paper, we report our results using the Nagao-Fujita reaction under various conditions with aldehyde 4 that enabled the first total synthesis of 3-(R)hydroxyeicosapentaenoic acid (3R-HEPE, 2). In addition, performing the Braun aldol reaction [27] on the same aldehyde, an efficient preparation of the enantiomer of 2 became available. To date, limited knowledge with respect to the role of the oxylipin 2 constitutes in microbial pathogenesis is available. In order to gain new knowledge of the biological processes mediated by this oxylipin, biosynthetic and biological studies are needed. However, such endeavors require access to sufficient amounts of chemically pure EPA-derived oxylipin 2, not easily available by isolation from biological sources. Moreover, the structural elucidations of these lipids have been based on GC/MS-and HPLC-analyses [13,17], due to the minute amount isolated from fungal sources. Hence, stereoselective synthesis is in demand in order to establish the tentatively assigned R-configuration of the PUFA derived 2. Furthermore, the need for new antifungal drugs is imminent [18].
Based on the above arguments, we aim for a convenient and stereoselective synthesis of (R)-3-hydroxyeicosapentaenoic acid (2). In order to meet this aim, the four all-Z double bonds present in 2 called for docosahexaenoic acid (3) as a convenient starting material, in particular based on our prior successful experiences [19,20]. This approach keeps the all-Z methylene interrupted double bonds present in 3, biosynthesized by highly specific desaturases, intact throughout the synthesis of 2.
The C18 aldehyde all-Z- (3,6,9,12,15)-octadeca-(3Z,6Z,9Z,12Z,15Z)-pentaenal (4) constitutes a useful starting material in order to meet these needs. Regarding the chiral center of the homoallylic alcohol present in 2, we noted that the absence of any large stereo-directing groups and the vast number of possible conformers of 2 limited as well as directed our retrosynthetic analyses towards diastereoselective acetate type aldol reactions. The Nagao-Fujita reaction [21] allows the use of different thiazolidinedione auxiliaries to participate in aldol reactions with a range of different aldehydes, including aliphatic ones. Early reports were on the tin-mediated aldol addition of various substituted N-acetyl-1,3-thiazolidine-2thione (5a-5e) to a wide range of aldehydes [21,22].
Later, Urpí and Vilarrasa expanded this methodology to titanium enolates using different bases, where diisopropylamine (DIPEA) gave high diastereoselection [23], also studied by Crimmins and co-workers with aliphatic aldehydes [24,25]. Moreover, Hodge and Olivo investigated the combined use of (−)-sparteine and titanium enolates of the iso-propyl substituted auxiliary 5b, which afforded high diastereoinduction of aliphatic aldehydes [26]. In this paper, we report our results using the Nagao-Fujita reaction under various conditions with aldehyde 4 that enabled the first total synthesis of 3-(R)-hydroxyeicosapentaenoic acid (3R-HEPE, 2). In addition, performing the Braun aldol reaction [27] on the same aldehyde, an efficient preparation of the enantiomer of 2 became available.

Results and Discussion
The preparation of aldehyde 4 was achieved as presented in Scheme 1 using docosahexaenoic acid (3) as the starting material. First, a modification of the Corey-lactonization protocol [28,29] was used for the preparation of δ-lactone 6 in 96% yield that was conveniently converted into epoxy methyl ester 7 in 64% yield. Reacting 7 with H 5 IO 6 and MeOH afforded the acetal 8 in a 46% yield that was used for long-term storage, while the more sensitive aldehyde 4 was formed by treatment with formic acid in dioxane in 89% overall yield from 8. The conservation of the five all-Z methylene interrupted double bonds in 4 was confirmed by 13 C NMR spectroscopy, since no characteristic signals for allylic carbons from E,Z-conjugated isomers were present within the limits of detection [30].

Results and Discussion
The preparation of aldehyde 4 was achieved as presented in Scheme 1 using docosahexaenoic acid (3) as the starting material. First, a modification of the Coreylactonization protocol [28,29] was used for the preparation of δ-lactone 6 in 96% yield that was conveniently converted into epoxy methyl ester 7 in 64% yield. Reacting 7 with H5IO6 and MeOH afforded the acetal 8 in a 46% yield that was used for long-term storage, while the more sensitive aldehyde 4 was formed by treatment with formic acid in dioxane in 89% overall yield from 8. The conservation of the five all-Z methylene interrupted double bonds in 4 was confirmed by 13 C NMR spectroscopy, since no characteristic signals for allylic carbons from E,Z-conjugated isomers were present within the limits of detection [30]. Next, the Nagao-Fujita reaction was investigated (see Scheme 2) with the aldehyde 4. The results are compiled in Table 1.

Scheme 2.
Investigations on the Nagao-Fujita acetate aldol reaction. Next, the Nagao-Fujita reaction was investigated (see Scheme 2) with the aldehyde 4. The results are compiled in Table 1.

Results and Discussion
The preparation of aldehyde 4 was achieved as presented in Scheme 1 using docosahexaenoic acid (3) as the starting material. First, a modification of the Coreylactonization protocol [28,29] was used for the preparation of δ-lactone 6 in 96% yield that was conveniently converted into epoxy methyl ester 7 in 64% yield. Reacting 7 with H5IO6 and MeOH afforded the acetal 8 in a 46% yield that was used for long-term storage, while the more sensitive aldehyde 4 was formed by treatment with formic acid in dioxane in 89% overall yield from 8. The conservation of the five all-Z methylene interrupted double bonds in 4 was confirmed by 13 C NMR spectroscopy, since no characteristic signals for allylic carbons from E,Z-conjugated isomers were present within the limits of detection [30]. Next, the Nagao-Fujita reaction was investigated (see Scheme 2) with the aldehyde 4. The results are compiled in Table 1.
Chromatographic purification by silica gel chromatography enabled the isolation of the major acetate syn aldol product of 9b. The assignment of the chiral center of the syn R-configured aldol product 9b was based on the arguments by Hodge and Olivo [26], by the typical chemical shift values for the alpha protons at 3.64 ppm (dd, J = 17.7, 2.6 Hz, 1 H) and at 3.51 ppm (dd, J = 11.5, 7.9 Hz, 1 H), for the less and the more shielded protons, respectively.
Next, reaction conditions to obtain the desired product 2 were attempted. However, the long-time storage of the aldol product 9b proved problematic, as it decomposed and formed its α, β-unsaturated isomer over time. Hence, TBS-protection was performed, which yielded 10 in 72% yield that was cleanly converted to the ethyl ester 11 in 78% yield. Then, 13 was converted using a two-step procedure (TBS-deprotection, hydrolysis, 53% yield) to afford 2 (see Scheme 3). The spectroscopic data (UV, IR, and NMR) were all in accord with the assigned structure and the MS-data were comparable to the literature [13]. Later, the treatment of 9b with diluted aqueous LiOH in THF and MeOH cleanly afforded (R)-3-hydroxyeicosapentaenoic acid (2) in 87% isolated yield (Scheme 3) [31]. The shortest sequence presented in Schemes 1-3 afforded 2 in 10% overall yield over the six steps.
Chromatographic purification by silica gel chromatography enabled the isolation of the major acetate syn aldol product of 9b. The assignment of the chiral center of the syn R-configured aldol product 9b was based on the arguments by Hodge and Olivo [26], by the typical chemical shift values for the alpha protons at 3.64 ppm (dd, J = 17.7, 2.6 Hz, 1 H) and at 3.51 ppm (dd, J = 11.5, 7.9 Hz, 1 H), for the less and the more shielded protons, respectively.
Next, reaction conditions to obtain the desired product 2 were attempted. However, the long-time storage of the aldol product 9b proved problematic, as it decomposed and formed its α, β-unsaturated isomer over time. Hence, TBS-protection was performed, which yielded 10 in 72% yield that was cleanly converted to the ethyl ester 11 in 78% yield. Then, 13 was converted using a two-step procedure (TBS-deprotection, hydrolysis, 53% yield) to afford 2 (see Scheme 3). The spectroscopic data (UV, IR, and NMR) were all in accord with the assigned structure and the MS-data were comparable to the literature [13]. Later, the treatment of 9b with diluted aqueous LiOH in THF and MeOH cleanly afforded (R)-3-hydroxyeicosapentaenoic acid (2) in 87% isolated yield (Scheme 3) [31]. The shortest sequence presented in Schemes 1-3 afforded 2 in 10% overall yield over the six steps. Then, we attempted the method reported by Braun and co-workers [27] using (R)-2hydroxy-1,2,2-triphenylethyl acetate (12) (see Scheme 4). In this case, the diastereomeric ratio was very high (95:5), and 13 was isolated in 70% yield after chromatographic purification. Hydrolysis with K2HPO4 in refluxing MeOH cleanly removed the ester auxiliary in 13 to give the methyl ester 14. Hydrolysis afforded the S enantiomer of 3R-HEPE, namely (S)-3-hydroxyeicosapentaenoic acid (2). The spectral data matched those of synthetic R-2. The optical rotation was observed to be dextrorotary ( = + 11.1, CHCl3, c Then, we attempted the method reported by Braun and co-workers [27] using (R)-2hydroxy-1,2,2-triphenylethyl acetate (12) (see Scheme 4). In this case, the diastereomeric ratio was very high (95:5), and 13 was isolated in 70% yield after chromatographic purification. Hydrolysis with K 2 HPO 4 in refluxing MeOH cleanly removed the ester auxiliary in 13 to give the methyl ester 14. Hydrolysis afforded the S enantiomer of 3R-HEPE, namely (S)-3-hydroxyeicosapentaenoic acid (2). The spectral data matched those of synthetic R-2. The optical rotation was observed to be dextrorotary ([α] 20 D = + 11.1, CHCl 3 , c 0.90), while (R)-3-hydroxyeicosapentaenoic acid gave a levorotary rotation ([α] 20 D = −10.5 • , CHCl 3 , c 1.08). Starting from DHA, the overall yield of S-2 was 12% over the eight steps. Hence, both enantiomers were made available for biological and biosynthetic investigations. 0.90), while (R)-3-hydroxyeicosapentaenoic acid gave a levorotary rotation ( = −10.5°, CHCl3, c 1.08). Starting from DHA, the overall yield of S-2 was 12% over the eight steps. Hence, both enantiomers were made available for biological and biosynthetic investigations. To prove the absolute configuration of each of the enantiomers of 3hydrocyeicosapentaenoic acid, we produced the Mosher ester derivatives [32]. However, looking at the 1 H NMR spectra of the two compounds, the diagnostic chemical shifts overlapped to such an extent that accurate interpretation was impossible. Fluorine NMR analysis has frequently been used to make assignments of the configuration of MTPA esters. Using Hoye and co-workers set-up [32,33], we reacted both enantiomers of 2 with both S-and R-MTBA corresponding esters (See Table 2). Looking at the presumed R-3hydrocyeicosapentaenoic acid (2) gave a chemical shift of −72.42 ppm for the S-MTBA derivative and a −72.38 ppm chemical shift for the corresponding R-MTBA derivative. The difference, thus, is negative, implying an R-configuration, according to Hoye's work and in accord with our own assumption.
A similar set-up for the presumed S-enantiomer of 2 gave a positive difference in chemical shift. This implies an S-configuration, again in accord with our assumption. The results are summarized in Table 2. The trifluoromethyl group prefers the illustrated rotamers (see Figure 3), in which it is eclipsed with the carbonyl group and is, therefore, within its deshielding cone. The fundamental arguments are again based on the assumption of the preferred conformations depicted in Figure 3. The relative extent of this preference for conformers is assumed to be largely dependent upon the magnitude of the steric interaction between the large phenyl group and the carbinol substituents L2 and L3. For example, if L3 is larger To prove the absolute configuration of each of the enantiomers of 3-hydrocyeicosapentaenoic acid, we produced the Mosher ester derivatives [32]. However, looking at the 1 H NMR spectra of the two compounds, the diagnostic chemical shifts overlapped to such an extent that accurate interpretation was impossible. Fluorine NMR analysis has frequently been used to make assignments of the configuration of MTPA esters. Using Hoye and co-workers set-up [32,33], we reacted both enantiomers of 2 with both Sand R-MTBA corresponding esters (See Table 2). Looking at the presumed R-3-hydrocyeicosapentaenoic acid (2) gave a chemical shift of −72.42 ppm for the S-MTBA derivative and a −72.38 ppm chemical shift for the corresponding R-MTBA derivative. The difference, thus, is negative, implying an R-configuration, according to Hoye's work and in accord with our own assumption. A similar set-up for the presumed S-enantiomer of 2 gave a positive difference in chemical shift. This implies an S-configuration, again in accord with our assumption. The results are summarized in Table 2.
The trifluoromethyl group prefers the illustrated rotamers (see Figure 3), in which it is eclipsed with the carbonyl group and is, therefore, within its deshielding cone. The fundamental arguments are again based on the assumption of the preferred conformations depicted in Figure 3. The relative extent of this preference for conformers is assumed to be largely dependent upon the magnitude of the steric interaction between the large phenyl group and the carbinol substituents L2 and L3. For example, if L3 is larger than L2, the destabilizing L3, or phenyl, the interaction in the conformer of the 8-MTPA ester 15 will result in the trifluoromethyl group spending less time in the carbonyl deshielding plane (relative to other conformers for the (R)-MTPA ester 15).
Kakisawa and Kashman [34] have cautioned that conclusions based on 19 F NMR analysis are often in error and must be scrutinized with care. However, the results are consistent with both our assumptions using well-established stereoselective reactions [22][23][24][25][26] and the conclusion by Hoye and co-workers, strongly indicating that the configuration was assigned correctly.  Kakisawa and Kashman [34] have cautioned that conclusions based on 19 F NMR analysis are often in error and must be scrutinized with care. However, the results are consistent with both our assumptions using well-established stereoselective reactions [22][23][24][25][26] and the conclusion by Hoye and co-workers, strongly indicating that the configuration was assigned correctly.

General Information:
Unless stated otherwise, all commercially available reagents and solvents were used in the form in which they were supplied without any further purification. The stated yields are based on isolated material. All reactions were performed under an argon atmosphere using Schlenk techniques. Thin layer chromatography was performed on silica gel 60 F254 aluminum-backed plates fabricated by Merck. Flash column chromatography was performed on silica gel 60 (40-63 μm) produced by Merck. NMR spectra were recorded on a Bruker Ascend TM 400 (Bruker, Billerica, MA,USA) or Bruker AVI600 spectrometer (Bruker, Billerica, MA,USA) at 600 MHz or 400 MHz, respectively, for 1 H NMR and at 150 MHz or 100 MHz, respectively, for 13 C NMR. Coupling constants (J) are reported in hertz and chemical shifts are reported in parts per million (δ) relative to the central residual protium solvent resonance in 1 H NMR (CDCl3 = δ 7.26, DMSO-d6 = δ 2.50 and MeOD-d4 = δ 3.31) and the central carbon solvent resonance in 13 C NMR (CDCl3 = δ 77.00 ppm, DMSO-d6 = δ 39.43 and MeOD-d4 = δ 49.00). 19 F NMR experiments were run in CDCl3 with signals calibrated against CF3CH2OH. Mass spectra were recorded at 70 eV on Waters Prospec Q spectrometer (Waters Corporation, Milford, MA, USA) using EI, ES or CI as the methods of ionization. High resolution mass spectra were recorded on Waters Prospec Q spectrometer using EI or ES as the methods of ionization. Optical rotations were measured using a 1 mL cell with a 1.0 dm path length on a Perkin Elmer 341 polarimeter (PerkinElmer Inc., Waltham, MA, USA). IR, MS, 1 H and 13 C NMR spectra as well as UV and HPLC chromatograms are found in the Supplementary Material.

General Information
Unless stated otherwise, all commercially available reagents and solvents were used in the form in which they were supplied without any further purification. The stated yields are based on isolated material. All reactions were performed under an argon atmosphere using Schlenk techniques. Thin layer chromatography was performed on silica gel 60 F254 aluminum-backed plates fabricated by Merck. Flash column chromatography was performed on silica gel 60 (40- 19 F NMR experiments were run in CDCl 3 with signals calibrated against CF 3 CH 2 OH. Mass spectra were recorded at 70 eV on Waters Prospec Q spectrometer (Waters Corporation, Milford, MA, USA) using EI, ES or CI as the methods of ionization. High resolution mass spectra were recorded on Waters Prospec Q spectrometer using EI or ES as the methods of ionization. Optical rotations were measured using a 1 mL cell with a 1.0 dm path length on a Perkin Elmer 341 polarimeter (PerkinElmer Inc., Waltham, MA, USA). IR, MS, 1 H and 13 C NMR spectra as well as UV and HPLC chromatograms are found in the Supplementary Material.
3.1.1. 5-((3Z,6Z,9Z,12Z,15Z)-1-Iodooctadeca-3,6,9,12,15-pentaenyl)dihydro-2(3H)-furanone (6) Iodolactone 6 was prepared according to a literature procedure by Ulven and coworkers [35]. 2,6-Lutidine (7.1 mL, 60.9 mmol, 2 eq.) was added dropwise to a solution of 3 (10.0 g, 30.4 mmol, 1 eq.) dissolved in CH 2 Cl 2 (100 mL). The mixture was cooled to 0 • C and I 2 (15.453 g, 60.9 mmol, 2 eq.) was added. The reaction continued overnight (15 h) at 0 • C. The reaction was quenched by adding saturated aqueous Na 2 S 2 O 3 (60 mL). Ethyl acetate was used for extraction (2 × 60 mL) and the combined organic layers were washed with saturated aqueous NaH 2 PO 4 (50 mL) and brine, and dried using Na 2 SO 4 . The solvent was removed in vacuo and the product was passed through a short silica plug (hexane/EtOAc 1:1) to afford iodolactone 6 (13.279 g) in a 96% yield. All spectroscopic and physical data were in agreement with those reported in the literature [35].  (7) Iodolactone 6 (8.482 g, 18.66 mmol, 1 eq.) was dissolved in MeOH (60 mL) and K 2 CO 3 (3.096 g, 22.40 mmol, 1.2 eq.) was added. The reaction mixture was stirred at ambient temperature for 3 h. Water (50 mL) was added and the aqueous phase was extracted using hexane (3 × 50 mL). The combined organic layers were dried (MgSO 4 ) and the solvent removed in vacuo. Epoxide 7 was obtained as a light brown oil (6.023 g) in a 90% yield. All spectroscopic and physical data were in agreement with those reported in the literature [36].  (8) Epoxide 7 (6.023 g, 16.80 mmol, 1 eq.) was dissolved in MeOH (120 mL) and periodic acid (4.595 g, 20.16 mmol, 1.2 eq.) was added. The reaction mixture was stirred for 6 h. Water (100 mL) was added and the aqueous phase was extracted (3 × 100 mL) using hexane. The combined organic layers were washed with brine and dried (MgSO 4 ). The solvent was removed in vacuo. The crude product was purified by column chromatography on silica (hexane/EtOAc 95:5) to obtain dimetylacetal 8 as a colorless oil (2.086 g) in a 41% yield. All spectroscopic and physical data were in agreement with those reported in the literature [35]. Acetal 8 (1.010 g, 3.32 mmol, 1 eq.) was dissolved in 1,4-dioxane (14 mL) and an 80% (w/w) aqueous solution of formic acid (16 mL) was added. The reaction mixture was stirred for 1.5 h at ambient temperature. The reaction was quenched by the addition of water (50 mL). The aqueous layer was extracted (3 × 30 mL) with hexane. The combined organic layers were washed with an aqueous NaHCO 3 solution (50 mL), then with brine (50 mL) and dried (MgSO 4 ). The solvent was removed in vacuo and aldehyde 4 (0.771 g) was obtained as a colorless oil in a 90% yield. All spectroscopic and physical data were in agreement with those reported in the literature [35]. (R)-valinol (1.010 g, 9.79 mmol, 1 eq.) was dissolved in EtOH (5 mL) and CS 2 (2.35 mL, 39.1 mmol, 4 eq.) was added. A 2.25 M solution of KOH (17.40 mL, 39.1 mmol, 4 eq.) in 1:1 EtOH/H 2 O was added dropwise at room temperature over 20 min. The reaction mixture was stirred and heated at reflux for 72 h. After cooling, the mixture was acidified by slowly adding 0.5 M HCl (20 mL). The slightly acidic mixture was extracted with CH 2 Cl 2 (3 × 20 mL) and the combined organic layers were concentrated in vacuo. The crude product was purified by column chromatography on silica (hexane/EtOAc 4:1) to afford 5b as a yellow oil (1.40 g) in a 88% yield. All spectroscopic and physical data were in agreement with those reported in the literature [37].  (1.40 g, 9.3 mmol, 1. eq.) was dissolved in dry THF (18 mL) and cooled to 0 • C. A 60% dispersion of sodium hydride in mineral oil (0.409 g, 10.2 mmol, 1.1 eq.) was dissolved in dry THF (18 mL) and the thioxothiazolidin solution was added dropwise. The mixture was stirred for 10 min at 0 • C, followed by the addition of AcCl (0.73 mL, 10.2 mmol, 1.1 eq.) and the mixture was stirred for another 10 min at the same temperature. The heterogenous solution was allowed to reach ambient temperature and stirred for 1 h. The reaction was quenched by the addition of an aqueous 5% HCl solution (15 mL) and the aqueous layer was extracted with EtOAc (3 × 20 mL). The combined organic layers were dried (Na 2 SO 4 ), filtrated and the solvent was removed in vacuo. The crude product was purified by column chromatography (hexane/EtOAc 9:1) to afford the chiral auxiliary 5b (1.70 g) as a bright yellow oil in a 90% yield. All spectroscopic and physical data were in agreement with those reported in the literature [37]. R f = 0.23 (hexanes/EtOAc 9:1 KMnO 4 stain); 1 H NMR (400 MHz, CDCl 3 ) δ 5.14 (ddd, J = 7.7, 6.1, 1. The aldol product 9b was prepared according to a literature procedure by Tungen and co-workers [31]. (R)-4-isopropyl-1,3-thiazolidine-2-thione (5b) (1.213 g, 5.97 mmol mmol, 2 eq.) was dissolved in CH 2 Cl 2 (60 mL) and 1 M TiCl 4 in CH 2 Cl 2 (6.56 mL, 6.56 mmol mmol, 2.2 eq.) was added at −78 • C. After 5 min, DIPEA (1.25 mL, 7.16 mmol, 2.4 eq.) was added. The reaction mixture was stirred at −78 • C for 1 h. Aldehyde 4 (0.771 g, 2.98 mmol, 1 eq.) in CH 2 Cl 2 (16 mL) was added dropwise over 20 min, and the reaction mixture was stirred for 4 h at −78 • C. The reaction was quenched by adding aqueous saturated NH 4 Cl (60 mL) and the mixture was allowed to reach ambient temperature. The aqueous phase was extracted using CH 2 Cl 2 (3 × 60 mL). The combined organic layers were dried (Na 2 SO 4 ) and the solvent removed in vacuo. The crude product was purified by column chromatography on silica (hexane/EtOAc 9:1) yielding aldol product 9b (630 mg, 46%) as a yellow oil. In addition, its diastereomer (85 mg, 6%) was also obtained. The aldol product 9b (0.085 g, 0.18 mmol) was dissolved in CH 2 Cl 2 (9.5 mL). The solution was cooled on a dry ice/ethanol bath. After 10 min, 2,6-lutidine (0.064 mL, 0.059 g, 0.55 mmol) was added. The reaction mixture was stirred for an additional 10 min, and TBSOTf (0.063 mL, 0.073 g, 0.28 mmol) was added dropwise. The reaction was stirred at this temperature for two hours before it was quenched with a saturated aqueous solution of NH 4 Cl (5 mL). The phases were separated. The water phase was extracted with CH 2 Cl 2 (3 × 10 mL). The combined organic phases were dried (Na 2 SO 4 ), filtered and concentrated in vacuo. The crude oil was purified by column chromatography on silica (EtOAc/hexane 5:95) to give compound 10 (0.077 g, 0.13 mmol, 72%) as a yellow oil. Rf Silyl ether 10 (0.041 g, 0.072 mmol) was dissolved in absolute EtOH (1.45 mL) and cooled to 0 • C. K 2 CO 3 (0.015 g, 0.11 mmol) was added to the solution, and the reaction mixture was left stirring for 3 h, then slowly allowed to reach room temperature, and stirred for an additional 23 h. The reaction was stopped by adding saturated aqueous NH 4 Cl (10 mL). The product was extracted with CH 2 Cl 2 (3 × 10 mL). The combined organic phase was washed with 1 M KOH (5 mL), saturated aqueous NaCl (10 mL) and dried over Na 2 SO 4 . Filtration and concentration in vacuo gave a crude oil, which was purified by column chromatography on silica (EtOAc/hexane 1:39) to give ethyl ester 11 (0.026 g, 0.056 mmol, 78%) as a colorless oil. Rf: 0.53 (EtOAc:hexane 1:9  (2) At 0 • C, TBAF in THF (1 M, 0.26 mL, 0.26 mmol) was added to a stirred solution of ethyl ester 11 (0.024 g, 0.052 mmol) in THF (1 mL). The reaction was stirred at this temperature for seven hours before it was quenched with phosphate buffer (pH = 7.29, 1 mL). Saturated aqueous NaCl (5 mL) and EtOAc (5 mL) was added, followed by the separation of the phases. The aqueous phase was extracted with EtOAc (2 × 5 mL). The combined organic phases were dried (Na 2 SO 4 ), filtered and concentrated in vacuo. The crude oil was purified by column chromatography on silica (EtOAc/hexane 3:7) to give the ethyl ester of 3R-HEPE (2)  was dissolved in THF/EtOH/H2O 2:2:1 (11 mL), followed by the addition of LiOH·H 2 O (0.153 g, 3.64 mmol) 0 • C. The reaction mixture was allowed to reach room temperature and stirred for 1.5 h. The solvent was evaporated in vacuo. EtOAc (10 mL) and NaH 2 PO 4 (5 mL) were added. The phases were separated. The water phase was extracted with EtOAc (2 × 10 mL). The combined organic phases were dried over Na 2 SO 4 , filtered and concentrated in vacuo. This gave acid 3R-HEPE (2) (0.030 g, 0.095 mmol, 95%) as a colorless oil.  13  The aldol product 9b (100 mg, 0.216 mmol, 1 eq.) was dissolved in a solution of THF/MeOH/H 2 O (2:2:1) (30 mL) and the solution was cooled to 0 • C. LiOH·H 2 O (0.318 g, 7.56 mmol, 35 eq.) was added and the reaction mixture was stirred for 2 h, allowed to reach ambient temperature during this time. The solvent was removed in vacuo and EtOAc (30 mL) was added. The solution was acidified with aqueous saturated NaH 2 PO 4 (20 mL) and the aqueous layer was extracted with EtOAc (2 × 30 mL). The combined organic layers were dried (Na 2 SO 4 ) and the solvent was removed in vacuo. The crude product was purified by column chromatography on silica (CH 2 Cl 2 /MeOH/HCO 2 H 95:5:0.0125) and 2 (60.0 mg) was obtained as a light-yellow oil in 87% yield. The aldol product 13 was prepared according to a literature procedure by Braun and co-workers [27]. Auxiliary 12 (81.5 mg, 0.245 mmol, 1 eq.) was added to dry THF (1.28 mL) under argon atmosphere. The slurry was cooled to −78 • C and the 2.5 M LDA solution (0.27 mL, 0.27 mmol, 1.1 eq.) was added dropwise over 10 min. The mixture was

Mosher Analysis
Mosher esters were prepared according to the procedure by Hoye et al. [32]. To a 4 mL glass vial with a Teflon-coated magnetic stir bar, methyl ester R-14 and S-14 (20 mg, 60 mmol) and dry pyridine (16 µL, 0.20 mmol) were added. CHCl 3 (1 mL) was passed through a pipette with fresh silica gel to make sure no alcohol or water were present, then added to the vial. MTPA-Cl (23 mL, 0.12 mmol) was added to the mixture. The vial was capped and left stirring in the dark at room temperature for two hours. Ether (3 mL) and water (1 mL) were added, and the layers were mixed thoroughly. The layers were separated, and the extraction was repeated two times. The combined organic layers were dried over anhydrous solid Na 2 SO 4 and the solvents were evaporated on rotary evaporator. 1 H and 19 F NMR were recorded. The 1 H NMR results were inclusive, while changes were seen in 19 F NMR. Data were analyzed as described by Hoye et al. [32,33]. Results are reported in Table 2.

Conclusions
3-Hydroxy polyunsaturated oxylipins, in particular 3-hydroxyeicosapentaenoic acid (2), show complex roles in fungal pathogenesis [39]. The oxylipin 2 may serve as a substrate for the biosynthesis of pro-inflammatory hydroxylated PUFAs [5]. Hence, biological investigations are in demand that require access to synthetic material. In this paper, the total synthesis of both enantiomers of 2 is presented, using acetate aldol methodologies. This methodology is of current interest in the total synthesis of natural products [40]. Sufficient amounts of materials for conducting biological and biosynthetic studies are now available. Results from such efforts will be reported in due time.