Towards the Total Synthesis of Jerangolids – Synthesis of an Advanced Intermediate for the Pharmacophore Substructure

The jerangolids are a class of natural products with a skipped diene substructure isolated from Sorangium cellulosum. Here, we present a new strategy for the total synthesis of these compounds based on a skipped diyne as central building block and a suitably substituted epoxy aldehyde as building block for the dihydropyran substructure. So far, we reached an advanced intermediate which is related to the pharmacophore subunit of

the jerangolids as well as of the ambruticins. A key step is a Shi epoxidation with high e.r. to form the epoxy aldehyde. Both building blocks are coupled in a Carreira alkynylation, where additional mechanistic studies based on DFT calculation were realized. The alkynylation is followed by a nucleophilic 6-endotet epoxide opening to form the pyran structure and a Nicholas reduction to remove a propargylic OH group.
The jerangolids are structurally related to the ambruticins [2] (Figure 2) which were isolated from Polyangium cellulosum var. fulvum [2a,2b] and Sorangium cellulosum So ce 10. [2c,2d] Since both, the jerangolids and the ambruticins, show potent antifungal activity, [1,2] it was assumed [3] that the common substructure of both natural compounds (C6-C17 and C20-C22 jerangolide numbering) is the pharmacophore responsible for the fungicidal effect. Due to the interesting structure and the pharmacological activity of the jerangolids total syntheses were developed for these substances. Whereas Marko et al. [4] synthesized jerangolid D (2), Hanessian and co-workers [5] synthesized jerangolid A (1) and Hahn et al. [6] very recently developed a total synthesis for jerangolid E (4). Besides these complete syntheses, also synthetic approaches for substructures of the jerangolids appeared in the literature. [7] We became interested in the jerangolids because of their cryptical symmetric structure ( Figure 1, blue structure) and their promising pharmacological activity. Here we would like to present preliminary results of our synthetic efforts to develop a synthetic strategy for jerangolid B (3), which should be adaptable also for the synthesis of all other jerangolids. Therefore we here focus on the synthesis of an advanced intermediate related to the C6-C17 substructure (closely related to the pharmacophore of jerangolids and ambruticins). Next, we synthesized building block 10 (Scheme 3). Starting from commercially available 5-trimethylsilyl-pent-4-yn-1-ol 12, hydroalumination with DIBALH and iodination with I 2 led to vinyl iodide 18. [11] Negishi coupling [12] with ethylzinc bromide gave unsaturated alcohol 11, which was converted to epoxy alcohol 19 through Shi epoxidation [13] with er = 98.5:1.5. [14] Finally, Dess-Martin oxidation resulted in epoxy aldehyde 10 in 46 % yield over four steps. Scheme 3. Synthesis of the building block 10 for the tetrahydropyran substructure of the jerangolids. Now, the stage was set for the planned Carreira alkynylation. [15] Prior to the coupling of skipped diyne 9 with epoxy aldehyde 10, we studied the alkynylation of 9 with isobutyraldehyde (20), 3-phenylpropionaldehyde (21) and with benzaldehyde (22) to find suitable reaction conditions. Astonishingly, no reaction occurred at room temperature, in contrast to Carreira′s standard conditions. [15] As control experiments, we pursued alkynylations with 1-octyne (23) and aldehyde 20 under standard conditions in toluene and in dichloromethane as solvents. These control experiments worked quite well, so the skipped diyne 9 must have a lower intrinsic reactivity than simple alkynes. [16] After extensive experimentation with 9 and isobutyraldehyde (20) we found that a Carreira alkynylation takes place at 40°C and requires 2.5 equiv. of Zn(OTf ) 2 , 2.5 equiv. Nmethyl ephedrine and 2.0 equiv. of Et 3 N. To prevent excessive formation of 26, aldehyde 20 had to be added very slowly (best with a syringe pump over several hours) to the reaction mixture (Scheme 4).
ynes with an active C-H bond in 3-position always showed isomerization to the corresponding allenynols alongside the alkynylation.
Obviously, isomerization of 9 to 17 occurs during alkynylation of the aldehyde. To gain more insight into isomerization during the Carreira reaction with 9, DFT calculations were performed. [17] Allenyne 17 was computed to be 11.5 kcal/mol more stable than the diyne 9, indicating that the allenyne is thermodynamically preferred. We also calculated various possible Zn complexes [18] from 9 and 17 ( Figure 3). For additional information of the Zn complexes 27-30 (bond lengths, bond angles) see SI.
Until now, there is no further information in the literature about the exact alkynylation mechanism of skipped diynes. Therefore we present our suggested mechanism for the alkynylation and occurring isomerization based on the results of the DFT calculations in Scheme 5. Scheme 5. Possible mechanism to explain products 24, 25 and 26. The blue reaction path must be fast compared to the red reaction path to explain the relative amounts of the products. nificantly and after deprotonation with triethylamine, the allenylzinc A or zincacetylide B are formed. Following this, the zincacetylide B reacts with the aldehyde to the particular propargylic alcohol 24, whereas the reactivity of the allenylzinc species is too low for further reactions and is consequently protonated by the ammonium ion. The allenyne 17 undergoes the same complexation to form the most stable zinc complex 27 and further reacts to the propargylic alcohol 25. Based on the DFT calculations, the blue reaction pathway is favored leading to the formation of the energetically lower allenyne 17 and the zinc complex 27, assuming thermodynamic control of the mechanism after the formation of intermediate A.
To circumvent the allene formation, the central building block 9 had to be replaced by a less volatile and more stable diyne, which can not isomerize to an allenyne like 17 or 25. Therefore, we used the tertiary alcohol 31 as a substitute for 9. 31 is a stable, less volatile compound, the OH-group can in principle be removed by the Nicholas reaction [19] and it is easily accesssible [20] (Scheme 6). We got crystals of 31 suitable for X-ray analysis. Details thereof can be found in the SI. With 31 in hand, the Carreira alkynylation with aldehyde 10 could be performed. Applying the conditions for diyne 9, we obtained the desired secondary alcohol 33 as a mixture of diastereomers (Scheme 7). Regarding the expected desymmetrization of the enantiotopic triple bonds in 31, the diastereoselectivity is very small (56:44), whereas the selectivity of the Carreira alkynylation is acceptable (87:13) (Scheme 7). [21] Scheme 7. Carreira alkynylation of 31 and epoxy aldehyde 10, cyclization and deprotection.
The mixture of diastereomers 33a and 33b was cyclized to tetrahydropyrans 34a and 34b through BF 3 -induced epoxide opening in 81 % yield and high 6-endo-tet selectivity (the same composition as 33a and 33b, but now, the diastereomers could be separated via flash chromatography). The regioselectivity is controlled by the silyl-group, which favors a nucleophilic attack at the α-carbon center. [10i,22] The TMS group attached to C15 could be removed in 69 % yield with TBAF in THF to give the advanced intermediate 35.
Here, the diastereomers also could be separated with flash chromatography.
At this stage, we studied the racemic deoxygenation of the tertiary alcohol via Nicholas reaction. [19] A stereoselective method was published by Kann. [23] To remove the OH group it has to be converted into an acetate since acetates work better in Nicholas reactions than free OH groups. [19] Thus, treatment of 35 with acetic anhydride and DMAP gave acetate 36, which was subjected to deoxygenation with Co 2 (CO) 8 , Et 3 SiH and BF 3 ·OEt 2 to give the skipped diyne 37 after oxidative regeneration of the triple bond with ceric ammonium nitrate in 35 % (Scheme 8).

Conclusions
In summary, we could synthesize compound 37, which is an advanced intermediate in our total synthesis of the jerangolids, in 11 steps and 7 % yield, with the longest linear sequence of 7 steps. This intermediate permits the completion of the synthesis of the natural products as well as the synthesis of analogues for pharmacological testing. Key steps are a Carreira alkynylation with a skipped diyne, a TMS directed 6-endo-tet epoxide opening and a Nicholas reaction with Et 3 SiH for the deoxygenation of a tertiary alcohol. The strategy is highly flexible, which opens the possibility to synthesize many structural analogues for pharmaceutical evaluation. Currently, we work on the completion of the synthesis of jerangolid B and the optimization of the Nicholas reaction. This work will be published in due course.

Experimental Section
General: All reactions were run under an N 2 atmosphere in dried (heat gun) glassware. All solvents used in reactions were purchased in HPLC grade quality and were additionally freshly distilled under N 2 . THF and toluene were distilled from sodium/benzophenone and dichloromethane was distilled from CaH 2 . Solvents for flash chromatography and for the extraction of aqueous phases were distilled under laboratory atmosphere with a rotavap. Petrol ether refers to a mixture of hexanes with a boiling range from 45-65°C. All other chemicals were used as purchased without additional purification. Silica gel for flash chromatography was from Merck, Darmstadt, Germany (Silica 60, particle size 40-63 μm). TLC plates for reaction monitoring were from Merck, Darmstadt, Germany (Si60 254 glass 5836 plates 50 × 100 mm). NMR spectra were recorded with a BRUKER AV 400 NMR spectrometer at 400 MHz ( 1 H) and 100 MHz ( 13 C) respectively. HRMS spectra were recorded on a Bruker SolariX 7T FTICR-MS. (14): Trimethylsilylacetylene (13) (21.9 mL, 153 mmol) is dissolved in dry THF (200 mL) and cooled to -78°C. nBuLi (72 mL of a 2.5 M solution in hexane, 180 mmol) is added over 15 min. The resulting yellow solution is stirred for 1.5 h at -78°C prior to slow addition of acetaldehyde (25.3 mL, 450 mmol) over 30 min. Stirring is continued at -78°C for three hours. Then, the reaction is quenched by adding sat. NH 4 Cl solution. The phases are separated and the aqueous phase is extracted with diethyl ether. The combined organic phases are washed with brine and dried with MgSO 4 . Filtration and evaporation of the solvent yield the product as slightly yellow oil (13.3 g, 95.5 mmol, 62 %), which is used without further purification. 1

3-Methyl-1,5-bis(trimethylsilyl)penta-1,4-diyne (16):
A solution of MeMgBr (3.0 M, 19.0 mL, 57 mmol) in diethyl ether is diluted by THF (60 mL) prior to dropwise addition of Trimethylsilylacetylene (13) (9.5 mL, 67 mmol) during 15 min. so that the reaction mixture is gently boiling. After cooling to room temperature, dry CuCl (180 mg, 3.6 mmol) is added and the reaction mixture is heated to reflux for 1.5 hours. After cooling to room temperature, a solution of 15 (11.1 g, 50 mmol) in THF (8.5 mL) is added dropwise during 30 min. The black solution is heated and stirred for a further 2.5 hours at reflux. After cooling to room temperature, the reaction mixture is added carefully to a mixture of 1 M HCl (50 mL) and ice. The mixture is extracted with diethyl ether and the combined organic extracts are washed with sat. NaHCO 3 solution and brine successively. Drying with MgSO 4 , filtration and evaporation of the solvent gives the crude product which is purified by distillation (Kugelrohr) at 5 Torr at 85-90°C. Yield: colourless oil (6.83 g, 31 mmol, 62 %). 3-Methylpenta-1,4-diyne (9): 3-Methyl-1,5-bis(trimethylsilyl)penta-1,4-diyne (16) (441 mg, 2 mmol) is dissolved in toluene at room temperature. Next, acetic acid (230 μL, 4 mmol) is added with stirring and a solution of TBAF (1.24 g, 4 mmol) in toluene (4 mL) is added very slowly dropwise during 1 hour. Stirring at room temperature is continued until 16 is consumed completely (ca. 60 hours, judged by 1 H-NMR spectroscopy). Sat. NaHCO 3 solution is added and the organic phase is washed twice with water. The combined aqueous phases are re-extracted once with toluene (4 mL). The combined organic phases are dried with MgSO 4 . After filtering the MgSO 4 , the yield and concentration of the product in toluene are determined by 1 H-NMR spectroscopy and the solution is used for the Carreira reaction. Yield: 1.26 mmol of 9 (63 %) in 10 mL of toluene (c = 0.126 mmol/mL). 1  (E)-5-Iodo-5-(trimethylsilyl)pent-4-en-1-ol (18): 1-(Trimethylsilyl)pent-4-yn-1-ol (12) (5.47 g, 35 mmol) are dissolved in diethyl ether (80 mL) and cooled to 0°C. At this temperature, a solution of DIBALH in toluene (1.2 M, 71.5 mL, 85.8 mmol) is slowly added dropwise. After all, DIBALH has been added, the reaction mixture is heated to reflux for 24 hours. After cooling to room temperature the mixture is cooled to -78°C and a solution of iodine (35.5 g, 140 mmol) in diethyl ether (75 mL) is carefully added dropwise whereupon the reaction mixture turns dark brown. After two hours at -78°C, the reaction mixture is allowed to reach room temperature (remove the cooling bath; stirring and warming overnight reduces the yield dramatically!) and is quenched with 1 M HCl (ca. 150 mL). The phases are separated and the aqueous phase is extracted with diethyl ether. The combined organic phases are washed successively with sat. Na 2 S 2 O 3 solution and brine and are dried with Na 2 SO 4 . After filtration and evaporation the crude product is purified via flash chromatography (petroleum ether/acetone, 3:1 v/v). Yield: colourless oil (9.12 g, 32 mmol, 92 %). 1

(Z)-5-(Trimethylsilyl)hept-4-en-1-ol (11):
To magnesium chips (8.5 g, 350 mmol) is added a solution of bromoethane (26.1 mL, 350 mmol) in THF (120 mL) dropwise so that the Grignard reaction starts and is boiling gently throughout the addition of bromoethane. After the addition is complete, the mixture is refluxed until all Mg is consumed (Ethylmagnesiumbromide solution A). Next, freshly dried ZnCl 2 (47.7 g, 350 mmol) is dissolved in THF (430 mL) at 0°C. To this solution, 120 mL of the Grignard solution A prepared above is slowly added dropwise at 0°C and then the mixture is stirred at room temperature for one hour (Ethylzincchloride solution B). In the meantime, 18 (9.95 g, 35 mmol) is dissolved in THF (75 mL) and to this solution, (Ph 3 P) 4 Pd (1.12 g, 1.05 mmol) is added and the mixture is stirred for 15 min. Then, solution B (250 mL) is added dropwise at room temperature until all 18 is consumed (TLC control, petroleum ether/diethyl ether, 2:1 v/v). Then an excess of ethylzincchloride is destroyed by the addition of a mixture of sat. NH 4 Cl solution and ice. The mixture is extracted with diethyl ether snd the combined organic extracts are washed successively with water and brine. Drying with MgSO 4 , filtration and evaporation of the solvent gives the crude product which is purified by flash chromatography (petroleum ether/diethyl ether, 5:1 → 2.5:1 v/v), yielding pure 11 (4.6 g, 25 mmol, 71 %).

3-[(2S,3R)-3-ethyl-3-(trimethylsilyl)oxiran-2-yl]propanal (10):
Epoxyalcohol 19 (2.4 g, 11.8 mmol) is dissolved in dichloromethane (75 mL). Pyridine (1.12 g, 1.14 mL, 14.2 mmol) is added with stirring prior to addition of Dess-Martin periodinane (6.02 g, 14.2 mmol). The reaction mixture is stirred for one hour at room temperature and is quenched by addition of sat. NaHCO 3 solution and sat. Na 2 S 2 O 3 solution. The phases are separated and the aqueous phase is extracted with dichloromethane. The combined organic phases are washed with brine and dried with MgSO 4 . After filtration, the dichloromethane is distilled off at 500-600 Torr and 55°C bath temperature. The residue is purified by flash chromatography using n-pentane/diethyl ether, 4:1 (v/v) as eluent. After distillation of the solvents, aldehyde 10 is obtained as colourless liquid (2.22 g, 11.1 mmol, 94 %). 1  (3R)-2,6-Dimethylocta-4,7-diyn-3-ol (24): Zinctriflate (2.42 g, 6.65 mmol) is dried under vacuum with a heat gun until a fine white powder is obtained. After cooling to room temperature and flushing with N 2 , (+)-N-methyl-ephedrine (1.19 g, 6.65 mmol) is added and the flask is evacuated and flushed with N 2 three to four times. Then, dry toluene (5 mL) is added and the mixture is stirred vigorously at room temperature. After 5 min triethylamine (737 μL, 5.32 mmol) is added dropwise and the mixture is stirred at room temperature for an additional hour. To the resulting biphasic suspension a solution of 3-methyl-1,4-pentadiyne (9) in toluene (21.1 mL of a 0.126 molar solution; see preparation of 9 above) is added at once. Stirring is continued for 1.5 hours at room temperature prior to heating the reaction mixture to 40°C. Then, a solution of isobutyraldehyde (291 μL, 3.19 mmol) in toluene (2 mL) is added during 4.5 hours via syringe pump. After further 30 min. at 40°C, the reaction mixture is cooled to room temperature and the reaction is quenched by addition of sat. NH 4 Cl solution (ca. 10 mL). The phases are separated and the aqueous phase is extracted with diethyl ether. Use the aqueous phase later for recovering of N-methyl-ephedrine! The combined organic phases are washed with brine and dried with MgSO 4 . After filtration and evaporation of the solvent, the crude product is purified by flash chromatography (petroleum ether/acetone, 6:1 v/v). Yield of 24: 80 mg as a clear oil (0.53 mmol, 20 %). Additionally, allenyne 25 (189 mg, 1.14 mmol, 43 %) and 26 (71 mg, 0.32 mmol, 12 %) are obtained. 25: 1

3-Methyl-1,5-bis(trimethylsilyl)penta-1,4-diyn-3-ol (32):
Magnesium turnings (2.67 g, 110 mmol) are placed in a dry flask. 1-bromopropane (13.53 g, 110 mmol) is dissolved in dry THF (40 mL). 5 mL of this solution is added at once to the magnesium turnings so that the Grignard reaction starts. In case the reaction starts vigorously the reaction temperature has to be controlled by gentle cooling with cold water. Then, the rest of the bromopropane solution is carefully added dropwise at such a rate, that the reaction mixture is gently boiling. When all the bromopropane has been added, the mixture is heated to reflux until all magnesium turnings were consumed (ca. 45 min.). Then, a solution of trimethylsilylacetylene (11.79 g, 17.08 mL, 120 mmol) in THF (20 mL) is carefully added dropwise to the hot solution of nPrMgBr (vigorous gas evolution!). After the addition is complete, the mixture is further refluxed for 30 min. Next, a solution of ethyl acetate (4.41 g, 4.88 mL, 50 mmol) in THF (10 mL) is slowly added dropwise while still refluxing the reaction mixture. When the addition of ethyl acetate is complete, the reaction mixture is refluxed for another 1.5 hours. After cooling to room temperature the excess of the Grignard reagent is destroyed by carefully quenching the reaction mixture with ice-cold water (30 mL; vigorous gas evolution!) and sat. NH 4 Cl solution (50 mL) under vigorous stirring. The phases are separated and the aqueous phase is acidified with 1 M HCl to dissolve precipitated salts and then extracted several times with diethyl ether. The combined organic phases were washed with brine and dried with MgSO 4 . After filtration, the solvent is evaporated and the product starts to crystallize. Yield of 32: 11.3 g (47.5 mmol, 95 %). 1 H-NMR (400 MHz, CDCl 3 , 20°C): δ = 1.73 (s,3 H,H6),0.17 (s,18 H,TMS) ppm. 13 C-NMR (100 MHz, CDCl 3 , 20°C) δ = 106.0 (C2 + C4), 86.9 (C1 + C5), 67.9 (C3), 31.9 (C6), -0.3 (TMS) ppm.

(6R)-8-[(2R,3R)-3-Ethyl-3-(trimethylsilyl)oxiran-2-yl]-3-methyl-
octa-1,4-diyne -3,6-diol (33a + 33b): Zinctriflate (7.85 g, 21.6 mmol) is dried under vacuum with a heat gun until a fine white powder is obtained. After cooling to room temperature and flushing with N 2 , (+)-N-methyl-ephedrine (3.94 g, 21.6 mmol) is added and the flask is evacuated and flushed with N 2 three to four times. Then, dry toluene (21 mL) is added and the mixture is stirred vigorously at room temperature. After 5 min triethylamine (1.62 g, 2.23 mL, 16.0 mmol) is added dropwise and the mixture is stirred at room temperature for an additional hour. To that biphasic suspension, 31 (824 mg, 8.8 mmol) is added at once. Stirring is continued for 1.5 hours prior to heating the reaction mixture to 40°C. Then, a solution of epoxy aldehyde 10 (1.68 g, 8.0 mmol) in toluene (8 mL) is added during 4.5 hours via syringe pump. After a further 30 min. at 40°C, the reaction mixture is cooled to room temperature and the reaction is quenched by addition of sat. NH 4 Cl solution (15 mL). The phases are separated and the aqueous phase is extracted with diethyl ether (ca. 20 mL) three times. Do not discard the aqueous phase! The combined organic phases are washed with brine and dried with MgSO 4 . After filtration of the MgSO 4 and evaporation of the solvent, the crude product is purified by flash chromatography (petroleum ether/acetone, 6:1 v/v). 33 is obtained as a mixture of diastereomers (1.92 g, 6.5 mmol, 81 %). 1  The aqueous phase from above is made basic by addition of NaOH solution. Extraction with diethyl ether, drying of the extract with Na 2 SO 4 is a means to recover the (+)-N-methyl ephedrine for further use.