NaTure CHemIsTry
Articles
12 and 20 could be obtained from 16 in one pot by a sequential Fortunately, the aldehyde could be converted to terminal alkene
reduction in the presence of Ph2SiH2 and LiBEt3H, followed by full (45) with moderate yield. PGE2 (2) could be obtained after one-step
or partial deprotection. This one-pot reaction could also proceed in cis-cross-metathesis of 45 according to the reported procedure27.
a stepwise manner (see Supplementary Fig. 2 for details). In the con-
jugate 1,4-reduction, Ph2SiH2 and Sn(nBu)3H both have similar per- Conclusion
formance in gaining excellent diastereoselectivity (d.r. >20:1). In In summary, we have successfully achieved the short, highly enan-
the later stereocontrolled 1,2-reduction, inexpensive super hydride tioselective and scalable syntheses of PGs with our enyne cycloi-
was found to be the best reductant upon screening. Finally, the TBS somerization as the key step from readily available starting materials.
and the acetal groups could be removed simultaneously with aque- In this synthesis, the asymmetric hydrogenation protocol developed
ous HCl solution to afford 12, or TBS was selectively detached in the by our group played a critical role in introducing key stereogenic
presence of tetrabutylammonium fluoride (TBAF) to give 20. The centres. All reactions could be carried out on a multi-gram scale and
relative configuration of these two key intermediates was further most on a decagram scale. Additionally, our common intermedi-
determined by X-ray crystallographic analysis. The single crystal ates in this work, alongside various α and ω side chains, facilitated
analysis of compound ( )-21 derived from racemic 20 showed that the divergent synthesis of PGs. These versatile common precursors
all stereocentres exactly matched those of PGF2α (Fig. 3; for details will help to expand the existing chemical space of PGs and pro-
see Supplementary Fig. 4).
vide access to more promising therapeutic analogues. We have also
We customized different synthetic methods for various ω side shown that the key enyne cycloisomerization could offer a strategic
chains (Fig. 4). Compounds 22 and 23 could be hydrogenated enan- insight into designing synthetic routes towards multi-functionalized
tioselectively on a gram scale with the protocol developed by our five-membered rings. In particular, this work has a high possibility
group9, with excellent yields and enantioselectivity. The resulting to be developed into industrial production.
diols 24 and 25 were then transformed into corresponding epoxides
26 and 27 through mono-tosylation and intramolecular nucleo- Online content
philic substitution. Treatment of the epoxides with deprotonated Any methods, additional references, Nature Research report-
trimethyl sulfonium iodide led to allylic alcohols 28 and 29, respec- ing summaries, source data, extended data, supplementary infor-
tively (Fig. 4a). The chiral tertiary allylic alcohol 32 was conveniently mation, acknowledgements, peer review information; details of
Substituted phenols 33 and 34 were subjected to epichlorohydrin in
Received: 26 October 2020; Accepted: 14 April 2021;
Published online: 27 May 2021
the presence of K2CO3, affording epoxides 35 and 36 in high yields.
Afterwards, following the same operations as employed for 26 and
27, epoxides 35 and 36 were converted to the relevant allylic alcohols
37 and 38 in 90% and 91% yields, respectively (Fig. 4c).
references
1. Gibson, K. H. Prostaglandins, thromboxanes, PGX: biosynthetic products
from arachidonic acid. Chem. Soc. Rev. 6, 489–510 (1977).
2. Curtis-Prior, P. B. Prostaglandins: Biology and Chemistry of Prostaglandins and
Related Eicosanoids (Churchill Livingstone, 1988).
With the enantioenriched key intermediates 12 and 28, the
cross-metathesis reaction was tested with the assistance of the
Hoveyda–Grubbs second-generation catalyst (Fig. 5a)24. The desired
product 15 was furnished in 66% yield. Finally, hemiacetal 15 under-
went a Wittig reaction with phosphonium salt 39 to afford PGF2α
in 55% yield. Starting from readily available material 11, the total
synthesis of PGF2α was thus accomplished in six steps from 11 in 15%
overall yield. From versatile building block 12, the synthesis of latano-
prost (3), carboprost (5) and cloprostenol (40) were also achieved
(Fig. 5b). Latanoprost (3) was synthesized in 5.7% overall yield after
eight steps from 11 (additional hydrogenation and esterification
steps were needed for latanoprost; Supplementary Fig. 6). Carboprost
(5) and cloprostenol (40) were synthesized in 23% and 19% overall
yields, respectively, in six steps from 11. According to our investiga-
tion, intermediate 20 was more stable under cross-metathesis condi-
tions and usually resulted in higher yields than 12. A one-more-step
longer yet more scalable route was thus invented based on interme-
diate 20. Taking the cross-metatheses of 20 and 38 as representa-
tive, 26g of acetal 41 could be obtained in 81% yield (93% based on
recovered starting material) from 23g of intermediate 20. Hydrolysis
of the acetal 41 in aqueous HCl followed by Wittig olefination
gave 23.1g of fluprostenol (42) in 81% yield. Travoprost (6) was then
gathered in 74% yield after a simple esterification.
In addition, the formal synthesis of PGE2 (2) from 16 was also
established (Fig. 5d). Another useful intermediate (43) possessing
a carbonyl group was obtained by conjugated 1,4-reduction and
simultaneous deprotection in one pot. Following cross-metathesis
of 43 and allylic alcohol 28, compound 44 was produced in 67%
yield. This precursor renders PGs containing carbonyl groups, such
as PGA, PGB and PGE6, relatively easier to access. In an effort to
obtain PGE2 directly, 44 was subjected to phosphonium salt 39,
adhering to many classic Wittig olefination protocols. However,
all attempts failed and only resulted in the decomposition of 44.
3. Marks, F. & Fürstenberger, G. Prostaglandins, Laukotrienes and Other
Eicosanoids (Wiley, 1999).
4. Dams, I., Wasyluk, J., Prost, M. & Kutner, A. Terapeutic uses of
prostaglandin F2α analogues in ocular disease and novel synthetic strategies.
Prostaglandins Other Lipid Mediat. 104–105, 109–121 (2013).
5. Peng, H. & Chen, F. E. Recent advances in asymmetric total synthesis of
prostaglandins. Org. Biomol. Chem. 15, 6281–6301 (2017).
6. Das, S., Chandrasekhar, S., Yadav, J. S. & Gree, R. Recent developments in the
synthesis of prostaglandins and analogues. Chem. Rev. 107, 3286–3337 (2007).
7. Li, J. J. in Name Reactions: A Collection of Detailed Mechanisms and Synthetic
Applications (ed. Li, J. J.) 652–653 (Springer, 2014).
8. Cao, P., Wang, B. & Zhang, X. Rh-catalyzed enyne cycloisomerization. J. Am.
Chem. Soc. 122, 6490–6491 (2000).
9. Wu, W. et al. Asymmetric hydrogenation of α-hydroxy ketones with an
iridium/f-amphox catalyst: efcient access to chiral 1,2-diols. Org. Chem.
Front. 4, 555–559 (2017).
10. Von Euler, U. S. Information on the pharmacological efect of natural
secretions and extracts from male accessory sexual glands. Arch. Exp. Pathol.
Pharm. 175, 78–84 (1934).
11. Bergström, S. et al. Te isolation of prostaglandin. Acta Chem. Scand. 11,
1086–1086 (1957).
12. Flower, R. J. Prostaglandins, bioassay and infammation. Br. J. Pharmacol.
147, 182–192 (2006).
13. Bergström, S. Prostaglandins: members of a new hormonal system. Tese
physiologically very potent compounds of ubiquitous occurrence are formed
from essential fatty acids. Science 157, 382–391 (1967).
14. Funk, C. D. Prostaglandins and leukotrienes: advances in eicosanoid biology.
Science 294, 1871–1875 (2001).
15. Corey, E. J., Weinshenker, N. M., Schaaf, T. K. & Huber, W. Stereo-
controlled synthesis of dl-prostaglandins F2α and E2. J. Am. Chem. Soc. 91,
5675–5677 (1969).
16. Woodward, R. B. et al. Novel synthesis of prostaglandin F2α. J. Am. Chem. Soc.
95, 6853–6855 (1973).
17. Stork, G. & Isobe, M. General approach to prostaglandins via
methylenecyclopentanones. Total synthesis of ( )-prostaglandin F2α.
J. Am. Chem. Soc. 97, 4745–4746 (1975).
696