28 April 2025 Advanced Synthetic Strategies toward Ginkgolide Diterpenoids: Total Synthesis of (±)-Ginkgolide C and Formal Syntheses of (±)-Ginkgolides A and B Louis Barriault1, Martin Hébert2, Gabriel Bellavance1 1. University of Ottawa 2. university of ottawa Abstract Ginkgolides are highly oxygenated diterpenes isolated from Ginkgo biloba that exhibit potent anti- inflammatory and neuroprotective properties. Their compact hexacyclic architecture—featuring multiple contiguous stereocenters, a spirocyclic core, and a rare tert-butyl group—presents a formidable challenge in synthetic organic chemistry. Herein, we report the first total synthesis of ginkgolide C (3), the most structurally complex member of this family, completed in 26 steps from commercially available materials. The synthesis is guided by a functional group–driven strategy that enables the convergent construction of the polycyclic core through key diastereoselective carbon–carbon bond formations, selective oxidations, and late-stage epoxide-opening lactonizations. In parallel, the formal syntheses of ginkgolides A (1) and B (2) were accomplished via interception of a late-stage intermediate in 17 steps, the shortest route to these targets reported to date. This work provides a unified synthetic platform for accessing the ginkgolide family and offers new opportunities for the synthesis and biological evaluation of related analogues. Keywords total synthesis, natural products, stereoselective Posted on 28 April 2025 — CC-BY-NC 4.0 — This is a preprint and has not been peer reviewed. Data may be preliminary. — https:// doi.org/10.26434/chemrxiv-2025-ppm63 1 Advanced Synthetic Strategies toward Ginkgolide Diterpenoids: Total Synthesis of (±)-Ginkgolide C and Formal Syntheses of (±)-Ginkgolides A and B Martin Héberta, Gabriel Bellavanceb and Louis Barriault* Centre for Catalysis, Research and Innovation, Department of Chemistry and Biomolecular Sciences, University of Ottawa, Ottawa, Canada K1N 6N5. Louis.Barriault@uottawa.ca RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required according to the journal that you are submitting your paper to) ABSTRACT: Ginkgolides are highly oxygenated diterpenes isolated from Ginkgo biloba that exhibit potent anti-inflammatory and neuroprotective properties. Their compact hexacyclic architecture— featuring multiple contiguous stereocenters, a spirocyclic core, and a rare tert-butyl group—presents a formidable challenge in synthetic organic chemistry. Herein, we report the first total synthesis of ginkgolide C (3), the most structurally complex member of this family, completed in 26 steps from commercially available materials. The synthesis is guided by a functional group–driven strategy that O O O OOHHO OO O HO OH O O AcO O Ph OMe Ginkgolide C MeO2C OH O O O O OOHR OO O HO Ginkgolide A (R = H) Ginkgolide B (R = OH) 12 steps https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 2 enables the convergent construction of the polycyclic core through key diastereoselective carbon–carbon bond formations, selective oxidations, and late-stage epoxide-opening lactonizations. In parallel, the formal syntheses of ginkgolides A (1) and B (2) were accomplished via interception of a late-stage intermediate in 17 steps, the shortest route to these targets reported to date. This work provides a unified synthetic platform for accessing the ginkgolide family and offers new opportunities for the synthesis and biological evaluation of related analogues. INTRODUCTION Ginkgolides are diterpenoids isolated from the ancient ginkgo tree (Ginkgo biloba), a living relic with a lineage dating back 280 million years. For centuries, the tree and its extracts have been used extensively for their beneficial properties, especially in China, Japan and India. In 1932, intrigued by the therapeutic potential of ginkgo extracts, Furukuwa isolated four ginkgolides from the bark of Ginkgo biloba, believing them to be responsible for its activity.1 In 1967, Nakanishi elucidated the structures of these ginkgolides, identifying them as ginkgolides A (1), B (2), C (3), and M (4).2 Ginkgolide J (5) was discovered and characterized by Weinges and co-workers in 1987 (Figure 1).3 Since then, additional ginkgolides, including K, L, P, and Q, have been isolated.4,5 Ginkgolides have been shown potential for cognitive enhancement and may help mitigate conditions such as dementia and Alzheimer’s disease as well as other CNS ailment such as Parkinson and multiple sclerosis.6-8 In addition, these compounds, notably ginkgolide B (2), exhibit potent platelet-activating factor (PAF) antagonist properties, holding promise for diverse therapeutic applications.9 https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 3 Figure 1. Structure of ginkgolides From a structural perspective, ginkgolides exhibit a unique compact highly polyoxygenated cage-like structure characterized by six five-membered rings, two adjacent quaternary carbon centers, and tert-butyl group that is rarely found in natural products. The most complex member of this family, ginkgolide C (3), contains a total of 12 contiguous stereocenters. The unique and highly intricate structure of ginkgolides, particularly their polycyclic, oxygenated framework and spirocyclic core, presents a formidable yet intellectually stimulating platform for the design of synthetic strategies. These unique features, combined with their impressive therapeutic profile and high structural complexity, have made ginkgolides a fascinating synthetic target. To date, only a handful of synthetic studies on these molecules have been reported,10,11 including the total syntheses of ginkgolides A (1) and B (2) by the research groups of Corey12 and Crimmins.13 Containing 12 contiguous stereocenters and 11 oxygenated carbons, ginkgolide C (3) is the most complex diterpene of its family. Before this work, it had not been successfully synthesized in its entirety. In this Article, we disclose a detailed synthetic strategy that has evolved over time, leading to the first total synthesis of ginkgolide C (3) and the formal syntheses of ginkgolides A (1) and B (2). RESULTS AND DISCUSSION Retrosynthetic analysis (first approach): From the outset, the densely packed polycyclic structure, characterized by its intricate network of interconnected rings and stereochemical complexity, served as the guiding blueprint in shaping our initial synthetic strategy. The structural intricacy, including the precise arrangement of multiple quaternary O O O OOHR2 OO O R1 R3 O O R3 O O O OR1 R2 O HO Ginkgolide A (1) R1= OH, R2 = R3 = H Ginkgolide B (2) R1= OH, R2 = OH, R3 = H Ginkgolide C (3) R1= OH, R2 = OH, R3 = OH Ginkgolide M (4) R1= OH, R2 = H, R3 = OH Ginkgolide J (5) R1= H, R2 = OH, R3 = OH C A D E F 1 3 4 5 7 8 9 10 11 13 12 14 15 B 10 8 14 4 6 7 12 2 https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 4 centers and spirocyclic frameworks, underscored the challenges inherent in the synthesis of this target molecule. As depicted in Scheme 1, ginkgolide C (3) can be derived from ketone 6 through a diastereoselective formation of the C–C and C–O bonds at positions C3, C13, and C12. The construction of rings D and E in intermediate 6 is envisioned to proceed via a strategic addition of a cyano group or its equivalent at C4, facilitating the subsequent formation of the lactone ring D. This step would be followed by the development of ring E through acetal formation in the bis-spirocyclic compound 7, ensuring stereochemical control. At this juncture, the introduction of the tert-butyl group at C8 and the C–O bond at C6 would be achieved via an allylic oxidation and a diastereoselective conjugate addition, respectively, favoring the sterically less hindered face of compound 8. Scheme 1. First approach and key disconnections To arrive at compound 8, we propose employing a Claisen rearrangement using intermediates 9 and 10, a critical transformation that effectively establishes the essential quaternary centers at C5 and C9. This rearrangement not only defines the stereochemical integrity of the molecule but also serves as a cornerstone for subsequent synthetic manipulations. The synthesis involves a ring-closing metathesis to O O O OOHHO OO O HO O O O OO O O O OH OR O H OMe O RO MeO OMe Ginkgolide C (3) 6 8 9 aldol [O] alkylation [O] Cuprate addition [3,3] RCM 10 5 7 8 9 13 3 2 8 14 12 11 [O] O ORO HO OMe 7 6 OR OH TBSO OTBS 9 OMe CN 4 10 5 https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 5 construct rings A and B, bringing the core polycyclic framework to completion. This retrosynthetic analysis presents a strategy centered on the compact and intricate ginkgolide framework, aiming to achieve the diastereoselective formation of key C–C and C–O bonds. The synthesis began with the transformation of ketone 914 into the corresponding ketal, which in the presence of allylic alcohol 1015 and propionic acid, generated the enol ether 11 in situ via a transketalization. This intermediate subsequently underwent a Claisen rearrangement, effectively generating the adjacent quaternary carbon centers C5 and C9, resulting in the formation of ketone 12 (Scheme 2A).16 The latter was next subjected to a ring-closing metathesis reaction catalyzed by Grubbs II catalyst, leading to the formation of spiroacetal 13. Subsequent treatment of 13 with CSA (1 mol%) yielded hemiketal 14, with an overall yield of 19% over four steps. This sequence enabled the construction of three rings and established two adjacent quaternary centers. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 6 Scheme 2. The primary alcohol was silylated under standard conditions, yielding 62%. This was followed by allylic oxidation using excess CrO3 and 3,5-dimethylpyrazole, resulting in a mixture of regioisomeric enones 15 and 16 with yields of 24% and 30%, respectively. Next, the tert-butyl group was introduced via conjugate addition to enone 16 using tert-BuLi and CuCN in the presence of TMSCl, giving enol ether O OH O MeO 9 10 OH TBSO OTBS 1. Amberlyst 15 (5% w/w) CH(OMe)3, MeOH, reflux 30 min 2. 10, EtCO2H (10 mol%), 85° to 150°C, 3 h, neat O OROR O OTBSTBSO 95 CSA (1 mol%), MeOH, 40°C (19%, over 3 steps) 11 R = TBS OTBS OMeOO OTBS OMeO TMSO O OMeOO OMeOO OMe 1. TBSCl, DMAP imidazole, DCM, rt (62%) N H N 2. CrO3, DCM, rt OTBS O MeO O + + CuCN, t-BuLi TMSCl, THF -78°C (63%) A 15 (24%) 16 (30%) 1. TBAF, THF, rt (71%) 2. DMP, DCM, rt (92%) 12 14 171819 1. DMP, DCM, rt (82%) 2. [Ph3PCH2OMe][Cl], t-BuOK, THF, rt (95%, E/Z = 1:1.9) 14 OMeO 20 OMe OMeO 21 (not observed) OMe OMe 22 (88%) MeO MeO O OMe CSA (5 mol%) MeOH, rt + CSA (5 mol%) dioxane/MeOH/H2O 80°C (90%, 23a/23b = 1.5:1) O X O Y O X O Y O O X O Y O 23a X = OMe, Y = H 23b X = H, Y = OMe 24a X = OMe, Y = H (47%) 24b X = H, Y = OMe (45%) 25a X = OMe, Y = H (48%) 25b X = H, Y = OMe (0%) CuCN, t-BuLi TMSCl, THF -78°C O X O Y TMSO 26a X = OMe, Y = H (19%) 26b X = H, Y = OMe (30%) TBAF, THF, rt (76%) + N H N CrO3, DCM, rt B Grubbs II (2 mol%) DCM, 40°C 13 O OTBS OTBS https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 7 16 in a 63% yield. Notably, this addition proceeded with high diastereoselectivity (dr > 20:1) adjacent to a quaternary center. Subsequent treatment with TBAF (71%) led to the corresponding primary alcohol which was subsequently oxidized using Dess-Martin periodinane to give aldehyde 18 in a 92% yield. Unfortunately, all efforts to synthesize enol ether 19, or its equivalent, via olefination were unsuccessful, with the starting material being predominantly recovered in most attempts. This outcome suggests that steric hindrance from the tert-butyl group may be blocking access to the aldehyde, thereby inhibiting the reaction. Taking a step back, we opted to generate the F ring first before the installation of the tert-butyl group. Returning to alcohol 14, oxidation with Dess-Martin periodinane led to the corresponding aldehyde in 82% yield which was then subjected to Wittig olefination to give the desired enol ether 20 in 95% yield (Scheme 2B). Attempts to convert the enol ether 19 into the corresponding dimethyl acetal 21 were unsuccessful, instead leading to the formation of the ketal 21. After screening various solvent mixtures, it was found that performing the reaction in dioxane/methanol/water (7:2.75:0.25) system produced acetals 23a and 23b in 90% yield with a ratio of 1.5 to 1. Following the previously established allylic oxidation protocol, enones 24a and 24b were obtained in 47% and 45% yields respectively. Interestingly, the addition of tert-butyl cuprate to enone 24a resulted in a mixture ketone 25a (48%) and TMS-enol ether 26a (19%), whereas enone 24b led exclusively to the formation 26b in a modest 30% yield. Both TMS-enol ethers 26a and 26b were subsequently converted back to their corresponding ketones 25a and 25b using TBAF. With ketones 25a and 25b in hand, our next objective was to investigate C–C bond formation at C4 and C–H oxidation at C10 (Scheme 3). We hypothesized that the latter transformation could be achieved through a Norrish Type II photochemical functionalization. Although both ketone moieties at C4 and C6 could be excited and participate in the reaction, we speculated that the spatial proximity of C4 and C10 would favor a Norrish–Yang reaction, leading to the formation of cyclobutanol intermediate 27, which could then be converted to 28 through appropriate functional group manipulations. Unfortunately, despite multiple attempts, no detectable traces of the desired product 27 were observed in the crude https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 8 reaction mixtures. Faced with this challenge, we then shifted our focus to a more direct approach– oxidation of the C10 C–H bond using strong oxidants such as RuCl₃ and RuO₄. However, this strategy also proved difficult, yielding only trace amounts of oxidized products 29 and 30. In parallel, we explored the formation of a C–C bond at C4 using ketone 13 through nucleophilic addition or enol formation strategies. Unfortunately, these attempts did not yield the desired products 31 or 32, highlighting the inherent challenges associated with selective functionalization at this position. Scheme 3. We suspected that the steric hindrance from the TBS ether moieties at C9 created significant steric hindrance, preventing nucleophilic addition and effectively blocking access to the reactive site. To address this, we investigated replacing the TBS protecting group with an isopropylidene moiety (Scheme 4). Beginning with allylic alcohol 33, we successfully synthesized spiroketal 34 with an overall yield of 33%. Cyanosilylation of ketone 34 using PEG (10 mol%), nBuLi (10 mol%), and TMSCN (1.5 equiv.) yielded the TMS-cyanohydrin 35 in 89% as a single diastereomer. The stereochemical outcome was confirmed via acidic deprotection of the isopropylidene group, producing lactone 36. Unfortunately, the O O O OMe 10 [O] C-C formation 4 25 O O O OMe 25 hν O HOO OMe 27 O O O OMe 28 E E+ O O O OMe 25 O O O OMe 29 OH Oxidation at C10 O O O OMe 30 O RuCl3 xH2O NaIO4 solvent (<10%) + C-C Bond Formation at C4 13 O OTBS OTBS 32 RO OTBS OTBS 31 OH OTBS OTBS R 4 104 6 https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 9 cyanosilylation yielded the undesired diastereomer. As an alternative strategy, vinyl triflate 37 was subjected to palladium-catalyzed cross-coupling with sodium cyanide, forming the corresponding α,β- unsaturated nitrile. Subsequent oxidation with tert-butyl hydroperoxide and TBAF produced epoxides 38 and 39 in 59% and 33% yields, respectively. However, NMR analysis revealed that the desired diastereomer was the minor product. As a result, we abandoned this approach. Scheme 4. Although these approaches did not lead to desirable outcome, they nonetheless provided key insights into the synthetic strategy, highlighting critical factors that influence reaction efficiency, stereoselectivity, and scalability. We confirmed that the construction of the spirocycle (rings A and B) via a Claisen rearrangement and ring-closing metathesis reaction is highly robust and scalable to the decagram level. The incorporation of the tert-butyl group via cuprate addition proceeds with high diastereoselectivity, ensuring efficient stereochemical control. Additionally, we established that the oxidation state at C10 must be properly set early in the synthesis to facilitate downstream transformations. Finally, the steric environment around C4 plays a critical role in facilitating both C–C and C–O bond formation, influencing the overall success of key synthetic steps. O 9 33 OH O O 1. Amberlyst 15 (5% w/w) CH(OMe)3, MeOH, reflux 30 min 2. 33, EtCO2H (10 mol%), 85° to 150°C, 3 h, neat 3. Grubbs II (2 mol%) DCM, 40°C (33% over 3 steps) + 34 O O O 34 O O O 35 TMSO O O NC 37 TfO O O 38 (59%) CN O O O 39 (33%) CN O O O 36 HO OH O O PEG (10 mol%) n-BuLi (10 mol%) TMSCN, THF 0°C to rt (89%) PTSA (5 mol%) MeOH, reflux (78%) KHMDS, PhNTf2 THF, -78 °C to -45°C (94%) 1. NaCN, Pd(PPh3)4 CuI, MeCN, reflux (99%) 2. t-BuO2H, TBAF DMSO, rt + https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 10 Retrosynthetic analysis (second approach): Building on these findings, we explored an alternative synthetic route to the lactone moiety (ring D) in ketone 6, envisioning that it could arise from a 5-exo-dig cyclization at C15 on the alkyne group (Scheme 5). This transformation would be preceded by a stereoselective cuprate addition of a tert-butyl group at C8 on enone 40, a key intermediate for the synthesis of ginkgolides A (1), B (2), and C (3). To ensure precise control over the adjacent quaternary carbon centers (C5 and C9), we adopted a Sonogashira coupling strategy with phenylacetylene 43 (C14– C15) followed by alkylation with iodoalkane 42 on spiroalkane 41. This Pd-catalyzed reaction was designed not only to enable the formation of the D-ring lactone but also to create a steric environment favoring diastereoselective nucleophilic addition at C9, facilitating the formation of the bond C9–C12 on spiroalkane 41. The spiroalkane framework could be efficiently accessed from commercially available ketone 9 and allyl alcohol 44, following the well-established Claisen rearrangement securing the C5 quaternary center, followed by RCM between C7 and C8 to generate the B-ring. Together, these refinements would strengthen our synthetic strategy, enhancing both efficiency and selectivity while providing a scalable and diastero-controlled route to the ginkgolide framework. O O O OOHHO OO O HO O O O OO O O O OH OAc H O O RO O Ph OMe Ph Ginkgolide C (3) 6 40 aldol [O] 5-exo dig cuprate add. SN2 [O] Pd couling [3,3] RCM 43 I OTBS 42 13 3 2 86 11 12 15 15 14 12 11 [O] O 9 44 5 7 8 9 OH OMe O 41 TfO O OMe https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 11 Scheme 5. Using previous established conditions, ketone 9 was directly converted to spiroketone 45 via a Claisen rearrangement with allylic alcohol 4417 followed by a Grubbs II-catalyzed RCM, yielding the desired ring-closed adduct in 78% yield over three steps (Scheme 6). To facilitate further functionalization of the system, the resulting alkene was subjected to DBU-mediated conjugation, efficiently generating α,β- unsaturated ester 46 in 99% yield, a crucial intermediate for subsequent transformations. Next, the ketone moiety in 46 was converted into vinyl triflate 41 in 96% yield followed by a Sonogashira coupling with phenylacetylene 43 to furnish enyne 47 in 99% yield.18 With the enyne group in place, the second quaternary carbon center was introduced with high stereocontrol. A vinylogous deprotonation of α,β- unsaturated ester 47 at C7 using KHMDS/18-crown-6 generated the enolate intermediate, which was subsequently alkylated with iodoalkane 4219 to afford α-alkylated ester 48 in 93% yield with excellent diastereoselectivity (dr >20:1). Notably, conjugation of the olefin C8-C9 with the ester group in 47 prior to alkylation was crucial. Indeed, the enolate intermediate cannot be generated when the olefin in 47 was positioned at C7-C8 with KHMDS/18-crown-6. This can be attributed to the steric hindrance imposed by the close proximity of alkyne chain to C9, which obstructed access to KHMDS, preventing deprotonation. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 12 Scheme 6. The TBS and ester protecting groups were sequentially removed using an aqueous solution of NaOH, followed by acidic treatment at 75 °C, to afford the corresponding hydroxy carboxylic acid in 90% yield. Formation of the E ring was accomplished via an iodolactonization, delivering lactone 49 in 85% yield. Oxidation of the primary alcohol to the corresponding aldehyde, followed by conversion to the dimethyl acetal 50, proceeded in high yield. Reduction of the lactone and subsequent treatment with dry HCl generated an anomeric mixture of tetracycles 51a and 51b in 51% and 41% yields, respectively. Unfortunately, the previously effective Cr(VI)/3,5-dimethylpyrazole protocol failed to promote the allylic oxidation of 51a and 51b, instead leading to complete degradation. To overcome this issue, a one-pot procedure was developed: initial SeO2-mediated allylic oxidation in 1,4-dioxane at 110 °C, followed by oxidation of the intermediate using Dess–Martin periodinane in a 1:1 mixture of 1,4-dioxane and DCM. This sequence successfully afforded enones 52a and 52b in 89% and 90% yields, respectively. At this 9 1. Amberlyst 15 (5% w/w) CH(OMe)3, MeOH, reflux 30 min 2. 10, EtCO2H (10 mol%), 85° to 150°C, 3 h, neat 3. Grubbs II (0.25 mol%) DCM, 40°C (78% over 3 steps) 45 O CO2Me 4. DBU, PhMe 80°C (99%) 46 O CO2Me 5. PhNTf2 LiHMDS THF, -45°C (96%) 41 TfO CO2Me 44 OH OMe O+ CO2Me Ph 6. 43, Pd(Pph3)2Cl2 (1 mol%) CuI (1 mol%) Et3N, 60°C (99%) 7. KHMDS, 18-Crown-6, THF then 42, -78°C (93%, dr > 20:1) CO2Me 48 OTBS Ph 8. NaOH (2M) THF/MeOH 75°C then HCl, rt (90%) 9. I2, KI NaHCO3 THF/H2O, rt (85%) 49 O Ph 50 OMe O I O O I Ph 10. DMP, DCM, rt then Amberlyst 15 (5% w/w), CH(OMe)3, MeOH, rt (92%) 11. Dibal-H, THF -78°C to 0°C 12. HCl, THF, rt O O O I Ph OMe 53 O O O I Ph OMe 54 OH O O OO I OMe O O I Ph 51a X = OMe, Y = H (51%) 51b X = H, Y = OMe (41%) X Y Y X 52a X = OMe, Y = H (89%) 52b X = H, Y = OMe (80%) 13. SeO2, dioxane 110°C, 3h then 14. DMP, dioxane/DCM rt to 60°C, 2h 15. CuCN, t-BuLi TMSI, THF -78°C then TBAF (46%) 16. LiBH4, THF 0°C to rt (75%) 17. [IPrAu(MeCN)]SbF6 (10 mol%) DCM, -78°C to rt (93%) 18. O3, DCM, -78°C then DMS (28%) 47 55 7 8 9 O Ph OH O I OMe https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 13 stage, the synthesis was continued using the b-anomer 52a. The desired conjugate addition adduct 53 was obtained in 46% yield as a single diastereomer via a one-pot sequence involving tert-butyl cuprate addition, activation/trapping with TMSI, and subsequent deprotection of the TMS enol ether using TBAF. The resulting ketone was then reduced diastereoselectively with LiBH₄ in THF to afford alcohol 54 in 75% yield. Formation of the D-ring was accomplished through a gold(I)-catalyzed 5-exo-dig cyclization of the alcohol onto the pendant alkyne, employing [IPrAu(MeCN)]SbF₆ as the catalyst.20 This transformation furnished the corresponding enol ether in 93% yield, which, upon ozonolysis, provided the pentacyclic lactone 55 in 28% yield. The initial rationale for installing the alkyl iodide on the A-ring was to enable downstream functionalization via SN2 substitution, E2 elimination, or Kornblum oxidation21 (Scheme 7). However, due to the limited availability of lactone 55, we redirected our efforts toward A-ring modification using intermediate 54. Attempts to perform Ag(I)-catalyzed substitution of the iodide with an acetate to access compound 56 were unsuccessful under a variety of conditions. Similarly, efforts to induce elimination of the iodide using either DBU or t-BuOK to obtain compound 57 resulted in no observable conversion. We next evaluated Kornblum oxidation conditions on 54, which did not furnish the desired product. Given the minimal steric hindrance present in the molecule, we hypothesized that compound 50 might undergo Kornblum oxidation. However, treatment of 50 with AgBF4 in DMSO did not yield hemiketal 58 but instead led to ether 59 in 68% yield. An alternative approach involved subjecting enol ether 60 to Kornblum oxidation. Despite screening various conditions, no conversion to 61 was observed. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 14 Scheme 7. Another strategy we explored involved replacing the alkyne with a vinyl group, based on the hypothesis that the alkene would undergo oxidative cleavage more readily than the alkyne. This transformation was also expected to reduce steric congestion around the B-ring ketone, thereby facilitating its subsequent functionalization. Starting from triflate 41, a Stille coupling using tributyl(vinyl)tin in the presence of Pd(Ph3)4 and LiCl gave diene 62 in 81% yield (Scheme 8). We next carried out the deconjugative alkylation reaction using alkyl iode 42, which proceeded with complete conversion to the desired product 63. However, despite successful formation, 63 was only isolated in small amounts due to rapid degradation observed almost instantly at –78 °C. Attempts to quench the reaction immediately after addition of the alkyl iodide were unsuccessful, and reducing the amount of KHMDS/18-crown-6 did not resolve the issue. To address the instability observed in previous alkylation attempts, we subjected diene 62 to a chemoselective epoxidation using mCPBA in a 1:1 mixture of DCM and saturated aqueous NaHCO₃, affording epoxides 64a and 64b in 87% yield with a 6.36:1 diastereomeric ratio. While the modification enabled a successful deconjugative alkylation—producing compound 65 as a single diastereomer in 75% yield—the strategy was ultimately abandoned. This was because the target O O I Ph OMe 54 OH O O AcO Ph OMe 56 OH O O Ph OMe 57 OH O Ph 50 OH O I O O I Ph OMe 51 O O I Ph 60 O O Ph 61 O O Ph 58 (0%) O O O Ph 59 (68%) O O AgBF4 NaHCO3 DMSO, 100°C + PPTS, py PhCl, 135°C (19%) OH https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 15 diastereomer 64b remained only a minor component, and subsequent attempts to open the epoxide moiety in 65 generated multiple side products. Scheme 8. Capitalizing on the high chemoselectivity observed in the epoxidation of 62, we revisited our retrosynthetic plan to feature an enyne epoxidation as a key transformation (Scheme 9). Epoxidation of compound 48 with mCPBA proceeded with excellent selectivity for the enyne moiety. Unexpectedly, the reaction favored the more sterically hindered face (top face), delivering epoxide 66 as the major product in 51% yield. This facial selectivity may arise from hydrogen bonding between mCPBA and the adjacent methyl ester. Interestingly, the minor bottom-face epoxide underwent an intramolecular ring-opening, likely acid-catalyzed by the m-chlorobenzoic acid by-product, furnishing compound 67 in 35% yield. Supporting this mechanism, repeating the epoxidation in a 1:1 mixture of DCM and saturated aqueous NaHCO₃ suppressed the rearrangement and cleanly yielded both epoxide diastereomers without formation of the ring-opened product. Both compounds 66 and 67 were successfully converted to hydroxylactone 68. Treatment of epoxide 66 with KOAc in DMSO at 145 °C, which unexpectedly promoted lactonization onto the methyl ester and simultaneous deprotection of the primary alcohol, delivering 68 in 44% yield. In a complementary route, lactone 68 was converted to alcohol 68 in 78% yield via a one-pot sequence involving acetylation treatment with Ac2O followed by TBS deprotection using TBAF. The primary alcohol was oxidized to the corresponding aldehyde using Dess–Martin periodinane in DCM, followed by conversion to dimethyl acetal 69 using trimethyl orthoformate and Amberlyst® 15 in 41 TfO CO2Me 62 CO2Me Bu3Sn Pd(PPh3)4 (2.5 mol%) LiCl, THF, rt (81%) CO2Me 63 OTBS 64a (major) CO2Me O CO2Me OCO2Me 65 OTBS O 2. m-CPBA, DCM/NaHCO3 (sat.) 0°C to rt (87%, dr = 6.4:1) + 64b (minor) 3. KHMDS, 18-Crown-6 THF then 42, -78°C (75%, dr > 20:1) I OTBS 42 1. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 16 a 1:1 mixture of DCM and MeOH under reflux overnight. This one-pot sequence afforded 69 in 75% yield. Subsequent treatment with DIBALH (–78 °C to 0 °C) removed the acetate and reduced the lactone, delivering lactol 70 in 90% yield (dr = 5:1). The crude lactol mixture was treated with anhydrous HCl (4 M in dioxane), promoting cyclization onto the dimethyl acetal to form the F-ring. Acetylation of the resulting secondary alcohol with Ac2O in the same pot afforded the separable anomers 71a and 71b in 96% combined yield (dr = 1.59:1). The anomers were separated at this stage to facilitate further functionalization. From 71a, allylic oxidation with SeO2 followed by Dess–Martin oxidation of the resulting alcohol furnished enone 72a as a single regioisomer in 85% overall yield (one-pot). In contrast, oxidation of 71b under the identical conditions produced a mixture of regioisomeric enones 78% combined yield (rr = 6.43:1), with 72b isolated as the major regioisomer in 68% yield. With both enones 72a and 72b, we pursue the synthesis of ginkgolides the stereoselective cuprate addition of the tert-Butyl group using previously developed conditions. In both cases, the desired ketones 73a and 73b were obtained in 81% and 80% yields respectively as the sole diastereomer. At this stage, the synthesis diverged to focus initially on the formal syntheses of ginkgolides A (1) and B (2). Beginning with 73b, the ketone was stereoselectively reduced using LiBH4 to afford the corresponding alcohol which upon treatment with 1 M NaOH induced a 5-exo-dig cyclization, forming the D-ring of 26 in 92% yield. Under these basic conditions, the acetate protecting group was also hydrolyzed, revealing the free alcohol at C3. Ozonolysis of enol ether 74b furnished led to the corresponding hydroxylactone in 91% yield, which was subsequently oxidized with 2-iodoxybenzoic acid (IBX) and 4-methoxypyridine N-oxide (MPO) in DMSO at 75°C to afford enone 75b in 41% yield.22 This sequence completed our 17-step synthesis of Corey’s intermediate (75b), which has been previously converted to ginkgolides A (1) and B (2) in 10 and 6 additional steps, respectively.12 The successful preparation of this late-stage intermediate therefore constitutes a formal synthesis of these complex natural products. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 17 Scheme 9. Having completed the formal syntheses of 1 and 2, we next turned our attention to the “pièce de resistance” of the series: ginkgolide C (3). The synthesis of ginkgolide C (3) requires, among other key transformations, the regioselective installation of a hydroxyl group at C7 (Scheme 10A). Our initial strategy involved α-oxygenation of ketone 73a using strong bases such as KHMDS, LiHMDS, or LDA in combination with Davis’ oxaziridine. Alternatively, we explored direct Rubottom oxidation of enol ether 77a, which was accessed via tert-butyl cuprate addition. Despite extensive experimentation, neither approach yielded the desired hydroxylated product 76a. Furthermore, attempts to generate enol ether 76a CO2Me Ph OTBS CO2Me Ph O O HO Ph 67 (35%) O O AcO Ph 68 O 48 1. m-CPBA DCM, 0°C to rt 66 (51%) + 3. KOAc, DMSO 145°C (44%) 2. Ac2O, Et3N, DMAP, DCM then TBAF, 65°C (78%) [one-pot] O AcO Ph 69 OMe O OMe OHO HO Ph OMe 70 OMe 4. DMP, DCM, rt then CH(OMe)3,Amberlyst 15 DCM/MeOH, rt (75%) [one-pot] 5. DIBALH, THF -78° to 0°C (90%, dr = 5:1) O O AcO Ph 71a (59%) OMe 6. HCl, THF then Ac2O, Et3N, DMAP, THF [one-pot] O O AcO O Ph 72b OMe 9. SeO2, dioxane, rt then DMP, dioxane/DCM, rt (78%, rr = 6.4 :1) [one-pot] 10. t-BuLi, CuCN TMSI, THF, -78°C then TBAF (80%) [one-pot] O O AcO O Ph 73b OMe O O AcO Ph OMe 71b (37%) O O AcO O Ph 72a OMe 7. SeO2, dioxane, rt then DMP, dioxane/DCM, rt (85%) [one-pot] 8. t-BuLi, CuCN TMSI, THF, -78°C then TBAF (81%) [one-pot] O O AcO O Ph 73a OMe 11. LiBH4, THF, rt then NaOH, THF/MeOH/H2O, rt (92%) [one-pot] O O O HO Ph OMe 74b 12. O3, DCM, -78°C then DMS (91%) 13. IBX, MPO DMSO, 75°C (41%) O O OO OMe O 75b O O O OOHR OO O HO Ginkgolide A (1) R = H Ginkgolide B (2) R = OH known steps OTBS OTBS OH In both case https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 18 directly from ketone 73a were unsuccessful, suggesting that either the ketone or the corresponding enol ether is too sterically hindered to undergo the desired transformation. Although the tert-butyl group at C8 sterically shields the bottom face of the B ring in compound 73a, we hypothesized that the phenylacetylene substituent adopts a conformation that projects over the top face of the ketone, impeding both enolate formation and subsequent electrophilic trapping (Scheme 10B). This spatial constraint likely prevents α-deprotonation or coordination to the electrophile from either face. As shown in 78a, we reasoned that repositioning the phenylacetylene moiety away from the ring system could relieve this steric congestion, thus enabling enolate generation and facilitating a diastereoselective α-oxygenation pathway. To this end, aqueous NaOH (1 M) was introduced following the addition of the tert-butyl cuprate reagent, effecting hydrolysis of the acetate group and simultaneously promoting a 5-endo-dig cyclization of the C3 hydroxyl onto the proximal alkyne to furnish ketone 78a in 80% yield. Gratifyingly, treatment of ketone 78a with KHMDS and Davis’ oxaziridine resulted in highly diastereoselective α-hydroxylation, delivering the desired α-hydroxy ketone 79a in 92% yield as a single diastereomer. This outcome suggests strong facial bias, likely governed by the steric environment imposed by the neighboring ring system. The α-hydroxy group was subsequently protected as a MOM acetal, affording compound 80a in nearly quantitative yield. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 19 Scheme 10. Having established a successful route to target intermediate 80a from enone 78a, we attempted to replicate this sequence with the a-anomer 72b. However, the transformation proved less straightforward than anticipated. The tert-butyl conjugate addition followed by 5-endo-dig cyclization proceeded efficiently, delivering ketone 78b in 77% yield as a one-pot transformation (Scheme 11A). In contrast to the b-anomer 78a, α-hydroxylation of ketone 78b with KHMDS and Davis’ oxaziridine yielded a separable diastereomeric mixture of the desired alcohol in 88% yield (C7 dr = 1.45:1). We attribute the reduced stereoselectivity to a steric interaction between the OMe group and the tert-butyl substituent, which likely forces the tert-butyl into a more equatorial orientation. This conformation diminishes the facial bias of the B-ring ketone, reducing discrimination between the two enolate faces during oxidation. The major α-hydroxy ketone was treated with MOMBr to afford 80b in 91% yield. The minor diastereomer was protected with MOMBr to furnish 81b as a mixture in 76% yield along with 12% of 80b. Building on this result, we took advantage of the facile epimerization at C7 to convert 81b to the desired epimer 80b. Treatment with tert-butoxide and 18-crown-6 in a THF/tert-butanol (70:1) mixture effected complete epimerization, affording 80b in 86% yield. O O AcO OTMS Ph 77a OMe O O OAc OMe 8 4 7 O Ph O O AcO O Ph 76a OMe OH 73a O O AcO O Ph 73a OMe [O] [O] O O O OMe 8 O 78a Ph [O] [O] 1. t-BuLi, CuCN TMSI, THF, -78°C then TBAF then NaOH THF/MeOH/H2O 75°C (80%) [one-pot] O O O 78a OMe 2. KHMDS, THF Davis’ ox., 78°C (92%, dr > 20:1) O Ph 72a O O O OMe O Ph OR 79a R = H 80a R = MOM 3. MOMBr, TBAI DIPEA, DCM 60°C (99%) A B C 7 https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 20 At this stage, both anomers were subjected to RuO₄/NaIO₄-mediated oxidative cleavage, affording lactols 82a/82b and lactones 83a/83b in good yields (Scheme 11B). The minor lactol intermediates 82a and 82b were readily oxidized to the corresponding lactones 83a and 83b using I2 and K2CO3, providing 91% and 55% yields, respectively. Formation of the D-ring lactone and deprotection of the C3 benzoate were accomplished in a one-pot sequence: diastereoselective reduction of the hemiketal moiety at C6 furnished the corresponding secondary alcohol, which upon treatment with NaOH revealed the C3 hydroxyl group. Subsequent addition of AcOH promoted lactonization, delivering lactones 84a and 84b in 73% and 76% yields, respectively. Both lactones were carried forward and oxidized with IBX and 4-methoxypyridine N-oxide (MPO) to provide the corresponding enones 85a and 85b, albeit in modest yields of 31% and 28%. Due to the inefficiency of this transformation, an alternative strategy was devised to improve the overall yield. Oxidation of the C3 alcohol followed by α-selenylation using anhydrous HCl and PhSeCl, and final oxidation with H2O2, furnished enone 85a in 75% yield over three steps. To our dismay, this protocol proved inconsistent and low yielding for the conversion of 84b to 85b. Next, both enones 85a and 85b were subjected to a thermally induced MeOH elimination in PhCl at 135 °C using PPTS and pyridine, following the protocol originally developed by Corey.12 Unfortunately, no desired product 86 (R = MOM) was isolated under these conditions. NMR analysis of the crude reaction mixture revealed the presence of an aldehyde byproduct, tentatively assigned as compound 87. We postulate that the MOM group at C7 was cleaved under the acidic conditions, unmasking a free hydroxyl that underwent intramolecular cyclization onto the adjacent E-ring, ultimately leading to F-ring opening and formation of 87. To prevent this undesired rearrangement, acetic anhydride was added to the reaction mixture to trap the liberated C7 alcohol as an acetate immediately upon MOM cleavage. With this modification, heating 85a and 85b at 135 °C in PhCl with PPTS, pyridine, and Ac2O successfully delivered the key intermediate 88 in 66% yield from 85a and 60% from 85b. Given that MOM deprotection could have posed significant challenges in later stages of the synthesis, this result not only https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 21 validated our modified conditions but also provided a more robust and convergent route to access ginkgolide (3). With our advanced intermediate 88 in hand, we embarked on the endgame of our synthetic sequence toward ginkgolide C (3). Following extensive experimentation to achieve regio- and stereoselective functionalization of the enone moiety, epoxidation across the C1–C2 olefin was successfully accomplished using trityl hydroperoxide (2.4 equivalents) in the presence of DBU (1.2 equivalents) in dichloromethane. This reaction furnished epoxide 89 as the sole diastereomer in 57% yield. Inspired by Corey’s precedent for C-ring construction via enolate alkylation, we next effected the formation of the final carbocyclic ring by treating 89 with the lithium enolate of tert-butyl propionate in a mixture of THF and HMPA (4:1).12 This transformation proceeded cleanly to afford the corresponding adduct in 59% yield. Subsequent acid-catalyzed epoxide opening and concomitant lactonization was achieved using camphorsulfonic acid (CSA), delivering tricyclic lactone 90 in 89% yield with complete stereocontrol. Efforts to install the F-ring lactone via osmium tetroxide-mediated dihydroxylation or epoxidation with meta-chloroperbenzoic acid (m-CPBA) proved unsuccessful, resulting in significant degradation of the substrate. Recognizing the sensitivity of this advanced intermediate, we adapted a milder oxidation protocol developed by Crimmins.13 Freshly prepared dimethyldioxirane (DMDO), generated in situ from acetone, enabled the formation of the highly labile epoxide 91, which was not isolated due to its instability. Instead, the crude epoxide was directly subjected to oxidation with Br₂ and NaOAc in a 1:1 mixture of acetic acid and water, furnishing the penultimate intermediate 92 in 57% yield over two steps. The diastereoselectivity observed in the epoxidation can be rationalized by steric shielding from the tert-butyl group, which blocks the external face and directs oxidation to occur on the interior face of the cage-like framework. The synthesis was completed by treating acetate 92 with potassium carbonate in methanol, effecting smooth deprotection to furnish (±)-ginkgolide C (3) in 95% yield. The spectral data (¹H NMR and ¹³C NMR) of the synthetic material were in full agreement with those reported for the natural product, thereby confirming the success of our total synthesis. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 22 Scheme 11. CONCLUSION In summary, we report the first total synthesis of (±)-ginkgolide C (3), completed in 26 steps from commercially available starting materials, concluding a synthetic journey that spanned over a decade. En O O O 78b OMe 2. KHMDS, THF Davis’ ox., -78°C (88%, dr = 1.4:1) 3. MOMBr, TBAI DIPEA, DCE 70°C O Ph O O O OMe O Ph MOMO 6. I2, K2CO3 DCM 60°C (91% and 55%) O O OHO BzO X 5. RuCl3 xH2O, NaIO4 CCl4/MeCN/H2O 50°C O O O OO BzO X O + 7. NaBH4, THF/H2O, 0°C then NaOH then AcOH, rt [one-pot] O O OO HO X 8. IBX, MPO DMSO, 75°C O O OO O MOMO OMOMOMOMH H MOMO 9. DMP, DCM, 0°C 10. PhSeCl, HCl, EtOAc/THF, rt 11. H2O2, Py, DCM/H2O, rt (75% over 3 steps for 85a only) 12. PPTS, Py Ac2O PhCl, 135°C (66% for 85a) (60% for 85b) O O OO O OR 13. Ph3COOH DBU DCM, -25°C (57%) O O OO O OAc 89 O 86 R = MOM 88 R = Ac O O O OOHHO OO O HO OAc O O O HO OO O HO OAc 14. EtCO2tBu LDA, THF/HMPA -78° to -30°C (59%) 15. CSA, DCM, rt (89%) 90 16. DMDO Acetone/H2O, rt 18. K2CO3 MeOH, rt (95%) 1. t-BuLi, CuCN TMSI, THF, -78°C then TBAF then NaOH THF/MeOH/H2O 75°C (77%) [one-pot] O O O OMe O Ph MOMO O O AcO O Ph 72b OMe + 80b (91%) 81b (76%) 4. t-BuOK, 18-Crown-6 THF/t-BuOH (70:1) -78°C (86%) O O O X O Ph MOMO 80a X = OMe, Y = H 80b X = H, Y = OMe Y 82a X = OMe, Y = H (18%) 82b X = H, Y = OMe (19%) 83a X = OMe, Y = H (73%) 83b X = H, Y = OMe (76%) Y Y Y 84a X = OMe, Y = H (83%) 84b X = H, Y = OMe (99%) O O O OOHHO OO O HO OH 85a X = OMe, Y = H (31%) 85b X = H, Y = OMe (28%) O O O HO OO O HO OAc 91 O [O] O O AcO O O O OHO HO 17. Br2, NaOAc AcOH/H2O, rt (57% over 2 steps) A B Ginkoglide C (3) 92 90 7 7 X Y 3 3 6 O O OO O O 87 https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 23 route, we also accomplished the formal syntheses of ginkgolides A (1) and B (2) by intercepting intermediate 75b in 17 steps—representing the shortest route to these targets reported to date. Key innovations include a scalable Claisen rearrangement to forge adjacent quaternary centers, a highly diastereoselective tert-butyl cuprate addition, and carefully choreographed C–C and C–O bond-forming transformations to assemble the intricate cage-like framework of ginkgolide C. Several late-stage challenges—most notably in regioselective epoxidation, enolate oxidation, and lactonization—were addressed through the development of one-pot protocols and strategic protecting group manipulations. Collectively, this work establishes a unified and stereocontrolled synthetic platform for accessing multiple members of the ginkgolide family, distinguished by their compact, hexacyclic, and highly oxygenated frameworks. The strategies developed herein lay the groundwork for the synthesis of additional ginkgolides and related analogues, enabling future exploration of their biological activity and therapeutic potential. ASSOCIATED CONTENT The data underlying this study are available in the published article and its Supporting Information SUPPORTING INFORMATION Detailed experimental procedures, full spectroscopic data for all new compounds and copies of NMR spectra ACKNOWLEDGMENTS This work was generously supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). M.H. and G.B. thank NSERC for doctoral graduate scholarship CGS-D. M.H. thanks the Government of Ontario and X-Chem for graduate scholarships (OGS and OGS-QEII). We thank Professor Michael Crimmins (University of North Carolina at Chapel Hill) for sharing some experimental data. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 24 REFERENCES a Present address: X-CHEM, 7171 rue Frederick Banting, Saint-Laurent, QC, Ca nada H4S1Z9. b Present address: Environment and Climate Change Canada, Place Vincent Massey, 351 Boulevard Saint-Joseph, Gatineau, QC J8Y 3Z4. (1) Furukawa, S. Studies on the Constituents of Ginkgo Biloba L. Leaves Part I and II. Sci. Pap. Inst. Phys. Chem. Res. 1932, 19, 27-42. (2) (a) Maruyama, M.; Terahara, A.; Itagaki, Y.; Nakanishi, K. The Ginkgolides. I. Isolation and Characterization of the Various Groups. Tetrahedron Lett. 1967, 8, 299-302. (b) Maruyama, M.; Terahara, A.; Itagaki, Y.; Nakanishi, K. The Ginkgolides. II. Derivation of Partial Structures. Tetrahedron Lett. 1967, 8, 303-308. (c) Maruyama, M.; Terahara, A.; Nakadaira, Y.; Woods, M. C.; Nakanishi, K. The Ginkgolides. III. The Structures of the Ginkgolides. Tetrahedron Lett. 1967, 8, 309- 313. (d) Maruyama, M.; Terahara, A.; Nakadaira, Y.; Woods, M. C.; Takagi, Y.; Nakanishi, K. The Ginkgolides. IV. Stereochemistry of the Ginkgolides. Tetrahedron Lett. 1967, 8, 315-319. (e) Woods, M. C.; Miura, I.; Nakadaira, Y.; Terahara, A.; Maruyama, M.; Nakanishi, K. The Ginkgolides. V. Some Aspects of their NMR Spectra. Tetrahedron Lett. 1967, 8, 321-326. (3) Weinges, K.; Hepp, M.; Jaggy, H. Chemistry of Ginkgolides, II. Isolation and Structural Elucidation of a New Ginkgolide. Liebigs Ann. Chem. 1987, 521-526. (4) Wang, Y.; Sheng, L. S.; Lou, F. C. Analysis and Structure Identification of Trace Constituent in the Total Ginkgolide by Using LC/DAD/ESI/MS. Acta. Pharmacol. Sin. 2001, 36, 606-608. (5) Liao, H.-J.; Zheng, Y.-F.; Li, H.-Y.; Peng, G.-P. Two New Ginkgolides from the Leaves of Ginkgo Biloba. Planta Med. 2011, 77, 1818-1821. (6) Bate, C.; Salmona, M.; Williams, A. Ginkgolide B Inhibits the Neurotoxicity of Prions or Amyloid-β1-42. J. Neuroinflammation 2004, 1, Article 4, 1-8. (7) Pedotti, R.; DeVoss, J. J.; Youssef, S.; Mitchell, D.; Wedemeyer, J.; Madanat, R.; Garren, H.; Fontoura, P.; Tsai, M.; Galli, S. J.; Sobel, R. A.; Steinman, L. Multiple Elements of the Allergic Arm of https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 25 the Immune Response Modulate Autoimmune Demyelination. Proc. Natl. Acad. Sci. 2003, 100, 1867- 1872. (8) Kim, B. K.; Shin, E.-J.; Kim, H.-C.; Chung, Y. H.; Dang, D.-K.; Jung, B.-D.; Park, D.-H.; Wie, M. B.; Kim, W.-K.; Shimizu, T.; Nabeshima, T.; Jeong, J. H. Platelet-Activating Factor Receptor Knockout Mice are Protected from MPTP-Induced Dopaminergic Degeneration. Neurochem. Int. 2013, 63, 121- 132 (9) Liu, Y.; Shields, L. B. E.; Gao, Z.; Wang, Y.; Zhang, Y. P.; Chu, T.; Zhu, Q.; Shields, C. B.; Cai, J. Current Understanding of Platelet-Activating Factor Signaling in Central Nervous System Diseases. Mol. Neurobiol. 2017, 54, 5563-5572. (10) (a) Villhauer, E. B.; Anderson, R. C. Synthesis of the CDE Ring System of the Ginkgolides. J. Org. Chem. 1987, 52, 1186 – 1189. (b) Schreiber, S.L.; Desmaele, D.; Porco, Jr.; J. A. On the Use of Unsymmetrically Substituted Furans in the Furan-Carbonyl Photocycloaddition reaction: Synthesis of Kadsurenone-Ginkgolide Hybrid. Tetrahedron Lett. 1988, 29, 6689 – 6692. (c) Corey, E. J; Gavai, A. V. Enantioselective Route to a Key Intermediate in the Total Synthesis of Ginkgolide B. Tetrahedron Lett. 1988, 21, 3201 – 3204. (d) Harrison, T.; Myers, P. L.; Pattenden, G. Radical Cyclizations onto 2(5H)-Furanone and Maleate Electrophore Leading to the Spiro- and Linear-Fused g-Lactone Ring Systems. Tetrahedron Lett. 1988, 29, 3869 – 3872. (e) Harrison, T.; Myers, P. L.; Pattenden, G. Radical Cyclizations onto 2(5H)-Furanone and Maleate Electrophore. An Approach to the Spiro- and Linear- Fused g-Lactone Ring Systems in the Ginkgolides. Tetrahedron 1989, 45, 5247 – 5262. (f) Corey, E. J.; Kamiyama, K. A Simple Stereoselective Synthesis of a Tetracyclic C14 Ginkgolide. Tetrahedron Lett. 1990, 31, 3995 – 3998. (g) Corey, E. J.; Rao, K. S. Enantioselective Total Synthesis of Ginkgolide derivatives Lacking the tert-Butyl Group, an Essential Structural Subunit for Antagonist of Platelet Activating Factor. Tetrahedron Lett. 1991, 32, 4623 – 4626. (h) DeLucas, M. R.; Magnus, P. Studies on the Synthesis of Ginkgolides. Rapid Construction of the Spiro[4.4]nonane A and B Rings. J. Chem. Soc., Perkins Trans 1991, 2661 – 2667. (i) Weinges, K.; Rümmler, M.; Schick, H.; Schilling, G. On the Preparation of the Ginkgolide Skeleton. Liebigs Ann. Chem. 1993, 287 – 291. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 26 (11) For the synthesis of related ginkgolide metabolite, see: (a) Baker, M. A.; Demoret, R. M.; Ohtaa, M.; Shenvi, R. Concise Asymmetric Synthesis of (-)-Bilobalide. Nature, 2019, 575, 643 – 646. (b) Corey, E. J.; Su, W. G. Total Synthesis of a C15 Ginkgolide, (±)-Bilobalide. J. Am. Chem. Soc. 1987, 109, 7534 – 7536. (c) Corey, E. J.; Su, W. G. Enantioselective Total Synthesis of Bilobalide, a C15 Ginkgolide. Tetrahedron Lett. 1988, 29, 3423 – 3426. (d) Crimmins, M. T., Jung, D. K.; Gray, J. L. Synthetic Studies on the Ginkgolides: Total Synthesis of (±)-Bilobalide. J. Am. Chem. Soc. 1993, 115, 3146 – 3155. (12) (a) Corey, E. J.; Kang, M.-C.; Desai, M. C.; Ghosh, A. K.; Houpis, I. N. Total Synthesis of (±)- Ginkgolide B. J. Am. Chem. Soc. 1988, 110, 649 - 651. (b) Corey, E. J.; Ghosh, A. K. Total Synthesis of Ginkgolide A. Tetrahedron Lett. 1988, 29, 3205 - 3206. (13) (a) Crimmins, M. T.; Pace, J. M.; Nantermet, P. G.; Kim-Meade, A. S.; Thomas, J. B.; Watterson, S. H.; Wagman, A. S. Total Synthesis of (±)-Ginkgolide B. J. Am. Chem. Soc. 1999, 121, 10249-10250. (b) Crimmins, M. T.; Pace, J. M.; Nantermet, P. G.; Kim-Meade, A. S.; Thomas, J. B.; Watterson, S. H.; Wagman, A. S. The Total Synthesis of (±)-Ginkgolide B. J. Am. Chem. Soc. 2000, 122, 8453 - 8463. (14) Posner, G. H.; Lentz, C. M. A Directing Effect of Neighboring Aromatic Groups on the Regiochemistry of Formation and on the Stereochemistry of Alkylation and Bromination of Ketone Lithium Enolates. Evidence for Lithium Arene Coordination and for a Dramatic Effect of Even Small Amounts of Copper(I) in Controlling Stereochemistry and in Limiting Polyalkylation. J. Am. Chem. Soc. 1979, 101, 934 – 946. (15) Hong J. H. Stereoselective Synthesis of Novel 4’-Branched and Biclyclo[3.1.0]hexane-Templated Nucleoside. Nucleosides Nucleotides Nucleic Acids 2004, 23, 625-635 (16) (a) Daub, G. W.; Griffith, D. A. The Stereoselectivity of the Ketal Claisen Rearrangements with Ketals of Simple Cyclic Ketones. Tetrahedron Lett. 1986, 27, 6311 - 6314. (b) Poulin, J.; Grisé-Bard, C.; Barriault, L. A Formal Synthesis of Vinigrol. Angew. Chem. Int. Ed. 2012, 51, 2111 – 2114. (17) Nakagawa, Y.; Chanthamath, S.; Shibatomi, K.; Iwasa, S. Ru(II)–Pheox-Catalyzed Asymmetric Intramolecular Cyclopropanation of Electron-Deficient Olefins. Org. Lett. 2015, 17, 2792-2795. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/ 27 (18) Chinchilla R.; Najera, C. The Sonogashira Reaction: A Booming Methodology in Synthetic Organic Chemistry. Chem Rev. 2007, 107, 874 – 922. (19) Heslin J. C.; Moody C. J. Rhodium carbenoid mediated cyclisations. Part 2. Synthesis of cyclic ethers. J. Chem. Soc., Perkin Trans. 1 1988, 1417-1423 (20) Echavarren, A. M.; Dorel, R. Gold(I)-Catalyzed Activation of Alkynes for the Construction of Molecular Complexity. Chem. Rev. 2015, 115, 9028-9072. (21) Kornblum, N.; Jones, W. J.; Anderson, G. J. A New and Selective Method of Oxidation. The Conversion of Alkyl Halides and Alkyl Tosylates to Aldehydes. J. Am. Chem. Soc. 1959, 81, 4113- 4114. (22) Nicolaou, K. C.; Montagnon, T.; Baran, P. S. Modulation of the Reactivity Profile of IBX by Ligand Complexation: Ambient Temperature Dehydrogenation of Aldehydes and Ketones to a,b- Unsaturated Carbonyl Compounds Angew. Chem. Int. Ed. 2002, 41, 993-996. https://doi.org/10.26434/chemrxiv-2025-ppm63 ORCID: https://orcid.org/0000-0003-2382-5382 Content not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0 https://doi.org/10.26434/chemrxiv-2025-ppm63 https://orcid.org/0000-0003-2382-5382 https://creativecommons.org/licenses/by-nc/4.0/