Postdoc Malone† & Günther Schlonk§*
Abstract
We report the total synthesis of the supernatural product mobiustrine A, starting from isoheptane. The key steps in our synthesis were a [2+2]+[2+2] cycloaddition, a Bonnie-Clyde cross-conjugal visitation, an enzymatic epoxide unzipping, and a Morbin rearrangement.
Mobiustrine A is a macrocycle polyketide, which was isolated from the fungus Kleptomyces negligencii in 2019 (Figure 1).2 Similar “extremophilic” organisms have been a rich font of interesting chemistry, most notably the enzyme Taq polymerase, which was originally obtained from bacteria inhabiting thermal springs.3 K. negligencii thrives in an environment far more inimical than a puddle of hot water: it grows in the fume hoods of Ivy-League universities. This mold embodies the idiom “life finds a way”. It has adapted to its environment masterfully, and subsists on a different kind of organic matter to all other detritivores: organic solvents. It is believed that the mobiustrines are enzymatic cofactors in a metabolic process that converts dichloromethane and hexanes to phosgene and CO2.4 This unique form of cellular respiration makes K. negligencii the only known petrol-powered fungus.
Quite aside from its fascinating and poorly-understood biochemistry, mobiustrine A also exhibits an interesting topological feature, in that it is a molecular Möbius strip. While not the first of its kind,5 it is self-evidently the coolest, and alongside its aforementioned metabolic import, this is a compelling motivator for its total synthesis. We decided to undertake this task, and we used a piece of paper as a model system to inform our retrosynthetic analysis (Scheme 1).
We began with amino aldehyde 2, which is prepared in two steps from the unnatural amino acid sarahpaline (Scheme 2). Addition of Hodenkobold’s reagent (3) followed by rearrangement and THP protection delivered allene 4. This compound was subjected to a Butkus oxidation with Dean–Martin’s phosiodinane (5) to obtain nitroso-furan 6. This was followed by a nitroso-Wittig reaction to generate intermediate 8, which underwent a spontaneous photochemical [2+2]+[2+2] cycloaddition. The product (9) was deprotected with radium triflate, followed by ring expansion, global tosylation and a Diels-Alder reaction to yield 12. The penultimate ring system was installed with a Bonnie–Clyde cross-conjugal visitation, followed by a Schutlipz cuprate addition, which set up compound 15 for an autophaliative conjugate addition. Finally, an intramolecular Hashwig–Buckwig–Hartwald coupling was used to generate the strained alkyne of compound 17, which was secured by X-ray crystallomancy.6 Turning next to the other half of mobiustrine A, we began by pouring a liter of isoheptane (24) and some bromine into a sunbed, removing the UV filters and turning the power to full blast. The ensuing mixture was distilled with the aid of a 10 m fractional column packed with buckyballs, and the fraction collected between 327 and 329 °C was determined to be our desired product (25). This tetrabromoheptane was subjected to elimination conditions, and the resultant diolefin was coupled with aldehyde 27 via a Wittig reaction. Subsequently, we used the conditions developed by the Arabian chemist Muhammad Salman Al-Dohl to condense the resultant aldehyde (28) with ester 29.7
With this coupling partner in hand, we set about the preparation of the azulene-containing counterpart (30). To this end, we reacted diarylzinc species 18 with iodide 19 in a Kumada reaction, and used Grubbs Lite© to perform an intramolecular olefin metathesis. A thermal rearrangement of the resultant phenanthrene (21) was used to construct the benz(e)azulene core (22). Care should be taken with this reaction, as the high temperatures can also cause the reactants to thermally rearrange into CO2 and H2O with some force, if oxygen is not excluded. We selected the Isuzu reaction to cross-couple the polyolefinic (30) and azulene (23) fragments on the grounds of novelty. This reaction is underutilized, probably because many chemists balk at using superstoichiometric quantities of rhodium as an expendable leaving group. We, however, just got a massive grant, so you can all suffer in your jocks. Also, stoichiometric rhodium isn’t really an issue when you only have 7 mg of material. The coupling proceeded smoothly, and we obtained a satisfactory yield of compound 31. Finally, we used barium perxenate to generate bromosobenzene in situ, which epoxidized 31 in excellent yield and abysmal stereoselectivity. This was of little import, however, as the next step was enzyme-catalyzed, and everyone knows that enzymes can do anything.
The enzyme davisase was developed in the laboratory of Kacey Nicolaou, to aid them in their eternal quest for a synthesis of maitotoxin. It functions as a “molecular zipper”, and was intended to facilitate the construction of maitotoxin’s core ring systems by chaining about 30 epoxides together. Unfortunately, it gets stuck after four epoxides, when another part of the substrate gets caught in the active site and jams it. It was perfectly sufficient for our purposes, however, and it elegantly collapsed all 256 isomers from reaction s into a single product (33). To unite the two halves of mobiustrine A, we subjected 33 and 17 to a second Diels-Alder-type [4+2] cyclisation, followed by a Vollpfosten radical confabulation to deliver the not-Möbius-strip 34. Finally, morbin time arrived, and we performed a Morbin rearrangement and imine reduction, which delivered mobiustrine A (1) in 29% yield. The mechanism of the morbin rearrangement, including all 29 transition states, will be made available on request.
CONCLUSION
Seasons change, the stars turn and empires rise and fall, yet some things endure. Benjamin Franklin wrote that only two things in this world are certain: death and taxes.1 He should have appended a third inevitability to his list, for as long as the sun shines, chemists will attempt to synthesize every obscure alkaloid, twisted terpene and perverse polyketide that ever nature devised. Untrammeled by lack of purpose or practicality, these obsessives will leave no stone unturned, in case there’s an algae harboring some novel compound underneath it. In this paper, we chose to make mobiustrine A, not because it was easy, but because we thought it would easily get into JACS. Damn.
REFERENCES
(1) Sparks, J. The Writings of Benjamin Franklin, Vol. X (1789-1790). 1856, Macmillan. p. 410.
(2) Davis, A.; Helsinki, V., It’s morbin time: a topologically challenged natural product. JAX, 2019, 5837, 2–6.
(3) According to Taq polymerase (personal communication)
(4) Ipsum, L. A petrol-powered fungus. CNS, 2019, 1, 4–444.
(5) Walba, D. M.; Richards, M. R.; Haltiwanger, C. R; Total synthesis of the first molecular Moebius strip, J. Am. Chem. Soc. 1982, 104 (11), 3219-3221.
(6) Rosiland, F.; Crockson, J.; Wick, F.; Schlonk, G. X-ray Crystallomancy: A Practical Guide. J. Immat. Sci. 2022, 2, 57.
(7) Al-Dohl, M. S. A novel condensation reaction, Orgas. Lett. 1899, 4, 650–666.