Total Synthesis of the Alleged Structure of Crenarchaeol Enables Structure Revision

Abstract Crenarchaeol is a glycerol dialkyl glycerol tetraether lipid produced exclusively in Archaea of the phylum Thaumarchaeota. This membrane‐spanning lipid is undoubtedly the structurally most sophisticated of all known archaeal lipids and an iconic molecule in organic geochemistry. The 66‐membered macrocycle possesses a unique chemical structure featuring 22 mostly remote stereocenters, and a cyclohexane ring connected by a single bond to a cyclopentane ring. Herein we report the first total synthesis of the proposed structure of crenarchaeol. Comparison with natural crenarchaeol allowed us to propose a revised structure of crenarchaeol, wherein one of the 22 stereocenters is inverted.


Introduction
In 1990, Woese proposed to classify all living organisms in three domains of life:A rchaea, Bacteria and Eukarya. [1] Before that, "archaeabacteria" were considered to belong to the Bacteria. Based on differences in their genome and lipidome,A rchaea were ultimately recognized as separate, third domain. [2] Foralong time,A rchaea were primarily associated with extreme habitats such as high temperature, extreme pH, and hypersaline environments. [3] Growing interest over the years,however,led to the discovery of meso-and extremophilic Archaea in virtually any habitat on Earth. [4] Thec ell membrane of Archaea is built up of diether or membrane-spanning tetraether lipids containing isoprenoid chains,contrary to the straight chain fatty acid glycerol ester lipids found in Bacteria and Eukarya. [5] Apart from the difference in lipid linkage,the stereochemistry of the glycerol backbone in archaeal isoprenoidal glycerol dialkyl glycerol tetraether lipids (GDGTs) is opposite to bacterial or eukaryotic glycerolipids,r aising questions on the evolution of archaeal and bacterial/eukaryotic lipid membranes. [6] The lipid composition of Archaea varies,depending on the species and environmental factors,a nd this is considered an adaptation to their habitat. [7] Thee ther-linkages provide chemical stability against hydrolysis,a nd the presence of methylbranches and cyclopentane moieties,w hich are formed by internal cyclization of the biphytanol chain, [8] leads to decreased membrane permeability,a llowing growth at extreme pH, salinity,a nd temperature. [9] One archaeal GDGT-named crenarchaeol-stands out from all other archaeal membrane lipids due to its unique chemical structure ( Figure 1). Crenarchaeol is produced by as pecific lineage of Archaea, the Thaumarchaeota, [10] and was first isolated from surface sediments of the Arabian Sea. After extensive GC-MS and NMR analysis,t he structure and stereochemistry of this unique GDGT was proposed, aconsiderable achievement given the fact that the molecular complexity originates merely from its unusual hydrocarbon framework. [11] It contains four 1,3-trans-substituted cyclopentane moieties.O ne of these is connected by asingle bond to acyclohexane ring, astructural feature rarely found in natural products. [12] This feature of crenarchaeol is likely formed by further internal cyclization of the bicyclic biphytanyl moiety. [5b] Crenarchaeol contains at otal of 22 stereocenters,m ost of which are remote, including an all-carbon quaternary stereocenter.R ecently, aparallel glycerol configuration of sedimentary crenarchaeol was inferred from chemical derivatization experiments. [13] Montenegro et al. confirmed the structure of the bicyclic biphytanyl moiety in archaeal GDGTs by total synthesis, [14] yet to date,t here is no proof of structure of the tricyclic biphytanyl moiety of crenarchaeol and no total synthesis.The 5-6 ring motif of crenarchaeol is particularly interesting due to its complexity and uniqueness in nature.I no rder to ultimately confirm the structure and stereochemistry of crenarchaeol, we embarked on its total synthesis.

Results and Discussion
Our retrosynthetic analysis of crenarchaeol made use of the inherent symmetry of the bicyclic biphytanyl chain of the molecule (Scheme 1). It started with the disconnection of the central CÀC-bond of the bicyclic biphytanyl moiety by intramolecular alkene metathesis and ether bond disconnection of 1.This led to two key intermediates,termed Fragment Aand B, and protected glycerol building block 2.Fragment A can be further simplified via dithiane disconnections to arrive at building blocks 4 and 6,b oth carrying am ethyl-branched stereocenter,and cyclopentanebuilding block 5.Inturn, 5 can be traced back to hydroxyketone 7,w hich is accessed from commercially available (S)-carvone via ring contraction. Syntheses of archaeal cis- [15] and trans-substituted [14] cyclopentane containing lipids have been previously reported. As we planned to build the macrocycle by alkylation of asuitably functionalized glycerol building block and ring-closing metathesis,w er equired differentially protected lipid chains containing the trans-substituted cyclopentane and the methyl-branches.B ased on the stereochemical assessment of the bicyclicb iphytane moiety in crenarchaeol [17] and its subsequent confirmation provided by Helmchen et al., [22] we planned the synthesis of the desired stereoisomer.
Retrosynthesis of Fragment Bcommenced with the CÀCbond disconnection of 8 arriving at dithiane 10 and iodide 9, the latter originating from Fragment A. Further simplification of 10 by asymmetric Cu-catalyzed Grignard alkylations and aWittig olefination delivered diacetate 11.The 5-6 ring motif of 11 was disconnected at the CÀC-bond joining the two carbocycles. [16] We realized that for this challenging transformation an advanced intermolecular Pd-catalyzed asymmetric allylic alkylation could be instrumental, inspired by the work of Tr ost. [17] By this,w ea rrived at building blocks 13 and 14, Scheme 1. Retrosynthetic analysis of crenarchaeol.

Angewandte Chemie
Research Articles readily accessible from pimelic acid and cyclopentadiene, respectively.

Synthesis of Fragment A
Thes ynthesis of Fragment Aw as initiated by the preparation of known b-hydroxyketone 7 from (S)-carvone (Scheme 2). Viaafour-step sequence involving ah ydrolytic ring contraction, [18] 7 was obtained as single diastereomer,a s confirmed by NOESY.N otably,t his sequence proved robust and scalable and allowed multigram synthesis of 7 (see Supporting Information). After acetal protection of 7,t he hydroxyl group of 15 was removed by Barton-McCombie deoxygenation, providing 16 in excellent yield. Notably, acetal protection was necessary to avoid elimination of the b-hydroxyl group in the synthesis of the xanthate intermediate.I nitially,w ee nvisioned to stereoselectively install the methyl stereocenter adjacent to the 5-membered ring by means of Cu-or Co-catalyzed asymmetric hydroboration. [19] No published method to perform the asymmetric hydroboration of the 1,1-disubstituted terminal alkene of 16 delivered 17 in acceptable yield and stereoselectivity,h owever.Thus,weresorted to non-stereoselective hydroborationoxidation of 16 followed by diastereomer separation, giving 17 in 43 %y ield as single stereoisomer.T he stereochemistry of the methyl-branched center in 17 was determined by amidation of its corresponding acid with phenylglycine methyl ester,f ollowed by 1 HNMR analysis (See Supporting Information). [20] In addition, the efficiencyo ft he synthesis was further increased by "recycling" of the undesired epi-17 by iodination and elimination, giving alkene 16 in 77 %yield over two steps. After silyl protection of 17,t he acetal moiety of 18 was removed. Optimization of the reaction conditions,t om inimize epimerization, resulted in treatment of 17 in acetone with FeCl 3 adsorbed to silica, [21] giving 19 in quantitative yield with 3% epimerization. Ketone 19 was converted to the corresponding terminal alkene by enol-triflation and Pdcatalyzed triflate reduction, delivering 20 in 80 %y ield over two steps.H ydroboration-oxidation of 20 gave alcohol 21 in 87 %y ield, which was converted to the corresponding bromide 5 in excellent yield. With 5 in hand, the stage was set for the first dithiane alkylation. [22] After optimization of the lithiation conditions of 4 (prepared using known methods, see Supporting Information), the alkylation proceeded in high yield (87 %) giving 22.D esilylation followed by Appel iodination delivered iodide 9,w hich serves as intermediate in the synthesis of both Fragment Aa nd B. In turn, after identification of the optimal lithiation conditions,deprotonation of dithiane 6 with n-BuLi at 0 8 8Cfollowed by addition of 9 produced bis-dithiane 3 in 68 %y ield. With the carbon skeleton of Fragment Ac onstructed, the dithiane moieties and the benzyl ether of 3 were removed by Raney-nickel reduction in good yield, thus concluding the synthesis of Fragment A.

Synthesis of Fragment B
Next, the considerably more complex Fragment Bwas to be constructed. Thesynthesis started with the preparation of two building blocks 14 and 27 (Scheme 3). Thes ynthesis of cyclopentene 14 started from meso-diacetate 24,accessible in two steps from cyclopentadiene. [23] Diacetate 24 was subjected to enzymatic desymmetrization [24] in excellent yield and ee,followed by silyl protection giving 14.Cyclohexanone 27 was prepared according to the method developed by the Stoltz laboratory from allyl cyclohexanone 13, [25] which was protected and subjected to hydroboration/oxidation to deliver 26. Omission of the protection of the ketone in 13 led to the formation of the corresponding hemiacetal. Benzylation and acetal hydrolysis provided the desired cyclohexanone 27 in 92 %y ield over two steps.
We started by screening ligands L1-L4 (Scheme 3) in combination with Pd 2 (dba) 3 CHCl 3 in order to achieve good  (Table 1, entry 1). Under the same conditions,( R)-t-ButylPHOX L2 failed to give chiral induction (Table 1, entry 2). When using DACH ligands L3 and L4,g ood diastereoselectivities of 81:19 and 86:14 were achieved (entries 3a nd 4), yet with al ow conversion of around 40 %a nd in the case of L4 only 27 % isolated yield. Since acceptable stereo-induction was achieved, we continued the optimization with L4.C hanging the solvent to toluene or DME (Table 1, entry 5a nd 6) did not result in higher conversion, but the latter gave the product with improved dr of 94:6. When using NaHMDS the conversion dropped significantly to around 10-15 % (Table 1, entry 7), while LDAp erformed comparable to LHMDS (entry 8). Ultimately,i ncreasing the equivalents of LHMDS to 1.6 and using LiCl as additive resulted in full conversion (Table 1, entry 9). Theproduct was isolated in 53 %yield with an excellent dr of 93:7.
We decided to apply these conditions to acetate 14 and cyclohexanone 27,and found this system to be superior to the model reaction. Product 30 was obtained in 67 %y ield with a dr > 20:1, and no undesired diastereomer detected (Scheme 3). This variant of the intermolecular Pd-catalyzed asymmetric allylic alkylation further expands the toolbox of this type of reaction and we expect it to open up new avenues for future asymmetric construction of joint ring systems in ac onvergent manner.
Progressing the synthesis of Fragment B, the ketone moiety was reduced and acetylated, giving 31 as single diastereomer.S ubsequent desilylation and acetylation delivered diacetate 11 in excellent yield. Notably,a ttempts to shorten this sequence by performing reduction, desilylation, and double acetylation led to significantly lower yields.T his was due to the formation of at ricyclic product arising from S N 2' addition of the non-allylic hydroxy group to the double bond (see Supporting Information). With diacetate 11 in hand, ar egioselective copper-catalyzed Grignard alkylation with 32 (prepared from (R)-citronellol, see Supporting Information) was performed providing ac rude dr of 4:1 and, after separation of the isomers,alkylation product 33 in 75 %yield as single stereoisomer.T he double bond of 33 was reduced by af lavin-catalyzed diimide reduction [26] followed by deacetylation providing 35 in 98 %y ield over two steps. Theh ydroxyl moiety of 35 was then removed by aB arton-McCombie deoxygenation reaction in excellent yield. After Pd-catalyzed hydrogenolysis of the benzyl ether in 36,alcohol 37 was oxidized to the corresponding aldehyde and subjected to aW ittig olefination delivering a,b-unsaturated thioester 39.The last methyl-branched stereocenter of Fragment Bwas then introduced in an excellent dr of 20:1 (see Supporting Information for details) by copper-catalyzed asymmetric conjugate addition of methylmagnesium bromide [27] producing 40 in 87 %y ield. With the last stereocenter of the biphytane core of crenarchaeol set, the dithiane moiety of 10 was installed, after MOM deprotection of 40,t hrough thioester reduction and treatment with 1,3-propanedithiol in the presence of BF 3 ·OEt 2 .D ithiane 10 was obtained in 82 % over the three steps.Notably,dithiane synthesis in presence of the MOM ether resulted in ac omplex mixture of 10 and various trans-acetalization products.W ith 10 in hand, the last dithiane alkylation was performed, in presence of the free hydroxyl group.A fter optimization of the lithiation conditions,t he reaction of lithiated 10 with iodide 9 smoothly provided the coupling product 8 in 67 %yield, containing the entire carbon-skeleton of Fragment B. Thes ynthesis of Fragment Bwas concluded by atwo-step sequence,involving removal of the dithianes with Raney-nickel, followed by Pdcatalyzed hydrogenolysis of the remaining benzyl ether.

Endgame-Completion of the Total Synthesis of the Proposed Structure of Crenarchaeol
After the successful stereoselective synthesis of both Fragment Aa nd B, the macrocycle of crenarchaeol was assembled (Scheme 4). Theendgame of the synthesis started with the O-alkylation of protected glycerol 2 with mesylate 41 prepared from Fragment A. During the reaction using sodium hydride in DMF,p artial cleavage of the TBDPS ether was observed. Therefore,a fter O-alkylation, the silyl ether was reintroduced, giving alkylation product 42 in 62 %yield. The trityl ether was removed delivering 43,t he substrate for the next ether synthesis,in94%yield. Thedouble O-alkylation of 43 with bis-mesylate 44 came about after considerable experimentation, by reaction with KOtBu as the base in toluene in the presence of TBAB as phase-transfer catalyst. After desilylation of the crude double alkylation product, the desired diol 45 was obtained in apoor yield of 27 %over the two steps.T here are multiple factors complicating this reaction. It is ad ouble O-alkylation of ab is-mesylate.T he sheer size and flexibility of this electrophile plays arole in the reaction rate as we expect that the site of alkylation is not always exposed for reaction with the weak alkoxide nucleophile.Inaddition, small amounts of elimination products were observed. Consequently,g iven the difficulty of this step,w e continued with the synthesis.Inorder to perform the final ring  13 CNMR of the crude product.
closure of the macrocycle, 45 was converted to bis-alkene 1 by oxidation and Wittig reaction. The66-membered macrocycle was closed by means of ring-closing metathesis with Grubbs 2 nd generation catalyst, am ethod often used for the construction of large rings. [28] This provided 46 in 65 %y ield, given the size of the produced macrocycle am ore than satisfactory result. In the final step,the double bond as well as the benzyl ethers were removed by hydrogenolysis with palladium on carbon in low yield of 34 %, which could be partially attributed to the scale of the reaction. This concluded the synthesis of this structurally complex lipid and provided 1.2 mg of synthetic crenarchaeol. With both synthetic crenarchaeol and the tricyclic intermediate Fragment B in hand we sought to investigate the chemical structure of natural crenarchaeol. Forthis purpose,were-isolated natural crenarchaeol in al aborious procedure (see Supporting Information) and made ac omparison of their NMR spectra. Furthermore,w ep erformed chemical derivatization in combination with GC-MS analysis.

Comparison of Natural Crenarchaeol and Fragment BbyGC-MS
TheB ligh Dyer extract of the thermophilic Thaumarchaeota "Ca. Nitrosotenuis uzonensis"(dominated by crenarchaeol and its cis-cyclopentyl isomer, [29] see Figure 2A)h as previously been treated with HI. This cleaves the ether bonds to produce amixture of biphytane diiodides. [29] Reduction of the iodides with H 2 /PtO 2 led to the corresponding hydrocarbons I-III,w hich were analyzed by GC-MS. [29] This showed ar atio of bi-and tricyclic biphytanes of approximately 1:1( Figure 2B). As ad irect comparison of the configuration of the tricyclic biphytane unit within synthetic and natural crenarchaeol was considered complicated, we subjected also Fragment Bt ot his derivatization (Figure 2A). [29,30] This enabled ap recise comparison by GC-MS. Tr eatment of fragment Bw ith HI followed by reduction yielded biphytane IV which appeared, as expected, as asingle peak in the gas chromatogram ( Figure 2C), but much to our surprise with asignificantly different retention time than the supposedly identical II derived from natural crenarchaeol. Them ismatch in chemical structure was confirmed by coinjection, showing retention time differences of IV and II or III of approximately 1.5 and 2min, respectively ( Figure 2D).
Next, we turned our attention to the mass spectra of II-IV (see Supporting Information). Thefragmentation patterns of natural II and III were equivalent to their previously reported mass spectra, [29,31] and featured the characteristic fragment m/z 262, originating from bond cleavage adjacent to the quaternary stereocenter.T his fragment was also clearly visible in the mass spectrum of synthetic IV.
Furthermore,the remaining fragmentation patterns of II/ III and IV are also virtually identical, providing strong evidence that the overall chemical connectivity of II/III and synthetic IV is identical. Thus,weconcluded that II and IVare stereoisomers.

Comparison of Fragment Bw ith Isolated Natural Crenarchaeol by NMR
In order to elucidate the exact structural difference between synthetic Fragment Ba nd the tricyclic biphytanyl moiety of natural crenarchaeol, we compared their NMR spectra. The 1 Hand 13 Csignals of natural crenarchaeol [11] and Fragment Bwere assigned by thorough 2D NMR analysis.In addition, the 13 Cs ignals of synthetic crenarchaeol were assigned based on the NMR analysis of Fragment B.
Thecomparison of selected 13 CNMR signals of Fragment Band synthetic crenarchaeol with those of natural crenarchaeol is shown in Table 2( see Supporting Information for atable with all signal assignments).
Thec arbon numbering is shown in Figure 3, and significant differences in 13 CNMR shifts between Fragment Ba nd natural crenarchaeol are marked in orange (Dd = 0.25-1ppm) and red (Dd > 1ppm). Upon comparison of the 13 CNMR signals of Fragment Bw ith those of natural crenarchaeol, [11] it becomes clear that the majority of the chemical shifts of Fragment Ba re in very good agreement (Dd < 0.25 ppm) with those of the tricyclic biphytane of crenarchaeol. In particular,the 13 Cchemical shifts of the three cyclopentane rings (which are not connected to the cyclohexane ring) and their alkyl substituents are virtually identical (see Supporting Information). Moderate chemical shift differences (Dd = 0.25-1 ppm) were ascribed to the cyclohexane ring carbons (A11',A 12',A 13' and A15')a nd the alkyl chain adjacent to the cyclohexyl ring (A16). Large differences (Dd > 1ppm) in chemical shift at the (sub)terminal carbons (A1, A1',A 2a nd A2')o f Fragment Bo riginate from the presence of primary hydroxyl moieties contrary to the ether linkages in crenarchaeol. More importantly, however, three carbon atoms  [a] Assignments of 13 CNMR chemical shifts of crenarchaeol [11] and Fragment B. Signals are reported relative to the solvent residual signal (CDCl 3 d = 77.16 ppm).
[b] Correspondings ignals of synthetic nominal crenarchaeol are shown in brackets. around the all-carbon quaternary stereocenter elicit large differences in chemical shift at positions A14' (Dd = À5.96 ppm), A16' (Dd = À4.29 ppm) and A20' (Dd =+ 7.57 ppm), indicating ad ifference in structure around these positions.Itisnoteworthy that the 13 Csignals of the remaining stereocenters of the 5-6-ring system (A10' and A7')i n Fragment Bs how no significant difference.I np articular the good agreement of A10' is indicative for the ascribed stereochemistry of the single bond connecting the 5-and 6membered ring. It is expected that ad ifference in stereochemistry on A11' would translate to as ignificant 13 C chemical shift difference in A10'.This indicates that, on these positions,t he chemical structure of natural crenarchaeol matches that of Fragment B. Thechemical shifts of synthetic nominal crenarchaeol (chemical shifts in brackets in Table 2) show the same pattern of chemical shift differences.Itshould be highlighted that there is no significant 13 Cc hemical shift difference between Fragment Band the tricyclic biphytane of synthetic crenarchaeol (except for the terminal carbons A1/ A1' and A2/A2')e xcluding an influence of the macrocyclic structure on the chemical shifts.
Besides the good agreement of most of the 13 CNMR chemical shifts of crenarchaeol and Fragment B, the 1 HNMR chemical shifts of A7',A10' and A11' correlate well (Table 3, see Supporting Information for all assignments). At position A19' (axial) and A20',o nly minor 1 Hs hift differences were observed. Only three positions show significant chemical shift differences:t he equatorial proton of A14' (Dd = 0.27 ppm), A16' (Dd = 0.47 ppm) and the equatorial proton of A19' (Dd = 0.15 ppm). This provides further evidence that the difference in structure of natural and synthetic crenarchaeol is located around these positions.
Since the relative and absolute stereochemistry of Fragment Bisknown, the methyl substituent A20' of Fragment B is assigned to be equatorial due to the 1,3-cis relationship of the methyl and cyclopentyl substituents on the cyclohexane ring. As ar esult of the deshielding g-gauche effect, the 13 CNMR chemical shift of axial substituents in cyclohexanes is more upfield relative to equatorial substituents. [32] In Fragment Bt he 13 Cs ignal of methyl group A20' resonates at 30.12 ppm, while the methyl group A20' of natural crenarchaeol is shifted more upfield at 22.55 ppm. This strongly suggests that the methyl group A20' in natural crenarchaeol is in axial position in contrast to the initially proposed structure.T ofurther support this,weconsidered the 13 Cc hemical shifts of A16'.I nF ragment B, the carbon atom A16' of the alkyl side-chain of the cyclohexyl ring is axially oriented. The 13 Cs ignal resonates at 33.51 ppm, whereas in crenarchaeol the 13 Cs ignal of A16' is shifted downfield to 37.80 ppm. Thus,t he downfield shift of A16' in natural crenarchaeol strongly suggests equatorial substitution of the alkyl chain substituent on the cyclohexyl ring.
Further support comes from the computationally calculated 13 Cshift values for A16' and A20'.First, MD simulations in chloroform were carried out on fragment Band its isomer to determine the lowest energy conformations.Subsequently, the energies of the conformers from the MD trajectory were evaluated using DFT calculations and the chemical shifts calculated (See Supporting Information for the protocol and the calculated shifts). TheD FT prediction is in good agreement with the upfield shift of methyl group A20' in natural crenarchaeol and the expected downfield shift of methylene A16'.
All in all this combined data provides overwhelming evidence for an inverted stereochemistry of crenarchaeol at A15' compared to Fragment B. On the basis of the evidence from chemical derivatization, NMR studies,and computation, we therefore propose ar evised structure of crenarchaeol ( Figure 4), in which the stereochemistry of the all-carbon quaternary stereocenter is inverted compared to the original proposal.

Conclusion
Thef irst total synthesis of the originally proposed structure of the thaumarchaeotal GDGT crenarchaeol has been achieved. Thes ynthesis involved the stereoselective construction of aunique 5-6 ring motif as well as alate-stage 66-membered macrocyclization by means of RCM. The structure determination of crenarchaeol has ac onsiderable history. [11] Due to the very complex structure,i ncluding 22 stereocenters,aswell as the highly aliphatic character and its lack of rigidity,N MR-based structural studies have been heavily complicated. Furthermore,s ince this lipidic molecule does not have the tendency to crystallize,X -ray diffraction was not possible.T he synthesis of the proposed structure of crenarchaeol and the key intermediate Fragment Benabled direct comparison with natural crenarchaeol by chemical derivatization and GC-MS analysis.T his revealed amismatch of the chemical structure of the tricyclic biphytane chain. Subsequently,detailed NMR analysis including computational simulation of 13 C chemical shifts,o fF ragment Ba nd synthetic cren- [a] Assignments of 1 HNMR chemical shifts of crenarchaeol [11] and Fragment B. Signals are reported relative to the solvent residual signal (CDCl 3 d = 77.16 ppm). archaeol, and comparison with natural crenarchaeol isolated from sea surface sediments was performed. Ultimately,f rom the spectroscopic data of fragment B, synthetic and natural crenarchaeol, we were able to revise the originally proposed structure beyond reasonable doubt. Through this extensive analysis we identified the inversion of just one out of the 22 stereocenters of crenarchaeol, namely the quaternary stereocenter embedded in the cyclohexane ring. Total synthesis not only comprises the access to complex molecules,b ut serves also as ab reeding ground for new synthetic methodology as well as probing current synthetic methods.M istakes in the proposed structure of an atural product are by no means ar are occurrence. [33] Thea rchitectural and stereochemical complexity of an ew unknown structure,incombination with very small amounts of isolated material often make assignments extremely difficult, in particular in ac ase such as crenarchaeol, which features almost no heteroatom functionalities and is highly flexible.By using the information gathered from the synthetic epimer of natural crenarchaeol, we were able to reassign the structure without the need to repeat the entire,very complex, synthesis.
Thec orrection of the structure of crenarchaeol has important implications for the study of its role in archaeal membranes.T he current hypothesis is that the presence of crenarchaeol regulates membrane fluidity and packing,a n important adaptation to temperature and pressure changes in the environment. As the stereochemistry of the quaternary center in crenarchaeol has as ignificant influence on its conformation, and thus membrane packing, we expect that an explanation (supported by for instance molecular dynamics simulations) for its role in membrane behavior is now within reach.