On the Total Synthesis of Archaeal and Mycobacterial Natural Products
Holzheimer, Mira
DOI:
10.33612/diss.150711132
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2021
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Holzheimer, M. (2021). On the Total Synthesis of Archaeal and Mycobacterial Natural Products. University of Groningen. https://doi.org/10.33612/diss.150711132
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Archaeal Membrane Lipids and
Crenarchaeol
This chapter describes the lipid composition of archaeal cell membranes as well as the differences compared to bacteria and eukarya, the biosynthesis of archaeal (membrane-spanning) lipids and the adaptation of lipid content to environmental factors. In addition, the proposed structure and geochemical implications of a unique archaeal Glycerol Dialkyl Glycerol Tetraether (GDGT) – crenarchaeol – are summarized. Lastly, in this chapter the initial retrosynthetic disconnections of crenarchaeol are outlined in regard to its chemical synthesis.
Introduction – Archaea and archaeal lipids
In 1990 Carl Woese formally proposed that all living organisms on Earth are divided in three domains of life: Archaea, Bacteria and Eukarya (Fig. 1).1 The
latter two were already considered as such before 1990 but Archaea have long been considered to be part of the Bacteria. Eventually Archaea were recognized as the third domain of life based on differences in their genome and lipidome.2-4 Archaea were for long mainly associated with extreme
environments (i.e. high temperatures, extreme pH, hypersalinity)5 but
growing interest led to the discovery of archaea in virtually any place on Earth.6, 7
Fig. 1 Phylogenetic tree proposed by Woese.1
Archaeal vs. bacterial and eukaryotic cell membrane lipids
One of the most striking features of archaea, which sets them apart from bacteria and eukarya, is the composition of their cell membrane. Archaeal cell membranes are composed of prenol (isoprenoid) lipid chains connected to a glycerol backbone by ether linkages, in contrast to the ester linkages predominantly found in eukaryotic and bacterial membrane lipids (Fig. 2). Lipids based on ether linkages are not entirely exclusive to archaea, but in bacteria and eukarya these only occur in very minor amounts, and in very few species in somewhat larger amounts.8 Besides the differences in lipid linkage,the stereochemistry of the glycerol backbone in archaea is an sn-glycerol-1-phosphate (sn-G1P). This is opposite to the stereochemistry in membrane lipids of bacteria and eukarya, which produce sn-glycerol-3-phosphate (sn-G3P) based lipids (Fig. 1.2). This, as far as we know now, strict difference in stereochemistry raised questions on how the evolution of archaeal and
Fig. 2 Glycerol backbone configuration and cell membrane composition of archaea (left) and
bacteria/eukarya (right).
Another difference of archaea compared to eukarya and bacteria is the fact that archaeal membranes are composed of isoprenoid diether lipids and/or macrocyclic, membrane-spanning glycerol dibiphytanyl glycerol tetraether (GDGT) lipids. Archaeal membranes composed of GDGTs are arranged as lipid monolayers, whereas the cell membranes of eukarya and bacteria are composed of linear fatty acid glycerol diesters resulting in formation of lipid bilayers.
Archaeal membrane lipid composition as adaptation to
environmental factors
It has been found that the membrane lipid composition of archaea varies greatly depending on the species as well as on environmental factors such as growth temperatures. After the analysis of membrane lipid content of many strains of archaea, a general trend became obvious: archaea growing at high temperatures tend to produce membranes mainly composed of membrane-spanning GDGTs with or without cyclopentane rings in their biphytanyl chains. Additionally, archaea thriving in extreme acidic environments tend to produce membranes with mainly GDGTs whereas archaea from extreme alkaline environments produce predominantly lipid bilayers with diether lipids.14-16 The presence of both cyclopentane rings and a cyclohexane ring
as found in crenarchaeol is, though, a distinct feature of archaea belonging to the thaumarchaeota phylum.17-19 Archaea have found to be very adaptive
to changes in growth temperatures. Thermoacidophilic and
(hyper)thermophilic archaea respond to increased growth temperatures by increasing the level of tetraether lipids compared to diether lipids as well as the number cyclopentane rings within the tetraether lipids in their
O OH PO3 2-OH O OH PO3 2-OH sn-G1P sn-G3P Archaea Bacteria/Eukarya HO OH OH sn1 sn2 sn3
membranes.20-23 Furthermore, cold adaptation has been observed in
psychrophilic archaea. Decreased growth temperatures result in an increased degree of unsaturation in the isoprenoid chains of the membrane lipids.24, 25
The distinct structural features of archaeal membrane lipids are thought to contribute to the ability of archaea to survive in hostile and extreme habitats. The ether linkages in archaeal diether and tetraether lipids are chemically more stable towards hydrolysis at high temperatures and extreme pH than the ester linkages in bacterial and eukaryotic membrane lipids. The low degree of unsaturation in archaeal lipid membranes reduces susceptibility to oxidation. The methyl branches and cyclopentane rings in the hydrocarbon chains of archaeal lipids lead to a more densely packed membrane, thus reducing permeability for ions and allowing archaea to grow in high salinity and strongly acidic conditions.26 The presence of membrane spanning lipids
which arrange as lipid monolayers is thought to result in a stiffened membrane and protects the archaeon from high temperature.27-32
Furthermore, the presence of methyl branches in archaeal lipids results in enhanced membrane fluidity, which is characterized by an increased temperature range in which the membrane is in a liquid crystalline state. This allows archaea to grow even at temperatures below the freezing point of water.33 These distinct properties of archaeal lipids and membranes not only
interest researchers from an evolutionary perspective but also from a practical perspective. Liposomes (archaeasomes) based on archaeal lipids are found to be attractive for biotechnological applications, such as drug and vaccine delivery systems.30, 34-37
Biosynthesis
The biosynthesis of archaeal membrane lipids can be subdivided in multiple stages: the synthesis of sn-G1P, the synthesis of isoprenoid pyrophosphate building blocks, and subsequently the condensation to long chain isoprenoids, the polar headgroup attachment, the cyclization to form the macrocyclic GDGT lipids and finally the formation of the cyclopentane rings and the cyclohexane ring of crenarchaeol. To this date, the full biosynthetic pathway has not been elucidated yet. Research is ongoing to fully understand key steps such as the mechanism of the macrocyclization and the formation of the five- and six-membered ring motifs, as well as the exact order of the
Biosynthesis of the sn-G1P backbone
Scheme 1 Biosynthesis of archaeal sn-G1P.
As said, the glycerol backbone of archaea is distinguished by its opposite stereochemistry compared to its bacterial and eukaryotic counterpart. Archaeal sn-G1P is biologically produced from dihydroxyacetone phosphate (DHAP) which originates from glycolysis or gluconeogenesis pathways (Scheme 1). The reduction of DHAP to sn-G1P is catalyzed by the enzyme G1P dehydrogenase, which utilizes nicotinamide adenine dinucleotide (NADH) or nicotinamide adenine dinucleotide phosphate (NADPH) in their reduced form as cofactor.40, 41 The archaeal G1P dehydrogenase – though
mediating the same reaction – does not share homology to the bacterial G3P dehydrogenase.42 G1P dehydrogenase is a Zn2+-dependent enzyme43
conserved in archaea and transfers the pro-R-hydrogen of NAD(P)H to DHAP.44
Isoprenoid pyrophosphate biosynthesis
For the synthesis of the isoprenoid lipid chains found in archaeal membrane lipids, two C5-building blocks are needed: isopentenyl pyrophosphate (IPP)
and its isomer dimethylallyl pyrophosphate (DMAPP). These small isoprene units are, depending on the organism, synthesized by different pathways, namely the mevalonate (MVA) and the 2-C-methyl-D-erythritol-4-phosphate (MEP) pathway. The former is prevalent in eukaryotes, archaea and few bacteria whereas the latter is the major pathway in bacteria and chloroplasts.45
To this date, besides the ‘classical’ MVA pathway, multiple alternative MVA pathways have been discovered and elucidated.46-48 The ‘classical’ MVA
pathway (Scheme 2, red arrows) comprises of seven enzymatic steps starting from acetyl-CoA. This pathway has only been identified in very few archaeal species (Sulfolobus tokadaii and S. solfararicus),49, 50 but is the major
pathway in eukaryotes.45 The first three enzymatic steps (Scheme 2, black
arrows) are shared among all described MVA pathways and consist of the condensation of two molecules acetyl-CoA catalyzed by acetoacetyl-CoA thiolase to give acetoacetyl-CoA. Acetoacetyl-CoA is converted to
3-hydroxy-O OH P O O O sn-G1P OH O OH P O O O DHAP G1P dehydrogenase glycolysis or gluconeogenesis NAD(P)H NAD(P)+ O
3-methylglutaryl-CoA (CoA) by the enzyme CoA synthase. HMG-CoA is subsequently reduced by HMG-HMG-CoA reductase to MVA.45 The classic
MVA pathway then continues with the phosphorylation of mevalonate to mevalonate-5-phosphate (MVAP) by mevalonate-5-kinase (MVK). A second phosphorylation catalyzed by phosphomevalonate-5-kinase (PMK) produces mevalonate-5-diphosphate (MVAPP), which is then decarboxylated by mevalonate-5-diphosphate decarboxylase (MDD), giving IPP.
Scheme 2 MVA pathways occurring in archaea for the biosynthesis of the isoprenoid building
blocks IPP and DMAPP. SCoA O acetyl-CoA acetoacetyl-CoA thiolase SCoA O O acetoacetyl-CoA SCoA O HMG-CoA synthase HO O OH HMG-CoA HMG-CoA reductase OH HO O OH mevalonate O HO O OH mevalonate-5-phosphate P O O O MVK O HO O OH mevalonate-5-diphosphate P O O O P O O O PMK M3K OH HO O O mevalonate-3-phosphate P O O O O HO O O P O O O P O O O M3P5K O P O O O mevalonate-3,5-bisphosphate MBD isopentenyl phosphate O HO O P O O O trans-anhydromevalonate-5-phosphate PMD AHMPD MPD O P O O O isopentenyl pyrophosphate P O O O IPK MDD OP O O O dimethylallyl pyrophosphate P O O O IPI MVA pathways ⟶ ‘classical’ ⟶ altMVA1 ⟶ altMVA2 ⟶ altMVA3
IPP is isomerized to DMAPP by isopentenyl pyrophosphate isomerase (IPI).49
Homologs of MVK, PMK and MDD are absent in most archaea which led to the hypothesis that archaea possess an alternative MVA pathway.51 The first
alternative MVA pathway (altMVA1) was discovered in Haloferax volcanii in 2013 (Scheme 2, orange arrows). In this pathway, MVAP is directly decarboxylated by mevanolate-5-phosphate decarboxylase (MPD) to isopentenyl phosphate (IP) which is phosphorylated by isopentenyl kinase (IPK) to produce IPP.52, 53 A second alternative MVA pathway (altMVA2) was
proposed due to the characterization of two enzymes in Thermoplasma
acidophilum with previously unknown activity (Scheme 2, blue arrows). It was
found that these enzymes are 3-kinase (M3K) and mevalonate-3-phosphate-5-kinase (M3P5K).47, 54 The last predicted enzyme for the
completion of the biosynthetic altMVA2 pathway was discovered in 2016 and denoted mevalonate-3,5-bisphosphate decarboxylase (MBD) as it catalyzes the decarboxylation of mevalonate-3,5-bisphosphate to IP.55 A third
alternative MVA pathway (altMVA3) has been discovered in the archaeon
Aeropyrum pernix proceeding through a previously unknown biosynthetic
intermediate (Scheme 1.2, green arrows). In altMVA3, MVAP is dehydrated to trans-anhydromevalonate-5-phosphate (tAHMP) mediated by the enzyme mevalonate-5-phosphate dehydratase (PMD). tAHMP is then converted to IP by trans-anhydromevalonate-5-phosphate decarboxylase (AHMPD).48 It has
been suggested that the atlMVA3 pathway is the pathway which is predominant in the majority of archaea, whereas the classical MVA, altMVA1 and altMVA2 are found predominantly in Sulfolobales, Haloarchaea and Thermoplasma, respectively.48
Archaeal isoprenoid lipid biosynthesis
The next stage of the biosynthesis commences with the condensation of DMAPP with several IPP molecules resulting in elongation of the isoprenoid chains by C5-units producing geranyl (GPP, C10), farnesyl (FPP, C15),
geranylgeranyl (GGPP, C20) and farnesylgeranyl (FGPP, C25) diphosphate
(Scheme 1.3). These chain elongation steps are mediated by enzymes that belong to the family of the prenyl transferases and are also called isoprenyl diphosphate (IPP) synthases. Prenyl transferases are conserved in all three domains of life.3, 56, 57
There are multiple different prenyl transferases known in archaea which catalyze IPP addition depending on the chain length. In archaeal diether
lipids, the chain length is mostly C20 (GGPP) or C25 (FGPP) and in the case
of archaeal GDGTs, a chain length of C40 is prevalent. The synthesis of these
long isoprenoid chains has been subject to numerous studies and has been reviewed extensively.3, 38, 39
Ether bond formation and headgroup attachment
The synthesis of the first and second ether linkage between sn-G1P and GGPP is catalyzed by the enzymes geranylgeranylglyceryl diphosphate (GGGP) synthase and di-O-geranylgeranylglyceryl diphosphate (DGGGP) synthase, respectively (Scheme 3).
GGGP synthase is conserved among all archaea that are able to synthesize lipids58 (an exception is Nanoarchaeum equitans,59 a symbiont which obtains
its lipids from the host). The second ether bond formation to give DGGGP is mediated by the intrinsic membrane enzyme DGGGP synthase, a member of the ubiquinone-biosynthetic (UbiA) prenyl transferases. This family of enzymes catalyzes the transfer of a prenyl group onto acceptors that generally feature apolar ring structures. The archaeal DGGGP synthase represents an exception to this as the acceptor in the archaeal ether synthesis does not contain ring structures. This enzyme shows rather strict substrate specificity regarding the prenyl moiety,60, 61 but yet accepts both
enantiomers of the glycerol backbone.62
Next in the biosynthetic pathway, DGGGP is activated by cytidine triphosphate (CTP) and converted into cytidine diphosphate-archaeol (CDP-archaeol). This reaction is catalyzed by CDP-archaeol synthase (CarS), which is a Mg2+-dependent enzyme and conserved among archaea (except
Nanoarchaeum equitans). The enzyme activity of CarS is specific for glycerol
diethers bearing unsaturated isoprenoid chains but not specific for the stereochemistry of the glycerol backbone.63, 64
Scheme 3 Archaeal isoprenoid lipid biosynthesis. Condensation of DMAPP with multiple units
IPP produces GGPP. Biosynthesis of ether linkages to sn-G1P by GGGP and DGGGP synthase followed by attachment of CDP produces CDP-archaeol.
The polar headgroups serine, myo-inositol, ethanolamine and glycerol are found in the phospholipids of archaea, bacteria and eukaryotes. In archaea, the polar headgroups are attached to CDP-archaeol by replacing the cytidine monophosphate (CMP) unit with the respective polar headgroup (Scheme 4). The enzymes which catalyze these reactions in archaea are members of the CDP-alcohol phosphatidyl transferase family and share homology with their bacterial counterparts.10 The enzyme archaetidylinositol phosphate (AIP)
synthase mediates the reaction between CDP-archaeol and L-myo-inositol-1-phosphate, and produces AIP. Dephosphorylation of AI phosphate by AIP
DMAPP PPO GPP IPP PPO FPP IPP synthase PPO GGPP O PO OH OH PO OH sn-G1P GGGP GGGP synthase O PO O DGGGP DGGGP synthase GGPP DHAP G1P dehydrogenase IPP IPP synthase IPP IPP synthase O O CDP-archaeol O O O O P O O OP NH2 O N HO HO N O CarS
phosphatase produces AI.65 The reaction between CDP-archaeol and L
-serine is catalyzed by archaetidyl-serine (AS) synthase.66 Furthermore,
homologs of the bacterial and eukaryotic phosphatidylserine (PS) decarboxylase and phosphatidylglycerol (PG) synthase have been identified in archaea and termed AS decarboxylase, which converts AS to archaetidylethanolamine (AE), and archaetidylglycerol (AG) synthase, which produces AG from CDP-archaeol.67
Scheme 4 Attachment of the polar headgroups ethanolamine, glycerol and inositol. GDGT synthesis – biosynthetic hypotheses
There are still multiple missing pieces in the biosynthesis of archaeal macrocyclic tetraether lipids. It is under debate at which stage of the biosynthesis the double bonds of the isoprenoid chains are reduced, how and
O O AG O O O OP OH OH O O AS O O O OP O OH H2N O O AE O O O OP NH2 O O O O OP O O O P O OH OH OH HO O O O O O OP OH OH OH OH HO O AIP AI CDP-archaeol AS decarboxylase AS synthase AG synthase AIP synthase AIP phosphatase
cyclopentane rings, found in many GDGTs, and the cyclohexane ring in crenarchaeol are synthesized.38, 39
Scheme 5 Current biosynthetic hypotheses for the formation of membrane-spanning GDGTs
from GGPP. Red arrows: macrocyclization from unsaturated precursors. Blue arrows: macrocyclization from saturated precursors.
The double bonds of the intermediates in the biosynthesis of archaeol are reduced stereospecifically through syn-addition of hydrogens by the enzyme digeranylgeranylglycerophospate reductase, a member of the family of geranylgeranyl reductases (GGR).68-70 The double bond nearest to the
glycerol backbone is reduced first by GGR and the one furthest away from the glycerol is hydrogenated last.71 To this date, it is not known at what stage
in the biosynthesis the reduction happens. It has been suggested that reduction takes place after synthesis of CDP-archaeol, as CarS – the enzyme responsible for the formation of CDP-archaeol – has a strong preference for unsaturated substrates.63 Furthermore, it has been proposed that
hydrogenation might take place after polar headgroup attachment.72
O O O headgroup O O O headgroup O O O headgroup O O O headgroup O O O headgroup O O O headgroup O O O headgroup O O O headgroup O O O H H O O O H H headgroup headgroup O O O H H O O O H H headgroup headgroup H H H H H H H H O O O headgroup O O O headgroup H H H H H H H H O O O headgroup O O O headgroup ‘GDGT synthase’ reductase GrsA reductase reductase ‘GDGT synthase’ GrsB GrsA GrsB
The step in the GDGT biosynthesis which, until now, sparks extensive debate is the macrocylization of archaeal diether lipids to the cyclic GDGTs. This step is thought to occur via head-to-head condensation of two diether precursors. It still needs to be determined what exactly the substrates and the enzyme of this reaction are. There are multiple hypotheses proposed regarding the head-to-head condensation to form GDGTs (Scheme 5). Pulse-chase experiments with [32P]-orthophosphate73 and [14C]-MVA showed first
incorporation of the radiolabel into archaeol (fully saturated form) until saturation of radioactivity and only then incorporation into caldarchaeol. Additionally, when performing pulse-chase protocols with [14C]-MVA in the
presence of terbinafine (a known inhibitor for tetraether formation), an accumulation of saturated diether lipids is observed. Upon removal of the inhibitor, the radioactivity of the saturated diether lipid decreased while radioactivity of the GDGT increased.74, 75 These results prompted the
conclusion that the saturated diether lipid acts as precursor for the head-to-head condensation. In another study, labelling experiments with mevalonolactone-d9 were performed, followed by purification, derivatization,
and NMR analysis of the isolated membrane lipids. It is hypothesized that a reversible double bond isomerization occurs as scrambling of the deuterium labels was observed. This led the authors to speculate that this double bond isomerization acts as trigger for the C–C bond formation in GDGTs.76 These
biosynthetic proposals have been challenged by the results of feeding experiments with [14C]-labelled archaeol and caldarchaeol. No incorporation
of radioactivity into the GDGTs was observed in presence of [14C]-archaeol
or [14C]-archaeol phosphate, suggesting the saturated diether is not a direct
precursor for head-to-head dimerization.77 Yet, another study describes
feeding experiments with deuterium labelled digeranylgeranylglycerol diether lipid derivatives. It was observed that labels were only incorporated into GDGTs when a terminal isopropylidene was present. A derivative lacking the double bond at the hydrophobic end was only incorporated into archaeol, and another derivative with a terminal double bond was incorporated in neither archaeol, nor GDGT. This result indicates that double bond isomerization prior head-to-head condensation does not occur in the biosynthesis of archaeal GDGTs.78 Ultimately, identification of the missing tetraether forming
enzyme and in vitro experiments are crucial in order to prove or disprove the current biosynthetic hypotheses.
A recent study from 2019 shed light on the biosynthesis of the cyclopentane rings found in GDGT-1 to GDGT-8 and crenarchaeol. Two enzymes of the S-adenosylmethionine (SAM) protein family have been identified in the genome of Sulfolobus acidocaldarius, an archaeon known to produce GDGTs with up to eight cyclopentane rings. Members of the SAM protein family are enzymes that are able to catalyze C–C bond formation, mainly via radical-based mechanisms.79 The newly identified enzymes in S. acidocaldarius were
termed GDGT ring synthase (Grs) A and B. Further experiments have shown that GrsA introduces cyclopentane rings exclusively at the C7 position
whereas GrsB produces cyclopentane rings specifically at C3 and has
preference for substrates with existing cyclopentane rings at C7. The results
of this study confirm that cyclopentane ring formation occurs after the generation of the ether bonds. The authors refrain though from speculating whether the cyclopentane rings are formed from saturated or unsaturated precursors. More research is needed to answer this question definitively.19, 80
Crenarchaeol
Occurrence and geochemical implications
In the late 1990s, GDGTs containing cyclopentane rings in their biphytanyl chains were detected and isolated from planktonic archaea in marine water columns and sediments as well as from Crenarchaeum symbiosum.81-83
These findings demonstrated that cyclopentane moieties in archaeal GDGTs are not limited to hyperthermophilic archaea but are also produced by mesophilic and hypothermophilic archaeal species. Among the isolated membrane lipids were also previously unknown GDGTs bearing a biphytanyl chain with two cyclopentane rings and one cyclohexane moiety. Extensive NMR analysis of isolated and purified material of this unknown GDGT – termed crenarchaeol – was conducted to elucidate its structure.84
Additionally, an isomer of crenarchaeol was detected in all samples (Fig. 3). With the exception of Group 1b Thaumarchaeota, the crenarchaeol isomer is only produced as minor component.17, 18 Since the discovery of crenarchaeol,
it has been detected in virtually any habitat on earth, from hot springs, to soils, stalagmites, river beds and lakes.85-93
Crenarchaeol production is strongly associated with marine archaea, in particular Thaumarchaeota, and it has been suggested that these are the predominant source for this particular GDGT and the highest concentrations
of crenarchaeol are found in marine environments. Thus, crenarchaeol was proposed to serve as biomarker for tracing ammonia-oxidizing Thaumarchaeota.16, 91, 94
It is known that certain archaeal GDGTs are very specific to particular archaeal phyla and that for the production of some GDGTs, environmental factors such as temperature are in direct correlation to chemical structure, in particular the number of cyclopentane rings in the biphytanyl chains. As a result of that, multiple proxies have been developed on the basis of the ratio between certain GDGTs. In 2011 the methane index has been proposed to trace the presence of methanotrophic archaea. It utilizes the ratio of total concentration of GDGT1, 2 and 3 over the concentration of GDGT1, 2, -3 and both crenarchaeol isomers.95
Fig. 3 Proposed structures of crenarchaeol and crenarchaeol isomer. The target biphytane
chain is highlighted in blue.84, 96
As it was established that growth temperature is one of the major factors influencing the relative number of cyclopentane rings in archaeal GDGTs, it was proposed to utilize this relationship as basis for a proxy for sea surface temperature (SST). In 2002 this proxy was introduced as the so-called TEX86
Crenarchaeol A12 A11 A19 A10 A9 A8 A7 A18 A6 A3 A1 O A17 A15’ A14’ A13’A12’ A11’ A19’ A20’ A16’ A10’ A18’ A7’ A8’ A9’ H H A17’ O A1’ A3’ A6’ A15A16 A20 C2 C3 O C3’ C2’ O B1’ B3’ B17’ B6’ B7’ B18’ B10’ B9’ B8’ B11’ B12’ B15’ B16’ B16 B19’ B20’ B15 B20 B12 B11 B19 B10 B9 B8 B7 B18 B6 B3 B1 B17 C1 HO H H H H C1’ OH O H H O O O HO H H H H OH Crenarchaeol isomer
The two most abundant GDGTs (crenarchaeol and GDGT-0) were excluded to diminish their overpowering effect on the proxy. This proxy has been applied vastly and has proven to correlate well to the annual mean SST. In addition to determination of annual SST, the TEX86 index is also applied as
paleothermometer by using sediments to reconstruct past lake and sea temperatures.16, 97-100
Chemical structure
The structure of crenarchaeol (Fig. 3) has been proposed after extensive HPLC-MS, and 1D and 2D NMR analysis of isolated and purified material.84
The absolute stereochemistry of the methyl branches (A17, A17’, B17, B17’, A19, B19, B19’, A20, B20, B20’) was assumed to be identical to the known stereochemistry of the methyl substituents in GDGT-0.101, 102 At the time of
the structure proposal, the absolute stereochemistry of the cyclopentane rings was not known but by means of NOESY NMR, the relative stereochemistry could be determined to be 1,3-trans. Later, the absolute stereochemistry of the cyclopentane rings was confirmed to be in agreement with the original proposal by total synthesis of four diastereomers and NMR comparison to natural archaeal biphytanyl chains.103 The analysis of the
cyclohexane moiety was based on the chemical shift differences of axial and equatorial protons and analysis of their coupling patterns. The quaternary stereocenter bearing the methyl group has been pinpointed to the position A15’ and not A11’ as no coupling to the protons of the neighboring cyclopentane ring to the protons of the quaternary methyl substituent was observed by HMBC. The stereochemistry of the quaternary stereocenter was based on two observations: 1) A strong 4J-coupling of the methyl protons to
the axial protons A14’ and A19’ in COSY was observed. 2) The chemical shift of A20’ is relatively downfield at 22.39 ppm, indicating that the methyl is in equatorial position.
In the initial structure proposal, the major isomer of crenarchaeol was assumed to possess an antiparallel glycerol configuration, and the minor to be the parallel isomer. This was later disproven by chemical degradation and HPLC-MS analysis. There, it was found that crenarchaeol exclusively occurs with parallel glycerol arrangement, in contrast to other GDGTs such as caldarchaeol which exists in both parallel and antiparallel glycerol arrangement.104, 105 Eventually, further analysis of the crenarchaeol isomer by
proposal that crenarchaeol isomer is a stereoisomer of crenarchaeol with a
cis-substituted cyclopentane ring adjacent to the cyclohexane ring (Fig. 3).96
Even though extensive analysis has been done to elucidate the structure of crenarchaeol (including relative and absolute stereochemistry), ultimate proof (or disproof) can only be obtained by comparison with synthetic material of defined molecular structure and stereochemistry.
Research aim
From a synthetic chemist’s point of view, crenarchaeol is an intriguing molecule. It is a heavily underfunctionalized molecule, with only six out of 86 carbons bearing heterofunctionalizations (four ether bonds and two hydroxyl groups), yet its carbon skeleton is highly complex. Crenarchaeol possesses no less than 22 (!) stereocenters, with multiple remote methyl groups in 1,5- and 1,4-arrays. The remainder of the stereocenters is embedded in five different carbocycles: four trans-substituted cyclopentane rings, three of which with an adjacent methyl stereocenter, and one cyclohexane ring bearing a quaternary stereocenter. Furthermore, the cyclohexane and the neighboring cyclopentane ring are connected by a single bond resulting in a very unusual skeletal architecture, a motif to this date not accessed by chemical synthesis.
Fig. 4 Chemical structure of the tricyclic biphytanediol unit of crenarchaeol, the target for total
synthesis.
The absolute stereochemistry of crenarchaeol as a whole, and in particular the cyclohexyl containing biphytanyl chain (Fig. 3, in blue), has yet to be confirmed. The ultimate aim of this study is to develop a synthetic route towards crenarchaeol to confirm or disprove its proposed chemical structure. The structure and stereochemistry of the bicyclic biphytane of crenarchaeol was proven by total synthesis already,103 in a study in which four
diastereomeric trans-cyclopentane fragments were synthesized. All of these were accessed from the chiral pool. The optical rotations and NMR spectra
Biphytanediol 1 HO H H OH H H
isolated bicyclic biphytane allowing the confirmation of its structure and absolute stereochemistry.
This thesis describes the synthesis of biphytanediol 1 (Fig. 4) as a single stereoisomer (chapters 2, 3 and 4). This allows comparison with analytical data obtained from natural material in order to unravel the exact chemical connectivity and stereochemistry of the 5-6-ring system of crenarchaeol. Furthermore, biphytanediol 1 itself provides an excellent platform to probe the limits of current synthetic (asymmetric) organic chemistry. An underfunctionalized yet highly complex target as 1 demands a synthesis route in which multiple C–C bond formation reactions need to be applied, some of which in asymmetric fashion. This will require installation of heterofunctionalities to employ their reactivity and removal of these functional handles at a later stage, notably without epimerization of adjacent stereocenters. Besides the installation of the remote methyl branches, constructing the trans-substituted cyclopentane rings stereoselectively will pose a challenge, as this substitution pattern has proved to be difficult to prepare enantio- and/or diastereoselectively. In addition, the complex 5-6-ring system will require development of a new synthetic method to access this motif with high stereocontrol. Due to the inherent flexibility of this lipid chain 1 of crenarchaeol, utilizing substrate control for installation of further stereocenters is not feasible. Instead, catalytic asymmetric reactions in presence of chiral ligands are envisioned for the construction of the stereocenters.
In summary, biphytanediol 1 is an excellent target for a total synthesis as it provides challenges in asymmetric synthesis, functional group installation and removal as well as the prospect to confirm or disprove/revise the proposed structure and stereochemistry.
Retrosynthesis – Initial disconnections
Scheme 6 Retrosynthetic analysis of the target biphytanediol 1 to arrive at two key
intermediates, termed Western and Eastern fragment.
In order to achieve a convergent synthesis route, it was planned to disconnect a central C–C bond in biphytanediol 1 (Scheme 6). Any C–C bond disconnection in the alkyl spacers of 1 always produces two fragments of which one bears an a-stereocenter. Consequently, a synthetic method which minimizes the risk of epimerization and elimination needs to be utilized. Strategies involving transition metal-catalyzed transformations for direct C(sp3)–C(sp3) coupling, such as Pd-catalyzed coupling of primary
organoboranes with primary alkyl halides,106 are not a viable option in this
system as the halides in this synthesis would possess an a-methyl stereocenter. a-Branching was anticipated to lead to significant amounts of side products due to competing b-hydride elimination. To this date, no a-branched halides have shown to undergo successful transition
metal-HO H H OH H H OMOM OTBDPS AcO O S S Biphytanediol 1 2 3 FGI Dithiane alkylation
Cu-cat. Grignard alkylation
Wittig
Cu-cat. Grignard alkylation
FGI BnO I H O (S)-Carvone S S Ring contraction FGI Dithiane alkylation Hydrofunctionalization + +
to that, swapping the halide and the alkene in the substrates would require an asymmetric hydroboration of a 1,1-disubstituted terminal alkene, a transformation which is inherently difficult to achieve in high yields and stereoselectivities (vide infra). As an alternative, dithiane chemistry was envisioned for the initial C–C bond disconnection. By deprotonation of the dithiane with a superbase (i.e. alkyl lithiums), a carbanion which acts as a strong nucleophile is generated and attacks the electrophile in SN2-fashion.
This disconnection provides two coupling partners: the Western Fragment (a primary alkyl iodide as electrophile) and the Eastern Fragment (a dithiane as carbon nucleophile). The Western Fragment is further simplified to commercially available (S)-carvone. The Eastern Fragment can be traced back to two cyclic building blocks, namely cyclopentene acetate 3 and 1,1-disubstituted cyclohexanone 2 carrying the all-carbon quaternary stereocenter. Both 2 and 3 can be accessed enantioselectively in few steps from commercial starting materials. The detailed retrosynthetic analysis of the Western and Eastern Fragment will be discussed in Chapters 2 and 3, respectively.
References
1. C. R. Woese, O. Kandler and M. L. Wheels, Proc. Natl. Acad. Sci. USA, 1990, 87,
4576-4579.
2. R. Cavicchioli, Nat. Rev. Microbiol., 2011, 9, 51-61.
3. Y. Koga and H. Morii, Microbiol. Mol. Biol. Rev., 2007, 71, 97-120.
4. C. R. Woese and G. E. Fox, Proc. Natl. Acad. Sci. USA, 1977, 74, 5088-5090.
5. C. R. Woese, L. J. Magrum and G. E. Fox, J. Mol. Evol., 1978, 11, 245-252.
6. E. F. DeLong, Curr. Op. Genet. Dev., 1998, 8, 649-654.
7. S. Schouten, E. C. Hopmans, R. D. Pancost and J. S. Sinninghe Damsté, Proc. Natl.
Acad. Sci. USA, 2000, 97, 14421-14426.
8. J. S. Sinninghe Damsté, W. I. C. Rijpstra, B. U. Foesel, K. J. Huber, J. Overmann, S.
Nakagawa, J. J. Kim, P. F. Dunfield, S. N. Dedysh and L. Villanueva, Org. Geochem., 2018, 124, 63-76.
9. M. Di Giulio, J. Mol. Evol., 2011, 72, 119-126.
10. Y. Koga, J. Mol. Evol., 2011, 72, 274-282.
11. Y. Koga, J. Mol. Evol., 2014, 78, 234-242.
12. J. Lombard, P. Lopez-Garcia and D. Moreira, Nat. Rev. Microbiol., 2012, 10, 507-515.
13. V. Sojo, BioEssays, 2019, 41, e1800251.
14. J. L. Macalady, M. M. Vestling, D. Baumler, N. Boekelheide, C. W. Kaspar and J. F.
Banfield, Extremophiles, 2004, 8, 411-419.
15. A. Sugai, I. Uda, Y. H. Itoh and T. Itoh, J. Oleo Sci., 2004, 53, 41-44.
16. S. Schouten, E. C. Hopmans and J. S. Sinninghe Damsté, Org. Geochem., 2013, 54,
19-61.
17. A. Pitcher, N. Rychlik, E. C. Hopmans, E. Spieck, W. I. Rijpstra, J. Ossebaar, S. Schouten, M. Wagner and J. S. Sinninghe Damsté, ISME J., 2010, 4, 542-552.
18. J. S. Sinninghe Damsté, W. I. Rijpstra, E. C. Hopmans, M. Y. Jung, J. G. Kim, S. K.
Rhee, M. Stieglmeier and C. Schleper, Appl. Environ. Microbiol., 2012, 78, 6866-6874.
19. Z. Zeng, X. L. Liu, K. R. Farley, J. H. Wei, W. W. Metcalf, R. E. Summons and P. V.
Welander, Proc. Natl. Acad. Sci. USA, 2019, 116, 22505-22511.
20. G. D. Sprott, M. Meloche and J. C. Richards, J. Bacteriol., 1991, 173, 3907-3910.
21. D. Lai, J. R. Springstead and H. G. Monbouquette, Extremophiles, 2008, 12, 271-278.
22. Y. Matsuno, A. Sugai, H. Higashibata, W. Fukuda, K. Ueda, I. Uda, I. Sato, T. Itoh, T.
Imanaka and S. Fujiwara, Biosci. Biotechnol. Biochem., 2009, 73, 104-108.
23. J. Feyhl-Buska, Y. Chen, C. Jia, J. X. Wang, C. L. Zhang and E. S. Boyd, Front.
Microbiol., 2016, 7, 1323.
24. D. S. Nichols, M. R. Miller, N. W. Davies, A. Goodchild, M. Raftery and R. Cavicchioli,
J. Bacteriol., 2004, 186, 8508-8515.
25. J. A. Gibson, M. R. Miller, N. W. Davies, G. P. Neill, D. S. Nichols and J. K. Volkman,
Syst. Appl. Microbiol., 2005, 28, 19-26.
26. O. Dannenmuller, K. Arakawa, T. Eguchi, K. Kakinuma, S. Blanc, A.-M. Albrecht, M.
Schmutz, Y. Nakatani and G. Ourisson, Chem. Eur. J., 2000, 6, 645-654.
27. A. Gliozzi, R. Rolandi, M. De Rosa and A. Gambacorta, J. Membr. Biol., 1983, 75,
45-56.
28. M. Kates, Experimentia, 1993, 49, 1027-1036.
29. H. Komatsu and P. L.-G. Chong, Biochem., 1998, 37, 107-115.
30. D. A. Brown, B. Venegas, P. H. Cooke, V. English and P. L. Chong, Chem. Phys.
Lipids, 2009, 159, 95-103.
31. A. Jacquemet, J. Barbeau, L. Lemiegre and T. Benvegnu, Biochimie, 2009, 91,
711-717.
32. P. L.-G. Chong, Chem. Phys. Lipids, 2010, 163, 253-265.
33. K. Arakawa, T. Eguchi and K. Kakinuma, Chem. Lett., 1998, 27, 901-902.
34. G. D. Sprott, D. L. Tolson and G. B. Patel, FEMS Microbiol. Lett., 1997, 154, 17-22.
37. L. Krishnan and G. D. Sprott, Vaccine, 2008, 26, 2043-2055.
38. S. Jain, A. Caforio and A. J. Driessen, Front. Microbiol., 2014, 5, 641.
39. L. Villanueva, J. S. Damste and S. Schouten, Nat. Rev. Microbiol., 2014, 12, 438-448.
40. M. Nishihara and Y. Koga, J. Biochem., 1995, 117, 933-935.
41. M. Nishihara, T. Yamazaki, T. Oshima and Y. Koga, J. Bacteriol., 1999, 181,
1330-1333.
42. H. Daiyasu, T. Hiroike, Y. Koga and T. Hiroyuki, Protein Eng., 2002, 15, 987-995.
43. J.-S. Han and K. Ishikawa, Archaea, 2005, 1, 311-317.
44. Y. Koga, N. Sone, S. Noguchi and H. Morii, Biosci. Biotechnol. Biochem., 2003, 67,
1605-1608.
45. T. Kuzuyama, Biosci. Biotechnol. Biochem., 2002, 66, 1619-1627.
46. A. Smit and A. Mushegian, Genome Res., 2000, 10, 1468-1484.
47. J. M. Vinokur, T. P. Korman, Z. Cao and J. U. Bowie, Biochem., 2014, 53, 4161-4168.
48. H. Hayakawa, K. Motoyama, F. Sobue, T. Ito, H. Kawaide, T. Yoshimura and H.
Hemmi, Proc. Natl. Acad. Sci. USA, 2018, 115, 10034-10039.
49. Y. Boucher, M. Kamekura and W. F. Doolittle, Mol. Microbiol., 2004, 52, 515-527.
50. H. Nishimura, Y. Azami, M. Miyagawa, C. Hashimoto, T. Yoshimura and H. Hemmi, J.
Biochem., 2013, 153, 415-420.
51. J. Lombard and D. Moreira, Mol. Biol. Evol., 2011, 28, 87-99.
52. N. Dellas, S. T. Thomas, G. Manning and J. P. Noel, eLife, 2013, 2, e00672.
53. J. C. Vannice, D. A. Skaff, A. Keightley, J. K. Addo, G. J. Wyckoff and H. M. Miziorko,
J. Bacteriol., 2014, 196, 1055-1063.
54. L. Rossoni, S. J. Hall, G. Eastham, P. Licence and G. Stephens, Appl. Environ.
Microbiol., 2015, 81, 2625-2634.
55. J. M. Vinokur, M. C. Cummins, T. P. Korman and J. U. Bowie, Sci. Rep., 2016, 6, 39737.
56. K. Wang and S.-I. Ohnuma, Trends Biochem. Sci., 1999, 24, 445-451.
57. S. Vandermoten, E. Haubruge and M. Cusson, Cell. Mol. Life Sci., 2009, 66,
3685-3695.
58. D. Peterhoff, B. Beer, C. Rajendran, E. P. Kumpula, E. Kapetaniou, H. Guldan, R. K.
Wierenga, R. Sterner and P. Babinger, Mol. Microbiol., 2014, 92, 885-899.
59. M. Podar, K. S. Makarova, D. E. Graham, Y. I. Wolf, E. V. Koonin and A.-L. Reysenbach, Biol. Dir., 2013, 8, 9.
60. H. Hemmi, K. Shibuya, Y. Takahashi, T. Nakayama and T. Nishino, J. Biol. Chem.,
2004, 279, 50197-55203.
61. W. Li, Trends Biochem. Sci., 2016, 41, 356-370.
62. H. Zhang, K. Shibuya, H. Hemmi, T. Nishino and G. D. Prestwich, Org. Lett., 2006, 8,
943-946.
63. H. Morii, M. Nishihara and Y. Koga, J. Biol. Chem., 2000, 275, 36568-36574.
64. S. Jain, A. Caforio, P. Fodran, J. S. Lolkema, A. J. Minnaard and A. J. M. Driessen,
Chem. Biol., 2014, 21, 1392-1401.
65. H. Morii, S. Kiyonari, Y. Ishino and Y. Koga, J. Biol. Chem., 2009, 284, 30766-30774.
66. H. Morii and Y. Koga, J. Bacteriol., 2003, 185, 1181-1189.
67. H. Daiyasu, K.-I. Kuma, T. Yokoi, H. Morii, Y. Koga and H. Toh, Archaea, 2005, 1, 399-410.
68. T. Eguchi, M. Morita and K. Kakinuma, J. Am. Chem. Soc., 1998, 120, 5427-5433.
69. Q. Xu, T. Eguchi, Mathews, II, C. L. Rife, H. J. Chiu, C. L. Farr, J. Feuerhelm, L. Jaroszewski, H. E. Klock, M. W. Knuth, M. D. Miller, D. Weekes, M. A. Elsliger, A. M. Deacon, A. Godzik, S. A. Lesley and I. A. Wilson, J. Mol. Biol., 2010, 404, 403-417.
70. D. Sasaki, M. Fujihashi, Y. Iwata, M. Murakami, T. Yoshimura, H. Hemmi and K. Miki,
J. Mol. Biol., 2011, 409, 543-557.
71. Y. Kung, R. P. McAndrew, X. Xie, C. C. Liu, J. H. Pereira, P. D. Adams and J. D. Keasling, Structure, 2014, 22, 1028-1036.
72. M. Murakami, K. Shibuya, T. Nakayama, T. Nishino, T. Yoshimura and H. Hemmi,
FEBS J, 2007, 274, 805-814.
73. M. Nishihara, H. Morii and Y. Koga, Biochem., 1989, 28, 95-102.
75. N. Nemoto, Y. Shida, H. Shimada, T. Oshima and A. Yamagishi, Extremophiles, 2003,
7, 235-243.
76. T. Eguchi, H. Takyo, M. Morita, K. Kakinuma and Y. Koga, Chem. Commun., 2000,
1545-1546.
77. C. D. Poulter, T. Aoki and L. Daniels, J. Am. Chem. Soc., 1988, 110, 2620-2624.
78. T. Eguchi, Y. Nishimura and K. Kakinuma, Tet. Lett., 2003, 44, 3275-3279.
79. K. Yokoyama and E. A. Lilla, Nat. Prod. Rep., 2018, 35, 660-694.
80. A. Pearson, Proc. Natl. Acad. Sci. USA, 2019, 116, 22423-22425.
81. M. J. L. Hoefs, S. Schouten, J. W. De Leeuw, L. L. King, S. G. Wakeham and J. S.
Sinninghe Damsté, Appl. Environ. Microbiol., 1997, 63, 3090-3095.
82. E. F. DeLong, L. L. King, R. Massana, H. Cittone, A. Murray, C. Schleper and S. G.
Wakeham, Appl. Environ. Microbiol., 1998, 64, 1133-1138.
83. S. Schouten, E. C. Hopmans, R. D. Pancost and J. S. Sinninghe Damsté, Proc. Natl.
Acad. Sci. USA, 2000, 97, 14421-14426.
84. J. S. Sinninghe Damsté, S. Schouten, E. C. Hopmans, A. C. T. van Duin and J. A. J.
Geenevasen, J. Lipid. Res., 2002, 43, 1641-1651.
85. A. Pearson, Z. Huang, A. E. Ingalls, C. S. Romanek, J. Wiegel, K. H. Freeman, R. H.
Smittenberg and C. L. Zhang, Appl. Environ. Microbiol., 2004, 70, 5229-5237.
86. J. W. H. Weijers, S. Schouten, M. van der Linden, B. van Geel and J. S. Sinninghe
Damsté, FEMS Microbiol. Lett., 2004, 239, 51-56.
87. S. Schouten, M. T. J. van der Meer, E. C. Hopmans, W. I. C. Rijpstra, A.-L. Reysenbach, D. M. Ward and J. S. Sinninghe Damsté, Appl. Environ. Microbiol., 2007,
73, 6181-6191.
88. A. Pitcher, S. Schouten and J. S. Sinninghe Damsté, Appl. Environ. Microbiol., 2009,
75, 4443-4451.
89. C. I. Blaga, G.-J. Reichart, S. Schouten, A. F. Lotter, J. P. Werne, S. Kosten, N. Mazzeo, G. Lacerot and J. S. Sinninghe Damsté, Org. Geochem., 2010, 41, 1225-1234.
90. X. Liu, J. L. Lipp and K.-U. Hinrichs, Org. Geochem., 2011, 42, 368-375.
91. A. Pitcher, C. Wuchter, K. Siedenberg, S. Schouten and J. S. Sinninghe Damsté,
Limnol. Oceanogr., 2011, 56, 2308-2318.
92. H. Yang, W. Ding, C. L. Zhang, X. Wu, X. Ma, G. He, J. Huang and S. Xie, Org.
Geochem., 2011, 42, 108-115.
93. C. Zell, J.-H. Kim, P. Moreira-Turcq, G. Abril, E. C. Hopmans, M.-P. Bonnet, R. L. Sobrinho and J. S. Sinninghe Damsté, Limnol. Oceanogr., 2013, 58, 343-353.
94. A. Bannert, C. Mueller-Niggemann, K. Kleineidam, L. Wissing, Z.-H. Cao, L. Schwark
and M. Schloter, Biol. Fertil. Soils, 2011, 47, 839-843.
95. Y. G. Zhang, C. L. Zhang, X.-L. Liu, L. Li, K.-U. Hinrichs and J. E. Noakes, Earth
Planet. Sci. Lett., 2011, 307, 525-534.
96. J. S. Sinninghe Damsté, W. I. C. Rijpstra, E. C. Hopmans, M. J. den Uijl, J. W. H. Weijers and S. Schouten, Org. Geochem., 2018, 124, 22-28.
97. S. Schouten, E. C. Hopmans, E. Schefuß and J. S. Sinninghe Damsté, Earth Planet.
Sci. Lett., 2002, 204, 265-274.
98. J.-H. Kim, S. Schouten, E. C. Hopmans, B. Donner and J. S. Sinninghe Damsté,
Geochim. Cosmochim. Acta, 2008, 72, 1154-1173.
99. Z. Liu, M. Pagani, D. Zinniker, R. DeConto, M. Huber, H. Brinkhuis, S. R. Shah, R. M.
Leckie and A. Pearson, Science, 2009, 323, 1187-1190.
100. J.-H. Kim, J. van der Meer, S. Schouten, P. Helmke, V. Willmott, F. Sangiorgi, N. Koç,
E. C. Hopmans and J. S. Sinninghe Damsté, Geochim. Cosmochim. Acta, 2010, 74, 4639-4654.
101. C. H. Heathcock, B. L. Finkelstein, E. T. Jarvi, J. A. Radel and C. R. Hadley, J. Org.
Chem., 1988, 53, 1922-1942.
102. T. Eguchi, K. Ibaragi and K. Kakinuma, J. Org. Chem., 1998, 63, 2689-2698.
103. E. Montenegro, B. Gabler, G. Paradies, M. Seemann and G. Helmchen, Angew.
105. X.-L. Liu, D. A. Russell, C. Bonfio and R. E. Summons, Org. Geochem., 2019, 128, 57-62.
106. S. R. Chemler, D. Trauner and S. J. Danishefsky, Angew. Chem. Int. Ed., 2001, 40,
