Ring-closing metathesis approaches towards the total synthesis of rhizoxins

Article abstract of DOI:10.3390/molecules25194527

Efforts are described towards the total synthesis of the bacterial macrolide rhizoxin F, which is a potent tubulin assembly and cancer cell growth inhibitor. A significant amount of work was expanded on the construction of the rhizoxin core macrocycle by ring-closing olefin metathesis (RCM) between C(9) and C(10), either directly or by using relay substrates, but in no case was ringclosure achieved. Macrocycle formation was possible by ring-closing alkyne metathesis (RCAM) at the C(9)/C(10) site. The requisite diyne was obtained from advanced intermediates that had been prepared as part of the synthesis of the RCM substrates. While the direct conversion of the triple bond formed in the ring-closing step into the C(9)-C(10) E double bond of the rhizoxin macrocycle proved to be elusive, the corresponding Z isomer was accessible with high selectivity by reductive decomplexation of the biscobalt hexacarbonyl complex of the triple bond with ethylpiperidinium hypophosphite. Radical-induced double bond isomerization, full elaboration of the C(15) side chain, and directed epoxidation of the C(11)-C(12) double bond completed the total synthesis of rhizoxin F.

Full text of DOI:10.3390/molecules25194527

molecules
Article
Ring-Closing Metathesis Approaches towards the
Total Synthesis of Rhizoxins
Marc Liniger , Christian M. Neuhaus and Karl-Heinz Altmann *
ETH Zürich, Department of Chemistry and Applied Biosciences, Institute of Pharmaceutical Sciences,
8093 Zürich, Switzerland; marc.liniger@givaudan.com (M.L.); chr.neuhaus1@gmail.com (C.M.N.)
* Correspondence: karl-heinz.altmann@pharma.ethz.ch; Tel.: +41-44-633-73-90
ꢀꢁꢂꢀꢃꢄꢅꢆꢇ
Academic Editor: Ari Koskinen
Received: 16 August 2020; Accepted: 28 September 2020; Published: 2 October 2020
ꢀꢁꢂꢃꢄꢅꢆ
Abstract: Efforts are described towards the total synthesis of the bacterial macrolide rhizoxin F,
which is a potent tubulin assembly and cancer cell growth inhibitor. A significant amount of work
was expanded on the construction of the rhizoxin core macrocycle by ring-closing olefin metathesis
(RCM) between C(9) and C(10), either directly or by using relay substrates, but in no case was
ring-closure achieved. Macrocycle formation was possible by ring-closing alkyne metathesis (RCAM)
at the C(9)/C(10) site. The requisite diyne was obtained from advanced intermediates that had been
prepared as part of the synthesis of the RCM substrates. While the direct conversion of the triple
bond formed in the ring-closing step into the C(9)-C(10) E double bond of the rhizoxin macrocycle
proved to be elusive, the corresponding Z isomer was accessible with high selectivity by reductive
decomplexation of the biscobalt hexacarbonyl complex of the triple bond with ethylpiperidinium
hypophosphite. Radical-induced double bond isomerization, full elaboration of the C(15) side chain,
and directed epoxidation of the C(11)-C(12) double bond completed the total synthesis of rhizoxin F.
Keywords: macrocyclization; macrolactone; natural products; paterson aldol reaction; reductive
decomplexation; rhizoxin; ring-closing alkyne metathesis; radical-induced double bond isomerization;
ring-closing olefin metathesis; total synthesis
1. Introduction
Rhizoxin (
pathogenic fungus Rhizopus chinensis by Iwasaki and co-workers [
configuration of rhizoxin ( ) by means of X-ray crystallography cf. [
1
) (Figure 1) is a 16-membered macrolide that was first isolated in 1984 from the plant
] (for the elucidation of the absolute
]). Fungi of the genus Rhizopus
) is the causative phytotoxin responsible for
1
1
2
are responsible for the rice seedling blight and rhizoxin (
the disease [1].
1
Figure 1. Molecular structures of rhizoxin (1), rhizoxin F (2), and rhizoxin D (3).
Mechanistically, rhizoxin (
1) binds to rice β-tubulin [3,4], which leads to inhibition of mitosis, cell
cycle arrest, and, ultimately, death of the plant. Rhizopus sp. were believed to be the producers of
Molecules 2020, 25, 4527; doi:10.3390/molecules25194527
www.mdpi.com/journal/molecules

Molecules 2020, 25, 4527
2 of 27
rhizoxin (
microsporus is not produced by the fungus itself, but by endosymbiotic bacteria of the genus Burkholderia
that were termed “Burkholderia rhizoxina” [ ]. Partida-Martinez and Hertweck also located, cloned,
and sequenced the entire gene locus that encodes rhizoxin biosynthesis in Burkholderia rhizoxina [ ].
1) until 2005, when Partida-Martinez and Hertweck demonstrated that rhizoxin in Rhizopus
5
6
In addition, the endosymbiont was isolated in pure form and fermented on large scale, resulting in the
recovery of a number of rhizoxin variants with highly potent biological activity, including WF-1360F
(2) [7] (Figure 1), which we have termed rhizoxin F [
8]. The isolation of rhizoxin F (2) had also been
reported previously by Iwasaki et al. [
9
] and Kiyoto et al. [10] from Rhizopus chinensis and Rhizopus
sp., respectively. In addition, rhizoxin variants have been isolated from fermentations of Pseudomonas
fluorescens Pf-5 [11,12] (see also in [13]).
Rhizoxin (
microtubules [14
As a consequence, rhizoxin (
in the single digit nM or even sub-nM range [11
mouse P388 leukemia cell line is 0.16 nM [10].) Rhizoxin (
-subunit of the -tubulin heterodimer at the same site as maytansine, and this binding site was
proposed to be distinct from either the vinblastine or the colchicine site [14 15]. These hypotheses
have recently been confirmed experimentally by means of X-ray crystallography on complexes of
-tubulin with rhizoxin F ( ) and maytansine, respectively [ ]. Importantly, compared to rhizoxin
), monoepoxy-based rhizoxin F ( ) has been reported to exhibit similar [10] or even significantly
higher [8] in vitro antiproliferative activity.
Rhizoxin ( ) has shown significant in vivo antitumor activity in animal models of cancer [11
1
) also binds to eukaryotic tubulin and prevents its assembly (“polymerization”) into
15]; in addition, preformed microtubules are depolymerized by the compound.
) is a potent inhibitor of cancer cell growth in vitro with IC50 values
16]. (For example, the IC50 value of 1 against the
) had been suggested to bind to the
,
1
,
1
β
α/β
,
α
/
β
2
8
(1
2
1
,16,17]
and the compound has been evaluated in humans in a number of Phase I and Phase II clinical studies.
However, the outcome of these trials was generally disappointing [18], although the reasons for the lack
of clinical efficacy of rhizoxin (1) have not been elucidated in detail (based on public literature). It has
been suggested that “further studies of this class of agents [rhizoxins] will depend on the identification
of more efficacious analogs in preclinical models” [18].
The chemistry of rhizoxins has been explored extensively and as of today, 10 total syntheses of a
naturally occurring rhizoxin have been reported in the literature. Out of these 10 syntheses, eight have
targeted the epoxide-free rhizoxin variant rhizoxin D (
described the final conversion of 3 into rhizoxin [19]. In addition, we have reported the total synthesis
of the monoepoxy variant rhizoxin F ( ) [20]. Early work in the field (up to 2004) has been reviewed in
3) (Figure 1) and only Ohno and co-workers have
2
detail in an excellent account by Hong and White [21] and shall not be discussed here, except to note
that a recurring theme in the majority of these syntheses is the closure of the macrolactone ring through
intramolecular Horner-Wittig-Emmons (HWE) reaction. Alternative approaches to ring-closure have
been reported by Pattenden and co-workers (Stille coupling between C(10) and C(11)) [22] and by
our own group (ring-closing alkyne metathesis (RCAM) between C(9) and C(10)) [20]. Most recently,
Fürstner and co-workers have described an intriguing formal total synthesis of rhizoxin D (3) that
included ring-closure by RCAM between C(11) and C(12) (!) [23]. The resulting triple bond was then
efficiently transformed into the required trisubstituted trans double bond by trans-stannylation and
subsequent C-methylation with excellent selectivity.
Our group has a long-standing interest in the chemistry, medicinal chemistry, and pharmacology
of bioactive macrocyclic natural products. In this context, we were interested to develop a new synthetic
approach towards rhizoxins that would be based on ring-closing olefin metathesis (RCM), which,
somewhat surprisingly, had not been investigated as part of any of the previous total syntheses of a
rhizoxin variant. The specific target to be pursued in our work was rhizoxin F (
this monoepoxy rhizoxin variant has been reported to be at least equipotent with rhizoxin (
regard to cancer cell proliferation in vitro. In a previous communication, we have reported the total
synthesis of rhizoxin F ( ) via RCAM [20]; in this communication, we also briefly touched upon our
2); as alluded to above,
1) with
2

Molecules 2020, 25, 4527
3 of 27
numerous futile attempts at RCM-based ring closure. In the current paper, we now disclose full details
on these attempts and on the successful total synthesis of 2 based on macrocyclization by RCAM.
2. Results and Discussion
2.1. Synthetic Planning
Apart from the exploration of the feasibility of an RCM-based macrocyclization strategy, our total
synthesis of
2 was also to provide a platform for the subsequent synthesis of side chain-modified
rhizoxin F analogs for SAR studies. In this context, the intermediate formed in the ring-closure reaction
was also meant to serve as a central node and point of diversification for the elaboration of differently
modified side chains in a single step. One possible strategy for the synthesis of rhizoxin F (
would meet our predefined boundary conditions is outlined in Scheme 1. This strategy envisioned
the late stage elaboration of the C15-side chain from macrocyclic vinyl iodide * by means of Stille
coupling. Vinyl iodide * could also serve as a versatile common intermediate for the synthesis of
2) that
4
4
differently modified side chain analogs by employing alternative vinyl stannanes as building blocks.
Stille coupling was to be followed by deprotection of the C13 hydroxy group and directed epoxidation
of the C(11)-C(12) double bond to deliver 2; the feasibility of this latter transformation had already
been demonstrated by Ohno and co-workers as part of their synthesis of rhizoxin (1), although for a
differently decorated macrocycle [19].
Scheme 1. Retrosynthesis of rhizoxin F (2) based on ring-closing olefin metathesis (RCM) of diene 5*.
PG, protecting group. The suffix “*” indicates that differently protected versions of the corresponding
intermediates were contemplated at the planning stage.
Out the three possible diene precursors, our RCM-based disconnection of macrocycle
4* led to
diene * as the most promising substrate, as it presents two unencumbered monosubstituted terminal
5

Molecules 2020, 25, 4527
4 of 27
double bonds as the sites of reaction. At the same time,
5
* was envisaged to be accessible by the
esterification of two fragments of similar size and structural complexity, i.e., acid
*; therefore, this approach also offered the highest level of convergency for the total synthesis.
The -lactone-containing carboxylic acid was envisioned to be accessed by a diastereoselective
allylation of aldehyde 11*, followed by RCM to close the -lactone ring. Alcohol * was planned to
be obtained by the substrate-controlled 1,3-anti reduction of a -hydroxy ketone precursor, which in
turn was to be furnished by stereoselective aldol reaction between known aldehyde 10 and methyl
ketone 11, either under Mukaiyama [24 25] or Paterson [26] conditions. Acid was envisioned to be
accessed by two-carbon extension of aldehyde * via HWE olefination with an alkyl dialkylphosphono
6 and alcohol
7
δ
6
δ
7
β
,
6
8
acetate; the former was envisaged to result from the stereoselective allylation of an aldehyde 11* with
an allylsilane/allyl stannane 12* in a Hosomi-Sakurai- [27] or Keck-type [28] allylation reaction and
subsequent functional group adjustment.
2.2. Synthesis of Alcohol 7*
As illustrated in Scheme 2, the synthesis of aldehyde 10 as one of the building blocks for the
construction of 7* was based on a high-yielding and highly stereoselective Evans aldol reaction between
aldehyde 13 and N-propionyl-oxazolidinone 14 (92% yield of a 15 as a single isomer), as has been
described previously by White and co-workers [29]; aldehyde 13 was obtained from diethyl methyl
malonate according to Menche et al. [30]. Imide 15 was then cleanly converted into methyl ester 16 by
treatment with sodium methoxide. The latter was elaborated into aldehyde 10 by methylation with
NaH and MeI, followed by DIBAL reduction of the ester group (91% overall yield) and Dess-Martin
periodinane (DMP) [31] oxidation of the ensuing primary alcohol. In order to prevent epimerization of
10 during chromatography, the crude aldehyde was directly used in the aldol step.
Scheme 2. Reagents and conditions: (
single isomer; (
) NaOMe, CH2Cl2/MeOH (3:1), 0 ◦C, 95%; (
) DIBAL, CH2Cl2,
78 ◦C to 0 ◦C, 97%; ( ) DMP, NaHCO3, 0 ◦C to rt. DIBAL, diisobutylaluminum
a)
14, nBu2BOTf, NEt3, CH2Cl2, 0 ◦C, then 13,
−
78 ◦C to 0 ◦C, 92%,
b
c
) NaH, MeI, THF/DMF (3:1), 0 ◦C, 94%;
(
d
−
e
hydride; DMP, Dess–Martin periodinane (1,1,1-tris(acetyloxy)-1,1-dihydro-1,2-benziodoxol-3-(1H)-one);
Tf, trifluormethansulfonyl.
In initial experiments, methyl ketone 11 was prepared via the known ester 19 [32] (Scheme 3),
which was obtained by Wittig reaction of phosphorane 17 with acrolein (18) in 70% yield. Ester 19
could be converted into 11 either by enolate trap Grignard addition or by one-pot Weinreb ketone
synthesis [33]. However, for both approaches the yield of 11 was moderate at best and the reactions
suffered from poor reproducibility. Alternatively, methyl ketone 11 was accessed more reliably from
2-butanone (20), which was
(Scheme 3). Reaction of the mixture of 21a
α
-brominated to yield an 11:1 mixture of bromides 21a and 21b in 47% yield
21b with PPh3 in acetonitrile gave a crude phosphorane 22
/
that reacted with acrolein (18) to furnish 11 as a single isomer in 57% yield. Notably, the treatment
of 21a/b with aqueous NaOH (which were the conditions used for the preparation of 17 from its
phosphonium salt precursor in high yield) did not give any of the desired phosphorane 22.

Molecules 2020, 25, 4527
5 of 27
Scheme 3. Reagents and conditions: (
15 ◦C to rt, 51%; (
) LiHMDS, MeMgCl, THF,
21a:21b = 11:1; ( ) (
) PPh3, MeCN, 70 ◦C; (ii) NaOH, MeCN, rt, 79%; (
TMSCl, NaI, CH3CN, 0 ◦C to 90 ◦C, 86%; ( 78 ◦C, 12%, dr 4:1. LiHMDS, lithium
) TiCl4, CH2Cl2,
bis(trimethylsilyl)amide; TMS, trimethylsilyl.
a
) CH2Cl2, rt to rf, 70%; (
78 ◦C to +10 ◦C, 35%; (
) CH2Cl2, rt to rf, 57%; (
b
) Me(OMe)NH
) Br2, AcOH, H2O, 65 ◦C, 47%,
) NEt3,
·
HCl, MeMgCl, THF,
−
c
i
−
d
e
f
g
h
−
With aldehyde 10 and methyl ketone 11 in hand, the stage was set for the projected aldol reaction.
To this end, ketone 11 was first transformed into its TMS-enol ether 23 as the necessary precursor
for a Mukaiyama aldol reaction [24
involved treatment of 11 with triethylamine, TMS-chloride, and sodium iodide in acetonitrile, a method
developed by Cazeau et al. for the special purpose of forming TMS-enol ethers from -unsaturated
,25] (Scheme 3). The best conditions for the formation of 23
α,β
ketones and aldehydes [34]. In analogy to the work of Pattenden et al. [22], enol ether 23 was then
submitted to reaction with aldehyde 10 in the presence of Me2AlCl; however, these conditions led to a
mixture of at least three compounds, whose spectroscopic data could not be interpreted. If TiCl4 was
used as a Lewis acid, the aldol product was indeed obtained, but only in 12% yield and with moderate
diastereoselectivity (4:1).
In light of the disappointing findings on the Mukaiyama aldol reaction, we turned our attention
to the possible construction of the C15 stereocenter (rhizoxin numbering, see Scheme 1) by a Paterson
aldol reaction [26] between aldehyde 26 and methyl ketone 27 (Scheme 4). The successful Paterson
aldol reaction between 27 and a more complex aldehyde comprising the C(3)–C(13) segment of rhizoxin
had been reported by White and co-workers as part of their synthesis of rhizoxin D [29].
Aldehyde 26 was readily accessible from ester 19 by reduction to the alcohol and subsequent
oxidation of the latter with activated MnO2 (Scheme 4). Methyl ketone 27 could be prepared in 3 steps
from aldol product 15 (cf. Scheme 2) as described by White [29] (with improved conditions for the
methylation step) in 85% overall yield. Alternatively, ester 16 (cf. Scheme 2) could be converted into 27
by one-pot Weinreb amide ketone synthesis in 72% yield.
Satisfyingly, the reaction of the boron enolate of ketone 27 (formed with NEt3 and (+)-DIPCl in
dichloromethane) with aldehyde 26 afforded β-hydroxy ketone 28 in 58% yield with a dr of 14.8:1.
While the aldol product was inseparable from the isopinocampheol formed upon oxidative workup,
the latter did not interfere with the subsequent reaction step and could then be easily removed
by chromatography. Importantly, a control reaction with (
stereoselectivity (dr 1.3:1), thus indicating that the combination of
represents a mismatched case for the Paterson aldol reaction. Reduction of
−
)-DIPCl proceeded without significant
-chiral ketone 27 and ( )-DIPCl
-hydroxy ketone 28 with
α
,β
−
β
NMe4BH(OAc)3 [35] gave diol 29 in 92% yield as a single isomer (Scheme 4). While attempts at the
regioselective TBS protection of the C(13)-OH group (rhizoxin numbering) returned only mixtures of
regioisomers, treatment of 29 with TIPS-OTf and 2,6-lutidine at
78 ◦C gave silyl ether 30 in 91% yield
−
as a single regioisomer.

Molecules 2020, 25, 4527
6 of 27
◦
Scheme 4. Reagents and conditions: (
(2 steps); ( ) Me(OMe)NH
THF/DMF (3:1), 0 ◦C, 95%; (
THF, ) (
10 ◦C to rt, 72%; (
phosphate buffer (pH 7), MeOH, 0 ◦C to rt, 58%, dr 14.8:1; (
a
) LiAlH4, Et2O, 0 C; (
b
) activ◦ated MnO2, CH2Cl2, rt, 78%
, 40 C to rt, 96%; (d) NaH, MeI,
◦
c
·
HCl, AlMe3, THF, 0 C to rt, then 15
−
e
) MeMgBr, THF,
27, NEt3, (+)-DIPCl, CH2Cl2,
−
20 ◦C to
−
5 ◦C, 93%; (
78 ◦C, then 26
) NMe4BH(OAc)3, CH3CN/AcOH (2:1),
f
) Me(OMe)NH
·
HCl, MeMgCl,
−
g
i
)
−
,
−
78 ◦C to 18 ◦C; (ii) H2O2,
−
h
−40 ◦C to rt, 92%, single isomer; (i) TIPSOTf, 2,6-lutidine, 3 Å molecular sieves, CH2Cl2, −78 ◦C, 91%.
DIPCl, B-chlorodiisopinocampheylborane; Tf, trifluormethansulfonyl; TIPS, triisopropylsilyl.
2.3. Synthesis of Acid 6
The synthesis of acid building block 6 at an early stage of our work involved exploration of the
stereoselective allylation of an aldehyde 11 with allylsilane 12a in a Hosomi-Sakurai reaction [27].
As depicted in Scheme 5, the preparation of allylsilane 12a commenced with the benzylation of
commercially available 3-butynol with BnBr/NaH, which delivered benzyl ether 31 in quantitative
yield. Hydroiodination of 31 with in situ formed HI (from TMSCl, NaI and water) [36] in the presence
◦
of 0.6 equivalents of water at 0 C led to concomitant loss of the benzyl group and gave the free
alcohol 32 in 40% yield. These optimized hydroiodination conditions do not generate more than 0.6
equivalents of HI, thereby effectively preventing diiodination.
Scheme 5. Reagents and conditions: (
(0.6 equiv), MeCN, 0 ◦C to rt, 40%, 96% pure; (
Pd(PPh3)4 (0.5 mol%), LiCl, Et2O, rt, 93%; ( 34/35a (R = TBS): (
SnCl4 (2.0 equiv), CH2Cl2,
78 ◦C; (ii 11a (1.0 equiv), CH2Cl2, –78 ◦C, 71% (2 steps, including step g)),
34a/35a = 1.7:1; 34/35b (R = TBDPS): 11b (1.0 equiv), 12a (2.2 equiv), TiCl4 (1.1 equiv), CH2Cl2,
78 ◦C;
37% (2 steps, including step g)), 34b 35b = 1.74:1; 34 35c (R = Bn): 11c (1.0 equiv), 12a (2.6 equiv),
TiCl4 (1.4 equiv), CH2Cl2, 35c = 1:3.3; ( ) a: TBSCl,
78 ◦C, 50% (2 steps, including step g)), 34c
imidazole, CH2Cl2, rt, 99%; b: TBDPSCl, imidazole, DMF, rt, 92%; c: benzyl 2,2,2-trichloroacetimidate,
TfOH (cat.), CH2Cl2/cyclohexane (1:1), rt, 77%; ( ) DIBAL (1.05 equiv), CH2Cl2,
78 ◦C, crude.
a
) NaH, BnBr, TBAI, THF, 0 ◦C to rt, quant.; (
) TBSCl, imidazole, DMF, rt, 98%; (
12a (2.1 equiv), 4 Å molecular sieves,
b
) TMSCl, NaI, H2O
c
d
) TMSCH2MgCl,
e)
i)
−
)
−
/
/
−
/
f
g
−
TBAI, tetra-n-butylammonium iodide; TBS, tert-butyldimethylsilyl; TBDPS, tert-butyldiphenylsilyl;
TMS, trimethylsilyl.

Molecules 2020, 25, 4527
7 of 27
Hydroiodination was also possible with free butyn-3-ol, but the desired iodo alcohol 32 was only
obtained in 22% isolated yield. Attempts to (re)protect 32 as a benzyl ether under basic conditions
(BnBr, NaH) predominantly led to elimination and gave an inseparable mixture of the desired product
and benzyl ether 31 in a combined yield of 41%. Attempts to conduct the benzylation under neutral
(Dudley’s reagent [37]) or acidic conditions (benzyl trichloroacetimidate/catalytic acid [38]) resulted
in complex reaction mixtures, from which the desired product could not be purified to homogeneity.
In contrast, treatment of 32 with TBSCl cleanly furnished TBS-ether 33 in 98% isolated yield. Subsequent
cross-coupling with TMS-methylmagnesium chloride either under Negishi [39] or Kumada [40,41]
(Ni and Pd) conditions generally worked well. In particular, the use of Pd(PPh3)4 and LiCl gave
excellent yields of allyl silane 12a at very low catalyst loadings (down to 0.5 mol%) even on large scale.
Allylsilane 12a was found to be highly acid sensitive and prone to protodesilylation. Thus, the eluent
for silica gel chromatography was buffered with 0.5% of NEt3 and deuterated benzene was employed
as the NMR solvent instead of chloroform. These conditions effectively prevented degradation of 12a
during purification and analysis.
The Hosomi-Sakurai reaction between 12a and aldehydes 11 was then investigated for a range of
Lewis acids. As expected, the non-coordinating aldehydes 11a and 11b gave the desired Felkin-Anh
(syn) products 34a or 34b, respectively, preferentially, but yields were generally low to moderate and
even in the most favorable case (aldehyde 11a in combination with BF3 Et2O) the dr did not exceed 4:1.
•
As expected, benzyl-protected aldehyde 11c reacted with chelation control, thus delivering the anti
product 35c in excess.
Rather than trying to improve the selectivity of the Sakurai reaction by using chiral Lewis
acid as, for example, developed by Yamamoto [42], we considered it more promising to assess the
stereoselective Keck-type allylation [28] of aldehydes 11 with allyl stannane 12b. Different precursors
were envisioned for the preparation of 12b by reaction with tributyltin chloride, including benzyl ether
38 (via the derived anion), allylic alcohol 39 (e.g., via the corresponding mesylate) or allylic chloride
41; in principle, 38 should also serve as a precursor for 39 and 41 (Scheme 6). Benzyl ether 38 was
readily accessible from 2-methylbut-1-en-4-ol (37) in excellent yield (97%) by benzylation with benzyl
bromide under basic conditions. Unfortunately, however, deprotonation of 38 with nBuLi/TMEDA [43
]
or Schlosser’s base [44 45] followed by quenching of the resulting anion with tributyltin chloride failed
,
to yield allyl stannane 12b. Allylic chlorination [46] of 38 was non-selective and the desired isomer 41
could not be separated from isomeric impurities. In contrast, Riley oxidation [47] furnished a separable
mixture of allylic alcohols 39 and 40, from which the desired isomer 39 could isolated in 23% yield;
40 was obtained in 28% yield.
Scheme 6. Reagents and conditions: (
a
) NaH, BnBr, TBAI, THF, 0 ◦C to rt, 97%; (
) CeCl3
) mCPBA, CHCl3, rt, 95%; ( ) TMPAlEt2, toluene,
b
) (
i
) SeO2 (10 mol%),
tBuOOH, CH2Cl2, rt; (ii) NaBH4, MeOH, 0 ◦C, 39: 23%, 38% brsm, 40: 28%; (
c
·7H2O, NaClO,
CH2Cl2/H2O (1:1), rt, 58% + inseparable regioisomers; (
0 ◦C, 39: 48%, 43: 16%. TBAI, tetra-n-butylammonium iodide; TMP = 2,2,6,6-tetramethylpiperidine.
d
e
Brsm, based on recovered starting material.

Molecules 2020, 25, 4527
8 of 27
As an alternative to direct allylic hydroxylation, we also explored the regioselective
β-elimination
from gem-disubstituted epoxide 42 by means of a sterically hindered aluminium amide base, following
an approach developed by Brückner [48]. However, while the treatment of the TBS-protected variant
of 42 with TMPAlEt2 has been reported to yield the allylic alcohol corresponding to 39 in high yield
as a single isomer [48], epoxide 42 furnished a ~3:1 mixture of allylic alcohols 39 and 43, which were
isolated in 48% and 16% yield, respectively. Thus, while allylic alcohol 39 could be accessed from 38 as
a single isomer, the yield was far from satisfactory.
In light of the difficulties to access 39 directly from 38, an alternative route was developed from
1,4-butandiol (44) which delivered 39 with higher overall efficiency and without the need for isomer
separation. As depicted in Scheme 7, monobenzylation of 44 followed by Swern oxidation furnished
an aldehyde that was
a procedure originally developed by Ogasawara and co-workers [49]. The resulting
α
-methylenated in situ (CH2NMe2Cl, DBU), employing a modification of
-methylene
α
aldehyde 46 was obtained in 86% yield. Reduction of 46 with LiAlH4 gave alcohol 39, which was
submitted to Appel reaction to furnish allylic chloride 41. In initial experiments, 46 was converted into
41 by NaBH4 reduction followed by Finkelstein reaction, but this approach proved to be inconvenient
on larger scale, due to the use of DMF and EtOH in large quantities.
Scheme 7. Reagents and conditions: (
NEt3, CH2Cl2,
78 ◦C to 0 ◦C, 2 h, then CH2NMe2Cl, DBU, rt, 22 h, 86%; (
94%; (
) LiAlH4, THF, 0 ◦C, 96%; ( ) MsCl, 2,6-lutidine, LiCl, DMF, 0 ◦C to rt, 5 h, 88%; (
MeCN, rt, 2 h, 95%; (
) Mg, Bu3SnCl, THF, ultrasound, 0 ◦C, 2h, then dibromoethane (cat.), refl., then rt,
overnight, quant., 80% purity; ( 11a, BF3 OEt2, CH2Cl2, ) ( ) BBr3, ligand
78 ◦C, 1 h, 65% (dr 1.9:1); (
48, CH2Cl2, 0 ◦C to rt, 1 h, (ii 12b, 0 ◦C to rt, 16 h, (iii 78 ◦C, 2 h, 74% (2 steps) (dr 10:1). DBU,
11a
1,8-diazabicyclo [5.4.0]undec-7-ene; Ms, mesyl.
a
) NaH, BnBr, THF, 0 ◦C to rt, 15 h, 97%; (
) NaBH4, EtOH, 0 ◦C, 1 h,
) CCl4, PPh3,
b) (COCl)2, DMSO,
−
c
d
e
f
g
h
)
·
−
i i
)
)
,
−
Conversion of chloride 41 into stannane 12b was then achieved by Barbier-type reaction of 41
,
Mg turnings, and tributyltin chloride with ultrasound irradiation (quantitative yield, 80% purity by
1H-NMR analysis) [50]. The purity of the material obtained after simple extractive work-up was
sufficient for the next step.
The BF3 OEt2-mediated allylation of aldehyde 11a with stannane 12b gave alcohol 47 only with
•
poor diastereoselectivity (dr 1.9:1). However, transmetalation of 12b with bromoborane complex
(prepared from BBr3 and bis-tosylated (R,R)-1,2-diphenylethylenediamine ((R,R)-DPEN) (48) [51] under
conditions elaborated by Williams et al. [52] and subsequent reaction with 11a at
78 ◦C furnished
−
secondary alcohol 47 in very good yield and with high diastereoselectivity (74% over 2 steps, dr 10:1,
94% recovery of 48). The configuration of the newly established stereocenter in 47 was confirmed by
Mosher ester analysis [53].
Esterification of 47 with acryloyl chloride afforded diene 52 in high yield (85%) as a single
isomer (Scheme 8). RCM of 52 with 10 mol% of Hoveyda-Grubbs 2nd generation catalyst in refluxing
dichloroethane delivered dihydropyrone 53 in an excellent 89% yield on small scale (44 mg of 52).
However, the reaction was found to be rather unreliable on multigram scale, where it frequently did

Molecules 2020, 25, 4527
9 of 27
not proceed beyond 50% conversion, thus requiring multiple recyclings of the starting diene to achieve
acceptable yields. Moreover, dimerization via the more reactive monosubstituted double bond started
to compete with ring formation over time. Although this pathway could be partially suppressed by
using higher dilution, this requirement imposed restrictions on scale-up. Therefore, an alternative
three-step synthesis of 53 from 47 was elaborated that was based on an intramolecular HWE reaction
for 6-membered ring formation, similar to what had been reported by Regan and co-workers as part of
their synthesis of the C(1)–C(9) fragment of rhizoxin [54].
Scheme 8. Reagents and conditions: (
Grubbs 2nd generation catalyst (14 mol%), DCE, reflux, 89%; (
dioxane/water, rt, 94%; (
55, CME-CDI, DMAP, CH2Cl2, 0 ◦C to rt, 93%; (
) H2 (9 bar), Pd(OH)2/C (5 mol%), EtOAc, rt, 98%. CDE-CDI = N-cyclohexyl-N0-(
morpholino]ethyl) carbodiimide p-toluenesulfonate; DMAP, 4-(dimethylamino)pyridine.
a
) Acryloyl chloride, DIPEA, CH2Cl2,
) OsO4 (cat.), NaIO4, 2,6-lutidine,
) NaH, THF, 0 ◦C, 83%;
-[N-methyl-
−
50 ◦C, 85%; (
b) Hoveyda-
c
d)
e
(
f
β
Thus, alkene 47 was subjected to Lemieux-Johnson oxidation, to furnish β-hydroxy ketone 54
in excellent 94% yield. In comparison, ozonolysis led to lower yields (76–84%) and the formation of
inseparable side products (~5–10%). Steglich esterification of 54 with diethylphosphonoacetic acid
(
55) yielded phosphono ester 56, which underwent intramolecular HWE reaction upon treatment
with NaH, to yield the desired dihydropyrone 53 in 83% yield. The stereoselective hydrogenation of
the double bond in 53 with Pearlman’s catalyst, according to the procedure reported by Regan [54
]
proceeded with concomitant cleavage of the benzyl ether moiety, thus furnishing lactone 57 as a single
isomer in 98% yield. The large trans-diaxial coupling constants of 11.9 Hz and 10.5 Hz in the 1H-NMR
spectrum of 57 (Scheme 8) confirmed the cis-substitution of the δ-lactone ring.
The further elaboration of lactone 57 into acid building block
extension at the free hydroxy end that was achieved by Swern oxidation, to yield aldehyde
6
in a first stage required a two-carbon
, followed
8
by Wittig olefination with stabilized ylide 58 (Scheme 9). Ester 59 was obtained as a single isomer
in 79% overall yield from 57. In contrast to the Wittig reaction with ylide 58, the HWE reaction of
aldehyde
yields. After installation of the carboxy functionality, completion of the carbon framework of acid
building block required the elaboration of the terminal double bond at the non-carboxy end of
8 with the tbutyl diethylphosphonoacetate gave ester 59 in much lower and highly variable
6
the structure. To this end, the TBS-ether moiety in 59 was cleaved with buffered (AcOH) TBAF,
to afford alcohol 60 in excellent yield (provided that the acid present was not neutralized with NaHCO3
during work-up). After Dess-Martin oxidation [31], aldehyde 61 was isolated as a crude product
after simple extractive work-up, as significant racemization on silica gel precluded chromatographic
purification. Surprisingly, the conversion of the carbonyl group into the terminal double bond proved

Molecules 2020, 25, 4527
10 of 27
to be highly challenging. Thus, Wittig methylenation with MePPh3Br/KHMDS was accompanied by
epimerization of the -stereocenter and the desired olefin was only obtained in 8% yield. A whole
α
series of alternative methods for double bond formation were then investigated, including Tebbe [55],
Takai [56], Oshima-Lombardo [57], Julia [58], as well as Rh- [59] and Cu-catalyzed [60] methylenations;
of these, only the Julia olefination of 61 with the phenyltetrazol sulfone 62 gave the desired olefin 63 in
a preparatively useful yield of 38% (based on alcohol 60); but the compound could only be isolated as
a mixture with residual sulfone 62. The latter was readily removed after tbutyl ester cleavage with
TFA, which gave the desired acid 6 in quantitative yield.
Scheme 9. Reagents and conditions: (a) (COCl)2, DMSO, NEt3, CH2Cl2, −78 ◦C, 94%; (b) 58, CH2Cl2,
0 ◦C to rt, 84%; ( ) TBAF, AcOH, CH2Cl2, 0 ◦C to rt, 96%; ( ) DMP, CH2Cl2, 0 ◦C to rt; (
c d e) NaHMDS,
PT-sulfone 62, THF, f) TFA/CH2Cl2 (ca. 14 equiv TFA), rt, quant. DMP,
78 ◦C to rt, 38% (2 steps); (
−
Dess-Martin periodinane (1,1,1-tris(acetyloxy)-1,1-dihydro-1,2-benziodoxol-3-(1H)-one); NaHMDS,
sodium bis(trimethylsilyl)amide; PT, phenyltetrazole; TFA, trifluoroacetic acid.
2.4. Building Block Assembly and Attempted Ring-Closing Olefin Metathesis
The esterification of acid
coupling agents, including EDCI/DMAP; 2,4,6-trichlorobenzoyl chloride (TCBC)/NEt3/DMAP,
2-methyl-6-nitrobenzoic anhydride (MNBA)/NEt3/DMAP, and N-cyclohexyl-N0-(
-[N-methyl-
morpholino]ethyl) carbodiimide p-toluenesulfonate (CME-CDI); however, except for the use of
CDI-DME, none of the desired ester could be isolated (Scheme 10). In general, alcohol 30 could
be recovered from the reaction mixtures in high yield (>80%), while acid completely decomposed.
6 with alcohol 30 was investigated with a number of different
β
5
6
CME-CDI gave the desired ester in a yield of 14%, which, obviously, was unacceptably low at this
stage of the synthesis (Scheme 10).
At the same time, there was ample evidence from previous total syntheses of rhizoxins that the
esterification of the C(15) hydroxy group in different rhizoxin precursors with dialkylphosphonoacetic
acids was unproblematic. This observation immediately suggested that the projected RCM precursor
5
should be accessible from existing advanced intermediates that we had prepared at this point with very
limited additional effort; more specifically, alcohol 30 was to be esterified with diethylphosphonoacetic
acid (55) and the phosphono ester would then be submitted to HWE reaction with aldehyde 67
(Scheme 11). The latter would be obtained from alcohol intermediate 57 (cf. Scheme 9) by orthogonal
protection of the primary hydroxy group, selective cleavage of the TBS-ether, and elaboration of the
free hydroxy group into the desired double bond.

Molecules 2020, 25, 4527
11 of 27
Scheme 10. Reagents and conditions: (
then DMAP, rt; (iii) 0 ◦C, 60 h, 14% (72% 30 recovered). CME-CDI, N-cyclohexyl-N0-(
morpholino]ethyl) carbodiimide p-toluenesulfonate; DMAP, 4-(dimethylamino)pyridine.
i
)
6
, CME-CDI, 0 ◦C, CH2Cl2, molecular sieves 4 Å; (ii
)
30, 0 ◦C,
β-[N-methyl-
Scheme 11. Reagents and conditions: (
(4:1), rt, 82%; (
) DMP, CH2Cl2, 0 ◦C to rt; (
) TBAF, AcOH, THF, 0 ◦C to rt, 87%, (
) (COCl)2, DMSO, NEt3, CH2Cl2,
CH3CN, 0 ◦C, 85%; ( py, py, THF, 0 ◦C to rt, 86%. DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene; DMP,
) HF
a
) TBDPSCl, imidazole, DMF, rt, 95%; (
) NaHMDS, PT-sulfone 62, THF,
78 ◦C to rt, 65% (2 steps);
78 ◦C, 95%; (
) DBU, LiCl,
b) NaIO4, THF/water
c
d
−
(
e
f
−
g
h
•
Dess–Martin periodinane (1,1,1-tris(acetyloxy)-1,1-dihydro-1,2-benziodoxol-3-(1H)-one); NaHMDS.
sodium bis(trimethylsilyl)amide; PT, phenyltetrazole; TBAF, tetra-n-butylammonium fluoride; TBDPS,
tert-butyldiphenylsilyl.
To implement this revised strategy, alcohol 57 was converted into TBDPS-ether 64 followed by
selective cleavage of the TBS-ether moiety with an excess of NaIO4 in aqueous THF [61] to afford
alcohol 65 in 78% overall yield (based on 57) (Scheme 11). Subsequent Dess-Martin oxidation gave
aldehyde 66 in low yields if purified by column chromatography, in contrast to what had been reported
for a closely related aldehyde (bearing a TIPS- in place of a TBDPS-ether group) by Pattenden and
co-workers [22]. The aldehyde was thus used crude in the subsequent Julia olefination with PT-sulfone
62, which delivered the desired olefin 67 in 65% overall yield from 65. TBAF-mediated TBDPS
deprotection followed by Swern oxidation then furnished aldehyde 68; the latter underwent smooth
HWE olefination with phosphono ester 69 (obtained by CEM-CDI-mediated esterification of 30 with
diethylphosphonoacetic acid (55)) under Masamune-Roush conditions [62] to provide the desired
RCM substrate 5 in 79% overall yield from 67.
With pentaene
5 in hand, the stage was set for the assessment of the key ring-closing olefin
metathesis reaction. To this end, a number of ruthenium catalysts was assessed, including the commonly
used 1st and 2nd generation Grubbs and Hoveyda-Grubbs catalysts, but also the less commonly

Molecules 2020, 25, 4527
12 of 27
employed Grubbs
3
and Piers-Grubbs
2
catalysts [63], as well as the slowly initiating pyridine-containing
complex (sIMes)(Cl)2Ru(CH(CH2)2-C,N-2-C5H4N) [64] (see Table S1). Unfortunately, whenever the
starting material was consumed, it was converted into an array of products. While the desired
macrocycle 4 was sometimes detectable by ESI-MS, it could never be isolated.
In order to alter the properties of the RCM-substrate, the TIPS protecting group was removed from
5
by treatment with HF
was carried out with TBAF,
disappointment, however, all attempts at ring-closing olefin metathesis with 70 were also unsuccessful
•
py, which gave alcohol 70 in high yield (86%). Notably, if the deprotection
5
was mainly decomposed and the yield of 70 was poor (27%). To our
under a set of standard conditions (see Table S2). As for TIPS-ether
5, the material was either
decomposed to numerous unidentified products or remained unchanged.
As the original, RCM-based synthetic plan could not be implemented, we needed to assess possible
alternatives for ring-closure (excepting the commonly employed intramolecular HWE olefination at
C(2) C(3); vide supra) that would allow us to capitalize on the work expended at this point to the
−
greatest extent possible. In this context, the concept of relay ring-closing olefin metathesis (RRCM) as
developed by Hoye and co-workers presented itself as a plausible and attractive approach [65]. RRCM
broadens the scope of RCM to substrates that are either sterically hindered or electronically deactivated
by attaching a five-carbon tether with an easily accessible terminal double bond. The latter is intended
to direct the metathesis catalyst to the site of choice via a favorable loss of cyclopentene. Our first
choice was to attach the tether to the diene moiety of
display low reactivity towards metathesis catalysts.
5, as this part of the substrate was presumed to
The requisite substrate, substrate 76, was prepared according to the same Paterson aldol-based
strategy that had led to ene-diene (vide supra) (Scheme 12). Thus, departing from a mixture of
5
cis- and trans 2,7-octadienol (71), aldehyde 73 was obtained by PCC oxidation, Wittig olefination of
the ensuing trans/trans aldehyde (obtained in 60% yield) with phosphonium ylide 17, reduction of
the resulting unsaturated ester 72 with LiAlH4 and oxidation of the ensuing primary alcohol with
activated MnO2. Aldehyde 73 was obtained in 37% overall yield for the 4-step sequence from 71
.
Subsequent Paterson aldol reaction of 73 with the boron enolate derived from ketone 27 gave -hydroxy
β
ketone 74 in 6% yield, although with lower selectivity (dr 7.2:1) than for reaction of 27 aldehyde 26
(dr 14.8:1, vide supra). Reduction of 74 followed by regioselective TIPS protection of the resulting diol
gave 75; esterification of the latter with phosphonoacetic acid (55) and, finally, HWE reaction of the
phosphono ester with aldehyde 68 uneventfully delivered relay substrate 76 in 51% overall yield under
the conditions elaborated previously.
Various conditions were then screened to effect macrocyclization. Unfortunately, however,
macrocycle
reaction, to produce
by tether cleavage from 76 was observed at concentrations as low as 1.2 µM.
4
was never obtained; instead, the tether was cleaved within seconds in an intermolecular
5
, which did not undergo RCM, as had been found previously. The formation of
5
In addition to the conditions for macrocyclization that had been investigated in previous screenings
(see Tables S1 and S2), experiments were also conducted where 76 was added slowly to a refluxing
solution of the catalyst via a syringe pump, a procedure that had been reported by Porco and co-workers
to suppress unproductive relay cleavage [66]. However, this method proved to be unsuccessful with 76
Likewise, microwave-assisted heating according to Robinson et al. [67] or the use of CuI as a phosphine
scavenger [68], did not effect RRCM.
.
The resistance of substrate 76 towards RRCM led us to consider shifting the site of ring-closure
from C(9)/C(10) to C(11)/C(12) as a possible alternative approach towards macrocycle
RRCM substrate was prepared based on the same overall strategy that had been followed for the
synthesis of and 76. Thus, phosphono ester 80 (Scheme 13) was obtained from ketone 27 via Paterson
aldol reaction with methacrolein (77), stereoselective reduction of the ensuing -hydroxy ketone 78
4. The requisite
5
β
,
regioselective TIPS protection of the resulting diol product and esterification of the monoprotected diol
79 with diethylphosphono acetic acid (55) in 49% overall yield.

Molecules 2020, 25, 4527
13 of 27
Scheme 12. Reagents and conditions: (
plus 33% trans:cis 4:1; (
) LiAlH4, Et2O, 0 ◦C, 94%; (
85%; ( ) ( 27, (+)-DIPCl, NEt3, CH2Cl2,
78 ◦C, then 73
(pH7), MeOH, 0 ◦C to rt, 68%, dr 7.2:1; (
) NMe4BH(OAc)3, CH3CN/AcOH (2:1),
(from aldol); ( ) TIPSOTf, 2,6-lutidine, 3 Å sieves, CH2Cl2, ) diethylphosphonoacetic acid
78 ◦C, 85%; (
55), CME-CDI, DMAP, 3 Å sieves, CH2Cl2, 0 ◦C to rt, 82%; ( 68, DBU, LiCl, CH3CN, 0 ◦C to rt, 85%.
CME-CDI, N-cyclohexyl-N0-(
-[N-methyl-morpholino]ethyl) carbodiimide p-toluenesulfonate; DBU,
1,8-diazabicyclo[5.4.0]undec-7-ene; DIPCl, B-chlorodiisopinocampheylborane; TIPS, triisopropylsilyl.
a
) PCC, celite, CH2Cl2, 82%; (
) activated MnO2, 3 Å molecular sieves, CH2Cl2, rt,
78 ◦C to 20 ◦C; (ii) H2O2, phosphate buffer
40 ◦C to rt, 86%, dr 7.8:1
b) 17, CH2Cl2, rf, 60% trans only,
c
d
e
i
)
−
,
−
−
f
−
g
−
h
(
i)
β
Scheme 13. Reagents and conditions: (
ii) H2O2, phosphate buffer (pH7), MeOH, 0 ◦C to rt, 67%, dr 19.6:1; (
(1:1), ) TIPSOTf, 2,6-lutidine, 3 Å molecular sieves, CH2Cl2,
40 ◦C to 18◦C to rt, 95%; (
d) diethylphosphonoacetic acid (55), CME-CDI, DMAP, 3 Å molecular sieves, CH2Cl2, 0 ◦C to rt, 81%.
CME-CDI, N-cyclohexyl-N0-(
-[N-methyl-morpholino]ethyl) carbodiimide p-toluenesulfonate; DMAP,
i
) (+)-DIPCl, NEt3, CH2Cl2,
−
78 ◦C, then 77,
) NMe4BH(OAc)3, CH3CN/AcOH
78 ◦C, 97%;
−
78 ◦C to
−
20 ◦C;
(
b
−
−
c
−
β
4-(dimethylamino)pyridine, DIPCl, B-chlorodiisopinocampheylborane; TIPS, triisopropylsilyl.
Notably, of all aldehydes submitted to Paterson aldol reaction with ketone 27, methacrolein (77
gave the cleanest reaction and furnished the aldol product with the highest selectivity ((dr 19.6:1).
)
As shown in Scheme 14, aldehyde 83 was obtained from aldehyde 66 by Julia-Kocienski olefination
with sulfone 82 (obtained from 2,7-octadienol (81) via Mitsunobu reaction and subsequent oxidation of
the thioether with Mo7O24(NH4)6, H2O2 in 57% yield) followed by TBDPS removal and Swern oxidation
of the resulting free alcohol in 30% overall yield. HWE reaction of 80 and 83 under Masamune-Roush
conditions then furnished the RRCM substrate 84 in 69% yield.

Molecules 2020, 25, 4527
14 of 27
Scheme 14. Reagents and conditions:
(
a
) 2-Mercaptophenyltetrazole, PPh3, DEAD, THF, rt;
) LiHMDS, aldehyde 66, 34% (E/Z >
78 ◦C, 95%; (
80, DBU,
(
b
) Mo7O24(NH4)6, H2O2, EtOH, 0 ◦C to rt, 57% (2 steps); (
c
30:1); (
d
) TBAF, AcOH, THF, 0 ◦C to rt, 89%; (
e) (COCl)2, DMSO, NEt3, CH2Cl2,
−
f)
LiCl, CH3CN, 0 ◦C to rt, 69%. DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene; DEAD, diethyl azodicarboxylate;
LiHMDS, lithium bis(trimethylsilyl)amide; TBAF, tetra-n-butylammonium fluoride.
Unfortunately, treatment of 84 with Grubbs 2nd generation catalyst in toluene, which was the only
system that had shown some initial promise in screening experiments, did not result in macrocyclization
(employing catalyst loadings of 5 mol% or 10 mol% at temperatures of 45 ◦C, 60 ◦C, or 90 ◦C). As for
76, only the tether was cleaved and more and more side products appeared over time.
As simply shifting the position of the (projected) site of cyclization from C(9)/C(10) in 76 to
C(11)/C(12) in 84 did not enable RRCM, we went back to investigate the possibility of RRCM at the
originally conceived site between C(9) and C(10), but now with the tether attached to the terminal
double bond of the eastern part of the original RCM substrate 5. Directing the catalyst to this double
bond was felt to offer some promise, as the experiments with RRCM substrate 76 had suggested that it
did not react with a ruthenium-carbene at C10. The corresponding RRCM substrate 88 was obtained
by HWE of phosphono ester 69 and aldehyde 87 under Masamune-Roush conditions (Scheme 15).
The latter was prepared from the known phenyltetrazole sulfone 86 [66] via Julia-Koczienski olefination,
followed by TBDPS removal and Swern oxidation.
Scheme 15. Reagents and conditions: (
a
) 2-Mercaptophenyltetrazole, DEAD, PPh3, THF, rt; (
b
) H2O2,
(NH4)6Mo7O◦24, EtOH/THF (5:3), 0 ◦C to rt, 78% (2 steps); (
c
) NaHMDS, THF,
−
78 ◦C to
−9
◦C,
◦
◦
then 66,
CH2Cl2,
−
78 C to 0 C, 62%, E only; (
d
) TBAF, AcOH, THF, 0 C to rt, 94%; (
e
) (COCl)2,◦DMSO,
◦
◦
−
78 C, then NEt3,
−78 C to rt, 92%, E:Z = 9.6:1; (
f
)
87, DBU, LiCl, CH3CN, 0 C to rt,
85%. DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene; DEAD, diethyl azodicarboxylate; NaHMDS, sodium
bis(trimethylsilyl)amide; TBAF, tetra-n-butylammonium fluoride.

Molecules 2020, 25, 4527
15 of 27
Disappointingly, 88, as all other previous substrates, did not undergo macrocyclization to any
detectable extent for a variety of solvents and catalysts.
In light of the various unsuccessful attempts at making the rhizoxin core macrocycle either by RCM
or RRCM, the question arose if specific structural features of the corresponding cyclization precursors
might perhaps prevent the adoption of a favorable conformation for productive ring-closure. While
it was not entirely clear what these features would be, we hypothesized that macrocylization could
perhaps be promoted by a more flexible cyclization precursor that did not incorporate the δ-lactone
ring that had been present in all RCM/RRCM substrates investigated up to this point. To address this
question, we targeted the potential RRCM substrate 89 (Figure 2), which incorporates two additional,
rotatable bonds when compared to 5, 76, 84, and 88.
Figure 2. Structure of the potential RRCM substrate 89.
During the initial planning phase for the synthesis of 89, we assumed that ester 90 could serve as
a suitable intermediate that would also be readily accessible from lactone 67 by simple treatment with
NaOMe (Scheme 16).
Scheme 16. Reagents and conditions: (a) NaOMe, or NEt3, MeOH, or LiOMe; (b) (i) LiOH, THF/water
(1:1), 0 ◦C to rt, 98%, (ii) TMSCHN2, Et2O/MeOH, rt, quant (67). TMS, trimethylsilyl.
However, while initial attempts using a large excess of NaOMe led to led to significant but still
incomplete conversion to hydroxy ester 90 according to TLC, even neutral or slightly acidic conditions
during attempted purification on silica gel or in deuterated chloroform were sufficient to shift the
equilibrium back to the starting material 67. This was also the case for the trapped lithium carboxylate
of acid 91, which immediately recyclized to δ-lactone 67 after work-up or TMS diazomethane treatment.
For this reason, a longer route toward 91 had to be elaborated which is summarized in Scheme 17.
Thus, 67 was reduced with LiAlH4, to furnish a diol which was selectively TBS-protected at the primary
hydroxy group in 98% overall yield (from 67). Lewis acid-catalyzed PMB protection followed by
selective TBS deprotection furnished a primary alcohol that was elaborated into methyl ester 93 in 53%
overall yield in a three-step sequence involving Swern oxidation, Pinnick oxidation of the ensuing
aldehyde to the acid, and methyl ester formation with TMS-diazomethane. TBDPS deprotection with
buffered TBAF furnished a 9:1 mixture of lactone 94 and the corresponding hydroxy (methyl)ester in
88% combined yield. After sequential treatment of this (separable) mixture with lithium methoxide
and Dess-Martin periodinane, aldehyde 95 was isolated in 68% yield (93% brsm). Subsequent HWE
reaction of 95 with phosphonate 69 under Masamune-Roush conditions furnished ester 89.

Molecules 2020, 25, 4527
16 of 27
Scheme 17. Reagents and conditions: (
rt, 98% (2 steps); ( ) Sc(OTf)3 (6 mol%), PMB-2,2,2-trichloroacetimidate, toluene, rt; (
(4:1), rt, 67% (2 steps); ( ) ( ) (COCl)2, DMSO, CH2Cl2,
78 ◦C, then NEt3, 78 ◦C to rt, (ii) NaClO2,
NaH2PO4, 2-methyl-2-butene, tBuOH, THF, rt, (iii) TMSCHN2, toluene/MeOH, 0 ◦C, 79% (3 steps);
) TBAF, AcOH, THF, rt, 79%; g) LiOMe, MeOH, rt; (
) DMP, CH2Cl2, 0 ◦C to rt, 68% (2 steps) (93%
brsm); (
75, LiCl, DBU, THF/MeCN (3:1), 0 ◦C to rt, 78%. DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene;
a
) LiAlH4, Et2O, 0 ◦C to rt, quant; (
b) TBSCl, imidazole, CH2Cl2,
c
d) NaIO4, THF/H2O
e
i
−
−
(
f
h
i
)
DMP, Dess–Martin periodinane (1,1,1-tris(acetyloxy)-1,1-dihydro-1,2-benziodoxol-3-(1H)-one); PMB,
para-methoxy benzyl; TBS, tert-butyldimethylsilyl; TMS, trimethylsilyl.
Unfortunately, as for other RRCM substrates, Grubbs 2nd and Hoveyda-Grubbs 2nd generation
catalysts in dichloroethane led to efficient cleavage of the relay tether, but no macrocyclization was
observed based on MS and TLC analysis of reaction mixtures. In refluxing toluene, both catalysts led
to a complex mixture of products as evidenced by HPLC and NMR analysis.
2.5. Macrocyclization by Ring-Closing Alkyne Metathesis (RCAM)
At the point where our numerous attempts at synthesizing the macrocyclic portion of rhizoxin F
(
2
) by means of RCM/RRCM had all met with failure, we felt that a radical change in strategy was
necessary. Triggered by Fürstner’s groundbreaking work on RCAM in the field of natural product
synthesis (for early examples, see in [69 78]; for more recent examples, see in [23 79 81], we reasoned
–
,
–
that this method would be a promising alternative to RCM. (For a review on the use of RCAM in
natural product synthesis, see in [82].) Thus, macrocyclization was planned to be achieved by RCAM
at C(9)/C(10), i.e., at the same site that was initially foreseen for RCM, leading to diyne 113 as the
ultimate cyclisation precursor (Scheme 18). Obviously, the triple bond in the RCAM product would
have to be converted into a trans double bond, but methods for the trans-selective reduction of alkynes
to olefins had been described (see, for example, in [83–85]).
Scheme 18. Global retrosynthesis of 5 via 113.

Molecules 2020, 25, 4527
17 of 27
Diyne 113 was to be obtained from phosphono ester 101 and aldehyde 109 in analogy to the
synthesis of the various RCM/RRCM substrates. This approach offered the advantage that it would
allow us to capitalize on the chemistry developed in the course of the synthesis of the corresponding
RCM/RRCM macrocyclization precursors. The largest part of the chemistry described in the following
has already been reported in [20], but is included here to provide a coherent and complete narrative of
the work that finally led to the successful synthesis of rhizoxin F.
2.5.1. Synthesis of Building Blocks
As illustrated in Scheme 19, the synthesis of phosphono ester 101 relied on the same type of aldol
chemistry that had been successfully exploited in the synthesis of the different RCM/RRCM substrates.
In this context, a short route was devised towards the requisite aldehyde 99, making use of vinyl iodide
102, which was already at hand from our previous synthesis of ketone 27. Thus, 102 was submitted
to Sonogashira cross-coupling with propyne to yield ene-yne 98, which was readily oxidized to the
desired aldehyde 99 (Scheme 19).
Scheme 19. Reagents and conditions: (
) 17, CH2Cl2, rf, 45%; (
) LiAlH4, Et2O, 0 ◦C; (
82%; ( ) propyne, DIPEA, CuI (20 mol%), PdCl2(PPh3)2 (15 mol%), THF,
(+)-DIPCl, NEt3, CH2Cl2,
78 ◦C, then 99, 78 ◦C to 18 ◦C; (ii) H2O2, phosphate buffer (pH7), MeOH,
0 ◦C to rt, ca. 73%, dr n.d.; ( 40 ◦C to rt, 90%; (
) TIPSOTf,
2,6-lutidine, 3 Å molecular sieves, CH2Cl2, ) diethylphosphonoacetic acid (55), CME-CDI,
a
) Activated MnO2, 3 Å molecular sieves, CH2Cl2, rt, 82%;
) activated MnO2, 3 Å molecular sieves, CH2Cl2, rt,
78 ◦C to rt, 63%; (
) ( ) 27,
(
b
c
d
e
−
f i
−
−
−
g
) NMe4BH(OAc)3, CH3CN/AcOH (2:1.5),
−
h
−
◦ 78 ◦C, 97%; (
i
DMAP, 3 Å molecular sieves, CH2Cl2, 0 C to rt, 82%. DIPCl, B-Chlorodiisopinocampheylborane;
DIPEA, N,N-diisopropylethylamine (Hünig’s base); TIPS, triisopropylsilyl.
However, the Sonogashira coupling proved to be very tedious; not only were vinyl iodide 102
and the coupling product 98 inseparable on silica gel (TLC), thus creating the need for reaction
monitoring by RP-HPLC and requiring full conversion of the starting material, but the coupling also
relied on the use of expensive propyne and a precious catalyst. Furthermore, it was found that the
optimized conditions in small-scale experiments were not amenable to scale up (>1 g), which was a
key requirement at this early stage of the synthesis. For these reasons an alternative route towards 99
was sought. In the event, oxidation of 2-butyn-1-ol (96) with activated MnO2 and subsequent Wittig
olefination of the resulting propargylic aldehyde with phosphonium ylide 17 furnished ester 97, which
was smoothly reduced to allylic alcohol 98 (Scheme 19). In contrast to gaseous propyne, 2-butyn-1-ol
(96) is an easy-to-handle and relatively cheap liquid. Furthermore, phosphonium ylide 77 could be

Molecules 2020, 25, 4527
18 of 27
conveniently synthesized in large quantities (>120 g). The synthesis of 99 from 96 could thus be scaled
up without difficulties.
With aldehyde 99 in hand, the synthesis of phosphono ester 101 could be addressed, relying
on previously established protocols. Thus, Paterson aldol reaction of 99 with ketone 27 followed by
directed reduction of β-hydroxy ketone 100 according to Evans [35], regioselective TIPS protection of
the ensuing diol, and esterification with diethylphosphonacetic acid (55) furnished 101 in 52% overall
yield as a single isomer.
The synthesis of aldehyde 109 started from crude aldehyde 66 (obtained by Dess-Martin oxidation
from primary alcohol 65, cf. Scheme 11). In an initial experiment on small scale, 66 could be directly
converted into terminal alkyne 103 by Seyferth-Gilbert alkynylation [86] (Scheme 20). However, 103
was obtained only in moderate yield (43%) and no efforts were spent on optimizing this transformation.
Instead, we decided to assess the feasibility of the alkynylation of 66 via a Corey-Fuchs reaction as an
alternative approach. In the event, dibromo olefin 104 could be prepared from 66 in good yield (80%)
but, disappointingly, the conversion of the dibromo alkene moiety into a triple bond failed completely
and the treatment of 104 with nBuLi led to complete decomposition of starting material. However,
the installation of the triple bond via the dibromo alkene moiety became possible after reduction of the
lactone group to the corresponding hemiacetal with DIBAL and protection of the latter with a TBS
group, to furnish mixed acetal 105 as a single isomer in excellent yield (87%) (Scheme 20). Treatment of
105 with nBuLi and trapping of the intermediate acetylide with MeI gave alkyne 106 in 63% yield over
the four step sequence from aldehyde 66.
Scheme 20. Reagents and conditions: (
phosphonate), THF, ) CBr4, PPh3, CH2Cl2,
78 ◦C, 43%; (
78 C, dr 1:1, (ii) TBSCl, imidazole, CH2Cl2, rt, 87% (2 steps) (single isomer); d)
78 ◦C to rt, 90%; ( ) TBAF, AcOH, THF, 0 ◦C to rt, quant. (107/108, 2:1); (
) TEMPO,
a
) KOtBu, Seyferth-Gilbert reagent (dimethyl (diazomethyl)
−
b
−
78 ◦C, 80% (2 steps, based on 65); (
c) (i)
◦
DIBAL, CH2Cl2,
nBuLi, MeI, THF,
−
−
e
f
BAIB, Yb(OTf)3 (cat.), CH2Cl2, 0 ◦C to rt, 109: 62%, 110: 13%. BAIB, bis(acetoxy)iodobenzene; DIBAL,
diisobutylaluminum hydride; TBAF, tetra-n-butylammonium fluoride; TBS, tert-butyldimethylsilyl;
TEMPO, 2,2,6,6-tetramethylpiperidine-1-oxyl.

Molecules 2020, 25, 4527
19 of 27
Simultaneous cleavage of both silyl ethers with TBAF/AcOH produced an inseparable mixture
of lactols 107 and 108 (each as a pair of diastereoisomers; 2:1 ratio in favor of the desired
regioisomer 107), which was treated with bisacetoxy iodobenzene (BAIB) (2.2 equiv) in the presence
of catalytic amounts of TEMPO (19 mol%) and Yb(OTf)3 (4 mol%) (this system is similar to the
iodosylbenzene/TEMPO/Yb(OTf)3 system reported by Vatèle [87]) to produce the desired building
block 109 as a single isomer in 62% yield (based on 106) along with the corresponding hydroxy lactone
110 (13%).
The above synthesis of aldehyde 109 from alcohol 57 involved an avoidable protection/deprotection
sequence (cf. Scheme 11); a slightly shorter route to 109 was thus implemented in the scale-up process
that relied on lactone reduction at the stage of alcohol 57 (Scheme 21).
Scheme 21. Reagents and conditions: (
CH2Cl2, rt, 90% (2 steps) (single isomer); (
to rt, 88% or (COCl)2, DMSO, CH2Cl2,
78 ◦C, then NEt3,
78 ◦C, 99%; ( 78 ◦C to rt, then MeI, rt, 94%; (
) nBuLi, THF,
(cat.), NMO, CH2Cl2, rt, 50% (2 steps). DIBAL, diisobutylaluminum hydride; DMP, Dess–Martin
periodinane (1,1,1-tris(acetyloxy)-1,1-dihydro-1,2-benziodoxol-3-(1H)-one); NMO, 4-methylmorpholine
N-oxide; TBAF, tetra-n-butylammonium fluoride; TPAP, tetrapropyl-ammonium perruthenate.
a
) DIBAL, CH2Cl2,
) NaIO4, THF/water (4:1), rt, 87%; (
78 ◦C to rt, 96%; (
) TBAF, AcOH, THF, 0 ◦C to rt; (
h
−
78 ◦C, dr 1:1; (
b
) TBDPSCl, imidazole,
) DMP, CH2Cl2, 0 ◦C
) CBr4, PPh3, CH2Cl2,
) TPAP
c
d
−
−
e
−
f
−
g
The resulting 1:1 mixture of hemiacetals was then converted into tris-silyl derivative 111, which
was obtained as a single isomer in 90% yield from 57. This finding is in line with Leahy’s observation
that the all-equatorial isomer of the lactol is formed preferentially after equilibration under basic
conditions [88]. Selective cleavage of the TBS-ether with NaIO4 [61] followed by either Dess-Martin
or Swern oxidation gave an aldehyde (in 88% or 96% yield, respectively) that could be converted
into dibromo olefin 112 in 99% yield. The latter was elaborated into aldehyde 109 by treatment with
nBuLi/MeI, followed by TBDPS removal and TPAP oxidation. In contrast to the previously used
TEMPO-mediated oxidation of 107
yield over 2 steps reproducibly. The former method proved to be rather unreliable for scale-up, with
yields decreasing down to 20%. Aldehyde 109 could also be obtained from 107 108 by Dess-Martin
/108, TPAP/NMO gave aldehyde 109 as a single compound in 50%
/
oxidation, but the material was very difficult to purify by column chromatography.
2.5.2. Building Block Assembly and Completion of the Synthesis of Rhizoxin F (2)
The HWE reaction between aldehyde 109 and phosphono ester 101 proceeded smoothly to deliver
diyne 113 in 81% yield (Scheme 22).
Gratifyingly, and very much to our relief, heating 113 with the active catalysts obtained by in
situ decomplexation of Mo-complexes 114 or 115 (complexes 114 and 115 were generous gifts from
Prof. Alois Fürstner, Max Planck Institute for Carbon Research, Mühlheim/Ruhr, Germany) with
MnCl2 to 125 ◦C in toluene for 2.45 h (for 114) or 27 h (for 115) gave the desired macrocycle 116 in
69% and 56%/31% yield (The two yields were obtained in two independent experiments on a 700 mg
scale of 113 that were performed under ostensibly identical conditions. The reasons for the significant
discrepancy in yields are unknown at this point.), respectively (Scheme 22). Heating of the reaction to
at least 120 ◦C was necessary to achieve a practical rate of cyclization; at the same time, this finding
also highlights the exceptional thermal stability of active catalysts derived from complexes 114 and
115 in the reaction mixture. While the yields obtained in the macrocyclization reaction as such may

Molecules 2020, 25, 4527
20 of 27
not be considered spectacular, given our prior extensive unsuccessful efforts at RCM, they were an
impressive demonstration of the power of RCAM is an efficient method for macrocyclization.
Scheme 22. Reagents and conditions: (a b) Mo-complex 114
) DBU, LiCl, CH3CN, 0 ◦C to rt, 81%; (
(35 mol%) or 115 (15 mol%), MnCl2, toluene, 5 Å molecular sieves, 125 ◦C; 114: 69%, 115: 56%/31%%.
DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene.
With the hurdle of the ring-closure finally cleared, we now had to address the E-selective alkyne
reduction of the newly formed triple bond. While the possibility of generating Z or E olefins from
alkynes selectively, for example, by hydrosilylation/desilylation, is usually regarded as one of the
benefits of RCAM, this transformation turned out to be a major challenge for 116. In spite of extensive
efforts, both the yield and selectivity of hydrosilylation and protodesilylation steps were unsatisfactory
and no conditions could be identified to affect clean conversion of the ene-yne portion in 116 into the
desired E,E diene moiety.
However, in the context of the above experiments, we had obtained a small amount of the C(9)/C(10)
Z isomer of
in benzene [89]. This observation suggested that an alternative approach towards the conversion of
116 into could be the Z-selective reduction of the triple bond followed by double bond isomerization.
5, which we found could be cleanly isomerized to 5 by refluxing with thiophenol and AIBN
4
As a consequence, experiments were conducted aiming to access C(9)/C(10)-Z-5 (i.e., 117, Scheme 23)
selectively; this included the examination of Red-Al-, Birch-, SmI2-, and hydrazine-promoted alkyne
reductions, and a plethora of hydrogenations (transfer methods or application of elemental H2) and
hydrometallations (with boron, tin, silicon), which are not to be discussed here in detail. Suffice it to
say, that none of these methods delivered the desired diene 117.
While alkyne 116 could not be reduced to the corresponding Z olefin directly, it could be
converted cleanly into its biscobalthexacarbonyl complex 116A (Scheme 23) by simple treatment with
Co2(CO)8 [90
the reductive decomplexation of endocyclic biscobalthexacarbonyl complexes with rhodium on
charcoal [92 93] or Wilkinson catalyst [94] under elevated hydrogen pressure to give olefins. They later
,91]. We considered this a promising finding, as Isobe and co-workers had reported
,
discovered that the reductive decomplexation could also be brought about not only under radical
conditions using nBu3SnH [94], but also with the less toxic sodium hypophosphite monohydrate [95].

Molecules 2020, 25, 4527
21 of 27
Scheme 23. Reagents and conditions:
hypophosphite, benzene, reflux, 74% over 3 cycles, Z only; (
E/Z = 20:1; ( , PdCl2(MeCN)2, DMF, rt, 68%; ( ) HF py, py, THF, 0 C to rt, 54%; (e) tBuOOH,
(a) (
i
) Co2(CO)8, CH2Cl2, rt; (ii) N-ethylpiperidinium
b
) AIBN, PhSH, benzene, reflux, 88%,
◦
c
)
3
d
·
VO(acac)2, benzene, 0 ◦C to rt, 65%, 29% after preparative HPLC. acac, acetylacetonato; AIBN,
azobisisobutyronitrile; py, pyridine.
In our case, reductive decomplexation with nBu3SnH was not compatible with the vinyl iodide
moiety and the use of hydrogen at elevated pressure with Wilkinson’s catalyst or rhodium on
charcoal was considered somewhat impracticable. Thus, the decomplexation of 116A with sodium
hypophosphite was deemed to be the most viable approach towards diene 117. Unfortunately, Isobe’s
original protocol led to transesterification of the δ-lactone moiety with the solvent 2-ethoxy ethanol
(cellosolve). At the same time, partial reduction of the complex was observed, thus suggesting that
117 might indeed be accessible from 116A if the decomplexation was carried out in a different solvent.
When the reaction was performed in DMSO or CH3CN/H2O (6:1), no reduction took place. In order to
allow for screening also in more lipophilic solvents, the tetrabutylammonium salt of hypophosphorous
acid was then prepared, which proved to be soluble in CH2Cl2 and THF. Treatment of 116A with this
salt in CH2Cl2 or THF afforded a mixture of diene 117 and ene-yne 116 (ca. 50% conversion).
The ultimate choice of reducing agent was N-ethylpiperidinium hypophosphite, which was
commercially available and which had been employed as a substitute for nBu3SnH in the radical
deoxygenation of alcohols [96]. N-Ethylpiperidinium hypophosphite proved to be more soluble than
tetrabutylammonium hypophosphite; the most soluble reductant investigated was diphenylphosphine
oxide (DPPO) [97], but this only led to decomposition of 116A. In analogy to Isobe’s protocol
with nBu3SnH [94], benzene was the solvent of choice for the reduction of 116A to 117 with
N-ethylpiperidinium hypophosphite. After some fine-tuning of conditions, the reaction reliably
afforded C(9)/C(10) Z olefin 117 in a highly selective fashion. In order to achieve full consumption
of alkyne 116, the two-step sequence of complex formation and reductive decomplexation had to be
performed three times, as 116 was inseparable from olefin 117 by silica gel chromatography. This
procedure initially provided 117 in 47% yield; while this yield was moderate, it was still considered
very satisfactory, given the fact that an array of well-established methods had failed in reducing
ene-yne 116. Gratifyingly, it was discovered in subsequent experiments (by serendipity) that the yield
for the conversion of 116 into 117 could be increased substantially, up to 74%, if an aqueous workup
after the hypophosphite reduction was omitted.

Molecules 2020, 25, 4527
22 of 27
Radical isomerization of 117 with AIBN and thiophenol then proceeded smoothly furnished the
desired E,E-diene in 88% yield (Scheme 23). Subsequent Stille coupling of with the known vinyl
stannane 118 29], followed by HF-promoted silyl ether cleavage, gave rhizoxin D ( ) in 37% overall
4
4
[
3
yield. For reasons unknown, this yield is lower than those reported in the literature for the same
two-step sequence (50% [29] and 76% [98]); in our hands, all intermediates with the C(15) side chain
attached were prone to decomposition, in agreement with reports by Hertweck [99] and Stella [100].
Finally, directed epoxidation of rhizoxin D (3) according to Ohno and co-workers [19] proceeded in
65% yield, thus completing the first synthesis of rhizoxin F (2).
3. Materials and Methods
Detailed protocols for the synthesis of new compounds and the associated analytical data can be
found in the Supplementary Material.
4. Conclusions
The total synthesis of the bacterial macrolide rhizoxin F (2) (also known as WF1360F) was
accomplished based on macrocyclization by ring-closing alkyne metathesis (RCAM) at the C(9)–C(10)
site, followed by a highly selective radical-based reduction/isomerization sequence to install the
macrocyclic (E,E)-diene unit. The synthesis comprised 24 steps for the longest linear sequence from
1,4-butanediol (44) and delivered the target structure in ~1.5% yield overall yield. Notably, in spite of
extensive efforts, all attempts to access macrocycle
4 by ring-closing olefin metathesis (RCM) met with
complete failure. Our findings not only further highlight the general potential of RCAM in natural
product synthesis, which by now is well established through the work of Fürstner and others; but it
also shows that RCAM can serve as a valuable alternative when macrocyclization cannot be effected
by RCM. We have also developed a new alkyne semireduction protocol, which reliably furnished
diene
the elaboration of macrocyclic alkynes obtained by RCAM into the corresponding olefins. Finally,
although this is not discussed here, macrocyclic vinyl iodide has been smoothly converted into a
number of side chain-modified analogs of ; in addition, our overall strategy for the synthesis of was
also adaptable to the synthesis of a (side chain-modified) lactam analog.
4 in good yields and with exquisite selectivity. This protocol provides an additional option for
4
2
2
Supplementary Materials: The following are available online. synthesis protocols and analytical data for all new
compounds, Table S1: RCM screening for substrate 5, Table S2: RCM screening for substrate 70.
Author Contributions: For research articles with several authors, a short paragraph specifying their individual
contributions must be provided. Conceptualization, K.-H.A., C.M.N., and M.L.; methodology, K.-H.A., C.M.N.,
and M.L.; formal analysis, C.M.N., M.L., and K.-H.A.; investigation, C.M.N. and M.L.; data curation, C.M.N.,
and M.L.; writing, K.-H.A.; supervision, K.-H.A.; funding acquisition, K.-H.A. All authors have read and agreed
to the published version of the manuscript.
Funding: This research was funded by the Swiss National Science Foundation (SNF projects 200021_126511 and
200020_143269).
Acknowledgments: We thank Kurt Hauenstein, ETH Zürich, for excellent technical assistance; Bernhard Pfeiffer,
ETH Zürich, for NMR support; and Louis Bertschi, ETH Zürich and the entire ETHZ-LOC MS-Service for
HRMS spectra acquisition. We are very grateful to Alois Fürstner, Max Planck Institute for Carbon Research,
Mülheim/Ruhr, Germany, for a generous gift of Mo-complexes 114 and 115. We also want to thank one of the
reviewers for the very careful reading of the manuscript. The comments of this reviewer have led to a significant
improvement of our manuscript.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the
study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to
publish the results.

Molecules 2020, 25, 4527
23 of 27
References
1.
Iwasaki, S.; Kobayashi, H.; Furukawa, J.; Namikoshi, M.; Okuda, S. Studies on macrocyclic lactone antibiotics.
vii. structure of a phytotoxin "Rhizoxin" produced by Rhizopus Chinensis. J. Antibiot. 1984, 37, 354–362.
[CrossRef] [PubMed]
2.
Iwasaki, S.; Namikoshi, M.; Kobayashi, H.; Furukawa, J.; Okuda, S. Studies on macrocyclic lactone antibiotics.
viii. Absolute structures of rhizoxin and a related compound. J. Antibiot. 1986, 39, 424–429. [CrossRef]
[PubMed]
3.
4.
5.
6.
Takahashi, M.; Matsumoto, S.; Iwasaki, S.; Yahara, I. Molecular basis for determining the sensitivity of
eucaryotes to the antimitotic drug rhizoxin. Mol. Gen. Genet. 1990, 222, 169–175. [CrossRef] [PubMed]
Koga-Ban, Y.; Niki, T.; Nagamura, Y.; Sasaki, T.; Minobe, Y. cDNA Sequences of Three Kinds of β-tubulins
from Rice. DNA Res. 1995, 2, 21–26. [CrossRef]
Partida-Martinez, L.P.; Hertweck, C. Pathogenic fungus harbours endosymbiotic bacteria for toxin production.
Nature 2005, 437, 884–888. [CrossRef]
Partida-Martinez, L.P.; Hertweck, C. A Gene Cluster Encoding Rhizoxin Biosynthesis in “Burkholderia
rhizoxina”, the Bacterial Endosymbiont of the Fungus Rhizopus microspores. ChemBioChem 2007, 8, 41–45.
[CrossRef]
7.
8.
9.
Scherlach, K.; Partida-Martinez, L.P.; Dahse, H.-M.; Hertweck, C. Antimitotic Rhizoxin Derivatives from a
Cultured Bacterial Endosymbiont of the Rice Pathogenic Fungus Rhizopus microsporus. J. Am. Chem. Soc.
2006, 128, 11529–11536. [CrossRef]
Prota, A.; Bargsten, K.; Díaz, J.F.; Marsh, M.; Cuevas, C.; Liniger, M.; Neuhaus, C.; Andreu, J.M.; Altmann, K.-H.;
Steinmetz, M.O. A new tubulin-binding site and pharmacophore for microtubule-destabilizing anticancer drugs.
Proc. Natl. Acad. Sci. USA 2014, 111, 13817–13821. [CrossRef]
Iwasaki, S.; Namikoshi, M.; Kobayashi, H.; Furukawa, J.; Okuda, S. Studies on macrocyclic lactone antibiotics.
IX: Novel macrolides from the fungus Rhizopus chinensis: Precursors of rhizoxin. Chem. Pharm. Bull. 1986
34, 1387–1390. [CrossRef]
,
10. Kiyoto, S.; Kawai, Y.; Kawakita, T.; Kino, E.; Okuhara, M.; Uchida, I.; Tanaka, H.; Hashimoto, M.; Terano, H.;
Kohsaka, M.; et al. A new antitumor complex, Wf-1360, Wf-1360A, B, C, D, E and F. J. Antibiot. 1986, 39,
762–772. [CrossRef]
11. Brendel, N.; Partida-Martinez, L.P.; Scherlach, K.; Hertweck, C. A cryptic PKS-NRPS gene locus in the plant
commensal Pseudomonas fluorescens Pf-5 codes for the biosynthesis of an antimitotic rhizoxin complex.
Org. Biomol. Chem. 2007, 5, 2211–2213. [CrossRef] [PubMed]
12. Loper, J.E.; Henkels, M.D.; Shaffer, B.T.; Valeriote, F.A.; Gross, H. Isolation and identification of rhizoxin
analogs from Pseudomonas fluorescens Pf-5 by using a genomic mining strategy. Appl. Environ. Microbiol.
2008, 74, 3085–3093. [CrossRef] [PubMed]
13. Loper, J.E.; Henkels, M.D.; Rangel, L.I.; Olcott, M.H.; Walker, F.L.; Bond, K.L.; Kidarsa, T.A.; Hesse, C.N.;
Sneh, B.; Stockwell, V.O.; et al. Rhizoxin analogs, orfamide A and chitinase production contribute to the
toxicity of Pseudomonas protegens strain Pf-5 to Drosophila melanogaster. Environ. Microbiol. 2016, 18,
3509–3521. [CrossRef] [PubMed]
14. Sullivan, A.S.; Prasad, V.; Roach, M.C.; Takahashi, M.; Iwasaki, S.; Luduena, R.F. Interaction of rhizoxin with
bovine brain tubulin. Cancer Res. 1990, 50, 4277–4280.
15. Takahashi, M.; Iwasaki, S.; Kobayashi, H.; Okuda, S.; Murai, T.; Sato, Y. Rhizoxin binding to tubulin at the
maytansine-binding site. Biochim. Biophys. Acta 1987, 926, 215–223. [CrossRef]
16. Hendriks, H.R.; Plowman, J.; Berger, D.P.; Paull, K.D.; Fiebig, H.H.; Fodstad, Ø.; van der Meulen, H.C.D.;
Henrar, R.E.C.; Pinedo, H.M.; Schwartsmann, G. Preclinical antitumour activity and animal toxicology
studies of rhizoxin, a novel tubulin-interacting agent. Ann. Oncol. 1992, 3, 755–763. [CrossRef]
17. Tsuruo, T.; Ohara, T.; Iida, H.; Tsukagoshi, S.; Sato, Z.; Matsuda, I.; Iwasaki, S.; Okuda, S.; Shimizu, F.;
Sasagawa, K.; et al. Rhizoxin, a macrocyclic lactone antibiotic, as a new antitumor agent against human and
murine tumor cells and their vincristine-resistant sublines. Cancer Res. 1986, 46, 381–385.
18. Hanauske, A.-R.; Catimel, G.; Aamdal, S.; Huinink, W.T.B.; Paridaens, R.; Pavlidis, N.; Kaye, S.B.; Velde, A.T.;
Wanders, J.; Verweij, J. Phase II clinical trials with rhizoxin in breast cancer and melanoma. Br. J. Cancer 1996
,
73, 397–399. [CrossRef]

Molecules 2020, 25, 4527
24 of 27
19. Nakada, M.; Kobayashi, S.; Shibasaki, M.; Iwasaki, S.; Ohno, M. The first total synthesis of the antitumor
macrolide, rhizoxin. Tetrahedron Lett. 1993, 34, 1039–1042. [CrossRef]
20. Neuhaus, C.M.; Liniger, M.; Stieger, M.; Altmann, K.-H. Total synthesis of the tubulin inhibitor WF-1360F
based on macrocycle formation through ring-closing alkyne metathesis. Angew. Chem. Int. Ed. 2012, 52,
5866–5870. [CrossRef]
21. Hong, J.; White, J.D. The chemistry and biology of rhizoxins, novel antitumor macrolides from Rhizopus
chinensis. Tetrahedron 2004, 60, 5653–5681. [CrossRef]
22. Mitchell, I.S.; Pattenden, G.; Stonehouse, J. A total synthesis of the antitumour macrolide rhizoxin. Org.
Biomol. Chem. 2005, 3, 4412–4431. [CrossRef] [PubMed]
23. Karier, P.; Ungeheuer, F.; Ahlers, A.; Anderl, F.; Wille, C.; Fürstner, A. Metathesis at an Implausible Site:
A Formal Total Synthesis of Rhizoxin D. Angew. Chem. Int. Ed. 2018, 58, 248–253. [CrossRef] [PubMed]
24. Mukayama, T.; Naraska, K.; Banno, K. A New Aldol Type Reaction. Chem. Lett. 1973, 2, 1011–1014. [CrossRef]
25. Mukaiyama, T.; Banno, K.; Narasaka, K. New Cross-Aldol reactions. Reactions of Silyl Enol Ethers with
Carbonyl Compounds Activated by Titanium Tetrachloride. J. Am. Chem. Soc. 1974, 96, 7503–7509. [CrossRef]
26. Paterson, I.; Goodman, M. Aldol reactions of methylketones using chiral boron reagents: A reversal in
aldehyde enantioface selectivity. Tetrahedron Lett. 1989, 30, 997–1000. [CrossRef]
27. Hosomi, A.; Sakurai, H. Syntheses of γ,δ-unsaturated alcohols from allylsilanes and carbonyl compounds in
the presence of titanium tetrachloride. Tetrahedron Lett. 1976, 16, 1295–1298. [CrossRef]
28. Keck, G.E.; Tarbet, K.H.; Geraci, L.S. Catalytic Asymmetric Allylation of Aldehydes. J. Am. Chem. Soc. 1993
115, 8467–8468. [CrossRef]
,
29. White, J.D.; Blakemore, P.R.; Green, N.J.; Hauser, E.B.; Holoboski, M.A.; Keown, L.E.; Nylund Kolz, C.S.;
Phillips, B.W. Total Synthesis of Rhizoxin D, a Potent Antimitotic Agent from the Fungus Rhizopus chinensis.
J. Org. Chem. 2002, 67, 7750–7760. [CrossRef]
30. Menche, D.; Hassfeld, J.; Li, J.; Rudolph, S. Total synthesis of archazolid A. J. Am. Chem. Soc. 2007, 129,
6100–6101. [CrossRef]
31. Dess, D.B.; Martin, J.C. A Useful 12-I-5 Triacetoxyperiodinane (the Dess-Martin Periodinane) for the Selective
Oxydation of Primary or Secondary Alcohols and a Variety of Related 12-I-5 Species. J. Am. Chem. Soc. 1991
113, 7277–7278. [CrossRef]
,
32. House, H.O.; Rasmussen, G.H. Stereoselective Synthesis of α-Substituted α,β-Unsaturated Esters. J. Org.
Chem. 1961, 26, 11–4278. [CrossRef]
33. Nahm, S.; Weinreb, S.M. N-methoxy-n-methylamides as effective acylating agents. Tetrahedron Lett. 1981, 22,
3815–3818. [CrossRef]
34. Cazeau, P.; Duboudin, F.; Moulines, F.; Babot, F.; Dunogues, J. A new practical synthesis of silyl enol ethers:
II. From α,β-unsaturated aldehydes and ketones. Tetrahedron 1987, 43, 2089–2093. [CrossRef]
35. Evans, D.A.; Chapman, K.T.; Carreira, E.M. Directed Reduction of
β-Hydroxy Ketones Employing
Tetramethylammonium Triacetoxyborohydride. J. Am. Chem. Soc. 1988, 110, 3560–3578. [CrossRef]
36. Sugiyama, H.; Yokokawa, F.; Shioiri, T. Total synthesis of mycothiazole, a polyketide heterocycle from marine
sponges. Tetrahedron 2003, 59, 6579–6593. [CrossRef]
37. Poon, K.W.C.; Dudley, G.B. Mix-and-Heat Benzylation of Alcohols Using a Bench-Stable Pyridinium Salt.
J. Org. Chem. 2006, 71, 3923–3927. [CrossRef]
38. Iversen, T.; Bundle, D.R. Benzyl trichloroacetimidate, a versatile reagent for acid-catalysed benzylation of
hydroxy-groups. Chem. Commun. 1981, 1240–1241. [CrossRef]
39. Negishi, E.-I. Palladium- or Nickel-Catalyzed Cross Coupling. A New Selective Method for Carbon-Carbon
Bond Formation. Acc. Chem. Res. 1982, 15, 340–348. [CrossRef]
40. Tamao, K.; Sumitani, K.; Kumada, M. Selective Carbon-Carbon Bond Formation by Cross-Coupling of
Grignard Reagents with Organic Halides. Catalysis by Nickel-Phosphine Complexes. J. Am. Chem. Soc.
1972, 94, 4374–4376. [CrossRef]
41. Knappke, C.E.; Von Wangelin, A.J. 35 years of palladium-catalyzed cross-coupling with Grignard reagents:
How far have we come? Chem. Soc. Rev. 2011, 40, 4948–4962. [CrossRef] [PubMed]
42. Ishihara, K.; Mouri, M.; Gao, Q.; Maruyama, T.; Furuta, K.; Yamamoto, H. Catalytic Asymmetric Allylation
Using a Chiral (Acyloxy)borane Complex as a Versatile Lewis Acid Catalyst. J. Am. Chem. Soc. 1993, 115,
11490–11495. [CrossRef]

Molecules 2020, 25, 4527
25 of 27
43. Williams, D.R.; Meyer, K.G. Total Synthesis of (+)-Amphidinolide K. J. Am. Chem. Soc. 2001, 123, 765–766.
[CrossRef]
44. Schlosser, M. Superbases for organic synthesis. Pure Appl. Chem. 1988, 60, 1627–1634. [CrossRef]
45. Prandi, C.; Venturello, P. α,β-Unsaturated Acetals as Precursors of α-Substituted Ethoxy Dienes. Useful
Reagents for Nucleophilic Acylation. J. Org. Chem. 1994, 59, 5458–5462. [CrossRef]
46. Moreno-Dorado, F.J.; Guerra, F.M.; Manzo, F.L.; Aladro, F.J.; Jorge, Z.D.; Massanet, G.M. CeCl3/NaClO: A
safe and efficient reagent for the allylic chlorination of terminal olefins. Tetrahedron Lett. 2003, 44, 6691–6693.
[CrossRef]
47. Smith, A.B.; Sfouggatakis, C.; Risatti, C.A.; Sperry, J.B.; Zhu, W.; Doughty, V.A.; Tomioka, T.; Gotchev, D.B.;
Bennett, C.S.; Sakamoto, S.; et al. Spongipyran synthetic studies. Evolution of a scalable total synthesis of
(+)-spongistatin 1. Tetrahedron 2009, 65, 6489–6509. [CrossRef]
48. Weigand, S.; Brückner, R. Direct Preparation of Allylstannanes from Allyl Alcohols: Convenient Synthesis of
β
-Substituted Allylstannanes and of Stereodefined
[CrossRef]
49. Takano, S.; Inomata, K.; Samizu, K.; Tomita, S.; Yanase, M.; Suzuki, M.; Iwabuchi, Y.; Sugihara, T.;
γ-Substituted Allylstannanes. Synthesis 1996, 475–482.
Ogasawara, K. A Convenient One-flask Synthesis of
α-Methylenealdehydes from Primary Alcohols.
Chem. Lett. 1989, 1283–1284. [CrossRef]
50. Naruta, Y.; Nishigaichi, Y.; Maruyama, K. Extremely Facile and Stereoselective Preparation of Allylstannanes
with Use of Ultrasound. Chem. Lett. 1986, 1857–1860. [CrossRef]
51. Corey, E.J.; Imwinkelried, R.; Pikul, S.; Xiang, Y.B. Practical enantioselective Diels-Alder and aldol reactions
using a new chiral controller system. J. Am. Chem. Soc. 1989, 111, 5493–5495. [CrossRef]
52. Williams, D.R.; Meyer, K.G.; Shamim, K.; Patnaik, S. Diastereoselectivity in asymmetric allylations: The role
of vicinal chirality in the allyl nucleophile for SE20 reactions with aldehydes. Can. J. Chem. 2004, 82, 120–130.
[CrossRef]
53. Hoye, T.R.; Jeffrey, C.S.; Shao, F. Mosher ester analysis for the determination of absolute configuration of
stereogenic (chiral) carbinol carbons. Nat. Protoc. 2007, 2, 2451–2458. [CrossRef] [PubMed]
54. Davenport, R.J.; Regan, A.C. Synthesis of a C1–C9 fragment of rhizoxin. Tetrahedron Lett. 2000, 41, 7619–7622.
[CrossRef]
55. Tebbe, F.N.; Parshall, G.W.; Reddy, G.S. Olefin Homologation with Titanium Methylene Compounds. J. Am.
Chem. Soc. 1978, 100, 3611–3613. [CrossRef]
56. Okazoe, T.; Hibino, J.-I.; Takai, K.; Nozaki, H. Chemoselective methylenation with a methylenedianion
synthon. Tetrahedron Lett. 1985, 26, 5581–5584. [CrossRef]
57. Takai, K.; Nitta, K.; Utimoto, K. Simple and selective method for aldehydes (RCHO). fwdarw.(E)-haloalkenes
(RCH:CHX) conversion by means of a haloform-chromous chloride system. J. Am. Chem. Soc. 1986, 108,
7408–7410. [CrossRef]
58. Aissa, C. Improved Julia Kocienski Conditions for the Methylenation of Aldehydes and Ketones. J. Org.
−
Chem. 2006, 71, 360–363. [CrossRef]
59. Lebel, H.; Paquet, V. Rhodium-Catalyzed Methylenation of Aldehydes. J. Am. Chem. Soc. 2004, 126, 320–328.
[CrossRef]
60. Lebel, H.; Parmentier, M. Copper-Catalyzed Methylenation Reaction:
Total Synthesis of
(+)-Desoxygaliellalactone. Org. Lett. 2007, 9, 3563–3566. [CrossRef]
61. Li, J.; Menche, D. Selective Deprotection of Silyl Ethers with Sodium Periodate. Synthesis 2009, 1904–1908.
[CrossRef]
62. Blanchette, M.A.; Choy, W.; Davis, J.T.; Essenfeld, A.P.; Masamune, S.; Roush, W.R.; Sakai, T.
Horner-Wadsworth-Emmons reaction: Use of lithium chloride and an amine for base-sensitive compounds.
Tetrahedron Lett. 1984, 25, 2183–2186. [CrossRef]
63. Schrodi, Y.; Pederson, R.L. Evolution and Applications of Second-Generation Ruthenium Olefin Metathesis
Catalysts. Aldrichim. Acta 2007, 40, 45–52.
64. Ung, T.; Hejl, A.; Grubbs, R.H.; Schrodi, Y. Latent Ruthenium Olefin Metathesis Catalysts That Contain an
N-Heterocyclic Carbene Ligand. Organometallics 2004, 23, 5399–5401. [CrossRef]
65. Hoye, T.R.; Jeffrey, C.S.; Tennakoon, M.A.; Wang, J.; Zhao, H. Relay Ring-Closing Metathesis (RRCM):
A Strategy for Directing Metal Movement Throughout Olefin Metathesis Sequences. J. Am. Chem. Soc. 2004
,
126, 10210–10211. [CrossRef]

Molecules 2020, 25, 4527
26 of 27
66. Wang, X.; Bowman, E.J.; Bowman, B.J.; Porco, J.A., Jr. Total Synthesis of the Salicylate Enamide Macrolide
Oximidine III: Application of Relay Ring-Closing Metathesis. Angew. Chem. Int. Ed. 2004, 43, 3601–3604.
[CrossRef]
67. Robinson, A.J.; Elaridi, J.; Van Lierop, B.J.; Mujcinovic, S.; Jackson, W.R. Microwave-assisted RCM for the
synthesis of carbocyclic peptides. J. Pept. Sci. 2007, 13, 280–285. [CrossRef]
68. Voigtritter, K.; Ghorai, S.; Lipshutz, B.H. Rate Enhanced Olefin Cross-Metathesis Reactions: The Copper
Iodide Effect. J. Org. Chem. 2011, 76, 4697–4702. [CrossRef]
69. Fürstner, A.; Mathes, C.; Lehmann, C.W. Mo[N(t-Bu)(Ar)]3 Complexes As Catalyst Precursors: In Situ
Activation and Application to Metathesis Reactions of Alkynes and Diynes. J. Am. Chem. Soc. 1999, 121,
9453–9454. [CrossRef]
70. Brewitz, L.; Llaveria, J.; Yada, A.; Fürstner, A. Formal Total Synthesis of the Algal Toxin ( )-Polycavernoside A.
−
Chem. Eur. J. 2013, 19, 4532–4537. [CrossRef]
71. Chaładaj, W.; Corbet, M.; Fürstner, A. Total Synthesis of Neurymenolide A Based on a Gold-Catalyzed
Synthesis of 4-Hydroxy-2-pyrones. Angew. Chem. Int. Ed. 2012, 51, 6929–6933. [CrossRef] [PubMed]
72. Willwacher, J.; Kausch-Busies, N.; Fürstner, A. Divergent Total Synthesis of the Antimitotic Agent
Leiodermatolide. Angew. Chem. Int. Ed. 2012, 51, 12041–12046. [CrossRef] [PubMed]
73. Lehr, K.; Fürstner, A. An efficient route to the musk odorant (R,Z)-5-muscenone via base-metal-catalysis.
Tetrahedron 2012, 68, 7695–7700. [CrossRef]
74. Hickmann, V.; Kondoh, A.; Gabor, B.; Alcarazo, M.; Fürstner, A. Catalysis-Based and Protecting-Group-Free
Total Syntheses of the Marine Oxylipins Hybridalactone and the Ecklonialactones A, B, and C. J. Am.
Chem. Soc. 2011, 133, 13471–13480. [CrossRef]
75. Benson, S.; Collin, M.-P.; Arlt, A.; Gabor, B.; Goddard, R.; Fürstner, A. Second-Generation Total Synthesis of
Spirastrellolide F Methyl Ester: The Alkyne Route. Angew. Chem. Int. Ed. 2011, 50, 8739–8744. [CrossRef]
76. Lehr, K.; Mariz, R.; Leseurre, L.; Gabor, B.; Fürstner, A. Total Synthesis of Tulearin C. Angew. Chem. Int. Ed.
2011. [CrossRef]
77. Micoine, K.; Fürstner, A. Concise Total Synthesis of the Potent Translation and Cell Migration Inhibitor
Lactimidomycin. J. Am. Chem. Soc. 2010, 132, 14064–14066. [CrossRef]
78. Hickmann, V.; Alcarazo, M.; Fürstner, A. Protecting-Group-Free and Catalysis-Based Total Synthesis of the
Ecklonialactones. J. Am. Chem. Soc. 2010, 132, 11042–11044. [CrossRef]
79. Schaubach, S.; Gebauer, K.; Ungeheuer, F.; Hoffmeister, L.; Ilg, M.K.; Wirtz, C.; Fürstner, A. A Two-Component
Alkyne Metathesis Catalyst System with an Improved Substrate Scope and Functional Group Tolerance:
Development and Applications to Natural Product Synthesis. Chem. Eur. J. 2016, 22, 8494–8507. [CrossRef]
80. Kwon, Y.; Schulthoff, S.; Dao, Q.M.; Wirtz, C.; Fürstner, A. Total Synthesis of Disciformycin A and B:
Unusually Exigent Targets of Biological Significance. Chem. Eur. J. 2018, 24, 109–114. [CrossRef]
81. Mata, G.; Woelfl, B.; Fürstner, A. Synthesis and Molecular Editing of Callyspongiolide, Part 1: The Alkyne
Metathesis/trans-Reduction Strategy. Chem. Eur. J. 2019, 25, 246–254. [CrossRef] [PubMed]
82. Davies, P.W. Ring-closing alkyne metathesis in natural product synthesis. In Metathesis in Natural Product
Synthesis; Cossy, J., Arseniyadis, S., Meyer, C., Eds.; Wiley-VCH Verlag GmbH & Co. KgaA: Weinheim,
Germany, 2010; pp. 205–223.
83. Trost, B.M.; Ball, Z.T. Alkyne Hydrosilylation Catalyzed by a Cationic Ruthenium Complex: Efficient and
General Trans Addition. J. Am. Chem. Soc. 2005, 127, 17644–17655. [CrossRef] [PubMed]
84. Lacombe, F.; Radkowski, K.; Seidel, G.; Fürstner, A. (E)-Cycloalkenes and (E,E)-cycloalkadienes by ring
closing diyne- or enyne-yne metathesis/semi-reduction. Tetrahedron 2004, 60, 7315–7324. [CrossRef]
85. Kleinbeck, F.; Carreira, E.M. Total Synthesis of Bafilomycin A1. Angew. Chem. Int. Ed. 2009, 121, 586–589.
[CrossRef]
86. Gilbert, J.C.; Weerasooriya, U. Diazoethenes: Their Attempted Synthesis from Aldehydes and Aromatic
Ketones by Way of the Horner-Emmons Modification of the Wittig Reaction. A Facile Synthesis of Alkynes1.
J. Org. Chem. 1982, 47, 1837–1845. [CrossRef]
87. Vatèle, J.-M. Yb(OTf)3-Catalyzed Oxidation of Alcohols with Iodosylbenzene Mediated by TEMPO. Synlett
2006, 2055–2058. [CrossRef]
88. Lafontaine, J.A.; Provencal, D.P.; Gardelli, C.; Leahy, J.W. Enantioselective Total Synthesis of the Antitumor
Macrolide Rhizoxin D. J. Org. Chem. 2003, 68, 4215–4234. [CrossRef]

Molecules 2020, 25, 4527
27 of 27
89. Hama, N.; Matsuda, T.; Sato, T.; Chida, N. Total Synthesis of (
−
)-Agelastatin A: The Application of a
Sequential Sigmatropic Rearrangement. Org. Lett. 2009, 11, 2687–2690. [CrossRef]
90. Sternberg, H.W.; Greenfield, H.; Friedel, R.H.; Wotiz, J.H.; Markby, R.; Wender, I. A new type of metallo-organic
complex derived fromdicobalt octacarbonyl and acetylenes. J. Am. Chem. Soc. 1954, 76, 1457–1458. [CrossRef]
91. Greenfield, H.; Sternberg, H.W.; Friedel, R.H.; Wotiz, J.H.; Markby, R.; Wender, I. Acetylenic Dicobalt
Hexacarbonyls. Organometallic Compounds Derived from Alkynes and Dicobalt Octacarbonyl. J. Am.
Chem. Soc. 1956, 78, 120–124. [CrossRef]
92. Yenjai, C.; Isobe, M. One-step recyclization of sugar acetylenes to form medium ether rings via
dicobalthexacarbonyl complexes. Tetrahedron 1998, 54, 2509–2520. [CrossRef]
93. Isobe, M.; Yenjai, C.; Tanaka, S. Medium Size Ether Ring Formation of C-Alkynylated Sugars via
Dicobalthexacarbonyl Complexes. Synlett 1994, 916–918. [CrossRef]
94. Hosokawa, S.; Isobe, M. Reductive decomplexation of biscobalthexacarbonyl acetylenes into olefins.
Tetrahedron Lett. 1998, 39, 2609–2612. [CrossRef]
95. Takai, S.; Ploypradith, P.; Hamajima, A.; Kira, K.; Isobe, M. Sodium Hypophosphite Decomplexation of
Acetylenebiscobalthexacarbonyls to cis-Olefins. Synlett 2002, 588–592. [CrossRef]
96. Chong, Y.K.; Moad, G.; Rizzardo, E.; Thang, S.H. Thiocarbonylthio End Group Removal from
RAFT-Synthesized Polymers by Radical-Induced Reduction. Macromolecules 2007, 40, 4446–4455. [CrossRef]
97. Jang, D.O.; Cho, D.H.; Kim, J. Diphenylphosphine Oxide: An Alternative to Organotin Hydrides in the
Radical Deoxygenation of Alcohols. Synth. Commun. 1998, 28, 3559–3565. [CrossRef]
98. Kende, A.S.; Blass, B.E.; Henry, J.R. Enantioselective total synthesis of didesepoxyrhizoxin. Tetrahedron Lett.
1995, 36, 4741–4744. [CrossRef]
99. Kusebauch, B.; Scherlach, K.; Kirchner, H.; Dahse, H.-M.; Hertweck, C. Antiproliferative Effects of Ester- and
Amide-Functionalized Rhizoxin Derivatives. Chemmedchem. 2011, 6, 1998–2001. [CrossRef]
100. Stella, V.J.; Umprayn, K.; Waugh, W.N. Development of parenteral formulations of experimental cytotoxic
agents. I. Rhizoxin (NSC-332598). Int. J. Pharm. 1988, 43, 191–199. [CrossRef]
Sample Availability: Samples of the compounds are not available from the authors.
©
2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).