Total Synthesis and Configurational Assignment of AscospiroketalA

Article abstract of DOI:10.1002/chem.201502754

The total synthesis of the marine fungus-derived natural product ascospiroketal is described. This concise synthesis relies on a unique Ag^I-promoted tandem cascade cyclization that provides direct access to the correctly configured tricyclic core of the natural product from a linear precursor. The synthesis of candidate stereostructures of ascospiroketalA allowed for the confident assignment of both the relative and absolute stereochemistry of this unusual octaketide.

Full text of DOI:10.1002/chem.201502754

DOI: 10.1002/chem.201502754
Full Paper
&
Natural Products
Total Synthesis and Configurational Assignment of
Ascospiroketal A
[
a]
Stanley Chang, Soo Hur, and Robert Britton*
Abstract: The total synthesis of the marine fungus-derived
cursor. The synthesis of candidate stereostructures of asco-
spiroketal A allowed for the confident assignment of both
the relative and absolute stereochemistry of this unusual oc-
taketide.
natural product ascospiroketal is described. This concise syn-
I
thesis relies on a unique Ag -promoted tandem cascade cyc-
lization that provides direct access to the correctly config-
ured tricyclic core of the natural product from a linear pre-
Introduction
Ascochyta salicorniae is a marine-derived fungus well known to
vide a full account of these efforts including the optimization
of the key silver(I)-promoted cyclization cascade.
produce unusual polyketide secondary metabolites including
[
1,2]
ascosalipyrrolidinone A, ascolactone A, and ascochitine.
In
2
007, following a systematic investigation of cultivation param-
eters, Kçnig and co-workers isolated two structurally unprece-
dented polyketides from A. salicorniae: ascospiroketals A (1)
[3]
and B (2) (Figure 1). Extensive analysis of their 1D and 2D
NMR spectroscopic data revealed that 1 and 2 both contain
a rare octaketide tricyclic core, and the quaternary stereogenic
center at C2 was postulated to arise from geminal biomethyla-
[4]
tion by S-adenosyl methionine (SAM). Biosynthetically, 1 and
were proposed to derive from the direct cyclization of the
C2-methylated polyketide 3 via distinct pathways that involve
2
[
3]
formation of a tricyclic tetrahydrofuran 4 or lactone 5. The
relative stereochemistry within the tricyclic core of the asco-
spiroketals was established following extensive analysis of 1D
NOESY spectra. Unfortunately, it was not possible to assign the
relative stereochemistry of the side chain in these molecules or
[3]
the absolute stereochemistry of their tricyclic cores. Structur-
ally, the ascospiroketals are related to a small family of 5,5-spi-
roacetal-containing fungal metabolites, several of which dem-
onstrate potentially useful biological activities. For example,
[5]
[6]
members of the cephalosporolides and penisporolides
share the common 5,5-spiroacetal-lactone motif, and both
cephalosporolides H (8) and I (9) display anti-inflammatory acti-
[5d]
vity. As part of our continued efforts to explore the utility of
chlorohydrins as building blocks for natural product synthesis,
we recently communicated the total synthesis of ascospiroke-
tal A (1), including the assignment of both relative and abso-
[7]
lute stereochemistry for this natural product. Here we pro-
[
a] S. Chang, S. Hur, Prof. R. Britton
Department of Chemistry, Simon Fraser University
Burnaby, British Columbia, V5A 1S6 (Canada)
E-mail: rbritton@sfu.ca
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201502754.
Figure 1. Biosynthesis of ascospiroketals A and B, and other 5,5-spiroacetal-
containing fungal metabolites.
Chem. Eur. J. 2015, 21, 16646 – 16653
16646
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Results and Discussion
We recently reported a convenient strategy for the synthesis of
I
spiroacetals that involves the Ag -promoted cyclization of keto-
[8]
chlorohydrins (e.g., 12!14, Scheme 1). This process was also
demonstrated in a short synthesis of cephalosporolides E (6)
and F (7), which simply involved oxidative cleavage of the
alkene function in 14 and further oxidation of the resulting
[
8]
lactol. Considering that the tricyclic core of the cephalosporo-
lides contains a 5,5-spiroacetal similar to that embedded in the
ascospiroketals (Figure 1), these studies also provided a tem-
plate for the synthesis of ascospiroketals A and B. In contem-
plating the application of this chlorohydrin-based spiroacetal
synthesis to the preparation of the ascospiroketals, it was en-
visaged that the side chain 15 would be coupled to the fully
[9]
functionalized tricyclic core 16 through a Suzuki or Sonoga-
[
10]
shira cross-coupling reaction. As the stereocenters at C15,
C2’, and C3’ were undefined, this late-stage coupling would
allow for rapid preparation of configurational isomers at these
centers. While the sequential alkylation and carbonylation of
enolates derived from cephalosporolide-like lactones (e.g., 6
and 7) was originally explored as a means to access the all-
carbon C2 quaternary stereocenter in the ascospiroketals, at-
tempts to effect this sequence of reactions on the related spi-
rocyclic lactone 6 and 7 proved unsuccessful and were compli-
[
11]
cated by elimination/hetero-Michael addition processes. As
an alternative and potentially more convergent strategy, direct
incorporation of a prochiral quaternary center in the chloroal-
dehyde (e.g., 21a–c) would enable the all-carbon quaternary
stereocenter found at C2 in the ascospiroketals to be intro-
duced at a late stage in the synthesis where substrate bias
may potentially be exploited for stereodifferentiation. Thus,
I
Scheme 1. Application of a Ag -promoted spiroacetal synthesis in the syn-
thesis of cephalosporolides E and F, and a retrosynthetic analysis for asco-
spiroketal A. Tf=trifluoro-methanesulfonyl, DMF=N,N-dimethylformamide.
[12]
several chloroaldehydes were conceived that included the
di-ester 21a, the di-nitrile 21b, and the oxetane 21c. Impor-
tantly, early incorporation of the C2 quaternary center would
also improve the stability of spiroacetal intermediates (e.g., 17)
by preventing b-elimination processes. Furthermore, consider-
ing the propensity of oxetanes to undergo Lewis-acid cata-
[13]
lyzed rearrangement, the oxetanyl aldehyde 21c was a par-
I
ticularly appealing substrate as the Ag salts required for spi-
[
8]
roacetal formation (e.g., 19!18) may also promote oxetane
to tetrahydrofuran rearrangement (e.g., 18c!17b).
As depicted in Scheme 2, we first prepared the di-ester and
di-nitrile aldehydes 24 and 27, as well as the oxetanyl alde-
hyde 30 following straightforward sequences of reactions.
Thus, treatment of diethyl methylmalonate 22 with NaH fol-
lowed by allyl bromide afforded 23. Ozonolytic cleavage of the
terminal alkene function followed by reductive workup provid-
Scheme 2. Synthesis of aldehydes 24, 27, and 30. a) NaH, THF, 08C, then
allyl bromide: 84%; b) O , CH Cl , À788C, then PPh : 63%; c) bromoacetalde-
[14]
ed the desired aldehyde 24 in good yield. The di-nitrile 27
3
2
2
3
hyde diethyl acetal, K
2
3
CO , DMF, 508C: 63%; d) NaH, THF, then iodomethane:
was accessed from the sequential alkylation of malonitrile with
[
15]
92%; e) 2m HCl, THF/H
2
O (1:1): 69%; f) MsCl, Et
Cl
3 2 2
N, CH Cl , 08C; g) KCN, DMF,
bromoacetaldehyde diethyl acetal and methyl iodide. Hydro-
lytic removal of the diethyl acetal protecting group afforded al-
dehyde 27. Finally, the oxetanyl aldehyde 30 was prepared fol-
lowing mesylation of 3-methyl-3-oxetanemethanol 28 and se-
quential displacement of the mesylate with cyanide and DIBAL
reduction of the resulting nitrile.
8
08C: 76% over two steps; h) DIBAL, CH
2
2
, À58C: 73%. THF=tetrahydro-
furan, MsCl=methanesulfonyl chloride, DIBAL=diisobutylaluminum hy-
dride.
With the unique aldehydes 24, 27, and 30 in hand, we next
explored their compatibility with the a-chlorination procedures
[
16]
[17]
[18]
reported by MacMillan, Jørgensen, and Christmann. As
Chem. Eur. J. 2015, 21, 16646 – 16653
www.chemeurj.org
16647 ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
summarized in Table 1, chlorination of the di-ester 24 follow-
ing Jørgensen’s procedure using l-proline led to the racemic
chloroaldehyde 21a in good yield (entry 1). Surprisingly, the a-
chlorination of 27 proved challenging and after an exhaustive
screen of organocatalysts and chlorinating agents, we failed to
identify a satisfactory chlorination procedure for this aldehyde
formation of this major product involves the initial production
of the desired ketochlorohydrin 38, followed by its rapid cycli-
zation to the corresponding g-butyrolactone 39. A sequential
b-elimination and decarboxylation/chloride elimination then
provides diene 40. Fortunately, these side reactions could be
avoided by use of the corresponding oxetanyl chloroaldehyde
21c, which underwent smooth lithium aldol reaction with the
enolate derived from methyl ketone 41 to provide ketochloro-
hydrin 42 in modest yield (49%). The silyl protecting group in
ketochlorohydrin 42 was readily removed by treatment with
TBAF to provide the key spiroacetalization substrate 43. With
[17]
(
(
entries 2–6). For example, l-proline catalysis
entry 2) led to a mixture of the desired a-chloroaldehyde 21b
with NCS
and the corresponding a,a-dichloroaldehyde (not shown)
along with recovered starting material. While increasing the re-
action time resulted in complete consumption of the chlorinat-
ing agent (NCS), there was no practical improvement in prod-
uct ratios (entry 3). Employing the chlorination protocols de-
I
this material in hand, a series of Ag -promoted spirocyclizations
[
8]
were examined using our previously optimized conditions.
[16]
scribed by MacMillan on this substrate produced none of
the desired a-chloroaldehyde 21b, even after prolonged stir-
ring at room temperature (entries 5 and 6). These unfortunate
results prompted us to abandon further attempts to produce
the di-nitrile chloroaldehyde 21b. As indicated in entry 7,
chlorination of the oxetanyl aldehyde 30 proceeded smoothly
using Jørgensen’s a-chlorination procedure with l-proline as
the organocatalyst. Notably, aldehyde 21c proved to be unsta-
ble even when stored in solution in a freezer (in THF, stored at
À58C), and was consequently used in subsequent reactions di-
rectly following its preparation (vide infra).
Pleasingly, when ketodiol 43 was treated with 2 equivalents of
AgOTf/Ag O and sonicated for four days, a mixture of the de-
2
sired spiroacetals 44 and 45 were produced in a combined
58% yield.
At this stage, it became crucial to explore the oxetane to tet-
rahydrofuran rearrangement on this model system. It was ex-
pected that this rearrangement would be facile considering
that the ring strain in an oxetane is similar to that present in
epoxides (~25 kcalmolÀ1).[19] Surprisingly, while intermolecular
[
20]
ring opening reactions of oxetanes are common, few exam-
ples of intramolecular oxetane opening reactions have been
[
13]
As detailed in Scheme 3, the aldol reaction of chloroalde-
hyde 21a with the lithium enolate derived from methyl ketone
reported outside the taxol literature. Notably, Bach has dem-
onstrated that the opening of hydroxyalkyl oxetanes proceeds
under basic conditions to afford ring expanded ethers (e.g.,
37 afforded none of the desired ketochlorohydrin. Instead, the
dienoate 40 was isolated in low yield from the crude reaction
mixture as the major product. A possible mechanism for the
Table 1. Organocatalytic a-chlorination of aldehydes 24, 27, and 30.
[
a]
Entry
Aldehyde
Conditions
Product
Yield [%]
63
1
2
3
4
5
6
7
24
27
27
27
27
27
30
A
A
A
B
C
D
A
21a
21b
21b
21b
21b
21b
21c
[b]
n.d.
n.d.
0
[c]
0
0
73
[
a] Conditions: A) 36, l-proline, CH
5, acetone; D) 34, LiCl, Cu(TFA)
day the ratio of 27/21b/a,a-dichloroaldehyde=5.8:1.7:1. [c] After
days the ratio of 27/21b/a,a-dichloroaldehyde=1.5:1.2:1; n.d.= not
2
Cl
2
; B) 36, l-prolinamide, CH
2 2
Cl ; C) 33,
Scheme 3. a) LDA, THF, À788C, then (Æ)-21a: 26%; b) LDA, THF, À788C,
3
1
6
2
, Na
2
S
2
O
8
, H O, acetonitrile. [b] After
2
then (Æ)-21c: 49%; c) TBAF, THF, 08C: 75%; d) AgOTf, Ag
2
O, acetone: 58%;
e) (salen)CoOTf, MeCN: 98% (46/47=3:2); f) AgBF
days: 66%. LDA=lithium diisopropylamide, TBAF=tetrabutylammonium
fluoride.
4 2
, Ag O, THF, 20–508C,
2
determined.
Chem. Eur. J. 2015, 21, 16646 – 16653
www.chemeurj.org
16648
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
21]
[29]
tetrahydrofurans), and Jacobsen has reported enantioselec-
bration process. Table 2 summarizes our attempts to effect
epimerization of the spiroacetal center in 46 using this strategy
(starting ratio 46/47=1:1). As indicated in entries 1 and 2, sev-
eral protic acids failed to induce equilibration or led to the for-
III
tive Co -catalyzed oxetane to tetrahydrofuran rearrange-
[
22]
ments. We were pleased then to find that treatment of hy-
droxy oxetane 44 with a catalytic amount of racemic (salen)-
[
22]
CoOTf produced the diastereomeric spiroacetals 46 and 47
in excellent yield. The tricycles were separated by flash chro-
matography and independently analyzed by NOESY experi-
mation of furan byproducts (entry 2). Addition of BF ·OEt2
3
(entry 3) to a mixture of spiroacetal epimers was effective in al-
tering the ratio of these compounds in favor of the desired
epimer 47, while TMSOTf (entry 4) promoted decomposition
and other Lewis acids (e.g., Sc(OTf) , Yb(OTf) , InCl , Ti(OiPr) )
ments in [D ]benzene, which confirmed the relative stereo-
6
chemistry at the newly formed all-carbon quaternary center as
shown. Importantly, both spiroacetals 46 and 47 possess the
relative configuration at C2 required for ascospiroketal A (1).
3
3
3
4
failed to effect anomerization. Fortunately, treatment of the
mixture of spiroacetals with ZnCl (entries 9 and 10) provided
2
1
Furthermore, re-inspection of H NMR spectra recorded on the
a mixture of spiroacetals favoring 47 in ratios of up to approxi-
I
crude Ag-promoted spirocyclization reaction products indicat-
mately 2:1. While this ratio is modest when compared to those
[
30]
[28]
ed that the tricycles 46 and 47 were indeed present as minor
products of this reaction. To evaluate whether the Ag-promot-
reported by Kishi
and Dudley
for related systems, the
I
equilibration provided an avenue for the recycling of 46 that
would ultimately find use in the total synthesis of ascospiroke-
tal A (vide infra).
ed spirocyclization/oxetane rearrangement could be further
I
[23]
optimized, we screened several Ag salt/base combinations.
We were delighted to find that when the ketodiol 43 was
treated with a combination of AgBF and Ag O (508C for 2
Table 2. Anomerization of tricycles 46 and 47.
4
2
days), the desired tricycles 46 and 47 were produced in a com-
bined isolated yield of 66% (46/47=3:2). Remarkably, this cas-
cade cyclization generates two new stereocenters (C2 and C6)
with complete control of the C2 all-carbon quaternary center
I
[a]
and represents a unique example of a Ag-promoted oxetane
Entry
Conditions
Acid
46/47
opening reaction (Scheme 4). The observed stereoselectivity
from the opening of the prochiral oxetane was attributed to
the preference for a transition structure involving bidentate
chelation between the central ring oxygen and that of the oxe-
1
2
3
4
5
6
7
8
9
A
A
B
C
B
C
C
B
A
A
PPTS
CSA
1.3:1
[
b]
1.7:1
1:1.4
BF
TMSOTf
Sc(OTf)
3
·OEt
2
[b]
N/A
1:1
I
I
3
tane to the Ag center. Notably, bidentate chelation of Ag salts
Yb(OTf)
InCl3
3
1.3:1
1.3:1
1.2:1
[24]
has previously been exploited in aldol reactions and [3+2]
[
25]
cycloadditions. The pro-S transition structure 49 does not si-
Ti(OiPr)
4
[
b]
I
ZnCl
ZnCl
2
2
1:1.9
1:1.6
multaneously allow for a bidentate chelation with Ag and re-
[
b,c]
1
0
quired orbital alignment.
[
a] Conditions: A) RT, 20 h; B) À788C, 30 min, C) À408C, 30 min.[b] De-
The formation of the tricyclic spiroacetals 46 and 47 as a 3:2
diastereomeric mixture is consistent with previous syntheses of
composition to furan also observed. [c] An excess of MgO was added. N/
A=not applicable.
[
26]
5
,5-spiroacetals as anomeric mixtures.
However, seminal
[
27]
[28]
studies by SolladiØ and Dudley provided inspiration for re-
cycling the undesired spiroacetal epimer 46 through an equili-
Having established a viable route to the tricyclic core of as-
cospiroketal A, our attention turned to addressing the stereo-
chemical uncertainties within the C12-C5’ side chain of the nat-
ural product. Preparation of a candidate stereostructure for
C12-C5’ side chain involved coupling of TBS-protected (2R,3R)-
[
31]
3
-hydroxy-2-methylbutyric acid (54) with homoallylic alcohol
[
32]
5
3
derived from TMS-acetylene addition to (À)-propylene
[33]
oxide, followed by DIBAL reduction of the alkyne function.
While esterification of the homoallylic alcohol 53 with acid 54
failed using standard DIC or DCC coupling reagents, a Yama-
[
34]
[
35]
guchi esterification eventually afforded the desired TBS-pro-
tected ester 55 in good yield (Scheme 5). At this point at-
tempts were made to convert the vinyl silane function in 55
into a vinyl boronic acid in preparation for an eventual Suzuki
coupling with a suitably functionalized tricyclic core of asco-
spiroketal A. Unfortunately, under a variety of conditions (e.g.,
[
36]
BCl , BBr ) only protodesilylation was observed and our in-
3
3
ability to effect a direct Si!B transformation prompted us to
consider an end game Sonogashira cross-coupling strategy.
Thus, the homopropargyl alcohol 56 was also esterified with
I
Scheme 4. Proposed TS model for Ag -promoted oxetane rearrangement.
a) AgBF
4
, Ag
2
O, THF, 20–508C, 2 days: 66%.
Chem. Eur. J. 2015, 21, 16646 – 16653
www.chemeurj.org
16649
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Plots depicting the difference between the chemical shift of select-
1
ed resonances in the H NMR spectra of: a) 57, b) 58, c) 59, d) 60
(
[D
6
]acetone, 400 MHz) and the equivalent resonances reported for ascospir-
[3]
oketal A (1) ([D
6
]acetone, 500 MHz).
Scheme 5. a) nBuLi, Et
2
O, À788C, then (S)-propylene oxide, BF
, 408C: 68%; c) 54, 2,4,6-trichlorobenzoyl chloride, Et
: 63%; d) DIC, DMAP, CH Cl : 96%; e) TBAF, THF: 83%.
3
·OEt
2
: 69%;
by Sato documenting the kinetic resolution of related vinylsi-
[
39]
b) DIBAL, CH
DMAP, CH Cl
2
Cl
2
3
N,
lanes, it was expected that the S configured allylic alcohol
2
2
2
2
6
sis
3 would be recovered from this reaction. Mosher ester analy-
TMS=trimethylsilyl, DMAP=4-dimethylaminopyridine, DIC=N,N’-diisopropyl-
carbodiimide.
[
40]
(see the Supporting Information) ultimately supported
this assignment. The enantiomeric excess of 63 was deter-
mined to be 98% following its conversion into the correspond-
ing benzoyl ester and analysis by chiral HPLC. Completion of
the synthesis of methyl ketone 64 then involved removal of
the acetal protecting group and direct protection of the secon-
dary alcohol as the corresponding trimethylsilyl ether. This six-
step process starting from commercial starting materials was
suitable for the production of gram quantities of the methyl
ketone 64.
(
2S,3R)-3-hydroxy-2-methylbutyric acid (51). Removal of the
two silyl protecting groups in 56 provided the corresponding
hydroxy-ester 57 in good yield.
With a viable synthetic route to the linear fragment 57 es-
tablished, the remaining three potential diastereomeric C12-
C5’ side chains were prepared following the aforementioned
reaction sequence but starting with (+)-propylene oxide and/
or (2S,3S)-3-hydroxy-2-methylbutyric acid. The spectral data de-
rived from the four candidate C12-C5’ side chains are summar-
ized in Figure 2 as difference plots compared to the equivalent
data reported for the corresponding segment of ascospiroketa-
The preparation of a-chloroaldehyde (+)-21c required enan-
tioselective chlorination of the aldehyde 30, prepared previ-
ously as part of our model studies (Scheme 6). Employing the
[3]
[16]
l A. From this analysis, we proposed that the natural product
possesses the 2’S*,3’R* relative configuration that is common
to esters 57 and 59. Unfortunately, we were unable to confi-
dently assign the relative configuration at C15 using this
method and, consequently, it would ultimately be necessary to
synthesize four potential candidate stereostructures for asco-
spiroketal A that incorporate the side chains 57, 59, and their
enantiomers (i.e., ent-57 and ent-59). It was hoped that the
comparison of the spectral data derived from these four candi-
date stereostructures for ascospiroketal A with that of the nat-
ural product would allow for confident assignment of the
global stereochemistry of the natural product.
a-chlorination conditions reported by MacMillan,
Jørgen-
[
17]
[18]
sen, and Christmann with this unusual substrate all result-
ed in the formation of the desired a-chloroaldehyde (+)-21c.
However, considering the operational simplicity of Christ-
mann’s procedure, a combination of MacMillan’s catalyst 34
and NCS was used to effect formation of the unstable chloroal-
dehyde (+)-21c, which was purified by trituration of the crude
reaction mixture with Et O. The configuration of the chlorome-
2
thine center was tentatively assigned based on analogous
chlorinations reported by Christmann, and the enantiopurity of
(+)-21c (85% ee) was determined by chiral HPLC following re-
duction of the aldehyde (NaBH ) to the chlorohdyrin and sub-
4
Considering that ascospiroketal A bears considerable struc-
tural similarities to the fungal metabolite cephalosporolides H
sequent benzoylation (not shown).
With an efficient synthesis of both aldol coupling partners
secured, the lithium enolate derived from methyl ketone 64
(Scheme 7) was treated with a-chloroaldehyde (+)-21c to pro-
vide the aldol adduct 65 in good yield and diastereoselectivity.
Removal of the trimethylsilyl ether moiety of 65 was effected
by treatment with a catalytic amount of PTSA in acetone/water
in advance of the key spirocyclization reaction. In the event,
the spirocyclization was facilitated by heating the ketochloro-
[
5d]
and I (Figure 1), we chose to target the 2R,3R,4R,6R,9S enan-
tiomer of the tricyclic core. Toward this goal, Scheme 6 depicts
the enantioselective synthesis of the methyl ketone 64, which
commenced with the addition of vinyl lithium to the known al-
[
37]
dehyde 61. Resolution of the resulting racemic allylic alco-
hols 63 was effected by Sharpless asymmetric epoxidation
[38]
using (À)-diisopropyl d-tartrate. Based on a previous report
Chem. Eur. J. 2015, 21, 16646 – 16653
www.chemeurj.org
16650
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 7. a) LDA, THF, À788C, then (+)-21c: 46% (72% based on recovered
Scheme 6. a) Et
À208C: 40%; c) PTSA, H
two steps; e) DIBAL, CH
over two steps. DIPT=diisopropyl tartrate, TBHP=tert-butyl hydroperoxide,
PTSA=p-toluenesulfonic acid monohydrate, TMSCl=chlorotrimethylsilane,
NCS=N-chlorosuccinimide.
2
O, À408C, then 62: 80%; b) Ti(OiPr)
O/acetone (1:10); d) TMSCl, Et
Cl
4
, (À)-DIPT, TBHP, CH
N, THF: 91% over
Cl , 58C: 64%
2 2
Cl ,
starting material, d.r.=12:1); b) PTSA, H
2
O/acetone (1:10): 83%; c) AgBF
4
,
2
3
Ag
O, THF, 20–508C, approximately 3 days: 82% (67/68 ~1:1); d) ZnCl
Cl : 67/68=2:1.
2
,
2
2
2
, À58C; f) 34 (20 mol%), NCS, CH
2
2
MgO, CH
2
2
hydrin 66 with a combination of AgBF and Ag O in THF over
4
2
three days. We were delighted to find that these conditions
delivered the tricyclic spiroacetals 67 and 68 in good com-
1
bined yield. Notably, the H NMR spectra derived from these
spiroacetals and those prepared as part of the model study
(
e.g., 46 and 47, Scheme 3) were closely matched in both
chemical shift and coupling constants for the resonances as-
signed to protons on rings B and C. Thus, the stereochemistry
of these two new tricyclic spiroacetals was assigned by analo-
I
gy, and this remarkably efficient Ag-promoted tandem spiro-
cyclization/oxetane rearrangement process secured both the
densely functionalized core and C2 quaternary all-carbon
center of ascospiroketal A. Notably, the undesired spiroacetal
6
7 was readily epimerized using Dudley’s conditions (ZnCl ,
2
[
28]
MgO) as a means to recycle the anomeric spiroacetal (67/
Scheme 8. a) DMP, NaHCO
3
, CH
2
Cl
2
3 2 2
: 67%; b) DMP, NaHCO , CH Cl : 79%;
6
8=2:1).
2 2 4 2 2
c) NaClO , NaH PO ·H O, 2-methyl-2-butene, H O/tBuOH (1:2.4): 87%; d) NIS,
HFIP, 08C: 67%. DMP=Dess–Martin periodinane, NIS=N-iodosuccinimide,
HFIP=1,1,1,3,3,3-hexafluoro-2-propanol.
At this point several modifications to tricyclic core 68 were
required. Specifically, the C1 hydroxymethylene function re-
quired oxidation to the corresponding carboxylic acid, and the
vinyl silane moiety would necessarily be converted into a vinyl
halide in preparation for a Sonogashira coupling. Toward this
end, we first probed the proposed reactions on the C6 epimer
As illustrated in Scheme 9, completion of the synthesis of
the first candidate stereostructure for ascospiroketal A involved
Sonogashira coupling between the side chain 57 and the vinyl
iodide 72. The resulting enyne (not shown) was subjected di-
rectly to a Lindlar reduction to provide the target molecule 73.
It is noteworthy that this final transformation required careful
monitoring, as prolonged reaction times resulted in concomi-
tant epimerization of the spiroacetal. Having successfully iden-
tified a synthetic route to a candidate stereostructure of asco-
spiroketal A, these final two reactions were repeated with the
stereochemically unique side chain fragments 59, ent-57, and
ent-59, providing the full ensemble of candidate stereostruc-
(
i.e., tricycle 67) where the primary alcohol was effectively oxi-
[41]
dized with Dess–Martin periodinane to provide aldehyde 69
Scheme 8). Repeating the same reaction on tricycle 68 with
(
an excess of NaHCO to scavenge residual acetic acid smoothly
3
provided aldehyde 70 without epimerization at the spiroacetal
center. A subsequent Pinnick oxidation afforded the desired
carboxylic acid from which suitable crystals were obtained by
[42]
slow evaporation from a mixture of Et O/pentane and allowed
2
confident assignment of the structure 71 by X-ray crystallo-
graphic analysis. Importantly, the H NMR spectral data derived
1
1
13
tures 73–76. Unfortunately, the H and C NMR spectra record-
ed on these materials (600 MHz) were similar and differed little
from the data reported for the natural product. Ultimately, dif-
ferentiation of the NMR spectra recorded on the candidate ste-
reostructures 73–76 from that of the natural product required
their direct comparison with the authentic spectral data (gen-
from 71 were also in close agreement with that reported for
[
3]
ascospiroketal A, providing preliminary support for the as-
signed stereochemistry of the natural product. Finally, the vinyl
silane 71 was converted into the corresponding vinyl iodide
[43]
72 using Zakarian’s conditions.
Chem. Eur. J. 2015, 21, 16646 – 16653
www.chemeurj.org
16651
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
2 3 2 2
Scheme 9. a) Compound 57, PdCl (PPh ) , CuI, DIPA, THF; b) H (balloon), Lin-
dlar catalyst, quinoline, EtOH: 46% over two steps. DIPA=diisopropylamine.
erously provided to us by Professor Kçnig) derived from the
natural product (500 MHz, [D ]acetone).
6
Following this subsequent analysis, resonances assigned to
the diastereotopic protons at C14 were clearly unique and dis-
tinguished the C15(S) and C15(R) stereostructures (Figure 3). In
1
particular, only the H NMR spectra derived from the C15(S) dia-
stereomers 73 and 76 contained H14a/H14b resonances char-
acteristic of ascospiroketal A. As illustrated in Figure 4, further
1
analysis of the H NMR spectra derived from the C15(S) diaste-
reomers 73 and 76 highlighted several subtle differences be-
tween the resonances assigned to H10, H11, H13, H14, and H3’
in 76 and the equivalent resonances in the spectrum of the
natural product (jDdj >0.01 ppm). Thus, based on this analysis,
Figure 4. Plots depicting the difference in Hz between the chemical shift of
1
selected resonances in the H NMR spectra of: a) 73, b) 74, c) 75, d) 76
(
[D
6
]acetone, 500 MHz) and the equivalent resonances reported for ascospir-
[3]
oketal A (1) ([D
6
]acetone, 500 MHz).
the relative configuration of the natural product was assigned
as that depicted for 76. Additionally, the sign of the optical ro-
20
tation measured on 76 ([a] = +5 (c 0.20 in MeOH)) matched
D
2
D
0
that reported for the natural product ([a] = +20 (c 0.45 in
MeOH)), confirming the absolute stereochemistry of ascospiro-
ketal A as shown for 76, which is also consistent with that of
the related tricyclic spiroacetal cephalosporolide I (Figure 1).
Conclusion
In summary, several synthetic approaches to the octaketide
Figure 3. Comparison of the resonances for protons at C14 in the candidate
stereostructures 73–76 and those of the natural product (500 MHz,
natural product (+)-ascospiroketal A were explored, ultimately
I
[
D
6
]acetone).
leading to the development of a tandem Ag-promoted spiro-
Chem. Eur. J. 2015, 21, 16646 – 16653
www.chemeurj.org
16652
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
cyclization/oxetane-rearrangement reaction that secured the
tricyclic core of the natural product as well as the all carbon
quaternary stereocenter at C2. This efficient, 14-step synthesis
allowed for the rapid production of four candidate stereostruc-
tures of the natural product. The ultimate assignment of the
relative and absolute stereochemistry of the natural product as
[16] a) M. P. Brochu, S. P. Brown, D. W. C. MacMillan, J. Am. Chem. Soc. 2004,
1
26, 4108–4109; b) M. Amatore, T. D. Beeson, S. P. Brown, D. W. C. Mac-
Millan, Angew. Chem. Int. Ed. 2009, 48, 5121–5124; Angew. Chem. 2009,
21, 5223–5226.
1
[17] a) N. Halland, A. Braunton, S. Bachmann, M. Marigo, K. A. Jørgensen, J.
Am. Chem. Soc. 2004, 126, 4790–4791; b) N. Halland, M. A. Lie, A. Kjærs-
gaard, M. Marigo, B. Schiøtt, K. A. Jørgensen, Chem. Eur. J. 2005, 11,
7083–7090.
2
R,3R,4R,6R,9S,15S,2’R,3’R relied on close inspection of NMR
[18] P. Winter, J. Swatschek, M. Willot, L. Radtke, T. Olbrisch, A. Schäfer, M.
spectra recorded on these synthetic candidate stereostructures
and comparison with that of ascospiroketal A. Considering the
biosynthetic and structural similarities between ascospiroketal-
s A (1) and B (2), the results from this study also provide in-
sight into the relative and absolute stereochemistry of asco-
spiroketal B.
Christmann, Chem. Commun. 2011, 47, 12200–12202.
[19] H. K. Eigenmann, D. M. Golden, S. W. Benson, J. Phys. Chem. 1973, 77,
687–1691.
20] For early examples, see: a) J. C. Mullis, W. P. Weber, J. Org. Chem. 1982,
7, 2873–2875; b) S. A. Carr, W. P. Weber, J. Org. Chem. 1985, 50, 2782–
1
[
4
2785; c) M. Yamaguchi, Y. Nobayashi, I. Hirao, Tetrahedron Lett. 1983, 24,
5121–5122; d) M. Yamaguchi, Y. Nobayashi, I. Hirao, Tetrahedron 1984,
4
0, 4261–4266.
[
[
[
21] T. Bach, K. Kather, O. Krämer, J. Org. Chem. 1998, 63, 1910–1918.
22] R. N. Loy, E. N. Jacobsen, J. Am. Chem. Soc. 2009, 131, 2786–2787.
23] For example, sonication of a solution of ketodiol 43 at 408C with
Acknowledgements
2
2
equiv of both AgOTf and Ag O produced an approximately 7:1 mix-
This work was supported by an NSERC Discovery Grant and
a Michael Smith Foundation for Health Research Career Investi-
gator Award to R.B. and an NSERC PGSD to S.C.
ture of bicyclic/tricyclic spiroacetals (44/45/46/47) in 43% yield after
41 h or 58% yield after 96 h.
[
[
24] A. Yanagisawa, Y. Matsumoto, K. Asakawa, H. Yamamoto, Tetrahedron
2
002, 58, 8331–8339.
25] A. M. Szpilman, E. M. Carreira, Cycloaddition Reactions, in Silver in Organ-
ic Chemistry (Ed.: M.Harmata), Wiley, Hoboken, 2010, pp. 46–73 and ref-
erences therein.
Keywords: ascospiroketal A
·
chlorohydrins
·
natural
products · spiroacetals · total synthesis
[
26] For previous total syntheses of cephalosporolides E and F, see: a) C.
Ramana, S. Suryawanshi, R. Gonnade, J. Org. Chem. 2009, 74, 2842–
[
[
1] C. Osterhage, R. Kaminsky, G. M. Kçnig, A. D. Wright, J. Org. Chem. 2000,
5, 6412–6417.
2] S. F. Seibert, E. Eguereva, A. Krick, S. Kehraus, E. Voloshina, G. Raabe, J.
Fleischhauer, E. Leistner, M. Wiese, H. Prinz, K. Alexandrov, P. Janning, H.
Waldmann, G. M. Kçnig, Org. Biomol. Chem. 2006, 4, 2233–2240.
2845; b) R. Fernandes, A. Ingle, Synlett 2010, 158–160; c) M. Brimble,
6
O. C. Finch, A. M. Heapy, J. D. Fraser, D. P. Furkert, P. D. O’Connor, Tetra-
hedron Lett. 2011, 67, 995–1001; d) L. Song, Y. Liu, R. Tong, Org. Lett.
2
013, 15, 5850–5853; e) C. N. Kona, C. V. Ramana, Tetrahedron 2014, 70,
653–3656.
27] G. Solladie, N. Huser, J. Fischer, A. Decian, J. Org. Chem. 1995, 60, 4988–
990.
3
[
[
3] S. F. Seibert, A. Krick, E. Eguereva, S. Kehraus, G. M. Kçnig, Org. Lett.
[
[
2007, 9, 239–242.
4
4] For other examples of polyketide geminal biomethylation via SAM, see:
a) A. M. Gehring, E. DeMoll, J. D. Fetherston, I. Mori, G. F. Mayhew, F. R.
Blattner, C. T. Walsh, R. D. Perry, Chem. Biol. 1998, 5, 573–586; b) D. A.
Miller, L. Luo, N. Hillson, T. A. Keating, C. T. Walsh, Chem. Biol. 2002, 9,
28] a) S. F. Tlais, G. B. Dudley, Org. Lett. 2010, 12, 4698–4701; b) S. F. Tlais,
G. B. Dudley, Beilstein J. Org. Chem. 2011, 7, 570–577; c) S. F. Tlais, G. B.
Dudley, Beilstein J. Org. Chem. 2012, 8, 1287–1292.
29] For a recent stereocontrolled synthesis of cephalosporolide E, see: O.
Cortezano-Arellano, L. Quintero, F. Sartillo-Piscil, J. Org. Chem. 2015, 80,
[
[
3
33–344.
[
5] a) M. J. Ackland, J. R. Hanson, P. B. Hitchcock, A. H. Ratcliffe, J. Chem.
Soc. Perkin Trans. 1 1985, 843–847; b) V. Rukachaisirikul, S. Pramjit, C.
Pakawatchai, M. Isaka, S. Supothina, J. Nat. Prod. 2004, 67, 1953–1955;
c) J. L. Oller-López, M. Iranzo, S. Mormeneo, E. Oliver, J. M. Cuerva, J. E.
Oltra, Org. Biomol. Chem. 2005, 3, 1172–1173; d) X. Li, Y. Yao, Y. Zheng,
I. Sattler, W. Lin, Arch. Pharmacal Res. 2007, 30, 812–815.
6] X. Li, I. Sattler, W. Lin, J. Antibiot. 2007, 60, 191–195.
7] S. Chang, S. Hur, R. Britton, Angew. Chem. Int. Ed. 2015, 54, 211 –214;
Angew. Chem. 2015, 127, 213–216.
2
601–2608.
30] A. Ueda, A. Yamamoto, D. Kato, Y. Kishi, J. Am. Chem. Soc. 2014, 136,
171–5176.
5
[
[
31] P. Renaud, D. Seebach, Helv. Chim. Acta 1986, 69, 1704–1710.
32] J. D. White, T. C. Somers, G. N. Reddy, J. Am. Chem. Soc. 1986, 108,
5352–5353.
[
[
[
[
[
33] a) I. E. Markó, D. J. Bayston, Tetrahedron 1994, 50, 7141–7156; b) A. P.
Dobbs, S. J. J. GuesnØ, Synlett 2005, 2101–2103.
34] B. Neises, W. Steglich, Angew. Chem. Int. Ed. Engl. 1978, 17, 522–524;
Angew. Chem. 1978, 90, 556–557.
35] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull. Chem. Soc.
Jpn. 1979, 52, 1989–1993.
36] D. A. Singleton, S.-W. Leung, J. Organomet. Chem. 1997, 544, 157–161.
37] A. P. Ramirez, A. M. Thomas, K. A. Woerpel, Org. Lett. 2009, 11, 507–510.
38] a) T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102, 5974–5976;
b) V. S. Martin, S. S. Woodward, T. Katsuki, Y. Yamada, M. Ikeda, K. B.
Sharpless, J. Am. Chem. Soc. 1981, 103, 6237–6240.
39] Y. Kitano, T. Matsumoto, F. Sato, Tetrahedron 1988, 44, 4073–4086.
40] a) J. A. Dale, D. L. Dull, H. S. Mosher, J. Org. Chem. 1969, 34, 2543–2549;
b) J. A. Dale, H. S. Mosher, J. Am. Chem. Soc. 1973, 95, 512–519; c) T. R.
Hoye, C. S. Jeffrey, F. Shao, Nat. Protoc. 2007, 2, 2451–2458.
41] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155–4156.
42] B. S. Bal, W. E. Childers, Jr., H. W. Pinnick, Tetrahedron 1981, 37, 2091–
[
[
8] S. Chang, R. Britton, Org. Lett. 2012, 14, 5844–5847.
9] a) N. Miyaura, A. Suzuki, J. Chem. Soc. Chem. Commun. 1979, 866–867;
b) N. Miyaura, K. Yamada, A. Suzuki, Tetrahedron Lett. 1979, 20, 3437–
3
440.
10] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 16, 4467–
470.
11] Epimerization of a synthetic precursor to cephalosporolide H via a bute-
nolide intermediate has previously been reported, see: R. A. Fernandes,
M. B. Halle, Asian J. Org. Chem. 2013, 2, 593–599.
12] For a recent review of the utility of a-haloaldehydes in synthesis, see:
R. Britton, B. Kang, Nat. Prod. Rep. 2013, 30, 227–236.
13] For selected examples, see: a) V. Farina, S. Huang, Tetrahedron Lett.
[
[
[
[
[
4
[
[
[
[
1
992, 33, 3979–3982; b) K. C. Nicolaou, J. Renaud, P. G. Nantermet, E. A.
Couladouros, R. K. Guy, W. Wrasidlo, J. Am. Chem. Soc. 1995, 117, 2409–
420; c) I. Ojima, S. Lin, T. Inoue, M. L. Miller, C. P. Borella, X. Geng, J. J.
[
[
2
2096.
Walsh, J. Am. Chem. Soc. 2000, 122, 5343–5353.
[
43] E. A. Ilardi, C. E. Stivala, A. Zakarian, Org. Lett. 2008, 10, 1727–1730.
[
[
14] D. T. Warner, J. Org. Chem. 1959, 24, 1536–1539.
15] M. M. Bio, F. Xu, M. Waters, J. M. Williams, K. A. Savary, C. J. Cowden, C.
Yang, E. Buck, Z. J. Song, D. M. Tschaen, R. P. Volante, R. A. Reamer,
E. J. J. Grabowski, J. Org. Chem. 2004, 69, 6257–6266.
Received: July 13, 2015
Published online on October 1, 2015
Chem. Eur. J. 2015, 21, 16646 – 16653
www.chemeurj.org
16653
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim