Stereochemical Revision, Total Synthesis, and Solution State Conformation of the Complex Chlorosulfolipid Mytilipin B

Article abstract of DOI:10.1021/jacs.9b05013

Chlorosulfolipids constitute a structurally intriguing and synthetically challenging class of marine natural products that are isolated from mussels and freshwater algae. The most complex structure from this family of compounds is currently represented by

Full text of DOI:10.1021/jacs.9b05013

Subscriber access provided by UNIV OF SOUTHERN INDIANA
Article
Stereochemical Revision, Total Synthesis and Solution State
Conformation of the Complex Chlorosulfolipid Mytilipin B
Philipp Sondermann, and Erick M Carreira
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b05013 • Publication Date (Web): 03 Jun 2019
Downloaded from http://pubs.acs.org on June 3, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Stereochemical Revision, Total Synthesis and Solution State Confor-
mation of the Complex Chlorosulfolipid Mytilipin B.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Philipp Sondermann, Erick M. Carreira*
Eidgenössische Technische Hochschule Zürich, Vladimir-Prelog-Weg 3, 8093 Zürich, Switzerland
Supporting Information Placeholder
sparked significant research interest with regard to their toxi-
cology, synthesis, and biological role.3 Members of the family
display antimicrobial and antiviral activity as well as protein ki-
nase inhibition.4 CSLs constitute 90% of the polar lipid content
in the flagellar membrane of O. Danica, which otherwise lacks
phospholipids.5 Unlike many complex natural products, whose
role are speculated to be defensive, CSL's are suggested to play
a role integral to structure of the organism’s membrane. This is
noteworthy because cell compartmentalization is a requisite for
the emergence of life6 and fundamentally different membrane
architectures are usually associated with different domains of
life7 or found in highly specialized organelles.8
ABSTRACT: Chlorosulfolipids constitute a structurally intri-
guing and synthetically challenging class of marine natural
products that are isolated from mussels and freshwater algae.
The most complex structure from this family of compounds is
currently represented by Mytilipin B, isolated in 2002 from the
culinary mussel Mytilus galloprovincialis, whose initially pro-
posed structure was shown to be incorrect. In this study we pre-
sent the synthesis of 4 diastereomers which allowed the reas-
signment of 8 stereocenters and stereochemical revision of Myt-
ilipin B, along with the determination of the dominant solution
state conformation.
To date the most complex CSL 1 (Figure 1, left) was isolated
from the Italian region Emilia Romagna in 20029 when a routine
check revealed a sample of the culinary mussel Mytilus gallo-
provincialis to be toxic in a mouse bioassay. The sample was
subsequently shown to contain a lipid whose structure was pro-
posed to be 1 (Figure 1). Although the primary producing or-
ganism remains elusive, it was hypothesized that the mussels
accumulate this compound through their dietary intake and the
similarity of 1 to other chlorosulfolipids that are produced by
microalgae have led to the suggestion of a similar origin for 1.
Introduction
Chlorosulfolipids (CSL) are a fascinating class of marine mem-
brane lipids that were first isolated by Elovson and Vagelos
from the freshwater algae Ochromonas danica in 1969.1 Nu-
merous members of this family have been isolated since, mostly
from the phylogenetic class Chrysophycae.2 Their unusual
structure raises interesting questions regarding their enigmatic
role as lipids. CSLs have been linked to seafood poisoning and
Figure 1. Originally proposed structure of Mytilipin B (1) and our strategy towards determining the revised structure of Mytilipin B (2).
ACS Paragon Plus Environment

Journal of the American Chemical Society
This lack of knowledge impedes access to further material for
chemical and biological study, currently rendering chemical
synthesis the only potential supply source
Page 2 of 10
1
2
3
4
5
6
7
8
9
In addition to missing provenance, there are additional compli-
cations concerning the natural product. A total synthesis of 1
from our group produced spectral data that was not in accord-
ance with that published for natural Mytilipin B.10 Confounding
the problem, the raw isolation data of 1 (such as high-resolution
NMR spectra or FIDs) are irrecoverable. Scanned spectra for
Figure 2. Strategy for the configurational assignment of 1 by
Ciminiello and co-workers.
1
1H, HMQC, and COSY as well as tabulated data from H and
13C-NMR, ESI-HRMS, and IR are the only available infor-
mation for 1. It is notable that a 13C-NMR spectrum of 1 is miss-
ing from the collection. Additionally, there are inconsistencies
with the published data for the peracetylated natural product
(Figure 2, nominally 3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1H-NMR shifts for the Mosher esters are provided in the text of
the manuscript only, and there is otherwise a lack of spectra in
SI. Nonetheless, our own analysis using the data provided for
these three centers was in accordance with the assignment of
configuration by Ciminiello as C11(R), C17(R), and C22(S).
This influenced our decision to look to reassignment at other
stereogenic centers of 1.
Our preliminary analysis suggested that the DEPT spectrum
was accumulated in d4-MeOD and not d6-acetone as indicated.
This discrepancy was also true for the HMQC of 1, which was
recorded in d4-methanol and not d6-acetone as well. Compound-
ing the problem, spectra of three Mosher ester derivatives of the
natural product, discussed by Ciminiello et. al. in the text for
configurational assignment of C1, C11, C17 and C22, are ab-
sent in the published isolation report. On these terms we set out
to revise the structure for Mytilipin B with the benefits of a
broad outline of a synthesis route. This enabled access to a set
of four diastereomers whose data proved critical for compara-
tive purposes and reassignments of Mytilipin B as 2 (Figure 1,
right).
Configuration assignment at remote stereocenter C1. The con-
figuration at C1 was determined by Ciminiello after reductive
cleavage of the palmitoyl ester and esterification with (R)-
Mosher’s acid chloride, following a temperature-dependent
protocol modelled after that devised by Riguera.14 However, it
is important to note this NMR-experiment was originally devel-
oped for methoxy-phenylacetic acid esters; and in an earlier
study, Riguera had noted that Mosher esters can only provide
limited information, as they are subject to more complex con-
formational equilibria.15 Accordingly, we hypothesized that the
C1 stereocenter might be subject to revision.
Discussion
Analysis of Published Configurational Assignment of 1.
Ciminiello and co-workers employed Murata's J-based
configurational analysis (JBCA)16 with data supplied for nomi-
nal pentaacetate 3 for the determination of relative configura-
tion. Given the limited quality of the supplied spectra (as exem-
plified by Figure 3), we could only extract some of the coupling
constants, and still, most proved difficult to determine accu-
rately. Consequently, we depended on the published table of
coupling constants and not the spectra themselves, which led to
no corrections. Murata has pointed out that two conformations
A3 and B3 (Figure 3) cannot be differentiated purely on the ba-
sis of homo- and heteronuclear coupling constants. Thus,
Ciminiello and co-workers recorded ROE (Rotating frame nu-
clear Overhauser Effect) experiments, of which a portion is
shown in Figure 3. Our analysis concurs with that of Ciminiello
that the conformation about C14–C15 appears to correspond to
B3, which results in assigning the relative configuration as anti
because no ROE signal can be observed between protons H13
and H16 (Figure 3, green circles). Additionally, the confor-
mation about C12 and C13 is judged to be A3, because an ROE
signal can be clearly identified for protons H11 and H14 (or-
ange circles, Figure 3). The authors state in the text that the
ROE interaction between H9 and H12 is consistent with con-
formation A3. Inspection of the ROESY spectrum reveals that
there is no such interaction (blue circles and boxes, Figure 3).
Connectivity of published structure. Our analysis began with
careful examination of reported 1H-NMR for the verification of
the proposed connectivity in 1. Comparison of the isolation
team’s 1H-NMR spectrum and their tabulated 1H-NMR data of
1 recorded in d6-acetone revealed discrepancies between the
two. For example, one of the most pronounced deviations is
seen in the methylene signal corresponding to the primary alco-
hol, which appears at δ 3.60 as a broad multiplet in the spectrum
but is tabulated at δ 4.26 and δ 4.01 ppm. Accordingly, 1H-NMR
data for 1 was judged unreliable. This was further complicated
by the fact that the respective HMQC of 1 appeared to be rec-
orded in d4-methanol, as suggested by the strong residual sol-
vent peak at δ ≈ 3.31 ppm (1H)/49.0 ppm (13C), with the other
spectra of 1 having been recorded in d6-acetone. Consequently,
we turned to the spectral data of nominal pentaacetate 311 (Fig-
1
ure 2). Close inspection of the published H-NMR and 13C-
NMR along with 2D NMR COSY, HSQC, and HMBC spectra
lead to the conclusion that no revision of the connectivity was
necessary.
Absolute configuration assignment at C11/C17/C22 by isola-
tion group. Mosher ester analysis employing Kakisawa's modi-
fied protocol12 was conducted by the isolation team following
preparation of the (R)- and (S)-Mosher-triester derivatives acyl-
ated at C11, C17, and C22 (Figure 2, alcohols indicated with a
green line).13 It should be noted that selected
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Figure 3. Analysis of the ROESY spectrum of 3 and the accordance with its proposed structure. For conformations A3 and B3 all five
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
significant coupling constants are of similar size: 3J1H-1H’: large, 2J13C’-1H: large, 2J1H’-13C: large, 3J13C’’-1H: small 3J1H’-13C’’’: small.
This is especially noteworthy given that the spectrum is missing
the expected twinned, cognate signal. Accordingly, we con-
clude that the data is consistent with B3 as the dominant con-
formation and assign relative anti-configuration for C10–C11.
Our assignment of configuration at C10 necessitates revision of
the original assignment at C5–C7 (S,S,R) and C9–C10 (R,S)
with five inverted stereocenters (C5–C7 (R,R,S) and C9–C10
(S,R)) as their configurations were determined relative to that at
C11. When combined with the uncertainty at C1, this leads to a
pair of C1 epimers 4a and 4b (Figure 3), which would need to
be prepared and studied to potentially identify the correct con-
figuration of natural Mytilipin B.
omer. A sequence of reactions which included acetonide for-
mation (73%), silyl ether cleavage (quant.), DMP oxidation of
the primary alcohol (97%), followed by Still–Gennari olefina-
tion (72%) provided cis-enoate 9. 1,2-Reduction with DIBAL
followed by Sharpless epoxidation then provided unstable cis-
2,3-epoxyalcohol 10. Opening of the epoxide at C3 with a chlo-
ride source proved to be troublesome, as reported methods led
to either no desired reactivity (PPh3Cl2, TMSCl) or mixtures of
regioisomers ranging from 1:1 to 1.5:1 (Ti(OiPr)4/NH4Cl,
TiCl(OiPr)3, CeCl3).19 Screening of a wide variety of metal
chlorides led to the identification of MgCl2 as a novel reagent
for the selective chlorolysis of cis-2,3-epoxy alcohols, provid-
ing the desired 1,2-diol 11 in high yield and regioselectivity in
refluxing ethyl acetate (86% 1,2-diol, 13% 1,3-diol).20 Interest-
ingly, the reaction showed little regioselectivity when run at
room temperature (1,2:1,3 diol ~ 1:1). Following 1,3-dioxolane
formation (79%) and hydrogenolysis of the benzyl ether in 95%
yield, the primary alcohol was subjected to a Mitsunobu reac-
tion to give 12 in quantitative yield, and subsequent oxidation
delivered sulfone 5 in 84% yield. The synthesis of aldehyde
fragment 6 containing the five reassigned stereocenters com-
menced from 13, which is accessible in seven steps from D(+)-
malic acid.21 Allylic alcohol 13 underwent Sharpless epoxida-
tion to give 4,5-dichloro-2,3-epoxy alcohol 14 (72%, Scheme
1). Esterification followed by dioxolane hydrolysis provided the
1,2-diol (76%) of which both hydroxyl groups were TBS-
protected (77%). Selective cleavage of the primary silyl ether to
15 (77%, 2 cycles), and Dess–Martin oxidation delivered an un-
stable aldehyde that was directly subjected to conditions
Synthesis and Analysis of 4a-b. The route to the originally pub-
lished structure 1 reported by our group was designed to address
the specific, proposed configurational relationship in the C5-
C10 domain. Accordingly, a new stereochemically-driven ret-
rosynthetic analysis targeting 4a was necessary. This led to two
fragments of similar complexity: sulfone 5 and aldehyde 6,
which in the synthetic direction would be joined by implemen-
tation of the Julia-Kociensky olefination reaction (Scheme 1).17
Sulfone fragment 5 was prepared from epoxide 7, itself availa-
ble in seven steps from 1,5-pentane-diol.10 The use of ZrCl4 for
epoxide opening of 7 did not scale well, consequently, we
looked to a different method for generation of the chlorohydrin.
Denmark's method for epoxide openings with SiCl4 and
HMPA18 proved reliable on large scale and cleanly delivered a
chlorohydrin, which without purification was converted to 1,3-
anti diol 8 with NaBH(OAc)3 in 92% yield as a single diastere-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
Scheme 1. Construction of Sulfone 5, Aldehyde 6 and Key Julia Olefination.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: Synthesis of sulfone fragment 5: a) SiCl4, HMPA, CH2Cl2, 96 h, 91%; b) NaBH(OAc)3, AcOH-MeCN, 0
°C, 92%; c) MeCH(OMe)CH2, PPTS, CH2Cl2, 23 °C, 73%; d) nBu4NF, AcOH, DMF, 23 °C, quant.; e) DMP, CH2Cl2, 23 °C, 97%;
f) (CF3CH2O)2P(O)CH2CO2Me, KN(SiMe3)2, THF, –78 °C, 72%; g) i-Bu2AlH, THF, –78 °C, 4 h, 96%; h) Ti(Oi-Pr)4, t-BuO2H, (+)-
diethyl L-tartrate, CH2Cl2, –20 °C, 83% (dr = 4:1); i) MgCl2, AcOEt, 80 °C, 86%; j) CuSO4, p-TsOH, acetone, 79%; k) Pd/C, H2,
EtOAc, 23 °C, 90 min, 95%; l) 1-phenyl-1H-tetrazole-5-thiol, (i-PrO2C)N2, PPh3, THF, 0 °C to 23 °C, quant.; m)
(NH4)6Mo7O24•4H2O, H2O2, THF–EtOH 60 °C, 84%. Synthesis of aldehyde fragment 6 and Julia-olefination: n) Ti(Oi-Pr)4, (–)-
diethyl D-tartrate, t-BuOOH, 4Å MS, CH2Cl2, 72%; o) AcCl, pyridine, CH2Cl2, 76%; p) CSA, MeOH, 80%; q) TBSOTf, 2,6-lutidine,
CH2Cl2, 77%; r) HF•py, py, THF, 77% (2 cycles); s) DMP, t-BuOH, CH2Cl2, quant.; t) 16, KHMDS, THF, 93%; u) Et4NCl3, CH2Cl2,
–78 °C, 92%, dr 5:1; v) K2CO3, MeOH, 0 °C, 95%; w) DMP, CH2Cl2; x) 5 (1.2 equiv.), NaHMDS, PhMe, –78 °C to rt, 49% (2 steps),
E:Z = 3:1. DMP = Dess–Martin periodinane, PPTS = pyridinium para-toluenesulfonate, pTsOH = para-toluenesulfonic acid.
for Wittig olefination with ylide 1610 to give cis-olefin 17. In-
verse addition of the starting aldehyde to the phosphonium ylide
was crucial to obtain a high yield of 17 (93%, 2 steps). The ole-
fin was then subjected to dichlorination employing Mi-
oskowki's reagent (Et4NCl3) to give 18. The stereochemical out-
come of the chlorination reaction was in accordance with
Vanderwal’s report that cis-allylic alcohol derivatives provide
the corresponding syn-syn dichloride.22 Initial attempts in exe-
cuting this reaction (0.05M in CH2Cl2, addition of Et4NCl3 to
the olefin at –78 °C) gave low yields due to competing for-
mation of a furan, derived from intramolecular attack of the
proximal benzyl ether on the intermediate epichloronium ion
with subsequent loss of benzyl chloride. A related mode of an-
chimeric attack was previously observed by Burns en route to
(−)-Danicalipin.23
ployed, yielding product 18 in high yield and good diastereose-
lectivity, as determined by NMR (92%, dr = 5:1).24 Hydrolysis
of the ester in 18 delivered 19 (95%), and DMP oxidation of the
primary alcohol afforded unstable aldehyde 6, which was cou-
pled with sulfone 5 in toluene in the presence of NaHMDS to
obtain the desired cis-olefin 20 (49% over 2 steps, dr = 3:1 as
determined by NMR).25 In this reaction the base was added last
to a solution of the sulfone and the aldehyde at –78 °C to avoid
E1cb elimination of the deprotonated sulfone.
Next, we attempted opening of trans-epoxide in 20 with reten-
tion of configuration at the allylic sterecoenter (Scheme 2) in
order to obtain the desired cis-chlorohydrin. Lewis acids such
as SiCl4, MgCl2, or Ph3PCl2 did not lead to any product for-
mation. Although TMSCl in methylene chloride effected the
conversion of 20 in 2 h to the corresponding chlorohydrin in
51% yield, analysis of the product indicated that inversion of
configuration had occurred at the allylic carbon.26 When 20 was
subjected to TMSCl in a 1:1 mixture of ethyl acetate and meth-
ylene chloride the desired syn-chlorohydrin 21 was obtained af-
ter long reaction times (5 days). In accordance with results pre-
viously reported by our group,26 addition of ethyl acetate as a
Whilst intramolecular opening of the chloronium ion can be ex-
pected to follow a first order rate law, the desired chloride ad-
dition to the chloronium ion may also depend on chloride
concentration. Consequently, inverse addition of the alkene 17
to a saturated solution of Et4NCl3 in CH2Cl2 at –78 °C was em-
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
made use of this strategy by preparing the pentaacetate of syn-
Scheme 2. Stereoretentive Epoxide Opening with Anchi-
1
2
3
4
5
6
7
8
meric Assistance.
thetic 4a (Scheme 3) by treatment with Ac2O and pyridine over
3 days to give 27. Although there were now considerable simi-
larities between the 1H-NMR spectra of Ciminiello’s pentaace-
tate (500 MHz) and 27 (400 MHz), there were notable differ-
ences extracted from the spectral data for the shifts of various
protons, for example the methylene protons at C19  2.55 (nom-
inal 3) versus  2.65/2.48 (27). These results suggest H-bonding
effects in the natural product pose a potential complication
when conducting spectral comparisons and strongly imply that
spectral comparisons should include the peracetylated deriva-
tives.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Taking into account our observations involving 27 together
with our reservations concerning the assignment of the config-
uration at C1, we decided to prepare the C1-epimer of com-
pound 4a (C1-epi 4a), or 4b. We identified bis-TBS ether 24 as
a suitable late stage entry point (Scheme 4 A).
Monodeprotection of the less hindered TBS ether at 0 °C deliv-
ered 28 (50%), which is poised for inversion at the allylic C1
by implementation of a Mitsunobu reaction (Scheme 4 A).
Reagents and conditions: TMSCl (30 equiv.), HCl in AcOEt
(1M, 7-9 equiv.), CH2Cl2–AcOEt (1:1), 49%.
Scheme 3. Synthesis of Diastereomer 4a and Peracetylation.
more polar solvent is thought to promote anchimeric assistance
through formation of polar chloretanium and chlorolanium ions
(Scheme 2). Unfortunately, this method proved to be even
slower and unreliable on large scale, producing complex mix-
tures of polar by-products. In attempts to understand the lack of
scalability, we focussed on TMSCl and hypothesized that it
served two functions. First, its reaction with adventitious water
on small scale would generate HCl, which would in turn effect
rapid epoxide opening. Second, its reaction with water also pre-
serves a strictly anhydrous reaction medium, thereby precluding
hydrolysis of the various labile protecting groups. Accordingly,
we found that slow addition of HCl in ethyl acetate in the pres-
ence of TMSCl led to rapid (2 h) formation of 21 (49%), which
proved to be highly sensitive to alkaline conditions.27
Subsequently, 21 underwent smooth dichlorination in 67%
yield with Et4NCl3 in CH2Cl2 when warmed to room tempera-
ture to give undecachloride 22 (Scheme 3) as the desired stere-
oisomer as determined by NMR.28 Debenzylation of 22 proved
surprisingly challenging, as standard conditions (Pd/C and H2
in AcOEt) failed to provide product. Fortunately, screening a
wide variety of palladium sources revealed Degussa type NE/W
Pd/C as suitable, delivering alcohol 23 (65%) within a few
minutes at 35 °C in a 1:1 mixture of MeOH–THF. Stepwise ad-
dition of Martin's sulfurane in small portions of 0.12 equivalents
proved vital to selective dehydration of 23 to access 24 (52%).
The trans-configuration of the olefin was established on the ba-
sis of the large vinylic C–H coupling (3JH–H = 15.4 Hz). Subse-
quent silyl ether cleavage delivered the corresponding triol that
was selectively acylated at the allylic alcohol to give 25
30
(67%).29 Subsequently, O-sulfation with DMF•SO3 gave
monosulfate 26 in 80% yield. Hydrolysis of the acetals then de-
livered the target structure 4a (72%), without transposition or
cleavage of either the ester or the sulfate groups. Unfortunately,
1
the H and 13C-NMR spectra we acquired for 4a in d6-acetone
Reagents and conditions: a) Et4NCl3, CH2Cl2, 57%; b) H2, Pd/C,
THF–MeOH, 35 °C, 65%; c) Martin's sulfurane, PhMe, 58%;
d) HF•py, py, MeCN, 95%; e) palmitoyl chloride, py, CH2Cl2,
–78 °C to –40 °C 67%; f) DMF•SO3, Na2SO4, DMF–py, 80%;
g) F3CCO2H–H2O (1:1), 72%, h) Ac2O, py. Martin's sulfurane
= [PhC(CF3)2O]2SPh2, py = pyridine.
1
differed substantially from H-NMR spectra and the tabulated
1H and 13C-NMR data reported by Ciminiello for natural Mytil-
ipin B in d6-acetone. In the original isolation and structural de-
termination work, as a consequence of multiply overlapping 1H-
NMR signals for the natural product (nominally 1), Ciminiello
synthesized the corresponding nominal pentaacetate 3. We
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
Scheme 4. Synthesis and Peracetylation of Key Diastereomers 4b, 35 and the Natural Product Mytilipin B (2).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: a) HF•py, py, MeCN, 23 °C to 0 °C, 50% 28, 35% 24; b) 2-picolinic acid, (i-PrO2C)N2, PPh3, THF, -20 °C
to 0 °C, 18%; c) Cu(OAc)2, MeOH, CHCl3, 100%; d) HF•py, py, MeCN, 73%; e) C15H31COCl, py, CH2Cl2, –78 °C to –40 °C 75%;
f) DMF•SO3, Na2SO4, DMF–py, 83%; g) F3CCO2H–H2O (1:1), 72%; h) Ti(Oi-Pr)4, t-BuO2H, (–)-diethyl D-tartrate, CH2Cl2, –20 °C,
77% (dr=4:1); ); i) MgCl2, AcOEt, 80 °C, 77%; j) CuSO4, p-TsOH, acetone, 81%; k) Pd/C, H2, EtOAc, 23 °C, 90 min, 95%; l) 1-
phenyl-1H-tetrazole-5-thiol, (i-PrO2C)N2, PPh3, THF, 0 °C to 23 °C, 91%; m) (NH4)6Mo7O24•4H2O, H2O2, THF–EtOH 60 °C, 70%,
n) 6, NaHMDS, PhMe, –78 °C to rt, 66% (2 steps), E:Z = 3:1; o) HCl in AcOEt (1M), AcOEt–CH2Cl2 (1:1), 55%; p) Et4NCl3, CH2Cl2,
66%; q) H2, Pd/C, THF–MeOH, 35 °C, 61%; r) Martin's sulfurane, PhMe, 60%; s) HF•py, py, MeCN, 66%; t) C15H31COCl, py,
CH2Cl2, –78 °C to –40 °C 76%; u) DMF•SO3, Na2SO4, DMF–py, 84%; v) F3CCO2H–H2O (1:1), 72%; w) 6-Me-2-picolinic acid, (i-
PrO2C)N2, PPh3, THF, -20 °C to 0 °C, 42%; x) Cu(OAc)2, MeOH–CHCl3 (1:1), 85%; y) palmitoyl chloride, py, CH2Cl2, –78 °C to –
40 °C 71%; z) DMF•SO3, Na2SO4, DMF–py, 90%; aa) F3CCO2H–H2O (1:1), 79%. py = pyridine, DMF = N,N-dimethylformamide,
Martin's sulfurane = [PhC(CF3)2O]2SPh2.
The use of standard conditions involving p-nitrobenzoic acid
proved to be unsuitable, because conditions for the subsequent
hydrolysis were incompatible with the base-sensitive chlorohy-
drins. Accordingly, picolinic acid was examined as nucleophile,
as it has been reported to undergo hydrolysis under neutral con-
ditions with Zn(OAc)2 or Cu(OAc)2 in methanol.31 Fortunately,
we were able to obtain ester 29 inverted at C1, albeit in only
18% yield. Subsequently, hydrolysis of 29 to 30 occurred in
quantitative fashion (3.8 equivalents Cu(OAc)2, 100 equiva-
lents MeOH in CHCl3),and silyl ether cleavage with
HF/pyridine delivered the triol (73%). The previously estab-
lished acylation-sulfation-acetal hydrolysis sequence gave ac-
cess to 4b (3 steps, 45%). Pentaol 4b was subjected to peracety-
lation to give 31. We then compared the data from 1H and 13C-
NMR spectra for 31 to the Ciminiello data for the peracetate of
the natural product. For 31 the methylene protons at C19 once
again showed the largest deviation, with Ciminiello tabulating
a multiplet at  2.60 for 3 while synthetic material 31 displayed
resonances at  2.63 and  2.48 ppm.
We proceeded to conduct configurational studies of 4b by
JBCA. The derived couplings in the C5–C16 and C21–C22 ar-
rays were in accordance with Ciminiello's observations within
these regions for peracetylated Mytilipin B (3) (see SI for de-
tails). With this in mind, we reasoned that the remaining spec-
tral discrepancies could only be explained by incorrect relative
configuration between the two spin systems C1-Me–C17 and
C19–C23. This leads to compounds 35 and 2 as targets of
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Observed coupling constants in d6-acetone and the resulting dominant solution-state conformation of Mytilipin B (2) determined
for C5–C17 and C21–C22 by J-based configurational analysis.
interest (Scheme 4 B). The stereochemical problem presented
by the new targets necessitated restarting the synthetic sequence
with allylic alcohol 32 (1 step from 9). Using D(–)-diethyl tar-
trate instead of L(+)-diethyl tartrate, Sharpless epoxidation of
32 delivered epoxide 3332 (77%, dr = 4:1). Elaboration to 34
was conducted as previously presented in Schemes 2 and 3 for
the respective C21-epi/C22-epi compounds. Palmitoylation
(76%), O-sulfation (84%), and acetonide hydrolysis (72%) de-
livered 35. Additionally, pentaol 35 was subjected to peracety-
lation, as previously described for 27, to give 36.
4 A). Subsequent hydrolysis of the more hindered 6-methyl pic-
olinate 37 to 38 was achieved through the use of higher amounts
of both methanol (CHCl3–MeOH 1:1) and Cu(OAc)2 (10 equiv-
alents) without any observable chlorohydrin reactions despite
three potential sites for oxirane formation (85%, Scheme 4 B).
It is interesting to note that unlike the sequence from 28 to 30,
the inversion sequence of 34 to 38 was conducted with the free
alcohol at C7, obviating the need for a second silyl deprotection
step in the reaction sequence. Elaboration of 38 was conducted
as previously described for 35 through palmitoylation (71%),
O-sulfation (90%), and acetonide deprotection (79%) (Scheme
4 B). Acylation with Ac2O and pyridine delivered 39. Compar-
ing the recorded spectra of the four synthetic diastereomeric
pentaacetates 27, 31, 36, and 39, the spectra of 39 optically
matched with those published by Ciminiello for the peracety-
lated natural product (nominal 3).34 In addition the HSQC of
synthetic pentaol 2 in d4-methanol also matched the HMQC of
the natural product (nominally 1). The 13C-NMR spectrum of
synthetic pentaol 2 in d4-methanol matched the DEPT spectrum
provided for the peracetylated natural product (nominally pen-
taacetate 3 and nominally recorded in d6-acetone), suggesting
the DEPT spectrum provided by Ciminiello was recorded of the
natural product in d4-methanol and incorrectly labeled.35 Thus,
we conclude that 2, inverted at the eight stereocenters
C1/C5/C6/C7/C9/C10/C21/C22 relative to the originally pro-
posed structure 1, represents the structure of Mytilipin B. Ad-
ditionally, these results may suggest that the related natural
product Mytilipin C lacking the C17-hydroxyl group is likely
The HSQC-spectrum of pentaol 35 in d4-methanol matched the
HMQC reported by Ciminiello for the natural product Mytilipin
B. However, comparing the spectra obtained for 36 to the
peracetylated natural product, a significantly different peak
shape of the two vinylic resonances at  5.69 and  5.79 ppm
was observed. We thus set out to synthesize the C1-inverted
compound 2 (Scheme 4 B). Given the small amount of material
available and the necessity to prepare both C1 diastereomers,
we were forced to improve the Mitsunobu inversion to gain ac-
cess to sufficient material. While changing the nucleophile from
picolinic acid to chloroacetic acid vastly improved the yield of
the inverted product (62%), we were unable to find hydrolytic
conditions that left the sensitive chlorohydrin intact.33 Further
experimentation revealed that the use of the less nucleophilic 6-
methyl-2-picolinic acid delivered 37 in a significantly improved
yield of 42% (vs. 18% for synthesis of 29 as shown in Scheme
ACS Paragon Plus Environment

Journal of the American Chemical Society
also correctly depicted with our revised stereochemical assign-
Page 8 of 10
1
2
ment at the respective eight stereocenters given the similarities
that have been pointed out for Mytilipin B and C.36
3
4
5
6
7
8
9
In addition, the dominant solution state structure of Mytilipin B
(2) was modeled based on the determined conformations (Fig-
ure 4). Intriguingly the C5–C17 carbon chain adopts a rigid con-
formation in which the chlorinated region appears U-shaped.
The sulfate and the core hydroxyl groups form the polar face of
Mytilipin B (2). The opposing face is dominated by the large
palmitoate sidechain along with numerous chloride substitu-
ents. However, it should be noted that regions C1–C4 and C18–
C20 are likely to have increased degree of rotational flexibility
in solution. Conventional phospholipids usually consist of a po-
lar, relatively small phosphate headgroup and a long nonpolar
fatty acid chain. In contrast, chlorosulfolipids often contain a
broad distribution of polar groups throughout the molecule.
Given the observed dominant solution state conformation of
Mytilipin B (2) we propose that this discrepancy may be re-
solved for 2 when considering its folded secondary structure.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Intriguingly, comparative analysis of the stereohexade C9–C14
of Mytilipin B relative to other known chlorosulfolipids reveals
partially conserved stereochemical sequences. As shown in
Figure 5 alignment of the aliphatic array of Mytilipin A and B,
Figure 5. Comparison of conserved stereoarray patterns
Danicalipin
A as well as Malhamensilipin A reveals
throughout a selection of different chlorosulfolipids. /green
denotes a substituent pointing inwards, /red outwards. Some
conserved stereocenters are formally inverted in the CIP
homologies. The complete stereohexade of Mytilipin A2c is
conserved in the revised structure of Mytilipin B (2).
Malhamensilipin A37 and Danicalipin A38 share a sequence of
five and four stereocenters with 2. This may hint to a common
biosynthetic pathway or lipid function and raises the question
what structural role this highly conserved sequence fulfills.
Comparison of the solution state structure of our Mytilipin B,
Mytilipin A (both in d6-acetone),2c Danicalipin A (in d4-
MeOD), as well as desulfated Fluoro- and Bromodanicalipin (in
−
nomenclature. R = SO3 .
results in subtle spectral differences among the diastereomers.
However, the synthesis of four key diastereomers 2, 4a, 4b and
35, enables direct comparison and identification of 2 as the most
likely structure of the natural product, illustrating the power of
J-based configurational analysis. Moreover, we conducted con-
formational analysis of the solution state structure of 2. This
may explain for the ability of chlorosulfolipids to form or insert
into lipid bilayers by taking into account the secondary struc-
ture. It is intriguing to speculate how Mytilipin B (2), in light of
its solution state structure, impacts the mechano-physical prop-
erties of the membrane in terms of density, rigidity, chemical
stability, curvature or permeability. Further biophysical studies
are required to provide insight into the phenomena.
CDCl3)39 and
a
reduced, desulfated derivative of
Malhamensilipin A (in C6D6)37a displays similar conformations
within conserved sequences.
In conclusion, we followed a strategy prioritizing changes in
configuration likely to have the largest impact on the spectral
signatures. This guided us through incremental, stepwise
correction for Mytilipin B. Subsequently, we synthesized a total
of four diastereomers, culminating in the reassignment of 8 out
of 15 stereocenters. The distance between stereochemical ar-
rays
ASSOCIATED CONTENT
The Supporting Information is available free of charge via the
Internet at http://pubs.acs.org.”
General methods; detailed experimental procedures;
spectral data; comparison of synthetic and natural Mytilipin B;
X-ray crystallographic data; and references
X-ray crystallographic data
AUTHOR INFORMATION
Corresponding Author
*E-mail:erickm.carreira@org.chem.ethz.
ACKNOWLEDGMENT
ETH Zürich and an ERC 320666_CHLIP Grant are gratefully
acknowledged for financial support.
8
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Based on Carbon−Proton Spin-Coupling Constants. A Method of Con-
References
figuration Analysis for Natural Products. J. Org. Chem. 1999, 64, 866-
876. (b) Nilewski, C.; Geisser, R. W.; Ebert, M.-O.; Carreira, E. M.
Conformational and Configurational Analysis in the Study and
Synthesis of Chlorinated Natural Products. J. Am. Chem. Soc. 2009,
131, 15866-15876. (c) For an alternative approach towards
determination of relative configurations see: Lee, J.; Kobayashi, Y.;
Tezuka, K.; Kishi, Y. Toward Creation of a Universal NMR database
for the Stereochemical Assignment of Acyclic Compounds: Proof of
Concept. Org. Lett. 1999, 1, 2181-2184.
1 Elovson, J.; Vagelos, P. R. A New Class of Lipids: Chlorosulfolipids.
Proc. Natl. Acad. Sci. 1969, 62, 957-963.
2
(a) Elovson, J.; Vagelos, P. R. Structure of the major Species of
Chlorosulfolipid from Ochromonas danica. 2,2,11,13,15,16-Hexa-
chloro-N-docosane 1,14-disulfate. Biochemistry, 1970, 9, 3110-3126.
(b) Chen, J. L.; Proteau, P. J.; Roberts, M. A.; Gerwick, W. H.; Slate,
D. L.; Lee R. H. Structure of Malhamensilipin A, an Inhibitor of Pro-
tein Tyrosine Kinase, from the Cultured Chrysophyte Poterioochromo-
nas malhamensis. J. Nat. Prod. 1994, 57, 524–527. (c) Ciminiello, P.;
Fattorusso, E.; Forino, M. Structural Elucidation of a New Cytotoxin
Isolated from Mussels of the Adriatic Sea. J. Org. Chem. 2001, 66, 578-
582.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
17 (a) Julia, M.; Paris, J. M. Syntheses a l'aide de sulfones V(+) - methode
de synthese generale de doubles liaisons. Tetrahedron Lett. 1973, 14,
4833-4836. (b) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley,
A. A Stereoselective Synthesis of trans-1,2-Disubstituted Alkenes
Based on the Condensation of Aldehydes with Metallated 1-Phenyl-
1H-tetrazol-5-yl Sulfones. Synlett, 1998, 1, 26-28.
3 (a) Ciminiello, P.; Fattorusso, E. Shellfish Toxins - Chemical Studies
on Northern Adriatic Mussels. Eur. J. Org. Chem, 2004, 2533-2551.
(b) Yasamuto, T.; Murata, M. Marine toxins. Chem. Rev. 1993, 93,
1897-1909. Bedke, K.; Vanderwal, C. D. Chlorosulfolipids: Structure,
Synthesis, and Biological Relevance Nat. Prod. Rep. 2011, 28, 15-25.
4 For the reported bioactivity of Malhamensilipin see ref. 2(b).
5 Chen, L. L.; Poursada, M.; Haines, T. H. The Flagellar Membrane of
Ochromonas danica. Lipid Composition. J. Biol. Chem. 1976, 251,
1835-1842.
18 Denmark, S. E.; Barsanti, P. A.; Wong, K.-T.; Stavenger, R. A. En-
antioselective Ring Opening of Epoxides with Silicon Tetrachloride in
the Presence of a Chiral Lewis Base. J. Org. Chem. 1998, 63, 2428-
2429.
19 (a) Caron, M.; Sharpless, K. B. Titanium Isopropoxide-Mediated Nu-
cleophilic Openings of 2,3-Epoxy Alcohols. A Mild Procedure for Re-
gioselective Ring-Opening. J. Org. Chem. 1985, 50, 1557-1560. (b)
Raifeld, Y. E. ; Nikitenko, A. A.; Arshava, B. M. Compounds of (i-
PrO)3TiX as novel reagents for regioselective oxirane ring opening.
Tetrahedron: Asymmetry, 1991, 2, 1083-1084. (c) Sabitha, G.; Babu,
R. S.; Rajkumar, M.; Reddy, C. S.; Yadav, J. S. Highly Regioselective
Ring Opening of Epoxides and Aziridines using Cerium(III) Chloride.
Tetrahedron Lett. 2001, 42, 3955-3958. (d) Wang, C.; Yamamoto, H.
Tungsten-, Molybdenum-, and Cerium-Promoted Regioselective and
Stereospecific Halogenation of 2,3-Epoxy Alcohols and 2,3-Epoxy
Sulfonamides. Org. Lett. 2014, 16, 5937-5939. (e) D. Díaz, T. Martín,
V. S. Martín J. Org. Chem. 2001, 66, 7231-7233.
6
Blain, J. C.; Szostak, J. W. Progress Toward Synthetic Cells. Annu.
Rev. Biochem. 2014, 83, 615-640.
7
Lombard, J.; Lopez-Garcia, P.; Moreira, D. The Early Evolution of
Lipid Membranes and the three Domains of Life. Nat. Rev. Microbiol.
2012, 10, 507-515.
8 van Niftrik, L. A.; Fuerst, J. A.; Damsté, J. S. S.; Kuenen, J. G.; Jetten,
M. S. M.; Strous, M. The Anammoxosome: an Intracytoplasmic
Compartment in Anammox Bacteria.. FEMS Microbiol. Lett. 2004,
233, 7-13. Moss, F. R. III; Shuken, S. R.; Mercer, J. A. M.; Cohen, C.
M.; Weiss, T. M.; Boxer, S. G.; Burns, N. Z. Ladderane Phospholipids
form a Densely Packed Membrane with Normal Hydrazine and
Anomalously low Proton/Hydroxide Permeability. Proc. Nat. Acad.
Sci. U.S.A. 2018 115, 9098-9103.
20 MgCl2 has been reported as a Lewis acid for the thiolysis of cyclo-
hexene oxide and cyclopentene oxide in which traces of the chlorohy-
drin were formed in CH2Cl2: Rani, R.; Pattanayak, S.; Agarwal, J.; Ped-
dinti, R. K. Magnesium Chloride-Catalyzed Thiolysis of Epoxides:
Synthesis of β-Hydroxy Sulfides. Synth. Commun. 2010, 40, 2658-
2666. In our case ethyl acetate and acetonitrile proved to be suitable
solvents at reflux.
9 Ciminiello, P.; Dell'Aversano, C.; Fattorusso, E.; Forino, M.; Magno,
S.; Di Rosa, M.; Ianaro, A.; Poletti, R. Structure and Stereochemistry
of a New Cytotoxic Polychlorinated Sulfolipid from Adriatic Shellfish.
J. Am. Chem. Soc. 2002 124, 13114-13120.
21 (a) Hanessian, S.; Ugolini, A.; Dube, D.; Glamyan, A. Facile access
to (S)-1,2,4-butanetriol and its derivatives. Can. J. Chem. 1984, 62,
2146-2147. (b) Tandon, V. K.; van Leusen, A. M.; Wynberg, H. Syn-
thesis of enantiomerically pure (S)-(+)-3-hydroxytetrahydrofuran, and
its (R)-enantiomer, from malic or tartaric acid. J. Org. Chem. 1983, 48,
2767-2769. See Ref. 10 for the synthesis of ent-13 from (S)-1,2,4-bu-
tanetriol.
10
Nilewski, C.; Deprez, N. R.; Fessard, T. C.; Li, D. B.; Geisser, R.
W.: Carreira, E. M. Synthesis of Undecachlorosulfolipid A: Re‐evalu-
ation of the Nominal Structure. Angew. Chem. Int. Ed. 2011, 50, 7940-
7943.
11 Peracetate 3 was synthesized by Ciminiello et. al. from natural Myt-
ilipin B to achieve better dispersion of the strongly overlapping NMR
signals of 1 (see ref. 9).
22 Shibuya, G. M.; Kanady, J. S.; Vanderwal, C. D. Stereoselective Di-
chlorination of Allylic Alcohol Derivatives to Access Key Stereochem-
ical Arrays of the Chlorosulfolipids. J. Am. Chem. Soc. 2008, 130,
12514-12518.
12
Ohtani, I.; Ksumi, T.; Kashman, Y.; Kakisawa, H. High-field FT
NMR application of Mosher's method. The absolute configurations of
marine terpenoids. J. Am. Chem. Soc. 1991, 113, 4092-4096.
13 It should be noted that the alcohols at C15 and C23 did not undergo
esterification: The ester at the primary alcohol might have undergone
saponification during the hydrolytic workup with K2CO3 and the alco-
hol at C15 likely undergoes epoxide closure with the more basic
DMAP as compared to pyridine used by Ciminiello et. al. in the
peracetylation of 1.
23 Landry, M. L.; Hu, D. X.; McKenna, G. M.; Burns, N. Z. Catalytic
Enantioselective Dihalogenation and the Selective Synthesis of (−)-
Deschloromytilipin A and (−)-Danicalipin A. J. Am. Chem. Soc. 2016,
138, 5150−5158.
24 Addition of cyclopentene allowed rapid and effective quenching of
excess Et4NCl3 at cryogenic temperatures. This obviates the need for
an aqueous quench, which proved to be troublesome at larger scales
and cryogenic temperatures. Both, excess cyclopentene and 1,2-dichlo-
rocyclopentane are easily removed under reduced pressure.
25 Aïssa, C. Mechanistic Manifold and New Developments of the Julia–
Kocienski Reaction. Eur. J. Org. Chem. 2009, 1831-1844.
14
Latypov, S. K.; Seco, J. M.; Quinoa, E.; Riguera, R. Are Both the
(R)- and the (S)-MPA Esters Really Needed for the Assignment of the
Absolute Configuration of Secondary Alcohols by NMR? The Use of
a Single Derivative. J. Am. Chem. Soc. 1998, 120, 877-882.
15 For a report regarding the three low energy conformations of Mosher
esters see: Latypov, S. K.; Seco, J. M.; Quinoa, E.; Riguera, R. MTPA
vs MPA in the Determination of the Absolute Configuration of Chiral
Alcohols by 1H-NMR. J. Org. Chem. 1996, 61, 8569-8577.
26 (a) Shemet, A.; Sarlah, D.; Carreira, E. M. Stereochemical Studies of
the Opening of Chloro Vinyl Epoxides: Cyclic Chloronium Ions as In-
termediates. Org. Lett. 2015, 17, 1878-1881. (b) Nilewski, C.; Geisser,
R. W.; Carreira, E. M. Total Synthesis of a Chlorosulfolipid Cytotoxin
Associated with Seafood Poisoning. Nature 2009, 457, 573-576.
16
(a) Matsumori, N.; Kaneno, D.; Murata, M. Nakamura, H.;
Tachibana, K. Stereochemical Determination of Acyclic Structures
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
Poterioochromonas malhamensis. J. Nat. Prod. 2010, 73, 279-283. (b)
Bedke, D. K.; Shibuya, G. M.; Pereira, A. R.; Gerwick, W. H. ;
Vanderwal, C. D. A Concise Enantioselective Synthesis of the
Chlorosulfolipid Malhamensilipin A. J. Am. Chem. Soc. 2010, 132,
2542-2543.
1
2
3
4
5
6
7
8
27
38
The stereochemical outcome was first probed by reclosure of the
(a) Haines, T. H. Halogen- and Sulfur-Containing Lipids of
epoxide with K2CO3 to give the corresponding cis-epoxide as judged
by the corresponding coupling constants (3JH–H = 3.9 Hz) and later con-
firmed by JBCA of 2 and 4b.
Ochromonas. Annu. Rev. Microbiol. 1973, 27, 403-411. (b) Kawahara,
T.; Kumaki, Y.; Kamada, T.; Ishii, T.; Okino, T. Absolute
Configuration of Chlorosulfolipids from the Chrysophyta Ochromonas
Danica. J. Org. Chem. 2009, 74, 6016-6024. (c) Bedke, D. K.; Shibuya,
G. M.; Pereira, A.; Gerwick, W. H.; Haines, T. H., Vanderwal, C. D.
Relative Stereochemistry Determination and Synthesis of the Major
Chlorosulfolipid from Ochromonas Danica. J. Am. Chem. Soc. 2009,
131, 7570-7572.
28 Our group has previously established the stereochemical outcome for
a similar, truncated system through J-based configurational analysis,
see: Ref. 10. The desired stereochemical outcome is confirmed by
JBCA of the synthetic Mytilipin (2) and its diastereomer 4b (see SI for
details). The dichlorination formally inverts the (R)/(S) indication of
the stereocenter opened with retention in the previous step.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
39Fischer, S; Huwyler, N.; Wolfrum, S.; Carreira, E. M. Synthesis and
Biological Evaluation of Bromo- and Fluorodanicalipin A. Angew.
Chem. Int. Ed. 2016, 55, 2555–2558.
29 Ishihara, K.; Kurihara, H.; Yamamoto; H. An extremely simple, con-
venient, and selective method for acetylating primary alcohols in the
presence of secondary alcohols. J. Org. Chem. 1993, 58, 3791-3793.
The regioselectivity of the reaction could be verified by inspection of
1
the HMBC spectrum and the strong downfield H-NMR shift of the
methine proton at C1.
30 A strong 13C downfield shift of the adjacent carbon was distinct for
the O-sulfation and was used to determine the regioselectivity. No bis-
sulfation was observed.
31
(a) Sammakia, T.; Jacobs, J. S. Picolinic Acid as a Partner in the
Mitsunobu reaction: Subsequent Hydrolysis of Picolinate Esters under
Essentially Neutral Conditions with Copper Acetate in Methanol. Tet-
rahedron Lett. 1999, 40, 2685-2688. (b) Ju, Y. B.; Shin, Y.-J.; Heung,
B. J.; Kwan, S. K. Picolinyl group as an efficient alcohol protecting
group: cleavage with Zn(OAc)2·2H2O under a neutral condition
Tetrahedron Lett. 2005, 46, 5143-5147.
32 This compound has been previously prepared on an exploratory route
in our group: Dissertation Roger Geisser, Diss. ETH Nr. 19364, Zürich
2010.
33 Chloroacetate esters are highly labile to hydrolysis, however quanti-
tative closure to the epoxide was observed with even mildly basic Na-
HCO3 in MeOH at 60 °C or NH3 in MeOH at room temperature.
34 The natural product provided d6-acetone spectra that were highly de-
pendent on concentration, water content and counterion. The closest
match withthe 1H-NMR spectrum of 1 was observed with a Na+ coun-
terion (prepared by dissolving 2 in aq. NaHCO3 and subsequent lyoph-
ilization). Our group has previously established a sulfolipid to be pro-
tonated after silica column chromatography through elemental analysis
of bromodanicalipin: Dissertation Stefan Fischer, Diss. ETH Nr.
24978, Zürich 2018. This irreproducibility was less pronounced in pro-
tic solvents and consequently the HMQC of natural Mytilipin B
matches the HSQC of synthetic 2 in d4-MeOD very well. See also:
Sigala, P. A.; Ruben, E. A.; Liu, C. W.; Piccoli, P. M. B.; Hohenstein,
E. G.; Martinez, T. J.; Schultz, A. J.; Herzschlag, D. Determination of
Hydrogen Bond Structure in Water versus Aprotic Environments To
Test the Relationship Between Length and Stability. J. Am. Chem. Soc.
2015, 137, 5730-5740.
Insert Table of Contents artwork here
35
The DEPT spectrum provided by Ciminiello was assigned as the
DEPT spectrum of 3 in d6-acetone. However, the residual solvent peak
at 49.0 ppm suggested it to be recorded in d4-methanol and the match
with our 13C-NMR spectrum of 2 suggests that the spectrum provided
by Ciminiello was also recorded of the peracetate.
36
Ciminiello, P.; Dell’Aversano, C.; Fattorusso, E.; Forino, M.;
Magno, S.; Meglio. P.; Di, Ianaro, A.; Poletti, R. A New Cytotoxic
Polychlorinated Sulfolipid from Contaminated Adriatic Mussels. Tet-
rahedron, 2004, 60, 7093-7098.
37(a) Pereira, A. R.; Byrum, T.; Shibuya, G. M.; Vanderwal, C. D.;
Gerwick, W. H. Structure Revision and Absolute Configuration of
Malhamensilipin
A
from
the
Freshwater
Chrysophyte
10
ACS Paragon Plus Environment