Stereodivergent Synthesis and Configurational Assignment of the C1-C15 Segment of Amphirionin-5

Article abstract of DOI:10.1021/acs.joc.6b01700

The relative configuration of the C3-C12 portion of amphirionin-5, a novel marine polyketide with potent cell proliferation-promoting activity, was established by the stereodivergent synthesis of six diastereomeric model compounds and comparison of their NMR spectroscopic data with those reported for the natural product. This study led to the elucidation of the relative configuration between C4/C5 and C9/C12 and to the reassignment of the proposed configuration of the C9 position of amphirionin-5.

Full text of DOI:10.1021/acs.joc.6b01700

Subscriber access provided by Northern Illinois University
Article
Stereodivergent Synthesis and Configurational
Assignment of the C1–C15 Segment of Amphirionin-5
Moemi Kanto, Sota Sato, Masashi Tsuda, and Makoto Sasaki
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.6b01700 • Publication Date (Web): 26 Aug 2016
Downloaded from http://pubs.acs.org on September 1, 2016
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 free 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 accessible to all
readers and 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.
The Journal of Organic Chemistry 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 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Stereodivergent Synthesis and Configurational Assignment of the C1–C15 Segment of
Amphirionin-5
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
†
‡,§
||
,†
Moemi Kanto, Sota Sato, Masashi Tsuda, and Makoto Sasaki*
†
Graduate School of Life Sciences, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577,
Japan
‡
JST, ERATO, Isobe Degenerate
π-Integration Project and Advanced Institute for Materials
Research, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577, Japan
§
Department of Chemistry, Graduate School of Science, Tohoku University, 6-3 Aramaki-Aza-Aoba,
Aoba-ku, Sendai 980-8578, Japan
||
Faculty of Agriculture and Marine Science and Center for Advanced Marine Core Research,
Kochi University, Kochi 783–8502, Japan
Eꢀmail: masasaki@m.tohoku.ac.jp
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
TOC graphic
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
ACS Paragon Plus Environment

Page 3 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
ABSTRACT: The relative configuration of the C3–C12 portion of amphirioninꢀ5, a novel marine
polyketide with potent cell proliferationꢀpromoting activity, was established by the stereodivergent
synthesis of six diastereomeric model compounds and comparison of their NMR spectroscopic data
with those reported for the natural product. This study led to the elucidation of the relative
configuration between C4/C5 and C9/C12 and to the reassignment of the proposed configuration of
the C9 position of amphirioninꢀ5.
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
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
INTRODUCTION
Dinoflagellates of the genus Amphidinium are an enormously rich source of structurally diverse
secondary metabolites with complex molecular structures and potent biological activities. In
particular, more than 45 cytotoxic macrolides, amphidinolides and iriomoteolides, have been
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
isolated from Amphidinium sp. to date. Recently, the novel complex tetrahydrofuran
2
3,4
5
ringꢀcontaining linear polyketides, amphirioninsꢀ5 (1), ꢀ4 (2), and ꢀ2 (3), which exhibit
intriguing biological activities, were identified from Amphidinium sp. by Tsuda and coꢀworkers
(Figure 1). Of these, amphirioninꢀ5 (1) was isolated, along with cytotoxic macrolides
6
7
iriomoteolidesꢀ1a and ꢀ3a, from cultivated algal cells of the benthic dinoflagellate Amphidinium
2
sp. strain KCA09053 collected off the coast of Iriomote Island, Okinawa Prefecture, Japan. The
gross structure and partial relative configuration of amphirioninꢀ5 were assigned through extensive
8
2
DꢀNMR studies and Jꢀbased configurational analyses. Structurally, amphirioninꢀ5 consists of a
linear polyketide skeleton containing two tetrahydrofuran rings, a transꢀepoxide, and eleven
stereogenic centers. However, despite detailed NMR analysis, the relative configurations of the
C4/C5 stereogenic centers and the stereochemistry of the two isolated C12 and C26 stereogenic
centers could not be resolved, and the absolute configuration also remained unknown. In particular,
the remote stereogenic centers at C12 and C19 could not be correlated with each other.
4
ACS Paragon Plus Environment

Page 5 of 80
The Journal of Organic Chemistry
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 1. Structures of amphirioninsꢀ5 (1), ꢀ4 (2), and ꢀ2 (3).
Most importantly, this linear polyketide natural product exhibited potent cell
proliferationꢀpromoting activity on murine bone marrow stromal STꢀ2 cells (282%) and murine
osteoblastic MC3T3ꢀE1 cells (320%) at a dose of 10 ng/mL, and it did not induce cellular
differentiation or morphological changes in the dose range of 0.001–1000 ng/mL or exhibit
2
cytotoxicity at higher doses (1–10 ꢀg/mL). This intriguing biological profile of amphirioninꢀ5
suggests that it may be a promising candidate for the regenerative treatment of bone and joint
disease and for the prevention or treatment of osteoporosis. However, further studies on the
mechanism of action of amphirioninꢀ5 have been hampered not only by its limited availability
from the natural source, but also by the incomplete stereochemical assignment of its structure. A
synthetic approach is therefore required to address these issues.
The promising biological properties of amphirioninꢀ5, as well as its complex molecular structure
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and undefined stereochemistry, prompted our current efforts toward the total synthesis and
complete configurational assignment of the compound. We describe herein the assignment of the
relative configuration of the C3–C12 portion of amphirioninꢀ5 through stereodivergent synthesis of
six diastereomeric model compounds and comparison of their NMR spectroscopic data with those
reported for the natural product. Portions of this work have previously been published in a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
preliminary form.
RESULTS AND DISCUSSION
Stereochemical-Determination Strategy. Based on our previous work on the stereochemical
assignment of acyclic portions of the large polycyclic ether natural products maitotoxin and
10,11
prymnesins,
we predicted that the relative configurations of C4/C5 and C9/C12 of
amphirioninꢀ5 could be assigned by the synthesis of appropriate designed model compounds and
12
subsequent comparison of their NMR data with those reported for the natural product (Figure 2).
The first phase of the stereochemicalꢀdetermination of the C1–C15 segment of amphirioninꢀ5 was
the assignment of the unresolved relative configuration between C4 and C5 (Step 1). We therefore
decided to prepare two possible diastereomeric model compounds, (3S,4S,5R)ꢀ4 and (3R,4R,5R)ꢀ5,
of the C1–C12 segment of amphirioninꢀ5 using a stereodivergent approach and to compare their
NMR data with those reported for the natural product. Once we assigned the relative configuration
6
ACS Paragon Plus Environment

Page 7 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
of C4 and C5, the next phase of our studies relied on the synthesis of two possible diastereomers,
(9R,12R)ꢀ6 and (9R,12S)ꢀ7, of the C1–C15 segment with the assigned configuration at C3–C5 for
comparison of their NMR characteristics with that of the natural product (Step 2). Although the two
stereogenic centers at C9 and C12 are separated by two methylene units, differences in their
relative configurations should be detected as small but distinct variations in their NMR
characteristics, allowing the two possible diastereomers to be distinguished by currently available
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
1
0c,e,13
NMR spectroscopic techniques.
Figure 2. Stereochemicalꢀdetermination strategy for the C1–C15 portion.
Synthetic Plan for the C1–C12 Segment. Our retrosynthetic analysis of the two possible
diastereomeric model compounds (3S,4S,5R)ꢀ4 and (3R,4R,5R)ꢀ5 for the C1–C12 segment of
amphirioninꢀ5 is depicted in Scheme 1. We envisioned that the 2,5ꢀtransꢀsubstituted
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tetrahydrofuran ring of 4 and 5 could be constructed through a domino Sharpless asymmetric
1
4
dihydroxylation /stereospecific 5ꢀexo cyclization of mesylates (3S,4S)ꢀ8 and (3R,4R)ꢀ9,
15
respectively. The two diastereomeric epoxides 8 and 9 would be accessed by branching from a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16
common intermediate, allylic alcohol 10, by Katsuki–Sharpless asymmetric epoxidation using
(+)ꢀ or (–)ꢀtartrate ester as a chiral ligand. Allylic alcohol 10 would be obtained by means of
17
Corey–Bakshi–Shibata reduction of the precursor α,βꢀunsaturated ketone, which in turn would be
assembled from the three fragments isobutyraldehyde (11), aldehyde 12, and sulfone 13 through
18
Julia–Kocienski olefination and Horner–Wadsworth–Emmons reaction in a convergent fashion.
Scheme 1. Retrosynthetic Analysis of the C1–C12 Segments 4 and 5
Synthesis of Allylic Alcohol 10. The synthesis of allylic alcohol 10 started with the known imide
8
ACS Paragon Plus Environment

Page 9 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
9
20
1
4. The chiral auxiliary of 14 was reductively removed with LiBH
4
in THF/MeOH to provide
21
22
primary alcohol 15 in 92% yield (Scheme 2). Parikh–Doering oxidation of 15 afforded the
1
8
corresponding aldehyde 12, which was subjected to Julia–Kocienski olefination using the known
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
23
phenyltetrazolyl sulfone 13 and KHMDS in DME at –55 °C to produce (E)ꢀalkene 16 in 64%
yield for the two steps as a single stereoisomer (E/Z >20:1). Oxidative removal of the
pꢀmethoxybenzyl (PMB) group of 16 with DDQ provided primary alcohol 17 in 87% yield.
2
2
Parikh–Doering oxidation of 17 gave the corresponding aldehyde, which was reacted with the
lithium anion generated from dimethyl methylphosphonate using nꢀBuLi to provide βꢀhydroxy
phosphonate 18 in 81% yield (two steps) as a 1:1 diastereomeric mixture of alcohols. Subsequent
oxidation of 18 with tetraꢀnꢀpropylammonium perruthenate (TPAP)/Nꢀmethylmorpholine Nꢀoxide
2
4
(
NMO) delivered βꢀketo phosphonate 19 in 82% yield. Horner–Wadsworth–Emmons reaction of
25
1
2
9 with isobutyraldehyde (11) under Masamune–Roush conditions (LiCl, iꢀPr NEt, MeCN)
provided (E)ꢀα,βꢀunsaturated ketone 20 in nearly quantitative yield as a single stereoisomer (E/Z
17
>
20:1). Finally, Corey–Bakshi–Shibata reduction of 20 using (R)ꢀ2ꢀmethylꢀCBSꢀoxazaborolidine
2
3
1 (1.0 equiv) and BH ·THF (2.0 equiv) in THF at –40 °C furnished the desired allylic alcohol 10
26
in 93% yield. The diastereomer ratio (dr) of 10 was determined to be 11:1 by HPLC analysis of
the corresponding benzoate derivative 22. The absolute configuration of the newly generated
2
7
28
stereogenic center at C5 was unambiguously established by a modified Mosher analysis after
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
derivatizing 10 to the corresponding (S)ꢀ and (R)ꢀMTPA esters 23a and 23b, respectively (Figure
3).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthesis of Allylic Alcohol 10.
Figure 3. Determination of the absolute configuration at C5 of alcohol 10. The numbers in red and
1
blue are the difference (ꢁδ) in the H NMR chemical shifts between 23a and 23b (ꢁδ = δ (23a) – δ
10
ACS Paragon Plus Environment

Page 11 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
(
3
23b) in CDCl ).
Stereodivergent Synthesis of the C1–C12 Segments 4 and 5. With the requisite allylic alcohol 10
in hand, we proceeded to the stereodivergent synthesis of the diastereomeric C1–C12 segments 4
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
16
and 5. Katsuki–Sharpless asymmetric epoxidation of allylic alcohol 10 using (+)ꢀdiisopropyl
tartrate (DIPT) as a chiral ligand provided a “matched” case, and allylic alcohol 10 (dr 11:1)
smoothly underwent epoxidation to afford the desired epoxy alcohol 24 in 90% yield with high
diastereoselectivity (dr ca. 23:1) after 2.5 h (Scheme 3). Alcohol 24 was then converted to the
3
corresponding mesylate 8 (MsCl, Et N), which was subjected to Sharpless asymmetric
14
dihydroxylation
using ADꢀmixꢀβ Diastereoselective dihydroxylation with concomitant
stereospecific 5ꢀexo cyclization took place to form a tetrahydrofuran ring, and the desired C1–C12
segment (3S,4S,5R)ꢀ4 was obtained in 90% yield for the two steps. The relative configuration of the
2,5ꢀtransꢀsubstituted tetrahydrofuran ring in 4 was confirmed by means of HMBC correlations and
NOE data, and the absolute configuration of the C9 stereogenic center was unambiguously
established by derivatization of 4 to the corresponding (S)ꢀ and (R)ꢀMTPA esters 25a and 25b,
28
respectively, and a modified Mosher analysis (Figure 4).
16
Although the Katsuki–Sharpless asymmetric epoxidation is well recognized as a reliable
asymmetric reaction, the relative configurations of the C3–C5 positions of epoxy alcohol 24 were
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
further confirmed by NMR analysis of a suitable tetrahydropyran derivative. Thus, Sharpless
14
asymmetric dihydroxylation of 24 using ADꢀmixꢀβ provided triol 26 (87%), which upon treatment
with PPTS (0.1 equiv) in (CH Cl) at 40 °C induced a 6ꢀexo epoxide ringꢀopening cyclization to
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
form tetrahydropyran 27 in 57% yield (Scheme 4). The relative configuration of 27 was
3
unambiguously established by means of HMBC spectra, NOE analysis, and JH,H data, as shown in
Figure 5.
Scheme 3. Synthesis of the C1–C12 Segment (3S,4S,5R)-4
12
ACS Paragon Plus Environment

Page 13 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Figure 4. Stereochemical assignment of alcohol 4. Singleꢀheaded arrows indicate HMBC
correlations and doubleꢀheaded dashed arrows denote key NOEs. The numbers in red and blue are
1
the difference (ꢁδ) in the H NMR chemical shifts between 25a and 25b (ꢁδ = δ (25a) – δ (25b) in
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
3
CDCl ).
Scheme 4. Synthesis of Tetrahydropyran Derivative 27
Figure 5. Stereochemical assignment of tetrahydropyran 27. Singleꢀheaded arrows indicate HMBC
correlations and doubleꢀheaded dashed arrows denote key NOEs.
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The diastereomeric C1–C12 segment (3R,4R,5R)ꢀ5 was prepared using the same sequence of
reactions from allylic alcohol 10 via epoxy alcohol 28 (Scheme 5). In this case, Katsuki–Sharpless
1
6b,c
asymmetric epoxidation of 10 using (–)ꢀDIPT provided a typical “mismatched” pair,
and thus
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the starting allylic alcohol 10 in a diastereomerically enriched form (dr 22:1 at C5) obtained by
kinetic resolution was used for asymmetric epoxidation. In the presence of (–)ꢀDIPT, the
epoxidation of allylic alcohol 10 was much slower than that using (+)ꢀDIPT (2.5 h) and required a
prolonged reaction time (15.5 h) for consumption of the starting material, and a 1.4:1 mixture of
diastereomeric epoxides 28 and 24 was obtained. This mixture of epoxides was readily separable by
flash column chromatography on silica gel to afford the desired epoxide 28 in 55% yield with a
diastereomer ratio of 22:1, along with 24 (40%, dr 27:1). Mesylation of 28, followed by oneꢀpot
1
4
Sharpless asymmetric dihydroxylation /5ꢀexo cyclization of the resultant mesylate 9, furnished the
desired (3R,4R,5R)ꢀ5 in 86% yield for the two steps. The relative configuration of the
tetrahydrofuran ring moiety of 5 was confirmed by NMR analysis (HMBC and NOEs), and the
absolute configuration of the C9 stereogenic center was established by the modified Mosher
28
method using the corresponding (S)ꢀ and (R)ꢀMTPA esters 29a and 29b, as shown in Figure 6.
Scheme 5. Synthesis of the Diastereomeric C1–C12 Segment (3R,4R,5R)-5
14
ACS Paragon Plus Environment

Page 15 of 80
The Journal of Organic Chemistry
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 6. Stereochemical assignment of alcohol 5. Singleꢀheaded arrow indicates HMBC
correlation and doubleꢀheaded dashed arrows denote key NOEs. The numbers in red and blue are
1
the difference (ꢁδ) in the H NMR chemical shifts between 29a and 29b (ꢁδ = δ (29a) – δ (29b) in
3
CDCl ).
1
13
NMR Comparison of Compounds 4 and 5 with the Natural Product. The H and C NMR
chemical shifts in the C1–C9 region of the two diastereomeric model compounds 4 and 5 thus
29
obtained were compared with those of the corresponding moiety of the natural product. As shown
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
in Figure 7A, the H NMR chemical shifts in the C1–C6 region of 4 were virtually identical with
those reported for the natural product (|ꢁδ| <0.02 ppm), whereas significant deviations of chemical
30
13
shifts (|ꢁδ| >0.06 ppm) were observed for diastereomer 5 in the C3–C5 region. Similarly, the C
NMR chemical shifts in the C1–C6 region of 4 were in good agreement with those of the natural
product (|ꢁδ| <0.7 ppm), whereas compound 5 displayed obviously different chemical shifts; in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
particular, the observed C NMR chemical shifts for C5 and C6 of 3 significantly deviated from
3
0
those of the natural product by over 1.0 ppm (Figure 7B). These results strongly suggested that the
relative configuration of the C3–C5 portion of amphirioninꢀ5 is represented by structure 4. However,
1
13
we were surprised to find that there were significant and similar discrepancies in the H and C
NMR chemical shifts in the rightꢀhand C7–C9 region for both compounds 4 and 5. In particular, the
1
13
largest deviations in the H and C NMR chemical shifts were observed for Hꢀ9 (ꢁδ = 0.27 ppm),
C9 (ꢁδ = 1.9 ppm), and the attached C30 methyl group (ꢁδ >1.3 ppm). From these large deviations
in the NMR chemical shifts between compounds 4/5 and the natural product, we inferred that the
C9 stereogenic center of amphirioninꢀ5 might have been incorrectly assigned and that the most
likely configuration of the C9 stereogenic center of amphirioninꢀ5 is inverted, as represented by the
revised structure 30 (Scheme 6).
16
ACS Paragon Plus Environment

Page 17 of 80
The Journal of Organic Chemistry
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
1
13
Figure 7. Differences in H and C NMR chemical shifts between amphirioninꢀ5 (500 and 125
MHz, respectively) and model compounds 4 and 5 (600 and 150 MHz, respectively). The xꢀ and
yꢀaxes represent the carbon number and ꢁδ = δ (natural product) – δ (model compound) in ppm
(
CDCl ), respectively.
3
Synthesis of the Diastereomers 30 and 31 and their NMR Comparison with the Natural
Product. Thus, the C9 hydroxy group of 4 and 5 was inverted using modified Mitsunobu
31
2 6 4 2 3
conditions (pꢀNO C H CO H, Ph P, diisopropyl azodicarboxylate (DIAD), THF) (Scheme 6).
2 3
Methanolysis of the resultant pꢀnitrobenzoates 32 and 33 (K CO , MeOH) led to alcohols
(
3S,4S,5R,9S)ꢀ30 and (3R,4R,5R,9S)ꢀ31, respectively.
1
13
The H and C NMR spectroscopic data of 30 and 31 were once again compared with those
1
13
reported for the natural product. As expected, the H and C NMR chemical shifts in the C1–C9
region for diastereomer 30 were virtually identical to those reported for the natural product (Figure
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
0
1
13
8
). In contrast, significant differences in the H and C NMR chemical shifts were observed for
other diastereomer 31 in the C3–C6 region, as was the case for compound 5 (Figure 7). In addition,
3
30
JH,H data of the C1–C9 portion of 30 corresponded well with the data of amphirioninꢀ5. These
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
results convincingly defined the relative configuration of the C1–C9 portion of amphirioninꢀ5 as
that represented by structure 30 with the (3S*,4S*,5R*,7R*,8R*,9S*)ꢀstereochemistry.
The relative configuration of the C8/C9 stereogenic centers of the natural amphirioninꢀ5 had
8
been elucidated to be 8R*,9R* applying Jꢀbased configuration analysis mainly based on coupling
3
2
2
2
3
constants ( JHꢀ8,Hꢀ9 = 4.1 Hz, JCꢀ8,Hꢀ9 = –7 Hz, and JCꢀ9,Hꢀ8 = –6 Hz). Because both of the JHꢀ8,Hꢀ9
values (3.2 and 4.6 Hz) for (8R,9R)ꢀ4 and (8R,9S)ꢀ30, respectively, were not completely different
2
from that for the natural amphirioninꢀ5 (4.1 Hz), elucidation of one of the other, or both of JCꢀ8,Hꢀ9
2
and JCꢀ9,Hꢀ8 values from the HETLOC spectrum of the natural amphirioninꢀ5 might be wrong.
Nevertheless, the remeasurement of the HETLOC spectrum of amphirioninꢀ5 was not possible
because of a lack of the natural sample.
Scheme 6. Synthesis of the Diastereomers 30 and 31
18
ACS Paragon Plus Environment

Page 19 of 80
The Journal of Organic Chemistry
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
1
13
Figure 8. Differences in H and C NMR chemical shifts between amphirioninꢀ5 (500 and 125
MHz, respectively) and model compounds 30 and 31 (600 and 150 MHz, respectively). The xꢀ and
yꢀaxes represent the carbon number and ꢁδ = δ (natural product) – δ (model compound) in ppm
3
(CDCl ), respectively.
Synthesis of the Diastereomeric C1–C15 Segments 34 and 35. Having established the relative
configuration of the C3–C9 portion of amphirioninꢀ5, we next sought to determine the unexplored
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
configuration of the remote C12 stereogenic center. For this purpose, we proceeded with the
synthesis of two possible diastereomeric model compounds, (9S,12R)ꢀ34 and (9S,12S)ꢀ35, for the
C1–C15 segments (Scheme 7). We originally planned to synthesize these two compounds by means
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
32
of an acetate aldol reaction of aldehyde 36 with a metal enolate derived from either the (R)ꢀ or
(S)ꢀenantiomer of Nꢀacetylꢀ4ꢀisopropylꢀ1,3ꢀthiazolidineꢀ2ꢀthione (37) developed by Nagao and
3
3
coꢀworkers.
Scheme 7. Initial Synthesis Plan for the C1–C15 Segments 34 and 35
The synthesis of aldehyde 36 commenced with alcohol 30, which was protected as its TBS ether
with TBSCl/imidazole in DMF at 50 °C to give bisꢀTBS ether 38 in 98% yield (Scheme 8).
Selective cleavage of the primary TBS ether of 38 under acidic conditions (CSA (0.1 equiv),
2 2
MeOH/CH Cl (1:1), 0 °C) led to primary alcohol 39 in 78% yield, along with diol 40 (20%).
34
Dess–Martin oxidation of 39 provided aldehyde 36 in 97% yield.
20
ACS Paragon Plus Environment

Page 21 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Scheme 8. Synthesis of Aldehyde 36
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
We
initially
attempted
an
aldol
reaction
of
36
with
3
3
(
R)ꢀNꢀacetylꢀ4ꢀisopropylꢀ1,3ꢀthiazolidineꢀ2ꢀthione ((R)ꢀ37) under Vilarrasa’s conditions (TiCl4,
3
5
2 2 2
iꢀPr NEt, CH Cl , –78 °C). However, the epoxide ring of 36 was extremely labile under Lewis
acidic conditions, resulting in a moderate yield of the desired aldol products 41a,b with low
diastereoselectivity, and formation of the epoxide ringꢀopening products 42a,b as the major
components (Scheme 9).
Scheme 9. Unsuccessful Acetate Aldol Reaction of 36
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In order to suppress the lability of the epoxide ring in 36, we next selected a
diastereoselective aldol reaction using 2ꢀacetoxyꢀ1,1,2ꢀtriphenylethanol 43 developed by Braun et
36
al. under basic conditions. Thus, diastereoselective aldol reaction of 36 with the lithium enolate
derived from (S)ꢀ43 using 2 equiv of LDA in THF at –78 °C gave the desired βꢀhydroxy ester 44
(Scheme 10). As the diastereomer ratio could not be determined at this point, ester 44 thus obtained
2 3
was further transesterified to the corresponding methyl ester 45 with K CO /MeOH in 78% yield
for the two steps. At this stage, the diastereomer ratio of the aldol reaction was estimated to be 10:1
1
by integration of the SiMe signal in the H NMR spectrum. Methyl ester 45 was converted to the
corresponding Weinreb amide 46 in 91% yield by reaction with N,Oꢀdimethylhydroxylamine
3
7
hydrochloride and nꢀBuLi (THF, –78 °C). Direct conversion of 44 to the corresponding Weinreb
amide 46 was also achieved in good yield (71%), but large excess amounts of MeNH(OMe)·HCl
(10 equiv) and nꢀBuLi (20 equiv) were required for complete consumption of 44. Therefore, ester
22
ACS Paragon Plus Environment

Page 23 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
4
4 was converted to amide 46 by a twoꢀstep sequence of reactions. Subsequent treatment of 46
3
8
with methyllithium (THF, –78 °C) provided methyl ketone 47 in 89% yield. The absolute
configuration of the C12 stereogenic center, which was newly generated in the aldol reaction, was
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
unambiguously established by derivatization of 47 to the corresponding (S)ꢀ and (R)ꢀMTPA esters
28
4
8a and 48b, respectively, and a modified Mosher analysis (Figure 9). Finally, cleavage of the
3
9
TBS ether of 47 using tris(dimethylamino)sulfonium difluorotrimethylsilicate (TASF) (DMF/H
2
O,
0 °C) furnished the desired C1–C15 segment (9S,12R)ꢀ34 in 85% yield after HPLC purification.
Scheme 10. Synthesis of the C1–C15 Segment 34 through Braun Acetate Aldol Reaction
Figure 9. Determination of the absolute configuration at C12 of alcohol 47. The numbers in red and
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
blue are the difference (ꢁδ) in the H NMR chemical shifts between 48a and 48b (ꢁδ = δ (48a) – δ
3
(48b) in CDCl ).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The other diastereomeric C1–C15 segment, (9S,12S)ꢀ35, was prepared in a similar manner
from aldehyde 36, as summarized in Scheme 11. Aldol reaction of 36 with the lithium enolate
2 3
derived from (R)ꢀ43 (LDA (2 equiv), THF, –78 °C) followed by methanolysis (K CO , MeOH)
provided βꢀhydroxy ester 50 (64%, two steps), which was converted to methyl ketone 52 via
Weinreb amide 51 in 97% yield for the two steps with a 7:1 diastereomer ratio at C12. The absolute
configuration of the C12 stereogenic center of 52 was established by the modified Mosher
28
method using the corresponding (S)ꢀ and (R)ꢀMTPA esters 53a and 53b, as shown in Figure 10.
4
0
Removal of the TBS group of 52 with TBAF gave rise to the requisite C1–C15 segment
(
9S,12S)ꢀ35 in 84% yield after HPLC purification.
Scheme 11. Synthesis of the Diastereomeric C1–C15 Segment 35
24
ACS Paragon Plus Environment

Page 25 of 80
The Journal of Organic Chemistry
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 10. Determination of the absolute configuration at C12 of alcohol 52. The numbers in red
1
and blue are the difference (ꢁδ) in the H NMR chemical shifts between 53a and 53b (ꢁδ = δ (53a)
–
3
δ (53b) in CDCl ).
1
Assignment of the Relative Configuration of the C1–C15 Segment of Amphirionin-5. The H
1
3
and C NMR chemical shifts in the region of C1–C12 for the two diastereomeric model
compounds 34 and 35 were compared with the data reported for the natural product. As shown in
1
Figure 11, diastereomer 34 displayed NMR chemical shifts, and particularly H NMR shifts, that
were virtually identical to those reported for the natural product. In contrast, upon careful
examination of the NMR data of diastereomer 35, small but distinct differences in the chemical
1
shifts were detected between 35 and the natural product. Remarkably, distinct deviations in the H
13
and C NMR chemical shifts were observed for Hꢀ11 (ꢁδ = 0.09 and –0.04 ppm) and C10 (ꢁδ =
–
0.6 ppm) that are deemed sufficiently so as to be distinguishable. It is critical that these protons
1
0e
and carbon are located in the region bridging by the two stereogenic centers at C9 and C12.
These results conclusively demonstrated that compound 34 represents the relative configuration of
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
1
the corresponding portion of the natural product amphirioninꢀ5. Consequently, we assigned the
relative configuration of the C3–C12 portion of amphirioninꢀ5 as 3S*,4S*,5R*,7R*,8R*, 9S*,12R*
(Figure 12).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
13
Figure 11. Differences in H and C NMR chemical shifts between amphirioninꢀ5 (500 and 125
MHz, respectively) and model compounds 34 and 35 (600 and 150 MHz, respectively). The xꢀ and
yꢀaxes represent the carbon number and ꢁδ = δ (natural product) – δ (model compound) in ppm
3
(CDCl ), respectively.
Figure 12. Assigned relative configuration of the C3–C12 portion of amphirioninꢀ5 (1).
26
ACS Paragon Plus Environment

Page 27 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
CONCLUSIONS
In summary, we assigned the relative configuration of the C3–C12 portion of amphirioninꢀ5 as
3
S*,4S*,5R*,7R*,8R*,9S*,12R* by stereodivergent synthesis of six diastereomeric model
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
compounds and by carefully comparing their NMR data with those reported for the natural product.
Four diastereomeric model compounds for the C1–C12 segment of amphirioninꢀ5 were synthesized
in a stereodivergent fashion. The key features of the synthesis route included (1) convergent
synthesis of the common intermediary allylic alcohol by exploiting Julia–Kocienski olefination,
Horner–Wadsworth–Emmons reaction, and Corey–Bakshi–Shibata reduction, (2) efficient
construction of the 2,5ꢀtransꢀsubstituted tetrahydrofuran ring by a domino Sharpless asymmetric
dihydroxylation/stereospecific 5ꢀexo cyclization, and (3) Mitsunobu inversion reaction. Comparison
of the NMR data of the four diastereomeric model compounds with those reported for the natural
product enabled not only assignment of the relative configuration of the C4/C5 stereogenic centers,
but also reassignment of the originally proposed configuration at C9 of amphirioninꢀ5. Furthermore,
synthesis of two possible diastereomeric model compounds for the C1–C15 segment through an
acetate aldol reaction and comparison of their NMR data with those of the natural product defined
the relative configuration of the remote stereogenic centers at C9 and C12 bridged by two
methylene units. Further studies aimed at the complete stereochemical assignment and total
synthesis of amphirioninꢀ5 are underway and will be reported in due course.
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 80
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
EXPERIMENTAL SECTION
General Methods. All reactions sensitive to moisture and/or air were carried out under an
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
atmosphere of argon in dry, freshly distilled solvents under anhydrous conditions using ovenꢀdried
2 2
glassware unless otherwise noted. Anhydrous CH Cl was purchased and anhydrous THF was
purified by a Glass Contour solvent purification system. Acetonitrile (MeCN), 1,2ꢀdichloroethane
(DCE), diisopropylamine, diisopropylethylamine, 1,2ꢀdimethoxyethane (DME), pyridine, and
3
triethylamine (Et N) were distilled from calcium hydride under an atmosphere of argon. DMF and
DMSO were distilled from magnesium sulfate under reduced pressure. All other chemicals were
purchased at highest commercial grade and used directly. Analytical thinꢀlayer chromatography
(TLC) was performed using preꢀcoated glass plate (silica gel 60 F254, 0.25ꢀmm thickness). Flash
column chromatography was carried out using silica gel (spherical, neutral, 40–100 mesh; granular,
200–400 mesh). Reverseꢀphase HPLC was performed using an UV/Visible detector. Optical
rotations were measured on a digital polarimeter at 589 nm. IR spectra were recorded as a thin film
–1
1
13
on a KBr disk using an FTꢀIR spectrometer and reported in wavenumbers (cm ). H and C NMR
spectra were recorded at 600 and 150 MHz NMR spectrometers, respectively. Chemical shift
values are reported in ppm (δ) downfield from tetramethylsilane with reference to internal residual
1
13
solvent [ H NMR, CHCl (7.26), C HD (7.16); C NMR, CDCl (77.0), C D (128.0)]. Coupling
3
6
5
3
6
6
28
ACS Paragon Plus Environment

Page 29 of 80
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
constants (J) are reported in hertz (Hz). The following abbreviations were used to designate the
multiplicities: s = singlet; d = doublet; t = triplet; q = quartet; m = multiplet or unresolved; br =
broad. Highꢀresolution mass spectra (HRMS) were measured on an ESIꢀTOF mass spectrometer.
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
1
Diastereomer ratio (dr) and E/Z isomer ratio were estimated by H NMR spectroscopic analysis,
unless otherwise noted.
19
(R)-4-((4-Methoxybenzyl)oxy)-2-methylbutan-1-ol (15). To a solution of imide 14 (8.68 g,
4
21.9 mmol) and MeOH (2.70 mL, 66.7 mmol) in THF (200 mL) at 0 °C was added LiBH (3 M
solution in THF, 22 mL, 66 mmol). The resultant solution was stirred at 0 °C for 1 h. The reaction
was carefully quenched with saturated potassium sodium tartrate solution and Et O at 0 °C. The
2
2
resultant mixture was diluted with Et O and saturated aqueous potassium sodium tartrate solution
and vigorously stirred at room temperature. The reaction mixture was extracted with EtOAc, and
2 4
the organic layer was washed with H O and brine, dried over MgSO , filtered, and concentrated
under reduced pressure. Purification of the residue by column chromatography (silica gel, 20 to 40
to 50% EtOAc/hexanes) gave alcohol 15 (4.50 g, 92%) as a colorless oil. The spectroscopic data of
2
1
5 were compared with those the earlier known compound 1 and found to be identical.
(R,E)-tert-Butyl((8-((4-methoxybenzyl)oxy)-6-methyloct-4-en-1-yl)oxy)dimethylsilane (16).
29
ACS Paragon Plus Environment