Synthesis of (-)-Piperitylmagnolol Featuring ortho-Selective Deiodination and Pd-Catalyzed Allylation

Article abstract of DOI:10.1021/acs.orglett.6b00706

A 1,4-addition strategy using an enone and a copper reagent was studied for the synthesis of (-)-piperitylmagnolol. A MOM-protected biphenol copper reagent was added to BF₃·OEt₂-activated 4-isopropylcyclohexenone, whereas 1,4-addition of protected monophenol reagents possessing an allyl group was found to be unsuccessful. The allyl group was later attached to the p-,p′-diiodo-biphenol ring by Pd-catalyzed coupling with allylborate. The aforementioned iodide was synthesized using a new method for ortho-selective deiodination of o-,p-diiodophenols.

Full text of DOI:10.1021/acs.orglett.6b00706

Letter
pubs.acs.org/OrgLett
Synthesis of (−)-Piperitylmagnolol Featuring ortho-Selective
Deiodination and Pd-Catalyzed Allylation
Atsushi Ikoma, Narihito Ogawa, Daiki Kondo, Hiroki Kawada, and Yuichi Kobayashi*
Department of Bioengineering, Tokyo Institute of Technology, Box B-52, Nagatsuta-cho 4259, Midori-ku, Yokohama 226-8501,
Japan
S
* Supporting Information
ABSTRACT: A 1,4-addition strategy using an enone and a copper
reagent was studied for the synthesis of (−)-piperitylmagnolol. A
MOM-protected biphenol copper reagent was added to BF3·OEt2-
activated 4-isopropylcyclohexenone, whereas 1,4-addition of pro-
tected monophenol reagents possessing an allyl group was found to
be unsuccessful. The allyl group was later attached to the p-,p′-
diiodo-biphenol ring by Pd-catalyzed coupling with allylborate. The
aforementioned iodide was synthesized using a new method for
ortho-selective deiodination of o-,p-diiodophenols.
Scheme 1. 1,4-Addition Strategy Using Mono- or Bis-Aryl
Copper Reagents
he medicinal plant Magnolia officinalis produces piper-
1
Titylmagnolol (1) (Figure 1), which exhibits inhibitory
activity against Epstein−Barr virus antigen activation,1a
antibacterial activity against vancomycin-resistant enterococci
(VRE) and methicillin-resistant Staphylococcus aureus
(MRSA),2a and moderate cytotoxic activity against several
cancer cell lines.2b These properties are highly attractive from a
pharmaceutical perspective. To the best of our knowledge,
there has been only one report on the total synthesis of 1.3
However, this synthesis would be insufficient for primary
biological studies, such as investigating the structure−activity
relationships of 1 using structurally related compounds, because
analogues to be synthesized would be limited to those based on
camphor, which was a starting compound.
install the allyl group was uncertain because the acidic CH2 in
the allyl group restricted the catalogue of compatible reactions.
Herein, synthesis of 1 along this line is presented, featuring (1)
the formation of an iodine-containing biphenol intermediate;
(2) the ortho-selective deiodination of an o-,p-diiodophenol
derivative; and (3) Pd-catalyzed allylation of the resulting p-,p′-
diiodophenol derivative.
Figure 1. Piperitylmagnolol and tetrahydrocannabinol.
First, we examined monoaryl copper reagents 8a and 8b
possessing an allyl group and an allyl equivalent, respectively
(Figure 2). The attempted preparation of copper reagent 8a via
the Br−Li exchange of bromide 7a with t-BuLi was
unsuccessful, and the corresponding monobromide was not
isolated after aqueous quenching (Scheme 2, eq 1), indicating
decomposition of the bromide before formation of the copper
Although the 1,4-addition of sterically bulky copper reagents
to BF3·OEt2-activated enones produces only marginally reactive
boron enolates, we recently developed a method to activate
such enolates using MeLi, which allows phosphorylation to
furnish enol phosphates.4 This method was successfully utilized
for the synthesis of tetrahydrocannabinol (2), which has
analgesic effects. Owing to the structural similarities between 1
and 2, we envisioned the synthesis of 1 from an enol phosphate
derived from enone 3 and an arylcopper reagent of the
monoaryl type 4 or bis-aryl type 6 (Scheme 1), though a step to
Received: March 11, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b00706
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. 1,4-Addition of Copper Reagents
Figure 2. Copper reagents and their bromide precursors.
reagent with CuCN.5 In contrast, Br−Li exchange of 7d with t-
BuLi and subsequent reaction with CuCN afforded copper
reagent 8d, which underwent 1,4-addition to racemic enone
rac-3. Subsequent phosphorylation of the resulting enolate with
ClP(O)(OEt)2 produced enol phosphate 10d in 74% yield (eq
4). The reaction of bromide 7b, containing a masked allyl
group, gave a mixture of unidentifiable products (eq 1),
whereas model bromide 7c was successfully converted to
copper reagent 8c. However, the subsequent 1,4-addition of 8c
to enone rac-3 did not take place (eq 2). In contrast, the 1,4-
addition to BF3·OEt2-activated rac-3 proceeded well, although
the resulting boron enolate 9c was almost unreactive toward
ClP(O)(OEt)2. As a result, instead of 10c, ketone 11c was
obtained in 67% yield after aqueous quenching (eq 3). The Br
atom was likely directing the Me group on the MeO substituent
to occupy the space near the Cu atom in 8c, thus increasing the
steric bulk around the 1,4-addition center. Overall, the 1,4-
addition strategy using monoaryl copper reagents possessing
the allyl group or its equivalent, (CH2)3OR, was unsuccessful.
Hence, we turned our attention to the alternative strategy using
bis-aryl copper reagent (6), knowing that it would be feasible to
install the allyl group at a later stage in the synthesis.
Bromide 7e, which contains no allyl groups, was synthesized
from 2,2′-biphenol by MOM protection and monobromination
as described in the Supporting Information. Copper reagent 8e,
prepared by Br−Li exchange of 7e with t-BuLi and subsequent
reaction with CuCN, underwent a 1,4-addition to BF3·OEt2-
activated enone rac-3. The resulting boron enolate rac-9e was
treated with MeLi before phosphorylation with ClP(O)(OEt)2
to furnish enol phosphate rac-10e in 70% isolated yield (eq 5).6
The stereochemistry was tentatively assigned as trans based on
4
1
previous examples of 1,4-addition and later confirmed by H
NMR spectroscopy of the final compound.
With the synthesis of enol phosphate rac-10e in hand, we
continued the synthesis of (−)-1 using 3 with 81−89% ee,
favoring the (S)-configuration. Enone 3 was prepared via the
previously reported asymmetric Michael reaction of methyl
vinyl ketone with isovaleraldehyde.7 The enantiomeric purity
was increased at a later stage in the synthesis. As shown in
Scheme 3, the 1,4-addition of 8e to 3 (84% ee), followed by
phosphorylation, gave 10e in a comparable yield (73%) to that
of rac-3. Nickel-catalyzed methylation of 10e with MeMgCl
afforded 12 in 91% yield.8,9 Recrystallization of the solid mass
of 12 followed by concentration of the mother liquor afforded
95% ee of 12 with 77% recovery (70% yield from 10e).
Attempted iodinations of 12 resulted in a complex mixture (I2
and AcOAg) or starting material recovery (KI, m-CPBA, 18-
crown-6).10 Alternatively, 12 was hydrolyzed to biphenol
derivative 13 in good yield. Iodination of 13 using NaI and
oxidants (NaOCl, m-CPBA/18-crown-6) gave an unidentifiable
mixture of products or the starting compound, whereas NIS
unexpectedly produced tetraiodofuran 14 in 74% yield from
12.11 Attempted deiodination of 14 at the ortho position using
aqueous amines such as pyridine and N-Me-morpholine
according to the literature12 gave unidentifiable mixtures of
products. In contrast, the reaction of 14 with BBr3, Bu4NI, and
pyridine successfully afforded triiodide 15 in good yield. In
comparison to the reported conversion of 1-iodo-2-alkoxides to
B
DOI: 10.1021/acs.orglett.6b00706
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Synthesis of (−)-1
1-olefins using Me2BBr and Bu4NI,13 the present method was
advantageous because of the ready availability of BBr3.14
With triiodide 15 in hand, we envisioned ortho-selective
deiodination of 15 through an intramolecular hydride attack,
which was inspired by the LiAlH4 reduction of iodobenzene to
benzene.15 However, the LiAlH4 reduction of model iodide 20
at 0 °C proceeded slowly and showed low regioselectivity (21 >
22, 23) (Table 1, entry 1). Hydride reduction was then
Scheme 4. Regioselective Deiodination
Table 1. Deiodination of Model Diiodide
ortho position did not affect the deiodination of 24−26,
affording 30−32 in good yields. Protected phenols 27−29 were
also good substrates for ortho-selective deiodination, affording
33−35. In general, the iodination of phenols affords o-,p-
diiodophenols and/or p-iodophenols depending on the
reaction conditions.10,12,16 The present Red-Al deiodination
would be useful in cases where the iodination of phenols results
in polyiodination.
The above deiodination was successfully applied to triiodide
15 to afford diiodide 16. Next, Pd-catalyzed allylation toward
the final product was attempted using allylborate 17, according
to a literature procedure.17 However, this allylation afforded an
unidentifiable mixture of products. We proposed that the Pd
catalyst activity was reduced by 16 chelating with the Pd atom
as a bidentate ligand. Hence, 16 was silylated to give monosilyl
ether 18 in 77% yield from 14. The site of the TBS group was
assigned tentatively on the basis of steric effects. The allylation
of 18 with borane 17 was performed using Pd(PPh3)4 or
Pd(PPh3)4/dppf as catalyst. TLC analysis indicated a somewhat
clean coupling using the latter, affording 19 in 74% yield.
Finally, desilylation of 19 furnished (−)-1 in 69% yield. The 1H
and 13C NMR spectra and the specific rotation were
a
b
c
entry
reagent
solvent
time, h
21:22:23:20
yield, %
d
1
2
3
4
5
LiAlH4
DIBAL
Red-Al
Red-Al
Red-Al
Et2O
0.5
0.5
0.5
0.5
3
30:18:8:44
0:0:0:100
96:0:4:0
nd
Et2O
−
79
Et2O
CH2Cl2
CH2Cl2
97:0:0:3
89
d
96:0:3:1
nd
a
b
0 °C for 30 min with 3−3.4 equiv of reagents. Determined by
calculating signals in NMR spectra. Isolated yield of 21. Not
c
d
determined.
investigated using various hydride reagents to identify
conditions for selective synthesis of 21 (see Table S1 in the
Supporting Information for metalation of 20 with n-, s-, t-BuLi,
and i-PrMgCl·LiCl followed by aqueous quenching and metal-
catalyzed deiodination). Briefly, DIBAL was not productive
(entry 2), whereas Red-Al gave 21 selectively (entries 3 and 4),
with CH2Cl2 giving better product selectivity and isolated yield
than Et2O. Further deiodination of 21 in the para position was
hardly observed even after reaction times of more than 30 min.
The scope of this reaction was briefly investigated with results
shown in Scheme 4. The presence of an alkyl group at the
C
DOI: 10.1021/acs.orglett.6b00706
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
comparable with the literature data.1a,b,2b,3 Furthermore, the
13C−APT (attached proton test) NMR spectrum clearly
differentiated all the aromatic carbons including the signals of
the quaternary carbon (δ 131.0 ppm) and others nearby at δ
130.9 and 131.2 ppm as shown in the Supporting Information.
In summary, the synthesis of (−)-piperitylmagnolol (1) is
presented, featuring (1) activation of the boron enolate,
product of the 1,4-addition of MOM-protected biphenol
copper reagent 8e to the BF3·OEt2-activated enone 3, to afford
the enol phosphate 10e; (2) the subsequent intriguing
transformation to triiodophenol 15; and (3) the new finding
of the ortho-selective deiodination of 15 using Red-Al. A total
yield including synthesis of 3 from isovaleraldehyde was 5.3%
over 11 steps, whereas that from (+)-camphor3 was 1.6% over
11 steps. The scope of this deiodination was also briefly
explored.
(8) (a) Hayashi, T.; Fujiwa, T.; Okamoto, Y.; Katsuro, Y.; Kumada,
M. Synthesis 1981, 1981, 1001. (b) Sahlberg, C.; Quader, A.; Claesson,
A. Tetrahedron Lett. 1983, 24, 5137. (c) William, A. D.; Kobayashi, Y. J.
Org. Chem. 2002, 67, 8771. (d) Kobayashi, Y.; Takeuchi, A.; Wang, Y.-
G. Org. Lett. 2006, 8, 2699.
(9) This product was also synthesized through 1,4-addition of a
monoaryl copper reagent to rac-3: (a) rac-3 with Li2Cu(CN)(2-
MOMOC6H4)2 then ClP(O)(OEt)2, 82%; (b) MeMgCl, Ni(acac)2
(cat.), 89%; (c) t-BuLi, ZnCl2 then Pd(PPh3)2Cl2 (cat.)/DIBAL, 2-I-
C6H4OMOM, 42% (three steps, 31%; cf. Scheme 2, two steps via 10a,
66%).
(10) Edgar, K. J.; Falling, S. N. J. Org. Chem. 1990, 55, 5287.
(11) Similar iodocyclization: (a) Miyagawa, T.; Nagai, K.; Yamada,
A.; Sugihara, Y.; Fukuda, T.; Fukuda, T.; Uchida, R.; Tomoda, H.;
Omura, S.; Nagamitsu, T. Org. Lett. 2011, 13, 1158. (b) Srebnik, M.;
̅
Mechoulam, R.; Yona, I. J. Chem. Soc., Perkin Trans. 1 1987, 1423.
(12) Talekar, R. S.; Chen, G. S.; Lai, S.-Y.; Chern, J.-W. J. Org. Chem.
2005, 70, 8590.
(13) Gauthier, J. Y.; Guindon, Y. Tetrahedron Lett. 1987, 28, 5985.
(14) Michel, B. Y. Synlett 2008, 2008, 2893.
(15) Brown, H. C.; Krishnamurthy, S. J. Org. Chem. 1969, 34, 3918.
(16) (a) Placzek, A. T.; Scanlan, T. S. Tetrahedron 2015, 71, 5946.
(b) Peters, M.; Trobe, M.; Tan, H.; Kleineweischede, R.; Breinbauer,
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b00706.
R. Chem. - Eur. J. 2013, 19, 2442. (c) Perez, C. R.; Lopez-Perez, D.;
́
́
́
Chmielewski, J.; Lipton, M. Chem. Biol. Drug Des. 2012, 79, 260.
(d) Zhou, C.-Y.; Li, J.; Peddibhotla, S.; Romo, D. Org. Lett. 2010, 12,
2104. (e) Bovonsombat, P.; Leykajarakul, J.; Khan, C.; Pla-on, K.;
Krause, M. M.; Khanthapura, P.; Ali, R.; Doowa, N. Tetrahedron Lett.
2009, 50, 2664. (f) Adimurthy, S.; Ramachandraiah, G.; Ghosh, P. K.;
Bedekar, A. V. Tetrahedron Lett. 2003, 44, 5099. (g) Jendralla, H.;
Chen, L.-J. Synthesis 1990, 1990, 827. (h) Kometani, T.; Watt, D. S. J.
Org. Chem. 1985, 50, 5384.
Experimental procedures and analytical data (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: ykobayas@bio.titech.ac.jp.
Notes
(17) Farmer, J. L.; Hunter, H. N.; Organ, M. G. J. Am. Chem. Soc.
2012, 134, 17470.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported in part by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Science, Sports, and
Culture, Japan.
REFERENCES
■
(1) (a) Konoshima, T.; Kozuka, M. J. Nat. Prod. 1991, 54, 816.
(b) Yahara, S.; Nishiyori, T.; Kohda, A.; Nohara, T.; Nishioka, I. Chem.
Pharm. Bull. 1991, 39, 2024. (c) Kuo, W.-L.; Chung, C.-Y.; Hwang, T.-
L.; Chen, J.-J. Phytochemistry 2013, 85, 153. (d) Li, S.; Wang, W.;
Tang, H.; Chen, K.; Yang, J.; He, L.; Ye, H.; Peng, A.; Chen, L. J.
Chromatogr. A 2014, 1344, 51.
(2) (a) Syu, W.-J.; Shen, C.-C.; Lu, J.-J.; Lee, G.-H.; Sun, C. M. Chem.
Biodiversity 2004, 1, 530. (b) Youn, U. J.; Lee, I. S.; Chen, Q. C.; Na,
M.-K.; Jung, H. J.; Lee, S. M.; Choi, J. S.; Woo, M. H.; Bae, K.-H.; Min,
B.-S. Nat. Prod. Sci. 2011, 17, 95.
(3) (a) Agharahimi, M. R.; LeBel, N. A. J. Org. Chem. 1995, 60, 1856.
(b) Vaillancourt, V.; Agharahimi, M. R.; Sundram, U. N.; Richou, O.;
Faulkner, D. J.; Albizati, K. F. J. Org. Chem. 1991, 56, 378.
(4) Kawada, H.; Ikoma, A.; Ogawa, N.; Kobayashi, Y. J. Org. Chem.
2015, 80, 9192.
(5) A copper reagent Li2Cu(CN)(4-BrC6H4)2 could be prepared
from 1,4-dibromobenzene (t-BuLi then CuCN) and used for 1,4-
addition to the BF3·OEt2-activated enone rac-3. The resulting boron
enolate was activated by MeLi and transformed to the enol phosphate
in 54% yield with ClP(O)(OEt)2.
(6) 1,4-Addition of 8e to inactivated rac-3 (without BF3·OEt2)
followed by phosphorylation gave only 7% yield of rac-10e.
(7) (a) Houjeiry, T. I.; Poe, S. L.; McQuade, D. T. Org. Lett. 2012,
́
14, 4394. (b) Chen, K.; Ishihara, Y.; Galan, M. M.; Baran, P. S.
Tetrahedron 2010, 66, 4738. (c) Chi, Y.; Gellman, S. H. Org. Lett.
2005, 7, 4253.
D
DOI: 10.1021/acs.orglett.6b00706
Org. Lett. XXXX, XXX, XXX−XXX