Mild Friedel-Crafts Reactions Enable a Robust Synthesis of Roseophilin

Article abstract of DOI:10.1021/acs.orglett.9b01100

An 11-step total synthesis of either enantiomer of roseophilin has been developed. The chemistry features effective production of a challenging carbocation on a macrocyclic segment 25 and a highly efficient intermolecular Friedel-Crafts alkylation reaction to integrate a complex furan-pyrrole unit 5 regioselectively with this carbocation under very mild reaction conditions.

Full text of DOI:10.1021/acs.orglett.9b01100

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Mild Friedel−Crafts Reactions Enable a Robust Synthesis of
Roseophilin
Xiao Zhang, Cuifang Zhang, Xiaobin Wang, and Guangxin Liang*
State Key Laboratory of Elemento-Organic Chemistry, College of Chemistry, Nankai University, Tianjin 300071, China
S
* Supporting Information
ABSTRACT: An 11-step total synthesis of either enantiomer of roseophilin has been developed. The chemistry features
effective production of a challenging carbocation on a macrocyclic segment 25 and a highly efficient intermolecular Friedel−
Crafts alkylation reaction to integrate a complex furan-pyrrole unit 5 regioselectively with this carbocation under very mild
reaction conditions.
(−)-Roseophilin (1, Figure 1) is the enantiomer of the
naturally occurring (+)-roseophilin (2, Figure 1), a structurally
and was identified as a new class of protein tyrosine
phosphatase inhibitors.5
Considering aberrant protein phosphorylation is responsible
for development of many human diseases such as cancer and
diabetes and the scarcity of specific protein phosphatase
inhibitors, 1 could serve as a promising scaffold for further
development of potent and selective agents for chemical
biology and medicinal research.5 It is noteworthy that the
mechanism of cytotoxicity observed on both enantiomers of
roseophilin is largely different from prodigiosin and its
derivatives.5 Unlike prodigiosins, roseophilins do not cleave
DNA in the presence of CuII and O2.5 In addition, inhibition of
phosphatase by 1 is speculated to be linked to the unknown
mechanism of cytotoxicity.5 Taken together, the appealing
biological activities, elusive mode of functions, and the intricate
topology render 1 a very attractive and rewarding target for
total synthesis.
Figure 1. (−)-Roseophilin and its naturally occurring antipode.
unique metabolite identified from the culture broth of a
member of actinomycete Streptomyces griseoviridis.1 Rose-
ophilin is believed to be biosynthesized partially through the
same pathway as the prodigiosin family of alkaloids.2 Given
that the related prodigiosin alkaloids have demonstrated a
range of appealing bioactivities such as antitumor,1 antibiotic,3a
antibacterial,3b antimalarial,3c and immunosuppressive activi-
ties,3d roseophilin has been enthusiastically studied for
biological functions since its isolation in 1992. Fascinatingly,
a very rare bioactivity feature of roseophilin was observed in
the previous assays: the non-natural enantiomer 1 exhibits
higher biological activities than its naturally occurring antipode
2 in multiple independent assays in both cytotoxicity and
phosphatase inhibition studies.4,5
With the exception of Harran’s total synthesis,6 previous
synthetic efforts all focused on the condensation of a furan-
pyrrole unit with a keto pyrrole macrocyclic segment to
construct the ansa-brideged 1-azafulvene core in the natural
product (Figure 2).4,7,8 This seemingly straightforward
proposal turned out to be a formidable task. Although model
studies worked nicely in simplified systems,9 the only method
that fulfilled the need in real total synthesis was through an
organocerium species 3 with N-SEM (2-(trimethylsilyl)-
ethoxymethyl) protected 4. This successful solution was
In 2001, Boger and Hong first disclosed that the unnatural
enantiomer 1 was approximately 2- to 10-fold more potent
than its naturally occurring antipode in a range of cytotoxic
originally reported by Furstner and Weintritt. They found
̈
that both the organocerium species 3 and a carefully chosen
coupling partner 4 were vital to the success of the
condensation. This breakthrough led to the very first total
assays.4 Later that year, Fu
̈
rstner and co-workers reinforced
this finding in their study searching for new protein tyrosine
phosphatase inhibitors based on roseophilin: while naturally
occurring 2 proved to be an inhibitor of phosphatases, 1 was 5-
to 6-fold more potent in inhibition of PTP1B as well as VHR
Received: March 28, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b01100
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 3. An intermolecular Friedel−Crafts alkylation strategy for
coupling of 5 with 10.
Figure 2. Previous solution to incorporate a furan-pyrrole unit to a
pyrrole macrocyclic segment.
a
Scheme 1. Convenient Preparation of 17
synthesis of ( )-roseophilin.7a Later on, in applying the same
chemistry in their total synthesis, the Tius group discovered
that precise control of reaction temperature and time at each
stage was critical to the success of the reaction.8 Eventually,
they contributed a detailed protocol for this delicate key
reaction, which helped Boger’s synthesis of (−) -roseophilin as
well.4 Tius also commented that even the particle size of CeCl3
could affect the transmetalation step which might be
responsible for the capricious results they encountered.8 The
perplexing condensation step might explain why only three
total syntheses have been accomplished this way, considering
nearly ten other synthetic efforts stopped at advanced
macrocyclic intermediates even though they are only two
steps away from the ultimate synthetic target,9 given that the
preparation of 5 is well-documented in the literature.10
The finicky nature of the reaction urged us to seek a more
user-friendly approach to incorporate a furan-pyrrole unit to
the pyrrole macrocyclic segment in our synthesis toward
roseophilins. We envisioned that a Friedel−Crafts alkylation
reaction between a stabilized carbocationic species such as 8/9
and a neutral mild nucleophile such as 5 could serve as a good
solution (Figure 3). This strategy will capitalize on the
inherent nucleophilicity at C8 in the furan11 so as to avoid the
requirement for regioselective prefunctionalization of the
nucleophile. Furthermore, such a solution could facilitate
easy preparation of roseophilin analogues bearing diverse
electron-rich species in the place of the original furan-pyrrole
unit. The desired carbocation 8 was expected to be formed
readily through activation of the hydroxyl group in 10 due to
the stabilization effect provided by the pyrrole in the resonance
form 9. It was encouraging that a compound similar to the
Friedel−Crafts product 7 had been successfully oxidized to
afford 1 in Harran’s synthesis.6
a
Reagents and conditions: (a) (1) LiHMDS, THF, −78 °C, then 4-
iodobut-1-ene (65%); (2) NaOH, MeOH/H2O; K2CO3, acetone,
BnBr; DMSO, (COCl)2, NEt3, CH2Cl2 (80% in 3 steps); (b) D-
proline, MeOH, NaOAc, 2-nitroocta-1,7-diene (97%); (c) BnNH2,
MgSO4, DMSO, rt−110 °C (70%); (d) (1) Grubbs I cat. (20 mol %),
Ti(Oi-Pr)4 (50%); (2) H2, Pd/C (94%); (e) PPh3, Cl3CCN, rt−40
°C (80%); LiHMDS = lithium bis(trimethylsilyl) amide, Bn = benzyl.
produced 14 as a precursor for the formation of pyrrole 15.13
Condensation between 14 and benzyl amine at room
temperature followed by Grob−Camenisch pyrrole synthesis
produced 15 in 70% yield.14 The one-pot transformation of 14
to 15 takes advantage of the versatile role of the nitro group
which serves as both an electrophile upon tautomerization and
a leaving group. Catalyzed by Grubbs’ first-generation catalyst,
ring-closing metathesis (RCM) of 15 produced a macrocyclic
alkene,15 which was subjected to hydrogen and Pd/C to
undergo hydrogenation of the alkene as well as removal of a
benzyl group to afford carboxylic acid 16. Upon activation with
a combination of PPh3 and trichloroacetonitrile, 16 underwent
an intramolecular Friedel−Crafts acylation reaction to afford
17 in 80% yield.16 This 9-step sequence from 12 provided
optically pure 17 in 13% overall yield, which allowed us to
accumulate multiple grams of 17 conveniently.
With 17 in hand, we carried out a reduction of the carbonyl
group (Scheme 2). Reduction using LiAlH4 went smoothly as
indicated by H NMR analysis on the crude reaction mixture.
However, upon purification with silica gel chromatography, 18
was cleanly converted to 19 in 80% yield. Presumably, 18
underwent a retro-Friedel−Crafts reaction even when exposed
to slightly acidic silica gel. Given that we planned to activate
Our exploration commenced from a concise preparation of
17 depicted in Scheme 1. An optically pure lactone 12, readily
derived in 2 steps from R-carvone,12 was used as a starting
material. Enolate alkylation using homoallylic iodide installed
the desired carbon chain with complete stereochemical control
in 65% yield. Subsequent saponification of the intermediate
lactone followed by formation of the benzyl ester and Swern
oxidation afforded 13 in 80% overall yield. Michael addition of
13 to 2-nitroocta-1,7-diene catalyzed by D-proline cleanly
1
B
DOI: 10.1021/acs.orglett.9b01100
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
the debenzylation nicely to convert 17 to 23 in 88% yield.17
LiAlH4 reduction of 23 proceeded uneventfully to give 24.
Compound 24 is still acid labile and also produced the
corresponding aldehyde as 18 did when exposed to slightly
acidic conditions. Unfortunately, addition of lithiated 5 to the
aldehyde derived from 24 failed, which urged us to seek a new
solution to accomplish the synthesis. Contemplating on the
retro-Friedel−Crafts reaction responsible for the formation of
aldehyde, we reasoned that if the pyrrole could be prevented
from protonating during activation of the hydroxyl group in
24, we should be able to suppress the retro-Friedel−Crafts
reaction and generate the desired carbocation 25. Guided by
this assumption, we tested [PPh3PBr]+ for specific activation of
the hydroxyl group in 24.18 Gratifyingly, this final solution
succeeded, awarding us a highly effective union of 5 and 24
under mild conditions. After reduction of 23 with LiAlH4, the
crude product 24 and 5 were dissolved in CH2Cl2 and slowly
added to a premixed solution of Ph3P, CBr4 in CH2Cl2 at 0 °C.
The intermolecular Friedel−Crafts alkylation was completed
immediately after the addition. Subsequent oxidation with
DDQ, and desilylation with a mixture of hydrofluoric acid and
hydrochloric acid, afforded (−)-roseophilin·HCl in 67% overall
yield from 23. The entire reaction sequence proved to be
robust, and a reaction on a 100 mg scale showed no loss of
efficiency. Both 1H and 13C NMR data for our synthetic 1·HCl
match those reported for the natural product. The optical
rotation of the synthetic sample [α]1D9 = −7650 (c = 0.01,
CHCl3) was of opposite sign to that of natural roseophilin,4
indicating preparation of unnatural enantiomer 1. Presumably,
the route we have developed for synthesis of 1 is applicable to
the synthesis of its natural enantiomer 2 as long as S-carvone is
used.19
Scheme 2. A Detour Originated from Unexpected
Production of 19
a
a
Reagents and conditions: (a) LiAlH4, THF, 0 °C, then silica gel
(80%); (b) sec-BuLi, 5, THF; then SnCl4, CH2Cl2 (55% in 2 steps);
(c) DDQ, CH2Cl2.
the hydroxyl group in 18 under acidic conditions so as to
produce a stable carbocation 8 in our initial proposal, the
unexpected acid lability of 18 temporarily prevented us from
exploring the original idea at this stage. Instead, we decided to
take a detour to investigate an intramolecular Friedel−Crafts
reaction to synthesize the target. To this end, we first treated
19 with lithiated 5 to produce substrate 20. To our delight, 20
is well suited for this strategy and it even partially cyclized to
give 21 during purification on silica gel chromatography.
Under optimal conditions, when subjected to SnCl4, crude 20
generated 21 as a single diastereomer in 55% yield in 2 steps
from 19. Oxidation of 21 with DDQ seemed to be effective as
indicated by the characteristic deep red color of an azafulvene
chromophore. However, all efforts to isolate compound 22
failed due to its instability. Speculating that the benzyl group in
22 might be removed through either SN2 reaction or oxidation,
we then attempted to convert 22 directly to roseophilin.
Unfortunately, these efforts were fruitless as well. The molecule
either stayed intact under mild conditions or decomposed
when more forcing conditions were applied.
To further explore the Friedel−Crafts reaction, we also used
unprotected indole and pyrrole as coupling partners. These
alternatives effectively provided roseophilin analogues 27 and
28 in 73% and 53% overall yield, respectively (Figure 4).
Given that the benzyl group is hard to remove in 22, we
decided to remove it at an earlier stage (Scheme 3). We found
that t-BuOK in a mixed solvent of THF and DMSO fulfilled
Figure 4. Roseophilin analogues produced through the protocol.
In summary, we have developed a robust total synthesis of
either enantiomer of roseophilins from economically friendly
carvones. The convenient sequence afforded roseophilin in 100
mg scale in 7.7% overall yield starting from 12.20 The
chemistry features a 9-step21 gram-scale preparation of
optically pure pyrrole macrocyclic segment 17 and an
intermolecular Friedel−Crafts reaction to solve the challenging
integration of a furan-pyrrole unit into this segment.
Furthermore, the chemistry enabled preparation of two new
roseophilin analogues in either enantiomeric form as well.22
We believe that the chemistry we have developed will help the
preparation of more roseophilin analogues as well and expand
the chemical biology and medicinal studies on this new family
of cytotoxic agents and protein tyrosine phosphatase inhibitors.
In addition, the neutral reaction conditions we developed for
the Friedel−Crafts reaction might be useful and broadly
applicable for acid sensitive substrates in related reactions.
a
Scheme 3. Total Synthesis of (−)-Roseophilin·HCl
a
Reagents and conditions: (a) t-BuOK, DMSO/THF, O2, 0 °C
(88%); (b) LiAlH4, THF, rt; 5, Ph3P, CBr4, CH2Cl2; DDQ, CH2Cl2;
HF, HCl, MeOH (67% from 23).
C
DOI: 10.1021/acs.orglett.9b01100
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Org. Chem. 2012, 77, 704−706. (i) Kerr, D. J.; Flynn, B. L. Org. Lett.
2012, 14, 1740−1743.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.9b01100.
■
S
(10) For preparation of furan-pyrrole unit 5, see: (a) Reference 7a.
(b) Frederich, J. H.; Matsui, J. K.; Chang, R. O.; Harran, P. G.
Tetrahedron Lett. 2013, 54, 2645−2647. (c) Kerr, D. J.; Flynn, B. L.
Aust. J. Chem. 2015, 68, 1821−1828. (d) It was prepared on gram
scale using a modified approach in 7 steps from commercially
available 2-formyl-3-chloropyrrole in our synthesis; see the Support-
ing information (SI) for details.
Experimental procedures, spectral data for all new
compounds (PDF)
(11) Nakatani, S.; Kirihara, M.; Yamada, K.; Terashima, S.
Tetrahedron Lett. 1995, 36, 8461−8464.
AUTHOR INFORMATION
■
(12) Paquette, L. A.; Dahnke, K.; Doyon, J.; He, W.; Wyant, K.;
Friedrich, D. J. Org. Chem. 1991, 56, 6199−6205.
Corresponding Author
*E-mail: lianggx@nankai.edu.cn.
ORCID
(13) List, B.; Pojarliev, P.; Martin, H. J. Org. Lett. 2001, 3, 2423−
2425.
(14) For seminal studies of the Grob−Camenisch reaction, see:
(a) Grob, C. A.; Camenish, K. Helv. Chim. Acta 1953, 36, 49−58.
(b) Grob, C. A.; Schad, H. P. Helv. Chim. Acta 1955, 38, 1121−1127.
(c) Gomez-Sanchez, A.; Mancera, M.; Rosado, F.; Bellanato, J. J.
Chem. Soc., Perkin Trans. 1 1980, 1, 1199−1205. (d) Meyer, H. Liebigs
Ann. Chem. 1981, 1981, 1534−1544. (e) Lim, S.; Jabin, I.; Revial, G.
Tetrahedron Lett. 1999, 40, 4177−4180. (f) Ploypradith, P.; Mahidol,
C.; Sahakitpichan, P.; Wongbundit, S.; Ruchirawat, S. Angew. Chem.,
Int. Ed. 2004, 43, 866−868. (g) Liermann, J. C.; Opatz, T. J. Org.
Chem. 2008, 73, 4526−4531.
Guangxin Liang: 0000-0003-3122-0332
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National Key Research and Development
Program of China (2017YFD0201404) and the National
Natural Science Foundation of China (21572104, 21772097)
for financial support. We also thank Prof. Hans Renata at
Scripps Florida and Prof. Christopher Beaudry at Oregon State
University for helping us with the manuscript preparation.
(15) For use of different RCM reactions to synthesize the
macrocycles in roseophilins, see refs 8, 9a, 9e, and 9i.
(16) Jang, D. O.; Park, D. J.; Kim, J. Tetrahedron Lett. 1999, 40,
5323−5326.
(17) Fu
combined yield using calcium in liquid ammonium followed by
reoxidation of carbonyl group with PDC; see: (a) Furstner, A.;
̈
rstner has reported the debenzylation of 17 in 50%
REFERENCES
■
̈
(1) Hayakawa, Y.; Kawakami, K.; Seto, H.; Furihata, K. Tetrahedron
Lett. 1992, 33, 2701−2704.
Weintritt, H. J. Am. Chem. Soc. 1997, 119, 2944−2945. (b) Haddach,
A. A.; Kelleman, A.; Deaton-Rewolinski, M. V. Tetrahedron Lett. 2002,
43, 399−402.
(2) (a) Wasserman, H. H.; Sykes, R. J.; Peverada, P.; Shaw, C. K.;
Cushley, R. J.; Lipsky, S. R. J. Am. Chem. Soc. 1973, 95, 6874−6875.
(b) Williamson, N. R.; Fineran, P. C.; Leeper, F. J.; Salmond, G. P. C.
Nat. Rev. Microbiol. 2006, 4, 887−899. (c) Withall, D. M.; Haynes, S.
W.; Challis, G. L. J. Am. Chem. Soc. 2015, 137, 7889−7897. For a
review on pyodigininenatual products, see: (d) Hu, D. X.; Withall, D.
M.; Challis, G. L.; Thomson, R. J. Chem. Rev. 2016, 116, 7818−7853.
(3) (a) Boger, D. L.; Patel, M. Tetrahedron Lett. 1987, 28, 2499−
2502. (b) Boger, D. L.; Patel, M. J. Org. Chem. 1988, 53, 1405−1415.
(c) Gerber, N. N. J. Antibiot. 1975, 28, 194−199. (d) Yamasaki, M.;
Tsuji, R. F.; Yamamoto, M.; Nakamura, A.; Kataoka, T.; Magae, J.;
Nagai, K. J. Antibiot. 1990, 43, 1293−1301.
(18) Appel, R. Angew. Chem., Int. Ed. Engl. 1975, 14, 801−811.
(19) For a recent review on total syntheses using carvone, see: Brill,
Z. G.; Condakes, M. L.; Ting, C. P.; Maimone, T. J. Chem. Rev. 2017,
117, 11753−11795.
(20) For the importance and examples of scalable synthesis, see:
Kuttruff, C. A.; Eastgate, M. D.; Baran, P. S. Nat. Prod. Rep. 2014, 31,
419−432.
(21) 2-Nitroocta-1,7-diene used in the sequence was prepared from
hex-5-enal in 5 steps; please see SI for details.
(22) For a recent example of the benefit in exploring both
enantiomers of the natural product for biological functions, see:
Logan, M. M.; Toma, T.; Thomas-Tran, R.; Du Bois, J. Science 2016,
354, 865−869.
(4) Boger, D. L.; Hong, J. J. Am. Chem. Soc. 2001, 123, 8515−8519.
(5) (a) Fu
709. (b) Fu
(c) Furstner, A.; Reinecke, K.; Prinz, H.; Waldmann, H.
̈
rstner, A.; Grabowski, E. J. ChemBioChem 2001, 2, 706−
̈
rstner, A. Angew. Chem., Int. Ed. 2003, 42, 3582−3603.
̈
ChemBioChem 2004, 5, 1575−1579. (d) Manger, M.; Scheck, M.;
Prinz, H.; Von Kries, J. P.; Langer, T.; Saxena, K.; Schwalbe, H.;
Furstner, A. ChemBioChem 2005, 6, 1749−1753.
̈
(6) Frederich, J. H.; Harran, P. G. J. Am. Chem. Soc. 2013, 135,
3788−3791.
(7) (a) Furstner, A.; Weintritt, H. J. Am. Chem. Soc. 1998, 120,
̈
2817−2825. (b) Furstner, A.; Gastner, T.; Weintritt, H. J. Org. Chem.
̈
1999, 64, 2361−2366.
(8) Harrington, P. E.; Tius, M. A. J. Am. Chem. Soc. 2001, 123,
8509−8514.
(9) (a) Kim, S. H.; Figueroa, I.; Fuchs, P. L. Tetrahedron Lett. 1997,
38, 2601−2604. (b) Mochizuki, T.; Itoh, E.; Shibata, N.; Nakatani, S.;
Katoh, T.; Terashima, S. Tetrahedron Lett. 1998, 39, 6911−6914.
(c) Robertson, J.; Hatley, R. J. D. Chem. Commun. 1999, 16, 1455−
1456. (d) Trost, B. M.; Doherty, G. A. J. Am. Chem. Soc. 2000, 122,
3801−3810. (e) Bamford, S. J.; Luker, T.; Speckamp, W. N.;
Hiemstra, H. Org. Lett. 2000, 2, 1157−1160. (f) Robertson, J.; Hatley,
R. J. D.; Watkin, D. J. J. Chem. Soc., Perkin Trans. 2000, 1, 3389−3396.
(g) Bitar, A. Y.; Frontier, A. J. Org. Lett. 2009, 11, 49−52. (h) Song,
C.; Liu, H.; Hong, M.; Liu, Y.; Jia, F.; Sun, L.; Pan, Z.; Chang, J. J.
D
DOI: 10.1021/acs.orglett.9b01100
Org. Lett. XXXX, XXX, XXX−XXX