Total synthesis of (+)-galbulin and unnatural lignans

Article abstract of DOI:10.1021/acs.orglett.0c02294

The total synthesis of (+)-galbulin was achieved in 15% yield and 99% ee over eight steps from commercially available 4-veratraldehyde. The key steps include Meyer's asymmetric tandem addition to a chiral 2-oxazoline-substituted naphthalene, a Pd-catalyzed stereospecific decarboxylative γ-arylation, and a formal anti-Markovnikov hydromethylation. In addition, five unnatural lignans were synthesized using the same synthetic strategy.

Full text of DOI:10.1021/acs.orglett.0c02294

pubs.acs.org/OrgLett
Letter
Total Synthesis of (+)-Galbulin and Unnatural Lignans
Florian Clausen and Armido Studer*
Cite This: https://dx.doi.org/10.1021/acs.orglett.0c02294
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The total synthesis of (+)-galbulin was achieved in 15% yield and 99% ee
over eight steps from commercially available 4-veratraldehyde. The key steps include
Meyer’s asymmetric tandem addition to a chiral 2-oxazoline-substituted naphthalene, a Pd-
catalyzed stereospecific decarboxylative γ-arylation, and a formal anti-Markovnikov
hydromethylation. In addition, five unnatural lignans were synthesized using the same
synthetic strategy.
albulin (1) is a naturally occurring tetrahydronaphtha-
lene lignan (THNL) which was isolated from the trees
Himantandra baccata Bail., found in north Queensland, and
Himantandra belgraveana F. Muell. found in New Guinea, at
subtropical altitudes (Figure 1).1 THNL and dihydronaph-
Scheme 1. Retrosynthesis of Galbulin (1)
G
Figure 1. Galbulin (1), podophyllotoxin (2), and etoposid (3) are
members of the THNL family and are important secondary
metabolites.
structure of 1.13 More recently, Li et al. described an
alternative strategy using an Evans asymmetric alkylation and
a Sharpless epoxidation as key steps for the stereoselective
construction of galbulin.14
The retrosynthetic analysis of our approach is depicted in
Scheme 2. The motivation for the design of the key building
block 5 for our galbulin synthesis arouse from the total
syntheses of podophyllotoxin15 (2) of Reynolds et al. and
epipodophyllotoxin16 of Engelhardt et al. that use chiral
dihydronaphthalenes as intermediates. Guided by their
syntheses, we planned to prepare the dihydronaphthalene 5
using Meyers’ asymmetric tandem silyl anion addition/
thalene lignans show broad biological activity including
neurotoxic,2 antifungal,3 anti-HIV, anticancer,4 and antiviral,5
rendering these secondary metabolites highly interesting target
compounds. Noteworthy, podophyllotoxin (2) and its
glycosylated derivative etoposid (3) have previously been
used for the treatment of external warts6 and numerous types
of cancer including lung cancer, testicular cancer, leukemia,
and ovarian cancer.7,8 The core structure of 2 is related to the
backbone of galbulin, and novel strategies for accessing the
latter might pave the way to prepare more complex THNL
derivatives.
To date, a variety of racemic total syntheses of galbulin (1)
have been reported (Scheme 1).9−12 However, to the best of
our knowledge, only two enantioselective variants have been
disclosed. In 2012, Hong et al. investigated an organocatalyzed
domino Michael−Michael−aldol condensation in combination
with a kinetic resolution as key steps to access the core
Received: July 10, 2020
https://dx.doi.org/10.1021/acs.orglett.0c02294
© XXXX American Chemical Society
Org. Lett. XXXX, XXX, XXX−XXX
A

Organic Letters
pubs.acs.org/OrgLett
Letter
a
Scheme 2. Synthesis of Building Block 5
Scheme 3. Decarboxylative γ-Arylation of 5 and Preparation
of Galbulin Precursor 18 via a Hydroboration−Matteson−
a
CH2 Homologation Sequence
a
Key: (a) 10 (1.2 equiv), THF, −78 to 0 °C; H2SO4 (20%), H2O/
MeOH;24 (b) HCl(g) (bubbled through solution, 2 h), EtOH, 0 °C;
(S)-valinol (2.0 equiv), CH2Cl2, 40 °C 24 h; (c) Me2PhSiLi (0.5 M in
THF, 3.0 equiv), −20 °C, 24 h; Me2SO4 (5.0 equiv); (d) HCl (3 M
in H2O/dioxane, 1 mL/mmol of 9).
alkylation of the appropriately substituted naphthalene
derivative 6,17 itself accessible from commercially available 4-
veratraldehyde (7). The second veratryl moiety should then be
installed by applying our stereospecific decarboxylative γ-
arylation on acid 5 to give 4.18−22 A formal anti-Markovnikov
hydromethylation, recently introduced by our group,23 should
finally serve to install the remaining two chirality centers
including the methyl group to eventually afford galbulin.
We commenced our investigations by developing a robust
protocol for the synthesis of the building block 5 (Scheme 2).
In analogy to a literature procedure24 disclosed by Yamada et
al., the commercially available 4-veratraldehyde (7) was treated
with a slight excess of the alkyllithium compound 10. The
corresponding benzylic alcohol thus generated was heated with
diluted aqueous sulfuric acid in methanol under reflux to
engage in an intramolecular Friedel−Crafts-type reaction.
Subsequent elimination of ethanol and water provided the
cyanonaphthaline 8 in 81% yield. Nitrile 8 was reacted with
gaseous HCl in ethanol to afford the corresponding ethyl
imidate, which was further converted to the oxazoline 6 upon
treatment with 2.0 equiv of (S)-valinol (82% yield, 99% ee). A
variation of Meyers’ asymmetric tandem addition strategy17
allowed to break the aromaticity and to introduce a methyl
group α to the oxazoline moiety. Hence, reaction of
naphthalene 6 with an excess of dimethylphenylsilyl lithium
followed by quenching of the adduct anion with dimethyl
sulfate provided the β-silylated α-methylated oxazoline 9 in
80% yield as a single diastereoisomer. Hydrolysis of the
oxazoline ring and simultaneous protodesilylation of the
dimethylphenylsilyl group using aqueous HCl in dioxane at
elevated temperature led to the carboxylic acid 5 in 68% yield
and 99% ee. The ee of acid 5 was determined on its methyl
ester readily prepared upon treatment of 5 with iodomethane
and potassium carbonate in DMF (see the Supporting
Information).
a
Key: (a) 4-veratryl bromide (12) (1.2 equiv), Cs2CO3 (1.3 equiv),
Pd(dba)2 (10 mol %), PhMe (0.3 M), 110 °C, 18 h; (b) Et3SiH (3.0
equiv), BCl3 (1.0 M in CH2Cl2, 3.0 equiv), CH2Cl2, 0 °C, 3 h; 2,2-
dimethyl-1,3-propanediol (3.0 equiv); (c) CH2BrI (10.0 equiv), n-
BuLi (1.6 M in hexanes, 8.0 equiv), THF, −78 to rt; NaOH(aq) (0.2
M), pinacol (5.0 equiv).
be highly sensitive toward oxidation and clean isolation
without decomposition was not possible. We therefore decided
to carry out the planned hydroboration−Matteson−CH2
homologation sequence on crude 4 without any further
purification. Of note, decarboxylative γ-arylation with 4-
veratryl iodide (11) provided a significant amount of undesired
dihydronaphthalene 13, resulting from Heck-type arylation of
targeted 4 (ratio 4:13 around 1:1 at 110° and around 3:1 at 90
°C; see the Supporting Information).
Diastereoselective hydroboration of crude 4 was achieved
with in situ generated Cl2BH. To this end, triethylsilane was
added to the crude mixture of 4, and the solution was then
filtered directly into a solution of boron trichloride in CH2Cl2
at 0 °C to give the hydroboration product 14 (dr around 5:1,
see below).25 Addition of pinacol provided the corresponding
boronic ester 15, which was again highly unstable, forcing us to
continue the reaction sequence without any further purifica-
tion. Unfortunately, under standard Matteson−CH2 homo-
logation conditions with dibromomethane as the source of the
carbenoid after a permutational interconversion, 15 could not
be converted to the boronic ester 17, likely for steric reasons.
Substitution of dibromomethane with chloroiodomethane led
to the desired homologation, though the conversion as
monitored by GC−FID was not satisfactory. We therefore
switched to the sterically less bulky neopentyl boronic ester 16,
which is readily obtained from 14 upon treatment with
neopentyl glycol. As already noted for the corresponding
pinacol ester 15, 16 turned out to be too labile for purification
and was therefore directly used in the next step. Pleasingly, in
combination with chloroiodomethane as carbenoid source, the
homologation proceeded smoothly to give 18 that, unfortu-
nately, was again very unstable. However, after transester-
ification with pinacol, the pinacol boronic ester 17 could be
With the desired building block 5 in hand, we started to
examine the stereospecific decarboxylative γ-arylation for the
introduction of the veratryl ring (Scheme 3). Careful
experimentation revealed that the reaction works best using
4-veratryl bromide (12) as the electrophile with Pd(dba)2 as
the catalyst (10 mol %) in toluene at 110 °C using Cs2CO3
(1.3 equiv) as the base. However, we found the product 4 to
B
https://dx.doi.org/10.1021/acs.orglett.0c02294
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
a
isolated in 51% overall yield with a diastereomeric ratio of 5:1
over three steps (one pot).
Scheme 6. Synthesis of Unnatural Lignans from Acid 20
Final protodeboronation of 17 leading to galbulin (1)
turned out to be highly challenging. Several known
strategies26−28 were tested on 17; however, all these
established protocols failed to deliver 1. To our delight, the
protodeboronation protocol recently developed by our group23
was applicable to this substrate. Hence, treatment of 17 with
phenyllithium gave the corresponding phenylboron−“ate”
complex 19. Subsequent oxidation under photoredox con-
ditions and trapping of the generated primary alkyl radical with
thiophenol gave (+)-galbulin (1) in 80% yield with a
diastereomeric ratio of 5:1 and an ee of 99% (Scheme 4).
a
Scheme 4. Final Transformation to (+)-Galbulin (1)
a
Key: (a) aryl iodide (1.1 equiv), Cs2CO3 (1.2 equiv), Pd(dba)2 (20
mol %), PhMe (0.3 M), 110 °C, 18 h, then Et3SiH (3.0 equiv), BCl3
(1.0 M in CH2Cl2, 3.0 equiv), CH2Cl2, 0 °C, 3 h; 2,2-dimethyl-1,3-
propanediol (5.0 equiv). then CH2ICl (10.0 equiv), n-BuLi (1.6 M in
hexanes, 8.0 equiv), THF, −78 to rt.; NaOH(aq.) (0.2 M), pinacol (5.0
equiv); (b) PhLi (1.1 equiv), THF, −78 °C to rt, 1 h, then
Ir(dFCF3ppy)2-(dtbbpy)PF6 (2 mol %), PhSH (1.1 equiv), MeOH/
acetone (1:1), blue LED, rt, 18 h.
a
Key: (a) PhLi (1.1 equiv), Et2O, 0 °C to rt, 1 h; (b) Ir(dFCF3ppy)2-
(dtbbpy)PF6 (2 mol %), PhSH (1.1 equiv), MeOH/acetone (1:1),
blue LED, rt, 18 h.
To further harvest the potential of this newly developed
strategy for the synthesis of lignans, galbulin analogues were
synthesized starting from the nonmethoxylated precursor 20.
The carboxylic acid 20 was prepared in four steps, starting
from commercially available 2-naphthoyl chloride (21) and
racemic tert-leucinol, to give oxazoline 22 in 58% yield
(Scheme 5). Tandem silylation/methylation provided 23 in
fore conducted without further purification to give the boronic
esters 25−29 in 30−55% isolated yields with diastereomeric
ratios varying from 1.3:1 to 10.5:1. Our protodeboronation
protocol afforded the lignans 30−34 in moderate to good
overall yields (41%−85%). The protodeboronation is a
stereospecific process and the varying diastereoisomeric ratio
in 30−34 as compared to the starting boronic esters 25−29 is
caused either by the different reactivity of the two isomeric
esters toward the protodeboronation or by isomer enrichment
during purification.
a
Scheme 5. Synthesis of rac-Carboxylic Acid (20)
In summary, we have reported an enantioselective total
synthesis of (+)-galbulin (1) starting from commercially
available 4-veratraldehyde (7). A Meyers’ asymmetric tandem
addition, a highly stereospecific γ-arylation, and a diaster-
eoselective formal anti-Markovnikov hydromethylation were
successfully used as key steps. The same strategy was also
applied to the synthesis of a small series of unnatural lignans
30−34.
ASSOCIATED CONTENT
■
a
sı
Key: (a) rac-tert-leucinol (1.5 equiv), NEt3 (1.5 equiv), CH2Cl2, rt,
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.0c02294.
overnight; SOCl2, 3 h; (b) Me2PhSiLi (0.9 M in THF, 3.0 equiv),
−20 °C, 24 h; MeI (4.0 equiv); (c) TBAF (6.2 equiv), THF, rt,
overnight; (d) H2SO4, dioxane/H2O (4:1).
General experimental procedures and characterization of
compounds including NMR spectra (PDF)
61% yield. Hydrodesilylation with TBAF in THF delivered the
oxazoline 24 (78%), which was hydrolyzed under acidic
conditions to afford the key carboxylic acid 20 (64%), ready
for further diversification toward lignans.
Decarboxylative γ-arylation was achieved using five different
aryl iodides (Scheme 6). As for the galbulin synthesis, these
intermediates turned out to be sensitive toward air oxidation
and the subsequent hydroboration−homologation was there-
AUTHOR INFORMATION
■
Corresponding Author
̈
Armido Studer − Organisch-Chemisches Institut, Westfalische
̈
̈
̈
Wilhelms-Universitat Munster, 48149 Munster, Germany;
orcid.org/0000-0002-1706-513X; Email: studer@wwu.de
C
https://dx.doi.org/10.1021/acs.orglett.0c02294
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Author
pubs.acs.org/OrgLett
Letter
(28) Povie, G.; Villa, G.; Ford, L.; Pozzi, D.; Schiesser, C. H.;
Renaud, P. Chem. Commun. 2010, 46, 803−805.
̈
Florian Clausen − Organisch-Chemisches Institut, Westfalische
Wilhelms-Universitat Munster, 48149 Munster, Germany
̈
̈
̈
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.orglett.0c02294
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was financially supported by the Deutsche
Forschungsgemeinschaft (DFG). We thank Dr. Grete
̈
Hoffmann (WWU Munster) for fruitful discussions.
REFERENCES
■
(1) Hughes, G.; Ritchie, E. Aust. J. Chem. 1954, 7, 104−112.
(2) Xu, P.; Sun, Q.; Wang, X.; Zhang, S.; An, S.; Cheng, J.; Gao, R.;
Xiao, H. NeuroToxicology 2010, 31, 680−686.
(3) Anil Kumar, K.; Kumar Singh, S.; Siva Kumar, B.; Doble, M.
Cent. Eur. J. Chem. 2007, 5, 880−897.
(4) Lv, M.; Xu, H. Mini-Rev. Med. Chem. 2011, 11, 901−909.
(5) Ayres, D. C.; Loike, J. D. Lignans: Chemical, Biological and
Clinical Properties; Cambridge University Press, 1990.
(6) Beutner, K. R.; Ferenczy, A. Am. J. Med. 1997, 102, 28−37.
(7) Fitzharris, B. M.; Kaye, S. B.; Saverymuttu, S.; Newlands, E. S.;
Barrett, A.; Peckham, M. J.; McElwain, T. J. Eur. J. Cancer 1980, 16,
1193−7.
(8) Loike, J. D. Cancer Chemother. Pharmacol. 1982, 7, 103−11.
(9) Muller, A.; Vajda, M. J. Org. Chem. 1952, 17, 800−806.
̈
(10) Landais, Y.; Lebrun, A.; Lenain, V.; Robin, J. P. Tetrahedron
Lett. 1987, 28, 5161−5164.
(11) Kasatkin, A. N.; Checksfield, G.; Whitby, R. J. J. Org. Chem.
2000, 65, 3236−3238.
(12) Datta, P.; Yau, K. C.; Hooper, T. S.; Yvon, B. L.; Charlton, J. L.
J. Org. Chem. 2001, 66, 8606−8611.
(13) Hong, B. C.; Hsu, C. S.; Lee, G. H. Chem. Commun. 2012, 48,
2385−2387.
(14) Li, X.; Jiao, X.; Liu, X.; Tian, C.; Dong, L.; Yao, Y.; Xie, P.
Tetrahedron Lett. 2014, 55, 6324−6327.
(15) Reynolds, A. J.; Scott, A. J.; Turner, C. I.; Sherburn, M. S. J. Am.
Chem. Soc. 2003, 125, 12108−12109.
(16) Engelhardt, U.; Sarkar, A.; Linker, T. Angew. Chem., Int. Ed.
2003, 42, 2487−2489.
(17) Meyers, A. I.; Roth, G. P.; Hoyer, D.; Barner, B. A.; Laucher, D.
J. Am. Chem. Soc. 1988, 110, 4611−4624.
(18) Chou, C. M.; Chatterjee, I.; Studer, A. Angew. Chem., Int. Ed.
2011, 50, 8614−8617.
(19) Klotter, F.; Studer, A. Angew. Chem., Int. Ed. 2014, 53, 2473−
2476.
(20) Klotter, F.; Studer, A. Angew. Chem., Int. Ed. 2015, 54, 8547−
8550.
(21) Scheipers, I.; Koch, E.; Studer, A. Org. Lett. 2017, 19, 1741−
1743.
(22) Scheipers, I.; Muck-Lichtenfeld, C.; Studer, A. Angew. Chem.,
̈
Int. Ed. 2019, 58, 6545−6548.
(23) Clausen, F.; Kischkewitz, M.; Bergander, K.; Studer, A. Chem.
Sci. 2019, 10, 6210−6214.
(24) Yamada, H.; Hirayama, F.; Koshio, H.; Matsumoto, Y.;
Yanagisawa, I. Production of Naphthalene Derivative and Its
Production Intermediate. JP2000219665 (A), 2000.
(25) Soundararajan, R.; Matteson, D. S. J. Org. Chem. 1990, 55,
2274−2275.
(26) Rasappan, R.; Aggarwal, V. K. Nat. Chem. 2014, 6, 810−814.
(27) Pozzi, D.; Scanlan, E. M.; Renaud, P. J. Am. Chem. Soc. 2005,
127, 14204−14205.
D
https://dx.doi.org/10.1021/acs.orglett.0c02294
Org. Lett. XXXX, XXX, XXX−XXX