Unconventional application of the Mitsunobu reaction: Selective flavonolignan dehydration yielding hydnocarpins

Article abstract of DOI:10.3762/bjoc.12.66

Various Mitsunobu conditions were investigated for a series of flavonolignans (silybin A, silybin B, isosilybin A, and silychristin A) to achieve either selective esterification in position C-23 or dehydration in a one-pot reaction yielding the biological

Full text of DOI:10.3762/bjoc.12.66

Unconventional application of the Mitsunobu reaction:
Selective flavonolignan dehydration yielding hydnocarpins
Guozheng Huang1,2, Simon Schramm1, Jörg Heilmann3, David Biedermann4,
Vladimír Křen4 and Michael Decker*1
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2016, 12, 662–669.
1Pharmazeutische und Medizinische Chemie, Institut für Pharmazie
doi:10.3762/bjoc.12.66
und Lebensmittelchemie, Julius-Maximilians-Universität Würzburg,
Am Hubland, D-97074 Würzburg, Germany, 2College of Life and
Environmental Sciences, Shanghai Normal University, Shanghai,
P. R. China, 3Lehrstuhl für Pharmazeutische Biologie, Institut für
Pharmazie, Universität Regensburg, Universitätsstraße 31, D-93053
Regensburg, Germany and 4Centre of Biotransformation and
Biocatalysis, Institute of Microbiology, Czech Academy of Sciences,
Videnska 1083, Prague 4, CZ-14220, Czech Republic
Received: 11 December 2015
Accepted: 24 March 2016
Published: 08 April 2016
Associate Editor: S. Bräse
© 2016 Huang et al; licensee Beilstein-Institut.
License and terms: see end of document.
Email:
Michael Decker* - michael.decker@uni-wuerzburg.de
*
Corresponding author
Keywords:
dehydration; flavonoid; hydnocarpin; Mitsunobu; silybin
Abstract
Various Mitsunobu conditions were investigated for a series of flavonolignans (silybin A, silybin B, isosilybin A, and silychristin
A) to achieve either selective esterification in position C-23 or dehydration in a one-pot reaction yielding the biologically important
enantiomers of hydnocarpin D, hydnocarpin and isohydnocarpin, respectively. This represents the only one-pot semi-synthetic
method to access these flavonolignans in high yields.
Introduction
Flavonolignans combine the structural moieties of flavonoids ized extract of S. marianum fruits contains the so-called sily-
and phenylpropanoids (lignans). Among the group of flavono- marin complex, and is used as the main active component mix-
lignans, the medicinally most important ones can be found in ture of Legalon®, a drug approved for treatment of chronic
Silybum marianum (L.) Gaertn., the milk thistle. The phenolic hepatitis, hepatoprotection in alcohol addicts [2], or intoxica-
secondary metabolites of this plant have been intensively tion with Amanita phalloides, the death cap [3]. Silibinin (1,
studied due to their multiple biological activities and promising Figure 1), the major component of silymarin, became the first
therapeutic applications, such as hepatoprotection, antitumor, isolated flavonolignan [4]. Besides 1, this extract consists also
antiproliferative and anti-oxidant properties [1]. The standard- of isosilibinin (3), silychristin (5), silydianin (11), and dehy-
662

Beilstein J. Org. Chem. 2016, 12, 662–669.
Figure 1: Structures of silibinin, isosilybin, and silychristin, and hydnocarpin-type flavonolignans.
drosilibinin [1,5]. Natural and commercially available silibinin The flavonolignan hydnocarpin (4, Figure 1) was first isolated
is the diastereomeric mixture (quasi equimolar) of silybin A in 1973 from Hydnocarpus wightiana [12], and its structure was
(
1a) and silybin B (1b).
at first assigned erroneously as what is now named hydno-
carpin D (2, Figure 1) (for details see below) [13]. The actual
Because of the well-known biological properties of silibinin and hydnocarpin D was first synthesized [14] and later isolated from
the other compounds of the silymarin complex for protection of various plant species (for reviews see [5,15]). In several refer-
liver damage, and for preventing skin tumor promotion, these ences the nomenclature of hydnocarpins, assignment of their
compounds have attracted efforts on their total synthesis (e.g., structures and/or enantiomeric purities are not used coherently
isosilybin) [6] and structural modification [5,7-9] to improve or experimental and full structural evidence have not been pro-
their water-solubility, bioavailability and biological activities. vided [15].
Recently, our group has developed a hybrid drug combining
silibinin with tacrine (a potent acetylcholinesterase inhibitor for Hydnocarpin-type compounds (2, 4 and 6; Figure 1) are formal-
treatment of Alzheimer’s disease), which shows neuro- and ly dehydrated analogs of silymarin flavonolignans with flava-
hepatoprotective effects exceeding the cytoprotective effects of none-3-ol (3-hydroxyflavanone) structure (silybin A and B,
silibinin and effectively counteracts tacrine’s dose-dependent isosilybin A and silychristin A). Hydnocarpin and its deriva-
hepatotoxicity in vitro as well as in vivo [10,11].
tives show interesting biological activities such as being effi-
663

Beilstein J. Org. Chem. 2016, 12, 662–669.
cient inhibitors of the multidrug resistance (MDR) efflux pump uct only (7%, catalyzed by Ag2CO3; 5% catalyzed by horse-
(
e.g., of Staphylococcus aureus): this activity had been serendi- radish peroxidase) [14].
piously employed in traditional medicine (without knowledge
of the mechanism) in the treatment of leprosy with chaul- To study the biological activities of hydnocarpin-type
moogra oil. This oil, which is obtained from fruits (kernels) of compounds, a new and robust method for their preparation is re-
H. wightiana, contains besides hydnocarpin mainly cyclopen- quired. It is known that elimination can occur in some cases as a
tenoic fatty acids [16], which show pronounced antibiotic activ- side reaction in Mitsunobu esterifications with structurally
ity and inhibit multiplication of mycobacteria [17]. Combina- hindered alcohols and bulky acids: Cherney and Wang ob-
tion of these antibiotics together with hydnocarpin (MDR inhib- served an elimination of a serine derivative during esterifica-
itor) helped to treat such a persistent disease like leprosy caused tion experiments [23]. During conversion of substituted benzyl
by M. leprae. Recently, hydnocarpin (isolated from Brucea alcohols to amines under Mitsunobu conditions, elimination
javanica; which was named as (−)-hydnocarpin, even though also occurred [24]. Dehydration under Mitsunobu conditions
the enantiomeric purity was not determined) has been de- has been applied in modifications of natural products [23,24],
scribed as potentiator of vincristine’s cytotoxicity due to MDR such as N-Boc neomycin B under epoxide formation [25].
inhibition [18]. Furthermore, 5-methoxyhydnocarpin has been Elson et al. systematically investigated the mechanism of the
described to enhance the antimicrobial activity of berberine reaction of menthol with p-nitrobenzoic acid using a
[
19]. In the light of steadily growing antibiotic resistance any Hendrickson or Mitsunobu reagent [26]. The Hendrickson
potent and nontoxic MDR inhibitor is of utmost importance. reagent led to conversion of menthol into the β-elimination
Hydnocarpin has also antineoplastic activity due to sensitizing product. In all the above cases the substrate structures are very
multidrug-resistant cancer cell lines towards chemotherapy and different from flavanone-3-ols, but since the 3-OH group
no toxicity was found in vivo [20].
of the silymarin flavonolignans is also hindered, we have envi-
sioned that Mitsunobu conditions might enable the conversion
Unlike for silybins [6], there are only a few examples of total of silybins to hydnocarpins. Herein, we summarize our
syntheses or even structural modifications of hydnocarpin-type achievements on converting silibinin (including the two
(
flavone) flavonolignans – simply because they are not syntheti- isolated diasteromers silybin A and silybin B) to hydnocarpin D
cally available [14,21]. A straightforward synthetic approach (and its two enantiomers) under various Mitsunobu reaction
such as a semi-synthetic one from readily available material will conditions, and applying these conditions to the other silymarin
provide access to this class of highly interesting natural prod- congeners, which provided an effective and expeditious semi-
ucts and derivatives with putatively differing biological activi- synthetic preparation of hydnocarpin, hydnocarpin D and
ties. A very recent example of conversion of silybins to hydno- isohydnocarpin.
carpins was described by Vimberg et al., who applied a four-
step synthesis and used Vilsmeier–Haack conditions (Figure 2) Results and Discussion
[
22]. Before this, the only reference for the synthetic prepara- Although silibinin, i.e., silybins A and B, bears three phenolic,
tion of hydnocarpin D used coniferyl alcohol and luteolin as one secondary and one primary hydroxy group, selective
starting materials, and obtained hydnocarpin D as a minor prod- Mitsunobu esterification typically occurs at the primary C-23
Figure 2: Synthetic strategy of semi-synthesis of hydnocarpins from silybins [22].
664

Beilstein J. Org. Chem. 2016, 12, 662–669.
OH (position 11 according to IUPAC nomenclature, but rarely (major product, 8a, 8b) and hydnocarpin D (minor product, 9a,
used in the literature) without attacking other hydroxy groups, 9b). 2,3-Olefination was further proven by hydrolysis of
as reported by Wang et al. [9], and by our group [10]. Two dif- compound 9a, which converted the esters into the racemic mix-
ferent sequences of addition of reactants were applied, i.e., ad- ture of (10R,11R)- and (10S,11S)-hydnocarpin D (enantiomers
dition of solution of acid to the mixture of silibinin, triphenyl- 2a and 2b). We then systematically investigated dehydration by
phosphine (Ph3P) and diethyl azodicarboxylate (DEAD) [9], varying the amounts of Ph3P and DIAD, respectively (Table 1).
versus addition of diisopropyl azodicarboxylate (DIAD) to the It was observed that excessive Ph3P and DIAD shifted the ratio
mixture of silibinin, acid and Ph3P [10]. In both cases, the of products toward the desired hydnocarpin D esters, while
desired C-23 ester was obtained in moderate yields, and we ob- using a lower amount of Mitsunobu reagents favored the sili-
served a side reaction, which we expected to be a dehydration binin ester formation (Table 1, entries 1 and 5). In absence of an
reaction.
acid, 2,3-dehydration did not occur even using excessive
amounts of Ph3P and DIAD (Table 1, entries 7 and 8). In the
We investigated the reaction under Mitsunobu conditions using absence of DIAD, silibinin was not converted into any other
the latter sequence of addition [10] for silibinin and benzoic compound (Table 1, entry 9). Benzoic acid had been applied to
acid, or 2,2-dimethyl-3-(nitrooxy)propanoic acid (7, Scheme 1), synthesize 23-O-benzoylsilibinin (8b) in 55% yield [9]. Using
an NO-donor compound with several biological properties, e.g., benzoic acid with 6 equiv of Ph3P and 4 equiv of DIAD, sili-
vasorelaxation [27]. Compound 7 had initially been used by us binin was converted into ester 9b with a yield of 33% after
since we wanted to obtain hybrid esters for medicinal chemical purification (Table 1, entry 4). Again hydrolysis of 9b yielded
purposes [27]. As expected, in all cases, two products were hydnocarpin D (2).
detected in different ratios according to the different ratios of
Ph3P and DIAD used (Table 1, entries 1–4). Spectral data of the Being encouraged by the feasibility of using a common acid but
two products confirmed their structures as esters of silibinin still disappointed by the yields, we carefully monitored the
Scheme 1: Synthesis of ester derivatives of silibinin and conversion to hydnocarpin-type compounds. Reaction conditions: a) 7 or benzoic acid, Ph3P,
diisopropyl azodicarboxylate (DIAD), THF, 10 °C to rt, 3 h.; b) LiOH·H2O, MeOH/H2O, rt, 2 h.
665

Beilstein J. Org. Chem. 2016, 12, 662–669.
Table 1: Reaction conditions for optimization of dehydration.
Reaction conditions
Products
Ph3P
DIAD
Acid
1
2
3
4
5
6
7
8
9
2.5 equiv
6.0 equiv
6.0 equiv
6.0 equiv
4.0 equiv
6.0 equiv
2.0 equiv
6.0 equiv
6.0 equiv
2.5 equiv
3.0 equiv
3.0 equiv
4.0 equiv
2.0 equiv
4.0 equiv
2.0 equiv
4.0 equiv
–
7, 2.0 equiv
7, 3.0 equiv
PhCO2H, 3.0 equiv
PhCO2H, 3.0 equiv
7, 2.0 equiv
7, 3.0 equiv
–
8a and 9a (39% and 6%)a
8a and 9a (11:41)b
8b and 9b (19:81)b
9b (33%)a
8a (82%)a
9a (40%)a
no reaction
–
no reaction
7, 3.0 equiv
no reaction
aIsolated yield. bRatio by HPLC, see Figures S1 and S2 in Supporting Information File 1.
process of the reaction and found that ester 8b was initially established. Also Guz et al. denote their compound as (±)-
formed, and it was then gradually converted into 9b by the hydnocarpin D, which means that no optically pure compound
addition of excess DIAD. In the course of the reaction, less was obtained [14].
polar new products were also formed, which we assumed
to be the oligo-esters. After total ester hydrolysis (of both The Mitsunobu reaction represents a powerful method to
mono- and oligo-esters) the hydnocarpin D-type entities convert primary and secondary alcohols into ester but also into
formed, therefore it was not necessary to isolate the monoester various derivatives. The mechanism is well described and
prior to hydrolysis. We chose p-nitrobenzoic acid to accom- includes the formation of the triphenylphosphine–DIAD adduct,
plish the reaction, since the corresponding esters are readily which then activates the alcohol making it a good leaving group
hydrolyzed [28]. A Mitsunobu reaction was conducted at susceptible to a nucleophilic attack. Application of the
higher temperature to prepare hydnocarpin D, and without sepa- Mitsunobu reaction for dehydration was also reported previ-
ration of the different esters formed the crude mixture was ously as shown in the above-mentioned cases [23-26]. The ratio
saponified in a one-pot way, further improving the (overall) between possible substitution and elimination products could
yield to 56%.
not be determined in our experiments, but conditions are de-
scribed now leading to the elimination product. Figure 3 shows
Our method is robust, simple and the starting material (silibinin) a putative reaction mechanism leading to the formation of the
is cheap and readily available. As mentioned above, natural sili- eliminated product.
binin represents a mixture of two diastereomers, therefore the
dehydration product is also composed of two stereomers (enan- As (+)-catechin (10, Scheme 2) cannot produce the desired
tiomers). The use of diastereomerically pure compounds dehydrated compound under our conditions, we assume the car-
(
silybin A and silybin B, each of them with a de of 94%), there- bonyl group is necessary in the reaction described here, since
fore yielded the individual enantiomers (10R,11R)- and then β-elimination generates a thermodynamically more stable
10S,11S)-hydnocarpin D (2a, 2b). We measured optical rota- compound (Scheme 2). Therefore, flavanones dehydration
(
tions and ECD and confirmed the proposed structures (see under Mitsunobu conditions requires a 3-hydroxyflavanone
Figures S3 and S4 in Supporting Information File 1). To our (flavanonol) structure.
knowledge, this is the second case only [22] that these two en-
antiomers were obtained as pure and fully characterized Accordingly, we converted other 3-hydroxyflavanones into
compounds, namely by chiroptical methods. It should be their dehydrated analogs: isosilybin A (containing typically 5%
mentioned that previous studies on hydnocarpin D failed to de- of the B isomer), and silychristin A yielded hydnocarpin (24%),
termine its absolute configuration. In 1973 Ranganathan and and isohydnocarpin (22%), respectively. This demonstrates the
Seshadri isolated hydnocarpin D from H. wightiana for the first broad applicability of this method and provides until now the
time and determined the stereochemistry of the substituents at only semi-synthetic preparation of these flavonolignans
C-10 and C-11 [12]. However, neither in this nor in later papers (Scheme 3). Therefore, our method provides a simple and
on hydnocarpin D the absolute configuration had been properly robust semi-synthetic access to all hydnocarpins with pro-
666

Beilstein J. Org. Chem. 2016, 12, 662–669.
Figure 3: Putative mechanism of dehydration of flavanonols under Mitsunobu conditions.
Scheme 2: Attempt to dehydrate catechin. Reagents and conditions: a) p-nitrobenzoic acid, Ph3P, DIAD, THF, rt, 20 h; b) 2 N NaOH, rt, 1 h.
nounced biological activities. When applying the optimized and gives 56% yield starting from commercially available sili-
Mitsunobu conditions for elimination of 3-hydroxyflavanone to binin in a one-pot reaction. Evaluation of Mitsunobu conditions
silydianin (11, Scheme 3), another major component of the sily- and reagents applied for esterification and dehydration, respec-
marin complex, decomposition was observed yielding several tively, enabled us to exclusively obtain either the hydnocarpin
products (Scheme 3). Since the cyclic hemiacetal structure (of a compound esters (and therefore hydnocarpin-type compounds
diketone) represents the only functional difference to the other after hydrolysis) or esterification. The recently reported method
flavanonols employed, it is obviously unstable under the condi- by Vimberg et al. [22], that was published during the prepara-
tions used.
tion of this article, describes a four-step synthesis using
Vilsmeier–Haack conditions. Our study exhibits good atom
economics and remains the only specific application of
Conclusion
To our knowledge, this is the first semi-synthesis of optically Mitsunobu conditions for a one-pot dehydration of this impor-
pure (10R,11R)- and (10S,11S)-hydnocarpin D described to date tant class of natural products.
667

Beilstein J. Org. Chem. 2016, 12, 662–669.
Scheme 3: Preparation of hydnocarpin (4) and isohydnocarpin (6) and attempt to dehydrate silydianin A (11). Reagents and conditions: a) p-nitro-
benzoic acid, Ph3P, DIAD, THF, rt, 20 h; b) 2 N NaOH, rt, 1 h.
References
Supporting Information
1.
2.
3.
Gažák, R.; Walterová, D.; Křen, V. Curr. Med. Chem. 2007, 14,
15–338. doi:10.2174/092986707779941159
Abenavoli, L.; Capasso, R.; Milic, N.; Capasso, F. Phytother. Res.
010, 24, 1423–1432. doi:10.1002/ptr.3207
3
Supporting Information File 1
Experimental procedures, chiroptical and spectral data of
compounds 2, 2a, 2b, 4, 6, 8a, 9a and 9b.
2
Vogel, G.; Tuchweber, B.; Trost, W.; Mengs, U.
Toxicol. Appl. Pharmacol. 1984, 73, 355–362.
doi:10.1016/0041-008X(84)90087-5
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-12-66-S1.pdf]
4.
5.
6.
Pelter, A.; Hänsel, R. Tetrahedron Lett. 1968, 9, 2911–2916.
doi:10.1016/S0040-4039(00)89610-0
Biedermann, D.; Vavříková, E.; Cvak, L.; Křen, V. Nat. Prod. Rep.
Acknowledgements
We wish to thank Dr. Lucie Bednarova (Institute of Organic
2014, 31, 1138–1157. doi:10.1039/C3NP70122K
McDonald, B. R.; Nibbs, A. E.; Scheidt, K. A. Org. Lett. 2015, 17,
Chemistry and Biochemistry, Prague) for optical measurements
98–101. doi:10.1021/ol503303w
and Dr. Christopher S. Chambers (Institute of Microbiology, 7. Yang, L. X.; Huang, K. X.; Li, H. B.; Gong, J. X.; Wang, F.; Feng, Y. B.;
Tao, Q. F.; Wu, Y. H.; Li, X. K.; Wu, X. M.; Zeng, S.; Spencer, S.;
Prague) for language proofreading and valuable comments.
Zhao, Y.; Qu, J. J. Med. Chem. 2009, 52, 7732–7752.
M. Decker acknowledges the German Science Foundation
doi:10.1021/jm900735p
(
DFG) for financial support (DFG DE 1546/6-1); V. Křen
acknowledges MSMT grant LD15085 (COST action CM 1407).
668

Beilstein J. Org. Chem. 2016, 12, 662–669.
8
9
1
.
Zarrelli, A.; Sgambato, A.; Petito, V.; De Napoli, L.; Previtera, L.;
Di Fabio, G. Bioorg. Med. Chem. Lett. 2011, 21, 4389–4392.
doi:10.1016/j.bmcl.2011.06.049
License and Terms
.
Wang, F.; Huang, K.; Yang, L.; Gong, J.; Tao, Q.; Li, H.; Zhao, Y.;
Zeng, S.; Wu, X.; Stöckigt, J.; Li, X.; Qu, J. Bioorg. Med. Chem. 2009,
This is an Open Access article under the terms of the
Creative Commons Attribution License
17, 6380–6389. doi:10.1016/j.bmc.2009.07.023
(http://creativecommons.org/licenses/by/2.0), which
0.Chen, X.; Zenger, K.; Lupp, A.; Kling, B.; Heilmann, J.; Fleck, C.;
Kraus, B.; Decker, M. J. Med. Chem. 2012, 55, 5231–5242.
doi:10.1021/jm300246n
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
1
1
1
1
1
1
1.Zenger, K.; Chen, X.; Decker, M.; Kraus, B. J. Pharm. Pharmacol.
2
013, 65, 1765–1772. doi:10.1111/jphp.12070
2.Ranganathan, K. R.; Seshadri, T. R. Tetrahedron Lett. 1973, 14,
481–3482. doi:10.1016/S0040-4039(01)86949-5
3.Antus, S.; Seligmann, O.; Wagner, H. Liebigs Ann. Chem. 1986,
47–654. doi:10.1002/jlac.198619860406
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
3
(http://www.beilstein-journals.org/bjoc)
6
The definitive version of this article is the electronic one
which can be found at:
4.Guz, N. R.; Stermitz, F. R. J. Nat. Prod. 2000, 63, 1140–1145.
doi:10.1021/np000166d
5.Chambers, C. S.; Valentová, K.; Křen, V. Curr. Pharm. Des. 2015, 21,
doi:10.3762/bjoc.12.66
5489–5500. doi:10.2174/1381612821666151002112720
6.Sahoo, M. R.; Dhanabal, S. P.; Jadhav, A. N.; Reddy, V.; Muguli, G.;
Babu, U. V.; Rangesh, P. J. Ethnopharmacol. 2014, 154, 17–25.
doi:10.1016/j.jep.2014.03.029
17.Jacobsen, P. L.; Levy, L. Antimicrob. Agents Chemother. 1973, 3,
373–379. doi:10.1128/AAC.3.3.373
1
8.Bueno Perez, L.; Pan, L.; Sass, E.; Gupta, S. V.; Lehman, A.;
Kinghorn, A. D.; Lucas, D. M. Phytother. Res. 2013, 27, 1735–1738.
doi:10.1002/ptr.4903
1
9.Stermitz, F. R.; Lorenz, P.; Tawara, J. N.; Zenewicz, L. A.; Lewis, K.
Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 1433–1437.
doi:10.1073/pnas.030540597
2
2
2
0.Sharma, D. K.; Hall, I. H. J. Nat. Prod. 1991, 54, 1298–1302.
doi:10.1021/np50077a010
1.Ismailova, G.; Khilya, V. P.; Aitmambetov, A. Chem. Nat. Compd.
1999, 35, 159–161. doi:10.1007/BF02234923
2.Vimberg, V.; Kuzma, M.; Stodůlková, E.; Novák, P.; Bednárová, L.;
Šulc, M.; Gažák, R. J. Nat. Prod. 2015, 78, 2095–2103.
doi:10.1021/acs.jnatprod.5b00430
2
2
2
2
2
2
3.Cherney, R. J.; Wang, L. J. Org. Chem. 1996, 61, 2544–2546.
doi:10.1021/jo951958t
4.Nikam, S. S.; Kornberg, B. E.; Rafferty, M. F. J. Org. Chem. 1997, 62,
3754–3757. doi:10.1021/jo961738v
5.Quader, S.; Boyd, S. E.; Jenkins, I. D.; Houston, T. A.
Org. Biomol. Chem. 2006, 4, 36–37. doi:10.1039/B513113H
6.Elson, K. E.; Jenkins, I. D.; Loughlin, W. A. Org. Biomol. Chem. 2003,
1
, 2958–2965. doi:10.1039/B305375J
7.Decker, M.; König, A.; Glusa, E.; Lehmann, J. Bioorg. Med. Chem. Lett.
004, 14, 4995–4997. doi:10.1016/j.bmcl.2004.07.014
2
8.Martin, S. F.; Dodge, J. A. Tetrahedron Lett. 1991, 32, 3017–3020.
doi:10.1016/0040-4039(91)80675-V
669