Improving Physical Properties via C-H Oxidation: Chemical and Enzymatic Approaches

Article abstract of DOI:10.1002/anie.201407016

Physicochemical properties constitute a key factor for the success of a drug candidate. Whereas many strategies to improve the physicochemical properties of small heterocycle-type leads exist, complex hydrocarbon skeletons are more challenging to derivatize because of the absence of functional groups. A variety of C-H oxidation methods have been explored on the betulin skeleton to improve the solubility of this very bioactive, yet poorly water-soluble, natural product. Capitalizing on the innate reactivity of the molecule, as well as the few molecular handles present on the core, allowed oxidations at different positions across the pentacyclic structure. Enzymatic oxidations afforded several orthogonal oxidations to chemical methods. Solubility measurements showed an enhancement for many of the synthesized compounds. A handle on [O]: A variety of C-H oxidation methods were explored on the betulin skeleton to improve the solubility of this bioactive, yet poorly water-soluble, natural product. The innate reactivity of the molecule, as well as the molecular handles present on the core, allowed oxidations at different positions. Solubility enhancement was observed for many of the synthesized compounds.

Full text of DOI:10.1002/anie.201407016

Angewandte
Chemie
DOI: 10.1002/anie.201407016
Terpenoids
À
Improving Physical Properties via C H Oxidation: Chemical and
Enzymatic Approaches**
Quentin Michaudel, Guillaume Journot, Alicia Regueiro-Ren, Animesh Goswami, Zhiwei Guo,
Thomas P. Tully, Lufeng Zou, Raghunath O. Ramabhadran, Kendall N. Houk, and
Phil S. Baran*
Abstract: Physicochemical properties constitute a key factor
for the success of a drug candidate. Whereas many strategies to
improve the physicochemical properties of small heterocycle-
type leads exist, complex hydrocarbon skeletons are more
challenging to derivatize because of the absence of functional
source of drug failure.[3] A subtle balance between lipophi-
licity and polarity is therefore required to ensure a viable
future for a drug candidate. In the case of heterocyclic leads,
this balance can be achieved by substituting some positions of
the heteroarenes with hydrogen-bond donors and/or accept-
ors, as well as various fluorinated motifs. When the drug
candidate is a natural product derivative, however, the quest
for an ideal solubility and cell permeability is often hampered
by the few positions which can be accessed. This is especially
true in the case of lowly oxidized terpenes, where only a few
carbon atoms of the skeleton can be modified using conven-
tional chemical means. Our laboratory has been developing
À
groups. A variety of C H oxidation methods have been
explored on the betulin skeleton to improve the solubility of
this very bioactive, yet poorly water-soluble, natural product.
Capitalizing on the innate reactivity of the molecule, as well as
the few molecular handles present on the core, allowed
oxidations at different positions across the pentacyclic struc-
ture. Enzymatic oxidations afforded several orthogonal oxi-
dations to chemical methods. Solubility measurements showed
an enhancement for many of the synthesized compounds.
À
various aliphatic C H functionalization methods and strat-
egies for a biomimetic ꢀoxidase phaseꢁ in natural-product
synthesis.[4] It was in this context that a collaboration was
forged with Bristol-Myers Squibb to probe the applicability of
these studies in drug discovery and more precisely the
improvement of drug physicochemical properties. The
lupane natural products betulin (1) and betulinic acid (2)
are ideal substrates to explore this concept because of their
promising in vitro bioactivity coupled with their extreme
insolubility, thus limiting their medicinal potential (Figure 1).
Outlined herein is a strategy for improving the solubility of
this promising natural-product class using both chemical and
enzymatic means.[5] This combination of orthogonal tech-
niques has led to a dramatic improvement in solubility of the
lupane core (Figure 1), thus setting the stage to realize their
full medicinal potential.
T
he rates of attrition for small-molecule drug candidates
remain excessively high at any stage of drug development
despite a continuously improving understanding of drug–
target interactions, as well as drug distribution and metabo-
lism.[1] Consequently, the cost of discovering and developing
a drug in 2010 was estimated to be approximately $1.8
billion.[2] Historically, some of the main reasons behind drug
failure are the lack of efficacy, toxicity, and poor pharmaco-
kinetic properties or bioavailability. Unsatisfactory physico-
chemical properties have specifically emerged as a common
[*] Q. Michaudel, Dr. G. Journot, Prof. Dr. P. S. Baran
Department of Chemistry, The Scripps Research Institute
10550 North Torrey Pines Road, La Jolla, CA 92037 (USA)
E-mail: pbaran@scripps.edu
Betulin (1) and betulinic acid (2) are two natural
pentacyclic triterpenes isolated from the bark of the birch
tree and only differ by the oxidation at C28 (Figure 1).[6]
Betulin is more abundant than its carboxylic acid counterpart
but is generally less bioactive.[7] Several methods have been
developed to convert 1 into 2.[8] Betulinic acid displays many
intriguing pharmacological properties such as anti-inflamma-
tory, anticancer, and anti-HIV, the latter two being the most
promising for pharmaceutical applications.[7,9] However, the
low solubility of 2 in H2O constitutes a serious limitation for
its use as a therapeutic, since it would make formulation for
oral delivery difficult.[7] To render 1 and 2 more soluble in
aqueous media, the lipophilic carbon skeleton must be
decorated with heteroatoms. However, oxidation of the
lupane core presents many challenges because of the very
few functional groups present on the skeleton. Prior art has
mainly focused on oxidation of the A ring that already
possesses an alcohol group at C3 as a handle,[10] and
A. Regueiro-Ren
Discovery Chemistry, Bristol-Myers Squibb
5 Research Parkway, Wallingford, CT 06492 (USA)
A. Goswami, Z. Guo, T. P. Tully
Chemical Development, Bristol-Myers Squibb
One Squibb Drive, New Brunswick, NJ 08903 (USA)
L. Zou, R. O. Ramabhadran, K. N. Houk
Department of Chemistry and Biochemistry, University of California,
Los Angeles, Los Angeles, CA 90095-1569 (USA)
[**] Financial support for this work was provided by the NIH
(GM097444-03) and the Swiss National Science Foundation
(postdoctoral fellowship to G.J.). We are grateful to Prof. A.
Rheingold and Dr. C. E. Moore (UCSD) for X-ray crystallographic
analysis, Dr. D.-H. Huang, Dr. L. Pasternack (TSRI), Dr. Frank A.
Rinaldi, Dr. Xiaohua Huang, and Jacob Swidorski (BMS) for
assistance with NMR spectroscopy, Gerry Everlof for solubility
measurements and Dr. John Kadow for insightful discussions.
À
modification of the isopropenyl group by allylic C H
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201407016.
oxidation or direct transformation of the alkene.[10c,11] In
Angew. Chem. Int. Ed. 2014, 53, 1 – 7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Dozens of oxidants were screened,
but most of them resulted in no
reaction or an inextricable mixture
of products, including the reported
CrO3 conditions.[13a,b] Fortunately,
it was found that methyl(tri-
fluoromethyl)dioxirane (TFDO)
afforded the C16 ketone product
14 in 30% yield as a major product,
as confirmed by X-ray crystallog-
raphy. No minor products were
isolated in quantities sufficient for
characterization, but it is of note
that none of these products con-
tained
a ketone motif. While
TFDO has been previously used
to selectively oxidize a sesquiter-
pene skeleton in the total synthesis
of eudesmantetraol,[4a] this reac-
tion constitutes the first example
of selective methylene oxidation in
À
Figure 1. Diversification of the lupane core by C H oxidation.
the context of a complex triterpene
substrate. The striking selectivity
contrast, known oxidations of the B, C, and D rings are scarce
and have mostly utilized the C28 alcohol as a directing
group.[12] To the best of our knowledge, only one group has
reported a nondirected oxidation of lupanes, leading to
functionalization of the C ring.[13] However, the reaction
conditions (CrO3 in boiling acetic acid) afforded many
compounds in yields all below 3%.[13a,b] Finally, it is of note
for C16 cannot be explained solely by electronic factors since
six carbon atoms are more electron-rich based on 13C NMR
spectroscopy.
Calculations using the UB3LYP functional were per-
formed to determine the origins of this surprising selectivity.
Activation energies for attack at the ring carbon hydrogen
bonds by dimethyldioxirane (DMDO) were computed (Fig-
À
À
that several C H bonds can be biochemically oxidized by
ure 3A). Based on our previous computational studies, C H
oxidation using DMDO or TFDO involves a concerted
feeding 2 to microorganisms.[7a]
À
With that precedent in mind, and by analogy to studies
reaction, initiated by C H abstraction followed by a barrier-
conducted on eudesmane terpenes,[4a] two different strategies
less oxygen rebound from a radical pair.[17] DMDO is less
À
were applied to oxidize various unactivated aliphatic C H
reactive than TFDO and more selective.[17] The calculated
ꢀ
À
À
bonds of 1 or 2. Thus, nondirected (innate) C H oxidation
DH value for the oxidation of the equatorial C16 H is found
to be 3.5 kcalmolÀ1 lower than that of C6 H and is consid-
À
À
would capitalize on the default reactivity of the 48 C H bonds
of 1 (46 for 2), whereas directed (guided) C H oxidation
À
erably lower than activation energies computed for all other
positions (see Figure 3A). This difference drops to 1.6 kcal
molÀ1 with TFDO (see the Supporting Information). The
abstraction of the C16 equatorial hydrogen atom is consistent
would employ the hydroxy groups at C3, C28, or C20 (arising
from hydration of the alkene).[14] To achieve innate C H
À
oxidation, the two hydroxy groups and the alkene required
protection, since they are the most prone to oxidation.
Carreiraꢁs hydration conditions[15] followed by diacetylation
afforded substrate 12 in good yield (Figure 2B). X-ray
crystallographic analysis of 12 suggested that it would be
oxidized at a methylene position since all the methine
positions are fairly hindered. The electronic character of all
À
with our previous finding that equatorial C H bonds are
more reactive because of both the strain-release effect[17,18]
and higher steric accessibilities. Thus, Figure 3B shows an
oxygen atom of TFDO just touching the surface of the
hydrogen at C16 and C6. The former occurs with no clashing
À
of van der Waals surfaces, while approach at C6 H involves
the C H bonds was evaluated through 13C NMR studies.[4a,16]
significant steric repulsions of the dimethyl group at C4 of
betulin and the CF3 groups of TFDO. These interactions are
confirmed in the transition state geometries given in Fig-
ure S13 in the Supporting Information. The stability of
radicals formed by hydrogen abstraction at all positions
À
Every carbon peak was unambiguously assigned using 13C–13C
INADEQUATE NMR spectroscopy and the chemical-shift
trend was determined as follows (see the Supporting Infor-
mation for actual values):
À
were also calculated (Figure 3A). Remarkably, the C H
oxidation process is extremely selective, and hydrogen
abstraction occurs only at C16. This selectivity is present
despite the fact that the diradical transition state has
appreciable diradical character at C16 in 12, where a radical
is intrinsically of low stability. Finally, calculations on an
analogue of 12 where the C17 CH2OAc group is replaced by
dC6 < dC11 < dC2 < dC15 ꢀ dC21 ꢀ dC12 ꢀ dC16 < dC22 ꢀ dC7 < dC1
!
dC28
Based on this analysis, the C6 to C16 positions are the most
electron-rich methylenes, and thus most likely to be oxidized.
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 7
These are not the final page numbers!

Angewandte
Chemie
Figure 2. Divergent synthesis of oxidized compounds from a) betulinic acid and b) betulin. Reagents and conditions: a) BnBr, K2CO3, 608C, 3.5 h,
92%; b) 1 mol% [{Ir(cod)OMe}2], Et2SiH2, RT, 12 h; c) 2 mol% [{Ir(cod)OMe}2], 5 mol% Me4Phen, norbornene, 1208C, 36 h; d) H2O2, KHCO3,
508C, 12 h, 24% (over 3 steps); e) 50 mol% Pd(OAc)2, Et3N, tBuMe2SiH, 608C, 2.5 h; f) TBAF, RT, 1.5 h, 70% (over 2 steps); g) Sodium-bis-[N-
salicylidene-2-amino-isobutyrato]-cobaltate(III) (20), PhSiH3, O2, RT, 10 h; h) Ac2O, pyr, RT, 99% (70% for 12 over 2 steps); i) 20, Et3SiH, O2, RT,
14 h, 71%; j) TFDO, 08C, 2 h, 30%; k) Zn(BH4)2, RT, 1 h, 80% (1:1 d.r.); l) HCl, 08C (À418C for 4); m) NaOH, RT, 12 h, 72% for 4 (over 2 steps)
(70% for 3 (over 3 steps), 92% for 7); n) LiAlH4, À958C, 1 h; o) I2, PIDA, hn, RT, 15 min, 30% of 16 from 14; 34% of 15 and 15% of 16 from 13;
p) 40 mol% 4-NHAcTEMPO, PIDA, RT 3 h; q) NaClO2, NaH2PO4·H2O, 2-methyl-2-butene, RT, 2.5 h, 73% (over 3 steps); r) Al(OiPr)3, iPrOH,
1008C, 1.5 h; s) Pd/C, H2, RT, 12 h, 79% (over 3 steps); t) I2, Pb(OAc)4, hn, RT, 15 min; u) AgOAc, RT, 20 h, 50% (over 2 steps); v) LiAlH4, 08C,
1 h, 74%. cod=1,5-cyclooctadinene, TBAF=tetra-n-butylammonium fluoride, PIDA=phenyliodine(III) diacetate, TEMPO=2,2,6,6-tetramethyl-
piperidin-1-oxyl.
a methyl group (see the Supporting Information) showed that
Derivative 14 was then reduced to the axial alcohol using
LiAlH4 and 3 was obtained after elimination and saponifica-
tion. Zn(BH4)2 reduction of 14 afforded a 1:1 mixture of
diastereomers which could be separated, and equatorial
alcohol 4 was obtained with a similar sequence as that for 3.
À
the C H abstraction barrier at C16 is lowered by 1.6 kcal
molÀ1 by removing the inductively electron-withdrawing
ester, thus ruling out a potential neighboring effect from the
acetate group.
Angew. Chem. Int. Ed. 2014, 53, 1 – 7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
attention.[25] To examine this directed oxidation, peroxide 13
was synthesized in two steps in a fashion very similar to that of
12. Using triethylsilane instead of phenylsilane and removing
the reductive work-up allowed the preparation of the desired
peroxide. This substrate was then subjected to the same
Suꢂrez conditions and two products were isolated. The minor
product was the previously isolated tetrahydropyran 16 (15%
yield), whereas the major product was the desaturated
compound 15 (34% yield). This compound corresponds to
Figure 3. Analysis of site selectivities on 12 for TFDO. a) Relative
À
activation enthalpies for equatorial C H abstraction by DMDO
(DDHꢀ) and relative stabilities of the free radicals (DDE, in parenthe-
ˇ
´
Cekovicꢁs findings, but interestingly, their exact conditions
applied to 13 only afforded 16. However, adding Cu(OAc)2 to
our reaction conditions enabled a slight improvement of the
alkene yield, which is consistent with a mechanism involving
hydrogen abstraction, followed by radical oxidation and
proton elimination.[25f] Hydrolysis of the acetate groups
delivered 7 which could serve as a platform for further
modifications of the C ring.
ses) (kcalmolÀ1). b) Juxtaposition of space-filling models of TFDO
À
À
approaching C6 H and C16 H in 12, to indicate the steric effects
À
À
disfavoring C H abstraction from C6 H. The relative orientations of
TFDO and 12 were determined from the transition-state geometries
shown in Figure S13 in the Supporting Information.
The structure of 4 has been attributed to the natural product
heliantriol B2 in the literature. However, our NMR data did
not perfectly match the reported data of this natural
product,[19] which suggests a potential misassignment of
heliantriol B2, since the structure of 4 was confirmed by X-
ray crystallography. However, the lack of details in the
isolation report might explain the observed discrepancies.
Functionalization of the A ring was then pursued using
the C3 alcohol as a directing group. After benzylation of 2,
Hartwigꢁs elegant 1,3-diol methodology was applied.[20] Sily-
A few reports have described oxidation of C13 using
a hydroxyl radical at C28. The substrate for this transforma-
tion is the hydrogenated monoacetate compound 17, and
different authors reported the formation of the corresponding
tetrahydrofuran product, albeit generally in low yields, after
thermal or photoactivation of the hydroxyl radical. In our
hands, however, this product could not be isolated despite the
many reaction conditions that were tried. Instead, the
formation of an iodinated product was observed when using
lead tetraacetate and iodine under a sunlamp.[26] This product
could not be purified and was subsequently treated with silver
acetate in acetone.[22a] This two-step procedure initiated
a rearrangement of the carbon skeleton, thus providing 18
in 50% yield (X-ray analysis), presumably by a 1,2-shift of the
À
lation of the C3 alcohol, followed by C H silylation at C23
and Tamao–Fleming oxidation afforded 11 in 24% yield
(Figure 2A). A two-step debenzylation delivered the natural
product C23-hydroxybetulinic acid (5). Compound 5 has been
previously synthesized from betulinic acid using a Baldwin-
[21]
À
À
type C H oxidation. In that study, transformation of the
C13 C14 bond. From a medicinal chemistꢁs point of view, this
À
alcohol into an oxime directing group, C H oxidation with
unexpected transformation allows the exploration of the
bioactivity and physicochemical properties of a new family of
compounds possessing an intriguing 6,6,5,7,5 pentacyclic
structure. Finally, reduction of 18 with an excess of LiAlH4
afforded triol 6.
stoichiometric palladium, and then two-step regeneration of
the alcohol was required. Hartwigꢁs conditions avoided these
four functional-group manipulation steps and do not neces-
sitate protection of the C20–C21 alkene.
Functionalization of the C ring was achieved via the
intermediacy of a hydroxyl radical at C20. Indeed, an oxygen
radical could be generated using Suꢂrezꢁs conditions on
In parallel to these chemical studies, BMS screened
several microorganisms to oxidize 2 (Figure 4). It was found
that Streptomyces fragilis tranformed 2 into diol 9, whereas
Bacillus megaterium afforded diol 10 instead.[27]
substrate 12,[22] resulting in C H abstraction, iodine trap, and
À
displacement leading to the formation of tetrahydropyran
product 16 in 30% yield. While this type of transformation
generally leads to tetrahydrofuran formation, the specific
geometry of the system results in 1,6-hydrogen abstraction
rather than 1,5-hydrogen abstraction. Compound 16 has been
previously synthesized in a related transformation, but using
lead tetraacetate in refluxing benzene for several hours.[23]
Derivative 16 was then transformed into carboxylic acid 8 by
removal of the acetate groups and a selective two-step
oxidation of the primary alcohol. The tetrahydropyran ring
in 16 proved to be very stable, and no hydrolysis conditions
could be developed. Consequently, the venerable Barton
nitrite ester reaction was considered to install a ketone at C12
in place of the cyclic ether.[24] Unfortunately, formation of the
nitrite ester of the tertiary alcohol with NOCl in pyridine was
ineffective, probably as a result of steric hindrance. The
Figure 4. Microbial oxidations of betulinic acid (2).
À
The panel of aliphatic C H oxidations that were devel-
oped on the lupane core are summarized in Figure 5. Alcohol
12 can serve both for the directed oxidation of C12 and
nondirected oxidation of C16 with TFDO (entries 1, 2, and 4).
No other oxidants were found to enable the same reaction
À
benzylic C H oxidation directed by a peroxyl radical
ˇ
´
reported by both Kropf and Cekovic then caught our
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 7
These are not the final page numbers!

Angewandte
Chemie
in assay 2 than that of 2, albeit somewhat lower in assay 1.
Consequently, adding one hydroxy group to 1 and 2 drasti-
cally improved their solubility. This work clearly demon-
strates the unpredictable nature of skeletal oxidation on
solubility since 4 has almost identical solubility to 1 despite
having one more alcohol, and diols 9 and 10 are less soluble
than 2 (only a slight improvement seen in assay 2). Finally, the
solubility of the rearranged product 6 is the lowest of all the
family.
To the best of our knowledge, this is the first published
systematic study using both chemical and enzymatic methods
À
for aliphatic C H oxidation with the goal of improving
physical properties. It is our hope that this study will help to
alleviate some of the empiricism in the process of screening
oxidants and that the combination of spectroscopy, crystal-
À
lography, and computation along with a systematic C H
À
oxidation panel will serve to accelerate future efforts. C H
oxidation clearly merits consideration by medicinal chemists
confronted with the challenge of an exciting chemotype
whose further development is hampered by extreme insol-
ubility.
Received: July 8, 2014
Published online: && &&, &&&&
À
Figure 5. Summary of aliphatic C H oxidation panel used on the
lupane skeleton. [a] R1 =CO2Bn.
À
Keywords: C H activation · medicinal chemistry · oxidation ·
radical reactions · terpenoids
.
(entry 5–9).[28] From hydroperoxide 13 arose alkene 15 and
ether 16, two compounds oxidized at C12 (entry 1). Alcohol
17 is the precursor of the C13 oxidation followed by skeletal
rearrangement (entry 2) and benzylated 2 is that of C23
oxidation (entry 10). Interestingly, all the developed enzy-
[1] a) I. Kola, J. Landis, Nat. Rev. Drug Discovery 2004, 3, 711; b) M.
Hay, D. W. Thomas, J. L. Craighead, C. Economides, J. Rose-
ntha, Nat. Biotechnol. 2014, 32, 40.
À
À
[2] S. M. Paul, D. S. Mytelka, C. T. Dunwiddie, C. C. Persinger, B. H.
Munos, S. R. Lindborg, A. L. Schacht, Nat. Rev. Drug Discovery
2010, 9, 203.
[3] N. A. Meanwell, Chem. Res. Toxicol. 2011, 24, 1420.
[4] a) K. Chen, P. S. Baran, Nature 2009, 459, 824; b) Y. Ishihara,
P. S. Baran, Synlett 2010, 1733; c) H. Renata, Q. Zhou, P. S.
Baran, Science 2013, 339, 59; d) L. Jørgensen, S. J. McKerrall,
C. A. Kuttruff, F. Ungeheuer, J. Felding, P. S. Baran, Science
2013, 341, 878.
matic C H oxidations are orthogonal to the chemical C H
oxidations (entry 11 and 12), which illustrates the comple-
mentarity of these two approaches.
The solubility of these new oxidized compounds was
tested in simulated intestinal fluids and compared to refer-
ence compounds (1 or 2) possessing the same oxidation state
at C28 (Table 1). The solubility was measured in both fasted
state (assay 1) and fed state (assay 2).[29] The solubility of 3
and 7 in assay 1 is over 100-fold higher than that of 1, while
nearly identical in assay 2. The solubility of 8 is 17-fold higher
À
[5] A dual chemical/enzymatic C H oxidation strategy on pharma-
ceutical substrates was previously published. See: a) J. Geno-
vino, S. Lꢃtz, D. Sames, B. B. Tourꢄ, J. Am. Chem. Soc. 2013, 135,
12346. For a purely chemical approach to both predict and block
À
metabolic hot spots using C H functionalization, see: b) F.
Table 1: Relative solubility enhancement of the oxidized compounds.
OꢁHara, A. C. Burns, M. R. Collins, D. Dalvie, M. A. Ornelas,
A. D. N. Vaz, Y. Fujiwara, P. S. Baran, J. Med. Chem. 2014, 57,
1616.
Entry Substrate R1
Relative Solubility
Enhancement:
Relative Solubility
Enhancement:
Assay 1 (FaSSIF)[a]
Assay 2 (FeSSIF)[b]
[6] a) J. T. Lçwitz, Crellꢀs Chem. Ann. 1788, 1, 312; b) V. Bruckner, J.
Kovacs, I. Koczka, J. Chem. Soc. Abstr. 1948, 948; c) E. W. H.
Hayek, U. Jordis, W. Moche, F. Sauter, Phytochemistry 1989, 28,
2229.
[7] For reviews, see: a) R. H. Cichewicz, S. A. Kouzi, Med. Res. Rev.
2003, 24, 90; b) S. Alakurtti, T. Mꢅkelꢅ, S. Koskimies, J. Yli-
Kauhaluoma, Eur. J. Pharm. Sci. 2006, 29, 1.
[8] a) L. Ruzicka, A. H. Lamberton, E. W. Christie, Helv. Chim.
Acta 1938, 21, 1706; b) A. Robertson, G. Soliman, E. C. Owen, J.
Chem. Soc. 1939, 1267; c) D. S. H. L. Kim, Z. Chen, V. T.
Nguyen, J. M. Pezzuto, S. Qiu, Z. Lu, Synth. Commun. 1997,
27, 1607.
1
2
3
4
5
6
7
8
3
4
7
6
5
8
9
10
CH2OH 274ꢀ
CH2OH 8.00ꢀ
CH2OH 121ꢀ
no change
0.077ꢀ
0.357ꢀ
CH2OH no change
0.077ꢀ
CO2H
CO2H
CO2H
CO2H
0.056ꢀ[c]
0.112ꢀ[c]
0.019ꢀ[c]
0.002ꢀ[c]
0.115ꢀ[c]
17.4ꢀ[c]
3.38ꢀ[c]
0.462ꢀ[c]
[a] Solubility ratio substrate/1 in the fasted state simulated intestinal
fluid. [b] Solubility ratio substrate/1 in the fed state simulated intestinal
fluid. [c] Solubility ratio substrate/2. R1 refers to the position shown in
the structure of Figure 5 (C17).
[9] For reviews, see: a) E. Pisha, H. Chai, I. S. Lee, T. E. Chagwe-
dera, N. R. Farnsworth, G. A. Cordell, C. W. W. Beecher,
Angew. Chem. Int. Ed. 2014, 53, 1 – 7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
H. H. S. Fong, A. D. Kinghorn, D. M. Brown, M. C. Wani, M. E.
Wall, T. J. Hieken, T. K. Das Gupta, J. M. Pezzuto, Nat. Med.
1995, 1, 1046; b) E. De Clercq, Med. Res. Rev. 2000, 20, 323.
[10] a) Y. Deng, J. K. Snyder, J. Org. Chem. 2002, 67, 2864; b) R.
Mukherjee, M. Jaggi, M. J. A. Siddiqui, S. K. Srivastava, P.
Rajendran, A. Vardhan, A. C. Burman, Bioorg. Med. Chem.
Lett. 2004, 14, 4087; c) R. C. Santos, J. A. R. Salvador, S. Marꢆn,
M. Cascante, Bioorg. Med. Chem. 2009, 17, 6241.
[16] For the first demonstration that 13C shifts could be used to assist
À
in the prediction of C H oxidation, see Ref. [4a]. See also: M. A.
Bigi, P. Liu, L. Zou, K. N. Houk, M. C. White, Synlett 2012, 2768.
[17] L. Zou, R. S. Paton, A. Eschenmoser, T. R. Newhouse, P. S.
Baran, K. N. Houk, J. Org. Chem. 2013, 78, 4037.
[18] K. Chen, A. Eschenmoser, P. S. Baran, Angew. Chem. Int. Ed.
2009, 48, 9705; Angew. Chem. 2009, 121, 9885.
[19] J. S. Pyrek, Pol. J. Chem. 1979, 53, 2465.
[20] a) E. M. Simmons, J. F. Hartwig, Nature 2012, 483, 70; b) B. Li,
M. Driess, J. F. Hartwig, J. Am. Chem. Soc. 2014, 136, 6586.
[21] F. Sun, P. Zhu, H. Yao, X. Wu, J. Xu, J. Chem. Res. 2012, 254.
[22] a) J. I. Concepción, C. G. Francisco, R. Hernꢂndez, J. A. Salazar,
E. Suꢂrez, Tetrahedron Lett. 1984, 25, 1953; b) B. M. Trost, H. C.
Shen, J.-P. Surivet, J. Am. Chem. Soc. 2004, 126, 12565.
[11] a) E. Suokas, T. Hase, Acta Chem. Scand. Ser. B 1975, 29, 139;
b) E. Suokas, T. Hase, Acta Chem. Scand. Ser. B 1977, 31, 182;
c) G. Dutta, S. N. Bose, Tetrahedron Lett. 1988, 29, 5807; d) I. C.
Sun, H. K. Wang, Y. Kashiwada, J. K. Shen, L. M. Cosentino,
C. H. Chen, L. M. Yang, K. H. Lee, J. Med. Chem. 1998, 41, 4648;
ˇ
ˇ
e) J. Sarek, J. Klinot, P. Dzubꢂk, E. Klinotovꢂ, V. Noskovꢂ, V.
ˇ
[23] V. Pouzar, J. Protiva, E. Lisꢂ, E. Klinotovꢂ, A. Vystrcil, Collect.
ˇ ˇ
ˇ
ˇ
Krecek, G. Korꢆnkovꢂ, J. O. Thomson, A. Janostꢁꢂkovꢂ, S. Wang,
S. Parsons, P. M. Fischer, N. Z. Zhelev, M. Hajdfflch, J. Med.
Chem. 2003, 46, 5402; f) S. Alakurtti, T. Heiska, A. Kiriazis, N.
Sacerdoti-Sierra, C. L. Jaffe, J. Yli-Kauhaluoma, Bioorg. Med.
Chem. 2010, 18, 1573.
Czech. Chem. Commun. 1975, 40, 3046.
[24] a) D. H. R. Barton, J. M. Beaton, J. Am. Chem. Soc. 1960, 82,
2640; b) D. H. R. Barton, J. M. Beaton, L. E. Geller, M. M.
Pechet, J. Am. Chem. Soc. 1961, 83, 4076.
[25] a) H. Kropf, H. von Wallis, Synthesis 1981, 237; b) H. Kropf, H.
von Wallis, Synthesis 1983, 633; c) H. Kropf, H. von Wallis,
Liebigs Ann. Chem. 1983, 610; d) H. Kropf, H. von Wallis, L.
[12] a) J. M. Allison, W. Lawie, J. McLean, G. R. Taylor, J. Chem.
ˇ
Soc. 1961, 3353; b) A. Vystrcil, J. Protiva, Collect. Czech. Chem.
Commun. 1974, 39, 1382; c) S. K. Nag, S. N. Bose, Tetrahedron
Lett. 1989, 30, 2855.
ˇ
ˇ
ˇ
´
Bartnick, Liebigs Ann. Chem. 1983, 624; e) Z. Cekovic, M. M.
ˇ
´
Green, J. Am. Chem. Soc. 1974, 96, 3000; f) Z. Cekovic, Lj.
ˇ ˇ
´
[13] a) J. Sejbal, J. Klinot, M. Budesꢆnsky, J. Protiva, Collect. Czech.
´
´
´
Dimitrijevic, G. Djokic, T. Srnic, Tetrahedron 1979, 35, 2021.
[26] J. Kalvoda, K. Heusler, Synthesis 1971, 601.
Chem. Commun. 1991, 56, 2936; b) J. Sejbal, J. Klinot, M.
ˇ ˇ
´
Budesꢆnsky, J. Protiva, Collect. Czech. Chem. Commun. 1997, 62,
1905. Oxidation of C19 using dry ozonation or peracids in ca.
10% yield have been disclosed as well: c) E. Suokas, T. Hase,
Acta Chem. Scand. Ser. B 1978, 32, 623; d) M. Tori, R. Matsuda,
Y. Asakawa, Chem. Lett. 1985, 167; e) J. Sejbal, J. Klinot, A.
[27] A manuscript disclosing the full biotransformation studies is in
preparation.
[28] a) S. Lee, P. L. Fuchs, J. Am. Chem. Soc. 2002, 124, 13978; b) E.
McNeill, J. Du Bois, J. Am. Chem. Soc. 2010, 132, 10202; c) J.
Graeupner, T. P. Brewster, J. D. Blakemore, N. D. Schley, J. M.
Thomsen, G. W. Brudvig, N. Hazari, R. H. Crabtree, Organo-
metallics 2012, 31, 7158.
[29] For the use of physiologically relevant media like FeSSIF or
FaSSIF for solubility measurements of drug-like substrates, see:
J. H. Fagerberg, O. Tsinman, N. Sun, K. Tsinman, A. Avdeef,
C. A. S. Bergstrçm, Mol. Pharm. 2010, 7, 1419.
ˇ
Vystrcil, Collect. Czech. Chem. Commun. 1987, 52, 487.
[14] T. Brꢃckl, R. D. Baxter, Y. Ishihara, P. S. Baran, Acc. Chem. Res.
2012, 45, 826.
[15] J. Waser, B. Gaspar, H. Nambu, E. M. Carreira, J. Am. Chem.
Soc. 2006, 128, 11693.
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 7
These are not the final page numbers!

Angewandte
Chemie
Communications
Terpenoids
Q. Michaudel, G. Journot,
A. Regueiro-Ren, A. Goswami, Z. Guo,
T. P. Tully, L. Zou, R. O. Ramabhadran,
K. N. Houk, P. S. Baran*
&&&& — &&&&
À
Improving Physical Properties via C H
Oxidation: Chemical and Enzymatic
Approaches
À
A handle on [O]: A variety of C H
the molecule, as well as the molecular
oxidation methods were explored on the
betulin skeleton to improve the solubility
of this bioactive, yet poorly water-soluble,
natural product. The innate reactivity of
handles present on the core, allowed
oxidations at different positions. Solubil-
ity enhancement was observed for many
of the synthesized compounds.
Angew. Chem. Int. Ed. 2014, 53, 1 – 7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7
These are not the final page numbers!