Synthesis of oxa-B-ring analogs of colchicine through Rh-catalyzed intramolecular [5+2] cycloaddition

Article abstract of DOI:10.1002/ejoc.201200677

A synthetic approach towards (5R)-5-methyl-6-oxa-desacetamido colchicine as a conformationally defined non-natural colchicine analog with a modified B-ring was undertaken. The synthetic strategy was based on a Rh-catalyzed cascade reaction involving a [5+

Full text of DOI:10.1002/ejoc.201200677

Job/Unit: O20677
/KAP1
Date: 09-07-12 11:21:13
Pages: 8
FULL PAPER
DOI: 10.1002/ejoc.201200677
Synthesis of Oxa-B-Ring Analogs of Colchicine through Rh-Catalyzed
Intramolecular [5+2] Cycloaddition
Andreas Ole Termath,[a][‡] Stefanie Ritter,[a][‡] Marcel König,[b] Darius Paul Kranz,[a]
Jörg-Martin Neudörfl,[a] Aram Prokop,*[b] and Hans-Günther Schmalz*[a]
Dedicated to Professor E. Winterfeldt on the occasion of his 80th birthday
Keywords: Natural products / Homogeneous catalysis / Cycloaddition / Ylides / Cascade reactions / Antitumor agents
A synthetic approach towards (5R)-5-methyl-6-oxa-desacet-
amido colchicine as a conformationally defined non-natural
colchicine analog with a modified B-ring was undertaken.
The synthetic strategy was based on a Rh-catalyzed cascade
reaction involving a [5+2] cycloaddition of a carbonyl ylide
intermediate as a key step, in which both seven-membered
rings of the polycyclic framework are formed in a single oper-
ation. Starting from 2-iodo-3,4,5-trimethoxy-acetophenone,
an upper side-chain was constructed through enantioselec-
tive CBS reduction (up to 75%ee) and propargylation, while
a lower succinoyl side-chain was attached either through
iodine–magnesium–copper exchange and subsequent reac-
tion with methyl 4-chloro-4-oxobutanoate, or by Pd-cata-
lyzed Stille cross-coupling with 2-tributylstannyl-5-meth-
oxyfuran followed by hydrolytic furan-opening. Treatment of
an α-diazoketone intermediate with Rh2(OAc)4 (3 mol-%)
initiated the diastereoselective key cyclization cascade
(Ն97:3dr). Treatment of the cycloadduct 3 with Et2AlCl af-
forded an interesting 11,12-dihydrocolchicine analog 24,
which, however, could not be oxidized to the corresponding
tropolone. Structural assignments were confirmed by X-ray
crystallography. While compounds 3 and 24 did not exhibit
noteworthy cytotoxic activity by themselves, they were found
to strongly enhance the cytostatic (apoptosis-inducing) ac-
tivity of doxorubicin against resistant Nalm-6 cells (i.e., in a
synergy effect).
While the trimethoxyarene (ring A) and the tropolone
methyl ether (or a related hydroxyarene, ring C) are sub-
structures essential to ensuring strong binding to the colchi-
cine binding site of tubulin,[7] the impact of ring B is not
yet fully understood. However, it has been demonstrated
that a helical twist around the chiral biaryl axis and its ab-
solute configuration (aR in the natural product) is a crucial
parameter.[8] Modifications of ring B considerably influence
the conformational preorganization (and dynamics) of col-
chicine analogs with respect to the atropisomeric equilib-
rium.[9]
Introduction
Even though significant progress has been made in the
treatment of various types of cancer, the development of
novel and effective anticancer agents, especially those able
to overcome resistance, remains a relevant scientific task.
One of the most prominent targets addressed in cancer che-
motherapy is the protein tubulin, which plays a crucial role
in the cell-division process.[1] Colchicine (1), first isolated in
1820,[2] is the oldest natural product known to bind to tubu-
lin (Figure 1).[3] Although 1 is able to effectively inhibit tub-
ulin polymerization (i.e., microtubule formation) which im-
mediately leads to mitosis arrest,[4] the molecule’s potential
for cancer therapy is limited due to its high general toxic-
ity.[5] Over the last decades, various synthetic and semisyn-
thetic analogs of 1 have been prepared and evaluated biolo-
gically in order to deduce structure–activity relationships.[6]
[a] Department of Chemistry, Universität zu Köln
Greinstr. 4, 50939 Köln, Germany
Fax: +49-221-470-3064
Figure 1. (7S,aR)-Colchicine (1) and its 5-methyl-6-oxadesacet-
amido analog 2.
E-mail: schmalz@uni-koeln.de
[b] Department of Pediatric Hematology/Oncology,
Children’s Hospital Köln
Amsterdamerstr. 59, 50735 Köln, Germany
[‡] These authors contributed equally to this work
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200677.
Therefore, the investigation of new, conformationally de-
fined B-ring analogs of 1 could lead to an improved under-
standing of subtle conformational effects and possibly to a
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: O20677
/KAP1
Date: 09-07-12 11:21:13
Pages: 8
A. Prokop, H.-G. Schmalz et al.
FULL PAPER
future, more “rational” design of molecules with improved Synthetic Strategy
pharmacological properties. Remarkably, virtually no prac-
Our retrosynthetic analysis, shown in Scheme 1, re-
sembles the strategy developed in the course of the total
synthesis of colchicine (1).[11] In the key step, dioxatetra-
cyclic pretarget compound 3 should be formed in a rhodi-
um(II)-catalyzed cascade process involving a formal [5+2]
cycloaddition of carbonyl ylide intermediate 4, generated in
situ from diazoketone precursor 5. This compound in turn
should be prepared (via 6) from 3,4,5-trimethoxyaceto-
phenone derivative 7 through reduction/O-propargylation
(to set up the “upper” chain) and introduction of the
“lower” succinoyl substituent in a formal acylation process.
A successful realization of this plan in the synthesis of 2
tical entries to colchicine B-ring analogs (preserving the tro-
polone ether moiety) have been reported, except for a very
few semisynthetic protocols.[10] In this paper, we report a
synthesis of a colchicine B-ring analog exploiting a strategy
we had previously developed in the course of our total syn-
thesis of 1.[11]
Results and Discussion
Design
Our design of a promising new colchicine B-ring analog
was guided by the following considerations: (1) Ring A (tri- would also prove the general applicability of our strategy in
methoxyarene) and ring C (tropolone ether) should remain the synthesis of novel ring-B-modified colchicine analogs
unchanged. (2) The size of ring B (seven-membered) should (exploiting an intramolecular 1,3-dipolar cycloaddition to
also be maintained. (3) The chemical characteristics of simultaneously build up rings B and C). It should be men-
ring B should be modified by introduction of an oxygen tioned that Padwa, who pioneered such Rh-triggered carb-
heteroatom. (4) A substituent at C5 (or C7) should induce a onyl ylide formation/[3+2] cycloaddition cascades,[13] ex-
defined conformational preorganization. (5) The compound plicitly emphasized that the method is not suitable for the
should realistically be accessible in a limited number of construction of seven-membered rings.[13c] However, as we
steps (Յ15). As a target structure potentially fulfilling these have previously shown, this limitation can be overcome in
requirements, we selected the 6-oxadesacetamidocolchicine certain cases by performing the reaction at elevated tem-
derivative 2 with a methyl substituent at C5. Conforma- peratures under high dilution conditions.[11,14]
tional analysis of 2 using computational methods[12] re-
vealed an energy difference of 7.4 kJ/mol between the two
possible atropdiastereomers 2a and 2b. Thus, the 5R-
enantiomer of 2 should preferentially adopt a conformation
Probing the Strategy in the Racemic Series
To investigate the feasibility of the key Rh-catalyzed cy-
cloaddition step in context, we first developed a straight-
forward six-step synthesis of cyclization precursor rac-5,
omitting configuration control at the stereogenic center
(Scheme 2).
(2a) with a colchicine-like (aR) configuration of the biaryl
axis (Figure 2).
First, commercially available 3,4,5-trimethoxyaceto-
phenone (8) was monoiodinated using I2 in combination
with silver trifluoroacetate.[15] Subsequent Pd-catalyzed
cross-coupling of resulting iodoarene 9 with 2-methoxyfur-
ylstannane 10 (as a masked succinoyl equivalent) proceeded
smoothly under standard Stille conditions.[16] Reduction of
ketone 11 with NaBH4 in an ethanol/1,4-dioxane solvent
Figure 2. Prediction of the conformational equilibrium of the
atropdiastereomers 2a and 2b based on calculated relative energies. mixture afforded alcohol rac-12, which was readily con-
Scheme 1. Retrosynthetic analysis of 2.
2
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20677
/KAP1
Date: 09-07-12 11:21:13
Pages: 8
Analogs of Colchicine through Rh-Catalyzed Intramolecular Cycloaddition
Scheme 2. Synthesis of the cyclization precursor in the racemic series. Reagents and conditions: (a) I2 (1 equiv.) Ag(O2CCF3), CHCl3,
r.t., 6 h, 75%. (b) [Pd(PPh3)4] (3 mol-%), toluene, 110 °C, 16 h, 86%. (c) NaBH4, EtOH/dioxane, r.t., 4 h, 87%. (d) NaH, DMF, r.t., 1.5 h;
then propargyl bromide, r.t., 4 h, 63%. (e) 2 n HCl, THF, r.t., 30 min, 82%; then 2 n LiOH, THF, r.t., 1.5 h, 72%. (f) ClCO2iBu, NEt3,
THF/Et2O (1:1), –20 °C, 1 h; then CH2N2/Et2O, –5 °C, 12 h, 65%. DMF = N,N-dimethylformamide.
verted into propargyl ether rac-13 under SN2 conditions
As shown in Scheme 3, we assume that the transforma-
(NaH, propargyl bromide in DMF). The succinoyl side tion of 5 into 3 is initiated by formation of electrophilic
chain was then released by acidic hydrolysis of the meth- Rh–carbene complex 14, which is subsequently attacked by
oxyfuryl moiety of rac-13 and saponification of the initially the oxygen atom of the benzylic ketone. Resulting 1,3-di-
formed methyl ester. Resulting carboxylic acid rac-6 was fi- pole 4 then either undergoes the desired intramolecular
nally converted into diazoketone rac-5 through activation [3+2] cycloaddition with the tethered alkyne, or is iso-
with isobutyl chlorocarbonate (to form a mixed anhydride) merized to enol–lactone 15.[17] As in the case of our colchi-
followed by reaction with diazomethane.
cine synthesis,[11] the cycloaddition step (5 to 3) proceeded
With cyclization precursor rac-5 prepared, we went on to with virtually complete diastereoselectivity. The relative
test the intramolecular RhII-triggered key step. Much to our configuration of 3 was unambiguously confirmed by X-ray
satisfaction, the desired transformation could indeed be crystallography (Figure 3). By calculating the energy differ-
achieved (Scheme 3). After slowly adding a solution of rac-
5 to a vigorously stirred suspension of Rh2(OAc)4 (3 mol-
%) in refluxing toluene, cycloaddition product rac-3 was
isolated as a pure diastereomer (Ն97:3dr according to
NMR analysis) along with enol lactone rac-15 as a byprod-
uct. Initial experiments conducted in the racemic series (on
a 0.1 mmol scale) typically afforded a 2.5:1 mixture of the
two products rac-3 and rac-15 in a combined yield of ca.
80%. Later, when the reaction was conducted in the nonra-
cemic series on a 2 mmol scale, key intermediate 3 was iso-
lated in 72% yield after purification.
Figure 3. Structures of cycloadduct 3 (left) and chlorinated inter-
mediate 23 (right) in the crystalline state.
Scheme 3. Mechanism of the Rh-catalyzed key transformation.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3

Job/Unit: O20677
/KAP1
Date: 09-07-12 11:21:13
Pages: 8
A. Prokop, H.-G. Schmalz et al.
FULL PAPER
ence between the two possible diastereomeric cycloadducts with CuCN·2LiCl to generate a cuprate.[21] Dropwise ad-
(ΔE = 15.5 kJ/mol) it was ascertained that the observed cy- dition of methyl 4-chloro-4-oxobutanoate and subsequent
cloadduct 3, formed under kinetically controlled conditions, ester saponification finally afforded acid 6 after chromato-
is also the thermodynamically more stable diastereomer.[12] graphic purification. The deiodinated starting material was
isolated as a byproduct. The conversion of 6 into diazo-
ketone 5, and further into the key intermediate 3, was then
Nonracemic Series
smoothly accomplished using the protocols developed in
the racemic series (Scheme 4). The enantiomeric purity did
not change significantly in the course of the sequence, as
confirmed by chiral HPLC analysis of 3 (73%ee).
To establish the R-configured stereocenter at C5 enantio-
selectively, we decided to employ the oxazaborolidine-cata-
lyzed ketone hydroboration (CBS reduction).[18] Under
standard CBS conditions (using oxazaborolidine catalyst
16), 3,4,5-trimethoxyacetophenone (8) was reduced to the
corresponding alcohol in quantitative yield and with high
enantioselectivity (80%ee; unoptimized). Unfortunately,
Endgame
As we had already experienced in the course of the total
however, the subsequent iodination step (to give 17) could synthesis of colchicine, the final conversion of cycloadduct
not be accomplished in a satisfactory manner in the pres- 3 into a tropolone target structure represents an astonish-
ence of the benzylic hydroxy group. On the other hand, re- ingly tough challenge.[11b] Unfortunately, the sequence de-
lated acetophenones with an additional ortho-substituent veloped in the colchicine series (which involved an initial
gave substantially lower enantioselectivities in the CBS re- reduction of the ketone functionality, followed by opening
duction. For instance, reduction of ketone 11 afforded the of the oxygen bridge with TMSOTf and subsequent Swern
product with only 58%ee. However, under optimized reac- oxidation) could not be applied in the 6-oxa series. While
tion conditions (20 mol-% of 16, THF, 25 °C) we succeeded the reduction of 3 with L-Selectride afforded desired
in reducing aryl iodide 9 to alcohol 17 with at least 75%ee alcohol 19 diastereoselectively in 95% yield, all our
(Scheme 4). The absolute configuration of 17 was deter- attempts to achieve the subsequent ring opening to generate
mined by X-ray crystal structure analysis of its 3,5-dini- diol 20 failed (Scheme 5).
trobenzoate ester (see Supporting Information) and is in
Instead, the reaction of 19 with TMSOTf afforded epox-
ide 21, which we considered to be a dead end with respect
agreement with Corey’s transition-state model.[19]
The transformation of 17 into nonracemic cyclization to the synthetic goal. Presumably, the allylic carbocation
precursor 5 was then carried out, using a different and even (related to 22, but with a TMS group instead of the AlEt2
more efficient approach to build up the succinoyl side chain group; see Scheme 6), initially formed by TMSOTf-pro-
(Scheme 4). For this purpose, 17 was first converted into its moted ring opening of 19, has a low tendency to lose a
propargyl ether (NaH, propargyl bromide, DMF), and the proton. The expected product 20 could not be detected in
terminal alkyne was protected by trimethylsilylation (LDA, the product mixture, from which the epoxide 21 was iso-
TMSCl). To prepare the introduction of the succinoyl side lated as the only major component in 42% yield
chain, 18 was subjected to a Knochel iodine–magnesium (Scheme 5). Interestingly, treatment of 3 with Et2AlCl re-
exchange (iPrMgCl·LiCl)[20] to give the corresponding sulted in the rapid formation (10 min, 0 °C) of chlorohydrin
Grignard species, which was directly further transmetalated 23, possibly resulting from chloride transfer within the first-
Scheme 4. Enantioselective synthesis of the key intermediate 3. Reagents and conditions: (a) 16 (20 mol-%), BH3·SMe2, THF, r.t., 3 h;
then MeOH, quant., 75%ee. (b) NaH, propargyl bromide, DMF, 0 °C Ǟ r.t., 12 h, 93%. (c) LDA, THF, –78 °C, 1 h; then TMSCl, –78 °C
Ǟ r.t., 3 h, 77%. (d) iPrMgCl·LiCl, THF, –30 °C Ǟ –10 °C, 30 min.; then CuCN·2LiCl, –30 °C Ǟ r.t., 2.5 h; then methyl 4-chloro-4-
oxobutanoate, –30 °C Ǟ r.t., 2 h. (e) 2 n LiOH, THF, 0 °C Ǟ r.t., 20 h, 58% over 2 steps. (f) ClCO2iBu, NEt3, THF/Et2O (1:1), –20 °C,
1 h; then CH2N2/Et2O, –5 °C, 12 h, 73%. (g) [Rh2(OAc)4] (3 mol-%), toluene, 130 °C, 5 h, 72%.
4
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20677
/KAP1
Date: 09-07-12 11:21:13
Pages: 8
Analogs of Colchicine through Rh-Catalyzed Intramolecular Cycloaddition
pre-B leukemia cells). Neither compound exhibited any
noteworthy unspecific cytotoxicity (necrosis), as determined
by a lactate dehydrogenase (LDH) release assay after 1 h.
While compound 24 inhibited proliferation of BJAB cells at
higher concentrations (50% inhibition at 1–10 μm, deter-
mined after 24 h by cell counting) associated with a weak
Scheme 5. Formation of epoxide 21 in the attempt to transform 3
into tropolone precursor 20. Reagents and conditions: (a) L-Se-
lectride, THF, –78 °C, 3 h, 95%, Ͼ96:4dr. (b) TMSOTf, Et3N,
CH2Cl2, –60 °C Ǟ –10 °C, 1.5 h; then K2CO3, MeOH, –60 °C Ǟ
r.t., 1 h, 42%.
formed ring-opened intermediate 22 (Scheme 6). The struc-
ture of 23, a rather unstable compound, was proved by X-
ray crystallography (Figure 3).[22] When the reaction mix-
ture was stirred at 0 °C for a longer period of time (2 h),
dihydrotropolone 24 (formed from 23 by elimination of
HCl) became the only main product (Scheme 6).
Remarkably, all our attempts to convert (oxidize) 23 or
24 to the target tropolone 25 (corresponding to 2) failed. In
addition to IBX-based methods[23] we tested DDQ, Swern-
type and chromium(VI)-based oxidation protocols. How-
ever, in all cases, either no conversion of the starting mate-
rial, or, under forcing conditions, the formation of complex
mixtures (decomposition), was observed.
Biological Investigations
Figure 4. Synergistic effects of apoptosis-induction in Nalm-6 cells
of 24 (top) and 3 (bottom) in combination with doxorubicin. Cells
were cultivated with different concentrations of either 3 (or 24,
respectively) and doxorubicin only or with mixtures of both com-
pounds. Depicted are mean values and standard deviations from
three measurements. The yellow area shows the synergistic effects
of 3, where the substance was able to sensitize the cells for doxorub-
While the originally selected target structure, that is, tro-
polone ether 2, could not be prepared, we nevertheless de-
cided to at least investigate the cytotoxic (antitumoral)
properties of dihydrotropolone 24 and its isomeric oxa-
bridged precursor 3 using BJAB (Burkitt-like lymphoma)
and doxorubicin-resistant Nalm-6 cells (human peripheral icin.
Scheme 6. Lewis acid promoted ring-opening of 3. Reagents and conditions: (a) Et2AlCl (5 equiv.), CH2Cl2, –78 Ǟ 0 °C, 2 h, 57% of 24.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5

Job/Unit: O20677
/KAP1
Date: 09-07-12 11:21:13
Pages: 8
A. Prokop, H.-G. Schmalz et al.
FULL PAPER
(5S)-6-Oxo-5-methyl-9-hydroxy-1,2,3-trimethoxy-5,7,11,12-tetra-
hydrobenzo[a]-heptalene-10-one (24): Et2AlCl (0.430 mL, 1.0 m in
apoptosis induction (18% at 50 μm after 72 h), compound
3 did not show any significant activity. However, an inter-
esting observation was made when the compounds were ap-
plied in combination with doxorubicin to Nalm-6 cells (Fig-
ure 4).
Coapplication of different cytostatics is a common way
to increase the effectiveness of cancer therapies, because
apoptosis-inducing effects of the combined cytostatics are
sometimes higher than the sum of the individual cytotoxici-
hexane, 0.43 mmol) was added to
a solution of 3 (30 mg,
0.087 mmol) in dry dichloromethane (3 mL) at –78 °C. The solu-
tion was stirred at 0 °C for 2 h, then re-cooled to –78 °C, and care-
fully quenched by dropwise addition of MeOH. Then, saturated
aqueous NH4Cl (5 mL) was added, and the mixture was extracted
with dichloromethane (3ϫ 5 mL). The combined organic layers
were dried with Na2SO4, and the solvent was removed under re-
duced pressure. The residue was purified by column chromatog-
ties (i.e., a synergy effect). The synergistic potential of new raphy (EtOAc/cyclohexane, 1:4) to afford 24 (17 mg, 57%) as a
colorless oil.
compounds can be experimentally assessed by determining
the apoptosis induction (through FACScan flow cytometry)
after cultivation of malignant cells treated with the sub-
stances alone and in combination with existing chemothera-
peutic drugs (such as doxorubicin). Indeed, it was found
that both compounds exhibited unusually pronounced syn-
ergistic effects with doxorubicin (Figure 4). For instance,
the combination of 3 (10 μm) and doxorubicin (1 nM) re-
sulted in a synergistic effect of more than 400%.
Supporting Information (see footnote on the first page of this arti-
cle): Detailed experimental and spectroscopic data.[26]
Acknowledgments
This work was supported by the Volkswagen Stiftung. Donations
of chemicals from Chemetall and Evonik are gratefully acknowl-
edged. The authors also would like to thank Dr. D. Blunk and Dr.
M. Drayß, Department of Chemistry, for assistance with the DFT
calculations, and Dr. M. Pietch and J. E. Stein, Department of
Pharmacology, for testing compounds 3 and 24 in a GST inhibition
assay.
Conclusions
In summary, we have exploited a [5+2] cycloaddition
strategy, based on a Rh-triggered cyclization cascade,[11] as
a key transformation for the synthesis of new 6-oxacolchic-
ine-related compounds.[24] While the final elaboration of the
tropolone ether substructure turned out to be a highly diffi-
cult task (which could not be achieved in the course of this
study), the cycloaddition product 3 was efficiently con-
verted into the interesting 11,12-dihydro-6-oxacolchicine
analog 24 in a single (Et2AlCl-mediated) step. The synthesis
of 24 was achieved in only 9 steps in 15% overall yield start-
ing from 3,4,5-trimethoxyacetophenone (8). From a chemi-
cal point of view, the unexpected difficulties in the final
functionalization reflect a surprisingly subtle dependence of
the reactivity of ring C on structural variations in ring B.
An important outcome of the study is the finding that both
cycloadduct 3 and (isomeric) dihydrotropolone 24 (while
not showing any colchicine-related biological activity)
strongly amplify the antitumoral activity of doxorubicin
against Nalm-6 cells. The molecular basis of the observed
synergistic effect may deserve further investigation in the
course of the development of future chemotherapeutic stra-
tegies.[25]
[1] E. Nogales, S. G. Wolf, K. H. Downing, Nature 1998, 391, 199–
203.
[2] P. J. Pelletier, J. B. Caventou, Ann. Chim. Phys. 1820, 14, 69–
83.
[3] a) O. Boyé, A. Brossi in The Alkaloids (Eds.: A. Brossi, G. A.
Cordell), Academic Press, San Diego, 1992, vol. 41, pp. 125–
176; b) C. LeHello in The Alkaloids (Ed.: G. A. Cordell), Aca-
demic Press, San Diego, 2000, vol. 53, ch. 5; for a comprehen-
sive review on colchicine total syntheses, see: c) T. Graening,
H.-G. Schmalz, Angew. Chem. 2004, 116, 3292–3318; Angew.
Chem. Int. Ed. 2004, 43, 3230–3256.
[4] a) R. C. Weisenberg, G. G. Borisy, E. W. Taylor, Biochemistry
1968, 7, 4466–4479; b) S. Bane Hastie, Pharmacol. Ther. 1991,
51, 377–401; c) A. Brossi, H. J. Yeh, M. Chrzanowska, J. Wolff,
E. Hamel, C. M. Lin, F. Quin, M. Sulfness, J. Silverton, Med.
Res. Rev. 1988, 8, 77–94.
[5] a) W. E. Dixon, W. Malden, J. Physiol. (London) 1908, 37, 50–
76; b) R. M. Naidus, R. Rodvein, H. Mielke, Arch. Int. Med.
1977, 137, 394–396.
[6] a) Q. Shi, P. Verdier-Pinard, A. Brossi, E. Hamel, A. T.
McPhail, K. H. Lee, J. Med. Chem. 1997, 40, 961–966; b) B.
Bhattacharyya, D. Panda, S. Gupta, M. Banerjee, Med. Res.
Rev. 2008, 28, 155–183; c) N. Nicolaus, J. Zapke, P. Riesterer,
J.-M. Neudörfl, A. Prokop, H. Oschkinat, H.-G. Schmalz,
ChemMedChem 2010, 5, 661–665.
[7] a) E. Hamel, H. H. Ho, G. J. Kang, C. M. Lin, Biochem. Phar-
macol. 1988, 37, 2445–2449; b) B. Bhattacharyya, J. Wolff,
Proc. Natl. Acad. Sci. USA 1974, 71, 1134–1140; c) M. Rosner,
H. G. Capraro, A. E. Jacobsen, L. Atwell, A. Brossi, M. A.
Iorio, T. H. Williams, R. H. Sik, C. F. Chignell, J. Med. Chem.
1981, 24, 257–261; d) G. G. Choudhury, A. Banerjee, B. Bhat-
tacharyya, B. B. Biswas, FEBS Lett. 1983, 161, 55–59; e) B.
Bhattacharyya, J. Wolff, J. Biol. Chem. 1984, 259, 11836–11843;
f) S. B. Hastie, R. C. Williams Jr., D. Puett, T. L. McDonald,
J. Biol. Chem. 1989, 264, 6682–6688; for a publication dealing
with colchicide, see: g) C. Dumortier, Q. Yan, S. Bane, Y. En-
gelborghs, Biochem. J. 1997, 327, 685–688; for publications
dealing with thiocolchicine, see: h) L. Velluz, G. Müller, Bull.
Soc. Chim. Fr. 1954, 755–757; i) L. Velluz, G. Müller, Bull. Soc.
Chim. Fr. 1954, 1072–1075; j) S. B. Hastie, Biochemistry 1989,
28, 7753–7760.
Experimental Section
(5R,9S,12aS)-6-Oxo-5-methyl-1,2,3-trimethoxy-9,12a-epoxy-
5,7,9,11,12,12a-hexahydrobenzo[a]heptalene-10-one (3): A solution
of 5 (0.70 g, 1.87 mmol) in toluene (50 mL) was slowly added to an
intensely stirred suspension of [Rh2(OAc)4] (24 mg, 56 μmol, 3 mol-
%) in refluxing toluene (150 mL; oil-bath temperature: 135 °C) over
a period of 4 h by means of a syringe pump. The reaction mixture
was heated at reflux for an additional 30 min, and subsequently
cooled to room temp. The solvent was removed under reduced pres-
sure, and the residue was purified by column chromatography
(EtOAc/cyclohexane, 1:4) to afford 3 (464 mg, 72%) as a colorless
solid.
6
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20677
/KAP1
Date: 09-07-12 11:21:13
Pages: 8
Analogs of Colchicine through Rh-Catalyzed Intramolecular Cycloaddition
[8] U. Berg, J. Deinum, P. Lincoln, J. Kvassman, Bioorg. Chem.
1991, 19, 53–65.
C. P. Chen, V. K. Singh, J. Am. Chem. Soc. 1987, 109, 7925–
7926.
[9] a) G. Chakrabarti, S. Sengupta, B. Bhattacharyya, J. Biol.
Chem. 1996, 271, 2897–2901; b) A. R. Chaudhuri, P. Seethara-
malu, P. M. Schwarz, F. H. Hausheer, R. F. Ludueña, J. Mol.
Biol. 2000, 303, 679–692.
[10] a) B. Danieli, G. Lesma, D. Passarella, D. Prosperi, A. Sacch-
etti, A. Silvani, R. Destro, E. May, E. Bombardelli, Helv. Chim.
Acta 2003, 86, 2082; b) B. Bhattach, J. Wolff, Proc. Natl. Acad.
Sci. USA 1974, 71, 2627.
[11] a) T. Graening, W. Friedrichsen, J. Lex, H.-G. Schmalz, Angew.
Chem. 2002, 114, 1594–1597; Angew. Chem. Int. Ed. 2002, 41,
1524–1526; b) T. Graening, V. Bette, J. Neudörfl, J. Lex, H.-G.
Schmalz, Org. Lett. 2005, 7, 4317–4320.
[19]
[20]
[21]
[22]
E. J. Corey, C. J. Helal, Angew. Chem. 1998, 110, 2092–2118;
Angew. Chem. Int. Ed. 1998, 37, 1986–2012.
A. Krasovskiy, P. Knochel, Angew. Chem. 2004, 116, 3396–
3399; Angew. Chem. Int. Ed. 2004, 43, 3333–3336.
O. Baron, P. Knochel, Angew. Chem. 2005, 117, 3193–3195;
Angew. Chem. Int. Ed. 2005, 44, 3133–3135.
The X-ray structure analysis was performed with a crystal of
ent-23 prepared in the course of initial studies in which the
enantiomeric CBS catalyst had been employed (Scheme 5).
[23] K. C. Nicolaou, T. Montagnon, P. S. Baran, Y.-L. Zhong, J.
Am. Chem. Soc. 2002, 124, 2245–2258.
[12] DFT calculations were performed using the B3LYP-functional
with a 6–311G(d) basis set as implemented in the Gaussian
03 Program suite. A frequency analysis was performed for all
optimized structures to prove the minima.
[13] a) A. Padwa, G. E. Fryxell, L. Zhi, J. Am. Chem. Soc. 1990,
112, 3100–3109; b) A. Padwa, R. L. Chinn, S. F. Hornbuckle,
Z. J. Zhang, J. Org. Chem. 1991, 56, 3271–3278; c) A. Padwa,
M. D. Weingarten, Chem. Rev. 1996, 96, 223–269; d) A. Padwa,
Helv. Chim. Acta 2005, 88, 1357–1374.
[14] For a related temperature effect on a [3+2] cyclization, see: A.
Gromov, V. Enev, J. Mulzer, Tetrahedron Lett. 2009, 50, 5283–
5284.
[15] D. E. Janssen, C. V. Wilson, Org. Synth. 1956, 36, 46.
[16] J. K. Stille, Angew. Chem. 1986, 98, 504–519; Angew. Chem. Int.
Ed. Engl. 1986, 25, 508–524.
[17] Byproducts of this type have also been observed in the total
synthesis of colchicine (ref.[11b]) as well as in other cases
(ref.[13a,14]). Their formation can be understood in terms of a
[1,4] sigmatropic H-shift: A. C. Lottes, J. A. Landgrebe, K.
Larsen, Tetrahedron Lett. 1989, 30, 4089–4092.
[18] a) E. J. Corey, R. K. Bakshi, S. Shibata, J. Am. Chem. Soc.
1987, 109, 5551–5553; b) E. J. Corey, R. K. Bakshi, S. Shibata,
[24] a) For some recent publications on 6-oxaallocolchicine deriva-
tives, see: N. Nicolaus, S. Strauss, J.-M. Neudörfl, A. Prokop,
H.-G. Schmalz, Org. Lett. 2009, 11, 341–344; b) L. Keller, P.
Dauban, J. Med. Chem. 2008, 51, 3414–3421.
[25] So far, we only could exclude that the observed activity of 3
and 24 results from the inhibition of glutathione S-transferase
from human placenta (hpGST). The assays were performed ac-
cording to: a) W. H. Habig, M. J. Pabst, W. B. Jakoby, J. Biol.
Chem. 1974, 249, 7130–7139; b) W. H. Ang, L. J. Parker, A.
De Luca, L. Juillerat-Jeanneret, C. J. Morton, M. Lo Bello,
M. W. Parker, P. J. Dyson, Angew. Chem. 2009, 121, 3912–3915;
Angew. Chem. Int. Ed. 2009, 48, 3854–3857; for a review on
GST, see: c) J. D. Hayes, J. U. Flanagan, I. R. Jowsey, Annu.
Rev. Pharmacol. Toxicol. 2005, 45, 51–88.
[26] CCDC-881272 (for the 3,5-dinitrobenzoate of 17), -881273 (for
3), and -881274 (for ent-23) contain the supplementary crystal-
lographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
Received: May 19, 2012
Published Online: ᭿
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7

Job/Unit: O20677
/KAP1
Date: 09-07-12 11:21:13
Pages: 8
A. Prokop, H.-G. Schmalz et al.
FULL PAPER
Natural Product Analogs
A. O. Termath, S. Ritter, M. König,
D. P. Kranz, J.-M. Neudörfl, A. Prokop,*
H.-G. Schmalz* ................................ 1–8
Synthesis of Oxa-B-Ring Analogs of Col-
chicine through Rh-Catalyzed Intramol-
ecular [5+2] Cycloaddition
Exploiting two successive metal-mediated
transformations, the diazoketone A is ef-
ficiently converted via B into the dihydro-
tropolone C. While B and C did not exhibit
cytotoxic properties, surprisingly, they were
found to sensitize resistant tumor cells
(Nalm 6) against the clinically used an-
ticancer drug doxorubicin.
Keywords: Natural products
/ Homo-
geneous catalysis / Cycloaddition / Ylides /
Cascade reactions / Antitumor agents
8
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0