A concise palladium-catalyzed carboamination route to (±)- tylophorine

Article abstract of DOI:10.1021/jo902114u

(Chemical Equation Presented) Atotal synthesis of the racemic natural product tylophorine [(±)-1] has been demonstrated using the palladium-catalyzed carboamination method developed byWolfe and co-workers. In this case, an electron-rich aryl bromide 18 wa

Full text of DOI:10.1021/jo902114u

pubs.acs.org/joc
approximately 3- to 4-fold increase in cytotoxicity compared
A Concise Palladium-Catalyzed Carboamination
Route to (()-Tylophorine
to its natural (R)-(-)-isomer (avg. GI50 13 vs 43 nM,
respectively).2b
In the past decade, a number of concise synthetic prepara-
tions of racemic tylophorine have appeared in the literature,3
and a few approaches to the discrete R- and S-enantiomers of
1 have also been published.4 We nevertheless sought to
contribute our own novel approach, based on the palla-
dium-catalyzed carboamination process recently established
by Wolfe and co-workers.5 As both absolute configurations
of tylophorine still serve as viable starting points for struc-
tural analogue modifications for drug discovery investiga-
tions,2 we decided to demonstrate the feasibility for a natural
product synthesis of racemic tylophorine using the Wolfe
carboamination strategy. This communication summarizes
our results toward this goal.
Lana M. Rossiter, Meagan L. Slater, Rachel E. Giessert,
Samuel A. Sakwa, and R. Jason Herr*
Medicinal Chemistry Department, Albany Molecular
Research, Inc. (AMRI), P.O. Box 15098, 26 Corporate Circle,
Albany, New York 12212-5098
rjason.herr@amriglobal.com
Received October 1, 2009
A palladium-catalyzed carboamination reaction of γ-amino-
alkenes 2 with aryl bromides has recently been developed by the
Wolfe group to define a general synthesis of substituted pyrro-
lidines 3 (n= 1), as well as larger heterocycles (Scheme 1).5
This process relies on the ability of amines to form a
metallo-amide species 4 with a reactive arylpalladium sub-
strate to direct the metal toward a tethered olefinic moiety.
The palladium intermediate may then undergo an insertion
reaction into the olefin, generating new organometallic
intermediate 5 that contains the newly formed heterocycle.
Reductive elimination of the metal from the species 5 may
then provide the amine heterocycle 3 and regenerate the
active palladium species to perpetuate the catalytic cycle.
While the earliest published method was limited to N-aryl
nucleophiles, more recent communications have demon-
strated the use of carbamate substrates 2 (e.g., PG = Cbz,
Boc) to provide protected products 4 that can be easily
converted into the free amine under mild conditions.6,7
To test the feasibility of the method toward an approach to
racemic tylophorine, we first explored the carboamination
reaction between the two commercially available coupling
precursors 9-bromophenanthrene (6a, X = Br) and tert-
butyl pent-4-enylcarbamate (7), as shown in Scheme 2.
A total synthesis of the racemic natural product tylophorine
[(()-1] has been demonstrated using the palladium-cata-
lyzed carboamination method developed by Wolfe and co-
workers. In this case, an electron-rich aryl bromide 18 was
prepared in four steps and subjected to palladium-catalyzed
Wolfe carboamination conditions with olefinic carbamate 7
to provide the racemic 2-(arylmethyl)pyrrolidine (()-19 in
good yield and was further elaborated to racemic tylopho-
rine. This application of the Wolfe carboamination protocol
as a key step to construct a natural product provides further
evidence of the utility of the method.
Tylophorine [(R)-(-)-1, Figure 1] is a phenanthroindoli-
zidine alkaloid that was first isolated in 1935 from the
perennial climbing plant Tylophora indica1 and has recently
re-emerged as an alkaloid target family of interest, due
largely to its cytotoxic activity by a novel mechanism of
action.2 Its synthetically derived (S)-(þ)-antipode has gene-
rated even greater interest due to its superior cell growth
inhibition activity versus its natural isomer. For instance, in
three carcinoma cell lines, (S)-(-)-tylophorine exhibited an
(3) For the most recent examples, see: (a) Yamashita, S.; Kurono, N.;
Senboku, H.; Tokuda, M.; Orito, K. Eur. J. Org. Chem. 2009, 1173.
(b) McIver, A.; Young, D. D.; Deiters, A. Chem. Commun. 2008, 39, 4750.
(c) Wang, K.-L.; Lue, M.-Y.; Wang, Q.-M.; Huang, R.-Q. Tetrahedron 2008,
64, 7504. (d) Chuang, T.-H.; Lee, S.-J.; Yang, C.-W.; Wu, P.-L. Org. Biomol.
Chem. 2006, 4, 860. (e) Li, Z.; Jin, Z.; Huang, R. Synthesis 2001, 16, 2365.
(4) For the most recent examples, see: (a) Zeng, W.; Chemler, S. R. J. Org.
Chem. 2008, 73, 6045. (b) Wang, K.; Wang, Q.; Huang, R. J. Org. Chem. 2007,
€
72, 8416. (c) Furstner, A.; Kennedy, J. W. J. Chem.;Eur. J. 2006, 12, 7398.
(1) (a) Wu, P.-L.; Rao, K. V.; Su, C.-H.; Kuoh, C.-S.; Wu, T.-S.
Heterocycles 2002, 57, 2401. (b) Abe, F.; Iwase, Y.; Yamauchi, T.; Honda,
K.; Hayashi, N. Phytochemistry 1995, 39, 695. (c) Govindachari, T. R.;
Lakshmikantham, M. V.; Nagarajan, K.; Pai, B. R. Chem. Ind. 1957, 1484.
(d) Ratnagiriswaran, A. N.; Venkatachalam, K. Ind. J. Med. Res. 1935,
22, 433.
(d) Jin, Z.; Li, S. P.; Wang, Q. M.; Huang, R. Q. Chin. Chem. Lett. 2004, 15,
1164.
(5) (a) Wolfe, J. P. Synlett 2008, 2913. (b) Wolfe, J. P. Eur. J. Org. Chem.
2007, 571. (c) Ney, J. E.; Hay, M. B.; Yang, Q.; Wolfe, J. P. Adv. Synth. Catal.
2005, 347, 1614.
(6) (a) Bertrand, M. B.; Neukom, J. D.; Wolfe, J. P. J. Org. Chem. 2008,
73, 8851. (b) Bertrand, M. B.; Wolfe, J. P. Org. Lett. 2007, 9, 3073.
(c) Bertrand, M. B.; Leathen, M. L.; Wolfe, J. P. Org. Lett. 2007, 9, 457.
(d) Bertrand, M. B.; Wolfe, J. P. Tetrahedron 2005, 61, 6447.
(7) For related palladium- and copper-catalyzed carboamination work,
see: (a) Chemler, S. R. Org. Biomol. Chem. 2009, 7, 3009. (b) Rosewall, C. F.;
Sibbald, P. A.; Liskin, D. V.; Michael, F. E. J. Am. Chem. Soc. 2009, 131,
(2) (a) Yang, X.; Shi, Q.; Liu, Y.-N.; Zhao, G.; Bastow, K. F.; Lin, J.-C.;
Yang, S.-C.; Yang, P.-C.; Lee, K.-H. J. Med. Chem. 2009, 52, 5262. (b) Gao,
€
W.; Chen, A. P.-C.; Leung, C.-H.; Gullen, E. A.; Furstner, A.; Shi, Q.; Wei,
L.; Lee, K.-H.; Cheng, Y.-C. Bioorg. Med. Chem. Lett. 2008, 18, 704.
(c) Banwell, M. G.; Bezos, A.; Burns, C.; Kruszelnicki, I.; Parish, C. R.;
Su, S.; Sydnes, M. O. Bioorg. Med. Chem. Lett. 2006, 16, 181. (d) Gao, W.;
Lam, W.; Zhong, S.; Kaczmarek, C.; Baker, D. C.; Cheng, Y.-C. Cancer Res.
2004, 64, 678. (e) Strk, D.; Lykkeberg, A. K.; Christensen, J.; Budnik, B. A.;
Abe, F.; Jaroszewski, J. W. J. Nat. Prod. 2002, 65, 1299.
€
~
9488. (c) Hovelmann, C. H.; Streuff, J.; Brelot, L.; Muniz, K. Chem.
Commun. 2008, 2334. (d) Minatti, A.; Muniz, K. Chem. Soc. Rev. 2007, 36,
~
1142. (e) Chemler, S. R.; Fuller, P. H. Chem. Soc. Rev. 2007, 36, 1153.
9554 J. Org. Chem. 2009, 74, 9554–9557
Published on Web 11/17/2009
DOI: 10.1021/jo902114u
r
2009 American Chemical Society

Rossiter et al.
JOCNote
SCHEME 2. Model System Proof of Principle for the Key Step
FIGURE 1. Natural product tylophorine and its unnatural antipode.
SCHEME 1. Mechanism of the Wolfe Carboamination Reaction
SCHEME 3. Preparation of the Aryl Bromide Substrate 18
We were pleased to find that our initial attempt using the
conditions reported by the Wolfe group for carbamate
nucleophiles6 produced our desired 2-benzylpyrrolidine 8
in 57% yield, a result consistent with yields reported in the
literature for this process. In this case, the only other
compounds isolated were small amounts of phenanthrene,
resulting from the protodebromination of 6a, and trace
amounts of the byproduct 10, originating from a Heck
reaction but without cyclization. Interestingly, when we
subjected commercially available 9-iodophenanthrene (6b,
X = I) to the reaction conditions, the two major products
isolated were the requisite pyrrolidine 8 in only 28% yield
and substantial amounts of the protodehalogenated bypro-
duct from the precursor 6b. Isolated in minor amounts were
both the (Z)- and (E)-olefins derived from the Heck reaction
(9% of 9 and 8% of 10) and the N-arylation byproduct 11 in
4% yield.
Formation of the pentacycle 12 from the 2-benzylic pyr-
rolidine carbamate 8 by nitrogen deprotection with conco-
mitant electrophilic ring-closing reaction with formalin was
achieved in 63% yield (eq 1). While this moderate conversion
is typical for a Pictet-Spengler cyclomethylenation reaction
of an unactivated ring system, it nevertheless provided
confirmation for our proof of principle.
Double reduction of 15 to the corresponding bibenzyl 16
under catalytic hydrogenation conditions was problematic at
first, requiring multiple subjections of fresh palladium cata-
lyst to the reaction mixture (Scheme 3). However, after
examining a number of solvent systems and additives, we
found that a mixture of methanol and THF in aqueous HCl9
worked well to effect the full reduction of 15 on a Parr shaker
(40 psi of hydrogen) to 16 in yields ranging from 57 to 94%.
Alternatively, we found that the reduction could be achieved
under a balloon of hydrogen gas (1 atm) when the reduction
of 15 to 16 was run in ethyl acetate with a small amount of
TFA as a cosolvent, although the reaction took a longer time
to complete (typically days rather than overnight), albeit
without multiple catalyst loadings.
We then turned our attention to the preparation of the
requisite carboamination precursor 18, as shown in Scheme 3.
Sonogashira coupling between the commercially available
reagents 4-ethynyl-1,2-dimethoxybenzene (13) and 4-iodo-1,2-
dimethoxybenzene (14), based on conditions for an analogous
system,8 furnished the symmetrical acetylene 15 in good isolated
yields, ranging from 61 to 88%.
We surveyed several conditions for the oxidative cycliza-
tion reaction10 of the electron-rich 1,2-diarylethane 16 to
provide the symmetrical phenanthrene 17, many of which
(9) Bui, V. P.; Hudlicky, T.; Hansen, T. V.; Stenstrom, Y. Tetrahedron
Lett. 2002, 43, 2839.
(8) (a) Foster, E. J.; Babuin, J.; Nguyen, N.; Williams, V. E. Chem.
Commun. 2004, 18, 2052. (b) Odlo, K.; Klaveness, J.; Rongved, P.; Hansen,
T. V. Tetrahedron Lett. 2006, 47, 1101.
(10) (a) Moreno, I.; Tellitu, I.; San Martın, R.; Domınguez, E. Synlett
2001, 7, 1161. (b) Moreno, I.; Tellitu, I.; Domınguez, E.; San Martın, R. Eur.
J. Org. Chem. 2002, 13, 2126.
J. Org. Chem. Vol. 74, No. 24, 2009 9555

Rossiter et al.
JOCNote
TABLE 1. Selected Results of the Wolfe Carboamination Reaction with
Phenanthryl Bromide 18
albeit in only 16% yield, although the starting phenanthryl
bromide 18 was completely consumed (Table 1, entry 1).
Upon closer examination of the reaction byproducts, we also
isolated and identified a 24% yield of a mixture of (E)- and
(Z)-olefinic byproducts 20 resulting from a Heck coupling
reaction between 18 and 7, but without the cyclization
process having taken place, an observation also made by
Wolfe and co-workers in various instances. We also isolated
a significant amount of the phenanthrene 17, resulting from
the protodebromination of 18 in the reaction environment.
However, as we set about optimizing the reaction para-
meters, we were eventually able to increase the yield of
desired (()-19, as described by the representative experi-
ments summarized in Table 1.
We next subjected phenanthryl bromide 18 and olefinic
carbamate 7 to the reaction conditions using DMF as a
solvent at 80 °C (Table 1, entry 2), as we were initially
concerned about the apparent insolubility of aryl bromide
18 in refluxing dioxane. However, no increase in the isolated
yield of (()-19 was observed. Returning to the use of
refluxing dioxane but using sodium tert-butoxide as a base
(entry 3) provided a significant increase in yield of (()-19,
although it was accompanied by the isolation of 18% of
a new byproduct 21 resulting from a palladium-catalyzed
N-arylation process. Using these new conditions with a change
to Pd2dba3 as an active catalyst (entry 4) increased the isolated
of (()-19 even further to 67%, although the formation of
N-arylation byproduct 21 only increased to 21%.
Investigation of other bidentate ligands for palladium,
including BINAP (entry 5) and Xantphos (entry 6), led to
diminished isolated yields of 19, although it was found that
increasing the ratio of catalyst to Xantphos ligand from 1:1
to 1:2 brought the isolated yield back to a synthetically useful
process and virtually eliminated the formation of byproduct
21 (entry 7). Interestingly, the switch back to the initial reac-
tion conditions, but using Xantphos as a ligand (entry 8), did
not provide any significant amounts of the desired product.
Investigations using the entry 7 conditions in other sol-
vents including toluene (entry 9) and DME (entry 10) led to
yields that were roughly equivalent, although the use of
DMF (entry 11) led to a decrease in the amount of 19 isolated
and a significant increase in the protodebromination of 18.
To complete the total synthesis of (()-tylophorine, the
benzylic pyrrolidine (()-19 was first subjected to anhydrous
deprotection conditions to generate the free pyrrolidine (()-
22 as a hydrochloride salt (Scheme 4). This material was
typically not isolated, but instead formalin and ethanol were
added and the resulting mixture was heated at reflux to effect
the Pictet-Spengler cyclomethylenation reaction,3c,4a,12
providing the pentacyclic product (()-1 in near-quantitative
isolated yields for the two-step process. While there is one
example in the literature for the one-step deprotection/
cyclization sequence from 19 (prepared by a different
strategy) to tylophorine,4c we found the protocol to be fairly
low-yielding in our hands.
yield of
19 (%)
entry
catalyst/ligand
base
conditions
1
2
3
4
5
6
7
8
9
1:1 Pd(OAc)2, DPE-phos Cs2CO3 dioxane, 100 °C
1:1 Pd(OAc)2, DPE-phos Cs2CO3 DMF, 80 °C
1:1 Pd(OAc)2, DPE-phos NaOtBu dioxane, 100 °C
1:1 Pd2dba3, DPE-phos
1:1 Pd2dba3, BINAP
1:1 Pd2dba3, Xantphos
1:2 Pd2dba3, Xantphos
1:1 Pd(OAc)2, Xantphos Cs2CO3 dioxane, 100 °C
1:2 Pd2dba3, Xantphos
1:2 Pd2dba3, Xantphos
1:2 Pd2dba3, Xantphos
16
12
42
67
17
49
65
<5
61
56
16
NaOtBu dioxane, 100 °C
NaOtBu dioxane, 100 °C
NaOtBu dioxane, 100 °C
NaOtBu dioxane, 100 °C
NaOtBu toluene, 100 °C
NaOtBu DME, 85 °C
NaOtBu DMF, 100 °C
10
11
aIsolated yields. All reactions were conducted at the temperatures
shown for 16 h, and all starting materials were consumed unless
otherwise noted.
worked moderately well (Scheme 3). However, we found that
the best results were achieved when we subjected 16 to the
PIFA-mediated oxidative cyclization conditions reported by
Zeng and Chemler.4a Our modification of running the reac-
tion at -20 °C rather than at room temperature successfully
generated phenanthrene 17 in good yields (64-78% range),
with only a trituration of the crude reaction residue necessary
to isolate the purified product. As our preliminary experi-
ments suggested that an aryl bromide would be a more
suitable coupling partner versus the analogous aryl iodide
(Scheme 2), we decided to explore the halogenation of 17 to
prepare phenanthryl bromide 18. Thus, careful addition of
1.0 equiv of NBS11 to 17 at 0 °C cleanly formed the mono-
brominated adduct 18 in excellent isolated yields (98-99%),
whereas the use of greater than an equimolar amount of NBS
led to unsymmetrical bisbromination byproducts that were
inseparable from the desired product.
Finally, the phenanthryl bromide 18 was subjected to the
Wolfe carboamination cross-coupling/cyclization condi-
tions with the commercially available olefinic carbamate 7
(Table 1). We set up our first set of conditions for the
palladium-catalyzed protocol as described by the Wolfe
group for carbamate internal nucleophiles, using palladium
acetate as a precatalyst, DPE-phos as a ligand, and cesium
carbonate as a base in refluxing 1,4-dioxane.6 We observed
that our desired benzylic pyrrolidine (()-19 was isolated,
In summary, the present work further demonstrates the
synthetic utility of the Wolfe carboamination reaction as
an application to the construction of the natural product
(11) (a) Takeuchi, K.; Ishita, A.; Matsuo, J.; Ishibashi, H. Tetrahedron
2007, 63, 11101. (b) Lin, S.; Chen, Q.; You, T. Synlett 2007, 13, 2101.
ꢀ
(c) Brule, C.; Laali, K. K.; Okazaki, T.; Musafia, T.; Baird, W. M. Eur. J.
Org. Chem. 2007, 487.
(12) See also (a) Fu, Y.; Lee, S. K.; Min, H.-Y.; Lee, T.; Lee, J.; Cheng,
M.; Kim, S. Bioorg. Med. Chem. Lett. 2007, 17, 97. (b) Lebrun, S.; Couture,
A.; Deniau, E.; Grandclaudon, P. Tetrahedron 1999, 55, 2659. (c) Nordlander,
J. E.; Njoroge, F. G. J. Org. Chem. 1987, 52, 1627.
9556 J. Org. Chem. Vol. 74, No. 24, 2009

Rossiter et al.
JOCNote
SCHEME 4. Elaboration of Pyrrolidine 19 to (()-Tylophorine
after which the mixture was stirred for 1 h and then slowly
warmed to room temperature, stirring for a total of 7 h. The
mixture was diluted with 5% Na2S2O3 solution (50 mL) and
extracted with methylene chloride (2 ꢀ 15 mL). The combined
organic extracts were washed with saturated NaHCO3 and brine
solutions (20 mL each), dried over sodium sulfate, filtered, and
the solvent was removed under reduced pressure to provide
9-bromo-2,3,6,7-tetramethoxyphenanthrene (18) as a brown
solid (204 mg, 98%): see Supporting Information for spectro-
scopic and characterization data.
(()-tert-Butyl 2-((2,3,6,7-tetramethoxyphenanthren-9-yl)methyl)-
pyrrolidine-1-carboxylate [(()-19]4a,c. A degassed mixture of
9-bromo-2,3,6,7-tetramethoxyphenanthrene (18, 101.9 mg, 0.27
mmol), tert-butyl pent-4-enylcarbamate (7, 65.1 mg, 0.35 mmol),
tris(dibenzylideneacetone)dipalladium(0) (51.2 mg, 0.06 mmol),
4,5-bis(diphenylphosphino)-9,9-dimethylxanthene
(Xantphos,
tylophorine. The preparation of the requisite aryl bromide
precursor 18 was achieved in four steps, followed by a
palladium-catalyzed cross-coupling/cyclization reaction
with the commercially available olefinic carbamate 7 to
provide the 2-(2-arylmethyl)pyrrolidine (()-19 in 67% iso-
lated yield for the key step. A one-pot deprotection and
Pictet-Spengler cyclomethylenation of 19 with formalin
provided a near-quantitative yield of racemic tylophorine
[(()-1], providing an overall 35% yield of the natural pro-
duct over the six-step process. This application of the Wolfe
carboamination protocol as a key step to construct a natural
product provides further evidence of the utility of the
method, and further work is under investigation in our
laboratory to establish a catalytic asymmetric variant for
the preparation of nonracemic natural products.
62.0 mg, 0.11 mmol), and sodium tert-butoxide (68.4 mg, 0.71
mmol) in anhydrous 1,4-dioxane (5.0 mL) was heated at 100 °C
under nitrogen for 12 h. The cooled mixture was diluted with ethyl
acetate (25 mL) and partially purified by filtration through a plug of
Celite, eluting with ethyl acetate, and the solvents were removed
under reduced pressure. The residue was purified by flash column
chromatography on silica gel, eluting with ethyl acetate/methylene
chloride (gradient of 1:99 to 15:85), to provide racemic tert-butyl
2-((2,3,6,7-tetramethoxyphenanthren-9-yl)methyl)pyrrolidine-
1-carboxylate [(()-19] as an off-white solid (89.9 mg, 65%): see
Supporting Information for spectroscopic and characterization
data.
Acknowledgment. The authors wish to thank AMRI
scientists Paul E. Zhichkin and Larry Yet for many helpful
discussions, and Kent Kaltenborn for acquiring high-resolu-
tion mass spectrometry data.
Experimental Section
9-Bromo-2,3,6,7-tetramethoxyphenanthrene (18). N-Bromo-
succinimide (NBS, 98 mg, 0.55 mmol) was added to a solution
of 2,3,6,7-tetramethoxyphenanthrene (17, 172 mg, 0.58 mmol)
in anhydrous methylene chloride (5 mL) at 0 °C under nitrogen,
Supporting Information Available: Full experimental pro-
cedures as well as spectroscopic and characterization data for all
compounds prepared. This material is available free of charge
via the Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 74, No. 24, 2009 9557