A model study for constructing the DEF-benzoxocin ring system of menogaril and nogalamycin via a reductive heck cyclization

Article abstract of DOI:10.1021/ol300072h

A novel reductive Heck cyclization approach was developed in order to construct a model DEF-benzoxocin ring system that is present in nogalamycin, menogaril, and related anthracycline antitumor antibiotics.

Full text of DOI:10.1021/ol300072h

ORGANIC
LETTERS
2012
Vol. 14, No. 8
1962–1965
A Model Study for Constructing the
DEF-Benzoxocin Ring System of Menogaril and
Nogalamycin via a Reductive Heck Cyclization
Ruogu Peng and Michael S. VanNieuwenhze*
Department of Chemistry, 800 East Kirkwood Avenue, Bloomington, Indiana 47405-7102,
United States
mvannieu@indiana.edu
Received January 12, 2012
ABSTRACT
A novel reductive Heck cyclization approach was developed in order to construct a model DEF-benzoxocin ring system that is present in
nogalamycin, menogaril, and related anthracycline antitumor antibiotics.
The anthracyclines constitute a widely used family of
chemotherapeutic agents. Nogalamycin 1, isolated from
Streptomyces nogalator,1 has potent biological activity
versus Gram-positive bacteria and shows prominent cyto-
toxicity against L1210 and KB cell lines in vitro. Studies
have shown that nogalamycin intercalates DNA with the
amino sugar binding in the major groove and the nogalose
subunit binding within the minor groove.2 The binding of
nogalamycin to an upstream site can induce highly specific
topoisomerase I mediated DNA cleavage.3 However, nogal-
amycin was found to show only weak activity against solid
tumors in vivo and have an unacceptable toxicity profile in
large animals.4 Its semisynthetic derivative, 7-con-O-methyl-
nogarol (menogaril) 2, possessed better antitumor activity
and was chosen for evaluation in a clinical trial.4
Nogalamycin and menogaril both contain a synthetically
interesting DEF-benzoxocin ring system. The F-ring unit,
also called nogalamine, has an L-glucose configuration. It is
attached to the anthracycline aglycone to form the E-ring by
an aryl C-glycosidic linkage and an O-glycosidic linkage.
Because of their interesting structure, along with their
promising antimicrobial and antitumor activities, there
has been much effort focused toward the total synthesis
of nogalamycin and related congeners. A total synthesis of
nogalamycin hasyet tobe achieved althoughseveral model
studies, most having focused on the enantioselective syn-
thesis of the DEF-ring system, have been reported.5
(5) (a) Krohn, K.; Ekkundi, V. S.; Doring, D.; Jones, P. J. Carbohyd.
Chem. 1998, 17, 153–170. (b) Krohn, K.; Florke, U.; Keine, J.; Terstiege,
I. J. Carbohyd. Chem. 1998, 17, 171–195. (c) Krohn, K.; Terstiege, I.;
Florke, U. J. Carbohyd. Chem. 1998, 17, 197–215. (d) Htwe, Y.; Franck,
R. W.; Chen, S. L.; Quigley, G. J.; Todaro, L. J. Org. Chem. 1992, 57,
644–651. (e) Deshong, P.; Li, W.; Kennington, J. W.; Ammon, H. L.;
Leginus, J. M. J. Org. Chem. 1991, 56, 1364–1373. (f) Kawasaki,
M.; Matsuda, F.; Terashima, S. Tetrahedron 1988, 44, 5713–5725.
(g) Kawasaki, M.; Matsuda, F.; Terashima, S. Tetrahedron 1988, 44,
5695–5711. (h) Bates, M. A.; Sammes, P. G.; Thomson, G. A. J. Chem.
Soc., Perkin Trans. 1 1988, 3037–3045. (i) Vatele, J. M. Tetrahedron
1986, 42, 4443–4450. (j) Joyce, R. P.; Parvez, M.; Weinreb, S. M.
Tetrahedon Lett. 1986, 27, 4885–4888. (h) Kawasaki, M.; Matsuda, F.;
Terashima, S. Tetrahedron 1988, 44, 5727–5743.
(1) Wiley, P. F.; Mackellar, F. A; Caron, E. L.; Kelly, R. B. Tetra-
hedron Lett. 1968, 663–668.
(2) (a) Sim, S. P.; Gatto, B.; Yu, C.; Liu, A. A.; Li, T. K.; Pilch, D. S.;
LaVoie, E. J.; Liu, L. F. Biochemistry 1997, 36, 13285–13291. (b) Liaw,
Y. C.; Gao, Y. G.; Robinson, H.; Vandermarel, G. A.; Vanboom, J. H.;
Wang, A. H. J. Biochemistry 1989, 28, 9913–9918.
(3) Sim, S. P.; Pilch, D. S.; Liu, L. F. Biochemistry 2000, 39, 9928–
9934.
(4) Wiley, P. F.; Elrod, D. W.; Houser, D. J.; Richard, F. A. J. Med.
Chem. 1982, 25, 560–567.
r
10.1021/ol300072h
Published on Web 04/05/2012
2012 American Chemical Society

In addition, syntheses of (þ)- and (()-7-con-O-methyl-
nogarol (menogaril 2) have been reported.5h,6 The overall
strategies for most of the reported syntheses are similar.
The C-glycosidic bond is established first. Subsequent
introduction of the O-glycosidic bond completes the
carbohydrate-bridged DEF-ring system.
Scheme 1. Proposed Catalytic Cycle for Reductive Heck
Cyclization of 3
An alternative strategy would be to develop an approach
wherein the O-glycosidic bond is established first. This in
turn will allow one to exploit the stereoelectronic preferences
dictated by the anomeric effect to install the quaternary aryl-
C-glycosidic bond with the desired stereochemistry. We are
aware of two reports that sought to exploit this strategy. In
one report, the formation of the C-glycosidic bond was
attempted via an intramolecular FriedelꢀCrafts alkylation
onto an electrophile-activated exocyclic 5,60-olefin.7 No
cyclization was observed, a result ascribed to the insufficient
nucleophilicity of the aromatic moiety. In a second report,
an aryl radical cyclization onto an exocyclic 5,60-olefin
provided none of the desired cyclization product and
afforded predominantly the direct reduction product of
the glycosidic aryl bromide radical precursor.8
Despite the failures of these attempts, we were intrigued
by this strategy since it could leverage the stereoelectronic
preference(s) dictated by the anomeric effect to introduce
the C-glycosidic bond with the desired stereochemistry. In
addition, this approach offered the possibility for late-
stage introduction of the bridging F-ring carbohydrate if a
suitably functionalized aromatic precursor could be pre-
pared. Overman,9 Grigg,10 and several others11 have de-
monstrated that the Heck cyclization, whether in the
normal or reductive mode, is particularly well suited for
the construction of quaternary CꢀC bonds. Successful
application of a reductive Heck cyclization may enable late
stage introduction of the DEF-ring system on a suitably
protected and fully functionalized anthracycline core.
Our proposed model for a reductive Heck cyclization
construct, along with pertinent mechanistic considerations,
is illustrated in Scheme 1. We envisaged aryl glycosides 3 and
4 as model substrates for these reactions. Although each is a
D-sugar (the nogalamine residue in nogalamycin has the
L-configuration), like nogalamine, each has the gluco relative
configuration and each is readily accessible from commer-
cially available precursors. In addition, the dimethyl ketal
present in 4 confers an added element of rigidity, relative to 3,
that could impact the efficiency of the cyclization reaction.
The desired reaction would take place via initial oxidative
addition of the Pd(0) catalyst to the aryl-Br bond generating
an arylpalladium(II) intermediate (e.g., 5 or 7). This inter-
mediate would then undergo an olefin insertion reaction
with the exocyclic olefin and generate an alkylpalladium(II)
intermediate (8). Capture of this intermediate by a suitable
hydride source, to generate a palladium hydride, followed
by reductive elimination would provide the desired cycliza-
tion product and regenerate the Pd(0) catalyst.
Several mechanistic aspects of this reaction mustworkin
our favor in order for this construct to be successful. First,
the olefin insertion reaction must proceed at a favorable
rate, relative to direct capture of the arylpalladium(II)
intermediate (e.g., 5) by hydride, in order to suppress
formation of the direct reduction product 6. Second, the
olefin insertion reaction must take place preferentially
in a 6-exo mode relative to the alternative 7-endo mode
(intermediate not shown). Our intuition suggested that the
desired mode of olefin insertion ought to be intrinsically
favored on both kinetic and thermodynamic grounds.
Finally, our construct required a kinetically competent
hydride source in order to ensure efficient capture of the
alkylpalladium(II) intermediate8 and enable catalyst turn-
over; however, if the olefin insertion reaction is sluggish,
the reactivity of the hydride source must be sufficiently
moderate in order to avoid direct capture of the aryl-
palladium(II) intermediate that would lead to the reduc-
tion product 6.
(6) Hauser, F. M.; Chakrapani, S.; Ellenberger, W. P. J. Org. Chem.
1991, 56, 5248–5250.
(7) Smith, T. H.; Wu, H. Y. J. Org. Chem. 1987, 52, 3566–3573.
(8) Meschwitz, S. M. Thesis (PhD), Brown University, 1989.
(9) (a) Overman, L. E. Pure Appl. Chem. 1994, 66, 1423–1430. (b)
Abelman, M. M.; Oh, T.; Overman, L. E. J. Org. Chem. 1987, 52, 4130–
4133. (c) Abelman, M. M.; Overman, L. E. J. Am. Chem. Soc. 1988, 110,
2328–2329. (d) Abelman, M. M.; Overman, L. E.; Tran, V. D. J. Am.
Chem. Soc. 1990, 112, 6959–6964.
(10) (a) Grigg, R.; Sridharan, V.; Stevenson, P.; Worakun, T.
J. Chem. Soc., Chem. Commun. 1986, 1697–1699. (b) Grigg, R.;
Sridharan, V.; Stevenson, P.; Sukirthalingam, S. Tetrahedron 1989, 45,
3557–3568. (c) Burns, B.; Grigg, R.; Santhakumar, V.; Sridharan, V.;
Stevenson, P.; Worakun, T. Tetrahedron 1992, 48, 7297–7320. (d) Grigg,
R.; Sukirthalingam, S.; Sridharan, V. Tetrahedon Lett. 1991, 32, 2545–2548.
(11) (a) Donets, P. A.; van der Eycken, E. V. Org. Lett. 2007, 9, 3017–
3020. (b) Banerjee, M.; Mukhopadhyay, R.; Achari, B.; Banerjee, A. K.
J. Org. Chem. 2006, 71, 2787–2796. (c) Ichikawa, M.; Takahashi, M.;
Aoyagi, S.; Kibayashi, C. J. Am. Chem. Soc. 2004, 126, 16553–16558. (d)
Li, Z.; Watkins, E. B.; Liu, H.; Chittiboyina, A. G.; Carvalho, P. B.;
Avery, M. A. J. Org. Chem. 2008, 73, 7764–7767. (e) Trost, B. M.; Thiel,
O. R.; Tsui, H. C. J. Am. Chem. Soc. 2003, 125, 13155–13164. (f) Minatti,
A.; Zheng, X. L.; Buchwald, S. L. J. Org. Chem. 2007, 72, 9253–9258.
The synthesis of 3 (Scheme 2) began with sulfoxide 10,
which was prepared in four steps from commercially
Org. Lett., Vol. 14, No. 8, 2012
1963

Scheme 2. Synthesis of 3
Scheme 3. Synthesis of 4
low conversions and undesirable product ratios.16 A large
increase in yield and product ratio was realized by using
Pd2(dba)3/(o-Tol)3P as the catalyst (entry 3). In this ex-
ample, the desired cyclization product was obtained in
65% isolated yield. We also found that mixtures of DMF
and water gave poorer product ratios while little or
no reaction was observed in dioxane and toluene. With
(o-Tol)3P as the ligand, neither Pd(OAc)2 (entry 4) nor the
Herrmann catalyst [trans-di-(μ-acetate)-bis-[o-(di-o-tolyl-
phosphine)-benzyl]dipalladium(II)] (entry 5) provided a
significant improvement in product ratios or conversion
when compared to Pd2(dba)3. Other hydride donors, such
as ammonium formate or triethylsilane, gave undesirable
product ratios and/or decomposition.16 Product ratios (9:6)
were unchanged by running the reactions at higher dilution.
available methyl-R-D-glucopyranoside.12 Sulfoxide 10 was
then subjected to Kahne glycosylation conditions (Tf2O,
TTBP, Et2O, 11, 68%) to give an inseparable mixture of
O-aryl glycosides 12.13 After cleavage of the acetate protect-
ing group (K2CO3, THF/H2O, 76%), the respective anom-
ers were cleanly separated by chromatographic purification.
The R-anomer 13 was subjected to iodination (I2, PPh3,
pyridine, toluene) followed by elimination of the primary
iodide to provide 3in good overall yield (70% for two steps).
Our synthesis of 4 (Scheme 3) began with thioglycoside
14 that was readily available in eight steps from pentaace-
tyl-R-D-glucopyranoside.14 Glycosylation with 11 (Tf2O,
TTBP, Ph2SO, DCM)15 followedbycleavageofthe acetate
protecting group provided 16 in good overall yield (49%
fortwosteps). Iodination of16(I2, PPh3, pyridine, toluene)
followed by elimination (DBU, MeCN, 61% overall) and
chromatographic separation of the respective R- and
β-anomers provided exocyclic olefin 4 in good overall yield.
Our initial experiments (summarized in Table 1) utilized
3 and were designed in order to identify an optimal source
of Pd(0). Catalytic Pd(PPh3)4 (0.1 equiv) gave a 70%
conversion of the substrate and provided a 4:1 ratio of
the cyclization (9)/direct reduction (6) products (entry 1).
Stoichiometric Pd(PPh3)4 reduced the ratio of 9/6 to
approximately 2.5:1, again with complete consumption
of 3 (entry 2). Increasing the temperature to 125 °C led
to decomposition. Common Heck reaction additives such
as silver salts or tetraethylammonium chloride resulted in
Table 1. Conditions for Reductive Heck Cyclization of 3
yieldb
entry
catalyst
ligand
ratioa
(%)
1
2
3
4
5
0.1 equiv Pd(PPh3)4
1 equiv Pd(PPh3)4
0.1 equiv Pd2(dba)3,
0.2 equiv Pd(OAc)2,
0.1 equiv
none
none
30:56:14 NDc
0:73:27
38
0.8 equiv (o-Tol)3P 0:88:12
0.8 equiv (o-Tol)3P 0:73:27
0.4 equiv (o-Tol)3P 0:78:22
68(65d)
43
(12) (a) Lu, W.; Navidpour, L.; Taylor, S. D. Carbohydr. Res. 2005,
340, 1213–1217. (b) Yang, G. B.; Ding, X. L.; Kong, F. Z. Tetrahedon
Lett. 1997, 38, 6725–6728. (c) Gildersleeve, J.; Pascal, R. A.; Kahne, D.
J. Am. Chem. Soc. 1998, 120, 5961–5969.
49
Herrmann catalyst
(13) Kahne, D.; Walker, S.; Cheng, Y.; Vanengen, D. J. Am. Chem.
Soc. 1989, 111, 6881–6882.
a Ratio of 3:9:6, determined by comparing the anomeric proton peak
1
area of each compound in crude H NMR. b NMR yield. c Not deter-
(14) (a) Dubois, E. P.; Pannell, L. J. Org. Chem. 1997, 62, 2832–2846.
(b) Dubois, E. P.; Robbins, J. B.; Pozsgay, V. Bioorg. Med. Chem. Lett.
1996, 6, 1387–1392. (c) Pozsgay, V.; Robbins, J. B. Carbohydr. Res. 1995,
277, 51–66. (d) Berry, J. M.; Dutton, G. G. S. Can. J. Chem. 1974, 52, 681–
683. (e) Hodge, J. E.; Rist, C. E. J. Am. Chem. Soc. 1952, 74, 1498–1500.
(15) van den Bos, L. J.; Dinkelaar, J.; Overkleeft, H. S.; van der Marel,
G. A. J. Am. Chem. Soc. 2006, 128, 13066–13067.
mined. d Isolated yield. All reactions were carried out with 1.2 equiv of
sodium formate (HCOONa).
Using the results above, a series of phosphine ligands was
screened with the Pd2(dba)3 catalyst (Table 2). (1-Naphthyl)3P
(entry 4) gave a comparable result to (o-Tol)3P. All the other
(16) Complete experimental details are reported in the Supporting
Information.
1964
Org. Lett., Vol. 14, No. 8, 2012

Table 2. Ligand Screening for the Reductive Heck Cyclization of 3
Table 3. Reductive Heck Cyclization of 4
yieldb
yield
(%)
entry
ligand
(p-Tol)3P
ratioa
(%)
entry
catalyst
ligand
ratioa
1
38:40:22
41:26:33
0:68:32
0:93:7
31
20
55
65
0
1
2
3
0.1 equiv Pd2(dba)3
0.5 equiv Pd2(dba)3
1 equiv Pd(PPh3)4
0.8 equiv (o-Tol)3P
4 equiv (o-Tol)3P
none
10:32:68
0:22:78
0:72:28
NDb
ND
44c
2
(p-F-phenyl)3P
(o-OMe-phenyl)3P
(1-Naphthyl)3P
(2,4,6-tri-OMe-Ph)3P
(2,4,6-tri-Me-Ph)3P
(CyHex)3P
3
4
a Ratio of 4:17:18, determined by integration of the anomeric proton
peak area of each compound in crude 1H NMR. b Not determined.
c Isolated yield.
5
17:0:83
25:3:72
0:13:87
90:1:9
6
NDc
7
9
8
nBu3P
ND
11
8
9
tBu3P
dpppd
dppfd
DPEphosd
0:14:86
67:9:24
53:0:47
21:0:79
the arylpalladium(II) intermediate tries to access the tran-
sition state for olefin insertion (e.g., as in the conversion of
7 to 8).
10
11
12
0
0
Table 2 also reveals that ligand selection may also exert a
significant influence on product ratios. Reactions using
bidentate and bulky monodentate ligands gave the direct
reduction product (6) almost exclusively. This may reflect
the lack of an accessible coordination site for the olefin
prior to olefin insertion. Alternatively, it may also suggest
steric implications for the catalyst ensemble that suppress
access to the transition state for olefin insertion.
In conclusion, we have developed a novel approach for
the construction of the DEF-benzoxocin ring system of
nogalamycin, and related congeners, via a reductive Heck
cyclization. The product ratios obtained from these reac-
tions show a marked dependence on the choice of catalyst
precursor, ligand, and substrate. The successful construc-
tion of a model DEF-benzoxocin ring system has catalyzed
subsequent efforts in our laboratory directed at the total
synthesis of the nogalamycin family of anthracycline anti-
tumor antibiotics. The results of these studies will be
presented in due course.
a Ratio of 3:9:6, determined by integration of the anomeric proton
peak area of each compound in crude 1H NMR. b NMR yield. c Not
determined. d 0.2 equiv of ligand was used.
phosphines, triarylphosphines (entries 1ꢀ3, 5ꢀ6), trialkyl-
phosphines (entries 7ꢀ9), or bidentate phosphines (entries
10ꢀ12), resulted in incomplete conversions or undesirable
product ratios.
The optimal reductive Heck cyclization conditions were
then tried on substrate 4 (Table 3). Under these conditions
(0.1 equiv Pd2(dba)3, 0.8 equiv (o-Tol)3P, DMF), the reaction
of 4 gave incomplete conversion and a 2:1 ratio of products
favoring the direct reduction product (entry 1). With stoichio-
metric Pd2(dba)3, complete consumption of starting
material was observed, but the ratio of direct reduction
to cyclization product (18/17) increased (entry 2). A
reaction employing stoichiometric Pd(PPh3)4 resulted
in a complete reversal of product selectivity; the desired
cyclization product (17) was isolated as the major
product of a 72:28 product mixture (entry 3).
It is interesting to note that the results obtained with 3
and 4are qualitativelyidentical withrespecttotheextentof
conversion and product yields/ratios when stoichiometric
Pd(PPh3)4 was used (cf., Table 1, entry 2 and Table 3, entry 3).
When substoichiometric amounts of catalyst were used,
substrates 3 and 4 showed opposing product preferences; 3
preferentially provided the desired cyclization product
while reactions with 4 resulted in preferential formation
of a direct reduction product (e.g., 18). This observation
may be the result of additional strain energy in 4 due to the
fused five-membered ketal ring that is exacerbated as
Acknowledgment. Financial support for this work was
provided by Indiana University. M.S.V. acknowledges
Eli Lilly and Company for an Eli Lilly New Faculty
Award. R.P. gratefully acknowledges the Department of
Chemistry at Indiana University for support in the form of
a Carmack Fellowship.
Supporting Information Available. Experimental details
and spectral data for all new compounds. This material
is available free of charge via the Internet at http://
pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 8, 2012
1965