Asymmetric synthesis of the cyclopentanones related to NCS and N1999A2 antitumor antibiotics

Article abstract of DOI:10.1016/S0040-4039(03)00565-3

Optically pure protected mono- or dihydroxylated cyclopentanones, precursors for the core of the antitumor antibiotics NCS and N1999A2, 1 and 2, were obtained in six to eight steps with excellent overall yields (up to 52%).

Full text of DOI:10.1016/S0040-4039(03)00565-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3391–3395
Asymmetric synthesis of the cyclopentanones related to
NCS and N1999A2 antitumor antibiotics
Philippe Bertus,† Jing-Heng Zhang, Geoffroy Sir, Jean-Marc Weibel and Patrick Pale*
Laboratoire de synthe`se et re´activite´ organique, associe´ au CNRS, Institut Le Bel, Universite´ L. Pasteur, 67000 Strasbourg,
France
Received 23 January 2003; revised 24 February 2003; accepted 27 February 2003
Abstract—Optically pure protected mono- or dihydroxylated cyclopentanones, precursors for the core of the antitumor antibiotics
NCS and N1999A2, 1 and 2, were obtained in six to eight steps with excellent overall yields (up to 52%). © 2003 Elsevier Science
Ltd. All rights reserved.
Neocarzinostatin Chromophore (NCS-Chrom; Scheme
1)1 and N1999A2 (Scheme 1)2 belong to the dienediyne
(DEDY) class of antitumor antibiotics.3 The potent
biological activities of these DEDY and their unique
structures have initiated increasing efforts from syn-
thetic chemists,4 which recently culminated with their,
so far sole, synthesis.5,6 A retrosynthetic analysis gives a
possible expedient route towards the synthesis of NCS-
Chrom. and related compounds based on coupling
reactions between appropriate diethynyloxiranes and
hydroxylated cyclopentenyl derivatives. Such a discon-
nection leads to two fragments of equal complexity
each including a ring, two unsaturated bonds and one
to three stereogenic centers, it would also provide the
opportunity to use combinatorial tools at the coupling
stage (Scheme 2).
conveniently prepared from either tartaric acid or malic
acid. Since all stereoisomers of these acids are commer-
cially available, any stereoisomers of 1 and 2 could be
As a prelude to the synthesis of NCS-Chrom7 and
analogs based on this strategy, we already reported on
new conditions for coupling polyfunctionalized
compounds8 and for direct coupling of silylated alky-
nes.9 We now described a convenient and rapid access
to optically pure five-membered ring precursors of the
NCS-Chrom and N1999A2 nuclei.
Scheme 1. Structures of NCS-Chrom. and N1999A2.
These functionalized cyclopentane derivatives can be
derived from the corresponding hydroxylated cyclopen-
tanones 1–2. These ketones could be obtained through
a ring closure reaction between a formyl dianion syn-
thon and mono- or dihydroxylated butane derivatives 3
or 4 (Scheme 2). These chiral derivatives could be
* Corresponding author.
† Present address: Laboratoire des re´actions se´lectives et applications
associe´ au CNRS, Universite´ de Reims-Champagne Ardenne, BP
1039, 51687 Reims, France.
Scheme 2. Retrosynthetic analysis of NCS and N1999A2.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00565-3

3392
P. Bertus et al. / Tetrahedron Letters 44 (2003) 3391–3395
obtained and used for structure–activity relationship
studies. The two protecting groups in 1 must be differ-
ent in order to later on functionalize it.
Despite their usefulness especially as intermediate for
prostano¨ıds syntheses, there are only a few reports
dealing with the synthesis of 1 or its enantiomer, pro-
tected or not. Surprisingly enough, only one synthesis
of 2 was mentioned in a paper dealing with the kinetic
resolution of racemic allylic alcohols.10 The four known
syntheses of 1 or ent-1 started from (R,R)- or (S,S)-tar-
taric acid and were based on a common strategy identi-
cal in essence to the one disclosed above. They,
however, led to C2-symmetrical compounds. Gero et al.
described a rather lengthy procedure (12 steps, overall
yield unknown).11 A similar procedure was reported by
Corey et al.12 More recently, Rokach et al. and Suffert
et al. reported improved versions of these routes (7–8
steps, 33–34% overall yield).13,14 In all cases, various
formyl anion equivalents were investigated with limited
success11,13,14 and the final recovery of the ketone from
the masked cyclopentanone derivatives so obtained
proved to be always troublesome, usually low yielding
and giving mixtures.13,14 In a synthesis of 1-nitro-1-
cyclopentenes, was mentioned a lengthy access to
Scheme 3. Mechanism of the cyclization with methylene–
triphenylphosphorane.
monoprotected 1 from D-glucose (11 steps, 34% overall
yield).15
To overcome these problems and nevertheless have a
rapid and scalable route to 1 and 2, we decided to
explore the potential of another type of formyl anion
equivalents, i.e. the methylenetriphenylphosphorane.
During his extensive studies on alkylidenephosphorane
reactivity,16 Bestmann described a ring closure reaction
of 1,v-dibromo or ditosyloxyalkanes by addition of
alkylidene phosphorane.17 It is worth noting that the
cyclic phosphonium salt so obtained could be isolated
and further oxidized to the corresponding cycloalka-
none.18 Therefore, this sequence would offer a rapid
access toward 1 and 2, and furthermore, would be the
first application of this reaction toward natural product
synthesis.
Scheme 4. Reagents and conditions: (a) pMeOPhCHO, CSA,
PhH; (b) NaBH4, LiCl, THF; (c) TsCl, Py, 93% over three
steps; (d) DIBAH, Tol, 89%; (e) DHP, PPTS, CH2Cl2 (PG=
THP, 3b 96%, 3e 100%); (f) CH2ꢀCHOEt, PPTS, CH2Cl2
(PG=EE: 3c 98%, 3f 100%); (g) NaI, Me2CO, 99%.
alcohol 3a was then protected, only the tetra-
hydropyranyl (THP)22 and ethoxyethyl (EE)23 groups
could however be introduced.24 The protected bistosyl-
ates 3b–c were obtained in almost quantitative yields
using conventional reaction conditions. In order to
check the scope of the Bestmann cyclization, we also
prepared the iodinated analogs 3e–f using usual condi-
tions.25 With these butane derivatives bearing different
protecting and leaving groups in hand, we then
explored the key cyclization step. Selected results of our
screening experiments are shown in Table 1.
Applied to our goal, this cyclization requires that 2
equiv. of methylenetriphenylphosphorane,19 must be
added to the mono- or dihydroxylated butanes 3 or 4
(Scheme 3). The so formed cyclic phosphonium 5 or 6
could then be isolated and oxidized by oxygen after
conversion to the corresponding ylide or it could be in
situ treated with base then with an aldehyde. The
corresponding alkylidene cyclopentane 7 or 8 can then
be ozonized to the desired cyclopentanone 1 or 2.
We were unable to isolate the expected cyclopentanone
1 from the complex mixtures obtained in any of the
conditions described for the direct oxidation of phos-
phonium ylides.17,18 We thus trapped the phosphonium
ylide in situ formed after addition of 2 equiv. of base by
simply adding excess paraformaldehyde to the reaction
mixture. The methylenecyclopentane derivatives 7a–b
(PG=THP or EE respectively) were thus isolated in
moderate to good yields depending on the conditions
(Table 1).26 The right combination of leaving and pro-
tecting groups seemed important for an efficient
cyclization (entries 1, 3 versus 2, 4), the best combina-
tion being iodide and EE respectively. However, the
most critical factor proved to be the nature of the base,
phenyllithium always giving the best results (see 1a,
entry 7 versus 4).27
The synthesis of 1 started from (S,S)-dimethyl tartrate
(Scheme 4). In order to differentiate its hydroxyl
groups, a para-methoxybenzylidene group was intro-
duced since it can be later reductively cleaved. After
quantitative protection of the diol and reduction of the
diester moiety through a known procedure,20 the so
formed diol was fully tosylated by an excess of tosyl
chloride in pyridine giving 9 with an excellent overall
yield. This bistosyl ketal was best opened with DIBAH
in toluene at room temperature.21 The corresponding

P. Bertus et al. / Tetrahedron Letters 44 (2003) 3391–3395
Table 1. Cyclization of tartrate derivatives27
3393
Entry
LG
PG
Conditions
Yields (%)a
1
2
3
4
5
6
7
OTs
I
OTs
I
OTs
I
I
THP
THP
EE
EE
THP
EE
3b
3e
3c
3f
3b
3f
3f
nBuLi, THF, 66°C
nBuLi, THF, 66°C
nBuLi, THF, 66°C
nBuLi, THF, 66°C
PhLi, THF, rt
32
35
25
59
55
65
79
PhLi, THF, rt
PhLi, THF, 66°C
EE
a Yields of pure isolated products.
gradually formed and the solution color discharged After
cooling to room temperature, PhLi (2 equiv.) was added
again. The suspension vanished to a dark solution. After
5 min, excess of formaldehyde was added. The color
discharged again. After stirring for 1 h, water was added
and the mixture was extracted three times with diethyl
ether. The combined organic layers were then dried with
MgSO4, filtered and concentrated under reduced pres-
sure. The crude product was then purified by chromato-
graphy.
S-Malic acid could be transformed in a similar way.
After esterification, the free hydroxyl group in 10 was
protected with different protecting groups without any
problem. The benzyl group was best introduced with the
benzyl transfer conditions catalyzed by Lewis acid.28 The
protected diesters 11a–c were then reduced and the
corresponding diols tosylated leading to 4a–c in good
yields. The diiodo derivatives 4d–f were also prepared
(Scheme 5).
These butane derivatives were then submitted to cycliza-
tion (Table 2). In these series, the leaving group as well
as the protecting group have a dramatic influence on the
reaction outcome. The compounds protected with a silyl
group could not be readily converted to the expected
methylenecylopentane 8b (entries 2, 5), while the
ethoxyethyl group only allowed for low to moderate
yields (entries 1, 4). The benzyl group proved to be the
best one giving up to 61% of the expected cyclized
product 8c (entries 3, 6). Here again, the iodides always
gave the best results (entries 4–6 versus 1–3).
1
7a (PG=THP; mixture of diastereoisomers): H NMR
(250 MHz, CDCl3): 1.48–1.90 (6H, m), 2.22–2.46 (2H,
m), 2.61–2.85 (2H, m), 3.46–3.58 (1H, m), 3.81 (3H, s),
3.81–4.05 (2H, m), 4.07–4.32 (1H, m), 4.51 (2H, br. s) and
4.54 (1H, d, J=11.0) 4.58 (1H, d, J=11.0), 4.66–4.80
(1H, m), 4.88–4.93 (2H, m), 6.87 and 6.88 (2H, d, J=8.6),
7.26 and 7.28 (2H, d, J=8.6); 13C NMR (62 MHz,
CDCl3): 19.44 and 19.74 (q), 25.44 (t), 30.92 and 31.01
(t), 36.40 and 36.68 (t), 37.09 and 37.96 (t), 55.24 (q),
62.32 and 62.69 (t), 70.87 and 70.95 (t), 79.52 and 81.04
(d), 82.66 and 82.98 (d), 96.95 and 98.67 (d), 107.64 (t),
113.73 (d), 129.09 (d), 130.66 and 130.84 (s), 146.28 and
146,78 (s), 159.08 and 159.14 (s).
The cyclization proved to be more efficient in the tartaric
series than in the malic series but in each case, mono- or
dihydroxylated methylenecyclopentanes 7 or 8 can be
rapidly obtained (5–7 steps). The corresponding ketones
1 or 2 were then conveniently obtained in good yields
(77–88%) through ozonolysis.
In conclusion, the above-mentioned results clearly
demonstrated that the Bestmann cyclization first can be
applied to highly functionalized compounds, and second
is a good alternative to other methods leading to
cyclopentane derivatives. Indeed, the protected mono- or
dihydroxylated cyclopentanones 1 or 2 were obtained in
6–8 steps with excellent overall yields (up to 52%)
Supplementary material
Scheme 5. Reagents and conditions: (a) AcCl, MeOH, 98%; (b)
11a PG=EE: CH2ꢀCHOEt, PPTS, CH2Cl2, 98%; 11b PG=
Typical procedure for the cyclization: To a suspension of
MePh3P+Br− (2 equiv.) in THF (0.12 M) was added a
commercial solution of PhLi (2 equiv.). After 30 min, a
0.25 M THF solution of 3 or 4 (1 equiv.) was added and
the resulting mixture was refluxed for 16 h. A precipitate
t
TPS: BuPh2SiCl, DMAP, Et3N, CH2Cl2, 93%; 11c PG=Bn:
BnOC(ꢀNH)CCl3, CH2Cl2, 79%; (c)NaBH4, LiCl, THF (12a
96%, 12b 79%, 12c 78%); (d) TsCl, Py (4a 90%, 4b 78%, 4c 79%);
(e) NaI, Me2CO (4d 69%, 4e 79%, 4f 88%).

3394
P. Bertus et al. / Tetrahedron Letters 44 (2003) 3391–3395
Table 2. Cyclization of malate derivatives
Entry
LG
PG
Conditions
Yields (%)a
1
2
3
4
5
6
OTs
OTs
OTs
I
I
I
EE
TPS
Bn
EE
TPS
Bn
4a
4b
4c
4d
4e
4f
PhLi, THF, 66°C
PhLi, THF, 66°C
PhLi, THF, 66°C
PhLi, THF, 66°C
PhLi, THF, 66°C
PhLi, THF, 66°C
20
0
42
45
5
61
a Yields of pure isolated products.
1
7b (PG=EE; mixture of diastereoisomers): H NMR
(250 MHz, CDCl3): 1.19 (3H, t, J=7.1), 1.31 (3H, d,
J=5,3), 2.24–2.40 (2H, m), 2.62–2.80 (2H, m), 3.39–
3.54 (1H, m), 3.59–3.73 (1H, m), 3.81 (3H, s), 3.84–4.00
(1H, m), 4.08–4.19 (1H, m), 4.49 and 4.52 (2H, br. s),
4.74 and 4.80 (1H, q, J=5.3), 4.86–4.93 (2H, m), 6.88
and 6.89 (2H, d, J=8.6), 7.26 (2H, d, J=8.6); 13C
NMR (62 MHz, CDCl3): 15.19 (q), 20.58 and 20.67 (q),
36.60 and 36.72 (t), 37.26 and 37.82 (t), 55.16 (q), 60.62
(t), 70.95 (t), 78.98 and 79.80 (d), 83.13 (d), 98.86 and
99.55 (d), 107.74 (t), 113.70 (d), 129.06 (d), 130.54 (s),
146.05 and 146,35 (s), 159.11 (s).
Mizugaki, M.; Koide, Y.; Seto, H.; Furihata, K.; Otake,
N.; Ishida, N. Tetrahedron Lett. 1985, 26, 331–334.
2. (a) Ishii, M.; Ando, T.; Kajiura, T.; Kameyama, T.;
Nihey, Y. Jpn. Kokai Tokkyo Koho JP 07291955; Chem.
Abstr. 1996, 124, 115564h; (b) Ando, T.; Ishii, M.; Kaji-
ura, T.; Kameyama, T.; Miwa, K.; Sugiura, Y. Tetra-
hedron Lett. 1998, 39, 6495–6498.
3. Kedarcidin: (a) Hofstead, S. J.; Matson, J. A.; Malacko,
A. R.; Marquard, H. J. Antibiot. 1992, 45, 1250–1254;
(b) Leet, J. E.; Schroeder, D. R.; Hofstead, S. J.; Golik,
J.; Colson, K. L.; Huang, S.; Klohr, S. E.; Doyle, T.
W.; Matson, J. A. J. Am. Chem. Soc. 1992, 114, 7946–
7948; C-1027: (c) Zhen, Y.-S.; Ming, X.-Y.; Yu, B.;
Otani, T.; Saito, H.; Yamada, Y. J. Antibiot. 1989, 42,
1294–1298; (d) Yoshida, K.-I.; Minami, Y.; Azuma, R.;
Saeki, M.; Otani, T. Tetrahedron Lett. 1993, 34, 2637–
2640; Maduropeptin: (e) Hanada, M.; Ohkuma, H.;
Yonemoto, T.; Tomita, K.; Ohbayashi, M.; Kamei, H.;
Miyaki, T.; Konishi, M. M.; Kawaguchi, H.; Forenza,
S. J. Antibiot. 1991, 44, 403–414; (f) Schroeder, D. R.;
Colson, K. L.; Klohr, S. E.; Zein, N.; Langley, D. R.;
Lee, M.; Matson, J. A.; Doyle, T. W. J. Am. Chem.
Soc. 1994, 116, 9351–9352.
4. For recent reviews, see: (a) Nicolaou, K. C.; Smith, A.
L. In Modern Acetylene Chemistry, Stang, P. J.;
Diederich, F., Eds.; VCH: Weinheim, 1995; pp. 203–283;
(b) Lhermitte, H.; Grierson, D. S. Contemporary
Organic Synthesis, Parts 1 and 2, 1996; pp. 41–63 and
93–124; (c) Grissom, J. W.; Gunawardena, G. U.;
Klingberg, D.; Huang, D. Tetrahedron 1996, 52, 6453–
6518.
5. (a) Myers, A. G.; Liang, J.; Hammond, M.; Harrington,
P. M.; Wu, Y.; Kuo, E. Y. J. Am. Chem. Soc. 1998,
120, 5319–5320; (b) Myers, A. G.; Glattar, R.; Ham-
mond, M.; Harrington, P. M.; Kuo, E. Y.; Liang, J.;
Schaus, S. E.; Wu, Y.; Xiang, J.-N. J. Am. Chem. Soc.
2002, 124, 5380–5401.
1
8a (PG=EE; mixture of diastereoisomers): H NMR
(200 MHz, CDCl3): 1.21 (3H, t, J=7.1), 1.30 (3H, d,
J=5.4), 1.65–2.08 (2H, m), 2.14–2.62 (4H, m), 3.49
(2H, q, J=7.1) and 3.62 (2H, q, J=7.1), 4.20 (1H, m),
4.72 (1H, q, J=5.4) and 4.76 (1H, q, J=5.4), 4.88 (2H,
m); 13C NMR (50 MHz, CDCl3): 15.36 (q), 20.69 (q),
30.22 and 30.39 (t), 32.05 and 32.34 (t), 35.60 and 36.28
(t), 60.31 (t), 76.35 (t), 98.58 and 98.76 (d), 115.79 (d),
140.17 (s); I.R. (neat): 1601, 1382, 1244, 1098, 987.
1
8c (PG=Bn): H NMR (200 MHz, CDCl3): 1.83–1.94
(2H, m), 2.18–2.64 (4H, m), 4.08 (1H, m), 4.53 (2H, s),
4.91 (2H, m), 7.22–7.39 (5H, s); 13C NMR (50 MHz,
CDCl3): 30.15 (t), 32.10 (t), 39.51 (t), 70.78 (t), 79.89
(d), 106.51 (t), 127.51 (d), 127.65 (d), 128.41 (d), 138.88
(s), 149.85 (s).
Acknowledgements
P.B. thanks the ‘Ministe`re de la Recherche et de la
Technologie’ for a doctoral fellowship. P.P. thanks the
‘Institut Universitaire de France’ for support.
6. (a) Kobayashi, S.; Reddy, R. S.; Sugiura, Y.; Sasaki, D.;
Miyagawa, N.; Hirama, M. J. Am. Chem. Soc. 2001,
123, 2887–2888; (b) Kobayashi, S.; Ashizawa, S.; Taka-
hashi, Y.; Sugiura, Y.; Nagaoka, M.; Lear, M.; Hirama,
M. J. Am. Chem. Soc. 2001, 123, 11294–11295.
References
1. (a) Ishida, N.; Miyazaki, K.; Kumagai, K. M.;
Rikimura, M. J. Antibiot. 1965, 18, 68–76; (b) Edo, K.;

P. Bertus et al. / Tetrahedron Letters 44 (2003) 3391–3395
3395
7. For related approaches towards NCS Chrom., see: (a)
Nakatani, K.; Arai, K.; Hirayama, N.; Matsuda, F.;
Terashima, S. Tetrahedron 1992, 48, 633–650; (b)
Nakatani, K.; Arai, K.; Yamada, K.; Terashima, S.
Tetrahedron 1992, 48, 3045–3060; (c) Suffert, J.; Bru¨ck-
ner, R. Synlett 1994, 51–53.
8. (a) Bertus, P.; Pale, P. Tetrahedron Lett. 1996, 37, 2019–
2022; (b) Bertus, P.; Pale, P. Tetrahedron Lett. 1997, 38,
8193–8196; (c) Bertus, P.; Pale, P. J. Organomet. Chem.
1998, 567, 173–180; (d) Halbes, U.; Bertus, P.; Pale, P.
Eur. J. Org. Chem. 2001, 4391–4393; (e) Dillinger, S.;
Bertus, P.; Pale, P. Org. Lett. 2001, 3, 1661–1664.
9. (a) Halbes, U.; Bertus, P.; Pale, P. Tetrahedron Lett.
2001, 42, 8641–8644; (b) Halbes, U.; Pale, P. Tetrahedron
Lett. 2002, 43, 2039–2042.
18. (a) Bestmann, H. J.; Kratzer, O. Chem. Ber. 1963, 96,
1899–1908; (b) Bestmann, H. J.; Armsen, R.; Wagner, H.
Chem. Ber. 1969, 102, 2259–2269.
19. The use of one equivalent of methylenephosphorane and
further treatment by bases did provide cyclized products
but in lower yield. Metallated methylenephosphorane
(Corey, E. J.; Kang, J. J. Am. Chem. Soc. 1982, 104,
4724–4725; Corey, E. J.; Kang, J.; Kyler, K. J. Tetra-
hedron Lett. 1985, 26, 555–558) did not yield an efficient
cyclization.
20. Somfai, P.; Olsson, R. Tetrahedron 1993, 49, 6645–6650.
21. Takano, S.; Akiyama, M.; Satoh, S.; Ogasawara, K.
Chem. Lett. 1983, 1593–1596.
22. Miyashita, M.; Yoshikoshi, A.; Grieco, P. A. J. Org.
Chem. 1977, 42, 3772–3774.
23. Fukusawa, A.; Sato, H.; Masamune, T. Tetrahedron Lett.
1987, 28, 4303–4306.
10. Kitamura, M.; Kasahara, I.; Manabe, K.; Noyori, R. J.
Org. Chem. 1988, 53, 708–710.
24. From the common silyl groups, only the triethyl- and
trimethylsilyl could be introduced, but in low yield. The
bulkiness of the reactive intermediate as well as the one
of the hydroxyl group in 3a might account for this
limitation.
25. Tipson, R. S.; Clapp, M. A.; Cretcher, L. H. J. Org.
Chem. 1947, 12, 133.
11. Barrie`re, F.; Barrie`re, J.-C.; Barton, D. H. R.; Cleophax,
J.; Gateau-Olesker, A.; Ge´ro, S. D.; Tadj, F. Tetrahedron
Lett. 1985, 26, 3119–3120.
12. Corey, E. J.; Niimura, K.; Konishi, Y.; Hashimoto, S.;
Hamada, Y. Tetrahedron Lett. 1986, 27, 2199–2202.
13. Khanapure, S.; Najafi, N.; Manna, S.; Yang, J.; Rokach,
J. J. Org. Chem. 1995, 60, 7548–7551.
26. As judged from TLC, these cyclization reactions are quite
clean. No double substitution product was detected using
our conditions as far as judged from NMR of the crude
mixture.
14. Raeppel, S.; Toussaint, D.; Suffert, J. Synlett 1997, 1061–
1062.
15. Torii, S.; Inokuchi, T.; Oi, R.; Kondo, K.; Kobayashi, T.
J. Org. Chem. 1986, 51, 254–256.
27. The reason why PhLi is the most effective base among
those investigated is still unclear.
16. Bestmann, H. J. Angew. Chem., Int. Ed. 1965, 4, 645–660
and 830–838.
28. Wessel, H. P.; Iversen, T.; Bundle, D. R. J. Chem. Soc.,
Perkin Trans. 1 1985, 2247–2249.
17. Bestmann, H. J.; Kranz, E. Chem. Ber. 1969, 102, 1802–
1815.