Microbial and reducing agents catalyze the rearrangement of taxanes

Article abstract of DOI:10.1016/S0968-0896(01)00106-7

5α, 7β, 9α, 10β, 13α-Pentahydroxy-4(20),11(12)-taxadiene derivative 1 was converted to two unprecedented 1(15→11)abeo-taxanes and a taxane derivative with a C10-C11 double bond by Absidia coerula ATCC 10738a. A similar compound was obtained from treatment with zinc of a triacetoxy-4(20),11(12)-taxadiene derivative.

Full text of DOI:10.1016/S0968-0896(01)00106-7

Bioorganic & Medicinal Chemistry 9 (2001) 1985–1992
Microbial and Reducing Agents Catalyze the Rearrangement
of Taxanes
Di-An Sun,a Anastasia Nikolakakis,a Francoise Sauriol,b Orval Mamerc
and Lolita O. Zamira,d,
*
a
´ ´
Human Health Research Center, INRS-Institut Armand-Frappier, Universite du Quebec, 531Boulevard des Prairies,
´
Laval, Quebec, Canada H7V 1B7
bDepartment of Chemistry, Queen’s University, Kingston, Ontario, Canada K7L 3N6
c
´
Biomedical Mass Spectrometry Unit, McGill University, 1130 Pine Avenue west, Montreal Quebec, Canada H3A 1A3
dMcGill Center for Translational Research in Cancer, Sir Mortimer B. Davis-Jewish General Hospital, 3755 Coˆte Ste.
´
Catherine Rd., Suite D-127, Montreal, Quebec, Canada H3T 1E2
Received 13 December 2000; accepted 28 February 2001
Abstract—5a, 7b, 9a, 1 b0, 1 a3-Pentahydroxy-4(20),11(12)-taxadiene derivative
1 was converted to two unprecedented
1(15!11)abeo-taxanes and a taxane derivative with a C10–C11 double bond by Absidia coerula ATCC 10738a. A similar com-
pound was obtained from treatment with zinc of a triacetoxy-4(20),11(12)-taxadiene derivative. # 2001Elsevier Science Ltd. All
rights reserved.
The use of enzymes in organic synthesis has been stud-
ied extensively since the 1980’s.1ꢀ5 Impressive advances
have been reported using enzymes, which are functional
in organic, aqueous or biphasic systems.6ꢀ8 Recently,
filamentous fungi have been used for taxanes9ꢀ24 and
stereospecific hydroxylations were discovered, some of
them in relatively high yields.12 These authors reported
the C-1(39% yield) and C-41 hydroxylations (26%
yield) of 5a, 7b, 9a, 1 0b, 1 3a-pentaacetoxy-4(20),11(12)-
taxadiene by Absidia coerula. We have also recently
reported25 the biotransformation of a 4(20),11(12)-tax-
adiene derivative by different microorganisms. The pro-
ducts included an efficient C-1-hydroxylation, and
conversion to a C-15-hydroxylated 11(15!1)abeo-tax-
ane. In addition, reported for the first time were iso-
merizations (trans- to cis-cinnamoyl) and a C-17-
hydroxylation, which were detected but in low yields.
The main disadvantages of the use of microorganisms is
that the specificity of the substrate and the products
obtained cannot be predicted. In addition, the yields are
usually very low.
(Schemes 1and 2) with A. coerula ATCC 10738a. Tax-
ane 1 was the only one biotransformed. The taxane
product 5 (Scheme 2) obtained in 32% yield has a C-13
ketone and a C10–C11 double bond, resulting from
oxidation–reduction steps. In addition, two unprece-
dented 1(15!11)abeo-taxanes 6 and 7 (Scheme 2) were
obtained in 42 and 4% yields, respectively. A chemical
reduction (zinc/acetic acid) of a related 4(20),11(12)-
taxadiene derivative 3 (Schemes 1and 3) with a ketone
on C-13, gave a major product 8 which has a similar
structure to the taxane 5 obtained by biotransformation.
Results and Discussion
We intended to study the specificity of substrates with
different fungi. A. coerula was known to catalyze C-1
and C-14 hydroxylations of 5a, 7b, 9a, 1 0b, 1 3a-pen-
taacetoxy-4(20),11(12)-taxadiene.12 We have indeed
reproduced these results when we used this same sub-
strate and A. coerula ATCC 10738a. We found that the
partially acetylated substrates, 5a,13a-dihydroxy, 7b,
9a, 1 0b-triacetoxy-4(20), 11(12)-taxadiene 226 and 5a-
hydroxy, 7b, 9a, 1 0b-triacetoxy-4(20), 11(12)-taxadien-
13-one 3,27 gave no reaction when incubated with A.
coerula ATCC 10738a. Similarly, 5a, 7b, 9a, 1 0b-tetra-
hydroxy-4(20), 11(12)-taxadien-13-one 4 (Scheme 1) was
unreacted after incubation with this fungi. On the other
In this publication, we are reporting the biotransforma-
tion of the 4(20),11(12)-taxadiene derivatives 1–4
*Corresponding author. Tel.: +1-450-687-5010, extn 4260; fax:
(office) +1-450-686-5501, (residence): +1-514-481-2797; e-mail: lolita.
zamir@iaf.uquebec.ca
0968-0896/01/$ - see front matter # 2001Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00106-7

1986
D.-A. Sun et al. / Bioorg. Med. Chem. 9 (2001) 1985–1992
hand, the taxane derived from decinnamoylation and
complete deacetylation of 2-deacetoxytaxinine J: 5a, 7b,
9a, 1 b0, 1 a3-pentahydroxy-4(20),11(12)-taxadiene 1,28
(Scheme 1) was a good substrate for A. coerula ATCC
10738a. Surprisingly, the expected C-1 and C-14 hydro-
xylations were not observed. Two major taxanes 5 (32%
yield) and 6 (42% yield) were obtained as well as a
minor product 7 (4%) (Scheme 2). The compound 5 was
characterized as shown in Scheme 2 with a ketone at C-
13 and a double bond at C10–C11. This is a new taxane,
but taxoids with a C10–C11 double bond have pre-
viously been obtained by chemical reactions.29 Taxanes
6 and 7 were found to be unprecedented 1(15!11)abeo-
taxanes (Scheme 2).
Upon treating the derived 2-deacetoxydecinnamoyltax-
inine J26 with excess LiOH.H2O for 7 h at room tem-
perature, the major product involved complete
deacetylation (1, 78% yield) and a minor compound
with C-13-deacetylated (2, 7% yield). On the other
hand, treatment of the same startiꢁng material with
compound 2 whose oxidation in situ yielded the sub-
strate 3 with an overall yield of 55%. Compound 3 is
not a new compound, it was previously isolated as a
natural product.27 The taxane derivative 4 was obtained
by basic deacetylation of 3.
.
excess LiOH H2O for 75 min at 4 C gave a major
5ꢀ, 7ꢁ, 9ꢀ-trihydroxy-4(20),10(11)-taxadien-13-one (5,
Scheme 2, Table 1)
An attempt to obtain a 12,13-isotaxane,30 from the tax-
ane derivative 1 by treatment with zinc/acetic acid failed
and no taxanes similar to 5–7 were obtained. On the
other hand, the triacetylated ketone 3 when treated with
zinc/acetic acid gave a major compound 8 (23% yield)
(Scheme 3). Its structure is similar to the bio-
transformation product 5. The semi-syntheses of the
substrates, characterization of the new taxoids (5–8) as
well as plausible mechanisms for their formation are
described in the following sections.
1
Comparison of the H NMR of 5 (Table 1) to 1 (see
Experimental) shows that ring C remains unchanged
(Me-19, H-7, H-6a,b, H-5, H-20a,b). On the other hand,
there are major differences in rings A and B, both in the
1H and 13C NMR. The first obvious finding in 5 is the
ketone at C-13 (d 211.3 ppm) and a characteristic CH3–
CH demonstrated by the H-12 quartet at d 3.67 ppm
(J=6.3 Hz) coupled with the Me-18 doublet at d 1.30
ppm (J=6.3 Hz). This Me-18 is correlated in the
HMBC to two quaternary carbons: the ketone C-13 and
the olefinic carbon C-11. In the HMQC of 5, we detec-
ted that the H-10 is attached to the olefinic carbon C-10.
These observations proved that the original C11–C12
double bond has shifted to C10–C11 in 5. As expected,
H-10 is a doublet at d 5.68 ppm (J=11.0 Hz) coupled
Semi-synthesis of 5ꢀ, 7ꢁ, 9ꢀ, 10ꢁ, 13ꢀ-pentahydroxy-
4(20),11(12)-taxadiene, 1; 5ꢀ, 13ꢀ-dihydroxy-7ꢁ, 9ꢀ, 10ꢁ-
triacetoxy-4(20),11(12)-taxadiene, 2; 5ꢀ-hydroxy,7ꢁ, 9ꢀ,
10ꢁ-triacetoxy-4(20),11(12)-taxadien-13-one, 3; and 5ꢀ,
7ꢁ, 9ꢀ, 10ꢁ-tetrahydroxy-4(20),11(12)-taxadien-13-one,
4 (Scheme 1)
The taxane derivatives (1–4) were semi-synthesized from
2-deacetoxytaxinine J.31 The first step involved C-5-
decinnamoylation according to a reported procedure.32
Scheme 2. Microbial transformation products and yields of taxane 1
by fungi A. coerula ATCC 10738a: (a) yield of taxane 5: 32%; (b) yield
of taxane 6: 42%; (c) yield of taxane 7: 4%.
Scheme 1. Semi-syntheses of substrates 1–4. Reagents and conditions:
(a) NH2OH.HCl/NaOAc in EtOH/H2O, 80 ꢁC, 24 h,26,32 90%; (b)
LiOH, MeOH/H2O, 1:1, 23 ꢁC, 7 h, taxane 1 78%, taxane 2 7%; (c)
LiOH, MeOH/H2O, 1:1, 4 ꢁC, 75 min taxane 2 not isolated but oxi-
dized in situ according to (d); (d) CrO3 in pyridine/CH2Cl2, 1:1, 23 ꢁC,
2 h, taxane 3 55%; (e) 1N NaOH in MeOH, 23 ꢁC, 10 h, taxane 4
63%.
Scheme 3. Reduction of the ab-unsaturated ketone 3. Reagents and
conditions: activated zinc and acetic acid 23 ꢁC, 42 h, taxane 8 23%.

D.-A. Sun et al. / Bioorg. Med. Chem. 9 (2001) 1985–1992
1987
with H-9 (d 5.08 ppm, o.d. J=11.0 Hz). The HMBC
correlations in Table 1for taxane 5 confirm these
assignments. The NOESY data proves the stereo-
chemistry shown in Scheme 2 for taxane 5. In parti-
cular, the stereochemistry at the new stereocenter C-12
is demonstrated by the NOESY correlations: H-9 is b
oriented because of its strong correlations with Me-19
and Me-17. In addition, a similarly strong correlation
between H-12 and Me-16 proves that H-12 is also b
oriented whereas Me-18 is a. High resolution mass
spectrometry confirmed the elemental composition of
the potassium quasimolecular ion of taxane 5.
Table 1. 1H and 13C NMR for taxane 5 in CDCl3
1
Position
d H mult.a J (J in Hz)
d
13Cb
HMBC
NOESYc
1
2.12 o.m
1.88 ddd (16.4, 5.9, 1.7)
1.75 ddd(16.4, 5.9, 2.0)
2.22 br.d (5.7)
—
4.29 t (2.9)
2.03 ddd (13.5, 5.0, 2.9)
1.54 ddd (13.5, 11.4, 3.2)
4.37 dd (11.4, 5.0)
—
44.1
26.3
11
w
w
2a
2b
3
4
5
6a
6b
7
8
9
10
11
12
13
14a
14b
15
16
17
18
19
20a
20b
1, 3, 8, 15
1 , 2, 8, 1 5
1, 2, 8
2bs, 9/20aw, 1 7 , 1 9
2a, 1/14bm, 20bs
s
w
w
35.6
7s, 1 0 , 1 4b
152.4
73.3
38.2
6bw, 20a/2s
5w, 6bs, 7w
s
5s, 6as, 1 9
s
68.1
51.3
77.6
125.5
144.2
52.1
211.3
42.3
6, 19
3s, 1 0
s
s
5.08 o.d (11.0)
5.68 d (11.0)
—
3.67 q (6.3)
—
2.77 dd (18.5, 8.5)
2.09 o.d (18.5)
—
7, 10, 11, 19
9, 11, 12, 15
2am, 1 7, 1 9
s
3w, 7s, 1 8
m
10, 11, 13, 18
16s, 1 8
w
1, 2, 13
1, 2, 12, 13, 14, 15
14bs, 1 6
2bw, 1 4a, 1 6
s
s
41.2
30.5
26.5
11.8
11.3
110.8
s
s
s
1.40 s
1.60 s
1.30 d (6.3)
0.98 s
5.09 s
1, 11, 15, 17
1, 11, 15, 16
11, 12, 13
3, 7, 8, 9
3, 8
1w, 1 2, 1 4a, 1 7
1/14bw, 2as, 9s, 1 6, 1 9
s
w
s
10s, 1 2
2aw, 6bw, 9w
5s, 20bs
4.74 d (1.0)
3, 4, 5, 8
2bs, 20as
aMult. multiplicity: s, singlet; d, doublet; t, triplet; q, quartet; dd, doublet of doublet; br, broad; m, multiplet; o, overlapping. The precision of the
coupling constants is ꢂ0.5 Hz.
bThe 13C chemical shifts were extracted from the HMQC and HMBC experiments (for quaternary carbons) (ꢂ0.2 ppm).
cNOESY intensities are marked as strong (s), medium (m) or weak (w).
Table 2. 1H and 13C NMR for taxane 6 in CDCl3
Position
d 1H mult.a J (J in Hz)
COSY
d
13Cb
HMBC
NOESYc
w
s
1
2.83 dt (9.2, 4.9, 4.9)
2.09 o.m
1.16 ddd (14.2, 5.6, 1.8)
2.01o.m
—
4.27 t (2.9)
2.05 o.m
1.59 ddd (14.0, 11.4, 3.0)
2a, 2b, 14a
Overlap
1, 2a/3
39.3
32.1—
11, 12, 15
2aw, 9s, 1 4a , 1 6
2a
2b
3
4
5
6a
6b
7
8
9
10
11
12
13
14a
14b
15
16
17
18
19
20a
20b
s
3/14, 4, 8
2, 4, 8, 20
2as, 1 4b, 20bs
Overlap
37.9
150.4
73.3
37.4
–
6a, 6b
Overlap
5, 6a, 7
6a, 6b
6bw, 20aw
—
7
6, 9, 19
5m, 6as
4.32 dd (11.4, 5.1)
—
4.04 d (10.6)
4.81d ( 10.6)
68.3
44.2
82.0
77.0
10s, 6as/3s/18s
m
s
10
9
7, 10, 11, 19
9, 11, 15
1s, 1 6 , 1 9
7s, 6as/3s/18s
—
—
60.4
140.8
126.7
40.6
5.49 d (1.5)
2.28 dm
1.73 br.dd (15.4, 1.5)
2a, 14b
1, 13, 14b, 18
13, 14a, 18
2 (weak)
2, 11, 12
12, 13
18m
s
1s, 1 4b, 1 6
2bs, 1 3 , 1 4a
w
m
s
—
1.47
1.28
2.04
0.94
5.00 s
4.61br.s
89.6
30.2
26.8
15.4
11.8
110.7
w
11, 15, 17
11, 15, 16
11, 12, 13
3, 7, 8, 9
4, 5, 6, 8
4, 5, 6, 8
1s, 9s, 1 4a
18w
Overlap
2aw, 6bm, 9s, 20bw
5s, 20bs
3, 20b
3, 20a
20a s, 2bm
aMult. multiplicity: s, singlet; d, doublet; t, triplet; dd, doublet of doublet; br, broad; m, multiplet; o, overlapping. The precision of the coupling
constants is 0.5 Hz.
bThe 13C chemical shifts were extracted from the HMQC and HMBC experiments (for quaternary carbons) (0.2 ppm).
cNOESY intensities are marked as strong (s), medium (m) or weak (w).

1988
D.-A. Sun et al. / Bioorg. Med. Chem. 9 (2001) 1985–1992
Rearrangement
1(15!11)abeo-taxane (6, Scheme 2, Table 2)
to
a
major
unprecedented
Table 3. 1H and 13C NMR for taxane 7 in CDCl3
Position
d 1H mult.a J
(J in Hz)
d
13Cb
HMBC
NOESY
Overlap
The compound with the highest yield obtained from this
biotransformation was taxane 6 (42%). Comparison of
1
2.81o.m
1.67 o.m
1.56 dd (15.0, 3.8)
2.83 o.m
—
44.1
23.6
1
the H NMR of 1 (see Experimental) and 6 (Table 2)
shows that ring C in 6 is practically unchanged (Me-19,
2a
2b
3
4
5
6a
6b
7
3, 4, 8, 11 1/3, 2a, 20b
1
34.1
150.7
73.3
H-7, H-6a,b, H-5, H-20a,b). On the other hand, the H
and 13C NMR of ring A and B of taxane 6 are different,
in particular ring A. The signal for H-13 has dis-
appeared from the HMQC spectra and a new olefinic
proton appearing as a doublet at d 5.49 ppm (J=1.5
Hz) is connected to an olefinic carbon at d 126.7 ppm.
As this olefinic carbon is correlated with Me-18 in the
HMBC experiment, and the olefinic proton shows
COSY correlation to methylene proton H-14, the only
position for this carbon is on C-13. At first, the struc-
ture of 6 seemed to be that of a usual 11(15!1)abeo-
taxane with two newly introduced quaternary carbons
at d 60.4 ppm and d 89.6 ppm. A close look at the
HMBC experiment demonstrates that the gem-dimethyl
group is attached to C-11 and not C-1. Indeed in the
HMBC experiment, Me-18 shows correlations to the
expected three carbons, C-11, C-12, and C-13. The
assignment of the aliphatic type of quaternary carbon as
C-11 is supported by several other correlated protons
surrounding it within 2–3 bonds. The observation that
the protons of Me-16/17 (d 1.47 and 1.28 ppm) are cor-
related in the HMBC with C-11 (d 60.4 ppm), C-15 (d
89.6 ppm), and Me-16/17 (d 30.2 and 26.8 ppm)
4.33 o.m
2.11 o.m
1.71 o.m
4.36 o.m
—
20a
38.4
69.6
43.7
8
9
3.98 d (7.5)
4.18 d (7.5)
—
76.9
70.8
64.8
146.1
126.2
34.9
17b, 19
10
11
12
13
14a
14b
15
16
17a
17b
18
19
20a
20b
—
5.54 br.s
2.33 br.dd (15.2, 10.2)
2.14 br.dd (15.2, 8.4)
—
18
14b
1/3, 14a, 18
144.0
22.6
114.9
1.71 s
5.15 s
5.05 s
1.75 br.s
0.97 s
5.07 s
11, 15, 17
11, 16
15, 16
11, 12, 13
3, 7, 8, 9
3, 5
1/3, 17a
16, 17b
1, 9, 17a
13, 16
2a, 9
5, 20b
2b, 20a
16.8
13.4
110.6
4.68 s
3, 4, 5
aMult. multiplicity: s, singlet; d, doublet; t, triplet; dd, doublet of
doublet; br, broad; m, multiplet; o, overlapping. The precision of the
coupling constants is ꢂ0.5 Hz.
bThe 13C chemical shifts were extracted from the HMQC and HMBC
experiments (for quaternary carbons) (ꢂ0.2 ppm).
Table 4. 1H and 13C NMR for taxane 8 in CDCl3
Position
d 1H mult.a J (J in Hz)
d
13Cb
HMBC
NOESYc
1
2.12 o.m
1.97 o.m
1.76 o.m
2.34 br.d (5.4)
—
4.27 t (2.9)
1.89 o.m
1.59 o.m
5.60 dd (11.6, 5.2)
—
43.8
25.8
2a
2b
3
4
5
6a
6b
7
1, 3, 8
1, 2, 8, 18
3, 20
w
m
35.8
7s, 1 0 , 1 4b
151.4
72.7
36.2
6am, 6bm, 20as
s
67.6
51.1
6, 19, 169.6
3s, 6as, 1 0
8
s
m
9
6.12 d (11.3)
5.72 d (11.4)
—
3.64 q (6.4)
—
2.77 dd (18.5, 8.8)
2.03 o.m
—
77.2
7, 8, 10, 11, 19, 170.4
8, 9, 11
2am, 1 7, 1 9
s
10
11
12
13
14a
14b
15
16
17
18
19
20a
20b
OAc
120.9
145.6
52.3
211.2
42.2
3w, 7s, 1 8
m
10, 11, 13, 18
16s, 1 8
m
1, 2, 13
1, 2, 12, 15
14bs, 1 6
41.0
30.5
26.0
11.5
11.8
111.1
s
m
s
1.37 s
1.71 s
1.32 d (6.4)
0.87 s
5.11 s
4.77 s
2.03 s
2.01s
1, 11, 15, 17
1, 11, 15, 16
11, 12, 13
3, 7, 8, 9
3, 5
1m, 1 2, 1 4a , 1 7
1s, 2as, 9s, 1 6
s
10s, 1 2
s
2am, 6bm, 9s
5s, 20bs
3, 4, 5
2bs, 20as
aMult. multiplicity: s, singlet; d, doublet; t, triplet; q, quartet; dd, doublet of doublet; br, broad; m, multiplet; o, overlapping. The precision of the
coupling constants is ꢂ0.5 Hz.
bThe 13C chemical shifts were extracted from the HMQC and HMBC experiments (for quaternary carbons) (ꢂ0.2 ppm).
cNOESY intensities are marked as strong (s), medium (m) or weak (w).

D.-A. Sun et al. / Bioorg. Med. Chem. 9 (2001) 1985–1992
1989
confirms that the structure 6 is an 1(15!11)abeo-taxane.
Similarly, the proton H-1, which shows COSY correla-
tions with H-2 and H-14, is correlated in HMBC with
C-11 (d 60.4 ppm), C-12 (d 140.8 ppm) and C-15 (d 89.6
ppm). These observations also confirm the structure
assignment of 6. The relative stereochemistry of taxane
6 was established using the information contained in the
NOESY experiment shown in Table 2. High resolution
mass spectrometry confirmed the elemental composition
of the potassium quasimolecular ion of taxane 6.
also has HMBC correlations with C-11 and two olefinic
carbons C-15 and C-17 (d 144.0 and 114.9 ppm, respec-
tively). The H-17 protons, being correlated in the
HMBC with C-11, C-15 and Me-16, further confirm
that 7 has the structure shown in Scheme 2.
5ꢀ-Hydroxy-7ꢁ, 9ꢀ-diacetoxy-4(20),10(11)-taxadien-
13-one (8, Scheme 3, Table 4)
Comparison of the 1H NMR of 8 and 5 (Table 1) shows
the presence of two extra acetyl methyl groups (d 2.03
and 2.01ppm). All the protons in these two compounds
are observed at similar shifts except for H-7 (d 5.60
ppm) and H-9 (d 6.12 ppm) that are significantly more
deshielded than one can expect upon replacing the
hydroxy group by acetoxy. The presence of a CH3–CH
group is deduced from the H-12 quartet at d 3.64 ppm
coupled to the Me-18 doublet at d 1.32 ppm (J=6.4
Hz). This methyl shows HMBC correlation to the
ketone C-13 and to the olefinic quaternary carbon C-11.
The H-10 olefinic doublet (d 5.72 ppm), coupled to H-9
(d 6.12 ppm), has an HMBC correlation to C-11 con-
firming the position of the shifted C10–C11 double
bond. The position of each acetoxy group at C-7 and C-
9 has been confirmed by HMBC, as the H-7 and H-9
protons show correlation to the carbonyl ester group. In
the NOESY experiment, the b orientation of H-9 is
confirmed by its interaction with Me-19 and Me-17. The
a-stereochemistry of Me-18 in 8 is determined by the
NOESY correlations of H-12 with Me-16 which are
therefore both in the b-configuration. The correlation
between H-3, H-7, H-10, and H-14b allow us to
conclude that these protons have an a orientation.
Rearrangement
to
a
minor
unprecedented
1(15!11)abeo-taxane (7, Scheme 2, Table 3)
A minor product (4% yield) was also isolated, purified
1
and characterized. The significant differences in the H
NMR of 7 and 6 is the appearance of two extra olefinic
singlets at d 5.15 ppm and d 5.05 ppm and only three
methyl singlets (d 1.71, 1.75, and 0.97 ppm) instead of
four. The 13C NMR also shows the appearance of two
shielded olefinic carbons (d 110.6 and 114.9 ppm), one
for C-20 and the other one located at C-17. Similarly to
6, we note the presence of the olefinic H-13/C-13 corre-
lated to the Me-18 protons (d 1.75 ppm) in the HMBC
experiment. This methyl is further correlated to two
other quaternary carbons, C-12 (d 146.1 ppm) and C-11
(d 64.8 ppm). The other deshielded methyl (d 1.71 ppm)
Attempts to deacetylate 8 and convert it to 5 under differ-
ent conditions [NaOH, MeOH; LiOH, MeOH; NaOMe,
.
MeOH; NH2NH2 H2O; ZnBr2; Ba(OH)2 8H2O] failed.
.
Putative mechanisms for the formation of taxanes 5–8
(Schemes 4 and 5)
One of the plausible mechanisms for the formation of
these unusual 1(15!11)abeo-taxanes 6 and 7 is the fol-
lowing. The first step could be activation of the C13-OH
by adding a good leaving group, such as pyrophos-
phate, etc. Upon leaving, it will generate a carbonium
ion, which will cause the formation of the abeo-five-
membered ring. The new carbonium ion is neutralized
by addition of a hydroxyl group or formation of a
double bond C15–C17 (Scheme 4). The reaction with
zinc and acetic acid involves protonation of the car-
bonyl and donation of 2eꢀ from the zinc causing the
shift of the double bond and elimination of the OAc
anion (Scheme 5).
Scheme 4. Putative mechanism for the formation of the
1(15!11)abeo-taxanes 6 and 7.
Conclusion
In this publication, we have shown that very similar
taxanes with C10–C11 double bonds could be obtained
from microbial reactions as well as from conditions with
a reducing agent. This unexpected result is promising
and indeed suggests that a reductase with a similar
Scheme 5. Putative mechanism for the formation of taxane 8.

1990
D.-A. Sun et al. / Bioorg. Med. Chem. 9 (2001) 1985–1992
mechanism as the treatment with zinc might be one
of the enzymes in A. coerula ATCC 10738a. Indeed,
further research in these fields might lead to other
similarities between chemical and microbial reactions,
which could help us understand the mechanisms
involved. This will eventually help us to understand the
selectivity rules of these otherwise empirical reactions.
The unprecedented 1(15!11)abeo-taxanes produced
are presently under investigation in our laboratory to
determine if addition of side chains might lead to
bioactive taxanes.
double-focussing instrument using a xenon beam having
8 kV energy at 1mA equivalent neutral current. Low
resolution mass spectra were obtained in glycerol. Sam-
ples were dissolved in 0.2 mL DMSO before addition of
0.5 mL glycerol. FAB-HR-MS was similarly obtained in
glycerol-DMSO at a resolving power of 12,000.
Substrates: 5ꢀ, 7ꢁ, 9ꢀ, 10ꢁ, 13ꢀ-pentahydroxy-
4(20),11(12)-taxadiene, 1; 5ꢀ, 13ꢀ-dihydroxy-7ꢁ, 9ꢀ,
10ꢁ-triacetoxy-4(20),11(12)-taxadiene, 2; 5ꢀ-hydroxy,
7ꢁ, 9ꢀ, 10ꢁ-triacetoxy-4(20),11(12)-taxadien-13-one, 3;
5ꢀ, 7ꢁ, 9ꢀ, 10ꢁ-tetrahydroxy-4(20),11(12)-taxadien-13-
one, 4 (Scheme 1). 2-Deacetoxytaxinine J is abundant
in various Taxus species.31 Decinnamoylation of 2-dea-
cetoxytaxinine J was accomplished on treatment with
hydroxylamine as reported.26,32 The derived 2-deace-
toxydecinnamoyltaxinine J (244 mg, 0.47 mmol) in a
solution of methanol/water (1:1, 5 mL) was treated with
LiOH.H2O (160 mg; 3.80 mmol). After stirring at room
temperature for 7 h, the reaction was worked up by
dilution with ethyl acetate (40 mL) and washing with
10% NaHCO3, water and brine. After drying, filtering
and evaporating the solvent, the residue was purified by
preparative HPLC. Two compounds were isolated: tax-
ane 5a,7b,9a,10b,13a-pentahydroxy-4(20),11(12)-tax-
Experimental
Instrumentation
Flash chromatography was performed on Silica gel 60
(230–400 mesh EM Science). Thin layer chromato-
graphy was conducted on Silica Gel 60 F254 pre-coated
TLC plates (0.25 mm, EM Science). The compounds
were visualized on TLC plates with 10% sulfuric acid in
ethanol and heating on a hot plate. Na2SO4 was the
drying agent used in all work up procedures unless
otherwise noted. Analytical HPLC was performed on a
Waters 600 FHU delivery system coupled to a PDA 996
detector. Preparative and semi-preparative HPLC were
carried out on a Waters Delta Prep 3000 instrument
coupled to a UV 486 Tunable Absorbance detector set
at 227 nm (Waters, Montreal, Quebec, Canada). Ana-
lytical HPLC was performed with two Whatman partisil
10 ODS-2 analytical columns (4.6ꢃ250 mm) in series.
Semi-preparative HPLC was performed with two
Whatman partisil 10 ODS-2 Mag-9 semi-preparative
columns (9.4ꢃ250 mm) in series. Preparative HPLC was
performed with one partisil 10 ODS-2 MAG-20 pre-
parative column (22ꢃ500 mm). The products were
eluted with a 50 min linear gradient of acetonitrile (25–
100%) in water at a flow rate of 18 mL/min (preparative
HPLC) and 3 mL/min (semi-preparative HPLC). All
the reagents were of the best available commercial
quality and were used without further purification.
adiene
1 (128 mg, 78%) and 5a,13a-dihydroxy-
7b,9a,10b-triacetoxy-4(20),11(12)-taxadiene 2 (16.5 mg,
7%).
1. [a]2D2 ꢀ199.2ꢁ (c 0.07, CHCl3); H NMR (500 MHz,
1
CDCl3) d 5.10 (s, 1H, H-20a), 4.99 (d, J=9.5 Hz, 1H,
H-10), 4.78 (s, 1H, H-20b), 4.40 (dd, J=11.5, 5.0 Hz,
1H, H-7), 4.36 (br. d, J=10.7 Hz, 1H, H-13), 4.32 (t,
J=2.5 Hz, 1H, H-5), 4.11 (d, J=9.5 Hz, 1H, H-9), 3.09
(br.s, 1H, H-3), 2.81 (ddd, J=15.5, 10.5, 8.7 Hz, 1H, H-
14a), 2.07 (br.s, 3H, Me-18), 1.97 (br.ddd, J=2.0, 4.0,
13.5 Hz, 1H, H-6a), 1.68-1.70 (o.m, 2H, H-2), 1.66 (o.m,
1H, H-1), 1.63 (o.m, 1H, H-6b), 1.44 (br.s, 3H, Me-17),
1.18 (dd, J=15.5, 3.6 Hz, 1H, H-14b), 0.97 (br.s, 3H,
Me-16), 0.96 (s, 3H, Me-19); 13C NMR (125 MHz,
CDCl3) d 150.5 (C-4), 138.9 (C-11), 138.8 (C-12), 111.7
(C-20), 80.5 (C-9), 73.7 (C-5), 72.6 (C-10), 70.7 (C-7),
68.4 (C-13), 45.8 (C-8), 39.6 (C-1), 39.6 (C-6), 38.7 (C-
15), 34.9 (C-3), 32.9 (C-14), 32.9 (C-16), 26.7 (C-2), 25.9
(C-17), 17.0 (C-18), 12.1 (C-19); FAB-HR-MS for
C20H32O5K [M+K]+ requires: 391.1887; found:
391.1887.
NMR and mass spectrometry measurement
All the NMR data were obtained at room temperature
on a Bruker Avance-500 spectrometer operating at
500.13 MHz for proton and at 125.77 MHz for carbon-
13. Deuterated chloroform was used as an internal
reference (7.25 ppm for proton and 77.0 ppm for car-
bon-13). The various 2-D spectra were acquired and
processed using standard procedures. For phase-sensi-
tive 2-D experiments (NOESY and HMQC), the data
were acquired using the TPPI phase mode. The NOESY
experiment was obtained using a mixing time of 0.3 s
and a relaxation delay of 1s. The intensity of the cross-
peaks in the NOESY experiment is designated as strong
(s), medium (m) and weak (w). Abbreviations regarding
NMR data are: multiplicity (mult), singlet (s), doublet
(d), triplet (t), quartet (q), doublet of doublet (dd),
broad (br), multiplet (m), overlapping (o). Positive ion
Fast Atom Bombardment Mass Spectra (FAB-MS)
were obtained with a Vacuum Generators ZAB-HS
2. [a]2D2 ꢀ16.4ꢁ (c 0.33, CHCl3); H NMR (500 MHz,
1
CDCl3) d 6.22 (d, J=10.8 Hz, 1H, H-10), 5.81 (d,
J=10.8 Hz, 1H, H-9), 5.67 (dd, J=11.4, 5.1 Hz, 1H, H-
7), 5.16 (s, 1H, H-20a), 4.84 (d, J=1.3 Hz, 1H, H-20b),
4.37 (o.m, 1H, H-5), 4.37 (o.m, 1H, H-13), 1.74 (o.m,
1H, H-1), 1.89 (o.m, 1H, H-2a), 1.79 (o.m, 1H, H-2b),
3.21(d, J=5.4 Hz, 1H, H-3), 2.84 (ddd, J=15.3, 10.3,
8.4 Hz, 1H, H-14a), 2.32 (d, J=1.1 Hz, 3H, Me-18),
2.07 (s, 3H, COCH3), 2.04 (s, 3H, COCH3), 1.99 (s, 3H,
COCH3), 1.95 (o.m, 1H, H-6a), 1.67 (o.m, 1H, H-6b),
1.53 (s, 1H, Me-17), 1.19 (o.m, 1H, H-14b), 0.92 (s, 3H,
Me-16), 0.82 (s, 3H, Me-19); 13C NMR (125 MHz,
CDCl3) d 150.2 (C-4), 142.2 (C-12), 134.8 (C-11), 112.4
(C-20), 76.9 (C-9), 73.4 (C-5), 72.7 (C-10), 70.0 (C-7),
68.5 (C-13), 46.7 (C-8), 39.6 (C-1), 38.8 (C-15), 37.2 (C-6),

D.-A. Sun et al. / Bioorg. Med. Chem. 9 (2001) 1985–1992
1991
35.8 (C-14), 35.7 (C-3), 32.5 (C-16), 26.8 (C-2), 26.1 (C-17),
21.5 (Ac), 21.1(Ac), 20.9 (Ac), 16.7 (C-18), 12.5 (C-19);
FAB-HR-MS for C26H38O8K [M+K]+ requires:
517.2204; found: 517.2206.
to prevent mutation.33 The seed culture was prepared by
the addition of several grains of silica gel, which absor-
bed the fungus, to 30 mL of medium in a 125 mL
Erlenmeyer flask. The culture was incubated for 3 days.
To 1L of potato dextrose broth in a 2 L flask was
added 5 mL of the seed culture previously homogenized
with a Polytron homogenizer. The culture was incu-
bated for 1day after which time it was inoculated with
the substrate 5a, 7b, 9a, 1 b0, 1 a3-pentahydroxy-
4(20),11(12)-taxadiene, 1 (40 mg in 1mL DMSO). The
culture was further incubated for 8 days, then homo-
genized and extracted with CH2Cl2 (400 mLꢃ3) afford-
ing 91mg extract. Cultures with the same medium and
substrate but without fungi were used as controls in the
same experimental conditions.
The
same
starting
material
2-deacetoxy-
decinnamoyltaxinine J (60 mg; 0.12 mmol) in a solution
of methanol/water (1:1, 1.2 mL) was treated with
ꢁ
.
LiOH H2O (36 mg; 0.86 mmol) at 4 C for 75 min. The
reaction mixture was quenched with brine and extracted
with ethyl acetate. The organic phase was washed with
brine, dried with MgSO4, filtered and evaporated. To a
solution of the residue in CH2Cl2 (1.8 mL) and pyridine
(1.8 mL) was added CrO3 (34 mg; 0.34 mmol). The
reaction mixture was stirred at 23 ꢁC for 2 h. The mix-
ture was diluted with brine and extracted with CH2Cl2.
The organic phase was washed with brine, dried with
MgSO4, filtered and evaporated. The residue was pur-
ified by preparative HPLC affording 5a-hydroxy,7b, 9a,
10b-triacetoxy-4(20),11(12)-taxadien-13-one: 3 (30 mg,
55% yield, tR=31.6 min) and recovered starting mate-
rial 2-deacetoxydecinnamoyltaxinine J (9 mg, 15%
yield, tR=36.0 min).
Isolation and purification of the taxanes obtained from
incubation of 5ꢀ, 7ꢁ, 9ꢀ, 10ꢁ, 13ꢀ-pentahydroxy-
4(20),11(12)-taxadiene,
1 with A. coerulea ATCC
10738a. The above CH2Cl2 extract (91mg) was first
applied on a silica gel flash chromatography column (10
mL cylinder, 4 g), eluted with CH2Cl2/acetone (100:0,
20:1, 10:1: 5:1, 1:1, and 0:100 each 8 mL) and MeOH
(10 mL) and fifteen 3 mL fractions were collected.
Fractions 10–13 (37.9 mg) containing the taxanes were
combined, evaporated and further purified by pre-
parative HPLC to obtain the pure compounds: 5 (12
mg, 32%, tR=22.8 min, colorless gum), 6 (17 mg, 42%,
tR=29.1min, colorless gum), 7 (1.5 mg, 4%, tR=36.3
min, colorless gum).
3. Specific rotation, 1H and 13C NMR (500 MHz,
CDCl3) as well as FABHRMS identical to literature.27
To a solution of 5a-hydroxy,7b, 9a, 1 b0-triacetoxy-
4(20),11(12)-taxadien-13-one 3 (71mg; 0.149 mmol) in
methanol (14 mL) at 0 ꢁC, was added 1N NaOH (0.85
mL). The ice bath was removed and the solution was
stirred at 23 ꢁC for 10 h. The mixture was neutralized
with 5% HCl and evaporated to remove the methanol.
The aqueous residue was diluted in brine and extracted
with ethyl acetate. The organic phase was dried
(MgSO4), filtered and evaporated to dryness. The resi-
due was purified by preparative HPLC affording 5a, 7b,
9a, 1 0b-tetrahydroxy-4(20),11(12)-taxadien-13-one 4 (33
mg, 63%, tR=16.1 min).
5ꢀ, 7ꢁ, 9ꢀ-Trihydroxy-4(20),10(11)-taxadien-13-one, 5
(Scheme 2, Table 1). [a]2D2 ꢀ148.8ꢁ (c 0.24, CHCl3); H
1
NMR (500 MHz, CDCl3) and 13C NMR (125 MHz,
CDCl3) see Table 1; FAB-HR-MS for C20H30O4K
[M+K]+ requires: 373.1781; found: 373.1783.
1(15!11)abeo-taxane 6 (Scheme 2, Table 2). [a]D22
ꢁ
1
ꢀ58.5 (c 0.55, CHCl3); H NMR (500 MHz, CDCl3)
and 13C NMR (125 MHz, CDCl3) see Table 2; FAB-
HR-MS for C20H30O4K [M-H2O+K]+ requires:
373.1781; found: 373.1783.
4. [a]2D2 +169.0ꢁ (c 0.10, CHCl3); H NMR (500 MHz,
1
CDCl3) 5.13 (o.s, 1H, H-20a), 5.11 (o.d, 1H, H-10),
4.83 (s, 1H, H-20b), 4.51 (dd, J=11.1, 4.8 Hz, 1H, H-7),
4.32 (t, J=1.9 Hz, 1H, H-5), 4.27 (d, J=9.3 Hz, 1H, H-
9), 3.30 (br.d, J=4.5 Hz, 1H, H-3), 2.88 (dd, J=19.4,
7.4 Hz, 1H, H-14a), 2.12 (s, 3H, H-18), 2.04 (o.m, 1H,
H-2a), 2.00 (o.d, J=19.4 Hz, 1H, H-14b), 1.71 (s, 3H,
H-17), 1.25 (s, 3H, H-16), 1.11 (s, 3H, H-19); 13C NMR
(CDCl3, HMQC data) 200.0 (C-13), 156.3 (C-4), 154.6
(C-11), 135.0 (C-12), 109.0 (C-20), 80.2 (C-9), 73.1 (C-
5), 72.8 (C-10), 71.0 (C-7), 46.0 (C-8), 41.0 (C-1), 40.1
(C-15), 39.6 (C-14), 39.4 (C-16), 36.8 (C-6), 34.6 (C-3),
25.9 (C-2), 25.2 (C-17), 13.4 (C-18), 12.3 (C-19); FAB-
HR-MS for C20H30O5Na [M+Na]+ requires: 373.1991;
found: 373.1992.
1(15!11)abeo-taxane 7 (Scheme 2, Table 3). [a]D22
ꢀ24.0ꢁ (c 0.03, CHCl3); H NMR (500 MHz, CDCl3)
1
and 13C NMR (125 MHz, CDCl3) see Table 3; FAB-
HR-MS: C20H30O4K [M+K]+ required: 373.1781;
found: 373.1783.
Reduction of 5ꢀ-hydroxy,7ꢁ, 9ꢀ, 10ꢁ-triacetoxy-
4(20),11(12)-taxadien-13-one 3 with zinc/acetic acid.
Taxane 3 (55 mg; 0.12 mmol) was dissolved in glacial
acetic acid (0.65 mL) and treated with activated zinc
(0.60 g; 9.18 mmol) at 23 ꢁC for 42 h. The mixture was
diluted with ethyl acetate and filtered through Celite.
Heptane was added and the mixture was evaporated
under reduced pressure. The residue was purified by
semi-preparative HPLC affording a major product 8 (11
mg, 23% yield, tR=37.6 min).
Microorganism and biotransformation procedure
A. coerulea ATCC 10738a was purchased from Amer-
ican Type Culture Collection. Cultures were grown in
potato dextrose broth (24 g/L, DIFCO Laboratories)
and incubated at 25 ꢁC and 125 rpm. A. coerulea ATCC
10738a was first preserved on silica gel and kept at 4 ꢁC
5ꢀ-Hydroxy-7ꢁ,
9ꢀ-diacetoxy-4(20),10(11)-taxadien-
13-one, 8 (Scheme 3, Table 4). [a]2D2 ꢀ123.7ꢁ (c 0.30,
1
CHCl3); H NMR (500 MHz, CDCl3) and 13C NMR

1992
D.-A. Sun et al. / Bioorg. Med. Chem. 9 (2001) 1985–1992
(125 MHz, CDCl3) see Table 4; FAB-HR-MS:
C24H34O6K [M+K]+ required: 457.1993; found:
457.1993.
15. Hu, S.; Tian, X.; Zhu, W.; Fang, Q. Tetrahedron 1996, 52,
8739.
16. Hanson, R. L.; Howell, J. M.; Brzozowski, B. B.; Sullivan,
S. A.; Patel, R. N.; Szarka, L. J. Biotechnol. Appl. Biochem.
1997, 26, 153.
17. Xu, H.; Cheng, K. D.; Fang, W. S.; Zhu, P.; Meng, C.
Chinese Chemical Lett. 1997, 8, 1055.
Acknowledgements
18. Zocher, R.; Weckwerth, W.; Hacker, C.; Kammer, B.;
Hornbogen, T.; Ewald, D. Biochem. Biophys. Res. Commun.
1996, 229, 1 6.
19. Zhang, J. Z.; Zhang, L. H.; Sun, D. A.; Gu, J. Q.; Fang,
Q. C. Chinese Chem. Lett. 1996, 7, 1091.
20. Nanduri, V. K.; Hanson, R. L.; LaPorte, T. L.; Ko, R. Y.;
Patel, R. N.; Szarka, L. J. Biotechnol. Bioeng. 1995, 48, 547.
21. Patel, R. N.; Banerjee, A.; Howell, J. M.; McNamee,
C. G.; Brzozowski, D.; Nanduri, V. K.; Thothathil, J. K.;
Szarka, L. J. Ann. N.Y. Acad. Sci. 1995, 166.
22. Patel, R. N.; Banerjee, A.; Ko, R. Y.; Howell, J. M.; Li,
W.-S.; Comezoglu, F. T.; Partyka, R. A.; Szarka, L. J. Bio-
technol. Appl. Biochem. 1994, 20, 23.
The authors thank the Natural Science and Engineering
Research Council of Canada and the Societe de
Recherche sur le Cancer via operating grants to L.O.Z.
We thank the INRS-Institut Armand-Frappier for a
predoctoral fellowship to D. A. Sun. Dr. Li Zhichao
[Zhongling (Huizhou) High Science and Technology
Co. Ltd, P.R. China] is gratefully acknowledged for a
generous supply gift of a sample of 2-deacetoxytaxinine J.
23. Brieva, R.; Crich, J. Z.; Sih, C. J. J. Org. Chem. 1993, 58,
1068.
References and Notes
24. Hanson, R. L.; Wasylyk, J. M.; Nanduri, V. K.; Cazzu-
lino, D. L.; Patel, R. N.; Szarka, L. J. J. Biol. Chem. 1994, 269,
22145.
1. Whitesides, G. M.; Wong, C. H. Angew. Chem. Int. Ed.
Engl. 1985, 24, 617.
2. Simon, H.; Bader, J.; Gunther, H.; Newmann, S.; Thanos,
J. Angew. Chem., Int. Ed. Engl. 1985, 24, 539.
3. Jones, J. B. Tetrahedron 1986, 42, 3351.
4. Yamada, H.; Shimizu, S. Angew. Chem., Int. Ed. Engl.
1988, 27, 622.
5. Wong, C.-H. Science 1989, 244, 1145.
6. Koskinen, A. M. P.; Klibanov, A. M., Eds. Enzymatic
Reactions in Organic Media; Blackie: London, 1996.
7. Klibanov, A. M. Trends Biotechnol. 1997, 15, 97.
8. Klibanov, A. M.; Samokhin, G. P.; Martinek, K.; Berezin,
I. V. Biotechnol. Bioeng. 2000, 67, 737.
25. Sun, D.A.; Sauriol, F.; Mamer, O.; Zamir L. O. Bioorg.
Med. Chem. 2000, 9, 793.
26. Appendino, G.; Belloro, E.; Jakupovic, S.; Danieli, B.;
Jakupovic, J.; Bombardelli, E. Tetrahedron 1999, 55, 6567.
27. Morita, H.; Gonda, A.; Wei, L.; Yamamura, Y.; Waka-
bayashi, H.; Takeya, K.; Itokawa, H. Phytochemistry 1998, 48,
857.
28. Appendino, G.; Cravotto, G.; Enriu , R.; Gariboldi, P.;
`
Barboni, L.; Torregiani, E.; Gabetta, B.; Zini, G.; Bombar-
delli, E. J. Nat. Prod. 1994, 57, 607.
29. Kant, J.; O’Keeffe, W. S.; Chen, S.-H.; Farina, V.; Fair-
child, C.; Johnston, K.; Kadow, J. F.; Long, B. H.; Vyas, D.
Tetrahedron Lett. 1994, 35, 5543.
9. Hu, S.; Sun, D. A.; Scott, A. I. Tetrahedron Lett. 2000, 41,
1703.
10. Patel, R. N. Ann. Rev. Microbiol. 1998, 98, 361.
11. Lee, D.; Kim, K.-C.; Kim, M.-J. Tetrahedron Lett. 1998,
39, 9039.
30. Kelly, R. C.; Wicnienski, N. A.; Gebhard, I.; Qualls, S. J.;
Han, F.; Dobrowolski, P. J.; Nidy, E. G.; Johnson, R. A. J.
Am. Chem. Soc. 1996, 118, 919.
12. Hu, S.; Sun, D. A.; Tian, X.; Fang, Q. Tetrahedron Lett.
1997, 38, 2721.
13. Hu, S.; Tian, X.; Zhu, W.; Fang, Q. Biocatal. Biotrans-
form. 1997, 14, 241.
14. Hu, S.; Tian, X.; Zhu, W.; Fang, Q. J. Nat. Prod. 1996,
59, 1006.
31. Parmar, V. S.; Jha, A.; Bisht, K. S.; Taneja, P.; Singh,
S. K.; Kumar, A.; Poonam, A.; Jain, R.; Olsen, C. E. Phy-
tochemistry 1999, 50, 1267.
32. Bathini, Y.; Micetich, R. G.; Daneshtalab, M. Synth.
Commun. 1994, 24, 1513.
33. Perkins, D. D. Can. J. Microbiol. 1962, 8, 591.