Application of an amidyl radical cascade to the total synthesis of (±)-fortucine leading to the structural revision of kirkine

Article abstract of DOI:10.1016/j.tet.2009.04.027

A radical cascade initiated by a nitrogen-centred (amidyl) radical was developed, allowing the rapid construction of galanthan frameworks. It was applied to a concise, stereo/regio-selective and tin-free total synthesis of the natural product (±)-fortucin

Full text of DOI:10.1016/j.tet.2009.04.027

Tetrahedron 65 (2009) 6730–6738
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Application of an amidyl radical cascade to the total synthesis
of (ꢀ)-fortucine leading to the structural revision of kirkine
Aurelien Biechy , Shuji Hachisu , Beatrice Quiclet-Sire , Louis Ricard , Samir Z. Zard
a
a
a
b
a,
*
´
´
a Laboratoire de Synthe`se Organique associe´ au CNRS, Ecole Polytechnique, Route de Saclay, Palaiseau, 91128, France
b Laboratoire «He´te´ro´ele´ments et Coordination», Ecole Polytechnique, Route de Saclay, Palaiseau, 91128, France
a r t i c l e i n f o
a b s t r a c t
Article history:
A radical cascade initiated by a nitrogen-centred (amidyl) radical was developed, allowing the rapid
construction of galanthan frameworks. It was applied to a concise, stereo/regio-selective and tin-free
total synthesis of the natural product (ꢀ)-fortucine. This in turn resulted in the correction of the
structure of kirkine.
Received 12 January 2009
Received in revised form 7 April 2009
Accepted 7 April 2009
Available online 16 April 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Dedicated with admiration to Professor
Larry E. Overman
1. Introduction
Ever since the structure of lycorine was first described in 1955,5
the tetracyclic galanthan framework common to lycorine alkaloids
Over the past decade, drug discovery from medicinal plants has
played a large role in furnishing new drugs and drug leads against
various diseases. There are numerous pharmacologically active
compounds that belong to the Amaryllidaceae family, which in-
cludes nearly 500 structurally diverse alkaloids showing significant
biological activity.1
has been the subject of intense research.1 Most lycorine alkaloids
have a trans-B,C-ring junction; however, some compounds have
a cis-B,C-ring junction, notably g-lycorane 1, fortucine (kirkine, 2)
and siculinine 3 (Fig. 1).4c,1
Fortucine 2 was isolated from the Fortune variety of narcissus by
a Russian group in 1988.6a Bastida et al. subsequently isolated kir-
kine from Crinum kirkii, a common grassland plant from East Africa
used as a purgative, rat poison and for the treatment of sores.7 They
assigned to kirkine the same structure as fortucine 2.6b Notwith-
standing that the Spanish study is more recent, it was apparently
conducted without knowledge of the work by the Russian group.
Furthermore, the respective NMR spectra show discrepancies be-
tween the two compounds, which cannot be readily explained,
even when taking into consideration the fact that the NMR spectra
were recorded under different conditions. We decided to undertake
the total synthesis of structure 2 to clarify this anomaly and to
provide sufficient quantities of 2 to explore its biological activity.
We also sought to establish a general strategy to access galanthan
Lycorine alkaloids2 were isolated from the Amaryllidaceae
family. These alkaloids have long been known for their medicinal
virtues3 and show antineoplastic, antimitotic, insect antifeedant,
and antiviral activities. They also inhibit growth and cell division in
higher plants, protein and DNA synthesis in murine cells and in vivo
growth of a murine transplantable ascite tumour. Tests for anti-
fungal and antimicrobial activities demonstrated that bacteria are
resistant to lycorine whereas yeasts are sensitive to it.4
OH
HO
C
H
H
H
O
O
N D
A
H
H
B
OMe
N
2
1
HO
3
HO
OH
galanthan
(-)-lycorine
H
N
H
H
H
N
4
MeO
HO
O
O
O
O
H
H
H
N
γ−lycorane 1
siculinine 3
fortucine 2
* Corresponding author. Tel.: þ33 (0)1 69 33 59 71; fax: þ33 (0)1 69 33 59 72.
E-mail address: zard@dcso.polytechnique.fr (S.Z. Zard).
Figure 1. Lycorine type alkaloids.
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.04.027

A. Biechy et al. / Tetrahedron 65 (2009) 6730–6738
6731
O
H
O
O
HO
H
OTBS
O
O
O
H
H
N
OTBS
OTBS
MeO
HO
MeO
BnO
H
O
H
N
N
2
N
6
4
NPhCS2Me
12
O
O
O
O
1. NaBH4
2. BzCl
O
TFA/H2O
(97%)
O
7
(85%, 2 steps)
O
OTBS
OTBS
MeO
BnO
CH3
O
O
6
N
N
N
8
NPhCS2Me
NPhCS2Me
O
NPhCS2Me
O
5
H
O
OTBS
OTBS
SMe
H
O
Scheme 1. Retrosynthetic analysis of fortucine 2.
(51%)
DLP
N
N
N
11
O
O
frameworks with a cis-B,C-ring junction. In the present article we
give a full account of our work.8
10
Ph
PhCl, reflux
C11H23
S
DLP
-H+
Our synthetic approach to structure 2 (henceforth fortucine) is
based on a radical cascade that is initiated by the generation of an
amidyl radical that undergoes cyclisation on to the neighbouring
alkene followed by an oxidative cyclisation on to the aromatic ring
(Scheme 1). This cascade is designed to form rings B and D in one step.
Moreover, we expected the cyclisations to be stereoselective leading
to the requisite cis-B,C-ring junction, as well as regioselective to give
para (as opposed to ortho) cyclisation with respect to the benzyloxy
group. Precursor 6 is constructed by the union of p-benzoquinone
ethylene-monoketal 79 with the azaenolate of hydrazone 8.10
Hydrazone 6 can be a useful building block for the formation of
lycorine type alkaloids, through reduction and benzoylation of the
ensuing hydrazide to give key intermediates such as 5.
O
O
O
OTBS
OTBS
N
N
O
O
Scheme 3. Synthesis of pentacycle 12.
treated with benzoyl chloride to yield radical precursor 10. Pleas-
ingly, treatment of radical precursor 10 in refluxing chlorobenzene
with lauroyl peroxide (DLP, 2.0 equiv) effected the desired radical
cascade to give pentacycle 11 in 51% yield. Even though we have
studied this type of cyclisation before,12 we were delighted (but not
surprised) to see that the cascade process had occurred stereo-
selectively to give the all-cis cyclised product. The protected-ketone
was then unmasked by treatment with TFA in water to give enone
12 in 97% yield (Scheme 3).
At this point, we examined the reductive elimination of the al-
lylic silyl ether and the concomitant migration of the double bond.
Unfortunately, treatment of 12 with DIBAL, Red-Al, or LiAlH4 only
gave an amino alcohol (Scheme 4).
We then attempted treating 12 with zinc in AcOH, NiCl2/NaBH4,
SmI2 and Ca/NH3, but these reactions led to the decomposition of
12 or reduction of the enone double bond without the elimination
of the silyl ether, which would have led to the installation of the
double bond in the desired D3,4 location.
Notwithstanding the failure to induce reductive elimination of
the silyl ether, we decided to explore Wittig, Peterson and Julia
chemistry to install the requisite D3,4 double bond. To this end, we
attempted the 1,4-addition of triphenylphosphine and tri-
organosilyl cuprates, but without success. By contrast, the addition
of p-methylthiophenol onto 12 gave thioether 13 in a good yield.
Oxidation of sulfide 13 to the sulfone 14 by m-CPBA was possible,
but the reaction was not very clean. Moreover, subsequent re-
duction with LiAlH4, DIBAL or BH3$THF was not successful and we
2. Results and discussion
We began the synthesis by trapping the azaenolate derived from
hydrazone 8 with p-monoketal 7 to provide alcohol 9 in a quanti-
tative yield. Such a satisfying result was not expected, since neither
1,4-addition onto the enone nor intramolecular cyclisation of the
azaenolate onto the thiocarbonyl group were observed. The azae-
nolate of 8 was stable at ꢁ78 ꢂC, and it was indeed possible to trap it
with a variety of a,b-unsaturated ketones. These coupled products
can in principle be used to construct dihydropyrroles via iminyl
radicals,11 and galanthan frameworks via amidyl radicals12 through
cyclisations onto the neighbouring alkenes (Scheme 2).
At this point we attempted the direct transformation of alcohol
9 into the radical precursor 5 via reduction and reaction with an
appropriate aroyl chloride, but the free di-allylic tertiary alcohol
was too unstable to undergo the benzoylation reaction. We there-
fore decided to protect it as a silyl ether, and attempted to prepare
silyl ether 6 in the same pot by quenching the coupling reaction of
hydrazone 8 with monoketal 7 with TBSOTf as opposed to the ad-
dition of aqueous NH4Cl. In the event, it proved necessary to isolate
and purify alcohol 9 and to react it with TBSOTf in a separate step to
afford silyl ether 6 in an acceptable yield.
2.1. Model study
HO
H
A model study was first carried out in which ring A was replaced
by a phenyl ring. Hydrazone 6 was reduced with NaBH4 and then
O
H
H
N
O
O
O
O
OTBS
O
O
O
H
O
7
OH
OTBS
HO
LDA
TBSOTf
(quant.)
N
DIBAL or
Red-Al or LiAlH4
H
N
CH3
NPhCS2Me
OTBS
N
N
12
(quant.)
N
9
6
8
H
NPhCS2Me
NPhCS2Me
Scheme 2. Synthesis of silyl ether 6.
Scheme 4. Reduction with aluminium based reagents.

6732
A. Biechy et al. / Tetrahedron 65 (2009) 6730–6738
O
O
O
H
N
HO
STol
OTBS
O
H
H
N
OTBS
TFA/H2O
OTBS
MeO
BnO
H
N
MeO
BnO
OTBS
oxydation
HO
H
O
H
O
H
O
N
H
(88%)
Ts
12
15
4
H
N
16
OTBS
LiAlH4
TolSH, Et3N
H
TBAF
HClaq
Zn, AcOH
Zn, AcOH
O
STol
O
Ts
H
N
OTBS
mCPBA
H
N
OTBS
O
O
H
O
H
O
H
reduction
H
OH
MeO
BnO
MeO
BnO
Zn, AcOH
13
H
O
14
H
N
N
Scheme 5. Routes to model sulfone.
O
Scheme 7. Attempts at direct reductive elimination.
observed the b-elimination of the sulfone moiety. We also tried to
reduce both amide and ketone prior to the oxidation of sulphide 15,
but neither m-CPBA in the presence of TFA nor treatment with
AcOH/AcO2H gave the desired product (Scheme 5).
alcohol was converted into oxalate 19 and treated with tributyltin
hydride and a small amount of AIBN. We hoped that the in-
termediate tertiary allylic radical in this variation of the Barton-
McCombie deoxygenation13 would react on C-4, which appeared to
be the less hindered position. Unfortunately, only the unwanted
deoxygenated isomer 19a was observed (Scheme 8).
2.2. Total synthesis of ( )-fortucine
At this point, we decided to pursue more persistently the ad-
dition of thiols, which was explored in the model study. Addition of
p-methylthiophenol to 16 gave thioether 20 as a mixture of di-
astereoisomers. It was now necessary to oxidise the sulphide and
reduce both amide and ketone. The previous results from the model
study led us to believe that the reduction should be carried out
prior to the oxidation. Thioether 20 was treated with LiAlH4 to
reduce both the amide and the ketone to the tertiary amine and
alcohol, respectively. After many attempts to oxidise the sulphide
group, we found that dimethyldioxirane in the presence of TFA to
protect the tertiary amine was the most satisfactory. Having in hand
sulfone 21, we tried to provoke direct elimination of TBSO– and
sulfone groups, but the phenol had to be unmasked by hydro-
genolysis over Pd/C (22) before processing to the Julia-type olefi-
nation using 6% Na/Hg amalgam to finally give fortucine (2).
However, the fact that addition of p-methylthiophenol to enone 16
was not diastereoselective complicated considerably the in-
terpretation for the NMR spectra. The NMR spectra of our first batch
of compound 2, recorded with a very small amount of product, did
not match with those described in Bastida’s paper6b but matched
the Russian spectral data (Scheme 9).6a
Despite these setbacks in the model studies, we decided to
concentrate our efforts on the real system by introducing the ap-
propriate ring A of fortucine 2 via the acylation of the intermediate
hydrazide with 3-benzyloxy-4-methoxybenzoyl chloride (Bz0Cl,
Scheme 6) to give amidyl radical precursor 5 in 81% from hydrazone
6. Treatment of this substance in refluxing 1,2-dichloroethane with
lauroyl peroxide (DLP, 1.4 equiv) effected the radical cascade to give
pentacycle 4 in 60% yield. We were pleased to note that the yield of
this reaction increased with scale (optimal at 1–2 mmol).10 Grati-
fyingly, pentacycle 4 was obtained with complete stereoselectivity
to afford the cis ring junction. Oxidative cyclisation on to the aro-
matic ring occurred regioselectively in a 14:1 ratio (para/ortho with
respect to the benzyloxy group), the steric clash between the ethy-
lene ketal and benzyloxy functionality being sufficient in dis-
favouring the undesired ortho cyclisation.
The formation of densely functionalized pentacycle 4 in five
steps is remarkably efficient. The structural diversity of lycorine
type alkaloids generally originates from functional groups on rings
A and C. The strategic positioning of the alkene and the protected-
ketone in 7 as well as the choice of the aryl chloride in the for-
mation of 5 can potentially be used to target other complex
members of the lycorine alkaloids (see Section 2.4).
In order to better clarify the NMR of our intermediates, we de-
cided to replace p-methylthiophenol by the more bulky 2-
methoxythiophenol and obtained thioether 23 from enone 16 in
The remaining task was to carry out a reductive cleavage of the
allylic silyl ether with the migration of the double bond (D2,3
/D3,4),
and the installation of an axial alcohol at C-1. Treatment of penta-
cycle 4 with TFA/H2O gave enone 16 in 88% yield. As anticipated
from the model study, all our efforts to induce direct reductive
elimination of the TBSO-group failed (Scheme 7).
First, the silyl ether in pentacycle 4 was deprotected with TBAF
to give tertiary alcohol 17, which was then converted into a mesy-
late group and the ketone unmasked with TFA. Compound 18 thus
obtained was subjected to zinc in acetic acid but this furnished the
undesired conjugated ketone 18a. In another approach, the tertiary
O
O
O
H
N
H
N
1. MsCl, Et3N
2. TFA/H2O
OMs
OH
MeO
BnO
MeO
BnO
TBAF
H
O
H
O
4
18
17
1. (COCl)2
2. MeOH
Zn, AcOH
O
R1
O
O
H
O
O
CO2Me
Bu3SnH, AIBN
H
N
H
MeO
BnO
MeO
BnO
O
O
H
O
H
OTBS
O
N
19
H
O
OTBS
MeO
BnO
MeO
BnO
1. NaBH4
2. Bz'Cl
DLP
6
H
O
N
N
(60%)
R1 = O: 18a
1 = -OCH2CH2O-: 19a
NPhCS2Me
5
O
R
(81%, 2 steps)
4
Scheme 6. Synthesis of pentacycle 4.
Scheme 8. Replacement of TBSO– with leaving groups.

A. Biechy et al. / Tetrahedron 65 (2009) 6730–6738
6733
O
STol
HO
SO2Tol
H
N
H
OTBS
OTBS
MeO
BnO
MeO
RO
1. LiAlH4
H
O
H
Na/Hg 6%
N
2. DMDO, TFA
20
21
R = Bn
22
R = H
TolSH, Et3N
H2, Pd/C
(quant.)
20 °C
0 °C
16
2
Ar =
SAr
ArSH, Et3N
(80%)
O
OMe
HO
H
SO2Ar
OTBS
H
OTBS
MeO
BnO
MeO
1. LiAlH4 (66%)
Na/Hg 6%
(50%)
H
O
H
N
N
2. DMDO, TFA
(92%)
RO
24
R = Bn
H-6
H-12
-20 °C
6.0
H-4a
H-10
25
23
H2, Pd/C
(quant.)
R = H
7.0
ppm (t1)
5.0
4.0
3.0
Scheme 9. Synthesis of fortucine 2.
Figure 3. 1H NMR spectra of 2 at low temperatures.
2
The NMR spectrum of our product was consistent with the data
of the Russian group6a but not with those of Bastida et al.6b (Table
1). An authentic sample of kirkine, kindly supplied by Prof. Bastida,
proved different from our synthetic material (by TLC and NMR).
Prominent differences observed involved protons in the cis-junc-
tion (4a and 10b), those next to the nitrogen (6 and 12) and in the
vicinity of the double bond (3 and 11).
1
HO
3
H
10
10b
MeO
HO
11
12
4a
H
N
7
6
Figure 2. X-ray crystallographic structure of fortucine 2.
Interestingly, protons 10 (in the vicinity of the alcohol), 4a, 6 and
12 (adjacent to the nitrogen) of the 1H NMR spectrum in our study
were ill resolved, apparently due to slow conformational changes in
ring C as indicated by molecular modelling. A better resolution was
obtained by recording the spectra at ꢁ27 ꢂC, at which point the
compound was essentially frozen in its most stable conformation
(Fig. 3).
It appeared that kirkine and fortucine were either di-
astereoisomers or positional isomers of the double bond. In order
to distinguish between these two possibilities, we subjected our
synthetic material and authentic kirkine to catalytic hydrogena-
tion and obtained the same compound 27 by NMR, TLC and HRMS
(the only difference being that the synthetic sample is racemic).
We therefore concluded that kirkine must have structure 26,
which contains a double bond in the same location as siculinine
3.6c Structure 26 is the only isomer compatible with the NMR
spectra and which would lead to 27 upon catalytic hydrogenation
(Scheme 10).
80% yield as one diastereoisomer. As expected, the reduction of
ketone with LiAlH4 occurred stereoselectively from the convex face
to deliver exclusively the axial alcohol in 66% yield. Mechanical loss
in the isolation of the polar amino alcohol is presumably the cause
for the relatively modest yield. Oxidation of the sulfide group with
DMDO/TFA furnished sulfone 24 in 92% yield. Hydrogenolysis over
Pd/C unmasked quantitatively phenol 25, which was subjected to
Julia-type olefination using 6% Na/Hg amalgam to give finally the
target molecule fortucine 2 in 50% yield. Product stemming from
desulfonylation without elimination of the TBSO– group was also
observed in the crude mixture, explaining in part the moderate
yield in the last step.
2.3. Revision of the structure of kirkine
The structure of our synthetic (ꢀ)-fortucine was unambiguously
confirmed by X-ray crystallography (Fig. 2).
2.4. Hydrazone 6 as a building block for the synthesis of
pyrrolophenanthridine alkaloids
Table 1
Differences between 1H NMR spectra of fortucine/kirkine
The pyrrolo[3,2,1-de]phenanthridine ring system 28 constitutes
the core structural framework of the pyrrolophenanthridine alka-
loids.14 Oxoassoanine 29, assoanine 30, hippadine 31 and pratosine
32 are members of the pyrrolophenanthridine class of compounds
Authentic fortucinea,6a Fortucine 2b Protons Fortucine 2c Authentic kirkined,6b
4.25
2.30–2.60
5.52
4.30
1
4.27
4.37
2.34–2.46
5.87
2.40–2.50
5.55
2
3
2.31
5.47
2.68
3.27–3.87
6.53
2.71
3.30–3.90
6.57
4a
6
7
3.24
3.45–3.66
6.53
4.24
4.40–4.61
6.60
HO
HO
H2, Pd/C
HO
H
N
H
N
6.79
2.95
2.30–2.60
2.30–2.60
3.83
6.80
3.00
2.30–2.50
2.30–3.30
3.85
10
10b
11
12
OMe
7.18
3.19
2.40–2.67
2.40–3.00
3.81
7.08
3.52
2.81–2.87
3.71–3.83
3.87
MeO
HO
MeO
HO
H
H
H
N
H
2
26
MeO
HO
H
a
Recorded on 360 MHz in CDCl3.
b
c
Recorded on 400 MHz in CDCl3 at ꢁ27 ꢂC.
Recorded on 400 MHz in CD3OD at ꢁ20 ꢂC.
Recorded on 500 MHz in CD3OD.
27
d
Scheme 10. Hydrogenation of synthetic fortucine 2 and authentic kirkine.

6734
A. Biechy et al. / Tetrahedron 65 (2009) 6730–6738
3. Conclusion
C
N
MeO
MeO
In summary, we have implemented a short first total synthesis
D
A
B
N
of (ꢀ)-fortucine 2 (12 steps, 9% overall yield) and corrected the
structure of kirkine, which should now be revised to 26. The
strategy is noteworthy for its stereo/regio-selective radical cascade
and the multifaceted use of hydrazone 8. We also have explored
briefly the possibility of reaching pyrrolophenanthridine alkaloids
by a radical cascade.
R
28
R = O: Oxoassoanine (29)
R = H2: Assoanine (30)
MeO
O
O
N
4. Experimental
4.1. General
N
MeO
O
O
Pratosine (32)
Hippadine (31)
All reactions requiring anhydrous conditions were conducted
under an inert atmosphere. Glassware, syringes and needles were
dried at 120 ꢂC and allowed to either cool in a desiccator over CaCl2,
or under a positive pressure of Ar before use. THF was distilled from
benzophenone ketyl; CH2Cl2, MeOH, and NEt3 from CaH2. Reactions
were monitored by TLC, using plates pre-coated with a 0.25 mm
layer of silica containing a fluorescent indicator. Visualisation of
reaction components was achieved with 254 nm light, and with
anisaldehyde, vanillin or Dragendorff reagent. Organic layers were
dried using MgSO4. Column chromatography was carried out on
Figure 4. Pyrrolophenanthridine alkaloids.
O
O
O
O
O
O
H
N
OTBS
DLP
OTBS
OTBS
ArCOCl
R
R
H
O
N
N
NPhCS2Me
I
NPhCS2Me
6
II
O
Kieselgel 60 (40–63 mm). Petroleum ether refers to the fraction of
petroleum boiling between 30 ꢂC and 40 ꢂC. IR spectra were
recorded as thin films. Melting points were recorded on a Kofler hot
block, and are uncorrected. 1H and 13C NMR spectra of compounds
were recorded in CDCl3 at 25 ꢂC. Chemical shifts are reported rel-
DDQ
LiAlH4
R
R
R
N
N
N
IV
III
V
O
O
ative to CDCl3 (dH 7.26, dC (central line of t) 77.23). Coupling con-
Scheme 11. Strategic approach to Pyrrolophenanthridine alkaloids.
stants (J) are given in hertz, multiplicities are given as multiplet (m),
singlet (s), doublet (d), triplet (t), quartet (q), or broad (br). COSY,
HSQC, NOESY and DEPT were use to aid spectral assignments.
(Fig. 4). They are isolated from various species of Amaryllidaceae15
and present potentially useful pharmacological activities.16
Hydrazone 6 could in principle be utilised as a synthetic build-
ing block to access this class of compounds. Aromatic ring A can be
introduced via acylation of the hydrazide with an appropriate aryl
chloride to get radical precursor I. The radical cascade should
provide tetracyclic structure II as previously demonstrated and
pyrrolophenanthridine alkaloids III can then be obtained by aro-
matisation of ring C. Related compounds could also be obtained by
further reduction (IV) or oxidation (V) (Scheme 11).
In a preliminary study, and in order to define the conditions for
the aromatisation of ring C, we decided to use p-tolyl chloride for
the formation of amidyl radical precursor 33. The ketone and ter-
tiary alcohol in 34 were unmasked to give 35. Reduction with
NaBH4 and subsequent treatment with APTS in refluxing toluene
yielded 36 with the desired pyrrolophenanthridine structure
(Scheme 12).
4.2. Hydrazone (8)
Acetaldehyde (3.80 mL, 68.0 mmol) was added to a solution of 1-
(methylthio)thiocarbonyl-1-phenyl-hydrazine17 (795 mg, 4.01 mmol)
in EtOH (15 mL). After 2 h 30 min, the mixture was concen-
trated in vacuo, and purified by flash column chromatography
(gradient elution: 0–20% Et2O/petroleum ether) to give hydra-
zone 8 (881 mg, 98%) as a brown crystalline solid. Rf¼0.3 (Et2O/
petroleum ether; 10%); mp 37–40 ꢂC; nmax/cmꢁ1 2913, 2359,
1631, 1594, 1486, 1399, 1359, 1320, 1219, 1167, 1134, 1069, 1036,
959, 929; 1H NMR (400 MHz, CDCl3) 7.57–7.52 (2H, m, 2CHAr),
7.50–7.45 (1H, m, CHAr), 7.18–7.15 (2H, m, 2CHAr), 6.79 (1H, q,
J¼5.4 Hz, NCH), 2.62 (3H, s, SCH3), 2.01 (3H, d, J¼5.4 Hz,
NCHCH3); 13C NMR (100 MHz, CDCl3) 202.0 (CS2), 145.8 (NCH),
139.6 (Cq), 130.5 (2CHAr), 129.6 (CHAr), 129.2 (2CHAr), 19.1
(CS2CH3), 18.8 (NCHCH3); m/z (CI) 225 (MþHþ, 100%) Found: M,
224.0445. C10H12N2S2 requires M, 224.0442.
O
O
O
H
N
O
OTBS
4.3. Alcohol (9)
OTBS
DLP
Me
Me
Cl
1. NaBH4
Me
H
O
6
N
n-BuLi (3.54 mL, 1.6 M in THF, 5.67 mmol) was added dropwise
to a stirred solution of diisopropylamine (0.88 mL, 6.24 mmol) in
THF (8 mL) at ꢁ78 ꢂC. After 15 min, a solution of hydrazone 8
(1.27 g, 5.67 mmol) in THF (15 mL) was added dropwise to the re-
action mixture and stirred for 1 h. After that time, a solution of p-
benzoquinone ethylene-monoketal (7)9 (288 mg, 1.89 mmol) in
THF (6 mL) was added dropwise. After 15 min, saturated aqueous
solution of NH4Cl (30 mL) was added to the reaction mixture and
allowed to warm to 20 ꢂC. The aqueous layer was extracted with
Et2O (3ꢃ50 mL), and the combined organic extracts were washed
with brine (150 mL). The organic layer was dried, concentrated in
vacuo, and purified by flash column chromatography (gradient
(55%)
34
NPhCS2Me
33
2.
O
O
1. TFA/H2O
2. TBAF
(60% 2 steps)
(85%, 2 steps)
O
H
OH
Me
Me
1. NaBH4/MeOH
H
N
N
2. APTS, toluene, reflux
(35% 2 steps)
35
36
O
O
Scheme 12. Model study towards pyrrolophenanthridine.

A. Biechy et al. / Tetrahedron 65 (2009) 6730–6738
6735
elution: 0–100% Et2O/petroleum ether) to give recovered excess
hydrazone 8 (807 mg, 95%) and alcohol 9 (708 mg, 99%) as a white
crystalline solid. Rf¼0.6 (Et2O); mp 134–136 ꢂC; nmax/cmꢁ1 3398,
2885, 2361, 1488, 1390, 1324, 1218, 1116, 1069, 1013, 965; 1H NMR
(400 MHz, CDCl3) 7.53–7.49 (2H, m, 2CHAr), 7.46–7.42 (1H, m,
CHAr), 7.16–7.13 (2H, m, 2CHAr), 6.60 (1H, t, J¼5.7 Hz, NCH), 5.95
(2H, d, J¼10.5 Hz, 2CH]CH), 5.73 (2H, d, J¼10.5 Hz, 2CH]CH), 3.99
(4H, s, 2OCH2), 3.10 (1H, s, OH), 2.59 (2H, d, J¼5.7 Hz, NCHCH2), 2.59
(3H, s, SCH3); 13C NMR (100 MHz, CDCl3) 202.3 (CS2), 144.5 (NCH),
139.0 (Cq), 135.4 (2CH]CH), 130.3 (2CHAr), 129.5 (CHAr), 129.1
(2CHAr), 127.8 (2CH]CH), 98.7 (OCO), 67.1 (COH), 65.3 (OCH2), 65.1
(OCH2), 42.9 (NCHCH3), 19.1 (SCH3); m/z (CI) 377 (MþHþ, 20%), 225
(100%) Found: M, 376.0918. C18H20O3N2S2 requires M, 376.0916.
flash column chromatography (gradient elution: 0–5% EtOAc/tolu-
ene) to give radical precursor 5 (1.21 g, 81% over two steps) as
a yellow foam; Rf¼0.4 (EtOAc/toluene; 8%).
Radical precursor 5 (951 mg, 1.30 mmol) was dissolved in 1,2-
dichloroethane (10.3 mL) and lauroyl peroxide (DLP, 100 mg por-
tions, 700 mg, 140%) was added into the reaction mixture at 1 h
intervals for 7 h. Immediately after the first addition of DLP, the
reaction mixture was heated to reflux for 7 h. After cooling, the
resulting solution was concentrated in vacuo and directly purified
by column chromatography (gradient elution: 0–100% EtOAc/pe-
troleum ether) to give pentacycle 4 (427 mg, 60%) as a colourless
foam. Rf¼0.4 (EtOAc); nmax/cmꢁ1 2928, 2856, 2360, 1654, 1602,
1508, 1450, 1388, 1350, 1293, 1261, 1218, 1129, 1051, 1009, 950; 1H
NMR (400 MHz, CDCl3) 7.67 (1H, s, H-7), 7.48–7.46 (2H, m, 2CHAr),
7.38–7.34 (2H, m, 2CHAr), 7.31–7.27 (1H, m, CHAr), 6.89 (1H, s, H-
10), 5.82 (1H, dd, J¼10.2, 1.7 Hz, H-3), 5.74 (1H, d, J¼10.2 Hz, H-2),
5.24 (1H, d, J¼11.9 Hz, OCH2Ph), 5.17 (1H, d, J¼11.9 Hz, OCH2Ph),
4.00 (1H, dd, J¼3.9, 1.5 Hz, H-4a), 3.92 (3H, s, OCH3), 3.91–3.88 (1H,
m, H-12), 3.75–3.62 (3H, s, OCH2CH2O), 3.31 (1H, d, J¼3.9 Hz, H-
4.4. Silyl ether (6)
2,6-Lutidine (6.36 mL, 54.6 mmol) and TBSOTf (6.02 mL,
26.2 mmol) were added to a stirred solution of alcohol 9 (4.11 g,
10.9 mmol) in CH2Cl2 (70 mL). After 1 h 15 min, H2O (100 mL) was
added to the reaction mixture, and the aqueous layer was extracted
with Et2O (3 ꢃ200 mL). The combined organic extracts were se-
quentially washed with 0.1 M aqueous HCl (3ꢃ400 mL), saturated
aqueous solution of NaHCO3 (2ꢃ400 mL) and brine (400 mL). The
organic layer was dried, concentrated in vacuo and purified by flash
column chromatography (20% Et2O/petroleum ether) to give silyl
ether 6 (5.36 g, 100%) as a white crystalline solid. Rf¼0.2 (Et2O/
petroleum ether; 20%); mp 84–86 ꢂC; nmax/cmꢁ1 3036, 2954, 2927,
2883, 2855, 2360, 1593, 1488, 1471, 1408, 1388, 1325, 1253, 1220,
1116,1070,1014, 969, 946; 1H NMR (400 MHz, CDCl3) 7.54–7.50 (2H,
m, 2CHAr), 7.46–7.42 (1H, m, CHAr), 7.17–7.15 (2H, m, 2CHAr), 6.78
(1H, t, J¼5.9 Hz, NCH), 5.88 (2H, d, J¼10.2 Hz, 2CH]CH), 5.75 (2H, d,
J¼10.2 Hz, 2CH]CH), 4.03 (2H, d, J¼3.2 Hz, OCH2), 4.02 (2H, d,
J¼3.2 Hz, OCH2), 2.60 (3H, s, SCH3), 2.55 (2H, d, J¼5.9 Hz, NCHCH2),
0.70 (9H, s, SiC(CH3)3), ꢁ0.04 (6H, s, Si(CH3)2); 13C NMR (100 MHz,
CDCl3) 202.4 (CS2), 145.9 (NCH), 139.4 (Cq), 136.1 (2CH]CH), 130.4
(2CHAr), 129.5 (CHAr), 129.1 (2CHAr), 127.1 (2CH]CH), 98.9 (OCO),
69.2 (Cq), 65.5 (OCH2), 65.0 (OCH2), 45.8 (NCHCH2), 25.7 (3SiCCH3),
19.1 (SCH3), 18.0 (SiC(CH3)3), ꢁ2.8 (2SiCH3); m/z (CI) 491 (MþHþ,
100%), 355 (20%), 238 (10%) Found: M, 490.1769. C24H34O3N2S2Si
requires M, 490.1780.
10b), 3.25 (1H, dt, J¼11.8, 7.1 Hz, H-12), 2.44–2.39 (1H, m, OCH2
-
CH2O), 2.33–2.15 (1H, m, H-11), 2.15–2.08 (1H, m, H-11), 0.86 (9H, s,
3SiCCH3), 0.15 (3H, s, SiCH3), 0.15 (3H, s, SiCH3); 13C NMR (100 MHz,
CDCl3) 163.6 (C-6), 151.5 (Cq), 147.8 (Cq), 136.8 (Cq), 131.5 (C-3),
130.6 (C-2), 130.0 (Cq), 128.7 (2CHAr), 128.1 (CHAr), 127.8 (2CHAr),
125.5 (Cq), 112.1 (C-7), 112.1 (C-10), 103.6 (OCO), 76.4 (Cq), 71.0
(OCH2Ph), 66.3 (OCH2CH2O), 64.6 (OCH2CH2O), 63.7 (C-4a), 56.4
(OCH3), 42.2 (C-12), 41.8 (C-10b), 38.4 (C-11), 25.8 (3SiCCH3), 18.1
(SiC(CH3)3), 2.1 (SiCH3), 2.1 (SiCH3); m/z (CI) 550 (MþHþ, 100%), 285
(15%) Found: M, 549.2536. C31H39O6NSi requires M, 549.2547.
4.6. Ketone (16)
A mixture of H2O (15 mL) and TFA (15 mL) was added to a stirred
solution of pentacycle 4 (427 mg, 0.78 mmol) in THF (3.5 mL). After
1 h, the reaction mixture was diluted with EtOAc (100 mL), the or-
ganic layer was washed with H2O (50 mL), cautiously basified by
washing with saturated aqueous solution of Na2CO3 (3ꢃ100 mL),
washed with brine (200 mL), dried, concentrated in vacuo and pu-
rified by flash column chromatography (gradient elution: 0–50%
EtOAc/petroleum ether) to give ketone 16 (346 mg, 88%) as a col-
ourless foam. Rf¼0.5 (EtOAc/petroleum ether; 50%); nmax/cmꢁ1
2928,1684,1651,1601,1511,1149,1385,1352,1293,1255,1221,1136;
1H NMR (400 MHz, CDCl3) 7.61 (1H, s, H-7), 7.48–7.45 (2H, m,
2CHAr), 7.38–7.34 (2H, m, 2CHAr), 7.31–7.27 (1H, m, CHAr), 6.88 (1H,
s, H-10), 6.72 (1H, dd, J¼10.2, 2.2 Hz, H-3), 6.15 (1H, d, J¼10.2 Hz, H-
2), 5.21 (1H, d, J¼12.1 Hz, OCH2Ph), 5.17 (1H, d, J¼12.1 Hz, OCH2Ph),
4.27 (1H, dd, J¼4.3, 2.1 Hz, H-4a), 3.97 (3H, s, OCH3), 3.96–3.90 (1H,
m, H-12), 3.70 (1H, d, J¼4.3 Hz, H-10b), 3.29–3.21 (1H, m, H-12),
2.27–2.22 (2H, m, H-11), 0.90 (9H, s, 3SiCCH3), 0.20 (3H, s, SiCH3),
0.18 (3H, s, SiCH3); 13C NMR (100 MHz, CDCl3) 194.7 (C-1), 163.0 (C-
6), 152.3 (Cq), 148.0 (Cq), 146.8 (C-3), 136.7 (Cq), 130.1 (C-2), 129.0
(Cq),128.6 (2CHAr),128.0 (CHAr),127.6 (2CHAr),121.8 (Cq),113.0 (C-
10), 112.0 (C-7), 75.6 (Cq), 70.9 (OCH2Ph), 63.9 (C-4a), 56.2 (OCH3),
45.2 (C-10b), 42.1 (C-12), 37.5 (C-11), 25.6 (3SiCCH3), 18.0
(SiC(CH3)3), ꢁ2.1 (SiCH3), ꢁ2.2 (SiCH3); m/z (CI) 506 (MþHþ, 100%)
Found: M, 505.2282. C29H35O5NSi requires M, 505.2285.
4.5. Pentacycle (4)
NaBH4 (2.31 g, 61.13 mmol) was added portion wise over 1 h to
a stirred solution of silyl ether 6 (1.00 g, 2.04 mmol) in MeOH
(20 mL) and THF (20 mL) at 0 ꢂC. The reaction mixture was heated
to reflux for 30 min. After cooling, H2O (30 mL) was added and the
mixture was concentrated in vacuo to remove MeOH. The aqueous
layer was then extracted with Et2O (2ꢃ50 mL) and the combined
organic extracts were washed with brine (100 mL), dried and
concentrated in vacuo to give the hydrazide as yellow oil.
Oxalyl chloride (2.85 mL, 32.61 mmol) was added to a stirred
mixture of 3-benzyloxy-4-methoxybenzoic acid (2.14 g, 8.15 mmol)
and DMF (three drops) in CH2Cl2 (25 mL). After 1 h 30 min, the
reaction mixture was concentrated in vacuo to give 3-benzyloxy-4-
methoxybenzoyl chloride as a yellow crystalline solid, which was
used in the next step without further purification.
The above 3-benzyloxy-4-methoxybenzoyl chloride was added
portion wise to a stirred mixture of the above hydrazide, DMAP
(50 mg, 0.41 mmol) and Et3N (3.39 mL, 24.46 mmol) in CH2Cl2
(20 mL). After 24 h, H2O (50 mL) and saturated aqueous solution of
NaHCO3 (50 mL) were added to the reaction mixture, and the
aqueous layer was extracted with in CH2Cl2 (2ꢃ50 mL). The organic
layer was dried, concentrated in vacuo and the crude material was
triturated with ether. The mixture was filtered to remove acid an-
hydride, the filtrate was then concentrated in vacuo and purified by
4.7. Thioether (23)
Ketone 16 (150 mg, 0.30 mmol) and o-methoxythiophenol
(0.90 mL, 7.41 mmol) were dissolved in THF (5.9 mL) and Et3N
(1.05 mL, 7.41 mmol) was added. After 1 h 15 min, the mixture was
concentrated in vacuo, and purified by flash column chromatog-
raphy (gradient elution: 0–40% EtOAc/petroleum ether) to give
thioether 23 (153 mg, 80%) as a colourless foam. Rf¼0.5 (EtOAc/
petroleum ether; 40%); nmax/cmꢁ1 2952, 2932, 2858, 2360, 2336,

6736
A. Biechy et al. / Tetrahedron 65 (2009) 6730–6738
1728, 1659, 1601, 1513, 1451, 1386, 1350, 1255, 1219, 1132, 1108, 1027,
918; 1H NMR (400 MHz, CDCl3) 7.67 (1H, s, H-7), 7.47–7.46 (2H, m,
2CHAr), 7.41–7.33 (3H, m, 3CHAr), 7.33–7.29 (2H, m, 2CHAr), 6.91–
6.86 (2H, m, 2CHAr), 6.77 (1H, s, H-10), 5.22–5.15 (2H, m, OCH2Ph),
4.34 (1H, d, J¼6.1 Hz, H-4a), 4.13–4.04 (1H, m, H-12), 4.13–4.04 (1H,
m, H-3), 3.92 (3H, s, OCH3), 3.90 (3H, s, OCH3), 3.74 (1H, d, J¼6.1 Hz,
H-10b), 3.64 (1H, m, H-12), 2.87 (1H, dd, J¼16.6, 4.5 Hz, H-2), 2.64
(1H, m, H-11), 2.38 (1H, dd, J¼16.6, 9.0 Hz, H-2), 2.23 (1H, m, H-11),
0.92 (9H, s, 3SiCCH3), 0.26 (3H, s, SiCH3), 0.23 (3H, s, SiCH3); 13C
NMR (100 MHz, CDCl3) 205.1 (C-1), 162.6 (C-6), 159.5 (Cq), 152.3
(Cq), 148.1 (Cq), 136.6 (Cq), 136.1 (CHAr), 130.1 (CHAr), 128.5
(2CHAr), 127.9 (CHAr), 127.5 (2CHAr), 127.3 (CHAr), 121.5 (Cq), 121.2
(CHAr), 120.1 (Cq), 112.6 (C-10), 111.9 (C-7), 111.2 (CHAr), 83.0 (C-4),
70.8 (OCH2Ph), 68.1 (C-4a), 56.1 (OCH3), 55.7 (OCH3), 50.9 (C-3),
49.1 (C-10b), 43.8 (C-12), 42.2 (C-2), 34.3 (C-11), 25.7 (3SiCCH3),18.2
(SiC(CH3)3), ꢁ2.4 (SiCH3), ꢁ2.5 (SiCH3); m/z (EI) 645 Found: M,
645.2596. C36H43O6NSSi requires M, 645.2580.
7.44–7.42 (2H, m, 2CHAr), 7.38–7.35 (2H, m, 2CHAr), 7.32–7.28 (1H,
m, CHAr), 7.10 (1H, t, J¼7.6, Hz, CHAr), 7.04 (1H, d, J¼8.4 Hz, CHAr),
6.68 (1H, s, H-7), 6.61 (1H, s, H-10), 5.10 (2H, m, OCH2Ph), 4.03 (1H,
d, J¼14.3 Hz, H-6), 4.00 (1H, m, H-1), 3.98 (3H, s, OCH3), 3.94 (1H,
dd, J¼12.2, 6.1 Hz, H-3), 3.83 (3H, s, OCH3), 3.49 (1H, d, J¼13.4 Hz, H-
6), 3.29 (1H, t, J¼7.8 Hz, H-12), 3.01–2.93 (1H, m, H-11), 2.89 (1H, m,
H-10b), 2.86 (1H, m, H-4a), 2.66 (1H, td, J¼11.0, 7.5 Hz, H-12), 2.31
(1H, dd, J¼14.1, 5.5 Hz, H-11), 2.23 (1H, dd, J¼15.0, 8.4 Hz, H-2),
2.16–2.08 (1H, m, H-2), 0.90 (9H, s, 3SiCCH3), 0.19 (3H, s, SiCH3), 0.19
(3H, s, SiCH3); 13C NMR (100 MHz, CDCl3) 157.1 (Cq), 148.6 (Cq),
147.2 (Cq), 137.1 (Cq), 135.2 (CHAr), 131.1 (CHAr), 128.5 (CHAr), 128.2
(CHAr), 128.0 (Cq), 127.8 (CHAr), 127.4 (Cq), 127.2 (CHAr), 125.2
(CHAr), 120.8 (CHAr), 112.4 (CHAr), 111.8 (C-10), 111.6 (C-7), 82.0 (C-
4), 75.1 (C-4a), 71.1 (OCH2Ph), 68.3 (C-1), 65.1 (C-3), 56.2 (OCH3),
55.9 (OCH3), 55.2 (C-6), 52.3 (C-12), 39.7 (C-10b), 34.9 (C-11), 33.0
(C-2), 26.0 (3SiCCH3),18.3 (SiC(CH3)3), ꢁ1.6 (SiCH3), ꢁ1.8 (SiCH3); m/
z (EI) 665 Found: M, 665.2857. C36H47O7NSSi requires M, 649.2843.
4.8. Amine
4.10. Phenol (25)
LiAlH4 (270 mg, 7.11 mmol) was added to a stirred solution of
thioether 23 (153 mg, 0.23 mmol) in THF (30 mL) at ꢁ78 ꢂC. The re-
action mixture was warmed to 20 ꢂC and stirred for 3 h 15 min. H2O
(0.28 mL), 1 M aqueous KOH (0.28 mL) and H2O (0.28 mL) were se-
quentially added (cautiously with some cooling/ice-bath) to the re-
action mixture. The resulting mixture was filtered through Celite
(CH2Cl2 as eluant), concentrated in vacuo and purified by flash col-
umn chromatography (gradient elution: 0–50% EtOAc/petroleum
ether) to give the amine (100 mg, 66%) as a colourless foam. Rf¼0.42
(EtOAc/petroleum ether/NEt3; 49/49/2); nmax/cmꢁ1 2928, 2856, 1608,
1576, 1515, 1467, 1317, 1255, 1125, 1080, 1016; 1H NMR (400 MHz,
CDCl3) 7.48–7.46 (1H, dd, J¼7.6, 1.7 Hz, CHAr), 7.45–7.43 (2H, m,
2CHAr), 7.38–7.34 (2H, m, 2CHAr), 7.31–7.27 (2H, m, 2CHAr), 6.2 (1H,
dd, J¼7.5, 1.1 Hz, CHAr), 6.89 (1H, m, CHAr), 6.75 (1H, s, H-7), 6.60 (1H,
s, H-10), 5.11 (2H, m, OCH2Ph), 4.78 (1H, br s, OH), 4.08 (1H, m, H-1),
4.05 (1H, d, J¼14.3 Hz, H-6), 3.90 (3H, s, OCH3), 3.87(3H, s, OCH3), 3.64
(1H, t, J¼5.8 Hz, H-3), 3.46 (1H, dd, J¼9.9, 5.9 Hz, H-12), 3.40 (1H, d,
J¼14.4 Hz, H-6), 2.93 (1H, t, J¼3.8 Hz, H-10b), 2.75 (1H, d, J¼3.0 Hz, H-
4a), 2.58 (1H, dd, J¼13.4, 5.7 Hz, H-11), 2.52 (1H, dd, J¼13.8, 6.8 Hz, H-
12), 2.36 (1H, td, J¼14.7, 5.1 Hz, H-2), 2.23 (1H, ddd, J¼14.9, 6.2, 3.5 Hz,
H-2), 1.94 (1H, m, H-11), 0.86 (9H, s, 3SiCCH3), 0.07 (3H, s, SiCH3), 0.07
(3H, s, SiCH3); 13C NMR (100 MHz, CDCl3) 158.9 (Cq), 148.6 (Cq), 147.0
(Cq), 137.3 (Cq), 134.1 (CHAr), 128.8 (CHAr), 128.5 (2CHAr), 128.4
(CHAr), 128.4 (Cq), 127.7 (CHAr), 127.4 (Cq), 127.2 (2CHAr), 124.4 (Cq),
120.9 (CHAr), 111.8 (C-10), 111.4 (C-7), 110.9 (CHAr), 82.3 (C-4), 71.1
(OCH2Ph), 70.3 (C-4a), 70.1 (C-1), 56.3 (C-6), 55.9 (OCH3), 55.7 (OCH3),
52.9 (C-12), 49.2 (C-3), 40.2 (C-10b), 35.43 (C-11), 35.37 (C-2), 25.7
(3SiCCH3), 18.1 (SiC(CH3)3), ꢁ2.6 (SiCH3), ꢁ2.7 (SiCH3); m/z (EI) 633
Found: M, 633.2972. C36H47O5NSSi requires M, 633.2944.
Sulfone 24 (96 mg, 0.14 mmol) was dissolved in THF (7 mL) and
10% Pd/C (115 mg, 0.11 mmol) was added. The reaction mixture was
placed under an atmosphere of H2 (2ꢃdoubled balloon) and stirred
for 15 h. The reaction mixture was then filtered through Celite
(CH2Cl2 as eluant) and concentrated in vacuo to give phenol 25
(78 mg, 94%) as a pale solid. Rf¼0.54(MeOH/CH2Cl2; 10%); nmax
/
cmꢁ1 3049, 2982, 2684, 2306, 1782, 1424, 1262, 1154, 1034; 1H NMR
(400 MHz, CDCl3) 7.93 (1H, dd, J¼7.8, 1.7, Hz, CHAr), 7.57 (1H, ddd,
J¼9.1, 7.5, 1.7 Hz, CHAr), 7.10 (1H, t, J¼7.6 Hz, CHAr), 7.04 (1H, d,
J¼8.4 Hz, CHAr), 6.64 (1H, s, H-7), 6.63 (1H, s, H-10), 5.58 (1H, br s,
OH), 4.06 (1H, d, J¼14.3 Hz, H-6), 3.98 (3H, s, OCH3), 3.93 (1H, dd,
J¼12.4, 6.0 Hz, H-3), 3.83 (3H, s, OCH3), 3.50 (1H, d, J¼14.0 Hz, H-6),
3.29 (1H, t, J¼7.9 Hz, H-11), 3.00–2.92 (1H, m, H-12), 2.89–2.87 (1H,
m, H-10b), 2.86 (1H, m, H-4a), 2.65 (1H, ddd, J¼11.3, 8.7, 5.9 Hz, H-
11), 2.30 (1H, dd, J¼14.2, 5.5 Hz, H-12), 2.28–2.21 (1H, m, H-2), 2.15–
2.07 (1H, m, H-2), 0.90 (9H, s, 3SiCCH3), 0.18 (6H, s, 2SiCH3); 13C
NMR (100 MHz, CDCl3) 157.2 (Cq), 145.6 (Cq), 144.5 (Cq), 135.2
(CHAr), 131.2 (CHAr), 128.14 (Cq), 128.07 (Cq), 126.0 (Cq), 120.8
(CHAr), 112.4 (CHAr), 112.2 (CHAr), 110.3 (CHAr), 82.1 (C-4), 75.4 (C-
4a), 68.5 (C-1), 65.1 (C-3), 56.2 (OCH3), 55.8 (OCH3), 55.2 (C-6), 52.4
(C-11), 39.8 (C-10b), 34.9 (C-12), 33.0 (C-2), 26.0 (3SiCCH3), 18.4
(SiC(CH3)3), ꢁ1.5 (SiCH3), ꢁ1.8 (SiCH3); m/z (EI) 575 Found: M,
575.2416. C29H41O7NSSi requires M, 575.2373.
4.11. Fortucine (2)
Sodium amalgam (1.59 g, 6% Na in Hg, 4.14 mmol) was added to
a stirred solution of phenol 25 (119 mg, 0.21 mmol) in refluxing
MeOH (6.2 mL). After 50 min, reaction mixture was cooled and H2O
(50 mL) was added followed by 1 M aqueous KOH until pH>10. The
aqueous layer was extracted with EtOAc (2ꢃ50 mL). The combined
organic extracts were dried, concentrated in vacuo and purified by
column chromatography (gradient elution: 0–30% MeOH/CH2Cl2) to
give fortucine 26a (28 mg, 50%) as a colourless solid, which was
recrystallised from isopropanol/hexane (1:2). Rf¼0.28 (MeOH/EtOAc;
15%); mp 188 ꢂC (dec); nmax/cmꢁ1 3362, 2922, 2848,1609,1590,1514,
1447, 1313, 1276, 1248, 1198, 1160, 1127, 1104, 1074, 1033; 1H NMR
(400 MHz, CDCl3) 6.80 (1H, s, H-10), 6.57 (1H, s, H-7), 5.55 (1H, d,
J¼6.7 Hz, H-3), 4.30 (1H, m, H-1), 3.90 (1H, d, J¼14.0 Hz, H-6), 3.85
(3H, s, OCH3), 3.30(1H, d, J¼13.9 Hz, H-6), 3.30(1H, m, H-12)3.00 (1H,
d, J¼5.3 Hz, H-10b), 2.71 (1H, d, J¼5.7 Hz, H-4a), 2.50–2.40 (2H, m, H-
2), 2.5–2.3 (2H, m, H-11), 2.3 (1H, m, H-12); 13C NMR (100 MHz,
CDCl3) 145.8 (Cq),143.8 (Cq),139.3 (Cq),127.8 (Cq),125.5 (Cq),116.2 (C-
3), 112.2 (C-7), 109.4 (C-10), 69.7 (C-1), 59.0 (C-4a), 55.5 (OCH3), 55.0
(C-6), 51.7 (C-12), 40.8 (C-10b), 33.0 (C-11), 27.0 (C-2); m/z (CI) 274
(MþHþ, 100%) Found: M, 273.1367. C16H19O3N requires M, 273.1365.
4.9. Sulfone (24)
Freshly prepared dimethyldioxirane18 (26 mL, w0.1 M in ace-
tone, 1.9 mmol) was added to a stirred solution of the amine and
TFA (3.61 mL, 48.6 mmol) in CH2Cl2 (7.5 mL) at ꢁ78 ꢂC. The reaction
mixture was warmed to 0 ꢂC and after 45 min Me2S (1.43 mL,
19.4 mmol) was added. The mixture was concentrated in vacuo and
dissolved in a mixture of EtOAc (30 mL) and 1 M aqueous KOH
(30 mL). The aqueous layer was extracted with EtOAc (3ꢃ25 mL),
and the combined organic extracts were dried, concentrated in
vacuo and purified by column chromatography (gradient elution:
0–60% EtOAc/petroleum ether containing 2% of NEt3) to give sul-
fone 24 (149 mg, 92%) as a colourless foam. Rf¼0.47 (EtOAc/petro-
leum ether; 70%); nmax/cmꢁ1 2927, 2855, 2336, 2341, 1589, 1515,
1463, 1316, 1258, 1152, 1022, 941; 1H NMR (400 MHz, CDCl3) 7.93
(1H, dd, J¼7.8, 1.7 Hz, CHAr), 7.57 (1H, ddd, J¼9.1, 7.6, 1.7, CHAr),

A. Biechy et al. / Tetrahedron 65 (2009) 6730–6738
6737
4.12. Catalytic hydrogenation of fortucine (2) and authentic
kirkine
temperature. After 1 h, H2O (5 mL) was added and the mixture was
concentrated in vacuo to remove MeOH. The aqueous layer was
then extracted with EtOAc (3ꢃ10 mL) and the combined organic
extracts were washed with brine (30 mL), dried, concentrated in
vacuo and redissolved in toluene (2 mL). A catalytic amount of
APTS$H2O was added and the solution was heated to reflux. After
1 h, the solution was concentrated in vacuo, and directly purified by
column chromatography (100% EtOAc) to give 36 (24 mg, 30%) as
a pale solid. Rf¼0.5 (MeOH/EtOAc; 10%); mp 185 ꢂC; nmax/cmꢁ11657,
1609; 1H (400 MHz, CDCl3) 8.42 (1H, d, J¼8.2 Hz), 7.97 (1H, s), 7.88
(1H, d, J¼7.6 Hz), 7.40 (1H, dd, J¼8.2, 1.1 Hz), 7.30 (1H, dd, J¼7.3,
1.0 Hz), 7.19 (1H, t, J¼7.6 Hz), 4.46 (2H, t, J¼8.4 Hz, NCH2), 3.41 (2H,
t, J¼8.3 Hz, NCH2CH2), 2.56 (3H, s, CH3); 13C (100 MHz, CDCl3) 160.1
(Cq), 142.5 (Cq), 140.0 (Cq), 133.8 (Cq), 130.7 (CHAr), 129.2 (CHAr),
128.3 (CHAr), 125.0 (Cq), 124.3 (CHAr), 123.2 (CHAr), 122.2 (CHAr),
119.8 (CHAr),116.7 (Cq), 46.3 (NCH2), 27.3 (CH2), 22.0 (CH3); m/z (CI)
236 (MþHþ, 100%) Found: M, 235.0997. C16H13ON requires M,
235.0997.
Small amounts (1 mg) of fortucine 2 and natural kirkine6b
(supplied by Prof. Bastida) were dissolved in THF (200 mL) in two
separate flasks and 10% Pd/C (1 mg) was added in each. The reaction
mixtures were placed under an atmosphere of H2 (2ꢃdoubled
balloons) and stirred for 2 h. The reaction mixtures were then fil-
tered through Celite (CH2Cl2 as eluant) and concentrated in vacuo.
The two compounds were found identical by TLC (20% MeOH/
CH2Cl2) and by 1H NMR spectroscopy (CDCl3, 20 ꢂC).
Reduced fortucine 27: m/z (EI) 275 Found: M, 275.1519.
C16H21O3N requires M, 575.1521.
Reduced kirkine 27: m/z (EI) 275 Found: M, 275.1528. C16H21O3N
requires M, 575.1521.
4.13. Pyrrolophenanthridine 36
NaBH4 (1.85 g, 48.9 mmol) was added portion wise over 1 h to
a stirred solution of silyl ether 6 (800 mg, 1.63 mmol) in MeOH
(12 mL) and THF (12 mL) at 0 ꢂC. The reaction mixture was
heated to reflux for 30 min. After cooling, H2O (20 mL) was
added and the mixture was concentrated in vacuo to remove
MeOH. The aqueous layer was then extracted with Et2O
(2ꢃ20 mL) and the combined organic extracts were washed with
brine (50 mL), dried and concentrated in vacuo to give the hy-
drazide as yellow oil.
4-Methylbenzoyl chloride (1.08 mL, 8.15 mmol) in solution in
dry CH2Cl2 (8 mL) was added dropwise to a stirred mixture of the
above hydrazide, DMAP (40 mg, 0.33 mmol) and Et3N (3.4 mL,
24.45 mmol) in CH2Cl2 (16 mL). After 24 h, H2O (50 mL) and satu-
rated aqueous solution of NaHCO3 (50 mL) were added to the re-
action mixture, and the aqueous layer was extracted with in CH2Cl2
(2ꢃ50 mL). The organic layer was dried, concentrated in vacuo and
the crude material was triturated with ether. The mixture was fil-
tered to remove acid anhydride, the filtrate was then concentrated
in vacuo and purified by flash column chromatography (gradient
elution: 0–20% EtOAc/petroleum ether) to give radical precursor 33
(843 mg, 85% over two steps) as a white foam. Rf¼0.35 (EtOAc/
petroleum ether; 20%).
Radical precursor 33 (843 mg, 1.375 mmol) was dissolved in
chlorobenzene (14 mL) and lauroyl peroxide (DLP, 110 mg portions,
770 mg, 140%) was added into the reaction mixture at 20 min in-
tervals for 2 h 40 min. Immediately after the first addition of DLP,
the reaction mixture was heated to reflux for 7 h. After cooling, the
resulting solution was concentrated in vacuo and directly purified
by column chromatography (gradient elution: 0–100% EtOAc/pe-
troleum ether) to give pentacycle 34 (321 mg, 55%) as a colorless
foam. Rf¼0.4 (EtOAc).
4.14. X-ray crystallographic data
Single crystal X-ray diffraction was used for determining the
solid-state structure of fortucine (2). All data were collected on
a Nonius KappaCCD diffractometer at 150(1) K using Mo K
a
(
l
¼0.71073 Å) X-ray source and a graphite monochromator. Ex-
perimental details are described in Table 1. The crystal structures
were solved in SIR 9719 and refined in SHELXL-9720 by full-matrix
least-squares using anisotropic thermal displacement parameters
for all non-hydrogen atoms. All the hydrogen atoms were placed in
geometrically calculated positions.
Crystal data for fortucine 2
Molecular formula
Molecular weight
Crystal habit
Crystal dimensions (mm)
Crystal system
Space group
a (Å)
C16H19NO3
273.32
Colourless plate
0.12ꢃ0.12ꢃ0.05
Orthorhombic
Pbca
8.766(1)
b (Å)
c (Å)
13.240(1)
22.830(1)
90.00
90.00
90.00
2649.7(4)
8
1.370
1168
0.094
a
b
g
(ꢂ)
(ꢂ)
(ꢂ)
V (Å3)
Z
d (g cmꢁ3
F(000)
)
m
(cmꢁ1
)
Absorption corrections
multi-scan; 0.9888 min,
0.9953 max
KappaCCD
Diffractometer
X-ray source
MoK
a
l
(Å)
0.71069
Graphite
150.0(1)
Phi and omega scans
A mixture of H2O (7.5 mL) and TFA (7.5 mL) was added to
a stirred solution of pentacycle 34 (321 mg, 0.75 mmol) in THF
(3.75 mL). After 2 h, the reaction mixture was diluted with EtOAc
(50 mL), the organic layer was washed with H2O (50 mL), cau-
tiously basified by washing with saturated aqueous solution of
Na2CO3 (3 ꢃ50 mL), washed with brine (100 mL), dried and con-
centrated in vacuo to give the ketone (290 mg, 99%) as a colorless
foam. Rf¼0.6 (EtOAc).
Monochromator
T (K)
Scan mode
Maximum
q
21.48
HKL ranges
ꢁ9.8; ꢁ10.13; ꢁ23.21
Reflections measured
Unique data
Rint
Reflections used
Criterion
Refinement type
Hydrogen atoms
Parameters refined
Reflections/parameter
wR2
8521
1509
0.1228
1117
(I>2
sI)
This ketone (236 mg, 0.61 mmol) was dissolved in THF (6 mL)
and TBAF (2.75 mL,1 M in THF) and acetic acid (190 mL) were added.
Fsqd
Mixed
184
6
0.1446
0.0645
0.0657;0.8957
1.103
After 20 h, the resulting mixture was concentrated in vacuo and
purified by column chromatography (gradient elution: 50–100%
EtOAc/petroleum ether, then 1–5% MeOH/EtOAc) to give alcohol 35
(100 mg, 60%) as white crystals. Rf¼0.45 (MeOH/EtOAc; 5%); m/z
(CI) 272 (MþHþ, 100%).
R1
Weights a, b
GoF
NaBH4 (12 mg, 0.33 mmol) was added to a stirred solution of
alcohol 35 (94 mg, 0.33 mmol) in MeOH (3.2 mL) at room
Difference peak/hole (e Åꢁ3
)
0.201(0.058)/ꢁ0.239(0.058)

6738
A. Biechy et al. / Tetrahedron 65 (2009) 6730–6738
10. Formed by the condensation of acetaldehyde with 1 (methylthio)thiocarbonyl-
Acknowledgements
1-phenyl-hydrazine. See Ref. 9a.
11. (a) Callier-Dublanchet, A. C.; Quiclet-Sire, B.; Zard, S. Z. Tetrahedron Lett. 1995,
36, 8791; (b) Callier-Dublanchet, A. C.; Quiclet-Sire, B.; Zard, S. Z. Tetrahedron
Lett. 1997, 38, 2463; (c) Depature, M.; Grimaldi, J.; Hatem, J. Eur. J. Org. Chem.
2001, 941.
We wish to thank Professor J. Bastida (Barcelona) for a sample of
kirkine and Ecole Polytechnique for fellowships (to A.B. and S.H.).
12. (a) Hoang-Cong, X.; Quiclet-Sire, B.; Zard, S. Z. Tetrahedron Lett. 1999, 40, 2125;
(b) Barclay, G. L.; Quiclet-Sire, B.; Sanchez-Jimenez, G. Org. Biomol. Chem. 2005,
3, 823.
References and notes
1. For a review on Amaryllidaceae alkaloids, see: Jin, Z. Nat. Prod. Rep. 2007, 24, 886.
2. Dalton, D. R. The Alkaloids: The Fundamental Chemistry-A Biogenetic Approach;
Marcel Dekker: New York, NY, 1979.
13. Dolan, S. C.; MacMillan, J. J. Chem. Soc., Chem. Commun. 1985, 1588.
14. For recent synthetic studies, see: (a) Parnes, J. S.; Carter, D. S.; Kurz, L. J.;
Flippin, L. A. J. Org. Chem. 1994, 59, 3497; (b) Hutchings, R. H.; Meyers, A. I. J.
Org. Chem. 1996, 61, 1004; (c) Padwa, A.; Dimitroff, M.; Waterson, A. G.; Wu, T.
J. Org. Chem. 1998, 63, 3986; (d) Harayama, T.; Hori, A.; Abe, H.; Takeuchi, Y.
Tetrahedron 2004, 60, 1611; (e) Torres, J. C.; Pinto, A. C.; Garden, S. J. Tetrahe-
dron 2004, 60, 9889.
15. Oxoassoanine, assoanine: (a) Llabres, J. M.; Viladomat, F.; Bastida, J.; Codina, C.;
Rubiralta, M. Phytochemistry 1986, 25, 2637; Pratosine, pratorinine, prator-
imine: (b) Ghosal, S.; Saini, K. S.; Frahm, A. W. Phytochemistry 1983, 22, 2305;
Hippadine: (c) Ghosal, S.; Rao, P. H.; Jaiswal, D. K.; Kumar, Y.; Frahm, A. W.
Phytochemistry 1981, 20, 2003; (d) El Mehgazi, A. M.; Ali, A. A.; Mesbah, M. K.
Planta Med. 1975, 28, 336.
16. (a) Chattopadhyay, S. C.; Chattopadhyay, U.; Marthur, P. P.; Saini, K. S.; Ghosal, S.
Planta Med. 1983, 49, 252; (b) Petti, G. R.; Gaddamidi, V.; Goswani, A.; Cragg, G.
M. J. Nat. Prod. 1984, 47, 796; (c) Ghosal, S.; Lochan, R.; Ashutosh, K.; Kumar, Y.;
Srivastava, R. S. Phytochemistry 1985, 24, 1825; (d) Zee-Cheng, R. K. Y.; Yan, S. J.;
Chen, C. C. J. Med. Chem. 1978, 21, 199; (e) Chen, C. C.; Zee-Cheng, R. K. Y.
Heterocycles 1981, 15, 1275.
3. Hartwell, J. L. Lloydia 1967, 30, 379.
4. (a) De Laurentis, C.; Rosato, A.; Vitali, C.; Leone, L.; Milillo, M. A. Riv. Ital. EPPOS
2004, 38, 19; (b) De Laurentis, C.; Rosato, A.; Vitali, C.; Leone, L.; Milillo, M. A.
Chem. Abstr. 2005, 143, 129863; (c) Martin, S. F. In The Alkaloids; Brossi, A., Ed.;
Academic: New York, NY, 1987; Vol. 30, pp 251–376; (d) Ghosal, S.; Saini, K. S.;
Razdan, S. Phytochemistry 1985, 24, 2141; (e) Mineshita, T.; Yayaguchi, K.;
Takeda, K.; Kotera, K. Ann. Rep. Shinogi Res. Lab. 1956, 6, 119; (f) Chattopadhyay,
U.; Chaudhuri, L.; Das, S.; Kumar, Y.; Ghosal, S. Pharmazie 1984, 39, 855; (g) Leo
De, P.; Dalessandro, G.; De Santis, A.; Arigoni, O. Plant Cell Physiol. 1973, 14, 487.
5. (a) Takagi, S.; Taylor, W. I.; Uyeo, S.; Yajima, H. J. J. Chem. Soc. 1955, 4003; (b)
Fales, H.; Warnhoff, E. M.; Wildman, W. C. J. Am. Chem. Soc. 1955, 77, 5885.
6. (a) Tokhtabaeva, G. M.; Sheichenko, V. I.; Yartseva, I. V.; Tolkachev, O. N. N.
Khim. Prir. Soedin. 1987, 727; (b) Bastida, J.; Codina, C.; Peeters, P.; Rubiralta, M.;
Orozco, M.; Luque, F. J.; Chharbra, S. C. Phytochemistry 1995, 40, 1291; (c) Ri-
chomme, P.; Pabuccuoglu, V.; Gozler, T.; Freyer, A. J.; Shamma, M. J. Nat. Prod.
1989, 52, 1150.
7. (a) Watt, J. M.; Breyer-Brandwijk, M. G. Medicinal and Poisonous Plants of
Southern and Eastern Africa; E. & S. Livingstone: Edinburgh, 1962; (b) Kokwaro,
J. O. Medicinal Plants of East Africa; East Africa Literature Bureau: Kampala, 1976.
8. Biechy, A.; Hachisu, S.; Quiclet-Sire, B.; Ricard, L.; Zard, S. Z. Angew. Chem., Int.
Ed. 2008, 47, 1436.
17. Jensen, K. A.; Anthoni, U.; Holm, A. Acta Chem. Scand. 1969, 23, 1916.
18. Eaton, P. E.; Wicks, G. E. J. Org. Chem. 1988, 53, 5353.
19. Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.; Giacovazzo, C.; Guagliardi,
A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. SIR97, an integrated package of
computer programs for the solution and refinement of crystal structures using
single crystal data.
9. Tran-Huu-Dau, M. E.; Wartchow, R.; Winterfeldt, E.; Wong, Y. S. Chem.dEur. J.
2001, 7, 2349.
20. Sheldrick, G. M. SHELXL-97; Universita¨t Go¨ttingen: Go¨ttingen, Germany, 1997.