Stereoselective synthesis of phosphorothioate and alkylphosphinate analogs using a L-tryptophan derived chiral auxiliary

Article abstract of DOI:10.1016/S0040-4020(00)00308-2

A novel phosphorus containing ring system incorporating both imidazole and indole moieties has been synthesized and investigated. A new L-tryptophan derived chiral auxiliary which incorporates those elements has been prepared and used for the stereoselective synthesis of novel indolophosphorothioate and alkylphosphinate analogs. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00308-2

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 4355±4365
Stereoselective Synthesis of Phosphorothioate and
Alkylphosphinate Analogs Using a l-Tryptophan Derived
Chiral Auxiliary
*
Yixin Lu and George Just
Department of Chemistry, McGill University, Montreal, Quebec H3A 2K6, Canada
Received 4 February 2000; revised 11 April 2000; accepted 12 April 2000
AbstractÐA novel phosphorus containing ring system incorporating both imidazole and indole moieties has been synthesized and investi-
gated. A new l-tryptophan derived chiral auxiliary which incorporates those elements has been prepared and used for the stereoselective
synthesis of novel indolophosphorothioate and alkylphosphinate analogs. q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
drawbacks of PS-oligos is their polydiastereomerism due to
the generation of a new chiral center at each phosphoro-
thioate linkage. The PS-oligos currently employed in
clinical trials and biological studies are obtained as mixtures
The therapeutic potential of oligonucleotide phosphoro-
thioates (PS-oligos) is well recognized.1 One of the major
Scheme 1.
Keywords: diastereoselection; imidazoles; indoles; nucleotides.
* Corresponding author. Tel.: 11-514-398-6232; fax: 11-514-398-3797; e-mail: just@chemistry.mcgill.ca
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00308-2

4356
Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
Scheme 2.
of diastereomers. Several approaches2 have been designed
to address the stereoselective synthesis of PS-oligos. The
most advanced one to date is the oxathiaphospholane
method developed by Stec and coworkers.3 However, it
suffers from the fact that the required chiral precursors
have to be separated by chromatography.
synthesized before, and it therefore would be a worthwhile
target for synthesis and evaluation of reactivity.
Results and Discussion
Model studies
We have been involved for some time in synthesizing
diastereomerically pure phosphorothioates.4 Our investi-
gations on the stereoselective synthesis of phosphoro-
thioates and their potential application to solid phase
synthesis have led to the following results.
Two key issues needed to be addressed, which are whether
the six-membered ring will be formed when such an aux-
iliary reacts with dichlorophosphorus compounds, and
whether the imidazole and indole moieties will leave
sequentially when reacted with nucleosides. We therefore
synthesized achiral and racemic model compounds to test
the feasibility of the methodology.
In the imidazo-oxazaphosphorine approach4c as shown in
Scheme 1, the reaction of the chiral hydroxypropylimida-
zole 1 with ethyl dichlorophosphite and triethylamine gave
a mixture of diastereomeric imidazophosphorines 2. Heat-
ing the reaction mixture effected a presumably triethyl-
ammonium chloride catalyzed epimerization to a single
diastereoisomer of 2. Reaction of the latter with a variety
of alcohols converted 2 to the phosphite triester 3, and hence
to the corresponding phosphorothioate in a stereospeci®c
manner. Extreme hydrolytic instability of 2, and our in-
ability to synthesize an analog of 2 in which the ethyl
group is replaced by a nucleoside convinced us to investi-
gate a more stable azole, such as indole, as a leaving group.
Treatment of 1-(phenylsulfonyl) indole with n-butyllithium,
followed by quenching with DMF, gave 1-(phenylsulfonyl)-
2-indolecarboxaldehyde.5 N-trityl imidazole was then
reacted with n-butyllithium, and added to the aldehyde.
However, no desired coupling product was obtained,
presumably due to the steric hindrance introduced by two
large protecting groups on indole and imidazole. In order to
circumvent this problem, we carried out an analogous reac-
tion with an unprotected indole. As shown in Scheme 3,
indole-2-carboxylic acid 8 was transformed via its acid
chloride 9 to Weinreb amide 10, and the latter was treated
with two equivalents of the N-tritylimidazole anion to give
11. Deprotection gave achiral ketone 12. Although not
appropriate for the synthesis of a chiral phosphonate or
phosphorothioate, we investigated the reaction of 12 with
As illustrated in the indolo-oxazaphosphorine approach,4f
the reaction of hydroxyalkylindoles 4 with phosphorus
trichloride provided 5, which was converted to a mixture
of diastereomeric indolophosphorines 6 by reaction with
50-O-protected thymidine. The latter could not be equi-
librated to a single diastereomer under basic or acidic con-
ditions without decomposition. The indole residue could act
as a leaving group, provided the 50-hydroxy group of thymi-
dine was activated by a strong base such as 1,8-diazabi-
cyclo[5.4.0]undec-7-ene (DBU). The displacement reaction
proceeded in a stereospeci®c manner.
It seemed logical to link the two leaving groups by a chiral
substituted methylene group. This would provide a recover-
able chiral auxiliary suitable for the synthesis of both chiral
phosphonates and phosphate analogs, as illustrated in
Scheme 2. In addition, ring system B has never been
Scheme 3. (i) SOCl2, Et3N, DME, 08C to RT; (ii) CH3NH(OCH3)´HCl,
CH2Cl2, 08C to RT; (iii) n-BuLi, N-trityl imidazole, THF, 2788C to RT;
(iv) AcOH/MeOH, re¯ux.

Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
4357
meant thymidine not only displaced imidazole, but also
displaced the indole moiety. Since indole is not a good
leaving group under such mild conditions, we reasoned
that the conjugation of indole with the adjacent electron
withdrawing carbonyl group made it a better leaving
group. Having established that ring systems of type 13 are
readily formed, we then proceeded to reduce the ketone 12
to alcohol 15, which was protected as its TBDMS ether 16.
When 16 was treated with methyl dichlorophosphite, 31P
NMR revealed two resonances at 88.6 ppm (major) and
80.3 ppm (minor) corresponding to 17. After heating at
558C to equilibrate those two products, the resonance at
Scheme 4. (i) MeOPCl2, 2.2 equiv. Et3N, CH2Cl2, 08C to RT; (ii) T3OH,
CH2Cl2, RT.
methyl dichlorophosphite to establish the stability and reac-
tivity of cyclic phosphite derivative 13.
0
88.6 ppm shifted to 80.3 ppm. Without puri®cation, T3 OH
in dry dichloromethane was added into the solution of 17 in
the NMR tube. No reaction happened initially. The solution
was allowed to stand overnight at room temperature. A
reaction started to occur, but was very slow and after several
days, less than 10% of 17 had reacted. In contrast to the
normally very reactive and therefore unstable imidazo-
oxazaphosphorine intermediate, the cyclic compound 17
could be separated by silica gel column chromatography
and fully characterized (Scheme 5).
A model reaction with 12 is shown in Scheme 4. Ketone 12
in freshly distilled dichloromethane containing triethyl-
amine in a dry NMR tube was treated with methyl dichloro-
phosphite at 08C. 31P NMR showed that the original
resonance at around 180 ppm corresponding to methyl
dichlorophosphite disappeared and the new resonance at
80 ppm corresponding to 13 was formed immediately.
6
0
0
5 -O-tert-butyldimethylsilyl thymidine (T3 OH) dissolved
in dichloromethane was then added to the reaction mixture.
In contrast to imidazo-oxazaphosphorine,4c the reaction
went slowly. After allowing the reaction to proceed at
ambient temperature for about 15 h, the original resonance
at 80 ppm disappeared and a new resonance at 141 ppm was
observed. The product turned out to have structure 14. That
We suspected that the lack of reactivity of 17 was due to
steric hindrance, and thus carried out molecular modeling to
establish the most stable conformation of 17. Cyclic
compound 17 equilibrated to give the thermodynamically
more stable conformation. For related systems,4g,7 this had
Scheme 5. (i) LiAlH4, THF, 08C to RT; (ii) TBDMSCl, imidazole, DMF; (iii) MeOPCl2, 2.2 equiv. Et3N, CH2Cl2.
Figure 1.

4358
Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
0
Scheme 6. (i) MeOH, RT; (ii) T3 OH, DBU, CH2Cl2, RT; (iii) Beaucage's reagent.
meant that an exocyclic methoxy group is axially or pseudo-
axially disposed, whereas the bulky silyloxy group prefers
an equatorial arrangement (i.e. A, boat conformation).
Depending on whether the phosphorus containing six-
membered ring adopts a boat or chair-like conformation,
the two substituents have a trans or cis relationship. Much
to our surprise, calculation (Macromodel, version 5.5)8 indi-
cated that the most stable conformation was a boat, in which
the TBDMS group was axially disposed and the methoxy
group took up the pseudo-equatorial conformation (17b)
(Fig. 1). This is presumably due to the fact that the hydrogen
attached to the 3-position of indole exerts enough steric
hindrance to make the relatively unhindered axial confor-
mation an energy minimum, in which the only interaction of
any consequence is a ¯ag±pole interaction with the phos-
phorus lone pair. 2D-NOESY spectrum of 17 showed a
cross-peak between the carbinol proton and H-3 of indole.
No cross-peaks were observed between the silyloxy protons
and the methoxyl protons or H-3 of indole. These results are
in agreement with conformation C. This conformation also
explains why the displacement of the imidazole is so slow,
since the approach of a nucleophile is blocked by the bulky
OTBDMS group.
An easy way to construct such a ring system containing a
chiral center is based on the Pictet±Spengler reaction of a
tryptophan derivative with an appropriate aldehyde, as
outlined in Scheme 7. Reaction of imidazole 2-carbox-
aldehyde10 and Nb-benzyltryptophan methyl ester11 in the
presence of p-toluenesulfonic acid gave the desired product
as one diastereomer. The reported12,13 uncatalyzed reaction
proceeded in an unsatisfactory manner. Cook and
coworkers11 have shown that analogous reactions led to
the formation of the trans diastereomer, and this was
con®rmed by a 2D-NOESY spectrum of 22 which did not
show any cross-peaks between H1 and H2. The stereo-
chemistry of the chiral auxiliary 22 was assigned accord-
ingly.
Stereoselective synthesis of phosphorothioate
In order to con®rm that steric hindrance was the cause for
0
the slow reaction of 17 with T3 OH, we repeated the reaction
with a smaller alcohol (Scheme 6). Upon the addition of
methanol to a solution of 17 in dichloromethane, 31P
NMR revealed the new peak within 20 min at around
146 ppm which was assigned to dimethyl phosphite 18.
Reaction of 22 with methyl dichlorophosphite and triethyl-
amine in dichloromethane was carried out in a dry NMR
tube (Scheme 8). 31P NMR revealed the appearance of a
single new peak at 88 ppm, corresponding to the formation
of 26 as a single diastereomer. Addition of one equivalent of
0
When 18 was treated with T3 OH in the presence of DBU,
0
0
5 -O-TBDMS-thymidine (T3 OH) resulted in the rapid
a further displacement occurred. The 31P NMR shifted to
140 ppm, indicating the formation of phosphite triester.
Sulfurization by Beaucage's reagent (3H-1,2-benzo-
dithiole-3-one 1,1-dioxide)9 yielded phosphorothioate 19.
After puri®cation, 16 was recovered in 80% yield.
formation of 30 (31P NMRˆ143 ppm), accompanied by
the formation of a small amount of material with 31P
NMR at 141 ppm. We suspected that the latter material
Synthesis of the chiral auxiliary
So far, we demonstrated that six-membered ring system A
incorporating the heteroatoms of an indole and imidazole
connected to a trivalent phosphorus can be readily formed.
Due to steric hindrance, the normally extremely rapid
displacement of the imidazole moiety went ef®ciently
only when a small alcohol such as methanol was used. In
order to investigate whether the displacements of imidazole
and (base catalyzed) indole could be carried out sequentially
and in a stereoselective and expeditious manner, we decided
to substitute the X group in A (17, XˆOTBDMS) by a ring
(see dotted lines). Molecular modeling indicated this
structure was much less hindered to the displacement at
phosphorus.
Scheme 7.

Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
4359
Scheme 8. (i) MeOPCl2, 2.2 equiv. Et3N, CH2Cl2 or CDCl3; (ii) 3.0 equiv. 50-O-TBDMS-thymidine in CH2Cl2; (iii) Heated at 558C, 15 h; (iv) Beaucage's
reagent.
was the product of a double displacement, and therefore
having the characteristic thiophosphate triester 31P NMR
resonances at around 70 ppm. Surprisingly, it consisted
of a mixture of the expected dithymidyl thiophosphate
34, accompanied by up to 33% of the dimethoxy
thiophosphate 35, which were separated by ¯ash
chromatography.
0
repeated the reaction with an excess of T3 OH. The major
product 30 (31P NMRˆ143 ppm) formed immediately.
Heating the mixture at 558C for 15 h gave exclusively
material with 31P NMR at 141 ppm. Sulfurization with
Beaucage's reagent proceeded as expected to give materials
Scheme 9.

4360
Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
Scheme 10. (i) PCl3, 2.2 equiv. Et3N, CH2Cl2; (ii) 1.0 equiv. 50-O-TBDMS-thymidine in CH2Cl2; (iii) 30-O-TBDMS-thymidine; (iv) Beaucage's reagent.
(v) Triton B in methanol.
Rigorous exclusion of possible contamination by changing
the solvent used, and running the reactions on the ethyl ester
23, Nb-isopropyl auxiliary 24 and descarbomethoxy aux-
iliary 25 established that the source of methanol necessary
to form the dimethoxy derivative 35 was the phosphorus
attached methoxy group. The sequence of reactions
proposed in Scheme 9 explains the formation of 35.
N-benzyltrimethylammonium hydroxide did not proceed in
a stereoselective manner, and phosphorothioate 40 was
obtained as a 3:2 mixture of diastereomers. This is again
most likely due to competing single and double displace-
ments, as outlined in Scheme 11.
Stereoselective synthesis of alkylphosphinate
0
T3 OH ®rst displaces imidazole to yield a species of type
a. Since imidazole is a good nucleophile, it can displace the
methoxy group to form methanol and c. Alternatively, the
displacement of indole by imidazole gives b. Reactions of b
0
with either T3 OH or methanol, followed by sulfurization
would give 34 or 35.
The stereoselective synthesis of alkylphosphinate is illustrated
in Scheme 12. Chiral auxiliary 22 in dry dichloromethane
containing triethylamine was added to dichloroethyl phos-
phine in dry dichloromethane at 2788C, and the mixture
was warmed up slowly to room temperature. Intermediate
41 was formed immediately, with 31P NMR resonances at
An indolophosphorothioate 39 could be obtained in a stereo-
selective manner as described in Scheme 10. Reaction of
chiral auxiliary 22 with phosphorus trichloride in dichloro-
methane containing triethylamine gave a 10:1 mixture of
diastereomeric chlorophosphites 36 with 31P NMR at
0
about 89.6 ppm. Addition of T3 OH gave 37 which had a
31P MMR resonance around 87 ppm, typical for this type of
0
0
compound. Addition of 3 -O-TBDMS-thymidine (T5 OH) in
dichloromethane provided 38, with 31P NMR signal at
142 ppm, within 10 min. Virtually no double displacement
happened at this stage. Sulfurization with Beaucage's
reagent and isolation gave the novel indolophosphorothioate
39, with 31P NMR at 61.2 and 59.5 ppm, as a mixture of
diastereomers in a ratio of 10:1. After chromatographic
puri®cation, the ratio increased to 20:1.
Unfortunately, the displacement of the chiral auxiliary with
Scheme 11.

Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
4361
85% H3PO4 signal as 0 ppm. THF was distilled from sodium
benzophenone ketyl, triethylamine from calcium hydride,
dichloromethane from phosphorus pentoxide and methanol
from magnesium. Phosphorus trichloride was ®rst degassed
by re¯uxing for 2 h under argon followed by fractional
distillation and was stored under argon. Beaucage's reagent
was a gift from Isis Pharmaceuticals, Carlsbad, CA. All
other reagents were purchased from Aldrich.
Indole-2-carboxylic acid Weinreb amide 10
To a solution of indole-2-carboxylic acid 8 (645 mg,
4.0 mmol) in anhydrous 1,2-dimethoxyethane (DME) (15
ml) was added triethylamine (0.80 ml, 5.7 mmol). Thionyl
chloride (0.58 ml, 8.0 mmol) was then added very slowly at
08C over a period of 10 min, and the reaction mixture stirred
at 08C for half an hour. The precipitate was ®ltered off, and
the ®ltrate was concentrated in vacuo to give crude 9 as a
dark brown solid. Crude 9 and N,O-dimethylhydroxylamine
hydrochloride (390 mg, 4.0 mmol) were dissolved in dry
dichloromethane (10 ml), the solution was cooled down to
08C and pyridine (0.71 ml, 8.8 mmol) was added. The
mixture was stirred at ambient temperature for 1 h and
evaporated in vacuo. The residue was partitioned between
brine and a 1:1 mixture of dichloromethane and ether, and
the organic layer dried over magnesium sulfate. Puri®cation
by column chromatography gave the desired Weinreb amide
10 (0.60 g, 73% from acid) as a bright yellow solid.
Scheme 12. EtPCl2, 2.2 equiv. Et3N, CH2Cl2; (ii) 1.0 equiv. 50-O-TBDMS-
thymidine in CH2Cl2; (iii) DBU, 30-O-TBDMS-thymidine.
0
76.4 and 74.9 ppm, in a ratio of 1:20. Then T3 OH was
added slowly at 2788C. The mixture was allowed to
warm up to room temperature. 31P NMR showed two new
resonances at 142.5 and 142.0 ppm in a ratio of 20:1.
Puri®cation by ¯ash chromatography gave alkylphosphin-
amidite 42 as a single diastereomer. Stereospeci®c displace-
0
ment of the indole moiety with T5 OH and DBU in
dichloromethane was not successful, and a mixture of
epimeric alkylphosphinates 43 were obtained. The epimeri-
zation process is most likely due to partial double displace-
ment, as depicted in Scheme 11.
1H NMR (500 MHz, CDCl3) d 9.78 (br, 1H, NH), 7.71 (d,
1H, H-4, J4-5ˆ8.0 Hz), 7.46 (d, 1H, H-7, J7-6ˆ7.5 Hz), 7.31
(m, 1H, H-6), 7.26 (s, 1H, H-3), 7.15 (m, 1H, H-5), 3.82 (s,
3H, OCH3), 3.42 (s, 3H, NCH3); 13C NMR (125.7 MHz,
CDCl3) d 161.52, 135.74, 128.08, 127.81, 124.66, 122.36,
120.23, 111.67, 107.81, 61.18, 33.11; MS (CI, NH3) m/z 205
((M11)1, 60.4%), 174 (35.3%), 144 (100%).
In order to avoid double displacement, attempts were made
to block the nitrogen of imidazole by acyl, trityl,
2-(trimethylsilyl)ethoxylmethyl (SEM) or methyl groups.
Unfortunately, all attempts to do so were unsuccessful.
Since imidazole is a good nucleophile, its participation can
be understood. An alternative approach to solve the problem
would be to use other more acidic and hence less nucleo-
philic heterocyclic compounds such as tetrazole. We are
currently developing this methodology.
2-Indolyl-N-trityl-2-imidazolyl ketone 11
n-Butyllithium solution in hexane (1.6 M) (1.37 ml, 2.2
mmol) was added to a stirred solution of N-trityl imidazole
(620 mg, 2.0 mmol) in dry THF (20 ml) at 2788C under
argon. The resulting solution was stirred at 2788C for half
an hour and warmed gradually to room temperature for an
hour. The original colorless solution turned to red gradually.
The mixture was re-cooled to 2788C and indole-2-Weinreb
amide 10 (204 mg, 1.0 mmol) in dry THF (10 ml) was
added. The mixture was allowed to warm up to room
temperature and stirred for an additional hour. The reaction
mixture was then poured into a mixture of 5% hydrochloric
acid and ether at 08C, the mixture was partitioned between
brine and a mixture of ether and dichloromethane (1:1).
Puri®cation by ¯ash chromatography afforded the desired
product 11 (280 mg, 62%) as a yellow solid and recovered
Weinreb amide 10 (70 mg, 35%).
In conclusion, we demonstrated the new type of chiral
auxiliary incorporating indole and imidazole can be easily
prepared from tryptophan, from which a novel chiral
indolophosphorothioate could be obtained. Our investi-
gation on the stereoselective synthesis of phosphorothioate
and alkylphosphinate led us to develop a new type of chiral
auxiliary incorporating tetrazole and indole.
Experimental
General methods
Mass spectra were recorded on MS25RFA and ZAB 2F HS
mass spectrometers. 1H NMR and 13C NMR were recorded
on a Varian XL 200 and Unity 500 spectrometers and are
referenced with respect to the residual signals of the solvent.
The assignments of proton spectra are based on COSY
experiments. 31P NMR spectra were recorded on a Varian
Unity 500 spectrometer and are referenced to the external
1H NMR (500 MHz, CDCl3) d 10.50 (br, 1H, NH), 7.62 (d,
1H, H-4, J4-5ˆ8.0 Hz), 7.35 (m, 2H, H-7 and H of imidazole),
7.24±7.29 (m, 11H, 9H of trityl, H-6, H of imidazole),
7.12±7.18 (m, 7H, 6H of trityl, H-3), 7.06 (m, 1H, H-5);
13C NMR (125.7 MHz, CDCl3) d 170.80, 144.91, 142.67,
137.10, 135.68, 129.81, 127.61, 127.58, 127.32, 127.27,

4362
Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
127.21, 125.52, 122.97, 120.36, 112.10, 111.34, 78.08; MS
(FAB, NBA) m/z 454 (M11)1.
added to a solution of 15 (213 mg, 1.0 mmol) in DMF
(10 ml) containing imidazole (136 mg, 2.0 mmol) at room
temperature. The solution was stirred for 2 h. The solvent
was removed under high vacuum, the residue was dissolved
in ethyl acetate, washed with water and brine, respectively.
Puri®cation by ¯ash chromatography provided the desired
product as a light yellow solid (310 mg, 95% from 12).
2-Indolyl-2-imidazolyl ketone 12
2-Indolyl-N-trityl-2-imidazolyl ketone 11 (440 mg, 1.0
mmol) was dissolved in 5% acetic acid/methanol (10 ml),
the mixture was brought to re¯ux for half an hour. The
solvent was removed in vacuo and the residue was taken
up in ethyl acetate, washed with saturated sodium bicarbo-
nate solution, distilled water and brine, respectively, and
dried over magnesium sulfate. Puri®cation by ¯ash chroma-
tography yielded the product as a yellow solid (0.20 g,
97%).
1H NMR (500 MHz, CDCl3) d 9.38 (br, 1H, NH of imida-
zole), 8.90 (br, 1H, NH of indole), 7.57 (d, 1H, H-4,
J4-5ˆ7.5 Hz), 7.29 (d, 1H, H-7, J6-7ˆ8.0 Hz), 7.13 (m, 1H,
H-6), 7.07 (m, 1H, H-5), 7.03 (s, 1H, H of imidazole), 6.96
(s, 1H, H of imidazole), 6.46 (s, 1H, H-3), 6.18 (s, 1H,
HCOTBDMS), 0.97 (s, 9H, SiC(CH3)3), 0.14 (s, 3H,
Si(CH3)), 0.07 (s, 3H, Si(CH3)); 13C NMR (125.7 MHz,
CDCl3) d 148.48, 138.34, 136.07, 128.21, 127.95, 121.70,
120.38, 119.59, 115.40, 110.92, 99.22, 66.63, 25.65, 25.23,
25.46; HRMS (EI) C18H25N3OSi calcd: 327.1767, found:
327.1770.
1H NMR (500 MHz, CDCl3) d 12.46 (br, 1H, NH of imida-
zole), 11.47 (br, 1H, NH of indole), 8.01 (s, 1H, H of imida-
zole), 7.76 (m, 1H, H-4), 7.65 (m, 1H, H-7), 7.52 (s, 1H, H
of imidazole), 7.36 (s, 1H, H-3), 7.33 (m, 1H, H-6), 7.12 (m,
1H, H-5); 13C NMR (125.7 MHz, CDCl3) d 172.74, 146.26,
138.99, 135.22, 131.79, 128.45, 126.61, 123.75, 121.72,
121.36, 113.64, 112.57; HRMS (EI) C12H9N3O calcd:
211.0746, found: 211.0748.
Cyclic phosphite monoester 17
To a solution of methyl dichlorophosphite (6.4 ml, 0.067
mmol) in dry dichloromethane (0.2 ml) in a scrupulously
dried NMR tube at 08C was added slowly a solution of 16
(20 mg, 0.06 mmol) in dry dichloromethane (0.6 ml)
containing triethylamine (18.4 ml, 0.132 mmol). The NMR
tube was then sealed and the reaction was followed by 31P
NMR. After 24 h at 508C, the crude mixture was puri®ed by
¯ash chromatography to afford 17 as a white solid (16 mg,
70%).
Phosphite triester 14
To
a solution of methyl dichlorophosphite (7.4 ml,
0.078 mmol) in dry dichloromethane (0.2 ml) in a scrupu-
lously dried NMR tube at 08C was added slowly a solution
of 12 (15 mg, 0.07 mmol) in dry dichloromethane (0.6 ml)
containing triethylamine (21.7 ml, 0.156 mmol). The NMR
tube was then sealed. When 31P NMR showed the dis-
appearance of starting material and the formation of a new
resonance at 80 ppm assigned to 13, 50-O-TBDMS thymi-
dine6 (30 mg, 0.084 mmol) in dry dichloromethane was
added. The NMR tube was allowed to stand at room
temperature for 15 h. The crude mixture was puri®ed by
¯ash chromatography to afford the product as a white
solid (26 mg, 81%).
31P NMR (202.3 MHz, CDCl3) d 80.47; 1H NMR
(500 MHz, CDCl3) 7.78 (d, 1H, H-4, J4-5ˆ8.0 Hz), 7.64
(d, 1H, H-7, J6-7ˆ7.5 Hz), 7.33 (m, 1H, H-5), 7.24±7.27
(m, 2H, H-6 and H of imidazole), 7.21 (m, 1H, H of imida-
zole), 6.73 (s, 1H, H-3), 6.06 (s, 1H, HCOTBDMS), 3.52 (d,
3H, OCH3, 3JH-Pˆ9.0 Hz), 0.80 (s, 9H, SiC(CH3)3), 0.26 (s,
3H, Si(CH3)), 0.19 (s, 3H, Si(CH3)); 13C NMR (125.7 MHz,
3
CDCl3) d 147.27 (d, C-2 of imidazole, JP-Cˆ5.5 Hz),
31P NMR (202.3 MHz, CDCl3) d 140.84; 1H NMR
(500 MHz, CDCl3) d 9.23 (br, 1H, NH), 9.19 (br, 1H,
NH), 7.45 (2s's, 2H, 2£H-6), 6.36 (m, 2H, 2£H-10), 4.79±
4.89 (m, 2H, 2£H-30), 4.10 (m, 2H, 2£H-40), 3.79±3.90 (m,
4H, 2£H-50), 3.56 (d, 3H, OCH3), 2.42±2.52 (m, 2H, H-20),
2.11 (m, 2H, H-20), 1.89 (2 s's, 6H, 2£CvCMe), 0.93 (2 s's,
18H, 2£SiC(CH3)3), 0.12 (2 s's, 12H, 2£Si(CH3)2); MS
(FAB, NBA/NaCl) m/z 795 (M1Na21).
140.17 (d, C-2, JP-Cˆ24.7 Hz), 139.43 (d, C adjacent to
2
N21 of indole, 2JC-Pˆ7.3 Hz), 130.35 (d, C-5 of imidazole,
2JC-Pˆ8.24 Hz), 129.49 (d, C between C-3 and C-4 of
3
indole, JC-Pˆ3.66 Hz), 123.60 (C-5), 122.00 (C-6),
3
121.46 (d, C-4 of imidazole, JP-Cˆ19.23 Hz), 121.17 (d,
3
4
C-7, JC-Pˆ12.8 Hz), 110.89 (d) C-4, JC-Pˆ17.4 Hz),
107.74 (C-3), 61.45 (CHOTBDMS), 52.92 (d, OCH3,
2JP-Cˆ8.13 Hz), 25.55 (SiC(CH3)3), 18.05 (SiC(CH3)3),
24.91 (SiCH3), 25.33 (SiCH3); HRMS (FAB) (M1H)1,
calcd: 388.1610, found: 388.1613.
2-Indolyl-2-imidazolyl carbinol 15
2-Indolyl-2-imidazolyl ketone 13 (210 mg, 1.0 mmol) in
dry THF was added slowly into a suspension of lithium
aluminum hydride (38 mg, 1.0 mmol) at 08C, and the
mixture was allowed to stir at room temperature for
10 min. A few drops of ethyl acetate were added to the
reaction mixture to destroy the excess lithium aluminum
hydride. Fieser work-up14 gave alcohol 15, which was
used in the next step without further puri®cation.
Dimethyl indolyl phosphite 18
Methanol (0.10 ml) was added to the mixture containing 17
obtained from the previous experiment in an NMR tube.
Within 30 min, 31P NMR indicated the completion of the
reaction. The crude mixture was puri®ed by ¯ash chroma-
tography to afford 18 as a white solid (16 mg, 65% from 16).
31P NMR (202.3 MHz, CDCl3) d 147.50; 1H NMR
(500 MHz, CDCl3) d 9.18 (br, 1H, NH of imidazole), 8.67
(br, 1H, NH of indole), 7.89 (d, 1H, H-4, J4-5ˆ8.0 Hz), 7.53
(d, 1H, H-7, J7-6ˆ7.5 Hz), 7.09±7.15 (m, 2H, H-5, H-6),
Silyloxy ether 16
tert-Butyldimethylsilyl chloride (180 mg, 1.2 mmol) was

Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
4363
6.96 (m, 2H, 2H of imidazole), 6.68 (s, 1H, H-3), 6.35 (s,
1H, CH(OTBDMS)), 3.45 (d, 3H, OCH3, JH-Pˆ12.5 Hz),
3.28 (d, 3H, OCH3, JH-Pˆ13.0 Hz), 0.93 (s, 9H,
SiC(CH3)3), 0.062 (s, 3H, Si(CH3)), 0.036 (s, 3H,
Si(CH3)); 13C NMR (125.7 MHz, CDCl3) d 148.64,
142.59, 142.45, 138.54, 138.51, 129.76, 128.92, 123.74,
21.18, 14.35; HRMS (EI) C23H22N4O2 calcd: 386.1743,
found: 386.1745.
2-Isopropyl-3-(ethoxycarbonyl)-1,2,3,4-tetrahydro-9H-
pyrido[3,4-b]indole-2-imidazole 23
4
121.29, 120.71, 115.56 (d, JC-Pˆ2.77 Hz), 115.02, 106.03
Ethyl ester 23 was prepared using the same procedure as
described for the preparation of 22.
3
2
(d, JC-Pˆ3.77 Hz), 99.39, 52.28 (d, JC-Pˆ17.5 Hz), 51.65
2
(d, JC-Pˆ18.4 Hz), 25.80, 24.91, 25.33.
1H NMR (500 MHz, CD2Cl2) d 11.67 (br, 1H, NH of imida-
zole), 10.02 (br, 1H, NH of indole), 7.45 (d, 1H, H-4, J4-
Phosphorothioate 19
ˆ7.5 Hz), 7.29 (d, 1H, H-7, J7-6ˆ8.0 Hz), 6.95±7.07 (m,
5
4H, H-5, H-6, 2H of imidazole), 5.36 (s, 1H,
(NH)NvCCHN(i-Pr)), 4.22 (m, 1H, CH2CHCOOEt),
4.00±4.10 (m, 2H, COOCH2CH3), 3.50 (d, 1H,
CH2CHCOOEt, Jˆ15.0 Hz), 3.22 (m, 1H, CH(CH3)2),
2.98 (ddd, 1H, CH2CHCOOEt, Jˆ15.0, 8.0 and 1.5 Hz),
1.15 (m, 6H, CH(CH3)2, COOCH2CH3), 1.07 (d, 3H,
CH(CH3)2, Jˆ7.0 Hz); MS (EI) m/z 352 (M1).
To a solution of 18 (20 mg, 0.06 mmol) in dry dichloro-
methane (0.25 ml) in a NMR tube was added a solution of
50-O-TBDMS-thymidine (42 mg, 0.12 mmol) in dry
dichloromethane (0.45 ml) containing DBU (45 ml,
0.30 mmol) at room temperature. The NMR tube was then
sealed. 31P NMR indicated the reaction went to completion
within 20 min. The mixture was then loaded to the top of
column to run a fast column to ®lter off DBU (acetonitrile:
dichloromethaneˆ1:1). The solvent was removed in vacuo
and the residue was re-dissolved in dichloromethane and
Beaucage's reagent (24 mg, 0.12 mmol) was added. After
10 min, the mixture was diluted with dichloromethane,
washed with brine and dried over magnesium sulfate. Puri-
®cation by ¯ash chromatography provided the desired
product 19 as a white solid (24 mg, 82%) and recovered
16 (16 mg, 80%).
2-Isopropyl-3-(methoxycarbonyl)-1,2,3,4-tetrahydro-
9H-pyrido[3,4-b]indole-2-imidazole 24
Nb-isopropyl auxiliary 24 was prepared using the same
procedure as described for the preparation of 22.
1H NMR (500 MHz, CD2Cl2) d 11.62 (br, 1H, NH of imida-
zole), 9.90 (br, 1H, NH of indole), 7.47 (d, 1H, H-4,
J4-5ˆ7.5 Hz), 7.30 (d, 1H, H-7, J7-6ˆ8.0 Hz), 6.98±7.09
(m, 4H, H-5, H-6, 2H of imidazole), 5.37 (s, 1H,
(NH)NvCCHN(i-Pr)), 4.26 (dd, 1H, CH2CHCOOCH3,
Jˆ6.0 and 1.5 Hz), 3.63 (s, 3H, COOCH3), 3.51 (d, 1H,
CH2CHCOOCH3, Jˆ15.0 Hz), 3.24 (m, 1H, CH(CH3)2),
3.00 (ddd, 1H, CH2CHCOOCH3, Jˆ15.0, 6.6 and 2.0 Hz),
1.18 (d, 3H, CH(CH3)2, Jˆ6.5 Hz), 1.07 (d, 3H, CH(CH3)2,
Jˆ6.5 Hz); 13C NMR (125.7 MHz, CD2Cl2) d 177.88,
150.76, 136.00, 131,59, 128.50, 126.60, 121.64, 119.13,
118.02, 115.53, 111.14, 104.82, 54.70, 54.57, 52.64,
52.25, 24.82, 20.93, 19.65; MS (EI) m/z 338 (M1).
31P NMR (202.3 MHz, CDCl3) d 70.98; 1H NMR
(500 MHz, CDCl3) d 8.02 (br, 1H, NH), 7.51 (s, 1H,
H-6), 6.38 (dd, 1H, H-10, J1 -2 ˆ9.0, 5.5 Hz), 5.11 (m, 1H,
H-30), 4.26 (m, 1H, H-40), 3.90 (m, 2H, H-50), 3.76 (2 d's,
6H, 2£OCH3, JH-Pˆ13.5 Hz), 2.50 (m, 1H, H-20), 2.11 (m,
1H, H-20), 1.92 (s, 3H, CvCMe), 0.94 (9H, SiC(CH3)3),
0.14 (s, 6H, Si(CH3)2); 13C NMR (125.7 MHz, CDCl3) d
163.21, 149.99, 135.02, 111.21, 85.81 (d, JC-Pˆ4.5 Hz),
84.65, 79.29, 63.39, 54.74, 39.22 (d, JC-Pˆ5.53 Hz),
29.70, 25.76, 18.18, 12.33, 25.56, 25.59; MS (ES) m/zˆ
479 (M2H), 959 (M1M2H).
0
0
2-Benzyl-1,2,3,4-tetrahydro-9H-pyrido[3,4-b]indole-2-
imidazole 25
2-Benzyl-3-(methoxycarbonyl)-1,2,3,4-tetrahydro-9H-
pyrido[3,4-b]indole-2-imidazole 22
Decarbomethoxy 25 was prepared using the same procedure
as described for the preparation of 22.
Nb-Benzyl-l-tryptophan methyl ester 20 (310 mg,
1.0 mmol), imidazole-2-carboxylic aldehyde 21 (960 mg,
1.0 mmol) and p-toluene sulfonic acid (380 g, 0.2 mmol)
were dissolved in benzene (50 ml). The mixture was then
brought to re¯ux using Dean±Stark water trap to remove
water. After 18 h, the solvent was removed in vacuo. The
residue was puri®ed by ¯ash chromatography to furnish the
desired product 22 as a yellow solid (251 mg, 65%).
1H NMR (500 MHz, CD2Cl2) d 9.37 (br, 1H, NH of imida-
zole), 8.82 (br, 1H, NH of indole), 7.48 (d, 1H, H-4,
J4-5ˆ7.5 Hz), 7.25 (d, 1H, H-7, J7-6ˆ7.5 Hz), 7.05±7.12
(m, 2H, H-5, H-6), 6.94 (2 s's, 2H of imidazole), 5.22 (s,
1H, (NH)NvCCHN(i-Pr)), 3.19 (m, 1H, CH2CH2N(i-Pr),
2.94 (m, 1H, NCH(CH3)2), 2.73±2.83 (m, 3H, CH2CH2N(i-
Pr)), 1.19 (d, 3H, NCH(CH3)2, Jˆ6.5 Hz), 1.04 (d, 3H,
NCH(CH3)2, Jˆ6.5 Hz); 13C NMR (125.7 MHz, CD2Cl2)
d 149.04, 136.32, 132.48, 128.17, 126.57, 121.38, 118.70,
117.99, 115.44, 110.93, 108.15, 55.64, 50.57, 41.59, 21.63,
14.39; MS (EI) m/z 280 (100%, M1).
1H NMR (500 MHz, CD2Cl2) d 9.91 (br, 1H, NH of indole),
7.50 (d, 1H, H-4, J4-5ˆ7.5 Hz), 7.28±7.36 (m, 5H, H of
phenyl), 7.22 (d, 1H, H-7, J7-6ˆ7.0 Hz), 7.01±7.10 (m,
2H, H-5, H-6), 6.93 (s, 2H, H of imidazole), 5.60 (s, 1H,
(NH)NvCCHN(Bn)), 3.98±4.02 (m, 2H, CHCOOCH3, 1H
of CH2Ph), 3.86 (d, 1H, 1H of CH2Ph, Jˆ14.0 Hz), 3.70 (s,
3H, COOCH3), 3.21 (m, 2H, CH2CHCOOCH3); 13C NMR
(125.7 MHz, CD2Cl2) d 173.24, 148.67, 139.00, 137.25,
131.98, 128.89, 127.74, 126.98, 121.95, 119.39, 118.35,
111.60, 106.04, 60.61, 57.76, 55.94, 52.02, 51.20, 23.54,
Cyclic phosphite monoesters 26, 27, 28, 29
To a solution of methyl dichlorophosphite (4.1 ml, 0.044
mmol) in dry dichloromethane (0.45 ml) in a NMR tube at
08C was added slowly a solution of 22 (17 mg, 0.044 mmol)

4364
Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
dissolved in dichloromethane (0.25 ml) containing triethyl-
amine (12.8 ml, 0.092 mmol) under argon. The NMR tube
was then sealed and allowed to stand at room temperature
until 31P NMR showed the disappearance of the resonance at
180 ppm and the formation of the new resonances at around
88 and 91 ppm (10:1). The crude 26 was used in the next
reaction without further puri®cation.
TBDMS-thymidine (19 mg, 0.054 mmol) in dry dichloro-
methane (0.4 ml) containing triethylamine (7.9 ml,
0.056 mmol) was added slowly at 08C, the NMR tube was
sealed and allowed to stand at room temperature. After 31P
NMR showing the formation of a major resonance at
87.0 ppm,
a
solution of 30-O-TBDMS-thymidine6
(19.9 mg, 0.056 mmol) in dry dichloromethane (0.4 ml)
was added to the reaction mixture at 08C. The NMR tube
was sealed and warmed up slowly. 31P NMR indicated
the formation of resonance at 142 ppm within 10 min.
Beaucage's reagent (12 mg, 0.061 mmol) was then added
to the mixture, a major phosphorus resonance at 61 ppm
was formed within 5 min. The reaction mixture was diluted
with dichloromethane and washed with water and brine,
dried over magnesium sulfate. Puri®cation by ¯ash chroma-
tography furnished the desired product 39 (25 mg, 42.3%)
as a white solid.
Similarly, compounds 27, 28 and 29 were prepared from 23,
24 and 25, respectively.
Phosphorothioates 34 and 35
To the previous NMR tube containing 26 was added 50-O-
TBDMS-thymidine (47 mg, 0.132 mmol) in dry dichloro-
methane (0.5 ml) at room temperature. The NMR tube
was then sealed. 31P NMR indicated the formation of new
resonance at 142.1 ppm immediately. Upon heating at 558C
overnight, peaks at 142.1 ppm shifted to 140 ppm.
Beaucage's reagent (11 mg, 0.053 mmol) was added to the
reaction mixture. After 10 min, the reaction mixture was
diluted with dichloromethane and washed with water
and brine. Puri®cation by ¯ash chromatography afforded
34 (16 mg, 46.6%) and 35 (5 mg, 23.6%), both as white
solids.
31P NMR (202.3 MHz, CDCl3) d 61.2 ppm; 1H NMR
(500 MHz, CDCl3) d 9.77 (br, 1H, NH of imidazole), 8.69
(2 br's, 2H, NH of thymidines), 8.05 (d, 1H, H-4 of indole,
J4-5ˆ8.0 Hz), 7.55 (d, 1H, H-7 of indole, J7-6ˆ7.5 Hz); 7.46
(2 s's, CvCH of thymidines); 7.21±7.36 (m, 7H, H-5, H-6
of indole and benzyl aromatic protons), 6.98 (s, 1H, H of
imidazole), 6.77 (s, 1H, H of imidazole), 6.21 (t, 1H, H-10),
5.51 (s, 1H, NCHimidazole), 5.42 (m, 1H, H-10), 5.08 (m,
1H, H-30), 4.17±4.22 (2H, H-30 and CH2Ph), 4.02 (m, 1H,
CH2CHCOOCH3), 3.51±4.04 (m, 10H, H-40 and H-50 of
thymidines, CH2Ph, COOCH3), 3.08±3.16 (m, AB of
ABX system, CH2CHCOOCH3), 1.98±2.17 (m, 3H,
H±20), 1.89 (2s's, 6H, 2£CH3CvC), 1.74 (m, 1H, H-20),
0.81 (2s's, 18H, 2£Si(CH3)2C(CH3)3), 20.03 (4s's, 12H,
2£Si(CH3)2C(CH3)3; 13C NMR (125.7 MHz, CDCl3) d
172.49, 163.47, 150.13, 150.01, 147.55, 138.72, 137.85,
137.78, 135.56, 134.66, 132.69, 132.65, 129.03, 128.77,
127.52, 127.45, 123.94, 122.82, 118.94, 116.86, 116.18,
116.12, 114.73, 111.35, 111.22, 85.20 (d, C-40,
When 27, 28 or 29 were used instead of 26, similar results
were obtained.
Dithymidine phosphorothioate 34: 31P NMR (202.3 MHz,
CDCl3) d 68.03; H NMR (500 MHz, CDCl3) d 8.87 (br,
1
2H, NH of thymidines), 7.46 (2H, H-6 of thymidines), 6.36
(m, 2H, H-10), 5.09 (m, 2H, H-30), 4.24 (m, 2H, H-40), 3.88
(m, 4H, H-50), 3.76 (d, 3H, OCH3, JP-Hˆ13.5 Hz), 3.52 (m,
2H, H-20), 2.11 (m, 2H, H-20), 1.88 (s, 6H, 2£CH3), 0.92
(2S's, 18H, 2£Si(C(CH3)3), 0.12 (12H, 2£Si(CH3)2); 13C
NMR (125.7 MHz, CD2Cl2) d 163.62, 150.33, 134.92,
111.25, 85.68, 84.62 (d, JC-Pˆ4.5 Hz), 79.67, 63.34,
54.80, 39.15 (d, JC-Pˆ4.7 Hz), 25.91, 25.62, 18.33, 12.52,
23.61, 25.39, 25.46; MS (CI) m/z 804 (M11)1.
3
3JC-Pˆ5.4 Hz), 85.13, 84.82 (d, C-40; JC-Pˆ10.1 Hz),
2
84.67, 79.61 (d, C-30, JC-Pˆ2.77 Hz), 71.85, 67.56 (d,
JC-Pˆ7.29 Hz), 63.14, 55.87, 55.03, 52.26, 52.10, 40.48,
38.98 (PCH2CH3, JC-Pˆ1.1 Hz), 25.88, 25.80, 25.63,
20.33, 18.15, 17.79, 12.50, 12.48, 24.67, 24.88, 25.54,
25.56; HRMS (EI) C55H75N8O12PSSi2 calcd: 1158.4501,
found: 1158.4505.
Dimethoxy phosphorothioate 35: 31P NMR (202.3 MHz,
CDCl3) d 70.98; H NMR (500 MHz, CDCl3) d 8.02 (br,
1
1H, NH), 7.51 (s, 1H, H-6), 6.38 (dd, 1H, H-10,
0
0
0
J1 -2 ˆ9.0 Hz, 5.5 Hz), 5.11 (m, 1H, H-3 ), 4.26 (m, 1H,
H-40), 3.90 (m, 2H, H-50), 3.76 (2 d's, 6H, 2£OCH3,
JH-Pˆ13.5 Hz), 2.50 (m, 1H, H-20), 2.11 (m, 1H, H-20),
1.92 (s, 3H, CvCMe), 0.94 (9H, SiC(CH3)3), 0.14 (s, 6H,
Si(CH3)2); 13C NMR (125.7 MHz, CDCl3) d 163.21, 149.99,
135.02, 111.21, 85.81 (d, JC-Pˆ4.5 Hz), 84.65, 79.29, 63.39,
54.74, 39.22 (d, JC-Pˆ5.5 Hz), 29.70, 25.76, 18.18, 12.33,
25.56, 25.59; MS (ES) m/zˆ479 (M2H), 959
(M1M2H).
Alkyl phosphinate 42
To
a solution of dichloroethyl phosphine (12.5 ml,
0.11 mmol) in dry dichloromethane (0.20 ml) in a scrupu-
lously dried NMR tube at 2788C was added slowly a solu-
tion of 22 (44 mg, 0.11 mmol) in dry dichloromethane
containing triethylamine (35 ml, 0.24 mmol). The NMR
was then sealed. When 31P NMR showed resonance at
around 75 ppm, 50-O-TBDMS thymidine (49 mg, 0.13
mmol) in dichloromethane (0.5 ml) was added to the
NMR tube at 08C. The reaction mixture was allowed to
warm up to room temperature and stand for half an hour.
The reaction mixture was diluted with dichloromethane,
washed with water and brine. Puri®cation by ¯ash column
chromatography afforded the desired product 42 (50 mg,
57%) as a white solid.
Indolophosphorothioate 39
To a cooled solution of phosphorus trichloride (4.43 ml,
0.051 mmol) in dry dichloromethane in a NMR tube at
08C was added slowly a solution of 22 (19.6 mg,
0.051 mmol) in dichloromethane containing triethylamine
(15.6 ml, 0.11 mmol). The NMR tube was sealed. 31P
NMR indicated the formation of peaks at 90.9 and
89.6 ppm in a ratio of 1 to 11. A solution of 50-O-
31P NMR (202.3 MHz, CDCl3) d 142.70; 1H NMR

Y. Lu, G. Just / Tetrahedron 56 (2000) 4355±4365
4365
2. Stec, W. J.; Wilk, A. Angew. Chem. Int. Ed. Engl. 1994, 33, 709.
3. Stec, W. J.; Grajkowski, A.; Kobylanska, A.; Karwowski, B.;
Koziolkiewicz, M.; Misiura, K.; Okruszek, A.; Wilk, A.; Guga, P.;
Boczkowska, M. J. Am. Chem. Soc. 1995, 117, 12019.
4. (a) Xin, Z.; Just, G. Tetrahedron Lett. 1996, 37, 969. (b) Jin, Y.;
Biancotto, G.; Just, G. Tetrahedron Lett. 1996, 37, 973. (c)
Marsault, E.; Just, G. Tetrahedron Lett. 1996, 37, 977. (d)
Marsault, E.; Just, G. Tetrahedron 1997, 53, 16 945. (e) Wang,
J.; Just, G. Tetrahedron Lett. 1997, 38, 705. (f) Wang, J.; Just, G.
Tetrahedron Lett. 1997, 38, 3797. (g) Jin, Y.; Just, G. J. Org.
Chem. 1998, 63, 3647.
(500 MHz, CDCl3) d 9.35 (br, 1H, NH of imidazole), 8.27
(br, 1H, NH of thymidine), 7.83 (d, 1H, H-4 of indole,
J4-5ˆ8.0 Hz), 7.52 (d, 1H, H-7 of indole, J7-6ˆ7.5 Hz),
7.15±7.34 (m, 7H, H-5, H-6 of indole and benzyl aromatic
protons), 6.95 (s, 1H, H of imidazole), 6.89 (s, 1H, H of
imidazole), 6.09 (t, 1H, H-10), 5.40 (s, 1H, NCHimidazole),
4.45 (m, 1H, H-30), 4.08 (m, 1H, H-40), 3.84±3.95 (m, 3H,
H-50, CH2CHCOOCH3, CH2Ph), 3.68±3.73 (m, 5H, H- 50,
CH2Ph, COOCH3), 3.01±3.10 (m, AB of ABX system,
CH2CHCOOCH3), 2.44±2.49 (m, 1H, H±20), 2.00 (m, 1H,
H-20), 1.79±1.87 (m, 5H, CH3CvC, PCH2CH3), 0.95±1.01
(m, 3H, PCH2CH3), 0.91 (s, 9H, Si(CH3)2C(CH3)3), 0.07 (s,
5. Saulnier, M. G.; Gribble, G. W. J. Org. Chem. 1982, 47, 757.
6. Xin, Zhili, Masters thesis, McGill University, 1994.
7. (a) Huang, Y.; Arif, A. M.; Bentrude, W. G. J. Org. Chem.
1993, 58, 6235. (b) Huang, Y.; Yu, J.; Bentrude, W. G. J. Org.
Chem. 1995, 60, 4767.
3H, Si(CH3)2C(CH3)3), 0.06 (s, 3H, Si(CH3)2C(CH3)3); 13
C
NMR (125.7 MHz, CDCl3) d 172.64, 163.38, 149.84,
147.96, 138.94, 135.47, 129.95, 128.63, 128.51, 128.28,
127.50, 122.65, 120.97, 118.83, 115.90, 114.65, 110.70,
3
86.23 (d, C-40, JC-Pˆ8.29 Hz), 84.98, 79.47 (d, C-30,
8. (a) Each structure was built up ®rst and pre-minimized. The
Monte-Carlo search was followed to ®nd the global minimum of
each structure. The force ®eld used for the calculation was
AMBER and the minimization method was TNCG. (b) Conformer
17b is about 4 kcal/mol more stable than conformer 17a.
9. (a) Iyer, R. P.; Egan, W.; Regan, J. B.; Beaucage, S. L. J. Am.
Chem. Soc. 1990, 112, 1253. (b) Iyer, R. P.; Phillips, L. R.; Egan,
W.; Regan, J. B.; Beaucage, S. L. J. Org. Chem. 1990, 55, 4693.
10. Kirk, K. L. J. Org. Chem. 1978, 43, 4381.
2JC-Pˆ18.35 Hz), 63.27, 57.20, 55.17, 55.06, 53.14, 52.01,
38.81 (PCH2CH3, JC-Pˆ4.53 Hz), 25.96, 24.22 (d, C-20,
3JC-Pˆ10.9 Hz), 21.39, 18.36, 7.50 (d, PCH2CH3, JC-P
ˆ
19.2 Hz), 25.34, 25.37, HRMS (EI) C41H53N6O7PSi
calcd: 800.3483, found: 800.3486.
Acknowledgements
11. Soerens, D.; Sandrin, J.; Ungemach, F.; Mokry, P.; Wu, G. S.;
Yamanaka, E.; Hutchins, L.; DiPierro, M.; Cook, J. M. J. Org.
Chem. 1979, 44, 535.
We wish to thank Natural Science and Engineering
Research Council of Canada for generous ®nancial support
and Mr Nadim Saadeh and Dr Orval Mamer, McGill
University biomedical mass spectroscopy unit, for recording
mass spectra.
12. Ungemach, F.; DiPierro, M.; Weber, R.; Cook, J. M. J. Org.
Chem. 1981, 46, 164.
13. Cox, E. D.; Hamaker, L. K.; Li, J.; Yu, P.; Czerwinski, K. M.;
Deng, L.; Bennett, D. W.; Cook, J. M.J. Org. Chem. 1997, 62, 44.
14. Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis, vol. 1,
Wiley: New York, 1967; p. 583.
References
1. Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 543.