Improved utility of photolabile solid phase synthesis supports for the synthesis of oligonucleotides containing 3'-hydroxyl termini

Article abstract of DOI:10.1021/jo951550w

Oligonucleotides are synthesized on, and cleaved from, a solid phase support (6) using the o-nitrobenzyl intramolecular photochemical redox reaction. The yields of isolated oligonucleotides relative to yields obtained using conventional hydrolytic cleavage vary between 67% and 82.5%. Synthesis of oligonucleotides using phosphoramidites that do not contain N-benzoyl protecting groups enables one to photolytically cleave the biopolymers in good yields using a commonly available UV irradiation source. Tritium labeling indicates that less than 3% thymidine·thymidine photodimers are formed during photolytic cleavage of polythymidylates from 6 using a transilluminator. No UV-induced damage is detected via HPLC analysis of enzymatically digested oligonucleotides that were obtained following photolytic cleavage from 6.

Full text of DOI:10.1021/jo951550w

J . Org. Chem. 1996, 61, 525-529
525
Im p r oved Utility of P h otola bile Solid P h a se Syn th esis Su p p or ts
for th e Syn th esis of Oligon u cleotid es Con ta in in g 3′-Hyd r oxyl
Ter m in i
Hariharan Venkatesan and Marc M. Greenberg*
Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523
Received August 22, 1995X
Oligonucleotides are synthesized on, and cleaved from, a solid phase support (6) using the
o-nitrobenzyl intramolecular photochemical redox reaction. The yields of isolated oligonucleotides
relative to yields obtained using conventional hydrolytic cleavage vary between 67% and 82.5%.
Synthesis of oligonucleotides using phosphoramidites that do not contain N-benzoyl protecting
groups enables one to photolytically cleave the biopolymers in good yields using a commonly available
UV irradiation source. Tritium labeling indicates that less than 3% thymidine‚thymidine
photodimers are formed during photolytic cleavage of polythymidylates from 6 using a transillu-
minator. No UV-induced damage is detected via HPLC analysis of enzymatically digested
oligonucleotides that were obtained following photolytic cleavage from 6.
The o-nitrobenzyl photoredox process (eq 1) has been
employed in a variety of synthetic strategies and as a
tool for probing biological processes.1 We have utilized
this venerable reaction to construct solid phase oligo-
nucleotide synthesis supports that enable one to cleave
the biopolymers from their solid supports at room tem-
perature under neutral conditions. The supports are
compatible with commercially available oligonucleotide
synthesis reagents, and do not require altering of stan-
dard automated synthesis cycles. Elimination of NH4-
OH from the cleavage process enables one to release
protected oligonucleotides containing a single functional
group at their 3′ termini that is suitable for conjugation
(e.g., 1).2 Other supports (2, 3) yield biopolymers con-
taining 3′-hydroxy termini and are useful for site-specific
incorporation of alkaline labile nucleosides (e.g., 4, 5) in
oligonucleotides, which are in turn amenable to enzy-
matic manipulation.3,4
photolytic cleavage of the biopolymers from the solid
phase supports using light sources, such as transillumi-
nators, which are routinely available in chemistry and
biology laboratories. Good yields of polythymidylates are
obtained from 1 using a transilluminator which emits at
Despite the general utility of o-nitrobenzyl supports 2
and 3, there is need for improvement. Carbonate support
3 produces moderate yields (61%) of oligonucleotides after
9 h of band pass (λmax ) 400 nm) photolysis. Seven
percent of the contiguous thymidines present are trans-
formed into thymine‚thymine photodimers under these
photolysis conditions.3 Photolabile supports that are
more efficient are needed in order to reduce this type of
damage. It is also highly desirable to carry out the
365 nm. However, similar treatment of heteropolymers5
containing the most commonly used commercially avail-
able phosphoramidite protecting groups yields less than
15% of isolable, deprotected oligonucleotides. Similar
results were obtained using a Rayonet photoreactor
equipped with lamps exhibiting a maximum output at
350 nm. We now report a photolabile solid phase
synthesis support (6) that releases oligonucleotides con-
X Abstract published in Advance ACS Abstracts, J anuary 15, 1996.
(1) (a) Pillai, V. N. R. In Organic Photochemistry, Vol. 9; Padwa,
A., Ed.; Marcel Dekker: New York, 1987. (b) Corrie, J . E. T.; Trentham,
D. R. In Bioorganic Photochemistry; Morrison, H., Ed.; J ohn Wiley:
New York, 1993; Vol. 2, pp 243-305.
(2) Yoo, D. J .; Greenberg, M. M. J . Org. Chem. 1995, 60, 3358.
(3) Greenberg, M. M.; Gilmore, J . L. J . Org. Chem. 1994, 59, 746.
(4) (a) Matray, T. J .; Greenberg, M. M. J . Am. Chem. Soc. 1994,
116, 6931. (b) Barvian, M. R.; Greenberg, M. M. J . Am. Chem. Soc.
1995, 117, 8291.
(5) A heteropolymer is defined as an oligonucleotide containing
thymidine and other nucleosides.
0022-3263/96/1961-0525$12.00/0 © 1996 American Chemical Society

526 J . Org. Chem., Vol. 61, No. 2, 1996
Venkatesan and Greenberg
Sch em e 1a
taining 3′-hydroxy termini, which is more effective than
2 and 3. Using 6 as a vehicle, we also report: (1) an
improved synthesis of the veratrole derivative, which is
the reactive chromophore in 6 and other photolabile solid
phase synthesis supports2 and (2) a strategy that enables
one to synthesize heteropolymers from 6 and other
photolabile solid phase synthesis supports, using conve-
nient sources of light, such as a transilluminator.
Design a n d Syn th esis of 6. Solid phase support 6
contains a carbonate linker between the o-nitrobenzyl
chromophore and the nucleoside component. This fea-
ture was incorporated in 6, because prior work with 2
established that photolabile supports containing carbon-
ate linkages between the 3′-terminal nucleotide and the
photolabile moiety are readily synthesizable.3 Similarly,
the increased photochemical efficiency observed with 1
attributable to alkoxyl substitution of the o-nitrobenzyl
group led us to incorporate this feature in 6 as well.2
Finally, we found that the three-carbon linker between
the o-nitrobenzyl group and the solid support (such as is
found in 1) is prone to elimination during synthesis of
the respective activated esters (e.g., 8) that are used to
load the solid phase supports. This limits the range of
reactions which can be carried out in the presence of this
side chain. Consequently, 6 was designed so as to contain
a four-carbon linker between the aromatic chromophore
and the solid phase support.
a
Key: (a) phosgene (1 M) in toluene, THF, 25 °C; (b) NaH,
THF, 25 °C; (c) CsF, DMF, 35 °C; (d) 2,4,5-trichlorophenol, DCC,
CH2Cl2, 25 °C; (e) LCAA-CPG, DMF.
Sch em e 2a
We routinely prepare modified solid phase oligonucle-
otide synthesis supports by carrying out the loading of
the support by an activated ester as the last step in the
process. This strategy minimizes the introduction of
impurities to the support (Scheme 1). Because activated
esters (e.g., 2,4,5-trichlorophenyl, 8) are unstable to the
conditions needed to effect coupling of the chloroformate
of 12 to the nucleoside, the choice of the carboxylate
protecting group in 12 was crucial. It was necessary that
the ester be cleavable under conditions which would not
cleave the dimethoxytrityl group (nonacidic) or hydrolyze
the carbonate (nonnucleophilic). The (trimethylsilyl)-
ethyl protecting group was chosen, because it could be
removed under anhydrous conditions using CsF.
The requisite alcohol (12) was obtained in a simple,
albeit circuitous, manner (Scheme 2). Saponification and
subsequent esterification of the carboxylate was neces-
sary, because attempted alkylation of the phenol with
4-chlorobutyric acid yielded the respective benzoate of
4-hydroxybutyric acid. Presumably, this product arises
from lactonization of the acid in situ, followed by nucleo-
philic attack of the phenolate under the mildly basic
reaction conditions. Moreover, the (trimethylsilyl)ethyl
ester was unstable to electrophilic aromatic nitration
reaction conditions. Despite the necessary protecting
group manipulations, 12 was obtainable in multigram
quantities.
a
Key: (a) methyl 4-chlorobutyrate, K2CO3, n-Bu4N+I-, CH3CN,
reflux; (b) LiOH, dioxane, 25 °C; (c) fuming HNO3, AcOH, 25°C;
(d) 2-(trimethylsilyl)ethanol, DCC, CH2Cl2, 25°C; (e) NaBH4,
EtOH, 0 °C.
P h otoch em ica l Relea se of Oligon u cleotid es fr om
6. Photolytic cleavage yields of an eicosameric poly-
thymidylate (T20) were optimized separately using the
400 nm band pass filtered output of a high-pressure Hg/
Xe lamp and a transilluminator (λmax ) 365 nm). Pho-
tolytically cleaved oligonucleotides were subsequently
deprotected with NH4OH and isolated by denaturing gel
electrophoresis. Isolated yields in all experiments are
expressed as percent yields relative to those obtained for
the oligonucleotide without irradiation, but with identical
NH4OH cleavage. The yields presented are averages of
those obtained over three separate experiments and vary
by less than 15% of the average value. Maximum yields
of T20 (67%) were obtained in one-fourth the time (1.5 h)
needed to obtain comparable yields (within experimental
error) of the identical oligonucleotide from 2 using the
same light source (Figure 1). Prolonged irradiation of
oligonucleotides synthesized on 6 results in diminution
of isolated yields of biopolymers. This is attributed to
formation of alklaine labile lesions in the biopolymer.

Efficiency and Utility of Oligonucleotide Synthesis Supports
J . Org. Chem., Vol. 61, No. 2, 1996 527
isobutyryl group (dGiBu). We anticipated that the ben-
zamide groups were the most likely to be photoactive via
either direct absorption of the irradiation or energy
transfer within the protected oligonucleotide.
In order to test this hypothesis, an eicosamer (14)
containing T, dG, and dC in which the exocyclic amine
of dC was protected as its acetate (dCAc) was prepared
on 6. Standard isobutyryl-protected dG was utilized.
Fully deprotected 14 was obtained in 80.5% yield, fol-
lowing photolytic cleavage with a transilluminator for 2
h. Further confirmation of the detrimental effect of the
benzamide protecting groups was ascertained via isola-
tion of 13 in 67% yield using dCAc and N-phenoxyacetyl-
protected deoxyadenosine (dAPac).6 A comparable yield
(68.5%) of 13 was obtained using all commercially avail-
able “fast deprotecting” phosphoramidites.6,7
In tegr ity of th e P h otolytica lly Clea ved Oligo-
n u cleotid es. Enzymatic digestion of oligonucleotides
released from 1 and 3, following band pass filtered
photolysis, did not reveal the presence of any damaged
nucleosides.2,3 Oligonucleotide 13, prepared using fast-
deprotecting phosphoramidites and thymidine, was di-
gested using snake venom phosphodiesterase, and the
nucleotides released were dephosphorylated with calf
alkaline phosphatase (Figure 2).6,8 The area ratio of
nucleosides obtained from the photolytically (transillu-
minator, 2 h) cleaved oligonucleotide was identical (within
experimental error) to that measured from the sample
subjected directly to ammonolysis. The broad featureless
peak present in both samples (tR ∼22.5 min) is found in
a blank injection as well (data not shown).
Thymidine photodimer formation is a prevalent type
of damage induced via UV irradiation. Extrapolation
from previous experiments utilizing band pass filtered
photolysis (λmax ) 400 nm) suggests that less than 2% of
thymidine‚thymidine photodimers will be formed upon
cleavage from 6 under these conditions.3 There is no
previous data involving the transilluminator from which
to extrapolate. Consequently, thymidine photodimer
formation was assayed for by HPLC analysis of formic
acid cleaved material, as previously described.3,9 Detec-
tion sensitivity of the released thymine‚thymine dimers
was enhanced by incorporating tritiated thymidine (2.55
Ci/mol) at the 11th thymidine (5′ numbering) in T20.
After accounting for background tritium, less than 3%
of thymine‚thymine dimers (detection limits) are formed
during a 2 h photolysis.
Su m m a r y. Photolabile solid phase oligonucleotide
synthesis support 6 enables one to synthesize biopoly-
mers containing 3′-hydroxy termini in good yields, under
conditions which induce minimal amounts of UV photo-
damage. The yields obtained using 6 are superior than
those obtained from 3. Comparable yields of identical
oligonucleotides are obtained from 2 and 6. However,
F igu r e 1. Isolated yields of T20 as a function of irradiation
time. Irradiation source: transilluminator (365 nm), solid bar;
Hg/Xe (400 nm), striped bar. Photocleavage was carried out
on detritylated material, prior to NH4OH deprotection. Pho-
tolytically cleaved oligonucleotide was subsequently depro-
tected with NH4OH and isolated by denaturing gel electro-
phoresis. Isolated yields are expressed relative to yield of
oligonucleotide obtained without irradiation, but with identical
NH4OH cleavage.
These lesions give rise to shorter oligonucleotides via
strand breaks upon ammonolysis. The shorter fragments
are separated from the intact material upon purification.
Extrapolation of previous experiments using tritiated
oligonucleotides indicates that these photolysis conditions
will produce less than 2% thymidine photodimers.3
While a slightly longer (2 h) photolysis time was needed
to optimize yields of T20 using the transilluminator, the
yields (82.5%) were higher than those obtained using the
Hg/Xe lamp. The isolated yields of T20 and required
photolysis times from 6 are similar to those measured
for a 3′-alkyl carboxylic acid containing polythymidylates
from 1.2
As expected, photolysis with a transilluminator of a
protected eicosameric heteropolymer prepared on 6 yielded
negligible amounts (e15%) of product (13) following
ammonolysis and gel electrophoresis. The recurring
observation that irradiation sources which emit maxi-
mally in the 350-365 nm region could not be used to
prepare heteropolymers from 6, or other o-nitrobenzyl-
based supports, posed a significant limitation to the
application of this methodology. Considering the wide-
spread use of the o-nitrobenzyl photoredox reaction, we
suspected that the problem lay in the properties of the
protected oligonucleotides and not the actual photochemi-
cal reaction. Two observations suggested that the source
of low yields of 13 upon irradiation at 365 nm were due
to photochemistry associated with the protecting groups
on the nucleobases: (1) isolated yields of polythymidy-
lates (which do not contain nucleobase protecting groups)
are not strongly dependent upon wavelength and (2)
oligonucleotides whose nucleobases are deprotected are
unaffected by irradiation at 350-365 nm.3,4
(6) Schulhof, J . C.; Molko, D.; Teoule, R. Tetrahedron Lett. 1987,
28, 51.
(7) Similar observations and suggested solutions were reported while
this manuscript was under review. See: Pirrung, M. C.; Bradley, J .-
C. J . Org. Chem. 1995, 60, 6270.
(8) Eadie, J . S.; McBride, L. J .; Efcavitch, J . W.; Hoff, L. B.; Cathcart,
R. Anal. Biochem. 1987, 165, 442.
(9) (a) Cadet, J .; Voituriez, L.; Hruska, F. E.; Kan, L.-S.; de Leeuw,
F. A. A. M.; Altona, C. Can. J . Chem. 1985, 63, 2861. (b) Cadet, J .;
Voituriez, L.; Hahn, B.-S.; Wang, S. Y. J . Chromatogr. 1980, 195, 139.
The most commonly utilized commercially available
nucleobase protecting groups contain benzoyl groups on
deoxyadenosine (dABz) and deoxycytidine (dCBz). The
exocyclic amine of deoxyguanosine is protected by an

528 J . Org. Chem., Vol. 61, No. 2, 1996
Venkatesan and Greenberg
Exp er im en ta l Section
Gen er a l Meth od s. 1H NMR spectra were recorded at 300,
270, or 200 MHz. HPLC chromatography was carried out on
a Rainin Microsorb-MV C-18 (5 µm) column. Photolyses were
carried using a Oriel 1000 W high-pressure Hg/Xe lamp or a
UVP dual wavelength transilluminator. The transilluminator
contained 4 × 8 W bulbs. The band pass filter was from Oriel
(λmax ) 400 nm, no. 59820). [methyl-3H]Thymidine was
purchased from Amersham. Tritiated thymidine phosphora-
midite was prepared as described previously (2.54 Ci/mol).3
Enzymatic digests and polyacrylamide gel electrophoresis were
carried out as previously described.2,3
All reactions were run under a nitrogen atmosphere in oven-
dried glassware, unless specified otherwise. Acetonitrile, CH2-
Cl2, and DMF were freshly distilled from CaH2. THF was
freshly distilled from Na/benzophenone ketyl. Long chain
alkylamine controlled pore glass support (CPG) was purchased
from Sigma. Snake venom phosphodiesterase was obtained
from Boehringer-Mannheim. Calf intestine alkaline phos-
phatase was from New England Biolabs.
Oligonucleotides were synthesized as previously described.3
“Fast deprotecting” phosphoramidites were from Pharmacia.6
N-Acetyldeoxycytidine and all other phosphoramidites were
from Glen Research.
Gen er a l P r oced u r e for P h otolytic Clea va ge of Oligo-
n u cleotid es fr om Solid P h a se Su p p or ts Usin g a Tr a n -
sillu m in a tor . Resin (∼1.0 mg) was stirred in a Pyrex tube
containing 3 mL of a 9:1 mixture of CH3CN:H2O. The mixture
was sparged with N2 for ∼20 min before photolysis. The
volume above the solution was continuously purged with N2
during photolysis. The resin was filtered through a 0.45 µm
filter upon completion of the photolysis. The tube and filter
were washed with H2O (3 × 1 mL) and CH3CN (3 × 1 mL).
The filtrates were concentrated, combined, and subjected to
the appropriate deprotection and/or purification method.
P r ep a r a tion of 9. Methyl 4-chlorobutyrate (5.4 g, 39.5
mmol) was refluxed in CH3CN (80 mL) with vanillin (5 g, 33
mmol), K2CO3 (10.9 g, 79 mmol), and n-Bu4N+I- (2.43 g, 6.5
mmol) for 22 h. The reaction mixture was cooled and diluted
with ether (200 mL) and H2O (50 mL). The aqueous layer was
extracted with ether (2 × 100 mL). The combined organics
were washed with brine (50 mL) and dried over MgSO4. After
removing excess methyl 4-chlorobutyrate under vacuum, the
crude product was recrystallized from EtOAc to yield 6.6 g
(80%) of alkylated vanillin. The product was carried on to the
next step in the synthetic procedure without further purifica-
tion. Mp: 67-68.5 °C. 1H NMR (CDCl3): δ 9.76 (s, 1H), 7.36
(m, 2H), 6.94 (d, 1H, J ) 8 Hz), 4.1 (t, 2H, J ) 6 Hz), 3.86 (s,
1H), 3.64 (s, 1H), 2.52 (t, 2H, J ) 7 Hz), 2.2 (tt, 2H, J ) 7,7
Hz). 13C NMR (CDCl3): δ 190.6, 173.1, 153.6, 149.6, 129.9,
126.4, 111.4, 109.1, 67.6, 55.7, 51.4, 30.1, 24. IR (film): 3079,
2951, 2882, 1732, 1682, 1586, 1510, 1467, 1425, 1397, 1267,
1196, 1031, 867, 731, 654 cm-1
.
LiOH (1 M, 28.5 mL) was added to a solution of the above
alkylated vanillin (3 g, 11.9 mmol) in dioxane (90 mL). After
8 h of stirring at room temperature, the reaction mixture was
acidified to pH 3 with 10% HCl. Additional H2O (40 mL) was
added and the aqueous layer extracted with EtOAc (3 × 150
mL). The combined organics were washed with brine (50 mL)
and dried over MgSO4. Removal of solvent yielded 2.8 g (99%)
of 9, which was used without further purification. Mp: 155
°C. 1H NMR (CD3OD): δ 9.62 (s, 1H), 7.51 (d, 1H, J ) 8 Hz),
7.42 (s, 1H), 7.12 (d, 1H, J ) 8 Hz), 4.16 (t, 2H, J ) 7 Hz),
3.88 (s, 1H), 2.37 (t, 2H, J ) 7 Hz), 2.14 (tt, 2H, J ) 7, 7 Hz).
13C NMR (CD3OD): δ 182.7, 156.3, 151.7, 132.1, 128.5, 124.7,
113.8, 111.6, 70.8, 57.2, 35.8, 27.7. IR (film): 3584, 3520, 2957,
2922, 1712, 1694, 1673, 1585, 1556, 1361, 1315, 1273, 1261,
F igu r e 2. Base composition analysis of 13 via enzymatic
digestion. (A) Oligonucleotide cleaved using concentrated NH4-
OH. (B) Oligonucleotide cleaved photochemically.
significantly shorter irradiation times are required when
using 6. Utilization of commercially available phos-
phoramidites which do not contain N-benzoyl protecting
groups allows one to prepare oligonucleotides containing
all four common naturally occurring nucleosides, using
a UV irradiation source which is commonly available in
chemistry and biology laboratories. Finally, the utility
of 6 for the incorporation of alkaline labile nucleosides
has been demonstrated via the synthesis of oligonucle-
otides containing 4 and 5.10
1231, 1158, 1132, 1092, 933, 681, 654 cm-1
.
P r ep a r a tion of 11. A mixture of fuming nitric acid (7.5
mL) and glacial acetic acid (30 mL) was added dropwise to
crude 9 (3 g, 12.76 mmol). After stirring for 6 h, the reaction
mixture was poured into ice water (50 mL) and extracted with
EtOAc (3 × 150 mL). The combined organics were washed
with a saturated solution of NaHCO3 (ca u tion ) until all
effervescence ceased. The aqueous layer was acidified to pH
(10) (a) Greenberg, M. M.; Barvian, M. R.; Cook, G. P.; Matray, T.
J .; Venkatesan, H. J . Am. Chem. Soc. (submitted). (b) Matray, T. J .;
Greenberg, M. M. (unpublished results).

Efficiency and Utility of Oligonucleotide Synthesis Supports
J . Org. Chem., Vol. 61, No. 2, 1996 529
3 with 10% HCl (ca u tion ) and extracted with EtOAc (3 × 150
mL). The combined organics were washed with brine (75 mL)
and dried over MgSO4. Removal of solvent yielded 3.3 g (95%)
of 10 and its regioisomers in a ratio of 4:1:1.
6H), 3.49 (m, 2H), 2.55-2.14 (m, 6H), 1.66 (s, 3H), 0.99-0.94
(m, 2H), 0.02 (s, 9H). 13C NMR (CDCl3): δ 172.8, 163.3, 158.6,
154, 153.7, 150.1, 147.5, 143.9, 139.3, 135.1, 134.9, 129.89,
127.89, 127.87, 127.1, 125.9, 113.16, 111.56, 109.8, 109.4,
87.13, 84.12, 83.64, 79.18, 68.18, 66.6, 63.6, 62.6, 56.4, 55.1,
37.7, 30.5, 24.1, 17.15, 11.4, -1.7. IR (film): 3064, 2932, 1751,
1690, 1608, 1580, 1559, 1522, 1509, 1463, 1329, 1278, 1219,
2-(Trimethylsilyl)ethanol (1.65 g, 14 mmol) was added to
crude 10 (3.3 g, 11.7 mmol) in CH2Cl2 (65 mL) and EtOAc (10
mL). A solution of DCC (2.9 g, 14 mmol) in CH2Cl2 (5 mL)
was added dropwise, and the reaction mixture was stirred for
5 h at room temperature. The mixture was filtered to remove
DCU and poured into H2O (75 mL). The aqueous layer was
extracted with CH2Cl2 (3 × 100 mL). The combined organics
were washed with 10% HCl (25 mL) and brine (50 mL) and
dried over MgSO4. Flash chromatography (EtOAc:hexanes,
1:4) yielded 2.7 g (56%) of 11. The product was carried on to
the next step in the synthetic procedure without further
purification. 1H NMR (CDCl3): δ 10.41 (s, 1H), 7.58 (s, 1H),
7.38 (s, 1H), 4.2-4.1 (m, 4H), 3.97 (s, 3H), 2.53-2.48 (m, 2H),
2.23-2.14 (m, 2H), 0.99-0.92 (m, 2H), 0.01 (s, 9H). 13C NMR
(CDCl3): δ 187.5, 172.6, 153.3, 151.6, 143.6, 125.3, 109.7, 107.9,
68.5, 62.7, 56.4, 30.3, 23.9, 17.2, -1.7. IR (film): 2953, 1731,
1689, 1573, 1520, 1469, 1403, 1335, 1283, 1223, 1060, 837, 737
1154, 1081, 956, 910, 754, 730, 702, 647 cm-1
.
P r ep a r a tion of 8. CsF (0.18 g, 1.2 mmol), which was dried
over P2O5, was added to 7 (0.23 g, 0.24 mmol) in DMF (1.6
mL). After stirring at 35 °C for 36 h, EtOAc (40 mL) was
added. Water (10 mL) was added, and the aqueous layer was
extracted with EtOAc (3 × 40 mL). The combined organics
were washed with brine (20 mL) and dried over MgSO4. Flash
chromatography (EtOAc:hexanes:MeOH, 50:40:10) yielded
0.12 g (59%) of the carboxylic acid, which was carried on in
the synthetic sequence without further purification. 1H NMR
(CDCl3): δ 9.37 (s, 1H), 7.72 (s, 1H), 7.61 (s, 1H), 7.46-7.22
(m, 9H), 7.03 (s, 1H), 6.83 (d, 4H, J ) 8 Hz), 6.45 (dd, 1H, J )
5, 9 Hz), 5.54 (s, 2H), 5.38 (d, 1H, J ) 6 Hz), 4.23 (s, 1H),
4.14-4.06 (m, 2H), 3.93 (s, 3H), 3.76 (s, 6H), 3.52 (m, 2H),
2.61-2.12 (m, 6H), 1.36 (s, 3H). 13C NMR (CDCl3): δ 179,
164.24, 158.8, 154.22, 153.93, 150.55, 147.66, 144.1, 139.5,
135.49, 135.08, 135.13, 130.05, 128.07, 127.24, 126.15, 113.33,
111.68, 110.1, 109.69, 87.3, 84.37, 83.83, 79.3, 68.23, 66.72,
63.72, 56.55, 55.22, 37.97, 30.21, 24.04, 11.55. IR (film): 3178,
3006, 2933, 1748, 1699, 1652, 1608, 1580, 1558, 1522, 1509,
cm-1
.
P r ep a r a tion of 12. NaBH4 (0.3 g, 7.8 mmol) was added
to 11 (2.5 g, 6.5 mmol) in EtOH (54 mL) at 0 °C. The reaction
mixture was warmed to room temperature and stirred for 3
h. After quenching with a saturated solution of NH4Cl (2 mL),
EtOH was removed in vacuo. H2O (50 mL) was added, and
the aqueous layer was extracted with EtOAc (3 × 100 mL).
The combined organics were washed with brine (50 mL) and
dried over MgSO4. Flash chromatography (EtOAc:hexanes,
2:3) yielded 1.8 g (72%) of 12. Mp: 84 °C. 1H NMR (CDCl3):
δ 7.67 (s, 1H), 7.14 (s, 1H), 4.94 (d, 2H, J ) 6 Hz), 4.2-4.08
(m, 4H), 3.95 (s, 3H), 2.75 (t, 1H, J ) 7 Hz), 2.53 (t, 2H, J )
7 Hz), 2.18-2.11 (m, 2H), 1.0-0.93 (m, 2H), 0.02 (s, 9H). 13C
(CDCl3): δ 173, 154.3, 147.1, 139.5, 132.5, 111, 109.6, 68.34,
62.7, 62.6, 56.3, 30.7, 24.3, 17.3, -1.6. IR (film): 3533, 3101,
2953, 1731, 1615, 1577, 1519, 1466, 1435, 1386, 1325, 1273,
1215, 1170, 939, 756, 664 cm-1. Anal. Calcd for C17H27NO7-
Si: C, 52.97; H, 7.06; N, 3.63. Found: C, 53.17; H, 7.06; N,
3.69.
1445, 1371, 1252, 980, 829, 791, 648 cm-1
.
2,4,5-Trichlorophenol (53 mg, 0.27 mmol) was added to the
above carboxylic acid (0.15 g, 0.18 mmol) in CH2Cl2 (1 mL).
DCC (55 mg, 0.27 mmol) in CH2Cl2 (0.5 mL) was added
dropwise to the mixture. After stirring for 12 h at 25 °C, the
reaction mixture was diluted with CH2Cl2 (40 mL) and H2O
(10 mL) was added. The aqueous layer was extracted with
CH2Cl2 (3 × 40 mL). The combined organics were washed with
brine (20 mL) and dried over MgSO4. Flash chromatography
(EtOAc:hexanes, 3:2) yielded 0.14 g (78%) of 8. Mp: 97 °C.
1H NMR (CDCl3): δ 8.29 (s, 1H), 7.74 (s, 1H), 7.59 (s, 1H),
7.53 (s, 1H), 7.37-7.23 (m, 10H), 7.05 (s, 1H), 6.83 (d, 4H, J )
8 Hz), 6.49 (dd, 1H, J ) 5, 9 Hz), 5.56 (s, 2H), 5.39 (d, 1H, J
) 5 Hz), 4.22-4.17 (m, 3H), 3.97 (s, 3H), 3.53 (s, 6H), 3.53-
3.43 (m, 2H), 2.88 (t, 2H, J ) 7 Hz), 2.58-2.21 (m, 4H), 1.36
(s, 3H). 13C NMR (CDCl3): δ 207.95, 169.17, 163.20, 158.63,
154.04, 153.763, 150.55, 147.34, 145.53, 143.91, 139.24, 135.1,
134.84, 131.31, 130.86, 130.47, 129.86, 127.89, 127.11, 126.25,
125.88, 125.12, 113.16, 111.58, 109.83, 109.51, 87.14, 84.11,
83.65, 79.25, 67.85, 66.56, 63.57, 56.45, 55.08, 37.72, 30.19,
23.97, 11.41. IR (film): 3064, 2932, 1751, 1690, 1608, 1580,
1559, 1522, 1509, 1463, 1329, 1278, 1219, 1154, 1081, 956, 910,
754, 730, 702, 647 cm-1. HRMS FAB (M+): calcd 1033.1994,
found 1033.2004.
P r ep a r a tion of 7. Phosgene in toluene (1.2 mL, 1.9 M
solution) was added to 12 (0.15 g, 0.39 mmol) in THF (2 mL).
After the reaction mixture was stirred for 4 h, N2 was bubbled
through the solution for 1 h to remove excess phosgene. After
removal of the solvent in vacuo, an aliquot of the crude product
was analyzed by IR and 1H NMR. IR showed two carbonyl
stretches at 1778 and 1730 cm-1 1H NMR showed a shift of
.
the benzylic protons from 4.94 to 5.82 ppm. This reaction was
carried out on this scale in two separate flasks. The above
analytical methods indicated that each reaction was quantita-
tive. Significant reductions in yield of the subsequent coupling
reaction were observed when the chloroformate formation was
carried out on a larger scale. The sodium alkoxide of (dimethox-
ytrityl)thymidine (0.353 g, 0.65 mmol) in THF (2.3 mL)
prepared from NaH was added to the chloroformate (0.35 g,
0.76 mmol) in THF (2 mL), and the reaction mixture was
stirred for 2 h at 25 °C. The reaction was diluted with EtOAc
(40 mL), and water (10 mL) was added. The aqueous layer
was extracted with EtOAc (3 × 40 mL). The combined
organics were washed with brine (20 mL) and dried over
MgSO4. Flash chromatography (EtOAc:hexanes:MeOH, 50:
48:2) yielded 0.45 g (72%) of 7. The product was carried on to
the next step in the synthetic procedure without further
purification. Mp: 89 °C. 1H NMR (CDCl3): δ 8.67 (s, 1H),
7.72 (s, 1H), 7.6 (s, 1H), 7.37-7.22 (m, 9H), 7.02 (s, 1H), 6.84
(d, 4H, J ) 8 Hz), 6.49 (dd, 1H, J ) 5, 9 Hz), 5.55 (s, 2H), 5.39
(d, 1H, J ) 5 Hz), 4.22-4.09 (m, 5H), 3.96 (s, 3H), 3.77 (s,
Ack n ow led gm en t. We are grateful for support of
this research from the National Science Foundation
(CHE-9424040). Mass spectra were obtained on instru-
ments supported by the National Institutes of Health
shared instrumentation grant GM49631.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
compounds 7, 8, 9, and 11 (4 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J O951550W