The 3-(N-tert-butylcarboxamido)-1-propyl group as an attractive phosphate/thiophosphate protecting group for solid-phase oligodeoxyribonucleotide synthesis

Article abstract of DOI:10.1021/jo0258608

Among the various phosphate/thiophosphate protecting groups suitable for solid-phase oligonucleotide synthesis, the 3-(N-tert-butylcarboxamido)-1-propyl group is one of the most convenient, as it can be readily removed, as needed, under thermolytic conditions at neutral pH. The deprotection reaction proceeds rapidly (t_1/2 ～ 100 s) through an intramolecular cyclodeesterification reaction involving the amide function and the release of the phosphate/thiophosphate group as a 2-(tertbutylimino)tetrahydrofuran salt. Incorporation of the 3-(N-tert-butylcarboxamido)-1-propyl group into the deoxyribonucleoside phosphoramidites 1a-d is achieved using inexpensive raw materials. The coupling efficiency of 1a-d in the solid-phase synthesis of d(ATCCGTAGCTAAGGTCATGC) and its phosphorothioate analogue is comparable to that of commercial 2-cyanoethyl deoxyribonucleoside phosphoramidites. These oligonucleotides were phosphate/thiophosphate-deprotected within 30 min upon heating at 90 °C in Phosphate-Buffered Saline (PBS buffer, pH 7.2). Since no detectable nucleobase modification or significant phosphorothioate desulfurization occurs, the 3-(N-tert-butylcarboxamido)-1-propyl group represents an attractive alternative to the 2-cyanoethyl group toward the large-scale preparation of therapeutic oligonucleotides.

Full text of DOI:10.1021/jo0258608

Th e 3-(N-ter t-Bu tylca r boxa m id o)-1-p r op yl Gr ou p a s a n Attr a ctive
P h osp h a te/Th iop h osp h a te P r otectin g Gr ou p for Solid -P h a se
Oligod eoxyr ibon u cleotid e Syn th esis
Andrzej Wilk, Marcin K. Chmielewski, Andrzej Grajkowski, Lawrence R. Phillips,† and
Serge L. Beaucage*
Division of Therapeutic Proteins, Center for Biologics Evaluation and Research, Food and Drug
Administration, 8800 Rockville Pike, Bethesda, Maryland 20892, and Biological Testing Branch,
Developmental Therapeutics Program, National Cancer Institute, Frederick, Maryland 21701
beaucage@phosphoramidite.cber.nih.gov
Received April 25, 2002
Among the various phosphate/thiophosphate protecting groups suitable for solid-phase oligonu-
cleotide synthesis, the 3-(N-tert-butylcarboxamido)-1-propyl group is one of the most convenient,
as it can be readily removed, as needed, under thermolytic conditions at neutral pH. The deprotection
reaction proceeds rapidly (t1/2 ∼100 s) through an intramolecular cyclodeesterification reaction
involving the amide function and the release of the phosphate/thiophosphate group as a 2-(tert-
butylimino)tetrahydrofuran salt. Incorporation of the 3-(N-tert-butylcarboxamido)-1-propyl group
into the deoxyribonucleoside phosphoramidites 1a -d is achieved using inexpensive raw materials.
The coupling efficiency of 1a -d in the solid-phase synthesis of d(ATCCGTAGCTAAGGTCATGC)
and its phosphorothioate analogue is comparable to that of commercial 2-cyanoethyl deoxyribo-
nucleoside phosphoramidites. These oligonucleotides were phosphate/thiophosphate-deprotected
within 30 min upon heating at 90 °C in Phosphate-Buffered Saline (PBS buffer, pH 7.2). Since no
detectable nucleobase modification or significant phosphorothioate desulfurization occurs, the 3-(N-
tert-butylcarboxamido)-1-propyl group represents an attractive alternative to the 2-cyanoethyl group
toward the large-scale preparation of therapeutic oligonucleotides.
In tr od u ction
alkylating properties of acrylonitrile have been investi-
gated by us5 and by others.6 The extent to which
nucleobases are alkylated by acrylonitrile during nano-
mole- or micromole-scale oligonucleotide deprotection is
insignificant, presumably due to the low concentration
of acrylonitrile being generated. However, nucleobase
alkylation during millimole-scale oligonucleotide depro-
tection is a valid concern5 as this threatens the desired
biological activity of synthetically derived therapeutic
oligonucleotides against a variety of human diseases
including cancer and viral infections. To prevent the loss
of bioactivity, modifications to nucleobases during oligo-
nucleotide synthesis and deprotection must not occur.
Thus, synthesis of therapeutic oligonucleotides must
utilize phosphate/thiophosphate protecting groups that
will not expose nucleobases to alkylating or otherwise
modifying agents during oligonucleotide deprotection.
With the advent of deoxyribonucleoside phosphor-
amidites,1 rapid and highly efficient solid-phase synthe-
ses of DNA oligonucleotides2 are now available and have,
over the past two decades, revolutionized the biomedical
sciences. This technology has expanded oligonucleotide
syntheses for therapeutic indications3 from nanomole- to
millimole-scale. The magnitude of this up-scaling has
seriously challenged the suitability of 2-cyanoethyl deoxy-
ribonucleoside phosphoramidites for oligonucleotide syn-
theses, since acrylonitrile is produced as a side-product
during oligonucleotide deprotection.4 The potent DNA-
* To whom correspondence should be addressed. Tel: (301)-827-
5162. Fax: (301)-480-3256.
† National Cancer Institute.
(1) (a) Beaucage, S. L.; Caruthers, M. H. Tetrahedron Lett. 1981,
22, 1859-1862. (b) McBride, L. J .; Caruthers, M. H. Tetrahedron Lett.
1983, 24, 245-248. (c) Sinha, N. D.; Biernat, J .; McManus, J .; Ko¨ster,
H. Nucl. Acids Res. 1984, 12, 4539-4557. (d) Beaucage, S. L.; Iyer, R.
P. Tetrahedron 1992, 48, 2223-2311. (e) Beaucage, S. L. In Methods
in Molecular Biology, Vol. 20: Protocols for Oligonucleotides and
Analogs; Agrawal, S., Ed.; Humana Press: Totowa, 1993; pp 33-61.
(f) Beaucage, S. L.; Caruthers, M. H. In Bioorganic Chemistry: Nucleic
Acids; Hecht, S. M., Ed.; Oxford University Press: New York, 1996;
pp 36-74. (g) Beaucage, S. L.; Caruthers, M. H. In Current Protocols
in Nucleic Acid Chemistry; Beaucage, S. L., Bergstrom, D. E., Glick,
G. D., J ones, R. A, Eds.; J ohn Wiley & Sons: New York, 1999; p 3.3.1-
3.3.20.
(4) Tener, G. M. J . Am. Chem. Soc. 1961, 83, 159-168.
(5) (a) Wilk, A.; Grajkowski, A.; Phillips, L. R.; Beaucage, S. L. J .
Org. Chem. 1999, 64, 7515-7522. (b) Wilk, A.; Grajkowski, A.;
Srinivasachar, K.; Beaucage, S. L. Antisense Nucleic Acid Drug Dev.
1999, 9, 361-366.
(6) (a) Crippa, S.; Di Gennaro, P.; Lucini, R.; Orlandi, M.; Rindone,
B. Gazz. Chim. Ital. 1993, 123, 197-203. (b) Prokopczyk, B.; Bertinato,
P.; Hoffman, D. Carcinogenesis 1988, 9, 2125-2128. (c) Mag, M.;
Engels, J . W. Nucl. Acids Res. 1988, 16, 3525-3543. (d) Solomon, J .
J .; Cote, I. L.; Wortman, M.; Decker, K.; Segal, A. Chem.-Biol. Interact.
1984, 51, 167-190. (e) Ogilvie, K. K.; Beaucage, S. L. Nucl. Acids Res.
1979, 7, 805-823. (f) Chambers, R. W. Biochemistry 1965, 4, 219-
226.
(2) Adams, S. P.; Kavka, K. S.; Wykes, E. J .; Holder, S. B.; Galuppi,
G. R. J . Am. Chem. Soc. 1983, 105, 661-663.
(3) Guzaev, A. P.; Manoharan, M. J . Am. Chem. Soc. 2001, 123,
783-793.
10.1021/jo0258608 This article not subject to U.S. Copyright. Published 2002 American Chemical Society
Published on Web 08/15/2002
6430 J . Org. Chem. 2002, 67, 6430-6438

3-(N-tert-Butylcarboxamido)-1-propyl Protecting Group
Earlier, we reported the use of deoxyribonucleoside
phosphoramidites in the solid-phase synthesis of DNA
oligonucleotides having the 4-[N-methyl-N-(2,2,2-trifluo-
roacetyl)amino]butyl group as a phosphate/thiophosphate
protecting group.5 This protective group is easily removed
from oligonucleotides by simple treatment with concen-
trated ammonium hydroxide or by pressurized ammonia
gas.7 The nonmutagenic side products, 2,2,2-trifluoro-
acetamide and N-methylpyrrolidine, are formed during
the course of the deprotection reaction. Although these
phosphoramidites lead to alkylation-free DNA oligonucle-
otide products, 4-(N-methylamino)butan-1-ol, which is
required in the synthesis of the phosphoramidites, is not
commercially available and, consequently, must be syn-
thesized. Because of this added expenditure, the 4-[N-
methyl-N-(2,2,2-trifluoroacetyl)amino]butyl deoxyribo-
nucleoside phosphoramidites might not effectively compete
with the conventional but less desirable 2-cyanoethyl
deoxyribonucleoside phosphoramidites to produce oligo-
nucleotide drugs at the most affordable cost per dose.
phosphoramidites 1a -d . The usefulness of these phos-
phoramidites is demonstrated in the solid-phase synthe-
sis of d(ATCCGTAGCTAAGGTCATGC) and its phospho-
rothioate analogue. The rapid and facile removal of the
3-(N-tert-butylcarboxamido)-1-propyl phosphate/thiophos-
phate protecting group under thermolytic conditions at
neutral pH is shown, and the factors influencing depro-
tection kinetics under these conditions are addressed.
In an effort to reduce the cost of therapeutic oligo-
nucleotides, we initiated a search for inexpensive phos-
phate/thiophosphate protecting groups. Recently, we
described the use of the 4-oxopentyl group as a labile
phosphate/thiophosphate protecting group in the syn-
thesis of DNA oligonucleotides.8 The starting material
for this protecting group, 4-oxopentan-1-ol, is commer-
cially available and considerably less expensive than
4-[N-methyl-N-(2,2,2-trifluoroacetyl)amino]butan-1-ol.
DNA oligonucleotides are readily prepared from 4-oxo-
pentyl deoxyribonucleoside phosphoramidites and the
4-oxopentyl phosphate/thiophosphate protecting group is
cleanly removed under very mild conditions by treatment
with pressurized ammonia, methylamine gas, or concen-
trated ammonium hydroxide at ambient temperature.7,8
Interestingly, the 4-oxopentyl phosphate/thiophosphate
protecting group can also be cleaved from oligonucleotides
in less than 1 h (in the absence of gaseous or aqueous
amines) simply by heating at 90 °C in a neutral aqueous
buffer.8 In this context, the 2-(N-formyl-N-methylamino)-
ethyl group used to protect phosphate/thiophosphate
diesters in the solid-phase synthesis of oligodeoxyribo-
nucleotides is also removable under thermolytic condi-
tions at neutral pH.9 The 2-(N-formyl-N-methylamino)-
ethyl deoxyribonucleoside phosphoramidites are prepared
from inexpensive reagents and, hence, are suitable for
cost-efficient oligonucleotide syntheses. Since the 2-(N-
formyl-N-methylamino)ethyl phosphate/thiophosphate pro-
tecting group cannot be removed reliably by standard
ammonolysis, it is cleaved from oligonucleotides only
upon heating at 90 °C in an aqueous buffer (pH 7.0) over
a period of at least 3 h.9 Although the group is cleanly
removed from phosphate or phosphorothioate triesters,
its long deprotection time detracts from an otherwise
efficient process. We now propose the use of the 3-(N-
tert-butylcarboxamido)-1-propyl group for enhanced phos-
phate/thiophosphate protection, and report its incorpo-
ration into oligonucleotides via the deoxyribonucleoside
Resu lts a n d Discu ssion
A direct approach to the preparation of N-tert-butyl-
4-hydroxybutyramide (2) in essentially quantitative yields
involves heating γ-butyrolactone and tert-butylamine at
90 °C in a sealed glass container. Condensation of 2 with
O,O-diethyl chlorophosphate and triethylamine in ben-
zene affords O-[3-(N-tert-butylcarboxamido)-1-propyl]-
O,O-diethyl phosphate (3) as a simple phosphotriester
model (Scheme 1).
SCHEME 1
Removal of the 3-(N-tert-butylcarboxamido)-1-propyl
group from purified 3 is then monitored by 31P NMR
spectroscopy in D2O at 80 °C.10 Within 35 min under
these conditions, phosphotriester 3 (δP 0.8 ppm) is
efficiently converted to the diethyl phosphate salt 4 (δP
2.0 ppm) via an intramolecular cyclization of the amide
as shown in Figure 1. Our findings are consistent with
the known intramolecular cyclization of N-substituted
amides of ω-bromocarboxylic acids, as reported by Stirling
more than 40 years ago.11 The identity of the 2-(tert-
butylimino)tetrahydrofuran cation in 4 is confirmed
through its synthesis by reacting tert-butylamine with
4-bromobutyryl chloride, followed by heating the amide
product to ∼140 °C.11 The H and proton-decoupled 13C
1
NMR spectra of salt 4 and of 2-(tert-butylimino)tetrahy-
drofuran hydrobromide show diagnostic signals having
identical chemical shifts in D2O.12
(7) Boal, J . H.; Wilk, A.; Harindranath, N.; Max, E. E.; Kempe, T.;
Beaucage, S. L. Nucl. Acids Res. 1996, 24, 3115-3117.
(8) Wilk, A.; Chmielewski, M. K.; Grajkowski, A.; Phillips, L. R.;
Beaucage, S. L. Tetrahedron Lett. 2001, 42, 5635-5639.
(9) Grajkowski, A.; Wilk, A.; Chmielewski, M. K.; Phillips, L. R.;
Beaucage, S. L. Org. Lett. 2001, 3, 1287-1290.
(10) Data shown in Supporting Information.
(11) Stirling, C. J . M. J . Chem Soc. 1960, 255-262.
J . Org. Chem, Vol. 67, No. 18, 2002 6431

Wilk et al.
SCHEME 2a ,b
F IGURE 1. Proposed mechanism for the removal of the 3-(N-
tert-butylcarboxamido)-1-propyl phosphate protecting group
under thermolytic conditions at neutral pH.
The suitability of the 3-(N-tert-butylcarboxamido)-1-
propyl group for phosphate/thiophosphate protection in
DNA oligonucleotide syntheses is evaluated first on a
dinucleoside phosphate triester and its phosphorothioate
analogue. These are prepared from the reaction of 5′-O-
(4,4′-dimethoxytrityl)thymidine (5) with an equimolar
amount of tris(diethylamino)phosphine and diethylam-
monium tetrazolide in dry methylene chloride.13 The
resulting nucleoside 3′-O-phosphordiamidite intermedi-
ate 6 is not isolated, but is immediately condensed with
a slight excess (1.2 equiv) of amido alcohol 2 to afford
the desired deoxyribonucleoside phosphoramidite
7
(Scheme 2). The crude phosphoramidite is purified by
silica gel chromatography and, then, used in a manual
solid-phase synthesis of dinucleoside phosphotriesters 11
and 12 (Scheme 2) according to published procedures.9
The dinucleotides 11 and 12 are released from the solid
support upon treatment with pressurized methylamine
gas, and are then purified by reversed-phase (RP) HPLC.
Heating either 11 or 12 in 0.1 M triethylammonium
acetate (pH 7.0) at 90 °C leads to complete removal of
the 3-(N-tert-butylcarboxamido)-1-propyl group to afford
either the dinucleoside phosphodiester 16 or RP -17 and
SP -17, respectively, as the sole nucleic acid products. The
reaction half-time of the phosphate/thiophosphate depro-
tection is ∼100 s, as determined by monitoring 16 or 17
by RP-HPLC at various time points during the course of
the reaction (see Experimental Section). There is no
indication of diastereomeric bias in the removal of the
3-(N-tert-butylcarboxamido)-1-propyl group from either
phosphate or thiophosphate triesters under neutral
aqueous conditions at 90 °C. Indeed, the thiophosphate
deprotection kinetics of RP-HPLC purified RP -12 or SP -
12 are, under similar conditions, indistinguishable. As
expected, deprotection proceeds with retention of config-
uration at phosphorus with no significant desulfurization
of the phosphorothioate.9
a
Reaction conditions: (i) (Et2N)3P/diethylammonium tetra-
zolide/CH2Cl2, 25 °C, 30 min; (ii) 2 or N-tert-butyl-5-hydroxyval-
eramide or N-tert-butyl-3-hydroxypropionamide or N-tert-butyl-
2-hydroxyacetamide, 25 °C, 8 h followed by purification on silica
gel; (iii) T-LCAA-CPG/1H-tetrazole/MeCN; (iv) 0.02 M I2 in THF/
pyridine/water or 0.05 M 3H-1,2-benzodithiol-3-one 1,1-dioxide in
MeCN; (v) 3% TCA/CH2Cl2; (vi) MeNH2 gas (∼ 2 bar), 1 min; (vii)
b
RP-HPLC purification; (viii) 0.1 M TEAA, pH 7.0, 90 °C. Key:
Thy, thymin-1-yl; T, thymidine 3′-O-succinyl; LCAA, long chain
alkylamine; CPG, controlled-pore glass; TEAA, triethylammonium
acetate.
tatively proposed that a suitably located carbonyl group
relative to the phosphate/thiophosphate leaving group is
all that is necessary for deprotection. Consistent with this
postulate, the 2-[N-(2-fluoroacetyl)amino]-1-phenyleth-
yl,14 2-(N-acetylamino)ethyl,9 2-(N-acetyl-N-methylami-
no)ethyl,9 2-(N-formyl-N-methylamino)ethyl,9 2-(N,N-
dimethylaminocarbonyloxy)ethyl,9 and several 2-benz-
amidoethyl groups3 are all cleaved from respective phos-
phate/thiophosphate triesters under neutral aqueous
conditions, and all have a carbonyl group located at a
similar distance from the phosphate/thiophosphate leav-
ing group.
Since the cleavage of the 4-oxopentyl phosphate/
thiophosphate protecting group occurs under neutral
aqueous conditions at elevated temperature,8 it is ten-
(12) The following NMR signals are found in the spectrum of 4 and
that of 2-(tert-butylimino)tetrahydrofuran hydrobromide: 1H NMR (300
MHz, D2O), δ 1.28 (s, 9H), 2.21 (tt, J ) 7.3 and 8.4 Hz, 2H), 2.98 (t,
J ) 8.4 Hz, 2H), 4.79 (t, J ) 7.3 Hz, 2H). 13C NMR (75 MHz, D2O), δ
26.5, 32.4, 36.9, 62.4, 85.6, 185.8.
(13) Yamana, K.; Nishijima, Y.; Oka, A.; Nakano, H.; Sangen, O.;
Ozaki, H.; Shimidzu, T. Tetrahedron 1989, 45, 4135-4140. See also:
Wilk, A.; Srinivasachar, K.; Beaucage, S. L. J . Org. Chem. 1997, 20,
6712-6713.
(14) Wilk, A.; Grajkowski, A.; Phillips, L. R.; Beaucage, S. L. J . Am.
Chem. Soc. 2000, 122, 2149-2156.
6432 J . Org. Chem., Vol. 67, No. 18, 2002

3-(N-tert-Butylcarboxamido)-1-propyl Protecting Group
The nature of the spatial arrangement of the carbonyl
group for the removal of the 3-(N-tert-butylcarboxamido)-
1-propyl thiophosphate protecting group was investigated
further. Thus, dinucleoside phosphorothioates 13-15
carrying a 4-(N-tert-butylcarboxamido)-1-butyl group, a
2-(N-tert-butylcarboxamido)ethyl group, and a (N-tert-
butylcarboxamido)methyl group, respectively, were pre-
pared from the corresponding deoxyribonucleoside phos-
phoramidites 8-10 (Scheme 2) in a manner identical to
that used for the preparation of 12.
Heating 13 in 0.1 M triethylammonium acetate (pH
7.0) at 90 °C leads to its complete conversion to 17. The
half time of thiophosphate deprotection was determined
as described for 12 (vide supra) and was found to be 42
min. Thus, moving the carbonyl group one bond-length
away from the thiophosphate leaving group results in a
∼25-fold reduction in the deprotection rate as compared
to that determined for 12. However, when moving the
carbonyl group one bond-length closer to the thiophos-
phate leaving group (such as in 14), deprotection of the
2-(N-tert-butylcarboxamido)ethyl group is very sluggish
under the conditions used for deprotecting 12; the half-
time of thiophosphate deprotection is estimated to be in
excess of 65 h. Since an intramolecular cyclization of the
amide leading to the formation of a four-membered
iminolactone is very unlikely, a â-elimination mechanism
with simultaneous formation of N-tert-butylacrylamide
may be preferred in this case, as the formation of 17 is
pH-dependent. Indeed, heating 14 in 20% NH4OH (pH
12) at 90 °C produces 17 to the extent of 50% within 25
min, whereas heating 14 in 0.1 M triethylammonium
acetate (pH 4) at the same temperature generates 17 very
slowly (t1/2 > 100 h). These findings strongly suggest that
under thermolytic conditions at pH 7.0, the 2-(N-tert-
butylcarboxamido)ethyl group may predominantly be
removed from 14 via a â-elimination mechanism rather
than through a cyclodeesterification mechanism.
and facile process that strongly suggests its application
to solid-phase synthesis of oligodeoxyribonucleotides. Low
cost and ease of preparation of a particular amido alcohol
are major factors influencing its selection as a phosphate/
thiophosphate protecting group for an economical and
alkylation-free oligodeoxyribonucleotide synthesis. Con-
sidering all factors, the 3-(N-tert-butylcarboxamido)-1-
propyl group is currently the superior choice, and its
utility is now demonstrated for phosphate/thiophosphate
protection in the solid-phase synthesis of a 20-mer,
d(ATCCGTAGCTAAGGTCATGC), and its phosphorothio-
ate analogue.
Thus, reaction of N-tert-butyl-4-hydroxybutyramide (2)
with bis(N,N-diisopropylamino)chlorophosphine, which is
generated immediately prior to use from phosphorus
trichloride and N,N-diisopropylamine in dry benzene,
affords the O-[3-(N-tert-butylcarboxamido)-1-propyl]-
N,N,N′,N′-tetraisopropyl phosphordiamidite 22.
Heating 15 to effect thiophosphate deprotection is also
observed to proceed slowly (t1/2 > 16 h). This may be
expected because of the steric constraints associated with
intramolecular cyclization of the amide thiophosphate
protecting group to an imino-R-lactone. Thus, phosphate/
thiophosphate deprotection experiments performed with
the dinucleoside phosphotriesters 11-15 have collectively
shown a critical relationship between the deprotection
reaction rate and the distance separating the amidic
carbonyl from the phosphate/thiophosphate function.
We also investigated 3-(N-methylcarboxamido)-1-pro-
panol and 3-(N-isopropylcarboxamido)-1-propanol as po-
tential protecting groups for phosphates/thiophosphates
in solid-phase oligonucleotide synthesis. These amido
alcohols were prepared similarly as described in Scheme
1 and then incorporated into model dinucleoside phos-
phorothioate triesters 20 and 21, which were synthesized
from deoxyribonucleoside phosphoramidites 18 and 19,
respectively, in a manner identical to that shown in
Scheme 2. As in the case of 12, heating RP-HPLC purified
20 or 21 in 0.1 M triethylammonium acetate (pH 7.0) at
90 °C leads to 17 with a reaction half time of ∼300 s. On
the basis of RP-HPLC analysis, deprotection of 20 or 21
occurs without significant desulfurization of the phos-
phorothioate diester.9
Crude 22 may then be activated using 1H-tetrazole and
condensed with 5′-O-(4,4′-dimethoxytrityl)-N-protected
deoxyribonucleosides in dry dichloromethane to give the
corresponding deoxyribonucleoside phosphoramidites 1a-
d . These phosphoramidites were purified by chromatog-
raphy on silica gel and then used in the automated solid-
phase synthesis of d(ATCCGTAGCTAAGGTCATGC) and
its phosphorothioate analogue using standard synthesis
cycle parameters. Upon completion of the syntheses, each
oligonucleotide is treated with pressurized ammonia gas
(∼10 bar) for 10 h at 25 °C to effect both its release from
the CPG support and N-deprotection of the nucleobases.
The unmodified 20-mer is eluted from the CPG column
with PBS buffer, pH 7.2, and then heated at 90 °C for 30
min to remove the phosphate protecting groups.15 Using
polyacrylamide gel electrophoresis (PAGE), the coupling
efficiency of 1a -d in the preparation of the 20-mer can
be compared with that of commercial 2-cyanoethyl de-
oxyribonucleoside phosphoramidites (see Figure 2). The
(15) We noticed that the removal of the 3-(N-tert-butylcarboxamido)-
1-propyl phosphate protecting group from 3 under thermolytic condi-
tions resulted in an acidic solution. Although this pH drop from
neutrality did not affect the outcome of the deprotection reaction, it is
nonetheless recommended to use PBS buffer (pH 7.2) for the ther-
molytic cleavage of 3-(N-tert-butylcarboxamido)-1-propyl phosphate/
thiophosphate protecting group from DNA oligonucleotides to ensure
neutrality throughout the deprotection reaction.
Thiophosphate deprotection of the model dinucleoside
phosphorothioate triesters 12, 20, and 21 is an attractive
J . Org. Chem, Vol. 67, No. 18, 2002 6433

Wilk et al.
F IGURE 3. 121 MHz 31P NMR spectrum of d(APSTPSCPSCPS
PSTPSAPSGPSCPSTPSAPSAPSGPSGPSTPSCPS APSTPSGPSC) in D2O.
-
G
deuterated water indicated that ∼60% of the 3-(N-tert-
butylcarboxamido)-1-propyl thiophosphate protecting
groups are cleaved under these conditions, as estimated
by the ratio of the integrated signals corresponding to
phosphotriester (δP 64 ppm) and phosphodiester (δP 53
ppm), respectively.10 Complete thiophosphate deprotec-
tion is accomplished within 25 min upon heating the
oligonucleotide solution at 60 °C in the NMR probe (see
Figure 3). Desulfurized material is observed (δP ∼ 4 to
-4 ppm) in amounts comparable to that obtained when
using the 2-cyanoethyl group for thiophosphate protec-
tion.
F IGURE 2. Polyacrylamide gel electrophoresis analysis of
d(ATCCGTAGCTAAGGTCATGC) under denaturing condi-
tions (7 M urea, 1X TBE buffer, pH 8.3). Left lane: crude
oligomer synthesized from commercial 2-cyanoethyl deoxyri-
bonucleoside phosphoramidites and deprotected by treatment
with concentrated NH4OH for 10 h at 55 °C. Right lane: crude
oligomer synthesized from 1a -d and deprotected under pres-
surized ammonia gas (∼10 bar) for 10 h at 25 °C followed by
heating in PBS buffer (pH 7.2) for 30 min at 90 °C. Oligo-
nucleotides are visualized as blue bands upon staining the gel
with Stains-all. Bromophenol blue is used as a marker and
appears as a large band, in each lane, at the bottom of the
gel.
To further evaluate phosphorothioate desulfurization
in the context of large-scale oligonucleotide deprotection,
a solution of 0.5 M potassium O,O-diethylthiophosphate
and 0.75 M 2-(tert-butylimino)tetrahydrofuran hydrobro-
mide in 0.1 M triethylammonium acetate (pH 7.0) or PBS
buffer (pH 7.2) was heated for 1 h at 90 °C. Analysis of
the reaction mixture by 31P NMR did not show any
significant conversion of O,O-diethylthiophosphate (δP
∼55 ppm) to O,O-diethyl phosphate (δP ∼2 ppm).10
PAGE profiles of these two syntheses are very similar
in terms of product yields and distribution of shorter than
full-length oligonucleotides.
The likelihood of nucleobase modification under the
conditions used for chain assembly and complete oligo-
nucleotide deprotection when using the 3-(N-tert-butyl-
carboxamido)-1-propyl group for phosphate protection
deserves further evaluation. Thus, after subjecting the
crude deprotected oligonucleotide to snake venom phos-
phodiesterase and bacterial alkaline phosphatase, RP-
HPLC analysis of the digest indicates that the oligonu-
cleotide is completely converted to the expected nucleo-
sides, free of any detectable nucleobase modifications.10
The possibility of nucleobase modification occurring
under conditions mimicking large-scale oligonucleotide
deprotection was also investigated. Specifically, a solution
composed of thymidine, 2′-deoxycytidine, 2′-deoxyadenos-
ine, 2′-deoxyguanosine (125 mM each), and 1 M 2-(tert-
butylimino)tetrahydrofuran hydrobromide in 0.1 M tri-
ethylammonium acetate (pH 7.0) or PBS buffer (pH 7.2)
was heated to 90 °C and maintained at that temperature
for 1 h. RP-HPLC analysis of the reaction mixture shows
that each 2′-deoxyribonucleoside is unaffected under such
conditions and demonstrates the suitability of the 3-(N-
tert-butylcarboxamido)-1-propyl group for phosphate/
thiophosphate protection in large-scale oligonucleotide
syntheses.
Con clu sion
We have shown that the 3-(N-tert-butylcarboxamido)-
1-propyl group provides convenient phosphate/thiophos-
phate protection in solid-phase syntheses of oligonucleo-
tides and their phosphorothioated analogues. The pro-
tecting group is prepared from inexpensive and readily
available starting materials and is incorporated into the
deoxyribonucleoside phosphoramidites 1a -d in a very
straightforward manner. Small-scale (0.2 µmol) solid-
phase synthesis of oligodeoxyribonucleotides using either
1a -d or commercial 2-cyanoethyl deoxyribonucleoside
phosphoramidites produces unmodified and phospho-
rothioated oligonucleotides in similar yields and purity.
When a large-scale oligonucleotide deprotection reaction
is simulated, no detectable nucleobase modification and
no significant desulfurization of the phosphorothioates
are observed. The 2-(tert-butylimino)tetrahydrofuran side-
product that is generated during the removal of the 3-(N-
tert-butylcarboxamido)-1-propyl group is innocuous to-
ward the reaction mixture, as there are no unwanted
secondary reactions resulting from its presence. Thus, use
of the 3-(N-tert-butylcarboxamido)-1-propyl group for
phosphate/thiophosphate protection in the large-scale
preparation of therapeutic oligonucleotides is an obvious
The stability of the 3-(N-tert-butylcarboxamido)-1-
propyl phosphate/thiophosphate protecting group to pres-
surized ammonia gas under the conditions used for
deprotecting nucleobases and releasing oligonucleotides
from the CPG support (10 bar, 10 h, 25 °C) was evaluated
using the phosphorothioated 20-mer as a model. 31P NMR
analysis of the oligomer eluted from the CPG column with
6434 J . Org. Chem., Vol. 67, No. 18, 2002

3-(N-tert-Butylcarboxamido)-1-propyl Protecting Group
m/z 50 and 550 at the rate of 1.5 scan/s. Data were collected
between 2.5 and 17.8 min postinjection.
choice and therefore recommended to ensure economic
manufacture and optimal potency of the drugs.
N-ter t-Bu tyl-4-h yd r oxybu tyr a m id e (2). γ-Butyrolactone
(1.64 g, 19.0 mmol) and tert-butylamine (5.00 g, 68.4 mmol)
are placed in a 20 mL thick-walled glass ampule. The container
is flame-sealed and then heated at 90 °C for 3 days (Ca u tion !
Th e gla ss con ta in er m u st be sh ield ed a s p r essu r e d e-
velop s). The glass ampule is cooled to ambient temperature
and opened. The reaction mixture is concentrated under
reduced pressure until a crystalline material is obtained. The
solid is recrystallized from chloroform/hexane affording 2 (2.96
g, 18.6 mmol, 98%) as white crystals (mp 73 °C). 1H NMR (300
MHz, DMSO-d6): δ 1.23 (s, 9H), 1.59 (dt, J ) 6.6, 7.5 Hz, 2H),
2.04 (t, J ) 7.5 Hz, 2H), 3.34 (t, J ) 6.6 Hz, 2H), 3.38 (s, 1H),
7.35 (s, 1H). 13C NMR (75 MHz, DMSO-d6): δ 28.5, 28.7, 32.9,
49.8, 60.3, 171.8. EI-MS (70 eV): m/z (% relative abundance),
159 (M+, 4), 144 (7), 115 (31), 87 (11), 86 (14), 59 (19), 58 (100),
57 (29).
One may speculate that the 3-(N-tert-butylcarboxami-
do)-1-propyl group will find further application as a
phosphate/thiophosphate protecting group in the solid-
phase synthesis of ribonucleotides, or in the preparation
of phosphosugars, phosphopeptides, and inositol phos-
phates or thiophosphates. Currently, we are searching
for novel thermolytic groups that can be used for hydroxyl
protection in nucleic acid chemistry, the results of which
will be the subject of future reports.
Exp er im en ta l Section
Ma ter ia ls a n d Meth od s. Common chemicals and solvents
were purchased from commercial sources and used without
further purification. Lecture bottles of ammonia and methyl-
amine gases, anhydrous pyridine, benzene, δ-valerolactone,
γ-butyrolactone, 4-bromobutyryl chloride, aluminum chloride,
O,O-diethyl chlorophosphate, O,O-diethyl thiophosphate (po-
tassium salt), 3-hydroxypropionitrile, 1,1′-carbonyldiimidazole,
methyl glycolate, tert-butylamine, isopropylamine, tris(diethyl-
amino)phosphine, sublimed 1H-tetrazole, and Stains-all were
obtained commercially and used as received. Acetonitrile,
dichloromethane, N,N-diisopropylamine, triethylamine, 1,2-
dichloroethane, and petroleum ether were refluxed over cal-
cium hydride, distilled, and stored over 4 Å molecular sieves.
Phosphorus trichloride (Aldrich) was distilled immediately
prior to use. 4,4′-Dimethoxytrityl chloride and suitably pro-
tected 2′-deoxyribonucleosides were purchased from Chem-
Impex International and used without further purification.
Unprotected 2′-deoxynucleosides were obtained from Sigma
and used as received.
O-[3-[(N-ter t-Bu tyl)ca r boxa m id o]-1-p r op yl]-O,O-d ieth -
yl P h osp h a te (3). O,O-Diethyl chlorophosphate (180 mg, 1.04
mmol) is added to a suspension of (N-tert-butyl)-4-hydroxy-
butyramide (2, 160 mg, 1.00 mmol) in 1 mL of dry benzene:
triethylamine (1:1 v/v). The reaction mixture is stirred at
ambient temperature for 12 h. The reaction mixture is then
filtered to remove triethylamine hydrochloride and extracted
with water (3 mL). The organic phase is collected and
evaporated to dryness under reduced pressure. The residue is
dissolved in chloroform (0.5 mL) and loaded onto a silica gel
column (1.3 cm × 5 cm) preequilibrated in chloroform. The
product is eluted from the column using a gradient of methanol
(0-10% v/v) in chloroform. The desired fractions are collected
and concentrated under reduced pressure affording 210 mg of
3 (0.71 mmol, 71%) as a colorless viscous oil. 1H NMR (300
4
MHz, D2O): δ 1.10 (s, 9H), 1.14 (dt, J ) 7.1 Hz, J PH ) 1.0
Hz, 6H), 1.75 (tt, J ) 6.3, 7.1 Hz, 2H), 2.08 (t, J ) 7.1 Hz,
2H), 3.90 (dt, J ) J PH ) 6.3 Hz, 2H), 3.97 (dq, J ) 7.1 Hz,
Preparative chromatographic purifications were performed
on columns packed with Merck silica gel 60 (230-400 mesh),
whereas analytical thin-layer chromatography (TLC) was
conducted on 2.5 cm × 7.5 cm glass plates coated with a 0.25
mm thick layer of silica gel 60 F254 (Whatman).
J PH ) 8.1 Hz, 4H). 13C NMR (75 MHz, C6D6): δ 27.2 (d, J PC
6.4 Hz), 28.8, 33.1, 50.8, 63.5 (d, J PC ) 6.4 Hz), 64.9, 66.5 (d,
J PC ) 6.4 Hz), 170.9. 31P NMR (121 MHz): δ 1.44 (C6D6) or
0.69 (D2O).
)
2
1
NMR spectra were recorded at 7.05 T (300 MHz for H). 1H
2-(N-ter t-Bu tylim in o)tetr a h yd r ofu r a n Hyd r obr om id e.
This compound is prepared according to the procedure of
Sterling.11 A solution of tert-butylamine (2.0 g, 27 mmol) in
ether (10 mL) is added over a period of 15 min to a vigorously
stirred etheral solution of 4-bromobutyryl chloride (2.0 g, 11
mmol). The reaction mixture is then immediately washed with
water (3 × 20 mL), and the organic phase is dried over
anhydrous MgSO4 and evaporated to dryness under reduced
pressure. Upon further drying under high vacuum, the gummy
mass crystallized. Crude N-tert-butyl-4-bromobutyramide (1.00
g, 4.25 mmol) is heated in a flame-sealed glass ampule until
it melted at ∼140 °C. The melt resolidified within minutes to
give a crystalline product. Slow diffusion of ether into an
ethanol solution of the crude product gave pure 2-(N-tert-
butylimino)tetrahydrofuran hydrobromide as colorless crystals
(mp 174-175 °C) in essentially quantitative yield. 1H NMR
(300 MHz, D2O), δ 1.28 (s, 9H), 2.21 (tt, J ) 7.3 and 8.4 Hz,
2H), 2.98 (t, J ) 8.4 Hz, 2H), 4.79 (t, J ) 7.3 Hz, 2H). 13C
NMR (75 MHz, D2O), δ 26.5, 32.4, 36.9, 62.4, 85.6, 185.8.
N-ter t-Bu tyl-5-h yd r oxyva ler a m id e. This compound is
prepared according to a literature procedure.16 Aluminum
chloride (13.0 g, 97.5 mmol) is suspended in 1,2-dichloroethane
(50 mL) and cooled to 5 °C in an ice bath. Anhydrous
triethylamine (10.0 g, 98.8 mmol) in 1,2-dichloroethane (30
mL) is added to the stirred suspension. The reaction mixture
is left stirring for 15 min and then allowed to warm to room
temperature. A solution of δ-valerolactone (7.0 g, 70 mmol)
and tert-butylamine (5.7 g, 78 mmol) in 1,2-dichloroethane (20
mL) is then added to the suspension, dropwise, over 30 min.
and proton-decoupled 31P NMR spectra were obtained using
deuterated solvents. Unless otherwise indicated, tetrameth-
ylsilane (TMS) was used as internal reference for 1H NMR
spectra, and 85% phosphoric acid in deuterium oxide as an
external reference for 31P NMR spectra. Proton-decoupled 13C
NMR spectra were recorded in CDCl3, C6D6, or DMSO-d6 using
TMS as an internal reference. Chemical shifts δ are reported
in parts per million (ppm). NMR spectra were run at 25 °C or
as indicated.
Low and high-resolution FAB mass spectra were acquired
from samples dissolved in a thioglycerol matrix containing
sodium iodide as the reference. Samples were then subjected
to bombardment with 8 keV fast cesium ions and accurate
mass measurements (n ) 3) were obtained by peak matching.
Gas chromatogaphy-mass spectrometry analyses were
performed using a gas chromatograph equipped with a capil-
lary inlet and a mass selective detector, controlled through a
DOS-series MS ChemStation. Sample separations were per-
formed on a 15 m × 0.25 mm fused-silica capillary column,
wall-coated with 0.25 µm HP-1 cross-linked methyl siloxane
(Hewlett-Packard). Helium was employed as the carrier gas
at a linear velocity of 52 cm/s using flow rates of 3 and 50
mL/min at the septum purge and split vents, respectively.
Temperatures were 250 °C at the injection port and 280 °C at
the transfer line to the detector. Injections were made at an
initial oven temperature of 60 °C. The inlet purge was
activated at 2.0 min postinjection. The oven temperature was
held isothermally at 60 °C for 2 min and then increased
linearly to 275 °C at 20 °C/min. The final temperature was
maintained for 5.0 min. Mass spectral detection (electron-
ionization, 70 eV) was performed by scanning ions between
(16) (a) Lesimple, P.; Bigg, D. C. H. Synthesis 1991, 306-308. (b)
Bigg, D. C. H.; Lesimple, P. Synthesis 1992, 277-278.
J . Org. Chem, Vol. 67, No. 18, 2002 6435

Wilk et al.
After additional stirring (2 h) at room temperature, the
reaction mixture is quenched with ice/water (∼200 mL). Upon
addition of sodium carbonate (100 g), the mixture is extracted
with ethyl acetate (10 × 30 mL). The organic fractions are
combined, dried over magnesium sulfate, and concentrated
under reduced pressure. The residue is dissolved in dichlo-
romethane (30 mL) and filtered through a 45 µm nylon
membrane filter. The solution is brought to a boil, and hexane
is added until opalescence of the solution, which is left to cool
slowly to room temperature and then stored at 4 °C for 12 h.
The crystalline product that is formed is filtered off, washed
with hexane, and air-dried to give N-tert-butyl-5-hydroxyval-
eramide (6.7 g, 39 mmol, 56%) as white crystals (mp 70 °C).
1H NMR (300 MHz, CDCl3): δ 1.34 (s, 9H), 1.60 (m, 2H), 1.70
(m, 2H), 2.15 (t, J ) 7.0 Hz, 2H), 3.64 (t, J ) 6.2 Hz, 2H), 4.09
(s, 1H), 5.33 (s, 1H). 13C NMR (75 MHz, CDCl3): δ 21.5, 28.9,
32.1, 36.9, 51.2, 62.1, 172.5. EI-MS (70 eV): m/z (% relative
abundance), 173 (M+, 2), 158 (3), 155 (11), 115 (6), 101 (15),
100 (15), 83 (6), 72 (6), 59(15), 58 (100).
N-ter t-Bu t ylglycola m id e. Methyl glycolate (9.00 g, 100
mmol) is mixed with tert-butylamine (16.0 g, 219 mmol) in a
50 mL round-bottom flask equipped with a reflux condenser
and a drying tube. The reaction mixture is refluxed for 16 h.
Excess amine is evaporated under reduced pressure, and the
residue is further purified by recrystallization from chloroform:
hexane. N-tert-Butylglycolamide (4.8 g, 37 mmol) is obtained
as white crystals (mp 75 °C).
1H NMR (300 MHz, CDCl3): δ
1.39 (s, 9H), 3.97 (s, 2H), 6.19 (s, 1H). 13C NMR (75 MHz,
CDCl3): δ 28.9, 51.2, 62.4, 170.6. EI-MS (70 eV): m/z (%
relative abundance), 131 (M+, 20), 116 (51), 100 (7), 76 (16),
58 (85), 57 (100).
N-Meth yl-4-h yd r oxybu tyr a m id e. This compound is pre-
pared and characterized as reported earlier.5a EI-MS (70 eV):
m/z (% relative abundance), 117 (M+, 0.3), 99 (5), 87 (42), 86
(10), 73 (100), 69 (14), 59 (9), 58 (69).
N-Isop r op yl-4-h yd r oxybu tyr a m id e. This compound is
prepared in a manner identical to that described for the
preparation of 2 and isolated as a viscous oil. 1H NMR (300
MHz, CDCl3): δ 1.15 (d, J ) 6.6 Hz, 6H), 1.86 (m, 2H), 2.32
(t, J ) 6.4 Hz, 2H), 3.29 (bs, 1H), 3.68 (t, J ) 5.8 Hz, 2H), 4.06
(m, 1H), 5.94 (bs, 1H). 13C NMR (75 MHz, CDCl3): δ 22.7, 28.4,
34.2, 41.6, 62.3, 172.6. EI-MS (70 eV): m/z (% relative
abundance), 145 (M+, 7), 130 (7), 115 (21), 114 (22), 112 (59),
101 (100), 86 (50), 70 (26).
Gen er a l P r oced u r e for th e P r ep a r a tion of th e Deoxy-
r ibon u cleosid e P h osp h or a m id ites 7-10, 18, a n d 19. 5′-
O-(4,4′-Dimethoxytrityl)thymidine (5, 270 mg, 0.50 mmol) and
diethylammonium tetrazolide (34 mg, 0.49 mmol) are dried
under high vacuum for 5 h. Dry methylene chloride (5 mL) is
added under a dry argon atmosphere followed by tris(diethy-
lamino)phosphine (125 mg, 0.50 mmol). After 20 min, the
appropriate amido alcohol (0.60 mmol) is added to the solution,
and the reaction mixture is stirred at room temperature for 8
h. The reaction mixture is then evaporated to dryness under
reduced pressure, resuspended in 1 mL benzene:triethylamine
(9:1 v/v), and loaded onto a silica gel chromatography column
(13 mm × 50 mm). The reaction product is eluted from the
column using benzene:triethylamine (9:1 v/v) as the eluent.
Selected fractions are pooled and concentrated to dryness
under vacuum. The deoxyribonucleoside phosphoramidites
7-10, 18, and 19 are isolated as white amorphous solids in
yields ranging from 75 to 95%.
N-ter t-Bu tyl-3-h yd r oxyp r op ion a m id e. 3-Hydroxypropi-
onic acid17 (1.00 g, 11.1 mmol) is coevaporated twice with
pyridine (50 mL) and then concentrated to volume of ∼20 mL.
4,4′-Dimethoxytrityl chloride (4.00 g, 11.8 mmol) is added to
the solution, and the reaction mixture is left stirring at room
temperature for 12 h. The reaction mixture is then poured into
a saturated solution of sodium bicarbonate (100 mL) and
extracted with ethyl acetate (3 × 50 mL). The organic fractions
are collected and evaporated under reduced pressure, affording
a foam. The product is then dissolved in chloroform and loaded
onto a silica gel column (2.5 cm × 5 cm) preequilibrated in
chloroform. The desired product is eluted from the column
using a gradient of methanol (0-10% v/v) in chloroform. It is
important to ensure that the chloroform used for column
equilibration and the preparation of various eluents contains
pyridine (1% v/v) to prevent premature cleavage of the 4,4′-
dimethoxytrityl ether. Appropriate fractions are pooled, con-
centrated, and dried under high vacuum to give 4.3 g of
pyridinium 3-(4,4′-dimethoxytrityloxy)propionate as a viscous
oil. The dry material is mixed with 1,1′-carbonyldiimidazole
(3.5 g, 22 mmol) and dry THF (20 mL), and the resulting
solution is heated at 50 °C for 2 h. Then, tert-butylamine (2.0
g, 27 mmol) and triethylamine (2 mL) are added to the
solution, which is left stirring at 50 °C for an additional 3 h.
The reaction mixture is evaporated to dryness and kept under
high vacuum overnight. The material is triturated in ethyl
acetate:hexane (1:1 v/v), and the sludge is applied onto a silica
gel column (6 cm × 5 cm) preequilibrated in ethyl acetate:
hexane (1:1 v/v). The product is eluted from the column using
a gradient of ethyl acetate (50-70% v/v) in hexane. Selected
fractions are collected, concentrated, and dried under high
vacuum, affording 1.47 g of N-tert-butyl-3-(4,4′-dimethoxytri-
tyloxy)propionamide as an amorphous solid. The 4,4′-dimethox-
ytrityl ether derivative is dissolved in dichloromethane (10
mL), and solid p-toluenesulfonic acid monohydrate (650 mg,
3.42 mmol) is added in one portion to the solution. After 15
min, the reaction mixture is loaded onto a silica gel column
(13 mm × 5 cm) preequilibrated in chloroform. The product is
eluted from the column using a gradient of methanol (0-5%
v/v) in chloroform. Elution of the desired product is monitored
by thin-layer chromatography and phosphomolybdic acid
staining. Product-containing fractions are combined, concen-
trated, and dried under high vacuum. N-tert-Butyl-3-hydrox-
ypropionamide is recrystallized from methylene chloride:
hexane (1:4 v/v) to produce large white crystals (180 mg)
5′-O-(4,4′-Dim eth oxytr ityl)-3′-O-(N,N-d ieth yla m in o)[3-
(N-ter t-b u t ylca r b oxa m id o)-1-p r op yloxy]p h osp h in yl-2′-
d eoxyth ym id in e (7). 31PNMR (121 MHz, C6D6): δ 148.6,
149.3.
5′-O-(4,4′-Dim et h oxyt r it yl)-3′-O-(N,N-d iet h yla m in o)-
[4-(N-ter t-bu tylca r boxa m id o)bu tyloxy]p h osp h in yl-2′-d e-
oxyth ym id in e (8). 31P NMR (121 MHz, C6D6): δ 149.1, 149.4.
5′-O-(4,4′-Dim et h oxyt r it yl)-3′-O-(N,N-d iet h yla m in o)-
[2-(N-ter t-bu tylca r boxa m id o)eth yloxy]p h osp h in yl-2′-d e-
oxyth ym id in e (9). 31P NMR (121 MHz, C6D6): δ 148.9, 149.5.
5′-O-(4,4′-Dim et h oxyt r it yl)-3′-O-(N,N-d iet h yla m in o)-
[(N-ter t-bu tylca r boxa m id o)m eth yloxy]p h osp h in yl-2′-d e-
oxyth ym id in e (10). 31P NMR (121 MHz, C6D6): δ 149.8,
150.7.
5′-O-(4,4′-Dim eth oxytr ityl)-3′-O-(N,N-d ieth yla m in o)[3-
(N-m eth ylcar boxam ido)-1-pr opyloxy]ph osph in yl-2′-deoxy-
th ym id in e (18). 31PNMR (121 MHz, C6D6): δ 144.8, 145.2.
5′-O-(4,4′-Dim eth oxytr ityl)-3′-O-(N,N-d ieth yla m in o)[3-
(N-isop r op ylca r b oxa m id o)-1-p r op yloxy]p h osp h in yl-2′-
d eoxyth ym id in e (19). 31PNMR (121 MHz, C6D6): δ 148.6,
149.2.
Gen er a l P r oced u r e for th e P r ep a r a tion of th e Di-
n u cleosid e P h osp h a te Tr iester 11 a n d P h osp h or oth ioa te
Tr iester s 12-15, 20, 21. Manual solid-phase dinucleotide
syntheses are performed on a 0.2 µmol scale under conditions
recommended for standard automated DNA synthesis. The
sulfurization reaction required in the preparation of dinucleo-
side phosphorothiates is effected by treatment with 0.05 M
3H-1,2-benzodithiol-3-one 1,1-dioxide (Glen Research) in ac-
1
melting at 93 °C. H NMR (300 MHz, CDCl3): δ 1.36 (s, 9H),
2.35 (t, J ) 5.5 Hz, 2H), 3.39 (s, 1H), 3.85 (t, J ) 5.5 Hz, 2H),
5.70 (s, 1H). 13C NMR (75 MHz, CDCl3): δ 29.0, 38.9, 51.5,
59.2, 172.1. EI-MS (70 eV): m/z (% relative abundance), 145
(M+, 9), 130 (26), 90 (8), 73 (12), 58 (100), 57 (19).
(17) Read, R. R. Organic Syntheses; Wiley: New York, 1941; Collect.
Vol. I, pp 321-322.
6436 J . Org. Chem., Vol. 67, No. 18, 2002

3-(N-tert-Butylcarboxamido)-1-propyl Protecting Group
TABLE 1. RP -HP LC Reten tion Tim e (RT) of
acetonitrile (1:1 v/v, 20 mL) is then added to the stirred
suspension and allowed to react for 2 h at 25 °C. Diisopropyl-
ammonium hydrochloride is filtered off, and the filtrate is
concentrated under reduced pressure affording 22 as a yel-
lowish oil. Crude 22 (δP 123.2 ppm in C6D6)10 is immediately
used without further purification in the synthesis of deoxyri-
bonucleoside phosphoramidites 1a -d . For analytical purposes,
22 (0.6 g) is dissolved in benzene/triethylamine (9:1 v/v, 3 mL)
and chromatographed on a silica gel column (2.5 cm × 6 cm)
using benzene/triethylamine (9:1 v/v) as an eluent. Fractions
are collected, and those containing the product are pooled and
evaporated to dryness under reduced pressure affording 22
(0.5 g) as a white crystalline solid (mp 71-72 °C). 1H NMR
(300 MHz, C6D6):10 δ 1.19 (d, J ) 6.8 Hz, 12H), 1.22 (d, J )
6.8 Hz, 12H), 1.25 (s, 9H), 2.00 (m, 2H), 2.05 (m, 2H), 3.48
Din u cleosid e P h osp h otr iester s
dinucleotide
RT (min)
11
12
13
14
15
20
21
26.4, 26.7
32.7, 33.1
34.6, 35.1
31.3, 31.7
24.6, 25.1
28.2, 28.4
32.2a
a
RP and SP diastereomers are not resolved under the chro-
matographic conditions used.
etonitrile as recommended in the literature.18 Cleavage of the
dinucleoside phosphotriesters from the solid support is achieved
upon exposure to gaseous methylamine (∼2 bar) for 1 min.7
The phosphotriesters 11-15, 20, and 21 are then eluted from
the respective columns with 0.1 M triethylammonium acetate
(pH 7.0, 0.3 mL). Each of the dinucleoside phosphotriesters
are purified by RP-HPLC using a 5 µm Supelcosil LC-18S
column (25 cm × 4.6 mm) and a linear gradient of 1% MeCN/
min, starting from 0.1 M triethylammonium acetate, pH 7.0,
at a flow rate of 1 mL/min. Peaks corresponding to either the
RP or SP diastereomer of dinucleoside phosphotriesters 11-
15, 20, and 21 are collected individually. The RP-HPLC
retention time (RT) of each pair of diastereomers are shown
in Table 1.
P h osp h a te/Th iop h osp h a te Dep r otection Stu d ies of
th e RP -HP LC P u r ified Din u cleosid e P h osp h otr iester s
11-15, 20, a n d 21. These studies are aimed at providing
relatively accurate assessments of the deprotection rates of
selected nucleotidic phosphate/phosphorothioate protecting
groups under thermolytic conditions. Typically, each of five
150 µL-aliquots of a RP-HPLC-purified and, when possible,
P-stereopure dinucleoside phosphate/phosphorothioate triester
in 0.1 M triethylammonium acetate (pH 7.0) is flame-sealed
in a 1 mL glass ampule. The samples are then placed in a
thermostated heat block at 90 ( 2 °C. At appropriate time-
points, samples are taken out the heat block and immediately
frozen in dry ice. Reference samples (t ) 0) are handled
identically without being heated. Thawed samples are ana-
lyzed by RP-HPLC under the conditions used for the purifica-
tion of the dinucleoside phosphotriesters (see Experimental
Section). Phosphate/thiophosphate deprotection during RP-
HPLC analysis at 25 °C is negligible as evidenced by the
reference samples (t ) 0). The amounts of dinucleoside
phosphodiester formation recorded for each time point is
calculated from the integrated peak area of 16 or RP /SP -17
relative to that determined for unreacted dinucleoside phos-
photriesters. The half time (t1/2) of each deprotection reaction
is determined from the best fit of the first-order curve and is
reported, where appropriate, in the Results and Discussion
section. The idendity of 16 or RP /SP -17 is corroborated with
authentic commercial samples or with samples that were
synthesized using 2-cyanoethyl deoxyribonucleoside phos-
phoramidites under standard conditions.
(sept, J ) 6.8 Hz, 2H), 3.52 (sept, J ) 6.8 Hz, 2H), 3.63 (dt,
3
J ) 6.0 Hz, J PH ) 7.3 Hz, 2H). 13C NMR (75 MHz, C6D6):10
δ
3
24.0, 24.1, 24.7, 24.8, 28.1 (d, J PC ) 8.5 Hz), 28.8, 34.0, 44.6,
44.7, 50.7, 63.7 (d, 2J PC ) 21.2 Hz), 171.1. 31P NMR (121 MHz,
C6D6): δ 123.3.
Gen er a l P r oced u r e for th e P r ep a r a tion of th e Deoxy-
r ibon u cleosid e P h osp h or a m id ites 1a -d . A suitably pro-
tected 2′-deoxyribonucleoside (2.0 mmol) is dried under high
vacuum for 4 h in a 50 mL round-bottom flask. Anhydrous
methylene chloride (10 mL) is added under an argon atmo-
sphere followed by crude 22 (818 mg, ∼2.1 mmol). To the
magnetically stirred solution is added sublimed 1H-tetrazole
(140 mg, 2.0 mmol), portionwise, over a period of 1 h. The
progress of the reaction is monitored by TLC using either
benzene:triethylamine (9:1 v/v) or methylene chloride:hexane:
triethylamine (3:6:1 v/v/v) as eluents. Phoshinylation of pro-
tected 2′-deoxyribonucleosides is usually complete within 3 h
at 25 °C. However, in the case of N2-isobutyryl-5′-O-(4,4′-
dimethoxytrityl)-2′-deoxyguanosine, the phosphinylation reac-
tion time is extended to 12 h to ensure optimum yields of 1d .
The reaction mixture is then evaporated to dryness under
vacuum, resuspended in 1 mL of methylene chloride:hexane:
triethylamine (1:8:1 v/v/v), and chromatographed on a silica
gel column (2.5 cm × 10.0 cm) preequilibrated in methylene
chloride:hexane:triethylamine (1:8:1 v/v/v). The column is first
eluted with methylene chloride:hexane:triethylamine (1:8:1
v/v/v), and then the concentration of methylene chloride is
increased until final solutions composed of 3:6:1 (1a ), 5:4:1
(1b), 7:2:1 (1c), and 8:1:1 (1d ) (v/v/v) methylene chloride:
hexane:triethylamine elute pure phosphoramidites. Appropri-
ate fractions are collected, pooled together, and evaporated to
dryness under vacuum. Phosphoramidites 1a -d are obtained
as white amorphous solids in yields ranging from 70 to 90%.
5′-O-(4,4′-Dim eth oxytr ityl)-3′-O-(N,N-diisopr opylam in o)-
[3-(N-ter t-b u t ylca r b oxa m id o)-1-p r op yloxy]p h osp h in yl-
2′-d eoxyth ym id in e (1a ). 31P NMR (121 MHz, C6D6): δ 148.3,
148.5. FAB-HRMS: calcd for C45H61N4O9P (M + Na)+ 855.4074,
found 855.4083.
N4-Ben zoyl-5′-O-(4,4′-d im eth oxytr ityl)-3′-O-(N,N-d iiso-
p r op yla m in o)[3-(N-ter t-bu tylca r boxa m id o)-1-p r op yloxy]-
p h osp h in yl-2′-d eoxycytid in e (1b). 31P NMR (121 MHz,
C6D6): δ 148.3, 149.1. FAB-HRMS: calcd for C51H64N5O9P
(M + Cs)+ 1054.3496, found 1054.3480.
O-[3-(N-ter t-Bu t ylca r boxa m id o)-1-p r op yl] N,N,N′,N′-
tetr a isop r op ylp h osp h or d ia m id ite (22). Anhydrous diiso-
propylamine (19.6 mL, 140 mmol) is added dropwise under
an argon atmosphere to a solution of freshly distilled phos-
phorus trichloride (1.75 mL, 20 mmol) in dry benzene (100
mL). The reaction mixture is left stirring at 25 °C until
complete transformation of (N,N-diisopropylamino)dichloro-
phosphine (δP 168.2 ppm in C6D6) to bis(N,N-diisopropylami-
no)chlorophosphine (δP 134.1 ppm in C6D6) is achieved (∼3-4
d), as indicated by 31P NMR spectroscopy. A solution of N-tert-
butyl-4-hydroxybutyramide (2, 3.18 g, 20 mmol) in benzene:
N6-Ben zoyl-5′-O-(4,4′-Dim eth oxytr ityl)-3′-O-(N,N-d iiso-
p r op yla m in o)[3-(N-ter t-bu tylca r boxa m id o)-1-p r op yloxy]-
p h osp h in yl-2′-d eoxya d en osin e (1c). 31P NMR (121 MHz,
C6D6): δ 148.6, 148.7. FAB-HRMS: calcd for C52H64N7O8P
(M + Na)+ 968.4453, found 968.4430.
N2-Isobu tyr yl-5′-O-(4,4′-Dim eth oxytr ityl)-3′-O-(N,N-d i-
isop r op yla m in o)[3-(N-ter t-bu tylca r boxa m id o)-1-p r op yl-
oxy]p h osp h in yl-2′-d eoxygu a n osin e (1d ). 31P NMR (121
MHz, C6D6): δ 146.3, 148.3. FAB-HRMS: calcd for C49H66
-
N7O9P (M + Na)+ 950.4559, found 950.4560.
P r ep a r a tion of Oligon u cleotid es. Solid-phase synthesis
of d(ATCCGTAGCTAAGGTCATGC) and its phosphorothioate
analogue is performed using a DNA synthesizer according to
the manufacturer’s recommendations. 2-Cyanoethyl deoxyri-
(18) Iyer, R. P.; Phillips, L. R.; Egan, W.; Regan, J . B.; Beaucage, S.
L. J . Org. Chem. 1990, 55, 4693-4699; see also: Regan, J . B.; Phillips,
L. R.; Beaucage, S. L. Org. Prep. Proc. Int. 1992, 24, 488-492.
J . Org. Chem, Vol. 67, No. 18, 2002 6437

Wilk et al.
bonucleoside phosphoramidites and all the reagents pertaining
to the automated preparation of oligonucleotides were pur-
chased from Perkin-Elmer and used as recommended by the
manufacturer. 2-Cyanoethyl deoxyribonucleoside phosphor-
amidites and phosphoramidites 1a -d are used as 0.1 M
solutions in dry acetonitrile. The sulfurization reaction re-
quired in the preparation of oligodeoxyribonucleoside phos-
phorothioates is effected by treatment with 0.05 M 3H-1,2-
benzodithiol-3-one 1,1-dioxide (Glen Research) in acetonitrile
as recommended in the literature.18
LC-18S column (25 cm × 4.6 mm) and a linear gradient of 1%
MeCN/min, starting from 0.1 M triethylammonium acetate pH
7.0, at a flow rate of 1 mL/min. RP-HPLC profiles of the digests
are shown in Supporting Information. These did not reveal
detectable amounts of nucleobase modification or incomplete
oligonucleotide deprotection.
Ack n ow led gm en t. This research is supported in
part by an appointment to the Postgraduate Research
Participation Program at the Center for Biologics Evalu-
ation and Research administered by the Oak Ridge
Institute for Science and Education through an inter-
agency agreement between the U.S. Department of
Energy and the U.S. Food and Drug Administration.
P u r ifica tion a n d Ch a r a cter iza tion of Oligon u cleo-
tid es. Crude oligomers synthesized from 1a -d that are still
covalently linked to the CPG support are treated with pres-
surized ammonia gas (∼10 bar) for 10 h at 25 °C. The crude
oligonucleotides are then eluted from the support with PBS
buffer pH 7.2 (1 mL). The solutions are heated in a closed vial
at 90 °C for 30 min. Crude oligonucleotides synthesized from
commercial 2-cyanoethyl deoxyribonucleoside phosphorami-
dites are released from CPG and deprotected by treatment
with concentrated ammonium hydroxide for 10 h at 55 °C.
Fully deprotected oligomers are then electrophoresed on 20%
polyacrylamide-7 M urea gels (40 cm × 20 cm × 0.75 mm),
which were prepared using electrophoresis purity reagents
(Bio-Rad).
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra of 2, N-tert-butyl-5-hydroxyvaleramide, N-tert-butyl-
3-hydroxypropionamide, N-tert-butylglycolamide, N-methyl-4-
hydroxybutyramide, N-isopropyl-4-hydroxybutyramide, 3, 4,
2-(tert-butylimino)tetrahydrofuran hydrobromide, and 22. 31
P
NMR spectra of 1a -d , 7-10, 18, 19, and 22. RP-HPLC chro-
matograms of hydrolysates resulting from snake venom phos-
phodiesterase and bacterial alkaline phosphatase digestion of
crude d(ATCCGTAGCTAAGGTCATGC), which was prepared
from either phosphoramidites 1a -d or commercial 2-cyano-
ethyl deoxyribonucleoside phosphoramidites. Kinetic studies
of the cleavage of the 3-(N-tert-butylcarboxamido)-1-propyl
Gels are stained by soaking in a solution of Stains-all as
described elsewhere.5a Figure 2 shows a photograph of such a
gel.
group from purified 3 and crude d(APSTPSCPSCPSGPSTPSAPSGPS
-
Oligonucleoside phosphorothioates are analyzed for desul-
furization by 31P NMR spectroscopy as shown in Figure 3 and
in Supporting Information.
C
PSTPSAPSAPSGPSGPSTPSCPSAPSTPSGPSC) in D2O at 80 °C and
60 °C, respectively. 31P NMR spectrum of the interaction of
O,O-diethyl phosphorothioate with 2-(tert-butylimino)tetrahy-
drofuran hydrobromide in D2O at 90 °C. This material is
available free of charge via the Internet at http://pubs.acs.org.
Enzymatic digestion of crude d(ATCCGTAGCTAAGGT-
CATGC) synthesized using either 1a -d or conventional 2-cy-
anoethyl deoxyribonucleoside phosphoramidites is performed
with snake venom phosphodiesterase (Crotalus durissus,
Boehringer) and bacterial alkaline phosphatase (Sigma) ac-
cording to a published procedure.19 An aliquot of either digest
is analyzed by reversed-phase HPLC using a 5 µm Supelcosil
J O0258608
(19) Scremin, C. L.; Zhou, L.; Srinivasachar, K.; Beaucage, S. L. J .
Org. Chem. 1994, 59, 1963-1966.
6438 J . Org. Chem., Vol. 67, No. 18, 2002