Syntheses of aliphatic polycarbonates from 2′-deoxyribonucleosides

Article abstract of DOI:10.1021/bm101328j

Poly(2′-deoxyadenosine) and poly(thymidine) constructed of carbonate linkages were synthesized by polycondensation between silyl ether and carbonylimidazolide at the 3′- and 5′-positions of the 2′-deoxyribonucleoside monomers. The N-benzoyl-2′-deoxyadenos

Full text of DOI:10.1021/bm101328j

ARTICLE
pubs.acs.org/Biomac
Syntheses of Aliphatic Polycarbonates from 20-Deoxyribonucleosides
Masato Suzuki,* Toyokazu Sekido, Shin-ichi Matsuoka, and Koji Takagi
Department of Materials Science and Engineering, Nagoya Institute of Technology, Nagoya 466-8555, Japan
S Supporting Information
b
ABSTRACT: Poly(20-deoxyadenosine) and poly(thymidine)
constructed of carbonate linkages were synthesized by poly-
condensation between silyl ether and carbonylimidazolide at
the 30- and 50-positions of the 20-deoxyribonucleoside mono-
mers. The N-benzoyl-20-deoxyadenosine monomer afforded
the corresponding polycarbonate together with the cyclic
oligomers. However, the deprotection of the N-benzoyl group
resulted in the scission of the polymer main chain. Thus, the N-unprotected 20-deoxyadenosine monomers were examined for
polycondensation. However, there was involved the undesired reaction between the adenine amino group and the carbonylimi-
dazolide to form the carbamate linkage. In order to exclude this unfavorable reaction, dynamic protection was employed. Strong
hydrogen bonding was used in place of the usual covalent bonding for reducing the nucleophilicity of the adenine amino group.
Herein, 30,50-O-diacylthymidines that form the complementary hydrogen bonding with the adenine amino group were added to
the polymerization system of the N-unprotected 20-deoxyadenosine monomer. Consequently, although the oligomers (Mn =
1000ꢀ1500) were produced, the contents of the carbamate group were greatly reduced. The dynamic protection reagents were
easily and quantitatively recovered as the MeOH soluble parts from the polymerization mixtures. In the polycondensation of the
thymidine monomer, there tended to be involved another unfavorable reaction of carbonate exchange, which consequently formed
the irregular carbonate linkages at not only the 30-50 but also the 30-30 and 50-50 positions. Employing the well-designed monomer
suppressed the carbonate exchange reaction to produce poly(thymidine) with the almost regular 30-50carbonate linkages.
’ INTRODUCTION
Although a green process of enzymatic polymerization would be
preferable especially for polymer synthesis from renewable
resources, it can be applied to limited monomers for the
preparation of polycarbonate.3 A literature survey has actually
revealed that carbohydrate-based polycarbonates are prepared by
ring-opening polymerization as well as polycondensation using
chemical reagents.4 Our project needs the regular formation of
the intermolecular 30-50 carbonate linkage between the ribose
rings. To our first expectation, regio-selective ring-opening
polymerization could satisfy our demand together with con-
trolling the molecular weight. However, unfortunately, the
corresponding cyclic monomer was not obtained from 20-deox-
yadenosine. Thus, the second choice was polycondensation of an
AB type monomer, which should produce the regioregular
connection. In contrast, chemical or enzymatic polycondensa-
tion between AA and BB type monomers, typically the combina-
tion of phosgene or dialkyl carbonate with diol for polycarbonate,
is unfavorable for the regioselective formation of asymmetric
carbonate linkages. Thus, we decided to employ the literature
method for polycarbonate synthesis.5 Therein, cyclohexane-1,4-
diol is transformed to the AB monomer having silyl ether and
carbonylimidazolide and subjected to polycondensation, which is
conducted by desilylation with the added fluorine anion and
DNA is a polyphosphate that is composed of 20-deoxyribonu-
cleosides with the regular 30-50 phosphate linkages. We have
planned to use 20-deoxyribonucleosides, which are a kind of
renewable resources, as monomer components and to construct
novel poly(20-deoxyribonucleoside)s with various linking
groups. The produced biobase polymers are expected to have
specific characteristics due to the complementary hydrogen
bonding between nucleobase pairs, showing biocompatibility
and -degradability. Herein, syntheses of the corresponding
polycarbonates from 20-deoxyadenosine and thymidine have
been studied.
There are many known works about oligomers, mainly dimers
and trimers, of 20-deoxyribonucleosides with various linking
groups such as carbonate, carbamate, ester, ether, silyl ether,
and triazole.1 They are prepared for the study of antisense
molecules. There are many other works using the complemen-
tary hydrogen bonding between nucleobase pairs for synthetic
polymer materials. Nucleobases are introduced to various poly-
mers as the pendant or terminal groups.2 There have been
disclosed unique organizations of the polymers owing to the
complementary hydrogen bonding.
As mentioned above, this article deals with the syntheses of
poly(20-deoxyadenosine) and polythymidine constructed with
the regular 30-50 carbonate linkages. There are two synthetic
ways, ring-opening polymerization and polycondensation, for
aliphatic polycarbonate, using chemical reagents or enzymes.
Received: November 8, 2010
Revised:
March 2, 2011
Published: March 05, 2011
r
2011 American Chemical Society
1449
dx.doi.org/10.1021/bm101328j Biomacromolecules 2011, 12, 1449–1459
|

Biomacromolecules
ARTICLE
subsequent attack of the generated alkoxide anion to the active
carbonyl group. This methodology is expected to be applicable to
our project; the corresponding monomers can be prepared by the
individual transformation of the 30- and 50-hydroxyl groups of 20-
deoxyadenosine and thymidine ribose rings to a combination of
the silyl ether and the carbonylimidazolide.
and the white precipitate was isolated by centrifugation. Successive
wash with methanol (50 mL ꢁ 4) and diethyl ether was followed by
drying under vacuum to give a white powdery polymer (198 mg, 78%
yield).
1H NMR (DMSO-d6, 200 MHz, ppm) δ 11.16 (1H), 8.78ꢀ8.65
(2H), 8.01 (2H), 7.57 (3H), 6.53 (1H), 5.53ꢀ5.43 (1H), 4.56ꢀ4.29
(3H), 3.30 (1H, overlapped with H2O peak), 2.70 (1H). 13C NMR
(DMSO-d6, 50 MHz, ppm) δ 165.8, 153.6, 152.0, 151.8, 150.5, 143.3,
’ EXPERIMENTAL SECTION
133.4, 132.5, 128.5, 125.9, 84.1, 81.3, 78.2, 67.2, 35.4. IR (ATR, cmꢀ1
1748, 1695, 1608, 1581, 1454, 1246, 1073, 708.
)
Measurements. The 1H and 13C NMR spectra were recorded on
Bruker AVANCE-200FT-NMR and -600FT-NMR spectrometers in
CDCl3 (0.03 v/v% TMS as the internal standard), DMSO-d6 (the
signals due to the residual protons and the carbon were used for the
standards), and DMF (1 v/v% TMS as the internal standard). The IR
spectra were recorded on a JASCO FT/IR-460 plus spectrometer by
KBr pellet and ATR methods. The MALDI-TOF mass spectra were
recorded on a SHIMADZU/KRATOS AXIMA-CFR plus spectrometer
(matrix, 2-(4-hydroxyphenylazo)benzoic acid; cast solvent, 1,1,1,3,3,3-
hexafluoro-2-propanol). Gel permeation chromatography (GPC) was
conducted on tandem columns of TOSOH TSKgel ALPHA5000 and
ALPHA3000 (eluent, DMF with KBr (0.5 g/L); calibration, polystyrene
standards). Silica gel 60N (Kanto Kagaku) was used for column
chromatography.
30-O-(Imidazol-1-yl)carbonyl-50-O-(t-butyl)dimethylsilyl-
N-monomethoxytrityl-20-deoxyadenosine (2). Carbonyl diimi-
dazole (0.78 mg, 4.88 mmol) was dissolved into dry THF (3 mL) under
N2 and stirred with a catalytic amount (50ꢀ80 mg) of powdery KOH
at 0 °C. Then, the dry THF solution (9 mL) of 50-O-(t-bu-
tyl)dimethylsilyl-N-MMTr-20-deoxyadenosine (2.01 mg, 3.15 mmol)
was added dropwise. The reaction was monitored by TLC, which
detected no spot due to the starting material (Rf = 0.61, EtOAc) after
40 min at 0 °C. The reaction mixture was diluted with EtOAc, washed
with three portions of water, and dried over MgSO4. Vacuum drying
gave a white solid of 30-O-(imidazol-1-yl)carbonyl-50-O-TBDMS-N-
MMTr-20-deoxyadenosine (1.88 g, 81%yield). No impurity was de-
tected by TLC (Rf = 0.51, EtOAc) as well as 1H NMR spectroscopy;
therefore, the product was used for polymerization without further
purification. 1H NMR (CDCl3, 200 MHz, ppm) δ 8.17 (s, 1H), 8.09
(s,1H), 8.06 (s, 1H), 7.44 (s, 1H), 7.35ꢀ7.23 (m, 12H), 7.11 (s, 1H),
6.93 (s, 1H), 6.79 (d, 2H, J = 8.8 Hz), 6.51 (dd, 1H, J = 8.7, 5.9 Hz),
5.68 (d-like, 1H, J = 5.4 Hz), 4.39 (m, 1H), 3.97 (d-like, 2H, J = 3.0 Hz),
3.78 (s, 3H), 2.87 (m, 2H), 0.92 (s, 9H), 0.12 (s, 6H). 13C NMR
(CDCl3, 50 MHz, ppm) δ 158.4, 154.2, 152.4, 148.6, 148.2, 145.2,
137.8, 137.2, 137.1, 131.1, 130.3, 128.9, 127.9, 126.9, 121.2, 117.1,
113.2, 85.1, 84.2, 79.8, 71.1, 63.5, 55.2, 38.6, 26.0, 18.4, ꢀ5.3, ꢀ5.5. IR
(KBr, cmꢀ1) 3413, 2954, 2929, 2856, 1763, 1605, 1471, 1400, 1290,
1250, 1179, 1002, 835.
Materials. DMF, CH3CN, and Et3N used for the monomer
syntheses and the polymerizations were dried over CaH2 and distilled
under N2. Other reagents were used as received. The preparation of
starting compounds for monomer syntheses is described in Supporting
Information.
30-O-(Imidazol-1-yl)carbonyl-50-O-triisopropylsilyl-N-ben-
zoyl-20-deoxyadenosine (1). To a solution of 50-O-triisopropylsilyl-
N-benzoyl-20-deoxyadenosine (1.27 g, 2.47 mmol) in dry toluene
(13 mL) were added a catalytic amount (50ꢀ80 mg) of powdery
KOH and subsequently carbonyl diimidazole (616 mg, 3.80 mmol)
under N2. TLC analyses (AcOEt/MeOH = 9/1) suggested that the
starting material (Rf = 0.67) faded out and quantitatively gave the
product (Rf = 0.60) for 1.5 h at r.t. The reaction mixture was diluted with
toluene (50 mL) and washed successively with two portions of water and
brine. Drying the organic layer over MgSO4 and excluding toluene in
vacuo gave 30-O-(imidazol-1-yl)carbonyl-50-O-triisopropylsilyl-N-ben-
zoyl-20-deoxyadenosine (1.45 g, 97%). No impurity was detected by
TLC as well as 1H NMR spectroscopy; therefore, the product was used
for polymerization without further purification. 1H NMR (CDCl3, 200
MHz) δ 9.01 (s, 1H), 8.82 (s, 1H), 8.34 (s, 1H), 8.21 (s, 1H), 8.03 (s,
1H), 7.66ꢀ7.50 (m, 3H), 7.49 (s, 1H), 7.14 (s, 1H), 6.63 (dd, 1H, J = 8.5
Hz, J = 5.8 Hz), 5.82 (d-like, 1H, J = 5.6 Hz), 4.45 (s-like, 1H), 4.09 (d-
like, 2H, J = 3.2 Hz), 3.03 (m, 1H), 2.88 (m, 1H), 1.08ꢀ1.26 (m, 21H).
13C NMR (CDCl3, 50 MHz) δ 165.0, 152.7, 151.6, 149.8, 148.1, 141.0,
137.2, 133.6, 132.8, 131.1, 128.8, 128.0, 123.5, 117.1, 85.4, 84.3, 79.4,
63.6, 38.5, 18.0, 11.9. IR (ATR, cmꢀ1) 3146, 3008, 2941, 2865, 1769,
1690, 1581, 1513, 1393, 1283, 1127, 994, 755.
Polycondensation of N-Protected Adenosine Monomer 2.
Monomer 2 was polymerized by the same procedure as monomer 1,
giving the corresponding polycarbonate. 1H NMR (DMSO-d6, 200
MHz, ppm) δ 8.38 (1H), 7.90 (1H), 7.41ꢀ7.20 (13H), 6.82 (2H), 6.34
(1H), 5.35 (1H), 4.30 (3H), 3.70 (3H), 3.25ꢀ2.60 (2H, overlapped
with H2O and solvent peaks). IR (KBr, cmꢀ1) 3411, 1754, 1605, 1510,
1471, 1252, 1182, 1034, 706.
30-O-(Imidazol-1-yl)carbonyl-50-O-triisopropylsilyl-20-de-
oxyadenosine (3). Carbonyl diimidazole (569 mg, 3.51 mmol) was
dissolved into dry THF (4 mL) under N2 and stirred with a catalytic
amount (50ꢀ80 mg) of powdery KOH at 0 °C. Then, the dry THF
solution (6 mL) of 50-O-TIPS-20-deoxyadenosine (1.01 g, 2.48 mmol)
was added dropwise. The reaction mixture was stirred for 1 h at 0 °C, was
diluted with EtOAc, washed with three portions of water, and dried over
MgSO4. The organic phase was concentrated, and hexane was slowly
added under ice cooling to give a white precipitate, which was collected
by filtration and dried in vacuo. No impurity was detected by TLC (Rf =
0.28, EtOAc/MeOH = 9:1) as well as 1H NMR spectroscopy; therefore,
the obtained white powder of 30-O-(imidazol-1-yl)carbonyl-50-O-TIPS-
20-deoxyadenosine (1.15 g, 92%yield) was used for polymerization
without further purification. 1H NMR (CDCl3, 200 MHz, ppm) δ
8.36 (s, 1H), 8.19 (s, 1H), 8.15 (s, 1H), 7.46 (s, 1H), 7.26 (s, 1H), 6.55
(dd, 1H, J = 8.5, 5.8 Hz), 5.80 (d-like, 1H, J = 5.7 Hz), 5.62 (s, 2H), 4.42
(m, 1H), 4.08 (d-like, 2H, J = 3.3 Hz), 2.99 (m,1H), 2.83 (m, 1H),
1.11ꢀ1.08 (m, 21H). 13C NMR (CDCl3, 50 MHz, ppm) δ 155.8, 153.3,
149.7, 148.2, 138.5, 137.3, 131.2, 120.0, 117.2, 85.3, 84.2, 79.6, 63.7, 38.6,
18.1, 12.0. IR (ATR, cmꢀ1) 3321, 3194, 2942, 2866, 1780, 1631, 1596,
1287, 1237, 1170, 1051, 1005, 648.
Polycondensation of N-Protected Adenosine Monomer
1. With CsF (No. 1, Table 1). CsF (0.53 g, 3.5 mmol) in a test tube was
heated with a heat gun under vacuum for 5 min. Into the test tube under
N2 were added a magnetic stirrer chip, dry DMF (1.3 mL), and 1 (401
mg, 0.66 mmol). The reaction mixture became a white gel and hardly
stirred. After 20 h at r.t., it was poured into water (50 mL) to give a white
precipitate, which was isolated by centrifugation and successively
washed with water (50 mL ꢁ 4), methanol (50 mL ꢁ 2), and diethyl
ether. Drying under vacuum gave a white powdery polymer (176 mg,
70%yield).
With TBAT (No. 2, Table 1). Tetrabutylammonium triphenyldifluor-
osilicate (TBAT) (440 mg, 0.81 mmol) was added to the solution of 1
(405 mg, 0.67 mmol) in dry DMF (1.3 mL) under N2 and stirred for
20 h at r.t. The reaction mixture was poured into methanol (50 mL),
30-O-(Imidazol-1-yl)carbonyl-50-O-(t-butyl)dimethylsilyl-
20-deoxyadenosine (4). Carbonyl diimidazole (2.2 g, 13.6 mmol)
1450
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
was dissolved into dry THF (15 mL) under N2 and stirred with a
catalytic amount (50ꢀ80 mg) of powdery KOH at 0 °C. Then, the dry
THF solution (25 mL) of 50-O-TBDMS-20-deoxyadenosine (3.53 g,
9.64 mmol) was added dropwise, and the mixture was stirred at 0 °C.
The reaction was monitored by TLC, which showed no spot due to the
starting material (Rf = 0.41, EtOAc/MeOH = 9:1) after 7 h. Thus, the
reaction mixture was diluted with EtOAc, washed with three portions of
water, and dried over MgSO4. The residue obtained by evaporation was
dissolved into EtOAc, and hexane was slowly added under ice cooling to
give a white precipitate, which was collected by filtration and dried in
vacuo. The obtained white powder of 30-O-(imidazol-1-yl)carbonyl-50-
O-TBDMS-20-deoxyadenosine (3.37 g, 76%yield) had no impurity
detected by TLC (Rf = 0.35, EtOAc/MeOH = 9:1) as well as 1H
NMR spectroscopy; therefore, it was used for polymerization without
further purification. 1H NMR (CDCl3, 200 MHz, ppm) δ 8.36 (s, 1H),
8.19 (s, 1H), 8.18 (s, 1H) 7.46 (s, 1H), 7.12 (s, 1H), 6.57 (dd, 1H, J = 7.9,
6.4 Hz), 5.88 (s, 2H), 5.70 (m, 1H), 4.42 (m, 1H), 3.99 (d-like, 2H, J =
3.0 Hz), 2.89 (m, 2H), 0.93 (s, 9H), 0.14 (s, 6H). 13C NMR (CDCl3, 50
MHz, ppm) δ 156.0, 152.9, 149.3, 147.9, 138.0, 137.0, 130.8, 119.6,
(m, 1H), 2.29 (m, 1H), 1.93 (s, 3H), 1.14ꢀ1.12 (m, 21H). 13C NMR
(CDCl3, 50 MHz, ppm) δ 163.8, 150.6, 148.4, 137.3, 134.6, 131.1,
117.2, 111.7, 85.1, 84.5, 79.3, 63.8, 38.1, 18.1, 12.5, 12.0. IR
(KBr, cmꢀ1) 2945, 2867, 1764, 1691, 1469, 1401, 1320, 1279, 1242,
1127, 1003, 883, 769, 650.
30-O-(Imidazol-1-yl)carbonyl-50-O-(t-butyl)dimethylsilylthy-
midine (9). To the solution of 50-O-TBDMSthymidine (1.38 g, 3.87
mmol) in dry THF (13 mL) were successively added a catalytic amount
(50ꢀ80 mg) of KOH powder and carbonyl diimidazole (0.95 g, 5.9
mmol). The reaction was monitored by TLC, which showed no spot due
to the starting material (Rf = 0.45, EtOAc) after 2 h at r.t. Thus, the
reaction mixture was diluted with EtOAc, washed with three portions of
water, and dried over MgSO4. The residue obtained by evaporation was
dissolved into toluene, and hexane was slowly added to give a white
precipitate, which was collected by filtration and dried in vacuo. The
obtained white powder of 30-O-(imidazol-1-yl)carbonyl-50-O-
TBDMSthymidine (1.64 g, 94%yield) had no impurity detected by
TLC (Rf = 0.30, EtOAc) as well as 1H NMR spectroscopy; therefore, it
was used for polymerization without further purification. The melting
point was measured for the sample recrystallized from toluene with
hexane (mp; 177ꢀ179 °C). 1H NMR (CDCl3, 200 MHz, ppm) δ 8.53
(br, 1H), 8.18 (s, 1H), 7.53 (s, 1H), 7.44 (s, 1H), 7.12 (s, 1H), 6.42 (dd,
1H, J = 9.3, 5.3 Hz), 5.51 (d-like, 1H. J = 5.8 Hz), 4.32 (s-like, 1H), 3.99
(d-like, 2H, J = 1.7 Hz), 2.65 (m, 1H), 2.26 (m, 1H), 1.95 (s, 3H), 0.95
(s, 9H), 0.17 (s, 6H). 13C NMR (CDCl3, 50 MHz, ppm) δ 164.0, 150.7,
148.3, 137.3, 134.5, 130.9, 117.1, 111.5, 84.9, 84.7, 79.6, 63.5, 37.9, 25.9,
18.3, 12.5, ꢀ5.4, ꢀ5.5. IR (KBr, cmꢀ1) 3296, 3097, 2950, 2929, 2857,
1743, 1706, 1688, 1469, 1407, 1323, 1276, 1262, 1179, 1126, 1006,
835, 783.
116.9, 84.9, 84.1, 79.6, 63.3, 38.5, 25.7, 18.2, ꢀ5.6, ꢀ5.7. IR (KBr, cmꢀ1
)
2926, 2859, 1765, 1602, 1397, 1322, 1282, 1241, 1124, 1005, 836,
772, 650
Polycondensation of 20-Deoxyadenosine Monomer 4
with Dynamic Protection. The typical procedures are as follows.
To the DMF (2.8 mL) solution of 30-O-(imidazol-1-yl)carbonyl-50-O-
TBDMS-20-deoxyadenosine (328 mg, 0.71 mmol) and 30,50-O-dipivar-
oyl thymidine (886 mg, 2.16 mmol) was added (n-Bu)4NPh3SiF2
(TBAT) (462 mg, 0.86 mmol) under N2. The mixture was stirred for
22.5 h and then poured into MeOH to give a white precipitate, which
was collected by centrifugation and washed with four portions ofMeOH
and one portion of Et2O. Drying the precipitate in vacuo gave a white
powder of the objective polycarbonate (172.9 mg, 86%yield, Mn = 1,000,
Mw/Mn = 2.64 (GPC; DMFþLiBr 0.5 g/L, 40 °C, PSt std)). The MeOH-
and Et2O-soluble parts were concentrated and subjected to silica gel column
chromatography (EtOAc/hexane = 1:2, Rf = 0.15) to recover 30,50-O-
dipivaroylthymidine (842 mg, 95%yield). 1H NMR (DMSO-d6, 200 MHz,
ppm) δ 8.31 (1H), 8.15 (1H), 7.31 (2H), 6.43ꢀ6.36 (1H), 5.38 (1H),
4.44ꢀ4.35 (3H), 3.25 (1H, overlapped with H2O peak), 2.60 (1H,
overlapped with DMSO peak), 0.80 (end group, Si(t-Bu)), ꢀ0.01 (end
group, SiMe2). 13C NMR (DMSO-d6, 200 MHz, ppm) δ 156.2, 153.6,
152.7, 149.2, 139.8, 119.4, 83.8, 81.2, 78.2, 67.4, 35.0, 25.7 (end group,
Si(C(CH3))Me2), 17.9 (end group, Si(C(CH3))Me2), ꢀ5.55 (end
group, Si(t-Bu)(CH3)2). IR (KBr, cmꢀ1) 3334, 3186, 1753, 1640, 1599,
1475, 1245, 1080, 932, 786, 651.
30-O-(t-Butyl)dimethylsilyl-50-O-(imidazol-1-yl)carbonyl-
thymidine (10). Carbonyl diimidazole (1.3 g, 8.0 mmol) was dis-
solved into dry THF (10 mL) under N2 and stirred with a catalytic
amount (50ꢀ80 mg) of powdery KOH at 0 °C. Then, the dry THF
solution (20 mL) of 30-O-TBDMSthymidine (1.85 g, 5.2 mmol) was
added dropwise, and the mixture was stirred at 0 °C. The reaction was
monitored by TLC (EtOAc), which showed no spot due to the starting
material (Rf = 0.47) after 1 h. Thus, the reaction mixture was diluted with
EtOAc, washed with three portions of water, and dried over MgSO4.
Vacuum drying gave a white solid of 30-O-TBDMS-50-O-(imidazol-1-
yl)carbonylthymidine (1.97, 84%yield), which had no impurity detected
1
by TLC (Rf = 0.30, EtOAc/MeOH = 9:1) as well as H NMR spec-
troscopy and whose recrystallization was impossible to achieve. Accord-
1
ingly, it was used for polymerization without further purification. H
NMR (CDCl3, 200 MHz, ppm) δ 8.54 (br, 1H), 8.18 (s, 1H), 7.43 (s,
1H), 7.11 (s, 1H), 7.07 (s, 1H), 6.12 (t-like, 1H, J = 6.6 Hz), 4.70ꢀ4.41
(m, 3H), 4.12 (m, 1H), 2.43ꢀ2.20 (m, 2H), 1.87 (s, 3H), 0.90 (s, 9H),
0.10 (s, 6H). 13C NMR (CDCl3, 200 MHz, ppm) δ 164.2, 150.3, 148.3,
137.1, 135.9, 130.6, 116.9, 111.1, 86.4, 83.6, 71.3, 66.6, 40.0, 25.5, 17.7,
12.3, ꢀ4.8, ꢀ5.1. IR (KBr, cmꢀ1) 2954, 2931, 2858, 1768, 1693, 1472,
1404, 1292, 1241, 1104, 1005, 836, 780.
Polycondensation of Thymidine Monomers Representa-
tively Using 10. With TBAT (No. 9, Table 5). To a dry DMF
(1.36 mL) solution of 10 (307 mg, 0.68 mmol) under N2 was added
TBAT (0.48 g, 0.88 mmol). The mixture was stirred for 43.5 h at r.t. and
then poured into water. The white precipitate generated was collected by
centrifugation and successively washed with five portions of MeOH and
one portion of diethyl ether. Vacuum drying gave a white powder of the
polymer (95 mg, 52%).
30-O-(Imidazol-1-yl)carbonyl-50-O-triisopropylsilylthymi-
dine (8). Carbonyl diimidazole (492 mg, 3.0 mmol) was dissolved into
dry THF (5 mL) under N2 and stirred with a catalytic amount (50ꢀ80
mg) of powdery KOH at 0 °C. Then, the dry THF solution (5 mL) of
50-O-TIPSthymidine (1.01 g, 2.53 mmol) was added dropwise, and the
mixture was stirred at 0 °C. The reaction was monitored by TLC
(EtOAc/hexane = 9:1), which showed no spot due to the starting
material (Rf = 0.46) after 0.5 h. Thus, the addition of water and NaCl to
the reaction mixture was followed by the extraction with three potions
of EtOAc. The organic phase combined was washed with three portions
of water and dried over MgSO4. The residue obtained by evaporation
was dissolved into EtOAc, and hexane was slowly added under ice
cooling to give a white precipitate, which was collected by filtration and
dried in vacuo. The obtained white powder of 30-O-(imidazol-1-
yl)carbonyl-50-O-TIPSthymidine (923 mg, 74%yield) had no impurity
detected by TLC (Rf = 0.30, EtOAc/MeOH = 9:1) as well as 1H NMR
spectroscopy; therefore, it was used for polymerization without further
purification. 1H NMR (CDCl3, 200 MHz, ppm) δ 9.54 (s, 1H), 8.21 (s,
1H), 7.49 (s, 1H), 7.46 (s, 1H), 7.11 (s, 1H), 6.43 (dd, 1H, J = 9.0, 5.1
Hz), 5.63 (d-like, 1H, J = 5.7 Hz), 4.31 (s-like, 1H), 4.08 (m, 2H), 2.67
With CsF (No. 12, Table 5). CsF (0.49 g, 3.3 mmol) in a test tube was
heated with a heat gun under vacuum for 5 min. Into the test tube
under N2 were added a magnetic stirrer chip and dry DMF (1.3 mL).
Then, 10 (401 mg, 0.66 mmol) was added at 0 °C, and the mixture was
stirred for 20 h at this temperature. The reaction mixture became a
white gel and hardly stirred during that period. It was poured into
1451
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
Scheme 1. Polycondensation of N-Bz-20-Deoxyadenosine Monomer 1
Table 1. Polycondensation of N-Bz-20-Deoxyadenosine Monomer 1 at r.t.
fluoride anion
no.
source
eq.
monomer conc. (M)
time (h)
yielda (%)
molecular weightbMp
1
2
3
CsF
5.3
1.2
5.2
0.51
0.51
0.25
20
20
65
70
78
81
3,800, 14,600
2,000, 5,700
1,500, 7,300
TBAT
TBAT
a A MeOH-insoluble part. b Mp: the molecular weight at the peak top of the GPC profile (DMFþLiBr 0.5 g/L, 40 °C, PSt std.). TBAT: (n-Bu)4NPh3SiF2
polymerization by the nucleophilic attack to the carbonyli-
midazolide; therefore, N-protected adenosine monomers
were examined. In order to get preliminary information, the
most popular protective group, the N-benzoyl, was employed
for the first monomer, 1, consequently found to be poly-
merized with the aid of CsF or (n-Bu)4NPh3SiF2 (TBAT) in
DMF at r.t. (Scheme 1 and Table 1). CsF is insoluble in
DMF but more efficiently promoted the polymerization than
DMF-soluble TBAT, affording the higher molecular weight
product.
The product polymers showed bimodal GPC profiles
(Figure 1), which were consistent with MALDI-TOF mass
spectra (Figure 2). As observed in Figure 2, there are two peak
series with the regular intervals due to the objective repeating
unit, and they are assignable to the cyclic and linear polycarbo-
1
nates, respectively. The H and 13C NMR and IR spectra also
supported the polymer structure (Supporting Information,
Figures S1ꢀS4).
Although we supposed that the deprotection of the N-benzoyl
Figure 1. GPC profiles of N-Bz-20-deoxyadenosine polycarbonates: no. 1
(a), no. 2(b), andno. 3(c) inTable1(DMFþLiBr 0.5 g/L, 40 °C, PSt std.).
group without affecting the carbonate linkage would be difficult,
it was examined. Usual deprotection conditions, the treatment
with 28% NH3 aq. at 50 °C, resulted in complete deprotec-
tion but simultaneously the frequent cleavage of the carbonate
linkage.
Accordingly, the monomethoxytrityl group, which could be
excluded under milder conditions, was employed for N-protec-
tion. The corresponding adenosine monomer 2 was found to be
polymerized under the same conditions with 1 (Supporting
Information, Scheme S1 and Table S1), however, giving the
lower molecular weight oligomer (Mn = 1700ꢀ1900) probably
due to the steric hindrance of the monomethoxytrityl group.
1H NMR, IR, and MALDI-TOF mass spectra supported the
polymer structure (Supporting Information, Figures S5ꢀS7).
The MALDI-TOF mass spectra suggested the production of the
linear and cyclic oligomers, showing the predominant formation
of the cyclic trimer.
water (50 mL) to give a white precipitate, which was collected by
centrifugation and successively washed with two portions of water and
three portions of MeOH, and one portion of Et2O. Drying under
vacuum gave a white powdery polymer (169 mg, 96%yield).
1H NMR (DMSO-d6, 200 MHz, ppm) δ 11.34 (1H), 7.52ꢀ7.48
(1H), 6.19 (1H), 5.21 (1H), 4.39 and 4.30 (3H), 2.44 (2H), 1.80 (3H),
0.85 and 0.08 (Si(t-Bu)Me2 end group). 13C NMR (DMSO-d6, 50 MHz,
ppm) δ 163.7, 153.6, 150.5, 136.0, 110.2, 84.3, 80.7, 77.8, 67.5, 35.5, 12.2.
IR (KBr, cmꢀ1) 1753, 1687, 1473, 1406, 1249, 1104, 958, 891, 784.
’ RESULTS AND DISCUSSION
Polymerization of N-Protected Adenosine Monomers.
The amino group of adenine was speculated to disturb the
1452
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
Figure 2. MALDI-TOF mass spectrum of N-Bz-20-deoxyadenosine polycarbonate (no. 3, Table 1). 2-(4-Hydroxyphenylazo)benzoic acid as a
matrix.
Table 2. Polycondensation of N-Unprotected 20-Deoxyadenosine Monomers 3 and 4 at r.t.
fluoride anion
molecular weightb
no.
monomer
source
eq.
solv.
monomer conc. (M)
time (h)
yielda (%)
Mn
Mw/Mn
carbamate contentc (%)
1
3
4
5
6
7
8
3
3
3
4
4
4
4
ꢀ
DMF
DMF
THF
DMF
THF
DMF
DMF
0.50
0.50
0.50
0.50
0.25
0.25
0.50
30.5
20
TBAT
TBAT
CsF
1.2
1.2
5.0
1.2
1.3
1.3
quant.
70
4,200
3,400
3.56
6.15
38
35
ꢀ
22
18.5
23
ꢀ
DMF-insoluble
DMF-insoluble
TBAT
TBAT
TBAT
ꢀ
ꢀ
20
quant.
2,900
2.18
3.89
32
37
24
quant.
3,400
a A MeOH-insoluble part. b GPC (DMFþLiBr 0.5 g/L, 40 °C, PSt std.). c Calculated from 1H NMR spectra.
Polymerization of N-Unprotected 20-Deoxyadenosine
Monomers. During the preparation of N-protected monomers
1 and 2, we found them contaminated with a trace amount of
the corresponding N-unprotected compounds. This means that
the carbonylimidazolide group is unexpectedly tolerant to the
adenine amino group. Thus, we were encouraged to prepare the
N-unprotected 20-deoxyadenosine monomers and succeeded in
getting two monomers 3 and 4.
Table 2 summarizes the polymerization results of monomers 3
and 4. The control reaction without the fluoride anion showed
that the monomers were stable in DMF at r.t. (no. 1), while the
addition of the fluoride anion promoted polymerization. How-
ever, the product polymers were found to be composed of not
only the carbonate but also the carbamate linkages, the latter of
which were provided by the reaction between the carbonylimi-
dazolide and the adenine amino groups, giving as a result the
hyperbranched structure (Scheme 2). This undesired reaction
was not suppressed in all runs of Table 2; the relative contents of
1
the carbamate unit were evaluated to be 32ꢀ38% by the H
NMR spectra (Figure 3). This finding means that the tolerance of
the carbonylimidazolide group to the adenine amino group is lost
in the presence of the fluoride anion, which presumably increases
the nucleophilicity of the adenine amino group through Fꢀ-HN
hydrogen bonding in addition to the activation of silyl ether.
There is another possibility that the carbamate linkage could be
formed by the reaction of the adenine amino group with the
1453
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
Scheme 2. Polycondensation of N-Unprotected 20-Deoxyadenosine Monomers 3 and 4
Figure 3. 1H NMR spectrum (DMSO-d6, 200 MHz) of polymer from 4 (no. 7, Table 2).
carbonate linkage, but it was excluded by the control experiment.
No formation of the carbamate linkage was observed when
oligo(20-deoxyadenosine carbonate) prepared from 4 with dy-
namic protection (see the next section) was treated with TBAT
(1.2 equiv) in DMF at r.t. for 23 h similarly to the polymerization
conditions. Thus, it is concluded that the adenine amino group
activated with the fluoride anion has the ability to react with the
carbonylimidazolide group but not with the carbonate group for
forming the carbamate linkage.
In order to exclude the undesired carbamate formation, we
came up with the idea of dynamic protection, where strong
hydrogen bonding would work to reduce the nucleophilic re-
activity of the amino group instead of the usual covalent bonding
protection. 1H NMR spectroscopy revealed that the complemen-
tary H-bonding between the adenine and the thymine moieties is
formed even in DMF. Mixing monomer 3 with 30,50-O-diacylthy-
midine caused both the adenine and thymine NH proton signals
to undergo a downfield shift depending on their concentration
1454
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
1
Table 3. H NMR Study of H-Bonding Interaction between the 20-Deoxyadenosine TBDMS Monomer 4 with 30,50-O-
Diacylthymidines
down field shift (ppm)a
adenine NH2 thymine NH
hydrogen bonding pair
adenosine (M)
thymidine (M)
20-deoxyadenosine TBDMS monomer 4 þ 30,50-O-diacetylthymidine 5
0.25
0.25
0.50
0.50
0.50
0.25
0.25
0.25
0.75
0.50
1.00
1.50
0.25
0.75
0.03
0.07
0.02
0.04
0.08
0.02
0.07
0.34
0.31
0.58
0.55
0.50
0.35
0.33
20-deoxyadenosine TBDMS monomer 4 þ 30,50-O-dipivaroylthymidine 6
a 1H NMR spectroscopy in DMF at 25 °C.
Scheme 3. Polymerization of 20-Deoxyadenosine TBDMS Monomer 4 with Dynamic Protection
Table 4. Polycondensation of 20-Deoxyadenosine TBDMS Monomer 4 with Dynamic Protection
fluoride anion
dynamic protection
molecular weightb
no.
source
eq.
additive
eq.
solv.
conc. of 4 (M) time (h) yielda (%)
Mn
Mw/Mn
carbamatecontentc (%)
1
TBAT
TBAT
TBAT
TBAT
CsF
1.2
1.3
1.4
1.2
5.0
1.2
1.3
1.2
1.2
1.2
1.2
1.2
1.2
5
1.6
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.1
1.0
1.0
1.0
3.1
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
THF
THF
THF
DMF
DMF
0.25
0.25
0.25
0.50
0.19
0.25
0.50
0.125
0.24
1.00
0.50
0.25
0.25
20
88
72
81
105
89
92
96
80
89
85
98
85
90
1,500
1,100
1,300
1,300
1,000
1,000
1,200
1,000
1,000
1,400
1,400
1,200
1,100
1.57
1.51
1.60
1.58
2.26
2.07
2.00
1.89
1.63
1.94
1.98
1.89
1.50
7.2
2.3
1.7
5.4
14.5
1.1
0
2
5
21
3
5
67
4
5
24
5
5
43.5
21
6
TBAT
TBAT
TBAT
TBAT
TBAT
TBAT
TBAT
TBAT
6
7
6
21
8
7
24
0
9
5
20
3.8
6.8
3.6
6.8
1.7
10
11
12
13
6
16.5
16.5
20
6
Thymine
Thymine
20
a A MeOH-insoluble part. b GPC (DMFþLiBr 0.5 g/L, 40 °C, PSt std.). c Calculated from 1H NMR spectra.
1455
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
Figure 4. 1H NMR spectrum (DMSO-d6, 200 MHz) of the oligomer from 4 with dynamic protection (no. 7, Table 4).
Figure 5. MALDI-TOF mass spectrum of 20-deoxyadenosine oligocarbonate (no. 7, Table 4). 2-(4-Hydroxyphenylazo)benzoic acid as a matrix.
1456
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
Scheme 4. Polymerization of Thymidine Monomers
Table 5. Polycondensation of Thymidine Monomers
fluoride anion
molecular weightb
monomer
proportion of 30-50
carbonate linkagec (%)
no.
monomer
source
eq.
solv.
conc. (M)
temp. (°C)
time (h)
yielda (%)
Mn
Mw/Mn
1
8
TBAT
TBAT
CsF
1.2
1.2
4.1
1.2
1.2
5.0
4.2
5.1
1.3
5.1
5.0
4.9
DMF
DMF
DMF
DMF
THF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
0.50
0.50
0.58
0.50
0.25
0.50
0.48
0.48
0.50
0.25
0.50
0.51
40
40
r.t.
r.t.
r.t.
r.t.
0
24
85
61
66
75
94
90
93
98
52
91
87
96
3,300
2,100
4,700
2,700
4,000
4,000
3,100
3,200
1,700
3,900
3,900
3,900
1.95
1.44
1.86
1.72
2.37
2.12
2.35
2.32
1.40
2.13
1.92
2.03
44
ꢀ
2
72
3
23
74
ꢀ
4
9
TBAT
TBAT
CsF
70
5
94.5
43.5
21.5
44.5
43.5
46.5
42
52
66
78
71
70
75
78
>90
6
7
CsF
8
CsF
0
9
10
TBAT
CsF
r.t.
r.t.
r.t.
0
10
11
12
CsF
CsF
20
a A MeOH-insoluble part. b GPC (DMFþLiBr 0.5 g/L, 40 °C, PSt std.). c Calculated from 13C NMR spectra.
(Table 3 and Supporting Information, Figure S8). The shift values
of the adenine amino proton signal were small, as compared with
those of the thymine NH proton signal, but increased with
increasing relative concentrations of the thymidine derivative,
showing the complementary H-bonding interaction.
The polycondensation of 20-deoxyadenosine TBDMS mono-
mer 4 was conducted in the presence of 30,50-O-diacylthymidine
5, 6, or 7 (Scheme 3 and Table 4). As compared with the
polymerization without them (Table 2), the formation of
the carbamate linkage was extremely reduced. The efficiency of
the dynamic protection was increased by using a three times
excess amount of 30,50-O-diacylthymidine with TBAT as the
1
fluoride anion source. The representative H NMR spectrum
(Figure 4) shows no signal due to the adenine protons that form
the carbamate linkage, which are observed in Figure 3. However,
the obtained product was the low molecular weight oligomer;
therefore, improvements in reaction conditions were investi-
gated. However, unfortunately, the longer reaction time (no. 3)
and the higher monomer concentration (no. 4) hardly increased
the molecular weight. Using THF in place of DMF as the
solvent gave almost comparable results (no. 9ꢀ11). Thymine,
which could have a smaller steric hindrance than 30,50-O-
diacylthymidine, was also found to act as the dynamic protec-
tion reagent (no. 12ꢀ13), but the product was also the
Figure 6. 13C NMR spectra of the product polymers from thymidine
monomers: nos. 1 (a), 7 (b), and 12 (c) in Table 5.
oligomer. The complementary H bonding interaction causes
1457
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
Scheme 5. Carbonate Exchange Reaction during Polymerization
this low reactivity by steric hindrance, probably affecting the
conformation of the ribose ring causing it to be unfavorable for
polycondensation. Although this drawback should be over-
come, the dynamic protection is a convenient methodology.
Actually, the protection reagent, 30,50-O-diacylthymidine, was
easily and almost quantitatively recovered as the MeOH-
soluble part from the reaction mixture. MALDI-TOF mass,
13C NMR, and IR spectra further supported the identification
of the product (Figure 5 and Supporting Information, Figures
S9 and S10); mass spectrometry suggested that linear and
cyclic oligocarbonates were produced.
Polymerization of Thymidine Monomers. Since thymidine
has no nucleophilic amino group, no problem was expected
for the monomer synthesis as well as the polymerization
(Scheme 4). Thymidine TIPS monomer 8 was found to require
heating at 40 °C for the polymerization in the presence of TBAT,
while CsF conducted the polymerization even at r.t. (no. 1ꢀ3,
Table 5). Comparing 8 with the corresponding adenosine
monomers 1 and 3, which were polymerized at r.t. with the aid
of TBAT (Tables 1, 2, and 4), suggests that the nucleobase would
indirectly affect the reactivity of the polymerization sites on the
ribose ring.
carbon signals, which were assigned to the 30-30, 30-50, and
50-50 carbonate linkages by comparison with the model dimers
(Supporting Information). This finding is ascribable to the
unfavorable side reaction of the carbonate exchange. Therein
the alkoxide anion attacks not the carbonylimidazolide for the
chain elongation but the carbonate linkage in the polymer main
chain, regenerating another alkoxide anion (Scheme 5). The
frequent carbonate exchange reactions consequently cause the
carbonate connection to lose regioregularity.
In order to exclude this unfavorable reaction, monomer 9 was
examined since the t-BuMe2Si group of 9 was expected to
decrease the polymerization temperature due to the higher
reactivity compared to that of the (i-Pr)3Si group of 8. Actually,
TBAT and CsF conducted the polycondensation of 9 at r.t.
and 0 °C, respectively (no. 4ꢀ8, Table 5). However, the
carbonate exchange reaction was comparably involved in produ-
cing the irregular carbonate linkages (Figure 6b). Accordingly,
we designed monomer 10, whose polymerization sites are
mutually replaced in contrast with 8 and 9. In monomer 10,
the silyl ether group at the 30-position generates more sterically
hindered alkoxide, while the carbonylimidazolide group at the
50-position has the higher reactivity due to the smaller steric
hindrance. These factors would be favorable for suppressing the
reaction of the alkoxide with the carbonate linkage of the
polymer main chain and relatively promoting that with the
The product polymers from 8, contrary to expectations,
contained irregular structures, as revealed by the 13C NMR
spectra (Figure 6a). There were observed three carbonyl
1458
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459

Biomacromolecules
ARTICLE
carbonylimidazolide group. The polymerization results of 10
met our expectations (no. 9ꢀ12, Table 5). The contents of the
30-50 carbonate linkage were increased; especially, the polymer
produced at 0 °C using CsF (no. 12) hardly showed the 13C
NMR signals due to the 30-30 and 50-50 carbonate carbonyl
carbons (Figure 6c). This regio regular polycarbonate from
thymidine was identified also by 1H NMR, IR, and MALDI-TOF
mass spectra (Supporting Information, Figures S11, S12, and
S13), suggesting the involvement of both the linear and the
cyclic polymers similarly to the polycarbonates from adenine
monomers 1 and 4.
(2) (a) Lutz, J.-F.; Th€unemann, A. F.; Rurack, K. Macromolecules
2005, 38, 8124. (b) Spijker, H. J.; van Delft, F. L.; van Hest, J. C. M.
Macromolecules 2007, 40, 12. (c) Mather, B. D.; Baker, M. B.; Beyer,
F. L.; Berg, M. A. G.; Green, M. D.; Long, T. E. Macromolecules 2007,
40, 6834. (d) Lo, P. K.; Sleiman, H. F. J. Am. Chem. Soc. 2009, 131, 4182.
(e) Karikari, A. S.; Mather, B. D.; Long, T. E. Biomacromolecules 2007,
8, 302. (f) Jatsch, A.; Kopyshev, A.; Mena-Osteritz, E.; Buerle, P. Org.
Lett. 2008, 10, 961. (g) Lin, I.-H.; Cheng, C.-C.; Yen, Y.-C.; Chang, F.-C.
Macromolecules 2010, 43, 1245and references cited in these articles.
(3) (a) Kobayashi, S.; Uyama, H.; Kimura, S. Chem. Rev. 2001,
101, 3793. (b) Gross, R. A.; B. Kalra, B; Kumar, A. Appl. Microbiol.
Biotechnol. 2001, 55, 655. (c) Zhaozhong, J.; Chen, L.; Wenchun, X.;
Gross, R. A. Macromolecules 2007, 40, 7934. (d) Wu, R.; Al-Azemi, T. F.;
Bisht, K. S. Biomacromolecules 2008, 9, 2921. (e) Yamamoto, Y.; Kaihara,
S.; Toshima, K.; Matsumura, S. Macromol. Biosci. 2009, 9, 968. (f) Wang,
C.-F.; Lin, Y.-X.; Jiang, T.; He, F.; Zhuo, R.-X. Biomaterials 2009,
30, 4824and references cited in these articles.
(4) (a) Acemoglu, M.; Bantle, S.; Mindt, T.; Nimmerfall, F. Macro-
molecules 1995, 28, 3030. (b) Shen, Y.; Chen, X.; Gross, R. A. Macro-
molecules 1999, 32, 2799. (c) Haba, O.; Tomizuka, H.; Endo, T.
Macromolecules 2005, 38, 3562. (d) Yokoe, M.; Aoi, K; Okada, M.
J. Polym. Sci., Polym. Chem. 2003, 41, 2312. (e) García-Martín, M. G.;
Pꢀerez, R. R.; Hernꢀandez, E. B.; Espartero, J. L.; Mu~noz-Guerra, S.;
Galbis, J. A. Macromolecules 2005, 38, 8664.
’ CONCLUSIONS
As mentioned above, we have synthesized the oligomers and
polymers from 20-deoxyadenosine and thymidine with the 30-50
regular carbonate linkages. However, further investigations in-
cluding not only chemical but also enzymatic reaction systems
are necessary for increasing the molecular weights of the product
polymers and extending this project to another pair, i.e., 20-
deoxy-guanosine and -cytidine, as well as other linking groups.
The additional interest coming next is the study of supramole-
cular interaction between two kinds of polycarbonates from the
complementary 20-deoxynucleosides.
(5) Bolton, D. H.; Wooley, K. L. J. Polym. Sci., Polym. Chem. 1997,
35, 1133.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures to
b
prepare the starting materials for the monomer syntheses, figures
showing 1H and 13C NMR and IR spectra of the polycarbonate
from 1, partial 13C NMR spectra of the model dimers, 1H NMR,
IR, and MALDI-TOF mass spectra of the polycarbonate from 2,
1H NMR spectra for the observation of hydrogen bonding
interaction between 4 and 5, 13C NMR and IR spectra of the
polycarbonate from 4 with the dynamic protection, 13C NMR
and MALDI-TOF mass spectra of the polycarbonate from 10,
and a scheme and a table of the polycondensation of monomer 2.
This material is available free of charge via the Internet at http://
pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: suzuki.masahito@nitech.ac.jp.
’ ACKNOWLEDGMENT
We thank Yuki Gosei Kogyo Co., Ltd. for gifting us 20-deoxy-
adenosine, thymidine, and triisopropylsilyl chloride. We appreciate
the contribution of Mr. Tomoaki Goto, a member of our laboratory,
for performing the control experiment to prove no reaction of the
adenine amino group with the carbonate linkage.
’ REFERENCES
(1) (a) Mertes, M. P; Coats, E. A. J. Med. Chem. 1969, 12, 154. (b)
Tittensor, J. R. J. Chem. Soc. (C) 1971, 2656. (c) Gait, M. J.; Jones, A. S.;
Walker, R. T. J. Chem. Soc. Perkin 1. 1974, 1684. (d) Calcerrada-Mu~noz,
N.; O’Neil, I.; Cosstick, R. Nucleosides, Nucleotides Nucleic Acids 2001,
20, 1347. (e) Sengupta, J.; Bhattacharjya, A. J. Org. Chem. 2008, 73, 6860.
(f) Yeo, A. N. H.; Williams, D. H. J. Chem. Soc., C 1968, 2667. (g)
Ogilvie, K. K.; Cormier, J. F. Tetrahedron Lett. 1985, 26, 4159. (h) Isobe,
H; Fujino, T.; Yamazaki, N.; Guillot-Nieckowski, M.; Nakamura, E. Org.
Lett. 2008, 10, 3729.
1459
dx.doi.org/10.1021/bm101328j |Biomacromolecules 2011, 12, 1449–1459