Synthesis of 3-guaninyl- and 3-adeninyl-5-hydroxymethyl-2-pyrrolidinone nucleosides

Article abstract of DOI:10.1021/jo2004617

l- And d-glutamic acids, as well as trans-4-hydroxy-l-proline, are converted to the corresponding 3-guaninyl-5-hydroxymethyl-2-pyrrolidinone (4) or 3-adeninyl-5-hydroxymethyl-2-pyrrolidinone (5) nucleoside analog. The protecting group used to block the lactam nitrogen in key intermediates has a significant effect on the diastereoselectivity of the coupling reaction with adenine or guanine.

Full text of DOI:10.1021/jo2004617

ARTICLE
pubs.acs.org/joc
Synthesis of 3-Guaninyl- and 3-Adeninyl-5-hydroxymethyl-2-
pyrrolidinone Nucleosides
Abdullah Saleh,‡ John G. D’Angelo,§ Martha D. Morton, Jesse Quinn, Kendra Redden,|| Rafal W. Mielguz,†
Christopher Pavlik, and Michael B. Smith*
Department of Chemistry, 55 N. Eagleville Road, University of Connecticut, Storrs, Connecticut 06269-3060, United States
S Supporting Information
b
ABSTRACT: L- And D-glutamic acids, aswellastrans-4-hydroxy-
L-proline, are converted to the corresponding 3-guaninyl-5-
hydroxymethyl-2-pyrrolidinone (4) or 3-adeninyl-5-hydroxy-
methyl-2-pyrrolidinone (5) nucleoside analog. The protecting
group used to block the lactam nitrogen in key intermediates has
a significant effect on the diastereoselectivity of the coupling
reaction with adenine or guanine.
’ INTRODUCTION
synthesis and biological studies.9,8c,d Our interest in this area
began with our previous work that used enantiopure lactams as
chiral templates for synthesis. We published several papers that
converted L-glutamic acid to chiral 2-pyrrolidinone derivatives
via pyroglutamate esters.10 The planarity of the lactam ring in
5-substituted-2-pyrrolidinone derivatives is well-established.10
We speculated that 2-pyrrolidinone could be used as a re-
placement for the cyclopentene ring in 3 or similar compounds.
Such a target required the attachment of a purine or pyrimidine
base to C3 of a 5-hydroxymethyl-2-pyrrolidinone derivative. We
therefore examined synthetic routes to establish the efficacy of
using pyroglutamate and 3-hydroxyproline as chiral templates in
a proof-of-principle study to prepare the two compounds shown
in Scheme 1: 3-guaninyl-5-hydroxymethyl-2-pyrrolidinone (4)
and the 3-adeninyl-5-hydroxymethyl-2-pyrrolidinone (5). We
have completed synthetic routes to both compounds and report
their total synthesis.
Human immune-deficiency virus type 1 (HIV-1) is the
causative organism for acquired immune-deficiency syndrome
(AIDS),1 and despite great progress in chemotherapeutic treat-
ment and prevention,2 millions have lost their lives. The World
Health Organization estimates that as of 2009 33.3 million
people were living with AIDS, and 1.8 million died in 2009.3
Several forms of chemotherapy are based on key events in the life
cycle of HIV-1, including interception of the viral enzyme,
reverse transcriptase (RT).4 The HIV-RT enzyme converts the
viral RNA to proviral DNA, and there are two general classes of
RT inhibitors: nucleoside-based reverse transcriptase inhibitors
(NRTI’s)5 and non-nucleoside reverse transcriptase inhibitors
(NNRTI’s).6 Significant toxicity is associated with many NRTIs,
and there is evidence that much of this toxicity results from the
inhibition of mitochondrial DNA replication. AZT (1), for
example, is known to cause bone marrow suppression, but the
delay of disease progression often outweighs the complications
caused by treatment with AZT.7 One FDA-approved anti-HIV
NRTI treatment is abacavir (2),8 which after the intracellular
monophosphorylation, is converted to carbovir monophosphate,
which is further phosphorylated to the biologically active carbo-
vir triphosphate. Carbovir (3),8 an anti-HIV drug marketed by
GlaxoSmithKline, was first identified as a potent anti-HIV agent
in 1990. It has comparable activity to the clinically used AZT and
lower toxicity. There have been several syntheses of carbovir,
with the first in 1990 by Vince and co-workers,9a,d the discoverers
of carbovir. In the scheme reported by Vince and co-workers, a
guanine derivative was structurally modified to incorporate the
cyclopentene unit, rather than begin with a cyclopentene and
then attach a guanine unit.
’ RESULTS AND DISCUSSION
L-Glutamic acid is an attractive starting material due to its
commercial availability, low cost, and widespread use. In a study
that is highly relevant to our targeted compounds, Nielsen and
co-workers reported a synthesis of conformationally restricted
peptide nucleic acid (PNA) derivatives from D-glutamic acid
(6).11 As shown in Scheme 2, formation of 2-pyrrolidinone-5-
carboxylic acid (pyroglutamic acid) was followed by reduction to
the hydroxymethyl derivative and protection of the alcohol and
nitrogen to give the O-TBDPS, N-Boc derivative 7. R-Hydro-
xylation with MoOPH, via the lactam enolate anion, gave 8 as a
key intermediate. The PNA monomer was prepared by conver-
sion of 8 to 9 in several steps, followed by a Mitsunobu coupling
that incorporated adenine, with clean inversion of configuration
Analyses of 1ꢀ3 (see Scheme 1), as well as other related
antiviral drugs, show that a nucleobase and a hydroxymethyl unit
are attached to a relatively flat five-membered ring. The synthesis
of such compounds remains an area of interest with respect to
Received: March 2, 2011
Published: May 27, 2011
r
2011 American Chemical Society
5574
dx.doi.org/10.1021/jo2004617 J. Org. Chem. 2011, 76, 5574–5583
|

The Journal of Organic Chemistry
ARTICLE
Scheme 1. AZT (1), Abacavir (2), and Carbovir (3) and the 3-Guaninyl-5-hydroxymethyl-2-pyrrolidinone (4) and 3-Adeninyl-5-
hydroxymethyl-2-pyrrolidinone (5) Targets
Scheme 2. The Nielsen Synthesis of Peptide Nucleic Acid 10
Scheme 3. Conversion of L-Glutamic Acid to Adeninyl Derivative 17 via the Nielsen Approach
to give 10.11 We anticipated a straightforward preparation of 4 or
5 by simply modifying Nielsen’s synthetic route to obtain the free
lactam rather than 10. We chose to target the adenine derivative
first because of the extensive protecting group manipulation
that is required for coupling guanine. Initially, we converted
L-glutamic acid (11) to ethyl pyroglutamate, 12. We chose 11 for
the development work due to its greater availability and lower
price relative to D-glutamic acid, 6. Following Nielsen’s
protocol,11 the ester group was reduced with sodium borohy-
dride to give 13 and the alcohol moiety was protected as the
TBDMS derivative (14) as shown in Scheme 3. Our target was
the free lactam, so the lactam nitrogen required a protecting
group, which is a structural difference when compared to
Nielsens’s synthesis that required the preparation of 9. We chose
the N-Boc group, and 15 was generated in 23% overall yield from
12. Subsequent generation of the enolate anion and reaction with
MoOPH gave 16 in 35% yield. Coupling with adenine proceeded
smoothly under Nielsen’s Mitsunobu conditions to give 17,
but there was a problem. We obtained 17 as a 1:1 mixture of
diastereomers at C3: S-[(tert-butyldimethylsilyloxy)methyl]-3R-
adenin-9-yl-2-pyrrolidinone-N-tert-butylcarbamate þ S-[(tert-
butyldimethylsilyloxy)methyl]-3S-adenin-9-yl-2-pyrrolidinone-
N-tert-butylcarbamate. It was clear that the stereochemical integ-
rity of C3 had been compromised. For that reason, we did not
deprotect 17 to give the final target. It was not clear why this
product should be configurationally unstable when Nielsen pre-
pared 10 without a problem. The main structural difference
between 10 and 17 was the presence of the N-Boc group, and
we speculated that the presence of this group allowed epimeriza-
tion to occur. While the cause of this effect is not proven, we
5575
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583

The Journal of Organic Chemistry
ARTICLE
Scheme 4. Synthesis of 23 from 4-Hydroxy-L-proline
Scheme 5. Enantioselective Synthesis of 5 from L-Glutamic Acid and of 33 from D-Glutamic Acid
speculate that the Boc group enhances the acidity of the C3
proton, which would make it subject to epimerization.
mixture of diastereomers at C3. Interestingly, we recovered
unreacted 22 and found that the C3 stereocenter had epimerized.
In other words, the enantiopure starting material, 22, was
recovered as a mixture of diastereomers at C3 after the reaction.
Unfortunately, these experiments did not provide a suitable
solution to preparing a configurationally stable adeninyl lactam.
From previous work in our laboratory,10a and work reported
by others,14 we were aware of the influence of the nitrogen
protecting group on the reactivity of chiral 2-pyrrolidinone
derivatives. Once again, we speculate that the presence of the
N-Boc group leads to enhanced acidity of the proton at C3, which
in turn leads to configurational instability. It was clear that
Nielsen’s route led to an adenine derivative without loss of
stereochemical integrity. Proceeding on the assumption that the
presence of the N-Boc group was responsible for the configura-
tional instability, the synthetic sequence was changed to modify
the protecting group on nitrogen.
In order to probe this issue, we prepared the 3-adeninyl lactam
by an alternative route that allowed two different coupling
procedures, as outlined in Scheme 4. We modified the synthetic
sequence to use commercially available trans-4-hydroxy-L-pro-
line as a starting material. Conversion to the N-Boc methyl ester
18 (79% yield) was followed by protection of the alcohol moiety
to give 19 in 71% yield. Oxidation used the protocol of Zhang
et al.12 (ruthenium oxide and periodate)13 to give lactam 20 in
76% yield, and deprotection of the alcohol moiety with TBAF
gave 21 in 63% yield. Two routes were used to prepare the
targeted 23. The first used a Mitsunobu coupling of 21 with
unprotected adenine that gave 23 directly, in 46% yield, as a 1:1
mixture of diastereomers at C3: methyl N-tert-butoxycarbonyl-
3R-adeninylpyroglutamate and methyl N-tert-butoxycarbonyl-
3S-adeninylpyroglutamate. The second route was the less effi-
cient conversion of 21 to tosylate 22 (75% yield). Isolation of 22
confirmed that this compound was enantiopure and configur-
ationally stable. Subsequent reaction with adenine in DMF with
potassium carbonate gave 23, but as observed in the Mitsunobu
coupling with 21, the use of enantiopure 22 led to 23 as a 1:1
Frydman and co-workers reported the synthesis of bicyclic
derivative 24,15 and we also prepared it in an earlier work.10a
Conversion of glutamic acid to 24 provides protection of both
the hydroxymethyl group as well as the lactam nitrogen. The
reactions in Scheme 5 show the conversion of L-glutamic acid 11
5576
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583

The Journal of Organic Chemistry
ARTICLE
Scheme 6. Enantioselective Synthesis of 5 from trans-4-Hydroxy-L-proline
Scheme 7. Synthesis of 4
to ethyl pyroglutamate, followed by reduction to give 13 in 73%
overall yield. We prepared enantiopure 24 from 13 in 55%
isolated yield using Frydman’s procedure. Hara et al previously
reported the MoOPH hydroxylation of 24 to give 25 via the
enolate anion.16 Interestingly, 25 was obtained in 75% yield as a
single diastereomer, where syn-hydroxylation resulted from co-
ordination of the MoOPH reagent with the enolate anion derived
from 24.16 Unfortunately, the targeted stereochemistry based on
1ꢀ3 required the opposite stereochemistry for the C3 hydroxyl
group (see Scheme 1), but Mitsunobu inversion17 provided 26 in
51% yield. A second Mitsunobu reaction incorporated the
adenine moiety to give 27 in 65% yield as a single stereoisomer.
We were gratified to find that 27 was configurationally stable, and
catalytic hydrogenation provided a quantitative yield of the
targeted 5, as a completely stable compound. These results
clearly suggest that the N-Boc group in 16 and in 22 was
responsible for the configurational instability at C3 of the lactam,
whereas 26 provided a configurationally stable vehicle to the
targeted nucleobase lactam.
Given our previous problems with configurational stability,
and in order to provide further proof of stereocontrol in the
synthesis, we targeted the antipode of 5, using D-glutamic acid
(6) as the starting material. The synthetic sequence is identical
and shown for comparative purposes in Scheme 5. Initial
conversion to the hydroxymethyl lactam 28 was followed by
conversion to 29 by reaction with benzaldehyde and tosic acid.
Hydroxylation via MoOPH gave 30 and Mitsunobu inversion led
to 31. Mitsunobu coupling with adenine gave 32, and catalytic
hydrogenation gave the expected 33. As shown in Scheme 5, the
isolated yields are virtually identical to those obtained with
L-glutamic acid, except for the conversion of 32 to 33. The
diminished yield of 33 relative to the conversion of 27 to 28
5577
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583

The Journal of Organic Chemistry
ARTICLE
represents a single experiment and is not optimized. It is clear
that antipodes 5 and 33 can be prepared with equal facility using
the glutamic acid route, from L-glutamic acid and D-glutamic acid,
respectively
for several hours, filtered, and then vacuum distilled. n-Butyllithium was
standardized before use, using diphenylacetic acid.20 Water-sensitive and
air-sensitive reactions were carried out under a nitrogen blanket. Flash
chromatography was performed using silica gel (60 Å, 32ꢀ63 μm).
MoOPH was prepared according to literature methods.21 The follow-
ing compounds used for the preparation of key compounds were
prepared using literature procedures. Details for the synthesis of each
compound listed here are provided in the cited reference, and in the
Supporting Information: ethyl 2-pyrrolidone-5S-carboxylate (12),11a,d,e
ethyl 2-pyrrolidinone-5R-carboxylate,22 trans-4-hydroxy-L-proline methyl
ester hydrochloride salt,23 5S-(hydroxymethyl)-2-pyrrolidinone (13),24
N-tert-butoxycarbonyl-trans-4-hydroxy-L-proline methyl ester (18),25
5R-(hydroxymethyl)-2-pyrrolidinone (28),11dꢀf,h N-tert-butoxycarbonyl-
trans-3-tert-butyldimethylsilyl ether L-proline methyl ester (19),26 methyl
5S-N-tert-butoxycarbonyl-3R-tert-butyldimethylsilyloxypyrogluta-
mate (20),27 methyl 5S-N-tert-butoxycarbonyl-3R-hydroxypyroglutamate
(21),28 5S,8R-phenyltetrahydropyrrolo[1,2-O]oxazol-2-one (24),29 3S-
hydroxy-5S,8R-phenyltetrahydropyrrolo[1,2-O]oxazol-2-one (25),29 3S-hy-
droxy-5S,8S-phenyltetrahydropyrrolo[1,2-O]oxazol-2-one (26),30,31 N2-
isobutyrylguanine monohydrate (37),33a,b 9-acetyl-N2-isobutyrylguanine
(38),33 and N2-isobutyryl-O6-[2-(p-nitrophenyl)ethyl]guanine (39).33
GCꢀMS spectra were obtained with an HP-1 column gas chromato-
gramꢀmass spectrometer. 1H and 13C NMR were respectively collected
at 300.13 MHz for proton and at 75.48 MHz for carbon and at 400.144
MHz for proton and 100.65 MHz for carbon. In all cases, CDCl3 was
used as a solvent unless otherwise stated; chemical shifts are in ppm (δ)
relative to TMS as an internal standard. All IR spectra are reported
in cmꢀ1. Mass spectra (MS) [displayed as m/z (% base peak)] were
recorded via GCꢀMS or via LCꢀMS.
We next targeted 5 using trans-4-hydroxy-L-proline as the
chiral template for comparative purposes, but we modified
the previous sequence in Scheme 4 to prepare 26 as the key
intermediate. Using the methodology described in Scheme 4
we prepared 19, and standard deprotection at nitrogen gave
34 in 63% yield, as shown in Scheme 6. In this case, 34 gave
poor yields of 35 when reduced with sodium borohydride.
Reduction of the ester moiety with lithium borohydride,18
however, gave a satisfactory 76% yield of 35. The reaction of
35 with benzaldehyde and tosic acid gave the expected 36, but
in only 36% yield. This reaction is the result of a single
experiment and was not optimized. Deprotection of the
3-hydroxy unit gave 26 in 70% yield, and subsequent reaction
with adenine under Mitsunobu conditions gave 27 in 25%
yield. Deprotection by catalytic hydrogenation gave 5 in
quantitative yield.
We next applied this methodology to the synthesis of the
guanine derivative, which is the formal analog of 2 or 3 from
Scheme 1, using 26 as the key intermediate for introduction of
the hydroxymethyl lactam unit. It is well-known that the free
amine group in guanine requires protection prior to coupling
at N9. Guanine was therefore converted to isobutryl amide 37
and then acetylated to give 38 in 85% and 80% yield,
respectively (see Scheme 7). The lactam moiety of 38 was
protected as a nitrophenylethanol moiety to give 39 in 65%
yield. This elaborate protection strategy allowed 39 to under-
go a Mitsunobu coupling reaction at N9 with 26 to give 40 in
75% yield, with the expected inversion of configuration at the
R-carbon of the 2-pyrrolidinone unit. Deprotection of the
nitrophenylethanol unit gave 41, but attempts to reduce the
oxazolidine ring by catalytic hydrogenation gave the N-benzyl
lactam 42 rather than the unsubstituted lactam. However, the
reaction of 41 with 7 N ammonia-in-methanol19 gave 43 in
94% yield, and subsequent hydrogenation gave the targeted 4,
but in only 19% yield. The regiochemistry at N9 was estab-
lished by HMBC, which clearly showed the correlation
between H30 and C4.
(5S)-{[(tert-Butyl)dimethylsilyloxy]methyl}-2-pyrrolidi-
none, 14. A mixture of 3.0 g of 13 (26.0 mmol) and 4.4 g (65.0 mmol)
of imidazole dissolved in 15 mL of DMF was treated with 4.7 g (31.3
mmol) of chloro-tert-butyldimethylsilane. The reaction mixture was
stirred for 24 h at ambient temperature. At this time, 75 mL of Et2O was
added and the crude product was washed with water and brine and dried
over anhydrous Na2SO4, and the solvents were removed in vacuo.
Purification by flash chromatography (100% EtOAc) gave 4.45 g of
(5S)-{[(tert-butyl)dimethylsilyloxy]methyl}-2-pyrrolidinone, 14,11 as a
yellow oil (19.40 mmol, 75%). 1H NMR: δ 0.00 (s, 6H), 0.87 (s, 9H),
1.60ꢀ2.50 (m, 4H), 3.35ꢀ3.80 (m, 3H), 6.65 (br s, NH). 13C NMR: δ
0.4, 18.5, 23.3, 25.2, 30.3, 56.2, 67.0; 178.8. IR (film): 3225 (NH); 1700
(CdO) cmꢀ1. MS: 229 (1, M), 200 (4), 172 (100, B), 158 (54), 155
(10), 128 (22), 116 (24), 84 (50), 75 (53), 74 (24), 73 (51), 59 (23), 58
(18), 55 (17). HR-TOF MS: calcd for C11H24NO2Si (M þ Hþ) m/z
230.2576, found m/z 230.1569; calcd for C11H23NO2SiNa m/z
252.1396 (M þ Naþ), found m/z 252.1388.
In conclusion, we have shown that it is possible to attach a
purine base to the C3 position of 5-hydroxymethyl-2-pyrrolidi-
none. The choice of protecting groups is quite important, as the
use of N-Boc derivatives for coupling with the nucleobase led to
configurational instability, which may be due to enhanced acidity
of the proton at C3. The bicyclic derivative 26 is prepared from
L-glutamic acid or from trans-4-hydroxy-L-proline, and 31 can be
prepared from D-glutamic acid. Both the 3-adeninyl derivative 5
and 3-guaninyl derivative 4 can be prepared by these methods.
Incorporation of other nucleobases via similar methodology may
be feasible, although we anticipate protecting group issues in the
synthesis of other derivatives. This work constitutes a successful
proof-of-principle for the preparation of 3-nucleobase analogs of
the antiviral compounds 1ꢀ3.
(2S)-5{[(tert-Butyl)dimethylsilyloxy]methyl}-2-pyrrolidi-
none-1-N-tert-butylcarbamate, 15. A solution of 1.34 g (6.0
mmol) of 14 in 25 mL of CH2Cl2 was treated with 2.61 g (11.6 mmol)
of (Boc)2O, 0.71 g (6.0 mmol) of DMAP, and 0.81 mL (6.0 mmol) of
Et3N, respectively. The reaction mixture was stirred at ambient tem-
perature for 24 h, and 25 mL of Et2O was added. The ether was washed
with 10% citric acid, satd NaHCO3, and brine. The organic layer was
dried over anhydrous Na2SO4, and the solvents were evaporated in
vacuo. Purification by flash chromatography (CH2Cl2ꢀEtOAc, 15:1)
gave 1.9 g of (2S)-5{[(tert-butyl)dimethylsilyloxy]methyl}-2-pyrrolidi-
none-1-N-tert-butylcarbamate, 15,11 as a yellow oil (5.5 mmol, 93%). 1H
NMR: δ 0.00 (s, 6H), 0.85 (s, 9H), 1.30 (s, 9H), 1.70ꢀ2.60 (m, 4H),
3.40ꢀ4.00 (m, 3H); 13C NMR: δ 1.2, 18.4, 21.4, 25.9, 28.3, 32.6, 59.1,
64.6, 82.6, 150.2, 174.9. MS: 257 (2). 256 (9), 216 (41), 172 (100, B),
156 (51), 100 (13), 84 (21), 75 (20), 73 (33), 59 (11), 58 911), 57 (68),
55 (13). HR-TOF MS: calcd for C16H32NO4Si m/z 330.2101, found
m/z 330.2094; calcd for C16H32NO4SiNa m/z 352.1920 (M þ Naþ),
found m/z 352.1902.
’ EXPERIMENTAL SECTION
All chemicals and reagents were used as received unless otherwise
stated. All solvents were dried according to standard procedures. THF
was distilled from sodium benzophenone; methylene chloride, toluene,
and benzene were distilled from CaH2; and DMF was stirred with CaH2
5578
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583

The Journal of Organic Chemistry
ARTICLE
5S-[(tert-Butyldimethylsilyloxy)methyl]-3R-hydroxy-2-
pyrrolidinone-N-tert-butylcarbamate, 16. Dropwise addition of
9.0 mL (9.0 mmol, 1.0 M in hexane) of n-butyllithium to a solution of
1.9 mL (9.0 mmol) of hexamethyldisilazane dissolved in 20 mL of THF
was followed by cooling to ꢀ78 °C and stirring at ꢀ78 °C for 30 min. At
this time, 1.36 g (3.0 mmol) of 15, dissolved in 20 mL of THF and
cooled to ꢀ78 °C, was added in 10 portions. The reaction mixture was
stirred at ꢀ78 °C for 30 min and then warmed to ꢀ40 °C over a period
of 40 min. A total of 2.61 g (6.0 mmol) of MoOPH was added in two
portions, and the reaction mixture assumed a greenish color. After
stirring for 1 h at ꢀ40 °C, half-saturated NH4Cl was added to quench the
reaction, and THF was removed in vacuo. The aqueous phase was
extracted with EtOAc, the combined organic phases were washed with
brine and dried (anhydrous Na2SO4), and the solvents were evaporated
in vacuo. The crude materials were combined and purified by flash
chromatography (gradient solvent system: EtOAc in hexane from 2:3 to
2:1) to give 0.49 g of 16 (1.4 mmol, 47%).34 1H NMR: δ 0.00 (s, 6H),
0.85 (s, 9H), 1.50 (s, 9H), 2.00ꢀ2.50 (m, 2H), 3.50ꢀ4.00 (m, 2H),
4.10ꢀ4.20 (m, 1H), 4.55ꢀ4.65 74 (m, 1H). 13C NMR: δ 1.4, 18.6, 26.2,
28.6, 32.0, 56.5, 64.2, 70.4, 83.8, 150.1, 175.9. MS: 272 (4), 232 (33), 189
(13), 188 (100, B), 140 (17), 116 (43), 100 (21), 75 (24), 73 (27), 59
(13), 57 (51), 56 (11). HR-TOF MS: calcd for C16H32NO5Si m/z
346.2050, found m/z 346.2047; calcd for C16H31NO5SiNa m/z
368.1869 (M þ Naþ), found m/z 368.1862.
suspension, which was stirred for a total of 4 days, during which time a
pink color was observed. The THF was evaporated in vacuo and the
residue taken up in CHCl3 and filtered to remove adenine and other
chloroform-insoluble byproducts. Evaporation gave a rust-colored oil
that was purified by flash chromatography (silica gel, EtOAc, then 10%
MeOH/EtOAc) to give 0.32 g of 23 (0.85 mmol, 45%) as a 1:1 mixture
of C3 diastereomers (designated A and B) as determined by NMR. 1H
NMR: δ 1.48 (s, 9H), 2.65ꢀ2.70 (m, 1H), 2.99ꢀ3.08 (m, 1H), 3.77 (B),
3.82 (A) (d, 3H, J = 7.5 Hz), 4.63ꢀ4.67 (B, t), 4.82ꢀ4.85 (A, d, J = 9 Hz)
(1H), 5.35ꢀ5.38 (A, t, J = 11 Hz), 5.42ꢀ5.47 (B, t) (1H), 6.15 (A), 6.18
(B) (m, 2H), 7.82 (A), 7.91 (B) (2s, 1H), 8.25, 8.27 (2s, 1H). 13C NMR:
δ 28.25, 28.29, 28.8, 30.3, 31.3, 53.4, 53.51, 55.52, 56.1, 77.7, 85.1, 85.6,
140.2, 149.3, 150.3, 153.5, 156.1, 168.0, 171.5, 180.1. IR (KBr): 3327,
3165, 2963, 1792, 1759 cmꢀ1. HR-TOF MS: calcd for C16H20N6O5Na
m/z 399.1393 (M þ Na), found m/z 399.1413.
Method B. Adenine (0.108 g, 0.8 mmol) was added to 20 mL of DMF
under a N2 purge. Sodium hydride (60% dispersion in oil, 0.032 g,
0.8 mmol) was added and the suspension became a solution after a few
minutes, at which time a solution of 0.3 g (0.7 mmol) of 22 in DMF was
added via an addition funnel. After stirring overnight, the solution was an
orange color, and stirring continued at ambient temperature for a total of
3 days. The solvent was evaporated in vacuo and purification by flash
chromatography (silica gel, 5% MeOH/ethyl acetate) gave 56 mg of 23
(150 mmol, 21%) as a 1:1 mixture of C3 diastereomers.
3S-N9-Adeninyl-5S,8R-phenyltetrahydropyrrolo[1,2-O]
oxazol-2-one, 27. A solution of 177.5 mg (0.81 mmol) of 26 in
dioxane was treated with 0.524 g (2.0 mmol) of PPh3, followed by 0.546 g
(4.04 mmol) of adenine. Addition of 0.39 mL (0.40 g, 2.0 mmol) of DIAD
to this suspension, in two portions at ambient temperature, was
followed by stirring at ambient temperature for 24 h. The solvents
were evaporated in vacuo, and the residue was purified by column
chromatography (EtOAc then 15% MeOH in CH2Cl2) to give 0.177
g of 27 as a white solid (0.53 mmol, 65%). Mp: >300 °C. 1H NMR:
δ 2.35ꢀ2.45 (m, 1H) and 2.90ꢀ3.10 (m, 1H), 3.85ꢀ3.90 (m, 1H),
4.10ꢀ4.15 (m, 1H), 4.28ꢀ4.35 (m, 1H), 5.50ꢀ5.55 (m, 1H), 6.41
(s, 1H), 7.25ꢀ7.45 (m, 5H), 7.84 (s, 1H), 8.27 (s, 1H). 13C NMR:
δ 33.0, 55.3, 58.3, 72.3, 87.7, 120.0, 128.8, 129.2, 133.5, 133.6, 137.9,
140.1, 153.3, 170.5. HR-TF MS: calcd for C17H17N6O2 m/z
337.1413 (M þ Hþ), found m/z 337.1407; calcd for C17H16N6O2Na
m/z 359.1232 (M þ Naþ), found m/z 359.1236.
5S-[(tert-Butyldimethylsilyloxy)methyl]-3R-adenin-9-yl-2-
pyrrolidinone-N-tert-butylcarbamate, 17. Slow addition of
0.656 g (2.5 mmol) of PPh3 followed by 0.675 g (5.0 mmol) of adenine
and 0.404 g (2.0 mmol) of diisopropyldiazodicarboxylate (DIAD) to a
solution of 0.345 g (1.0 mmol) of 16 dissolved in 10 mL of dry THF was
followed by stirring for 48 h. Solvents were evaporated in vacuo, and
purification by flash chromatography (100% EtOAc then 5% MeOH in
EtOAc and 10% MeOH in EtOAc) afforded 0.218 g of 17 (0.5 mmol,
50%). 1H NMR: δ 0.00 (s, 6H), 0.85 (s, 9H), 1.55 (s, 9H), 2.60ꢀ2.90
(m, 2H), 3.7ꢀ4.2 (m, 2H), 4.38ꢀ4.43 (m, 1H), 5.78ꢀ5.88 (m, 1H),
7.80 (s, 1H), 8.30 (s, 1H). 13C NMR: δ 1.5, 18.7, 26.4, 28.5, 30.1, 56.5,
56.9, 64.7, 84.4, 120.2, 140.4, 150.1, 150.6, 153.5, 155.9, 169.4. HR-TOF
MS: calcd for C21H35N6O4Si m/z 463.2489 (M þ Hþ), found m/z
463.2496; calcd for C21H34N6O4SiNa m/z 485.2308 (M þ Na), found
m/z 485.2299.
Methyl 5S-N-tert-Butoxycarbonylpyroglutamate 3R-p-To-
luenesulfonate, 22. A solution of 21 (2.5 g, 9.6 mmol) in dry THF
(25 mL) was immersed in a dry ice/acetone bath. Triethylamine
(3.1 mL, 22.2 mmol) was added dropwise via a plastic syringe. Tosyl
chloride (1.8 g, 9.4 mmol) in 15 mL of THF was added by addition
funnel and the dry ice/acetone bath was allowed to warm to ambient
temperature. The reaction was subsequently stirred under nitrogen for
4 days. The resulting mixture was partitioned between 100 mL of CHCl3
and 100 mL of water, and the organic layer was washed with 1 M HCl
and then water (3 ꢁ 30 mL each). The combined organic phases were
washed with brine and dried over MgSO4. Filtration was followed by
evaporation of the solvent in vacuo to give a solid that was crystallized
from diethyl ether, to give 2.62 g of 22 (6.34 mmol, 66%) as a white solid.
Mp: 144ꢀ147 °C. 1H NMR: δ 1.45 (s, 9H), 2.4ꢀ2.48 (m, 1H), 2.42 (s,
3H), 2.57ꢀ2.63 (m, 1H), 3.77 (s, 3H), 4.58ꢀ4.61 (m, 1H), 5.03ꢀ5.07
(m, 1H), 7.31ꢀ7.33 (m, 2H), 7.82ꢀ7.84 (m, 2H). 13C NMR: δ 21.7,
27.8, 29.6, 53.0, 55.2, 74.1, 84.8, 128.2, 129.0, 132.7, 145.5, 148.7, 166.1,
170.9. IR (KBr): 3008, 1773, 1752, 1722, 1309 cmꢀ1. HR-TOF MS:
calcd for C18H23NO8NaS m/z 436.1042, found m/z 436.1009.
Methyl N-tert-Butoxycarbonyl-3-adeninylpyroglutamate,
23. Method A. A mixture of 21 (0.150 g, 1.9 mmol), adenine (1.28 g,
9.5 mmol), and triphenylphosphine (1.74 g, 6.65 mmol) in 20 mL of
THF was stirred for several minutes, and diisopropylazadicarboxylate
(0.77 mL, 3.9 mmol) was added dropwise via plastic syringe, over a
period of 80 min. The resultant reaction mixture was a creamy orange
3S-Adenin-9-yl-5S-(hydroxymethyl)-2-pyrrolidinone, 5. A
suspension of 26 mg of 10% Pd-on-C in MeOH was purged at least five
times by evacuation of the reaction flask and then filling with hydrogen,
while the solution was stirred vigorously. A solution of 0.021 g
(0.06 mmol) of 27 in 10 mL of MeOH was added, followed by the
addition of 0.1 mL of concd HCl. The reaction was stirred for 13 h under
a hydrogen atmosphere (a hydrogen-filled balloon). At this time, 0.2 g of
solid NaHCO3 was added, the suspension was filtered through a Celite
mat, and the solids were washed with MeOH. The solvents were
removed in vacuo and the crude product was purified by flash chroma-
tography (20% MeOH in CH2Cl2) to give 0.0154 g of 5 as a white solid
1
(0.06 mmol, 100%). Mp: >300 °C. H NMR (D2O): δ 2.30ꢀ2.40
(m, 1H) and 2.82ꢀ2.98 (m, 1H), 3.70ꢀ3.80 (m, 1H) and 3.90ꢀ3.98
(m, 1H), 4.10ꢀ4.20 (m, 1H), 5.60ꢀ5.70 (m, 1H), 8.29 (s, 1H), 8.30
(s, 1H). 13C NMR: δ 32.7, 56.0, 58.7, 65.9, 144.7, 152.5, 155.5, 158.0,
167.9, 177.9. HR-TOF MS: calcd for C10H13N6O2 þ Hþ m/z 249.1081,
found m/z 249.1100; calcd for C10H12N6O2Naþ m/z 271.0900, found
m/z 271.0919.
5R,8S-Phenyltetrahydropyrrolo[1,2-O]oxazol-2-one, 29. A
solution of 4.9 g of 28 (42.6 mmol), 5.94 g (56.0 mmol) of benzalde-
hyde, and 0.10 g (0.58 mmol) of p-toluenesulfonic was heated at reflux in
30 mL of toluene for 16 h with a DeanꢀStark trap. The reaction was
cooled and then washed with 5% NaHCO3, satd sodium bisulfite, water,
and brine. The toluene was evaporated in vacuo and the brown oil was
5579
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583

The Journal of Organic Chemistry
ARTICLE
purified by K€ugelrohr distillation to give 4.43 g of 29 as a colorless oil
4.20ꢀ4.35 (m, 2H), 5.70ꢀ5.85 (m, 1H), 6.20 (s, 1H), 7.25ꢀ7.45
(m, 5H), 7.84 (s, 1H), 8.17 (s, 1H). 13C NMR (CD3OD): δ 33.0, 57.3,
60.3, 73.5, 88.9, 127.6, 129.7, 130.0, 130.2, 138.8, 142.6, 150.3, 157.6,
172.4. HR-TOF MS: calcd for C17H17N6O2 m/z 337.1413 (M þ Hþ),
found m/z 337.1410; calcd for C17H16N6O2Na m/z 359.1232
(M þ Naþ), found m/z 359.1238.
1
(22.0 mmol, 52%). H NMR: δ 1.80ꢀ1.90 (m, 1H) and 2.20ꢀ2.35
(m, 1H), 2.4ꢀ2.55 (m, 1H), 2.60ꢀ2.80 (m, 1H), 3.40 (t, 1H),
4.00ꢀ4.15 (m, 1H), 4.16ꢀ4.20 (dd, 1H), 6.30 (s, 1H), 7.20ꢀ7.45
(m, 5H). 13C NMR: δ 22.7, 33.4, 58.8, 71.6, 87.0, 125.9, 128.4, 129.8,
138.8, 178.1. MS (m/z): 203 (21, Mþ), 202 (100, B), 173 (9), 145 (15),
144 (20), 126 (12), 119 (11), 117 (24), 105 (60), 104 (23), 97 (10), 92
(10), 91 (23), 90 (23), 89 (22), 77 (37), 69 (12), 63 (10), 55 (20), 51
(27), 50 (11); calcd for C12H14NO2 m/z 204.1025 (M þ Hþ), found
m/z 204.1033.
3R-Adenin-9-yl-5R-(hydroxymethyl)-2-pyrrolidinone, 33.
A suspension of 120 mg of 10% Pd-on-C in MeOH was purged five
times by evacuation of the reaction flask and then filling with hydrogen,
while the reaction was stirred vigorously. A solution of 0.080 g of 32
(0.24 mmol) in 10 mL of MeOH was added, followed by 1.0 mL of
concd HCl. The reaction was stirred for 22 h under a hydrogen
atmosphere (hydrogen balloon), and 0.9 g of solid NaHCO3 was added.
The suspension was filtered through a Celite mat that was washed with
MeOH, and the solvents were evaporated in vacuo. The crude product
was purified by flash chromatography (20% MeOH in CH2Cl2) and the
resulting yellowish solid washed several times with chloroform to give
0.040 g of 33 as a white solid (0.16 mmol, 67%). Mp: >300 °C. 1H NMR
(D2O): δ 2.15ꢀ2.29 (m, 1H) and 2.67ꢀ2.80 (m, 1H), 3.52ꢀ3.77
(m, 2H) and 3.80ꢀ4.00 (m, 1H), 5.36ꢀ5.48 (m, 1H), 8.00 (s, 1H).
13C NMR (CD3OD): δ 31.5, 54.6, 57.0, 65.1, 142.4, 153.9, 157.4, 174.2.
HR-TOF MS: calcd for C10H13N6O2 m/z 249.1100 (M þ Hþ), found
m/z 249.1081; calcd for C10H12N6O2Na m/z 271.0919 (M þ Naþ),
found m/z 271.0900.
3R-Hydroxy-5R,8S-phenyltetrahydropyrrolo[1,2-O]oxazol-
2-one, 30. LDA was prepared by mixing 0.98 mL (0.71 g, 7.0 mmol) of
diisopropylamine with 9.5 mL (7.0 mmol) of n-butyllithium (0.74 M in
hexane) in dry THF at 0 °C for 1 h. Addition of 1.02 g (5.02 mmol) of 29
in THF to the LDA solution at ꢀ78 °C was followed by stirring at this
temperature for 30 min. At this time, 3.43 g (7.90 mmol) of MoOPH was
added at ꢀ78 °C and the mixture was warmed to ꢀ40 °C over a period
of 2 h. The reaction was quenched by addition of a satd aq solution of
Na2SO3 the mixture extracted with EtOAc, washed with 5% HCl and
brine, and then dried over Na2SO4. Solvents were evaporated in vacuo.
The residue was purified by column chromatography (3:1 hexaneꢀ
1
ether) to give 0.82 g of 30 as yellow oil (3.7 mmol, 75%). H NMR:
δ 1.80ꢀ1.90 (m, 1H) and 2.75ꢀ2.85 (m, 82 1H), 3.55ꢀ3.65 (m, 1H),
3.90ꢀ4.03 (m, 1H), 4.25ꢀ4.35 (m, 1H), 4.70ꢀ4.80 (m, 1H), 6.32
(s, 1H), 7.30ꢀ7.50 (m, 5H). 13C NMR: δ 36, 55, 72.97, 73.02, 87.3,
126.5, 129, 129.3, 138.3, 176.9. MS (m/z): 219 (44, Mþ), 218 (63), 175
(9), 132 (27), 107 (49), 106 (23), 105 (100, B), 98 (8), 91 (37), 90 (27),
89 (23), 79 (28), 77 (43), 69 (24), 51 (22); calcd for C12H13NO3Na
m/z 242.0793 (M þ Naþ), found m/z 242.0813.
Methyl 3R-(O-tert-Butyldimethylsilyl)-5S-pyroglutamate,
34. A stirred solution of 20 (1.00 g, 2.67 mmol) in dry CH2Cl2
(30 mL) was treated with 2.1 mL (26.7 mmol) of trifluoroacetic acid
(TFA) over a period of 20 min. The resulting mixture was stirred for 5 h,
diluted with CH2Cl2 (50 mL), and washed with satd aq NaHCO3 (3 ꢁ
20 mL) and brine. The organic phase was dried over MgSO4 and filtered,
and solvents were evaporated in vacuo to give 0.71 g of 34 as an oil
(2.6 mmol, 97%). 1H NMR: δ 0.13, 0.14 (2s, 6H), 0.89 (s, 9H),
2.27ꢀ2.34 (m, 1H), 2.44ꢀ2.5 (m, 1H), 3.76 (s, 3H), 4.24ꢀ4.27
(m, 1H), 4.30ꢀ4.34 (t, 1H, J = 7 Hz), 6.54 (s, 1H). 13C NMR:
δ ꢀ5.2, ꢀ4.6, 18.2, 25.6, 25.7, 25.8, 35.1, 52.4, 52.6, 68.9, 172.3, and
175.9. IR (neat): 3251, 2954, 2856, 1724 cmꢀ1. HR-TOF MS: calcd for
C12H24NO4Si m/z 274.1475, found m/z 274.1464.
5S-Hydroxymethyl-3R-(O-tert-butyldimethylsilyl)-2-pyr-
rolidinone, 35. The flask containing a solution of 6.99 g of 34
(25.60 mmol) in 160 mL of THF was fitted with a septum, and a N2
purge was added. The reaction was cooled to 0 °C with stirring under N2
and a 2 M solution of LiBH4 in THF (46.1 mL, 92.2 mmol) was added
over about 30 min. After stirring for 1 h at 0 °C and 1 h at ambient
temperature, the reaction was cooled to 0 °C and 2.3 mL water/mmol of
ester (59 mL) was slowly added, at which time a white solid separated.
After addition was complete, 0.9 mL of 1:1 HCl:H2O/mmol of ester
(23 mL) was carefully added (the initial addition induced a violent
reaction). The aqueous layer was extracted with CH2Cl2 (3 ꢁ 100 mL),
and the combined organic phases were washed with 1 M KOH (1 ꢁ
100 mL), water (1 ꢁ 100 mL), and brine (1 ꢁ 100 mL). Drying over
MgSO4, filtration, and evaporation of solvents in vacuo gave 4.83 g of 35
as a foam (19.68 mmol, 77%). 1H NMR: δ 0.45 (s, 6H) 0.87 (s, 9H),
2.45ꢀ2.64 (AB quartet, 2H, J = 15 Hz, 37 Hz), 3.62 (d, 2H, J = 7 H),
3.82ꢀ4.00 (m, 1H), 4.25ꢀ4.50 (m, 2H), 7.24 (s, 1H). 13C NMR:
δ ꢀ4.5, ꢀ4.0, 18.8, 26.3, 26.4, 34.6, 54.1, 65.8, 70.9, 178.3. HR-TOF
MS: calcd for C11H23NO3SiNa m/z 268.1345 (M þ Naþ), found m/z
268.1344.
3S-Hydroxy-5R,8S-phenyltetrahydropyrrolo[1,2-O]oxazol-
2-one, 31. A solution of 0.4 g (1.8 mmol) of 30 in 10 mL of dry THF
was treated with a solution of 1.10 g (9.0 mmol) of benzoic acid in 15 mL
of toluene at 0 °C and stirred for 2 min. This mixture was treated with 1.8 g
(8.9 mmol) of DIAD dropwise and the reaction mixture warmed to
ambient temperature and stirred for 24 h. EtOAc was added and the
mixture was washed with 0.5 M aq citric acid, brine, satd aq NaHCO3,
and brine again. The organic phase was dried over Na2SO4 and filtered,
and the solvents were evaporated in vacuo. The crude product was
purified by flash chromatography (4:1 hexaneꢀether then 1:1 hexaneꢀ
ether). This material (0.92 g) was dissolved in MeOH and cooled to 0 °C,
and 0.18 g (3.3 mmol) of NaOMe in MeOH was added, dropwise. The
mixture was stirred for 30 min and the reaction was quenched by addition
of half-saturated aq NH4Cl. The organic phase was extracted with EtOAc
followed by CH2Cl2. The combined organic phases were dried over
Na2SO4 and filtered, and the solvents were evaporated in vacuo. The
crude product was purified by flash chromatography (EtOAc then
9:1 CH2Cl2ꢀMeOH to give 0.200 g of 31 (0.9 mmol, 50%). 1H NMR:
δ 2.10ꢀ2.20 (m, 1H) and 2.20ꢀ2.25 (m, 1H), 3.30ꢀ3.38 (m, 1H),
4.10ꢀ4.20 (m, 2H), 4.40ꢀ4.45 (m, 1H), 6.17 (s, 1H), 7.20ꢀ7.35
(m, 5H). 13C NMR: δ 31.8, 57.21, 71.7, 74, 87.1, 126.1, 128.7, 129.0,
138.2, 177.5. MS (m/z): 219 (44, Mþ), 218 (63), 15 (9), 132 (27), 107
(49), 106 (23), 105 (100, B), 98 (8), 91 (37), 90 (27), 89 (23), 79 (28),
77 (43), 69 (24), 51 (22). HR-TOF MS: calcd for C12H13NO3Na m/z
242.0793 (M þ Naþ), found m/z 242.0813.
3R-N9-Adeninyl-5R,8S-phenyltetrahydropyrrolo[1,2-O]-
oxazol-2-one, 32. A solution of 0.100 g (0.46 mmol) of 31 in
anhydrous THF was treated with 0.300 g (1.14 mmol) of PPh3, followed
by 0.310 g (2.29 mmol) of adenine. Addition of 0.190 g (0.94 mmol) of
DIAD in two portions to this suspension, at ambient temperature, was
followed by stirring at ambient temperature for 24 h. The solvents were
evaporated in vacuo, and the residue was purified by flash chromatog-
raphy (EtOAc then 15% MeOH in CH2Cl2) to give 0.100 g of 32 as a
white solid (0.3 mmol, 65%). Mp: >300 °C. 1H NMR (CD3OD):
δ 2.51ꢀ2.61 (m, 1H) and 2.88ꢀ3.00 (m, 1H), 3.85ꢀ3.90 (m, 1H),
3R-(tert-Butyldimethylsilyloxy)-5S,8R-phenyltetrahydro-
pyrrolo[1,2-O]oxazol-2-one, 36. Benzaldehyde (2.66 mL, 25.53
mmol), 35 (4.80 g, 19.64 mmol), and p-toluenesulfonic acid (0.1 g, cat.)
were added to 100 mL tof oluene in a flask fitted with a DeanꢀStark trap,
and the solution was heated at reflux for 60 h. After cooling, EtOAc
(20 mL) was added and the organic layer was washed with 5% aq
NaHCO3 (2 ꢁ 20 mL), satd aq sodium metabisulfite (3 ꢁ 20 mL), and
5580
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583

The Journal of Organic Chemistry
ARTICLE
finally brine (1 ꢁ 50 mL). The organic layer was dried with MgSO4 and
filtered, and the solvents were evaporated in vacuo. Purification by flash
chromatography (120 g silica, 2:1 pentaneꢀether, Rf = 0.39) gave 2.33 g
of 36 as an oil (7.0 mmol, 36%). 1H NMR: δ 0.15. 0.16 (2s, 6H), 0.91
(s, 9H), 2.08ꢀ2.28 (m, 2H), 3.93ꢀ3.96 (m, 1H), 4.22ꢀ4.27 (m, 2H),
4.44ꢀ4.47 (m, 1H), 6.26 (s, 1H), 7.32ꢀ7.47 (m, 5H); 13C NMR:
δ ꢀ5.1, ꢀ4.7, 18.2, 25.7, 34.3, 56.9, 71.5, 74.9, 86.8, 126.0, 128.4, 128.6,
138.4, 176.0. IR (neat): 2930, 2857, 22.50, 1716 cmꢀ1. MS (m/z): 32,
276, 170, 129, 75. HR-TOF MS: calcd for C18H27NO3Si m/z 334.1838,
found m/z 334.1845.
6S-Hydroxy-3-phenyltetrahydropyrrolo[1,2-O]oxazol-5-
one, 26. A solution of 2.11 g of 36 (6.33 mmol) in 20 mL of dry THF
was treated with 13 mL (13 mmol) of a 1 M solution of tetrabutylam-
monium fluoride (TBAF) in THF via syringe. The mixture was stirred
for 5 h, AcOH (0.72 mL, 12.6 mmol) was added, and the mixture was
stirred for 15 min. Solvents were evaporated in vacuo directly onto silica,
and flash chromatography (60 g silica, 100% ether) gave 26 contami-
nated with AcOH. This material was stirred in pentane overnight and
filtered to give 0.97 g of 21 (4.42 mmol, 70%) as a white solid.31 Mp:
102ꢀ104 °C. 1H NMR: δ 2.18ꢀ2.33 (m, 2H), 3.39ꢀ3.45 (m, 1H), ∼4
(bs, 1H), 4.24ꢀ4.3 (m, 2H), 4.52ꢀ4.55 (m, 1H), 6.25 (s, 1H),
7.34ꢀ7.46 (m, 5H). 13C NMR: δ 31.6, 57.0, 71.4, 73.7, 86.7, 125.9,
128.4, 128.6, 137.9, 177.3. IR (KBr): 3339, 1695 cmꢀ1. MS (m/z): 218,
175, 132, 105, 91, 77.
6.20 (s, 1H), 7.20ꢀ7.60 (m, 9H), 8.15 (s, 1H). 13C NMR: δ 19.9, 33.1,
36.1, 37, 57.3, 60.3, 61.7, 68.2, 73.6, 88.8, 124, 6, 127.6, 129.7, 131.5,
132.2, 133.6, 138.8, 144, 147.8, 148.3, 153.7, 163.9, 172.4, 178.5. HR-
TOF MS: calcd for C29H30N7O6 m/z 572.2258 (M þ Hþ), found m/z
572.2254; calcd for C29H29N7O6Na m/z 594.2077 (M þ Naþ), found
m/z 594.2073.
3S-(N2-Isobutyrylguanin-9-yl)-5S,8R-phenyltetrahydro-
pyrrolo[1,2-O]oxazol-5-one, 41. Dropwise addition of 1.95 g of
DBU (1,8-diazobicyclo[5.4.0]undec-7-ene) (12.83 mmol) to a solution
of 40 (1.36 g, 12.83 mmol) in pyridine at 0 °C was followed by warming
to ambient temperature and stirring for 24 h. Glacial acetic acid (1 mL,
7.4 mmol) was added, and all solvents were evaporated in vacuo. At this
time, 1 mL of toluene was added to the resulting solid, which was purged
with a stream of nitrogen to dryness. The yellow product was washed
with EtOAc to give 41 as a white solid (0.361 g, 0.85 mmol, 36%). Mp:
1
>300 °C. H NMR (CD3ODꢀCDCl3): δ 1.13 (d, 6H, J = 10 Hz),
2.40ꢀ2.65 (m, 1H), 2.65ꢀ2.79 (m, 1H), 2.80ꢀ3.00 (m, 1H), 3.29
(d, 1H, J = 15 Hz), 3.85ꢀ3.90 (m, 1H), 4.15ꢀ4.20 (m, 1H), 4.20ꢀ4.28
(m, 1H), 5.50ꢀ5.65 (t, 1H, J = 12 Hz), 6.40 (s, 1H), 7.20ꢀ7.40 (m, 5H),
7.75 (s, 1H). 13C NMR: δ 18.9, 33.1, 35.9, 54.8, 58.4, 71.7, 87.4, 126.0,
128.7, 129.2, 137.9, 139.3, 148.0, 149.0, 156.0, 158.0, 170.9, 180.0. IR
(film): 1167, 1604, 1553 (3 CdO) cmꢀ1. HR-TOF MS: calcd for
C21H23N6O4 m/z 423.1781 (M þ Hþ), found m/z 423.1777; calcd for
C21H 22N6O4Na m/z 445.1600 (M þ Naþ), found m/z 445.1591.
3S-(2-(N-Isobutyrylguanin-9-yl))-5S-(hydroxymethyl)-N-
benzyl-2-pyrrolidinone, 42. A suspension of 300 mg of 10% Pd-on-
C in 40 mL of MeOH was purged five times by evacuation and filling
with hydrogen, while the mixture was vigorously stirred. A solution of
0.090 g of 41 (0.21 mmol) in 10.0 mL of MeOH and 1.0 mL of concd
HCl was added. The reaction was stirred for 48 h under a hydrogen
atmosphere (H2 balloon). At this time, 1.0 g of solid NaHCO3 was
added and the suspension was filtered through a Celite mat, which was
washed with MeOH. All solvents were evaporated in vacuo. The crude
product was precipitated from CHCl3ꢀMeOH to give 0.018 g
(0.05 mmol, 24%) of 42 as an off-white solid. Mp: >300 °C. 1H
NMR: δ 1.25 (d, 6H, J = 10 Hz), 2.40ꢀ2.55 (m 1H), 2.60ꢀ2.75
(m, 1H), 3.32 (s, 2H), 3.54ꢀ3.73 (m), 4.03 (d, 1H, J = 15 Hz), 4.15 and
5.21 (AB q, 2H, J = 10 Hz, 320 Hz35), 5.25ꢀ5.38 (m), 7.24ꢀ7.40
(m, 5H), 7.50 (s), 7.96 (s). HR-TOF MS: calcd for C21H25N6O4
m/z 425.1941 (M þ Hþ), found m/z 425.1973; calcd for C21H24N6O4-
Na m/z 447.1752 (M þ Naþ), found m/z 447.1757.
3R-Adeninyl-5S-hydroxymethyl 2-pyrrolidinone, 5. A solu-
tion of 26 (0.49 g, 2.23 mmol), PPh3 (0.9 g, 3.4 mmol), and adenine
(0.31 g, 2.23 mmol) in 12 mL of dry THF under N2 was treated with
DIAD (0.67 mL, 3.4 mmol) by slow addition. The mixture was stirred at
ambient temperature overnight, diluted with CHCl3 (50 mL), and
filtered to remove unreacted adenine. The CHCl3 was evaporated in
vacuo, and the residue purified by flash chromatography (30 g silica, 10%
MeOH/EtOAc) to give 0.19 g of 27 (0.56 mmol, 25%), which was taken
directly to the next step without further purification. 1H NMR:
δ 2.53ꢀ2.6 (m, 1H), 3.04ꢀ3.11 (m, 1H), 3.96ꢀ4.0 (t, 1H), 4.20ꢀ4.27
(m, 1H), 4.38ꢀ4.42 (dd, 1H), 5.61ꢀ5.66 (dd, 1H), 5.73 (bs, 1H), 6.49
(s, 1H), 7.36ꢀ7.50 (m, 5H), 7.91 (s, 1H), 8.35 (s, 1H). 13C NMR: δ 32.8,
55.0, 58.1, 72.1, 87.3, 126.1, 128.6, 129, 137.7, 138.8, 153.1, 170.3.
A slurry of 10% Pd/C (120 mg) in 40 mL of MeOH was purged three
times with hydrogen, and then 0.19 g of the freshly prepared 27 (0.56
mmol) in MeOH (20 mL) was added in one portion. After addition of
1 mL of concd HCl, the reaction was stirred at ambient temperature
under hydrogen (H2 balloon) for 17 h. Sodium bicarbonate (2 g) was
added, the suspension was filtered through Celite, and the Celite was
washed several times with MeOH. The MeOH was evaporated in vacuo
to give a pale yellow solid that was purified by flash chromatography
(10 g silica, 20% MeOH/CH2Cl2 and then 1:1 MeOH:CH2Cl2) to give
0.138 g of 5 as white solid (0.56 mmol, 99%). Mp: >300 °C. 1H NMR
(D2O, 80 °C): δ 2.70ꢀ2.79 (m, 1H), 3.09ꢀ3.23 (m, 1H), 3.33ꢀ3.40
(m, 1H), 4.15ꢀ4.33 (m, 2H), 4.48ꢀ4.57 (m, 1H), 5.95ꢀ6.03 (m, 1H),
8.66, 8.71 (2s, 2H). 13C NMR: δ 33.0, 55.3, 58.3, 72.27, 87.3, 120.0,
128.8, 129.2, 133.5, 133.6, 137.9, 140.1, 153.3. 170.5. IR (KBr): 3337,
3S-Guanin-9-yl-5S,8R-phenyltetrahydropyrrolo[1,2-O]-
oxazol-5-one, 43. A suspension of 41 (0.036 g, 0.09 mmol) and 5 mL
of 7 N NH3 in MeOH was stirred in a sealed tube at ambient tem-
perature for 60 h. The tube was opened, all volatiles were evaporated in
vacuo, and the residue was washed with 3 ꢁ CH2Cl2 to give 43 as a white
solid (0.030 g, 0.085 mmol, 94%). Mp >300 °C. 1H NMR (DMSO-d6):
δ 2.46ꢀ2.54 (m, 1H), 2.80ꢀ2.90 (m, 1H), 3.84ꢀ3.89 (m, 1H),
4.24ꢀ4.26 (m, 1H), 4.29ꢀ4.38 (m, 1H), 5.60ꢀ5.67 (m, 1H), 6.15
(s, 1H), 7.37ꢀ7.48 (m, 5H), 7.78 (s, 1H). 13C NMR (DMSO-d6):
δ 31.3, 55.7, 58, 72.5, 117.1, 126.9, 128.9, 129.3, 137.6, 139.2, 151.6,
153.8, 157.2, 171.1. HR-TOF MS: calcd for C17H17N6O3 m/z 353.1362
(M þ Hþ), found m/z 353.1366; calcd for C17H16N6O3Na m/z
375.1182 (M þ Naþ), found m/z 375.1191.
1701, 1655 cmꢀ1
.
3S-N2-Isobutyryl-O6-[2-{(p-nitrophenyl)ethyl}guanin-9-
yl)-5S,8R-phenyltetrahydropyrrolo[1,2-O]oxazol-5-one, 40.
A solution of 0.100 g (0.46 mmol) of alcohol 26 and 300 mg of PPh3
in anhydrous THF was treated with 450 mg of 39. A total of 190 mg of
DIAD was added to this suspension in two portions at ambient
temperature, and the reaction was stirred at ambient temperature for
24 h. The solvents were evaporated in vacuo and the residue was purified
by flash chromatography (EtOAc then 15% MeOH in CH2Cl2) to give
40 as a white solid (0.200 g, 0.35 mmol, 76%). Mp: >300 °C. 1H NMR
(CD3OD): δ 1.10 (d, 6H, J = 9 H), 1.90 (s, 2H), 2.52ꢀ2.65 (m, 1H),
2.65ꢀ2.79 (m, 1H), 2.80ꢀ2.90 (m, 1H), 3.20 (bd s, 1H), 3.85ꢀ4.00 (m,
2H), 4.22ꢀ4.35 (m, 1H), 4.75 (m, 2H), 5.80ꢀ5.90 (t, 1H, J = 13.5 H),
3S-Guanin-9-yl-5S-(hydroxymethyl)-2-pyrrolidinone, 4. A
suspension of 100 mg of 10% Pd-on-C in MeOH was purged seven times
by evacuation and then filled with hydrogen while the mixture was
vigorously stirred. A solution of 0.043 g of 43 (0.12 mmol) in 30 mL of
MeOH and 1.0 mL of concd HCl was added, and the reaction was stirred
for 13 h under hydrogen atmosphere (H2 balloon). At this time, 2.0 g of
solid NaHCO3 was added and the suspension was filtered through a
Celite mat, and the solids were washed with MeOH. All solvents were
evaporated in vacuo to give an amorphous, off-white solid. This solid
was insoluble in most solvents and could not be crystallized or
5581
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583

The Journal of Organic Chemistry
ARTICLE
chromatographed. Addition to 0.5 mL of D2O containing two flakes of
KOH allowed dissolution of the solid in order to obtain a proton NMR
that was consistent with that of 4 (0.006 g, 0.02 mmol, 19%). Mp: >
Lavelle, G.; Qualls, J.; Weislow, O.; Kiser, R.; Canonico, P.; Schultz, R.;
Narayanan, V.; Mayo, J.; Shoemaker, R.; Boyd, M. Biochem. Biophys. Res.
Commun. 1988, 156, 1046–1053. (e) White, E.; Parker, W.; Macy, L.;
Shaddix, S.; McCaleb, G.; Secrist, J.; Vince, R.; Shannon, W. Biochem.
Biophys. Res. Commun. 1989, 161, 393–398.
1
300 °C. H NMR (D2OꢀKOH): 2.30ꢀ2.40 (m, 1H) and 2.85ꢀ2.92
(m, 1H), 3.62 (t, 1H), 3.70ꢀ3.80 (m, 1H), 3.92ꢀ4.12 (m, 1H), 4.50 and
5.02 (AB quartet, 2H), 5.35 (m, 1H), 7.40ꢀ7.55 (m, 2H), 7.80 (s, 1H).
HR-TOF MS: calcd for C10H13N6O3 m/z 265.1049 (M þ Hþ), found
m/z 265.1059.
(9) (a) Agrofoglio, L.; Suhas, E.; Farese, A.; Condom, R.; Challand,
S. R.; Earl, R. A.; Guedj, R. Tetrahedron 1994, 50, 10611–10670.
(b) Crimmins, M. Tetrahedron 1998, 54, 9229–9272.
(10) (a) Smith, M. B. Recent Res. Dev. Org. Chem. 1998, 2, 215–240.
(b) Grant-Young, K.; Smith, M. B.; Ovaska, T.; Wang, C. J. Synth.
Commun. 1998, 28, 4233–4239. (c) Chen, P.; Tao, S.; Smith, M. B.
Synth. Commun. 1998, 28, 1641–1648. (d) Keusenkothen, P. F.; Smith,
M. B. J. Chem. Soc., Perkin Trans. 1 1994, 2485–2492. (e) Smith, M. B.;
Dembofsky, B. T.; Son, Y. C. J. Org. Chem. 1994, 59, 1719–1725.
(f) Kwon, T. W.; Keusenkothen, P.; Smith, M. B. J. Org. Chem. 1992,
57, 6169–6173. (g) Smith, M. B.; Kwon, T. W. Synth. Commun. 1992,
22, 2865–2873. (h) Keusenkothen, P. F.; Smith, M. B. Tetrahedron 1992,
48, 2977–2992. (i) Smith, M. B.; Keusenkothen, P. F.; Dembofsky, B.;
Wang, C.-J.; Zezza, C. A.; Kwon, T. W.; Sheu, J.; Son, Y. C.; Menezes,
R. F.; Fay, J. N. Chem. Lett. 1992, 247–250. (j) Keusenkothen, P. F.;
Smith, M. B. Tetrahedron Lett. 1989, 30, 3369–3372. (k) Menezes,
R. F.; Zezza, C. A.; Sheu, J. L.; Smith, M. B. Tetrahedron Lett. 1989,
30, 3295–3298.
’ ASSOCIATED CONTENT
Supporting Information. 1H and 13C NMR of new
S
b
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: michael.smith@uconn.edu
Present Addresses
‡Department of Chemistry, The Hashemite University, Zarqa
13115, Jordan.
(11) P€uschl, A.; Boesen, T.; Zuccarello, G.; Dahl, O.; Pitsch, S.;
Nielsen, P. E. J. Org. Chem. 2001, 66, 707–712.
(12) Zhang, X.; Schmitt, A. C.; Jiang, W. Tetrahedron Lett. 2001,
§Department of Chemistry, Alfred University, College of Liberal
Arts & Sciences, 1 Saxon Drive, Alfred, NY 14802.
42, 5335–5338.
(13) See(a) Sharma, N. K.; Ganesh, K. N. Tetrahedron Lett. 2004,
45, 1403–1406. (b) Zhang, X.; Schmitt, A. C.; Jiang, W. Tetrahedron Lett.
2001, 42, 5335–5338.
(14) (a) Thottathil, J. K.; Moniot, J. L.; Mueller, R. H.; Wong, M. K. Y.;
Kissick, T. P. J. Org. Chem. 1986, 51, 3140–3143. (b) Hanessian, S.;
Ratovelomanana, V. Synlett 1990, 501–503. Also see(c) Silverman, R. B.;
Levy, M. A. J. Org. Chem. 1980, 45, 815–818. (d) Ackermann, J.; Matthes,
M.; Tamm, C. Helv. Chim. Acta 1990, 73, 122–132.
DISCLOSURE
NSF R.E.U. Summer Scholar, 2002.
Notes
†Deceased, May 29, 2010.
(15) Valasinas, A.; Frydman, B.; Friedmann, H. C. J. Org. Chem.
1992, 57, 2158–2160.
’ ACKNOWLEDGMENT
The authors thank Mr. Marvin Thompson for his assistance in
the electrospray mass spectral analysis of several compounds. We
also thank Dr. Dennis Hill, Tzipporah Kentesz, and Dr. Srikanth
Rapole, for obtaining the high-resolution mass spectral data.
(16) Hara, O.; Takizawa, J.; Yamatake, T.; Makino, K.; Hamada, Y.
Tetrahedron Lett. 1999, 40, 7787–7790.
(17) (a) Mitsunobu, O. Synthesis 1981, 1–28. (b) The Merck Index,
14th ed.; Merck & Co., Inc.: Whitehouse Station, NJ, 2006, p ONR-62.
(18) (a) Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis; Wiley,
New York, 1967; Vol. 1, p 603; (b) SeeKeusenkothen, P. F.; Smith,
M. B. Synth. Commun. 1989, 19, 2859–2868.
’ REFERENCES
(19) Borthwick, A.; Crame, A.; Exall, A.; Weingaraen, G.; Mahmoudian,
M. Tetrahedron Lett. 1995, 36, 6929–6932.
(1) (a) Petroll, A. E.; Hare, C. B.; Pinkerton, S. D. J. Infusion Nursing
2008, 31, 228–235. (b) Siddiqui, N.; Ahsan, W.; Ahuja, P.; Alam, M. S.;
Habib, S. Int. J. Pharm. Sci. Rev. Res. 2010, 4, 77–87.
(2) Gupta, S. D.; Ranteke, S.; Gupta, A.; Moorthy, N. S. H. N. Indian
J. Pharm. Sci. 2007, 69, 173–179.
(20) Duhamel, L.; Plaquevent, J. J. Org. Chem. 1979, 44, 3404–3405.
(21) Vedejs, E.; Larsen, S. Organic Syntheses ; Wiley: New York,
1990; Collect. Vol. 7, 277ꢀ282.
(22) Evans, D. A.; Lundy, K. M. J. Am. Chem. Soc. 1992, 114,
1495–1496.
(3) World Health Statistics 2011; World Health Organization:
Geneva, 2011; pp 61ꢀ62.
(23) (a) Jordis, U.; Sauter, F.; Siddiqi, S.; Suhaib, M.; Kueenburg, B.;
Bhattacharya, K. Synthesis 1990, 925–930. (b) Evans, P. J. Org. Chem.
2007, 72, 1830–1833.
(24) Ackermann, J.; Matthes, M.; Tamm, C. Helv. Chim. Acta 1990,
73, 122–132.
(4) (a) Granich, R.; Crowley, S.; Vitoria, M.; Smyth, C.; Kahn, J. G.;
Bennet, R.; Lo, Y.-R.; Souteyrand, Y.; Williams, B. Curr. Opin. HIV AIDS
2010, 5, 298–304. (b) Shehu-Xhilaga, M.; Tachedjian, G.; Crowe, S. M.;
Kedzierska, K. Curr. Med. Chem. 2005, 12, 1705–1719.
(5) Macchi, B, M.A. Pharmacol. Res. 2002, 46, 473–482.
(25) (a) Dormoy, J.-R. Synthesis 1982, 753–756. (b) Chen, K. X.;
Njoroge, F. G.; Vibulbhan, B.; Buevich, A.; Chan, T.-M.; Girijavallabhan,
V. J. Org. Chem. 2002, 67, 2730–2733.
(26) Rossen, T.; Chu, D. T. W.; Lico, I. M.; Fernandes, P. B.; Marsh,
K.; Shen, L.; Cepa, V. G.; Pernet, A. G. J. Med. Chem. 1988, 31,
1598–1611.
(27) Merino, P.; Arono, S.; Franco, S.; Merchan, F. L.; Tejero, T.;
Turon, V. J. Org. Chem. 2000, 65, 1590–1596.
(28) Zhang, X.; Schmitt, A. C.; Jiang, W. Tetrahedron Lett. 2001,
42, 5335–5338.
(6) Figueiredo, A.; Moore, K. L.; Mak, J.; Sluis-Cremer, N.;
de Bethune, M.-P.; Tachedjian, G. PLoS Pathogens 2006, 2, 1051–1059.
(7) (a) Barre-Sinoussi, F.; Chermann, J.; Rey, F.; Nugeyre, M.;
Chamaret, S.; Gruest., J.; Dauget, C.; Axler-Blin, C.; Vezinet-Brun, F.;
Rouzioux, X.; Rozenbaum, W.; Mantaininer, L. Science 1983,
220, 868–871. (b) Gallo, R.; Salahuddin, S.; Popovic, M.; Shearen, G.;
Kaplan, M.; Haynes, B.; Palker, T.; Redfield, R.; Oleske, J.; Safai, B.
Science 1984, 224, 500–503. (c) Gallo, R. Science 2002, 298, 1728–1730.
(8) (a) Vince, R.;Hua, M. J. Med. Chem. 1990, 33, 17–21. (b) Daluge, S.
M. U. S. Therapeutic Nucleosides. U. S. Patent 5,034,394; 1991.
(c) Huryn, D. M.; Okabe, M. Chem. Rev. 1992, 92, 1745–1768.
(d) Vince, R.; Hua, M.; Brownell, J.; Daluge, S.; Lee, F.; Shannon, W.;
(29) Thottathil, J.; Moniot, J.; Mueller, R.; Wong, M.; Kissick, T.
J. Org. Chem. 1986, 51, 3140–3143.
5582
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583

The Journal of Organic Chemistry
ARTICLE
(30) Hara, O.; Takizawa, J.; Yamatake, T.; Makino, K.; Hamada, Y.
Tetrahedron Lett. 1999, 40, 7787–7790.
(31) Thottathil, J.; Moniot, J.; Mueller, R.; Wong, M.; Kissick, T.
J. Org. Chem. 1986, 51, 3140–3143.
(32) (a) Robins, M. J.; Zou, R.; Guo, Z.; Wnuk, S. F. J. Org. Chem.
1996, 61, 9207–9212. (b) This work noted that the carbons of guanine
did not appear in the 13C NMR.
(33) (a) Jenny, T. F.; Schneider, K. C.; Benner, S. A. Nucleosides
Nucleotides 1992, 11, 1257–1261. (b) See also Tsai, J.-Y.; Bouhadir,
K. H.; Zhou, J.-L.; Webb, T. R.; Sun, Y.; Shevlin, P. B. J. Org. Chem. 2003,
68, 1235–1241.
(34) P€uschl, A.; Boesen, T.; Zuccarello, G.; Dahl, O.; Pitsch, S.;
Nielsen, P. J. Org. Chem. 2001, 66, 707–712.
(35) The large coupling constant for an AB quartet in this pyrroli-
dinone system is well-documented; see: (a) Smith, M. B.; Dembofsky,
B. T.; Son, Y. C. J. Org. Chem. 1994, 59, 1719–1725. (b) Latypov, S.;
Riguera, R.; Smith, M. B.; Polivkova, J. J. Org. Chem. 1998, 63,
8682–8688. (c) Chen, P.; Tao, S.; Smith, M. B. Synth. Commun. 1998,
28, 1641–1648.
5583
dx.doi.org/10.1021/jo2004617 |J. Org. Chem. 2011, 76, 5574–5583