Solid phase synthesis of novel pyrrolidinedione analogs as potent HIV-1 integrase inhibitors

Article abstract of DOI:10.1021/cc9001026

A novel series of HIV-1 integrase inhibitors were identified from a 100 member (4R₁ × 5R₂ × 5R₃) library of pyrrolidinedione amides. A solid-phase route was developed which facilitates the simultaneous variation at R₁, R₂, and R₃ of the pyrrolidinedione scaffold. The resulting library samples were assayed for HIV-1 integrase activity and analyzed to determine the R₁, R ₂, and R₃ reagent contributions towards the activity.

Full text of DOI:10.1021/cc9001026

84
J. Comb. Chem. 2010, 12, 84–90
Solid Phase Synthesis of Novel Pyrrolidinedione Analogs as Potent
HIV-1 Integrase Inhibitors
Annapurna Pendri,* Timothy L. Troyer, Michael J. Sofia, Michael A. Walker,
B. Narasimhulu Naidu, Jacques Banville, Nicholas A. Meanwell, Ira Dicker, Zeyu Lin,
Mark Krystal, and Samuel W. Gerritz
Bristol-Myers Squibb Research and DeVelopment, 5 Research Parkway, Wallingford, Connecticut 06492
ReceiVed July 15, 2009
A novel series of HIV-1 integrase inhibitors were identified from a 100 member (4R1 × 5R2 × 5R3)
library of pyrrolidinedione amides. A solid-phase route was developed which facilitates the simultaneous
variation at R1, R2, and R3 of the pyrrolidinedione scaffold. The resulting library samples were assayed
for HIV-1 integrase activity and analyzed to determine the R1, R2, and R3 reagent contributions towards
the activity.
The human immunodeficiency virus-1 (HIV-1) remains a
3-methoxyphenoxy (FMP) resin4 via standard reductive
amination5 conditions to provide resin-bound amine 4. The
coupling of 4 with 2,2-dimethyl-5-(carboxymethylene)-1,3-
dioxolan-4-one 5 using PyBOP afforded resin-bound amide
6. Amide 6 can also be prepared by the coupling of the acid
chloride generated from dioxolanone 5 with 4. Dioxolanone
5 was readily synthesized from malic acid in five steps via
a procedure developed in-house.6 Amide 6 was then trans-
formed into pyrrolidinedione 9 via a two-step, one-pot
Mannich cyclization procedure which efficiently installs the
second (R2) and third (R3) points of diversity via aldehyde
7 and amine 8, respectively. This unusual transformation
merits additional discussion. In general, the Mannich reac-
tion7 utilizes three components: an aldehyde, a primary or
secondary amine, and a reactant bearing an active hydrogen.
In our protocol, dioxolanone amide 6 serves as the active
hydrogen component. The R2 aldehyde component and R3
amine component for the Mannich reaction constitute the
other two diversity elements in the pyrrolidinedione scaffold.
Preforming the Schiff base at 75 °C from 7 and 8 using
N-methylpyrrolidine (NMP) with 20% methanol as the
solvent followed by the addition of 6 generated the final
compound 2 in 10-30% overall yield, after cleavage from
the solid support. Final products were obtained as racemic
mixtures and no attempt was made to separate the enanti-
omers in the present work.
highly infectious and incurable disease, but the discovery
of chemotherapeutic agents which inhibit HIV-1 replication
has provided dramatic improvements in both the quality and
duration of life for patients infected with HIV-1. The targets
of approved antiretroviral agents are viral enzymes and
proteins that are indispensable for the virus to complete its
replication cycle, namely, virus entry, fusion, reverse tran-
scription, integration, and proteolytic maturation.1 The
administration of a combination of these medications rou-
tinely suppresses the HIV-1 titer to undetectable levels. While
the clinical benefit of these combination therapies is con-
siderable, the emergence of multiple drug-resistant viral
strains necessitates the discovery of new and improved anti-
HIV-1 drugs. HIV integrase is an enzyme involved in the
integration of reverse-transcribed viral DNA into host cell
DNA. HIV-1 integrase is essential for viral replication, but
is absent in host cells, making it a very attractive target for
therapeutic intervention. An inhibitor of HIV-1 integrase has
proved to be an effective addition to antiretroviral therapy.2
As part of an effort to prepare heterocyclic variants of the
diketoacid inhibitor 1,3 the pyrrolidinedione scaffold 2 was
identified as a promising HIV-1 integrase inhibitor chemotype.
We have observed that a small amount of methanol is
required for dioxolanone 6 to undergo the Mannich reaction.
We hypothesize that dioxolanone 6 itself is unreactive and
methanol converts it to the reactive enol methyl ester in situ.
Although we have never isolated the enol ester from the
reaction mixture, we have made the following observations
with the analogous solution phase reaction: (1) in the absence
of methanol the benzamide derivative of dioxolanone 5 does
not undergo the Mannich reaction; (2) the corresponding
benzamide enol methyl ester (synthesized independently)
affords the desired Mannich product; and (3) in methanolic
solutions the benzamide derivative of dioxolonone ester 5
From a library perspective, 2 represents an attractive
chemotype because it contains three points of diversity
(denoted R1, R2, and R3) which can be accessed using readily
available starting materials (vide infra). As outlined in
Scheme 1, we have developed a solid-phase synthesis of 2
that facilitates the simultaneous variation of R1, R2, and R3
under mild conditions. The first diversity element (R1) in 2
was introduced by loading primary amine 3 onto 4-formyl-
* Corresponding author. Fax: 203-677-7702. E-mail: annapurna.pendri@
bms.com.
10.1021/cc9001026 CCC: $40.75  2010 American Chemical Society
Published on Web 11/10/2009

Synthesis of Novel Pyrrolidinedione Analogs
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 1 85
Scheme 1a
a Reagents and conditions: (a) 3, 2% HOAc in DMF/(CH3O)3CH (7:3), 0.25 M NaBH(OAc)3, 25 °C, 72 h; (b) 5, PyBOP, DIEA, CH2Cl2, 25 °C, 24 h;
(c) 0.25 M R2CHO, 0.2 M R3NH2, NMP/CH3OH (4:1), 75 °C, 2 h, then add 6, 75 °C, 72 h; (d) 50% TFA/DCM, RT, 1 h.
Table 1. R1, R2, and R3 Reagent Chemsets Utilized for the Synthesis of 100 Analogs of 2
is readily converted into the corresponding methyl ester.
These observations are consistent with our hypothesis.
The application of Mannich chemistry to resin bound
substrates has been reported recently.8 However, the solid-
phase synthesis of 4-carboxamido-2,3-pyrrolidinediones via
the Mannich reaction is not precedented.9
Having developed a general solid-phase route to 2, we
synthesized a 100-member library as a 4 R1 × 5 R2 × 5 R3
array using the reagent chemsets listed in Table 1. Each
compound was analyzed by HPLC10 with UV (λ ) 220 nm)
and evaporative light scattering (ELS) detection.11 The
molecular masses of the compounds were determined by MS
using a Waters LCT time-of-flight mass analyzer, with a
multiplexed electrospray (MUX) four-way source. Of 100
compounds prepared, 75 compounds (75%) passed our
criteria of correct mass and g80% purity. The crude yields
were determined gravimetrically. The overall yield based on
initial FMP resin loading for the four step sequence ranged
between 10-30%. The yields of representative products were
also determined by 1H NMR, using 2,5-dimethylfuran
(DMFu) as an internal standard,12 and were found to be in
close agreement with the gravimetric method.
All 75 library members were tested using a previously
described HIV-1 integrase strand transfer inhibition assay,13
to assess the potency of compounds. In general, this library
was quite active; the dose-response HIV integrase inhibition
data for all 75 analogs of 2 is provided in the Supporting
Information. A summary of the data is illustrated in Table
2-15% of the library members exhibited an IC50 e 100 nM,

86 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 1
Table 2. Sample Counts by Activity Cutoff
Pendri et al.
bubble diagram in Figure 1 suggests that the median is more
representative of the activity distribution for the R1
substituentssthere are clearly more samples with IC50 < 0.1
µM for R1{1} than R1{4}. The situation is similar for Figures
2 and 3, and thus, a line representing the median is included
on each bubble diagram and the median will be used
exclusively for all further activity comparisons between
substituents.
activity cutoff
number of samples (percentage)
e50 nM
3 (4%)
e100 nM
e250 nM
e500 nM
e1000 nM
11 (15%)
19 (25%)
26 (35%)
40 (53%)
and more than half of the library members provided
submicromolar activity in the integrase assay.
A comparison of Figures 1-3 offers some insight into the
relative sensitivity of each point of diversity to structural
variation. For example, Figure 1 (R1) shows a wide range
of median IC50 values across the four substituents included
in the library, whereas the five R3 substituents in Figure 3
afford similar median IC50 values. One interpretation of this
data is that the R3 position is more tolerant to substitution
than R1, and thus, the R3 position offers an opportunity to
modulate other properties (i.e., lipophilicity, metabolic stabil-
ity) with minimal effect on integrase inhibitory potency.
Further inspection of Figure 1 suggests that the contribu-
tion of the R1 reagent toward integrase inhibitory activity
follows the trend R1{1} > R1{4} > R1{2} ) R1{3}, with the
3,4-dichlorobenzyl group for R1 (R1{1}) conferring optimal
potency for HIV-1 integrase inhibition. In fact, this group is
incorporated in every sample affording an IC50 < 0.1 µM.
Figure 2 suggests that ortho O-alkyl-substituted phenyl
groups (R2{1-4}) confer equivalent potency, even when one
of the groups (R2{3}) bears a carboxylic acid and another
bears a phenyl ring (R2{4}); in contrast, replacement of the
O-alkyl group with a CF3 substituent (R2{5}) results in a
significant loss of activity. As previously described, variation
in the R3 reagent (Figure 3; R3{1-5}) did not significantly
affect the activity, but it should be noted that R3{5} did not
appear in a single sample with an IC50 < 0.1 µM. This is
particularly interesting in light of the strong structural
similarity between R3{4} and R3{5}.
Rather than simply identifying the most potent analogs
from the library, we were interested in obtaining a deeper
understanding of the underlying structure activity relation-
ships (SARs) by leveraging the statistical analysis opportuni-
ties inherent to a library approach. Specifically, each diversity
reagent appears in ∼15-20 analogs, and this allows the use
of traditional statistical techniques to address two key
parameters: (1) the overall contribution of each point of
diversity (i.e., R1, R2, or R3) to integrase inhibition and (2)
the contribution of each reagent within a given point of
diversity to integrase inhibition. Our approach to conducting
this type of library data analysis is shown in Figures 1-3,
where each figure corresponds to a single point of diversity
and comprises a “bubble diagram” (constructed using Spot-
fireTM) and a statistical summary table. In our experience,
these bubble diagrams display qualitative and quantitative
comparisons better than tables alone for large multivariate
data sets. For example, two common ways to estimate the
center of a set of data are the sample mean and the sample
median. As shown in Figure 1, if one compares the mean
IC50 values, it appears that R1{4} (mean IC50 ) 2.47 µM)
demonstrated a comparable inhibitory profile to that of R1{1}
(mean IC50 ) 1.34 µM). However, comparison of the median
IC50 values of these two reagents reveals that samples
containing R1{4} (median IC50 ) 0.74 µM) afford substan-
tially weaker integrase activity than samples containing
R1{1} (median IC50 ) 0.09 µM). Visual inspection of the
Figure 1. Library analysis of R1 reagents vs integrase IC50 (IC50 values are represented in log scale on the y axis).

Synthesis of Novel Pyrrolidinedione Analogs
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 1 87
Figure 2. Library analysis of R2 reagents vs integrase IC50 (IC50 values are represented in log scale on the y axis).
Figure 3. Library analysis of R3 reagents vs integrase IC50 (IC50 values are represented in log scale on the y axis).
A significant outcome of this study is the observation that
the previously unexplored R2 position afforded a number of
potent compounds and thus provided a novel vector on the
pyrrolidinedione scaffold. Additionally, this study revealed
that variation in the R3 position has minimal effect on
potency, suggesting that this group can be varied to impart
better pharmaceutical properties to the molecule. Several
active samples from the library were resynthesized and
purified by preparative HPLC and the IC50 values were
determined (Table 3). The close correlation observed for the
IC50 values of crude and purified samples suggests that
reliable biological data can be obtained from the crude library
samples of established purity.
In conclusion, a novel series of HIV-1 integrase inhibitors
were identified from a 100 member (4 × 5 × 5) library of
pryrrolidinedione amides synthesized on solid support. Even
though the original library comprised crude samples (>80%
pure by HPLC), the activity of crude and resynthesized
samples showed a good correlation. This solid phase
synthesis route facilitated the rapid assembly of compounds
for SAR exploration at three diversity points on the pyrro-
lidinedione scaffold and provided access to a relatively

88 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 1
Table 3. IC50 Values for Crude and Resynthesized Library Samples
Pendri et al.
unexplored vector at R2. Statistical analysis of library sample
biological data using bubble diagrams provides an effective
visualizing tool as to the contribution of each point of
diversity as well as the contribution of each reagent within
a given point of diversity to biological activity.
3-methoxyphenoxy (FMP) resin, (Polymer Laboratories, 1.2
mmol/g, 75-150 uM mesh) and an Rf tag were loaded into
each MicroKan using the IRORI dry-resin loader and Rf tag
dispenser. All MicroKans were sorted into bins for each
diversity step using the IRORI AutoSort 10K. After every
reaction, the solvent was drained from the reaction vessel,
and each reaction batch was rinsed with DMF (2×). The
MicroKans were combined and then washed in the IRORI
Wash Station for thorough washing. Each wash included two
degas cycles where a vacuum was applied to remove air
pockets from the MicroKans and two cycles of stirring for
10 min. Finally MicroKans were washed with DMF (2×),
1:1 DMF-MeOH (2×), and DCM (3×). MicroKans were
dried under vacuum for 12 h in a vacuum oven at 30 °C
before each reaction. Cleavage of the final products from
the resin was performed in Bohdan Mini Blocks.
Experimental Section
Each compound was analyzed by HPLC with UV and ELS
detection to determine the purity of the compounds (condi-
tions: Xterra MS-C18 (2.1 × 50 mm); eluted with 0-100%
B, 2.75 min gradient (A ) 100% water-0.1% TFA and B
) acetonitrile with 0.1% TFA); flow rate at 1 mL/min; UV
detection at 220 nm and with evaporative light scattering
(ELS) detection. The molecular mass of the compounds were
determined by MS (ES) by the formula m/z using Waters
LCT time-of-flight mass analyzer, with a multiplexed elec-
trospray (MUX) 4-way source.
General Procedure for Reductive Amination (step 1).
Twenty-five MicroKans in a 100 mL glass bottle were
suspended in 2% HOAc in DMF/TMOF (50 mL), and the
appropriate R1 amine (13.5 mmol, 15 X) and NaBH(OAc)3
General Reaction Conditions with MicroKans. IRORI
MicroKan technology was used for execution of the 100
member library. Approximately 30 mg (36 µmol) of 4-formyl-

Synthesis of Novel Pyrrolidinedione Analogs
Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 1 89
(13.5 mmol, 15 X) added to make a 0.25 M solution and
shaken for 72 h. The solvent was drained into a solvent
waste, followed by initial wash with DMF (40 mL) in the
reaction vessel. After the initial wash, all the MicroKans were
pooled in the IRORI auto washer washed as described above.
General Procedure for Acylation. The MicroKans after
the step 1 reaction and washing were pooled into a 500 mL
glass bottle and were suspended in anhydrous DCM (250
mL). To this suspension was added 2,2-dimethyl-5- (car-
boxymethylene)-1,3-dioxalan-4-one (3) (2X, 1.548 g, 9
mmol), PYBOP (2X, 4.68 g, 9 mmol), and iPr2NEt (2.3 g,
18 mmol). The MicroKans were shaken for 48 h at room
temperature. The solvent was drained and the MicroKans
were washed in the IRORI Wash Station as described.
General Procedure for Cyclization. Imines were pre-
formed in 20 mL reaction vials by heating a 0.25 M solution
of R2 aldehyde (40 µmol) and R3 amine (36 µmol) in a 1.1:1
ratio in NMP (1 mL) containing 20% methanol at 70-80
°C for 2 h using a turbo coil. The MicroKans from step 2
were sorted into 5 separate bins using the IRORI Autosorter.
These were added to the reagent vials containing preformed
imine and the reaction vessels were heated with agitation at
80 °C for 72 h. The MicroKans were pooled in the IRORI
Wash Station and washed successively with DMF, DMF/
MeOH (1:1), THF, DCM, and dried in a vacuum oven.
General Procedure for Final Cleavage. The MicroKans
from step 3 were sorted into 48 well Red and Blue Bohdan
Mini Blocks fitted with polypropylene tubes. The cleavage
solution (1.5 mL, trifluoroacetic acid, DCM 1:1) was added,
and the MicroKans were shaken for 60 min. Products in the
cleavage solution in both Red and Blue Mini Blocks were
drained into a 96 deep-well block. MicroKans were rinsed
with methanol for 20 min, and the combined solutions were
dried under reduced pressure. The residues were dissolved
in MeOH and analyzed by HPLC (conditions: Xterra MS-
C18 (2.1-50 mm); eluted with 0-100% B, 2.75 min
gradient (A ) 100% water-0.1% TFA and B ) acetonitrile
with 0.1% TFA); flow rate at 1 mL/min. UV detection at
220 nm. The molecular mass of the compounds were
determined by MS (ES) by the formula m/z using Waters
LCT time-of-flight mass analyzer, with a multiplexed elec-
trospray (MUX) 4-way source.
CH2), 3.71 (m, 1H), 4.29-4.5 (2H, m, CH2), 5.463 (1H, s,
CH), 5.943-5.962 (2H, m, OCH2), 6.6-6.829 (3H, m,
aromatic), 6.829-7.412 (3H, m, aromatic), LCMS, RT 2.277
min, [M + H] 592.09 and 594.07.
Entry 3: 2{1,4,1}. 1H NMR 500 MHz (MeOD) δ (ppm):
0.95 (3H, bS, CH3), 1.564-1.68 (2H, m, CH2), 2.6-2.89
(2H, m, CH2), 3.4-3.42, (2H, m, CH2), 3.54-3.58, (2H, m,
CH2), 4.19-4.337 (2H, q, CH2), 5.0-5.07 (m, 2H, OCH2),
5.883 (1H, s, CH), 6.5-7.2 (12H, m, aromatic), LCMS, RT
2.207 min, [M + H] 569.12 and 571.12.
Entry 4: 2{1,2,2}. 1H NMR 500 MHz (MeOD) δ (ppm):
1.1-1.657 (9H, m, CH2, CH3), 1.895 (3H, s, CH3), 2.68 (2H,
m, CH2), 3.71 (m, 1H), 3.9-4.1(2H, m, OCH2) 4.2-4.4 (2H,
m, CH2), 5.83 (1H, s, CH), 6.8-7.082 (3H, m, aromatic),
7.311-7.415 (4H, m, aromatic), LCMS, RT 2.420 min, [M
+ H] 592.32.
Entry 5: 2{3,5,1}. 1H NMR 500 MHz (MeOD) δ (ppm):
0.2-0.4 (4H, m, cyclopropyl), 0.6 (1H, m, cyclopropyl),
2.8-3.0 (2H, m, CH2), 3.58 (3H, s, OCH3), 4.2-4.4 (2H,
m, CH2), 5.549 (1H, s, CH), 6.646-6.663 (2H, m, aro-
matic),6.937-6.954 (3H, m, aromatic) 7.3-7.4 (2H, m,
aromatic), 7.6 (1H, m, aromatic), LCMS, RT 2.505 min, [M
+ H] 461.31.
Acknowledgment. The authors would like to thank Peter
Tidswell, Thomas Swann, Marianne Vath, and Scott Lapaglia
for the analysis of the library.
1
Supporting Information Available. H NMR data of
purified compounds followed by LCMS data (UV absor-
bance, TIC, and mass spectrum). This material is available
free of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) (a) Greene, W. C.; Debyser, Z.; Ikeda, Y.; Freed, E. O.;
Stephens, E.; Yonemoto, W.; Buckheit, R. W.; Este, J. A.;
Cihlar, T. AntiViral Res. 2008, 80, 251–265. (b) Young, S. D.
Annu. Rep. Med. Chem. 2003, 38, 173–182. (c) De Clercq,
E. Int. J. Biochem. Cell Biol. 2004, 36, 1800–1822. (d) Este,
J. A. Curr. Med. Chem. 2003, 10, 1617–1632.
(2) (a) Nair, V.; Chi, G. ReV. Med. Virol. 2007, 17, 277–295. (b)
Evering, T. H.; Markowitz, M. Expert Opin. InVest. Drugs
2008, 17, 413–422. (c) Pais, G. C. G.; Burke, T. R., Jr. Drugs
Future 2002, 27, 1101–1111. (d) Chiu, T. K.; Davies, D. R.
Curr. Top. Med. Chem. 2004, 4, 965–979.
General Procedure for the Purification of Resynthe-
sized Compounds. Five samples from Table 3 entries 1-5
were resynthesized on solid support using loose resin. The
resulting product was purified by chromatography (YMC
Combiprep ODS-A, 30 mm × 50 mm, MeOH/H2O/0.1% TFA)
to yield the title compounds: entry 1 (36 mg, 21% yield), entry
2 (34 mg, 10% yield), entry 3 (56 mg, 27% yield), entry 4 (65
mg, 10% yield), entry 5 (56 mg, 18% yield).
(3) Wai, J. S.; Egbertson, M. S.; Payne, L. S.; Fisher, T. E.;
Embrey, M. W.; Tran, L. O.; Melamed, J. Y.; Langford, H. M.;
Guare, J. P., Jr.; Zhuang, L.; Grey, V. E.; Vacca, J. P.;
Holloway, M. K.; Naylor-Olsen, A. l. M.; Hazed, D. J.; Felock,
P. J.; Wolfe, A. L.; Stillmock, K. A.; Schleif, W. A.;
Gabryelski, L. J.; Young, S. D. J. Med. Chem. 2000, 43, 4923–
4926.
(4) Boojamra, C. G.; Burow, K. M.; Thompson, L. A.; Ellman,
J. A. J. Org. Chem. 1997, 62, 1240–1256.
(5) Bilodeau, M. T.; Cunningham, A. M. J. Org. Chem. 1998,
Analytical Data for Selected Compounds. Entry 1:
63, 2800–2801.
1
2{1,1,1}. H NMR 500 MHz (MeOD) δ (ppm): 0.25-0.5
(6) (a) Walker, M. A.; Ma, Z.; Naidu, B. N.; Sorenson, M. E.;
Pendri, A.; Banville, J.; Plamondon, S.; Remillard, R. HIV
Integrase Inhibitors. WO 2004004657. (b) Walker, M. A.;
Banville, J.; Plamondon, S.; Remillard, R. HIV Integrase
Inhibitors. WO 2004004657. For previous methods for the
synthesis of pyrrolidinones. (a) Pace, P.; Spieser, S. A. H.;
Summa, V. Bioorg. Med. Chem. Lett. 2008, 8, 3865–3869.
(b) Kawasuji, T.; Fuji, M.; Yoshinaga, T.; Sato, A.; Fujiwara,
T.; Kiyama, R. Bioorg. Med. Chem. Lett. 2007, 15, 5487–
(4H, m, cyclopropyl), 0.9 (1H, m, cyclopropyl), 2.5 (2H, m,
CH2), 4.29-4.38 (2H, m, CH2), 5.463 (1H, s, CH),
5.845-5.863 (2H, m, OCH2), 6.732-6.758 (3H, m, aro-
matic), 7.266-7.341 (3H, m, aromatic), LCMS, RT 2.587
min, [M + H] 475.23 and 477.22.
Entry 2: 2{1,1,2}. 1H NMR 500 MHz (MeOD) δ (ppm):
1.3-1.9 (6H, m, CH2), 1.992 (3H, s, CH3), 2.68 (2H, m,

90 Journal of Combinatorial Chemistry, 2010 Vol. 12, No. 1
Pendri et al.
5492. (c) Davam, R.; Al-Mawsawi, L. Q.; Neamati, N. Bioorg.
Med. Chem. Lett. 2007, 17, 6155–6159. (d) Madhav, R.;
Snyder, Carroll, A.; Southwick, P. L. J. Heterocycl. Chem.
1980, 17, 1231–1235. (e) Campbell, J. B.; Firor, J. W. J. Org.
Chem. 1995, 60, 7687–7689. (f) Larsen, S. D.; Spilman, C. H.;
Bell, F. P.; Dinh, D. M.; Martinborough, E.; Wilson, G. J.
J. Med. Chem. 1991, 34, 1721–1727. (g) Mylari, B. L.; Beyer,
T. A.; Siegel, T. W. J. Med. Chem. 1991, 34, 1011–1018. (h)
Baldwin, J. E.; Chan, M.; Fai; Gallacher, G.; Otsuka, M.
Tetrahedron 1984, 40, 4513–4526. (i) Southwick, S. J. Am.
Chem. Soc. 1949, 71, 2532–2537.
6350. (c) Jonsson, D.; Molin, H.; Unden, A. Tetrahedron Lett.
1998, 39, 1059–1062. (d) Zhang, H. C.; Brumfield, K. K.;
Jaroslova, I. Tetrahedron Lett. 1998, 39, 4449–4452.
(9) Solid phase synthesis of 3-acyl-2,4-pyrrolidinediones was
reported using amino acid ester loaded Merrifield resin, an
aldehyde and Meldrum’s acid. Romoff, T. T.; Ma, L.; Wang,
Y.; Campbell, D. A. Synlett. 1998, 12, 1341–1342.
(10) HPLC conditions: Xterra MS-C18 (2.1 × 50 mm); eluted with
0-100% B, 2.75 min gradient (A ) 100% water-0.1% TFA
and B ) acetonitrile with 0.1% TFA); flow rate at 1 mL/min.
(11) Kyranos, J. N.; Lee, H.; Goetzinger, W. K.; Li, L. Y. T.
J. Comb. Chem. 2004, 6, 796–804.
(12) Gerritz, S. W.; Sefler, A. M. J. Comb. Chem. 2000, 2, 39–41.
(13) (a) Walker, M. A.; Johnson, T.; Ma, Z.; Banville, J.;
Remillard, R.; Kim, O.; Zhang, Y.; Staab, A.; Wong, H.;
Torri, A.; Samanta, H.; Lin, Z.; Deminie, C.; Terry, B.;
Krystal, M.; Meanwell, N. Bioorg. Med. Chem. Lett. 2006,
16, 2920–2924. (b) Engelman, A.; Craigie, R. J. Virol. 1995,
69, 5908–5911.
(7) (a) Ma, D.; Ma, J.; Ding, W.; Dai, L. Tetrahedron: Asymm.
1996, 7, 2365–2370. (b) Besbes, R.; Villieras, M.; Amri, H.
Ind. J. Chem., Sect. B: Org. Chem. Incl. Med. Chem. 1997,
36, 5–8. (c) Campbell, S. E.; Comer, M. C.; Derbyshire, P. A.;
Despinoy, X. L. M.; Mcnab, H.; Morrisson, R.; Sommerville,
C. C.; Thornley, C. J. Chem. Soc., Perkin Trans. 1 1997, 15,
2195–2202. (d) Manabe, K.; Mori, Y.; Kobayashi, S. Synlett
1999, 9, 1401–1402.
(8) (a) Kobayashi, S.; Moriwaki, M.; Akiyama, R.; Suzuki, S.;
Hachiya, I. Tetrahedron Lett. 1996, 7783–7786. (b) Young-
man, M. A.; Dax, S. L. Tetrahedron Lett. 1997, 38, 6347–
CC9001026