Discovery of raltegravir, a potent, selective orally bioavailable HIV-integrase inhibitor for the treatment of HIV-AIDS infection

Article abstract of DOI:10.1021/jm800245z

Human immunodeficiency virus type-1 (HIV-1) integrase is one of the three virally encoded enzymes required for replication and therefore a rational target for chemotherapeutic intervention in the treatment of HIV-1 infection. We report here the discovery of Raltegravir, the first HIV-integrase inhibitor approved by FDA for the treatment of HIV infection. It derives from the evolution of 5,6-dihydroxypyrimidine-4-carboxamides and N-methyl-4- hydroxypyrimidinone-carboxamides, which exhibited potent inhibition of the HIV-integrase catalyzed strand transfer process. Structural modifications on these molecules were made in order to maximize potency as HIV-integrase inhibitors against the wild type virus, a selection of mutants, and optimize the selectivity, pharmacokinetic, and metabolic profiles in preclinical species. The good profile of Raltegravir has enabled its progression toward the end of phase III clinical trials for the treatment of HIV-1 infection and culminated with the FDA approval as the first HIV-integrase inhibitor for the treatment of HIV-1 infection.

Full text of DOI:10.1021/jm800245z

J. Med. Chem. 2008, 51, 5843–5855
5843
Discovery of Raltegravir, a Potent, Selective Orally Bioavailable HIV-Integrase Inhibitor for the
Treatment of HIV-AIDS Infection
Vincenzo Summa,*,† Alessia Petrocchi,† Fabio Bonelli,† Benedetta Crescenzi,† Monica Donghi,† Marco Ferrara,†
Fabrizio Fiore,† Cristina Gardelli,† Odalys Gonzalez Paz,† Daria J. Hazuda,‡ Philip Jones,† Olaf Kinzel,† Ralph Laufer,†
Edith Monteagudo,† Ester Muraglia,† Emanuela Nizi,† Federica Orvieto,† Paola Pace,† Giovanna Pescatore,† Rita Scarpelli,†
Kara Stillmock,‡ Marc V. Witmer,‡ and Michael Rowley†
Istituto Di Ricerche Di Biologia Molecolare, P. Angeletti SpA (Merck Research Laboratories, Rome), Via Pontina Km 30, 600,
00040 Pomezia, Italy, Department of AntiViral Research, Merck Research Laboratories, West Point, PennsylVania
ReceiVed March 6, 2008
Human immunodeficiency virus type-1 (HIV-1) integrase is one of the three virally encoded enzymes required
for replication and therefore a rational target for chemotherapeutic intervention in the treatment of HIV-1
infection. We report here the discovery of Raltegravir, the first HIV-integrase inhibitor approved by FDA
for the treatment of HIV infection. It derives from the evolution of 5,6-dihydroxypyrimidine-4-carboxamides
and N-methyl-4-hydroxypyrimidinone-carboxamides, which exhibited potent inhibition of the HIV-integrase
catalyzed strand transfer process. Structural modifications on these molecules were made in order to maximize
potency as HIV-integrase inhibitors against the wild type virus, a selection of mutants, and optimize the
selectivity, pharmacokinetic, and metabolic profiles in preclinical species. The good profile of Raltegravir
has enabled its progression toward the end of phase III clinical trials for the treatment of HIV-1 infection
and culminated with the FDA approval as the first HIV-integrase inhibitor for the treatment of HIV-1 infection.
Introduction
pursued as novel anti-HIV agents: Enfuvirtide,2 an injectable,
peptidic anti-HIV drug blocking gp-41-mediated fusion, was
licensed in 2003, and Maraviroc,3 which has recently been
approved. Although over the years a plethora of HIV-1 integrase
inhibitors have been discovered,4 Raltegravir is the first drug
with this mechanism of action to be approved by FDA.
Proof-of-concept for the mechanism was first shown in an
experimental model of retroviral infection using rhesus macaques
infected with SHIV-89.6P with 8-hydroxy[1,6]-naphthyridine
carboxamides.5 The robust antiviral activity in this model
supplied evidence of in vivo efficacy for integrase inhibitors in
retroviral infections, thereby establishing a solid base for the
treatment of HIV-1 infection.6 Furthermore, proof-of-concept
has also been achieved in man.7
HIV-1 integrase catalyzes the insertion of the viral DNA into
the cellular genome of the host cell through a multistep process
that includes two catalytic reactions: 3′ endonucleolytic process-
ing of the viral DNA ends and strand transfer, i.e., joining of
the viral and cellular DNAs.8 Strand transfer is temporally and
spatially separated from 3′ processing and occurs after transport
of the preintegration complex from the cytoplasm into the
nucleus. Divalent metals, such as Mg2+, are required for not
only 3′ processing and strand transfer but also assembly of
integrase onto specific viral donor DNA to form a complex
competent to carry out either function. The architecture of the
amino acids within the catalytic site and the geometry of the
catalytic metals are highly conserved among a superfamily of
nucleases and polynucleotidyl transferases, including also the
NS5b RNA-dependent RNA polymerase of HCV virus. 4-Aryl-
2,4-diketobutanoic acids (DKAs) were the first class of true
HIV-1 strand transfer inhibitors and provided the first proof-
of-concept for HIV-integrase inhibitors as antiviral agents in
the cell based assay.9 DKAs were also discovered as active site
HCV NS5b RNA-dependent RNA polymerase inhibitors.10 The
mechanism of action of these inhibitors is likely a consequence
of the interaction with metals in the active site, resulting in a
functional impairment by chelation of the critical metal cofac-
Human immunodeficiency virus type 1 (HIV-1) is the
causative agent of acquired immunodeficiency syndrome (AIDS).
Despite decades of research and the successful development of
combination therapies known as highly active antiretroviral
therapy (HAART),a HIV-1 remains one of the most serious
health problems in the world. It is estimated that approximately
39 million people are living with HIV/AIDS worldwide, with
infection and death rates of around 4 million and 3 million per
year, respectively.1 Currently, 22 drugs are approved by the FDA
for the treatment of HIV infection. These are categorized
according to their mode of action into four main groups: (i) the
nucleoside reverse transcriptase inhibitors (NRTI), (ii) the non-
nucleoside reverse transcriptase inhibitors (NNRTI), (iii) the
protease inhibitors (PI), and (iv) entry inhibitors. Multidrug
cocktails consisting of a protease inhibitor or a non-nucleoside
reverse transcriptase inhibitor in combination with two nucleo-
side reverse transcriptase inhibitors is the current standard for
HIV therapy (HAART). While HAART is undeniably effective,
it can fail to control HIV replication in patients due to several
limitations such as lack of therapy adherence, occurrence of
side effects significantly affecting quality of life, and most
importantly, the loss of drug effectiveness over time caused by
development of resistance, including multidrug resistance and
cross-resistance. It is therefore essential to continue to develop
new antiretroviral drugs with potency against an ever-wider
range of viral mutants including those which engender resistance
to multidrug classes. Recently, viral entry inhibitors have been
* To whom correspondence should be addressed. Phone: +39 0691093680.
Fax: +39 06 91093654. E-mail: vincenzo_summa@merck.com.
†
Istituto Di Ricerche Di Biologia Molecolare, P. Angeletti SpA (Merck
Research Laboratories, Rome).
‡
Department of Antiviral Research, Merck Research Laboratories.
a Abbreviations: FDA, Food and Drug Administration; HAART, highly
active antiretroviral therapy; NRTI, nucleoside reverse transcriptase inhibitor;
NNRTI, Non-nucleoside reverse transcriptase inhibitor; PI, protease inhibi-
tor; DKA, diketoacid; FBS, fetal bovine serum; NHS, normal human serum.
10.1021/jm800245z CCC: $40.75
2008 American Chemical Society
Published on Web 09/03/2008

5844 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Summa et al.
Figure 1. Optimization of dihydroxypyrimidine and N-methyl pyrimidinone carboxamides as HIV-integrase strand transfer inhibitors.
tors.11 Within Merck Research Laboratories (MRL), the two
groups working on HIV-integrase and HCV polymerase dis-
covered a series of scaffolds characterized by improved drug-
like qualities that dialed out the undesirable DKA properties.
In particular, we discovered that the dihydroxypyrimidine
carboxamide derived from the evolution of DKA in the HCV
polymerase program12 was a potent, reversible, and selective
HIV-integrase strand transfer inhibitor showing nanomolar
activity in the enzymatic assay (IC50 ) 0.085 µM) while being
completely inactive on the HCV polymerase.13 Extensive
structure activity relationship studies on the carboxamide moiety
led to the identification of the p-fluorobenzyl as the optimal
amide residue14 (1, Figure 1) and of the gem-dimethyl as the
optimal 2-substituent for the dihydroxy pyrimidine core15 (2).
Parallel efforts led to the identification of the related N-
methylpyrimidone scaffold showing equal or enhanced activity
for the HIV-integrase (3). Both 2 and 3 showed good potency
in the antiviral cell based assay in the presence of 50% normal
human serum (NHS) and favorable oral bioavailability in
preclinical species.16
Raltegravir (27) is the result of our continued efforts to
optimize these inhibitors, addressing issues presented by previ-
ously reported inhibitors. These include metabolic stability,
pharmacokinetic profile, antiviral activity against a panel of
HIV-1 integrase mutations previously shown to be associated
with resistance to 1,3-diketoacid inhibitors and other classes of
inhibitors, and genotoxicity. The antiviral properties and pre-
clinical pharmacological profile of Raltegravir are described
here: these qualities have enabled its progression toward the
end of phase III clinical trials for the treatment of HIV-1
infection and AIDS, positioning it as the first HIV-integrase
inhibitor approved by FDA for the treatment of HIV-1 infec-
tion.17
in the enzymatic assay (5 IC50 ) 0.003 µM; 6 IC50 ) 0.002
µM) and good potency in the antiviral cell based assays in low
serum condition (5 CIC95 ) 0.065 µM; 6 CIC95 ) 0.250 µM
10% FBS). However, a large potency shift was observed in the
presence of 50% NHS for both analogues, which led us to
explore other modifications. The simple acetamide derivative
7 showed a good potency in all assays and limited shift in high
serum conditions. Various acetamide replacements such as
sulfonamide 8, urea 9, and sulfamide 10 were also prepared.
While for all these analogues the potency in the enzyme assay
was in the nanomolar range (IC50 ) 0.007-0.018 µM), in the
antiviral cell based assay, the compounds showed a larger
variability in the activity values, ranging from 0.062 to >1.0
µM in low serum conditions. Only 7 and 10 retained activity
below 1 µM in high serum conditions (CIC95 ) 0.4 and 0.5
µM 50% NHS for 7 and 10, respectively).
The amide functionality offered the opportunity to explore
very rapidly a large series of fragments by rapid analogue
synthesis. We discovered that the introduction of an oxalamide
moiety on the scaffold resulted in a very potent compound both
in the enzymatic and antiviral cell based assays (11, Table 2):
18
compound 11 showed an IC50 ) 0.010 µM on the enzyme
and a marginal shift in the cell based assay, having CIC95
)
0.045 and 0.074 µM in 10% FBS and 50% NHS, respectively.
The corresponding ester 12 and acid 13 showed the same level
of potency on the enzyme but were significantly less potent in
the cell based assays. Methylation of the oxalamide NH was
detrimental: compound 14 showed both decreased enzymatic
potency and activity in the cell based assay compared to 11
(IC50 ) 0.015 µM; CIC95 ) 0.062, 0.125 µM 10% FBS and
50% NHS, respectively). A number of analogues were prepared
to explore replacement of the N,N-dimethyl fragment of the
oxalamide moiety. Among those, two of the most relevant
examples are the morpholine derivative 15 and the N-methyl
piperazine derivative 16, both of which showed good levels of
potency. However, the N,N-dimethyloxalamide substituent
remained the most potent substituent within this series.
The benzylamide moiety of 11 was also revisited with a
selection of amides known from previous work on the project
to modulate the physicochemical properties of our inhibitors.13
In particular, it was observed that the protein binding was
influenced by the substituent present on the aromatic ring. The
N-Me carboxamide (17, Table 3) had low activity on the
enzyme, reinforcing the importance of the benzyl amide as key
fragment for enzymatic activity. Introduction of more polar
substituents with respect to 4-F on the phenyl ring of the
benzylamides were generally tolerated in terms of enzymatic
potency, while the 2,3-dimethoxy derivative 18 was found to
be a weak inhibitor in the cellular based assay, the 2-methyl-
sulfone-4-fluoro 19 showed high activity with no shift between
low and high serum conditions. The more lipophilic benzylamide
Results and Discussion
As part of our continued effort to optimize the two classes
of inhibitors represented by 2 (IC50 0.05 µM; CIC95 0.06 µM
10% FBS; CIC95 0.078 µM 50% NHS) and 3 (IC50 0.06 µM;
CIC95 0.06 µM 10% FBS; CIC95 0.10 µM 50% NHS), we
combined the optimal properties of each series in a single
molecule in order to capitalize on the different physicochemical
properties of the scaffolds. The straightforward modification was
to convert the dihydroxypyrimidine derivative 215 into the
corresponding N-methyl pyrimidinone 4 (Figure 1). The modi-
fication did not lead to the desired result, as the compound was
less potent both in the enzymatic and cell based assays (Table
1). Initial structure activity relationship studies were conducted
in order to improve the activity against integrase, focusing in
particular on the basic amine and the gem-dimethyl moieties of
inhibitor 4. This effort led to compounds such as morpholine 5
and tetrahydropyran 6, both showing low nanomolar activity

HIV-1 Integrase
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5845
Table 1. 2-Substituted-N-methyl Pyrimidone Derivatives
a Assays were performed with recombinant HIV-1 integrase (0.01 µM) preassembled on immobilized oligonucleotides. Inhibitors were added after assembly
and washings, and IC50 is the concentration of inhibitor that reduces HIV-integrase activity by 50%. Results are the mean of at least 3 independent experiments;
SD was always ( 8% of the value. b Spread assay: 95% cell culture inhibitory concentrations for inhibition of the spread of HIV-1 infection in cell culture,
using HIV-1IIIb and MT-4 T-lymphoid cells, in a medium containing 10% heat-inactivated fetal bovine serum or 50% normal human serum. Results are the
mean of at least 3 independent experiments, SD was always < (10% of the value. For details see ref 10.
analogue 20 proved to be a potent inhibitor in the cellular assay
in low serum conditions, although a 4-fold shift in activity was
observed in 50% NHS (IC50 ) 0.009 µM; CIC95 ) 0.031 µM
10% FBS; CIC95 ) 0.130 µM 50% NHS). Overall, these results
highlighted that the 4-fluorobenzylamide remains the optimal
substituent even within an optimized scaffold such as the one
of analogue 11.
membered ring heterocycles were also investigated in depth,
and compounds bearing two or three heteroatoms were the most
potent in the cell based assays, with heterocycles having a
heteroatom in the 2-position being the most active. Oxazole 24,
thiazole 25, and imidazole 26 all showed single-digit nanomolar
activity on the enzyme. In contrast, in the cellular antiviral assay,
they showed different degrees of activity. Oxazole 24 had CIC95
) 0.50 µM in both serum conditions, while thiazole 25 had
higher activity in low serum and showed 4-fold shift in high
serum and imidazole 26 showed the best activity in high serum
conditions (CIC95 ) 0.250 µM 50% NHS). Among the
heterocycles having three heteroatoms, oxadiazole 27 was the
best, being one of the most potent compounds in the cell based
assay (CIC95 ) 0.019 and 0.031 µM in 10% FBS and 50% NHS,
respectively). Other heterocycles having a similar heteroatom
arrangement, including the corresponding triazole 28, showed
a lower activity in the antiviral assay.
Next, we moved to investigate aromatic and heteroaromatic
amides as capping groups of the amine portion of the scaffold
11 (Table 4). While the aromatic derivatives showed modest
potency in the antiviral cell based assay (data not shown), the
heteroaromatic ones proved to be very active. The original
fragment screening was conducted with the idea to have
heteroatoms in the same positions where the heteroatoms of
the oxalamide are located. The simple 2-pyridine carboxamide
21 showed low nanomolar activity on the enzyme, a 6-fold shift
in the spread assay in low serum, but reduced activity in high
serum (IC50 ) 0.02 µM; CIC95 ) 0.125 µM 10% FBS; CIC95
) 1.00 µM 50% NHS). The other pyridine isomers were less
potent in the cell based assay (data not shown). The introduction
of a second nitrogen in the ring led to the more polar pyridazine
derivative 22, which showed a reduced shift in potency between
the low and high serum conditions (CIC95 ) 0.062 µM 10%
FBS; CIC95 ) 0.50 µM 50% NHS). The isomeric pyrimidine
derivative 23 was potent in all assays (IC50 ) 0.007 µM; CIC95
) 0.02 µM 10% FBS; CIC95 ) 0.050 µM 50% NHS). Five-
Selectivity, Drug Metabolism and Pharmacokinetics
The most potent compounds in the antiviral cellular assays
were further profiled, and between them 11 and 27 were
preferred over the others. 11 and 27 were tested and were found
to be essentially inactive (IC50’s >50 µM) against a series of
related enzymes such as HCV polymerase, HIV RT, HIV
RNase-H, and human R-, ꢀ-, and γ-polymerases. This shows
the high degree of selectivity for integrase over the other

5846 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Table 2. Oxalamide Derivatives
Summa et al.
a Assays were performed with recombinant HIV-1 integrase (0.01 µM) preassembled on immobilized oligonucleotides. Inhibitors were added after assembly
and washings, and IC50 is the concentration of inhibitor that reduces HIV-integrase activity by 50%. Results are the mean of at least 3 independent experiments;
SD was always ( 8% of the value. b Spread assay: 95% cell culture inhibitory concentrations for inhibition of the spread of HIV-1 infection in cell culture,
using HIV-1IIIb and MT-4 T-lymphoid cells, in a medium containing 10% heat-inactivated fetal bovine serum or 50% normal human serum. Results are the
mean of at least 3 independent experiments, SD was always < (10% of the value. For details see ref 10.
Table 3. Benzylamide Moiety Derivatives
a Assays were performed with recombinant HIV-1 integrase (0.01 µM) preassembled on immobilized oligonucleotides. Inhibitors were added after assembly
and washings, and IC50 is the concentration of inhibitor that reduces HIV-integrase activity by 50%. Results are the mean of at least 3 independent experiments;
SD was always ( 8% of the value. b Spread assay: 95% cell culture inhibitory concentrations for inhibition of the spread of HIV-1 infection in cell culture,
using HIV-1IIIb and MT-4 T-lymphoid cells, in a medium containing 10% heat-inactivated fetal bovine serum or 50% normal human serum. Results are the
mean of at least 3 independent experiments, SD was always < (10% of the value. For details see ref 10.
enzymes that work through a Mg2+ catalyzed reaction. To assess
the selectivity toward other unrelated targets with different
mechanisms of action, the two agents were also tested in the
Pan Laboratories screen having 150 enzymes, channels, and
receptors, and they showed no significant activity <10 µM.
Neither of the two compounds inhibited any of the major
Cytochrome P450 isoforms, with IC50’s > 50 µM for 1A2, 2C9,
2D6, 3A4, and 2C9, nor showed time dependent inhibition of
the 3A4 isoform. Both compounds had binding affinity >50
µM on hERG channels.

HIV-1 Integrase
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5847
Table 4. Heteroaromatic Capping Group
a Assays were performed with recombinant HIV-1 integrase (0.01 µM) preassembled on immobilized oligonucleotides. Inhibitors were added after assembly
and washings, and IC50 is the concentration of inhibitor that reduces HIV-integrase activity by 50%. Results are the mean of at least 3 independent experiments;
SD was always ( 8% of the value. b Spread assay: 95% cell culture inhibitory concentrations for inhibition of the spread of HIV-1 infection in cell culture,
using HIV-1IIIb and MT-4 T-lymphoid cells, in a medium containing 10% heat-inactivated fetal bovine serum or 50% normal human serum. Results are the
mean of at least 3 independent experiments, SD was always < (10% of the value. For details see ref 10.
Table 5. iv Pharmacokinetic Parameters for Rats, Dogs, and Rhesus
Monkeys for 11 and 27
to the previous formulation (1% methyl cellulose AUC ) 23
µM·h, Cmax ) 12 µM; PEG400 AUC ) 49 µM·h, Cmax ) 31
c
d
e
f
Clp
compd species (mL/min/kg)
RT1/2
(h)
ꢀT1/2
(h)
Vd
dose
µM). The same effect was observed dosing the compound in
dogs at 10 mg/kg in 1% methyl cellulose, where the sodium
salt gave higher exposure, Cmax, and prolonged T1/2 compared
to the free OH, (Na+ salt AUC ) 47 µM·h, Cmax ) 32 µM,
T1/2 ) 11.9 h; free OH AUC ) 23 µM·h, Cmax ) 11.1 µM,
T1/2 ) 4.9 h). No advantages were observed dosing the free
OH in PEG400, as the exposure was similar to the Na+ salt in
methylcellulose (AUC PEG400-MC Na+ salt ) 44 vs 47
µM·h). 11 was also dosed po at 10 mg/kg in rhesus monkeys
as free OH in 1% methyl cellulose: the exposure and plasma
half-life were modest in comparison to the other preclinical
species (AUC ) 5.2 µM · h, Cmax ) 4.5 µM, T1/2 ) 1.9 h),
probably due to the low solubility in the conditions of the study.
These findings led to the decision to use the Na+ salt as the
optimal form for safety studies.
Values of plasma clearance (Clp), volume of distribution
(Vdss), and plasma half-life (T1/2) in rats, dogs, and rhesus
monkeys were calculated dosing 11 iv; data are reported in
Table 5. Considering the liver blood flow, clearance was
moderate for all species with a rank order of rhesus, rat, and
dog. The volume of distribution was modest for all species.
Plasma half-life showed a multiphasic elimination, with a
relatively short R-phase and a prolonged ꢀ-phase for all species.
Good oral bioavailability was observed in dog and rat dosing
11 as Na+ salt (F ) 93 and 36% for dog and rat, respectively),
(L/kg) (mg/kg)
11
11
11
27
27
27
rata
20
11
14
39
6
0.4
1.2
0.9
ND
ND
ND
6
16
3
2
11
4
1.3
1.1
2.9
2.0
0.9
1.2
3
5
1
3
1
1
doga
rhesusb
rata
doga
rhesusa
18
a Vehicle iv DMSO/saline 20%/80%. b Vehicle iv DMSO. c Plasma
clearance. d Plasma primary phase half-life following iv administration.
e Terminal plasma half-life following iv administration. f Volume of
Distribution. ND Not done.
The pharmacokinetic profiles of 11 and 27 were determined
in Sprague-Dawley rats, dogs, and rhesus monkeys. In rats,
various formulations and salts were studied. Dosing 11 in rats
at 3 mg/kg po using methyl cellulose as vehicle, the exposure
and Cmax were increased and T1/2 was prolonged using the
sodium salt with respect to the free acid (Na+ salt AUC ) 2.2
µM·h, Cmax ) 4.1 µM, T1/2 ) 2.3 h; free OH AUC ) 0.7 µM·h,
Cmax ) 0.6 µM, T1/2 ) 0.9 h; Table 6). Dose proportionality
studies were performed using methyl cellulose as a vehicle for
11 as free OH dosed po at 1, 4, 15, 50 as solutions and 100
mg/kg as a suspension, where the AUC and Cmax increased
linearly (R2 0.95 and 0.94, respectively). Using PEG400 as
vehicle at 100 mg/kg, the compound was completely soluble,
and the exposure and the Cmax increased 2.5-fold with respect

5848 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Summa et al.
Table 6. Rat, Dog, and Rhesus Monkey po Pharmacokinetic Parameters for 11 and 27
AUCd
PPBe
c
compd Fa(%) rat/dog/rhesus Cmaxb(µM) rat/dog/rhesus po t1/2 (h) rat/dog/rhesus (µM·h)
rat/dog/rhesus (% free) rat/dog/rhesus/human
(mg/kg)
11 Na+
11 OH
27 Na+
27 K+
27 OH
36/93/nd
nd/nd/24
nd/nd/nd
45/69-85/nd
37/45/8
4.1(3)/32(10)/nd
0.6(3)/11(10)/4.5(10)
1.0/nd/nd
1.6/4.6(2)-24(10)/nd
1.2(3)/3(2)-8(10)/0.3
2.3/11.9/nd
0.9/4.9/1.9
nd/nd/nd
7.5/13/nd
nd/nd/7
2.2(3)/47(10)/nd
0.7(3)/23(10)/5.2(10)
1.4(3)/nd/nd
1.3(3)/11(2)-45(10)/nd
1.0(3)/7(2)-21(10)/1.8(10)
35/38/18/28
26/29/15/17
a Oral bioavailability. b Maximum plasma concentration after po administration. c Terminal phase plasma half-life following po administration. d Area
under the curve following po administration. e Plasma protein binding (free fraction). nd ) not done.
while lower oral exposure was achieved in rat and rhesus dosing
11 as free OH, probably due to lower solubility and related
incomplete absorption (F ) 34 and 24% rat and rhesus,
respectively). Plasma protein binding was measured for human
and the three preclinical species: the free fraction was high for
all species with a rank order rat, dog, human, and rhesus (35,
38, 28, and 18%, respectively, all data reported in Table 6).
The pharmacokinetic profile of 27 was also studied using
different crystalline salts with 1% methyl cellulose as the vehicle
of choice to conduct po dosing. This choice was based on the
previous experience with 11, where the PEG400 was equivalent
or worse than the Na+ salt in 1% methyl cellulose. Furthermore,
1% methyl cellulose is one of the most frequently used vehicles
to conduct safety studies in preclinical species. 27 was dosed
in rat at 3 mg/kg using three different forms (free OH, Na+,
and K+ salts), and higher exposures were achieved with the
salts (AUC Na+ ) 1.4, K+ ) 1.3, free OH 1.0 µM·h,
respectively). Dose proportionality studies using 1, 4, 15, 60,
100 mg/kg were performed both on the crystalline free OH and
the crystalline K+ salt. The first was less than dose proportional
although the AUC increased (data not reported). Much better
results were obtained with the crystalline K+ salt, with AUC
increasing linearly and in a dose proportional manner (R2 )
0.95, AUC from 0.8 to 35 µM·h). In contrast, the Cmax was
less than dose proportional. In dog, 27 was dosed at 2 and 10
mg/kg as free OH and K+ salt, both crystalline. The AUCs and
Cmax were higher for the K+ salt at both doses (AUC 2 mg/kg
) 7 vs 11; 10 mg/ kg ) 21 vs 45 µM·h; Cmax 2 mg/kg 3 vs
4.5, 10 mg/kg 8 vs 24 free OH vs K+, respectively). The
terminal plasma half-life (T1/2) after 10 mg/kg po of the K+
salt was 13 h. The crystalline free OH was dosed at 10 mg/kg
in rhesus monkey showing a modest exposure and Cmax (AUC
) 1.8 µM·h; Cmax ) 0.3 µM). 27 was dosed iv using as vehicle
20% DMSO/40% Peg400/40% H2O at 3 mg/kg in rat and 1
mg/kg in dog and rhesus. The corresponding plasma clearances
(Clp) were 39, 6, and 18 mL/min/kg and volume of distribution
(Vdss) 2.0, 0.9, and 1.2 L/kg for rat, dog, and rhesus, respectively.
The oral bioavailabilities, calculated for the po experiments,
were 45% in rat, 69-85% in dog dosing the crystalline K+ salt,
and 37, 45, and 8% in rat, dog, and rhesus, respectively, with
the crystalline free OH (Table 6).
microsomes in the presence of NADPH, showing that oxidative
metabolism does not play a significant role. The major route of
metabolism proved to be glucuronidation, with the rate of
metabolism in the presence of UDPGA showing a clear rank
order between species, (rhesus > rat > human > dog) with
values equal to 46, 20, 8, and 3 µL/min/kg for 11 and 36, 34,
9, and 2 for 27. The same metabolism profile was observed in
the presence of hepatocytes. The compounds did not react with
glutathione after incubation at 37 °C for 24 h. [3H]-11 and [3H]-
27 were used to measure the irreversible protein covalent binding
in vivo (rat 25 mpk po) to liver, kidney, and plasma and in
vitro to human and rat liver microsomes. In both assays, the
covalent binding was <50 pmol-equivalent/mg-protein for both
compounds. The in vivo metabolism was also investigated in
rat using the radiolabeled material and in rat and dog by NMR.
In rats, 11 was eliminated mainly by metabolism. Urinary and
biliary secretion of intact drug amounted to 8% and 5% of the
administered dose, respectively. In dogs, 13% of the dose was
excreted unchanged in urine after po dosing. The major
metabolite of 11 detected in rat urine and bile and dog urine
was the 5-O-glucuronide conjugate. In rat and dog excreta, no
other metabolites were detected by 19F-NMR: 90% of total
fluorine signal accounts for the 5-O-glucuronide. Dosing 27 iv
in rats a recovery of 50% of the dose in bile and 20% in urine
(glucuronide plus parent compound) was observed. The amount
of glucuronide in bile and urine after iv dosing was 90% of the
total fluorine signal and the intact compound was 10%.
Following oral dosing, around 20% of the dose was recovered
in rat urine (glucuronide plus parent compound). In dogs, after
iv dosing of 27, around 30% of the dose was recovered in urine
(0-24 h). The amount of intact compound in the urine was
65% of the total fluorine signal and the glucuronide was 35%.
Following oral dosing the amount of 27 excreted in urine was
10% of the dose (glucuronide plus parent compound). The same
profiles were found using the 3H-materials. These data were in
perfect agreement with the microsome stability data, showing
glucoridation as the major route of metabolism. The glucuronide
of 27 was fully characterized using 1H, 13C, and 1H-13C
correlation NMR spectroscopy, which confirmed the formation
of the glucuronide conjugate of the hydroxyl group in position
1
5 of the pyrimidinone ring. The full assignment of all H and
The PK data conducted with different vehicles and salt forms
indicated that the Na+ or K+ salt showed better po profile for
the exposure and solubility properties, and these crystalline salt
forms were therefore selected for further studies in safety
assessment.
The plasma protein binding was measured for preclinical
species and human, showing modest interspecies variability of
the free fraction values (rat 26.6%, dog 29.1%, rhesus 15.4%,
and human 17.2% free, Table 6).
The metabolism of the two agents was studied in liver
microsomes of rat, dog, rhesus, and human. The compounds
had the same route of metabolism previously reported for this
class of molecules in that they were very stable in liver
13C signals ascribed to these ꢀ-glucuronides was achieved using
the crude sample of urine after solid phase extraction.19
Comparing the in vivo and in vitro data, the human phar-
macokinetic profile was predicted to be similar to the dog.
Looking at the po profile, it was clear that both 11 and 27
showed biphasic elimination with a short R-phase and a
prolonged ꢀ-phase. This behavior was a very important feature
to maintain the plasma concentration well above the CIC95 at
12 h (11 Na+ salt CIC95 ) 125 nM 50% NHS; dog C12 h 246
nM at 10 mg/kg; 27 K+ salt CIC95 ) 31nM 50% NHS; dog
C12 h 160 and 350 nM at 2 and 10 mg/kg, respectively). Taking
into account the plasma protein binding, the metabolic stability

HIV-1 Integrase
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5849
Table 7. Resistance Profiles with HIV-1 Containing Site-Directed Integrase Mutationsa
mutations
T66I
V151I
F121Y
T125K
T66I*b M154I
T66I*b S153Y
N155S*b
T125K*b F121Y
T66I*/L74Mb V151I
L-870810
11
27
1
3
1
1
1
1
4
3
12
3
0.5
1
1
3
5
1
1.3
8
1
10
40
10
12
19
8
9
35
6
S-1360
12
14
2
20
50
>50
50
100
a Shift in IC50 relative to wild type HIV-1-single cycle infectivity assay. b * ) Viruses that exhibit >50% reduction in specific infectivity.
(intrinsic clearance), and the plasma clearance for all species,
the human dosing regimen was predicted to be BID.
approved by the FDA as the first HIV-integrase inhibitor for
the treatment of HIV AIDS.
The compounds were also evaluated in the Ames test and
resulted negative both in the presence and in the absence of S9
fraction. This finding was in contrast to what had previously
been observed for 3, which proved positive in the same assay
showing a reversion on the strain TA 1535 and was therefore
not further developed.
Biology
Compounds were assessed for activity against the purified
HIV-1 integrase enzyme.6 Integrase mediated strand transfer
activity was determined as described in footnote Table 1.
Compounds were also tested in HIV-1 replication assays,
performed in MT-4 human T-lymphoid cells as described in
ref 10. Cells were infected en masse at low multiplicity (0.01).
HIV-1 replication was assessed by p24 core antigen ELISA.
Control cultures in the absence of inhibitor were fully infected
at 4 days.
11 and 27 shared very similar DMPK profiles, with 27 clearly
being the more potent of the two in the antiviral assay in
presence of high serum. We decided to include another
parameter to discriminate between the two compounds. Con-
sidering how important it is to profile HIV drugs against mutants
raised in the clinic and the correlation that they have with the
mutants generated in cell culture, we decided to profile the two
agents against a series of HIV-integrase mutants, which had
been raised during the course of many years of research in our
laboratories using different classes of inhibitors both from our
research and coming from the work of other institutions or
pharmaceutical companies (Table 7). The assay employed is a
single cycle infectivity assay in presence of 10% FBS. The
numbers are reported as fold shift in IC50 relative to wild type
HIV-1. Viruses which exhibit >50% reduction in specific
infectivity (impaired replication capacity) are starred.
Chemistry
The general synthetic route to the core N-methylpyrimidone-
4-carboxamide scaffold is depicted in Scheme 1. Conversion
of nitriles 29 to amidoximes 30 and reaction with dimethyl
acetylenedicarboxylate provided, via formation of the intermedi-
ate adducts 31, the dihydroxypyrimidine-2-carboxylates 32.21
Benzoylation of the 5-hydroxyl and subsequent methylation
afforded the 2-substituted methyl 5-(benzoyloxy)-1-methyl-6-
oxo-1,6-dihydropyrimidine-4-carboxylates 33 together with
minor amounts of the O-methylated derivatives 34, which could
be easily separated via chromatography. The pyrimidones 33
were reacted with p-fluorobenzylamine and then hydrogenated
to give the benzylic amides 35, which served as a useful
diversification point for the construction of the final compounds.
Reductive amination of 35 with formaldehyde gave 4 and 6,
while sulfonamide 8, urea 9, sulfonyl urea 10, and amides 12,
21-26, and 28 were obtained using standard protocols.
While the general synthetic route is outlined in Scheme 1, in
some cases, the functionality at the amino substituent in the
2-position of the pyrimidone ring of 33a was better introduced
prior to the installation of the benzylic amide in the 4-position
(Scheme 2). In particular, the synthesis of 27 by the reaction
sequence in Scheme 1 gave erratic results when it was scaled
up to more than one gram of material due to the acid liability
of the oxadiazole ring during RP-HPLC purification. The
alternative approach described in Scheme 2 allowed the removal
of side products either via silica gel column chromatography
or crystallization (seeExperimental Section), and the synthesis
could be scaled up in our laboratories to 10 g of 27 as crystalline
free OH at a time.
L-870810 is another Merck HIV strand transfer inhibitor.20
S-1360 is a diketo heteroaryl derivative from Shionogi that was
in clinical development. Comparing the four agents, 27 showed
a similar profile to L-870810 and was better than 11, with all
Merck compounds showing a better profile than that of S-1360.
Taking into account the intrinsic potency in the cell based
assay in high serum condition, the PK profile, and the mutant
profile, compound 27 was selected over 11 for further develop-
ment. 27 is also called Raltegravir or MK-0518, and in October
2007 was approved by the FDA as the first HIV-integrase
inhibitor for the treatment of HIV AIDS infected patients.
Conclusion
The discovery of Raltegravir (27) came from a series of potent
and structurally new HIV-integrase inhibitors designed as
valuable drug-like DKA replacements for the HCV polymerase.
The knowledge of the catalytic machinery of HIV-integrase and
HCV polymerase led us to test the new class against HIV-
integrase. We identified a compound that was active in the
enzymatic assay, and by the optimization of the antiviral activity
in cell culture in the presence of 50% NHS, physicochemical
properties and pharmacokinetic profiles identified a number of
suitable candidates. Counterscreening and activity against a
series of mutants selected in vitro was the final screening funnel
and led to the identification of Raltegravir or MK-0518.
Raltegravir met all preclinical criteria to be considered a very
effective and safe HIV agent; the prediction of human PK
provided the hypothesis that it would be a BID drug, which
was later confirmed in clinical studies. Raltegravir has been
Compound 12 was further diversified to 11, 15, and 16 via
reaction with the appropriate amine. In some cases, the
carboxylic acid 13 was isolated from the reaction mixture as a
byproduct (Scheme 4). To obtain compound 14, the same
chemistry as for the preparation of 11 was applied on intermedi-
ate 37, obtained from 35a by a stepwise reductive amination
with benzaldehyde and formaldehyde, followed by hydrogenation.
To explore the benzyl amide moiety of 11, the reaction
sequence in Scheme 5 was followed, and compounds 18-20

5850 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Summa et al.
Scheme 1. General Scheme for the Synthesis of N-Methylpyrimidone-4-carboxamidesa
a Reagents and conditions: (a) NH2OH.HCl, KOH, MeOH; (b) DMAD, CHCl3, 60 °C; (c) Xylene, 150 °C; (d) i: Bz2O, pyridine, rt; ii: Me2SO4, LiH,
dioxane, 60 °C; (e) i: 4-F-C6H4CH2NH2, MeOH or NMP, refl.; ii: H2, Pd/C, MeOH; (f) HCHO, NaCNBH3, AcOH, Et3N (or NaOAc), MeOH; (g) see
Experimental Section.
Scheme 2. Alternative Synthesis of
N-Methylpyrimidone-4-carboxamidea
Scheme 3. Synthesis of Compound 5a
a Reagents and conditions: (a) Morpholine (10 equiv), NMP, 120 °C.
2795 apparatus, equipped with a diode array and a ZQ mass
spectrometer, using an X-Terra C18 column (5 µm, 4.6 mm ×50
mm). Solvent system was acetonitrile-0.1% HCOOH/water-0.1%
HCOOH, gradient 10-90% over 6 min with a flow rate of 1 mL/
min. Purity of final compounds was more than 99% by area.
Preparative reversed-phase high-performance liquid chromatography
(RP-HPLC) was performed on a Shimadzu 10AV-VP operating
with a flow rate of 15 mL/min and incorporating a diode array
detector SPD10AV-VP. The stationary phases used were Waters
Symmetry C18 5 µm 19 mm × 100 mm, C18 7 µm 19 mm × 150
mm, and C187 µm 19 mm × 300 mm. The mobile phase comprised
a linear gradient of binary mixtures of acetonitrile (containing 0.1%
TFA) and H2O (containing 0.1% TFA). Nuclear magnetic resonance
spectra 1H NMR recorded at 400 or 300 MHz, 13C NMR recorded
at 100 or 75 MHz were obtained on Bruker AMX spectrometers
and are referenced in ppm relative to TMS. Unless indicated, spectra
were acquired at 300 K. Low resolution mass spectra (m/z) were
recorded on a Perkin-Elmer API 100 (electrospray ionization) mass
spectrometer. High resolution mass measurements were carried out
by electrospray ionization performed on a TSQ Quantum Ultra AM
operating in enhanced mass resolution mode. The accurate mass
measurements were carried out on a chromatographic time scale
by means of flow injection of 5 µL of each working solution (20
a Reagents and conditions: (a) H2, Pd/C, MeOH; (b) R1COCl, Et3N,
CH2Cl2; (c) ArCH2NH2, MeOH, refl. Compound 5 was prepared by
treating 4 with an excess of morpholine in N-methylpyrrolidone at 120
°C (Scheme 3).
were prepared in parallel (as part of a larger array), with
introduction of the benzylic amide as the ultimate step.
Experimental Section
Solvents and reagents were obtained from commercial suppliers
and were used without further purification. Flash chromatography
purifications were performed on Merck silica gel (200-400 mesh)
as the stationary phase or were conducted using prepacked cartridges
on a Biotage system, eluting with petroleum ether/ethyl acetate
mixtures. HPLC-MS analyses were performed on a Waters Alliance

HIV-1 Integrase
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5851
Scheme 4. Synthesis of Compounds 11-16a
a Reagents and conditions: (a) R1R2NH, MeOH or THF; (b) (i) PhCHO, HC(OMe)3 then NaCNBH3, MeOH; (ii) HCHO, NaCNBH3; (iii) H2, Pd/C,
MeOH; (c) (i) MeOCOCOCl, Et3N, CHCl3; (ii) Me2NH, THF.
Scheme 5. Synthesis of Compounds 18-20a
a Reagents and conditions: (a) (i) MeOCOCOCl, Et3N, CHCl3; (ii) Me2NH, THF; (b) ArCH2NH2, NMP, 100 °C.
µg/mL concentration) in acetonitrile on a Waters C18 Sunfire (20
mm × 2.1 mm, 5 µm) column using an isocratic elution (solvent
A: solvent B, 50:50) of solvent A: water + 0.1% formic acid and
solvent B: acetonitrile + 0.1% formic acid at a flow rate of 1000
µL/min.22
Benzyl 1-Cyano-1-methylethylcarbamate (29a). To a suspension
of 2-amino-2-methylpropanenitrile23 (0.5 mol) in water (200 mL)
an equimolar amount of Na2CO3 (52 g, 0.5 mol) and a slight excess
(1.1 equiv) of benzylchloroformate were added with external
cooling. The reaction mixture was stirred overnight at room
temperature, extracted in EtOAc, and the organic phase was washed
with NaHCO3ss, dried (Na2SO4), filtered, and concentrated. Title
product was obtained as a white solid (92 g, 0.43 mol, 85% yield).
1H NMR (CDCl3) δ 7.48-7.33 (bs, 5 H), 5.15 (s, 2 H), 4.98 (bs,
1 H), 1.68 (s, 6 H). 13C NMR (CDCl3) δ 153.33, 13.81, 127.81,
127.63, 127.55, 120.64, 66.56, 46.19, 26.67. MS m/z 219 (M +
H)+.
dimethylacetylenedicarboxylate and the reaction was refluxed
overnight. After cooling to room temperature, volatiles were
evaporated and the residue containing crude 31a was taken into
xylene and heated at 145 °C for 48 h. The reaction mixture was
stirred at room temperature overnight to allow the precipitation of
the product (32a) as a light-brown solid. This solid was collected
by filtration and washed with diethyl ether (56 g, 0.16 mol, 41%
yield). 1H NMR (DMSO-d6) δ 12.54 (s, 1 H), 10.21 (s, 1 H), 7.44
(bs, 1 H), 7.30 (bs, 5 H), 4.95 (s, 2 H), 3.80 (s, 3 H), 1.47 (s, 6 H).
MS m/z 362 (M + H)+.
Methyl 2-(4-{[(Benzyloxy)carbonyl]amino}tetrahydro-2H-py-
ran-4-yl)-5,6-dihydroxypyrimidine-4-carboxylate (32b). was pre-
pared from benzyl (4-cyanotetrahydro-2H-pyran-4-yl)carbamate 29b
as described for the synthesis of 32a. 1H NMR (DMSO-d6) δ 12.58
(s, 1H), 10.30 (s, 1H), 7.45 (bs, 1H), 7.40-7.31 (m, 5H), 4.98 (s,
2H), 3.82 (s, 3H), 3.66-3.62 (m, 4H), 2.20-2.09 (m, 2H),
1.97-1.93 (m, 2H). MS m/z 404 (M + H)+.
Benzyl 2-Amino-2-(hydroxyimino)-1,1-dimethylethylcarbamate
(30a). Hydroxylamine hydrochloride (30 g, 0.43 mol) in methanol
(100 mL) was added to an equimolar stirred solution of potassium
hydroxide in methanol (100 mL). The mixture was stirred for 15
min and the precipitated potassium chloride was removed by
filtration. The filtrate was added to an equimolar amount of the
above nitrile 29a, and the solution was stirred overnight at 40 °C,
then cooled to room temperature and concentrated. The resulting
residue was triturated with water giving, after drying under vacuum,
a white solid consisting mainly of the title product 30a (95 g, 0.38
mol, 88% yield). 1H NMR (DMSO-d6) δ 9.12 (bs, 1 H), 7.48 (bs,
5 H), 7.08 (bs, 1 H), 5.33 (bs, 2 H), 4.98 (s, 2 H), 1.39 (s, 6 H).
MS m/z 252 (M + H)+.
Methyl 5-(Benzoyloxy)-2-(1-{[(benzyloxy)carbonyl]amino}-1-me-
thylethyl)-1-methyl-6-oxo-1,6-dihydropyrimidine-4-carboxylate (33a).
Step 1: to a stirred solution of methyl 2-(1-{[(benzyloxy)carbony-
l]amino}-1-methylethyl)-5,6-dihydroxypyrimidine-4-carboxylate
32a (6.0 g, 16.6 mmol) in pyridine (20 mL), 1.1 equiv of benzoic
anhydride was added and the resulting mixture was stirred at room
temperature overnight. Pyridine was evaporated, and the residue
was taken into ethyl acetate and washed with 1 N HCl and brine.
The organic layer was separated, dried (Na2SO4), filtered, and
concentrated by rotary evaporation. The residue was purified by
flash column chromatography (SiO2, petroleum ether/ethyl acetate
60/40 v/v as eluant) to give methyl 5-(benzoyloxy)-2-(1-{[(ben-
zyloxy)carbonyl]amino}-1-methylethyl)-6-hydroxypyrimidine-4-
carboxylate (85% yield). 1H NMR (CDCl3) δ 12.2 (bs, 1 H), 8.15
(d, J ) 7.4 Hz, 2 H), 7.65 (t, J ) 7.4 Hz, 1 H), 7.50 (t, J ) 7.5 Hz,
2 H), 7.32 (bs, 5 H), 5.54 (bs, 1 H), 5.05 (s, 2 H), 3.82 (s, 3 H),
1.67 (s, 6 H). MS m/z 466 (M + H)+. Step 2: To a stirred solution
Methyl 2-(1-{[(Benzyloxy)carbonyl]amino}-1-methylethyl)-5,6-
dihydroxypyrimidine-4-carboxylate (32a). Benzyl 2-amino-2-(hy-
droxyimino)-1,1-dimethylethylcarbamate (30a, 0.38 mol) was
suspended in chloroform (200 mL) and treated with 1.2 equiv of

5852 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Summa et al.
of LiH (1.1 equiv) in dioxane (15 mL), the previous intermediate
(6.5 g, 14 mmol) was added and the mixture was stirred at 38 °C
for 45 min. After cooling to room temperature, dimethyl sulfate
(1.3 equiv) was added and the reaction mixture was heated at 60
°C for 2 h. The mixture was then cooled to room temperature,
dioxane evaporated, and the residue was purified by flash chroma-
tography (SiO2), eluting with 65/55 v/v petroleum ether/ethyl acetate
luoroacetic acid as eluent). Collection and lyophilization of ap-
propriate fractions afforded compound 4 as its trifluoroacetate salt
(23% yield). 1H NMR (DMSO-d6) δ 12.58 (bs, 1 H), 9.71 (bt, J )
6.6 Hz, 1 H), 9.37 (bs, 1 H), 7.37 (dd, J ) 8.4, 5.7 Hz, 2 H), 7.19
(t, J ) 8.8 Hz, 2 H), 4.48 (d, J ) 6.6 Hz, 2 H), 3.56 (s, 3 H), 2.79
(s, 6 H), 1.72 (s, 6 H). MS m/z 363 (M + H)+. HRMS calcd for
C18H24O3N4F (M + H)+: 363.18270. Found: 363.18274.
1
to give 33a (4.5 g, 67% yield). H NMR (CDCl3) δ 8.19 (d, J )
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-(1-methyl-1-morpho-
lin-4-ylethyl)-6-oxo-1,6-dihydropyrimidine-4-carboxamide (5). A
solution of 4 (25 mg, 0.052 mmol) in N-methylpyrrolidone (1 mL)
was treated with 20 equiv of morpholine and the mixture was stirred
at 120 °C for 2 days, then cooled to room temperature and purified
by preparative RP-HPLC (C18, water/acetonitrile with 0.1% trif-
luoroacetic acid as eluent) to give 5 as its trifluoroacetate salt (15
mg, 56% yield). 1H NMR (DMSO-d6) δ 12.20 (bs, 1 H), 8.96 (bt,
J ) 6.9 Hz, 1 H), 7.49 (dd, J ) 8.5, 5.6 Hz, 2 H), 7.18 (t, J ) 8.8
Hz, 2 H), 4.50 (d, J ) 6.9 Hz, 2 H), 3.98 (s, 3 H), 3.57 (bs, 4 H),
2.44 (bs, 4 H), 1.44 (s, 6 H). MS m/z 405 (M + H)+. HRMS calcd
for C20H26O4N4F (M + H)+: 405.19326. Found: 405.19302.
2-[4-(Dimethylamino)tetrahydro-2H-pyran-4-yl]-N-(4-fluoroben-
zyl)-5-hydroxy-1-methyl-6-oxo-1,6-dihydropyrimidine-4-carboxa-
mide (6). A solution of 35b (0.05 g, 0.1 mmol) in MeOH (1 mL)
was treated with triethylamine (0.017 mL, 0.12 mmol), sodium
acetate (0.013 g, 0.16 mmol), formaldehyde 37% w/w aq soln (0.022
mL, 0.3 mmol), and sodium cyanoborohydride (0.009 g, 0.14
mmol). The mixture was stirred at room temperature for 3 h.
Volatiles were removed under vacuum and the crude was purified
by preparative RP-HPLC, giving the title compound 6 as trifluo-
roacetate salt. 1H NMR (CD3CN + TFA) δ 8.33 (bt, 1 H),
7.57-7.52 (m, 2 H), 7.29-7.23 (m, 2 H), 4.75 (d, J ) 6.6 Hz, 2
H), 4.18-4.09 (m, 2 H), 3.80 (s, 3 H), 3.63-3.58 (m, 2 H),
3.08-3.03 (m, 8 H), 2.40-2.30 (m, 2 H). MS m/z 405 (M + H+).
HRMS calcd for C20H26O4N4F (M + H)+: 405.19326. Found:
405.19290.
2-[1-(Acetylamino)-1-methylethyl]-N-(4-fluorobenzyl)-5-hydroxy-
1-methyl-6-oxo-1,6-dihydropyrimidine-4-carboxamide (7). To a
stirred solution of 36 (0.14 mmol) and triethylamine (6 equiv) in
dichloromethane (1 mL), 3 equiv of acetyl chloride were added.
The mixture was stirred at room temperature for 3 h then diluted
with chloroform and washed with 1 N HCl, brine, and NaHCO3ss.
The dried organic phase was concentrated, and the residue was taken
up in methanol (1 mL) and treated with 2.5 equiv of p-fluoroben-
zylamine. The mixture was refluxed for 4 h then purified by
preparative RP-HPLC to give the title compound in 14% yield. 1H
NMR (DMSO-d6) δ 12.11 (bs, 1 H), 8.99 (bt, J ) 5.7 Hz, 1 H),
8.43 (s, 1 H), 7.48 (dd, J ) 8.5, 5.7 Hz, 2 H), 7.17 (t, J ) 8.8 Hz,
2 H), 4.49 (d, J ) 5.7 Hz, 2 H), 3.48 (s, 3 H), 1.84 (s, 3 H), 1.59
(s, 6 H). MS m/z 377 (M + H)+. HRMS calcd for C18H22O4N4F
(M + H)+: 377.16196. Found: 377.16203.
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-{1-methyl-1-[(meth-
ylsulfonyl)amino]ethyl}-6-oxo-1,6-dihydropyrimidine-4-carboxam-
ide (8). Methyl sulfonyl chloride (0.028 mL, 0.36 mmol, 1.2 equiv)
was added dropwise to a solution of 35a (0.1 g, 0.30 mmol, 1.0
equiv) and Et3N (0.050 mL, 0.36 mmol, 1.2 equiv) in dichlo-
romethane (10 mL) at 0 °C, and stirring was continued at room
temperature. After 1 h, volatiles were evaporated in vacuum, the
residue was dissolved in dimethylamine (5 mL, 2.0 M solution in
THF), and stirring was continued overnight. The solvent was
reduced in vacuo, and the residue was dissolved in EtOAc, washed
with 1 N HCl and brine, and dried over Na2SO4. Evaporation of
the solvent gave a residue, which was purified by preparative RP-
HPLC, yielding 8 as a solid (0.043 g, 35% yield). 1H NMR (DMSO-
d6 + TFA) δ 12.14 (bs, 1 H), 8.98 (t, J ) 6.2 Hz, 1 H), 7.81 (s, 1
H), 7.38 (dd, J ) 8.4, 5.8 Hz, 2 H), 7.16 (t, J ) 8.8 Hz, 2 H), 4.49
(d, J ) 6.4 Hz, 2 H), 3.68 (s, 3 H), 3.00 (s, 3 H), 1.70 (s, 6 H). MS
m/z 413 (M + H)+. HRMS calcd for C17H22O5N4FS (M + H)+:
413.12895. Found: 413.12906.
7.3 Hz, 2 H), 7.65 (t, J ) 7.3 Hz, 1 H), 7.51 (t, J ) 7.6 Hz, 2 H),
7.33 (bs, 5 H), 5.63 (bs, 1 H), 5.03 (s, 2 H), 3.80 (s, 3 H), 3.63 (bs,
3 H), 1.72 (s, 6 H). MS m/z 480 (M + H)+.
Methyl 5-(Benzoyloxy)-2-(4-{[(benzyloxy)carbonyl]amino}tetra-
hydro-2H-pyran-4-yl)-1-methyl-6-oxo-1,6-dihydropyrimidine-4-
carboxylate (33b). was prepared in 54% yield from 32b following
the same procedure described for 33a. 1H NMR (DMSO-d6) δ 8.39
(bs, 1 H), 8.10 (d, J ) 7.6 Hz, 2 H), 7.79 (t, J ) 7.6 Hz, 1 H), 7.64
(t, J ) 7.6 Hz, 2 H), 7.42-7.28 (m, 5 H), 5.02 (s, 2 H), 3.78 (s, 3
H), 3.77-3.65 (m, 4 H), 3.54 (s, 3 H), 2.34-2.27 (m, 2 H),
2.09-2.04 (m, 2 H). MS m/z 522 (M + H)+.
2-(1-Amino-1-methylethyl)-N-(4-fluorobenzyl)-5-hydroxy-1-methyl-
6-oxo-1,6-dihydropyrimidine-4-carboxamide (35a). Step 1: to a
methanolic solution of 33a (4.5 g, 9.3 mmol), p-fluorobenzylamine
(3 equiv) was added and the resulting mixture was refluxed
overnight. After evaporation of methanol, the residue was taken
into EtOAc, washed with aq 1 N HCl and brine, dried (Na2SO4),
filtered, and evaporated to obtain benzyl 1-(4-{[(4-fluorobenzy-
l)amino]carbonyl}-5-hydroxy-1-methyl-6-oxo-1,6-dihydropyrimi-
1
din-2-yl)-1-methylethylcarbamate (98% yield). H NMR (CDCl3)
δ 11.9 (bs, 1 H), 7.79 (bt, 1 H), 7.35-7.29 (m, 7 H), 7.07 (t, J )
8.6 Hz, 2 H), 5.27 (bs, 1 H), 5.02 (bs, 2 H), 4.58 (d, J ) 6.2 Hz,
2 H), 3.67 (s, 3 H), 1.70 (s, 6 H). MS m/z 469 (M + H)+. Step 2:
a methanolic solution of compound from step 1 was stirred
overnight under a hydrogen atmosphere in the presence of catalytic
10% Pd/C. Catalyst was then filtered off through celite, and the
filtrate was concentrated. The title product was obtained after
1
trituration with ethyl ether (100% yield). H NMR (DMSO-d6) δ
12.31 (bs, 1 H), 9.68 (bt, J ) 6.6 Hz, 1 H), 8.60 (bs, 2 H), 7.43
(dd, J ) 8.4 Hz, J ) 5.7 Hz, 2 H), 7.20 (t, J ) 8.8 Hz, 2 H), 4.54
(d, J ) 6.6 Hz, 2 H), 3.56 (s, 3 H), 1.73 (s, 6 H). MS m/z 335 (M
+ H)+.
2-(4-Aminotetrahydro-2H-pyran-4-yl)-N-(4-fluorobenzyl)-5-hy-
droxy-1-methyl-6-oxo-1,6-dihydropyrimidine-4-carboxamide (35b).
The title compound was obtained as described for the preparation
of 35a, using in step 1 methyl 5-(benzoyloxy)-2-(4-{[(benzyl-
oxy)carbonyl]amino}tetrahydro-2H-pyran-4-yl)-1-methyl-6-oxo-1,6-
dihydropyrimidine-4-carboxylate (33b). 1H NMR (DMSO-d6) δ
12.38 (bs, 1 H), 9.82 (bs, 1 H), 8.98 (s, 2 H), 7.50-7.45 (m, 2 H),
7.25-7.19 (m, 2 H), 4.57 (d, J ) 6.2 Hz, 2 H), 3.95-3.88 (m, 4
H), 3.66 (s, 3 H), 2.72-2.58 (m, 2 H), 1.85-1.79 (m, 2 H). MS
m/z 377 (M + H)+.
2-[5-(Benzoyloxy)-4-(methoxycarbonyl)-1-methyl-6-oxo-1,6-di-
hydropyrimidin-2-yl]propan-2-amine Hydrochloride (36). A solu-
tion of methyl 5-(benzoyloxy)-2-(1-{[(benzyloxy)carbonyl]amino}-
1-methylethyl)-1-methyl-6-oxo-1,6-dihydropyrimidine-4-
carboxylate (33a) (3.86 g, 8.06 mmol) in MeOH (160 mL) and 1
equiv of 6 N HCl (1.34 mL) was stirred in the presence of 10%
Pd/C (400 mg) under an H2 atmosphere at atmospheric pressure
for 90 min. The mixture was then filtered and the Pd residues
washed with MeOH (150 mL). The combined organics were
concentrated under reduced pressure to yield the title compound
as hydrochloride salt (3.08 g, quantitative yield). 1H NMR (DMSO-
d6) δ 8.72 (br. s, 3 H), 8.09 (d, J ) 7.5 Hz, 2 H), 7.80 (t, J ) 7.5
Hz, 1 H), 7.64 (t, J ) 7.5 Hz, 2 H), 3.79 (s, 3 H), 3.63 (s, 3 H),
1.84 (s, 6 H). MS m/z 346 (M + H)+.
2-[1-(Dimethylamino)-1-methylethyl]-N-(4-fluorobenzyl)-5-hy-
droxy-1-methyl-6-oxo-1,6-dihydropyrimidine-4-carboxamide (4).
To a stirred solution of 35a in methanol, 1.5 equiv of acetic acid
were added followed by NaBH3CN (2 equiv) and formaldehyde
(37% aqueous solution, 2.5 equiv). The mixture was stirred at room
temperature for 2 h, concentrated by rotary evaporation, and purified
by preparative RP-HPLC (C18, water/acetonitrile with 0.1% trif-
2-(1-{[Dimethylamino)carbonyl]amino}-1-methylethyl)-N-(4-
fluorobenzyl)-5-hydroxy-1-methyl-6-oxo-1,6-dihydropyrimidine-4-
carboxamide (9). A solution of 35a (0.038 g) in dichloromethane
(2 mL) was added to a solution of di-tert-butyl dicarbonate (1.4

HIV-1 Integrase
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5853
equiv) and DMAP (1.0 equiv) in dichloromethane (1 mL), and the
reaction mixture was stirred for 30 min at room temperature. Then
a 2 M solution of dimethylamine in THF (2.0 equiv) was added,
and the mixture was stirred at 40 °C for 2 h. After evaporation of
volatiles, the residue was purified by preparative RP-HPLC to give
some drops of 1 N HCl and charged on cation-exchange resin
cartridges (Varian Mega Bond Elute SCX). The cartridges were
washed with MeOH, and the crude product was eluted with 1 M
ammonia in methanol. The pooled eluents were concentrated to
dryness under reduced pressure to give a solid-pink residue
containing the title product (0.280 g, 72% yield). MS m/z 439 (M
+ H)+. Step 2: N-(4-fluorobenzyl)-5-hydroxy-1-methyl-2-[1-meth-
yl-1-(methylamino)ethyl]-6-oxo-1,6-dihydropyrimidine-4-carboxa-
mide. Compound from previous step (0.28 g, 0.64 mmol) was taken
up in methanol (10 mL) and hydrogenated under hydrogen
atmosphere in the presence of 10% Pd/C. After overnight stirring,
the catalyst was filtered off and the filtrate concentrated under
reduced pressure after addition of a few drops of 1 N HCl. Product
was obtained by trituration with diethyl ether as a white solid (HCl
1
compound 9 in 23% yield. H NMR (DMSO-d6) δ 12.07 (bs, 1
H), 8.94 (t, J ) 7.0 Hz, 1 H), 7.49 (dd, J ) 8.5, 5.6 Hz, 2 H), 7.17
(t, J ) 8.8 Hz, 2 H), 6.52 (s, 1 H), 4.51 (d, J ) 7.0 Hz, 2 H), 3.53
(s, 3 H), 2.79 (s, 6 H), 1.62 (s, 6 H); MS m/z 406 (M + H)+.
HRMS calcd for C19H25O4N5F (M + H)+: 406.18851. Found:
406.18753.
2-(1-{[(Dimethylamino)sulfonyl]amino}-1-methylethyl)-N-(4-
fluorobenzyl)-5-hydroxy-1-methyl-6-oxo-1,6-dihydropyrimidine-4-
carboxamide (10). To a solution of 35a (100 mg, 0.30 mmol, 1.0
equiv) and Et3N (0.05 mL, 0.35 mmol, 1.2 equiv) in dichlo-
romethane (10 mL), dimethylsulfamoyl chloride (0.038 mL, 0.35
mmol, 1.2 equiv) was added. After stirring for 18 h at room
temperature, volatiles were evaporated in vacuo, the residue was
dissolved in dimethylamine (5 mL, 2.0 M in THF), and stirring
was continued overnight. The solvent was reduced in vacuo and
the residue was dissolved in EtOAc, washed with 1 N HCl and
brine, and dried over Na2SO4. After evaporation of volatiles, crude
was purified by RP-HPLC to give compound 10. 1H NMR (DMSO-
d6) δ 12.20 (bs, 1 H), 8.97 (t, J ) 6.2 Hz, 1 H), 7.63 (s, 1 H), 7.37
(dd, J ) 8.6, 5.7 Hz, 2 H), 7.16 (t, J ) 9.1 Hz, 2 H), 4.49 (d, J )
6.4 Hz, 2 H), 3.72 (s, 3 H), 2.66 (s, 6 H), 1.67 (s, 6 H). MS m/z
442 (M + H)+.
1
salt, 98% yield). H NMR (DMSO-d6, 330 K) δ 12.39 (bs, 1 H),
9.90 (bs, 1 H), 9.42 (bs, 2 H), 7.43 (m, 2 H), 7.14 (t, J ) 8.7 Hz,
2 H), 4.52 (d, J ) 6.7 Hz, 2 H), 3.54 (s, 3 H), 2.57 (s, 3 H), 1.77
(s, 6 H). MS m/z 349 (M + H)+. Step 3: following the same
procedure described for 12, treatment of N-(4-fluorobenzyl)-5-
hydroxy-1-methyl-2-[1-methyl-1-(methylamino)ethyl]-6-oxo-1,6-di-
hydropyrimidine-4-carboxamide from above with methyl chloroxoac-
etate gave a residue that was reacted with dimethylamine as
described for 11. Purification by preparative RP-HPLC afforded
1
14 in 56% yield. H NMR (DMSO-d6, 330 K) δ 12.09 (bs, 1 H),
9.03 (t, J ) 5.9 Hz, 1 H), 7.39 (dd, J ) 8.5, 5.7 Hz, 2 H), 7.14 (t,
J ) 8.8 Hz, 2 H), 4.52 (d, J ) 5.9 Hz, 2 H), 3.43 (s, 3 H), 3.01 (s,
3 H), 2.87 (s, 3 H), 2.86 (s, 3 H), 1.68 (s, 6 H). MS m/z 448 (M +
H)+. HRMS calcd for C21H27O5N5F (M + H)+: 448.19907. Found:
448.19934.
Methyl {[1-(4-{[(4-Fluorobenzyl)amino]carbonyl}-5-hydroxy-
1-methyl-6-oxo-1,6-dihydropyrimidin-2-yl)-1-methylethyl]-
amino}(oxo)acetate (12). To a stirred mixture of 35a (3.44 g, 9.3
mmol) and triethylamine (3 equiv) in chloroform (20 mL), methyl
chloroxoacetate (1.5 equiv) was added with external cooling. After
the addition, the ice bath was removed and the mixture was stirred
at room temperature for 3 h. Reaction mixture was then partitioned
between chloroform and 1 N HCl. Organic layer was separated,
washed with brine, dried (Na2SO4), filtered, and concentrated to
obtain title product 12 (80% yield). An aliquot of the crude product
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-(1-methyl-1-{[mor-
pholin-4-yl(oxo)acetyl]amino}ethyl)-6-oxo-1,6-dihydropyrimidine-
4-carboxamide (15). Following the same procedure described for
11, treatment of 12 with morpholine and purification by preparative
1
RP-HPLC gave 15 in 55% yield. H NMR (DMSO-d6, 330 K) δ
12.18 (bs, 1 H), 9.24 (s, 1 H), 8.89 (bt, 1 H), 7.39 (dd, J ) 8.4, 5.4
Hz, 2 H), 7.14 (t, J ) 8.7 Hz, 2 H), 4.51 (d, J ) 5.6 Hz, 2 H), 3.60
(m, 4 H), 3.56 (s, 3 H), 3.49 (m, 2 H), 3.40 (m, 2 H), 1.69 (s, 6 H).
MS m/z 476 (M + H)+. HRMS calcd for C22H27O6N5F (M + H)+:
476.19399. Found: 476.19390.
1
was further purified by preparative RP-HPLC. H NMR (DMSO-
d6) δ 12.2 (bs, 1 H), 9.47 (s, 1 H), 9.04 (t, J ) 6.3 Hz, 1 H), 7.38
(dd, J ) 8.4 Hz, J ) 5.7 Hz, 2 H), 7.16 (t, J ) 8.8 Hz, 2 H), 4.50
(d, J ) 6.3 Hz, 2 H), 3.78 (s, 3 H), 3.45 (s, 3 H), 1.67 (s, 6 H). MS
m/z 421 (M + H)+. HRMS calcd for C19H22O6N4F (M + H)+:
421.15179. Found:. 421.15140.
{[1-(4-{[(4-Fluorobenzyl)amino]carbonyl}-5-hydroxy-1-methyl-
6-oxo-1,6-dihydropyrimidin-2-yl)-1-methylethyl]amino}(oxo)acetic
acid (13). The compound was isolated as byproduct of the above
1
reaction. H NMR (DMSO-d6, 330 K) δ 12.20 (bs, 1 H), 9.38 (s,
1 H), 9.05 (t, J ) 6.5 Hz, 1 H), 7.48 (dd, J ) 8.6, 5.7 Hz, 2 H),
7.16 (t, J ) 8.9 Hz, 2 H), 4.52 (d, J ) 6.5 Hz, 2 H), 3.44 (s, 3 H),
1.68 (s, 6 H). MS m/z 407 (M + H)+. HRMS calcd for
C18H20O6N4F (M + H)+: 407.13614. Found: 407.13564.
N-[1-(4-{[(4-Fluorobenzyl)amino]carbonyl}-5-hydroxy-1-methyl-
6-oxo-1,6-dihydropyrimidin-2-yl)-1-methylethyl]-N,N-dimethyle-
thanediamide (11). Compound 12 (3.14 g, 7.5 mmol) was refluxed
in an excess of 2 M solution of dimethylamine in THF for 2 h.
The reaction mixture was cooled to room temperature, evaporated,
and the residue was purified by RP-HPLC (C18, water/acetonitrile
containing 0.1% of trifluoroacetic acid as eluant) to give 11 in 75%
yield. 1H NMR (DMSO-d6) δ 12.19 (s, 1 H), 9.32 (s, 1 H), 9.06 (t,
J ) 6.4 Hz, 1 H), 7.40 (dd, J ) 8.5, 5.7 Hz, 2 H), 7.18 (t, J ) 8.8
Hz, 2 H), 4.51 (d, J ) 6.4 Hz, 2 H), 3.55 (s, 3 H), 2.93 (s, 3 H),
2.87 (s, 3 H), 1.68 (s, 6 H). 13C NMR (DMSO-d6) δ 168.23, 163.76,
163.09, 161.20 (d, J ) 96.4 Hz), 158.46, 151.90, 145.49, 134.77,
129.40 (d, J ) 3.2 Hz), 124.29, 115.05 (d, J ) 8.5 Hz), 56.50,
41.51, 35.46, 33.42, 32.68, 26.85. 19F-NMR (DMSO-d6) δ -116.71.
MS m/z 434 (M + H)+; mp (Na+ salt): 199-201 °C. HRMS calcd
for C20H25O5N5F (M + H)+: 434.18342. Found: 434.18357.
N-{[1-(4-{[(4-Fluorobenzyl)amino]carbonyl}-5-hydroxy-1-meth-
yl-6-oxo-1,6-dihydropyrimidin-2-yl)-1-methylethyl]}-N,N′,N′-trim-
ethylethanediamide (14). Step 1: 2-{1-[Benzyl(methyl)amino]-1-
methylethyl}-N-(4-fluorobenzyl)-5-hydroxy-1-methyl-6-oxo-1,6-
dihydropyrimidine-4-carboxamide. A solution of 35a (0.33 g, 0.89
mmol) in trimethylorthoformate (5 mL) was treated with 4 equiv
of benzaldehyde, and the mixture stirred for two days at room
temperature. Volatiles were evaporated, and the residue was taken
up in methanol and 1 equiv of NaCNBH3 was added. After 1.5 h
stirring, 2 equiv of formaldehyde and an additional 2 equiv of
NaCNBH3 were added and the resulting mixture was left overnight
at room temperature. The crude mixture was then acidified with
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-(1-methyl-1-{[(4-me-
thylpiperazin-1-yl)(oxo)acetyl]amino}ethyl)-6-oxo-1,6-dihydropy-
rimidine-4-carboxamide (16). Following the same procedure de-
scribed for 11, treatment of 12 with N-methylpiperazine and
purification by preparative RP-HPLC gave 16 (as trifluoroacetic
1
salt) in 67% yield. H NMR (DMSO-d6) δ 12.23 (bs, 1 H), 9.86
(bs, 1 H), 9.48 (s, 1 H), 9.04 (t, J ) 6.9 Hz, 1 H), 7.42 (dd, J )
8.5, 5.7 Hz, 2 H), 7.19 (t, J ) 8.9 Hz, 2 H), 4.56 (d, J ) 7.0 Hz,
2 H), 4.40 (bs, 1 H), 4.00 (bs, 1 H), 3.58 (s, 3 H), 3.53 (bs, partially
obscured by water, 2 H), 3.08 (bs, 4 H), 2.87 (bs, 3 H), 1.69 (s, 6
H). MS m/z 489 (M + H)+. HRMS calcd for C23H30O5N6F (M +
H)+: 489.22562. Found: 489.22460.
N′-[1-(4-{[(2,3-Dimethoxybenzyl)amino]carbonyl}-5-hydroxy-1-
methyl-6-oxo-1,6-dihydropyrimidin-2-yl)-1-methylethyl]-N,N-dim-
ethylethanediamide (18). As for the synthesis of 12, replacing 35a
with 36 gave a residue that was reacted with dimethylamine as
described for 11 to afford crude methyl 2-(1-{[(dimethylami-
no)(oxo)acetyl]amino}-1-methylethyl)-5-hydroxy-1-methyl-6-oxo-
1,6-dihydropyrimidine-4-carboxylate 39. MS m/z 341 (M + H)+.
A portion of this material was taken in N-methylpyrrolidone and
treated with about 1.2 equiv of 2,3-dimethoxybenzylamine at 100
°C for 1 h. Purification by preparative RP-HPLC afforded the title
compound. 1H NMR (DMSO-d6) δ 12.10 (bs, 1 H), 9.31 (s, 1 H),
8.85 (t, J ) 7.3 Hz, 1 H), 7.05-6.96 (m, 2 H), 6.82 (dd, J ) 8.0,

5854 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Summa et al.
5.5 Hz, 1 H), 4.54 (d, J ) 7.3 Hz, 2 H), 3.80 (s, 3 H), 3.78 (s, 3
H), 2.92 (s, 3 H), 2.86 (s, 3 H), 1.68 (s, 6 H). MS m/z 476 (M +
H)+. HRMS calcd for C22H30O7N5 (M + H)+: 476.21398. Found:
476.21387.
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-{1-methyl-1-[(1,3-thia-
zol-4-ylcarbonyl)amino]ethyl}-6-oxo-1,6-dihydropyrimidine-4-car-
boxamide (25). Following the procedure described for 21, treatment
of 35a with 1,3-thiazole-4-carboxylic acid and purification by
preparative RP-HPLC afforded the title compound in 15% yield.
1H NMR (DMSO-d6) δ 12.15 (bs, 1 H), 9.21 (s, 1 H), 9.03 (bt, 1
H), 8.89 (s, 1 H), 8.28 (s, 1 H), 7.43-7.32 (m, 2 H), 7.17 (t, J )
8.8 Hz, 2 H), 4.51 (d, J ) 6.0 Hz, 2 H), 3.48 (s, 3 H), 1.78 (s, 6
H). MS m/z 446 (M + H)+. HRMS calcd for C20H21O4N5FS (M +
H)+: 446.12928. Found: 446.13049.
N-(4-Fluorobenzyl)-5-hydroxy-2-{1-[(1H-imidazol-5-ylcarbony-
l)amino]-1-methylethyl}-1-methyl-6-oxo-1,6-dihydropyrimidine-4-
carboxamide (26). Following the procedure described for 21,
treatment of 35a with 1H-imidazole-5-carboxylic acid and purifica-
tion by preparative RP-HPLC afforded the title compound in 15%
yield. 1H NMR (DMSO-d6) δ 12.18 (bs, 1 H), 9.07 (bt, 1 H), 8.77
(bs, 1 H), 8.45 (bs, 1 H), 8.42 (bs, 1 H), 7.41 (dd, J ) 8.5. 5.6 Hz,
2 H), 7.18 (t, J ) 8.9 Hz, 2 H), 4.52 (d, J ) 6.4 Hz, 2 H), 3.48 (s,
3 H), 1.73 (s, 6 H). MS m/z 429 (M + H)+. HRMS calcd for
C20H22O4N6F (M + H)+: 429.16811. Found: 429.16754.
N′-{1-[4-({[4-Fluoro-2-(methylsulfonyl)benzyl]amino}carbonyl)-
5-hydroxy-1-methyl-6-oxo-1,6-dihydropyrimidin-2-yl]-1-methylethyl}-
N,N-dimethylethanediamide (19). As for the synthesis of 18,
treatment of crude 39 with 4-fluoro-2-(methylsulfonyl)benzylamine
followed by preparative RP-HPLC afforded the title compound.
1H NMR (DMSO-d6) δ 11.78 (bs, 1 H), 9.36 (s, 1 H), 9.00 (t, J )
6.9 Hz, 1 H), 7.72 (m, 1 H), 7.67-7.59 (m, 2 H), 4.86 (d, J ) 6.9
Hz, 2 H), 3.54 (s, 3 H), 2.93 (s, 3 H), 2.87 (s, 3 H), 1.68 (s, 6 H),
a methyl signal obscured by water. MS m/z 512 (M + H)+. HRMS
calcd for C21H27O7N5FS (M + H)+: 512.16152. Found: 512.16148.
N′-[1-(4-{[(3-Chloro-4-fluorobenzyl)amino]carbonyl}-5-hydroxy-
1-methyl-6-oxo-1,6-dihydropyrimidin-2-yl)-1-methylethyl]-N,N-
dimethylethanediamide (20). As for the synthesis of 18, treatment
of crude 39 with 3-chloro-4-fluorobenzylamine followed by pre-
parative RP-HPLC afforded the title compound. 1H NMR (DMSO-
d6) δ 12.03 (bs, 1 H), 9.31 (s, 1 H), 9.08 (t, J ) 6.4 Hz, 1 H), 7.55
(d, J ) 5.6 Hz, 1 H), 7.40-7.37 (m, 2 H), 4.49 (d, J ) 6.4 Hz, 2
H), 3.54 (s, 3 H), 2.91 (s, 3 H), 2.85 (s, 3 H), 1.67 (s, 6 H). MS
m/z 468 (M + H)+. HRMS calcd for C20H24O5N5ClF (M + H)+:
468.14445. Found: 468.14464.
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-{1-methyl-1-[(pyridin-
2-ylcarbonyl)amino]ethyl}-6-oxo-1,6-dihydropyrimidine-4-carboxa-
mide (21). A solution of 35a (0.040 g, 0.11 mmol) in chloroform
(1 mL) was treated with pyridine-2-carboxylic acid (0.027 g, 0.22
mmol), EDCI (0.041 g, 0.022 mmol), HOBT (0.033 g, 0.22 mmol),
and N,N-diisopropylethylamine (0.22 mmol). The mixture was
stirred for 24 h, then concentrated and purified by preparative RP-
HPLC to give 22 in 62% yield. 1H NMR (DMSO-d6) δ 12.16 (bs,
1 H), 9.16 (s, 1 H), 9.07 (t, J ) 6.5 Hz, 1 H), 8.68 (d, J ) 3.6 Hz,
1 H), 8.02 (dd, J ) 6.8, 7.6 Hz, 1 H), 7.92 (d, J ) 6.8 Hz, 1 H),
7.63 (dd, J ) 7.6, 3.6 Hz, 1 H), 7.48 (dd, J ) 8.6, 5.4 Hz, 2 H),
7.27 (t, J ) 8.8 Hz, 2 H), 4.53 (d, J ) 6.5 Hz, 2 H), 3.46 (s, 3 H),
1.75 (s, 6 H). MS m/z 440 (M + H)+. HRMS calcd for
C22H23O4N5F (M + H)+: 440.17286. Found: 440.17206.
N-[1-(4-{[(4-Fluorobenzyl)amino]carbonyl}-5-hydroxy-1-methyl-
6-oxo-1,6-dihydropyrimidin-2-yl)-1-methylethyl]pyridazine-3-car-
boxamide (22). Following the procedure described for 21, treatment
of 35a with pyridazine-3-carboxylic acid and purification by
preparative RP-HPLC afforded the title compound in 25% yield.
1H NMR (DMSO-d6) δ 12.18 (bs, 1 H), 9.68 (s, 1 H), 9.43 (dd, J
) 4.9, 1.8 Hz, 1 H), 9.08 (t, J ) 6.7 Hz, 1 H), 8.11 (dd, J ) 8.8,
1.8 Hz, 1 H), 7.91 (dd, J ) 8.8, 4.9 Hz, 1 H), 7.40 (dd, J ) 8.4,
5.6 Hz, 2 H), 7.27 (t, J ) 8.9 Hz, 2 H), 4.54 (d, J ) 6.7 Hz, 2 H),
3.49 (s, 3 H), 1.79 (s, 6 H). MS m/z 441 (M + H)+. HRMS calcd
for C21H22O4N6F (M + H)+: 441.16811. Found: 441.16696.
N-[1-(4-{[(4-Fluorobenzyl)amino]carbonyl}-5-hydroxy-1-methyl-
6-oxo-1,6-dihydropyrimidin-2-yl)-1-methylethyl]pyrimidine-2-car-
boxamide (23). Following the procedure described for 21, treatment
of 35a with pyrimidine-2-carboxylic acid and purification by
preparative RP-HPLC afforded the title compound in 40% yield.
1H NMR (DMSO-d6) δ 12.15 (bs, 1 H), 9.31 (s, 1 H), 9.08 (t, J )
6.2 Hz, 1 H), 8.96 (d, J ) 4.9 Hz, 2 H), 7.69 (t, J ) 4.9 Hz, 1 H),
7.40 (dd, J ) 8.4, 5.7 Hz, 2 H), 7.17 (t, J ) 8.8 Hz, 2 H), 4.52 (d,
J ) 6.2 Hz, 2 H), 3.50 (s, 3 H), 1.76 (s, 6 H). MS m/z 441 (M +
H)+. HRMS calcd for C21H22O4N6F (M + H)+: 441.16811. Found:
441.16678.
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-{1-methyl-1-[(1,3-ox-
azol-5-ylcarbonyl)amino]ethyl}-6-oxo-1,6-dihydropyrimidine-4-car-
boxamide (24). Following the general procedure described for 21,
treatment of 35a with 1,3-oxaxole-5-carboxylic acid and purification
by preparative RP-HPLC afforded the title compound in 33% yield.
1H NMR (DMSO-d6) δ 12.27 (bs, 1 H), 9.12 (m, 2 H), 8.63 (s, 1
H), 7.91 (s, 1 H), 7.45 (dd, J ) 8.6, 5.4 Hz, 2 H), 7.23 (t, J ) 8.8
Hz, 2 H), 4.57 (d, J ) 6.7 Hz, 2 H), 3.51 (s, 3 H), 1.80 (s, 6 H).
MS m/z 430 (M + H)+. HRMS calcd for C20H21O5N5F (M + H)+:
430.15212. Found: 430.15204.
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-(1-methyl-1-{[(5-methyl-
1,3,4-oxadiazol-2-yl)carbonyl]amino}ethyl)-6-oxo-1,6-dihydropyri-
midine-4-carboxamide (27). Step 1: to a stirred suspension of
potassium 5-methyl-1,3,4-oxadiazole-2-carboxylate (2.67 g, 16.1
mmol) in dichloromethane (300 mL) at room temperature, a solution
of oxalyl chloride (32.2 mmol) in dichloromethane (2.0 M, 16.1
mL) was added dropwise over 10 min. DMF (15 drops, ca. 0.1
mL) was added, and vigorous gas evolution was observed. The
resulting reaction mixture was stirred for a further 30 min, after
which time a second portion of DMF (ca. 0.1 mL) was added and
stirring was continued for a further hour, by which point almost
the entire solid had dissolved. The reaction mixture was then
concentrated under reduced pressure. Meanwhile, 2-[5-(benzoy-
loxy)-4-(methoxycarbonyl)-1-methyl-6-oxo-1,6-dihydropyrimidin-
2-yl]propan-2-amine hydrochloride (36, 3.08 g, 8.05 mmol) was
taken up in dichloromethane (300 mL) and Et3N (4.49 mL, 32.2
mmol) and cloudy solution of the freshly prepared 5-methyl-1,3,4-
oxadiazole-2-carbonyl chloride in dichloromethane (150 mL) was
added dropwise over 15 min. The resulting solution was stirred
for a further hour, and then the mixture was washed with saturated
aq NaHCO3 solution (150 mL), H2O (150 mL), and brine (150 mL).
The mixture was dried (Na2SO4) and concentrated under reduced
pressure. The material was purified by column chromatography on
silica using 100% EtOAc as eluent to yield methyl 5-(benzoyloxy)-
1-methyl-2-(1-methyl-1-{[(5-methyl-1,3,4-oxadiazol-2-
yl)carbonyl]amino}ethyl)-6-oxo-1,6-dihydropyrimidine-4-carbox-
ylate. (2.85 g, 78%). 1H NMR (DMSO-d6) δ 10.02 (bs, 1 H), 8.09
(d, J ) 7.5 Hz, 2 H), 7.76 (t, J ) 7.5 Hz, 1 H), 7.63 (t, J ) 7.5 Hz,
2 H), 3.77 (s, 3 H), 3.56 (s, 3 H), 2.58 (s, 3 H), 1.75 (s, 6 H). MS
m/z 456 (M + H)+. Step 2: a mixture of compound from previous
step (2.85 g, 6.25 mmol) and 4-fluorobenzyl amine (1.72 mL, 15.0
mmol) in MeOH (50 mL) was heated at reflux for 14 h, then cooled
to room temperature and concentrated under reduced pressure. The
resulting mixture was taken up in 0.5 N NaOH solution (240 mL)
and was washed with Et2O (3 × 200 mL). The aqueous layer was
cautiously neutralized with 6 N HCl solution (20 mL) and then
extracted with dichloromethane (200 mL and 4 × 100 mL). The
combined organic extracts were dried (Na2SO4) and concentrated
under reduced pressure to yield 27 (1.94 g, 70% yield), which was
recrystallized from i-PrOH (190 mL, recovery ) 1.30 g, 67%). 1H
NMR (DMSO-d6) δ 12.19 (s, 1 H), 9.83 (s, 1 H), 9.25-8.90 (bs,
1 H), 7.39 (dd, J ) 8.5, 5.6 Hz, 2 H), 7.16 (app. t, J ) 8.8 Hz, 2
H), 4.51 (d, J ) 6.4 Hz, 2 H), 3.48 (s, 3 H), 2.56 (s, 3 H), 1.74 (s,
6 H). MS m/z 445 (M + H)+. HRMS calcd for C20H22O5N6F (M
+ H)+: 445.16302. Found: 445.16278. Melting point 216 °C.
Preparation of the Potassium salt of 27. A standard 0.5 N
solution of KOH in H2O (2.28 mL, 1.14 mmol) was added to a
solution of 27 (510 mg, 1.14 mmol) in MeCN (20 mL), and the
mixture was stirred at room temperature for 20 min. The resulting
solution was frozen in a -78 °C bath and then freeze-dried to yield
the desired potassium salt (553 mg, quant). This material was then

HIV-1 Integrase
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5855
talysis. AdV. Virus Res. 1999, 12, 2331–2338.
dissolved in refluxing EtOH (40 mL) and within a few minutes
material crystallized out. The mixture was further refluxed for 10
min, cooled to rt, and left to stand for 4 h The resulting crystals
were filtered and dried yielding potassium 4-{[(4-fluorobenzyl)ami-
no]carbonyl}-1-methyl-2-(1-methyl-1-{[(5-methyl-1,3,4-oxadiazol-
2-yl)carbonyl]amino}ethyl)-6-oxo-1,6-dihydropyrimidin-5-olate (401
mg, 72%). 1H NMR (DMSO-d6) 11.70-11.20 (bs, 1 H), 9.75 (s, 1
H), 7.33 (dd, J ) 8.8, 5.8 Hz, 2 H), 7.12 (app. t, J ) 8.8 Hz, 2 H),
4.44 (d, J ) 5.8 Hz, 2 H), 3.40 (s, 3 H), 2.56 (s, 3 H), 1.70 (s, 6
H). Melting point 282 °C.
N-(4-Fluorobenzyl)-5-hydroxy-1-methyl-2-(1-methyl-1-{[(5-methyl-
4H-1,2,4-triazol-3-yl)carbonyl]amino}ethyl)-6-oxo-1,6-dihydropy-
rimidine-4-carboxamide (28). Following the procedure described
for 21, treatment of 35a with 5-methyl-4H-1,2,4-triazole-3-car-
boxylic acid and purification by preparative RP-HPLC afforded the
title compound in 72% yield. 1H NMR (DMSO-d6) δ 12.18 (bs, 1
H), 9.07 (bt, 1 H), 8.98 (bs, 1 H), 7.39 (dd, J ) 8.5, 5.7 Hz, 2 H),
7.13 (t, J ) 8.8 Hz, 2 H), 4.52 (d, J ) 6.4 Hz, 2 H), 3.43 (s, 3 H),
2.48 (s, 3 H), 1.72 (s, 6 H). MS m/z 444 (M + H)+. HRMS calcd
for C20H23O4NF7 (M + H)+: 444.17901. Found: 444.17925.
(9) Hazuda, D. J.; Felock, P. J.; Witmer, M.; Wolfe, A.; Stillmock,
K.; Grobler, J. A.; Espeseth, A.; Gabryelski, L.; Schleif, W.; Blau,
C.; Miller, M. D. Inhibitors of strand transfer that prevent
integration and inhibits HIV-1 replication in cells. Science 2000,
287, 646–650.
(10) Summa, V.; Petrocchi, A.; Pace, P.; Matassa, V. G.; De Francesco,
R.; Altamura, S.; Tomei, L.; Koch, U.; Neuner, P. Discovery of R,γ-
diketo acids as potent selective and reversible inhibitors of hepatitis
C virus NS5b RNA-dependent RNA polymerase. J. Med. Chem. 2004,
47, 14–17.
(11) Grobler, J. A.; Stillmock, K.; Hu, B.; Witmer, M.; Felock, P.; Espeseth,
A. S.; Wolfe, A.; Egbertso, M.; Bourgeois, M.; Melamed, J.; Wai,
J. S.; Young, S.; Vacca, J.; Hazuda, D. J. Diketo acid inhibitor
mechanism and HIV-1 integrase: implication for metal binding in the
active site of phosphotranferase enzymes. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 6661–6666.
(12) Summa, V.; Petrocchi, A.; Matassa, V. G.; Taliani, M.; Laufer, R.;
De Francesco, R.; Altamura, S.; Pace, P. HCV NS5b RNA-dependent
RNA polymerase inhibitors: from R,γ-diketoacids to 4,5-dihydroxy-
pyrimidine- or 3-methyl-5-hydroxypyrimidinone carboxylic acids.
Design and synthesis. J. Med. Chem. 2004, 47, 5336–5339.
(13) Summa, V.; Petrocchi, A.; Matassa, V. G.; Gardelli, C.; Muraglia,
E.; Rowley, M.; Gonzalez, Paz, O.; Laufer, R.; Monteagudo, E.; Pace,
P. 4,5-Dihydroxypyrimidine Carboxamides and N-Alkyl-5-hydroxy-
pyrimidinone Carboxamides Are Potent, Selective HIV-Integrase
Inhibitors with Good Pharmacokinetic Profiles in Preclinical Species.
J. Med. Chem. 2006, 49, 6646–6649.
(14) Petrocchi, A.; Koch, U.; Matassa, V. G.; Pacini, B.; Stillmock, K. A.;
Summa, V. From dihydroxypyrimidine carboxylic acids to carboxa-
mide HIV-integrase inhibitors: SAR around the amide mojety. Bioorg.
Med. Chem. Lett. 2007, 17, 350–353.
(15) Pace, P.; Di Francesco, M. E.; Gardelli, C.; Harper, S.; Muraglia, E.;
Nizi, E.; Orvieto, F.; Petrocchi, A.; Poma, M.; Rowley, M.; Scarpelli,
R.; Laufer, R.; Gonzalez Paz, O.; Monteagudo, E.; Bonelli, F.; Hazuda,
D.; Stillmock, K. A.; Summa, V. Dihydroxypyrimidine-4-carboxam-
ides as novel potent and selective HIV-integrase inhibitors. J. Med.
Chem. 2007, 50, 2225–2239.
(16) Gardelli, C.; Nizi, E.; Muraglia, E.; Crescenzi, B.; Ferrara, M.; Orvieto,
F.; Pace, P.; Pescatore, G.; Poma, M.; Rico Ferreira, M. R.; Scarpelli,
R.; Homnick, C. F.; Ikemoto, N.; Alfieri, A.; Verdirame, M.; Bonelli,
F.; Gonzalez Paz, O.; Monteagudo, E.; Taliani, M.; Pesci, S.; Laufer,
R.; Felock, P.; Stillmock, K. A.; Hazuda, D.; Rowley, M.; Summa,
V. Discovery and synthesis of HIV-integrase inhibitors: development
of potent and orally bioavailable N-methyl pyrimidones. J. Med. Chem.
2007, 50, 4953–4975.
Acknowledgment. We thank Anna Alfieri and Francesca
Naimo for plasma protein binding determination and for
analytical chemistry, Silvia Pesci for NMR spectrometry,
Vincenzo Pucci for accurate mass measurements, Peter Felock
for the HIV-integrase enzymatic assays, William Schleif for the
antiviral cell based assays, Massimiliano Fonsi and Marina
Taliani for the microsome stability data, and Annalise Di Marco
for the hepatocytes stability data.
Supporting Information Available: HPLC retention times are
reported. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) 2006 Report on the Global AIDS Epidemic; UNAIDS: Geneva,
Switzerland 2006.
(2) Cervia, J. S.; Smith, M. A. Enfuvirtide (T20): a novel human
immunodeficiency virus type 1 fusion inhibitor. Clin. Infect. Dis. 2003,
37, 1102–1106.
(17) Raltegravir. www.aidsinfo.nih.gov/DrugsdNews/2007. Accessed 2007.
(18) Guare, J. P.; Wai, J. S.; Gomez, R. P.; Anthony, N. J.; Jolly, S. M.;
Cortes, A. R.; Vacca, J. P.; Felock, P. J.; Stillmock, K. A.; Schleif,
W. A.; Moyer, G.; Gabryelski, L. J.; Jin, L.; Chen, I. W.; Hazuda,
D. J.; and Young, S. D. A series of 5-aminosubstituted 4-fluorobenzyl-
8-hydroxy-[1,6]naphthyridine-7-carboxamide HIV-1 integrase inhibi-
tors. Bioorg. Med. Chem. Lett. 2006, 16, 2900–2904.
(3) Dorr, P.; Westby, M.; Dobbs, S.; Griffin, P.; Irvine, B.; Macartney,
M.; Mori, J.; Rickett, G.; Smith-Burchnell, C.; Napier, C.; Webster,
R.; Armour, D.; Price, D.; Stammen, B.; Wood, A.; Perros, M.
Maraviroc (UK-427,857), a potent, orally bioavailable, and selective
small-molecule inhibitor of chemokine receptor CCR5 with broad-
spectrum anti-human immunodeficiency virus type 1 activity. Anti-
microb. Agents Chemother. 2005, 4721–4732.
(4) Pommier, Y.; Johnson, A. A.; Marchand, C. Integrase inhibitors to
treat HIV/AIDS. Nat. ReV. Drug DiscoVery 2005, 4, 236–248.
(5) Zhuang, L.; Wai, J. S.; Embrey, M. W.; Fisher, T. E.; Egbertson, M. S.;
Payne, L. S.; Guare, J. P.; Vacca, J. P.; Hazuda, D. J.; Felock, P. J.;
Wolfe, A. L.; Stillmock, K. A.; Witmer, M. V.; Moyer, G.; Schleif,
W. A.; Gabryelski, L. J.; Leonard, Y. M.; Linch, J. J.; Michelson,
S. R.; Young, S. D. Design and synthesis of 8-hydroxy-[1,6]naph-
thyridines as novel inhibitors of HIV-1 integrase in vitro and in infected
cells. J. Med. Chem. 2003, 46, 453–456.
(6) Hazuda, D. J.; Young, S. D.; Guare, J. P.; Neville, A. J.; Gomez,
R. P.; Wai, J. S.; Vacca, J. P.; Handt, L.; Motzel, S. L.; Klein, H. J.;
Dornadula, G.; Danovich, R. M.; Witmer, M. V.; Wilson, K. A. A.;
Tussey, L.; Schleif, W. A.; Gabryelski, L. S.; Jin, L.; Miller, M. D.;
Casimiro, D. R.; Emini, E. A.; Shiver, J. W. Integrase inhibitors and
cellular immunity suppress retroviral replication in rhesus macaques.
Science 2004, 305, 528–532.
(7) Ramirez, M. 10th European AIDS Conference 2006, Dublin, Ireland,
2006, Abstract LPBS1/6. Beatriz Grinsztejn, B.; Nguyen, B.; Katlama,
C.; Gatell, J. M.; Lazzarin, A.; Vittecoq, D.; Gonzalez, C. J.; Chen,
J.; Harvey, C. M.; Isaacs, R. D. Safety and efficacy of the HIV-1
integrase inhibitor raltegravir (MK-0518) in treatment-experienced
patients with multidrug-resistant virus: a phase II randomised controlled
trial. Lancet 2007, 369, 1261-1269.
(8) For reviews on biology of HIV-1 integrase, see: (a) Esposito, D.;
Craigie, R. HIV-Integrase Structure and Function. AdV. Virus Res.
1999, 52, 319–333. (b) Asante-Appiah, E.; Skalka, A. M. HIV-1
integrase: structural organization, conformational changes, and ca-
(19) Monteagudo, E.; Pesci, S.; Taliani, M.; Fiore, F.; Petrocchi, A.; Nizi,
E.; Rowley, M.; Laufer, R.; Summa, V. Studies of metabolism and
disposition of potent human immunodeficiency virus (HIV) integrase
inhibitors using 19F-NMR spectroscopy. Xenobiotica 2007, 37, 1000–
1012.
(20) Hazuda, D. J.; Anthony, N. J.; Gomez, R. P.; Jolly, S. M.; Wai, J. S.;
Zhuang, L.; Fisher, T. E.; Embrey, M.; Guare, J. P., Jr.; Egbertson,
M. S.; Vacca, J. P.; Huff, J. R.; Felock, P. J.; Witmer, M. V.; Stillmock,
K. A.; Danovich, R.; Grobler, J.; Miller, M. D.; Espeseth, A. S.; Jin,
L.; Chen, I. W.; Lin, J. H.; Kassahun, K.; Ellis, J. D.; Wong, B. K.;
Xu, W.; Pearson, P. G.; Schleif, W. A.; Cortese, R.; Emini, E.; Summa,
V.; Holloway, M. K.; and Young, S. D. A naphthyridine carboxamide
provides evidence for discordant resistance between mechanistically
identical inhibitors of HIV-1 integrase. Proc. Natl. Acad. Sci. U.S.A.
2004, 101, 11233–11238.
(21) Pye, P. J.; Zhong, Y.-L.; Jones, G. O.; Reamer, R. A.; Houk, K. N.;
Askin, D. A Polar Radical Pair Pathway to Assemble the Pyrimidinone
Core of the HIV-Integrase Inhibitor Raltegravir Potassium. Angew.
Chem., Int. Ed. 2008, 47, 4134–4136.
(22) Pucci, V.; Bonelli, F. Development of a simple and reliable accurate
mass liquid chromatography/electrospray ionization mass spectrometry
method for high resolution accurate mass determinations of new drug
entities on a triple quadrupole mass spectrometer. Rapid Commun.
Mass Spectrom. 2007, 21, 3051–3059.
(23) Org. Synth. Coll. 1943, 2, 29.
JM800245Z