Novel HIV-1 protease inhibitors active against multiple PI-Resistant viral strains: Coadministration with indinavir

Article abstract of DOI:10.1016/j.bmcl.2003.08.049

HIV-1 protease inhibitors (PI) with an N-arylpyrrole moiety in the P ₃ position afforded excellent antiviral potency and substantially improved aqueous solubility over previously reported variants. The rapid in vitro clearance of these compounds in human liver microsomes prompted oral coadministration with indinavir to hinder their metabolism by the cyctochrome P450 3A4 isozyme and allow for in vivo PK assessment.

Full text of DOI:10.1016/j.bmcl.2003.08.049

Bioorganic & Medicinal Chemistry Letters 13 (2003) 4027–4030
Novel HIV-1 Protease Inhibitors Active against Multiple
PI-Resistant Viral Strains: Coadministration with Indinavir
Nancy J. Kevin,a Joseph L. Duffy,a,* Brian A. Kirk,a Kevin T. Chapman,a
William A. Schleif,b David B. Olsen,c Mark Stahlhut,c Carrie A. Rutkowski,c
Lawrence C. Kuo,d Lixia Jin,e Jiunn H. Lin,e Emilio A. Eminib and James R. Tataa
aDepartment of Basic Chemistry, Merck Research Laboratories, Rahway, NJ 07065, USA
bDepartment of Virus and Cell Biology, Merck Research Laboratories, West Point, PA 19486, USA
cDepartment of Biological Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
dDepartment of Structural Biology, Merck Research Laboratories, West Point, PA 19486, USA
eDepartment of Drug Metabolism, Merck Research Laboratories, West Point, PA 19486, USA
Received 17 June 2003; accepted 25 August 2003
Abstract—HIV-1 protease inhibitors (PI) with an N-arylpyrrole moiety in the P3 position afforded excellent antiviral potency and
substantially improved aqueous solubility over previously reported variants. The rapid in vitro clearance ofthese compounds in
human liver microsomes prompted oral coadministration with indinavir to hinder their metabolism by the cyctochrome P450 3A4
isozyme and allow for in vivo PK assessment.
# 2003 Elsevier Ltd. All rights reserved.
0
HIV-1 Protease inhibitors (PI), including indinavir,
have demonstrated efficacy as a component ofmulti-
drug therapy in the treatment ofAIDS. 1 The emergence
ofstrains ofHIV that are resistant to current anti-
retroviral therapy2 has prompted intensive research
toward more broadly active chemotherapeutic agents.3
The next generation ofcompounds must posses greater
intrinsic potency against both the wild-type virus and
PI-resistant variants. They must also achieve high and
sustained drug concentration in vivo for more complete
Transposition ofthe pyridyl nitrogen to the P 1 position
(where Y=N) afforded compounds with a more favor-
able P450 inhibition profile in vitro, but with poor
solubility and very low bioavailability.6 We sought to
increase the aqueous solubility ofthis class ofcom-
pounds, while maintaining the lipophilicity required for
compound cell penetration,7 by replacing the furanyl
heterocycle in the P3 biaryl substituent with a pyrrole
moiety (Fig. 1).
4
suppression ofviral replication. Finally, these com-
pounds must be developed as a component ofa multi-
drug regimen. It is therefore imperative that potential
metabolic interactions between chemotherapeutic
agents be well controlled.
Previously we described the synthesis ofprotease inhib-
itors based on the indinavir scaffold with an arylfuran
substituent in the P3 position (Fig. 1).5 These com-
pounds afforded substantially increased potency against
both wild type and PI-resistant viral strains. However,
the most bioavailable compounds (where X=N) were
found to be inhibitors of multiple P450 enzymes.
*Corresponding author. Fax: +1-732-594-5350; e-mail: joseph_duffy@
merck.com
Figure 1. The P3 biaryl substituents: either arylfuran, where X or
Y=N, or N-arylpyrrole, as in 3.
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.08.049

4028
N. J. Kevin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4027–4030
The synthesis ofa representative ofthis class ofcom-
pounds is illustrated in Scheme 1. A modified Paal–
Knorr condensation with aniline afforded the aldehyde
substituent I.8 The aldehyde was condensed with the 3-
pyridyl penultimate intermediate II6 by reductive
amination, affording the fully elaborated product 3.
per batch) using the route from Scheme 1. The modified
Paal–Knorr reactions were carried out in test tubes on a
heating block and monitored by HPLC-MS. These
mixtures were extracted in parallel with saturated
NaHCO3 and dichloromethane. The biphasic mixtures
were passed by gravity filtration through disposable poly-
propylene reaction vessels fitted with 20 micron frit filters,
which afforded passage ofonly the dichloromethane
The compounds synthesized were tested for the ability
to inhibit HIV-1 protease. The enzymes employed were
derived from both the wild-type (NL4-3) virus and from
a panel of clinical viral isolates from patients infected
with highly PI-resistant strains ofHIV-1 (K-60, Q-60,
and V-18). The HIV-1 strains employed were those that
we have observed to exhibit the highest resistance to
indinavir, and the genotype and phenotype ofthese
strains have been reported.5,9 The compounds were also
tested for the ability to inhibit the spread of viral infec-
tion in MT4 human T-lymphoid cells in culture using
viral constructs derived from the strains described
above. The concentrations required to inhibit viral
spread by 95% (CIC95) are presented.10
Table 2. Enzyme inhibitory concentrations (IC50) and viral spread
inhibitory concentrations (CIC95) ofHIV protease inhibitors
PI
P3-Substituent
R
IC50
(nM)
NL4-3 NL4-3 K-60C V-18C Q-60C
Viral spread CIC95 (nM)a
4
5
6
7
8
–CF3 1.32
–CN 0.46
38
46
500
250
62
125
62
500
62
31
31
–F
–CH3 0.16
–Cl 0.18
0.16
15
ꢁ8
31
31
ꢁ8
31
9
–CF3 0.05
–CN 0.15
15
62
31
62
31
31
ꢁ8
31
10
11
12
13
–F
–CH3 0.10
–Cl 0.05
0.05
15
ꢁ8
15
ꢁ8
As we have reported previously, the furanyl analogue 2
afforded substantial improvements in inhibitory potency
(IC50) over indinavir (1) against all viral strains investi-
gated (Table 1). These improvements translated directly
into improved antiviral potency (CIC95). The analogous
pyrrole-substituted compound 3 was substantially less
potent than 2 in the enzymatic assays, particularly
against the enzymes derived from resistant strains of
HIV-1. However, the compound remained a remarkably
potent antiviral in the viral spread assays, affording a
CIC95 of <40 nM against each strain investigated. The
aqueous solubility of 3 (0.74 mg/mL) was similar to
indinavir at pH 5.2, and was substantially improved
over 2.
ꢁ8
ꢁ8
ꢁ8
ꢁ8
ꢁ8
14
15
16
17
18
–CF3 0.04
–CN 0.10
–F 0.05
–CH3 0.06
15
62
15
15
15
125
15
15
15
ꢁ8
31
15
ꢁ8
15
ꢁ8
–Cl
0.03
ꢁ8
ꢁ8
19
20
–H
–Cl
0.02
0.02
15
ꢁ8
31
31
15
15
ꢁ8
ꢁ8
21
0.05
ꢁ8
15
ꢁ8
ꢁ8
aThe ꢁ and ꢂ values denote the lower and upper concentrations tested in
We sought to optimize the phenyl moiety in 3 by paral-
lel solution-phase library syntheses (6–12 compounds
our assays.
Scheme 1. (a) AcOH, 90 ꢀC, 30 min (89%); (b) NaBH(OAc)3, DMF, AcOH, 1 h, (66%).
Table 1. Enzyme inhibitory concentrations (IC50) viral spread inhibitory concentrations (CIC95), and aqueous solubility ofHIV protease inhibitors
0
PI
P3
Substituent
P1
Substituent
HIV-1 Protease inhibition IC50 (nM)
Viral spread CIC95 (nM)a
Solubility
at pH 5.2
(mg/mL)
NL4-3
0.60
K-60C
61.2
V-18C
43.6
Q-60C
20.1
NL4-3
50
K-60C
V-18C
Q-60C
1
2
Indanivir
Benzyl
>1000
>1000
>1000
0.69
0.01
3-Pyridyl
0.02
0.11
0.06
2.70
0.15
2.49
0.07
n.d.
ꢁ8
ꢁ8
ꢁ8
ꢁ8
ꢁ8
ꢁ8
3
3-Pyridyl
15
31
0.74
aThe ꢁand ꢂvalues denote the lower and upper concentrations tested in our assays.

N. J. Kevin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4027–4030
4029
layers. These filtered solutions were then applied
directly to disposable silica gel columns and purified
using a parallel chromatography apparatus.11 The pur-
ified pyrrole carboxaldehydes thus obtained were con-
densed with II by parallel reductive amination, and the
crude reaction mixtures were again applied directly to
the disposable silica gel columns for similar parallel
purification.
navir, both in human liver microsomes and in vivo. The
pyrrole-substituted compounds show no apparent
correlation between CYP3A4 inhibitory potency and
clearance rate in human liver microsomes. One possible
explanation may be that these compounds are metabo-
lized by additional or alternate P450 isoforms. How-
ever, this hypothesis has not been rigorously
investigated.
The inhibitory potencies (IC50) and antiviral activity
(CIC95) ofa representative subset ofcompounds syn-
thesized in this fashion are presented in Table 2. It is
evident that N-arylpyrroles bearing an ortho-substituent
(4–8) are suboptimal in antiviral activity against resis-
tant strains ofHIV-1. The meta-substituted aryl moi-
eties (9–13) and para-substituted compounds (14–18)
afforded generally higher levels ofantiviral potency. Of
these compounds the chloro-substituted N-arylpyrroles
13 and 18 exhibited the greatest antiviral potency. The
effect ofthese chloro-substituents was not synergistic, as
19 and 20 exhibited decreased antiviral activity. However,
the polyfluorinated N-arylpyrrole 21 was equipotent with
the mono-chlorinated species.
Clinical investigations have clearly established the uti-
lity ofcoadministration ofa PI substrate ofCYP3A4
with a more potent inhibitor ofCYP3A4. This strategy,
known as duel PI therapy,14 may afford substantially
enhanced oral exposure ofthe more metabolically labile
compound. Given the relatively high rate ofin vitro
clearance observed with the compounds presented
(Table 3), we reasoned that there was a poor likelihood
ofachieving high and sustained plasma levels ofthese
compounds following oral dosing. We therefore pur-
sued a strategy ofcoadministration ofthese compounds
in dogs with an equal dose ofindinavir. The results are
presented in Table 4.15 The poor pharmacokinetic pro-
file for 2 when dosed alone (5 mpk) has been published6
and is presented for comparison. Coadministration of a
10 mpk dose ofindinavir with 2 (10 mpk) resulted in
substantially improved plasma levels of 2. Indeed, the
enhanced A.U.C. of 2 with coadministration (10.0 mM h)
is comparable to that achieved with indinavir alone. The
more soluble N-arylpyrrole compounds 3 and 21 affor-
ded lower plasma exposure when dosed with indinavir.
This further supports the possibility that isoforms other
than CYP3A4 are responsible for the metabolism of 3
and 21. Conversely, coadministration ofthe potent
CYP3A4 inhibitor 21 with indinavir did afford sig-
nificantly enhanced exposure ofindinavir. The A.U.C.
ofindinavir was more than doubled by this coadmin-
The most potent antiviral compounds synthesized were
investigated further to determine their aqueous solubi-
lity and metabolic profile in vitro, and these results are
presented in Table 3. The halogenated N-arylpyrrole
substituted compounds 13, 18, and 21 exceeded the
gains in aqueous solubility afforded by 3 and were sig-
nificantly more soluble than indinavir itself. The pyr-
role-substituted compounds were also found to be
competitive inhibitors ofthe metabolic P450 isoofrm
CYP3A4, in a similar fashion to indinavir and all other
currently available PI.12 We have previously established
that CYP3A4 is the isoform primarily responsible for
13
.
the metabolism ofindinavir. Thus the inhibition of
this isoform contributes to the slower clearance of indi-
stration (26.6 mM h) as compared to an equal dose of
indinavir alone.
Table 3. In vitro metabolism assays for HIV protease inhibitors
The N-arylpyrrole substituted HIV-1 PI are sub-
stantially more soluble than the phenylfuran derivative
2 while affording comparable antiviral potency. How-
ever, the high metabolic clearance ofthese compounds
in human liver microsomes necessitated coadministra-
tion with the CYP3A4 inhibitor indinavir for pharma-
cokinetic studies. In these investigations higher plasma
levels were achieved with 2 than any ofthe more soluble
pyrrole substituted variants. This is possibly due to the
involvement ofother CYP isoof rms in the metabolism
ofthese compounds.
a
PI
Solubility
pH 5.2
(mg/mL)
P450 Isoform IC50 (mM)
CLint
(mL/min/kg)
CYP3A4
CYP2D6
Indinavir
2
3
13
18
21
0.69
0.01
0.74
1.20
0.98
1.00
0.150
0.430
0.400
0.080
0.050
0.050
ꢂ30
15.6
12.0
n.d.
n.d.
n.d.
49.5
238
332
307
229
218
aDetermined in human liver microsomes.
Table 4. Pharmacokinetics oforally dosed HIV-1 PI (dogs), both alone or with indinavir
PI
Dose PI/indinavir (mg/kg)
PI
Indavir codose
A.U.C. (mM h)
Cmax (mM)
A.U.C. (mM h)
Cmax (mM)
Indinavir
2
2
3
21
0/10
5/0
10/10
10/10
10/10
—
0.1
6.1
5.3
1.9
—
0.08
10.0
7.0
11.4
—
6.5
12.1
13.2
12.5
—
11.2
18.9
26.6
4.4

4030
N. J. Kevin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4027–4030
1999, 274, 23699. (b) Condra, J. H.; Holder, D. J.; Schleif,
References and Notes
W. A.; Blahy, O. M.; Danovich, R. M.; Gabyelski, L. J.;
Graham, D. J.; Laird, D.; Quintero, J. C.; Rhodes, A.; Rob-
bins, H. L.; Roth, E.; Shivaprakash, M.; Yang, T.; Chodake-
witz, J. A.; Deutsch, P. J.; Leavitt, R. Y.; Massari, F. E.;
Mellors, J. W.; Squires, K. E.; Steigbigel, R. T.; Teppler, H.;
Emini, E. A. J. Virol. 1996, 70, 8270.
10. Vacca, J. P.; Dorsey, B. D.; Schleif, W. A.; Levin, R. B.;
McDaniel, S. L.; Darke, P. L.; Zugay, J.; Quintero, J. C.;
Blahy, O. M.; Roth, E.; Sardana, V. V.; Schlabach, A. J.;
Graham, P. I.; Condra, J. H.; Gotlib, L.; Holloway, M. K.;
Lin, J.; Chen, I.-W.; Vastag, K.; Ostovic, D.; Anderson,
P. S.; Emini, E. A.; Huff, J. R. Proc. Nat. Acad. Sci. 1994,
91, 4096.
1. Kempf, D. J.; Molla, A.; Hsu, A. In Antiretroviral Therapy;
De Clercq, E. Ed.; ASM: Washington DC., 2001; p 147.
2. Romano, L.; Venturi, G.; Giomi, S.; Pippi, L.; Valensin,
P. E.; Zazzi, M. J. Med. Virol. 2002, 66, 143.
3. Rusconi, S.; Catamancio, S. Expert Opin. Investig. Drugs
2002, 11, 387.
4. Dorsey, B. D.; McDoough, C.; McDaniel, S. L.; Levin,
R. B.; Newton, C. L.; Hoffman, J. M.; Darke, P. L.; Zugay-
Murphy, J. A.; Emini, E. A.; Schleif, W. A.; Olsen, D. B.;
Stahlhut, M. W.; Rutkowski, C. A.; Kuo, L. C.; Lin, J. H.;
Chen, I.-W.; Michelson, S. R.; Holloway, M. K.; Huff, J. R.;
Vacca, J. P. J. Med. Chem. 2000, 43, 3386.
5. Duffy, J. L.; Rano, T. A.; Kevin, N. J.; Chapman, K. T.;
Schleif, W. A.; Olsen, D. B.; Stahlhut, M.; Rutkowski, C. M.;
Kuo, L. C.; Jin, L.; Lin, J. H.; Emini, E. A.; Tata, J. R.
Bioorg. Med. Chem. Lett. 2003, 13, 2569.
11. An early prototype ofthe QUAD3 apparatus rfom Bio-
tage, Inc. was utilized for parallel chromatography.
12. Williams, G. C.; Sinko, P. J. Adv. Drug Delivery Reviews
1999, 39, 211.
6. Duffy, J. L.; Kirk, B. A.; Kevin, N. J.; Chapman, K. T.;
Schleif, W. A.; Olsen, D. B.; Stahlhut, M.; Rutkowski, C. A.;
Kuo, L. C.; Jin, L.; Lin, J. H.; Emini, E. A.; Tata, J. R.
Bioorg. Med. Chem. Lett. 2003, 13, 3323.
7. Dorsey, B. D.; McDaniel, S. L.; Levin, R. B.; Vacca, J. P.;
Darke, P. L.; Zugay, J. A.; Emini, E. A.; Schleif, W. A.; Lin,
J. H.; Chen, I.-W.; Holloway, M. K.; Anderson, P. S.; Huff,
J. R. Bioorg. Med. Chem. Lett. 1994, 4, 2769.
8. Sundberg, R. J. In Comprehensive Heterocyclic Chemistry
II; Bird, C. W. Ed.; Elsevier Science, Tarrytown, NY, 1996,
Vol. 2, p 149.
9. (a) Olsen, D. B.; Stahlhut, M. W.; Rutkowski, C. A.;
Schock, H. B.; vanOlden, A. L.; Kuo, L. C. J. Biol. Chem.
13. Chiba, M.; Hensleigh, M.; Nishime, J. A.; Balani, S. K.;
Lin, J. H. Drug Metab. Disp. 1996, 24, 307.
14. Kempf, D. J.; Akhterzzaman, M.; Hsu, A. In Anti-
retroviral Therapy; De Clercq, E., Ed.; ASM Press: Washing-
ton, DC, 2001, Chapter 8.
15. The use ofanimals was done under the purview ofan
Institutional Animal Care and Use Committee, and all
applicable regulations and laws pertaining to the use of
laboratory animals were followed. For experimental methods
for both in vitro and in vivo pharmacokinetic investigations,
see: Lin, J. H.; Chiba, M.; Balani, S. K.; Chen, I.-W.; Kwei,
G. Y.-S.; Vastag, K. J.; Nishime, J. A. Drug Metab. Disp.
1996, 24, 1111.