Discovery of a novel and potent class of anti-HIV-1 maturation inhibitors with improved virology profile against gag polymorphisms

Article abstract of DOI:10.1016/j.bmcl.2017.04.042

A new class of betulin-derived α-keto amides was identified as HIV-1 maturation inhibitors. Through lead optimization, GSK8999 was identified with IC₅₀ values of 17?nM, 23?nM, 25?nM, and 8?nM for wild type, Q369H, V370A, and T371A respectively. When tested in a panel of 62 HIV-1 isolates covering a diversity of CA-SP1 genotypes including A, AE, B, C, and G using a PBMC based assay, GSK8999 was potent against 57 of 62 isolates demonstrating an improvement over the first generation maturation inhibitor BVM. The data disclosed here also demonstrated that the new α-keto amide GSK8999 has a mechanism of action consistent with inhibition of the proteolytic cleavage of CA-SP1.

Full text of DOI:10.1016/j.bmcl.2017.04.042

Accepted Manuscript
Discovery of a Novel and Potent Class of anti-HIV-1 Maturation Inhibitors with
Improved Virology Profile against Gag Polymorphisms
Jun Tang, Stacey A. Jones, Jerry L. Jeffrey, Sonia R. Miranda, Cristin M.
Galardi, David M. Irlbeck, Kevin W. Brown, Charlene B. McDanal, Brian A.
Johns
PII:
S0960-894X(17)30418-3
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.04.042
BMCL 24895
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
22 February 2017
11 April 2017
12 April 2017
Please cite this article as: Tang, J., Jones, S.A., Jeffrey, J.L., Miranda, S.R., Galardi, C.M., Irlbeck, D.M., Brown,
K.W., McDanal, C.B., Johns, B.A., Discovery of a Novel and Potent Class of anti-HIV-1 Maturation Inhibitors with
Improved Virology Profile against Gag Polymorphisms, Bioorganic & Medicinal Chemistry Letters (2017), doi:
http://dx.doi.org/10.1016/j.bmcl.2017.04.042
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Discovery of a Novel and Potent Class of anti-HIV-1 Maturation Inhibitors
with Improved Virology Profile against Gag Polymorphisms
Jun Tang a,*, Stacey A. Jones a, Jerry L. Jeffrey a, Sonia R. Miranda a, Cristin M.
Galardi a, David M. Irlbeck a, Kevin W. Brown a, Charlene B. McDanal a, and
Brian A. Johnsa
aGlaxoSmithKline Research & Development, Infectious Diseases Therapy Area Unit,
Research Triangle Park, NC 27709, USA
Keywords: Betulin, Bevirimat, HIV-1, GSK8999, Maturation inhibitor
More than 35 anti-retroviral drugs have been approved for HIV-1 treatment
and shown to significantly increase patient survival. The once deadly disease can be
managed as a chronic infection if patients access and are compliant with effective
treatments. However, HIV-1 continues to infect millions of people around the world
every year and leads to morbidity or death in the majority of those over time.
Interestingly while treatments have been nothing less than miraculous, statistics show
that for every ten people who are put on highly active anti-retroviral therapy (HAART),
sixteen more people are newly infected suggesting the epidemic is not going to end
anytime soon. Currently, efforts to slow down the impact of HIV infection rates using
pre-exposure prophylaxis (PrEP) for prevention have made huge strides however these
are only part of the strategy to combat the disease. Cure research has also gained
momentum in earnest but will take time to develop basic science and bring forth
approaches that can be safe and effective. In the meantime, HAART needs to be further
optimized with the addition of novel antiviral mechanisms to address the emergence of
drug resistant HIV-1 variants as well as continue to improve convenience and options for
patients both in early lines of therapy and treatment experienced populations.1
Since the first-in-class HIV-1 maturation inhibitor bevirimat (BVM, PA-457,
MPC-4326) was halted for further development in 2010 due to baseline gag
polymorphisms at positions Q369, V370, or T371,2 many efforts have focused on finding
alternatives to overcome these issues.3,4 Recent studies also confirmed that maturation
inhibitors such as BVM, act by stabilizing HIV-1 CA-SP1 lattice, which results in
preventing the viral protease from accessing the proteolytic cleavage sites. The dramatic
reduction of the antiviral activities of gag polymorphisms at positions Q369, V370, or
T371, may be related to a less stabilized lattice with BVM.5 One initial attempt to
overcome this hurdle with macrocyclized betulin derivatives ended with only modest

success.6 Herein, we disclose initial lead optimization results of a series of potent HIV-1
maturation inhibitors through optimization of substituents at C-17, which are also
potent against BVM resistant gag polymorphisms at positions Q369, V370, and T371.
Figure 1. Design of C-28 α-keto acid 1 with reduced lipophilicity
One of our major efforts has focused on a novel template, compound 1 (Fig.1),
which consists of several modifications of the BVM first generation stucture. Two
additional carbonyl groups were designed into the triterpene skeleton in order to
increase the molecule polarity with an intention to decrease plasma protein binding
shift effects on antiviral potency. As described in Scheme 1, commercially available
betulin 2 was first protected with acetyl groups at both the C-3, and C-28 hydroxyls to
give compound 3. Without further purification, the crude material underwent a double
bond migration from the C-20 isopropenyl to the C-18 and C-19 under acidic condition.
The overall yield was about 67% from compound 2 to 4.
Scheme 1. Synthesis of compound 10
Compound 4 was easily oxidized with Na2Cr2O7 to introduce a carbonyl at the
C-21 to give enone 5. The crude material 5 was subjected to a selective hydrolysis with
KOH to afford alcohol 6. The hydroxyl group at the C-28 of compound 6 was transformed

into a formyl group by pyridinium chlorochromate in the presence of silica gel, and the
overall yield from 4 to 7 was about 53%. Next, Henry reaction was performed to add the
nitromethane to the formyl group (compound 8), followed by Dess-Martin
Periodinane (DMP) oxidation to give compound 9. Further treatment of compound 9
with sodium nitrite in DMSO yielded the common starting material compound 10.
The corresponding α-keto amides could be made through multiple common
coupling routes from compound 10 or other procedures.7 The most efficient way is
described in Scheme 2 as exemplified with compound 18. The α-keto carboxylic acid was
sequentially treated with oxalyl chloride, followed by the addition of the amine, HCl
hydrolysis for the removal of the acetyl group at the C-3, and the installation of the 2,
2-dimethyl-4-oxobutanoic acid tail to afford compound 18.
Scheme 2. Synthesis of α-keto amides as exemplified with compound 18
A variety of α-keto amides at the C-17 with different amines were synthesized.
We focused on activity against gag polymorphisms to triage these compounds to direct
medicinal chemistry efforts to discover novel pharmacophore elements. Initial antiviral
activity was assessed in a MT4 assay using NL4-3 with a consensus clade B CA-SP1
region (wt) and compared to a point mutant V370A (Site Directed Mutagenesis). The
compounds in Table 1 represent a selection of the C-28 α-keto amides that were screened
for inhibition with both assays.
Table 1. Anti-HIV-1 activity of compounds in wt & V370A7

MT4 (NL.wt)
IC50 µM
MT4 (NL.V370A)
IC50 µM
MT4 (NL.wt) MT4 (NL.V370A)
Compound #
Compound #
IC50 µM
IC50 µM
mean
mean
mean
mean
(+/- SD)
(+/- SD)
(+/- SD)
(+/- SD)
0.017
0.486
1
> 10
NA
20
(0.005)
(0.988)
1.316
6.667
0.016
(N=1)
0.37
11
12
21
22
(2.227)
(5.163)
(N=1)
5.600
0.012
0.087
> 10
(6.223)
(0.009)
(0.116)
0.028
0.329
0.085
1.071
13
14
23
24
(0.013)
(0.548)
(0.001)
(0.170)
0.011
0.027
0.037
0.193
(0.006)
(0.027)
(0.011)
(0.283)
O
1.000
0.135
1.315
N
15
> 10
25
H
(0.990)
O
(0.108)
(1.372)
CF3
1.183
7.500
0.099
(N=1)
8
16
17
18
19
26
27
28
29
(0.540)
(3.536)
(N=1)
0.251
0.664
0.021
1.625
(0.243)
(0.469)
(0.007)
(1.475)
0.13
1.8
6.200
> 10
(N=1)
(N=1)
(5.374)
0.016
0.202
0.018
0.019
(0.006)
(0.171)
(0.001)
(0.001)
It was very intriguing that compound 1, closely resembled BVM, was inactive
against wt. N-mono-methylsubstituted α-keto amide (compound 11) only showed
marginal activities against both wt and V370A. α-Keto azetidine amide (compound 12)
did not work well either. Encouragingly, lipophilic cyclohexyl and cyclohexylmethyl
α-keto amides (compounds 13, 14) regained activity not only against wt, but also against
V370A. Compound 14 showed for the first time 27 nM potency against the V370A
polymorph. On the other hand, the incorporation of a distal amine as in compounds 15
and 16 negatively impacted potency. α-Keto amides with anilines (compounds 17, 18)
only demonstrated moderate activities, while the benzyl amine derivative compound 19
exhibited improved activity. Based on the results from compounds 14 and 19, we
anticipated that an optimal lipophilic amine substituent was essential for antiviral
activities against both wt and V370A. As benzyl amine is a good handle for rapid

diversification, further exploration focused on various substitution patterns at the
benzyl ring. Among all the substitution patterns at the phenyl ring, para-chlorobenzyl
amine (compound 22) gave the best results in both assays. Other substitution patterns
such as ortho- (compound 20), and meta- (compound 21) demonstrated inferior activities.
Different patterns of dichloro substitutions (compounds 23, 24) were not able to improve
the potencies. Furthermore, the replacement of the para-Cl at the phenyl ring with
either an electron withdrawing group (compound 25) or a donating group (compound 26)
only led to the loss of activities due to either electronic effects or intolerable size of the
substituents at the para-position of the phenyl ring. Thienylmethyl amine (compound
27) resembling benzyl amine in size also proved to be an inferior substitution. A basic
functionality such as a pyridine ring (compound 28), was also poorly tolerated. Benzylic
disubstitution (compound 29) showed equipotency against both wt (18 nM) and V370A
(19 nM) viruses.
The next avenue of exploration focused on further benzylic position
substitutions. Q369H, T371A mutants were also added to the screening panel in order to
have overall more complete assessment of the improved virology profile against gag
polymorphisms.
Table 2. Anti-HIV-1 activity of compounds in wt, Q369H, V370A, and T371A 7

MT4 (NL.wt)
IC50 µM
MT4 (NL.Q369H) MT4 (NL.V370A) MT4 (NL.T371A)
Compound #
IC50 µM
IC50 µM
IC50 µM
mean
mean
mean
mean
(+/- SD)
(+/- SD)
(+/- SD)
(+/- SD)
0.039
0.060
0.474
0.019
30
31
(0.013)
(0.019)
(0.501)
(0.003)
0.018
0.025
0.025
0.005
(0.009)
(0.012)
(0.029)
(0.002)
0.054
0.062
0.372
0.026
32
33
34
(0.018)
(0.013)
(0.132)
(0.002)
0.028
0.074
0.064
0.012
(0.001)
(0.006)
(0.014)
(0.006)
0.019
0.017
0.017
0.011
(0.004)
(0.003)
(0.006)
(0.002)
35
0.017
0.023
0.025
0.008
GSK8999
(0.010)
(0.016)
(0.027)
(0.007)
0.029
0.048
0.065
0.017
36
37
(0.005)
(0.015)
(0.018)
(0.001)
0.066
(N=1)
0.1
0.11
0.033
(N=1)
(N=1)
(N=1)
As shown in Table 2, benzylic position substitutions including stereochemical
effects had a strong impact on the virology profile against gag polymorphisms Q369H,
V370A, and T371A. Compound 31 with a (R)-methyl group at the benzylic position
demonstrated much better activities across all QVT mutants in this panel in comparison
to its counterpart (S)-methyl diasteromer 30. When the substituent size was increased
from methyl to ethyl (compounds 32, 33), a slightly reduction of the antiviral activities
on all QVT mutants was observed. Benzylic dimethyl substituted compound 34 again
showed almost same activities against wt, Q369H, V370A, and T371A. Benzylic
cyclopropyl moiety (compound 35, GSK8999) exhibited similar activities when compared
to compound 31, however, was more desirable due to the lack of an additional chiral
center. Further ring size enlargement (compounds 36, 37) started to cause the loss of
activities across wt and its mutants.
Due to a robust polymorph antiviral profile, GSK89998 was selected to evaluate
against a panel of BVM-resistant viruses (Table 3) in a MT4 assay and a PBMC assay.7,9
GSK8999 exhibited greatly improved inhibition across the BVM-resistant viruses

tested. In addition, ASJM108 carries the V370A polymorphism plus some other altered
residues resulting in a profile which is highly resistant to BVM. Fortunately, GSK8999
showed little loss of potency to this challenging isolate. Beyond these, GSK8999 also
demonstrated consistent potency against a clade A viral isolate I-2496. Overall, the
results confirmed that our exploration of α-keto amides provided sets of compounds that
achieved excellent inhibition of wt, Q369H, V370A, T371A and beyond.
Table 3. Activity of BVM and GSK8999 against a panel of viruses
The effect of human serum on GSK8999 antiviral potency was assessed in a life
cycle HIV assay. With the presence of 40% human serum, GSK8999 exhibited a 9.7 fold
reduction of antiviral activity compared to BVM at the same condition with a 52 fold
reduction. The mechanism of action of GSK8999 was also confirmed by a capsid western
blot assay10 as depicted in Figure 2. Similar to BVM, GSK8999 demonstrated inhibition
of the CA-SP1 proteolytic cleavage resulting in the buildup of the CA-SP1 protein.
Figure 2. Mechanism of action: GSK8999 inhibits cleavage of CA-SP1

GSK8999 was further tested in a panel of 62 isolates covering various clades
including A, AE, B, C, and G using a PBMC based assay, only 5 isolates (91BR014,
96THNP1538, 00KEKR1008, 96USNG031, 01ET188) were partially or not inhibited by
GSK8999 with IC50 greater than 60 nM, compared to BVM which lost potency to more
than 60% of the isolates in the same panel. The panel also included viruses with known
resistance genotypes to show that GSK8999 is not cross resistant with marketed
anti-retroviral products.
Table 4. GSK8999 has an improved isolate coverage in a panel of 62 isolates (N=1)
IC50 > 100 nM
In summary, a new class of betulin-derived α-keto amides was identified as
HIV-1 maturation inhibitors. The efficient conversion of commercially available betulin
compound 2 to α-keto acid compound 10 allowed for lead optimization of α-keto amides.
Among these compounds, GSK8999 was identified with IC50 values of 17 nM, 23 nM, 25
nM, and 8 nM for wt, Q369H, V370A, and T371A respectively. When tested in a panel of
62 HIV-1 isolates covering a diversity of CA-SP1 genotypes including A, AE, B, C, and G
using a PBMC based assay, GSK8999 was potent against 57 of 62 isolates demonstrating
an improvement over the first generation maturation inhibitor BVM. The data disclosed
here also demonstrated that the new α-keto amide GSK8999 has a mechanism of action
consistent with inhibition of the proteolytic cleavage of CA-SP1. These data establish
that these new α-keto amides as HIV maturation inhibitors can be broad spectrum
inhibitors of wild type, its polymorphisms and are a viable mechanism for further

optimization.
Acknowledgments
The authors gratefully acknowledge the experimental support from Nianhe
Han, Daxin Gao, Yongyong Wu, Fangxian Ning, Heping Yang, and Zhimin Jin at
ShangPharma Discovery Chemistry Services, Zhangjiang High-tech Park, Pudong,
Shanghai 201203, China.
References and notes
[1] (a) Adamson, S. C.; Freed, O. E. Novel approaches to inhibiting HIV-1 replication.
Antiviral Res. 2010, 85(1), 119-141. (b) Lataillade, M.; Chiarella, J.; Yang, R.;
Schnittman, S.; Wirtz, V.; Uy, J.; Seekins, D.; Krystal, M.; Mancini, M.; McGrath,
D.; Simen, B.; Egholm, M.; Kozal, M. Prevalence and clinical significance of HIV
drug resistance mutations by ultra-deep sequencing in antiretroviral-naïve subjects
in the CASTLE study. PLoS One 2010, 5(6), e10952. (c) Jakobsen, M. E.; Jakobsen,
M. R; Tolstrup, M.; Søgaard, O. S.; Jørgensen, L. B.; Gorry, P. R.; Laursen, A.;
Ostergaard, L. Transmission of HIV-1 drug-resistant variants: Prevalence and effect
on treatment outcome. Clin. Infect. Dis. 2010, 50, 566-573.
[2] Dang, Z.; Ho, P.; Zhu, L.; Qian, K.; Lee,K. H.; Huang, L.; Chen,C. H. New betulinic
acid derivatives for bevirimat-resistant human immunodeficiency virus type-1. J.
Med. Chem., 2013, 56, 2029-2037 and references therein.
[3] a) Qian, K.; Bori, D. I.; Chen, C. H.; Huang, L.; Lee, K. H. Anti-AIDS agents 90.
Novel C‑28 modified bevirimat analogues as potent HIV maturation inhibitors. J.
Med. Chem., 2012, 55(18), 8128-8136. (b) Coric, P.; Turcaud, S.; Souquet, F.; Briant,
L.; Gay, B.; Royer, J.; Chazal, N.; Bouaziz, S. Synthesis and biological evaluation of a
new derivative of bevirimat that targets the Gag CA-SP1 cleavage site. Eur. J. Med.
Chem., 2013, 62, 453-465.
[4] a) Zhao, Y.; Gu, Q.; Morris-Natschke, S. L.; Chen, C. H.; Lee, K. H. Incorporation of
privileged structures into bevirimat can improve activity against wild-type and
bevirimat-resistant HIV-1. J. Med. Chem., 2016, 59 (19), 9262–9268. (b)
Nowicka-Sans, B.; Protack, T.; Lin, Z.; Li, Z.; Zhang, S.; Sun, Y.; Samanta, H.; Terry,
B.; Liu, Z.; Chen, Y.; Sin, N.; Sit, S.-Y.; Swidorski, J. J.; Chen, J.; Venables, B. L.;
Healy, M.; Meanwell, N. A.; Cockett, M.; Hanumegowda, U.; Regueiro-Ren, A.;
Krystal, M.; Dicker, I. B. Identification and characterization of BMS-955176, a
second-generation HIV-1 maturation inhibitor with improved potency, antiviral

spectrum, and gag polymorphic coverage. Antimicrobial Agents and Chemotherapy
2016, 60(7), 3956-3969.
[5] a) Schur, F. K. M.; Obr, M.; Hagen, W. J. H.; Wan, W.; Jakobi, A. J.; Kirkpatrick, J.
M.; Sachse, C.; Kraeusslich, H.-G.; Briggs, J. A. G. An atomic model of HIV-1
capsid-SP1 reveals structures regulating
assembly
and
maturation.
Science 2016, 353, 506-508. b) Fontana, J.; Keller, P. W.; Urano, E.; Ablan, S. D.;
Steven, A. C.; Freed, E. O. Identification of an HIV-1 mutation in spacer peptide 1
that stabilizes the immature CA-SP1 lattice. J. of Virol., 2016, 90(2), 972-978.
[6] Tang, J.; Jones, S. A.; Jeffrey, J. L.; Miranda, S. R.; Galardi, C. M.; Irlbeck, D. M.;
Brown, K. W.; McDanal, C. B.; Han, N.; Gao, D.; Wu, Y.; Shen, B.; Liu, C.; Xi, C.;
Yang, H.; Li, R.; Yu, Y.; Sun, Y.; Jin, Z.; Wang, E.; Johns, B. A. Synthesis and
biological evaluation of macrocyclized betulin derivatives as a novel class of
anti-HIV-1 maturation inhibitors. The Open Med Chem J. 2014, 8, 23-27.
[7] The description of these assays and synthetic routes for each compound can be
found: Gao, D.; Han, N.; Jin, Z.; Ning, F.; Tang, J.; Wu, Y.; Yang, H. Derivatives of
betulin. WO2,011,100,308, August 18, 2011.
[8] GSK8999:
4-((3aR,5aR,5bR,7aR,9S,11aR,11bR,13aS)-3a-(2-(1-(4-chlorophenyl)cyclopropylamin
o)-2-oxoacetyl)-1-isopropyl-5a,5b,8,8,11a-pentamethyl-2-oxo-3,3a,4,5,5a,5b,6,7,7a,8,
9,10,11,11a,11b,12,13,13a-octadecahydro-2H-cyclopenta[a]chrysen-9-yloxy)-2,2-dim
ethyl-4-oxobutanoic acid, 1HNMR (400 MHz, CDCl3) δ ppm 7.41 (1H, s), 7.26 (2H, d,
J = 8.8 Hz), 7.20 (2H, d, J = 8.4 Hz ), 4.52 (1H, dd, J = 11.2, 5.2 Hz), 3.21-3.14 (1H, m),
2.72-2.68 (2H, m), 2.61 (1H, d, J = 5.2 Hz), 2.57 (1H, s), 2.45 (1H, t, J = 8.4, 7.2 Hz),
2.19 (1H, d, J = 18.8 Hz), 1.89 (2H, d, J = 2.4 Hz), 1.77-0.78 (47H, m). LC/MS: m/z
calculated 775.4, found 776.2 (M+1) +. For GSK8999, CC50 was higher than 10 µM in
a MT4 cytotoxicity assay.
[9] The description of this PBMC assay can be found: Hatcher, A. M.; Johns, A. B.;
Martin, T. M.; Tabet, A. E.; Tang, J. Derivatives of betulin. WO2,013,090,664, June
20, 2013.
[10]Li, F.; Goila-Gaur, R.; Salzwedel, K.; Kilgore, N. R.; Reddick, M.; Matallana, C.; Ca
stillo, A.; Zoumplis, D.; Martin, D. E.; Orenstein, J. M.; Allaway, G. P.; Freed, E. O.;
Wild, C. T. PA-457: A potent HIV inhibitor that disrupts core condensation by
targeting a late step in Gag processing. PNAS 2003, 100(23), 13555-13560.
Footnotes
*Corresponding author. For J.T.: phone, +1-919-483-4173; FaX: +1-919-315-6923; e-mail,
jun.x.tang@gsk.com.

Discovery of a Novel and Potent Class of anti-HIV-1 Maturation Inhibitors
with Improved Virology Profile against Gag Polymorphisms
Jun Tang a,*, Stacey A. Jones a, Jerry L. Jeffrey a, Sonia R. Miranda a, Cristin M.
Galardi a, David M. Irlbeck a, Kevin W. Brown a, Charlene B. McDanal a, and
Brian A. Johnsa
aGlaxoSmithKline Research & Development, Infectious Diseases Therapy Area Unit,
Research Triangle Park, NC 27709, USA
Abstract: A new class of betulin-derived α-keto amides was identified as HIV-1
maturation inhibitors. Through lead optimization, GSK8999 was identified with IC50
values of 17 nM, 23 nM, 25 nM, and 8 nM for wild type, Q369H, V370A, and T371A
respectively. When tested in a panel of 62 HIV-1 isolates covering a diversity of CA-SP1
genotypes including A, AE, B, C, and G using a PBMC based assay, GSK8999 was potent
against 57 of 62 isolates demonstrating an improvement over the first generation
maturation inhibitor BVM. The data disclosed here also demonstrated that the new
α-keto amide GSK8999 has a mechanism of action consistent with inhibition of the
proteolytic cleavage of CA-SP1.
Keywords: Betulin, Bevirimat, HIV-1, GSK8999, Maturation inhibitor