The design, synthesis and structure-activity relationships associated with C28 amine-based betulinic acid derivatives as inhibitors of HIV-1 maturation

Article abstract of DOI:10.1016/j.bmcl.2018.03.067

The design and synthesis of a series of C28 amine-based betulinic acid derivatives as HIV-1 maturation inhibitors is described. This series represents a continuation of efforts following on from previous studies of C-3 benzoic acid-substituted betulinic acid derivatives as HIV-1 maturation inhibitors (MIs) that were explored in the context of C-28 amide substituents. Compared to the C-28 amide series, the C-28 amine derivatives exhibited further improvements in HIV-1 inhibitory activity toward polymorphisms in the Gag polyprotein as well as improved activity in the presence of human serum. However, plasma exposure of basic amines following oral administration to rats was generally low, leading to a focus on moderating the basicity of the amine moiety distal from the triterpene core. The thiomorpholine dioxide (TMD) 20 emerged from this study as a compound with the optimal antiviral activity and an acceptable pharmacokinetic profile in the C-28 amine series. Compared to the C-28 amide 3, 20 offers a 2- to 4-fold improvement in potency towards the screening viruses, exhibits low shifts in the EC₅₀ values toward the V370A and ΔV370 viruses in the presence of human serum or human serum albumin, and demonstrates improved potency towards the polymorphic T371A and V362I virus variants.

Full text of DOI:10.1016/j.bmcl.2018.03.067

Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The design, synthesis and structure-activity relationships associated
with C28 amine-based betulinic acid derivatives as inhibitors of HIV-1
maturation
a,
a
a
a
a
a
Yan Chen a, , Sing-Yuen Sit , Jie Chen , Jacob J. Swidorski , Zheng Liu , Ny Sin , Brian L. Venables ,
⇑
⇑
Dawn D. Parker b,d, Beata Nowicka-Sans e, Zeyu Lin c, Zhufang Li c,d, Brian J. Terry c, Tricia Protack c,
Sandhya Rahematpura b, Umesh Hanumegowda b,d, Susan Jenkins b,d, Mark Krystal c,d, Ira D. Dicker c,d
Nicholas A. Meanwell a, Alicia Regueiro-Ren a
,
a Department of Discovery Chemistry and Molecular Technologies, Bristol-Myers Squibb Research and Development, 5 Research Parkway, Wallingford, CT 06492, USA
b Department of Pharmaceutical Candidate Optimization, Bristol-Myers Squibb Research and Development, 5 Research Parkway, Wallingford, CT 06492, USA
c Department of Virology, Bristol-Myers Squibb Research and Development, 5 Research Parkway, Wallingford, CT 06492, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The design and synthesis of a series of C28 amine-based betulinic acid derivatives as HIV-1 maturation
inhibitors is described. This series represents a continuation of efforts following on from previous studies
of C-3 benzoic acid-substituted betulinic acid derivatives as HIV-1 maturation inhibitors (MIs) that were
explored in the context of C-28 amide substituents. Compared to the C-28 amide series, the C-28 amine
derivatives exhibited further improvements in HIV-1 inhibitory activity toward polymorphisms in the
Gag polyprotein as well as improved activity in the presence of human serum. However, plasma exposure
of basic amines following oral administration to rats was generally low, leading to a focus on moderating
the basicity of the amine moiety distal from the triterpene core. The thiomorpholine dioxide (TMD) 20
emerged from this study as a compound with the optimal antiviral activity and an acceptable pharma-
cokinetic profile in the C-28 amine series. Compared to the C-28 amide 3, 20 offers a 2- to 4-fold improve-
ment in potency towards the screening viruses, exhibits low shifts in the EC50 values toward the V370A
Received 11 January 2018
Revised 17 March 2018
Accepted 23 March 2018
Available online xxxx
Keywords:
HIV-1
HIV-1 maturation inhibitor
C-28 amine
and
DV370 viruses in the presence of human serum or human serum albumin, and demonstrates
improved potency towards the polymorphic T371A and V362I virus variants.
Ó 2018 Elsevier Ltd. All rights reserved.
The first effective therapy for the treatment of human immun-
odeficiency virus type 1 (HIV-1) infection was the nucleoside
reverse transcriptase inhibitor (NRTI) zidovudine (AZT) approved
in the US by the FDA in 1987.1 Since then, more than 20 drugs that
are used in 14 combinations for the treatment of HIV-1 infection
have been approved that can be grouped into the following six
mechanistic classes: NRTIs, non-nucleoside reverse transcriptase
inhibitors (NNRTIs), protease inhibitors (PIs), integrase strand
transfer inhibitors (INSTIs), fusion inhibitors (FIs) and chemokine
receptor 5 antagonists (CCR5 antagonists). Each class of antiviral
agent targets a unique step along the HIV-1 viral life cycle.2 The
current standard of care for HIV/AIDS patients is to combine mul-
tiple drugs from at least two different classes of these antiretroviral
agents to suppress virus replication, a regimen termed combina-
tion antiretroviral therapy (cART).3 While most of the regimens
consist of three or more drugs, JulucaÒ, the first complete treat-
ment regimen containing only two drugs (the INSTI dolutegravir
and the NNRTI rilpirivine) was approved by FDA in November
2017 to treat certain adults with HIV-1 infection.3 This approval
marks the start of a new era in HIV treatment. Since the advent
of potent combination therapy in 1995, cART has dramatically
reduced HIV-1-associated morbidity and mortality by effectively
suppressing viral replication, thus transforming HIV-1 infection
from a life-threatening disease into a manageable, chronic condi-
tion.4,5 However, despite the success of cART in recent years,
HIV-1 continues to infect millions of people and remains one of
the leading causes of death worldwide.6 As a lifetime therapy, cART
is associated with a variety of issues related to long term drug tox-
icity, intolerability, barriers to adherence, unfavorable drug-drug
⇑
Corresponding authors.
E-mail address: yan.chen@bms.com (Y. Chen).
Current address: ViiV Healthcare, 36 East Industrial Road, Branford, CT 06405,
d
USA.
e
Deceased April 8th, 2016.
https://doi.org/10.1016/j.bmcl.2018.03.067
0960-894X/Ó 2018 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Chen Y., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.067

2
Y. Chen et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
interactions and the eventual development of drug resistance. In
addition, cross-resistance among existing antiretroviral agents also
limits future treatment options, circumstances that collectively set
the stage for the development of novel antiretroviral therapies that
are endowed with distinct mechanisms in the HIV-1 viral life cycle.
Maturation inhibitors (MIs) represent a class of antiretroviral
agents with a unique mechanism of action that act by binding to
a region near the CA-SP1 segment of the Gag polyprotein and inter-
fering with access of HIV-1 protease to one specific cleavage site
within the Gag polyprotein.7,8 Only the final and rate-limiting
cleavage of the Gag protein between capsid p24 and spacer peptide
1 (SP1) into its structural components is blocked by MIs, leading to
the production of immature and non-infectious viral particles. 3-O-
(3030-Dimethylsuccinyl) betulinic acid (1, bevirimat, BVM, Fig. 1)
represents the first-in-class HIV-1 MI that showed clinical efficacy
and an acceptable safety profile in HIV-1 infected patients.9–15
However, development of 1 was halted in 2010 after reaching
Phase IIb clinical trials due to the finding that antiviral activity
was sensitive to the presence of baseline polymorphisms at resi-
dues 369–371 within the CA-SP1 region of the Gag protein, the
so-called glutamine-valine-threonine (QVT) motif, that compro-
mised efficacy in ꢀ50% of the patients in the study.16,17 Residue
362 was subsequently shown to be important in the context of nat-
urally-occurring reduced susceptibility to 1.18 In addition, 1 suf-
fered from high human serum binding and was associated with
formulation challenges, observations that have stimulated efforts
to identify MIs with improved profiles.19,20
was accomplished by reducing the C28 amide moiety to the corre-
sponding amine and resulted in the series of analogues 7–35 com-
piled in Table 1.
The synthetic approach to the majority of the C28 amine ana-
logues is described in Scheme 1. This three step procedure began
with oxidation of 4 using PCC to afford aldehyde 5 in good yield,
which was combined with a series of amines in a reductive amina-
tion process to give 6.21a Unmasking of the tert-butyl ester of 6 or 60
with CF3CO2H (TFA) in CH2Cl2 at room temperature followed by
HPLC purification afforded the target compounds 7–28 and 30–
34 which were isolated as their TFA salts.
Following the same in vitro screening paradigm as described for
the C28 amide series, novel C28 amine targets were screened for
antiviral activity toward WT HIV-1 and two of the polymorphic
viruses, V370A and
DV370, that are known to reduce sensitivity
to 1.16–18 The cytotoxicity of all of the compounds toward the host
cell (CC50 value) was also determined while select compounds
were rescreened in the WT HIV-1 cell culture assay in the presence
of human serum albumin (HSA), data that are presented in Table
1.21b Representative compounds were evaluated for their in vivo
pharmacokinetic profile either in an abbreviated, 6 h snapshot rat
PK screen, which was used as an expedited assay, or in a full 24
h rat PK study. The compounds were dosed at 5 mg/kg PO using
a mixture of poly(ethylene glycol) 300 (PEG 300), ethanol and
TW80 (tween 80, 89:10:1 v/v) as the vehicle unless otherwise
noted, with the results presented in Table 1.
As shown in Table 1, compound 7, the C-28 amine derived by
reduction of amide 3, exhibited comparable potency towards WT
We have previously reported on the discovery of a C3 benzoic
acid moiety that acts as an advantageous surrogate for the
dimethyl succinate ester installed at the C3 position of 1.21a
Compound 2 (Fig. 1) demonstrates potency against wild-type
(WT) HIV-1 that is comparable to 1 and is less affected by the
presence of human serum, with a reduction in potency of only
10-fold compared to the nearly 100-fold change observed with 1.
and the two polymorphic V370A and
DV370 viruses when com-
pared with 3, and with a similar shift (ꢀ4-fold) in the presence
of HSA. Homologation of the ethylene linker (8) resulted in an
order of magnitude decline in potency towards all 3 viruses while
the bulkier aliphatic dialkyl amines 9 and 10 retained the antiviral
potency of 7 as did the primary amine 11. The piperidine deriva-
tives 12 and 13 exhibited disparate potency, with SARs that were
the inverse of the acyclic analogues 7 and 8, with the homologated
13 more potent than 12. For the piperazines 14–16, the propyl lin-
ker in 16 offered superior antiviral properties compared to the
ethylene linker in 15 while the effect of dealkylation of the N atom
was not significant (14). An in vivo rat PK study with 7 revealed
that the plasma exposure (AUC) of the compound after oral dosing
was 33% of that for the C-28 amide analogue 3 and 10-fold less
than for 1 while plasma exposure for 8 was considerably lower.
These results were broadly consistent with previous observations
that had revealed that compounds incorporating basic side chains
were generally associated with favorable antiviral properties while
polar or negatively charged side chains conferred improved in vivo
exposure.21b Thus, the belief was that a delicate balance of
physicochemical properties would be required in order to
identify a suitable compromise. Adding further to the burden
associated with basic diamine-containing C-28 substituents was
the specter of cytotoxicity that emerged with 11, 13, 15 and 16
which was more pronounced than for compounds 7–10 and,
Efforts to broaden the polymorphic coverage of
2 through
modifications at the C28 position of the triterpenoid core were
successful when it was discovered that the basic amine-
containing C28 carboxylamide 3 (Fig. 1) demonstrated promising
levels of antiviral activity toward the major polymorphic viruses,
including V370A,
DV370 and some of the more recalcitrant
viruses that were not inhibited by first generation MIs.21b,22
Moreover, the potency of 3 was reduced by only a modest 4-fold
in the presence of human serum.21b Against this backdrop,
attention was focused on further exploring the role that the basic
amine moiety installed at C28 played in broadening MI activity
toward polymorphisms in the Gag polyprotein.
In this iteration of compound optimization, attention was
focused on modifications directed towards modulating the basicity
and polarity of the C28 sidechain, with the objective of preserving
the overall virology profile of 3 while improving the pharmacoki-
netic (PK) characteristics. One strategy examined was the incorpo-
ration of
a second basic element into the molecule while
maintaining the overall shape close to that of prototype 3. This
although not
a
consistent observation, was of sufficient
persistence to be of concern.23 Taken together, these observations
led to a focus on moderating the basicity of the amine distal from
the core by incorporating this element into ring systems containing
electron withdrawing functionality (17–23) or by introducing or
integrating into substituents that would either attenuate (24–31)
or completely quench (32–34) the basicity.24
The two morpholine derivatives 17 and 18 (pKa = 8.2) demon-
strated potent antiviral activity toward all 3 viruses that, in the
case of 18 was not associated with a significant serum shift; how-
ever, this compound demonstrated a low plasma AUC over 6 h in
rats following oral administration. Replacing the morpholine rings
of 17 and 18 with a thiomorpholine-1,1-dioxide (TMD) heterocy-
Fig. 1. Structures of HIV-1 maturation inhibitors.
Please cite this article in press as: Chen Y., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.067

Y. Chen et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
3
Table 1
Antiviral activity towards wild type HIV-1, V370A and
D
V370 viruses, serum shift, fold change and in vivo pharmacokinetic properties of C28 amine derivatives.
#
R
EC50 (nM)a
WT
CC50 (mM)
AUC-6/24 h
(nMꢁh)c
WT + 15 mg/ml
HSAb
Fold
Change
V370A
DV370
1
See Fig. 1
See Fig. 1
See Fig. 1
10 11
16 13
2.3 1.3
2.2 0.8
970 710
150 99
97
9
4
552 633
233 305
7.8 8.8
2.5 2.2
1900
17 14.5
27 38.5
6390/13,763
5724/NAg
1623/3814
NA/1285
2
>6000
31 19
21 14
3d
7
11
9
3
1.8
8.7
4
1.2 0.2
8e
27 22
160
NA
6
79 104
200
35
4.5
20
10/NA
NA
9
4.3
NA
1.6
10
6.7
NA
NA
4.4
35
3.7
NA
11
12
2.0
51
NA
NA
4
3.6
40
37
0.12
4.0
NA
NA
210
390
13
1.4
NA
NA
2.4
6.4
0.24
NA
14
15
5.0
7.7
26
23
5
3
44
29
NA
14
2.3
NA
NA
0.97
(continued on next page)
Please cite this article in press as: Chen Y., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.067

4
Y. Chen et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
Table 1 (continued)
#
R
EC50 (nM)a
CC50 (mM)
AUC-6/24 h
(nMꢁh)c
WT
WT + 15 mg/ml
HSAb
Fold
Change
V370A
0.4
DV370
16
0.7
NA
NA
16
0.7
NA
17
1.5
NA
NA
2.0
15
5.0
NA
18
19
2.8 2.9
8.2
NA
3
3.8 3.5
28
2.2
2.6
45/NA
0.7
NA
1.3
6
2049/NA
20
1.0 0.4
2.0
NA
2
2.5 1.4
14 8.4
3.5 1.6
1708/2815
4834/9062
21f
2.0 0.7
NA
6.0 1.8
26 13
10 1.5
22
23
2.0
NA
NA
NA
NA
5.0
11
43
57
7
NA
NA
6
8.5
24
25
4.7
2.7
NA
NA
NA
NA
2.9
13
60
55
11
962/NA
NA
4.6
26
3.3
NA
NA
0.42
20
2.0
NA
Please cite this article in press as: Chen Y., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.067

Y. Chen et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
5
Table 1 (continued)
#
R
EC50 (nM)a
CC50 (mM)
0.32
AUC-6/24 h
(nMꢁh)c
WT
WT + 15 mg/ml
HSAb
Fold
Change
V370A
6.7
DV370
27
28
2.7
NA
NA
18
NA
NA
0.34
NA
NA
1.0
8.4
6.8
29
30
31
0.78
2.5
16
3.8
3.2
8
5
8.8
4.0
19
27
16
12
7.4
540/NA
223/NA
NA
1.3
0.5
9.3
NA
32
33
1.7 0.9
4.2 3.0
2.5
3
3.6 3.5
1.7 0.7
37 25
6.3 2.0
5.8 0.8
NA
0.68 0.2
2.0
9.9 8.3
1044/NA
34
35
1.4
14
NA
21
NA
1.5
3.4
62
16
2.4
30
288/NA
650
4387/NA
a
The antiviral activity of test compounds was assessed in a multiple cycle assay in MT-2 cells in the presence of 10% fetal bovine serum (FBS) (see Supplementary
Information (SI)). The data provided are the means of at least triplicate values (n ꢃ 3) with a mean CV (confidence value) of 83 50% for all of the experiments, and all CV
values are <200%. Where experiments were performed 4 or more times, the corresponding standard deviations are noted.
b
Compounds were evaluated for the effects of serum in 10% FBS + 15 mg/mL HSA; values are the means from experiments performed a minimum of three times (n ꢃ 3).
c
Dose = 5 mg/kg po, n = 2; see SI for vehicle information. Vehicle:
d
78% PEG300; 10% 0.1 N NaOH; 10% EtOH; 2% TPGS.
4.4% 0.1 N HCl; 9.6% EtOH; 85.5% PEG 300; 0.5% TW.
5% 0.1 N NaOH; 10% EtOH; 84.5% PEG 300; 0.5% TW.
NA: not available.
e
f
g
cle, which is less basic (pKa = 4.5) but more polar (cLogP of N-CH3-
TMD = ꢂ0.80 compared to 0.17 for N-methylmorpholine) than
morpholine, afforded 19 and 20. This structural modification not
only maintained the good antiviral profile associated with the
more basic C-28 amines but also retained activity in the presence
of human serum, with just a 2-fold shift in the EC50 value for 20.
More importantly, the TMD moiety conferred
a significant
improvement in plasma exposure following oral dosing to rats,
Please cite this article in press as: Chen Y., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.067

6
Y. Chen et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
H
H
O
OH
H
H
H
H
a
H
H
O
O
4
5
O
O
b
H
N
H
N
H
H
R
R
H
H
H
H
c
H
H
O
HO2C
7 - 20, 23-28, 30 - 34
6
O
d
R'
N
H
R'
N
H
R
H
R
H
c
H
H
H
H
HO
O
21, 22
6'
O
O
Scheme 1. Synthesis of C28 amine derivatives 7–28, 30–34 from the C3 benzoic acid-modified triterpenoid 4. Reagents and conditions: (a) PCC, CH2Cl2, RT., 6.5 h, 80%; (b)
NaBH(OAc)3, AcOH, amine reactant, CH2Cl2, RT, overnight; (c) CF3CO2H, CH2Cl2, 6–16 h; (d) NaBH(OAc)3, AcOH, aldehyde or ketone reactants, CH2Cl2, RT, overnight.
with the 24 h AUC values for 20 2-fold higher than the AUC of 7
measured over the same time period in the rat while the 6 h AUC
value for 19 was higher than the 24 h AUC value measured for 3.
The plasma AUC for 20 was comparable to that of 3 measured over
both the 6 and 24 h time frames, while the concentration of drug in
plasma 24 h post-dose (C24h = 34 nM) was improved compared to 1
(C24h = 6 nM) and similar to compound 3 (C24h = 27 nM). The TMD
moiety, in addition to modulating the basicity of the side chain,
might be viewed in this context as functioning as a non-charged
analogue of a carboxylic acid moiety, leading to enhanced PK prop-
erties.25 It was observed in the C28 amides series that the combi-
Fig. 2. Structure of GSK3532795, formerly known as BMS-955176 (36).
nation of a basic amine with a carboxylic acid side chain could
the rat was modest over the 6 h AUC that was measured. Installing
hydroxyl substituents at the terminus was associated with strong
antiviral properties, compounds 25–28, although sub-micromolar
half-maximal cytotoxicity was observed for 27. The sarcosine ana-
logue 29 exhibited good antiviral properties but was poorly orally
bioavailable. Of the 2 heterocycles examined, imidazole 30 was a
superior HIV-1 inhibitor compared to pyridine 31 but 30 gave only
a low plasma AUC following oral dosing to rats. The amides 32 and
33 and the aryl sulfonamide 34 exhibited potent virus inhibition
but 33 demonstrated only modest plasma exposure after oral
administration to rats while hybridizing a basic amine with a car-
boxylic acid moiety (35) resulted in a significant erosion of antivi-
ral activity, particularly toward the polymorphic viruses. However,
35 demonstrated plasma exposure in rats over 6 h that was more
than 2-fold higher than for prototype 3, recapitulating observa-
tions in the amide series where a carboxylic acid moiety was often
associated with enhanced plasma exposure following oral
administration.
lead to significantly improved oral exposure although, in that case,
the antiviral coverage was negatively affected.21b
Shielding of the C-28 amine proximal to the triterpenoid skele-
ton by alkylation with methyl (21) or a basicity-moderating cyclo-
propyl (22) moiety afforded tertiary amines that resulted in a 2-
fold reduction in antiviral potency toward the WT as well as the
polymorphic viruses when compared with 20.26a,b However, oral
dosing of 21 to rats was associated with
a noticeable
improvement in the rat plasma AUC measured over 6 and 24 h
(both about 3-fold) compared to 20, demonstrating design
principles that would be of importance in the identification of
GSK3532795, formerly know as BMS-955176 (36, Fig. 2).27 The
final example in this series incorporates a 4,4-difluoropiperidine
heterocycle (23), which might be considered as a less hydrophilic
(cLogP of 4,4-difluoro-1-methylpiperidine = 0.77) but more basic
(pKa = 8.6) bioisostere of the morpholine and sulfone moieties, that
resulted in an overall loss of potency of ꢀ5- to 10-fold compared
with 19.26c
The TMD analogue 20 emerged as the compound with the best
combination of antiviral and pharmacokinetic profiles in the C-28
amine series representing a 2- to 4-fold improvement in potency
The weakly basic N-phenylamine 24 retained good potency
toward WT virus and the V370A polymorph but was less effective
toward the
DV370 polymorphic variant, while plasma exposure in
Please cite this article in press as: Chen Y., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.067

Y. Chen et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
7
Table 2
Polymorphic coverage of TMD analogue 20, C-28 amide 3 and 1 (BVM).
EC50 (nM)a
WT
T371A
V362I
WT + 15 mg/mL
HSAc
Fold
Change
WT + HS (40%)
Fold
Change
V370A + HS
Fold Change
D
V370 + 15 mg/ml
FC
(40%) + HSAb
HSAc
1
3
20
10 11
2.3 1.3
10.4
3.0
1.7
74 59
23 24
0.94 0.4
970 710
97
4
2
1120 1000
42 29
7
112
14
7
dNA
NA
15
NA
NA
7
NA
201
20
NA
7
1.4
11
2
9
1
0.4
6
a
b
c
Values are the means from experiments performed a minimum of three times (n ꢃ 3).
Compounds were evaluated for serum effects in 10% FBS + 40% HS + 27 mg/ml HSA.
Compounds were evaluated for serum effects in 10% FBS + 15 mg/ml HAS.
NA: not available.
d
towards the screening viruses when compared to amide 3; this
improvement was 10- to 135-fold when compared to 1, with
almost a 50-fold improved potency in the presence of human
serum albumin compared to 1. To further understand the potential
The in vitro data generated for 20 qualified it as a compound
with the profile sought in a second generation MI and this com-
pound was advanced into a dose escalation study in rats in order
to more fully assess its potential for further development. Oral
administration of 20 to SD rats at doses of 10, 30 and 100 mg/kg
formulated as a solution in 20% sulfobutylether cyclodextrin and
80% 50 mM citrate buffer (pH 3.0) resulted in a dose-dependent
but less than dose-proportional increase in the plasma AUC
between 5 and 100 mg/kg (AUCnorm from 650 to 291 nM*h/dose,
respectively) (Fig. 3, Table 3). The C24h exposure was approxi-
mately dose-proportional from 5 to 100 mg/kg, with very little
inter-animal variability in exposure observed at each dose. All of
the doses were well-tolerated, confirming the TMD in 20 as an
effective handle with which to provide both the antiviral charac-
teristics and plasma exposure following oral dosing for this type
of MI. The improvement in overall PK properties of 20, including
plasma exposure after oral administration as well as the ability
to dose escalate, when compared to 3 were very encouraging.
However, while the antiviral profile of 20 was close to that targeted
(EC50 values of <10 nM toward WT and key polymorphic viruses),
the PK profile in rats left room for further improvement. Neverthe-
less, the two key learnings from this study that subtended the sub-
sequent discovery of 36 were the antiviral and PK properties
conferred by the TMD heterocycle, which appears to act somewhat
analogously to a carboxylic acid moiety but maintaining charge
neutrality in that part of the molecule, and the benefit on oral
exposure of shielding the amine proximal to the betulin-based
core.
of 20, this compound was evaluated for V370A and
DV370 inhibi-
tion in the presence of human serum, and for activity toward two
additional polymorphic virus variants: T371A and V362I both of
which exhibit resistance to 1. The results are summarized in Table 2
and the data indicate that 20 exhibits low fold-shifts in the EC50
values toward V370A and
DV370 in the presence of human serum
or human serum albumin and greatly improved potency towards
the polymorphic T371A and V362I viruses. However, virus harbor-
ing the A364V mutation that emerges in response to selective pres-
sure with 1, remained insensitive to 20 (EC50 >4 mM).28
1000
100
Compound 20 Rat- 5 mg/kg PO
10
Compound 20 Rat- 10 mg/kg PO
Compound 20 Rat- 30 mg/kg PO
Compound 20 Rat- 100 mg/kg PO
In summary, the TMD construct 20 with its moderated basicity
installed distal to the C28 nitrogen of this triterpene core improved
in vivo exposure in rats compared to more basic analogues, as well
as the overall virology profile versus WT virus and some of the
most prevalent HIV-1 Gag polymorphisms. In addition, compounds
containing a TMD displayed only a modest 2-fold reduction of
antiviral potency in cell culture in the presence of human serum.
All of these observations suggest that the TMD moiety has the
characteristics of an optimal side chain and it was subsequently
1
0
8
16
24
32
40
48
Time (h)
Fig. 3. Exposure of 20 in rats following doses of 5, 10, 30 and 100 mpk formulated
as a solution in 20% sulfobutylether cyclodextrin and 80% 50 mM citrate buffer (pH
3.0).
Table 3
Pharmacokinetic parameters for 20 in the rat in a dose escalation study following oral administration.
Dose (mg)
Cmax (nM)
Tmax (hours)
C24h (nM)
11.8 (soln)a
285
4.7
91
35.0 (soln)b
451
5.3
124 (soln)c
697
17.3
635
5 (soln)
232
6.7
234
34
C24h
AUClast
AUCtot nM*h
AUCnorm nM*h/dose
%AUCextra
(nM/dose)
7.7
5307
5441
461
2
6.7
5.1
6.8
norm
11,360
13,225
378
23,222
29,113
235
2991
3252
650
8
14
17
% Bioavailability
8
7
4
12
a
b
c
Weight of 20 in the HCl salt form corresponding to 10 mg of the free form of the drug.
Weight of 20 in the HCl salt form corresponding to 30 mg of the free form of the drug.
Weight of 20 in the HCl salt form corresponding to 100 mg of the free form of the drug.
Please cite this article in press as: Chen Y., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.067

8
Y. Chen et al. / Bioorganic & Medicinal Chemistry Letters xxx (2018) xxx–xxx
15. Martin DE, Blum R, Wilton J, et al. Safety and pharmacokinetics of bevirimat
(PA-457), a novel inhibitor of human immunodeficiency virus maturation, in
healthy volunteers. Antimicrob Agents Chemother. 2007;51:3063–3066.
16. McCallister S, Lalezari J, Richmond G, et al. HIV-1 Gag polymorphisms
determine treatment response to bevirimat (PA-457). Antiviral Ther. 2008;13:
A10.
combined with a C-17 amine modification to give 36, a compound
that exhibited superior solubility, and PK characteristics and was
advanced into clinical studies as a second generation HIV-1 matu-
ration inhibitor.27–29 Even though the first TMD containing MI
compound 20 was not the ideal candidate to advance for further
development, the value of the thiomorpholine dioxide was persis-
tent as a PK- and activity-enhancing element and it has continued
to be a key pharmacophoric element in the SAR studies for the ben-
zoic acid series of betulinic acid derivatives.
17. Baelen KV, Salzwedel K, Rondelez E, et al. Susceptibility of human
immunodeficiency virus type
1 to the maturation inhibitor bevirimat is
modulated by baseline polymorphisms in Gag spacer peptide 1. Antimicrob
Agents Chemother. 2009;53:2185–2188.
18. Margot NA, Gibbs CS, Miller MD. Phenotypic susceptibility to bevirimat in
isolates from HIV-1-infected patients without prior exposure to bevirimat.
Antimicrob Agents Chemother. 2010;54:2345–2353.
19. (a) Qian K, Bori DI, Chen CH, Huang L, Lee KH. Anti-AIDS agents 90. Novel C-28
modified bevirimat analogues as potent HIV maturation inhibitors. J Med Chem.
2012;55:8128–8136;
Acknowledgments
The authors gratefully acknowledge the bioanalytical group at
Bristol-Myers Squibb, Dr. Stella Huang for spectroscopic analysis
and Chiradeep Panja, Richard A. Rampulla, Vengateshwaran S and
the Department of Discovery Synthesis at Bristol-Myers Squibb-
Biocon Research Center (BBRC) for scale up of intermediates.
(b) Urano E, Ablan S, Mandt R, et al. Alkyl amine bevirimat derivatives are
potent and broadly active HIV-1 maturation inhibitors. Antimicrob Agents
Chemother. 2016;60:190–197.
20. Coric P, Turcaud S, Souquet F, et al. 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.
21. (a) Liu Z, Swidorski JJ, Nowicka-Sans B, et al. C-3 benzoic acid derivatives of C-3
deoxybetulinic acid and deoxybetulin as HIV-1 maturation inhibitors. Bioorg
Med Chem. 2016;24:1757–1770;
A. Supplementary data
(b) Swidorski JJ, Liu Z, Sit SY, et al. Inhibitors of HIV-1 maturation: development
of structure-activity relationship for C-28 amides based on C-3 benzoic acid-
modified triterpenoids. Bioorg Med Chem Lett. 2016;26:1925.
22. Lin Z, Cantone J, Lu H, et al. Mechanistic studies and modeling reveal the origin
of differential inhibition of Gag polymorphic viruses by HIV-1 maturation
inhibitors. PLoS Pathog. 2016;12:e1005990. https://doi.org/10.1371/journal.
ppat.1005990.
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.bmcl.2018.03.067.
References
23. (a) Peters J-U, Schnider P, Mattei P, Kansy M. Pharmacological promiscuity:
dependency on compound properties and target specificity in a set of recent
Roche compounds. ChemMedChem. 2009;4:680;
1. (a) U.S Approves Drug to Prolong Lives of AIDS Patients. New York Times, 1987-
03-21; (b) Lichtenberg F, The Effect of New Drugs on Mortality from Rare
Disease and HIV. NBER Working Paper 8677, December 2001.
2. http://emedicine.medscape.com/article/1533218-overview.
3. (a) Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and
Adolescents. https://aidsinfo.nih.gov/guidelines/html/1/adult-and-adolescent-
arv/0; (b) https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/
ucm586305.htm.
4. (a) Palella FJ, Delaney KM, Moorman AC, et al. Declining morbidity and
mortality among patients with advanced human immunodefficiency virus
infections. N Engl J Med. 1998;338:853–860(b) http://www.ncbi.nlm.nih.gov/
pubmed/19770621.
(b) Kresja CM, Horvath D, Rogalski SL, Penzotti JE, Mao B. Predicting ADME
properties and side effects: the BioPrint approach. Curr Opin Drug Discov Dev.
2003;6:470.
24. Martin RE, Plancq B, Gavelle O, et al. Remote modulation of amine basicity by a
phenylsulfone and a phenylthio group. ChemMedChem. 2007;2:285–287.
25. (a) Kleeb S, Pang L, Mayer K, et al. FimH antagonists: bioisosteres to improve
the in vitro and in vivo PK/PD profile. J Med Chem. 2015;58:2221–2239;
(b) Meanwell NA. Synopsis of some recent tactical application of bioisosteres in
drug design. J Med Chem. 2011;54:2529–2591;
(c) Lassalas P, Oukoloff K, Makani V, et al. Evaluation of oxetan-3-ol, thietan-3-
ol, and derivatives thereof as bioisosteres of the carboxylic acid functional
group. ACS Med Chem Lett. 2017;8:864–868;
(d) Murafuji H, Sakai H, Goto M, Imajo S, Sugawara H, Muto T. Discovery and
structure-activity relationship study of 1,3,6-trisubstituted 1,4-diazepane-7-
5. Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early
antiretroviral therapy. N Engl J Med. 2011;365:493–505.
6. Fact Sheet – Latest Global and Regional Statistics on the Status of the AIDS
Empidemic
http://www.unaids.org/sites/default/files/media_asset/UNAIDS_
FactSheet_en.pdf.
ones as novel human kallikrein
2017;27:5272–5276.
7 inhibitors. Bioorg Med Chem Lett.
7. Sundquist WI, Kräusslich HG. HIV-1 assembly, budding, and maturation. Cold
Spring Harb Perspect Med. 2012;7:a006924.
8. Freed EO. HIV-1 assembly, release and maturation. Nat Rev Microbiol.
2015;13:484–496.
26. (a) Perrin CL, Fabian MA, Rivero IA. Basicities of cycloalkylamines: Baeyer strain
theory revisited. Tetrahedron. 1999;55:5773–5780;
(b) Christensen JJ, Izatt RM, Wrathall DP, Hansen LD. Thermodynamics of
9. Martin DE, Blum R, Doto J, Galbraith H, Ballow C. Multiple-dose
proton ionization in dilute aqueous solution. Part X1. pK,
D
H°, and
DS° values
pharmacokinetics and safety of bevirimat,
a novel inhibitor of HIV
for proton ionization from protonated amines at 25 °C.
J
Chem Soc A.
maturation, in healthy volunteers. Clin Pharmacokinet. 2007;46:589–598.
10. Kanamoto T, Kashiwada Y, Kanbara K, et al. Anti-human immunodeficiency
virus activity of YK-FH312 (a betulinic acid derivative), a novel compound
blocking viral maturation. Antimicrob Agents Chemother. 2001;45:1225–1230.
11. Qian K, Nitz TJ, Yu D, Allaway GP, Morris-Natschke SL, Lee KH. Anti-AIDS agents
76. From natural product to clinical trials: bevirimat, a plant-derived anti-AIDS
drug. In: Buss T, Butler M, eds. Natural Product Chemistry for Drug Discovery.
Royal Society of Chemistry Biomolecular Sciences Series. Singapore: Merlion
Pharmaceuticals Pte Ltd Press; 2009:374–391.
12. Smith PF, Ogundele A, Forrest A, et al. Phase I and II study of the safety, virologic
effect, and pharmacokinetics/pharmacodynamics of single-dose 3-O-(30,30-
dimethylsuccinyl)betulinic acid (bevirimat) against human immunodeficiency
virus infection. Antimicrob Agents Chemother. 2007;51:3574–3581.
1969;1212–1223;
(c) Krabbe SW, Chan VS, Franczyk TS, et al. Copper-catalyzed aerobic oxidative
amidation of benzyl alcohols. J Org Chem. 2016;81:10688–10697.
27. Regueiro-Ren A, Liu Z, Chen Y, et al. Discovery of BMS-955176, a second
generation HIV-1 maturation inhibitor with broad spectrum antiviral activity.
ACS Med Chem Lett. 2016;7:568–572.
28. Nowicka-Sans B, Protack T, Lin Z, et al. Identification and characterization of
BMS-955176, a second-generation HIV-1 maturation inhibitor with improved
potency, antiviral spectrum, and gag polymorphic coverage. Antimicrob Agents
Chemother. 2016;60:3956–3969.
29. Hwang C, Schürmann D, Sobotha C, et al. Antiviral activity, safety, and
exposure-response relationships of GSK3532795, a second-generation human
immunodeficiency virus type
1 maturation inhibitor, administered as
13. Li F, Goila-Gaur R, Salzwedel K, et al. PA-457, a potent HIV inhibitor that
disrupts core condensation by targeting a late step in Gag processing. Proc Natl
Acad Sci USA. 2003;100:13555–13560.
monotherapy or in combination with atazanavir with or without ritonavir in
a phase 2a randomized, dose-ranging, controlled trial (AI468002). Clin Infect
Dis. 2017;65:442–452.
14. http://investor.myriad.com/releasedetail.cfm?ReleaseID=360258.
Please cite this article in press as: Chen Y., et al. Bioorg. Med. Chem. Lett. (2018), https://doi.org/10.1016/j.bmcl.2018.03.067