Synthesis of the biotinylated anti-HIV compound BMMP and the target identification study

Article abstract of DOI:10.1016/j.bmcl.2015.11.036

BMMP [2-(benzothiazol-2-ylmethylthio)-4-methylpyrimidine], an inhibitor of HIV-1 replication, was linked to biotin to study the interaction with the presumed target, HIV-1 Pr55^Gag or CA, by means of surface plasmon resonance. The synthesized Bi

Full text of DOI:10.1016/j.bmcl.2015.11.036

Bioorganic & Medicinal Chemistry Letters 26 (2016) 43–45
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of the biotinylated anti-HIV compound BMMP
and the target identification study
b,
Masahiro Kamo a, Hiroshi Tateishi a, Ryoko Koga a, Yoshinari Okamoto a, Masami Otsuka a, , Mikako Fujita
⇑
⇑
a Department of Bioorganic Medicinal Chemistry, Faculty of Life Sciences, Kumamoto University, 5-1 Oe-honmachi, Chuo-ku, Kumamoto 862-0973, Japan
b Research Institute for Drug Discovery, School of Pharmacy, Kumamoto University, 5-1 Oe-honmachi, Chuo-ku, Kumamoto 862-0973, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
BMMP [2-(benzothiazol-2-ylmethylthio)-4-methylpyrimidine], an inhibitor of HIV-1 replication, was
linked to biotin to study the interaction with the presumed target, HIV-1 Pr55Gag or CA, by means of sur-
face plasmon resonance. The synthesized Biotin–BMMP inhibited HIV-1 replication to a similar extent as
BMMP alone, but did not interact with Pr55Gag or CA, suggesting that BMMP exerts its activity by a dif-
ferent mechanism.
Received 6 October 2015
Revised 8 November 2015
Accepted 12 November 2015
Available online 12 November 2015
Ó 2015 Elsevier Ltd. All rights reserved.
Keywords:
BMMP
Biotin
HIV-1
Pr55Gag
CA
Acquired immunodeficiency syndrome (AIDS), caused by
human immunodeficiency virus type 1 (HIV-1), was previously
fatal in all patients. Since the development of anti-retroviral ther-
apy (ART),1 the use of multiple drugs targeting the viral reverse
transcriptase, protease and integrase as well as viral entry has
allowed patients to survive the disease. However, the disadvan-
tages of ART therapy include drug resistance and other side effects.
Thus, efforts to develop new anti-HIV drugs acting through novel
mechanisms are ongoing.
One attractive approach is to treat HIV to block the function of
the HIV-1 protein Gag.2 When it is present in a cell which is pro-
ducing virus particles, Gag mainly exists as the polyprotein precur-
sor Pr55Gag. Concurrently or immediately after budding of the
virion, the HIV-1 protease cleaves Pr55Gag into four main proteins
(matrix: MA, capsid: CA, nucleocapsid: NC and p6) and two spacer
peptides (p2 and p1), to produce the mature virion capable of tar-
get cell infection. In both producer and target cells, immature and
mature Gag proteins play various central roles.3
methylpyrimidine (BMMP) (Fig. 1) as an inhibitor of HIV-1 replica-
tion by an unknown mechanism. Contrary to the generally
accepted prediction of its mechanism of action, BMMP did not inhi-
bit the virion release stage.4 Instead, BMMP suppressed the reverse
transcription of the virus genome in target cells, and the protein
responsible for interacting with BMMP was shown to be CA.4 This
suggests the possibility that BMMP directly binds to the CA domain
of Pr55Gag or the CA protein. To examine this hypothesis, we syn-
thesized biotinylated BMMP (Biotin–BMMP), confirmed that Bio-
tin–BMMP retains the activity of BMMP alone, and examined its
interaction with Pr55Gag or CA using biotin–avidin technology.
Biotin–BMMP (Fig. 1) carries a biotin on the pyrimidyl–methyl
group of BMMP through a methylene linker. Synthesis of Biotin–
BMMP began with the construction of the mercaptopyrimidine-
linker fragment which was then coupled with the benzothiazole
moiety, followed by conjugation with biotin as shown in Scheme 1.
Benzothiazole-2-carboxaldehyde 1 was reduced by NaBH4 to
afford 2 quantitatively. Treatment of 2 with PBr3 gave bromide 3
in 57% yield. N-(5-Bromopentyl)phthalimide 4 was treated with
the sodium enolate of ethyl acetoacetate to furnish 5 in 79% yield.
Acid hydrolysis and decarboxylation of 5 afforded 6 in 90% yield.
Reaction of 6 and N,N-dimethylformamide dimethyl acetal in the
presence of BF3 gave 7 in 64% yield. Treatment of 7 with thiourea
and sodium ethoxide furnished the mercaptopyrimidine com-
pound 8 that was used without further purification and coupled
with the benzothiazole 3 in the presence of Et3N to afford 9 in
49% overall yield. Treatment of 9 with hydrazine gave the free
In 2011, Urano et al. explored small molecule inhibitors of the
oligomerization of HIV-1 Pr55Gag,4 a key process essential for virion
production,5 and identified 2-(benzothiazol-2-ylmethylthio)-4-
⇑
Corresponding authors. Tel.: +81 96 371 4620; fax: +81 96 371 4620 (M.O.); tel.:
+81 96 371 4622; fax: +81 96 371 4108 (M.F.).
E-mail addresses: 141y1023@st.kumamoto-u.ac.jp (M. Kamo), 155y2008@st.
kumamoto-u.ac.jp (H. Tateishi), kk1205@kumamoto-u.ac.jp (R. Koga), okamoto@
gpo.kumamoto-u.ac.jp (Y. Okamoto), motsuka@gpo.kumamoto-u.ac.jp (M. Otsuka),
mfujita@kumamoto-u.ac.jp (M. Fujita).
http://dx.doi.org/10.1016/j.bmcl.2015.11.036
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

44
M. Kamo et al. / Bioorg. Med. Chem. Lett. 26 (2016) 43–45
ker compound 10 had inhibitory activity, although the activity of
S
N
N
N
Biotin–BMMP was somewhat weaker than that of BMMP. In con-
trast, the biotin methyl ester, which represents the biotin moiety
of Biotin–BMMP, did not show any activity. This indicated that
the BMMP part of Biotin–BMMP was responsible for the inhibition
of HIV-1 replication, presumably by binding to the Pr55Gag or CA
target protein. It is likely that the 4-methyl group of the pyrimidine
ring is located in an appropriate position without affecting the
binding and thus, function, of the BMMP.
S
%003
O
S
N
NH
HN
N
N
%LRWLQꢀ%003
H
H
S
The interaction between BMMP and Pr55Gag or CA protein was
examined using Biotin–BMMP. We previously established a highly
sensitive surface plasmon resonance method7 to analyze the
hydrophilic and hydrophobic interactions between Pr55Gag/MA
and inositol phosphates conjugated to biotin. This method was
applied to the analysis of the BMMP–Pr55Gag or –CA interaction.
Biotin–BMMP in buffer was incubated with streptavidin covalently
immobilized on the surface of a BIACORE sensor chip, and conjuga-
tion of Biotin–BMMP with the sensor chip was confirmed by its
response. Then, Pr55Gag or CA protein, both prepared from 293T
cells overexpressing the protein of interest, was passed over the
sensor chip. As shown in Figure 2B, no interactions between either
of the proteins with the conjugated Biotin–BMMP were observed.
Since the interaction of Pr55Gag and phosphoinositide was
observed using the same method,7a we concluded that BMMP did
not interact with Pr55Gag or CA in solution.
There have been reports that the heterocyclic compounds PF-
3450074,8 I-XW-0539 and BI-1/210 interact with CA, and inhibit
HIV replication at the post-entry step. These compounds were
shown to directly bind to specific portions of CA. In this study,
we concluded that BMMP inhibits HIV replication through a mech-
anism which is different from that employed by the heterocyclic
compounds. As BMMP was identified as an inhibitor of HIV replica-
H
N
S
6
O
Figure 1. Structures of BMMP and Biotin–BMMP.
amine 10 in 48% yield. Finally, 10 was conjugated with biotin using
EDC to furnish Biotin–BMMP in 96% yield. The total yield of Biotin–
BMMP from 4 in 7 steps was 10%.
To confirm that Biotin–BMMP retained the biological activity of
BMMP, its inhibitory effect on HIV-1 replication was examined and
compared with BMMP, the synthetic intermediate 10, and the bio-
tin methyl ester. To obtain cell-free HIV-1 for infection, pNL4-36
coding the full length HIV-1 was transfected into 293T human
embryonic kidney cells and incubated. The supernatant at 2 days
post transfection containing virus was inoculated into the human
T-lymphoblastoid cell line M8166, and incubated for approxi-
mately 5 days in the presence of each chemical (20 lM). The virus
amounts in the supernatant were monitored by p24 (Gag CA)
ELISA, and the obtained growth curve is shown in Figure 2A.
In agreement with the previous Letter,4 BMMP suppressed HIV-
1 replication. It was shown that Biotin–BMMP and the BMMP-lin-
O
S
N
S
S
N
NaBH4
MeOH
PBr3
CH2Cl2
N
OH
Br
O
H
,
,
2h r.t.
2h r.t.
100%
57%
1
2
3
O
O
O
O
O
O
CH
3COOH, HCl,
H2O
OC2H5
Na
N
Br
N
N
5
5
reflux
EtOH,
1h,
reflux
79%
90%
O
OC2H5
6h,
O
O
O
4
5
6
S
NMe2
OMe
BF3 • Et2O
O
O
O
N
H2N
NH2
MeO
HO
NH
N
N
5
HS
N
5
Na
EtOH,
reflux
22h,
reflux
3d,
O
O
64%
7
8
S
N
N
N
S
N
O
3
H2NNH2 • H2O
EtOH
S
N
HO
S
NH
Et
EtOH
3N,
6h,
5
N
NH2
reflux
3h,
r.t.
49%
6
48%
2
steps
O
O
9
10
HN
H
NH
H
HO
S
O
EDC
Pyridine
Biotin-BMMP
r.t.
3d,
96%
Scheme 1. Synthesis of Biotin–BMMP.

M. Kamo et al. / Bioorg. Med. Chem. Lett. 26 (2016) 43–45
45
cess of determining the true target of BMMP by proteomic analysis
of viral and host factors using Biotin–BMMP.
600
500
400
300
200
100
0
ꢁ$ꢂ
Acknowledgment
PRFN
ꢀꢃ
This work was supported in part by a grant from the Ministry of
Health, Labour, and Welfare of Japan (Research on HIV/AIDS pro-
ject No. H24-005) (to M.O. and M.F.).
%003
%LRWLQꢀ%003
ꢇꢈ
+,9ꢀꢇꢃ
ELRWLQꢃPHWK\Oꢃ
HVWHU
Supplementary data
0
20
40
60
80
100
120
Supplementary data (experimental procedures and characteri-
zation of the compounds) associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.bmcl.2015.11.
036.
+RXUVꢃDIWHUꢃDGGLWLRQꢃRIꢃFKHPLFDOVꢃꢁKꢂ
ꢁ%ꢂ
100
References and notes
80
60
40
20
0
1. (a) Kuritzkes, D. R.; Koup, R. A., 6th ed. In Fields Virology; Knipe, D. M., Howley,
P. M., Cohen, J. I., Griffin, D. E., Lamb, R. A., Martin, M. A., Racaniello, V. R.,
Roizman, B., Eds.; Lippincott Williams & Wilkins: Philadelphia, PA, 2013; Vol.
2,. pp 1561–1583 (b) Looney, D.; Ma, A.; Johns, S. Curr. Top. Microbiol. Immunol.
2015, 389, 1.
2. Tedbury, P. R.; Freed, E. O. Curr. Top. Microbiol. Immunol. 2015, 389, 171.
3. Freed, E. O.; Martin, M. A., 6th ed. In Fields Virology; Knipe, D. M., Howley, P. M.,
Cohen, J. I., Griffin, D. E., Lamb, R. A., Martin, M. A., Racaniello, V. R., Roizman, B.,
Eds.; Lippincott Williams & Wilkins: Philadelphia, PA, 2013; Vol. 2,. pp 1502–
1560.
4. Urano, E.; Kuramochi, N.; Ichikawa, R.; Murayama, S. Y.; Miyauchi, K.; Tomoda,
H.; Takebe, Y.; Nermut, M.; Komano, J.; Morikawa, Y. Antimicrob. Agents
Chemother. 2011, 55, 4251.
5. (a) von Schwedler, U. K.; Stray, K. M.; Garrus, J. E.; Sundquist, W. I. J. Virol. 2003,
77, 5439; (b) Lingappa, J. R.; Reed, J. C.; Tanaka, M.; Chutiraka, K.; Robinson, B.
A. Virus Res. 2014, 193, 89; (c) Maldonado, J. O.; Martin, J. L.; Mueller, J. D.;
Zhang, W.; Mansky, L. M. Front. Microbiol. 2014, 5, 302.
PRFN
&$
3Uꢉꢉ*DJ
0
50
100
150
200
250
300
350
400
7LPHꢃꢁVꢂ
Figure 2. Biological experiments using Biotin–BMMP. (A) Growth kinetics of HIV-1
NL4-3 strain in M8166 cells in the presence of Biotin–BMMP and its related
chemicals (20 l
M). The supernatant of 293T cells transfected with pNL4-36 was
6. Adachi, A.; Gendelman, H. E.; Koenig, S.; Folks, T.; Willey, R.; Rabson, A.; Martin,
M. A. J. Virol. 1986, 59, 284.
inoculated into M8166 cells. After 16 h, chemicals were added and cultured. At the
selected time points, the virus amounts in the supernatant were examined by p24
ELISA. As a control, the supernatant of 293T cells transfected with pUC19 was used
instead of the virus solution. (B) The binding activity of CA or Pr55Gag protein to
Biotin–BMMP. The proteins were prepared as previously described.7a Then each
protein was incubated on a Biotin–BMMP immobilized BIACORE sensor chip.
7. (a) Anraku, K.; Fukuda, R.; Takamune, N.; Misumi, S.; Okamoto, Y.; Otsuka, M.;
Fujita, M. Biochemistry 2010, 49, 5109; (b) Tateishi, H.; Anraku, K.; Koga, R.;
Okamoto, Y.; Fujita, M.; Otsuka, M. Org. Biomol. Chem. 2014, 12, 5006.
8. Blair, W. S.; Pickford, C.; Irving, S. L.; Brown, D. G.; Anderson, M.; Bazin, R.; Cao,
J.; Ciaramella, G.; Isaacson, J.; Jackson, L.; Hunt, R.; Kjerrstrom, A.; Nieman, J. A.;
Patick, A. K.; Perros, M.; Scott, A. D.; Whitby, K.; Wu, H.; Butler, S. L. PLoS Pathog.
2010, 6, e1001220.
9. Kortagere, S.; Madani, N.; Mankowski, M. K.; Schon, A.; Zentner, I.;
Swaminathan, G.; Princiotto, A.; Anthony, K.; Oza, A.; Sierra, L. J.; Passic, S. R.;
Wang, X.; Jones, D. M.; Stavale, E.; Krebs, F. C.; Martin-Garcia, J.; Freire, E.; Ptak,
R. G.; Sodroski, J.; Cocklin, S.; Smith, A. B., 3rd J. Virol. 2012, 86, 8472.
10. Lamorte, L.; Titolo, S.; Lemke, C. T.; Goudreau, N.; Mercier, J. F.; Wardrop, E.;
Shah, V. B.; von Schwedler, U. K.; Langelier, C.; Banik, S. S.; Aiken, C.; Sundquist,
W. I.; Mason, S. W. Antimicrob. Agents Chemother. 2013, 57, 4622.
tion by cell-based screening and not by a molecular target
approach,4 it is not inconceivable that the direct target of BMMP
is something other than Pr55Gag or CA protein.
In summary, we synthesized Biotin–BMMP, and confirmed its
usefulness as a tool for HIV target identification. Surface plasmon
resonance experiments using Biotin–BMMP suggested the target
of BMMP other than Pr55Gag or CA protein. We are now in the pro-