Use of a hexasubstituted benzene scaffold in the development of multivalent HIV-1 integrase inhibitors

Article abstract of DOI:10.1248/cpb.c14-00074

The highly directional hexasubstituted benzene moiety was used as the central scaffold to create new human immunodeficiency virus (HIV)-1 integrase inhibitors through the attachment of multiple active groups. A series of potential inhibitors having substi

Full text of DOI:10.1248/cpb.c14-00074

754
Regular Article
Chem. Pharm. Bull. 62(8) 754–763 (2014)
Vol. 62, No. 8
Use of a Hexasubstituted Benzene Scaffold in the Development of
Multivalent HIV-1 Integrase Inhibitors
Wipa Tupchiangmai,a Saowanaporn Choksakulporn,a Supinya Tewtrakul,b
,a
Somsak Pianwanit,a and Yongsak Sritana-anant*
a Department of Chemistry, Faculty of Science, Chulalongkorn University; Phayathai Rd, Pathumwan, Bangkok
b
10330, Thailand: and Department of Pharmacognosy and Pharmaceutical Botany, Faculty of Pharmaceutical
Sciences, Prince of Songkla University; Hat-Yai, Songkhla 90112, Thailand.
Received January 23, 2014; accepted April 30, 2014
The highly directional hexasubstituted benzene moiety was used as the central scaffold to create new
human immunodeficiency virus (HIV)-1 integrase inhibitors through the attachment of multiple active
groups. A series of potential inhibitors having substituted polyhydroxylated mono, bis and tris-cinnamoyl
derivatives connected on the scaffold were prepared through Claisen–Schmidt condensations with substi-
tuted benzaldehydes, followed by partial demethylation to uncover the active phenolic groups required for
the interactions with the integrase enzyme active sites. Using a multiplate integration assay method, four
compounds carrying at least two sets of interacting moieties were found to be relatively potent integrase
inhibitors with IC50 values in the low micromolar range. The results confirmed that multiple polyhydroxyl-
ated groups were required on the platform in order to effectively interact with the enzyme. The results from
molecular docking studies consistently complemented the experimental results and revealed the nature of the
potential key binding interactions responsible for the apparent activity of the active compounds.
Key words multivalent drug; hexasubstituted benzene; integrase inhibitor; cinnamoyl; polyhydroxylated aro-
matic
Multivalency, or polyvalency, the presence of multiple tentially provides favorable synergistic interactions within the
structural units connected to a central core that could induce binding sites of the target enzyme. Such pre-organized control
simultaneous interactions with another entity or receptor, has of directionality is quite unique and rarely found in other plat-
become a ubiquitous strategy for the design of new biologi- forms.
cally active agents.1–3) Multivalent compounds are potentially
One of the potential targets for this multivalent scaffold is
more effective than traditional one-site ligands due to the the human immunodeficiency virus (HIV) integrase enzyme
favorable multiple interactions with the desired targets. The (IN). This enzyme has no counterpart in the human host cell
design also allows one to diversify the interactive groups to and hence its inhibitors tend to show little or no side effect
suit the existing binding sites of the target, and facilitate the in humans. Currently there are only a few drugs of this class
process of identifications of these binding units. The potential available.7,8) Among many agents investigated, the polyhydrox-
complementary binding, and diversity available on such a ylated aromatics and their related derivatives repeatedly ap-
platform allows these molecules to closely resemble the ac- pear as one of the most potent classes of active IN inhibitors.
tual biochemical interactions of the target enzymes with their Their activities were shown to inhibit both 3′-processing and
substrates or inhibitors. Among many reported core platforms, strand transfer reactions of the enzyme in the submicromolar
hexasubstituted benzene derivatives are of particular inter- or nanomolar ranges. Compounds with two catechol units
est.4,5) They are easy to prepare and can, in principle, hold up separated by a central linker have been shown to be more po-
to six active groups. These groups are also uniquely arranged tent than compounds having only one such unit, indicating the
into two sets where three meta-arranged substituents point apparent benefits of multivalent binding. Some of the leading
perpendicularly outward from one face of the central aromatic active compounds in this class are divalent, including caffeic
plane, while the other alternating arms are directed out from acid phenethyl ester (CAPE), L-chicoric acid and its 3,5-dicaf-
the opposite face6) (Fig. 1). This conformational rigidity po- feoyl derivatives,9–13) and a series of cinnamoyl or chalcone
pharmacophores.14–17)
Although current treatments for AIDS and HIV infected
patients have shown significant improvements over those
used previously, the eventual eradication of the virus has not
yet been achieved.18,19) With the emergence of new strains of
the virus showing resistance against the available drugs, the
continued search for new therapeutic agents and targets at
different stages of the viral life cycle is essential. It is widely
accepted that the most effective strategy for HIV treatment re-
quires several inhibitors combined in one dose, hereby known
as highly active retroviral therapy (HAART). An addition of
IN inhibitors as part of this strategy would expand the alterna-
tives for new formulations and potentially raise their activi-
ties, especially against the resistant virus strains.
Fig. 1. The Arrangement of the Substituents on the Hexasubstituted
Benzene Core
The authors declare no conflict of interest.
© 2014 The Pharmaceutical Society of Japan
*To whom correspondence should be addressed. e-mail: yongsak.s@chula.ac.th

August 2014
755
In this report, we tested the feasibility of the hexasubsti-
tuted aromatic platform as a central linker between groups of
Results and Discussion
Synthesis The triacetyl derivative of phloroglucinol 2 was
polyhydroxylated aromatics as the basis for new inhibitors. Up selected as the hexasubstituted benzene core precursor due to
to three groups were connected onto the platform as mono, its preparative simplicity and the possibility for further func-
bis and tris-cinnamoyl or chalcone-like analogs with various tionalization through both the acetyl-, and hydroxyl groups
patterns of hydroxyl-, or methoxy substituents on the flank- as shown in Chart 1.23–25) The compound could be made in
ing aromatic rings. Without the ester linkages employed in good yield from the acetylation of phloroglucinol 1 based on
many other scaffolds, the possible susceptibility toward hy- reported procedures or using our own developed methodol-
drolytic enzymes was eliminated.12) The effect of the crowded ogy.26) In this report, we focused on the functionalization of
rigid platform would be to stretch out the cinnamoyl groups 2 only through the acetyl groups. To obtain the alternating ar-
and constrain them into the preferable syn disposition of the rangements as shown in Fig. 1, compound 2 was subjected to
carbonyls with respect to their vinylic double bonds.14,20) To excessive methylation of all phenolic hydroxyl groups to yield
figure out the nature of the fundamental interactions and the 1,3,5-triacetyl-2,4,6-trimethoxybenzene 3.27) Claisen–Schmidt
multivalency effect of the binding groups, the positions and condensations of the acetyl groups with selected benzaldehyde
numbers of free hydroxyl groups, and the number of cin- derivatives allowed up to three pharmacophores to be attached
namoyl moieties attached onto the core platform were varied onto the central platform, giving rise to the multi-cinnamoyl
over the series. The inhibitory activities of the synthesized analogs 5–7.14,28,29) Typical conditions as reported previously
compounds were evaluated against the integrase enzyme were attempted first with unsatisfactory results, providing
using a multiplate integration assay.21,22) Finally, computational only low yields of mixtures of incomplete condensed products,
docking studies of the potential enzyme-substrate complexes even at very long reaction times. It was later found that un-
were performed in an attempt to determine the interactions usually high concentrations of hydroxide solution were needed
between the active moieties of these potential inhibitors and to induce the effective condensations at room temperature to
the bound amino acid residues within the enzyme. Correla- afford reasonable yields of mixtures of mono- (5), bis- (6)
tions between theory and the experimental bioactivity results and tris-cinnamoyl (7) products, in various ratios. Because
are also discussed herein.
the main purpose is to obtain compounds having a range of
substitution patterns in order to screen and compare their po-
tential activities, the synthetic procedures developed here were
Reagents and conditions: (i) AcCl, AlCl3; (ii) (CH3)2SO4, K2CO3, CH3CN; (iii) 80% KOH, MeOH and benzaldehyde derivatives 4 (4a: 3-hydroxybenzaldehyde, 4b: 4-hy-
droxybenzaldehyde, 4c: vanillin); (iv) AlCl3, pyridine/CH2Cl2.
Chart 1

756
Vol. 62, No. 8
not optimized to obtain the maximum yields of any particular lecular additions of the free phenolic hydroxyl groups to the
products. Rather, we deliberately interrupted the reactions to adjacent unsaturated ketones were unexpectedly obtained as
obtain a mixture of the three desired products from each of products.31–33)
the consecutive three steps of condensations, and subsequently
IN Inhibition Assays The studies on HIV-1 IN inhibi-
separated each of them by column chromatography to allow tory activity were carried out using the reported multiplate
the availability of all three condensed products for bioactivity integration assay (MIA) method.21,22) Suramin was used as the
tests. Longer reaction times and higher stoichiometric ratios of positive control. Only the compounds showing relatively high
the substituted benzaldehyde derivatives would be expected to activities were carried on to subsequent tests at lower concen-
drive the reactions towards completion, and therefore provide trations. The results are shown in Table 1.
higher yields of the tris-cinnamoyl products.
The results showed that the core structure compounds 2 and
Under the conditions described above, mono-cinnamoyl 3 without the extending cinnamoyl arms appeared inactive.
compounds 5a–c were obtained as the major products from The rather compact size of 2 and the lack of hydroxyl binding
their corresponding condensations, along with sequentially groups in 3 were assumed to be possible reasons for their in-
lesser amounts of bis-cinnamoyl compounds 6a–c, and tris- activity. These results eliminated the possibility that the core
cinnamoyl compounds 7a–c, respectively. The reactions of structure alone might be involved in the interactions with the
these benzaldehyde derivatives were rather sluggish and enzyme binding sites. The series of mono-cinnamoyl analogs
sometimes resulted in the formation of unknown byproducts 5a–d also showed no activity against the integrase enzyme,
that complicated the successive purification processes. Conse- suggesting that the presence of only one binding group was
quently, we prepared the multi-phenolic derivatives 5d, 6d,e not sufficient to show any detectable effect, emphasizing the
and 7d–f from their methoxy precursors 5c, 6c and 7c re- necessity of building in multivalent active groups. Adding one
spectively. Demethylations using AlCl3 and pyridine could be more cinnamoyl side chain (compound 6 series) showed im-
controlled to remove some or all of the methyl groups to give provement, with the derivative carrying two catechol moieties
both symmetric and non-symmetric compounds.30) Interest- (6e) becoming highly active, in comparison to 6d, which has
ingly, the inner three methoxy groups on the central aromatic one group and was almost inactive. These results confirmed
core were not affected by this reaction and all remained intact that a polyhydroxylated aromatic compound bearing more
in the products.
than one catecholic group was preferred, or even required, to
Because the presence of multiple phenolic groups in the become an effective integrase inhibitor.
molecule potentially raises the activity, complete demethyl-
The series of tris-cinnamoyl analogs 7a–f further under-
ations including those methoxy groups on the hexasubstituted lined the above observations. While compound 7d with only
benzene core would generate another desirable group of ac- one catecholic group was inactive, both bis-catechol 7e and
tive compounds. After our earlier failed attempts on core tris-catechol 7f appeared to be quite potent inhibitors. Al-
demethylation reactions, we decided to build these series from though the differences of the IC50 values among 7e,f and that
the 1,3,5-triacetyl-2,4,6-trihydroxybenzene precursor using a of bis-catechol 6e previously discussed were not substantial,
similar condensation strategy. Unfortunately, we once again with only slightly higher potency for 7f than 7e, this implied
encountered sluggish reactions, presumably resulting from that at least two catecholic groups were needed for the inhibi-
the presence of the extra phenolic groups. Small amounts of tion potency. The concept of multivalency in inhibitor design
cyclized flavonoid-like compounds derived from intramo- was then justified, but these results indicated that there was no
Table 1. Percentage Inhibition Results for HIV-1 IN Inhibitors
% Inhibition at concentration
Compound
IC50 (µM)
100µM
30µM
10µM
2
3
21.02 3.73
11.74 3.80
17.91 1.61
NPa)
—
—
>100
>100
>100
>100
>100
>100
>100
>100
>100
89.4
—
—
5a
—
—
5b
5c
—
—
25.63 4.89
24.94 3.16
36.12 4.08
26.24 2.56
33.07 0.05
55.95 1.02
90.30 0.03
79.96 2.13
40.45 1.27
35.75 4.98
44.36 3.30
74.78 2.61
93.60 0.94
99.89 0.41
—
—
5d
6a
—
—
—
—
6b
6c
—
—
—
—
6d
6e
16.25 1.20
86.88 1.60
68.24 1.09
—
2.74 4.32
59.46 1.91
49.15 2.33
—
3.5
7a
9.5
7b
7c
>100
>100
>100
16.5
—
—
7d
7e
—
—
56.58 1.39
81.03 0.13
89.40 1.73
44.01 2.70
75.75 1.65
69.72 1.56
7f
7.0
Suramin
2.7
a) NP=No positive result.

August 2014
757
significant additive effect from adding more than two active ness scores and potential hydrogen bonding distances between
pendant groups onto the molecules, at least for this particular each ligand and IN for all 15 compounds as well as suramin
platform.
are given in Table 2. For the series of inactive mono-cinnamo-
An interesting result was observed with the unexpected yl analogs 5a–d, the number of the predicted H-bonding inter-
activity of compound 7a, which lacks the catechol moiety but actions was lower than that of the active compounds in the se-
has three meta-hydroxyl groups on the flanking cinnamoyl ries. In addition, these were found to bind to different amino
appendages, showing a potent IC50 value of 9.5µM. The result acid residues than those of the active compounds. Increasing
indicated that the main interactions of the hydroxyl groups the number of pendant cinnamoyl moieties enlarged the size
from either phenol or catechol side chains of the compounds of the molecules, and enabled more interactions (higher fitness
with the enzyme relied heavily on those groups at the meta- scores). The bis-cinnamoyl analogs 6a,b, which contained
positions, relative to the connections to the core linker. In this only one hydroxyl group per pendant residue, formed only one
case, only the cinnamoyl derivative with three meta-phenolic H-bond with IN while compound 6c having both hydroxyl and
groups could induce such interactions sufficiently. All deriva- methoxy groups could potentially form more. Although com-
tives prepared from the condensations with vanillin, in which pounds 6d,e formed comparable number of H-bonds to 6c,
the meta-hydroxyl group was masked as a methoxy group, (5c, the binding patterns were different. They were found to bind
6c, 7c,d) were inactive.
to both LYS156 and LYS159 whereas 6c bound to only one of
Molecular Docking Study The correlations between these residues.34,35) These binding differences as well as mul-
inhibitory activity and structure of the inhibitor series was tiple H-bond interactions were the possible reasons why only
probed through undertaking a molecular docking study. Fit- 6d,e showed significant activity. The fact that 6e formed more
Table 2. The Docking Fitness Scores of the Ligands with Their Interacting Residues within IN
Compound
Score
Ligand
Amino acid residues
ASP64
Distance (Å)
5a
5b
47.30
51.38
1-Ph-m-OH
1.99
1-Ph-p-OH
1-Ph-p-OH
ASN155
ASP64
2.44
2.09
5c
51.59
51.12
1-Ph-p-OH
1-Ph-p-OH
ASN155
ASP64
1.82
2.08
5d
1-Ph-m,p-di-OH
1-Ph-p-OH
ASP64
ASN155
1.73, 1.64
2.27
6a
6b
6c
64.47
60.54
59.54
1-Ph2-m-OH
3-Ph1-p-HO
GLN148
LYS156
2.11
2.06
1-Ph1-m-OMe
5-Acetyl-CO
LYS159
GLN148
2.02
1.64, 2.41
6d
6e
61.15
58.15
1-Ph1-m-OH
1-Ph1-p-OH
LYS159
LYS156
1.93, 2.29
2.06
1-Ph1-m-OH
1-Ph1-p-OH
1-Ph1-p-OH
3-Ph2-m-OH
5-Acetyl-CO
LYS159
LYS159
LYS156
PHE139
GLN148
2.00
2.49
2.09
1.64
1.63
7a
7b
73.49
67.26
1-Ph1-m-OH
3-Ph2-m-OH
LYS159
PHE139
2.26, 2.44
1.64
1-Ph1-p-OH
3-Ph2-p-OH
5-Ph2-p-OH
LYS156
SER116
GLY140
1.95
1.92
2.40
7c
70.62
72.85
1-Ph1-m-OMe
1-Ph1-p-OH
LYS159
LYS159
2.10
2.41
7d
1-Ph1-p-OMe
3-Ph2-p-OH
6-OMe
LYS156
SER119
GLN148
1.65
2.05
1.93
7e
7f
70.95
69.32
73.18
1-Ph1-m-OH
1-Ph1-p-OH
1-Ph1-p-OH
5-Ph3-CO
LYS159
LYS159
LYS156
GLN148
1.89
2.46
2.08
1.66
1-Ph1-m-OH
1-Ph1-p-OH
1-Ph1-p-OH
5-Ph3-CO
LYS159
LYS159
LYS156
GLN148
1.92, 2.45
2.39
2.08
1.97
Suramin
1-NPTH-1-SO3H
GLU 152
LYS156
LYS159
ASP116
SER119
1.66
2.39
1.76
1.52
2.12
1-NPTH-3-SO3H
2-NPTH-1-SO3H
2-NPTH-3-SO3H
NPTH=naphthalene.

758
Vol. 62, No. 8
Fig. 2. Predicted Protein–Ligand Interactions of Superimposed Structures of 5d, 6e and 7f with Amino Acid Residues of IN (Small Stick) and Mg2+
Fig. 3. Predicted Protein–Ligand Interaction of Superimposed Structures of 6e, 7f and 7a with Amino Acid Residues of IN (Small Stick) and Mg2+
H-bonding interactions than 6d due to more catechol groups bis-cinnamoyl derivative 6e is bound to IN in very similar
was reflected in its higher activity.
fashion to the tris-cinnamoyl compound 7f. Both compounds
In the case of tris-cinnamoyl analogs, most compounds appear to have comparable activities. In addition, this result
formed more than one H-bonding interaction with IN. The validated our earlier speculation that the third additional cat-
reason just discussed for the apparent activity of the bis- echol group in 7f compared to 6e is not necessary for the en-
cinnamoyl 6d and e, i.e., the binding of the ligand to both zyme binding and its addition does not enhance the associated
LYS156 and LYS159, could as well be used to explain the biological activity.
greater activity of the tris-cinnamoyl 7e and f. The surpris-
The three most active compounds from the bioassays as
ingly active tris-m-hydroxycinnamoyl analog 7a was found reported in Table 1 (6e, 7a,f) shared similar binding patterns
to form H-bonding interactions with PHE139, and LYS159 as (Fig. 3), in which all compounds formed H-bonding interac-
shown for 6e, the most active compound. This result under- tions with the LYS159 residue of IN. Although the orientation
lined the leading importance of having free hydroxyl groups upon binding of 7a was partly different from those of 6e and
at meta-positions of the flanking cinnamoyl arms as a require- 7f, it could also form H-bonding interactions with PHE139,
ment to enhance binding to the IN active site.
similar to 6e. In addition, the binding of these three com-
In order to investigate the effect of the number of active pounds are also similar to that of Suramin, which forms H-
groups on the binding efficiency, compounds 5d, 6e and bonding interactions with GLU152, LYS156, LYS159, ASP116
7f, which contained one, two, and three catechol groups re- and SER119, reflecting in their comparable activities.
spectively, were selected as a visual comparison (Fig. 2). It
is clearly seen that the mono-cinnamoyl molecule 5d is lo-
Conclusion
cated quite far from the IN amino acid residues LYS156 and
We have successfully prepared a series of cinnamoyl ana-
LYS159, thus explaining its inactivity. On the other hand, the logs based on the hexasubstituted benzene platform through

August 2014
759
the Claisen–Schmidt condensations of 1,3,5-triacetyl-2,4,6- Compound 2 (5.99g, 23.8mmol) was dissolved in acetonitrile
trimethoxybenzene and substituted benzaldehydes. The syn- (250mL). Dimethyl sulfate (13.5mL, 142.8mmol) and K2CO3
thesized mono-, bis- and tris-cinnamoyl compounds displayed (49.26g, 357.0mmol) were added and the mixture was stirred
various patterns of hydroxyl and methoxy substituents on the under reflux for 13h. The solvent was then removed and water
flanking aromatic rings. These compounds became the first (150mL) was added. The resulted precipitate was filtered,
group of hexasubstituted benzene derivatives to be tested as washed with cool water and recrystallized from ethanol to
potential biological inhibitors. The bioassays of these com- give a white solid of 3: 6.50g, yield 93%; mp 124–126°C
pounds indicated that two catecholic groups were needed to be (lit.23) mp 122–124°C); 1H-NMR (CDCl3) δ: 3.75 (9H, s,
present on the platform in order for the molecules to become –OCH3), 2.54 (9H, s, –COCH3); 13C-NMR (CDCl3) δ: 200.8,
active against the HIV-1 integrase enzyme. An additional 155.4, 127.3, 64.6, 32.6; HR-MS (ESI–methanol) m/z: (M+
binding group would just slightly improved the activity and Na)+ 317.1019 (Calcd for C15H18NaO6: 317.0996).
may not be necessary to be included on the platform. Howev-
General Procedure A: Synthesis of Multi-cinnamoyl An-
er, an exception was noted when each of the flanking aromatic alogs (5a–c, 6a–c and 7a–c)14,28,29) A mixture of compound
ring carried only one hydroxyl group at its meta-position 3 (588mg, 2.0mmol) and one of the corresponding benzal-
relative to the connection to the core, as demonstrated by the dehyde derivatives 4a–c (4.0mmol) in methanol (20–30mL)
activity of tris-cinnamoyl derivatives 7a. These results support was added slowly into a well stirred aqueous suspension of
the earlier observations regarding the requirement of mul- 80% w/v KOH (6.0mL). The mixture was stirred at room
tivalency for potential inhibitors. Molecular docking studies temperature for 1–2d and worked up with 10% HCl and ex-
of these compounds supported the above results and suggested tracted with ethyl acetate (3×50mL). The organic extracts
the importance of LYS156 and LYS159 of the IN being the were collected and washed with brine (3×50mL), dried with
key residues that participate in forming binding interactions anhydrous Na2SO4 and the solvent was evaporated. The crude
with the active inhibitors.34) Now that the active functional product was purified by gradient chromatography on silica
groups needed to be present on the potential inhibitors were gel column (hexane/ethyl acetate, 70:30, 60:40 and 20:80)
identified, the hexasubstituted benzene core could be further to yield the corresponding pure products 5a–c, 6a–c or 7a–c.
modified to incorporate other groups that would enhance its
(E)-1-(3,5-Diacetyl-2,4,6-trimethoxyphenyl)-3-(3-
inhibitory activity through other means such as solubility, hydroxyphenyl)prop-2-enone (5a): Yellow solid (242mg, 28%
1
hydrophobicity or even different functional groups capable of yield), mp 107–110°C; H-NMR (acetone-d6) δ: 7.46 (1H, d,
binding to other residues of the enzyme. Since chalcones are J=16.0Hz, H-β), 7.29 (1H, t, J=8.0Hz, ArH), 7.20 (1H, d,
known to be quite a versatile class of compounds that exhibit J=8.0Hz, ArH), 7.17 (1H, s, ArH), 7.07 (1H, d, J=16.0Hz,
a variety of pharmacological activities, further tests on other H-α), 6.96 (1H, d, J=8.0Hz, ArH), 3.79 (3H, s, –OCH3), 3.73
targets could reveal a wider therapeutic potential for these (6H, s, –OCH3), 2.53 (6H, s, –COCH3); 13C-NMR (acetone-
multivalent compounds.28–30,36)
d6) δ: 200.6, 192.9, 158.7, 156.8, 156.1, 146.9, 136.8, 130.9,
128.9, 127.5, 125.5, 121.2, 118.9, 115.8, 64.5, 64.2, 32.6;
HR-MS (ESI–methanol) m/z: (M+Na)+ 421.1275 (Calcd for
Experimental
General Information 1H- and 13C-NMR spectra were C22H22O7Na: 421.1258).
recorded on Varian Mercury 400 or Bruker Avance 400
spectrometers, operated at 400MHz for H and 100MHz for enoyl]-2,4,6-trimethoxyphenyl]-3-(3-hydroxyphenyl)prop-2-
(E)-1-[3-Acetyl-5-[(E)-3-(3-hydroxyphenyl)prop-2-
1
13C nuclei. High resolution mass spectra were determined on enone (6a): Yellow solid (148mg, 14% yield), mp 164–166°C;
Bruker Daltonik GmbH micrOTOF-Q II. Melting points were 1H-NMR (acetone-d6) δ: 7.51 (2H, d, J=16.0Hz, H-β), 7.28
determined using a Stuart Scientific Melting Point SMP1 (2H, t, J=8.0Hz, ArH), 7.20 (2H, d, J=8.0Hz, ArH), 7.19 (2H,
(Bibby Sterlin Ltd.). Thin layer chromatography (TLC) was s, ArH), 7.08 (2H, d, J=16.0Hz, H-α), 6.95 (2H, d, J=8.0Hz,
carried out using precoated silica gel on aluminum sheets ArH), 3.77 (6H, s, –OCH3), 2.55 (3H, s, –COCH3), 3.70 (3H,
(Merck Kieselgel 60 F254 with layer thickness 0.25mm). s, –OCH3); 13C-NMR (acetone-d6) δ: 200.7, 193.1, 158.8, 157.5,
Flash column chromatography was performed using Merck 156.9, 146.9, 136.8, 130.9, 128.9, 127.5, 125.4, 122.2, 121.2,
Kieselgel 60 Art 9385 (230/400 mesh). All reagents and sol- 118.9, 115.7, 64.2, 63.9, 32.6; HR-MS (ESI–methanol) m/z:
vents were used as purchased or distilled prior to use.
(M+Na)+ 525.1540 (Calcd for C29H26O8Na: 525.1520).
Synthesis of 1-(3,5-Diacetyl-2,4,6-trihydroxyphenyl)-
(E)-1-[3,5-Bis[(E)-3-(3-hydroxyphenyl)prop-2-enoyl]-2,4,6-
ethanone (1,3,5-Triacetylphloro-glucinol, 2)23,26) Acetyl trimethoxyphenyl]-3-(3-hydroxyphenyl)prop-2-enone (7a): Yel-
1
chloride (50mL) was added to phloroglucinol dihydrate low solid (43mg, 3% yield), mp 244–246°C; H-NMR (ace-
(C6H6O3·2H2O) 1 (5.01g, 30.9mmol) and anhydrous AlCl3 tone-d6) δ: 7.56 (3H, d, J=16.0Hz, H-β), 7.28 (3H, t, J=8.0Hz,
(20.47g, 154.5mmol). The reaction was stirred under reflux ArH), 3.74 (9H, s, –OCH3), 7.21 (3H, d, J=8.0Hz, ArH),
for 1h and quenched with 10% HCl (10mL), water (150mL) 7.20 (3H, s, ArH), 7.09 (3H, d, J=16.0Hz, H-α), 6.95 (3H, d,
and then filtered. The collected precipitate was recrystallized J=8.0Hz, ArH); 13C-NMR (acetone-d6) δ: 193.2, 158.8, 157.9,
from ethanol to give colorless needle crystals of 2: 7.09g, 146.8, 136.8, 130.9, 129.0, 125.4, 121.3, 118.9, 115.6, 63.9;
1
91% yield, mp 155–156°C (lit.23,24) mp 155–156°C); H-NMR HR-MS (ESI–methanol) m/z: (M+Na)+ 629.1780 (Calcd for
(CDCl3) δ: 17.16 (3H, s, –OH), 2.72 (9H, s, –COCH3); C36H30O9Na: 629.1782).
13C-NMR (CDCl3) δ: 205.1, 175.9, 103.3, 33.0; high resolution
(E)-1-(3,5-Diacetyl-2,4,6-trimethoxyphenyl)-3-(4-
(HR)-MS (electrospray ionization (ESI)/methanol) m/z: (M+ hydroxyphenyl)prop-2-enone (5b): Yellow solid (399mg, 49%
1
Na)+ 275.0539 (Calcd for C12H12NaO6: 275.0532).
yield), mp 129–131°C; H-NMR (acetone-d6) δ: 7.61 (2H, d,
Synthesis of 1-(3,5-Diacetyl-2,4,6-trimethoxyphenyl)- J=8.0Hz, ArH), 7.45 (1H, d, J=16.0Hz, H-β), 6.97 (1H, d,
ethanone (1,3,5-Triacetyl-2,4,6-trimethoxybenzene, 3)27) J=16.0Hz, H-α), 6.92 (2H, d, J=8.0Hz, ArH), 3.78 (3H, s,

760
Vol. 62, No. 8
–OCH3), 3.73 (6H, s, –OCH3), 2.53 (6H, s, –COCH3); 13C- for 10min and pyridine (1.0mL) was added. The reaction
NMR (acetone-d6) δ: 199.8, 191.9, 160.4, 155.8, 154.9, was then stirred under reflux for 30h and worked up with
146.4, 130.8, 126.6, 126.0, 125.2, 116.0, 63.6, 63.2, 31.7; 10% HCl. The separated water layer was extracted with ethyl
HR-MS (ESI–methanol) m/z: (M+Na)+ 421.1260 (Calcd for acetate (3×50mL). The organic extracts were combined with
C22H22O7Na: 421.1258).
the previous dichloromethane layer and washed with brine
(E)-1-[3-Acetyl-5-[(E)-3-(4-hydroxyphenyl)prop-2- (3×50mL) and dried with anhydrous Na2SO4. Evaporation of
enoyl]-2,4,6-trimethoxyphenyl]-3-(4-hydroxyphenyl)prop-2- the solvent and purification of the crude product by gradient
enone (6b): Yellow solid (196mg, 19% yield), mp 216–219°C; chromatography on silica gel column (hexane–ethyl acetate=
1H-NMR (acetone-d6) δ: 7.60 (2H, d, J=16.0Hz, H-β), 7.47 80:30, 70:40) afforded unreacted 5c: 60mg (29% recovered)
(4H, d, J=8.0Hz, ArH), 6.95 (2H, d, J=16.0Hz, H-α), 6.90 and pure product 5d.
(4H, d, J=8.0Hz, ArH), 3.74 (6H, s, –OCH3), 3.66 (3H, s,
(E)-1-(3,5-Diacetyl-2,4,6-trimethoxyphenyl)-3-(3,4-
–OCH3), 2.52 (6H, s, –COCH3); 13C-NMR (acetone-d6) δ: dihydroxyphenyl)prop-2-enone (5d): Yellow solid (34mg, 18%
1
199.9, 192.2, 160.4, 156.4, 155.6, 146.5, 130.8, 126.6, 125.9, yield), mp 142–144°C; H-NMR (acetone-d6) δ: 7.38 (1H, d,
125.2, 124.8, 115.9, 63.2, 62.9, 31.7; HR-MS (ESI–methanol) J=16.0Hz, H-β), 7.22 (1H, s, ArH), 7.09 (1H, d, J=8.0Hz,
m/z: (M+Na)+ 525.1530 (Calcd for C29H26O8Na: 525.1520).
ArH), 6.90 (1H, d, J=16.0Hz, H-α), 6.89 (1H, d, J=8.0Hz,
(E)-1-[3,5-Bis[(E)-3-(4-hydroxyphenyl)prop-2-enoyl]-2,4,6- ArH), 3.78 (3H, s, –OCH3), 3.72 (6H, s, –OCH3), 2.52 (6H,
trimethoxyphenyl]-3-(4-hydroxyphenyl)prop-2-enone
(7b): s, –COCH3); 13C-NMR (acetone-d6) δ: 200.7, 192.8, 156.7,
Yellow solid (43mg, 3% yield), mp 239–240°C; 1H-NMR 155.8, 149.5, 147.7, 146.4, 127.5, 126.1, 125.9, 123.5, 116.5,
(acetone-d6) δ: 7.61 (6H, d, J=8.0Hz, ArH), 7.51 (3H, d, 115.6, 64.5, 64.1, 32.6; HR-MS (ESI–methanol) m/z: (M+Na)+
J=16.0Hz, H-β), 6.96 (3H, d, J=16.0Hz, H-α), 6.90 (6H, d, 437.1217 (Calcd for C22H22O8Na: 437.1207).
J=8.0Hz, ArH), 3.70 (9H, s, –OCH3); 13C-NMR (acetone-d6)
Following the general procedure B, compound 6c (240mg,
δ: 192.3, 160.2, 156.3, 146.3, 130.8, 126.1, 125.4, 124.8, 115.9, 0.4mmol) was dissolved in dichloromethane (30mL) and an-
62.9; HR-MS (ESI–methanol) m/z: (M+Na)+ 629.1775 (Calcd hydrous AlCl3 (400mg, 2.9mmol) was added. The mixture
for C36H30O9Na: 629.1782).
was stirred at room temperature for 10min, pyridine (1.5mL)
(E)-1-(3,5-Diacetyl-2,4,6-trimethoxyphenyl)-3-(4-hydroxy-3- was added, and then the mixture was heated under reflux for
methoxyphenyl)prop-2-enone (5c): Yellow solid (430mg, 48% 10h. After the usual work up, the crude product was purified
1
yield), mp 129–131°C; H-NMR (acetone-d6) δ: 7.43 (1H, d, by gradient chromatography on silica gel column (hexane–
J=16.0Hz, H-β), 7.41 (1H, s, ArH), 7.19 (1H, d, J=8.0Hz, ethyl acetate=60:30, 40:60) to recover the unreacted 6c:
ArH), 7.01 (1H, d, J=16.0Hz, H-α), 6.89 (1H, d, J=8.0Hz, 47mg (19% recovered) and pure products 6d and 6e.
ArH), 3.93 (3H, s, –OCH3), 3.78 (3H, s, –OCH3), 3.73 (6H,
(E)-1-[3-Acetyl-5-[(E)-3-(3-hydroxy-4-methoxyphenyl)-
s, –OCH3), 2.52 (6H, s, –COCH3); 13C-NMR (acetone-d6) prop -2-enoyl]-2,4,6 -t r imethoxy phenyl]-3-(3,4 -di-
δ: 199.7, 191.9, 155.7, 154.9, 152.0, 149.8, 147.9, 146.8, 126.6, hydroxyphenyl)prop-2-enone (6d): Yellow solid (30mg, 12%
1
126.5, 125.4, 125.1, 124.8, 124.1, 121.7, 115.3, 110.8, 63.6, 63.2, yield), mp 118–120°C; H-NMR (acetone-d6) δ: 7.46 (1H, d,
55.5, 31.7; HR-MS (ESI–methanol) m/z: (M+Na)+ 451.1370 J=20.0Hz, H-β), 7.42 (1H, d, J=16.0Hz, H-α), 7.40 (1H, s,
(Calcd for C23H24O8Na: 451.1363).
ArH), 7.24 (1H, s, ArH), 7.20 (1H, d, J=12.0Hz, ArH), 7.10
(E)-1-[3-Acetyl-5-[(E)-3-(4-hydroxy-3-methoxyphenyl)- (1H, d, J=8.0Hz, ArH), 7.00 (1H, d, J=16.0Hz, H-β), 6.90
prop-2-enoyl]-2,4,6-trimethoxyphenyl]-3-(4-hydroxy-3- (1H, d, J=16.0Hz, H-α), 6.89 (2H, d, J=8.0Hz, ArH), 3.92
methoxyphenyl)prop-2-enone (6c): Yellow solid (210mg, 18% (3H, s, –OCH3), 3.76 (6H, s, –OCH3), 3.67 (3H, s, –OCH3),
1
yield), mp 188–190°C; H-NMR (acetone-d6) δ: 7.46 (2H, d, 2.53 (3H, s, –COCH3); 13C-NMR (acetone-d6) δ: 200.8, 193.1,
J=16.0Hz, H-β), 7.39 (2H, s, ArH), 7.19 (2H, d, J=8.0Hz, 157.3, 156.6, 150.7, 149.5, 148.9, 147.7, 147.6, 146.5, 127.6,
ArH), 6.99 (2H, d, J=16.0Hz, H-α), 6.87 (2H, d, J=8.0Hz, 127.4, 126.5, 126.2, 125.8, 124.9, 123.5, 116.5, 116.2, 115.5,
ArH), 3.90 (6H, s, –OCH3), 3.74 (3H, s, –OCH3), 3.66 (3H, 111.7, 64.1, 63.8, 56.4, 32.6; HR-MS (ESI–methanol) m/z: (M+
s, –OCH3), 2.52 (3H, s, –COCH3); 13C-NMR (acetone-d6) δ: Na)+ 571.1599 (Calcd for C30H28O10Na: 571.1575).
199.9, 192.2, 156.3, 155.5, 149.8, 147.9, 146.8, 126.5, 125.5,
(E)-1-[3-Acetyl-5-[(E)-3-(3,4-dihydroxyphenyl)prop-2-
124.9, 124.1, 115.3, 110.7, 63.2, 62.9, 55.5, 31.7; HR-MS (ESI– enoyl]-2,4,6-trimethoxy-phenyl]-3-(3,4-dihydroxyphenyl)-
methanol) m/z: (M+Na)+ 585.1739 (Calcd for C31H30O10Na: prop-2-enone (6e): Yellow solid (20mg, 8% yield), mp
1
585.1731).
122–124°C; H-NMR (acetone-d6) δ: 7.39 (2H, d, J=16.0Hz,
(E)-1-[3,5-Bis[(E)-3-(4-hydroxy-3-methoxyphenyl)- H-β), 7.22 (2H, s, ArH), 7.08 (2H, d, J=8.0Hz, ArH), 6.88
prop-2-enoyl]-2,4,6-trimethoxy-phenyl]-3-(4-hydroxy-3- (2H, d, J=16.0Hz, H-α), 6.87 (2H, d, J=8.0Hz, ArH), 3.73
methoxyphenyl)prop-2-enone (7c): Yellow solid (70mg, 5% (6H, s, –OCH3), 3.65 (3H, s, –OCH3), 2.51 (3H, s, –COCH3);
1
yield), mp 228–230°C; H-NMR (acetone-d6) δ: 7.51 (3H, d, 13C-NMR (acetone-d6) δ: 201.0, 192.9, 168.3, 163.5, 156.5,
J=16.0Hz, H-β), 7.41 (3H, s, ArH), 7.23 (3H, d, J=8.0Hz, 149.4, 147.6, 147.4, 146.9, 146.4, 127.4, 126.3, 124.8, 123.7,
ArH), 7.01 (3H, d, J=16.0Hz, H-α), 6.89 (6H, d, J=8.0Hz, 123.5, 116.5, 115.5, 115.4, 64.1, 63.8, 32.6; HR-MS (ESI–
ArH), 3.92 (9H, s, –OCH3), 3.72 (9H, s, –OCH3); 13C-NMR methanol) m/z: (M+Na)+ 557.1435 (Calcd for C29H26O10Na:
(acetone-d6) δ: 192.4, 156.2, 149.8, 148.0, 146.7, 126.6, 125.7, 557.1418).
124.9, 124.0, 115.3, 110.8, 62.9, 55.5; HR-MS (ESI–methanol)
Following the general procedure B, compound 7c (400mg,
0.5mmol) dissolved in dichloromethane (30mL) and AlCl3
m/z: (M+Na)+ 719.2112 (Calcd for C39H36O12Na: 719.2099).
General Procedure B: Demethylations of the Methoxy (540mg, 4.0mmol) was added. The mixture was stirred at
Groups30) Anhydrous AlCl3 (450mg, 2.9mmol) was added room temperature for 10min, pyridine (1.5mL) was added
to compound 5c (210mg, 0.4mmol) dissolved in dichlorometh- and the mixture heated under reflux for 10h. After the usual
ane (80mL). The mixture was stirred at room temperature work up, the crude product was purified by preparative TLC

August 2014
761
(CH2Cl2–methanol=99:1) to recover the unreacted 7c: 37mg 1mM EDTA and 100mM KCl). Both solutions were heated at
(9% recovered) and pure products 7d, 7e and 7f. 85°C for 15min in water bath. Each sample solution was then
(E)-1-[3,5-Bis[(E)-3-(4-hydroxy-3-methoxyphenyl)prop-2- cooled gradually to room temperature in water bath for 3h
enoyl]-2,4,6-trimethoxy-phenyl]-3-(3,4-dihydroxyphenyl)- and stored at −20°C.
prop-2-enone (7d): Orange solid (31mg, 7.4% yield), mp
Pretreatment of the Multiplate To a 96-well plate was
1
152–154°C; H-NMR (acetone-d6) δ: 7.49 (2H, d, J=16.0Hz, added 50µL of a steptavidin solution containing 40µg/mL
H-β), 7.44 (1H, d, J=16.0Hz, H-α), 7.38 (2H, s, ArH), 7.28 steptavidin, 90mM Na2CO3 and 10mM KCl. The microplate
(2H, d, J=8.0Hz, ArH), 7.21 (1H, s, ArH), 7.06 (1H, d, was covered with plastic seal and allowed to stand gently
J=8.0Hz, ArH), 6.99 (2H, d, J=16.0Hz, H-β), 6.88 (2H, d, overnight at 4°C (for coating). After discarding the steptavi-
J=4.0Hz, ArH), 6.86 (1H, d, J=16.0Hz, H-α), 6.83 (1H, d, din coating solution, the coated microplate was washed with
J=4.0Hz, ArH), 3.91 (6H, s, –OCH3), 3.71 (9H, s, –OCH3); phosphate buffer saline (PBS) solution (300µL) twice. The
13C-NMR (acetone-d6) δ: 193.4, 193.1, 157.1, 157.0, 151.6, blocking buffer (300µL) containing 1% skin milk in PBS
150.9, 148.9, 147.7, 147.4, 127.3, 126.5, 125.9, 125.8, 125.5, was added into each well and the plate was kept at room
124.8, 123.6, 116.6, 116.3, 115.1, 111.8, 63.9, 63.8, 56.4; temperature for 30min (for blocking). After discarding block-
HR-MS (ESI–methanol) m/z: (M+Na)+ 705.1965 (Calcd for ing buffer, each well was washed with PBS solution (300µL)
C38H34O12Na: 705.1952).
twice and the PBS solution was completely removed. A bio-
(E ) -3 - (3,4 -Di hyd rox y phe nyl) -1-[3 -[(E ) -3 - (3,4 - tinylated LTR donor DNA (50µL) solution containing 10mM
dihydroxyphenyl)prop-2-enoyl]-5-[(E)-3-(4-hydroxy-3- Tris–HCl (pH 8.0), 1mM NaCl and 40mol/mL of LTR donor
methoxyphenyl)prop-2-enoyl]-2,4,6-trimethoxyphenyl]prop-2- DNA was added into each well and mixed gently at room tem-
enone (7e): Orange solid (27mg, 6.5% yield), mp 168–170°C; perature for 30min (for adsorption). After discarding the LTR
1H-NMR (acetone-d6) δ: 7.48 (1H, d, J=16.0Hz, H-β), 7.42 donor solution, the microplate was washed with PBS solution
(1H, s, ArH), 7.39 (2H, d, J=20Hz, H-α), 7.23 (2H, s, ArH), (300 µL) twice and each well was filled with 300µL of PBS (if
7.22 (1H, d, J=8.0Hz, ArH), 7.08 (2H, d, J=8.0Hz, ArH), necessary, stored at 4°C). Just before the integration reaction,
6.99 (1H, d, J=16.0Hz, H-β), 6.88 (1H, d, J=4.0Hz, ArH), the PBS solution was discarded and each well was rinsed with
6.87 (2H, d, J=16.0Hz, H-α), 6.84 (2H, d, J=4.0Hz, ArH), 300µL distilled water.
3.90 (3H, s, –OCH3), 3.71 (9H, s, –OCH3); 13C-NMR (acetone-
Multiplate Integration Assay (MIA) The integra-
d6) δ: 193.4, 193.2, 157.2, 157.1, 150.8, 150.7, 148.9, 148.0, tion reaction was evaluated using the method previously
147.6, 147.1, 127.4, 126.9, 126.5, 125.9, 125.8, 124.8, 123.3, described.21,22) A mixture (45µL) composed of 12µL of IN
116.5, 116.3, 115.5, 111.8, 63.9, 56.4; HR-MS (ESI–methanol) buffer [containing 150mM 3-(N-morpholino)propane sulfonic
m/z: (M+Na)+ 691.1801 (Calcd for C37H32O12Na: 691.1786).
acid pH 7.2 (MOPS), 75mM MnCl2, 5mM dithiothritol (DTT),
(E)-1-[3,5-Bis[(E)-3-(3,4-dihydroxyphenyl)prop-2- 25% glycerol and 500µg/mL bovine serum albumin], 1µL of
enoyl]-2,4,6-trimethoxyphenyl]-3-(3,4-dihydroxyphenyl)- 5pmol/mL digoxigenin-labelled target DNA and 32µL of ster-
prop-2-enone (7f): Orange solid (10mg, 2.4% yield), mp ilized water were added into each well of a 96-well plate. Sub-
1
302–304°C; H-NMR (DMSO-d6) δ: 7.35 (3H, d, J=16.0Hz, sequently, 6µL of a sample solution and 9µL of 1/5 dilution of
H-β),7.14 (3H, s, ArH), 7.07 (3H, d, J=8.4Hz, ArH), 6.86 integrase enzyme were added to the mixture in the designated
(3H, d, J=16.0Hz, H-α), 6.79 (3H, d, J=8.4Hz, ArH), 3.60 well and incubated at 37°C for 80min. The content in each
(9H, s, –OCH3); 13C-NMR (DMSO-d6) δ: 191.9, 155.6, 149.3, well was washed with PBS four times, and then 100µL of
146.9, 145.7, 126.4, 124.6, 124.2, 122.6, 115.8, 114.9, 62.9; 500mU/mL alkaline phosphatase (AP) labelled anti-digoxigen-
HR-MS (ESI–methanol) m/z: (M+Na)+ 677.1630 (Calcd for in antibody was added prior to further incubation at 37°C for
C36H30O12Na: 677.1629).
1h. The mixture was washed again with washing buffer con-
IN Inhibition Assay. Enzyme HIV-1 IN protein was taining 0.05% Tween 20 in PBS four times and then with plain
kindly provided by Dr. Robert Craigie (the National Institute PBS four times. Then, AP buffer (150µL) containing 100mM
of Health, Bethesda, Maryland, U.S.A.), and stored at −80°C Tris–HCl (pH 9.5), 100mM NaCl, 5mM MgCl2 and 10mM
before use.
p-nitrophenyl phosphate was added to each well followed by
Oligonucleotide Substrates Oligonucleotides of long incubation at 37°C for 1h. Finally, the absorbance of each
terminal repeated bases from donor DNA (LTRD) and from solution on the plate was measured with a microplate reader
target substrate DNA (TS) were purchased from QIAGEN at the wavelength of 405nm. A control composed of the simi-
Operon, U.S.A., and stored at −25°C before use. The se- lar reaction mixture with 50% DMSO in place of the sample
quence of biotinylated LTRD and its unlabelled complement solution, while a blank was buffer-E containing 20mM MOPS
were 5′-biotin-ACCCTTTTAGTCAGTGTGGAAAATCTC (pH 7.2), 400mM potassium glutamate, 1mM ethylenediami-
TAGCAGT-3′ (LTR-D1) and 3′-GAAAATCAGTCACACCTT netetraacetate disodium salt (EDTA·2Na), 0.1% Nonidet-P 40
TTAGAGATCGTCA-5′ (LTR-D2), respectively, while those (NP-40), 20% glycerol, 1mM DTT and 4M urea without the
of the target substrate DNA (digoxigenin-labelled target DNA, integrase enzyme. Suramin, a polyanionic HIV-1 IN inhibitor
TS-1) and its 3′-labelled complement were 5′-TGACCAAGG was used as a positive control. Percentage Inhibition against
G CTAATTCACT-digoxigenin and digoxigenin-ACTGGT HIV-1 IN could be calculated from the equation below:
TCCCGATTAAGTGA-5′ (TS-2), respectively.
ODcontrol −ODsample


Annealing of the Substrate DNA LTR-D1 and LTR-D2,
and TS-1 and TS-2 were mixed separately and then the for-
mer solution was diluted to a concentration of 2pmol/mL,
% inhibition against HIV-1 IN=
×100


ODcontrol


while the latter one was made to 5pmol/mL by diluting with where OD=absorbance detected from each well.
KTE (a buffer solution containing 10mM Tris–HCl (pH 8.0),
The results of anti-HIV-1 IN activity were expressed as

762
Vol. 62, No. 8
3) Mintzer M. A., Dane E. L., O’Toole G. A., Grinstaff M. W., Mol.
Pharm., 9, 342–354 (2012).
4) Hennrich G., Anslyn E. V., Chem. Eur. J., 8, 2219–2225 (2002).
5) Zysman-Colman E., Denis C., Coord. Chem. Rev., 256, 1742–1761
(2012).
6) Iverson D. J., Hunter G., Blount J. F., Damewood J. R. Jr., Mislow
K., J. Am. Chem. Soc., 103, 6073–6083 (1981).
7) Al-Mawasawi L. Q., Al-Safi R. I., Neamati N., Expert Opin. Emerg.
Drugs, 13, 213–225 (2008).
8) “HIV-1 Integrase: Mechanism and Inhibitor Design,” ed. by Neama-
ti N., Wiley, Hoboken, 2011.
9) Fesen M. R., Kohn K. W., Leteurtre F., Pommier Y., Proc. Natl.
Acad. Sci. U.S.A., 90, 2399–2403 (1993).
10) Fesen M. R., Pommier Y., Leteurtre F., Hiroguchi S., Yung J., Kohn
K., Biochem. Pharmacol., 48, 595–608 (1994).
11) Lee J. Y., Yoon K. J., Lee Y. S., Bioorg. Med. Chem. Lett., 13,
4331–4334 (2003).
mean S.D. from four determinations. The IC50 values were
calculated using the Microsoft Excel® program.
Molecular Modeling. Structure of Inhibitor The 3D
structures of the multi-cinnamoyl compounds 5a–a, 6a–e and
7a–f were generated using Hyperchem professional 8.0. Since
the hexasubstituted benzene could have their substituents ar-
ranged into eight possible conformations based on their posi-
tions relative to the benzene ring whether they were “above”
or “below” the benzene plane,6) compound 7a was selected as
a model to identify the lowest energy conformation by quan-
tum mechanical method at B3LYP/6-31G(d) level of theory
using the Gaussian 03 program. The obtained lowest energy
conformation of 7a was then used as a guideline to build the
appropriate conformers for all other compounds. Subsequent-
ly, geometry optimization at the B3LYP/6-31G(d) level was
performed for each compound. Finally, the resulting structures
were applied for docking calculations.
12) Charvat T. T., Lee D. J., Robinson W. E., Chamberlin A. R., Bioorg.
Med. Chem., 14, 4552–4567 (2006).
Structure of Enzyme As it is known that an inhibi-
tor binds to HIV-1 IN at the catalytic core domain, the only
available X-ray structure of IN core domain in the complex
13) Crosby D. C., Robinson W. E. Jr., “HIV-1 Integrase: Mechanism
and Inhibitor Design,” ed. by Neamati N., Wiley, Hoboken, 2011,
pp. 341–362.
14) Artico M., Di Santo R., Costi R., Novellino E., Greco G., Massa S.,
Tramontano E., Marongiu M. E., De Montis A., La Colla P., J. Med.
Chem., 41, 3948–3960 (1998).
15) Costi R., Di Santo R., Artico M., Massa S., Ragno R., Loddo R., La
Colla M., Tramontano E., La Colla P., Pani A., Bioorg. Med. Chem.,
12, 199–215 (2004).
16) Burke T. R. Jr., Fesen M. R., Mazumder A., Wang J., Carothers
A. M., Grunberger D., Driscoll J., Kohn K., Pommier Y., J. Med.
Chem., 38, 4171–4178 (1995).
17) Deng J., Sanchez T., Al-Mawsawi L. Q., Dayam R., Yunes R. A.,
Garofalo A., Bolger M. B., Neamati N., Bioorg. Med. Chem., 15,
4985–5002 (2007).
18) De Clercq E., J. Med. Chem., 38, 2491–2517 (1995).
19) Zolopa A. R., Antiviral Res., 85, 241–244 (2010).
20) Veber D. F., Johnson S. R., Cheng H. Y., Smith B. R., Ward K. W.,
Kopple K. D., J. Med. Chem., 45, 2615–2623 (2002).
21) Tewtrakul S., Nakamura N., Hattori M., Fujiwara T., Supavita T.,
Chem. Pharm. Bull., 50, 630–635 (2002).
22) Hazuda D. J., Hastings J. C., Wolfe A. C., Emini E. A., Nucleic
Acids Res., 22, 1121–1122 (1994).
23) Dean F. M., Robertson A., J. Chem. Soc., 1241–1249 (1953).
24) Göschke A., Tambor J., Ber. Dtsch. Chem. Ges., 45, 1237–1239
(1912).
with
1-(5-chloroindol-3-yl)-3-hydroxy-3-(2H-tetrazol-5-yl)-
propenone (5CITEP) was used in our studies.37) The structure
was obtained from the protein data bank (PDB) code 1QS4.
Since 5CITEP is bound to chain A, only chain A was selected
for the current study. All water molecules and 5CITEP were
removed while a magnesium ion at the active site was main-
tained. Finally, missing residues and hydrogen atoms were
added. The protonation state of Lys156 and Lys159 were set to
protonated forms as identified by Nunthaboot et al.38)
Molecular Docking Molecular docking calculations were
performed using version 4.1.1 of the GOLD software package.
To define the binding site, all atoms within 10Å around the
Mg2+ ion were chosen. The CHEMPLP scoring function was
used. The GA run was 50. All other parameters were main-
tained at their default GOLD software settings. All conform-
ers of the compounds were docked into the binding site of the
IN. Based on the GOLD fitness score for each molecule, a
bound structure with the highest fitness score was considered
as the best bound conformation. In order to validate the dock-
ing procedure, 5CITEP was re-docked back into the IN. The
docked structure agrees well with the X-ray complex structure
(1QS4).
25) Hansen P. E., Kamounah F. S., Zhiryakova D., Manolova Y., An-
tonov L., Tetrahedron Lett., 55, 354–357 (2014).
26) Choksakulporn S., “Synthesis of Hexasubstituted Benzene Deriva-
tives from Phloroglucinol,” M.S. thesis, Chulalongkorn University,
Thailand, 2005.
27) Simaan S., Siegel J. S., Biali S. E., J. Org. Chem., 68, 3699–3701
(2003).
28) Jin F., Jin X. Y., Jin Y. L., Sohn D. W., Kim S.-A., Sohn D. H., Kim
Y. C., Kim H. S., Arch. Pharm. Res., 30, 1359–1367 (2007).
29) Boumendjel A., Boccard J., Carrupt P.-A., Nicolle E., Blanc M.,
Geze A., Choisnard L., Wouessidjewe D., Matera E.-L., Dumontet
C., J. Med. Chem., 51, 2307–2310 (2008).
30) Aponte J. C., Verástegui M., Málaga E., Zimic M., Quiliano M.,
Vaisberg A. J., Gilman R. H., Hammond G. B., J. Med. Chem., 51,
6230–6234 (2008).
31) Climent M. J., Corma A., Iborra S., Primo J., J. Catal., 151, 60–66
(1995).
32) Yao N., Song A., Wang X., Dixon S., Lam K. S., J. Comb. Chem., 9,
668–676 (2007).
33) Neves M. P., Cravo S., Lima R. T., Vasconcelos M. H., Nascimento
M. S. J., Silva A. M. S., Pinto M., Cidade H., Corrêa A. G., Bioorg.
Acknowledgments We are very grateful to Dr. Robert
Craigie, National Institute of Health, Bethesda, Maryland,
U.S.A., for kindly providing the HIV-1 integrase enzyme.
We also thank Dr. Christopher Smith for his assistance on
language editing during manuscript preparation. The re-
search funds from the National Research University project
of CHE and the Ratchada phiseksomphot Endowment Fund
(HR1155A), Chulalongkorn University, are highly appreciated.
W. T. acknowledges the scholarship from Suan Dusit Rajabhat
University. S. C. would like to thank the Development and
Promotion of Science and Technology Talents Project (DPST)
for the scholarship during the study.
References
1) Choi S.-K., “Synthetic Multivalent Molecules: Concepts and Bio-
medical Applications,” Wiley, New York, 2004.
2) Vance D., Shah M., Joshi A., Kane R. S., Biotechnol. Bioeng., 101,
429–434 (2008).

August 2014
Med. Chem., 20, 25–33 (2012).
763
Pharm. Bull., 55, 64–71 (2007).
34) Sotriffer C. A., Ni H., McCammon J. A., J. Med. Chem., 43, 4109–
4117 (2000).
35) Bodiwala H. S., Sabde S., Gupta P., Mukherjee R., Kumar R., Garg
P., Bhutani K. K., Mitra D., Singh I. P., Bioorg. Med. Chem., 19,
1256–1263 (2011).
37) Goldgur Y., Craigie R., Cohen G. H., Fujiwara T., Yoshinaga T.,
Fujishita T., Sugimoto H., Endo T., Murai H., Davies D. R., Proc.
Natl. Acad. Sci. U.S.A., 96, 13040–13043 (1999).
38) Nunthaboot N., Pianwanit S., Parasuk V., Kokpol S., Wolschann P.,
J. Mol. Struct., 844–845, 208–214 (2007).
36) Handler N., Jaeger W., Puschacher H., Leisser K., Erker T., Chem.