Synthesis, biological evaluation and molecular modeling of 2-Hydroxyisoquinoline-1,3-dione analogues as inhibitors of HIV reverse transcriptase associated ribonuclease H and polymerase

Article abstract of DOI:10.1016/j.ejmech.2017.03.059

Human immunodeficiency virus (HIV) reverse transcriptase (RT) associated ribonuclease H (RNase H) remains the only virally encoded enzymatic function not clinically validated as an antiviral target. 2-Hydroxyisoquinoline-1,3-dione (HID) is known to confer

Full text of DOI:10.1016/j.ejmech.2017.03.059

Accepted Manuscript
Synthesis, biological evaluation and molecular modeling of 2-
Hydroxyisoquinoline-1,3-dione analogues as inhibitors of HIV reverse transcriptase
associated ribonuclease H and polymerase
Jing Tang, Sanjeev Kumar V. Vernekar, Yue-Lei Chen, Lena Miller, Andrew D. Huber,
Nataliya Myshakina, Stefan G. Sarafianos, Michael A. Parniak, Zhengqiang Wang
PII:
S0223-5234(17)30226-X
10.1016/j.ejmech.2017.03.059
EJMECH 9319
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 17 November 2016
Revised Date: 21 March 2017
Accepted Date: 24 March 2017
Please cite this article as: J. Tang, S.K.V. Vernekar, Y.-L. Chen, L. Miller, A.D. Huber, N. Myshakina,
S.G. Sarafianos, M.A. Parniak, Z. Wang, Synthesis, biological evaluation and molecular modeling
of 2-Hydroxyisoquinoline-1,3-dione analogues as inhibitors of HIV reverse transcriptase associated
ribonuclease H and polymerase, European Journal of Medicinal Chemistry (2017), doi: 10.1016/
j.ejmech.2017.03.059.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract

ACCEPTED MANUSCRIPT
Synthesis, Biological Evaluation and Molecular Modeling of 2-
Hydroxyisoquinoline-1,3-dione Analogues as Inhibitors of HIV
Reverse Transcriptase Associated Ribonuclease H and Polymerase
†
a
†a
ξ
a
b
c
Jing Tang , Sanjeev Kumar V. Vernekar , Yue-Lei Chen , Lena Miller , Andrew D. Huber ,
d
c
b
a
Nataliya Myshakina , Stefan G. Sarafianos , Michael A. Parniak and Zhengqiang Wang*
a
Center for Drug Design, Academic Health Center, University of Minnesota, Minneapolis, MN 55455, USA
b
Department of Microbiology & Molecular Genetics, University of Pittsburgh School of Medicine, Pittsburgh, PA 15219, USA
c
Department of Molecular Microbiology and Immunology and Department of Biochemistry, University of Missouri School
of Medicine, Christopher S. Bond Life Sciences Center, Columbia, MO 65211, USA
d
Department of Natural Science, Chatham University, 1 Woodland Road, Pittsburgh, PA, 15232, USA
Footnote
†
ξ
These authors contributed equally.
Present address: Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Room
407, 555 Zuchongzhi Road, Shanghai 201203, PR China.
5
*
6
To whom correspondence should be addressed. Email: wangx472@umn.edu; Phone: +1 (612)
26-7025; Fax: +1 (612) 625-8154.

ACCEPTED MANUSCRIPT
Abstract: Human immunodeficiency virus (HIV) reverse transcriptase (RT) associated
ribonuclease H (RNase H) remains the only virally encoded enzymatic function not clinically
validated as an antiviral target. 2-Hydroxyisoquinoline-1,3-dione (HID) is known to confer
active site directed inhibition of divalent metal-dependent enzymatic functions, such as HIV
RNase H, integrase (IN) and hepatitis C virus (HCV) NS5B polymerase. We report herein the
synthesis and biochemical evaluation of a few C-5, C-6 or C-7 substituted HID subtypes as HIV
RNase H inhibitors. Our data indicate that while some of these subtypes inhibited both the
RNase H and polymerase (pol) functions of RT, potent and selective RNase H inhibition was
achieved with subtypes 8–9 as exemplified with compounds 8c and 9c.
Keywords: 2-hydroxyisoquinoline-1,3-diones, HIV, RNase H, reverse transcriptase, inhibitor
1
. Introduction
HIV encodes three enzymes required for viral replication: RT, IN and protease (PR) [1].
Inhibitors of these enzymes constitute the cornerstone of highly active antiretroviral therapy
HAART) [2] which provide the only means for clinically managing HIV. Particularly
(
interesting is the RT which consists of two distinct enzymatic domains: the pol domain and the
RNase H domain. The primary function of the RNase H domain is to selectively degrade the
intermediate RNA/DNA heteroduplex [3]. Interestingly, while RT pol has been extensively
targeted by many nucleoside RT inhibitors (NRTIs) [4] and nonnucleoside RT inhibitors
(NNRTIs) [5-7], drug discovery efforts targeting the RNase H function have not produced a
single candidate in development pipeline[8]. Nevertheless, the attenuated biochemical activity of
RNase H resulted from active site mutation was found to correlate closely with reduced HIV
replication in cell culture[9], strongly suggesting that effective RNase H inhibition should confer

ACCEPTED MANUSCRIPT
the desired antiviral phenotype. Due to their unique mechanism of action, antivirals targeting
RNase H are expected to inhibit HIV strains resistant to known drugs.
RNase H adopts an active site fold that defines a family of homologous enzymes termed
retroviral integrase super family (RISF) [10] . Specifically, its catalytic function hinges on a
DEDD motif at active site that can engage with two divalent metal ions.[11] Consequently,
RNase H inhibition typically entails a chemical core capable of chelating two divalent metals, a
pharmacophore feature shared by inhibitors of two homologous enzymes: HIV IN and HCV
NS5B polymerase. The HID core contains a perfectly aligned chelating triad around its 6-
membered ring (1,[11] Fig. 1a), thus has been widely explored as inhibitors of HIV RNase H (1
[11], 2 [12, 13], 3 [14], Fig. 1a), HIV IN (2[15], Fig. 1a) and HCV NS5B polymerase (4—5
[16], Fig. 1b). Given that HID subtypes have been reported to potently inhibit HIV RNase H, we
were prompted to study the RNase H inhibitory activity of the two HID variants (4—5)
previously reported as HCV NS5B inhibitors. Our current study also extends to previously
unknown aryl substituted HID regiosiomers (6—7), 5-arylmethyl HID analogues (8) and 6-
arylcarboxamide substituted HID analogues (9, Fig.1b). Regioisomeric HID scaffold of 8 has
been reported to potently inhibit HIV RNase H and pol [14], and that of 9 was found to inhibit
HIV IN [12].
Insert Figure 1
2
. Results and Discussion
2
.1 Chemistry
HID subtypes 4 and 5 were synthesized based on our previously reported procedures
(Schemes 1–2) [16]. Both routes feature a synthetic sequence consisting of a Sandmeyer reaction
[17] on an amino homophthalic acid compound (7 in Scheme 1 and 14 in Scheme 2) to prepare

ACCEPTED MANUSCRIPT
the iodo intermediate (8 in Scheme 1 and 15 in Scheme 2), a Suzuki coupling reaction[18] to
install the aryl group (9 in Scheme 1 and 16 in Scheme 2), and a cyclization to construct the HID
ring (10 in Scheme 1 and 17 in Scheme 2).
Insert Scheme 1
The two routes differ in that the requisite amino homophthalic acid compound 7 is
commercially available for the synthesis of 7-substituted HID (subtype 4), whereas the
regioisomer (14) for the synthesis of 6-substituted HID (subtype 5) was prepared in three steps
featuring a Hurtley reaction [19] (Scheme 2, 11 to 12).
Insert Scheme 2
The synthesis of subtypes 6—8 all started with a halogenated indanone (21 for subtype 6 and
2
5 for subtypes 7—8). Following a reported procedure [20], acylation of the indanone with
diethyl oxalate and the subsequent oxidation with alkaline hydrogen peroxide produced key
homophthalic acid intermediates 23, 26 and 31 (Schemes 3—5). The introduction of the aryl
group was carried out right before (for subtypes 6 and 8, Schemes 3 and 5) or after (for subtype
7
, Scheme 4) the formation of the homophthalic acid. The construction of HID from the
homophthalic acid was effected using the same synthetic sequence as for subtypes 4—5.
Insert Scheme 3
Insert Scheme 4
Insert Scheme 5
Additionally, subtype 9 in which the 6-aryl moiety is separated by an amide linkage was also
prepared. C-7 regioisomers of these analogues were reported to inhibit HIV IN. The synthesis of
9
was effected via a simple acylation of the amido diacid intermediate 17 followed by the HID
construction (Scheme 6).

ACCEPTED MANUSCRIPT
Insert Scheme 6
.2 Biology
2
All synthesized HID analogues were evaluated for biochemical inhibition of both the RNase
H and pol functions of HIV RT, and for antiviral activity in a cytoprotection assay. The RNase H
assay was conducted using an 18 bp oligonucleotide duplex as described previously [14, 21]. A
known inhibitor THBNH [22] was used as the positive control in all assays. This substrate
measures the random and internal cleavage which is believed to represent the major mode of
RNA cutting. Detailed biochemical assay results are summarized in Table 1. Overall, all HID
subtypes tested (4–9) inhibited RNase H in sub to low micromolar range (IC = 0.20–2.0 µM).
5
0
The position of the aryl substitution did not appear to impact RNase H inhibition significantly, as
shown with analogues 4a (C-7, IC = 1.2 µM), 5a (C-6, IC = 1.1 µM) and 7a (C-5, IC = 2.0
5
0
50
50
µM). For the C-6 aryl substituted analogues, introduction of an additional methoxy group at C-7
slightly decreased RNase H inhibition (6a vs 5a, Table 1). Interestingly, for the C-5 substituted
analogues, introducing a methylene linker led to increased RNase H inhibition (7c vs 8c, Table
1
). It is noteworthy that three of the four subtype 8 inhibited RNase H in submicromolar range
(8b–8d). The most potent RNase H inhibition, however, was observed with subtype 9. As shown
in Table 1, all HID analogues with an amide linkage (9a–9d) inhibited RNase H at nanomolar
concentrations (IC50 = 0.20–0.30 µM), reflecting a 5–10 fold of improvement in potency over the
ones without linkage (subtypes 4–7). As for RT pol inhibition, analogues of subtypes 4–5 and 9
all potently inhibited the pol function, resulting in virtually no selectivity between RNase H and
pol inhibition for 4–5, and modest selectivity for 9 (2–7 fold). A significantly lower level of RT
pol inhibition was observed with subtypes 6 and 8, amounting to a selectivity of up to 6 fold and
1
2 fold, respectively. Two of the three analogues of subtype 7 exhibited only weak inhibition (7b

ACCEPTED MANUSCRIPT
and 7c, Table 1) against pol. However, their biochemical inhibition against RNase H was
substantially less potent than subtypes 8–9. These results suggest that the 5-arylmethyl
substituted subtype (8) and the 6-arylamido substituted subtype (9) represent the most promising
scaffolds for potently and selectively inhibiting HIV RT associated RNase H, as shown with
compounds 8c (IC = 0.8 µM, selectivity = 12 fold) and 9c (IC = 0.2 µM, selectivity = 7 fold)
5
0
50
both of which inhibited HIV RNase H with comparable potency to and better selectivity than
THBNH (Table 1).
Insert Table 1
All these analogues were also screened in a previously reported cytoprotection antiviral assay
[23]. Unfortunately, no significant reduction in cytopathic effect (CPE) was observed at 10 µM
with any analogue described herein (data not shown).
2
.3 Molecular modeling
To better understand the observed dual inhibition against RT RNase H and pol, we conducted
extensive molecular modeling analysis with two representative compounds 8c and 9c to probe
their predicted binding to the active sites of RNase H and pol as well as the NNRTI binding site
of the pol.
Insert Figure 2
RNase H active site. Computer modeling shows two major binding modes for compounds 8c
and 9c in RNase H active center. We compare these binding modes with a position of LP8 (3-
cyclopentyl-1,4-dihydroxy-1,8-naphthyridin- 2(1H)-one) co-crystallized in 3LP1 structure (Fig.
2
a). Compounds 8c, 9c and LP8 all bear a chelating triad of electronegative atoms, thus similar
metal coordination mode could be expected for these compounds. Structural analysis shows
significant differences in the overall position of the ligands as well as in metal coordination for

ACCEPTED MANUSCRIPT
the first binding mode (Fig. 2) which has the highest docking scores for both compounds (-8.706
kcal/mol for 8c and -8.428 kcal/mol for 9c).
Insert Figure 3
The second binding mode simulated for 8c and 9c is represented by poses with lower docking
scores (-7.588 kcal/mol for 8c and -6.795 kcal/mol for 9c), but has more similarity in overall
pose and metal coordination pattern to LP8 ligand (Fig. 3a). Overall docking poses and per-
residue interaction scores (Fig. 3a and 3c) show that both compounds create similar contacts with
surrounding amino acid side chains. Arg557 is involved in pi-pi stacking interactions with
hydroxyisoquinoline rings of 8c and 9c. The phenyl ring of compound 9c also has an additional
pi-pi stacking interaction with Arg448 (Fig. 3a), an interaction absent for compound 8c. Another
contact, by which 8c and 9c are differ by, is Gln475 side chain which is about 2.4Å from
compound 8c, but further away from 9c since the substituent amide linker and phenyl ring of 9c
face opposite directions. To evaluate Arg448 contribution to overall binding of compound 9c and
8
c to RNase H active center we mutated Arg448 to alanine in silico and docked both compounds
to modified RT structure. RNase H active site is shallow and wide which limits contacts an
active site inhibitor could make with surrounding residues while effectively chelating metal
cations. Thus, elimination of bulky Arg side chain, which guanidinium group is capable for pi-pi
stacking interactions with aromatic moiety of 9c, would affect binding and docking scores.
Results obtained for simulations involving mutated structure were quite unexpected. For both
compounds 8c and 9c lower docking scores were calculated which indicates better binding to
R448A mutant. In absence of long and bulky Arg448, both compounds bound close to the loop
formed by residues 444-448 and thus created more contacts which resulted in more negative
docking scores (see Supplemental information). Based on this theoretical finding we allow full

ACCEPTED MANUSCRIPT
flexibility of Arg448 and adjacent residue side chains and re-docked both compounds into RNase
H active center. This docking experiment did result in more contacts with site chain residues and
more negative docking scores (Fig. 3b and 3c). Large and negative interaction scores for Arg448
and Arg557 (Fig. 3c) are due large negative Coulombic term defining interactions between
positively charged Arg and negatively charged molecules of 8c and 9c. Docking into structure
with flexible Arg448 side chain results in pose where less flexible compound 8c moves away
from Arg448 (Fig. 3b). Besides potential for strong pi-pi stacking interaction between compound
9
c and Arg448, our modeling data does not show significant superiority of 8c over 9c in binding
to the RNase H active site. Thus substantial difference in RNase H inhibition is not expected.
Indeed, the biochemical data for both compounds differ by less than one order of magnitude and
such a difference is probably too subtle to be picked up by molecular modeling.
Insert Figure 4
Polymerase active site. Docking results show that both compounds can bind into the RT
polymerase active center without any steric hindrance (Fig. 4 insert) and close to residues Lys65,
Arg72, and Gln151, residues which play key roles in the polymerization activity of HIV-1
RT.[24-27] However, for 8c and 9c in the polymerase active site, inhibition through a
competitive mode of action is unlikely. In our polymerase assays, RT is pre-incubated with
primer-template duplex for which RT has a substantially greater binding affinity than for small
molecules such as 8c and 9c. The latter are unlikely to compete for binding with the nucleic acid
duplex, and thus cannot really block the polymerase active site. Compounds 8c and 9c likely
bind to the RT-primer-template complex, creating a steric clash with the template strand since 8c
and 9c are longer molecules compared to the natural nucleotide substrates (Fig. 4).
Insert Figure 5

ACCEPTED MANUSCRIPT
NNRTI Binding Pocket. Inhibition of RT polymerase activity by compounds 8c and 9c most
likely results from allosteric inhibition due to binding to NNRTI binding pocket. However, the
NNRTI binding pocket presents a challenge for computational simulations since the pocket does
not exist prior to inhibitor binding. NNRTIs constitute a broad variety of chemical structures, yet
they all bind at the same hydrophobic allosteric pocket [5, 28, 29]. This binding extends the p66
thumb domain due to conformational changes in amino acid residues Tyr181 and Tyr188. The
conformational change of RT upon NNRTI binding would be too global to be successfully
picked up by present docking algorithms for induced fit binding. Thus, we considered a crystal
structure in which the NNRTI pocket is already present. We chose structure 1KLM with co-
crystalized delavirdine (Fig. 5b), a compound with more structural similarities to our studied
compounds than other NNRTIs.
Our modeling results show that both 8c and 9c bind better to the NNRTI pocket in protonated
neutral form. Calculated docking scores for neutral 8c and 9c are -9.488 kcal/mol and -9.797
kcal/mol respectively while the best binding scores are predicted at -8.994 kcal/mol for
deprotonated 8c and -8.881 kcal/mol for deprotonated 9c. Lower binding of 8c and 9c
deprotonated forms could be explained by the significant hydrophobicity of the NNRTI pocket.
Calculated docking scores for 8c and 9c are lower than the score calculated for delavirdine (-
1
1.209 kcal/mol). Both 8c and 9c bind very similarly to delavirdine (Fig. 5a) in proximity to
Tyr181, Tyr188, Asn103, and Trp229. Both the value of the docking scores and the larger
number of van der Waals contacts (Fig. 5c) indicates that 9c interacts more strongly with
residues in the NNRTI binding pocket, which corroborates the observation that 9c inhibited RT
polymerase with about 10-fold greater potency than 8c. Difference in RT polymerase inhibition
measured for 8c and 9c could be attributed to structural features of compound 9c which is more

ACCEPTED MANUSCRIPT
extended and has a more hydrophobic non-substituted phenyl ring compared with 8c, attributes
that might favor hydrophobic interactions. Other crystal structures with co-crystallized NNRTIs
were explored, however those docking simulations did not indicate significant interactions with
side chains of Tyr181, Tyr188, Asn103, and Trp229 (additional data available in Supplemental
Information) .
Docking scores calculated for 8c and 9c in the NNRTI binding pocket are higher than those
calculated for binding into RNase H active site, yet the RNase H inhibition of both compounds
was found to be higher than the inhibition of RT polymerase. Several factors need to be taken
into consideration when correlating docking scores to inhibitory activities. First, our docking
experiments were performed with different RT structures, and docking scores cannot be directly
compared because they were calculated for different protein receptors. Second, more negative
docking scores are indicative of stronger binding; however, they also reflect number of
interactions between a ligand and protein. NNRTI binding pocket and RNase H active sites are
different in shape and size with NNRTI binding pocket being more confined and RNase H active
site being wide and shallow. This difference is reflected in the number of contacts with 8c and 9c
in RNase H active site (Fig. 3d) and NNRTI binding pocket (Fig. 5c). The key binding of metal
chelation at RNase H active site is often underestimated by scoring algorithms which results in
overall lower docking scores calculated when ligand-metal interactions are involved.
3
. Conclusion
A few HID subtypes were synthesized and tested for their ability to inhibit HIV RT-
associated RNase H. Interestingly, subtypes 4, 5 and 9 showed inhibitory activity against both
the RNase H and pol of HIV RT at sub to low micromolar concentrations, whereas subtypes 6
and 8 inhibited RT with modest selectivity toward RNase H inhibition and subtype 7

ACCEPTED MANUSCRIPT
predominantly inhibited RNase H. Extensive molecular modeling largely corroborated these
observations. Our studies identified analogues 8c (IC = 0.8 µM, selectivity = 12 fold) and 9c
5
0
(
IC50 = 0.2 µM, selectivity = 7 fold) as potent and selective inhibitors of HIV RNase H.
4
. Experimental part
4
.1. General - Chemistry
All commercial chemicals were used as supplied unless otherwise indicated. Dry solvents
(THF, toluene, and dioxane) were dispensed under argon from an anhydrous solvent system with
two packed columns of neutral alumina or molecular sieves. Flash chromatography was
performed on a Teledyne Combiflash RF-200 with RediSep columns (silica) and indicated
mobile phase. All moisture sensitive reactions were performed under an inert atmosphere of
1
13
ultra-pure argon with oven-dried glassware. H and C NMR spectra were recorded on a Varian
00 MHz spectrometer. Mass data were acquired on an Agilent TOF II TOS/MS spectrometer
6
capable of ESI and APCI ion sources. Analysis of sample purity was performed on a Varian
Prepstar SD-1 HPLC system with a Phenomenex Gemini, 5 micron C18 column (250mm x 4.6
mm). HPLC conditions: solvent A = H O, solvent B = MeCN; flow rate = 1.0 mL/min;
2
2
compounds were eluted with a gradient of 10% MeCN/H O to 100% MeCN for 30 min. Purity
was determined by total absorbance at 254 nm. All tested compounds have a purity ≥ 98%.
Compounds of subtypes 4 (4a–4e) and 5 (5a–5h) were previously reported [14].
4
.1.1. General procedure 1 for Suzuki coupling 22a-22d.
The mixture of 5-iodo-6-methoxy-1-indanone 21 (200 mg, 0.69 mmol), boronic acid (0.83
mmol, 1.2 equiv.), K CO (2.7 mmol, 4.0 equiv.), toluene/H O (9.5:0.5, 4.0 mL) and Pd[P(Ph) ]
3 4
2
3
2
(30 mg) were microwaved at 110 ºC for 30 min. The reaction was monitored by TLC. The black
residue formed was filtered and the filtrate was diluted with EtOAc. The EtOAc layer was

ACCEPTED MANUSCRIPT
2 4
washed with brine, the combined organic layers were dried over Na SO , filtered and
concentrated. The crude product was purified by flash column chromatography (EtOAc/hexane;
1
4
3
:9) to afford desired compound (22a-d).
.1.2. General procedure 2 for the synthesis of homophthalic acid derivatives 23a-23d, 26 and
1a-31d.
A freshly prepared solution of sodium methoxide (2.45 mmol, 2.0 equiv.) was added
dropwise to a solution of compound 22 (0.3 g, 1.22 mmol) and diethyl oxalate (1.96 mmol, 1.6
equiv.) in toluene (5 ml) at 0 °C. The mixture was stirred overnight at ambient temperature. The
solvent was evaporated to dryness and the residue was suspended in MeOH (4 mL) and solid
KOH (7.37 mmol, 6.0 equiv.) was added portion wise keeping the temperature below 50 °C and
stirred at room temperature for 3 h. Then hydrogen peroxide (30%, 12.2 mmol, 10.0 equiv.) was
added dropwise to the reaction mixture keeping the temperature below 50 °C and then stirred at
room temperature for 18 h. Gas was evolved during H O addition. The mixture was filtered, and
2
2
the filtrate was partially reduced in vacuo to remove MeOH. The remaining aqueous filtrate was
washed with ether, and the organic layer was discarded. The aqueous layer was acidified with
HCl until pH 2-3. The acidic aqueous layer was extracted with EtOAc (3 x 20 mL), the
combined organic layers were washed with brine, dried over Na SO , filtered and concentrated.
2
4
The crude product was recrystallized from ether to yield desired compound (23a-d), 26 and
31a-d) as a solid.
(
4
.1.3. General procedure 3 for cyclization and deprotection 6a-6d, 7a-7c and 8a-8d.
A solution of dicarboxylic acid 23 (0.1 g, 0.35 mmol, 1.0 equiv.) and NH OTHP (0.70 mmol,
2
2
.0 equiv.) in toluene (10 mL) was refluxed for 5 minutes. To the mixture, a solution of CDI
(0.28 mmol, 0.8 equiv.) in DCM was added dropwise. The suspension turned clear and stirred at

ACCEPTED MANUSCRIPT
reflux for 8 h, a black solid separated from solution, and the reaction monitored by TLC and MS.
The reaction mixture was passed through a short pad of silica gel which then rinsed with
(EtOAc/hexane; 1:3), the combined filtrate was evaporated to dryness. The product was used for
the next reaction without further purification. The cyclized product was dissolved in MeOH (5.0
mL) and treated with pTSA hydrate (0.35 mmol, 1.0 equiv.) and stirred at room temperature for
2
-3 h. The reaction was monitored by TLC and MS. The mixture was evaporated to dryness to
get pale yellow solid. The solid obtained was triturated with water and then with ether, and dried
at room temperature to afford a desired pure compound (6a-d), (7a-c) and (8a-d) as a solid.
4
.1.4. General procedure 4 for Suzuki coupling 27a-27c.
The mixture of 6-bromo homophthalic acid 26 (0.2 g, 0.77 mmol), boronic acid (1.55 mmol,
.0 equiv.), K CO (3.1 mmol, 4.0 equiv.), EtOH/ H O (9:1, 4.0 mL) and Pd[P(Ph) ] (30 mg)
2
2
3
2
3 4
were microwaved at 150 ºC for 30 min. The reaction was monitored by TLC. The black residue
formed was filtered and the EtOH was evaporated. The aqueous layer was acidified with HCl
until pH 2-3 and extracted with EtOAc (3 x 20 mL). The combined organic layers were dried
over Na
ether/ EtOAc to afford desired compound (27a-c) as a solid.
.1.5. Synthesis of 4-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)-2,3-dihydro-1H-inden-1-one
29).
The mixture of 4-Bromo-1-indanone 25 (1.0 g, 4.7 mmol, 1.0 equiv.), bis(pinacolato)diboron
1.8 g, 7.1 mmol, 1.5 equiv.), KOAc (0.93 g, 9.4 mmol, 2.0 equiv.), dioxane (15 mL) and
2 4
SO , filtered and concentrated. The crude product was purified by recrystallization with
4
(
(
PdCl (dppf) (0.1 g) were microwaved at 120 ºC for 20 min. The suspension was filtered and
2
2
washed with EtOAc. The solvent was evaporated and the residue was diluted with EtOAc and
washed with brine, organic layer was dried over Na SO , filtered and concentrated. The crude
2
4

ACCEPTED MANUSCRIPT
product was purified by flash column chromatography (EtOAc/hexane; 1:9) to afford desired
1
compound 29 (0.93 g, 3.6 mmol, 76%) as a yellow solid. H NMR (CDCl , 600 MHz) δ 8.04 (d,
3
J = 6.6 Hz, 1H), 7.84 (d, J = 7.8 Hz, 1H), 7.37 (t, J = 7.2 Hz, 1H), 3.34-3.33 (m, 2H), 2.73-2.66
(m, 2H), 1.29 (s, 12H).
4
.1.6. General procedure 5 for Suzuki coupling 30a-30d.
The mixture of boronic ester 29 (0.5 g, 1.9 mmol, 1.0 equiv.), ArCH Br (5.8 mmol, 3.0
2
2 3 2 3 4
equiv.), Na CO (3.8 mmol, 2.0 equiv.), THF/H O (9:1, 10.0 mL) and Pd[P(Ph) ] (30 mg) were
microwaved at 80 ºC for 45 min. The reaction was monitored by TLC. The solvent was
evaporated and the residue was diluted with EtOAc and washed with brine, organic layer was
dried over Na SO , filtered and concentrated. The crude product was purified by flash column
2
4
chromatography (EtOAc/hexane; 1:9) to afford desired compound (30a-d).
.1.7. 5-(Furan-2-yl)-6-methoxy-2,3-dihydro-1H-inden-1-one (22a). Yield 72%. H NMR
CDCl , 600 MHz) δ 7.90 (s, 1H), 7.51 (s, 1H), 7.22 (s, 1H), 7.12 (s, 1H), 6.51 (s, 1H), 3.94 (s,
1
4
(
3
3
H), 3.07-3.05 (m, 2H), 2.70-2.68 (m, 2H).
.1.8. 2-(Carboxymethyl)-4-(furan-2-yl)-5-methoxybenzoic acid (23a).Yield 51%. H NMR
1
4
(DMSO-d6, 600 MHz) δ 12.43 (br, 2H), 7.79 (s, 1H), 7.66 (s, 1H), 7.55 (s, 1H), 7.05 (s, 1H),
6
.62 (s, 1H), 3.94 (s, 3H), 3.91 (s, 2H).
.1.9. 6-(Furan-2-yl)-2-hydroxy-7-methoxyisoquinoline-1,3(2H,4H)-dione (6a). Yield 42%. H
1
4
NMR (DMSO-d6, 600 MHz) δ 10.37 (s, 1H), 7.84 (s, 1H), 7.71 (s, 1H), 7.58 (s, 1H), 7.13 (s,
1
3
1
1
H), 6.64 (s, 1H), 4.22 (s, 2H), 3.98 (s, 3H); C NMR (DMSO-d6, 150 MHz) δ 166.9, 161.8,
54.2, 148.5, 144.0, 127.4, 124.3, 124.1, 113.6, 112.9, 109.7, 56.3, 36.7; HRMS-ESI(-) m/z calcd
-
for C14
H10NO
5
272.0559 [M-H] , found 272.0596.

ACCEPTED MANUSCRIPT
1
4
6
3
4
3
.1.10. 6-Methoxy-5-phenyl-2,3-dihydro-1H-inden-1-one (22b). Yield 68%. H NMR (CDCl ,
00 MHz) δ 7.53-7.52 (m, 2H), 7.44-7.42 (m, 2H), 7.39-7.37 (m, 2H), 7.29 (s, 1H), 3.83 (s, 3H),
.11-3.09 (m, 2H), 2.74-2.72 (m, 2H).
1
.1.11. 5-(Carboxymethyl)-2-methoxy-[1,1'-biphenyl]-4-carboxylic acid (23b).Yield 54%. H
NMR (DMSO-d6, 600 MHz) δ 12.21 (br, 2H), 7.56 (s, 1H), 7.51-7.49 (m, 2H), 7.44-7.42 (m,
2
H), 7.38-7.36 (m, 1H), 7.27 (s, 1H), 3.91 (s, 2H), 3.80 (s, 3H).
.1.12. 2-Hydroxy-7-methoxy-6-phenylisoquinoline-1,3(2H,4H)-dione (6b). Yield 32%. H NMR
1
4
(DMSO-d6, 600 MHz) δ 10.21 (s, 1H), 7.59 (s, 1H), 7.50 (d, J = 7.2 Hz, 2H), 7.44-7.36 (m, 3H),
1
3
7
1
.32 (s, 1H), 4.20 (s, 2H), 3.83 (s, 3H); C NMR (DMSO-d6, 150 MHz) δ 166.9, 162.0, 155.7,
37.0, 135.8, 130.2, 129.6, 128.6, 128.2, 127.4, 125.4, 109.6, 56.3, 36.7; HRMS-ESI(-) m/z calcd
-
for C H NO 282.0766 [M-H] , found 282.0798.
1
6
12
4
1
4
.1.13. 5-(4-Fluorophenyl)-6-methoxy-2,3-dihydro-1H-inden-1-one (22c). Yield 61%. H NMR
(
CDCl , 600 MHz) δ 7.52-7.50 (m, 2H), 7.38 (s, 1H), 7.29 (s, 1H), 7.14-7.11 (m, 2H), 3.85 (s,
3
3
4
6
7
4
5
1
1
H), 3.13-3.11 (m, 2H), 2.76-2.74 (m, 2H).
.1.14. 5-(Carboxymethyl)-4'-fluoro-2-methoxy-[1,1'-biphenyl]-4-carboxylic acid (23c).Yield
1
4%. H NMR (DMSO-d6, 600 MHz) δ 12.32 (br, 2H), 7.53-7.52 (m, 3H), 7.35 (s, 1H), 7.26-
.23 (m, 2H), 3.81 (s, 3H), 3.77 (s, 2H).
.1.15. 6-(4-Fluorophenyl)-2-hydroxy-7-methoxyisoquinoline-1,3(2H,4H)-dione (6c). Yield
1
0%. H NMR (DMSO-d6, 600 MHz) δ 10.49 (s, 1H), 7.69 (s, 1H), 7.66-7.63 (m, 2H), 7.43 (s,
1
3
H), 7.37 (t, J = 8.4 Hz, 2H), 4.29 (s, 2H), 3.94 (s, 3H); C NMR (DMSO-d6, 150 MHz) δ
66.9, 161.9, 155.7, 134.6, 133.3, 131.7, 130.2, 127.4, 125.5, 115.6, 109.6, 56.3, 36.6; HRMS-
-
ESI(-) m/z calcd for C16
H11NFO
4
300.0672 [M-H] , found 300.0712.

ACCEPTED MANUSCRIPT
1
4
.1.16. 5-(2,4-Difluorophenyl)-6-methoxy-2,3-dihydro-1H-inden-1-one (22d). Yield 58%. H
NMR (CDCl , 600 MHz) δ 7.35 (s, 1H), 7.30 (s, 1H), 7.27 (s, 1H), 6.96-6.91 (m, 2H), 3.84 (s,
3
3
H), 3.12-3.11 (m, 2H), 2.76-2.75 (m, 2H).
4
.1.17. 5-(Carboxymethyl)-2',4'-difluoro-2-methoxy-[1,1'-biphenyl]-4-carboxylic acid (23d).
1
Yield 61%. H NMR (CD OD, 600 MHz) δ 7.55 (s, 1H), 7.41 (d, J = 7.2 Hz, 1H), 7.31 (s, 1H),
3
7
4
3
.21 (s, 1H), 7.17-7.16 (m, 1H), 3.89 (s, 2H), 3.78 (s, 3H).
.1.18. 6-(2,4-Difluorophenyl)-2-hydroxy-7-methoxyisoquinoline-1,3(2H,4H)-dione (6d). Yield
1
9%. H NMR (DMSO-d6, 600 MHz) δ 10.41 (s, 1H), 7.61 (s, 1H), 7.43-7.39 (m, 1H), 7.34-7.30
1
3
(
m, 1H), 7.28 (s, 1H), 7.17 (t, J = 8.4 Hz, 1H), 4.19 (s, 2H), 3.82 (s, 3H); C NMR (DMSO-d6,
1
5
4
50 MHz) δ 166.8, 161.9, 156.1, 133.0, 130.8, 129.5, 127.2, 126.4, 121.3, 111.9, 109.3, 104.5,
-
6.4, 36.6; HRMS-ESI(-) m/z calcd for C H NF O 318.0578 [M-H] , found 318.0607.
1
6
10
2
4
1
.1.19. 3-Bromo-2-(carboxymethyl)benzoic acid (26).Yield 56%. H NMR (DMSO-d6, 600
MHz) δ 12.61 (br, 2H), 7.61-7.58 (m, 2H), 7.32 (t, J = 7.8 Hz, 1H), 3.36 (s, 2H).
1
4
.1.20. 2-(Carboxymethyl)-3-(furan-2-yl)benzoic acid (27a). Yield 71%. H NMR (DMSO-d6,
00 MHz) δ 12.61 (br, 2H), 7.86 (d, J = 7.8 Hz, 1H), 7.79 (s, 1H), 7.72 (d, J = 6.8 Hz, 1H), 7.43
6
(
t, J = 7.2 Hz, 1H), 6.60 (s, 2H), 4.04 (s, 2H).
.1.21. 5-(Furan-2-yl)-2-hydroxyisoquinoline-1,3(2H,4H)-dione (7a). Yield 41%. H NMR
DMSO-d6, 600 MHz) δ 10.42 (s, 1H), 8.05 (d, J = 7.2 Hz, 1H), 7.98 (d, J = 7.8 Hz, 1H), 7.86
1
4
(
(
1
3
s, 1H), 7.56 (t, J = 7.8 Hz, 1H), 6.94 (s, 2H), 6.67 (s, 2H), 4.33 (s, 2H); C NMR (DMSO-d6,
1
3
4
50 MHz) δ 166.1, 162.0, 150.8, 144.0, 131.8, 130.7, 128.9, 128.1, 127.8, 126.8, 112.5, 111.0,
-
7.3; HRMS-ESI(-) m/z calcd for C H NO 242.0453 [M-H] , found 242.0483.
1
3
8
4
1
.1.22. 2-(Carboxymethyl)-[1,1'-biphenyl]-3-carboxylic acid (27b). Yield 55%. H NMR
(DMSO-d6, 600 MHz) δ 12.41 (br, 2H), 7.92 (d, J = 7.8 Hz, 1H), 7.88 (d, J = 7.8 Hz, 1H), 7.48

ACCEPTED MANUSCRIPT
(t, J = 7.8 Hz, 1H), 7.43-7.42 (m, 2H), 7.31 (d, J = 6.6 Hz, 1H), 7.24 (d, J = 7.2 Hz, 2H), 3.91 (s,
2
H).
.1.23. 2-Hydroxy-5-phenylisoquinoline-1,3(2H,4H)-dione (7b). Yield 44%. H NMR (DMSO-
1
4
d6, 600 MHz) δ 10.44 (s, 1H), 8.09 (d, J = 6.6 Hz, 1H), 7.56-7.53 (m, 2H), 7.49-7.47 (m, 2H),
1
3
7
.44 (d, J = 6.6 Hz, 1H), 7.39 (d, J = 7.2 Hz, 2H), 4.02 (s, 2H); C NMR (DMSO-d6, 150 MHz)
δ 166.3, 162.2, 140.8, 139.0, 134.9, 132.3, 129.3, 129.0, 128.3, 128.0, 127.7, 126.1, 37.0;
-
HRMS-ESI(-) m/z calcd for C15
H10NO
3
252.0661 [M-H] , found 252.0697.
1
4
.1.24. 2-(Carboxymethyl)-2',4'-difluoro-[1,1'-biphenyl]-3-carboxylic acid (27c). Yield 63%. H
NMR (DMSO-d6, 600 MHz) δ 12.41 (br, 2H), 7.88 (d, J = 8.4 Hz, 1H), 7.48 (t, J = 7.2 Hz, 1H),
7
.36-7.35 (m, 2H), 7.31-7.29 (m, 2H), 3.91 (s, 2H).
.1.25. 5-(2,4-Difluorophenyl)-2-hydroxyisoquinoline-1,3(2H,4H)-dione (7c). Yield 38%. H
1
4
NMR (DMSO-d6, 600 MHz) δ 10.47 (s, 1H), 8.15 (d, J = 6.6 Hz, 1H), 7.59 (s, 2H), 7.47-7.41
13
(
m, 2H), 7.24 (t, J = 6.6 Hz, 1H), 3.90 (s, 2H); C NMR (DMSO-d6, 150 MHz) δ 166.0, 162.0,
1
57.7, 135.7, 133.6, 133.4, 128.5, 128.0, 126.3, 104.9, 36.4; HRMS-ESI(-) m/z calcd for
-
C
15
H
8
NF
2
O
3
288.0472 [M-H] , found 288.0506.
1
4
.1.26. 4-(4-Methylbenzyl)-2,3-dihydro-1H-inden-1-one (30a). Yield 66%. H NMR (CDCl
MHz) δ 7.65 (d, J = 7.8 Hz, 1H), 7.39 (d, J = 7.2 Hz, 1H), 7.33 (t, J = 7.8 Hz, 1H), 7.11 (d, J =
.8 Hz, 2H), 7.05 (d, J = 8.4 Hz, 2H), 4.01 (s, 2H), 2.99-2.97 (m, 2H), 2.67-2.65 (m, 2H), 2.32
s, 3H).
.1.27. 2-(Carboxymethyl)-3-(4-methylbenzyl)benzoic acid (31a). Yield 51%. H NMR (DMSO-
d6, 600 MHz) δ 12.43 (br, 2H), 7.71 (d, J = 7.8 Hz, 1H), 7.42 (t, J = 7.8 Hz, 1H), 7.30-7.27 (m,
H), 7.07 (d, J = 7.8 Hz, 2H), 6.96 (d, J = 8.4 Hz, 2H), 3.90 (s, 2H), 2.28 (s, 3H).
3
, 600
7
(
1
4
1

ACCEPTED MANUSCRIPT
1
4
.1.28. 2-Hydroxy-5-(4-methylbenzyl)isoquinoline-1,3(2H,4H)-dione (8a). Yield 39%. H NMR
(DMSO-d6, 600 MHz) δ 10.36 (s, 1H), 7.94-7.92 (m, 1H), 7.43-7.42 (m, 2H), 7.10 (d, J = 7.8
1
3
Hz, 2H), 7.04 (d, J = 7.8 Hz, 2H), 4.05 (s, 2H), 3.97 (s, 2H), 2.24 (s, 3H); C NMR (DMSO-d6,
1
3
4
50 MHz) δ 166.3, 162.2, 139.2, 136.1, 135.7, 134.9, 133.3, 129.6, 129.1, 127.8, 126.5, 126.0,
-
7.2, 35.6, 21.0; HRMS-ESI(-) m/z calcd for C H NO 280.0974 [M-H] , found 280.0995.
1
7
14
3
1
.1.29. 4-(4-(Trifluoromethyl)benzyl)-2,3-dihydro-1H-inden-1-one (30b). Yield 62%. H NMR
, 600 MHz) δ 7.62 (d, J = 7.2 Hz, 1H), 7.49 (d, J = 7.8 Hz, 2H), 7.31-7.29 (m, 2H), 7.21-
.18 (m, 2H), 4.04 (s, 2H), 2.89-2.87 (m, 2H), 2.61-2.59 (m, 2H).
.1.30. 2-(Carboxymethyl)-3-(4-(trifluoromethyl)benzyl)benzoic acid (31b). Yield 56%. H NMR
(CDCl
3
7
4
1
(DMSO-d6, 600 MHz) δ 12.39 (br, 2H), 7.67 (d, J = 7.8 Hz, 1H), 7.62 (d, J = 7.8 Hz, 1H), 7.49-
7
.48 (m, 1H), 7.44 (d, J = 8.4 Hz, 2H), 7.31 (d, J = 7.8 Hz, 2H), 4.13 (s, 2H), 3.90 (s, 2H).
.1.31. 2-Hydroxy-5-(4-(trifluoromethyl)benzyl)isoquinoline-1,3(2H,4H)-dione (8b). Yield 44%.
4
1
H NMR (CD OD, 600 MHz) δ 8.11 (d, J = 7.8 Hz, 1H), 7.61 (d, J = 8.4 Hz, 2H), 7.50-7.48 (m,
3
1
3
2
1
3
4
6
4
4
H), 7.37 (d, J = 8.4 Hz, 2H), 4.84 (s, 2H), 4.17 (s, 2H); C NMR (DMSO-d6, 150 MHz) δ
66.6, 162.3, 143.4, 137.7, 135.1, 132.6, 129.0, 128.6, 127.5, 126.8, 125.6, 125.2, 125.1, 48.1,
-
7.2; HRMS-ESI(-) m/z calcd for C17
H
11NF
3
O
3
334.0691 [M-H] , found 343.0724.
1
.1.32. 4-(2,4-Difluorobenzyl)-2,3-dihydro-1H-inden-1-one (30c). Yield 58%. H NMR (CDCl ,
3
00 MHz) δ 7.67 (d, J = 7.2 Hz, 1H), 7.35-7.32 (m, 2H), 7.05-7.02 (m, 1H), 6.84-6.80 (m, 2H),
.01 (s, 2H), 3.03-3.01 (m, 2H), 2.70-2.68 (m, 2H).
1
.1.33. 2-(Carboxymethyl)-3-(2,4-difluorobenzyl)benzoic acid (31c). Yield 49%. H NMR
(DMSO-d6, 600 MHz) δ 12.41 (br, 2H), 7.73 (d, J = 7.8 Hz, 1H), 7.31-7.26 (m, 2H), 7.22-7.24
(m, 1H), 7.05-7.02 (m, 1H), 6.98-6.96 (m, 1H), 3.99 (s, 2H), 3.94 (s, 2H).

ACCEPTED MANUSCRIPT
1
4
.1.34. 5-(2,4-Difluorobenzyl)-2-hydroxyisoquinoline-1,3(2H,4H)-dione (8c). Yield 36%. H
NMR (DMSO-d6, 600 MHz) δ 10.39 (s, 1H), 7.95 (d, J = 7.8 Hz, 1H), 7.42 (t, J = 7.8 Hz, 1H),
7
.30 (d, J = 7.8 Hz, 1H), 7.25-7.20 (m, 2H), 7.03 (t, J = 7.8 Hz, 1H), 4.14 (s, 2H), 4.00 (s, 2H);
C NMR (DMSO-d6, 150 MHz) δ 166.3, 162.2, 160.0, 137.4, 134.1, 133.4, 132.6, 132.5, 127.9,
1
3
1
26.8, 126.0, 122.3, 122.1, 112.2, 104.5, 104.2, 35.5, 30.3; HRMS-ESI(-) m/z calcd for
-
C H NF O 302.0629 [M-H] , found 302.0654.
16
10
2
3
1
4
.1.35. 4-(4-Bromobenzyl)-2,3-dihydro-1H-inden-1-one (30d). Yield 52%. H NMR (CDCl
3
, 600
MHz) δ 7.60 (d, J = 7.8 Hz, 1H), 7.35 (d, J = 8.4 Hz, 2H), 7.29-7.27 (m, 2H), 6.97 (d, J = 7.8 Hz,
2
H), 3.93 (s, 2H), 2.89-2.87 (m, 2H), 2.60-2.58 (m, 2H).
.1.36. 3-(4-Bromobenzyl)-2-(carboxymethyl)benzoic acid (31d). Yield 45%. H NMR (DMSO-
1
4
d6, 600 MHz) δ 12.81 (br, 1H), 12.09 (br, 1H), 7.72 (d, J = 7.8 Hz, 1H), 7.43 (d, J = 7.8 Hz, 2H),
7
2
4
.34 (d, J = 7.2 Hz, 1H), 7.29 (t, J = 7.2 Hz, 1H), 7.03 (d, J = 7.8 Hz, 2H), 3.98 (s, 2H), 3.87 (s,
H).
1
.1.37. 5-(4-Bromobenzyl)-2-hydroxyisoquinoline-1,3(2H,4H)-dione (8d). Yield 30%. H NMR
(DMSO-d6, 600 MHz) δ 10.38 (s, 1H), 7.95 (t, J = 2.4 Hz, 1H), 7.49 (d, J = 8.4 Hz, 2H), 7.44-
13
7
.43 (m, 2H), 7.13 (d, J = 7.8 Hz, 2H), 4.05 (s, 2H), 4.00 (s, 2H); C NMR (DMSO-d6, 150
MHz) δ 166.3, 162.2, 138.8, 138.5, 135.0, 133.4, 131.9, 131.5, 127.9, 126.7, 126.1, 119.8, 36.9,
-
3
5.6; HRMS-ESI(-) m/z calcd for C H NBrO 343.9922 [M-H] , found 343.9956.
1
6
11
3
4
.1.38. General procedure for the preparation of homophthalic anhydrides (33). [10]
To homophthalic acid 17 [16] (150 mg, 0.77 mmol) in round bottom flask was added
substituted benzoyl chloride (6.15 mmol, 8 equiv). This reaction mixture was heated under
refluxing for 1.5 h then cooled to room temperature. The precipitate was filtered and washed
with ether and cold methanol to give the desired compounds.

ACCEPTED MANUSCRIPT
1
4
.1.39. 6-((4-Fluorobenzyl)amino)isochroman-1,3-dione (33a). Yield: 85%. HNMR (600 MHz,
DMSO-d6) δ10.61 (s, 1H), 7.97 (d, J = 8.4 Hz, 1H), 7.70 (s, 1H), 7.59 (d, J = 8.4 Hz, 1H), 7.33
t, J = 7.2 Hz, 2H), 7.25 (t, J = 7.2 Hz, 2H); MS (ESI-) m/z: 284.26.
.1.40. N-(1,3-Dioxoisochroman-6-yl)-4-fluorobenzamide (33b). Yield: 98%. HNMR (600
MHz, DMSO-d6) δ 10.13 (s, 1H), 7.96 (m, 2H), 7.78 (m, 2 H), 7.65 (s, 1H), 7.51 (t, J = 7.8 Hz,
H), 7.48 (s, 1H), 3.98 (s, 2H); MS (ESI-) m/z: 298.25.
.1.41. 6-(Benzylamino)isochroman-1,3-dione (33c). Yield: 96%. HNMR (600 MHz, CD OD)
(
1
4
1
1
4
3
δ 7.91 (d, J = 8.4 Hz, 1 H), 7.53 (d, J = 8.4 Hz, 1H), 7.44 (s, 1H), 7.25 (m, 4H), 7.16 (t, J = 7.2,
1
H), 3.89 (s, 2H), 3.61 (s, 2H); MS (ESI-) m/z: 266.28.
.1.42. N-(1,3-Dioxoisochroman-6-yl)benzamide (33d). Yield: 95%. HNMR (600 MHz,
1
4
DMSO- d6) δ 10.12 (s, 1H), 7.94 (m, 3H), 7.81 (m, 1H), 7.74 (s, 1H), 7.59 (t, J = 7.8 Hz, 1H),
7
.53 (t, J = 7.2 Hz, 2H), 3.98 (s, 2H); MS (ESI-) m/z: 280.26.
4
.1.43. General procedure for the synthesis of 34 derivatives.
A solution of 33a (100 mg, 0.36 mmol) and O-benzyl hydroxylamine (53 mg, 0.42 mmol, 1.2
equiv) in toluene 120 mL was heated under refluxing overnight. Solvent was removed and the
residue was re-dissolved in DMF. To this solution was added CDI (64 mg, 1.1 equiv), and the
o
resulting mixture was heated at 100 C for 1 hour, then cooled to rt. The solvent was removed
and the residue was purified by column chromatography to give desired compounds.
4
.1.44. 2-(Benzyloxy)-6-((4-fluorobenzyl)amino)isoquinoline-1,3(2H,4H)-dione (34a). Yield:
1
6
1
0%. HNMR (600 MHz, CD
3
OD) δ 7.89 (d, J = 9.0 Hz, 1H), 7.53 (d, J =9.0 Hz, 1H), 7.45 (s,
H), 7.26 (m, 7H), 6.95 (t, J = 9.0 Hz, 2H), 3.58 (s, 2H), 3.71 (s, 2H); MS (ESI-) m/z: 389.41.

ACCEPTED MANUSCRIPT
4
.1.45. N-(2-(Benzyloxy)-1,3-dioxo-1,2,3,4-tetrahydroisoquinolin-6-yl)-4-fluorobenzamide (34b).
1
Yield: 80%. HNMR (600 MHz, DMSO-d6) δ 10.70 (s, 1H), 8.05 (m, 2H), 7.89 (d, J = 9.0 Hz,
H), 7.79 (d, J = 9.0 Hz, 1H), 7.72 (s, 1H), 7.41 (m, 7H), 4.78 (s, 2H), 3.76 (s, 2H); MS (ESI-)
m/z: 403.39
1
4
.1.46. 6-(Benzylamino)-2-(benzyloxy)isoquinoline-1,3(2H,4H)-dione (34c). Yield: 82%.
1
HNMR (600 MHz, CD
3
OD) δ 7.89 (d, J = 9.0 Hz, 1H), 7.75 (s, 1H), 7.66 (s, 1H), 7.50 (m, 3H),
7
.28 (m, 6H), 7.16 (m, 1H), 4.98 (s, 2H), 3.62 (s, 2H); MS (ESI-) m/z: 371.42.
4
.1.47. N-(2-(Benzyloxy)-1,3-dioxo-1,2,3,4-tetrahydroisoquinolin-6-yl)benzamid (34d). Yield:
1
7
7
5
4
4 %. HNMR (600 MHz, DMSO- d6) δ 10.70 (s, 1H), 8.12 (d, J = 9.0 Hz, 1 H), 8.05 (m, 2H),
.99 (s, 1H), 7.94 (d, J = 9.0 Hz, 1H), 7.10 (t, J = 7.2 Hz, 1H), 7.64 (m, 4H), 7.50 (m, 3H),
.11 (s, 2H), 4.37 (s, 2H), MS (ESI-) m/z: 386.40
.1.48. General procedure for the synthesis of 9 derivatives.
Compounds 34b (40 mg, 0.1 mmol) was dissolved in 0.5 mL of CH
2 2
Cl , and boron tribromide
(
1.0 M solution in CH Cl , 7.5 mL, 0.75 mmol) was added dropwise at room temperature. The
2
2
reaction mixture was stirred for1 hour at rt (reaction was followed by TLC analysis) and
quenched by adding 2 mL of H O dropwise. The resulting precipitate was filtered, washed with
2
H O and cold ether. The crude products were recrystallized in methanol to give desired
2
compounds.
4
.1.49.
2-(4-Fluorophenyl)-N-(2-hydroxy-1,3-dioxo-1,2,3,4-tetrahydroisoquinolin-6-
1
3
yl)acetamide (9a). Yield: 80 %. HNMR (600 MHz, CD OD) δ 8.06 (d, J = 9.0 Hz, 1H), 7.75 (s,
1
H), 7.58 (d, J = 9.0 Hz, 1H), 7.35 (t, J = 8.4 Hz, 2H), 7.05 (t, J = 8.4 Hz, 2H), 4.18 (s, 2H), 3.69
13
(
s, 2H); C NMR (150 MHz, DMSO-d6) δ 170.2, 166.7, 161.7, 144.0, 136.3, 132.131.5, 131.4,

ACCEPTED MANUSCRIPT
-
1
3
4
20.2, 118.2, 117.0, 115.6, 115.4, 42.8, 37.5; HRMS (ESI-) calcd. for C17
27.0978 found 327.0977.
2 4
H12FN O [M-H]
.1.50. 4-Fluoro-N-(2-hydroxy-1,3-dioxo-1,2,3,4-tetrahydroisoquinolin-6-yl)benzamide (9b).
1
Yield: 75%; HNMR (600 MHz, CD OD) δ 8.05 (d, J = 8.4 Hz, 1H), 7.93 (m, 1H), 7.81 (s, 1H),
3
1
3
7
.67 (d, J = 9.0 Hz, 1H), 7.59 (m, 1H), 7.14 (t, J = 8.4 Hz, 2H), 4.13 (s, 2H); C NMR (150
MHz, DMSO-d6) δ 166.7, 161.7,161.8, 144.0, 136.1, 133.1, 131.0, 129.2, 120.5, 119.3, 118.1,
-
1
16.0, 115.8, 37.6; HRMS (ESI-) calcd. for C H FN O [M-H] 313.0824 found 313.0824.
1
6
10
2
4
4
.1.51. N-(2-Hydroxy-1,3-dioxo-1,2,3,4-tetrahydroisoquinolin-6-yl)-2-phenylacetamide (9c).
1
Yield: 67%. HNMR (600 MHz, CD OD) δ 7.97 (d, J = 9.0 Hz, 1H), 7.67 (s, 1H), 7.50 (d, J =
3
1
3
9
.0 Hz, 1H), 7.26 (m, 4H), 7.20 (t, J = 7.2 Hz, 1H), 4.13 (s, 2H), 3.62 (s, 2H); C NMR (150
MHz, DMSO-d6) δ 170.3, 166.7, 161.8, 144.0, 136.3, 136.0, 129.5, 129.3, 128.7, 127.1, 120.1,
-
1
3
18.2, 117.0, 43.8, 37.4; HRMS (ESI-) calcd. for C H N O [M-H] 309.1110, found
1
7
13
2
4
09.1112.
.1.52. N-(2-Hydroxy-1,3-dioxo-1,2,3,4-tetrahydroisoquinolin-6-yl)benzamide (9d). Yield: 60%.
4
1
3
HNMR (600 MHz, CD OD) δ 8.40 (d, J = 8.4 Hz, 1 H), 8.23 (d, J = 6.6 Hz, 2H), 8.18 (s, 1H),
1
3
8
.08 (d, J = 9.0 Hz, 1H), 7.89 (t, J = 7.2 Hz, 1H), 7.82 (t, J = 7.2 Hz, 2H), 4.53 (s, 2H); C
NMR (150 MHz, DMSO--d6) δ 172.7, 168.1, 166.3, 138.1, 133.3, 132.2, 132.0, 129.7, 129.0,
-
1
2
4
4
28.8, 128.2, 123.8, 118.5, 40.8; HRMS (ESI-) calcd. for C16
H
11
N
2
O
4
[M-H] 295.1057 found
95.1055.
.2. Biology
.2.1. RNase H assay
RNase H activity was measured essentially as previously described.[14, 21] Full-length HIV
RT was incubated with the RNA/DNA duplex substrateHTS-1 (RNA 5’-gaucugagccugggagcu -

ACCEPTED MANUSCRIPT
3
’-fluorescein annealed to DNA 3’-CTAGACTCGGACCCTCGA -5’-Dabcyl), a high sensitivity
duplex that assesses non-specific internal cleavage of the RNA strand.
4
.2.2. HIV-1 cytoprotection assay [23]
The HIV Cytoprotection assay used CEM-SS cells and the IIIB strain of HIV-1. Briefly virus
and cells were mixed in the presence of test compound and incubated for 6 days. The virus was
pre-titered such that control wells exhibit 70 to 95% loss of cell viability due to virus replication.
Therefore, antiviral effect or cytoprotection was observed when compounds prevent virus
replication. Each assay plate contained cell control wells (cells only), virus control wells (cells
plus virus), compound toxicity control wells (cells plus compound only), compound colorimetric
control wells (compound only) as well as experimental wells (compound plus cells plus virus).
4
.3. Molecular modeling
All modeling experiments were done employing Schrodinger Small Molecule Discovery
Suite.[30]
.3.1. Small Molecules
D structures of compounds 8c and 9c were prepared with the LigPrep™ module, which
4
2
converts 2D structures to 3D, performs energy minimization, and predicts protonated and
deprotonated states at physiological pH based on calculated pK values for ionizable hydrogens.
a
For both studied compounds pK for the hydroxyl attached to nitrogen atom was predicted to be
a
around 7.07±1 pH unit, which indicates that both protonated and deprotonated forms of 8c and
9
c will likely be present at physiological pH. However, in chelating divalent cations in active
center the deprotonated form will be more effective due to negative charge and better availability
of oxygen lone pairs to overlap with metal orbitals. Thus deprotonated forms of 8c and 9c were
used for docking into active sites. For docking into non-nucleoside RT inhibitor (NNRTI)

ACCEPTED MANUSCRIPT
binding pocket both deprotonated and protonated forms were tested. For both compounds in
protonated and deprotonated forms conformers were generated using advanced conformational
search implemented in ConfiGen™ module. All obtained conformers were subjected to docking
into RNase H active center, poly active center, and NNRTI binding pocket.
4
.3.2. Protein Receptors
Selected HIV-1 RT structures were obtained from the Protein Data Bank. To add hydrogen
atoms, minimize energy, and create appropriate protonation states of amino acid side chains the
Protein Preparation Wizard ™ module was used. Docking grid was centered on co-crystallized
ligands. Van der Waals radii of non-polar atoms were scaled by 0.8 scaling factor to account for
some flex ability of protein backbone and amino acid side chains. 3LP1[31] crystal structure was
used as a receptor for dicking to RNase H active site. Docking to polymerase active site was
carried out with 1RTD[32] structure. This structure contains RNase H active site inhibitor. RT
structure 1KLM[33] served as a receptor for docking into NNRTI binding pocket.
4
.3.3. Docking
Docking simulations were performed with Glide ™ docking algorithm at standard precision.
Coulombic and van der Waals interactions of 8c, 9c and delavirdine with individual amino acid
residues were calculated. Ligand poses with the most negative docking scores were chosen for
further analysis.
Acknowledgements
This research was supported in part by the National Institutes of Health (AI100890 to SGS,
MAP and ZW), and by the Research Development and Seed Grant Program of the Center for
Drug Design, University of Minnesota.
Appendix A. Supplementary data

ACCEPTED MANUSCRIPT
Additional docking data for compounds 8c and 9c. This material is available free of charge
via the Internet at t http://dx.doi.org
Abbreviations
HIV
human immunodeficiency virus
reverse transcriptase
ribonuclease H
RT
RNase H
HID
2-hydroxyisoquinoline-1,3-dione
integrase
IN
Pol
polymerase
HCV
Pol
hepatitis C virus
polymerase
PR
protease
HAART
NRTI
NNRTI
highly active antiretroviral therapy
nucleoside RT inhibitor
nonnucleoside RT inhibitor

ACCEPTED MANUSCRIPT
RISF
retroviral integrase super family
LID
ligand interaction diagram
References

ACCEPTED MANUSCRIPT
[
1] E.O. Freed, HIV-1 replication, Somat. Cell Mol. Genet. 26(1-6) (2001) 13-33.
2] D.o.H.a.H. Services., Panel on Antiretroviral Guidelines for Adults and Adolescents.
Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. ,
[
2
015. http://aidsinfo.nih.gov/contentfiles/AdultandAdolescentGL.pdf.
[
3] S.G. Sarafianos, B. Marchand, K. Das, D.M. Himmel, M.A. Parniak, S.H. Hughes, E. Arnold,
Structure and function of HIV-1 reverse transcriptase: molecular mechanisms of
polymerization and inhibition, J. Mol. Biol. 385(3) (2009) 693-713.
[
4] T. Cihlar, A.S. Ray, Nucleoside and nucleotide HIV reverse transcriptase inhibitors: 25 years
after zidovudine, Antiviral Res. 85(1) (2010) 39-58.
[
5] M.P. de Bethune, Non-nucleoside reverse transcriptase inhibitors (NNRTIs), their discovery,
development, and use in the treatment of HIV-1 infection: a review of the last 20 years
(1989-2009), Antiviral Res. 85(1) (2010) 75-90.
[
6] Z.L. Xu, J.M. Guo, Y. Yang, M.D. Zhang, M.Y. Ba, Z.Z. Li, Y.L. Cao, R.C. He, M. Yu, H.
Zhou, X.X. Li, X.S. Huang, Y. Guo, C.B. Guo, 2,4,5-Trisubstituted thiazole derivatives as
HIV-1 NNRTIs effective on both wild-type and mutant HIV-1 reverse transcriptase:
Optimization of the substitution of positions 4 and 5, Eur. J. Med. Chem. 123 (2016) 309-
3
16.
[
7] P. Simon, O. Baszczynski, D. Saman, G. Stepan, E. Hu, E.B. Lansdon, P. Jansa, Z. Janeba,
Novel (2,6-difluorophenyl)(2-(phenylamino)pyrimidin-4-yl) methanones with restricted
conformation as potent non-nucleoside reverse transcriptase inhibitors against HIV-1, Eur. J.
Med. Chem. 122 (2016) 185-195.
[
8] T. Ilina, K. Labarge, S.G. Sarafianos, R. Ishima, M.A. Parniak, Inhibitors of HIV-1 Reverse
Transcriptase-Associated Ribonuclease H Activity, Biology 1(3) (2012) 521-541.

ACCEPTED MANUSCRIPT
[
9] Parniak, M. P. unpublished data
10] M. Nowotny, Retroviral integrase superfamily: the structural perspective, EMBO rep. 10(2)
2009) 144-151.
[
(
[
11] K. Klumpp, J.Q. Hang, S. Rajendran, Y.L. Yang, A. Derosier, P.W.K. In, H. Overton,
K.E.B. Parkes, N. Cammack, J.A. Martin, Two-metal ion mechanism of RNA cleavage by
HIV RNase H and mechanism-based design of selective HIV RNase H inhibitors, Nucleic
Acids Res. 31(23) (2003) 6852-6859.
[
12] M. Billamboz, F. Bailly, M.L. Barreca, L. De Luca, J.F. Mouscadet, C. Calmels, M.L.
Andreola, M. Witvrouw, F. Christ, Z. Debyser, P. Cotelle, Design, Synthesis, and Biological
Evaluation of a Series of 2-Hydroxyisoquinoline-1,3(2H,4H)-diones as Dual Inhibitors of
Human Immunodeficiency Virus Type 1 Integrase and the Reverse Transcriptase RNase H
Domain, J. Med. Chem. 51(24) (2008) 7717-7730.
[
13] M. Billamboz, F. Bailly, C. Lion, C. Calmels, M.L. Andreola, M. Witvrouw, F. Christ, Z.
Debyser, L. De Luca, A. Chimirri, P. Cotelle, 2-Hydroxyisoquinoline-1,3(2H,4H)-diones as
inhibitors of HIV-1 integrase and reverse transcriptase RNase H domain: Influence of the
alkylation of position 4, Eur. J. Med. Chem. 46(2) (2011) 535-546.
[
14] S.K. Vernekar, Z. Liu, E. Nagy, L. Miller, K.A. Kirby, D.J. Wilson, J. Kankanala, S.G.
Sarafianos, M.A. Parniak, Z. Wang, Design, synthesis, biochemical, and antiviral
evaluations of C6 benzyl and C6 biarylmethyl substituted 2-hydroxylisoquinoline-1,3-
diones: dual inhibition against HIV reverse transcriptase-associated RNase H and
polymerase with antiviral activities, J. Med. Chem. 58(2) (2015) 651-664.
[15] M. Billamboz, V. Suchaud, F. Bailly, C. Lion, J. Demeulemeester, C. Calmels, M.L.
Andreola, F. Christ, Z. Debyser, P. Cotelle, 4-Substituted 2-Hydroxyisoquinoline-