Structure-Activity Relationship for Small Molecule Inhibitors of Nicotinamide N-Methyltransferase

Article abstract of DOI:10.1021/acs.jmedchem.7b00389

Nicotinamide N-methyltransferase (NNMT) is a fundamental cytosolic biotransforming enzyme that catalyzes the N-methylation of endogenous and exogenous xenobiotics. We have identified small molecule inhibitors of NNMT with >1000-fold range of activity and developed comprehensive structure-Activity relationships (SARs) for NNMT inhibitors. Screening of N-methylated quinolinium, isoquinolinium, pyrididium, and benzimidazolium/benzothiazolium analogues resulted in the identification of quinoliniums as a promising scaffold with very low micromolar (IC₅₀ à 1 μM) NNMT inhibition. Computer-based docking of inhibitors to the NNMT substrate (nicotinamide)-binding site produced a robust correlation between ligand-enzyme interaction docking scores and experimentally calculated IC₅₀ values. Predicted binding orientation of the quinolinium analogues revealed selective binding to the NNMT substrate-binding site residues and essential chemical features driving protein-ligand intermolecular interactions and NNMT inhibition. The development of this new series of small molecule NNMT inhibitors direct the future design of lead drug-like inhibitors to treat several metabolic and chronic disease conditions characterized by abnormal NNMT activity.

Full text of DOI:10.1021/acs.jmedchem.7b00389

Subscriber access provided by Binghamton University | Libraries
Article
Structure Activity Relationship for Small Molecule
Inhibitors of Nicotinamide N-Methyltransferase
Harshini Neelakantan, Hua-Yu Wang, Virginia Vance, Jonathan
D Hommel, Stanton F McHardy, and Stanley J. Watowich
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 26 May 2017
Downloaded from http://pubs.acs.org on May 26, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure Activity Relationship for Small Molecule Inhibitors of Nicotinamide N-
Methyltransferase
1
2
1
3
Harshini Neelakantan, HuaꢀYu Wang, Virginia Vance, Jonathan D. Hommel, Stanton F.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
1*
McHardy, Stanley J. Watowich
1
Department of Biochemistry and Molecular Biology, University of Texas Medical Branch,
Galveston, Texas 77550 USA
2
Department of Chemistry and Center for Innovative Drug Discovery, University of Texas at San
Antonio, San Antonio, Texas 78249 USA
3
Department of Pharmacology and Toxicology, University of Texas Medical Branch, Galveston,
Texas 77550 USA
*
Corresponding Author
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
Nicotinamide Nꢀmethyltransferase (NNMT) is a fundamental cytosolic biotransforming enzyme
that catalyzes the Nꢀmethylation of endogenous and exogenous xenobiotics. We have identified
small molecule inhibitors of NNMT with >1000ꢀfold range of activity and developed
comprehensive structureꢀactivity relationships (SARs) for NNMT inhibitors. Screening of Nꢀ
methylated quinolinium, isoquinolinium, pyrididium, and benzimidazolium/benzothiazolium
analogs resulted in the identification of quinoliniums as a promising scaffold with very low
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
micromolar (IC ~1 ꢁM) NNMT inhibition. Computerꢀbased docking of inhibitors to the NNMT
50
substrate (nicotinamide)ꢀbinding site produced a robust correlation between ligandꢀenzyme
interaction docking scores and experimentally calculated IC values. Predicted binding
5
0
orientation of the quinolinium analogs revealed selective binding to the NNMT substrateꢀbinding
site residues and essential chemical features driving proteinꢀligand intermolecular interactions
and NNMT inhibition. The development of this new series of small molecule NNMT inhibitors
direct the future design of lead drugꢀlike inhibitors to treat several metabolic and chronic disease
conditions characterized by abnormal NNMT activity.
2
ACS Paragon Plus Environment

Page 3 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Nicotinamide Nꢀmethyltransferase (NNMT) is a cytosolic enzyme that catalyzes the
transfer of methyl group from the coꢀfactor Sꢀ(5’ꢀadenosyl)ꢀLꢀmethionine (SAM) to substrates
such as nicotinamide (NCA), pyridine, and related analogs (e.g., quinoline, isoquinoline, 1,2,3,4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
tetrahydroisoquinoline ), directly regulating the detoxification of endogenous and exogenous
drugs/xenobiotics by the formation of methylated metabolic products (1ꢀmethyl nicotinamide [1ꢀ
2
MNA], methylated pyridiniums, and methylated related analogs). Given its primary
3
ꢀ5
metabolizing function, NNMT is predominantly expressed in the liver, but significant levels of
the enzyme are also present in other tissues, including the adipose tissue, kidney, brain, lung,
2
, 6
heart, and muscle. Enhanced expression and enzymatic activity of NNMT has been linked to a
number of chronic disease conditions making it a significant and relevant target for drug
development. For example, several studies have demonstrated a causal relationship between
increased NNMT expression and enhanced cell proliferation/progression in a variety of cancer
cell lines with potential implications for NNMT as a biomarker for cancer prognosis and a target
7ꢀ11
for antiꢀcancer therapeutic development.
NNMT expression has also been reported to be
upregulated in patients with Parkinson’s disease, which is suggested to be linked to the
production of neurotoxins such as Nꢀmethylpyridinium ions that underlie neurodegeneration.
1
2ꢀ13
1
4ꢀ15
14, 16ꢀ17
Furthermore, studies in both animals
and humans
have shown that NNMT expression
and activity was increased in obesity and related chronic metabolic conditions (e.g., typeꢀ2
diabetes). Knockdown of NNMT expression using an antisense oligonucleotide was reported to
suppress body weight gain, reduce fat mass, and increase energy expenditure in mice fed a high
1
5
fat diet. While the underlying molecular mechanisms that link decreased NNMT activity to
increased adipocyte metabolism are not well understood, NNMT may modulate intracellular
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
metabolite turnover in the methionineꢀhomocysteine cycle and/or the nicotinamide adenine
1
5
dinucleotide (NAD+) synthesis pathway critical for cellular energy expenditure. Therefore,
targeted small molecule inhibitors of the NNMT could be significantly beneficial as molecular
probes for mechanistic investigations and for the development of therapeutics to treat metabolic
and chronic diseases that are characterized by abnormal NNMT activity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
SAMꢀdependent methyltransferases represent a major class of biotransforming enzymes
that catalyze the methylation of various substrates, including proteins, nucleic acids, and
endogenous small molecules (e.g., intracellular metabolites) using the coꢀfactor SAM as a
18ꢀ20
universal methyl donor.
Many inhibitors of SAMꢀdependent methyltransferases (e.g., the
broadꢀspectrum methyltransferase inhibitor sinefungin, histone methyltransferase inhibitor EPZꢀ
2
1ꢀ22
5676) mimic the chemical structure of SAM and interact with the SAMꢀbinding site,
thereby
23
lacking selectivity for specific methyltransferase enzymes. However, the recent determination
2
4
of the structure of NNMT bound to the NCA substrate and SAH product provide information
that can be used to design selective and specific small molecule NNMT inhibitors.
Based on consistent observations of 1ꢀMNA, the primary endogenous Nꢀmethylated
product of NNMT, as an NNMT inhibitor in biochemical and pharmacological studies in vitro
2
, 15
and in vivo,
in the present study we explored the NNMT inhibitory activity of Nꢀmethylated
heterocyclic small molecules containing several different aromatic scaffolds including quinoline,
1
, 25
isoquinoline, pyridine, benzimidazole, and benzothiazole.
Preliminary sets of Nꢀmethylated
analogs for each of the chemical scaffolds were purchased from commercial suppliers and
screened for NNMT inhibitory activity using purified recombinant protein. We expanded several
promising chemical series to develop predictive structureꢀactivity relationships (SARs) with
emphasis on substitution effect on aryl ring and the N1ꢀatom for each chemical series. Further,
4
ACS Paragon Plus Environment

Page 5 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
computerꢀbased docking of inhibitors to the NNMT NCA substrateꢀbinding site yielded a
correlation between inhibitorꢀenzyme interaction docking scores and experimentally determined
IC values that served as a guiding tool to identify NNMT residues responsible for proteinꢀ
5
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ligand interactions and further generate binding hypotheses to design novel inhibitors with
improved binding. The SAR described herein demonstrates significant promise for future
exploration of the chemical space surrounding the template moieties to develop potent, selective,
and drugꢀlike NNMT inhibitors.
RESULTS AND DISCUSSION
Chemistry. The general approach to prepare the N1ꢀmethylated quinolinium and N2ꢀmethylated
isoquinolinium salts was carried out as shown in Scheme 1. Commercially available and/or
synthesized quinoline and isoquinoline with various substitution patterns were treated with
iodomethane (MeI) in 2ꢀpropanol (IPA) or methyl trifluoromethanesulfonate (MeOTf) in toluene
to give the corresponding analogs in good yields. Full details on the synthesis of the substituted
aminoꢀquinoline and perꢀmethylated quinoline intermediates/precursors are described in the
Supporting Information.
Biological evaluation of 1-methylquinolinium (1-MQ) analogs. To test the hypothesis that Nꢀ
methylated quinolinium containing compounds function as NNMT inhibitors, we began by
probing the NNMT inhibitory activity of 1ꢀMQ (1a), which exhibited an IC value of 12 ꢁM
5
0
(Table 1). Based on the activity exhibited by 1a, we explored SAR for NNMT inhibition on the
1ꢀMQ scaffold by testing 48 additional analogs (obtained either from commercial suppliers or
25ꢀ26
synthesized inꢀhouse) using a modified HPLCꢀbased biochemical assay.
In the first series of
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compounds, we explored the inhibitory activities of monoꢀsubstituted 1ꢀMQ with various
polar/apolar substituents and different N1ꢀsubstituted quinolinium analogs (1b-z; Table 1).
Generally, small substituents such as methyl or amine groups were well tolerated in different
substitution patterns on the 1ꢀMQ ring with NNMT inhibitory activity in the low micromolar
range (Table 1). Among all the monoꢀmethylated quinolinium analogs tested with various
substitution patterns at C2ꢀC8ꢀpositions of 1ꢀMQ (1b, 1e, 1i, 1l, 1n, 1r and 1t; Table 1), a methyl
substituent at the C8ꢀposition (1t) gave the most significant improvement in inhibition (6.7ꢀfold
lower IC value, IC = 1.8 ꢁM) as compared to the parent compound 1ꢀMQ (1a). While a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
50
methyl substituent at the C3ꢀposition also exhibited low micromolar NNMT inhibition (1e, IC50
4.1 ꢁM; Table 1), an amino substitution at the same position further improved the IC value to
=
50
2.9 ꢁM (1d) suggesting relatively favorable interactions between the NNMT residues and the
C3ꢀamino substitution. To further explore the favorable nature of polar groups at the C3ꢀ
position, we tested analogs 1f-h; analog 1f with a C3ꢀmethylamino substitution and analog 1g
with a cyano substituent at the C3ꢀposition had ~3ꢀfold and ~8ꢀfold higher IC values,
5
0
respectively, relative to compound 1d. Importantly, analog 1h with a bulky phenyl ring at the
C3ꢀposition had significantly reduced inhibitory activity (IC > 1 mM). Taken together, the data
5
0
suggests that potentially favorable interactions can be imparted by small polar group substituents
such as an amine, while unfavorable steric interactions occur when large chemical groups occupy
the C3ꢀposition of the 1ꢀMQ scaffold.
To further explore 1ꢀMQ scaffold positions that could form favorable hydrogenꢀbonding
interactions with NNMT binding site residues to improve inhibitory activity, a systematic SAR
was conducted among different aminoꢀquinolinium derivatives (1c, 1d, 1j, 1k, 1o, 1q; Table 1).
Many of the 1ꢀMQ analogs with amino pharmacophores (e.g., 1c, 1k, 1q) demonstrated excellent
6
ACS Paragon Plus Environment

Page 7 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibitory activities that were comparable to 1d (IC = 2.9 ꢁM); the exceptions were analogs 1j
50
(C4ꢀamino substituent) and 1o (C6ꢀamino substituent) that exhibited ~4ꢀfold and ~10ꢀfold
reduction in activity (IC = 11 ꢁM and 34 ꢁM), respectively, compared to 1d. An amino
5
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
substitution at the C5ꢀposition (1k) produced potent NNMT inhibition (IC = 1.2 ꢁM) with 10ꢀ
5
0
and 2.4ꢀfold lower IC values compared to 1a and 1d, respectively (Figure 1; Table 1). This
5
0
suggests that the polar amine group at the C5ꢀposition favors strong intermolecular interactions
with the backbone serine residues in the NNMT substrateꢀbinding site (refer to molecular
docking discussion below). In contrast to amino substitutions, 1ꢀMQ analogs with other polar
substitutions such as hydroxyl (1s and 1u) and methoxy (1p) showed significantly weaker
NNMT inhibition activities compared to the parent compound 1a. For example, a hydroxyl
group substituent at the C7ꢀposition demonstrated ~50% inhibition of NNMT at nearly a
millimolar concentration (1s, IC = 709 ꢁM), whereas the comparable aminoꢀsubstituted analog
5
0
(
1q) was a potent NNMT inhibitor (IC = 2.6 ꢁM). While this observation remains intriguing, it
50
can be speculated based on our molecular docking analyses using predicted binding orientations
of the analogs 1s and 1q that the N atom of the C7ꢀamino forms a stronger hydrogen bond to the
carbonyl oxygen side chain of Ser213 residue of NNMT, relative to the more repulsive hydrogen
bond interaction between donor hydrogen atom of C7ꢀhydroxyl and the Ser213 residue.
Additionally, under the aqueous buffer reaction condition (pH 8.6) used in the current study, it is
likely that the C7ꢀhydroxy quinolinium analog undergoes tautomerization into its dominant form
(i.e., C7ꢀoxoꢀdihydroquinolinium analog), losing the charge on the N1 atom and significantly
diminishing the overall binding within the NNMT substrateꢀbinding site. Similarly, replacement
of the methyl group (1t) by a polar hydroxyl substituent at the C8ꢀposition exhibited >50ꢀfold
lower IC (1u, IC = 95.2 ꢁM) that may also be due to preferred ionization of the polar
5
0
50
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
hydroxyl group in aqueous buffer resulting in potential disruption of the tight binding within the
apolar pocket of the NNMT binding site (refer to molecular docking discussion below). Lastly,
we examined the relationship between quinolinium analogs containing N1ꢀsubstitutions with
varied steric and electronic properties (1vꢀz; Table 1) and NNMT inhibition. The results
revealed that the activity of N1ꢀsubstituted quinoliniums was inversely correlated with the size of
the substituent. For example, N1ꢀethyl quinolinium analog 1v was an ~2ꢀfold weaker inhibitor
compared to 1ꢀMQ (1a) and analogs with bulky substituents (1wꢀz) showed very poor inhibition
(IC > 1000 ꢁM). These data suggest that small moieties such as methyl or ethyl groups directly
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
adjacent to the N1ꢀposition may be crucial for maintaining potent NNMT inhibitory activity of
the quinolinium scaffold.
To further develop an SAR model for the quinolinium scaffold, we next explored the
activities of analogs with dual (2a-o; Table 2) and multiꢀpositional substitutions (3a-h; Table 3).
Consistent with our previously observed inverse relationship between NNMT inhibition and N1ꢀ
substituent size, N1ꢀethyl quinolinium analogs with C4ꢀ and C8ꢀmethyl substituents (2aꢀb;
Table 2) showed lower activities compared to N1ꢀmethyl quinolinium congeners (1i and 1t;
Table 1). Substituting a hydroxyethyl (2c) for an ethyl group (2a) at the N1ꢀposition further
reduced the activity by ~3.5ꢀfold, reinforcing the observation that a smaller alkyl group (e.g.,
methyl) at the N1ꢀposition is valuable for maintaining good inhibitory potency. The relationship
between NNMT inhibition and likely steric effects from N1ꢀsubstitutions was further
exemplified by analogs 2eꢀi, which displayed IC values that either exceeded 1000 ꢁM or
5
0
showed no inhibition of NNMT, regardless of the pharmacophore (i.e., methyl, chloro and
hydroxyl groups) placed at the C6ꢀposition. Singleꢀpoint SAR studies indicated that a C8ꢀ
hydroxyl containing analog (1u; Table 1) decreased the IC value by ~50ꢀfold relative to a
5
0
8
ACS Paragon Plus Environment

Page 9 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
more favorable methyl substituent (1t), while twoꢀpoint SAR studies incorporating a hydroxyl
group at the C8ꢀposition and a relatively bulky C2ꢀoxopropyl at the N1ꢀposition (2d) did not
additively impair activity; instead, the IC value for 2d was ~2ꢀfold lower relative to 1u. This
5
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
may imply that substitutions at C8ꢀposition combined with branched groups at the N1ꢀposition
might alter the binding mode of the analog and disrupt the stericꢀdependent inhibition pattern
observed with bulky groups at the quinolinium N1ꢀposition.
Interestingly, when a fluorine was added to the C6ꢀposition of the C3ꢀaminoꢀanalog (1d),
we observed supraꢀadditive effects on NNMT inhibition; the C3ꢀamino C6ꢀfluoroꢀ1ꢀMQ analog
(
2j) had an IC of 1.2 ꢁM, which was ~2.5ꢀfold more potent than the nonꢀfluorinated derivative
50
1
d (IC = 2.9 ꢁM) and ~5ꢀfold more potent than the C6ꢀfluoroꢀ1ꢀMQ analog (1m; IC = 5.7
5
0
5
0
ꢁM), highlighting the significance of both C3ꢀamino and C6ꢀfluoro substituents in improving
activity. The improvement observed in NNMT inhibitory activity with the introduction of a
fluorine substituent is consistent with the general observation that electronꢀwithdrawing halogen
groups such as fluorine may engage in multipolar interactions (e.g., orthogonalꢀtype dipolar
interaction with carbonyl groups in the binding pocket of a target protein), causing improvements
2
7ꢀ28
in compound binding and affinity.
Analogs with CF substitutions have also been observed to
3
form strong multipolar intermolecular interactions and improve binding affinities relative to the
2
9
corresponding methyl derivatives. CF substituents were tested using several 1ꢀMQ analogs;
3
CF substituent (2k; Table 2) placed at C5ꢀposition within the 1ꢀMQ scaffold impaired activity
3
relative to an analog lacking this substituent (1t) and a CF substituent at the C8ꢀposition (2l;
3
Table 2) in addition to an impairing chloro substituent at the C4ꢀposition significantly
deteriorated the activity of the compound. Further, we tested for other potentially synergistic
effects by screening dualꢀsubstituent 1ꢀMQ analogs containing numerous small apolar and amino
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
pharmacophores. We found that the C2,3ꢀdiaminoꢀ1ꢀMQ analog 2m showed inhibitory activity
(
IC = 2.8 ꢁM; Table 2) that was similar to the C3ꢀamino 1ꢀMQ analog 1d (IC = 2.9 ꢁM;
50 50
Table 1) without additive improvements, reinforcing the observation that an aminoꢀsubstituent at
the C3ꢀposition is an important feature among 1ꢀMQ analogs to maintain potent inhibitory
activity. Similarly, other dimethyl substitutions in analogs (2n-o; Table 2) did not cause
synergistic improvements in compound activity compared to the corresponding monoꢀmethyl
substitutions (1b, 1n, 1t; Table 1), but instead rendered the dimethyl quinolinium compounds as
poor inhibitors of NNMT (5ꢀ50ꢀfold higher IC values, 2n-o; Table 2).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
In our studies of the activity of multiꢀpositionally substituted N1ꢀquinoliniums, a C6ꢀ
fluorinated analog 3a (IC = 3.6 ꢁM; Table 3) showed a ~2.5ꢀfold increase in NNMT inhibitory
50
activity compared to the nonꢀfluorinated derivative 2a (IC = 8.7 ꢁM; Table 2), consistent with
5
0
earlier observations that C6ꢀfluoro substitution positively impacts quinolinium inhibitory
activity. Substitution of the C6ꢀfluoro of analog 3a to a C6ꢀtrifluoromethyl group (3b) resulted in
a significant loss of compound inhibitory activity (IC > 1000 ꢁM), comparable to the drastic
5
0
reductions in NNMT inhibition observed when CF substituents were incorporated into the
3
quinolinium scaffold (analogs 2kꢀl; Table 2).
General synthetic approaches were applied to improve the lipophilicity and drugꢀlike
properties (e.g., permeability) of this class of quinoliniumꢀbased NNMT inhibitors, in which the
lipophilicity was evaluated in silico by measuring the calculated logarithmic form of the partition
30
coefficient using ChemAxon software. To this end, we tested several perꢀmethylated
quinolinium analogs with methyl substituents at multiple ring positions (C2ꢀC8). Similar to the
results obtained with dimethylꢀsubstituted analogs, the addition of multiple methyl substituents
largely decreased the ability of the analogs to inhibit NNMT (3c-f; Table 3). A C2, C4, C8ꢀ
1
0
ACS Paragon Plus Environment

Page 11 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
trimethylꢀanalog 3c showed ~5ꢀ64ꢀfold lower inhibitory activity (IC = 115.4 ꢁM) as compared
5
0
to monoꢀmethylated analogs such as 1b (IC = 21 ꢁM), 1h (IC = 7.5 ꢁM) and 1q (IC = 1.8
50
50
50
ꢁ
M). Graphical analysis of molecular docking simulations suggested that multiꢀmethylated 1ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MQ analogs make intermolecular apolar contacts with several hydrophobic residues within the
NNMT binding site that are likely detrimental to the overall binding energetics between the
ligand and enzyme; this conclusion was supported by poor docking scores generated by Vina
ligand docking program. Furthermore, unlike the activity improvements observed for 1ꢀMQ
analogs with C6ꢀfluoro group substitutions (see compounds 1k, 2j, 3a), the addition of a fluoro
group to the C5 or C7 positions of a perꢀmethylated quinolinium scaffold (3c) did not improve
analog inhibitory potency (3g, IC = 94.6 ꢁM; 3h, IC = 109.2 ꢁM; Table 3) suggesting that
50
50
the C6ꢀposition may be highly “fluorophillic” to promote dipoleꢀtype interactions within the
NNMT substrateꢀbinding site. However, additional analogs with positional fluoro substitutions
within the 1ꢀMQ scaffold need to be tested to confirm this hypothesis.
Biological evaluation of N2-methyl isoquinolinium analogs. Previous studies have shown that
the isoquinoline scaffold is amendable to the development of drugꢀlike small molecule
31ꢀ32
1, 25
inhibitors.
Since isoquinoline is also an NNMT substrate,
we examined the NNMT
inhibitory activities of a series of analogs designed around the N2ꢀmethyl isoquinolinium
chemotype (Table 4). NNMT methylates isoquinoline to produce the reaction product N2ꢀ
1, 25
methyl isoquinolinium (4a),
which inhibits NNMT with an IC of 14.9 ꢁM similar to that of
50
1
ꢀMQ (1a; IC = 12.1 ꢁM). Guided in part by the SAR established for the 1ꢀMQ scaffold
50
(Tables 1-3) and molecular docking calculations, we explored the activities of a small series of
monoꢀsubstituted isoquinoline analogs (4b-f; Table 4). As summarized in Table 4, a C3ꢀamino
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
substituent (4b) improved compound activity ~2.4ꢀfold (IC = 6.3 ꢁM) compared to the parent
5
0
N2ꢀmethyl isoquinolinium (4a); the relative improvement in activity that resulted from adding a
C3ꢀamino group to the isoquinoline parent was smaller compared to the activity increase
observed when the substitutional congener was added to the 1ꢀMQ scaffold (1d, IC = 2.9 ꢁM).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
Instead, 4b had activity similar to analog 1c (IC = 6.3 ꢁM); both analogs have an amino group
5
0
adjacent to the Nꢀmethyl moiety, which could potentially shift the binding orientations of these
analogs and disrupt some of the hydrophobic interactions within the NNMT substrateꢀbinding
site (a detailed discussion is presented in the molecular docking section below). Incorporation of
a bromo group at the C4ꢀposition (4c) or an amino group at the C6ꢀposition (4d) reduced the
potency of these analogs by ~2ꢀfold relative to the parent analog (4a), while adding a bromo
group at the C6ꢀposition (4e) significantly decreased analog activity (IC = 505.7 ꢁM). In
5
0
contrast to the 1ꢀMQ chemotype (1a) where a C7ꢀamino (1n) improved analog potency by 5ꢀ
fold, a comparable amino substitution pattern on the N2ꢀmethyl isoquinolinium scaffold
decreased analog potency by ~3ꢀfold (4f; IC = 39.4 ꢁM). These observations are consistent
5
0
with calculated changes to the binding orientations and intermolecular interactions between
NNMT and N1ꢀMQ and N2ꢀmethyl isoquinolinium analogs, respectively (refer to molecular
docking discussion below).
Inhibitory activities of 1-methylpyridinium analogs. 1ꢀMethylnicotinamide (1ꢀMNA; 5a) is
2, 15
the endogenous product of the NNMT catalyzed reaction and an established NNMT inhibitor.
3
3
Using a recently developed sensitive fluorescenceꢀbased assay we determined the IC of 1ꢀ
50
MNA to be 9.0 ꢁM (Table 5), which is comparable to the IC value determined for 1ꢀMQ (1a).
5
0
Based on this result, 1ꢀMNA analogs with dimethyl groups at C4ꢀ and C6ꢀpositions (5b) and a
1
2
ACS Paragon Plus Environment

Page 13 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
sterically demanding C2ꢀacylꢀthiofuranyl group at the N1ꢀposition (5c) were tested. Compound
5
b exhibited slightly decreased potency (IC = 11.8 ꢁM) relative to 5a (Table 5) and was
50
similar to that of 1ꢀMQ (1a). Again, consistent with the general observations that sterically
hindered substituents at the N1ꢀposition significantly reduce the activity of analogs within the
NNMT substrateꢀbinding site, analog 5c was observed to have very poor activity (IC > 1000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
0
ꢁ
M).
Inhibitory activities of benzimidazoliums/benzothiazoliums and other structural analogs.
Lastly, other structurallyꢀrelated chemotypes, i.e., compounds containing benzimidazolium and
benzothiazolium scaffolds (6aꢀf; Table 6) and other heteroaryl templates (7aꢀi; Table 7) were
tested for NNMT inhibition. Compound 6a, with apolar methyl group substituents at both the
N1ꢀ and N3ꢀpositions of the benzimidazole scaffold exhibited an IC value of 16.7 ꢁM, which
5
0
was ~2ꢀfold lower relative to 1ꢀMNA (5a), 1ꢀMQ (1a), and N2ꢀmethyl isoquinolinium analog
4a). Consistent with results observed for additive groups to the 1ꢀMQ scaffold, an ethyl
substitution at the N1ꢀposition (6b) was well tolerated and generated comparable activity (IC =
(
5
0
22.8 ꢁM) to the parent structure 6a. This is supported, in part, by overlapping intermolecular
interactions that are predicted to occur between ligands containing distinct core scaffolds (1ꢀMQ,
N2ꢀmethyl isoquinolinium, pyridinium, and benzimidazolium/benzothiazolium) and NNMT
substrateꢀbinding site residues; however, molecular docking studies also revealed dissimilar
interactions for the ligands with the different scaffolds within the NNMT substrateꢀbinding site
(refer to molecular docking discussion below). For example, the C3ꢀamino benzimidazolium
analog 6c demonstrated a ~5ꢀfold higher IC value (IC = 82.4 ꢁM) relative to the parent
5
0
50
structure 6a; this was in contrast to the 1ꢀMQ and N2ꢀmethyl isoquinolinium scaffolds where an
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
amino group at the C3ꢀposition improved ligand activity (1d and 4b). Other analogs composed
of the benzimidazolium (6d) and benzothiazolium scaffolds (6e-f) were very poor inhibitors of
NNMT (IC > 1000 ꢁM). As summarized in Table 7, only analogs (7a-b) containing
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
phenanthridinium and acridinium structures showed even modest ability to inhibit NNMT (IC50
values, 100ꢀ300 ꢁM); all other tested chemotypes (7c-i) showed very poor NNMT inhibition
(
IC > 1000 ꢁM). Importantly, analogs containing tetrahydroquinoline/quinolone scaffolds (e.g.,
50
7d, 7f) that were uncharged exhibited very poor NNMT inhibition, suggesting that a positive
charge at the N1ꢀposition may be a necessary chemical feature for small molecule inhibitors of
NNMT.
Structure-based docking between NNMT and small molecule inhibitors. The Vina computer
docking program was able to reproduce the NCA molecular orientation observed in the Xꢀray
24
structure of the NNMTꢀNCAꢀSAH ternary complex with an all atom rootꢀmeanꢀsquare
deviation of 0.74 Å, suggesting that these calculations would be suitable to investigate likely
orientations and intermolecular interactions of small molecule inhibitors bound to NNMT.
Binding orientations and docking scores between NNMT and each of the ~44 combined
quinolinium, isoquinolinium, pyridinium, and benzimidazolium analogs containing Nꢀmethyl
substitutions, respectively, were calculated. As shown in Figure 2 for the combined set of
2
analogs, a positive linear correlation (r = 0.63, R = 0.4, P[twoꢀtailed] = 0.0003) was observed
between Vina docking score and log(IC ) values. In these calculations, low (i.e., more negative)
5
0
docking scores suggest more favorable binding conformations and intermolecular interactions
between the ligand and macromolecular target. Consistent with this observation, analogs that
were potent NNMT inhibitors (e.g., compounds 1k, 1t; IC = 1.2ꢀ1.8 ꢁM) had the lowest
5
0
1
4
ACS Paragon Plus Environment

Page 15 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
calculated Vina docking scores (scores between ꢀ8.3 and ꢀ8.1). Conversely, many analogs (e.g.,
perꢀmethylated compounds 3e-f) that were either poor NNMT inhibitors (IC > 1000 ꢁM) or
50
failed to inhibit NNMT had higher calculated Vina docking scores (scores between ꢀ6.0 and ꢀ5.0)
or did not bind in the nicotinamide binding pocket. Taken together, Vina docking scores and
inhibitor potencies were modestly correlative for many analogs containing the N1ꢀmethyl
quinolinium scaffold (Figure 2), suggesting that docking calculations could serve as a predictive
tool to guide the identification and design of small molecule NNMT inhibitors similar to other
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4
reported systems. However, there were exceptional N1ꢀmethylquinolium ligands and analogs
containing other chemical scaffolds (e.g., N2ꢀmethyl isoquinolinium, benzimidazolium) that
were inconsistent, likely, in part, due to inherent limitations of the Vina computational program
and/or inherent analog properties. The Vina program makes assumptions on the protonation
states/partial charge distribution in molecules not accounting for solvent, dipole, entropic, and
3
5
tautomerization effects. As a result, disparity between calculated docking scores and
experimentally generated IC values were observed for some of the tested analogs. For example,
5
0
analogs with relatively good docking scores (e.g., analogs 1s, 2l; docking scores between ꢀ7 and
ꢀ8) exhibited poor IC values (> 500 ꢁM) and improvements imparted by C6ꢀfluoro
5
0
substitutions in the IC values compared to the nonꢀfluorinated derivatives, presumably via
5
0
enhanced dipolar interactions, were not truly reflected in the docking scores generated by the
Vina calculations [e.g., analogs 1a (IC = 12.1 ꢁM) and 1m (IC = 5.7 ꢁM) had similar Vina
5
0
50
docking scores ~ ꢀ7.7)].
Vina docking calculations can predict likely orientations and conformations of small
molecules bound to a target protein. 1ꢀmethylquinolinium (1a) was used to establish a
prototypical structure for 1ꢀMQ analogs bound to the NNMT substrateꢀbinding pocket (Figure
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
). The predicted binding orientation of 1ꢀMQ (and related analogs) was similar to the bound
structures of the endogenous substrate NCA and docked 1ꢀMNA, with the N1ꢀatoms of these
ligands superimposable and with similar intermolecular interactions to residues within the
NNMT substrateꢀbinding pocket (Figure 3A, B). Moreover, the inhibitors were positioned in the
nicotinamide binding pocket such that their N1ꢀmethyl groups were directed towards the
adjacent SAM binding pocket (Figure 3C). 1ꢀMQ and related analogs are predicted to form
strong π−π, cationꢀπ, and hydrophobic interactions with residues Tyr204 and Leu164 that define
the apolar pocket surrounding the quinolinium N1ꢀcontaining ring (Figure 3B, D), similar to the
manner in which these aromatic residues interact with the pyridine ring of NCA in the NNMT
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
substrateꢀbinding site. In addition, NNMT residues Ala198, Tyr242, and Tyr 24 were also
indicated to contact via hydrophobic interactions with the quinolinium ring of 1ꢀMQ (Figure 3B,
D), two of which (Ala198 and Tyr242) have also been indicated to contact with the pyridinium
ring of the substrate NCA and 1ꢀMNA (Figure 3A). Further, the NCA amide group has been
reported to form hydrogen bonds with the hydroxyl groups of Ser201/Ser213 residues and the
2
4 24
Asp197 carboxylate.
The predicted binding orientation of quinolinium analog 1k, however,
suggests that the C5ꢀamino group forms hydrogen bonds with Ser201, and hydrophobic
interactions with Ser213 and Tyr20 (figure not shown). These distinct predicted interactions
between analog 1k and NNMT compared to NCA suggest more favorable intermolecular
bonding interactions with residues within the NNMT substrateꢀbinding pocket, which is
supported by a lower Vina docking score (Vina score = ꢀ8.1) for 1k relative to 1ꢀMNA (Vina
score = ꢀ6.8), and the potent inhibitory activity exhibited by this analog (IC = 1.2 ꢁM).
5
0
The predicted conformations, binding orientations, and intermolecular interactions with
NNMT residues for the most potent analogs with the pyridinium scaffold (1,2,4ꢀtrimethylꢀ5ꢀ
1
6
ACS Paragon Plus Environment

Page 17 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
carboxamidepyridinium, 5b), benzimidazolium scaffold (1,2,3ꢀtrimethylꢀ1Hꢀ3,1ꢀbenzimidazolꢀ
3ꢀium, 6a), and N2ꢀmethyl isoquinolinium scaffold (3ꢀamineꢀ2ꢀmethylisoquinolinim, 4b) were
examined and are summarized below. Inhibitors with different chemotypes were predicted to
bind NNMT via hydrophobic interactions with the same aromatic residues (Leu164 and Tyr204)
in the apolar substrateꢀbinding pocket and make additional contacts with apolar atoms of Tyr24,
Tyr242, and Ala198 around the respective scaffold ring structures, relative to the prototype
analog 1k. Analogs 5b and 6a with the pyridinium and benzimidazolium scaffolds, respectively,
either formed only hydrogen bonds or hydrophobic contacts with Ser201/213 residues, and made
additional hydrophobic contacts with NNMT residues (e.g., 5b, Asp167, Ala247; 6a, Ala168,
Asp167, Asp197) compared to 1k, which potentially rendered these inhibitors less potent (at
least ~10ꢀfold lower potency relative to 1k). Lastly, the predicted orientation of isoquinolinium
analogs (e.g., 4b) was shifted within the NNMT substrateꢀbinding pocket, relative to the
quinolinium analogs. Specifically, the carbon atoms around the N2ꢀmethyl isoquinolinium rings
were predicted to form distinct hydrophobic interactions with NNMT residues Tyr203 and
Ala247, and more importantly form hydrogen bond and hydrophobic interactions with the phenol
hydroxyl group and ring carbon atoms in the backbone of Tyr20 residue, respectively. Given that
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
4
Tyr20 is critical for the functional activity of NNMT and the alignment of both methyl donor
2
4
and acceptor molecules (SAM and NCA) in the binding pocket, interactions with this residue
might be relevant in guiding the inhibitor activity.
Summary of SAR for 1-MQ analogs. The most potent NNMT inhibitor analogs identified in
the current study (e.g., 1k, 2j; IC ~1.2 ꢁM) contained the quinolinium scaffold. These
5
0
compounds were predicted to bind the NNMT substrateꢀbinding pocket analogous to NCA and
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
exhibited favorable intermolecular interactions with NNMT residues. A predictive SAR
developed for this series of analogs highlights a few important chemical features required for
potent NNMT inhibition as summarized in Figure 4: (i) apolar methyl substitutions at N1ꢀ and
C8ꢀpositions promote strong hydrophobic interactions with aromatic NNMT residues (Leu164,
Tyr204) and the positive charge on the N1ꢀatom is required to retain potent inhibitory activity;
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(ii) methyl or amino substituents at C2–C5ꢀ, and C7ꢀpositions provide favorable C and N atoms
that can contact with Tyr (Tyr20, Tyr24, Ty242) and Ser (Ser201 and Ser213) residues via
multiple hydrophobic and/or hydrogen bond interactions; and (iii) an electronꢀwithdrawing
fluoro substitution at the C6ꢀposition engages in multipolar interactions and significantly
enhances analog potency.
CONCLUSIONS
Exploration of an initial series of commercially available and synthesized Nꢀmethylated analogs
of different chemotypes, including quinolinium, isoquinolinium, pyrididium, and
benzimidazolium/benzothiazolium, resulted in the rapid development of structureꢀactivity
relationships (SARs) for NNMT inhibitors demonstrating >1000ꢀfold range of activity.
Computerꢀbased docking of inhibitors to the NNMT NCA substrateꢀbinding site produced a
robust correlation between ligandꢀenzyme interaction docking scores and experimentally
established IC values. Computational calculations and predicted binding orientation of
5
0
inhibitors enabled identification of NNMT residues responsible for proteinꢀligand intermolecular
interactions. The quinolinium analog series that demonstrated significant NNMT inhibition were
further expanded by synthesis, guided in part by calculations to identify analogs with improved
docking scores and binding. Potent quinolinium analogs identified in the current study exhibited
1
8
ACS Paragon Plus Environment

Page 19 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
very low micromolar potency (IC ~1 ꢁM) that fully correlated with very low ligandꢀenzyme
5
0
docking scores. The SAR described herein demonstrates quinolinium as a promising scaffold to
selectively bind to the NNMT substrateꢀbinding site with a confirmation analogous to the
endogenous NNMT substrate NCA and highlights some of the key structural features of
quinolinium analogs for NNMT inhibition. These SAR results motivate future exploration of the
chemical space surrounding the template moieties to develop potent, selective, and drugꢀlike
NNMT inhibitors. Ongoing research will investigate selectivity, confirm inhibitor activities using
orthogonal assays, and develop potent NNMT inhibitors with improved “drugꢀlike”
physicochemical properties. Future studies will elucidate inhibitor mechanismꢀofꢀaction and the
structures of the NNMTꢀSAMꢀinhibitor ternary complexes, which will validate the molecular
predictions derived from the current findings.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Chemistry. Analogs and Chemicals. PubChem and Zinc similarity search programs were
employed to identify commercially available analogs containing quinolinium, isoquinolinium,
pyridinium, benzimidazolium, and benzothiazolium core structures and the analogs were
purchased from established commercial suppliers, including Sigma Aldrich (St. Louis, MO),
Chembridge Corporation (San Diego, CA), ChemDiv (San Diego, CA), and Specs (Hopkinton,
RI), Cayman (Ann Arbor, MI) and Pfaltz & Bauer (Waterbury, CT). The identity of all the tested
1
compounds was confirmed by H NMR and HPLCꢀMS, and the purity was ensured to be ≥ 95%
(
see Supporting Information for representative examples of HPLC analyses for compound
purity). SAM was obtained from Sigma Aldrich and nicotinamide from Fluka Analytical
Kwazulu Natal, South Africa; distributed by Sigma Aldrich in the USA). 1ꢀMNA chloride and
(
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Sꢀ5'ꢀadenosylhomocysteine (SAH) were obtained from Cayman Chemical Company (Ann
Arbor, MI). All compounds were made in double distilled water.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
General procedures. Unless otherwise indicated all reactions were conducted in standard
commercially available glassware using standard synthetic chemistry methods and setup. All airꢀ
and moistureꢀsensitive reactions were performed under nitrogen atmosphere with dried solvents
and glassware under anhydrous conditions. Starting materials and reagents were commercial
compounds of the highest purity available and were used without purification. Solvents used for
reactions were indicated as of commercial dry or extraꢀdry or analytical grade. Analytical thin
layer chromatography was performed on aluminium plates coated with Merck Kieselgel 60F254
and visualized by UV irradiation (254 nm) or by staining with a solution of potassium
permanganate. Flash column chromatography was performed on Biotage Isolera One 2.2 using
commercial columns that were preꢀpacked with Merck Kieselgel 60 (230– 400 mesh) silica gel.
1
Final compounds for biological testing are all ≥95% purity as determined by HPLCꢀMS and H
NMR.
1
NMR. H NMR experiments were recorded on Agilent DD2 400MHz spectrometers at ambient
temperature. Samples were dissolved and prepared in deuterated solvents (CDCl , CD OD and
3
3
DMSOd ) with residual solvents being used as the internal standard in all cases. All deuterated
6
solvent peaks were corrected to the standard chemical shifts (CDCl , d = 7.26 ppm; CD OD, d
H
3
H
3
=
3.31 ppm; DMSOꢀd , d = 2.50 ppm). Spectra were all manually integrated after automatic
6 H
baseline correction. Chemical shifts (d) are given in parts per million (ppm), and coupling
constants (J) are given in Hertz (Hz). The proton spectra are reported as follows: d (multiplicity,
2
0
ACS Paragon Plus Environment

Page 21 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
coupling constant J, number of protons). The following abbreviations were used to explain the
multiplicities: app = apparent, b = broad, d = doublet, dd = doublet of doublets, ddd = doublet of
doublet of doublets, dddd = doublet of doublet of doublet of doublets, m = multiplet, s = singlet,
t = triplet.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
HPLC-MS. All samples were analyzed on Agilent 1290 series HPLC system comprised of binary
pumps, degasser, UV detector, and autoꢀsampler coupled to an Agilent 6150 mass spectrometer.
Purity was determined via UV detection with a bandwidth of 170nm in the range from 230ꢀ
400nm. The general LC parameters were as follows: Column ꢀ Zorbax Eclipse Plus C18, size 2.1
X 50 mm; Solvent A: 0.10 % formic acid in water, Solvent B: 0.00 % formic acid in acetonitrile;
Flow rate – 0.7 mL/min; Gradient: 5 % B to 95 % B in 5 min and hold at 95 % B for 2 min; UV
detector – channel 1 = 254 nm, channel 2 = 254 nm. Mass detector Agilent Jet Stream – Electron
Ionization (AJSꢀES).
General Procedure A: quinolinyl or isoquinolinyl ring N-alkylation using MeI. A mixture of
appropriate quinoline or isoquinoline derivative (1 equiv) and MeI (1.5 equiv unless otherwise
indicated) in IPA (0.5M) was heated at 90 ºC for 12 h. The reaction was cooled to ambient
temperature and the resulting precipitate was isolated by vacuum filtration, washed with a
mixture of IPA/Et O (v:v/1:1), and dried in vacuo.
2
General Procedure B: quinolinyl or isoquinolinyl ring N-alkylation using MeOTf. A
mixture of appropriate quinoline or isoquinoline derivative (1 equiv) and MeOTf (3 equiv unless
otherwise indicated) in toluene (0.2M) was heated at 100 ºC for 12 h. The reaction was cooled to
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ambient temperature and added Et O to induce precipitation. The resulting precipitate was
2
isolated by vacuum filtration, washed with Et O, and dried in vacuo.
2
1,2-dimethylquinolin-1-ium trifluoromethanesulfonate (1b). According to general procedure
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
B, the title compound was obtained as pale orange powder (89% yield). H NMR (400 MHz,
DMSOꢀd ) δ 9.08 (d, J = 8.4 Hz, 1H), 8.58 (d, J = 9.2 Hz, 1H), 8.39 (d, J = 8.0 Hz, 1H), 8.22
6
(dd, J = 8.4, 7.6 Hz, 1H), 8.10 (d, J = 8.8 Hz, 1H), 7,99 (dd, J = 8.4, 7.6 Hz, 1H), 4.44 (s, 3H),
3
.07 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 1.13 min, m/z 158.2 [MꢀOSO CF ].
2 3
2-amino-1-methylquinolin-1-ium iodide (1c). According to general procedure A, the title
1
compound was obtained as grey powder (26% yield). H NMR (400 MHz, DMSOꢀd ) δ 9.45 (br,
6
1
H), 8.87 (br, 1H), 8.34 (d, J = 9.2 Hz, 1H), 8.01 (d, J = 8.8 Hz, 1H), 7.98 (d, J = 8.8 Hz, 1H),
.87 (dd, J = 8.4, 7.6 Hz, 1H), 7.57 (dd, J = 7.6, 7.6 Hz, 1H), 7.17 (d, J = 9.2 Hz, 1H), 3.87 (s,
7
3H); HPLCꢀMS (AJSꢀES): Rt 1.31 min, m/z 159.1 [MꢀI].
3-amino-1-methylquinolin-1-ium iodide (1d). According to general procedure A, the title
1
compound was obtained as orange powder (33% yield). H NMR (400 MHz, DMSOꢀd ) δ 8.81
6
(s, 1H), 8.21 (d, J = 8.8 Hz, 1H), 8.07 (d, J = 7.2 Hz, 1H), 7.98 (s, 1H), 7.78 (m, 2H), 6.56 (br,
2H), 4.51 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 1.08 min, m/z 159.1 [MꢀI].
1,3-dimethylquinolin-1-ium trifluoromethanesulfonate (1e). According to general procedure
1
A, the title compound was obtained as white powder (93% yield). H NMR (400 MHz, DMSOꢀ
d6) δ 9.47 (s, 1H), 9.08 (s, 1H), 8.46 (d, J = 9.2 Hz, 1H), 8.35 (d, J = 9.2 Hz, 1H), 8.21 (ddd, J =
2
2
ACS Paragon Plus Environment

Page 23 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
7
.6, 6.8, 1.6 Hz, 1H), 8.02 (ddd, J = 7.6, 7.6, 0.8 Hz, 1H), 4.60 (s, 3H), 2.64 (s, 3H); HPLCꢀMS
(AJSꢀES): Rt 0.39 min, m/z 158.1 [MꢀOSO CF ].
2 3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
-methyl-3-(methylamino)quinolin-1-ium iodide (1f). According to general procedure A, the
1
title compound was obtained as orange powder (72% yield). H NMR (400 MHz, DMSOꢀd ) δ
6
8
.90 (d, J = 2.4 Hz, 1H), 8.22 (dd, J = 4.8, 4.4 Hz, 1H), 8.10 (dd, J = 4.8, 4.4 Hz, 1H), 7.97 (d, J
1.6 Hz, 1H), 7.78 (dd, J = 4.8, 4.4 Hz, 1H), 7.13 (br, 1H), 4.54 (s, 3H), 2.90 (s, 3H); HPLCꢀMS
(AJSꢀES): Rt 0.78 min, m/z 173.1 [MꢀI].
=
1-methyl-3-(phenylamino)quinolin-1-ium iodide (1h). According to general procedure A, the
1
title compound was obtained as orangeꢀyellow powder (59% yield). H NMR (400 MHz,
DMSOꢀd ) δ 9.33 (s, 1H), 9.17 (d, J = 2.4 Hz, 1H), 8.57 (d, J = 2.0 Hz, 1H), 8.31 (d, J = 8.8 Hz,
6
1
H), 8.23 (d, J = 8.0 Hz, 1H), 7.93 (dd, J = 8.0, 7.6 Hz, 1H), 7.85 (dd, J = 8.0, 6.8 Hz, 1H), 7.44
(d, J = 8.0 Hz, 1H), 7.41 (d, J = 7.6 Hz, 1H), 7.34 (s, 1H), 7.32 (s, 1H), 7.11 (dd, J = 8.0, 7.2 Hz,
1H), 4.59 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 1.48 min, m/z 235.2 [MꢀI].
4-amino-1-methylquinolin-1-ium iodide (1j). According to general procedure A, the title
1
compound was obtained as grey powder (78% yield). H NMR (400 MHz, DMSOꢀd ) δ 8.97 (s,
6
2H), 8.51 (d, J = 7.6 Hz, 1H), 8.48 (d, J = 8.4 Hz, 1H), 8.06 (d, J = 3.6 Hz, 2H), 7.77 (m, 1H),
6.78 (d, J = 7.2 Hz, 1H), 4.11 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 0.69 min, m/z 159.1 [MꢀI].
5-amino-1-methylquinolin-1-ium iodide (1k). According to general procedure A, the title
compound was obtained using stoichiometric amount of MeI (1 equiv) to isolate the product as
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
red powder (62% yield). H NMR (400 MHz, DMSOꢀd ) δ 9.32 (d, J = 8.4 Hz, 1H), 9.28 (d, J =
6
5
7
1
.6 Hz, 1H), 7.89 (dd, J = 8.4, 8.0 Hz, 1H), 7.85 (d, J = 5.6 Hz, 1H), 7.35 (d, J = 8.8 Hz, 1H),
.11 (br, 2H), 7.02 (d, J = 8.0 Hz, 1H), 4.42 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 0.39 min, m/z
59.1 [MꢀI].
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
,5-dimethylquinolin-1-ium trifluoromethanesulfonate (1l). According to general procedure
1
B, the title compound was obtained as white powder (99% yield). H NMR (400 MHz, DMSOꢀ
d6) δ 9.48 (d, J = 5.6 Hz, 1H), 9.35 (d, J = 8.8 Hz, 1H), 8.35 (d, J = 8.8 Hz, 1H), 8.17 (dd, J =
8
.4, 1.6 Hz, 1H), 8.16 (d, J = 8.8 Hz, 1H), 7.90 (d, J = 7.2 Hz, 1H), 4.63 (s, 3H), 2.86 (s, 3H);
HPLCꢀMS (AJSꢀES): Rt 0.41 min, m/z 158.1 [MꢀOSO CF ].
2
3
6
-fluoro-1-methylquinolin-1-ium iodide (1m). According to general procedure A, the title
1
compound was obtained as yellow powder (75% yield). H NMR (400 MHz, DMSOꢀd ) δ 9.52
6
(
s, 1H), 9.23 (d, J = 8.8 Hz, 1H), 8.64 (dd, J = 9.6, 4.4 Hz, 1H), 8.37 (dd, J = 8.8, 2.8 Hz, 1H),
.26 (dddd, J = 9.2, 9.2, 2.8, 2.0 Hz, 1H), 8.21 (dd, J = 8.4, 5.6 Hz, 1H), 4.66 (s, 3H); HPLCꢀMS
(AJSꢀES): Rt 0.37 min, m/z 162.1 [MꢀI].
8
6
-amino-1-methylquinolin-1-ium iodide (1o). According to general procedure A, the title
1
compound was obtained as orangeꢀbrown powder (58% yield). H NMR (400 MHz, DMSOꢀd ) δ
6
8
.94 (s, 1H), 8.77 (d, J = 9.2 Hz, 1H), 8.18 (d, J = 9.2 Hz, 1H), 7.83 (m, 1H), 7.58 (d, J = 9.2 Hz,
1
H), 7.11 (s, 1H), 6.46 (br, 2H), 4.48 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 0.74 min, m/z 159.1 [Mꢀ
I].
2
4
ACS Paragon Plus Environment

Page 25 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
7
-amino-1-methylquinolin-1-ium iodide (1q). According to general procedure A, the title
compound was obtained using stoichiometric amount of MeI (1 equiv) to isolate the product as
1
orangeꢀbrown powder (56% yield). H NMR (400 MHz, DMSOꢀd ) δ 8.92 (d, J = 5.6 Hz, 1H),
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
.71 (d, J = 8.0 Hz, 1H), 8.02 (d, J = 9.2 Hz, 1H), 7.46 (dd, J = 7.6, 6.4 Hz, 1H), 7.38 (br, 2H),
7.29 (dd, J = 9.2, 1.6 Hz, 1H), 6.93 (d, J = 1.2 Hz, 1H), 4.23 (s, 3H); HPLCꢀMS (AJSꢀES): Rt
.52 min, m/z 159.1 [MꢀI].
0
1,7-dimethylquinolin-1-ium trifluoromethanesulfonate (1r). According to general procedure
1
B, the title compound was obtained as white powder (87% yield). H NMR (400 MHz, DMSOꢀ
d6) δ 9.41 (d, J = 6.0 Hz, 1H), 9.20 (d, J = 8.4 Hz, 1H), 8.36 (d, J = 8.4 Hz, 1H), 8.34 (s, 1H),
8.08 (dd, J = 8.4, 6.0 Hz, 1H), 7.91 (d, J = 8.4 Hz, 1H), 4.59 (s, 3H), 2.72 (s, 3H); HPLCꢀMS
(AJSꢀES): Rt 0.41 min, m/z 158.1 [MꢀOSO CF ].
2 3
3
-amino-6-fluoro-1-methylquinolin-1-ium iodide (2j). According to general procedure A, the
1
title compound was obtained as yellow powder (58% yield). H NMR (400 MHz, DMSOꢀd ) δ
6
8.84 (d, J = 2.0 Hz, 1H), 8.35 (dd, J = 9.6, 4.4 Hz, 1H), 7.99 (dd, J = 9.2, 2.8 Hz, 1H), 7.94 (d, J
=
2.0 Hz, 1H), 7.76 (ddd, J = 8.8, 8.8, 3.2 Hz, 1H), 6.78 (br, 2H), 4.56 (s, 3H); HPLCꢀMS (AJSꢀ
ES): Rt 0.22 min, m/z 177.1 [MꢀI].
1
,8-dimethyl-5-(trifluoromethyl)quinolin-1-ium trifluoromethanesulfonate (2k). According
to general procedure B, the title compound was obtained using excess amount of MeOTf (5
1
equiv) to isolate the product as pale grey powder (69% yield). H NMR (400 MHz, DMSOꢀd ) δ
6
9.56 (d, J = 6.0 Hz, 1H), 9.24 (d, J = 8.8 Hz, 1H), 8.38 (d, J = 7.6 Hz, 1H), 8.29 (dd, J = 8.8, 5.6
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Hz, 1H), 8.21 (d, J = 7.6 Hz, 1H), 4.89 (s, 3H), 3.15 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 0.92 min,
m/z 226.1 [MꢀOSO CF ].
2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4-chloro-1-methyl-8-(trifluoromethyl)quinolin-1-ium trifluoromethanesulfonate (2l).
According to general procedure B, the title compound was obtained using excess amount of
1
MeOTf (5 equiv) to isolate the product as pale grey powder (88% yield). H NMR (400 MHz,
DMSOꢀd ) δ 8.56 (m, 1H), 8.33 – 8.10 (m, 2H), 7.59 (m, 1H), 6.47 (m, 1H), 3.93 (m, 3H);
6
HPLCꢀMS (AJSꢀES): Rt 3.01 min, m/z 246.1 [MꢀOSO CF ].
2
3
2,3-diamino-1-methylquinolin-1-ium iodide (2m). According to general procedure A, the title
1
compound was obtained as tan powder (30% yield). H NMR (400 MHz, DMSOꢀd ) δ 8.76 (br,
6
2H), 7.89 (d, J = 8.4 Hz, 1H), 7.72 (d, J = 8.0 Hz, 1H), 7.54 (dd, J = 8.0, 7.2 Hz, 1H), 7.43 (dd, J
=
7.6, 7.6 Hz, 1H), 7.35 (br, 1H), 5.94 (br, 2H), 3.95 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 1.41 min,
m/z 174.1 [MꢀI].
1,2,8-trimethylquinolin-1-ium trifluoromethanesulfonate (2n). According to general
1
procedure B, the title compound was obtained as pale tan powder (84% yield). H NMR (400
MHz, DMSOꢀd ) δ 9.01 (d, J = 8.4 Hz, 1H), 8.18 (d, J = 8.0 Hz, 1H), 8.04 (d, J = 8.4 Hz, 1H),
6
8.02 (d, J = 8.4 Hz, 1H), 7.83 (dd, J = 7.6, 7.6 Hz, 1H), 4.45 (s, 3H), 3.03 (s, 3H), 2.97 (s, 3H);
HPLCꢀMS (AJSꢀES): Rt 0.31 min, m/z 172.1 [MꢀOSO CF ].
2
3
1,2,6-trimethylquinolin-1-ium iodide (2o). According to general procedure A, the title
1
compound was obtained as yellowꢀorange powder (36% yield). H NMR (400 MHz, DMSOꢀd )
6
2
6
ACS Paragon Plus Environment

Page 27 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
δ 8.98 (d, J = 8.4 Hz, 1H), 8.49 (d, J = 9.2 Hz, 1H), 8.16 (s, 1H), 8.08 (m, 1H), 8.06 (m, 1H),
4.42 (s, 3H), 3.05 (d, J = 4.0 Hz, 3H), 2.60 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 1.00 min, m/z 172.1
[
MꢀI].
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1,2,4,8-teramethylquinolin-1-ium trifluoromethanesulfonate (3c). According to general
1
procedure B, the title compound was obtained as white powder (94% yield). H NMR (400 MHz,
DMSOꢀd ) δ 8.25 (d, J = 8.4 Hz, 1H), 8.00 (d, J = 6.8 Hz, 1H), 7.96 (s, 1H), 7.84 (dd, J = 8.0,
6
7.6 Hz, 1H), 4.36 (s, 3H), 2.96 (s, 3H), 2.93 (s, 3H), 2.88 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 0.85
min, m/z 186.2 [MꢀOSO CF ].
2
3
1,2,4,5,8-pentamethylquinolin-1-ium trifluoromethanesulfonate (3d). According to general
1
procedure B, the title compound was obtained as tan powder (93% yield). H NMR (400 MHz,
DMSOꢀd ) δ 7.83 (s, 1H), 7.80 (d, J = 7.6 Hz, 1H), 7.61 (d, J = 8.0 Hz, 1H), 4.24 (s, 3H), 3.04
6
(
s, 3H), 2.91 (s, 6H), 2.78 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 1.01 min, m/z 200.2 [MꢀOSO CF ].
2 3
1,2,4,6,8-pentamethylquinolin-1-ium trifluoromethanesulfonate (3e). According to general
1
procedure B, the title compound was obtained as white powder (70% yield). H NMR (400 MHz,
DMSOꢀd ) δ 8.03 (s, 1H), 7.91 (s, 1H), 7.85 (s, 1H), 4.35 (s, 3H), 2.93 (s, 3H), 2.90 (s, 3H), 2.85
6
(s, 3H), 2.55 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 1.11 min, m/z 200.2 [MꢀOSO CF ].
2
3
1,2,4,7,8-pentamethylquinolin-1-ium trifluoromethanesulfonate (3f). According to general
1
procedure B, the title compound was obtained as white powder (87% yield). H NMR (400 MHz,
DMSOꢀd ) δ 8.13 (d, J = 8.4 Hz, 1H), 7.87 (s, 1H), 7.79 (d, J = 8.4 Hz, 1H), 4.28 (s, 3H), 2.94
6
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
s, 3H), 2.86 (s, 3H), 2.67 (s, 3H), 2.57 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 1.04 min, m/z 200.2
[
MꢀOSO CF ].
2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
-fluoro-1,2,4,8-tetramethylquinolin-1-ium trifluoromethanesulfonate (3g). According to
1
general procedure B, the title compound was obtained as tan powder (89% yield). H NMR (400
MHz, DMSOꢀd ) δ 7.99 (dd, J = 7.6, 6.8 Hz, 1H), 7.95 (s, 1H), 7.70 (dd, J = 11.2, 8.8 Hz, 1H),
6
4
.32 (s, 3H), 2.95 (s, 6H), 2.86 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 0.87 min, m/z 204.2 [Mꢀ
OSO CF ].
2
3
7
-fluoro-1,2,4,8-tetramethylquinolin-1-ium trifluoromethanesulfonate (3h). According to
1
general procedure B, the title compound was obtained as white powder (78% yield). H NMR
(400 MHz, DMSOꢀd ) δ 8.37 (dd, J = 9.6, 6.8 Hz, 1H), 7.95 (s, 1H), 7.89 (dd, J = 9.2, 9.2 Hz,
6
1
H), 4.35 (s, 3H), 2.96 (s, 3H), 2.89 (s, 3H), 2.74 (s, 3H); HPLCꢀMS (AJSꢀES): Rt 0.86 min, m/z
04.2 [MꢀOSO CF ].
2
2
3
3-amino-2-methylisoquinolin-2-ium iodide (4b). According to general procedure A, the title
1
compound was obtained as tanꢀorange powder (40% yield). H NMR (400 MHz, DMSOꢀd ) δ
6
9
.36 (s, 1H), 7.93 (d, J = 8.4 Hz, 1H), 7.81 (d, J = 8.8 Hz, 1H), 7.75 (dd, J = 6.4, 0.8 Hz, 1H),
7.71 (br, 2H), 7.41 (ddd, J = 7.6, 7.6, 0.8 Hz, 1H), 7.34 (s, 1H), 4.03 (s, 3H); HPLCꢀMS (AJSꢀ
ES): Rt 1.25 min, m/z 159.1 [MꢀI].
4
-bromo-2-methylisoquinolin-2-ium iodide (4c). According to general procedure A, the title
compound was obtained using excess MeI (5 equiv) to isolate the product as tanꢀyellow powder
2
8
ACS Paragon Plus Environment

Page 29 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
(
99% yield). H NMR (400 MHz, DMSOꢀd ) δ 10.09 (m, 1H), 9.24 (s, 1H), 8.56 (d, J = 8.0 Hz,
6
1
H), 8.43 – 8.33 (m, 2H), 8.17 (dd, J = 8.0, 8.0 Hz, 1H), 4.45 (d, J = 4.0 Hz, 3H).; HPLCꢀMS
(
AJSꢀES): Rt 0.63 min, m/z 222.0, 224.0 [MꢀI].
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6-amino-2-methylisoquinolin-2-ium iodide (4d). According to general procedure A, the title
1
compound was obtained as tan powder (68% yield). H NMR (400 MHz, DMSOꢀd ) δ 9.21 (s,
6
1H), 8.14 (dd, J = 6.8, 1.2 Hz, 1H), 8.00 (d, J = 8.8 Hz, 1H), 7.81 (d, J = 6.8 Hz, 1H), 7.35 (br,
2H), 7.27 (dd, J = 8.8, 2.0 Hz, 1H), 6.89 (d, J = 1.6 Hz, 1H), 4.16 (s, 3H); HPLCꢀMS (AJSꢀES):
Rt 0.78 min, m/z 159.1 [MꢀI].
6-bromo-2-methylisoquinolin-2-ium iodide (4e). According to general procedure A, the title
compound was obtained using excess MeI (5 equiv) to isolate the product as brownꢀgrey powder
1
(
99% yield). H NMR (400 MHz, DMSOꢀd ) δ 10.04 (s, 1H), 8.74 (d, J = 6.4 Hz, 1H), 8.69 (s,
6
1H), 8.49 (d, J = 6.4 Hz, 1H), 8.41 (d, J = 8.8 Hz, 1H), 8.21 (m, 1H), 4.45 (s, 3H).; HPLCꢀMS
(AJSꢀES): Rt 1.25 min, m/z 222.0, 224.0 [MꢀI].
7-amino-2-methylisoquinolin-2-ium iodide (4f). According to general procedure A, the title
1
compound was obtained as orangeꢀyellow powder (53% yield). H NMR (400 MHz, DMSOꢀd )
6
δ 9.48 (s, 1H), 8.23 (d, J = 6.8 Hz, 1H), 8.20 (d, J = 6.8 Hz, 1H), 7.99 (d, J = 8.8 Hz, 1H), 7.55
(dd, J = 8.8, 2.4 Hz, 1H), 7.13 (d, J = 2.0 Hz, 1H), 6.52 (br, 2H), 4.34 (s, 3H); HPLCꢀMS (AJSꢀ
ES): Rt 0.78 min, m/z 159.1 [MꢀI].
2
9
ACS Paragon Plus Environment