An N,N-Bis(benzimidazolylpicolinoyl)piperazine (BT-11): A Novel Lanthionine Synthetase C-Like 2-Based Therapeutic for Inflammatory Bowel Disease

Article abstract of DOI:10.1021/acs.jmedchem.6b00412

Lanthionine synthetase C-like 2 (LANCL2), a novel therapeutic target for inflammatory and autoimmune diseases and diabetes, exerts anti-inflammatory and insulin-sensitizing effects. This study reports the first LANCL2-based therapeutics for inflammatory bowel disease (IBD). Analogues of 1 (ABA) and 2 (NSC61610) were screened by molecular docking, then synthesized and analyzed for binding to LANCL2 by surface plasmon resonance. Piperazine-1,4-diylbis(6-benzo[d]imidazole-2-yl)pyridine-2-yl)methanone, 7, was identified as the lead LANCL2-binding compound for treating IBD. The oral treatment with 7 (8 mg/kg/d) in a mouse model of IBD resulted in lowering the disease activity index, decreasing colonic inflammatory lesions by 4-fold, and suppressing inflammatory markers (e.g., TNF-α, and interferon-γ) in the gut. Furthermore, studies in LANCL2-/- mice demonstrated that loss of LANCL2 abrogated beneficial actions of 7, suggesting high selectivity for the target. In conclusion, 7 merits continued development as a LANCL2-based, first-in-class orally active therapeutic for IBD.

Full text of DOI:10.1021/acs.jmedchem.6b00412

Subscriber access provided by UNIV OF WATERLOO
Article
An N,N-bis(benzimidazolylpicolinoyl)piperazine (BT-11): A Novel Lanthionine
Synthetase C-like 2-Based Therapeutic for Inflammatory Bowel Disease
Adria Carbo, Richard D. Gandour, Raquel Hontecillas,
Noah Philipson, Aykut Uren, and Josep Bassaganya-Riera
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00412 • Publication Date (Web): 04 Nov 2016
Downloaded from http://pubs.acs.org on November 8, 2016
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 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
An N,Nꢀbis(Benzimidazolylpicolinoyl)piperazine
(BTꢀ11): A Novel Lanthionine Synthetase Cꢀlike 2ꢀ
Based Therapeutic for Inflammatory Bowel Disease
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
Adria Carbo1, Richard D. Gandour1, Raquel Hontecillas1, Noah Philipson1, Aykut Uren2, and
Josep Bassaganya-Riera*1
1: Biotherapeutics Inc., 1800 Kraft Dr. Suite 200, Blacksburg, Virginia 24060
2: Georgetown University Medical Center, Washington, District of Columbia, 20057
Keywords: LANCL2, Crohn’s disease, ulcerative colitis, BTꢀ11, therapeutics, inflammatory
bowel disease
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 52
1
2
3
4
Abstract
5
6
7
8
9
Lanthionine synthetase Cꢀlike 2 (LANCL2), a novel therapeutic target for inflammatory and
autoimmune diseases, and diabetes, exerts antiꢀinflammatory and insulinꢀsensitizing effects. This
study reports the first LANCL2ꢀbased therapeutics for inflammatory bowel disease (IBD).
Analogs of 1 (ABA) and 2 (NSC61610) were screened by molecular docking, then synthesized
and analyzed for binding to LANCL2 by surface plasmon resonance. Piperazineꢀ1,4ꢀdiylbis(6ꢀ
benzo[d]imidazoleꢀ2ꢀyl)pyridineꢀ2ꢀyl)methanone, 7, was identified as the lead LANCL2ꢀbinding
compound for treating IBD. The oral treatment with 7 (8 mg/kg/d) in a mouse model of IBD
resulted in lowering the disease activity index, decreasing colonic inflammatory lesions by 4ꢀ
fold, and suppressing inflammatory markers (e.g., TNFꢀα, and interferonꢀγ) in the gut.
Furthermore, studies in LANCL2–/– mice demonstrated that loss of LANCL2 abrogated
beneficial actions of 7, suggesting high selectivity for the target. In conclusion, 7 merits
continued development as a LANCL2ꢀbased, firstꢀinꢀclass orally active therapeutic for IBD.
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
ACS Paragon Plus Environment

Page 3 of 52
Journal of Medicinal Chemistry
1
2
3
4
Introduction
5
6
7
8
Inflammatory bowel disease (IBD), a chronic recurring disease of the gastrointestinal (GI)
tract, afflicts over 1 million people in North America and 4 million worldwide.1 This widespread
and debilitating autoimmune disease results in decreased quality of life and significant health
careꢀrelated costs.2 IBD comprises two clinical manifestations: ulcerative colitis (UC) and
Crohn’s disease. Both pathologies are characterized by uncontrolled chronic inflammation in the
GI tract triggered by an unbalanced, proꢀinflammatory immune response to the commensal gut
microbiota. Current therapies for IBD, such as antiꢀTNFꢀα biologics, are modestly successful for
the longꢀterm management of the disease and have significant adverse side effects, including
infection, cancer, and death.
9
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
A recent review3 summarizes the current treatment paradigm for IBD—small molecules and
biologics. Aminosalicylates (e.g., mesalamine) inhibit interleukin (IL)ꢀ1, TNFꢀα, and platelet
activating factor. These locally acting immunosuppressors decrease antibody secretion by
nonspecifically inhibiting cytokines. Immunomodulators (e.g., methotrexate) block the de novo
pathway of purine synthesis. They show antiꢀproliferative effects and reduce inflammation.
Corticosteroids (e.g., prednisone), amongst other functions, block phospholipase A2 in the
arachidonic acid cascade and, thus, alter the balance between prostaglandins and leukotrienes.4
They stimulate apoptosis of lamina propria lymphocytes and suppress transcription of cytokines.
AntiꢀTNFꢀα drugs (e.g., infliximab) induce apoptosis in proinflammatory cells. They bind
specifically to TNFꢀα and block of the interaction of the receptor. They specifically inhibit
inflammatory cytokines, which result in immunosuppression. Antiꢀcell adhesion molecules (e.g.,
vedolizumab) inhibit migration of leukocytes into the GI mucosa. They specifically inhibit cell
adhesion molecules that mediate lymphocyte–endothelial interactions.5 Although current
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 52
1
2
3
4
5
6
7
8
9
treatments have improved, an unmet clinical need remains for safer and more effective IBD
therapeutics. For example, antiꢀTNFꢀα antibodies, used to treat patients with moderate to severe
Crohn’s disease, only work in 30–40% of the patients and have adverse effects that range from
infection to cancer and death.
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
Lanthionine synthetase Cꢀlike 2 (LANCL2) has emerged as a new therapeutic target for
treating inflammatory and immuneꢀmediated diseases, and diabetes.6 Its proposed natural ligand
is abscisic acid, 1 (ABA) (Chart 1),7 which exerts antiꢀinflammatory and antiꢀdiabetic effects.8ꢀ
12
LANCL2 is broadly expressed in epithelial cells and the immune system, including thymus,
spleen, colon, and Peyer’s patches; expression in these cells and tissues suggests a potential role
in modulating mucosal immune responses.
To further understand the function of LANCL2 through its structure and to investigate whether
1 activates LANCL2 via direct binding to the protein, molecular modeling of human LANCL2
was performed.13 Based on the predicted structure of LANCL2, an in silico docking approach
was used to screen ligands from public databases such as NCI Diversity Set II, ChemBridge,
ZINC natural products, and FDAꢀapproved drugs databases. These studies uncovered 2
(NSC61610, PubChem 247228) (Chart 1) as the topꢀranked ligand based on binding energy to
LANCL2. Biochemical in vitro and in vivo experiments demonstrated that treatment with 2
induces ILꢀ10ꢀmediated antiꢀinflammatory responses that ameliorate disease activity, and
decrease inflammatory lesions in the gut.14
Herein, we describe the process for identifying 7 (BTꢀ11)15 (Chart 1 and Scheme 1), as a lead
orally active compound for treating IBD. Our discovery process begins with designing analogs of
2 (Table S1) and 1 (Table S2) within limits of “freedom to operate” to develop new chemical
entities (NCEs). These designs are screened computationally by molecular docking to LANCL2.
ACS Paragon Plus Environment

Page 5 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
Compounds selected for synthesis are then assayed for binding to LANCL2 as measured by
surface plasmon resonance (SPR).
7
8
9
From these data, 7, our top lead, is assayed for antiꢀinflammatory efficacy in a dextran sodium
sulfate (DSS) mouse model of IBD. These in vivo results demonstrate that oral treatment with 7
(8 mg/kg/day) ameliorates colitis in mice.16, 17 Further, our initial safety assessment in rats
indicates that oral treatment with 7 at high doses has an excellent safety profile up to 1,000
mg/kg/day.18, 19 In addition to reporting the details of these preclinical efficacy studies,16, 17 we
report a selectivity study with knockout mice (LANCL2 –/–) to support our hypothesis that 7
protects from IBD by targeting LANCL2. These studies support a claim that 7 has potential to be
a safe and effective orally active therapeutic for IBD.
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
Results
Drug Design. The physicochemical properties of the two leads suggest that 1 fits the profile
(Table 1) for an oral drug20, while 2 might be too big and too hydrophobic. In analyses with
ChemAxon software (www.chemicalize.org) on drug likeness, 2 only passes one of the six
Lipinskiꢀlike filters; 1 passes all six. Lead 2 passed only the Veber filter (polar surface area ≤
140 Å2 and rotatable bonds ≤ 12). Yet, 2 effectively reduces gut inflammation in mouse
models,14 thereby validating its potential as a lead. The one caveat is that 2 is a very insoluble
compound. So, both leads must be considered as candidate parent compounds for developing
derivatives and analogs that are NCEs.
To maintain as much of the lead ligands as possible, the design of analogs of 2 and 1 began
with a conservative approach to retain any pharmacophores in the leads. We designed analogs
based on ease of synthesis and ability to probe possible separate binding sites on LANCL2 for
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 52
1
2
the two leads. Informed by the ranges for the characteristic physical properties of oral drugs20
and given the limited “freedom to operate” for constructing novel chemotypes, we designed
several sets of ligands to improve physical properties for both leads (Tables S1 and S2). The
designs for analogs of these two, very dissimilar leads provided a broad range of physical
properties.
3
4
5
6
7
8
9
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
For analogs (Table S1) of 2, we examined a few modifications—(i) changing paraꢀsubstituted
benzene to 1,4ꢀcyclohexyl to vary flexibility in the central scaffold, (ii) changing the two metaꢀ
substituted benzene rings to 2,6ꢀsubstituted pyridines to lower cLogP, and (iii) benzimidazole to
benzoxazole or 7ꢀazabenzimidazoles to decrease or increase the number of nitrogen atoms. For
the remainder of the analogs, we reversed the amide bond, a strategy known as an umpolung
approach. The central scaffold varied with several diamines, including piperazine,21, 22 which
changed secondary amides into tertiary amides, thereby removing two hydrogen bond donors.
Although the 1,4ꢀarrangement of the two nitrogens in piperazine provided the same relative
attachment as the paraꢀsubstituted benzene ring in 2, the number of bonds connecting the two
benzimidazoles were reduced, resulting in a slightly smaller molecule. Further, the increased
flexibility of the piperazine compared to a planar benzene could enhance interaction of these
analogs with LANCL2 compared to the rigid 2.
The isoprenoid phytohormone 1, discovered first in plants and then in mammals,23 provided an
even greater challenge for finding NCEs because a recent patent application24 covered a large
chemical space. With the thought that the keto and hydroxyl groups were the key parts of the
pharmacophore, simplifying the isophorone moiety to cyclohexanone offered opportunities for
analogs (Table S2). We added an (E)ꢀethenediyl spacer to link the cyclohexanone to sixꢀ
ACS Paragon Plus Environment

Page 7 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
membered and fiveꢀmembered aromatic rings. The positioning of the carboxyl group on these
aromatic rings was designed to probe the optimal geometry for binding.
7
8
9
To test our designs, we performed molecular docking, then selected compounds for synthesis,
assayed their binding to LANCL2 with SPR, and then selected one for in vivo testing in mice and
rats. The top lead was advanced towards a fullꢀscale preclinical efficacy and safety program.
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
Molecular Docking. To guide the selection of which compounds to synthesize for assays, the
interactions of the analogs with LANCL2 were analyzed by docking studies. The initial goal was
to find analogs with binding energies equal to or exceeding those for 2 and 1. Having identified
these analogs, we visually examined the fit with respect to sterics and chemical compatibility.
Among the analogs of 2, piperazine analogs were significantly better than the others. Thus, 7 and
8 emerged as targets for chemical synthesis (Scheme 1) and further assay. As the synthons and
chemistry were similar, we synthesized 9–14 and 19.
For analogs (Table S2) of 1, docking studies suggested that nearly all members of the sixꢀ
membered ring series especially 36, 38, and 41, showed better binding than 1. No analog was
especially better or worse at docking than 1. Analogs 24 and 28 were similar in binding to that of
1; 24, which has no nitrogen atoms, was chosen for comparison with 36, 38, and 41. Analogs 24,
28, 36, 38, and 41 were synthesized (Schemes 2 and 3) for further assay by SPR.
Chemistry
Analogs of 2 were synthesized by carbodiimide chemistry (Scheme 1). Synthesizing the
unsymmetrical 19 required making 17 and 18. The coupling of 18•HCl with 3 gave a higher
yield than the coupling of 17•HCl with 4. The yields for 7 and 8 were optimized and scaled to
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 52
1
2
3
4
gram quantities as these had better solubilities and activities than the others. All target
5
6
1
compounds were characterized by LCMS and H NMR (See supporting information.). Target
7
8
9
compounds were purified by chromatography to at least 95% as analyzed by LCMS.
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
Analogs of 1 (Schemes 2 and 3) were synthesized by organometallic chemistries. Heck and
Suzuki crossꢀcoupling reactions produced the analogs. Heck chemistry worked well for coupling
to the aryl iodide (21) and furanyl bromide (25); it gave poor yields for coupling to
bromopicolinates and bromoisonicotinate. Suzuki chemistry worked well for these pyridines.
Surprisingly coupling of 33 to 4ꢀbromopicolinate occurred with demethylation of the ester.
Notably, 4ꢀbromopicolinic acid was identified as a side product (See supporting information.). In
the course of optimizing the synthesis of 28, 30 formed as a major side product. Aware that
serendipity plays a role in drug discovery and after performing molecular docking, we developed
a pathway to 30. Interestingly, reversing the deꢀprotection reactions favored one product over the
1
other. All target compounds were characterized by LCMS and H NMR (See Supporting
Information.). Target compounds were purified by chromatography to at least 95% as analyzed
by LCMS.
Biochemistry and Biology
Below, we present four experiments—two in vitro assessments and two in vivo assessments—
that identify 7 as the lead and elucidate its role in targeting LANCL2 for treating IBD.
Experiment #1 is assaying several analogs for binding to LANCL2 by SPR. Experiment #2 is
measuring cAMP in WT and LANCL2–/– splenocytes following treatment with 7. Experiments
#3 and #4 are assessing efficacy of 7 in DSS mouse models of IBD. Experiment #3 is with WT
only; Experiment #4 is with WT compared to LANCL2–/–. These two experiments assess
ACS Paragon Plus Environment

Page 9 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
selectivity and mechanism of action (MoA) in vivo. Our goal is to demonstrate that 7 has antiꢀ
inflammatory effects by binding to LANCL2 by examining four key events: i) direct binding to
the LANCL2 protein, ii) target engagement/activation after binding of the LANCL2 pathway, iii)
high antiꢀinflammatory efficacy in animal pharmacology studies with mouse models of colonic
inflammation, and iv) abrogation of the protective actions of 7 in LANCL2–/– mice with IBD.
We illustrate the LANCL2 pathway in a cartoon representation (Figure 1A).
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
Experiment #1: Assaying several analogs for direct binding to LANCL2 by SPR.
To assess direct binding to the LANCL2 protein, we performed SPR studies with LANCL2. All
synthesized analogs were assayed. Binding sensorgrams for 7 (Figure 1B) showed a typical
small molecule–protein interaction with very fast on rates and very fast off rates. By plotting the
equilibrium binding level against the compound concentration, we measured the equilibrium
dissociation constant (KD) for each analog that bound (e.g., Figure 1C).
For analogs of 2, the benzoxazole analogs (8, 10, 12, 14) were insoluble in the assay
conditions. Analogs 7 and 19 displayed the best binding. We also tested 17, 18, and 3 in the
synthesis as these are potential hydrolysis products of the two top candidates. None of these
latter three compounds showed any binding. For analogs of 1, all six compounds showed modest
binding with 30 and 36 as the two top candidates. These results agreed with the docking studies
where analogs of 2 had higher binding energies than analogs of 1.
Out of the pool of analogs, we identified 7 as an ideal candidate to move forward in the
preclinical and clinical regulatory pipeline based on predicted binding, chemical structure,
patentability, and ease of synthesis. Further, 2 had activity14 with LANCL2; thus, an analog of 2
might be successful.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 52
1
2
3
4
5
6
7
8
Experiment #2: measuring cAMP in WT and LANCL2–/– splenocytes following
treatment with 7.
9
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
Results from previous studies14 showed that LANCL2 engagement produces an increase of
PKA, followed by an accumulation of cAMP in the cytoplasm (Figure 2). From a mechanistic
perspective, activating cAMP leads to suppressing TNFꢀα and increasing ILꢀ10 via cAMP
response elementꢀbinding (CREB) phopshorylation.25 Several endogenous mediators have been
described26, 27 as TNFꢀα attenuators, such as cAMP. To elucidate the specific steps in target
engagement triggered by binding of 7 to LANCL2, we performed in vitro studies with mouse
splenocytes and assess the levels of downstream signaling.
Results (Figure 2) showed a dose–response increase of cAMP production when WTꢀderived
splenocytes are treated with 7. With splenocytes extracted from LANCL2–/– mice, cAMP
production did not correlate with dose of 7. This in vitro experiment demonstrates 7 stimulates
cAMP production by activating the LANCL2 pathway.
Experiment #3: Assessment of efficacy of 7 in a WT DSS colitis mouse model.
Once we confirmed the direct binding by SPR and the target engagement of 7 to LANCL2 in
splenocytes, we assayed the efficacy of 7 in vivo using a mouse model of DSS colitis. Notably,
we reported14 that administering 2, the parent compound of 7, significantly improves disease
activity and gut inflammatory lesions. Our objective in this study was to investigate whether
administering 7 also activates LANCL2 and protects mice from IBD.
Daily oral treatment with 7 (8 mg/kg/d) significantly improved disease activity during a 7ꢀday
DSS challenge (Figure 3A). We next examined the effect of 7 on colonic inflammatory lesions.
ACS Paragon Plus Environment

Page 11 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In line with our observations of disease activity and gross lesions, histopathological analyses
confirmed that oral treatment with 7 significantly decreased by fivefold the inflammation in the
gut mucosa based on assessment of leukocytic infiltration (Figure 3G), epithelial erosion
(Figure 3H), and mucosal thickening (Figure 3I). Representative colonic photomicrographs
(Figure 3AꢀF) demonstrate that oral treatment with 7 in mice with DSS colitis significantly
ameliorates colonic lesion formation.
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
Furthermore, oral treatment with 7 showed a significant upregulation in colonic LANCL2
expression on day 3 (Figure 4A). The restitution of LANCL2 to homeostatic levels by 7 is
critical because LANCL2 mRNA is decreased (Figure 4B) during the peak of
inflammation/disease in a DSS colitis mouse model. Furthermore, oral treatment with 7 downꢀ
regulated the colonic expression of TNFꢀα (Figure 4C), and upregulated the levels of the antiꢀ
inflammatory cytokine ILꢀ10 (Figure 4D).
Experiment #4: Target selectivity Studies: Assessment of efficacy of 7 in WT and
LANCL2–/– mice with IBD.
To investigate the dependency of 7 for LANCL2, we performed preclinical studies using WT
mice and mice lacking LANCL2 with IBD. Our results provided molecular evidence in vivo that
LANCL2 is necessary for 7 to exert antiꢀinflammatory benefits, because absence of LANCL2
abrogated the therapeutic efficacy of 7 in DSS colitis (Figure 5A). During peak of inflammation
in DSS colitis (7 days after start of the challenge), the effect of 7 in lesion formation at the
colonic mucosa also depended on LANCL2. We assessed colonic histopathology in LANCL2–/–
mice treated with either vehicle or 7 and observed that the absence of LANCL2 completely
abrogates the protective effect of 7 in experimental IBD (Figure 5B).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 52
1
2
3
4
5
6
7
8
9
Our flow cytometry analyses of colonic cells from the two DSS colitis mouse models also
demonstrated that reducing TNFꢀα (Figure 5C), MHCꢀII+ CD11c+ granulocytes (Figure 5E),
and MCPꢀ1 (Figure 5F) depends on LANCL2, because the loss of lancl2 abrogates the effect of
7. In line with these results, upregulation of ILꢀ10 secretion after treatment with 7 was
completely abrogated in colons of LANCL2–/– mice (Figure 5D) but not in WT littermates.
These immunophenotyping results at the colonic mucosa level, together with disease activity and
histopathological results, demonstrated that 7 requires LANCL2 to decrease inflammation in
experimental IBD.
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
Discussion and Conclusion
This study achieved our goal of finding a novel orally active firstꢀinꢀclass therapeutic for IBD
(Crohn’s disease and UC) based on a unique MoA involving the LANCL2 pathway.
Computational and biochemical screening of libraries of analogs of 1 and 2 enabled identifying
our top lead, 7, which binds to LANCL2, downꢀregulates expression of proꢀinflammatory
cytokines (e.g., TNFꢀα or interferonꢀγ, which are hallmarks of IBD), and promotes ILꢀ10ꢀ
mediated antiꢀinflammatory responses in the GI tract. Guided by the average physical properties
of marketed oral drugs, in silico analyses led to the synthesizing 15 compounds for analysis by
SPR to assess binding to LANCL2. With the lowest KD, 7 emerged as the most likely candidate.
As an analog of 2, 7 is the best choice of the analogs analyzed herein for an IBD drug
development program. As stated earlier, 2 only passes one of the six Lipinskiꢀlike filters; 7
passes three. Comparing some key properties to those of 2, 7 has a significantly lower cLogP
(3.93 to 6.78), fewer rotatable bonds (4 to 6), lower MW (528 to 548), greater polar surface area
(123 to 116), and smaller total surface area (700 to 719). These changes improve the drugꢀlike
ACS Paragon Plus Environment

Page 13 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
properties compared to 2. More importantly, 7 has better solubility than 2, especially in acidic
media as found in the gut. Furthermore, 7 is stable in simulated gastric and intestinal fluids (data
not published).
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
As stated above, our goal is to develop an orally active IBD drug that acts in the GI tract
(ileum and colon). A drug with minimal systemic absorption (10% or less) provides many
advantages for a locally acting therapeutic. Not satisfying all rules for a good oral drug with
systemic distribution may be a good strategy for developing a drug for IBD. Here, we developed
a drug with physicochemical properties that might enable local action in the GI tract.
Our preclinical therapeutic efficacy studies in mice show that 7 reduces the disease activity and
improves gut inflammation by significantly decreasing leukocytic infiltration in the gut mucosa,
plus decreasing mucosal thickening and epithelial erosion. A previous study11 documents that
oral treatment with 1 produces antiꢀinflammatory effects in mice with IBD. Specifically, in a
DSS colitis mouse model, 1 significantly ameliorates disease activity, colitis, reduced colonic
leukocyte infiltration, and inflammation.11 These improvements are associated with
downregulation in vascular cell adhesion markerꢀ1, Eꢀselectin, and mucosal addressin adhesion
markerꢀ1 expression.11
In our study, gene expression analyses confirm that oral administration of 7 upregulates the
expression of ILꢀ10, and downregulates the expression of TNFꢀα mRNA in a DSS colitis mouse
model. Also, oral treatment with 7 upregulates LANCL2 expression in the GI tract. LANCL2 is
necessary to exert all the antiꢀinflammatory effects due to administration of 7, because the
absence of LANCL2 abrogates the beneficial effect of 7, as demonstrated by Experiment 4. From
a mechanistic perspective, small molecules (e.g., 1 and 2) that activate LANCL2 regulate
inflammatory and metabolic processes through adenylyl cyclase/cAMP dependent
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 52
1
2
mechanisms.23 LANCL2 works through cAMP, which ultimately regulates the expression of
3
4
5
6
cytokines. Notably, various cAMPꢀelevating agents, such as rolipram and cicaprost, precisely
7
8
9
29
inhibit TNFꢀα synthesis.28,
Also, accumulation of cAMP stimulates ILꢀ10 synthesis in
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
humans.30 The mediating signal transduction pathway engages PKA as suggested by the reversal
of effects with a cAMP inhibitor.30
Targeting activation of the LANCL2 pathway might open a new strategy to treat patients that
may not have responded well to other therapies. Activating LANCL2 with 7 positively regulates
the production of cAMP in the cytoplasm. In addition, LANCL2 is expressed in RAW 264.7
cells, mouse leukemic monocyte macrophage cell line. Further, lancl2 knockdown studies
provide evidence that ABAꢀmediated signaling depends on lancl2 expression.23 This leads to the
downstream activation of PKA, which serves to phosphorylate and activate many proteins such
as CREB, a cellular transcription factor. By binding to cAMP response element DNA sequences,
it increases or decreases transcription of downstream genes.25 CREBꢀregulated genes include: cꢀ
fos, the neurotrophin brainꢀderived neurotrophic factor, tyrosine hydroxylase, many
neuropeptides, and the antiꢀinflammatory cytokine, ILꢀ10.
Analog 7 was tested in safety studies18 in rats up to 1,000 mg/kg/d. Behavioral, clinical
pathology, and histopathology studies18, 19 in a highꢀsingleꢀdose and a 14ꢀday highꢀmultipleꢀdose
showed no toxicologically relevant differences between vehicle and 7ꢀtreated rats. Preliminary
data from a PK study (0.5–48 h) conducted in healthy rats given an oral dose (80 mg/kg) of 7
showed that the maximal concentration (Cmax) in blood plasma (28 ng/mL) was reached at 1 h.
The product was cleared with a t1/2 of 3.1 h with a rate constant of elimination of –0.225 L/h
from an apparent volume of distribution estimated at 3.38 L/kg. A second trial with 7 at 500
mg/kg resulted in a Cmax of 17 ng/mL and a time to Cmax of 4 h. That 7 did not demonstrate
ACS Paragon Plus Environment

Page 15 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
plasmaꢀconcentration proportionality following a >6ꢀfold difference in doses suggested that 7
has limited absorption from the GI tract. One would expect greater doseꢀrelated differences in
plasma concentrations if 7 had a high bioavailability. Also, concentrations of 7 in colon and
colon contents were much higher than those in plasma and studies with oral versus rectal
administration showed no differences (Figure S1).
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
Combining the toxicological findings at doses up to 1,000 mg/kg/d, the low bioavailability,
and lack of dose proportionality with the results from Experiments #3 and #4 (8 mg/kg/d)
strongly suggest that 7 may protect from IBD by acting locally in the GI mucosa as opposed to
being systemically distributed and taken up by tissues. These two experiments further validate
our claim that LANCL2 is a unique therapeutic target for inflammatory diseases.
An unmet clinical need remains for novel, safer, and more efficacious drugs to treat both
Crohn’s disease and UC. We have presented a unique therapeutic target and a novel drug
candidate. The promising results for 7 in animal pharmacology in mice and safety studies in rats
support continued nonclinical and clinical development toward a firstꢀinꢀclass, safe, efficacious,
orally active therapeutic for IBD.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 52
1
2
3
4
Experimental Methods
5
6
7
Chemistry
8
9
Piperazineꢀ1,4ꢀdiylbis((6ꢀ(1Hꢀbenzo[d]imidazolꢀ2ꢀyl)pyridinꢀ2ꢀyl)methanone), (7 and
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
7•HCl).
A solution of 6ꢀ(1Hꢀbenzimidazolꢀ2ꢀyl)pyridineꢀ2ꢀcarboxylic acid31 (3, 12 g) in DMF (100 mL)
was cooled to 0 °C, and then sequentially added EDCI (1.5 equiv), HOBt (1.5 equiv) and DIPEA
(1.2 equiv). The mixture was stirred for 10 min at 0 °C. Piperazine (0.5 equiv) was added; the
reaction mixture was stirred and gradually warmed to RT over 16 h. The reaction mixture was
then poured onto ice (~300 mL). The precipitated solid was filtered, washed with iceꢀcold water
and dried to give a pale brown solid (10 g), mp 326.2–328.0 °C: 1H NMR (400 MHz, DMSOꢀ
d6), δ 12.97 (s, 1H), 12.79 (s, 1H), 8.42 (d, 1H, J = 8.0), 8.34 (d, 1H, J = 8.0), 8.16 (t, 1H, J =
8.0), 8.09 (t, 1H, J = 8.0), 7.77–7.70 (m, 2H), 7.67 (d, 2H, J = 8.0), 7.57 (d, 1H, J = 7.2), 7.53 (d,
1H, J = 6.8), 7.33–7.16 (m, 4H), 3.90 (br, 2H), 3.80 (br, 2H), 3.65 (br, 2H), 3.56 (br, 2H); 99.9 %
pure by LCMSꢀES 529.44 [M+H]+, 265.46 [(M+2H)/2]++.
A suspension of 7 (1.0 equiv) in a minimal amount of MeOH was cooled to 0 °C, and 4M
methanolic HCl (excess, 15 mL/g) was added dropwise over a period of 15–20 min. The mixture
was stirred and gradually warmed to RT over 3 h. Concentration of the reaction mixture gave a
viscous mass, which was washed with MeOH/DCM 1:9 and then lyophilized to get an offꢀwhite
1
solid (850 mg), mp 338.8–340.1 °C: H NMR (300 MHz, DMSOꢀd6), δ 8.63 (d, 1H, J = 7.8), δ
8.54 (d, 1H, J = 7.8), 8.27 (t, 1H, J = 7.8), 8.25 (t, 1H, J = 7.8), 7.92–7.74 (m, 6H), 7.56–7.41
(m, 4H), 3.91 (br, 2H), 3.81 (br, 2H), 3.64 (br, 2H), 3.55 (br, 2H); 95.2 % pure by LCMSꢀES
529.56 [M+H]+.
ACS Paragon Plus Environment

Page 17 of 52
Journal of Medicinal Chemistry
1
2
3
4
Piperazineꢀ1,4ꢀdiylbis((6ꢀ(benzo[d]oxazolꢀ2ꢀyl)pyridinꢀ2ꢀyl)methanone), (8).
5
6
7
8
A solution of 6ꢀ(benzoxazolꢀ2ꢀyl)pyridineꢀ2ꢀcarboxylic acid32 (4, 4.05 g) in DMF/DCM 1:9 was
treated with EDCI (1.5 equiv), HOBt (1.5 equiv), DIPEA (1.2 equiv), and piperazine (0.5 equiv)
at 0 °C. The mixture was stirred and gradually warmed to RT over 16 h. A light brown solid
formed and was filtered in a sinterꢀglass funnel. Washing with water and lyophilizing gave a
9
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
light brown solid (3.2 g), mp 342.0–343.3 °C: H NMR (300 MHz, CDCl3), δ 8.47 (d, 1H, J =
8.1), 8.40 (d, 1H, J = 8.1), 8.15–7.98 (m, 2H), 7.94–7.83 (m, 3H), 7.79 (d, 1H, J = 7.8), 7.71 (d,
1H, J = 6.9), 7.55 (d, 1H), 7.52–7.29 (m, 4H), 4.09–3.91 (m, 8H); 99.3% pure by HPLC and
95.4% pure by qHNMR.
Piperazineꢀ1,4ꢀdiylbis((3ꢀ(1Hꢀbenzo[d]imidazolꢀ2ꢀyl)phenyl)methanone), (9 and 9•HCl).
A solution of 3ꢀ(1Hꢀbenzimidazolꢀ2ꢀyl)benzoic acid33 (5, 100 mg) in DMF (6 mL) was treated
with EDCI (1.5 equiv), HOBt (1.5 equiv), DIPEA (1 equiv), and piperazine (0.5 equiv) at 0 °C.
The mixture was stirred and gradually warmed to RT over 16 h. After workup and washings with
1
Et2O, a light brown solid (30 mg) was isolated: H NMR (400 MHz, DMSOꢀd6), δ 13.01 (s, 2H),
8.23 (br, 4H), 7.66 (br, 4H), 7.55 (br, 4H), 7.22 (br, 4H), 3.91–3.42 (br, 8H); 95.5% pure by
LCMSꢀES 527.36 [M+H]+, 264.50 [(M+2H)/2]++.
To a solution of 9 (30 mg) in dioxane, 4M HCl was added at RT. After stirring for 3 h, the
mixture was concentrated to give a solid. Washing with Et2O gave a light brown solid (10 mg),
1
mp 348.8–349.9 °C: H NMR (400 MHz, DMSOꢀd6), δ 8.48 (br, 4H), 7.91–7.74 (br, 8H), 7.56
(br, 4H), 3.81 (br, 4H) , 3.57 (br, 4H); 98.3% pure by LCMSꢀES 527.44 [M+H]+, 264.50
[(M+2H)/2]++.
Piperazineꢀ1,4ꢀdiylbis((3ꢀ(benzo[d]oxazolꢀ2ꢀyl)phenyl)methanone), (10).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 52
1
2
In DMF (10 mL), 3ꢀ(2ꢀbenzoxazolyl)benzoic acid34 (6, 50 mg) was treated with EDCI (1.25
equiv), HOBt (1.25 equiv), DIPEA (1 equiv), and piperazine (0.5 equiv) at 0 °C. The mixture
was stirred and gradually warmed to RT over 16 h. After diluting the reaction mixture with ice
cold water, a precipitate formed. Filtration, followed by drying, gave a light brown solid (30
3
4
5
6
7
8
9
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
mg): H NMR (300 MHz, TFA), δ 8.72–8.54 (br, 4H), 8.18–7.80 (12H), 4.37–4.05 (4H), 4.04–
3.66 (4H); 82.8% pure by LCMSꢀES 529.48 [M+H]+.
N,N'ꢀ(1,4ꢀPhenylene)bis(6ꢀ(1Hꢀbenzo[d]imidazolꢀ2ꢀyl)picolinamide), (11 and 11•HCl).
In DMF (10 mL), 3 (100 mg) was treated with EDCI (1.5 equiv), HOBt (1.5 equiv), DIPEA (3
equiv), and benzeneꢀ1,4ꢀdiamine (0.5 equiv) at 0 °C. The mixture was stirred and gradually
warmed to RT over 16 h. After pouring the reaction mixture into iceꢀcold water, the precipitate
was filtered and dried to give a pale brown solid (60 mg).
In dioxane (3 mL) at 0 °C, the solid (50 mg) was treated with 4M HCl. The mixture was stirred
and gradually warmed to RT over 4 h. Evaporation of the excess dioxane HCl gave a brown solid
1
(30 mg), mp 394.6–398.9 °C: H NMR (300 MHz, DMSOꢀd6), δ 10.96 (s, 2H), 8.61 (t, 2H, J =
4.5), 8.35 (d, 4H, J = 4.5), 8.06 (s, 4H), 7.88–7.81 (m, 4H), 7.47–7.40 (m, 4H); 97.5% pure by
LCMSꢀES 551.84 [M+H]+.
N,N'ꢀ(1,4ꢀPhenylene)bis(6ꢀ(benzo[d]oxazolꢀ2ꢀyl)picolinamide), (12).
In DMF (10 mL), 4 (100 mg) was treated with EDCI (1.5 equiv), HOBt (1.5 equiv), DIPEA (1.2
equiv), and benzeneꢀ1,4ꢀdiamine (0.5 equiv) at 0 °C. The mixture was stirred and gradually
warmed to RT over 16 h. Dilution of the reaction mixture with ice cold water gave a precipitate.
1
Filtering, followed by drying, gave a light brown solid (70 mg): H NMR (400 MHz, TFA), δ
ACS Paragon Plus Environment

Page 19 of 52
Journal of Medicinal Chemistry
1
2
3
4
8.88 (d, 2H, J = 7.6), 8.82 (d, 2H, J = 7.2), 8.57 (t, 2H, J = 8.0), 8.11 (br, 4H), 8.02–7.86 (m,
5
6
8H). 99.5% pure by LCMSꢀES 553.28 [M+H]+.
7
8
9
N,N'ꢀ(1,4ꢀPhenylene)bis(3ꢀ(1Hꢀbenzo[d]imidazolꢀ2ꢀyl)benzamide), (13 and 13•HCl).
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
In DMF (6 mL), 5 (100 mg) was treated with EDCI (1.5 equiv), HOBt (1.5 equiv), DIPEA (1
equiv), and benzeneꢀ1,4ꢀdiamine (0.5 equiv) at 0 °C. The mixture was stirred and gradually
warmed to RT over 16 h. TLC (10% methanol: DCM) shows formation of nonꢀpolar spot and
absence of starting material. Workup and washing with Et2O gave a light brown solid (60 mg):
1H NMR (300 MHz, DMSOꢀd6), δ 13.10 (s, 2H), 10.49 (s, 2H), 8.77 (s, 2H), 8.38 (d, 2H, J =
7.8), 8.07 (d, 2H, J = 7.8), 7.83 (s, 4H), 7.73 (t, 4H, J = 7.8), 7.57 (d, 2H, J = 6.9) 7.30–7.18 (m,
4H); 98.3% pure by LCMSꢀES 549.0 [M+H]+ 275.1 [(M+2H)/2]++.
In dioxane, 13 (60 mg) was treated with 4M HCl for 3 h. Concentrating the solution formed a
1
solid, which was washed with Et2O to give a light brown solid (50 mg): H NMR (300 MHz,
DMSOꢀd6), δ 10.60 (s, 2H), 8.99 (s, 2H), 8.53 (d, 2H, J = 7.5), 8.31 (d, 2H, J = 7.8), 7.94–7.81
m, 10H), 7.58–7.51 (m, 4H); 96.2% pure by LCMSꢀES 549.3 [M+H]+ 275.3 [(M+2H)/2]++.
N,N'ꢀ(1,4ꢀPhenylene)bis(3ꢀ(benzo[d]oxazolꢀ2ꢀyl)benzamide) (14).
In DMF (10 mL), 6 (50 mg) was treated with EDCI (1.25 equiv), HOBt (1.25 equiv), DIPEA (1
equiv), and benzeneꢀ1,4ꢀdiamine (0.5 equiv) at 0 °C. The mixture was stirred and gradually
warmed to RT over 16 h. Diluting the reaction mixture with ice cold water formed a precipitate,
1
which was filtered and dried to give a light brown solid (30 mg): H NMR (300 MHz, TFA), δ
9.18 (br, 2H), 8.71 (d, 2H, J = 7.2), 8.55 (br, 2H), 8.10–7.97 (br, 6H), 7.97–7.76 (br, 8H).
Insolubility prevented accurate analysis by LCMSꢀES.
tertꢀButyl 4ꢀ(6ꢀ(1Hꢀbenzo[d]imidazolꢀ2ꢀyl)picolinoyl)piperazineꢀ1ꢀcarboxylate, (15).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 52
1
2
3
4
5
6
7
8
9
In DMF (10 mL), 3 (500 mg) was treated with EDCI (1.5 equiv), HOBt (1.5 equiv), DIPEA (3
equiv), and tertꢀbutyl piperazineꢀ1ꢀcarboxylate (1.1 equiv) at 0 °C. The mixture was stirred and
gradually warmed to RT over 16 h. After pouring the reaction mixture into iceꢀcold water, the
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
precipitate was filtered and dried to give a pale brown solid (600 mg): H NMR (300 MHz,
DMSOꢀd6), δ 12.85 (s, 1H), 8.39 (d, 1H, J = 7.5), 8.12 (t, 1H, J = 7.5), 7.73(d, 1H, J = 7.2), 7.66
(d, 1H, J = 7.8), 7.58 (d, 1H, J = 7.2), 7.32–7.19 (m, 2H), 3.70 (br, 2H), 3.49 (br, 2H), 3.41 (br,
4H), 1.42 (s, 9H); 99.3% pure by LCMSꢀES 408.54 [M+H]+.
tertꢀButyl 4ꢀ(6ꢀ(benzo[d]oxazolꢀ2ꢀyl)picolinoyl)piperazineꢀ1ꢀcarboxylate, (16).
A solution of 4 (500 mg) in DMF (10 mL) was treated with EDCI (1.5 equiv), HOBt (1.5 equiv),
DIPEA (3 equiv), and tertꢀbutyl piperazineꢀ1ꢀcarboxylate (1.1 equiv) at 0 °C. The mixture was
stirred and gradually warmed to RT over 16 h. After concentration of the mixture, the residue
was extracted into EtOAc and washed with water. The organic layer was concentrated to give a
1
solid, which was washed with pentane to yield a light brown solid (120 mg): H NMR (300
MHz, DMSOꢀd6), δ 8.42 (d, 1H, J = 7.2), 8.20 (t, 1H, J = 7.8), 7.93–7.85 (m, 2H), 7.84–7.79 (m,
1H), 7.56–7.44 (m, 2H), 3.70 (br, 2H), 3.48 (br, 4H), 3.38 (br, 2H), 1.42 (s, 9H); 99.4% pure by
LCMSꢀES 431.37 [M+Na]+, 447.36 [M+K]+.
4ꢀ(6ꢀ(1HꢀBenzo[d]imidazolꢀ2ꢀyl)picolinoyl)piperazinꢀ1ꢀium chloride, (17•HCl).
Compound 15 (600 mg) was treated with methanolic HCl (6 mL) at 0 °C. The mixture was
stirred and gradually warmed to RT over 3 h. Evaporation of the excess methanolic HCl gave a
light brown solid (500 mg): 1H NMR (300 MHz, DMSOꢀd6), δ 9.61 (br, 2H), 8.67 (br, 1H), 8.31
(t, 1H J = 7.8), 7.84 (br, 3H), 7.50 (br, 2H), 3.97 (br, 2H), 3.76 (br, 2H), 3.35–3.15 (br, 4H);
95.0% pure by LCMSꢀES 308.26 [M+H]+.
ACS Paragon Plus Environment

Page 21 of 52
Journal of Medicinal Chemistry
1
2
3
4
(6ꢀ(1HꢀBenzo[d]imidazolꢀ2ꢀyl)pyridinꢀ2ꢀyl)(piperazinꢀ1ꢀyl)methanone, (17).
5
6
7
Salt 17•HCl (12 mg) was neutralized with satd aq NaHCO3 followed by lyophilization to give a
8
9
1
pale brown solid (9 mg): H NMR (300 MHz, DMSOꢀd6), δ 12.93 (br, 1H), 8.4 (d, 1H, J = 7.2),
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
8.10 (t, 1H, J = 7.8), 7.72 (d, 1H, J = 7.5), 7.59 (t, 2H, J = 7.8), 7.31–7.19 (m, 2H), 3.63 (br, 2H),
3.32 (br, 2H, overlap with H2O signal), 2.81 (br, 2H), 2.68 (br, 2H); 98.0% pure by LCMSꢀES
308.36 [M+H]+.
4ꢀ(6ꢀ(Benzo[d]oxazolꢀ2ꢀyl)picolinoyl)piperazinꢀ1ꢀium chloride, (18•HCl).
Compound 16 (200 mg) was treated with methanolic HCl (6 mL) at 0 °C. The mixture was
stirred and gradually warmed to RT over 3 h. Evaporating the solvent, then washing with
1
pentane and Et2O gave of a light brown solid (160 mg): H NMR (300 MHz, DMSOꢀd6), δ 9.30
(br, 2H), 8.45 (d, 1H, J = 8.1), 8.24 (t, 1H, J = 7.8), 7.94–7.84 (m, 3H), 7.57–7.45 (m, 2H), 3.95
(br, 2H), 3.79 (br, 2H), 3.26 (br, 2H), 3.18 (br, 2H); 98.1% pure by LCMSꢀES 309.26 [M+H]+.
(6ꢀ(Benzo[d]oxazolꢀ2ꢀyl)pyridinꢀ2ꢀyl)(piperazinꢀ1ꢀyl)methanone, (18).
Salt 18•HCl (25 mg) was neutralized with satd aq NaHCO3 followed by lyophilization to give a
light brown solid (20 mg): 1H NMR (300 MHz, DMSOꢀd6), δ 8.40 (d, 1H, J = 7.8), 8.18 (t, 1H, J
= 7.8), 7.89 (t, 2H, J = 7.2), 7.78 (d, 1H, J = 7.5), 7.55–7.44 (m, 2H), 3.65 (br, 2H), 3.40 (br, 2H
overlap with H2O signal), 2.84 (br, 2H), 2.72 (br, 2H); 98.7% pure by LCMSꢀES 309.37
[M+H]+.
(4ꢀ(6ꢀ(1HꢀBenzo[d]imidazolꢀ2ꢀyl)picolinoyl)piperazinꢀ1ꢀyl)(6ꢀ(benzo[d]oxazolꢀ2ꢀyl)pyridinꢀ
2ꢀyl)methanone, (19).
Acid 3 (4.0 g) in DMF (5 mL) was treated with EDCI (1.5 equiv), HOBt (1.5 equiv), DIPEA (3
equiv), and 18•HCl (5.1 g) at 0 ºC. The mixture was stirred and gradually warmed to RT over 16
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 52
1
2
3
4
5
6
h. Dilution of the reaction mixture with ice cold water gave a pale brown solid. Filtration, drying,
and washing with Et2O and pentane gave a beige solid (5.0 g). Combiꢀflash chromatography
7
8
9
1
(MeOH/DCM) on 4.8 g yielded offꢀwhite solid (3.6 g), mp 322.6–326.2 °C: H NMR (400
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
MHz, DMSOꢀd6), δ 12.94 (s, 0.5H), 12.92 (s, 0.5H), 8.48–8.33 (m, 2H), 8.28–8.06 (m, 2H),
7.95–7.69 (m, 5H), 7.54–7.46 (m, 2H), 7.42–7.18 (m, 3H), 3.90 (s, 2H), 3.82 (br, 2H), 3.76–3.52
(br, 4H); 99.9% pure by LCMSꢀES 530.40 [M+H]+.
Ethyl (E)ꢀ3ꢀ(2ꢀ(8ꢀhydroxyꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀyl)vinyl)benzoate (22).
A solution of 8ꢀvinylꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀol35 (20, 500 mg, 1 equiv), ethyl 3ꢀ
iodobenzoate (21, 0.8 equiv), and PPh3 (0.02 equiv) in Et3N (8 mL) was degassed with argon for
10 min. Then, Pd(OAc)2 (0.02 equiv) was added, and the mixture was again degassed for 10 min.
The resulting reaction mixture was heated at 95 °C for 16 h. After workup, a pale brown solid
1
(500 mg) was isolated by column chromatography (EtOAc/hexane 3:7); H NMR (300 MHz,
DMSOꢀd6), δ 7.95 (s 1H), 7.80 (d 1H, J = 7.2), 7.71 (d 1H, J = 8.4), 7.47 (t 1H, J = 7.8), 6.65 (d
1H, J = 16.2), 6.49 (d, 1H, J = 15.9), 4.65 (br 1H), 4.32 (q, 2H, J = 6.9), 3.87 (s, 4H), 1.99–1.65
(m, 4H), 1.64–1.48 (m, 4H), 1.33 (t 3H, J = 7.2); 95.1% pure by LCMSꢀES 315.38 [M–17]+.
Eꢀ3ꢀ(2ꢀ(8ꢀHydroxyꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀyl)vinyl)benzoic acid, (23).
A solution of 22 (500 mg) in THF/H2O/EtOH (4:2:1, 17.5 mL) was cooled to 0 °C; LiOH (2.5
equiv) was added. The mixture was stirred while rising to RT over 16 h. Concentration of the
mixture gave a yellow solid, which was dissolved in the minimum amount of water and then
acidified with 1N HCl to pH 3–4. Purification by column chromatography (EtOAc/hexane 7:3)
gave a pale yellow solid (220 mg): 1H NMR (300 MHz, DMSOꢀd6), δ 12.95 (br 1H), 7.95 (s 1H),
7.78 (d 1H, J = 7.5), 7.67 (d 1H, J = 7.5), 7.44 (t 1H, J = 8.1), 6.64 (d 1H, J = 15.9), 6.48 (d, 1H,
ACS Paragon Plus Environment

Page 23 of 52
Journal of Medicinal Chemistry
1
2
3
4
J = 16.2), 4.65 (s 1H), 3.86 (s, 4H), 1.88–1.60 (m, 4H), 1.56–1.49 (m, 4H); 94.5% pure by
5
6
LCMSꢀES 287.34 [M–17]+.
7
8
9
Eꢀ3ꢀ(2ꢀ(1ꢀHydroxyꢀ4ꢀoxocyclohexyl)vinyl)benzoic acid, (24).
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
In THF at 0 °C with stirring, 2N HCl (1.5 mL) was added to 23 (100 mg). The mixture was
stirred and gradually warmed to RT over 6 h. The solution was then concentrated, diluted with
water, extracted with EtOAc, and reꢀconcentrated to give a pale yellow solid (20 mg), mp 148.5–
1
150.2 °C: H NMR (300 MHz, DMSOꢀd6), δ 12.99 (bs 1H), 7.98 (s, 1H), 7.80 (d 1H, J = 7.5),
7.68 (d 1H, J = 7.8), 7.46 (t 1H, J = 7.5), 6.74 (d 1H, J = 16.2), 6.56 (d, 1H, J = 16.2), 6.10 (t,
1H), 5.13 (s 1H), 2.74–2.59 (m, 2H), 2.16 (br, 1H), 2.11 (br, 1H), 2.03–1.84 (m, 4H); 99.4% pure
by LCMSꢀES 259.37 [M–H]– 519.48 [2M–H]–.
Methyl (E)ꢀ5ꢀ(2ꢀ(8ꢀhydroxyꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀyl)vinyl)furanꢀ2ꢀcarboxylate, (26).
A solution of 20 (200 mg, 1.0 equiv) and methyl 5ꢀbromofuranꢀ2ꢀcarboxylate (25, 1.5 equiv) in
Et3N (2 mL) was degassed with argon for 10 min. Then, Pd(OAc)2 (0.025 equiv) and DPPF (0.05
equiv) were added, and the solution was degassed for 10 min. The resulting reaction mixture was
stirred and heated at 100 °C for 16 h. After concentration of the reaction mixture, a light brown
1
solid (130 mg) was isolated by column chromatography (EtOAc/hexane 3:7): H NMR (400
MHz, DMSOꢀd6), δ 7.29 (d 1H, J = 4.0), 6.63 (d 1H, J = 3.6), 6.52 (d, 1H, J = 16.0), 6.46 (d, 1H,
J = 16.4), 4.76 (s, 1H), 3.86 (s, 4H), 3.80 (s, 3H), 1.87–1.78 (m, 2H), 1.74–1.64 (m, 2H), 1.60–
1.47 (m, 4H); 91.6% pure by LCMSꢀES 291.34 [M–OH]+.
Methyl (E)ꢀ5ꢀ(2ꢀ(1ꢀhydroxyꢀ4ꢀoxocyclohexyl)vinyl)furanꢀ2ꢀcarboxylate, (27).
To a mixture of 26 (130 mg) in acetone/H2O (1:1) was added TsOH (0.1 equiv) at 0 °C. The
mixture was stirred at 0 °C and gradually warmed to RT over 24 h. The solution was neutralized
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 52
1
2
3
4
5
6
with satd aq NaHCO3, concentrated, diluted with water, and extracted with EtOAc. The organic
solution was concentrated to give a brown solid, which was purified by column chromatography
7
8
9
1
with a gradient eluent, EtOAc/hexane 1:5 to 3:7, to give a light brown solid (100 mg): H NMR
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
(300 MHz, CDCl3), δ 7.15 (d 1H, J = 3.3), 6.60 (d 2H), 6.37 (d, 1H, J = 3.6), 3.90 (s, 3H), 2.86–
2.72 (m, 2H), 2.36 (br, 1H), 2.31 (br, 1H), 2.06 (d, 2H, J = 3.9), 2.03 (d, 2H, J = 3.6), 1.63 (s,
1H); 95.8% pure by LCMSꢀES 247.17 [M–OH]+.
Eꢀ5ꢀ(2ꢀ(1ꢀHydroxyꢀ4ꢀoxocyclohexyl)vinyl)furanꢀ2ꢀcarboxylic acid (28).
To a mixture of 27 (100 mg) in THF/H2O 3:2 was added LiOH•H2O (2.5 equiv) at 0 °C; the
mixture was stirred at 0 °C to RT over 16 h. The solution was concentrated and washed with
Et2O; the water layer was acidified with 2N HCl to pH 4. The aqueous solution was extracted
with EtOAc, which was concentrated to give a light brown solid (70 mg). A sample was purified
1
by column chromatography (EtOAc/hexane/HOAc 1:1:0.01) to give a beige solid (10 mg): H
NMR (400 MHz, DMSOꢀd6), δ 13.01 (br 1H), 7.20 (d 1H, J = 3.6), 6.60 (d 1H, J = 3.6), 6.60 (d
1H, J = 15.9), 6.49 (d 1H, J = 15.9), 5.21 (s, 1H), 2.76–2.58 (m, 2H), 2.13 (br, 1H), 2.09 (br,
1H), 2.04–1.80 (4H); 98.0% pure by LCMSꢀES 249.26 [M–H]–, 499.20 [2M–H]–, 233.19 [M–
OH]+
Eꢀ5ꢀ(2ꢀ(8ꢀHydroxyꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀyl)vinyl)furanꢀ2ꢀcarboxylic acid, (29).
To a solution of 26 (100 mg) in THF: H2O: MeOH (2:1:0.5 mL), LiOH (3 equiv) was added; the
resulting mixture was stirred at RT over 16 h. The mixture was then concentrated, and the solid
was dissolved in minimum amount of H2O and acidified with 2N HCl to pH 4. Extraction with
EtOAc followed by concentration of the organic layer yielded a light brown solid (54 mg), which
was used for next reaction without further purification. The solid was a binary mixture of 29/30
72:26. LCMSꢀES 29 277.26 [M–OH]+ and 30 233.25 [M+H].
ACS Paragon Plus Environment

Page 25 of 52
Journal of Medicinal Chemistry
1
2
3
4
Eꢀ5ꢀ(2ꢀ(4ꢀOxocyclohexꢀ1ꢀenꢀ1ꢀyl)vinyl)furanꢀ2ꢀcarboxylic acid, (30).
5
6
7
8
9
To the binary mixture of 29/30 (50 mg) in THF (0.3 mL), 3N HCl (0.1 mL) was added at 0 °C.
The mixture was stirred and gradually warmed to RT over 6 h. The mixture was concentrated,
diluted with water, extracted with EtOAc, and reꢀconcentrated to yield a brown solid (20 mg): 1H
1H NMR (300 MHz, DMSOꢀd6), δ 12.98 (br 1H), 7.20 (d 1H, J = 3.3), 6.95 (d 1H, J = 16.5),
6.65 (d 1H, J = 3.3), 6.45 (d, 1H, J = 15.9), 6.11 (t, 1H, J = 3.9), 3.05 (br, 2H), 2.65 (t, 2H, J =
6.9), 2.55–2.45 (2H overlapped with DMSOꢀd5); 99.3% pure by LCMSꢀES 233.21 [M+H]+,
231.27 [M–H]–, 463.15 [2M–H]–.
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
8ꢀEthynylꢀ8ꢀ((tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)oxy)ꢀ1,4ꢀdioxaspiro[4.5]decane, (32).
Dihydropyran (1.5 equiv) and PPTS (0.1 equiv) was added to a solution of 3136 (18.0 g, 1 equiv)
in DCM (600 mL) at 0 °C with stirring. The resulting solution with stirring gradually warmed to
RT over 16 h, then Et3N (1.0 mL) was added. A pale yellow liquid (16.0 g) was isolated by
column chromatography (EtOAc/hexane 1:19). The compound (1H NMR in supporting
information) was used in the next step without further purification.
Eꢀ4,4,5,5ꢀTetramethylꢀ2ꢀ(2ꢀ(8ꢀ((tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)oxy)ꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀ
yl)vinyl)ꢀ1,3,2ꢀdioxaborolane, (33).
To a solution of 32 (14.0 g, 1 equiv) 4,4,5,5ꢀtetramethylꢀ[1,3,2]dioxaborolane (1.2 equiv), and
Cp2ZrCl2H (0.15 equiv) cooled to 0 °C, Et3N (0.5 equiv) was slowly added at 0 °C with stirring.
When completed, the solution was heated to 70 °C for 16 h. The reaction mixture was diluted
with hexanes. The precipitate was removed by filtration over short pad of silica gel and washed
with hexanes. Upon concentrating the hexanes solution, a colorless oily liquid formed.
Purification by column chromatography (EtOAc/hexane 1:19) afforded a white solid (7.0 g): 1H
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 52
1
2
3
4
NMR (300 MHz, CDCl3), δ 6.59 (d 1H, J = 18.6), 5.57 (d 1H, J = 18.6), 4.62 (br, 1H), 4.02–3.88
5
6
(m, 5H), 3.48–3.37 (m, 1H), 2.09–1.45 (14H), 1.27 (s, 12H).
7
8
9
Methyl
(E)ꢀ6ꢀ(2ꢀ(8ꢀ((tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)oxy)ꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀ
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
yl)vinyl)picolinate, (34).
A solution of 33 (500 mg, 1.1 equiv), methyl 6ꢀbromopicolinate (1.0 equiv), K2CO3 (2.0 equiv)
in mixture of DME/H2O 9:1 (8 mL) was degassed with argon for 10 min. Then, Pd[(P(Ph)3]4
(0.04 equiv) was added. The resulting reaction mixture was heated at 100 °C for 16 h.
Concentration of the mixture followed by column chromatography (EtOAc/hexane 1:3) yielded a
pale yellow liquid (350 mg). The product (81.1% pure by LCMSꢀES 404.39 [M+H]+, 302.26
[M–101]+) was used without further purification.
Methyl (E)ꢀ6ꢀ(2ꢀ(1ꢀhydroxyꢀ4ꢀoxocyclohexyl)vinyl)picolinate, (35).
To a solution of 34 (230 mg, 1.0 equiv) in acetone/H2O 1:1 (6 mL), TsOH (0.1 equiv) was added
The resulting reaction mixture was stirred at RT over 16 h. Concentration of the mixture
followed by column chromatography (EtOAc/hexane 7:3) gave a pale yellow liquid (110 mg): 1H
NMR (300 MHz, DMSOꢀd6), δ 7.98–7.84 (m 2H), 7.72 (d, 1H, J = 7.5), 6.98 (d 1H, J = 16.2),
6.80 (d 1H, J = 15.9), 5.24 (s, 1H), 3.89 (s, 3H), 2.75–2.61 (m, 2H), 2.16 (br, 1H), 2.11 (br, 1H),
2.07–1.86 (m, 4H); 97.1% pure by LCMSꢀES 276.38 [M+H]+.
(E)ꢀ6ꢀ(2ꢀ(1ꢀHydroxyꢀ4ꢀoxocyclohexyl)vinyl)picolinic acid, (36)
To a solution of 35 (75 mg) in THF/H2O 3:1 (3 mL), LiOH (2.5 equiv) was added at 0 °C. The
mixture was stirred and gradually warmed to RT over 6 h. The reaction mixture was acidified
with citric acid and extracted with a mixture of THF and EtOAc. Concentrating the organic
solution gave an offꢀwhite solid (10 mg); mp 163.8–165.1 °C: 1H NMR (300 MHz, DMSOꢀd6), δ
ACS Paragon Plus Environment

Page 27 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
13.08 (br, 1H), 7.96–7.84 (m, 2H), 7.67 (d, 1H, J = 7.2), 7.05 (d, 1H, J = 16.2), 6.80 (d, 1H, J =
15.6), 5.22 (s, 1H), 2.75–2.61 (m, 2H), 2.16 (br 1H), 2.12 (br 1H), 2.07–1.85 (4H); 97.9% pure
by LCMSꢀES 262.27 [M+H]+.
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
(E)ꢀ4ꢀ(2ꢀ(8ꢀ((Tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)oxy)ꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀyl)vinyl)picolinic
acid, (37).
A solution of 33 (300 mg, 1.2 equiv), methyl 4ꢀbromopicolinate (1.0 equiv), K2CO3 (2.0 equiv)
in mixture of DME/H2O 9:1 (5 mL) was degassed with argon for 10 min. Then, Pd[(P(Ph)3]4
(0.04 equiv) was added. The resulting mixture was stirred and heated at 90 °C for 12 h.
Concentration of the solution followed by column chromatography (EtOAc/hexane 1:3) yielded
1
a pale yellow liquid (200 mg). The product (79.2% pure by LCMSꢀES 390.35 [M+H]+ and H
NMR in supporting information) was used without further purification.
(E)ꢀ4ꢀ(2ꢀ(1ꢀHydroxyꢀ4ꢀoxocyclohexyl)vinyl)picolinic acid, (38)
To a solution of 37 (200 mg, 1.0 equiv) in acetone/H2O 1:1 (6 mL), TsOH (0.1 equiv) was
added. The resulting reaction mixture was stirred at RT for 48 h. The reaction mixture was
acidified with citric acid and extracted with mixture of THF and EtOAc. The solution was
concentrated to give an offꢀwhite solid (18 mg): 1H NMR (300 MHz, DMSOꢀd6), δ 8.60 (d, 1H, J
= 4.5), 8.07 (br, 1H), 7.64 (dd, 1H, J = 4.5, 1.5), 6.93 (d, 1H, J = 16.2), 6.76 (d, 1H, J = 15.9),
5.26 (br, 1H), 2.72–2.59 (m, 2H), 2.18 (br 1H), 2.13 (br 1H), 2.06–1.83 (4H); 99.0% pure by
LCMSꢀES 262.27 [M+H]+.
Methyl
(E)ꢀ2ꢀ(2ꢀ(8ꢀ((tetrahydroꢀ2Hꢀpyranꢀ2ꢀyl)oxy)ꢀ1,4ꢀdioxaspiro[4.5]decanꢀ8ꢀ
yl)vinyl)isonicotinate, (39).
A solution of 33 (437 mg, 1.2 equiv), methyl 2ꢀbromoisonicotinate (1.0 equiv), K2CO3 (2.0
equiv) in mixture of DME/H2O 9:1 (7 mL) was degassed with argon for 10 min. Then,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 52
1
2
3
4
5
6
7
8
9
Pd[(P(Ph)3]4 (0.05 equiv) was added and the reaction mixture was degassed with argon for 10
min. The reaction mixture was stirred and heated at 90 °C for 12 h. Concentration of the solution
followed by column chromatography (EtOAc/hexane 1:3) yielded a pale yellow liquid (300 mg).
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
The product (72.6% pure by LCMSꢀES 404.54 [M+H]+ and 302.53 [M–101]+, H NMR in
supporting information) was used without further purification.
Methyl (E)ꢀ2ꢀ(2ꢀ(1ꢀhydroxyꢀ4ꢀoxocyclohexyl)vinyl)isonicotinate, (40).
To a solution of 39 (300 mg, 1.0 equiv) in acetone/H2O 1:1 (6 mL), TsOH (0.1 equiv) was
added. The resulting reaction mixture was stirred at RT over 48 h. Concentration of the mixture
followed by column chromatography (EtOAc/hexane 7:3) gave an offꢀwhite solid (160 mg). The
product (88.8% pure by LCMSꢀES 276.22 [M+H]+ and 1H NMR in supporting information) was
used without further purification.
(E)ꢀ2ꢀ(2ꢀ(1ꢀHydroxyꢀ4ꢀoxocyclohexyl)vinyl)isonicotinic acid, (41).
To a solution of 40 (100 mg) in THF/H2O 3:1 (3 mL), LiOH (2.5 equiv) was added at 0 °C with
stirring. The mixture was stirred and gradually warmed to RT over 16 h. The reaction mixture
was acidified with citric acid and extracted with mixture of THF and EtOAc. Concentrating the
extract gave an offꢀwhite solid (20 mg): 1H NMR (300 MHz, DMSOꢀd6), δ 13.62 (br, 1H), 8.69
(d, 1H, J = 4.8), 7.84 (br, 1H), 7.60 (dd, 1H, J = 4.8, 1.5), 7.01 (d, 1H, J = 15.6), 6.83 (d, 1H, J =
15.6), 5.21 (s, 1H), 2.75–2.43 (m, 2H), 2.16 (br, 1H), 2.11 (br, 1H), 2.07–1.84 (m, 4H); 92.6%
pure by LCMSꢀES 262.28 [M+H]+.
Surface Plasmon Resonance Studies
Direct binding between LANCL2 proteins and the analogs was measured by SPR method on a
Biacore Tꢀ200 instrument. Experiments were performed at the Biacore Molecular Interactions
ACS Paragon Plus Environment

Page 29 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Shared Resource of Georgetown University, Washington DC. All experiments were done at 25
°C. LANCL2 was immobilized on CM4 chip by amine coupling method (~4000 RU). MOPS (25
mM MOPS, 150 mM NaCl, 1% DMSO and 0.05% Tween 20, pH 6.5) buffer was used the
running buffer. For example, 7 was injected over the proteinꢀcoated surface at 5 different
concentrations (0, 1.25, 2.5, 5.0, 10, 20 µM) in triplicates. Each injection was 60 s with 300 s
stabilization time. The experiment did not require any regeneration buffer. Double reference
subtraction was performed by using an empty reference flow cell and buffer only injections over
protein coated surfaces. Data analysis was done by using steady state affinity model in
BiaEvaluation software.
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
Validation of Computational Modeling Methods
To undergo this process the original MOL2/PDB file LANCL2 was obtained and MOL2/PDB
files for the analogs were generated. Using the most published molecular docking software,
AutoDock Vina, an analog was manually docked against LANCL2 with the top 20 output files
saved as a single PDBQT file and separated for analysis in the molecular viewer UCSF Chimera.
All measures of docking are consistent with the “Vina Tutorial” provided online by The Scripps
Research Institute, who created the software. Only the number of conformations (num_modes =
100) and the exhaustiveness (exhaustiveness = 16) were altered. The exhaustiveness was set to
16 in order to double the amount of time spent identifying the global minimum conformation.
The first step was to initiate “blind” docking, in which the grid box in which the docking event
took place covered the entire protein. This experiment enabled analyses of the different cavities,
where conformers might bind. The grid box dimension for blind docking represented a
rectangular cuboid (74 Å × 68 Å × 60 Å) with grid points separated by 1.000 Å and centered at
the middle of the protein (47.046x, 53.123y, 48.510z). The top 20 docked conformations from
ACS Paragon Plus Environment