Discovery and profiling of a selective and efficacious syk inhibitor

Article abstract of DOI:10.1021/jm5018863

We describe the discovery of selective and potent Syk inhibitor 11, which exhibited favorable PK profiles in rat and dog and was found to be active in a collagen-induced arthritis model in rats. Compound 11 was selected for further profiling, but, unfortunately, in GLP toxicological studies it showed liver findings in rat and dog. Nevertheless, 11 could become a valuable tool compound to investigate the rich biology of Syk in vitro and in vivo.

Full text of DOI:10.1021/jm5018863

Subscriber access provided by University of Oklahoma
Article
Discovery and profiling of a selective and efficacious Syk inhibitor
Gebhard Thoma, Alexander Baxter Smith, Maurice van Eis, Eric Vangrevelinghe, Joachim Blanz,
Reiner Aichholz, Amanda Littlewood-Evans, Christian Lee, Hong Liu, and Hans-Guenter Zerwes
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 29 Jan 2015
Downloaded from http://pubs.acs.org on January 30, 2015
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 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery and profiling of a selective and efficacious Syk inhibitor
a,
a
a
a
Gebhard Thoma, * Alexander B. Smith, Maurice J. van Eis, Eric Vangrevelinghe, Joachim
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
b
b
c
d
d
Blanz, Reiner Aichholz, Amanda LittlewoodꢀEvans, Christian C. Lee, Hong Liu, and Hansꢀ
c
Günter Zerwes
a
b
c
Global Discovery Chemistry, Analytical Sciences & Imaging, Autoimmunity, Transplantation
and Inflammation Research, Novartis Institutes for Biomedical Research, 4056 Basel,
Switzerland;
d
Genomics Institute of the Novartis Research Foundation, 10675 John Jay Hopkins Drive, San
Diego, CA 92121, USA
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 50
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: We describe the discovery of the selective and potent Syk inhibitor 11, which
exhibited favorable PK profiles in rat and dog and was found to be active in a collagen induced
arthritis model in rats. Compound 11 was selected for further profiling but, unfortunately, in GLP
toxicological studies it showed liver findings in rat and dog. Nevertheless, 11 could become a
valuable tool compound to investigate the rich biology of Syk in vitro and in vivo.
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
INTRODUCTION
Spleen tyrosine kinase (Syk) is located in the cytoplasm of hematopoietic lineage cells except
mature T cells. It plays a key role in mediating signal transduction via multiple receptors
containing ITAM motifs (Bꢀcell receptor in Bꢀcells; Fc receptors in myeloid cells, basophils and
mast cells; adhesion receptors; Cꢀtype lectin receptors). Following receptor stimulation, kinases
of the src family phosphorylate tyrosine residues of the receptor intracellular ITAM domains,
which serve as docking sites for the two SH2 domains of Syk. The kinase is recruited to the
receptor and undergoes a conformational change leading to full catalytic activity. Syk
phosphorylates various substrates (such as BLNK in B cells, SLP76 in myeloid cells, PLCγ2a,
Vav, PI3ꢀkinase family members, Cbl) and becomes part of a multiꢀprotein signaling complex ꢀ
the “signalosome” ꢀ leading to the activation of downstream effector pathways such as PKC,
1
MAPK, NFκB. Depending on the cell type and the triggers, effector functions such as
proliferation, cytokine release and oxidative burst are initiated. These are the basis of
physiological responses associated with (auto)immune, inflammatory or allergic reactions.
Blockade of the catalytic activity of Syk kinase is expected to abrogate these signaling pathways
and to attenuate the biological responses. Thus, Syk is considered an attractive target for antiꢀ
inflammatory, antiꢀallergic and autoimmune diseases. In particular, prevention of activation of
2
ACS Paragon Plus Environment

Page 3 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
cells via immune complexes or antigen triggering Fc receptor signaling and prevention of B cell
2
receptor mediated events are believed to have therapeutic potential. To probe this hypothesis in
animals and in clinical trials, the availability of selective, drugꢀlike Syk inhibitors is highly
desirable.
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
Compounds targeting Syk were tested in a battery of assays. Potency and kinase selectivity were
assessed in enzymatic assays based on the Caliper microfluidic mobility shift technology.
Cellular activity was measured in Ramos Bꢀcells upon BCR stimulation with antiꢀIgM which
leads to phosphorylation of the adaptor protein BLNK (B cell linker protein), a direct Syk
substrate. Inhibition of Syk in presence of 90% human blood was monitored in monocytes
3
following FcγR stimulation with an anti CD32 antibody. This leads to the phosphorylation of the
adaptor protein SLPꢀ76 which is also a direct substrate of Syk. To monitor cellular offꢀtarget
inhibition (particularly Jak2) we used an assay measuring the IL3 dependent proliferation of
mouse bone marrow cells.
3
4
Syk is an established target first discovered in 1991. Industryꢀwide efforts have been committed
5
6
to identify selective antagonists but, to our knowledge, BIIBꢀ057 1 (Figure 1) is the only
selective Syk inhibitor, which has been evaluated in clinical studies. However, a planned phase II
7
trial in rheumatoid arthritis (RA) was withdrawn prior to patient enrollment. The compound was
potent in our enzymatic Syk assay (IC = 13 nM) and inhibited only 2 out of 77 additional
5
0
kinases with IC values below 100 nM (ZAP70, 68 nM; PKCα, 90 nM; Table 1). Substantial
5
0
activity was observed in the cellular Syk assay (IC = 178 nM). In human blood monocytes
5
0
stimulated by activating CD32 in presence of blood Syk activity was inhibited with an IC value
5
0
8
9
of ~1 µM. However, in both automated and manual patch clamp assays 1 significantly affected
hERG channel activity with IC values around 10 µM.
5
0
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 50
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
Recently, Gilead disclosed the structure of GSꢀ9973 2, which is being evaluated in cancer clinical
1
0
trials. Their compound was described as a highly potent and selective Syk inhibitor. However,
in our assays 2 showed only modest activity particularly in the presence of blood (IC > 19 µM,
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
Table 1). It is important to note that 2 considerably affected bone marrow cell proliferation which
cannot be explained by Syk inhibition (Table 1).
Figure 1. Structures of clinical compounds 1 and 2 as well as structures of reference compounds
3
, 4 and 5
4
ACS Paragon Plus Environment

Page 5 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Key in vitro data
a
b
a
a
a
c
Syk
Kinase
BM cell
prolif.
Syk
Syk
hERG
hERG
Compound
(enzyme) selectivity
(cell)
(blood)
Qꢀpatch
man. patch
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
42 % @ 10 µM
1
13 ± 4
2 (77)
5853
178 ± 8
952 ± 70
8600
72 % @ 30 µM
2
3
4
5
6
7
8
377 ± 90
5 ± 2
0 (58)
9 (36)
3 (55)
1 (55)
2 (59)
0 (59)
12 (67)
582
249
878 ± 88
778 ± 132
90 ± 15
19416 ±1474
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
4 ± 0.6
22 ± 1
3411
3515
2601
>10000
2110
307 ± 123
433 ± 159
n.d.
>30000
24900
8500
55 ± 2
19 ± 8
445 ± 75
438 ± 34
69 ± 15
165 ± 45
1 ± 0.2
n.d.
>30000
10000
1788 ± 98
1
3 % @ 3 µM
9
5 ± 1
5 ± 0.6
35 ± 4
4 (69)
6 (64)
0 (69)
5267
811
30 ± 2
104 ± 29
99 ± 7
302 ± 57
274 ± 48
367 ± 27
8100
3700
5
0 % @ 10 µM
n.d.
1
1
0
1
2
5900,
0 % @ 10 µM
12 % @ 30 µM
n.d.
9020
>
30000
d
1
1
2
3
1
6 (58)
0 (67)
5270
14 ± 1
177 ± 53
3700
94 ± 2
>10000
170 ± 54
782 ± 372
>30000
n.d.
a
b
c
d
IC in nM; n ≥ 3 for Syk enzyme, cell and blood assays (SEM shown).
5
0
number of kinases with IC < 100 nM in addition to Syk (number of kinases tested).
5
0
inhibition at indicated concentration; 3 independent measurements for each concentration
tested once
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 50
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
In a previous communication we reported a series of napthyridinone and pyridopyrimidinone Syk
1
1
inhibitors with bicyclic cores sharing the binding mode of compound 1. Initial compounds such
as 3 containing substituted anilines displayed poor kinase selectivity and moderate cellular
activity (IC = 778 nM, Table 1). In contrast, replacement of aniline with a 7ꢀaminoindole
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
substituent led to compounds such as 4 and 5 with acceptable kinase selectivity, good cellular
activity (IC < 100 nM), high potency in the presence of blood (IC < 500 nM) and little or no
5
0
50
hERG channel inhibition (Table 1). Unfortunately, we failed to improve the poor PK properties
of these compounds and, consequently, these series were abandoned. In efforts to overcome these
PK liabilities, we investigated monocyclic analogs of 1 based on pyridine (6, 7), pyrazine (8, 9),
triazine (10, 11, 12) and pyrimidine cores (13).
Here we describe discovery and profiling of the potent, highly selective and orally bioavailable
Syk inhibitor 11, which shares the binding mode of 1, 3, 4 and 5.
6
ACS Paragon Plus Environment

Page 7 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CHEMISTRY
The pyridine derivative 6 was prepared from building block 14 (Scheme 1). The reaction with
aniline 15 led to 16. Compound 15 was treated with an excess of LiHMDS (3 eq.) prior to the
addition of acid 14 to obtain the desired regioisomer 16 in good yield. Compound 16 was
transformed into the methyl ester 17. The Bocꢀprotected diaminocyclohexane 18 was introduced
giving 19, which was converted into amide 20. Removal of the protecting group afforded
compound 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
a
Scheme 1. Synthesis of compound 6
a
Reagents and conditions: (a) (1) 15, LiHMDS, THF, 1 h, ꢀ78⁰C, (2) add 14, ꢀ78→25⁰C, 1 h, (b)
(1) CDI, DMF, 16 h, 25⁰C, (2) add MeOH, 0.5 h, 25⁰C, (c) 18, DIPEA, NMP, 16 h, 120⁰C, (d)
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 50
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) LiOH, H O, dioxane, 2 h, 100⁰C, (2) COMU, DIPEA, DMF, NH OH, 1 h, 25⁰C, (e) TFA,
2
4
CH Cl , 1 h, 25⁰C.
2
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
Compound 7 was accessible from building block 21 (Scheme 2). The Bocꢀprotected
diaminocyclohexane 18 was introduced to give 22. Palladium catalyzed coupling with
aminoindole 23 led to 24 which was transformed into amide 25. Deprotection afforded compound
7
.
a
Scheme 2. Synthesis of compound 7
a
Reagents and conditions: (a) 18, NMP, NEt , 16 h, 75⁰C, (b) 23, Pd(OAc) , Xantphos, K CO ,
3
2
2
3
dioxane, 1.5 h, 150⁰C, (c) H O , NaOH, DMSO, EtOH, 3 h, 25⁰C, (d) HCl, CH Cl , 2 h,
2
2
2
2
0
→25⁰C.
8
ACS Paragon Plus Environment

Page 9 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Pyrazine 8 was obtained from building block 26, which was reacted with the Bocꢀprotected
diaminocyclohexane 18 to give 27 (Scheme 3). The regioisomer of 27 was also isolated (19%).
Introduction of aniline 15 led to 28, which was transformed into amide 29. Deprotection afforded
compound 8.
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
Scheme 3. Synthesis of compound 8
NH2
O
Cl
NHBoc
O
Cl
O
N
18, a)
N
O
N
NH
71%
27
N
NHBoc
Cl
26
N
N
N
N
N
N
N
H2N
15, b)
O
HN
N
c)
3
7%
O
80%
NH
2
8
NHBoc
N
N
N
N
N
N
O
HN
d)
O
HN
H2N
N
75%
H2N
N
N
N
NH
NH
2
9
8
NHBoc
NH2
a
Reagents and conditions: (a) 18, DMF, NEt , 16 h, 0→25⁰C, (b) 15, Pd(OAc) , Xantphos,
3
2
K CO , dioxane, 16 h, 90⁰C, (c) (1) LiOH, H O, dioxane, 2 h, 25⁰C, (2) COMU, DIPEA, DMF,
2
3
2
NH OH, 0.5 h, 25⁰C, (d) HCl, CH Cl , MeOH, 16 h, 25⁰C.
4
2
2
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 50
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
Indole derivative 9 was obtained from building block 30 which was reacted with the Bocꢀ
protected diaminocyclohexane 18 to give 31 (Scheme 4). Introduction of aminoindole 23 led to
32 which was transformed into amide 33. Deprotection led to compound 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
2
3
4
5
6
7
8
9
0
a
Scheme 4. Synthesis of compound 9
a
Reagents and conditions: (a) 18, DMF, NEt , 16 h, 0→25⁰C, (b) 23, Pd(OAc) , Xantphos,
3
2
K CO , dioxane, 16 h, 90⁰C, (c) H O , NaOH, DMSO, EtOH, 2 h, 0→25⁰C, (d) HCl, CH Cl ,
2
3
2
2
2
2
MeOH, 16 h, 25⁰C.
1
0
ACS Paragon Plus Environment

Page 11 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
Triazine 10 was prepared from building block 34 which was reacted with aniline 15 to give 35
(
Scheme 5). Compound 35 was transformed into amide 36. Oxidation of 36 with mꢀCPBA led to
the formation of a mixture of sulfoxide and sulfone, which was treated with the Bocꢀprotected
diaminocyclohexane 18 to give 37. Deprotected afforded compound 10.
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
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
a
Scheme 5. Synthesis of compound 10
a
Reagents and conditions: (a) 15, NMP, 0.5 h, 0→25⁰C, (b) NH (7M) in MeOH, 2 h, 25⁰C, (c)
3
(
1) mꢀCPBA, DMF, 2 h, 0→45⁰C, (2) 18, NEt , 2 h, 25→65⁰C, (d) TFA, CH Cl , 1 h, 25⁰C.
3
2
2
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 50
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
Indole derivative 11 was also prepared from building block 34 (Scheme 6). The introduction of 7ꢀ
aminoindole 23 led to 38. The indole was protected to give 39, which was transformed into amide
40. The sulfur atom of compound 40 was oxidized leading to a mixture of sulfoxide and sulfone.
Susequent treatment with the Bocꢀprotected diaminocyclohexane 18 gave 41. Deprotection
furnished compound 11. The deprotection had to be carefully monitored because of the formation
of side products, which were difficult to separate (probably dimerization of the indole due to the
acidic reaction conditions). Compound 12 was prepared following analogous procedures but
using 4ꢀaminoindole instead of 7ꢀaminoindole (see experimental section).
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
a
Scheme 6. Synthesis of compound 11 and 12
1
2
ACS Paragon Plus Environment

Page 13 of 50
Journal of Medicinal Chemistry
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
a
Reagents and conditions: (a) 23, NMP, 0.25 h, 25⁰C, (b) DMAP, (Boc) O, THF, 0.5 h,
2
0
→25⁰C, (c) NH (7M) in MeOH, 0.75 h, 25⁰C, (d) (1) m-CPBA, DMF, 0.75 h, 0⁰C, (2) 18,
3
NEt , 0.25 h, 65⁰C, (e) TFA, H O, CH Cl , 17.5 h, 0→25⁰C.
3
2
2
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
Pyrimidine 13 was prepared from building block 42, which was reacted with aminoindole 23 to
give 43. Introduction of benzyloxycarbonylꢀprotected diaminocyclohexane 44 led to 45, which
was deprotected to afford compound 13. The benzyloxycarbonylꢀprotected building block 44 was
used to avoid acidic deprotection conditions, which led to side product formation in the synthesis
of compound 11.
a
Scheme 7. Synthesis of compound 13
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 50
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
a
Reagents and conditions: (a) 22, THF, NEt , 16 h, 25⁰C, (b) 44, DMF, NEt , 3 h, 110⁰C, (c)
3
3
Pd/C, H , MeOH, 3.5 h, 25⁰C.
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
RESULTS AND DISCUSSION
Compounds 6-13 were broadly assessed and compared to competitor compound 1 (Table 1). In
the enzymatic Syk assay the triazolylꢀanilines 1, 6, 8 and 10 (IC values of 13, 19, 1 and 5 nM,
50
respectively) were generally found to be more potent than the corresponding 7ꢀaminoindoles 13,
7
, 9 and 11 (IC values of 94, 165, 5 and 35 nM, respectively). However, in the cellular Syk
50
assay compound pairs with the same core structure showed similar potency (pyrimidines 1 and
3, 178 and 170 nM; pyridines 6 and 7, 445 and 438 nM; pyrazines 8 and 9, 69 and 30 nM;
1
triazines 10 and 11, 104 and 99 nM). With the exception of pyridines 6 and 7 all new compounds
8ꢀ13 were at least equipotent to 1. As we had observed previously for related bicyclic compounds
11
3
, 4, and 5 (Table 1) the 7ꢀaminoindoles 13, 7, 9 and 11 were found to be more selective in our
enzymatic kinase panel than the corresponding triazolylꢀanilines 1, 6, 8 and 10. The inferior
selectivity of the triazolylꢀanilines compared to the 7ꢀaminoindoles was confirmed by more
pronounced effects in the bone marrow cell proliferation assay, which is independent of Syk
(
Table 1). The 4ꢀaminoindole 12 was found to be the most potent compound with IC values of 1
50
and 14 nM in the enzymatic and cellular assays, respectively. Unfortunately, 12 showed limited
selectivity inhibiting 6 out of 58 kinases with IC values below 100 nM but did not significantly
5
0
affect bone marrow cell proliferation (IC > 5 µM). We previously had observed the reduced
5
0
kinase selectivity of 4ꢀaminoindole derivatives with bicyclic cores sharing the binding mode of
1
1
the compounds discussed here.
1
4
ACS Paragon Plus Environment

Page 15 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To understand the superior kinase selectivity of the 7ꢀaminoindoles we compared the high
1
2
resolution Xꢀray coꢀstructures of compounds 10 and 11 bound to Syk catalytic domain. Both the
triazolylꢀaniline 10 and the corresponding indole 11 form two strong hydrogen bonds with the
hinge sequence of Syk (E449, A451; Figure 2). Both the aniline of 10 and the indole of 11 are
sandwiched between the side chain of L377 and P455 located at the proximal Pꢀloop and
downstream hinge regions, respectively. However, the orientation of the aromatic substituents
differs. The Nꢀlinked triazole of 10 is oriented toward the diaminocyclohexane whereas the 5ꢀ
membered ring of the indole of 11 is directed toward the downstream hinge region of Syk.
Overall, the bioactive (bound) conformation of aniline 10 is more compact compared to 11. The
indole of 11 can form an additional Hꢀbond with A451 11. However, this Hꢀbond is most likely
weak as the carbonyl oxygen of A451and the nitrogen of the indole are not optimally positioned.
The weakness of the hydrogen bond is supported by the slightly inferior activity of the 7ꢀ
aminoindoles in the biochemical assay. As we did not attempt to coꢀcrystallize our compounds
with other kinases, the rationale for the superior selectivity of the 7ꢀaminoindoles remained
elusive.
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
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 50
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
Figure 2. Crystal structures of triazines 10 (A) and 11 (B) bound to the kinase domain of Syk
Pyrimidine 1 and pyridine 6 showed superior kinase selectivity compared to pyrazine 8 and
triazine 10. In contrast to 1 and 6, compounds 10 and 8 can form an intramolecular Hꢀbond
between the amide NH and the triazine core, which is expected to reduce the rotation of the
amide out of the plane of the aromatic ring (Figure 3A). This assumption was supported by
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
12
DFT/B3LYP calculations in the gas phase indicating a more pronounced distortion of the amide
for 1 compared to 10 with dihedral angles of 16⁰ and 2⁰, obtained for the respective optimized
geometries (Figure 3B). As compound 3 with a perfectly planar bicyclic pyridopyrimidinone
core, which shares the binding mode of compounds 1, 6, 8 and 10 also showed limited kinase
1
1
selectivity (Table 1) , we speculated that a distortion of the hingeꢀbinding amide motif improves
the kinase selectivity of compounds with this binding mode. However, when we analyzed the
structures of pyrimidine 1 and triazine 10 bound to Syk catalytic domain we learned that the
bound conformations superimposed almost perfectly and showed very small distortions of the
amide with similar dihedral angles of ~8⁰ and ~4⁰ for 1 and 10, respectively.
Pyrazines 8 and 9 as well as triazines 10 and 11 which form an intramolecular Hꢀbond (flat core)
were more potent but less selective than the corresponding pyridines 6 and 7 or pyrimidines 1 and
13 which do not form an intramolecular Hꢀbond (Table 1). Thus, compounds 9 and 11 combining
a 7ꢀaminoindole substituent (inducing selectivity) and a “flat” core (inducing potency) were
considered most promising.
1
6
ACS Paragon Plus Environment

Page 17 of 50
Journal of Medicinal Chemistry
1
2
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
Figure 3. A) Compounds 10 and 8 can form an intramolecular Hꢀbond, compounds 1 and 6
cannot. Thus, an increased distortion of the amide is expected for 1 and 6. B) DFT/B3LYP
1
2
calculations in the gas phase indicate a more pronounced distortion of the amide for 1 compared
to 10 with dihedral angles of 16⁰ and 2⁰, measured for the respective optimized geometries. C)
The conformations of compounds 1 and 10 bound to the Syk catalytic domain superimpose
almost perfectly and show very small distortions of the amide with similar dihedral angles of ~8⁰
and ~4⁰ for 1 and 10, respectively. For simplicity, the amino acids of the protein are omitted.
Compounds 8, 9, 10, 11, 12 and 13 were tested for Syk inhibition in whole blood monocytes
(IC values of 1788, 302, 274, 367, 177 and 782 nM, respectively) and – with the exception of 8
5
0
–
were found to be superior to 1 (IC = 952 nM). Interestingly, compared to the 7ꢀaminoindole
50
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 50
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
1, the corresponding 4ꢀaminoindole 13 was only 2 fold more potent in the blood assay despite of
its clearly superior potencies in both enzymatic and cellular assays (35ꢀfold and 7ꢀfold,
respectively. Inhibition of the hERG channel was assessed in an automated patch clamp assay
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
8
(Qꢀpatch, Table 1). The 7ꢀaminoindoles 13, 7 and 11 showed reduced effects (IC > 20 µM)
5
0
compared to their triazolylꢀaniline analogs 1, 6 and 10 (IC < 10 µM). The pyrazines 8 and 9
50
showed similar IC values (~ 10 µM). The 4ꢀaminoindole 12 significantly inhibited hERG
5
0
channel activity with an IC value of 3.7 µM. Compounds 9 and 11 were also tested in a manual
50
9
patch clamp assay. Pyrazine 9 considerably inhibited the hERG channel activity (50% at 10 µM)
whereas triazine 11 did not show any inhibition at 10 µM and only 12 % inhibition at 30 µM and,
thus, differentiated favorably from compound 1.
Compounds 9 and 11 showed favorable PK properties in rat (Table 2). Medium clearance (CL)
and high volume of distribution (V ) led to long mean residence times (MRT). Both exposure
SS
(AUC) and oral bioavailability (BAV) were acceptable. Slow absorption led to flat PK curves
avoiding high peak concentrations (Figure 4). In dogs, compound 11 also showed good exposure,
acceptable oral bioavailability and a long MRT (Table 2). The corresponding 4ꢀaminoindole 12
1
3
showed very limited exposure, higher clearance and a shorter MRT. Due to the poor PK
properties, the hERG channel inhibition and limited selectivity the slightly more potent
compound 12 was not further considered (2ꢀfold more potent than 9 and 11 in the most relevant
whole blood assay, Table 1).
1
8
ACS Paragon Plus Environment

Page 19 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. PK data on compounds 9, 11 and 12
Parameter
species
9
11
12
a
a
b
a
rat
rat
dog
rat
83
10.3
2.1
522
19
4
ꢀ
1
ꢀ1
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
CL (mL min kg )
28
12.3
7.2
31
16.4
8.7
11
VSS (L/kg)
6.8
8.4
MRT (h)
c
AUC iv (nM h)
1'610
1'523
95
1'459
877
60
3414
1836
54
c
AUC po (nM h)
BAV (%)
c
Cmax (nM)
108
80
139
4
a
Cassette dosing in Sprague Dawley rats; i.v. 1 mg/kg, NMP:PEG200 (3:7); p.o. 3 mg/kg,
CMC:water:tween (0.5:99:0.5).
b
One in one dosing in Beagle dogs; i.v. 0.1 mg/kg, NMP:PEG200 (3:7); p.o. 0.3 mg/kg,
MC:water (0.5:99.5).
c
dose normalized.
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 50
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
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
Figure 4. PK curves for 9 and 11 following i.v. and p.o. dosing in rat
Compound 11 was selected for further profiling. For a broader exploration of the kinase
14
selectivity, the binding of 11 to a panel of 451 kinases was assessed (Figure 5). Only 14 of the
kinases demonstrated remarkable binding affinity to 11 and K ꢀvalues were determined.
d
Compound 11 was found to bind most strongly to Syk (K = 0.64 nM) exhibiting 20ꢀfold
d
selectivity over Pak7, 25ꢀfold selectivity over Pak4 and ≥100ꢀfold selectivity over the remaining
kinases. The K for Zap70 which is the closest analog of Syk was 65 nM resulting in a selectivity
d
factor of 100. In the biochemical Caliper assays, we observed only a 7ꢀfold selectivity over
Zap70 (Table 3). However, in a cellular assay interrogating Zap70 activity (antiCD3ꢀinduced
Zap70ꢀdependent phosphorylation of SLP76 in Jurkat cells), compound 11 was found to be 25ꢀ
fold less potent than in a similar assay interrogating Syk activity (anti IgMꢀinduced BLNK
2
0
ACS Paragon Plus Environment

Page 21 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
phosphorylation; Table 3). Compound 1 showed only 10ꢀfold selectivity in this most relevant
cellular assay. Overall, compound 11 exhibited a very promising kinase selectivity profile to
warrant further experiments.
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
Figure 5. Kinase selectivity profile of compound 11. In addition to Syk only 13 out of 451
kinases were affected at a concentration of 300 nM. For these kinases K values were determined.
d
Table 3. Selectivity of compound 11 over Zap70
Assay [nM]
Syk (enzyme)
Syk (Kd)
1
13 ± 4
n.d.
11
35 ± 4
0.64
Syk (cell)
178 ± 8
76 ± 13
n.d.
99 ± 7
251 ± 23
65
Zap70 (enzyme)
Zap70 (Kd)
Zap70 (cell)
1805 ± 305
2526 ± 493
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 50
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
We established a rat PK/PD model in which compound 11 was administered to Lewis rats (Figure
4). Blood samples were taken at different times post dosing for PK determinations as well as
assessment of inhibition of Sykꢀdependent signaling events. For this, the extent of SLP76
phosphorylation in monocytes in response to stimulation by anti CD32 was quantified. A
representative experiment is shown in Figure 6A. A dose of 22 mg/kg of 11 led to 90, 85 and 45
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
%
inhibition after 2, 4 and 24 h post administration, respectively. Blood concentrations were 900,
1100 and 300 nM, respectively. Analysis of individual samples from several experiments
involving different doses (30 ꢀ 3 mg/kg) and time points (2, 4, 24 h) indicated a dose and
exposureꢀdependent inhibition of the PD response ex vivo (FcgRꢀinduced PꢀSLP76 in peripheral
blood monocytes) with blood IC levels of 189 nM (Figure 6B).
5
0
2
2
ACS Paragon Plus Environment

Page 23 of 50
Journal of Medicinal Chemistry
1
2
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
Figure 6. PK/PD experiments with compound 11. (A) Single dose treatment of Lewis rats
followed by exꢀvivo assessment of blood compound levels (PK) and FcgRꢀinduced PꢀSLP76 in
peripheral blood monocytes (PD). (B) The IC value was determined by PK/PD analysis of
5
0
individual samples from several experiments involving several doses (30 ꢀ 3 mg/kg) as well as
several time points (2, 4, 24 h). The colors indicate data from distinct dosing groups (dark green:
30 mg/kg; light blue: 22 mg/kg; light green and dark blue: 10 mg/kg; yellow: 3 mg/kg).
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 50
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
Efficacy of 11 was demonstrated in vivo in a rat collagenꢀinduced arthritis model in which the
compound dosing was started when rats developed paw swelling (day 15) subsequent to
1
5
immunization (day 0) and boost (day 7) with porcine collagen. In this “therapeutic” model
compound 11 showed rapid and sustained reversal of paw swelling at 10 and 30 mg/kg p.o. q.d.
and inhibition of swelling at 3 mg/kg p.o. q.d. with slow onset but significant inhibition vs.
placebo (Figure 7A). The treated animals displayed higher body weight gain compared to
vehicleꢀtreated animals due to reduced joint swelling and good tolerability of compound 11
(Figure 7B). The reversal of joint swelling was also reflected in improved histology (reduced
inflammation and cartilage and proteoglycan loss; Figure 7C). Peak compound concentrations in
the blood of treated animals were 126 nM, 866 nM and 2891 nM for the doses of 3, 10 and 30
mg/kg, respectively (Figure 7D). It is worthwhile to mention that a statistically significant
reduction of joint swelling (3 mg/kg dose) was observed at exposure levels below the IC value
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
of 189 nM determined for 11 in the PK/PD model.
2
4
ACS Paragon Plus Environment

Page 25 of 50
Journal of Medicinal Chemistry
1
2
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
Figure 7. Inhibition of collagenꢀinduced arthritis in female Lewis rats by compound 11. (A) Rats
were immunized on day 1 with collagen and boosted after 7 days. Joint swelling developed in
hind paws. Rats with swollen joints were randomized to treatment groups. Compound 11 was
administered once daily at the indicated doses. (B) Body weight of treated and untreated rats. (C)
Histology scores for inflammation, bone and cartilage damage in hind paws at termination of the
experiment on day 28 (D) Blood levels of compound 11 for all doses were determined on day 18
rd
(
3 day of treatment) at the indicated time points.
Because of its favorable profile compound 11 was investigated in toxicology studies in rats and
dogs. Unfortunately, in 4 weeks GLP studies we observed liver findings such as single necrosis
of hepatocytes, bile plugs and portal inflammation. Furthermore, in the GLP in vitro hERG assay,
compound 11 showed an IC ꢀvalue of 5 µM, which was unexpected considering the nonꢀGLP
5
0
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 50
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
data (Table 1). However, in a dog telemetry study no prolongation of the QT interval was
observed. Compound 11 was nonetheless abandoned mainly because of an insufficient
therapeutic index for autoimmune indications.
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
As Syk expression is limited to hematopoietic lineage cells , we attributed the liver findings to
the chemical structure of 11 but not to its mode of action. Studies of the metabolism of 11 in vitro
(liver microsomes) and in vivo (rat) did not indicate the presence of reactive metabolites.
However, experiments with radiolabeled 11 in hepatocytes pointed to covalent adduct formation.
Furthermore, incubation of 11 with hepatocytes afforded cysteine conjugate 46 (Figure 8A)
which possibly originated from a reactive metabolite trapped by glutathione or cysteine. To the
best of our knowledge, the formation of a cysteine adduct such as 46 is unprecedented.
Metabolite 46 (0.9 µg) was isolated from incubation of parent compound 11 with monkey liver
slices. The structure was assigned by NMR based on the following observations (Figure 8B): (a)
The spin system Hꢀ4, Hꢀ5, Hꢀ6 was observed by TOCSY (total correlation spectroscopy)
indicating that there was no additional substitution compared to 11. (b) A proton of the indole
nitrogen (Nꢀ1) was not observed. (c) The signal for Hꢀ4 was shifted to 8.31 ppm (7.44 ppm in 11,
Figure 8C) which is in line with a deꢀshielding effect of the carbonyl function at Cꢀ3. (d) A single
proton at the nitrogen (Nꢀ18) of the cysteine was detected and the chemical shift of 4.12 ppm for
Hꢀ19 indicated an Nꢀlinked cysteine. Additional evidence came from MS experiments (Figure
8D). Collisionꢀinduced dissociation of the protonated molecular ion of 46 (m/z=500) resulted in
the loss of CO (m/z=456) followed by the loss of hydrogen sulfide (m/z=422). The loss of
2
hydrogen sulfide can only occur if the cysteine is linked to the indole via nitrogen but not if it is
linked via sulfur. The elemental formula of 46 and the fragment ions was supported by accurate
2
6
ACS Paragon Plus Environment

Page 27 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mass measurements. The metabolic pathway leading to 46, the structure of the primary
metabolite, as well as the contribution of this metabolite to the observed liver findings, remains
elusive.
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
Figure 8. Metabolite 46. (A) Formation in hepatocytes from different species. (B) HꢀNMR
1
spectrum of 46 and assignment of key signals (0.9 µg, DMSO-d6). (C) HꢀNMR spectrum of 11
and assignment of key signals (DMSO-d6). Additional signals and broadening of signals due to
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 50
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
existence of different rotamers. (D) Collisionꢀinduced fragmentation of the protonated molecular
ion of 46.
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
CONCLUSIONS
We have described the discovery and characterization of the efficacious and selective Syk
inhibitor 11. Unfortunately, compound 11 could not be advanced to clinical trials because of liver
findings in toxicology studies limiting the therapeutic index. However, we believe that compound
11 is a good tool compound to study the biology of Syk in vitro and in vivo.
EXPERIMENTAL SECTION
General. Reagents, solvents and heterocyclic building blocks (14, 21, 26, 30, 34, and 42) were
purchased from commercial sources and used without further purification. All reactions were
carried out under an atmosphere of Argon. Highꢀresolution LC/ESIꢀMS data were recorded using
a Thermo Scientific LTQ Orbitrap XL mass spectrometer with an electrospray ionization source
and a Shimadzu Nexera liquid chromatograph equipped with a diode array detector. The NMR
1
spectra were obtained using a 400 MHz spectrometer. All H NMR spectra are reported in δ units
(
ppm) and were recorded in DMSO-d6 or CD OD and referenced to the solvent peaks. Liquid
3
chromatography was performed on Acquity UPLC/MS systems (Waters, Milford, MA) equipped
with a binary solvent manager, a sample manager, a column manager, a photodiode array detector
(PDA) and a Waters ZQ2000 MS detector. Acquity UPLC columns (Waters, Milford, MA) with
dimensions of 2.1x50 mm and packed with High Strength Silica (HSS) T3 particles of 1.8µ
diameter and 100 Å pore size were used. The columns were housed at constant 60°C inside the
column manager with a tolerance band of 0.1°C. The flow was 1 ml/min. The mobile phase
2
8
ACS Paragon Plus Environment

Page 29 of 50
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
consisted of (A) water + 0.05% formic acid and (B) acetonitrile + 0.04% formic acid. The
runtime was 2 min. The following gradient was used: from A/B 95:5→ 2:98 within 1.40 min;
A/B 2:98 for 0.4 min; from A/B 2:98→95:5 within 0.1 min; A/B 95:5 for 0.1 min. UV absorption
was monitored at λ = 210ꢀ450 nM. The MS detector was operated in continuously
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
positive/negative ESI alternating mode with full scan from 120ꢀ1200 Da in 0.3 seconds. Mass
spectra were acquired and stored in centroid mode. MS based confirmation of molecular weight
+
ꢀ
was based on the formation of the pseudoꢀmolecular ions [M+H] in positive mode and [MꢀH] in
negative mode. All test compounds reported in this manuscript had a purity >95%.
1
13
Structure elucidation of 46: NMR spectra ( H, C, 2D ROESY (rotating frame nuclear
Overhauser effect), HSQC (heteronuclear single quantum coherence) and HMBC (heteronuclear
multiple bond correlation) were measured on a Bruker AVANCE III spectrometer (600 MHz
1
13 15
proton frequency) equipped with a 1.7 mm H{ C, N} Bruker Biospin’s TCI
1
13
MicroCryoProbe™. H and C shifts were referenced internally to the solvent signals at 2.50
ppm and 39.5 ppm, respectively. Each metabolite sample was dissolved in about 40 µL DMSO-
d6 and transferred to a 1.7 mm NMR tube. MS investigations were performed on a LTQ XL
Orbitrap (Linear Quadrupole 2D Ion Trap / Orbitrap, Thermo Scientific, CA, USA) mass
spectrometer equipped with a captive spray ionization (CSI) source (Microchrom Bioresources,
Auburn, USA) operating in positive mode electrospray ionization (ESI). The resolution was set to
30000 in full scan and 15000 in MS(n) mode. The omnipresent polysiloxane background ion
[
C H SiO] with m/z 445.12003 was used as external lock mass.
2
6
6
16. LiHMDS (21.7 ml of 1M solution in THF, 21.7 mmol) was added at ꢀ78⁰C to a solution of 15
1193 mg, 7.45 mmol) in THF (15 ml) and the mixture stirred for 1 h. A solution of 14 (1300 mg,
(
6.77 mmol) in THF (5 ml) was added at ꢀ78⁰C, the mixture warmed to 25⁰C and stirred for 1 h.
2
9
ACS Paragon Plus Environment