Conversion of carbazole carboxamide based reversible inhibitors of Bruton's tyrosine kinase (BTK) into potent, selective irreversible inhibitors in the carbazole, tetrahydrocarbazole, and a new 2,3-dimethylindole series

Article abstract of DOI:10.1016/j.bmcl.2018.07.041

Incorporation of a suitably-placed electrophilic group transformed a series of reversible BTK inhibitors based on carbazole-1-carboxamide and tetrahydrocarbazole-1-carboxamide into potent, irreversible inhibitors. Removal of one ring from the core of these compounds provided a potent irreversible series of 2,3-dimethylindole-7-carboxamides having excellent potency and improved selectivity, with the additional advantages of reduced lipophilicity and molecular weight.

Full text of DOI:10.1016/j.bmcl.2018.07.041

Accepted Manuscript
Conversion of Carbazole Carboxamide Based Reversible Inhibitors of Bru-
ton’s Tyrosine Kinase (BTK) into Potent, Selective Irreversible Inhibitors in the
Carbazole, Tetrahydrocarbazole, and a New 2,3-Dimethylindole Series
Qingjie Liu, Douglas G. Batt, Charu Chaudhry, Jonathan S. Lippy, Mark A.
Pattoli, Neha Surti, Songmei Xu, Percy H. Carter, James R. Burke, Joseph A.
Tino
PII:
S0960-894X(18)30636-X
https://doi.org/10.1016/j.bmcl.2018.07.041
BMCL 25972
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
10 May 2018
17 July 2018
29 July 2018
Please cite this article as: Liu, Q., Batt, D.G., Chaudhry, C., Lippy, J.S., Pattoli, M.A., Surti, N., Xu, S., Carter, P.H.,
Burke, J.R., Tino, J.A., Conversion of Carbazole Carboxamide Based Reversible Inhibitors of Bruton’s Tyrosine
Kinase (BTK) into Potent, Selective Irreversible Inhibitors in the Carbazole, Tetrahydrocarbazole, and a New 2,3-
Dimethylindole Series, Bioorganic & Medicinal Chemistry Letters (2018), doi: https://doi.org/10.1016/j.bmcl.
2018.07.041
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Bioorganic & Medicinal Chemistry Letters
Conversion of Carbazole Carboxamide Based Reversible Inhibitors of Bruton’s
Tyrosine Kinase (BTK) into Potent, Selective Irreversible Inhibitors in the
Carbazole, Tetrahydrocarbazole, and a New 2,3-Dimethylindole Series
Qingjie Liu, Douglas G. Batt*, Charu Chaudhry, Jonathan S. Lippy, Mark A. Pattoli, Neha Surti, Songmei
Xu, Percy H. Carter, James R. Burke, Joseph A. Tino
Bristol-Myers Squibb Company, PO Box 4000, Princeton, NJ 08543-4000, USA
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
Incorporation of a suitably-placed electrophilic group transformed a series of reversible BTK
inhibitors based on carbazole-1-carboxamide and tetrahydrocarbazole-1-carboxamide into
potent, irreversible inhibitors. Removal of one ring from the core of these compounds provided a
potent irreversible series of 2,3-dimethylindole-7-carboxamides having excellent potency and
improved selectivity, with the additional advantages of reduced lipophilicity and molecular
weight.
Keywords:
Bruton’s Tyrosine Kinase
BTK
2018 Elsevier Ltd. All rights reserved.
Indole carboxamide
Carbazole
Irreversible inhibitor
Bruton’s tyrosine kinase (BTK) is a member of the TEC
family of nonreceptor tyrosine kinases which plays an important
role in B cell receptor mediated signaling. Because of the
involvement of this signaling pathway in the function and
proliferation of B lymphocytes, inhibitors of BTK have been
actively pursued in the pharmaceutical industry for the treatment
of autoimmune diseases such as rheumatoid arthritis and
systemic lupus erythematosis, as well as various B cell related
cancers.1 Ibrutinib (Imbruvica®) is the first marketed inhibitor of
BTK, approved for the treatment of chronic lymphocytic
Due to the potential for exceptional potency and selectivity,
there has been widespread interest in recent years in irreversible
kinase inhibitors, although there are obvious concerns about the
potential for off-target or idiopathic toxicity.6 Ibrutinib is an
irreversible inhibitor of BTK, and many recent publications of
experimental BTK inhibitors in the peer-reviewed and patent
literature also disclose irreversible inhibitors.1 These compounds
contain electrophiles, generally acrylamide or related Michael
acceptors, designed to alkylate cysteine 481 which is located in
the ATP binding site of BTK. A cysteine at this location is
relatively rare in the kinome, occurring in only 10 other kinases,
so selectivity is generally accepted to be a manageable concern.
leukemia,
mantle
cell
lymphoma,
Waldenström’s
macroglobulinemia, marginal zone lymphoma and chronic graft
versus host disease.2
Since we first described 4,7-disubstituted carbazole-1-
carboxamides as reversible BTK inhibitors,3 two additional
reports from our laboratories have detailed further structure-
activity relationships in the carbazole and tetrahydrocarbazole
series. This work culminated in the discovery of BMS-935177
(1)4 and the single atropisomer BMS-986142 (2),5 two very
potent and selective reversible inhibitors of BTK. The latter
compound, unlike the former, was a single enantiomer with
respect to the 4-substituent, due to severely restricted rotation
about the atropisomeric bonds to the linking phenylene moiety.
Compound 2 showed significant improvements in potency and
selectivity over 1, and is currently being studied clinically for the
treatment of autoimmune diseases.

Additional analogs, also shown in Table 1, were prepared to
briefly explore the SAR of the electrophilic moiety. Mono- (6) or
di-substitution (7) at the -position of the acrylamide reduced
potency against the enzyme, in the case of 7 back to the same
level as 1. In both compounds, cell potency was reduced even
further. This is consistent with conjugate addition of the thiol of
Cys481 to the acrylamide, since steric hindrance from the methyl
substituents would be expected to impede this process. These
changes also returned selectivities against other kinases to levels
similar to those seen with 1. N-Methylation of the acrylamide
moiety, accompanied by removal of the methyl substituent at the
R2' position (compound 8), improved potency even further,
although cell potency was reduced. Replacement of the
acrylamide with the nearly isosteric and non-electrophilic
propionamide (9) gave a compound with even less potency than
1.
We now report the conversion of our carbazole- and
tetrahydrocarbazole-based series of BTK inhibitors into
irreversible inhibitors by incorporation of a suitably-placed
electrophilic group. These compounds further improved the
potency and selectivity of our previous series. Also, dissection of
the tetrahydro ring of the latter series led to a new series of 2,3-
dimethylindole-7-carboxamide inhibitors (3), which provides a
promising lead series for the development of additional
irreversible inhibitors having excellent potency and selectivity
combined with physicochemical properties better than those of
the carbazoles.
Examination of a previously reported4 X-ray structure of 4, a
close analog of 1, bound to BTK showed that the quinazolinone
moiety of this compound was positioned very close to Cys481
(Figure 1). Indeed, the thiol of Cys481 appears to be ideally
oriented to react with an electrophilic center near the space
occupied by C-5 of the quinazolinone, immediately suggesting
replacement of that ring system by an acrylamide.
It should be noted that the potency of a compound having a
BTK IC50 value significantly less than 1 nM may, in fact, be
underestimated, since this approaches the lower limit of the
enzyme assay.7 Therefore, such a compound may also be even
more selective relative to other kinases than the values indicated.
A.
Table 1. Analogs of 1 bearing electrophilic groups
Compd
R2'
CH3
CH3
CH3
CH3
H
R3’
BTKa
2.6
Ramosa
26
3-quinazolin-4-one
-NHCOCH=CH2
-NHCOCH=CH-CH3
1
5
6
7
8
9
0.46
1.2
7.1
68
b
-NHCOCH=C(CH3)2
-N(CH3)COCH=CH2
-NHCOCH2-CH3
2.4
75
0.16
7.3
23
CH3
59
a IC50, nM.7 b trans stereochemistry.
In light of the known SAR for this series,3,4 the increase in
potency for compounds bearing smaller, electrophilic groups
over the optimized quinazolinone is suggestive of, but does not
prove, irreversible inhibition. Therefore, a dialysis assay was
used to compare the rate of dissociation from BTK of 8 with that
of the sterically similar 9.8 After incubation of 9 (at a
concentration 25-fold higher than the IC50) with BTK for 90
minutes, followed by two 6-hour periods of dialysis against assay
buffer, all of the original BTK activity was recovered, while the
same treatment with 8 at the same relative concentration yielded
essentially inactive BTK (101% activity and 3.5% activity
recovered for 9 and 8, respectively). This demonstrated that 8
does indeed bind essentially irreversibly with BTK, supporting
this as the basis for the increased potency.
B.
Figure 1. (A) X-ray crystal structure of compound 4 bound to
the BTK kinase domain. (PDB code 5JRS).4 (B) Structure of 4
and replacement of the quinazolinone by acrylamide.
The tetrahydrocarbazole series (represented by 2) was
previously shown to improve potency and selectivity over the
fully aromatic series, and also provided improved oral exposure.5
This is in line with current thought that reductions in aromatic
ring count and planarity are desirable tactics in drug discovery.9
We therefore prepared irreversible analogs (in racemic form) in
this partially saturated series as well, with results shown in Table
2.
As in the carbazole series, replacing the quinazolinone of 10
with acrylamide (11) increased the potency significantly both in
the enzyme and cell assays to an even greater extent than was
seen in the fully aromatic series (44- and 26-fold, respectively).
Likewise, dramatic increases in selectivity with respect to LCK
The resulting analog (5) was readily prepared, and comparison
with 1 supported this hypothesis. As shown in Table 1, 5 gave a
nearly 6-fold increase in potency over 1 against recombinant
human BTK, and an almost 4-fold improvement in potency in a
Ramos cell-based assay measuring inhibition of BTK-induced
calcium flux.7 Although selectivity for BTK over other TEC
family kinases, which also have Cys481, was not significantly
changed or even reduced, selectivities over LCK and JAK2, two
other kinases involved in immune cell function which lack
Cys481, were increased by 4.6- and 4-fold, respectively, relative
to 1. (Data not shown.)

(from 98-fold to >2000-fold) and JAK2 (from 610-fold to
15,000-fold) were observed, although again the selectivity
differences in the other TEC family kinases were relatively small
(see below). Both -methylation (12) and -methylation (13) of
the olefin reduced potency and selectivity back to the levels of
the quinazolinone, the former possibly through steric hindrance
We first prepared examples bearing the quinazolinone moiety
(Table 3), to establish baseline potency relative to an
unsubstituted carbazole (22), itself significantly less potent than 1
due to removal of the tertiary carbinol moiety. The unsubstituted
indole 23 lost over 30-fold in potency, while adding one methyl
substituent at either R2 (24) or R3 (25) regained some potency
(only 5- to 6-fold reduced from 22). Gratifyingly, the 2,3-
dimethyl analog 26 regained the enzyme potency of the
carbazole, and showed a 2-fold increase in cell potency over 22.
to conjugate addition and the latter perhaps through
a
conformational effect. Interestingly, substitution of 12 with a
dimethylamino group (14), designed to activate the thiol of
Cys481 by deprotonation,2 regained much of the enzyme
potency. However, cell potency was reduced, presumably due to
poorer cell penetration by this charged species. Removal of the
methyl R2' substituent (15) had little effect, nor did moving the
methyl onto the acrylamide nitrogen (16), although this
compound did show reduced cell potency. The N, 2'-dimethylated
compound 17 also lost some potency, both against the enzyme
and in the cell assay.
Table 3. Reversible indole inhibitors of BTK: 2- and 3-
substituent variation
Compd
22
R2
R3
BTKa
16
Ramosa
1700
nt b
-CH=CH-CH=CH-
H
CH3
H
H
H
550
81
23
>2000
>2000
890
24
CH3
CH3
92
25
CH3
12
26
Table 2. Tetrahydrocarbazoles bearing electrophilic groups
a IC50, nM.7 b Not tested.
Compd
10
R2'
CH3
CH3
CH3
CH3
CH3
H
R3’
BTKa
9.3
Ramosa
120
4.6
3-quinazolin-4-one
-NHCOCH=CH2
-NHCOCH=CH-CH3
As shown in Table 4, replacing the quinazolinone of 26 with
acrylamide (27) provided an expected boost in potency (32-fold),
in line with the results seen in the tetrahydrocarbazole series.
Although removal of both indole methyl substituents, or of the 2-
methyl, did reduce potency, the loss was much less severe than in
the case of the quinazolinones shown in Table 3. The 2-methyl
analog 29 was as potent as the dimethyl analog against the
enzyme, although cell potency dropped by 3- to 4-fold. While the
3-methyl analog 30 was similar to the unsubstituted analog 28
against the enzyme, it was similar to 27 in cells. All four
compounds showed excellent selectivity against JAK2, LCK
(>1500-fold) and ITK (>200-fold).
0.21
12
11
a
370
290
1000
5.1
12
-NHCOC(CH3)=CH2
-NHCOCH=CH-CH2N(CH3)2
-NHCOCH=CH2
9.1
13
0.82
0.17
0.24
1.1
14
15
H
-N(CH3)COCH=CH2
-N(CH3)COCH=CH2
-NHSO2CH=CH2
28
16
CH3
CH3
H
15
nt c
17
0.079
0.18
3.5
18
-N(CH3)SO2CH=CH2
-CH2-NHCOCH=CH2
-C(CH3)2-NHCOCH=CH2
8.8
19
H
200
2000
20
H
36
21
a IC50, nM.7 b trans stereochemistry. c Not tested.
Replacement of the acrylamide with vinyl sulfonamide, as an
alternative electrophilic group, gave 18 which showed an even
greater increase in potency against BTK relative to 10. However,
this and other vinyl sulfonamides prepared (see below) appeared
to be somewhat unstable, showing significant decomposition on
storage for several days at room temperature. N-Methylation to
give 19 provided a compound comparable to the N-unsubstituted
acrylamide.
Table 4 Irreversible indole inhibitors of BTK
Two compounds were prepared with a one carbon spacer
between the aromatic ring and the acrylamide moiety (20 and
21), but both caused loss of potency, dramatically in the gem-
dimethyl substituted case. The ideal placement of the electrophile
for reaction with Cys481 thus appeared to be that of 11 or 15.
In addition to improving selectivity as well as potency (by 45-
fold for 11 compared to 10), replacing the quinazolinone
substituent by acrylamide also reduced the molecular weight and
heavy atom count by about 15%, and lowered the calculated log
P by 0.7 units. Many literature reports have promoted the
desirability of reducing these characteristics of drug candidates to
improve solubility and oral bioavailability.10 Since the tetrahydro
ring of that series extends toward solvent,5 we explored the
possibility of further reducing the size and lipophilicity of these
molecules by excising the ring completely to give indole-7-
carboxamides, relying on the added potency offered by
irreversibility to offset any lost binding interactions.
Compd
27
R2, R3
CH3, CH3
H, H
R2'
CH3
CH3
CH3
CH3
CH3
CH3
CH3
H
R3’
BTKa
0.38
1.1
Ramosa
9.2
57
-NHCOCH=CH2
-NHCOCH=CH2
-NHCOCH=CH2
-NHCOCH=CH2
-NHSO2CH=CH2
-N(CH3)COCH=CH2
-N(CH3)SO2CH=CH2
-NHCOCH=CH2
-NHSO2CH=CH2
-N(CH3)COCH=CH2
-N(CH3)SO2CH=CH2
28
CH3, H
0.32
1.2
34
29
H, CH3
15
30
CH3, CH3
CH3, CH3
CH3, CH3
CH3, CH3
CH3, CH3
CH3, CH3
CH3, CH3
0.14
0.59
0.11
0.21
0.077
0.23
0.068
23
31
61
32
26
33
16
34
H
6.4
170
63
35
H
36
H
37
a IC50, nM.7
Replacement of the acrylamide by a vinyl sulfone (31) gave a
slight increase in enzyme potency, as was seen in the

tetrahydrocarbazole series, but this was accompanied by a slight
decrease in cell potency. N-Methylation (32, 33) gave very little
change in potency, although a significant drop in cell potency
was seen in the case of the acrylamide. Removal of the methyl at
C-2' (34 – 37) caused two-fold increase in enzyme potency, but a
two- to three-fold decrease in cell potency except in the case of
the vinyl sulfone 35. The loss of cell potency for 34, 36 and 37
relative to the methylated analogs may reflect poorer cell
penetration, perhaps due to decreased shielding of the polarity
and hydrogen-bonding ability of the amide moiety.
Selectivity for BTK over other kinases is summarized in Table
5 for three irreversible compounds (11, a tetrahydrocarbazole,
and two indoles, 27 and 31), along with the tetrahydrocarbazole
quinazolinone 10 for comparison. Selectivity over LCK, which is
involved in leukocyte function, is greatly increased for all three
compounds relative to 10, dramatically so for the indole vinyl
sulfone. Compound 10 is already very selective with respect to
JAK2, but this selectivity is greatly improved (5000- to 16,000-
fold) for the three irreversible compounds. ITK is also involved
in leukocyte function and, like BTK, is a member of the TEC
family. Despite also having a cysteine homologous to Cys481 in
the ATP binding site, selectivity over ITK is improved in the
irreversible compounds relative to that seen with 2, especially for
the indole acrylamide 27. Differences in selectivity are less
profound and more varied for the remaining TEC family kinases
(BMX, TEC and TXK). Selectivities for two other kinases
having homologous cysteines are also shown. Compound 10 was
already very selective for JAK3; the irreversible compounds were
still selective but less so. Against HER4, the acrylamides showed
good selectivity, while the vinyl sulfonamide was about 10-fold
less selective.
these molecules, particularly in the indole series, combines with
the increased potency (seen despite the loss of the tertiary
carbinol in 27 and 31) to give significant increases in ligand
efficiency.
Unfortunately, compared to the quinazolinone-based
reversible inhibitors, the dimethylindole acrylamides and vinyl
sulfones showed reduced metabolic stability on incubation with
microsomes from human and (especially) mouse hepatocytes.
Because of this, although in vivo studies were not performed on
the irreversible compounds described here, their pharmacokinetic
profiles are expected to be poor. It should be noted, though, that
irreversible inhibitors are expected to require less prolonged and
elevated plasma levels than reversible inhibitors, since any
inhibitor bound to BTK would equilibrate only very slowly, if at
all, with plasma, leading to a favorable disconnect between
pharmacokinetics and pharmacodynamics. Further SAR, as well
as study and optimization of the ADME properties and efficacy
of this new class of inhibitors, will be reported elsewhere.
Table 6. Physicochemical properties and metabolic stability for
three irreversible BTK inhibitors compared to reversible inhibitor
10
Property
MWa
10
506.60
38
11
431.53
32
27
347.42
26
31
383.47
27
HACb
HBD/Ac
PSAd
3/4
4/3
3/2
3/3
111.78
4.61
0.19
84
108.21
3.91
0.29
90
87.98
3.76
0.35
71
105.05
2.85
0.36
71
clogPe
LEf
hLMg
mLMg
86
100
1.6
12
amolecular weight. bheavy atom count. cH-bond donors/acceptors.
dPolar surface area. eCalculated log P. fLigand efficiency.
GPercent remaining after 15 min incubation with human or mouse
liver microsomes.
Table 5. Kinase selectivity for three irreversible BTK inhibitors
compared to reversible inhibitor 10a,b
10
11
27
31
(THC,
reversible)
(THC,
(indole,
(indole,
Kinase
BTK
irreversible) irreversible) irreversible)
9.3
0.21
0.38
0.14
LCK
910
(98)
5700
(610)
290
(31)
200
(22)
56
420
>2000
(>5300)
>2000
(>5300)
88
(230)
13
(34)
>2000
(>14000)
1100
(7900)
8.3
(59)
3.3
(24)
0.9
(2000)
3300
(16,000)
23
(110)
5.2
JAK2
ITK
BMX
TEC
(25)
2.3
6.4
(6.0)
nt
(11)
15
(71)
41
(200)
150
(17)
26
(68)
28
(74)
240
(6.4)
2.3
(16)
11
(79)
10
TXK
JAK3
HER4
15,000
(1600)
nt
(710)
(630)
(71)
a
IC50 (nM). Numbers in parentheses are selectivities for BTK
over the indicated kinase. nt
b
=
not tested.
THC =
tetrahydrocarbazole
Some physicochemical properties for the same four
compounds are compared in Table 6. As mentioned previously,
replacement of the quinazolinone of 10 with the acrylamide of 11
decreases molecular weight, heavy atom count, polar surface area
and calculated log P, clearly improving these parameters of
“drug-likeness”.10 The trend obviously continues in moving from
11 to the dimethylindole 27, along with further improvement in
hydrogen bond donor/acceptor count and polar surface area
through removal of the tertiary carbinol. The smaller sizes of
Scheme 1. Reagents and conditions: (i) bis-pinacol diborane, KOAc,
PdCl2(dppf)2, dioxane, 110 oC, 16 h (50-70%). (ii) Pd(PPh3)4, R4Br,
Na2CO3 or K2CO3, toluene/EtOH, 90 °C, 16
h (19-65%). (iii)
ClC(=O)C(R)=C(R2)R1 or ClSO2CH2CH2Cl, Hünig’s base, CH2Cl2, rt,
16 h (17-94%). (iv) -78 °C to rt, THF, 4 h (37-45%). (v) SEM-Cl, NaH,
THF, 0 °C to rt, 3 h (82-95% crude). (vi) n-BuLi, THF, -78 °C, 10 min;
then CO2, -78 °C to rt, 4 h (93-95% crude). (vii) NH4OH, HOBT, EDC,
THF/DCM, rt, 16 h (36-83%). (viii) TBAF, ethylenediamine, THF, 50
°C, 1-3 days (49-74%). (ix) HOAc, 110 °C, 16 h (50-80%).

9.
(a) Ritchie TR, Macdonald SJF. Drug Discov. Today 2009, 14:1011-
1120. (b) Young RJ, Green DVS, Luscombe CN, Hill AP. Drug Discov.
Today 2011,16:822-830.
Synthesis of the compounds described generally followed
chemistry disclosed previously.4,5,8,11 (Scheme 1). Carbazole,
tetrahydrocarbazole or indole derivatives 38 bearing a bromide
para to the primary carboxamide were converted to boronate
esters 39, which underwent Suzuki coupling with an appropriate
bromobenzene derivative bearing the desired electrophilic moiety
to give the desired compounds. (Alternatively, the boronate and
bromide coupling partners could be exchanged: a 4-bromoindole
10. (a) Sutherland JJ, Raymond JW, Stevens JL et al. J. Med. Chem. 2012,
55:6455-6466. (b) Tarcsay A, Keseru GM. J. Med. Chem. 2013,
56:1789-1795.
11. Ahmad S, Tino JA, Macor JE et al. PCT WO2016065226A1, 28 Apr
2016.
12. (a) Dobbs AP, Voyle M, Whittall N. Synlett 1999, 1594-1596. (b)
Dobbs A. J. Org. Chem. 2001, 66:638-641.
13. Li L, Martins A. Tetrahedron Lett. 2003, 44:5987-5990.
14. Kamata J, Okada T, Kotake Y et al. Chem. Pharm. Bull. 2004,
52:p1071-1081.
or 4-bromocarbazole derivative could be reacted with
a
phenylboronate substituted with the desired electrophilic moiety,
prepared from the corresponding bromobenzene derivative.)
Bromides 41 and 42 were simply prepared by acylation of the
appropriate bromoaniline 40 with acryloyl chloride or a
substituted derivative, or in the case of vinyl sulfonamides 42
with 2-chloroethylsulfonyl chloride. In the latter case,
elimination of HCl during the sulfonylation reaction gave the
vinyl sulfonamide directly. 4-Bromoindole carboxamides 46
were prepared by two different routes. In the cases where one or
both indole substituents was H, 2,5-dibromonitrobenzene 43 was
treated with an appropriate vinylic Grignard reagent (the Bartoli
indole synthesis) to give the dibromoindole 44.12 After protection
as the SEM derivative, lithium-halogen exchange occurred
selectively at the 7-position,13 and treatment with carbon dioxide
gave the corresponding acid 45. Conversion to the primary
carboxamide followed by deprotection provided 46.
Alternatively, in the case of the 2,3-dimethylindole core, the
known carboxyhydrazine 47 underwent condensation with 2-
butanone (the Fischer indole synthesis14) to provide the acid 48,
which was then converted to the carboxamide 46.
In summary, two series of potent reversible BTK inhibitors
based on carbazole-1-carboxamide or tetrahydrocarbazole-1-
carboxamide were converted to even more potent and selective
irreversible inhibitors by replacing the previously optimized
substituents at C-3' of the pendent 4-phenyl group with an
electrophile designed to alkylate Cys481 of the enzyme.
Acrylamides and vinyl sulfonamides displayed sub-nanomolar
potency against isolated BTK enzyme, and were also very potent
in a cell-based calcium flux assay. A further significant reduction
in molecular weight and lipophilicity, achieved by removing one
ring of the carbazole core to provide the corresponding indole,
retained the excellent potency and increased selectivity against
some kinases, particularly LCK and ITK. The compounds
reported here provide good starting points for further SAR study
as well as optimization of potency, selectivity, and oral
bioavailability and efficacy.
References
1.
2.
3.
(a) Lou Y, Owens TD, Kuglstatter A. J. Med. Chem. 2012, 55:4539-
4550. (b) Norman P. Exp. Opin. Invest. Drugs 2016, 25:891-899. (c)
Thompson PA, Burger JA. Exp. Opin. Invest. Drugs 2018, 27:31-42.
(a) Gayko U, Fung M, Clow F et al. Ann. N. Y. Acad. Sci. 2015,
1358:82-94. (b) Roskoski R. Pharmacol. Res. 2016, 113A:395-408. (c)
Deeks ED. Drugs 2017, 77:225-236.
Liu Q, Batt DG, Lippy JS et al. Bioorg. Med. Chem. Lett. 2015,
25:4265-4269.
4.
5.
De Lucca GV, Shi Q, Liu Q et al. J. Med. Chem. 2016, 59:7915-7935.
Watterson S, De Lucca GV, Shi Q et al. J. Med. Chem. 2016, 59:9173-
9200.
6.
(a) Liu Q, Sabnis Y, Zhao Z et al. Chem. Biol. 2013, 20:146-159. (b) De
Cesco S, Kurian J, Dufresne C et al. Eur. J. Med. Chem. 2017, 138:96-
114. (c) Chaikuad A, Koch P, Laufer SA, Knapp S. Angewandte Chem.
2018, 57:2-16.
7.
8.
For details of BTK enzyme and Ramos cell assays, see Liu Q, Batt DG,
De Lucca GV et al. U. S. Patent 8,084,620, 27 Dec 2011, columns 33
and 36.
For details of the dialysis assay, see Ahmad S, Batt DG, Liu Q et al.
PCT WO2016065236A1, 28 Apr 2016, pages 227-228.

Graphical Abstract:

Highlights:
Electrophilic groups increased potency and selectivity of carbazole BTK inhibitors.
Related indole analogs retained this potency and further increased selectivity.
Irreversible inhibition was demonstrated using a dialysis assay.
Better physicochemical characteristics accompanied the improved biochemical profile.