Cys909 in JAK3 were strongly inhibited with less than 10% activity remaining.
More detailed kinetic investigations were performed and demonstrated moderate
selectivity for JAK3 (k inact /K I ¼ 3.7 Â 10
5 M
À1 s
À1 ) over TEC family
kinases with an equivalently positioned cysteine (BMX, BLK, BTK, ITC, TEC,
TXK; k inact /K I ranging from 1.0 Â 10
4 to 8.6 Â 10
2 M
À1 s
À1 ). Pronounced selectivity
over the kinase TXK (k inact /K I ¼ 2.3 Â 10
3 M
À1 s
À1 ) was observed in kinetic
analyses although the IC 50 values were in the same range for both enzymes
highlighting that IC 50 data of covalent inhibitors should always be interpreted with
caution. Remarkably, selectivity over TEC family kinases is caused by the more
efficient covalent inactivation of JAK3 (k inact ¼ 2.3 s
À1 ), while K I values were
significantly lower for the mentioned TEC kinases.
In a comprehensive set of cellular assays, compound 56 was able to suppress
STAT phosphorylation only in the JAK3-dependent pathways triggered by IL-2,
IL-4, IL-7, IL-15, and IL-21 with IC 50 values in the range of 200–400 nM, while no
inhibition of downstream STAT phosphorylation was observed after stimulation
with IL-6, IL-10, IL-12, IL-23, IL-27, IFN-α, IFN-γ G-CSF, and EPO.
A comparison of 56, pan-JAK inhibitor tofacitinib, and JAK1-selective inhibitor
PF-02384554 in a human whole blood assay showed that tofacitinib was able to
inhibit the STAT phosphorylation after stimulation with IL-15 and IL-21 slightly
more efficient (fivefold to sevenfold) than 56. The latter, however, was equipotent to
JAK1-selective compound PF-02384554 in the same assay format. These results
suggest once more that a dominant role of JAK1 over JAK3 should be considered
unlikely.
The successful covalent targeting of Cys909 by 56 was also confirmed by X-ray
crystallography, and the ligand adopted the predicted binding mode. The heterocyclic core is directed to the hinge region where it addresses the protein backbone via
the two typical hydrogen bonds. Meanwhile, the piperidinyl side chain orients
toward Cys909 in the proposed “anti-conformation” allowing the acrylamide to
react with the sulfhydryl group. It is noteworthy that all hydrogen bond donors
and acceptors of 56 are engaged in a hydrogen bonding network built up from the
ligand, the protein backbone, and water bridging molecules (Fig. 21b).
Unspecific protein binding of 56 was shown to be very low as demonstrated using
human serum albumin (HSA) as a surrogate protein and in functional human
hepatocytes. In the latter case, more than 98% of radiolabeled 56 remained unbound
after incubation. Moreover, 56 was tested in several in vivo models and showed
good pharmacokinetic properties although a certain interspecies variability was
observed. In an adjuvant-induced arthritis (AIA) rat model, 56 was able to decrease
paw swelling in a dose-dependent manner after oral administration (3–30 mg/kg, p.
o., QD).
At the moment, 56 is being investigated in several phase II clinical trials
for the treatment of rheumatoid arthritis (NCT02969044), ulcerative colitis
(NCT02958865), and alopecia areata (NCT02974868).
A compound denominated III-4 (59, Fig. 22), representing a close structural
analog of tofacitinib and PF-06651600, has very recently been characterized as
a covalent JAK3 inhibitor by Chen an co-workers [58]. Analogous
Covalent Janus Kinase 3 Inhibitors
245
More detailed kinetic investigations were performed and demonstrated moderate
selectivity for JAK3 (k inact /K I ¼ 3.7 Â 10
5 M
À1 s
À1 ) over TEC family
kinases with an equivalently positioned cysteine (BMX, BLK, BTK, ITC, TEC,
TXK; k inact /K I ranging from 1.0 Â 10
4 to 8.6 Â 10
2 M
À1 s
À1 ). Pronounced selectivity
over the kinase TXK (k inact /K I ¼ 2.3 Â 10
3 M
À1 s
À1 ) was observed in kinetic
analyses although the IC 50 values were in the same range for both enzymes
highlighting that IC 50 data of covalent inhibitors should always be interpreted with
caution. Remarkably, selectivity over TEC family kinases is caused by the more
efficient covalent inactivation of JAK3 (k inact ¼ 2.3 s
À1 ), while K I values were
significantly lower for the mentioned TEC kinases.
In a comprehensive set of cellular assays, compound 56 was able to suppress
STAT phosphorylation only in the JAK3-dependent pathways triggered by IL-2,
IL-4, IL-7, IL-15, and IL-21 with IC 50 values in the range of 200–400 nM, while no
inhibition of downstream STAT phosphorylation was observed after stimulation
with IL-6, IL-10, IL-12, IL-23, IL-27, IFN-α, IFN-γ G-CSF, and EPO.
A comparison of 56, pan-JAK inhibitor tofacitinib, and JAK1-selective inhibitor
PF-02384554 in a human whole blood assay showed that tofacitinib was able to
inhibit the STAT phosphorylation after stimulation with IL-15 and IL-21 slightly
more efficient (fivefold to sevenfold) than 56. The latter, however, was equipotent to
JAK1-selective compound PF-02384554 in the same assay format. These results
suggest once more that a dominant role of JAK1 over JAK3 should be considered
unlikely.
The successful covalent targeting of Cys909 by 56 was also confirmed by X-ray
crystallography, and the ligand adopted the predicted binding mode. The heterocyclic core is directed to the hinge region where it addresses the protein backbone via
the two typical hydrogen bonds. Meanwhile, the piperidinyl side chain orients
toward Cys909 in the proposed “anti-conformation” allowing the acrylamide to
react with the sulfhydryl group. It is noteworthy that all hydrogen bond donors
and acceptors of 56 are engaged in a hydrogen bonding network built up from the
ligand, the protein backbone, and water bridging molecules (Fig. 21b).
Unspecific protein binding of 56 was shown to be very low as demonstrated using
human serum albumin (HSA) as a surrogate protein and in functional human
hepatocytes. In the latter case, more than 98% of radiolabeled 56 remained unbound
after incubation. Moreover, 56 was tested in several in vivo models and showed
good pharmacokinetic properties although a certain interspecies variability was
observed. In an adjuvant-induced arthritis (AIA) rat model, 56 was able to decrease
paw swelling in a dose-dependent manner after oral administration (3–30 mg/kg, p.
o., QD).
At the moment, 56 is being investigated in several phase II clinical trials
for the treatment of rheumatoid arthritis (NCT02969044), ulcerative colitis
(NCT02958865), and alopecia areata (NCT02974868).
A compound denominated III-4 (59, Fig. 22), representing a close structural
analog of tofacitinib and PF-06651600, has very recently been characterized as
a covalent JAK3 inhibitor by Chen an co-workers [58]. Analogous
Covalent Janus Kinase 3 Inhibitors
245