Janus kinase (JAK) inhibitors in the treatment of inflammatory and neoplastic diseases

Robert Roskoski Jr.

Blue Ridge Institute for Medical Research

3754 Brevard Road, Suite 116, Box 19

Horse Shoe, North Carolina 28742-8814, United States

Phone: 1-828-891-5637

Fax: 1-828-890-8130

E-mail address: [email protected]

Key words: ATP-binding site; Catalytic spine; K/E/D/D; Regulatory spine; Ruxolitinib; Tofacitinib

Chemical compounds studied in this article:

Tofacitinib (PubMed CID: 9926791), Oclacitinib (PubMed CID: 44631938), Baricitinib (PubMed CID: 44205240), Ruxolitinib (PubMed CID: 25126798), Decernotinib (PubMed CID: 59422203), Peficitinib (PubMed CID: 57928403), Filgotinib (PubMed CID: 49831257), Fedratinib (PubMed CID: 16722836), Momelitinib (PubMed CID: 25062766), Lestaurtinib (PubMed CID: 126565)

Abbreviations: ALL, acute lymphocytic leukemia; AML, acute myelogenous leukemia; AS, activation segment; CS or C-spine, catalytic spine; CL, catalytic loop; HΦ or Φ, hydrophobic; IFN, interferon; IL, interleukin; JAK1/2, Janus kinases 1 and 2; MW, molecular weight; NSCLC, non-small cell lung cancer; PKA, protein kinase A; RA,

rheumatoid arthritis; RS or R-spine, regulatory spine; Sh1, shell residue 1; TNF, tumor necrosis factor.

Graphical Abstract

ABSTRACT

The Janus kinase (JAK) family of non-receptor protein-tyrosine kinases consists of JAK1, JAK2, JAK3, and TYK2 (tyrosine kinase-2). Each of these proteins contains a JAK homology pseudokinase (JH2) domain that regulates the adjacent protein kinase domain (JH1). JAK1/2 and TYK2 are ubiquitously expressed whereas JAK3 is found predominantly in hematopoietic cells. The Janus kinase family is regulated by numerous cytokines including interleukins, interferons, and hormones such as erythropoietin,
thrombopoietin, and growth hormone. Ligand binding to cytokine and hormone receptors leads to the activation of associated Janus kinases, which then mediate the phosphorylation of the receptors. The SH2 domain of STATs (signal transducers and

activators of transcription) binds to the receptor phosphotyrosines thereby promoting STAT phosphorylation by the Janus kinases and consequent activation. STAT dimers are translocated to the nucleus where they participate in the regulation of the expression of thousands of proteins. JAK-STAT dysregulation results in autoimmune disorders such as rheumatoid arthritis, ulcerative colitis, and Crohn disease. JAK-STAT dysregulation also plays a role in the pathogenesis of myelofibrosis, polycythemia vera, and other myeloproliferative illnesses. An activating JAK2 V617F mutation occurs in 95% of people with polycythemia vera and in a lower percentage of people with other neoplasms. JAK1/3 signaling participates in the pathogenesis of inflammatory afflictions while JAK1/2 signaling participates in the development of several malignancies including leukemias and lymphomas as well as myeloproliferative neoplasms. Tofacitinib is a pan- JAK inhibitor that is approved by the FDA for the treatment of rheumatoid arthritis and ruxolitinib is a JAK1/2 inhibitor that is approved for the treatment of polycythemia vera and myelofibrosis.
Contents

1.Introduction to JAK-STAT signaling

2.Janus kinase biochemistry

2.1Janus kinase architecture

2.2Active JH1 and inactive JH2 hydrophobic spines

2.3Mechanism of Janus kinase activation

3.Selected Janus kinase disease targets

3.1Inflammatory disorders

3.1.1Rheumatoid arthritis

3.1.2Canine dermatitis

3.1.3Psoriasis

3.1.4Ulcerative colitis

3.1.5Crohn disease

3.2Neoplastic disorders

3.2.1Myelofibrosis

3.2.2Polycythemia vera

3.2.3Primary thrombocythemia

4.Janus kinase mutations in human illnesses

5.Selected Janus kinase inhibitors in clinical trials

5.1First-generation drugs

5.1.1Tofacitinib

5.1.2Oclacitinib

5.1.3Baricitinib

5.1.4Ruxolitinib

5.2Second-generation drugs

5.2.1Decernotinib

5.2.2Peficitinib

5.2.3Filgotinib

5.2.4Fedratinib

5.2.5Momelotinib

5.2.6Lestaurtinib

5.2.7Gandotinib

5.2.8Pacritinib

6.Non-conserved amino acid residues near the Janus kinase ATP-binding sites

7.Epilogue Conflict of interest Acknowledgement References

1.Introduction to JAK-STAT signaling

The Janus kinase (JAK) family of protein-tyrosine kinases is made up of four members: JAK1, JAK2, JAK3, and TYK2 (Tyrosine kinase 2) [1]. The Janus kinases share seven distinct JAK homology (JH1-JH7) domains. These proteins possess an inactive pseudokinase domain (JH2) adjacent to an active carboxyterminal protein kinase domain (JH1). The pseudokinase domain ordinarily inhibits the functional protein kinase domain. Janus is a two-faced (looking forwards and backwards) Roman God whose name was applied to this enzyme family owing to the presence of the two protein kinase domains within a single polypeptide chain. JAK was earlier conceived as Just Another Kinase [2]. JAK1/2 and TYK2 are ubiquitously expressed whereas JAK3 is confined to hematopoietic, myeloid, and lymphoid cells [3]. Mature blood cells have a limited life span and are thus continuously renewed in an intricate multi-step process. The Janus kinases play an important role in normal hematopoiesis; accordingly, Janus kinase dysregulation can result in a variety of hematological illnesses. These enzymes also function in a wide variety of processes including post-natal growth, metabolism, and satiety.

Manning et al. identified 478 classical and 40 non-classical or atypical human protein kinase family genes (total 518) that correspond to nearly 2% of the human genome [4]. Based upon the identity of the phosphorylated –OH group, these enzymes are catalogued as protein-serine/threonine kinases (385 members), protein-tyrosine kinases (90), or tyrosine-kinase like proteins (43). A small group of proteins including MEK1 and MEK2, which catalyze the phosphorylation of both threonine and tyrosine residues on target proteins, are catalogued as dual specificity kinases. Additionally, there are 106 protein kinases existing as pseudogenes within the human genome. Of the 90 protein-tyrosine kinases, 58 are transmembrane receptors and 32 are cytosolic non- receptor kinases, including the Janus kinase family. This enzyme family catalyzes the following reaction:
MgATP1– + protein-tyrosine–O:H  protein-tyrosine–O:PO32– + MgADP + H+ Note that it is not the phosphate (OPO32–) group, but rather it is the phosphoryl group (PO32–) that is transferred from ATP to a tyrosine residue within the protein substrate. Divalent cations such as Mg2+ or Mn2+ are required for the reaction, although Mg2+ is the physiological ion owing to its significantly greater intracellular content.
The JAK-STAT (signal transducer and activator of transcription) pathway transduces extracellular signals from a variety of cytokines, growth factors, and hormones to the nucleus and is responsible for the expression of thousands of protein- encoding genes [5]. Each of the Janus kinase proteins binds to the juxtamembrane region of specific cytokine receptors. The genes specifying each high-affinity cytokine receptor homo- or hetero-dimer are listed in Table 1. Receptors for Type I cytokines share a common amino acid motif (WSxWS) in the extracellular portion adjacent to the cell

membrane. Receptors for Type II cytokines are similar to those for type I cytokines except that these receptors lack the WSxWS signature. The cytokine receptors consist of an extracellular domain, a transmembrane segment, and an intracellular domain that lacks catalytic activity, but which binds to specific Janus kinases.
Several steps are involved in the conversion of an extracellular signal into a transcriptional response. First, ligand binding mediates structural changes in the cytokine receptors that result in protein-tyrosine kinase activation following phosphorylation of two tyrosine residues within the activation segment of the JH1 domains as catalyzed by a partner Janus kinase enzyme. This intermolecular reaction is referred to as phosphorylation in trans; an intramolecular reaction is referred to as phosphorylation in cis. Following activation, the JH1 domain catalyzes the phosphorylation of tyrosine residues within the cytokine receptor that attracts the SH2 domain of STATs. The JH1 domain then catalyzes the phosphorylation of the STAT molecules themselves. Phosphorylated STATS form dimers that are translocated into the nucleus where they mediate the transcription of target genes (Fig. 1). Alternatively, STATs may preexist as dimers and phosphorylation may produce a conformational change resulting in activation [7].
Note that the Type I IL-2 or IL-4 receptors along with the common γ-chain bind to JAK1 and JAK3 while the Type I receptors with the common β-chain bind to JAK2 (Table 1). The Type I receptors using gp130 (glycoprotein 130) and those that are activated by heterodimeric cytokines preferentially bind to JAK1/2 and TYK2. The Type I receptors that are activated by the hormone-like cytokines bind to and activate JAK2. The Type II receptors for interferons bind to JAK1 and TYK2 whereas those for the IL-

10 family of cytokines bind to JAK1/2 and TYK2. Accordingly, specific Janus kinases bind to the appropriate receptor subunits thereby imparting signaling specificity. The IL- 10 receptor is a tetramer consisting of two IL10RA and two IL10RB encoded subunits while most of the other receptors consist of two subunits as listed in Table 1. Janus kinase activation by EGFR and PDGF is downstream from the activated receptor and may involve other protein kinases such as Src [8].
There are seven STAT genes in humans: STAT1, STAT2, STAT3, STAT4, STAT5A, STAT5B, and STAT6 (Table 2). Each STAT protein contains six domains that are illustrated in Fig. 2A including a DNA binding domain, an SH2 domain, and a transcriptional activation domain (TAD), first described for STAT1 [9]. The transcriptional activation domain contains a tyrosine residue that is phosphorylated in a reaction catalyzed by an upstream Janus kinase. Following phosphorylation, the STAT forms a homo- or heterodimer with another phosphorylated STAT that is mediated by the binding of a phosphotyrosine (pY) to its partner’s SH2 domain (Fig. 2B). The dimer travels to the nucleus where it binds to target DNA sequences alone or in combination with other transcription factors that either enhance or repress DNA transcription.
STAT1/2/4/6 have restricted roles in signaling. STAT1 participates in interferon, interleukin-9 (IL-9), thrombopoietin, and IL-10 family signaling while STAT2 is chiefly involved in interferon signaling (Table 1). STAT4 plays a role in signaling by the Type I heterodimeric cytokines (IL-12/23) and Type II interferon family cytokines (IFN-α/β, IL- 28/29). STAT6 participates in IL-3/4/5 and PDGF signaling. STAT3 and STAT5a/b have broader functions. STAT3 participates in the signaling initiated by many of the Type I, II, and IL-10 families of cytokines while STAT5a/b participate in signaling initiated by

Type I cytokines with the common γ-chain or common β-chain receptor subunits and the hormone-like cytokines. The information provided in Table 1 is qualitative in nature and the precise pathway from ligand and receptor to a specific Janus kinase and a specific STAT depends upon the cellular and physiological context. Moreover, more than two dozen cytokines including IL-8, transforming growth factor-β, and RANTES interact with receptors that do not signal directly through the JAK-STAT pathway [6].
The activity of all protein kinases is stringently regulated owing to their overall importance in multiple signaling pathways [10]. Various intracellular phosphoprotein phosphatases including SHP1, SHP2, PTP1B, and TCPTP mediate the dephosphorylation and inactivation of Janus kinases [7]. CD45 is a receptor phosphoprotein phosphatase that is able to catalyze the dephosphorylation of the activation segment phosphotyrosines of
all four Janus kinases [11]. The SOCS (suppressor of cytokine signaling) family of eight proteins also negatively regulates cytokine and Janus kinase signaling [7]. The SOCS proteins are ubiquitin ligases that promote the proteasomal degradation of activated Janus kinase-associated cytokine receptors. Moreover, SOCS1 and SOCS3 can bind to the protein-substrate binding grove and sterically inhibit the catalytic activity of JAK1/2 and TYK2, but not JAK3.
2.Janus kinase biochemistry

2.1Janus kinase family architecture

The Janus kinases contain about 1100 amino acid residues and they contain seven JAK homology modules: JH7-JH1 as we go from the N-terminus to the C-terminus. These have subsequently been reorganized into four functional domains [12]. The JH7- JH6 module corresponds to a FERM domain (F for 4.1 protein, E for ezrin, R for radixin

and M for moesin), which is a ubiquitous protein module of ≈ 350 amino-acids and is involved in targeting proteins to the plasma membrane. The JH5-JH3 module corresponds to an atypical SH2 domain that is unable to bind protein-tyrosine phosphate.
The JH3 domain is followed by a JH2 pseudokinase domain and the carboxyterminal JH1 segment represents the catalytically active protein-tyrosine kinase domain (Fig. 2C).
The Janus kinases contain a small amino-terminal lobe and large carboxyterminal lobe that contains several conserved α-helices and β-strands, first described by Knighton et al. for PKA in 1991 [13, 14]. The small lobe is dominated by a twisted five-stranded antiparallel β-sheet (β1–β5) [15]. It also contains an important regulatory αC-helix (Fig. 3A and B). The small lobe of JAK1 contains a conserved glycine-rich (GEGHFG) ATP- phosphate-binding loop, which occurs between the β1- and β2-strands. The canonical sequence of this loop in protein kinases is GxGxΦG, where x refers to any amino acid and Φ is a hydrophobic residue, usually phenylalanine or tyrosine. The β1 and β2-strands occur above the adenine component of ATP. The β3-strand typically contains an Ala- Xxx-Lys sequence, the lysine of which (JAK1 K908) couples the α- and β-phosphates of ATP to the αC-helix. A conserved glutamate occurs near the center of the αC-helix (JAK1 E925) in protein kinases. The presence of a salt-bridge between the β3-lysine and the αC-glutamate is a prerequisite for the formation of the activate state, which corresponds to an “αC-in” conformation (Fig. 3A). In contrast the Lys622 and Glu637 of the pseudokinase of the JAK1 JH2 domain fail to make contact, which corresponds to an “αC-out” conformation (Fig. 3C). Moreover, the G-loop in the JH2 domain consists of a shortened GRG sequence between the β1- and β2-strands. The αC-in conformation is necessary, but not sufficient, for the expression of full protein kinase activity.

The large lobe of the JAK1 protein kinase domain is mainly α-helical with six conserved segments (αD-αI) that occur in all protein kinases [15]. It also contains four short conserved β-strands (β6-β9). The primary structures of the β-strands occur between those of the αE- and αF-helices and include residues just before the catalytic loop (β6), after the catalytic loop and below the adenine ring of ATP (β7 and β8), and within the activation segment (β9). The activation segment of the JAK1 JH1 active state forms an open structure that allows protein/peptide binding (Fig. 3A). The activation segment in the dormant JH2 domain forms a closed, compact structure that blocks protein/peptide binding (Fig. 3C). The active JAK1 protein kinase domain and the pseudokinase domain contain an additional helix (αEF) near the end of the activation segment and this additional helix occurs in most protein kinases.
In an important analysis, Hanks and Hunter identified 12 subdomains (I–VIa, VIb–XI) with canonical sequences that make up the indispensable catalytic residues of protein kinases [16]. The following four amino acids, which are part of a K/E/D/D core, exemplify the mechanistic properties of JAK1. An invariant β3-strand lysine (K908, the K of K/E/D/D) forms ionic bonds with the αC-glutamate (E925, the E of K/E/D/D) and also the α- and β-phosphate groups of ATP (Fig. 3B). The catalytic loop surrounding the actual site of phosphoryl transfer consists of HRDLAARN in most protein-tyrosine kinases including JAK1/3 and TYK2; the sequence HRDLATRN occurs in JAK2. The JAK1 JH2 pseudokinase domain contains HGNVCTKN. The catalytic aspartate (D1003) in JAK1, which is the first D of K/E/D/D) serves as a base that accepts a proton from the tyrosyl –OH group (Fig. 4). This residue in the pseudokinase domains of the Janus kinases is an asparagine, which cannot function as a base thereby contributing to the

catalytic dormancy of the pseudokinase domains. The AAR sequence in the catalytic loop represents a receptor protein-tyrosine kinase signature and RAA represents a non- receptor protein-tyrosine kinase; the occurrence of AAR in the JAK non-receptor protein kinase is thus anomalous.
The second aspartate of the K/E/D/D signature of JAK1, D1021, is the first residue of the activation segment. The activation segment of nearly all protein kinases including those of the Janus kinases begins with DFG (Asp-Phe-Gly) and ends with APE (Ala-Pro-Glu). DFG-Asp1021 binds Mg2+(1), which in turn coordinates the α- β- and γ- phosphates of ATP. The HRDxxxxN (N1008) binds Mg2+(2), which in turn coordinates the α- and γ-phosphates of ATP (Fig. 3B). The exocyclic 6-amino nitrogen of ATP characteristically forms a hydrogen bond with the carbonyl backbone residue of the first hinge residue (E957 of JAK1 JH1 domain) that connects the small and large lobes of the protein kinase domain and the N1 nitrogen of the adenine base forms a second hydrogen bond with the N–H group of third hinge residue (L595) (Fig. 3B). As discussed later,
most small-molecule inhibitors of the Janus kinases that are steady-state ATP competitive inhibitors also make hydrogen bonds with the backbone residues of the connecting hinge. Fig. 3D is the superposition of the active JAK1 JH1 domain and the dormant JAK1 JH2 domain, which depicts the relative positions of the αC-helix and the activation segments. The large lobe characteristically binds the peptide/protein substrates near the end of the activation segment as shown in Fig. 4.
The activation segment influences both protein-substrate binding and catalytic efficiency [17]. This segment in Janus kinases contains two phosphorylatable tyrosines (Fig 3A). The segment is located close in the three-dimensional sense to the amino-

terminus of the αC-helix and the conserved HRD portion of the catalytic loop. The interaction of these components is hydrophobic in nature. As is the case for most protein kinases [18], phosphorylation of a residue within the activation segment (tyrosine for the Janus kinases) converts an inactive JAK JH1 domain to an active one [19]. The activation segment of the Janus kinases and other protein kinases is further stabilized by hydrophobic bonds between an amino acid two residues N-terminal to the HRD-H of the catalytic loop and seven residues C-terminal to the DFG-D within the activation segment. Functionally important human Janus kinase JH1 residues are listed in Table 3.
2.2Active JH1 and inactive JH2 hydrophobic spines

Kornev et al. analyzed the structures of active and inactive conformations of 23 protein kinases and they uncovered the role of important residues by a local spatial pattern alignment algorithm [20, 21]. Their analysis led to a classification of eight hydrophobic residues as a catalytic or C-spine and four hydrophobic residues that constitute a regulatory or R-spine. Each of the two spines consists of amino acids occurring in both the small N-terminal lobe and the large C-terminal lobe. The regulatory spine contains a residue from the αC-helix and from the activation segment, both of which are important in determining active and dormant states. The adenine portion of ATP is one component of the catalytic spine. The R-spine positions the protein substrate and the C-spine positions ATP thereby enabling catalysis. Note that the structure of the spines differs between the active JH1 and dormant JH2 domains. Moreover, the correct alignment of both spines is necessary for the assembly of an active protein kinase domain as described for the cyclin-dependent protein kinases, Src, EGFR, ERK1/2, and MEK1/2 [22–27].

The canonical protein kinase R-spine consists of the HRD-His of the catalytic loop, the phenylalanine of the activation segment DFG, a residue near the C-terminal end of the αC-helix (four residues carboxyterminal to the conserved αC-glutamate), and a residue at the beginning of the β4-strand. The backbone of the HRD-His is anchored to the very hydrophobic αF-helix by a hydrogen bond to a conserved aspartate residue. Going from the aspartate within the αF-helix up to the top residue of the spine within the β4-strand, Meharena et al. named the R-spine residues as follows: RS0, RS1, RS2, RS3, and RS4 (Fig. 5A and B) [17]. The regulatory spine of the active JH1 protein kinase domain is nearly linear while that of the JH2 pseudokinase domain is broken with RS3 displaced (Fig. 5B and C).
The protein kinase catalytic spines are made up of residues from both the small N- lobes and large C-lobes lobes; the C-spine is completed by the adenine base of ATP (Fig. 5A) [17, 21]. The two residues of the small lobe of protein kinase domains that bind to
the adenine component of ATP include a valine residue at the beginning of the β2-strand (CS7) and an alanine from the conserved AxK of the β3-strand (CS8). Additionally, a hydrophobic residue from the β7-strand (CS6) juxtaposes with the adenine base. This CS6 residue is flanked by two hydrophobic residues named CS4 and CS5 that bind to the CS3 residue near the beginning of the αD-helix. Finally, CS3 and CS4 interact with the CS1 and CS2 residues of the αF-helix to produce a completed C-spine (Fig. 5A and B) [27]. It is important to notice that the hydrophobic αF-helix anchors both the C-spine and R-spine. Moreover, the spines play an important function in positioning the protein kinase catalytic residues in their active state. When comparing the locations of the spinal

residues, the greatest divergence in the structures of the JH1 and JH2 domains involve RS3 and RS4 along with CS7 and CS8 (Fig. 5D).
Using site-directed mutagenesis, Meharena et al. identified three residues in murine PKA that stabilize the R-spine that they labeled Sh1, Sh2, and Sh3, where Sh refers to shell [17]. The Sh2 residue corresponds to the gatekeeper residue. The name gatekeeper signifies the role that this residue plays in controlling access to the back cleft. The back cleft is sometimes called the back pocket or hydrophobic pocket II (HPII). The residues that constitute the spines were identified by their positions in active and inactive enzymes based upon their X-ray crystallographic structures [20, 21]. This contrasts with the identification of the DFG, APE, or HRD amino acid signatures based upon their primary structures [16]. Table 4 provides a summary of the spine and shell residues of the JAK JH1 and JH2 domains. Small molecule protein kinase therapeutic antagonists generally interact with residues that make up the C-spine and sometimes the R-spine and shell residues [27].
2.3Mechanism of Janus kinase activation

Protein kinases are stringently regulated and have not evolved to continuously catalyze the phosphorylation of thousands of molecules per minute like hexokinase, a general metabolic enzyme. For example, when a Janus kinase is activated by its cytokine receptor, the chief phosphorylated product is the enzyme itself mediated by the trans- phosphorylation of one Janus kinase protein by another. Phosphorylation of the cytokine receptor and STATs also occurs. The concentration of protein kinases and their substrates are nearly equivalent and high turnover numbers are not required. The mechanisms for

the interconversion or switching of dormant and active protein kinases vary with the particular enzyme and are generally quite intricate as seen for the Janus kinases.
The interaction of a cytokine ligand with its receptor triggers a change in the receptor conformation that leads to a repositioning of the associated Janus kinase proteins. This rearrangement promotes the trans-phosphorylation of activation segment tyrosine residues. Feng et al. prepared Y1007F and Y1008F activation segment single and double mutants of JAK2 [28]. They found that the Y1007F and the double mutant were kinase impaired whereas the Y1008F mutant was active in vitro. They also found that the wild type and Y1008F mutants were active in erythropoietin signaling in human fibrosarcoma γ2 cells in culture whereas the Y1007F mutant was not. These studies
indicate that phosphorylation of only the first tyrosine residue in the activation segment is necessary for JAK2 enzyme activation.
Zhou et al. prepared Y980F and Y981F activation segment mutants of JAK3 [29]. They demonstrated that both of the activation segment tyrosines can be phosphorylated. They measured the protein kinase activity of various mutants in COS-7 cell immunoprecipitates expressing the proteins. They observed that the Y980F mutant exhibited about 30% of the activity as the wild-type enzyme. In contrast, the Y981F mutant had about 230% of the activity of the wild-type protein while the Y980F and Y981F double mutant had about the same activity as the wild-type enzyme. These results suggest that Y981 per se in JAK2 is somehow inhibitory.
Gauzzi et al. studied the role of the activation segment tyrosine residues on TYK2 activity in mutant human fibrosarcoma cell immunoprecipitates that do not express this protein [30]. They demonstrated that the Y1054F/Y1055F double mutant had decreased

catalytic activity when compared with the wild-type enzyme. Moreover, the extent of autophosphorylation of the double mutant was markedly decreased. These investigators did not perform experiments with Y1054F or Y1055F single mutants so that we cannot establish with certainty whether TYK2 behaves like JAK2 and JAK3, which indicate that phosphorylation of the first tyrosine residue of the activation segment results in an increase in catalytic activity.
Saharinen et al. reported that the activity of a construct containing only the JAK2 JH1 domain was 50 times more active in catalyzing the phosphorylation of an exogenous peptide substrate that a JH2-JH1 construct [31]. Thus, the JH2 domain inhibits the activity of the JH1 protein kinase domain. As noted later, an activating JAK2 V617F mutation occurs in several myeloproliferative neoplasms and this mutation occurs within the JH2 domain. Lupardus et al. determined the X-ray crystal structure of the TYK2 JH1- JH2 construct [32]. Each of their domains has the characteristic protein kinase fold (Fig. 6A). The JH1 domain has an active structure with the αC-in conformation and with the activation segment in its open conformation despite being non-phosphorylated. The JH2 domain has an inactive structure with the αC-out conformation and with a compact or closed activation segment.
The interaction between the TYK2 pseudokinase and kinase domains is mediated by the N-lobes of each domain (Fig. 6B). The pseudokinase interface surface is primarily composed of residues of the αC-β4 loop and the N-terminal exon 12 segment (TYK2 residues 579–591). Important contacts in the JH1 domain include I900 of the β1-strand, L913 of the β2-strand, M926 of the β3-strand, and Y980 within the hinge. The X-ray crystal structural analysis failed to identify the location of a 17-amino acid pseudokinase-

kinase interdomain linker. The JH2 inhibition of JH1 catalytic activity may result from the immobilization of the JH1 N-lobe. In this scheme, the blockade of catalytic activity involves an intraprotein or cis-inhibitory mechanism (Fig. 7A).
In contrast, Waters et al. proposed an interprotein or trans-inhibitory mechanism [33]. Accordingly, the activation segment of a pseudokinase interacts with the active site catalytic domains of a JH1 partner domain and thereby blocks its activity (Fig. 7B). Cytokine receptor activation induces a conformational shift that promotes the trans- phosphorylation of the activation segment of the Janus kinase partner leading to enzyme activation. In support of the trans mechanism, Lu et al. found that the constitutively activated V617F mutant of JAK2 does not manifest its activity unless the erythropoietin receptor is present [34]. Brooks et al. studied the activation of JAK2 by the growth hormone receptor using fluorescence resonance energy transfer (FRET) to monitor positioning of the JAK2 binding motif in the receptor dimer [35]. They reported that activation of the receptor dimer induced a separation of its JAK2 binding motifs, driven by a ligand-induced transition from a parallel transmembrane helix pair to a crossover arrangement. This separation leads to the removal of the pseudokinase domain from the kinase domain of the partner JAK2 and pairing of the two kinase domains, thereby facilitating trans-activation. In either the cis- or trans-mechanisms, immobilization of the small lobe by any means would decrease enzyme activity owing to the requirement that this lobe must be flexible in order to bind ATP, catalyze the reaction, and then release ADP.
Besides activation segment phosphorylation, other activating and inhibiting phosphorylation events participate in Janus kinase regulation. Little is known about the

phosphorylation of JAK1/3 and TYK2 outside of the activation segment. However, there is considerable data on the phosphorylation of JAK2. This protein is constitutively phosphorylated on Ser523 [36, 37]. Besides the phosphorylation of Tyr1007 and Tyr1008 within the activation segment, phosphorylation of the following tyrosines increases JAK2 protein kinase activity: Tyr637 (within the αD’ helix of JH2), Tyr813 (in the last α-helix of JH2), Tyr868 (at the end of the β2-strand of the JH1 domain), and Tyr966 (near the
end of the αE helix, and Tyr972 (before the catalytic loop) [38, 39] In contrast, phosphorylation of Ser523 (within the SH2 domain) and tyrosines 119, 221, 317 (each within the FERM domain), Tyr570 (within the β2′-β3′ JH2 loop), and Tyr913 (the RS4 residue at the beginning of the β4 loop of the JH1 domain) decrease JAK2 activity.
Adding to the complexity of JAK2 phosphorylation, Ungureanu et al. reported that the pseudokinase domain of JAK2 is a dual-specificity protein kinase that catalyzes the phosphorylation of two negative regulatory sites in this protein: Ser523 and Tyr570 [40]. These investigators demonstrated that the rate of JAK2 JH2 autophosphorylation is about 10% that of the JH1 domain and showed that the β3′-strand K581A mutant was kinase dead. Ser523 phosphorylation occurs before that of Tyr570. They found that Ser523 is constitutively phosphorylated in unstimulated cells and phosphorylation of Tyr570 and other sites occurs after cytokine stimulation. Moreover, Ser523 autophosphorylation appears to be an intramolecular reaction (in cis). Thus, under basal conditions Ser523 phosphorylation serves as a negative regulator of JAK2 protein kinase activity. Thus far it is uncertain whether the other Janus kinases behave similarly.
3.Selected Janus kinase disease targets

3.1Inflammatory disorders

3.1.1Rheumatoid arthritis

Rheumatoid arthritis (RA) is a common disorder of unknown etiology affecting about 1% of the adult population [41] with a female: male ratio of about 3:1 [42, 43]. Uncontrolled active rheumatoid arthritis causes joint damage, disability, decreased quality of life, and cardiovascular and other co-morbidities [41]. The illness is characterized by synovitis of multiple small joints, pain, and morning stiffness. The metacarpophalangeal joints (between the wrist and fingers) are commonly affected. Radiographic changes are diagnostic. Serological findings in RA include the presence of rheumatoid factor (RF) and anti-citrullinated protein antibodies (ACPA). The replacement of the ketimine group (=NH) by a ketone group (=O) in the amino acid arginine in proteins yields a protein-citrulline. The immune system often attacks such proteins leading to autoimmune disorders such as RA. The anti-citrullinated protein antibody level serves as a valuable biomarker for the diagnosis of this illness. The erythrocyte sedimentation rate and plasma C-reactive protein levels, which are signs of inflammation, are usually elevated. The pathogenic mechanisms for the development of rheumatoid arthritis result from a complex interplay of immunological, environmental, and genetic factors that produce dysregulation of the immune system and a breakdown of self-tolerance.
Disease-modifying antirheumatic drugs (DMARDs), the key therapeutic agents, along with low doses of prednisone reduce synovitis and systemic inflammation and improve function. The leading DMARD is methotrexate, which can be combined with other drugs such as sulfasalazine and leflunomide. Methotrexate is a folate antagonist that inhibits nucleotide biosynthesis; the drug also inhibits enzymes involved in purine

metabolism resulting in the accumulation of adenosine thereby leading to inhibition of T cell and B cell activity. Leflunomide is a pyrimidine synthesis inhibitor that blocks dihydroorotate dehydrogenase. The mechanism of anti-inflammatory activity of sulfasalazine is unclear.
Biological agents are used when arthritis is uncontrolled or toxic effects arise with DMARDs [41]. The tumor necrosis factor-α inhibitor infliximab was the first biological agent approved for inflammatory disease, followed by adalimumab, etanercept,
abatacept, rituximab, and tocilizumab. Adalimumab (a human construct) and infliximab (a mouse-human chimeric construct) are two monoclonal antibodies directed against TNF-α while etanercept is a protein that functions as a decoy receptor that binds TNF-α. Abatacept is a fusion protein composed of the Fc region of the immunoglobulin IgG1 fused to the extracellular domain of CTLA-4 (a protein receptor that down-regulates the immune response). Rituximab is a chimeric mouse-human monoclonal antibody against CD20 that destroys B cells, which express CD20. Tocilizumab is a chimeric mouse-
human monoclonal antibody directed against the soluble and membrane forms of the IL-6 receptor. There are at least 20 additional agents called biosimilars that are either approved or are in clinical trials for the treatment of inflammatory disorders [44].
3.1.2Canine dermatitis

Canine dermatitis is an inflammatory process that affects the skin of the face, axillae, and distal extremities of dogs [45, 46]. Depending on the seriousness of the signs of the dermatitis and the attitudes of the owners, the symptomatic treatment and interventional therapy for environmental allergies (allergen avoidance and allergen- specific desensitization) may be implemented [47]. Symptomatic treatment includes the

use of various anti-inflammatory agents including glucocorticoids (systemically or topically) and cyclosporine. Other treatment modalities of lower or less proven efficacies include antihistamines, feline interferon-ω, and pentoxifylline (a phosphodiesterase inhibitor that reduces inflammation).
3.1.3Psoriasis

Psoriasis is a common illness that affects 2–4% of the population and it occurs in men and women with an equal frequency. It is a chronic autoimmune disease that occurs at any age and is typified by patches of abnormal skin [48]. These patches are red, scaly, and they produce itching. They vary in size from small localized lesions to those that completely cover the body. Patients suffer more during the winter and exacerbations can arise after the use of medications such as β-blockers or NSAIDs (non-steroidal anti- inflammatory drugs). The underlying pathogenesis involves an immune reaction to dermal and epidermal cells.
Although there is no cure for psoriasis, various treatments can provide symptomatic relief including such therapies as steroid ointments, vitamin D3 ointments, ultraviolet light, and methotrexate. About three-quarters of cases can be managed with topical ointments alone. People with the disease have an increased risk of cardiovascular problems, Crohn disease, lymphomas, and psoriatic arthritis. Psoriatic arthritis occurs in up to 30% of individuals with the disease.
Psoriasis is characterized by the rapid growth of the epidermis. Accordingly, epidermal cells are not replaced in the usual 28–30 days, but rather they are replaced every 3–5 days. These changes arise from the premature maturation of keratinocytes that result from an inflammatory cascade in the dermis that is produced by various immune

cells including macrophages, T cells, and dendritic cells. These various cells travel from the dermis to the epidermis and secret cytokines such as interleukin-1β (IL-1β), IL-6, IL- 22, along with tumor necrosis factor-α (TNF-α). It appears that these cytokines induce the growth and proliferation of keratinocytes.
3.1.4Ulcerative colitis

Ulcerative colitis is a chronic relapsing form of inflammatory bowel disease that causes inflammation and ulcers in the colon [49]. The illness usually occurs in young adults with symptoms ranging from mild to severe. The hallmark symptom of active disease is bloody diarrhea. The incidence of newly diagnosed ulcerative colitis ranges from 1–20 people per 100,000. Ulcerative colitis can be treated with a number of medications, including 5-aminosalicylic acid (ASA) pro-drugs such as sulfasalazine and mesalazine, which inhibit the synthesis of inflammatory cytokines.
Immunosuppressive corticosteroids such as prednisone are used for short-term, but not long-term treatment [49]. Other immunosuppressive medications such as azathioprine, which is metabolized to 6-mercaptopurine thereby inhibiting cellular replication and purine synthesis, are also used. Biological agents such as infliximab and adalimumab are given only if people cannot achieve remission with 5-aminosalicylic acid and corticosteroids. Biologic treatments such as the TNF inhibitors infliximab, adalimumab, golimumab (a fully human antibody directed against TNF-α), and vedolizumab (a humanized monoclonal antibody directed against α4β7 integrin) are also used to treat patients with ulcerative colitis who are no longer responding to corticosteroids. Usually these medications are used only after other options have been exhausted (i.e., the patient has received and not responded favorably to high-dose

corticosteroids and immunomodulators such as azathioprine and mesalazine). In contrast to Crohn disease as described next, methotrexate is ineffective in the treatment of ulcerative colitis. Surgical removal of the colon (colectemy) can cure ulcerative colitis. This is usually performed only in patients with severe symptomatology that is unresponsive to medical management.
3.1.5Crohn disease

In contrast to ulcerative colitis which affects only the colon, Crohn disease is an inflammatory bowel disease (IBD) that may affect any part of the gastrointestinal tract [50]. It is an uncommon disorder with a prevalence of about 300 per 1,000,000 people in North America and Europe while it is even less common in Africa and Asia. Typical signs and symptoms include abdominal pain, diarrhea, and fever along with weight loss. The pathogenesis of Crohn disease involves a combination of bacterial, genetic, environmental, and immunological factors. It is a chronic disorder resulting from an immune response to the gastrointestinal tract that may be directed at microbial antigens. Because the immune system is not reacting to indigenous body cells, Crohn disease is not
an autoimmune disease per se. Despite considerable work, the mechanistic basis of Crohn disease is puzzling.
There are no medications or surgical procedures that can cure Crohn disease [50]. Moreover, surgery is contraindicated owing to the development of fistulas. Treatment options can only reduce symptoms, maintain remission, and prevent relapse. Medications used to treat the symptoms of Crohn disease include 5-aminosalicylic acid (5-ASA) formulations, prednisone, immunomodulators such as azathioprine, methotrexate, infliximab, adalimumab, and vedolizumab. Corticosteroids are used in severe attacks of

Crohn disease. Biological therapies are medications used to avoid long-term steroid use and to decrease inflammation.
3.2Neoplastic disorders

3.2.1Myelofibrosis

Myelofibrosis is a rare idiopathic disorder that results from the proliferation of hematopoietic stem cells in the bone marrow and results in the replacement of these cells with scar tissue [51]. Primary myelofibrosis occurs spontaneously whereas secondary myelofibrosis develops after polycythemia vera or after thrombocythemia. Besides bone marrow fibrosis, primary myelofibrosis is a chronic illness characterized by ineffective red cell production, excess production of dysplastic megakaryocytes, extramedullary hematopoiesis, systemic inflammation with excess circulating levels of proinflammatory cytokines, and shortened life expectancy. The occurrence of extramedullary hematopoiesis in the spleen is commonly observed.
The median age at the time of the initial diagnosis of primary myelofibrosis is about 64 years [52] and the signs and symptoms include fatigue, early satiety, insomnia, abdominal discomfort, night sweats, abdominal pain, dizziness, cough, bone pain, itching, weight loss, headache, and fever [51]. Laboratory examination typically reveals anemia with tear-drop shaped erythrocytes, nucleated erythrocytes, and immature myeloid cells as a consequence of extramedullary hematopoiesis. About half of the patients with primary myelofibrosis posses a JAK2 V615F mutation, about 30% possess a calreticulin mutation, and about 10% possess a mutation in the thrombopoietin receptor (MPL); these mutations are mutually exclusive [53]. These genetic changes are considered to be the principal drivers of neoplastic myeloproliferation [51]. The V617F

mutation appears to make hematopoietic cells more sensitive to growth factors that need JAK2 for signal transduction, which include erythropoietin and thrombopoietin (Table 1). Fibroblasts producing the scar tissue in this disorder are polyclonal and not part of a neoplastic clone. Median life expectancy for primary myelofibrosis is about six years.
Myelofibrosis has been treated with immunomodulatory agents such as thalidomide, lenalidomide, or pomalidomide. Hydroxyurea (a ribonucleotide reductase inhibitor), melphalan (a DNA-alkylating agent), busulfan (an alkylating agent), and cladribine (an adenosine deaminase inhibitor) have been used for the treatment of the splenomegaly (enlarged spleen) of primary myelofibrosis with only minimal success. Other treatment modalities for myelofibrosis include transfusion for anemia or radiation therapy for painful sites of extramedullary hematopoiesis. Stem cell transplantation in selected patients is curative, but is infrequently used owing to the lack of suitable donors.
3.2.2Polycythemia vera

Polycythemia vera is a clonal neoplastic stem cell disorder characterized by excessive red blood cell production; this condition may also result in the overproduction of white blood cells and platelets [54]. The overproduction of red cells is independent of erythropoietin and the serum levels of this agent are low. Patients with polycythemia vera may be asymptomatic and the disease is discovered on routine blood analysis. The
median age a diagnosis is about 61 years and the incidence is about 22 per 100,000 population [55]. The signs and symptoms include those listed above for primary myelofibrosis. The JAK2 V615F mutation occurs in about 95% of these people. This condition has the capacity to transform into myelofibrosis 10 to 20 years after diagnosis or into acute myelogenous leukemia in 5 to 20 years after diagnosis. The overall median

life expectancy is about 13.5 years [51]. The goal of therapy is to decrease the hematocrit to less than 45%. This is most commonly achieved by periodic phlebotomies. Moreover, hydroxyurea is a commonly used medication for cytoreductive therapy. This agent inhibits ribonucleotide reductase by scavenging tyrosyl free radicals that are involved in the production of deoxyribonucleoside diphosphates (dADP, dGDP, dCDP, and dUDP). Low-dose aspirin is used to reduce the risk of vascular events.
3.2.3Primary thrombocythemia

Primary thrombocythemia (thrombocytosis) is a clonal disorder of unknown etiology involving a multipotent hematopoietic progenitor cell manifested clinically by the overproduction of blood platelets without erythrocyte overproduction [56]. The median age at diagnosis is in the 50-60 year range and the incidence is about 24 per 100,000 people [57]. Most patients are asymptomatic and elevated platelets are observed on routine hematological analysis. The hematocrit is normal, but there may be a mild elevation of leukocytes. Splenomegaly occurs in about one-quarter of the patients. Although the JAK2 V617F mutation occurs in about half of the people with primary thrombocythemia and myelofibrosis and in 95% of patients with polycythemia vera, Spivak et al. argue that this mutation cannot be the sole explanation for the pathogenesis of these rare disorders [58]. This affliction transforms into myelofibrosis in 8–16 years after the onset of thrombocythemia.
4.Janus kinase mutations in human illnesses

Somatic mutations in JAK1 are found in the SH2 domain and the small and large lobes of the pseudokinase domain that result in AML, ALL, breast ductal carcinoma, and NSCLC [32]. The incidence of activating mutations in T-cell ALL ranges from 6–27%

and the incidence in B-cell ALL is about 1.5% while that of T-cell prolymphocytic leukemia ranges from 8–12% [59]. More than three dozen JAK2 mutations in exon 12 (535MVFHKIRNEDLIF547) result in polycythemia vera while mutations in the small lobe of the pseudokinase domain have been reported in ALL and polycythemia vera [32]. The majority of JAK2 mutations in lymphoid lineage neoplasms involve R683; this residue is at the end of the β7′-strand below the ATP-binding site of JH2 and potentially interacts with the β2–β3 loop of the small lobe of JH1. Mutations of the JAK3 pseudokinase N- lobe have been reported in acute megakaryoblastic leukemia (15%) while large lobe mutations occur in breast ductal carcinomas and ALL. TYK2 mutations have been described in about 21% of T-cell ALL [59]. TYK2 deficiencies are associated with susceptibility to viral and bacterial infections and several TYK2 polymorphisms show a strong linkage to autoimmune afflictions such as multiple sclerosis, systemic lupus erythematosus, Crohn disease, primary biliary cirrhosis, and type I diabetes. Moreover, TYK2 polymorphisms have also been linked to AML.
Rare chromosomal rearrangements linking the JH1 domain-coding portion of the JAK2 gene to the oligomerization domain of either TEL, BCR, PCM1, Pax5, or ETV6 have been reported in atypical chronic myelogenous leukemia [59]. The fusion proteins oligomerize and promote the transphosphorylation of the activation segments resulting in JAK2 JH1 activation and subsequent downstream signaling including the STAT, phosphatidylinositol-3 kinase, or MAP kinase pathways. Moreover, the JAK2 locus is amplified in 30–50% of Hodgkin lymphomas and in some B-cell lymphomas.
5.Selected Janus kinase inhibitors in clinical trials

5.1First-generation drugs

5.1.1Tofacitinib

Tofacitinib bears a pyrrolo[2,3-d]pyrimidine scaffold (Fig. 8A). It is an orally effective drug that was approved by the FDA for the treatment of rheumatoid arthritis, an autoimmune disorder, in 2012 (www.brimr.org/PKI/PKIs.htm). Tofacitinib inhibits JAK1/2 with sub-nanomolar EC50 values; it is also a potent (EC50 less than 5 nM) inhibitor of JAK3 and TYK2 (Table 5). Van Vollenhoven et al. reported on a randomized phase III clinical trial involving 717 patients who were receiving methotrexate who were randomly assigned to cohorts receiving 5 mg of tofacitinib twice daily, 10 mg of tofacitinib twice daily, 40 mg of adalimumab by injection once every 2 weeks, or placebo [42]. After 6 months, positive outcomes were higher among patients receiving 5 mg or 10 mg of tofacitinib (52% and 53%, respectively) and among those receiving adalimumab (47%) than among those receiving placebo (28%). Positive outcomes were defined as a 20% reduction in the number of tender and swollen joints along with improvements in three of five other standard assessment parameters.
Adverse events occurred more frequently with tofacitinib and adalimumab than with placebo [42]. Herpes zoster and cellulitus infections occurred in the two tofacitinib groups, but none in the antibody or placebo groups. Tofacitinib was associated with an increase in both low-density and high-density lipoprotein cholesterol levels and with a reduction of blood neutrophils. A small decrease in hemoglobin (1-3 g/dl) occurred in about 8% of patients in each of the four groups. Owing to the inhibition of erythropoietin receptor signaling by tofacitinib, this lack of an adverse response is somewhat surprising. The authors concluded that tofacitinib was significantly superior to placebo and was numerically similar to adalimumab in efficacy. Based upon this clinical trial, tofacitinib

was approved for the treatment of rheumatoid arthritis in 2012 (www.brimr.org/PKI/PKIs.htm).
Lee et al. reported on a randomized clinical trial involving 958 patients comparing first-line treatment of methotrexate with either 5 mg or 10 mg doses of tofacitinib (twice daily) in previously untreated patients [43]. About 26% of the 5 mg group and 38% of the 10 mg group had positive outcomes compared with 12% of the methotrexate cohort. Positive outcomes were defined as a 70% reduction in the number of tender and swollen joints along with five additional criteria including decreased erythrocyte sedimentation rates, C-reactive protein levels, pain, assessment by the patient, and assessment by the physician. The most common adverse events included infections (herpes zoster) and gastrointestinal symptoms. Herpes zoster occurred in 13 of 373 patients in the 5 mg group, 18 of 397 in the 10 mg group, and 2 of 186 in the methotrexate group. Decreases
in neutrophils and increases in serum creatinine were observed in all groups. Low-density lipoprotein levels increased about 20% in the tofacitinib groups and 4% in the methotrexate group while increases in the high-density lipoprotein levels increased by about 17% in the tofacitinib groups and 7% in the methotrexate group. Tofacitinib is in numerous clinical trials in patients with various inflammatory disorders as indicated in Table 6.
X-ray crystal structures of complexes of tofacitinib with the Janus kinases have been reported [1, 60]. Tofacitinib binds to the active conformation of the active protein kinase domains of these proteins and is therefore a type I inhibitor [27]. Tofacitinib binds within the adenine pocket of these proteins (Fig. 9A). The N7 N–H of the pyrrolo[2,3- d]pyrimidine scaffold forms one hydrogen bond with the hinge residue carbonyl group of

Glu957 and the N1 nitrogen of the scaffold forms a second hydrogen bond with the N–H group of Leu959 in JAK1 (Glu 930 and Leu932 in JAK2; PDB ID: 3FUP and Glu903 and Leu905 in JAK3; PDB ID: 3LXK). A hydrogen bond is observed between (i) β3- K908 and αC-E925 and (ii) β3-K908 and DFG-D1021 (Fig. 9A). The terminal cyanoacetyl handle extends into the front cleft underneath the G-rich loop and the methyl group of the piperidine ring is also found within the front cleft. On the whole, tofacitinib makes hydrophobic contacts with the β1-strand L881 before the G-rich loop, V889
(CS7), A906 (CS8), and M956, F958, and L959 of the hinge, and L1010 (CS6). The drug makes van der Waals contact with β3-K908, DFG-D1021, and R1007 and N1008 within the catalytic loop. The drug makes similar hydrophobic and van der Waals contacts with JAK2/3.
In TYK2, the N7 N–H of the pyrrolo[2,3-d]pyrimidine core forms one hydrogen bond with the hinge residue carbonyl group of Glu979, the N1 nitrogen of the scaffold forms a second hydrogen bond with the N–H group of Val981, and a third hydrogen bond forms between the cyano group and Gly906 in the G-rich loop. A hydrogen bond is observed between (i) β3-K930 and αC-E947, (ii) β3-K930 and DFG-D1041, and (iii) HRD-H1021 and DFG-D1041 (Fig. 9B). Additionally, tofacitinib makes hydrophobic contacts with the β1-strand L903 before the G-rich loop, V911 in the β2 strand, A928 (CS8), and Leu1030 (CS6). The drug makes van der Waals contact with Tyr980 of the hinge, N1028 at the end of the catalytic loop, and DFG-D1041. Although tofacitinib makes one more hydrogen bond with TYK2 than JAK1, its binding affinity is an order of magnitude less probably because of fewer favorable hydrophobic interactions.
5.1.2Oclacitinib

Oclacitinib is an orally effective cyclohexylamino pyrrolo[2,3-d]pyrimidine derivative (Fig. 8B) that was approved by the FDA in 2013 for the control of pruritus associated with allergic dermatitis and for the control of atopic dermatitis in dogs. The drug targets human JAK1/2 (Table 5). Gonzales et al. found that oclacitinib was much less effective in blocking JAK2 activity in human cells in culture with IC50 values near 1400 nM compared with 18 nM in vitro, thereby indicating that findings with purified enzymes may not represent activities in the cellular or whole animal context [61]. Differences in potency of one or two orders of magnitude between values obtained with purified proteins and with cellular assays should be considered in all cases. The human and canine proteins exhibit about 95% homology.
Cosgrove et al. reported on a randomized clinical study evaluating oclacitinib vs. placebo [68]. They reported that the drug provided rapid and safe control of atopic dermatitis with improvements in skin lesions. This same group performed a double- blinded and randomized clinical trial comparing the efficacy and safety of oral oclacitinib and oral cyclosporine for the control of atopic dermatitis in client-owned dogs [69]. They concluded that both drugs were effective, but that oclacitinib brought about a faster resolution of symptoms. Most adverse events with oclacitinib included vomiting and diarrhea early in the course of treatment and these symptoms were resolved following standard practices. There is no indication that oclacitinib is being tested in human clinical trials.
No X-ray structural studies of oclacitinib bound to any Janus kinase have been reported. To obtain an idea on the possible interaction of this drug with JAK1, the Schrödinger Glide Suite (2016-1 release) was used to dock the drug into human JAK1

(with initially bound tofacitinib, PDB ID: 3EYG) [70]. The pyrrolopyrimidine core in the resulting model makes hydrogen bonds with the carbonyl oxygen of E957 and with the N–H group of L959, both within the hinge (Fig. 9C). The drug makes hydrophobic contacts with L881 immediately before the G-rich loop, V889 immediately after the G- rich loop, A906 (CS7), V938 (Sh1), M956 within the hinge, and L1010 (CS6). These interactions parallel those observed in the model of ruxolitinib binding to JAK2 as described later.
5.1.3Baricitinib

Baricitinib is an orally effective pyrrolo[2,3-d]pyrimidine derivative like tofacitinib, oclacitinib, and ruxolitinib (Fig. 8C). This congener is undergoing clinical trials for RA and psoriasis. Keystone et al. performed a phase II study involving 301 patients that were randomized 2:1:1:1:1 to receive once daily doses of placebo or 1, 2, 4 or 8 mg of baricitinib for 12 weeks [71]. The patients included in this study failed to respond to prior methotrexate treatment. The primary endpoint was the proportion of patients in the combined 4 and 8 mg groups achieving a standard response at 12 weeks as defined by the American College of Rheumatology. They reported that more patients in the combined baricitinib 4 and 8 mg groups compared with placebo achieved such a response (76% vs. 41%). Serious infections developed in three patients receiving baricitinib. Dose-dependent decreases in hemoglobin were observed with baricitinib. They found that baricitinib improved the signs and symptoms of RA in these people who did not respond to methotrexate. Moreover, baricitinib was well tolerated with no unexpected safety findings after 24 weeks of treatment.

Genovese et al. performed a phase III study involving 527 patients with refractory rheumatoid arthritis with an inadequate response to or unacceptable side effects associated with one or more TNF inhibitors as described in Section 7.1.2, other disease- modifying antirheumatic drugs (DMARDs), or both [72]. They randomly assigned the
patients in a 1:1:1 ratio to baricitinib at a dose of 2 or 4 mg daily or placebo for 24 weeks. They used standard responses as defined by the American College of Rheumatology and other standard criteria in their assessment of efficacy. They reported that significantly more patients receiving baricitinib at the 4-mg dose than those receiving placebo had a positive response at 12 weeks (55% vs. 27%). The rates of serious adverse events were 4%, 10%, and 7% in the three groups, respectively. Two non-melanoma skin cancers and two major adverse cardiovascular events, including a fatal stroke, occurred in the higher- dose group. Baricitinib was associated with a small reduction in neutrophils and increases in serum creatinine and low-density lipoprotein cholesterol levels. They concluded that patients with RA and a previously inadequate response to biologic DMARDs exhibited significant clinical improvement at 12 weeks.
Papp et al. reported on the results of a phase II clinical trial on the effectiveness of baricitinib in patients with moderate-to-severe psoriasis. Patients were randomized (n = 271) to receive placebo or oral baricitinib at 2, 4, 8 or 10 mg once daily for 12 weeks
[73]. Dose adjustment for 12 additional weeks was based on improvements in the Psoriasis Area and Severity Index (PASI) score. The primary end point was PASI 75% (PASI-75) after 12 weeks for North American patients (n = 238); secondary end points were safety and efficacy measures. After 12 weeks, they found that more North American patients in the 8-mg (43%) and 10-mg (54%) baricitinib groups achieved a PASI-75 score

than in the placebo group (17%). All baricitinib-treated groups had greater mean changes from baseline in their PASI scores at 12 weeks. More than 81% of PASI-75 responders maintained their scores through 24 weeks. Study discontinuations due to adverse events were 0%, 0%, 2.8%, 6.3% and 5.8% for placebo and 2-, 4-, 8- and 10-mg baricitinib groups, respectively. No opportunistic infections were observed in any treatment group. Dose-dependent changes in laboratory values were observed. These included small decreases in neutrophil and hemoglobin levels and small increases in creatinine and low- and high-density lipoproteins. Lymphocyte levels increased during the first week of treatment and returned to normal levels after the second week. They concluded that patients with moderate-to-severe psoriasis treated with baricitinib for 12 weeks achieved significant improvements in their PASI-75 scores.
No X-ray structural studies of baricitinib bound to the Janus kinases have been reported. To obtain an idea on the possible interaction of this drug with JAK1, the Schrödinger Glide Suite (2016-1 release) was used to dock the drug into JAK1 (with initially bound tofacitinib, PDB ID: 3EYG) [70]. The pyrrolopyrimidine core in the resulting pose makes hydrogen bonds with the carbonyl oxygen of E957 and with the N– H group of L959, both within the hinge (Fig. 9D). The drug makes hydrophobic contacts with L881 immediately before the G-rich loop, V889 immediately after the G-rich loop, A906 (CS7), V938 (Sh1), M956 within the hinge, and L1010 (CS6). It also makes van der Waals contact with DFG-D1021. Thus far the clinical trails of baricitinib in the treatment of RA and of psoriasis look promising.
5.1.4Ruxolitinib

Ruxolitinib, like the three previous drugs, is a pyrrolo[2,3-d]pyrimidine derivative (Fig. 8D) and is approved for the treatment of myelofibrosis and polycythemia vera (www.brimr.org/PKI/PKIs.htm). Ruxolitinib is an orally effective JAK1/2 inhibitor; its IC50 values for these proteins are in the sub-nanomolar range while those for JAK3 and TYK2 are in the low nanomolar range (Table 5). Verstovsek et al. examined the efficacy of ruxolitinib vs. placebo in the treatment of 309 randomized intermediate- and high-risk groups of patients with myelofibrosis in the COMFORT-I clinical trial [74]. The primary endpoint was the proportion of patients with a decrease in spleen volume (determined by magnetic resonance imaging) of 35% or more at 24 weeks. Secondary endpoints included the durability of response, changes in symptom burden, and overall survival. About 42% of the ruxolitinib group (n = 155) achieved the primary response vs. 0.7% of the placebo group (n = 154). About 46% of the ruxolitinib group experienced symptom improvement compared with 5% who received placebo. They reported that 13 deaths occurred in the group receiving drug compared with 24 in the control group. Anemia occurred in 96% and thrombocytopenia occurred in 70% of the drug-treated group compared with 87%
and 31% of the control group. However, these adverse events did not lead to the discontinuance of the drug. Fatigue, diarrhea, peripheral edema were the most common non-hematologic adverse responses. In a parallel study, Harrison reported that ruxolitinib was equally effective in patients with or without the V617F JAK2 mutation [75]. As a result of these studies, the FDA approved the use of ruxolitinib in the treatment of primary myelofibrosis in 2012.
In a randomized study, Verstovsek et al. evaluated ruxolitinib vs. best available therapy for efficacy/safety in hydroxyurea-resistant or intolerant patients with

polycythemia vera in the 80-week follow-up of the RESPONSE clinical trial [76]. The best available therapeutic options included interferon, pegylated interferon, pipobroman (an alkylating agent), anagrelide (a phosphodiesterase inhibitor), or immunomodulators (lenalidomide, thalidomide). They reported that the hematocrit normalized in 60% vs. 19% and a 35% or more reduction in spleen volume occurred in 40% vs. 0.9% of patients taking ruxolitinib vs. the control group. New or worsening hematologic laboratory abnormalities in ruxolitinib-treated patients were primarily grade 1 or 2 decreases in hemoglobin, lymphocytes, and platelets (where grade 4 is most serious). The thromboembolic event rate per 100 patient-years was 1.8 with ruxolitinib treatment vs. 8.2 with best available therapy. These data support ruxolitinib as an effective long-term treatment option for hydroxyurea-resistant or intolerant patients with polycythemia vera. The most common non-hematologic adverse events each occurring in about 15% of patients were back pain, cough, diarrhea, and fever. Ruxolitinib was approved for the treatment of polycythemia vera in 2014 and currently is in several clinical trials in patients with hematologic and solid tumors along with inflammatory disorders as listed in Table 6.
No X-ray structural studies of ruxolitinib bound to the Janus kinases have been reported. To obtain an idea on the possible interaction of this drug with JAK2, the Schrödinger Glide Suite (2016-1 release) was used to dock ruxolitinib into JAK2 (with initially bound tofacitinib, PDB ID: 3FUP) [70]. In the pose of the resulting model, the N–H group from the core forms a hydrogen bond with E930 carbonyl group of the hinge and the N–H group from L932 of the hinge forms a hydrogen bond with a nitrogen of the fused pyrimidine ring of the drug (Fig. 9E). The β3-K882 makes polar contacts with αC-

E988 and with DFG-D994. The drug makes hydrophobic contact with L855 before the G- rich loop, V863 in the β2-strand, A880 (CS7) in the β3-strand, V911 (Sh1) in the αC-β4 back loop, M929 before the hinge, and L983 (CS6). The protein is in its active conformation thus making ruxolitinib a type I inhibitor of JAK2 [27]. The computer- derived interaction of ruxolitinib with JAK1/3 and TYK2 is similar to that described here for JAK2.
Ruxolitinib is also a type I inhibitor of Src (PDB:ID 4U5J) [27]. However, the disposition of the pyrrolo[2,3-d]pyrimidine core is flipped about 180° with the same side of the core interacting with the hinge when compared with the binding to Src; we refer to the experimentally determined Src pose as orientation I and the in silico version with JAK2 as orientation II. Ruxolitinib is a type I inhibitor of Src that binds within the cleft of its active conformation and it forms two hydrogen bonds with hinge residue M341 [27]. The drug also makes hydrophobic contacts with several residues including CS6, CS7, and CS8.
5.2Second-generation drugs

5.2.1Decernotinib

Decernotinib is an orally effective pyrrolo[2,3-b]pyridine derivative (Fig. 8E) that is in clinical trials for the treatment of RA. In a pre-clinical study, Mahajan et al. evaluated the potency and selectivity of decernotinib against JAK3 kinase activity and inhibition of JAK3-mediated signaling in vitro and JAK3-dependent physiologic processes in vivo [77]. They found that decernotinib potently inhibits JAK3 in enzyme assays (Ki = 2.5 nM) and cellular assays dependent on JAK3 activity (IC50 range, 50-170 nM), with limited or no measurable potency against other Janus kinase isotypes or non-

JAK enzymes. These data again indicate that potencies based upon purified proteins can differ markedly from cellular or whole animal measurements. Decernotinib resulted in dose-dependent reduction in ankle swelling in a rat collagen-induced arthritis model. In a mouse model of oxazolone-induced delayed-type hypersensitivity, decernotinib reduced the T cell-mediated inflammatory response in skin. These investigators concluded that the data support evaluation of decernotinib for treatment of patients with autoimmune and inflammatory problems such as rheumatoid arthritis.
Genovese et al. reported on a double-blinded, randomized, phase II clinical trial in patients with RA receiving decernotinib at doses of 100 mg (n=11), 200 mg (n=10), 300 mg (n=10), or placebo (n=12) once daily for 12 weeks [78]. They determined the American College of Rheumatology score (ACR20; improvement of ≥20%) and disease activity score (DAS) using C-reactive protein (CRP) and the RA MRI (magnetic resonance imaging) scoring (RAMRIS) systems. They found that the ACR20 response at 12 weeks was 63%, 60%, and 60% in the decernotinib 100-mg, 200-mg and 300-mg groups, respectively, compared with 25% in the placebo group. They also reported that the DAS-CRP scores decreased in a dose-dependent fashion. Decreases in RAMRIS synovitis scores were significantly different from placebo for all of the decernotinib-dose cohorts and for the RAMRIS osteitis scores for the decernotinib-300 mg cohort. They reported that treatment was generally well tolerated. These investigators also found that decernotinib combined with methotrexate improved the signs and symptoms of RA after 12 and 24 weeks when compared with the placebo group.
Decernotinib binds to the active conformation of JAK2 and is therefore a type I inhibitor [27]. The pyridine nitrogen forms a hydrogen bond with the N–H group of the

hinge L932 while the pyrrolo N–H group hydrogen bonds with the L932 carbonyl oxygen (Fig. 9F). The drug makes hydrophobic contacts with L855 before the G-rich loop, V863 after the G-rich loop, A880 (CS7), V911 (Sh1), I982 (CS4), L983 (CS6), V984 (CS5), and N981 of the catalytic loop (HRDLATRN). The drug makes van der Waals contact with DFG-D994 and Y931 of the hinge. The decernotinib clinical trials for the treatment of RA thus far appear promising.
5.2.2Peficitinib

Peficitinib is an orally effective pyrrolo[2,3-b]pyridine derivative (Fig. 8F) that is in clinical trials for the treatment of psoriasis, RA, and ulcerative colitis. Takeuchio et al. evaluated the efficacy, safety and dose response of peficitinib as monotherapy in Japanese patients with moderate to severe rheumatoid arthritis [79]. In a 12-week,
double-blind study, 281 adult patients with active RA and not on concomitant disease- modifying antirheumatic drug (DMARD) therapy were randomized equally to once-daily placebo or peficitinib at a dosage of 25, 50, 100 or 150 mg. The primary endpoint was American College of Rheumatology (ACR) 20 response in the peficitinib treatment groups vs. placebo after 12 weeks. The mean age of the cohort was 53 years and one quarter had previously received anti-TNF therapy.
Peficitinib treatment prescribed at 50, 100 and 150 mg amounts each showed statistically significantly higher ACR20 response rates compared to the placebo and response rates increased up to the 150 mg dosage. Adverse events included neutropenia, headache, and abdominal pain. The treatment-emergent adverse events occurring more frequently in the peficitinib group compared with the placebo group included diarrhea, nasopharyngitis, and increased serum creatine phosphokinase activity. No cases of

serious infections were reported. Herpes zoster occurred in four patients (two each in the peficitinib 25 and 100 mg cohorts). The authors concluded that treatment with peficitinib as monotherapy for 12 weeks in Japanese patients with moderate to severe RA is efficacious and showed an acceptable safety profile.
No X-ray structural studies of peficitinib bound to the Janus kinases have been reported. To obtain an idea on the possible interaction of this drug with JAK3, the Schrödinger Glide Suite (2016-1 release) was used to dock the drug into JAK3 (with initially bound tofacitinib, PDB ID: 3LXK) [70]. The pose in the computed model indicates that the pyrrolo N–H group makes a hydrogen bond with the carbonyl group of E903 of the hinge and the pyridine nitrogen hydrogen bonds with the N–H group of L905 while the exocyclic N–H group hydrogen bonds with the carbonyl group of L905, also within the hinge (Fig. 9G). In this model, the drug makes hydrophobic contacts with
L828 immediately before the G-rich loop, V835 immediately after the G-rich loop, A853 (CS7), V884 (Sh1), M902 within the hinge, and I955 (CS4) and L956 (CS6). It also makes van der Waals contact with DFG-D967. Clinical studies with this drug are in the early stages.
5.2.3Filgotinib

Filgotinib is an orally effective triazolopyridine congener (Fig. 8G) that is in clinical trials for the treatment of RA and Crohn disease. Filgotinib is an orally effective JAK1 (EC50 of 10 nM), JAK2 (2.8 nM), and TYK2 (11 nM) inhibitor; it is a less effective JAK3 inhibitor (81 nM) [64]. Clinical trials are in early stages thus far establishing a
daily dose of 200 mg [80]; no studies reporting clinical efficacy have been forthcoming.

X-ray crystal structures of complexes of figlotinib with JAK2 have been reported [64]. The drug binds to the active conformation of the protein kinase domain and is therefore a type I inhibitor [27]. Figlotinib binds within the adenine pocket of these proteins. The N1 of the triazolopyridine scaffold forms one hydrogen bond with the hinge residue N–H group of L932 of the hinge and the exocyclic N–H of the drug forms a hydrogen bond with the carbonyl group of L932 (Fig 9H). On the whole, figlotinib makes hydrophobic contacts with the β1-strand L885 before the G-rich loop, V863 (CS7), A886 (CS8), and M929 and Y931 of the hinge, and L983 (CS6). The drug also makes van der Waals contact with β3-K882. There are no available crystal structures for the next five drugs and the glide and the induced-fit docking programs [70, 81] failed to yield satisfactory models.
5.2.4Fedratinib

Fedratinib is an anilinopyrimidine derivative (Fig. 8H) that inhibits JAK2 and is in clinical trials for the treatment of myelofibrosis (Table 6). Pardanani et al. performed a phase I trial with fedratinib in 59 patients with high- or intermediate-risk primary myelofibrosis or myelofibrosis developing after polycythemia vera or essential
thrombocythemia [82]. The maximum-tolerated dose was 680 mg daily and dose-limiting toxicity was a reversible asymptomatic increase in the serum amylase level. Forty-three patients (73%) continued treatment beyond six cycles; the median cumulative exposure to fedratinib was 380 days. Grade 3 to 4 incidence rates of adverse events included nausea (3%), vomiting (3%), diarrhea (10%), anemia (35%), and thrombocytopenia (24%). Fedratinib treatment had a modest effect on serum cytokine levels, but more than half of the patients achieved rapid and durable improvement in their constitutional symptoms

including early satiety, night sweats, fatigue, pruritus, and cough. After six and 12 cycles of treatment, 39% and 47% of the patients, respectively, exhibited a decreased spleen size. The majority of patients with leukocytosis or thrombocytosis at the initial phase of the study (n = 28 and n = 10, respectively) achieved normalization of their blood counts after six (57% and 90%) and 12 (56% and 88%) cycles. A significant decrease in the JAK2 V617F allele burden was observed at six months in the 51 mutation-positive patients and the decrease lasted for more that a year. The authors concluded that fedratinib is well tolerated and produces a significant reduction in the disease burden and a durable clinical benefit in patients with myelofibrosis.
5.2.5Momelotinib

Momelotinib is an anilinopyrimidine derivative (Fig. 8I) that is in clinical trials for the treatment of myeloproliferative neoplasms as well as NSCLC and pancreatic cancer (Table 6). In pre-clinical studies, Tyner et al. found that between 0.5 and 1.5 μM momelotinib produced growth suppression and apoptosis in JAK2-dependent human and murine hematopoietic cell lines, while non-hematopoietic cell lines were unaffected [83]. Based upon work with an experimental murine myeloproliferative system, Tyner et al. found that momelotinib restored physiologic levels of inflammatory cytokines and it also normalized the hematocrit, white blood cell counts, and spleen size [83]. Although these responses included a reduction of the JAK2 V617F allele burden, the JAK2 V617F mutant-containing cells persisted. Consistent with the results of clinical trails with JAK2 inhibitors, myeloproliferative growth resumes following cessation of treatment
Pardanani et al. conducted a Phase I/II clinical trial of momelotinib in patients with high- or intermediate-risk primary or post-polycythemia vera/essential

thrombocythemia myelofibrosis [84]. In the dose-escalation phase (n=21), the maximum- tolerated oral dose was 300 mg daily based on reversible grade 3 headache and an increase in serum lipase activity. Anemia and spleen responses, based upon standardized criteria, were 59% in a cohort receiving 150 mg daily. Among 33 patients who received blood transfusions in the month prior to study entry, 70% achieved a minimum 12-week period without transfusions. Most patients experienced an improvement of their constitutional symptoms. Grade 3 or 4 adverse reactions included thrombocytopenia (32%), hyperlipasemia (5%), elevated liver transaminases (3%), and headache (3%). The plasma levels of more than a dozen cytokines decreased in a large percentage of patients. They reported that momelotinib is well tolerated and produces significant improvements in the signs and symptoms including anemia and spleen size in myelofibrosis patients.
5.2.6Lestaurtinib

Lestaurtinib is an orally available indolocarbazole (Fig. 8J) that is in clinical trials for the treatment of psoriasis and a variety of neoplastic disorders (Table 6). This drug is a potent inhibitor of the Janus kinases (Table 5) as well as FLT3, TRKA, and PRK1 (pubchem.ncbi.nlm.nih.gov). FLT3 is a transmembrane receptor protein-tyrosine kinase that is stimulated by FLT3LG and regulates the growth, proliferation, differentiation, and survival of hematopoietic progenitor and dendritic cells. TRKA is a transmembrane receptor protein-tyrosine kinase that is stimulated by nerve growth factor and regulates the growth, proliferation, differentiation, and survival of neurons in both the central and peripheral nervous systems. PRK1 is a protein kinase C-related cytosolic
serine/threonine-protein kinase that is activated by Rho proteins and is involved in the regulation of transcription, cell motility, and tumor cell invasion.

Santos et al. conducted a phase II clinical study of lestaurtinib in 22 JAK2 V617F mutant-positive myelofibrosis patients (80 mg orally twice daily) [85]. The response rate was 27%. Median time to response was three months and the response lasted more than 14 months. These investigators found that there was no improvement in bone marrow fibrosis nor was there any improvement in the JAK2 V617F allele burden. However, they found that lestaurtinib decreased the phosphorylated STAT3 levels. They reported that eight patients (36%) encountered various grade 3 or 4 toxicities and 6 (27%) required a reduction in dosage. The main side effects that resulted from lestaurtinib treatment included myelosuppression with anemia and thrombocytopenia as well as gastrointestinal symptoms including nausea, vomiting, and diarrhea. Based upon the results of this clinical trial, the authors concluded that lestaurtinib produced modest beneficial
responses in myelofibrosis.

5.2.7Gandotinib

Gandotinib is an imidazopyridazine derivative (Fig. 8K) that inhibits JAK2 and is in clinical trials for myeloproliferative disorders. Results from a phase I/II trial with 19 subjects demonstrated a decrease in spleen size in 22% of the patients with myeloproliferative neoplasms [86]. The JAK2 V716F allele burden was unaffected. Main adverse events included anemia, diarrhea, electrolyte imbalance, and nausea.
5.2.8Pacritinib

Pacritinib is a macrocyclic JAK2 inhibitor (Fig. 8L, Table 5) that is in clinical trials for the treatment of myelofibrosis, leukemia, and solid tumors (Table 6). Chow et al. reported that the drug reduced spleen size and ameliorated bone pain, fatigue, fever, itching, night sweats, and weight loss [87]. Common adverse events included nausea,

vomiting, and diarrhea, while the development of severe hematologic toxicities including anemia and thrombocytopenia was limited. However, the FDA placed all ongoing clinical trials with pacritinib on hold in February 2016 owing to possibility of the increased danger of intracranial hemorrhages and cardiac events.
6.Non-conserved amino acid residues near the Janus kinase ATP-binding sites

The amino acid residues that make up the Janus kinase ATP-binding site and the surrounding region are highly conserved thereby making it difficult to design inhibitors that are specific for a particular family member. Non-conserved residues are listed in Table 7 and their locations within the JH1 domains are depicted in Fig. 10. Of importance is the Cys909 in JAK3 that serves as a potential target for the covalent attachment of an inhibitor. Drugs that possess an α,β-unsaturated carbonyl group undergo a Michael reaction that involves the addition of a nucleophile (the –SH of cysteine) to the double bond to form a covalent Michael adduct. Non-covalent interactions position the drug in a productive orientation within the ATP-binding pocket that allows the covalent modification to proceed. Three FDA-approved drugs use this inhibitory mechanism including afatinib (targeting EGFR in the treatment of NSCLC), ibrutinib (targeting BTK in mantle cell lymphoma), and osimertinib (targeting EGFR T970M mutants in NSCLC) (www.brimr.org/PKI/PKIs.htm).
The gatekeeper residue in the four Janus kinases is the rather bulky methionine. This residue serves as a molecular gate that controls access to a hydrophobic pocket or back cavity adjacent to the adenine binding site [88, 89]. Its shape and size are primarily controlled by the nature of the gatekeeper residue, which is the last residue of the β5- strand for the Janus kinases or the first residue of the hinge connecting the small and

large lobes, depending upon the particular enzyme. When the gatekeeper is small, the hydrophobic pocket expands toward the β5-strand and the αC-helix and large antagonists can be accommodated. When the gatekeeper residue has a bulky side chain, as in the cases of the Janus kinases, the back cavity is small and delimited by the β4-strand and by the gatekeeper side chain itself. This limits the size of potential Janus kinase antagonists owing to the gatekeeper methionine blockade of the back pocket. In agreement with this supposition, none of the drugs described in Section 5 extend toward the β5-strand or the αC-helix.
7.Epilogue

Of all of the human protein kinase proteins, only five contain both a pseudokinase and an active kinase: JAK1/2/3, TYK2, and GCN2 (a serine/threonine protein kinase) (Supplementary Material in Ref. [4]). The latter enzyme is encoded by the EIF2AK4 gene and catalyzes the phosphorylation of protein synthesis elongation factor-2. MSK1/2, obscurin, RSK1/2/3/4, and SPEG are protein-serine/threonine kinases that contain two active catalytic domains. The MSK and RSK enzymes catalyze the phosphorylation of ribosomal proteins and SPEG is a striated muscle preferentially expressed gene that participates in muscle cell differentiation. Obscurin is the product of the OBSCN gene
and consists of 7968 amino acid residues with a molecular mass of 868 kDa! This enzyme participates in myofibrillogenesis.
There are 32 small molecule protein kinase inhibitors approved by the FDA for the treatment of human illnesses (www.brimr.org/PKI/PKIs.htm) and one drug (oclacitinib) that is approved for the treatment of canine dermatitis. Whereas the cost of the drugs for human sicknesses is around $10,000 per month in the United States and is

steadily increasing, the cost of the canine drug is about $60 per month. As noted previously, the high cost of small molecule inhibitors produces financial toxicity [22], a concern that is not addressed in clinical trials and only becomes apparent when the patient is billed. Even with health insurance, the cost of co-payments is often prohibitive.
There are more than 200 different small molecule protein kinase inhibitors in clinical trials worldwide. Although one aim in the development of such inhibitors has been to target a single enzyme, it has turned out that many, if not most, of the selective inhibitors have been found at later stages to inadvertently inhibit multiple targets. It may be counterintuitive, but the majority of the approved protein kinase antagonists are multikinase inhibitors. Moreover, their therapeutic effectiveness may be related to the simultaneous inhibition of more than a single target. Accordingly, we have the question of whether magic shotguns are to be preferred over magic bullets [90].
Nearly all of the initial approved small molecule protein kinase inhibitors were directed at various malignancies. The exceptions include: (i) sirolimus (an mTOR inhibitor) that was approved for the prevention of renal graft-vs.-host disease in 1999, tofacitinib (a pan-JAK inhibitor) that was approved for the treatment of RA in 2012, and nintedanib (a fibroblast growth factor receptor multi-kinase inhibitor) that was approved for the treatment of idiopathic pulmonary fibrosis in 2014. As the number of disease targets increases, we can expect advances in clinical efficacy and subsequent approval of new drugs targeting additional protein kinases as well as additional illnesses such as autoimmune afflictions (as described in this paper), hypertension, and Parkinson disease [91].
Conflict of interest

The author is unaware of any affiliations, memberships, or financial holdings that might be perceived as affecting the objectivity of this review.
Acknowledgment

The author thanks Laura M. Roskoski for providing editorial and bibliographic assistance.
References

[1]Williams NK, Bamert RS, Patel O, Wang C, Walden PM, Wilks AF, et al. Dissecting specificity in the Janus kinases: the structures of JAK-specific inhibitors complexed to the JAK1 and JAK2 protein tyrosine kinase domains. J Mol Biol 2009;387:219–32.
[2]Wilks AF. The JAK kinases: not just another kinase drug discovery target. Semin Cell Dev Biol 2008;19:319–28.
[3]Kawamura M, McVicar DW, Johnston JA, Blake TB, Chen YQ, Lal BK, et al. Molecular cloning of L-JAK, a Janus family protein-tyrosine kinase expressed in natural killer cells and activated leukocytes. Proc Natl Acad Sci U S A 1994;91:6374–8.
[4]Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The protein kinase complement of the human genome. Science 2002;298:1912–34.
[5]O’Shea JJ, Schwartz DM, Villarino AV, Gadina M, McInnes IB, Laurence A. The JAK-STAT pathway: impact on human disease and therapeutic intervention. Annu Rev Med 2015;66:311–28.
[6]BF Haynes, KA Soderberg, AS Fauci, Introduction to the immune system, in: DL Longo, DL Kasper, JL Jameson, AS Fauci, SL Hauser, J Loscalzo (Eds.) Harrison’s Principles of Internal Medicine (18th edition), McGraw Hill Medical, New York, 2012, pp. 2650–2685.

[7]Babon JJ, Lucet IS, Murphy JM, Nicola NA, Varghese LN. The molecular regulation of Janus kinase (JAK) activation. Biochem J 2014;462:1–13.
[8]Abroun S, Saki N, Ahmadvand M, Asghari F, Salari F, Rahim F. STATs: An Old Story, Yet Mesmerizing. Cell J 2015 l;17:395–411.
[9]Chen X, Vinkemeier U, Zhao Y, Jeruzalmi D, Darnell JE Jr, Kuriyan J. Crystal structure of a tyrosine phosphorylated STAT-1 dimer bound to DNA. Cell 1998;93:827– 39.
[10]Roskoski R Jr. A historical overview of protein kinases and their targeted small molecule inhibitors. Pharmacol Res 2015;100:1–23.
[11]Irie-Sasaki J, Sasaki T, Matsumoto W, Opavsky A, Cheng M, Welstead G, et al. CD45 is a JAK phosphatase and negatively regulates cytokine receptor signalling. Nature. 2001 18;409:349–54.
[12]Yamaoka K, Saharinen P, Pesu M, Holt VE 3rd, Silvennoinen O, O’Shea JJ. The Janus kinases (Jaks). Genome Biol 2004;5:253.
[13]Knighton DR, Zheng JH, Ten Eyck LF, Ashford VA, Xuong NH, Taylor SS, et al. Crystal structure of the catalytic subunit of cyclic adenosine monophosphate-dependent protein kinase. Science 1991;253:407–14.
[14]Knighton DR, Zheng JH, Ten Eyck LF, Xuong NH, Taylor SS, Sowadski JM. Structure of a peptide inhibitor bound to the catalytic subunit of cyclic adenosine monophosphate-dependent protein kinase. Science 1991;253:414–20.
[15]Taylor SS, Keshwani MM, Steichen JM, Kornev AP. Evolution of the eukaryotic protein kinases as dynamic molecular switches. Philos Trans R Soc Lond B Biol Sci 2012;367:2517–28.

[16]Hanks SK, Hunter T. Protein kinases 6. The eukaryotic protein kinase superfamily: kinase (catalytic) domain structure and classification. FASEB J 1995;9:576–96.
[17]Meharena HS, Chang P, Keshwani MM, Oruganty K, Nene AK, Kannan N, et al. Deciphering the structural basis of eukaryotic protein kinase regulation. PLoS Biol 2013;11:e1001680.
[18]Nolen B, Taylor S, Ghosh G. Regulation of protein kinases; controlling activity through activation segment conformation. Molecular Cell 2004;15:661–75.
[19]Chatti K, Farrar WL, Duhé RJ. Tyrosine phosphorylation of the Janus kinase 2 activation loop is essential for a high-activity catalytic state but dispensable for a basal catalytic state. Biochemistry 2004;43:4272–83.
[20]Kornev AP, Haste NM, Taylor SS, Ten Eyck LF. Surface comparison of active and inactive protein kinases identifies a conserved activation mechanism. Proc Natl Acad Sci U S A 2006;103:17783–8.
[21]Kornev AP, Taylor SS, Ten Eyck LF. A helix scaffold for the assembly of active protein kinases. Proc Natl Acad Sci U S A 2008;105:14377–82.
[22]Roskoski R Jr. Cyclin-dependent protein kinase inhibitors including palbociclib as anticancer drugs. Pharmacol Res. 2016;107:249–75.
[23]Roskoski R Jr. Src protein-tyrosine kinase structure, mechanism, and small molecule inhibitors. Pharmacol Res 2015;94:9–25.
[24]Roskoski R Jr. ErbB/HER protein-tyrosine kinases: Structures and small molecule inhibitors. Pharmacol Res 2014;87:42–59.
[25]Roskoski R Jr. ERK1/2 MAP kinases: structure, function, and regulation. Pharmacol Res 2012;66:105–43.

[26]Roskoski R Jr. MEK1/2 dual-specificity protein kinases: structure and regulation. Biochem Biophys Res Commun 2012;417:5–10.
[27]Roskoski R Jr. Classification of small molecule protein kinase inhibitors based upon the structures of their drug-enzyme complexes. Pharmacol Res 2016;103:26–48.
[28]Feng J, Witthuhn BA, Matsuda T, Kohlhuber F, Kerr IM, Ihle JN. Activation of Jak2 catalytic activity requires phosphorylation of Y1007 in the kinase activation loop. Mol Cell Biol 1997;17:2497–501.
[29]Zhou YJ, Hanson EP, Chen YQ, Magnuson K, Chen M, Swann PG, et al. Distinct tyrosine phosphorylation sites in JAK3 kinase domain positively and negatively regulate its enzymatic activity. Proc Natl Acad Sci U S A 1997;94:13850–5.
[30]Gauzzi MC, Velazquez L, McKendry R, Mogensen KE, Fellous M, Pellegrini S. Interferon-α-dependent activation of Tyk2 requires phosphorylation of positive regulatory tyrosines by another kinase. J Biol Chem 1996;271:20494–500.
[31]Saharinen P, Takaluoma K, Silvennoinen O. Regulation of the Jak2 tyrosine kinase by its pseudokinase domain. Mol Cell Biol 2000;20:3387–95.
[32]Lupardus PJ, Ultsch M, Wallweber H, Bir Kohli P, Johnson AR, et al. Structure of the pseudokinase-kinase domains from protein kinase TYK2 reveals a mechanism for Janus kinase (JAK) autoinhibition. Proc Natl Acad Sci U S A 2014;111:8025–30.
[33]Waters MJ, Brooks AJ. JAK2 activation by growth hormone and other cytokines. Biochem J 2015;466:1–11.
[34]Lu X, Levine R, Tong W, Wernig G, Pikman Y, Zarnegar S, et al. Expression of a homodimeric type I cytokine receptor is required for JAK2V617F-mediated transformation. Proc Natl Acad Sci U S A 2005;102:18962–7.

[35]Waters MJ, Brooks AJ, Chhabra Y. A new mechanism for growth hormone receptor activation of JAK2, and implications for related cytokine receptors. JAKSTAT 2014;3:e29569.
[36]Ishida-Takahashi R, Rosario F, Gong Y, Kopp K, Stancheva Z, Chen X, et al. Phosphorylation of Jak2 on Ser523 inhibits Jak2-dependent leptin receptor signaling. Mol Cell Biol 2006;26:4063–73. Erratum in: Mol Cell Biol 2006;26:6309.
[37]Mazurkiewicz-Munoz AM, Argetsinger LS, Kouadio JL, Stensballe A, Jensen ON, Cline JM, et al. Phosphorylation of JAK2 at serine 523: a negative regulator of JAK2 that is stimulated by growth hormone and epidermal growth factor. Mol Cell Biol 2006;26:4052–62.
[38]Argetsinger LS, Kouadio JL, Steen H, Stensballe A, Jensen ON, Carter-Su C. Autophosphorylation of JAK2 on tyrosines 221 and 570 regulates its activity. Mol Cell Biol 2004;24:4955–67.
[39]Argetsinger LS, Stuckey JA, Robertson SA, Koleva RI, Cline JM, Marto JA, et al. Tyrosines 868, 966, and 972 in the kinase domain of JAK2 are autophosphorylated and required for maximal JAK2 kinase activity. Mol Endocrinol 2010;24:1062–76.
[40]Ungureanu D, Wu J, Pekkala T, Niranjan Y, Young C, Jensen ON, et al. The pseudokinase domain of JAK2 is a dual-specificity protein kinase that negatively regulates cytokine signaling. Nat Struct Mol Biol 2011;18:971–6.
[41]Scott DL, Wolfe F, Huizinga TW. Rheumatoid arthritis. Lancet 2010;376:1094–108.

[42]van Vollenhoven RF, Fleischmann R, Cohen S, Lee EB, García Meijide JA, Wagner S, et al. Tofacitinib or adalimumab versus placebo in rheumatoid arthritis. N Engl J Med 2012;367:508–19.

[43]Lee EB, Fleischmann R, Hall S, Wilkinson B, Bradley JD, Gruben D, et al. Tofacitinib versus methotrexate in rheumatoid arthritis. N Engl J Med 2014;370:2377– 86.
[44]Dörner T, Strand V, Cornes P, Gonçalves J, Gulácsi L, Kay J, et al. The changing landscape of biosimilars in rheumatology. Ann Rheum Dis 2016;75:974–82.
[45]Pucheu-Haston CM. Atopic dermatitis in the domestic dog. Clin Dermatol 2016;34:299–303.
[46]Santoro D, Rodrigues Hoffmann A. Canine and Human Atopic Dermatitis: Two Faces of the Same Host-Microbe Interaction. J Invest Dermatol 2016;136:1087–9.
[47]Saridomichelakis MN, Olivry T. An update on the treatment of canine atopic dermatitis. Vet J 2016;207:29–37.
[48]Schleicher SM. Psoriasis: pathogenesis, assessment, and therapeutic update. Clin Podiatr Med Surg 2016;33:355–66.
[49]Sofia MA, Rubin DT. Current approaches for optimizing the benefit of biologic therapy in ulcerative colitis. Therap Adv Gastroenterol 2016;9:548–59.
[50]Baumgart DC, Sandborn WJ. Crohn’s disease. Lancet 2012;380:1590–605.

[51]Bryan JC, Verstovsek S. Overcoming treatment challenges in myelofibrosis and polycythemia vera: the role of ruxolitinib. Cancer Chemother Pharmacol 2016. doi: 10.1007/s00280-016-3012-z.
[52]Cervantes F, Dupriez B, Pereira A, Passamonti F, Reilly JT, Morra E, et al. New prognostic scoring system for primary myelofibrosis based on a study of the International Working Group for Myelofibrosis Research and Treatment. Blood 2009;113:2895–901.

[53]Klampfl T, Gisslinger H, Harutyunyan AS, Nivarthi H, Rumi E, Milosevic JD, et al. Somatic mutations of calreticulin in myeloproliferative neoplasms. N Engl J Med 2013;369:2379–90.
[54]Geyer H, Scherber R, Kosiorek H, Dueck AC, Kiladjian JJ, Xiao Z, et al. Symptomatic profiles of patients with polycythemia vera: implications of inadequately controlled disease. J Clin Oncol 2016;34:151–9.
[55]Ma X, Vanasse G, Cartmel B, Wang Y, Selinger HA. Prevalence of polycythemia vera and essential thrombocythemia. Am J Hematol 2008;83:359–62.
[56]Cerquozzi S, Tefferi A. Blast transformation and fibrotic progression in polycythemia vera and essential thrombocythemia: a literature review of incidence and risk factors. Blood Cancer J 2015:e366.
[57]Mesa RA, Silverstein MN, Jacobsen SJ, Wollan PC, Tefferi A. Population-based incidence and survival figures in essential thrombocythemia and agnogenic myeloid metaplasia: an Olmsted County Study, 1976-1995. Am J Hematol 1999;61:10–5.
[58]Spivak JL, Considine M, Williams DM, Talbot CC Jr, Rogers O, Moliterno AR, et al. Two clinical phenotypes in polycythemia vera. N Engl J Med 2014;371:808–17.
[59]Springuel L, Renauld JC, Knoops L. JAK kinase targeting in hematologic malignancies: a sinuous pathway from identification of genetic alterations towards clinical indications. Haematologica 2015;100:1240–53.
[60]Davis MI, Hunt JP, Herrgard S, Ciceri P, Wodicka LM, Pallares G, et al. Comprehensive analysis of kinase inhibitor selectivity. Nat Biotechnol 2011;29:1046-51.

[61]Gonzales AJ, Bowman JW, Fici GJ, Zhang M, Mann DW, Mitton-Fry M. Oclacitinib (APOQUEL®) is a novel Janus kinase inhibitor with activity against cytokines involved in allergy. J Vet Pharmacol Ther 2014;37:317–24.
[62]Farmer LJ, Ledeboer MW, Hoock T, Arnost MJ, Bethiel RS, Bennani YL, et al. Discovery of VX-509 (decernotinib): a potent and selective janus kinase 3 inhibitor for the treatment of autoimmune diseases. J Med Chem 2015;58:7195-216.
[63]Takeuchi T, Tanaka Y, Iwasaki M, Ishikura H, Saeki S, Kaneko Y. Efficacy and safety of the oral Janus kinase inhibitor peficitinib (ASP015K) monotherapy in patients with moderate to severe rheumatoid arthritis in Japan: a 12-week, randomised, double- blind, placebo-controlled phase IIb study. Ann Rheum Dis 2016;75:1057-64.
[64]Menet CJ, Fletcher SR, Van Lommen G, Geney R, Blanc J, Smits K, et al. Triazolopyridines as selective JAK1 inhibitors: from hit identification to GLPG0634. J Med Chem 2014;57:9323-42.
[65]Ma L, Clayton JR, Walgren RA, Zhao B, Evans RJ, Smith MC, et al. Discovery and characterization of LY2784544, a small-molecule tyrosine kinase inhibitor of JAK2V617F. Blood Cancer J 2013;3:e109.
[66]Poulsen A, William A, Blanchard S, Lee A, Nagaraj H, Wang H, et al. Structure- based design of oxygen-linked macrocyclic kinase inhibitors: discovery of SB1518 and SB1578, potent inhibitors of Janus kinase 2 (JAK2) and Fms-like tyrosine kinase-3 (FLT3). J Comput Aided Mol Des 2012;26:437-50.
[67]Chrencik JE, Patny A, Leung IK, Korniski B, Emmons TL, Hall T, et al. Structural and thermodynamic characterization of the TYK2 and JAK3 kinase domains in complex with CP-690550 and CMP-6. J Mol Biol 2010;400:413–33.

[68]Cosgrove SB, Wren JA, Cleaver DM, Walsh KF, Follis SI, King VI, et al. A blinded, randomized, placebo-controlled trial of the efficacy and safety of the Janus kinase inhibitor oclacitinib (Apoquel®) in client-owned dogs with atopic dermatitis. Vet Dermatol 2013;24:587–97, e141–2.
[69]Little PR, King VL, Davis KR, Cosgrove SB, Stegemann MR. A blinded, randomized clinical trial comparing the efficacy and safety of oclacitinib and ciclosporin for the control of atopic dermatitis in client-owned dogs. Vet Dermatol 2015;26:23–30.
[70]Friesner RA, Banks JL, Murphy RB, Halgren TA, Klicic JJ, Mainz DT, et al. Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J Med Chem 2004;47:1739–49.
[71]Keystone EC, Taylor PC, Drescher E, Schlichting DE, Beattie SD, Berclaz PY, et al. Safety and efficacy of baricitinib at 24 weeks in patients with rheumatoid arthritis who have had an inadequate response to methotrexate. Ann Rheum Dis 2015;74:333–40.
[72]Genovese MC, Kremer J, Zamani O, Ludivico C, Krogulec M, Xie L, et al. Baricitinib in patients with refractory rheumatoid arthritis. N Engl J Med 2016;374:1243– 52.
[73]Papp KA, Menter MA, Raman M, Disch D, Schlichting DE, Gaich C, et al. A randomized phase 2b trial of baricitinib, an oral Janus kinase (JAK) 1/JAK2 inhibitor, in patients with moderate-to-severe psoriasis. Br J Dermatol 2016;174:1266–76.
[74]Verstovsek S, Mesa RA, Gotlib J, Levy RS, Gupta V, DiPersio JF, et al. A double- blind, placebo-controlled trial of ruxolitinib for myelofibrosis. N Engl J Med 2012;366:799–807.

[75]Harrison C, Kiladjian JJ, Al-Ali HK, Gisslinger H, Waltzman R, Stalbovskaya V, et al. JAK inhibition with ruxolitinib versus best available therapy for myelofibrosis. N Engl J Med 2012;366:787–98.
[76]Verstovsek S, Passamonti F, Rambaldi A, Barosi G, Rosen PJ, Rumi E, et al. A phase 2 study of ruxolitinib, an oral JAK1 and JAK2 Inhibitor, in patients with advanced polycythemia vera who are refractory or intolerant to hydroxyurea. Cancer 2014;120:513–20.
[77]Mahajan S, Hogan JK, Shlyakhter D, Oh L, Salituro FG, Farmer L, et al. VX-509 (decernotinib) is a potent and selective janus kinase 3 inhibitor that attenuates inflammation in animal models of autoimmune disease. J Pharmacol Exp Ther 2015;353:405–14.
[78]Genovese MC, Yang F, Østergaard M, Kinnman N. Efficacy of VX-509 (decernotinib) in combination with a disease-modifying antirheumatic drug in patients with rheumatoid arthritis: clinical and MRI findings. Ann Rheum Dis 2016 Apr 15. pii: annrheumdis-2015-208901. doi: 10.1136/annrheumdis-2015-208901. [Epub ahead of print]
[79]Takeuchi T, Tanaka Y, Iwasaki M, Ishikura H, Saeki S, Kaneko Y. Efficacy and safety of the oral Janus kinase inhibitor peficitinib (ASP015K) monotherapy in patients with moderate to severe rheumatoid arthritis in Japan: a 12-week, randomised, double- blind, placebo-controlled phase IIb study. Ann Rheum Dis 2016;75:1057–64.
[80]Namour F, Diderichsen PM, Cox E, Vayssière B, Van der Aa A, Tasset C, et al. Pharmacokinetics and pharmacokinetic/pharmacodynamic modeling of filgotinib

(glpg0634), a selective jak1 inhibitor, in support of phase iib dose selection. Clin Pharmacokinet 2015;54:859–74.
[81]Sherman W, Day T, Jacobson MP, Friesner RA, Farid R. Novel procedure for modeling ligand/receptor induced fit effects. J Med Chem 2006;49:534–53.
[82]Pardanani A, Gotlib JR, Jamieson C, Cortes JE, Talpaz M, Stone RM, et al. Safety and efficacy of TG101348, a selective JAK2 inhibitor, in myelofibrosis. J Clin Oncol 2011;29:789–96.
[83]Tyner JW, Bumm TG, Deininger J, Wood L, Aichberger KJ, Loriaux MM, et al. CYT387, a novel JAK2 inhibitor, induces hematologic responses and normalizes inflammatory cytokines in murine myeloproliferative neoplasms. Blood 2010;115:5232– 40.
[84]Pardanani A, Laborde RR, Lasho TL, Finke C, Begna K, Al-Kali A, et al. Safety and efficacy of CYT387, a JAK1 and JAK2 inhibitor, in myelofibrosis. Leukemia 2013;27:1322–7.
[85]Santos FP, Kantarjian HM, Jain N, Manshouri T, Thomas DA, Garcia-Manero G, et al. Phase 2 study of CEP-701, an orally available JAK2 inhibitor, in patients with primary or post-polycythemia vera/essential thrombocythemia myelofibrosis. Blood 2010;115:1131–6.
[86]Furqan M, Mukhi N, Lee B, Liu D. Dysregulation of JAK-STAT pathway in hematological malignancies and JAK inhibitors for clinical application. Biomark Res 2013;1:5.
[87]Chow V, Weissman A, O’Connell CL, Mehrvar A, Akhtari M. Emerging treatment options for myelofibrosis: focus on pacritinib. Onco Targets Ther 2016;9:2655–65.

[88]Shah K, Liu Y, Deirmengian C, Shokat KM. Engineering unnatural nucleotide specificity for Rous sarcoma virus tyrosine kinase to uniquely label its direct substrates. Proc Natl Acad Sci U S A 1997;94:3565–70.
[89]Liu Y, Shah K, Yang F, Witucki L, Shokat KM. A molecular gate which controls unnatural ATP analogue recognition by the tyrosine kinase v-Src. Bioorganic Med Chem 1998;6:1219–26.
[90]Roth BL, Sheffler DJ, Kroeze WK. Magic shotguns versus magic bullets: selectively non-selective drugs for mood disorders and schizophrenia. Nat Rev Drug Discov 2004;3:353–9.
[91]Cohen P, Alessi DR. Kinase drug discovery––what’s next in the field? ACS Chem Biol 2013;8:96–104.

Figure Captions

Fig. 1. The JAK-STAT signal transduction pathway.

Fig. 2. Domain architecture of the STAT and JAK families. (A) STAT domains. TAD, transcriptional activation domain. (B) Interaction of the STAT domains involved in dimer formation. (C) JAK architecture. The amino acid residue numbers correspond to human JAK3. pY, phosphotyrosine. A and B were adapted with permission from Ref. [8].

Fig. 3. (A) Structure of the active JAK1 JH1 protein kinase domain. (B) Mode of ATP binding to JAK1. (C) Structure of the inactive JAK1 JH2 pseudokinase domain. (D) Superposition of active and inactive protein kinase domains illustrating the differences in position of the αC helix and the activation segment (AS). Figures 3, 4, 5, 6, 9, and 10 were prepared using the PyMOL Molecular Graphics System Version 1.5.0.4 Schrödinger, LLC.

Fig. 4. Mechanism of the JAK1 protein kinase reaction illustrating the role of the catalytic aspartate (D1003). The phenolic oxygen of the tyrosyl residue attacks the γ- phosphate of ATP. JAK1 is derived from PDB ID: 3EYG; ATP and the metal ions are from PDB ID: 1ATP (murine PKA), and the peptide substrate is from PDB ID: 2GS6 (human EGFR). The carbon atoms of the peptide substrate are sky blue.

Fig. 5. (A) Frontal view and numbering system of the C- and R-spines. (B) Spines from the active JH1 domain of JAK1. (C) Spines from the inactive JH2 domain of JAK1. (D) Superposition of active and inactive spines. The carbon atoms of the spine residues in an active conformation are gray and those from an inactive conformation are blue.

Fig. 6. (A) Frontal view of the JH1 domain apposed to the inactive JH2 domain. (B) Top view of the JH1 domain depicting the interaction of the N-lobes. Components of JH2 are depicted with a prime mark. The asterisk at the end of the β4′-strand indicates the location of the JAK2 V617F activating mutation; in the case of TYK2 this residue is V678.

Fig. 7. (A) JAK intramolecular inactivation in cis. (B) JAK intermolecular inactivation in trans.

Fig. 8. Structures of selected JAK antagonists.

Fig. 9. Binding of selected drugs to the JAK family. The PDB IDs used to prepare the structures are given. Otherwise the Schrödinger Glide program (Version 2016-1) was used to generate the poses of the ligand docking sites [70].

Fig. 10. Non-conserved residues in the ATP-binding site listed in the amino to carboxyl direction. The carbon atoms have the following colors: JAK1, blue; JAK2, gray; JAK3, magenta; TYK2, green. The figure is a superposition of JAK1/2/3 and TYK2 PDB IDs 3EYG/3FUP/3LXK and 3LXN, respectively.

Table 1
Cytokine and growth factor stimulation of JAK-STAT signalinga
Cytokine/hormoneb Genes of human receptor subunits Downstream JAKs Downstream STATs Selected functions
Type I, common γ-chain (gene IL2RG) cytokines
IL-2 IL2RA/B-IL2RG JAK1/3 STAT3/5 Regulates T cell, B cell, and NK cell activities
IL-4 IL4RA-IL2RG JAK1/3 STAT6 Induces differentiation of helper T cells; anti- inflammatory action on T cells and monocytes
IL-7 IL7RA-IL2RG JAK1/3 STAT3/5 T cell development and homeostasis
IL-9 IL9R-IL2RG JAK1/3 STAT1/3/5 T cell growth and differentiation
IL-15 IL15RA-IL2RG JAK1/3 STAT3/5 Promotes T cell activation and proliferation
IL-21 IL21R-IL2RG JAK1/3 STAT1/3/5 Down regulates NK cell activation
Type I/common β-chain (gene CSF2RB) cytokines
GM-CSF CSF2RA-CSF2RB JAK2 STAT3/5 Growth of macrophages and granulocytes; stimulation and differentiation of stem cells; used to reverse neutropenia after chemotherapy
IL-3 IL3RA/B-CSF2RB JAK2 STAT3/5/6 Differentiation of stem cells; proliferation of all cells in the myeloid lineage
IL-5 IL5R-CSF2RB JAK2 STAT3/5/6 Stimulates B cell growth and immunoglobulin

secretion
Type I, gp130 (gene IL6RB = IL6ST) cytokines
IL-6 IL6RA/B-IL6RB JAK1/2, TYK2 STAT1/3 Prototypic pro-inflammatory cytokine increases acute-phase protein production
IL-11 IL11RA-IL6RB JAK1/2, TYK2 STAT3 Induces megakaryocyte colony formation and maturation
IL-27 IL27RA-IL6RB JAK1/2, TYK2 STAT1/2/3/4/5 Regulation of B and T cell activity
Type I, heterodimeric cytokines
IL-12 (35kD/40kD) IL12RB1-IL12RB2 JAK1,
TYK2 STAT4 Induces TH1 T helper cell formation
IL-23 IL23R-IL12RB1 JAK1,
TYK2 STAT3/4 Pro-inflammatory cytokine
Type I, hormone-like cytokines
Erythropoietin EPOR-EPOR JAK2 STAT5 Control of red blood cell production; used to treat anemia
Thrombopoietin TPOR-TPOR JAK2 STAT1/3/5 Differentiation of megakaryocytes and platelets
G-CSF CSF3R-CSF3R JAK2 STAT5 Production of stem cells and granulocytes; used to treat neutropenia
Growth hormone GHR-GHR JAK2 STAT3/5a Regulates post-natal body growth

Leptin LEPR-LEPR JAK2 STAT3/5a Coordination of energy metabolism; increases satiety
Type II, IFN family cytokines
IFN-α/β IFNAR1-IFNAR2 JAK1,
TYK2 STAT1/2/4 Promotes antiviral activity; stimulates T cell, macrophage, and NK cell activity; used to treat multiple sclerosis
IFN-γ IFNGR1-IFNGR2 JAK1,
TYK2 STAT1 Regulates macrophage and NK cell activation; used to treat chronic granulomatous disease and osteopetrosis
IL-28 IFNLR1-IL10RB JAK1,
TYK2 STAT1/2/3/4/5 Enhances immunity against infection
IL-29 IFNLR1-IL10RB JAK1,
TYK2 STAT1/2/3/4/5 Enhances immunity against infection
Type II, IL-10 family cytokines
IL-10 IL10RA-IL10RB JAK2,
TYK2 STAT3 Anti-inflammatory actions
IL-19 IL20RA-IL20RB JAK1/2, TYK2 STAT3 B-cell activation
IL-20 IL20RA-IL20RB JAK1/2, TYK2 STAT3 Regulates differentiation and proliferation of keratinocytes during inflammation

IL-22 IL22RA1-IL10RB JAK1/2, TYK2 STAT1/3/5 Targets non-hematopoietic cells such as hepatocytes, keratinocytes, lung, and intestinal epithelial cells
IL-24 IL20RA-IL20RB JAK1/2, TYK2 STAT1/3 Targets skin, lung, and reproductive cells
IL-26 IL10RB-IL20RA JAK1/2, TYK2 STAT1/3 Enhances IL-10 secretion from monocytes and IL-8 secretion from macrophages
Growth factors
EGF EGFR-EGFR JAK1 STAT1//3/5a Cell proliferation
PDGF PDGFRA/B- PDGFRA/B JAK1/2 STAT1/2/3/5/6 Stem cell proliferation
aAdapted from Ref. [5, 6].
bGM-CSF, granulocyte-macrophage colony stimulating factor; G-CSF, granulocyte colony stimulating factor.

Table 2

Human STAT properties

Gene Isoform Amino

acids MW,

kDa Uniprot ID pY Upstream enzyme
STAT1 α 750 87 P42224 701 JAK1, TYK2
STAT1 β 712 83 P42224 701 JAK1, TYK2
STAT2 I 851 98 P52630 690 JAK1/2, TYK2
STAT2 II 847 97 P52630 686 JAK1/2, TYK2
STAT3 I 770 88 P40763 705 JAK1/2, TYK2
STAT3 II 769 88 P40763 704 JAK1/2, TYK2
STAT3 III 722 83 P40763 705 JAK1/2, TYK2
STAT4 I 748 86 Q14765 693 JAK1, TYK2
STAT5 a 794 91 P42229 694 JAK1/2, TYK2
STAT5 b 787 90 P51692 682 JAK1/2, TYK2
STAT6 I 847 94 P42226 641 JAK1/2/3

STAT6 II 673 74 P42226 467 JAK1/2/3
STAT6 III 737 82 P42226 531 JAK1/2/3

Table 3
Important residues in the human Janus kinasesa
JAK1 JAK2 JAK3 TYK2 Inferred function Hanks no.b
N-lobe
Glycine-rich loop: GxGxΦG 882GEGHFG887 856GKGNFG861 829GKGNFG834 904GEGHFG909 Anchors ATP β-phosphate I
β3-K (K of K/E/D/D) K908 K882 K855 K930 Forms ion pair with ATP α- and β-phosphates II
αC-E (E of K/E/D/D) E925 E898 E871 E947 Forms ion pair with β3-K III
Hinge residues 957EFLPG961 931YKPYG936 903EYLPSG908 979EYVPLG984 Connects N- and C-lobes V
C-lobe
αE-AS loop and AS HΦ-interaction Y999-A1027 Y972-V1000 C945-L973 Y1019-A1047 Stabilizes AS VIb-VII
Catalytic loop HRD (first D of K/E/D/D) D1003 D976 D949 D1023 Catalytic base (abstracts proton) VIb
Catalytic loop Asn (N) N1008 N981 N954 N1028 Chelates Mg2+(2) VIb
Activation segment 1021–1051 994-1024 967-997 1041-1071 Positions protein substrate VII-
VIII
AS DFG (second D of D1021 D994 D967 D1041 Chelates Mg2+(1) VII

K/E/D/D)
AS phosphorylation site Y1034/Y1035 Y1007/Y1008 Y980/Y981 Y1054/Y1055 Stabilizes the AS after phosphorylation VIII
APE, end of AS 1049–1051 1022-1024 995-997 1069-1071 Interacts with the αHI loop and stabilizes the AS VIII
UniProt KB ID P23458 O60674 P52333 P29597
aAS, activation segment bFrom Ref. [16]

Table 4
Spine and shell residues of murine PKA and human Janus kinase and pseudokinase (Ψ) domains
Symbol PKAa JAK1 Ψ JAK1 JAK2 Ψ JAK2 JAK3 Ψ JAK3 TYK2 Ψ TYK2
Regulatory spine
β4-strand (N-lobe) RS4 L106 L653 Y940 N612 Y913 L587 Y886 V673 Y962
C-helix (N-lobe) RS3 L95 M641 L929 M600 L902 L575 L875 L661 L951
Activation loop F of DFG (C-lobe) RS2 F185 P740 F1022 P700 F995 P672 F968 P760 F1042
Catalytic loop His/Tyr (C- lobe) RS1 Y164 H712 H1001 H671 H974 H645 H947 H732 H1021
F-helix (C-lobe) RS0 D220 D775 D1063 D735 D1036 D707 D1009 D796 D1083
R-shell
Two residues upstream from the gatekeeper Sh3 M118 M665 L954 L624 L927 M598 L900 M685 L976
Gatekeeper, end of β5-strand Sh2 M120 E667 M956 Q626 M929 Q600 M902 T687 M978
αC-β4 loop Sh1 V104 V651 V938 V610 V911 V585 V884 A671 I960
Catalytic spine
β2-strand (N-lobe) CS8 V57 I597 V889 I559 V863 I535 V863 V603 V911
β3-AxK motif (N-lobe) CS7 A70 I620 A906 L579 A880 L554 A853 V640 A928
β7-strand (C-lobe) CS6 L173 L721 L1010 L680 L983 L654 L956 L741 L1030

β7-strand (C-lobe) CS5 I174 L722 V1011 L681 V984 L655 V057 L742 L1031
β7-strand (C-lobe) CS4 L172 L720 V1009 I679 I982 V653 I955 I740 V1029
D-helix (C-lobe) CS3 M128 L675 L964 L634 L937 I608 L910 L695 L986
F-helix (C-lobe) CS2 L227 I782 T1070 T742 V1043 T714 S1012 I807 T1090
F-helix (C-lobe) CS1 M231 I786 L1074 I746 L1047 V718 V1016 T803 L1094
a From Ref. [20, 21].

Table 5
Janus kinase inhibitor EC50 values (nM)a
Drug JAK1 JAK2 JAK3 TYK2
First generation
Tofacitinibb 0.16 0.58 1.6 4.8
Oclacitinibc 10 18 99 84
Baricitinibd 4 6.6 259 21.1
Ruxolitinibd 0.09 0.036 2 0.4
Second generation
Decernotinibe 11 13 2 11
Peficitinibf 3.9 5.0 0.71 4.8
Filgotinibg 10 2.8 81 11.6
Fedratinibb 18 1.1 ? ?
Momelotinibd 11 18 155 17
Lestaurtinibd 8.8 3.7 2.3 15
Gandotinibh 25 3 60 ?
Pacritinibi 1280 23 520 50
aEC50 (effective concentration) refers to IC50, Kd, or Ki. bRef. [60].

cRef. [61]. dpubchem.ncbi.nlm.nih.gov eRef. [62].
fRef. [63]. gRef. [64]. hRef. [65]. iRef. [66].

Table 6
Properties of small molecule Janus kinase inhibitors in clinical trialsa, b
Name, code, trade name® Targets PubChem CID Formula MW D/Ac cLogPd Indications and clinical trials
First generation
Tofacitinib, CP690550, Xeljanz® JAK1/2/3 9926791 C16H20N6O 312.37 1/7 1.076 RAe, psoriasis, alopacia areata, atopic eczema, spondyloarthritis, systemic lupus, ulcerative colitis, acute host-vs.- graft disease
Oclacitinib, OF03394197, Apoquel® JAK1/2 44631938 C15H23N5O2S 337.44 2/6 1.528 Canine allergic dermatitisf
Baricitinib, INCB28050, LY3009104 JAK1/2 44205240 C16H17N7O2S 371.42 1/7 0.330 RA, psoriasis, autoinflammatory disease
Ruxolitinib, INC424, Jakafi® JAK1/2 25126798 C17H18N6 306.37 1/6 1.967 Myelofibrosisg, polycythemia verag, ALL, AML, CLL, CML, NSCLC, breast, colorectal, head and neck, prostate, and pancreatic cancers, RA, psoriasis

Second generation
Decernotinib, VX509 JAK3 59422203 C18H19F3N6O 392.38 3/8 2.021 RA
Peficitinib, ASP015K JAK3 57928403 C18H22N4O2 326.39 4/4 2.046 Psoriasis, RA, ulcerative colitis
Filgotinib, GLPG0634 JAK1/2 49831257 C21H23N5O3S 425.50 1/6 1.958 RA, Crohn disease
Fedratinib, SAR302503, TG101348 JAK2 16722836 C27H36N6O3S 524.68 3/9 4.934 Myelofibrosis, polycythemia vera, primary thrombocythemia
Momelitinib, Cyt387 JAK1/2 25062766 C23H22N6O2 414.46 2/7 2.352 Myelofibrosis, polycythemia vera, NSCLC, pancreatic carcinoma
Lestaurtinib, CEP-701 JAK2,
FLT3, TRKA/B/C 126565 C26H21N3O4 439.36 3/4 2.816 Myelofibrosis, psoriasis, polycythemia vera, ALL, AML, prostate cancer, multiple myeloma, neuroblastoma, Hodgkin lymphoma
Gandotinib, LY2784544 JAK2 46213929 C23H25ClFN7O 469.94 2/7 3.661 Myeloproliferative disorders
Pacritinib, SB1518 JAK2 46216796 C28H32N4O3 472.58 1/7 4.499 Myelodysplastic syndromes, myelofibrosis, AML, CLL, NSCLC,

colorectal cancer
aclinicaltrials.gov
bALL, acute lymphoblastic leukemia; AML, acute myelogenous leukemia; CLL, chronic lymphocytic leukemia; CML, chronic myelogenous leukemia; NSCLC, non-small cell lung cancer; RA, rheumatoid arthritis.
bNo. of hydrogen bond donors/acceptors.
dCalculated log of the partition coefficient as determined by MedChem Designer® v.1.0.1.15. eFDA-approved treatment in 2012.
fFDA-approved treatment in 2013. gFDA-approved treatment in 2011.

Table 7

Non-conserved residues near the ATP-binding site of Janus kinases

Locationa JAK1 JAK2 JAK3 TYK2
β1-strand Arg879 Gln853 Ser826 Arg901
G-rich loop His885 Asn859 Asn832 His907
Hinge 1 Phe958 Tyr931 Tyr904 Tyr980
Hinge 2 Ser961 Tyr934 Ser907 Leu983
Hinge 3 Ser963 Ser936 Cys909 Ser985
αDE loop Pro969 Gln942 Gln915 Pro991
Before DFG Gly1020 Gly993 Ala966 Gly1040
Catalytic loop Ala1005 Ala978 Ala951 Ala1025
After catalytic loop Val1009 Ile982 Ile955 Val1029
aDepicted in Fig. 10.