Characterization of JAK1 Pseudokinase Domain in Cytokine Signaling

cancers Article Characterization of JAK1 Pseudokinase Domain in Cytokine Signaling Juuli Raivola 1, Teemu Haikarainen 1and Olli Silvennoinen 1,2,3,* 1Faculty of Medicine and Life Sciences, Tampere University, 33014 Tampere, Finland; [email protected] (J.R.); [email protected] (T.H.) 2Institute of Biotechnology, Helsinki Institute of Life Science HiLIFE, University of Helsinki, 00014 Helsinki, Finland 3Fimlab Laboratories, Fimlab, 33520 Tampere, Finland *Correspondence: [email protected] Received: 30 September 2019; Accepted: 23 December 2019; Published: 27 December 2019   Abstract: The Janus kinase-signal transducer and activator of transcription protein (JAK-STAT) pathway mediates essential biological functions from immune responses to haematopoiesis. Deregulated JAK-STAT signaling causes myeloproliferative neoplasms, leukaemia, and lymphomas, as well as autoimmune diseases. Thereby JAKs have gained significant relevance as therapeutic targets. However, there is still a clinical need for better JAK inhibitors and novel strategies targeting regions outside the conserved kinase domain have gained interest. In-depth knowledge about the molecular details of JAK activation is required. For example, whether the function and regulation between receptors is conserved remains an open question. We used JAK-deficient cell-lines and structure-based mutagenesis to study the function of JAK1 and its pseudokinase domain (JH2) in cytokine signaling pathways that employ JAK1 with different JAK heterodimerization partner. In interleukin-2 (IL-2)-induced STAT5 activation JAK1 was dominant over JAK3 but in interferonγ (IFN γ ) and interferonα (IFN α ) signaling both JAK1 and heteromeric partner JAK2 or TYK2 were both indispensable for STAT1 activation. Moreover, IL-2 signaling was strictly dependent on both JAK1 JH1 and JH2 but in IFN γ signaling JAK1 JH2 rather than kinase activity was required for STAT1 activation. To investigate the regulatory function, we focused on two allosteric regions in JAK1 JH2, the ATP-binding pocket and the α C-helix. Mutating L633 at the α C reduced basal and cytokine induced activation of STAT in both JAK1 wild-type (WT) and constitutively activated mutant backgrounds. Moreover, biochemical characterization and comparison of JH2s let us depict differences in the JH2 ATP-binding and strengthen the hypothesis that de-stabilization of the domain disturbs the regulatory JH1-JH2 interaction. Collectively, our results bring mechanistic understanding about the function of JAK1 in different receptor complexes that likely have relevance for the design of specific JAK modulators. Keywords: JAK; STAT; cytokine; cytokine receptor; cancer; inflammation 1. Introduction Janus kinases (JAK1–3 and Tyrosine Kinase 2, TYK2) are non-receptor tyrosine kinases that play a critical role in cell signaling via type I/II cytokines and interferons (IFNs) [ 1 ]. JAKs are constitutively bound to the intracellular part of the receptors that dimerize after ligand binding, thus enabling the transphophorylation and activation of JAKs and subsequent phosphorylation and translocation of STATs (signal transducer and activator of transcription proteins) into nucleus to regulate transcription. Structurally all JAKs consist of four domains; the N-terminal FERM (4.1-band, ezrin, radexin, moiesin) domain that together with the SH2-like domain (SH2) compose the main receptor binding moiety, Cancers 2020,12, 78; doi:10.3390/cancers12010078 www.mdpi.com/journal/cancers Cancers 2020,12, 78 2 of 20 followed by the C-terminal JAK-homologue (JH) 2 and JH1 domains. JH1 is an active tyrosine kinase while JH2 has an important role in regulating both basal and cytokine-induced activation [ 2 – 5 ]. Some residues required for the phosphotransferase reaction from ATP to a substrate are not conserved in JH2, such as altered Gly-rich region and lack of catalytic Asp, hence giving the domain “pseudokinase domain” designation [ 6 ]. Despite the absent or low catalytic activity, JH2s regulates JAK activity through allosteric mechanisms. For example, JH2 forms an autoinhibitory interaction with JH1 by binding to the hinge region and stabilizing the inactive conformation of JH1 [ 7 ]. Furthermore, the ATP binding to JH2 stabilizes the domain and nucleotide binding is critical for pathogenic activation of JAK2 [ 8 , 9 ]. In TYK2, a small ATP pocket-binding compound has been shown to efficiently decrease the activation [ 10 ]. In addition, the SH2-JH2 linker is an important regulative region that controls the activity of JAKs [4]. JAK1, as well as other JAK family members, harbor pathogenic gain-of-function mutations (GOFs) that cause acute lymphoid and -myeloid leukemia (ALL and AML, respectively). JAK1 mutations are found in varying types of cancers; e.g., 9% of hepatocellular carcinoma patients have been found to have JAK1 mutations [ 10 ]. In addition to GOFs, JAK loss-of-function mutations (LOFs) have been identified in JAK3, which cause severe combined immune deficiency (SCID); a disease resulting a depletion of B-cells and complete loss of Tand NK-cells [ 11 , 12 ]. A majority of the pathogenic mutants clusters in JH2 highlighting the regulative role of the domain [ 10 ]. The most common mutation, JAK2 V617F also resides in JH2. JAK2 V617F and accounts for ~95% of patients with polycythaemia vera and about 50% of patients with essential thrombocytosis and primary myelofibrosis [ 13 – 15 ]. Mechanistically, the mutation stabilizes JAKs the α C-helix ( α C) in the N-lobe of JH2, and induces cytokine independent dimerization of the receptors, possibly via JH2-dimerization [ 16 , 17 ]. The α C resides in the JH1-JH2 interface but also lines the ATP-binding pocket. The region is important in cytokine-induced activation of kinases, and modulating it with mutations can inhibit constitutive activation of JAKs [ 17 – 19 ]. However, the mechanisms of function for many mutations is yet unknown. Homologous mutation to JAK2 V617F in JAK1 (V658F) causes acute lymphoblastic leukemia [ 20 ]. In addition to the JAK2 V617F and its homologues, a distinctive cluster of GOFs resides in the JH1-JH2 interface and these mutations disrupt the autoinhibitory interaction between the domains. For example, JAK1 R724E has one of the highest incidence rates among the JAK1 pathologic mutations. It resides in the N-lobe of JH2 (similarly to homologous JAK2 R683S) and interacts with JH1. Other JAK1 mutants in the JH1-JH2 interface are A634D and L653F [21,22]. Currently there are three JAK JH1-targeted inhibitors in clinical use, but recently targeting JAK JH2s with small molecular compounds has gained interest as a potential treatment for constitutively active cytokine signaling [ 10 , 23 , 24 ]. The unique mode of ATP binding in JH2 may allow increased specificity over other eukaryotic protein kinases [25]. Although the general principles of the JAK-STAT signaling are well established, the underlying differences between the function of JAKs and their individual domains in different cytokine signaling pathways are not fully defined. Moreover, the unbalanced number of JAKs (four) versus the cytokine pathways they transduce (over fifty) raises the question about the mechanisms that allow the versatile function of JAKs. For example, JAK1 is employed by the IL-2, IL-4, IL-10, and gp130 (including e.g., IL-6) receptor families as well as type I and type II interferons (e.g., IFN α and IFN γ , respectively). Intrigued by these questions, we set to study the role of JAK1 and particularly its JH2 in various signaling pathways. Our results show that JAK1 has varying roles in different receptor complexes, and that JH2 mediates important allosteric regulation of the JAK activity. 2. Results 2.1. JAK1 Is Dominant STAT Activator in IL-2, but not in IFNγand IFNαSignaling To investigate the role of individual JAKs in heterodimeric receptor complexes, we focused on IL-2, IFN γ , and IFN α receptor systems. These pathways utilize JAK1 but consists of different JAK Cancers 2020,12, 78 3 of 20 dimers: IL-2-signaling is driven by JAK1-JAK3, IFNγby JAK1-JAK2, and IFNαby JAK1-TYK2. First, we studied how dependent the signaling systems are to the presence of an individual JAK. For this, we used cell lines that lack specific JAK expression. JAK1 and JAK3 deficient U4C γβ -, JAK2 deficient γ 2Aor TYK2 deficient 11.1 human fibroblast cells were transiently transfected with one or both JAKs relevant to the pathway, and the expression of JAKs and STATs was detected from the cell lysates. In addition, the phosphorylation of STAT1 (pSTAT1) and STAT5 (pSTAT5) was assayed. Of note, HA-STAT5 was transfected into the U4C γβ cells, since endogenous STAT5 could not be detected with this method. All of the studied cytokine receptor pathways required the expression of two different wild-type (WT) JAKs for the cytokine-dependent activation of STAT (Figure S1A,B). In IFN γ and IFN α systems, STAT1 activation required the expression of both JAK1 and JAK2 or TYK2, respectively. In the IL-2 receptor complex both JAK1 and JAK3 were also required for cytokine dependent signaling; JAK1 alone activated STAT5 but the activation was unresponsive to cytokine while JAK3 could not induce STAT5 phosphorylation (even with IL-2) without the presence of JAK1 (Figure S1C). Together these results confirm the previous findings that cytokine induced signaling requires heterodimerization of JAK WT [26,27] as well as validated the experimental system. Next, we studied the JAK dependency of the three signaling systems in the context of hyperactive mutations. JAK2 V617F mutant or homologous mutations in JAK1 and TYK2 (see Table 1) were transfected in JAK1 deficient cells either alone or with the relevant partner JAK. Interestingly, we observed that unlike in the WT-setting, JAK2 V617F and homologous mutations in JAK1 and TYK2 could induce (reduced) pSTAT1 even in the absence of the partnering JAK. However, the presence of the partner JAK increased the activation (Figure 1A). JAK1 V658F also activated STAT5 independently of JAK3 in the IL-2 system (Figure 1A [ 9 ]). As JAK1, JAK2 and TYK2 are roughly same size (~130 kDa) they appear as a single band in the western blot (with less intensive band when only one JAK is transfected with the vector). Table 1. Mutations used in this study qualified as loss-of-function mutations (LOFs) or gain-of-function mutations (GOFs) based on the shown effects (-, designates as neutral). JAK Mutation Effect Short Description. JAK1 L633K LOF At the solvent exposed face of the JH2 αC-helix, homologous to the JAK2 E592R. I597F GOF/-Residing JH2 ATP-binding site and designed to inhibit ATP binding. Homologous to JAK2 I559F. K622A LOF Removes conserved β3 lysine in JH2. Designed to inhibit ATP binding. Homologous mutations shown to inhibit hyperactivation in JAK2 and JAK3. V658F GOF Homologous to JAK2 V617F and TYK2 V678FF. Resides in the JH2 β4-β5 loop and potentially disturbs the SH2-JH2 linker and causes cytokine independent activation. Mutation in JAK1 or JAK2 cause ALL. JAK2 E592R LOF At the solvent exposed face of the JH2 αC-helix, shown to inhibit JAK2 V617F-driven dimerization of EPOR [19]. I559F LOF In the JH2 β2. Designed to sterically inhibit ATP binding and shown to inhibit ATP binding in recombinant JH2 [8]. V617F GOF Homologous to JAK1 V658F and TYK2 V678F. TYK2 L653R LOF At the solvent exposed face of the JH2 αC-helix, homologous to the JAK2 E592R. V603F LOF At the ATP, binding pocket of JH2, designed to inhibit ATP binding. Homologous to JAK2 I559F. V678F GOF Constitutive active mutation in JH2. Homologous to JAK1 V658F and JAK1 V617F. Cancers 2020,12, 78 4 of 20 Table 1. Cont. JAK Mutation Effect Short Description. JAK3 E567R LOF Resides at the solvent exposed face of the JH2 α C-helix. Homologous to the JAK2 E592R. I535F LOF At the ATP binding pocket of JH2 and homologous to JAK2 I559F.Shown to inhibit constitutive JAK3 activation [9]. R657Q GOF Activating mutation found in ALL patient. Resides in the JH1-JH2 interface. M592F GOF Homologous to constitutively active JAK1 V658F, JAK2 V617F and TYK2 V678F. L570F - Mutation designed to create the WT state as in JAK1, JAK2 and TYK2 (F595 in JAK2). Stacks with the mutated JAK2 V617F and enables the hyperactivation via the FFV-triad formation (see text). M592F +L570F GOF Double mutant designed to create a complete FFV-triad into JAK3 (see above). In contrast, the pathogenic JAK3 R657Q residing in the JH2-JH1 interface was strongly dependent on the presence of JAK1 WT, shown as abolished STAT5 activation when JAK1 was not transfected into the U4C γβ cells (Figure 1B [ 9 ]). Of note, equal amounts of DNA for each JAK were transfected although the JAK1 HA-signal is consistently weaker than JAK3 HA. As we have previously noticed that overexpression of JAK1 quickly increases the activation status of the basal STAT5 [ 9 ], and to keep the experimental set-up reliable, we did not want to increase the expression of JAK1 despite the weaker bands obtained with the immune labeling. In addition, the proper expression of JAK1 is evident also from the IL-2 responsive STAT5 activation, which cannot be induced without JAK1. JAK3 R657Q was chosen as a representative activating JAK3 JH2 mutation because JAK3 lacks the residue homologous to JAK2 V617 that is present in other JAK JH2s (see Table 1). Furthermore, JAK3 has a leucine in the JH2 α C that in all other JAKs is occupied by a phenylalanine (Phe) (Table 1). This Phe is part of the so-called FFV-triad (Phe-Phe-Val) that is important for JAK2 V617F activation. The residue stack with the V617F, rigidify the α C and interacts with the SH2-JH2 linker via (JAK2) Phe 575, as well as alters the ATP-site cleft, leading to hyperactivation [ 28 – 30 ]. We were interested whether the difference in JAK3 FFV-triad (or the usage of non-homologous activating mutation) is causing the lack of STAT-activation in the absence of JAK1. Hence, we reconstituted the FFV-triad by double-mutating JAK3 at M592F (mimicking the JAK2 V617F) and at L570F to introduce the phenylalanine present in JAK1 WT (Phe F636) and JAK2 (Phe 595) (Figure 1C,D). The JAK3 M592F +L570F caused constitutive activation of STAT5 in the presence of JAK1 WT, but not in the absence of JAK1 WT. The single mutants were also analyzed for their STAT5-activation potential, and while JAK3 M592F showed some hyperactivation, L570F did not affect the STAT5 activation. The unaltered STAT activation in response to L570F was expected, as it simulates the WT-situation in JAK1 and JAK2. These results suggest that the dominance of JAK1 in the IL-2 system is not an intrinsic feature of JAK3, but rather a property of the IL2R complex. Next, we compared the roles of JAK1 JH1 and JH2 in the IL-2 and IFN γ signaling. U4C γβ cells were transiently transfected with JAK1 constructs where either JH1 or JH2 was deleted (Figure 1E). We have previously shown that in IL-2 signaling both JAK1 JH1 and JH2 are crucial for STAT5 activation (Figure 1E) [ 9 ]. However, in the IFN γ system JH1 deleted JAK1 maintained IFN γ responsiveness albeit the inductivity was considerably lowered, while JH2 was critical for any cytokine responsiveness (Figure 1E). This observation is in line with the study of Eletto et al., where JAK1 JH2, but not JH1, was found to be essential in IFNγsignaling, while both JH1 and JH2 were crucial for IFNαsignaling [31]. Based on these results, the role of JAK1, and its JH2, is different between IL-2 and IFN γ (or IFN α ) systems. In IL-2 signaling, JAK1 dominantly mediated the STAT5 phosphorylation and the activation requires both JH1 and JH2 domains. In the IFN γ and IFN α signaling, STAT1 activation requires both JAK1 and JAK2/TYK2, and IFN γ signaling shows to be dependent specifically on the pseudokinase domain of JAK1. Cancers 2020,12, 78 5 of 20 Figure 1. Comparison of JAK JH2s. ( a ) Janus kinases (JAK1, JAK2 and TYK2) hyperactive mutants can activate STAT1 without the partner JAK. pSTAT1 analysis and the transcriptional activity of STAT1 were detected from U4C γβ and γ 2A cells that were transfected for 24 h with JAK2 V617F, TYK2 V678F or JAK1 V658F. Wild-type (WT) partner JAK or vector was co-transfected with the activating mutants. Cells were starved overnight, and left untreated or stimulated with 100 ng/mL interferon γ (IFN γ ) and interferon α (IFN α ) for 20 min after which the pSTAT1 was detected by immunolabeling. For the STAT1 transcriptional activity detection, IRF-GAS or ISRE-luc plasmids) were co-transfected for 43 h with renilla plasmid (pRL-TK) (see Materials and Methods). Cells were stimulated or starved for 5 h, and the luciferase activity was measured. The values were divided with the renilla values to reduce the effects the possible differences in the transfection efficiency might have. Errors are SD of triplicates. Below are representative immunoblots of whole-cell lysates from U4C γβ and γ 2A cells transiently transfected with full-length hemagglutinin (HA)-tagged JAK mutants with or without JAK WT, as indicated. The cell lysates were immunolabeled with pSTAT1 (STAT1 Y701 phosphorylation), HA and Cancers 2020,12, 78 6 of 20 STAT1 antibodies. The experiment was repeated twice with similar results. ( b ) JAK3 R657Q cannot induce STAT5 activation without JAK1. U4C γβ cells were transfected with JAK WT or hyperactive JAK1 V658F or JAK3 R658Q and left untreated or stimulated with IL-2 (100 ng/mL) for 5 h. STAT5 specific SPI-Luc 2 plasmid was used for the detection of STAT5 transcriptional activity and the pRL-TK was used as a control. Errors are SD of triplicates. Below: Whole cell lysates from transiently transfected U4C γβ cells were labelled with pSTAT5 (phosphorylation at Y694) and HA antibodies. HA-tagged STAT5 was transfected with JAK-HA constructs. Blot is a representative from three experiments. ( c ) Reconstituted JAK2 V617F homolog in JAK3 cannot signal without JAK1. JAK3 M592F, L570F, and double mutant were transiently transfected with JAK1 WT or vector. The U4C γβ cells were starved and/or stimulated with 100 ng/mL IL-2, and the activity of the STAT5 responsive SPI-Luc vector measured as described above. (d) Illustration of JAK1 V658, F636 and homologous mutations in JAK3 pseudokinase domain (JH2) with schematic presentation of the four domains and approximate location of the JH2 mutations. Also the amino acid range for domains (according to the UniProt database) are shown in the scheme below each domain. Structures were visualized with PyMol using JAK1 JH2 structure (PDB 4L00) and JAK3 JH2 homology models (modelled based on TYK2 structure; Protein data bank 4OLI). ( e ) JAK1 JH2 is critical for IL-2 and IFN γ signaling. U4C γβ cells were transfected with full-length JAK1 or with JAK1 JH1 or JH2 deletions. STAT5 and STAT1 responsive Luc-vectors were used as described before to detect the IL-2 and IFN γ responsiveness of the constructs. Errors are SD of triplicates, and p-values according to two-tailed student t-test (*—indicating p<0.05 and **—p<0.001). Expression of the HA-tagged, unstimulated JAK1 (and JAK3 in the IL-2 system) was confirmed by immunolabeling the whole cell lysates with HA-antibody. The band below the JAK1 WT and JAK3 bands in the left side panel WT/WT sample is due unspecific binding of the HA antibody. 2.2. L633K in the Outer Face of JAK1 JH2 aC-helix Inhibits WT and Hyperactive IL-2, IFNγ, and IFNα Signaling at Variable Degrees We continued the analysis of JAK1 JH2 with available information from other JAK JH2s and focused on regions that have been identified to allosterically regulate JAK activation. We focused on the hydrophilic outer face of the JH2 α C and introduced a mutation in JAK1 that corresponds to JAK2 E592R (Figure 2A). JAK2 E592R inhibits hyperactivation in JAK2 and reduces V617F-driven dimerization [17]. α C-helix is an essential conserved region in protein kinases, and in the active kinase conformation the N-terminus of the α C-helix typically interacts with the activation loop phosphate. Furthermore, the conserved β 3 Lys 72 (numbering based on Protein Kinase A, PKA) couples the ATP phosphates to the α C-helix [ 25 ]. The clinical relevance of this region in pathogenic JAK1 signaling was evaluated by searching for patient-derived mutations that cluster in the area. Based on the COSMIC database [ 32 ] together with a literary research by Hammar é n et al., the JAK1 JH2 α C-helix and the surrounding α B and β 4 linkers were found to be highly mutated in human cancers. Based on mutations from Hammar é n et al. [ 10 ], 11 out of the 29 residues (38%) were mutated and with all COSMIC mutations included, total of 16 mutations could be depicted in this region (Figure 2A, mutated residues shown as grey-shaded, bolded letters in the sequence). To test the effect of the α C modulation in cell-based assays, JAK1 L633K was transfected into U4C γβ cells with STAT1 or STAT5 responsive luciferase reporters and the normalized luciferase values were compared to cells transfected with JAK1 WT. L633K reduced the cytokine responsiveness in JAK1-JAK3 driven IL-2 signaling and JAK1-TYK2 driven IFN α signaling (Figure 2B). In addition, the basal STAT5 activity in L633K transfected cells was reduced compared to WT. On the contrary, JAK1 L633K did not markedly affect the IFN γ responsiveness (driven by JAK1-JAK2), but slightly reduced the basal STAT1 activity. Basal STAT1 activity profile correlated with the reduced pSTAT1 in L633K transfected cells (Figure S2A,B). Homologous mutations JAK2 E592R and TYK2 L653R also reduced the cytokine responsiveness as measured with STAT1 responsive IRF-GASand ISRE-luciferase reporters (Figure S2C). Cancers 2020,12, 78 7 of 20 Figure 2. Analysis of the JAK1 JH2 α C-mutation. ( a ) JAK1 JH2 (PDB 4L00) structure is shown with the L633 in the α C-helix. The sequence around JAK1 L633 in JAK family is shown. Right: Mutations in the JAK1 α C-region including the adjacent α B-helix and the β 4-linker. The mutations (shaded residues with bold letters) are derived from the review of Hammar é n et al. 2018 [ 17 ] and the Catalogue of Somatic Mutations in Cancer (COSMIC)-database [ 32 ]. Deletions are shown as arrows. ( b ) Mutation in the JAK1 JH2 α C inhibits JAK1 driven IL-2 and IFN α signaling, but has lesser effect in the IFN γ -induced signaling. JAK1 WT (shown as blue dots) or JAK1 L633K (red dots) were transiently transfected with STAT1 and STAT5-responsive luciferase plasmids as described previously. The U4C γβ cells were then starved or stimulated with increasing amounts of cytokine, and the STAT-activation was detected. All 12 replicas are presented as dots in a logarithmic axis showing the cytokine amount versus the relative luciferase activity. Basal state is set to −1. (c) JAK1 L633K and homologous mutation in JAK3 JH2 α C reduces hyperactivation in cis. JAK1 and JAK3 mutations were studied with (100 ng/mL) and without IL-2 stimulation in U4C γβ cells that lack JAK1 and JAK3. In comparison to the activating mutations, the double mutants had reduced basal activity and responded to cytokine stimulation similarly as JAK WT transfected cells. STAT5 activity was studied with SPI-Luc luciferase system as described previously. The errors are SD of two separate experiments both having triplicate samples (n =6). Two-tailed t-test was performed and **—indicates p-value <0.001. Cancers 2020,12, 78 8 of 20 Since JAK1 L633K reduced IL-2-driven STAT activation, we were interested whether the mutation can inhibit constitutively activated JAKs in cis, as shown with JAK2 [ 17 ]. We created double mutants of JAK1 and JAK3, where the JAK1 L633K was combined with hyperactivating JAK1 V658F. In addition, similar mutations in JAK3 JH2 were tested (JAK3 E567R +R658Q, see Table 1). Both double mutants reduced the hyperactivation back to WT levels but retained the IL-2 inductivity (Figure 2C). Similar reduction of hyperactivation has been shown with homologous mutations in JAK2 [ 8 , 17 ] and were seen with TYK2 (Figure S2C). Of note, more fluctuation in the STAT activation levels was observed in the JAK1 L633K +V658F double mutant, shown as larger errors seen in Figure 2C. However, the reduction from JAK1 V658F transfected cells was significant (p<0.001). 2.3. Characterization of ATP Binding to JAK1 JH2 Next, we set to compare the inhibitory potential between the α C-mutant and another allosteric region of JH2, namely the ATP-binding site. First, we showed that in addition to JAK2 I559F and JAK3 I535F mutations that have previously been shown to inhibit ATP binding and JAK hyperactivation, [ 8 , 9 ] also homologous TYK2 V603F inhibits hyperactive TYK2 V678F in the IFN α system (Table 1, Figure S2D). The mutation was originally designed to create steric hindrance in the pocket and have been veritably shown to inhibit ATP binding into JAK2 JH2 [ 8 ]. We introduced a mutation in JAK1 JH2 ATP-site, JAK1 I597F, which is homologous to the above-mentioned JAK mutants. In addition, another ATP site mutant, JAK1 K622A was chosen as its homolog has been shown to inhibit JAK2 and JAK3 hyperactivation in cis [ 8 , 9 ]. This highly conserved lysine (Lys72 in PKA) is critical in making a salt bridge to the conserved Glu (91 in PKA) in the α C, and is required for coordinating nucleotide binding of multiple kinases and pseudokinases [33]. We have previously noted that JAK1 I597F is unable to inhibit hyperactive IL-2 signaling, contrasting the effect of the homologous mutants in JAK2 and JAK3 [ 8 , 9 ]. Here, we found that JAK1 I597F increased basal STAT5 activity and pSTAT5 in WT background, although to a lesser extent than hyperactive JAK1 and JAK3 mutants (Figure 3A,B). Furthermore, the IL-2 induction was disturbed in comparison to JAK1 WT, and although some increase was apparent in the STAT5 transcriptional activity assay, JAK1 I597F could not significantly respond to IL-2 addition (p=0.12 between the basal vs. IL-2, 50 ng/mL). The pSTAT5 analysis of the mutant showed more variability, but also in this setting both the increased basal activity and the disturbed cytokine responsiveness were detected (Figure 3A,B). Mutation of the conserved lysine K622 in the JAK1 JH2 ATP-binding site (Table 1) to alanine reduced the cytokine induced STAT activation, thus correlating with the behavior of the JAK2 [ 8 ] and JAK3 homologs (Figure 3B). To decipher the cause for the untypical behavior of the JAK1 JH2 ATP-site mutant I597F, we produced and purified JAK1 JH2s with I597F or K622A mutation. GST-tagged proteins were bound to glutathione sepharose, eluted by digestion with Tobacco Etch Virus TEV Protease and further purified in size-exclusion chromatography (SEC) (Figure 3C). The JAK1 JH2 mutants were then analyzed in differential scanning fluorimetry (DSF) with and without ATP. JAK1, JAK2, and TYK2 JH2s have been shown to bind ATP in the presence of divalent cations while JAK3 JH2 binds ATP without cations [ 8 , 9 ] and thus proteins were analyzed with and without MgCl 2 . Figure 3C shows the melting temperatures (Tm), as well as the change in the melting temperatures (dTm) relative to the WT JH2 apo-form (protein that does not bind any ligands). As expected, ATP binding did not increase the Tm in K622A. However, the mutant showed significantly increased thermal stability. Compared to the WT JH2 Tm (44 ◦ C), the I597F mutant showed reduced Tm of 40 ◦ C, and addition of 1 mM ATP and 1 mM MgCl2 increased the melting temperature by 2 ◦ C. The increase was only slightly differed from the observed 3 ◦ C increase in WT JH1. The Tm for apo-form WT, I597F, and K622A were 44.3 ◦C, 40.3 ◦C, and 53.3 ◦C, respectively. Taken together, our cell-based and recombinant protein assays suggest that the JAK1 JH2 ATP-site differs from other JAKs in that the I597F cannot efficiently block ATP binding and shows increased basal activity and no inhibition with hyperactivating mutants. These unexpected effects to the signaling may be due to reduced stability of the mutated JH2 with I597F. This hypothesis is supported by data Cancers 2020,12, 78 9 of 20 from JAK1 K622A that shows increased thermal stability and reduced signaling, probably stemming from the stabilizing interaction towards the α C that locks the whole domain in an inhibitory position. Figure 3. Characterization of the JAK1 JH2 ATP-binding site mutants. ( a ) Illustration of the JAK1 JH2 ATP-binding pocket, including the α C-helix of (PDB 4L00). The mutated residues K622 and I597 are shown, as well as ATP. ( b ) JAK1 I597F slightly increases the basal STAT5 activity and is responding to IL-2. JAK1 K622A shows reduced but cytokine-responsive STAT activation. STAT5 responsive luciferase system was used as previously described in U4C γβ cells transfected with JAK1 and JAK3 JH2 ATP-site mutants or JAK WT. The errors are SD triplicate samples. Below : pSTAT5and HAlabeled cell lysates from basal, and cytokine treated cells. Two-tailed t-test was performed and p<0.05 is indicated as *. Right : comparison of JAK1 I597F with WT and hyperactive JAK1 and JAK3. Immunoblots from whole-cell lysates were labelled with HA (JAK1/JAK3/STAT5) and pSTAT5 antibodies to detect the pSTAT5/STAT5 ratios for basal and IL-2 stimulated (100 ng/mL) cells. ( c ) Recombinant JAK1 JH2 mutants vary in their stability. Differential scanning fluorimetry (DSF) analysis of size-exclusion chromatography (SEC)-purified JAK1 JH2 proteins show that JAK1 I597F has lower Tm compared to WT, while K622A increases Tm. Left : the SDS page gels with JAK1 JH2 elutions beforeand after SEC purification. The size of the JAK1 JH2 is ~34 kDa. Right: DSF analysis showing protein Tm − /+ ATP and − /+ MgCl2. Errors are SD (n=6). Graph above presents the dTm that is normalized to the wild-type JAK1 JH2 in its apo-form (protein that does not bind any ligands). Cancers 2020,12, 78 16 of 20 4.3. Cell Transfection and Immunolabeling Human fibroblast cell lines U4C γβ , γ 2A, and 11.1 (deficient in JAK1 and JAK3, JAK2 or TYK2, respectively) were transiently transfected using Fugene HD reagent with different human JAK-hemagglutinin (HA) constructs in pCIneo vector (75 ng for each per 24 well plate well). If two JAK constructs were transfected simultaneously, equal amounts of DNA were used. If STAT5 phosphorylation was studied, 2 ng or HA-tagged STAT5 was co-transfected. After 48 h transfection, cell were stimulated (if needed) for 20 min and washed with cold phosphate buffered saline (PBS). Triton X-100 lysis buffer with protease and phosphatase inhibitors (2 mM vanadate, 1 mM phenylmethanesulfonyl fluoride, 8.3 µ g/mL aprotinin, and 4.2 µ g/mL pepstatin) was used to lyse the cells. Whole cell lysates were spun for 20 min at 16,000 g at 4 ◦ C, and the resulting supernatants were run on 4–15% Mini-PROTEAN ® TGX ™ Precast Gels (BioRad, Irvine, CA, USA). Immunoblots were blocked with bovine serum albumin (BSA) and incubated with primary antibodies for HA Tag (1:2000, OAEA00009, Aviva Systems Biology, San Diego, CA, USA), phosphorylated STAT1 (pSTAT1; 1:1000, #7649, Lot1 Cell Signaling), STAT1 (1:1000, 610116, BD Biosciences, San Jose, CA, USA), or phosphorylated STAT5 (pSTAT5; Cell Signaling, #4322, Lot9), and with a mixture of goat anti-rabbit and goat anti-mouse DyLight secondary antibodies (both from Thermo Fisher Scientific, Waltham, MA, USA). Blots were scanned with an Odyssey CLx (LI-COR Biosciences, Lincoln, NE, USA), and immunoblot signals were quantified with Image Studio software (LI-COR Biosciences) by manually assigning bands and dividing the phosphorylation (pSTAT1 or pSTAT5) signal values with the expression (STAT1 or HA) signals. 4.4. Luciferase and Dual Luciferase Assays STAT5 transcriptional activity was assessed by measuring the luciferase expression (SPI-Luc 2) driven by a STAT5 responsive promoter, as described previously [ 2 ]. For STAT1 transcription activity detection, either IRF-Gas or ISRE-Luc plasmids (30 ng) were transfected instead of SPI-Luc 2 (transfected with 15 ng plasmid). These are specific for IFN γ and IFN α stimulated activation of STAT1, respectively [ 2 , 51 ]. U4C γβ , 11.1, or γ 2A cells were transfected with indicated DNA constructs including the STAT-reporter and with 15 ng renilla plasmid (pRL-TK). The latter was co-transfected as an internal transfection control. Transient transfections were done in 96 well plates with FuGENE HD (Promega) according to manufactures instructions. Then, 42 h after transfection, cells were stimulated (in starvation media) or starved for 5 h after which luciferase assays were analyzed using the dual luciferase reporter assay system (Promega) according to manufactures instructions. Luciferase values were measured with EnVision 2104 Multilabel Reader (Perkin Elmer, Waltham, MA, USA). The results are presented as relative luciferase activity (arbitrary units: a.u.) corresponding to the firefly luciferase light emission values divided by renilla luciferase light emission values. 4.5. Recombinant Protein Production and Purification JAK1 JH2 constructs spanning residues 561–852 of human JAK1, of the wild-type sequence or either I597F or K622A mutations were sub-cloned into pFastBac vector for expression as N-terminal glutathione-S-transferase (GST) fusion proteins with a tobacco etch virus (TEV) proteins. Constructs were expressed in High Five insect cells (Thermo Fisher Scientific) using the Bac-to-Bac baculovirus expression system (Invitrogen, Carlsbad, CA, USA) according to manufacturer’s instructions. After protein expression (10% P3 virus, 48 h, 27 ◦ C), the cells were collected by centrifugation, resuspended in lysis buffer containing 50 mM Tris HCl pH 8.0, 10% Glycerol, 500 nM NaCl, 1 mM TCEP supplemented with phosphatase and protease inhibitors (100 mM sodium orthovanadate, 100 mM PMSF, 10 µ g/mL pepstatin A), and lysed by applying four freeze-thaw cycles. The lysates were clarified by centrifugation and incubated 1.5 h with GSH-coupled resin beads (Protino Glutathione Agarose 4B). Beads were washed and the protein detached by digesting with a tobacco etch virus (TEV) protease (overnight at 4 ◦ C). The flow-through was concentrated and run on a Superdex 200 gel filtration column equilibrated Cancers 2020,12, 78 17 of 20 in final buffer (20 mM Tris pH 8.0, 500 mM NaCl, 10% glycerol, 4 mM DTT and 0.02% CHAPS). The eluted peak was concentrated and stored at −80 ◦C. 4.6. Differential Scanning Fluorometry (DSF) Thermal-shift experiments were carried out in 96-well PCR plates in a final volume of 25 µ L with the following reagent concentrations: 6x Sypro Orange (Molecular Probes, cat. no. S6551), 3 µ M protein Ni-NTA eluate, 1 mM MgCl2, 20 mM Tris pH 8.0, 500 mM NaCl, 10% glycerol, 4 mM DTT and 0.02% CHAPS, and 1 mM ATP. Reactions were heated in a real-time CFX96 PCR cycler (Bio-Rad) at 1 ◦ C per min from 4 ◦ C to 95 ◦ C with a fluorescence read every 1 ◦ C. Fluorescence data were then normalized to represent unfolding curves, which were fitted to a Boltzmann sigmoidal equation with GraphPad Prism to obtain average Tm values with errors as SD. 5. Conclusions Our studies show that JAK1 has varying roles in IL-2, IFN α , and IFN γ systems. Our results demonstrate that the pseudokinase domain (JH2) of JAK1 is an important regulatory region and modulating it can either down, or up-regulate JAK1-driven signaling. Specifically the outer face of the JH2 α C-helix was effective in reducing wild-type and constitutive STAT activation, although the potency dependent on the JAK and the related signaling complex. Furthermore, we highlighted biochemical and biological differences between the ATP-binding sites of JAK JH2s, which can be beneficial in design novel JAK modulators targeting outside the conserved kinase domain. Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6694/12/1/78/s1, Figure S1. Both JAKs are required for cytokine responsive STAT activation in IFN α , IFN γ , and IL-2 systems: Figure S2. Modulating JH2 domain can suppress WT and hyperactive JAKs. ( a ) JAK1 L633K reduces basal pSTAT1 as seen with STAT-activity assay in Figure 2B; Figure S3. Effects of the α C-mutants in trans; Figure S4. Whole blots for figures. Author Contributions: J.R. designed and performed the experiments and wrote the article, T.H. designed and cloned the recombinant JAK1 JH2 constructs and assisted in composing the article, O.S. supervised the work and critically read and edited the article. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Academy of Finland, Sigrid Jus é lius Foundation, Finnish Cancer Foundation, Jane and Aatos Erkko Foundation, Tampere Tuberculosis Foundation, Pirkanmaa hospital district competitive research funding. Acknowledgments: The authors thank Hans-Günter Zerwes (Novartis Institute for Biomedical Research) and Claude (University of Luxembourg) for the kind gift of the modified U4C cell line. We thank Henrik Hammar é n for providing an excellent background for this study, and for great scientific discussions. As always, we thank Merja Peltomaa, Marika Vähä-Jaakkola and Krista Lehtinen for their great technical support. Conflicts of Interest: The authors declare no conflict of interest. References 1. Schwartz, D.M.; Bonelli, M.; Gadina, M.; O’Shea, J.J. Type I/II cytokines, JAKs, and new strategies for treating autoimmune diseases. Nat. Rev. Rheumatol. 2016,12, 25–36. [CrossRef] [PubMed] 2. Saharinen, P.; Silvennoinen, O. The pseudokinase domain is required for suppression of basal activity of Jak2 and Jak3 tyrosine kinases and for cytokine-inducible activation of signal transduction. J. Biol. Chem. 2002 , 277, 47954–47963. [CrossRef] [PubMed] 3. Saharinen, P.; Takaluoma, K.; Silvennoinen, O. Regulation of the Jak2 tyrosine kinase by its pseudokinase domain. Mol. Cell. Biol. 2000,20, 3387–3395. [CrossRef] [PubMed] 4. Sanz Sanz, A.; Niranjan, Y.; Hammar é n, H.; Ungureanu, D.; Ruijtenbeek, R.; Touw, I.P.; Silvennoinen, O.; Hilhorst, R. The JH2 domain and SH2-JH2 linker regulate JAK2 activity: A detailed kinetic analysis of wild type and V617F mutant kinase domains. Biochim. Biophys. Acta BBA Proteins Proteom. 2014 ,1844, 1835–1841. [CrossRef] Cancers 2020,12, 78 18 of 20 5. Lupardus, P.J.; Ultsch, M.; Wallweber, H.; Bir Kohli, P.; Johnson, A.R.; Eigenbrot, C. Structure of the Pseudokinase-kinase domains from protein kinase TYK2 reveals a mechanism for janus kinase (JAK) autoinhibition. Proc. Natl. Acad. Sci. USA 2014,111, 8025–8030. [CrossRef] 6. Ungureanu, D.; Wu, J.; Pekkala, T.; Niranjan, Y.; Young, C.; Jensen, O.N.; Xu, C.F.; Neubert, T.A.; Skoda, R.C.; Hubbard, S.R.; et al. The pseudokinase domain of JAK2 is a dual-specificity protein kinase that negatively regulates cytokine signaling. Nat. Struct. Mol. Biol. 2011,18, 971–976. [CrossRef] 7. Shan, Y.; Gnanasambandan, K.; Ungureanu, D.; Kim, E.T.; Hammar é n, H.; Yamashita, K.; Silvennoinen, O.; Shaw, D.E.; Hubbard, S.R. Molecular basis for pseudokinase-dependent autoinhibition of JAK2 Tyrosine kinase. Nat. Struct. Mol. Biol. 2014,21, 579–584. [CrossRef] 8. Hammar é n, H.M.; Ungureanu, D.; Grisouard, J.; Skoda, R.C.; Hubbard, S.R.; Silvennoinen, O. ATP Binding to the pseudokinase domain of JAK2 is critical for pathogenic activation. Proc. Natl. Acad. Sci. USA. 2015 , 112, 4642–4647. [CrossRef] 9. Raivola, J.; Hammar é n, H.M.; Virtanen, A.T.; Bulleeraz, V.; Ward, A.C.; Silvennoinen, O. Hyperactivation of oncogenic JAK3 mutants depend on ATP binding to the pseudokinase domain. Front. Oncol. 2018 ,8, 560. [CrossRef] 10. Hammar é n, H.M.; Virtanen, A.T.; Raivola, J.; Silvennoinen, O. The regulation of JAKs in Cytokine signaling and its breakdown in disease. Cytokine 2019,118, 48–63. [CrossRef] 11. Candotti, F.; Oakes, S.A.; Johnston, J.A.; Giliani, S.; Schumacher, R.F.; Mella, P.; Fiorini, M.; Ugazio, A.G.; Badolato, R.; Notarangelo, L.D.; et al. Structural and functional basis for JAK3-deficient severe combined immunodeficiency. Blood 1997,90, 3996–4003. [CrossRef] 12. O’Shea, J.J.; Husa, M.; Li, D.; Hofmann, S.R.; Watford, W.; Roberts, J.L.; Buckley, R.H.; Changelian, P.; Candotti, F. Jak3 and the pathogenesis of severe combined immunodeficiency. Mol. Immunol. 2004 ,41, 727–737. [CrossRef] 13. Levine, R.L.; Loriaux, M.; Huntly, B.J.; Loh, M.L.; Beran, M.; Stoffregen, E.; Berger, R.; Clark, J.J.; Willis, S.G.; Nguyen, K.T.; et al. The JAK2V617F activating mutation occurs in chronic myelomonocytic leukemia and acute myeloid leukemia, but not in acute lymphoblastic leukemia or chronic lymphocytic leukemia. Blood 2005,106, 3377–3379. [CrossRef] [PubMed] 14. Jelinek, J.; Oki, Y.; Gharibyan, V.; Bueso-Ramos, C.; Prchal, J.T.; Verstovsek, S.; Beran, M.; Estey, E.; Kantarjian, H.M.; Issa, J.P. JAK2 mutation 1849G>T is rare in acute leukemias but can be found in CMML, philadelphia chromosome-negative CML, and megakaryocytic leukemia. Blood 2005 ,106, 3370–3373. [CrossRef] [PubMed] 15. Scott, L.M.; Scott, M.A.; Campbell, P.J.; Green, A.R. Progenitors homozygous for the V617F mutation occur in most patients with polycythemia vera, but not essential thrombocythemia. Blood 2006 ,108, 2435–2437. [CrossRef] [PubMed] 16. Bandaranayake, R.M.; Ungureanu, D.; Shan, Y.; Shaw, D.E.; Silvennoinen, O.; Hubbard, S.R. Crystal structures of the JAK2 Pseudokinase domain and the pathogenic mutant V617F. Nat. Struct. Mol. Biol. 2012 ,19, 754–759. [CrossRef] 17. Hammar é n, H.M.; Virtanen, A.T.; Abraham, B.G.; Peussa, H.; Hubbard, S.R.; Silvennoinen, O. Janus kinase 2 activation mechanisms revealed by analysis of suppressing mutations. J. Allergy Clin. Immunol. 2019 ,143, 1549–1559. [CrossRef] 18. Leroy, E.; Balligand, T.; Pecquet, C.; Mouton, C.; Colau, D.; Shiau, A.K.; Dusa, A.; Constantinescu, S.N. Differential effect of inhibitory strategies of the V617 mutant of JAK2 on cytokine receptor signaling. J. Allergy Clin. Immunol. 2019,144, 224–235. [CrossRef] 19. Kornev, A.P.; Taylor, S.S. Dynamics-driven allostery in protein kinases. Trends Biochem. Sci. 2015 ,40, 628–647. [CrossRef] 20. Jeong, E.G.; Kim, M.S.; Nam, H.K.; Min, C.K.; Lee, S.; Chung, Y.J.; Yoo, N.J.; Lee, S.H. Somatic mutations of JAK1 and JAK3 in acute leukemias and solid cancers. Clin. Cancer Res. 2008,14, 3716–3721. [CrossRef] 21. Cant é -Barrett, K.; Uitdehaag, J.C.M.; Meijerink, J.P.P. Structural modeling of JAK1 mutations in T-Cell acute lymphoblastic leukemia reveals a second contact site between pseudokinase and kinase domains. Haematologica 2016,101, 189. [CrossRef] [PubMed] 22. Flex, E.; Petrangeli, V.; Stella, L.; Chiaretti, S.; Hornakova, T.; Knoops, L.; Ariola, C.; Fodale, V.; Clappier, E.; Paoloni, F.; et al. Somatically acquired JAK1 mutations in adult acute lymphoblastic leukemia. J. Exp. Med. 2008,205, 751–758. [CrossRef] [PubMed] Cancers 2020,12, 78 19 of 20 23. Haan, C.; Behrmann, I.; Haan, S. Perspectives for the use of structural information and chemical genetics to develop inhibitors of janus kinases. J. Cell. Mol. Med. 2010,14, 504–527. [CrossRef] [PubMed] 24. Moslin, R.; Gardner, D.; Santella, J.; Zhang, Y.; Duncia, J.V.; Liu, C.; Lin, J.; Tokarski, J.S.; Strnad, J.; Pedicord, D.; et al. Identification of imidazo[1,2-B] pyridazine TYK2 pseudokinase ligands as potent and selective allosteric inhibitors of TYK2 signalling. Med. Chem. Commun. 2017,8, 700–712. [CrossRef] 25. Taylor, S.S.; Kornev, A.P. Protein Kinases: Evolution of dynamic regulatory proteins. Trends Biochem. Sci. 2011,36, 65–77. [CrossRef] 26. Haan, C.; Rolvering, C.; Raulf, F.; Kapp, M.; Drückes, P.; Thoma, G.; Behrmann, I.; Zerwes, H. Jak1 has a dominant role over Jak3 in signal transduction through Γ c-containing cytokine receptors. Chem. Biol. 2011 , 18, 314–323. [CrossRef] 27. Ferrao, R.D.; Wallweber, H.J.; Lupardus, P.J. Receptor-mediated dimerization of JAK2 FERM domains is required for JAK2 activation. eLife 2018,7, 38089. [CrossRef] 28. Dusa, A.; Mouton, C.; Pecquet, C.; Herman, M.; Constantinescu, S.N. JAK2 V617F constitutive activation requires JH2 residue F595: A pseudokinase domain target for specific inhibitors. PLoS ONE 2010 ,5, e11157. [CrossRef] 29. Gnanasambandan, K.; Magis, A.; Sayeski, P.P. The Constitutive activation of Jak2-V617F is mediated by a Pi stacking mechanism involving phenylalanines 595 and 617. Biochemistry 2010,49, 9972–9984. [CrossRef] 30. Toms, A.; Deshpande, A.; McNally, R.; Jeong, Y.; Rogers, J.; Un Kim, C.; Gruner, S.; Ficarro, S.; Marto, J.A.; Sattler, M.; et al. Structure of a pseudokinase domain switch that controls oncogenic activation of Jak kinases. Nat. Struct. Mol. Biol. 2013,20, 1221. [CrossRef] 31. Eletto, D.; Burns, S.O.; Angulo, I.; Plagnol, V.; Gilmour, K.C.; Henriquez, F.; Curtis, J.; Gaspar, M.; Nowak, K.; Daza-Cajigal, V.; et al. Biallelic JAK1 mutations in immunodeficient patient with mycobacterial infection. Nat. Commun. 2016,7, 13992. [CrossRef] [PubMed] 32. Tate, J.G.; Bamford, S.; Jubb, H.C.; Sondka, Z.; Beare, D.M.; Bindal, N.; Boutselakis, H.; Cole, C.G.; Creatore, C.; Dawson, E.; et al. COSMIC: The catalogue of somatic mutations in cancer. Nucleic Acids Res. 2018 ,47, 941–947. [CrossRef] [PubMed] 33. Hammar é n, H.M.; Virtanen, A.T.; Silvennoinen, O. Nucleotide-binding mechanisms in pseudokinases. Biosci. Rep. 2016, 36. [CrossRef] [PubMed] 34. Rodig, S.J.; Meraz, M.A.; White, J.M.; Lampe, P.A.; Riley, J.K.; Arthur, C.D.; King, K.L.; Sheehan, K.C.; Yin, L.; Pennica, D.; et al. Disruption of the Jak1 gene demonstrates obligatory and nonredundant roles of the jaks in cytokine-induced biologic responses. Cell 1998,93, 373–383. [CrossRef] 35. Vainchenker, W.; Constantinescu, S.N. JAK/STAT signaling in hematological malignancies. Oncogene 2012 , 32, 2601. [CrossRef] 36. Kleppe, M.; Spitzer, M.H.; Li, S.; Hill, C.E.; Dong, L.; Papalexi, E.; De Groote, S.; Bowman, R.L.; Keller, M.; Koppikar, P.; et al. Jak1 integrates cytokine sensing to regulate hematopoietic stem cell function and stress hematopoiesis. Cell Stem Cell 2017,21, 489–501. [CrossRef] 37. Briscoe, J.; Rogers, N.C.; Witthuhn, B.A.; Watling, D.; Harpur, A.G.; Wilks, A.F.; Stark, G.R.; Ihle, J.N.; Kerr, I.M. Kinase-negative mutants of JAK1 can sustain interferon-gamma-inducible gene expression but not an antiviral state. EMBO J. 1996,15, 799–809. [CrossRef] 38. Haan, S.; Margue, C.; Engrand, A.; Rolvering, C.; Schmitz-Van, d.L.; Heinrich, P.C.; Behrmann, I.; Haan, C. Dual role of the Jak1 FERM and kinase domains in cytokine receptor binding and in stimulation-dependent jak activation. J. Immunol. 2008,180, 998. [CrossRef] 39. Greenlund, A.C.; Farrar, M.A.; Viviano, B.L.; Schreiber, R.D. Ligand-Induced IFN gamma receptor tyrosine phosphorylation couples the receptor to its signal transduction system (p91). EMBO J. 1994 ,13, 1591–1600. [CrossRef] 40. Keil, E.; Finkenst ä dt, D.; Wufka, C.; Trilling, M.; Liebfried, P.; Strobl, B.; M Ã1 4 ller, M.; Pfeffer, K. Important scaffold function of the janus kinase 2 uncovered by a novel mouse model harboring a Jak2 activation-loop mutation. Blood 2014,123, 520. [CrossRef] 41. Wallweber, H.J.; Tam, C.; Franke, Y.; Starovasnik, M.A.; Lupardus, P.J. Structural Basis of recognition of interferon-alpha receptor by tyrosine kinase 2. Nat. Struct. Mol. Biol. 2014 ,21, 443–448. [CrossRef] [PubMed] 42. Ferrao, R.; Wallweber, H.J.; Ho, H.; Tam, C.; Franke, Y.; Quinn, J.; Lupardus, P.J. The structural basis for class II cytokine receptor recognition by JAK1. Structure 2016,24, 897–905. [CrossRef] [PubMed] Cancers 2020,12, 78 20 of 20 43. Zhang, D.; Wlodawer, A.; Lubkowski, J. Crystal Structure of a complex of the intracellular domain of interferon λ receptor 1 (IFNLR1) and the FERM/SH2 domains of human JAK1. J. Mol. Biol. 2016 ,428, 4651–4668. [CrossRef] [PubMed] 44. Ferrao, R.; Lupardus, P.J. The janus kinase (JAK) FERM and SH2 domains: Bringing specificity to JAK - receptor interactions. Front. Endocrinol. 2017,8, 71. [CrossRef] [PubMed] 45. Hubbard, S.R. Mechanistic insights into regulation of JAK2 tyrosine kinase. Front. Endocrinol. 2018 ,8, 361. [CrossRef] [PubMed] 46. Lupardus, P.J.; Skiniotis, G.; Rice, A.J.; Thomas, C.; Fischer, S.; Walz, T.; Garcia, K.C. Structural snapshots of full-length Jak1, a transmembrane gp130/IL-6/IL-6Ralpha cytokine receptor complex, and the receptor-Jak1 holocomplex. Structure 2011,19, 45–55. [CrossRef] [PubMed] 47. Koppikar, P.; Bhagwat, N.; Kilpivaara, O.; Manshouri, T.; Adli, M.; Hricik, T.; Liu, F.; Saunders, L.M.; Mullally, A.; Abdel-Wahab, O.; et al. Heterodimeric JAK-STAT activation as a mechanism of persistence to JAK2 inhibitor therapy. Nature 2012,489, 155–159. [CrossRef] 48. Meyer, S.C.; Keller, M.D.; Chiu, S.; Koppikar, P.; Guryanova, O.A.; Rapaport, F.; Xu, K.; Manova, K.; Pankov, D.; O’Reilly, R.J.; et al. CHZ868, a type II JAK2 inhibitor, reverses type I JAK inhibitor persistence and demonstrates efficacy in myeloproliferative neoplasms. Cancer. Cell. 2015,28, 15–28. [CrossRef] 49. Tvorogov, D.; Thomas, D.; Liau, N.P.D.; Dottore, M.; Barry, E.F.; Lathi, M.; Kan, W.L.; Hercus, T.R.; Stomski, F.; Hughes, T.P.; et al. Accumulation of JAK activation loop phosphorylation is linked to type I JAK inhibitor withdrawal syndrome in myelofibrosis. Sci. Adv. 2018,4, 3834. [CrossRef] 50. Leroy, E.; Dusa, A.; Colau, D.; Motamedi, A.; Cahu, X.; Mouton, C.; Huang, L.J.; Shiau, A.K.; Constantinescu, S.N. Uncoupling JAK2 V617F activation from cytokine-induced signalling by modulation of JH2 αC helix. Biochem. J. 2016,473, 1579–1591. [CrossRef] 51. Pine, R.; Canova, A.; Schindler, C. Tyrosine phosphorylated p91 binds to a single element in the ISGF2/IRF-1 promoter to mediate induction by IFN alpha and IFN gamma, and is likely to autoregulate the p91 gene. EMBO J. 1994,13, 158–167. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).