Expression, purification, and characterization of TYK-2 kinase domain, a member of the Janus kinase family

Biochemical and Biophysical Research Communications 396 (2010) 543–548

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

Expression, purification, and characterization of TYK-2 kinase domain, a member of the Janus kinase family Brian Korniski, Arthur J. Wittwer, Thomas L. Emmons, Troii Hall, Stacy Brown, Ann D. Wrightstone, Jeffrey L. Hirsch, Jennifer A. Gormley, Robin A. Weinberg, Joseph W. Leone, Jacqueline E. Day, Jill E. Chrencik, Cynthia D. Sommers, H. David Fischer, Alfredo G. Tomasselli * Pfizer Inc., Global Research and Development, St. Louis Laboratories, 700 Chesterfield Parkway West, St. Louis, MO 63017-1732, USA

a r t i c l e

i n f o

Article history: Received 20 April 2010 Available online 8 May 2010 Keywords: TYK-2 Phosphorylation Purification Kinetic Kinase inhibitors JAK-2 JAK-3

a b s t r a c t The Janus kinase family consists of four members: JAK-1, -2, -3 and TYK-2. While JAK-2 and JAK-3 have been well characterized biochemically, there is little data on TYK-2. Recent work suggests that TYK-2 may play a critical role in the development of a number of inflammatory processes. We have carried out a series of biochemical studies to better understand TYK-2 enzymology and its inhibition profile, in particular how the TYK-2 phosphorylated forms differ from each other and from the other JAK family members. We have expressed and purified milligram quantities of the TYK-2 kinase domain (KD) to high purity and developed a method to separate the non-, mono- (pY1054) and di-phosphorylated forms of the enzyme. Kinetic studies (kcat(app)/Km(app)) indicated that phosphorylation of the TYK-2-KD (pY1054) increased the catalytic efficiency 4.4-fold compared to its non-phosphorylated form, while further phosphorylation to generate the di-phosphorylated enzyme imparted no further increase in activity. These results are in contrast to those obtained with the JAK-2-KD and JAK-3-KD, where little or no increase in activity occurred upon mono-phosphorylation, while di-phosphorylation resulted in a 5.1-fold increase in activity for the JAK-2-KD. Moreover, ATP-competitive inhibitors demonstrated 10–30-fold shifts in potency (Ki(app)) as a result of the TYK-2-KD phosphorylation state, while the shifts for JAK-3-KD were only 2– 3-fold and showed little or no change for JAK-2-KD. Thus, the phosphorlyation state imparted differential effects on both activity and inhibition within the JAK family of kinases. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction The four members of the Janus tyrosine kinase (JAK) family, JAK-1, JAK-2, JAK-3, and non-receptor protein tyrosine kinase (TYK-2), and the seven members of the Signal Transducers and Activators of Transcription (STAT), STAT1–4, -5a, -5b, and -6, are critically involved in downstream signaling of many cytokine receptors [1,2]. Human and murine genetic studies clearly link the cytokine receptor/JAK/STAT pathway to proper functioning of the immune system: inactivating mutations within JAK-3 result in severe combined immunodeficiency (SCID) syndrome [3–5]; myeloproliferative disorders and more specifically polycythemia vera have been linked to a V617F mutation within JAK-2 [6–8]; and a patient diagnosed with multiple opportunistic infections was found to lack functional TYK-2 activity [9]. In addition, several lines of evidence support the notion that IFN/Th1, IL-12 and IL-23 signaling via TYK-2 may serve as a regulator of the inflammatory * Corresponding author. Present address: 1540 Garden Valley Dr., Wildwood, MO 63038, USA. E-mail address: [email protected] (A.G. Tomasselli). 0006-291X/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2010.04.141

response [2]. These findings and the proven success of anti-cytokine/anti-cytokine receptor therapy against rheumatoid arthritis (RA), psoriasis, and inflammatory bowel disease (IBD), have fostered interest in targeting the JAK family to treat these diseases. Indeed, inhibitors of JAK-2 and JAK-3 are already in advanced clinical trials. Historically the development of selective ATP-competitive inhibitors against a target kinase has proven challenging due to the structural similarity among kinase binding pockets. Targeting a specific form of the kinase, e.g., inhibiting the non-phosphorylated rather than phosphorylated enzyme might provide a mechanism to improve inhibitor selectivity. For instance, the commercial inhibitor, Gleevec (imatinib), recognizes the inactive and unique conformational state of the tyrosine kinase, c-Abl [10,11]. A similar approach towards developing therapeutics targeting the JAK kinases would be greatly enhanced by an understanding of how phosphorylation affects activity and structural conformation. All four JAK enzymes conserve a motif consisting of two tandem tyrosine residues within the activation loop that is critical for regulating enzyme activity [12–15]. Substantial experimental evidence with both the full length enzyme and the truncated kinase

544

B. Korniski et al. / Biochemical and Biophysical Research Communications 396 (2010) 543–548

domain have defined the activation of JAK-2 and JAK-3 through phosphorylation of tyrosine residues Y1007 and Y980, respectively, within the activation loop for each isozyme [12,15]. The equivalent YY motif within TYK-2, Y1054 and Y1055, are predicted to play similar roles in TYK-2 activation. Indeed, site-directed mutagenesis experiments have identified the YY residues to be critical in cytokine dependent activation in cell based assays [13]. Herein we describe the expression, purification and characterization of non-, mono-, and di-phosphorylated forms of the TYK2 kinase domain (KD). An enzyme assay was developed that allowed for the comparative investigation of the effect of phosphorylation state on the activity of JAK-2, JAK-3, and TYK-2. Inhibition studies using various ATP-competitive inhibitors examined the relationship between phosphorylation state and inhibitor potency. In addition, the methods developed provided sufficient quantities of purified TYK-2-KD to support a structural biology effort [16]. 2. Materials and methods 2.1. Materials The following reagents were employed. Tris, NaCl, MgCl2, glycerol, ATP, HEPES, BSA, dithiothreitol, Tween-20, EDTA and DMSO, from Sigma Chemical Co., St. Louis, MO. CompleteÒ-EDTA free tablets from Roche Diagnostics, Indianapolis, IN. The Resource™-Q, HiLoad™, HisTrap™ columns, and BioWaveÒ reactor bags from GE Healthcare, Piscataway, NJ, and the Ni2+-NTA Superflow resin from Qiagen Inc., Valencia, CA. BenzonaseÒ, BacMagic™ and imidazole from EMD Biosciences, Madison, WI. The QuikChange™ kit was from Stratagene, La Jolla, CA. Tris-(2-carboxyethyl)phosphine, hydrochloride (TCEP) from Invitrogen, Carlsbad, CA. Sequencing grade trypsin from Promega, Madison, WI. IRS-1 Peptide (5FAMKKSRGDYMTMQIG-NH2) and JAKtide (FITC-C6-KGGEEEEYFELVKKNH2) were synthesized by American Peptide Company, Sunnyvale, CA. Coating-3 Reagent from Caliper Life Sciences, Hopkinton, MA. JAK Inhibitor I (Pyridone 6) and Staurosporine from Calbiochem, La Jolla, CA. CP-690550 and PF-956980 from Pfizer, Inc. (Chesterfield, MO). 2.2. Cloning, expression, and purification of JAK-2, JAK-3 and TYK-2 kinase domains The expression and purification of JAK-2-KD and JAK-3-KD were carried out as described previously [17]. With respect to TYK-2-KD, recombinant baculovirus stocks were generated using the BacMagic™ system according to the manufacturer. Expression optimization studies were completed by varying the virus concentration, length of infection and cell line. Infection course was monitored by cell diameter and/or cell viability counts. A Sf21 insect cell line was chosen as host for the production of recombinant TYK-2-KD. For large scale production (10 L or 20 L), cells were grown in BioWaveÒ reactor bags in serum free media. Cell counts at the time of infection were approximately 1  106 cells/mL with >97% viability, as determined by trypan blue staining. Baculovirus stocks were added to give a final multiplicity of infection of 1 pfu/cell and the cells harvested at 65 h postinfection. Expression batches were performed at Pfizer Labs, Chesterfield, MO; EMD Biosciences; and Blue Sky Biotech, Worcester, MA. All TYK-2-KD purification procedures were performed at 5 °C unless otherwise noted. Cell pellets from a 5 L fermentation were resuspended into 250 mL lysis buffer (50 mM Tris, pH 8.5, 200 mM NaCl, 1 mM TCEP, 10% glycerol, CompleteÒ-EDTA free protease inhibitor cocktail; 1 tablet/50 mL, 500 U benzonase/L cell paste) at room temperature followed by continued stirring for 30 min at 5 °C. The lysate was clarified by centrifugation at

30,000g for 60 min. The clarified supernatant was batch bound for 1 h to 5 mL of Ni2+-NTA Superflow resin. The resin slurry was packed into a BioRad EconoÒ-column and washed with 20 column volumes of wash buffer (50 mM Tris, pH 8.5, 400 mM NaCl, 20 mM imidazole, 1 mM TCEP, 10% glycerol) and the protein eluted with 3 column volumes of wash buffer containing 250 mM imidazole. The eluate was applied to a HiLoad™ 26/60 Superdex 200 prep grade column pre-equilibrated with 50 mM Tris, pH 8.5, 1 mM TCEP, 10% glycerol, and 10 mM NaCl (Q-buffer). Monomeric TYK-2-KD was applied to a 6 mL Resource™-Q column and eluted using a 60 column volume linear gradient run from 10 to 130 mM NaCl in Q-buffer to resolve the different TYK-2-KD phosphorylation forms. Fractions were analyzed by SDS–PAGE and protein concentrations determined by A280nm using an extinction coefficient of 1.258 AU mL mg 1. 2.3. Mass spectrometry of JAK-2, JAK-3, and TYK-2 KD phosphorylated states Masses of the various non-, mono-, and di-phosphorylated forms of JAK-2 and JAK-3 kinase domains were determined as described earlier [17]. Electrospray ionization mass spectrometry (ESI-MS) and ESI-MS/MS were applied to determine the molecular weights and the phosphorylation site, respectively, for the purified states of TYK-2-KD using a Waters-Micromass Q-TOF Micro mass spectrometer. Each peak isolated from the anion-exchange column was buffer exchanged into 20% acetonitrile with 0.1% formic acid using a 10 kDa MWCO spin concentrator. The ESI-MS data were acquired by direct infusion into the spectrometer using a capillary voltage of 3000 V and cone voltage of 35–45 V. The raw mass spectral data was deconvoluted using the MaxEnt algorithm provided with Masslynx mass spectrometry software. The results reported are from centered spectra using appropriate parameters. For identification of the phosphorylation sites, the TYK-2-KD was digested with sequencing grade trypsin at 1:50 trypsin: TYK2-KD (w/w) ratio in 20% acetonitrile and 20 mM Tris pH 7.8 at room temperature overnight. The digested samples were injected onto a Waters capillary HPLC system equipped with a reversephase C18 VydacÒ EverestÒ trap and column, (100  0.3 mm, 5 lm particle size) with a set flow rate of 3 lL/min. The peptides were eluted with a linear gradient of 0–45% B in A over a period of 80 min, where A is 2% acetonitrile/water with 1.0% formic acid, and B is 90% acetonitrile/water with 1.0% formic acid. The ESIMS/MS data were acquired in positive ion mode from m/z 200 to 1950 amu s 1. During data-dependent analysis three precursors/ scan were allowed at a threshold of 20 counts. TOF MS/MS spectra were acquired for 2 s using an interscan time of 0.1 s and data analyzed using the MassLynx 4.1 and MASCOT software. All phosphopeptides identified during MASCOT searches were confirmed by manual interpretation of the spectra. 2.4. Activity and inhibition assays TYK-2-KD activity was measured using a LabChipÒ 3000 microfluidics instrument (Caliper Life Sciences). In the assay, phosphorylation of IRS-1 Peptide was detected as a shift in mobility in the microfluidics system. This method allowed rapid separation and fluorescence quantification of both substrate and product. Reaction mixtures of 10 lL volume contained 1 lM 5FAM-KKSRGDYMTMQIG-NH2 substrate, ATP and enzyme as indicated, and 2% DMSO, in a buffer containing 20 mM HEPES, pH 7.5, 10 mM MgCl2, 0.01% bovine serum albumin, 0.0005% Tween-20, and 1 mM dithiothreitol. For enzyme inhibition assays, the final ATP concentration was fixed at the apparent Km, inhibitors were introduced as DMSO solutions to yield a final DMSO concentration of 2%, and the reactions initiated by the addition of enzyme (20 nM non-phosphor-

545

B. Korniski et al. / Biochemical and Biophysical Research Communications 396 (2010) 543–548

ylated TYK-2-KD, 2 nM mono-phosphorylated TYK-2-KD, or 1.25 nM di-phosphorylated TYK-2-KD, final nominal concentrations). After incubating for 60 min at room temperature the reactions were quenched by adding 20 lL of 140 mM HEPES, pH 7.5, 22.5 mM EDTA, and 0.15% Coating-3 Reagent, and the samples assayed. For kinetic assays, the reaction volume was increased to 80 lL and small samples assayed without quenching every few minutes. Kinetic assays were used to verify linearity of the reaction with time, to obtain initial rates for Km determinations, and to determine the amount of enzyme and time required to yield 15– 20% substrate conversion for the enzyme inhibition assays. JAK-2-KD and JAK-3-KD activity were measured as described above, except that FITC-C6-KGGEEEEYFELVKK-NH2 was used as the substrate and the final nominal concentrations of enzyme and incubation times for the enzyme inhibition assays were as follows: non-phosphorylated JAK-2-KD, 12.5 nM and 135 min; monophosphorylated JAK-2-KD, 2.5 nM and 80 min; di-phosphorylated JAK-2-KD, 1 nM and 30 min; non-phosphorylated JAK-3-KD, 10 nM and 70 min; mono-phosphorylated JAK-3-KD, 10 nM and 45 min. Apparent Vmax and ATP Km values were determined by non-linear regression analysis of initial rate data [18] obtained at ATP concentrations ranging from at least 10-fold above to 10-fold below the Km value. These values were reported as apparent values because they were determined at a sub-saturating level of peptide substrate. Although the Km values for the peptide substrates employed in these assays were not known, it was clear that the 1 lM concentration used in these assays was substantially below the Km as evidenced by the linearity of activity as a function of peptide concentration in the 0–8 lM range. Active enzyme concentrations were determined by titration with two different reversible inhibitors at three to four enzyme concentrations under tightbinding conditions (active enzyme concentrations greater than the apparent inhibitor Ki) and globally fitting all these dose–response data to the Morrison equation described by Copeland [19]. Apparent kcat values were calculated from apparent Vmax values by dividing by the active enzyme concentration.

kDa

1

2

3

4

5

IC50 values calculated from enzyme inhibition assay dose–response data were sometimes the same order of magnitude as the active enzyme concentration. To properly compare the data between enzyme forms, these IC50 values were converted to apparent Ki values by subtracting one-half the active enzyme concentration [19]. 3. Results and discussion 3.1. TYK-2-KD expression TYK-2-KD constructs were designed based on the existing structural information for the JAK-2-KD and JAK-3-KD enzymes [20,21]. Three different N-termini, D888, S884 and V880 and two carboxy terminal boundaries, G1178 and S1182, were selected and their expression evaluated in Sf9 and Sf21 cell lines at three different multiplicities of infection. The duration of infection was also varied to ensure that all conditions were appropriately assessed. Surprisingly, overall expression yields were poor and large amounts of protein were observed to be aggregated. To address these problems, the internal cysteine residues were mutated to alanine as designed in our JAK-2-KD and JAK-3-KD constructs [17]. Additionally, three variants with differing surface mutations, (M1119S, L1122T), (K1074A, E1075A, K1077A) and (Q969A, E971A, K972A), were generated for structural biology studies. These mutations resulted in significant improvements in the oligomerization state of the protein to the monomeric form. The construct spanning TYK-2 residues D888 to S1182 and including mutations C936A, C1142A, Q969A, E971A, and K972A, was chosen for further studies due to both increases in protein expression yields as well as significant reductions in the aggregation profile. 3.2. Isolation of non-, mono-, and di-phosphorylated TYK-2-KD Purification of the mutated TYK-2-KD construct from 5 L crude lysate by Ni2+-NTA chromatography yielded a protein that was >85% pure and migrated at approximately 36 kDa on SDS–PAGE (Fig. 1). A final purification step utilizing size exclusion chromatog6

7

8

9

10

11

12

188 98 62 49 38 28

14 6 3

Fig. 1. SDS–PAGE gel of the purification of TYK-2-KD. Samples from the purification of TYK-2-KD were analyzed using a 4–12% Bis-Tris NuPage SDS–PAGE gel run under reducing conditions. Lane 1: SeeBlue Plus2 Standard (Invitrogen); Lane 2: Clarified cell lysate; Lane 3: Ni2+-NTA flow through; Lane 4: Ni2+-NTA resin wash; Lane 5: Ni2+-NTA resin eluate; Lane 6: SEC monomer Pool/Q load; Lane 7: Q Peak 1, non-phosphorylated pool; Lane 8: Q Peak 2, mono-phosphorylated pool; Lane 9: Q Peak 3, diphosphorylated pool; Lane 10: Q Peak 4, di- and tri- phosphorylated pool; Lane 11: Q Peak 5 di- and tri-phosphorylated pool; and Lane 12: Q Peak 6, high salt elution of non-, mono-, and di-phosphorylated pool.

546

B. Korniski et al. / Biochemical and Biophysical Research Communications 396 (2010) 543–548

Fig. 2. Resolution of the multiple phosphorylated isoforms of TYK-2-KD by anion-exchange chromatography. A linear gradient from 10 to 130 mM NaCl over 60 column volumes followed by a 1 M NaCl wash was used to separate the various phosphorylated species. Six peaks, designated 1–6, were resolved and characterized by ESI-MS. The peaks were characterized to contain the following TYK-2-KD isoforms: (1) non-phosphorylated; (2) mono-phosphorylated; (3) di-phosphorylated; (4) di- and triphosphorylated; (5) di- and tri-phosphorylated; and (6) high salt elution of non-, mono-, and di-phosphorylated.

raphy (SEC) yielded 6.3 mg of monomeric TYK-2-KD at >98% purity. Analysis of the monomeric TYK-2-KD by ESI-MS identified a mixture of non-, mono-, and di-phosphorylated TYK-2-KD, with the mono-phosphorylated form being the most abundant. Indeed, when this TYK-2 construct was co-purified with CP-690550 or JAK Inhibitor I, it formed crystals that diffracted to high resolution and were characterized to be mono-phosphorylated [16]. To characterize the role of phosphorylation on TYK-2-KD activity and inhibitor potency the different isoforms were isolated. An anion-exchange column combined with a shallow salt gradient successfully resolved six major peaks (Fig. 2) which ESI-MS identified as acetylated TYK-2-KD lacking the N-terminal methionine with zero to three phosphates. The first three major peaks correlated to non-, mono-, and di-phosphorylated TYK-2-KD, respectively. The fourth and fifth peaks contained mixtures of di- and tri-phosphorylated TYK-2-KD. Raising the salt concentration from 130 mM NaCl to 1 M NaCl resulted in a sixth peak that contained a mixture of non-, mono-, and di-phosphorylated TYK-2-KD. The enzyme activity of TYK-2-KD was monitored during the purification process and specific activities calculated (Table 1). Trypsin digestion of the mono-phosphorylated TYK-2-KD produced a doubly charged parent ion of mass 650.80 Da and MS/ MS analysis identified the phosphorylation at Y1054 in the peptide AVPEGHEY1054Y1055R. This site of phosphorylation was in agree-

Table 1 Purification summary of the separated phosphorylated states of TYK-2-KD from 5 L of a Sf21 expression batch. Purification step

Total protein (mg)

Specific activity (nmol/min/mg)

Total activity (nmol/min)

Ni2+-NTA load Ni2+-NTA eluate/SEC load SEC pool/Q load (mixture) Q pool 1 (non) Q pool 2 (mono) Q pool 3 (di) Q pool 4 (di- and tri-) Q pool 5 di- and triQ pool 6 (mixture)

8070 13.1 6.3 0.16 0.62 0.04 0.02 0.005 0.45

0.445 53.6 126 13.0 216 263 269 253 10.8

3590 702 796 2.09 134 10.5 5.38 1.26 4.85

ment with the crystal structure of the mono-phosphorylated TYK-2-KD [16]. Digestion of the di-phosphorylated pool did not result in an identifiable site for the second phosphorylation, but is likely to occur at Y1055. 3.3. Catalytic activity of phosphorylated isoforms of the JAK-2, JAK-3, and TYK-2 kinase domains The active site titration percentages, Km(app), and calculated kcat(app)/Km(app) values of the purified JAK-2-KD, JAK-3-KD, and TYK-2-KD, phosphorylation isoforms are reported in Table 2 and compared with previously reported results [17]. Each enzyme exhibited a distinct effect of phosphorylation state on activity. It is of interest that for a given enzyme the Km(app) values were within 2-fold regardless of phosphorylation state meaning that differences in kcat(app)/Km(app) were primarily due to kcat(app). Comparison of mono-phosphorylated isoforms with the non-phosphorylated isoforms, TYK-2-KD showed a 4.8-fold increase in kcat(app)/Km(app) upon phosphorylation, while JAK-2-KD and JAK-3-KD showed little or no increase. In contrast, when mono- and di-phosphorylated isoforms were compared, TYK-2-KD did not show an increase in kcat(app)/Km(app) upon additional phosphorylation, while JAK-2-KD demonstrated a 5.4-fold increase. These results suggest that whereas the primary phosphorylation of TYK-2-KD is sufficient for activation, di-phosphorylation is required for JAK-2-KD and possibly JAK-3-KD activation. As given in Table 2, earlier reports [17] of Km(app) and kcat(app)/ Km(app) for JAK-2-KD and JAK-3-KD phosphorylation isoforms purified by similar methods described herein. While the Km(app) values agreed within 2-fold, there were significant differences in the values of kcat(app)/Km(app) between the present and previously reported results. In particular, the previous work showed a 9-fold increase in the activity of JAK-2-KD upon mono-phosphorylation, while the values in Table 2 revealed a modest 1.5-fold increase. This could be explained in part by the use of nominal protein concentrations for the calculation of kcat(app)/Km(app) carried out earlier [17], while concentrations determined by active site titration were used here. The active site concentration of the non-phosphorylated JAK-2-KD was only 1.0% of nominal, compared to 14.6%, and 9.1%, for the

547

B. Korniski et al. / Biochemical and Biophysical Research Communications 396 (2010) 543–548

Table 2 Kinetic parameters of phosphorylated isoforms for JAK-2, JAK-3, and TYK-2 kinase domains. Isoform

*

Recalculated* from previous study [17]

Present study Active site titration (% nominal)

Km(app) (lM)

kcat(app)/Km(app) (lM 1min 1)

Km(app) (lM)

kcat(app)/Km(app) (lM 1min 1)

kcat(app)/Km(app) (lM (recalculated)

JAK-2 (non) JAK-2 (mono) JAK-2 (di)

1.0 ± 0.1 14.6 ± 1.1 9.1 ± 1.2

9.7 ± 1.1 5.3 ± 0.9 11.5 ± 1.3

11 ± 2 17 ± 3 87 ± 16

5.68 ± 0.81 10.9 ± 0.46 7.40 ± 0.19

0.04 0.35 2.07

4.0 2.4 23

JAK-3 (non) JAK-3 (mono)

22.8 ± 1.6 21.5 ± 1.8

1.9 ± 0.2 3.9 ± 0.4

1.9 ± 0.3 2.0 ± 0.3

3.08 ± 0.54 5.15 ± 0.67

0.28 0.23

1.2 1.1

TYK-2 (non) TYK-2 (mono) TYK-2 (di)

6.8 ± 0.6 7.5 ± 1.0 9.4 ± 0.5

11.6 ± 1.0 14.4 ± 1.7 11.8 ± 0.4

1.9 ± 0.2 8.3 ± 1.4 8.2 ± 0.5

– – –

– – –

– – –

1

min

1

)

Using the active site titration percentages determined in the present study.

Table 3 Inhibition profile of phosphorylated isoforms of JAK-2, JAK-3, and TYK-2 kinase domains with four inhibitors. Isoform

Ki(app) (nM) (mean ± standard error, n = 4, n = 8 for CP-690550) CP-690550

PF-956980

JAK Inhibitor I

Staurosporine

JAK-2 (non) JAK-2 (mono) JAK-2 (di)

2.1 ± 0.3 1.3 ± 0.2 1.4 ± 0.2

3.4 ± 0.3 2.2 ± 0.1 2.9 ± 0.3

0.69 ± 0.1 0.47 ± 0.08 0.49 ± 0.02

1.4 ± 0.2 1.7 ± 0.2 1.9 ± 0.3

JAK-3 (non) JAK-3 (mono)

2.6 ± 0.5 1.0 ± 0.2

9.0 ± 0.5 2.9 ± 0.3

5.5 ± 0.3 2.4 ± 0.1

6.5 ± 0.8 5.4 ± 0.3

595 ± 33 75 ± 3 49 ± 2

6.2 ± 0.3 0.59 ± 0.06 0.21 ± 0.02

15 ± 2 2.3 ± 0.1 1.2 ± 0.2

TYK-2 (non) TYK-2 (mono) TYK-2 (di)

146 ± 8 24 ± 2 14 ± 2

mono-, and di-phosphorylated isoforms. Active site titration was not reported previously [17]; however, if the active site percentages determined in this study were applied retrospectively to the previous data, the conclusion would also have been that the activities of non- and mono-phosphorylated JAK-2-KD were similar (see Table 2). Further studies will be required to understand the reason for the very low amount of active enzyme in the non-phosphorylated JAK-2-KD preparation and what influence this might have on the accuracy of the kcat(app)/Km(app) values for non-phosphorylated JAK-2-KD in Table 2.

an inactive state, thereby preventing activation. Our data suggest a possible difference between the JAK enzyme-inhibitor binding profiles that may help elucidate biological JAK selectivity profiles with these inhibitors. Among the JAK family members studied, TYK-2-KD was unique in possessing a clear difference in inhibitor potency depending on the phosphorylation state. Based on this change in inhibitor potency, it would appear that phosphorylation of TYK-2-KD resulted in a more dramatic change in the binding pocket compared to JAK-2-KD and JAK-3-KD. This unique property of TYK-2 might be exploited in the design of TYK-2 specific inhibitors.

3.4. Inhibition profile of the phosphorylation isoforms of the JAK-2, JAK-3, and TYK-2 kinase domains

References

The potency of inhibition of the phosphorylation isoforms of JAK-2-KD, JAK-3-KD, and TYK-2-KD by three inhibitors of the JAK family of kinases, CP-690550 [22], JAK Inhibitor I [23] and PF986980 [24] as well as staurosporine, a broad spectrum protein kinase inhibitor, is presented in Table 3. The three kinases exhibited differing effects of phosphorylation on inhibitor potency. The effect on inhibitor potency was less than 2-fold when comparing the phosphorylation isoforms of JAK-2-KD. In the case of JAK-3-KD, however, there was a modest 2–3-fold increase in potency upon phosphorylation for the JAK inhibitors while staurosporine potency was essentially unchanged. In contrast, TYK-2-KD revealed a pronounced 6.2–10.5-fold increase in potency upon mono-phosphorylation with an additional 1.5–2.7-fold increase upon diphosphorylation. Kinases are believed to interconvert between active and inactive conformations based on the phosphorylation state of key tyrosine residues [25]. Inhibitors can target both conformations with similar affinity [26] or selectively target the active or inactive conformations [10]. Compounds that bind both conformations may have a therapeutic advantage by shifting the equilibrium towards

[1] A.F. Wilks, The JAK kinases: not just another kinase drug discovery target, Semin. Cell Dev. Biol. 19 (2008) 319–328. [2] K. Ghoreschi, A. Laurence, J.J. O’Shea, Janus kinases in immune cell signaling, Immunol. Rev. 228 (2009) 273–287. [3] N.A. Cacalano, T.S. Mignone, F. Bazan, et al., Autosomal SCID caused by a point mutation in the N-terminus of Jak3: mapping of the Jak3-receptor interaction domain, EMBO J. 18 (1999) 1549–1558. [4] D.C. Thomas, W. Lee, L.J. Berg, T cells from Jak3-deficient mice have intact TCR signaling, but increased apoptosis, J. Immunol. 159 (1997) 4708–4719. [5] K. Suzuki, H. Nakajima, Y. Saito, Kinase 3 (Jak3) is essential for common cytokine receptor c chain (cc)-dependent signaling: comparative analysis of cc, Jak3, and cc and Jak3 double-deficient mice, Int. Immunol. 12 (2000) 123– 132. [6] E.J. Baxter, L.M. Scott, P.J. Campbell, et al., Cancer genome project. Acquired mutation of the tyrosine kinase Jak2 in human myeloproliferative disorders, Lancet 365 (2005) 1054–1061. [7] C. James, V. Ugo, J.P. Le Couedic, et al., A unique clonal JAK2 mutation leading to constitutive signaling causes polycythaemia vera, Nature 434 (2005) 1144– 1148. [8] R. Kralavics, F. Passamonti, A.S. Buser, et al., A gain-of–function mutation of JAK2 in myeloproliferative disorders, N. Engl. J. Med. 352 (2005) 1779–1790. [9] Y. Minegishi, Human tyrosine kinase-2 deficiency reveals its requisite roles in multiple cytokine signals involved in innate and acquired immunity, Immunity 25 (2006) 745–755. [10] T. Schindler, W. Bornmann, P. Pellicena, et al., J. Structural mechanism for STI571 inhibition of Abelson tyrosine kinase, Science 289 (2000) 1938–1942.

548

B. Korniski et al. / Biochemical and Biophysical Research Communications 396 (2010) 543–548

[11] B. Nagar, O. Hantschel, M.A. Young, Structural basis for the autoinhibition of cAbl tyrosine kinase, Cell 112 (2003) 859–871. [12] Y.J. Zhou, E.P. Hanson, Y.Q. Chen, et al., Distinct tyrosine phosphorulation sites in JAK3 kinase domain positively and negatively regulate its enzymatic activity, Proc. Natl. Acad. Sci 94 (1997) 13850–13855. [13] M.C. Gauzzi, L. Velazquez, R. McKendry, et al., Interferon-a-dependent activation of Tyk2 requires phosphorylation of positive regulatory tyrosines by another kinase, J. Biol. Chem. 271 (1996) 20494–20500. [14] K.D. Liu, S.L. Gaffen, M.A. Goldsmith, W.C. Greene, Janus kinases in interleukin2-mediated signaling: JAK1 and JAK3 are differentially regulated by tyrosine phosphorylation, Curr. Biol. 7 (1997) 817–826. [15] F. Feng, B.A. Witthuhn, T. Matsuda, et al., Activation of Jak2 catalytic activity requires phosphorylation of Y1007 in the kinase activation loop, Mol. Cell Biol. 17 (1997) 2497–2501. [16] J.E. Chrencik, A. Patny, I.K. Leung, et al., Structural and thermodynamic characterization of the TYK2 and JAK3 kinase domains in complex with PAN inhibitors, J. Mol. Biol. (2010), in press. [17] T. Hall, T.E. Emmons, J. Chrencik, et al., Expression, purification, characterization and crystallization of non- and phosphorylated states of JAK2 and JAK3 kinase domain, Protein Expr. Purif. 69 (2010) 54–63.

[18] R.J. Leatherbarrow, Erithacus Software Ltd., 2001. Grafit Version 5. [19] R.A. Copeland, Enzymes: A Practical Introduction to Structure, Mechanism, and Data Analysis, Wiley-VCH, New York, 2000. [20] I.S. Lucet, E. Fantino, M. Styles, et al., The structural basis of Janus kinase 2 inhibition by a potent and specific pan-Janus kinase inhibitor, Blood 107 (2006) 176–183. [21] T.J. Boggon, Y. Li, P.W. Manley, M.J. Eck, Crystal structure of the Jak3 kinase domain in complex with a staurosporine analog, Blood 106 (2005) 996–1002. [22] P.S. Changelian, M.E. Flanagan, D.J. Ball, et al., Prevention of organ allograft rejection by a specific Janus kinase 3 inhibitor, Science 302 (2003) 875–878. [23] J.E. Thompson, R.M. Cubbon, R.T. Cummings, et al., Photochemical preparation of a pyridone containing tetracycle: a Jak protein kinase inhibitor, Bioorg. Med. Chem. Lett. 12 (2002) 1219–1223. [24] P.S. Changelian, D. Moshinsky, C.F. Kuhn, M.E. Flanagan, et al., The specificity of JAK3 kinase inhibitors, Blood 111 (2008) 2155–2157. [25] Z.A. Knight, K.M. Shokat, Features of selective kinase inhibitors, Chem. Biol. 12 (2005) 621–637. [26] B. Frantz, T. Klatt, M. Pang, et al., The activation state of p38 mitogen-activated protein kinase determines the efficiency of ATP competition for pyridinylimidazole inhibitor binding, Biochemistry 37 (1998) 13846–13853.