Joseph W. Leone

Structural and thermodynamic characterization of the TYK2 and JAK3 kinase domains in complex with CP-690550 and CMP-6.

Janus kinases (JAKs) are critical regulators of cytokine pathways and attractive targets of therapeutic value in both inflammatory and myeloproliferative diseases. Although the crystal structures of active JAK1 and JAK2 kinase domains have been reported recently with the clinical compound CP-690550, the structures of both TYK2 and JAK3 with CP-690550 have remained outstanding. Here, we report the crystal structures of TYK2, a first in class structure, and JAK3 in complex with PAN-JAK inhibitors CP-690550 ((3R,4R)-3-[4-methyl-3-[N-methyl-N-(7H-pyrrolo[2,3-d]pyrimidin-4-yl)amino]piperidin-1-yl]-3-oxopropionitrile) and CMP-6 (tetracyclic pyridone 2-t-butyl-9-fluoro-3,6-dihydro-7H-benz[h]-imidaz[4,5-f]isoquinoline-7-one), both of which bind in the ATP-binding cavities of both JAK isozymes in orientations similar to that observed in crystal structures of JAK1 and JAK2. Additionally, a complete thermodynamic characterization of JAK/CP-690550 complex formation was completed by isothermal titration calorimetry, indicating the critical role of the nitrile group from the CP-690550 compound. Finally, computational analysis using WaterMap further highlights the critical positioning of the CP-690550 nitrile group in the displacement of an unfavorable water molecule beneath the glycine-rich loop. Taken together, the data emphasize the outstanding properties of the kinome-selective JAK inhibitor CP-690550, as well as the challenges in obtaining JAK isozyme-selective inhibitors due to the overall structural and sequence similarities between the TYK2, JAK1, JAK2 and JAK3 isozymes. Nevertheless, subtle amino acid variations of residues lining the ligand-binding cavity of the JAK enzymes, as well as the global positioning of the glycine-rich loop, might provide the initial clues to obtaining JAK-isozyme selective inhibitors.

High Resolution Crystal Structure of the Catalytic Domain of ADAMTS-5 (Aggrecanase-2)

Aggrecanase-2 (a disintegrin and metalloproteinase with thrombospondin motifs-5 (ADAMTS-5)), a member of the ADAMTS protein family, is critically involved in arthritic diseases because of its direct role in cleaving the cartilage component aggrecan. The catalytic domain of aggrecanase-2 has been refolded, purified, and crystallized, and its three-dimensional structure determined to 1.4Å resolution in the presence of an inhibitor. A high resolution structure of an ADAMTS/aggrecanase protein provides an opportunity for the development of therapeutics to treat osteoarthritis.

Identification of an ADAMTS-4 Cleavage Motif Using Phage Display Leads to the Development of Fluorogenic Peptide Substrates and Reveals Matrilin-3 as a Novel Substrate

ADAMTS-4 and ADAMTS-5 are aggrecanases responsible for the breakdown of cartilage aggrecan in osteoarthritis. Multiple ADAMTS-4 cleavage sites have been described in several matrix proteins including aggrecan, versican, and brevican, but no concise predictive cleavage motif has been identified for this protease. By screening a 13-mer peptide library with a diversity of 108, we have identified the ADAMTS-4 cleavage motif E-(AFVLMY)-X(0,1)-(RK)-X(2,3)-(ST)-(VYIFWMLA), with Glu representing P1. Several 13-mer peptides containing this motif, including DVQEFRGVTAVIR and HNEFRQRETYMVF, were shown to be substrates for ADAMTS-4. These peptides were found to be specific substrates for ADAMTS-4 as they were not cleaved by ADAMTS-5. Modification of these peptides with donor (6-FAM) and acceptor (QSY-9) molecules resulted in the development of fluorescence-based substrates with a Km of ∼35 μm. Furthermore, the role of Glu at P1 and Phe at P1′ in binding and catalysis was studied by exploring substitution of these amino acids with the d-isomeric forms. Substitution of P1 with dGlu was tolerable for binding, but not catalysis, whereas substitution of P1′ with dPhe precluded both binding and catalysis. Similarly, replacement of Glu with Asp at P1 abolished recognition and cleavage of the peptide. Finally, BLAST results of the ADAMTS-4 cleavage motif identified matrilin-3 as a new substrate for ADAMTS-4. When tested, recombinant ADAMTS-4 effectively cleaved intact matrilin-3 at the predicted motif at Glu435/Ala436 generating two species of 45 and 5 kDa.

ADAM-8 isolated from human osteoarthritic chondrocytes cleaves fibronectin at Ala271

OBJECTIVE Fibronectin fragments are thought to play a critical role in the initiation and progression of cartilage degradation in arthritis. In a recent study, fibronectin neoepitopes resulting from cleavage of intact fibronectin at the Ala(271)/Val(272) scissile bond, generating an approximately 30-kd fragment with the new C-terminus VRAA(271) and an approximately 50-85-kd fragment with the new N-terminus (272)VYQP, were identified in osteoarthritis (OA) cartilage. The present study was undertaken to isolate the enzymes responsible for this cleavage from human OA chondrocytes. METHODS Fibronectin-degrading activity in human OA chondrocyte-conditioned medium (OACCM) was purified using conventional chromatography. A fluorescent peptide was developed based on the fibronectin scissile bond (269)RAA downward arrowVal(272), and this peptide was used to track fibronectinase activity during purification. Western blotting with antibodies that detect the fibronectin neoepitopes VRAA(271) and (272)VYQP was used to confirm cleavage of intact fibronectin by the enzymatically active fractions. Mass spectrometry was used to identify the proteins found in the fibronectinase-enriched fractions, with further confirmation by Western blotting. In addition, a recombinant enzyme identified by mass spectrometry was tested by Western blotting and dimethylmethylene blue assay for its ability to produce fibronectin neoepitopes in OA cartilage. RESULTS Purification of OACCM by chromatography resulted in isolation of a fibronectin-degrading enzyme, and mass spectrometry identified ADAM-8 as the fibronectinase present in these preparations. Furthermore, treatment of OA cartilage with recombinant human ADAM-8 promoted cartilage catabolism. CONCLUSION The results of this study identify ADAM-8 as a fibronectinase in human OA chondrocytes. Because ADAM-8 is capable of producing the fibronectin neoepitopes VRAA(271) and (272)VYQP in human OA cartilage, this enzyme may be an important mediator of cartilage catabolism.

Autoactivation of human ADAM8: a novel pre- processing step is required for catalytic activity

Members of the ADAM (a disintegrin and metalloproteinase) family of proteins possess a multidomain architecture which permits functionalities as adhesion molecules, signalling intermediates and proteolytic enzymes. ADAM8 is found on immune cells and is induced by multiple pro-inflammatory stimuli suggesting a role in inflammation. Here we describe an activation mechanism for recombinant human ADAM8 that is independent from classical PC (pro-protein convertase)-mediated activation. N-terminal sequencing revealed that, unlike other ADAMs, ADAM8 undergoes pre-processing at Glu(158), which fractures the Pro (pro-segment)-domain before terminal activation takes place to remove the putative cysteine switch (Cys(167)). ADAM8 lacking the DIS (disintegrin) and/or CR (cysteine-rich) and EGF (epidermal growth factor) domains displayed impaired ability to complete this event. Thus pre-processing of the Pro-domain is co-ordinated by DIS and CR/EGF domains. Furthermore, by placing an EK (enterokinase) recognition motif between the Pro- and catalytic domains of multiple constructs, we were able to artificially remove the pro-segment prior to pre-processing. In the absence of pre-processing of the Pro-domain a marked decrease in specific activity was observed with the autoactivated enzyme, suggesting that the Pro-domain continued to associate and inhibit active enzyme. Thus, pre-processing of the Pro-domain of human ADAM8 is important for enzyme maturation by preventing re-association of the pro-segment with the catalytic domain. Given the observed necessity of DIS and CR/EGF for pre-processing, we conclude that these domains are crucial for the proper activation and maturation of human ADAM8.

Expression, purification, characterization and crystallization of non- and phosphorylated states of JAK2 and JAK3 kinase domain

Janus-associated kinases (JAKs) play critical roles in cytokine signaling, and have emerged as viable therapeutic targets in inflammation and oncology related diseases. To date, targeting JAK proteins with highly selective inhibitor compounds have remained elusive. We have expressed the active kinase domains for both JAK2 and JAK3 and devised purification protocols to resolve the non-, mono- (Y1007) and diphosphorylated (Y1007 and Y1008) states of JAK2 and non- and monophosphorylated states of JAK3 (Y980). An optimal purified protein yield of 20, 29 and 69mg per 20L cell culture was obtained for the three JAK2 forms, respectively, and 12.2 and 2.3mg per 10L fermentation for the two JAK3 forms allowing detailed biochemical and biophysical studies. To monitor the purification process we developed a novel HPLC activity assay where a sequential order of phosphorylation was observed whereby the first tyrosine residue was completely phosphorylated prior to phosphorylation of the tandem tyrosine residue. A Caliper-based microfluidics assay was used to determine the kinetic parameters (K(m) and k(cat)) for each phosphorylated state, showing that monophosphorylated (Y1007) JAK2 enzyme activity increased 9-fold over that of the nonphosphorylated species, and increased an additional 6-fold for the diphosphorylated (Y1007/Y1008) species, while phosphorylation of JAK3 resulted in a negligible increase in activity. Moreover, crystal structures have been generated for each isolated state of JAK2 and JAK3 with resolutions better than 2.4A. The generation of these reagents has enabled kinetic and structural characterization to inform the design of potent and selective inhibitors of the JAK family.

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

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- (pY(1054)) and di-phosphorylated forms of the enzyme. Kinetic studies (k(cat(app))/K(m(app))) indicated that phosphorylation of the TYK-2-KD (pY(1054)) 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 (K(i(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.

Intrasteric inhibition of ATP binding is not required to prevent unregulated autophosphorylation or signaling by the insulin receptor.

ABSTRACT Receptor tyrosine kinases may use intrasteric inhibition to suppress autophosphorylation prior to growth factor stimulation. To test this hypothesis we made an Asp1161Ala mutant in the activation loop that relieved intrasteric inhibition of the unphosphorylated insulin receptor (IR) and its recombinant cytoplasmic kinase domain (IRKD) without affecting the activated state. Solution studies with the unphosphorylated mutant IRKD demonstrated conformational changes and greater catalytic efficiency from a 10-fold increase ink cat and a 15-fold-lowerK m ATP althoughK m peptide was unchanged. Kinetic parameters of the autophosphorylated mutant and wild-type kinase domains were virtually identical. The Asp1161Ala mutation increased the rate of in vitro autophosphorylation of the IRKD or IR at low ATP concentrations and in the absence of insulin. However, saturation with ATP (for the IRKD) or the presence of insulin (for the IR) yielded equivalent rates of autophosphorylation for mutant versus wild-type kinases. Despite a biochemically more active kinase domain, the mutant IR expressed in C2C12 myoblasts was not constitutively autophosphorylated. However, it displayed a 2.5-fold-lower 50% effective concentration for insulin stimulation of autophosphorylation and was dephosphorylated more slowly following withdrawal of insulin than wild-type IR. In tests of the regulation of the unphosphorylated basal state, these results demonstrate that neither intrasteric inhibition against ATP binding nor suppression of kinase activity is required to prevent premature autophosphorylation of the IR. Finally, the lower rate of dephosphorylation suggests invariant residues of the activation loop such as Asp1161 may function at multiple junctures in cellular regulation of receptor tyrosine kinases.

Purification and characterization of recombinant human soluble guanylate cyclase produced from baculovirus-infected insect cells.

Soluble guanylate cyclase (sGC) has been purified from 100 L cell culture infected by baculovirus using the newer and highly effective titerless infected-cells preservation and scale-up (TIPS) method. Successive passage of the enzyme through DEAE, Ni(2+)-NTA, and POROS Q columns obtained approximately 100mg of protein. The sGC obtained by this procedure was already about 90% pure and suitable for various studies which include high throughput screening (HTS) and hit follow-up. However, in order to obtain enzyme of greater homogeneity and purity for crystallographic and high precision spectroscopic and kinetic studies of sGC with select stimulators, the sGC solution after the POROS Q step was further purified by GTP-agarose affinity chromatography. This additional step led to the generation of 26 mg of enzyme that was about 99% pure. This highly pure and active enzyme exhibited a M(r)=144,933 by static light scattering supportive of a dimeric structure. It migrated as a two-band protein, each of equal intensity, on SDS-PAGE corresponding to the alpha (M(r) approximately 77,000) and beta (M(r) approximately 70,000) sGC subunits. It showed an A(430)/A(280)=1.01, indicating one heme per heterodimer, and a maximum of the Soret band at 430 nm indicative of a penta-coordinated ferrous heme with a histidine as the axial ligand. The Soret band shifted to 398 nm in the presence of an NO donor as expected for the formation of a penta-coordinated nitrosyl-heme complex. Non-stimulated sGC had k(cat)/K(m)=1.7 x 10(-3)s(-1)microM(-1) that increased to 5.8 x 10(-1)s(-1)microM(-1) upon stimulation with an NO donor which represents a 340-fold increase due to stimulation. The novel combination of using the TIPS method for co-expression of a heterodimeric heme-containing enzyme, along with the application of a reproducible ligand affinity purification method, has enabled us to obtain recombinant human sGC of both the quality and quantity needed to study structure-function relationships.

Removal of N‐Terminal Blocking Groups from Proteins

Two enzymatic methods commonly used in N‐terminal sequence analysis of blocked proteins are presented: one uses pyroglutamate aminopeptidase for Nα‐pyrrolidone carboxyl‐proteins in solution or blotted onto a membrane, and the other uses acylaminoacyl‐peptide hydrolase for Nα‐acyl‐proteins blocked with other acyl groups. A Support Protocol describes a colorimetric assay for pyroglutamate aminopeptidase activity. Sequencing with acylaminoacyl‐peptide hydrolase must include fragmentation of the protein before unblocking, so procedures are provided for chemically blocking newly generated peptides with either succinic anhydride or phenylisothiocyanate/performic acid. The hydrolase is then applied to the total mixture of peptides, only one of which, the acylated N‐terminal peptide, should be a substrate for hydrolase. After incubation, the mixture of peptides is subjected to sequence analysis. Curr. Protoc. Protein Sci. 63:11.7.1‐11.7.20.