Research Summary Protein Structure Our laboratory is investigating the structural properties of molecules involved in cell signaling events. For many years our focus has been on the ligands of chemokine receptors, chemokines and defensins, with the aim of correlating structural and functional properties of these molecules. During last few years we were also engaged in studies of human glutamate carboxypeptidase II (GCPII), an enzyme involved in carcinogenesis and metastasis of prostate cancer and in several neurodegenerative diseases. That work was aimed at the development of novel, highly specific ligands/inhibitors of GCPII, with a potential in diagnosis/monitoring of prostate cancer and in the treatment of neuropathologies. Recently, we extended our research portfolio to two additional families, triggering receptors expressed on myeloid cells (TREMs) and Janus kinases (JAKs), with the latter area becoming the primary target of our current research. Whereas x-ray crystallography is our primary research tool, our research techniques also extend into biochemistry and molecular biology. Defensins Defensins are small, basic proteins that are primarily known for their potent antimicrobial properties. However, human β-defensins are also potent chemoattractants specifically interacting with CCR6-the receptor for the chemokine MIP-3α. The mechanisms underlying the antimicrobial and chemotactic activity of defensins are not well understood, although both o properties are of great practical interest. The crystal structures of human β-defensin-2 (hBD2), solved by our laboratory, were the first published for human β-defensins. These structures revealed a topology and an oligomerization mode for defensins that had not been reported before. Our findings shed new light on a possible mechanism for the antimicrobial properties of these proteins. Since these original findings, we completed extensive structural and functional studies for other human β-defensins (hBD1 and hBD4), their multiple mutants, and for all known human α-defensins (including many of their derivatives). The results of these efforts allowed identification of the N-terminal fragment of β-defensins molecules as the primary determinant of their CCR6-mediated chemotactic properties. We also described the roles of all conserved motifs/residues in defensins molecules and enhanced an understanding of antimicrobial properties of these proteins at the molecular level. In the future, we plan to extend and generalize the correlations between the structural and chemotactic properties of β-defensins as well as to study the structural basis of (recently identified) interactions between these proteins and other cellular receptors (i.e. Toll-like receptors). Human Glutamate Carboxypeptidase II GCPII (EC 3.4.17.21) is a glycosylated type II membrane metallopeptidase with an approximate molecular weight of 90-100 kDa. In healthy individuals, GCPII is expressed predominantly in the nervous system, small intestine, kidney, and prostate. Within the nervous system, GCPII hydrolyses N-acetyl-aspartyl-glutamate (NAAG), the most abundant neurotransmitter in human brain. Dysregulation of the glutamatergic transmission leads to various neuropathologies, and a therapeutic effect of GCPII inhibitors was demonstrated in animal models of stroke, ALS, schizophrenia, and inflammatory pain. One of the goals of our studies of GCPII and its homologs, GCPIII and NAALADase L, is to assist a development of novel inhibitors, targeting the "brain form" of GCPII, i.e. are capable to cross efficiently the blood-brain barrier (BBB). Our collaboration with Drs. Konvalinka, Slusher, Tsukamoto, and Pomper, supported by nearly 30 structures of GCPII complexes, allowed (i) an identification of key structural motifs/interactions defining the affinity and specificity of GCPII inhibitors, (ii) a detail description of the enzymatic mechanism and, (iii) development of a next generation of inhibitors with increased ability to permeate BBB. Whereas the function of GCPII in prostate is unknown, hGCPII was shown as an excellent target for prostate cancer imaging and therapy. Currently, a monoclonal antibody targeting intracellular domain of hGCPII is used in clinical setting for visualization of the prostate cancer progression. In addition to several limitations of this approach, utilizing the antibody greatly increases the costs of a cancer monitoring. A similar goal may be achieved by using small molecule markers targeting the active site of hGCPII, and the development of such agents is the second goal of our research on GCPII. TREMs TREMs are type I transmembrane proteins composed of a single extracellular immunoglobulin-like domain (IgD), a transmembrane segment (TD), and a cytoplasmic region (CD). The most thoroughly characterized members are TREM-1, TREM-2, and the TREM-like transcript-1 (TLT-1). TREM-1 and -2 are characterized by the presence of a charged residue in TD and a small size of CD (10-15 a.a. residues). In contrast, TLT-1 lacks any charged residue in the TD and its CD is significantly larger (approximately 130 a.a. residues). An engagement of TREM-1 was shown to enhance responses of host cytokines in mouse models of sepsis. The presence of a soluble form of TREM-1 in the blood was linked to several inflammatory conditions. The x-ray structure of IgD from TREM-1 was the first reported for any of the members of the TREM family. Binding of TREM-2 to the cell surface in cultured microglia results in suppression of cellular activation and inflammation. Mutations in the TREM-2 gene were linked to occurrence of early dementia. A recent finding that describes TREM-2 binding to semaphorin plexin A-1, for the first time links this protein to proper brain and bone function. To date, human TLT-1 has been detected only in megakaryocytes and platelets, and was shown to be sequestered in α-granules prior platelet activation. Recently, TLT-1 has been shown to play a role in platelet aggregation in vitro. In the past, we have developed protocols for efficient expression of the extracellular domain of TLT-1 (IgD-TLT-1), which we subsequently crystallized and determined its structure at the resolution of 1.2 Å. The structure allowed a clear illustration of the differences between the molecules of TLT-1 and other TREMS, such as TREM-1 or NKp44. However, the most interesting was identification of a similarity between 17 amino acid fragments of TLT-1 and TREM-1, which prompted us to hypothesize that this stretch of TLT-1 may function in blocking its cellular activity. That hypothesis is currently under investigation by our collaborators. Our present effort in structural characterization of the TREM family proteins focuses on the extracellular domain of TREM-2. JAKs Four Janus kinases are present in mammals (Jak1-3 and Tyk2). In contrast, a genome of Drosophila codes for only one JAK, Hopscotch, whereas no JAKs are present in yeast and in more evolutionary ancient organisms. Due to their central role in the immune host responses and the association with various pathologies, JAK/STAT signal transduction pathways are well-recognized as attractive molecular targets for therapeutic intervention. Due to the complexity of the JAK/STAT system, expressed by steadily growing number of component molecules, polymorphism, and cross-dependence on other signaling pathways, many critical steps of signal flow and the properties of molecules involved in JAK/STAT signaling remains poorly understood. The biology of the JAK/STAT pathways can be better consolidated, understood, and utilized if complemented by data derived from structural studies. Structural description of the molecular components of these systems is, however, highly unbalanced. While the extracellular components, cytokines and the extracellular fragments of cytokine receptors are currently reasonably well-characterized in structural terms and a substantial information is also available for STATs and SOCSs, a profound "blank spot" exists at the receptor/JAK interface. No structural data are available for the transmembrane and the intracellular domains of cytokine receptors and the only structures available for JAKs describe their protein tyrosine kinase domains (PTK), comprising less than 30% of the entire molecules. The molecular defects (mutations) in non-PTK regions of JAKs, however, are directly linked to several pathologies, particularly those of the immune and blood cells. One of the best known examples is an association between the mutation V617F (pseudokinase domain) of Jak2 and various myeloproliterative disorders (MPDs). Similar correlations are also described for mutations located in other (SH2-like and FERM) domains of JAKs. In addition to leading to blood disorders and malignancies, defects within the N-terminal fragment of Jak3 or TYK2 (FERM domain) have been identified in the patients with severe combined immunodeficiency (SCID). The importance of JAKs in human biology and disease, their relationship with cancer, and the lack of structural information generated our interest in studies the structural properties of these proteins. Our ultimate goal is to determine the structure of an entire JAK molecule, possibly in complex with a molecular partner (i.e. receptor, inhibitor, etc.). Due to the intrinsic properties of JAKs (size, multi-domain composition, potential lack of structural integrity), we are also pursuing research on their individual domains and on smaller multi-domain fragments. Whereas this project was initiated only recently, it now became the primary research activity of our Section.