Increases in npoh-OLF thermal stability in the presence of GAGs are modest (Table 2), and while npoh-OLF devoid of bound nucleotides can bind a heparin column, this is only in buffers with non-physiological, low ionic strength (Fig 4D). GSK2982772 organisms. Even though these proteins have been implicated in human disorders ranging from cancers to attention deficit disorder to glaucoma, little is known about their structure(s) and function(s). Here we biophysically, biochemically, and structurally characterize OLF domains from olfactomedin-1 (npoh-OLF, also called noelin, pancortin, OLFM1, and hOlfA), and gliomedin (glio-OLF, also called collomin, collmin, and CRG-L2), and compare them with available structures of myocilin (myoc-OLF) recently reported by us and glio-OLF and latrophilin-3 (lat3-OLF) by others. Although the five-bladed -propeller architecture remains unchanged, numerous physicochemical characteristics differ among these OLF domains. First, npoh-OLF and glio-OLF exhibit prominent, yet distinct, positive surface charges and copurify with polynucleotides. Second, whereas npoh-OLF and myoc-OLF exhibit thermal stabilities typical of human proteins near 55C, and most myoc-OLF variants are destabilized and highly prone to aggregation, glio-OLF is nearly 20C more stable and significantly more resistant to chemical denaturation. Phylogenetically, glio-OLF is most similar to primitive OLFs, and structurally, glio-OLF is missing distinguishing features seen in OLFs such as the disulfide bond formed by N- and C- terminal cysteines, the sequestered Ca2+ ion within the propeller central hydrophilic cavity, and a key loop-stabilizing cation- interaction on the top face of npoh-OLF and myoc-OLF. While deciphering the explicit biological functions, ligands, and binding partners for OLF domains will likely continue to be a challenging long-term experimental pursuit, we used structural insights gained here to generate a new antibody selective for myoc-OLF over npoh-OLF and glio-OLF as a first step in overcoming the impasse in detailed functional characterization of these biomedically important protein domains. Introduction Olfactomedins comprise a large protein family (PFAM: PF02191) with seven phylogenetic branches [1]. These multidomain proteins contain a ~30 kDa olfactomedin (OLF) domain and are predominantly expressed extracellularly in a variety of tissues of multicellular organisms, particularly in vertebrates as well as selected invertebrates [2]. Although the specific biological functions, binding partners, and mechanism(s) of action still remain largely unknown, the involvement of OLF domain-containing proteins in diseases is broadly documented, Sfpi1 particularly in the case of glaucoma [3], but also in a host of cancers [2], inflammatory bowel disorder and Crohns/colitis [4], defense against infection [5], attention deficit disorder [6], and childhood obesity [7]. The difficulty in assigning discrete biological function to OLFs is due in part to the fact that in general, reports of partial deletion mutants or knock-out mice of a variety of OLF domain-containing proteins do indicate a strong phenotype, e.g. gross abnormalities or systemic disease [8C11]. This observation, combined with considerable sequence similarity [1], has suggested that OLF domains might exhibit somewhat compensatory functions [2]. However, the extents of such overlap, or interconnectedness in function and/or binding partners, remain major open questions. For example, we recently reported the crystal structure of the best studied GSK2982772 OLF domain from myocilin (myoc-OLF) [12], a protein linked to inherited forms of glaucoma in populations throughout GSK2982772 the world. The leading proposed pathogenic mechanism involves intracellular aggregation leading to GSK2982772 cell death in the trabecular meshwork, a tissue of the eye implicated in maintaining pressure; high pressure is a major risk factor for glaucoma [3]. Myoc-OLF variants are exquisitely prone to misfolding, corresponding aggregates exhibit characteristics of amyloid [13, 14] and in cells [14], and have aberrant interactions with molecular chaperones [15, 16]. However, the lack of myocilin does not cause glaucoma, and in spite of considerable research efforts over the past 20 years, the normal functional role of wild type myocilin in the trabecular meshwork remains unclear [3]. The availability of the myoc-OLF 5-bladed -propeller structure, along with two others recently reported, (gliomedin (glio-OLF) [17] and latrophilin-3 (lat3-OLF) [18]), should provide new insight and enable the development of selective reagents, such as antibodies or small molecules, to better probe OLF function and pathophysiology. Here we present biophysical, biochemical, and structural characterization of the OLF domain of a second glio-OLF, that from (94%.