Introduction of raw starch-binding domains into Bacillus subtilis alpha-amylase by fusion with the starch-binding domain of Bacillus cyclomaltodextrin glucanotransferase

We constructed two types of chimeric enzymes, Ch1 Amy and Ch2 Amy. Ch1 Amy consisted of a catalytic domain of Bacillus subtilis X-23 alpha-amylase (Ba-S) and the raw starch-binding domain (domain E) of Bacillus A2-5a cyclomaltodextrin glucanotransferase (A2-5a CGT). Ch2 Amy consisted of Ba-S and D (function unknown) plus E domains of A2-5a CGT. Ch1 Amy acquired raw starch-binding and -digesting abilities which were not present in the catalytic part (Ba-S). Furthermore, the specific activity of Ch1 Amy was almost identical when enzyme activity was evaluated on a molar basis. Although Ch2 Amy exhibited even higher raw starch-binding and -digesting abilities than Ch1 Amy, the specific activity was lower than that of Ba-S. We did not detect any differences in other enzymatic characteristics (amylolytic pattern, transglycosylation ability, effects of pH, and temperature on stability and activity) among Ba-S, Ch1 Amy, and Ch2 Amy.     

Identification, structure, and differential expression of members of a BURP domain containing protein family in soybean.

Expressed sequence tags (ESTs) exhibiting homology to a BURP domain containing gene family were identified from the Glycine max (L.) Merr. EST database. These ESTs were assembled into 16 contigs of variable sizes and lengths. Consistent with the structure of known BURP domain containing proteins, the translation products exhibit a modular structure consisting of a C-terminal BURP domain, an N-terminal signal sequence, and a variable internal region. The soybean family members exhibit 35-98% similarity in a -100-amino-acid C-terminal region, and a phylogenetic tree constructed using this region shows that some soybean family members group together in closely related pairs, triplets, and quartets, whereas others remain as singletons. The structure of these groups suggests that multiple gene duplication events occurred during the evolutionary history of this family. The depth and diversity of G. max EST libraries allowed tissue-specific expression patterns of the putative soybean BURPs to be examined. Consistent with known BURP proteins, the newly identified soybean BURPs have diverse expression patterns. Furthermore, putative paralogs can have both spatially and quantitatively distinct expression patterns. We discuss the functional and evolutionary implications of these findings, as well as the utility of EST-based analyses for identifying and characterizing gene families.     

Structure, specificity and function of cyclomaltodextrinase, a multispecific enzyme of the alpha-amylase family

Cyclomaltodextrinase (CDase, EC 3.2.1.54), maltogenic amylase (EC 3. 2.1.133), and neopullulanase (EC 3.2.1.135) are reported to be capable of hydrolyzing all or two of the following three types of substrates: cyclomaltodextrins (CDs); pullulan; and starch. These enzymes hydrolyze CDs and starch to maltose and pullulan to panose by cleavage of alpha-1,4 glycosidic bonds whereas alpha-amylases essentially lack activity on CDs and pullulan. They also catalyze transglycosylation of oligosaccharides to the C3-, C4- or C6-hydroxyl groups of various acceptor sugar molecules. The present review surveys the biochemical, enzymatic, and structural properties of three types of such enzymes as defined based on the substrate specificity toward the CDs: type I, cyclomaltodextrinase and maltogenic amylase that hydrolyze CDs much faster than pullulan and starch; type II, Thermoactinomyces vulgaris amylase II (TVA II) that hydrolyzes CDs much less efficiently than pullulan; and type III, neopullulanase that hydrolyzes pullulan efficiently, but remains to be reported to hydrolyze CDs. These three types of enzymes exhibit 40-60% amino acid sequence identity. They occur in the cytoplasm of bacteria and have molecular masses from 62 to 90 kDa which are slightly larger than those of most alpha-amylases. Multiple amino acid sequence alignment and crystal structures of maltogenic amylase and TVA II reveal the presence of an N-terminal extension of approximately 130 residues not found in alpha-amylases. This unique N-terminal domain as seen in the crystal structures apparently contributes to the active site structure leading to the distinct substrate specificity through a dimer formation. In aqueous solution, most of these enzymes show a monomer-dimer equilibrium. The present review discusses the multiple specificity in the light of the oligomerization and the molecular structures arriving at a clarified enzyme classification. Finally, a physiological role of the enzymes is proposed.     

Improved adsorption to starch of a beta-galactosidase fusion protein containing the starch-binding domain from Aspergillus glucoamylase.

We have previously shown (Chen et al., 1991) that a beta-galactosidase (beta-gal) fusion protein (BSB133) containing 133 amino acids (aa) from the C-terminus of Aspergillus glucoamylase (GA) adsorbs strongly to starch compared to beta-gal, due to the presence of the GA starch-binding domain. We have now made deletions at the N-terminus of this 133-aa region to test the minimal size required for starch binding of beta-gal fusion proteins. Three fusion proteins (BSB119, BSB103, and BSB80) were genetically engineered, containing 119, 103, and 80 C-terminal aa from GA, respectively. The fusion proteins were expressed in Escherichia coli and purified. Purified BSB119 adsorbed to native starch at least 2-fold more strongly than did BSB133 or fusion proteins with shorter tails. Adsorption isotherms generated over a wide range of initial concentrations indicated a 10-fold difference in the loading capacity of starch for BSB119 (36.5 mg of protein/g of starch) compared to beta-gal (3.7 mg of protein/g of starch). Adsorption constants calculated from the initial slopes of the isotherms indicated a nearly 30-fold difference in affinity to starch for BSB119 (Kad = 63 mL/g of starch) compared to beta-gal (Kad = 2.3 mL/g of starch). BSB119 in the presence of crude enzyme extracts also bound to starch with a high affinity compared to a beta-gal control. Potential applications of the starch-binding tail include enzyme immobilization to starch or recovery and purification of target proteins from crude extracts.     

Complex structures of Thermoactinomyces vulgaris R-47 alpha-amylase 1 with malto-oligosaccharides demonstrate the role of domain N acting as a starch-binding domain

The X-ray structures of complexes of Thermoactinomyces vulgaris R-47 alpha-amylase 1 (TVAI) with an inhibitor acarbose and an inactive mutant TVAI with malto-hexaose and malto-tridecaose have been determined at 2.6, 2.0 and 1.8A resolution, and the structures have been refined to R-factors of 0.185 (R(free)=0.225), 0.184 (0.217) and 0.164 (0.200), respectively, with good chemical geometries. Acarbose binds to the catalytic site of TVAI, and interactions between acarbose and the enzyme are very similar to those found in other structure-solved alpha-amylase/acarbose complexes, supporting the proposed catalytic mechanism. Based on the structure of the TVAI/acarbose complex, the binding mode of pullulan containing alpha-(1,6) glucoside linkages could be deduced. Due to the structural difference caused by the replaced amino acid residue (Gln396 for Glu) in the catalytic site, malto-hexaose and malto-tridecaose partially bind to the catalytic site, giving a mimic of the enzyme/product complex. Besides the catalytic site, four sugar-binding sites on the molecular surface are found in these X-ray structures. Two sugar-binding sites in domain N hold the oligosaccharides with a regular helical structure of amylose, which suggests that the domain N is a starch-binding domain acting as an anchor to starch in the catalytic reaction of the enzyme. An assay of hydrolyzing activity for the raw starches confirmed that TVAI can efficiently hydrolyze raw starch.     

The role of the C-terminal domain in collagenase and stromelysin specificity.

Recombinant human interstitial collagenase, an N-terminal truncated form, delta 243-450 collagenase, recombinant human stromelysin-1, and an N-terminal truncated form, delta 248-460 stromelysin, have been stably expressed in myeloma cells and purified. The truncated enzymes were similar in properties to their wild-type counterparts with respect to activation requirements and the ability to degrade casein, gelatin, and a peptide substrate, but truncated collagenase failed to cleave native collagen. Removal of the C-terminal domain from collagenase also modified its interaction with tissue inhibitor of metalloproteinases-1. Hybrid enzymes consisting of N-terminal (1-242) collagenase.C-terminal (248-460) stromelysin and N-terminal (1-233) stromelysin.C-terminal (229-450) collagenase, representing an exchange of the complete catalytic and C-terminal domains of the two enzymes, were expressed in a transient system using Chinese hamster ovary cells and purified. Both proteins showed similar activity to their N-terminal parent and neither was able to degrade collagen. Analysis of the ability of the different forms of recombinant enzyme to bind to collagen by ELISA showed that both pro and active stromelysin and N-terminal collagenase.C-terminal stromelysin bound to collagen equally well. In contrast, only the active forms of collagenase and N-terminal stromelysin.C-terminal collagenase bound well to collagen, as compared with their pro forms.     

Phylogeny, taxonomy, and evolution of the endothelin receptor gene family

A gene phylogeny provides the natural historical order to classify genes and to understand their functional, structural, and genomic diversity. The gene family of endothelin receptors (EDNR) is responsible for many key physiological and developmental processes of tetrapods and teleosts. This study provides a well-defined gene phylogeny for the EDNR family, which is used to classify its members and to assess their evolution. The EDNR phylogeny supports the recognition of the EDNRA, EDNRB, and EDNRC subfamilies, as well as more lineage-specific duplicates of teleosts and the African clawed frog. The duplications for these nominal genes are related to the various whole-genome amplifications of vertebrates, jawed vertebrates, fishes, and frog. The EDNR phylogeny also identifies several gene losses, including that of EDNRC from placental and marsupial (therian) mammals. When coupled with structural and biochemical information, site-specific analyses of evolutionary rate shifts reveal two distinct patterns of potential functional changes at the sequence level between therian versus non-therian EDNRA and EDNRB (i.e., between groups without and with EDNRC). An analysis of linkage maps and tetrapod synteny further suggests that the loss of therian EDNRC may be related to a chromosomal deletion in its common ancestor.     

Caveolin-3 directly interacts with the C-terminal tail of beta -dystroglycan. Identification of a central WW-like domain within caveolin family members

Caveolin-3, the most recently recognized member of the caveolin gene family, is muscle-specific and is found in both cardiac and skeletal muscle, as well as smooth muscle cells. Several independent lines of evidence indicate that caveolin-3 is localized to the sarcolemma, where it associates with the dystrophin-glycoprotein complex. However, it remains unknown which component of the dystrophin complex interacts with caveolin-3. Here, we demonstrate that caveolin-3 directly interacts with beta-dystroglycan, an integral membrane component of the dystrophin complex. Our results indicate that caveolin-3 co-localizes, co-fractionates, and co-immunoprecipitates with a fusion protein containing the cytoplasmic tail of beta-dystroglycan. In addition, we show that a novel WW-like domain within caveolin-3 directly recognizes the extreme C terminus of beta-dystroglycan that contains a PPXY motif. As the WW domain of dystrophin recognizes the same site within beta-dystroglycan, we also demonstrate that caveolin-3 can effectively block the interaction of dystrophin with beta-dystroglycan. In this regard, interaction of caveolin-3 with beta-dystroglycan may competitively regulate the recruitment of dystrophin to the sarcolemma. We discuss the possible implications of our findings in the context of Duchenne muscular dystrophy.     

The intermolecular interaction between the PH domain and the C-terminal domain of Arabidopsis dynamin-like 6 determines lipid binding specificity

Dynamin and its related proteins are a group of mechanochemical proteins involved in the modulation of lipid membranes in various biological processes. Here we investigate the nature of membrane binding of the Arabidopsis dynamin-like 6 (ADL6) involved in vesicle trafficking from the trans-Golgi network to the central vacuole. Fractionation experiments by continuous sucrose gradients and gel filtration revealed that the majority of ADL6 is associated with membranes in vivo. Amino acid sequence analysis revealed that ADL6 has a putative pleckstrin homology (PH) domain. In vitro lipid binding assays demonstrated that ADL6 showed high affinity binding to phosphatidylinositol 3-phosphate (PtdIns-3-P) and that the PH domain was responsible for this interaction. However, the PH domain alone binds equally well to both PtdIns-3-P and phosphatidylinositol 4-phosphate (PtdIns-4-P). Interestingly, the high affinity binding of the PH domain to PtdIns-3-P was restored by a protein-protein interaction between the PH domain and the C-terminal region. In addition, deletion of the inserted regions within the PH domain results in high affinity binding of the PH domain to PtdIns-3-P. These results suggest that ADL6 binds specifically to PtdIns-3-P and that the lipid binding specificity is determined by the interaction between the PH domain and the C-terminal domain of ADL6.     

The mouse alpha-amylase multigene family. Sequence organization of members expressed in the pancreas, salivary gland and liver

The DNAs that specify the α-amylase messenger RNAs found in the pancreas, salivary gland and liver of mouse strain A have been isolated by molecular cloning in phage λ. Amylase clones were studied by mRNA/DNA hybrid analysis in the electron microscope, restriction endonuclease site mapping and DNA sequencing. The Amy-2^a gene, which specifies pancreatic α-amylase mRNA, measures 10·1 kb from cap to polyadenylation site and is interrupted by at least 9 intervening sequences. Amy-1^a, which specifies both salivary gland and liver α-amylase mRNAs contains at least 10 introns. The distance between the cap and polyadenylation sites used in the salivary gland and the liver measures 22·9 kb and 20 kb, respectively. Introns are located at very similar, if not identical, positions within comparable regions of Amy-1^a and Amy-2^a. The first intron of Amy-1^a, which interrupts sequences specifying 5′ non-translated regions of salivary gland and liver α-amylase mRNAs, has no counterpart in Amy-2^a. Some introns exhibit considerable sequence homology, suggesting that Amy-1^a and Amy-2^a have evolved by duplication from a common split ancestor sequence. Repetitive sequence elements occur in the introns and flanking regions of these genes. Gene titration by quantitative autoradiography reveals only one copy of Amy-1^a, but two copies of Amy-2^a per haploid mouse genome. In addition to Amy-1^a and Amy-2^a, several other amylase-like DNA sequences exist in the mouse genome. No gross rearrangements of amylase DNA sequences can be detected between germline DNA and that of various mouse tissues.     

Relationship of sequence and structure to specificity in the alpha-amylase family of enzymes

The hydrolases and transferases that constitute the alpha-amylase family are multidomain proteins, but each has a catalytic domain in the form of a (beta/alpha)(8)-barrel, with the active site being at the C-terminal end of the barrel beta-strands. Although the enzymes are believed to share the same catalytic acids and a common mechanism of action, they have been assigned to three separate families - 13, 70 and 77 - in the classification scheme for glycoside hydrolases and transferases that is based on amino acid sequence similarities. Each enzyme has one glutamic acid and two aspartic acid residues necessary for activity, while most enzymes of the family also contain two histidine residues critical for transition state stabilisation. These five residues occur in four short sequences conserved throughout the family, and within such sequences some key amino acid residues are related to enzyme specificity. A table is given showing motifs distinctive for each specificity as extracted from 316 sequences, which should aid in identifying the enzyme from primary structure information. Where appropriate, existing problems with identification of some enzymes of the family are pointed out. For enzymes of known three-dimensional structure, action is discussed in terms of molecular architecture. The sequence-specificity and structure-specificity relationships described may provide useful pointers for rational protein engineering.     

Amylosucrase, a glucan-synthesizing enzyme from the alpha-amylase family

Amylosucrase (E.C. 2.4.1.4) is a member of Family 13 of the glycoside hydrolases (the alpha-amylases), although its biological function is the synthesis of amylose-like polymers from sucrose. The structure of amylosucrase from Neisseria polysaccharea is divided into five domains: an all helical N-terminal domain that is not similar to any known fold, a (beta/alpha)(8)-barrel A-domain, B- and B'-domains displaying alpha/beta-structure, and a C-terminal eight-stranded beta-sheet domain. In contrast to other Family 13 hydrolases that have the active site in the bottom of a large cleft, the active site of amylosucrase is at the bottom of a pocket at the molecular surface. A substrate binding site resembling the amylase 2 subsite is not found in amylosucrase. The site is blocked by a salt bridge between residues in the second and eight loops of the (beta/alpha)(8)-barrel. The result is an exo-acting enzyme. Loop 7 in the amylosucrase barrel is prolonged compared with the loop structure found in other hydrolases, and this insertion (forming domain B') is suggested to be important for the polymer synthase activity of the enzyme. The topology of the B'-domain creates an active site entrance with several ravines in the molecular surface that could be used specifically by the substrates/products (sucrose, glucan polymer, and fructose) that have to get in and out of the active site pocket.     

Monoclonal antibody with specificity to a conserved epitope in the C-terminal domain of histone H1 variants.

A monoclonal type M-immunoglobulin (IgM) was generated in mice against a nuclease-urea extract of HeLa metaphase chromosomes. This antibody stains metaphase chromosomes from a variety of mammalian cultured cell types by indirect immunofluorescence. Antibody 12C7 reacts by western transfer technique with histone H1 in all the cell lines tested. The antibody cross-reacts with H1, and H1(0) in human cells. Proteolytic digestions of H1 suggest that the epitope is localized in the carboxy-terminal domain of the histone H1 molecule. Digestion with trypsin demonstrates that the antibody 12C7 does not react with the globular domain of histone H1. The C-terminal domain of H1 subtypes therefore seems to have a conserved determinant which does exist in H1, H1(0), and probably in H5. This antibody has applications in studying the role of that domain of H1 in processes like chromosome condensation and variations in chromatin structure which influence gene expression.     

A structural model of the plant acyl-acyl carrier protein thioesterase FatB comprises two helix/4-stranded sheet domains, the N-terminal domain containing residues that affect specificity and the C-terminal domain containing catalytic residues

Plant acyl-acyl carrier protein thioesterases (TEs) terminate the acyl-acyl carrier protein track of fatty acid biosynthesis and play an essential role in determining the amount and composition of fatty acids entering the storage lipid pool. A combination of bioinformatics tools was used to predict a three-dimensional model for Arabidopsis FatB (AtFatB), which comprises a fold similar to that of Escherichia coli TEII, an enzyme that is functionally similar to plant TEs but lacks significant sequence similarity and displays different inhibitor sensitivity. The catalytic residues in AtFatB, Cys-264 and His-229, localize to the same region of the model as catalytic residues found in other enzymes with helix/multi-stranded sheet motifs (hot dog folds). Based on the model, we identified Asn-227 as a possible third member of the proposed papain-like catalytic triad. The conversion of Asn-227 to Ala resulted in a loss of detectable activity (>200-fold reduction), similar to the result seen for the equivalent mutation in papain. Mapping of plant TE specificity-affecting mutations onto the structural model showed that these mutations all cluster around the catalytic triad. Also, superposition of the crystallographically determined structures of the complexes of 4-hydroxybenzoyl-CoA TE with substrate and beta-hydroxydecanoyl thiol ester dehydrase with inhibitor onto the AtFatB model showed that the substrate and inhibitor localize to the same region as the AtFatB catalytic triad in their respective structures. Together these data corroborate the structural model and show that the hot dog fold is common to enzymes from both prokaryotes and eukaryotes and that this fold supports at least three different catalytic mechanisms.     

A family IIb xylan-binding domain has a similar secondary structure to a homologous family IIa cellulose-binding domain but different ligand specificity.

BACKGROUND: Many enzymes that digest polysaccharides contain separate polysaccharide-binding domains. Structures have been previously determined for a number of cellulose-binding domains (CBDs) from cellulases. RESULTS: The family IIb xylan-binding domain 1 (XBD1) from Cellulomonas fimi xylanase D is shown to bind xylan but not cellulose. Its structure is similar to that of the homologous family IIa CBD from C. fimi Cex, consisting of two four-stranded beta sheets that form a twisted 'beta sandwich'. The xylan-binding site is a groove made from two tryptophan residues that stack against the faces of the sugar rings, plus several hydrogen-bonding polar residues. CONCLUSIONS: The biggest difference between the family IIa and IIb domains is that in the former the solvent-exposed tryptophan sidechains are coplanar, whereas in the latter they are perpendicular, forming a twisted binding site. The binding sites are therefore complementary to the secondary structures of the ligands cellulose and xylan. XBD1 and CexCBD represent a striking example of two proteins that have high sequence similarity but a different function.     

Identification of the Otopetrin Domain,a conserved domain in vertebrate otopetrins and invertebrate otopetrin-like family members

Background  

Otopetrin 1 (Otop1) encodes a multi-transmembrane domain protein with no homology to known transporters, channels, exchangers, or receptors. Otop1 is necessary for the formation of otoconia and otoliths, calcium carbonate biominerals within the inner ear of mammals and teleost fish that are required for the detection of linear acceleration and gravity. Vertebrate Otop1 and its paralogues Otop2 and Otop3 define a new gene family with homology to the invertebrate Domain of Unknown Function 270 genes (DUF270; pfam03189).     

The dermaseptin precursors: a protein family with a common preproregion and a variable C-terminal antimicrobial domain.

Preprodermaseptins are a group of antimicrobial peptide precursors found in the skin of a variety of frog species. Precursors of this family have very similar N-terminal preprosequences followed by markedly different C-terminal domains that correspond to mature antimicrobial peptides. Some of these peptides are 24-34 amino acids long and form well-behaved amphipathic alpha-helices, others are disulfide-linked peptides of 20-46 residues, still others, highly hydrophobic, are the smallest antimicrobial peptides known so far being only 10-13 residues in length. All these peptides are broad-spectrum microbicides that kill many bacteria, protozoa, yeasts and fungi by destroying or permeating the microbial membrane. In frogs belonging to the genus Phyllomedusinae, preprodermaseptins encoded peptides also include dermorphins and deltorphins, D-amino acid-containing heptapeptides which are very potent and specific agonists of the mu- or delta-opioid receptors. The remarkable similarity between preproregions of precursors that give rise to peptides with very different primary structures, conformations and activities suggests that the corresponding genes originate from a common ancestor. The high conservation of the precursor prepropart indicates that this region must have an important function.     

Structural evolution of C-terminal domains in the p53 family

The tetrameric state of p53, p63, and p73 has been considered one of the hallmarks of this protein family. While the DNA binding domain (DBD) is highly conserved among vertebrates and invertebrates, sequences C-terminal to the DBD are highly divergent. In particular, the oligomerization domain (OD) of the p53 forms of the model organisms Caenorhabditis elegans and Drosophila cannot be identified by sequence analysis. Here, we present the solution structures of their ODs and show that they both differ significantly from each other as well as from human p53. CEP-1 contains a composite domain of an OD and a sterile alpha motif (SAM) domain, and forms dimers instead of tetramers. The Dmp53 structure is characterized by an additional N-terminal beta-strand and a C-terminal helix. Truncation analysis in both domains reveals that the additional structural elements are necessary to stabilize the structure of the OD, suggesting a new function for the SAM domain. Furthermore, these structures show a potential path of evolution from an ancestral dimeric form over a tetrameric form, with additional stabilization elements, to the tetramerization domain of mammalian p53.     

The dimerization and topological specificity functions of MinE reside in a structurally autonomous C-terminal domain

Correct placement of the division septum in Escherichia coli requires the co-ordinated action of three proteins, MinC, MinD and MinE. MinC and MinD interact to form a non-specific division inhibitor that blocks septation at all potential division sites. MinE is able to antagonize MinCD in a topologically sensitive manner, as it restricts MinCD activity to the unwanted division sites at the cell poles. Here, we show that the topological specificity function of MinE residues in a structurally autonomous, trypsin-resistant domain comprising residues 31-88. Nuclear magnetic resonance (NMR) and circular dichroic spectroscopy indicate that this domain includes both alpha and beta secondary structure, while analytical ultracentrifugation reveals that it also contains a region responsible for MinE homodimerization. While trypsin digestion indicates that the anti-MinCD domain of MinE (residues 1-22) does not form a tightly folded structural domain, NMR analysis of a peptide corresponding to MinE1-22 indicates that this region forms a nascent helix in which the peptide rapidly interconverts between disordered (random coil) and alpha-helical conformations. This suggests that the N-terminal region of MinE may be poised to adopt an alpha-helical conformation when it interacts with the target of its anti-MinCD activity, presumably MinD.     

Structure, function and tissue forms of the C-terminal globular domain of collagen XVIII containing the angiogenesis inhibitor endostatin.

The C-terminal domain NC1 of mouse collagen XVIII (38 kDa) and the shorter mouse and human endostatins (22 kDa) were prepared in recombinant form from transfected mammalian cells. The NC1 domain aggregated non-covalently into a globular trimer which was partially cleaved by endogenous proteolysis into several monomers (25-32 kDa) related to endostatin. Endostatins were obtained in a highly soluble, monomeric form and showed a single N-terminal sequence which, together with other data, indicated a compact folding. Endostatins and NC1 showed a comparable binding activity for the microfibrillar fibulin-1 and fibulin-2, and for heparin. Domain NC1, however, was a distinctly stronger ligand than endostatin for sulfatides and the basement membrane proteins laminin-1 and perlecan. Immunological assays demonstrated endostatin epitopes on several tissue components (22-38 kDa) and in serum (120-300 ng/ml), the latter representing the smaller variants. The data indicated that the NC1 domain consists of an N-terminal association region (approximately 50 residues), a central protease-sensitive hinge region (approximately 70 residues) and a C-terminal stable endostatin domain (approximately 180 residues). They also demonstrated that proteolytic release of endostatin can occur through several pathways, which may lead to a switch from a matrix-associated to a more soluble endocrine form.