Location of the disulfide bonds in human coagulation factor XI: the presence of tandem apple domains

Factor XI is a plasma glycoprotein that participates in the blood coagulation cascade. Of the 19 disulfide bonds present in each of the subunits of the human protein, 16 were determined by amino acid sequence analysis of peptide fragments produced by chemical and enzymatic digestion. Four apple domains of 90 or 91 amino acids were identified in the tandem repeats present in the amino-terminal portion of each subunit of factor XI. The disulfide bonds in the carboxyl-terminal portion of the molecule were similar to those in the catalytic region of other serine proteases. The two identical subunits of factor XI were connected by a single disulfide bond at Cys321 linking each of the fourth apple domains while each of the Cys residues at position 11 in the first apple domains forms a disulfide bond with another Cys residue.     

Amino acid sequence of human factor XI, a blood coagulation factor with four tandem repeats that are highly homologous with plasma prekallikrein

A lambda gtll cDNA library prepared from human liver poly(A) RNA has been screened with affinity-purified antibody to human factor XI, a blood coagulation factor composed of two identical polypeptide chains linked by a disulfide bond(s). A cDNA insert coding for factor XI was isolated and shown to contain 2097 nucleotides, including 54 nucleotides coding for a leader peptide of 18 amino acids and 1821 nucleotides coding for 607 amino acids that are present in each of the 2 chains of the mature protein. The cDNA for factor XI also contained a stop codon (TGA), a potential polyadenylation or processing sequence (AACAAA), and a poly(A) tail at the 3' end. Five potential N-glycosylation sites were found in each of the two chains of factor XI. The cleavage site for the activation of factor XI by factor XIIa was identified as an internal peptide bond between Arg-369 and Ile-370 in each polypeptide chain. This was based upon the amino acid sequence predicted by the cDNA and the amino acid sequence previously reported for the amino-terminal portion of the light chain of factor XI. Each heavy chain of factor XIa (369 amino acids) was found to contain 4 tandem repeats of 90 (or 91) amino acids plus a short connecting peptide. Each repeat probably forms a separate domain containing three internal disulfide bonds. The light chains of factor XIa (each 238 amino acids) contain the catalytic portion of the enzyme with sequences that are typical of the trypsin family of serine proteases. The amino acid sequence of factor XI shows 58% identity with human plasma prekallikrein.     

Disulfide bonds are localized within the short consensus repeat units of complement regulatory proteins: C4b-binding protein

Several plasma and membrane proteins belong to a superfamily of structurally related proteins that contain internal homology of a variable number (2-30) of repeating units. Each SCR (short consensus repeat) unit is approximately 60 amino acid residues in length, with the positions of 1 Trp, 2 Pro, and 4 Cys residues being conserved. The aim of this study was to provide experimental evidence that each SCR may exist as an independent structural domain maintained by disulfide bonds. The well-characterized C4b-binding protein (C4BP) with eight SCR units in each of its seven identical chains was chosen for this study. Analysis of the disulfide-bonding pattern indicated that intrachain disulfide bonds may be localized within each SCR unit, with the first and third and the second and fourth half-cystines in each unit being linked. This pattern of disulfides may confer to C4BP (and to other structurally related proteins) a conformation which apparently allows the assembly of the SCR units (4-30) in a tandem fashion. Such an arrangement of the polypeptide chain(s) may explain, in part, the elongated shape of these protein molecules. The structural motif of the SCR units of C4BP is discussed in relation to those previously described for the type II domain of fibronectin and the kringle structure present in various proteins of the coagulation system.     

Human plasma prekallikrein, a zymogen to a serine protease that contains four tandem repeats

The amino acid sequence of human plasma prekallikrein was determined by a combination of automated Edman degradation and cDNA sequencing techniques. Human plasma prekallikrein was fragmented with cyanogen bromide, and 13 homogeneous peptides were isolated and sequenced. Cyanogen bromide peptides containing carbohydrate were further digested with trypsin, and the peptides containing carbohydrate were isolated and sequenced. Five asparagine-linked carbohydrate attachment sites were identified. The sequence determined by Edman degradation was aligned with the amino acid sequence predicted from cDNAs isolated from a lambda gt11 expression library. This library contained cDNA inserts prepared from human liver poly(A) RNA. Analysis of the cDNA indicated that human plasma prekallikrein is synthesized as a precursor with a signal peptide of 19 amino acids. The mature form of the protein that circulates in blood is a single-chain polypeptide of 619 amino acids. Plasma prekallikrein is converted to plasma kallikrein by factor XIIa by the cleavage of an internal Arg-Ile bond. Plasma kallikrein is composed of a heavy chain (371 amino acids) and a light chain (248 amino acids), and these 2 chains are held together by a disulfide bond. The heavy chain of plasma kallikrein originates from the amino-terminal end of the zymogen and is composed of 4 tandem repeats that are 90 or 91 amino acid residues in length. These repeat sequences are also homologous to those in human factor XI. The light chain of plasma kallikrein contains the catalytic portion of the enzyme and is homologous to the trypsin family of serine proteases.     

Introduction to the blood coagulation cascade and cloning of blood coagulation factors

The coagulation cascade that occurs in mammalian plasma involves a large number of plasma proteins that participate in a stepwise manner and eventually give rise to the formation of thrombin. This enzyme then converts fibrinogen to an insoluble fibrin clot. This series of reactions involves a number of glycoproteins that particupate as enzymes as well as cofactors. These proteins that circulate in the blood in a precursor or zymogen form are multifunctional proteins that share many common segments or domains. One group includes the vitamin K-dependent glycoproteins (prothrombin, factor IX, factor X, and protein C) that show considerable homology in both their amino acid sequences and their gene structures. The proteins that participate in the contact or early phase of the blood coagulation cascade include plasma prekallikrein, factor XII, and factor IX. The amino-terminal regions of both factor XI and plasma prekallikrein contain four tandem repeats of about 90 amino acids, and these tandem repeats show considerable amino acid sequence homology. Factor XII contains four different domains in the amino-terminai region of the protein, including a kringle structure, two growth factor domains, and type I and type II finger domains. The finger domains were first identified in fibronectin. The carboxyl-terminal portion of plasma prekallikrein, factor XII, and factor XI contains the serine or protease portion of the molecule. These various plasma proteins that share common domains appear to have evolved by gene shuffling that may have, in some cases, involved introns.     

Disulfide formation and stability of a cysteine-rich repeat protein from Helicobacter pylori

Helicobacter pylori cysteine-rich proteins (Hcps) are disulfide-containing repeat proteins. The repeating unit is a 36-residue, disulfide-bridged, helix-loop-helix motif. We use the protein HcpB, which has four repeats and four disulfide bridges arrayed in tandem, as a model to determine the thermodynamic stability of a disulfide-rich repeat protein and to study the formation and the contribution to stability of the disulfide bonds. When the disulfide bonds are intact, the chemical unfolding of HcpB at pH 5 is cooperative and can be described by a two-state reaction. Thermal unfolding is reversible between pH 2 and 5 and irreversible at higher pH 5. Differential scanning calorimetry shows noncooperative structural changes preceding the main thermal unfolding transition. Unfolding of the oxidized protein is not an all-or-none two-state process, and the disulfide bonds prevent complete unfolding of the polypeptide chain. The reduced protein is significantly less stable and does not unfold in a cooperative way. During oxidative refolding of the fully reduced protein, all the possible disulfide intermediates with a correct disulfide bond are formed. Formation of "wrong" (non-native) disulfide bonds could not be demonstrated, indicating that the reduced protein already has some partial repeating structure. There is a major folding intermediate with disulfides in the second, third, and fourth repeat and reduced cysteines in the first repeat. Disulfide formation in the first repeat limits the overall rate of oxidative refolding and contributes about half of the thermodynamic stability to native HcpB, estimated as 27 kJ mol(-1) at 25 degrees C and pH 7. The high contribution to stability of the first repeat may be explained by the repeat acting as a cap to protect the hydrophobic interior of the molecule.     

Apple four in human blood coagulation factor XI mediates dimer formation.

Human blood coagulation factor XI is a dimer composed of two identical subunits. Each subunit contains four apple domains as tandem repeats followed by a serine protease region. A disulfide bridge between Cys321 of each fourth apple domain links the subunits together. The role of Cys321 in the dimerization of factor XI was examined by mutagenesis followed by expression of its cDNA in baby hamster kidney cells. The recombinant proteins were then purified from the tissue culture medium and shown to have full biological activity. Normal recombinant factor XI was secreted as a dimer as determined by SDS-PAGE, while recombinant factor XI-Cys321 Ser migrated as a monomer under these conditions. Gel filtration studies, however, revealed that each protein existed as a dimer under native conditions, indicating that the disulfide bond between Cys321 of each factor XI monomer was not necessary for dimer formation. The fourth apple domain (apple4) of factor XI was then introduced into tissue plasminogen activator (tPA) to investigate its role in the dimerization of other polypeptide chains. The fusion protein, containing apple4 (apple4-tPA), formed dimers as detected by SDS-PAGE and gel filtration. Furthermore, dimerization was specific to apple4, while apple3 had no effect on dimerization. These data further indicated that the apple4 domain of factor XI mediates dimerization of the two subunits and the interchain disulfide bond involving Cys321 was not essential for dimer formation.     

Mapping of the discontinuous H-kininogen binding site of plasma prekallikrein. Evidence for a critical role of apple domain-2.

Plasma prekallikrein, a zymogen of the contact phase system, circulates in plasma as heterodimeric complex with H-kininogen. The binding is mediated by the prekallikrein heavy chain consisting of four apple domains, A1 to A4, to which H-kininogen binds with high specificity and affinity (K(D) = 1.2 x 10(-8) M). Previous work had demonstrated that a discontinuous kininogen-binding site is formed by a proximal part located in A1, a distal part exposed by A4, and other yet unidentified portion(s) of the kallikrein heavy chain. To detect relevant binding segment(s) we recombinantly expressed single apple domains and found a rank order of binding affinity for kininogen of A2 > A4 approximately A1 > A3. Removal of single apple domains in prekallikrein deletion mutants reduced kininogen binding by 21 (A1), 64 (A2), and 24% (A4), respectively, whereas deletion of A3 was without effect. Transposition of homologous A2 domain from prekallikrein to factor XI conferred high-affinity kininogen binding from the former to the latter. The principal role of A2 for H-kininogen docking to the prekallikrein heavy chain was further substantiated by the finding that cleavage of a single peptide bond in A2 drastically diminished the H-kininogen binding affinity. Furthermore, the epitope of monoclonal antibody PKH6 which blocks kallikrein-kininogen complex formation with an IC(50) of 8 nM mapped to the center portion of domain A2. Our data indicate that domain A2 and two flanking sequence segments of A1 and A4 form a discontinuous binding platform for H-kininogen on the prekallikrein heavy chain. Domain-specific antibodies directed to these critical sites efficiently interfered with contact phase-induced bradykinin release from H-kininogen.     

The acid-stable proteinase inhibitor of human mucous secretions (HUSI-I, antileukoprotease). Complete amino acid sequence as revealed by protein and cDNA sequencing and structural homology to whey proteins and Red Sea turtle proteinase inhibitor

The complete amino acid sequence of human antileukoprotease has been determined by direct sequencing of the inhibitory active protein isolated from seminal plasma (HUSI-I) and by sequence analysis of cDNA reverse-transcribed from mRNA prepared from cervical tissue. The inhibitor (Mr 11726) consists of 107 amino acid residues including 16 cysteines presumably forming disulfide bonds. The molecule comprises two consecutive domains which are homologous to each other, to the second domain of the basic protease inhibitor from Red Sea turtle (chelonianin) and to both domains of the whey proteins of rat and mouse. Both domains contain a pattern of cysteines known as the 'four-disulfide-core' that has also been found in wheat germ agglutinin and neurophysin.     

The structure of the murine Fc receptor for IgG. Assignment of intrachain disulfide bonds, identification of N-linked glycosylation sites, and evidence for a fourth form of Fc receptor

The Fc receptor (Fc gamma R) of the murine macrophage cell line, J774, was purified by immunoaffinity chromatography then subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and amino-terminal sequencing. FcR material judged to be pure by these criteria was digested with a number of enzymes to identify the cysteine residues engaged in disulfide bonds within the native structure. The results clearly establish that the mouse macrophage Fc gamma R contains two intrachain disulfide bonds, each of which connects adjacent cysteine residues within the two putative extracellular domains of the molecule. In addition, each disulfide-bonded domain was shown to contain two authentic sites of N-linked glycosylation. Extensive peptide sequencing resulted in the unexpected identification of peptide fragments from a fourth Fc gamma R whose sequences were highly homologous to sequences surrounding the two Cys residues in the amino-terminal domain of both alpha and beta 1 Fc gamma R. The fourth Fc gamma R contains a disulfide-bonded amino-terminal domain similar to beta 1 Fc gamma R.     

Characterization of the H-kininogen-binding site on factor XI: a comparison of factor XI and plasma prekallikrein.

Factor XI (FXI), the zymogen of the blood coagulation protease FXIa, and the structurally homologous protein plasma prekallikrein circulate in plasma in noncovalent complexes with H-kininogen (HK). HK binds to the heavy chains of FXI and of prekallikrein. Each chain contains four apple domains (F1-F4 for FXI and P1-P4 for prekallikrein). Previous studies indicated that the HK-binding site on FXI is located in F1, whereas the major HK-binding site on prekallikrein is in P2. To determine the contribution of each FXI apple domain to HK-FXI complex formation, we examined binding of recombinant single apple domain-tissue plasminogen activator fusion proteins to HK. The order of affinity from highest to lowest is F2 F4 > F1 F3. Monoclonal antibodies against F2 are superior to F4 or F1 antibodies as inhibitors of HK binding to FXI. Antibody alphaP2, raised against prekallikrein, cross-reacts with FXI F2 and inhibits FXI-HK binding with an IC(50) of 8 nm. HK binding to a platelet-specific FXI variant lacking the N-terminal half of F2 is reduced > 5-fold compared with full-length FXI. A chimeric FXI molecule in which F2 is replaced by P2 is cleaved within P2 during activation by factor XIIa, resulting in greatly reduced HK binding capacity. In contrast, wild-type FXI is not cleaved within F2, and its binding capacity for HK is unaffected by factor XIIa. Our data show that HK binding to FXI involves multiple apple domains, with F2 being most important. The findings demonstrate a similarity in mechanism for FXI and prekallikrein binding to HK.     

Complete determination of disulfide bonds localized within the short consensus repeat units of decay accelerating factor (CD55 antigen).

Decay accelerating factor (DAF) has 4 SCR (short consensus repeat) units. Each SCR unit consists of approx. 60 amino acids characterized by having four conserved cysteine residues and several other highly conserved residues which include proline, tryptophan, tyrosine/phenylalanine and glycine. To determine the disulfide-bonding pattern, we used the urine form of DAF. After thermolysin and trypsin digestion, we isolated seven disulfide-linked peptides by HPLC purification. Because all of the cysteine residues are disulfide-bonded, DAF should contain eight disulfide bonds. After subtilisin and trypsin digestion, we isolated the eighth disulfide-bonded peptides by HPLC purification. From sequence analyses of these peptides, we could identify all disulfide bonds in the 4 SCR units of DAF as being between the first and the third and between the second and the fourth half-cystines within each SCR unit.     

Disulfide bond formation during the folding of influenza virus hemagglutinin

To study the importance of individual sulfhydryl residues during the folding and assembly in vivo of influenza virus hemagglutinin (HA), we have constructed and expressed a series of mutant HA proteins in which cysteines involved in three disulfide bonds have been substituted by serine residues. Investigations of the structure and intracellular transport of the mutant proteins indicate that (a) cysteine residues in the ectodomain are essential both for efficient folding of HA and for stabilization of the folded molecule; (b) cysteine residues in the globular portion of the ectodomain are likely to form native disulfide bonds rapidly and directly, without involvement of intermediate, nonnative linkages; and (c) cysteine residues in the stalk portion of the ectodomain also appear not to form intermediate disulfide bonds, even though they have the opportunity to do so, being separated from their correct partners by hundreds of amino acids including two or more other sulfhydryl residues. We propose a role for the cellular protein BiP in shielding the cysteine residues of the stalk domain during the folding process, thus preventing them from forming intermediate, nonnative disulfide bonds.     

Thrombin-binding affinities of different disulfide-bonded isomers of the fifth EGF-like domain of thrombomodulin.

The fifth EGF-like domain of thrombomodulin (TM), both with and without the amino acids that connect the fifth domain to the sixth domain, has been synthesized and refolded to form several different disulfide-bonded isomers. The domain without the connecting region formed three disulfide-bonded isomers upon refolding under redox conditions. Of these three isomers, the (1-2,3-4,5-6) bonded isomer was the best inhibitor of fibrinogen clotting and also of the thrombin-TM interaction that results in protein C activation, but all the isomers were inhibitors in both assays. The isomer containing an EGF-like disulfide-bonding pattern (1-3,2-4,5-6) was not found among the oxidation products. The domain with the connecting region amino acids (DIDE) at the C-terminus formed two isolable products upon refolding in redox buffer. These products had the same two disulfide-bonding patterns as the earliest and latest eluting isomers of the domain without the DIDE. In order to compare the thrombin-binding affinities of these isomers to the isomer with the EGF-like disulfide bonds, acetamidomethyl protection of the second and fourth cysteines was used to force the disulfide bonds into the EGF-like pattern. Thrombin-binding affinity, measured as inhibition of fibrinogen clotting and as inhibition of protein C activation correlated inversely with the number of crossed disulfide bonds. As was found for the domain without the connecting region, the isomer that was the best inhibitor of fibrinogen clotting and of protein C activation was the isomer with no crossing disulfide bonds (1-2,3-4,5-6).(ABSTRACT TRUNCATED AT 250 WORDS)     

The position of the disulfide bonds in human plasma alpha 2 HS-glycoprotein and the repeating double disulfide bonds in the domain structure

The positions of the inter- and intra-chain disulfide bonds of human plasma alpha 2 HS-glycoprotein were determined. alpha 2 HS-glycoprotein was digested with acid proteinase and then with thermolysin. The disulfide bonds containing peptides were separated by reversed-phase HPLC and detected by SBD-F (7-fluorobenzo-2-oxa-1,3-diasole-4-sulfonic acid ammonium salt) method. One inter-disulfide bond containing peptide and five intra-disulfide bond containing peptides (A-chain) were purified and identified as Cys-18 (B-chain)--Cys-14 (A-chain), Cys-71--Cys-82, Cys-96--Cys-114, Cys-128--Cys-131, Cys-190--Cys-201 and Cys-212--Cys-229, respectively. The location of the intra-disulfide bonds revealed that the A-chain of alpha 2 HS-glycoprotein is composed of three domains. Two domains were shown to possess intramolecular homology judging from the total chain length of the domains, size of the loops formed by the S--S bonds, the location of two disulfide loops near the C-terminal end of domains A and B, the distance between two S--S bonds of each domain, the amino acid sequence homology between these two domains (22.6%), number of amino acid residues between the second S--S loops and the end of domains A and B, and the positions of the ordered structures.     

Crystal structure of a sweet tasting protein thaumatin I, at 1.65 A resolution.

The crystal structure of thaumatin I, a potently sweet protein isolated from the fruits of the West African shrub, Thaumatococcus danielli Benth, has been refined at a resolution better than 1.65 A using a combination of energy minimization and stereochemically restrained least-squares methods. The final model consists of all 207 amino acids, 28 alternate amino acid conformers and 236 waters, with a crystallographic R-factor of 0.145 for 19,877 reflections having F > 4 sigma F between 10.0 A and 1.65 A (R = 0.167 for all 24,022 reflections). The model has good stereochemistry, with root-mean-square deviations from ideal values for bond and angle distances of 0.014 A and 0.029 A, respectively. The estimated root-mean-square co-ordinate error is 0.15 A. The current model confirms the previously reported 3.1 A C alpha trace in both main chain connectivity and disulfide topology, including two disulfide bonds, that differed from the earlier reported biochemical determination. The structure contains three domains. The core of the molecule consists of an eleven-stranded, flattened beta-sandwich folded into two Greek key motifs. All beta-strands in this sandwich are antiparallel except the parallel N-terminal and the C-terminal strands. The average hydrogen bond length in this sandwich is 2.89 A, with an angle of 155.1 degrees. Two beta-bulges are found in one of the sheets. The second domain consists of two beta-strands forming a beta-ribbon and connected by an omega-loop, and contains a proline residue in cis conformation. This structural motif folds back against the main sandwich to form a smaller sandwich-like structure. The third domain is a disulfide-rich region stretching away from the sandwich portion of the molecule. It contains one alpha-helix and three short helical fragments. Two of the helical segments are connected by an unusually sharp turn, stabilized by a disulfide bridge. One of the three disulfide bonds in this domain takes on two conformations.     

Domain-dependent protein folding is indicated by the intracellular kinetics of disulfide bond formation of human chorionic gonadotropin beta subunit.

We have measured the intracellular rates of formation of the six disulfide bonds in the human chorionic gonadotropin beta subunit (hCG-beta) to determine whether the folding pathway of this molecule can be described by a simple sequential model. If such a model is correct, the formation of disulfide bonds, which is indicative of tertiary structural changes during protein folding, should occur in a discrete order. The individual rates of disulfide bridging were determined by identifying the extent of disulfide bond formation in hCG-beta intermediates purified from choriocarcinoma cells that had been metabolically labeled for 40 to 120 s and chased for 0 to 25 min. The results of these kinetic studies describe a folding pathway in which the disulfide bonds between cysteines 34-88, 38-57, 9-90 and 23-72 stabilize, in a discrete order, the putative domain(s) involving amino acids 1-90 of hCG-beta. However, the S-S bonds 93-100 and 26-110 begin to form before the complete formation of the disulfide bonds that stabilize the amino acid 1-90 domain(s), and continue to form after complete formation of these disulfide bonds, suggesting that hCG-beta does not fold by a simple sequential pathway. The order of completion of each of the six disulfide bonds of hCG-beta is: 34-88 (t1/2 = 1-2 min), 38-57 (t1/2 = 2-3 min), 9-90 and 23-72, 93-100, and 26-110. Moreover, 60-100% of each of the six disulfide bonds form posttranslationally, and nonnative disulfide bonds do not form in detectable amounts during intracellular folding of hCG-beta.     

Structure, dynamics and domain organization of the repeat protein Cin1 from the apple scab fungus

Venturia inaequalis is a hemi-biotrophic fungus that causes scab disease of apple. A recently-identified gene from this fungus, cin1 (cellophane-induced 1), is up-regulated over 1000-fold in planta and considerably on cellophane membranes, and encodes a cysteine-rich secreted protein of 523 residues with eight imperfect tandem repeats of ~60 amino acids. The Cin1 sequence has no homology to known proteins and appears to be genus-specific; however, Cin1 repeats and other repeat domains may be structurally similar. An NMR-derived structure of the first two repeat domains of Cin1 (Cin1-D1D2) and a low-resolution model of the full-length protein (Cin1-FL) using SAXS data were determined. The structure of Cin1-D1D2 reveals that each domain comprises a core helix-loop-helix (HLH) motif as part of a three-helix bundle, and is stabilized by two intra-domain disulfide bonds. Cin1-D1D2 adopts a unique protein fold as DALI and PDBeFOLD analysis identified no structural homology. A (15)N backbone NMR dynamic analysis of Cin1-D1D2 showed that a short stretch of the inter-domain linker has large amplitude motions that give rise to reciprocal domain-domain mobility. This observation was supported by SAXS data modeling, where the scattering length density envelope remains thick at the domain-domain boundary, indicative of inter-domain dynamics. Cin1-FL SAXS data models a loosely-packed arrangement of domains, rather than the canonical parallel packing of adjacent HLH repeats observed in α-solenoid repeat proteins. Together, these data suggest that the repeat domains of Cin1 display a "beads-on-a-string" organization with inherent inter-domain flexibility that is likely to facilitate interactions with target ligands.     

Complete primary structure of bovine beta 2-glycoprotein I: localization of the disulfide bridges.

The complete primary structure of bovine beta 2-glycoprotein I was determined by a combination of cDNA and peptide sequencing. Bovine beta 2-glycoprotein I was purified from citrated plasma, and by sequencing selected peptides, the complete disulfide bridge patterns of the 11 disulfide bridges were established as well as the positions of the five asparagine-linked carbohydrate groups. Bovine beta 2-glycoprotein I comprises five mutually homologous domains or Short Consensus Repeats, each containing two disulfide bridges, except for the fifth most C-terminal domain which diverges from the Short Consensus Repeat consensus by containing an additional disulfide bridge. In the four N-terminal domains, the first and third and the second and fourth half-cystines are disulfide-linked, while in the fifth domain the first and fourth, the second and fifth, and the third and sixth half-cystines are disulfide-linked.     

Characterization of the amino-terminal segment in type III procollagen.

Native type III collagen and procollagen were prepared from fetal bovine skin. Examination of the cleavage products produced by digestion with tadpole collagenase demonstrated that the three palpha1(III) chains of type III procollagen were linked together by disulfide bonds occurring at both the amino-terminal and carboxy-terminal portions of the molecule. Type III collagen contained interchain disulfide bonds only in the carboxy-terminal region of the molecule. After digestion of procollagen with bacterial collagenase an amino-terminal, triple-stranded peptide fragment was isolated. The reduced and alkylated chain constituents of this fragment had molecular weights of about 21 000. After digestion of procollagen with cyanogen bromide a related triple-stranded fragment was isolated. The chains of the cyanogen bromide fragment had a molecular weight of about 27 000. When the collagenase-derived peptide was fully reduced and alkylated, it became susceptible to further digestion with bacterial collagenase. This treatment released a fragment of about 97 amino acid residues which contained 12 cystein residues and had an amino acid composition typical for globular proteins. A second, non-helical fragment of about 48 amino acid residues contained three cysteines. This latter fragment is formed from sequences that overlap the amino-terminal region in the collagen alpha1(III) chain by 20 amino acids and possesses an antigenic determinant specific for the alpha1(III) chain. The collagenase-sensitive region exposed by reduction comprised about 33 amino acid residues. It was recovered as a mixture of small peptides. These results indicate that the amino-terminal region of type III procollagen has the same type of structure as the homologous region of type I procollagen. It consists of a globular, a collagen-like and a non-helical domain. Interchain disulfide bonding and the occurrence of cysteines in the non-helical domain are, however, unique for type III procollagen.