Human complement component C1s. Partial sequence determination of the heavy chain and identification of the peptide bond cleaved during activation

Proteogenomic analysis of human cerebrospinal fluid identifies neurologically relevant regulation and informs causal proteins for Alzheimer's disease

The integration of quantitative trait loci (QTL) with disease genome-wide association studies (GWAS) has proven successful at prioritizing candidate genes at disease-associated loci. QTL mapping has mainly been focused on multi-tissue expression QTL or plasma protein QTL (pQTL). Here we generated the largest-to-date cerebrospinal fluid (CSF) pQTL atlas by analyzing 7,028 proteins in 3,107 samples. We identified 3,373 independent study-wide associations for 1,961 proteins, including 2,448 novel pQTLs of which 1,585 are unique to CSF, demonstrating unique genetic regulation of the CSF proteome. In addition to the established chr6p22.2-21.32 HLA region, we identified pleiotropic regions on chr3q28 near OSTN and chr19q13.32 near APOE that were enriched for neuron-specificity and neurological development. We also integrated this pQTL atlas with the latest Alzheimer's disease (AD) GWAS through PWAS, colocalization and Mendelian Randomization and identified 42 putative causal proteins for AD, 15 of which have drugs available. Finally, we developed a proteomics-based risk score for AD that outperforms genetics-based polygenic risk scores. These findings will be instrumental to further understand the biology and identify causal and druggable proteins for brain and neurological traits.

Protease propeptide structures, mechanisms of activation, and functions

Proteases are a diverse group of hydrolytic enzymes, ranging from single-domain catalytic molecules to sophisticated multi-functional macromolecules. Human proteases are divided into five mechanistic classes: aspartate, cysteine, metallo, serine and threonine proteases, based on the catalytic mechanism of hydrolysis. As a protective mechanism against uncontrolled proteolysis, proteases are often produced and secreted as inactive precursors, called zymogens, containing inhibitory N-terminal propeptides. Protease propeptide structures vary considerably in length, ranging from dipeptides and propeptides of about 10 amino acids to complex multifunctional prodomains with hundreds of residues. Interestingly, sequence analysis of the different protease domains has demonstrated that propeptide sequences present higher heterogeneity compared with their catalytic domains. Therefore, we suggest that protease inhibition targeting propeptides might be more specific and have less off-target effects than classical inhibitors. The roles of propeptides, besides keeping protease latency, include correct folding of proteases, compartmentalization, liganding, and functional modulation. Changes in the propeptide sequence, thus, have a tremendous impact on the cognate enzymes. Small modifications of the propeptide sequences modulate the activity of the enzymes, which may be useful as a therapeutic strategy. This review provides an overview of known human proteases, with a focus on the role of their propeptides. We review propeptide functions, activation mechanisms, and possible therapeutic applications.

A functional model of the human C1 complex Emergence of a functional model

Precise structural data on C1s-C1r-C1r-C1s, the catalytic subunit of C1 (the first component of the classical pathway of human complement), led to the emergence of a structural and functional model of this complex protease. Now with new structural information on the amino acid sequence of the protease responsible for C1 activation (C1r), Gérard Arlaud and his colleagues propose a refinement of their original C1 model, and an overall scheme of the intramolecular events associated with the activation and control of C1.

A novel human complement-related protein, C1r-like protease (C1r-LP), specifically cleaves pro-C1s

The availability of the human genome sequence allowed us to identify a human complement-related, C1r-like protease gene (c1r-LP) located 2 kb centromeric of the C1r gene (c1r). Compared with c1r, c1r-LP carries a large deletion corresponding to exons 4-8 of c1r. The open reading frame of the C1r-LP cDNA predicts a 50 kDa modular protein displaying 52% amino acid residue identity with the corresponding regions of C1r and 75% identity with a previously described murine C1r-LP. The serine protease domain of C1r-LP, despite an overall similarity with the AGY group of complement serine proteases, has certain structural features characteristic of C2 and factor B, thus raising interesting evolutionary questions. Northern blotting demonstrated the expression of C1r-LP mRNA mainly in the liver and ELISA demonstrated the presence of the protein in human serum at a concentration of 5.5+/-0.9 microg/ml. Immunoprecipitation experiments failed to demonstrate an association of C1r-LP with the C1 complex in serum. Recombinant C1r-LP exhibits esterolytic activity against peptide thioesters with arginine at the P1 position, but its catalytic efficiency (kcat/K(m)) is lower than that of C1r and C1s. The enzymic activity of C1r-LP is inhibited by di-isopropyl fluorophosphate and also by C1 inhibitor, which forms stable complexes with the protease. Most importantly, C1r-LP also expresses proteolytic activity, cleaving pro-C1s into two fragments of sizes identical with those of the two chains of active C1s. Thus C1r-LP may provide a novel means for the formation of the classical pathway C3/C5 convertase.

Control of insulin-like growth factor binding protein-5 protease synthesis and secretion by human fibroblasts and porcine aortic smooth muscle cells

IGF binding protein-5 (IGFBP-5) is an important trophic factor for controlling the actions of IGF-I in human dermal fibroblasts and porcine aortic smooth muscle cells. When IGFBP-5 is associated with extracellular matrix, it acts to enhance the cell growth response to IGF-I. The amount of IGFBP-5 within the extracellular matrix is related in part to the amount that is present in conditioned medium, which is related to its rate of synthesis and degradation. A serine protease that degrades IGFBP-5 is present in the conditioned medium of both of these cell types. Because the IGFBP-5 protease activity that is secreted by fibroblasts has been shown to be due to the complement components C1r and C1s, these studies were undertaken to determine whether smooth muscle cells also secreted these proteases and to identify some of the factors that regulate their secretion by both cell types. Both smooth muscle cells and human fibroblasts were shown to release C1r and C1s into conditioned medium. Both C1r and C1s were detected as activated forms, as determined by SDS-PAGE using reducing conditions. The addition of increasing concentrations of either IL-1beta or TNFalpha resulted in increased synthesis of C1s by fibroblasts and smooth muscle cells, and they each increased C1r release. TNFalpha (50 ng/ml) and IL-1beta (20 ng/ml) resulted in maximum stimulation of release of both proteases. In contrast dexamethasone (10(-7) M) had no effect on C1s release and stimulated C1r release only by smooth muscle cells. To determine the physiological significance of this increase in C1r and C1s, the amount of IGFBP-5 protease activity that was present in conditioned medium was determined before and after exposure to TNFalpha, IL-beta, and dexamethasone. All three compounds resulted in an increase in the amount of IGFBP-5 proteolytic activity. Dexamethasone inhibited the release of C(1) inhibitor from fibroblasts, and this contributed to the net increase in proteolytic activity. TNFalpha inhibited the smooth muscle cell DNA synthesis response to IGF-I, but the effect of IGF-I was partially restored by the addition of C1 inhibitor. In conclusion, both C1r and C1s are released by cultured fibroblasts, and the release of each into fibroblast or porcine smooth muscle cells medium is stimulated by TNFalpha and IL-1beta. This increase results in a net increase in IGFBP-5 proteolysis, which has the potential to modify IGF-I and IGFBP-5 actions.

A novel murine complement-related gene encoding a C1r-like serum protein

C1r and C1s are highly specific serine proteases that initiate the classical pathway of complement activation. We recently demonstrated that, in the mouse, the genes encoding these proteins are duplicated. Analysis of the 5'-flanking region of the murine C1rA gene, the homologue of human C1r, revealed the presence of a novel gene encoding a C1r-like protein (c1r-LP). Although this gene carries a large deletion, it shows an overall structure similar to that of c1rA, suggesting that it may have arisen from a duplication of the C1r gene. The c1r-LP gene is expressed primarily in the liver, and is not regulated by lipopolysaccharide. The open reading frame of full-length cDNA clones encodes a pre-protein with a calculated molecular mass of 50.6 kDa which, except for an internal deletion of several modules, has a modular organization similar to that of C1r and shows 51% overall amino acid identity to corresponding regions of C1rA. Western blot analysis demonstrates the presence of C1r-LP in mouse serum. The serine protease domain of C1r-LP displays 60% amino acid residue identity to that of C1rA, however, certain atypical features of the active center, and primarily the absence of the activation/cleavage site, suggest that C1r-LP is either an atypical enzyme, or it lacks proteolytic activity, perhaps serving a regulatory function in the classical pathway.

One Active C1r Subunit Is Sufficient for the Activity of the Complement C1 Complex: Stabilization of C1r in the Zymogen Form by Point Mutations

The binding of C1 (the first component of complement) to immune complexes leads to the autoactivation of C1r through the cleavage of the Arg463-Ile464 bond in the catalytic domain. Spontaneous activation of C1r (and C1) also occurs in the fluid phase, preventing the characterization of the zymogen form of C1r. To overcome this difficulty, the zymogen form of human C1r was stabilized by mutating the Arg in the Arg463-Ile464 bond to Gln. This mutant was designated as mutant QI. Recombinant C1r (wild type (wt) or mutant) was expressed in insect cells using serum-free medium in functionally pure form; therefore, the cell culture supernatant was suitable to reconstruct C1 for the hemolytic assay. Mutant QI was a stable, nonactivable zymogen and showed no hemolytic activity in reconstituted C1. However, this stable zymogen C1r mutant could form an active mixed dimer with the wt C1r, indicating that one active C1r subunit in the C1 complex is sufficient for the full activity of the entire complex. Our experiments also showed that the exchange of C1r monomers between the C1r dimers is completed in less than 16 h even at pH 7 and 4°C. Two other mutants were also constructed by changing Arg463 to Lys, or Ile464 to Phe, and were designated as mutants KI and RF, respectively. Although these substitutions did increase the stability of the proenzyme in the cell culture supernatant, the mutant proteins retained their ability to autoactivate, and both had a wt-like hemolytic activity.

Baculovirus-mediated expression of truncated modular fragments from the catalytic region of human complement serine protease C1s. Evidence for the involvement of both complement control protein modules in the recognition of the C4 protein substrate

C1s is the modular serine protease responsible for cleavage of C4 and C2, the protein substrates of the first component of complement. Its catalytic region (gamma-B) comprises two complement control protein (CCP) modules, a short activation peptide (ap), and a serine protease domain (SP). A baculovirus-mediated expression system was used to produce recombinant truncated fragments from this region, deleted either from the first CCP module (CCP2-ap-SP) or from both CCP modules (ap-SP). The aglycosylated fragment CCP2-ap-SPag was also expressed by using tunicamycin. The fragments were produced at yields of 0.6-3 mg/liter of culture, isolated, and characterized chemically and then tested functionally by comparison with intact C1s and its proteolytic gamma-B fragment. All recombinant fragments were expressed in a proenzyme form and cleaved by C1r to generate active enzymes expressing esterolytic activity and reactivity toward C1 inhibitor comparable to those of intact C1s. Likewise, the activated fragments gamma-B, CCP2-ap-SP, and ap-SP retained C1s ability to cleave C2 in the fluid phase. In contrast, whereas fragment gamma-B cleaved C4 as efficiently as C1s, the C4-cleaving activity of CCP2-ap-SP was greatly reduced (about 70-fold) and that of ap-SP was abolished. It is concluded that C4 cleavage involves substrate recognition sites located in both CCP modules of C1s, whereas C2 cleavage is affected mainly by the serine protease domain. Evidence is also provided that the carbohydrate moiety linked to the second CCP module of C1s has no significant effect on catalytic activity.

Proteinase-activated receptor-2: expression by human neutrophils

Neutrophils were shown to express the proteinase-activated receptor-2 (PAR-2), a seven transmembrane domain receptor, which is activated by cleavage by trypsin. Granulocytes from 14 donors stained positively for PAR-2 with affinity-purified rabbit antibodies raised against a peptide corresponding to the trypsin cleavage site of human PAR-2. Neutrophil activation in response to a receptor activating peptide (RAP) varied between donors. RAP (Ser-Leu-Ile-Gly-Lys-Val-NH2) alone induced an increase in the forward and side light scatter after 5–10 minutes and a small increase in the expression of the activation molecule CD11b. The increased expression of CD11b induced by RAP was markedly enhanced by priming the neutrophils with a low concentration (1 nM) of formyl-Leu-Met-Phe. Trypsin and RAP also induced an increase in intracellular calcium, but there were large variations in the magnitude of responses between donors also in this assay. The effects of RAP in the different assays were specific; acetylated RAP was completely without activity.

Analysis of the N-linked oligosaccharides of human C1s using electrospray ionisation mass spectrometry

Information on the structures of the oligosaccharides linked to Asn residues 159 and 391 of the human complement protease C1s was obtained using mass spectrometric and monosaccharide analyses. Asn159 is linked to a complex-type biantennary, bisialylated oligosaccharide NeuAc2 Gal2 GlcNAc4 Man3 (molecular mass = 2206 +/- 1). Asn391 is occupied by either a biantennary, bisialylated oligosaccharide, or a triantennary, trisialylated species NeuAc3 Gal3 GlcNAc5 Man3 (molecular mass = 2861 +/- 1), or a fucosylated triatennary, trisialylated species NeuAc3 Gal3 GlcNAc5 Man3 Fuc1 (molecular mass = 3007 +/- 1), in relative proportions of approximately 1:1:1. The carbohydrate heterogeneity at Asn391 gives rise to three major types of C1s molecules of molecular masses 79,318 +/- 8 (A), 79,971 +/- 8 (B), and 80,131 +/- 8 (C), with an average mass of 79,807 +/- 8. A minor modification, yielding an extra mass of 132 +/- 2, is also detected within positions 1-153.

Neutron scattering study of the (gamma-B) catalytic domains of complement proteases activated C1r and C1s

The catalytic domains of activated C1r and C1s, comprising the C-terminal region of the A chain (gamma), disulphide-linked to the B chain, were obtained by limited proteolysis of the native proteases with chymotrypsin and plasmin, respectively, and studied by small angle neutron scattering. For activated C1s (gamma-B), a molar mass of 45,000 +/- 5000 g/mol, and a relatively large radius of gyration (Rg) of 28 +/- 1 A were determined, excluding a single globular domain. The corresponding values for activated C1r (gamma-B)2 (90,000 g/mol, Rg = 34 +/- 1 A) are consistent with a dimer involving the loose packing of two (gamma-B) subunits. Various models of the dimer are discussed in the light of neutron scattering and other data.

Chemical and functional characterization of a fragment of C1-s containing the epidermal growth factor homology region

C1-s, one of the three subcomponents of C1-, the first component of complement, is a serine protease comprising two disulfide-linked chains, the B chain, containing the catalytic site, and the A chain, involved in Ca2+ binding and Ca2(+)-dependent interaction(s) with the other C1- subcomponents. In an attempt to identify the regions responsible for the latter functions, C1-s was submitted to limited proteolysis with plasmin, a treatment that split the A chain into three major fragments, alpha 1, alpha 2, and gamma. Fragment alpha 2, which comprised the epidermal growth factor-like (EGF-like) region of C1-s, was heterogeneous, starting at serine 97 or phenylalanine 105 and ending at lysine 195. This fragment was reduced and alkylated and then digested with elastase, and three peptides covering positions 131-135, 131-139, and 131-140 were characterized by amino acid analysis, Edman degradation, and mass spectrometry, showing that position 134 of C1-s is occupied partly by an asparagine (47%) and partly by an erythro-beta-hydroxyasparagine, in contrast with the homologous position (150) of C1-r which only contains erythro-beta-hydroxyasparagine. As measured by equilibrium dialysis, native alpha 2, like the other plasmin-cleavage fragments, did not retain the ability of intact C1-s to bind Ca2+. In the same way, plasmin cleavage abolished the ability of C1-s to dimerize or to associate with C1-r in the presence of Ca2+. In contrast, both alpha 2 and the N-terminal alpha 1 fragment, starting at serine 24 of the A chain, were able to compete significantly with intact C1s for the formation of the Ca2(+)-dependent C1-s-C1r-C1-r-C1-s tetramer.(ABSTRACT TRUNCATED AT 250 WORDS)

Human genes for complement components C1r and C1s in a close tail-to-tail arrangement

Complementary DNA clones for human C1s were isolated from cDNA libraries that were prepared with poly(A)+ RNAs of human liver and HepG2 cells. A clone with the largest cDNA insert of 2664 base pairs (bp) was analyzed for its complete nucleotide sequence. It contained 202 bp of a 5' untranslated region, 45 bp of coding for a signal peptide (15 amino acid residues), 2019 bp for complement component C1s zymogen (673 amino acid residues), 378 bp for a 3' untranslated region, a stop codon, and 17 bp of a poly(A) tail. The amino acid sequence of C1s was 40.5% identical to that of C1r, with excellent matches of tentative disulfide bond locations conserving the overall domain structure of C1r. DNA blotting and sequencing analyses of genomic DNA and of an isolated genomic DNA clone clearly showed that the human genes for C1r and C1s are closely located in a "tail-to-tail" arrangement at a distance of about 9.5 kilobases. Furthermore, RNA blot analyses showed that both C1r and C1s genes are primarily expressed in liver, whereas most other tissues expressed both C1r and C1s genes at much lower levels (less than 10% of that in liver). Multiple molecular sizes of specific mRNAs were observed in the RNA blot analyses for both C1r and C1s, indicating that alternative RNA processing(s), likely an alternative polyadenylylation, might take place for both genes.

Characterization of a T-lymphocyte Epstein-Barr virus/C3d receptor (CD21)

The Epstein-Barr virus/C3d receptor (EBVR-CR2) was detected on three T-lymphoblastoid cell lines. The apparent Mrs of purified EBVR-CR2 of T-cell and B-cell origin were identical. The N-terminal amino acid sequence from the T-cell EBVR-CR2 confirmed the placement of this receptor in a multigene family of complement regulatory proteins. All EBVR-CR2-positive T-cell lines were T6 and T4-T8 antigen positive.

Molecular cloning of cDNA for human complement component C1s. The complete amino acid sequence

The complete amino acid sequence (673 residues plus 15 residues of leader sequence) of human complement component C1s has been determined by nucleotide sequencing of cDNA clones from a human liver library probed with synthetic oligonucleotides. Much of the sequence is supported by independent amino acid sequence information. The cDNA sequence contains an anomalous "intron-like" sequence, including a stop codon, that can be discounted because of the amino acid sequence evidence. The N-terminal chain (422 residues) of C1s, like that of C1r with which it is broadly homologous, contains five domains: domains I and III are homologous to one another and to similar regions in C1r, domain II is homologous to the epidermal growth factor sequence found in C1r and several other proteins, and domains IV and V are homologous to one another and to the 60-residue repeating sequence found in C1r, C2, factor B, C4-binding protein and some apparently unrelated proteins. The sequence of the C-terminal chain (251 residues) agrees with that already established to be the "serine protease" domain of C1s.

Occurrence of beta-hydroxylated asparagine residues in non-vitamin K-dependent proteins containing epidermal growth factor-like domains

Vitamin K-dependent bovine protein S has been shown to contain a posttranslationally hydroxylated asparagine within a conserved sequence in three of its epidermal growth factor (EGF)-like domains. In a review of amino acid sequences deduced from cDNA data, we have observed that a conserved sequence containing a potential asparagine hydroxylation site exists within EGF-like domains of a variety of functionally diverse proteins. We have studied a number of these and report the presence of erythro-beta-hydroxyasparagine (e-beta Hyn) in three non-vitamin K-dependent proteins: the plasma complement proteins C1r and C1s (where overbar indicates activated form) and the urinary protein uromodulin. For each protein, e-beta Hyn was identified in enzyme digests following the initial observation of erythro-beta-hydroxyaspartic acid (e-beta Hya) in acid hydrolysates of the proteins. e beta Hya and e-beta Hyn residues are detected by a postcolumn derivatization cation-exchange HPLC method herein described. HPLC isolation of the presumptive e-beta Hyn residue from enzyme digests of intact C1r allowed confirmation of its structure by GC/MS. Based upon available cDNA sequence data and observation of e-beta Hya in acid hydrolysates, we suggest other proteins in which e-beta Hyn may occur.

Lipopolysaccharide-sensitive serine-protease zymogen (factor C) of horseshoe crab hemocytes. Identification and alignment of proteolytic fragments produced during the activation show that it is a novel type of serine protease

The horseshoe crab clotting factor, factor C, present in the hemocytes is a serine-protease zymogen activated with lipopolysaccharide. It is a two-chain glycoprotein (Mr = 123,000) composed of a heavy chain (Mr = 80,000) and a light chain (Mr = 43,000) [T. Nakamura et al. (1986) Eur. J. Biochem. 154, 511-521]. In our continued study of this zymogen, we have now also found a single-chain form of factor C (Mr = 123,000) in the hemocyte lysate. The heavy chain had the NH2-terminal sequence of Ser-Gly-Val-Asp-, consistent with that of the single-chain factor C, indicating that the heavy chain is derived from the NH2-terminal part of the molecule. The light chain had an NH2-terminal sequence of Ser-Ser-Gln-Pro-. Incubation of the two-chain zymogen with lipopolysaccharide resulted in the cleavage of a Phe-Ile bond between residues 72 and 73 of the light chain. Concomitant with this cleavage, the A (72 amino acid residues) and B chains derived from the light chain were formed. The complete amino acid sequence of the A chain was determined by automated Edman degradation. The A chain contained a typical segment which is similar in sequence to a family of repeats in human beta 2-glycoprotein I, complement factors B, protein H, C4b-binding protein, and coagulation factor XIII b subunit. The NH2-terminal sequence of the B chain was Ile-Trp-Asn-Gly-. This chain contained the serine-active site sequence-Asp-Ala-Cys-Ser-Gly-Asp-Ser-Gly-Gly-Pro-. These results indicate that horseshoe crab factor C exists in the hemocytes in a single-chain zymogen form and is converted to an active serine protease by hydrolysis of a specific Phe-Ile peptide bond.

Genetic studies of low-abundance human plasma proteins. VII. Heterogeneity of the C1S subcomponent of the first complement component

Charge-based structural variation has been observed in the C1s subcomponent of the first complement component C1 after isoelectric focusing and immunoblotting. One common and two uncommon autosomal co-dominantly expressed alleles, designated C1S*1, C1S*2 and C1S*3, have been recognized at the C1S structural locus. The frequency of these alleles was 0.979, 0.016 and 0.005, respectively, in a U.S. white population. No variation at the C1S locus was observed in a U.S. black sample (n = 95).

Complete amino acid sequence of the A chain of human complement-classical-pathway enzyme C1r

The amino acid sequence of human C1r A chain was determined, from sequence analysis performed on fragments obtained from C1r autolytic cleavage, cleavage of methionyl bonds, tryptic cleavages at arginine and lysine residues, and cleavages by staphylococcal proteinase. The polypeptide chain has an N-terminal serine residue and contains 446 amino acid residues (Mr 51,200). The sequence data allow chemical characterization of fragments alpha (positions 1-211), beta (positions 212-279) and gamma (positions 280-446) yielded from C1r autolytic cleavage, and identification of the two major cleavage sites generating these fragments. Position 150 of C1r A chain is occupied by a modified amino acid residue that, upon acid hydrolysis, yields erythro-beta-hydroxyaspartic acid, and that is located in a sequence homologous to the beta-hydroxyaspartic acid-containing regions of Factor IX, Factor X, protein C and protein Z. Sequence comparison reveals internal homology between two segments (positions 10-78 and 186-257). Two carbohydrate moieties are attached to the polypeptide chain, both via asparagine residues at positions 108 and 204. Combined with the previously determined sequence of C1r B chain [Arlaud & Gagnon (1983) Biochemistry 22, 1758-1764], these data give the complete sequence of human C1r.

Cloning and sequencing of full-length cDNA encoding the precursor of human complement component C1r

The sequencing of human liver cDNA clones encoding the entire C1r precursor protein has confirmed the previously determined peptide sequence and has shown that there is a leader peptide which is 17 amino acids long. A residue tentatively identified as beta-hydroxyaspartic acid [Arlaud, Willis & Gagnon (1986) Biochem. J., in the press] located in the C1r A-chain, within an epidermal-growth-factor consensus sequence, was found to be encoded as asparagine. Two sequence elements, tandemly located in the A-chain, are related to a sequence widespread among proteins which interact with C3b or C4b. Structural comparisons between different clones indicate that multiple polyadenylation sites are responsible for the length heterogeneity observed for C1r mRNA from liver and Hep G2 cells.