Comparative properties of three functionally different but structurally related serpin variants from horse plasma

An engineered protein-based submicromolar competitive inhibitor of the Staphylococcus aureus virulence factor aureolysin

Aureolysin, a secreted metallopeptidase (MP) from the thermolysin family, functions as a major virulence factor in Staphylococcus aureus. No specific aureolysin inhibitors have yet been described, making this an important target for the development of novel antimicrobial drugs in times of rampant antibiotic resistance. Although small-molecule inhibitors are currently more common in the clinic, therapeutic proteins and peptides (TPs) are favourable due to their high selectivity, which reduces off-target toxicity and allows dosage tuning. The greater wax moth Galleria mellonella produces a unique defensive protein known as the insect metallopeptidase inhibitor (IMPI), which selectively inhibits some thermolysins from pathogenic bacteria. We determined the ability of IMPI to inhibit aureolysin in vitro and used crystal structures to ascertain its mechanism of action. This revealed that IMPI uses the “standard mechanism”, which has been poorly characterised for MPs in general. Accordingly, we designed a cohort of 12 single and multiple point mutants, the best of which (I⁵⁷F) inhibited aureolysin with an estimated inhibition constant (K_i) of 346 nM. Given that animals lack thermolysins, our strategy may facilitate the development of safe TPs against staphylococcal infections, including strains resistant to conventional antibiotics.

Effect of corticotherapy on proteomics of endometrial fluid from mares susceptible to persistent postbreeding endometritis

The objective was to determine the effects of corticotherapy, in the presence and absence of uterine inflammation, on proteomics of endometrial fluid from mares susceptible to endometritis. In 11 mares, estrus was induced seven times with 5 mg PGF(2α) given at 14-day intervals. The first estrus was a control (no treatment). During the third estrus, mares received glucocorticoid (GC) treatment (20 mg isoflupredone acetate) every 12 h, for three consecutive days. The fifth estrus was the Infected treatment (intrauterine infusion of 1 × 10(9) colony-forming unit/mL Streptococcus equi subspecies zooepidemicus). Finally, the seventh was a combination of GC + Infected treatment (infusion of bacteria 24 h after the first GC treatment). At 12 h after the end of each treatment, uterine samples were collected and submitted to two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) for protein separation and mass spectrometry. Both GC treatment and uterine lumen infection induced proteomic alterations in the endometrial fluid of susceptible mares, characterized by an increase, decrease, or both in the relative optic density and/or frequency of inflammatory acute phase proteins (APP), with major alterations occurring when corticotherapy was applied in the presence of an infectious process. Corticotherapy in the presence of infection increased α(1)-antitrypsin (AAT), transthyretin (TT), and actin, but reduced immunoglobulin G, whereas intrauterine infection increased haptoglobin (Hp) and apolipoprotein A-1 (ApoA-1) and decreased transferrin (TF). Infection reduced levels of α(1)-antitrypsin and transthyretin, whereas corticotherapy in the presence of infection increased their frequency. We concluded that GC influenced the immune response, not only as suppressors, but also as enhancers of local defense mechanisms, through an immunomodulatory action. Short-term corticotherapy could be beneficial for treatment of uterine infectious processes in the mare.

Kinetics of equine neutrophil elastase release and superoxide anion generation following secretagogue activation: a potential mechanism for antiproteinase inactivation

Man and horses both suffer from neutrophil mediated pulmonary diseases however there are striking species differences in the underlying pathology. In particular while pulmonary emphysema is a common pathological sequel to human respiratory disease it is not a major feature of the common equine neutrophil mediated condition, chronic obstructive pulmonary disease (COPD). The proposed reason for this difference is that equine neutrophils contain less elastase than equivalent human cells and therefore there is a reduced risk of excess and/or uninhibited elastase activity, which is considered the major cause of pulmonary emphysema in man, in the horse lung. In previous studies equine neutrophil elastase (ENE) has been assayed by measuring elastinolytic activity whereas human neutrophil elastase content has been determined using immunological techniques. Neutrophils contain several intracellular protease inhibitors therefore measurement of elastase activity may underestimate the total NE content. The aim of the current study was to develop immunological techniques to allow investigation of the cellular content, distribution and release of ENE from purified equine neutrophils. Equine neutrophil elastase 2A (ENE 2A), the most abundant elastase in equine neutrophils, and equine alpha-1-proteinase inhibitor (API), the main inhibitor of elastase were found to be present at 0.813 pg +/- 0.179 and 0.021 pg +/- 0.003 (mean +/- SEM, n = 11 individual horses) per neutrophil, respectively. This represents twice as much elastase as previously found in the equine neutrophil and a comparable amount to that reported in human neutrophils. Immunolocalisation demonstrated that ENE 2A has a granular distribution within the cytosol of neutrophils, whereas API exhibits a uniform non-granular cytoplasmic appearance. In addition the kinetics of simultaneous generation and release of superoxide anions (SOA) and release of ENE 2A from equine neutrophils, stimulated in vitro by zymosan-activated serum (ZAS) in the presence and absence of the cation chelator ethylene glycol-N,N,N',N'-tetraacetic acid (EGTA), showed a close relationship between total SOA generation and total ENE 2A release during the initial 90 min post-ZAS stimulation and the dependence of both events on extracellular cations. In conclusion these studies have shown that horse and human neutrophil elastase content and mediator release functions are more closely matched than was previously thought. This suggests that the species differences in pathology resulting from neutrophil-mediated respiratory disease are determined by other factors such as differences in the abundance and function of intra- and extra-cellular protease inhibitors.

Improved hepatic and pancreatic localisation of the equine alpha-1-proteinase inhibitor family of serpins using an antigen enhancement technique and a monoclonal antibody

Equine alpha-1-proteinase inhibitor (API) consists of three, occasionally four, serum glycoproteins. This study investigated the immunohistochemical localisation of equine API in paraformaldehyde fixed, paraffin embedded equine tissue samples of liver, lung, stomach, pancreas, jejunum and colon in five horses using affinity purified sheep polyclonal and protein A purified mouse monoclonal antibodies, whose specificities were verified by Western blotting. Exposing tissue sections to boiling citrate buffer greatly enhanced antigen recovery and improved immunostaining with both antibodies, resulting in discovery of novel tissue distribution patterns for the horse. In the horses studied, all hepatocytes showed some degree of cytoplasmic staining, many having perinuclear intense granular inclusions. This finding is contrary to findings in human studies where hepatocytes of Pi MM phenotype have proven difficult to stain for human API, despite evidence at the molecular level suggesting hepatocytes as the major source of serum API. This discrepancy may be due to the use of different tissue fixation and antigen recovery techniques. In all other tissues examined, the distribution of equine API was similar to human studies.

Identification of lysines within alpha 1-antichymotrypsin important for DNA binding. An unusual combination of DNA-binding elements

The human serum serine protease inhibitor (serpin) alpha 1-antichymotrypsin (ACT) appears to be unique among serpins in its ability to bind to double-stranded DNA. Using site-directed mutagenesis and chemical modification, a tri-lysine sequence (residues 210-212) falling within a solvent exposed loop and the C-terminal peptide containing two lysines (residues 391 and 396) were shown to be important for DNA binding. Mutation of residues 210-212 from lysines to either glutamates or threonines abolished DNA binding. The Lys210-Thr211-212 and Thr210-Th4(211)-Lys212 variants displayed reduced affinity for DNA, especially at higher ionic strength. Limited acetylation of rACT with acetic anhydride led to loss of DNA binding and, conversely, DNA protected rACT from acetylation. A combination of CNBr digestion, peptide separation, and peptide sequencing identified Lys396, two residues from the C terminus, as the most reactive lysine in rACT. Acetylation of Lys396 is strongly decreased in the presence of DNA. The double mutant K391T/K396T-rACT had very little affinity for DNA. The epsilon-amines of lysines 210-212 are 8-15 A across a cleft from the epsilon-amines in Lys391 and Lys396, and together these two elements may form an unusual DNA binding domain. Attempts to isolate a DNA sequence to which ACT binds specifically have been unsuccessful to date, raising the possibility that nonspecific binding of ACT to DNA suffices to account for the ACT found in certain cell nuclei. ACT variants not binding to double-stranded DNA retain ACT protease inhibitory activity, a potentially important result for the use of ACT variants as therapeutic agents.

Pig plasma alpha-protease inhibitors PI2, PI3 and PI4 are members of the antichymotrypsin family

Three related alpha-protease inhibitors, PI2 I, PI3 C and PI4 C2, of blood serum of the pig (Sus scrofa) were isolated. PI2 I inhibited both trypsin and chymotrypsin; PI3 C and PI4 C2 strongly inhibited chymotrypsin, but did not significantly inhibit trypsin. By using SDS-PAGE, the three proteins were found to be composed of single polypeptide chains, and molecular weights were 63,000 for PI2 I, 58,000 for PI3 C and 64,000 for PI4 C2. All three proteins were shown to be glycoproteins. In PI3 C, eight sialic acid residues were found, and in PI4 C2 (similarly as in PI2 F) 10-11 residues were found. Amino acid composition as well as N-terminal sequences of the three proteins were very similar, indicating close homology. Comparison of these partial amino acid sequences with the cDNA-deduced amino acid sequence of pig alpha-antichymotrypsin (AACT; Buchman, 1989, GenBank, Accession No. M29508) revealed great similarities, the sequence of PI2 I being virtually identical with the pig AACT. On the basis of all available results, PI2 is proposed to be pig AACT, an orthologue of human AACT.

Characterization of two serpins from bovine plasma and milk

An inhibitor of pancreatic elastase (EI), which can also inhibit chymotrypsin, and an inhibitor of trypsin (TI), which can also inhibit plasmin, have been isolated from bovine plasma. EI and TI belong to the serpin family of inhibitors. The size of both inhibitors is approx. 60 kDa and they are able to form SDS-stable complexes with proteinases. Curiously, TI dimerizes in the presence of SDS, a feature which has been observed previously only in non-denaturing gels of human alpha 1-antitrypsin (alpha 1PI). EI and TI are glycosylated [16% and 19% (w/w) respectively] and their amino acid compositions are similar to those of other plasma serpins. Neither EI nor TI is the equivalent of bovine alpha 1PI, as revealed by partial sequence analysis of their N-termini and reactive sites. Rather, both inhibitors appear to be related to human alpha 1-antichymotrypsin. Inhibition of pancreatic elastase and chymotrypsin by EI occurs with a kass. approximately 10(5) M-1.s-1. TI inhibits trypsin with a kass. approximately 10(5) M-1.s-1. Plasmin is inhibited by TI with a kass. approximately 10(3) M-1.s-1. The values of the kinetic constants are similar to those determined for the well-studied human serpins. Antibodies to EI and TI reveal a set of four antigenically related proteins of similar size in plasma. In addition, they detect the same set of proteins in milk. The inhibitors isolated from milk are identical to EI and TI from plasma. EI could control the activity of chymotrypsin-like proteinases in milk. In contrast, no target proteinases of TI in milk can be suggested.

Molecular weight alterations of alpha-1 proteinase inhibitor in equine bronchoalveolar lavage fluid

The equine alpha-1 proteinase inhibitor (alpha 1PI) system differs from that of man in that the equine system consists of four closely-linked genes (Spi1-Spi4) whereas in man, a single gene encodes for alpha 1PI. We have previously found differences in the proportion of the Spi proteins in equine serum and bronchoalveolar lavage fluid (BALF). We therefore wished to determine whether, as reported in man, there was any molecular weight difference between the Spi proteins in serum and BALF. alpha 1PI and albumin from equine BALF migrated further towards the anode compared with serum alpha 1PI on native polyacrylamide gel electrophoresis (PAGE) although the difference was only significant for alpha 1PI. Sodium dodecyl sulphate-PAGE (SDS-PAGE) showed that a mean decrease in molecular weight of 1.5 kDa for alpha 1PI and 1.3 kDa for albumin had occurred in BALF. These findings were observed in control animals and in those with symptomatic or asymptomatic chronic obstructive pulmonary disease. The mechanism of this decrease in molecular weight of alpha 1PI is likely to differ from reports of alpha 1PI cleavage by bacterial proteinases in man since the molecular weight change was relatively small and loss of trypsin inhibitory activity did not occur. Nor, in our system, was there evidence of bacterial infection. Damage by endogenous proteinases or glycosidases at a site other than the reactive site may be involved but the resultant effect on the efficiency of the antiproteinase screen of the lower respiratory tract is uncertain.

Measurement by ELISA of equine alpha-1-proteinase inhibitor in uterine flushings from mares

An enzyme linked immunosorbent assay (ELISA) was developed and used to estimate the concentrations of the serine proteinase inhibitor, alpha-1 proteinase inhibitor (API), in uterine flushings recovered from mares at different stages of the oestrous cycle and before and after the induction of experimental endometritis. There was a significant increase in the concentrations of API and albumin relative to total protein in flushings recovered during oestrus compared with dioestrus but no difference was observed in the concentrations of these proteins relative to total protein before and after the induction of endometritis. A regression analysis revealed a significant correlation between the concentrations of albumin and API in the flushings examined, suggesting that the API was derived entirely from serum and was not produced locally in the uterus.

Investigation of association between alpha-1 proteinase inhibitor haplotype and endometritis in the thoroughbred mare

Failure to inhibit proteinases can lead to excessive tissue damage. The possibility that the severity of endometritis in Thoroughbred mares correlates with the haplotypes of plasma alpha 1-proteinase inhibitor (alpha 1-PI) expressed was investigated in two groups of mares. In mares with pyometritis before treatment, the frequency of the N haplotype, which is already high in the Thoroughbred population, was significantly increased when compared with that in a large published population. In mares with acute endometritis which persisted after treatment followed by sexual rest, the absence of S and T haplotypes was significant, suggesting that, when present, they may have a protective function.

Primary structure of a porcine leukocyte serpin

Inhibitors of neutral serine proteinases (serpins) have been shown to be colocalized with their target enzymes in leukocytes of several mammalian species. Here we report the purification and complete primary structure of a cytosolic inhibitor from porcine granulocytes which is directed against neutrophil elastase. Two molecular mass forms of the leukocyte neutral proteinase inhibitor (LNPI) were isolated by affinity and ion-exchange chromatography followed by gel filtration, and identified as the inhibitorily active monomer and homodimer of the inhibitor protein. According to the amino acid sequence the molecular mass of the non-glycosylated inhibitor was calculated to 42,597 Da (378 amino acid residues). A sequence identity of 81% was found between LNPI and the homologous elastase inhibitors from both human and equine leukocytes, whereas only 50% of the positions are identical in LNPI and human plasminogen activator inhibitor 2. These data suggest that LNPI is a member of a new group of cytosolic serpins closely related to the ovalbumin branch of the superfamily.

Comparative studies of the Spi1 proteins of three equine alpha-1-proteinase inhibitor haplotypes following isolation by affinity chromatography

1. Antiproteinase deficiency can result in excessive proteinase-induced tissue damage. The major anti-elastase (Spi1) protein of equine alpha 1-proteinase inhibitor (alpha 1-PI) was isolated from the plasma/serum of three common haplotypes (I, L and U). 2. The N-terminal amino acid sequences of the three inhibitors were identical, but were only approx 65-77% homologous with two other published equine Spi1 sequences. 3. All three inhibitors complexed quickly and irreversibly with equine leucocyte proteinase 2A (kass = 2 x 10(7) M-1 sec-1). They were also efficient inhibitors of chymase (rat mast cell proteinase-II; kass = 2 x 10(5) M-1 sec-1; Ki = 2 x 10(-10) M). There was therefore no evidence of deficient inhibition in the Spi1 variants of the I,L and U haplotypes.

Molecular cloning and expression of an intracellular serpin: an elastase inhibitor from horse leucocytes

Horse blood leucocytes contain an elastase inhibitor (HLEI) belonging to the serpin family. Poly(A)+RNA isolated from these cells was used to construct a cDNA library in lambda gt10, which was first screened with a synthetic degenerate oligonucleotide probe corresponding to the amino acid sequence of the reactive centre of the inhibitor. Three clones were obtained covering the entire coding region of the protein. Sequencing of these clones showed identity with the amino acid sequence obtained from Edman degradation of the elastase inhibitor. The coding sequence of the HLEI cDNA was cloned into the bacterial expression vector pKK233-2 and expressed in Escherichia coli cells. Transformed bacteria expressed significant amounts of the protein, which was immunoprecipitated with a specific anti-HLEI antiserum. Furthermore, HLEI expressed in bacteria inhibited the activity of elastase but not trypsin.

Neutrophil chemotaxis in the horse is not mediated by a complex of equine neutrophil elastase and equine alpha-1-proteinase inhibitor

Studies have demonstrated that as a result of proteolytic inactivation or complex formation (with neutrophil elastase), human alpha-1-proteinase inhibitor (API) becomes a potent chemoattractant for human neutrophils. The present study aimed to investigate the in vitro chemotactic response of equine neutrophils to an equivalent complex of equine API and neutrophil elastase. No evidence of neutrophil migration was observed towards purified complex derived from equine neutrophil elastase and the Spi 1 isoform of equine API, or to crude mixtures of porcine pancreatic elastase and unseparated equine API isoforms, although the same neutrophil preparations actively migrated towards zymosan activated plasma. It was concluded that, in the horse, complexes of API are not involved in the migration of neutrophils to sites of inflammation.

Retardation of serum albumin migration in native polyacrylamide gel electrophoresis by the incorporation of blue dextran in the gel matrix

Blue dextran was incorporated into the stacking layer of one-dimensional polyacrylamide gels to prevent the migration of albumin into the resolving gel during the electrophoresis of native plasma samples. Human and horse plasma proteins normally masked by albumin were revealed by this method.

The equine major plasma serpin multigene family: partial characterization including sequence of the reactive-site regions

The equine Pi system, which is highly polymorphic and was considered to be controlled by a single locus, has been shown to be controlled by four loci (named Spi 1-4). This system is the equine equivalent of the major human plasma serpin (serine protease inhibitor), human alpha 1 PI. Twenty-two haplotypes of the equine Pi system have been characterized by two-dimensional electrophoresis, resulting in the assignment of pI, Mr, and bovine trypsin and chymotrypsin inhibition characteristics to 109 proteins. These proteins have been analyzed further to determine their relatedness to each other as well as to human alpha 1 PI using immunochemical, structural, and functional criteria. The amino acid sequences of the N termini and reactive-site regions have been determined on proteins from each of the four equine Spi loci. This allowed the designation of the proteins from the Spi 1 locus as being METserpins and the functional equivalents of human alpha 1 PI. The Spi 4 proteins are ARGserpins, and by alignment the Spi 2 proteins are ILEserpins, the first so far described. The P1 residue for the Spi 3 proteins was unable to be determined. The limited peptide and immunopeptide mapping revealed that proteins from all four loci were closely related, but within the four there were two pairs (Spi 1 and 2 and Spi 3 and 4) which were more related. All were probably derived from the same gene that gave rise to human alpha 1 PI.