Isolation and characterization of alpha2-plasmin inhibitor from human plasma. A novel proteinase inhibitor which inhibits activator-induced clot lysis.

A procedure is presented for purifying a novel proteinase inhibitor in human plasma whose apparent unique biological property is to inhibit efficiently the lysis of fibrin clots induced by plasminogen activator. The final product is homogeneous as judged by disc gel electrophoresis, and immunoelectrophoresis. Its molecular weight estimated by sodium dodecyl sulfate gel electrophoresis or sedimentation equilibrium is 67,000 and 63,000, respectively. The inhibitor is a glycoprotein consisting polypeptide chain containing 11.7% carbohyrate. It migrates in the alpha2-globulin region in immunoelectrophoresis. The inhibitor is chemically and immunologically different from all the other known inhibitors in plasma. Inhibition of plasmin by the inhibitor is almost instantaneous even at 0 degrees, in contrast to the slow inhibition of urokinase (plasminogen activator in urine). Plasminogen activation by urokinase-induced clot lysis is inhibited by the inhibitor mainly through a mechanism of instantaneous inhibition of plasmin formed and not through the inhibition of urokinase. The inhibitor also inhibits trypsin. Consequently, it is suggested that this newly identified inhibitor is named alpha2-plasmin inhibitor or alpha2-proteinase inhibitor. A specific antibody directed against the inhibitor neutralizes virtually all inhibitory activity of plasma to activator-induced clot lysis. Immunochemical quantitation of the inhibitor was specific antiserum to the inhibitor and the purified inhibitor as a standard indicates that the concentration of the inhibitory in the serum of a healthy man is in or near the range of 5 to 7 mg/100 ml, which is the lowest concentration among the concentration of the proteinase inhibitors in plasma. The inhibitor and plasmin, trypsin, or urokinase form a complex which cannot be dissociated with denaturing and reducing agents. The formation of the enzyme-inhibitor complex occurs on a 1:1 molar basis and is associated with the cleavage of a unique peptide bone, which is most clearly demonstrated in the interaction of the inhibitor and beta-trypsin. In the complex formation between the inhibitor and plasmin, the inhibitor is cross-linked with the light chain which contains the active site of plasmin. It is suggested that, in a fashion analogous to complex formation between alpha1-antitrypsin and trypsin, the cross-links are formed between the active site serine of the enzyme and the newly formed COOH-terminal residue of the inhibitor, with cleavage of a peptide bond.     

Differential inhibition of serine proteinases by rabbit alpha 1-proteinase inhibitors F and S

Inhibition of six serine proteinases (bovine trypsin and chymotrypsin, equine leucocyte proteinases type 1 and 2A, porcine pancreatic elastase type III and rabbit plasmin) by rabbit alpha 1-proteinase inhibitors F and S was studied. In each case examined, the F form reacted more rapidly. The number of moles of an enzyme inhibited by one mole of alpha 1-proteinase inhibitor in a complete reaction (molar inhibitory capacity) ranged from 0.26 (leucocyte proteinase type 1) to 1.01 (trypsin). More significantly, however, the molar inhibitory capacities of both alpha 1-proteinase inhibitors differed for the same enzymes. The highest F/S inhibitory ratio was recorded with chymotrypsin (1.88), and the lowest with elastase (0.69). These differences in molar inhibitory capacities are likely to reflect the dual nature of the reaction between the inhibitor and a proteinase, that is, either complex formation or inactivation of alpha 1-proteinase inhibitor without enzyme inhibition. No evidence was obtained to suggest that differential reactivity and differential inhibitory capacity are interdependent. The observations are consistent with the view that rabbit alpha 1-proteinase inhibitors F and S are closely related yet functionally distinct proteins.     

Inactivation of alpha- and beta-thrombin by antithrombin-III, alpha 2-macroglobulin and alpha 1-proteinase inhibitor.

Inactivation of alpha- and beta-thrombin by alpha 2-macroglobulin, by alpha 1-proteinase inhibitor and by antithrombin-III and heparin was studied. The amount of alpha- and beta-thrombin inactivated by antithrombin-III was proportional to the concentration of the inhibitor, but the inactivation rates of the two forms of thrombin were different. Heparin facilitated complex-formation between alpha-thrombin and antithrombin-III, whereas inactivation of beta-thrombin by antithrombin was only slightly influenced, even at a heparin concentration two orders of magnitude higher. alpha 2-Macroglobulin inhibited both alpha- and beta-thrombin activity similarly, i.e. the amount of alpha- and beta-thrombin inactivated as well as the rates of their inhibition were the same. alpha 1-Proteinase inhibitor also formed a complex with alpha- and beta-thrombin, similarly to antithrombin-III, although the inactivation of the enzyme needed high inhibitor concentration and long incubation time. These results suggest that the inactivation of beta-thrombin, if it occurs in the plasma, is also controlled by plasma inhibitors.     

Effects of human and ovine pancreatic secretory trypsin inhibitors on the proliferation of normal human fibroblasts

Human pancreatic secretory trypsin inhibitor inhibited cell-surface proteolytic activity in human fibroblasts. In the range of concentrations which caused proteinase inhibition, fibroblast proliferation was also inhibited by this reagent and by the ovine equivalent. At lower concentrations, there was some evidence for a mitogenic effect, and this was confirmed by obvious stimulation of DNA synthesis at these concentrations. Human alpha 1-proteinase inhibitor, previously demonstrated to be an inhibitor of fibroblast proliferation, was also mitogenic at concentrations lower than those which inhibited proteolytic activity and cell proliferation. Human pancreatic secretory trypsin inhibitor and epidermal growth factor apparently work through independent mechanisms, since their mitogenic effects are additive.     

The reactive site of human alpha 2-antiplasmin

Human alpha 2-antiplasmin rapidly forms a stable, equimolar complex with either its target enzyme, plasmin, or with trypsin. Perturbation of the inhibitor-trypsin complex results in peptide bond cleavage at the reactive site of the inhibitor with the concomitant release of a small peptide fragment which apparently represents the carboxyl-terminal segment of the inhibitor. Sequence analysis of this fragment, together with that of an overlapping peptide obtained by treatment of native inhibitor with either Staphylococcus aureus V8 proteinase or human neutrophil elastase, yields data which indicate that the reactive site of alpha 2-antiplasmin encompasses a P1-P'1 Arg-Met sequence. However, unlike alpha 1-1-proteinase inhibitor which has a Met residue in the P1-position, oxidation of alpha 2-antiplasmin has no effect on its inhibitory activity toward either plasmin, trypsin, or chymotrypsin, indicating the lesser mechanistic importance of the P'1-residue during enzyme inactivation by this inhibitor.     

Interactions of the complex between human urinary trypsin inhibitor and human leukocyte elastase with alpha 1-proteinase inhibitor and alpha 2-macroglobulin

The urinary trypsin inhibitor was recently shown to inhibit human leukocyte elastase. Complexes of human urinary trypsin inhibitor with human leukocyte elastase or human trypsin were produced and subjected to gel filtration. The complexes were found to be sufficiently stable up to 24 h incubation (at least 70% recovery). When human serum was added, elastase and trypsin dissociated from the urinary trypsin inhibitor and associated with alpha 1-proteinase inhibitor or alpha 2-macroglobulin. The addition of alpha 1-proteinase inhibitor to a complex of urinary trypsin inhibitor and leukocyte elastase caused a rapid dissociation of the complex (kdiss = 3.2 X 10(-2) s-1).     

The major plasma kallikrein inhibitor of guinea pig plasma.

A plasma kallikrein inhibitor in guinea pig plasma (KIP) was purified to homogeneity. KIP is a single chain protein and the apparent molecular weight is estimated to be 59,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. In amino acid composition, KIP is similar to human and mouse alpha 1-proteinase inhibitors and mouse contrapsin. KIP forms an equimolar complex with plasma kallikrein in a dose- and time-dependent fashion. The association rate constants for the inhibition of guinea pig plasma kallikrein by KIP, alpha 2-macroglobulin, C1-inactivator and antithrombin III were 2.5 +/- 0.3.10(4), 2.4 +/- 0.4.10(4), 6.6 +/- 0.5.10(4) and 9.1 +/- 0.6.10(2), respectively. Comparison of the association rate constants and the normal plasma concentrations of the four inhibitors demonstrates that KIP is ten-times as effective as alpha 2-MG and other two inhibitors are marginally effective in the inhibition of kallikrein. KIP inhibits trypsin and elastase rapidly, and thrombin and plasmin slowly, but is inactive for chymotrypsin and gland kallikrein. These results suggest that KIP is the major kallikrein inhibitor in guinea pig plasma and the proteinase inhibitory spectrum is unique to KIP in spite of the molecular similarity to alpha 1-proteinase inhibitor.     

Differences between the binding of trypsin and chymotrypsin by alpha 1-proteinase inhibitor

Complexes of alpha 1-proteinase inhibitor with proteases were examined by SDS-PAGE in 7.5% polyacrylamide gel and in a gel gradient. While the inhibitor-chymotrypsin complex was stable under both sets of conditions, the inhibitor-trypsin complex quantitatively dissociated under the second set of conditions, indicating that trypsin, unlike chymotrypsin, is not linked covalently to the inhibitor. Although the inhibitor sustained at least two discrete cleavages by trypsin, its overall recovery after dissociation was 100%. Due to an increased rate of autolytic breakdown in the presence of the inhibitor, the recovery of trypsin after dissociation was appreciably less than 100%. Based on these observations, a new theory of trypsin inhibition by alpha 1-proteinase inhibitor is proposed. This method is suitable for the examination of other inhibition systems as well.     

Interactions in vitro and in vivo between anionic and cationic porcine trypsin and porcine plasma proteinase inhibitors

A method for purifying porcine anionic and cationic trypsin is presented. Reaction mixtures with increasing amounts of the two porcine trypsins and porcine serum were studied in vitro to evaluate the relative importance of alpha 1-macroglobulin and alpha 2-macroglobulin as well as alpha 1-proteinase inhibitor in the rapid binding of porcine anionic and cationic trypsin. Porcine cationic trypsin was preferentially bound to alpha 1-macroglobulin, while anionic trypsin exhibited equal binding to both alpha-macroglobulins. Both trypsins were also bound by the alpha 1-proteinase inhibitor but not until alpha 1-macroglobulin approached saturation. Trypsin-alpha-macroglobulin complexes were cleared from plasma with a half-life of 6 min. For trypsin-alpha 1-proteinase inhibitor-complexes the half-life was 120 min. These findings are in accordance with results for other mammalian species, including man.     

The molecular stoichiometry of trypsin inhibition by human alpha-1-proteinase inhibitor

The stoichiometry of interaction of human alpha-1-proteinase inhibitor with porcine trypsin has been determined using a highly purified preparation of inhibitor. In contrast to the reports of others, one mole of alpha-1-proteinase inhibitor was found to inhibit two moles of trypsin. Disc gel electrophoresis indicates that the 2:1 complex is preferentially formed even when free alpha-1-proteinase inhibitor is still present.     

Stoichiometry of the interaction of human plasma alpha 1-proteinase inhibitor with subtilisin BPN'

Interaction of human plasma alpha 1-proteinase inhibitor (alpha 1PI) with subtilisin BPN' was assessed by spectrophotometric determination of the inhibitory capacity and by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). During the course of incubation of the enzyme and the inhibitor (E : I = 1 : 7.5) at pH 8.0 about 17% of the enzyme activity which had been inhibited initially was regenerated, indicating a temporary type of inhibition. The results of the titration experiments indicate that 9.8 mol of the inhibitor is required to inhibit 1 mol of the enzyme completely. However, patterns of 5% disc SDS-PAGE under non-reducing conditions revealed only an equimolar complex (Mr80K) of alpha 1PI with the enzyme and no other higher Mr component than the native inhibitor (Mr 56K). On the other hand, complete dissociation of the complex occurred under reducing conditions, producing an enzymatically modified inhibitor. When 5 21% gradient slab SDS-PAGE was employed, no complex formation was observed under either reducing or non-reducing conditions. With the gradient gel system, dissociation of the equimolar complex produced different forms of the inhibitor, that is, regeneration of an intact alpha 1PI under non-reducing conditions and an enzymatically modified form under reducing conditions. All these results indicate that the complex formed between subtilisin BPN' and human alpha 1PI is not so stable as that of the inhibitor with bovine chymotrypsin and that no covalent bond may be involved in the complex formation. The results also indicate that human alpha 1PI is not an effective inhibitor of subtilisin BPN' and behaves like a substrate for the enzyme.     

Recombinant DNA-derived forms of human alpha 1-proteinase inhibitor. Studies on the alanine 358 and cysteine 358 substituted mutants

The specificity and reactivity of human alpha 1-proteinase inhibitor has been investigated by in vitro mutagenesis of the reactive site P1 methionine 358 residue to alanine 358 and cysteine 358. A comparison of the second-order association rates of both uncharged mutants with 9 serine proteinases indicated that each reacted similarly to either the normal plasma inhibitor or to a mutant containing valine in this position (Travis, J., Owen, M., George, P., Carrell, R., Rosenberg, S., Hallewell, R. A., and Barr, P. J. (1985) J. Biol. Chem. 260, 4384-4389) when tested against either neutrophil or pancreatic elastase. However, oxidation, carboxymethylation, or aminoethylation of the cysteine mutant to yield a charged P1 residue resulted in a significant decrease in association rates with both elastolytic enzymes, and aminoethylation created an excellent trypsin and plasmin inhibitor. These results indicate that the specificity of alpha 1-proteinase inhibitor is determined in a general manner by the class of amino acid residue in the P1 position. Substitution within the same category, such as from valine to alanine or cysteine among the aliphatic hydrophobic residues, has little effect on association rates with the elastolytic enzymes tested. However, alteration from an uncharged to a charged residue may cause considerable changes in both inhibitor specificity and reactivity as noted here with the cysteine derivatives and also previously with a natural variant in which methionine 358 to arginine 358 conversion resulted in the production of a potent thrombin inhibitor (Owen, M. C., Brennan, S. O., Lewis, J. H., and Carrell, R. W. (1983) N. Engl. J. Med. 309, 694-698).     

Heparin strongly decreases the rate of inhibition of neutrophil elastase by alpha 1-proteinase inhibitor

Heparin depresses the second-order rate constant ka for the inhibition of neutrophil elastase by alpha 1-proteinase inhibitor. High molecular mass heparin decreases ka from 1.3 x 10(7) M-1 s-1 to a limit of 4.6 x 10(4) M-1 s-1. Low molecular mass heparin is about 7-fold less effective. Dermatan sulfate and chondroitin sulfate are less efficient. Heparin preparations used in clinical care also strongly depress ka when tested at concentrations corresponding to their clinical efficacy. Heparin also decreases the ka for the elastase/eglin c and the cathepsin G/alpha 1-proteinase inhibitor systems but not that for the alpha 1-proteinase inhibitor/pancreatic elastase or trypsin pairs. These results, together with Sepharose-heparin binding studies, indicate that the ka-depressing effect of the polymer is related to its ability to form a tight complex with elastase but not with alpha 1-proteinase inhibitor. One mol of high molecular mass heparin binds 3 mol of neutrophil elastase with a Kd of 3.3 nM. Low molecular mass heparin binds elastase with a 1:1 stoichiometry and a Kd of 89 nM. For both heparins ka is lowest when elastase is fully saturated with heparin. From this we conclude that heparin decreases ka, because the heparin-elastase complex is able to slowly react with alpha 1-proteinase inhibitor and not because the inhibitor slowly dissociates the heparin-elastase complex. These findings may have important pathophysiological bearing.     

Active site distortion is sufficient for proteinase inhibition by serpins: structure of the covalent complex of alpha1-proteinase inhibitor with porcine pancreatic elastase

We report here the x-ray structure of a covalent serpin-proteinase complex, alpha1-proteinase inhibitor (alpha1PI) with porcine pancreatic elastase (PPE), which differs from the only other x-ray structure of such a complex, that of alpha1PI with trypsin, in showing nearly complete definition of the proteinase. alpha1PI complexes with trypsin, PPE, and human neutrophil elastase (HNE) showed similar rates of deacylation and enhanced susceptibility to proteolysis by exogenous proteinases in solution. The differences between the two x-ray structures therefore cannot arise from intrinsic differences in the inhibition mechanism. However, self-proteolysis of purified complex resulted in rapid cleavage of the trypsin complex, slower cleavage of the PPE complex, and only minimal cleavage of the HNE complex. This suggests that the earlier alpha1 PI-trypsin complex may have been proteolyzed and that the present structure is more likely to be representative of serpin-proteinase complexes. The present structure shows that active site distortion alone is sufficient for inhibition and suggests that enhanced proteolysis is not necessarily exploited in vivo.     

Canonical inhibitor-like interactions explain reactivity of alpha1-proteinase inhibitor Pittsburgh and antithrombin with proteinases

The serpin antithrombin is a slow thrombin inhibitor that requires heparin to enhance its reaction rate. In contrast, alpha1-proteinase inhibitor (alpha1PI) Pittsburgh (P1 Met --> Arg natural variant) inhibits thrombin 17 times faster than pentasaccharide heparin-activated antithrombin. We present here x-ray structures of free and S195A trypsin-bound alpha1PI Pittsburgh, which show that the reactive center loop (RCL) possesses a canonical conformation in the free serpin that does not change upon binding to S195A trypsin and that contacts the proteinase only between P2 and P2'. By inference from the structure of heparin cofactor II bound to S195A thrombin, this RCL conformation is also appropriate for binding to thrombin. Reaction rates of trypsin and thrombin with alpha1PI Pittsburgh and antithrombin and their P2 variants show that the low antithrombin-thrombin reaction rate results from the antithrombin RCL sequence at P2 and implies that, in solution, the antithrombin RCL must be in a similar canonical conformation to that found here for alpha1PI Pittsburgh, even in the nonheparin-activated state. This suggests a general, limited, canonical-like interaction between serpins and proteinases in their Michaelis complexes.     

Fluorescence labeling of a single sulfhydryl group in human plasma alpha 1-proteinase inhibitor and characterization of the labeled inhibitor

A single cysteine residue present in human plasma alpha 1-proteinase inhibitor was labeled with a fluorescent sulfhydryl reagent, N-iodoacetyl-N'-(5-sulfo-1-naphthyl)ethylenediamine. The resulting fluorescent inhibitor retained nearly full inhibitory activity and formed complexes with bovine chymotrypsin, porcine pancreatic elastase, and bovine trypsin as revealed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Association rate constants for the interactions of the labeled inhibitor with the proteinases were determined to be 1.5 (+/- 0.4) X 10(6), 3.3 (+/- 0.3) X 10(5), and 1.4 (+/- 0.3) X 10(5) M-1 X s-1 for chymotrypsin, elastase, and trypsin, respectively. These values were found to be only slightly lower than those of the unlabeled inhibitor. Fluorescence emission spectra of the labeled inhibitor in the absence and presence of each proteinase were also examined, and little difference was observed between them.     

In vivo catabolism of heparin cofactor II and its complex with thrombin: evidence for a common receptor-mediated clearance pathway for three serine proteinase inhibitors

The plasma clearance of 125I-labeled human heparin cofactor II and its complex with thrombin was studied in mice to determine whether a specific mechanism exists for the catabolism of the inhibitor-proteinase complex. Initial studies demonstrated that murine plasma contains a heparin cofactor II-like inhibitor as shown by the presence of a dermatan sulfate-sensitive thrombin inhibitor. Human heparin cofactor II cleared from the circulation of mice with an apparent half-life of 80 min while heparin cofactor II-thrombin complexes cleared with an apparent half-life of only 10 min. The specificity of the clearance mechanism was investigated by clearance competition studies involving coinjection of excess unlabeled heparin cofactor II-alpha-thrombin, antithrombin III-alpha-thrombin, or alpha 1-proteinase inhibitor-elastase, and by tissue distribution studies. The results demonstrated that the clearance of 125I-labeled heparin cofactor II-alpha-thrombin is a receptor-mediated process, and that the same hepatocyte receptor system recognizes complexes containing heparin cofactor II, antithrombin III, and alpha 1-proteinase inhibitor.     

The kinetics of inhibition of alpha-thrombin in human plasma

Methods have been developed for kinetic studies of the inhibition of exogenous unmodified thrombin in human plasma containing normal levels of fibrinogen and calcium ion. To prevent interference by other proteases, factor VIII-deficient plasma was used and contained 50 nM Phe-Phe-Arg-chloromethyl ketone and 1 kallikrein-inactivating unit/ml aprotinin; neither inhibited thrombin at these levels. Two independent assays were used. The first was the discontinuous amidolytic assay of thrombin activity, which measures both free thrombin and thrombin-alpha 2-macroglobulin complex, and was used to estimate the rates of inhibition both by "inactivating" inhibitors, such as anti-thrombin and alpha 1-protease inhibitor, and by alpha 2-macroglobulin (alpha 2M). The contribution of alpha 2M was confirmed by a second method, which measured with time the generation of amidolytic activity due to the thrombin-alpha 2M complex. The total rate of thrombin inhibition in plasma containing 4 mM free Ca2+ was of the order of 1.9 min-1, of which 0.4 min-1 was due to alpha 2M and 0.9 min-1 was due to inhibitors that were removed when plasma was passed through heparin-agarose. Thrombin inhibition was also measured in varying dilutions of plasma and confirmed that total inhibition rate is approximately linearly related to plasma (and thus inhibitor) concentration. Negatively charged phospholipid micelles had very little effect on thrombin inhibition rate, but platelets accelerated inhibition to more than 2.5 min-1.     

The transferable tail: fusion of the N-terminal acidic extension of heparin cofactor II to alpha1-proteinase inhibitor M358R specifically increases the rate of thrombin inhibition

The conversion of the reactive center bond of the serpin alpha1-proteinase inhibitor (alpha1-PI, also known as alpha1-antitrypsin) from Met-Ser to Arg-Ser decreases the rate at which it inhibits neutrophil elastase and endows it with the ability to inhibit thrombin and activated protein C (APC). Another serpin, heparin cofactor II (HCII), contains a unique N-terminal extension that binds thrombin exosite 1. We fused residues 1-75 of HCII to the N-terminus of alpha1-PI M358R, forming an HCII-alpha1-PI chimera (HAPI M358R). It inhibited alpha-thrombin 21-fold faster than alpha1-PI M358R, with second-order rate constants of 2.3 x 10(8) M(-1) min(-1) versus 1.1 x 10(7) M(-1) min(-1), respectively. When gammaT-thrombin, which lacks an intact exosite 1, was substituted for alpha-thrombin, the kinetic advantage of HAPI M358R over alpha1-PI M358R was reduced to 9-fold, whereas APC and trypsin, proteases lacking exosite 1-like regions, were inhibited only 1.3- and 2-fold more rapidly by HAPI M358R than by alpha1-PI M358R, respectively. Maximal enhancement of alpha1-PI M358R activity required the acidic residues found between HCII residues 55 and 75, because no enhancement was observed either by fusion of residues 1-54 alone or by fusion of a mutated HCII acidic extension in which all Glu and Asp residues between positions 55 and 75 were neutralized by mutation. Fusing residues 55-75 to alpha1-PI M358R yielded a relative rate enhancement of only 6-fold, suggesting a need for the full tail region to achieve maximal enhancement. Our results suggest that transfer of the N-terminal acidic extension of HCII to alpha1-PI M358R enhanced its inhibition of thrombin by conferring the ability to bind exosite 1 on HAPI M358R. This enhancement may aid in efforts to tailor this inhibitor for therapeutic use.     

1-deoxynojirimycin impairs oligosaccharide processing of alpha 1-proteinase inhibitor and inhibits its secretion in primary cultures of rat hepatocytes

1-Deoxynojirimycin was found to inhibit oligosaccharide processing of rat alpha 1-proteinase inhibitor. In normal hepatocytes alpha 1-proteinase inhibitor was present in the cells as a 49,000 Mr high mannose type glycoprotein with oligosaccharide side chains having the composition Man9GlcNAc and Man8GlcNAc with the former in a higher proportion. Hepatocytes treated with 5 mM 1-deoxynojirimycin accumulated alpha 1-proteinase inhibitor as a 51,000 Mr glycoprotein with carbohydrate side chains of the high mannose type, containing glucose as measured by their sensitivity against alpha-glucosidase, the largest species being Glc3Man9GlcNAc. Conversion to complex oligosaccharides was inhibited by the drug. In addition, increasing concentrations of 1-deoxynojirimycin inhibited glycosylation resulting in the formation of some alpha 1-proteinase inhibitor with two instead of three oligosaccharide side chains. 5 mM 1-deoxynojirimycin inhibited the secretion of alpha 1-proteinase inhibitor by about 50%, whereas secretion of albumin was unaffected. The oligosaccharides of alpha 1-proteinase inhibitor secreted from 1-deoxynojirimycin-treated cells were characterized by their susceptibility to endoglucosaminidase H, incorporation of [3H]galactose, and [3H]fucose and concanavalin A-Sepharose chromatography. It was found that 1-deoxynojirimycin did not completely block oligosaccharide processing, resulting in the formation of alpha 1-proteinase inhibitor molecules carrying one or two complex type oligosaccharides. Only these alpha 1-proteinase inhibitor molecules processed to the complex type in one or two of their oligosaccharide chains were nearly exclusively secreted. This finding demonstrates the importance of oligosaccharide processing for the secretion of alpha 1-proteinase inhibitor.