Carbohydrate-controlled serine protease inhibitor (serpin) production in Bifidobacterium longum subsp. longum.

The Serine Protease Inhibitor (serpin) protein has been suggested to play a key role in the interaction of bifidobacteria with the host. By inhibiting intestinal serine proteases, it might allow bifidobacteria to reside in specific gut niches. In inflammatory diseases where serine proteases contribute to the innate defense mechanism of the host, serpin may dampen the damaging effects of inflammation. In view of the beneficial roles of this protein, it is important to understand how its production is regulated. Here we demonstrate that Bifidobacterium longum NCC 2705 serpin production is tightly regulated by carbohydrates. Galactose and fructose increase the production of this protein while glucose prevents it, suggesting the involvement of catabolite repression. We identified that di- and oligosaccharides containing galactose (GOS) and fructose (FOS) moieties, including the human milk oligosaccharide Lacto-N-tetraose (LNT), are able to activate serpin production. Moreover, we show that the carbohydrate mediated regulation is conserved within B. longum subsp. longum strains but not in other bifidobacterial taxons harboring the serpin coding gene, highlighting that the serpin regulation circuits are not only species- but also subspecies- specific. Our work demonstrates that environmental conditions can modulate expression of an important effector molecule of B. longum, having potential important implications for probiotic manufacturing and supporting the postulated role of serpin in the ability of bifidobacteria to colonize the intestinal tract.

A Commensal Bifidobacterium longum Strain Prevents Gluten-Related Immunopathology in Mice through Expression of a Serine Protease Inhibitor.

Microbiota-modulating strategies, including probiotic administration, have been tested for the treatment of chronic gastrointestinal diseases despite limited information regarding their mechanisms of action. We previously demonstrated that patients with active celiac disease have decreased duodenal expression of elafin, a human serine protease inhibitor, and supplementation of elafin by a recombinant Lactococcus lactis strain prevents gliadin-induced immunopathology in the NOD/DQ8 mouse model of gluten sensitivity. The commensal probiotic strain Bifidobacterium longum NCC2705 produces a serine protease inhibitor (Srp) that exhibits immune-modulating properties. Here, we demonstrate that B. longum NCC2705, but not a srp knockout mutant, attenuates gliadin-induced immunopathology and impacts intestinal microbial composition in NOD/DQ8 mice. Our results highlight the beneficial effects of a serine protease inhibitor produced by commensal B. longum strains.IMPORTANCE Probiotic therapies have been widely used to treat gastrointestinal disorders with variable success and poor mechanistic insight. Delivery of specific anti-inflammatory molecules has been limited to the use of genetically modified organisms, which has raised some public and regulatory concerns. By examining a specific microbial product naturally expressed by a commensal bacterial strain, we provide insight into a mechanistic basis for the use of B. longum NCC2705 to help treat gluten-related disorders.

Luminal contents from the gut of colicky infants induce visceral hypersensitivity in mice.

BACKGROUND: The pathophysiology of infantile colic is poorly understood, though various studies report gut microbiota dysbiosis in colicky infants. We aimed to test the hypothesis that colic-related dysbiosis is associated with visceral hypersensitivity triggered by an altered luminal milieu. METHODS: Fecal samples from seven colicky and seven non-colicky infants were studied. Fecal supernatants (FS) were infused into the colons of C57/Bl6 mice (n=10/specimen). Visceral sensitivity was subsequently assessed in the animals by recording their abdominal muscle response to colorectal distension (CRD) by electromyography (EMG). Serine and cysteine protease activities were assessed in FS with specific substrates. Infant fecal microbiota composition was analyzed by DNA extraction and 16S rRNA gene pyrosequencing. KEY RESULTS: FS from colicky infants triggered higher EMG activity than FS from non-colicky infants in response to both the largest CRD volumes and overall, as assessed by the area under the curve of the EMG across all CRD volumes. Infant crying time strongly correlated with mouse EMG activity. Microbiota richness and phylogenetic diversity were increased in the colicky group, without showing prominent microbial composition alterations. Only Bacteroides vulgatus and Bilophila wadsworthia were increased in the colicky group. Bacteroides vulgatus abundance positively correlated with visceral sensitivity. No differences were found in protease activities. CONCLUSIONS & INFERENCES: Luminal contents from colicky infants trigger visceral hypersensitivity, which may explain the excessive crying behavior of these infants. Additional studies are required to determine the nature of the compounds involved, their mechanism of action, and the potential implications of intestinal microbiota in their generation.

Bifidobacterium longum NCC3001 inhibits AH neuron excitability.

BACKGROUND: Bifidobacterium longum (B. longum) NCC3001 can affect behavior and brain biochemistry, but identification of the cellular targets needs further investigation. Our hypothesis was that the communication with the brain might start with action on enteric sensory neurons. METHODS: Ileal segments from adult mice were used to create a longitudinal muscle-myenteric-plexus preparation to expose sensory after-hyperpolarizing (AH) neurons in the myenteric plexus to allow access with microelectrodes. The intrinsic excitability of AH neurons was tested in response to the perfusion of conditioned media (B. longum culture supernatant) or unconditioned media (growth medium, MRS). KEY RESULTS: B. longum conditioned medium significantly reduced the excitability of AH neurons compared to perfusion with the unconditioned medium. Specifically, a reduction was seen in the number of action potentials fired per depolarizing test pulse, the instantaneous and time-dependent input resistances and the magnitude of the hyperpolarization-activated cationic current (Ih ). CONCLUSIONS & INFERENCES: The probiotic B. longum reduces excitability of AH sensory neurons likely via opening of potassium channels and closing of hyperpolarization-activated cation channels.

The anxiolytic effect of Bifidobacterium longum NCC3001 involves vagal pathways for gut-brain communication.

BACKGROUND: The probiotic Bifidobacterium longum NCC3001 normalizes anxiety-like behavior and hippocampal brain derived neurotrophic factor (BDNF) in mice with infectious colitis. Using a model of chemical colitis we test whether the anxiolytic effect of B. longum involves vagal integrity, and changes in neural cell function. Methods Mice received dextran sodium sulfate (DSS, 3%) in drinking water during three 1-week cycles. Bifidobacterium longum or placebo were gavaged daily during the last cycle. Some mice underwent subdiaphragmatic vagotomy. Behavior was assessed by step-down test, inflammation by myeloperoxidase (MPO) activity and histology. BDNF mRNA was measured in neuroblastoma SH-SY5Y cells after incubation with sera from B. longum- or placebo-treated mice. The effect of B. longum on myenteric neuron excitability was measured using intracellular microelectrodes. KEY RESULTS: Chronic colitis was associated with anxiety-like behavior, which was absent in previously vagotomized mice. B. longum normalized behavior but had no effect on MPO activity or histological scores. Its anxiolytic effect was absent in mice with established anxiety that were vagotomized before the third DSS cycle. B. longum metabolites did not affect BDNF mRNA expression in SH-SY5Y cells but decreased excitability of enteric neurons. CONCLUSIONS & INFERENCES: In this colitis model, anxiety-like behavior is vagally mediated. The anxiolytic effect of B. longum requires vagal integrity but does not involve gut immuno-modulation or production of BDNF by neuronal cells. As B. longum decreases excitability of enteric neurons, it may signal to the central nervous system by activating vagal pathways at the level of the enteric nervous system.

Magnet Tracking: a new tool for in vivo studies of the rat gastrointestinal motility.

Digestive motility was studied in the rat using a miniaturized version of the Magnet Tracking system which monitored the progression of a small magnetic pill through the entire digestive tract. The dynamics of movement was followed and three-dimensional (3-D) images of digestive tract were generated. After a retention period in the stomach and rapid passage through duodenum, the magnet progressed along the small intestine with gradually decreasing speed and longer stationary periods. It remained in the caecum for variable intervals. In the colon, periods of progress alternated with long quiescent periods. Gastric activity oscillated at 5-6 min(-1). In the small intestine, two frequency domains coexisted, showing independent modulations and proximo-distal gradients (40 to >32 and 28 to >20 min(-1)). Caecal oscillations were of 1.5 min(-1). The data allowed the magnet location and calculation of gastric and small intestinal transit times (58 +/- 36 and 83 +/- 14 min respectively), both significantly prolonged by oleate administration (243 +/- 130 and 170 +/- 45 min respectively). Magnet Tracking is a non-invasive tool to study the in vivo spatial and temporal organization of gastrointestinal motility in the rat.

Essential oils as components of a diet-based approach to management of Helicobacter infection.

An increased density of Helicobacter pylori in the gastric mucosa can be associated with more severe gastritis and an increased incidence of peptic ulcers. Therefore, people with asymptomatic gastritis would certainly benefit from a nutritional approach to help them manage the infection and therefore decrease the risk of development of associated pathologies. We analyzed the activities of 60 essential oils against H. pylori P1 and identified 30 oils that affected growth, with in vitro inhibition zones ranging between 0.7 and 6.3 cm in diameter. We further analyzed the effects of 16 oils with different activities on H. pylori P1 viability. Fifteen showed strong bactericidal activities, with minimal bactericidal concentrations after 24 h ranging from 0.02 to 0.1 g/liter at pH 7.4. Even though slight variations in activities were observed, the essential oils that displayed the strongest bactericidal potentials against H. pylori P1 were also active against other Helicobacter strains tested. Among the pure constituents of different essential oils tested, carvacrol, isoeugenol, nerol, citral, and sabinene exhibited the strongest anti-H. pylori activities. Although oral treatment of H. pylori SS1-infected mice with carrot seed oil did not result in significant decreases in the bacterial loads in the treated animals compared to those in the control animals, in all experiments performed, the infection was cleared in 20 to 30% of carrot seed oil-treated animals. Our results indicate that essential oils are unlikely to be efficient anti-Helicobacter agents in vivo. However, their effects may not be irrelevant if one plans to use them as food additives to complement present therapies.

Interplay between galanin and leptin in the hypothalamic control of feeding via corticotropin-releasing hormone and neuropeptide Y.

Over long periods, feeding and metabolism are tightly regulated at the central level. The total amount of nutrients ingested is thought to result from a delicate balance between orexigenic and anorexigenic factors expressed and secreted by specialized hypothalamic neuronal populations. We have developed a system of perifused hypothalamic neurons to characterize the relationships existing between the orexigenic peptide galanin and two other physiological modulators of feeding: neuropeptide Y (NPY) and corticotropin-releasing hormone (CRH). We demonstrated that galanin stimulates CRH and NPY secretion from hypothalamic neurons in a dose-dependent manner. Exposure to leptin for 24 h before galanin stimulation decreased NPY secretion by 30%, leaving the responsiveness of CRH neurons intact. These results suggest that CRH and NPY neurons participate to the intrahypothalamic signaling pathway of galanin, an observation that can explain the lower potency of galanin to stimulate food intake in vivo compared with NPY. The differential effects exerted by leptin on CRH and NPY suggest that there exists a subset of NPY neurons that are exquisitely sensitive to marked variations in leptin levels, and that the CRH neurons are less responsive to increases in leptin concentrations.

Cloning and characterization of Helicobacter pylori succinyl CoA:acetoacetate CoA-transferase, a novel prokaryotic member of the CoA-transferase family.

Sequencing of a fragment of Helicobacter pylori genome led to the identification of two open reading frames showing striking homology with Coenzyme A (CoA) transferases, enzymes catalyzing the reversible transfer of CoA from one carboxylic acid to another. The genes were present in all H. pylori strains tested by polymerase chain reaction or slot blotting but not in Campylobacter jejuni. Genes for the putative A and B subunits of H. pylori CoA-transferase were introduced into the bacterial expression vector pKK223-3 and expressed in Escherichia coli JM105 cells. Amino acid sequence comparisons, combined with measurements of enzyme activities using different CoA donors and acceptors, identified the H. pylori CoA-transferase as a succinyl CoA:acetoacetate CoA-transferase. This activity was consistently observed in different H. pylori strains. Antibodies raised against either recombinant A or B subunits recognized two distinct subunits of Mr approximately 26,000 and 24, 000 that are both necessary for H. pylori CoA-transferase function. The lack of alpha-ketoglutarate dehydrogenase and of succinyl CoA synthetase activities indicates that the generation of succinyl CoA is not mediated by the tricarboxylic acid cycle in H. pylori. We postulate the existence of an alternative pathway where the CoA-transferase is essential for energy metabolism.

Plasminogen: an important hemostatic parameter in septic patients.

BACKGROUND: Previously we observed in some but not all septic patients a low plasma concentration of plasminogen. OBJECTIVES: To investigate prospectively whether plasma levels of plasminogen or the ratio of plasminogen to alpha-2-antiplasmin have a prognostic value for survival from sepsis and to study the variation of other hemostatic parameters during septicemia. PATIENTS: The study population consisted of 45 consecutive patients with septicemia, 15 non-septic patients from the same intensive care unit and 30 healthy volunteers. MEASUREMENTS AND MAIN RESULTS: Plasminogen concentrations were significantly lower (p < 0.001) in plasma of septic patients (median 0,62 IU/ml range: 0.15-1,06) than in plasma of healthy controls (median 1.00 IU/ml, range: 0.75-1.10) or of non-septic intensive care patients (median 1.00 IU/ml, range: 0.82-1.08). Among the other parameters tested, plasminogen activator inhibitor (PAI-1) antigen concentration and PAI activity were similar in septic and non-septic intensive care patients, but higher than in healthy controls. Concentrations of elastase-alpha-1-protease inhibitor or of thrombin-antithrombin complexes were higher in septic patients than in non-septic intensive care patients or healthy controls. A degraded form of plasminogen of 38 kDa was detected by Western blot analysis in the plasma of septic patients, but not in plasma of non-septic intensive care patients or controls. Plasminogen alone or the ratio of plasminogen to antiplasmin were good markers for survival from septicemia. E.g. for plasminogen at a cut off of 0.65 IU/ml, sensitivity was 90.5% and specificity 66.7%, whereas for the ratio of plasminogen over antiplasmin at a cut off ratio of 0,65 IU/ml, sensitivity was 95.2% and specificity 70.8%. CONCLUSION: Plasminogen or the ratio of plasminogen to antiplasmin are sensitive markers for survival in patients with septicemia.

Hypertension increases connexin43 in a tissue-specific manner.

BACKGROUND: Connexin43 (Cx43), a membrane protein involved in the control of cell-to-cell communication, is thought to play a role in the contractility of the vascular wall and in the electrical coupling of cardiac myocytes. The aim of this study was to investigate the effects of experimental hypertension on Cx43 expression in rat aorta and heart. METHODS AND RESULTS: Rats were made hypertensive after one renal artery was clipped (two kidney, one-clip renal model) or after the administration of deoxycorticosterone and salt (DOCA-salt model). After 4 weeks, all rats showed a similar increase in intra-arterial mean blood pressure and in the thickness of both the aortic wall and the heart. Northern blot analysis of aorta mRNA and immunolabeling for Cx43 showed that hypertensive rats expressed twice as much Cx43 in aorta as the control animals. In contrast, no difference in Cx43 mRNA or in the immunolabeled protein was observed in heart. CONCLUSIONS: The results show that rats exhibiting a similar degree of blood pressure elevation, as the result of different mechanisms, feature a comparable increase in Cx43 gene expression, which was observed in the aortic but not in the cardiac muscle. These data suggest that localized mechanical forces induced by hypertension are major tissue-specific regulators of Cx43 expression.

Renin and angiotensin II receptor gene expression in kidneys of renal hypertensive rats.

The aim of this investigation was to examine the interrelation between renal mRNA levels of renin and angiotensin II receptor type 1 (AT1) in a renin-dependent form of experimental hypertension. Rats were studied 4 weeks after unilateral renal artery clipping. Mean blood pressure and plasma renin activity were significantly higher in the hypertensive rats (n = 10 206 +/- mm Hg and 72.4 +/- 20.9 ng/mL-1/h-1, respectively) than in sham-operated controls (n = 10, 136 +/- 3 mm Hg and 3.3 +/- 0.5 ng/mL-1/h, respectively). Northern blot analysis of polyA+ RNA obtained from the kidneys of renal hypertensive rats showed increased levels of renin mRNA in the clipped kidney, whereas a decrease was observed in the unclipped kidney. Plasma renin activity was directly correlated with renin mRNA expression of the poststenotic kidney (r = .94, P < .01). AT1 mRNA expression was lower in both kidneys of the hypertensive rats. This downregulation was specific for the AT1A subtype since the renal expression of the AT1B subtype remained normal in hypertensive rats. The downregulation of the renal AT1A receptor may be due to high circulating angiotensin II levels. This is supported by the significant inverse correlation (r = .71, P < .01) between plasma renin activity and AT1A mRNA expression measured in the clipped kidney of the hypertensive rats.

Localization and production of plasminogen activator inhibitor-1 in human healthy and atherosclerotic arteries.

High plasma levels of plasminogen activator inhibitor type-1 (PAI-1), the principal inhibitor of the fibrinolytic system, have been associated with thrombotic and arterial disease. To study PAI-1 expression in healthy and atherosclerotic human arteries, a detailed analysis was made by light and electron microscopy immunocytochemistry and by in situ hybridization. In healthy arteries PAI-1 was found both at the level of endothelial cells and of smooth muscle cells (SMCs) of the arterial media. In early atherosclerotic lesions PAI-1 was also detected in intimal SMCs and in extracellular areas in association with vitronectin. Immunogold analysis by electron microscopy revealed PAI-1 in vesicular structures in endothelial cells and in SMCs with normal or foam cell characteristics. In advanced atheromatous plaques, PAI-1 mRNA expression in SMCs within the fibrous cap was increased compared with SMCs located in the adjacent media or in normal arterial tissue. PAI-1 mRNA was also detected in macrophages located at the periphery of the necrotic core. The increased synthesis of PAI-1 by cellular components of the atherosclerotic plaque and the extracellular accumulation of PAI-1 may contribute to the thrombotic complications associated with plaque rupture and possibly play a role in the accumulation of extracellular matrix deposits.

Transcriptional antagonism of phorbol ester-mediated induction of plasminogen activator inhibitor types 1 and 2 by cyclic adenosine 3',5'-monophosphate.

Gene expression of plasminogen activator inhibitor (PAI) types 1 and 2 is modulated by the protein kinase-C (PKC) and cAMP-dependent protein kinase-A (PKA) signal transduction pathways. To determine whether the PKC and PKA pathways functionally interact during modulation of PAI gene expression, we assessed changes in gene transcription rates, mRNA, and antigen levels of PAI-1 and PAI-2 in HT-1080 fibrosarcoma cells treated with the PKC activator phorbol 12-myristate 13-acetate (PMA), alone or in combination with cAMP agonists and analogs. PMA produced a transient increase in PAI-1 and a sustained increase in PAI-2, which was evident at the level of gene transcription and mRNA. Treatment with the cAMP agonist forskolin or the cAMP analog 8-bromo-cAMP decreased constitutive and PMA-mediated expression of PAI-1 mRNA. PAI-2 mRNA was below detection limits in nontreated and cAMP-treated cells. However, elevated levels of cAMP reduced the stimulatory effect of PMA on PAI-2 mRNA. The antagonism of the PMA effect by cAMP was evident at the level of gene transcription, suggesting that the end point of the functional interplay between the PKC and PKA pathways requires modulation of a nuclear transcription factor(s). Our results suggest that the PKC- and PKA-dependent signaling pathways have counteractive effects on transcriptional expression of the PAI-1 and PAI-2 genes in HT-1080 cells.

Constitutive plasminogen activator inhibitor 1 (PAI-1) biosynthesis in human Hep G2 hepatoma cells is maintained by an autocrine factor.

We studied the production of PAI-1 by human Hep G2 hepatoma cells. When culturing these cells in fresh medium (FM), PAI-1 accumulation rate was not linear: PAI-1 antigen after 24 h was 200 ng/10(6) cells while in subsequent 24 h periods, it was on average 1,000 ng/10(6) cells. Culture of Hep G2 cells in regular changes of a 12 h conditioned medium (CM) obtained from other Hep G2, instead of in FM, resulted in a 2-fold higher PAI-1 accumulation. In cells incubated in FM, PAI-1 mRNA declined rapidly after medium change and returned to basal levels after 24 h. In contrast, PAI-1 mRNA remained relatively stable when CM was used. The acute phase mediator interleukin 6 (IL-6) was not responsible for the autocrine stimulation of PAI-1: neither IL-6 nor antibodies to IL-6 altered the observed variations in PAI-1 expression. Our studies suggest the presence of an unknown PAI-1 stimulating factor.

Induction of plasminogen activator inhibitor 1 biosynthesis by hyperthermia.

Hyperthermia is a clinical sign of inflammation and constitutes in itself an adaptive defense mechanism. The fibrinolytic system, a highly regulated proteolytic system, is involved in inflammatory processes. Plasminogen activator inhibitor 1 (PAI-1) is the principal inhibitor of the two activators of the fibrinolytic system: tissue- and urokinase-type PAs (t-PA and u-PA). Our present paper provides the first evidence that hyperthermia can directly induce PAI-1. A moderate heat stress, sufficient to induce heat shock protein 70 mRNA approximately 100-fold, resulted in a two- to three-fold increase in functionally active PAI-1 in the conditioned medium of human HT-1080 fibrosarcoma and Hep G2 hepatoma cells. Exposure of these cells to 42 degrees C led to a similar two-fold and two- to five-fold induction of PAI-1 mRNA expression in HT-1080 and Hep G2 cells, respectively, as has been determined by using both oligo d(T) selected and total RNA preparations. These results suggest that the observed increase in PAI-1 accumulation is due to an induction of PAI-1 biosynthesis. Run-on transcription analysis indicates that the induction of PAI-1 biosynthesis by hyperthermia is mediated by a stimulation of PAI-1 gene transcription. No significant effect of hyperthermia was found on t-PA or u-PA at the level of antigen accumulation, mRNA, and gene transcription in human HT-1080 fibrosarcoma cells. These results point to an additional regulatory mechanism of fibrinolysis in the context of inflammation.

Regulatory elements involved in constitutive and phorbol ester-inducible expression of the plasminogen activator inhibitor type 2 gene promoter.

Gene transcription rates and mRNA levels of plasminogen activator inhibitor type 2 (PAI-2) are markedly induced by the tumor promoting agent phorbol 12-myristate 13-acetate (PMA) in human HT1080 fibrosarcoma cells. To identify promoter elements required for basal-, and phorbol ester-inducible expression, deletion mutants of the PAI-1 promoter fused to the chloramphenicol acetyl transferase (CAT) reporter gene, were transiently expressed in HT1080 cells. Constitutive CAT activity was expressed from constructs containing more than 215 bp of promoter sequence, whereas deletion to position -91 bp abolished CAT gene expression. Treatment of transfected cells with PMA resulted in a three- to ten-fold increase in CAT expression from all constructs except from the construct shortened to position -91. DNAse1 protection analysis of the promoter region between -215 and the transcription initiation site revealed numerous protected regions, including two AP1-like binding sites (AP1a and AP1b) and one CRE-like element. Site-directed mutagenesis of the AP1a site or of the CRE-like site resulted in the loss of basal CAT activity and abolished the PMA effect, whereas mutagenesis of AP1b only partially inhibited basal and PMA-mediated expression. Our results suggest that the PAI-2 promoter contains at least two elements required for basal gene transcription and PMA-mediated induction.

Plasminogen-activator inhibitor type 1 is a potent natural inhibitor of extracellular matrix degradation by fibrosarcoma and colon carcinoma cells.

We have analyzed the role of plasminogen-activator inhibitor type 1 (PAI-1) in the regulation of tumor cell-mediated extracellular matrix degradation. Immunocytochemical analysis revealed PAI-1 associated with microgranular and fibrillar material of the extracellular matrix and demonstrated the presence of PAI-1 as a cell surface-associated antigen. Transforming growth factor beta significantly reduced matrix degradation mediated by HT-1080 human fibrosarcoma cells. This inhibition was correlated with an increase in PAI-1 antigen expression, whereas urinary-type plasminogen activator (u-PA) secretion was unaffected. In this experimental system, PAI-1 regulated extracellular matrix breakdown, as added PAI-1 inhibited matrix solubilization, whereas monoclonal antibodies to PAI-1 increased it. A cell line (LPAI) producing high levels of biologically active PAI-1 was established by transfection of a human PAI-1 cDNA clone into mouse L cells. Coculture experiments demonstrated that LPAI cells prevented matrix degradation by Lu-PA cells (L cells expressing high levels of u-PA) or Co-115 human colon carcinoma cells (expressing tissue-type plasminogen activator). These results indicate that PAI-1 may play a critical role in the regulation of extracellular matrix degradation during tumor cell invasion.

Mini-plasminogen like molecule in septic patients.

Plasma from 7 septic patients with positive blood cultures were studied. None of them presented either clinical or laboratory evidence of Disseminated Intravascular Coagulation. The white cells count varied between 5 and 45 X 10(9)/l. In plasma functional plasminogen levels varied between 25 and 45%, while those of alpha 2-antiplasmin were normal (80-105%). The levels of elastase ranged between 250 and 750 micrograms/ml. Leukocyte elastase digests plasminogen "in vitro" and is able to produce several fragments; one of them called mini-plasminogen lacking lysine binding sites; therefore it does not bind to lysine-Sepharose 4B. Two different behaviors were observed in the plasmatic plasminogen of these patients with respect to their binding capacity to lysine-Sepharose 4 B. 3 patients had plasminogen which did not bind to lysine-Sepharose 4 B; the other 4 had two different components, one of which bound to lysine-Sepharose 4 B and another one which did not bind. Previous studies "in vitro" have shown that leukocyte elastase modifies alpha 2-antiplasmin, initially producing a non-plasminogen binding form. A free alpha 2-antiplasmin (non-plasminogen binding form) was detected in the plasma of these patients with sepsis by crossed immunoelectrophoresis with plasminogen in the first dimension. It seems tenable that high levels of leukocyte elastase could be responsible for these findings although, the possible relationships to leukocyte elastase still remain to be proven but could possibly explain this effect.