Complement fragment C3a in plasma of asphyxiated neonates

Recent clinical studies with adult polytrauma patients indicate that elevated plasma levels of anaphylatoxin C3a correlate with the subsequent development of the adult respiratory distress syndrome (ARDS). However, there are no parameters which allow a reliable diagnosis of ARDS in neonates. As the most predisposing condition for ARDS seems to be shock, plasma C3a was determined in 30 ventilated premature infants and neonates with respiratory distress syndrome (birth weights 660-3350 g) within the first 24 h post partum or 6-24 h after acute asphyxia or shock during the neonatal period. The range of C3a, measured by ELISA, was between 57 and 1000 ng/ml. In the asphyxia group (n = 15) peak levels of C3a in plasma (mean 388 ng/ml) were significantly higher (P less than 0.001) than in the control group (mean 153 ng/ml). In some neonates with suspected ARDS, additional samples were taken. A rise in C3a between days 2 and 8 was associated with a fatal outcome of the disease. As in adults, C3a might be a useful indicator for ARDS in neonates.

Inherited neuroaxonal dystrophy in C6 deficient rabbits

We report the occurrence of a progressive neurological syndrome clinically characterized by subacute motor neuropathy in offspring of C6 deficient rabbits. On the basis of the pedigree analysis, the disease appears to be genetically transmitted, most probably with an autosomal recessive mode of inheritance. Pathological studies of affected animals revealed: transmitted, most probably with an autosomal recessive mode of inheritance. Pathological studies of affected animals revealed: 1) severe axonal degeneration in the sciatic nerve system involving mainly motor fibers; 2) occasional peripheral axonal enlargement closely associated with axonal degeneration; 3) presence of structured abnormal material in normal-size myelinated fibers of central nervous system (CNS) and peripheral nervous system (PNS); and 4) widespread occurrence of dystrophic axons and axonal spheroids in the gray matter of CNS. By ultrastructural examination, dystrophic axons are filled with tubulovesicular material, stalks of parallel membranes and dense bodies similar to what is described in human neuroaxonal dystrophies (NAD). The disease manifested by C6 deficient rabbits may represent an animal model of primary human NAD.

Generation of anaphylatoxin C3a in plasma and bronchoalveolar lavage fluid in trauma patients at risk for the adult respiratory distress syndrome

OBJECTIVE

To determine the generation of anaphylatoxin C3a in plasma and bronchoalveolar lavage fluid in trauma patients at risk for the adult respiratory distress syndrome (ARDS).

DESIGN

Prospective study.

SETTING

ICU in a university hospital.

PATIENTS

Severely traumatized patients at risk for the ARDS (n = 25).

INTERVENTION

EDTA plasma samples and bronchoalveolar lavage fluid were obtained.

MEASUREMENTS AND MAIN RESULTS

Complement proteins C3, C4, C5, and the inhibitors C1-inhibitor, Factor H, and Factor I were quantitated in EDTA-plasma samples obtained every 6 hrs during the first 48 hrs after ICU admission and every morning from days 4 to 14 after injury. In bronchoalveolar lavage fluid, the complement activation production of C3a-desArg was quantitated and the volume of epithelial lining fluid was calculated. All patients showed a decrease of the complement proteins C3, C4, C5 and of the inhibitors C1-inhibitor, Factor H, and Factor I during the first 24 hrs, indicating complement consumption. Patients developing ARDS (n = 11) showed significantly higher C3 concentrations and a higher C3a/C3 ratio in the first few hours after multitrauma. Follow-up bronchoalveolar lavages demonstrated highly increased amounts of C3a in epithelial lining fluid during the first 24 hrs, mainly in ARDS patients and, to a lesser degree, in non-ARDS patients. To determine the origin of C3a in bronchoalveolar lavages, the ratio of C3a in epithelial lining fluid and plasma was calculated.

CONCLUSION

The C3a of epithelial lining fluid to plasma ratio was extremely high in patients developing ARDS, but even the non-ARDS group had a ratio greater than 1, indicating that a substantial local complement activation occurs in the lung.

Characterization of I/F1 glycoprotein as a receptor for Mycoplasma pneumoniae

Serologic evidence of anti-I and anti-Fl cold agglutinins occurring in mycoplasma infections led to the isolation of I/Fl glycoprotein from human erythrocyte membranes. Mycoplasma pneumoniae bound to purified I/Fl glycoprotein in a dose-dependent fashion depending on sialylated carbohydrate determinants. This was shown by the decreased binding of mycoplasmas to either sialidase-treated I/Fl glycoprotein (dot blot analysis) or sialidase-treated erythrocytes (hemagglutination test). Structural properties of the receptor for optimal binding could be explored by hemagglutination inhibition assays. Glycophorins were excluded as receptors. These results indicate that Fl (and I) antigens are receptors for M. pneumoniae.

Heparin-coated versus non-coated surfaces for extracorporeal circulation

Studies were made to compare completely heparin-bonded (HBS) and conventional extracorporeal circulation surfaces using capillary membrane oxygenators (CMO) in sheep and dogs for up to five days. The aims were: to investigate the need for systemic heparinization in the case of heparin-coated surfaces, to assess blood compatibility and gas exchange performance of both systems and the extent of complement activation, and to find solutions for plasma leakage by the use of CMO. All studies were performed under standardized conditions, such as drugs, surgery, priming, blood flow rate etc. For heparin-coated surface studies all blood interfaces (CMO, catheters, tubes, etc) were coated. It was possible to eliminate systemic heparinization totally when HBS were used. During the five-day non-heparin application period blood coagulation parameters were almost unchanged and in the physiological range, platelets did not drop below 80%, hemolysis was negligible and gas exchange performance was unaffected. Less complement activation occurred with HBS than with non-coated surfaces.

The complement system in HIV disease

Different aspects of the relationship between the HIV infection and the complement system were studied. 1. No significant differences were found between seronegative controls, asymptomatic, and symptomatic (ARC, AIDS) HIV-seropositive patients in the plasma levels of complement components C4, Bf, and C3. 2. Using sensitive ELISA assays, a significant increase was observed in the levels of protein-protein complexes which are formed at the activation of the classical (C1r-C1s-C1-INH) and alternative (C3b-Bb-P) pathways, indicating that both complement pathways are activated in the HIV disease. No significant differences were found, however, in the levels of these complexes between the groups of asymptomatic and symptomatic HIV-infected patients. 3. Artificial immune complexes of synthetic peptides representing some immunodominant epitopes of HIV envelope (gp120, and gp41) proteins, and human polyclonal anti-HIV IgG were found to weakly activate both the classical and alternative complement pathways. 4. An elevated percentage of the lymphocytes carrying a complement activation fragment, C3d, was detected in the blood of HIV seropositive patients as compared to the seronegative controls. No significant positive correlation was found between the percentage of these cells and that of any T cell subsets tested.

[Post-traumatic alveolar changes in lung contusion]

Bronchoalveolar lavage (BAL) specimens taken from nine patients with lung contusion following multiple trauma were compared with specimens from different control groups. Early interstitial and intra-alveolar reactions are PMN degranulation, mediator release and high protein leakage. The alveolar reactions are similar in extent to the reaction found in post-traumatic ARDS.

Complement activation by house dust: reduced reactivity of serum complement in patients with bronchial asthma

Among ten different allergens, house-dust extract proved to be the most potent complement activator. It was therefore chosen to investigate the susceptibility of complement in the serum and bronchoalveolar lavage fluid of patients with extrinsic asthma and control persons. Complement activation in serum was assessed by the appearance of C3d as well as the activation-specific protein-protein complexes C1rs-C1inhibitor (classical pathway) and C3b(Bb)P (alternative pathway). Complement was activated via both the classical and the alternative pathway in a dose- and time-dependent manner. In contrast to earlier observations, however, complement was less affected in the serum of asthmatics than in the serum of normal individuals. Differences were restricted to alternative-pathway activation, probably due to preactivation and/or a significantly higher serum concentration of the regulatory protein factor H in asthmatic patients. In vitro generation of C3a in bronchoalveolar lavage fluid could not be achieved, although the presence of alternative pathway proteins C3, B and D was demonstrated.

Differences in the complement activation induced by preformed and nascent immune complexes

While preformed BSA-anti-BSA immune complexes (PIC) bind efficiently to human RBC after their interaction with human complement, nascent BSA-anti-BSA immune complexes (NIC) formed in the presence of complement do not bind to autologous RBC. The same results were obtained with tetanus toxoid-anti-tetanus toxoid PIC and NIC. In order to elucidate the causes of this marked difference between the RBC-binding properties of PIC and NIC, the profile of complement activation induced by them was compared using haemolytic assays and sensitive ELISA tests. BSA-anti-BSA NIC activated C1 more efficiently than PIC. This was reflected in a higher C4 content of the isolated NIC and higher C1 INH-C1s and lower C1q-fibronectin complex level in the NIC-treated serum as compared to the PIC-treated ones. Although isolated NIC contained more C3 than isolated PIC did, there was no significant difference in the AP activation. These findings suggest that the failure of NIC to bind to RBC is not due to a lack of C4-binding, or C3-binding and/or activation, but rather to the special structure of this type of complex.

Neolacto (type-2 chain)-sialoautoantigens recognized by human cold agglutinins

The antigens Fl and Vo, recognized by cold agglutinins, are protease-resistant differentiation antigens like I and i antigens. In contrast to Ii, Fl and Vo are sialidase-susceptible antigens like the protease-resistant Gd antigens. Because I/i antigens are branched/linear type-2 (poly-N-acetyllactosamine or neolacto series) chains which are cleaved from red cells by endo-beta-galactosidase, the effect of the enzyme on Fl, Vo and Gd antigens was studied. Fl and Gd antigens, previously shown to be sialylated branched (Fl), and linear and branched (Gd) type-2 sequences, were not inactivated on red cells by the enzyme, whereas Vo antigenicity was completely abolished. The data suggest that Vo antigen is expressed on sialylated linear poly-N-acetyllactosamine sequence.

[Recurrent post-infectious immune complex glomerulonephritis with persistent activation of the alternative complement pathway]

Recurring upper respiratory tract infections in a previously healthy 4 years old boy led to an acute nephritic syndrome requiring haemodialysis. Renal biopsy showed poststreptococcal like immune complex glomerulonephritis with subepithelial humps, C3 deposition and crescent formation. Renal function improved after methyl-prednisolone pulse therapy. Two relapses occurred in the following six months which responded to immunosuppressive therapy. 21 months following the first presentation the boy has normal glomerular filtration rate but persistent glomerular proteinuria and haematuria. Detailed complement analyses revealed a persistent activation of the alternate pathway.

Complement activation and the prognostic value of C3a in patients at risk of adult respiratory distress syndrome

In vivo and in vitro studies have shown that complement activation plays an important role in the pathogenesis of the adult respiratory distress syndrome (ARDS). In a prospective study of polytrauma patients at risk of ARDS (n = 38) complement parameters were determined over a period of 14 days in serial plasma samples (obtained every 6 h during the first 48 h). Polytrauma induced a rapid and remarkable complement activation. Low levels of the complement proteins C3, C4, C1 inhibitor (C1 INH) factor I and factor H during the first 48 h indicated complement consumption in all patients. Elevated C3a levels in the first few hours after injury were associated with the later development of ARDS. A more sensitive indicator than C3a alone was the calculated C3a:C3 ratio discriminating ARDS and non-ARDS patients. A second rise of C3a levels and C3a:C3 ratio from day 4 on paralleled the course of extravascular lung water. To assess the mode of complement activation, the activation-specific protein complexes C1rC1s-C1 INH and C3b(Bb)P were measured in some of the patients. We demonstrate that in the first 48 h complement activation occurred via the alternative pathway only and was later followed by an additional activation via the classical pathway. Our observations suggest that monitoring of C3a and C3 in plasma can identify polytrauma patients at high risk for ARDS at an early stage of the disease.

Binding and activation of C4 and C3 on the red cell surface by non-complement enzymes

We investigated the binding of C4 and C3 to red cell surfaces by non-complement enzymes. Cell bound C components were quantitated by a radioimmunoassay, the chain structure of bound components was analyzed by Western blotting and the hemolytic activity of bound components was determined. Trypsin, chymotrypsin, plasmin, elastase, thrombin, kallikrein and enzymes from Bacillus subtilis, Staphylococcus aureus and Streptomyces griseus all were found capable of binding C4b and C3b to sheep red cells. C4b bound by any of these enzymes was hemolytically active; both classical and alternate pathway activity of C3 could be demonstrated for most enzymes except plasmin and thrombin. In addition, trypsin and the bacterial enzymes were also able to generate the classical pathway C3-convertase from C4b + C2. The hemolytic efficiency of enzyme bound C4b and C3b was about the same as for these molecules bound by complement enzymes. In contrast, the process of binding by the non-complement enzymes was several hundred-fold less efficient than by cell bound complement enzymes. The results demonstrate that several enzymes can replace the C1 and C42 enzymes in the classical pathway and are able to initiate the alternative pathway by activating C3 and binding C3b to the cell surface.

Functional analysis and quantification of the complement C3 derived anaphylatoxin C3a with a monoclonal antibody

The C3 fragment C3a belongs to the anaphylatoxins. It has immune regulatory activity and contributes to the pathogenesis of the adult respiratory distress syndrome (ARDS). The low molecular weight (9 kD) of C3a complicates the production of antibodies to C3a. We obtained a monoclonal antibody (designated H13) to human C3a. It reacts with C3a or C3a-desArg and with native C3 but not with C5 or C5a. In immunoblot analysis it reacts with the alpha- but not with beta-chain of C3 and binds to a protein with a mol. wt of about 10 kD present in zymosan-activated sera which is only marginally detectable in nonactivated serum and absent in plasma. H13 crossreacts with the analogous proteins of rabbit, guinea pig and sheep. H13 has the capacity to bind 125I-radiolabelled C3a efficiently but fails totally to react with 125I-C5a or with other C3 alpha-chain fragments. H13 blocks C3a functional activity. It markedly inhibits C3a-induced 3H-serotonin release from platelets in vitro and similarly inhibits the C3a-induced extravasation of Evans blue into the skin in vivo. H13 does not interfere with the haemolytic activity of C3. An ELISA system was established using H13 which permits quantification of C3a in sera of polytrauma patients. The antibody H13 should facilitate further functional analysis of C3a in experimental systems. It should be useful for quantification of C3a in diagnostic assays and also for application in immunopathology.

Augmentation of hemolytic efficiency of anti-hapten IgG antibody by anti-IgG antibody is primarily the function of the anti-hapten IgG

IgG anti-cell antibodies are inefficient in inducing cell lysis by complement. C mediated lysis by anti-cell IgG can be augmented by the use of C fixing anti-antibody. We have studied augmentation of rabbit anti-Forssman IgG, rabbit anti-methotrexate IgG and rabbit anti-folinic acid IgG antibodies by rabbit anti-allotype IgG antibodies. Large variability (from a low of 2 to a high of over 100) was found in augmentation efficiency even with the same anti-allotype antibody; the variability was primarily the function of the anti-hapten antibody. It was concluded that quantitative studies on anti-hapten IgG production relying on augmented hemolysis may lead to misleading conclusions.

Different in vitro generation of C3d by cuprammonium and polycarbonate capillary hemodialyzers

Cellulosic membranes have been shown to activate C3 via the alternative pathway. Generation of the stable product C3d was compared in polycarbonate (PC) and cuprammonium (CU) hollow-fiber minidialyzers with identical geometry, using an in vitro recirculation system with platelet-poor plasma (recirculation rate 20 ml/min; 1,600 fibers; 0.115 m2; 40 ml system priming volume; recirculation time 120 min). After extensive washing, residual protein was removed by reverse transmembrane pressure (1.5 bar) from the air-filled dialysate compartment. Such protein fraction, functionally defined as 'secondary membrane', was similar in quantity for PC and CU fibers. Plasma proteins were present in the membrane without selective adsorption (flat-bed polyacrylamide gel electrophoresis and rocket immunoelectrophoresis using monospecific antibodies). C3d, measured by immunoelectrophoresis, increased with time in the fluid phase compartment of both dialysers, but C3d generation was consistently and significantly (p less than 0.05) less in PC than in CU dialysers. The observation documents an activation of C3 with PC, in contrast to previous reports of no activation. However, C3 activation with PC was less intense than with CU.