Eur. J. Immunol. 1990.20: 1669-1675 HClene Dauchel, Nathalie Julen, Claudie Lemercier, Maryvonne Daveau, Denyse Ozanne, Marc Fontaine and Jean Ripoche INSERM U-78, Bois-Guillaume
IL 1 and dexamethasone regulate C secretion by HUVEC
Expression of complement alternative pathway proteins by endothelial cells. Differential regulation by interleukin 1 and glucocorticoids* We have studied the secretion of proteins of the alternative pathway of complement C3, factor B and factor H by human umbilical vein endothelial cells (HUVEC). Results showed that factor H and factor B are quantitatively secreted in abundance whereas C3 could only be detected when the cells are maintained in culture during long periods of time. Interferon-y stimulated factor H, factor B and, to a lesser extent, C3 secretions. Interleukin (IL) 1had a differential effect on spontaneous C3, factor B and factor H secretions. In the presence of IL 1, there was a significant secretion of C3 occurring within a short period of culture. IL 1also stimulated factor B secretion.There was a synergistic stimulating effect between IL 1and interferon-y to bring C3 and factor B productions by HUVEC to very high levels. In contrast, factor H secretion was consistently inhibited by IL 1. Local increase in C3 and factor B secretions by endothelial cells in the presence of IL 1may have important implications in the inflammatory reaction. In striking contrast, the glucocorticoid dexamethasone (DXM) had modulatory effects which are consistent with its anti-inflammatory properties. DXM, at therapeutic concentrations, decreased C3 and factor B secretions and increased factor H secretion. Local modulation of complement protein secretion by DXM appears to be a new mechanism by which this glucocorticoid may control inflammation.
1 Introduction Endothelial cells are known to play an active role in hemostasis and the inflammatory reaction. In these processes, the activation of the endothelium by cytokines and the subsequent expression of specific membrane-bound and/or soluble proteins, appear t o be a crucial step (for review see [ l , 21) The secretion of complement proteins by endothelial cells has recently been a focus of interest. Human umbilical vein endothelial cells (HUVEC) have been shown to express proteins of the alternative pathway of complement, C3, factor B, and the regulatory proteins factor H and factor I [3-61. These findings would suggest that these cells can assemble the complete alternative pathway of complement. Little is known about the regulation of this secretion. The modulation of the secretion of complement components by HUVEC in inflammation may have important implications in the physiopathology of the inflammatory reaction. For instance, alternative complement protein secretion by HUVEC may take an active part in the local deposition of iC3b on the endothelial cells, which in turn could mediate the CDllb-dependent adhesion of myelomonocytic cells to the endothelium [7]. In this report, we have quantitated the secretion of the proteins of the alternative pathway of complement C3 and factor B and the regulatory protein factor H by HUVEC and studied the regulation of this expression. Our results
[I 82291
*
1669
Supported by INSERM, Rouen University and the Fondation pour la Recherche Medicale.
Correspondence:HCliine Dauchel, INSERM U-78, BP 73, F-76233 Bois-Guillaume Cedex, France Abbreviations: HUVEC: Human umbilical vein endothelial cells DXM: Dexamethasone 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
show that IL la and the synthetic glucocorticoid dexamethasone (DXM) have profound and opposite modulatory effect on C3, factor B and factor H secretion by HUVEC. IL 1 favors the secretion of proteins acting positively in complement activation C3 and factor B and has a suppressive effect on the secretion of the control protein factor H. This is a new mechanism by which IL 1 can influence inflammation via the endothelium. On the contrary, the modulatory effects of DXM are acting by down-regulating complement activation, indicating that DXM may exert a control over inflammation by regulating complement protein secretion.
2 Materials and methods 2.1 Endothelial cells Primary cultures of HUVEC were obtained from freshly collected umbilical cords. Collagenase (0.1% w/v; Boehringer Mannheim, Meylan, France) was used for treatment of the umbilical vein instead of 0.2% as originally described [S]. Cells were grown at 37°C in a-MEM containing 15% heat-inactivated FCS, 20 pg/ml endothelial cell supplement (No E9005, Sigma via Eurobio, Paris, France), 2 m M L-glutamine, 90 pg/ml heparin (Sigma), 2.5 pg/ml fungizone, 50 pg/ml penicillin, 50 pg/ml streptomycin (this medium being designated as standard medium). Cell culture reagents were from Gibco (Gibco/BFU, CergyPontoise, France) unless otherwise stated. Cells were identified by their characteristic morphology and by the expression of factor VIII antigen [8]. They were grown to confluence in 75- or 25-cm2 flasks (Nunc, Kamstrup, Denmark) coated with 1% gelatin. Subculturingof the cells was done by treating the cell monolayer with trypsin-EDTA and cells were routinely used in the second through fourth passage for experimental work. After the second passage, there is no visible contamination by monocytes. 0014-2980/90/0808-1669$3.50+ .25/0
1670
Eur. J. Immunol. 1990. 20: 1669-1675
H. Dauchel, N. Julen, C. Lemercier et al.
incubated overnight at 37°C. After four washes with PBS-Tween bound proteins were detected by incubation Human rIFN-y was a gift from Roussel Uclaf, Paris, France. with the peroxidase-conjugated appropriate antibody for IL la was human rIL-la purified from E. coli, a gift from 5 h at 37 "C. After four washes with PBS-Tween, 100 pl of Dr. PT. Lomedico, Hoffman-La Roche Inc., Nutley, NJ. A the peroxidase substrate 2,2-azino-di-(3-ethylbenzthiazolsecond human rIL-la preparation, purified from E. coli, ine sulfonic acid; Boehringer Mannheim) in citrate buffer used in control experiments, and rabbit polyclonal anti- (0.2 M trisodium citrate, 0.2 M citric acid, pH 4.5) containIL la antibodies were a generous gift from Dr. c . Rordorf ing 1%0 (v/v) H202 was added to each well. Measurement of (Ciba-Geigy, Basel, Switzerland). DXM, sodium phos- the absorbance at 405 nm was performed after a further 1-h phate salt (sterile, apyrogenic solution for human thera- incubation at 37°C using an automatic ELISA reader peutic use) was from Merck, Sharp and Dohme-Chibret (Flow, Puteaux, France). For each assay, a standard curve was established using dilutions of a pool of 200 normal Lab., Paris, France. human sera and a known amount of the appropriate purified protein. Concentrations were determined from the 2.3 Proteins and antisera standard curves. Significance of difference between treated cells and control was measured by paired Student's Factor H and C3 were purified from human plasma as t-test. described previously [9]. An additional purification step was performed for factor H by HPLC on a Spherogel TSK column (Phenyl5TW, Beckman, Palo Alto, CA). Factor B 2.6 Northern blot analysis was purified from human plasma as described [lo]. Polyspecific polyclonal anti-factor H and anti-C3 antibodies Total cellular RNA was extracted from HUVEC monowere obtained by immunization of rabbits with the respec- layers by lysis with guanidinium thiocyanate and included a tive purified human proteins. Monospecific anti-factor H cesium chloride density gradient ultracentrifugation [ 121. and anti-C3 polyclonal antibodies were obtained by chro- RNA was denatured using formamide and formaldehyde matography of the corresponding polyspecific polyclonal and electrophoresis was done on gels containing formaldeantibodies on factor H- and C3-Sepharose, respectively. hyde. Double-stranded cDNA probes were as follows: Monospecificanti-factor B polyclonal antibodies were from B38.1 encoding the first N-terminal third of factor H and Atlantic Antibodies, Scarborough, ME. Antibodies were recognizing, in addition to the 4.3-kb full length message, a conjugated with peroxidase (Boehringer Mannheim) by a 1.8-kb message encoding a truncated form of factor H containing the N-terminal region of the molecule [13] and one-step procedure as described [111. pC3.11 encoding the C-termina1(90%) of human C3 [14], a gift from Georg Fey (Research Institute of Scripps Clinic, La Jolla, CA). Factor B probe pFB-3b encoding the near 2.4 Experimental procedures complete sequence of the Ba fragment of factor B, was a gift Confluent HUVEC monolayers were incubated in standard from Duncan Campbell (MRC Immunochemistry Unit, medium containing the appropriate stimuli. Working stimu- Oxford, GB). Probes were labeled with 32Pby random li concentrations were 50 IU/ml for IL l a , 200 lU/ml for priming (Boehringer Mannheim). Autoradiographs were IFN-y and 1 O P 6 ~for DXM. For the study of long-term scanned using a Gelman DCD16 densitometer Aubervillsecretion of complement proteins by HUVEC, the medium ers, France. was decanted every 48 h and replaced by fresh standard medium containing the indicated stimuli. Replacement of 2.7 IL 1 assays the medium at regular intervals is necessary for long-term culture studies, as the cells tend to detach when they have IL 1 activity was measured in a co-stimulatory assay using being maintained for > 72 h without changing the medium. murine thymocytes as described previously [15]. In brief, At indicated time intervals, the SN was decanted, supple- thymocytes (from six-week-old C3H/HeJ mice) were culmented with proteinase inhibitors [5mM aprotinin; Choay- tured in 96-well round-bottom microplate (10Vwell) for Chimie, Notre Dame de Bondeville, France), 50 pg/ml 72 h in RPMI containing 5% FCS, antibiotics, 2.5 x lop5M soybean trypsin inhibitor (Worthington, Millipore Corp., 2-ME and 1 pg/ml PHA. Cultures in triplicate were pulsed NJ), 1 pg/ml pepstatin A (Sigma), 2 mM PMSF (Sigma)], for 16 h with 1 pCi/well = 37 kBq/well of [3H]dThd final concentrations, and frozen immediately at - 70°C (Amersham Int., Amersham, GB). The incorporation of until being assayed. Cells were observed under microscope [3H]dThd in the cell SN was compared with that in the and counted after trypsinization. In some experiments, the presence of human rlL-1 (Hoffman-La-Roche) as a stancells were cultured in the presence of polymyxin B (Pfizer dard. IL 1 was also measured using an RIA kit (Medgenix, Lab., Orsay, France) at a final concentration of 2 yg/ml. Fleurus, Belgium).This assay is based upon the principle of
2.2 Cytokines preparations and other reagents
2.5 ELISA Assays were performed in 96-well microtiter plates (Falcon, Becton Dickinson, France). Wells were coated 5 h at 37 "C with the appropriate polyclonal antibody (0.5 to 1.5 pg/well in 0.1 M NaHC03, pH 9.0, buffer). After one wash with PBS-Tween buffer [PBS containing 0.17% (w/v) Tween-201 and aspiration to remove excess of fluid, 100 p1 of the sample to be tested was added to the wells and plates were
a competition between radiolabeled IL 1and the IL 1to be tested for a specific antibody.The kit was used according to the manufacturer's instructions.
3 Results 3.1 Secretion of factor H, factor B and C3 by HUVEC
Since HUVEC are known to express factor H , factor B and C3 messages, we first attempted to demonstrate and
Eur. J. Immunol. 1990. 20: 1669-1675
IL 1 and dexamethasone regulate C secretion by HUVEC
1671
quantitate the secretion of these proteins in HUVEC SN. This was done by ELISA, as described in Sect. 2.5, in the SN of 48-h and 72-h stimulated cells. Results (Fig. 1) show that HUVEC spontaneously secreted factor H, factor B and C3. Factor H and factor B are readily detectable after 48 h in the SN of HUVEC. IFN-y significantlyincreased the secretion of factor B and factor H (p < 0.001). There was a good correlation between the increase in specifictranscripts (Fig. 2) and the increase in protein secretion indicatingthat IFN-y is acting at a pretranslational level to increase factor B and factor H secretion. Both factor H mRNA species were increased to a similar extent by IFW-y. In contrast to factor H and factor B, C3 was quantitatively a minor secretion product under basal conditions. Indeed, after 48 h of culture, C3 concentrations in HUVEC SN were below the lower level of assay detection (5 ng/ml). C3 could only be consistently demonstrated after 72 h of culture and still was in very low concentration (20 ng/106 cells; Fig. 1). On a molar basis, there was 6- to 7-fold more factor H than C3 in HUVEC SN after 72 h of culture. C3 secretion was increased approximately 3-fold by IFN-y after 72 h of stimulation (p cO.05). This increase in C3 secretion correlated with a parallel increase in C3 specific messenger (Fig. 2).This stimulatory effect of IFN-y on C3, factor B and factor H secretions was also effective in the presence of polymyxin B, indicating that this effect was not related to a endotoxin contamination of the rIFN-y preparation.
Figure 2. Differential modulation of factor H, factor B and C3 by I L 1 and DXM: Northern blot analysis. Confluent HUVEC monolayers were stimulated during 72 h with the indicated stimuli and total RNA prepared as described in Sect. 2.6. Total FZNA, 15 pg for each track, was run under denaturing conditions and blotted onto nitrocellulose. The filter was hybridized with factor H-specific cDNA probe B38-1 (A), factor B-specific cDNA probe pFB3b (B), C3-specifkcDNA probepC3.11 (C). (Lane 1)control HUVEC; (lane 2) IFN-y-stimulated HUVEC (200 IU/ml); (lane 3) DXM-stimulated HUVEC M); (lane 4) IL lastimulated HUVEC (50 IU/ml); (lane 5) DXM plus IFN-ystimulated HUVEC (loT6M and 200 IU/ml, respectively); (lane 6) IL la plus EN-y-stimulated HUVEC (50 IU/ml and 200 IU/ml, respectively).The minor band seen in (C) is a rest of signal from the previous probing with factor H-specific cDNA.
3.2 Differential modulation of the secretion of C3, factor B and factor H by IL l a
IL la had a dramatic stimulatory effect on C3 secretion and a reproducible stimulatory effect on factor B secretion (Fig. l).Thestimulatoryeffect of IL la on C3 secretion was very significant (p c 0.02), the secretion of C3 by HUVEC being multiplied by more than 7-fold in the presence of IL la after 72 h of culture. In the presence of IL la, C3 could be readily detected after 48 h of culture, in contrast to control cells. As for C3, IL la induced an approximately 2-fold increase in factor B secretion after 48 h of culture (p
IL 1 and dexamethasone regulate C secretion by HUVEC
Expression of complement alternative pathway proteins by endothelial cells. Differential regulation by interleukin 1 and glucocorticoids* We have studied the secretion of proteins of the alternative pathway of complement C3, factor B and factor H by human umbilical vein endothelial cells (HUVEC). Results showed that factor H and factor B are quantitatively secreted in abundance whereas C3 could only be detected when the cells are maintained in culture during long periods of time. Interferon-y stimulated factor H, factor B and, to a lesser extent, C3 secretions. Interleukin (IL) 1had a differential effect on spontaneous C3, factor B and factor H secretions. In the presence of IL 1, there was a significant secretion of C3 occurring within a short period of culture. IL 1also stimulated factor B secretion.There was a synergistic stimulating effect between IL 1and interferon-y to bring C3 and factor B productions by HUVEC to very high levels. In contrast, factor H secretion was consistently inhibited by IL 1. Local increase in C3 and factor B secretions by endothelial cells in the presence of IL 1may have important implications in the inflammatory reaction. In striking contrast, the glucocorticoid dexamethasone (DXM) had modulatory effects which are consistent with its anti-inflammatory properties. DXM, at therapeutic concentrations, decreased C3 and factor B secretions and increased factor H secretion. Local modulation of complement protein secretion by DXM appears to be a new mechanism by which this glucocorticoid may control inflammation.
1 Introduction Endothelial cells are known to play an active role in hemostasis and the inflammatory reaction. In these processes, the activation of the endothelium by cytokines and the subsequent expression of specific membrane-bound and/or soluble proteins, appear t o be a crucial step (for review see [ l , 21) The secretion of complement proteins by endothelial cells has recently been a focus of interest. Human umbilical vein endothelial cells (HUVEC) have been shown to express proteins of the alternative pathway of complement, C3, factor B, and the regulatory proteins factor H and factor I [3-61. These findings would suggest that these cells can assemble the complete alternative pathway of complement. Little is known about the regulation of this secretion. The modulation of the secretion of complement components by HUVEC in inflammation may have important implications in the physiopathology of the inflammatory reaction. For instance, alternative complement protein secretion by HUVEC may take an active part in the local deposition of iC3b on the endothelial cells, which in turn could mediate the CDllb-dependent adhesion of myelomonocytic cells to the endothelium [7]. In this report, we have quantitated the secretion of the proteins of the alternative pathway of complement C3 and factor B and the regulatory protein factor H by HUVEC and studied the regulation of this expression. Our results
[I 82291
*
1669
Supported by INSERM, Rouen University and the Fondation pour la Recherche Medicale.
Correspondence:HCliine Dauchel, INSERM U-78, BP 73, F-76233 Bois-Guillaume Cedex, France Abbreviations: HUVEC: Human umbilical vein endothelial cells DXM: Dexamethasone 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
show that IL la and the synthetic glucocorticoid dexamethasone (DXM) have profound and opposite modulatory effect on C3, factor B and factor H secretion by HUVEC. IL 1 favors the secretion of proteins acting positively in complement activation C3 and factor B and has a suppressive effect on the secretion of the control protein factor H. This is a new mechanism by which IL 1 can influence inflammation via the endothelium. On the contrary, the modulatory effects of DXM are acting by down-regulating complement activation, indicating that DXM may exert a control over inflammation by regulating complement protein secretion.
2 Materials and methods 2.1 Endothelial cells Primary cultures of HUVEC were obtained from freshly collected umbilical cords. Collagenase (0.1% w/v; Boehringer Mannheim, Meylan, France) was used for treatment of the umbilical vein instead of 0.2% as originally described [S]. Cells were grown at 37°C in a-MEM containing 15% heat-inactivated FCS, 20 pg/ml endothelial cell supplement (No E9005, Sigma via Eurobio, Paris, France), 2 m M L-glutamine, 90 pg/ml heparin (Sigma), 2.5 pg/ml fungizone, 50 pg/ml penicillin, 50 pg/ml streptomycin (this medium being designated as standard medium). Cell culture reagents were from Gibco (Gibco/BFU, CergyPontoise, France) unless otherwise stated. Cells were identified by their characteristic morphology and by the expression of factor VIII antigen [8]. They were grown to confluence in 75- or 25-cm2 flasks (Nunc, Kamstrup, Denmark) coated with 1% gelatin. Subculturingof the cells was done by treating the cell monolayer with trypsin-EDTA and cells were routinely used in the second through fourth passage for experimental work. After the second passage, there is no visible contamination by monocytes. 0014-2980/90/0808-1669$3.50+ .25/0
1670
Eur. J. Immunol. 1990. 20: 1669-1675
H. Dauchel, N. Julen, C. Lemercier et al.
incubated overnight at 37°C. After four washes with PBS-Tween bound proteins were detected by incubation Human rIFN-y was a gift from Roussel Uclaf, Paris, France. with the peroxidase-conjugated appropriate antibody for IL la was human rIL-la purified from E. coli, a gift from 5 h at 37 "C. After four washes with PBS-Tween, 100 pl of Dr. PT. Lomedico, Hoffman-La Roche Inc., Nutley, NJ. A the peroxidase substrate 2,2-azino-di-(3-ethylbenzthiazolsecond human rIL-la preparation, purified from E. coli, ine sulfonic acid; Boehringer Mannheim) in citrate buffer used in control experiments, and rabbit polyclonal anti- (0.2 M trisodium citrate, 0.2 M citric acid, pH 4.5) containIL la antibodies were a generous gift from Dr. c . Rordorf ing 1%0 (v/v) H202 was added to each well. Measurement of (Ciba-Geigy, Basel, Switzerland). DXM, sodium phos- the absorbance at 405 nm was performed after a further 1-h phate salt (sterile, apyrogenic solution for human thera- incubation at 37°C using an automatic ELISA reader peutic use) was from Merck, Sharp and Dohme-Chibret (Flow, Puteaux, France). For each assay, a standard curve was established using dilutions of a pool of 200 normal Lab., Paris, France. human sera and a known amount of the appropriate purified protein. Concentrations were determined from the 2.3 Proteins and antisera standard curves. Significance of difference between treated cells and control was measured by paired Student's Factor H and C3 were purified from human plasma as t-test. described previously [9]. An additional purification step was performed for factor H by HPLC on a Spherogel TSK column (Phenyl5TW, Beckman, Palo Alto, CA). Factor B 2.6 Northern blot analysis was purified from human plasma as described [lo]. Polyspecific polyclonal anti-factor H and anti-C3 antibodies Total cellular RNA was extracted from HUVEC monowere obtained by immunization of rabbits with the respec- layers by lysis with guanidinium thiocyanate and included a tive purified human proteins. Monospecific anti-factor H cesium chloride density gradient ultracentrifugation [ 121. and anti-C3 polyclonal antibodies were obtained by chro- RNA was denatured using formamide and formaldehyde matography of the corresponding polyspecific polyclonal and electrophoresis was done on gels containing formaldeantibodies on factor H- and C3-Sepharose, respectively. hyde. Double-stranded cDNA probes were as follows: Monospecificanti-factor B polyclonal antibodies were from B38.1 encoding the first N-terminal third of factor H and Atlantic Antibodies, Scarborough, ME. Antibodies were recognizing, in addition to the 4.3-kb full length message, a conjugated with peroxidase (Boehringer Mannheim) by a 1.8-kb message encoding a truncated form of factor H containing the N-terminal region of the molecule [13] and one-step procedure as described [111. pC3.11 encoding the C-termina1(90%) of human C3 [14], a gift from Georg Fey (Research Institute of Scripps Clinic, La Jolla, CA). Factor B probe pFB-3b encoding the near 2.4 Experimental procedures complete sequence of the Ba fragment of factor B, was a gift Confluent HUVEC monolayers were incubated in standard from Duncan Campbell (MRC Immunochemistry Unit, medium containing the appropriate stimuli. Working stimu- Oxford, GB). Probes were labeled with 32Pby random li concentrations were 50 IU/ml for IL l a , 200 lU/ml for priming (Boehringer Mannheim). Autoradiographs were IFN-y and 1 O P 6 ~for DXM. For the study of long-term scanned using a Gelman DCD16 densitometer Aubervillsecretion of complement proteins by HUVEC, the medium ers, France. was decanted every 48 h and replaced by fresh standard medium containing the indicated stimuli. Replacement of 2.7 IL 1 assays the medium at regular intervals is necessary for long-term culture studies, as the cells tend to detach when they have IL 1 activity was measured in a co-stimulatory assay using being maintained for > 72 h without changing the medium. murine thymocytes as described previously [15]. In brief, At indicated time intervals, the SN was decanted, supple- thymocytes (from six-week-old C3H/HeJ mice) were culmented with proteinase inhibitors [5mM aprotinin; Choay- tured in 96-well round-bottom microplate (10Vwell) for Chimie, Notre Dame de Bondeville, France), 50 pg/ml 72 h in RPMI containing 5% FCS, antibiotics, 2.5 x lop5M soybean trypsin inhibitor (Worthington, Millipore Corp., 2-ME and 1 pg/ml PHA. Cultures in triplicate were pulsed NJ), 1 pg/ml pepstatin A (Sigma), 2 mM PMSF (Sigma)], for 16 h with 1 pCi/well = 37 kBq/well of [3H]dThd final concentrations, and frozen immediately at - 70°C (Amersham Int., Amersham, GB). The incorporation of until being assayed. Cells were observed under microscope [3H]dThd in the cell SN was compared with that in the and counted after trypsinization. In some experiments, the presence of human rlL-1 (Hoffman-La-Roche) as a stancells were cultured in the presence of polymyxin B (Pfizer dard. IL 1 was also measured using an RIA kit (Medgenix, Lab., Orsay, France) at a final concentration of 2 yg/ml. Fleurus, Belgium).This assay is based upon the principle of
2.2 Cytokines preparations and other reagents
2.5 ELISA Assays were performed in 96-well microtiter plates (Falcon, Becton Dickinson, France). Wells were coated 5 h at 37 "C with the appropriate polyclonal antibody (0.5 to 1.5 pg/well in 0.1 M NaHC03, pH 9.0, buffer). After one wash with PBS-Tween buffer [PBS containing 0.17% (w/v) Tween-201 and aspiration to remove excess of fluid, 100 p1 of the sample to be tested was added to the wells and plates were
a competition between radiolabeled IL 1and the IL 1to be tested for a specific antibody.The kit was used according to the manufacturer's instructions.
3 Results 3.1 Secretion of factor H, factor B and C3 by HUVEC
Since HUVEC are known to express factor H , factor B and C3 messages, we first attempted to demonstrate and
Eur. J. Immunol. 1990. 20: 1669-1675
IL 1 and dexamethasone regulate C secretion by HUVEC
1671
quantitate the secretion of these proteins in HUVEC SN. This was done by ELISA, as described in Sect. 2.5, in the SN of 48-h and 72-h stimulated cells. Results (Fig. 1) show that HUVEC spontaneously secreted factor H, factor B and C3. Factor H and factor B are readily detectable after 48 h in the SN of HUVEC. IFN-y significantlyincreased the secretion of factor B and factor H (p < 0.001). There was a good correlation between the increase in specifictranscripts (Fig. 2) and the increase in protein secretion indicatingthat IFN-y is acting at a pretranslational level to increase factor B and factor H secretion. Both factor H mRNA species were increased to a similar extent by IFW-y. In contrast to factor H and factor B, C3 was quantitatively a minor secretion product under basal conditions. Indeed, after 48 h of culture, C3 concentrations in HUVEC SN were below the lower level of assay detection (5 ng/ml). C3 could only be consistently demonstrated after 72 h of culture and still was in very low concentration (20 ng/106 cells; Fig. 1). On a molar basis, there was 6- to 7-fold more factor H than C3 in HUVEC SN after 72 h of culture. C3 secretion was increased approximately 3-fold by IFN-y after 72 h of stimulation (p cO.05). This increase in C3 secretion correlated with a parallel increase in C3 specific messenger (Fig. 2).This stimulatory effect of IFN-y on C3, factor B and factor H secretions was also effective in the presence of polymyxin B, indicating that this effect was not related to a endotoxin contamination of the rIFN-y preparation.
Figure 2. Differential modulation of factor H, factor B and C3 by I L 1 and DXM: Northern blot analysis. Confluent HUVEC monolayers were stimulated during 72 h with the indicated stimuli and total RNA prepared as described in Sect. 2.6. Total FZNA, 15 pg for each track, was run under denaturing conditions and blotted onto nitrocellulose. The filter was hybridized with factor H-specific cDNA probe B38-1 (A), factor B-specific cDNA probe pFB3b (B), C3-specifkcDNA probepC3.11 (C). (Lane 1)control HUVEC; (lane 2) IFN-y-stimulated HUVEC (200 IU/ml); (lane 3) DXM-stimulated HUVEC M); (lane 4) IL lastimulated HUVEC (50 IU/ml); (lane 5) DXM plus IFN-ystimulated HUVEC (loT6M and 200 IU/ml, respectively); (lane 6) IL la plus EN-y-stimulated HUVEC (50 IU/ml and 200 IU/ml, respectively).The minor band seen in (C) is a rest of signal from the previous probing with factor H-specific cDNA.
3.2 Differential modulation of the secretion of C3, factor B and factor H by IL l a
IL la had a dramatic stimulatory effect on C3 secretion and a reproducible stimulatory effect on factor B secretion (Fig. l).Thestimulatoryeffect of IL la on C3 secretion was very significant (p c 0.02), the secretion of C3 by HUVEC being multiplied by more than 7-fold in the presence of IL la after 72 h of culture. In the presence of IL la, C3 could be readily detected after 48 h of culture, in contrast to control cells. As for C3, IL la induced an approximately 2-fold increase in factor B secretion after 48 h of culture (p
