Inflammation, Vol. 16, No. 3, 1992
R E C O M B I N A N T H U M A N I N T E R L E U K I N l fl AND TUMOR NECROSIS FACTOR AFFECT GLYCOSYLATION OF SERUM oq-ACID GLYCOPROTEIN
IN R A T S
CHRISTIAN POOS, ~ LAURENCE CHAUVELOT-MOACHON, 2 MARYVONNE LECOUSTILLIER,~ and GENEVIEVE DURAND j ~D~partement de Biochimie G~ndrale UFR des Sciences Pharmaceutiques et Biologiques (Universitd Paris X1) 5, rue J.B. Cldment 92296 Ch~tenay-Malabry Cddex, France 2D~partement de Pharmacologie CNRS URA 595 H6pital Cochin 27 rue du Faubourg St Jacques 75674 Paris Cddex 14, France
Abstract--Serum concentration and glycosylation of rat a~-acid glycoprotein (oqAGP) were evaluated after the in vivo administration of recombinant human interleukin-lfl (rhlL-1/3) anti tumor necrosis factor u (rhTNF-ce), alone or associated. The effect of LPS and turpentine was also studied. In all models, serum a~-AGP concentrations were increased and glycosylation was altered. The c~I-AGP levels reached 1.8 g/liter with cytokines alone, 2.1 g/liter with cytokines associated or LPS, and 3.4 g/liter with turpent ne. Analysis by concanavalin A (Con A) affinoimmunoelectrophoresis (CAIE) revealed that the relative proportion of Con A unreactive form always decreased whatever the inducing agent. On the other hand, the resulting effect on the concentrations cf Con A unreactive oq-mGP concentrations was an increase with cytokines alone or LPS and a decrease with cytokines associated or turpentine. These results suggest a dissociation between the alteration in the level of oq-AGP synthesis and in the pattern of its glycosylation in the various inflammatory models.
INTRODUCTION The acute-phase response involves a series of events including major changes in liver functions. The hepatic biosynthesis of some proteins quickly rises with 197 0360-3997/92/0600-0197506.50/09 1992PlenumPublishingCorporation
198
Pofis et al.
a subsequent increase in their plasma concentrations. It has been previously s h o w n that the increase in serum levels o f different glycoproteins during experimental i n f l a m m a t i o n such as turpentine injection or laparotomy is associated with a modification o f their glycosylation (1). It is n o w k n o w n that at least a part o f the acute-phase response is mediated by inflammatory cytokines (2-5) and glucocorticoids (6). T h e m a i n cytokines i n v o l v e d are interleukin-1 (IL-1), t u m o r necrosis factor ( T N F ) , and interleukin-6 (IL-6). A m o n g the acute-phase proteins (APPs), ~xl-acid glycoprotein ( c q - A G P ) , which possesses an important g l y c a n moiety, is well suited to study some aspects of the posttranslational maturation o f glycoproteins. It was previously shown that o q - A G P is one of the A P P s that are more sensitive to IL-1 and T N F than to IL-6 in vitro (7) and that IL-6 induces only a weak increase in serum o q - A G P concentrations in vivo (8). H o w e v e r , there are no data available c o n c e r n i n g the effects o f IL-1 and T N F o n the glycosylation o f rat e q - A G P in vivo. T h e aim o f this study was to determine the effect of IL-1 and T N F on serum c q - A G P glycosylation by c o m p a r i s o n with the effects o f turpentine and lipopolysaccharide (LPS) after their administration to rats.
MATERIALS
AND METHODS
rhlL-l/3 and rhTNF-a were generous gifts from Laboratoires Roussel Uclaf (Romainville, France) and Knoll AG/BASF (Ludwigshafen, Germany), respectively; their contamination with bacterial endotoxin was less than 0.625 pg/#g protein. RPMI 1640 medium supplemented with 2 mM glutamine was purchased from Gibco. Escherichia coli LPS (055 : B5) was from Wellcome Research Laboratories (Beckenham, Kent, England). Con A was obtained from l'Industrie Biologique Frangaise (Villeneuve-la-Garenne, France). Treatment of Animals. Male Fisher 344 rats (8-10 weeks old, 190-220 g) (CNRS, Villejuif, France) were used in all experiments. The cytokines and LPS were diluted in RPMI medium. The 1-ml intravenous injections were performed in the caudal vein 24 h prior to sacrifice by decapitation except for the highest dose of LPS (2.5 mg/kg), which was injected intraperitoneally. This time corresponds to the approximate peak of serum ~xrAGP (9). Turpentine was injected subcutaneously (5 ml/kg), and blood was collected 48 h later. Serum samples were kept at -80~ until assay. c~-AGP Concentration Measurement. Serum a~-AGP was quantified by rocket immunoelectrophoresis according to Laurell (10) in a 1% (w/v) agarose, 4 % (w/v) polyethylene glycol 6000 gel containing 0.3 % (v/v) of rabbit anti-rat oq-AGP antiserum (11). Con A CAIE Patterns of Serum arAGP. Con A -CAIE of mt serum was performed according to Monnet et al. (12) using 1.25 mg/ml of Con A in the first dimension and 1.2% of anti-rat as-AGP antiserum in the second dimension. Peaks were integrated using a planimeter (A. Ott Kemplen Bayern). The peak areas were expressed in two ways: (1) as the ratio of Con A-unreactive cxrAGP (the fraction of Con A-unreactive oq-AGP was divided by the sum of the four fractious), and (2) as the absolute concentration of serum arAGP corresponding to each peak. Statistics. All the results are expressed as the mean _.+_SE of four rats per group. Comparisons were performed using Student's t test.
rhlL-1, TNF, and ~q-AGP
199
RESULTS
Effects of Cytokines, LPS, and Turpentine on Serum ~z-AGP Concentrations. The administration of IL-1 and TNF alone or associated dose-dependently increased serum ~I-AGP concentrations (Figure 1). The maximal effect after cytokine administration (2.2 g/liter) was observed with the association of both cytokines and was closer to that observed with the highest dose of LPS (2.1 g/liter) than with turpentine, which induced serum cq-AGP up to 3.4 g/liter (Table 1). Effect of Cytokines, LPS, and Turpentine on Serum oq-AGP Con A Unreac2.5
2.0
1.5 8
: -
n
(5
R E C O M B I N A N T H U M A N I N T E R L E U K I N l fl AND TUMOR NECROSIS FACTOR AFFECT GLYCOSYLATION OF SERUM oq-ACID GLYCOPROTEIN
IN R A T S
CHRISTIAN POOS, ~ LAURENCE CHAUVELOT-MOACHON, 2 MARYVONNE LECOUSTILLIER,~ and GENEVIEVE DURAND j ~D~partement de Biochimie G~ndrale UFR des Sciences Pharmaceutiques et Biologiques (Universitd Paris X1) 5, rue J.B. Cldment 92296 Ch~tenay-Malabry Cddex, France 2D~partement de Pharmacologie CNRS URA 595 H6pital Cochin 27 rue du Faubourg St Jacques 75674 Paris Cddex 14, France
Abstract--Serum concentration and glycosylation of rat a~-acid glycoprotein (oqAGP) were evaluated after the in vivo administration of recombinant human interleukin-lfl (rhlL-1/3) anti tumor necrosis factor u (rhTNF-ce), alone or associated. The effect of LPS and turpentine was also studied. In all models, serum a~-AGP concentrations were increased and glycosylation was altered. The c~I-AGP levels reached 1.8 g/liter with cytokines alone, 2.1 g/liter with cytokines associated or LPS, and 3.4 g/liter with turpent ne. Analysis by concanavalin A (Con A) affinoimmunoelectrophoresis (CAIE) revealed that the relative proportion of Con A unreactive form always decreased whatever the inducing agent. On the other hand, the resulting effect on the concentrations cf Con A unreactive oq-mGP concentrations was an increase with cytokines alone or LPS and a decrease with cytokines associated or turpentine. These results suggest a dissociation between the alteration in the level of oq-AGP synthesis and in the pattern of its glycosylation in the various inflammatory models.
INTRODUCTION The acute-phase response involves a series of events including major changes in liver functions. The hepatic biosynthesis of some proteins quickly rises with 197 0360-3997/92/0600-0197506.50/09 1992PlenumPublishingCorporation
198
Pofis et al.
a subsequent increase in their plasma concentrations. It has been previously s h o w n that the increase in serum levels o f different glycoproteins during experimental i n f l a m m a t i o n such as turpentine injection or laparotomy is associated with a modification o f their glycosylation (1). It is n o w k n o w n that at least a part o f the acute-phase response is mediated by inflammatory cytokines (2-5) and glucocorticoids (6). T h e m a i n cytokines i n v o l v e d are interleukin-1 (IL-1), t u m o r necrosis factor ( T N F ) , and interleukin-6 (IL-6). A m o n g the acute-phase proteins (APPs), ~xl-acid glycoprotein ( c q - A G P ) , which possesses an important g l y c a n moiety, is well suited to study some aspects of the posttranslational maturation o f glycoproteins. It was previously shown that o q - A G P is one of the A P P s that are more sensitive to IL-1 and T N F than to IL-6 in vitro (7) and that IL-6 induces only a weak increase in serum o q - A G P concentrations in vivo (8). H o w e v e r , there are no data available c o n c e r n i n g the effects o f IL-1 and T N F o n the glycosylation o f rat e q - A G P in vivo. T h e aim o f this study was to determine the effect of IL-1 and T N F on serum c q - A G P glycosylation by c o m p a r i s o n with the effects o f turpentine and lipopolysaccharide (LPS) after their administration to rats.
MATERIALS
AND METHODS
rhlL-l/3 and rhTNF-a were generous gifts from Laboratoires Roussel Uclaf (Romainville, France) and Knoll AG/BASF (Ludwigshafen, Germany), respectively; their contamination with bacterial endotoxin was less than 0.625 pg/#g protein. RPMI 1640 medium supplemented with 2 mM glutamine was purchased from Gibco. Escherichia coli LPS (055 : B5) was from Wellcome Research Laboratories (Beckenham, Kent, England). Con A was obtained from l'Industrie Biologique Frangaise (Villeneuve-la-Garenne, France). Treatment of Animals. Male Fisher 344 rats (8-10 weeks old, 190-220 g) (CNRS, Villejuif, France) were used in all experiments. The cytokines and LPS were diluted in RPMI medium. The 1-ml intravenous injections were performed in the caudal vein 24 h prior to sacrifice by decapitation except for the highest dose of LPS (2.5 mg/kg), which was injected intraperitoneally. This time corresponds to the approximate peak of serum ~xrAGP (9). Turpentine was injected subcutaneously (5 ml/kg), and blood was collected 48 h later. Serum samples were kept at -80~ until assay. c~-AGP Concentration Measurement. Serum a~-AGP was quantified by rocket immunoelectrophoresis according to Laurell (10) in a 1% (w/v) agarose, 4 % (w/v) polyethylene glycol 6000 gel containing 0.3 % (v/v) of rabbit anti-rat oq-AGP antiserum (11). Con A CAIE Patterns of Serum arAGP. Con A -CAIE of mt serum was performed according to Monnet et al. (12) using 1.25 mg/ml of Con A in the first dimension and 1.2% of anti-rat as-AGP antiserum in the second dimension. Peaks were integrated using a planimeter (A. Ott Kemplen Bayern). The peak areas were expressed in two ways: (1) as the ratio of Con A-unreactive cxrAGP (the fraction of Con A-unreactive oq-AGP was divided by the sum of the four fractious), and (2) as the absolute concentration of serum arAGP corresponding to each peak. Statistics. All the results are expressed as the mean _.+_SE of four rats per group. Comparisons were performed using Student's t test.
rhlL-1, TNF, and ~q-AGP
199
RESULTS
Effects of Cytokines, LPS, and Turpentine on Serum ~z-AGP Concentrations. The administration of IL-1 and TNF alone or associated dose-dependently increased serum ~I-AGP concentrations (Figure 1). The maximal effect after cytokine administration (2.2 g/liter) was observed with the association of both cytokines and was closer to that observed with the highest dose of LPS (2.1 g/liter) than with turpentine, which induced serum cq-AGP up to 3.4 g/liter (Table 1). Effect of Cytokines, LPS, and Turpentine on Serum oq-AGP Con A Unreac2.5
2.0
1.5 8
: -
n
(5
