Clinical Science (1992) 82,439-446 (Printed in Great Britain)
439
Variations in the glycoforms of serum a ,-antichymotrypsin in liver diseases and after liver transplantation E. HACHULLA*, A. LAINE*, V. HEDOUINt, C. FOURNIERf, F. MAURYt, C. MATHlEUt, F. R. PRUVOT!, N. DECLERCK!, J. C. PARlSt and P. DEGAND* “Unit6 16 de I’lnstitut National de la SantC et de la Recherche MCdicale, Lille, France, Clinique des Maladies de I’Apporeil Digestif, CHU HBpital Claude Huriez, Lille, France, Slaboratoire de Pharmocodynamie Clinique, Centre Oscar Lambret, Lille, France, and SClinique Chirurgicale, CHU HBpital Calmette, Lille, France
t
(Received
4 Junel30 October 1991; accepted 13 November 1991)
1. Using crossed immunoafinity electrophoresis with
free concanavalin A in the first dimension, we studied the microheterogeneity of a,-antichymotrypsin due to various glycoforms in sera from patients with various liver diseases and after liver transplantation. 2. Studies by isoelectric focusing and immunoblotting and by crossed immunoelectrophoresis without concanavalin A in the first dimension allowed us to show that there is no dramatic variation in electrophoretic heterogeneity of a,-antichymotrypsin in the serum of patients with liver diseases or after liver transplantation when compared with that of normal subjects. Therefore the heterogeneity observed in crossed immunoaffinity electrophoresis is due to various interactions with concanavalin A. 3. The results were expressed as the ratio of concanavalin A non-reactive glycoforms plus concanavalin A weakly reactive glycoforms to concanavalin A reactive glycoforms, called R a ,-ACT. R a ,-ACT was significantly higher in patients with cirrhosis (n= 53) when compared with normal subjects (n= 30). The median R a ,-ACT was 1 (range 0.72-1.25) in normal subjects. It was 1.6 (range 1.18-3.02), 1.45 (range 0.65-4.12) and 2.24 (range 1.03-19) in cirrhosis of Child’s grade A, B and C, respectively. There was a dramatic decrease in glycoforms with mostly biantennary glycans in some patients with Child’s grade C cirrhosis. Serum levels of a ,-antichymotrypsin were lower than normal only in some patients with Child’s grade C cirrhosis. 4. Among the patients with acute viral hepatitis studied ( n = 17), five were studied longitudinally. Ra,-ACT increased after the peak level of serum transaminases and remained high while the level of serum transaminases decreased and liver regeneration occurred. In these patients there was no alteration in glycoforms with mostly biantennary glycans.
5. Four patients were studied after liver transplantation. During the first days after transplantation Ra,-ACT was low because of an increase in glycoforms with mostly biantennary glycans. After the first week, Ra ,-ACT rose above the normal range because of an increase in glycoforms with pluriantennary glycans and it returned to normal within 3 post-operative weeks. R a ,-ACT remained high in a patient who presented with liver rejection. 6. The present study suggests that Ra,-ACT is a helpful additional marker for the evaluation of hepatocyte function.
INTRODUCTION For most of the serum acute-phase reactants, the major, if not the only, site of synthesis is the liver, and specifically the hepatocyte [l]. Like C-reactive protein, a,-antichymotrypsin (a,-ACT) is one of the acute-phase reactants whose concentration increases within 12 h of an inflammatory process [l,21. a,-ACT, with an M , of about 59000 and a carbohydrate content of 24% [3], is known to inhibit chymotrypsin-like proteinases [4, 51. It belongs to the superfamily of serpins (serine proteinase inhibitors), a group of proteins very similar to each other in amino acid sequence but differing in biological function [6].Furthermore, a,-ACT plays a role in the modulation of the immune response [7] but does not modify natural killing cytotoxicity [S]. Many glycoproteins exhibit molecular heterogeneity, essentially due to different carbohydrate structures, and consist of various glycoforms. The variety of the glycoforms is easily demonstrated directly in the serum by crossed immunoaffinity electrophoresis (CIAE) with lectin [9]. In CIAE with free concanavalin A (Con A ) in the first dimension, human serum a,-ACT exhibits three
Key words: a,-antichymotrypsin, glycoforms, liver disease, liver transplantation, microheterogeneity. Abbreviations:a,-ACT, a,-antichymotrypsin; CIAE, crossed immunoaffinity electrophoresis: Con A, concanavalin A Ra,-ACT, (concanavalin A non-reactive glycoforms -t concanavalin A weakly reactive g1ycoforms)lconcanavalin A reactive glycoforms ratio. Correspondence: D r Eric Hachulla, Clinique Medicale A, (Pr B. Devulder), H6pital Claude Huriez, 59037 Lille Cedex, France.
440
E. Hachulla e t al.
Fig. I. CIAE of human serum with specific antiserum against a,-ACT. The figure shows typical serum samples from ( a ) a normal subject (Ra,-ACT 0.93, serum a,-ACT level 0.38 gll), (6) a patient with a Child’s grade A cirrhosis (Ra,-ACT 1.45, serum a,-ACT level 0.82 gll), (c) a patient with a Child’s grade C cirrhosis (Ra,-ACT 19, serum a,-ACT level 0.36 gil), (d) a patient on the first day after liver transplantation, who was previously a cirrhotic patient (Ra,-ACT 0.38, serum a,-ACT level 0.88 g/l), (e) the same patient 15 days after the transplantation (Ra,-ACT 1.68, serum a,-ACT level 1.86 gll) and (0 a patient with acute viral hepatitis, after the peak level of serum transaminases (Ra,-ACT 2.22, serum a,-ACT level 0.77 gll), !+, first dimension; 2t, second dimension.
peaks, or four peaks when sugar (displacer) is added to the gel of the second dimension [ l o , 111. The complete amino acid sequence of a,-ACT deduced from the nucleotide sequence [6] reveals four potential glycosylation sites. In a recent work [12], we determined the complete primary structure of the glycans of a,-ACT by high-resolution IH-n.m.r. spectroscopy. The results indicated the presence of disialyl biantennary and trisialyl triantennary type glycanic structures and traces of disialylated triantennary oligosaccharide. Significant variations in the proportions of the glycoforms separated by CIAE were detected in various inflammatory syndromes for a,-ACT as well as for other acute-phase glycoproteins [lo, 13-15]. In liver diseases like cirrhosis o r acute hepatitis, an increase in Con A non-reactive glycoforms was detected for a,-HS glycoprotein, a,-acid glycoprotein and serotransferrin [ 16-19]. In the present study we investigated the variations in the glycoforms of a,-ACT in various liver diseases to determine if these variations were correlated with hepatocellular insufficiency and thus could provide a functional parameter useful for predicting the ultimate outcome. To our knowledge, the variations in the proportions of the glycoforms of plasma glycoproteins which are
synthesized in the liver have not, until now, been studied before and after liver transplantation. This study would be of particular interest for evaluating this aspect of hepatocyte function.
MATERIALS AND METHODS Blood samples were obtained from 30 healthy subjects (13 men and 17 women, aged 27-80 years), none of which was receiving hormonal medication, from 53 patients with biopsy-proven cirrhosis (20 men and 33 women, aged 29-80 years: 44 patients with alcoholic cirrhosis, seven patients with cirrhosis due to hepatitis B virus and two patients with cirrhosis due to hepatitis C virus) and from 17 patients with acute viral hepatitis (six men and 11 women, aged 18-41 years). A longitudinal study was performed on the serum of four patients with cirrhosis who had a liver transplantation (3-4 weeks after the operation) and of five of the 17 patients with acute viral hepatitis; one of these five patients developed a fulminant hepatic failure with a spontaneous favourable evolution. For the patients with cirrhosis the state of liver function was graded by the modified Child’s scoring system [20], which is based on the grading (A, B, C ) of five clinical and
Serum a,-antichymotrypsin glycoforms
biological complications of the cirrhotic process (ascitis, encephalopathy, albuminaemia, bilirubinaemia, prothrombin time). At the time of Child’s scoring, blood samples were also drawn from the patients for measurement of serum a,-ACT and proaccelerin levels and for CIAE. Besides serum a ,-ACT, albumin and proaccelerin levels, serum transaminases (serum glutamic-pyruvic transaminase activity) were assessed every time a blood sample was taken from the patients who had undergone liver transplantation and from the patients suffering from acute viral hepatitis at the same time as CIAE was performed. Serum a,-ACT levels were determined by a single conventional radial immunodifhsion technique (Behringwerke). CIAE was performed using a modification of the method of B0g-Hansen et al. [9] as previously described [13].C U E was performed in thin-layer 1% (w/v) agarose gels (10 cm x 10 cm) using free Con A (routinely 250 pg/ cm’) placed in a 1 cm x 8 cm trap-gel in the first dimension. A bovine albumin solution prepared in concentrated Bromophenol Blue solution applied above the sample well was used as a coloured marker to follow the migration in the plate. The first dimension was run for 80 min at 10 V/cm, i.e. until the coloured bovine albumin had crossed the plate. The second dimension was run for 18 h at 2 V/cm in a 1% (w/v) agarose gel, poured onto the upper two-thirds of the plate, which contained 0.3 pl of specific antiserum/cm* against a,-ACT and a-methyl-Dglucoside (2%). Three microlitres of serum was applied for each analysis. a,-ACT separates into four peaks on CIAE numbered 1-4. We have previously shown that peak 1 is due to Con A non-reactive glycoforms containing four triantennary glycans, peak 2 is due to Con A weakly reactive glycoforms with three triantennary and one biantennary glycan, and peaks 3 and 4 to Con A reactive glycoforms with an average of one triantennary and three biantennary glycans per molecule [12,13]. Peak areas enclosed by precipitates were measured by weighing a paper tracing of each pattern. A vertical line was drawn to separate peaks 1 and 2 from peaks 3 and 4. A ratio was calculated for each serum sample according to the formula: proportions of peak 1+peak 2 proportions of peaks (3+ 4) and was called Ra,-ACT. The precision of the measurements was within 2%. We checked the within-run, the within-day and the day-to-day reproducibilities: the variations were less than 5%. Isoelectric focusing was carried out as described in [ 111 with Servalyt (Serva Feinbio-chemica) carrier ampholytes pH 3-6 and pH 4-4.5 in a 1:2 ratio. Immunoblotting, immunological detection of proteins on nitrocellulose and conventional crossed immunoelectrophoresis were performed as previously indicated [ 111. The Mann-Whitney non-parametric test was used for statistical evaluation. Correlation coefficients ( Y) were calculated by using Spearman’srank correlation test.
44 I
Fig. 2. Crossed immunoelectrophoretic patterns of a,-ACT in the serum of (a) a healthy donor, ( b ) a patient with Child’s grade C cirrhosis and (c) a patient on the first day after liver transplantation, with specific antiserum against a,-ACT (from Dakopatts) in the second dimension
RESULTS Typical patterns produced by a,-ACT on CIAE are shown for a healthy subject in Fig. 1(a), for patients with cirrhosis with Child’s grade A and C in Figs. l ( b ) and 1(c), respectively, in a patient after liver transplantation [Fig. l(d), day 1; Fig. 1(e), day 151 and for a patient with acute viral hepatitis, after the peak level of serum transaminases, in Fig. lcf).As compared with the normal subjects, there was an increase in peak 1 and peak 2 in patients with cirrhosis and acute hepatitis (Figs. 1b, 1c and If). Moreover, in the patients with cirrhosis with Child’s grade C, there was a dramatic decrease in peaks 3 and 4 as compared with normal subjects; for three patients (as in Fig. l c ) there was no individualized peak 4 at all and just a small peak 3. In the light of these various patterns we studied the migration in conventional crossed immunoelectrophoresis of a,-ACT in the same serum samples. No noticeable change in migration was observed from one sample to another. Fig. 2 shows the results obtained for the patients
442
E. Hachulla et al.
1
2
3
4
5
I
PI
I
2
3
4
5
I
Fig. 3. Isoelectric focusing in polyacrylamide gel in a pH 3-6 gradient. (a) Pattern after fixation and staining with Coomassie Brilliant Blue R 250. ( b ) lmmunoblotting pattern with antiserum against a,-ACT after transfer on t o nitrocellulose. The Figure shows typical serum samples from I, a healthy donor; 2, a patient with Child’s grade C cirrhosis; 3, a patient on the first day after liver transplantation; 4, the same patient 15 days after the liver transplantation: (5) another healthy donor. I, a,-ACT purified from the serum o f a healthy subject; PI, isoelectric focusing calibration kit (low pl).
who exhibited the patterns that differed the most from those of normal subjects, i.e. a patient with Child’s grade C cirrhosis (Fig. Ic, Fig. 2b) and a patient on the first day after liver transplantation (Fig. Id, Fig. 2c). The heterogeneity of the precipitate observed in CIAE is certainly caused by the interactions with Con A and not by changes in electrophoretic migration owing, for example, to desialylation or to a decrease in charge due to another heterogeneity. To confirm this observation and to allow the direct comparison of samples, studies by isoelectric focusing, immunoblotting and immunological detection of a ,-ACT with specific antiserum were performed and the results are shown in Fig. 3. We observed that no dramatic changes in PI values were obtained for a,-ACT from the patients when compared with the healthy subjects. The more acid a,-ACT bands are observed for the patient with Child’s grade C cirrhosis. For the other samples the bands have more or less the same migration, indicating no noticeable changes in charge. The !?a,-ACT values and the serum levels of a,-ACT for the normal subjects and for the patients with cirrhosis and acute viral hepatitis at the first presentation are shown in Table 1. The serum levels of a,-ACT did not differ significantly between patients with acute viral hepatitis and normal subjects. Significantly higher serum levels of a,-ACT were observed in patients with cirrhosis with Child’s grade A or B compared with normal subjects. On the other hand, there was no significant difference between patients with Child’s grade C cirrhosis and normal subjects. The median Ra,-ACT was 1 (range 0.72-1.25) in normal subjects. It was 1.4 (range 0.72-3.34) in patients with acute viral hepatitis and there was just a slight, but significant, difference when
compared with normal subjects ( P = 0.02). In patients with cirrhosis, the median Ra,-ACT was 1.6 (range 1.15-3.02), 1.45 (range 0.65-4.12) and 2.24 (range 1.03-19) in Child’s grade A, B and C, respectively, and differed significantly from that in normal subjects ( P < 1W6,P< and P< respectively). Ra,-ACT differed significantly in grade C cirrhosis when compared with grade A and grade B cirrhosis (P
439
Variations in the glycoforms of serum a ,-antichymotrypsin in liver diseases and after liver transplantation E. HACHULLA*, A. LAINE*, V. HEDOUINt, C. FOURNIERf, F. MAURYt, C. MATHlEUt, F. R. PRUVOT!, N. DECLERCK!, J. C. PARlSt and P. DEGAND* “Unit6 16 de I’lnstitut National de la SantC et de la Recherche MCdicale, Lille, France, Clinique des Maladies de I’Apporeil Digestif, CHU HBpital Claude Huriez, Lille, France, Slaboratoire de Pharmocodynamie Clinique, Centre Oscar Lambret, Lille, France, and SClinique Chirurgicale, CHU HBpital Calmette, Lille, France
t
(Received
4 Junel30 October 1991; accepted 13 November 1991)
1. Using crossed immunoafinity electrophoresis with
free concanavalin A in the first dimension, we studied the microheterogeneity of a,-antichymotrypsin due to various glycoforms in sera from patients with various liver diseases and after liver transplantation. 2. Studies by isoelectric focusing and immunoblotting and by crossed immunoelectrophoresis without concanavalin A in the first dimension allowed us to show that there is no dramatic variation in electrophoretic heterogeneity of a,-antichymotrypsin in the serum of patients with liver diseases or after liver transplantation when compared with that of normal subjects. Therefore the heterogeneity observed in crossed immunoaffinity electrophoresis is due to various interactions with concanavalin A. 3. The results were expressed as the ratio of concanavalin A non-reactive glycoforms plus concanavalin A weakly reactive glycoforms to concanavalin A reactive glycoforms, called R a ,-ACT. R a ,-ACT was significantly higher in patients with cirrhosis (n= 53) when compared with normal subjects (n= 30). The median R a ,-ACT was 1 (range 0.72-1.25) in normal subjects. It was 1.6 (range 1.18-3.02), 1.45 (range 0.65-4.12) and 2.24 (range 1.03-19) in cirrhosis of Child’s grade A, B and C, respectively. There was a dramatic decrease in glycoforms with mostly biantennary glycans in some patients with Child’s grade C cirrhosis. Serum levels of a ,-antichymotrypsin were lower than normal only in some patients with Child’s grade C cirrhosis. 4. Among the patients with acute viral hepatitis studied ( n = 17), five were studied longitudinally. Ra,-ACT increased after the peak level of serum transaminases and remained high while the level of serum transaminases decreased and liver regeneration occurred. In these patients there was no alteration in glycoforms with mostly biantennary glycans.
5. Four patients were studied after liver transplantation. During the first days after transplantation Ra,-ACT was low because of an increase in glycoforms with mostly biantennary glycans. After the first week, Ra ,-ACT rose above the normal range because of an increase in glycoforms with pluriantennary glycans and it returned to normal within 3 post-operative weeks. R a ,-ACT remained high in a patient who presented with liver rejection. 6. The present study suggests that Ra,-ACT is a helpful additional marker for the evaluation of hepatocyte function.
INTRODUCTION For most of the serum acute-phase reactants, the major, if not the only, site of synthesis is the liver, and specifically the hepatocyte [l]. Like C-reactive protein, a,-antichymotrypsin (a,-ACT) is one of the acute-phase reactants whose concentration increases within 12 h of an inflammatory process [l,21. a,-ACT, with an M , of about 59000 and a carbohydrate content of 24% [3], is known to inhibit chymotrypsin-like proteinases [4, 51. It belongs to the superfamily of serpins (serine proteinase inhibitors), a group of proteins very similar to each other in amino acid sequence but differing in biological function [6].Furthermore, a,-ACT plays a role in the modulation of the immune response [7] but does not modify natural killing cytotoxicity [S]. Many glycoproteins exhibit molecular heterogeneity, essentially due to different carbohydrate structures, and consist of various glycoforms. The variety of the glycoforms is easily demonstrated directly in the serum by crossed immunoaffinity electrophoresis (CIAE) with lectin [9]. In CIAE with free concanavalin A (Con A ) in the first dimension, human serum a,-ACT exhibits three
Key words: a,-antichymotrypsin, glycoforms, liver disease, liver transplantation, microheterogeneity. Abbreviations:a,-ACT, a,-antichymotrypsin; CIAE, crossed immunoaffinity electrophoresis: Con A, concanavalin A Ra,-ACT, (concanavalin A non-reactive glycoforms -t concanavalin A weakly reactive g1ycoforms)lconcanavalin A reactive glycoforms ratio. Correspondence: D r Eric Hachulla, Clinique Medicale A, (Pr B. Devulder), H6pital Claude Huriez, 59037 Lille Cedex, France.
440
E. Hachulla e t al.
Fig. I. CIAE of human serum with specific antiserum against a,-ACT. The figure shows typical serum samples from ( a ) a normal subject (Ra,-ACT 0.93, serum a,-ACT level 0.38 gll), (6) a patient with a Child’s grade A cirrhosis (Ra,-ACT 1.45, serum a,-ACT level 0.82 gll), (c) a patient with a Child’s grade C cirrhosis (Ra,-ACT 19, serum a,-ACT level 0.36 gil), (d) a patient on the first day after liver transplantation, who was previously a cirrhotic patient (Ra,-ACT 0.38, serum a,-ACT level 0.88 g/l), (e) the same patient 15 days after the transplantation (Ra,-ACT 1.68, serum a,-ACT level 1.86 gll) and (0 a patient with acute viral hepatitis, after the peak level of serum transaminases (Ra,-ACT 2.22, serum a,-ACT level 0.77 gll), !+, first dimension; 2t, second dimension.
peaks, or four peaks when sugar (displacer) is added to the gel of the second dimension [ l o , 111. The complete amino acid sequence of a,-ACT deduced from the nucleotide sequence [6] reveals four potential glycosylation sites. In a recent work [12], we determined the complete primary structure of the glycans of a,-ACT by high-resolution IH-n.m.r. spectroscopy. The results indicated the presence of disialyl biantennary and trisialyl triantennary type glycanic structures and traces of disialylated triantennary oligosaccharide. Significant variations in the proportions of the glycoforms separated by CIAE were detected in various inflammatory syndromes for a,-ACT as well as for other acute-phase glycoproteins [lo, 13-15]. In liver diseases like cirrhosis o r acute hepatitis, an increase in Con A non-reactive glycoforms was detected for a,-HS glycoprotein, a,-acid glycoprotein and serotransferrin [ 16-19]. In the present study we investigated the variations in the glycoforms of a,-ACT in various liver diseases to determine if these variations were correlated with hepatocellular insufficiency and thus could provide a functional parameter useful for predicting the ultimate outcome. To our knowledge, the variations in the proportions of the glycoforms of plasma glycoproteins which are
synthesized in the liver have not, until now, been studied before and after liver transplantation. This study would be of particular interest for evaluating this aspect of hepatocyte function.
MATERIALS AND METHODS Blood samples were obtained from 30 healthy subjects (13 men and 17 women, aged 27-80 years), none of which was receiving hormonal medication, from 53 patients with biopsy-proven cirrhosis (20 men and 33 women, aged 29-80 years: 44 patients with alcoholic cirrhosis, seven patients with cirrhosis due to hepatitis B virus and two patients with cirrhosis due to hepatitis C virus) and from 17 patients with acute viral hepatitis (six men and 11 women, aged 18-41 years). A longitudinal study was performed on the serum of four patients with cirrhosis who had a liver transplantation (3-4 weeks after the operation) and of five of the 17 patients with acute viral hepatitis; one of these five patients developed a fulminant hepatic failure with a spontaneous favourable evolution. For the patients with cirrhosis the state of liver function was graded by the modified Child’s scoring system [20], which is based on the grading (A, B, C ) of five clinical and
Serum a,-antichymotrypsin glycoforms
biological complications of the cirrhotic process (ascitis, encephalopathy, albuminaemia, bilirubinaemia, prothrombin time). At the time of Child’s scoring, blood samples were also drawn from the patients for measurement of serum a,-ACT and proaccelerin levels and for CIAE. Besides serum a ,-ACT, albumin and proaccelerin levels, serum transaminases (serum glutamic-pyruvic transaminase activity) were assessed every time a blood sample was taken from the patients who had undergone liver transplantation and from the patients suffering from acute viral hepatitis at the same time as CIAE was performed. Serum a,-ACT levels were determined by a single conventional radial immunodifhsion technique (Behringwerke). CIAE was performed using a modification of the method of B0g-Hansen et al. [9] as previously described [13].C U E was performed in thin-layer 1% (w/v) agarose gels (10 cm x 10 cm) using free Con A (routinely 250 pg/ cm’) placed in a 1 cm x 8 cm trap-gel in the first dimension. A bovine albumin solution prepared in concentrated Bromophenol Blue solution applied above the sample well was used as a coloured marker to follow the migration in the plate. The first dimension was run for 80 min at 10 V/cm, i.e. until the coloured bovine albumin had crossed the plate. The second dimension was run for 18 h at 2 V/cm in a 1% (w/v) agarose gel, poured onto the upper two-thirds of the plate, which contained 0.3 pl of specific antiserum/cm* against a,-ACT and a-methyl-Dglucoside (2%). Three microlitres of serum was applied for each analysis. a,-ACT separates into four peaks on CIAE numbered 1-4. We have previously shown that peak 1 is due to Con A non-reactive glycoforms containing four triantennary glycans, peak 2 is due to Con A weakly reactive glycoforms with three triantennary and one biantennary glycan, and peaks 3 and 4 to Con A reactive glycoforms with an average of one triantennary and three biantennary glycans per molecule [12,13]. Peak areas enclosed by precipitates were measured by weighing a paper tracing of each pattern. A vertical line was drawn to separate peaks 1 and 2 from peaks 3 and 4. A ratio was calculated for each serum sample according to the formula: proportions of peak 1+peak 2 proportions of peaks (3+ 4) and was called Ra,-ACT. The precision of the measurements was within 2%. We checked the within-run, the within-day and the day-to-day reproducibilities: the variations were less than 5%. Isoelectric focusing was carried out as described in [ 111 with Servalyt (Serva Feinbio-chemica) carrier ampholytes pH 3-6 and pH 4-4.5 in a 1:2 ratio. Immunoblotting, immunological detection of proteins on nitrocellulose and conventional crossed immunoelectrophoresis were performed as previously indicated [ 111. The Mann-Whitney non-parametric test was used for statistical evaluation. Correlation coefficients ( Y) were calculated by using Spearman’srank correlation test.
44 I
Fig. 2. Crossed immunoelectrophoretic patterns of a,-ACT in the serum of (a) a healthy donor, ( b ) a patient with Child’s grade C cirrhosis and (c) a patient on the first day after liver transplantation, with specific antiserum against a,-ACT (from Dakopatts) in the second dimension
RESULTS Typical patterns produced by a,-ACT on CIAE are shown for a healthy subject in Fig. 1(a), for patients with cirrhosis with Child’s grade A and C in Figs. l ( b ) and 1(c), respectively, in a patient after liver transplantation [Fig. l(d), day 1; Fig. 1(e), day 151 and for a patient with acute viral hepatitis, after the peak level of serum transaminases, in Fig. lcf).As compared with the normal subjects, there was an increase in peak 1 and peak 2 in patients with cirrhosis and acute hepatitis (Figs. 1b, 1c and If). Moreover, in the patients with cirrhosis with Child’s grade C, there was a dramatic decrease in peaks 3 and 4 as compared with normal subjects; for three patients (as in Fig. l c ) there was no individualized peak 4 at all and just a small peak 3. In the light of these various patterns we studied the migration in conventional crossed immunoelectrophoresis of a,-ACT in the same serum samples. No noticeable change in migration was observed from one sample to another. Fig. 2 shows the results obtained for the patients
442
E. Hachulla et al.
1
2
3
4
5
I
PI
I
2
3
4
5
I
Fig. 3. Isoelectric focusing in polyacrylamide gel in a pH 3-6 gradient. (a) Pattern after fixation and staining with Coomassie Brilliant Blue R 250. ( b ) lmmunoblotting pattern with antiserum against a,-ACT after transfer on t o nitrocellulose. The Figure shows typical serum samples from I, a healthy donor; 2, a patient with Child’s grade C cirrhosis; 3, a patient on the first day after liver transplantation; 4, the same patient 15 days after the liver transplantation: (5) another healthy donor. I, a,-ACT purified from the serum o f a healthy subject; PI, isoelectric focusing calibration kit (low pl).
who exhibited the patterns that differed the most from those of normal subjects, i.e. a patient with Child’s grade C cirrhosis (Fig. Ic, Fig. 2b) and a patient on the first day after liver transplantation (Fig. Id, Fig. 2c). The heterogeneity of the precipitate observed in CIAE is certainly caused by the interactions with Con A and not by changes in electrophoretic migration owing, for example, to desialylation or to a decrease in charge due to another heterogeneity. To confirm this observation and to allow the direct comparison of samples, studies by isoelectric focusing, immunoblotting and immunological detection of a ,-ACT with specific antiserum were performed and the results are shown in Fig. 3. We observed that no dramatic changes in PI values were obtained for a,-ACT from the patients when compared with the healthy subjects. The more acid a,-ACT bands are observed for the patient with Child’s grade C cirrhosis. For the other samples the bands have more or less the same migration, indicating no noticeable changes in charge. The !?a,-ACT values and the serum levels of a,-ACT for the normal subjects and for the patients with cirrhosis and acute viral hepatitis at the first presentation are shown in Table 1. The serum levels of a,-ACT did not differ significantly between patients with acute viral hepatitis and normal subjects. Significantly higher serum levels of a,-ACT were observed in patients with cirrhosis with Child’s grade A or B compared with normal subjects. On the other hand, there was no significant difference between patients with Child’s grade C cirrhosis and normal subjects. The median Ra,-ACT was 1 (range 0.72-1.25) in normal subjects. It was 1.4 (range 0.72-3.34) in patients with acute viral hepatitis and there was just a slight, but significant, difference when
compared with normal subjects ( P = 0.02). In patients with cirrhosis, the median Ra,-ACT was 1.6 (range 1.15-3.02), 1.45 (range 0.65-4.12) and 2.24 (range 1.03-19) in Child’s grade A, B and C, respectively, and differed significantly from that in normal subjects ( P < 1W6,P< and P< respectively). Ra,-ACT differed significantly in grade C cirrhosis when compared with grade A and grade B cirrhosis (P
