- Email: [email protected]
Elevated serum aminoterminal procollagen type-III-peptide parallels collagen accumulation in rats with secondary biliary fibrosis
Journal of Hepatology 1996; 25:79-84 Printed in Denmark • All rights reserved Munksgaard. Copenhagen
Copyright © EuropeanAssociation for the Study of the Liver 1996 Journal of Hepatology
ISSN 0168-8278
Elevated serum aminoterminal procollagen type-III-peptide parallels collagen accumulation in rats with secondary biliary fibrosis Burkard Gerling 1, Michael Becker I, JiJrgen Waldschmidt 2, Martina Rehmann 2 and Detlef Schuppan 3 tDepartment of Pediatrics, KAVH, Virchow-Klinikum, Humboldt Universi~ of Berlin, 2Department of Pediatric Surge~ and3Department of Gastroenterology, Klinikum Benjamin Franklin, Free University of Berlin, Germany
Background~Aims: The aminoterminal procollagen type-III-peptide, which can be released d u r i n g collagen type III deposition, has been suggested as a s e r u m m a r k e r of fibrogenesis in patients with chronic liver disease. However, longitudinal studies correlating procollagen type-III-peptide concentrations in s e r u m with the evolution of liver fibrosis are still needed. The purpose of the present study was to prove the significance of procollagen type-III-peptide concentrations in s e r u m as a noninvasive m a r k e r of liver fibrogenesis in an animal model that best resembles progressive h u m a n liver fibrosis. Methods: In 88 female Wistar rats the biliary system was occluded by double ligation of the choledochal duct followed by retrograde injection of a mixture of prolamine/ethanole (Ethibloc®). Sixteen rats served as controls. Groups of 8--10 rats were sacrificed at days 2, 7, 14, 21, 30, 32, 35, 60 and 90 after bile duct occlusion. In the groups histological staging (fibrosis score), determination of total liver hydroxyproline, m e a s u r e m e n t of s e r u m
procollagen type-III-peptide and routine liver function tests were performed. Results: First histological signs of liver fibrosis were seen as early as 7 days after bile duct occlusion. Progressive fibrosis was paralleled by an increase of s e r u m procollagen type-III-peptide. There was a significant correlation between s e r u m procollagen type-III-peptide and histological stages of fibrosis (r=0.80; p<0.0001) as well as between s e r u m procollagen type-III-peptide and hydroxyproline in total liver tissue (r=0.85; p<0.0001) Conclusions: These results indicate that: (1) complete bile duct occlusion in rats produces progressive hepatic fibrosis resembling h u m a n secondary biliary fibrosis, and (2) procollagen type-III-peptide concentrations in s e r u m reflect ongoing collagen formation in the liver unrelated to s e r u m m a r k e r s of cholestasis and inflammation.
liver disease, fibrosis has been quantified I NbyCHRONIC histological staging or by measurement of hy-
pecially in the light of attempts to prevent or even reverse liver fibrosis by pharmacological intervention. These methods should ideally allow assessment of fibrogenesis, i.e. the rate of de novo deposition of connective tissue, or fibrolysis, i.e. the rate of removal of excess connective tissue, or fibrosis, i.e. the amount of connective tissue deposited in the liver. Several serum assays have been developed using putative markers of fibrogenesis, fibrolysis or fibrosis (5-7). However, none of these markers has been unequivocally shown to reflect one of these processes. The aminoterminal procollagen type-III-peptide (PIIINP), which is mainly released from the surface of collagen fibrils during fibril growth, has been suggested as a
droxyproline in biopsy specimens (1-4). However, quantification of fibrosis in human liver disease is subject to a significant sampling error (4). Furthermore, frequent biopsies are practically and ethically not possible. Therefore, non-invasive methods to determine the progress of liver fibrosis are needed, esReceived 10 July; revised 28 November; accepted 30 November 1995
Correspondence: Prof. Dr. Michael Becker, VirchowKlinikum, Humboldt University Kinderklinik, Augustenburger Platz 1, D-13353 Berlin, Germany. Tel. 0049-304566171, Fax. 0049-30-45066917.
Key words: Aminoterminal procollagen-III-peptide; Bile duct occlusion; Cholestasis; Hydroxyproline; Liver fibrosis; Rat.
79
B. Gerling et al.
marker of fibrogenesis, but the data in humans are still controversial or based on cross-sectional studies (2,3,8-24). On the other hand, the rat models used to validate PIIINP in relation to histological progression (carbon tetrachloride-intoxication, nutritional deficiency) do not reflect the pathology of the common human liver diseases (1,25,26). We therefore chose to determine serum PIIINP in a rat model of progressive biliary fibrosis secondary to complete occlusion of the biliary systems by Ethibloc. This model is not accompanied by overt hepatic inflammation and thus resembles human primary and especially secondary biliary fibrosis
700
• []
A L A T (U/I) bilirubin (~mol/I)
• AP
600
(U/I)
500
4O0
300
200
1 O0
0 0
Materials and Methods In 88 female Wistar rats (Wistar RA 25, aged 3 months, mean weight 339+23.7 g) the extra- and intrahepatic biliary systems were completely occluded by double ligation of the choledochal duct followed by retrograde injection of 1.5 ml/kg bw of a mixture of prolamine and ethanol (Ethibloc, Eticon, Hamburg, Germany). Sixteen sham-operated rats served as controls. In order to exclude the influence of growth-related variations of serum PIIINP (22,23), PIIINP serum levels were determined in 8 control rats aged 3 months (beginning of the study), and in 8 control rats at the age of 6 months (end of the study). The bile-duct-occluded animals were divided into 9 groups of 8-10 animals that were sacrificed at days 2, 7, 14, 21, 30, 32, 35, 60 or 90 after the procedure. At this time, blood was obtained by puncture of the right cardiac ventricle and the liver was removed. Experiments were performed according to the usual ethical principles and were approved by the senator of health, reference number IV A 4-5855 and IV A 4 30/88. Histological evaluation Tissue sections of the right and left lobe of each liver were prepared and fixed in 5% formalin, followed by staining with hematoxylin and eosin, and with Gomori's silver impregnation. Staging of fibrosis was done blindly by an experienced pathologist, according to the method of Ruwart et al. (1), but with slight modifications: Stage 0: normal liver. Stage 1: definite increase of portal collagen, without septum formation. Stage 2: further increase of portal collagen with incomplete septum formation. Stage 3: further increase of portal collagen with complete formation of thin septa. 80
20
, 40
•
"1" 60
.
, 80
1()0
time (days) Fig. 1. Alanine-aminotransferase (ALAT, U/l) (11), bilirubin (gmol/1) (u) and alkaline phosphatase (AP, U/l) (s) in the serum of rats after bile duct occlusion.
Stage 4: complete cirrhosis (formation of thick septa with nodule formation) Scores from the right and left lobe were averaged to give a single numeric value for each liver. Hydroxyproline was measured colorimetrically from hydrolyzed liver tissue (27). In brief, 200 mg tissue were taken from the left and the right lobe of the liver, homogenized in 4 ml 6N HCI and hydrolyzed under nitrogen at ll0°C for 16 h. The hydrolysate was filtered and 50 ~tl were dried under vacuum and the sediment was dissolved in 1.2 ml 50% isopropanol. Then 0.2 ml of a 0.84% solution of chloramine-T was added, followed by incubation for 10 min at room temperature. Next, 1.0 ml of 12% Ehrlich's reagent was added and the mixture incubated at 50°C for 90 min. Hydroxyproline was then quantitated photometrically at 558 nm from a standard curve with the aminoacid and against a reagent blank. Serum PIIINP was analyzed with a radioimmunoassay developed for rat PIIINP, as described previously (1,25). Briefly, inhibition assays (sequential saturation) were carried out using a fixed dilution of rabbit-anti-rat-PIIINP antiserum (1:50,0001:100,000) incubated with 200 ~tl of the inhibitor (standard or serum in 3 sequential dilutions) for 16 h. Then 100 ~tl of 1 ng/100 ~1 ~25Iodine-labeled rat PIIINP was added and incubated for another 4 h, followed by precipitation of immune-complexes with goat antiserum to rabbit IgG, and quantitation of precipitated radioactivity in a scintillation counter. Concentrations were calculated graphically by the 50%intercept method, according to Rohde et al. (28).
PIIINP in progressive liver fibrosis in rats
12
Fig. 2a: Histological findings at day 7 after bile duct occlusion in liver tissue sections stained with Gomori "s silver impregnation: increase of portal collagen without septa (stage 1). (Original magnification x125) Fig. 2b: Histological findings at day 30 after bile duct occlusion in liver tissue sections stained with Gomori "s silver impregnation: increase of portal collagen with complete septa (stage 3). (Original magnification x125) Fig. 2c: Histological findings at day 60 after bile duct occlusion in liver tissue sections stained with Gomori "s silver impregnation: complete cirrhosis (stage 4). (Original magnification x125)
Serum bilirubin, alanine aminotransferase (ALAT) and alkaline phosphatase (AP) were measured by standard methods. Statistical evaluation All results are given as mean+SD. Intergroup comparisons were performed by Kruskal-Wallis ANOVA and the Mann-Whitney U-test. Correlation coeffi-
TABLE 1 C o r r e l a t i o n s b e t w e e n s e r u m P I I I N E l i v e r h y d r o x y p r o l i n e a n d histol o g i c a l s t a g e s o f l i v e r fibrosis w i t h a l a n i n e - a m i n o t r a n s f e r a s e ( A L A T ) , alkaline phosphatase (AP) and bilirubin Coefficient of
Significance
correlation PIIINP/hydroxyproline PllINP/histology
r = 0.85 r -= 0 . 8 0
p < 0.0001 p < 0.0001
PIIINP/ALAT
r = ~).40
p < 0.005
PIIINP/AP PllINP/bilirubin Hydroxyproline/histology Hydroxyproline/ALAT
r r r r
n.s. n.s. p < 0.0001 p < 0.0001
Hydroxyproline/AP Hydroxyproline/bilirubin
r = 0.19 r = 0.20
n.s. n.s.
Histology/ALAT Histology/AP
r = -0.58 r = 0.14
p < 0.0001 n.s.
Histology/bilirubin
r = 0.19
n.s.
= = = =
0.17 0.25 0.94 -0.53
cients were calculated using Spearman rank correlation.
Results In all animals double ligation of the common bile duct and injection of prolamine/ethanol resulted in complete occlusion of the extra- and intrahepatic bile ducts. Histological evaluation showed significant hepatic fibrosis seen as early as 7 days after bile duct occlusion; it increased continuously without significant inflammatory activity and resulted in complete cirrhosis 60 to 90 days after occlusion. After the intervention, there were sharp initial increases in AP, bilirubin and ALAT, with a subsequent normalization of ALAT activity, whereas bilirubin and AP activity remained elevated (Fig. 1 and 2). Total liver hydroxyproline (Table 1) paralleled the progression of histologically verified fibrosis (r=0.94, p<0.0001). As early as 7 days after bile duct occlusion serum PIIINP was significantly higher than in the controls (p<0.001), only to increase further up to day 90 (Kruskal-Wallis ANOVA: p<0.0001). Growthrelated changes in serum PIIINP and hydroxyproline could be excluded, since in controls these parameters did not differ significantly at the beginning of the study from those at day 90 (Fig. 3). There was a sig81
B. Gerling et al. Discussion
80
140 130 120
l
110
~L
100 90
70 60
J 50
80
70-
40 n Z " 30 E
"° 60"c~ 50-
0
[]
4030-
20
20-
10
1 '~~
.
.
,
0
.
,
20
40 time
.
,
,
,
60 (days)
.
80
0
100
Fig. 3. Total liver hydroxyproline (u) and serum PIIINP (s) in rats with progressive biliary fbrosis after bile duct occlusion and total liver hydroxyproline (ll) and serum PIIIINP (O) in controls (means + SD). Different from control rats: #p
80
___
•
•, , ,
60
//
40 E,
•
•
.205x
20
•
;/
~m
•
•
-
(r =
0.85)
•
• m.m(__m_ • "
0 0
, 20
,~ 0
' 60
total hydroxyproline (gmol) Fig. 4. Correlation of serum PIIINP with total liver hydroxyproline in the 88 rats with progressive biliary fibrosis due to bile duct occlusion (r=0.85, p
nificant correlation between serum PIIINP and the histological stages of fibrosis (Table 1; r--0.80; p<0.0001), as well as between PIIINP and total liver hydroxyproline (Fig. 4; r--0.85; p<0.0001). A negative correlation was found between ALAT and PIIINE histology and hydroxyproline. However, there was no correlation between PIIINR histology or total liver hydroxyproline and bilirubin as well as AP (Table 1). 82
In the search for a rat model of progressive and reproducible liver fibrosis, we completely occluded the biliary system by double ligation of the common bile duct and retrograde injection of prolamine and ethanol. We could show that this procedure resulted in progressive biliary fibrosis with only minor inflammatory activity, as demonstrated by histology and an only moderately increased ALAT in the groups sacririced at day 2 and 7 after intervention. Significant fibrosis was seen as early as 7 days after occlusion, and complete cirrhosis occurred in all animals after 60 to 90 days. Since in previous studies in rats double ligation of the bile duct (without occlusion of the intrahepatic bile ducts by retrograde injection of prolamine/ethanol), did not result in complete cirrhosis in all animals (29), our method appears to be better suited for inducing a reproducible portal fibrosis, a model that could be of major importance for the testing of potential antifibrotic agents. In addition, the lack of significant histological inflammation and necrosis in our model might allow distinction between pure antifibrotic and anti-inflammatory, radical scavanging or membrane protecting agents. This is of major importance, since previously used models such as carbon-tetrachloride- or galactosamineinduced liver fibrosis did not lead to the discovery or validation of effective antifibrotic agents in man. Furthermore, since PIIINP levels behave similarly in rat and in man (25), this model suggests the usefulness of sequential measurements of serum PIIINP examinations in chronic human fibrotic liver disease that are devoid of major inflammatory activity. We could demonstrate that there was a significant correlation between serum PIIINP, the histological stages of fibrosis, and total liver hydroxyproline. Although we found a negative correlation of ALAT with PIIINP, total liver hydroxyproline and histological stages of fibrosis, this correlation was only due to the initial cell injury after the injection of prolamine and ethanol, resulting in an increase in ALAT observed only within the first 7 days. No correlation was found between serum PIIINP or hydroxyproline and bilirubin or alkaline phosphatase, supporting the empirical conclusions of previous reports (1,2,3,824). Numerous studies have reported elevated serum PIIINP levels in patients with hepatic fibrosis (2,3,824). However, most of these were of a cross-sectional design and liver biopsies were not or only rarely performed. A significant correlation between PIIINP in serum and the histological stage of liver fibrosis was demonstrated in patients with alcoholic fibrosis (15,19), primary biliary cirrhosis (9,18) and schisto-
PIlINP in progressive liver fibrosis in rats
somiasis of the liver (11), findings that were not confirmed by others (2,12). In addition, in severe chronic active hepatitis no correlation was found between serum PIIINP and the histological stage of fbrosis (17). These differences may be explained by the dynamic nature of chronic liver disease, with fibrogenesis being defined as the deposition of connective tissue in time, fibrolysis as the removal of connective tissue in time, whereas fibrosis, as a static term, defines the connective tissue that has accumulated over a given time (5-7). By definition, in progressive fibrosis, fibrogenesis exceeds fibrolysis. This appears to be reflected in increased PIIINP serum levels. Accordingly, in fibrosis with little fibrogenic activity, PIIINP serum concentrations have been found to be only slightly increased or even normal (24,30). Since for practical and ethical reasons, in these studies minimal fibrogenic activity could not be evaluated by sequential liver biopsies, it could not be clarified if PIIINP discriminates patients with progressive fibrosis, non-active fibrosis or fibrosis in regression (fibrolysis). The secondary biliary fibrosis of the present study was progressive with a large excess of fibrogenesis over fibrolysis. This was clearly demonstrated by the continuous increase of histologically verified fibrosis paralleled by the augmentation of total liver hydroxyproline. In conclusion, this new animal model with complete bile duct occlusion leads to progressive and reproducible liver fibrosis with little inflammation, and major similarity to human secondary biliary fibrosis, underlining its suitability for the testing of true antifibrotic agents. Since in this model serum PIIINP showed an excellent correlation with ongoing fibrogenesis and collagen accumulation, this noninvasive parameter should be used as a marker in longitudinal follow-up studies of progressive liver fibrosis in man.
4. 5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
References 1. Ruwart MJ, Wilkinson KF, Rush BD, Vidmar TJ, Peters KM, Henley KS, Appelman HD, Kim KY, Schuppan D, Hahn EG. The integrated value of serum procollagen III peptide over time predicts hepatic hydroxyproline content and stainable collagen in a model of dietary cirrhosis in the rat. Hepatology 1989; 10: 801-6. 2. Colombo M, Annoni G, Donato MF, Conte D, Martines D, Zaramella MG, Bianchi PA, Pipeono A, Tribelli C. Serum type III procollagen peptide in alcoholic liver disease and idiopathic hemochromatosis: its relationship to hepatic fibrosis, activity of the disease and iron overload. Hepatology 1985; 3: 475-9. 3. Frei A, Zimmermann A, Weigand K. The N terminal propeptide of collagen type III in serum reflects activity and degree
17.
18.
19.
20.
of fibrosis in patients with chronic liver disease. Hepatology 1984; 5: 830-4. Abdi W, Millan JC, Mezey E. Sampling variability in percutaneous liver biopsy. Arch Intern Med 1979; 139: 667-9. Schuppan D. Connective tissue polypeptides in serum as parameters to monitor antifibrotic treatment in hepatic fibrogenesis. J Hepatol 1991; 13: Suppl 3: 17-25. Schuppan D, St61zel U, Oesterling C, Somasundaram R. Serum assays for liver fibrosis. J Hepatol 1995; 22: Suppl 2: 82-8. Risteli L, Risteli J. Noninvasive methods for detection of organ fibrosis, In: Rojkind M, ed. Focus on Connective Tissue in Health and Disease. Boca Raton: CRC Press, 1990: 6198. Annoni G, Cargnel A, Colombo M, Hahn EG. Persistent elevation of the aminoterminal peptide of procollagen type III in serum of patients with acute viral hepatitis distinguishes chronic active hepatitis from resolving or chronic persistent hepatitis. J Hepatol 1986; 2: 379-88. Babbs C, Smith A, Hunt LP, Rowan B, Haboubi NY, Warnes TW. Type III procollagen peptide: a marker of disease activity and prognosis in primary biliary cirrhosis. Lancet 1988; i: 1021-4. Bentsen KD, Horn T, Risteli J, Risteli L Engstr6m-Laurent A, HCrslev-Petersen K, Lorenz I. Serum aminoterminal procollagen peptide and 7S domain of type IV collagen in patients with alcohol abuse. Liver 1987; 7: 339-46. Bentsen KD, Horslev-Petersen K, Junker P, Juhl E, Lorenzen I. Serum amino-terminal procollagen type III peptide in acute viral hepatitis. A long term follow-up study. Liver 1987; 7: 96-105. Beukers R, van Zanthen RA, Schalm SW. Serial determination of type III procollagen amino peptide serum levels in patients with histologically progressive and non-progressive primary biliary cirrhosis. J Hepatol 1992; 14: 22-9. E1-Mohandes M, Hassanein H, E1-Batrawy N, Voss B, Gerlach U. Serum concentration of N-terminal procollagen peptide of collagen type III in schistosomal liver fibrosis. Exp Mol Pathol 1986; 46: 383-90. Erikson S, Zettevall O. The N-terminal propeptide of collagen type III in serum as a prognostic indicator in primary biliary cirrhosis. J Hepatol 1986; 2: 370-8. Galambos MR, Collins DC, Galambos JT. A radioimmunoassay procedure for type III procollagen: its use for detection of hepatic fibrosis. Hepatology 1985; 1: 38-42. McCullough AJ, Stassen WN, Wiesner RH, Czaja AJ. Serial determination of the aminoterminal peptide of type III procollagen in severe chronic active hepatitis. J Lab Clin Med 1987; 1: 55-61. McCullough AJ, Strassen WN, Wiesner RH, Czaja AJ. Serum type III procollagen peptide concentrations in severe chronic active hepatitis: relationship to cirrhosis and disease activity. Hepatology 1987; 7: 49-54. Niemel~i O, Risteli L, Sotaniemi E, Stenb~ickF, Risteli J. Serum basement-membrane and type III procollagen - related antigens in primary biliary cirrhosis. J Hepatol 1988; 6: 30714. Sato S, Tashihiko N, Worner TM, Lieber CS. Liver fibrosis in alcoholics. Detection by lab radioimmunoassay of serum procollagen III peptides. JAMA 1986; 256: 1471-3. Surrenti C, Casini A, Milani S, Ambu S, Cecatelli P, D'Agata A. Is determination of serum N-terminal procollagen type III
83
B. Gerling et al.
21.
22.
23.
24.
25.
84
peptide (sPIIIP) a marker of hepatic fibrosis? Dig Dis Sci 1987; 7: 705-9. Torres-Salinas M, Pares A, Caballeria J, Jimenez W, Heredia D, Bruguera M, Rodes I. Serum procollagen type III peptide as a marker of hepatic fibrogenesis in alcoholic hepatitis. Gastroenterology 1986; 90: 1241-6. Trivedi P, Cheeseman P, Porman B, Hegarty J, Mowat AE Variation in serum type III procollagen peptide with age in healthy subjects and its comparative value in the assessment of disease activity in children and adults with chronic active hepatitis. Eur J Clin Invest 1985; 15: 69-74. Trivedi P, Risteli J, Risteli L, Tanner MS, Bhave S, Pandit AN, Mowat A.E Serum type III procollagen and basement membrane proteins as non-invasive markers of hepatic pathology in Indian childhood cirrhosis. Hepatology 1987; 6: 1249-53. Weigand K, Zaugg PY, Frei A, Zimmermann A. Long-term follow-up of serum N-terminal propeptide of collagen type III levels in patients with chronic liver disease. Hepatology 1984; 4: 835-8. Schuppan D, Dumont JM, Kim KY, Hennings G, Hahn EG. Serum concentration of the aminoterminal procollagen type III peptide in the rat reflects early formation of connective
26.
27.
28.
29.
30.
tissue in experimental liver cirrhosis. J Hepatol 1986; 3: 2737. Hayasaka A, Koch J, Schuppan D, Maddrey WC, Hahn EG. The serum concentrations of the aminoterminal propeptide of procollagen type III and the hepatic content of mRNA for the alpha 1 chain of procollagen type III in carbon tetrachloride-induced rat liver fibrogenesis. J Hepatol 1991; 13: 32838. Jamal JS, Finelli VN, Que Hee SS. A simple method to determine nanogram levels of hydroxyproline in biological tissues. Anal Biochem 1981; 112: 70-5. Rohde H, Vargas L, Hahn EG, Kalbfleisch H, Bruguera M, Timpl R. Radioimmunoassay for type III procollagen peptide and its application for human liver disease. Eur J Clin Invest 1979; 9: 451-9. Schweizer P, Schubert G.E, Reifferscheid E Das unterschiedliche morphologische Bild der Leber des Schweins und der Ratte nach Gallengangsverschlul3. Z Kinderchir 1974; 14: 61-74. Ballardini G, Faccani A, Bianchi FB, Famlani M, Patrono D, Capelli M, Pisi E. Steroid treatment lowers hepatic fibroplasia, as explored by serum aminoterminal procollagen III peptide, in chronic liver disease. Liver 1984; 4: 348-52.
Copyright © EuropeanAssociation for the Study of the Liver 1996 Journal of Hepatology
ISSN 0168-8278
Elevated serum aminoterminal procollagen type-III-peptide parallels collagen accumulation in rats with secondary biliary fibrosis Burkard Gerling 1, Michael Becker I, JiJrgen Waldschmidt 2, Martina Rehmann 2 and Detlef Schuppan 3 tDepartment of Pediatrics, KAVH, Virchow-Klinikum, Humboldt Universi~ of Berlin, 2Department of Pediatric Surge~ and3Department of Gastroenterology, Klinikum Benjamin Franklin, Free University of Berlin, Germany
Background~Aims: The aminoterminal procollagen type-III-peptide, which can be released d u r i n g collagen type III deposition, has been suggested as a s e r u m m a r k e r of fibrogenesis in patients with chronic liver disease. However, longitudinal studies correlating procollagen type-III-peptide concentrations in s e r u m with the evolution of liver fibrosis are still needed. The purpose of the present study was to prove the significance of procollagen type-III-peptide concentrations in s e r u m as a noninvasive m a r k e r of liver fibrogenesis in an animal model that best resembles progressive h u m a n liver fibrosis. Methods: In 88 female Wistar rats the biliary system was occluded by double ligation of the choledochal duct followed by retrograde injection of a mixture of prolamine/ethanole (Ethibloc®). Sixteen rats served as controls. Groups of 8--10 rats were sacrificed at days 2, 7, 14, 21, 30, 32, 35, 60 and 90 after bile duct occlusion. In the groups histological staging (fibrosis score), determination of total liver hydroxyproline, m e a s u r e m e n t of s e r u m
procollagen type-III-peptide and routine liver function tests were performed. Results: First histological signs of liver fibrosis were seen as early as 7 days after bile duct occlusion. Progressive fibrosis was paralleled by an increase of s e r u m procollagen type-III-peptide. There was a significant correlation between s e r u m procollagen type-III-peptide and histological stages of fibrosis (r=0.80; p<0.0001) as well as between s e r u m procollagen type-III-peptide and hydroxyproline in total liver tissue (r=0.85; p<0.0001) Conclusions: These results indicate that: (1) complete bile duct occlusion in rats produces progressive hepatic fibrosis resembling h u m a n secondary biliary fibrosis, and (2) procollagen type-III-peptide concentrations in s e r u m reflect ongoing collagen formation in the liver unrelated to s e r u m m a r k e r s of cholestasis and inflammation.
liver disease, fibrosis has been quantified I NbyCHRONIC histological staging or by measurement of hy-
pecially in the light of attempts to prevent or even reverse liver fibrosis by pharmacological intervention. These methods should ideally allow assessment of fibrogenesis, i.e. the rate of de novo deposition of connective tissue, or fibrolysis, i.e. the rate of removal of excess connective tissue, or fibrosis, i.e. the amount of connective tissue deposited in the liver. Several serum assays have been developed using putative markers of fibrogenesis, fibrolysis or fibrosis (5-7). However, none of these markers has been unequivocally shown to reflect one of these processes. The aminoterminal procollagen type-III-peptide (PIIINP), which is mainly released from the surface of collagen fibrils during fibril growth, has been suggested as a
droxyproline in biopsy specimens (1-4). However, quantification of fibrosis in human liver disease is subject to a significant sampling error (4). Furthermore, frequent biopsies are practically and ethically not possible. Therefore, non-invasive methods to determine the progress of liver fibrosis are needed, esReceived 10 July; revised 28 November; accepted 30 November 1995
Correspondence: Prof. Dr. Michael Becker, VirchowKlinikum, Humboldt University Kinderklinik, Augustenburger Platz 1, D-13353 Berlin, Germany. Tel. 0049-304566171, Fax. 0049-30-45066917.
Key words: Aminoterminal procollagen-III-peptide; Bile duct occlusion; Cholestasis; Hydroxyproline; Liver fibrosis; Rat.
79
B. Gerling et al.
marker of fibrogenesis, but the data in humans are still controversial or based on cross-sectional studies (2,3,8-24). On the other hand, the rat models used to validate PIIINP in relation to histological progression (carbon tetrachloride-intoxication, nutritional deficiency) do not reflect the pathology of the common human liver diseases (1,25,26). We therefore chose to determine serum PIIINP in a rat model of progressive biliary fibrosis secondary to complete occlusion of the biliary systems by Ethibloc. This model is not accompanied by overt hepatic inflammation and thus resembles human primary and especially secondary biliary fibrosis
700
• []
A L A T (U/I) bilirubin (~mol/I)
• AP
600
(U/I)
500
4O0
300
200
1 O0
0 0
Materials and Methods In 88 female Wistar rats (Wistar RA 25, aged 3 months, mean weight 339+23.7 g) the extra- and intrahepatic biliary systems were completely occluded by double ligation of the choledochal duct followed by retrograde injection of 1.5 ml/kg bw of a mixture of prolamine and ethanol (Ethibloc, Eticon, Hamburg, Germany). Sixteen sham-operated rats served as controls. In order to exclude the influence of growth-related variations of serum PIIINP (22,23), PIIINP serum levels were determined in 8 control rats aged 3 months (beginning of the study), and in 8 control rats at the age of 6 months (end of the study). The bile-duct-occluded animals were divided into 9 groups of 8-10 animals that were sacrificed at days 2, 7, 14, 21, 30, 32, 35, 60 or 90 after the procedure. At this time, blood was obtained by puncture of the right cardiac ventricle and the liver was removed. Experiments were performed according to the usual ethical principles and were approved by the senator of health, reference number IV A 4-5855 and IV A 4 30/88. Histological evaluation Tissue sections of the right and left lobe of each liver were prepared and fixed in 5% formalin, followed by staining with hematoxylin and eosin, and with Gomori's silver impregnation. Staging of fibrosis was done blindly by an experienced pathologist, according to the method of Ruwart et al. (1), but with slight modifications: Stage 0: normal liver. Stage 1: definite increase of portal collagen, without septum formation. Stage 2: further increase of portal collagen with incomplete septum formation. Stage 3: further increase of portal collagen with complete formation of thin septa. 80
20
, 40
•
"1" 60
.
, 80
1()0
time (days) Fig. 1. Alanine-aminotransferase (ALAT, U/l) (11), bilirubin (gmol/1) (u) and alkaline phosphatase (AP, U/l) (s) in the serum of rats after bile duct occlusion.
Stage 4: complete cirrhosis (formation of thick septa with nodule formation) Scores from the right and left lobe were averaged to give a single numeric value for each liver. Hydroxyproline was measured colorimetrically from hydrolyzed liver tissue (27). In brief, 200 mg tissue were taken from the left and the right lobe of the liver, homogenized in 4 ml 6N HCI and hydrolyzed under nitrogen at ll0°C for 16 h. The hydrolysate was filtered and 50 ~tl were dried under vacuum and the sediment was dissolved in 1.2 ml 50% isopropanol. Then 0.2 ml of a 0.84% solution of chloramine-T was added, followed by incubation for 10 min at room temperature. Next, 1.0 ml of 12% Ehrlich's reagent was added and the mixture incubated at 50°C for 90 min. Hydroxyproline was then quantitated photometrically at 558 nm from a standard curve with the aminoacid and against a reagent blank. Serum PIIINP was analyzed with a radioimmunoassay developed for rat PIIINP, as described previously (1,25). Briefly, inhibition assays (sequential saturation) were carried out using a fixed dilution of rabbit-anti-rat-PIIINP antiserum (1:50,0001:100,000) incubated with 200 ~tl of the inhibitor (standard or serum in 3 sequential dilutions) for 16 h. Then 100 ~tl of 1 ng/100 ~1 ~25Iodine-labeled rat PIIINP was added and incubated for another 4 h, followed by precipitation of immune-complexes with goat antiserum to rabbit IgG, and quantitation of precipitated radioactivity in a scintillation counter. Concentrations were calculated graphically by the 50%intercept method, according to Rohde et al. (28).
PIIINP in progressive liver fibrosis in rats
12
Fig. 2a: Histological findings at day 7 after bile duct occlusion in liver tissue sections stained with Gomori "s silver impregnation: increase of portal collagen without septa (stage 1). (Original magnification x125) Fig. 2b: Histological findings at day 30 after bile duct occlusion in liver tissue sections stained with Gomori "s silver impregnation: increase of portal collagen with complete septa (stage 3). (Original magnification x125) Fig. 2c: Histological findings at day 60 after bile duct occlusion in liver tissue sections stained with Gomori "s silver impregnation: complete cirrhosis (stage 4). (Original magnification x125)
Serum bilirubin, alanine aminotransferase (ALAT) and alkaline phosphatase (AP) were measured by standard methods. Statistical evaluation All results are given as mean+SD. Intergroup comparisons were performed by Kruskal-Wallis ANOVA and the Mann-Whitney U-test. Correlation coeffi-
TABLE 1 C o r r e l a t i o n s b e t w e e n s e r u m P I I I N E l i v e r h y d r o x y p r o l i n e a n d histol o g i c a l s t a g e s o f l i v e r fibrosis w i t h a l a n i n e - a m i n o t r a n s f e r a s e ( A L A T ) , alkaline phosphatase (AP) and bilirubin Coefficient of
Significance
correlation PIIINP/hydroxyproline PllINP/histology
r = 0.85 r -= 0 . 8 0
p < 0.0001 p < 0.0001
PIIINP/ALAT
r = ~).40
p < 0.005
PIIINP/AP PllINP/bilirubin Hydroxyproline/histology Hydroxyproline/ALAT
r r r r
n.s. n.s. p < 0.0001 p < 0.0001
Hydroxyproline/AP Hydroxyproline/bilirubin
r = 0.19 r = 0.20
n.s. n.s.
Histology/ALAT Histology/AP
r = -0.58 r = 0.14
p < 0.0001 n.s.
Histology/bilirubin
r = 0.19
n.s.
= = = =
0.17 0.25 0.94 -0.53
cients were calculated using Spearman rank correlation.
Results In all animals double ligation of the common bile duct and injection of prolamine/ethanol resulted in complete occlusion of the extra- and intrahepatic bile ducts. Histological evaluation showed significant hepatic fibrosis seen as early as 7 days after bile duct occlusion; it increased continuously without significant inflammatory activity and resulted in complete cirrhosis 60 to 90 days after occlusion. After the intervention, there were sharp initial increases in AP, bilirubin and ALAT, with a subsequent normalization of ALAT activity, whereas bilirubin and AP activity remained elevated (Fig. 1 and 2). Total liver hydroxyproline (Table 1) paralleled the progression of histologically verified fibrosis (r=0.94, p<0.0001). As early as 7 days after bile duct occlusion serum PIIINP was significantly higher than in the controls (p<0.001), only to increase further up to day 90 (Kruskal-Wallis ANOVA: p<0.0001). Growthrelated changes in serum PIIINP and hydroxyproline could be excluded, since in controls these parameters did not differ significantly at the beginning of the study from those at day 90 (Fig. 3). There was a sig81
B. Gerling et al. Discussion
80
140 130 120
l
110
~L
100 90
70 60
J 50
80
70-
40 n Z " 30 E
"° 60"c~ 50-
0
[]
4030-
20
20-
10
1 '~~
.
.
,
0
.
,
20
40 time
.
,
,
,
60 (days)
.
80
0
100
Fig. 3. Total liver hydroxyproline (u) and serum PIIINP (s) in rats with progressive biliary fbrosis after bile duct occlusion and total liver hydroxyproline (ll) and serum PIIIINP (O) in controls (means + SD). Different from control rats: #p
80
___
•
•, , ,
60
//
40 E,
•
•
.205x
20
•
;/
~m
•
•
-
(r =
0.85)
•
• m.m(__m_ • "
0 0
, 20
,~ 0
' 60
total hydroxyproline (gmol) Fig. 4. Correlation of serum PIIINP with total liver hydroxyproline in the 88 rats with progressive biliary fibrosis due to bile duct occlusion (r=0.85, p
nificant correlation between serum PIIINP and the histological stages of fibrosis (Table 1; r--0.80; p<0.0001), as well as between PIIINP and total liver hydroxyproline (Fig. 4; r--0.85; p<0.0001). A negative correlation was found between ALAT and PIIINE histology and hydroxyproline. However, there was no correlation between PIIINR histology or total liver hydroxyproline and bilirubin as well as AP (Table 1). 82
In the search for a rat model of progressive and reproducible liver fibrosis, we completely occluded the biliary system by double ligation of the common bile duct and retrograde injection of prolamine and ethanol. We could show that this procedure resulted in progressive biliary fibrosis with only minor inflammatory activity, as demonstrated by histology and an only moderately increased ALAT in the groups sacririced at day 2 and 7 after intervention. Significant fibrosis was seen as early as 7 days after occlusion, and complete cirrhosis occurred in all animals after 60 to 90 days. Since in previous studies in rats double ligation of the bile duct (without occlusion of the intrahepatic bile ducts by retrograde injection of prolamine/ethanol), did not result in complete cirrhosis in all animals (29), our method appears to be better suited for inducing a reproducible portal fibrosis, a model that could be of major importance for the testing of potential antifibrotic agents. In addition, the lack of significant histological inflammation and necrosis in our model might allow distinction between pure antifibrotic and anti-inflammatory, radical scavanging or membrane protecting agents. This is of major importance, since previously used models such as carbon-tetrachloride- or galactosamineinduced liver fibrosis did not lead to the discovery or validation of effective antifibrotic agents in man. Furthermore, since PIIINP levels behave similarly in rat and in man (25), this model suggests the usefulness of sequential measurements of serum PIIINP examinations in chronic human fibrotic liver disease that are devoid of major inflammatory activity. We could demonstrate that there was a significant correlation between serum PIIINP, the histological stages of fibrosis, and total liver hydroxyproline. Although we found a negative correlation of ALAT with PIIINP, total liver hydroxyproline and histological stages of fibrosis, this correlation was only due to the initial cell injury after the injection of prolamine and ethanol, resulting in an increase in ALAT observed only within the first 7 days. No correlation was found between serum PIIINP or hydroxyproline and bilirubin or alkaline phosphatase, supporting the empirical conclusions of previous reports (1,2,3,824). Numerous studies have reported elevated serum PIIINP levels in patients with hepatic fibrosis (2,3,824). However, most of these were of a cross-sectional design and liver biopsies were not or only rarely performed. A significant correlation between PIIINP in serum and the histological stage of liver fibrosis was demonstrated in patients with alcoholic fibrosis (15,19), primary biliary cirrhosis (9,18) and schisto-
PIlINP in progressive liver fibrosis in rats
somiasis of the liver (11), findings that were not confirmed by others (2,12). In addition, in severe chronic active hepatitis no correlation was found between serum PIIINP and the histological stage of fbrosis (17). These differences may be explained by the dynamic nature of chronic liver disease, with fibrogenesis being defined as the deposition of connective tissue in time, fibrolysis as the removal of connective tissue in time, whereas fibrosis, as a static term, defines the connective tissue that has accumulated over a given time (5-7). By definition, in progressive fibrosis, fibrogenesis exceeds fibrolysis. This appears to be reflected in increased PIIINP serum levels. Accordingly, in fibrosis with little fibrogenic activity, PIIINP serum concentrations have been found to be only slightly increased or even normal (24,30). Since for practical and ethical reasons, in these studies minimal fibrogenic activity could not be evaluated by sequential liver biopsies, it could not be clarified if PIIINP discriminates patients with progressive fibrosis, non-active fibrosis or fibrosis in regression (fibrolysis). The secondary biliary fibrosis of the present study was progressive with a large excess of fibrogenesis over fibrolysis. This was clearly demonstrated by the continuous increase of histologically verified fibrosis paralleled by the augmentation of total liver hydroxyproline. In conclusion, this new animal model with complete bile duct occlusion leads to progressive and reproducible liver fibrosis with little inflammation, and major similarity to human secondary biliary fibrosis, underlining its suitability for the testing of true antifibrotic agents. Since in this model serum PIIINP showed an excellent correlation with ongoing fibrogenesis and collagen accumulation, this noninvasive parameter should be used as a marker in longitudinal follow-up studies of progressive liver fibrosis in man.
4. 5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
References 1. Ruwart MJ, Wilkinson KF, Rush BD, Vidmar TJ, Peters KM, Henley KS, Appelman HD, Kim KY, Schuppan D, Hahn EG. The integrated value of serum procollagen III peptide over time predicts hepatic hydroxyproline content and stainable collagen in a model of dietary cirrhosis in the rat. Hepatology 1989; 10: 801-6. 2. Colombo M, Annoni G, Donato MF, Conte D, Martines D, Zaramella MG, Bianchi PA, Pipeono A, Tribelli C. Serum type III procollagen peptide in alcoholic liver disease and idiopathic hemochromatosis: its relationship to hepatic fibrosis, activity of the disease and iron overload. Hepatology 1985; 3: 475-9. 3. Frei A, Zimmermann A, Weigand K. The N terminal propeptide of collagen type III in serum reflects activity and degree
17.
18.
19.
20.
of fibrosis in patients with chronic liver disease. Hepatology 1984; 5: 830-4. Abdi W, Millan JC, Mezey E. Sampling variability in percutaneous liver biopsy. Arch Intern Med 1979; 139: 667-9. Schuppan D. Connective tissue polypeptides in serum as parameters to monitor antifibrotic treatment in hepatic fibrogenesis. J Hepatol 1991; 13: Suppl 3: 17-25. Schuppan D, St61zel U, Oesterling C, Somasundaram R. Serum assays for liver fibrosis. J Hepatol 1995; 22: Suppl 2: 82-8. Risteli L, Risteli J. Noninvasive methods for detection of organ fibrosis, In: Rojkind M, ed. Focus on Connective Tissue in Health and Disease. Boca Raton: CRC Press, 1990: 6198. Annoni G, Cargnel A, Colombo M, Hahn EG. Persistent elevation of the aminoterminal peptide of procollagen type III in serum of patients with acute viral hepatitis distinguishes chronic active hepatitis from resolving or chronic persistent hepatitis. J Hepatol 1986; 2: 379-88. Babbs C, Smith A, Hunt LP, Rowan B, Haboubi NY, Warnes TW. Type III procollagen peptide: a marker of disease activity and prognosis in primary biliary cirrhosis. Lancet 1988; i: 1021-4. Bentsen KD, Horn T, Risteli J, Risteli L Engstr6m-Laurent A, HCrslev-Petersen K, Lorenz I. Serum aminoterminal procollagen peptide and 7S domain of type IV collagen in patients with alcohol abuse. Liver 1987; 7: 339-46. Bentsen KD, Horslev-Petersen K, Junker P, Juhl E, Lorenzen I. Serum amino-terminal procollagen type III peptide in acute viral hepatitis. A long term follow-up study. Liver 1987; 7: 96-105. Beukers R, van Zanthen RA, Schalm SW. Serial determination of type III procollagen amino peptide serum levels in patients with histologically progressive and non-progressive primary biliary cirrhosis. J Hepatol 1992; 14: 22-9. E1-Mohandes M, Hassanein H, E1-Batrawy N, Voss B, Gerlach U. Serum concentration of N-terminal procollagen peptide of collagen type III in schistosomal liver fibrosis. Exp Mol Pathol 1986; 46: 383-90. Erikson S, Zettevall O. The N-terminal propeptide of collagen type III in serum as a prognostic indicator in primary biliary cirrhosis. J Hepatol 1986; 2: 370-8. Galambos MR, Collins DC, Galambos JT. A radioimmunoassay procedure for type III procollagen: its use for detection of hepatic fibrosis. Hepatology 1985; 1: 38-42. McCullough AJ, Stassen WN, Wiesner RH, Czaja AJ. Serial determination of the aminoterminal peptide of type III procollagen in severe chronic active hepatitis. J Lab Clin Med 1987; 1: 55-61. McCullough AJ, Strassen WN, Wiesner RH, Czaja AJ. Serum type III procollagen peptide concentrations in severe chronic active hepatitis: relationship to cirrhosis and disease activity. Hepatology 1987; 7: 49-54. Niemel~i O, Risteli L, Sotaniemi E, Stenb~ickF, Risteli J. Serum basement-membrane and type III procollagen - related antigens in primary biliary cirrhosis. J Hepatol 1988; 6: 30714. Sato S, Tashihiko N, Worner TM, Lieber CS. Liver fibrosis in alcoholics. Detection by lab radioimmunoassay of serum procollagen III peptides. JAMA 1986; 256: 1471-3. Surrenti C, Casini A, Milani S, Ambu S, Cecatelli P, D'Agata A. Is determination of serum N-terminal procollagen type III
83
B. Gerling et al.
21.
22.
23.
24.
25.
84
peptide (sPIIIP) a marker of hepatic fibrosis? Dig Dis Sci 1987; 7: 705-9. Torres-Salinas M, Pares A, Caballeria J, Jimenez W, Heredia D, Bruguera M, Rodes I. Serum procollagen type III peptide as a marker of hepatic fibrogenesis in alcoholic hepatitis. Gastroenterology 1986; 90: 1241-6. Trivedi P, Cheeseman P, Porman B, Hegarty J, Mowat AE Variation in serum type III procollagen peptide with age in healthy subjects and its comparative value in the assessment of disease activity in children and adults with chronic active hepatitis. Eur J Clin Invest 1985; 15: 69-74. Trivedi P, Risteli J, Risteli L, Tanner MS, Bhave S, Pandit AN, Mowat A.E Serum type III procollagen and basement membrane proteins as non-invasive markers of hepatic pathology in Indian childhood cirrhosis. Hepatology 1987; 6: 1249-53. Weigand K, Zaugg PY, Frei A, Zimmermann A. Long-term follow-up of serum N-terminal propeptide of collagen type III levels in patients with chronic liver disease. Hepatology 1984; 4: 835-8. Schuppan D, Dumont JM, Kim KY, Hennings G, Hahn EG. Serum concentration of the aminoterminal procollagen type III peptide in the rat reflects early formation of connective
26.
27.
28.
29.
30.
tissue in experimental liver cirrhosis. J Hepatol 1986; 3: 2737. Hayasaka A, Koch J, Schuppan D, Maddrey WC, Hahn EG. The serum concentrations of the aminoterminal propeptide of procollagen type III and the hepatic content of mRNA for the alpha 1 chain of procollagen type III in carbon tetrachloride-induced rat liver fibrogenesis. J Hepatol 1991; 13: 32838. Jamal JS, Finelli VN, Que Hee SS. A simple method to determine nanogram levels of hydroxyproline in biological tissues. Anal Biochem 1981; 112: 70-5. Rohde H, Vargas L, Hahn EG, Kalbfleisch H, Bruguera M, Timpl R. Radioimmunoassay for type III procollagen peptide and its application for human liver disease. Eur J Clin Invest 1979; 9: 451-9. Schweizer P, Schubert G.E, Reifferscheid E Das unterschiedliche morphologische Bild der Leber des Schweins und der Ratte nach Gallengangsverschlul3. Z Kinderchir 1974; 14: 61-74. Ballardini G, Faccani A, Bianchi FB, Famlani M, Patrono D, Capelli M, Pisi E. Steroid treatment lowers hepatic fibroplasia, as explored by serum aminoterminal procollagen III peptide, in chronic liver disease. Liver 1984; 4: 348-52.