- Email: [email protected]
Morphology of liver stellate cells and liver vitamin A content in 3,4,3′,4′-tetrachlorobiphenyl-treated rats
Journal of Hepatology 1991; 21: 545-553 Printed in Denmark AN rights reserved Munksgaard Copenhagen
Copyrighr 6 European Association for the Study of the Liver I997
Journal of Hepatology ISSN 0168-8278
Morphology of liver stellate cells and liver vitamin A content in 3,4,3’,4’-tetrachlorobiphenyl-treated rats Veronique Azais-Braesco ‘INRA UMMM,
‘, Marc L. Hautekeete*,
Vitamin Research Group, Clermont-Ferrand,
Isabelle Dodeman’
and Albert Geerts*
France and 2Luboratory for Cell Biology and Histology, Free University of Brussels, Brussels, Belgium
Background/Aims: Because xenobiotics decrease the vitamin A stores localized in the liver stellate cells, we investigated morphological alterations in the liver of rats exposed to 3,4,3’,4’-tetrachlorobiphenyl. Special attention was given to the morphology of the liver stellate cells and to their relationship to the liver vitamin A content. Met/&s: Six rats received an intraperitoneal injection of 3,4,3’,4’-tetrachlorobiphenyl(300 mol/kg) in soyabean oil. A further six rats received the vehicle alone. After 7 days, all rats were killed and their livers assayed for vitamin A. Liver stellate cells were examined and counted on liver sections, stained with toluidine blue or immunocytochemically for desmin and, for some animals, for a-smooth muscle actin. Results: In the livers of 3,4,3’,4’-tetrachlorobiphenyltreated rats, we found spotty and bridging necrosis, with inflammation and accumulation of desmin-positive liver stellate cells. Steatosis and mild portal inflammation were also observed. 3,4,3’,4’-Tetrachlorobiphenyl decreased the liver vitamin A content by
38%, whereas morphometric analyses showed a 40% decrease of the number of toluidine blue-detected liver stellate cells and an 11% increase of de&n-detected liver stellate cells, indicating a likely differentiation of liver stellate cells into myofibroblast-like cells 3,4,3’,4’-Tetrachlorobiphenyl treatment did not modify the expression of a-smooth muscle actin. Morphological alterations were more pronounced in periportal than in pericentral areas. The liver vitamin A content was positively correlated (r=0.56, p
V
storing cells, Ito cells or perisinusoidal cells, are localized in the space of Disse and are capable of expressing a dual phenotype (3). In normal liver, they express the quiescent phenotype. They are long-lived cells with low proliferative activity and contain characteristic lipid droplets which can concentrate more than 75% of hepatic retinoids (4). The number of these droplets and their content in retinyl esters is closely related to the vitamin A status of the animal ($6). In chronically diseased liver, LSCs acquire the activated phenotype and differentiate first into transitional and further into myofibroblast-like cells, a process whereby they lose their lipid droplets (3). Such a transition from LSCs to transitional and myofibroblast-like cells has been observed in several experimental conditions in animals, namely, following administration of carbon tetra-
A iS an eSSt?ntial IniCrOnUtrient, necessary for vision, growth and cellular differentiation. In normally nourished humans or animals, the dietary vitamin A supply is more than sufficient to fulfill the body’s needs and the excess is stored mainly in the liver. Under normal conditions, the liver vitamin A store, which represents approximately 90% of the total body reserve, consists mainly of retinyl esters, located in the lipid droplets of liver stellate cells (LSCs) (1,2). These cells, which are also known as lipocytes, fatITAMIN
Received 29 July 1996; revised 2 April: accepted 17 April 1997
Correspondence: Wronique Azais-Braesco, INRA, UMMM. Equipe Vitamines, CRNH BP 321. 63009 Clermont-Ferrand Cedex 01, France. Tel: 33-73.60.82.70. Fax: 33-73.60.82.72
Key words: Differentiation; Hepatocytes; Liver stellate cells; Myofibroblast; Vitamin A; Xenobiotic.
545
V. AzaD-Braesco
et al.
chloride (7-l l), or dimethylnitrosamine (12) and after ethanol consumption (13). A simultaneous decrease in liver vitamin A has often been reported during similar treatments (14-16). Polyhalogenated aromatic hydrocarbons, a group of widely used and persistent industrial chemicals, can cause a severe decrease in the hepatic vitamin A store of exposed animals (reviewed in 17). As an example, we have previously shown that the 3,4,3’,4’-tetrachlorobiphenyl (TCB) can induce a 40% decrease of the liver vitamin A content in rats receiving a single intraperitoneal injection of 300 pmol/kg (18). Necrotic and fibrotic lesions of the liver have been described following animal exposure to polyhalogenated hydrocarbons (19,20), but no data are available on the effect of these xenobiotics on LSCs. Although modulation of LSC phenotype and decrease in liver vitamin A seem to be closely related phenomena, few authors have addressed their relationship in vivo. The purpose of the present work was to study the relationship between the xenobiotic-induced vitamin A decrease and morphological alterations in the liver of rats exposed to a single dose of TCB, with a special emphasis on the modulation of LSC phenotype.
Materials and Methods Animals and treatment Twelve male Wistar rats were born in our laboratory from vitamin A-restricted mothers. During 1 week after weaning, they received a vitamin A-free semi-synthetic diet, in order to empty their liver vitamin A stores, while avoiding the occurrence of clinical symptoms of vitamin A deficiency Then, they received the semi-synthetic diet supplemented with 25 000 IU of vitamin A (retinyl palmitate)/kg. After 48 days, they were randomized into two groups: six rats received an i.p. injection of TCB (300 pmollkg) dissolved in soyabean oil. The six remaining rats received an i.p. injection of the vehicle (7.8 ml/kg) and were pair-fed with the treated rats. After 7 days, and following an overnight fast, all the animals were ether-anesthesized and killed by whole body perfusion with ice-cold phosphate buffer saline through the portal vein. The liver was weighed and sampled as described below. In a second experiment performed under exactly similar conditions, three rats received TCB and two received the vehicle. The protocol was designed and followed according to the usual guidelines concerning animal care. Biochemical determinations The caudal liver lobe was homogenized in ice-cold saccharose buffer (Saccharose 250 mM, EDTA 1 mM,
546
Tris-HCl 50 mM, pH 7.4), in a l/7 (w/v) ratio and aliquots were kept at - 80°C until analysis. Retinoids (retinol and retinyl esters) were measured on a hexane extract, as previously described (21) on a reverse phase HPLC column (Nucleosil 5 pm, 250x4.6 mm) eluted with pure methanol at 2 ml/min, with UV detection at 325 nm. Retinoids were quantified by external and internal calibrations (21). Toluidine blue staining Small liver pieces were fixed upon sampling by immersion for 4 h in sodium phosphate buffer (0.1 M, pH 7.4) containing 2% formaldehyde, 2% glutaraldehyde and 2% saccharose. After a 2-h post fixation in sodium phosphate buffer (0.1 M, pH 7.4) containing 2% osmium tetroxide, the samples were progressively dehydrated in a graded series of alcohol and embedded into Epon. Sections l-pm-thick were colored with toluidine blue (1% dissolved in 1% Na tetraborate buffer, pH 9.25) and mounted in Depex (Laborimpex, Belgium) Desmin staining Small blocks of liver were frozen in liquid nitrogen. Sections 10 ,um thick were treated as previously described (10). Briefly, the sections were fixed for 10 min in acetone at -20°C incubated for 15 min with 2% bovine serum albumin in phosphate buffer saline (PBS) to reduce unspecific binding and then incubated with 10 pg/ml mouse monoclonal antidesmin antibody (Boehringer, Ingelheim, Germany) for 1 h. After washing, the sections were incubated with 20 @ml of peroxidaselabeled affinity-purified goat anti-mouse IgG antibody (Tag0 Inc., Burlingame, CA, USA) for 1 h. The sections were washed and incubated for 10 min in diaminobenzidine (DAB)/H202 medium containing 20 mg DAB dissolved in 50 ml 0.14 M NaiK phosphate buffer (final pH 6.9), 1 ml of 1% CoC12 and 0.8 ml of 1% Ni(NH&(SO& and 14 ~1 of 30% H202 (22). Thereafter, sections were rinsed, dehydrated in a series of alcohol and mounted in Depex. Control experiments included the omission of the primary antibody and the use of mouse serum instead of the primary antibody. Actin staining Sections 10 pm thick were fixed and unspecific binding was blocked as described for desmin staining. Then, the primary antibody, anti-a-smooth muscle actin (monoclonal mouse antibody, Sigma, St. Louis, MO), was applied and sections were incubated overnight at 4°C. They were then incubated with horseradish peroxidase-conjugated sheep anti-mouse IgG secondary antibody (Amersham, UK) at room temperature. After rinsing, peroxidase was visualized by DAB/H202 as
Stellate cells and vitamin A in injured liver TABLE
1
Liver vitamin-A
content
Control rats TCB-treated rats
in control
and TCB-treated
Retinol
Winy1
(IU/g)
(IU/g)
(IV/g)
71*4a 152+25b
3981?358= 2345?~273~
40515359” 2498Z2Wjb
Six rats in each group. Data are expressed Different superscripts in the same column ences (ANOVA; p
esters
droplets in the parenchymal cells. Fat droplets in LSCs were not included in this analysis.
rats Total vitamin
as mean+SEM. indicate significant
A
Statistical analyses Variance analysis was performed by ANOVA, and levels of significance determined by Fisher’s test. Correlations were established by simple regressions. differ-
Results
described above. Dehydration, mounting and negative controls were carried out as described for desmin staining. Cell counts For each rat, at least three cryostat sections of different areas of the liver lobe were examined. LSCs were identified by the presence of their characteristic lipid droplets in toluidine blue-stained preparations, or by desmin staining. We determined the mean number of LSCs per mm*, both in the toluidine blue- and in the desmin-stained sections. In the toluidine blue-stained sections, we also determined the mean number of lipid droplets per LSC and the “Ito cell index” (i.e. the number of LSCs per 1000 hepatocytes) (23). Only nucleated LSCs were taken into account. Cells were counted by projecting the microscopic image on a video screen using a Zeiss Axiophot microscope and a Sony Color Video camera DXC-M2P Morphometric analysis was done by projecting a graded standard slide using the same magnification and using the Image 1.41 software (W. Rasband, National Institute of Health, USA), on a Macintosh Quadra 700 Computer. Cells were counted separately in three randomly selected periportal and pericentral areas per section, and on at least two liver lobules. Portal tracts were not considered for morphometric analysis of LSCs. The percentage of steatosis was determined in the fields examined by measuring the percentage of the microscopic field filled by lipid
TABLE
Liver vitamin A content One week after the TCB treatment, the liver vitamin A store was decreased by 38%, as indicated in Table 1. This decrease concerned only the esterified form of vitamin A: free retinol was significantly increased in treated rats. Liver histology In the livers of TCB-treated animals, various types of liver damage were found, with varying degrees of anomalies from one animal to another. Spotty necrosis accompanied by inflammation was found in all livers. Bridging necrosis was also noted in all livers, but its intensity varied from one animal to another (Fig. 1, panel B, Fig. 2, panel C) In areas of bridging necrosis, an accumulation of desmin-positive cells was seen (Fig. 2, panel C). Other features found included a mild degree of portal inflammation and macro- and microvesicular steatosis, which was massive in some livers, but only mild in others. Pathological changes were more pronounced in periportal than in centrolobular areas. Number of LSCs and of their lipid droplets Quantitative data about the number of LSCs and of their lipid droplets are summarized in Table 2 (whole lobule) and Table 3 (periportal vs pericentral areas). In TCB-treated livers, there was a 40% decrease in the number of LSCs per mm* detected by toluidine blue staining, as compared to controls, whereas the number of LSCs per mm2 detected with desmin staining was increased by 11% (Table 2). The Ito cell index (deter-
2
Morphometric
analysis
Control rats TCB-treated rats
of liver stellate cells Number of LSCs per mm* as detected by toluidine blue
Number of LSCs per mm2 as detected by desmin
Ito cell index
Mean number lipid droplets per LSC
87?2a 52Z6b
108-C3a 120k4b
109*3a 75kl3b
4.1%0.3 3.9-r-0.6
Six rats in each group. Data are meantSEM. Different superscripts in the same column indicate significant Ito cell index (i.e. the number of LSCs per 1000 hepatocytes), stained liver sections.
differences (ANOVA; p
of
per LSC were determined
% of steatosis
0.5-tO.2” 7.7k2.3b
on toluidine
blue-
547
V. Azais-Braesco
et al.
Fig. I. Rat liver sections, stained with toluidine blue. Panel A: Control rat. Panels B and C: TCB-treated bridging necrosis. Panel C: massive mucrovesicul~r steatosis. The bar represents IO pm.
548
rat. Panel B:
Stellate cells and vitamin A in injured liver
Fig. 2. Rut liver sections, immunostained for desmin. Panel A: Control rat. Panels B and C: TCB-treated LSCs in jibrosis area. Panel C: bridging necrosis in a periportal area. The bar represents 10 ,um.
rat. Panel B.
549
V. Azais-Braesco et al. TABLE
3
Number
of liver stellate cells per mm* in periportal
Control rats TCB-treated rats
and pericentral
areas
as detected by toluidine blue in periportal areas
as detected by toluidine blue in pericentral areas
as detected by desmin in periportal areas
as detected by desmin in pericentral areas
91~2~* 50t7b
84t2a 52?5b
118?1”** 132?6b*
92k4 10125
Six rats in each group. Data are mean?SEM. Different superscripts in the same column indicate significant differences (ANOVA; p
mined on toluidine blue-stained sections) decreased by 31%. Because of the difference in section thickness, no direct statistical comparison was possible of the absolute numbers of LSCs detected by two different technique, but the observed trends are clear and present in every animal. The number of lipid droplets in LSCs was not significantly altered. When periportal and pericentral areas were compared (Table 3), LSCs were found to be more numerous in periportal than in pericentral areas in the control rats. In the TCB-treated rats, the changes described above (a decrease in the number of LSCs detected in the toluidine blue-stained sections versus an increase in the number of cells detected by desmin staining) were more pronounced in periportal areas than in pericentral ones.
in different
liver areas.
In control rats, a-smooth muscle actin stained the media of blood vessel of portal tracts and central veins. In TCB-treated rats, the staining pattern was similar to that of control rats. No cx-smooth muscle actin-positive cells were detected in areas of necrosis (Fig. 3). Relationships between and the morphological (Table 4)
the vitamin A status of the rats characteristics of their livers
Regression analysis on all rats showed a positive correlation between the liver content in vitamin A and the number of toluidine blue-detected LSCs (r=0.56, p
Fig. 3. u-smooth muscle actin staining. Panel A: TCB-treated rut. Panel B: Control rat. In both cases, a-smoo lth muscle actin stains only the media of blood vessels. The bar represents 100 ,um. 550
Stellate cells and vitamin A in injured liver TABLE
4
Correlations between liver vitamin A content, % of steatosis lobule and in the periportal and pericentral areas Total vitamin
A
LSCS Tol. blue
and number
LSCS desmin
It0 cell index
of LSCs detected
by toluidine
No. of lipid % droplets steatosis
Whole lobule 1
0.56
-
_
0.53
_
-
1 _
_ 1
0.78
_
0.72 _
Ito cell index No. of lipid droplets% steatosis -
0.8 -
_
1 0.95 0.81
0.57 1 0.91
0.82 _
Periportal areas
LSCs tol. blue LSCs desmin
0.87 -
0.98 _
0.90
_ -
_ _
0.77
Pericentral
LSCs tol. blue LSCs desmin
0.67 _
0.98 _
_ _
_
_ _
0.71 _
In the whole lobule
total vitamin
A
LSCs tol. blue LSCs desmin
_
1
blue or desmin staining,
in the whole liver
LSCS Tol. blue
LSCS desmin
LSCS Tol. blue
LSCS desmin
Periportal
areas
Pericentral
areas
0.55 0.99
_ _
0.56 0.99
_ _
0.77 _ _
0.95 _ 0.48 _
0.77 _ _
0.63 0.54 _
I _
_ 1
0.96 _
_ _
0.94
_ _
1 _
1
The upper right part of the table gives correlations obtained on all animals (n= 12). The bottom left part gives correlations obtained on TCB-treated rats (n=6). Only significant correlations QKO.05) are indicated.
when regression analyses were performed on the TCBtreated group only. Free retinol in the liver was correlated to the level of steatosis.
Discussion Consistent with our previous experiments (18) and with other reports (17-23), we found that TCB treatment led to a rapid and dramatic decrease in rat liver vitamin A stores. Because vitamin A is mainly stored in the lipid droplets of LSCs, we examined the morphological changes in LSCs, which we identified by two methods: toluidine blue, which identifies LSCs by their characteristic lipid droplets, and thus allows the detection of vitamin A-rich LSCs in their “quiescent” state, and desmin immunocytochemistry, which identifies stellate cells independent of the presence of vitamin A-containing lipid droplets (9). In normal rat liver, desmin stains up to 100% of stellate cells in the periportal areas, whereas this proportion can decrease to 50% in the centrolobular areas (23,24). When LSCs differentiate into myofibroblast-like cells, desmin expression increases, and can be considered an activation marker (8,10,12,25,26). Because the transmodulation of LSCs into myofibroblasts has been associated both with an increased desmin content and with the presence of a-smooth muscle actin (27) this latter marker was also assayed. In the livers of TCB-treated rats, the number of toluidine blue-detected LSCs was lowered in the same proportion (-40%) as the vitamin A content (-38%) while the mean number of lipid droplets in each cell remained unchanged. Simultaneously, the number of LSCs identified by desmin immunocytochemistry was increased. We also found a positive correlation between .
the liver vitamin A content and the number of stellate cells identified in the toluidine blue-stained sections. These findings suggest that the TCB-induced vitamin A decrease was mainly due to differentiation of LSCs into myofibroblast-like cells, with increased desmin expression (8,10,12,27) and a loss of vitamin A-containing lipid droplets. It cannot be excluded that the increase of desmin-positive cells is also caused by the proliferation of LSCs. Despite the clear upregulation of desmin expression by LSCs in TCB-treated rats, there was no increased expression of a-smooth muscle actin. This surprising finding can perhaps be explained by the fact that asmooth muscle actin is a more advanced marker of activation of rat LSCs than desmin. Indeed, our group has recently shown that following 3 weeks of CC& administration in rats, only 20% of the total number of pericentral stellate cells express a-smooth muscle actin, whereas a large majority of the cells express desmin (A. Geerts et al., unpublished observation). This is consistent with the findings of Tanaka et al. (28), who found that cr-smooth muscle actin is only expressed in a portion of desmin-positive cells in CC&-induced liver fibrosis. Possibly, the lesions induced by a single administration of TCB are insufficient to cause increased ar-smooth muscle actin expression. Furthermore, the mechanisms of hepatocytic damage and subsequent LSC activation may be different in TCB- and CC&induced injuries. The latter are caused by the metabolizing of CCL, to the trichloromethyl radical, which damages hepatocytes extensively, leading to necrosis and release of liver enzymes into the serum (29). The precise mechanism of action of TCB, although still unknown, is probably quite different, according to the
551
V. Azai6Braesco
et al.
differences in the morphological lesions observed. Consequently, LSC activation in TCB-induced liver injury may not be as strong as that induced by CC14, resulting in a different expression of activation markers. Further studies are required to understand these various patterns of LSC activation and to determine whether cx-smooth muscle actin expression by LSCs would be induced by repeated administrations of TCB. A similar decrease in liver vitamin A content has been observed after carbon tetrachloride administration (1516) and chronic ethanol consumption (14), conditions which also lead to a decrease in the liver vitamin A content (7-l 1,13). Our findings are comparable to those of Mak et al. (30) who found a 0.79 correlation between the number of LSCs and the content in vitamin A in alcohol-injured baboon livers. We have previously found a similar correlation in human alcoholic and viral diseases (31). However, it cannot be excluded that other mechanisms contribute to the TCB-induced vitamin A decrease. Xenobiotics in general, and TCB in particular, might alter the activity of some of the enzymes involved in the uptake, metabolism and release of retinoids. Vitamin A-depleting compounds, such as ethanol, induce specific isoforms of cytochrome P-450 and other microsomal enzymes (32,33), which are able to metabolize retinol. Nilsson et al. (34) have shown that the activity of Lecithin Retino1 Acyl Transferase was diminished in LSCs isolated from dioxin-treated rats. We recently described (21) the hydrolysis into retinol of retinyl esters located in LSC lipid droplets by a Retinyl Ester Hydrolase, active at an acidic pH. We observed here an increased proportion of retinol, relative to retinyl esters. A possible induction of this enzyme activity is currently under investigation in our laboratory. In the liver of TCB-treated animals we found spotty and bridging hepatocytic necrosis. This is in agreement with the observation of Durham et al. (35) who found that from 1 to 7 days after exposure TCB distributes mainly (~80%) to hepatocytes, as compared to LSCs and endothelial and Kupffer cells. It seems therefore likely that LSC activation is triggered by products released by damaged hepatocytes and subsequently by cytokines secreted by Kupffer cells, inflammatory cells and platelets (3,36). A direct effect of TCB on LSCs cannot be excluded. It can be presumed that stellate cells accumulating in areas of necrosis differentiate into myofibroblast-like cells and are responsible for the subsequent development of fibrosis. This could explain the occurrence of fibrotic lesions that have been described after TCB administration (19,20).
552
The morphological changes observed are comparable to those induced by other agents that lead to acute liver necrosis, e.g. carbon tetrachloride (7-11) and dimethylnitrosamine (12). In contrast to what is observed following carbon tetrachloride administration, the changes that we noted in TCB-treated rats were more pronounced in periportal than in pericentral areas. This can be explained by the fact that TCB, which does not need to be metabolized to exert its toxicity, first reaches periportal hepatocytes and generates pathological alterations preferentially in this area of the liver lobule. We conclude that TCB administration leads to the differentiation of LSCs into myofibroblast-like cells, probably through an activation due to inflammatory products released by TCB-damaged hepatocytes and Kupffer or endothelial cells. Although a direct effect of TCB on hepatic vitamin A metabolism is not excluded, the decrease in the liver vitamin A content is likely to be primarily due to this differentiation of LSCS.
Acknowledgements I. Dodeman was the recipient of a CIFRE grant from Produits Roche and the ANRT (France). This work was supported by the National Fund for Scientific Research (Belgium). The authors wish to thank Carine Seynaeve for technical help and Chris Derom for the photographic work. They are grateful to Prof. L.W. Robertson for providing the TCB and for his careful reading of the manuscript.
References 1. Wake K. “Sternzellen”
2.
3.
4.
5.
6.
in the liver: perisinusoidal cells with special reference to storage of vitamin A. Am J Anat 1971; 132: 429-62. Blomhoff R, Green MH, Green JB, Norum KR. Vitamin A metabolism: new perspectives on absorption, transport and storage. Physiol Rev 1991; 71: 951-90. Geerts A, De Bleser P, Hautekeete M, Niki T, Wisse E. Fatstoring (Ito) cell biology. In: Arias IM, Boyer JL, Faust0 N, Jakoby WB, Schachter DA, Schaffritz DA, editors. The Liver: Biology and Pathobiology, 3rd ed. New York: Raven Press; 1994. p. 819-38. Hendriks HFJ, Verhoofstad WAMM, Brouwer A, DeLeeuw AM, Knook DL. Perisinusoidal fat-storing cells are the main vitamin A storage site in rat liver. Exp Cell Res 1985; 160: 13849. Ito T. Recent advances in the study on the fine structure of the hepatic sinusoidal wall: a review. Gunna Rep Med Sci 1973; 6: 119-63. Blaner WS. Retinoid (vitamin A) metabolism and the liver. In: Arias IM, Boyer JL, Faust0 N, Jakoby WB, Schachter DA, Schaffritz DA, editors. The Liver: Biology and Pathobiology, 3rd ed. New York: Raven Press; 1994. p. 52941.
Stellate cells and vitamin A in injured liver 7. Burt A, Robertson J, Heir J, MacSween R. Desmin-containing stellate cells in rat liver: distribution in normal animal and response to experimental acute liver injury. J Path01 1986; 150: 29-35. 8. Ogawa K, Suzuki J, Mukai H, Mori M. Sequential changes of extracellular matrix and proliferation of Ito cells with enhanced expression of desmin and actin in focal hepatic injury. Am J Path01 1986; 125: 611-9. 9. Yokoi Y, Namihisha T, Matsuzaki K, Miyazaki A, Yamaguchi Y. Distribution of Ito cells in experimental liver fibrosis. Liver 1988; 8: 48-52. 10. Geerts A, Lazou J, De Bleser P Wisse E. Tissue distribution, quantitation and proliferation kinetics of fat-storing cells in CC&-injured rat liver. Hepatology 1991; 13: 1193-202. 11. Schmitt-Graff A, Chakroun G, Gabbiani G. Modulation of perisinusoidal and cytoskeletal features during experimental hepatic fibrosis. Virchows Arch A Path01 Anat 1993; 422: 99107. 12. Jezequel A, Mancini R, Rinaldesi M, Ballardini G, Fallani G, Bianchi F, et al. Dimethylnitrosamine-induced cirrhosis. Evidence for an immunological mechanism. J Hepatol 1989; 11: 206-14. 13. French SW, Miyamoto K, Wong K, Jui L, Briere L. Role of the Ito cell in liver parenchymal fibrosis in rats fed alcohol and a high fat low protein diet. Am J Path01 1988; 132: 7385. 14. Sato M, Lieber CS. Hepatic vitamin A depletion after chronic ethanol consumption in baboons and rats. J Nutr 1981; 111: 2015-23. 15. Fruncillo RJ, Ruth EK, DiGregorio GJ. The effect of acute hepatic injury on the disposition of vitamin A in the rat. Res Commun Chem Path01 Pharmacol 1986; 54: 283-6. 16. Yamane M, Tanaka Y, Marumo F, Sato C. Role of hepatic vitamin A and lipocyte distribution in experimental hepatic fibrosis. Liver 1993; 13: 282-7. 17. Zile MH. Vitamin A homeostasis endangered by environmental pollutants. Proc Sot Exp Biol Med 1992; 202: 14153. 18. Azais V, Arand M, Schramm H, Rauch P, Bellenand P, Narbonne JF, et al. A time-course investigation of vitamin A levels and drug metabolizing enzymes activities in rats following a single oral treatment with prototypic polychlorinated biphenyls and DDT. Toxicology 1987; 44: 341-54. 19. Allen JR, Abrahamson LJ. Morphological and biochemical changes in the liver of rats fed chlorinated biphenyls. Arch Environ Contam Toxic01 1973; 1: 265-72. 20. Koller LD, Zinkl JG. Pathology of polychlorinated biphenyls in rabbits. Am J Path01 1973; 70: 363-73. 2 1. Azais-Braesco V, Dodeman I, Delpal S, Alexandre-Gouabau MC, Partier A, Bore1 P et al. Vitamin A contained in the lipid droplets of rat liver stellate cells is substrate for acid retinyl ester hydrolase. Biochim Biophys Acta 1995; 1259: 271-6.
22. Adams J. Heavy metal intensification of diaminobenzidinebased horseradish peroxidase reaction product. J Histochem Cytochem 1981; 29: 775. 23. Horwath E, Kovacs K, Ross RC. Liver ultrastructure in methotrexate treatment of psoriasis. Arch Dermatol 1973; 108: 427-8. 24. Chen LC, Berberian I, Koch B, Mercier M, Azdis-Braesco V, Glauert HP, et al. Polychlorinated and polybrominated biphenyl congeners and retinoid levels in rat tissues: structureactivity relationships. Toxic01 Appl Pharmacol 1992; 114: 4755. 25. Ballardini G, Groff P, DeGiorgi LB, Schuppan D, Bianchi FB. Ito cell heterogeneity: desmin-negative Ito cells in normal rat liver. Hepatology 1994; 19: 440-6. 26. Niki T, De Bleser PJ, Xu G, Van den Berg K. Comparison of glial fibrillary acidic protein and desmin staining in normal and in CCL-induced fibrotic rat livers. Hepatology 1996; 23: 153845. 27. Ballardini G, Fallani M, Biagini G, Bianchi EB, Pisi E. Desmin and actin in the identification of Ito cells and monitoring their evolution to myofibroblasts in experimental liver fibrosis. Virchows Arch B Cell Path01 1988; 56: 45-9. 28. Tanaka Y, Nouchi T, Yamane M, Irie T, Miyakawa H, Sato C, et al. Phenotypic modulation in lipocytes in experimental liver fibrosis. J Path01 1991; 164: 273-8. 29. Parkinson A. Biotransformation of xenobiotics. In: Klaassen CD, editor. Casarett & Doull’s Toxicology. The Basic Science of Poisons, 5th ed. New York: McGraw-Hill; 1996. p. 12059. 30. Mak KM, Leo MA, Lieber CS. Alcoholic liver injury in baboons: transformation of lipocytes to transitional cells. Gastroenterology 1984; 87: 188-200. 31. Hautekeete M, Dodeman I, Geerts A., Azdis-Braesco V, Seynaeve C, Lazou JM, et al. Correlation between number of fat-storing cells (FSC), mean number of lipid droplets, and liver vitamin A content in human liver disease. In: Wisse E, Knook DL, Wake K, editors. Cells of the Hepatic Sinusoid, vol 5. Leiden: Kupffer Cell Foundation; 1995. p. 410-2. 32. Leo AM, Lowe N, Lieber CS. Decreased hepatic vitamin A after drug administration in men and in rats. Am J Clin Nutr 1984; 40: 1131-6. 33. Leo AM, Lowe N, Lieber CS. Interaction of drugs and retinol. Biochem Pharmacol 1986; 35: 3949-53. 34. Nilsson C, Hanberg A, Trossvik C, Hakansson H. 2,3,7,8tetrachlorodibenzo-p-dioxin affects retinol esterification in rat hepatic stellate cells and kidney. Environ Toxic01 Pharmaco1 1996; 2: 17-23. 35. Durham SK, Brouwer A. 3,4,3’,4’-Tetrachlorobiphenylinduced effects in the rat liver: II: microscopic autoradiographic localization of 3H-TCB. Toxic01 Path01 1989; 17: 782-8. 36. Gressner AM. Perisinusoidal lipocytes and fibrogenesis. Gut 1994; 35: 1331-3.
553
Copyrighr 6 European Association for the Study of the Liver I997
Journal of Hepatology ISSN 0168-8278
Morphology of liver stellate cells and liver vitamin A content in 3,4,3’,4’-tetrachlorobiphenyl-treated rats Veronique Azais-Braesco ‘INRA UMMM,
‘, Marc L. Hautekeete*,
Vitamin Research Group, Clermont-Ferrand,
Isabelle Dodeman’
and Albert Geerts*
France and 2Luboratory for Cell Biology and Histology, Free University of Brussels, Brussels, Belgium
Background/Aims: Because xenobiotics decrease the vitamin A stores localized in the liver stellate cells, we investigated morphological alterations in the liver of rats exposed to 3,4,3’,4’-tetrachlorobiphenyl. Special attention was given to the morphology of the liver stellate cells and to their relationship to the liver vitamin A content. Met/&s: Six rats received an intraperitoneal injection of 3,4,3’,4’-tetrachlorobiphenyl(300 mol/kg) in soyabean oil. A further six rats received the vehicle alone. After 7 days, all rats were killed and their livers assayed for vitamin A. Liver stellate cells were examined and counted on liver sections, stained with toluidine blue or immunocytochemically for desmin and, for some animals, for a-smooth muscle actin. Results: In the livers of 3,4,3’,4’-tetrachlorobiphenyltreated rats, we found spotty and bridging necrosis, with inflammation and accumulation of desmin-positive liver stellate cells. Steatosis and mild portal inflammation were also observed. 3,4,3’,4’-Tetrachlorobiphenyl decreased the liver vitamin A content by
38%, whereas morphometric analyses showed a 40% decrease of the number of toluidine blue-detected liver stellate cells and an 11% increase of de&n-detected liver stellate cells, indicating a likely differentiation of liver stellate cells into myofibroblast-like cells 3,4,3’,4’-Tetrachlorobiphenyl treatment did not modify the expression of a-smooth muscle actin. Morphological alterations were more pronounced in periportal than in pericentral areas. The liver vitamin A content was positively correlated (r=0.56, p
V
storing cells, Ito cells or perisinusoidal cells, are localized in the space of Disse and are capable of expressing a dual phenotype (3). In normal liver, they express the quiescent phenotype. They are long-lived cells with low proliferative activity and contain characteristic lipid droplets which can concentrate more than 75% of hepatic retinoids (4). The number of these droplets and their content in retinyl esters is closely related to the vitamin A status of the animal ($6). In chronically diseased liver, LSCs acquire the activated phenotype and differentiate first into transitional and further into myofibroblast-like cells, a process whereby they lose their lipid droplets (3). Such a transition from LSCs to transitional and myofibroblast-like cells has been observed in several experimental conditions in animals, namely, following administration of carbon tetra-
A iS an eSSt?ntial IniCrOnUtrient, necessary for vision, growth and cellular differentiation. In normally nourished humans or animals, the dietary vitamin A supply is more than sufficient to fulfill the body’s needs and the excess is stored mainly in the liver. Under normal conditions, the liver vitamin A store, which represents approximately 90% of the total body reserve, consists mainly of retinyl esters, located in the lipid droplets of liver stellate cells (LSCs) (1,2). These cells, which are also known as lipocytes, fatITAMIN
Received 29 July 1996; revised 2 April: accepted 17 April 1997
Correspondence: Wronique Azais-Braesco, INRA, UMMM. Equipe Vitamines, CRNH BP 321. 63009 Clermont-Ferrand Cedex 01, France. Tel: 33-73.60.82.70. Fax: 33-73.60.82.72
Key words: Differentiation; Hepatocytes; Liver stellate cells; Myofibroblast; Vitamin A; Xenobiotic.
545
V. AzaD-Braesco
et al.
chloride (7-l l), or dimethylnitrosamine (12) and after ethanol consumption (13). A simultaneous decrease in liver vitamin A has often been reported during similar treatments (14-16). Polyhalogenated aromatic hydrocarbons, a group of widely used and persistent industrial chemicals, can cause a severe decrease in the hepatic vitamin A store of exposed animals (reviewed in 17). As an example, we have previously shown that the 3,4,3’,4’-tetrachlorobiphenyl (TCB) can induce a 40% decrease of the liver vitamin A content in rats receiving a single intraperitoneal injection of 300 pmol/kg (18). Necrotic and fibrotic lesions of the liver have been described following animal exposure to polyhalogenated hydrocarbons (19,20), but no data are available on the effect of these xenobiotics on LSCs. Although modulation of LSC phenotype and decrease in liver vitamin A seem to be closely related phenomena, few authors have addressed their relationship in vivo. The purpose of the present work was to study the relationship between the xenobiotic-induced vitamin A decrease and morphological alterations in the liver of rats exposed to a single dose of TCB, with a special emphasis on the modulation of LSC phenotype.
Materials and Methods Animals and treatment Twelve male Wistar rats were born in our laboratory from vitamin A-restricted mothers. During 1 week after weaning, they received a vitamin A-free semi-synthetic diet, in order to empty their liver vitamin A stores, while avoiding the occurrence of clinical symptoms of vitamin A deficiency Then, they received the semi-synthetic diet supplemented with 25 000 IU of vitamin A (retinyl palmitate)/kg. After 48 days, they were randomized into two groups: six rats received an i.p. injection of TCB (300 pmollkg) dissolved in soyabean oil. The six remaining rats received an i.p. injection of the vehicle (7.8 ml/kg) and were pair-fed with the treated rats. After 7 days, and following an overnight fast, all the animals were ether-anesthesized and killed by whole body perfusion with ice-cold phosphate buffer saline through the portal vein. The liver was weighed and sampled as described below. In a second experiment performed under exactly similar conditions, three rats received TCB and two received the vehicle. The protocol was designed and followed according to the usual guidelines concerning animal care. Biochemical determinations The caudal liver lobe was homogenized in ice-cold saccharose buffer (Saccharose 250 mM, EDTA 1 mM,
546
Tris-HCl 50 mM, pH 7.4), in a l/7 (w/v) ratio and aliquots were kept at - 80°C until analysis. Retinoids (retinol and retinyl esters) were measured on a hexane extract, as previously described (21) on a reverse phase HPLC column (Nucleosil 5 pm, 250x4.6 mm) eluted with pure methanol at 2 ml/min, with UV detection at 325 nm. Retinoids were quantified by external and internal calibrations (21). Toluidine blue staining Small liver pieces were fixed upon sampling by immersion for 4 h in sodium phosphate buffer (0.1 M, pH 7.4) containing 2% formaldehyde, 2% glutaraldehyde and 2% saccharose. After a 2-h post fixation in sodium phosphate buffer (0.1 M, pH 7.4) containing 2% osmium tetroxide, the samples were progressively dehydrated in a graded series of alcohol and embedded into Epon. Sections l-pm-thick were colored with toluidine blue (1% dissolved in 1% Na tetraborate buffer, pH 9.25) and mounted in Depex (Laborimpex, Belgium) Desmin staining Small blocks of liver were frozen in liquid nitrogen. Sections 10 ,um thick were treated as previously described (10). Briefly, the sections were fixed for 10 min in acetone at -20°C incubated for 15 min with 2% bovine serum albumin in phosphate buffer saline (PBS) to reduce unspecific binding and then incubated with 10 pg/ml mouse monoclonal antidesmin antibody (Boehringer, Ingelheim, Germany) for 1 h. After washing, the sections were incubated with 20 @ml of peroxidaselabeled affinity-purified goat anti-mouse IgG antibody (Tag0 Inc., Burlingame, CA, USA) for 1 h. The sections were washed and incubated for 10 min in diaminobenzidine (DAB)/H202 medium containing 20 mg DAB dissolved in 50 ml 0.14 M NaiK phosphate buffer (final pH 6.9), 1 ml of 1% CoC12 and 0.8 ml of 1% Ni(NH&(SO& and 14 ~1 of 30% H202 (22). Thereafter, sections were rinsed, dehydrated in a series of alcohol and mounted in Depex. Control experiments included the omission of the primary antibody and the use of mouse serum instead of the primary antibody. Actin staining Sections 10 pm thick were fixed and unspecific binding was blocked as described for desmin staining. Then, the primary antibody, anti-a-smooth muscle actin (monoclonal mouse antibody, Sigma, St. Louis, MO), was applied and sections were incubated overnight at 4°C. They were then incubated with horseradish peroxidase-conjugated sheep anti-mouse IgG secondary antibody (Amersham, UK) at room temperature. After rinsing, peroxidase was visualized by DAB/H202 as
Stellate cells and vitamin A in injured liver TABLE
1
Liver vitamin-A
content
Control rats TCB-treated rats
in control
and TCB-treated
Retinol
Winy1
(IU/g)
(IU/g)
(IV/g)
71*4a 152+25b
3981?358= 2345?~273~
40515359” 2498Z2Wjb
Six rats in each group. Data are expressed Different superscripts in the same column ences (ANOVA; p
esters
droplets in the parenchymal cells. Fat droplets in LSCs were not included in this analysis.
rats Total vitamin
as mean+SEM. indicate significant
A
Statistical analyses Variance analysis was performed by ANOVA, and levels of significance determined by Fisher’s test. Correlations were established by simple regressions. differ-
Results
described above. Dehydration, mounting and negative controls were carried out as described for desmin staining. Cell counts For each rat, at least three cryostat sections of different areas of the liver lobe were examined. LSCs were identified by the presence of their characteristic lipid droplets in toluidine blue-stained preparations, or by desmin staining. We determined the mean number of LSCs per mm*, both in the toluidine blue- and in the desmin-stained sections. In the toluidine blue-stained sections, we also determined the mean number of lipid droplets per LSC and the “Ito cell index” (i.e. the number of LSCs per 1000 hepatocytes) (23). Only nucleated LSCs were taken into account. Cells were counted by projecting the microscopic image on a video screen using a Zeiss Axiophot microscope and a Sony Color Video camera DXC-M2P Morphometric analysis was done by projecting a graded standard slide using the same magnification and using the Image 1.41 software (W. Rasband, National Institute of Health, USA), on a Macintosh Quadra 700 Computer. Cells were counted separately in three randomly selected periportal and pericentral areas per section, and on at least two liver lobules. Portal tracts were not considered for morphometric analysis of LSCs. The percentage of steatosis was determined in the fields examined by measuring the percentage of the microscopic field filled by lipid
TABLE
Liver vitamin A content One week after the TCB treatment, the liver vitamin A store was decreased by 38%, as indicated in Table 1. This decrease concerned only the esterified form of vitamin A: free retinol was significantly increased in treated rats. Liver histology In the livers of TCB-treated animals, various types of liver damage were found, with varying degrees of anomalies from one animal to another. Spotty necrosis accompanied by inflammation was found in all livers. Bridging necrosis was also noted in all livers, but its intensity varied from one animal to another (Fig. 1, panel B, Fig. 2, panel C) In areas of bridging necrosis, an accumulation of desmin-positive cells was seen (Fig. 2, panel C). Other features found included a mild degree of portal inflammation and macro- and microvesicular steatosis, which was massive in some livers, but only mild in others. Pathological changes were more pronounced in periportal than in centrolobular areas. Number of LSCs and of their lipid droplets Quantitative data about the number of LSCs and of their lipid droplets are summarized in Table 2 (whole lobule) and Table 3 (periportal vs pericentral areas). In TCB-treated livers, there was a 40% decrease in the number of LSCs per mm* detected by toluidine blue staining, as compared to controls, whereas the number of LSCs per mm2 detected with desmin staining was increased by 11% (Table 2). The Ito cell index (deter-
2
Morphometric
analysis
Control rats TCB-treated rats
of liver stellate cells Number of LSCs per mm* as detected by toluidine blue
Number of LSCs per mm2 as detected by desmin
Ito cell index
Mean number lipid droplets per LSC
87?2a 52Z6b
108-C3a 120k4b
109*3a 75kl3b
4.1%0.3 3.9-r-0.6
Six rats in each group. Data are meantSEM. Different superscripts in the same column indicate significant Ito cell index (i.e. the number of LSCs per 1000 hepatocytes), stained liver sections.
differences (ANOVA; p
of
per LSC were determined
% of steatosis
0.5-tO.2” 7.7k2.3b
on toluidine
blue-
547
V. Azais-Braesco
et al.
Fig. I. Rat liver sections, stained with toluidine blue. Panel A: Control rat. Panels B and C: TCB-treated bridging necrosis. Panel C: massive mucrovesicul~r steatosis. The bar represents IO pm.
548
rat. Panel B:
Stellate cells and vitamin A in injured liver
Fig. 2. Rut liver sections, immunostained for desmin. Panel A: Control rat. Panels B and C: TCB-treated LSCs in jibrosis area. Panel C: bridging necrosis in a periportal area. The bar represents 10 ,um.
rat. Panel B.
549
V. Azais-Braesco et al. TABLE
3
Number
of liver stellate cells per mm* in periportal
Control rats TCB-treated rats
and pericentral
areas
as detected by toluidine blue in periportal areas
as detected by toluidine blue in pericentral areas
as detected by desmin in periportal areas
as detected by desmin in pericentral areas
91~2~* 50t7b
84t2a 52?5b
118?1”** 132?6b*
92k4 10125
Six rats in each group. Data are mean?SEM. Different superscripts in the same column indicate significant differences (ANOVA; p
mined on toluidine blue-stained sections) decreased by 31%. Because of the difference in section thickness, no direct statistical comparison was possible of the absolute numbers of LSCs detected by two different technique, but the observed trends are clear and present in every animal. The number of lipid droplets in LSCs was not significantly altered. When periportal and pericentral areas were compared (Table 3), LSCs were found to be more numerous in periportal than in pericentral areas in the control rats. In the TCB-treated rats, the changes described above (a decrease in the number of LSCs detected in the toluidine blue-stained sections versus an increase in the number of cells detected by desmin staining) were more pronounced in periportal areas than in pericentral ones.
in different
liver areas.
In control rats, a-smooth muscle actin stained the media of blood vessel of portal tracts and central veins. In TCB-treated rats, the staining pattern was similar to that of control rats. No cx-smooth muscle actin-positive cells were detected in areas of necrosis (Fig. 3). Relationships between and the morphological (Table 4)
the vitamin A status of the rats characteristics of their livers
Regression analysis on all rats showed a positive correlation between the liver content in vitamin A and the number of toluidine blue-detected LSCs (r=0.56, p
Fig. 3. u-smooth muscle actin staining. Panel A: TCB-treated rut. Panel B: Control rat. In both cases, a-smoo lth muscle actin stains only the media of blood vessels. The bar represents 100 ,um. 550
Stellate cells and vitamin A in injured liver TABLE
4
Correlations between liver vitamin A content, % of steatosis lobule and in the periportal and pericentral areas Total vitamin
A
LSCS Tol. blue
and number
LSCS desmin
It0 cell index
of LSCs detected
by toluidine
No. of lipid % droplets steatosis
Whole lobule 1
0.56
-
_
0.53
_
-
1 _
_ 1
0.78
_
0.72 _
Ito cell index No. of lipid droplets% steatosis -
0.8 -
_
1 0.95 0.81
0.57 1 0.91
0.82 _
Periportal areas
LSCs tol. blue LSCs desmin
0.87 -
0.98 _
0.90
_ -
_ _
0.77
Pericentral
LSCs tol. blue LSCs desmin
0.67 _
0.98 _
_ _
_
_ _
0.71 _
In the whole lobule
total vitamin
A
LSCs tol. blue LSCs desmin
_
1
blue or desmin staining,
in the whole liver
LSCS Tol. blue
LSCS desmin
LSCS Tol. blue
LSCS desmin
Periportal
areas
Pericentral
areas
0.55 0.99
_ _
0.56 0.99
_ _
0.77 _ _
0.95 _ 0.48 _
0.77 _ _
0.63 0.54 _
I _
_ 1
0.96 _
_ _
0.94
_ _
1 _
1
The upper right part of the table gives correlations obtained on all animals (n= 12). The bottom left part gives correlations obtained on TCB-treated rats (n=6). Only significant correlations QKO.05) are indicated.
when regression analyses were performed on the TCBtreated group only. Free retinol in the liver was correlated to the level of steatosis.
Discussion Consistent with our previous experiments (18) and with other reports (17-23), we found that TCB treatment led to a rapid and dramatic decrease in rat liver vitamin A stores. Because vitamin A is mainly stored in the lipid droplets of LSCs, we examined the morphological changes in LSCs, which we identified by two methods: toluidine blue, which identifies LSCs by their characteristic lipid droplets, and thus allows the detection of vitamin A-rich LSCs in their “quiescent” state, and desmin immunocytochemistry, which identifies stellate cells independent of the presence of vitamin A-containing lipid droplets (9). In normal rat liver, desmin stains up to 100% of stellate cells in the periportal areas, whereas this proportion can decrease to 50% in the centrolobular areas (23,24). When LSCs differentiate into myofibroblast-like cells, desmin expression increases, and can be considered an activation marker (8,10,12,25,26). Because the transmodulation of LSCs into myofibroblasts has been associated both with an increased desmin content and with the presence of a-smooth muscle actin (27) this latter marker was also assayed. In the livers of TCB-treated rats, the number of toluidine blue-detected LSCs was lowered in the same proportion (-40%) as the vitamin A content (-38%) while the mean number of lipid droplets in each cell remained unchanged. Simultaneously, the number of LSCs identified by desmin immunocytochemistry was increased. We also found a positive correlation between .
the liver vitamin A content and the number of stellate cells identified in the toluidine blue-stained sections. These findings suggest that the TCB-induced vitamin A decrease was mainly due to differentiation of LSCs into myofibroblast-like cells, with increased desmin expression (8,10,12,27) and a loss of vitamin A-containing lipid droplets. It cannot be excluded that the increase of desmin-positive cells is also caused by the proliferation of LSCs. Despite the clear upregulation of desmin expression by LSCs in TCB-treated rats, there was no increased expression of a-smooth muscle actin. This surprising finding can perhaps be explained by the fact that asmooth muscle actin is a more advanced marker of activation of rat LSCs than desmin. Indeed, our group has recently shown that following 3 weeks of CC& administration in rats, only 20% of the total number of pericentral stellate cells express a-smooth muscle actin, whereas a large majority of the cells express desmin (A. Geerts et al., unpublished observation). This is consistent with the findings of Tanaka et al. (28), who found that cr-smooth muscle actin is only expressed in a portion of desmin-positive cells in CC&-induced liver fibrosis. Possibly, the lesions induced by a single administration of TCB are insufficient to cause increased ar-smooth muscle actin expression. Furthermore, the mechanisms of hepatocytic damage and subsequent LSC activation may be different in TCB- and CC&induced injuries. The latter are caused by the metabolizing of CCL, to the trichloromethyl radical, which damages hepatocytes extensively, leading to necrosis and release of liver enzymes into the serum (29). The precise mechanism of action of TCB, although still unknown, is probably quite different, according to the
551
V. Azai6Braesco
et al.
differences in the morphological lesions observed. Consequently, LSC activation in TCB-induced liver injury may not be as strong as that induced by CC14, resulting in a different expression of activation markers. Further studies are required to understand these various patterns of LSC activation and to determine whether cx-smooth muscle actin expression by LSCs would be induced by repeated administrations of TCB. A similar decrease in liver vitamin A content has been observed after carbon tetrachloride administration (1516) and chronic ethanol consumption (14), conditions which also lead to a decrease in the liver vitamin A content (7-l 1,13). Our findings are comparable to those of Mak et al. (30) who found a 0.79 correlation between the number of LSCs and the content in vitamin A in alcohol-injured baboon livers. We have previously found a similar correlation in human alcoholic and viral diseases (31). However, it cannot be excluded that other mechanisms contribute to the TCB-induced vitamin A decrease. Xenobiotics in general, and TCB in particular, might alter the activity of some of the enzymes involved in the uptake, metabolism and release of retinoids. Vitamin A-depleting compounds, such as ethanol, induce specific isoforms of cytochrome P-450 and other microsomal enzymes (32,33), which are able to metabolize retinol. Nilsson et al. (34) have shown that the activity of Lecithin Retino1 Acyl Transferase was diminished in LSCs isolated from dioxin-treated rats. We recently described (21) the hydrolysis into retinol of retinyl esters located in LSC lipid droplets by a Retinyl Ester Hydrolase, active at an acidic pH. We observed here an increased proportion of retinol, relative to retinyl esters. A possible induction of this enzyme activity is currently under investigation in our laboratory. In the liver of TCB-treated animals we found spotty and bridging hepatocytic necrosis. This is in agreement with the observation of Durham et al. (35) who found that from 1 to 7 days after exposure TCB distributes mainly (~80%) to hepatocytes, as compared to LSCs and endothelial and Kupffer cells. It seems therefore likely that LSC activation is triggered by products released by damaged hepatocytes and subsequently by cytokines secreted by Kupffer cells, inflammatory cells and platelets (3,36). A direct effect of TCB on LSCs cannot be excluded. It can be presumed that stellate cells accumulating in areas of necrosis differentiate into myofibroblast-like cells and are responsible for the subsequent development of fibrosis. This could explain the occurrence of fibrotic lesions that have been described after TCB administration (19,20).
552
The morphological changes observed are comparable to those induced by other agents that lead to acute liver necrosis, e.g. carbon tetrachloride (7-11) and dimethylnitrosamine (12). In contrast to what is observed following carbon tetrachloride administration, the changes that we noted in TCB-treated rats were more pronounced in periportal than in pericentral areas. This can be explained by the fact that TCB, which does not need to be metabolized to exert its toxicity, first reaches periportal hepatocytes and generates pathological alterations preferentially in this area of the liver lobule. We conclude that TCB administration leads to the differentiation of LSCs into myofibroblast-like cells, probably through an activation due to inflammatory products released by TCB-damaged hepatocytes and Kupffer or endothelial cells. Although a direct effect of TCB on hepatic vitamin A metabolism is not excluded, the decrease in the liver vitamin A content is likely to be primarily due to this differentiation of LSCS.
Acknowledgements I. Dodeman was the recipient of a CIFRE grant from Produits Roche and the ANRT (France). This work was supported by the National Fund for Scientific Research (Belgium). The authors wish to thank Carine Seynaeve for technical help and Chris Derom for the photographic work. They are grateful to Prof. L.W. Robertson for providing the TCB and for his careful reading of the manuscript.
References 1. Wake K. “Sternzellen”
2.
3.
4.
5.
6.
in the liver: perisinusoidal cells with special reference to storage of vitamin A. Am J Anat 1971; 132: 429-62. Blomhoff R, Green MH, Green JB, Norum KR. Vitamin A metabolism: new perspectives on absorption, transport and storage. Physiol Rev 1991; 71: 951-90. Geerts A, De Bleser P, Hautekeete M, Niki T, Wisse E. Fatstoring (Ito) cell biology. In: Arias IM, Boyer JL, Faust0 N, Jakoby WB, Schachter DA, Schaffritz DA, editors. The Liver: Biology and Pathobiology, 3rd ed. New York: Raven Press; 1994. p. 819-38. Hendriks HFJ, Verhoofstad WAMM, Brouwer A, DeLeeuw AM, Knook DL. Perisinusoidal fat-storing cells are the main vitamin A storage site in rat liver. Exp Cell Res 1985; 160: 13849. Ito T. Recent advances in the study on the fine structure of the hepatic sinusoidal wall: a review. Gunna Rep Med Sci 1973; 6: 119-63. Blaner WS. Retinoid (vitamin A) metabolism and the liver. In: Arias IM, Boyer JL, Faust0 N, Jakoby WB, Schachter DA, Schaffritz DA, editors. The Liver: Biology and Pathobiology, 3rd ed. New York: Raven Press; 1994. p. 52941.
Stellate cells and vitamin A in injured liver 7. Burt A, Robertson J, Heir J, MacSween R. Desmin-containing stellate cells in rat liver: distribution in normal animal and response to experimental acute liver injury. J Path01 1986; 150: 29-35. 8. Ogawa K, Suzuki J, Mukai H, Mori M. Sequential changes of extracellular matrix and proliferation of Ito cells with enhanced expression of desmin and actin in focal hepatic injury. Am J Path01 1986; 125: 611-9. 9. Yokoi Y, Namihisha T, Matsuzaki K, Miyazaki A, Yamaguchi Y. Distribution of Ito cells in experimental liver fibrosis. Liver 1988; 8: 48-52. 10. Geerts A, Lazou J, De Bleser P Wisse E. Tissue distribution, quantitation and proliferation kinetics of fat-storing cells in CC&-injured rat liver. Hepatology 1991; 13: 1193-202. 11. Schmitt-Graff A, Chakroun G, Gabbiani G. Modulation of perisinusoidal and cytoskeletal features during experimental hepatic fibrosis. Virchows Arch A Path01 Anat 1993; 422: 99107. 12. Jezequel A, Mancini R, Rinaldesi M, Ballardini G, Fallani G, Bianchi F, et al. Dimethylnitrosamine-induced cirrhosis. Evidence for an immunological mechanism. J Hepatol 1989; 11: 206-14. 13. French SW, Miyamoto K, Wong K, Jui L, Briere L. Role of the Ito cell in liver parenchymal fibrosis in rats fed alcohol and a high fat low protein diet. Am J Path01 1988; 132: 7385. 14. Sato M, Lieber CS. Hepatic vitamin A depletion after chronic ethanol consumption in baboons and rats. J Nutr 1981; 111: 2015-23. 15. Fruncillo RJ, Ruth EK, DiGregorio GJ. The effect of acute hepatic injury on the disposition of vitamin A in the rat. Res Commun Chem Path01 Pharmacol 1986; 54: 283-6. 16. Yamane M, Tanaka Y, Marumo F, Sato C. Role of hepatic vitamin A and lipocyte distribution in experimental hepatic fibrosis. Liver 1993; 13: 282-7. 17. Zile MH. Vitamin A homeostasis endangered by environmental pollutants. Proc Sot Exp Biol Med 1992; 202: 14153. 18. Azais V, Arand M, Schramm H, Rauch P, Bellenand P, Narbonne JF, et al. A time-course investigation of vitamin A levels and drug metabolizing enzymes activities in rats following a single oral treatment with prototypic polychlorinated biphenyls and DDT. Toxicology 1987; 44: 341-54. 19. Allen JR, Abrahamson LJ. Morphological and biochemical changes in the liver of rats fed chlorinated biphenyls. Arch Environ Contam Toxic01 1973; 1: 265-72. 20. Koller LD, Zinkl JG. Pathology of polychlorinated biphenyls in rabbits. Am J Path01 1973; 70: 363-73. 2 1. Azais-Braesco V, Dodeman I, Delpal S, Alexandre-Gouabau MC, Partier A, Bore1 P et al. Vitamin A contained in the lipid droplets of rat liver stellate cells is substrate for acid retinyl ester hydrolase. Biochim Biophys Acta 1995; 1259: 271-6.
22. Adams J. Heavy metal intensification of diaminobenzidinebased horseradish peroxidase reaction product. J Histochem Cytochem 1981; 29: 775. 23. Horwath E, Kovacs K, Ross RC. Liver ultrastructure in methotrexate treatment of psoriasis. Arch Dermatol 1973; 108: 427-8. 24. Chen LC, Berberian I, Koch B, Mercier M, Azdis-Braesco V, Glauert HP, et al. Polychlorinated and polybrominated biphenyl congeners and retinoid levels in rat tissues: structureactivity relationships. Toxic01 Appl Pharmacol 1992; 114: 4755. 25. Ballardini G, Groff P, DeGiorgi LB, Schuppan D, Bianchi FB. Ito cell heterogeneity: desmin-negative Ito cells in normal rat liver. Hepatology 1994; 19: 440-6. 26. Niki T, De Bleser PJ, Xu G, Van den Berg K. Comparison of glial fibrillary acidic protein and desmin staining in normal and in CCL-induced fibrotic rat livers. Hepatology 1996; 23: 153845. 27. Ballardini G, Fallani M, Biagini G, Bianchi EB, Pisi E. Desmin and actin in the identification of Ito cells and monitoring their evolution to myofibroblasts in experimental liver fibrosis. Virchows Arch B Cell Path01 1988; 56: 45-9. 28. Tanaka Y, Nouchi T, Yamane M, Irie T, Miyakawa H, Sato C, et al. Phenotypic modulation in lipocytes in experimental liver fibrosis. J Path01 1991; 164: 273-8. 29. Parkinson A. Biotransformation of xenobiotics. In: Klaassen CD, editor. Casarett & Doull’s Toxicology. The Basic Science of Poisons, 5th ed. New York: McGraw-Hill; 1996. p. 12059. 30. Mak KM, Leo MA, Lieber CS. Alcoholic liver injury in baboons: transformation of lipocytes to transitional cells. Gastroenterology 1984; 87: 188-200. 31. Hautekeete M, Dodeman I, Geerts A., Azdis-Braesco V, Seynaeve C, Lazou JM, et al. Correlation between number of fat-storing cells (FSC), mean number of lipid droplets, and liver vitamin A content in human liver disease. In: Wisse E, Knook DL, Wake K, editors. Cells of the Hepatic Sinusoid, vol 5. Leiden: Kupffer Cell Foundation; 1995. p. 410-2. 32. Leo AM, Lowe N, Lieber CS. Decreased hepatic vitamin A after drug administration in men and in rats. Am J Clin Nutr 1984; 40: 1131-6. 33. Leo AM, Lowe N, Lieber CS. Interaction of drugs and retinol. Biochem Pharmacol 1986; 35: 3949-53. 34. Nilsson C, Hanberg A, Trossvik C, Hakansson H. 2,3,7,8tetrachlorodibenzo-p-dioxin affects retinol esterification in rat hepatic stellate cells and kidney. Environ Toxic01 Pharmaco1 1996; 2: 17-23. 35. Durham SK, Brouwer A. 3,4,3’,4’-Tetrachlorobiphenylinduced effects in the rat liver: II: microscopic autoradiographic localization of 3H-TCB. Toxic01 Path01 1989; 17: 782-8. 36. Gressner AM. Perisinusoidal lipocytes and fibrogenesis. Gut 1994; 35: 1331-3.
553