Transplantation of fetal liver tissue suspension into the spleens ofadult syngenic rats: Effects of β-naphthoflavone, Phenobarbital and dexamethasone on cytochrome P450 isoforms expression and on glycogen storage

Exp ToxIc Pathol 1998: 50: 173- 183 Gustav Fischer Verlag

'Institute of Pharmacology and Toxicology, 2Department of Anatomy I, Friedrich Schiller University Jena, Germany; 3Department of Physiology, University of Kuopio, Finland

Transplantation of fetal liver tissue suspension into the spleens of adult syngenic rats: Effects of ~-naphthoflavone, phenobarbital and dexamethasone on cytochrome P450 isoforms expression and on glycogen storage A. Lupp', N. LUCAS l , P. LINDSTROM-SEPPA" K. KOPONEN\

o. HANNINEN\ M. DANZ2, and W. KLINGER I

With 4 figures and 3 tables Received: January 16, 1998; Accepted: January 27,1998 Address for correspondence: Dr. med. A. Lupp, Institute of Pharmacology and Toxicology, Friedrich Schiller University Jena, Lobderstr. 1,07740 Jena. Germany. Key words: Fetal liver suspensions, transplantation; Transplantation, fetal liver suspensions; Liver, fetal, suspensions; Spleen, transplantation into; ~-Naphthoflavone; Phenobarbital; Dexamethasone; Cytochrome P450 isoforms; Glycogen storage; Hepatocytes, fetal. Abbreviations: BNF: ~-naphthoflavone; b.wt.: body weight; BSA: bovine serum albumin; DEX: dexamethasone; DMSO: dimethylsulfoxide; HE: hematoxylin and eosin; P450: cytochrome P450; PAS: periodic acid-Schiff; PB: phenobarbital; PBS: phosphate buffered saline.

Summary In the present study, the effect of ~-naphthoflavone (BNF), phenobarbital (PB) and dexamethasone (DEX) on the expression of three cytochrome P450 (P450) isoforms, 1A I. 2B I and 3A2, and on glycogen storage was investigated in intrasplenic liver cell explants in comparison to adult liver. Fetal liver tissue suspensions were transplanted into the spleens of adult male syngenic Fisher inbred rats. Four months after surgery. transplant recipients and age matched controls were orally treated with BNF (1 x 50 mg/kg body weight (b.wt.», PB (I x 50 mg/kg b.wL), DEX (for 3 days 4 mg/kg b.wl. per day), or the respective solvents (dimethylsulfoxide or 0.9% NaCl). The animals were sacrificed 24 (BNF, DEX) or 48 (PB) hours after the last treatment. The livers of both solvent treated transplant recipients and control rats displayed only in few liver lobules a slight P450 tAl. but in all lobules a strong P450 2Bl and 3A2 expression, which was all mainly located in the hepatocytes around the central veins (zone ITT, according to Rappaport). After BNF administration a P450 I A 1expression was induced in the hepatocytes of the peripheral regions of the liver lobules (zone L according to Rappaport), whereas the staining of the hepatocytes around the central veins disappeared. Also the staining for P450 2B 1 in the hepatocytes of zone III became slightly more pronounced. Following PB treatment the

P450 1A 1 expression in the hepatocytes of the central regions (zone III), as seen in few lobules after solvent treatment only, was reduced, whereas the staining for P450 2B I and 3A2 was more pronounced in the hepatocytes of the intermedial and central regions of the liver lobules (zone II and III). D EX treatment diminished P450 I A I and 2B 1 expression within the livers of both transplant recipients and control rats. In contrast. the staining for P450 3A2 was enhanced in all regions of the liver lobules. Transplantation of fetal liver tissue suspensions into the spleens did not influence the inducibi lity of P450 isoforms expression within the respective livers of the animals. Spleens of control rats displayed no P450 isoforms expression without as well as with induction. In the explant containing spleens, however, similar to normal liver, the transplanted hepatocytes displayed nearly no P450 I AI, but a strong P450 2B 1 and 3A2 expression. After BNF treatment a staining for P450 I A I was induced and also the P450 2B 1 expression was slightly more pronounced. PB treatment caused an increase in the staining for P450 2B 1 and 3A2 and DEX administration for P450 3A2 within the transplanted hepatocytes. Additionally, after DEX treatment some bile ducts of the explants displayed a slight staining for P450 IA 1, 2B 1 and 3A2. All hepatocytes within the livers of both solvent treated transplant recipients and control rats displayed a slightly Exp Toxic Pathol50 (1998) 3

173

PAS-positive cytoplasm a and, in most cases, homogeneously distributed, fine-grained, strongly PAS-stained granules indicating glycogen storage. No regional variance in the glycogen content of the hepatocytes was seen within the liver lobules, but there was a marked difference between the individual hepatocytes of the same lobular region in the extent of glycogen accumulation. The hepatocytes within the explants displayed the same type of glycogen storage as did the adult liver cells. BNF treatment did not display any effect on the glycogen accumulation in livers and intrasplenic liver cell explants. After PB administration, only in livers, but not in the transplants, the glycogen content in the hepatocytes around the central veins was slightly reduced. DEX treatment lead to an excessive storage of fat within the hepatocytes of both livers and spleens. Thus, the glycogen was displaced, leading to a "spoke-wheel" like pattern of glycogen storage. Additionally, within the hepatocytes of both livers and liver cell explants a higher amount of glycogen seemed to be stored and the granules appeared to be more coarse-grained. These results demonstrate that transplanted liver cells originating from syngenic fetal liver tissue suspensions are able to survive in host organs like the spleen at least for 4 months. Similar to normal adult liver cells, the transplanted hepatocytes display a P4S0 isoforms expression, which can be induced by treatment with BNF, PB or DEX. Additionally, they are able to store glycogen.

Introduction Up to now only few studies on the biochemical function of fetal hepatocytes after transplantation into a host organ like the spleen have been conducted, mainly by the group of KATO and coworkers (KATO et al. 1992, 1994a, 1994b, 1996a, 1996b, 1997). These studies focussed on the developmental expression of different cytochrome P4S0 (P4S0) SUbtypes and their inducibility by 70% hepatectomy, dexamethasone (DEX) or hepatocyte growth factor using immunoblot analysis or immunohistochemistry. We recently reported about the developmental expression of three cytochrome P4S0 (P4S0) isoforms, 1AI, 2B 1 and 3A2, in intrasplenic fetal liver cell explants and the ability of these cells to store glycogen in comparison to adult and fetal liver (Lupp et al. 1998). In this study we had been able to demonstrate that the transplanted liver cells can survive in the spleens for at least 1 year, which is in line with other data from literature (see e.g. KUSANO and MITO 1982; KOKUDO et al. 1995). The transplanted fetal hepatocytes proliferated and differentiated, with a variable developmental decline in P450 lAl and an increase in P450 2B 1 and 3A2 subtypes expression. After differentiation, the P450 isoforms expression within the transplanted hepatocytes exhibited the same pattern as in the respective cells of normal adult liver (very low P450 lAl, high P450 2Bl and 3A2 subtypes expression). Additionally, like normal adult liver cells, the transplanted hepatocytes were able to store glycogen. The aim of the present study was to gain more information about the P450 subtypes which are constitutionally expressed and/or which could be induced within the 174

Exp Toxic PatholSO (1998) 3

hepatocytes transplanted into the spleens in comparison to normal adult liver. For that purpose we investigated the effects of three well known inducers of different P450 SUbtypes (GIBSON and SKETT 1994; CORREIA 1995), [3naphthoflavone (BNF; P450 lA), phenobarbital (PB; P450 2A, 2B, 2C, 3A) and DEX (P450 3A) on the expression of three P450 isoforms, lAl, 2Bl and 3A2, in spleens 4 months after intrasplenic transplantation of a syngenic fetal liver tissue suspension. Additionally, the influence of the three inducers on glycogen storage within the transplanted hepatocytes was examined. Effects were compared to those seen in liver. A possible influence of the transplant on the intact liver of the transplant recipients was also elucidated.

Material and methods Animals Fisher 344 inbred rats from our own institute's breed were used. The animals were housed in plastic cages unter standardized conditions (light-dark cycle 12/12 h, temperature 22 ± 2°C, humidity SO ± 10 %, pellet diet Altromin 1316, water ad libitum). Donor-fetuses were taken from pregnant Fisher inbred rats at the 21 st day of gestation. The fetal livers were immediately removed, pooled and minced by razor blades in 4°C cold Hank's balanced salt solution (I: 1 w/v) until a homogenous suspension was obtained. Recipients were syngenic 60-90 days old male rats. After laparotomy, spleens of the recipients were injected 0.2 ml of the fetal liver tissue suspension in longitudinal direction. Subsequently the abdomen was closed again with two sutures on top of each other. Age matched control animals received no surgery. No immunosuppressant was given to the transplant recipients.

Induction experiments Rats which had received a transplantation of a syngenic fetal liver tissue suspension 4 months ago and age matched control rats were divided into four groups as follows (n = 6 per group and treatment): 1) transplant recipients treated with BNF, PB or DEX; 2) transplant recipients treated with the solvents (dimethylsulfoxide (DMSO) or 0.9 % NaCl); 3) control rats treated with BNF, PB or DEX; 4) control rats treated with the solvents. BNF was given once orally at a dosage of SO mg/kg body weight (b.wt.), dissolved in 2.S ml/kg b.wt. DMSO. Solvent treated rats received once orally 2.S ml/kg b.wt. DMSO. Animals were sacrificed in ether anesthesia 24 hours after the treatment. PB was given once orally at a dosage of SO mg/kg b.wt., dissolved in 2.S ml/kg b.wt. DMSO. Solvent treated rats received once orally 2.S ml/kg b.wt. DMSO. Animals were sacrificed in ether anesthesia 48 hours after the treatment. DEX was given orally in the morning of three consecutive days at a dosage of 4 mg/kg b.wt., dissolved in 2.0 ml/kg b.wt. 0.9 % NaCl. Solvent treated rats received three times orally 2.0 ml/kg b.wt. 0.9% NaCl. Animals were sacrificed in ether anesthesia 24 hours after the last treatment.

Histology After sacrifice the livers and spleens were removed instantly, weighed and fixed in 10% buffered formaldehyde. The middle parts of the spleens were cut into 3 mm sections (5 per spleen), which were embedded together in one paraffin block. Similarly, from the respective livers two 3 mm sections extending from the hilus to the margin of the left lateral lobe were placed in one block. Subsequently 5 11m sections were prepared from the paraffin blocks. Paraffin sections of each block were deparaffinated and stained with hematoxylin and eosin (HE) or by means of the periodic acid-Schiff (PAS) procedure for glycogen. For immunohistochemistry the sections were deparaffinated and hydrated during which they were additionally incubated in 0.5 % H)O)/methanol for 45 min in order to block endogenous peroxida~e activity. Staining ofP450 1AI. 2B I and 3A2 was performed by an indirect peroxidase labelling method according to SMOLOWITZ et al. (1991). The sections were first incubated in normal rabbit serum to block nonspecific attachment of the secondary antibody (rabbit anti-goat IgG) and then kept overnight in an optimal dilution of goat anti-rat P450 I AI, 2B 1. or 3A2 (Daiichi Pure Chemicals Co., Tokyo, Japan) in 1 % bovine serum albumin (BSA)lphosphate buffered saline (PBS). The sections were then incubated in rabbit anti-goat IgG and peroxidase labelled nonspecific goat IgG. The colour was developed by incubation in 3-amino-9-ethylcarbazole in acetate buffer. Each incubation was followed by washes in I % BSA/PBS. The sections were then rinsed, counterstained with Mayer's hematoxylin, and mounted in glycerol. As controls. sections of each block were stained with the same procedure but using normal goat serum instead of goat anti-rat P450 antibodies.

Statistics The explant size and the percentage of bile ducts were estimated by means of the "intersection counting procedure" using a net plate within the ocular of the microscope. The 5 different sections of the spleen of each animal were evaluated and a mean value per spleen of one animal was calculated. These data served for further statistics on the different experimental groups. The animals investigated per group comprised n = 6. The results are expressed as arithmetic means ± S.E.M. For statistical analysis the Mann-Whitney test (p ::; 0.05) was applied.

higher (about 16 %; data not shown). Spleen weights of both groups of animals were not affected by BNF administration. PB treatment caused a significant increase in liver weights of both transplanted as well as control rats by about 15 % (fig. 1). No significant differences were seen between the liver weights of controls and transplanted animals. The spleen weights of transplanted rats were significantly higher than those of the controls without as well as with PB treatment (see above; fig. I). There was only a slight (and statistically not significant) increase in the weights of the transplant containing spleens due to PB treatment (fig. 1). DEX treatment for three consecutive days led to a significant and marked decrease in body weights by about 12 % of the original weights with both control and transplanted rats in comparison to solvent treated animals. Sol5

i

.c

Cl

liver

4

-

*

I

.c

020

:l ":~

S

UI

025

~

*

C) C)

030

spleen

J

2

Cl

c

j

~ 0

0

control

tr;lnsp!antatlon

() 15

'.j, ..

010

.,~

'j ~ IU

:j .::!

:i

I

:

.,.,,

:

""ntrol

005 000

Fig. 1. Liver and spleen weights of control rats and of transplant recipients after administration of the solvent dimethylsulfoxide (0) or of phenobarbital (1El) 4 months after surgery, Ordinate scale: Organ wet weights expressed in g related to 100 g body weight. Data are given as arithmetic means ± S.E.M. Asterisks mark statistical significant differences from the values of the solvent treated rats (*p::; 0.05; Mann-Whitney test; n = 6 for each group).

i

6

.c

5

C) C)

4

Cl

....

S

UI

liver

*

Ji -~-

"

0.4

spleen

*

~

~

-----

, '-, ..

.c

';

03

*

-~

*

02

Cl 2

Results Body and organ weights After BNF treatment no significant effects on relative and absolute liver weights of both transplant recipients and control rats were observed (data not shown). Additionally, also no significant differences were seen between the liver weights of both groups of animals. Compared to control rats, the relative and absolute spleen weights of transplant recipients were significantly

C

IU

...Cl0

o~ 1

0-0-:-:O"ltrol

transplantatIon

control

transplantatIon

00

Fig. 2. Liver and spleen weights of control rats and of transplant recipients after administration of the solvent 0.9% NaCI (0) or of dexamethasone (I""l) 4 months after surgery. Ordinate scale: Organ wet weights expressed in g related to 100 g body weight. Data are given as arithmetic means ± S.E.M. Asterisks mark statistical significant differences from the values of the solvent treated rats (*p::; 0.05; MannWhitney test; n = 6 for each group). Exp Toxic Pathol 50 (1998) 3

175

:I~ Fig. 3. Localization of cytochrome P450 (P450) lAI (a; b) and 2Bl (c; d) within transplant containing spleens 4 months after surgery. a: P450 IAI expression following treatment with the solvent dimethylsulfoxide; b: P450 lAI expression after administration of 13-naphthoflavone; c: P450 2BI expression following treatment with the solvent dimethylsulfoxide; d: P450 2Bl expression after administration of phenobarbital. Immunohistochemistry, counterstaining with HE; original magnification: x 400 (b); x 200 (a; c; d). 176

Exp Toxic Patho150 (1998) 3

~

cld

Fig. 4. Localization of cytochrome P450 (P450) 3A2 (a; b) and of glycogen (c; d) within transplant containing spleens 4 months after surgery. a: P450 3A2 expression following treatment with the solvent 0.9% NaCI; b: P450 3A2 expression after administration of dexamethasone; c: glycogen storage following treatment with the solvent 0.9% NaCI; d: glycogen storage after administration of dexamethasone. a and b: Immunohistochemistry, counterstaining with HE; c and d: PAS; original magnification: x 200 (a; b); x 100 (c; d). Exp Toxic Pathol 50 (1998) 3

177

vent treated rats of both groups of animals displayed no significant changes in body weights during this time period (data not shown). The relative liver weights of both transplanted as well as control rats significantly increased by about 36 % (fig. 2) and the absolute liver weights by about 28 % after DEX treatment. Compared to rats without transplantation the relative and absolute spleen weights were significantly higher in transplant recipients. With both groups of animals the relative and absolute spleen weights were decreased by about 22 % or 37 %, respectively, due to DEX treatment, the spleen weights of the transplanted rats still being significantly higher than those of the controls (fig. 2). Additionally, DEX lead to a significant reduction in relative weights of adrenal glands and of thymus glands of both groups of animals by about 18 % and 49 %, respectively (data not shown).

Histological appearance Four months after transplantation of syngenic fetal liver tissue suspensions into the spleens of adult rats, the intrasplenic explants of solvent treated transplant recipients comprised about 41 % of the total spleen mass (table 1). The transplanted hepatocytes were mostly organized into liver-typical cord structures, but also single cells were visible (fig. 3a, 3c, 4a, 4c). Besides the hepatocyte masses, and in nearly all cases distinctly separated from the hepatocytes, big bulks of bile ducts were observed which comprised even about 19 % of the explants (table I). All the bile ducts displayed noticeable large lumina, but in some cases the bile duct cells were atrophic and the ducts were surrounded by fibrotic material (figs. 3a, 3c, 4a, 4c). In contrast to the hepatocytes, which were settling only in the red pulp around the spleen follicles, some bile ducts also expanded into the white pulp. After BNF treatment no effects were seen on liver morphology of both transplant recipients and control rats. Additionally, there was no influence of BNF on size or morphology (fig. 3b) nor on percentage ofhepatocytes or bile ducts within the intrasplenic explants (table 1). PB administration caused an increase in size of the liver lobules and an enlargement of the hepatocytes within the livers of both transplant recipients and control rats. Additionally, quite often mitoses were seen within the hepatocytes. With the intrasplenic transplants a significant extension of the explant size was observed after PB treatment (table 1), which was mainly due to a significant increase in the absolute number of bile ducts. Additionally, also the percentage of bile ducts within the intrasplenic explants was enhanced (table I). This effect of PB on the bile ducts of the explants was not observed with the bile ducts in the livers. Also a slight increase in the absolute number of the intrasplenic hepatocytes and an enlargement of these cells was seen. Similar to liver, quite often mitoses were noticed in the hepatocytes of the explants. 178

Exp Toxic Pathol 50 (1998) 3

Table 1. Size of the intrasplenic explants and percentage of bile ducts 4 months after transplantation of fetal liver tissue suspensions into the spleens of adult syngenic rats and after pretreatment of the rats with the solvents DMSO or 0.9 % NaCI or with BNF, PB or DEX before sacrifice. Data are given as arithmetic means ± S.E.M. Treatment Explant [% of spleen tissue]

Bile ducts [% of explant]

n

DMSO 40.43 ± 1.90 0.9% NaCl 42.63 ± 4 .10 BNF 38.45 ± 4.45 PB 51.56 ± 3.50 * DEX 54.37 ± 5.51

21.40 ± 2.04 17.11±3.81 21.64 ± 4.35 28.03 ± 0.33 * 28.53 ± 3.05 *

6 6 6 6 6

*: significant difference from the values of the respective solvent treated rats (Mann-Whitney test; p :::::; 0.05).

After DEX treatment a noticeable fatty degeneration and inflammation was observed in the hepatocytes of the livers in both groups of rats. Additionally, following DEX administration in the livers of both transplant recipients and control rats many round areas with necrotic or degenerating hepatocytes were noticed mainly located in the intermedial part of the liver lobules (zone II; not shown), probably representing a peliosis hepatis, which is known to occur after e.g. steroid hormone treatment (BAGHERI and BOYER 1974; STANG-VOSS and ApPELL 1981; VAN ERPECUM et al. 1988; BALAZS 1990. With the spleens of both transplant recipients and control rats DEX treatment caused a remarkable reduction in specific spleen tissue, thus leading to a (though not significantly) enhanced relative explant size within the transplant containing spleens (table 1). Similar to liver, also within the transplanted hepatocytes an excessive fat storage was observed after DEX administration (figs. 4b, 4d). Following DEX treatment only within the explants, but not within the livers, an increase in the absolute and relative number of bile ducts was seen (table O. Since there was no "real" enlargement of the explant size, but an increase in the amount of bile ducts, even a decrease in the absolute number of hepatocytes has to be assumed.

P450 isoforms expression Within the livers of both solvent treated transplant recipients 4 months after surgery and solvent treated age matched control rats, nearly no P450 lAl expression was detectable (table 2). Only in few liver lobules a mild to moderate staining for P450 1A 1 mainly in the hepatocytes around the central veins (zone III) was seen. In contrast, a strong P450 2B 1 and 3A2 isoforms expression predominantly in the centrilobular hepatocytes (zone III) was observed (table 2). Marked differences in the extent of the expression of these P450 SUbtypes were not only seen be-

'" ...,

rn

\D

.....,

w

\D \D 00

01

=

2-

S-

~

'"tl

n '"

0

"Cl

Hepatocytes (zone I) Hepatocytes (zone II) Hepatocytes (zone III) Bile ducts

Hepatocytes (zone I) Hepatocytes (zone II) Hepatocytes (zone III) Bile ducts

Hepatocytes (zone I) Hepatocytes (zone II) Hepatocytes (zone III) Bile ducts

lAl

2BI

3A2 +++ +++ +++

+++ +++ +++

+/++

Occurrence

-/+ +/++ ++

-/+ + ++

+1++

Intensity

DMSOIO.9% NaCI

+++ +++ +++

+++ +++ +++

+++ +

Occurrence

BNF

-/+ + ++

-/+ + ++/+++

++ +

Intensity

+++ +++ +++

+++ +++ +++

Occurrence

PB

-/+ +/++ ++/+++

-/+ ++ ++/+++

Intensity

+++ +++ +++

+++ +++ +++

Occurrence

DEX

+ ++ ++/+++

-/+ -/+ +

Intensity

Hepatocytes Bile ducts Spleen tissue ++

++

-/+

+/++/+++

+/++

-/+

++

++

++

Occurrence

Occurrence

Intensity

BNF

DMSO/0.9% NaCI

+/++/+++

+/++1+++

+/++/+++

Intensity

++

++

Occurrence

PB

++/+++

++1+++

Intensity

Occurrence: - negative; + seldom; ++ frequent; +++ diffuse. Intensity: - no staining: -/+ very mild; + mild; ++ moderate; +++ strong .

3A2

Hepatocytes Bile ducts Spleen tissue

Hepatocytes Bile ducts Spleen tissue

IAI

2BI

Cell type

P450 isoform

++ ++

++ ++

++

Occurrence

DEX

++/+++ -/+

+1++/+++ +

+/++

Intensity

Table 3. P450 IAI, 2BI and 3A2 expression in intrasplenic liver cell explants 4 months after transplantation of fetal liver tissue suspensions into the spleens of adult syngenic rats and after pretreatment of the animals with the solvents DMSO or 0.9% NaCl or with BNF, PB or DEX before sacrifice.

Occurrence: - negative; + seldom; ++ frequent; +++ diffuse. Intensity: -no staining; -/+ very mild: + mild; ++ moderate: +++ strong.

Cell type

P450 isoform

Table 2. P450 IAI, 2B I and 3A2 expression in the livers of adult rats after pretreatment of the animals with the solvents DMSO or 0.9% NaCI or with BNF, PB or DEX before sacrifice.

tween the different regions of the liver lobules, but also between the individual hepatocytes within one region. Also within the individual cell bodies the expression was not homogeneous. No P450 isoforms expression was observed in bile duct cells. Between the livers of transplant recipients and control rats no differences were visible in the staining for the three P450 sUbtypes investigated. Similarly, in the hepatocytes of the intrasplenic explants nearly no P450 IAI, but a strong P450 2BI and 3A2 expression was seen after treatment with the solvents (figs. 3a, 3c, 4a; table 3). The extent of staining for P450 2B I and 3A2 in the transplanted hepatocytes showed marked differences between the individual cells ranging from very mild to very strong (table 3). Additionally, as in liver tissue, within the individual cell bodies the staining was quite inhomogeneous. The bile ducts within the intrasplenic explants showed no P450 isoforms expression after treatment with the solvents only (table 3). After BNF treatment, a P450 IAI expression appeared within the livers of both transplant recipients and control rats, which was mainly located in the hepatocytes of the peripheral region of the liver lobules (zone I; table 2). In contrast, the staining for P450 lAI of the centrilobular hepatocytes, observed in few lobules after treatment with the solvents only (see above), was not present any more. The P450 2B 1 expression in the hepatocytes around the central veins became more pronounced than after treatment with the solvent, whereas the staining for P450 3A2 seemed to be slightly reduced (table 2). With bile ducts no P450 isoforms expression was seen after BNF treatment as it was the case after treatment with the solvents only (see above). No differences were seen between the livers of transplant recipients and control rats. With the intrasplenic explants, BNF treatment caused a P450 lAI expression in some of the hepatocytes, but there were marked differences in the extent of expression between the individual cells (fig. 3b; table 3). The P450 2B I expression was slightly more pronounced following BNF administration and, additionally, the differences in the staining between the individual cells seemed to be slightly reduced (table 3). Concerning the P450 3A2 expression in the transplanted hepatocytes, however, no differences were seen between BNF or solvent treated rats (table 3). In the bile duct cells of the explants also after BNF treatment no P450 isoforms expression was observed. After PB treatment the slight P450 lAI expression disappeared within the livers of both transplant recipients and control rats (table 2). In contrast, there was a more pronounced staining for P450 2B 1 and 3A2 in the intermedial and central regions of the liver lobules (zone II and III; table 2). Also after treatment with PB, the bile ducts displayed no P450 isoforms expression. There was no difference between the livers of both transplant recipients and control rats in the P450 isoforms expression after PB treatment. Similar as in livers, the intrasplenically transplanted hepatocytes did not express P450 IAI after PB treatment 180

Exp Toxic Pathol50 (1998) 3

and there was an increase in the staining for P450 2B I (fig. 3d; table 3) and also 3A2 (table 3). Additionally, the P450 2B 1 and 3A2 expression within the individual hepatocytes became more homogeneous. No P450 isoforms expression was seen within the bile duct cells of the explants (table 3). DEX treatment diminished P450 lA 1 and 2B I expression in the livers of both transplant recipients and control rats (table 2). In contrast, the staining for P450 3A2 was similarly enhanced in all regions of the liver lobules (table 2). In the bile duct cells no P450 isoforms expression was visible. No difference was seen in the staining for the P450 SUbtypes after DEX treatment between the livers of transplant recipients and control rats. Also within the intrasplenic explants no P450 1A 1 expression was observed after DEX administration (table 3). No influence of the treatment was seen on the staining for P450 2BI (table 3). In contrast, there was an increase in the P450 3A2 expression (fig. 4b; table 3). Additionally, the staining within and between the transplanted hepatocytes for this P450 isoform was more homogeneous in comparison to the treatment with the solvent only. With some bile ducts within the explants a mild to moderate P450 IAI and a mild or very mild P450 2B 1 or 3A2 expression, respectively, appeared after DEX treatment (table 3). Spleens of control rats displayed no P450 isoforms expression wi thout as well as after treatment with any of the inducers.

Glycogen storage All hepatocytes within the livers of both solvent treated transplant recipients and control rats displayed a slightly PAS-positive cytoplasma and, in most cases, homogeneously distributed fine-grained strongly stained granules. No regional variance in the glycogen content of the hepatocytes was seen within the liver lobules (zone I-III), but there was a marked difference between the individual hepatocytes of the same lobular region in tpe extent of glycogen storage. In bile duct cells, no glycogen storage was visible. No differences in the glycogen content were observed between the livers of transplant recipients and control rats. The intrasplenically transplanted hepatocytes displayed the same type of glycogen storage as the adult liver cells (fig. 4c). As it was also the case with the normal liver cells, the transplanted hepatocytes displayed pronounced individual differences in their glycogen content. No glycogen storage was seen in the bile duct cells of the explants. There was no influence of BNF administration on glycogen content within the livers of transplant recipients and control rats. Similarly, also within the intrasplenic liver cell explants BNF treatment did not display any effect on glycogen storage (not shown). After PB treatment, the glycogen content in the centrilobular hepatocytes of the livers of both transplant recipi-

ents and control rats was slightly reduced in a similar manner. but there was no visible influence on the glycogen storage of the intrasplenicall.v transplanted hepafocyfes (not shown). After DEX treatment massive changes were observed: due to the excessive storage of fat within the hepatocytes of the livers of both transplant recipients and control rats and also within the hepatocytes of the intraspienic liver cell explant.~, the glycogen was displaced and thus accumulated at the margins of the cell bodies. around the nuclei and also around the fat droplets. leading to a "spoke-wheel" like pattern of glycogen storage (fig. 4d). Additionally. within the hepatocytes of both livers and liver cell explants, a higher amount of glycogen seemed to be stored and the granules appeared to be more coarsegrained.

Discussion The present study was conducted in consequence of previous investigations on P450 isoforms expression in fetal and adult livers and in intrasplenic fetal liver cell explants at different times after transplantation (Lupp et al. 1998). In these previous investigations we were able to demonstrate that, whereas in spleens of control or sham operated rats no P450 isoforms expression was detectable, after transplantation significant activities were present fro m 3 days after surgery on. The transplanted hepatocytes displayed a variable developmental decline in P450 I Al and an increase in P450 2B I and 3A2 SUbtypes expression. After differentiation. the transplanted hepatocytes exhibited the same pattern of P450 isoforms expression as the respective cells from normal adult liver. Additionally, like normal adult liver, the transplanted hepatocytes were able to store glycogen. The aim of the present study was to gain more information about the P450 subtypes which are constitutionally expressed andlor which can be induced within the hepatocytes transplanted into the spleens in comparison to normal liver. For this purpose, the effects of BNF. PB or DEX, inducing different P450 subtypes, on the expression of three P450 isoforms, IA I , 2B I and 3A2. were investigated 4 months after transplantation. Additionally. the influence of the three inducers on glycogen storage in the transplanted hepatocytes was investigated in comparison to normal liver. As we had already seen previously (Lupp et a1. 1998). weights of transplant containing spleens were significantly higher in comparison to those of control spleens. This increase in weights is obviously due to the hepatocytes residing and multiplying within these spleens. Similar to our previous investigations (Lupp et al. 1999), at the time point investigated (4 months after surgery) the explants comprised already about 41 % of the total spleen mas s and 19 % of the explants consisted of big bulks of bile ducts with large lumina and in some cases already atrophic bile duct cells. As expected, liver weights of both transplanted and

control rats were significantly increased by PB treatment. PB displays a mitogenic effect on hepatocytes thus increasing their multiplication. On the other hand PB also causes an expansion of the endoplasmic reticulum within the cells thus leading to an enlargement of the cell bodies. Both effects concomitantly lead to an augmentation in liver mass (BOHM and MOSER 1976; NISHIKAWA et a1. 1987; KAST et a1. 1988). These effects of PB were also seen within the intrasplenic explants, where an increase in the number and size of hepatocytes (but also an increase in the amount of bile ducts) and thus an enhancement in explant size was observed after PB treatment. The only minor effect of PB on the weights of the transplant containing spleens may be explained by the fact that the percentage of explants within the spleens is not high enough to detect an increase in cell mass by about 15 % as observed with the livers after PB induction. DEX treated rats displayed a marked decrease in body weights, probably due to the catabolic metabolism, in comparison to solvent treated animals. Relative as well as absolute liver weights were markedly increased. This enhancement in weight may mainly be due to the remarkable fatty degeneration observed in these livers. This excessive fat storage was also seen within the intrasplenic explants. Additionally, within the intrasplenic explants an increase in the number of bile ducts was seen after DEX treatment. The stimulus for this proliferation ofbile ducts may be an amplification of an already preexisting cholestasis within the explants (although not histologically visible). An additional cholestatic effect, known for steroid hormones (LINDBERG 1992; DESCOTES et a1. 1996; ERUNGER 1997), may increase the already existing problems of the transplanted liver cells with the excretion of the bile they produce since the explants are not connected to the ductus choledochus and thus have to increte their bile (much slower) into blood. As to be expected, after DEX treatment weights of adrenal glands and, due to the immunosuppressive action of DEX, weights of lymphatic organs like thymus glands and spleens were also strongly reduced, the weights of transplant containing spleens still being significantly higher than those of the control organs. The percentage of reduction in spleen weights after DEX administration was about the same in transplant containing and in control organs. again possibly due to the fact that the increase in the amount of transplanted liver cells within the transplant containing spleens is not high enough to counteract the strong decrease in weight of the spleen mass by about 37 %. Thus. the observed increase in the explant size (which is expressed in % of the total spleen mass) after DEX treatment must also be attributed to the massive decrease in specific spleen tissue in addi tion to a "real" enlargement of the explants due to the excessive fat storage within the hepatocytes or to the augmentation in the number of bile ducts. In line with data from literature (see e.g. WOLF et al. 1984; BARON et al. 1984; RATANASAVANH et al. 1991) and with our previous results (Lupp et a1. 1998), also in the present investigation the normal adult livers displayed Exp Toxic Patho] 50 ( 1998) 3

181

almost no P450 lAl, but a strong P450 2Bl and 3A2 isoforms expression, which was all mainly located in the hepatocytes around the central veins of the liver lobules (zone III). Corresponding to our previous findings with transplanted liver cells 4 months after surgery, in the present study the intrasplenically transplanted hepatocytes displayed nearly no P450 lAl, but a strong P450 2B 1 and 3A2 isoforms expression. BNF is known to predominantly induce P450 lA isoforms in rats, whereas PB mainly induces P450 2A, 2B, 2C and 3A, and DEX P450 predominantly 3A SUbtypes (GIBSON and SKETI 1994; CORREIA 1995). Thus, as to be expected, in the livers of both control rats and transplant recipients as well as in the intrasplenic liver cell explants P450 1A I (but also 2B I) expression was induced after BNF administration, P450 2B 1 and 3A2 by PB treatment and P450 3A2 after DEX administration. In contrast, a slight reduction in the staining for P450 3A2 was seen after BNF treatment, ofP450 lAI after PB, and ofP450 IAI and 2B 1 after DEX administration. This may possibly be due to a competitive inhibition of the synthesis of these P450 isoforms because of the excessive synthesis of the induced P450 SUbtypes. Similar observations for the treatment with DEX have already been described in the literature for rainbow trout liver and hepatocytes with respect to the P450 lAI subtype (LEE et al. 1993; DASMAHAPTRA et al. 1993). After DEX administration, the bile ducts within the explants displayed a P450 1A I, 2B 1 and 3A2 expression, which was not present after treatment with the solvents only. This P450 SUbtypes expression may be explained by the excessive proliferation of the bile ducts within the explants after DEX treatment, since P450 isoforms expression in bile ducts correlates with their proliferative activity as we have already shown in our previous investigations (Lupp et al. 1998). Like in our previous investigations (Lupp et a1. 1998), in the livers of the transplant recipients and control rats all the hepatocytes contained glycogen and no zonal differences of the glycogen storage were observed within the liver lobules, since also in this investigation the animals were sacrificed at 8.00 h a.m., known to be the time of the maximal glycogen storage (FREDRIKS et al. 1987). A possible regional variance in the glycogen content at later times, when consumption of glycogen has already started, was not investigated. Also corresponding to our previous findings with transplanted liver cells at 4 months after surgery (Lupp et al. 1998), in the present study the intrasplenically transplanted hepatocytes displayed the same type of glycogen storage as the adult liver cells. In line with data from literature (N ISHIKAWA et al. 1987; KAST et al. 1988), also in the present study the glycogen storage in the hepatocytes around the central veins of the liver lobules was slightly reduced after treatment with PB. In the intrasplenic explants, however, no influence of PB administration was visible. Since there were marked differences in the extent of glycogen storage between the individual hepatocytes with no obvious morphological 182

Exp Toxic Pathol 50 (1998) 3

regularity, a reduction, if present in individual hepatocytes, was too small to be detected by light microscopy. After DEX treatment an increase in glycogen content was observed with both livers and intrasplenic hepatocytes. This glycogenic effect on liver hepatocytes is well known for glucocorticoids. The results of the present investigation demonstrate that transplanted liver cells originating from syngenic fetal liver tissue suspensions are able to survive in host organs like the spleen at least for several months. The transplanted hepatocytes display a P450 isoforms expression and the ability to store glycogen, which both can be influenced by treatment with BNF, PB or DEX, like with normal adult liver cells.

References BAGHERI SA, BOYER JL: Peliosis hepatis associated with androgenic-anabolic steroid therapy. A severe form of hepatic injury. Ann Intern Med 1974; 81 :610-618. BALAZS M: Primary hepatocellular tumours during longterm androgenic steroid therapy. A light and electron microscopic study of 11 cases with emphasis on microvasculature of the tumours. Acta Morphol Hung 1991; 39: 201-216. BARON J, KAWABATA T, KNAPP S, et a1.: Intrahepatic distribution of xenobiotic metabolizing enzymes. In: CALDWELL J, PAULSON GD (eds.): Foreign compound metabolism. London Taylor & Francis 1984 pp. 17-36. BOHM N, MOSER B: Reversible hyperplasia and hypertrophy of the mouse liver induced by a functional charge with phenobarbital. Beitr Patho11976; 3: 283-300. CORREIA MA: Rat and human liver cytochrome P450: substrate and inhibitor specificities and functional markers. In: ORTIZ DE MONTELLANO PR (ed.): Cytochrome P450: structure, mechanism and biochemistry. Plenum Press New York, London 2nd edition, 1995; pp. 607-630. DASMAHAPTRA AK. LEE PC: Down regulation of CYPIAI by glucocorticoids in trout hepatocytes in vitro. In Vitro Cell Dev BioI Anim 1993; 29A: 643-648. DESCOTES G, OSMAN H, GALLAS JF, et a1.: Functional, biochemical and morphological hepatobiliary effects in rats chronically exposed to steroidal antiandrogen. Toxicol Appl Pharmacol1996; 137: 23-33. ERLINGER S: Drug-induced cholestasis. J Hepatol1997; 26: Suppll: 1-4. FREDRIKS WM, MARX F, V AN-NoORDEN CJ: Quantitative histochemical assessment of the heterogeneity of glycogen phosphorylase activity in liver parenchyma of fasted rats using the semipermeable membrane technique and the PAS reaction. Histochem J 1987; 19: 150-156. GIBSON GG, SKETI P: Introduction to drug metabolism, Blackie Academic & Professional London, Glasgow, New York, Tokyo, Melbourne, Madras. 2nd edition 1994. KAST A, NISHIKAWA J, YABE T, et al.: Circadian rhythm of liver parameters (cellular structures, mitotic activity, glycogen and lipids in liver and serum) during three consecutive cycles in phenobarbital-treated rats. Chronobiol Int 1988; 5:363-385. KATO K, HODGSON WJB , ONODERA K, et al.: Developmental regulation of cytochrome P-450 of intrasplenically

transplanted fetal hepatocytes in spontaneously hypertensive rats. Transplant Proc 1992; 24: 2956-2959. KATO K, HODGSON WJB, LUTION JD, et al.: Developmental expression of cytochrome P450s within intrasplenically transplanted fetal hepatocytes from spontaneously hypertensive rats. Cell Transplant 1994a; 3: 15-21. KATO K, HODGSON WJB, ABRAHAM NG, et al.: Effect of 70 % partial hepatectomy on developmental expression of cytochrome P450 IVAI within intrasplenically transplanted fetal hepatocytes from spontaneously hypertensive rats. Intern Hepatol Comm 1994b; 3: 254-259. KATO K, HODGSON WJB, ABRAHAM NG, et al.: Expression and inducibility of cytochrome P450 IlIA family within intrasplenically transplanted fetal hepatocytes. Cell Transplant 1996a; 5: 117-122. KATO K, ONODERA K, SAWA M, et al.: Effect of hepatocyte growth factor on the proliferation of transplanted hepatocytes in rats. Biochem Biophys Res Comm 1996b; 222: 101-106. KA TO K, KASAl I, ONODERA K, et al.: Effect of hepatocyte growth factor on the proliferation of intrasplenically transplanted hepatocytes in the spleen. Transplant Proc 1997;29: 2029-2031. KOKUKO N, OTSU I, OKAZAKI T, et al.: Long-term effects of intrasplenically transplanted adult hepatocytes and fetal liver in hyperbilirubinemic Gunn rats. Transpl Int 1995; 8: 262-267. KUSANO M, MITO M: Observation on the fine structure of long-survived isolated hepatocytes inoculated into rat spleen. Gastroenterol 1982; 82: 616-628. LEE PC, YOON HY, HAASCH ML, et al.: Negative control of cytochrome I A I (CYP I AI) by glucocorticoids in rainbow trout liver. Comp Biochem Physiol C 1993; 104: 457-461.

LINDBERG MC: Hepatobiliary complications of oral contraceptives. J Gen Intern Med 1992; 7: 199-209. Lupp A, LUCAS N, LINDSTROM-SEPPA P, et al.: Developmental expression of cytochrome P450 isoforms after transplantation of fetal liver tissue suspension into the spleens of adult syngenic rats. Exp Toxic Patho11998; 50: 41-51. NISHIKAWA J, KAST A, ALBERT H: Circadian rhythm of the liver of male rats dosed with phenobarbital. 1. Organ weight, cellular structures, glycogen contents and mitotic activity. Chronobiol Int 1987; 4: 161-173. RATANASAVANH D, BEAUNE P, MOREL F, et al.: Intralobular distribution and quantitation of cytochrome P-450 enzymes in human liver as a function of age. Hepatology 1991; 13: 1142-1151. SMOLOWITZ RM, HAHN ME, STEGEMAN 11: Immunohistochemical localization of cytochrome P-450IA I induced by 3,3',4,4' -tetrachlorobiphenyl and by 2,3,7,8-tetrachlorodibenzofuran in liver and extrahepatic tissues of the teleost Stenotomus chrysops (scup). Drug Metab Dispos 1991: 19: 113-123. STANG-VOSS C, ApPELL HJ: Structural alterations of liver parenchyma induced by anabolic steroids. Int J Sports Med 1981; 2: 101-105. WOLF CR, MOLL E, FRIEDBERG T, et al.: Characterization, localization and regulation of a novel phenobarbital-inducible form of cytochrome P-450, compared with three further P-450 isoenzymes, NADPH P-450 reductase, glutathione transferases and microsomal epoxide hydrolases. Carcinogenesis 1984; 5: 993-1001. VAN ERPECUM KJ, JANSSENS AR, KREUNING J, et al.: Generalized peliosis hepatis and cirrhosis after long-term use of oral contraceptives. Am J Gastroenterol 1988; 83: 572-575.

Exp TOXIC Pathol 50 ( 1998) 3

183