Molecular and Cellular Biochemistry 108: 151-156, 1991. © 1991 Kluwer Academic Publishers. Printed in the Netherlands.

A mechanism for the loss of cytochrome P-450 in primary mouse hepatocytes

Gurmit Singh and Karen L. Veltri Ontario Cancer Foundation, Hamilton Regional Cancer Centre and McMaster University, Department of Pathology, Hamilton, Ontario, Canada L8N 3Z5

Received 10 April 1991, accepted 24 June 1991

Key words: cytochrome P-450, mitochondria, heme, hepatocytes, mitochondrial DNA, cell culture

Abstract

This study examined various biochemical parameters such as mitochondria and mitochondrial DNA (mtDNA), total heme and cyto P450 content in fresh hepatocytes and dedifferentiated hepatocytes. These parameters were chosen in order to understand the dramatic decrease in drug metabolism in cultured hepatocytes. The data in this study shows a temporal decrease in cytochrome P450, total heme and also a decrease in mitochondria. Also, the ratio of mtDNA content to mitochondrial density was found to increase as hepatocytes underwent dedifferentiation. Stereological analysis of cell preparations provided a measure of mitochondrial density per cell area and mtDNA content was assessed by the use of a specific radiolabelled probe. This study demonstrates that a loss of the organelle which is partially responsible for synthesis of heine correlates with a decrease in cytochrome P450.

Introduction

Examination of primary hepatocytes provides an accurate comparison to in vivo hepatocytic proper- ties due to the short culture period and similarity to cells of origin. During 48 h in culture, hepatocytes begin to lose many of their differentiated charac- teristics; the most marked being a reduction in cytochrome P450 [1, 2]. Other studies have shown that within 48 h hepatocytes undergo 'dedifferen- tiation' [3, 4].

Primary hepatocyte cultures represent a homo- geneous system whereby various aspects of mam- malian development can be assessed. Hemopoetic function in fetal rat and mouse liver hepatocytes has been elucidated through ultrastructural and biochemical analyses [5-8] and sinusoidal cell de- velopment in the fetal liver has been determined by

electron microscopy and histochemical techniques 19, 101.

We have demonstrated that when cells differ- entiated in vitro a change in mitochondrial DNA content is observed [11]. Also in vivo differentiated organs have distinct mtDNA genomic copies [12]. In this study we examined whether in vitro we can observe changes in mitochondria and mitochon- drial DNA content as hepatocytes dedifferentiate. Mitochondrial fractional area was determined by stereological analysis of cell preparations using electron microscopy. MtDNA content per cell was estimated using a specific mtDNA probe.

The data in this study shows a temporal decrease in cytochrome P450, total heme and also a decrease in mitochondrial surface area. Further experiments are required to show that cytochrome P450 levels in primary hepatocytes may be maintained if mito-

152

chondrial and heme levels can be restored in the hepatocytes.

Materials and methods

Preparation of primary mouse hepatocyte cultures

C57BL/6J male mice (25g) were anaesthetized with sodium pentobarbitol and hepatocytes pre- pared by a modified method of Jakoby & Pastan [4].

Briefly the mouse liver was perfused by cannu- lating the inferior vena cava in a retrograde fashion with calcium-free perfusion media for 2 min at a rate of 10 ml/min followed by collagenase in calci- um free media for 5 min and again with calcium free media for an additional 1 min. Then the liver was excised and cells released in media. The cells were then washed and centrifuged 3 times to obtain a predominantly hepatocyte population. This meth- od yielded approximately 108 viable hepatocytes. The cells were attached to the plates and those that were floating were discarded as dead cells.

Hepatocytes were maintained at 37°C with 5% CO2 in a humidified atmosphere in Dulbecco's modified Eagles' medium supplemented with 10% fetal bovine serum and 100/~g/ml antibiotic-anti- mycotic. Cells harvested after 1 h and 48 h in cul- ture represent the l h fresh hepatocytes and 48 h (dedifferentiated) hepatocytes, respectively.

Electron microscopy

Hepatocytes were harvested by centrifugation and supernatant was replaced with 2% glutaraldehyde [in 0.1 M sodium cacodylate, pH 7.4] and incubated on ice for 1 h. After fixation in 1% osmium tetrox- ide (BDH) [in 0.1M sodium cacodylate, pH7.4 (Marivac)] for 1 h at 4 ° C, dehydration in graded ethanol followed. The final step involved embed- ding the cells in Spurr's resin [13]. Semi-thin sec- tions were stained with toluidine blue stain [TAAB] for light microscopic assessment. Thin sections of 70 nm were cut and stained with uranyl

acetate [saturated in 50% ethanol[ and Reynold's lead citrate.

Using the JEOL-BIOSYSTEM electron micro- scope micrographs were prepared. Stereoiogical analyses were done using the Standard Program Image Analysis System of the Kontron MOP- VIDEOPLAN (Zeiss). Mitochondrial area was obtained by measuring the area of each mitochon- drion within each micrograph; reference area was obtained by measuring the area of each micro- graph. The cumulative mitochondrial area for a series of micrographs is the 'total mitochondrial area'. The 'total reference area' is the cumulative area of a series of micrographs. To ensure statisti- cal significance, a total reference area of approxi- mately 5100/~m 2 was analyzed for each cell type.

MtDNA isolation from cells

1 ml aliquots of various cell concentrations were lysed for 2 h at 60 ° C with 10% SDS (sodium dode- cyl sulfate); 10 mM Tris-HC1/1 mM EDTA (pH 8); 0.4 M Tris-HC1/100 mM EDTA (pH 8); and Protei- nase K [0.2 mg/ml] to inhibit nucleases. DNA was extracted using phenol/chloroform [1:1] and chlo- roform/isoamyl alcohol [24:1] and precipitated with ethanol in the presence of 3 M sodium acetate. An aliquot of DNA was then removed and electro- phoresed on a 0.7% agarose gel. Similar amounts of DNA were then removed and electrophoresed in control wells for comparison of radioactivity. This procedure usually yielded a recovery of 85- 90% of recovery for mtDNA. DNA was denatured in 3 M NaOH for 1 h at 60 ° C and neutralized in 2 N ammonium acetate. The DNA was then trans- ferred onto nitrocellulose filter (S & S, 0.45tzm size).

MtDNA hybridization

Nitrocellulose filters were prehybridized for 6 h at 68 ° C in buffer [20 × SSC, 50 × Denhardt's, yeast tranfer RNA (20 mg/ml)]. The mtDNA probe was labelled with ct- [32p] _ dATP (3000 ci/mmol) by a

Table 1. Cytochrome P450 and total heme content

Cell type [cytochrome P450] Total heme (nmol/cell) (nmoles/1 × 106

cells)

Hepatocytes (1 h) 8.9 x 10 7 0.175 + 0.024 Hepatocytes (48h) 0.3 × 10 .7* > 0.01 + 0.005*

*P< 0.05.

modified nick translation procedure [14]. Unin- corporated nucleotides were removed by sephadex G-50 spin column equilibrated with 10 mM Tris- HC1 p H 8.0, 5 0 m M NaC1, 0 .1mM E D T A pH8.0 . Probes with high specific activity approximately 1 x 108 dpm//xg D N A were obtained with this pro- cedure. Filters containing bound m t D N A were hy- bridized with [32p] dATP-labelled m t D N A probe

for 18 h at 68 ° C. Vector PSP64 contains the mouse m t D N A genome inserted into the Sac I restriction site (provided by Dr. D. Clayton, Stanford Uni- versity). Quanti tat ion of m t D N A on nitrocellulose filters was done by using the B io -RAD model 620 densi tometer .

Standard curves with various concentrations of m t D N A in triplicate were used to estimate m t D N A in the various cell types. D N A dilutions were done in order to estimate the linear range.

Determination of total heme and cytochrome P-450

Total heme was measured by the pyridine hema- chromogen method described by Falk [15]. The results were calculated f rom the difference spec-

Table 2. MtDNA quantitation in hepatocytes

153

trum at 558 nm and expressed as nmoles of heme per mg of protein using an extinction coefficient of 31 mM -8 cm -s. Cytochrome P-450 was measured by the method of Omura and Sato [16].

Results

(a) Changes in cytochrome P-450 and total heme Fresh hepatocyte cultures ( l h ) cytochrome P450 content was 8.9 × 10 -7 nmol/cell. How- ever, within 48 h the content was reduced to less than 4% i.e. 0.3 × 10 -7 nmol/cell. Simi- larly the concentration of total heme was re- duced from 0.175 nmoles/106 cells in fresh he-

patocyte cultures to barely detectable levels in hepatocytes that dedifferentiated after 48 h in culture. These data are shown in Table 1.

(b) Mitochondrial DNA content There was a significant decrease in hepatocyte m t D N A content from l h to 48h in culture (Table 2). Accounting for differences in mi- tochondrial area (MT Area) per cell, the ratio of m t D N A to mitochondrial area was substan- tially lower in 1 h hepatocyte culture relative to dedifferentiated hepatocytes.

Using the two-sample t-test, a significant dif- ference at P < 0.05 in values of mtDNA/cel l was found between l h and 48h hepatocyte cultures. These data are presented in table 2.

(c) Mitochondrial morphometry The total mitochondrial area measured f rom a series of micrographs is represented by ' ~ Ar- eaMT'. Examinat ion of hepatocytes indicated an approximate 7-fold difference in mitochon-

Cell type MtDNA content/cell (~g) Copies mtDNA cell MT area/cell (/xm 2) Ratio of mtDNA/cell to MT area/cell (copies/txm 2

Hepatocytes (1 h) 19.2 x 10 .8 + 5.2 x 10 -8 12800 _+ 346 83.8 + 6.4 153 Hepatocytes (48h) 7.6 x 10 .8* _ 3.4 x 10 -8 5067* _+ 226 7.9* _+ 0.6 641"

Values for mtDNA content per cell represent mean + S.D. 1 copy corresponds to 1.5 x 10-11txg DNA. *p< 0.05.

154

Fig. 1. Electron micrographs of: (A) Hepatocytes (1 h), (B) Hepatocytes (48 h). M - Mitochondria. Mag. Bar is (1/zm).

drial area density between 1 h and 48 h cultur- es; 1 h cultures representing the higher density (Table 3). Cell-specific mitochondrial area densities are clearly seen in electron micro- graphs (Fig. 1 A, B). Mitochondrial size was also shown to be cell-specific. In i h hepatocyte cultures, mitochondria were about 3.5 times larger than their dedifferentiated counterparts (Table 1).

Using the two-sample test, a significant dif- ference was found at P < 0.05 in values of mean area per mitochondrion and mean area per cell for 1 h and 48 h hepatocyte cultures.

Discussion

Examination of flesh primary hepatocytes and de- differentiated hepatocytes revealed differences in the ratio of mtDNA content to mitochondrial den- sity. In all cell types, mtDNA content was inversely

proportional to mitochondrial density. The higher ratio of mtDNA to mitochondrial area in dediffe- rentiated hepatocytes compared to fresh ( lh cul- tures) hepatocytes indicates that mtDNA content per mitochondrial density increased as hepatocytes reverted back to a dedifferentiated state.

Previous ultrastructural studies reported a loss of granular endoplasmic reticulum and deterioration of cellular morphology associated with the hepato- cytic dedifferentiated state [1, 2]. Other investiga- tors have confirmed these observations [17, 18]. In this study, similar morphological characteristics of dedifferentiation were evident in 48 h hepatocyte cultures. We interpreted hepatocyte dedifferentia- tion on the basis of morphological similarity to previous studies [1, 2, 17, 18]. We also observed an approximate 96% decrease in cytochrome P450 concentration in hepatocytes maintained for 48 h in culture. A 98% decline in cytochrome P450 con- tent was reported in another study using primary

Table 3. Stereological analysis

Cell type Y~ AreaMT (/~m 2) MT area density Mean area/MT (/zm 2) Mean area/cell (/zm 2)

Hepatocytes ( l h ) 1007 0.20 2.1 _ 1.1 419 _+ 150 Hepatocytes (48 h) 140" 0.03* 0.6 ± 0.4* 2 6 4 - 102

Values for mean area per MT and mean area per cell represent mean + S.D. )~ AreaMT refers to total mitochondrial surface area per 5100 ~m2. * P > 0.05.

rat hepatocyte cultures maintained for 48 h or long- er [2].

Earlier electron microscopic analysis of tumour cells and functionally differentiated liver and mus- cle tissue has established that mitoch~ndtia from tumour cells exhibited relatively fewer cristae and more open spaces within the matrix [19].

A previous investigation examined the devel- opment of fetal rat liver hepatocytes using ster- eological and ultrastructural criteria [20]. It con- cluded that structural changes occurring in matur- ing hepatocytes were characterized by a cubic cel- lular appearance, oblong-shaped mitochondria and well-differentiated Golgi apparatus and rough endoplasmic reticulum [20]. Golgi apparatus and rough endoplasmic reticulum was readily visible and quite prominent in fully matured hepatocytes (1 h) in the present study.

Mitochondria in cells can be upregulated with agents such as clofibrate and hormones such as Thyroxine [21, 22]. MtDNA in hepatocytes can be lowered using ethidium bromide [23]. Thus experi- ments involving upregulation or down regulation of mtDNA could influence a change in cytochrome P-450 loss. However, it is well recognized that agents such as clofibrate induce the level of specific isozymes of cytochrome P-450 and thus may not be an ideal upregulator. Thus more precise agents that can upregulate mtDNA without affecting cyto- chrome P-450 may be required to examine the mechanism for loss of cytochrome P-450.

It is recognized that a number of changes occur in the hepatocytes during dedifferentiation. Thus a correlation between changes in mitochondria and loss of cytochrome P-450 may not be causal. How- ever, since cytochrome P-450 is a hemoprotein and the loss of mitochondria is directly related to loss of heme, it is most likely that cytochrome P-450 loss is also related. Further experiments as suggested, whereby the level of heme or mitochondrial stabil- ity can be achieved, a direct causal relationship between loss of cytochrome P-450 and mitochon- dria may be established.

In summary this study shows that mtDNA con- tent decreases as cells dedifferentiate and this may result in decreased mitochondria. The temporal decline in mitochondria is consistent with the turn-

155

over of mitochondria. Decrease in mitochondria could be responsible for decline in total heme which contributes to a decrease in cytochrome P-450 levels.

Acknowledgements

We would like to acknowledge financial assistance from the Medical Research Council of Canada.

References

1. Bissell MJ: The differentiated state of normal and malig- nant cells or how to define a 'normal' cell in culture, fnt Rev Cytol 70: 27-100, 1981

2. Bissell DM, Guzelian PS: Phenotypic stability of adult rat hepatocytes in primary monolayer culture. Ann NY Acad Sci 349: 85-98, 1981

3. Buonassisi V, Sato G, Cohen A: Hormone producing cul- tures of adrenal and pituitary tumor origin. Proc Natl Acad Sci USA 48: 1184-1190, 1962

4. Jakoby W, Pastan I: Cell culture. In: W Jakoby and I Pastan (eds.) Methods in Enbzymology, vol. 58. Academic Press, New York, 1979

5. Holtzman E, Novikoff AB: Cells and Organelles (3rd ed.) Saunders College Publishing, New York, 1984

6. Asano H, Kobayashi M, Hoshino T: The hemopoetic mi- croenvironment in the fetal liver of mice: relationship be- tween developing hepatocytes and erythroblasts. J Elect Micros 36: 15-25, 1987

7. Luzzato AC: Hepatocyte differentiation during early fetal development in the rat. Cell and Tissue Res 215: 133-142, 1981

8. Medlock ES, Haar JL: The liver hemopoetic environment: 1. Developing hepatocytes and their role in fetal hemopoie- sis. Anatom Rec 207: 31-41, 1983

9. Bankston PW, Pino RM: The development of the sinusoids of fetal rat liver: morphology of endothelial cells, Kupffer cells, and the transmural migration of blood cells into the sinusoids. Am J Anatomy 159: 1-15, 1980

10. Pino RM, Bankston PW: The development of the sinusoids of fetal rat liver: localization of endogenous peroxidase in fetal Kupffer cells. J Histochem Cytochem 27: 643-652, 1979

11. Singh G, Veltri KL: Effect of differentiation of embryonal carcinoma cells (P19) on mitochondrial DNA content in

vitro. 1991 (In press) 12. Vettri KL, Espiritu M, Singh G: Distinct genomic copy

number in mitochondria of different mammalian organs. J Cellular Physiology 143: 160-164, 1990

13. Rigby PW, Dieckmann M, Rhodes C, Berg P: Labelling

156

DNA to high specific activity in vitro by Nick-translation with DNA polymerase 1. J Mol Biol 113: 237-251, 1977

14. Spurt AR: A low-viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res 26: 31, 1969

15. Falk JE: Porphyrins and metalloporphyrins. BBA Libr 2: 232, 1964

16. Omura T, Satu R: The carbon monoxide-binding pigment of liver microsomes. 1. Evidence for its hemoprotein na- ture. J Biol Chem 239: 2370-2378, 1964

17. Reid LM, Jefferson DM: Culturing hepatocytes and other differentiated cells. Hepatology 4: 548-559, 1984

18. Suolinna EM, Pitkaranta T: Effect of culture age on drug metabolizing enzymes and their induction in primary cul- tures of rat hepatocytes. Biochem Pharmacol 35: 2241- 2245, 1986

19. Nass NMK: Mitochondrial DNA: Advances, problems and goals. Science 165: 25-35, 1969

20. Vassy J, Kraemer M, Chalumeau MT et al.: Development

of the fetal rat liver: Utrastructural and stereological study of hepatocytes. Cell Differ 24: 9-24, 1988

21. Meijer J, Afzelius BA: Effects of clofibrate treatment and of starvation on peroxisomes, itochondria and lipid droplets in mouse hepatocytes: A morphometric study. J Ultrastruct Mol Struct Res 102: 87-94, 1989

22. Leung ACF, McKee EE: Mitochondrial protein synthesis during thyroxine-induced cardiac hypertrophy. Am J Phy- siol Endocrinol Metab 258: E511-E518, 1990

23. KingMP, AttardiG: Injectionofmitochondriaintohuman cells leads to a rapid replacement of the endogenous mi- tochondrial DNA. Cell 52: 811-819, 1988

Address for offprints: G. Singh, Department of Pathology, HSC 3N26D, McMaster University, 1280 Main Street West, Hamil- ton, Ontario, L8N 3Z5 Canada