Maintenance of microsomal cytochromes b5 and P-450 in primary cultures of parenchymal liver cells on collagen membranes

In previous reports from various laboratories, the levels of the microsomal cytochromes b₅ and P-450 in hepatocytes in primary culture have been found to be very low and difficult to measure. The studies reported in this paper demonstrate that cytochromes b₅ and P-450 in hepatocytes cultured on floating collagen membranes for periods of at least 10 days are maintained at levels readily measured by conventional techniques and comparable to those of liver $\text{in}$$\text{vivo}$. Addition of high levels of hydrocortisone (10^?4M) to the culture medium for periods up to 10 days resulted in further increases in the levels of these cytochromes. Cells cultured in the presence of hydrocortisone exhibited the appearance of cytochrome P-448, in contrast to the cells cultured in the absence of hydrocortisone, where cytochrome P-450 was maintained.     

Inhibition of overall protein and RNA synthesis as a mechanism for the tunicamycin induced decrease in cytochrome P-450 in rat hepatocytes

In rat hepatocytes maintained in culture, cytochrome P-450 and NADPH cytochrome c reductase activities were decreased by tunicamycin in a dose and time dependent fashion. The effect of tunicamycin was mainly due to inhibition of protein synthesis. Tunicamycin decreased L-[35S] methionine incorporation into many proteins, including a 52 kDa cytochrome P-450 isozyme. Tunicamycin also reduced RNA synthesis. These results indicate that tunicamycin decreased cytochrome P-450 levels in hepatocytes by inhibiting protein and RNA synthesis.     

Long-term survival of functional hepatocytes from adult rat in the presence of phenobarbital in primary culture

In the presence of phenobarbital (PB) at 3 mM, hepatocytes isolated from adult rats by a collagenase-perfusion technique survived well on plastic dishes for at least 49 days after initiation of primary culture. PB at concentrations less than 3 mM was ineffective for the maintenance of hepatocytes, and the maintenance of them was attained only in the continuous presence of 3 mM PB. The hepatocytes surviving in the presence of 3 mM PB were morphologically indistinguishable from the hepatocytes after 1-day attachment period, except for the presence of prominent nucleoli in the former. Although both the albumin secretion and tyrosine aminotransferase (TAT) activities of the cells decreased gradually up to day 7 with time in culture, both were thereafter maintained at relatively high levels at least up to day 35 of primary culture. The addition of 10 microM dexamethasone caused a 3-5-fold induction in TAT activity, and the cells were capable of responding to the hormone in this manner at least up to day 28 of primary culture. Furthermore, the cells also had glucose-6-phosphatase activity, even though the level of this enzyme activity was relatively low as compared with that of TAT activity. Survival of hepatocytes in the presence of 3 mM PB was further enhanced by simultaneous addition of dexamethasone (10 microM) and insulin (10 micrograms/ml). The sensitivity of hepatocytes to 3'-methyl-4-dimethylaminoazobenzene (0.24 mM) was remarkably reduced by treatment with PB at 3 mM. PB treatment decreased efficiently the falling rate of total cytochrome P-450 content, but did not induce P-450PB, which is the specific form of cytochrome P-450 induced by PB, in primary cultured hepatocytes. On the other hand, 3-methylcholanthrene (MC, 10 microM) caused an increase of both contents of total cytochrome P-450 and P-450MC, which is the specific form of cytochrome P-450 induced by MC, in primary cultured hepatocytes. However, MC was ineffective for the maintenance of hepatocytes in primary culture. The possible biological actions of PB on primary cultured hepatocytes are discussed on the basis of the experimental data obtained.     

Primary cultures of hepatocytes on human fibroblasts.

Parenchymal hepatocytes isolated from adult rats were cultured on three types of collagen-containing substrata: collagen-coated plates, collagen membranes and confluent diploid human fibroblasts. Hepatocytes on the latter two substrata maintained characteristic morphology for at least 10 days in culture, whereas degenerative changes (cell death and formation of multinucleated hepatocytes) and growth of nonparenchymal elements were seen after 5 days in cultures on collagen-coated plates. Parallel findings were seen on basal and induced levels of cytochrome P-450 and NADPH-cytochrome C reductase. The basal levels of cytochrome P-450 were not measurable after day 3 in hepatocytes cultured on collagen-coated plates, whereas measurable levels were maintained in the hepatocytes cultured on the other two substrata. Addition of phenobarbital or methylcholanthrene at day 5 in culture caused an increase in cytochromes P-450 and P-448, respectively, only in hepatocytes cultured on collagen membranes and confluent fibroblasts. Analogous results were seen for the enzyme NADPH-cytochrome C reductase. The similarities in performance between hepatocytes on collagen membranes and on human fibroblasts show that a continuous collagen-containing substratum is important for optimal performance of hepatocytes in primary culture. The possible importance of cultures of hepatocytes on human fibroblasts for carcinogenesis studies is discussed.     

Primary cultures of hepatocytes on human fibroblasts

Summary Parenchymal hepatocytes isolated from adult rats were cultured on three types of collagen-containing substrata: collagen-coated plates, collagen membranes and confluent diploid human fibroblasts. Hepatocytes on the latter two substrata maintained characteristic morphology for at least 10 days in culture, whereas degenerative changes (cell death and formation of multinucleated hepatocytes) and growth of nonparenchymal elements were seen after 5 days in cultures on collagen-coated plates. Parallel findings were seen on basal and induced levels of cytochrome P-450 and NADPH-cytochrome C reductase. The basal levels of cytochrome P-450 were not measurable after day 3 in hepatocytes cultured on collagen-coated plates, whereas measurable levels were maintained in the hepatocytes cultured on the other two substrata. Addition of phenobarbital or methylcholanthrene at day 5 in culture caused an increase in cytochromes P-450 and P-448, respectively, only in hepatocytes cultured on collagen membranes and confluent fibroblasts. Analogous results were seen for the enzyme NADPH-cytochrome C reductase. The similarities in performance between hepatocytes on collagen membranes and on human fibroblasts show that a continuous collagen-containing substratum is important for optimal performance of hepatocytes in primary culture. The possible importance of cultures of hepatocytes on human fibroblasts for carcinogenesis studies is discussed.     

Cryopreservation of isolated rat hepatocytes

Summary Isolated parenchymal hepatocytes from adult rats were frozen in media containing 10% glycerol, 10% dimethylsulfoxide (DMSO), or 20% DMSO. Three microsome-associated functions were compared in nonfrozen cells and cells frozen in each of the above cryoprotectant solutions. Freezing in DMSO maintains cytochromes P-450 and b₅ and NADPH-cytochrome C reductase at levels nearer to control values than does freezing in glycerol. Cells frozen and subsequently thawed and cultured for 24 h lose a greater amount of cytochrome P-450 than do nonfrozen cultured cells. The levels of cytochrome b₅ and reductase in frozen-thawed cells remain close to control values. Cell viability (trypan blue dye exclusion and percentage of attached cells) after freezing is maintained better using DMSO as a cryoprotectant. Dimethylsulfoxide protects the hepatocytes from freeze-induced damage to the extent that many viable cells attach to collagen-coated petri dishes, survive for at least 24 h, and still maintain significant levels of enzymes of importance to drug and carcinogen metabolism. This work was supported by Grant CA-30241 from the National Institutes of Health, Bethesda, Maryland.     

Effect of low temperature preservation and cell density on metabolic function in a bioartificial liver

Difficulties associated with bioartificial liver (BAL) preservation limit not only the commercialization of BAL, but also its clinical trials. In this study, the possibility of cold preservation of BAL cartridges containing porcine hepatocytes was examined at 4 °C. In anin vitro perfusion culture system, BAL cartridges maintained cytochrome P450 metabolic function for at least 50 days. However, all BAL cartridges completely lost their ammonia eliminating ability when stored at 4 °C. We also studied the effect of cell density on the maintenance of BAL liver function in a highly differentiated and healthy state. As expected, BALs containing a larger number of hepatocytes demonstrated higher metabolic functions. When metabolic functions were compared per gram of hepatocytes, no large differences were observed between devices containing different densities of hepatocytes. Decreased cell density did not successfully prolong BAL function. The viability and function of isolated hepatocytes highly depend on the culture conditions, such as cell density, substrata, culture media, and additives to the culture media. Perfusion culture of BAL cartridges at 4°C gave a promosing result with respect to the maintenance of P450 activity. However, as indicated by the rapid loss of ammonia metabolic activity, many factors still remain to be optimized for preservation of BAL keeping high metabolic functions for a longer time.     

Effects of Hormones on Changes in Cytochrome P-450, Prolyl Hydroxylase,and Glycerol Phosphate Acyltransferase in Primary Monolayer Cultures of Parenchymal Cells from Adult Rat Liver

Previous studies have shown that isolation and primary culture of rat hepatocytes in a standard, chemically defined medium is associated with selective changes in microsomal function. These changes were found to be selectively sensitive to addition of hormones to the culture medium. The concentration of cytochrome P-450 declined dramatically during the first 24 hours of incubation. However, cytochrome P₁-450, a form of the hemoprotein induced by polycyclic aromatic hydrocarbons, was resistant to this change. Cytochrome P₁-450 levels selectively rose during the first ten hours in culture and, thereafter, declined at a less rapid rate than did the cytochrome P-450 in normal hepatocytes or in cells prepared from phenobarbital pretreated animals. Addition of dexamethasone to the medium at the time of cell plating partially prevented the fall of cytochrome P-450 and of ¹⁴C-heme in microsomes prepared from hepatocytes derived from rats given 5¹⁴[C]-δ-aminolevulinic acid. This suggests that the steroid decreases degradation of the hemoprotein. As compared to the loss of cytochrome P-450 in cultures of normal hepatocytes, the hemoprotein fell to lower levels in hepatocytes prepared from regenerated liver four days after partial hepatectomy. This result may be related to the accelerated formation of the monolayer in the cultures of regenerated hepatocytes. Both sn-glycerol-3-phosphate acyltransferase activity and glycerol kinase activity declined in the first 24 hours of culture. The fall in the latter enzyme was partially prevented by addition of estradiol. Collagen prolyl hydroxylase, a newly discovered microsomal constituent of the hepatocyte, rose slightly during the first 24 hours in culture. This change was augmented threefold by addition of insulin to the medium. We conclude that the present hepatocyte culture system with its attendant changes in functional phenotype may be useful in better defining the role of hormones in modulating metabolic processes in the liver.     

Effects of ethanol and clofibrate on expression of cytochrome P-450 enzymes and epoxide hydrolase in cultures and cocultures of rat hepatocytes

Cultured and cocultured rat hepatocytes were used to study the effects of ethanol and clofibrate on cytochrome P-450 (P-450) enzymes and epoxide hydrolase. We showed that in the presence of ethanol, clofibrate or both compounds, rat hepatocytes were able to express, after 3 days of pure culture, quantitatively and qualitatively reasonable levels of most cytochrome P-450 enzymes and epoxide hydrolase, compared to freshly isolated hepatocytes. However, ethanol induced the P-450IA subfamily, and clofibrate the P-450-IVA subfamily. In cocultures, after 6 days, most P-450 enzymes were still expressed while P-450IIC11 was completely lost. Ethanol and clofibrate had the same effect as in pure culture. These results show, by modifying culture medium conditions and cell-cell interactions, that it is possible to maintain reasonable xenobiotic-metabolizing-enzyme expression; however, these conditions have to be improved in order to preserve better P-450 expression. The mechanism of these effects and the inducibility of these systems remain to be elucidated by a study at molecular level.     

Di-rhamnolipids improve effect of trehalose on both hypothermic preservation and cryopreservation of rat hepatocytes

Trehalose, a disaccharide of glucose, is a highly hydrophilic small molecule (MW?=?342D) and a bioprotectant normally impermeable to the membrane phospholipid bilayer. Di-rhamnolipids, a major component of rhamnolipids, were applied to increase the effect of trehalose in cryopreservation and hypothermic preservation. We found that di-rhamnolipids (10 mg/L) increased the survival of hepatocytes after cryopreservation or hypothermic preservation as indicated by cell viability using trypan blue exclusion and methyl thiazolyl tetrazolium assay. Correspondingly, after hepatocytes were preserved in the presence of di-rhamnolipids, their hepatospecific functions were comparable to those of freshly cultured cells in terms of intracellular glutathione level, albumin secretion, urea production, and metabolic activities of cytochrome P450 isoforms. Measurement of trehalose intracellular concentration showed that its accumulation increased in the presence of di-rhamnolipids (10 mg/L) but was not altered by two other well-known surfactants, Tween-80, and Pluronic 127. Hence, di-rhamnolipids, which are non-toxic, effective, and commercially available, could be a promising protectant by potentiating the function of trehalose against hypothermic or cryopreservation cell damage.     

Studies on the maintenance of cytochromes P-450 and b5, monooxygenases and cytochrome reductases in primary cultures of rat hepatocytes

The cytochrome P-450 content of rat hepatocytes declined rapidly over 72 h in culture, due primarily to denaturation to cytochrome P-420. Six different media were investigated for their ability to conserve cytochrome P-450 during culture, and the most successful was a modified Earle's medium. After 72 h culture in this medium, cytochromes P-450 and b5, NADH-cytochrome b5- and NADPH-cytochrome c-reductases were maintained at 40, 100, 35 and 52% of fresh cell values, respectively. Cytochrome P-450 showed differential functional stability during culture with ethoxyresorufin O-deethylation being more stable than either pentoxyphenoxazone O-depentylation or biphenyl 4-hydroxylation. Monooxygenase than did cytochrome P-450 content. This discrepancy was not explained by loss of flavin nucleotides, FMN or FAD.     

The B-type cytochrome in endoplasmic reticulum of mammary gland epithelium and milk fat globule membranes consists of two components cytochrome b5 and cytochrome P-420

The cytochrome contents of rough endoplasmic reticulum (ER) of lactating bovine and rat mammary epithelial cells and of the membranes surrounding the fat globules in bovine and human milk (MFGM) were analyzed with spectrophotometric (at +20 °C and ?196 °C) and immunological methods. Two cytochrome components were found. One was identified as cytochrome b₅ by the characteristic split of the α-band in reduced versus oxidized difference spectra at low temperature, by the reduction with NADH, which was insensitive against rotenone and antimycin, and by the solubility upon trypsin treatment. This component showed cross-reaction with the microsomal cytochrome b₅ from rat hepatocytes using rabbit antibodies against the purified cytochrome b₅ fragment released from rat liver microsomes by trypsin treatment. The in situ localization of cytochrome b₅ in mammary epithelial cells was demonstrated by indirect immunofluorescence microscopy in both frozen sections of tissue and cultured cells. The second cytochrome component was identified as cytochrome P-420 by the characteristic spectral bands in the CO-difference spectrum and the dithionite-difference spectrum, by the reaction with cyanide, and by the insolubility upon trypsin treatment of the membranes. In addition, we found evidence for the existence of a form of cytochrome P-420 in these membranes which does not bind CO. The presence of cytochrome P-420 in mammary gland ER and MFGM fractions was not due to preparative artifacts. No cytochrome P-450 was observed in these membranes. The significance of the occurrence of these redox components in ER and surface membrane of mammary gland epithelium and other cells is discussed.     

Regulation of cytochrome P-450c by glucocorticoids and polycyclic aromatic hydrocarbons in cultured fetal rat hepatocytes

The actions of polycyclic aromatic hydrocarbons and glucocorticoids to regulate the synthesis of cytochrome P-450c (the major isozyme induced by polycyclic aromatic hydrocarbons) were investigated in fetal rat hepatocytes maintained in primary monolayer culture. Treatment of hepatocytes in culture with 1,2-benzanthracene resulted in a 50-fold increase in 7-ethoxycoumarin O-deethylase activity. The level of P-450c increased in the cells in a time-dependent fashion as determined by immunoelectrophoretic analysis. The inductive effect of BA was potentiated approximately 1.6- to 2.3-fold when 1 microM dexamethasone was included in the culture medium. However, dexamethasone alone had little or no effect on the induction of P-450c. The rate of synthesis of P-450c was examined by immunoisolation of the specific isozyme from total cellular proteins radiolabeled with [35S]methionine and from the protein products formed during in vitro translation of the isolated mRNA. In addition, the amount of mRNA specific for cytochrome P-450c was determined by Northern blot analysis of RNA extracted from cultured cells. The changes in the rates of synthesis and mRNA levels were found to parallel the changes in enzyme activity. The concentration of dexamethasone required to cause a half-maximal increase in P-450c content in the presence of 1,2-benzanthracene was between 10(-8) and 10(-7) M. It is concluded that glucocorticoids act synergistically with polycyclic aromatic hydrocarbons to increase the levels of P-450c expressed in the fetal rat liver, and that this action is likely mediated by the classical type II glucocorticoid receptor.     

Immunological evidence for phenobarbital-stimulated synthesis of cytochrome P-450 in primary cultures of chick embryo hepatocytes

The induction of cytochrome P-450 by phenobarbital was studied in primary cultures of chick embryo hepatocytes. The rate of the de novo synthesis of the induced form of cytochrome P-450 was measured directly and specificially, using form-specific anti-cytochrome antibodies that quantitatively immunoprecipitated this form from the radiolabeled hepatocytes. Additionally, the steady-state levels of the cytochrome were estimated spectrophotometrically and electrophoretically. In the presence of phenobarbital the synthesis of cytochrome P-450_PB by cultured hepatocytes was markedly accelerated. Furthermore, the same cytochrome P-450_PB form was induced by phenobarbital in vivo in chicken liver and in the cultured chick embryo hepatocytes. Their identity was judged from immunological and electrophoretic properties of these induced cytochromes. Immunological cross-reactivity was also detected between the cytochrome P-450_PB forms from chick embryo hepatocytes and from adult rat liver. The immunological cross-reactivity observed between the phenobarbital-induced cytochrome P-450 forms from different species was not observed between the different cytochrome forms with the same liver (Thomas, P.E., Reik, L.M., Ryan, D.E. and Levin, W. (1981) J. Biol. Chem. 256, 1044–1052). Implications as to the evolutionary origin of the different cytochrome forms are discussed.     

Phenobarbital induction of cytochromes P-450. High-level long-term responsiveness of primary rat hepatocyte cultures to drug induction, and glucocorticoid dependence of the phenobarbital response.

The induction of hepatic cytochromes P-450 by phenobarbital (PB) was studied in rat hepatocytes cultured for up to 5 weeks on Vitrogen-coated plates in serum-free modified Chee's medium then exposed to PB (0.75 mM) for an additional 4 days. Immunoblotting analysis indicated that P-450 forms PB4 (IIB1) and PB5 (IIB2) were induced dramatically (greater than 50-fold increase), up to levels nearly as high as those achieved in PB-induced rat liver in vivo. The newly synthesized cytochrome P-450 was enzymically active, as shown by the major induction of the P-450 PB4-dependent steroid 16 beta-hydroxylase and pentoxyresorufin O-dealkylase activities in the PB-induced hepatocyte microsomes (up to 90-fold increase). PB induction of these P-450s was markedly enhanced by the presence of dexamethasone (50 nM-1 microM), which alone was not an affective inducing agent, and was inhibited by greater than 90% by 10% fetal bovine serum. The PB response was also inhibited (greater than 85%) by growth hormone (250 ng/ml), indicating that this hormone probably acts directly on the hepatocyte when it antagonizes the induction of P-450 PB4 in intact rats. In untreated hepatocytes, P-450 RLM2 (IIA2), P-450 3 (IIA1) and NADPH P-450 reductase levels were substantially maintained in the cultures for 10-20 days. The latter two enzymes were also inducible by PB to an extent (3-4 fold elevation) that is comparable with that observed in the liver in vivo. Moreover, P-450c (IA1) and P-450 3 (IIA1) were highly inducible by 3-methylcholanthrene (5 microM; 48 h exposure) even after 3 weeks in culture. In contrast, the male-specific pituitary-regulated P-450 form 2c (IIC11) was rapidly lost upon culturing the hepatocytes, suggesting that supplementation of appropriate hormonal factors may be necessary for its expression. The present hepatocyte culture system exhibits a responsiveness to drug inducers that is qualitatively and quantitatively comparable with that observed in vivo, and should prove valuable for more detailed investigations of the molecular and mechanistic basis of the response to PB and its modulation by endogenous hormones.     

Hypothermic preservation of hepatocytes on gelatin gels

Hypothermic preservation of hepatocytes on gelatin gels allows hepatocytes to be stored for at least 9 days. The procedure is easy, inexpensive and does not require specialised equipment. The cells retain their morphology and are released as separate spherical entities by dissolving the gelatin layer at 37°. The recovered cells have an intact plasma membrane as judged by lactate dehydrogenase activity, attachment efficiency and subsequent monolayer formation in culture. Functional tests show that the cells recover quickly from the storage conditions. Rates of protein synthesis are maintained over a 6 day period and remain at 62% of the initial level on day 9. The liver specific, hormonal induction of the enzyme tyrosine aminotransferase is apparent throughout the culture period. In addition a phenotypic marker enzyme, γ-glutamyl transpeptidase, remained at basal levels.     

Regulation of gene expression in adult rat hepatocytes cultured on a basement membrane matrix

Freshly isolated adult rat hepatocytes, when cultured on type I collagen (commercially available as Vitrogen), assume a polygonal shape, form a stable monolayer within 24 hours, but lose the capacity to express some liver-specific functions over time in culture. We incubated hepatocytes in a serum-free medium on a reconstituted basement membrane gel, "matrigel" (prepared from an extract of extracellular matrix of the murine Engelbreth-Holm-Swarm sarcoma), and observed that the cells adhered firmly, remained rounded as single cells or clusters, and maintained liver-specific gene expression for more than 1 week in vitro. Hepatocytes on matrigel secreted substantially higher amounts of albumin, transferrin, haptoglobin, and hemopexin, Northern blot analyses of extracted cellular RNA, expressed increased amounts of mRNA for the liver-specific protein albumin (as compared with cells on vitrogen). In cultures treated with phenobarbital, cytochrome P-450b, and cytochrome P-450e, mRNAs and proteins were barely detectable in cells on Vitrogen but were induced to levels similar to those in the liver in vivo in matrigel cultures. Likewise, the use of matrigel greatly enhanced the induction of mRNA and protein for P-450c by 3-methylcholanthrene and for P-450p by steroidal and nonsteroidal inducers. However, neither substratum permitted induction of P-450d by 3-methylcholanthrene, suggesting that the effects of matrigel are selective even for expression in liver of members of the superfamily of cytochrome P-450 genes. Within 5 days in cultures on Vitrogen, hepatocytes expressed detectable amounts of fetal liver aldolase activity and also mRNA for vimentin and type I collagen, each considered a phenotypic change reflecting hepatocyte "dedifferentiation." None of these was present in cells on matrigel. Responsiveness to mitogenic stimuli, as judged by incorporation of 3H-thymidine into DNA, was also decreased in hepatocytes cultured on matrigel. Finally, there was a remarkable increase in the levels of both matrices during the first 2 days in culture. However, the continuously cytoskeleton mRNA over time in culture than did the rounded cells on matrigel. We conclude that hepatocytes cultured on matrigel, as opposed to the standard collagen, exhibit remarkably enhanced expression of many liver-specific functions.