Analysis of discontinuous variation in albumin production by hepatoma cells at the cellular level

The clonal variation in the rate of albumin production in cultured rat hepatoma cells has been studied on a cellular basis by immunoperoxidase techniques using specific antisera against rat serum albumin. Previously, it has been shown that an array of clonal hepatoma cell populations that produce serum albumin at different rates can be isolated simply by subcloning a single clonal hepatoma cell line (Fu5). The present study demonstrates conclusively that this phenotypic variation is the result of quantal shifts in the rate of albumin production in the individual cells and is not due to changes in the percentage of albumin-producing cells. Also, by analyzing individual colonies as they develop from single cells, it was possible to establish that the rate of variation in albumin content in several hepatoma cell clones is on the order of 0.5-1.4 10(-2) per cell per generation. This variation in albumin content probably reflects shifts in the rate of albumin synthesis. Even after several sequential subclonings, the same clonal variation persists. The variants are not the result of fluctuations in albumin synthesis with different phases of the cell cycle.     

A mouse hepatoma cell line which secretes several serum proteins including albumin and alpha-foetoprotein.

A permanent cell line (BW) was established from a transplantable mouse hepatoma, BW7756, which produces alpha-foetoprotein (AFP). Three clones were isolated from the uncloned culture: BW1, BW2 and BWTG3. The cells of the latter clone, which was isolated after selection in the presence of thioguanine, are deficient in the enzyme hypoxanthine-guanine-phosphoribosyl transferase. Both BW1 and BWTG3 cells have mean chromosome number of 64 (60 telocentric and 4 metacentric chromosomes). All three clones secrete at least five serum proteins into the culture medium: albumin, AFP, and alpha 2 globulin, transferrin and C3, the third component of complement. The approximate rate of albumin secretion by BW1 and BWTG3 cells is 10 mug/24 h/10(6) cells. Both albumin and AFP can easily be detected in cell extracts. The simultaneous production of AFP and a hepatocyte specific marker (albumin) by cloned hepatoma cells show that the production of AFP by the tumour is due to the tumoural hepatocytes themselves.     

Extinction, Retention and Induction of Serum Protein Secretion in Hepatoma-fibroblast Hybrids

The production of four serum proteins has been analysed in several hepatoma-fibroblast hybrids. Extinction of albumin and alpha-foetoprotein production occurs systematically in intra and interspecific (rat X mouse) hybrids derived from mouse hepatoma cells (BW). Similar hybrids derived from two related clones of rat hepatoma cells either do not produce albumin (Fa32-derived hybrids), as the BW-derived hybrids, or retain the capacity to produce it, but at a reduced rate (Fu5-derived hybrids); some differences in the control of albumin production thus seem to exist between clonal hepatoma cell lines. The mouse hepatoma cell hybrids retain the capacity to secrete transferrin at a reduced rate, and C3 (the third component of complement) at a high rate. Further analysis of C3 production in interspecific hybrids showed that both parental genomes actively contribute to C3 production: induction of C3 secretion is thus observed in these hybrids.     

A Mouse Hepatoma Cell Line which Secretes Several Serum Proteins Including Albumin and α-foetoprotein

A permanent cell line (BW) was established from a transplantable mouse hepatoma, BW7756, which produces α-foetoprotein (AFP).Three clones were isolated from the uncloned culture: BW1, BW2 and B WTG3. The cells of the latter clone, which was isolated after selection in the presence of thioguanine, are deficient in the enzyme hypoxanthine-guanine-phosphoribosyl transferase. Both B W1 and BWTG3 cells have mean chromosome number of 64 (60 telocentric and 4 metacentric chromosomes). All three clones secrete at least'five serum proteins into the culture medium: albumin, AFP, an a2 globulin, transferrin and C3, the third component of complement. The approximate rate of albumin secretion by BW1 and BWTG3 cells is 10 μg/24 h/10⁶ cells. Both albumin and AFP can easily be detected in cell extracts. The simultaneous production of AFP and a hepatocyte specific marker (albumin) by cloned hepatoma cells show that the production of AFP by the tumour is due to the tumoural hepatocytes themselves.     

Coordinated modulation of albumin synthesis and mRNA levels in cultured hepatoma cells by hydrocortisone and cyclic AMP analogs.

The treatment of Hepa-2 cells, a permanent mouse hepatoma cell line, for 72 h with hydrocortisone (10(-6) M), N6,O2-dibutyryl cyclic AMP (10(-3) M), or 8-bromocyclic AMP (10(-3) M) results in a 2-,3- or 4-fold increase, respectively, in rates of synthesis and secretion of mouse serum albumin. Simultaneous treatment with hydrocortisone and N6,O2-dibutyryl cyclic AMP results in a 10-fold stimulation in these parameters, an effect that is significantly more than additive for the two compounds tested. The number of albumin mRNA sequences, determined by hybridization of total cell RNA to albumin complementary DNA, was increased in direct proportion to the increases in albumin synthesis in all experiments. The relative rate of albumin synthesis approaches in vivo levels in cells treated simultaneously with hydrocortisone and N6,O2-dibutyryl cyclic AMP. We propose that these factors may be necessary to maintain the maximal level of differentiated function in the continuous culture of Hepa-2 cells.     

The control of serum protein synthesis in hepatoma-fibroblast hybrids.

Hybrids between mouse hepatoma cells (which secrete several serum proteins) and mouse or rat fibroblasts (which do not secrete these proteins) produce transferrin and the third component of complement (C3) like the parental hepatoma cells, while they do not secrete either albumin or alpha-fetoprotein (AFP). This lack of albumin and AFP secretion is probably due to a lack of synthesis, rather than to a simple defect in secretion. The cessation of albumin and AFP production is not dependent upon the parental fibroblast nor upon the selection conditions; it is best explained by a shut-off synthesis and could thus reflect the existence of a regulatory mechanism. This would imply a difference between the control of albumin and AFP synthesis and that of transferrin and C3 synthesis. On the other hand, in agreement with Peterson and Weiss (1972), hybrids between rat hepatoma cells and mouse fibroblasts continue to product rat albumin. This suggests that the mouse hepatoma cells differ from the rat hepatoma cells in the way they control albumin production.     

Synthesis of alpha-fetoprotein, albumin and transferrin by long-term cultured cells of murine hepatoma]

The capacity of continuous cell of XXIIa mouse hepatoma (strain MHXXIIa) to synthesize alpha-fetoprotein, albumin and transferrin was studied by immunoautoradiography. Albumin and transferrin were detected in the polyethylene glycol concentrated growth medium of hepatoma cells on the 5th year (the 55th month) of their cultivation. alpha-fetoprotein was not found. Only transferrin was revealed in the growth medium of hepa toma cells of the 8th year (the 92d month) of cultivation. Two clonal cultures obtained on the 8th year of hepatoma cell cultivation were also characterized by the ability to synthesize transferrin. The continuous mouse hepatoma cells retained their malignancy. The agar micro-precipitation reaction showed the presence of alpha-fetofetoprotein in lyfogel concentrated serum of mice with tumors formed after inoculation of the hepatoma cells of the 5th year of cultivation. However, alpha-fetoprotein was not detected in the serum of mice with tumors induced by inoculation of the hepatoma cells of the 8th year of cultivation.     

ESTABLISHMENT OF A CLONAL STRAIN OF HEPATOMA CELLS WHICH SECRETE ALBUMIN

A clonal strain of epithelial cells (designated MH₁C₁) has been established from the transplantable Morris hepatoma No. 7795. The cells have maintained distinctive morphology throughout more than 20 subcultures (split 1:5) at 2- to 4-week intervals in supplemented Ham's F 10 medium. They contain many highly refractile, round, cytoplasmic bodies which stain bright red with Oil Red O. The population doubling time was 2 wk when the clonal strain was first established. It has gradually decreased to 1 wk. The cells synthesize rat serum albumin and secrete it into the culture medium as determined immunologically by microcomplement fixation and double diffusion. Albumin secretion (3–6 µg albumin/mg cell nitrogen/24 hr) occurs throughout the logarithmic phase of cell proliferation and has not diminished during serial propagation since the strain was initiated 15 months ago.     

Effect of anti-lymphotoxin on cell-mediated cytotoxicity. Evidence for two pathways, one involving lymphotoxin and the other requiring intimate contact between the plasma membranes of killer and target cells.

A rabbit anti-lymphotoxin serum produced against partially purified, antigeninduced, guinea pig lymphotoxin, was used to study the role of lymphotoxin in lymphocyte-mediated cytotoxicity in vitro. The anti-lymphotoxin serum inhibited cytolysis resulting from incubation of ovalbumin-immune guinea pig spleen cells with either mouse (P815 mastocytoma) or guinea pig (line 10 hepatoma) target cells in the presence of soluble ovalbumin. The antiserum also inhibited the cytolysis of ovalbumin-coupled target cells by ovalbumin-immune guinea pig spleen cells. In contrast, the anti-lymphotoxin serum did not inhibit: (a) the lysis of line 10 (strain 2) hepatoma cells by spleen cells from alloimmunized Hartley or strain 13 animals; (b) the lysis of line 10 hepatoma cells by spleen cells from tumor-bearing syngeneic animals; or (c) the lysis of P815-mastocytoma cells by spleen cells from P815-immune guinea pigs. These results support the hypothesis that there are at least two distinct pathways by which immune lymphocytes can destroy target cells in vitro, one which involves secretion of a nonspecific soluble factor, i.e., lymphotoxin, and another which probably requires intimate contact between the plasma membranes of the target and killer cells.     

Polyribosome size analysis. Measurement of number-average polyribosome sizes

The analysis of translational efficiencies of specific mRNAs requires a determination of the polyribosome size. The appropriate value to use in such calculations is the number-average size. A method is described for accurately measuring the number-average size of total and of specific protein synthesizing polyribosomes using isokinetic sucrose density gradients and 125I-labeled antibodies. By this method, we demonstrated that albumin synthesizing polyribosomes from a serum albumin secreting mouse hepatoma cell line exist over a broad range from trimers to 20-mers (mean 6-10). The specificity of antibody interaction with polyribosomes was demonstrated using cells not synthesizing mouse serum albumin, and by demonstrating that 125I-anti ovalbumin does not bind to mouse hepatoma polyribosomes. Treatment of the mouse hepatoma cells with 1 MUM cycloheximide shifted practically all of the monomers into polyribosomes resulting in an increase in the number-average size of the albumin synthesizing polyribosomes. Cycloheximide treatment, however, did not eliminate the size heterogeneity in the albumin synthesizing polyribosomes.     

Phenotypic and molecular expression of albumin in rat hepatoma X L cell hybrids

A rat hepatoma cell line (H₄A_ZC₂) was characterized with respect to seven liver-specific phenotypes. Ten clones from the fusion of H₄A_ZC₂ and mouse L cell were analyzed for the expression of these phenotypes. The only hepatic function retained by the hybrid clones was rat albumin synthesis which continued at reduced levels relative to the hepatoma parent. Rat albumin cDNA analysis of RNA from parental and hybrid cells indicated that the reduction in albumin production observed in the hybrids was reflected in coordinate reduction of cytoplasmic rat albumin mRNA.     

Regulation of mouse haptoglobin synthesis

A cloned line of mouse hepatoma cells (Hepa-1) responded to treatment with dexamethasone by a 30-80-fold increase in synthesis and secretion of functional haptoglobin. Under the same conditions, the production of albumin was only slightly elevated whereas that of alpha 1-fetoprotein was reduced by 50%. The hormone concentration for half-maximal stimulation of haptoglobin synthesis was between 1 and 2 X 10(-8) M. The time course of induction is characteristic for a glucocorticoid- regulated protein. Cell-free translation of RNA indicated an increase in the amount of functional haptoglobin mRNA that can account for the change in the protein production. To correlate our findings on Hepa-1 cells with those on nontransformed liver cells, we tested the hormonal response of isolated hepatocytes in tissue culture. Haptoglobin was first synthesized and secreted by hepatocytes from 17-19-d-old fetuses. But neither prenatal nor adult hepatocytes showed a dexamethasone- dependent increase in haptoglobin synthesis. However, when several independent clones of hybrid cells formed from adult mouse hepatocytes and rat hepatoma cells were treated with dexamethasone, the synthesis of mouse haptoglobin was in all cases elevated. It appears that haptoglobin expression in mouse liver cells is potentially sensitive to glucocorticoids, but this modulation is manifested only in transformed cells and their derivatives.     

Expression of human hepatic genes in somatic cell hybrids

Four diploid human cell types (lymphocytes, fibroblasts, amniotic fluid cells, and hepatocytes) were fused to mouse hepatoma cells, HH. HH synthesized and secreted several liver-specific gene products including albumin, transferrin, and alpha-fetoprotein. The resulting interspecific hybrids were compared to determine whether or not the pattern of human hepatic gene expression was similar when these various cells were fused with the mouse hepatoma line. The expression of six human hepatic genes was examined, including albumin, alpha-fetoprotein, ceruloplasmin, transferrin, alpha-1-antitrypsin, and haptoglobin. Albumin was most frequently expressed while alpha-fetoprotein was not detected in any of the hybrids studied. The patterns of expression of human serum proteins differed between the hybrid series. Hybrids derived from human fibroblasts produced primarily albumin, while those derived from lymphoblastoid cells and amniocytes had a higher frequency of clones secreting alpha-1-antitrypsin. The findings reported here suggest that the frequency of hybrid clones expressing human hepatic gene products and the array of proteins produced are influenced by the histogenetic state of the human parental cell type.     

Estrogen-dependent expression of the chicken very low density apolipoprotein II gene in serum-free cultures of LMH cells

Summary The estrogen-responsive Leghorn strain M chicken hepatoma (LMH) cell line provides a model system for studying the estrogen-dependent, liver-specific expression of avian genes. Serum-free culture conditions have been established that allow expression of apolipoprotein B, very low density apolipoprotein II (apoVLDLII), serum albumin, and transferrin at levels detectable by Northern blot analysis. Regulation of apoVLDLII mRNA by estrogen occurred in an appropriate time-and dose-dependent manner in serum-free cultures of the LMH cells. The expression of apoVLDLII mRNA in serum-free culture was at least 100-fold higher than that expressed in cultures containing 10% serum. The level of estrogen receptors in LMH cells cultured with 10% serum was approximately 2000 receptors per cell, and in serum-free culture approximately 1000 receptors per cell. When these cells were transfected with estrogen receptor DNA and cultured in serum-free medium, apoVLDLII mRNA was decreased relative to that expressed in cells transfected with a control plasmid. These results indicate that when the LMH cells are cultured without serum, estrogen receptors are not the limiting factor for the expression of the apoVLDLII gene.     

Detection of lecithin: cholesterol acyltransferase (LCAT) in a human hepatoma cell line

A human hepatoma cell line (HepG-2) was probed for the presence of lecithin: cholesterol acyltransferase (LCAT) using an antiserum to human plasma LCAT. Double immunodiffusion analysis using antiserum to human plasma LCAT revealed a single precipitin line in the sonicated cell homogenate. This precipitin line showed a reaction of identity with highly purified plasma LCAT. The presence of LCAT within the hepatoma cells was also confirmed by an immunofluorescence test. In contrast, the cell culture supernate showed a weak and inconsistent precipitin line. These data suggest that HepG-2 cells synthesize LCAT but secretion of the enzyme by these cells into the culture medium may be partially or totally impaired.     

Expression of differentiated functions in dexamethasone-resistant hepatoma cells

Abstract. The expression of liver-specific functions of different dexamethasone-resistant variants derived from a well-differentiated dexamethasone-sensitive Reuber H35 rat hepatoma cell line (Faza 967) was examined during long-term cultivation. The dexamethasone-sensitive Faza 967 cells are characterized by the activity of tyrosine aminotransferase (TAT) and gluconeogenic enzymes, secretion of serum albumin, and the presence of liver isozymes of alcohol dehydrogenase (L-ADH), aldolase (aldolase-B), and five isoenzymes of lactate dehydrogenase (LDH). The hormone-resistant cells undergo a very dramatic change in expression of most liver-specific functions (dedifferentiation) during long-term culture, in contrast to the sensitive cells in which only certain functions (TAT activity, inducibility, and synthesis of serum albumin) exhibit considerable changes. The hormone-dependent growth sensitivity and the expression of other differentiated functions is not controlled in coordinated way in Faza 967 cells. The time course of the expression of liver-specific functions shows that the cells are resistant before they became 'dedifferentiated', i.e., loss of these liver-specific functions is not a prerequisite of the establishment of the hormone-resistant state.     

Expression of differentiated functions in dexamethasone-resistant hepatoma cells

The expression of liver-specific functions of different dexamethasone-resistant variants derived from a well-differentiated dexamethasone-sensitive Reuber H35 rat hepatoma cell line (Faza 967) was examined during long-term cultivation. The dexamethasone-sensitive Faza 967 cells are characterized by the activity of tyrosine aminotransferase (TAT) and gluconeogenic enzymes, secretion of serum albumin, and the presence of liver isozymes of alcohol dehydrogenase (L-ADH), aldolase (aldolase-B), and five isoenzymes of lactate dehydrogenase (LDH). The hormone-resistant cells undergo a very dramatic change in expression of most liver-specific functions (dedifferentiation) during long-term culture, in contrast to the sensitive cells in which only certain functions (TAT activity, inducibility, and synthesis of serum albumin) exhibit considerable changes. The hormone-dependent growth sensitivity and the expression of other differentiated functions is not controlled in coordinated way in Faza 967 cells. The time course of the expression of liver-specific functions shows that the cells are resistant before they became 'dedifferentiated', i.e., loss of these liver-specific functions is not a prerequisite of the establishment of the hormone-resistant state.     

Regulation of alpha-fetoprotein and transferrin synthesis and secretion in mouse hepatoma cells

Significant differences in the glucocorticoid- and cyclic nucleotide-mediated regulation of the secretory glycoproteins, α-fetoprotein and transferrin, have been observed to develop in a mouse hepatoma cell line, Hepa-2, after many passages in culture. Treatment of low-passage cells with hydrocortisone (10^?6m), N⁶,O²-dibutyryl cyclic AMP (10^?3m), or 8-bromo-cyclic AMP (10^?3m) results in 1.5-, 2- to 4-, and 5.5- to 6-fold increases, respectively, in the rates of synthesis and secretion of α-fetoprotein. As expected of secretory proteins, the ratio of synthesis to secretion is 1 and remains unaltered when treatment with hydrocoritsone, N⁶,O²-dibutyryl cyclic AMP, and 8-bromo-cyclic AMP causes a stimulation of synthesis and secretion. Similar studies showing that albumin and transferrin synthesis and secretion are also balanced in these low-passage cells have been published and indicate that the regulation of synthesis and secretion remains coupled in these low-passage cells. In high-passage Hepa-2 cells, however, we have shown that the relative rate of α-fetoprotein synthesis is higher than its rate of secretion and that the ratio of synthesis to secretion is 4. Similarly, the ratio of transferrin synthesis to secretion is 3.6, whereas it remains unaltered for albumin. When the high-passage cells are treated with N⁶,O²-dibutyryl cyclic AMP, there is a greater increase in the rate of secretion for both glycoproteins, resulting in a reduction of the ratio of synthesis to secretion from 4 to 1.63 for α-fetoprotein and from 3.6 to 2.3 for transferrin. This effect on the secretion of α-fetoprotein and transferrin is specific for the cyclic nucleotides and occurs only in high-passage cells. Hydrocortisone treatment causes an increase in α-fetoprotein synthesis and secretion. However, the ratio of synthesis to secretion increases from 3.96 in control to 5.5 in treated cells. Our studies show, therefore, that there is an increase in this ratio because of a slightly greater effect on synthesis which is not reflected in secretion. Similarly, hydrocortisone exerts a greater increase in transferrin synthesis than secretion and causes the ratio of synthesis to secretion to increase from 3.6 to 6.2. We propose that during continued subculturing a Hepa-2 variant is selected in which the regulation of serum glycoprotein synthesis and secretion is uncoupled. Furthermore, this effect is specific for secretory glycoproteins since the regulation of albumin synthesis and secretion by hydrocortisone and cyclic nucleotides remained unaltered.     

Generation and chracterization of variants of mouse hepatoma cells with defects in hepato-specific gene expression. I. Albumin synthesis variants

Clonal variants of mouse hepatoma cells that either fail to produce albumin (variant 19/2) or show significantly reduced levels (100-fold less) of albumin production (variant 1/c/1) were isolated from the parental line. Hepa la, after a single exposure to N-methyl-N'-nitrosoguanidine (MNNG). Intracellular levels of albumin in both variants were below detection by our assay. Analyses by cDNA-RNA reassociation kinetics indicate that there are approximately 3900 molecules of cytoplasmic albumin mRNA per cell in the parent and less than 10 molecules per cell in both variants. Southern blotting of the Eco RI restriction fragments of cellular DNA from the parent and variants did not indicate any major deletions in the albumin gene DNA sequences. We conclude that in the two variants studied, processes that regulate albumin production via alterations in the level of cytoplasmic albumin mRNA have been affected. Our analyses have also shown that alpha-fetoprotein (AFP) production is lacking in one variant (19/2) and is slightly reduced in the other (1/c/1). Transferrin secretion is lower than the parental line in both variants. Thus multiple nonlethal defects in hepatic gene expression can be obtained in Hepa la cells in culture that will be useful in determining the number and kinds of genes that control the expression of liver-specific loci.