Transcriptional and post-transcriptional regulation of the genes ...

Vol. 263,No.18. Issue of June 25, pp. ffi71-8676,1988 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 1988 by The American Society for Biochemistry and Molecular Biology, Inc.

@

Transcriptional and Post-transcriptional Regulation of the Genes Encoding Cytochromes P-45Oc and P-450din Vivo and inPrimary Hepatocyte Cultures* (Received for publication, December 31, 1987)

David S . Pasco$(I,Kent W.BoyumO, Sitara N. Merchant$, Stephen C.ChalbergOll, and John B. Fagan5 From the Molecular Biology Laboratory, Departments of $Chemistry and $Biology, Maharishi International University, Fairfield, Iowa 52556

In both primary cell cultures of rat hepatocytes and inliver, polycyclic aromatichydrocarbons (PAHs) were found to influence the accumulation of the cytochrome P-46Oc and P-46Od mRNAsby both transcriptional and post-transcriptional mechanisms. Following treatment withPAHs, cytochrome P-46Oc mRNA levels increased approximately 100-fold in both hepatocyte cultures and in liver, while transcription rates, measured by run-on transcription of isolated nuclei, increased %fold in hepatocyte cultures and 10-fold in liver. The difference in the -fold increases of mRNA level and transcription rate suggests that post-transcriptional, as well as transcriptional, mechanisms cont2ibuted to the regulation of cytochrome P-46Oc mRNA levels. Following treatment with PAHs, cytochrome P-46Od mRNA levels increased 200-fold in hepatocyte cultures and 70-fold in liver, while transcription rates remained unchanged in hepatocyte cultures andincreased only 1.7-fold in liver. This suggests that post-transcriptional mechanisms were of primary importance in regulating cytochrome P-46Od mRNA levels. The newly developed hepatocyte primary cell culture system used in these studies differs frompreviously reported systems in that the cytochrome P460d gene, as well as the cytochrome P-46Oc gene, were expressed in response to PAHs. In this cell culture system the regulation of these two genes was quite similar, although not identical, to that found in liver. The mechanisms controlling thetissue-specific espression of the genes encoding cytochromes P-4SOc and P460d were also examined. The cytochrome P-46Oc mRNA was found in kidney, heart, and lung, as well as in liver, of PAH-treated rats, while the mature cytochrome P-46Od mRNA was detected only in liver. The substantial increase cytochrome in P-45Oc mRNA in kidney in response to 8-napthoflavone was not associated with a detectable change in the transcription rate of cytochrome P-46Oc gene, indicating that cytochrome P-46Oc mRNA levels must be regulated primarily post-transcriptionally in kidney. Even though mature cytochrome P-4SOd mRNA could not be detected in kidney, the cytochrome P-46Od gene was transcribed at a substantial rate in thistissue; therefore, the lack of accumulation of mature cytochrome

* This project was supported by Research Grant R01-CA38655-02 from the National Cancer Institute andby a grant from the Maharishi International University Research Fund. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. B Recipient of National Research Service Award GM 09310-2. 11 To whom correspondence should be addressed.

P-460d mRNA in kidney must have been due to posttranscriptional control.

Cytochrome P-45Oc’ and P-450d from rat and the corresponding enzymes from mouse are induced by PAHs and are capable of both detoxifying these compounds and converting them to highly active carcinogens and mutagens. In recent years the transcriptional regulation of rat P-45Oc and its murine ortholog, PI-450, has been studied in detail (1-8). Although it has been reported (9) that post-transcriptional processes may play a role in murine P1-450 and P3-450 gene regulation, post-transcriptional regulation has not been extensively investigated in themouse and no reports in this area have been published for rat. We have used both in vivo and cell culture approaches to investigate the contributions of transcriptional and post-transcriptional processes to theregulation of the levels of the rat P-45Oc and P-450d mRNAs. These studies indicate that, in liver and in cultured hepatocytes, both transcriptional and post-transcriptional mechanisms are important in controlling P-45Oc mRNA levels, while post-transcriptional mechanisms are of primary importance in controlling P-450d mRNA levels. Tissue-specific expression of specific forms of P-450 may play an important role in determining tissue-specific responses of the organism to carcinogens and pharmacological agents. Evidence, primarily at the protein level, has demonstrated thatmany of the P-450 isozymes characterized to date are expressed in a tissue-specific and inducer-specific manner. These include the phenobarbital-induced P-450s inrabbit (10-12) and rat (13, 14) the steroid biosynthetic P-450s (15), as well as thePAH-induced P-450s in rabbit (16),mouse (9), and rat (17). We have measured the expression of the P-45Oc and P-450d genes in four different rat tissues. These experiments show that the mRNAs encoding P-45Oc and P-450d have distinct tissue-specific patterns of expression and that, at least in liver and kidney, there exist tissue-specific differences in the contributions of transcriptional and post-transcriptional mechanisms to theregulation of these genes. The research reported here makes use of a newly developed rat hepatocyte primary cell culture system in which P-450d, as well as P-450c, can be expressed in response to PAHs. This system promises to be of substantial use in the future investigation of the regulation of the P-45Oc and P-450d genes. The abbreviations used are: P-450, cytochrome P-450; BNF, 8naphthoflavone; 20 X SSC, 0.3 M NaCl, 0.3 M sodium citrate; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; PAH, polycyclic aromatic hydrocarbons; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid.

8671

8672

DifferentialRegulation of P-45Oc and P-450d mRNA Levels EXPERIMENTAL PROCEDURES

RNA Isolation-Rats (180-200 g) were treated with 3-methylcholanthrene (50 mg/kg), BNF (50 mg/kg), or TCDD (10 pg/kg) by intraperitoneal injection in 0.5 ml of corn oil. Tissues were removed 18-24 h after treatment and total RNA wasisolated by the guanidine thiocyanate-CSC1 (18) procedure. Total RNAwas extracted from cultured hepatocytes and from tissues for the run-on transcription experiments utilizing the acid guanidine thiocyanate-phenol-chloroform technique (19). The yield of total RNA from one 60-mm dish was 60-80 pg. Northern Blots-Total RNA (10 pg/lane) was denatured and electrophoresed through formaldehyde-agarose gels as described by Lehrach et al. (20). The RNA was transferred to nitrocellulose in 20 X SSC overnight (21), and the nitrocellulose filters were air dried and baked in uucw for 2 h at 80 "C. Hybridization Conditions and Probes-Prehybridization, hybridization (42 "C, 24 h), and washing were performed as described (21), using probes labeled with [3ZP]dCTPby the random primer technique (22) to a specific activity of 5-10 X 10' cpm/pg DNA. Probes specific for P-45Oc, plasmid p210 (23) and P-450d, plasmid p72 (23) have been described previously. A rat albumin cDNA plasmid (pRSA 57) was a gift from T. Sargent (National Institutes of Health, Bethesda, MD), and pAC.H8-H, a cardiac actin gene, was a gift from N. Battula (National Cancer Institute, Bethesda, MD). A full length @-fibrinogen cDNA has been isolated and sequenced inour laboratory (to be reported elsewhere); a 650-base pair probe, corresponding to the 3' end of the @-fibrinogenmRNA, was used in the experiments reported here. All plasmids were free of repetitive sequences. Autoradiograms were quantitated by densitometry using a Hoefer GS-300 scanning densitometer. Exposure times used were in the linear response range for the film used, Kodak XAR-5. Nuclear Run-on Transcription-Liver (24) and kidney (25) nuclei were isolated from rats 18-20 h after treatment with corn oil or BNF in corn oil and frozen at -80 "C until use. Cultured rat hepatocytes were washed with ice-cold phosphate-buffered saline, scraped, and lysed by Dounce homogenization (5 strokes with pestle B) in ice-cold reticulocyte standard buffer (10 mM Tris-HC1, pH 7.5, 10 mM NaC1, 3 mM MgCl,) containing 0.5% Nonidet P-40. Cell lysis was monitored by phase contrast microscopy. Nuclei (2-5 X lo') were incubated (30 min, 26 "C) in 60 pl of 20mM Tris-HC1 (pH 7.91, 30% glycerol, 2.5 mM dithiothreitol, 5.0 mM MgCl2,3 mM spermidine, 10 mM each of ATP, CTP, and UTP,130 mM KC1,0.05 mM GTP, and84 pCi of [a3ZP]GTP(Du Pont-NewEngland Nuclear, 3,000 Ci/mmol). The KC1 and GTP concentrations were optimized for maximal incorporation. At the end of the reaction 1ml of cold reticulocyte standard buffer was added, nuclei were pelleted a t 15,000 X g for a few seconds, and the supernatant aspirated to remove most of the unincorporated [a32P]GTP.Calf liver tRNA (50 pg) was added to each reaction along with 600 pl of guanidine thiocyanate solution (19) and total RNA extracted by the acid guanidine thiocyanate-phenol-chloroform procedure (19). Nitrocellulose filters containing 1 pgof each of the indicated plasmids were hybridized a t 42 'C for 3 days with 3.0 X 10' cpm of 32P-labeledRNA. Filters were washed and treated with 20 pg/ ml DNase-free RNase A in 2 X SSC at 37 "C for 30 min (26). Hybridizations were linear with increasing 32P-labeledRNA input, cDNAs were in excess, and theaddition of cu-amanitinto thereaction (1 pg/ml) decreased the transcription signal for all genes tested by 80-90%. Autoradiograms were scanned and the densitometric signal corresponding to each mRNA was normalized to that corresponding to actin sequences. Actin mRNA levels quantitated by Northern hybridization did not change in response to TCDD orBNF treatment in uiuo or in vitro. Densitometric data not normalized to actin signals exhibited the same trends asactin-normalized data but with slightly greater variability. Isolation of Rat Hepatocytes-Male Sprague-Dawley rats (200-250 g) were anesthetized with ether and then the livers were perfused through the inferior uenu caua with 200 ml of solution A (0.14 M NaCI, 6.0 mM KCl, 1.5 mg/ml glucose, 0.5 pg/ml insulin, and 10 mM Hepes, pH 7.4) at 37 "C with a flow rate of 12 ml/min. This was then followed bya 10-15 min perfusion with solution A containing 5.7 mM CaC12and 0.05% (w/v) collagenase (Type CLS, Cooper Biomedical). The liver was then removed, minced with scissors in solution A, and filtered through two layers of gauze. The cells were washed twice (50 X g for 2 min) and then subjected to an iso-density Percoll centrifugation step to remove dead and damaged hepatocytes (27). Viability was always over 90%, and yields were from 90 to 160 X lo6 cells/liver. Culture Condition and Procedures-Hepatocytes (1.0 X 106/ml)

were plated in plastic culture dishes (Falcon) with 3 ml (60-mm dishes) or 8.3 ml(100-mm dishes) of medium, which wasa 1:l mixture of Dulbecco's modified Eagles (4.5 mg/ml glucose) and Ham's F-12 with hypoxanthine (Hazelton Research Products, Lenexa, KS). This medium was supplemented with gentamycin sulfate (25 pg/ml), amphoterecin B (0.7 pglml), insulin (5 pg/ml from bovine pancrease, Sigma), and dexamethasone (10" M) throughout the culture period. During the first 24 h the medium contained 10% Nu-Serum (Collaborative Research Inc., Bedford, MA) and, thereafter, 5% Nu-Serum. The culture medium was replaced every 24 h, and cultures were maintained a t 37 "C in a humidified atmosphere containing 5% COZ and 95% air. TCDD (final concentration 1.0 nM) was added to the culture medium in dimethyl sulfoxide (1 pl/ml). RESULTS

Regulation of P-45Oc and P-450d mRNAs in Primary Cultures of Rat Hepatocytes and in Liver-Previous studies have reported the inability to express rat P-450d mRNA, as well as the murine equivalent of this mRNA, in hepatocytes or hepatocyte-derived cell culture systems (28-31). We have developed a rat hepatocyte primary culture system in which the P-450d mRNA, as well as the P-45Oc mRNA, were expressed in response to PAHs to levels similar to those seen in uiuo. We found that the PAH-induced expression of P-450d mRNA did not require the use of exotic substrate ormedium components, but only required the appropriate combination of factors that have already been shown to promote differentiated hepatocyte function. These included 1) the removal of dead and damaged cells before culture throughthe use of isodensity Percoll centrifugation (27); 2) the inclusion of insulin and dexamethasone in theculture medium. Exclusion of these two components reduced TCDD-induced P-450d mRNA accumulation by 25 and 70%, respectively; 3) the inclusion of 5% Nu-Serum in the culture medium. Replacement with 5% fetal bovine serum substantially reduced expression after the first 36 h in culture. Nu-Serum is a medium supplement containing 25% (v/v) newborn bovine calf serum, various growth factors, insulin,transferrin, andother hormones, amino acids, vitamins, and nutrients. 4) Although the studies reported herein used hepatocytes maintained on standard tissueculture plastic dishes, maintenance on a matrix of solubilized tissue basement membrane derived from Engelbreth-Holm-Swarm mouse tumors (Matrigel) furtherenhanced P-450d mRNA inducibility and maintained inducibility for longer periods in culture. We (23) and others (32) have previously reported that the mRNAs for P-45Oc and P-450d were induced with different kinetics in the liver. As shown in Fig. 1, we have reexamined these kinetics using the hepatocyte culture system described above. In these cells, P-45Oc mRNA increased rapidly in response to TCDD, reaching half-maximal levels about 7 h after adding inducer and reaching maximal levels by about 11 h. In contrast, P-450d mRNA increased more slowly,reaching half-maximal levels at about 24 h and reaching maximal levels at about 33 h. The kinetics of induction of both mRNAs were similar to those reported in vivo (23, 32). Although it is not apparent from Fig. 1, there were small but consistent changes in the basal levels of P-45Oc and P450d mRNA after hepatocytes were placed in culture. Basal levels of P-450d mRNA were initially similar to those found in uiuo, but declined after 24 h, reaching levels that were 5% of the initial value after 3 days in culture. The reduced basal P-450d mRNA level accounts for the observation that, although the TCDD-induced levels of P-450d mRNA in this culture system were similar to those found in the livers of PAH-treatedrats, the -fold increase was much greater in culture, approaching 200. Whereas the basal level of P-450d mRNA decreased after placing hepatocytes in culture, that of

Differential Regulation 10

of P-45Oc and P-450d mRNA Levels

8673

-

8 6 4 -

2 0

3

Or 0.0

0 105

25 20 15

30 40 35

M TCDD CONT

C

0 24 48 3 6 1124 33 48

10.0

15.0

20.0

25.0

Time (hour)

45 50

Time (hours)

5.0

H ACTIN

R

O

1

0.5

1.0

3.0

I

P-FGN

P-450d P-450~

ALBUMIN PUC-18 FIG.2. Run-on transcription of the P-45Oc and P-450d genes as a function of time after addition of TCDD. Upper

FIG.1. Cytochrome P-45Oc and P-450d mRNA levels as a function of time after adding TCDD. Twenty-four h afterplacing hepatocytes in culture,TCDD was added to 1.0 nM (squares),or cells were left untreated (diamonds).At each time after TCDD addition, indicated in the figure, total cellular RNA was extracted from one 60-mm dish. A Northern blot of these RNAs was hybridized with cDNA probes specific for P-45Oc mRNA (open symbols) or P-450d mRNA (closed symbols). The upper panel presents the densitometric quantitation of the autoradiogram shown in the lowerpanel. The data shown is from one experiment that was repeated twice with similar results. In the lower panel the lanes labeled C and M contained total RNA from the livers of untreated and 3-methylcholanthrene-treated rats, respectively. CONT, control.

panel, at theindicated times after the addition of TCDD to 1.0 nM, nuclei were isolated from primary hepatocyte cultures (2-3 100-mm dishes/group). Nuclear RNA was labeled with [3ZP]GTPby the runon transcription reaction, isolated, and hybridized to slot blots of the plasmids indicated in the lower panel. After RNase treatment, the slot blots were quantitated by autoradiography and densitometry. Results are expressed as -fold increase over control transcriptional rate +- range (P-45Oc (open squares) and P-450d (solid squares)).The data are from five separate experiments; n = 2 for the 0.5 and 1.0 h time points, n = 3 for the 3 and 12 h points, and n = 1 for the 24 h points. Lowerpanel, autoradiogram from a representative experiment. Numbers indicate time afteraddition of TCDD to theculture medium; 8-FGN, 8-fibrinogen.

transcription rate during the first 24 h following addition of the inducer TCDD. Even at 24 h, when P-450d mRNA levels were increasing rapidly, run-on transcription activity of the P-450d gene was not significantly greater than in controls. This indicates that P-450d mRNA levels must be controlled predominantly by post-transcriptional mechanisms in primary cultures of rat hepatocytes. Tissue-specific Regulation of the P-45Oc and P-450d P-45Oc mRNA increased during the first24 h in culture from mRNAs-Previous studies (23, 32) have shown that the Pthe undetectable level found in the livers of untreated animals 450c and P-450d mRNAs were noncoordinately regulated in to a low but measurable level that was maintained throughout rat liver. Fig. 3 shows that these mRNAs were also noncoordinately regulated in three other tissues. The Northern blot 7 days in culture. Fig. 2 shows that the 100-fold increase in P-45Oc mRNA in Fig.3, upper panel, shows that the 2.9-kilobase P-45Oc level,shown in Fig. 1, was accompanied by only a %fold mRNA was present in measurable amounts in kidney, lung, increase in run-on transcription. The difference in the -fold and cardiac muscle, as well as in liver, from PAH-treated rats, increase in mRNA levelsand in run-on transcription suggests but was not detected in tissues of untreated rats. The P-45Oc that post-transcriptional as well as transcriptional mecha- mRNA was induced to 5-fold higher levels in liver than in nisms must contribute to the regulation ofP-45Oc mRNA other tissues, with the exception of kidney from TCDDlevels in thesecultured hepatocytes. As expected, the kinetics treated ratsin which this mRNA reached levels equal to those of induction of transcription preceded those of mRNA accu- in liver. Fig. 3, lower panel shows that, in contrast toP-45Oc mulation. Run-on transcription activity of the P-45Oc gene mRNA, the 2.1-kilobase P-450d mRNA was detectable only reach half-maximal level within 1h and was maximal by 3 h. in liver, not inkidney, lung, or cardiac muscle, even in animals Fig. 2 also shows that, despite the 200-fold increase in P-450d treated with high levels of TCDD. Similar results were obmRNA, shown in Fig. 1, there was no significant change in tained using MC as inducer (data notshown). Long exposures

DifferentialRegulation of P-45Oc and P-450d mRNA Levels

8674 Li

Ki

Lu

C M T

C M T

C M T

1

I

1

1

l

l

I

l

l

He C M T I

1

+

P-450~

1

I

LIVER KiLuHe C M T T T T

P-450d+

TABLE I Transcription rates and mRNA levels in kidney and liver of untreated and BNF-treated rats Nuclei and total RNA were isolated in parallel from liver and kidney of untreated and BNF-treated rats. The total RNA preparations were used to quantitatemRNA levels by Northern hybridization and the nuclei were used to measure run-on transcription rates. Messenger RNA levels were expressed as average arbitrary densitometric units f S.D. ( n = 3). The abundance of a specific mRNA can be compared between liver and kidney, but comparisons between different mRNAs are notvalid because of differences in probe specific activities. Run-on transcription data was normalized to actin and is expressed as average arbitrary densitometric units f S.D. ( n = 3). The transcription rates of different genes cannot be directly compared, nor can these rates be directly compared between tissues. Relative transcription mRNA level rate Tissue BNFControl BNF Control O.D. X"l0.D. actin O.D. units Liver 9.6 f 2.1 59.8 f 10.4 192 f 69 489 f 107 &Fibrinogen 59 f 10 101 f 21 P-450d 1.0 f 0.04 71.3 f 21 4 6 f 12 493 f 147 P-450~ NDb 39.1 f 5.7 Albumin 1 2 6 f 34 5 4 f 14 5 5 6 f 210 171 f 22 Kidney 0.2 f 0.01 55 f 7 50 f 3 ,%Fibrinogen ND P-450d ND ND 4 1 4f 95 f 6 P-450~ ND 1.8 f 0.2 42 f 2 39 f 7 Albumin ND ND ND a X,P-45Od, P-450c, @-fibrinogen,or albumin. bND, not detectable. The lower limit of detection on Northern blots was