Mapping sequence specific DNA-protein interactions: a versatile, quantitative method and its application to transcription factor XF1.

We have developed a method for the quantitative, exhaustive sequence specificity determination of DNA-binding proteins. The QuESSD method overcomes the limitations inherent in other published in vitro selection methods, not only defining the consensus sequence, but also quantifying the effect on DNA-protein affinity of replacing each base in the recognition domain with every other base. The features distinguishing this method from other in vitro selection approaches are: (1) instead of synthesizing one target oligonucleotide population containing a long randomized domain, we synthesize several oligonucleotide populations, each randomized at two positions. (2) Instead of carrying out several cycles of selection and amplification, we carry out a single cycle. (3) We have developed data collection and analysis procedures that eliminate artifacts and allow generation of quantitative results. The QuESSD method yields accurate measures of: (a) the selectivity of the protein for each base at each position within the recognition domain (normalized relative selectivity), (b) the contributions of individual sites within the recognition domain to the binding affinity (selectivity variance), (c) the relative binding affinity of any given sequence (global selectivity). We confirmed results by (1) tabulating directly the frequency of appearance of individual species in the pool of protein-bound oligonucleotides by cloning and sequencing individual oligonucleotides, and (2) competition EMSA analysis of oligonucleotides designed on the basis of QuESSD data. We have used this method to map the sequence specificity of the nuclear protein XF1 and to distinguish the sequence specificities of XF1 and the AH receptor complex, both of which bind to XRE1, a xenobiotic responsive element (XRE) located upstream of the CYP1A1 gene. Using data obtained by the QuESSD method, we designed oligonucleotides specific for XF1 or for the AH receptor, and prepared CAT reporter gene constructs carrying these oligonucleotides, or wild-type XRE1, upstream of a minimal promoter. Transfection studies using these constructs indicated that XF1 can function as a weak activator of basal transcription, and can, under some circumstances, compete with the AH receptor for binding to XRE1.

Down-regulation of nuclear aryl hydrocarbon receptor DNA-binding and transactivation functions: requirement for a labile or inducible factor.

Aryl hydrocarbons (AHs) such as 2,3,7,8-tetrachlorodibenzo-p-dioxin and benzo[a]pyrene activate the sequence-specific DNA-binding activity of the AH receptor. In the rat hepatocyte-derived cell line LCS7, DNA-binding activity peaked after 30 min and was then down-regulated, reaching negligible levels by 2 h. Down-regulation could be blocked, and DNA-binding activity maintained at maximum for many hours by inhibiting protein or RNA synthesis, implying that down-regulation is a mediated process requiring a labile or inducible protein. CYP1A1 transcription and in vivo DNA-protein interactions at xenobiotic response elements were down-regulated in parallel with DNA-binding activity in nuclear extracts, and these changes could also be blocked by inhibitors of protein synthesis. The correlation between AH receptor DNA-binding activity, intensity of in vivo footprints at xenobiotic response elements, and CYP1A1 transcription rate implies that down-regulation of AH receptor DNA-binding activity is important in regulating CYP1A1 transcription and that receptor is required continuously to maintain transcription. This correlation extends to the murine hepatoma cell line Hepa-1c1c7, in which slower kinetics of activation and down-regulation of CYP1A1 transcription paralleled slower activation and down-regulation of AH receptor DNA-binding activity. The difference in kinetics between cell lines also implies that AH receptor DNA-binding activity is modulated by a mechanism that may be influenced by cell-specific regulatory pathways. The above observations in conjunction with mixing experiments and comparisons of cytoplasmic and nuclear extracts indicate that down-regulation of AH receptor DNA-binding activity is probably due either to degradation or to conversion of the receptor to form that is inactive in both DNA binding and transactivation.

Aryl hydrocarbon-induced interactions at multiple DNA elements of diverse sequence--a multicomponent mechanism for activation of cytochrome P4501A1 (CYP1A1) gene transcription.

In vivo footprinting experiments, augmented with gel shift and transfection analyses suggest that activation of the CYP1A1 gene by aryl hydrocarbons may be a multicomponent process. During the first 30 minutes of exposure to aryl hydrocarbon carcinogens and environmental contaminants, in vivo footprints appear at nine distinct sites within a 281 bp region centered 950 bp upstream of the CYP1A1 transcription start site. Six of these sites are unrelated in sequence to the three xenobiotic response elements (XREs) within this region, at which the aryl hydrocarbon (AH) receptor is known to bind. These six display a variety of footprint patterns, are diverse in sequence and range in G-C content from 60 to 75%. This diversity suggests that multiple nuclear factors may be responsible for these six in vivo footprints. These observations are consistent with competition gel shift experiments showing that the nuclear factors binding at two of these sites are different from each other, as well as from the AH receptor. Gel shifts also indicate that the sequence-specific factors binding at these sites are expressed constitutively. This is consistent with a model in which in vivo footprints are induced at these six sites, not through direct activation or de novo synthesis of DNA-binding factors, but through a two phase mechanism in which binding of the nuclear AH receptor complex to XREs facilitates the binding of constitutive factors at these sites. This facilitation could be mediated either through specific protein-protein interactions or through alterations in chromatin structure that make these sites accessible to constitutive nuclear factors. A function for the sequences at which aryl hydrocarbons induce in vivo footprints is suggested by transfection experiments showing that one of these sequences cooperates with a weak XRE to confer on a reporter gene responsiveness to aryl hydrocarbons.

Down-regulation of P4501A1 and P4501A2 mRNA expression in isolated hepatocytes by oxidative stress.

We have previously shown that the inflammatory mediator interleukin-1 suppressed transcription of CYP1A1 and CYP1A2 mRNAs (Barker, C.W., Fagan, J.B., and Pasco, D.S. (1992) J. Biol. Chem. 267, 8050-8055). Since many of the actions of inflammatory mediators are mimicked by oxidative stress, we treated isolated hepatocytes with 0.25-1.0 mM H2O2 to determine whether expression of these genes is also modulated by oxidative stress. Inducer-dependent accumulation of CYP1A1 and CYP1A2 mRNAs were maximally reduced approximately 50 and 70%, respectively, by 1.0 mM H2O2. Run-on transcription analysis suggested that the effect of H2O2 was mediated transcriptionally. The reduction in CYP1A mRNA levels was not due to a reduction in the levels of all mRNAs due to some general toxic effect since H2O2 did not reduce glyceraldehyde-3-phosphate dehydrogenase, alpha-tubulin, beta-fibrinogen, or albumin mRNA levels, and did not increase lactate dehydrogenase released into the medium. Insulin-mimicked H2O2 action, reducing the expression of both mRNAs, and N-acetylcysteine, which increases intracellular glutathione levels, completely reversed the insulin effect on both mRNAs and the H2O2 effect on CYP1A1 mRNA, but only partially reversed the H2O2 effect on CYP1A2 mRNA. This study indicates that the CYP1A1 and CYP1A2 genes are responsive to oxidative stress and that the majority of this responsiveness can be modified by cellular redox potential.

Inducer-dependent transcriptional activation of the P4501A2 gene in vivo and in isolated hepatocytes.

In vitro nuclear run-on transcription analysis using probes directed against different regions of CYP1A2 revealed that the 70-100-fold induction of CYP1A2 mRNA by polycyclic aromatic compounds is associated with a corresponding increase in the transcriptional activation of this gene in rat liver. Probes from regions of the 1st, 2nd, and 4th introns detected approximately 50-100-fold higher CYP1A2 run-on transcription in liver nuclei from inducer-treated animals than in nuclei from untreated animals. The run-on signals from untreated rats were 3-5-fold above background signals. Additional experiments using single-stranded DNA probes and a probe from a region 5' to the CYP1A2 transcription start site revealed that the inducer-dependent transcripts were colinear with the CYP1A2 mRNA and that they did not result from read through of an initiation event 5' to CYP1A2. Run-on transcription analyses were also carried out with nuclei from isolated hepatocytes using the same series of probes spanning CYP1A2. These analyses indicated that the inducer-dependent accumulation of CYP1A2 mRNA in hepatocytes is associated with at least a 20-fold increase in CYP1A2 transcription. In contrast to liver and hepatocytes, these probes failed to detect run-on transcripts from kidney nuclei, indicating that the lack of CYP1A2 mRNA in this tissue is due to the lack of transcriptional activation of this gene by polycyclic aromatic compounds.

Interleukin-1 beta suppresses the induction of P4501A1 and P4501A2 mRNAs in isolated hepatocytes.

Animals subjected to immunostimulatory conditions exhibit reduced tissue levels of total cytochrome P450 and P450-dependent drug metabolism. We have investigated the possibility that depressed levels of two carcinogen-metabolizing cytochrome P450s may be due to decreased levels of the mRNAs encoding these enzymes by studying the effect of monocyte-derived cytokines on the induction of CYP1A1 and CYP1A2 mRNAs in isolated rat hepatocytes. Medium conditioned by activated human peripheral blood monocytes or by the U937 monocyte cell line suppressed the induction of both mRNAs by 2,3,7,8-tetrachlorodibenzo-p-dioxin, whereas beta-fibrinogen mRNA levels increased 30-40-fold. CYP1A2 mRNA induction was maximally inhibited more than CYP1A1 mRNA (approximately 95 and 65%, respectively), and lower concentrations of conditioned medium suppressed CYP1A2 mRNA induction (half-maximal at 1.9 and 3.1%, respectively). Low concentrations of recombinant interleukin-1 suppressed the inducer-dependent accumulation of both CYP1A1 and CYP1A2 mRNAs in a dose-dependent fashion (half-maximal at 2 and 0.5 units/ml, respectively), while two other monocyte-derived cytokines, interleukin-6 and transforming growth factor-beta, did not. Run-on transcription analysis demonstrated that conditioned medium and interleukin-1 rapidly suppressed the transcription rate of CYP1A1 and CYP1A2 in inducer-treated hepatocytes. The close correspondence between the reductions in CYP1A1 and CYP1A2 transcription rates and mRNA levels suggest that conditioned medium and interleukin-1 suppress the induction of these mRNAs principally through a transcriptional mechanism.

Inhibition of hypochlorous acid by lidocaine and native components of alveolar epithelial lining fluid.

Alveolar epithelial lining fluid (ELF) contains several antioxidant substances that may provide in vivo protection. We studied the ability of ELF and ELF components to inhibit the neutrophil oxidant hypochlorous acid (HOCI). Normal bronchoalveolar lavage fluid containing ELF was incubated with physiologically relevant concentrations of HOCI (0.04 mM). After incubation, residual HOCI activity was titered by the iodide method. The inhibitory activity of lavage fluid was unexpectedly strong. For example, lavage fluid diluted 20-fold in the assay system quenched 49% of starting HOCI. We initially postulated that ELF total protein and glutathione would account for most of the inhibition of HOCI. However, several experimental approaches demonstrated that the total protein and glutathione concentrations in diluted lavage fluid were too low to explain the observed inhibition. Instead, the majority of HOCI inhibition was due to the lidocaine used for upper airway anesthesia. Reagent lidocaine exhibited strong reactivity in the HOCI assay system. Furthermore, the lavage fluid lidocaine concentration (32.4 +/- 6.9 micrograms/ml) was sufficient to explain most of the observed quenching activity. Additional experiments explored the hypothetical quenching activity of ELF components devoid of lidocaine. These findings demonstrate the technical problems posed by lidocaine in antioxidant studies involving lavage fluid or ELF.

Multiple DNA-binding factors interact with overlapping specificities at the aryl hydrocarbon response element of the cytochrome P450IA1 gene.

Three nuclear factors, the Ah receptor, XF1, and XF2, bind sequence specifically to the Ah response elements or xenobiotic response elements (XREs) of the cytochrome P450IA1 (P450c) gene. The interactions of these factors with the Ah response element XRE1 were compared by three independent methods, methylation interference footprinting, orthophenanthroline-Cu+ footprinting, and mobility shift competition experiments, using a series of synthetic oligonucleotides with systematic alterations in the XRE core sequence. These studies established the following (i) all three factors interact sequence specifically with the core sequence of XRE1; (ii) the pattern of contacts made with this sequence by the Ah receptor are different from those made by XF1 and XF2; and (iii) although XF1 and XF2 can be distinguished by the mobility shift assay, the sequence specificities of their interactions with XRE1 are indistinguishable. Further characterization revealed the following additional differences among these three factors: (i) XF1 and XF2 could be extracted from nuclei under conditions quite different from those required for extraction of the Ah receptor; (ii) XF1 and XF2 were present in the nuclei of untreated cells and did not respond to polycyclic compounds, such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and beta-napthoflavone, while nuclear Ah receptor was undetectable in untreated cells and rapidly increased in response to TCDD; (iii) inhibition of protein synthesis did not affect the TCDD-induced appearance of the Ah receptor but substantially decreased the constitutive activities of XF1 and XF2, suggesting that the Ah receptor must be present in untreated cells in an inactive form that can be rapidly activated by polycyclic compounds, while the constitutive expression of XF1 and XF2 depends on the continued synthesis of a relatively unstable protein; (iv) the receptor-deficient and nuclear translocation-defective mutants of the hepatoma cell line Hepa1, which are known to lack nuclear Ah receptor, expressed normal levels of XF1 and XF2, suggesting that the former factor is genetically distinct from the latter two; and (v) a divalent metal ion, probably Zn2+, is known to be an essential cofactor for the Ah receptor but was not required for the DNA-binding activities of XF1 and XF2. Together, these findings indicate that the Ah receptor is distinct from XF1 and XF2, while the latter two activities may be related. Because the DNA-binding domains of these three factors overlap substantially, their binding to XREs is probably mutually exclusive, which suggests that the interplay of these factors at Ah response elements may be important to the regulation of CYP1A1 gene transcription. The results of preliminary transfection experiments with constructs harboring XREs upstream of the chloramphenicol acetyltransferase gene driven by a minimal simian virus 40 promoter are presented that are consistent with this hypothesis.

Aryl hydrocarbon (Ah) receptor DNA-binding activity. Sequence specificity and Zn2+ requirement.

The aryl hydrocarbon (Ah) receptor, also called the xenobiotic or TCDD receptor, mediates transcriptional activation of the cytochrome P-450c (CYP1A1) gene by interacting with Ah or xenobiotic response elements. This paper presents evidence that a metal ion, probably Zn2+, is an essential cofactor for the Ah receptor. This paper also maps in detail the interactions between the Ah response element XRE1 and the Ah receptor from the rat hepatocyte-derived cell line LCS7. Interactions were mapped by three methods, 1) methylation interference footprinting, 2) mobility shift competition experiments, using a series of synthetic oligonucleotides with systematic alterations in the Ah response element core sequence, and 3) orthophenanthroline/Cu+ footprinting. These findings suggest the following consensus sequence for DNA recognition by the Ah receptor: CNA/TNA/TCACGCA/TA/T. The chelators 1,10-phenanthroline and oxalic acid inhibited the sequence-specific DNA-binding activity of the AH receptor in a concentration dependent manner, suggesting that the DNA-binding activity of the receptor requires divalent metal ions. Inhibition was due to metal-chelation, since: 1) inhibition was almost completely prevented by the presence of Zn2+, or other divalent metal ions having high affinity for the chelators used, while metal ions with low affinity did not protect; 2) the DNA-binding activity of the receptor could be restored by dialysis to remove 1,10-phenanthroline, but only in the presence of Zn2+, while dialysis in the absence of metal ions reversed inhibition by the nonchelating isomer 4,7-phenanthroline. The involvement of a divalent cation in receptor function, possibly bound via sulfhydryls, was also suggested by the finding that Cd2+ and Co2+ inhibited DNA-binding activity. Once bound to the XRE1 DNA sequence, the receptor could not be inhibited by 1,10-phenanthroline, suggesting that the essential metal ion must become inaccessible to chelation when the receptor binds DNA. The Zn2+ requirement of the Ah receptor is similar to that of the estrogen and the glucocorticoid receptors and is consistent with the hypothesis that the Ah receptor is a member of the steroid and thyroid hormone receptor superfamily.

Efficient DNA-mediated gene transfer into primary cultures of adult rat hepatocytes.

An efficient, simple, and reproducible DNA-mediated gene transfer procedure has been developed for primary cultures of adult rat hepatocytes. Calcium phosphate-DNA precipitate is formed in complete culture medium during 5 hr incubation with cells. Unabsorbed precipitate is then washed out, and 40 hr later gene expression is measured. Under optimal conditions, up to 20-25% of cells in cultures transfected with the beta-galactosidase (lacZ) gene stain positively for this activity, and cells transfected with the chloramphenicol acetyl transferase (CAT) gene, fused to a strong promoter, express CAT activities of 10-14 nmoles/min per mg protein. Five conditions were optimized based on transfection efficiency, CAT expression, and cell viability. (i) Medium composition: the presence of protein, such as fetal bovine serum or bovine serum albumin, in the medium was essential. (ii) Cell substratum: tissue culture plastic was superior to calf skin collagen and Matrigel. (iii) Cell density: 0.5-1.0 X 10(6) cells/60-mm dish were superior to higher densities. (iv) Duration of exposure to calcium phosphate-DNA: 5-8 hr was better than shorter or longer times. (v) Length of time hepatocytes were maintained in culture before initiating transfection: 2-3 days was superior to earlier times. This procedure was successful with reporter genes linked to three different eukaryotic promoters. These included a chimeric promoter containing the polycyclic aromatic hydrocarbon-responsive enhancer of the cytochrome P450c gene (CYP1A1), which was shown to confer upon the CAT gene responsiveness to polycyclic aromatic hydrocarbons comparable to that of the native P450c gene. This transfection procedure should be of considerable use for the study of liver-specific gene expression in primary hepatocyte cultures.

Transient superinducibility of cytochrome P450c (CYP1A1) mRNA and transcription.

Simultaneous treatment of the rat hepatocyte-derived cell line LCS7 with cycloheximide and polycyclic aromatic compounds increased CYP1A1 (cytochrome P450c) gene transcription rate four- to sixfold and mRNA levels 20-fold relative to the levels in cells treated with inducers alone. When cycloheximide was added up to 1 hr after inducer, a similar degree of superinduction occurred. However, if cycloheximide was added at 1.5 hr or later, superinduction did not occur, even though an increased transcription rate continued at these times in cells treated with inducers alone. Thus, treatment with cycloheximide revealed two phases to the response of the CYP1A1 gene to inducer. During the early phase, inhibition of protein synthesis could amplify the effect of inducer. During the later phase, transcription rate and CYP1A1 mRNA levels remained elevated due to inducer treatment, but could not be further elevated by inhibiting protein synthesis. The superinduction of CYP1A1 mRNA was also examined in primary hepatocyte cultures and in explant cultures of three tissues. These varied substantially in their superinduction response. All of these exhibited elevated levels of CYP1A1 mRNA following simultaneous treatment with inducer and cycloheximide; transcription rate was superinduced three- to fourfold in primary hepatocytes, and CYP1A1 mRNA levels were superinduced 20- to 25-fold in both kidney and lung explants. However, the delayed addition of cycloheximide had varying effects in different culture systems. In primary hepatocytes, addition of cycloheximide as long as 4 hr after addition of inducer resulted in superinduction equal to that which occurred when these agents were added together. In contrast, adding cycloheximide only 2.5 hr after adding inducer resulted in undetectable superinduction in kidney explants and diminished superinduction 70% in lung explants. Although the time course of cycloheximide responsiveness following inducer treatment has been studied in detail only in LCS7 cells, it appears that the length of the cycloheximide-responsive phase is different in different cell culture systems in in explants from different tissues. This may be related to the previously reported tissue-specific differences in CYP1A1 gene expression.

Transcriptional and post-transcriptional regulation of the genes encoding cytochromes P-450c and P-450d in vivo and in primary hepatocyte cultures.

In both primary cell cultures of rat hepatocytes and in liver, polycyclic aromatic hydrocarbons (PAHs) were found to influence the accumulation of the cytochrome P-450c and P-450d mRNAs by both transcriptional and post-transcriptional mechanisms. Following treatment with PAHs, cytochrome P-450c 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 3-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 contributed to the regulation of cytochrome P-450c mRNA levels. Following treatment with PAHs, cytochrome P-450d mRNA levels increased 200-fold in hepatocyte cultures and 70-fold in liver, while transcription rates remained unchanged in hepatocyte cultures and increased only 1.7-fold in liver. This suggests that post-transcriptional mechanisms were of primary importance in regulating cytochrome P-450d mRNA levels. The newly developed hepatocyte primary cell culture system used in these studies differs from previously reported systems in that the cytochrome P-450d gene, as well as the cytochrome P-450c 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 the tissue-specific expression of the genes encoding cytochromes P-450c and P-450d were also examined. The cytochrome P-450c mRNA was found in kidney, heart, and lung, as well as in liver, of PAH-treated rats, while the mature cytochrome P-450d mRNA was detected only in liver. The substantial increase in cytochrome P-450c mRNA in kidney in response to beta-napthoflavone was not associated with a detectable change in the transcription rate of cytochrome P-450c gene, indicating that cytochrome P-450c mRNA levels must be regulated primarily post-transcriptionally in kidney. Even though mature cytochrome P-450d mRNA could not be detected in kidney, the cytochrome P-450d gene was transcribed at a substantial rate in this tissue; therefore, the lack of accumulation of mature cytochrome P-450d mRNA in kidney must have been due to post-transcriptional control.

Noncoordinate regulation of the mRNAs encoding cytochromes P-450BNF/MC-B and P-450ISF/BNF-G.

The mRNAs encoding the major polycyclic aromatic hydrocarbon-induced cytochromes P-450 from rat, P-450BNF/MC-B and P-450ISF/BNF-G, were characterized using three classes of recombinant plasmids: those complementary to (a) only P-450BNF/MC-B mRNA, (b) only P-450ISF/BNF-G mRNA, and (c) both mRNAs. These classes were identified by hybridization-selected translation and immunoprecipitation using six monoclonal and polyclonal antibodies and were later sequenced to confirm their identity and specificity. These findings indicated that the mRNAs encoding these two P-450s have regions that are unique, as well as regions that are homologous. Hybridization-selected translation also showed that the primary in vitro translation products of the P-450BNF/MC-B and P-450ISF/BNF-G mRNAs are 55 and 52 kDa, respectively, and have both unique and common structural characteristics that can be distinguished immunologically. By Northern hybridization, the P-450BNF/MC-B mRNA was found to be 2900 bases long, while the P-450ISF/BNF-G mRNA was 2100 bases long. Precursors of 3500 and 5200 bases were detected for P-450BNF/MC-B mRNA, while a 3100-base precursor was detected for P-450ISF/BNF-G mRNA. These two mRNAs were induced by beta-naphthoflavone, isosafrole, and 3-methylcholanthrene, but not by phenobarbital. In untreated rats, the P-450BNF/MC-B mRNA was consistently present at very low levels while the P-450ISF/BNF-G mRNA was present in variable amounts, suggesting that the latter mRNA can be induced by dietary or other environmental factors. The kinetics of induction of the P-450BNF/MC-B and P-450ISF/BNF-G mRNAs were measured by dot blot hybridization. P-450BNF/MC-B mRNA increased rapidly, reaching half-maximum by 4 h after treatment with 3-methylcholanthrene, while the P-450ISF/BNF-G mRNA increased more slowly, reaching half-maximum after 12 h. The levels of both mRNAs peaked at 24 h, but decreased thereafter at different rates; P-450BNF/MC-B mRNA dropped by about 20% during the next 24 h, while P-450ISF/BNF-G mRNA dropped by 50 to 70%. These differences in the kinetics of induction and the apparent stabilities of the P-450BNF/MC-B and P-450ISF/BNF-G mRNAs, in conjunction with the observed differences in their levels in untreated rats, suggested that these two mRNAs were not coordinately regulated even though they were induced by the same compounds.(ABSTRACT TRUNCATED AT 400 WORDS)

Synthesis and degradation of 3-methylcholanthrene-inducible cytochromes P-450 and their mRNAs in primary monolayer cultures of adult rat hepatocytes.

We used primary nonproliferating cultures of adult rat hepatocytes to investigate the regulation of P-450c and P-450d, immunochemically related protein products of separate cytochromes P-450 genes that are coinduced by 3-methylcholanthrene and related compounds. In cultures of hepatocytes prepared from untreated rats and incubated in media containing 3-methylcholanthrene, beta-naphthoflavone, 3,4,3',4'-tetrachlorobiphenyl, and Aroclor 1254 (a mixture of chlorinated biphenyls) there was a 5- to 15-fold accumulation of P-450c protein (quantitated by immunoblotting), accompanied by an increased rate of P-450c synthesis (measured as incorporation of [3H]leucine into immunoprecipitable protein) and an increased amount of P-450c mRNA hybridizable to a specific cloned cDNA (p210). In contrast, there were no increases in the concentration of P-450d protein, its rate of synthesis, or the amount of P-450d mRNA hybridizable to its specific cDNA (p72). Similarly, when "preinduced" hepatocytes (isolated from rats treated with Aroclor 1254) were incubated for 4 days in culture medium, the amount of P-450c, its rate of synthesis, and the amount of P-450c mRNA remained elevated, whereas synthesis of P-450d and the amount of P-450d mRNA fell precipitously to less than 10% of the initial values despite the presence or absence of Aroclor 1254 or of isosafrole in the medium. However, the loss of P-450d protein in these cultures was almost completely prevented when isosafrole was added to the culture medium and was partially prevented when safrole, Aroclor 1254, and 3,4,5,2',4',5'-hexachlorobiphenyl, but not 3-methylcholanthrene, beta-naphthoflavone, or 3,4,3'4'-tetrachlorobiphenyl, were in the culture medium. Moreover, in similar cultures of "preinduced" hepatocytes that were pulse-labeled with [3H]leucine, the presence of isosafrole in the culture medium extended the apparent half-life for loss of radioactivity in immunoprecipitable P-450d to a value of 72 h (3-fold longer than in standard medium) but was without effect on the rate of disappearance of radiolabeled P-450c. We conclude that control of P-450d degradation is an important factor in the regulation of this hemoprotein and that induction of P-450c and P-450d proceed by separate pathways that are spontaneously divorced under standard conditions for primary culture of adult rat hepatocytes.

Identification and quantitation of a 2.0-kilobase messenger ribonucleic acid coding for 3-methylcholanthrene-induced cytochrome P-450 using cloned cytochrome P-450 complementary deoxyribonucleic acid.

We have used a plasmid containing DNA complementary to one of the two size classes of mRNA coding for 3-methylcholanthrene-induced cytochrome P-450 from rat liver to characterize and quantitate that mRNA. The plasmid used was constructed and identified as follows: Total poly(A+) RNA from 3-methylcholanthrene-induced liver was used as a template for cDNA synthesis. Double-stranded cDNA was inserted into plasmid pBR322 by the G-C tailing procedure. Recombinants were screened by colony hybridization using as probe [32P]cDNA prepared from partially purified cytochrome P-450 mRNA. A differential screening approach was used in which duplicate filters were hybridized with probe from either 3-methylcholanthrene treated or untreated rats. Plasmid p23 was strongly positive by colony hybridization. It was conclusively shown to contain cytochrome P-450 cDNA sequences by demonstrating that the mRNA which specifically hybridized to nitrocellulose-bound plasmid p23 could be translated in vitro into peptides that were immunoprecipitable with monoclonal antibodies specific for 3-methylcholanthrene-induced cytochrome P-450. The size and quantity of the mRNA complementary to plasmid p23 were determined by hybridization of the 32P-labeled plasmid to rat liver RNA that had been fractionated by electrophoresis under fully denaturing conditions and transferred to diazobenzyl-oxymethyl-paper. Plasmid p23 hybridized strictly to a single size of mRNA that was about 2000 nucleotides in length, the smaller of the two size classes of mRNA coding for 3-methylcholanthrene-induced cytochrome P-450. From this we concluded that, at least within the region of the mRNA contained within the insert of plasmid p23, the two size classes of 3-methylcholanthrene-induced cytochrome P-450 mRNA were very different in sequence. The mRNA complementary to plasmid p23 was increased about 4-fold after in vivo administration of 3-methylcholanthrene under conditions that result in maximal induction of 3-methylcholanthrene-induced cytochrome P-450 enzymatic activity. This increase in cytochrome P-450 mRNA parallels the increase in cytochrome P-450 enzymatic activity observed after 3-methylcholanthrene administration. These data suggest that the regulation of mRNA levels is an important point of control of cytochrome P-450 gene expression.

Molecular mechanism for the capture and excision of the transforming gene of avian sarcoma virus as suggested by analysis of recombinant clones.

Structural analysis of two cDNA clones, derived from reverse transcripts of avian sarcoma virus 21S mRNA's, reveals unusual features in the organization and expression of the integrated avian sarcoma virus (ASV) proviral DNA and predicts a mechanism for recombination events that will lead to either the capture or the excision of the transforming gene of this virus. The latter is supported by our observation that there is an extensive homologous region on either side of the transforming gene that will allow site-specific deletion or integration to occur. Comparison of the clone derived from the src-specific 21S mRNA coding for the transforming gene product to that derived from the env-specific 21S mRNA coding for the envelope glycoprotein show that the common c region present at the 3' terminus of the ASV genome is 326 bases long. Within this c region are nucleotide sequences that may play key roles in the life cycle of this virus. These regulatory sequences include (i) probable promoter sites for the initiation of transcription, (ii) a polyadenylation signal, and (iii) a sequence that is complementary to the 3' termini of both the env and the src regions, which will allow the generation of transformation-defective deletions.

Sequence rearrangement and duplication of double stranded fibronectin cDNA probably occurring during cDNA synthesis by AMV reverse transcriptase and Escherichia coli DNA polymerase I.

Two cloned cDNAs derived from the mRNA for cell fibronectin have been sequenced, providing evidence that transcription with AMV reverse transcriptase or Escherichia coli DNA polymerase I may not always result in double stranded cDNA that is exactly homologous with its mRNA template. Instead, the sequences of these cloned cDNAs are consistent with the duplication and rearrangement of sequences during synthesis of double stranded cDNA.

Partial purification and characterization of the messenger RNA for cell fibronectin.

Fibronectin mRNA has been partially purified by guanidine extraction, oligo-(dT)-cellulose chromatography and sucrose density gradient centrifugation. We obtain a fraction which programs a wheat germ in vitro translation system to synthesize a polypeptide species which co-electrophoreses with fibronectin in SDS-polyacrylamide gels and which is immunoprecipitated with affinity purified fibronectin-specific IgG. Analysis of this RNA fraction by methyl mercury hydroxide-agarose gel electrophoresis reveals the presence of a band accounting for 30 percent to 50 percent of the ethidium bromide-staining material in the fraction. The RNA of this band has an estimated molecular weight of about 3 million daltons and is greatly reduced in the corresponding RNA fraction from RSV transformed CEF. This RNA has been tentatively identified as fibronectin mRNA.