Isolation and Characterization of Rat Cytochrome P-450IIB Gene

0 downloads 0 Views 6MB Size Report
Apr 25, 2015 - gency hybridization oE gene IV fragments to probes comprising the ... periments indicated that the transcription initiation ... transcripts were elevated after PB treatment. .... sion of the unlabeled EX1 oligomer as described above, and unlabeled ..... implicated as hot spots for gene conversion in several gene.
THEJOURNAL

OF

BIOLOGICAL CHEMISTRY

Vol. 264, No. 12, Issue of April 25, pp. 7046-7053,1969 Printed in U.S.A.

0 1969 by The American Society for Biochemistry and Molecular Biology, Inc.

Isolation and Characterizationof Rat CytochromeP-450IIB Gene Family Members USE OF THE POLYMERASE CHAIN REACTION TODETECT EXPRESSION OF A NOVEL P-450 mRNA* (Received for publication, August 15, 1988)

Cecilia M. Ciachelli, Jennifer Lin-Jones, and CurtisJ. Omiecinski From the Department of Environmental Health, School of Public Health and Community Medicine, University of Washington, Seattle, Washington 98195

The rat cytochrome P-450IIB gene family consists of at least 10 members, 2 of which, P-450b and P450e, have been well characterized and are activated transcriptionally by phenobarbital (PB) in the liver. The remaininggenes in thisfamily have not been studied extensively. In this report, data are presented that provideadditionalcharacterization of aP-450IIB gene, termed gene IV. The complete gene IV and flanking regions were isolated from a rat liver Charon 4A genomic library and subjected to a variety of analyses. Structural homology to the P-450b gene was confirmed by comparative restriction mapping and high stringency hybridization oE gene IV fragments to probes comprising the entire cDNA for P-450b. The 5‘-portion of gene IV, including its flanking region, was sequenced and contained an open reading frame for58 amino acids which were 62% related to exon 1 of the P-450bJe genes. Typical TATA and CAAT promoter elements, as well as two Spl core sequences and a site related to a glucocorticoid responsive element were found in gene IV. Northern blot studies with an oligomer probe specific to gene IV sequence indicated a 4.3kilobase pair polyadenylated transcript present at low levels in untreated rat liver and inducible approximately 6-fold by PB treatment. Primer extension experiments indicated that the transcription initiation site mapped to a position on gene IV that was analogous to that reported for the structurally similar P-450e gene. Dueto thelow levels of hepatic RNA expression, we employed the polymerase chain reaction to facilitate characterization of gene IV transcripts. The polymerase chain reaction data verified that gene IV transcripts were elevated after PB treatment. Significantly, the polymerase chain reaction resultsdemonstrated that gene IV transcripts were associated with hepatic polysome fractions, indicatingtheir active utilization in this tissue. Polymerasechainreaction analysis of RNA also indicated constitutiveexpression of gene IV transcripts in rat lung, kidney, and testis, as well as fetal rat Iiver. Together, these data show that gene IV is a transcribed member of the P-450IIB gene family and, like the well characterized P-450b and P-450egenes, is positively regulated by PB inrat liver. * This work was supported by National Institutes of Health Grants GM32281, ES04696, and ES07032. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement’’ in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paperhas been submitted totheGenBankTM/EMBLDataBankwith accession nurnbeds) 504808.

Cytochrome P-450s (P-450s)’ comprise a multigene family whose products are essential inthe biotransformation of numerous endogenously and exogenously derived chemicals (1). Although primarily detoxifying enzymes, P-450s can bioactivate certain substrates, including the procarcinogenic polycyclic aromatic hydrocarbons and myocotoxins, to extremely toxic intermediates. To date, over 60 forms of P-450 have been characterized in numerous species, allowing for the classification of isozymes into 10 families based on amino acid homology (2). P-450s are furtherdivided into subfamilies whose members share at least 68% sequence relatedness. The rat P-450b (P-450IIB1) and P-450e (P-450IIB2) genes are members of the P-450IIB gene subfamily. They share98% homology in their mRNA and amino acid sequences (3, 4) and are identical in intron-exon arrangement (5). Both genes are PB-inducible, but display variation in their constitutive (6), tissue-specific (7), and developmental expression (8, 9). Previous studies utilizing Southern blotanalysis and genomic clone isolation (10,ll) have suggested that atleast ninegenes make up the P-450IIB family. These genes are closely linked on chromosome 1 in the rat (12). Gene conversion represents an important evolutionary mechanism for the maintenance of highly conserved regions within multigene families (13). Indeed, partial sequence analysis of the 3’ portions of the P-450IIB family has provided strong evidence that extensive gene conversion has occurred between its members (14). The striking conservation maintained in these genes suggests the possibility that other members of the family may represent functionally important P450 isozymes. Many questions pertaining to theP-4501IB subfamily, however, still exist. Which other members of the subfamily, besides P-450B and P-450e, represent transcribed genes? Are they PB-responsive? How do the genes differ in their 5‘flanking sequences orotherpotential regulatory regions? Comparative sequence analysis of the P-450IIB subfamily genes can begin to address the question of possible conserved regulatory motifs. However, to ascertain whether other genes in this subfamily are expressed requires the development of sensitive assays which can distinguish highly related gene products. In thispaper, we report the isolation and characterization of rat P-450IIB gene family members, including two genes that have not been reported previously. One member, designated gene IV, was subjected to a variety of analyses. Together with gene sequence, hybridization, and primer extension data, we describe the use of the polymerase chain I The abbreviations used are: P-450, cytochrome P-450; PB, phenobarbital; P-450b, P-450IIB1; P-450e, P-450IIB2; kb, kilobase pairs; bp, base pairs.

7046

Rat P-45OIIB Gene Family reaction to definitivelyshow that gene IV is transcribed constitutively and is inducible by PB in rat liver. MATERIALS ANDMETHODS

7047

antisense-primed reactions, 1250 units/ml of RNasin (Amersham) and 850 units/ml of avian myeloblastosis virus reverse transcriptase (Du Pont-New England Nuclear), and incubated at 42‘C for 1 h. Extensions were stopped by the addition of 2 pl of 2 N NaOH and incubation at 37 “C for 10 min. Samples were neutralized by adding 2 pl of 2 M NH, acetate, and 50 fig of yeast tRNA was added as a carrier. Nucleic acids were precipitated with EtOH, and theresulting pellets were washed twice with 70% EtOH and resuspended in 8 pl of 10 mM Tris-HC1, pH 8, 0.1 mM EDTA. 4 pl of each sample was combined with 2 pl of loading buffer, heated at 80 “C for 2 min, and applied to an 8% acrylamide/6 M urea sequencing gel together with sequencing reactions as size standards. Polymerase Chain Reactions-Tubes treated with Sigmacote (Sigma) were rinsed extensively with diethyl pyrocarbonate-treated water, autoclaved, and used in all reactions. Polymerase chain reactions were performed essentially as described by Saiki etal. (21), with the following modifications: single-stranded cDNA from 5 pgof poly(A+),total, or polysomal RNA was synthesized by primer extension of the unlabeled EX1 oligomer as described above, and unlabeled nucleotides were incorporated into the primer extension products. One-tenth of the cDNA product was combined directly with 50 mM Tris-HCl, pH 8.3,50 mM KCl, 5.0 mM MgClz, 100 pg/ml bovine serum albumin, 0.2 mM each dATP, dCTP, dGTP, and dTTP, and 0.5 p~ each EX1 and gene IV sense primers in a total volume of 50 pl and topped with mineral oil. The entiremixture was heated at 9195 “C for 5-10 min, cooled to 54 “C, and then 2 units of Taq polymerase (New England Biolabs) were added. Primers were allowed to anneal a t 54 “C for 2 min, extended at 70 “C for 1 min, and then denatured for 1 min at 91-95 ‘C. This cycle was repeated 30 times. As controls, polymerase chain reaction was performed on either an equal aliquot of RNA prior to cDNA processing, 0.2 pg of p17A.4 DNA, or 2 pg of rat liver genomic DNA, using the same conditions. Products were assessed by ethidium bromide staining of agarose gels followed by Southern blot analysis or sequencing as detailed above.

Unless otherwise stated, materials were purchased from sources described in previous publications (8,9). Animal Treatment-Three-week-old Sprague-Dawley rats(approximately 75 g)were purchased from Tyler Laboratories, Inc. (Bellevue, WA) and treated intraperitoneally with either 50 mg/kg PB, 40 mg/kg methylcholanthrene, 100 mg/kg dexamethasone from Aldrich, or 500 mg/kg Aroclor 1254obtained from The Foxboro Corp. Genomic Library Screening-The genomic library utilized in this study was provided by Dr. J. Bonner (California Institute of Technology, Pasadena, CA) and was prepared from HaeIII-digested liver DNA from a Sprague-Dawleyrat, cloned into theEcoRI site of Charon 4A phage. Three rounds of library screening were performed as previously described (15), each time plating approximately 500,000 recombinants. The probes used werecDNA and genomic clones representing successively more 5’-P-450b or -e gene sequence as follows. Screen 1 was performed utilizing a mixture of p73 (exons 7 and 8 of the P-450e gene) and clone 10 (part of exon 8 and all of exon 9 of the P-450b gene). Both clones were isolated from cDNA libraries prepared in our laboratory using standard techniques (16). Screen 2 was performed using the 2.7-kb EcoRI fragment of the P-450b gene contained in recombinant clone 18C (exons 5 and 6 and surrounding introns, see Fig. I), isolated during screen 1.Screen 3 was performed using the PstI-BamHI fragment of PB7 (a cDNA containing part of exon 1 and all of exons 2, 3, and 4 of the P-450b gene, see Fig. 2) obtained from Dr. Alan Anderson (17). AllcDNA labelings and hybridizations were performed as previously described (8, 9). After each screen, putative positives were rescreened a t least twice and plaque-purified. Phage were isolated using standard methods (18). Oligomer Probes-The B and E antisense oligomers, 5’-d(GGTTG GTAGCCGGTGTGA)-3’ and 5’-d(GGATGGTGGCCTGTGAGA)3’, are targeted to the hypervariable exon 7 region of P-450b and -e RESULTS genes, respectively. The characterization of these probes has been Isolation and Characterization of P-45OIIB Subfamily Gedescribed (6-9). The EX1 antisense oligomer has the sequence 5’d(GGGACGAGGTCCTGGTGGGAAGTTG)-3’ and is targeted to a nomic Clones-Screening of the rat liver DNA library was region of identity in exon 1 of the P-450b and -e genes. Its comple- undertaken as described under “Materials and Methods” in mentary sequence is shown in parentheses inFig. 4. The sequence of order to obtain a clone bank of P-450IIB gene family memthe gene IV sense oligomer is 5’-d(TCACTTTCACTGTGGGCTT)phage were isolated, a number 3’ and its target site in gene IV is shown bracketed in Fig. 4; an bers. A total of 40 recombinant oligomer containing the complementary sequence was also synthe- of which appeared to be multiple copies of the same genes. sized and is termed the gene IV antisense oligomer. The oligodeoxy- Twenty-two of the fortyclones seemed unique and were nucleotides were synthesized on a Biosearch DNA synthesizer in the characterized by restriction analysis and Sourthern blot hyDept. of Microbiology at theUniversity of Washington. bridization with various cDNA and oligomer probes as shown (5 pg) were Restriction Analysis of Genomic Clones-Phage DNA digested with at least a %fold excess of restriction enzyme for 1 h. in Fig. 1. Based on this information, the clones were placed I-VII, with each clone representSouthern blotting and hybridizations were performed exactly as de- into seven different groups, ing a portion of one gene (indicated by the linkage map). scribed (9). Subcluning and Sequence Analysis-Clone 17A was digested with Clones representing a portion of the P-450b gene (group I) PstI, the resulting 4-kb fragment was isolated in low gelling agar and nearly all of the P-450e gene (group 11) were isolated, as (FMC Corp.) and ligated into the PstI site of pUC18. Exonuclease well as five other members of the P-450IIB gene subfamily. clones were generated using the procedure of Henikoff (19). Polymerase chain reaction products were prepared for sequencing using a Clones in groups I-V all displayed 3’ relatedness to the PCentricon-30 microconcentrator from Amicon. Sequencing was per- 450b/e genes, while groups 11, 111, and IV also hybridized to formed using the dideoxy method of Chen and Seeburg (20) with probes containing extreme 5’-P-450b/esequence(seealso avian myeloblastosis virus reverse transcriptase or Klenow from Du Fig. 2), suggesting that these genes were represented in their Pont-New England Nuclear, or Sequenase from United States Bio- entirety (or near entirety) in our clone bank. Clones i n group chemical Corp. Sequence comparisons were facilitated by Genepro VI hybridized to p73, an exon 7 and 8 cDNA probe, but not computer software (Riverside Scientific). RNA Analyses-Total, poly(A’), and polysomal RNA were isolated to probes representing exons 2 , 3 , 4 , or 9. Probing of rat liver as previously described (8,9). Northernblot analyses were performed genomic blots with p73 suggested that these phage did not exactly as reported (9). Primer extension reactions were performed contain artifactsof the cloning procedure since a fragment of as follows: 50-200 pg of total RNA was combined with 5 pmol of similar size was identified as an intensely hybridizing species unlabeled oligomer (in the case of the gene IV antisense primer), or (data not shown). Group VI1 clones showed hybridization to 32P-oligomergenerated by 5”kinasing (in thecase of the EX1 primer), in 230 p1 of buffer containing 0.75 M NaCl, 20mM Tris-HC1, pH 8.0, cDNAs containing 5’-P-450b/e information exclusively. To and 4 mM EDTA. The mixture was heated to 55 ‘C and slow cooled assess sequences in the exon 7 region of our clones, we used to 42 “C, and hybridizations were continued at this temperature for t h e B and E oligomers which are targeted to the hypervariable 2 h. Hybridizations were terminated by the addition of 2 volumes of regions of P-450b and -e, respectively. Four genes, including 100% EtOH, and nucleic acids were precipitated a t -20 “C. Pellets gene IV, contained t h e B oligomer binding site, and a single were washed twice with 70% EtOH and dried briefly in a Speed-Vac concentrator (Savant). The pellets were resuspended in 20 p1 of a gene, gene 11, contained the E oligomer site. T w o genes, V and VII, did not hybridize to either of these oligomers. Based solution containing 50 mM Tris, pH 8.3, 50 mM KCl, 6 mMMgC12, 0.25 mM each dATP, dTTP, dGTP, and either unlabeled dCTP for on the restriction and hybridization data, we tentatively idenEXI-primed reactions or [32P]dCTP(3000 Ci/mmol) for gene IV tify gene I11 as analogous to the gene isolated by Mizukami et

Rat P-45OIIB Gene Family

7048 23

I

78

4 5 6

I

I

E

B

A

-1tb

P450E

I

B

1

B

E

9

n I

E

{

E

E EfB E N LIP

23

D

9.4

D

6.6

>

44

D

22

>

{

EM LIP

D

-

i

"'

3E uj 8 8

Eta

mmmmmp 112

group VU

PB7(150)

I

E

21

group IV

C

PB7(B/N)

PB7(400)

-

gLy prnpr

8 olilorrm L Z LE ~ o l i ~ m clone 10

I D J J I D Pn pBl(W

9 7 9 7. 8

FIG. 2. Southern blot analysis of clone 17A. Clone 17A was digested with EcoRI alone ( E ) ,or in combination with BamHI ( E / B ) , HindIII (EIH), or PstI (E/P),and hybridized to probes representing different regions of the P-450b cDNA derived from clone PB7 as illustrated below the photographs. Only the E/H and E/P lanes are shown for the autoradiogram probed with PB7(150).Exon 9 homology was determined using clone 10 (see Fig. 1).Molecular sizes in kilobases are indicated by arrows and were determined from the migration of size markers run concurrently.

12.3.4

FIG.1. Characterization of P-450IIB gene clones. Twentytwo genomic clones (numbered lines) were isolated from a rat liver DNA library, restricted with EcoRI, and hybridized to the various probes listed in the key below the diagram. Based on restriction and hybridization data which are indicated in the figure by vertical lines (EcoRI sites) and shading (see key), respectively, the clones were placed into 7 gene groups, I-VII. For gene IV, the boundary between the PB7(400) andthe exon7binding portions of the gene was determined by EcoRI/BamHI digestion (see Fig. 2). A linkage map illustrating thecomposite gene, derived from overlapping clone information,is shown at the top of each group. For comparison, the structure of the P-450e gene (3) is given a t the top of the figure. E = EcoRI; B = BamHI.

A. -1kb

P450B 5'

E

23 111

BHBP

I B

I

I I I I I

I I II

HE H H P

8

P450E

18

E H

I

4-

I

9

m

E H

I

H

3'

I

. I

E

H

E

8

17A

I

PPEB

23

1

I

I E

al. ( l l ) , represented by clones pgP-450pb-1 and pgP-450pb-8,

and gene IV, isolated by Atchison and Adesnick (10). Gene VI may be identical with the gene represented by Mizukami's clone pgP-450pb-5 (11).However, these identifications remain tentative until sequence comparisons of the different forms are performed. Genes V and VI1 displayed unique EcoRI restriction patterns thathave not been observed previously for other P-450IIB genes. Genomic blot analyses with Sprague-Dawley DNA indicated the presence of EcoRI bands for both of these genes (data notshown). Restriction Analysis of Clone 17A"We chose to focus our initial efforts onthe furthercharacterization of clone 17A. By comparison of restrictionfragment size and hybridization patterns with the P-450e gene (Fig. l),it appeared that gene IV was P-450e-like in terms of its overall size and was contained most likely within 15 kb of DNA. However, gene IV contained a B-oligomer binding site in its hypervariable region, indicating a similarityto theP-450b gene in thisrespect. Clone 17A, which contained the most 5' information, as well as significant hybridization to theexon 9 probe, was analyzed by a seriesof single and double restriction digestions followed by hybridization to fragments comprising 5'- and 3'-portions

56

4

D

I

I I

4"

& I P H

P

H

5 6

4

I l l P P B

PB

18

I I E H

BBP

I+

H

9

"

I

I

H

E

E

II HE

E

4hb

4-

FIG. 3. A, partial restriction map of clone 17A. Genomic clone 17A was mapped by single and double restriction digestions with BamHI ( B ) ,EcoRI ( E ) ,HindIII (H), and PstI (P).Regions homologous to various portions of the P-450b ( 5 ) and P-450e (3) genes which were determined by hybridization analysis (see Fig. 2) are indicated by common shading patterns B, sequencing strategy of p17A.4. The 4kb PstI ( P ) fragment of 17A was subcloned into pUC18 (p17A.4) and sequenced as indicated by the arrows. Hc = HincII; N = NcoI.

of the P-450b cDNA, as illustrated in Fig. 2. Based on this information, a restriction map of gene IV was generated and is diagrammed in Fig.3. Although areas of hybridization relatedness are prominent (Fig. 2, panels A, B, and C), there is virtually no similarity in restriction pattern between P450b, P-450e, and the portion of gene IV contained in clone 17A (Fig. 3, panel A). From the data observed in panel C of Fig. 2, it was evident that exon 1 homology existed between the P-450b gene and the4-kb PstI fragment of clone 17A. To

Rat P-45OIIB Gene Family

7049

both directions from the target site. The sequence obtained from one of these oligomers confirmed the presence of a highly conserved sequence in gene IV sharing 23/25 bases with the EX1 probe, hence the observed binding properties of this probe to gene IV (see Fig. 4). Finally, unidirectional deletion clones were generated from a unique HincIIsite in clone 17A.4 to obtain the remaining 5’ sequence. These data are presented in Fig. 4 along with the corresponding P-45Oe gene sequence for comparison. Each fragment was sequenced at least three times. An open reading frame of 58 amino acids (one more than for exon 1 of the P-45Ob or -e genes) was apparent in this portion of gene IV, beginning with a methionine residue and ending with a putative splice donor site identical with that for exon I of the P-450e gene. Thirty-seven out of fifty-eight of these amino acids are conserved in this region with P-450b and P-450e. Structural features of this region include: 1) a highly hydrophobic amino-terminal sequence followedby highly charged amino acid residues reminiscent of a membrane insertion and halt-transfer signal motif, and 2) a proline-rich region; both of which are conserved in the NH2terminal sequences of membrane bound P-450s. In the flanking region, a modified TATA sequence at -25 was observed in gene IV which differed from that of the P450e gene by a single base substitution. A CAAT box homolIVctgcagagacttga-tgttccaqggtaggcagatagatacccaqtqqqaqcc~ ** **** ** ** ** ** ******* * * ogy was present in gene IV at position -220 and is notably Ect----attcttgtcaactcaaacataatcacatg--tacccaq--gacacaaaaaacat absent in the P-450e gene. Four blocks of varying sequence -ctqaqagg----agaaqgagaggqaaggatgggtagagqga--ctatgcagtgqttgta relatedness were apparent comparing gene IV sequence with **** t *** * *** ** ** acagaqaagccccaataatttaagattatacatgtaaatataccctaqaca-tqcaaqaa the 5“flanking sequence of the P-450e gene. The first was a region of relatively high conservation exhibiting 73% homolggqgggatcttgag-gaggaqgcq-attaattgggatataaaatqattaaataaataaat * * **. I * * * * * * * * ogy (57/78 residues) and included nucleotides +1 to -77 (see aagaccacccagtgcatctaqactcagacaaaqaaatttacatcqgtacqtttatatcaq Fig. 4). From position -78 to -129, a sequence element was -302 caatgaaaaat-gtaatgttccttcatata-tgcacacaca-qcaqctccatgcc----t ****** *** **** ** ******* present in gene IV that had no apparent counterpart in the ***** * aaatgatctttcacataggaaaagcatatagaacacqcacac~~tcccatqccctagt P-450e gene. This sequence has been excised arbitrarily in aagtaaac-gaqcaqatgaaactaaactgacaaqtgc-cagctaatcqcaaatgtqttaq Fig. 4 to maximize the alignment of gene IV with the P-450e **.***** **** ** ***** ******.*** ** ** gene. It is interesting to notepotentialhairpinstructure aagtaaacagagctgacaaaactqaqttqacaagtgcacacccatttacataaaacaaqa within the excised sequence due to internalcomplementarity. t~~g--cc--ttctctcctgtqc--tatctc-atttcatqatgtctttqccaacat ** * * ** ** *** ** **** ** **** ********** The base of the hairpin loop would beflanked by the sequence ggcctaagtcccagtgcccttttgtcctgtqtatctgtttcqtg~gtccttqccaacat 5’-d(GAG(N),ACAGG)-3’immediately 5 ’ , and in theflipped -1 29 orientation on the 3’ side. A repeat sequence of 5’-7s ~qqa~cccaaacaggaaqtatgqgtg~tcacatacctqgacaqtgatqaqa~ .._ d(GAACAG)-3’ is present flanking either side of the extractaaggttgggataaaqgaatqqggagt~~acagqtcac~qggagqtg~~acagctqq neous sequence at positions -67 and -141. A third region in *** .** ** *** *..*** ******** ** ***** * ***** *** ctatggtgtgqgtaagqgaatgagqagtqaatagccaaaqcaqgaqgcgtgaacatctga gene IV extended from -130 to -310 and was 50% homolo-25 +1 gous (90/180) with corresponding sequence of the P-450e agttgcttaagtgagtaggg tcatsatasaaqatcctgctggaqtg--ttcC **I(*** *** ****. *!EF*.* *****************.. gene. Within this region, a short alternating purine-pyrimiagtt9cataactqaqtqtaggqgcagattcag~agatcctgctqgaqaqcatqcA dine stretchof (CA), is observed at position -302 in gene IV. CTGATGTCMCAGTGGTTAGACCAGGMCATGG~CTCGGTGTCTTCCTCCTC~CACTT A much longer (CA),, stretch is seen in this region for the P**** *** ** * * * * * . a ******* ****** ** **** ******** ** CPG~GTCTACCGTGGTTACACCAGGACCATGGAGCCCAGTATePTGCTCCTCCTTGCTC 450e gene. The fourth segment comprised residues -311 to -503 which were only 36% (70/192) similar to the analogous TCACTGTGGGCTT~GCTATTATTAGCCAGCCAAAATCGGCC~GACCCATGGCCACT ** ******.****** .* **** *** *e.*** **. **** *.* P-450e structure.Preliminary sequence data from regions TC~PTGTGGGCTTCTTGTTACPCTTAGTCAG---GGGACACCCAAAGTCCCGTG@MCT further 5’ of this most divergent block have not indicated TACCACCAGGAC~CGTCCCCTGCCCTTC~GGGGMCCTCTTACAGATGMCAGMGAG continuation of b/e gene homology. However, a second alter.................... *** ************** .** ** ****. *** TCCCACCAGGACCTCGTCCCCCCCT~GGGGMC nating purine/pyrimidine stretch of (CA)SO exists at position GCCTTCTGCGTTCTTTCATGCAGgtgagacattqcacaggacttcctgg -650 of gene IV (not shown). This second (CA), repeat is not **** ** *** **************. *** * * * * * a **** found in the 5’-flanking sequence of the P-450e gene seGCCTCCTCAATTCCTTCATGCAGqtqagatatt-cacaqgqc---ctgq quenced to date (22). FIG. 4. Sequence analysis of the exon I and 5’-flanking TWO putative SpI binding sites (23) are present in gene IV region of gene IV. The sequence of gene IV (upper line, ZV) is compared to the analogous region of the P-450e gene (lower line, E ) . and are absentin P-450e (one site is found in the P-450b gene but appears in the inverted orientation). A sequence sharing The transcription initiation site of the P-450e gene is designated +1. Putative promoter elements (TATA and CAAT boxes) are underlined. 11/15 bases with a consensus glucocorticoid responsive eler Core Spl binding sequences are boxed. A sequence related to a ment (24) was also found in gene IV but appeared to be glucocorticoid responsive element is ouerlined. The oligomer binding nonfunctional since gene IV was not appreciably induced by sites used for sequencing are indicated by parentheses (EX1 sense) or brackets (gene IV sense). Asterisks denote conserved nucleotides. dexamethasone in the liver or the adrenah2 A search for Arrows indicate the site of potential hairpin formation in gene IV. other known regulatory sequences including the xenobiotic

study this relatedness, the 4-kb PstI fragment of 17Awas subcloned into pUC18 (subclone p17A.4) and partially sequenced using a variety of primers. Sequence Analysis of Subclone pl7A.4-A schematic illustrating the strategy undertaken to sequence the 5‘ portion of gene IV contained in subclone p17A.4 is presented in Fig. 3B. To sequence the putative 5’-exonic and flanking regions of gene IV, an oligomer probe, EXI, was targeted to a conserved region in exon 1 of the P-450b and P-450e genes shown parenthetically in Fig. 4. We hypothesized that this oligomer would cross-react with gene IV based on the homology observed from hybridization experiments between gene IV and the PB7(150) probe. Supporting our hypothesis, the EX1 probe did cross-hybridize under moderate stringency to subclone p17A.4 (data not shown). The EX1oligomer was then used as a primerfor dideoxy sequencing of clone 17A.4. Over 100 bp of sequence was obtained in this manner. From this information, two complementary oligomer probes (gene IV sense and antisense) were targeted to a specific region of subclone p17A.4 (shown bracketed in Fig.4) and used as primers for further sequence analysis. The oligomers were made complementary to each other to allow sequencing in

*.

Gaps (denoted by dashes) and excised DNA have been introduced to maximize the alignment. Exonic and intronic sequences are denoted by upper and lower case lettering, respectively.

* C . M. Giachelli, J. Lin-Jones, and C . J. Omiecinski, unpublished observations.

Rat P-45OIIB Gene Family

7050

responsive element described by Fujii-Kuriyama and co-workers (25) proved negative. The absence of the latter is consistent with the inability of 3-methylcholanthrene to induce gene IV mRNA expression in the tissues examined (see below). Atchison and Adesnick (14) have reported the isolation of several members of the P-450IIB gene family, two of which (genes 2 and 3) share almost identical restriction maps to clone 17A. These investigators have sequenced genes 2 and 3 in a region containing exons 7 and 8 and portions of the surrounding introns. Genes 2 and 3 differed from the P-450e gene by 19 and 16 bases, respectively, and from each other in 10 bases within these exonic regions. To determine whether clone 17A represented either of these genes, we sequenced a portion of the exon 7 region of gene IV using our B oligomer as primer (data not presented). This sequence information was compared to theP-450e gene and Adesnick’s genes 2 and 3. Together with the restriction map similarity, the sequence data identified 17A as identical with Adesnick’s gene 2. Determination of the Transcription Initiation Site forGene IV-The transcription initiation site for gene IV was mapped using the primer extension technique, and the results are shown in Fig. 5. The panelon the left shows an autoradiogram of the fragments generated using the gene IV antisense oligomer as a primer with uninduced rat liver total RNA. Because of the low level of expression of gene IV in either uninduced or induced rats (Fig. 6),a modification of the typical primer extension reaction was employed to include incorporation of high specific activity radiolabeled nucleotides into the extension product (see “Materials and Methods”). Using this technique, two major bands were detected; one a t 85 bp and another at 75 bp, along with one or more minor bands surrounding these. It isnotable that at the site denoted +1 in Fig. 3, which is analogous to the cap site of the P-450e gene, we find the sequence CCC which represents a single base transversion from the CAC sequence found in the P-

Iv

M1

PrA C G T ext.

A C G T ext.

Pr.

J

FIG.5. Primer extension analysis of gene IV mRNA transcripts. Either the gene IV antisense (ZV)or EX1 oligomer (ISXI) was used as primers for extensionreactions as described under “Materials and Methods.” Lengths of fragments, in bases, were determined from concurrently runstandards (A, C, G, 7’) andare indicated adjacent to the primer extension (pr. ext.) lanes. Asterisks mark the fragmentsexpected if transcriptioninitiationsitesare identical with that reported for the P-450e gene (5).

A CP

B C P

661

4.4,

23* 20,

FIG.6. Northern blot analysis of gene IV mRNA expression. 15 pg of poly(A’) RNA from control (C) or PB-treated (P)rats was Northern blotted and probed with either 3*P-gene IV antisense oligomer ( A ) or ‘*P-EXI oligomer (B).Exposure times were as follows: A, 72 h; B, 3 h. Radiolabeled DNA standards were electrophoresed concurrently and areshown in the lefthandportion of the figure with molecular sizes indicated in kilobases. 450egene. Although most mammalian genes (about 50%) initiate at an A residue, about 25% initiate with a C residue (26). The second fragment may represent an alternative start site for gene IV, priming of a closely related gene, or spurious priming by the oligomer. In contrast,due to therelative abundance of their mRNAs, the P-450b/e transcripts could be readily generated with the EX1 primer using PB-treated rat liver total RNA (Fig.5, EXZ) and the conventional oligomer end-labeling technique (27). A major fragment of 138 bp was synthesized which is the size expected for both P-450b and P-450e products based on the published start site of transcription for the P-450e gene (3). Northern Blot Analysis of GeneI V mRNA Expression in the Rat-Since none of the sequence analyses thus far precluded the possibility that gene IV could encode a P-450 related to P-450b/e, it was of interest to characterize the expression of gene IV in the rat. Consequently, we probed poly(A+) RNA isolated from the liver of control or PB-treated ratswith the gene IV antisense oligomer, which interacts specifically with the putative gene IV mRNA. As shown in Fig. 6A, an mRNA migrating at about 4.3 kb was detected after prolonged autoradiography of the Northern blot in both control and PBtreated rat liver poly(A+) RNA. As predicted, the gene IV antisense oligomer did not cross-hybridize with P-450b/e mRNAs, despite their presence at very high concentrations in the induced samples (as evidenced by binding of the EX1 probe to these same preparations; Fig.6, panel B). In the liver, levels of gene IV transcript did not appear to be inducible by 3-methylcholanthrene administration, and Aroclor1254 treatment was no more effective than PBin eliciting increases in steady state levels of the 4.3-kb mRNA (data not shown). Detection of Gene I V mRNA Transcripts via the Polymerase Chain Reaction-We made use of the polymerase chain reaction to definitively show that gene IV was transcribed in the

Rat P-45OIIB Gene Fami1.y

7051 b.

uninduced and PB-treated rat liver. In this experiment, two A. Con A+ PB A+ Con poly PB poly oligonucleotide primers were used which flank the DNA of interest, and repeated cycles of heat denaturationof the DNA, - + - + - + - + Q annealing of the primers to their complementary sequences, Tuq DNA polymand extensionof the annealed primers with erase were performed.Using the EX1 and gene IV sense oligomers as primers, we expected to amplify a n 86-bp fragment from theexon I portion of gene IV (see Fig. 4). To test this assumption, we first used the oligomers to amplify the 86-bp fragmentfrom the p17A.4 plasmid and rat liver genomic DNA. The results are shown in Fig. 7A. As predicted, a single 86-bp fragment was generated in both cases. Sequencing of these fragments, shown inFig. 7B, indicated that thechosen primers amplify the correct gene IV product from genomic DNA sincethey clearly contain a diagnostic sequence 5’b. d(AGCCAGCCAAAA)-3’ found only in gene IV and not in Con A+ PB A+ Con poly PB poly P-450b, -e, or other known P-450s. Ghost banding in the * A * * - + - + - + - + a sequencing lanes is most likely attributed to a low level of nonspecifically primed polymerase chain reaction products which are notreadily detectable by ethidium bromide staining (Fig. 7A), and/or nonspecific priming during the sequencing reaction. Single-stranded cDNA was then prepared from rat liver RNA using the EX1 oligomer as a primer for reverse transcriptaseand gene IV specific cDNAs were amplified by subjecting the cDNA to the polymerase chain reaction. Fig. 8A shows the ethidium bromide-stained agarose gelof the FIG. 8. Analysis of gene IV mRNA expression in rat liver polymerase chain reaction products from hepatic cDNA pre- by the polymerase chain reaction. A, hepatic poly(A’) (A’) or pared from eitheruntreatedorPB-treated poly(A+) and polysomal (poly) RNA from untreated (con) or PB-treated ( P B ) rats polysomal-derived RNA. Aband co-migrating with the 86-bp was subjected to the polymerase chain reaction either prior to procgene IV fragment derived from p17A.4 is detected in all cases. essing (-) or after synthesis of single-stranded cDNA (+). ApproxiP B responsiveness of gene IV is further substantiatedby the mately 200 ng of the polymerase chain reaction product derived from p17A.4 was run asa size marker. B, the gel was transferred toa nylon polymerase chain reaction results since more 86-bp product filter and probed with 3ZP-PB7(150). is evident in cDNA derived from PB-treated rat liver RNAs

5

* * * * -

2

B.

A

B gen. DNA

pl7A.4 A

A A A A C C G A C C G A

C

G

T

A

C

G

T

A A A A C C G A C C G A

FIG. 7. Characterization of gene IV polymerase chain reaction products. A, ethidium bromide-stained agarosegel of polymerase chain reaction products amplified from p17A.4 and rat liver genomic DNA using theEX1 and gene IV sense oligomers as primers. Sizes indicated in base pairs were determined from standard derived from HaeIII-digested 0x174 DNA. B, sequence analysis of gene IV polymerase chain reaction products. Polymerase chain reaction products were sequenced using the gene IV sense primer. Shown is the region of the sequencing gel which contains a diagnostic sequence for gene IV, 5’-d(AGCCAGCCAAAA)-3’.

than from untreated samples. A possibility that contaminating genomic DNA in the original RNA preparation might be responsible for fragment generation was ruled out since polymerase chain reaction on the unprocessed, i.e. non-reversetranscribed, RNA sample(-) failed to amplify any fragments. Comparable results were obtained whenpolymerase chain reaction was performedon cDNA derivedfrom two additional sources of control ratliver total RNA (data not shown). The gel in Fig. 8A was transferred to a nylon membrane and hybridized to the PB7(150)probe. The autoradiogram is shown in Fig. 8B. The major hybridizing species in all processed samples is the 86-bp fragment, as expected. A greater signal intensity was observed for PB-treated samples compared to control samples. No signal was evident in unprocessed RNA samples.At leasta 6-fold induction by P B of gene IV mRNA is estimated by densitometric analysis of the autoradiogram (data not shown). Polymerase chain reaction also was used to assess expression of gene IV in various other tissues of the rat. The gene IV mRNA was detected in lung, kidney, and testis. Thegene is also active in the developing fetal liver as the appropriate polymerase chain reaction products were observed in RNA fractions derived from untreated rats at gestational ages 15, 19, and 22 (data not shown). DISCUSSION

Our ultimate aim is to define structures within specific members of the P-450IIBsubfamily genesthat might correlate with their transcriptional activity and responsiveness to PB. In this study,we isolated portions of seven genes comprising the P-450IIB subfamily, including one designated gene IV, and two other genes,V and VII, not isolated previously.

7052

Rat P-45OIIB Gene Family

Sequencing of gene IV indicated that it contained a region with 70% nucleotide homology to exon I of the P-450e gene. The 5’-flanking sequence of gene IV also possessed homology to P-450e and included a TATA sequence at a position identical with that of the P-450e gene. Results from primer extension experiments indicated that transcription initiation for gene Iv occurs at ananalogous position as that for P-450e. The fact that gene IV is also PB-responsive suggests that features conserved between it and otherPB-responsive genes might contribute to this unique regulatory feature. The conserved promoter region along with the CATATA . . . (CA), sequence in gene IV and P-450e are good candidates for important regulatory sequences in this regard. The attenuated response of gene IV to PB (-6-fold) compared to the >loofold induction of the b and e genes by this compound might be attributed to the extra sequence found in gene IV. This sequence bears a resemblance to the mobile elements found in both procaryotes and eucaryotes (27, 28): both typically contain an inverted repeat structure and a short duplicated sequence at the chromosomal insertion site. Such sequences might disrupt theputative regulatory region spatially or might introduce negative regulatory elements (e.g. repressors) leading to decreased gene expression, ashas been shown for bacterial insertion sequences (27). This hypothesis is being tested by deletion analysis of gene IV 5’ regulatory regions in transient transfection assays and by introduction of the insertion directly into the P-450b and -e gene 5“flanking constructs. Alternating purine/pyrimidine sequences (“TG elements”), which have the potential for forming 2-DNA, have been implicated as hot spots for gene conversion in several gene families (29,30). The proximity of the second divergent region in gene IV to two (CA), stretches makes attractive the hypothesis that this region has come about through gene conversion with another related gene. Sequence analysis of other members of the P-450IIB family may help resolve this question. TG elements have also been shown to modulate gene expression in an enhancer-likefashion when attachedto reporter genes in transient transfection assays (31). Clearly, further analysis of the P-450IIB family members is necessary to determine the exact role of TG elements in P-450 gene regulation and/or evolution. Aside from the relatively well studied P-45Ob and P-450e genes, very little data regarding the functionality of other members of the P-450IIB subfamily exists. Results of Friedberg et al. (32) suggest that gene 4 (analogous to our gene 111) is selectively expressed in the rat preputial gland, but not in the liver or any other tissue examined. A second gene, termed P-450IIB3 (not represented in our genomic clones), appears to be expressed in the liver since a cDNA clone whose DNA sequence perfectly matches the limited sequence available for exons 7 and 8 of this gene (14) has been recently isolated from a liver cDNA library (33). Notably, the mRNA detected was constitutively expressed and not appreciably induced by PB. Sequence analysis of gene IV by Atchison and Adesnick (termed gene 2 in that report) indicated high sequence relatedness between gene IV and P-450e in the regions encoding exons 7 and 8 and intron 7 (14). This high degree of conservation suggests that selective pressure tomaintainthese sequences in gene IV is high and may therefore have functional significance. The concept that other P-450IIB gene members are potentially expressed is further supported by evidence for the expression of as many as five different P450IIB gene members in the rabbit (34), and a similar heterogeneic expression of P-4501IB family members in the mouse (41).

Initialindications that gene IV might be actively transcribed come from Atchison and Adesnick (10) in their paper characterizing certain members of the P-450IIB gene family. In thatwork, 32P-labelednuclear RNA from control and PBtreated animals showed hybridization under stringent conditions toclones for gene IV (termed gene 2 in that report) and other P-450IIB family members. These researchers hypothesized that other genes besides P-450b and -e might be expressed and that some may even be induced to a limited extent by PB. The direct test of this hypothesis requires methodology enabling highly sensitive and discriminatory detection of RNA transcripts. In this report, we have presented data using an exquisitely sensitive technique, the polymerase chain reaction, to identify gene IV transcripts in rat liver RNA. Our results conclusively indicate accumulation of gene IV mRNA in uninduced rat liver and anelevation of its steady state level in the PB-treated ratliver. We suggest that this novel P-450 mRNA and corresponding gene be termed P-450IIB4 based on the new nomenclature (2). Although the presence of an mRNA product does not rule out the possiblity that gene IV is a pseudogene, several lines of evidence suggest that this is not the case. First, gene IV mRNA is associated with polysomal RNA, strongly suggesting its active translation in the liver. Second, the portions of the gene sequenced to datecontain correct intron/exon junctions and no premature termination codons. However, final proof of the expression of gene IV as a functional isozymewill require the isolation and identification of the putative protein product. The low levels of gene IV mRNA in whole tissue extracts raise the question whether this P-450 gene might be selectively expressed at a high level in a subpopulation of hepatocytes or in a different cell type altogether in the liver, e.g. Kupfer or endothelial cells. Heterogeneity in hepatocytes with regard to P-450 expression has been described for the liver by other researchers (35,36), andat least one reportsuggests P450 expression in nonhepatocyte parenchymal cells based on enzymatic activity (37). In situ hybridization studies are currently underway in our laboratory to address this possibility. Since the number of tissues or inducing agents we studied were not exhaustive, it remains to be seen whether gene IV might have the capacity for higher level expression under altered circumstances. As our understanding of mammalian gene regulation increases, we may be able to predict conditions which promote higher level expression of gene IV. The polymerase chain reaction allowed us to detect the very low abundance gene IV mRNA in the rat liver. This assay has been used to successfully detect an mRNA present at 0.01% in 100 ng of cDNA (21). By employing this sensitive methodology, wenow should be able to definitively assess whether particular P-450 genes are being expressed under certain inductive conditions, in various tissues, or at different stages of development. These questions are of extreme importance since even the presence of low levels of P-450-mediated biotransformation may determine organ-specific or developmental toxicities resulting from exposure to environmental chemicals. Furthermore, diagnostic screens for individuals exhibiting unusualdrug-metabolizing activities could become routine by the use of this technique for detecting susceptible polymorphisms in P-450 genes. In this regard, polymerase chain reaction has been used successfully to detect base substitutions in &thalassemia(38), an HLA DQ2 allele associated with pemphigus vulgaris (39), and for direct detection of HIV1 in DNA of peripheral blood mononuclear cells (40). Acknowledgments-We wish to thank Dr. Alan Anderson for his gift of PB(7) and Dr.James Bonner for the Charon 4A rat liver DNA

Rat P-45011BGene Family library. The excellent technical assistance of Richard Ramsden is greatly appreciated. REFERENCES 1. Black, S. D., and Coon, M. J . (1987) Adv. Enzymol. Relat. Areas

Mol. Biol. 60, 35-87 2. Nebert, D. W., and Gonzalez, F. J. (1987) Annu. Reu. Biochem. 56,945-993 3. Mizukami, Y., Sogawa, K., Suwa, Y., Muramatsu, M., and FujiiKuriyama, Y. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 39583962 4. Yuan, P. M., Ryan, D. E., Levin, W., and Shively, J . E. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 1169-1173 5. Suwa, Y., Mizukami, Y.,Sogawa,K., and Fujii-Kuriyama, Y. (1985) J. Biol. Chem. 260,7980-7984 6. Omiecinski, C. J., Walz, F. G., Jr., and Vlasuk, G. P. (1985) J. Biol. Chem. 260,3247-3250 7. Omiecinski, C. J . (1986) Nucleic Acids Res. 1 4 , 1525-1539 8. Giachelli, C. M., and Omiecinski, C. J. (1986)J. Biol. Chem. 261, 1359-1363 9. Giachelli, C.M., and Omiecinski, C. J. (1987) Mol. Pharmacol. 31,477-484 10. Atchison, M., and Adesnick, M. (1983) J . Biol. Chem. 258, 11285-11295 11. Mizukami, Y., Fujii-Kuriyama, Y., and Muramatsu, M. (1983) Biochemistry 22, 1223-1229 12. Rampersaud, A., and Walz, F. G., Jr. (1987) J . Biol. Chem. 262, 5646-5653 13. Maeda, N., and Smithies, 0. (1986) Annu. Reu. Genet. 20, 81108 14. Atchison, M., and Adesnick, M. (1986) Proc. Natl. Acad.Sci. U. S. A. 8 3 , 2300-2304 15. Benton, W. D., and Davis, R. W. (1977) Science 196,180-181 16. Gubler, U., and Hoffman, B. (1983) Gene (Amst.) 25,263-269 17. Affolter, M., Labbi., D., Jean, A., Raymond, M., Noel, D., Labelle, Y., Parent-Vaugeois, C., Lambert, M., Bojanowski, R., and Anderson, A. (1986) DNA ( N Y )5 , 209-218 18. Maniatis, T., Fritsch, E. F., and Sambrook, J . (1982) Molecular C1oning:A LaboratoryManual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 19. Henikoff, S. (1984) Gene (Amst.) 28, 351-359 20. Chen, E. Y., and Seeburg, P. H. (1985) DNA ( N Y )4, 165-170 21. Saiki, R. K., Gelfand, D. H., Stoffel, S., Scharf, S. J., Higuchi,

7053

R., Horn, G. T., Mullis, K.B., and Erlich, H. A. (1988) Science 239,487-491 22. Jaiswal, A. K., Rivkin, E., and Adesnik, M. (1987) Nucleic Acids Res. 15,6755 23. Dynan, W. S., and Tjian, R. (1985) Nature 316,774-778 24. Jantzen, H.-M., Strahle, U., Gloss, B., Stewart, F., Schmid, W., Boshart, M., Miksicek, R., and Schiiltz, G. (1987) Cell 49, 2938 25. Sogawa, K., Fujisawa-Sehara, A., Yamane, M., and Fujii-Kuriyama, Y. (1986) Proc. Natl. Acod. Sci. U. S. A. 83,8044-8048 26. Darnell, J., Lodish, H., and Baltimore, D. (1986) Molecular Cell Biology, pp. 436-462, Scientific American Books, Inc., New York 27. Calzone, F. J., Britten, R. J., and Davidson, E. H. (1987) Methods ERZYWZ.152,611-632 28. Kleckner, N. (1981) Annu. Reu. Genet. 15,341-404 29. O’Hare, K., and Rubin, G. M. (1983) Cell 34, 25-35 30. Proudfoot, N. J., and Maniatis, T. (1980) Cell 21,537-544 31. Hamada, H., Seidman, M., Howard, B.H., and Gorman, C. M. (1984) Mol. Cell. Biol. 4, 2622-2630 32. Friedberg, T.,Waxman, D. J., Atchison, M., Kumar, A., Haaparanta, T., Raphael, C., and Adesnick, M. (1986) Biochemistry 25,7975-7983 33. Labbe, D., Jean, A., and Anderson, A. (1988) DNA ( N Y ) 7,253260 34. Komori, M., Imai, Y., Tsunasawa, S., and Sato, R. (1988) Biochemistry 27, 73-80 35. Baron, J., Redick, J. A., and Guengerich, F. P. (1981) J. Biol. Chem. 256,5931-5938 36. Schwarz, M., Peres, G., Buchmann, A., Friedberg, T., Waxman, D. J., and Kunz, W. (1987) Carcinogenesis 8,1355-1357 37. Steinberg, P., Lafranconi, W.M., Wolf, C. R., Waxman, D. J., Oesch, F., and Friedberg, T. (1987) Mol. Phnrmacol. 32, 463470 38. Diaz-Chico, J. C., Yang, K., Yang, K., Efremov, D. G., Stoming, T. A., and Huisman, H. J. (1988) Biochim. Biophys. Acta 949, 43-48 39. Sinha, A. A., Brautbar, C., Szafer, F., Friedmann, A., Tzfoni, E., Todd, J. A., Steinman, L., and McDevitt, H. 0. (1988) Science 239,1026-1029 40. Ou, C.-Y., Kwok, S., Mitchell, S. W., Mack, D. H., Sninsky, J. J., Krebs, J. W., Feorino, P., Warfield, D., and Schochetman, G . (1988) Science 239,295-297 41. Masaki, R., Lasko, M., Noshiro, M., and Negishi, M. (1988) FASEB J. 2, A1340 (Abstr. 6068)