Molecularcloning and expression the mouse ornithine ... · developmental, andcell-growth-related influences (2). Recently, study of the mode of regulation of enzyme expression hasbeenfacilitated

Proc. Natl. Acad. Sci. USAVol. 81, pp. 540-544, January 1984Genetics

Molecular cloning and expression of the mouse ornithinedecarboxylase gene

(gene amplification/hormonal induction/cDNA)

LISA MCCONLOGUE*, MADHU GUPTA*, LILY WU*, AND PHILIP COFFINO*ttDepartment of *Microbiology and Immunology and Department of tMedicine, University of California, San Francisco, San Francisco, CA 94143

Communicated by Robert T. Schimke, September 30, 1983

ABSTRACT We used mRNA from a mutant S49 mouselymphoma cell line that produces ornithine decarboxylase(OrnDCase) as its major protein product to synthesize andclone cDNA. Plasmids containing OrnDCase cDNA were iden-tified by hybrid selection of OrnDCase mRNA and in vitrotranslation. The two of these with the largest inserts togetherspan 2.05 kilobases of cDNA. Southern blot analysis of DNAfrom wild-type or mutant S49 cells, cleaved with EcoRI orwith BamHI, revealed multiple bands homologous to OrnD-Case cDNA, only one of which was amplified in the mutantcells. RNA transfer blot analysis showed that the major OrnD-Case mRNA in the mouse lymphoma cells is 2.0 kilobases long.A similar size mRNA was found in mouse kidney and wasmore abundant in the kidneys of mice treated with testoster-one, an inducer of OrnDCase activity in that tissue.

Ornithine decarboxylase (OrnDCase; EC 4.1.1.17) is the ini-tial enzyme in the synthesis of polyamines by animal cells(1). Its activity is subject to a dazzling array of hormonal,developmental, and cell-growth-related influences (2).

Recently, study of the mode of regulation of enzymeexpression has been facilitated by purification of OrnDCaseto homogeneity (3-6), development of a method to specifi-cally label the protein by use of a radioactive enzyme-acti-vated suicide substrate (4, 7), and establishment of a quanti-tative serological assay (6, 8, 9). Application of these meth-ods has shown that in the best-studied system, androgen-induced mouse kidney, OrnDCase activity is determined bythe amount of enzyme protein (6, 9). The large change inlevel of protein was shown to result only in part from changein enzyme stability. This suggests that regulation of mRNAlevel or translatability also occurs.A DNA probe is needed to determine the molecular basis

of regulation of the mRNA that encodes OrnDCase. Howev-er, OrnDCase is a low-abundance protein; it constitutes0.01-0.05% of cytosolic protein in androgen-induced mousekidney (6, 9), the mammalian tissue whose activity is thehighest of any described. Cloning a gene that encodes a pro-tein of such low abundance remains a nontrivial task. There-fore, we have generated and utilized a mutant cell line thatvastly overproduces the enzyme (10). We used mRNA fromthat mutant to synthesize cDNA and to isolate plasmids thatcontain sequences homologous to the OrnDCase gene.

METHODSRNA. Total RNA was purified from wild-type or D4.1 mu-

tant S49 cells (10) according to ref. 11, using cesium chloridecentrifugation. Male BALB/c mice, 6 weeks of age, werepurchased from the University of California, Berkeley,breeding colony. To induce kidney OrnDCase, animals weretreated by subcutaneous injection with testosterone pro-prionate in corn oil, 4 mg/ml, at a dose of 100 mg/kg, and the

mice were killed 3 days later by cervical dislocation. To pre-pare RNA from mouse kidneys, the kidneys were excised,frozen immediately in liquid nitrogen, pulverized in a Waringblender, and stored until use at -80'C. RNA was then puri-fied as described above. Poly(A)+ RNA was isolated fromtotal RNA according to ref. 12.cDNA Cloning. Double-strand cDNA was copied from

poly(A)+ D4.1 RNA as described in ref. 13 with the follow-ing modifications: first-strand synthesis was carried out in abuffer containing 0.2 M ammonium acetate in place of KCI.After first-strand synthesis, the RNA template was hydro-lyzed at 680C for 15 min with 0.1 M KOH in place of NaOH.The solution was neutralized by addition of HCl and the sec-ond-strand synthesis reaction was carried out in the sametube to avoid loss of product otherwise associated withtransfer or ethanol precipitation of the single-strand cDNA.S1 nuclease treatment was done with 2 units of enzyme perng of double-strand cDNA at room temperature for 20 min.The product was fractionated on a P-200 column and the firsttwo-thirds of the peak eluting in the void volume was pooled.Homopolymeric tailing with dCTP was performed accordingto ref. 14. The C-tailed double-strand cDNA was annealed topBR322 G-tailed at the Pst I site and used to transform Esch-erichia coli strain RR1. The resulting tetracycline-resistanttransformants, -500 in number, were transferred to filters inreplicate and the filters were processed according to ref. 15.32P-labeled single-strand cDNA was reverse-transcribedfrom poly(A)+ RNA from either wild-type or D4.1 S49 cellsand used to screen bacterial colonies containing recombinantplasmids by differential hybridization (16).mRNA Hybrid Selection and In Vitro Translation. To iden-

tify recombinant plasmids with sequences homologous toOrnDCase mRNA, plasmid DNA was purified, made linearwith EcoRI, and fixed to nitrocellulose filters 7 mm in diame-ter as described (17). Hybrid selection was done in 150 41containing 250 ,g of poly(A)+ RNA from D4.1 cells per ml,50% formamide, 10 mM Pipes (pH 6.4), 0.5 M NaCl, 2 mMEDTA. The filters were washed as described in ref. 18 toremove nonspecifically bound RNA, except the washingbuffer salt concentration was 75 mM NaCl. In vitro transla-tion of hybrid-selected RNA or of unselected poly(A)+ RNAwas performed according to ref. 19. 35S-labeled in vitrotranslation products were immunoprecipitated with OrnD-Case antibody (8) or with normal rabbit serum according toref. 20. In vitro translation products or those products afterimmunoprecipitation were analyzed by NaDodSO4/10%polyacrylamide gel electrophoresis (20, 21).OrnDCase Assay. Frozen pulverized mouse kidney, pre-

pared as for RNA purification, was thawed on ice in a solu-tion containing 25 mM Tris (pH 7.5), 0.1 M EDTA, 5 mMdithiothreitol, 10% glycerol and made soluble by Polytronhomogenization. After centrifugation at 100,000 x g for 60

Abbreviations: kb, kilobase(s); OrnDCase, ornithine decarboxylase.tTo whom reprint requests should be addressed at: Department ofMicrobiology & Immunology, S 412, University of California, SanFrancisco, San Francisco, CA 94143.

540

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Proc. NatL. Acad Sci. USA 81 (1984) 541

min at 20C, the supernatant solution was stored at -80'C.OrnDCase activity was determined as described (22), using1.6 mM ornithine as the substrate concentration. One en-zyme unit represents 1 nmol of CO2 evolved per mg of pro-tein per 30 min.Other Procedures. High molecular weight cellular DNA

was isolated by a minor modification of the procedure de-scribed in ref. 23. The DNA was digested to completion withan excess quantity of restriction enzyme under the condi-tions recommended by the supplier. Southern blot analysis(24) was performed by using 0.9% agarose gels for fraction-ation. RNA transfer blot analysis was carried out in 1% agar-ose gels containing formaldehyde (25) and the fractionatedRNAs were transferred to nitrocellulose paper (24). RNAdot analysis was carried out according to ref. 26. Filtersto be hybridized for differential screening, Southern orRNA transfer blot analysis, or RNA dot analysis were pre-hybridized overnight at 420C in buffer containing 50% form-amide, 5x concentrated Denhardt's solution (27), 0.45 MNaCl/0.045 M sodium citrate, and 50 ,ug of salmon spermDNA per ml. Hybridization was carried out for 2 days underthe same conditions with the addition of 32P-labeled probe inthe form of single-strand cDNA or plasmid 32P-labeled bynick-translation. Filters were washed twice for 30 min in 15mM NaCl/1.5 mM sodium citrate/0.1% NaDodSO4 at 500C.Filters were subjected to autoradiography at -70'C with anintensifying screen.

Materials. Materials and their sources were: restriction en-zymes, Bethesda Research Laboratories or New EnglandBioLabs; terminal transferase and oligo(dT)12_18, P-L Bio-chemicals; reverse transcriptase, Life Sciences (St. Peters-burg, FL); S1 nuclease, Sigma; DNA polymerase I andKlenow fragment ofDNA polymerase I, Bethesda ResearchLaboratories; pBR322 G-tailed at the Pst I site and [35S]-methionine, New England Nuclear; P-200 gel, Bio-Rad; oli-go(dT)-cellulose type 2, Collaborative Research (Waltham,MA); and nitrocellulose filters, Schleicher and Scheull. Oth-er reagents were the purest available from commercialsources. The mouse muscle actin cDNA plasmid describedin ref. 28 was from William Marzluff.

RESULTSConstruction and Identification of Recombinant Clones

Containing OrnDCase-Specific Sequences. We have previous-ly described the S49 cell mutant D4.1, a cell line in whichOrnDCase is overproduced at least 300-fold compared to thewild-type parental cell and in which 15% of total protein syn-thesis is devoted to OrnDCase (10). Poly(A)+ RNA purifiedfrom D4.1 cells was used as a template for the synthesis ofdouble-strand cDNA, which was then inserted into the Pst Isite of pBR322 by the dG-dC homopolymeric tailing tech-nique. Recombinant plasmids were used to transform E. coliand tetracycline-resistant transformants were selected. Toidentify potential OrnDCase clones, the transformants werereplica plated to filters and hybridized to 32P-labeled cDNAcopied from poly(A)+ RNA of either wild-type or D4.1 S49cells. Of 450 colonies, 60 showed apparent positive differen-tial hybridization with the D4.1-derived probe. These 60were spotted in an ordered array onto fresh plates and againwere screened by differential hybridization. The 30 clonesthat were most strongly positive in the second screen werefurther analyzed.

Mini-lysates from these clones were prepared and theplasmids were digested with Pst I. Most of these containedrecombinant inserts with common-size internal fragmentsand were subsequently shown to cross-hybridize to theOrnDCase plasmid pOD48 described below (data notshown). Among recombinant plasmids containing the com-mon fragment, two with the largest cDNA inserts, now des-

ignated pOD48 and pOD32, were selected for further analy-sis by hybrid selection and in vitro translation. Plasmid DNAfrom each was purified, bound to nitrocellulose filters, andhybridized with poly(A)+ RNA from D4.1 cells. Filters with-out DNA, with DNA from pBR322, or with DNA from oneof our cDNA recombinant plasmids not related to OrnDCasewere used as negative controls. A mouse muscle actin cDNAplasmid that has been shown to hybrid select mouse non-muscle actin mRNA (28) was used as a positive control. TheRNAs binding to the filters were eluted, translated in a rabbitreticulocyte lysate system, and analyzed by gel electropho-resis (Fig. 1 Left). Clones pOD48 and pOD32 selectedmRNAs that, when translated in vitro, produced a Mr 53,000polypeptide of the same mobility as OrnDCase made in vivo(10, 20) or in vitro (10). Additional smaller polypeptides, pre-sumptive products of incomplete translation, were seen aswell and are also seen as a translation product of RNA notsubjected to hybrid selection (10). The identity of these prod-ucts was confirmed by immunoprecipitation with antibodydirected against OrnDCase (Fig. 1 Right). The actin plasmidgave rise to a polypeptide that had the molecular weight ofactin and did not immunoprecipitate with OrnDCase anti-body. We have shown that the OrnDCase antibody is specif-ic for OrnDCase; it does not significantly crossreact withother polypeptides synthesized in either wild-type or D4.1cells (10, 20). The identities of the OrnDCase and actin prod-ucts of hybrid selection and in vitro translation were alsoconfirmed by two-dimensional gel electrophoresis (data notshown). Therefore, clones pOD48 and pOD32 encode se-quences homologous to the OrnDCase mRNA.These plasmids contain cDNA inserts of 1.6 and 1.95 kilo-

bases (kb), respectively. Restriction analysis of the cDNA

12345678 1 23456

_-r w

FIG. 1. Identification of OrnDCase cDNA plasmids by hybridselection and in vitro translation. Linearized plasmid DNA immobi-lized on nitrocellulose filters, 10 u~g per filter, was used to selectpoly(A)+ RNA isolated from D4.1 cells. The bound mRNA was re-leased and translated in a rabbit reticulocyte lysate containing[35S]methionine. Samples were analyzed by gel electrophoresis(Left) or were immunoprecipitated with OrnDCase antibody or withcontrol antiserum before analysis (Right). (Left) Products of transla-tion encoded by no exogenous RNA (lane 1) or hybrid-selectedRNA bound by pMG20 (lane 2), pBR322 (lane 3), pOD48 (lane 4),pOD32 (lane 5), or mouse actin cDNA (lane 6) or a filter withoutDNA (lane 7). Lane 8 shows the translation product of 0.1 ,ug ofpoly(A)+ RNA not subjected to hybrid selection. (Right) The hybridselection translation product of pOD48 (lanes 1 and 2), pOD32 (lanes3 and 4), and mouse actin cDNA (lanes 5 and 6) immunoprecipitatedwith OrnDCase antibody (lanes 1, 3, and 5) or with normal rabbitserum (lanes 2, 4, and 6). Arrowheads mark the positions of OrnD-Case (upper) and actin (lower). Molecular weight standards migrat-ed in the gels at the positions indicated by bars and had Mrs of45,000, 66,200, and 92,500.

Genetics: McConlogue et aL

0"llijaw lso" ll -

542 Genetics: McConlogue et aL

.5 _

c C. . . a.

10, A_.

IL IL-

1 2345678 9

0bp L-

Soo 1000Il

1500 2000

FIG. 2. Restriction maps of OrnDCase cDNA in recombinantplasmids pOD48 and pOD32. Plasmid DNAs were digested by theindicated restriction enzymes or by pairs of enzymes and the size ofthe resulting fragments was determined on 1% agarose gels or on 5%polyacrylamide gels. The cDNAs are inserted at the unique Pst I siteof pBR322 and are so oriented that, as depicted, the unique EcoRIsite of the vector lies toward the right of pOD48 and toward the leftof pOD32. Additional restriction enzymes that did not cleave eithercDNA include BamHI, Bcl I, Bgl II, EcoRI, EcoRV, Kpn I, Pvu I,

Sac I, Sac II, Sma I, Sst I, Stu I, Xba I, and Xho I. bp, Base pairs.

inserts of pOD48 and pOD32 (Fig. 2) showed that they con-tain a 1.5-kb region of common sequence and together span2.05 kb. The two inserts have opposite polarity with respectto the vector. Plasmids from the clones positive in the sec-ond differential screen were tested for homology to pOD48.Plasmids were prepared by a mini-lysate procedure and di-gested with Pst I, and the products were analyzed by South-ern blot analysis with 32P-labeled pOD48 as a probe. Sixteenof 30 contained inserts with homology to the probe (data notshown).Southern Blot Analysis of Wild-type and D4.1 Genomic

DNAs. High molecular weight DNA was prepared from wild-type and D4.1 cells, digested with EcoRI or BamHI, frac-tionated and blotted by the Southern procedure, and hybrid-

1 2 3 4

FIG. 3. Southern blot analysis ofDNA from wild-type and D4.1cells. High molecular weight DNA from wild-type (lanes 1 and 3) orD4.1 (lanes 2 and 4) cells was digested to completion with EcoRI(lanes 1 and 2) or with BamHI (lanes 3 and 4), fractionated, andtransferred to a nitrocellulose filter by the Southern procedure, andDNA bands homologous to OrnDCase sequences were visualized byhybridization to 32P-labeled pOD48 and autoradiography. DNA mo-lecular weight standards migrated in the gels at the indicated posi-tions and were 4.3, 6.7, 9.4, and 23.7 kb in size.

..~ ~ ~ ~ .. ~ u

.".

FIG. 4. RNA transfer blot analysis of RNAs from S49 cells andmouse kidney. RNAs were fractionated by formaldehyde/agarosegel electrophoresis and transferred to a nitrocellulose filter, andRNA bands homologous to OrnDCase cDNA were visualized byhybridization to 32P-labeled pOD32 and autoradiography. TotalRNA was from wild-type S49-cells (lane 1), D4.1 S49 cells (lane 2),kidneys of individual untreated mice (lanes 3, 5, and 7), and kidneysof individual testosterone-treated mice (lanes 4, 6, and 8). In eachlane, 20 jig ofRNA was analyzed. A shorter autoradiographic expo-

sure of lane 2 is shown as lane 9. DNA molecular weight standardsmigrated in the gels at the indicated positions and were 0.52, 1.63,and 4.3 kb in size.

ized to a 32P-labeled probe obtained by nick-translation ofpOD32. As seen in Fig. 3, multiple hybridizing bands are

present with each enzyme. At least seven EcoRI fragmentsare seen. Among these, one, of about 6.9 kb, is present invery much higher copy number in D4.1 than in wild-typeDNA. Similarly, among the BamHI fragments, one, of about6.7 kb, is amplified in D4.1 DNA. The presence of multiplebands, one present in high copy number in D4.1 DNA, sug-gests that OrnDCase may belong to a multigene family andthat a single member of that family is amplified in the D4.1mutant.

Size and Abundance of OrnDCase mRNAs. We measuredthe size and abundance of OrnDCase RNAs in cells in whichenzyme levels were modulated by genetic alteration or hor-monal treatment. D4.1 mutant cells overproduce OrnDCasecompared to parental wild-type S49 cells. In the kidneys ofBALB/c mice, OrnDCase is inducible by testosterone andthis induction is due, in part, to a stabilization of the OrnD-Case polypeptide and, in part, to an increased rate of pro-duction of active enzyme (6, 29). RNA was isolated fromwild-type and D4.1 S49 cells and from the kidneys of threeuntreated BALB/c mice and three mice treated with testos-terone. Cytosolic lysates were also prepared from the mousekidneys for measurement of OrnDCase activity. Equalamounts of these RNAs were fractionated by formalde-hyde/agarose gel electrophoresis and were analyzed by theRNA transfer blot procedure, using 32P-labeled pOD48 as a

probe (Fig. 4). All RNAs displayed a band 2.0 kb in size, butthe relatively low amount of OrnDCase mRNA in wild-typeS49 cells precluded its visualization in the experiment shownhere. Additional bands of lesser intensity were seen as well.The relative amount of OrnDCase mRNA in the two kinds ofS49 cells was commensurate with the amount of OrnDCasepolypeptide synthesized by each. OrnDCase mRNA was in-duced substantially in mouse kidney by testosterone treat-ment.

pOD4Sa=_ w

C-

5n---.-Y

Proc. NatL Acad Sci. USA 81 (1984)

:gains

a

iiI _-

Proc. NatL Acad Sci USA 81 (1984) 543

1 2 3 4 5

4 * * - 10 &

* .

0,.v

':

& ~ 9

FIG. 5. RNA dot analysis of RNAs from kand testosterone-treated mice. Poly(A)+ RNA vcate in the amount indicated. Each row represetion of RNA from a single mouse. Rows 1, 3, anrows 2, 4, and 6, testosterone-treated mice.

The amount of OrnDCase mRNA in SD4.1 cells and in testosterone-treated and clneys was quantitated by RNA dot hybridiz,32P-labeled pOD48 as a probe. By comparirhybridization signal among RNA spots app]tion series, it was found that D4.1 S49 cells300-fold excess of OrnDCase mRNA compcells and that testosterone treatment inducemRNA by less than 10-fold in mouse kitshown). When OrnDCase mRNA in kidnemore accurately by performing the RNA dodilution series, hormonal induction was foulevel by a factor of 4-5 (Fig. 5). OrnDCasewas measured in the kidneys of the same anities in three control mice were 13.0, 13.(mean, 10.9 units) and in three treated micand 153 units (mean, 173 units). Therefore,tion was 16-fold.

DISCUSSIONWe have constructed recombinant plasmids containing se-quences present in the mRNA of mouse OrnDCase. Mutantcells that have amplified genes can provide an abundantsource of the homologous mRNA and thus facilitate cDNAcloning (30-32). Our task of cloning was simplified by usingmRNA extracted from the D4.1 variant of S49 cells, in whichOrnDCase is the single most highly synthesized protein (10).The two OmDCase cDNA inserts together span 2.05 kb andtherefore presumably contain most of the sequence of the 2.0kb mRNA. The identity of the cloned sequence in these plas-mids was determined by hybrid selection, in vitro transla-tion, and specific immunoprecipitation of the in vitro transla-tion products. Plasmids pOD48 and pOD32 specificallybound OrnDCase mRNA, whereas plasmid pBR322, a plas-

6 mid containing sequences homologous to mouse muscle ac-tin, and pMG20, another of our plasmids containing unrelat-ed cDNA, did not do so. The actin cDNA bound mRNA,whose in vitro translation product had the molecular weight

3 of actin and was not immunoprecipitable by OrnDCase anti-3J4I9 body.The D4.1 mutant has increased levels of in vitro translat-

able OrnDCase mRNA and a homogeneously staining regionon chromosome 14 (10), suggesting that gene amplificationunderlies the overproduction of OrnDCase in this cell line.

A Southern blot analysis of DNA from wild-type and D4.1 S491.5 cells confirmed this conclusion. Although there are multiple

* EcoRI and BamHI fragments homologous to the OrnDCasegene, only one of each is amplified in the D4.1 mutant. Thisindicates that the restriction fragments amplified in D4.1 en-code a functional OrnDCase gene. The identity of the otherbands with sequence homology is not yet determined. They

* may represent additional functional OrnDCase genes, pseu-0.75 dogenes, or DNA that cross-hybridizes to a noncoding re-

* * gion of the cDNA probe.Our previous studies have shown that the increased pro-

duction of OrnDCase in the D4.1 mutant is accompanied byan increased amount of in vitro translatable OrnDCasemRNA. The RNA dot hybridization and RNA transfer anal-ysis show that increased OrnDCase production is also ac-

0.38 companied by an increased amount of hybridizable OrnD-* Case mRNA and that the most abundant size class ofmRNA

is the same in S49 wild-type and D4.1 cells. Androgen treat-ment of male BALB/c mice has been reported to induce

idneys of untreated OrnDCase activity in kidney by >50-fold, in part due to in-was spotted in dupli- creased stability of the OrnDCase polypeptide (6, 29). In thents a serial 1:2 dilu- experiments reported here, the induction of enzyme activityd 5, untreated mice; was 16-fold. We found that hybridizable OrnDCase mRNA

was induced -5-fold and that the size classes of OrnDCasemRNAs are identical to that in uninduced mouse kidney.

49 wild-type and These results imply that an increase in OrnDCase mRNAontrol mouse kid- comprises one component of androgen-mediated inductionation, again using of OrnDCase in mouse kidney. The size of the major OrnD-ig the intensity of Case mRNA species in S49 cells is identical to that inlied in a 1:10 dilu- BALB/c mouse kidney, the mouse strain from which S49contain at least a lymphoma cells were derived. Therefore, the structure ofoared to wild-type OrnDCase mRNA has probably not become altered as a re-d the level of that sult of oncogenic transformation, establishment in culture,Iney (results not maintenance, and selection for gene amplification in this cell-y was measured line.it analysis in a 1:2 OrnDCase activity exhibits regulation during developmentnd to increase its and cellular transformation and regulation initiated by a pan-enzyme activity oply of hormones and growth factors. Activity of the enzyme

imals. The activi- is controlled in a variety of tissues and cultured cells in such6, and 6.1 units a way that, in general, cells that are proliferating exhibite were 209, 156, more activity than their nonproliferating counterparts. Thethe mean induc- isolation of the OrnDCase cDNA clones will allow detailed

analysis of the mechanisms of regulation of this enzyme.

We thank L. Persson for a gift of OrnDCase antiserum. This workwas supported by National Institutes of Health Grant CA 29048,American Cancer Society Grant CD 132, and National ScienceFoundation Grant PCM 78-07382.

1. Pegg, A. E. & Williams-Ashman, H. G. (1981) in Polyaminesin Biology and Medicine, eds. Morris, D. R. & Marton, L. J.(Dekker, New York), pp. 3-42.

2. Janne, J., Poso, H. & Raina, A. (1978) Biochim. Biophys. Acta473, 241-293.

3. Kameji, T., Murakami, Y., Fujita, K., Noguchi, T. & Hayashi,S. (1981) Med. Biol. 59, 296-299.

4. Seely, J. E., Poso, H. & Pegg, A. E. (1982) Biochemistry 21,3394-3399.

5. Kitani, T. & Fujisawa, H. (1983) J. Biol. Chem. 258, 235-239.6. Isomaa, V. V., Pajunen, A. E. I., Bardin, C. W. & Janne,

0. A. (1983) J. Biol. Chem. 258, 6735-6740.

Genetics: McConlogue et aL

544 Genetics: McConlogue et aL

7. Pritchard, M. L., Seely, J. E., Poso, H., Jefferson, L. S. &Pegg, A. E. (1981) Biochem. Biophys. Res. Commun. 100,1597-1603.

8. Persson, L. (1983) Acta Chem. Scand., Ser. B 36, 685-88.9. Seely, J. E. & Pegg, A. E. (1983) J. Biol. Chem. 258, 2496-

2500.10. McConlogue, L. & Coffino, P. (1983) J. Biol. Chem. 258,

12083-12086.11. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. & Rutter,

W. J. (1979) Biochemistry 18, 5294-5299.12. Aviv, H. & Leder, P. (1972) Proc. Natl. Acad. Sci. USA 69,

1408-1412.13. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular

Cloning (Cold Spring Harbor Laboratory, Cold Spring Harbor,NY), pp. 230-238.

14. Michelson, A. M. & Orkin, S. H. (1982) J. Biol. Chem. 257,14773-14782.

15. Taub, F. & Thompson, E. B. (1982) Anal. Biochem. 126, 222-230.

16. St. John, T. P. & Davis, R. W. (1979) Cell 16, 443-452.17. Palmiter, R. D., Chen, H. Y. & Brinster, R. L. (1982) Cell 29,

701-710.18. Parnes, J. R., Velan, B., Felsenfeld, A., Ramanathan, L., Fer-

rini, U., Appella, E. & Sidman, J. G. (1981) Proc. Natl. Acad.Sci. USA 78, 2253-2257.

19. Pelham, H. R. B. & Jackson, R. J. B. (1976) Eur. J. Biochem.67, 247-256.

20. McConlogue, L. & Coffino, P. (1983) J. Biol. Chem. 258, 8384-8388.

21. Laemmli, U. K. (1970) Nature (London) 227, 680-685.22. McConlogue, L. & Coffino, P. (1983) J. Cell. Biol. 96, 762-767.23. Blin, N. & Stafford, D. W. (1976) Nucleic Acids Res. 3, 2303-

2308.24. Southern, E. (1975) J. Mol. Biol. 98, 503-517.25. Rave, N., Crokvenjakov, R. & Boedtker, H. (1979) Nucleic

Acids Res. 6, 3559-3567.26. Thomas, P. S. (1980) Proc. Natl. Acad. Sci. USA 77, 5201-

5205.27. Denhardt, D. T. (1966) Biochem. Biophys. Res. Commun. 23,

641-646.28. Minty, A. J., Caravatti, M., Robert, B., Cohen, A., Daubas,

P., Weydert, A., Gros, F. & Buckingham, M. E. (1981) J. Biol.Chem. 256, 1008-1014.

29. Seely, J. E., P6so, H. & Pegg, A. E. (1982) J. Biol. Chem. 257,7549-7553.

30. Chang, A. C. Y., Nunberg, J. H., Kaufman, R. J., Erlich,H. A., Schimke, R. T. & Cohen, S. N. (1978) Nature (Lon-don) 275, 617-624.

31. Wahl, G. M., Padgett, R. A. & Stark, G. R. (1979) J. Biol.Chem. 254, 8679-8689.

32. Brennand, J., Chinault, A. C., Konecki, D. S., Melton, D. W.& Caskey, C. T. (1982) Proc. Natl. Acad. Sci. USA 79, 1950-1954.

Proc. Natl. Acad ScL USA 81 (1984)