Rat prolactin-like protein A partial gene and promoter structure: promoter activity in placental and pituitary cells

~olecu~r and Ceffular E~oc~~fo~, 0 1993 Elsevier Scientific Publishers

91

96 (1993) 91-98 Ireland, Ltd. 0303-72~7/93/$06.~

MCE 03074

Rat prolactin-like

protein A partial gene and promoter structure: promoter activity in placental and pituitary cells

Jean-Claude Vuille a,‘, Peter A. Cattini a, Margaret E. Bock ‘, Annemieke Verstuyf b, Ingo C. Schroedter a, Mary Lynn Duckworth a and Henry G. Friesen a ’ deportment ~~Physfofo~, Uniuersity o~~a~~toba, Winnipeg, ~unitoba, Canada; h Rega Institute for Medical Research, Catholic Unit*ersityLoul~ain, Louvain, Relgium (Received

Key words: Prolactin;

Placenta;

Gene;

Cell-specific

12 March

1993; accepted

15 June 1993)

regulation

Summary Rat prolactin-like protein A (rPLP-A) is a member of a rapidly expanding family of prolactin-related proteins that are expressed during pregnancy by the rat placenta according to specific developmental patterns. Although the factors involved in the pituita~-specific expression of the prolactin and growth hormone genes themselves have been extensively studied, essentially nothing is known of the factors responsible for the placental expression of these new family members. In this paper we describe the isolation of rPLP-A genomic clones, analyze a portion of the 5’ flanking sequence of this gene and use the recently described rat choriocarcinoma cell line, Rcho, in transient transfection studies to show that a 975 base-pair (bp) fragment of 5’ flanking sequence is sufficient to specify placental expression of the rPLP-A gene.

Introduction Rat prolactin-like protein A is a member of an expanding family of prolactin-like proteins that are synthesized by the developing placentas of several species including the rat (Duckworth et al., 1986a,b; 1988), mouse (Linzer and Nathans, 1984, 1985) and cow (Schuler and Hurley, 1987; Tanaka et al., 1989). The rat placenta produces at least six such prolactinrelated proteins which include rat placental lactogens (rPLs) I, II and I variant (Robertson et al., 1990, 1991; Duckworth et al., 1986; Deb et al., 1991a), rat prolactin-like proteins (rPLPs) A, B, and C (Duckworth et al., 1986, 1988; Deb et al., 1991b,c). In addition a group of growth hormone (GH) related proteins has also been identified in the late term rat placenta (Ogilvie et al., 1990).

Correspondence to: Dr. M.L. Duckworth, Dept. of Physiology, Faculty of Medicine. University of Manitoba, 770 Bannatyne Avenue, Winnipeg, Manitoba, Canada R3E 0W3. Tel: (204) 789 3746; Fax (204) 774 9.517. ’ Present address: Division de Diabetologie, HBpital Cantonai Universitaire de Genkve, 24, rue Micheli-du-Crest, 1211 GenZve 4, Switzerland.

Each of the rat prolactin-related proteins is expressed according to a specific temporal and cellular pattern, suggesting that they have specialized functions at different times during pregnancy. The rPLP-A mRNA is synthesized from approximately day 14 of pregnancy until term at day 21. In situ hybridization studies have localized the expression of rPLP-A to the cytotrophoblasts (or spongiotrophoblasts) and some giant cells of the basal zone of the rat placenta, a region which has been reported to have direct access to the maternal blood supply (Duckworth et al., 1990). The factors that are responsible for this rPLP-A developmental pattern of expression are not known. The cis- and trans-acting factors that are involved in the developmental expression of the pituitary members of the prolactin/growth hormone gene family have been extensively studied. The nuclear protein pitl/GHF-1 (Ingraham et al., 1988; Bodner et al., 1988) has been isolated and characterized as a homeoboxcontaining DNA binding protein with a further motif, termed the POU domain. It has been shown that this protein not only regulates the transcription of the prolactin and growth hormone genes, but also plays a role in the organogenesis of the pituitary. Snell and Jackson dwarf mice which have both been shown to

92

have a mutation in the homeobox domain of the pit-l gene are depleted in pituitary somatrophs, lactotrophs and thyrotrophs (Castillo et al., 1990; Li et al., 1990). An understanding of the factors that regulate placental expression of the members of this gene family has lagged behind that of the pituitary expressed genes primarily because of the lack of appropriate cell lines in which to carry out studies. Recently a rat choriocarcinema cell line, Rcho (Verstuyf et al., 1990) has been developed from a pure transplantable choricarcinoma tumour (Verstuyf et al., 1989; Teshima et al., 1983). Thcsc cells grow as small undifferentiated cells which arc able to differentiate into placental giant cells in culture. We and others (Faria et al.. 1991; Duckworth et al., in press) have shown that these cells synthesize rPLP-A as well as rPLI and rPLI1. In order to begin an analysis of the DNA sequences and the rruns-acting protein factors that are responsible for the developmental regulation of this family of prolactin-related placental proteins, and to test the uscfulncss of the Rcho cell line for these studies, we have isolated rPLP-A genomic clones, sequenced a portion of the 5’ flanking region and carried out transient transfcction experiments in the Rcho cells. Our results indicate that the Rcho cells represent a useful model system in which to study the tissue specific regulation of the placental genes of the prolactin family. Materials

and methods

Some restriction enzymes, Klenow DNA polymerase and T4 polynucleotide kinease were from Pharmacia; fetal bovine serum, RPM1 1640 medium, sodium pyruvate, Hepes, streptomycin, penicillin, agarose and some restriction enzymes were from Gibco/BRL; Dispase was from Boehringer-Mannheim; 25 cm’ Primaria tissue culture flasks were from Falcon; the vector pGEM7Zf( + ) and Erase-a-base kit for exonuclease III digestions were from Promega; Sequenase was from United States Biochemical; chemicals and MS1 Nitroplus nitrocellulose transfer membrane were from Fisher Scientific; [Q P]dCTP, [ y- ” P]ATP and [ ‘“S]dATP were from Dupont/NEN; the oligonucleotides used as probes and primer were synthesized in the DNA Synthesis Laboratory, University of Manitoba; the Pustell DNA sequence analysis software was from International Biotechnologies. Isolution und characterization of rPLP-A genomic clones A Charon 35 lambda phage library derived from SULI~A partially digested Wistar-Furth rat liver DNA (provided by Dr. C.B. Kaspar, University of Wisconsin) was screened according to standard protocols (Benton and Davis, 1977) using a nick-translated 823 bp rPLP-A cDNA clone, pRP6-5 (Duckworth et al., 1986b). Posi-

tive clones were purified and characterized by restriction enzyme mapping and Southern blot analysis. Four further probes were used for this analysis: a synthetic 1Smer oligonucleotide complementary to 9 nucleotides of the 5’ untranslated region and the first three codons of the signal peptide (nucleotides 1 to 18 of pRP6-5); an EcoRI/BgIII fragment corresponding to nucleotides 1 to 201 of pRP6-5; a HindIII/P.st I fragment corresponding to nucleotides 655 to 823 of pRP6-5, and a 21mer oligonucleotide complementary to nucleotides 761 to 781 in the 3’ untranslated region of pRP6-5. Sequence of the 5’ flanking region A 1.6 kilobase (kb) Hind111 fragment from the genomic clone grPLP-A-11 was identified by the above analysis as containing 5’ flanking information which was contiguous with the mRNA sequence. This fragment was subcloned into the vector pGEM7Zf(+ ). Exonuclease III deletions (Henikoff, 1984) were generated from both directions and sequenced by the dideoxy method (Sanger et al., 1977) using Sequenase and [ “‘S]dATP. Primer extension analysis Primer extension of day 18 rat placental poly(A+) RNA was carried out essentially as described in Duckworth et al. (1988). The primer was complementary to the first 18 nucleotides of the rPLP-A clone, pRP6-5 (5’-CAGATGCATCTCTGAGAT-3’). Reaction products were analyzed on a 6% polyacrylamide/7 M urea sequencing gel. To determine the end nucleotide of the primer extended product, a set of dideoxy sequencing reactions using this same primer with the 1.6 kilobase Hind111 fragment described above, were also electrophoresed on the gel. Cell culture Rcho cells were grown on Falcon Primaria 25 cm2 flasks in RPM1 1640 medium with 20% heat-inactivated fetal calf serum, 1 mM sodium pyruvate and 50 PM P-mercaptoethanol, 10 mM Hepes, 50 U/ml streptomycin and 50 pg/ml penicillin. Cells were harvested using Dispase and routinely passaged every three days before reaching confluency. Rcho cells for transient transfection studies were cultured for 15 days, by which time they had been confluent for about 10 days. The medium was changed every day during that time. Monolayer rat anterior pituitary GC cells were grown as described in Cattini and Eberhardt (1987). RNA analysis Rcho cells were grown as described for 3, 5, 7, 9, 11, 13, 15, and 18 days after plating. Total RNA was isolated according to the procedure of Chomczynski and Sacchi (19871, fractionated in a 1% agarose/2.2 M

93

formaldehyde gel (Maniatis et al., 1982), and transferred onto nitrocellulose. The blot was hybridized to the “2P-labelled rPLP-A cDNA clone pRP6-5, as previously described (Duckworth et al., 1986a) and exposed to Kodak XAR film at -70°C with one intensifying screen.

~o~structio~l of promoter-CAT c~~rneric&smids The E. co/i chloramphenicol transacetylase (CAT) reporter vector pCHAT (Cattini and Eberhardt, 19871, digested with HindIII/BglII to remove SV40 promoter sequences, was used in all CAT constructs. The CAT cassette contains an SV40 polyadenylation signal se-

A B

-A

-D

Bgl II (201)

C -E

Hind III (655) I

I

rPLP-AcDNA

u

0.1 Kb

1

ATG (IO)

‘TAA (693)

Bani HI ECORI I

Hind IIl

Hind III

I

grPLP-A- 11

Hind Ill

I

I

1.0 Kb

Barn HI Hind81

ECORI I

1

Hind111 HindIII

EC0 RI

Hind III

ECORI

I

grPLP-A-13

I

grPLP-A-16-2

Hind III EC0 RI

Hind III

A-

B-

c

*

4

D-

E-

B PSI I

1

BamHI

Hind III

-975rPLP-Ap.cat

-c975

CAT +45

Fm

-46~PLP-Ap.~at

td -4m

RI

Hind III

,uw - 975

, t45

CA?.

Fig. 1. (A) Restriction enzyme maps of three overlapping rPLP-A lambda genomic clones, grPLP-A-I 1. grPLP-A-13 and grPLP-A-16-2. The probes (A to EI used for the analysis of these clones are shown in relation to the rPLP-A cDNA clone, pRP6-5. Probe A is an oligonucleotide complementary to nucleotides 1 to 18; probe B is an EcuRI/Bg[II fragment containing nucleotides 1 to 201; probe C is the complete 823 nudeotide pRP6-5 cDNA clone, containing 9 nucleotides of 5’ untranslated region, and the entire coding and 3’ untranslated regions; probe D is a HindIII/EcoRI fragment containing nucleotides 655 to 823; probe E is an oligonucleotide complementary to nucleotides 761 to 781 in the 3’ untranslated region. The extent of the hybridization of the three genomic clones to each of these probes is indicated by arrows below the genomic clone maps; all fragments between the arrows hybridize to the specific probes. Probe E which contains only 3’ untranslated information hybridizes to the same fragments as the longer probe D. (B) A restriction enzyme map of the 5’ region of the rPLP-A gene used in the cur constructs; the region between the arrows has been sequenced and is given in Fig. 3. The first exon, representing nucleotides + I to +9S. is illustrated by the stippled box. The two cat constructs used in the transfection experiments, - 975rPLP-Ap.cat and ~ 46OOrPLP-Ap.cat are also illustrated. The -4600 is an approximate number based on restriction enzyme mapping. The TATA box is represented by the hatched box; both clones are ligated to the rut gene at a PsrI site at position + 45.

94

quence (German et al., 1982). The following fragments were cloned into this digested vector: -4600rPLPAp.cat - an EcoRl/PstI rPLP-A fragment from approximately - 4600 to +45 was blunt-ended using Klenow polymerase and cloned into the blunt-ended pCHAT; - 975rPLP-Ap.cat - a HindIII/Pst I rPLP-A fragment - 975 to + 45 in which the PstI site of the clone and the &/II site of the vector were blunt-ended by Klenow polymerase; -8OOrPLIIp.cat - a PztuII/ Pr,uII 5’ flanking fragment from approximately -800 to +70 of the rPLI1 gene was cloned into the bluntended pCHAT. Other constructs used were a Rous sarcoma enhancer/ promoter cat construct (RSVp.cat) (obtained from Dr. M. Walker, University of Califormia, San Fransciso, CA), a promoterless cut gene ( - p.cat), and the human chorionic somatomammotropin A promoter ligated to the cut gene (-496 to + lCS-Ap.cat) (Nachtigal et al., 1989).

GATC 201, :ii% . 160,

123~

90,

76, 67,

470

Transient transfection studies Rcho cells grown for 15 days as described were transiently transfected with 5 pg of plasmid per 25 cm2 flask by the calcium phosphate method as described in Cattini and Eberhardt (1987). Monolayer rat anterior pituitary GC cells were grown for l-2 days and transfected as described in Nickel et al. (1990). CAT assays The CAT activity was measured by a two-phase fluor diffusion assay (Neumann et al., 1987). Cells were harvested and lysed as described in Nickel et al. (1990). To standardize for plate to plate variation in plasmid DNA uptake, nuclear DNA from transfected cells was blotted onto nitrocellulose in a slot blot apparatus, probed with the ‘2P-labelled cut gene and analyzed by autoradiography and densitometry (Nickel et al., 1991). Statistical significance (P < 0.002) was established using an unpaired Student’s t-test.

Fig. 2. Primer extension analysis of the rPLP-A mRNA. The transcription start site of rPLP-A mRNA was determined hy primer extension analysis. A synthetic oligonucleotide (See Fig. 3) complementary to nucleotides l-18 in the cDNA sequence was end-labelled with [y-“‘P]ATP and annealed to 6 pg of day I8 rat placental poly A‘ RNA. The primer was extended using avian reverse transcriptase and the products were fractionated on a 6% polyactylamide/7 M urea sequencing gel. End-lahelled Mspl-cut pBR322 fragments were used as size markers as indicated. Two bands. one nucleotide apart are seen at positions 70/69. A dideoxy sequencing reaction using the same primer to extend a HindIII/HindIII fragment of grPLP-A-11 was also fractionated on the gel and identifies the precise transcription initiation site in the 5’ genomic sequence (Fig. 3).

Results Isolation of rPLP-A genomic clones Four different positive clones were isolated from the Charon 35 rat genomic DNA library using the rPLP-A cDNA clone pRP6-5 (Duckworth et al., 1986b). By restriction enzyme mapping and Southern blot analysis, three of these clones, designated grPLP-A-11 (10.8 kb), grPLP-A-13 (14.8 kb) and grPLP-A-16-2 (14.2 kb), were shown to be overlapping clones which contain the entire coding region as well as 5’ and 3’ flanking information. Restriction enzyme maps and hybridization patterns, using pRP6-5 and fragments of this clone as probes, are illustrated in Fig. 1. A further clone, grPLP-A-15 (data not shown) hybridizes with only the 3’ region probes (D and E of Fig. l), and contains

approximately 4 kilobases of more 3’ information than that found in grPLP-A-16. The overlapping restriction enzyme maps of grPLP-A-13 and grPLP-A-16 predict two EcoRI fragments of 11 kb and 2.8 kb which hybridize with pRP6-5 and correspond closely to the two hybridizing bands seen with this probe in EcoRI digests of rat genomic DNA (Duckworth et al., 1986b). Identification of the transcription initiation start site Using day 18 rat placental poly(A)+ RNA and a primer complementary to the first 18 nucleotides of the pRP6-5 sequence, a single major start site was identified which corresponded to a 70 bp product; a second minor start site was also seen at the adjacent 3’ nucleotide (Fig. 2). Since this primer contains 9 bp of the

9s

Fig. 3. DNA sequence of the 5’ flanking region of the rPLP-A gene including the first exon and a portion of the intervening sequence. A partial nucleotide sequence of the 1.3 kb HindIII/BamHI fragment from grPLP-A-I 1 is shown. Nucleotide numbering is relative to the transcription initiation start site (C) identified by primer extension which is designated + 1. The sequence contains 975 bp of 5’ flanking sequence. the first exon containing 61 bp of untranslated region and 34 bp of coding sequence, and 117 bp of the first intron. The intron/exon splice junction occurs between the first and second nucleotide of a tryptophan codon. Various features of the 5’ flanking region identified in the text are marked as follows: the putative TATA box is double underlined. nucleotide sequences related to binding sites for known DNA binding proteins are underlined and identified, direct and inverted repeats are shown with arrows indicating the direction of the repeat, the region of the primer used in the primer extension studies and as a probe is shown as a broken arrow.

5’ untranslated region, the result predicts an rPLP-A mRNA untranslated region of 61 nucleotides. Isolation and sequence of a fragment containing the 5’ flanking sequence of rPLP-A Gene structure. A 1.6 kb Hind111 fragment from grPLP-A-11 hybridized with the most 5’ 18 bp oligonucleotide used in the primer extension analysis (Fig. 1). It was therefore subcloned into the vector pGEM 7Zf(+ 1 and sequenced. Details of the sequence are shown in Fig. 3. The fragment contains 975 nucleotides of the 5’ flanking region, the entire untranslated region of 61 nucleotides as predicted from the primer extension data and 34 nucleotides of coding region which includes codons for the first eleven amino acids of the signal peptide plus one nucleotide of the next amino acid codon. In addition it contains 11.7 nucleotides of the first intron. The intron/exon splice junction is a type 1 junction (Mount, 19821, interrupting a tryptophan codon between the first and second nucleotides.

5’ Flanking sequence. Features of the 5’ flanking sequence are also shown in Fig. 3. A putative TATA box (TATATA) is located at positions -30 to -25 relative to the initiation start site of the mRNA (Corden et al., 1980). A detailed analysis of this 5’ flanking region reveals three potential AP2 binding sites (Faisst and Meyer, 1992) at - 107 to - 100, -948 to -941 and - 958 to - 951 and two potential PEA3 sites (Wasyslyk et al., 1989) at - 79 to - 74 and -543 to -538. At -515 to - 507 there is a related pit-l binding site in the antisense orientation which varies by one nucleotide from the consensus sequence (Nelson et al., 19881, ATGGATATG, versus ATGNATAT/ AT/A, but contains a critical pit-l binding motif, ATGA/GATAT/A, as identified by scanning mutagenesis (Elsholtz et al., 1990). Two other sequences, ATGTAAAT (- 798 to - 791) and ATGAAAAT (-710 to -703) both in the antisense orientation are related to the octamer motif ATTTGCAT which is recognized by the ubiquitous Ott-1 (Sturm et al., 1988) and the lymphoid-specific Ott-2 (Landolfi et al., 1986; Staudt et al., 1986) transcription factors. In addition to these sites for which specific binding proteins have been identified, the 5’ flanking sequence of rPLP-A gene contains an eleven-nucleotide direct repeat AAATAAATAAA, at - 794 to - 784 and - 823 to -813, a ten-nucleotide inverted repeat TTACTTCTCC, at - 436 to - 427 and - 460 to - 451 and a seven nucleotide inverted repeat AAAATTC at -317 to -311 and -307 to -301.

Transient transfection studies in placental and pituitary cells The giant cells of the rat placental basal zone are known to express rPLP-A in vivo, although they are not the major source of this placental mRNA. The rat choriocarcinoma cell line, Rcho, which differentiates into placental giant cells in culture (Verstuyf et al., 1989, 1990) also synthesizes rPLP-A mRNA and protein. Fig. 4 shows rPLP-A mRNA synthesis in Rcho cultures from 3 to 18 days after plating. In order to determine if the 5’ flanking region of the rPLP-A gene contained sequences that are responsible for the placental specific expression of this mRNA, Rcho cells cultured for 15 days and rat pituitary GC cells were transiently transfected with CAT chimeric constructs in which a - 975 to +45 HindIII/PstI fragment and a 4.6 kb EcoRI/PstI fragment from the 5’ flanking region were ligated to the E. coli cat gene. These results are shown in Table 1. Both 5’ flanking rPLP-A constructs show significant CAT activity (P < 0.002) compared to a promoterless control when transfected into the Rcho cells. In six experiments the chimera containing 4.6 kb of the 5’ flanking region of rPLP-A shows a consistent 5-fold greater level of CAT expres-

Oh

I

428 S

(18s * 1Kb

3

5 7 9 11 13 15 18 day of culture

Fig. 4. Expression of rPLP-A mRNA in rat choriocarcinoma cells. Total RNA was isolated from the rat choriocarcinoma cell line. Rcho, at days 3, 5, 7, 9, 11, 13, 15 and 18 after plating. 30 pg of RNA per lane were fractionated on a I%, agarose/2.2 M formaldehyde gel. blotted onto MS1 membrane and hybridized to a nick-translated rPLP-A cDNA probe. The I-kb rPLP-A mRNA begins to appear at approximately day 5 after plating, a time when there are giant cells in the cultures. Cultures used in the transfection studies were grown for 15 days, a time at which the rPLP-A mRNA is strongly expressed.

sion in the Rcho cells than the shorter clone. Neither of these constructs, however, shows any CAT activity when transfected into the pituitary GC cells although a human chorionic somatomammotropin A promoter construct (-498 to + lCS-Ap.cat) is able to direct

1

TABLE HYBRID

rPLP-A

GENE

EXPRESSION

IN Rcho AND GC CELLS

Hybrid rPLP-A-cat gene expression in placental and pituita~ cell lines. Hybrid cur constructs containing 5’ flanking sequences from rPLP-A f - 4.6 and - 9751, rat placental tactogen II f - 800 rPLII1, human chorionic somatomammotropin (CS-Al, Rous sarcoma virus (RSV) or no promoter ( - pl were transfected into Rcho rat choriocarcinoma cells or rat pituitary CC cells as described in Materials and methods. Details of the constructs are also described in Materials and methods. CAT activity is expressed as cpm/min/mg cell protein (means k SEM). All values were corrected for plasmid uptake as described. Values are for CAT activity above background levels measured in the appropriate non-transfected cell line; a value of 116.0+ 14.2 cpm/min/mg protein was obtained from 6 determinations for Rcho. and 18.0& 0.4 cpm/min/mg protein for GC cells. ND is not determined. Statistical significance (P < 0.002) was determined using an unpaired Student’s t-test. Hybrid

gene construct

- p.cut RSVp.rat -4.6rPLP-Ap.cat - 975rPLP-Ap.cat - XOOrPLIIp.cat CS-Ap.cat

CAT activity (cpm/min/mg protein) in Rcho cells

CAT activity (cpm/min/mg protein) in GC cells

4.8k 2.4 3317 k420.1 156.8+ 11.0 33.2+ 3.9 4.6t 2.4 ND

5.5+ 1.6 898.0+ 108.6 0 0 ND 37.0& 4.4

CAT expression in these cells. In addition, a chimera containing approximately 800 bp of the 5’ flanking region of the rat placental lactogen II (rPLI1) gene shows no CAT activity in the Rcho cells which are known to express this mRNA (Faria and Soares, 1991; Duckworth et al., in press). Discussion The number of members of the rat prolactin/growth family of hormones has increased dramatically over the last several years with the identification of the six new members that are expressed almost exclusively in the developing placenta. Although the factors that regulate the highly tissue specific nature of the expression of these various family members has been extensively studied for the pituitary expressed prolactin and growth hormone, there is no information to date about the factors that regulate the expression of the placental members of this gene family. In this paper we have presented data on genomic clones of the rPLP-A, isolated and analyzed 5’ flanking sequence from this gene and tested the rat choriocarcinoma cell line, Rcho, in transfection experiments to determine if it would be a useful model system in which to identify the factors involved in placental specific gene expression. Three overlapping rPLP-A genomic clones have been identified and restriction enzyme mapped. The result with the EcoRI digests of these clones corresponds well with the previous genomic DNA data (Duckworth et al., 1986) which suggested that the rPLP-A gene was a single copy gene greater in size than the 2.1 kb rat growth hormone gene (Barta et al., 1981) and potentially as large as the 10 kb rat prolactin gene (Cooke and Baxter, 19811. These current studies would appear to indicate an rPLP-A gene of approximately 10 kb. The intron/exon boundary that has been mapped and sequenced between exon 1 and intron A shows a class 1 type of splice junction (Mount, 1982) in which a tryptophan codon in the signal peptide is interrupted between the first and second nucleotide. This type of splice junction between the first exon and intron has been identified in all members of the family whose structure has been studied to date. In addition this junction is in an essentially identical location as in other members of the family. Our primer extension studies show that there is a single major transcription start site 61 bp from the putative initiator methionine. Upstream of this start site at positions - 30 to - 25 is the sequence TATATA which is a typical location for a functioning TATA sequence (Corden et al., 1980). There are few consensus binding sequences for known regulatory proteins within the 975 bp 5’ flanking DNA that have been examined. Some that have been identified by sequence comparison include three potential AP2 binding sites

97

which are associated with responsiveness to CAMP, phorbol esters and retinoic acid (Hyman et al., 1989; Liischer et al., 1989) and two sites which contain core PEA3 binding motifs. It has been suggested that the PEA3 protein is a primary target for signal transduction by such factors as serum, phorbol esters and a variety of non-nuclear oncogenes (Wasylyk et al., 1989). In addition there are three sequences that are related but not identical to motifs that bind homeobox containing proteins. These include the sequences at -798 to -791 and -710 to - 703 which are closely related to the Ott-1 and Ott-2 binding sites. Perhaps more interesting, given the relationship of rPLP-A to the rPRL and rGH genes, is a sequence at - 5 15 to - 507 which is only one base different from a canonical pit-l binding site and identical to a critical pit-l binding motif as identified by scanning mutagenesis (Elsholtz et al., 1990). It is the binding of pit-l to its specific sequence that is a determining factor in the pituitary specific expression of the prolactin and growth hormone genes (Ingraham et al., 1988; Bodner et al., 1988). This pit-l site in the rPLP-A gene is more 5’ than the proximal pit-l binding sites in either of the pituitary expressed genes. There is no other notable homology between the rPLP-A 5’ flanking sequence and that of the prolactin and growth hormone genes. This is in marked contrast to the very striking homology between the human growth hormone and the human chorionic somatomammotropin genes in their 5’ flanking regions (Parks, 1989). It has been reported that in the case of the human chorionic somatomammotropin/placental lactogen gene, hCS-B (hPL3), placental specificity is due to sequences located in the distal 3’ flanking region of this gene (Walker et al., 1990). The transient transfection experiments in the rat choriocarcinoma cell line, Rcho, and the rat pituitary GC cells provides a more functional measure of whether this region of DNA is sufficient to permit placental specific gene expression. Until the development of this cell line it has not been possible to address directly the question of what factors may be responsible for the tissue specific expression of the rat placental members of this hormone family. This cell line is able to differentiate into giant cells which is one of the cell types which expresses the rPLP-A mRNA in vivo ~Duckworth et al., 1990) and Rcho cells have been shown to express this mRNA in culture. Our data show that both a 4.6 kb fragment and a smaller 975 bp fragment from the 5’ flanking region of the rPLP-A gene are capable of directing the expression of the reporter cur gene in the Rcho cells, but not in pituitary GC cells. The larger fragment seems to permit higher levels of expression of the reporter than the smaller fragment, perhaps because of the presence of further enhancing elements, but the 975 bp 5’ flanking DNA appears to contain sufficient information to permit

placental expression. In marked contrast to the placental cells there is no expression of CAT in the pituitary GC cells, suggesting that the related pit-l site in the 5’ flanking sequence of rPLP-A does not function and may simply be a remnant of the gene duplication events that appear to be responsible for the evolutionary development of these placental genes. Whether any of the other sequences within this region that have been identified, such as the direct and inverted repeats, have a role in the placental cell expression remains to be determined. In contrast to the rPLP-A sequence, an 800 bp 5’ flanking fragment from the rPLI1 gene does not appear to contain information that allows reporter gene expression in the Rcho cells. We and others have shown that Rcho cells express the rPLI1 mRNA. Our transient transfection studies indicate that the rat choriocarcinoma cell line, Rcho, is a good model system for studying placenta1 specific rPLP-A expression and could prove useful for studying the tissue specific expression of several of the rat placental members of the proIactin/growth hormone family. These cells have been reported to express rPL1, rPLI1, rPLP-A and rPLP-C (Faria and Soares, 1991; Duckworth et al., in press). A very similar rat choriocarcinoma cell line to the one used in our studies has recently been used to examine the placenta1 specific expression of the mouse PLI gene (Shida et al., 1993). The Rcho cells provide not only a transfectable cell system for the identification of the c&acting sequences responsible for placental specific expression of these genes, but should aiso prove to be a rich source of the trans-acting protein factors which bind to these sequences.

This work was supported by the Medical Council of Canada with grants MT-10853 MT-12054 (M.L.D.) and MT-1862 (H.G.F.)

Research (P.A.C.),

References Barta, A., Richards, R.I.. Baxter, J.D., and Shine, J. (1981) Proc. Nati. Acad. Sci. USA 78, 4867-4871. Benton, W.D. and Davis, R.W. (1977) Science 196, 180-182. Bodner. M. Castrillo, J.-L., Theill, L.E., Deemick, T.. Ellisman. M. and Karin, M. (1988) Cell 55, 505-518. Castillo. J.-L., Theill, L.E. and Karin, M. (1990) Science 253, 197-199. Cattini, P.A., and Eberhardt, N.L. (1987) Nucleic Acids Res. 15, 1297- 1309. Chomczynski, P. and Sacchi, N. (1987) Anal. Biochem. 162, 156-159. Cooke, N.E. and Baxter, J.D. (1982) Nature 297, 603-606. Corden, J., Wasylyk, B., Buchwalder, A., Sassone-Corsi, P., Kedinger, C. and Chambon, P. (1980) Science 209, 1406-1414. Deb, S., Faria, T.N., Roby, K.F., Larsen, D., Kwok, SCM., Talamantes, F. and Soares, M.J. (1991a) J. Biol. Chem. 266, 16051610.

YX Deb, S., Roby, K.F., Faria, T.N., Larsen, D. and Soares, M.J. (1991b) Endocrinology 128, 3066-3072. Deb, S, Roby, K.F., Faria, T.N., Szpirer, C., Kwok, S.C.M. and Soares, M.J. (lY91c) J. Biol. Chem. 266, 23027-23032. Duckworth, M.L., Kirk, K.L. and Friesen, H.G. (1986a) J. Biol. Chem. 261, 10871-10878. Duckworth, M.L., Peden, L.M. and Friesen, H.G. (1986,) J. Biol. Chem. 261, 10879-10884. Duckworth, M.L., Peden. L.M. and Friesen, H.G. (1988) Mol. Endocrinol. 2, 912-920. Duckworth, M.L., Schroedter, I.C. and Friesen, H.G. (19901 Placenta 11, 143-15s. Du~kworth, ML., Robertson, MC., Schroedter, IX., Szpirer, C. and Friesen, H.G. in Trophoblast Cells: Pathways for Maternai-Embryonic Communications. Serono Symposia USA (in press). Elsholtz, H.P., Albert, V.R., Treacey, M.N. and Rosenfeld, M.G. (1990) Genes Dev. 4, 43-51. Faisst, S. and Meyer. S. (1992) Nucleic Acids Res. 20, 3-26. Faria, T.N. and Soares. M.J. (1991) Endocrinology 129, 2895-2906. Gorman. CM., Moffat, L.F. and Howard, B.H. (19821 Mol. Cell Biol. 2, 1044-1051. Hyman, S.. Comb, M., Pearlberg, J. and Goodman, H. (19891 Mol. Cell. Biol. 9, 321-324. Ingraham, H.A., Chen, R., Mangalam, H.J., Elsholtz, H.P., Flynn, SE.. Lin, CR., Simmons, D.M., Swanson, L. and Rosenfetd, M.G. (1988) Cell 55, 519-529. Landolfi, N.F., Capra, J.D. and Tucker. P.W. (1986) Nature 323, 548-55 1. Li, S., Crenshaw, E.B., Rawson, E.J., Simmons, D.M., Swanson, L.W. and Rosenfeld, M.G. (1990) Nature 347, 528-533. Linzer, D.I.H. and Nathans, D. (1984) Proc. Natl. Acad. Sci. USA Xl. 4255-4259. Linzer, D.I.H. and Nathans, D. (19851 EMBO J. 4, 1419-1423. Liischer, B., Mitchell, P.J., Williams, T. and Tjian, R. (1989) Genes Dev. 3, 1507-1517.

Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual Cold Spring Harbor Laboratories, Cold Spring Harbor, N.Y. Nelson, C., Albert, V.R.. Elsholtz, H.P., Lu, L.I.W. and Rosenfeld, M.G. (1988) Science 239, 1400-1405. Nachtigal, M.W., Nickel B.E., KIassen, M.E., Zhang, W., Eberhard, N.L. and Cattini, P.A. (1989) Nucleic Acids Res. 17, 4327-4337. Nickel, B.E., Kardami, E. and Cattini, P.A. (1990) Biochem. J. 267, 653-658. Nickel, B.E., Nachtigal, M.W., Bock, M.E. and Cattini, P.A. (1991) Mol. Cell. Biochem. 106, 181-187. Ogilvie, S., Buhi, W.C., Olson, J.A. and Shiverick, K.T. (1990) EndocrinoIo~ 126, 3271-3273. Parks, J.S. (1989) Acta Paediatr. Stand. (SuppI. 349, 127-135. Schuler, L.A. and Huriey, W.L. (19871 Proc. Natl. Acad. Sci. USA 84, 5650-5654. Shida, M.M., Ng, Y.-K., Soares. M.J. and Linzer, D.I.H. (1993) Mol. Endocrinol. 7, 181-188. Staudt, L.M., Singh, II., Sen, R., LeBowitz, J.H., Sharp, P.A. and Baltimore, D. (1986) Nature 323, 640-643. Sturm, R.A., Gokul, D. and Herr, W. (1988) Genes Dev. 2. 15821599. Tanaka, M., Yamakawa, M. Watahiki, M., Yamamoto, M. and Nakashima, K. (19891 Biochim. Biophys. Acta 1008, 1933197. Teshima, S., Shimosato, Y.. Koide, T., Kuroki, M., Kiruchi, M. and Aizawa, M. (1983) Gann 74, 205-212. Verstuyf. A., Sobis, H. and Vandeputte, M. (1989) Int. J. Cancer 44, 879-884. Verstuyf, A., Sobis, H., Goebels, J., Fonteyn, E., Cassiman, J.J. and Vandeputte, M. (19901 Int. J. Cancer 45, 752-756. Walker, W.H., Fitzpatrick, S.L. and Saunders. G.F. (1990) J. Biol. Chem. 265, 12940-12948. Wasylyk, C., Flares, P., Gutman, A. and Wasylyk, B. (1989) EMBO J. 8, 3371-3378.