5'-Heterogeneity in Human
Progesterone Receptor Transcripts
Predicts a New Amino-Terminal
Truncated "C"-Receptor and Unique
A-Receptor Messages
Lisa L. Wei, Carolina Gonzalez-Aller, William M. Wood,
Louise A. Miller, and Kathryn B. Horwitz
Departments of Medicine (L.L.W., C.G.A., W.M.W., K.B.H.) and
Pathology (L.A.M., K.B.H.)
University of Colorado Health Sciences Center
Denver, Colorado 80262
Human progesterone receptors (PR) are thought to
comprise two naturally occurring hormone-binding
proteins: 94-kDa A-receptors and 120-kDa B-receptors. In this paper we present evidence for a third
human PR, an N-terminally truncated, 45- to 50-kDa
species, termed the C-receptor. To determine the
translational origin of B- and A-receptors we mapped
the multiple messages that code for human PR by
Northern blot analyses, using a series of oligonucleotides and cDNA fragment probes corresponding to
different regions of the PR message. In addition to
the six transcripts of 2.5, 3.2, 4.5, 5.2, 6.1, and 11.4
kilobases (kb) originally described, we found that
the 11.4-kb species is a complex of four bands that
we have termed I—IV. Analysis of poly(A)+ RNA derived from T47DV human breast cancer cells using a
variety of 5'-specific probes has identified three
separate structural classes of human PR transcripts,
indicating extensive 5'-termini heterogeneity. Class
A messages, the 2.5- and 5.2-kb species, lack the
sequences surrounding AUGB (codon 1), which is
the translation initiation site for B-receptors, but
contain AUGA (codon 165), the initiation site for Areceptors, and, therefore, potentially encode only
the latter. Class B messages, consisting of the 3.2-,
4.5-, and 6.1-kb species as well as bands I and II of
the 11.4-kb complex contain both AUGB and AUGA
and could encode both receptor forms. Class C
transcripts, bands III and IV in the 11.4-kb complex,
lack AUGB and AUGA and, therefore, encode neither
A- nor B-receptors, but contain down-stream sequences that hybridize to probes complementary to
the DNA- and hormone-binding domains of PR. We
propose that by utilization of an initiator methionine
at codon 595 in exon 2, these messages direct the
synthesis of a 45- to 50-kDa protein that lacks the
N-terminus and first DNA-binding finger of PR, but
contains the second DNA-binding finger, the hinge
region, and the hormone-binding domain. This new
human receptor, the C-receptor, appears to be abundantly expressed in progesterone target cells. (Molecular Endocrinology 4: 1833-1840, 1990)
Progesterone receptors (PR) belong to a superfamily of
ligand-activated transcription factors that share structural similarities. These receptors are assembled from
highly conserved functional segments that include an
N-terminal transcriptional activation domain, a centrally
located DNA-binding domain, and a C-terminal hormone-binding domain (1-3). In the steroid hormone
receptor subfamily, a unique protein, encoded by a
single gene, mediates the effect of each ligand. PR are,
however, an exception, having two functional protein
forms, the B- and A-receptors (4,5). The 94-kDa human
A-receptors are N-terminally truncated, naturally occurring variants of the 120-kDa B-receptors (6,7). The two
proteins are present in approximately equimolar
amounts in human tissues (8), where they may subserve different frans-activating functions (9). Several
mechanisms have been proposed for the synthesis of
colinear proteins from a single gene, including translation initiation from internal AUG codons (10). Conneely
et al. (11) have proposed that in the chick oviduct, the
A- and B-receptors originate from alternate translation
of the same mRNA by initiation at two in-frame AUG
codons. While alternate codon utilization appears to be
the case, it remains unclear (12) whether the downstream AUGA that encodes A-receptors is used in transcripts that contain the up-stream B-receptor codon
(AUGB).
0888-8809/90/1833-1840$02.00/0
Molecular Endocrinology
Copyright © 1990 by The Endocrine Society
Analysis of PR synthesis has been further complicated by the existence of multiple PR transcripts (7,
13-17). In human tissues, at least six messages have
INTRODUCTION
1833
Vol4No. 12
MOL ENDO-1990
1834
been described, ranging in size from 2.5-11.4 kilobases
(kb) (7, 13-15). All are coordinately up-regulated by
estrogens and down-regulated by progestins, and all
are present in normal and malignant progesterone target tissues (13, 14). All six mRNAs are deficient in PRnegative cells that, nevertheless, contain receptors for
other steroid hormones (13,14). Taken together, these
data strongly suggest that all six messages are PR
specific.
To determine the structural relationships among the
multiple PR messages and to identify the mRNAs that
code for A-receptors, we have used oligonucleotide
mapping. We report here that there are not six, but
nine, PR transcripts. Two messages, the 2.5- and 5.2kb species, lack AUGB and contain only AUGA, the Areceptor translation start site. We also found that the
largest transcript, an 11.4-kb species, is, in fact, a
complex of four mRNAs. Two of the messages in the
11.4-kb complex contain neither B- nor A-translation
initiation sites, but contain DNA- and hormone-binding
domain sequences. Translation from an internal initiation AUG encoded in exon 2 in these two messages
would lead to the synthesis of a third, approximately
45- to 50-kDa PR, which we call the C-receptor. We
describe the evidence that such a progesterone-binding
protein is indeed synthesized.
RESULTS
We and others have described six PR mRNAs of 11.4,
6.1, 5.2, 4.5, 3.2, and 2.5 kb in T47D human breast
cancer cells expressing high receptor levels (7,13,14).
The 11.4- and 5.2-kb messages are the major hybridizing species. To map the sequence content of the six
PR messages we synthesized five oligonucleotides
ranging in size from 26-61 nucleotides (nt), spanning
the putative 5' end of the messages including the two
translation start sites at methionine codons 1 (AUGB)
and 165 (AUGA) and two down-stream methionine codons at 289 and 301. We also isolated and subcloned
three PR cDNA fragments that hybridize to the three
major functional domains specific for transcriptional
activation, DNA binding, and hormone binding. The
hPR-50, a 3' terminal PR cDNA that was initially used
to demonstrate the six human PR mRNA species (13,
14), was used as a control. The relative location and
sizes of all of the probes are shown diagramatically in
Fig. 1. Also shown are the major functional domains of
PR and the positions of the putative translation initiation
sites for B- and A-receptors.
For Northern blot analyses, poly(A)+ RNA purified
from T47D human breast cancer cells was used. The
hybridization pattern of each probe was compared to
that obtained with hPR-50 on two halves of the same
nitrocellulose sheet (Fig. 2, lane C). This allowed precise
alignment of mRNA bands. Poly(A)+ RNA from PRnegative MDA cells was included to control for nonspecific binding. While nonspecific binding was not observed with hPR-50 (Fig. 2, lane B), some nonspecific
bands, occasionally overlapping specific bands, were
detected with the oligonucleotide probes (Fig. 2, lane
D). Careful interpretation of the autoradiograms was,
therefore, critical.
Figures 2 and 3 show the hybridization patterns using
hPR-50 and oligo-1. The 33-nt oligonucleotide probe is
of particular interest since it is complementary to a
region including the first initiator methionine (AUGB)- It
extends 32 nt into the coding region and 1 nt into the
5' untranslated region (UTR). The hPR-50 hybridizes to
all of the PR messages in poly(A)+ RNA from T47D
cells, with the 11.4- and 5.2-kb species most abundant
(Fig. 2, lane A). No signal was detected by hPR-50 in
poly(A)+ RNA from MDA cells (lane B). Oligo-1 failed to
detect the 5.2- and 2.5-kb messages from T47D cells.
It also appeared to detect the 11.4-kb complex with
reduced intensity compared to hPR-50 (Fig. 2, lanes C
and E).
To analyze the larger messages more accurately,
longer gels were adopted, the time of electrophoresis
was extended, and Northern analyses with hPR-50 and
oligo-1 were repeated (Fig. 3). Compared to hPR-50
(lanes A and C), oligo-1 sometimes showed a weakly
hybridizing 5.2-kb band, but despite longer exposures
no labeling of the 2.5-kb message was detected (lanes
C and D). A weak signal at 5.2 kb (<10% of hPR-50)
was seen with oligo-1 in two of five experiments, and
2666 2802
619 665
oligo 2
743 778
oligo l
865 925 999 1154
oligo 4 Irag 5
1234 1259
oligo 3
1602 1652
oligo-8
2381
2659
Irag 6
3388
3833
(rag 7
Fig. 1. Human PR cDNA Showing the Major Functional Domains of the Protein (Broad Bar) and the Location of Oligonucleotides
and cDNA Fragments Used for Northern Blotting of PR Messages
Numbers represent nt in PR cDNA according to Kastner et al. (7). The hPR-50 is a cDNA obtained from a T47D-pCD library (13).
AUGB and AUGA are translation initiation sites for B- and A-receptors, respectively (7). UTR are shown by narrow bars.
PR Messages
AUGB
1835
hPR-50
AUG A
V//A
ONA
AUGB
AUG A
hPR-50
Hormone^ ?%%&.
DNA
Hormone*
Stds (kb)
Stds (kb)
PR mRNA (kbJ
.11 .4 I
z > 1 1 .4 II
— 1 1 .4 III
~ ^ 1 1 .4 IV
PR mRNA (kb)
7.46-
7.46-
-3.2
-6.1
-5.2
-4.5
4.40-
-3.2
-2.5
-2.5
2.37Probes:
hPR-50
oligo-1
Fig. 2. Northern Blot Hybridization Analysis of Nine PR Transcripts with hPR-50 and Oligo-1, which Spans AUGB
Poly(A)+ RNA was purified from T47DV cells (lanes A, C, and
E) and MDA cells (lanes B and D), separated by gel electrophoresis, and blotted to nitrocellulose. The nitrocellulose sheet
was cut through lane C. The left half was hybridized with hPR50. The right half was hybridizated with oligo-1, whose location
is shown by the black bar in the diagram at the top. The
position of RNA size standards is shown on the left; the nine
PR messages are marked on the right. A band hybridizing
above the 6.1-kb transcript with oligo-1 is nonspecific (lane E),
as is a band in MDA cells overlapping with the authentic 4.5kb transcript (lane D).
no signal was seen in three experiments despite the
fact that the other small messages were well labeled
with this probe. We tentatively concluded that the 5.2and 2.5-kb mRNAs do not contain AUGB. This was
supported by hybridization with oligo-2, a 47-nt probe
complementary to the region immediately up-stream of
AUGB, which also failed to detect a signal at 5.2 and
2.5 kb (not shown). The data suggest that sequences
up-stream of and surrounding AUGB are missing from
these two transcripts and that they probably do not
encode B-receptors.
The extended electrophoresis shown in Fig. 3 led to
improved resolution of the bands in the 11.4-kb region
and showed that this is a complex of four separate
message species, termed 11.4 I—IV. Oligo-1 (Figs. 2
and 3) and oligo-2 (not shown) clearly failed to hybridize
to bands III and IV of the 11.4-kb complex, suggesting
that they, too, lack AUGB.
Figure 4 shows the hybridization pattern obtained
using oligo-3, a 26-nt probe spanning AUGA. Oligo-3
recognized all of the PR messages with intensities
similar to hPR-50, with the exception of bands III and
IV in the 11.4-kb complex, which were absent. Probe
5, a 155-nt cDNA corresponding to the region between
the two AUGs and ending 85 basepairs (bp) up-stream
of AUGA, hybridized to the same bands as oligo-3 (Fig.
Probes:
hPR-50
oligo-1
Fig. 3. Analysis of the 11.4-kb Message Complex on Long
Gels Using hPR-50 and Oligo-1
Poly(A)+ RNA was purified from T47DV cells (lanes A, C, and
D) and MDA cells (lanes B and E) and separated by electrophoresis on a 20 x 15-cm gel. After electroblotting, the nitrocellulose sheet was cut through lane C and hybridized with
hPR-50 (left) or oligo-1 (right), as described in Fig. 2.
5). This pattern was repeated with oligo-4, which ends
75 bp up-stream of the probe 5 start (not shown).
Based on these data, we conclude that the 5.2- and
2.5-kb bands contain AUGA, but that bands III and IV
of the 11.4-kb complex lack AUGA as well as the region
up-stream of this codon.
To further analyze the structure of bands III and IV in
the 11.4-kb complex, we used three additional downstream probes: oligo-8, a probe down-stream of AUGA
that spans two methionine codons at amino acids 289
and 301; probe-6, a cDNA fragment complementary to
the DNA-binding domain of human PR; and probe-7,
which is specific for exon 8 and includes the extreme
C-terminus and approximately 290 nt of 3' UTR. Figure
6 includes the results of these analyses and summarizes the data with all nine probes. The hPR-50 hybridized to all four bands. However, probes 2 , 1 , 4 , and 5,
all derived from regions up-stream of AUGA, failed to
recognize bands III and IV of the 11.4-kb complex, as
did probe-3 which spans AUGA. Oligo-8 also failed to
recognize bands III and IV, suggesting that additional
down-stream sequences are missing from these transcripts. However, probes 6 and 7, which bind downstream of the N-terminal transcriptional activation domain, do hybridize to these transcripts. Thus, bands III
and IV appear to lack both AUGB and AUGA, but contain,
at least in part, DNA- and hormone-binding sequences.
DISCUSSION
We have identified three functional classes of human
PR mRNAs. Class A transcripts, the 2.5- and 5.2-kb
Vol4No. 12
MOL ENDO-1990
1836
AUGB
hPR-50
AUGA
AUG B
6
1 4 5 3
4 5 3
1
y///
DNA
Stds
CkW
hPR-50
AUG A
A
Stds
CkbD
B
;
DNA
Hormone y
A B C
7.46-
7.464.40-2.5
4.40-
-2.5
2.37Probes:
hPR-50
oligo-3
Fig. 4 . Analysis of Nine Human PR Messages for Sequences
Surrounding AUG A
Poly(A) + RNA w a s purified from T47D V cells (lanes A, C, and
D) and M D A cells (lanes B and E) and separated by electrophoresis on a 2 0 x 15-cm gel. After electroblotting, the nitrocellulose sheet w a s cut through lane C and hybridized with
hPR-50 {left) or oligo-3 (right), w h o s e position is marked by
the black box at the top.
Probes:
hPR-50
frag-5
Fig. 5. Analysis of Nine PR Messages for Sequences between
AUG B and AUG A
Poly(A) + RNA w a s purified from T 4 7 D V cells (lanes A, B, D,
E, and F) and M D A cells (lanes C and G) and separated by
electrophoresis on a 15 x 15-cm gel. After electroblotting, the
nitrocellulose sheet w a s cut through lane D. The left side w a s
hybridized with hPR-50; the right with cDNA fragment 5,
w h o s e position is marked by the black box in the diagram.
species, lack AUGB (codon 1), the translation initiation
site for B-receptors, but contain AUGA (codon 165) and
could encode only the latter. Class B, consisting of the
3.2-, 4.5-, and 6.1-kb species as well as bands I and II
of the 11.4-kb complex, contain AUGB and AUGA and
could potentially give rise to both receptor forms by
alternate AUG initiation (11). It is possible however, that
they generate only B-receptors (12). Class C transcripts, bands III and IV in the 11.4-kb complex, lack
AUGB
14
AUGA
5 3
8
AUGB and AUGA and, therefore, can encode neither
receptor form. However, they contain down-stream
sequences homologous to the DNA- and hormonebinding domains of PR.
11.4
11.4
11.4
HID 1V"
Origin of Multiple PR Messages
Since the steroid receptors are products of a single
gene (15, 18-24), mRNA size heterogeneity has been
attributed to alternate polyadenylation site selection,
resulting in different lengths of the 3' UTR (18). While
this simple explanation may suffice for some of the
steroid receptors having, at most, two or three message
forms, it is unlikely to explain the number and complexity of human PR mRNAs. As the number of message
species increases, the mechanisms by which they are
Fig. 6. Bands III and IV of the 11.4-kb Complex Lack AUGB
and AUGA
Poly(A)+ RNA was purified from T47DV cells and separated
by electrophoresis in several lanes of 20 x 15-cm gels. After
electroblotting, nitrocellulose sheets were cut and hybridized
to each of the nine probes shown in the diagram. The blot
shown in the second lane was from hybridization with oligo-2
and is representative of the pattern seen with probes 1,4, and
5 (see Figs. 2, 3, and 5). Data were pooled from separate
experiments.
1837
PR Messages
generated are likely to become increasingly complex.
Thus, even for human glucocorticoid receptors, which
are encoded by three messages (18), an intricate pattern is emerging, with alternate RNA splicing leading to
two message classes that encode two structurally different proteins, one of which fails to bind hormone (18).
Multiple PR messages are found in all of the species
studied (7, 13-17, 25). In chick tissues four to six
messages, ranging in size from 1.8-8.2 kb, have been
described. Some contain identical coding sequences,
and thus, the differences in their size appear to be due
to variation in the length of the 3' UTR (7, 16, 17).
Jeltsch et al. (17) analyzed the chick PR transcripts and
found that the 4.5-kb message contains the entire 2.1kb coding region, and that alternative polyadenylation,
splicing variants, and 5' truncation generate five other
messages. Failure to splice intron 2, followed by a
down-stream TAA stop codon in the intron, generates
a dominant 3.4-kb message which would encode a
protein containing the N-terminal transcriptional activation domain and the first finger of the DNA-binding
domain. Evidence for a similar transcript was described
by Huckaby et al. (26). However, in chick oviducts at
least, the protein product is either not expressed or is
unstable, since it cannot be detected by immunological
methods (Weigel, N., personal communication).
The chick 4.1-kb message is of particular interest.
Jeltsch et al. (17) postulate that it is a 5'-truncated
transcript initiating between ATGB and ATGA and, thus,
capable of encoding only A-receptors. The mechanisms
for its generation are unknown. Recently, Kastner et al.
(7) have concluded that the 5.2- and 2.9-kb transcripts
of human PR are similarly 5'-truncated and are the
products of a down-stream promoter within the human
PR gene (called promoter A). Our mapping studies are
in agreement with theirs, suggesting also that the 5.2kb message and the smallest message (which we calculate to be 2.5 kb) can encode only form A (Fig. 7).
Thus, there appears to be considerable heterogeneity
in the 5'-termini of transcripts encoding human PR,
leading to structural and functional heterogeneity in the
proteins.
Resolution of the 11.4-kb species into four transcripts
indicates additional levels of complexity at the 5'-end.
Bands III and IV of the 11.4-kb complex do not contain
Met301, Met289, or sequences up-stream, including AUGA
(Met165), so they cannot have been initiated at promoter
A. These two messages do, however, contain DNAbinding domain and hormone-binding domain sequences. Complex scenarios can be envisioned for
synthesis of such N-terminal truncated variants, the
simplest of which are initiation from a putative third
promoter down-stream of Met301; or initiation from the
up-stream promoter B (7), followed by excision of exon
1 (Fig. 7). Such transcripts could use the next available
translation initiation site at Met595, a strong site by the
rules of Kozak (10). It is located at the end of exon 2.
The two messages would encode a 339-amino acid
protein, with a theoretical mass of 38,802 daltons,
containing the second DNA-binding zinc finger encoded
by exon 3 and all of the down-stream functional regions
of PR, including the hormone-binding domain, nuclear
localization signal, and sequences thought to be responsible for receptor dimerization (27, 28) assembled
from exons 4-8. Figure 7 shows a diagram that outlines
the structural organization of this protein, which we call
the C-receptor, and its relationship to B- and A-receptors. Since experimentally and theoretically derived molecular masses of PR differ considerably (the B-receptor
has a theoretical mass of 98,992 but an experimentally
derived mass of 114,000-120,000 daltons), the C-receptor should appear as a 45- to 50-kDa band on
electrophoretic gels. Why is this protein not produced
from all of the messages? It has been suggested that
in vivo AUGA is not used when AUGB is present (12). If
that argument is extended, then Met595 (AUGC) would
not be used in messages containing AUGA. In that case,
bands III and IV of the 11.4-kb complex would uniquely
encode C-receptors.
There is good evidence that Met595 can serve as a
translation initiation site and that C-receptors are expressed in breast cancer cell lines. First, in vitro translation of a truncated human PR cDNA containing ATG595
produces a 45-kDa protein product (6). Second, an
abundant 45- to 50-kDa protein is photoaffinity labeled
by [3H]R5020 in progesterone target cells (5). Third,
immunoblots of extracts from T47D cells using a new
antibody derived from a peptide in the C-terminus of
PR show a major protein of 45-50 kDa (Edwards, D.,
and N. Weigel, personal communication). Note that this
protein could not be detected by the anti-PR antibodies
currently available, whose epitopes have been mapped
to the N-terminus (29, 30).
Significance of A-, B-, and C-Receptors
The heterogeneity of PR protein structure is unique
among the steroid receptors. Considerable evidence
has now accrued to suggest that the B- and A-receptors
are two naturally occurring progestin-binding forms, at
least in chick and man (31 -33). Their unequal regulation
in chick oviducts under different physiological states
has been a curious observation for years (34, 35).
Indeed, the recent demonstration by Tora et al. (9) that
the two forms differentially specify target gene activation in transfection studies suggests that each protein
may have unique functions, perhaps involving interactions with cell-specific factors. Questions that remain
to be addressed include the cellular distribution of the
two forms. Do single cells synthesize only one or the
other receptor, or are both produced in one cell? The
latter raises the possibility that in addition to the two
homodimers, binding of A-B heterodimers to the transcription complex represents a third regulatory state of
the receptors.
We now postulate that additionally, a 45- to 50-kDa
C-receptor is abundantly synthesized in PR-positive
cells. It is translated from two related 11.4-kb transcripts whose levels, like those of the other PR messages, are estrogen regulated (13). This protein could
MOL ENDO-1990
1838
Vol4No. 12
Exon 1
Met B
Met A
Met 301
Met 2 8 9 |
A
l
1
A/B
11.4
6.1
4.5.
I, II
5.2.
2.5
,2,
Met C-595
.5 . 6 , 7 , 8
4
DD
NLS
M
i
I C
,11.4
ID I
III,
E
IV
I
3.2
PRB
PRA
PRc
Fig. 7. Structure and Proposed Origin of Three Human PR
The N-terminus of PR (A/B) is encoded by exon 1 of the PR gene. The two DNA-binding zinc fingers (C) are encoded by exons
2 and 3; the hinge region (D) is within exon 4, and the hormone-binding domain (E) is assembled from exons 4-8 (1). Proposed
sites for a nuclear localization signal (NLS) and dimerization sequences (DD) are also indicated (27, 28). Messages that encode
MetB and can direct synthesis of B-receptors (PRB) include bands I and II of the 11.4-kb complex and the 6.1-, 4.5-, and 3.2-kb
mRNAs. The 5.2- and 2.5-kb messages can only encode A-receptors (PRA) by initiation at MetA. Bands III and IV of the 11.4-kb
complex lack Met301 and all up-stream sequences, but contain C and E domain sequences. A protein of 45-50 kDa (PRC) could be
initiated at the next down-stream initiation site, "Metc" codon 595 in exon 2.
bind hormone and would be able to form not only
homodimers, but also heterodimers with the other PR
forms (27, 28). What the functional consequence or
physiological relevance of this might be is at present
unknown. If this N-truncated receptor is functionally
compromised (36), it could, as a heterodimer, modify
the activity of the B- and A-receptors. Its existence may
require reappraisal of the meaning of PR positive in
breast cancers and other human progestin target cells.
raphy. This fragment lies between AUGB and AUGA, two putative translation start sites (see Fig. 1). Fragment 6 was
generated from the gel-purified hPR-54 insert by digestion with
Aha\\\ and Haelll. This probe spans nt 2381-2659 in the DNAbinding domain. Fragment 7, nt 3388-3833, spans the 3'
terminal portion of the hormone-binding domain and the 3'
UTR of hPR cDNA and is exon 8 specific. This 445-nt fragment
was produced by Saiv3A digestion of purified hPR-50. Complementary DNA probes were nick translated using a commercial kit (Bethesda Research Laboratories, Gaithersburg, MD) and [a-32P]dCTP (3200 Ci/mmol; ICN, Irvine, CA).
The specific activity of the probes was typically 1 x 109 cpm/
HQ DNA.
MATERIALS AND METHODS
Synthetic Oligonucleotides
Cell Culture
O(igonucleotides ranging in size from 26-61 nt were synthesized by the Nucleotide Synthesis Core Facility of the University of Colorado Cancer Center. Oligo-1, a 36-mer, 5' GGA
GCC CGG GGA CCC TTT GCC TTC AGC TCA GTC ATG 3',
spans nt 743-778 of human PR (7). Oligo-2, a 47-mer, 5' TAG
GGG CAG AGG GAG GAG AAA GTG GGT GTT GAA TGT
GGC TGG ACC GG 3', is complementary to nt 619-665 in
the 3' UTR of human PR. Oligo-3, a 26-mer, 5' ACC TTG
CAC CCG GAC CGG CTG ATG AG 3', is complementary to
nt 1234-1259, but contains a single base mismatch at the
third nt of the AUGA codon for another purpose. This oligonucleotide extends 3 nt 5' of AUGA. Oligo-4, a 61-mer, 5' GGA
AGA GTA GCC CGT CCA GGG AGA TAG GTA TGG CCG
AAA CTT CAG GCA AGG TGT CCG AGG T 3', is complementary to nt 865-925 between AUGB and AUGA. Oligo-8, a
51-mer, 5' GAA ATC CAT CAC CGT GGT GGC CAG CGG
GGA GCG CCC GGG CGC CAT CGG CGC 3', is complementary to nt 1602-1652 and spans two Met codons at AUG289
and AUG30i. All oligomers were deprotected and gel purified
before use as probes. They were tested for snap-back formation and lack of homology to other sequences in the Genbank/NBRF DNA/protein data bank (Beckman, Palo Alto, CA).
For use as hybridization probes, oligomers were end labeled
with T4 polynucleotide kinase (Boehringer Mannheim Biochemical) and [a-32P]ATP (4500 Ci/mmol; ICN). The specific
activity of the probes ranged from 2-4 x 108 cpm/^g DNA.
T47DV, described by Graham et al. (37) are human breast
cancer cells in culture that are constitutively PR rich. MDA231 human breast cancer cell lines are PR negative (38). They
were obtained from D. P. Edwards (Denver, CO). Cells were
grown in Falcon (Becton-Dickinson, Oxnard, CA) plastic flasks
(175 cm2) and kept in humidified 5% CO2 and air at 37 C in
medium, as previously described (39).
Complementary DNA Probes
Two human PR cDNA clones, hPR-50 and hPR-54, isolated
from a T47D-pCD library were used for these studies. They
were a gift from B. W. O'Malley (Houston, TX). PR-50 includes
960 nt in the C-terminus of the protein-coding region and 115
nt of the 3' untranslated region (Fig. 1). It encodes part of the
second DNA-binding finger and all down-stream protein sequences. It was subcloned into pGEM-4 (Promega Biotec,
Madison, Wl). PR-54 includes the entire coding region, except
for 200 nt at the 5' end, and was subcloned into pGEM-3.
Fragment 5, which spans nt 999-1154, was generated from
hPR-54 by digestion with Pst\ and SsfEII (Boehringer Mannheim Biochemicals, Indianapolis, IN) and purification by 5%
acrylamide gel electrophoresis and ion exchange chromatog-
PR Messages
Northern Blot Analysis
Total RNA was isolated from cell pellets, and poly(A)+ RNA
was purified from total RNA by affinity chromatography on
oligo(dT)-cellulose, as previously described (13,40). For Northern blot analysis, aliquots of poly(A)+ RNA (7.5-20 ng) were
denatured with formaldehyde and separated by electrophoresis on 15 x 15-cm or 15 x 20-cm 0.8% agarose gels containing
6% formaldehyde, as previously described (13, 41).
When cDNAs were used as probes, nitrocellulose sheets
were prehybridized, then hybridized with 1.5 x 107 cpm/ml
cDNA, and washed as previously described (13). To resolve
all of the messages, including the 11.4-kb complex, longer
gels were usually used. To accurately align bands revealed by
different probes, T47D poly(A)+ RNA was run in at least three
lanes, and the blot was purposely cut through the center of a
lane. In the figures, hPR-50-labeled bands are to the left of the
cut, and bands from test probes are to the right of the cut.
For oligonucleotide probes, nitrocellulose sheets were prehybridized for approximately 16 h at 55 C in a mixture of 6 x
SSPE (1.0 M NaCI, 60 mui NaH2PO4 H2O, and 6 mui EDTA;
final concentration), 5 x Denhart's [0.05% (wt/vol) BSA, 0.05%
Ficoll, and 0.05% polyvinylpyrrolidone; final concentration],
and 0.1% sodium dodecyl sulfate (SDS). Blots were then
hybridized for 48 h at 50 C in 6 x SSPE, but containing 1 x
Denhart's [0.01% (wt/vol) each of BSA, Ficoll, and polyvinylpyrrolidone] and 0.01% SDS (42). Denatured tRNA (250 ng/
ml) was used to block nonspecific binding, and 1.0-1.5 x 107
cpm/ml end-labeled oligomers were added as probe. After
hybridization, filters were washed two or three times for 2-5
min at room temperature in 0.6 x SSPE-0.01% SDS and air
dried. Washing was carefully monitored using a Geiger counter
until background counts were 2-5 counts/sec. For autoradiography, blots were exposed to Kodak XAR-5 film (Eastman
Kodak, Rochester, NY) held in a metal cassette with two
DuPont intensifying screens (Wilmington, DE) at - 7 0 C for 1 7 days.
Acknowledgments
We are grateful to Bert O'Malley for his gift of hPR-50 and
hPR-54, and to our colleague David Gordon for technical
advice.
Received August 6,1990. Revision received September 18,
1990. Accepted September 18,1990.
Address requests for reprints to: Dr. Kathryn B. Horwitz,
Department of Medicine, B151 University of Colorado Health
Sciences Center, 4200 East Ninth Avenue, Denver,
Colorado 80262.
This work was supported by Grants DCB-8709790 from
the NSF, CA-26869 from the NIH, BC-668 from the American
Cancer Society, the National Foundation for Cancer Research,
and the core facilities of the University of Colorado Cancer
Center.
REFERENCES
1. Evans RM1988 The steroid and thyroid hormone receptor
superfamily. Science 240:889-895
2. Beato M 1989 Gene regulation by steroid hormones. Cell
56:335-344
3. O'Malley B 1990 The steroid receptor superfamily: more
excitement predicted for the future. Mol Endocrinol
4:363-369
4. Schrader WT, O'Malley BW 1972 Progesterone-binding
components of chick oviducts. Characterization of purified
subunits. J Biol Chem 247:51-59
1839
5. Horwitz KB, Alexander PS 1983 In situ photolinked nuclear progesterone receptors of human breast cancer
cells: subunit molecular weights after transformation and
translocation. Endocrinology 113:2195-2201
6. Krett NL, Wei LL, Francis MD, Nordeen SK, Gordon DF,
Wood WM, Horwitz KB 1988 Human progesterone Areceptors can be synthesized intracellularly and are biologically functional. Biochem Biophys Res Commun
157:278-285
7. Kastner P, Krust A, Turcotte B, Stropp U, Tora L, Gronemeyer H, Chambon P 1990 Two distinct estrogenregulated promoters generate transcripts encoding the
two functionally different human progesterone receptor
forms A and B. EMBO J 9:1603-1614
8. Horwitz KB, Francis MD, Wei LL 1985 Hormone-dependent covalent modification and processing of human progesterone receptors in the nucleus. DNA 4:451-460
9. Tora L, Gronemeyer H, Turcotte B, Gaub M-P, Chambon
P1988 The N-terminal region of the chicken progesterone
receptor specifies target gene activation. Nature
333:185-188
10. Kozak M 1986 Bifunctional messenger RNAs in eukaryotes. Cell 47:481-483
11. Conneely OM, Kettelberger DM, Tsai M-J, Schrader WT,
O'Malley BW 1989 The chicken progesterone receptor A
and B isoforms are products of an alternate translation
initiation event. J Biol Chem 264:14062-14064
12. Kastner P, Bocquel M-T, Turcotte B, Gamier J-M, Horwitz
KB, Chambon P 1990 Transient expression of human and
chicken progesterone receptors does not support alternative translational initiation from a single mRNA as the
mechanism generating two receptor isoforms. J Biol
Chem 265:12163-12167
13. Wei LL, Krett NL, Francis MD, Gordon DF, Wood WM,
O'Malley BW, Horwitz KB 1988 Multiple human progesterone receptor messenger ribonucleic acids and their
autoregulation by progestin agonists and antagonists in
breast cancer cells. Mol Endocrinol 2:62-72
14. Read LD, Snider CE, Miller JS, Greene GL, Katzenellenbogen BS 1988 Ligand-modulated regulation of progesterone receptor ribonucleic acid and protein in human
breast cancer cell lines. Mol Endocrinol 2:263-271
15. Misrahi M, Loosfelt H, Atger M, Meriel C, Zerah V, Dessen
P, Milgrom E 1988 Complete amino acid sequence of the
human progesterone receptor deduced from cloned
cDNA. Biochem Biophys Res Commun 143:740-748
16. Conneely OM, Dobson ADW, Tsai MJ, Beattie WG, Toft
DO, Huckaby CS, Zarucki T, Schrader WT, O'Malley BW
1987 Sequence and expression of a functional chicken
progesterone receptor. Mol Endocrinol 1:517-525
17. Jeltsch JM, Turcotte B, GamierJM, Lerouge T, Krozowski
Z, Gronemeyer H, Chambon P 1990 Characterization of
multiple mRNAs originating from the chicken progesterone receptor gene. Evidence for a specific transcript
encoding form A. J Biol Chem 265:3967-3974
18. Hollenberg SM, Weinberger C, Ong ES, Cerelli G, Oro A,
Lebo R, Thompson EB, Rosenfeld MG, Evans RM 1985
Primary structure and expression of a functional glucocorticoid receptor cDNA. Nature 318:635-641
19. Huckaby CS, Conneely OM, Beattie WG, Dobson ADW,
Tsai MJ, O'Malley BW 1987 Structure of the chromosomal
chicken progesterone receptor gene. Proc Natl Acad Sci
USA 84:8380-8384
20. Arriza JL, Weinberger C, Cerelli G, Glaser TM, Handelin
BL, Housman DE, Evans RM 1987 Cloning of human
mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science 237:268-275
21. Walter P, Green S, Green G, Krust A, Bornert JM, Jeltsch
JM, Staub A, Jensen E, Scrace G, Waterfield M, Chambon
P 1985 Cloning of the human estrogen receptor cDNA.
Proc Natl Acad Sci USA 82:7889-7893
22. Lubahn DB, Joesph DR, Sullivan PM, Willard HF, French
FS, Wilson EM 1988 Cloning of human androgen receptor
Vol4No. 12
MOL ENDO-1990
1840
23.
24.
25.
26.
27.
28.
29.
30.
31.
complementary DNA and localization to the X chromosome. Science 240:321 -330
McDonnell DP, Mangelsdorf DJ, Pike WJ, Haussler MR,
O'Malley BW 1987 Molecular cloning of complementary
DNA encoding the avian receptor for vitamin D. Science
235:1214-1217
Loosfelt H, Atger M, Misrahi M, Guichon-Mantel A, Meriel
C, Logeat F, Benarous R, Milgrom E 1986 Cloning and
sequence analysis of rabbit progesterone receptor complementary DNA. Proc Natl Acad Sci USA 83:9045-9049
Misrahi M, Loosfelt H, Atger M, Meriel C, Zerah V, Dessen
P, Milgrom E 1988 Organization of the entire rabbit progesterone receptor mRNA and of the promoter and 5'
flanking region of the gene. Nucleic Acids Res 26:54595472
Huckaby CS, Conneely OM, Beattie WG, Dobson ADW,
Tsai M-J, O'Malley BW 1987 Structure of the chromosomal chicken progesterone receptor gene. Proc Natl
Acad Sci USA 84:8380-8384
Guiochon-Mantel A, Loosfelt H, Lescop P, Sokhavuth S,
Atger M, Perrot-Applanat M, Milgrom E1989 Mechanisms
of nuclear localization of the progesterone receptor: evidence for interaction between monomers. Cell 57:11471154
Fawell SE, Lees JA, White R, Parker MG 1990 Characterization and colocalization of steroid binding and dimerization activities in the mouse estrogen receptor. Cell
60:953-962
Estes P, Suba E, Lawler-Heavner J, El-Ashry D, Wei LL,
Toft DO, Sullivan W, Horwitz KB, Edwards DP 1987
Immunologic analysis of human breast cancer progesterone receptors. I. Immunoaffinity purification of transformed receptors and production of monoclonal antibodies. Biochemistry 26:6250-6262
Groyer-Picard MT, Vu-Hai MT, Jolivet A, Milgrom E, Perrot-Applanat M 1990 Monoclonal antibodies for immunocytochemistry of progesterone receptors (PR) in various
laboratory rodents, livestock, humans and chickens: identification of two epitopes conserved in PR of all these
species. Endocrinology 126:1485-1491
Schrader WT, Birnbaumer M, Hughes MR, Weigel N,
Grody W, O'Malley B 1981 Studies on the structure and
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
function of the chicken progesterone receptor. Recent
Prog Horm Res 37:583-633
Horwitz KB, Wei LL, Francis MD 1986 Structural analyses
of progesterone receptors. J Steroid Biochem 24:109117
Clarke CL, Feil PL, Satyaswaroop PG 1987 Progesterone
receptor regulation by 17/3-estradiol in human endometrial
carcinoma grown in nude mice. Endocrinology 121:16421648
Boyd PA, Spelsberg TC 1979 Seasonal changes in the
molecular species and nuclear binding of the chick oviduct
progesterone receptor. Biochemistry 18:3685-3690
Boyd-Leinen P, Gosse B, Rasmussen K, Martin-Dani G,
Spelsberg TC 1984 Regulation of nuclear binding of the
avian oviduct progesterone receptor. Changes during estrogen-induced oviduct development, withdrawal, and
secondary stimulation. J Biol Chem 259:24^1-2421
Bocquel MT, Kumar V, Strieker C, Chambon Pj Gronemeyer H 1989 The contribution of the N- and C-Jerminal
regions of steroid receptors to activation of transcription
is both receptor and cell-specific. Nucleic Acids Res
r
17:2581-2595
Graham ML II, Dalquist KE, Horwitz KB 1989 Simultaneous measurement of progesterone receptors and DNA
indices by flow cytometry: analysis of breast cancer cell
mixtures and genetic instability of the T47D line. Cancer
Res 49:3943-3949
Horwitz KB, Zava DT, Thilagar AK, Jensen EM, McGuire
WL 1978 Steroid receptor analyses of nine human breast
cancer cell lines. Cancer Res 38:2434-2437
Horwitz KB, Mockus MB, Lessey BA 1982 Variant T47D
human breast cancer cells with high progesterone receptor levels despite estrogen and antiestrogen resistance.
Cell 28:633-642
Chirgwin JM, Pryzbyla AE, MacDonald RJ, Rutter WJ
1979 Isolation of biologically active ribonucleic acid from
sources enriched in ribonuclease. Biochemistry 18:52845299
Maniatis T, Fritsch EC, Sambrook J 1982 Molecular Cloning-A Laboratory Manual. Cold Spring Harbor Laboratory,
Cold Spring Harbor, pp 202-203
Thomas PS 1980 Hybridization of denatured RNA and
small DNA fragments transferred to nitrocellulose. Proc
Natl Acad Sci USA 77:5201-5205