|
Introduction:
Breast cancer is the most common cancer in women and the second leading
cause of female cancer mortality. The BRCA1 gene is mutated in 50% of
familial breast cancers. Estrogen plays a fundamental role in breast development
as well as and in the development of breast cancer, and operates through
activation of its nuclear receptors. Disruption of SRC-1 (steroid receptor
coactivator-1) in mice causes partial resistance to multiple hormones,
including estrogen. In addition, SRC-3 (AIB1) is highly amplified in 10%
of primary breast tumors, and its mRNA is over-expressed in 64% of primary
breast tumors. We discovered that, by western blotting with a BRCA1 monoclonal
antibody, it was possible to detect a Protein of 120 kilo Dalton. This
protein appeared to be concentrated in chromatin fractions from isolated
nuclei of multiple human cell lines. Direct mass spectrometric sequencing
of this 120 kD Protein in parallel with the 220 kD BRCA1-Protein, supported
the idea that p120 contained authentic BRCA1 sequences encoded by exons
2-11, while p220 contained peptides from these exons and the following
3` exons encoding the C-terminal region of full length BRCA1. This endogenous,
p120 kD polypeptide appeared to co-immunoprecipitate (co-IP) with endogenous
SRC1, using either BRCA1 monoclonal antibodies that recognize p120, or
monoclonal SRC1 - and 3 antibodies. This interaction could lead to new
insights into the breast and ovarian specificity of BRCA1 related cancer
development.
|
|
Progress
report:
1) To
complete the cloning of human p120 cDNA and to validate and decipher the
structure of the clone and its protein product.
We already know that multiple human cell lines produce a 4.6kb polyA+
RNA in addition to the 7.2kb full-length BRCA1 mRNA. RNA encoding a 120
kD protein from exons 2-11 would be approximately 4-5 Kb in length, depending
upon the nature of its untranslated region. With this as background, we
have probed a human cDNA library from IMR90 (primary fibroblasts) and
a human fetal liver cells library for clones that contain sequences from
exons 2-11 and intron 11. A number of clones have been identified and
together, they represent a cDNA molecule which contains the natural initiating
Methionin codon for p220 in exon 2 and extend all the way through exon
11, ending after ~300 nucleotides within intron 11. From direct sequence
analysis of the relevant overlapping partial clones, the predicted intact
cDNA would contain a ~4300 nucleotide open reading frame which would encode,
as originally predicted (see above), a protein containing all of the sequence
encoded by exons 2 to 11 fused to a piece of 34 amino acids encoded by
the 5` end of intron 11. To support of the hypothesis that the intact
cDNA actually encodes an exons 2-11/intron 11-encoded fusion protein,
we have detected a partial cDNA clone containing much of exon 11 and ~300
nucleotides of intron 11 ending in a poly dA tract with a suitable upstream
poly A addition motif (therefore we call this protein: In-frame Reading
of BRCA1 Intron 11 Splice variant: IRIS). Using 5` RACE, we have identified
a partial 5`-UTR (untranslated region). This 5`-UTR is different from
both, the original BRCA1 exon 1a and 1b 5`UTR. Furthermore, since introns
1a and 1b contain both BRCA1 FL described promoters, this result implies
but does not prove the existence of a new exon (and probably promoter)
somewhere upstream of BRCA1: exon 1a, which we are trying to clone now.
2) To
functionally characterize p120 BRCA1 by comparison with p220 BRCA1.
We transcribed and translated the clonal cDNA in vitro (IVT) in search
of a 35S-methionin labeled product. We have found, that it does co-migrate
with authentic IRIS isolated from a human cell line by BRCA1 monoclonal
Antibody IP. More importantly, we microsequenced these 35S-labelled products,
and found in addition to BRCA1 peptide sequence a fragment that corresponds
to the intron peptide sequence from both IVT and endogenous proteins.
However, we are trying to prove this same point by the partial V8 and/or
chemotryptic proteolysis maps. Additionally, we are generating a number
of rabbit polyclonal antibodies against the intron peptide sequence. We
have affinity purified two of those polyclonal antibodies. One of these
polyclonal Ab works well in IP experiments of the IVT. Another can recognize
the protein in direct western. Both are good in immunoflourescence (IF)
experiments. The IF-experiments showed IRIS to be a nuclear protein, that
is intimately localized to DNA (especially newly synthesized DNA) labeled
with BrdU. The comparison between BRCA1 and IRIS localization was done
as well, and showed that only rarely these two proteins are colocalized
in the nucleus. We will also investigate if this is the case after DNA
damage as well.
3) To
test the hypothesis that p120 is engaged, at least in part, in estrogen-mediated
signal transduction.
Experiments using multiple BRCA1 and SRC-1 specific Antibodies showed,
that IRIS and not BRCA1 co-IP with SRC-1. This phenomenon was not dependent
upon estrogen (E2) addition. Moreover, we have repeatedly observed co-IP
by chromatin IP (CHIP) of SRC-1, a protein with the same BRCA1 immunoreactivity
characteristics as IRIS, and a selected E2-responsive region of the cyclin
D1 promoter in the presence but not the absence of E2. No co-IP was observed
in these assays of a protein with p220 reactivity. Interestingly, on another
E2-responsive promoter, cathepsin D (CathD), no such co-IP effects were
observed, for both proteins. Since then, the important question whether
SRC-1 and 3 truly interact with cloned and endogenous IRIS, was tested.
IRIS and p220 protein binding behavior was tested by performing anti-epitope
tagged co-IP of cell extracts. It confirmed the interaction of SRCs with
IRIS, and not BRCA1 p220. This implies, that IRIS and not BRCA1/p220 is
engaged in some, but not all E2-mediated transcriptional regulatory events.
CHIP of epitope-tagged clonal IRIS and its endogenous counterpart showed
that they actually operate in this manner. Moreover, these CHIP experiments
extended our understanding of the IRIS complex to include integrators
as well, such as p300, CBP and pCAF. Furthermore, time point CHIP after
E2 stimulation was done as well to map the way IRIS involvement in the
ER-signal transduction. Our data suggests that some ER-responsive promoters
(e.g. Cyclin D1, and c-myc), but not all (e.g. CathD), were found to be
occupied by IRIS-complex in waves. A fastwave (~10 min, post E2-stimulation)
that includes co-activators and co-integrators in addition to ER. Another
later wave (~1-2 hrs) showed different composition of the IRIS-complex,
the biological significance of which we are currently investigating.
|
