Molecular Medicine |
From the Department of Medicine (M.L.N., K.V.V., E.B., P.F.B.) and the Program in Cellular and Molecular Medicine (M.L.N.), Johns Hopkins University School of Medicine, Baltimore, Md.
Correspondence to Paul F. Bray, MD, Baylor College of Medicine, Thrombosis Research Section, Department of Medicine, One Baylor Plaza, BCM 286, Room N1319, Houston, TX 77030. E-mail pbray{at}bcm.tmc.edu
| Abstract |
|---|
|
|
|---|
3.7 kDa larger than ERß in a variety of
cell lines (including those of prostate and breast origin). A rigorous
investigation of platelet ERß mRNA by reverse
transcriptasepolymerase chain reaction revealed normal transcripts
and a single alternately spliced mRNA. However, this variant form was
smaller, lacking exon 2, and could not account for the larger protein
size seen in platelets. Treatment of ERß with
N-glycosidase F, which removes
core carbohydrate residues, caused a more rapid migration through
polyacrylamide gels but had no effect on ERß from human cell lines.
We conclude that the larger form of ERß in human platelets is not
attributable to alternate mRNA splicing but primarily to
tissue-specific
glycosylation.
Key Words: platelet estrogen receptor glycosylation
| Introduction |
|---|
|
|
|---|
Estrogen can modulate changes nongenomically (by
affecting Ca2+ flux, lipid metabolism,
etc)5 7 14 15
and genomically (through the estrogen receptor
[ER]),8 16 17
but relatively little is known about the effects of estrogens on
platelets. There are 2 forms of the estrogen receptor, discovered 10
years apart, named ER
and ERß to designate the order in which they
were
identified.18 19
These two forms of the receptor share considerable sequence identity in
their DNA and hormone-binding domains; they diverge considerably in
their activation domains.20
ER
and ERß are coexpressed in many human tissues but have
different downstream effects in response to various estrogenic
ligands.21 22 23
ER
knockout mice show estrogen responsiveness, suggesting that ERß
can function in the absence of
ER
.24 25 Both
receptors have been cloned from human
tissue,18 26 but
neither has been cloned from platelet material.
Previously, we identified ERß transcript and protein in megakaryocytes, platelets, and human erythroleukemia cells and noted that platelet ERß protein was larger than the protein in breast or prostate.27 In this study, we demonstrate that platelet ERß protein is glycosylated, offering an explanation for the larger size of the protein in platelets. We also identify two ERß transcripts from platelets, one of the expected size and one in which exon 2 is deleted. This is the first novel isoform of ERß described in human platelets. The implications of tissue-specific glycosylation of this receptor are discussed.
| Materials and Methods |
|---|
|
|
|---|
Use of Human Subjects
Healthy human volunteers were recruited to give
blood, and informed consent was obtained. Approval for use of human
donors was obtained from the Institutional Review Board at Johns
Hopkins School of Medicine, and procedures used were within the
guidelines established by the university and the Helsinki Declaration
of 1975. A total of 5 healthy women (age range 24 to 27 years) and 13
men (age range 23 to 49 years) were used.
Western Blotting
Platelet and cell lysates were prepared as previously
described,27 with platelets
being lysed in buffers containing 1% Triton X-100 and cell lines being
lysed in buffers containing 1% Nonidet P-40, 0.5% sodium
deoxycholate, and 0.1% SDS. Protein concentration in the lysates was
quantified using the BCA protein assay (Pierce), and polypeptides were
separated by electrophoresis through 10% polyacrylamide gels. The gels
were transferred to nitrocellulose, and the blots were incubated with
different anti-ERß antibodies, as previously
described.27
RT-PCR Characterization of ERß mRNA
MCF-7 cells were grown in DMEM-F12 supplemented with
FBS (Gemini Bio-Products) and penicillin/streptomycin. T47D cells were
grown in RPMI 1640 medium supplemented with 10 µg/mL insulin and
penicillin/streptomycin. Human erythroleukemia cells (American Type
Culture Collection) and LNCaP cells were grown in RPMI 1640 medium
supplemented with FBS and penicillin/streptomycin. Cell lines and
platelets were pelleted, and total RNA was isolated with RNAStat-60
(Tel-Test) according to the manufacturers instructions. RT-PCR was
carried out as described
previously.27 Amplified
products were cloned into bacterial vector pCR 2.1 using the TA cloning
system (Invitrogen), and regions of interest were sequenced using
universal primers on an automated DNA sequencer.
Protein Predictions
PredictProtein
(http://www.embl-heidelberg.de/predictprotein/predictprotein.html) was
used to predict domain structure and potential modification sites of
ERß. The published sequence of ERß (Genbank No. AB006590) was used;
topology of membrane proteins, as well as identification of potential
membrane helices, signal peptides, O-linked glycosylations, N-linked
glycosylations, phosphorylation sites, and myristoylation sites, was
determined.
N-Glycosidase
F Digestions
LNCaP (20 µg) and platelet lysates were treated
with 1 U of N-glycosidase F
(Roche Molecular Biochemicals) in a final volume of 20 µL for 18
hours at 30°C and subjected to Western blotting. As a positive
control for deglycosylation, an anti-integrin
v antibody (Life Technologies) was
used.
| Results |
|---|
|
|
|---|
3.7 kDa larger (range
from 7 different experiments, 1.9 to 6.6 kDa) than ERß protein from
nonplatelet sources. In addition to the LNCaP and T47D cells, platelet
ERß was larger than ERß in Dami (human megakaryocytic), Du145
(human prostate), PC-3 (human prostate), HS578T (human breast), and
Chinese hamster ovary cell lines (data not shown). The size difference
was consistently present in platelets from both female and male
subjects.
|
We considered whether alternate splicing could account
for the larger ERß seen in platelets. Because platelets are anucleate
and have only trace amounts of RNA, we used RT-PCR to characterize
ERß mRNA. The
Table
and
Figure 2A
contain the nucleotide sequence and location of
the primers used in these studies. Primers from the coding region were
able to amplify ERß from both platelets and nonplatelet sources
(Figures 2B
through 2D).
Figure 2B
shows that 1 of the 2 sequences amplified from
platelet cDNA using primers MN5 and MN14 was a smaller product than
expected. Amplified products from this region were subcloned and
sequenced, and nucleotide sequences were compared with published
sequences. This comparison confirmed that both products correspond to
ERß. The smaller product produced from the RT-PCR of platelet cDNA
(Figure 2B
) corresponds to a splicing variation described
previously in human pituitary
adenomas.31 As shown in
Figure 3
, exon 2 is skipped in this transcript, which
results in a frameshift and premature termination of the transcript. If
this truncated protein were expressed, it would contain only 122 amino
acids with a size of
13.5 kDa and, therefore, could not account for
the larger size of ERß seen in platelets. The larger product
amplified from primers MN5 and MN14 was determined to be cDNA
corresponding to the known sequence (ie, containing exon 2). No
sequence divergence was observed in this part of the coding region of
platelet ERß that would explain the larger size seen in
Figure 1
. Primers corresponding to the remaining regions of
ERß were used to amplify ERß from platelets and cell-line sources;
no size anomaly was noted in the rest of the cDNA
(Figures 2C
and 2D
). Thus, neither alternate mRNA splicing nor
other nucleotide sequence differences could account for the larger size
of ERß in platelets.
|
|
|
Posttranslational modifications represent another
means by which protein size may vary. Recent reports identifying
membrane-associated forms of estrogen
receptors32 33 34 35
raise the possibility that the size difference between platelet ERß
and nonplatelet ERß was attributable to glycosylation. We digested
both platelet and LNCaP lysates with several glycosidases to see if
carbohydrate removal changed the electrophoretic mobility of ERß.
Figure 4
shows a Western blot of lysates digested with
N-glycosidase F, an enzyme that
removes sugar chains joined to a polypeptide at asparagine residues
(N-linked). After enzyme digestion, the size of platelet ERß is much
more similar to LNCaP ERß (shown in 2 different experiments in
Figures 4A
and 4B
), suggesting that the ERß size difference
we observed was largely attributable to N-linked glycosylation. This
finding was consistent over several experiments. The lack of mobility
shift in LNCaP ERß was not attributable to the inability of
N-glycosidase F to act in these
lysates, because the mobility of glycoprotein integrin
v was reduced as expected
(Figure 4C
). Digestion with enzymes to remove O-linked
glycosylation yielded no change in the mobility of platelet ERß, even
after the potential complex sugar chains were reduced by neuraminidase
digestion (data not shown).
|
| Discussion |
|---|
|
|
|---|
3.7 kDa larger than ERß from
other tissues; and, third, the finding of tissue differences in ERß
glycosylation. These findings raise additional mechanistic
possibilities to explain tissue-specific responses to
estrogen. We found that the platelet form of the full-length ERß protein migrates more slowly on a polyacrylamide gel than ERß from other human cell lines. This difference was consistent across many platelet samples from both genders and across a wide range of donor ages. Alternate mRNA splicing is one mechanism that could result in a larger protein product. Although a larger splicing variant has been observed in rat tissue,39 our RT-PCR data show that only the full-length transcript and the exon 2deleted form were present in human platelets. The most likely explanation for our data is that the platelet ERß transcripts of normal size detected by RT-PCR are responsible for the larger than expected polypeptide seen by immunoblotting. The polypeptide could be posttranslationally processed, which would result in slower migration than would be predicted from its amino acid sequence.
Lu et al40 first described forms of human ERß lacking exons, finding deletions of exon 5 or of exons 5 and 6 in human breast tumors. Using RT-PCR followed by cloning and sequencing, we have identified a splicing variant of ERß in platelets that lacks the second exon. This is the first description of an exon-deleted form of ERß from human platelets; a similar splice variant of ERß has been identified in pituitary adenomas.31 Although we did not formally quantify the 2 transcripts, the relative intensities of the amplified PCR products suggested substantial quantities of the exon 2deleted form. If expressed, the resulting polypeptide would be small (13.5 kDa) and would have limited functional capacity as an estrogen receptor, because it would lack the ligand-binding domain, including activation function 2 and all of the DNA-binding domain. However, it could serve as a sink for coregulators or other factors that may interact with the ERß N-terminal domain, decreasing the ability of wild-type ERß to function normally.
We investigated several possible posttranslational mechanisms that might account for the increase in ERß size in platelets. Phosphatase treatment of platelet lysates caused no change in ERß-apparent mobility, making phosphorylation an unlikely contributor to this size difference (data not shown). Glycosylation is another posttranslational mechanism that could potentially increase the mass of the protein. Typically, glycosylation occurs in the rough endoplasmic reticulum, and entrance into it requires a hydrophobic signal sequence, most often in the N-terminus. We used several methodologies to predict ERß hydrophobicity but found nothing suggestive of a signal peptide (data not shown). Nevertheless, there are examples of proteins translocating across membranes despite the absence of a signal peptide (eg, human plasminogen activator41 and ovalbumin42 ). In addition, rat ERß has been reported to be glycosylated on serine or threonine (O-linked glycosylation),43 but this modification is not sensitive to N-glycosidase F. We determined that deglycosylating enzymes for removing N-linked, but not O-linked, sugars resulted in a form of ERß that was closer to the predicted size of the polypeptide. This suggested that most of the size difference between ERß from platelets and cell lines is attributable to glycosylation. Analysis with the PredictProtein program noted several potential sites for N-linked glycosylation sites (asparagine residues in the appropriate amino acid context) on ERß at residues N17, N55, N61, and N407. This glycosylation may be regulated by a tissue-specific mechanism, and additional cell lineages will need to be tested to see if this is unique to platelets. Tissue-specific glycosylation has been described previously,44 45 46 often attributable to tissue-specific expression of the glycosyltransferase responsible for adding the glycosyl moiety to the protein.47 48 Addressing the functional consequences of platelet ERß glycosylation is beyond the scope of these experiments, but 3 asparagine residues lie within the region of ERß known to contain transcriptional activity, and 1 asparagine is in the ligand-binding domain. Additional studies with purified megakaryocytes are required to address this possibility.
Our findings do not speak directly to the role of ERß in platelets. ERß-deficient mice are viable, but their platelet function has not been studied.49 50 The receptor may exert its effects solely at the level of transcription regulation (in the precursor megakaryocyte), and its presence in platelets could be merely residual. It is possible that the exon-deleted isoform may heterodimerize with the form of ERß we observe in platelets. Perhaps this unusual heterodimer could modulate the effects of estrogen in a tissue-specific manner. Alternatively, ERß could have nongenomic effects on platelets and megakaryocytes. We have shown previously that platelet reactivity varies with both the phase of the menstrual cycle and levels of sex hormones,13 51 findings that could be mediated by platelet ERß. Whether platelet-specific ERß glycosylation represents a novel mechanism for regulating the effects of estrogens awaits additional studies.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
and ERß at
AP1 sites. Science. 1997;277:15081510.
and estrogen receptor ß to partial
estrogen agonists/antagonists. Mol
Pharmacol. 1998;54:105112.
and ß is cell type- and promoter-dependent.
J Biol Chem. 1999;274:3200832014.
and ß. Biochem
Biophys Res Commun. 1997;236:140145.[Medline]
[Order article via Infotrieve]
and ß isoforms in human pituitary tumors.
J Clin Endocrinol Metab. 1998;83:39653972.
and ERß expressed in Chinese hamster ovary cells.
Mol Endocrinol. 1999;13:307319.
and a tumor-specific splice variant.
J Clin Endocrinol Metab. 1998;83:33083315.This article has been cited by other articles:
![]() |
K. Aschbacher, R. von Kanel, P. J. Mills, S. Hong, S. K. Roepke, B. T. Mausbach, T. L. Patterson, M. G. Ziegler, J. E. Dimsdale, S. Ancoli-Israel, et al. Combination of Caregiving Stress and Hormone Replacement Therapy is Associated With Prolonged Platelet Activation to Acute Stress Among Postmenopausal Women Psychosom Med, November 1, 2007; 69(9): 910 - 917. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Vitseva, D. A. Flockhart, Y. Jin, S. Varghese, and J. E. Freedman The Effects of Tamoxifen and Its Metabolites on Platelet Function and Release of Reactive Oxygen Intermediates J. Pharmacol. Exp. Ther., March 1, 2005; 312(3): 1144 - 1150. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Q. Chen, M. Delannoy, C. Cooke, and J. D. Yager Mitochondrial localization of ER{alpha} and ER{beta} in human MCF7 cells Am J Physiol Endocrinol Metab, June 1, 2004; 286(6): E1011 - E1022. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Maccarrone, M. Bari, N. Battista, and A. Finazzi-Agro Estrogen stimulates arachidonoylethanolamide release from human endothelial cells and platelet activation Blood, December 1, 2002; 100(12): 4040 - 4048. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Research Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |