0021-972X/00/$03.00/0
The Journal of Clinical Endocrinology & Metabolism
Copyright © 2000 by The Endocrine Society
Vol. 85, No. 10
Printed in U.S.A.
Effects of Norplant on Endometrial Tissue Factor
Expression and Blood Vessel Structure*
RADMILA RUNIC, FREDERICK SCHATZ, LIVIA WAN, RITA DEMOPOULOS,
GRACIELA KRIKUN, AND CHARLES J. LOCKWOOD
Department of Obstetrics and Gynecology, New York University School of Medicine, New York, New
York 10016
ABSTRACT
Abnormal uterine bleeding after Norplant administration is primarily responsible for the high discontinuation rate of this safe and
effective long-acting implantable progestin-only contraceptive agent.
Although tissue factor (TF) is the primary initiator of hemostasis,
previous studies indicated that Norplant-associated bleeding persists
despite relatively high TF levels in the stromal compartment. Recently, we determined that progestin-enhanced TF expression during
decidualization of human endometrial stromal cells involves both the
epidermal growth factor receptor and progesterone receptor (PR}. The
current study evaluated TF levels in endometrial bleeding (BL) and
nonbleeding (NBL) sites obtained by camera-guided hysteroscopy
during Norplant contraception. After 1 yr of therapy, immunohistochemical TF levels were unexpectedly higher at BL than at NBL sites.
Use of immunohistochemistry and Western blotting indicated that
L
OW-DOSE, progestin-only contraceptives are specifically recommended when estrogen-containing formulations are contraindicated (e.g. during lactation and for
women at risk for thrombosis) (1). Norplant, the archetypal
implantable form, is especially long acting and is more
convenient to use than oral or injectable progestin-only contraceptives. The Norplant system consists of subdermally
implanted Silastic capsules, which steadily release contraceptive blood levels of levonorgestrel for a 5-yr period (2).
Contraceptive efficacy (⬎99%) results from thickening of the
cervical mucus, which acts as a barrier to sperm penetration,
and suppression or alteration of ovulation. Unfortunately,
irregular and prolonged abnormal uterine bleeding is a frequent cause for the discontinuation of this otherwise safe,
economical, and effective contraceptive agent (3). The majority of women experience prolonged and irregular breakthrough bleeding, and spotting between cycles. These symptoms account for more than half of the removals during the
first year of use and a 30% 5-yr discontinuation rate (4, 5).
Several reports have associated bleeding in endometria
exposed to long-term progestin-only contraceptives with
compromised endometrial microvessels. Abnormal vascular
changes include enlargement and dilatation (6 – 8), increased
Received December 20, 1999. Revision received May 18, 2000. Accepted June 28, 2000.
Address correspondence and requests for reprints to: Frederick
Schatz, Ph.D., Department of Obstetrics and Gynecology, New York
University School of Medicine, 550 First Avenue, New York, New York
10016.
* Supported in part by grants from the NIH (5-R01-HD-33937– 05; to
C.J.L.) and from the General Clinical Research Center (M01-RR-00096).
both sites displayed elevated epidermal growth factor receptor levels
and that the BL sites exhibited high levels of the PR, as well as the
PRA and the PRB isoforms. Microscopic examination of 1-yr biopsies
revealed that significantly larger numbers of enlarged, distended
vessels were present in BL, compared with NBL sites. Elevated TF
levels and abnormally enlarged blood vessels in the BL sites are
consistent with the recently discovered angiogenic role of TF. By
promoting aberrant angiogenesis, chronic endometrial overexpression of TF could produce fragile vessels, which are at increased risk
to bleed. Analysis of endometrial BL and NBL sites, during Norplant
contraception, offers the potential of elucidating local mechanisms
that control enhanced TF expression, leading to abnormal angiogenesis at specific endometrial sites. (J Clin Endocrinol Metab 85: 3853–
3859, 2000)
microvascular density (9, 10), capillary endothelial proliferation (11), and endothelial gaps and hemostatic plugs (12). By
contrast with the spontaneous focal and transient episodes of
hemorrhage from these fragile microvessels, normal menstrual bleeding originates primarily from spiral arterioles in
response to withdrawal of circulating progesterone from the
estradiol-primed endometrium (13, 14).
In the current study, camera-guided hysteroscopy was
used to sample endometrial bleeding (BL) and nonbleeding
(NBL) sites, from the same patients, up through 12 months
of Norplant contraception. The biopsies were evaluated for
expression of tissue factor (TF), the primary initiator of hemostasis (15). Previous studies revealed that TF levels were
elevated in human endometrial stromal cells (HESCs) undergoing decidualization in vivo and in vitro (16, 17). Recently, we determined that progestin-enhancement of TF
expression during in vitro decidualization of HESCs required
costimulation by epidermal growth factor receptor (EGFR)
agonists (18). Moreover, EGFR levels were observed to increase in response to progestin exposure in the HESCs (18).
Therefore, to assess factors that are likely to regulate endometrial TF expression during Norplant contraception, immunoreactive progesterone receptor (PR) and EGFR levels
were also measured in the endometrial biopsies in both BL
and NBL sites.
Materials and Methods
Tissues and bleeding
Written informed consent from patients and approval by the Institutional Board of Research Associates of New York University Medical
Center and Bellevue Hospital were obtained before sampling. Endo-
3853
3854
metrial specimens were obtained by blind pipelle biopsy (Unimar, Willon, CT) across normal menstrual cycles before starting Norplant contraception (control group) and by a 5-mm operative hysteroscope (Karl
Storz Endoscopy-America Inc., Culver City, CA) connected to a video
camera to facilitate separate sampling of BL and NBL (Fig. 1) sites after
3 and 12 months of Norplant treatment. Biopsy specimens were full
thickness with those displaying significant myometrial components excluded from further analysis. BL sites were ascertained by increased
vascularity, and either ecchymosis or overt streaming. Three biopsies
each of BL or NBL sites were obtained per patient. In cases where a BL
site covered a large area, only a portion was biopsied. During this study,
both NBL and BL sites were obtained from the fundus of the uterus to
avoid sampling the lower uterine segment near the cervical canal, where
trauma resulting from dilating the cervix could cause bleeding artifacts.
Control and Norplant-derived biopsies were either fixed in 4% paraformaldehyde and embedded in paraffin or frozen in liquid nitrogen
before storage at ⫺80 C. Table 1 summarizes the bleeding pattern observed after 3 and 12 months of Norplant therapy.
Immunohistochemistry
Five-micron sections were placed on 1% poly-l-lysine-treated slides
(Newcomer Supply, Middletown, WI), then deparaffinized and dehydrated with xylene and ethanol. Endogenous peroxidase was quenched
with 5% hydrogen peroxide in 100% methanol. Before incubation with
primary antibodies, sections were either microwave-heated or, in the
case of the EGFR antibody, were untreated. They were then incubated
overnight at 4 C with either: 1) 0.3 ␮g/mL anti-EGFR monoclonal antibody (Oncogene Science, Inc., Cambridge, MA); 2) 1:80 dilution of
anti-PR monoclonal antibody (Novocastra Laboratories, Newcastle,
UK); 3) 10 mg/mL of anti-PRB isoform monoclonal antibody (clone KC
146) from Dr. G. Greene (University of Chicago, Chicago, IL); 4) 1:500
dilution of antihuman TF rabbit polyclonal antibody from Dr. Y. Nemerson (Mount Sinai School of Medicine, New York, NY); 5) 1:20 dilution
anti-EGFR monoclonal antibody (Zymed Laboratories, Inc., San Francisco, CA); or 6) a prediluted preparation of CD-34 monoclonal antibody
(BioGenex Laboratories, Inc. San Ramon, CA). Negative controls involved substituting nonimmune mouse or rabbit serum for the primary
antibody. Washed sections were treated with antimouse or antirabbitperoxidase conjugate, and color was developed with the Vectastain ABC
kit (Vector Laboratories, Inc. Burlingame, CA). Hematoxylin was used
for counterstaining. Luteal-phase specimens were employed as positive
controls, because they contain high concentrations of stromal cell TF, PR,
and EGFR.
Quantitation of microscopic measurements
At both NBL and BL sites, intensity of immunohistochemical (IHC)
staining and measurements of vessel density, lumen width (estimated
by red blood cell diameter), endothelial cell width, and vascular smooth
muscle thickness were quantitated by two blinded independent observers (R.R. and R.D.). Previously, we described a semiquantitative scoring
system (ranging from none, weak, moderate, and strong) to assess relative intensity of IHC staining in endometrial specimens obtained dur-
FIG. 1. Hysteroscopic biopsy of human
endometrial BL and NBL sites after 3
months of Norplant contraception. BL
site: TO, Tubal ostium in the uterine
fundus; NBL site: the biopsy forceps are
shown.
JCE & M • 2000
Vol. 85 • No. 10
RUNIC ET AL.
ing Norplant contraception (19). Statistical differences between BL and
NBL sites were determined using the Wilcoxon/Kruskal-Wallis rank
sum test.
Western blotting
Frozen tissues were disrupted in a Dounce homogenizer in 4 vol of
ice-cold lysis buffer (25 mmol/L TRIS, 150 mmol/L NaCl, 10 mmol/L
EGTA, 2 mmol/L EDTA, 0.5% Nonidet P-40, pH 7.6), containing a
protease inhibitor cocktail (16), and were centrifuged at 800 ⫻ g for 5 min
at 4 C. The resulting supernatant was centrifuged at 100,000 ⫻ g for 1 h
at 4 C, yielding a cytosolic fraction, which was concentrated using an
Ultrafree-4 centrifugal filter with a 30-kDa cutoff (Millipore Corp., Bedford, MA) and a membrane fraction, which was dissolved in lysis buffer.
Each fraction was resolved on 7.5% SDS PAGE under reducing conditions and subjected to Western blotting. Incubations were carried out
overnight at 4 C with 1:200 dilution of a monoclonal anti-PR antibody
(NEOMARKERS, Fremont, CA) for the cytosolic fraction and a 1:100
dilution rabbit polyclonal anti-EGFR antibody (Oncogene Science, Inc.)
for the membrane fraction. Detection was carried out with enhanced
chemiluminescence (Amersham Pharmacia Biotech, Piscataway NJ).
Densitometry was performed with the Sigmastat program (Jandel Scientific, San Raphael, CA).
Results
Endometrial EGFR expression after Norplant treatment
We have now observed that IHC staining for the EGFR is
weak in stromal cells of proliferative-phase endometrium,
stronger in predecidualized stromal cells localized around
blood vessels of periovulatory specimens, and more intense
in decidualized stromal cells around blood vessels and adjacent to glands in mid- to late-secretory-phase specimens
(results not shown). These findings indicate that, as previously shown for TF expression (16, 17) and for EGFR levels
in HESC monolayers (18), EGFR expression increases conTABLE 1. Total number of bleeding days in norplant patients
studied by hysteroscopic biopsy
Bleeding days in
preceding 90 days
3-Month biopsy
(n ⫽ 20)
12-Month biopsy
(n ⫽ 17)
0 –9
10 –19
20 –29
30 –39
40 – 49
50 –90
Total no. patients
10
2
0
4
2
2
20
7
4
5
0
0
1
17
NORPLANT AND ENDOMETRIAL TISSUE FACTOR
3855
FIG. 3. Western blotting for EGFR in pre-Norplant endometrial controls and BL and NBL sites after 3 and 12 months of Norplant.
Membrane fractions were prepared from the endometria of the various groups, as described in Materials and Methods. Sixty micrograms
protein/lane and molecular weight standards were resolved by 7.5%
SDS-PAGE under denaturing conditions. The results are typical of
three preparations.
comitantly with the progesterone-regulated decidualization
reaction.
As expected, EGFR immunostaining in the proliferativephase control specimen, obtained before Norplant treatment,
is weakly diffuse (Fig. 2A). By contrast, much more prominent EGFR immunoreactivity, equivalent to that observed
for normal secretory endometrium, was evident after 3
months (not shown), and 12 months of Norplant treatment
at both BL (Fig. 2B) and NBL sites (Fig. 2C). The inset shown
in Fig. 2B indicates that, as is the case for periovulatory- and
secretory-phase endometria, IHC staining for the EGFR is
intense in perivascular stromal cells. In the Western blot
shown in Fig. 3: 1) endometrial extracts from both preNorplant secretory-phase control and Norplant-derived
specimens display a major band at 170 kDa, which corresponds to the electrophoretic mobility of the EGFR previously demonstrated in human endometrial extracts (18, 20);
and 2) after Norplant treatment, EGFR levels at 3- and 12month BL sites and at 3-month NBL sites are comparable
with those present in the secretory-phase control specimen,
whereas the 12-month NBL sites contain slightly elevated
EGFR levels. Thus, the Western blotting results for the effects
of Norplant on endometrial EGFR levels (Fig. 3) are consistent with the IHC results (Fig. 2).
PR and its isoforms in Norplant-exposed endometria
FIG. 2. IHC staining for the EGFR in cycling and Norplant-exposed
human endometrium (magnification, ⫻200). A, Pre-Norplant proliferative-phase endometrium. Stromal cells (s) show weak staining,
and the glands (g) are negative. B, Norplant-exposed endometrium,
BL site (12 months). Prominent EGFR staining in the stromal cells.
Note that the inset indicates that EGFR staining is concentrated
around the vessels (arrows) in the stromal compartment and that the
endothelial cells are negative. C, Norplant-exposed endometrium,
NBL site (12 months). EGFR staining is similar to that observed in
B. Results are typical of 5 preparations.
Of the two PR isoforms described, PRB is functionally
active, whereas PRA can antagonize the actions of PRB (21,
22). However, the PRA isoform may mediate progesterone
effects in secretory-phase human endometrial stroma (23).
Figure 4, A and B, shows PR levels in a day-23 secretoryphase pre-Norplant control endometrial specimen. Prominent staining, using an antibody that recognizes a common
epitope on both the PRA and PRB isoforms (Fig. 4A) and an
antibody specific for the PRB isoform (Fig. 4B), is evident in
the nuclei of glands and stroma. Total PR was maintained at
high levels in the nuclei of glands and stroma of both
3-month BL (Fig. 4C) and NBL (Fig. 4D) sites and of 12-month
BL (4E) sites. At 12 months, lower PR levels were evident in
the NBL (Fig. 4F) sites. Similar IHC staining results were
observed for the presence of PRB at BL and NBL sites after
3 months and 12 months of Norplant treatment (not shown).
3856
JCE & M • 2000
Vol. 85 • No. 10
RUNIC ET AL.
FIG. 4. IHC staining for total PR and PRB in cycling and Norplant-exposed human endometrium (magnification, ⫻400). A and B,
Pre-Norplant secretory-phase control endometrium. Total PR IHC staining (A) is present at high levels in the nuclei of stromal cells and
glands. Staining for PRB (B) is comparable with that observed for total PR in the nuclei of both stromal cells and glands. C and D,
Norplant-exposed endometrial BL (C) and NBL (D) sites at 3 months. Total PR is maintained at high levels in the nuclei of endometrial
stromal cells and glands after 3 months of Norplant treatment at both BL and NBL sites. Comparable levels of staining for PRB were seen
(data not shown). E and F, Norplant-exposed endometria BL (E) and NBL (F) sites at 12 months. Maximal levels of total PR were observed
after 12 months of Norplant treatment in nuclei of endometrial stromal cells and glands at the BL sites (E). Comparable IHC staining
for PRB was seen (data not shown). Note that the endothelial cells of the vessel (v) failed to stain for the PR (E). Results are typical of
six specimens.
NBL sites (P ⬍ 0.04). By contrast, no significant differences
in vessel density or in vessel wall or endothelial cell thickness
were found.
TF levels in Norplant exposed endometrium
FIG. 5. Western blotting for PRB and PRA isoforms in Norplantexposed endometria. Cytosolic fractions were prepared from the endometria of the various groups, as described in Materials and Methods. Sixty micrograms protein/lane and molecular weight standards
were resolved by 7.5% SDS-PAGE under denaturing conditions. The
results are typical of four preparations.
The IHC staining results for PR and for PRB are typical of five
preparations.
The effects of Norplant contraception on the expression of
PRA and PRB were assessed by Western blotting with the
antibody recognizing both isoforms. Figure 5 demonstrates
that both PRA (84 kDa) and PRB (116 kDa) were maintained
at levels comparable to those shown for the day-21 secretoryphase specimen in BL and NBL sites of the Norplant-derived
endometrial extracts.
Norplant effects on endometrial vascular morphology and
TF expression
Studies from our laboratory (19), as well as those of others
(6 –12), indicate that Norplant contraception affects the density and integrity of the endometrial microvasculature. The
current study extends these microscopic observations to include biopsies from BL and NBL sites after IHC staining with
the endothelial cell marker CD-34. As illustrated in Fig. 6,
after 3 months of Norplant treatment, a trend towards increased microvascular density was discerned in the BL sites
(6B). Moreover, after 12 months of Norplant treatment, markedly enlarged endometrial vessels were more prevalent in BL
(Fig. 6D) vs. NBL (Fig. 6E) sites. In 10 endometrial specimens
obtained after 12 months of Norplant, a mean 40% increase
in the average lumen width was found in vessels from BL vs.
Consistent with our previous report (19), levels of immunoreactive TF were qualitatively reduced after 3 months of
Norplant treatment (results not shown), compared with TF
levels evident in decidualized stromal cells of secretoryphase endometrial sections. After 12 months of treatment, TF
levels were enhanced at BL (Fig. 7B), compared with NBL
(Fig. 7C) sites. These elevated levels were comparable with
those of the secretory-phase specimen shown in (Fig. 7A).
Thus, in apparent contradiction with the classical role of TF
as the primary initiator of hemostasis (15), 12 months of
Norplant treatment resulted in enhanced TF expression at
the BL sites, which also displayed enlarged, distended vessels (Fig. 6D).
Discussion
The coagulation cascade is initiated by binding of the
coagulation factor VII zymogen or its active form VIIa to
the transmembrane receptor, TF (15). TF is not normally
expressed by cells in contact with circulating blood, such
as endothelial cells or monocytes. However, its expression
by perivascular cells, such as those in the smooth-muscle
layer of large vessels or in the stroma surrounding microvessels, forms a protective hemostatic envelope (24).
That TF affects hemostasis in human endometrium was
suggested by the results of IHC staining and in situ hybridization, which revealed that TF protein and messenger
RNA levels were elevated in decidualized stromal cells of
sections of luteal-phase and gestational endometrium (16,
19). During human implantation, trophoblasts breach endometrial blood vessels embedded in a decidual cell matrix. This process provides the embryo with a vital source
of oxygen and nutrients, but it risks hemorrhage. Decidual
cell-expressed TF can counteract this risk by promoting
local hemostasis.
NORPLANT AND ENDOMETRIAL TISSUE FACTOR
3857
FIG. 6. Immunostaining for CD 34 in pre-Norplant and Norplant-exposed endometrial BL and NBL sites (magnification, ⫻400). A, Pre-Norplant
secretory-phase endometrium. B and C, Norplant-exposed endometrium (3 months) [BL (B) and NBL (C) site]. Blood vessels (arrows) are
delineated by immunostaining with endothelial cell-specific CD 34. D and E, Norplant-exposed endometrium (12 months) [BL (D) and NBL (E)
site]. Blood vessels are denoted by arrows, as above. Results are typical of 10 specimens.
FIG. 7. Immunostaining for TF in
pre-Norplant and Norplant-exposed
endometrial BL and NBL sites (magnification, ⫻400). After 12 months of
Norplant treatment, levels of TF in
the stroma of the BL site (B) were comparable with those of the secretoryphase control specimen (A) and were
much higher than those of the NBL (C)
site. The results are typical of five
preparations.
3858
JCE & M • 2000
Vol. 85 • No. 10
RUNIC ET AL.
The classical hemostasis-mediating role of TF prompted
us to carry out a retrospective study to determine whether
altered endometrial stromal cell TF expression contributed
to Norplant-derived endometrial bleeding. The results revealed that 3– 6 months of Norplant treatment lowered TF
messenger RNA and protein levels, compared with preNorplant controls (19). Thus, the intense bleeding that is
characteristic of this post-Norplant period coincided with
reduced local hemostatic capacity secondary to lowered
TF levels. However, bleeding continued at a lower frequency at 12 months, on NP, despite restoration of TF
expression to similar levels as those found in the secretory
phase of normal menstrual cycles. Moreover, these 12month specimens displayed abnormally enlarged and dilated vessels (19). Thus, continued endometrial bleeding
seemed to reflect vessel fragility and not simply impaired
hemostasis.
The current study sought to elucidate mechanisms underlying prolonged bleeding on Norplant. To extend our
previous retrospective study in which endometria were
biopsied at random, camera-guided hysteroscopy was
now used to specifically sample endometrial BL and NBL
sites up through 12 months of Norplant contraception. In
addition to microscopic examination of the blood vessels
and assessment of TF expression, the biopsies were evaluated for the presence of the EGFR and PR (as well as the
PR) isoforms, PRA and PRB. These endpoints were chosen
because of our recent observations that progestinenhanced TF expression in cultured HESCs required costimulation of the EGFR and that EGFR levels were progestin-enhanced in vitro (18). Recently, Critchley and
colleagues (25) assessed the PR isoform status of Norplantexposed endometria. Given the unavailability of an
antibody against the PRA isoform, IHC staining was
performed with an antibody against total PR, which
recognizes both PR isoforms, and an antibody against
the PRB isoform. They then inferred PRA levels by
subtracting the intensity of PRB immunostaining from
total PR immunostaining. To reduce the subjectivity of
such measurements, the current study evaluated PR isoform status by using IHC staining together with Western
Blotting.
In the current study, after 12 months of Norplant treatment, microvessels with enlarged lumens were preferentially localized at the BL sites. Unexpectedly, TF levels
were selectively up-regulated at the BL sites as well. The
BL sites also contained ample levels of EGFR, as well as
total PR and PRA and PRB isoforms. The specific coexpression of TF, EGFR, and PR at BL sites is consistent with
the absolute requirement for ligand binding to both EGFR
and PR for maximal TF expression in endometrial stromal
cells (18). Beyond its hemostatic role, TF is now known to
mediate angiogenesis (24, 26) and to induce expression of
the primary angiogenic agent, vascular endothelial cell
growth factor (VEGF) (24, 27, 28). Norplant administration
is also reported to increase VEGF levels in endometrial
glands and stroma (10). Recently, we determined that decidualization-related transcriptional enhancement of TF
expression involves mediation by the Sp1 transcription
factor (29, 30). The VEGF gene promoter contains a cluster
of Sp1 factor binding sites (14), suggesting that Sp1 may
be involved in up-regulation of both TF and VEGF expression. Because bleeding can also increase TF levels, we
hypothesize that Norplant administration promotes a
feed-forward loop between PR/EGFR-induced elevated
TF and VEGF levels and abnormally enlarged vessels. In
this model, vessel fragility ultimately overwhelms TFmediated hemostasis, and bleeding ensues. The use of
camera-guided hysteroscopy, to separately biopsy endometrial BL and NBL sites, offers the potential of further
elucidating local mechanisms that directly or indirectly
lead to fragile, easily disrupted blood vessels.
References
1. Boonkasemsanti W, Reinprayoon D, Pruksananonda K, et al. 1996 The effect
of transdermal oestradiol on bleeding pattern, hormonal profiles and sex
steroid receptor distribution in the endometrium of Norplant users. Hum
Reprod. 11(Suppl 2):115–123.
2. Shoupe D, Ballagh S, Mishell DR. 1992 Bleeding patterns in NORPLANT
contraceptive implant users. In: Alexander NJ, d’Arcangues C, eds. Steroid
hormones and uterine bleeding. Washington, DC: AAAS Press; 49.
3. Belsey EM. 1988 The association between vaginal bleeding patterns and reasons for discontinuation of contraceptive use. Contraception. 38:207–225.
4. Sivin I, Sanchez FA, Diaz S, et al. 1983 Three-year experience with NORPLANT subdermal contraception. Fertil Steril. 39:799 – 808.
5. Su-Juan G, Ming-Kun D, Ling-De Z, Ying-Lin L, Shu-Hua W, Sivin I. 1994
A 5-year evaluation of NORPLANT contraceptive implants in China. Obstet
Gynecol. 83:673– 678.
6. Johannison E. 1990 Endometrial morphology during the normal cycle and
under the influence of contraceptive steroids. In: d’Arcangues C, Fraser IS,
Newton JR, Odlind V, eds. Contraception and mechanisms of endometrial
bleeding. Cambridge: Cambridge University Press; 53– 81.
7. Rogers PAW. 1996 Endometrial vasculature in NORPLANT users. Hum
Reprod. 2(Suppl 2):45–50.
8. Hickey M, Fraser I, Dwarte D, Graham S. 1996 Endometrial vasculature in
NORPLANT users: preliminary results from a hysteroscopic study. Hum
Reprod. 11(Suppl 2):35– 44.
9. Rogers PA, Au CL, Affandi B. 1993 Endometrial microvascular density during
the normal menstrual cycle and following exposure to long-term levonorgestrel. Hum Reprod. 8:1396 –1404.
10. Lau TM, Affandi B, Rogers PA. 1999 The effects of levonorgestrel implants on
vascular endothelial growth factor expression in the endometrium. Mol Hum
Reprod. 5:57– 63.
11. Blaustein A, Shenker L, Post RE. 1968 The effects of oral contraceptives on the
endometrium. I. Blood vessels. Int J Fertil. 13:466 – 475.
12. Hourihan HM, Sheppard BL, Belsey EM, Brosens IA. 1991 Endometrial
vascular features prior to and following exposure to levonorgestrel. Contraception. 43:375–385.
13. King RJB, d’Arcangues C. 1992 Steroid hormone effects on uterine blood
vessels. In: Alexander NJ, d’Arcangues C, eds. Steroid hormones and uterine
bleeding. Washington, DC: AAAS Press; 25.
14. Smith SK. 1998 Angiogenesis, vascular endothelial growth factor and the
endometrium. Hum Reprod Update. 4:509 –519.
15. Nemerson Y. 1988 Tissue factor and hemostasis. Blood. 71:1– 8.
16. Lockwood CJ, Nemerson Y, Guller S, et al. 1993 Progestational regulation of
human endometrial stromal cell tissue factor expression during decidualization. J Clin Endocrinol Metab. 76:231–236.
17. Lockwood CJ, Nemerson Y, Krikun G, et al. 1993 Steroid-modulated stromal
cell tissue factor expression: a model for the regulation of endometrial hemostasis and menstruation. J Clin Endocrinol Metab. 77:1014 –1019.
18. Lockwood CJ, Krikun G, Runic R, Schwartz L, Mesia A, Schatz F. 2000
Progestin-epidermal growth factor regulation of tissue factor expression during decidualization of human endometrial stromal cells. J Clin Endocrinol
Metab. 85:297–301.
19. Runic R, Schatz F, Krey L, et al. 1997 Alterations in endometrial stromal cell
tissue factor protein and messenger ribonucleic acid expression in patients
experiencing abnormal uterine bleeding while using NORPLANT-2 contraception. J Clin Endocrinol Metab. 82:1983–1988.
20. Imai T, Kurachi H, Adachi K, et al. 1995 Changes in epidermal growth factor
receptor and levels of its ligands during menstrual cycle in human endometrium. Biol Reprod. 52:928 –938.
21. Tung L, Mohamed MK, Hoeffler JP, Takimoto GS, Horowitz KB. 1993
Antagonist-occupied human progesterone B-receptors activate transcription
without binding to progesterone response elements and are dominantly inhibited by A-receptors. Mol Endocrinol. 7:1256.
NORPLANT AND ENDOMETRIAL TISSUE FACTOR
22. Graham JD, Clarke CL. 1997 Physiological action of progesterone in target
tissue. Endocr Rev. 18:502–520.
23. Wang H, Critchley HO, Kelly RW, Shen D, Baird DT. 1998 Progesterone
receptor subtype B is differentially regulated in human endometrial stroma.
Mol Hum Reprod. 4:407– 412.
24. Carmeliet P, Collen D. 1998 Tissue factor. Int J Biochem Cell Biol. 30:661– 667.
25. Critchley HO, Kelly RW, Wang H, Gebbioe AE, Glasier AF. 1998 Progestin
receptor isoforms and prostaglandin dehydrogenase in the endometrium of
women using a levonorgestrel-releasing intrauterine system. Hum Reprod.
13:1210 –1217.
26. Nakagawa K, Zhang Y, Tsuji H, et al. 1998 The angiogenic effect of tissue
factor on tumors and wounds. Semin Thromb Hemost. 24:297–210.
3859
27. Ollivier V, Bentolila S, Chabbat J, Hakim J, de Prost D. 1998 Tissue factordependent vascular endothelial growth factor production by human fibroblasts in response to activated factor VII. Blood. 91:2698 –2703.
28. Abe K, Shoji M, Chen J, et al. 1999 Regulation of vascular endothelial growth
factor production and angiogenesis by the cytoplasmic tail of tissue factor. Proc
Natl Acad Sci USA. 96:8663– 8668.
29. Krikun G, Schatz F, Mackman N, Guller S, Lockwood CJ. 1988 Transcriptional regulation of the tissue factor gene by progestins in human endometrial
stromal cells. J Clin Endocrinol Metab. 83:926 –930.
30. Krikun G, Schatz F, Mackman N, Guller S, Lockwood CJ. 2000 Regulation
of tissue factor gene expression in human endometrium by transcription factors SP1 and SP3. Mol Endocrinol. 14:393– 400.