| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division of Reproductive Sciences, Oregon Regional Primate Research Center (O.D.S., R.M.B.), Beaverton, Oregon 97006; the Laboratory of Cellular and Molecular Biology, National Cancer Institute (J.S.R.), Bethesda, Maryland 20892; and Amgen, Inc. (D.L.L.), Thousand Oaks, California 91320
Address all correspondence and requests for reprints to: Dr. Ov D. Slayden, Division of Reproductive Sciences, Oregon Regional Primate Research Center, Beaverton, Oregon 97006.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
KGF, also known as fibroblastic growth factor-7 (FGF-7), is a stromally derived, secreted peptide (11) that acts as a mitogen for epithelial cells in culture, but does not stimulate proliferation in fibroblasts, melanocytes, or endothelial cells in vitro (12, 13, 14). The KGF transcript has been detected in the endometrium of a variety of species, including mouse (15), rat (16), human (10, 17), and macaque (2), as well as in the macaque placenta (18). Endometrial KGF transcript levels are increased by P treatment in the macaque (2) and mouse (19) and are elevated during the luteal phase of the menstrual cycle in women (17). In macaques, KGF protein is also more abundant in the endometrium during the luteal phase or after treatment with E2 plus P than in the follicular phase or in animals treated with E2 alone. In the P-treated macaque, this increase in endometrial KGF expression was blocked by cotreatment with the antiprogestin RU 486, indicating that the effects of P on endometrial KGF are mediated by the P receptor (2). Cellular localization studies of KGF transcript in the uterus by in situ hybridization (2) indicate that during P treatment, KGF transcript was most intensely expressed in the stromal cells of the basalis region, the perivascular stroma, and the musculature of spiral arteries.
The KGF receptor (KGFR; FGFR-IIb) has been characterized by both molecular and binding studies. It is a membrane-spanning tyrosine kinase that is present in epithelial cells and absent from all fibroblast cell lines tested (20, 21, 22). Binding studies have indicated that KGF and acidic FGF bind to KGFR with equally high affinity. However, basic FGF binds KGFR with only low affinity and is effectively displaced by low concentrations of both KGF and acidic FGF (20). Structural analysis of mouse KGFR and human KGFR revealed that KGFR is a splice variant of the FGFR-II/bek gene and contains a unique 49-amino acid region near the transmembrane domain (22). Binding studies of alternate splice variants of the FGFR-II gene have confirmed that the 49-amino acid sequence is responsible for the specific binding of KGF to KGFR.
The KGFR has been detected in human endometrium (10, 17), human endometrial carcinoma cells (23), and macaque endometrium (24). Human endometrial KGFR messenger ribonucleic acid (mRNA) is reported to be increased by estrogen (17) and the relative levels of KGFR mRNA in endometrial adenocarcinoma are reported to be similar to those in cycling endometrium (23). In the macaque, RT-PCR analysis revealed that KGFR mRNA was present in endometrial samples from both E2-treated and E2- plus P-treated macaques, but there was no indication that KGFR transcript levels were dramatically affected by E2 or P. Ligand histochemistry with a chimeric KGF-HFc molecule showed that after E2 plus P treatment, binding sites were primarily evident in the basolateral membranes of the glands of the basalis and the spiral arteries (3). These data suggest that during exogenous P treatment or in the natural luteal phase, the basalis glands and the spiral arteries may be sites of KGF action.
At the end of the menstrual cycle, P levels fall, and the abundance of endometrial KGF transcripts decreases about 9-fold. Withdrawal of P induces the menstrual cascade beginning with shrinkage of the entire endometrium, constriction and atrophy of the spiral arteries, up-regulation of matrix metalloproteinases in the upper functionalis zone, dissolution and sloughing of the functionalis accompanied by extensive bleeding, and final healing of the surface and cessation of bleeding (25). In the basalis zones, regression also occurs, but it consists not of tissue sloughing but of extensive apoptosis and glandular atrophy (1). Estrogen-dependent mitosis begins in the residual functionalis around day 5 after P withdrawal (1). The 4- to 5-day period following P withdrawal in the artificial cycle is known as the luteal-follicular transition (LFT) (1).
Because KGF levels fall so dramatically during the LFT, we hypothesized that replacement with exogenous KGF during the LFT would prevent some of the endometrial changes seen after P withdrawal. Here we describe two studies of the effects of exogenously administered KGF during the LFT. First, we administered KGF systemically to ovariectomized, juvenile rhesus macaques during LFT induced by hormonal manipulations. Immature animals weighing approximately 1 kg, were used to conserve KGF peptide and to observe general systemic effects. KGF had dramatic proliferative effects on the bladder (26) and salivary glands (unpublished results, Amgen, Inc., Thousand Oaks, CA), known targets of KGF, but did not affect cell proliferation in the endometrium. Instead, KGF inhibited apoptosis in the basalis zone and had trophic effects on the spiral arteries. In the second experiment, we installed oviductal catheters in ovariectomized adult rhesus macaques and infused KGF directly into the uterine lumen during a hormonally induced LFT. Again, arteriotrophic and antiapoptotic effects were observed in the absence of any effect on cell proliferation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animal care throughout these studies was provided by the
veterinary staff of the Oregon Regional Primate Research Center in
accordance with the NIH Guide for the Care and Use of Laboratory
Animals as approved by the Oregon Regional Primate Research Center
institutional animal care and use committee. The experimental design
for treating juvenile rhesus macaques with KGF is depicted in Fig. 1a
. Twelve juvenile macaques (
1 yr
old) were ovariectomized and sequentially treated with
E2 and P to induce artificial menstrual cycles.
To create these cycles, a 1-cm SILASTIC brand capsule (id, 0.34 cm; od,
0.64 cm; Dow Corning Corp., Midland, MI) packed with
crystalline E2 (Steraloids, Inc., Wilton, NH) was
inserted sc at the time of ovariectomy to stimulate an artificial
proliferative phase. After 14 days of E2 priming,
a 2-cm SILASTIC capsule containing crystalline P (Steraloids, Inc.) was
inserted sc for 14 days to stimulate an artificial secretory phase.
Removal of the P implant (with the E2 implant
left in place) completed each cycle and induced a LFT marked by 2–3
days of menstruation. Serum levels of E2 and P
were measured by RIA (27). To validate that the juvenile macaque
endometrium responded normally to such sequential hormonal
administration, uteri from two animals were taken after 14 days of
E2 treatment and uteri from two animals were
taken after a further 14 days of E2 and P
treatment and processed for histological examination (see
Morphological analysis below). Examination of these tissues
indicated that the juvenile macaque responded normally to cyclic
steroid hormone administration (see Fig. 2). The remaining eight juvenile
macaques, after a complete cycle, were injected iv with either
recombinant human KGF (5 mg/kg; n = 4, Amgen, Inc.)
or vehicle [phosphate-buffered saline (PBS); n = 4;
Sigma, St Louis, MO] daily for 5 days, beginning 1 day
before P withdrawal and continuing through the LFT. At the end of
treatment (LFT day 4), the juvenile macaques were injected iv with
5-bromo-2'-deoxyuridine (Br-dU; 50 mg/kg) dissolved in Hanks’ Balanced
Salt Solution (HBSS; Life Technologies, Inc.,
Gaithersburg, MD) to label cells synthesizing DNA. One hour later the
animals were deeply anesthetized with an overdose of sodium
pentobarbital (>50 mg/kg) and exsanguinated, and the reproductive
tracts were immediately collected. Samples of endometrium were frozen
and cryosectioned for immunohistochemical studies or fixed and embedded
in glycolmethacrylate (GMA) for histological and morphometric
analysis.
|
|
Twelve ovariectomized adult rhesus macaques were treated
sequentially as described above to stimulate artificial menstrual
cycles. The adult macaques were treated with larger implants than those
used in the juvenile macaques, specifically 3-cm
E2-filled SILASTIC capsules and 6-cm P-filled
SILASTIC capsules. Serum levels of E2 and P,
typical of the natural menstrual cycle (1) were found by RIA. During
the test cycle, eight of the monkeys were laparotomized on the day of P
withdrawal, and an oviductal cannula leading to a sc port (VAP access
port model TI 200, Access Technologies, Skokie, IL) was installed (see
Fig. 1b
). The oviductal cannulas (id, 0.64 mm; od, 1.12 mm; Access
Technologies) were fitted with two polyurethane retention beads at 0.5
and 1 cm from the distal tip, inserted 1–1.5 cm via the oviductal
ostium, and secured with two encircling ligatures of silk suture that
engaged the retention beads. The cannula-oviductal connection was
tested by infusing 1 ml 0.1% blue food color (McCormick & Co., Hunt
Valley, MD) dissolved in HBSS, and vaginal swabs were made to assure
that dye-marked fluid exited through the cervix into the vagina. The P
implants were then removed, and 1 ml KGF (0.5 mg/ml) or vehicle (1 ml;
HBSS) was slowly flushed through the reproductive tract over a period
of 1–2 min. Infusions were repeated daily for 4 days during the LFT.
At the end of treatment (LFT day 4), the reproductive tracts of all
animals were flushed via the oviduct with Br-dU (10 mg in 1 ml HBSS).
One hour later, the animals were laparotomized, and the reproductive
tracts were collected and processed as described above. In some of the
monkeys we infused a small quantity of blue dye (200 µL) through the
tract at the end of treatment (immediately preceding Br-dU) to assure
that the cannulas were functional. We previously validated the above
infusion technique by showing that intrauterine infusion of
[3H]thymidine and Br-dU effectively labeled
large numbers of DNA-synthesizing cells throughout the full thickness
of the adult macaque endometrium (data not shown)
Tissue-sampling procedure
To prepare tissue sections for immunohistochemical and
morphometric analyses, the uterine corpus was first cut in half along
the longitudinal axis from fundus to cervical end with a single-edged
razor blade. Each uterine half was then quartered by another
longitudinal cut along the same axis. Cross-sectional blocks of tissue
(5 mm thick) were made by cutting perpendicular to the longitudinal
axis of each quarter. Each block contained a full representation of all
of the endometrial zones, including the luminal surface, functionalis,
basalis, spiral arteries, and uppermost myometrium. The blocks were
then microwaved, fixed, or frozen as described below.
Immunohistochemistry
Microwave irradiation of fresh tissues before cryosectioning greatly improves the morphology and immunohistochemistry of many antigens (28). Therefore, fresh samples of endometrium were irradiated with an Amana Radarrange Touchmatic microwave oven (1500 watts; Amana, IA) for 7 s in 0.5 ml HBSS as previously described (28). The samples were then mounted in Tissue-Tek II OCT (Miles, Inc., Elkhart, IN) and frozen in liquid propane. Frozen sections (5 µm) were cut with a Hacker-Bright cryostat (Fairfield, NJ) and thaw-mounted on SuperFrost Plus slides (Fisher Scientific, Pittsburgh, PA).
Slides bearing cryosections were analyzed for Br-dU incorporation by immunohistochemistry as follows. Slides were microwave irradiated for 2 s, then lightly fixed for 10 min at room temperature in 0.2% picric acid-2% paraformaldehyde in PBS (PAPF). After fixation, the slides were immersed twice for 2 min each time in 85% ethanol and 1.5% polyvinylpyrollidone (PVP) at 4 C, rinsed in PBS, and then immersed twice for 7 min each time in 0.37% glycine in PBS and PVP. After rinsing again with PBS and PVP, the slides were immersed in 0.1% gelatin in PBS and PVP at 4 C. To inhibit endogenous peroxidase activity, the sections were incubated with a solution containing glucose oxidase (1 U/mL), sodium azide (1 mmol/L), and glucose (10 mmol/L) in PBS for 45 min at 25 C, then rinsed. To relax DNA, slides were treated with 2 N HCl at 25 C for 30 min. The sections were incubated for 20 min at 4 C with a nonspecific (equine) serum and then incubated overnight at 4 C with a Br-dU-specific mouse monoclonal antibody (catalogue no. 691991, ICN Biomedicals, Inc., Costa Mesa, CA). After incubation with primary anti Br-dU antibody, the slides were rinsed several times for 2–3 min each time with 0.1% gelatin-0.075% BRIJ (Sigma) in PBS (4 C) and reincubated with nonspecific (equine) serum and then with a biotinylated second antibody (equine antimurine IgG) for 30 min (25 C). The slides were rinsed several times with 0.1% gelatin-0.075% BRIJ in PBS, and the biotinylated antibody complexes were reacted with an avidin-biotin peroxidase kit (Vector Laboratories, Inc., Burlingame, CA) and treated with 3,3'-diaminobenzidine/4HCl (Dojindos DAB, WAKO Chemicals, Richmond, VA). The slides were rinsed several times with deionized H2O and then treated with 0.05% osmium tetroxide for 1 min, rinsed several times with deionized H2O, fixed for 3 min in PAPF, rinsed several times with deionized H2O, dehydrated with ethanol, and coverslipped with Permount (Fisher Scientific, Pittsburgh, PA). For comparison with Br-dU labeling, some sections were stained for Ki-67 antigen. Ki-67 immunocytochemistry was performed as previously described (29).
Apoptosis was evaluated by cytological criteria in GMA sections and using a histochemical technique that detects intranuclear DNA fragments in cryosections (30). For the latter, cryosections from microwave-irradiated samples were fixed with 4% paraformaldehyde in PBS and then immersed twice in 0.37% glycine in PBS and twice in PBS alone. The sections were further fixed with ethanol-acetic acid (2:1) and rewashed twice with PBS. The sections were then analyzed by terminal transferase labeling of DNA fragments with digoxigenin-deoxy-UTP, according to the ApopTag Plus in situ apoptosis detection kit (Oncor, Gaithersburg, MD). For negative controls, the terminal dexoxynucleotidyl transferase enzyme was replaced with H2O. Morphological evidence for basalis zone apoptosis during LFT was previously published (31).
Morphological analysis
GMA sections were prepared and stained with hematoxylin as described previously (29). Low power photographs were made with an Olympus Corp. OM-system 38-mm macro lens on Technical Pan film (Eastman Kodak Co., Rochester, NY). Negatives were digitized with a Polaroid Sprintscan 35 Plus film scanner. High power micrographs were captured through Carl Zeiss planapochromatic lenses with the Optronics DEI-750TD CCD camera (Optronics Engineering, Goleta, CA). Digital images were adjusted for sharpness and contrast with Adobe Pho-toshop (Adobe Systems, Seattle, WA), and photomicrographs were printed with a Sony Mavigraph dye sublimation printer (Sony Corp., Tokyo, Japan).
Morphometrics
The abundance of basalis zone apoptotic epithelial cells in GMA sections, and the abundance of basalis zone ApoTag-positive and Br-dU-positive epithelial cells in immunohistochemical preparations was determined by a trained observer who used an ocular micrometer grid to define microscope fields and counted between 1200–5000 epithelial cells/animal with the aid of a mechanical tabulator. The mitotic nuclei in GMA sections were also counted (3–4 sections/animal) and expressed as the number of mitotic cells per 20 mm2. Endometrial gland area values and spiral artery area values were measured with the Optimas (Optimas, Inc., Seattle, WA) image analysis software package on digital images captured through a Dage-MTI CCD 72 video camera (Dage Corp., Michigan City, ID). To make these measurements, the Optimas program was calibrated for x100 magnification with a stage micrometer (American Optical, Buffalo, NY), and the glands and spiral arteries in the basalis zone of each GMA section (1–2 sections/animal) were traced with the Optimas area morphometry function. The calculated area of the glands and spiral arteries in the basalis zone of each section was then expressed as a percentage of the total endometrial basalis area in the section. Spiral artery areas were also measured by tracing cross-sections of arteries (n = 24/GMA slide) at x250 magnification. The spiral artery wall thickness values were calculated by subtracting the area of the lumen from the cross-sectional area of the arteries and expressed as integrated wall thickness (square microns) per slide. In some cases linear measurements of wall thickness were made with the Optimas line tool to confirm the integrated thickness measurements. The area of secretory vacuoles in the glandular epithelium was also measured with the Optimas percent area function on images of glands from 3 GMA sections/animal, and results were presented as a percentage of the glandular epithelium area. Glandular epithelium thickness was determined by measuring the cell height of 50 epithelial cells/animal at x250 original magnification. Statistical comparisons between vehicle- and KGF-treated monkeys for all morphometric measurements and counts were made by Student’s t test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Sequential treatment of juvenile monkeys with small steroid-filled
SILASTIC implants produced serum levels of E2
(50–100 pg/ml) and P (5–6 ng/ml) indistinguishable from levels
in artificially cycled adult animals (Table 1) and similar to those during the
natural menstrual cycle. In the juvenile animals, withdrawal of the P
implant, as anticipated, resulted in a 2-day menstrual flow beginning
on day 2 of the LFT. To our knowledge, the endometrial histology of
hormone-treated, ovariectomized juvenile rhesus macaques has not been
previously illustrated. The induced proliferative phase endometria were
essentially identical to those of adult macaques. During the induced
secretory phase there was a heightened degree of edema in the upper
functionalis zone compared to that normally seen in similarly treated
adult macaques, but all other aspects of the secretory phase
endometria, such as spiral artery development, glandular secretion, and
pattern of menstrual breakdown were typical of the adult macaque (Fig. 2).
|
Effects of KGF treatment on the endometrial histology of juvenile
macaques are shown in Fig. 3, and the
morphometric data are presented in Table 2. The endometria from both KGF- and
PBS-injected animals showed normal menstrual breakdown of the upper
functionalis (Fig. 3, a and b). However, treatment with KGF
significantly reduced the number of apoptotic cells in the basalis zone
compared to PBS injected controls (P < 0.05; Table 2
and Fig. 3, c and d). This striking inhibition by KGF of basalis
epithelial cell death was evident both by assessment of cytological
criteria in GMA sections and by staining for DNA fragmentation with the
ApoTag Kit. Further, KGF treatment resulted in enhanced sacculation of
the glands in which cell death was inhibited (Fig. 3, a and b). This
enhanced sacculation resulted in a measurable increase in the basalis
gland area and gland epithelium thickness (P < 0.05
compared to vehicle controls). There was no evidence for an effect of
KGF treatment on glandular secretion, at least when assessed
morphometrically by measuring the size and extent (area) of secretory
vesicles in the glands (see Table 2).
|
|
). The thickening of the arterial
walls was apparently due to hypertrophy of the vascular smooth muscle
cells together with an increase in fibrillar tissue that histologically
resembled elastic fibers, interspersed between the layers of smooth
muscle cells (compare Fig. 3, e and f). There was no effect of KGF
treatment on Br-dU labeling of the spiral arteries, indicating that KGF
did not stimulate DNA synthesis or cell proliferation in the spiral
arterial system.
Moreover, in this specific experimental paradigm, we found no evidence
of a mitogenic effect of KGF on any endometrial epithelial cell type.
KGF did not increase rate of Br-dU incorporation, mitotic counts (Table 2), or Ki-67 labeling (not shown) in either the glands of the
functionalis or basalis, the vascular endothelium, or the endometrial
stroma.
However, KGF treatment did significantly increase Br-dU labeling in the
epithelium of the bladder and the salivary glands (Fig. 4), indicating that systemic levels of
KGF in these juvenile macaques were adequate to induce a mitogenic
response in known target cells. Additionally, we noted hypertrophy of
the hard palate, hypertrophy of the gingiva, increased salivation, and
reddening of the skin, all indexes of the systemic action of KGF
compared to vehicle controls. Details of the effects of KGF
administration on nonreproductive systems will be presented in a
separate report.
|
As in the juvenile macaques, administration of KGF did not prevent
menstruation (Fig. 5, a and b), but did
significantly reduce the abundance of apoptotic cells in the basalis
zone assessed both cytologically (Fig. 5, c and d, and Table 3) and by histochemical detection of DNA
fragmentation (Fig. 6, c and d). Also, as
in the juvenile macaques, KGF infusion significantly increased the
basalis glandular area due to more extensive sacculation compared to
the basalis in vehicle-infused animals(P < 0.05; Table 3 and Fig. 5, a and b). This increase in sacculation included both an
increase in gland area and an increase in gland epithelium thickness
(Fig. 6, a and b, and Table 3).
|
|
|
). As in juvenile macaques, this increase was largely due to a
significant increase in integrated spiral artery wall thickness and was
characterized by an apparent hypertrophy of the vascular smooth muscle
cells and deposition of increased amounts of fibrous material with the
histological properties of elastin in the arterial wall (Fig. 5, e and
f).
To determine whether the effect of KGF on the spiral arteries was
to maintain them in the hypertrophied state induced by P treatment or
to stimulate additional hypertrophy, we compared the thickness of the
arteries after an artificial luteal phase with their thickness at the
end of KGF infusion. The mean arterial wall integrated thickness in the
14-day E2- plus P-treated animals was 4518
± 510 µm2, which was significantly less than
9631 ± 712 µm2 observed in the
KGF-treated adults (Table 3). We also assessed wall thickness with a
linear measurement. Linear wall thickness was significantly greater
(P < 0.05) in the KGF group (37.2 ± 2.41 µm)
than in the vehicle (16.1 ± 2.03 µm) or
E2 plus P (19.5 ± 2.86 µm) group, exactly
as with the integrated thickness measurements. Therefore, the effect of
KGF during the LFT was apparently to stimulate further arterial
hypertrophy after P was withdrawn rather than simply maintain the
arteries in the hypertrophied state induced by P treatment.
As in the juvenile macaques, there was no evidence for a mitogenic
action of KGF in the endometrium. Br-dU incorporation (Table 3), Ki-67
immunostaining, and a count of mitotic cells per unit area were not
different between the KGF-infused animals and the vehicle controls in
glandular epithelial cells or vascular endothelial cells. Also, there
were no obvious differences in stromal Br-dU uptake or Ki-67 labeling,
although this was not quantified.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because the upper zones menstruated normally after P withdrawal in both adult and juvenile macaques in the presence of elevated KGF, we suggest that the decline in KGF plays no role in menstrual breakdown. However, apoptosis in the basalis glands and atrophy of the spiral arteries were both blocked by administration of KGF during the LFT. Therefore, KGF may function as both an antiapoptotic and an arteriotrophic factor during the P-dominated luteal phase.
KGF has been shown to have cytoprotective and antiapo-ptotic effects in a variety of systems (19, 26, 32), and the antiapoptotic activity of other FGFs has been associated with various pathways, such as rat sarcoma (RAS)/mitogen-activated protein kinase (33, 34), protein kinase C (35), and phosphatidylinositol 3-kinase (36). However, the precise intracellular step at which KGF acts to block apoptosis in endometrial cells is currently unknown.
A further effect of KGF during the LFT was increased sacculation and folding of the glands in the basalis region. The effect was probably due to the blockade of apoptotic cell death, which resulted in higher than normal numbers of viable cells in the basal glands. The cell crowding that resulted probably led to sacculation and folding as an accommodative response to the increased cell number. KGF treatment also caused a significant increase in glandular cell height, which would contribute to the overall increase in glandular area.
These morphological changes indicate that the basalis glandular epithelium is a target for KGF action, but there was no evidence for a KGF-induced increase in DNA synthesis, Ki-67 antigen, or mitotic figures in these cells. Therefore, this effect of KGF was not a mitogenic one, unless stimulation of DNA synthesis was transient and no longer occurring when tissue was sampled.
The failure to see endometrial proliferative effects was not due to inadequate KGF blood levels, as in the juveniles, the KGF injection induced DNA synthesis in the bladder and salivary glands, known KGF target organs (26) (unpublished results on salivary glands, Amgen, Inc.). Also, in a preliminary study (37) we found that the vaginal and oviductal epithelium of 1-yr-old rhesus monkeys treated for 6 days with 0.5 mg/kg KGF showed increased DNA synthesis, whereas no such increase occurred in the endometrial glands of the same animals.
In another preliminary report we found an antiapoptotic effect of KGF in the uterine luminal epithelium of adult ovariectomized mice (19) treated with E2 and P. However, no mitogenic effects of KGF were observed in the luminal epithelium of these hormone-treated uteri. An effect of exogenous KGF on increased endometrial gland development was reported in neonatal mice (38), and this increase was associated with heightened glandular cell proliferation. The neonatal mouse endometrium is apparently more sensitive than the juvenile or adult macaque endometrium to the mitogenic action of KGF.
Normally P stimulates an increase in endometrial KGF around the midluteal phase of the cycle, a time when the functionalis glands have stopped proliferating and mitosis in the basalis glands is declining. Because the normal P-induced elevation in endometrial KGF is associated with a decline in glandular proliferation, and because KGF did not induce glandular mitosis when infused during the LFT, we suggest that KGF does not function as an endometrial glandular mitogen during the nonfertile primate menstrual cycle.
Our studies are the first to show that KGF can stimulate an apparent hypertrophy of the spiral arteries of the primate endometrium. The mechanism underlying this effect is not clear, as the KGF receptor has not been reported in endothelium or vascular smooth muscle. KGF mRNA, but not KGF receptor mRNA, was detected in cultures of vascular smooth muscle cells and in samples of human arteries, although spiral arteries were not specifically sampled (39). We have reported that a chimeric KGF-HFc molecule (40) binds to spiral arteries in cryosections, but in situ hybridization trials have not revealed strong signals for KGF receptor in these arteries.
Either there is a KGF receptor in spiral arteries that has not yet been detected or there is an indirect mediator of KGF action involved. A likely candidate for such an indirect vascular mediator is vascular endothelial growth factor (VEGF). There are reports that KGF can up-regulate VEGF expression (41), and we have evidence that VEGF receptors I and II are both present in the rhesus macaque spiral arteries (data not shown). Studies to explore possible interactions between KGF and VEGF in primate endometria, especially in the spiral arterial system, are therefore warranted.
In summary, we have established that KGF, a P-dependent growth factor in the primate endometrium, has arteriotrophic and antiapoptotic effects when administered either systemically to juvenile macaques or by local infusion to adult macaques during the LFT. The endometrial basalis glands and the spiral arteries were clear targets of KGF action. Although KGF can clearly act as a mitogen on many epithelial cell types, we found no evidence for this role in the primate endometrium. Its most important roles may be to stimulate spiral artery growth and inhibit glandular apoptosis during the nonfertile menstrual cycle. Because its expression rises coincident with the time of implantation, the role of KGF in the fertile menstrual cycle deserves further study.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received April 22, 1999.
Revised June 15, 1999.
Revised August 12, 1999.
Accepted August 26, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-mediated apoptosis
requires Ras and the activation of mitogen-activated protein kinase. J Biol Chem. 271:14560–14566.This article has been cited by other articles:
 |
|
 |
|
  E. van Esch, J. M. Cline, E. Buse, and G. F. Weinbauer The Macaque Endometrium, with Special Reference to the Cynomolgus Monkey (Macaca fascicularis) Toxicol Pathol, December 1, 2008; 36(7_suppl): 67S - 100S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Song, T. Rutherford, F. Naftolin, S. Brown, and G. Mor Hormonal regulation of apoptosis and the Fas and Fas ligand system in human endometrial cells Mol. Hum. Reprod., May 1, 2002; 8(5): 447 - 455. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. R. Nayak and R. M. Brenner Vascular Proliferation and Vascular Endothelial Growth Factor Expression in the Rhesus Macaque Endometrium J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1845 - 1855. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
