Uterine Growth, Cell Proliferation, and c-fos Proto-Oncogene ...

BIOLOGY OF REPRODUCTION 56, 393-401 (1997)

Uterine Growth, Cell Proliferation, and c-fos Proto-Oncogene Expression Throughout the Estrous Cycle in Ewes' Mary Lynn Johnson,3 Dale A. Redmer, and Lawrence P. Reynolds 2 Department of Animal and Range Sciences, North Dakota State University, Fargo, North Dakota 58105

Uterine growth, cell proliferation, and endometrial expression of the c-fos proto-oncogene were evaluated for nonpregnant ewes (n = 6 ewes per day) on Days 0 (estrus), 2, 4, 8, 12, and 15 of the estrous cycle. Fresh and dry weights of uterine horns decreased (p < 0.01) linearly from Days 0 to 8 and then remained similar from Day 8 through Day 15. Tissue water content (1 - [dry weight/fresh weight]) was greater on Day 0 (p < 0.01) than on Days 2, 4, and 8, and the latter values were greater (p < 0.01) than on Days 12 and 15. DNA content was similar on Days 0, 2, 4, 8, and 15 but increased (p < 0.01) on Day 12. Although DNA content was greatest on Day 12, the ratios of RNA to DNA and of protein to DNA were least (p < 0.01) on that day. Thus, changes in uterine fresh weight were associated primarily with changes in tissue hypertrophy (RNA:DNA and protein:DNA ratios) and water content. In addition, changes in uterine fresh weight were associated with changes in the ratio of estradiol to progesterone in systemic blood, which was greatest (p < 0.01) on Days 0 and 2, decreased (p < 0.01) from Day 2 to Day 8, remained low through Day 12, and then was elevated again (p < 0.01) on Day 15. Moreover, compartmentalized changes in endometrial cell proliferation (labeling index [LI]; percentage of cells exhibiting nuclear incorporation of bromodeoxyuridine, a thymidine analogue) also were associated with the changes in fresh weight. The epithelium of the uterine lumen and luminal glands exhibited the greatest changes in rate of cell proliferation and accounted for most of the changes in LI seen across days of the estrous cycle. Endometrial expression of c-fos mRNA and protein also reflected changes in uterine weight and the systemic estradiol:progesterone ratio. The level of c-fos mRNA was greatest at estrus, low on Days 2-8, and elevated slightly on Days 12 and 15; Fos protein was greatest on Days 0 through 4, least on Day 8, and intermediate on Days 12 and 15. Characterization of uterine growth, cell proliferation, and c-fos expression throughout the estrous cycle will provide a foundation for future studies of gene expression regulating growth of the nonpregnant uterus.

uterotropic effects of estrogens and inhibits estrogen-induced uterine growth [4]. Growth and differentiation of normal tissues is thought to be regulated by proto-oncogenes such as c-fos, which is an "early response" gene and transcription factor [5-7]. In a variety of systems, including the rodent uterus, induction of cell proliferation by steroids and growth factors is associated with increased c-fos expression [8-16]. In ewes, the increased cell proliferation that accompanies uterine growth and vascular development during early pregnancy is reflected by an increase in endometrial expression of c-fos mRNA [17-19]. Thus, expression of c-fos in the nonpregnant uterus might be expected to reflect the cyclic changes in uterine cell growth and differentiation. Growth of the uterus during early pregnancy has been quantified in sows and ewes [18-20]. Uterine growth during the estrous cycle has been described histologically in cows and ewes [21, 22]. However, we are not aware of any studies that quantitatively evaluated the patterns of uterine growth and regression and the accompanying changes in systemic concentrations of ovarian steroids in cyclic ewes. We hypothesized that uterine growth during the estrous cycle is controlled by changes in the physiological concentrations of estradiol and progesterone and that the growthpromoting effects of steroids are mediated through increased uterine expression of c-fos. Thus, the purposes of this study were to 1) quantify uterine growth at the gross and cellular levels, 2) evaluate endometrial expression of the c-fos proto-oncogene, and 3) evaluate the relationships between systemic concentrations of estradiol and progesterone and relate them to patterns of growth and c-fos expression throughout the estrous cycle in ewes. Ewes were used in this study because publications on ovine uterine function are numerous, ovine uterine tissue is abundant, and the use of sheep as experimental animals is relatively common, particularly in biomedical research.

INTRODUCTION

MATERIALS AND METHODS

Markee [1, 2] showed that when endometrial explants from rabbits or monkeys were transplanted to the anterior chamber of the eye, they recruited a vascular supply and underwent cyclic periods of growth and regression that could be regulated by exogenous steroids. Similarly, Abel [3] observed vascularization and cyclic growth patterns in human endometrial explants transplanted to the hamster cheek pouch. The uterotropic effects of estrogens, which stimulate endometrial hyperplasia and hypertrophy, are well known [4]. In many species, progesterone antagonizes these

Animals and Sample Collection

ABSTRACT

Ewes were killed on Days 0 (estrus), 2, 4, 8, 12, and 15 after estrus (n = 6 ewes per day). The average length of the estrous cycle in this group of ewes, as determined by twice-daily exposure to vasectomized rams, was 16.5 days, and all ewes had exhibited at least one estrous cycle of normal duration (15-18 days) before being used for this study [23]. A sample of jugular venous blood was collected from each ewe immediately before slaughter, and the plasma was frozen (-20°C) until steroid concentrations were determined by RIA. An i.v. injection of bromodeoxyuridine (BrdU, 5 mg/kg BW; Aldrich, Milwaukee, WI), a thymidine analogue, was given 1 h before animals were killed for evaluation of the rate of cell proliferation [19, 23]. Because BrdU is a carcinogen and a teratogen, carcasses were disposed of by incineration. With use of procedures similar to those reported by

Accepted September 19, 1996. Received May 24, 1996. 'Supported in part by NIH grant HD22559 to L.P.R. and D.A.R. A publication of the North Dakota Agric. Exp. Sta., Projects 1782 and 1795. 2 Correspondence. FAX: (701) 231-7590. 3 Current address: Cell Biology Center, Hultz Hall 119b, North Dakota State University, Fargo, ND 58105.

393

394

JOHNSON ET AL.

Reynolds and Redmer [18], the ovaries, oviducts, cervix, and mesometrium were removed, and each uterus was then bisected between the uterine horns from the external bifurcation through the uterine body. Each uterine horn (including approximately one half of the uterine body) was incised longitudinally along the antimesometrial border; the luminal surface was rinsed gently with PBS (0.01 M sodium phosphate, 0.14 M NaCl, pH 7.3) to remove uterine fluids; and the fresh weight was determined. A portion of each horn was reweighed and freeze-dried to determine the dry weight:fresh weight ratio [18]. Cross sections (0.5 cm wide) from the midportion of the uterine horn adjacent to the luteal/follicular ovary were fixed in Carnoy's solution for histological evaluation of BrdU incorporation [18, 23]. Samples (-1 g each) of endometrium and myometrium were snap frozen in liquid nitrogen for later homogenization and determination of DNA, RNA, and protein concentrations. Snap-frozen endometrial samples also were used for Northern and Western analyses of c-fos gene expression. Uterine DNA, RNA, and Protein Concentrations To assess cellular growth of uterine tissues [18, 24], samples of endometrium and myometrium were homogenized with use of a Polytron (Brinkmann, Westbury, NY) in 10 volumes of TNE buffer (50 mM Tris, 2.0 M NaCl, 2 mM EDTA, pH 7.4) containing 0.02% (w:v) NaN 3. The diphenylamine procedure was used to determine the DNA concentrations, the orcinol procedure was used to determine RNA concentrations [18, 24], and the Coomassie brilliant blue G (Bradford) assay was used for protein concentrations [25] of homogenates. Standards were DNA type I from calf thymus, RNA type V from calf liver, and BSA fraction V, all from Sigma Chemical Company (St. Louis, MO). Whole uterine tissue concentrations of DNA were estimated as the average of endometrial plus myometrial DNA concentrations and were used to determine uterine DNA content. Uterine DNA content was calculated by multiplying uterine weight by whole uterine DNA concentration. DNA content was used as an index of hyperplasia, and RNA:DNA and protein:DNA ratios were used as indexes of hypertrophy [18, 24]. All ratios were first calculated within an animal and then subjected to statistical analyses. Endometrial Labeling Index Uterine cross sections that had been fixed in Carnoy's were embedded in paraffin, sectioned at 6 pxm, and mounted onto positively charged glass slides (ProbeOn Plus; Fisher Scientific, Pittsburgh, PA). BrdU was immunolocalized in these tissue sections as described previously [19, 23, 26]. Briefly, after a 30-min treatment with 2 N HC1 to denature nuclear chromatin, tissue sections were treated for 30 min with blocking buffer (PBS + 0.3% [v:v] Triton X-100 [Mallinckrodt, Paris, KY] + 1% [v:v] normal horse serum [Vector Labs., Burlingame, CA]). Sections were incubated for 1 h at 22°C with mouse anti-BrdU monoclonal antibody (1:100 [1 g/ml] in blocking buffer; Boehringer Mannheim, Indianapolis, IN). Control sections were incubated with mouse ascites fluid (ICN Biochemicals, Costa Mesa, CA) in place of the primary antibody. Detection of primary antibody was accomplished with the use of a biotinylated secondary antibody (horse anti-mouse IgG) and the avidinbiotinylated peroxidase complex system (Vectastain; Vector Labs.). Tissue sections were counterstained briefly with Harris hematoxylin. Stained sections were photographed at

a magnification of x200 using Kodak TMax-100 (Eastman Kodak, Rochester, NY) half-tone film. Cell proliferation (BrdU incorporation) was evaluated quantitatively by determining the BrdU labeling index (LI; percentage of total cells that were labeled for BrdU) as we have described previously [19, 23, 26]. To determine the LI of caruncular (CAR) and intercaruncular (ICAR) endometrium of each uterine horn, three half-tone prints (20 25 cm) were made for each tissue (i.e., three prints each for CAR, ICAR, and myometrium of each uterine horn). For luminal epithelium of CAR and ICAR, all unlabeled and BrdU-labeled cells were counted for the entire epithelium of each print. For glandular epithelium of ICAR, stroma of CAR and ICAR, and myometrium, all unlabeled and BrdUlabeled cells were counted within two randomly chosen 58-cm 2 areas of each print. In addition, for CAR and ICAR stroma and ICAR glands, luminal (0-360 pxm [-20 cell diameters] away from the uterine lumen) and deep (3601080 ,um from the uterine lumen) regions were evaluated [18]. For each cell type within each tissue and region of each ewe, the LI was based upon evaluation of -30005000 nuclei. Northern Analysis of Endometrial c-fos mRNA Northern analysis and densitometry of c-fos mRNA were conducted as we have described previously [19, 27]. Snapfrozen endometrial tissue samples were homogenized with a Polytron in cold (4°C) guanidinium-isothiocyanate buffer (-1 g tissue/8 ml buffer [28]). Total cellular (tc) RNA was isolated by CsCl density gradient centrifugation and quantified by absorbance at 260 nm [28]. Aliquots of tcRNA were pooled within each day of the estrous cycle (n = 6 ewes per day), and poly(A)-enriched RNA was selected by using the Mini-Oligo(dT) Cellulose Spin Column kit (5 Prime--3 Prime, Inc., Boulder, CO) according to the manufacturer's directions. For analysis, 10 [Lg of poly(A) RNA was electrophoresed on a 1.2% agarose, 2.2 M formaldehyde gel, electroblotted (TE50; Hoefer Scientific Instruments, San Francisco, CA) to a Zeta-Probe nylon membrane (Bio-Rad, Hercules, CA), and baked for 2 h at 80°C. Hybridization to the Pst ISst I insert of v-fos (American Type Culture Collection #41040; Rockville, MD), which was random-primer labeled (Bethesda Research Laboratories, Gaithersburg, MD) with 32 P-dATP (New England Nuclear, Boston, MA), was at -42 0 C for 16 h in 50% formamide, 1% SDS, 1 M NaCl, 10% dextran sulfate. Membranes were washed twice for 20 min each in double-strength SSC (single-strength SSC = 0.15 M NaCl, 0.015 M sodium citrate) at room temperature (rt), twice for 15 min each in 0.5-strength SSC with 0.25% SDS at 65 0C, and once in 0.1-strength SSC with 0.1% SDS for 1 h at rt. To detect hybridization bands, membranes were subjected to autoradiography (Fuji film; Fisher Scientific, Itasca, IL) at -70°C for 18 h. Densitometry of the autoradiographs was performed with a soft laser scanning densitometer (Model SLR-2D/1D; Biomed Instruments, Fullerton, CA). Hybridization to a biotin-labeled oligo(dT) 30-mer probe (R & D Systems, Minneapolis, MN), followed by autoradiography using chemiluminescence detection (ECL; Amersham Life Science, Arlington Heights, IL) and densitometry, was used to verify equal lane-to-lane loading and transfer of the poly(A) RNA to the membrane, as we have described and validated previously [27].

395

UTERINE GROWTH IN CYCLIC EWES Western Immunoblot Analysis of Endometrial Fos Protein

TABLE 1. Plasma estradiol-173 and progesterone concentrations throughout the estrous cycle in ewes.*

Procedures for Western analysis were similar to those we have already reported [25]. As for Northern analysis, frozen tissues were pooled within each treatment group and then were homogenized in ice-cold PBS buffer containing 1% (w:v) cholate + 0.1% (v:v) SDS (0.2 mg tissue/ml buffer). The pooled homogenates were sonicated for 30 sec and centrifuged briefly to remove cellular debris. Clarified homogenates were assayed for protein using the Coomassie brilliant blue G assay [25] with BSA fraction V (Sigma) as the standard. For analysis, 10 g of protein from each pooled sample was separated by using 12% SDS-PAGE [29], electroblotted to an Immobilon P membrane (Millipore, Bedford, MA), and immunoblotted. For immunoblotting, the Immobilon P membrane was wetted in methanol, rinsed five times in dH 20, and blocked for 1 h at rt in PBST (0.02 M sodium phosphate monobasic [NaH 2PO 4], 0.08 M sodium phosphate dibasic [Na 2HPO 4 ], 0.1 M NaCl [pH 7.5] + 0.1% Tween 20 [Amresco, Solon, OH]) containing 5% skim milk (Difco Laboratories, Detroit, MI). After two rinses of 5 min each in PBST, the membrane was incubated for 1 h at rt with a rabbit polyclonal antibody against Fos (#sc-52; Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1:1000 (0.1 ng/ml) in PBST containing 1% skim milk. The membrane was then rinsed three times (15, 5, and 5 min) in PBST at rt and incubated for 1 h at rt with a horseradish peroxidase-labeled anti-rabbit secondary antibody (Amersham) diluted 1:2000 in PBST containing 1% skim milk. After three washes in PBST and a final rinse in PBS without Tween 20, bound secondary antibody was detected by using the ECL solutions and protocol supplied by Amersham. Fuji film was used for autoradiography, and exposure was for 6 min at rt. Protein binding was quantified with densitometry [27]. To verify that the densitometric analysis was not saturated and could detect changes in Fos protein, various amounts (5-40 Ig protein per lane) of a pooled sample of endometrial protein (n = 6 ewes per day, 100 ,ag/ewe) were electrophoresed, immunoblotted, and analyzed. Specificity of antibody binding was confirmed by immunoadsorption of the primary antibody with Fos peptide (#sc-52P; Santa Cruz), followed by immunoblotting.

Progesterone (ng/ml)

Estradiol (pg/ml)

Day

2.6 2.4 2.8 2.2 2.9 3.9

0 2 4 8 12 15

+ ± ± ± + -

a

0.3 0.4 1.3 3.8 3.9 1.2

0.2 0.2a 0.2a 0.4a 0.4"a b b 1.2

Estradiol: progesterone (pg:ng) 11.1 8.7 2.5 0.6 0.6 5.4

+ 0.1a a + 0.1 0

.1 b

+ 0.5' c + 0.5 b + 0.5

a

+ 2.2 - 1.5a 0

.5b

5 0.1 + 0 .2 b ± 1.3

* Means - SEM for n = 6 ewes per day. a.b,c Within a column, means with different superscripts differ (p < 0.10 for estradiol, p < 0.01 for progesterone, and p < 0.01 for estradiol: progesterone ratio).

gesterone and 11.4% for estradiol. Assay low limit of detection was 0.2 ng/ml for progesterone and 1 pg/ml for estradiol. Statistical Analyses Data were examined statistically by using least-squares (General Linear Models) analysis of variance with day and, when appropriate, tissue type and day x tissue type interaction included in the model [33]. When an F-test was significant, differences between specific means were evaluated by using Bonferroni's t-test [34]. RESULTS Plasma Steroids Plasma concentrations of estradiol were greatest (p < 0.10) on Day 15 of the estrous cycle and were similar on Days 0, 2, 4, and 8; Day 12 values were not different from those for any of the other days (Table 1). Progesterone concentrations were least (p < 0.01) on Days 0 and 2, were increased (p < 0.01) threefold by Day 4, and were greatest on Days 8 and 12 (Table 1). Progesterone concentrations on Day 15 were less (p < 0.01) than on Days 8 and 12 and similar to those on Day 4 (Table 1). The ratio of estradiol to progesterone was greatest (p < 0.01) on Days 0 and 2; it decreased (p < 0.01) from Day 2 through Day 8 and began increasing (p < 0.01) again on Day 15 (Table 1).

Plasma Steroids Progesterone was extracted from plasma (200 il) with benzene-hexane, and estradiol was extracted from plasma (1 ml) with benzene; plasma steroid concentrations were determined by RIA as we have previously reported [3032]. For each steroid, a single assay was used for all samples; intraassay coefficients of variation were 5.8% for pro-

Uterine Growth and Cell Proliferation Fresh and dry weights of uterine horns decreased (p < 0.01) linearly from Days 0 to 8 and then remained constant on Day 12 and Day 15 of the estrous cycle (Table 2). The ratio of dry weight to fresh weight was less (p < 0.01) on Day 0 than on Days 2, 4, and 8, and on these days the

TABLE 2. Uterine fresh and dry weights, DNA contents, and RNA:DNA and protein:DNA ratios throughout the estrous cycle in ewes.* Day 0 2 4 8 12 15

Fresh wt. (g)+ 45.6 40.4 38.2 32.2 34.1 34.3 +

+ 1.7 + 2.2 + 2.0 + 4.1 + 3.3 + 2.5

t Dry wt. (g)

6.9 6.4 6.0 5.1 5.6 5.7

+ 0.4 + 0.3 + 0.3 + 0.6 + 0.5 + 0.4

Dry wt: fresh wt. 0.148 0.159 0.158 0.159 0.164 0.168

+ + + + + +

0.005a b 0.001 0.003 b 0.001 b 0.001 c 0.003'

DNA (mg) 103 125 114 108 215 129

+ 8' + 16a + 13' + 28 a ± 19b

- 17a

RNA:DNA 1.35 1.10 1.16 0.96 0.46 0.66

+ + + +

0.11' 0.07' a 0.06 0.08a ± 0.01b - 0.06 b

Protein:DNA 18.6 + 17.8 ± 20.8 17.1 + 10.5 13.8 +

1.5' 0.9 1.4a 1.3' 0.5b 1.3"

SEM for n = 6 ewes per day. * Means t Fresh wt. and dry wt. decreased (p < 0.01) linearly from Day 0 through Day 8 and then remained constant on Day 12 and Day 15. a,b,cWithin a column, means with different superscripts differ (p < 0.01).

396

JOHNSON ET AL.

FIG. 1. Immunohistochemical localization of BrdU in sections of endometrial (A)CAR or (B)ICAR tissues of ewes throughout the estrous cycle. Day of the estrous cycle is indicated in the lower right of each photomicrograph. e = luminal epithelium, s = stroma, and g = glandular epithelium. Some of the BrdU-labeled nuclei are indicated with arrowheads. Bars = 50 ptm.

ratios were less (p < 0.01) than on Days 12 and 15 (Table 2), thus indicating that tissue water content (1 - [dry weight/fresh weight]) was greatest on Day 0, intermediate on Days 2, 4, and 8, and least on Days 12 and 15 of the estrous cycle.

Total endometrial DNA content was greater (p < 0.01) on Day 12 than on any other day (Table 2). The ratios of RNA to DNA and protein to DNA, however, were least (p < 0.01) on Days 12 and 15 and similar on Days 0, 2, 4, and 8 of the estrous cycle (Table 2).

UTERINE GROWTH IN CYCLIC EWES

397

FIG. 1. Continued

The histological patterns of endometrial cell proliferation (BrdU labeling) throughout the estrous cycle are shown in Figure 1. For CAR (Fig. 1A), cell proliferation of luminal epithelium appeared to be relatively high on Days 2 and 4 but reduced on the other days. In addition, for CAR, cell proliferation of luminal stroma seemed greatest on Days 0 and 4 and less on the other days (Fig. 1A), whereas cell

proliferation of deep stroma (not shown) appeared low throughout the estrous cycle. For ICAR, cell proliferation appeared to be relatively high in luminal epithelium on Days 2 and 4, in luminal glands on Days 0 and 2, and in luminal stroma on Days 0 and 4 (Fig. 1B). For deep glands (not shown), cell proliferation appeared greatest on Day 4 and relatively low on the other days.

398

JOHNSON ET AL. 4I: r

.

A SE = 0.96

- 12 X o 10 C

i

-

Caruncular

14

F

l

Lum na Epithelium Luminal Stroma

D Deep

Stroma

8

rg 6 .' 4 2 0

0

2

4

8

12

15

Day of the Estrous Cycle Luminal E x X

c

Epithelium Luminal Stroma Luminal Glands Deep Glands

.C -0

Cu -j U

2

4

8

12

15

Day of the Estrous Cycle FIG. 2. LI of endometrial (A) CAR and (B)ICAR tissues of ewes throughout the estrous cycle. A) For CAR, LI for luminal epithelium was greatest on Days 2 and 4 (p < 0.01); for luminal stroma, LI was greatest on Days

0 and 4 (p < 0.01). B) For ICAR, LI of luminal epithelium was greatest on Day 2 (p < 0.01); LI of luminal stroma was greatest on Days 0 and 4 (p < 0.01). For luminal glands of ICAR, LI was greatest on Day 0 and decreased linearly from Day 0 to Day 8 (p < 0.01).

The LI, which is a quantitative estimate of the rate of cell proliferation, was similar to the histological observations for both CAR and ICAR tissues (Fig. 2). Thus for CAR (Fig. 2A), the luminal epithelium exhibited a high (12.72 1.42%; p < 0.01) LI on Days 2 and 4 but even on Days 8-15 exhibited a LI of > 1%; in contrast, the LI of luminal stroma of CAR was greatest (p < 0.01) on Days 0 and 4, and averaged 2.34 + 0.24%. The LI of deep stroma of CAR was low (< 1%) throughout the estrous cycle (Fig. 2A). For ICAR (Fig. 2B), the LI of luminal epithelium was greatest (17.57 + 2.17%; p < 0.01) on Day 2, intermediate (2.61 0.65%) on Days 0 and 4, and low (< 1%) on Days 8 through 15. The LI of luminal stroma of ICAR was greatest (2.10 + 0.49%; p < 0.01) on Days 0 and 4 and relatively low (< 1%) on Days 2, 8, 12, and 15 (Fig. 2B). For luminal glands of ICAR, the LI was relatively high (10.18 ± 3.25%) on Day 0, decreased linearly (p < 0.01) from Day 0 to Day 8, and then remained low (< 1%) through Day 15. For deep glands, the LI was greatest (3.58 1.14%) on Day 4, intermediate (1.20 0.28%) on Day 8, and low (< 1%) on the other days. Across all CAR and ICAR tissue compartments, endometrial LI was twofold greater (p < 0.01) on Day 2 (5.45 ± 0.84%) than on Days 0 (2.57 0.60%) or 4 (3.83 + 0.60%). In addition, the overall endometrial LI was greater (p < 0.01) on Days 0 and 4 than on Day 8 (0.44 0.11), Day 12 (0.27 0.09), or Day 15 (0.49 0.15). Across all days and tissue compartments, the LI of luminal areas of the endometrium (2.81 0.30%) was greater (p < 0.01) than for deep areas (0.63 + 0.13%).

FIG. 3. A) Northern analysis of c-fos mRNA in endometrial tissues of ewes throughout the estrous cycle. B) Densitometric evaluation of the Northern analysis of endometrial c-fos mRNA levels. In A, the 32P-labeled probe was hybridized to 10-[xg samples of endometrial poly(A) RNA pooled within each day (n = 1 pooled sample of 6 ewes per day). Lanes are as follows: 1, Day 0 (estrus); 2, Day 2; 3, Day 4; 4, Day 8; 5, Day 12; and 6, Day 15 after estrus. In B, area is expressed as arbitrary optical density units.

Endometrial Expression of c-fos mRNA and Protein Endometrial c-fos mRNA exhibited two bands: one very prominent band at -2.3 kilobases (kb) and one of reduced intensity at -1.9 kb (Fig. 3A). These sizes have been reported for c-fos mRNA expression in the fetal liver and for the ovine uterus during early pregnancy [9, 17, 191. Endometrial c-fos mRNA levels were greatest at estrus (Day 0), low on Days 2 through 8, and elevated slightly on Days 12 and 15 (Fig. 3, A and B). The levels of endometrial Fos protein were greatest on Days 0, 2, and 4, were least on Day 8, and were intermediate on Days 12 and 15 (Fig. 4, A and B). Specificity of the Fos antibody as confirmed by preadsorption with peptide and subsequent immunoblotting is shown in Figure 4C. The protein band at -62 kDa was no longer present when immunoblotting was performed with the preadsorbed antibody, indicating that the band at -62 kDa was the specific Fos protein band. In addition, dilutions of a pooled protein sample from ewes at estrus (Day 0) confirmed that the densitometric analysis could detect linear differences in c-fos expression in protein samples ranging from 10 through 40 Ixg (data not shown).


FIG. 4. A) Western analysis of Fos protein in endometrial tissues of ewes throughout the estrous cycle. B) Densitometric evaluation of the Western analysis of endometrial Fos protein levels. C) Verification of antibody specificity by preabsorption of Fos primary antibody. In A, each lane contained 10 pIgof protein pooled within each day (n = 1 pooled sample of 6 ewes per day). Lanes are as follows: 1, Day 0 (estrus); 2, Day 2; 3, Day 4; 4, Day 8; 5, Day 12; and 6, Day 15 after estrus. In B, area is expressed as arbitrary optical density units. In C, note the absence of the -62-kDa protein band (compare with A); lane 1 contains 100 ~g of protein pooled from ewes (n = 9) at estrus, and lane 2 contains 100 jg of protein pooled from ewes (n = 9) at the luteal (Day 10) stage of the estrous cycle.

DISCUSSION The uterus provides an environment in which growth and nourishment of offspring can occur [35]. In the nonpregnant uterus, cyclic changes in growth and cell proliferation occur in preparation for fertilization and successful implantation of the conceptus. Although the ovarian steroids estradiol-17 and progesterone can influence the pattern of uterine growth, exactly how these effects are mediated is still poorly understood [4, 36]. Ovine uterine growth has been described quantitatively during early pregnancy [18, 20] and descriptively during the estrous cycle

399

[21, 22]. Quantitative evaluation of uterine growth during the estrous cycle, as in the present study, will provide a foundation for understanding how this cyclic pattern of growth is controlled. Other researchers have shown that plasma concentrations of ovarian steroids fluctuate throughout the estrous cycle in ewes [37-39]. We found that changes in the levels of estradiol were less dramatic than changes in the levels of progesterone throughout the estrous cycle. But we feel the changes in estradiol shown herein, although different only at a value of p < 0.10, are an indication of a biologically important increase in estradiol per se immediately preceding estrus. This is further confirmed by our observations, in approximately 20-30 long-term (> 4 wk)-ovariectomized ewes, that estradiol replacement achieving peripheral plasma concentrations similar to those observed for estrous animals in the present study results in standing estrus in all animals within 12-24 h after estradiol treatment (unpublished results). Thus, the ewe's entire reproductive system appears to be exquisitely sensitive to small changes in systemic estradiol concentrations. Moreover, the ratio of estradiol to progesterone in plasma most closely reflects the pattern of uterine growth and demonstrates the most dramatic change across the estrous cycle. Thus, although this ratio may be an artificial tool (based on the 1000-fold difference in actual concentrations), we feel that it expresses a biologically important relationship in the ewe. Progesterone effects on uterine growth have previously been thought to be antagonistic to estrogen, but more likely both steroids have modifying effects on each other and may even have synergistic growthpromoting effects in the uterus [4, 36]. Perhaps when progesterone is dominant (reflected by the lowest estradiol: progesterone ratio), there are fewer available estradiol receptors and the growth-stimulatory effects of estradiol are inhibited by the loss of receptors [4]. Then, when the ratio of estradiol to progesterone is greatest (at estrus), these modifying effects of progesterone are overcome. These effects are likely regulated at the level of gene transcription, where progesterone could stop cell cycle progression by production of a specific, unknown factor [4, 36]. We know that the estrogen-receptor (ER) complex binds to an estrogen-response element (ERE) within a gene such as the c-fos gene to promote cell proliferation. Perhaps the progesterone-receptor complex activates a different gene and produces an ERE-binding protein that interferes with ER binding and production of c-fos mRNA [36]. In the present study, uterine fresh and dry weights were greatest at estrus, being approximately 40% greater than at midcycle. At estrus, uterine tissue water content is thought to increase, and the present data confirmed this because the dry weight:fresh weight ratio was least on Day 0. However, changes in uterine weight throughout the estrous cycle were primarily due to changes in tissue mass, because fresh and dry weights fluctuated by 30-40% whereas the dry weight: wet weight ratio changed by only -10%. DNA content of the tissue (hyperplasia) indicated that cell numbers were greatest on Day 12 when uterine weight was least. However, the RNA:DNA and protein:DNA ratios (hypertrophy) indicated that overall cell size also was least on Day 12 and, in fact, was threefold less than on Day 0. Thus, during the estrous cycle, cellular hypertrophy seems to have a greater influence on uterine size than does cellular hyperplasia. The increase in cell numbers on Day 12 may be due to the effects of progesterone, since progesterone

400

JOHNSON ET AL.

can stimulate proliferation of uterine epithelial cells [25, 36]. The LI provides an index of the rate of cell proliferation within a specific tissue at any given time [40]. For endometrial tissues, LI was greatest from Days 0 to 4, but least on Days 12-15, even though the total number of cells apparently peaked on Day 12. Thus, accumulation or destruction of cells in the endometrium seems to be a complex process that balances cell proliferation against cell death or apoptosis [41]. In addition, changes in myometrial cell numbers and cell proliferation, which were not determined in the present study, may have a dramatic effect on total uterine DNA content. Comparing the patterns of estradiol and progesterone, and their ratio with endometrial LI, the patterns suggest that in ewes, progesterone may be the primary controlling factor for uterine cell proliferation. Thus, when progesterone levels were least (Days 0-4), endometrial LI was greatest. When progesterone levels were elevated on Days 8 and 12, concomitant with the development of the mature corpus luteum, endometrial LI remained low. When progesterone levels began to decrease on Day 15, endometrial LI was again beginning to increase, at least in CAR tissues. In a previous study, exogenous estrogen stimulated cell proliferation in all uterine tissue compartments of ovariectomized ewes [26]. In contrast, progesterone alone stimulated proliferation only in luminal epithelium of CAR and ICAR and inhibited estrogen-stimulated cell proliferation in the other tissue compartments. Thus, the effects of steroids on cell proliferation in the ovine uterus appear to be complex and require further study. Luminal and deep areas of the endometrium exhibited dramatically different patterns and levels of cell proliferation. The luminal areas exhibited the most dramatic changes in LI, whereas changes in the deep areas were less dramatic and were observed primarily in the deep glands on Day 4. These data agree with our observations of the patterns of cell proliferation of the ovine uterus during early pregnancy and in ovariectomized steroid-treated ewes [19, 26]. Thus, the luminal compartments in the sheep may be analogous to the superficial, or functionalis, region of the primate endometrium, which exhibits the greatest rates of cell proliferation several days before and after ovulation [42]. The deep endometrial region in sheep may be analogous to the deep endometrial basalis of primates, which proliferates most dramatically during the early- to mid-luteal phase of the menstrual cycle [42]. The relatively low level of expression of c-fos mRNA during the estrous cycle made the use of pooled samples and selection of poly(A) RNA necessary for detection of changes across samples [27]. But even though these data are preliminary in nature, we feel they are important indicators of changes in c-fos expression during ovine uterine growth that warrant further investigation. In the uterus of mice and humans, the c-fos gene has an ERE and thus is directly stimulated by estrogens, as mentioned above [12, 43]. Endometrial expression of c-fos mRNA increased in gilts on Day 12 of pregnancy, when the conceptus transiently secretes estradiol [44]. In the cycling ewes in the present study, expression of c-fos mRNA was greatest at estrus (Day 0), when the estradiol:progesterone ratio also was the greatest, with much lower c-fos expression throughout the remainder of the estrous cycle. This may indicate that increased endometrial c-fos mRNA in cyclic ewes also resulted from direct stimulation by estradiol. However, since endometrial c-fos mRNA increased

with little change in plasma estradiol concentration, increased endometrial c-fos mRNA may actually result from removing the inhibitory influence of progesterone. Alternatively, local increases in endometrial estradiol or progesterone may be important [45]. The levels of c-fos mRNA were shown to increase dramatically on Day 18 of early pregnancy in ewes, a time when uterine estradiol content and cell proliferation were increasing [17-19, 46]. The increase in uterine LI and c-fos expression on Day 18 of early pregnancy preceded the increase in weight of gravid uterine horns, which was not evident until Day 24 of pregnancy. In the present study, c-fos expression also preceded, by 2 days, the maximal LI of the endometrium, suggesting that increased expression of c-fos mRNA precedes, and is probably involved in promoting, increased cell proliferation. However, these data should be interpreted with caution because c-fos also responds to a variety of growth- and differentiation-promoting stimuli [47, 48]. Levels of Fos protein in whole endometrium varied less across days than did levels of c-fos mRNA. The reason for this is not clear. These differences must be resolved by in situ hybridizations and immunocytochemical studies so that the numbers and types of cells expressing c-fos mRNA and protein can be determined for the ovine uterus, as has been reported for the rat and mouse [49, 50]. In summary, in the present study we quantified growth, cell proliferation, and c-fos expression in the ovine endometrium throughout the estrous cycle. Changes in the patterns of uterine growth at the gross and cellular level and in endometrial expression of c-fos mRNA and protein appear to be related to systemic levels of estradiol and progesterone. This study has laid the groundwork for more definitive studies on steroidal control of cell proliferation and gene expression in the ovine uterus. It is important for us to understand how cell proliferation is controlled in steroid-responsive normal adult tissues such as the uterus, since this knowledge may eventually contribute to improved reproductive performance. In addition, understanding how steroids influence normal cell growth may help determine the role that steroids play during pathological cell growth. ACKNOWLEDGMENTS We thank Dr. A.J. Conley, J. Corbin, J.D. Kirsch, and K.C. Kraft for expert technical assistance, Dr. M.A. Sheridan for use of the densitometer, and J. Berg for clerical assistance.

REFERENCES 1. Markee JE. An analysis of the rhythmic vascular changes in the uterus of the rabbit. Am J Physiol 1932; 100:374-383. 2. Markee JE. Menstruation in intraocular endometrial transplants in the rhesus monkey. Contrib Embryol 1940; 28:221-308. 3. Abel MH. Prostanoids and menstruation. In: Baird DT, Michie EA (eds.), Mechanisms of Menstrual Bleeding. New York: Raven Press; Serono Symposia 1985; 25:139-156. 4. Clark JH, Markaverich BM. Actions of ovarian steroid hormones. In: Knobil E, Neill JD (eds.), The Physiology of Reproduction. New York: Raven Press; 1988: 675-724. 5. Kaczmarek L. Protooncogene expression during the cell cycle. Lab Invest 1986; 54:365-376. 6. Travali S, Koniecki J, Petralia S, Baserga R. Oncogenes in growth and development. FASEB J 1990; 4:3209-3214. 7. Distel RJ, Spiegelman BM. Protooncogene c-fos as a transcription factor. Adv Cancer Res 1990; 55:37-55. 8. Kasik JW, Wan Y-JY, Ozato K. A burst of c-fos gene expression in the mouse occurs at birth. Mol Cell Biol 1987; 7:3349-3352. 9. Zhang X-K, Wang Z, Lee A, Huang A, Chiu JE Differential expres-


10.

11. 12. 13.

14. 15.

16.

17.

18.

19.

20. 21.

22.

23.

24.

25.

26.

27.

28.

sion of cellular oncogenes during rat liver development. Cancer Lett 1988; 41:147-155. Sheng M, Greenberg ME. The regulation and function of c-fos and other immediate early genes in the nervous system. Neuron 1990; 4: 477-485. Kaczmarek L, Kaminska B. Molecular biology of cell activation. Exp Cell Res 1989; 183:24-35. Hyder SM, Stancel GM. Presence of an estradiol response region in the mouse c-fos oncogene. Steroids 1991; 56:498-504. Cicatiello L, Sica V, Bresciani F, Weisz A. Identification of a specific pattern of "immediate-early" gene activation induced by estrogen during mitogenic stimulation of rat uterine cells. Receptor 1993; 3:1730. Cochran BH. Regulation of immediate early gene expression. NIDA Res Monogr 1993; 125:3-24. Khan SA, Nephew KP, Wang H. The jun protooncogene family in uterine growth and differentiation. In: Khan SA, Stancel GM (eds.), Protooncogenes and Growth Factors in Steroid Hormone Induced Growth and Differentiation. Boca Raton, FL: CRC Press; 1994: 105123. Stancel GM, Boettger-Tong H, Hyder SM, Kirkland JL, Loose-Mitchell DS. Uterine c-fos: a paradigm for control of gene expression during estrogen stimulated target tissue growth. In: Khan SA, Stancel GM (eds.), Protooncogenes and Growth Factors in Steroid Hormone Induced Growth and Differentiation. Boca Raton, FL: CRC Press; 1994: 87-104. Johnson ML, Reynolds LP Endometrial labeling index and c-fos proto-oncogene expression during early pregnancy in ewes. J Reprod Fertil Abstr Ser 1992; 9:69. Reynolds LP, Redmer DA. Growth and microvascular development of the uterus during early pregnancy in ewes. Biol Reprod 1992; 47: 698-708. Zheng J, Johnson ML, Redmer DA, Reynolds LP. Estrogen and progesterone receptors, cell proliferation, and c-fos expression in the ovine uterus during early pregnancy. Endocrinology 1996; 137:340348. Wu MC, Shin WJ, Dziuk PJ. Influence of pig embryos on uterine growth. J Anim Sci 1988; 66:1721-1726. Mckenzie FE Allen E, Guthrie MJ, Warbritton V, Terrill CE, Casida LE, Nahm LJ, Kennedy JW. Reproduction in the ewe-a preliminary report. Proc Am Soc Anim Prod 1933: 278-281. Eckstein P, Zuckerman S. Changes in the accessory reproductive organs of the non-pregnant female. In: Parkes AS (ed.), Marshall's Physiology of Reproduction, 3rd ed., vol. I. New York: Longmans Green; 1952: 543-654. Jablonka-Shariff A, Grazul-Bilska AT, Redmer DA, Reynolds LP Growth and cellular proliferation of ovine corpora lutea throughout the estrous cycle. Endocrinology 1993; 133:1871-1879. Reynolds LP, Millaway DS, Kirsch JD, Infeld JE, Redmer DA. Growth and in-vitro metabolism of placental tissues of cows from day 100 to day 250 of gestation. J Reprod Fertil 1990; 89:213-222. Grazul-Bilska AT, Redmer DA, Zheng J, Killilea SD, Reynolds SP Initial characterization of mitogenic activity of ovine corpora lutea from early pregnancy. Growth Factors 1995; 12:131-144. Johnson DA, Redmer DA, Reynolds LP Histological evaluation of cell proliferation in the uterus of intact and steroid-treated ovariectomized ewes. J Anim Sci 1994; 72(suppl 2):74. Johnson ML, Redmer DA, Reynolds LP Quantification of lane-to-lane loading of poly(A) RNA using an oligo(dT) probe and chemiluminescent detection. BioTechniques 1995; 19:712-715. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith

29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43.

44. 45. 46. 47. 48. 49. 50.

401

JA, Struhl K. Current Protocols in Molecular Biology. New York: John Wiley; 1989: 4.9.4-4.9.8. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227:680-685. Taraska T, Reynolds LP, Redmer DA. In vitro secretion of angiogenic activity by ovine follicles. In: Hirshfield AN (ed.), Growth Factors and the Ovary. New York: Plenum; 1989: 267-272. Grazul-Bilska AT, Redmer DA, Reynolds LP Secretion of angiogenic activity and progesterone by ovine luteal cell types in vitro. J Anim Sci 1991; 69:2099-2107. Redmer DA, Kirsch JD, Reynolds LP Production of mitogenic factors by cell types of bovine large estrogen-active and estrogen-inactive follicles. J Anim Sci 1991; 69:237-245. SAS. SAS User's Guide, Statistics, 5th ed. Cary, NC: Statistical Analysis System Institute, Inc.; 1985. Kirk RE. Experimental Design: Procedures for the Behavioral Sciences, 2nd ed. Belmont, CA: Brooks/Cole; 1982: 106-109. Hafez ESE. Anatomy of Female Reproduction. In: Hafez ESE (ed.), Reproduction in Farm Animals, chapter 3. Philadelphia, PA: Lea & Febiger; 1987: 35-64. Clarke CL, Sutherland RL. Progestin regulation of cellular proliferation. Endocr Rev 1990; 11:266-301. Moore NW, Barrett S, Brown JB, Schindler MA, Smith B. Oestrogen and progesterone content of ovarian vein blood of the ewe during oestrous cycle. J Endocrinol 1969; 44:55-62. Thorburn GD, Bassett JM, Smith ID. Progesterone concentration in the peripheral plasma of sheep during the oestrous cycle. J Endocrinol 1969; 45:459-469. Mattner PE, Braden AWH. Secretion of oestradiol-173 by the ovine ovary during the luteal phase of the oestrous cycle in relation to ovulation. J Reprod Fertil 1972; 28:136-137. Baserga R. The Biology of Cell Reproduction. Cambridge: Harvard University Press; 1985: 46-58. Brenner RM, Slayden OD. Cyclic changes in the primate oviduct and endometrium. In: Knobil E, Neill J (eds.), The Physiology of Reproduction, vol 1. New York: Raven Press; 1994: 541-569. Padykula HA. Regeneration in the primate uterus-the role of stem: cells. In: Wynn RM, Jollie WP (eds.), Biology of the Uterus. New York: Plenum Publishing; 1989: 279-288. Hyder SM, Stancel GM. In vitro interaction of uterine estrogen receptor with the estrogen response element present in the 3'-flanking region of the murine c-fos protooncogene. J Steroid Biochem Mol Biol 1994; 48:69-79. Dubois CH, Simmen RCM, Fliss A, Smith LC, Bazer FW. Expression of c-fos in porcine endometrium during the estrous cycle and early pregnancy. Biol Reprod 1993; 49:943-950. Pope WE Maurer RR, Stormshak E Distribution of progesterone in the uterus, broad ligament, and uterine arteries of beef cows. Anat Rec 1982; 203:245-250. Reynolds LP, Magness RR, Ford SP. Uterine blood flow during early pregnancy in ewes: interaction between the conceptus and the ovary bearing the corpus luteum. J Anim Sci 1984; 58:423-429. Rivera VM, Greenberg ME. Growth factor-induced gene expression: the ups and down of c-fos regulation. New Biol 1990; 2:751-758. Hyman SE, Kosofsky BE, Nguyen TV, Cohen BM, Comb MJ. Everything activates c-fos-how can it matter? NIDA Res Monogr 1993; 125:25-38. Papa M, Mezzogiorno V, Bresciani F, Weisz A. Estrogen induces c-fos expression specifically in the luminal and glandular epithelia of adult rat uterus. Biochem Biophys Res Commun 1991; 175:480-485. Baker DJ, Nagy E Nieder GL. Localization of c-fos-like proteins in the mouse endometrium during the peri-implantation period. Biol Reprod 1992; 47:492-501.