BIOLOGY OF REPRODUCTION 54, 249-254 (1996)
Role of Eosinophils in Uterine Responses to Estrogen'
Maria C. Perez, 3 Emma E. Furth,4 Pablo Damian Matzumura, 3 and C. Richard Lyttle 2 '3
Department of Obstetrics and Gynecology, Division of Reproductive Biology3
and the Department of Pathology,4 University of Pennsylvania School of Medicine
Philadelphia,Pennsylvania 19104
ABSTRACT
Administration of estradiol (E2)to ovariectomized mice results in a dramatic increase in uterine growth and an influx of eosinophilic
leukocytes. This influx is mediated by stimulation of an E2-dependent eosinophilic chemotactic factor inthe uterus (ECF-U). The role of
this eosinophil infiltration in uterus is presently unknown but could involve early growth and/or remodeling processes. In an attempt to
better define eosinophil function in uterine tissue, we produced ovariectomized mice severely depleted of circulating eosinophils by
administration of a purified rat IgG monoclonal antibody against interleukin-5 (IL-5). Seven days later, animals were submitted to estradiol
treatment. Experimental groups included mice treated with saline alone, saline followed by E2, IgG followed by E2, and anti-lL-5 followed
by E2. Pretreatment with IL-5 antibodies led to no significant alteration inE2-induced increase inuterine wet weight. However, histological
evaluation demonstrated a clear and almost complete blockade of E2-stimulated influx of eosinophils in anti-lL-5 treated animals. In
addition, IL-5 antibody administration significantly reduced E2-induced increase in peroxidase activity. Dramatic reduction of eosinophils
did not affect E2 stimulation of ECF-U activity by stromal cells or complement C3 synthesis by the epithelial cells. Thus, it appears that
differences in E2 responses ineosinophil-deficient mice are not directly associated with presence or absence of eosinophils. Taken together,
these data suggest that eosinophils most likely do not contribute to early growth in the E2-stimulated uterus. A possible role in other
events such as remodeling remains to be elucidated.
INTRODUCTION
The presence of eosinophils in the endometrium of rodents, specifically during the estrous cycle and after estrogen administration to immature animals, has been known
for many years [1-31. Although we have little knowledge
about either the mechanism responsible for eosinophil infiltration or the role that eosinophils play in uterine function, it is well documented that the number of eosinophils
present in uterine tissue of immature or late-pregnancy rats
is reduced compared to nonpregnant mature animals [1]. In
addition, uterine eosinophil infiltration varies during the estrous cycle with the largest number present in estrus [4].
Even though the function of eosinophils in uterus remains
obscure, data suggest that these cells are active elements in
the physiological and pathophysiological events occurring
in the uterus [5].
Exposure of immature rat uteri to estradiol has been reported to cause a variety of molecular events such as uterine
growth [6], synthesis and secretion of complement C3 (C3)
by epithelial cells [7, 8], increases in several growth factors,
and enhancement of uterine eosinophil infiltration by inducing production of an eosinophil chemotactic factor
(ECF-U) in stromal cells. However, the role that eosinophils
play in estrogen-dependent molecular events occurring in
uteri remains to be elucidated.
Several studies have shown that cytokines are responsible for the production, differentiation, and immunological
functions of mature eosinophils [9]. Interleukin-5 (IL-5) in
particular is a unipotential hemopoietin that stimulates
clonal proliferation and differentiation of eosinophilic precursors without affecting other myeloid or lymphoid clones
[9-11].
Thus, abolition of IL-5 actions results in a selective suppression of eosinophils, offering a means to evaluate the
eosinophils role in E2-induced uterine trophism. We have
evaluated the role of eosinophils in E actions in uteri by
analyzing the effects of the steroid on molecular markers
from mice depleted of eosinophils by passive immunization
against IL-5. This treatment produces an animal model that
is virtually devoid of eosinophilic white cells providing an
extremely useful means to evaluate this paradigm.
MATERIALS AND METHODS
Antibodies
Hybridomas producing rat monoclonal antibodies
(IgG1) specific for IL-5 (TRFK-5) were developed by Coffman et al. [12, 13] and were generously provided by Dr.
Phillip Scott (Veterinarian Medical School, University of
Pennsylvania, Philadelphia, PA).
Hybridoma cell lines were grown in RPMI 1640 medium
supplemented with 25 mM HEPES (Sigma Chemical Co., St.
Louis, MO), 1% Nutridoma SP (100X; Boehringer Mannheim Corp., Indianapolis, IN), 0.1% Fungizone (JHR Bio-
Accepted September 12, 1995.
Received February 24, 1995.
'Supported by grants HD-20025 and HD-06274 (C.R.L.) from the National Institutes of
Health, Bethesda, MD, and by the Rockefeller Foundation. New York, NY. Presented inpart
at the 25th Annual Meeting of the Society for the Study of Reproduction, July 13-15, 1992.
Raleigh, NC.
2 Correspondence: C.Richard Lyttle, Ph.D., Women's Health Research Institute, WyethAyerst, P.O. Box 8299, Philadelphia, PA 19101. FAX: (610) 989-4831.
249
250
PEREZ ET AL.
science, Lenexa, KS), 100 U/ml penicillin, and 1 tg/ml
streptomycin (Gibco Laboratories, Life Technologies, Inc.,
Grand Island, NY), 1% L-glutamine (100 X, Gibco), 1 mM
sodium pyruvate (Gibco), and 5.6 jiM 2-mercaptoethanol
(Fisher Scientific, Pittsburgh, PA). Highly purified antibodies
(98-99%) were obtained from tissue culture supernatant by
ammonium sulfate precipitation as determined by ELISA
and gel electrophoresis [14].
motactic factors in a physiological manner [18]. Briefly, cells
were cultured in RPMI-1640 with 10% heat-inactivated fetal
calf serum (Hyclone Labs, Logan, UT) in a humidified atmosphere of 5% CO:95% air at 37°C. One week before chemotactic assays, 0.3 mM butyric acid was added to the medium and the cells were cultured for 7 additional days. Cells
were then collected, washed as previously described, and
resuspended in PBS at 1.0 X 106 viable HL-60 Eos cells/ml
for chemotactic assays [18].
Animals
Chemotactic Assay
Three weeks postbilateral ovariectomy, mature female
(25-30 mg) BALB/c mice (Charles River Laboratories, Wilmington, MA) were randomly divided into 4 groups (A-D)
each consisting of 6 animals. Mice were housed in a temperature-controlled room (22°C) with a 14L:10D cycle and
food and water provided ad libitum. On Day 0, mice from
groups A and B were injected with vehicle (saline solution).
Mice in group C received 2 mg of anti-IL-5, and group D
received 2 mg of an isotype match rat IgG (Jackson Immunoresearch Labs, West Grove, PA). All injections were
intraperitoneal. On Days 7, 8, and 9, mice from group A
were injected with vehicle and groups B, C, and D received
1 .tg/day estradiol. On Day 10, animals were weighed and
killed. A one milliliter blood sample per animal was taken
for peripheral eosinophil count. Uteri were removed,
cleaned of fat, blotted, weighed, and divided in two sections. One uterine section was cut longitudinally, rinsed
three times in minimal essential medium (MEM; Sigma), and
incubated for 6 h at 37 0C in an atmosphere of 95% air:5%
CO 2 in methionine-free MEM containing penicillin (100 U/
ml), streptomycin (1 g/ml), and 50 PCi/ml of 35S-methionine (Amersham Inc., Arlington Heights, IL). Medium was
collected after 6 h, and radioactivity that had incorporated
into proteins was determined by trichloroacetic acid (TCA)
precipitation [7]. Proteins were analyzed by SDS-PAGE as
previously described [7, 8, 15]. Fluorograms were prepared
and developed after 48-72-h exposure. The remaining section of each uterus was homogenized at 40C in 10 mM TrisHCI (pH 7.4) buffer (Sigma) at 50 mg/ml in a Kinematica
Polytron homogenizer (Brinkman Instruments, Westbury,
NY). The homogenate was centrifuged at 30 000 x g for 45
min at 40C, and the supernatant was collected and kept on
ice until used in the chemotactic assays. Uterine pellets were
resuspended in T10 C500 buffer (10 mM Tris-HCL containing 0.5 M CaCI, at 25 mg/ml; Sigma), rehomogenized, and
centrifuged as described earlier. The supernatant was assayed for peroxidase activity using guaiacol and H2 02 as
previously described [16].
All uteri from every animal were fixed in Bouin's (Columbia Scientific, Springfield, VA). After processing, the tissue
was embedded in paraffin, and 5-plm sections were cut and
stained with hemotoxylin eosin as previously described [20].
Without knowledge of the treatment groups, pathologic examination was done. The number of eosinophils per highpower field were counted in a minimum of 4 fields per section, and the average eosinophil counts was calculated
(Table 1).
Preparationof Chemotactic Cells
Clone 15 HL-60 cells were differentiated to eosinophillike cells (HL-60/Eos) as previously described [17]. Differentiated HL-60/Eos respond to a variety of eosinophil che-
HematologicalValues
One-milliliter blood samples were taken from every experimental animal. Erythrocytes were lysed by a 1:100 dilution of blood in ammonium oxalate, and cells were
The chemotactic assay was carried out with use of a 48well chemotactic chamber (Neuro Probe Inc., Cabin John,
MD) with a 5-gm pore polycarbonate membrane (Costar Scientific Corporation, Cambridge, MA) [19]. Uterine extract
from control and treated animals were warmed to 37°C, vortexed to expel dissolved gasses, and loaded in quadruplicates in the bottom wells (29 pil/well). To minimize the formation of bubbles in the cell suspension from air trapped in
the membrane's pores, the membrane was saturated with
extract by incubation for 15 min at 37°C in humidified 5%
CO 2:95% air. A cell suspension containing 50 000 viable eosinophils was added to each upper well of the chamber, and
the chamber was incubated for an additional 45 min. The
membrane was then fixed in methanol and stained with
Wright-Giemsa (Hemacolor; EMI Diagnostics, Gibbstown,
NJ). Chemotaxis was assessed by counting the eosinophils
that had migrated through the membrane and were fixed to
the surface next to the uterine extract. Each extract was run
in a minimum of four wells, and the results were averaged
to obtain a value (expressed as cells per well) for the individual animal. Chemotactic activity for each sample was analyzed by one way analysis of variance on either the raw data
or log transformed data depending on normality and variance homocedasticity criteria. A Student Newman-Keuls test
was used to determine specific differences among groups.
and p values of < 0.001 were considered significant.
IHistology
251
EOSINOPHILS AND RESPONSES TO ESTROGEN
TABLE 1. Percentage of eosinophil granulocytes in blood per ml and number in
uterus (field section) in every experimental group.*
Treatment group
Control
Estradiol
CigG + Estradiol
MablL-5 + Estradiol
*Mean
%Blood eosinophils/ml
0.3 + 0.06 (6)
0.3 + 0.00 (6)
0.3 + 0.03 (6)
0.03 0.04 (6)
-,o)
0.00
37.22
32.22
0.33
4
'5)
# Eosinophils per section
(high power field)
3
+ 0.00
± 2.22
±+2.78
± 0.19
D
3
SE. Number in parenthesis is sample size.
2
counted using a hemacytometer. One blood smear per each
experimental animal was used to evaluate the percentage
of eosinophils per total nucleated white cells, and the total
eosinophil count was calculated. The average number of
eosinophils present in each experimental group was analyzed by one-way analysis of variance on either the raw data
or log transformed data depending on normality and variance homocedasticity criteria. A Student Newman-Keuls test
was used to determine specific differences among groups,
and p values of < 0.001 were considered significant.
RESULTS
Treatment of ovariectomized mice with 1 ig estradiol resulted in a significant increase in uterine weight when compared with control animals. Administration of TRFK-5 had
no effect on the estrogen-induced increase in uterine wet
weight (Fig. 1). Previous studies from our laboratory have
demonstrated that estrogen administration regulates the expression of an ECF-U in the stromal cells of the immature
rat uterus [6]. Estradiol administration to ovariectomized
mice resulted in an increase of eosinophil chemotactic activity in mouse uterus, and coadministration of the IL-5
monoclonal antibody did not affect the chemotactic activity
induced by estradiol (Fig. 2). These findings confirm our
previous observations that ECF-U activity is increased by
estrogen administration in the rat [6] and extends this finding to the mouse. No differences in response were seen with
the injection of control IgG.
Estrogen-dependent increases in uterine peroxidase activity have been described for more than 30 yr [21]. Several
studies have indicated that there are two possible sources
of peroxidase activity [22-24]. While some data support the
possibility that epithelial cells contain peroxidase activity,
the majority of the uterine peroxidase activity is due to infiltration by eosinophils [22, 25]. In the mouse uterus, a statistically significant increase in peroxidase activity was observed following the injection of estrogen. Coadministration
of control IgG had no significant effect on this estrogeninduced increase in peroxidase activity. In contrast, animals
injected with the monoclonal antibody to IL-5 (TRFK-5)
showed a significant reduction in uterine peroxidase activity
(Fig. 3). The small but significant increase in peroxidase
activity may be due to an incomplete block in the number
of eosinophils or to peroxidase activity from other sources.
,3
1
0
CONT
E2
CIgG + MabIL5 +
E2
E2
FIG. 1. Effect of estradiol on relative uterine weight (mg/100 g BW) associated with
or without preadministration TRFK-5. Ovariectomized mice were injected with 2 mg
of anti-IL-5 (MablL5) or 2 mg of control IgG (CIgG) seven days before administration
of E2 (1 g/day on 3 consecutive days). Controls were treated with vehicle alone. *p
> 0001.
Uteri from animals treated with 1 ig of estradiol demonstrated a significant increase in the number of eosinophils
(Fig. 4). Histological evaluation reinforces our earlier observations since minimal uterine infiltration of eosinophils
was observed following injection with monoclonal antibod700
600
500
L-i
400
0
On
0 300
04
LII
200
Z
100
z
0
CONT
E2
CIgG + MabIL5 + T10
E2
E2
FIG. 2. Estrogen induction of ECF-U activity. Ovariectomized mice were given 2
mg of Mab IL-5 or 2 mg of control IgG seven days before administration of E2 (1 Iug/
day on 3 consecutive days). Controls were treated with vehicle alone. Uterine extracts were prepared 72 h after the first estradiol administration as described in
Materials and Methods. *p < 0.001 for control vs. estradiol. Negative controls for
chemotactic assay used 10 mM Tris-HCI buffer (T10).
252
PEREZ ET AL.
30
0
20
I
©
10
0
0
CONT
E2
CIgG + MabIL5 +
E2
E2
FIG. 3. Effect of estradiol on uterine peroxidase activity in ovariectomized mice in
the presence or absence of Mab IL-5. *p < 0001.
ies to IL-5 (Table 1). This finding correlates with the statistically significant decrease of eosinophils in blood smears
of animals treated with TRFK-5 (Table 1). Administration of
the IL-5 antibody did not appear to alter the general morphology of the estrogen stimulated uterus, and heights of
the epithelial cells were comparable with other animals receiving estrogen. In order to determine whether the decrease in uteri eosinophil number resulted in the modulation of estrogen responses of specific cell types, we
examined the expression of C3 in the mouse uterus. Analysis
of the radiolabeled secreted proteins in the medium clearly
demonstrated that estrogen increased C3 synthesis. As in the
case of the uterine wet weight and ECF-U, the synthesis of
C3 was unaffected by the administration of IL-5 or control
immunoglobulins (Fig. 5).
DISCUSSION
In the rodent uterus, early responses to estrogen include
changes in vascular permeability, water imbibition, and cellular infiltration [26-28]. Many of these responses are similar
to the inflammatory response possibly regulated by various
mediators of inflammation. Some of these inflammatory responses are generated by histamine, serotonin, bradykinin,
prostaglandin E2, and leukotrienes. Complement components and leukotrienes are involved in the chemotactic process, causing a direct migration of cells along a concentration gradient [29, 30]. In this study we used ovariectomized
mice severely depleted of circulating eosinophils as a model
to determine the role of eosinophils in early uterine responses to estrogen. The ability of TRFK-5 to block the production of circulating eosinophils was clearly demonstrated
FIG. 4. Effect of estradiol on uterine eosinophilic infiltration. A) Uterus of control
mouse. B)Uterus from mouse treated with 1 pg of estradiol for 3 consecutive days,
showed extensive cellular infiltration. C) Uterus from mouse injected with a single
dose of 2 mg TRFK-5 and seven days later with 1 g of estradiol for three consecutive days. No statistically significant cellular infiltration was observed. Mice treated
with control IgG and estradiol were the same as the estradiol animals (data not
shown). (Arrows indicate eosinophilic infiltration). All sections were examined at
the same magnification (x21).
EOSINOPHILS AND RESPONSES TO ESTROGEN
253
FIG. 5. Synthesis and secretion of 35-S-labeled uterine proteins from control and estradiol-treated mice seven days after the
administration of 2 mg per mouse of TRFK-5 as described in Material and Methods. Each lane was loaded with equal amounts of
TCA-precipitable radioactive proteins obtained from the media after a 6-h incubation, under reducing and nonreducing conditions.
Proteins were separated on a 10% SDS-PAGE followed by fluorography. Upper arrow indicates the position of the 180-kDa protein
(C3).Arrows on right fluorography indicate the position of 115- and 65-kDa protein subunits of C3.
in our data by both blood cell counts and the absence of
infiltrating cells in mice uteri after treatment with the monoclonal antibody.
The data presented here clearly demonstrate that the results of estradiol administration in the ovariectomized
uterus are similar whether or not eosinophils had infiltrated
the uterus. Thus changes in morphology, uterine wet
weight, protein synthesis, and C3 synthesis were identical
in both the absence and presence of eosinophils.
The estrogen-dependent increase in uterine peroxidase
was reported over 20 years ago, and, because of its ease of
assay, this elevation in peroxidase activity has been used as
a marker for estrogenic stimulation. The greater part of the
peroxidase activity extracted from the stromal and myometrial layers of the rat uterus is probably due to eosinophil
accumulation following estrogen treatment [21, 25, 31, 32].
Our data demonstrate a significant reduction in the number
of circulating eosinophils. This is in agreement with the results of Coffman et al. [121 following in vivo treatment in
mice with an antibody to IL-5. Estradiol stimulation of
mouse uterine peroxidase activity was significantly reduced
under those conditions. These data collectively support the
view that the majority of uterine peroxidase activity is of
eosinophil origin and agree well with several lines of evidence [22, 24] that demonstrate that 95-98% of the uterine
peroxidase activity originates in these cells. However, a
minimum increase in peroxidase activity still remains in animals depleted of eosinophils. The source of this peroxidase
activity remains to be determined. Eosinophil peroxidase is
a potent cytotoxic element when it is combined with halide
and H20 2, having the capacity to kill a variety of targets such
as parasites, tumor cells, and mast cells. The role of peroxidase activity in uterine function is not known at present.
The mechanism controlling estrogen-induced infiltration
of eosinophils into the uterus appears to be regulated in
part by an eosinophil chemotactic factor previously described by Lee et al. [6]. Extracts of estrogen-treated uteri
contained significantly greater eosinophil chemotactic activity than extracts from control uteri. However, estrogen
regulation of the chemotactic protein was not abolished in
the absence of eosinophils. Data from our present study
support the view that estrogens may modulate this activity
and that it originates from the uterine cells as previously
suggested [6]. Our present data further indicate that the infiltration of eosinophils does not modulate the activity of
this factor by down-regulating its activity. These findings
254
PEREZ ET AL.
agree well with previous studies from our laboratory [33] in
which estradiol-stimulated uteri from rats, in the presence
or absence of pertussis toxin, resulted in blockage of uterine
eosinophil infiltration. However, although the circulating
concentration of eosinophils was not reduced, no effect was
seen on the estradiol-stimulated increase in uterine weight
or C3 secretion [33]. The outcome of that experiment, however, could have been potentially compromised by possible
effects of pertussis toxin on other G-protein-stimulated responses.
The presence of eosinophil chemotactic activity in mice
treated with the IL-5 antibody suggests that the estrogenregulated chemotactic factor is independent of IL-5. However, further experimentation is required to settle the question. The results of this study strongly indicate that
infiltrating eosinophils play little role in early estradiol-induced changes occurring in the uterus (growth, protein synthesis, epithelial C3 synthesis, and stromal production of an
eosinophil chemotactic factor).
In the proestrous rodent, eosinophils are observed at the
stromal-myometrial junction, and not in the epithelium. Just
before the estrous phase, eosinophils degranulate and are
then phagocytized by macrophages [34]. This sequence of
events suggests that eosinophil granules might provide the
enzymes necessary for endometrial stromal remodeling [6].
Whether eosinophils do play such a role in the tissue remodeling of the uterus is currently under investigation. The
potential role would agree with the observation of an eosinophil infiltration during early postpartum when considerable uterine remodeling is taking place [35].
REFERENCES
1. Ross R, Klebanoff SJ.The eosinophilic leukocyte. J Exp Med 1966: 124:653-659.
2. Tchernitchin A. The role of eosinophil receptors in the non-genonomic response to
oestrogens in the uterus. J Steroid Biochem 1979; 11:417-424.
3. Tchernitchin AN. Eosinophil-mediated non-genomic parameters of estrogen stimulation-a separate group of responses mediated by an independent mechanism. J Steroid
Biochem 1983: 19:95-100.
4. Ryt6mnaa T. Organ distribution and histochemical properties of eosinophil granulocytes
in rat. Acta Pathol Microbiol Scand Suppl 1960; 140:50-51.
5. Gleich GJ. Current understanding of eosinophil function. Hosp Pract 1988; 3:137-160.
6. Lee YH, Howe RS,Sha S, Teuscher C, Sheehan DM, Lyttle CR. Estrogen regulation of an
eosinophil chemotactic factor in the immature rat uterus. Endocrinology 1989 125:30223028.
7. Komm BS, Rusling DJ. Little CR. Estrogen regulation of protein synthesis in the immature
rat uterus: the analysis of proteins released into the medium during in vitro incubation.
Endocrinology 1986; 118:2411-2416.
8. Little CR, Wheeler C, Komm BS. Hormonal regulation of rat uterine secretory protein
synthesis. Adv Exp Med Biol 1987; 230:119-136.
9. Silberstein DS, Austen KF, Owen WF. Hemopoietins for eosinophils. Glycoprotein hor-
ones that regulate the development of inflammation in eosinophilia-associated disease
Hematol Oncol Clin North Am 1989; 3:511-533.
10. Yamaguchi Y, Suda T, Suda J, Eguchi M, Miura Y, Harada N. Tominaga A, Takatsu K.
Purified interleukin 5 supports the terminal differentiation and proliferation of murine
eosinophilic precursors. J Exp Med 1988; 167:43-56.
11 Lopez AF, Begley CG. Wiliamson DJ, Warren DJ, Vadas MA. Sanderson CJ. Murine eto
sinophil differentiation factor. An eosinophil-specific colony-stimulating factor with attivity for human cells.J Exp Med 1986; 163:1085-1099.
12. Coffman RL, Seymour BW, Hudak S, Jackson J. Rennick D. Antibody to interleukin-5
inhibits helminth-induced eosinophilia in mice. Science 1989; 245:308-310
13 SchumacherJH, O'Garra A, Shrader B, Van-Kimmenade A. Bond MW, Mosmann TR.
Coffman RT. The characterization of four monoclonal antibodies specific for mouse IL5 and development of mouse and human IL-5 enzyme-linked immunosorbent. J Immunol 1988; 141:1576-1581.
14. Harlow E, Lane D. Antibodies. A Laboratory Manual. New York: Cold Spring Harbor
Laboratory Publications; 1988: 245-281.
15. Sundstrom SA. Komm BS, Xu Q. Boundy V, Lyttle CR. The stimulation of uterine conmplement component C gene expression by antiestrogens. Endocrinology 1990:
126:1449-1456.
16. Himmelhoch SR, Evans WH, Mage MG, Peterson EA. Purification of myeloperoxidases
from the bone marrow of the guinea pig. Biochemistry 1969; 8:914-921.
17. FishkoffSA, Pollak A, Gleich GJ. TestaJR, Misawa 5, Reber TJ. Eosinophilic differentiation
of the human promyelocytic cell line, HL-60. J Exp Med 1984 160:179-196.
18. Howe RS,Fischkoff SA, Rossi RM, Lyttle CR. Chemotactic capabilities of HL-60 human
tmyeloid leukemia cells differentiated to eosinophils. Exp Hematol 1990: 18:299-303.
19. Falk W, Goodwin RC. Leonard EJ. A 48-well micro chemotaxis assembly for rapid and
accurate measurement of leukocyte migration.J Immunol Methods 1980; 33:239-241.
20. Sheehan DC, Hrapchak BB. Theory and practice of histotechnology. St. Louis, MO: CV
Mosby Co. Publications; 1973: 14-15.
21. Lucas FV, Neufeld HA, Utterback JG, Martin AP. Stotz E. The effect of estrogen on the
production of a peroxidase in the rat uterus.J Biol Chem 1955; 214:775-780.
22. Keeping HS, Lyttle CR. Monoclonal antibody to rat uterine peroxidase and its uise in
identification of the peroxidase as being of eosinophil origin. Biochim Biophvs Acta
1984; 802:309-406.
23. Keeping HS, Lyttle CR. Regulation of uterine steroid receptor levels and peroxidase
activity during the puerperium of the nonlactating rat. Biol Reprod 1982 27:1011-1022.
24. Lyttle CR, Jellinck PH. Subcellular localization of oestrogen-induced uterine peroxidase.
BiochemJ 1976: 160:237-241.
25. Klebanoff SJ. Inactivation of estradiol by rat uterine preparations. Endocrinology 1965:
76:301-311.
26. McRae A, Kennedy TG. Estrogen, progesterone, and the blood-uterine lumen permeability barrier in rats. Biol Reprod 1981; 25:314-317.
27. Daya S. Clark D, Delvin C, Jarrall J. Preliminary characterization of two types of sup
pressor cells in the human uterus. Fertil Steril 1988; 44:778-785.
28. Galman MS, Sundstrom SA, Lyttle CR. Antagonism of estrogen and antiestrogen-induced
uterine complement component C3 expression by ICI 164,384. J Steroid Biochem 1990;
36:28-36.
29. Issekutz AC, Movat HZ. The effect of vasodilator prostaglandins on polymorphonuclear
leukocyte infiltration and vascular injury. Am J Pathol 1982; 107:300-309.
30. Czarnetski BM, Rosenbach T. Chemotaxis of human neutrophils and eosinophils tosnard
leukotriene B4 and its 20-w-oxidation products in vitro. Prostaglandins 1986; 31:851-858.
31. DeSombre ER, Lyttle CR. Steroid hormone regulation of uterine peroxidase activity Adv
Exp Med Biol 1979: 17:157-171.
32. Sanderson CJ, Campell HD, Young IG. Molecular and cellular biology of eosinophil
differentiation factor (interleukin-5) and its effects on human and mouse B cells. mnmunol Rev 1988;: 102:29-50.
33. Lyttle CR, Teuscher C, Medlock KL, Sheehan DM. Estracliol-stimulated increases in uter
ine eosinophils and nuclear type II estrogen-binding sites are prevented by pertussis
toxin. Endocrinology 1988; 125:2773-2779.
34. Rytomaa T. Organ distribution and histochemical properties of eosinophil granulocytes
in rat. Acta Pathol Microbiol Scand 1960; 50(suppl 140):11-106.
35. Leiva MC, Xut Q, Galman M, Lyttle CR. Ontogeny of the production of an estrogen
regulated eosinophil chemotactic factor in the rat uterus. Biol Reprod 1991; 45:818-823.