| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Developmental and Molecular Biology (V.G.-E., J.W.P.), Obstetrics & Gynecology and Women’s Health (J.W.P.) and Center for the Study of Reproductive Biology and Women’s Health, Albert Einstein College of Medicine, New York, New York 10461
Address all correspondence and requests for reprints to: Jeffrey W. Pollard, Ph.D., Albert Einstein College of Medicine, Department of Developmental and Molecular Biology, 1300 Morris Park Avenue, Bronx, New York 10461. E-mail: pollard{at}aecom.yu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
dendritic
cells, eosinophils, and neutrophils are the predominant leukocytes
homing to the cycling uterus (6, 7, 8, 9, 10, 11). Considerable literature exists showing that eosinophils infiltrate cyclically the rodent uterus. Infiltration of the rat uterus by eosinophils, coincident with the estrus cycle (12), was observed during the 1950s, and this observation has been confirmed and extended to different species such as mouse and human (9, 10, 13, 14). Since then, numerous investigations have shown that: 1) injection of E2 into ovariectomized or immature rodents causes a dramatic increase in uterine eosinophils (5) and an increase in uterine peroxidase activity (15) that was shown later to be restricted to eosinophils (16, 17); 2) uterine eosinophil numbers vary more than 100-fold during the normal estrous cycle (18); 3) eosinophils undergo lysis in the uterus around the time of estrus releasing their contents into the extracellular spaces (10, 19); and 4) the E-induced uterine eosinophilia is associated with marked uterine edema (20). Despite the fact that eosinophil homing into the cycling uterus is well established, the role of these cells and their chemoattractant remain to be elucidated.
Of the cytokines implicated in modulating eosinophilic inflammation, only IL-5 and eotaxin have been identified to selectively regulate eosinophil trafficking (21, 22, 23). IL-5 is responsible for the proliferation, differentiation, recruitment, and activation of eosinophils. Studies in mice lacking IL-5 caused by passive immunization against IL-5 or by the generation of a null mutation of the IL-5 gene showed that these mice are virtually devoid of eosinophils in the blood stream and that consequently their uterine eosinophil population was depleted strongly (24, 25, 26). Nevertheless, these studies did not demonstrate the physiological chemoattractant effect of IL-5 for eosinophil homing from the blood stream into the uterus. Eotaxin, a member of the C-C chemokine family, has been originally identified as a novel chemotactic agent for eosinophils into the lung in a guinea pig model of eosinophilic airway hypersensitivity (22). Eotaxin mRNA is induced in multiple animal models of eosinophilic inflammation, in human tissue in response to allergen challenge (23, 27, 28), and also in the mouse mammary gland during postnatal development (29). Recently, the expression of eotaxin and its receptor have been reported in human endometrium through the menstrual cycle (30). However, the functions of eotaxin during the estrus cycle remain unknown.
The purpose of this study was to examine the uterine eosinophil population in eotaxin-deficient mice and determine the physiological importance of eotaxin and eotaxin-dependent eosinophils in reproduction.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The onset of puberty was assessed in two ways: 1) vaginal opening and
2) the onset of complete estrous cycles (two cycles). Mice were
examined daily from 4 wk of age to determine whether vaginal opening
had occurred. Analysis of estrous cycles was performed from 5 wk of
age. Vaginal smears were obtained on a daily basis, stained with
hematoxylin/eosin, and examined for cellular content. Four stages of
the cycle were defined as followed: proestrus (100% intact and alive
epithelial cells), estrus (100% cornified epithelial cells), metestrus
(
50% cornified epithelial cells and 50% leukocytes), and diestrus
(80–100% leukocytes). For determination of the age of first
parturition, seven females of each group were mated with males from the
same group from 6 wk of age, and the age of the first parturition was
determined. Successful pregnancy interval was also determined in weeks
by mating nine wild-type females and eight mutant female with males
from the same group. The male was maintained with the female from the
mating period to d 2 after parturition, then separated, and finally
replaced with the female when the weaning period occurred (3 wk after
parturition). Interval times between two successful pregnancies were
noted for two to six consecutive pregnancies for each female per
group.
Adult female mice (10–12 wk) were bilaterally ovariectomized on d 1, primed at d 7 and d 8 by sc injection of 100 ng of E2 in 0.1 ml of peanut oil, and then treated with 3 consecutive sc injections of 100 ng of E2 in 0.1 ml of peanut oil at d 15, 16, and 17, or with 0.1 ml of peanut oil for the control mice. Mice were killed 24 h or 42 h after the last injection, and uteri removed for analysis. The wet weight of uteri in gram was measured for 6–8 mice of each group, and statistical evaluations using standard deviations were performed with a two-tailed t test.
Histochemistry
Uteri were fixed overnight in formalin (Sigma, St.
Louis, MO) and paraffin wax embedded for hematoxylin/eosin staining, or
OCT embedded (Tissue Tek; Bayer Corp., Elkart, IN)
and frozen for peroxidase activity analysis. Five-micrometer paraffin
sections were consecutively stained with hematoxylin and lightly with
eosin Y to identify eosinophils by their pink cytoplasmic granules and
segmented nucleus. Five-micrometer frozen sections were used to
detect the endogenous peroxidase activity restricted to eosinophils in
uterine tissue (17). Briefly, frozen sections were fixed
in acetone at 4 C for 10 min, rinsed in PBS, and then incubated for
2–5 min at room temperature in diaminobenzidine in presence of
hydrogen peroxide in a ratio determined by the Vector Laboratories, Inc. peroxidase kit (Burlingame, CA). Sections
were finally rinsed in water and counterstained with hematoxylin.
Northern blot analysis
Total RNA from uteri of one to three mice from each group
was isolated by the method of Chomczynski and Sacchi (32).
Ten micrograms of total RNA was separated by formaldehyde-agarose gel
electrophoresis, transferred to nylon filters, and probed with a
[32P]dCTP-labeled cDNA probe for eotaxin
[ATCC (Manassas, VA) No. 1463042; GenBank accession no.
AA711712] using the method previously described (33).
Western blot analysis
Each individual uterus was homogenized in RIPA buffer (9.1
mM dibasic sodium phosphate, 1.7 mM monobasic
sodium phosphate, 150 mM NaCl, 1% Nonidet P-40, 0.5%
sodium deoxycholate, 0.1% SDS, pH 7.4), and protein concentration
determined by a BCA protein assay (Pierce Chemical Co.,
Rockford, IL). Equal amount of protein (80 µg) under reducing
conditions were separated by electrophoresis on 15% SDS-polyacrylamide
gels and transferred onto polyvinylidendifluoride Immobilon-P membranes
(0.45 µm; Millipore Corp., Bedford, MA). The membranes
were blocked with 5% nonfat dry milk and 0.1% Tween 20 in
Tris-buffered saline (TBS: 150 mM NaCl, 20 mM
Tris, pH 7.6) for 1 h at RT. They were then incubated with the
IgG-purified rabbit antimurine eotaxin antibody (generously provided by
Dr. Steven Kunkel, University of Michigan, Ann Arbor, MI) at 0.5
µg/ml in blocking solution for 1 h at RT, washed 3 times for 15
min in TBS-Tween buffer, and subsequently incubated with a 1:3,000
dilution of horseradish peroxidase-linked secondary antibody
(Santa Cruz Biotechnology, Inc., Santa Cruz, CA) for 45
min at RT. After washing with TBS-Tween, immunodetection was achieved
with an enhanced chemiluminescence system (ECL; Amersham Pharmacia Biotech, Piscataway, NJ). Recombinant murine eotaxin
(Chemicon International Inc., Temecula, CA) was used as a reference
control and an 
-guanine nucleotide dissociation inhibitor (GDI) (a
gift from Dr. Perry Bickel, Washington University, St. Louis, MO) was
used as internal control for protein abundance.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). In
the adult, they were elevated during proestrus (P) returning to a
significantly lower level at diestrus (D) (P = 0.0004,
two-tailed t test) (Fig. 1B). In ovariectomized mice,
transcript levels of eotaxin were almost undetectable, but they were
induced dramatically by E2 treatment, 39-fold at 24 h and 31-fold
at 42 h after injection (Fig. 1C, top and bottom
panels). As predicted from the size of the eotaxin cDNA, the major
mRNA transcript was about 1 kb. It is of interest that in the adult
uterus but not in the peripubertal uterus, smaller transcripts were
systematically detected with a major one at 0.6 kb after E2 stimulation
either during the estrous cycle (Fig. 1B) or following E2 treatment of
ovariectomized mice (Fig. 1C). In the latter case, the induction of the
0.6-kb transcript paralleled perfectly the induction of the larger
transcript, although it was virtually undetectable in the untreated
ovariectomized uterus (Fig. 1C, bottom panel). Even though
this smaller transcript has never been described in mice, it has been
reported in certain human tissues such as the small intestine, colon,
and heart (28). The nature of the 0.6-kb transcript is not
certain, but is sufficiently large to encode the 8.4-kDa eotaxin
protein and may represent use of an alternative polyadenylation site
(34). The eotaxin protein levels from total uterine
homogenates during the estrous cycle were analyzed in parallel to the
transcript levels by Western blot analysis using an anti-murine eotaxin
antibody. A band at approximately 8 kDa, which corresponded to the
reference control protein, indicated the presence of mouse eotaxin in
the uterus. Interestingly, eotaxin protein was only detected during the
proestrus phase (P/E), which was in this case at late proestrus, a
stage of maximal transcript expression. The inability to detect protein
in estrus or diestrus may reflect the sensibility of the antibody or
rapid degradation of the protein.
|
), whereas they were totally absent in
eotaxin-deficient mice (Fig. 2B). In adult mice, eosinophil recruitment
was analyzed following E2 treatment of ovariectomized mice. In
untreated mice, eosinophils were virtually absent (data not shown).
However, in confirmation of previous studies (5, 15), upon
E2-stimulation, the number of eosin-positive or peroxidase-positive
eosinophils increased gradually, reaching the highest number at a time
corresponding to an experimental estrus (determined by vaginal smear)
in wild-type mice 42 h after E2 treatment (Fig. 2, C, E, and F).
Eosinophils were recruited into the uterine stroma in the vicinity of
the luminal epithelium (Fig. 2, C and E) and in the myometrium
surrounding the stroma (Fig. 2F). In contrast, no eosin-positive or
peroxidase-positive eosinophils could be seen in the uterus of
eotaxin-deficient mice at the experimental estrus at any location (Fig. 2, D and G). Thus, eotaxin transcript level correlates with the
recruitment of eosinophils into the uterus of the pubertal and adult
mice. It is of interest to note that eotaxin protein is detected early
during the estrous cycle at proestrus when eosinophil homing into the
uterus begins, whereas the protein is undetectable by Western blot at
estrus at the time when eosinophil recruitment is the highest. This
suggested that eotaxin is an early signal for eosinophil homing into
the uterine stroma. In the absence of eotaxin, eosinophils are
completely absent at every stage of development.
|
Reproductive phenotypes of eotaxin-deficient mice
Because of the early presence of eosinophils in the uterus of
pubertal mice, we examined the consequence of their absence on pubertal
uterine development, as well as on the onset of puberty and on the
reproductive functions of eotaxin-deficient mice.
The number of uterine glands is a useful parameter to evaluate
the maturation of uterine morphogenesis upon puberty (35).
The number of glands was enumerated in uterine cross-sections of
wild-type and mutant 5-wk-old mice. The gland number was similar in
both groups, indicating that the morphologic development of the uterus
of mutant mice seemed normal (Fig. 3A).
Vaginal opening, as one measure of the onset of puberty, was examined,
and found to occur at a similar age in both groups between 4.5–5 wk of
age. The age of the onset of estrous cycle was next determined by
analyzing hematoxylin/eosin staining of daily vaginal smears starting
from 5 wk. Once the estrus phase was followed by metestrus, diestrus,
and proestrus phases, the estrous cycle was considered established. The
first estrous cycle started at 8.6 ± 0.2 wk in eotaxin-deficient
mice, significantly delayed by 2 wk compared with 6.6 ± 0.2 wk in
wild-type mice (Table 1). The age of
first pregnancy was assessed by mating seven females from each group at
the age of 6 wk and by determining the age of their first parturition.
Consistent with the delay in the onset of estrous cycle, the age of the
first parturition was significantly delayed by 2 wk in the mutant mice
occurring at 11.3 ± 1.5 wk of age compared with 9.7 ± 1.6
wk of age in wild-type mice (Fig. 3B). However, once the onset of
puberty occurred in mutant mice, the estrous cycle was not disturbed,
and it displayed a normal length compared with wild-type mice (Table 1). Consequently, once mice reached adulthood, the reproduction
function of eotaxin-deficient females was normal with similar litter
size and interval time between consecutive successful pregnancies
compared with wild-type mice (Table 1).
|
|
). Our data demonstrate that even though eosinophils influence uterine development upon the onset of puberty, the lack of eosinophils does not affect the cycling uterus and hence the reproductive functions of the adult mutant mice. These data suggest that eosinophils are not required for the proper development of the cycling uterus.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A previous study has shown the existence in the immature rat uterus of an eosinophil chemotactic factor (ECF-U) whose synthesis is E2 receptor mediated (37). After isolation and characterization of uterine proteins displaying eosinophil chemoactivity, an amino acid sequence analysis indicated that one of these proteins could be cyclophilin, a protein that specifically binds cyclosporin A (38). Obviously ECF-U/cyclophilin may contribute to the uterine eosinophil infiltration; nevertheless there is no evidence showing that they are necessary for this purpose. In the present study, we demonstrate clearly that eotaxin is the chemokine necessary for uterine eosinophil recruitment. It may be that eotaxin is in fact ECF-U, because the apparent molecular mass of mouse eotaxin under nonreducing condition is 17 kDa, similar to the 20-kDa rat ECF-U. However, because cyclophilin is a peptidyl propyl-isomerase that can act as a chaperone (39), it may be that eotaxin copurifies with it under nonreducing conditions. Eotaxin also meets the E-dependent requirement of the expected uterine eosinophil chemofactor because its mRNA level is dramatically induced in ovariectomized mice after E2 treatment. Moreover, both eotaxin protein and mRNA are detected at proestrus, when eosinophil homing into uterus begins. Our findings are consistent with the detection of eotaxin messengers by RT-PCR at estrus in mice (40) and detection of eotaxin protein in human endometrium (30).
Among the known eosinophil chemotactic factor, IL-5 was also considered as a candidate acting in the uterus. Although the eosinophil population was strongly depleted in the blood of IL-5-deficient mice, the temporal fluctuations in eosinophil infiltration and localization in the uterine stroma exhibited by the small residual population was unchanged in the IL-5-deficient mice during the estrous cycle (26). Moreover, the coadministration of the monoclonal antibody anti-IL-5 with E2 to ovariectomized mice did not affect the increase in the eosinophil chemotactic activity in the mouse uterus (25). Therefore, it seems that the uterus illustrates a cooperative and not synergistic role of IL-5 and eotaxin in recruiting eosinophils in a specific organ, which has been previously reported in other systems such as lung and skin (41, 42). These observations agree with the notion that IL-5 provides the signal for the release of a pool of eosinophils from the bone marrow, whereas eotaxin remains the critical local chemoattractant for E-dependent recruitment of eosinophils into the uterine stroma.
The eotaxin-deficient mouse, as a uterine eosinophil-free mouse, was therefore a useful model to study the role of eosinophils in uterine development. In these mice, eosinophils are at normal levels in the bone marrow and peripheral blood (31, 43). Moreover, the overexpression of IL-5 transgene is associated with an increased level of gastrointestinal eosinophils in the absence of eotaxin (43). Consequently, impaired eosinophil homing in tissues or release in the circulation are unlikely to account for reduced eosinophil uterine levels or for nonresponsiveness of eosinophils to other chemotactic agents such as IL-5 in eotaxin-deficient mice. The early detection of eotaxin mRNA at 5 wk of age was in line with the infiltration of eosinophils in the pubertal uterus. In the absence of eosinophils, there was a 2-wk delay in the onset of estrous cycles. This finding could be the result of a defect of the hypothalamic/pituitary axis or ovarian functions that are necessary for a proper establishment of the onset of puberty (see Ref. 44 for review). However, the vaginal opening, which is also regulated by the hypothalamic/pituitary axis, occurs at a similar time in the eotaxin-deficient mice compared with wild-type mice. This argues for a functional hypothalamic/pituitary axis in eotaxin-deficient mice. Moreover, eosinophils are completely absent from ovarian tissue of wild-type mice at puberty (at 5 and 7 wk of age, data not shown), excluding a local role of eosinophils on ovarian-dependent steroid hormonal levels, necessary for the onset of puberty (see Ref. 44 for review). Thus, the delay of the onset of estrous cycles in eotaxin-deficient mice, consistent with the delay of the age of first parturition, does not seem to be due to a systemic hormonal defect but rather suggests a local role of eosinophils in the pubertal uterus in preparing the mature uterus for pregnancy.
In contrast to this novel finding of the role of eosinophil in pubertal uterine development, investigations into the role of the E-induced infiltration of eosinophils into the uterine stroma during the estrus cycle have been extensive. Because other events such as increases in uterine growth, edema, protein synthesis, and epithelial C3 synthesis are also E-dependent in cycling mice, it has been thought for long time that uterine recruitment of eosinophils was directly associated with these processes (20, 45, 46, 47, 48). However, our data show very clearly that in the absence of eosinophils, the increase in wet weight as a measure of uterine edema was not affected. Our observations were consistent with others studies using different mouse models that strongly depleted uterine eosinophils by using pertussis toxin-treated mice, IL-5-deficient mice, or anti-IL-5 antibody-treated mice (11, 25, 26). In the two last studies, increase of complement C3 synthesis and changes in uterine morphology during estrus cycle were not modified in relative absence of eosinophils. In addition, a precise study of reproduction capacities of IL-5-deficient mice indicated no major abnormality (26), as judged by a normal estrus cycle length, although the estrus phase was significantly longer in the mutant mice, and even though the time to observe the copulatory plug in mutant mice was significantly reduced by 2 d, the length of gestation and the outcomes of pregnancies were normal. In our uterine eosinophil-free mouse model, displaying a different background (129/SvEv), we confirmed the normal estrus cycle length and litter size as well as the ability to have successive pregnancies occurring at a normal frequency. Consequently, eosinophils do not have a nonredundant role in regulating cyclicity in the adult uterus. This surprising conclusion does still not rule out a potential role for eotaxin in the estrous cycle, because, in humans, the eotaxin receptor (CCR3) is not only expressed on eosinophils but also on endometrial epithelial cells (30). If this is similar in mice, then chemokines other than eotaxin that are known ligands for CCR3 (RANTES, MCP-3) (49) could act directly on the luminal epithelium to control the morphologic modifications occurring upon the estrus cycle, even in absence of eosinophils.
In summary, in this study, we analyzed the uterus of eotaxin-deficient mice, and identified eotaxin as the necessary E-induced chemokine for eosinophil homing to the uterine stroma during puberty and adulthood. Our data suggest that eosinophils have a local role in the pubertal uterus in preparing the mature uterus for pregnancy but demonstrate that these cells are not required for regulating the duration of the estrous cycle and fertility of mice.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: ECF-U, Immature rat uterus of an eosinophil chemotactic factor; TBS, Tris-buffered saline.
Received March 14, 2001.
Accepted for publication July 6, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  W. V. Ingman and R. L. Jones Cytokine knockouts in reproduction: the use of gene ablation to dissect roles of cytokines in reproductive biology Hum. Reprod. Update, March 1, 2008; 14(2): 179 - 192. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Salati, E. Bianchi, R. Zini, E. Tenedini, D. Quaglino, R. Manfredini, and S. Ferrari Eosinophils, but not neutrophils, exhibit an efficient DNA repair machinery and high nucleolar activity Haematologica, October 1, 2007; 92(10): 1311 - 1318. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Naciff, G. J. Overmann, S. M. Torontali, G. J. Carr, Z. S. Khambatta, J. P. Tiesman, B. D. Richardson, and G. P. Daston Uterine Temporal Response to Acute Exposure to 17{alpha}-Ethinyl Estradiol in the Immature Rat Toxicol. Sci., June 1, 2007; 97(2): 467 - 490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Pan, L. Zhu, Y. Deng, and J. W. Pollard Microarray Analysis of Uterine Epithelial Gene Expression during the Implantation Window in the Mouse Endocrinology, October 1, 2006; 147(10): 4904 - 4916. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Niklaus and J. W. Pollard Mining the Mouse Transcriptome of Receptive Endometrium Reveals Distinct Molecular Signatures for the Luminal and Glandular Epithelium Endocrinology, July 1, 2006; 147(7): 3375 - 3390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Kwekel, L. D. Burgoon, J. W. Burt, J. R. Harkema, and T. R. Zacharewski A cross-species analysis of the rodent uterotrophic program: elucidation of conserved responses and targets of estrogen signaling Physiol Genomics, November 17, 2005; 23(3): 327 - 342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Chaves-Pozo, V. Mulero, J. Meseguer, and A. G. Ayala Professional phagocytic granulocytes of the bony fish gilthead seabream display functional adaptation to testicular microenvironment J. Leukoc. Biol., August 1, 2005; 78(2): 345 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. C. Fertuck, J. E. Eckel, C. Gennings, and T. R. Zacharewski Identification of temporal patterns of gene expression in the uteri of immature, ovariectomized mice following exposure to ethynylestradiol Physiol Genomics, October 17, 2003; 15(2): 127 - 141. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Barber and J. W. Pollard The Uterine NK Cell Population Requires IL-15 but These Cells Are Not Required for Pregnancy nor the Resolution of a Listeria monocytogenes Infection J. Immunol., July 1, 2003; 171(1): 37 - 46. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. N. Sferruzzi-Perri, S. A. Robertson, and L. A. Dent Interleukin-5 Transgene Expression and Eosinophilia Are Associated with Retarded Mammary Gland Development in Mice Biol Reprod, July 1, 2003; 69(1): 224 - 233. [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 |
