Full text of DOI:10.1093/humupd/6.1.16

Human Reproduction Update 2000, Vol. 6 No. 1 pp. 16–27
© European Society of Human Reproduction and Embryology
Endometrial leukocytes and menstruation
Lois A.Salamonsen1 and Louise J.Lathbury
Prince Henry’s Institute of Medical Research, PO Box 5152, Clayton, Victoria 3168, Australia
Received on May 7, 1999; accepted on September 9, 1999
This review examines evidence supporting the concept that menstruation occurs as a result of an inflammatory process. In
the endometrium, leukocyte numbers rise in the late secretory phase following the fall in serum progesterone
concentrations. It is postulated that products released following activation of these leukocytes are critically important for
menstruation. Mast cells, eosinophils, neutrophils and macrophages in particular are involved. Endometrial granular
lymphocytes may also play a role, although their increase in numbers is somewhat earlier during the menstrual cycle than
that of the others, suggesting perhaps a primary role in embryo implantation. Leukocyte products include a range of
proteases, chemokines and cytokines which in concert result in focal production and activation of matrix
metalloproteinases by endometrial cells and the subsequent breakdown of tissue that characterizes menstruation.
Regulation of leukocyte entry, proliferation, differentiation and activation within the endometrium is not yet well
understood, although both chemokines and cytokines produced locally by endometrial cells are clearly implicated. The role
of progesterone in regulating these events is still not understood although the lack of progesterone receptors on
endometrial leukocytes suggests indirect actions.
Key words: leukocyte/menstruation/chemokine/matrix metalloproteinase/progesterone
of 5–6 days. Re-epithelialization occurs simultaneously with the
tissue destruction and these processes are followed by regenera-
tion of the stromal components. This remarkable tissue remodel-
ling in an adult tissue is unparalleled in any other organ.
TABLE OF CONTENTS
Menstruation
Cellular composition of human endometrium
Endometrial leukocytes
Monocytes/macrophages
Regulation of leukocyte entry into endometrium
Leukocyte proliferation
16
17
17
18
20
22
23
23
23
25
25
25
25
The entire cyclical series of events in the endometrium is
orchestrated by the ovarian steroid hormones. Menstruation is
initiated by the fall in oestrogen and progesterone that results
from the demise of the corpus luteum at the end of a non-fertile
cycle. It can be mimicked by withdrawal of exogenous
hormones (as in the case of oral contraceptive use) or by admin-
istration of progesterone receptor (PR) antagonists, e.g. mifepris-
tone at the appropriate time of the cycle (Fraser, 1997).
However, although oestrogen and progesterone fall rapidly with
corpus luteum degeneration in all mammals with oestrous
cycles, only women and some Old World primates menstruate.
Thus, there must be features peculiar to the primate
endometrium that regulate menstruation.
Classic concepts of the mechanism of menstruation, which
remain the current dogma, originate from the work of Markee
(1940) who examined the morphological changes in autologous
endometrium transplanted to the anterior eye of the rhesus
monkey. He observed that prior to menstruation there was vaso-
constriction of the spiral arterioles followed by vasodilatation
Leukocyte activation
Role of leukocytes in menstruation
Matrix metalloproteinases
Leukocytes and tissue regeneration
Conclusions
Acknowledgements
References
Menstruation
Menstruation is the shedding of cell debris accompanied by
uterine bleeding which occurs in the endometrium at the end of
each normal menstrual cycle. It arises from the disintegration of
the superficial or functionalis layer of the endometrium and
results in almost complete loss of the functionalis over a period
1To whom correspondence should be addressed at: Prince Henry’s Institute of Medical Research, PO Box 5152, Clayton, Victoria 3168, Australia.
Phone: 61 3 9594 4373; Fax: 61 3 9594 6125; Email: lois.salamonsen@med.monash.edu.au

Endometrial leukocytes and menstruation
Endometrial leukocytes
17
and postulated that the tissue destruction occurred by necrosis
resulting from anoxia. However, overall endometrial blood flow
is not significantly reduced just prior to or during menstruation
(Fraser and Peek, 1992), and the occurrence of widespread
anoxia has not been substantiated in human endometrium. In
addition, it has been demonstrated that although the onset of
bleeding is the first outward sign of menstruation, it is clearly not
the first important event in the process. During the late luteal
phase of the cycle, widespread degeneration in the basal lamina
supporting the decidualized endometrial cells and the endothe-
lium of blood vessels has been reported (Roberts et al., 1992).
Furthermore, scanning electron microscopy has revealed small
lesions in the luminal epithelium on day 28 of the cycle, the
appearance of these being followed by very rapid but incomplete
degeneration of the functionalis layer (Ludwig and Spornitz,
1991).
The concept that menstruation arises as a result of an inflam-
matory process was first postulated by Finn (1986) and has
gained considerable credibility from a large body of data
published during the past decade. The hypothesis was originally
based on a number of features of endometrium during the late
secretory phase: the presence of tissue oedema, the influx of
migratory cells and the presence of decidual cells which have
some features representative of granulation tissue fibroblasts.
The entry of inflammatory cells into the endometrium, their
release of potent mediators, possible control mechanisms for
these cells, and the potential for the mediators to initiate tissue
destruction at menstruation will be reviewed in this article.
Leukocytes in the endometrium have been largely identified by
immunostaining or fluorescence-activated cell sorting (FACS)
analysis of specific surface antigens although some histochem-
ical stains (e.g. phloxine tartrazine for granular lymphocytes) are
still widely used. Caution must be applied when using these
techniques as variations in fixation procedures can substantially
alter the numbers of cells staining and probably account for
differences between reports from individual laboratories. Such
discrepancies may relate also to the difficulties in cell counting
when cells are not distributed evenly through the tissue (for
example, T cells; Starkey et al., 1991), where cells are so
numerous that it is difficult to score each separately, and where
there is variability in the morphology of the tissue, particularly
the degree of oedema and the density of the glands in the section.
There may also be variability in cell distribution in the basalis
compared with the functionalis. Studies in our laboratory and
others correct for at least some of these anomalies by expressing
leukocyte numbers per 1000 stromal cells and by assessing
distribution between cellular compartments (Vincent et al.,
1999). Evidence is also emerging that at any one time, subsets of
each cell type, represented by different phenotypes, are present
in the endometrium and that apparently similar cells in the
endometrium can be functionally different from those in the
blood of the same individual (Shi et al., 1995; White et al.,
1997). Thus at least some of the cells which traffic into the
endometrium must undergo phenotypic changes in response to
the new local environment. Most of the phenotypic changes so
far recorded relate to regulatory molecules rather than surface
proteins and are detailed in Table I.
Cellular composition of human endometrium
Neutrophils
The cellular composition of the human endometrium is
dynamic, altering almost on a daily basis in parallel with the
hormonal changes during the normal menstrual cycle. The first
event following menstruation is re-epithelialization, initiated
from the stem cells in the glands in the basalis layer and possibly
also from residual rafts of luminal epithelium. Subsequently,
throughout the proliferative phase of the cycle, both stromal
fibroblasts and cells associated with blood vessels undergo
mitosis until the full thickness of the endometrium is restored.
During the secretory phase of the cycle, differentiative changes
occur in many of these cell types including the decidualization of
some stromal cells. In the absence of an embryo, and following
the demise of the corpus luteum, there is a diminution in the
thickness of the endometrium, thought to result from loss of
water, and menstruation follows. Discussion of endometrial cell
biology generally ignores the population of lymphomyeloid
cells. The total and relative numbers of these cells vary with the
different phases of the cycle and are at greatest abundance during
the premenstrual phase (Klentzeris et al., 1992; Loke and King,
1995; Salamonsen and Woolley, 1999).
Neutrophils are the most abundant leukocytes in the human
immune system. They contain specific secretory granules which
store a heterogeneous range of regulatory molecules and which
differ with the stage of neutrophil maturation. These cells are
closely associated with tissue damage in inflammatory disor-
ders. Neutrophils have been identified in endometrial tissue by
their morphology and also by immunolocalization of the
neutrophil-specific protease, elastase. Endometrial neutrophils
bright
+
+
have also been defined as CD11b
, CD66b and CD16
(Yeaman et al., 1998). During most of the cycle, neutrophils are
barely detectable in normal endometrium but the numbers rise
dramatically perimenstrually, making up 6–15% of the total cell
number in the tissue at this time (Salamonsen and Woolley,
1999). Neutrophils have also been identified in increased
numbers in areas of endometrial breakdown in patients treated
with high-dose oral progestins (Song et al., 1998) or implanted
®
levonorgestrel (Norplant ; The Population Council, New York,
NY, USA) (Vincent et al., 1999), where they reach densities
similar to those seen in menstrual endometrium. There appears
to be more than one neutrophil phenotype in endometrium as

18
L.A.Salamonsen and L.J.Lathbury
Table I. Endometrial leukocyte phenotype
Leukocyte
Neutrophil
Abundance:
days 26–28a
Surface phenotype
Enzyme content
Other proteins
Actions
U
6–15%
CD11bbright CD66b+
CD16+
elastase,
MMP-9,
IFN-γ,
inhibin βA
MT1-MMP
ECP-1, ECP-2,
MMP-9
Eosinophil
3–5%
U
eotaxin
U
U
Macrophage
6–15%
CD68+
MMP-9
IL1-receptor antagonist,
inhibin βA
U
IFN-γ
Mast cell
T lymphocyte
3–5%
1–2%
U
tryptase, chymase
U
U
CD3+, CD4+/
cytolytic
CD8+ICAM-1+,
VLA-1+, CD69+HLA-
DR,DP,DQ+
eGL
6–5%
CD56bright, CD16-,
CD2+, CD122+,
CD49a,d+CD69+, HLA-
DR+, LFA-1+, CD3-
perforin, granzyme A,
MT1-MMP
TIA-1
cytolytic, proliferate in
response to IL-2
MMP-9 = matrix metalloproteinase-9; MT1–MMP = membrane type-1–matrix metalloproteinase; IFN-γ = interferon γ; ECP = eosinophil cationic
protein; IL-1 = interleukin-1; TIA-1 = T cell intracytoplasmic antigen-1; eGL = endometrial granular leukocyte; U = Unknown.
aExpressed as a percentage of total endometrial cells at this time in the normal menstrual cycle. From Salamonsen and Woolley (1999).
some, but not all of the neutrophils are immunopositive for
matrix metalloproteinase (MMP)-9, (Vincent et al., 1999),
marked phenotypic heterogeneity and respond to a wide range of
signals depending on the immediate tissue microenvironment.
These cells are an integral part of the immune response in
chronic inflammatory lesions and are found at sites of active
activin β , which has likely functions in cellular differentia-
A
tion or apoptosis (Leung et al., 1998) and membrane-type
(MT)1–MMP which activates latent MMP-2 (Zhang et al.,
2000a). Interferon (IFN)-γ has also been identified in intra-
epithelial neutrophils in human endometrium (Yeaman et al.,
1998). IFN-γ has a role in macrophage activation (Haddad et al.,
1997), suggesting at least one potential interaction between adja-
cent leukocytes at menstruation.
+
tissue remodelling. Macrophages (CD68 ) have been detected in
endometrium throughout the menstrual cycle with an increase in
numbers from the proliferative through to the menstrual phase
(Bonatz et al., 1992). On days 27–28, their numbers are similar
to those of neutrophils, comprising between 6–15% of the total
cells in the functionalis at this time (Salamonsen and Woolley,
1999). The cells are scattered throughout the tissue, with some
aggregates and some concentration around endometrial glands
(Song and Fraser, 1995). Increased numbers of macrophages are
also found in endometrium from Norplant users (Clark et al.,
1996) particularly in tissue from those women with abnormal
uterine bleeding. Endometrial macrophages show phenotypic
differences, and subtypes have been described with respect to
Eosinophils
Eosinophils function as effector cells that play an important role in
the pathogenesis of late-type inflammatory reactions (Bousquet
et al., 1990). Eosinophils have been detected in human
endometrium by immunolocalization of eosinophil cationic
proteins (ECP1 and ECP2). Like neutrophils, they are absent from
normal endometrium during most of the cycle but immediately
prior to menstruation there is a dramatic increase in their numbers
in the tissue. Most of them are found as aggregates and the extra-
cellular location of the ECP suggests activation of the cells
(Jeziorska et al., 1995). Some, but not all of the eosinophils are
positive for MMP-9 (Jeziorska et al., 1995) (Vincent et al., 1999).
As expected, endometrial eosinophils immunostain positively for
both the chemokine eotaxin and for its receptor CCR3 (Zhang
et al., 2000b). These cells are negative for both MT1–MMP
expression of MMP-9 (Jeziorska et al., 1995), activin β (Leung
B
et al., 1998) and MT1-MMP (Zhang et al., 2000a). Interleukin
(IL)-1 receptor antagonist, a component of the IL-1 system that
prevents binding of IL-1 to its type 1 receptor, has also been
detected in these cells (Tabibzadeh and Sun, 1992).
Mast cells
Mast cells secrete many vasoactive, nociceptive and proinflam-
matory molecules and participate in inflammation through
vasodilation and enhancing leukocyte infiltration as well as
causing direct tissue damage by their release of proteases. Mast
cells are present in the endometrium throughout the menstrual
cycle. They are detected by immunostaining for the two mast
cell-specific serine proteinases, tryptase and chymase, and the
extracellular location of the enzymes is indicative of activation.
(Zhang et al., 2000a) and activin β (Leung et al., 1998).
A
Monocytes/macrophages
Resident macrophages in tissues arise by differentiation from
monocytes which have migrated from the blood. They display

Endometrial leukocytes and menstruation
19
Figure 1. Cascade of matrix metalloproteinase (MMP) activation postulated to be initiated by mast cell degranulation. (Reprinted from Salamonsen and
Woolley 1996, with permission).
Endometrial mast cell activation corresponds closely to the
phases of tissue oedema and is most marked immediately prior
to menstruation (Jeziorska et al., 1995). Mast cells fall into two
well-known phenotypic subtypes. Mucosal mast cells are posi-
tive for tryptase but not chymase, whereas connective tissue
mast cells contain both enzymes. Both of these subtypes have
been detected in endometrium, with regional differences in loca-
tion. Those in the basalis region express both enzymes whereas
those in the functionalis are positive only for tryptase (Jeziorska
et al., 1995). While it is the functionalis that is shed at menstrua-
tion, there is considerable tissue destruction at the basalis–
functionalis interface and it is likely that both enzymes
contribute to menstruation. Tryptase, and to a lesser extent
chymase, can play an important role in establishing a cascade of
MMP activation (Figure 1) and this could represent a critical
function for these cells at menstruation. Mast cells also produce
histamine, heparin, arachidonate products and a variety of multi-
functional cytokines and growth factors, all of which have
marked effects on endothelial cell function and locally induced
oedema. The biological consequence of mast cell degranulation
within the endometrium is likely to be extremely complex
because it appears that the synthesis and release of specific
cytokines and arachidonate products may occur at specific
stages post activation. Thus a complex sequence of signals can
be produced within the localized domain of the activated mast
cell (Norrby and Woolley, 1993).
as basal lymphoid aggregates, scattered throughout the stroma
and in intraepithelial sites (Loke and King, 1995; Song and
Fraser, 1995). Interestingly, the CD4 (T helper phenotype) to
CD8 (cytotoxic T cell phenotype) ratio in endometrium is
inverted when compared with peripheral blood T cells (66%
+
+
CD8 to 33% CD4 in endometrium). The cells are cytolytically
active during the proliferative phase of the cycle but this activity
is diminished during the secretory phase. Therefore, proges-
terone may down-regulate this property (White et al., 1997),
although whether this activity is restored when progesterone
concentrations fall or whether it exists at menstruation remain to
be established. Neither is it known whether the alteration in
phenotype around mid-cycle results from a change in the activity
of existing cells, the deletion of a minor subset of these cells or
the selective migration of specific cell subsets into or away from
+
the endometrium. The observation that peripheral blood CD3
cells remain cytolytically active throughout the menstrual cycle
supports the concept of differentiation within the tissue. Subsets
of endometrial T cells express MMP-9 (Vincent et al., 1999) and
+
a proportion of the intraepithelial CD3 T cells are IFN-γ posi-
tive (Yeaman et al., 1998).
+
CD45RA B lymphocytes are also detected in low numbers
during the entire cycle and are present in peri-menstrual tissue in
clusters among the stromal cells (Salamonsen and Woolley,
1999). Whether or not these cells produce immunoglobulin
within the tissue is not known.
B and T lymphocytes
Endometrial granular lymphocytes (NK cells)
+
CD3 T cells are present in the endometrium throughout the
Granulated cells were recognized in non-pregnant endometrium
as early as 1921 (Weill, 1921). Subsequently, these phenotypi-
cally unusual CD56 , CD2 , CD38 , CD16 , and CD3
lymphocytes, commonly known as endometrial granular
lymphocytes (eGL) (Bulmer et al., 1991; King et al., 1989;
cycle and their numbers increase prior to menstruation.
However, their total numbers are much less than those of other
leukocytes (only 1–2% of the total lymphomyeloid cells)
(Salamonsen and Woolley, 1999). They are found in three sites:
+
+/_
+
_
_

20
L.A.Salamonsen and L.J.Lathbury
Klentzeris et al., 1992) have been shown to be among the most
numerous haematopoietic cells in perimenstrual endometrium.
They have been identified by staining with phloxine–tartrazine
and by immunolocalization. Importantly, the CD56 antigen is
not readily detectable in formalin-fixed tissue, unless an antigen
retrieval step is used to expose the epitope. The CD43 antigen
which survives this fixation is often utilized as an alternative,
although it also detects some macrophages and T cells. Very few
eGLs are present during the proliferative phase but numbers
increase during the secretory phase and have been assessed vari-
ously as being up to 15% (Salamonsen and Woolley, 1999) or
25% (Bulmer et al., 1991) of the total number of cells in the
stroma perimenstrually. The cells are found scattered throughout
the stroma and in intraepithelial locations (Song and Fraser,
1995). Due to their expression of CD56, these cells are said to be
of the natural killer (NK) lineage (King et al., 1991) but their
exact relationship to the CD56 CD16 NK cells in peripheral
blood is not yet clear (King et al., 1991). From the late prolifera-
tive phase on, eGL have cytotoxic activity comparable to that of
peripheral blood NK cells (Jones et al., 1998b), suggesting they
may be involved in protecting against infection. In support of
this notion, the proportion of eGL expressing the activation anti-
gens CD69 and human leukocyte antigen (HLA)-DR was
highest in the proliferative phase and decreased during the
menstrual phase (Kodama et al., 1998). Further, this suppression
of activation antigens in secretory endometrium suggests local
functional regulation at this time. eGL contain perforin,
granzyme A and T cell intracytoplasmic antigen (TIA)-1 gran-
ules (King et al., 1993) and some stain positively for MT1–
MMP (Zhang et al., 1999a). All of these factors could contribute
to cell and tissue degradation. Whether eGLs have a role in
menstruation is not known: it may be that their role at the end of
the cycle is one of preparation for pregnancy rather than one
specifically directed towards the process of menstruation. This
hypothesis was tested in part in a recent in-vivo study in women
to whom progesterone was administered and then withdrawn;
the eGL numbers in endometrium did not alter following with-
drawal (Critchley et al., 1999). However, additional studies are
required to finally support or to disprove this concept.
large family of nuclear transcription factors which regulate the
expression of numerous genes. Two forms or subtypes of each
receptor have been identified with the physiological significance
of each form currently unknown. The predominant form of ER
in all cell types in the uterus throughout the menstrual cycle is
ERα, although weak expression of mRNA for the β-subtype is
also detectable (Matsuzaki et al., 1999). At the time of menstru-
ation, expression of ERα in endometrial cells is at a nadir. The
isoforms of PR are known as PRA and PRB. In human
endometrium, PRA is present in glandular and stromal cell
nuclei during the proliferative phase but is present only in
stromal cells by the end of the secretory phase (Wang et al.,
1998). PRB is also present in both cellular compartments in the
proliferative phase and absent in the late secretory phase. Thus
those PR which have been localized in cells around the vessels in
the secretory phase are probably PRA and are likely to mediate
any effects of progesterone in the endometrium during the late
secretory phase.
+
+
Whether or not human endometrial leukocytes express either
ER or PR has been a subject of some controversy, not helped by
the discovery of the different receptor subtypes. The most recent
+
immunohistochemical data shows that uterine CD45 leuko-
cytes (including endometrial lymphocytes, macrophages and T
cells) express neither ER nor PR (King et al., 1996; Stewart et
al., 1998) and thus the effects of oestrogen and progesterone on
these cells are likely to be indirect. In rodents, steroid hormone
receptors have been demonstrated in a variety of immune cells.
Despite their very low binding affinities, numerous effects of the
steroids on these cells in rodents support direct actions (for
review see Miller and Hunt, 1996). It still remains to be conclu-
sively established whether or not any of the subsets of human
endometrial leukocytes express low concentrations of ER or PR
and whether there are any direct responses of these cells to the
hormones.
Chemokines
The influx of migratory cells into the endometrium is likely to be
regulated by chemokines, potent chemoattractant cytokines that
promote the recruitment of multiple lineages of leukocytes and
which act via specific receptors on attracted cells (Kunkel et al.,
1995). The chemokine superfamily has been divided into four
groups, CC, C, CXC and CXXXC. The CC and CXC chemok-
ines are the best studied and are differentiated both structurally
and by a general distinction in biological properties: most CXC
chemokines are chemoattractants for neutrophils (and to some
extent lymphocytes) but not monocytes, whereas CC chemok-
ines appear to attract monocytes, lymphocytes, basophils and
eosinophils but not neutrophils.
Regulation of leukocyte entry into endometrium
It is not yet known whether in-situ proliferation or migration
from the peripheral circulation is responsible for the increase in
leukocyte subpopulations in the late luteal phase but it is likely
that both of these processes are relevant. Whether these proc-
esses are driven directly by ovarian hormones or whether they
are mediated via other factors such as chemokines or adhesion
molecules is still not clear.
Chemokines mediate their effects via interactions with G
protein-linked receptors on the surface of target cells. There are
a large number of these receptors with most being promiscuous
in their ligand binding. A relatively new concept is that leuko-
cytes can respond sequentially to chemokines in a multi-step
Steroid hormones
The female steroid hormones, oestrogen and progesterone, act
via specific receptors (ER and PR respectively) that belong to a

Endometrial leukocytes and menstruation
21
Table II. Chemokines in human endometrium
Chemokine
RANTES
Chemoattractant for:
Cellular location
stroma
Reference
Hornung et al., 1997
Eosinophils, basophils,
monocytes, activated T cells, NK
cells
IL-8
Neutrophils, T cells, basophils
epithelium,
Arici et al., 1998
perivascular cells
Critchley et al., 1994;
Jones et al., 1997
MCP-1
Monocytes, activated T cells, NK epithelium
Jolicoeur et al., 1998
Hampton and Salamonsen,
unpublished observations
Jones et al., 1997
cells, basophils
stroma
perivascular cells
MCP-2
eotaxin
monocytes,activated T cells, NK epithelium
Hampton and Salamonsen,
unpublished observations
cells, basophils, eosinophils
stroma
perivascular cells
epithelium
eosinophils, basophils, T
Zhang et al., 2000b
lymphocytes that co-localize with eosinophils
eosinophils perivascular cells,
decidualized cells
RANTES = regulated upon activation, normal T cell expressed and secreted; IL-8 = interleukin-8;
MCP = monocyte chemotactic protein.
navigation mode with a corresponding sequential expression of
different chemokine receptors (Foxman et al., 1997).
(Critchley et al., 1994; Jones et al., 1997). However, in data from
another laboratory, the strongest staining was detected in the
surface epithelium and glands, with less intense staining in the
walls of blood vessels and no staining in the stroma (Arici et al.,
1998). The epithelial staining intensity increased across the cycle
from proliferative to late secretory endometrium. Of particular
relevance is a study of in-vivo progesterone withdrawal in
women, in which endometrial IL-8 protein rose significantly at
48 h after withdrawal (Critchley et al., 1999), suggesting that the
fall in progesterone late in the cycle could contribute to a rise
prior to menstruation. Cell culture studies using passaged
endometrial stromal cells demonstrated increased production of
IL-8 in the presence of IL-1 and TNFα (Arici et al., 1996) and a
slight decrease with transforming growth factor (TGF)β
(Arici et al., 1996). As stromal staining is not observed in
situ, whether in-vitro stromal production of IL-8 is an artefact
of the cell culture conditions remains to be established.
The major known biological actions of the monocyte chemo-
tactic proteins (MCP) are monocyte activation and recruitment
of these cells into sites of inflammation. Four structurally related
molecules (MCP-1, MCP-2, MCP-3, MCP-4) act via different
receptors and may have some functional differences (Van
Coillie et al., 1997). MCP-1 at least can also activate basophils,
mast cells and NK cells (Allavena et al., 1994; Feliciani et al.,
1995). MCP-1 has been immunolocalized in human
endometrium in several studies. In one study, it was localized to
perivascular cells with intense immunostaining in the early–late
proliferative and late secretory phases (Jones et al., 1997). In
contrast, other investigators have localized MCP-1 primarily in
epithelial cells (Jolicoeur et al., 1998). In our laboratory, MCP-1
immunostaining was extremely variable between individual
endometrial samples even on the same day of the cycle, with
only 77% of tissues showing epithelial staining, 52% of tissues
Regulation of chemokine production is still only partly under-
stood but typical members of the major groups are stimulated by
IL-1 and inhibited by glucocorticoids (Ben-Baruch et al., 1995).
Given the structural similarity between progesterone and gluco-
corticoids and their receptors, it is highly likely that progesterone
could act in progesterone-dependent tissues in a similar manner
to glucocorticoids in other tissues. Certainly, in mice with
targeted disruption of the progesterone receptor, homozygous
offspring have inflammatory abnormalities of the uterus (Lydon
et al., 1995).
The chemokines which have been examined in the human
endometrium are summarized in Table II. In some cases, data
regarding the cellular localization is somewhat conflicting,
possibly reflecting variability between recognition sequences of
the different antisera used for immunolocalization or differences
between the sensitivity of different immunohistochemical proto-
cols.
RANTES (regulated upon activation, normal T cell expressed
and secreted), acts as a chemoattractant for several leukocyte
subsets including monocytes and activated T cells. mRNA and
protein for this chemokine have been localized primarily to the
stromal compartment of human endometrium, with glandular
epithelium appearing mostly free of the antigen (Hornung et al.,
1997). However, premenstrual endometrium was not examined.
In vitro, both IL-1 and tumour necrosis factor (TNF)α, but not
oestrogen or progesterone stimulated RANTES production by
purified endometrial stromal cells.
IL-8 has neutrophil chemoattractant/activating and T cell and
fibroblast chemotactic activities. In one study of endometrial
tissue this chemokine was localized to perivascular cells of blood
vessels, increasing significantly in the late secretory phase

22
L.A.Salamonsen and L.J.Lathbury
demonstrating staining in the stroma and 34% of the tissues with
vascular staining. No variation across the menstrual cycle was
observed (A.L.Hampton and L.A.Salamonsen, unpublished
observations). Secretion studies on stimulated endometrial
stromal and epithelial cells in culture (Arici et al., 1995) and in-
situ hybridization studies (Jolicoeur et al., 1998) also support a
primary epithelial source of MCP-1. MCP-2, a related chem-
okine, is likewise expressed in human endometrium with a
predominantly epithelial localization. This chemokine is also
detectable in uterine secretions, suggesting a potential role other
than the attraction of monocytes (A.L.Hampton and L.A.Salam-
onsen, unpublished observations). In other tissues, production of
MCP-1 is inducible and can be stimulated in vitro by a number
of different cytokines including IL-1, TNFα and interferon γ,
with combinations of stimuli resulting variously in additive or
synergistic induction of MCP-2 compared with MCP-1. When
synergy occurs, MCP-1 and MCP-2 expression concentrations
can reach a comparable maximum (Struyf et al., 1998).
Eotaxin is one of the most selective of the chemokines, acting
primarily on eosinophils through the CCR3 receptor. In the
endometrium, eotaxin was detected in perivascular cells in the
late secretory phase (Zhang et al., 2000b), a time when eosi-
nophils are found in the tissue (Jeziorska et al., 1995; Poropatich
et al., 1987). However, this chemokine was also strongly
expressed by both luminal and glandular epithelial cells
throughout the entire menstrual cycle, and was detected in
epithelial and stromal endometrial cells in culture (Zhang et al.,
2000b). CCR3 is also strongly expressed in human
endometrium, predominantly in eosinophils, but also in epithe-
lium at all stages of the cycle and in perivascular cells in a
proportion of blood vessels (Zhang et al., 2000b). As CCR3 is
expressed strongly by the same cells that produce eotaxin, this
chemokine may act as a paracrine growth factor. It is important
to note that although eotaxin is relatively specific for eosinophils,
CCR3 also binds RANTES, MCP-3 and MCP-4.
the vasculature of the endometrium with respect to the expres-
sion of various adhesion molecules (Tabibzadeh et al., 1994).
The intercellular adhesion molecule (ICAM)-1 has been identi-
fied in the stroma of the functionalis during the menstrual phase
of the cycle ((Tawia et al., 1993; Thomson et al., 1999) and
evidence from in-vitro studies suggests this molecule may be
constitutively shed from the surface of endometrial stromal cells
(Somiglianna et al., 1996). ICAM-1 was also expressed by
+
CD3 cells in the lymphoid aggregates in the basalis layer and
was present on vascular endothelium throughout the cycle with
an overall peak in expression at menstruation (Tawia et al.,
1993) although this was not consistent across all vessel types. In
contrast, ICAM-2 expression was restricted to the vascular
endothelium, without change across the menstrual cycle
(Thomson et al., 1999). Platelet endothelial cell adhesion mole-
cule (PECAM) was also intensely stained in the stroma during
menstruation and in endothelial cells of all vessel types (Tawia
et al., 1993). VCAM-1 and E-selectin appeared in the stromal
cells in the upper functionalis in the secretory phase (Tabibzadeh
et al., 1994). Furthermore, immunoreactivity for VCAM-1 was
the distinguishing feature that separated aggregated lymphoid
cells in the tissues from non-aggregated cells, suggesting a
function in adhesion of these cells to one another. It is likely that
unique cell and site-specific expression of adhesion molecules in
the endometrium may account at least in part for the distinct
distribution of leukocytes in this tissue
Leukocyte proliferation
Lymphoid cells may enter the endometrium from the blood as
appears likely for such cells in the decidua of pregnancy
(Marzusch et al., 1993). However, there is some evidence that
their increase in the late secretory phase is due to proliferation in
+
+
+
+
situ. CD45 , CD3 T cells, CD11c macrophages and CD56
eGLs express the proliferation markers Ki67 and/or BrdU within
endometrium throughout the menstrual cycle with a marked
increase in the secretory phase (Pace et al., 1989; Tabibzadeh,
1990; Jones et al., 1998a). However, in perimenstrual tissue
eGLs undergo morphological changes indicative of apoptosis,
although their high expression of bcl-2 (which inhibits apop-
tosis) at this time suggests they are still functional (Jones et al.,
1998a). eGLs are particularly numerous in decidual tissue in
pregnancy (Ritson and Bulmer, 1989; King et al., 1991) and
these display a repertoire of killer inhibitory and activatory
receptors (Hilby et al., 1997). Both decidual eGLs and those in
cycling endometrium proliferate in response to IL-2 (Jones et al.,
1998b; Searle et al., 1999): however, there is no evidence for the
presence of IL-2 within normal endometrium (Jokhi et al., 1994;
King et al., 1995; Saito et al., 1993). In-vitro studies have
The overall picture emerging from chemokine studies in the
endometrium is that chemokines could be involved in control-
ling the influx of leukocytes prior to menstruation. Once the cells
have entered the tissue the epithelial chemokines may attract
them to the subepithelial space and could also ‘prime them’ for
subsequent responses. The experimental data reviewed here also
strongly suggests additional paracrine roles for these pleiotropic
molecules in endometrial remodelling. Known chemokine
actions of possible relevance include regulation of precursor cell
cycling and differentiation, and growth factor and angiogenic
activities (Schall, 1994). Further studies are required to elucidate
the precise biological activities of the chemokines of epithelial
cell origin in this tissue and their paracrine targets.
+
demonstrated that CD56 cells also proliferate in the presence of
Adhesion molecules
endometrial stromal cells and progesterone (Inoue et al., 1996),
suggesting that progesterone-dependent stromal factors may be
mitogenic for these cells. Furthermore, proliferation of periph-
eral blood monocytes in culture was influenced by a soluble
Adhesion molecules are necessary for the attachment of leuko-
cytes to endothelium and for their extravasation and trafficking
through tissues. Considerable heterogeneity has been shown in

Endometrial leukocytes and menstruation
23
factor produced by uterine epithelial cells in response to
cytokines (Prabhala et al., 1998). Thus, in-situ proliferation,
regulated by locally produced factors, is apparently one mecha-
nism by which the pre-menstrual increase in these cells is
achieved.
phases, coincident with the increased numbers of leukocytes in
the endometrium (reviewed in Tabibzadeh and Sun, 1992;
Salamonsen and Woolley 1996) and are likely to effect leuko-
cyte differentiation and activation. Other local endometrial
mediators, whose production appears to be steroid-hormone
modulated in the human and which have potential but as yet
undefined interactions with immune cells in the endometrium,
include leukaemia inhibitory factor, prostaglandins, TGFβs,
endothelin and nitric oxide.
Leukocyte activation
The presence of leukocytes in a tissue does not necessarily mean
that they continually contribute to the function of that tissue.
Many leukocytes can remain inactive in a tissue; their activation
results in the release of a plethora of bioactive regulators. For
example, mast cells within the endometrium are inactive during
most of the cycle, and phases of activation (as detected by extra-
cellular tryptase) coincide with tissue oedema, particularly
immediately prior to menstruation (Jeziorska et al., 1996). Like-
wise, eosinophil activation (as evidenced by extracellular eosi-
nophil cationic proteins) also occurs during the perimenstrual
phase of the cycle (Jeziorska et al., 1996). T cell activity may be
related to local secretion of cytokines such as IFNγ, which is
apparent in lymphoid aggregates in the stratum basalis but less
consistently in functionalis throughout the cycle (Stewart et al.,
1992). Other markers of T cell activation (CD69 and DR) are
similar in both the proliferative and secretory phases and suggest
that in the endometrium, T lymphocytes are in a state of recent
and persistent activation (Chen et al., 1995). Neutrophil elastase
has also been found extracellularly, indicating neutrophil activa-
tion prior to menstruation, although extracellular IFNγ origi-
nating from neutrophils was located immediately beneath the
luminal epithelium through all stages of the cycle (Yeaman et
al., 1998). There is no information regarding macrophage activa-
tion at menstruation though this might be expected. Evidence is
now emerging that the regulation of leukocytes can also occur
via the inhibition of activation, and multiple inhibitory receptors
have been identified on mast cells, macrophages and NK cells
(Katz and Austen, 1997). The mechanisms by which cells are
either activated or inhibited from activation in the endometrium
are completely unknown, although it is likely that there are
considerable interactions between cell types.
Steroid hormones may act directly on leukocyte differentia-
tion and activation (Miller and Hunt 1996). For example in
mouse uterine mast cells oestrogen promotes, and progesterone
inhibits production of inducible nitric oxide synthase and TNFα
while oestrogen has no effect on macrophages, in which proges-
terone inhibits production of the same two substances (for
review, see Hunt et al., 1997). Indirect actions of steroid
hormones via production of paracrine mediators by endometrial
epithelial and stromal cells are also likely. For example, IFNγ
which can modulate haematopoietic cell maturation, differentia-
tion, activation and apoptosis is produced by mouse uterine
epithelial cells during oestrus, when oestrogen is predominent.
In human endometrium, many cytokines, including IL-1 and
TNFα are produced in endometrial stromal or epithelial cells
reaching peak production during the mid- to late secretory
Role of leukocytes in menstruation
Leukocytes have the potential to contribute both directly and
indirectly to menstruation. A number of their products may
directly promote tissue degradation, while others may contribute
indirectly through the activation of latent enzymes produced by
adjacent cells, or by stimulating adjacent cells to produce other
biologically active molecules. Some of the proteolytic enzymes
produced by leukocytes that may play a role in menstruation are
listed in Table III and are discussed in further detail below.
Matrix metalloproteinases
Matrix metalloproteinases (MMP) are enzymes that degrade
components of both interstitial and basement membrane extra-
cellular matrix, their synthesis being negligible in normal
connective tissue (Birkedal-Hansen et al., 1993). The MMP can
be divided into four major subfamilies, collagenases (MMP-1,
MMP-8), gelatinases (MMP-2, MMP-9), stromelysins
(including MMP-3) and MT-MMP. Most MMP are secreted as
latent zymogens that can be activated in vitro by a number of
natural proteases including other MMP. Their expression is
differentially regulated by tissue and cell-type specific mecha-
nisms involving cytokines, steroid hormones (including proges-
terone) and other regulatory molecules. There is now strong
evidence for a role for these enzymes in menstruation
(Salamonsen and Woolley, 1996; Salamonsen and Woolley,
1999). Most of the MMP so far examined are produced by
endometrial stromal cells, although MMP-7 and MMP-2 are
epithelial cell products. Furthermore, a number of leukocytes
Figure 2. Actions of neutrophil enzymes on matrix metalloproteinases
(MMPs)/tissue inhibitors of matrix metalloproteinase (TIMPs) in
solution.

24
L.A.Salamonsen and L.J.Lathbury
Table III. Leukocyte proteases of potential relevance to menstruation
Leukocyte
Mast cell
Protease
tryptase
chymase
Potential substrate
activates proMMP-3, uPA
activates proMMP-1, proMMP-3,
inactivates bradykinin
broad spectrum
chymotrypsin
plasminogen activator
elastase
MMP-8
MMP-9
plasminogen
Neutrophil
elastin, proteoglycans, collagens III, IV
collagens,I,III,VII, VIII, X
collagen IV, V, VII elastin, FN, LN
activates proMMP-2, proMMP-13,
degrades FN, tenascin, nidogen,
perlecan, collagen I,III
heparan sulphate proteoglycans
elastin, proteoglycans collagen III, IV
collagen I, III, VII, VIII, X
as above
MT1-MMP
heparanase
cathepsin G
MMP-1
Eosinophil
MMP-9
β glucuronidase
aryl sulphatase
MMP-9
proteoglycans
proteoglycans, GAG
as above
Macrophage
metalloelastase
elastase, collagen IV, LN, FN, VN,
heparin, chondroitin sulphates
as above
as above
as above
MT1–MMP
Plasminogen activator
MMP-2
T lymphocyte
eGL
MMP-9
MT1-MMP
as above
as above
MMP = matrix metalloproteinase; MT1–MMP = membrane type-1–matrix
metalloproteinase; FN = fibronectin; LN = laminin; VN = vitronectin;
GAG = glycosaminoglycan
produce MMP (see Table III). These and other factors produced
by leukocytes have the potential to stimulate the production of
latent MMP produced by adjacent cells or participate in MMP
activation. In particular, once active MMP-3 is present, a
cascade of MMP activation can be initiated, as seen in the case
of activation by mast cell tryptase (Figure 1) or neutrophil
elastase (Figure 2). The recent finding that MT1-MMP, which
activates latent MMP-2, is present in membranes of neutrophils
and eGL (Zhang et al., 2000b), would provide for local activa-
tion of MMP-2 around these cells. The actions of MMPs are
apposed by the binding of their active forms by specific tissue
inhibitors of MMPs, the TIMPs, which are abundant in human
endometrium (Zhang and Salamonsen, 1997).
heparin (which stabilizes tryptase activity). Other experiments
examined interactions within co-cultures of freshly prepared
peripheral blood neutrophils and endometrial stromal cells
(Lathbury and Salamonson, 1998). Conditioned medium from
neutrophils alone contained proMMP-9 and elastase, while
conditioned medium from co-cultures contained active MMP-9
and MMP-3 and fragments of degraded MMP-2. MMP-3 acti-
vation and MMP-2 degradation were not observed when the
neutrophil conditioned medium was pre-treated with anti-
elastase. Therefore it is likely that elastase is involved in the acti-
vation of MMP in the endometrium in the vicinity of activated
neutrophils.
Studies in our laboratory have examined the regulation of
matrix metalloproteinase production and activation by endome-
trial stromal cells in vitro. The human mast cell line (HMC-1)
produces the mast cell products tryptase, IL-1 and TNFα. When
co-cultured with endometrial stromal cells, HMC-1 cells stimu-
lated stromal cell proMMP-1 and proMMP-3 and to a lesser
extent proMMP-2 production, with increasing stimulation as
mast cell numbers increased (Zhang et al., 1998). These effects
were largely abrogated by pre-adsorption of the mast cell condi-
tioned medium with antisera against both IL-1 and TNFα.
Further, medium from mast cells activated latent MMP-3 and
MMP-1 in stromal cell conditioned medium in the presence of
Other degradative enzymes
Many of these enzymes are listed in Table III and only some will
be discussed here. The activation of MMP-3 is a critical step in
the cascade of MMP activation and any leukocyte products
which facilitate this process will aid in the generation of a full
complement of enzymes, capable of degrading all the compo-
nents of the extracellular matrix. Plasminogen activator,
produced by mast cells and macrophages, converts plasminogen
to plasmin, which activates both MMP-1 and MMP-3. These
two latent enzymes can also be activated by the mast cell-
specific enzymes tryptase and chymase. The neutrophil-specific

Endometrial leukocytes and menstruation
25
enzymes elastase and cathepsin G can activate MMP-3. Elastase
can also degrade elastin, proteoglycans and type III and type IV
collagens, all of which are found within endometrium. Macro-
phages at sites of inflammation produce a variety of cysteine,
serine and metalloenzymes which may be involved in tissue
degradation.
fragile endometrium which is predisposed to bleed, presumably
as a result of inadequate repair mechanisms. It is likely that
leukocytes will play an important role in this process as we now
understand they do in the corollary, menstruation.
Conclusions
There is now strong evidence for a role for leukocytes in the
normal biology of human endometrium, particularly in endome-
trial remodelling, implantation and menstruation. Due to the
marked differences in endometrial tissue and leukocytes
between species, and the difficulty in performing other than
correlative studies in the human, the critical studies providing
evidence for the contribution of these cells to the above functions
are not possible. Further information is required regarding
control of entry of the cells and/or their proliferation in the tissue,
regulation of their phenotype and activation, and the means by
which the changing steroidal milieu exerts its effects on these
parameters. While animal models may provide some informa-
tion, experimentation on human material will ultimately be
required to provide answers relevant to the human and which
may assist in treatment of endometrial disorders in women.
Other regulatory molecules with potential roles at
menstruation
Description of the plethora of regulatory molecules produced by
leukocytes is beyond the scope of this article and has been
reviewed by other authors, particularly with respect to mast cells
(Galli, 1993; Schwartz, 1994), macrophages (Nathan, 1987) and
eosinophils (Kroegel et al., 1994). They include cytokines,
vasoactive substances such as nitric oxide and prostaglandins
and reactive oxygen species, all of which may play important
roles at menstruation (Salamonsen et al., 1999).
Leukocytes and tissue regeneration
Repair of the endometrium begins as early as 36 h after the onset
of menstrual bleeding, while tissue breakdown is still in
progress, emphasizing the very focal nature of the degradative
and repair processes (Ludwig et al., 1990). Regeneration of
epithelium from stumps of glands occurs first, followed by
thickening of the stroma and mitosis in the endothelium which
begins around days 4–5. Given the abundance of leukocytes in
the tissue at this time it is highly likely that they contribute to the
process of regeneration. Certainly, macrophages are responsible
to some extent for the removal of detritus (Ludwig et al., 1990).
The process of wound healing, in general, appears to be initiated
by a wealth of cytokines and growth factors, and platelets recruit
inflammatory cells, particularly neutrophils and macrophages
(for review, see Martin, 1997), which produce these necessary
factors. In particular, macrophages at wound sites express
TGFα, fibroblast growth factors (FGFs), platelet-derived growth
factor (PDGF), vascular endothelial growth factor (VEGF),
Acknowledgements
Work cited from the author’s laboratory was funded by the
National Health and Medical Research Council of Australia
(Grant 971292), the National Institutes of Health (HD33233)
and the Special Programme of Research, Development and
Research Training in Human Reproduction of the World Health
Organization (Grant 96908) to LAS. LJL is supported by a
Perpetual Trustees Fellowship.
References
Allavena, P., Bianchi, G., Zhou, D. et al. (1994) Induction of natural killer cell
migration by monocyte chemotactic protein-1, -2 and -3. Eur. J. Immunol.,
24, 3233–3236.
Arici, A., MacDonald, P.C. and Casey, M.L. (1995) Regulation of monocyte
chemotactic protein-1 gene expression in human endometrial cells in
culture. Mol. Cell. Endocrinol., 107, 189–197.
Arici, A., MacDonald, P.C. and Casey, M.L. (1996) Modulation of the levels of
interleukin-8 messenger ribonucleic acid and interleukin-8 protein
synthesis in human endometrial stromal cells by transforming growth
factor-β1. J. Clin. Endocrinol. Metab., 81, 3004–3009.
Arici, A., Seli, E., Senturk, L.M. et al. (1998) Interleukin-8 in the human
endometrium. J. Clin. Endocrinol. Metab., 83, 1783–1787.
TGFβ and activin β (Martin, 1997), while products of other
B
leukocytes can activate the macrophages. Eosinophils in healing
wounds express TGFα and TGFβ (Todd et al., 1991; Wong
1
et al., 1993), while lymphocytes are capable of producing the
growth factors bFGF and leukocyte-derived growth factor
(Blotnick et al., 1994; Iida et al., 1996). Isoforms of activin are
involved in the proliferation of fibroblasts at a wound margin
(Hubner et al., 1996), while TGFβ encourages the synthesis of
new collagen-rich matrices (Eckes et al., 1996) and mast cell
tryptase can act as a mitogen for epithelial cells (Cairns and
Walls, 1996). Endometrial repair is likely to have similarities to
tissue repair in embryos, in which scarring is not found and
which uses many different regulatory molecules. Almost
nothing is known at this time of the cellular and molecular mech-
anisms of endometrial repair. There is a need to understand this
process, given that much abnormal uterine bleeding results from
Ben-Baruch, A., Michiel, D.F. and Oppenheim, J.J. (1995) Signals and
receptors involved in recruitment of inflammatory cells. J. Biol. Chem.,
270, 11703–11706.
Birkedal-Hansen, H., Moore, W.G.I., Bodden, M.K. et al. (1993) Matrix
metalloproteinases: a review. Crit. Rev. Oral Biol. Med., 4, 197–250.
Blotnick, S., Peoples, G.E., Freeman, M.R. et al. (1994) T lymphocytes
synthesize and export heparin-binding epidermal growth factor-like growth
factor and basic fibroblast growth factor, mitogens for vascular cells and
+
+
fibroblasts: differential production and release by CD4 and CD8 T cells.
Proc. Natl Acad. Sci. USA, 91, 2890–2894.
Bonatz, G., Hansmann, M.L., Bucholz, F. et al. (1992) Macrophage- and
lymphocyte-subsets in the endometrium during different phases of the
ovarian cycle. Int. J. Gynaecol. Obstet., 37, 29–36.

26
L.A.Salamonsen and L.J.Lathbury
Bousquet, J., Chanez, P., Lacoste, J.Y. et al. (1990) Eosinophilic inflammation
in asthma. N. Engl. J. Med., 323, 1033–1039.
Bulmer, J.N., Morrison, L., Longfellow, M. et al. (1991) Granulated
populations by the reverse-transcriptase polymerase chain reaction. J.
Immunol., 153, 4427–4435.
Jolicoeur, C., Boutouil, M., Drouin, R. et al. (1998) Increased expression of
monocyte chemotactic protein-1 in the endometrium of women with
endometriosis. Am. J. Pathol., 152, 125–133.
lymphocytes
in
human
endometrium:
histochemical
and
immunohistochemical studies. Hum. Reprod., 6, 791–798.
Cairns, J.A. and Walls, A.F. (1996) Mast cell tryptase is a mitogen for epithelial
cells. Stimulation of IL-8 production and intercellular adhesion molecule-1
expression. J. Immunol., 156, 275–283.
Chen, C.K., Huang, S.C., Chen, C.L. et al. (1995) Increased expressions of
CD69 and HLA-DR but not of CD25 or CD71 on endometrial T
lymphocytes of non-pregnant women. Hum. Immunol., 42, 227–232.
Clark, D.A., Wang, S., Rogers, P. et al. (1996) Endometrial lymphomyeloid
cells in abnormal uterine bleeding due to levonorgestrel (Norplant). Hum.
Reprod., 11, 1438–1444.
Jones, R.L., Kelly, R.W. and Critchley, H.O. (1997) Chemokine and
cyclooxygenase-2 expression in human endometrium coincides with
leukocyte accumulation. Hum. Reprod., 12, 1300–1306.
Jones, R.K., Searle, R.F., Stewart, J.A. et al. (1998a) Apoptosis, bcl-2
expression, and proliferative activity in human endometrial stroma and
endometrial granulated lymphocytes. Biol. Reprod., 58, 995–1002.
Jones, R.K., Bulmer, J.N. and Searle, R.F. (1998b) Phenotypic and functional
studies of leukocytes in human endometrium and endometriosis. Hum.
Reprod. Update, 4, 702–709.
Critchley, H.O., Kelly, R.W. and Kooy, J. (1994) Perivascular location of a
chemokine interleukin-8 in human endometrium: a preliminary report.
Hum. Reprod., 9, 1406–1409.
Critchley, H.O.D., Jones, R.L., Lea, R.G. et al. (1999) Role of inflammatory
mediators in human endometrium during progesterone withdrawal and
early pregnancy. J. Clin. Endocrinol. Metab., 84, 240–248.
Eckes, B., Aumailley, T. and Krieg, T. (1996) Collagens and the re-
establishment of dermal integrity. In Clarke, R.A.F. (ed), The Molecular
and Cellular Biology of Wound Repair. Plenum Press, New York, pp. 493–
512.
Katz, H.R. and Austen, K.F. (1997) A newly recognized pathway for the
negative regulation of mast cell-dependent hypersensitivity and
inflammation mediated by an endogenous cell surface receptor of the gp49
family. J. Immunol., 158, 5065–5070.
King, A., Wellings, V., Gardner, L. et al. (1989) Immunocytochemical
characterisation of the unusual large granular lymphocytes in human
endometrium throughout the menstrual cycle. Hum. Immunol., 24, 195–
205.
King, A., Balendran, N., Wooding, P. et al. (1991) CD3 leukocytes present in
the human uterus during early placentation: phenotypic and morphologic
+
Feliciani, C., Theoharides, T.C., Mammarella, S. et al. (1995) Activation of
characterisation of the CD56 population. Dev. Immunol., 1, 169–190.
mast cells by MCP-1. J. Invest. Dermatol., 104, 685.
Finn, C.A. (1986) Implantation, menstruation and inflammation. Biol. Rev., 61,
313–328.
Foxman, E.F., Campbell, J.J. and Butcher, E.C. (1997) Multistep navigation
and the combinatorial control of leukocyte chemotaxis. J. Cell Biol., 139,
1349–1360.
Fraser, I.S. (1997) Mechanisms of endometrial bleeding. Reprod. Fertil. Dev.,
2, 193–198.
Fraser, I.S. and Peek, M.J. (1992) Effects of exogenous hormones on
endometrial capillaries. In Alexander, N.J. and d’Arcangues, C. (eds),
Steroid Hormones and Uterine Bleeding. AAAS Publications,
Washington, pp. 65–79.
Galli, S.J. (1993) New concepts about the mast cell. N. Engl. J. Med., 328, 257–
265.
Haddad, E.K., Duclos, A.J., Antecka, E. et al. (1997) Role of interferon-gamma
in the priming of decidual macrophages for nitric oxide production and
early pregnancy loss. Cell. Immunol., 181, 68–75.
Hilby, S.E., King, A., Sharkey, A.M., et al. (1997) Human uterine NK cells
have a similar repertoire of killer inhibitory and activatory receptors to
those found in blood, as demonstrated by RT–PCR and sequencing. Mol.
Immunol., 34, 419–430.
Hunt, J.S., Miller, L., Roby, K.F., et al. (1997) Female steroid hormones
regulate production of pro-inflammatory molecules in uterine leukocytes.
J. Reprod. Immunol. 15, 87–99.
Hornung, D., Ryan, I.P., Chao, V.A. et al. (1997) Immunolocalization and
regulation of the chemokine RANTES in human endometrial and
endometriosis tissues and cells. J. Clin. Endocrinol. Metab., 82, 1621–
1628.
Hubner, G., Hu, Q., Smola, H. et al. (1996) Strong induction of activin
expression after injury suggests an important role of activin in wound
repair. Dev. Biol., 173, 490–498.
Iida, N., Haisa, M., Igarashi, A. et al. (1996) Leukocyte-derived growth factor
links the PDGF and CXC chemokine families of peptides. FASEB J., 10,
1336–1345.
King, A., Wooding, P., Gardner, L. et al. (1993) Expression of perforin,
+
granzyme A and TIA-1 by human uterine CD56 NK cells implies they are
activated and capable of effector functions. Hum. Reprod., 8, 2061–2067.
King, A., Jokhi, P.P., Smith, S.K.et al. (1995) Screening for cytokine mRNA in
human villous and extravillous trophoblasts using the reverse-transcriptase
polymerase chain reaction (RT-PCR). Cytokine, 7, 364–371.
King, A., Gardner, L. and Loke, Y.W. (1996) Evaluation of oestrogen and
progesterone receptor expression in uterine mucosal lymphocytes. Hum.
Reprod., 11, 1079–1082.
Klentzeris, L.D., Bulmer, J.N., Warren, A. et al. (1992) Endometrial lymphoid
tissue in the timed endometrial biopsy: morphometric and
immunohistochemical aspects. Am. J. Obstet. Gynecol., 167, 667–674.
Kodama, T., Hara, T., Okamoto, E. et al. (1998) Characteristic changes of large
granular lymphocytes that strongly express CD56 in endometrium during
the menstrual cycle and early pregnancy. Hum. Reprod., 13, 1036–1043.
Kroegel, C., Virchow, J.-C., Luttmann, W. et al. (1994) Pulmonary immune
cells in health and disease: the eosinophil leukocyte (Part 1). Eur. Respir. J.,
7, 519–543.
Kunkel, S.L., Streiter, R.M., Lindley, I.J.D., et al. (1995) Chemokines: new
ligands, receptors and activities. Immunol. Today, 16, 559–561.
Lathbury, L. and Salamonsen, L.A. (1998) Role of neutrophil enzymes in the
activation of endometrial matrix metalloproteinases. [ (Abstr.] Proc. Aust.
Soc. Immunol., 28, 10–13.
Leung, P.H.Y., Salamonsen, L.A. and Findlay, J.K. (1998) Immunolocalization
of inhibin and activin subunits in human endometrium across the menstrual
cycle. Hum. Reprod., 13, 3469–3477.
Loke, Y.W. and King, A. (1995) Human Implantation. Cell Biology and
Immunology. Cambridge University Press, Cambridge, UK.
Ludwig, H., Metzger, H. and Frauli, M. (1990) Endometrium: tissue
remodelling and regeneration. In d’Arcangues, C., Fraser, I.S., Newton,
J.R., and Odlind, V. (eds), Contraception and Mechanisms of Endometrial
Bleeding. Cambridge University Press, Cambridge, UK, pp. 441–466.
Ludwig, H. and Spornitz, U.M. (1991) Microarchitecture of the endometrium
by scanning electron microscopy: menstrual desquamation and
remodelling. Ann. N. Y. Acad. Sci., 622, 28–46.
Lydon, J.P., DeMayo, F.J., Funk, C.R. et al. (1995) Mice lacking progesterone
receptor exhibit pleiotropic reproductive abnormalities. Genes Dev., 9,
2266–2278.
Markee, J.E. (1940) Menstruation in intraocular endometrial transplants in the
rhesus monkey. Contrib. Embryol., 177, 220–230.
Inoue, T., Kanzaki, H., Imai, K. et al. (1996) Progesterone stimulates the
+
induction of human endometrial CD56 lymphocytes in an in vitro culture
system. J. Clin. Endocrinol. Metab., 81, 1502–1507.
Jeziorska, M., Nagase, H., Salamonsen, L.A. et al. (1996) Immunolocalization
of the matrix metalloproteinases, gelatinase B and stromelysin-1 in human
endometrium throughout the menstrual cycle. J. Reprod. Fertil., 107, 43–
51.
Martin, P. (1997) Wound healing – aiming for perfect skin regeneration.
Jeziorska, M., Salamonsen, L.A. and Woolley, D.E. (1995) Mast cell and
eosinophil distribution and activation in human endometrium throughout
the menstrual cycle. Biol. Reprod., 53, 312–320.
Jokhi, P.P., King, A., Sharkey, A.M. et al. (1994) Screening for cytokine
messenger ribonucleic acids in purified human decidual lymphocyte
Science, 276, 75–81.
Marzusch, K., Ruck, P., Geiselhart, A., et al. (1993) Distribution of cell
++
adhesion molecules on CD56
lymphocytes and endothelial cells in first-trimester human decidua. Hum.
Reprod., 8, 1203–1208.
, CD3–, CD16– large granular

Endometrial leukocytes and menstruation
27
Matsuzaki, S., Fukaya, T., Suzuki, T., et al. (1999) Oestrogen receptor alpha and
beta mRNA expression in human endometrium throughout the menstrual
cycle. Mol. Hum. Reprod., 5, 559–564.
Miller, L. and Hunt, J.S. (1996) Sex steroid hormones and macrophage
function. Life Sci., 59, 1–14.
Stewart, J.A., Bulmer, J.D. and Murdoch, A.P. (1998) Endometrial leukocytes:
expression of steroid hormone receptors. J. Clin. Pathol., 51, 121–126.
Struyf, S., Van Coillie, E., Paemen, L. et al. (1998) Synergistic induction of
MCP-1 and –2 by IL-1beta and interferons in fibroblasts and epithelial
cells. J. Leukocyte Biol., 63, 364–372.
Tabibzadeh, S. (1990) Proliferative activity of lymphoid cells in human
endometrium throughout the menstrual cycle. J. Clin. Endocrinol. Metab.,
70, 437–443.
Tabibzadeh, S. and Sun, X.Z. (1992) Cytokine expression in human
endometrium throughout the menstrual cycle. Hum. Reprod., 7, 1214–
1221.
Tabibzadeh, S., Kong, Q.F. and Babaknia, A. (1994) Expression of adhesion
molecules in human endometrial vasculature throughout the menstrual
cycle. J. Clin. Endocrinol. Metab., 79, 1024–1032.
Tawia, S.A., Beaton, L.A. and Rogers, P.A. (1993) Immunolocalization of the
cellular adhesion molecules, intercellular adhesion molecule-1 (ICAM)-1
and platelet endothelial cell adhesion molecule (PECAM), in human
endometrium throughout the menstrual cycle. Hum. Reprod., 8, 175–181.
Thomson, A., Greer, M.R., Young, A. et al. (1999) Expression of intercellular
adhesion molecules ICAM-1 and ICAM-2 in human endometrium:
regulation by interferon-gamma. Mol. Hum. Reprod., 5, 64–70.
Todd, R., Donoff, B.R., Chiang, T. et al. (1991) The eosinophil as a cellular
source of transforming growth factor alpha in healing cutaneous wounds.
Am. J. Pathol., 138, 1307–1313.
Nathan, C. (1987) Secretory products of macrophages. J. Clin. Invest., 79, 319–
326.
Norrby, K. and Woolley, D.E. (1993) Role of mast cells in mitogenesis and
angiogenesis in normal tissue and tumour tissue. Adv. Biosci., 89, 71–115.
Pace, D., Morrison, L. and Bulmer, J.N. (1989) Proliferative activity in
endometrial stromal granulocytes throughout the menstrual cycle and early
pregnancy. J. Clin. Pathol., 42, 35–39.
Poropatich, C., Rojas, M. and Silverberg, S.C. (1987) Polymorphonuclear
leukocytes in the endometrium during the normal menstrual cycle. Int. J.
Gynecol. Pathol., 6, 230–234.
Prabhala, R.H., Fahey, J.V., Humphrey, S.L. et al. (1998) Regulation by human
uterine cells of PBMC proliferation: influence of the phase of the menstrual
cycle and menopause. J. Reprod. Immunol., 40, 25–45.
Ritson, A. and Bulmer, J.N. (1989) Isolation and functional studies of
granulated lymphocytes in first trimester human decidua. Clin. Exp.
Immunol., 77, 268.
Roberts, D.K., Parmley, T.H., Walker, N.J. et al. (1992) Ultrastructure of the
microvasculature in the human endometrium throughout the normal
menstrual cycle. Am. J. Obstet. Gynecol., 166, 1393–1406.
Saito, S., Nishikawa, K., Morii, T. et al. (1993) Cytokine production by CD!^-
CD56 bright natural killer cell in the human early pregnancy decidua. Int.
Immunol., 5, 559–563.
Van Coillie, E., Nomiyama, H., Sakaki, Y. et al. (1997) The human MCP-2
gene (SCYA8): cloning, sequence analysis, tissue expression, and
assignment to the CC chemokine gene contig on chromosone 17q112.
Genomics, 40, 323–331.
Salamonsen, L.A. and Woolley, D.E. (1996) Matrix metalloproteinases and
their tissue inhibitors in endometrial remodelling and menstruation.
Reprod. Med. Rev., 5, 185–203.
Salamonsen, L.A. and Woolley, D.E. (1999) Menstruation: induction by matrix
metalloproteinases and inflammatory cells. J. Reprod. Immunol., 44, 1–27.
Salamonsen, L.A., Kovacs, G. and Findlay, J.K. (1999) Current concepts of the
mechanisms of menstruation. In Smith, S.K. (ed.), Baillière’s Clinical
Obstetrics and Gynaecology, Dysfunctional Uterine Bleeding. Baillière
Tyndall, London, UK, pp. 161–180.
Schall, T.J. (1994) The chemokines. In Thomson, A. (ed.), The Cytokine
Handbook. Academic Press, New York, USA, pp. 419–460.
Schwartz, L.B. (1994) Mast cells: function and contents. Curr. Opin. Immunol.,
6, 91–97.
Searle, R.F., Jones, R.K. and Bulmer, J.D. (1999) Phenotypic analysis and
proliferative responses of human endometrial granulated lymphocytes
during the menstrual cycle. Biol. Reprod., 60, 871–878.
Shi, W., Mognetti, B., Campana, A. et al. (1995) Metalloproteinase secretion by
endometrial leukocyte subsets. Am. J. Reprod. Immunol., 34, 299–310.
Somiglianna, E., Vigano, P., Gaffuri, B. et al. (1996) Human endometrial
stromal cells as a source of soluble intercellular adhesion molecule
(ICAM)-1 molecules. Hum. Reprod., 11, 1190–1194.
Song, J.Y. and Fraser, I.S. (1995) Effects of progestogens on human
endometrium. Obstet. Gynaecol. Surv., 50, 385–394.
Song, J.Y., Russell, P., Markham, R. et al. (1998) Effects of high dose
progestogens on white cells and necrosis in human endometrium. Hum.
Reprod., 11, 1713–1718.
Vincent, A.J., Malakooti, N., Rogers, P.A.W. et al. (1999) Endometrial
breakdown in women using Norplant is associated with migratory cell
expression of matrix metalloproteinase-9 (gelatinase B). Hum. Reprod., 14,
807–815.
Wang, H., Critchley, H.O.D., Kelly, R.W. et al. (1998) Progesterone receptor
subtype B is differentially regulated in human endometrial stroma. Mol.
Hum. Reprod., 4, 407–412.
Weill, P. (1921) Etudes sur les leukocytes. I. Les cellules granuleuses des
muqueuses intestinale et uterine. Arch. d’Anatomie Microscopique, 17, 77–
82.
+
+
White, H.D., Crassi, K.M., Givan, A.L. et al. (1997) CD3 CD8 CTL activity
within the human female reproductive tract. J. Immunol., 158, 3017–3027.
Wong, D.T.W., Donoff, B.R., Yang, J. et al. (1993) Sequential expression of
transforming growth factors α and β by eosinophils during cutaneous
1
wound healing in the hamster. Am. J. Pathol., 143, 130–141.
Yeaman, G.R., Collins, J.E., Currie, J.K. et al. (1998) IFN-gamma is produced
by polymorphonuclear neutrophils in human uterine endometrium and by
cultured peripheral blood polymorphonuclear neutrophils. J. Immunol.,
160, 5145–5153.
Zhang, J. and Salamonsen, L.A. (1997) Tissue inhibitor of metalloproteinases
(TIMP)-1, –2 and –3 in human endometrium during the menstrual cycle.
Mol. Hum. Reprod., 3, 735–741.
Zhang, J., Nie, G.-Y., Wang, J. et al. (1998) Mast cell regulation of endometrial
stromal cell matrix metalloproteinases:
a mechanism underlying
menstruation. Biol. Reprod., 59, 693–703.
Zhang, J., Hampton, A.L., Nie, G.-Y. et al. (2000a) Progesterone inhibits
activation of latent matrix metalloproteinase (MMP)-2 by membrane-type
1 MMP: enzymes coordinately expressed in human endometrium. Biol.
Reprod., 62, 85–94.
Zhang, J., Lathbury, L. and Salamonsen, L.A. (2000b) Expression of the
chemokine eotaxin and its receptor CCR3 in human endometrium. Biol.
Reprod., 62, in press.
Starkey, P.M., Clover, L.M. and Rees, M.C.P. (1991) Variation during the
menstrual cycle of immune cell populations in human endometrium. Eur.
J. Obstet. Gynecol. Reprod. Biol., 39, 203–207.
Stewart, C.J., Farquharson, M.A. and Foulis, A.K. (1992) The distribution and
possible function of gamma interferon-immunoreactive cells in normal
endometrium and myometrium. Virchows Arch.
Histopathol., 420, 419–424.
A Pathol. Anat.
Received on May 7, 1999; accepted on September 9, 1999