Leukocytes infiltrate the myometrium during human parturition: further evidence that labour is an inflammatory process

Andrew J. Thomson1,3, Joan F. Telfer1, Anne Young1, Steve Campbell1, Colin J.R. Stewart2, Iain T. Cameron1, Ian A. Greer1 and Jane E. Norman1

1 Department of Obstetrics and Gynaecology and 2 Department of Pathology, University of Glasgow, 10 Alexandra Parade, Glasgow, G31 2ER, Scotland, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Inflammatory mediators in the cervix, placenta and fetal membranes play a crucial role in human parturition. The aim of this study was to determine whether the upper and lower segments of the myometrium are infiltrated by inflammatory cells during pregnancy and parturition. Myometrial biopsies were obtained from non-pregnant women, and pregnant women at term before and after the onset of spontaneous labour. Subpopulations of inflammatory cells were identified using immunocytochemistry. The intercellular adhesion molecules, 1 and 2, platelet endothelial cell adhesion molecule, vascular cell adhesion molecule and E-selectin were immunolocalized to investigate their involvement in leukocyte accumulation. Histological analysis demonstrated that inflammatory cells, predominantly neutrophils and macrophages, infiltrate human myometrium during spontaneous labour at term. The infiltrate is predominant in the lower uterine segment but is also present in the upper segment. Increased expression of E-selectin was found on the vascular endothelium of biopsies obtained during labour, suggesting a role for this molecule in the accumulation of leukocytes. These results suggest that inflammatory cell infiltration is part of the physiological mechanisms that occur in the myometrium during parturition. Further understanding of this process may suggest new strategies aimed at preventing preterm delivery.

Key words: adhesion molecules/labour/macrophages/neutrophils/uterus


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Inflammatory mediators play a crucial role in human parturition (Kelly, 1996Go). Cervical ripening, an early event in normal labour, has been compared to an inflammatory reaction and is characterized by an accumulation of leukocytes in the cervical stroma (Junqueira et al., 1980Go; Liggins, 1981Go). The inflammatory infiltrate within the ripened cervix consists predominantly of neutrophils and macrophages and accumulates in the stroma before the onset of labour at term (Bokstrom et al., 1997Go). These leukocytes contribute to cervical remodelling by releasing proteolytic enzymes such as collagenase and elastase (Rajabi et al., 1988Go; Osmers et al., 1992Go). Elsewhere in the pregnant uterus, inflammatory cells are known to infiltrate the placenta, maternal decidua and the fetal membranes during parturition, and may play a role in spontaneous rupture of the membranes (Halgunset et al., 1994Go; Rosenberg et al., 1996Go).

There are no data on whether inflammatory cells infiltrate myometrium during labour. However, indirect evidence suggests that inflammatory cells may infiltrate the myometrium at the time of parturition. Rechberger and Woessner (1993) demonstrated that collagenase activity is increased in lower segment myometrium during labour and proposed that neutrophils may be responsible for much of the increase. Furthermore, Osmers et al. (1995) demonstrated a significant increase in interleukin-8 concentration in lower segment myometrium during labour, and proposed that this increase could result in an influx of neutrophils. The presence of these leukocytes in myometrium has been examined prior to the onset of labour only. Butterworth et al. (1991) found no neutrophils in the myometrium of both normal and pre-eclamptic women before the onset of labour.

The attachment and extravasation of circulating leukocytes is controlled by the expression of cell surface adhesion molecules on both the circulating cells and the vascular endothelium (Akyama et al., 1989Go; Bevilacqua, 1993Go). An increased expression of adhesion molecules on the endothelium occurs in inflammatory conditions and supports the recruitment and aggregation of leukocytes (Bevilacqua, 1993Go) The major endothelial adhesion molecules involved in leukocyte attachment and transendothelial migration include E-selectin, intercellular adhesion molecule-1 and 2 (ICAM-1 and 2), platelet endothelial cell adhesion molecule (PECAM) and vascular cell adhesion molecule-1 (VCAM-1).

The purpose of the present study was to investigate inflammatory cell subpopulations in the human myometrium during pregnancy and labour. Specifically, we aimed to determine whether myometrium in each of the lower and upper uterine segments, like cervical stroma, maternal decidua and the fetal membranes, is infiltrated by leukocytes during parturition. Furthermore, the expression and distribution of the cell adhesion molecules, E-selectin, ICAM-1 and 2, PECAM and VCAM-1, were assessed in myometrium during labour to investigate possible mechanisms of leukocyte accumulation.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Subjects and collection of tissue
Pregnant women delivered by lower uterine segment Caesarean section and non-pregnant women undergoing hysterectomy were recruited to the study. Informed consent was obtained from each woman prior to recruitment and the study was approved by the Local Research Ethics Committee.

Myometrial biopsies were collected from three groups of women: (i) Eighteen pregnant women who were delivered before the onset of labour at term (>37 weeks gestation); (ii) Eighteen pregnant women who were delivered during active labour at term (cervical dilatation >4 cm and <9 cm). Women were excluded from the study if they had a multiple pregnancy, evidence of active infection, or following induction of labour; (iii) Thirteen non-pregnant, pre-menopausal women with regular menstrual cycles undergoing hysterectomy for benign disease.

In groups (i) and (ii), a myometrial biopsy was obtained from the upper margin of the lower uterine segment incision during the Caesarean section. Additionally, in seven of the women in group (i), and five of the women in group (ii), a myometrial biopsy was obtained from the upper uterine segment by dissecting a strip of myometrium from the inner aspect of the posterior uterine wall and inserting a haemostatic suture when required. In group (iii), myometrial biopsies were taken from the anterior wall of the lower uterine body, immediately following hysterectomy.

Each of the 12 upper segment and the 13 non-pregnant myometrial biopsies was fixed in 10% neutral buffered formalin (BDH, Poole, UK) and embedded in paraffin. The 36 lower segment myometrial biopsies were divided; one half was fixed in 10% neutral buffered formalin and embedded in paraffin, and the other half was snap frozen in liquid nitrogen and stored at –70°C.

Identification of inflammatory cells
Inflammatory cells were detected using immunocytochemistry, with a panel of antibodies as shown in Table IGo. Serial sections were stained for naphthol AS-D chloroacetate esterase activity, an enzyme considered specific for cells of granulocytic lineage (neutrophils, eosinophils and basophils).


View this table:
[in this window]
[in a new window]
 
Table I. Primary antibodies used for leukocyte immunocytochemistry
 
Immunocytochemistry
Neutrophils, macrophages, T-lymphocytes and B-lymphocytes
Inflammatory cells were identified in each of the biopsies obtained before and after the onset of labour at term. Sections 8 µm thick were cut from the paraffin embedded tissues and mounted on silane-coated slides, heated to 60°C for 35 min, deparaffinized in xylene and rehydrated in a graded alcohol series. If required (Table IGo), the sections were then pretreated to retrieve the antigen by microwaving at full power for 4x5 min in citrate buffer (10 mM, pH 6.0), or by enzymatic digestion with a 0.1% (w/v) trypsin (Sigma, Poole, Dorset, UK) solution in Tris buffer (pH 7.6) containing 0.1% (w/v) calcium chloride, for 10 min at room temperature.

The sections were pre-incubated with 1.5% (w/v) horse serum in PBS (10 mM sodium phosphate, pH 7.5, 120 mM sodium chloride) for 30 min at room temperature. They were then incubated for 1 h with the primary antibody diluted in 1.5% horse serum. Next, sections were washed in PBS before incubation with biotinylated anti-mouse immunoglobulin. This antibody, from a Vectastain Elite ABC kit (Vector, Peterborough, UK) was first diluted in 1.5% horse serum and 1.5% normal human serum. After washing as before, sections were placed in 3% hydrogen peroxide in methanol for 10 min at room temperature. The sections were thoroughly washed again, then incubated for 30 min with avidin DH/biotinylated horseradish peroxidase H reagent (Vectastain Elite ABC kit) in PBS before final washing. The antigen was localized using 1 mg/ml diaminobenzidine tetrahydrochloride (DAB), 0.02% H2O2 in 50 mM Tris.Cl, pH 7.6, and appeared as a brown end-product.

Negative controls included sections incubated without the primary antibody and sections incubated with a mouse monoclonal antibody against IgG1 Aspergillus niger glucose oxidase (Dako Ltd, High Wycombe, UK), an enzyme which is not expressed in mammalian cell systems. Tonsillar tissue was used as positive controls for CD 3, CD 20 and CD 68.

Mast cells
Mast cells were localized in non-pregnant and pregnant myometrial biopsies. The paraffin embedded sections were prepared as before and digested in a trypsin solution to retrieve the antigen (Table IGo). The sections were then pre-incubated with 20% normal goat serum (SAPU, Carluke, UK) in PBS (10 mM sodium phosphate, pH 7.5, 120 mM sodium chloride) for 30 min at room temperature. They were then incubated for 90 min with a monoclonal antibody raised against mast cell tryptase (Dako Ltd) diluted 1/300 in 2% normal goat serum. The primary antibody was omitted from the negative control slides. Next, sections were washed in PBS before incubation with goat anti-mouse IgG alkaline phosphatase (Sigma) diluted 1/200 in 2% normal goat serum and 5% normal human serum. The sections were thoroughly washed again and immunoreactive tryptase was localized using Fast Red Substrate (Sigma). Finally, the sections were counter-stained with Harris haematoxylin.

Cell adhesion molecules
Sections 5 µm thick were cut from the frozen tissue and mounted on silane-coated slides. The sections were fixed in acetone for 10 min, washed in TBS, and placed in 0.5% hydrogen peroxide in methanol for 30 min at room temperature. The sections were then washed as before and pre-incubated with 20% (w/v) normal goat serum (SAPU) in TBS for 30 min at room temperature. The sections were then incubated for 16 h at 4°C with the primary antibody diluted in 1.5% horse serum. Table IIGo shows the characteristics of the primary antibodies. The primary antibody was omitted from the negative control slides. Next, sections were washed in TBS before incubation for one h with biotinylated goat anti-mouse immunoglobulin (Dako Ltd), diluted in 2% normal goat serum and 1.5% normal human serum. The sections were thoroughly washed again, then incubated for 30 min with streptavidin horseradish peroxidase (Dako Ltd) in TBS before final washing. The antigen was localized as previously described using DAB.


View this table:
[in this window]
[in a new window]
 
Table II. Primary antibodies used for cell adhesion molecule immunocytochemistry
 
Identification of granulocytes
Granulocytic cells were identified by staining for naphthol AS-D chloroacetate esterase activity. This activity was determined using a commercial kit (Sigma Diagnostics, Poole, UK). Briefly, 1 ml of sodium nitrite solution (0.1 mol/l) was added to 1 ml of Fast Red Violet LB Base solution (15 mg/ml fast red violet LB base in 0.4 mol/l hydrochloric acid with stabilizer) and allowed to stand for 2 min. This solution was then added to 40 ml prewarmed (37°C) deionized water, 5 ml of Trizmal 6.3 buffer concentrate (TRIZMA maleate, 1 mol/l with surfactant, pH 6.3) and 1 ml of naphthol AS-D chloroacetate solution (naphthol AS-D chloroacetate 8 mg/ml and stabilizer). The slides were incubated in this solution for 15 min at 37°C and then rinsed thoroughly in deionized water. Sites of activity showed violet granulation.

Quantification of inflammatory cells and statistical analysis
The inflammatory infiltrate was quantified (i) to determine whether inflammatory cells infiltrate the myometrium during labour at term (ii) to compare the density of the inflammatory cell infiltrate in upper and lower uterine segment myometrium and (iii) to compare mast cell density in non-pregnant and pregnant myometrium. Inflammatory cells were identified by either brown staining (neutrophils, macrophages, T cells and B cells) or red staining (mast cells). In each section of myometrium, the number of positive cells was counted in a high powered field (x400 magnification within parts of a lined grid covering an area of 0.02 mm2). Six different fields were counted by two observers who were blinded to the specimen details. Areas containing blood vessels were avoided and leukocytes within vessels were not included. The average number (arithmetic mean) of positive cells recorded per field by each observer was calculated, and then a mean of these two values obtained. Statistical comparisons of the means were performed using three-factor analysis of variance (ANOVA) with Scheffé's S as a post hoc test. Significant differences between groups were explored using the Mann–Whitney U-test. The sites of expression of each of the cell adhesion molecules were recorded by the observers who were blinded to the specimen details. Differences in the expression of these molecules before and after labour were analysed using the {chi}2 test.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Staining for naphthol AS-D chloroacetate esterase activity, which identifies inflammatory cells of granulocytic lineage, revealed that myometrial biopsies removed from labouring women [group (ii)] showed a marked inflammatory infiltrate in the muscle connective tissue in 17 of the 18 specimens (Figure 1Go). In contrast, the myometrium in the non-labouring women [group (i), n = 18] did not exhibit an inflammatory infiltrate. The infiltrate appeared most dense at both the luminal (decidual) edge of the myometrium and also in and around blood vessels. Further, the infiltrate was most striking in the myometrium obtained from the lower uterine segment, but was also present in the myometrial biopsies obtained from the uterine fundus during labour.



View larger version (137K):
[in this window]
[in a new window]
 
Figure 1. Identification of granulocytes in myometrium. The granulocytes are localized by staining for naphthol AS-D chloroacetate esterase activity and appear as violet granulation. (A) Granulocytes are sparse in myometrium collected from women not in labour (arrow) and (B) are abundant in myometrium obtained during labour. Bar = 50 µm.

 
Analysis of individual cell types using immunocytochemistry showed significantly more inflammatory cells in labouring myometrium compared to non-labouring biopsies (Scheffé's test, P = 0.0001), and significantly more inflammatory cells were present in the lower uterine segment compared to the upper uterine segment myometrium (Scheffé's test, P < 0.02).

The leukocyte subpopulations were characterized as follows:

Neutrophils
Neutrophils were sparse in myometrium obtained before the onset of labour and abundant in biopsies obtained during labour (Figure 2Go). A significant increase in myometrial neutrophil density occurred following the onset of labour in both the lower (Table IIIGo), and the upper uterine segments (Table IVGo). During labour, the neutrophil density was significantly greater in the lower than in the upper uterine segment (P < 0.02). Within the labouring biopsies, the elastase antigen was localized both within the neutrophil cytoplasm and extracellularly in the vicinity of the leukocytes, suggesting that a proportion of these cells had degranulated.



View larger version (130K):
[in this window]
[in a new window]
 
Figure 2. Identification of neutrophils in lower segment myometrium obtained (A) before and (B) during labour. The neutrophils are immunolocalized using an antibody directed against the enzyme human neutrophil elastase. Neutrophils are abundant in myometrium obtained during labour and sparse in myometrium obtained before the onset of labour (arrow). The negative controls (see text) exhibited no reactivity. Bar = 50 µm.

 

View this table:
[in this window]
[in a new window]
 
Table III. Number of inflammatory cells in lower segment myometrium before and during labour at term, median (interquartile range) per high powered field
 

View this table:
[in this window]
[in a new window]
 
Table IV. Number of inflammatory cells in upper segment myometrium before and during labour at term, median (interquartile range) per high powered field
 
Macrophages
A population of immunoreactive macrophages was identified in myometrium before the onset of labour. The number of macrophages was significantly increased in both lower and upper uterine segments of the myometrium following the onset of labour (Figure 3Go, Tables III and IVGoGo). Following the onset of labour, there was no significant difference in macrophage density between the upper and lower uterine segments.



View larger version (144K):
[in this window]
[in a new window]
 
Figure 3. Identification of macrophages in lower segment myometrium obtained (A) before and (B) during labour. The macrophages are immunolocalized using an antibody directed against the antigen CD 68. Whilst a population of immunoreactive macrophages is identified before the onset of labour (arrows), significantly more macrophages are localized in myometrium following the onset of labour. The negative controls (see text) exhibited no reactivity. Bar = 50 µm.

 
Lymphocytes
There was a significant increase in the density of T-lymphocytes in lower segment myometrium following the onset of labour (Table IIIGo). B-Lymphocytes were sparse in lower segment myometrium and no significant change in their density occurred with the onset of labour. There was no significant change in either T or B-lymphocyte density following the onset of labour in upper segment myometrial biopsies (Table IVGo).

Mast cells
There was no significant difference in mast cell density in myometrium before and after the onset of labour at term (Table IIIGo). Furthermore, mast cells were sparse in both upper and lower uterine myometrial biopsies (Table IVGo), with no significant differences in mast cell density between the biopsy sites. We assessed mast cell density in non-pregnant myometrium [group (iii)], and found that this was significantly greater than in biopsies obtained from each of the pregnant groups, term not in labour [group (i)], and term in labour [group (ii)] (Figure 4Go). These results are summarized in Table VGo.



View larger version (149K):
[in this window]
[in a new window]
 
Figure 4. Identification of mast cells in (A) non-pregnant myometrium and (B) term, pregnant myometrium. The mast cells are immunolocalized (arrows) using an antibody directed against the enzyme, mast cell tryptase. There are significantly more mast cells in non-pregnant myometrium than in pregnant myometrium. The negative controls (see text) exhibited no reactivity. Bar = 50 µm.

 

View this table:
[in this window]
[in a new window]
 
Table V. Presence of mast cells in human myometrium. Values are shown as median (interquartile range)
 
Cell adhesion molecules
The cell adhesion molecules were localized as brown staining in serial cryosections of lower segment myometrium before (n = 6) and after (n = 6) the onset of labour (Figure 5Go). There were no changes in the immunolocalization of ICAM-1, ICAM-2, PECAM or VCAM following the onset of labour. ICAM-1 immunostaining was of weak intensity and ICAM-2 staining was of moderate intensity in the myometrial vascular endothelium, although not all vessels expressed these molecules. PECAM expression was strong and consistent in all of the vascular endothelium. In contrast, VCAM expression was weak and absent in some vessels. E-selectin was not expressed in any of the tissues obtained before the onset of labour. However, three of the six biopsies obtained during labour showed immunostaining for E-selectin on the vascular endothelium ({chi}2 P < 0.05).



View larger version (130K):
[in this window]
[in a new window]
 
Figure 5. Localization of cell adhesion molecules to the vascular endothelium in serial cryosections of myometrium. There are no changes in the expression of (A) ICAM-1 (B) ICAM-2, (C) PECAM or (D) VCAM in the vascular endothelium following the onset of labour. E-selectin is not expressed in any of the biopsies obtained before the onset of labour but is identified in three of the six biopsies obtained during labour. (E) Localization of E-selectin in the capillary vascular endothelium in myometrium obtained during labour. The negative controls (see text) exhibited no reactivity. Bar = 50 µm.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
We have demonstrated that inflammatory cells, predominantly neutrophils and macrophages, infiltrate human myometrium during spontaneous labour at term. The inflammatory infiltrate is predominant in the lower uterine segment, but is also present in the myometrium of the uterine body. We also observed a significant increase in T-lymphocyte density in lower segment myometrium following the onset of labour. There was no change in B-lymphocyte and mast cell densities in the myometrium during parturition.

In this study, we have examined both lower and upper uterine segment myometrium for the presence of inflammatory cells. Whether lower segment tissue reflects the state of the cervix or that of the uterine fundus, has aroused considerable interest over the last half century (Calder, 1994Go). Some workers contest that lower uterine segment biopsies provide an alternative source of tissue that closely resembles cervix (Rajabi et al., 1988Go). Because of the difficulties in obtaining cervical tissue for study, lower segment biopsies have been used in place of cervical tissue to investigate collagenase activity during cervical dilatation (Rechberger and Woessner, 1993Go). Our results indicate that lower uterine segment myometrium behaves quite differently from reported behaviour in cervix, and has similarities to upper segment myometrium. Bokstrom et al. (1997) demonstrated an abundance of neutrophils and macrophages in cervical biopsies in late pregnancy before the onset of labour with no significant increase in their densities during labour. In contrast, we have shown that neutrophils and macrophages are sparse in lower and upper segment myometrium before, and abundant during labour. Although the upper and lower uterine segments behave in a similar manner, they are not identical. During labour, the inflammatory infiltrate is more dense in lower segment myometrium than in upper segment myometrium. Consistent with this histological finding, functional studies have demonstrated that in normal labour the upper uterine segment contracts more strongly than the lower, a situation which is reversed in abnormal labour (Caldeyro-Barcia and Poseiro, 1960Go; Margono et al., 1993Go).

Activated neutrophils and macrophages are a rich source of inflammatory mediators. These include plasminogen activators, eicosanoids, collagenase and elastase, and proinflammatory cytokines, including interleukin-1 and tumour necrosis factor-{alpha} (Nathan, 1987Go; Osmers et al., 1992Go; Casatella, 1995Go). Since these mediators have many diverse functions, the inflammatory infiltrate could have different roles in different regions of the uterus. Within the lower segment it could be involved in tissue remodelling and thereby facilitate cervical dilatation and passage of the fetus. In the upper segment, leukocyte products, including eicosanoids, interleukins and tumour necrosis factor-{alpha}, may stimulate uterine contractions directly, or indirectly by facilitating the production of uterotonic prostaglandins (Casey et al., 1990Go). Furthermore, inflammatory mediators may also initiate tissue remodelling in the uterine body. Granstrom et al. (1989) demonstrated that the connective tissue of the uterine isthmus (lower segment) and the uterine body undergoes a biochemical ripening process similar to that found in the cervix, with an increase in collagenolytic activity following the onset of labour. A breakdown of the connective tissue within the myometrium may facilitate the co-ordination of uterine contractions by allowing the formation of gap junctions (Garfield and Hayashi, 1981Go).

We found no change in myometrial mast cell density before and after the onset of labour at term. The function of mast cells in the pregnant uterus remains unclear, although it has been proposed that myometrial mast cells regulate uterine contractility during labour (Rudolph et al., 1993Go). Mast cells produce mediators (histamine and serotonin) and prostaglandins which can induce strong contractions in human myometrium in vitro (Cruz et al., 1989Go; Rudolph et al., 1990Go, 1993Go). Further, these cells are considered to play a pivotal role in wound healing, fibrosis and tissue remodelling (Galli, 1993Go), and might be involved in promoting collagen degradation and uterine involution in the postnatal period (Jeffrey et al., 1991Go). We were surprised at the low density of mast cells in the pregnant myometrial biopsies since this is in contrast with the known distribution of mast cells in non-pregnant myometrium (Mori et al., 1997Go). Since mast cell mediators are capable of stimulating uterine contractions, the low density of mast cells in pregnant myometrium may be involved in the maintenance of myometrial quiescence as the uterus expands during pregnancy. Whilst we could demonstrate no change in myometrial mast cell density following the onset of labour, mast cell mediators might be capable of stimulating uterine contractions at term since the sensitivity of human myometrium to histamine and serotonin is upregulated at the end of pregnancy (Cruz et al., 1989Go). This means that myometrial smooth muscle cells might be stimulated by mast cell mediators even when the mast cell density is reduced. In non-pregnant myometrium, the high density of mast cells has been proposed to have a role in implantation (Brandon and Evans, 1983Go; Hore and Mehrotra, 1988Go), or in remodelling uterine smooth muscle and extracellular matrix during the menstrual cycle (Mori et al., 1997Go).

The mechanisms involved in the accumulation, extravasation and degranulation of inflammatory cells in uterine tissues during parturition are poorly understood. Chemotactic cytokines, including interleukin-1, tumour necrosis factor-{alpha} and interleukin-8, seem to play a role (Barclay et al., 1993Go; Chwalisz et al., 1994Go; Osmers et al., 1995Go), as well as other chemotactic agents, such as C5a (El Maradny et al., 1995Go). E-selectin is involved in the infiltration of leukocytes to the maternal decidua and fetal membranes during labour (Rosenberg et al., 1996Go).

Since the leukocytes in our myometrial biopsies were concentrated in and around blood vessels, we postulated that an up-regulation in the expression of vascular cell adhesion molecules in lower segment myometrium was involved in the accumulation of leukocytes in this tissue. ICAM-1, ICAM-2, PECAM and VCAM are members of the immunoglobulin superfamily (Frenette and Wagner, 1996Go). ICAM-1 is important in the adhesion of monocytes, lymphocytes and neutrophils to activated endothelium, whilst VCAM binds to leukocyte integrins on many cells including eosinophils and activated T lymphocytes. E-selectin, a member of the selectin family of adhesion molecules, is expressed by cytokine activated endothelial cells and has a major role in attracting neutrophils, monocytes, eosinophils and some lymphocytes (Lasky 1992Go; Bevilacqua and Nelson, 1993Go).

The expression of cell adhesion molecules in the endometrium of the non-pregnant uterus is well described (Tawia et al., 1993Go; Tabibzadeh et al., 1994Go). A recent report has identified ICAM-1, VCAM, PECAM and E-selectin in pregnant human myometrium (Winkler et al., 1998Go). We have shown that ICAM-1, ICAM-2, PECAM and VCAM are expressed on the vascular endothelium in myometrium obtained before the onset of labour at term and we propose that these molecules play a role in regulating leukocyte trafficking into this tissue. We found no change in the localization and intensity of staining of ICAM-1, ICAM-2, PECAM and VCAM in the biopsies obtained during labour compared with those obtained before the onset of labour. In contrast, E-selectin expression was absent in all of the biopsies collected before labour, but was expressed in three of the six biopsies obtained during labour, suggesting a role for this molecule in the recruitment of leukocytes in at least some of the tissues. These results are in broad agreement with Winkler et al. (1998), who also found that E-selectin expression was up-regulated during labour. Both our own study and that of Winkler et al. (1998) have employed immunocytochemistry, a qualitative technique. In order to confirm the changes in cell adhesion molecule expression during parturition, further studies are required using quantitative techniques.

Factors responsible for the initiation of parturition remain obscure. We have demonstrated that leukocytes infiltrate both the upper and lower uterine segments of the myometrium during spontaneous labour at term, and we propose that these cells play a fundamental role in normal parturition. A better understanding of the mechanisms involved in the initiation of labour both at term and preterm would allow the development of novel strategies to prevent premature delivery. Our results suggest that strategies aimed at preventing the influx of inflammatory cells into the myometrium could be crucial in averting preterm delivery, and thus in reducing the excess perinatal mortality and morbidity associated with this condition.


    Acknowledgments
 
This work was supported by grants from Wellbeing (NI/95) and the Medical Research Council (G94 37277), for which we are very grateful.


    Notes
 
3 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Akyama, S.K., Yamada, S.S., Chen, W.T. and Yamada, K.M. (1989) Analysis of fibronectin receptor function with monoclonal antibodies: roles in cell adhesion, migration, matrix assembly and cytoskeletal organization. J. Cell. Biol., 109, 863–875.[Abstract]

Bevilacqua, M.P. (1993) Endothelial–leukocyte adhesion molecules. Annu. Rev. Immunol., 11, 767–804.[ISI][Medline]

Bevilacqua, M.P. and Nelson, R.M. (1993) Selectins. J. Clin. Invest., 91, 379–387.[ISI][Medline]

Barclay, C.G., Brennand, J.E., Kelly, R.W. and Calder, A.A. (1993) Interleukin-8 production by the human cervix. Am. J. Obstet. Gynecol., 169, 625–632[ISI][Medline]

Bokstrom, H., Brannstrom, M., Alexandersson, M. and Norstrom, A. (1997) Leukocyte subpopulations in human uterine cervical stroma at early and term pregnancy. Hum. Reprod., 12, 586–590.[ISI][Medline]

Brandon, J.M. and Evans, J.E. (1983) Change in mast cells during the estrous cycle in the Syrian hamster. Am. J. Anat., 167, 241–247.[ISI][Medline]

Butterworth, B.H., Greer, I.A., Liston, W.A. et al. (1991) Immunocytochemical localization of neutrophil elastase in term placenta, decidua and myometrium in pregnancy-induced hypertension. Br. J. Obstet. Gynaecol., 98, 929–933.[ISI][Medline]

Calder, A.A.. (1994) The cervix during pregnancy. In Chard, T. and Grudzinskas, J.G. (eds), The Uterus. Cambridge University Press, Cambridge, pp. 288–307.

Caldeyro-Barcia, R. and Poseiro, J.J. (1960) Physiology of the uterine contraction. Clin. Obstet. Gynecol., 3, 386–408.[ISI]

Casatella, M.N. (1995) The production of cytokines by polymorphonuclear neutrophils. Immunol. Today, 16, 21–26.[ISI][Medline]

Casey, M.L., Cox, S.M., Word, A. and MacDonald, P.C. (1990) Cytokines and infection-induced preterm labour. Reprod. Fertil. Dev., 2, 499–510.[ISI][Medline]

Chwalisz, K., Benson, M., Scholz, P. et al. (1994) Cervical ripening with the cytokines interleukin 8, interleukin 1ß and tumour necrosis factor {alpha} in guinea pigs. Hum. Reprod., 9, 2173–2181.[Abstract]

Cruz, M.A., Gonzalez, C., Acevedo, C.G. et al. (1989) Effects of histamine and 5-HT on the contractility of isolated pregnant and non-pregnant human myometrium. Gynecol. Obstet. Invest., 28, 1–4.[ISI][Medline]

El Maradny, E., Kanayama, N., Halim, A. et al. (1995) The effect of interleukin-1 in rabbit cervical ripening. Eur. J. Obstet. Gynecol. Reprod. Biol., 60, 75–80.[ISI][Medline]

Frenette, P.S. and Wagner, D.D. (1996) Adhesion molecules – part 1. N. Engl. J. Med., 334, 1526–1529.[Free Full Text]

Galli, S.J. (1993) New concepts about the mast cell. N. Engl. J. Med., 328, 257–265.[Free Full Text]

Garfield, R.E. and Hayashi, R.H. (1981) Appearance of gap junctions in the myometrium of women during labor. Am. J. Obstet. Gynecol., 140, 254–260.[ISI][Medline]

Granstrom, L., Ekman, G., Ulmsten, U. and Malmstrom, A. (1989) Changes in connective tissue of corpus and cervix uteri during ripening and labour in term pregnancy. Br. J. Obstet. Gynaecol., 96, 1198–1202.[ISI][Medline]

Halgunset, J., Johnsen, H., Kjollesdal, A.M. et al. (1994) Cytokine levels in amniotic fluid and inflammatory changes in the placenta from normal deliveries at term. Eur. J. Obstet. Gynecol. Reprod. Biol., 56, 153–160.[ISI][Medline]

Hore, A. and Mehrotra, P.N. (1988) Presence of a blastocyst and mast cell depletion of the mouse uterus. Acta. Anat., 132, 6–8.[ISI][Medline]

Jeffrey, J.J., Ehlich, L.S. and Roswit, W.T. (1991) Serotonin: an inducer of collagenase in myometrial smooth muscle cells. J. Cell. Physiol., 146, 399–406.[ISI][Medline]

Junqueira L.C.U., Zugaib M., Montes G.S. et al. (1980) Morphological and histochemical evidence for the occurrence of collagenolysis and for the role of neutrophilic polymorphonuclear leukocytes during cervical dilatation. Am. J. Obstet. Gynecol., 138, 273–281.[ISI][Medline]

Kelly, R.W. (1996) Inflammatory mediators and parturition. J. Reprod. Fertil., 106, 89–96.

Lasky, L.A. (1992) Selectins: interpreters of cell-specific carbohydrate information during inflammation. Science, 258, 964–996.[ISI][Medline]

Liggins, G.C. (1981) Cervical ripening as an inflammatory reaction. In Ellwood, D.A. and Anderson, A.B.M. (eds), The Cervix in Pregnancy and Labour: Clinical and Biochemical Investigations. Churchill Livingstone, Edinburgh, pp. 1–9.

Margono, F., Minkoff, H. and Chan, E. (1993) Intrauterine pressure wave characteristics of the upper and lower uterine segments in parturients with active-phase arrest. Obstet. Gynecol., 81, 481–485.[Abstract]

Mori, A., Zhai Y.L., Toki T. et al. (1997) Distribution and heterogeneity of mast cells in the human uterus. Hum. Reprod., 12, 368–372.[Abstract]

Nathan, C.F. (1987) Secretory products of macrophages. J. Clin. Invest., 79, 319–320.[ISI][Medline]

Osmers, R., Rath, W., Adelmann-Grill, B.C. et al. (1992) Origin of cervical collagenase during parturition. Am. J. Obstet. Gynecol., 166, 1455–1460.[ISI][Medline]

Osmers, R.G.W., Blaser, J., Kuhn, W. and Tschesche, H. (1995) Interleukin-8 synthesis and the onset of labor. Obstet. Gynecol., 86, 223–229.[Abstract/Free Full Text]

Rajabi, M.R., Dean, D.D., Beydoun, S.N. and Woessner J.R. (1988) Elevated tissue levels of collagenase during dilation of uterine cervix during human parturition. Am. J. Obstet. Gynecol., 159, 971–976.[ISI][Medline]

Rechberger, T. and Woessner, J.F. (1993) Collagenase, its inhibitors, and decorin in the lower uterine segment in pregnant women. Am. J. Obstet. Gynecol., 168, 1598–1603.[ISI][Medline]

Rosenberg S., Draper L.D., Heine R.P. et al. (1996) Polymorphonuclear leukocytes invade maternal decidua and fetal membranes via an e-selectin dependent pathway during human labour. J. Soc. Gynecol. Invest, 3, 553

Rudolph, M.I., Reinicke, K., Padilla, L. et al. (1990) The mast cells in pregnant uterus. Eur. J. Pharmacol., 183, 1765–1766.[ISI]

Rudolph, M.I., Reinicke, K., Cruz, M.A. et al. (1993) Distribution of mast cells and the effect of their mediators on contractility in human myometrium. Br. J. Obstet. Gynaecol., 100, 1125–1130.[ISI][Medline]

Tawia S.A., Beaton L.A. and Rogers P.A.W. (1993) Immunolocalization of the cellular adhesion molecules, intercellular adhesion molecule-1 (ICAM-1) and platelet endothelial cell adhesion molecule (PECAM) in the human endometrium throughout the menstrual cycle. Hum. Reprod., 8, 175–181.[Abstract]

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.[Abstract]

Winkler, M., Ruck, P., Horny, H.P. et al. (1998) Expression of cell adhesion molecules by endothelium in the human lower uterine segment during parturition at term. Am. J. Obstet. Gynecol., 178, 557–561.[ISI][Medline]

Submitted on May 7, 1998; accepted on September 24, 1998.