| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Immunology |
Department of Obstetrics and Gynaecology and Reproductive Medicine Unit, Adelaide University, Adelaide, South Australia, 5005, Australia
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
female reproductive tract, immunology, steroid hormones, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Dendritic cells, the most specialized and potent APC [1], display a number of different maturational stages and behavioral features. Precursor or "immature" dendritic cells, found in peripheral organs and characterized by their capacity to efficiently capture and retain exogenous antigens, undergo phenotypic modulation with appropriate stimuli and home to lymph nodes, where as "mature" dendritic cells they induce the activation, proliferation, and phenotypic polarization of naïve T lymphocytes in accordance with their differential expression of surface major histocompatibility class (MHC) II and costimulatory molecules and cytokine secretion profile. Phenotypic variation in dendritic cells results from their ontogeny and lineage (e.g., whether derived from myeloid or lymphoid precursors) and their exposure to cytokines, chemokines, and exogenous bioactive moieties (e.g., bacterial components) present in the tissue milieu during antigen encounter.
Macrophages exhibit many of the cell membrane and behavioral features typical of professional APCs, and under appropriate environmental conditions, they can differentiate into cells with features indistinguishable from those of dendritic cells [2–4]. Macrophages residing in peripheral tissues play a primary role in mediating innate immune protection via receptor-mediated internalization and intracellular degradation of invading pathogens and other particulate matter [5] and by secretion of bioactive and toxic substances [6, 7]. Their secretion of cytokines and other bioactive moieties can be a potent regulatory influence in the behavior of dendritic cells [6]. Thus, macrophages and dendritic cells together offer innate immune protection and initiate antigen-specific immune responses such that tissue homeostasis is maintained in the face of pathogen invasion or other insults.
Because macrophages and dendritic cells have related developmental pathways, perform similar functions, and often show indistinguishable morphologies and patterns of distribution within mucosal and lymphoid tissues, clear differentiation between the two cell types can be difficult. Furthermore, subpopulations of macrophages and dendritic cells display remarkable heterogeneity in their expression of cell-surface molecules. Whereas F4/80 is expressed by most macrophage populations [3, 8] and dendritic cells are noted for their constitutive expression of MHC class II [1], both these markers and others, including class A scavenger receptor [9, 10], macrosialin [8, 11], sialoadhesin [12, 13], and B7-2 [14], can be found on both cell types depending on the activation status and tissue microenvironment. Other markers, such as the ß2-integrins CD11b/CD18 and CD11c/ CD18 and the dendritic cell-associated molecule 33D1, also show varied and overlapping patterns of expression on myeloid APCs [2, 13, 15, 16]. Both DEC205 [1] and CD1 [17] can be expressed on dendritic cells of either myeloid or lymphoid origin as well as on some monocytes/macrophages [18–20]. Thus, it is not feasible to uniformly apply a single phenotypic marker to distinguish between subpopulations of APCs. Instead, a careful and multifaceted approach needs to be adopted to identify and delineate macrophages and dendritic cells on a tissue-specific basis.
Dendritic cells must be considered to be the primary candidate for initiating immune responses to antigens encountered in the uterine environment. Although multiplexed flow cytometry has been utilized to identify myeloid dendritic cells within human uterine tissues [21], populations of putative dendritic cells have been proposed to exist in rodent endometrium based only on single-marker immunohistochemical and morphometric analysis [22–24]. Thus, similar to the situation prior to early studies regarding APCs in the lung [25], eye [26], and gastrointestinal tract [27], it remains unknown whether these cells are truly distinct from or simply a subset of the abundant populations of macrophages in the rodent endometrium [28–30]. The primary aim of the present study therefore was to undertake a detailed analysis of the phenotypes and dependence on ovarian steroid hormones of uterine APCs in the mouse to evaluate the occurrence and relationship between distinct populations of macrophages and dendritic cells in this tissue.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tissues and cells obtained from adult (age, 6–12 wk) virgin Balb/c x C57Bl/6 (Balb/c F1, H-2db) females were used for immunohistochemistry and flow cytometry experiments, and for phytohemagglutinin (PHA) proliferation assays. Cells obtained from C57Bl/6 (H-2b) and Balb/c (H-2d) females were used for mixed-lymphocyte reactions (MLRs). For identification of estrous mice, vaginal smears were classified according to previously described criteria [31]. To remove ovary-derived steroid hormones, mice were anesthetized by i.p. injection of avertin (1 mg/ml of tribromoethyl alcohol in tertiary amyl alcohol diluted to 2.5% [v/v]) at a dosage of 15 µl/g body weight and then bilaterally ovariectomized through a dorsal incision as previously described [32]. Animals were allowed to recover for 2–3 wk before use. All investigations were approved by the University of Adelaide Animal Ethics Committee and were carried out in accordance with the Guiding Principles for the Care and Use of Research Animals endorsed by the Society for the Study of Reproduction.
Antibodies
The rat monoclonal antibodies (mAbs) 2F8 (reactive with class A scavenger receptor [33]), FA11 (reactive with macrosialin [11]), and 3D6 (reactive with sialoadhesin [34]) were kindly provided as tissue-culture supernatants by S. Gordon (University of Oxford, Oxford, U.K.). Rat mAb mouse thymic stroma (MTS) number 24 (reactive with uterine epithelial cells [35]) was kindly provided as tissue-culture supernatant by R. Boyd (Monash University, Prahan, VIC, Australia). Rat mAbs as tissue-culture supernatants generated from hybridomas obtained from the American Type Culture Collection (Rockville, MD) were F4/80 (reactive with a macrophage antigen of unknown function), TIB120 (reactive with MHC class II [Ia]), TIB122 (reactive with leukocyte common antigen), TIB128 (reactive with CD11b), and TIB227 (reactive with 33D1). Biotin-conjugation of TIB120 was performed in-house. Monoclonal antibodies GL1 (reactive with CD86), biotin-conjugated 1B1 (reactive with CD1), and biotin-conjugated HL3 (reactive with CD11c) were obtained from Pharmingen (San Diego, CA). The NLDC-145 (reactive with DEC205) was obtained from Serotec (Oxford, U.K.), and horseradish peroxidase (HRP)-conjugated goat anti-rat immunoglobulin (Ig), biotin-conjugated rabbit anti-rat Ig, alkaline phosphatase-conjugated rabbit anti-rat Ig, HRP-conjugated streptavidin, and phycoerythrin (PE)-conjugated streptavidin were obtained from Dako (Copenhagen, Denmark). Fluorescein isothiocyanate (FITC)-conjugated sheep anti-rat Ig was obtained from Silenus (Boronia, VIC, Australia).
Immunohistochemistry
For immunohistochemistry, freshly excised uteri were embedded vertically in OCT compound (Tissue Tex, Naperville, IL), and frozen immediately by immersion in liquid nitrogen-cooled isopentane. Air-dried tissue sections (thickness, 7 µm) were fixed in ethanol and labeled sequentially with rat mAbs, biotin-conjugated rabbit anti-rat Ig, and HRP- conjugated streptavidin. Within an experiment, uterine sections from all mice were processed together, and control sections were incubated in PBS with 10% normal mouse serum (NMS) or irrelevant, isotype-matched mAbs. Bound HRP was localized on sections with diaminobenzidine tetrachloride (DAB; Sigma, St Louis, MO). After counterstaining in Gill hematoxylin (Sigma), sections were dehydrated in absolute ethanol, cleared in Safsolvent (Ajax Chemicals, Auburn, NSW, Australia), mounted in Depex (BDH Laboratory Supplies, Poole, U.K.), and examined and photographed using an Olympus BH-2 light microscope (Tokyo, Japan). The DAB-labeled APCs in the endometrial stroma of duplicate sections from each uterus were quantified by video image analysis using VideoPro software (Faulding Imaging, Adelaide, SA, Australia). For 10 or more fields per section, data are expressed as the percentage positivity [(mean area of DAB stain/mean area of hematoxylin + DAB stain) x 100] for each tissue. Endogenous peroxidase activity was controlled for by subtracting the corresponding percentage positivity value in sections stained with isotype control mAbs.
For double-labeling immunohistochemistry, sections of uteri collected from estrous mice were sequentially labeled to detect F4/80 using a two- step Fast Red detection system (Sigma) and to detect MHC class II using a two-step DAB detection system (Sigma).
Preparation of Uterine Cell Suspensions
To obtain suspensions of uterine cells for flow cytometric analysis and cell culture, a modification of a method described previously [36] was used. Briefly, uteri were collected aseptically, trimmed of mesentery and fat, slit longitudinally with the aid of a dissecting microscope (Olympus SZ-PT), and then minced extensively using fine scissors. The tissue fragments of individual or pooled uteri were stirred gently at room temperature for 2 h in 1–2 ml/uterus collagenase/DNase (1 mg/ml of type I collagenase [Sigma] and 2.5 µg/ml of DNase I [Sigma] in RPMI 1640 [Sigma] with 10% heat-inactivated fetal calf serum [FCS; JRH Biosciences, Lenexa, KS]). Tissue clumps were disrupted by manual pipetting throughout the digestion. After 2 h, an equal volume of ice-cold, 5 mM EDTA (BDH Laboratory Supplies) in Ca2+Mg2+-free Hanks buffered saline solution (Gibco, Auckland, New Zealand) with 5% FCS and 0.01% sodium azide ("flow buffer"; Sigma) was added and stirred for a further 20 min. The digestion mixture was then filtered to remove undigested tissue and cell clumps, and the resulting uterine single-cell suspension (approximately 107 cells obtained per uterus) was washed and used for flow cytometric analysis or APC purification.
Purification of APCs from Uterine Cell Suspensions
For collection of cells expressing F4/80, uterine cell suspensions were incubated (1 h, 4°C) at 105–106 cells/ml in F4/80 hybridoma supernatant diluted 1:2 in RPMI-FCS with 10% heat-inactivated NMS. After one wash, cells (107 cells/ml) were incubated (30 min, 4°C) in RPMI-FCS with 10% heat-inactivated NMS and 20% (v/v) miniMACS goat anti-rat Ig- coated microbeads (Miltenyi Biotech, Bergisch Gladbach, Germany). Cells were washed twice, resuspended to 106–107 cells/ml, and then sorted magnetically using a miniMACS separation column (Miltenyi Biotech) at 4°C into F4/80-negative (–) and F4/80-positive (+) fractions according to the manufacturer's instructions. To collect MHC class II+/F4/80– uterine cells, the F4/80– fraction was subsequently incubated in TIB120 hybridoma supernatant diluted 1:2 in RPMI-FCS with 10% heat-inactivated NMS and processed using miniMACS to collect MHC class II+ cells. Addition of FITC-conjugated sheep anti-rat Ig and examination by ultraviolet microscopy revealed that more than 95% of selected cells expressed F4/80 and MHC class II, respectively.
Flow Cytometry
For single- and dual-color flow cytometric analysis, cells obtained by enzymatic digestion of uteri pooled from estrous or ovariectomized mice were directly or indirectly labeled with FITC and PE to detect cell-surface markers F4/80, MHC class II, CD11b, class A scavenger receptor, macrosialin, sialoadhesin, 33D1, DEC205, CD11c, and CD1. Briefly, aliquots containing 5 x 105 uterine cells were incubated in neat hybridoma supernatant with 10% heat-inactivated NMS or in flow buffer containing unconjugated, FITC-conjugated, or biotin-conjugated mAbs at approximately 10 µg/ml with 10% heat-inactivated NMS. Unconjugated rat Igs were subsequently detected by incubation in FITC-conjugated goat anti-rat Ig (diluted 1:40) or in biotin-conjugated rabbit anti-rat Ig (diluted 1:300) followed by incubation in PE-conjugated streptavidin (diluted 1:10). Biotin- conjugated mAbs were labeled by incubation in PE-conjugated streptavidin (diluted 1:10). Aliquots of cells were also labeled using isotype- matched control rat and hamster mAbs for each of the labeling steps in all experiments. In some experiments, cells were labeled in sequence to detect two cell-surface antigens. All incubations were 45 min at 4°C. Unbound antibodies were removed by three washes in 2 ml of ice-cold flow buffer after each labeling step. After the final washes, cells were stored (1–6 days) in the dark at 4°C in 0.5–1 ml of fresh 1% (w/v) buffered paraformaldehyde (Sigma) until analysis. Quantification of the number of labeled cells and intensity of FITC- and PE-labeling was analyzed after the establishment of regions and quadrants defining background fluorescence with reference to isotype control-labeled cells.
Collection of Splenic Lymphocytes
To purify splenic lymphocytes, spleens were manually homogenized in PBS-FCS, tissue debris was removed by filtration, and red blood cells were depleted by flash lysis. The remaining cells were then washed twice in 25 ml RPMI (Sigma) and incubated (1 h, room temperature) at 105– 106 cells/ml in F4/80, TIB120, and TIB227 hybridoma supernatants diluted 1:2 (v/v) in RPMI-FCS with 10% NMS. Following one wash, cells were resuspended at 105–106 cells/ml in RPMI with complement-rich guinea pig serum (diluted 1:20; Sigma) and incubated (1 h, 37°C) in a 10- cm Petri dish (Sarstedt, Adelaide, SA, Australia). Remaining nonadherent, viable cells (referred to as "purified splenic lymphocytes") were collected, filtered through a sterile 70-µm membrane (Falcon Labware, Sydney, NSW, Australia), washed three times, and immediately used in proliferation assays.
In Vitro Proliferation Assays
In vitro proliferation assays were performed in 200-µl, flat-bottomed, 96-well plates (Costar, Cambridge, MA) using complete media (RPMI- FCS supplemented with 5 x 10–5 M ß2-mercaptoethanol, L-glutamine, and antibiotics). In MLRs, F4/80+ and F4/80–/MHC class II+ uterine cells obtained from C57Bl/6 females were incubated with purified splenic lymphocytes obtained from Balb/c females. In PHA assays, F4/80+ and F4/ 80–/MHC class II+ uterine cells obtained from Balb/c F1 females were incubated with purified splenic lymphocytes obtained from Balb/c F1 females in the presence of 10 µg/ml of PHA (Sigma). To quantify proliferation, cells were pulsed with 1 µCi/ml of [3H]thymidine (Amersham, Chicago, IL) for the final 24 h of culture, and radioactivity was measured in harvested cells as disintegrations per minute in a liquid scintillation beta counter (Beckman Instruments, Inc., South Pasadena, CA).
Statistical Analysis
Data were analyzed by nonparametric analysis. Groups assigned a P value of less than 0.05 by Kruskal-Wallis one-way ANOVA were subsequently analyzed using the Mann-Whitney rank sum test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abundant populations of F4/80+ macrophages in the uterus of mice have been described [24, 28–30, 37], but to our knowledge, the presence of dendritic cells and their potential relationship with macrophages has not been studied. Initially, we undertook an immunohistochemical evaluation to determine if subsets of uterine APCs could be identified on the basis of distinct expression patterns of six macrophage and dendritic cell markers. The morphology and abundance of cells expressing F4/80, class A scavenger receptor, macrosialin, sialoadhesin, MHC class II (Ia), and B7-2 were investigated in tissues collected from estrous mice.
At estrus, uteri showed the characteristic appearance of being edematous and highly vascularized. Histologically, a thick endometrial layer with a large convoluted lumen and numerous endometrial glands were evident. The endometrium contained abundant cells expressing F4/80 (Fig. 1A), class A scavenger receptor (Fig. 1B), macrosialin, and sialoadhesin (not shown). Cells expressing MHC class II (Fig. 1C) were relatively less abundant, and B7-2+ cells were rarely observed. Cells labeled with different mAbs showed very similar patterns of localization, being distributed throughout the endometrium and evident immediately subjacent to the basal aspect of luminal and glandular epithelial cells. Most labeled cells were large and rounded in appearance, although some cells with an irregular dendriform shape were also present (Fig. 1, A–C). Cells expressing F4/ 80, macrosialin, sialoadhesin, and, to a lesser extent, MHC class II were also evident in the circular and longitudinal layers of the myometrium and in the mesometrial triangle, but cells expressing class A scavenger receptor and, particularly, B7-2 were relatively uncommon in these tissues.
|
Effect of Ovariectomy on APC Marker Expressionin Uterine Tissue
The F4/80+ uterine macrophages are reported to be dependent on ovarian steroid hormones for maintenance within the endometrial environment [28, 30, 38]. We examined the steroid hormone dependence of uterine cells expressing F4/80, class A scavenger receptor, macrosialin, sialoadhesin, MHC class II, and B7-2 by comparing the relative densities of labeled cells in the uteri of estrous and ovariectomized mice. Compared to uteri collected from estrous mice, uteri after ovariectomy were reduced in weight by more than 2-fold, were less vascular, and by histological examination exhibited a much thinner endometrium with few glands and a markedly reduced luminal area. Labeled cells in uteri of ovariectomized mice were relatively small and had a dendriform morphology. This was particularly evident in MHC class II+ cells, which were notable for their intensely labeled cell membranes (Fig. 1G). In comparison, cells expressing F4/80 (Fig. 1D), macrosialin, sialoadhesin, and, particularly, class A scavenger receptor (Fig. 1F) were only weakly labeled.
We observed a significant effect of ovariectomy on the abundance of uterine cells expressing F4/80, class A scavenger receptor, macrosialin, and sialoadhesin. Most strikingly, class A scavenger receptor+ cells were severely diminished by ovariectomy (P = 0.003) (Fig. 2B). The F4/ 80+ and sialoadhesin+ endometrial cells were sparsely distributed in endometrial tissues of ovariectomized mice, with a more than 8-fold reduction in density (P 
0.01) (Fig. 2, A and D). The macrosialin+ cells were also observed in the endometrium of ovariectomized mice, but their abundance was reduced by 3-fold compared to that at estrus (P < 0.004) (Fig. 2C). In contrast, the density of MHC class II+ endometrial cells (Fig. 1G) was not significantly affected by ovariectomy (P > 0.3) (Fig. 2E). Similarly, after ovariectomy, the abundance of MHC class II+ cells in the mesometrial triangle and the myometrium appeared similar to that of estrus, whereas cells expressing markers such as F4/80 and macrosialin were markedly fewer in number in these regions of the uterus. The B7-2+ cells were not evident in any part of the uterus of ovariectomized mice (Fig. 2).
|
Flow Cytometric Analysis of Subpopulations of F4/80+ and MHC class II+ Uterine Cells
Heterogeneity in the relative abundance of F4/80+ and MHC class II+ uterine cells in estrous and ovariectomized mice led us to focus on these two markers in particular. Initially, we utilized dual-color immunohistochemistry to analyze coexpression of F4/80 and MHC class II by uterine cells at estrus. The dual-labeling procedure detected cells with a similar morphology to those identified by single- color immunohistochemistry (Fig. 1); however, this approach also allowed identification of distinct subpopulations of F4/80+ and MHC class II+ cells (Fig. 3A). Cells that expressed F4/80 but not MHC class II (F4/80+/MHC class II–) or both F4/80 and MHC class II (F4/80+/MHC class II+) were evident in the superficial and deep endometrium as well as in the myometrium. A smaller population of cells, which expressed MHC class II but not F4/80 (F4/80–/MHC class II+), was also evident within the endometrium.
|
To better characterize and enumerate subpopulations of F4/80+ and MHC class II+ uterine cells, flow cytometry was utilized. The mixed population of single cells collected after enzymatic digestion of uteri collected from estrous mice produced a characteristic and highly reproducible profile by forward- and side-scatter analysis (Fig. 3B). The majority of released cells were determined to be leukocytes (60%) and epithelial cells (36%) with the use of mAbs specific for leukocyte common antigen and MTS number 24, respectively (data not shown). Expression of F4/80 and/or MHC class II was evident on 10–15% of all uterine cells, with the three phenotypes observed by dual-color immunohistochemistry also evident by flow cytometry. In five individual experiments, the populations, as a percentage of total cells, were approximately 6% F4/80+/MHC class II–, 4% F4/80+/MHC class II+, and 3% F4/80–/MHC class II+ (Table 1 and Fig. 3C).
|
Each of the three populations could be selectively enriched through the application of mutually exclusive gates to the forward- and side-scatter profiles of mixed uterine cells (Fig. 3B and Table 1). A high proportion (
85%) of F4/80+/MHC class II– cells was confined to gate 1 (Fig. 3D), a region comprising relatively small cells with medium to high intracellular complexity (Fig. 3B). The majority (50–65%) of F4/80+/MHC class II+ cells were contained within gate 2 (Fig. 3E), which enclosed relatively large cells with medium-high intracellular complexity (Fig. 3B). More than 70% of F4/80–/MHC class II+ cells were found within gate 3 (Fig. 3F), a region enclosing cells of low to medium size and relatively low intracellular complexity (Fig. 3B).
Effect of Ovariectomy on F4/80+ and MHC Class II+ Uterine Cells
A differential effect of ovariectomy on uterine leukocytes was clearly evident by immunohistochemistry, whereby MHC class II+ cells were resistant to removal of ovarian steroid hormones but cells expressing other markers were severely depleted (Figs. 1 and 2). To further characterize the specific subpopulations of leukocytes that were ovarian hormone-sensitive and -resistant, F4/80+ and MHC class II+ uterine cells obtained from the uteri of ovariectomized mice were analyzed by flow cytometry and compared to those of estrous mice.
Consistent with the immunohistochemical results, the total number of MHC class II+ uterine cells was not altered by removal of ovarian steroid hormones: Both F4/80+/ MHC class II+ and F4/80–/MHC class II+ cells comprised similar proportions of total cells after ovariectomy as were observed at estrus (Table 1). Thus, regardless of ovarian steroid hormone status, MHC class II+ cells formed approximately 7% of cell suspensions. In contrast, cells released from uteri of estrous mice consistently contained approximately 10% F4/80+ cells, but digests from ovariectomized mice contained only 4.4–4.8% F4/80+ cells, a deficit of more than 50%. This difference appeared to be caused by a preferential loss of F4/80+/MHC class II– cells, with at least 10-fold fewer cells of this phenotype following ovariectomy (Table 1).
Expression of Class A Scavenger Receptor, Macrosialin, Sialoadhesin, and CD11b by F4/80+ and F4/80– Uterine Cells
To further explore heterogeneity in uterine APCs, we examined the expression of class A scavenger receptor, macrosialin, sialoadhesin, and CD11b on F4/80+ and F4/80– uterine cells at estrus. Of F4/80+ cells in gate 1, 6–8% expressed class A scavenger receptor (Fig. 4A) and sialoadhesin (Fig. 4C), and 10–11% expressed macrosialin (Fig. 4B). A higher proportion of F4/80+ cells in gate 2 coexpressed these markers: 50% expressed class A scavenger receptor (Fig. 4E), 46% expressed macrosialin (Fig. 4F), and 42% expressed sialoadhesin (Fig. 4G). Similar proportions of F4/80+ cells in gates 1 and 2 (65–70%) expressed CD11b (Fig. 4, D and H).
|
Apart from a small population of F4/80–/macrosialin+ cells in gate 1 (Fig. 4B), essentially no class A scavenger receptor+, macrosialin+, or sialoadhesin+ uterine cells were found that did not coexpress F4/80.
Expression of 33D1, DEC205, CD11c, and CD1by F4/80+ and MHC Class II+ Uterine Cells
To determine whether markers associated with myeloid dendritic cells were present on uterine cells, we examined the expression of 33D1, DEC205, CD11c, and CD1 by F4/ 80+ and MHC class II+ uterine cells at estrus. All four markers were detected on varying proportions of cells in each of gates 1, 2, and 3 (Table 2). Among the F4/80+ cells in gate 1, 8% expressed 33D1, 5% expressed DEC205, 82% expressed CD11c, and 22% expressed CD1. In contrast, higher proportions of F4/80+ cells in gates 2 and 3 coexpressed dendritic cell markers: 25–29% expressed 33D1, 42–43% expressed DEC205, 43–70% expressed CD11c, and 89–100% expressed CD1.
|
The MHC class II+ uterine cells also expressed dendritic cell markers. Of MHC class II+ cells in gate 2, 55% expressed 33D1, 44% expressed DEC205, 57% expressed CD11c, and 77% expressed CD1 (Fig. 5, A–D). The MHC class II+ cells in gate 3 showed a similar phenotype: 23% expressed 33D1, 33% expressed DEC205, 56% expressed CD11c, and 95% expressed CD1 (Fig. 5, E–H).
|
Immunoaccessory Function of F4/80+ and MHC Class II+ Uterine Cells
We utilized in vitro MLR and PHA assays to compare the immunoaccessory phenotypes of F4/80+ and F4/80–/ MHC class II+ uterine cells purified by antibody-mediated affinity isolation on magnetic beads. Both assay formats revealed differences between the capacity of the two cell populations to stimulate proliferation of purified splenic lymphocytes. Depending on cell concentration, F4/80–/ MHC class II+ uterine cells were 2- to 10-fold more stimulatory of allogeneic lymphocyte proliferation than were F4/80+ uterine cells (Fig. 6A). The F4/80–/MHC class II+ uterine cells were approximately twice as stimulatory of PHA-induced syngeneic lymphocyte proliferation than were F4/80+ uterine cells at a ratio of 1:10 ratio of stimulator to responder cells (Fig. 6B).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The first aim of the present study was to identify whether a broad array of macrophage- and dendritic cell-specific mAbs could be used to identify subpopulations of uterine APCs by immunohistochemistry, as has been possible in other mucosal and lymphoid organs [8]. Abundant expression of class A scavenger receptor and macrosialin suggests that uterine APCs have the capacity to bind and internalize antigens rich in lectins [11], oxidized low-density lipoproteins [39], maleylated residues [40], polyanionic ligands [41], or even those of neighboring cell membranes [10]. Other authors have detected uterine cells expressing class A scavenger receptor and macrosialin during mid to late pregnancy [42, 43], but to our knowledge, their presence during the estrous cycle has not been described previously. Expression of sialoadhesin by uterine APCs is consistent with the capacity to be involved in adhesive, sialic acid- dependent interactions with other leukocytes [12, 34], including CD4+ lymphocytes [44]. However, the similar morphologies and comparable densities and distribution patterns of endometrial cells expressing these markers did not allow identification of distinct subpopulations of APCs on the basis of immunohistochemistry alone.
To further investigate heterogeneity in uterine APCs, we used flow cytometry to examine cells released from enzymatically digested uteri. Similar to previous studies [28, 37, 45], F4/80+ cells were shown to comprise approximately 10% of released cells at estrus. Two clearly distinct populations of F4/80+ cells could be distinguished in the uterus of estrous mice: a relatively homogeneous group of undifferentiated macrophages, and a population exhibiting heterogeneous features consistent with a more activated, mature phenotype. The first population, which comprised approximately 60% of total F4/80+ cells, did not express MHC class II and displayed a forward- and side-scatter profile indicative of relatively small size and intermediate cytoplasmic complexity. These F4/80+ cells showed other features of a relatively nonactivated phenotype, expressing little or no class A scavenger receptor, macrosialin, or sialoadhesin. The expression of both CD11b and CD11c by the majority of these cells is suggestive of recent interaction with endothelial cell adhesion molecules during recruitment from the blood and of their retention within the tissue by interacting with local ligands [15].
A second population of macrophages, comprising approximately 40% of total F4/80+ cells, was initially differentiated from the other F4/80+ cells on the basis of larger size and surface expression of MHC class II. These cells appeared to be of a considerably more activated phenotype than the F4/80+/MHC class II– cells, with expression of CD1 in addition to MHC class II indicative of a role in presentation of both protein and glycolipid antigens [17]. Such activity is also consistent with data showing that approximately half these cells expressed cell-surface class A scavenger receptor, macrosialin, and sialoadhesin, which is suggestive of their ability to internalize and process antigens [10, 11, 41] and to cluster T cells [44, 46].
The third population of APCs expressed MHC class II but not F4/80, a staining pattern suggestive of a lineage of dendritic cells. The F4/80–/MHC class II+ cells could be distinguished from the majority of differentiated macrophages based on their smaller size and lower density of intracellular organelles, which are characteristic features of dendritic cells from other mucosal tissues [25, 27]. A considerable proportion of the uterine F4/80–/MHC class II+ cells also expressed the cell-surface markers 33D1, DEC205, CD11b, and CD11c, which are expressed on myeloid dendritic cells in other peripheral organs [1, 2, 13, 15, 16]. Furthermore, as much as 95% expressed the antigen- presentation molecule CD1, which is found at high levels on splenic dendritic cells [17] and on monocyte-derived dendritic cells [20]. Finally, like equivalent cells in other tissues [4, 26, 27], F4/80–/MHC class II+ uterine cells displayed a potent immunostimulatory phenotype, which contrasted with the relatively poor capacity of uterine macrophages to stimulate lymphocyte proliferation as reported here and in previous studies [36, 47]. These data provide compelling evidence that the virgin mouse uterus hosts a population of myeloid dendritic cells.
Immunohistochemical and flow cytometric analysis of cells from ovariectomized mice allowed analysis of steroid hormone dependence among the three populations of uterine APCs. Previous experiments have shown that uterine macrophage numbers are dramatically reduced after ovariectomy [30, 48, 49], secondary to ablation of steroid hormone-regulated colony-stimulating factor (CSF)-1 expression in uterine epithelial cells [30, 48–50]. Our histochemical analysis similarly showed that ovariectomy resulted in a marked depletion of uterine F4/80+ cells and, furthermore, that the densities of cells expressing class A scavenger receptor, macrosialin, and sialoadhesin were dramatically reduced. In contrast, abundant MHC class II+ cells with a distinctive, highly dendriform morphology remained distributed throughout the uterine tissues after ovariectomy.
Flow cytometric analysis allowed the two phenotypes of F4/80+ cells to be distinguished and showed that undifferentiated macrophages were more significantly depleted following ovariectomy than their differentiated counterparts. The former cells presumably require active secretion of chemotactic molecules [45, 51] and growth factors, such as CSF-1 [24, 30], from steroid hormone-regulated uterine epithelial cells to be continually recruited or to remain viable. In contrast, mature macrophages and dendritic cells appeared to be relatively resistant to removal of steroid hormones, because they were present in comparable numbers after ovariectomy. The persistence of these cells could be a reflection of a relatively long life span, as seen in other populations of mature macrophages in mice [52], and indicates that trophic signals sufficient to maintain mature macrophages and dendritic cells within the local environment are synthesized in the uterus despite ovariectomy. It remains to be determined whether other phenotypic characteristics of mature macrophages and dendritic cells are also steroid hormone dependent; interestingly, our immunohistochemical data suggest that expression of class A scavenger receptor, macrosialin, and sialoadhesin by mature macrophages may be decreased following ovariectomy.
The developmental pathways by which uterine APCs are generated remain to be determined. It is reasonable to speculate that chemokines secreted by uterine epithelial cells [45, 51] recruit monocytes and other precursor cells to the uterus and that their subsequent development into APC subpopulations is driven by exposure to local cytokines and differentiating factors, including CSF-1, granulocyte-macrophage colony-stimulating factor (GM-CSF), interferon-
, and tumor necrosis factor 
[53]. A range of macrophage phenotypes might evolve depending on fluctuating concentrations of cytokines within the local microenvironment over the course of the estrous cycle and pregnancy [38]. An alternative differentiation outcome for precursor cells recruited to the uterus may be evolution into dendritic cells. In vitro studies suggest that cells having many features in common with peripheral tissue myeloid dendritic cells can develop from monocyte precursor cells [1], and F4/80+ macrophages recruited by localized GM-CSF expression have been shown to differentiate into dendritic cells in the lung [2]. Consistent with this possibility, and similar to descriptions of dendritic-like cells of the murine vagina [54], we found that both F4/80+ and MHC class II+ uterine cells can express dendritic cell markers. That MHC class II+ dendritic cells appear not to require ovarian steroid-driven cytokine or chemokine expression also supports the view that these cells differentiate in situ from F4/ 80+/MHC class II– precursor cells.
In conclusion, we have shown that the uterus of the cycling virgin mouse hosts at least three distinct populations of APCs. These include a relatively homogeneous population of steroid hormone-dependent undifferentiated macrophages, a more heterogeneous population of mature macrophages, and a population of myeloid dendritic cells. Our observations are consistent with a model for differentiation of mature macrophage and dendritic cell populations from undifferentiated precursor cells recruited into the uterus in response to steroid hormone-regulated signals. The present study lays the foundation for future investigations into the processes governing initiation of immune responses within the uterine environment and the molecular mechanisms by which uterine APCs can activate lymphocytes appropriate for antigens of diverse origin. Both mature APC populations express an array of antigen-binding molecules and ligands for lymphocyte activation that would well-equip these cells for dealing with the range of challenges comprised by microbial agents, sperm, and the conceptus. Plasticity in their ontogenetic and functional relationships would allow further complexity and versatility in the uterine immune response. Of particular interest will be to examine the recruitment and phenotypic modulation of uterine macrophages and dendritic cells during the inflammatory-like response to insemination [55] and as the antigenically disparate conceptus invades to form the placenta. Importantly, it remains to be determined how uterine APC populations are involved in generating the state of paternal antigen-specific T-cell tolerance that is required to accommodate pregnancy [51, 56, 57].
|
FOOTNOTES |
|---|
2 Correspondence: Sarah Hudson Keenihan, Department of Obstetrics and Gynaecology, University of Adelaide, Adelaide, SA 5005, Australia. FAX: 61 8 303 4099; sarah.hudson-keenihan{at}adelaide.edu.au
Received: 29 October 2003.
First decision: 4 December 2003.
Accepted: 22 January 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. M. Moldenhauer, K. R. Diener, D. M. Thring, M. P. Brown, J. D. Hayball, and S. A. Robertson Cross-Presentation of Male Seminal Fluid Antigens Elicits T Cell Activation to Initiate the Female Immune Response to Pregnancy J. Immunol., June 15, 2009; 182(12): 8080 - 8093. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Hume Macrophages as APC and the Dendritic Cell Myth J. Immunol., November 1, 2008; 181(9): 5829 - 5835. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T.-D. Liao, X.-P. Yang, Y.-H. Liu, E. G. Shesely, M. A. Cavasin, W. A. Kuziel, P. J. Pagano, and O. A. Carretero Role of Inflammation in the Development of Renal Damage and Dysfunction in Angiotensin II-Induced Hypertension Hypertension, August 1, 2008; 52(2): 256 - 263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Boyson, N. Nagarkatti, L. Nizam, M. A. Exley, and J. L. Strominger Gestation stage-dependent mechanisms of invariant natural killer T cell-mediated pregnancy loss PNAS, March 21, 2006; 103(12): 4580 - 4585. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E Minge, N. K. Ryan, K. H. V. D. Hoek, R. L. Robker, and R. J. Norman Troglitazone Regulates Peroxisome Proliferator-Activated Receptors and Inducible Nitric Oxide Synthase in Murine Ovarian Macrophages Biol Reprod, January 1, 2006; 74(1): 153 - 160. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Flores-Langarica, S. Meza-Perez, J. Calderon-Amador, T. Estrada-Garcia, G. MacPherson, S. Lebecque, S. Saeland, R. M. Steinman, and L. Flores-Romo Network of dendritic cells within the muscular layer of the mouse intestine PNAS, December 27, 2005; 102(52): 19039 - 19044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Gangnuss, M. L. Sutton-McDowall, S. A. Robertson, and D. T. Armstrong Seminal Plasma Regulates Corpora Lutea Macrophage Populations During Early Pregnancy in Mice Biol Reprod, October 1, 2004; 71(4): 1135 - 1141. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S O'Leary, M J Jasper, G M Warnes, D T Armstrong, and S A Robertson Seminal plasma regulates endometrial cytokine expression, leukocyte recruitment and embryo development in the pig Reproduction, August 1, 2004; 128(2): 237 - 247. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
