The peritoneum is more than a slippery, nonadhesive, and protective surface, or a mere mechanical covering to ease the gliding of opposed peritoneal surfaces. It also undertakes pivotal roles including the following: involvement in the transport of fluid and solutes between the circulation, the interstitial space, and the body cavities; antigen presentation; inflammation and tissue repair; coagulation and fibrinolysis; and tumor cell adhesion (Mutsaers, 2004). The perception of how the visceral mesothelium actively participates in clinical processes such as peritoneal dialysis, peritonitis, and peritoneal adhesion formation, as well as the formation and adhesion of some types of tumors, has given recent impetus to ultrastructural studies in this field.
Although the mesothelium is described as a layer of predominantly flattened, squamous-like cells, cuboidal mesothelial cells are found in the serosa covering different regions and organs of several species (Bird, 2004; Mutsaers, 2004). Changes in morphology appear to be associated with the biosynthetic activation status of the cells, with biosynthetically induced mesothelial cells displaying a tight cuboidal morphology, in contrast to their flatter, biosynthetically quiescent counterparts (Bird, 2004). Other signs of mesothelial cell activation are the abundance of microvilli, of mesothelial primary cilia (solitary cilia), and of secretory vesicles in their surfaces (Bird, 2004).
The serosal surface in the vicinity of the genital tract has been described in the golden hamster for the complete ovarian bursa (Martin et al., 1981), in the sow for the broad ligament (Doboszynska et al., 1999), and in the cow for the genital tract and adjacent structures (Yaniz et al., 2000). This study was undertaken to describe the surface features of the peritoneal mesothelium covering the genital tract and adjacent structures of the sow, and presents additional features that have not previously been described.
MATERIALS AND METHODS
A total of 14 large White × Landrace sows, 7 in the follicular phase and 7 in the luteal phase of the estrous cycle, were used in the optical microscopy and scanning electron microscopy (SEM) studies. Five additional sows, three in the follicular phase and two in the luteal phase of the estrous cycle, were used in the TEM study. They were checked twice daily with a mature boar to establish the onset of estrus. Day 1 corresponds to the follicular phase and days 11–14 to the luteal phase. The genital organs and adjacent ligaments were obtained 15–30 min after slaughter in an abattoir, and fixed by total immersion in a solution of 2.5% glutaraldehyde in phosphate buffered saline (PBS, pH 7.4).
Immediately after slaughter, segments adjacent to the SEM samples (see below) were rapidly cut, fixed with 4% formaldehyde in PBS (pH 7.4), dehydrated, and embedded in paraffin. The segments were cut into 5-μm-thick sections, stained with hematoxylin and eosin, and observed and photographed with a Leica DM 4500 (Wetzlar, Germany) microscope.
Tissue blocks from both the right and left sides of the genital area of each animal were prepared for SEM. Tissue specimens of 0.5 to 1 cm2 were taken from the following zones: the infundibulum, at a point adjacent to the free margin and another 1–2 cm distal; the ampulla and adjacent mesosalpinx (mesotubarium inferius), 5–6 cm distal to the ostium abdominale; the isthmus and adjacent mesosalpinx, 5–6 cm proximal to the utero-tubal junction; and the mesometrium, 2–3 cm distal to the utero-tubal junction. For the samples from the isthmus, ampulla, and mesosalpinx, both internal and external surfaces in relation to the ovarian bursa were studied, whereas for the remaining samples only the external surface was studied.
The pieces were spread on a flat surface, pinned, and totally immersed in 2.5% glutaraldehyde (Prolabo, Fontenay S/ Bois, France) in PBS (pH 7.4) for 24 hr. Fixed tissues were rinsed in PBS (pH 7.4), post-fixed for 2 hr in 1% osmium tetroxide (Merk, Darmstadt, Germany), and washed again in PBS. Fixation and washing were carried out at 4°C, and the tissues were then dehydrated in graded ethanol (25–100%) and substituted with acetone. Specimens were subjected to critical-point drying using liquid CO2 substitution (Anderson, 1951). The dried specimens were mounted on aluminium stubs, coated with gold in a Balzers Sputter Coater (Liechtenstein), and examined and photographed in a Zeiss DSM 940 (Oberkochen, Germany) scanning electron microscope at 15 kV.
Initial sampling and fixation methods were similar to those described for SEM pieces. After fixation and washing, small pieces of tissue (approximately 1 × 2 mm) were obtained from each zone, post-fixed in osmium tetroxide, dehydrated in ethanol, and embedded in araldite. The araldite blocks were trimmed and sectioned. Semi-thin sections (1 μm thickness) were stained with toluidine blue for evaluation under light microscopy. Subsequently, ultra-thin sections from selected areas were cut using a Reichert ultramicrotome and a diamond knife. The ultra-thin sections were picked up onto uncoated copper grids, counterstained with uranyl acetate and lead citrate, and examined as well as photographed in a Zeiss EM910 electron microscope at 80 kV.
The mesothelium covering the genital tract and adjacent structures in the sow is a single layer of loosely attached epithelial cells, with either a flattened or cuboidal appearance (Figs. 1–3). Cylindrical cells were also found in the infundibulum and the ampulla, being particularly abundant in the former (Fig. 1).
The SEM observations on the external side of the infundibulum revealed that the oviductal epithelium from the mucosa exceeded the free margin of the external side of the infundibulum, continuing to its external face (Fig. 4), forming a continuous band, where the presence of polypoid processes was frequently observed. The cells lining polypoid processes were predominantly secretory cells, showing numerous vesicles on their surface (Fig. 4). This oviductal epithelium showed cyclical variations with a predominance of ciliated cells during the follicular phase, while bulbous processes of secretory cells were predominant throughout the epithelial surface in the luteal phase. Transition between the mucosa and serosa was gradual, with oviductal epithelium, characterized by the presence of multiciliated cells, penetrating between mesothelial cells, either individually or in groups forming islets (Fig. 5). The presence of groups of cells from the oviductal epithelium was also observed in the serosa covering the ampulla (Fig. 6).
Scanning electron micrographs of the serosal surface of the peritoneum showed a mat of long microvilli in all specimens studied. Frequently, the surface was so densely covered by microvilli that the limits between cells were indistinguishable, making it difficult to precisely measure the surface area of cells (Fig. 7). In the few samples showing discernible cell limits, the surface area ranged between 27.9 and 251.8 μm2 (112.1 ± 102.4).
Numerous bulbous processes with a diameter ranging between 0.62 and 3.09 μm (1.51 ± 0.53) on the exposed surface of mesothelial cells were detected in most of the samples studied (Figs. 7–10). Their densities ranged between 5 and 36 processes/1,000 μm2 (23.5 ± 11.4). Where a bulbous process had been previously released, the cellular surface acquired a typical crater-like morphology (Figs. 7, 8). Furthermore, the presence of primary cilia in mesothelial cells was observed for all peritoneal surfaces, but was more frequent in transitional areas between the mesothelium and the oviductal epithelium from the infundibulum (Fig. 11).
TEM demonstrated the presence of rounded mesothelial cells, often with a basal process projecting into the underlying stroma. The nucleus showed an irregular outline and a prominent nucleolus (Fig. 12). The cytoplasm contained many free-ribosomes, abundant rough endoplasmic reticulum, well-developed Golgi apparatus (frequently various per cell), as well as a prominent vesicular system composed of vesicles, vacuoles, vesiculo-vacuolar complexes, and multivesicular bodies (Figs. 12–14). Single lamellar bodies showing membranes with identical thickness and concentric arrangement were disposed in the mesothelial cytoplasm (Fig. 12). The intercellular boundaries are complex and tortuous, with cells connected by tight junctions and a spot desmosome near the free surfaces of cells. Intercellular dilatations were frequently observed, usually associated with vacuoles or lamellar bodies that varied in size. Vesicles, vacuoles, and multivesicular bodies were projected to the external surface of the cells, showing an active process of secretion (Figs. 12, 13).
Anatomical findings for the serosal surfaces revealed closely comparable structures covering both the serosal surface of the reproductive tract and the adjacent ligaments. Similarly, no clear differences were found to be associated with the side of the ovarian bursa (internal vs. external), the laterality of the ovarian bursa (right vs. left sides), or the phase of the estrous cycle.
The results of this study highlight several new findings about the peritoneal mesothelium relating to the genital area and associated ligaments of the sow. Among these findings, of particular interest are the presence of cells of the oviductal epithelium in the serosa of the infundibulum and the ampulla, as well as indications of a high functional activity of the mesothelial cells in the areas studied. Oviductal epithelial cells were identified as being multiciliated, whereas mesothelial cells showed single or no cilia on their surfaces.
When considering the presence of epithelial cells from the oviductal mucosa in the serosal surface of the infundibulum and the ampulla, it should be recalled that the oviduct is the only tubular-shaped organ that opens toward the peritoneal cavity. This determines a continuity of the oviductal mucosa and peritoneal serosa in the infundibulum. In the cow, this transition was not located exactly in the free margin, as the oviductal mucosa exceeded this area, forming a continuous band so that transition between the mucosa and serosa was abrupt, and clearly distinguishable 2.5 to 10 mm from the free margin, while the presence of epithelial cells from the oviductal mucosa among the mesothelial cells of the infundibulum and the ampulla was not recorded (Yániz et al., 2000). The results of the present study indicate that, also in the pig, the transition is not produced in the free margin and, most noteworthy, it is not abrupt, with groups of cells or isolated cells of the oviductal epithelium throughout the external surface of the infundibulum and even in the ampulla. These cells of the oviductal epithelium are more prominent than mesothelial cells, they include ciliated and secretory cells and, in the histological sections, acquire a cylindrical morphology instead of a flat or cuboidal structure. As described previously for the mucosal surface of the infundibulum (Abe and Oikawa, 1992; Yániz et al., 2006), an extensive distribution of ciliated cells during the follicular phase and of bleb secretory cells during the luteal phase was also observed in the oviductal epithelium of the serosal side. During the follicular phase, the presence of a densely ciliated oviductal epithelium on the external side of the infundibulum and ampulla may be important in producing ciliary currents to ovum pick-up at ovulation. Participation of secretory cells in the formation of peritoneal fluid may be important in this species.
Despite the above indications, the presence of oviductal epithelial cells in the serosal layer of the infundibulum and ampulla may have a quite distinct and unusual origin. During mid-ventral laparotomy, one of the present authors has repeatedly observed that, shortly before the time of ovulation in gilts, the fimbriated extremity of the oviduct not only embraces the ovary but may also extend backward and envelop much of the ampulla, and even the isthmus and tip of a uterine horn (unpublished data). Thus, the endosalpingeal surface of the fimbriated infundibulum and ampulla would be making direct contact with the serosal surface of portions of the oviduct and uterine extremity. Frictional displacement of epithelial cells from the oviductal mucosa into the serosal surface could be envisaged, because of the heightened contractile activity of the ducts close to the time of ovulation.
For the biosynthetic activity of mesothelial cells, there is evidence that it is high in the peritoneal mesothelium of the female pig. TEM and histological sections showed a predominance of cuboidal mesothelial cells over flattened cells. The existence of flat and cuboidal mesothelial cells is widely reported in the literature, and numerous factors indicate that cuboidal mesothelial cells are more biosynthetically active than their squamous counterparts (Bird, 2004; Michailova and Usonoff, 2006). TEM observations revealed a cytoplasm richly supplied with organelles, reflecting high biosynthesis, and different structures projecting to the external surface of the cells. Lamellar bodies and multivesicular structures were seen in the cytoplasm and leaving the mesothelial surface by exocytosis. Complex multivesicular structures were also described in the peritoneum of the rat by Michailova (2004). This author proposed that most of these structures represent initial forms of lamellar bodies, difficult to distinguish by using routine fixation techniques. Lamellar bodies are secretory organelles found in type II pneumocytes (Smith et al., 1972; Douglas et al., 1975; Stratton, 1976) and, more recently, in mesothelial cells (Dobbie and Lloyd, 1989). Existence of lamellar bodies has been described in the mesothelium of different species (Dobbie, 1996; Michailova, 2004). Dobbie and Lloyd (1989) and Chailley-Heu et al. (1997) provided evidence that the mesothelium is specialized for biosynthesis of surfactant and to secrete lamellar bodies. The phospholipids stored in lamellar bodies, based on choline, serve as surfactant after being released from the cell. Surface-active material (surfactant, SAM), may have an important role in the regulation of peritoneal permeability and may be the major lubricant of serous cavities (Dobbie and Lloyd, 1989).
SEM observations showed that the cells of all areas studied have a high density of microvilli, which virtually impedes the measurement of the cell surface. Based on previous studies, Bird (2004) described mesothelial microvilli as labile and dynamic structures, which increased in quantity in response to higher cell biosynthetic activity. The presence of isolated cilia is also frequent. Recent studies have shown that a nonmotile primary cilium is not a vestigial membrane appendage, but rather a highly organized functional organelle that may have a major role in rapid cell response (Bird, 2004). Their abundance appears to be associated with the biosynthetic activation status of the individual cells, and they are more abundant on biosynthetically induced mesothelial cells that display a tighter cuboidal morphology than their flatter, biosynthetically quiescent counterparts (Bird, 2004). Cilia may grow in response to cell activation, analogous to microvilli, and both organelles may have a cooperative function (Madison et al., 1979; Doughty, 1998). Finally, for the majority of samples analyzed, bulbous protuberances and cavities, which appear after their liberation, were present, often in high densities. These bulbous formations probably correspond to the complex multivesicular structures described in TEM samples.
Taken together, histological and electron microscopy observations suggest that mesothelial cells covering the ovarian bursa in the sow are biosynthetically activated cells. Mesothelial cells of the areas studied are probably producing and releasing large amounts of surfactant. The presence of surfactant has been associated with different functions related with the inmunomodulation, fluid balance, lubrication, and protection (Hills, 1992; Johansson and Curstedt, 1997; Chen and Hills, 2000).
In contrast to previous studies (Doboszynska et al., 1999), we did not find clear stomata in the samples analyzed. Round openings were observed between the microvilli, but were associated to the release of round protrusions. Holes were observed in some samples, but were usually associated with artefacts.
The authors thank the Electron Microscopy Service of the University of Lleida for technical assistance, and the Matadero Municipal de Huesca and the The Pink Pig S.A. company for their help with procuring animal samples.