SEARCH

SEARCH BY CITATION

Keywords:

  • reptiles;
  • uterus;
  • viviparity

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Australian species of viviparous skinks have noninvasive epitheliochorial placentation where there is no breeching or interruption of the uterine epithelial cell barrier. This is contrary to some African and South American species of skinks which exhibit invading chorionic cells and a localized endotheliochorial placenta. The desmosomes, which maintain the adhesive properties of the junctional complex between uterine epithelial cells, were found to decrease as gestation progressed in the uterus of two highly placentotrophic Australian skinks, but no changes in desmosomal numbers were present in the uterus of two Australian oviparous skinks or viviparous skinks with a simple placenta. In mammals, desmosomes decrease in the uterine epithelium of species with invasive hemochorial placentation, where less chemical and mechanical adhesion between cells assists the invading trophoblast at the time of implantation. However, Australian viviparous skinks do not have an invasive trophoblast; yet, similarities in decreasing lateral cellular adhesion exist in the uterus of both invasive and noninvasive placental types. This similarity in cellular mechanisms suggests a conservation of plasma membrane changes across placentation irrespective of reptilian or mammalian origin. Anat Rec, 293:502–512, 2010. © 2010 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Desmosomes are a series of “spot welds” that attach adjoining epithelial cells along the lateral plasma membrane (Farquhar and Palade,1963) and occur principally in heavily stressed tissue, such as the heart and the epidermis, that experience distension forces (Dusek et al.,2007). Desmosomal adhesion molecules comprise four desmogleins (Dsg1-4) and three desmocollins (Dsc 1-3), which are Ca2+-binding desmosomal cadherins responsible for the filamentous appearance of the extracellular region between two desmosomal plaques (Burdett,1998; Bazzi and Christiano,2007). Desmosomal cadherins exhibit tissue-specific patterns of expression; desmoglein-2 and desmocollin-2 are expressed in all desmosome-containing tissue (Eshkind et al.,2002), whereas expression of desmoglein-1 and -3 is largely restricted to stratified epithelial tissue and desmoglein-4 is present in the hair shaft (Jahoda et al.,2004). Desmosomal ultrastructure as well as desmosomal proteins are highly conserved among vertebrates including humans, bovine, rats, trout, axolotls and lizards (Overton,1975; Greven and Robenek,1980; Cowin et al.,1984). Because of their adhesive properties, desmosomes would be expected to play an important role in maintaining the adhesive bond and integrity of the lateral plasma membrane of the uterine epithelium in viviparous and oviparous squamates due to the extensive stretching the uterus undergoes in response to embryonic growth. However, desmosomal distribution in other nonreptilian placental species indicates another possible function of the desmosome and it may be that similarities in lateral adhesion exist between both reptilian and mammalian placental species.

In species with hemochorial placentation, the junctional complex of the uterine epithelial cell barrier becomes compromised at the time of implantation, which allows the trophoblast cells to fuse, displace or invade between uterine epithelial cells (Schlafke and Enders,1967,1975; Finn and Lawn1968; Enders and Schlafke,1971; Lee et al.,1995; Igwebuike,2006). As a consequence of this intrusion, the junctional complex of the uterine epithelium, including the tight junction, adherens junction and desmosomes, is greatly affected (Enders and Schlafke,1971; Schlafke and Enders,1975). Changes in the junctional complexes between uterine epithelial cells are likely to facilitate trophoblast invasion (Illingworth et al.,2000), in particular changes in the number of desmosomes, which are the strongest attachment components of the junctional complex (Enders and Schlafke,1971). In both rats and mice, the number of desmosomes in uterine epithelial cells decreases during implantation and early pregnancy (Illingworth et al.,2000; Preston et al.,2004), resulting from a redistribution of desmosomes to the apical portion of the lateral plasma membrane as pregnancy proceeds (Preston et al.,2004). Thus, a decrease in lateral adhesion takes place between adjacent uterine epithelial cells during this period, leading to a decrease in cytoskeletal binding (Bazzi and Christiano,2007). Reduced desmosome numbers increases the plasticity of the uterine epithelium in preparation for implantation and so helps mediate the detachment of the epithelium (Illingworth et al.,2000; Preston et al.,2004). Additionally, trophoblast invasion of the uterine epithelium (Mossman,1974; Moffett and Loke,2006) effectively reduces the barrier for the transfer of water and transudates from the blood allowing hemotrophic nutrient transport (Freyer et al.,2003).

Unlike in the African lizard Trachylepis ivensi (Blackburn and Flemming,2009) where uterine epithelium is replaced by invading chorionic cells and in the South American Mabuya genus (Ramirez-Pinilla,2006; Vieira et al.,2007) which exhibited a localized endotheliochorial placenta, Australian viviparous skinks exhibit epitheliochorial placentation, where there is no invasion or breaching of the uterine epithelium (Amoroso,1952; Mossman,1974; Blackburn1993; Adams et al.,2005). In epitheliochorial placentation, the maternal and fetal capillaries are separated by six distinct layers, three from the maternal component and three from the fetal component (Grosser,1927). So to compensate for the barrier imposed by the epithelium, epitheliochorial squamates have evolved specializations such as histotrophy (Dantzer,1984; Adams et al.,2005; Biazik et al.,2007) to allow for effective nutrient transport via pathways other than blood in the omphaloplacentae. This is a form of placentotrophy whereby the developing embryo receives most of its nutrients from maternal placentotrophic transfer. The omphaloplacenta, which is formed by apposition between fetal and maternal tissue, is situated at the abembryonic pole of the egg and consists of tall, hypertrophied uterine epithelial cells that exhibit secretory properties, whereas membranes associated with gaseous exchange occur in the chorioallantoic placental region of the paraplacentome, where the uterine epithelium is attenuated and overlies an extensive vascular network (Blackburn,1993; Corso et al.,2000; Flemming and Branch,2001; Jerez and Ramirez-Pinilla,2001; Adams et al.,2005; Ramirez-Pinilla et al.,2006). Highly placentotrophic skinks also contain a placentome, which is comprised of interdigitations between the chorioallantoic membranes of the embryonic tissue and uterine epithelium, which exhibits similar morphology and ultrastructure as the omphaloplacenta (Stewart and Thompson,1998,2000; Corso et al.,2000; Jerez and Ramirez-Pinilla,2001; Blackburn and Vitt,2002) with a unique case of syncytial formation in the Mabuya species (Blackburn and Vitt,1992). Although highly specialized, the uterine epithelium in Australian viviparous skinks remains intact throughout gestation without any documented trophoblastic invasion (Mossman,1974; Adams et al.,2005; Biazik et al.,2008).

The aim of this study is to determine whether the adhesive properties between individual uterine epithelial cells change during the evolution of viviparity in Australian skinks to determine whether a reduction in desmosomes may help prime the uterus for invasive placentation. The use of Australian skink species assists this investigation because skinks exhibit a range of reproductive modes including oviparous populations and those with complex placentae. Changes in desmosomal numbers will be documented in the uterine epithelium of Saiphos equalis and Lerista bougainvillii, two species of Australian skinks from the Sphenomorphus lineage that exhibit bimodal reproduction (Smith and Shine,1997; Adams et al.,2007) and in two closely related highly placentotrophic skinks from the Eugongylus lineage, Pseudemoia entrecasteauxii and P. spenceri (Harrison and Weekes,1925; Weekes,1935; Stewart and Thompson,1996,1998). The microanatomy of uterine epithelial cells, in particular the lateral plasma membrane will be examined with transmission electron microscopy to conduct a desmosomal morphometric survey. In addition, an immunohistochemical analysis will be incorporated using anti- desmogelin-2 antibody to show desmosomal expression in the uterus. This investigation will potentially conclude whether mechanisms associated with lateral cellular adhesion are present in the viviparous species in this investigation, irrespective of whether they undergo invasive and noninvasive placentation and whether differences are observed between the two lineages of skinks chosen.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Species and Embryonic Stages

Pseudemoia entrecasteauxii and P. spenceri females were collected in Kanangra Boyd National Park, New South Wales for this investigation because they have the most complex reptilian placenta to be described in Australia (Harrison and Weekes,1925; Weekes,1935), with P. entrecasteauxii showing a higher degree of placentotrophy than P. spenceri. Two bimodally reproductive skinks, Saiphos equalis and Lerista bougainvillii were selected to provide separate intraspecific comparisons of oviparous populations and populations of viviparous skinks with simple placentae. Viviparous Saiphos equalis females were collected in Riamukka State Forest, New South Wales and oviparous S. equalis were collected from the University of Sydney grounds. Lerista bougainvilli females were collected from South Australia (the viviparous population was collected on Kangaroo Island and the oviparous population was collected on the adjacent mainland). Pseudemoiaentrecasteauxii and P. spenceri were transported to the University of Sydney and housed in glass aquaria lined with newspaper and rocks to provide basking sites. Aquaria were kept at 21°C with an 8:16-hr light/dark photoperiod. Lerista bougainvillii and S. equalis are fossorial; females were housed under similar conditions, but aquaria contained 20-mm thick substrate of moist peat moss for S. equalis and sand for L. bougainvillii which was collected from their habitat. Skinks were fed three times a week with crickets (Acheta domestica) dusted with phosphorus-free calcium (Per-Cal CA), and water was provided ad libitum in open dishes. Four individuals from each species were allocated to each of three different reproductive stages using the Dufaure and Hubert (1961) embryonic staging scheme: nonreproductive, early (embryonic stages 20–25), and late (38–40) stages. No individuals from oviparous populations could be sampled at the late gravid stage, because oviparous species oviposit at Stage 30 (Blackburn,1995).

Uterus Excision

At the appropriate reproductive stage, skinks were euthanized by an intrathoracic injection of 0.1 mL sodium pentobarbitome 60 mg/mL (Nembutal, Boehringer, Australia) and a central incision was made to expose the uterus. Incubation chambers of the uterus containing the embryos were isolated and uterine tissue was peeled away. The incubation chambers in P. entrecasteauxii and P. spenceri were separated into two regions, the chorioallantoic placenta (containing the placentome and paraplacentome) at the embryonic pole of the egg and the abembryonic region, the omphaloplacenta.

Transmission Electron Microscopy

Immediately following excision, tissues were fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer (PB) pH 7.4 for 1 hr, washed in three times 10 min changes of 0.1 M PB, and postfixed in a 1% osmium tetroxide solution containing 0.8% potassium ferrocyanide to enhance plasma membranes. Tissues were then washed in PBS, dehydrated in graded alcohols followed by acetone, and then infiltrated with 50% Spurr's resin/absolute acetone for 2 hr. The tissues were then re-infiltrated in fresh Spurrs resin (Agar Scientific, Essex, UK) overnight under gentle agitation and embedded in fresh Spurr's resin in flat cassette moulds (ProSci Tech, Queensland, Australia) and polymerized at 60°C for 24 hr. Three blocks per animal were cut using a Leica ultracut UCT ultramicrotome (Leica, Heerbrugg, Switzerland) and silver-gold sections (80–90 nm) were mounted onto 200-mesh copper grids. Sections were stained with a 2% solution of aqueous uranyl acetate for 10 min, followed by Reynold's lead citrate stain for 10 min. Stained sections were then viewed using a JEOL JEM-1010 (Tokyo, Japan) transmission electron microscope operating at 80 kV.

Cells that were selected for desmosome counting were longitudinally orientated, had a pronounced nucleus and the total lateral plasma membrane was visible from the apical border to the basal border. Cell height was calculated by averaging the height of four randomly chosen uterine epithelial cells and distance was measured by calibrating against a TEM measuring grid using Adobe Photoshop 6.0 software. Total desmosome numbers were counted on four randomly chosen lateral plasma membranes from four skinks allocated to each of the three reproductive stages. Four plasma membranes for each of the four animals were analyzed to determine the number of desmosomes. The total number of desmosomes per animal was averaged and this number was used for the statistical analysis.

Statistical Analysis

Statistical analyses were conducted using SAS Statistical Package version 9.1.2 (SAS Institute, 2004). Desmosome numbers in the uterine epithelium of P. entrecasteauxii and P. spenceri were compared between regions of the uterus and among embryonic stage using a two-factor analysis of variance (ANOVA). Differences in desmosome number as a function of location on the uterus was evaluated using a Student's t-test. Desmosome numbers in the uterine epithelium of the bimodally reproductive species L. bougainvillii and S. equalis as a function of parity modes were compared using a two-factor analysis of variance (ANOVA). The mean number of desmosomes is presented ± the standard error.

Antibody Used

An antibody directed against desmoglein-2 antigens was used in this study, thus, any fluorescent signal detected with the deconvolution microscope is that of desmoglein-2 expression.

Immunohistochemistry

Four sections (7 μm thick) were collected per gelatin coated slide using a Leica CM3050 cryostat at −21°C. Four slides were collected per animal. All slides were fixed in 100% acetone at −20°C overnight. Two experimental and two control slides were randomly selected. After 30 min of air drying, slides were incubated for 30 min in 1% Bovine serum albumin (BSA; Sigma-Aldrich, St Louis, MO) in phosphate buffered saline, pH 7.4 (PBS). Slides were incubated in 5 μg/mL mouse monoclonal antibody to desmoglein-2 (Progen Biotechnik GmbH, Germany) in 1% BSA/PBS for 1 hr at RT. After three rinses in PBS, sections were incubated in 14 μg/mL secondary fluorescein isothyocianate (FITC) conjugated goat anti-mouse IgG (Jackson ImmunoResearch Laboratories, Baltimore) for 30 min at RT. The positive controls for immunohistochemistry (skink and rat stomach and rat uterus) were treated with the same primary and secondary incubation protocol. Negative controls carried out on skink tissue and rat tissue were treated by omitting the primary antibody and nonimmune controls were carried out by substituting mouse matched IgG for the primary antibody. After rinsing with PBS, slides were mounted with vectashield (Vector laboratories, CA) and visualized using a Zeiss deconvolution microscope (Carl Zeiss Pty, Australasia) and images were acquired using a Zeiss AxioCam HR digital monochrome CCD camera. Digital images were taken using a Leica DFC480 digital camera (Carl Zeiss Pty. Ltd Australasia) Micrographs were imported into Adobe Photoshop 6.0 and a scale bar was added through the Axiovision 4.4 computer program.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Transmission Electron Microscopy Results for P. entrecasteauxii

As gestation progressed in P. entrecasteauxii, the stroma of the chorioallantoic placental region became more vascularized and the uterine epithelium overlying the developing capillaries was more attenuated (Fig. 1a). Cell height decreased from 5 μm ± 1 to 1 μm ± 0.2, reducing the surface area of the lateral plasma membrane available for desmosomes (Fig. 1b).

thumbnail image

Figure 1. A–D: Transmission electron micrographs of uterine epithelium in Pseudemoia entrecasteauxii. Lumen; (L), uterine capillaries; (uc), lysosomes (*) and desmosomes (white arrowheads). A: The uterine epithelium overlying the uterine capillary in the chorioallantoic placenta in early stage pregnancy (Stage 25) has not yet attenuated (3 μm thick). One desmosome is present along the plasma membrane. Scale bar = 1 μm. B: The attenuated uterine epithelium of the chorioallantoic placenta in late stage pregnancy (Stage 39) with one desmosome present along the attenuated (1 μm thick) plasma membrane between adjoining uterine epithelial cells. Scale bar = 1 μm. C: Junction between two uterine epithelial cells in early stage omphaloplacenta depicting 4–6 desmosomes along the lateral plasma membrane and lysosomes aggregating in the apical cytoplasm. Scale bar = 1 μm. D: In late stage pregnancy (stage 39) the uterine epithelium of the omphaloplacenta develops an extensive lysosome system in the apical cytoplasm and desmosome numbers decrease along the lateral plasma membrane. Scale bar = 1 μm. E–H: Transmission electron micrographs of the uterine epithelium in Pseudemoia spenceri. E: Several desmosomes present along the lateral plasma membrane between uterine epithelial cells of the chorioallantoic placenta in early stage pregnancy. Scale bar = 1 μm. F: In late stage pregnancy (Stage 39) the uterine epithelium of the chorioallantoic placenta attenuates over the uterine capillary (<1 μm thick). Desmosome numbers decrease along the lateral plasma membrane. G: Many desmosomes (5–6) present along the lateral plasma membrane of the uterine epithelial cells of the omphaloplacenta in early stage pregnancy and lysosomes begin to appear in the cytoplasm. Scale bar = 1 μm. H: In late stage pregnancy (Stage 38) the desmosome numbers decrease along the lateral plasma membrane and an abundant lysosomal system is established in the apical cytoplasm of uterine epithelial cells. Scale bar = 1 μm.

Download figure to PowerPoint

During early stage pregnancy, the height of uterine epithelium in the omphaloplacental region in P. entrecasteauxii ranged from 8 μm ± 3. Lysosomes aggregated in the apical region of the cytoplasm and desmosomes span the lateral plasma membrane at uniform intervals (Fig. 1c). As gestation progressed, an extensive lysosome system became established in the apical region of the cell cytoplasm and the number of desmosomes was reduced along the lateral plasma membrane (Fig. 1d).

Transmission Electron Microscopy Results for P. spenceri

Desmosomes were most numerous in uterine epithelium of the chorioallantoic placenta of P. spenceri during early stage pregnancy (Fig. 1e). Uterine epithelial cells of the chorioallantoic placenta ranged from 6 μm ± 2 in height. The number of desmosomes along the lateral plasma membrane fell as gestation progressed, and the uterine epithelium became more attenuated and reached ∼1 μm ± 0.3 in thickness (Fig. 1f), reducing the surface area of the lateral plasma membrane.

The height of uterine epithelial cells in the omphaloplacenta of P. spenceri did not change during pregnancy and ranged from 8 μm ± 2. During early stage pregnancy, 5–7 desmosomes were uniformly distributed along the lateral plasma membrane and lysosomes began to aggregate in the apical region of the cytoplasm (Fig. 1g). As pregnancy progressed, the number of lysosomes in the apical region of the cytoplasm increased and the number of desmosomes along the lateral plasma membrane decreased (Fig. 1h). Glands are also evident in late stage pregnancy in the omphaloplacental region with desmosomes present along the lateral plasma membrane of the glandular epithelium (Fig. 2g,h).

thumbnail image

Figure 2. A, B: Transmission electron micrographs of uterine epithelium in viviparous Saiphos equalis. Lumen; (L), uterine capillaries; (uc), lysosomes (*) and desmosomes (white arrowheads). A: Low power micrograph of the uterine epithelium (black arrowheads) in mid stage gestation showing attenuation of the epithelium over uterine capillaries and ciliated cells (c). Scale bar = 1 μm. B: Lateral junction between two uterine epithelial cells in late stage pregnancy (Stage 39) showing several desmosomes concentrated in the apical region of the lateral plasma membrane and lysosomes present in the apical cytoplasm. Scale bar = 1 μm. C, D: Transmission electron micrographs of uterine epithelium in oviparous Saiphos equalis. C: Several desmosomes present along a convoluted lateral plasma membrane in early stage pregnancy (Stage 27). Scale bar = 1 μm. D: Large desmosomes present in the apical region of the lateral plasma membrane in the uterine epithelium at the time of oviposition (Stage 30). Scale bar = 0.5 μm. EH: Transmission electron micrographs of the uterine epithelium in the bimodally reproductive Lerista bougainvilli. E: Lateral junction between uterine epithelial cells in late stage pregnancy (Stage 40) in viviparous population, depicting large desmosome in the apical region of the lateral plasma membrane and mitochondria (m). Scale bar = 0.5 μm. F: Large desmosomes (0.3 μm long) in the apical region of the lateral plasma membrane of adjoining uterine epithelial cells at the time of oviposition in oviparous population. Mitochondria (m). Scale bar = 0.5 μm. G: High power micrograph of three desmosomes (0.1 μm long) along the lateral plasma membrane in P. spenceriin early stage pregnancy (Stage 26) in the omphaloplacenta. Scale bar = 0.5 μm. H: Gland in the omphaloplacenta in late stage pregnancy in P. spenceri depicting several desmosomes between adjoining glandular epithelial cells and glandular lumen in the centre (L). Scale bar = 2 μm.

Download figure to PowerPoint

Transmission Electron Microscopy Results for S. equalis Species

As gestation progressed in viviparous S. equalis, the uterine epithelium overlying capillaries became more attenuated, some ciliated cells were present (Fig. 2a) and desmosomes were concentrated along the apical region of the lateral plasma membrane (Fig. 2b). The height of individual epithelial cells remained unaltered, 6 μm ± 2 as pregnancy progressed. The number of desmosomes did not change and the lateral plasma membrane became more interdigitated as pregnancy progressed (Fig. 2c). There was no change observed in the number of desmosomes between the embryonic and abembryonic poles. Similarly, the lateral plasma membrane of oviparous S. equalis became more interdigitated as gestation progressed. Desmosomes were predominantly found along the apical region of the lateral plasma membrane (Fig. 2d) and the number of desmosomes did not change during the reproductive cycle nor did cell height 5 μm ± 1. Ultrastructurally, desmosomes in S. equalis were twice the size of those in P. spenceri and P. entrecasteauxii.

Transmission Electron Microscopy Results for L. bougainvillii Species

In viviparous L. bougainvillii, mitochondria were abundant in the apical region of uterine epithelial cells and lysosomes numbers remained unchanged. In both viviparous (Fig. 2e) and oviparous (Fig. 2f) L. bougainvillii, desmosomes were concentrated in the apical region of the lateral plasma membrane, which becomes more interdigitated as gestation progressed. Cell height of uterine epithelial cells remained unchanged throughout the reproductive cycle, 6 μm ± 1 The number of desmosomes did not change during the reproductive cycle in either oviparous or viviparous L. bougainvillii and no change in the number of desmosomes was present between the embryonic and abembryonic poles, however desmosomes were twice the size of those in P. spenceri and P. entrecasteauxii.

Desmosome Numbers in P. entrecasteauxii

The number of desmosomes was significantly different at different reproductive Stages (F1,15 = 52.8, P < 0.001) and between embryonic and abembryonic hemispheres of the uterus (F1,15 = 47.4, P = 0.001; Fig. 3a). There was a decrease in the number of desmosomes in late pregnancy in the omphaloplacenta, whereas the number of desmosomes remained unchanged in the chorioallantoic placenta as gestation progressed. In the chorioallantoic placenta, the mean number of desmosomes along the lateral plasma membrane fell from 1.12 ± 0.21 in early stage pregnancy to 0.83 ± 0.28 late in pregnancy. In contrast, there was a five-fold decrease in the mean number of desmosomes along the lateral plasma membrane of uterine epithelial cells from the omphaloplacenta from 4.68 ± 0.27 in early pregnancy to 0.81 ± 0.02 in late pregnancy.

thumbnail image

Figure 3. A:Number of desmosomes in the uterine epithelium of P. entrecasteauxii.Chorioallantoic placental region in early (CA early) and late (CA late) stage pregnancy and omphaloplacental region in early (OM early) and late (OM late) stage pregnancy and nonreproductive stage (N/R). Asterisk indicates significantly (P < 0.01) a large number of desmosomes in the omphaloplacental region in early stage pregnancy in P. entrecasteauxii.B:Number of desmosomes in the uterine epithelium of P. spenceri.Chorioallantoic placental region in early (CA early) and late (CA late) stage pregnancy and omphaloplacental region in early (OM early) and late (OM late) stage pregnancy and nonreproductive stage (N/R). Asterisk indicates significantly (P < 0.01) a large number of desmosomes in the omphaloplacental region in early stage pregnancy in P. spenceri.C: Number of desmosomes in the uterine epithelium of bimodally reproductive species L. bougainvillii (Lb) and S. equalis (Se), with no significant changes observed between oviparous (O) and viviparous (V) populations and between nonreproductive animals (N/R). Error bars = mean ± S.E.

Download figure to PowerPoint

There is a strong interaction between the number of desmosomes as a function of stage and location (F1,15 = 38.45, P < 0.001), thus with increasing embryonic stage the number of desmosomes decreases in the omphaloplacenta but not in the chorioallantoic placenta. In contrast in the nonreproductive uterus, the mean number of desmosomes along the lateral plasma membrane was 4 ± 0.3.

Desmosome Numbers in P. spenceri

The number of desmosomes was significantly different at different reproductive stages (F1,15 = 24.77, P < 0.001) and between embryonic and abembryonic hemispheres of the uterus (F1,15 = 17.61, P = 0.001; Fig 3b). There was a decrease in the number of desmosomes in late stage pregnancy in the omphaloplacenta whereas the number of desmosomes remained unchanged in the chorioallantoic placenta as gestation progressed. In the chorioallantoic placenta, the mean number of desmosomes along the lateral plasma membrane fell from 1.63 ± 0.27 in early stage pregnancy to 1.31 ± 0.28 in late stage pregnancy. In contrast, the mean number of desmosomes along the lateral plasma membrane of the omphaloplacental region fell from 4.25 ± 0.27 in early stage pregnancy to 1.37 ± 0.45 in late stage pregnancy. There is a strong interaction between the number of desmosomes as a function of stage and location (F1,15 = 16.01, P < 0.002), thus with increasing embryonic stage the number of desmosomes decreases in the omphaloplacenta but not in the chorioallantoic placenta. In contrast in the nonreproductive uterus, the mean number of desmosomes along the lateral plasma membrane was 3 ± 0.7.

Desmosome Numbers in Bimodally Reproductive Species

There was no significant difference in the number of desmosomes in the uterine epithelium of bimodally reproductive L. bougainvillii and S. equalis between species (F1,15 = 0.17, P = 0.69; Fig. 3c) and parity mode (F1,15 = 0.0, P = 1). Thus, there is no interaction (F1,15 = 0.67, P = 0.43) between species or between parity modes.

Desmoglein-2 Immunofluorescence Results

Desmoglein-2 showed punctate staining along the whole lateral plasma membrane of uterine and glandular epithelial cells in P. spenceri (Fig. 4a). During early and mid stage pregnancy, desmoglein-2 was expressed along the apical region of the lateral plasma membranes of uterine epithelium in both species (Fig. 4b). As gestation progressed, desmoglein-2 disappeared from the lateral plasma membrane in the omphaloplacental region and was only expressed along the lateral plasma membrane in glandular epithelium (Fig. 4c,d). In the paraplacentomal region of the two Pseudemoia species, desmoglein-2 was almost completely lost from the lateral plasma membrane of attenuated uterine epithelial cells overlying capillaries (Fig. 4e). No desmoglein-2 was evident in the uterine epithelium of the chorioallantoic placenta of P. entrecasteauxii, but the fetal epithelium had an apical distribution of the protein during mid stage pregnancy (Fig. 4f). Nonreproductive P. spenceri and P. entrecasteauxii show desmogelin-2 staining in the uterine and glandular epithelium (Fig. 5a,b).

thumbnail image

Figure 4. A–D: Immunofluorescence micrographs of desmoglein-2 expression in the uterine epithelium of P. spenceri. Lumen; (L), uterine epithelium (arrows), glands; (*), capillaries (c). A: Low power micrograph of the chorioallantoic placenta in mid-stage pregnant uterus (Stage 35) showing apical desmoglein-2 expression along the lateral plasma membrane. Scale bar = 120 μm. B: High power micrograph depicting apical desmoglein-2 distribution along the lateral plasma membrane in mid stage (Stage 35) pregnant uterus. Scale bar = 20 μm. C: Late stage pregnancy (Stage 39) omphaloplacental region showing glandular desmoglein-2 expression, but not immunofluorescence in epithelial cells. Scale bar = 60 μm. D: High power micrograph showing glandular desmoglein-2 expression, but no immunofluorescence in epithelium. Scale bar = 30 μm. E: Micrograph of the chorioallantoic (placentome) region of P. entrecasteauxiiin late stage pregnancy (Stage 40). Attenuated epithelia overlying uterine capillaries showing faint amounts of desmoglein-2 expression. Scale bar = 30 μm. F: Micrograph of foetal-maternal interface in mid stage (stage 34) pregnancy in the chorioallantoic region of P. entrecasteauxii. No desmoglein-2 is present in the uterine epithelium (ue) (arrows) but a prominent desmoglein-2 expression is evident along the apical region of the lateral plasma membrane in fetal epithelium (fe) (arrow heads).

Download figure to PowerPoint

thumbnail image

Figure 5. A–E: Immunofluorescence micrographs of desmoglein-2 expression in the uterine epithelium of skinks. Uterine epithelium; (arrows), lumen; (L), glands; (*). A: Low power micrograph of nonpregnant P. spenceri depicting desmoglein-2 expression in uterine epithelial cells. Scale bar = 120 μm. B: High power micrograph showing desmoglein-2 expression along the whole lateral plasma membrane in the nonpregnant uterus of P. entrecasteauxii. Scale bar = 60 μm. C: Low power micrograph of nonpregnant S. equalisuterus. Desmoglein-2 expression in present along whole lateral plasma membrane between adjoining uterine epithelial cells.D: High power micrograph of pregnant S. equalisshowing glandular and uterine epithelial desmoglein-2 expression. Scale bar = 30 μm. E: Low power micrograph of L. bougainvilliiuterus with desmoglein-2 expression in glandular and uterine epithelium. Scale bar = 120 μm. F: Nonimmune controls using mouse matched IgG showed no immunofluorescent staining.

Download figure to PowerPoint

In nonreproductive viviparous S. equalis, both uterine and glandular epithelium stained for desmoglein-2 (Fig. 5c), and the staining pattern remained the same during gestation (Fig. 5d). A similar result was observed in oviparous S. equalis in the nonreproductive phase and during gravidity (not shown). Similarly, uterine and glandular epithelium showed desmoglein-2 expression in oviparous and viviparous populations of Lerista bougainvillii with no change observed between parity mode and reproductive status of the skink (Fig. 5e). Positive controls showed desmoglein-2 expression in epithelial cells. Negative and nonimmune controls showed no desmoglein-2 expression in any part of the uterus in any species at any stage of the reproductive cycle (Fig. 5f).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

This is the first ultrastructural study documenting changes in desmosomal numbers along the lateral plasma membrane of uterine epithelial cell of any species of amniote vertebrate with epitheliochorial placentation. Immunohistochemical data were also incorporated reflecting similar changes in desmoglein-2 protein expression, a major component of desmosomal plaques. The decrease in the number of desmosomes and expression of desmoglein-2 in uterine epithelium, but not in the glandular epithelium, of the highly placentotrophic skinks P. entrecasteauxii and P. spenceri suggests that the lateral adhesion between uterine epithelial cells decreases as gestation progresses, even though the average height of uterine epithelial cells of the omphaloplacenta in species from the Pseudemoia genus remains unchanged, 18 μm ± 2. Although, the degree of adhesiveness between the fetal and maternal membranes varies between the three main placental regions in Pseudemoia species, this adhesion never achieves any breaching or invasion of the uterine epithelium (Weekes,1935; Stewart and Thompson,1996;1998; Adams et al.,2005; Biazik et al.,2007). Therefore, the reasons for a decrease in desmosomal expression in skinks with noninvasive epitheliochorial placentae remain unclear. An increase in the number of desmosomes in uterine epithelium of a noninvasive placenta might be expected during pregnancy to provide greater cell–cell adhesion of the heavily stressed uterine epithelia of both oviparous and viviparous species as the conceptus increases in size (Dusek et al.,2007). Similarly, in other desmosome functional studies, an increase in the number of desmosomes had a profound effect on tumor suppression (Tselepis et al.,1998) as elevated desmosomal adhesion prevented tumor invasion into other tissue. However, as found in this study, a decrease in the number of desmosomes in the uterine epithelium of highly placentotrophic species may lead to a decrease in mechanical adhesion between cells, making the uterine epithelial cells more susceptible to deformation and remodeling. Previous desmosomal studies similarly found that a decrease in desmosomal numbers attributes to invasive behavior of malignant cells (Kocher et al.,1981). This apparent remodeling that the uterus undergoes in highly placentotrophic reptiles may facilitate the increasingly intimate contact that the trophoblast achieves in these species. Intimate contact has been previously reported in the Pseudemoia genus where increased convolution and interdigitation is achieved between fetal and maternal membranes (Weekes,1935; Adams et al.,2005), however no previous study has looked at the microanatomy of the plasma membrane, in particular the desmosomes to determine whether the integrity of the uterine membranes is affected with the evolution of viviparity.

Similar changes in desmosomal adhesion occur in the uterine epithelium at the time of implantation in mammals such as rats and mice, which have invasive hemochorial placentation (Illingworth et al.,2000; Preston et al.,2004). In rats and mice, the trophoblast adheres to the apical plasma membrane, displaces individual uterine epithelial cells and penetrates the basal lamina (Finn and Lawn,1968; Enders and Schlafke,1971; Schlafke and Enders,1975). During the invasion, the trophoblastic processes share in the uterine epithelium junctional complexes (Schlafke and Enders,1975) such that transient “giant” desmosomes form at the time of implantation in rats (Preston et al.,2004). The loss of lateral adhesion and cell-to-cell contact helps detach the uterine epithelium from the underlying basal lamina (Schlafke and Enders,1967). In epitheliochorial placentation, however, there is no invasion of the trophoblastic cells and the uterine epithelium remains intact. Nevertheless, restriction of desmosomes to the apical region and decrease of desmosomes along the lateral plasma membrane in uterine epithelium still occurs as gestation progresses in the noninvasive, epitheliochorial placentae of the two placentotrophic skinks studied.

Many structural and functional similarities exist between eutherian and squamate placentae, including the presence of an elaborate placentome found in both highly placentotrophic reptiles (Weekes,1935; Blackburn,1993; Corso et al.,2000; Blackburn and Vitt,2002; Adams et al.,2005) and mammals with epitheliochorial placentae (Friess et al.,1980; Steven et al.,1980; Lee et al.,1995; Abd-Elnaeim et al.,1999). The plasma membrane transformation, a series of morphological and biochemical changes in the plasma membrane of the uterine epithelium, and the formation of uterodomes, which both occur in most mammals in preparation for implantation (Murphy,1995,2000; Adams et al.,2002) have also been described in some species of viviparous skinks (Murphy et al.,2000; Hosie et al.,2003; Adams et al.,2005; Thomson et al.,2005), in which there is cellular adhesion between the trophoblast and uterine epithelium but no invasive implantation (Bjorkman,1973). Additionally, in the uterine epithelium of various skinks, a reported change in the distribution of tight junctional proteins occludin and claudin-5 was found which offered strong evidence that the paracellular pathway for exchange between the mother and the embryo becomes more restricted and prevents the passage of water and solutes as gestation progresses (Biazik et al.,2007,2008), similar to conditions found in rat uterus at the time of implantation (Orchard and Murphy,2002). A reduction in the number of desmosomes occurs during pregnancy in highly placentotrophic skinks and also during trophoblastic implantation in rats and mice (Illingworth et al.,2000; Preston et al.,2004), thus suggesting that these events are regulated by similar cellular mechanisms.

If cellular processes that contribute to trophoblast invasion are present in uterine epithelia of placentotrophic skinks, but trophoblast invasion does not occur, it is likely that the evolution of invasive placentae is not due to loss of desmosomes alone, but involves other factors such as characteristics of the trophoblast cells or pregnancy hormones (Edwards and Jones,2001; Dantzer and Paulesu,2002; Girling and Jones,2003). The similarities in desmosomal changes in the uterine epithelium of skinks and eutherian mammals with invasive placentae, suggests that other eutherian placental features such as genomic imprinting, a compromise for resources between mother and fetus, may be present in viviparous squamates. Genomic imprinting involves a nonequivalent contribution of parental genomes and has evolved at the same time as viviparity due to its dependence on the evolution of an efficient system of maternal-foetal exchange, the placenta (Amoroso et al.,1979; Ruvinsky,1999; Renfree et al.,2008). Recent studies on marsupials which have a relatively short lived placenta that provides little contribution to the growth of the offspring also shows imprinted gene expression (Renfree et al.,2008), thus making it valuable to investigate whether placentotrophic reptiles also demonstrate this phenomenon.

In contrast to lizards with complex placentae, no observable changes in desmosomal numbers were detected between oviparous and viviparous populations and at different stages of the reproductive cycle in the bimodally reproductive species, L. bougainvillii and S. equalis. Additionally, the uterine epithelium in the bimodally reproductive species exhibit desmosomes that were twice the size of desmosomes present in Pseudemoia species and the membranes between adjoining cells were more interdigitated, suggesting strong lateral adhesion between cells throughout the reproductive cycle. Viviparous L. bougainvillii and S. equalis both have a simple unspecialized chorioallantoic placenta (Weekes,1935), have no placentome and have no net uptake of dry matter across the placenta (Stewart and Thompson,1993; Thompson et al.,2001; Adams et al.,2007). Even though the uterine epithelium of the viviparous population of L. bougainvillii shows some signs of a plasma membrane transformation (Adams et al.,2007) no other features indicate that a highly specialized placenta exists in viviparous L. bougainvillii and S. equalis. As there are no differences in expression of desmosomes in oviparous and viviparous populations in these two species, and their desmosomes are larger, viviparous reptiles with simple placentae may exhibit more ancestral oviparous features.

This investigation offers strong evidence that the uterus of oviparous skinks and viviparous skinks with simple placentae is essentially an incubation chamber not associated with nutrient provision and does not undergo the same remodeling as observed in highly placentotrophic species. A reduction in the number of desmosomes and reduced desmoglein-2 expression in highly placentotrophic skinks from the genus Pseudemoia supports the hypothesis that the cellular mechanisms which mediate the invasive process in hemochorial placentation are also present in the noninvasive epitheliochorial skink placentae offering strong evidence that the adhesive properties of the lateral plasma membrane reflect common cellular mechanisms across amniote vertebrates. Future investigations into hormonal priming of the uterus, genomic imprinting and apical adhesion processes are necessary to better understand the sequence of events that may have led to the evolution of invasive placentae which are found in other highly placentotrophic skinks and most eutherian mammals.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Skinks were collected with permits from the NSW National Parks and Wildlife Service (S10693) and the work was conducted under The University of Sydney Animal Ethics Committee number L04/1-2005/3/4038. The authors thank all who have volunteered in the field and lab, especially Jacquie Herbert, Bridget Murphy, Jim Stewart, Trevor Wilson, and Dr Scott Parker who also assisted in the statistical analysis.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED