SEARCH

SEARCH BY CITATION

Keywords:

  • ovary;
  • vasculogenesis;
  • angiogenesis;
  • caveolin-1;
  • in situ hybridization

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

Expression screening for genes preferentially expressed in mouse fetal ovaries relative to testes identified Cav-1 as a candidate female-specific gene. Cav-1 encodes caveolin-1, a component of the cell membrane invaginations known as caveolae, which are involved in lipid regulation and signal transduction. In situ hybridization revealed high levels of Cav-1 mRNA in developing ovaries, compared with moderate or low levels in testes. Analysis of caveolin-1 protein distribution by immunofluorescence showed this difference to be due to the development of a dense and complex vascular network in the developing ovary. These observations point to a higher degree of differentiation and organization of the early stage mammalian ovary than previously suspected. © 2002 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

Since the discovery of SRY just over a decade ago, a great deal of attention has been focused on unraveling the molecular genetics and cell biology of testis development. Several key genes in the male sex-determining pathway have been identified (Swain and Lovell-Badge, 1999), and an understanding of the role of cell proliferation, differentiation, migration, and communication in the histogenesis of the fetal testis has begun to emerge (Capel, 2000; Koopman, 2001).

In contrast to these advances in understanding the development of the testis, ovarian development has remained mysterious (for review, see Loffler and Koopman, 2002). This situation reflects a failure to identify genes that direct developmental events in the ovary, a dearth of molecular markers with which to observe the behavior of individual cell types, and the absence of clear histologic organization until several days after testis differentiation begins in the mouse. A better understanding of the early development of the mammalian ovary is, therefore, an important goal in developmental and reproductive biology. Here, we show, by using caveolin-1 expression as a marker, a high degree of vascular development in the mouse fetal ovary from an early stage.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

We recently undertook an array-based expression screen with the aim of identifying genes preferentially expressed in the developing gonads of one sex or the other in the mouse fetus (Bowles et al., 2000; Bullejos et al., 2001). By using suppression PCR (Diatchenko et al., 1996), we enriched for cDNAs expressed sex-specifically in the developing testis and ovary, respectively (see Experimental Procedures section). A feature of this type of screen is its ability to identify not only genes that represent missing links in the pathway of male sex determination and testis development, but also genes that could provide a molecular entry point for analysis of early ovarian development. This screening has identified several candidate genes that appear to be up-regulated in fetal ovaries relative to testes. One such “female-enriched” clone, coded 13M37, was retrieved from our arrayed cDNA library and sequenced. BLAST homology searching revealed 100% sequence identity of 13M37 with a 277-bp 3′ untranslated region of GenBank sequence AB029929, which corresponds to the mouse Cav-1 gene.

Cav-1 encodes caveolin-1, a component of the cholesterol- and glycosphingolipid-rich cell membrane domains known as caveolae (Anderson, 1998; Kurzchalia and Parton, 1999). The function of caveolae remains controversial, but they are implicated in lipid regulation (Razani et al., 2002), endocytosis/transcytosis, and signal transduction (Okamoto et al., 1998; Drab et al., 2001; Matveev et al., 2001; Schlegel and Lisanti, 2001). Caveolae are particularly abundant in adipocytes and endothelial cells (Scherer et al., 1995). Caveolin-1 is known to cause invagination of caveolae from the plasma membrane (Fra et al., 1995), but no detailed expression profile has been reported for the Cav-1 gene.

To investigate a possible role for caveolin-1 in gonadal development, we first examined the distribution of Cav-1 mRNA by in situ hybridization of whole mouse embryos and explanted mouse fetal testes and ovaries at a range of developmental time points. At 9.5 days post coitum (dpc), Cav-1 was expressed strongly in the developing heart and weakly in the developing vasculature of the mouse embryo (Fig. 1A). In addition, expression in urogenital ridges was observed. In explanted urogenital ridges, genital ridge-specific expression of Cav-1 was observed in both males and females at 9.5 and 11.0 dpc, before sexual differentiation has occurred (Fig. 1B,C). At 12.25 dpc, shortly before the onset of testis differentiation, expression of Cav-1 was up-regulated in the female genital ridges relative to male (Fig. 1D). In the period from 12.5 to 16.5 dpc, during differentiation of the testis cords but before overt differentiation of the ovary, expression of Cav-1 became strongly up-regulated in the ovary, but remained at low to moderate levels in the testis (Fig. 1E–H).

thumbnail image

Figure 1. Cav-1 mRNA expression in developing mouse gonads. Whole-mount in situ hybridization using a probe specific for Cav-1. A: Whole mouse embryo at 9.5 dpc, showing Cav-1 expression in developing heart (ht), weak expression in vasculature such as the intersomitic vessels (isv), and prominent expression in the urogenital ridge (ugr). B–H: Male (M) and female (F) urogenital ridges at 9.5–16.5 dpc, showing female-specific up-regulation of Cav-1. I, comparison of Cav-1 (left) and Oct4 (germ cell marker, right) expression in germ cell-free We mutant ovaries at 13.5 dpc. Cav-1 staining is clearly seen in the absence of germ cells, consistent with somatic expression.

Download figure to PowerPoint

We tested whether expression of Cav-1 mRNA in the developing ovary was due to somatic or germ cells, by examining expression in ovaries from We mutant embryos, which lack germ cells. In We ovaries, expression of Cav-1 was retained, confirming somatic expression but not excluding a possible contribution from germ cells (Fig. 1I).

To confirm that caveolin-1 protein is expressed in developing gonads, and to investigate the cell type responsible for caveolin-1 expression, we used whole-mount immunofluorescence on urogenital ridges at 13.5 dpc (Fig. 2). In the male, caveolin-1 expression marked a vascular plexus, including blood vessels in the mesonephros, vessels surrounding the testis cords, and the prominent coelomic vessel at the periphery of the testis (Fig. 2A,D). Caveolin-1 expression in endothelial cells was confirmed by colocalization with the endothelial cell marker platelet endothelial cell adhesion molecule-1 (PECAM-1; Fig. 2A–F). These observations are consistent with reports of a high density of caveolae on endothelial cells (Scherer et al., 1995; Glienke et al., 2000), of caveolin-1 protein expression in vasculature of the fetal lung (Ramirez et al., 2002), and of more widespread Cav-1 mRNA expression in the developing cardiovascular system in mice (Fig. 1A and Bullejos et al., unpublished data). They concur also with previous observations relating to vascularization of the developing testis (Brennan et al., 1998; Schmahl et al., 2000).

thumbnail image

Figure 2. Caveolin-1 protein expression in developing mouse gonads. Two-colour immunofluorescence of 13.5 dpc testes (A–F) and ovaries (G–L), by using antibodies to caveolin-1 (A,D,G,J) and the endothelial/germ cell marker PECAM-1 (B,E,H,K), with merged images also shown (C,F,I,L). V, blood vessels; G, germ cells. Scale bar = 500 μm in A–C,G–I; 250 μm in D–F,J–L.

Download figure to PowerPoint

Surprisingly, caveolin-1 marked a dense network of blood vessels in the developing ovary at 13.5 dpc (Fig. 2G–L). These vessels were in close apposition to the strings of germ cells also marked by PECAM-1 expression and known as ovigerous cords (Odor and Blandau, 1969; Konishi et al., 1986) or germline cysts (Pepling and Spradling, 1998). The PECAM-1–positive germ cells did not appear to express caveolin-1 (Fig. 2G–L). The presence of a complex vascular network in the ovary at this early stage of development has not been reported previously and indicates a higher level of histologic organization in the early ovary than is commonly assumed.

In summary, our data reveal a high degree of angiogenic activity in the developing ovary from an early stage. This observation is surprising from several points of view. First, because testes grow much faster than ovaries in the period 12.5 to 14.5 dpc and show a high level of histologic organization in this period, it is often assumed that the testis will be more vascularized than the ovary as a result. Indeed, the prominent, blood-filled vessel on the coelomic surface of the testis distinguishes it from the ovary as early as 12.5 dpc. The complex ovarian vasculature described in the present study consists of smaller vessels that are not obvious under the dissecting microscope. Second, in contrast to the testis, the fetal ovary has no known hormonal output and, hence, does not require an extensive vasculature for the secretion of hormones into the bloodstream. Our observations raise the possibility that an extensive vascular network may be required in the ovary from an early stage for the delivery of exogenous growth factors to ovarian somatic or germ cells. Further work is required to test this hypothesis.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

Array-Based Differential Gene Expression Screen

Outbred Swiss Quackenbush mouse gonadal tissue was collected at 12.5 to 13.5 dpc, where 0.5 dpc is defined as noon on the day of plug. Fetal testis and ovary RNA was extracted and suppression-subtractive hybridization (PCR-Select cDNA Subtraction Kit, Clontech) used to make male vs. female (“male-enriched”) and female vs. male (“female-enriched”) cDNA pools (Bowles et al., 2000). The female-enriched pool was cloned, and a portion of the resulting library was arrayed on nylon membranes in duplicate. Membranes were hybridized with radioactively labelled male-enriched and female-enriched cDNA pools to allow selection of clones most likely to be genuinely female-specifically expressed in gonads at this time point.

Staging, Dissection, and Genotyping of Fetal Mouse Gonads

Timed matings were produced by using outbred Swiss Quackenbush mice. Fetuses were dissected between 9.5 and 16.5 dpc; accurate staging between 10.5 and 12.5 dpc was achieved by counting tail somites (Hacker et al., 1995; Bullejos and Koopman, 2001). Urogenital ridge tissue was also dissected from homozygous We (extreme dominant spotting) mutant mice (Lyon and Searle, 1989). These mice lack the cell surface tyrosine kinase receptor molecule c-kit, and as a result, most primordial germ cells die by apoptosis during the migratory phase.

Whole-Mount In Situ Hybridization

Whole-mount in situ hybridization was carried out as described (Hargrave and Koopman, 2000; Bullejos and Koopman, 2001). Two Cav-1 probes were used simultaneously: one was transcribed from clone 13M37, which corresponds to nucleotides 1945-2222 in the 3′ UTR of mouse Cav-1 (accession no. AB029929), the other was generated from clone VIP21 (Kurzchalia et al., 1992), which represents the canine Cav-1 cDNA (accession no. Z12161). The 644-bp canine probe was 91% homologous to the corresponding mouse sequence and was used to augment the signal generated by the short (277-bp) mouse probe. A 462-bp probe for the germ cell marker Oct4 (Schöler et al., 1990) was transcribed from a plasmid supplied by P. Rathjen, University of Adelaide, Australia.

Whole-Mount Immunofluorescence and Confocal Microscopy

Gonads were fixed in 4% paraformaldehyde in PBS overnight at 4°C. Double labelling was performed as described elsewhere (Hogan et al., 1994). Caveolin-1 was detected by using a 1:200 dilution of a rabbit polyclonal antibody (Transduction Laboratories, #C13630). Germ cells and vasculature were identified by using a 1:100 dilution of a rat monoclonal antibody (Pharmingen, #557355) against mouse PECAM. Images of labelled gonads were collected with 10× and 20× objectives by using a Bio-Rad Radiance 2000 confocal microscope system.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

We thank Rob Parton, Patrick Tam, and Kelly Loffler for critical comments on the manuscript and Albert Pol for experimental assistance. The caveolin-1 antibody and canine probe were gifts from Rob Parton; the Oct4 probe was a gift from Peter Rathjen. M.B. was supported by the Secretaría de Estado de Universidades, Investigación y Desarrollo, Spain. P.K. is an Australian Research Council Professorial Research Fellow.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES