|Home | About | Journals | Submit | Contact Us | Français|
Signaling by Bone morphogenetic proteins (Bmps) has multiple and diverse roles in patterning and morphogenesis of the kidney, eye, limbs and the neural tube. Here, we employed the Bmp7lacZ strain to perform a detailed analysis of Bmp7 expression and the null phenotype during development of the mouse urogenital system. The urethral compartment originates in mid-embryogenesis from the ventral part of the cloaca, a transient cavity at the caudal end of the hindgut. At mid-gestation, Bmp7 expression was detected within several specific domains in the cloacal epithelium and mesenchyme. In late embryogenesis, Bmp7 expression was present in the urethra, rectum, the urethral glands, corpus cavernosum, and in the male and female genital ducts. Importantly, loss of Bmp7 resulted in arrest in cloacal septation, and severe defects in morphogenesis of the genital urethra and mesenchyme. Together, our analysis of Bmp7 expression and the null phenotype, indicates that Bmp7 may play an important role in re-organization of the epithelium during cloacal septation and morphogenesis of the genital tubercle.
The mammalian urogenital system, the bladder, urethra and external genitalia, develop from the epithelium and mesenchyme of the cloaca, a transient embryonic cavity at the caudal end of the hindgut (Fig. 1A). Urogenital malformations occur at high frequency in humans, and often combine malformations of the genital tubercle (GT), the penis in males and clitoris in females, and defects in separation of the urethral and rectal compartments (Hendren, 1992;1996;1998;Kurzock et al., 1999; Baskin et al., 2001; Mo et al., 2001). Morphogenetic mechanisms that direct development of the urogenital system are not well understood. The caudal homeobox genes, Hoxa13 and Hoxd13, are expressed in the cloacal epithelium and mesenchyme, and are essential for morphogenesis of all cloacal derivatives (Warot et al., 1997; Morgan et al., 2003). Expression of p63 (Cheng et al., 2006) and Sonic hedgehog (Shh) (Haraguchi et al., 2001; 2007; Mo et al., 2001; Seifert et al., 2008) in the cloacal and urethral epithelium is also important both for septation of the cloaca, and development of the bladder and external genitalia. In addition, Fibroblast Growth Factors 8 and 10, Bmp4 and 7, and Wnt5a have been implicated in development of the GT (Yamada et al., 2006), but their role in morphogenesis of the cloaca has not been explored. In this study, we employed the Bmp7lacZ reporter strain (Godin et al., 1998) to perform a detailed analysis of Bmp7 expression and the null phenotype during development of the urogenital system in males and females.
Comparative analysis of Bmp7 expression by in situ hybridization with an antisense Bmp7 RNA probe (Lyons et al., 1995), and by X-gal staining of tissues in Bmp7lacZ/+ background has shown that LacZ activity accurately reflects the expression of the wild type Bmp7 allele in many embryonic tissues, including the kidney (Godin et al., 1998; Dudley et al., 1999) and the urethra (Fig. 4C, D). In addition, our analysis of Bmp7lacZ/+ GT (Fig.1–3) showed LacZ activity in the urethral plate and genital mesenchyme in complete agreement with previous reports on Bmp7 expression (Morgan et.al, 2003; Suzuki et al., 2003). Thus, we conclude that LacZ activity in Bmp7lacZ strain is an accurate readout of the activity Bmp7 promoter in the lower urogenital system and the hindgut.
The cloacal cavity is defined as the caudal end of the hindgut when, at embryonic day 9.5 (E9.5) in mice, it comes in contact with the ectoderm just anterior to the tail at the site of the future anal opening (Perriton et al., 2002; Hynes and Fraher, 2004a,b,c; Sasaki et al., 2004). Starting at E10, the ventral end of the cloaca grows anteriorly giving rise to the embryonic urogenital sinus (UGS), the primordium of the pelvic urethra and the bladder (Fig. 1). Studies in mouse models indicate that septation of the cloaca is driven by the caudal growth of the lateral mesenchyme known as the urorectal septum at E10.5 – E13.5, and is completed with disintegration of the cloacal membrane to uncover the anal and the caudal urethral orifices at E13.5 – E14.0 (Kluth et al., 1995; Hynes and Fraher, 2004a,b,c; Sasaki et al., 2004; Seifert et al., 2008). Interestingly, analysis of Bmp7lacZ/+ heterozygous embryos showed that Bmp7 is expressed specifically in the urorectal septum (URS) at E11.5 (Fig. 1C) and E12.5 (Fig. 2F), and expression is dramatically reduced at E13.5 (Fig. 3C, D). We further asked whether Bmp7 function may be required for cloacal septation. From E11.5 to E13.5, heterozygous embryos showed normal descent of the URS (Fig. 1C; ;2E;2E; ;3C;3C; Hynes and Fraher, 2004a-c; Sasaki et al., 2004), and progressive separation of the urethral and hindgut compartments from anterior to caudal (Fig. 2G–I; 3E–G; Hynes and Fraher, 2004a-c; Sasaki et al., 2004). Bmp7 null embryos showed normal invagination of the URS into dorsal cloaca at E11.5 (Fig. 1D). However, further septation was delayed at E12.5 (Fig. 2K, L compare to G, H) and E13.5 (Fig. 3I, J compare to E, F). At E12.5 and E13.5, transverse sections of Bmp7 null GT showed persistent cloaca, whereas corresponding sections of heterozygous animals showed separate urethral and hindgut compartments (Fig. 2, ,33).
The initial stages of development of the GT at E11 – E15 are gender-independent, and begin with the appearance of genital swellings laterally to the cloaca at E10.5 (Perriton et al., 2002). At E11.5, Bmp7 expression was detected in the mesenchyme of the genital swellings, and the ectoderm of the ventral body wall and the tail (Fig. 1C). Distal outgrowth of the genital swellings is accompanied by the extension of the urethral epithelium and lumen (Kuzrock et al., 1999; Yamada et al., 2003; 2006; Seifert et al., 2008). At E12, the distal urethral walls adhere to form a solid urethral plate (Fig. 2H, I; Kuzrock et al., 1999; Yamada et al., 2003; 2006; Seifert et al., 2008). Analysis at E12.5 showed Bmp7 expression in the urethral plate (Fig. 2G, H), the adjacent mesenchyme (Fig. 2H), and the ventral ectoderm (Fig. 2E). In the null, LacZ activity marked the urethral plate, and was present in a broad domain in the GT mesenchyme (Fig. 2K, L). Distal extension of the urethral plate appeared 50% shorter in the null (Fig. 2D, H, L), most likely due to unzipping of the urethral folds at the ventral side of the GT (Fig. 2C; and K, L, inserts) and formation of a large urethral groove (Fig. 2C).
At E13.5, Bmp7 expression appeared in the dorsal GT mesenchyme (Fig. 3C, D) which may contribute to the corpus cavernosum (Yamada et. al., 2003). Bmp7 expression was maintained in the ventral mesenchyme, but decreased in the URS (Fig. 3C, D). In the genital urethra, Bmp7 was expressed in a gradient with the strongest expression in the distal urethral epithelium (DUE) at the tip of the GT, which also expresses Fgf8 (Haraguchi et al., 2000), and the weakest at the urorectal septum (Fig. 3F, G). In the null, LacZ activity was lost in the medial and proximal parts of the urethral plate, but maintained in the DUE (Fig. 3K, L). In the ventral GT mesenchyme, Bmp7 expression was restricted to the urethral plate in heterozygous (Fig. 2H; 3G, H). In contrast, in the null, LacZ-positive mesenchyme spread laterally (Fig. 2K, L; 3K, L).
Starting at E15.5 in mice, the external genitalia undergo sex-specific differentiation largely controlled by androgens in the male (Drews et al., 2002; Kim et al., 2002; Yucel et al., 2003, 2004; Buckley et al, 2006). In E17.5 male embryos, Bmp7 expression was detected in the urethra (Fig. 4D, 5B,D, H), the bulbourethral and preputial glands (Fig. 4D, ,5H),5H), the vas deferens (Fig. 4D, insert) and the corpus cavernosum (Fig. 4D, 5E, F, H). LacZ activity was not detected in the bladder, nor in the testes (Fig. 4D, and insert). To precisely localize LacZ activity to the epithelial and mesenchymal compartments of the GT, we embedded X-gal stained tissues in paraffin and carried out analysis of histological sections (Fig. 5). Some of the sections were co-immunostained for cytokeratin 14 (CK14) which marked the squamous epithelium in the urethra and the rectum (Fig. 5). Bmp7 expression was detected in the dorsal and ventral epithelium of the urethra (Fig. 5B, H), in the preputial glands (Fig. 5D, H), and in the crypts of the intestine (Fig. 5H). Bmp7 was also present in the mesenchyme of the urethra (Fig. 5D, K), the rectum (Fig. 5K), and in the corpus cavernosum (Fig. 4D; 5E, F, H). Normal morphogenesis of the penile urethra involves closure of the original caudal orifice (Perriton et al., 2002; Yamada et al., 2003), fusion of the ventral urethral groove (Fig. 5C; D; Yamada et al., 2003), and displacement of the ventral urethral seam by the genital mesenchyme (Fig. 5A;Kuzrock et al., 1999; Yamada et al., 2003; Seifert et al., 2008). In heterozygous GT, Bmp7 was expressed in the mesenchyme surrounding the ventral epithelial fusion (Fig. 5D). LacZ-positive mesenchyme also localized to the abnormal genital urethra in Bmp7 null (Fig. 5P-T, bellow). Further analysis showed that Bmp7 null males developed complex rectourethral malformations, including rectourethral fistula and severe hypospadia (Fig. 4E, F; 5P-T; ;8B).8B). In the null, the hindgut lacked a separate anal opening (Fig. 4E, F; ;8B),8B), and a narrowed rectum opened directly into the urethra (Fig. 4E, F; 5P-S), consistent with the arrest in cloacal septation in mid-embryogenesis (Fig. 2, ,3).3). Strong LacZ activity was detected in the mesenchyme of the fistula and the cloaca (Fig. 4E, F; 5O, R, S). LacZ activity was also detected in the basal epithelium of the urethra and hindgut (Fig. 5 N, O). In Bmp7 null males, the caudal urethral opening was maintained (Fig. 4E; ;8B),8B), and the tubular penile urethra did not form (Fig. 5T compare to 5A, C). Instead, the urethral epithelium lined the ventral surface of the GT adjacent to the cloacal opening (Fig. 5T). Strong LacZ-activity was present in the mesenchyme adjacent to the urethral epithelium (Fig. 4E, F; 5R, S). Other urogenital abnormalities in Bmp7 null males included loss of bulbourethral glands (Fig. 4E compare to D) and enlarged os penis (Fig. 5M, and S compare to I, K).
In the female urogenital system (Fig. 6–8), the pattern of Bmp7 expression and the Bmp7 null phenotype were very similar to what was observed in the male (Fig. 4, ,55 and and8).8). In females, Bmp7 was expressed in the urethra (Fig. 6C; 7B, E), the rectum (Fig. 6C; ;7E),7E), the intestinal crypts (Fig. 7B), the preputial glands (Fig. 6C; ;7B),7B), and the corpus cavernosum (Fig. 7E). In addition, LacZ activity was detected in all Mullerian duct derivatives, the oviducts, uterus and developing vagina (Fig. 6C–E). LacZ activity was not present in the ovaries, nor in the bladder (Fig. 6C, D). Analysis of Bmp7 null females showed rectourethral fistula (Fig. 7I, J), enlarged caudal urethral/cloacal opening (Fig. 7L, ,8C),8C), and abnormalities in the topology of the genital epithelium (Fig. 7L) and mesenchyme very similar to those observed in the null males (Fig. 4, ,5,5, ,8B).8B). Strong LacZ activity was detected in the mesenchyme of the fistula and the ventral mesenchyme of the GT (Fig. 6E; 7G–L). Bmp7 null females also showed enlarged os clitoris (Fig. 7M, and I, L compare to C, F), and excessive branching of the genital vein (Fig. 7I and C) similar to what was reported in the Hoxa13−/− GT (Morgan et al., 2003).
Bmp signaling mediated by Smad1/5/8 transcriptional mediators has multiple and diverse roles in organogenesis, including regulation of programmed cell death (Morgan et al., 2003), cell survival (Dudley et al., 1999), neural and limb patterning (Chesnutt et al., 2004; Pizette et al., 2001) and cell adhesion (Linecum et al., 1998). Although many details in morphogenesis of the anus and external genitalia are still unclear, our analysis of Bmp7 expression and the null phenotype indicates that Bmp7 signaling may play an important role in reorganization of the urethral epithelium and mesenchyme during cloacal septation and formation of the penile urethra. Bmp7 targets in the GT include the homeobox genes Msx1,2 (Morgan et al., 2003) which can regulate cadherin-dependent cell adhesion (Linecum et al., 1998). Thus, further studies are needed to determine whether Bmp7 may regulate cell adhesion in the urethral epithelium, and contribute to the migration and differentiation of the genital mesenchyme.
Animal work was conducted under NYUSM IACUC approved animal protocol. Bmp7lacZ strain has been described previously (Godin et al., 1998; Dudley et al., 1999). Genotyping, whole mount X-gal staining, and preparation of histological sections have been performed as described previously (Grishina et al., 2005). Some of the sections were counterstained with hematoxylin and eosin. At each stage from E11.5 to E13.5, we analyzed six to ten animals of the same genotype. At E17.5 and P0, we analyzed six to ten animals of the same sex and genotype. Measurements of the genital structures was performed on transverse sections of the null and heterozygous embryos, and Student’s t-test was used to determine the significance of differences. For immunofluorescent analysis, paraffin sections were incubated with antibodies for cytokeratin 14 (CK14, Covance, PRB-155P) at 1:100 dilution over night at 4°C, followed by 546 Alexa Fluor secondary antibodies (Invitrogen) diluted at 1:1000 for 1 hour at room temperature. In situ hybridization analysis was performed in whole mount as previously described (Grishina et al., 2005) using 0.9 kb digoxigenin-labeled antisense RNA probe for Bmp7 (Lyons et al., 1995).
We thank Lee Niswander, Tung-Tien Sun, Cathy Mendelson, Gen Yamada, Ashley Seifert, Steve Munger, Xue-Ru Wu and Susan Logan for critical reading of the manuscript and/or useful discussions, and Elizabeth Robertson for the gift of Bmp7lacZ strain and Bmp7 cDNA. This work was supported by the Department of Urology, NYU Langone Medical Center; National Institute of Diabetes, Kidney and Digestive Diseases grant DK-068007 to IG; and the Kidney and Urology Foundation of America Award to IG.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.