|Home | About | Journals | Submit | Contact Us | Français|
We genetically disrupted the Wolffian duct (WD) in mice to study the affected organogenesis processes and to test the hypothesis that cell loss can be the developmental basis for a wide spectrum of congenital anomalies in the kidney and urinary tract.
We use Hoxb7-Cre transgenic lines (HC1 and HC2) to induce diphtheria toxin (DT) production from a ROSADTA allele, disrupting the wolffian duct and derivatives.
The first set of mutants (HC1;ROSADTA/+) exhibited agenesis of the kidney, ureter, and reproductive tracts. The second set of mutants (HC2;ROSADTA/+) exhibited diverse defects, including renal agenesis/hypoplasia, hydronephrosis, hydroureter, ureter-vas deferens fistulas in males and ureter-oviduct/uterus fistulas in females. The phenotypic differences correspond to the degree of apoptosisinduced caudal truncation of the wolffian duct, which is less severe and more variable in HC2;ROSADTA/+ mice. Whenever the wolffian duct failed to reach the urogenital sinus, the ureter failed to separate from the wolffian duct, suggesting that ureteral migration along the common nephric duct to the cloaca and the subsequent common nephric duct degeneration constitute the only pathway for separating the ureter and WD derivatives.
The diverse and severe defects observed emphasize the central role of the wolffian duct in providing progenitors and signals for urogenital development. These results also indicate that the quantitative difference in cell deathinduced caudal truncation of the wolffian duct can lead to a wide range of qualitatively distinct defects, and that cell death can serve as a single etiological cause of a wide spectrum of congenital kidney and urinary tract defects.
Congenital anomalies of the kidney and urinary tract represent a significant cause of morbidity and mortality in the pediatric population.1 These anomalies occur in many forms, including renal agenesis, renal hypoplasia, horseshoe kidney, hydronephrosis, hydroureter, vesicoureteral reflux (VUR), and others. Interestingly, multiple defects can be found within an affected family or even within an affected individual.1 Despite recent advancements, the genetic or environmental determinants in most cases are still difficult to ascertain.1 Excessive alcohol consumption during pregnancy can cause birth defects (fetal alcohol syndrome) that include congenital urinary tract defects.2 At least in the nervous system most abnormalities are attributed to cell death due to ethanol toxicity.3, 4 How cell death may lead to urinary tract defects has not been directly studied in fetal alcohol syndrome or other conditions involving congenital urinary tract defects.
The Wolffian duct, also called mesonephric or nephric duct, emerges from the intermediate mesoderm to connect the mesonephros to the cloaca. During mid gestation, the ureteric bud (UB) branches out from the WD (at about 28 days in human and E10.5 in mice) and invades the metanephric mesenchyme within the intermediate mesoderm. The UB branches inside the MM to form the collecting system and induce nephron formation.5–7 The UB stalk develops into the ureter. The WD and (MD) are present in both genders until sexual differentiation, when MD develops into the female reproductive tract consisting of the oviduct, uterus and upper third of the vagina, and the WD differentiates into the male genital tract, including the vas deferens, epididymis, and rete testis.
Cell loss is common in diseased tissues as a consequence of either necrosis or apoptosis.8 Cell loss of sufficient magnitude impairs organ function, resulting in clinically overt diseases.8 Surgical, chemical, and genetic cell ablation have been used to study the roles of specific cell populations and to generate disease models.9 Among these techniques genetically controlled cell ablation offers the highest precision and consistency. The A chain of diphtheria toxin (DTA) causes apoptosis through its inhibition of the eukaryotic elongation factor 2.10, 11 Since rodents have no DTA receptors, transgene controlled production of DTA induces apoptosis in target cells without affecting their neighbors.11 In this study, we use mice with WD-specific expression of Cre to induce DTA production and apoptosis in WD cells. In addition to revealing a central role of the WD in urogenital development, our results also showed that a single factor—cell death—can lead to a wide spectrum of congenital anomalies of the kidney and urinary tract.
All animal studies have been approved by the Institutional Animal Care and Use Committee at Washington University and conducted according to NIH guidelines. The Hoxb7-Cre-EGFP (HC1), Hoxb7-Cre (HC2), and the ROSADTA allele were described previously.11–14 Mice carrying HC1 (or HC2) were crossed with mice carrying ROSADTA to produce embryos/mice (designated as mutants) producing DTA in the WD and derivatives. Their littermates with no DTA production were controls. The ROSALacZ transgene15 was used to fate map Cre-expressing cells. ROSADTA encodes an attenuated DTA, DTA176, which prevents unintended cell death due to potential minor leakage of the promoters used.11
For histological analyses embryos were fixed with 4% paraformaldehyde and embedded in paraffin. Sections of 5 μm were collected and stained following standard protocol. For immunohistochemistry, sections were stained with an anti-BrdU antibody (Developmental Studies Hybridoma Bank, 1: 200 ratio) and an anti-Pax2 antibody (Covance, Princeton, New Jersey, 1:200). Appropriate AlexaFlour488 or 555-conjugated secondary antibodies (Molecular Probe, Eugene, Oregon, 1:1,000 ratio) were used to detect the corresponding primary antibodies. Whole-mount immunostaining was carried out with the anti-Pax2 antibody and an anti-E-Cadherin antibody (BD Transduction Laboratories™, 1:200 ratio) as described.16
β-galactosidase assay was performed as previously described.17 Briefly, embryos or tissues were collected from timed pregnant females and fixed in 0.2% glutaraldehyde for 15–30 minutes. After permeabilization (10% TritonX™ 100 in phosphate buffer) embryos and tissues were stained in X-gal solution (50 mM potassium ferricyanide, 50 mM potassium ferrocyanide, 1 mg/ml X-gal) and post-fixed with 4% paraformaldehyde.
BrdU was injected intraperitoneally into pregnant mice (100 μg/g body weight) 30 minutes before harvest and was detected by a mouse monoclonal anti-BrdU antibody (Developmental Studies Hybridoma Bank, 1: 200 ratio). TUNEL analysis was performed on paraffin-embedded sections using the ApopTag® plus peroxidase in situ apoptosis detection kit. Proliferation index is presented as the number of BrdU-positive cells per 100 cells counted, and apoptosis index as number of TUNEL-positive cells per 100 cells counted.
TUNEL and BrdU immunohistochemistry results were statistically analyzed using student's t-test. A p value of 0.05 or less was considered significant. At least 3 sets of samples were analyzed for each data point.
To study the impact of cell loss during nephrogenesis, we generated mice carrying a Hoxb7-Cre transgene (HC1) and a ROSADTA allele with a loxP-flanked transcriptional stop cassette placed between the ubiquitous ROSA promoter and the DTA coding sequence. This stop cassette prevents DTA transcription and subsequent cell death. When Cre is present, recombination between the loxP sites removes the stop cassette, leading to DTA transcription, translational arrest, and subsequent apoptosis (Fig. 1A). The HC1 transgene produces Cre specifically in the WD, as demonstrated by β-galactosidase assay on samples from mice carrying HC1 and ROSALacZ (Fig. 1B–E).15 Thus, mice carrying both HC1 and ROSADTA (mutants) would have DTA expression and cell death in the WD and its derivatives, including the ureteric epithelium. However, their littermates carrying 1 or none of these alleles (controls) would not have any transgene-induced cell loss.
Newborn mice carrying both HC1 and ROSADTA had kidney and ureter agenesis (Fig. 2A–D). The vas deferens and the seminal vesicles were also missing in the male mutants (Fig. 2B). Since the male reproductive tract derives from the WD, the severe developmental disruption of these structures is not surprising. However, the testiswas not affected. While the ovary develops normally in female mutants, the uterus is missing (Fig. 2D), indicating the complete dependency of the development of the uterus (a mullerian duct derivative not targeted by DTA) on the WD.
Since almost no 2 transgenes behave exactly the same way, to further examine if potential minor differences in the level of WD disruption would lead to different outcomes, we generated a second system in which a different Hoxb7-Cre transgene (HC2)12 was used to induce DTA production and cell death in the WD. Interestingly, HC2;ROSADTA/+ mutants exhibitd wide range of defects (Fig. 2E–L). Besides agenesis, HC2;ROSADTA/+ mutants also have renal hypoplasia, hydronephrosis and hydroureter. Most of these defects are characteristic for congenital anomalies of the kidney and urinary tract in humans.
In HC2;ROSADTA/+ mutants the kidney was connected to the testis instead of the bladder, in 50% of the malesand to the ovary in 39% of the females (Fig. 3). These aberrant connections were apparently the results of fistulae formed between the ureter and the vas deferens/epididymis (Fig. 3B, J and N) or uterus/oviduct (Fig. 3D, L and P). Fistulas start to form in mid gestation, as revealed by whole-mount immunostaining for Pax2 in the urogenital ridges of the developing embryos at E12.5 (Fig. 3E–H). DTA induced cells death in the WD affected its extension toward the cloaca, leading to caudal truncation of the WD. In most HC2;ROSADTA/+ mice, the caudal truncation of the WD prevent its contact with the urogenital sinus and the formation of the CND (Fig. 3).
In some mutants with disrupted CND formation, the ureteral bud could still reach the MM to initiate nephrogenesis. In such male mutants, the UB-derived ureter was never separated from the WD. As the WD differentiated into the male reproductive tract below the testis, the UB and the male reproductive tract were connected via a fistula. This connection disrupts the urinary path to the bladder and inevitably results in severe urinary tract obstruction and lethality when presented bilaterally. In females with disrupted CND formation fistulae between the ureter and the female reproductive tract occurred. These malformations differ from those in the males because the female reproductive tract derives from the MD but not the WD. MD emerges at E12.5 near the anterior end of the urogenital ridge in mice. It extends caudally alongside the WD and reaches the cloaca area at about E13.5.
When CND is missing due to the caudal truncation of the WD, instead of elongating towards the urogenital sinus (Fig. 3E), the MD turns to follow the ureters (Fig.3 F–H), and eventually formed the fistulae between the ureters and oviducts/uterus at later stages in mutant females (Fig. 3L and P). Such fistulas in both genders disrupt the urinary path to the bladder, leading to severe urinary tract obstruction. However, urinary tract obstruction also occurred in HC2;ROSADTA/+ mice without fistulae (Fig. 2 and data not shown). Such obstructions are not directly caused by WD caudal truncation. Rather, they appear to result from the apoptosis induced structural and functional alterations in the ureter that eventually affect pyeloureteral peristalsis.
To confirm that the variation in mutant phenotypes is associated with the differences in levels of cell death, we have examined proliferation and apoptosis on sections from control and mutant mice. The proliferation rate in the WD was significantly reduced in mutants (Fig. 4). More importantly, significantly more WD cells were undergoing apoptosis in the HC1;ROSADTA/+ mutants. Excessive cell death was also detected in the MM of these mutants. The HC2;ROSADTA/+ mutants also demonstrated significantly reduced proliferation and significantly increased apoptosis in the WD (Fig. 5). Reduced proliferation likely reflects the fact that DTA forced cells that would otherwise proliferate into the apoptosis pathway. Compared to the HC1;ROSADTA/+ mutants, alterations in the HC2;ROSADTA/+ mutants were less dramatic and more variable. The higher and more uniform level of apoptosis found in the HC1;ROSADTA/+ mice is consistent with the more severe and consistent defects (bilateral kidney and ureter agenesis) in these mutants. The lower and more variable levels of apoptosis in the HC2;ROSADTA/+ mutants appear to be the underlying reason for the milder and more variable defects found in the HC2;ROSADTA/+ mutants. Similar differences were detected at as early as E9.5 (data not shown). Variations in cell death of the WD at earlier stages lead to various degree of caudal truncation of the WD, which, in turn, causes a wide range of defects.
In this report, we use genetically controlled tissue-specific diphtheria toxin expression to disrupt the WD during embryogenesis to assess directly the impact of cell death on urogenital development. We found that various levels of cell death in the WD result in congenital urogenital defects ranging from agenesis to milder malformation of multiple urogenital organs. These results provide support for the hypothesis that a single factor, cell death, can be the cause of a wide spectrum of congenital anomalies of the kidney and urinary tract.
The severe agenesis and malformation of multiple urogenital organs in the mutants emphasize an essential role of the WD in urogenital development. Phenotypic differences observed in the 2 lines may result from subtle variations, beyond current experimental detection limits, in the timing and completeness of Cre expression. Regardless of the origins, such variations led to differences in the level apoptosis in targeted cell populations (Fig. 4–5) and subsequently in phenotypic outcomes.
In mutants from both lines variations in apoptosis in the WD would lead to the cumulative effect of shortening of the WD at its caudal end to different degrees (Fig. 3F). The WDs in most HC1;ROSADTA/+ mutants fail to extend to the UB budding site, resulting in renal agenesis. Caudal truncation of the WD in HC2;ROSADTA/+ mutants is less severe but more variable. When the WD extends beyond the UB budding site but fails to reach the cloaca, nephrogenesis can occur in the absence of ureter-bladder connection, resulting in ureteral fistulas. If the WD manages to reach the cloaca, nephrogenesis and ureter-bladder connection can occur, although frequently accompanied by renal hypoplasia, hydroureter, and hydronephrosis (Fig. 1 and data not shown). These later defects may result from the ongoing DTA-induced apoptosis in the ureteral bud derivative and the subsequent negative effects on MM differentiation. Thus, quantitative differences in apoptosis and caudal truncation can translate into qualitatively distinct outcomes (renal agenesis, urogenital fistulas, or other urinary tract defects) in affected individuals.
In humans, fistulas of the genitourinary tract can involve multiple structures, such as kidney, ureter, bladder, urethra and the reproductive ducts. These fistulae can be congenital or secondary to infection, inflammatory disease, neoplasms, trauma or surgery.18–21 Congenital ureteral fistulae connected to the male or the female reproductive tracts are also called ureteral ectopia and their etiology is unclear in most patients.22–25 In many male HC2;ROSADTA/+ mutants, the CND was missing while the ureteral bud reached the MM to initiate nephrogenesis. In such mutants, the UB failed to separate from the WD, resulting in aberrant connection of their derivatives, the ureter and the male reproductive tract. Thus, the transient CND is critical for the separation of the UB and the WD during ureter maturation. The developmental sequence involving the vertical migration of the UB along the CND, lateral migration of the ureteral bud away from the WD and the degeneration of the CND26 appears to be the only pathway for the separation of the ureter and WD derivatives.
After emerging at E12.5 near the anterior end of the urogenital ridge, the MD normally follows the WD toward the cloaca area.27–30 The directional change in MD elongation to follow the ureter in the event of WD caudal truncation (Fig. 3) suggests that the ureter retains the guidance ability of the WD for MD development. While previous studies suggest that the WD does not contribute cells to the MD,27 the presence of the ureter-oviduct/uterus fistulas raises the possibility that direct cell-cell exchange between the WD and the MD becomes possible under pathological conditions. Apoptosis in some of the WD cells may compromise the integrity of the WD, providing the tightly associated MD entry points and establishing the fistulas between WD and MD derivatives.
Urogenital defects associated with WD cell death emphasize the central role of the WD in providing progenitors and signals for urogenital development. Although the WD signals for regulating MD are still unclear, the ureter appears to be able to provide such signals for MD guidance. The observed ureter-oviduct/uterus fistulae suggest that cell death allows otherwise impossible direct epithelial connection between WD and MD. Results from this study also indicate that quantitative differences in cell deathinduced caudal truncation of the WD can lead to a wide range of qualitatively distinct defects and that cell death can serve as a single etiological cause of a wide spectrum of congenital kidney and urinary tract defects.
Drs. M. R. Capecchi, A. P. McMahon, and P. Soriano for the ROSADTA, Hoxb7-Cre, and the ROSALacZ mice, respectively.