Pediatr Surg Int (2014) 30:189–195 DOI 10.1007/s00383-013-3452-z
ORIGINAL ARTICLE
Abnormal neural crest innervation in Sox10-Venus mice with all-trans retinoic acid-induced anorectal malformations Ryota Suzuki • Katsumi Miyahara • Hiroshi Murakami • Takashi Doi • Geoffrey J. Lane • Yo Mabuchi • Nobuharu Suzuki Atsuyuki Yamataka • Chihiro Akazawa
•
Published online: 19 December 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Background/purpose Despite technical advances in the surgical/medical care of anorectal malformation (ARM), persistent unsatisfactory postoperative bowel habit has been attributed to histopathologic abnormalities of the distal rectum/pouch (DRP) and hypoplasia of anal sphincter muscles (ASM). We used Sox10-Venus mice with ARM induced by all-trans retinoic acid (ATRA) to investigate neural crest cell (NCC) innervation in the DRP and ASM. Method Pregnant Sox10-Venus mice were administered single doses of 50, 70, or 100 mg/kg of ATRA on embryonic day 8.5 (E8.5) then sacrificed on either E16.5 or E19.5. Bowel specimens comprising the anorectum were examined using fluorescence microscopy without immunohistochemical staining (FMIS). Anti-PGP9.5 was used to delineate ganglion cells and anti-SMA for smooth muscles. Results The appropriate dose of ATRA for inducing ARM was 50 mg/kg. Under FMIS, all ARM embryos (n = 5; all high type; 3 male:2 female) had less NCC innervation with thick Venus-positive nerve fibers in the DRP compared with normal embryos (n = 8); there was abnormal NCC innervation in the DRP and absent ASM in ARM mice.
R. Suzuki Y. Mabuchi N. Suzuki C. Akazawa (&) Department of Biochemistry and Biophysics, Graduate School of Health Care Sciences, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, 113-8510 Tokyo, Japan e-mail: [email protected] K. Miyahara H. Murakami T. Doi G. J. Lane A. Yamataka Department of Pediatric General and Urogenital Surgery, Juntendo University School of Medicine, 113-8421 Tokyo, Japan
Conclusion We are the first to delineate abnormal enteric nervous system innervation in the DRP of ARM mice without using immunohistochemical staining techniques thus allowing specimens to be examined without any distortion. Keywords Retinoic acid Anorectal malformation Neural crest innervation Anal sphincter Sox10 Transgenic mouse
Introduction Normal ENS innervation requires interaction between enteric neurons and glial cells, both derived from vagal and sacral neural crests, for proliferation, differentiation, and extensive migration [1, 2]. During these processes, the Sox10 gene is expressed closely with neural crest lineage cells and plays an essential role in the function and survival of neural crest cells (NCCs) [3–6]. Anorectal malformation (ARM), occurs at a frequency of one per five thousand live births [7, 8]. Although surgical intervention for ARM has improved, incomplete evacuation of feces, constipation, and incontinence [9–12] continue to be a problem for many patients postoperatively. Previous studies have shown that ARM is caused by abnormalities of the sacral roots, however the pathogenesis of ARM is poorly understood [13–17]. Research on the relation between ARM and histology of the distal rectum/ pouch (DRP) indicates there are histologic features of the DRP such as abnormal innervation, i.e., aganglionosis with similar characteristics to those seen in Hirschsprung’s disease (HD) [18] with HD being characterized by an absence of the enteric nervous system (ENS) in the terminal regions of the gut, because of an arrest of the migration of enteric neurons from the neural crest.
123
190
Recently, we created Sox10-Venus mice, which overexpress Venus, a modified green fluorescent protein, under the control of Sox10 promoter to visualize Sox10-Venuspositive NCCs in vivo [6, 19, 20]. In this study, we used a mouse model for ARM induced by all-trans retinoic acid (ATRA) to delineate morphological features without using complicated staining techniques, thus eliminating physical aberrations.
Materials and methods ARM-model mouse To analyze NCCs, we used Sox10-Venus transgenic mice reported by Shibata et al. [6]. The genotyping of Sox10Venus mice and embryos was carried out by the polymerase chain reaction protocol as previously described [6]. The day a vaginal plug was observed as a sign of successful mating was designated as embryonic day 0.5 (E0.5). Pregnant Sox10-Venus mice were administered single doses of 50, 70, or 100 mg/kg of ATRA (Sigma-Aldrich, USA) dissolved in olive oil (Wako, Japan) via gavage on E8.5 and were sacrificed on either E16.5 or E19.5 by cervical dislocation. Embryos from Sox10-Venus mice not given ATRA were used as controls (n = 8). All embryos were dissected free under fluorescence stereo microscopy (FSM; M165FC, Leica, Germany) and images taken with FSM. All animal protocols were approved by the Juntendo University School of Medicine Animal Care and Use Committee (Institutional review board No. 230033). Analysis of the morphology of the DRP using fluorescence stereo microscopy The entire distal rectum/pouch (DRP) was excised from each embryo, fixed in 4 % paraformaldehyde and examined. Images were taken using FSM.
Pediatr Surg Int (2014) 30:189–195
with 40 ,6 diamidino-2-phenylindole (DAPI) (Vector Laboratories, USA). The stained samples were observed with a TSC-SP5 laser-scanning microscope (Leica, Germany). Quantification of Venus positive cells and PGP9.5 positive cells The number of neural crest cells (Venus positive cells: Venus? cells) and enteric ganglion cells (PGP9.5 positive cells: PGP9.5? cells) were counted blindly by three investigators in three sections from each specimen blindly. Scores were normalized with the total number of cells (DAPI positive cells: DAPI? cells) and compared statistically. All numerical data were presented as mean ± standard deviation (SD). Statistical analysis Differences between groups were tested using the student’s t test. Statistical significance was defined as p \ 0.05.
Results ATRA induced ARM in Sox10-Venus mice It is known that teratogenic doses of ATRA induce hydronephrosis, vertebral malformation and ARM in murine embryos [21, 22]. To create ARM, Sox10-Venus mice were subjected to ATRA orally (Table 1). We found the optimal dose of ATRA for inducing ARM successfully was 50 mg/ kg. A trial of 100 mg/kg of ATRA was lethal in 62.5 % of pregnant mice and in 3 pregnant ATRA mice that survived, all their embryos died in utero by E16.5; and with a trial of 70 mg/kg of ATRA, 2/3 of pregnant mice survived, but all their embryos died. With a trial of 50 mg/kg, all pregnant mice survived. Eleven embryos were harvested from 3 pregnant mice, and 5/11 (45 %) had ARM. One of ARM embryo also developed meningomyelocele.
Immunohistochemistry After washing with phosphate buffered saline (PBS), specimens were immersed in 20 % sucrose, embedded in OCT compound (Sakura Finetek, USA) and sectioned 10 lm thick. Sections were first incubated overnight at 4 °C with anti-protein gene product 9.5 (PGP9.5) (rabbit polyclonal, 1:1,000, Enzo Life Science, USA) and anti-asmooth muscle actin (SMA) (mouse monoclonal, 1:200, Dako, Denmark). Immunoreactivity was visualized using secondary antibodies conjugated with Alexa Fluor 594 (goat anti-rabbit IgG, 1:1,000, Life Technologies, USA) and Alexa Fluor 594 (goat anti-mouse IgG, 1:200, life technologies, USA). Nuclear counterstaining was performed
123
Table 1 Induced ARM in pregnant Sox10-Venus mice Concentration of ATRA (mg/kg)
Surviving mice/all mice
Surviving mice with viable embryos/ surviving mice
ARM embryos/ viable embryos
50
6/6 (100)
3/6 (50)
5a/11 (45)
70
2/3 (66.7)
0/2 (0)
0/0 (0)
100
3/8 (37.5)
0/3 (0)
0/0 (0)
ATRA all-trans retinoic acid a
Three embryos were analyzed at E16.5; two embryos at E19.5
one embryo had meningomyelocele ( ) data expressed as a percentage
Pediatr Surg Int (2014) 30:189–195
Embryos were examined on E16.5 or E19.5 (Fig. 1). In all ARM embryos, the tail was absent and the anus was imperforate (Fig. 1a, b). In contrast, compared to the ARM embryos, anorectum was normal in control embryo (Fig. 1c, d). Under FSM, the ENS could be observed clearly, without any staining. We found the anal area in both control and ARM embryos were strongly positive for Sox10-Venus (Fig. 1b, d). Morphology and histopathology in ARM embryos In order to assess abnormal morphology in ARM mice, sagittal sections from E16.5 embryos were stained with hematoxylin and eosin (Fig. 2a, a0 ). Most organs in ARM
Fig. 1 Visualizing the E16.5 embryos using fluorescence stereo microscopy. Comparison of morphology between control and ARM embryos, visualized using bright field microscopy and fluorescence
191
embryos were normal except for the bladder, which was abnormally swollen and filled with fluid in comparison with control embryos (Fig. 2f, f0 ; asterisk). The terminal DRP was located at the neck of the urinary bladder in all ARM embryos, thus ARM was classified as being high. While Sox10-Venus? cells were observed widely from the rectum to the anus in control embryos (Fig. 2g, g0 ), Sox10Venus? cells were only present around the anus in ARM embryos (Fig. 2b, b0 ). SMA positive cells (SMA positive cells: SMA? cells) were found around the bladder, and the internal and external anal sphincter muscles (ASM) in controls (Fig. 2h, h0 ). In ARM embryos, SMA? cells were found only around the bladder (Fig. 2c, c0 ). These data indicate that although
stereo microscopy. The tail and genital tubercle (G) are labeled. The rectangle indicates the anus. The arrowhead indicates absent anus
123
192
123
Pediatr Surg Int (2014) 30:189–195
Pediatr Surg Int (2014) 30:189–195 b Fig. 2 Visualizing the anal sphincter muscle in E16.5 embryos. Comparison of specimens from E16.5 control and ARM embryos. a, f hematoxylin and eosin, b, g Venus, c, h SMA, d, i DAPI, e, j merged images, a–e Control embryos, f–j ARM embryos, a0 –j0 92 magnified section, asterisk Bladder, Scale bars 1 mm
anal morphology was abnormal, Sox10-Venus? cells were present in the anus of ARM embryos. Analysis of neural crest derived cells in ARM embryos To investigate the correlation between PGP9.5 staining and NCCs in ARM embryos, we performed a detailed analysis of Sox10 positive cells in the ENS of E19.5 embryos. Immunohistochemistry showed that Sox10-Venus? and
193
PGP9.5? cells were found around the whole circumference of the entire colon and DRP wall in control mice (Fig. 3i– p). In contrast, the number of Sox10-Venus? and PGP9.5? cells in the colon wall was remarkably decreased in ARM embryos (Fig. 3a–h). We counted Sox10-Venus? cells and PGP9.5? cells in the colon and DRP (Fig. 4) and found there were 67.3 ± 8.35 % Sox10-Venus? cells in control colon, 78.5 ± 2.56 % in control DRP, 40.8 ± 3.84 % in ARM colon, and 45.9 ± 5.15 % in ARM DRP. For PGP9.5? cells, there were 32.4 ± 3.66 % in control colon, 36.5 ± 3.13 % in control DRP, 27.6 ± 5.42 % in ARM colon, and 6.10 ± 1.49 % in ARM DRP. Interestingly, extrinsic nerve fibers were spread along the DRP wall from a clump of Sox10-Venus? cells in ARM embryos (Fig. 5).
Fig. 3 Visualizing the colon and rectum in E19.5 embryos. a, e, i, m Venus, b, f, j, n PGP9.5, c, g, k, o DAPI, d, h, l, p merged images, a–h Control embryos, i–p ARM embryos, Scale bars 150 lm
123
194
Pediatr Surg Int (2014) 30:189–195
The ENS is mainly derived from vagal and sacral NCCs. Vagal NCCs invade the gastrointestinal tract, migrate from rostral to caudal, and form a neural plexus. Sacral NCCs
that migrate from caudal to rostral colonize a small terminal portion of the hindgut [20]. HD is a developmental disorder of the ENS, characterized by an absence of ganglion cells in the distal colon resulting in functional obstruction. In HD, thick nerve fibers from sacral NCCs can sometimes be associated with the aganglionic segment because there is failure of vagal NCC migration [19]. In ARM there is also some disruption to migration that is not fully understood and it is not certain whether there is abnormal proliferation and differentiation of NCCs in ARM either. A recent study using ethylenethiourea-treated rat ARM model embryos analyzed ENS patterning using antibody staining against nitric oxide synthases (NOs). Although significant decreases in NOs and c-kit expression were described, there was no description about the ENS morphology [23] which prompted us to analyze ENS morphology by inducing ARM in newly developed Sox10Venus mice using ATRA, a known teratogen that induces various malformations depending on dosage and timing of exposure. ATRA inhibits cell migration in slice preparations of embryonic mouse forebrains [24] and disrupts normal morphogenesis in the hindgut and caudal region of mouse embryos [25], and ATRA prevents normal development of the placenta, resulting in embryonic death in utero. ARM has been reported to have been induced in mouse embryos by the administration of 100 mg/kg of ATRA, but in the present study, 50 mg/kg of ATRA was found to be sufficient. Further studies are required to investigate the appropriate timing and dosage for administering ATRA.
Fig. 5 Visualizing the DRP in E19.5 embryos. a, d Macroscopic, b, e Distribution of Sox10-VENUS? cells/nerve fibers in the rectum, c, f Expression of PGP9.5, Venus, and DRP in the distal colorectum, a–c
Control embryos, d–f ARM embryos, asterisk marker to indicate the same point, arrowheads indicate increased extrinsic nerve fibers, Scale bars 150 lm
Fig. 4 Venus?/DAPI? and PGP9.5?/DAPI? for control and ARM embryos. Columns represent mean ± SD for three different sections from each specimen. *p \ 0.05 statistically significant difference compared with controls
In comparison with controls, there were fewer Sox10Venus? cells in the DRP of ARM embryos.
Discussion
123
Pediatr Surg Int (2014) 30:189–195
When animal models are employed for the study of human diseases, the critical issue is that the model must mimic the pathological morbidity seen clinically as closely as possible. In Sox10-Venus mice, Venus faithfully mirrors endogenous Sox10 expression, thus they a useful tool for studying both the development and pathology of NCCs. Similarly, ARM induced by ATRA are similar to clinical ARM and bear some resemblance to HD. Thus, we hypothesized that Sox10-Venus mice with ARM induced by ATRA could serve as an animal model for evaluating ENS development in mouse embryos by visualizing NCCs. There are many reports describing patients with ARM having ongoing problems with continence after pullthrough surgery. Our data demonstrate that Sox10-Venus? cells and PGP9.5? cells were decreased in ARM colon and DRP compared with controls. Furthermore, PGP9.5? cells were dramatically reduced in ARM embryo DRP. Based on this, we hypothesized that ATRA may inhibit migration of NCCs and differentiation into PGP9.5? cells or NCC survival. Our data also show that SMA expression is not detected in internal and external ASM in ARM, allowing us to hypothesize that insufficient development of internal and external ASM is involved in the mechanism for persistent continence problems in postoperative ARM patients. From our results, it would also seem there is failure of vagal NCC migration in the terminal hindgut. This may result in the absence of ganglion cells and proliferation of extrinsic nerve fibers. This disruption of vagal NCC migration may also induce sacral NCC activation in the DRP. In summary, we found that NCC morphology and histology observed in Sox10-Venus mice with ATRA induced ARM could be a possible technique for studying ENS development. Further analyses are necessary to address the pathogenesis of ARM with regard to the internal and external ASM and NCC innervation. Acknowledgments We are very grateful for technical support from the Division of Biomedical Imaging Research, BioMedical Research Center, Juntendo Graduate School of Medicine. This work was supported by the Ministry of Health, Welfare, and Labour, Japan, and the Ministry of Education, Culture, Sports, Science and Technology, Japan.
References 1. Durbec PL, Larsson-Blomberg LB, Schuchardt A et al (1996) Common origin and developmental dependence on c-ret of subsets of enteric and sympathetic neuroblasts. Development 122:349–358 2. Taraviras S, Pachnis V (1999) Development of the mammalian enteric nervous system. Curr Opin Genet Dev 9:321–327 3. Heanue TA, Pachnis V (2007) Enteric nervous system development and Hirschsprung’s disease: advances in genetic and stem cell studies. Nat Rev Neurosci 8:466–479
195 4. Amiel J, Sproat-Emison E, Garcia-Barcelo M et al (2008) Hirschsprung’s disease, associated syndromes and genetics: a review. J Med Genet 45:1–14 5. Southard-Smith EM, Kos L, Pavan WJ (1998) Sox10 mutation disrupts neural crest development in Dom Hirschsprung’s mouse model. Nat Genet 18:60–64 6. Shibata S, Yasuda A, Renault-Mihara F et al (2010) Sox10-Venus mice: a new tool for real-time labeling of neural crest lineage cells and oligodendrocytes. Mol Brain 3:31 7. Levitt MA, Pen˜a A (2007) Anorectal malformations. Orphanet J Rare Dis 2:33 8. Cuschieri A, EUROCAT Working Group (2001) Descriptive epidemiology of isolated anal anomalies: a survey of 4.6 million births in Europe. Am J Med Genet 103:207–215 9. Pena A, Hong A (2000) Advances in the management of anorectal malformations. Am J Surg 180:370–376 10. Bai Y, Yuan Z, Wang W et al (2000) Quality of life for children with fecal incontinence after surgically corrected anorectal malformation. J Pediatr Surg 35:462–464 11. Holschneider AM, Koebke J, Meier-Ruge W et al (2001) Pathophysiology of chronic constipation in anorectal malformations. Long-term results and preliminary anatomical investigations. Eur J Pediatr Surg 11:305–310 12. Rintala R (1990) Postoperative internal sphincter function in anorectal malformation—a manometric study. Pediatr Surg Int 5:127–130 13. Li L, Li Z, Wang LY et al (1993) Anorectal anomaly: neuropathological changes in the sacral spinal cord. J Pediatr Surg 28:880–885 14. Meier-Ruge WA, Holschneider AM (2000) Histopathologic observations of anorectal abnormalities in anal atresia. Pediatr Surg Int 16:2–7 15. Fernandez-Fraga X, Azpiroz F, Malagelada JR (2002) Significance of pelvic floor muscles in anal incontinence. Gastroenterology 123:1441–1450 16. Yuan ZW, Lui VC, Tam PK (2003) Deficient motor innervation of the sphincter mechanism in fetal rats with anorectal malformation: a quantitative study by fluorogold retrograde tracing. J Pediatr Surg 38:1383–1388 17. Jia HM, Zhang KR, Zhang SC et al (2007) Quantitative analysis of sacral parasympathetic nucleus innervating the rectum in rats with anorectal malformation. J Pediatr Surg 42:1544–1548 18. Holschneider AM, Pfrommer W, Gerresheim B (1994) Results in the treatment of anorectal malformations with special regard to the histology of the rectal pouch. Eur J Pediatr Surg 4(5):303–309 19. Miyahara K, Kato Y, Koga H et al (2010) Abnormal enteric innervation identified without histopathologic staining in aganglionic colorectum from a mouse model of Hirschsprung’s disease. J Pediatr Surg 45:2403–2407 20. Miyahara K, Kato Y, Koga H et al (2011) Visualization of enteric neural crest cell migration in SOX10 transgenic mouse gut using time-lapse fluorescence imaging. J Pediatr Surg 46:2305–2308 21. Kubota Y, Shimotake T, Iwai N (2000) Congenital anomalies in mice induced by etretinate. Eur J Pediatr Surg 10:248–251 22. Kubota Y, Shimotake T, Iwai N et al (1998) Development of anorectal malformations using etretinate. J Pediatr Surg 33:127–129 23. Wang W, Jia H, Zhang H, Chen Q et al (2011) Abnormal innervation patterns in the anorectum of ETU-induced fetal rats with anorectal malformations. Neurosci Lett 495:88–92 24. Crandall JE, Goodman T, McCarthy DM et al (2011) Retinoic acid influences neuronal migration from the ganglionic eminence to the cerebral cortex. J Neurochem 119:723–735 25. Hashimoto R, Nagaya M, Ishiguro Y et al (2002) Relationship of fistulas to the rectum and genitourinary tract in mouse fetuses with high anorectal malformations induced by all-trans retinoic acid. Pediatr Surg Int 18:723–727
123
ORIGINAL ARTICLE
Abnormal neural crest innervation in Sox10-Venus mice with all-trans retinoic acid-induced anorectal malformations Ryota Suzuki • Katsumi Miyahara • Hiroshi Murakami • Takashi Doi • Geoffrey J. Lane • Yo Mabuchi • Nobuharu Suzuki Atsuyuki Yamataka • Chihiro Akazawa
•
Published online: 19 December 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Background/purpose Despite technical advances in the surgical/medical care of anorectal malformation (ARM), persistent unsatisfactory postoperative bowel habit has been attributed to histopathologic abnormalities of the distal rectum/pouch (DRP) and hypoplasia of anal sphincter muscles (ASM). We used Sox10-Venus mice with ARM induced by all-trans retinoic acid (ATRA) to investigate neural crest cell (NCC) innervation in the DRP and ASM. Method Pregnant Sox10-Venus mice were administered single doses of 50, 70, or 100 mg/kg of ATRA on embryonic day 8.5 (E8.5) then sacrificed on either E16.5 or E19.5. Bowel specimens comprising the anorectum were examined using fluorescence microscopy without immunohistochemical staining (FMIS). Anti-PGP9.5 was used to delineate ganglion cells and anti-SMA for smooth muscles. Results The appropriate dose of ATRA for inducing ARM was 50 mg/kg. Under FMIS, all ARM embryos (n = 5; all high type; 3 male:2 female) had less NCC innervation with thick Venus-positive nerve fibers in the DRP compared with normal embryos (n = 8); there was abnormal NCC innervation in the DRP and absent ASM in ARM mice.
R. Suzuki Y. Mabuchi N. Suzuki C. Akazawa (&) Department of Biochemistry and Biophysics, Graduate School of Health Care Sciences, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, 113-8510 Tokyo, Japan e-mail: [email protected] K. Miyahara H. Murakami T. Doi G. J. Lane A. Yamataka Department of Pediatric General and Urogenital Surgery, Juntendo University School of Medicine, 113-8421 Tokyo, Japan
Conclusion We are the first to delineate abnormal enteric nervous system innervation in the DRP of ARM mice without using immunohistochemical staining techniques thus allowing specimens to be examined without any distortion. Keywords Retinoic acid Anorectal malformation Neural crest innervation Anal sphincter Sox10 Transgenic mouse
Introduction Normal ENS innervation requires interaction between enteric neurons and glial cells, both derived from vagal and sacral neural crests, for proliferation, differentiation, and extensive migration [1, 2]. During these processes, the Sox10 gene is expressed closely with neural crest lineage cells and plays an essential role in the function and survival of neural crest cells (NCCs) [3–6]. Anorectal malformation (ARM), occurs at a frequency of one per five thousand live births [7, 8]. Although surgical intervention for ARM has improved, incomplete evacuation of feces, constipation, and incontinence [9–12] continue to be a problem for many patients postoperatively. Previous studies have shown that ARM is caused by abnormalities of the sacral roots, however the pathogenesis of ARM is poorly understood [13–17]. Research on the relation between ARM and histology of the distal rectum/ pouch (DRP) indicates there are histologic features of the DRP such as abnormal innervation, i.e., aganglionosis with similar characteristics to those seen in Hirschsprung’s disease (HD) [18] with HD being characterized by an absence of the enteric nervous system (ENS) in the terminal regions of the gut, because of an arrest of the migration of enteric neurons from the neural crest.
123
190
Recently, we created Sox10-Venus mice, which overexpress Venus, a modified green fluorescent protein, under the control of Sox10 promoter to visualize Sox10-Venuspositive NCCs in vivo [6, 19, 20]. In this study, we used a mouse model for ARM induced by all-trans retinoic acid (ATRA) to delineate morphological features without using complicated staining techniques, thus eliminating physical aberrations.
Materials and methods ARM-model mouse To analyze NCCs, we used Sox10-Venus transgenic mice reported by Shibata et al. [6]. The genotyping of Sox10Venus mice and embryos was carried out by the polymerase chain reaction protocol as previously described [6]. The day a vaginal plug was observed as a sign of successful mating was designated as embryonic day 0.5 (E0.5). Pregnant Sox10-Venus mice were administered single doses of 50, 70, or 100 mg/kg of ATRA (Sigma-Aldrich, USA) dissolved in olive oil (Wako, Japan) via gavage on E8.5 and were sacrificed on either E16.5 or E19.5 by cervical dislocation. Embryos from Sox10-Venus mice not given ATRA were used as controls (n = 8). All embryos were dissected free under fluorescence stereo microscopy (FSM; M165FC, Leica, Germany) and images taken with FSM. All animal protocols were approved by the Juntendo University School of Medicine Animal Care and Use Committee (Institutional review board No. 230033). Analysis of the morphology of the DRP using fluorescence stereo microscopy The entire distal rectum/pouch (DRP) was excised from each embryo, fixed in 4 % paraformaldehyde and examined. Images were taken using FSM.
Pediatr Surg Int (2014) 30:189–195
with 40 ,6 diamidino-2-phenylindole (DAPI) (Vector Laboratories, USA). The stained samples were observed with a TSC-SP5 laser-scanning microscope (Leica, Germany). Quantification of Venus positive cells and PGP9.5 positive cells The number of neural crest cells (Venus positive cells: Venus? cells) and enteric ganglion cells (PGP9.5 positive cells: PGP9.5? cells) were counted blindly by three investigators in three sections from each specimen blindly. Scores were normalized with the total number of cells (DAPI positive cells: DAPI? cells) and compared statistically. All numerical data were presented as mean ± standard deviation (SD). Statistical analysis Differences between groups were tested using the student’s t test. Statistical significance was defined as p \ 0.05.
Results ATRA induced ARM in Sox10-Venus mice It is known that teratogenic doses of ATRA induce hydronephrosis, vertebral malformation and ARM in murine embryos [21, 22]. To create ARM, Sox10-Venus mice were subjected to ATRA orally (Table 1). We found the optimal dose of ATRA for inducing ARM successfully was 50 mg/ kg. A trial of 100 mg/kg of ATRA was lethal in 62.5 % of pregnant mice and in 3 pregnant ATRA mice that survived, all their embryos died in utero by E16.5; and with a trial of 70 mg/kg of ATRA, 2/3 of pregnant mice survived, but all their embryos died. With a trial of 50 mg/kg, all pregnant mice survived. Eleven embryos were harvested from 3 pregnant mice, and 5/11 (45 %) had ARM. One of ARM embryo also developed meningomyelocele.
Immunohistochemistry After washing with phosphate buffered saline (PBS), specimens were immersed in 20 % sucrose, embedded in OCT compound (Sakura Finetek, USA) and sectioned 10 lm thick. Sections were first incubated overnight at 4 °C with anti-protein gene product 9.5 (PGP9.5) (rabbit polyclonal, 1:1,000, Enzo Life Science, USA) and anti-asmooth muscle actin (SMA) (mouse monoclonal, 1:200, Dako, Denmark). Immunoreactivity was visualized using secondary antibodies conjugated with Alexa Fluor 594 (goat anti-rabbit IgG, 1:1,000, Life Technologies, USA) and Alexa Fluor 594 (goat anti-mouse IgG, 1:200, life technologies, USA). Nuclear counterstaining was performed
123
Table 1 Induced ARM in pregnant Sox10-Venus mice Concentration of ATRA (mg/kg)
Surviving mice/all mice
Surviving mice with viable embryos/ surviving mice
ARM embryos/ viable embryos
50
6/6 (100)
3/6 (50)
5a/11 (45)
70
2/3 (66.7)
0/2 (0)
0/0 (0)
100
3/8 (37.5)
0/3 (0)
0/0 (0)
ATRA all-trans retinoic acid a
Three embryos were analyzed at E16.5; two embryos at E19.5
one embryo had meningomyelocele ( ) data expressed as a percentage
Pediatr Surg Int (2014) 30:189–195
Embryos were examined on E16.5 or E19.5 (Fig. 1). In all ARM embryos, the tail was absent and the anus was imperforate (Fig. 1a, b). In contrast, compared to the ARM embryos, anorectum was normal in control embryo (Fig. 1c, d). Under FSM, the ENS could be observed clearly, without any staining. We found the anal area in both control and ARM embryos were strongly positive for Sox10-Venus (Fig. 1b, d). Morphology and histopathology in ARM embryos In order to assess abnormal morphology in ARM mice, sagittal sections from E16.5 embryos were stained with hematoxylin and eosin (Fig. 2a, a0 ). Most organs in ARM
Fig. 1 Visualizing the E16.5 embryos using fluorescence stereo microscopy. Comparison of morphology between control and ARM embryos, visualized using bright field microscopy and fluorescence
191
embryos were normal except for the bladder, which was abnormally swollen and filled with fluid in comparison with control embryos (Fig. 2f, f0 ; asterisk). The terminal DRP was located at the neck of the urinary bladder in all ARM embryos, thus ARM was classified as being high. While Sox10-Venus? cells were observed widely from the rectum to the anus in control embryos (Fig. 2g, g0 ), Sox10Venus? cells were only present around the anus in ARM embryos (Fig. 2b, b0 ). SMA positive cells (SMA positive cells: SMA? cells) were found around the bladder, and the internal and external anal sphincter muscles (ASM) in controls (Fig. 2h, h0 ). In ARM embryos, SMA? cells were found only around the bladder (Fig. 2c, c0 ). These data indicate that although
stereo microscopy. The tail and genital tubercle (G) are labeled. The rectangle indicates the anus. The arrowhead indicates absent anus
123
192
123
Pediatr Surg Int (2014) 30:189–195
Pediatr Surg Int (2014) 30:189–195 b Fig. 2 Visualizing the anal sphincter muscle in E16.5 embryos. Comparison of specimens from E16.5 control and ARM embryos. a, f hematoxylin and eosin, b, g Venus, c, h SMA, d, i DAPI, e, j merged images, a–e Control embryos, f–j ARM embryos, a0 –j0 92 magnified section, asterisk Bladder, Scale bars 1 mm
anal morphology was abnormal, Sox10-Venus? cells were present in the anus of ARM embryos. Analysis of neural crest derived cells in ARM embryos To investigate the correlation between PGP9.5 staining and NCCs in ARM embryos, we performed a detailed analysis of Sox10 positive cells in the ENS of E19.5 embryos. Immunohistochemistry showed that Sox10-Venus? and
193
PGP9.5? cells were found around the whole circumference of the entire colon and DRP wall in control mice (Fig. 3i– p). In contrast, the number of Sox10-Venus? and PGP9.5? cells in the colon wall was remarkably decreased in ARM embryos (Fig. 3a–h). We counted Sox10-Venus? cells and PGP9.5? cells in the colon and DRP (Fig. 4) and found there were 67.3 ± 8.35 % Sox10-Venus? cells in control colon, 78.5 ± 2.56 % in control DRP, 40.8 ± 3.84 % in ARM colon, and 45.9 ± 5.15 % in ARM DRP. For PGP9.5? cells, there were 32.4 ± 3.66 % in control colon, 36.5 ± 3.13 % in control DRP, 27.6 ± 5.42 % in ARM colon, and 6.10 ± 1.49 % in ARM DRP. Interestingly, extrinsic nerve fibers were spread along the DRP wall from a clump of Sox10-Venus? cells in ARM embryos (Fig. 5).
Fig. 3 Visualizing the colon and rectum in E19.5 embryos. a, e, i, m Venus, b, f, j, n PGP9.5, c, g, k, o DAPI, d, h, l, p merged images, a–h Control embryos, i–p ARM embryos, Scale bars 150 lm
123
194
Pediatr Surg Int (2014) 30:189–195
The ENS is mainly derived from vagal and sacral NCCs. Vagal NCCs invade the gastrointestinal tract, migrate from rostral to caudal, and form a neural plexus. Sacral NCCs
that migrate from caudal to rostral colonize a small terminal portion of the hindgut [20]. HD is a developmental disorder of the ENS, characterized by an absence of ganglion cells in the distal colon resulting in functional obstruction. In HD, thick nerve fibers from sacral NCCs can sometimes be associated with the aganglionic segment because there is failure of vagal NCC migration [19]. In ARM there is also some disruption to migration that is not fully understood and it is not certain whether there is abnormal proliferation and differentiation of NCCs in ARM either. A recent study using ethylenethiourea-treated rat ARM model embryos analyzed ENS patterning using antibody staining against nitric oxide synthases (NOs). Although significant decreases in NOs and c-kit expression were described, there was no description about the ENS morphology [23] which prompted us to analyze ENS morphology by inducing ARM in newly developed Sox10Venus mice using ATRA, a known teratogen that induces various malformations depending on dosage and timing of exposure. ATRA inhibits cell migration in slice preparations of embryonic mouse forebrains [24] and disrupts normal morphogenesis in the hindgut and caudal region of mouse embryos [25], and ATRA prevents normal development of the placenta, resulting in embryonic death in utero. ARM has been reported to have been induced in mouse embryos by the administration of 100 mg/kg of ATRA, but in the present study, 50 mg/kg of ATRA was found to be sufficient. Further studies are required to investigate the appropriate timing and dosage for administering ATRA.
Fig. 5 Visualizing the DRP in E19.5 embryos. a, d Macroscopic, b, e Distribution of Sox10-VENUS? cells/nerve fibers in the rectum, c, f Expression of PGP9.5, Venus, and DRP in the distal colorectum, a–c
Control embryos, d–f ARM embryos, asterisk marker to indicate the same point, arrowheads indicate increased extrinsic nerve fibers, Scale bars 150 lm
Fig. 4 Venus?/DAPI? and PGP9.5?/DAPI? for control and ARM embryos. Columns represent mean ± SD for three different sections from each specimen. *p \ 0.05 statistically significant difference compared with controls
In comparison with controls, there were fewer Sox10Venus? cells in the DRP of ARM embryos.
Discussion
123
Pediatr Surg Int (2014) 30:189–195
When animal models are employed for the study of human diseases, the critical issue is that the model must mimic the pathological morbidity seen clinically as closely as possible. In Sox10-Venus mice, Venus faithfully mirrors endogenous Sox10 expression, thus they a useful tool for studying both the development and pathology of NCCs. Similarly, ARM induced by ATRA are similar to clinical ARM and bear some resemblance to HD. Thus, we hypothesized that Sox10-Venus mice with ARM induced by ATRA could serve as an animal model for evaluating ENS development in mouse embryos by visualizing NCCs. There are many reports describing patients with ARM having ongoing problems with continence after pullthrough surgery. Our data demonstrate that Sox10-Venus? cells and PGP9.5? cells were decreased in ARM colon and DRP compared with controls. Furthermore, PGP9.5? cells were dramatically reduced in ARM embryo DRP. Based on this, we hypothesized that ATRA may inhibit migration of NCCs and differentiation into PGP9.5? cells or NCC survival. Our data also show that SMA expression is not detected in internal and external ASM in ARM, allowing us to hypothesize that insufficient development of internal and external ASM is involved in the mechanism for persistent continence problems in postoperative ARM patients. From our results, it would also seem there is failure of vagal NCC migration in the terminal hindgut. This may result in the absence of ganglion cells and proliferation of extrinsic nerve fibers. This disruption of vagal NCC migration may also induce sacral NCC activation in the DRP. In summary, we found that NCC morphology and histology observed in Sox10-Venus mice with ATRA induced ARM could be a possible technique for studying ENS development. Further analyses are necessary to address the pathogenesis of ARM with regard to the internal and external ASM and NCC innervation. Acknowledgments We are very grateful for technical support from the Division of Biomedical Imaging Research, BioMedical Research Center, Juntendo Graduate School of Medicine. This work was supported by the Ministry of Health, Welfare, and Labour, Japan, and the Ministry of Education, Culture, Sports, Science and Technology, Japan.
References 1. Durbec PL, Larsson-Blomberg LB, Schuchardt A et al (1996) Common origin and developmental dependence on c-ret of subsets of enteric and sympathetic neuroblasts. Development 122:349–358 2. Taraviras S, Pachnis V (1999) Development of the mammalian enteric nervous system. Curr Opin Genet Dev 9:321–327 3. Heanue TA, Pachnis V (2007) Enteric nervous system development and Hirschsprung’s disease: advances in genetic and stem cell studies. Nat Rev Neurosci 8:466–479
195 4. Amiel J, Sproat-Emison E, Garcia-Barcelo M et al (2008) Hirschsprung’s disease, associated syndromes and genetics: a review. J Med Genet 45:1–14 5. Southard-Smith EM, Kos L, Pavan WJ (1998) Sox10 mutation disrupts neural crest development in Dom Hirschsprung’s mouse model. Nat Genet 18:60–64 6. Shibata S, Yasuda A, Renault-Mihara F et al (2010) Sox10-Venus mice: a new tool for real-time labeling of neural crest lineage cells and oligodendrocytes. Mol Brain 3:31 7. Levitt MA, Pen˜a A (2007) Anorectal malformations. Orphanet J Rare Dis 2:33 8. Cuschieri A, EUROCAT Working Group (2001) Descriptive epidemiology of isolated anal anomalies: a survey of 4.6 million births in Europe. Am J Med Genet 103:207–215 9. Pena A, Hong A (2000) Advances in the management of anorectal malformations. Am J Surg 180:370–376 10. Bai Y, Yuan Z, Wang W et al (2000) Quality of life for children with fecal incontinence after surgically corrected anorectal malformation. J Pediatr Surg 35:462–464 11. Holschneider AM, Koebke J, Meier-Ruge W et al (2001) Pathophysiology of chronic constipation in anorectal malformations. Long-term results and preliminary anatomical investigations. Eur J Pediatr Surg 11:305–310 12. Rintala R (1990) Postoperative internal sphincter function in anorectal malformation—a manometric study. Pediatr Surg Int 5:127–130 13. Li L, Li Z, Wang LY et al (1993) Anorectal anomaly: neuropathological changes in the sacral spinal cord. J Pediatr Surg 28:880–885 14. Meier-Ruge WA, Holschneider AM (2000) Histopathologic observations of anorectal abnormalities in anal atresia. Pediatr Surg Int 16:2–7 15. Fernandez-Fraga X, Azpiroz F, Malagelada JR (2002) Significance of pelvic floor muscles in anal incontinence. Gastroenterology 123:1441–1450 16. Yuan ZW, Lui VC, Tam PK (2003) Deficient motor innervation of the sphincter mechanism in fetal rats with anorectal malformation: a quantitative study by fluorogold retrograde tracing. J Pediatr Surg 38:1383–1388 17. Jia HM, Zhang KR, Zhang SC et al (2007) Quantitative analysis of sacral parasympathetic nucleus innervating the rectum in rats with anorectal malformation. J Pediatr Surg 42:1544–1548 18. Holschneider AM, Pfrommer W, Gerresheim B (1994) Results in the treatment of anorectal malformations with special regard to the histology of the rectal pouch. Eur J Pediatr Surg 4(5):303–309 19. Miyahara K, Kato Y, Koga H et al (2010) Abnormal enteric innervation identified without histopathologic staining in aganglionic colorectum from a mouse model of Hirschsprung’s disease. J Pediatr Surg 45:2403–2407 20. Miyahara K, Kato Y, Koga H et al (2011) Visualization of enteric neural crest cell migration in SOX10 transgenic mouse gut using time-lapse fluorescence imaging. J Pediatr Surg 46:2305–2308 21. Kubota Y, Shimotake T, Iwai N (2000) Congenital anomalies in mice induced by etretinate. Eur J Pediatr Surg 10:248–251 22. Kubota Y, Shimotake T, Iwai N et al (1998) Development of anorectal malformations using etretinate. J Pediatr Surg 33:127–129 23. Wang W, Jia H, Zhang H, Chen Q et al (2011) Abnormal innervation patterns in the anorectum of ETU-induced fetal rats with anorectal malformations. Neurosci Lett 495:88–92 24. Crandall JE, Goodman T, McCarthy DM et al (2011) Retinoic acid influences neuronal migration from the ganglionic eminence to the cerebral cortex. J Neurochem 119:723–735 25. Hashimoto R, Nagaya M, Ishiguro Y et al (2002) Relationship of fistulas to the rectum and genitourinary tract in mouse fetuses with high anorectal malformations induced by all-trans retinoic acid. Pediatr Surg Int 18:723–727
123
