Characterization of fetal ovine renal dysplasia after mid-gestation ureteral obstruction

Douglas G. Matsell,*~ MD, CM
Tracey Bennett*~
Alan D. Bocking,*#§ MD

Clin Invest Med 1996; 19 (6): 444-52.

[résumé]

From *the Departments of Paediatrics and #Obstetrics and Gynecology, ~Child Health Research Institute, §Lawson Research Institute, University of Western Ontario, London, Ont.

(Original manuscript submitted Feb. 6, 1996; received in revised form May 3, 1996; accepted June 10, 1996)

Paper reprints may be obtained from: Dr. Douglas G. Matsell, 2230­800 Commissioner's Rd. E, Children's Hospital of Western Ontario, London ON N6C 2V5; fax 519 685-8156; dmatsell@julian.uwo.ca

Contents

Abstract

 Objective: The etiology and pathogenesis of renal dysplasia are poorly understood. To characterize the histologic changes in fetal renal dysplasia, we studied a fetal ovine model of urinary obstruction.

Design: Animal study.

Animals: Seven fetal lambs, and other lambs of the same gestational age as controls.

Interventions: Unilateral ureteral ligation on fetal lambs at approximately 70 days' gestation (term for sheep is 145 days), during nephrogenesis. Kidneys were subsequently collected, examined histologically and characterized by immunohistochemical tests involving cytokeratin antiserum and a monoclonal antibody to alpha-actin.

Outcome measures: Histologic changes in ligated fetal lamb kidneys, based on comparison with normal fetal lamb kidneys.

Results: At near term (140 days' gestation), the ligated kidney showed distorted and less abundant renal parenchyma than a normal control kidney. Upon microscopic examination, the ligated kidney displayed marked architectural distortion of the outer cortex, with abundant interstitial fibrosis, primitive ductules and glomeruli, and cysts of varying sizes lined by squamous and cuboidal epithelia and surrounded by a loose mesenchyme. The renal medulla contained differentiated collecting ducts, which were structurally distorted and less abundant than in normal control kidneys. The proximal and distal tubule elements were primitive and markedly underdeveloped. Cytokeratin immunoreactivity was present in the collecting duct epithelium and in the cuboidal epithelium lining many of the cortical cysts. Smooth muscle alpha-actin immunoreactivity was localized in the cortical region of the kidney, which highlighted the abundance and disorganization of the undifferentiated mesenchyme and identified the fibromuscular collars of the primitive ductules of the cortex and the distorted collecting ducts of the medulla.

Conclusions: These results highlight the histologic changes resulting from unilateral ureteral ligation in fetal lambs. This model is useful in the study of the pathogenesis of fetal obstructive renal dysplasia.

Résumé

 Objectif : L'étiologie et la pathogénèse de la dysplasie rénale sont méconnues. Nous avons étudié un modèle ovin foetal d'obstruction urinaire afin de caractériser les changements histologiques survenant dans la dysplasie rénale foetale.

Devis : Étude animale.

Sujets : Sept foetus d'agneaux; d'autres agneaux du même âge gestationnel servaient de témoins.

Interventions : Une ligature urétérale unilatérale fut pratiquée à environ 70 jours de gestation durant la néphrogénèse (la durée de la gestation est de 145 jours chez le mouton). Par la suite, les reins furent prélevés et examinés histologiquement. Un examen immuno-histochimique fut aussi pratiqué avec un anticorps polyclonal anti-cytokératine ainsi qu'un anticorps monoclonal anti-a-actine.

Variables mesurées : Changements histologiques au niveau des reins ligaturés en comparaison avec les reins normaux (non ligaturés).

Résultats : À 140 jours de gestation, les reins ligaturés démontraient un parenchyme rénal déformé et moins abondant que les reins-témoins. L'examen microscopique révélait dans le rein ligaturé une distorsion marquée de l'architecture corticale avec fibrose interstitielle abondante, canaux et glomérules primitifs, de même que la présence de kystes de dimensions variables tapissés d'un épithélium pavimenteux et cubique et entourés d'un mésenchyme lâche. La partie médullaire du rein contenait des tubes collecteurs différenciés à l'architecture déformée et moins abondants que les reins-témoins. Les composantes des tubules proximaux et distaux étaient primitives et très sous-développées. La cytokératine était présente dans l'épithélium des tubes collecteurs et dans l'épithélium cuboïde tapissant de nombreux kystes corticaux. L'anticorps monoclonal démontra la présence d'a-actine du muscle lisse dans la région rénale corticale, soulignant l'abondance et la désorganisation du mésenchyme indifférencié et permettant l'identification des collets fibro-musculaires des canaux corticaux primitifs et les tubes collecteurs médullaires déformés.

Conclusion : Ces résultats mettent en relief les changements histologiques résultant de la ligature urétérale unilatérale chez le foetus d'agneau. Ce modèle est utile dans l'étude de la pathogénèse de la dysplasie rénale foetale d'origine obstructive.

[Table of contents]

Introduction

 Renal dysplasia is one of the most common causes of end-stage renal disease in children.[1] Several forms of renal dysplasia have been described, all of which are characterized by renal architectural disorganization, cystic changes and primitive ductules and glomeruli.[2] The precise etiology of the various forms of renal dysplasia is unknown. Possible causes include abnormal induction or differentiation of the metanephric blastema and injury to or obstruction of urine flow during critical stages of nephrogenesis.[3,4] In addition, a role for the insulin-like growth factor (IGF) system has been described in the progressive histopathologic changes that occur in human multicystic renal dysplasia from before to after birth.[5]

The sheep model of fetal unilateral ureteral ligation allows us to study renal dysplasia during normal mid- and late-gestation nephrogenesis, a period which cannot be studied in humans.[6] The fetal sheep kidney accurately represents human nephrogenesis: active induction of the undifferentiated metanephric blastema occurs as late as 70 days' gestation in fetal lambs, equivalent to approximately 20 weeks' gestation in human fetuses. The fetal sheep model also provides a system in which normal nephrogenesis can be altered and the effects of these alterations studied through to the development of dysplasia. In addition, extensive literature demonstrates that fetal sheep are a representative model for normal and abnormal fetal growth and development.[7,8]

We demonstrated the histopathologic changes that occur in the kidney as a result of in utero ureteral obstruction and, through the use of immunohistochemical techniques, highlighted the marked cystic, dysplastic and interstitial changes.

[Table of contents]

Methods

Sheep experiments

Unilateral ureteral obstruction was performed in seven fetal sheep at 70 to 76 days' gestation. The obstructed kidneys were then studied near term (at 140 days' gestation). Experiments were performed in accordance with the Animal Care Council at the University of Western Ontario, London, Ont. The presence of a fetus was confirmed in the pregnant ewe with real-time ultrasonography before surgery. The ewe was anesthetized with intravenous thiopentone, a tube was placed, and 0.5% to 1.5% halothane was administered with oxygen to maintain anesthesia. Before surgery, an antibiotic was injected into the ewe at a dose of 2 mL/45 kg body weight. A midline incision was made along the anterior wall of the ewe's abdomen. The uterus was incised over the fetus' hindquarters, which were then advanced through the incision. After being localized by gentle palpation, the fetal kidney was exposed through a flank incision. Both the kidney and ureter were carefully isolated by blunt dissection. The ureter was then ligated with a suture placed close to the ureteropelvic junction. After closure of the fetal wound, the fetus was returned to the uterus. The uterine incision, peritoneum, fascia and skin were closed. After surgery was completed, the halothane was discontinued and the ewe allowed to recover. The animal was then placed in a metabolic cage and the tube removed when the animal began making spontaneous respiratory actions. Intramuscular banamine was administered immediately after the operation, and food and water were given when the animal was able to stand. After 2 to 3 days of recovery in the animal quarters, the ewe was returned to the farm. At an estimated gestation of 140 days, the pregnant ewe was returned, killed and the fetal sheep rapidly delivered by cesarean section. The newborn lamb was then killed.

Tissue preparation

Kidneys were removed and sectioned. One half of the obstructed kidney was immediately fixed in a 4% paraformaldehyde and 0.2% gluteraldehyde solution in 70 mmol/L phosphate buffer. The tissue was then washed in phosphate-buffered saline (PBS) solution and 70% ethanol, then embedded in paraffin. The blocks were sectioned to 5 µm and stained by standard techniques with hematoxylin and eosin. The unobstructed contralateral kidney was similarly processed for future studies.

Immunohistochemical studies

Immunohistochemical studies were performed as described elsewhere.[9] Tissue sections were deparaffinized in xylene, rehydrated and washed in PBS solution. Endogenous peroxidase activity was quenched by incubating tissue sections in 1% hydrogen peroxide (in PBS) for 10 minutes. Sections were next incubated in 10% normal goat serum and then incubated with specific primary antisera overnight at 4°C. The primary antibodies used included a rabbit polyclonal antiserum against cytokeratin (1:1000) (Dako, Dimension Labs, Mississauga, Ont.), which recognizes a wide range of differentiated epithelia, including Bowman's capsule and proximal and distal tubule epithelia in the kidney,[10] and a mouse monoclonal antibody to smooth muscle alpha-actin (Sigma Chemicals, St. Louis, Mo.). Sections were then washed in PBS, and positive immunoreactivity (IR) was identified by the avidin­biotin­peroxidase technique using the Vectastain kit (Vector Laboratories, Burlingame, Calif.) with diaminobenzidine as the chromagen. Tissue sections were then counterstained with Carazzi's hematoxylin, dehydrated and mounted with Permount (Fisher Scientific, Fair Lawn, NJ).

[Table of contents]

Results

Sheep experiments

We performed unilateral ureteral obstruction in seven fetal lambs at 70 to 76 days' gestation. Only two animals were carried successfully to 140 days' gestation. Two experiments resulted in fetal death in utero, and three others resulted in premature delivery at the farm and subsequent death of the premature lamb. The kidneys of the two animals carried to term were collected and compared with kidneys of animals that were not surgically manipulated and were of the same gestational age. Two of the normal fetal sheep controls were twins of the surgically manipulated animals. For kidney weights, data were taken from three other normal sheep that had attained 140 days' gestation. These data were obtained from the protocols of other investigators at the Lawson Research Institute, London, Ont.

Both kidneys with ligated ureters had marked dilatation of the proximal ureter and renal pelvis. The kidneys appeared misshapen, with a pale capsule and a thin nodular rim of pale cortex. The contralateral kidney in both cases was larger than normal kidneys of fetal sheep of similar gestational age (24.0 g, standard deviation 2.8 g, in the two kidneys of experimental animals, v. 14.7 g, standard deviation 1.6 g, in the eight kidneys of control animals), a finding that suggests hypertrophy. The surgically manipulated fetal sheep appeared to have developed normally in all other respects. In addition, the amniotic fluid volume appeared qualitatively normal at postmortem.

Renal histologic findings

Nephrogenesis was under way in the normal kidneys of fetal sheep of 70 days' gestation (Fig. 1). At this stage of development, the kidney had a well-demarcated nephrogenic zone with identifiable metanephric blastema cells, ureteric duct, S-shape nephrons and maturing glomeruli. Contact and communication between the ureteric duct epithelium and the undifferentiated metanephric blastema cells resulted in inductive differentiation of the blastema into the varied epithelia of the nephron.

At 140 days' gestation, which approximates term, the normal fetal sheep kidney was fully developed (Fig. 2, A). The nephrogenic zone had completely differentiated, and the metanephric blastema cells, inducing ureteric duct and S-shape nephrons, were no longer identifiable. Maturing glomeruli, proximal and distal tubules, loop of Henle and collecting duct were present. In addition, even under low magnification, the cortex and medulla of the kidney had become distinct. Within the medulla there was a well-developed papilla, collecting duct system and vasa recta (Fig. 2, B).

In the obstructed kidneys, the gross reniform shape was maintained; however, there was marked parenchymal and architectural disorganization. The outer cortex was disrupted by cysts and an overly abundant mesenchyme (Fig. 2, C). Although the exact origin of the cysts was not ascertained, there appeared to be at least two types of cysts. Subcapsular cysts, lined by a flattened cuboidal epithelium and displaying strong cytokeratin immunoreactivity, were adjacent to and likely derived from the collecting duct. In addition, cystic glomeruli were identified. Specific segments of the renal cortex showed classic cystic dysplasia, with dilated primitive ductules surrounded by fibromuscular collars, primitive glomeruli and abundant disorganized interstitial mesenchyme. As in obstructive renal dysplasia in humans, the medulla of the obstructed kidney of a fetal sheep at term was remarkably underdeveloped, with rudimentary medullary pyramids, few and less developed ducts, loops of Henle and vasa recta, and an excess of connective tissue (Fig. 2, D). There was no identifiable ectopic cartilage.

Immunohistochemical studies

Smooth muscle alpha-actin

In the kidney of normal fetal sheep of 140 days' gestation, smooth muscle alpha-actin immunoreactivity (SMA-IR) was present in the renal capsule, in the smooth muscle walls of the intrarenal vessels and in the mesangial areas of the maturing glomeruli (Fig. 3, A). SMA-IR was absent from the normal medulla (Fig. 3, B). In kidneys of normal human fetuses of correspondingly younger gestational age, SMA-IR also present in the undifferentiated metanephric blastema and renal mesenchyme.[8,11] In the obstructed kidneys of fetuses of 140 days' gestation there was marked distortion in the distribution of SMA-IR (Fig. 3, C). SMA-IR appeared in the abnormally distributed vascular smooth muscle and, more dramatically, in the fibromuscular collars of the cysts and dilated collecting ducts, in the cortical and the medullary regions (Fig. 3, D).

Cytokeratin

In the kidney from a normal fetal sheep of 140 days' gestation, cytokeratin immunoreactivity (CK-IR) was localized most prominently to the differentiated epithelium of the collecting duct, with less intense staining in Bowman's capsule, proximal tubule and loop of Henle (Figs. 4, A and B). In the obstructed kidney, there was intense CK-IR in the epithelia of most but not all cysts (Fig. 4, C). The epithelia of the cysts derived from Bowman's capsule lacked CK-IR. In the medulla of the obstructed kidney, the underdeveloped collecting system was noted, and CK-IR was also present in the collecting duct and loop of Henle epithelia (Fig. 4, D).

[Table of contents]

Discussion

In this study we characterized the histologic and immunohistochemical features of renal dysplasia in the fetal sheep model of mid-gestation ureteral obstruction. Histologic examination showed that the obstructed kidneys develop cortical cysts in the nephrogenic zone; these cysts appear to be derived from dilated collecting ducts and Bowman's space. In humans, obstructive renal dysplasia most commonly results from bladder outlet obstruction, as seen in children with posterior urethral valves. The degree of dysplasia is related to the severity of obstruction, and, therefore, the severity of disease varies widely.

Our model likely does not fully represent the most severe forms of human obstructive renal dysplasia or multicystic renal dysplasia, given that the obstruction began mid-gestation and that consequent parenchymal disruption was patchy. Yet many of the changes we observed, including cystic dilatation of the tubules, pericystic fibromuscular reaction, abundant and disorganized interstitial connective tissue and underdevelopment of the renal medulla, are characteristics of both of the disorders seen in humans, although more typical of obstructive renal dysplasia.[12,13]

Normal kidney development in humans begins at approximately 6 weeks' gestation and continues to 36 weeks' gestation; thereafter, hypertrophy and hyperplasia of existing nephrons account for renal growth.[14] Renal development involves the induction of the undifferentiated metanephric ridge by the migrating ureteric duct. This induction represents a complex, intricate and poorly understood interaction between a differentiated epithelium and an undifferentiated mesenchyme.[15,16] In transgenic models of mouse kidney development, an alteration in the genes responsible for the branching morphogenesis of the inducing ureteric duct results in significant renal abnormalities, including changes similar to cystic renal dysplasia.[17,18]

Classic theories concerning the etiology of human multicystic renal dysplasia have included obstruction of urinary flow early in renal development.[19] This theory has been supported by the clinical observation of associated urinary tract abnormalities in children with cystic renal dysplasia, including ureteric atresia and bladder outflow obstruction.[20] Apart from the model of ureteric obstruction in fetal sheep, no animal model has successfully reproduced the elements of renal dysplasia. The success we had with fetal sheep is likely due to their relatively long gestation and their extended period of renal development. Ureteric obstruction in later fetal gestation fails to create renal dysplasia and results instead in simple hydronephrosis.[21]

This model of fetal urinary obstruction has been used to study in utero intervention in fetuses with hydronephrosis.[6] In this study we have defined the marked histopathologic changes that occur as a result of obstructive injury. If the pathogenesis of obstructive cystic renal dysplasia involves an insult early in nephrogenesis, then early relief of the obstruction is needed to prevent these irreversible histopathologic changes. However, timely surgical repair will remain a practical limitation of in utero intervention. In addition, it will be important to define the factors affecting the significant cystic epithelial changes as well as the marked interstitial reaction we found. This model should provide an important means to study the pathogenesis of renal dysplasia. It may also suggest alternative in utero interventions that make the best use of existing functioning kidney tissue and prevent further interstitial and mesenchymal disruption.

[Table of contents]

Acknowledgements

This work was supported by grants from the Kidney Foundation of Canada, St. Joseph's Hospital Research Trust Fund, (D.G.M.) and the Medical Research Council of Canada (A.D.B.).

[Table of contents]

References

  1. McEnery PT, Stablein DM, Arbus G, Tejani A. Renal transplantation in children: a report of the North American Pediatric Renal Transplant Cooperative Study. N Engl J Med 1992;326:1727-32.
  2. Bernstein J, Gilbert-Barness E. Congenital malformations of the kidney. In: Tisher CC, Brenner BM, editors. Renal pathology. Philadelphia: JB Lippincott, 1989:1278-308.
  3. Schwarz RD, Stephens FD, Cussen LJ. The pathogenesis of renal dysplasia I. Quantification of hypoplasia and dysplasia. II. The significance of lateral and medial ectopy of the ureteric orifice. Invest Urol 1981;19:94-6.
  4. Mackie GG, Stephens FD. Duplex kidneys: a correlation of renal dysplasia with position of the ureteral orifice. J Urol 1975;114:274-80.
  5. Matsell DG, Bennett T, Armstrong RA, Goodyer P, Goodyer C, Han VKM. Insulin-like growth factor (IGF) and IGF binding protein gene expression in multicystic renal dysplasia. J Am Soc Nephrol In press.
  6. Glick PL, Harrison MR, Adzick NS. Correction of congenital hydronephrosis in utero. IV. In utero decompression prevents renal dysplasia. J Pediatr Surg 1984;19:649-57.
  7. Robillard JE, Smith FG, Nakamura KT, Sato K, Segar J, Jose PA. Neural control of renal hemodynamics and function during development. Pediatr Nephrol 1990;4:436-41.
  8. Iwamoto HS, Oh W, Rudolph AM. Renal metabolism in fetal and newborn sheep. In: Jones CT, Nathaniels PW, editors. The physiological development of the fetus and newborn. Orlando (FL): Academic Press, 1985:37-57.
  9. Matsell DG, Delhanty PJD, Stepaniuk O, Goodyer C, Han VKM. Expression of insulin-like growth factor and binding protein genes during nephrogenesis. Kidney Int 1994;46:1031-42.
  10. Pinkus GS, O'Connor EM, Etheridge CL, Corson JM. Optimal immunoreactivity of keratin proteins in formalin-fixed, paraffin-embedded tissue requires preliminary trypsinization. An immunoperoxidase study of various tumours using polyclonal and monoclonal antibodies. J Histochem Cytochem 1985;33:465-73.
  11. Alper CE, Seifert RA, Hudkins KL, Johnson RJ, Bowen-Pope DF. Developmental patterns of PDGF B-chain, PDGF-receptor, and alpha actin expression in human glomerulogenesis. Kidney Int 1992;42:390-9.
  12. Taxy JB. Renal dysplasia: a review. Pathol Annu 1985;20(pt 2):139-59.
  13. Ericson N, Ivenmark B. Renal dysplasia and pyelonephritis in infants and children. Arch Pathol Lab Med 1958;66:255-69.
  14. Ekblom P. Embryology and prenatal development. In: Holliday MA, Barratt TM, Avner ED, editors. Pediatric nephrology. 3rd ed. Baltimore: Williams and Wilkins, 1994:2-20.
  15. Saxen L. Organogenesis of the kidney. New York: Cambridge University Press, 1987.
  16. Grobstein C. Trans-filter induction of tubules in mouse metanephrogenic mesenchyme. Exp Cell Res 1956;10:424-40.
  17. Schuchardt A, D'Agati V, Larsson-Blomberg L, Constantini F, Pachnis V. Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret. Nature 1994;367:380.
  18. Dressler GR, Wilkinson JE, Rothenpieler UW, Patterson LT, Williams-Simons L, Westphal H. Deregulation of Pax-2 expression in transgenic mice generates severe kidney abnormalities. Nature 1993;362:65-7.
  19. Fouser L, Avner ED. Normal and abnormal nephrogenesis. Am J Kidney Dis 1993;21:664-770.
  20. Rubenstein M, Meyer R, Bernstein J. Congenital abnormalities of the urinary system. I. A postmortem survey of developmental abnormalities and acquired congenital lesions in a children's hospital. J Pediatr 1961;58:356-66.
  21. Harrison MR, Nakayama DK, Noall R, deLorimier AA. Correction of congenital hydronephrosis in utero. II. Compression reverses the effects of obstruction on the fetal lung and urinary tract. J Pediatr Surg 1982;17:965-74.
| CIM: December 1996 / MCE: décembre 1996 |
CMA Webspinners / >