Ex Vivo Whole Embryonic Kidney Culture: A Novel Methodfor Research in Development, Regeneration and TransplantationStefano Giuliani,* Laura Perin,* Sargis Sedrakyan, Paul Kokorowski, Daniel Jinand Roger De Filippo†From the Division of Urology, Childrens Hospital Los Angeles and Saban Research Institute, Keck School of Medicine, University ofSouthern California (SG, LP, SS, PK, DJ, RDF), Los Angeles, California, and Department of Pediatric Surgery, University of Padova(SG), Padova, Italy

Purpose: Whole metanephric organ culture represents a novel investigatory approach with potential application in manyaspects of research in kidney regeneration and transplantation. We report the current status of embryonic kidney culture,discussing issues such as the appropriate culture conditions and methods, histological results, values of and limitations to thedifferent techniques used today. To optimize this system in vitro for the benefit of future studies we focused our efforts onevaluating and developing a new durable 3-dimensional organ culture system using a uniquely modified approach.Materials and Methods: Metanephric kidneys were microdissected from the embryos of timed pregnant WT C57BL/C6 miceon days 12 to 16 of gestation (embryonic days 12 to 16). Novel perfusion channels were created in the harvested embryonickidneys before placing them in culture. Embryonic kidneys were placed on a 0.4 �m pore size Transwell® membrane, culturedin base medium at a medium gas interphase and incubated at 37C with fully humidified 5% CO2. Histological andimmunocytochemical analysis was performed to evaluate for signs of necrosis, and the structural integrity and functionalityof organs during culture.Results: We confirmed histologically that our organ culture system was capable of maintaining normal kidney structuressignificantly longer (mean 10 days) than previously reported standard protocols. Condensation and aggregation of themetanephric mesenchyma at the tips of the ureteral bud were observed, including the formation of well developed nephronsand glomeruli without evidence of necrosis. Organ maturation occurred in a developmentally appropriate centrifugal patternand the expression of key regulatory factors was demonstrated.Conclusions: Our in vitro model replicates closely the in vivo processes involved in normal kidney development. We alsopresent what is to our knowledge the first demonstration of a durable 3-dimensional kidney culture system reported in theliterature. This system may represent an uncomplicated method for in vitro kidney culture that we hope will serve as aneffective adjunct to research focused on signaling pathways, development and regeneration as applied to the kidney.

Key Words: kidney, morphogenesis, embryo, tissue culture techniques, Rodentia

Mouse and rat embryonic kidneys are organs that arefrequently used in various modes of study, rangingfrom normal renal organogenesis to kidney engi-

neering and transplantation. Whole organ cultures allowinvestigators to treat developing kidneys with pharmacolog-ical reagents such as growth factors, enzymes or antibodiesto determine their effects on the entire organ during devel-opment. Starting in the mid 20th century various proce-dures for whole embryonic kidney culture have been re-ported in the literature but none show long-term resultswith histological data and no standardized techniques existthat can be easily replicated.

In normal renal development the metanephros starts toform when the metanephric blastema, derived from the cau-dal intermediate mesoderm, induces the UB to form a spuroff the mesonephric or wolffian duct.1 The UB makes contact

Submitted for publication March 16, 2007.* Equal study contribution.† Correspondence: Childrens Hospital of Los Angeles, 4650 Sunset

Blvd., Mailstop 114, Los Angeles, California 90041 (telephone: 323-669-2247; FAX: 323-906-8034; e-mail: rdefilippo@chla.usc.edu).

See Editorial on page 17 in this issue.

0022-5347/08/1791-0365/0THE JOURNAL OF UROLOGY®

Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION

365

with the MM at approximately E11 in mice and E13 in rats.The initial mesenchymal response to UB derived inductivesignals is the condensation of mesenchymal cells around thetip of the UB.2,3

Several key regulators are known to be essential for theinteraction between the UB and MM, including Pax-2 andGDNF. Pax-2 is initially expressed in the UB and then in theadjacent MM until expression is down-regulated as the MMdevelops into mature nephrons.4 GDNF is primarily expressedin the MM and it is considered essential for UB branching.5

As the UB proliferates outward along the radial axis ofthe kidney, the aggregate cells begin to show signs of epi-thelial polarity (the so-called renal vesicle).6,7 The renalvesicle progressively develops into comma-shaped and S-shaped bodies.1 The conversion of MM to tubular epithelium(the so-called mesenchymal-to-epithelial transformation) oc-curs to generate mature tubular structures in the kidney.These S-shaped bodies differentiate into the glomerulus,proximal convoluted tubule, loop of Henle, distal convolutedtubule and most likely the connecting segment, whereasderivates of the UB ultimately form the ureter, pelvic calices

and collecting ducts of the methanephros.1,8

Vol. 179, 365-370, January 2008Printed in U.S.A.

DOI:10.1016/j.juro.2007.08.092

EX VIVO WHOLE EMBRYONIC KIDNEY CULTURE366

The process of induction, morphogenesis and differentia-tion of the metanephros occurs in a centrifugal pattern,proceeding from center to periphery. In the nephrogeniczone the outermost cortical portion of the developing kidneycontains the youngest, most immature tubules, while theoldest, more mature nephrons reside in deeper regions of thekidney.9 Kidney tubule induction and ureteral morphogen-esis are regulated reciprocally.1 The process of nephron in-duction and differentiation continues repeatedly until thefull complement of nephrons is formed, which in rodentsconcludes about 1 week after birth.10,11,12

Literature based evidence supports that many early mor-phogenetic events in kidney development can be mimickedin vitro using organ culture methods. Kidney rudiments arecommonly dissected from a mouse or rat embryo on gesta-tional day 12 to 13 or E12 to E13, immediately after MMinduction with UB invasion. The expression of early markergenes and subsequent morphological changes in vitro aresimilar to in vivo patterns with some notable differences.In vitro cultured kidneys undergo only an initial burst of cellproliferation and they do not develop renal vasculature.They also do not show the characteristic ascending anddescending limbs of the loop of Henle.13 Despite these limi-tations early events of organogenesis have been modeledquite successfully in organ culture up to the S-shaped bodystage but then cultures cannot be maintained long enoughfor further development to occur (mean only 4 days).

We describe our techniques and protocols for embryonickidney culture that have successfully allowed us to followthe development of a well preserved organ up to 10 daysin vitro. We also present a brief review of the literature tobetter understand the facts and limitations of the organculture protocols that have been described to date.

MATERIALS AND METHODS

To ensure sterile culture conditions all solutions and equip-ment were sterile, and all tissue culture steps and animalsurgeries were done in a laminar flow hood.

Medium and SolutionsThe base medium used was Leibovitz’s L-15 (Gibco/Invitro-gen™). This solution is light sensitive and it requires darkstorage at 4C. Also used were 1 � PBS (pH 7.4), 1% penicil-lin/streptomycin and 2% FBS (Gibco/Invitrogen).

Tools for Organ DissectionThe microsurgery instruments used were straight tip scis-sors, No.5 Dumont forceps (Fine Science Tools, Foster City,California), glass and plastic pipettes (VWR, San Diego,California), 0.4 �m pore size Transwell membrane, a dis-secting microscope and ice.

Organ Dissection ProcedureMetanephric kidneys of embryos from timed pregnant WTC57BL/C6 mice at 12 to 16 days of gestation (E12 to E16)were harvested under sterile conditions using a dissectingmicroscope. Embryonic staging was verified using the crite-ria of Theiler.13 The embryos were placed in a 10 cm plasticPetri dish containing cold 1 � PBS. A midline laparotomywas made to the abdominal wall and the intra-abdominal

contents were eviscerated to expose the retroperitoneum

and kidneys. Organs were transferred in a sterile mannerinto a culture dish with sterile medium using plastic pi-pettes (fig. 1).

Perfusion channels were created in the embryonic kidneysunder direct microscopy with a 10 �m diameter glass needlepermitting homogeneous perfusion of medium throughout thewhole organ. Tissue preparation and dissection was done asrapidly as possible with care taken to keep the embryonicorgans in cold medium (4C) before placing them in culturewells. Developing kidneys were placed on a 0.4 �m pore sizeTranswell membrane and cultivated at the medium (liquid)-gas interface. Culture was maintained for 4 to 10 days in afully humidified 37C incubator with 5% CO2. Medium wasnot changed during the culture but it was replenished asneeded.

Fixation and StainingA pipette with medium was used to detach the organs from themembrane. The kidneys were then washed for 1 minute � 3with ice-cold 1 � PBS. The kidneys were fixed for 1 hour in4% paraformaldehyde in PBS. Only fresh paraformaldehydewas used (stored at 4C no longer than 1 week). Fixation wasfollowed by paraffin embedding, serial sectioning and stain-ing with hematoxylin and eosin (Sigma-Aldrich™).

For immunostaining paraffin slides were placed in VectorAntigen Retrieval solution (Vector Laboratories, Burlin-game, California) and chilled. Subsequently slides were pro-cessed first with Avidin/Biotin Block solution (Vector Labo-ratories), then with 20% hydrogen peroxide (Sigma-Aldrich)for 1 hour to block endogenous peroxidase and finally with20% goat serum in PBS block (Vector Laboratories) for 1hour. Sections were incubated overnight with anti-caspase-3primary antibody (Cell Signaling Technology®) followed by

FIG. 1. A, genitourinary tract of mouse embryo from which embry-onic kidneys were excised for ex vivo culture. B, ex vivo culture

system consisting of 24-well plate and insert filter (tm), whereembryonic kidneys (ek) are cultured.

EX VIVO WHOLE EMBRYONIC KIDNEY CULTURE 367

incubation with an anti-rabbit secondary antibody (VectorLaboratories).

For immunofluorescent staining sections were blockedwith Avidin/Biotin Block after deparaffinization, followed byprotein block for 30 minutes using the appropriate 5% goatserum in PBS. The slides were incubated in primary anti-body (Santa Cruz Biotechnology, Santa Cruz, California)overnight at 4C. Subsequently the slides were washed inPBS for 5 minutes � 3, incubated with secondary antibodyfor a half-hour at room temperature and washed in PBSbefore development with Fluorescein/Avidin DCS (VectorLaboratories). Finally, slides were mounted using 4=,6-diamidino-2-phenylindole mounting medium (Vector Labo-ratories).

RESULTS

Our protocol made possible what is to our knowledge thelongest reported culture time for embryonic kidneys (up to10 days) with the preservation of UB branching and finetubulogenesis in vitro. Histological examination of the spec-imens revealed initial distal UB ramification and condensa-tion of MM cells at the branching UB tips. No endothelialelements or primitive capillaries were present at this stageof embryonic gestation (E12 to E13) in the metanephricblastema. By 24 to 48 hours of in vitro culture nephrogenesishad proceeded through the early stages of normal tubulardevelopment in the explants (fig. 2).

Reciprocal inductive interactions between the 2 primor-dia (UB and MM) induced dichotomous branching of the UB,initiating morphogenesis of the collecting duct system. Con-densation and aggregation of the metanephric mesenchymaat the tips of the branching ureteral bud formed normalrenal vesicle structures. After primordial connection be-tween the MM and UB we noted that kidney growth pro-gressed toward the natural succession of primitive nephronstructures, comma-shaped and S-shaped body elements,with final maturation into end structures, including glomer-uli, a proximal and a distal convoluted tubule, and collectingduct canalization (fig. 3). Glomeruli were formed by a centralcore of well preserved, elongated epithelial cells and sur-rounded by a regular shaped capsule of squamous epithelialcells, also lacking in vascularization.

The degree of glomerular epithelial differentiation thatoccurs in our system is similar to what occurs in vivo.14,15

The lower and middle limbs of the S-shaped tubule persistedfor 72 to 96 hours of culture. The lower limb of the tubulesformed a double lumen hemisphere and the middle limbdifferentiated into discrete proximal tubules (fig. 3, A andB). The expression of GDNF, an essential inducer of UB, wasfound in E15 embryonic kidneys after 5 days of culture (fig.3, C). GDNF expression was localized to the MM adjacent tothe tip of the UB and immunofluorescence could be seen inUB cells, which express GDNF receptor. Immunofluores-cence staining for Pax-2 was noted on the branching UB andadjacent MM in an E15 embryonic kidney after 5 days ofculture (fig. 3, D).

The kidney culture environment remained intact and cellssurvived up to 10 days in culture without signs of necrosis. Aclear distinction between the cortex and medulla was main-tained, accompanied by good quality glomerular and tubular

structures without evidence of apoptosis (figs. 4 and 5).

DISCUSSION

It has long been recognized that kidney rudiments can becultured readily on membranes in vitro at a medium-gasinterphase, where they develop to a surprising extent.1

Mouse embryonic kidney explants are relatively easy togrow for several days in culture and they process throughthe early stages of nephrogenesis. For example, more than50 nephrons form from a mouse E11 kidney rudiment cul-tured for 4 days. These nephrons are well differentiated withglomerulus-like bodies (glomeruli without vascular endothe-lium) together with proximal and distal tubules plus themain segments of the secretory nephrons. These modelsallow the investigation of various developmental mecha-nisms, including critical mesenchyma derived signals suchas GDNF, in branching morphogenesis.3 The table lists theprotocols and highlights of similar studies in the literature.Although all of these studies were well designed, each issomewhat limited by a short duration in culture or lack of

FIG. 2. Embryonic kidney at E14 after 3 days of culture. Beyond 3days cultured kidneys maintain and continue normal nephrogen-esis. Renal primordia grow without necrosis and demonstrate nor-mal centrifugal pattern of development with mature glomeruli (mg)in periphery and primordial glomeruli (pg) in center of embryonickidney. Reduced from �10.

histological data.

EX VIVO WHOLE EMBRYONIC KIDNEY CULTURE368

In 1982 Avner et al introduced a new method for organo-typic, mouse metanephric whole organ culture.16 The novel-ties in their system included a new serum-free, completelycharacterized, hormonally supplemented medium and thedemonstration that metanephric development continuesin vitro beyond the early S-shaped tubule. They also pointedout that it was unnecessary to separate the metanephricblastema described in previous kidney culture protocols.Before their discovery whole organ metanephric cultureshad been used only incidentally as controls in transfilterculture experiments and none of the early studies were ableto show differentiation beyond the stage of early S-shapedtubule development. The fact that they did not use animalserum supplemented medium or tissue extracts permittedthem the unique ability to study normal renal hormonalontogeny as well as the interaction of various hormonal andbiochemical systems without the confounding effects of suchsubstances.

A milestone report that established the standard for mod-ern whole embryonic kidney culture was presented in 1987by Saxén.1 According to the original protocol whole kidneyrudiments dissected from 11 to 13-day-old embryos areplaced on a filter membrane and cultivated at the medium-

FIG. 3. A, embryonic kidney at E14 after 3 days of culture. Kidneydevelopment in culture was maintained with outstanding quality.Tip of branching UB (ubb) is recognized, connected with developingcollecting duct (cd). Renal vesicle (rv) will ultimately give rise tomature nephron. Reduced from �25. B, embryonic kidney at E14after 3 days of culture demonstrates different stages of kidneydevelopment. Branching UB is fused to collecting duct. Adjacent tocollecting duct is renal vesicle. Note mature glomerulus with Bow-man’s capsule (cb) and portion of proximal tubule (pt) connected toglomerulus. Reduced from �32. C, immunofluorescent staining forGDNF in E15 embryonic kidney after 5 days of culture. GDNF wasexpressed by MM (S-shape and C-shape) surrounding UB tip andalso by UB. Reduced from �10. D, immunoflorescent detection ofPax-2 in E15 embryonic kidney after 5 days of culture. Pax-2 wasexpressed in branching UB and in adjacent MM (mm). Reducedfrom �40.

gas interphase on a Trowell-type screen made of stainless

steel net. Cultures are kept in an incubator with 100%humidity in a mixture of 5% CO2 in air. Medium is changedonce if the culture is maintained for 5 to 6 days but onaverage it was not changed during the routine 3-day cultureprocess. The conclusion of this report was that cytodifferen-tiation of epithelial tubules advanced to a stage at which the3 segments of the nephron could be distinguished, includingglomerular podocytes, and the cells of distal and proximaltubules.

Since the original communication of Saxén,1 various mod-ified versions of the procedures for embryonic kidney culturehave been reported in the literature but no single standard-ized methodology has emerged, nor have any successfullong-term culture processes devoid of necrosis been re-

FIG. 4. A, embryonic kidney at E17 after 10 days of culture. Long-term culture of embryonic kidney provided distinction between me-dulla and cortex. Glomeruli (g) and tubular structures (t) were wellpreserved with time and moderate amount of apoptotic cells wasnoted. Reduced from �4. B, higher magnification reveals matureglomeruli (mg), complete proximal tubules and distal tubules (dt) atcenter of medulla. Reduced from �10. C and D, another example oflong-term culture (up to 10 days) of embryonic kidney at E16.Necrosis was present at only small portion of organ periphery, whileglomeruli and tubules were well preserved. pt, proximal tubules.Reduced from �5 (C) and �8 (D).

FIG. 5. Immunohistostaining for apoptosis. Observed pattern is

similar to that in normal renal development with more cells atperiphery undergoing programmed cell death. Reduced from �20.

EX VIVO WHOLE EMBRYONIC KIDNEY CULTURE 369

ported to our knowledge. In 1993 Hardman et al used forthe first time a biospore membrane coated with type Icollagen (Matrigel™) to culture kidneys.18 After 4 days ofculture they observed enhanced kidney growth and morpho-genesis. These results were likely unrelated to the use ofcollagen since there was little histological difference be-tween organs grown on ECM coated membranes, as opposedto controls grown without ECM. To our knowledge there areno reports that fully analyze histological patterns beyond 4days. On the other hand, a limited collection of extendedculture methods was reported. Rogers et al observed thatsurgically removed metanephric tissue from E13 rats couldgrow in vitro for up to 6 days.17 However, they did not focuson histological results but only pointed out that the meta-nephros increased in size by 50% and its surface changedfrom a smooth to a more convoluted appearance.17

All of this forced investigators to try new methods thatwould optimize culture efficiency. Gupta et al introduced anew modified protocol that introduced DMEM/F12 with 5%FBS as a better solution for culturing murine embryonickidneys, suggesting that defined medium was not essen-tial.19 There was no difference between changing mediumevery 24 hours or at all. They noticed that little growthoccurred in the vertical dimension and most kidney growthdeveloped along the plane of the culture membranes, repre-senting perhaps an adaptation by the organ for a moreefficient means of nutrient uptake from the culture medium.However, the observed growth was likely due to a combina-tion of factors, including an increase in cell number (hyper-plasia), a change in individual cell size (hypertrophy), ECMproduction, and a decrease in cell apoptosis and necrosis. Aprogressive increase in apoptosis was unfortunately appar-ent at later time points (72 to 96 hours). The diffuse patternof apoptosis observed differs from normal embryonic kidney

Whole organ embryonic kidney cu

ReferencesEmbryonic

Stage (animal) Culture Conditions

Avner et al16 E13.5 (mouse) 0.8 �m Filter, DMEM � 20% FBpenicillin-streptomycin � 50 U/m37C, 95% humidity, 5% CO2, meevery 24 hrs

Saxén1 E11–13 (mouse),E13–15 (rat)

Trowell-type stainless steel net, M10% FBS, I-MEM � 50 �g/ml irtransferrin, 37C, fully humidifiemedium replaced if culture long

Rogers et al17 E13 (rat) 0.8 �m Thickness � 13 mm diammembranes, DMEM/F12 � 25 mbuffer � sodium bicarbonate (1.nM Na2 � SeO3 � 5H2O � prostag(10�11 M) � iron saturated tranml) � gentamicin (50 �g/ml) � n(50,000 U/ml) � penicillin/streptU/ml), 37C, fully humidified 5%replaced every 24 hrs

Hardman et al18 E11.5 (mouse) Biopore membrane noncoated orECM component (collagen or MaNucleopore® filter, DMEM/F12amphotericin B � penicillin-strefully humidified 5% CO2, mediuevery 48 hrs

Gupta et al19 E13 (mouse) 0.45 �mol/l Polyethylene terephthmembranes, DMEM/F12 � 5% F37C, fully humidified 5% CO2, mchanged

development, in which the MM shows a higher rate of apop-

tosis than the UB. However, the investigators still focusedmost of their efforts on culture systems that preserved 3-dimensional structure and not on histological integrity.However, we believe that an alteration in the 3-dimensionalshape of the in vitro organ is less important than the avail-ability and maintenance of good quality embryonic kidneystructures. In fact, we expect these alterations in the uni-formity of organ in vitro.

From our perspective the real importance of the organculture is the maintenance of good cell quality, structure andfunction, rather than macroscopic shape preservation. Ourtechnique produces a good yield of developing cells and thegeneration of well shaped primordial structures, which arepreserved in culture longer than previously described withhistological and supporting evidence. This protocol allowsthe preservation not only of embryonic kidney morphology,but also of functionality. Important key regulators such asGDNF and Pax-2 are expressed as expected during culture,demonstrating appropriate in vitro development.4,5

Our efforts were focused on maintaining the expectedmicroscopic anatomy of primordial kidneys and limiting thenecrosis process that eventually occurs during ex vivo cul-ture. Several important considerations are emphasized. 1)Accurate dissection of the embryonic kidney must be donewith complete removal of the renal capsule, that is the thinconnective tissue surrounding the primordia. 2) Perfusionchannels are important for maintaining satisfactory deliv-ery of medium to the organ core, which is at most risk fornecrosis. 3) A low FBS concentration of 2% also assists inmaintaining better tissue viability.

We noted that using our protocols embryonic kidneyswere well preserved in culture and they continued withnormal nephrogenesis, growing devoid of necrosis and withpreserved expression of GDNF and Pax-2, which are essen-

methods described in literature

No.Culture

Days Conclusion

00 �g/mlcostatin,replaced

5 Organ system maintained normal structuralrelationships of epithelial � mesenchymalelements, � underwent organotypicdifferentiation in vitro

� 5%–dedCO2,n 3 days

7 Development of mesenchymal cells cultivated in3-dimensional organotypic explants closelyresembled that in vivo

pesml) � 10

E1ng (5 �g/inin (50medium

6 Growth in size � morphological complexity,metanephros long axis increased by 50% in 4days � metanephros surface changed fromsmooth to more convoluted appearance, littleadditional change in gross morphology between4 � 6 culture days

withl),

FBS �ycin, 37C,laced

4 Biospore membrane was excellent substratumfor mouse embryo organ culture, ECM coatingenhanced kidney growth � morphogenesis, �allowed easy visualization of these processes inwhole organ

others,never

4 DMEM/F12 � 5% FBS was best medium, planarsurface area measurements described kidneymorphogenesis because they correlated withkidney vol � UB branching, significantapoptosis at 96 hrs of culture.

lture

S � 1l my

dium

EMon load 5%er thaeterM He

1 mg/landinsferriystatomycCO2,

coatedtrige

� 10%ptomm rep

alateBS �edium

tial for UB branching and normal interactions between the 2

EX VIVO WHOLE EMBRYONIC KIDNEY CULTURE370

embryonic tissues. The centrifugal pattern of glomeruli de-velopment is maintained through to maturity at the periph-ery with primitive glomeruli at the center of the embryonickidneys. The tips of the branching UB also connect with thecollecting ducts, as normally observed in vivo (figs. 2 and 3).However, the more interesting features are found when wefollow the cultures more than 6 days. The distinction be-tween the medulla and cortex is pronounced (fig. 4). Matureglomeruli are well preserved, and proximal and distal tubu-lar structures can be seen elongating to define the medullaryregion. We noted early signs of necrosis beyond 10 days ofculture but this was limited to only a small portion of kidneyparenchyma. Therefore, the majority of cells and structuresare still intact and useful for processing and analysis.

CONCLUSIONS

There are many protocols described in the literature outlin-ing commonly applied techniques for whole embryonic kid-ney culture in vitro. However, embryonic organs are fragileby nature and the majority of these culture systems report amaximum of only 5 to 6 days of successful culture before theonset of diffuse necrosis, disrupting normal embryonic de-velopment and anatomy.

We describe a novel method for long-term embryonic kid-ney culture in which the onset of diffuse necrosis is delayed,while the morphology, histology and functionality of thekidney are preserved. Such protocols may prove useful in awide variety of investigations, such as defining developmen-tal pathways or tissue engineering applications. Long-termculture methods may be particularly useful in stem cell andregenerative research, which often requires a durable sys-tem that allows enough time to pass for differentiation ofpluripotential cells to occur.

Abbreviations and Acronyms

DMEM � Dulbecco’s MEME � embryonic day

ECM � extracellular matrixFBS � fetal bovine serum

GDNF � glial cell line-derived neurotrophic factorM � metanephric mesenchyma

MEM � modified Eagle’s mediumUB � ureteral bud

REFERENCES

1. Saxén L: Organogenesis of the kidney. In: Development and

Cell Biology Series 19. Edited by PW Barlow, PB Green

and CC White. Cambridge, United Kingdom: CambridgeUniversity Press 1987.

2. Grobstein C: Inductive epithelio-mesenchymal interactionin cultured organ rudiments of the mouse. Science 1953;118: 52.

3. Grobstein C: Transfilter induction of tubules in mouse meta-nephrogenic mesenchyme. Exp Cell Res 1956; 10: 424.

4. Dressler GR and Woolf AS: Pax-2 in development and renaldisease. Int J Dev Biol 1999; 43: 463.

5. Sariola H and Saarma M: GDNF and its receptors in theregulation of ureteric branching. Int J Dev Biol 1999; 43:413.

6. Barasch J, Pressler L, Connor J and Malik A: A ureteric budcell line induces nephrogenesis in two steps by two distinctsignals. Am J Physiol 1996; 271: F50.

7. Karavanova ID, Dove LF, Resau JH and Perantoni AO: Con-ditioned medium from a rat ureteric bud cell line in combi-nation with bFGF induces complete differentiation of iso-lated metanephric mesenchyme. Development 1996; 122:4159.

8. Barasch J, Yang J, Ware CB, Taga T, Yoshida K, Erdjument-Bromage H et al: Mesenchymal to epithelial conversion inrat metanephros is induced by LIF. Cell 1999; 99: 377.

9. Potter EL: Glomerular development in the kidney as an indexof foetal maturity. J Pediatr 1943; 22: 695.

10. Davies J and Bard J: Inductive interactions between the mes-enchyme and the ureteric bud. Exp Nephrol 1996; 4: 77.

11. Larsson L, Aperia A and Wilton P: Effect of normal devel-opment on compensatory renal growth. Kidney Int 1980;18: 29.

12. Solomon S: Developmental changes in nephron number, prox-imal tubular length and superficial nephron glomerularfiltration rate of rats. J Physiol (Lond) 1877; 272: 573.

13. Theiler K: The House Mouse: Atlas of Embryonic Develop-ment. New York, Springer-Verlag 1989.

14. Potter E: Development of the human glomerulus. Arch Pathol1965; 80: 241.

15. Kazimierczak J: Development of the renal corpuscle and thejuxtaglomerular apparatus: a light and electron micro-scopic study. Acta Pathol Microbiol Scan (A), suppl., 1971;218: 1.

16. Avner ED, Ellis D, Temple T and Jaffe R: Metanephric devel-opment in serum-free organ culture. In Vitro 1982; 18: 675.

17. Rogers SA, Ryan G and Hammermen MR: Insulin-like growthfactor I and II are produced in the metanephros and arerequired for growth and development in vitro. J Cell Biol1991; 113: 1447.

18. Hardman P, Klement BJ and Spooner BS: Growth and mor-phogenesis of embryonic mouse organs on non-coated andextracellular matrix-coated Biopore membrane. DevGrowth Differ 1993; 35: 683.

19. Gupta IR, Lapointe M and Yu OR: Morphogenesis duringmouse embryonic kidney explant culture. Kidney Int 2003;

63: 365.