The molecular basis of embryonic kidney development

The molecular basis of embryonic kidney development

Mechanisms of Development 62 (1997) 105–120 Review article The molecular basis of embryonic kidney development Mark S. Lechner, Gregory R. Dressler*...

5MB Sizes 0 Downloads 50 Views

Mechanisms of Development 62 (1997) 105–120

Review article

The molecular basis of embryonic kidney development Mark S. Lechner, Gregory R. Dressler* Howard Hughes Medical Institute and Department of Pathology, University of Michigan, Ann Arbor, MI 48109-0650, USA Received 20 January 1997; accepted 25 January 1997

Abstract The development of the mature mammalian kidney begins with the invasion of metanephric mesenchyme by ureteric bud. Mesenchymal cells near the bud become induced and convert to an epithelium which goes on to generate the functional filtering unit of the kidney, the nephron. The collecting duct system is elaborated by the branching ureter, the growth of which is dependent upon signals from the metanephric mesenchyme. The process of reciprocal induction between ureter and mesenchyme is repeated many times over during development and is the key step in generating the overall architecture of the kidney. Genetic studies in mice have allowed researchers to begin to unravel the molecular signals that govern these early events. These experiments have revealed that a number of essential gene products are required for distinct steps in kidney organogenesis. Here we review and summarize the developmental role played by some of these molecules, especially certain transcription factors and growth factors and their receptors. Although the factors involved are far from completely known a rough framework of a molecular cascade which governs embryonic kidney development is beginning to emerge.  1997 Elsevier Science Ireland Ltd. Keywords: Kidney organogenesis; Metanephros; Induction

1. Introduction In the mammalian embryo, the metanephric or adult kidney is derived from the reciprocal interactions of two primordial mesodermal derivatives, the ureteric bud and the metanephric mesenchyme. Upon induction by the ureteric bud, the metanephric mesenchyme undergoes a series of morphogenetic events that convert the mesenchyme to an epithelium and eventually generate most of the tubules of the mature nephron. In turn the mesenchymal tissue is required for continued growth and branching of the ureteric bud epithelium which connects to the maturing nephric tubules and gives rise to the renal collecting duct system. Many of these early events were elucidated first by manipulating amphibian and avian embryos and subsequently by utilizing a mammalian in vitro organ culture system (see Saxen, 1987 and references therein). Both methodologies * Corresponding author. HHMI, MSRB 1, Room 4510, 1150 W. Medical Center Drive, Ann Arbor, MI 48109-0650, USA. Tel.: +1 313 7646490; fax: +1 313 7636640; e-mail: [email protected]

0925-4773/97/$17.00  1997 Elsevier Science Ireland Ltd. All rights reserved PII S0925-4773 (97 )0 0667-9

have proven to be fundamental methods for studying embryonic induction. The development of many other tubular epithelial organs, such as lung, salivary glands, and gonads, also require epithelial-mesenchymal cell interactions for the controlled differentiation and proliferation of both cell types. Thus, nephrogenesis has often served as a model for organogenesis in general. The advent of molecular biological techniques has identified a large and growing number of gene products produced during kidney development. A list of these can now be found in a convenient database available on line (Davies and Brandli, 1994). Although the list of genes expressed in the kidney has provided a more detailed set of markers to follow development, mouse mutants generated by homologous recombination in embryonic stem cells have provided most of our new insights into the classic models of kidney development. Current studies using combinations of these techniques – expression analyses, gene knockout and organ culture – are providing the most complete information and will likely continue to do so. In this article, we concentrate on the most recent findings that have impacted our understanding

106

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

of kidney development since the publication of several comprehensive reviews (Bard et al., 1994; Patterson and Dressler, 1994; Davies and Bard, 1996; Sariola, 1996). In so doing, we focus more on the control of organogenesis and less on the execution of cellular differentiation. It should also be noted that for the sake of brevity we limit our discussion of the various genes and gene products to their specific role in early kidney development although many have multiple functions elsewhere in the embryo.

2. Morphogenesis and cell lineage In order to discuss the results of current experiments it is necessary to first describe kidney development in somewhat more detail. And although some work has been done in chick and amphibians most studies of nephrogenesis have been conducted in mice and rats. Thus, we will concentrate our description on the murine system and note exceptions where pertinent. The Wolffian or nephric duct is formed in a rostro-caudal direction in the intermediate mesoderm of the developing mouse embryo (see Fig. 1). As the duct elongates it induces a series of tubules in the adjacent mesoderm. The first two series of inductive events, which take place between embryonic day 8 and 10 (E8 and E10), produce the pronephros and mesonephros which are transient structures in mice and humans. In mice, pronephric tubules are rudimentary and exist for only a very short period of time, whereas mesonephric tubules range from simple epithelium more rostrally to longer, convoluted tubules complete with glomeruli, more caudally. Around E10.5 the ureteric bud evaginates from the caudal end of the Wolffian duct and invades the blastema of the metanephric mesenchyme (see Fig. 2), which at this time can be distinguished morphologically from surrounding mesenchyme. During the induction process, mesenchymal cells condense around the tips of the growing bud, then are split into two aggregates as the bud begins to branch. The induced mesenchymal aggregates undergo a burst of proliferation and convert to an epithelial cell type. These early epithelial cells form a spherical cyst called the renal vesicle. Recent analysis of cadherin gene expression indicates that the renal vesicle is already patterned such that cells adjacent to the ureteric bud express the adhesion molecule E-cadherin and cells more distal from the bud express K-cadherin (Fig. 3) (G.R. Dressler, unpublished observations). The renal vesicle then undergoes a series of invaginations and elongations to generate the comma-shaped and then the S-shaped bodies. Molecular markers can now distinguish populations of precursor cells within the S-shaped body indicating a segmental organization along the linear, or proximal-distal, axis of the tubular epithelium. Cells at one end express E-cadherin and connect to the adjacent ureteric bud epithelium to form the distal tubule and collecting duct, whereas the cells at the more distal end express the transcription factor WT1 (Pritchard-Jones et al., 1990; Armstrong et al., 1992) and

generate the glomerular epithelium. Between these two extremes are the proximal tubule progenitor cells that express the adhesion molecule K-cadherin. Mesenchymal cells that fail to generate epithelium express the transcription factor BF-2 (Hatini et al., 1996) and are thought to give rise to interstitial mesenchyme or undergo programmed cell death. Concurrent with these events, the most peripheral mesenchyme, that remains undifferentiated and perhaps uninduced, expresses the secreted peptide GDNF (Hellmich et al., 1996) and promotes the growth and repeated branching of the ureteric bud. These branches form the system of collecting ducts that join with the distal convoluted tubule of the nephron and empty into the ureter. These inductive steps are reiterated throughout the growing kidney so that older nephrons are located in the center and newer nephrons are added at the periphery. In the mouse, the process continues until 7–10 days after birth at which time there are up to 104 nephrons and the cell population has grown 1000fold. Cell lineage analysis has recently challenged the view that cells derived from the ureteric bud contribute solely

Fig. 1. Schematic representation of the nephric system in the developing mouse embryo. A composite is depicted so that pronephros (p), mesonephros (m) and metanephros are seen together, although by the time the ureteric bud (UB) has formed the presumptive pronephros has begun degenerating. The nephric or Wolffian duct (W) is first observed about the level of the fifth somite in the pronephric region and develops caudally toward the region of cloaca (not shown). The metanephric mesenchyme (MM, not drawn to scale) is found in the intermediate mesoderm near the level of developing hindlimb bud (HL).

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

107

Fig. 2. Series of schematics depicting the morphological events during metanephric kidney development. The Wolffian duct (W) elongates posteriorly and encounters the metanephric mesenchyme (MM) where the ureteric bud (UB) emerges. Mesenchymal cells near the tips of the branching ureter are induced and differentiate through a series of forms: aggregate (A), renal vesicle (RV), comma-shaped body (C) and S-shaped body (S). Also shown is the developing vasculature (V) within the glomerular cleft of an S-shaped body and the glomerulus (G) in a more mature nephron. The tubular segments (T) of the mature nephron empty into the collecting ducts (CD) and eventually the ureter (u). Development proceeds in a radial manner so that older nephrons are centrally located and newer nephrons are added at the periphery as indicated.

to the ureter and the collecting ducts. Retroviral tagging, used to follow cell fate in organ culture experiments, revealed that cells of the uninduced metanephric mesenchyme could be incorporated in several different cell types of the nephron (Herzlinger et al., 1992) as well as the collecting ducts (Qiao et al., 1995). These studies also reported that descendants of tagged clones could be found in multiple nephrons and that nephrons could be composed of both tagged and untagged cells. Retroviral infection after induction showed that tagged progeny were found in discrete nephron segments, consistent with the restriction of their cell fate after induction. In addition, there appeared to be a bias in the location of tagged cells, in that fewer cells were located in the glomerular portion. This result, combined with the observation that glomerular epithelial cell differentiation takes place earlier than cells in distal segments (Ekblom et al., 1981), is consistent with the notion that the cell types of the nephron are ‘born’ in a stereotypic

order. Also, it appears that progeny from several progenitor cells must contribute to a single nephron. Whether the mesenchymal cells are a truly homogeneous population of stem cells whose fates become restricted at the time of induction, or shortly thereafter, remains to be determined. It is also not clear how the mesenchymal cells are incorporated into the growing ureter. Within the collecting system, at least, it appears that ductal epithelium grows by expansion of ureteric bud-derived cells and recruitment of metanephric mesenchymal cells. While it has been suggested that the pronephric duct elongates by directed migration in axolotl (Poole and Steinberg, 1981) more recent studies in Xenopus show that the nephric duct incorporates cells from the surrounding mesoderm (Cornish and Etkin, 1993). Another area of focus has been in the determination of the origin of the renal vasculature and mesangium. Vessels are first observed in the cleft of the comma-shaped body pre-

108

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

figuring the juxtapositioning of the capillary network and podocytes of the glomerulus. In mice lacking plateletderived growth factor B (PDGF-B) (Leveen et al., 1994) or its receptor (PDGF-bR) (Soriano, 1994), mesangial cells are absent. The glomerular vasculature is present though it is not fully elaborated, possibly owing to the absence of mesangial cells. Alternatively, the mesangial cells may not develop without a proper capillary system. In humans, PDGF-bR is expressed initially in stromal cells and vasculature and PDGF-B in the developing epithelial of the glomerulus (Alpers et al., 1992). Later expression of both is switched on in mesangial cells. It has been proposed that, initially, the endothelial cells recruit mesangial cell progenitors by paracrine signaling between the receptor and ligand. Autocrine stimulation in mesangial cells expressing both ligand and receptor may be necessary for continued development. The fact that endothelial cells are present in the absence of mesangial cells in the PDGF knockout mice suggests that they are of separate lineage, although the exact origin of each remains unresolved. Organ culture and grafting of metanephric mesenchyme have yielded different results, some suggesting an endogenous origin of the endothelial cells and vasculature (Hyinck et al., 1996) while others provided evidence that the metanephric mesenchyme does not contain endothelial or hematopoetic cells and that these lineages are derived by invasive or angiogenic means (Ekblom et al., 1982). The confusion may arise from shortcomings in both culturing methods. That is, culturing of metanephric rudiments on filters may not allow proper development of the endothelial cell lineage and unspecified angiogenic factors in the mesenchyme may inappropriately stimulate invasion of host vasculature in grafts. Once more specific markers of the renal vasculature are found then we may be able to sort out this question both in vitro and in vivo.

3. Growth factors in induction Presently, there is a growing list of potential signaling molecules that includes many secreted factors and their receptors which are either expressed in the embryonic kidney or can affect development in the in vitro organ culture system. However, this list is markedly shortened if demonstrable effects are examined in vivo. Based on the pioneer-

109

ing work of Grobstein (Grobstein, 1953a; Grobstein, 1953b; Grobstein, 1956), it is postulated that there must be a signal that promotes outgrowth of the ureteric bud and an inductive signal from the ureter that dictates differentiation of the mesenchyme. However, it is clear that the process of induction in nephrogenesis is more complex and likely involves many intercellular signals which regulate cell proliferation, apoptosis, differentiation, and cell motility. Those growth factors and receptors that have proven to be absolutely critical for kidney development will be discussed within the sequence of early inductive events. 3.1. GDNF and RET: a signaling complex that regulates ureteric bud growth In the mouse, targeted inactivation of RET, a member of the receptor tyrosine kinase super-family, results in renal agenesis and hypodysplasia (Schuchardt et al., 1994). RET is expressed in the Wolffian duct from E8 to E11.5 and in the ureteric bud as it emerges and branches within the metanephric mesenchyme (Pachnis et al., 1993; Avantaggiato et al., 1994). Later (E13.5–17.5) expression is confined to the growing tips of the ureteric bud epithelium. The primary defect in the homozygous mutant mice (RET-k − ) is a failure of the ureteric bud to emerge and respond to signals from the metanephric blastema (Schuchardt et al., 1996). However, the RET-k − mesenchyme is capable of supporting growth and branching of wild-type ureter in culture and is also capable of differentiating into epithelium when induced with wild-type ureteric bud or spinal cord, a potent heterologous inducer. The ureteric bud from RET-k − mice was unable to respond to wild-type mesenchyme in vitro. Notably, mesonephric tubules are also present in RET-k − mice, though in slightly reduced numbers, further indicating that the ability of mesenchyme to differentiate is largely unaffected. Thus, the RET protein acts cell autonomously by receiving a signal, presumably from the mesenchyme, that activates the proliferation and branching pathway of the ureteric bud epithelium. While the importance of RET in kidney development was clearly demonstrated (Schuchardt et al., 1994), it is only recently that its ligand, glial cell line-derived neurotrophic factor (GDNF), has been identified. GDNF was originally described as a factor promoting survival of midbrain dopaminergic neurons (Lin et al., 1993). It is a secreted glyco-

Fig. 3. Expression of cadherins in the comma- and S-shaped bodies. Immunostaining for E-cadherin (green) and K-cadherin (red) in the E14 mouse kidney is shown. (A) E-cadherin is prominent in the ureteric bud (u) whereas the comma-shaped body (c) expresses E-cadherin at the end adjacent to the tip of the bud (solid arrow) and K-cadherin in the epithelium furthest from the bud tip (open arrow). (B) In the S-shaped body, E-cadherin is prominent in the epithelium connecting the ureteric bud to the prospective distal tubular epithelium (D), whereas K-cadherin is expressed in the prospective proximal tubular epithelium (p). The glomerular epithelium (g), including the podocyte cells express neither E- or K-cadherin at this time and eventually expresses P-cadherin. Fig. 4. Expression of Wnt-11 in the ureteric bud tips. In situ hybridization demonstrates wnt-11 RNA is restricted to the tips of the ureteric bud, where induction of the mesenchyme is ongoing. (A) Sagittal section through an E12 kidney showing several ureteric bud termini (arrows). (B) Dark field image of (A) showing hybridization of Wnt-11 probe. (C) At E14, induction is restricted to the peripheral nephrogenic zone where ureteric bud termini are present (arrows). (D) Dark field image of (C) showing Wnt-11 expression is still present in the ureteric bud ends but less prevalent in the more mature collecting duct epithelium.

110

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

protein that forms a homodimer and possesses a cystine knot motif that is found in other growth factors such as PDGF, NGF and TGF-b. GDNF is expressed in the metanephric mesenchyme as early as E11.5 (Hellmich et al., 1996), in addition to many other tissues. Later, expression is found in the most peripheral mesenchyme of the kidney where induction of new nephrons occurs, complementing the expression pattern of RET at the ureteric bud tips. In the GDNF knockout mice mesonephric tubules develop and the Wolffian duct forms normally but the ureteric bud is absent (Moore et al., 1996; Pichel et al., 1996; Sanchez et al., 1996), a phenotype strikingly similar to the RET-k − mouse (Schuchardt et al., 1994). The metanephric blastema can be observed in the homozygous GDNF mutant animals at early stages but this tissue undergoes apoptosis and disappears at later times. Heterozygous GDNF mutants also display defective kidney development suggesting that the level of GDNF is critical for signaling and directing ureteric bud growth and branching. In support of this hypothesis, beads soaked in recombinant GDNF were able to partially rescue ureteric branching in GDNF homozygous mutant kidney rudiments grown in culture (Pichel et al., 1996). In addition, wild-type kidney rudiments produced supernumerary branches of the ureter when GDNF-soaked beads or media containing GDNF was added to the cultures (Pichel et al., 1996; Vega et al., 1996). Moreover, ureter branching was not stimulated by GDNF-soaked beads in RET-k − kidney rudiments (Durbec et al., 1996). These results demonstrate that GDNF and RET are elements of the same signaling pathway that regulates ureter budding and branching morphogenesis. The GDNF/RET interaction can also be demonstrated biochemically. GDNF forms a complex with RET and stimulates autophosphorylation of the cytoplasmic tyrosine kinase domain (Trupp et al., 1996; Vega et al., 1996). However, binding to RET with high affinity requires that GDNF first bind to a receptor molecule called GDNFR-a (Jing et al., 1996; Treanor et al., 1996), which is attached to the cell surface via a GPI linkage and cannot signal on its own. This complex then binds to the RET receptor, causing dimerization and inducing tyrosine kinase activity. Thus, ligand binding occurs through an obligatory adapter molecule, which can also function when released from the membrane with phospholipase. Predictably, GDNFR-a mRNA is found in the epithelium of developing nephrons coincident with RET and adjacent to sites of GDNF expression (Treanor et al., 1996). Taken together, the data from these studies show that the major signal for outgrowth of the ureteric bud into the metanephric mesenchyme is generated by GDNF/ GDNFR-a/RET signaling complex. Continued branching and patterning in the developing kidney is probably generated by a refinement of the expression pattern. That is, expression is down-regulated in mature nephrons but continues at high levels in active areas of nephrogenesis. Thus, ureteric epithelium not expressing GDNFR-a or RET would not be expected to respond to exogenously supplied GDNF,

a readily testable prediction. It is not known if there are situations where RET and GDNFR-a are not coexpressed in the same tissue. Further characterization of expression is ongoing as well as the search for other GDNFR-a-like receptors. Although the significance of GDNF/GDNFR-a/ RET signaling complex in early kidney development is clear, the results naturally raise more questions. For example, a small fraction of ureteric buds are formed in RET-k − mice and some eventually make contact with the mesenchyme (Schuchardt et al., 1994). Similarly, a low percentage of RET-k − ureters display branching in vitro, suggesting that additional signaling molecules may provide cues for ureteric bud development (Schuchardt et al., 1996). The localization of GDNF expression to the metanephric mesenchyme as opposed to other nephrogenic mesenchyme may ensure correct positioning of ureter budding and clearly distinguishes this blastema from other mesenchyme surrounding the Wolffian duct as it grows along the urogenital ridge. The factors that control GDNF expression in the metanephric mesenchyme may function to pattern this posterior domain of intermediate mesoderm and thus predispose the mesenchyme to generate kidney tubules. Deficient ureter budding is also found in the mouse limb deformity (ld) mutation (Woychik et al., 1985). Although homozygous ld mice have agenic kidneys, the mutant metanephric mesenchyme can be rescued with heterologous inducer in vitro (Maas et al., 1994). These observations indicate that renal agenesis in ld mice stems from defects intrinsic to the ureter. Since ld transcripts encode phosphoproteins called formins which are proposed to function in intracellular protein-protein interactions (Woychik et al., 1990; Vogt et al., 1993; Chan et al., 1996; Uetz et al., 1996), they may be involved in signal transduction in cells of the ureter epithelium. Transcripts have been isolated from both ureter and the metanephric mesenchyme, but more recent data suggests that ld expression is more prevalent in the ureteric bud (Chan et al., 1995). One possibility is that formins act downstream of the RET/GDNF signaling pathway; such a hypothesis would be consistent with the ld phenotype and the lack of demonstrable effect in the mesenchyme. 3.2. Wnt-11 and ureteric bud growth. The Wnt gene family encodes secreted proteins that function in a variety of patterning events during vertebrate embryogenesis (Lee et al., 1995). They are homologues of the Drosophila morphogen wingless, which is required for correct dorsal-ventral patterning in the wing disc. That wnt proteins can function as inducers of metanephric mesenchyme in vitro was first demonstrated by Herzlinger et al. (1994). In that study, metanephric mesenchyme was induced to form epithelium when cultured over NIH3T3 fibroblasts transfected with a wnt-1 expression vector. Although wnt-1 is not normally expressed in ureteric bud, other members of the wnt gene family are expressed in

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

domains that suggest inducing function (Fig. 3). The Wnt11, Wnt-7b and Wnt-4 genes are expressed in unique domains of the developing mouse kidney (Stark et al., 1994; Kispert et al., 1996). Recent results demonstrate that Wnt-11 mRNA is localized to the developing Wolffian duct as it elongates, but becomes restricted to a discrete region in the duct exactly at the position adjacent to the metanephric blastema (Kispert et al., 1996). As the ureteric bud emerges, Wnt-11 expression can be found in the tip, but not the stalk. This dynamic pattern continues so that the Wnt-11 gene is always expressed at the ends of the growing ureter. Utilizing Danforth’s short-tail (Sd) (GluecksohnSchoenheimer, 1943) and Wnt-4 mutant mice (Stark et al., 1994), the authors show that the contact with the metanephric mesenchyme is required for maintaining Wnt-11 expression at the ureter tips. In vitro culture experiments demonstrated that sulfated proteoglycans are required for ureteric bud branching, but they are not required for differentiation of the metanephric mesenchyme (Lelongt et al., 1988; Klein et al., 1989; Platt et al., 1990; Davies et al., 1995). Disruption of the proteoglycan network in organ culture suppresses RET and Ros expression, but these changes are evident after the loss of Wnt-11 expression (Kispert et al., 1996). Signals from the metanephric mesenchyme may refine and limit the Wnt-11 expression pattern to the ureteric bud tips, which in turn may establish the expression pattern of RET. This hypothesis can be directly tested by modulating Wnt-11 expression in the developing embryo. The proteoglycan network and other components of the extracellular matrix may act as a scaffold to localize high concentrations of secreted factors at the tips of the ureteric buds. The ability of lung mesenchyme to support branching of the ureter was also demonstrated by Kispert et al. (1996). Surprisingly, the lung mesenchyme was able to induce ureteric bud branching in a pattern more consistent with lung epithelium, although the ureteric bud epithelium did retain expression of the marker genes Pax-2 and c-RET. Wnt-11 was still localized to the tips of the branching ureteric buds, indicating any mesenchymal-derived signals that limit Wnt11 expression are not entirely kidney-specific. Previous studies showed heterologous mesenchyme was unable to support ureteric branching (Saxen, 1987) suggesting that the metanephric mesenchyme possessed unique and specific cues for ureteric bud initiation and growth. However, it now appears that these mesenchymal factors are also found in other tissues and may be required more for survival and proliferation. A kidney-specific program may be intrinsic to the ureteric epithelium, and becomes modified by additional signals expressed by the metanephric mesenchyme after induction. 3.3. Wnt-4 and BMP-7: regulators of nephron maturation Another member of the Wnt gene family, Wnt-4 is required for epithelium formation in the induced metanephros. Wnt-4 transcripts are found in aggregating mesench-

111

ymal cells adjacent to the growing ureter and their descendants, the comma- and S-shaped bodies (Stark et al., 1994). Mice with targeted disruption of the Wnt-4 coding region die at birth apparently as result of renal agenesis (Stark et al., 1994). Primary induction is observed by E11.5 but particular defects in Wnt-4 mutant kidneys are clearly seen on E15. The mesenchyme remains largely undifferentiated, although some markers of early induction such as Pax-2 and N-myc are detected. There is little evidence of epithelial cell formation and no expression of the Pax-8 gene in the mutant kidneys. Clearly, wnt-4 is required for the progression of mesenchymal aggregates to the epithelial phenotype. Whether wnt-4 functions in an autocrine manner awaits the identification of its receptor. Nevertheless, the data point to a stepwise inductive process that requires both ureteric bud and mesenchymal derived signals for the conversion of mesenchyme to epithelium. In the absence of this progression, the ureteric bud branching pattern is also arrested. Among the secreted growth factors expressed in the embryonic kidney at the time of induction is Bmp-7 (also known as OP-1). Bmp proteins are members of the TGF-b superfamily of secreted signaling molecules which play roles in many morphogenetic events (Kingsley, 1994). Bmp-7 transcripts are found in the mesonephric duct and tubules and later in the metanephric condensates, commaand S-shaped bodies (Lyons et al., 1995). Expression can also be found in the collecting ducts formed by the branching ureter. Homozygous mice carrying a disrupted BMP-7 gene die soon after birth and possess dysplastic kidneys which appear arrested in development (Dudley et al., 1995), whereas heterozygotes appear to develop normally (Luo et al., 1995). The Bmp-7-deficient embryonic kidneys showed some development of comma- and S-shaped bodies by E14.5 but abnormal patterning is observed shortly thereafter. Thus, initiation of mesenchymal condensation and differentiation appear to take place, yet later morphogenetic events are prevented, resulting in few if any glomeruli and an increase in the amount of interstitial or stromal tissue in the mutant kidneys. Indeed, many markers of induced mesenchyme are expressed in the early Bmp-7 mutant kidneys, such as WT-1, Pax-2, Pax-8, Wnt-4 and RET. However, the expression levels are generally reduced and the patterns abnormal. The incomplete condensation and differentiation of the mesenchymal aggregates is more apparent at later times in development. Recent in vitro culture experiments demonstrate that purified Bmp-7 can induce differentiation of metanephric mesenchyme (Vukicevic et al., 1996). Conversely, inhibition of Bmp-7 activity via antisense oligonucleotides or blocking antibodies limits differentiation and tubulogenesis in culture which is evidenced by the failure to express many of the markers noted above. These results are somewhat contradictory, as mesenchymal development in vivo proceeds significantly further than in vitro when Bmp-7 activity is lost. Since Bmp-7 is expressed in the induced mesenchyme, in addition to the ureteric bud,

112

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

one might expect the signal to be propagated laterally, from induced mesenchyme to surrounding uninduced cells. Yet, this clearly is not the case, as classical studies show that induced mesenchyme cannot function as a source of inductive signals (Saxen and Saskela, 1971). Presently, it is not known whether Wnt-4 is required for proper Bmp-7 expression in the mesenchyme. Nonetheless, both factors appear to act after the initial induction of the mesenchyme and are required for subsequent mesenchymal differentiation. Interestingly, both are expressed in the embryonic spinal cord, a potent inducer of nephrogenesis in vitro (Saxen, 1987). In Bmp-7 mutant mice, mesenchymal cells undergo apoptosis (Luo et al., 1995), suggesting that induction fails to progress in these kidneys and that Bmp-7 may be a survival factor. Furthermore, it appears that Bmp-7 and Wnt-4 may autoregulate themselves, which prevents analysis of their developmental roles at later times. Another interesting inducer of cultured metanephric mesenchyme is LiCl (Davies and Garrod, 1995). Lithium has recently been shown to inhibit the activity of glycogen synthase kinase 3 (GSK-3) (Klein and Melton, 1996), which functions in the Wnt signaling pathway (Peifer et al., 1994). Inhibition of GSK-3 prohibits its negative regulation of bcatenin, leaving b-catenin free to signal to factors in the nucleus, such as HMG-box transcription factors (Miller and Moon, 1996). These recent studies conducted in Xenopus may point the way to finding downstream targets of Wnt in the developing kidney. 3.4. Additional factors Fibroblast growth factors (FGF) and their receptors have been found in tissues of the developing and adult kidney (Orr-Urtreger et al., 1993; Dono and Zeller, 1994). FGF2 (basic FGF) has been identified as a component of pituitary extract that is able to elicit condensation and the expression of WT-1 and c-met in metanephric mesenchyme cultures but tubulogenesis requires additional, unidentified components (Perantoni et al., 1995). Embryonic nervous system tissue appears to be a rich source of a number of growth factors involved in kidney development and helps to explain the potency of this heterologous inducer of nephrogenesis. Furthermore, FGFs are known to bind heparin sulfate proteoglycans and may signal in this milieu (Venkataraman et al., 1996). Thus, the activity of FGFs could be perturbed by the disruption of the kidney proteoglycan network. Another family of putative morphogens, the hedgehog proteins, may also affect kidney development. Sonic hedgehog (Shh) appears to be expressed early in developing urogenital system (Bitgood and McMahon, 1995), whereas the related gene Indian hedgehog (Ihh) is expressed in the more mature kidney with maximum mRNA levels found in proximal tubules (Valentini et al., 1997). Gene disruption of Shh in mouse results in severe defects in axial patterning and the absence of kidneys (Chiang et al., 1996). Shh is expressed in

the urogenital sinus near E11.5, but it is unclear from the knockout report if the Wolffian duct is formed in the mutant embryos. The lack of kidneys in this instance could reflect an aberration in the midline of the mutant embryos which in turn affects the general development of many axial tissues such as the intermediate mesoderm. It remains to be seen whether the hedgehog proteins have more specific roles in kidney induction. The importance of the midline in embryonic development of the kidney is illustrated by the urogenital abnormalities in Sd mice (Gluecksohn-Schoenheimer, 1943). The notochord degenerates in the caudal region of developing Sd embryos, affecting axis formation and kidney organogenesis in heterozygous and homozygous animals. Though the Wolffian duct is present, contact with metanephric blastema is delayed or absent. Both primordia remain competent to respond to induction in vitro, indicating that the defect is not intrinsic to the ureter or mesenchyme and is probably due to improper positioning between them (GluecksohnWaelsch and Rota, 1963). The Sd gene has not yet been identified, so it is not possible to speculate on the molecular nature of this defect.

4. Transcriptional regulation of kidney development The function of key transcription factors are believed to coordinate the expression of effector molecules which enact specific differentiation programs. These ‘master regulators’ are proposed to set up basic pattern formation in the developing embryo and the kidney is no exception. To date, the most conclusive data have come from studies of mouse mutants generated by gene targeting. Additionally, there are a large number of transcription factors whose expression domains in the developing kidney suggest functions during the differentiation of the mesenchyme or ureter bud, but whose definitive roles have not been established. 4.1. BF-2: stromal cell function. BF-2 is a member of the Winged Helix family of transcription factors which are related to the Drosophila fork head gene (Lai et al., 1993). Mice lacking BF-2 die within 24 h of birth and exhibit rudimentary kidneys that are smaller and fused (Hatini et al., 1996). The reduced size is apparent by E14.5 in the homozygous mutant embryos and the number of ureter branches is also reduced at this stage. Expression of BF-2 can be detected in cells immediately surrounding the condensed cells of the metanephric mesenchyme which express Pax-2. Thus, expression of BF-2 distinguishes a population of mesenchymal cells that are not induced to become epithelium, but are destined to become stromal cells. Interestingly, in the mutants these interstitial cells are present throughout embryonic kidney development. Yet, abnormal development appears to arise from defective ureter growth and subsequent conversion of

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

metanephric mesenchyme. Normally, RET is expressed in the periphery of the developing kidney by E14.5, but in the BF-2− kidneys RET can be found in the medulla. Furthermore, expression of RET is found throughout the ureter branch though it is normally localized to the ureter tips. Aberrant ureter growth may also explain why the kidneys have not ascended to their usual position near the adrenal glands in the BF-2− mutant mice. Another striking feature of the mutant kidneys is the absence of comma- and S-shaped bodies. The induced mesenchyme aggregates and cells express Wnt-4 and Pax-2, but morphogenesis is arrested at this point. The data argues that BF-2 expression in stromal cells is not required for their development but is necessary for regulation of the nephrogenesis in the induced cell population, which does not express BF-2 and is destined to make epithelium. Since initial events appear to take place normally the stromal cells likely provide factors for both ureteric bud growth and maturation of pretubular aggregates. BF-2-positive cells are found adjacent to the growing ureter as it enters the kidney, at the periphery of the developing kidney and interspersed in the medullary region. Thus, these cells are appropriately located with respect to their proposed function in regulating ureter growth and differentiation of the mesenchymal cells. Since the stromal cell lineage is poorly characterized there are few clues regarding the possible nature of deficiency within these cells. It will be necessary to take a closer look at the function of these cells and find new markers, given the importance of the findings described above. However, some headway may be gained simply by examining expression of some of the genes in hand. For example, do stromal cells express GDNF or produce FGFs? And is expression disturbed in the BF-2 − mutants? Conversely, using BF-2 as a marker it is now possible to study the effects on stromal cell lineage in both mouse mutants and organ culture experiments. 4.2. WT-1 and Pax genes: a regulatory circuit? Originally identified as a Wilms’ tumor suppressor gene (Call et al., 1990; Gessler et al., 1990), WT-1 encodes a transcription factor, with four zinc-finger domains, that is required for the embryonic kidney development (Kreidberg et al., 1993). WT-1 maps to human 11p13, a locus where the loss of heterozygosity is associated with embryonal kidney tumors (Wilms’ tumor) and urogenital malformations (Bruening and Pelletier, 1994; Hastie, 1994). In the developing kidney, low levels of WT-1 expression can first be detected in the uninduced metanephric mesenchyme with a subsequent increase in expression post-induction (Fig. 4) (Pritchard-Jones et al., 1990; Armstrong et al., 1992). In the homozygous mutant WT-1− mice, the ureteric bud fails to form, leading to the eventual degeneration of the metanephric mesenchyme and the absence of kidneys (Kreidberg et al., 1993). Similar to what had previously been observed in large deletions of the WT-1 locus (Glaser et al., 1990), WT1 heterozygotes show no obvious developmental problems

113

and are not predisposed to renal tumors suggesting that the mechanism of kidney tumor formation or the frequency of loss of heterozygosity is different between humans and mice. In the mesonephros, tubules are formed in WT-1− embryos, albeit at reduced number. The Wolffian duct is present and expresses Pax-2 protein, suggesting that, for the most part, its development is unaffected. In vitro, WT-1− metanephric mesenchyme cannot be induced by heterologous, wild-type spinal cord, and mutant mesenchyme fails to express Pax-2 both in vitro and in vivo. Lacking WT-1 activity, the mesenchyme is unable to respond to inductive signals and undergoes apoptosis by default. Although this defect is intrinsic to the mesenchymal component, absence of the ureteric bud implies that the mesenchyme also is unable to emit proper signals for ureter bud proliferation and guidance. Clearly, GDNF expression in the metanephric mesenchyme of WT-1− mutant embryos must be examined to determined if this underlies the lack of ureter budding. The importance of WT-1 in kidney development is further underscored by conservation of the gene across 400 million years and expression in the developing nephrogenic systems of chick and alligator (Kent et al., 1995). Moreover, sequence comparison among WT-1 orthologues shows strong conservation of sub-regions identified, such as the DNA-binding zinc fingers and the KTS alternative splice (Rauscher et al., 1990), that are necessary for DNA binding or transcriptional potentiation. Analyses of WT-1 transcriptional regulatory properties have shown that it can act as a repressor or activator depending upon the particular site studied (Madden et al., 1991; Wang et al., 1992; Rupprecht et al., 1994; Ryan et al., 1995). Although many of the genes potentially regulated by WT-1 are involved in control of cell growth, the most provocative targets may be the WT-1 gene itself (Rupprecht et al., 1994) and Pax-2 (Ryan et al., 1995). In both instances, transcriptional repression is mediated through GC-rich binding sites for WT-1 located within their promoter regions. This suggests that WT-1 attenuates its own expression and may be responsible for limiting Pax2 expression in vivo. In the metanephric mesenchyme, Pax2 protein is seen in the earliest condensates and commashaped bodies (Dressler et al., 1990). Expression begins to decline in S-shaped bodies and expression is not detected in the mature nephron (Fig. 5). Coincident with this, high levels of WT-1 protein are observed in the podocyte precursor cells of the S-shaped bodies precisely where Pax-2 expression is lost first (Ryan et al., 1995). The complementary expression pattern of Pax-2 and WT-1 thus support a functional interaction between these two transcription factors. In addition, high levels of Pax-2 have been reported in Wilms’ tumors, further supporting the notion that WT-1 down-regulates Pax-2 gene expression (Dressler and Douglass, 1992). More recent studies suggest that Pax proteins can activate expression of WT-1 (Dehbi and Pelletier, 1996; Dehbi et al., 1996), indicating a reciprocal relationship. Thus, Pax-2 and Pax-8 may up-regulate WT-1 expression

114

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

Fig. 5. Immunostaining for Pax-2 and WT-1 expression in the developing kidney. (A) At E12, Pax-2 (orange) is present in the ureteric bud epithelium (u) and in condensing mesenchymal cells near the tips of the ureteric bud. (B) Serial section to (A) showing low levels of WT-1 (orange) in the mesenchymal cells at E12. (C) By E14, Pax-2 protein (orange) is still prominent in the nephrogenic zone but levels begin to decline in the prospective podocyte cells of the Sshaped body (solid arrow). Pax-2 is not found in more mature glomeruli (open arrows). (D) Serial section to (C) showing up-regulation of WT-1 (orange) protein levels precisely in those podocyte precursors that down-regulate Pax-2 (solid arrow). WT-1 levels remain high in the podocytes of the glomerular epithelium (open arrows). In all sections, the ureteric bud epithelium is stained green with anti-E-cadherin antibodies.

in the responding mesenchyme which in turn begins to turn off Pax-2 gene expression as nephrogenesis is completed. Finally, studies have shown that the KTS form of WT-1 is associated with the RNA splicing apparatus (Larsson et al., 1995) and has a potential RNA binding motif (Kennedy et al., 1996). Whether WT-1 has any direct function in the splicing of mRNA precursors or if it can affect alternative splicing of kidney-specific genes has not been demonstrated. 4.3. Pax genes in the developing kidney The Pax gene family encodes evolutionary conserved transcription factors which share a DNA-binding domain, the paired box (Gruss and Walther, 1992). Within this family, Pax-2 and Pax-8 are expressed in overlapping domains within the developing kidney (Dressler et al., 1990; Plachov et al., 1990). The dynamic expression pattern of Pax-2 has been described above. Pax-8, which shares a significant degree of homology with Pax-2 outside the paired box, is expressed at a slighter later stage in the renal vesicles and persists in the S-shaped body (Poleev et

al., 1992). Since Pax-2, and not Pax-8, is still expressed in mesenchymal cells in Wnt-4 − mice it appears that the expression of these two genes in the developing tubules can be separated genetically and provides compelling evidence that nephrogenesis proceeds in a stepwise manner. Recently the Pax-2 gene has been knocked out in mice (Torres et al., 1995). Predictably the homozygous null mutants do not develop kidneys and fail to form mesonephric tubules and genitourinary tracts. This is most likely due to defects in Wolffian duct formation, which appears to degenerate between E10 and E12. Unlike the phenotypes of other knockout mice exhibiting renal defects, Pax-2related defects affect epithelial differentiation of the intermediate mesoderm in general. Heterozygous Pax-2 mutant kidneys are reduced in size and hypoplastic. Similar phenotypes have now been found in mice and humans with the same frame-shift mutation in one copy of the Pax-2 gene (Sanyanusin et al., 1995; Favor et al., 1996). Hence Pax-2, like many other Pax genes, displays haplo-insufficient disease phenotype indicating that a certain level of expression may be required for proper function during development.

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

A critical role for Pax-2 in mesenchymal cell differentiation, a function that could not be determined in the Pax-2 null mice since primary induction does not take place, was demonstrated by inhibiting expression with antisense oligonucleotides in organ culture (Rothenpieler and Dressler, 1993). Conversely in more differentiated renal epithelium, failure to repress Pax-2 in transgenic mice causes a nephrotic-like syndrome with both proximal and distal tubule cysts and effacement of the podocyte foot processes (Dressler et al., 1993). High levels of Pax-2 protein are also found in the cystic epithelia from human dysplastic kidneys and correlate with continued cell proliferation (Winyard et al., 1996). Together, these data clearly show that kidney development relies on coordinating the expression of Pax-2 in multiple areas. For example, Pax-2 is down-regulated in proximal tubule cells which do not express high levels of WT-1 (Ryan et al., 1995). This fact alone indicates that there must be multiple mechanisms to ensure the proper spatiotemporal pattern of expression. 4.4. Hox genes The vertebrate Hox gene clusters represent duplications of the Drosophila homeobox genes. They encode homeodomain transcription factors and are thought to specify positional information along the anterior-posterior axis based upon their nested expression domains and the temporal sequence in which they are expressed (Krumlauf, 1994). Many of the Hox genes have been analyzed in mice by targeted disruption and so far none of the single knockouts display kidney abnormalities. Breeding separate hox-a11 and hox-d11 mutant heterozygotes produced double mutants whose phenotypes were more severe than predicted by summing the defects of the single mutants (Davis et al., 1995). Double homozygous mutants often lack one or both kidneys, and kidneys that do develop are hypoplastic. Both are of these Hox genes are expressed in the metanephric mesenchyme (Hsieh-Li et al., 1995), yet deletion of either one does not affect kidney development (Small and Potter, 1993; Davis and Capecchi, 1994). This might be considered a case of redundancy in gene function, and the results could further indicate that the ‘Hox code’ set by two these genes is important for specifying the metanephric blastema at the posterior end of the intermediate mesoderm. In this regard, ectopic expression may provide more meaningful data on Hox gene function in kidney development, as recently shown in a small number of Hoxb-7 transgenic mice (Argao et al., 1995). Another homeodomain-containing transcription factor, Lim1, appears to be required for the formation of the nephric duct and the pro- and mesonephric tubules (Shawlot and Behringer, 1995). Because most embryos die near E10 the basis for a complete lack of kidneys was not determined. The Lim1 expression pattern during embryogenesis includes nephrogenic mesenchyme, Wolffian duct and induced tubules (Barnes et al., 1994; Fujii et al., 1994) and makes a convincing argument that

115

Lim1 has some fundamental role in the early steps of mammalian kidney formation.

5. Cell proliferation and apoptosis in kidney development The control and execution of cell differentiation pathways in the developing kidney require both general and kidney-specific factors. General factors are likely to be involved in a variety of events such as promoting cell division, generating early growth response, regulating cell motility, and establishing cell contacts. Little is known regarding the precise way in which these events are coordinated in the developing kidney; however, some clues are beginning to emerge. 5.1. p53 The product of tumor suppressor gene p53 is thought to regulate cell growth in two ways. It may act as a checkpoint in the cell cycle and in a separate manner control apoptosis in response to stress (Polyak et al., 1996). In the developing kidney, p53 is normally expressed in commaand S-shaped bodies (Schmid et al., 1991), though deletion of p53 in mice does not significantly affect development (Donehower et al., 1995). In contrast, overexpression of wild-type p53 in the ureter of transgenic animals leads to smaller kidneys with fewer nephrons (Godley et al., 1996). Differentiation of the mesenchyme is clearly incomplete, as several cell surface markers fail to be expressed. An increase in apoptosis was also observed in the transgenic kidneys. The defects in the mesenchyme may stem from interactions with the ureteric bud in which misexpression of RET was noted. Cells which normally express p53 in the kidney are dividing but some progeny are beginning to differentiate and pull out of the cell cycle. Pax-2 and WT-1 are also expressed in these cells at this stage. Experiments have shown that Pax-2 down-regulates the expression of p53 (Stuart et al., 1995) and that WT-1 binds p53 protein (Maheswaran et al., 1995). Although these functions have yet to be demonstrated in vivo, Pax-2 and WT-1 could limit p53 activity until the appropriate number of cell divisions has occurred and the more differentiated cells exit the mitotic cycle. 5.2. Apoptosis Widespread cell death occurs in the metanephric mesenchyme in the absence of inductive signals (Koseki et al., 1992). Thus, it seems likely that induction generates both survival and differentiation cues. Indeed, a number of growth factors, including EGF (Weller et al., 1991) and bFGF (Perantoni et al., 1995), can prevent cell death in cultured mesenchyme. In the kidney, as well as in other developing organs, many cells also undergo programmed

116

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

Fig. 6. Flow chart indicating some of the critical genes involved in metanephric kidney development. Boxes summarize the steps in organogenesis as illustrated in Fig. 2. Listed below each are genes known to be expressed at that time which are grouped according to expression in the Wolffian duct and the ureteric bud (W/UB) or the metanephric mesenchyme and its derivatives (MM). Genes which are required primarily for particular steps in development are shown in bold and underlined. Note that in this scheme we have listed gene products in a general order of appearance from induction to the generation of a single nephron and ignored expression that takes place later as the pattern of nephrogenesis is repeated throughout kidney development.

cell death as a normal consequence of development. Apoptotic cells are characteristically found as islands between the maturing nephrons in the mesenchyme and frequently located next to condensed bodies of cells or at the tails of S-shaped bodies (Koseki et al., 1992; Coles et al., 1993). Whether these cells share common characteristics is not known. A later phase of cell death occurs in the branches of the ureter after birth (Coles et al., 1993). Factors controlling apoptosis have been investigated and one of these, bcl2, is from a class of cytoprotective proteins (Nunez and Clarke, 1994). Mice lacking bcl-2 develop cystic kidneys that are hypoplastic and have fewer nephrons. Examination of the homozygous mutant kidneys in embryos showed that organogenesis proceeded normally up to E13 in one study (Nagata et al., 1996) and up to birth in another (Veis et al., 1993; Sorenson et al., 1996). A dramatic increase in apoptosis within the mesenchyme was observed, and to a lesser extent in the cysts and interstitium. These results demonstrate that regulation of apoptosis is required for normal kidney development and that bcl-2 may block apoptosis in many kidney cell types.

6. Conclusions and perspectives The development of the excretory system in animals requires the precise three-dimensional organization of specialized epithelial cell types into a series of secretory nephrons. The number of genes known to be required for different steps of kidney morphogenesis is increasing and has allowed for a genetic dissection of key steps in the initiation and progression of renal development (Fig. 6). The reciprocal inductive events that occur between the epithelium of the Wolffian duct and the adjacent mesenchyme require the GDNF/GDNFR-a/RET receptor-ligand complex for ureteric bud growth and branching morphogenesis, whereas the mesenchyme requires WT-1 and Pax-2 to respond to induction. It is quite possible that the reiterated use of GDNF paracrine signaling complex is responsible for generating the radial pattern of growth and induction in the developing kidney. Indeed, aberrant expression of RET in the ureter is consistently associated with abnormal or delayed branching. The signals emanating from the ureteric bud that induce the mesenchyme have not been conclusively

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

identified but may involve one or more members of the Wnt gene family. The evidence points to a multi-step inductive interaction that requires several signaling molecules before proliferation and differentiation of the mesenchyme into tubular epithelium is fully achieved. Critical signaling factors that may feed back to enhance and maintain the inductive response include wnt-4 and Bmp-7. As kidney development proceeds, the number of cell types generated expands so it is not surprising to find that more molecules appear to play a role later in nephrogenesis. Some factors may be used again or in combination to provide the necessary repertoire of signals as the tissue becomes more complex. This being the case, methods other than gene knockout will be needed to unravel functions which are required later in development. One possibility is to use ectopic expression with transgenes bearing specific promoter/enhancer elements. To this end, it will be necessary to begin characterizing the regulatory elements which confer proper expression patterns of the various genes in the embryonic kidney. Concomitantly, such studies will enable a search for the transcription factors which regulate expression and, as such, pattern the early metanephric mesenchyme from surrounding mesoderm. Though not thoroughly examined, the timing of the interaction between the ureteric bud and the metanephric mesenchyme in vivo is likely to be critical for the primary inductive events and subsequent development. When and how the metanephric mesenchyme becomes competent to receive signals from the ureter may also depend on signals received from the notochord or other tissues that are required for patterning the posterior mesoderm. To date, the classical studies on mesenchymal-epithelial inductive interactions have proven to be a sound basis for further investigation into the nature of kidney organogenesis. More recently, the field has received a significant boost by the identification of signaling molecules, receptors, and transcription factors that are expressed in restricted domains during kidney morphogenesis. Furthermore, the development of gene targeting technology and the refinement of in vitro organ culture methods have accelerated the assignment of specific functions to these new molecules within the conceptual framework established many years ago. Unlike gastrulation or early patterning, organogenesis specifies the end point for many highly specialized cell types. As such, the study of tissue development relates directly to many pathogenic conditions where such end points are disturbed. How this ever-increasing body of knowledge may be applied to the understanding and treatment of renal diseases should also be a priority, given the prevalence of such diseases within the general population.

References Alpers, C.E., Seifert, R.A., Hudkins, K.L., Johnson, R.J. and Bowen-Pope, D.R. (1992) Developmental patterns of PDGF B-chain, PDGF receptor

117

and a-actin expression in human glomerulogenesis. Kidney Int. 42, 390–399. Argao, E.A., Kern, M.J., Branford, W.W., Scott, Jr., W.J. and Potter, S.S. (1995) Malformations of the heart, kidney, palate, and skeleton in aMHC-Hoxb-7 transgenic. Mech. Dev. 52, 291–304. Armstrong, J.F., Pritchard-Jones, K., Bickmore, W.A., Hastie, N.D. and Bard, J.B.L. (1992) The expression of the Wilms’ tumor gene, WT1, in the developing mammalian embryo. Mech. Dev. 40, 85–97. Avantaggiato, V., Dathan, N.A., Grieco, M., et al. (1994) Developmental expression of the RET protooncogene. Cell Growth Differ. 5, 305–311. Bard, J.B.L., McConnell, J.E. and Davies, J.A. (1994) Towards a genetic basis for kidney development. Mech. Dev. 48, 3–11. Barnes, J.D., Crosby, J.L., Jones, C.M., Wright, C.V. and Hogan, B.L. (1994) Embryonic expression of Lim-1, the mouse homolog of Xenopus Xlim-1, suggests a role in lateral mesoderm differentiation and neurogenesis. Dev. Biol. 161, 168–178. Bitgood, M.J. and McMahon, A.P. (1995) Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. Dev. Biol. 172, 126–138. Bruening, W. and Pelletier, J. (1994) Denys-Drash syndrome: a role of the WT1 tumor suppressor gene in urogenital development. Semin. Dev. Biol. 5, 333–343. Call, K.M., Glaser, T., Ito, C.Y., Buckler, A.J., Pelletier, J., Haber, D.A., Rose, E.A., Kral, A., Yeger, H., Lewis, W.H., Jones, C. and Housman, D.E. (1990) Isolation and characterization of a zinc finger polypeptide gene at the human chromosome 11 Wilms tumor locus. Cell 60, 509– 520. Chan, D.C., Wynshaw-Boris, A. and Leder, P. (1995) Formin isoforms are differentially expressed in the mouse embryo and are required for normal expression of fgf-4 and shh in the limb bud. Development 121, 3151–3162. Chan, D.C., Bedord, M.T. and Leder, P. (1996) Formin binding proteins bear WWP/WW domains that bind proline-rich peptides and functionally resemble SH3 domains. EMBO J. 15, 1045–1054. Chiang, C., Litingtung, Y., Lee, E., Young, K.E., Corden, J.L., Westphal, H. and Beachy, P.A. (1996) Cyclopia and defective axial patterning in mice lacking sonic hedgehog gene function. Nature 383, 407– 413. Coles, H.S.R., Burne, J.F. and Raf, M.C. (1993) Large-scale normal cell death in the developing rat kidney and its reduction by epidermal grown factor. Development 118, 777–784. Cornish, J.A. and Etkin, L.D. (1993) The formation of the pronephric duct in Xenopus involves recruitment of posterior cells by migrating pronephric duct cells. Dev. Biol. 159, 338–345. Davies, J.A. and Bard, J.B.L. (1996) Inductive interactions between the mesenchyme and the ureteric bud. Exp. Nephrol. 4, 77–85. Davies, J.A. and Brandli, A.W. (1994) The Kidney Development Database: http//mbisg2.sbc.man.ac.uk/kidbase/kidhome.html and http:// www.ana.ed.ac.uk/anatomy/kidbase/kidhome.html. Davies, J.A. and Garrod, D.R. (1995) Induction of early stages of kidney tubule differentiation by lithium ions. Dev. Biol. 167, 50–60. Davies, J., Lyon, M., Gallagher, J. and Garrod, D. (1995) Sulphated proteoglycan is required for collecting duct growth and branching but not nephron formation during kidney development. Development 121, 507– 1517. Davis, A.P. and Capecchi, M.R. (1994) Axial homeosis and appendicular skeleton defects in mice with a targeted disruption of hoxd-11. Development 120, 2187–2198. Davis, A.P., Witte, D.P., Hsieh-Li, H.M., Potter, S.S. and Capecchi, M.R. (1995) Absence of radius and ulna in mice lacking hoxa-11 and hoxd11. Nature 375, 791–795. Dehbi, M. and Pelletier, J. (1996) PAX8-mediated activation of the wt1 tumor suppressor gene. EMBO J. 15, 4297–4306. Dehbi, M., Ghahremani, M., Lechner, M., Dressler, G. and Pelletier, J. (1996) The paired-box transcription factor, PAX2, positively modulates expression of the Wilms’ tumor suppressor gene (WT1). Oncogene 13, 447–453.

118

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

Donehower, L.A., Godley, L.A., Aldaz, C.M., Pyle, R., Shi, Y.-P., Pinkel, D., Gray, J., Brandley, A., Medina, D. and Varmus, H.E. (1995) Deficiency of p53 accelerates mammary tumorigenesis in Wnt-1 transgenic mice and promotes chromosomal instability. Genes Dev. 9, 882–895. Dono, R. and Zeller, R. (1994) Cell-type-specific nuclear translocation of fibroblast growth factor-2 isoforms during chicken kidney and limb morphogenesis. Dev. Biol. 163, 316–330. Dressler, G.R. and Douglass, E.C. (1992) Pax-2 is a DNA-binding protein expressed in embryonic kidney and Wilms tumor. Proc. Natl. Acad. Sci. USA 89, 1179–1183. Dressler, G.R., Deutsch, U., Chowdhury, K., Nornes, H.O. and Gruss, P. (1990) Pax2, a new murine paired-box containing gene and its expression in the developing excretory system. Development 109, 787–795. Dressler, G.R., Wilkinson, J.E., Rothenpieler, U.W., Patterson, L.T., Williams-Simons, L. and Westphal, H. (1993) Deregulation of Pax-2 expression in transgenic mice generates severe kidney abnormalities. Nature 362, 65–67. Dudley, A.T., Lyons, K.M. and Robertson, E.J. (1995) A requirement for bone morphogenetic protein-7 during development of the mammalian kidney and eye. Genes Dev. 9, 2795–2807. Durbec, P., Marcos-Gutierrez, C.V., Kilkenny, C., Grigoriou, M., Wartiowaara, K., Suvanto, P., Smith, D., Poonder, B., Costantini, F., Saarma, M., Sariola, H. and Pachnis, V. (1996) GDNF signalling through the Ret receptor tyrosine kinase. Nature 381, 789–793. Ekblom, P., Miettinen, A., Virtanen, I., Wahlstrom, T., Dawnay, A. and Saxen, L. (1981) In vitro segregation of the metanephric nephron. Dev. Biol. 84, 88–95. Ekblom, P. Sariola, H., Karkinen-Jaaskelainen, M. and Saxen, L. (1982) The origin of the glomerular endothelium. Cell Differ. 11, 35–39. Favor, J., Sandulache, R., Neuhauser-Klaus, A., Pretsch, W., Chatterjee, B., Senft, E., Wurst, W., Blanquet, V., Grimes, P., Sporle, R. and Schughart, K. (1996) The mouse Pax21Neu mutation is identical to a human PAX2 mutation in a family with renal-coloboma syndrome and results in developmental defects of the brain, ear, eye, and kidney. Proc. Natl. Acad. Sci. USA 93, 13870–13875. Fujii, T., Pichel, J.G., Taira, M., Toyama, R., Dawid, I.B. and Westphal, H. (1994) Expression patterns of the murine LIM class homeobox gene lim1 in the developing brain and excretory system. Dev. Dyn. 199, 73–83. Gessler, M., Poustka, A., Cavenee, W., Neve, R.L., Orkin, S. and Bruns, G.A.P. (1990) Homozygous deletion in Wilms’ tumors of a zinc-finger gene identified by chromosome jumping. Nature 343, 774–778. Glaser, T., Lane, J. and Housman, D.E. (1990) A mouse model for the aniridia-Wilms’ tumor deletion syndrome. Science 250, 823–827. Gluecksohn-Schoenheimer, S. (1943) The morphological manifestations of a dominant mutation in mice affecting tail and urogenital system. Genetics 28, 341–348. Gluecksohn-Waelsch, S. and Rota, T.R. (1963) Development in organ tissue culture of kidney rudiments from mutant mouse embryos. Dev. Biol. 7, 432–444. Godley, L.A., Kop, J.B., Eckhaus, M., Paglino, J.J., Owens, J. and Varmus, H.E. (1996) Wild-type-p53 transgenic mice exhibit altered differentiation of the ureteric bud and possess small kidneys. Genes Dev. 10, 836– 850. Grobstein, C. (1953a) Morphogenetic interaction between embryonic mouse tissues separated by a membrane filter. Nature 172, 869–871. Grobstein, C. (1953b) Inductive epithelio-mesenchymal interaction in cultured organ rudiments of the mouse. Science 118, 52–55. Grobstein, C. (1956) Trans-filter induction of tubules in mouse metanephric mesenchyme. Exp. Cell Res. 10, 424–440. Gruss, P. and Walther, C. (1992) Pax in development. Cell 69, 719–722. Hastie, N.D. (1994) The genetics of Wilms’ tumor: a case of disrupted development. Annu. Rev. Genet. 28, 523–558. Hatini, V., Huh, S.O., Herzlinger, D., Soares, V.C. and Lai, E. (1996) Essential role of stromal mesenchyme in kidney morphogenesis revealed by targeted disruption of Winged Helix transcription factor BF-2. Genes Dev. 10, 1467–1478.

Hellmich, H.L., Kos, L., Cho, E.S., Mahon, K.A. and Zimmer, A. (1996) Embryonic expression of glial cell-line derived neurotrophic factor (GDNF) suggests multiple developmental roles in neural differentiation and epithelial-mesenchymal interactions. Mech. Dev. 54, 95–106. Herzlinger, D., Koseki, C., Mikawa, T. and Al-Awqati, Q (1992) Metanephric mesenchyme contains multipotent stem cells whose fate is restricted after induction. Development 114, 565–572. Herzlinger, D., Qiao, J., Cohen, D., Ramakrishna, N. and Brown, A.M.C. (1994) Induction of kidney epithelial morphogenesis by cells expressing wnt-1. Dev. Biol. 166, 815–818. Hsieh-Li, H.M., Witte, D.P., Weinstein, M., Branford, W., Li, H., Small, K. and Potter, S.S. (1995) Hoxa 11 structure, extensive antisense transcription, and function in male and female fertility. Development 121, 1373–1385. Hyinck, D.P., Tucker, D.C., St. John, P.L., Leardkamolkarn, V., Accavitti, M.A., Abrass, C.K. and Abrahamson, D.R. (1996) Endogenous origin of glomerular endothelial and mesangial cells in grafts of embryonic kidneys. Am. J. Physiol. 270, F886–F899. Jing, S., Wen, D., Yu, Y., Holst, P.L., Luo, Y., Fang, M., Tamir, R., Antonio, L., Hu, Z., Cupples, R., Louis, J.-C., Hu, S., Altrock, B.W. and Fox, G.M. (1996) GDNF-induced activation of the Ret protein tyrosine kinase is mediated by GDNFR-a, a novel receptor for GDNF. Cell 85, 1113–1124. Kennedy, D., Ramsdale, T., Mattick, J. and Little, M. (1996) An RNA recognition motif in Wilms’ tumour protein (WT1) revealed by structural modelling. Nature Genet. 12, 329–332. Kent, J., Coriat, A.-M., Sharpe, P.T., Hastie, N.D. and van Heyningen, V. (1995) The evolution of WT1 sequence and expression pattern in the vertebrates. Oncogene 11, 1781–1792. Kingsley, D.M. (1994) The TGF-b superfamily: new members, new receptors, and new genetic tests of function in different organisms. Genes Dev. 8, 133–146. Kispert, A., Vainio, S., Shen, L., Rowitch, D.H. and McMahon, A.P. (1996) Proteoglycans are required for maintenance of Wnt-11 expression in the ureter tips. Development 122, 3627–3637. Klein, P.S. and Melton, D.A. (1996) A molecular mechanism for the effect of lithium on development. Proc. Natl. Acad. Sci. USA 93, 8455–8459. Klein, D.J., Brown, D.M., Moran, A., Oegema, T.R.J. and Platt, J.L. (1989) Chondroitin sulfate proteoglycan synthesis and reutilization of b-d-xyloside-initiated chondroitin/dermatan sulfate glycosaminoglycans in fetal kidney branching morphogenesis. Dev. Biol. 133, 515–528. Koseki, C., Herzlinger, D. and Al-Awqati, Q. (1992) Apoptosis in metanephric development. J. Cell Biol. 119, 1327–1333. Kreidberg, J.A., Sariola, H., Loring, J.M., Maeda, M., Pelletier, J., Housman, D. and Jaenisch, R. (1993) WT1 is required for early kidney development. Cell 74, 679–691. Krumlauf, R. (1994) Hox genes and vertebrate development. Cell 78, 191– 201. Lai, E., Clark, K.L., Burley, S.K. and Darnell, J.E. (1993) Hepatocyte factor 3/fork head or ‘winged helix’ proteins: a family of transcription factors of diverse biologic function. Proc. Natl. Acad. Sci. USA 90, 10421–10423. Larsson, S.H., Charlieu, J.-P., Miyagawa, K., Engelkamp, D., Rassoulzadegan, M., Ross, A., Cuzin, F., van Heyningen, V. and Hastie, N.D. (1995) Subnuclear localization of WT1 in splicing or transcription factor domains is regulated by alternative splicing. Cell 81, 391–401. Lee, S.M.K., Dickinson, M.E., Par, B.A., Vainio, S. and McMahon, A.P. (1995) Molecular genetic analysis of Wnt signals in mouse development. Semin. Dev. Biol. 6, 267–274. Lelongt, B., Makino, H., Dalecki, T.M. and Kanwar, Y.S. (1988) Role of proteoglycans in renal development. Dev. Biol. 138, 256–276. Leveen, P., Pekny, M., Gebre-Medhin, S., Swolin, B., Larsson, E. and Betsholtz, C. (1994) Mice deficient for PDGFD B show renal, cardiovascular, and hematological abnormalities. Genes Dev. 8, 1876–1887. Lin, L.-F.H., Doherty, D.H., Lile, J.D., Bektesh, S. and Collins, F. (1993) GDNF: a glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons. Science 260, 1130–1132.

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120 Luo, G., Hofmann, C., Bronckers, A.L.J.J., Sohocki, M., Bradley, A. and Karsenty, G. (1995) BMP-7 is an inducer of nephrogenesis, and is also required for eye development and skeletal patterning. Genes Dev. 9, 2808–2820. Lyons, K.M., Hogan, B.L.M. and Robertson, E.J. (1995) Colocalization of Bmp-7 and Bmp-2 RNAs suggests that these factors cooperatively mediate tissue interactions during murine development. Mech. Dev. 50, 71– 83. Maas, R., Elfering, S., Glaser, T. and Jepeal, L. (1994) Deficient outgrowth of the ureteric bud underlies the renal agenesis phenotype in mice manifesting the limb deformity (ld) mutation. Dev. Dyn. 199, 214–228. Madden, S.L., Cook, D.M., Morris, J.F., Gashier, A., Sukhatme, V.P. and Rauscher, III, F.J. (1991) Transcriptional repression mediated by the WT1 Wilms’ tumor gene product. Science 253, 1550–1553. Maheswaran, S., Englert, C., Bennett, P., Heinrich, G. and Haber, D.A. (1995) The WT1 gene product stabilizes p53 and inhibits p53-mediated apoptosis. Genes Dev. 9, 2143–2156. Miller, J.R. and Moon, R.T. (1996) Signal transduction through b-catenin and specification of cell fate during embryogenesis. Genes Dev. 10, 2527–2539. Moore, M.W., Klein, R.D., Farinas, I., Sauer, H., Armanini, M., Phillips, H., Reichardt, L.F., Ryans, A.M., Carver-Moore, K. and Rosenthal, A. (1996) Renal and neuronal abnormalities in mice lacking GDNF. Nature 382, 76–79. Nagata, M., Nakauchi, H., Nakayama, K., Nakayama, K., Loh, D. and Watanabe, T. (1996) Apoptosis during an early stage of nephrogenesis induces renal hypoplasia in bcl-2-deficient mice. Am. J. Pathol. 148, 1601–1611. Nunez, G. and Clarke, M.F. (1994) The Bcl-2 family of proteins: regulators of cell death and survival. Trends Cell Biol. 4, 399–403. Orr-Urtreger, A., Bedford, M.T., Burakova, T., Arman, E., Zimmer, Y., Yayon, A., Givol, D. and Lonai, P. (1993) Developmental localization of the splicing alternatives of fibroblast growth factor receptor-2 (FGFR2). Dev. Biol. 158, 475–486. Pachnis, V., Mankoo, B. and Costantini, F. (1993) Expression of the c-ret proto-oncogene during mouse embryogenesis. Development 119, 1005– 1017. Patterson, L.T. and Dressler, G.R. (1994) Regulation of kidney development: new insights from an old model. Curr. Opin. Genet. Dev. 4, 696– 702. Peifer, M.D., Sweeton, D., Casey, M. and Wieschaus, E. (1994) Wingless signal and Zeste-white 3 kinase trigger opposing changes in the intracellular distribution of Armadillo. Development 120, 369–380. Perantoni, A.O., Dove, L.F. and Karavanova, I. (1995) Basic fibroblast growth factor can mediate the early inductive events in renal development. Proc. Natl. Acad. Sci. USA 92, 4696–4700. Pichel, J.G., Shen, L., Sheng, H.Z., Granholm, A.-C., Drago, J., Grinberg, A., Lee, E.J., Huang, S.P., Saarma, M., Hoffer, B.J., Sariola, H. and Westphal, H. (1996) Defects in enteric innervation and kidney development in mice lacking GDNF. Nature 382, 73–76. Plachov, D., Chowdhury, K., Walther, C., Simon, D., Guenet, J.L. and Gruss, P. (1990) Pax8, a murine paired box gene expressed in the developing excretory system and thyroid gland. Development 110, 643–651. Platt, J.L., Trescony, P., Lindman, B. and Oregema, T. (1990) Heparin and heparin sulfate delimit nephron formation in fetal metanephric kidneys. Dev. Biol. 139, 338–348. Poleev, A., Fickenscher, H., Mundlos, S., Winterpacht, A., Zabel, B., Fidler, A., Gruss, P. and Plachov, D. (1992) PAX8, a human paired box gene: isolation and expression in developing thyroid, kidney, and Wilms’ tumor. Development 116, 611–623. Poole, T.J. and Steinberg, M.S. (1981) Amphibian pronephric duct morphogenesis: segregation cell rearrangement and directed migration of the Ambystoma duct rudiment. J. Embryol. Exp. Morphol. 63, 1–16. Polyak, K., Waldman, T., He, T.-C., Kinzler, K.W. and Vogelstein, B. (1996) Genetic determinants of p53-induced apoptosis and growth arrest. Genes Dev. 10, 1945–1952.

119

Pritchard-Jones, K., Fleming, S., Davidson, D., Bickmore, W., Porteous, D., Gosden, C., Bard, J., Buckler, A., Pelletier, J., Housman, D., van Heyningen, V. and Hastie, N. (1990) The candidate Wilms’ tumor gene is involved in genitourinary development. Nature 346, 194–197. Qiao, J., Cohen, D. and Herzlinger, D. (1995) The metanephric blastema differentiates into collecting system and nephron epithelia in vitro. Development 121, 3207–3214. Rauscher, III, F.J., Morris, J.F., Tournay, O.E., Cook, D.M. and Curran, T. (1990) Binding of the Wilms’ tumor locus zinc finger protein to the EGR-1 consensus sequence. Science 250, 1259–1262. Rothenpieler, U.W. and Dressler, G.R. (1993) Pax-2 is required for mesenchyme-to-epithelium conversion during kidney development. Development 119, 711–720. Rupprecht, H.D., Drummond, I.A., Madden, S.L., Rauscher, III, F.J. and Sukhatme, V.P. (1994) The Wilms’ tumor suppressor gene WT1 is negatively autoregulated. J. Biol. Chem. 269, 6198–6206. Ryan, G., Steele-Perkins, V., Morris, J., Rauscher, III, F.J. and Dressler, G.R. (1995) Repression of Pax-2 by WT1 during normal kidney development. Development 121, 867–875. Sanchez, M.P., Silos-Santiago, I., Frisen, J., He, B., Lira, S.A. and Barbacid, M. (1996) Renal agenesis and the absence of enteric neurons in mice lacking GDNF. Nature 382, 70–73. Sanyanusin, P., McNoe, L.A., Sullivan, M.J., Weaver, R.G. and Eccles, M.R. (1995) Mutation of Pax2 in two siblings with renal-coloboma syndrome. Hum. Mol. Genet. 4, 2183–2184. Sariola, H. (1996) Does the kidney express redundant or important molecules during nephrogenesis. Exp. Nephrol. 4, 70–76. Saxen, L. (1987) Organogenesis of the kidney. In Barlow, P.W., Green, P.B. and White, C.C. (eds.), Developmental and Cell Biology Series 19, Cambridge University Press, Cambridge, UK. Saxen, L. and Saskela, E. (1971) Transmission and spread of embryonic induction II: exclusion of an assimilatory transmission mechanism in kidney tubule induction. Exp. Cell Res. 66, 369–377. Schmid, P., Lorenz, A., Hameister, H. and Montenarh, M. (1991) Expression of p53 during mouse embryogenesis. Development 113, 857–865. Schuchardt, A., D’Agati, V., Larsson-Blomberg, L., Costantini, F. and Pachnis, V. (1994) Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret. Nature 367, 380– 383. Schuchardt, A., D’Agati, V., Pachnis, V. and Costantini, F. (1996) Renal agenesis and hypodysplasia in ret-k − mutant mice result from defects in ureteric bud development. Development 122, 1919–1929. Shawlot, W. and Behringer, R.R. (1995) Requirement of Lim1 in headorganizer function. Nature 374, 425–430. Small, K.M. and Potter, S.S. (1993) Homeotic transformations and limb defects in Hox A11 mutant mice. Genes Dev. 7, 2318–2328. Sorenson, C.M., Padanilam, B.J. and Hammerman, M.R. (1996) Abnormal postpartum renal development and cystogenesis in the bcl-2 (−/−) mouse. Am. J. Physiol. 271, F184–F193. Soriano, P. (1994) Abnormal kidney development and hematological disorders in PDGF b-receptor mutant mice. Genes Dev. 8, 1888–1896. Stark, K., Vainio, S., Vassileva, G. and McMahon, A.P. (1994) Epithelial transformation of metanephric mesenchyme in the developing kidney regulated by Wnt-4. Nature 372, 679–683. Stuart, E.T., Haffner, R., Oren, M. and Gruss, P. (1995) Loss of p53 function through Pax-mediated transcriptional repression. EMBO J. 14, 5638–5645. Torres, M., Gomez-Pardo, E., Dressler, G.R. and Gruss, P. (1995) Pax-2 controls multiple steps of urogenital development. Development 121, 4057–4065. Treanor, J.J.S., Goodman, L., de Sauvage, F., Stone, D.M., Poulsen, K.T., Beck, C.D., Gray, C., Armanini, M.P., Pollock, R.A., Hefti, F., Phillips, H.S., Goddard, A., Moore, M.W., Buj-Bello, A., Davies, A.M., Asai, N., Takahashi, M., Vandlen, R., Henderson, C.E. and Rosenthal, A. (1996) Characterization of a multicomponent receptor for GDNF. Nature 382, 80–83. Trupp, M., Arenas, E., Fainzilber, M., Nilsson, A.-S., Sieber, B.-A., Gri-

120

M.S. Lechner, G.R. Dressler / Mechanisms of Development 62 (1997) 105–120

goriou, M., Kilkenny, C., Salazar-Grueso, E., Pachnis, V., Arumae, U., Sariola, H., Saarma, M. and Ibanez, C.F. (1996) Functional receptor for GDNF encoded by the c-ret proto-oncogene. Nature 381, 785–789. Uetz, P., Fumagalli, S., James, D. and Zeller, R. (1996) Molecular interaction between limb deformity proteins (formins) and Src family kinases. J. Biol. Chem. 271, 33525–33530. Valentini, R.P., Brookhiser, W.T., Park, J., Yang, T., Briggs, J., Dressler, G. and Holzman, L.B. (1997) Post-translational processing and renal expression of mouse indian hedgehog. J. Biol. Chem., in press. Vega, Q.C., Worby, C.A., Lechner, M.S., Dixon, J.E. and Dressler, G.R. (1996) Glial cell line-derived neurotrophic factor activates RET and promotes kidney morphogenesis. Proc. Natl. Acad. Sci. USA 93, 10657–10661. Veis, D.J., Sorenson, C.M., Shutter, J.R. and Korsmeyer, S.J. (1993) Bcl2-deficient mice demonstrate fulminant lymphoid apoptosis, polycystic kidneys, and hypopigmented hair. Cell 75, 229–240. Venkataraman, G., Sasisekharan, V., Herr, A.B., Ornitz, D.M., Waksman, G., Cooney, C.L., Langer, R. and Sasisekharan, R. (1996) Preferential self-association of basic fibroblast growth factor is stabilized by heparin during receptor dimerization and activation. Proc. Natl. Acad. Sci. USA 93, 845–850. Vogt, T.F., Jackson-Grusby, L., Rush, J. and Leder, P. (1993) Formins:

phosphoprotein isoforms encoded by the mouse limb deformity locus. Proc. Natl. Acad. Sci. USA 90, 5554–5558. Vukicevic, S., Kopp, J.B., Luyten, F.P. and Sampath, T.K. (1996) Induction of nephrogenic mesenchyme by osteogenic protein 1 (bone morphogenetic protein7). Proc. Natl. Acad. Sci. USA 93, 9021–9026. Wang, Z.-Y., Madden, S.L., Deuel, T.F. and Rauscher, III, F.J. (1992) The human platelet-derived growth factor A-chain (PDGF-A) gene is a target for repression by the WT1 Wilms’ tumor protein. J. Biol. Chem. 267, 21999–22002. Weller, A., Sorokin, L., Illgen, E.-M. and Ekblom, P. (1991) Development and growth of mouse embryonic kidney in organ culture and modulation of development by soluble growth factors. Dev. Biol. 144, 248–261. Winyard, P.J.D., Risdon, R.A., Sams, V.R., Dressler, G.R. and Woolf, A.S. (1996) The PAX2 transcription factor is expressed in cystic and hyperproliferative dysplastic epithelia in human kidney malformations. J. Clin. Invest., 451–459. Woychik, R.P., Stewart, T.A., Davis, L.G., D’Eustachio, P. and Leder, P. (1985) An inherited limb deformity created by insertional mutagenesis in a transgenic mouse. Nature 318, 36–40. Woychik, R.P., Maas, R.L., Zeller, R., Vogt, T.F. and Leder, P. (1990) ‘Formins’: proteins deduced from the alternative transcripts of the limb deformity gene. Nature 346, 850–853.