Architectural patterns in branching morphogenesis in the kidney
1998; Elsevier BV; Volume: 54; Issue: 6 Linguagem: Inglês
10.1046/j.1523-1755.1998.00196.x
ISSN1523-1755
AutoresQais Al‐Awqati, Michael R. Goldberg,
Tópico(s)Genetic and Kidney Cyst Diseases
ResumoArchitectural patterns in branching morphogenesis in the kidney. During kidney development, several discrete steps generate its three-dimensional pattern including specific branch types, regional differential growth of stems, the specific axes of growth and temporal progression of the pattern. The ureteric bud undergoes three different types of branching. In the first, terminal bifid type, a lateral branch arises and immediately bifurcates to form two terminal branches whose tips induce the formation of nephrons. After 15 such divisions (in humans) of this specifically renal type of branching, several nephrons are induced whose connecting tubules fuse and elongate to form the arcades. Finally, the last generations undergo strictly lateral branching to form the cortical system. The stems of these branches elongate in a highly regulated pattern. The molecular basis of these processes is unknown and we briefly review their potential mediators. Differential growth in three different axes of the kidney (cortico-medullary, dorso-ventral and rostro-caudal) generate the characteristic shape of the kidney. Rapid advances in molecular genetics highlight the need for development of specific assays for each of these discrete steps, a prerequisite for identification of the involved pathways. The identification of molecules that control branching (the ultimate determinant of the number of nephrons) has acquired new urgency with the recent suggestion that a reduced nephron number predisposes humans to hypertension and to progression of renal failure. Architectural patterns in branching morphogenesis in the kidney. During kidney development, several discrete steps generate its three-dimensional pattern including specific branch types, regional differential growth of stems, the specific axes of growth and temporal progression of the pattern. The ureteric bud undergoes three different types of branching. In the first, terminal bifid type, a lateral branch arises and immediately bifurcates to form two terminal branches whose tips induce the formation of nephrons. After 15 such divisions (in humans) of this specifically renal type of branching, several nephrons are induced whose connecting tubules fuse and elongate to form the arcades. Finally, the last generations undergo strictly lateral branching to form the cortical system. The stems of these branches elongate in a highly regulated pattern. The molecular basis of these processes is unknown and we briefly review their potential mediators. Differential growth in three different axes of the kidney (cortico-medullary, dorso-ventral and rostro-caudal) generate the characteristic shape of the kidney. Rapid advances in molecular genetics highlight the need for development of specific assays for each of these discrete steps, a prerequisite for identification of the involved pathways. The identification of molecules that control branching (the ultimate determinant of the number of nephrons) has acquired new urgency with the recent suggestion that a reduced nephron number predisposes humans to hypertension and to progression of renal failure. Research in kidney development has entered a golden age driven by the application of the powerful tools of molecular genetics. Not only have nephrologists applied these reagents to the study of kidney development, but the rise of the technology of gene deletions in mammals has led to the identification of hitherto unsuspected critical factors in renal development. Renal organogenesis has the attraction of being readily observable in vitro[1.Grobstein C. Inductive epithelio-mesenchymal interactions in cultured organ rudiments of the mouse.Science. 1953; 118: 52-55Crossref PubMed Scopus (299) Google Scholar], which allows investigators to test several hypotheses using more direct methods than is applicable to other organs. Recent excellent reviews have emphasized the role of extracellular matrix molecules, transcription factors, signaling pathways and growth factors[2.Hammerman M.R. Growth factors in renal development.Semin Nephrol. 1995; 15: 291-299PubMed Google Scholar, 3.Ekblom P. Extracellular matrix and cell adhesion molecules in nephrogenesis.Exp Nephrol. 1996; 4: 92-96PubMed Google Scholar, 4.Kanwar Y.S. Carone F.A. Kumar A. Wada J. Ota K. Wallner E.I. Role of extracellular matrix, growth factors and proto-oncogenes in metanephric development.Kidney Int. 1997; 52: 589-606Abstract Full Text PDF PubMed Scopus (69) Google Scholar, 5.Lechner M.S. Dressler G.R. The molecular basis of embryonic kidney development.Mech Dev. 1997; 62: 105-120Crossref PubMed Scopus (158) Google Scholar, 6.Davies J. How to build a kidney.Semin Cell Biol. 1993; 4: 213-219Crossref PubMed Scopus (18) Google Scholar, 7.Sariola H. Sainio K. The tip-top branching ureter.Curr Opin Cell Biol. 1997; 9: 877-884Crossref PubMed Scopus (73) Google Scholar, 8.Vainio S. Muller U. Inductive tissue interactions, cell signaling and the control of kidney organogenesis.Cell. 1997; 90: 975-978Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar]. A comprehensive database has recently been compiled by J. Davies into a readily accessible internet resource[9.http://www.mbisg2.sbc.man.ac.uk/kidbase/kidhome.html (Alternative address: http://www.ana.ed.ac.uk/anatomy/database/kidbase/kidhome.html)Google Scholar]. This should facilitate progress in this field and eventually permit the development of an atlas of gene expression in the kidney during development. However, in this arcadian scenario, a number of problems have crept up, most of which deal with discrepancies between in vitro results and those produced by gene deletions. For instance, interruption of the signaling pathways of several growth factors such as insulin-like growth factor (IGF), nerve growth factor (NGF), hepatocyte growth factor (HGF), and basic fibroblast growth factor (bFGF) convincingly inhibit renal development in organ culture[4.Kanwar Y.S. Carone F.A. Kumar A. Wada J. Ota K. Wallner E.I. Role of extracellular matrix, growth factors and proto-oncogenes in metanephric development.Kidney Int. 1997; 52: 589-606Abstract Full Text PDF PubMed Scopus (69) Google Scholar], yet, knock-out of the genes for these factors or their receptors does not lead to a "renal phenotype." Consequently, terms such as redundancy have appeared in the literature with distressing frequency[10.Sariola H. Does the kidney express redundant or important molecules during nephrogenesis.Dev Nephrol. 1996; 4: 70-76Google Scholar]. An equally likely interpretation of the discrepancy between these two kinds of studies is that the correct assay for deciphering the action of this gene is yet to be devised, a position whose advantages (aside from humility) are that it encourages the search for new assays to test the function of these genes. The development of many organs starts with folding of an epithelial sheet that progresses into the formation of a bud. Elongation of the bud forces the rudiment of this proto-organ to invade another tissue, frequently composed of mesenchyme. Eventually the bud will undergo a series of divisions. Most organs are composed of units, be they nephrons in the kidney or lobules in exocrine glands. The arborization is stereotyped in each organ, resulting in a set number of divisions and a defined number of units. However, the branching patterns are quite specific to the organ. Although we will touch on branching morphogenesis in other systems, this review focuses on the role of this process in the generation of the mammalian kidney. Development of the kidney starts when an outpouch of the wolfian duct, the ureteric bud, grows and invades a group of mesenchymal cells, the metanephric mesenchyme. We previously found that the mesenchymal cells are composed of stem cells that are capable of populating the glomeruli, proximal loop and distal segments of the nephron. In addition, they also produced non-epithelial cells that were stromal in distribution[11.Herzlinger D. Koseki C. Mikawa T. Al-Awqati Q. Metanephric mesenchyme contains multipotent stem cells whose fate is restricted after induction.Development. 1992; 114: 565-572Crossref PubMed Google Scholar]. The collecting system of the nephron forms from branches of the ureteric bud. The branching of the ureteric bud is highly structured and shows several types of repeating patterns of divisions. Each tip of the branch is capable of inducing about 100 mesenchymal cells to survive[12.Koseki C. Herzlinger D. Al-Awqati Q. Apoptosis in metanephric development.J Cell Biol. 1992; 119: 1327-1333Crossref PubMed Scopus (287) Google Scholar], proliferate and to undergo mesenchymal to epithelial transformation leading to the generation of the epithelial cells of the nephron. The induced mesenchyme in turn secretes factors that promote further growth, proliferation and division of the ureteric bud reviewed in[13.Saxen Organogenesis of the Kidney. Cambridge University Press, New York1987Crossref Google Scholar]. In addition to this reciprocal interaction, the induced mesenchyme and ureteric bud must also produce factors that control the growth, differentiation and migration of endothelial cells, mesangial and other smooth muscle and interstitial cells. These interactions will eventually form the three-dimensional architectural pattern of the adult kidney. The development of this pattern was analyzed by Jean Oliver using dissections of the branching tree in human kidneys of different embryonic ages[14.Oliver J. Nephrons and Kidneys. Harper & Row, New York, Hoeber Medical Division1968Google Scholar]. Our review will re-cast Oliver's comprehensive analysis in the language of molecular cell biology. We will describe the development of this pattern as a function of time and, where available, introduce the recently acquired molecular information that bears on these processes. Analysis of the pattern of branching is best performed prospectively by studying its development in vitro or in vivo. Retrospective analysis of the final result, by three-dimensional reconstruction, or by the use of casts, is less powerful since it will miss the presence of any remodeling events that shape the final tree. Both approaches have been used in the kidney, but the most comprehensive analysis was performed by dissection of individual nephrons from human embryonic kidneys. All branching patterns are composed of only two structural elements: branch points and stems. In addition, other factors are also critical to form the three dimensional pattern, including the axes of branching, number of generations, the stereotypy (or lack of it) of the branching pattern, its pace of development and fasciculation of the branched elements. The formation and characteristics of each of these are likely to be separately regulated. Branching mechanisms can be divided into two very large classes; in one, branching occurs only by cellular re-arrangement, while the other requires cell proliferation as well. Tremendous strides have been made in the understanding of some of the mechanism of the former, especially in the branching of the Drosophila trachea (see below for a brief description). Vertebrate organogenesis, however, uses the second mode where cell proliferation is a sine qua non. Two types of branches can be differentiated. Lateral branches (also termed monopodial) originate from a main stem at equally spaced or random intervals. Lateral branches often determine a lobule of an organ, for instance in the breast, salivary glands and pancreas. This kind of pattern is also found frequently during the development of the blood vessels including capillary networks. In bifid (or dipodial) branches the main stem divides into two daughter branches similar in size to each other. Each daughter branch can divide in a bifid manner until a terminal set of branches is reached. The angle of bifurcation is variable, but in the first bifurcation of the ureteric bud it is as large as 180°. During kidney development, the rapid appearance of bifid branching in rapid succession often results in the appearance of trifid or a four way division (carrefours). This occurs because of the rapidity with which bifid branching occurs before elongation of the stem of each branch. Similarly, although bifid branching implies that the two resultant stems are of equal bore, in fact in the kidney it had been observed that many of them are of unequal size and thickness, especially in the early days of development. The analysis of these two types of branches is often very difficult, and is ideally performed in a time lapse study of an individual branch. The following morphological characteristics of the branch need to be delineated: is it always preceded by an ampulla ballooning out of the site of branching; is there enhanced cell division to generate the ampulla, or is the ballooning due to rearrangement of the epithelium of the ureteric bud or attraction of cells from the mesenchyme, as has been recently proposed[15.Qiao J. Cohen D. Herzlinger D. The metanephric blastema differentiates into collecting system and nephron epithelia in vitro.Development. 1995; 121: 3207-3214PubMed Google Scholar]. Formation of the characteristic system of the kidney requires both lateral and bifid branches, which then form three different patterns of branching that unfold one after the other. Each pattern bears a specific quantitative relation to the others. The number of branches in a tree composed of a simple repeating unit of bifid branches, where the tip of each bifid branch divides into another bifid branch is equal to 2n, where n is the number of bifid divisions; we term this an iterative bifid branching system. Figure 1a. However, microdissection of human embryonic kidneys have shown that this simple pattern does not occur in the kidney. This follows from the fact that each tip of the branch induces and becomes connected to a nephron, and that connection removes the tip from the possibility of further branching. The growth and division of the ureteric bud proceeds in two steps that occur in rapid succession. The first is a lateral branch that then divides into two terminal branches Figure 1b, referred to as a terminal bifid branching system (terminal, because each tip does not generate a further bifid branch). Micro-dissection of nephrons shows that this pattern develops rapidly such that it is sometimes not possible to see a pure lateral branch followed by a bifurcation at its tip. Rather, a frequent image is a ballooned ampulla that starts to bifurcate and to elongate in three axes, one to form a lateral branch and the other two to form the yoke of a bifid branch Figure 1b. This type of branching process is unique to the kidney and is repeated 15 times during human nephrogenesis. The same pattern is seen from the beginning. the budding of the ureteric bud from the wolffian duct is the first lateral branch that then divides into two producing the first bifid branch point. The stem carrying each tip elongates, as first suggested by Peter[16.Peter K. Unterssuchungen uber Bau und Entwicklung der Niere. Gustav Fischer, Jena1909Google Scholar], transporting it and its induced nephron towards the developing cortex. From the shaft of the elongated branch a new ampulla appears, which first grows to form a lateral branch that then gets eventually transformed into the yoke of a new bifid division capable of inducing two new nephrons Figure 1b. As can be shown in Figure 1b, the number of branches (nephrons) that results from this pattern would be equal to 2(2n) - 2, where n is again the number of bifid branches. This pattern has been called the closed divided system of branching by Oliver. For 15 of these terminal bifid divisions, the number of resulting branches is only 65,534 nephrons, clearly much lower than the number of nephrons in one human kidney. The terminal bifid pattern of branching is unique to the kidney and is a consequence of the fact that the branching points induce a nephron and hence no longer can divide, the tree forms by having the stem produce another lateral bud. Based on the dissection of early embryonic kidneys, it became apparent that the first five divisions and generations of ureteric bud undergo transformation into the pelvis and calyceal system by increased growth, and dilation of these early generation of tubules into the appropriate number of calyces. In the human kidney, there is an average of 8 papillae; into each papilla an average of 44 ducts of Bellini open. In some animals, there is a single papilla while in others the papillae fuse to form one complex papilla or even a ridge of papillary tissue (crests). This raises the question of whether these first few generations of branching are similar to the later ones; for instance, do they induce nephrons also or do they simply dilate to form the calyces? Oliver refers to studies performed by Kampmeier[17.Kampmeier O.F. Studien uber die Entwicklunggeschichte der bleibende Niere beim Menschen.Ztschrf Ges Anatomie I. 1924; 73 (Abt Bun): P 459Crossref Scopus (2) Google Scholar], who dissected two-month-old human embryonic kidneys and found that the early branches indeed induced nephrons, which in all likelihood were transported up to the cortex by differential elongation of the branches. The remainder of the nephrons develop by two additional methods. After the 15th division has occurred (sometimes after the 13th or 14th) several nephrons get induced simultaneously around the stem of the elongating branch. The connecting tubules of each of these nephrons then join the branch close to each other to form an arcade Figure 2a. There is some variation in the arcade system; as many as half of the late branches do not have arcades and the rest have one to five nephrons per arcade, but on average there are three nephrons per arcade. In the adult kidney, arcades are located in the deep cortex. After the arcades form, the terminal branch of the 15th generation begins to elongate and to develop a succession of ampullae whose shape is different from those that occurred during the earlier phase of bifid branches in that they are less ballooned, more pointed and sometimes triangular. These ampullae occur near the tip, but the branch continues to elongate into the cortex leaving the ampullae to induce nephrons on each side of the terminal branch Figure 2b. These branches are clearly lateral (monopodial) branches and have been termed the open divided system by Oliver. By the time this branch terminates in two final branches, an average of 10 nephrons have been induced per cortical system. When these numerical averages are used, one million nephrons per kidney are generated, which is the expected number (but see below for a further discussion of the number of nephrons). The molecular and cellular basis of the three mechanisms of branching in the kidney require elucidation. Whether the differences among them represent minor variations on one general theme or are the consequences of three fundamentally different mechanisms remain to be determined. There is certainly no reason to conclude at present that they will have the same mechanism. Different patterns of branching are not limited to the kidney. They also exist in the lung, where after a certain number of bifid branches occur, the terminal bronchiole opens into the alveolar sac that is composed of many cavities that open into a common atrium. To form the lateral and bifid branching structures, cells must organize themselves into tubes and a lumen must be created. In addition, branch points must be created and anchored. Controls must exist that regulate the sequence of branching events and the length and diameter of each branch. To develop an ampulla and later a branch, cells in that area have to escape from the restriction imposed by their cell adhesion molecules and by the basement membrane, and they then have to proliferate. Elongation of the branch, on the other hand, must represent a different process whereby within the general inhibitory confines of a sheath of ECM proteins, the appropriate matrix is provided for proliferation to occur. The first morphological evidence of branching is the appearance of the ampulla, which is a group of cells that appear to have rapidly proliferated in one restricted area of the tubule. Presumably a similar process occurs when the ureteric bud appears from the side of the wolffian duct. Specification of a region such as the ampulla implies that a specific transcription factor has been activated, and one such candidate is the Pax2 gene[18.Dressler G.R. Deutsch U. Choudhry K. Nones H. Gruss P. Pax2, a new murine paired box containing gene and its expression in the developing excretory system.Development. 1990; 109: 787-795PubMed Google Scholar],[19.Torres M. Gomez-Pardo E. Dressler G.R. Gruss P. Pax-2 controls multiple steps of urogenital development.Development. 1995; 121: 4057-4065Crossref PubMed Google Scholar]. In situ hybridization demonstrated that the ret receptor tyrosine kinase is highly expressed in a localized manner only in the ampulla and later in the tips of branches[20.Pachnis V. Mankoo B. Costantini F. Expression of the c-ret proto-oncogene during mouse embryogenesis.Development. 1993; 119: 1005-1017Crossref PubMed Google Scholar]. Recent studies also suggest that a cell adhesion molecule (Ksp-cadherin; for kidney specific cadherin) is not expressed in the cells of the ampulla but is present in the shaft of the dividing tubules[21.Thomson R.B. Biemesderfer D. Aronson P.S. Developmental regulation of Ksp-cadherin expression in rabbit kidneys.J Am Soc Nephrol. 1995; 6 (abstract): 711Google Scholar]. Electron microscopic examination of that region has demonstrated that the basement membrane is no longer continuous[13.Saxen Organogenesis of the Kidney. Cambridge University Press, New York1987Crossref Google Scholar],[15.Qiao J. Cohen D. Herzlinger D. The metanephric blastema differentiates into collecting system and nephron epithelia in vitro.Development. 1995; 121: 3207-3214PubMed Google Scholar]. Further, it appears that the induced mesenchyme expresses metalloproteinases, while the shaft of the branch expresses high concentrations of the tissue inhibitor of metalloproteinase-2[22.Barasch J. Yang J. Qiao J. Herzlinger D. Oliver J.A. Tissue inhibitor of metalloprotienase-2 (TIMP-2) is a metanephrogenic mesenchymal (MM) growth factor.J Am Soc Nephrol. 1997; 8 (abstract): 357AGoogle Scholar],[23.Lelongt B. Trugnan G. Murphy G. Ronco P.M. Matrix metalloproteinases MMP2 and MMP9 are produced in earlystages of kidney morphogenesis but only MMP9 is required for renal organogenesis in vitro.J Cell Biol. 1997; 136: 1363-1373Crossref PubMed Scopus (150) Google Scholar]. Hence, regulation of the integrity of the extracellular matrix and cell to cell adhesion by secreted or membrane bound factors could be critical in determining the region of the stem that will sprout a lateral branch. Induction of the metanephric mesenchyme is the central question in kidney development and the inducing unit, the ampulla of the ureteric bud is also the region of the branching point. Whether the inductive process is mediated by the same transcriptional program as that of branching remains to be determined. Induction occurs by a process of reciprocal exchange of information. The tip of the ureteric bud instructs the metanephric mesenchyme to change its cell type and the induced mesenchyme instructs the ureteric bud tip to grow and perhaps also to divide further. This interaction is reminiscent of that occurring in a synapse, where both pre- and post-synaptic cells send signals to produce strengthening of the synapse. Like the synapse, it is likely that this interaction involves many molecules. Because of the location of RET[20.Pachnis V. Mankoo B. Costantini F. Expression of the c-ret proto-oncogene during mouse embryogenesis.Development. 1993; 119: 1005-1017Crossref PubMed Google Scholar], in the tips of the inducing bud and the absence of kidneys when this gene is deleted[24.Schuchhardt A. D'agati V. Larsson-Blumberg L. Costantini F. Pachnis V. Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret.Nature. 1994; 367: 380-383Crossref Scopus (1416) Google Scholar], it is likely that this receptor tyrosine kinase is central to this phenomenon. GDNF is produced by the metanephric mesenchyme and acts as the ligand for RET[25.Robertson K. Mason I. The GDNF-RET signaling relationship.Trends Genet. 1997; 13: 1-3Abstract Full Text PDF PubMed Scopus (72) Google Scholar], and its deletion produces essentially the same phenotype as ret knockout, that is, no kidneys. This is the first identified loop of reciprocal induction, the ureteric bud produces growth factors that rescues the GDNF-producing mesenchyme from apoptosis and GDNF induces new branching. In another loop, a mesenchymal metalloproteinase with ureteric TIMP2 was recently identified[22.Barasch J. Yang J. Qiao J. Herzlinger D. Oliver J.A. Tissue inhibitor of metalloprotienase-2 (TIMP-2) is a metanephrogenic mesenchymal (MM) growth factor.J Am Soc Nephrol. 1997; 8 (abstract): 357AGoogle Scholar]. Because deletions of WNT4 and BMP7 also lead to severe defects in kidney formation, it is important to analyze these mutations for the stages at which they exert their effects. WNT4 deletion leads to a phenotype that is different from that of RET[26.Stark K. Vainio S. Vassileva G. McMahon A.P. Epithelial transformation of metanephric mesenchyme in the developing kidney regulated by Wnt-4.Nature. 1994; 372: 679-683Crossref PubMed Scopus (896) Google Scholar]; there seem to be at least the beginnings of an inductive process and a few nephrons form, but very rapidly the processes of development stop. Mice lacking BMP7 have gross defects in kidney development that become apparent soon after the initial induction, although it seems that there is some variability in the pattern of the defect[27.Dudley A.T. Lyons K.M. Robertson E.J. A requirement for bone morphogenetic protein-7 during development of the mammalian kidney and eye.Genes Dev. 1995; 9: 2795-2807Crossref PubMed Scopus (957) Google Scholar],[28.Luo G. Hofmann C. Bronckers A.L. Sohocki M. Bradley A. Karsenty G. BMP-7 is an inducer of nephrogenesis, and is also required for eye development and skeletal patterning.Genes Dev. 1995; 9: 2808-2820Crossref PubMed Scopus (872) Google Scholar]. It had been demonstrated that addition of antibodies or anti-sense RNA to some growth factors (such as, IGF[29.Hammerman M.R. Rogers S.A. Ryan G. Growth factors and metanephrogenesis.Am J Physiol. 1992; 262: F523-F532PubMed Google Scholar] or HGF[30.Schmidt C. Bladt F. Goedecke S. Brinkmann V. Zschiesche W. Sharpe M. Gherardi E. Birchmeier C. Scatter factor/hepatocyte growth factor is essential for liver development.Nature. 1995; 73: 699-702Crossref Scopus (1223) Google Scholar]) or receptors (such as NGFR[31.Sariola H. Saarma M. Sainio K. Arumae U. Palgi J. Vaahtokari A. Thesleff I. Karavanov A. Dependence of kidney morphogenesis on the expression of nerve growth factor receptor.Science. 1991; 254: 571-573Crossref PubMed Scopus (174) Google Scholar]) results in cessation of renal development, and yet knockout of the genes has no effect on nephrogenesis. To reconcile these differences, quantitative studies need to be performed to assay nephron number, pattern of branching, pattern of lengthening and other functions discussed above. The stem, the length between each branch point, is a critical generator of the architectural pattern. One of the major advances in renal embryology was the identification of the differential growth of a zone of the stem that lies immediately adjacent to the inducing tip[16.Peter K. Unterssuchungen uber Bau und Entwicklung der Niere. Gustav Fischer, Jena1909Google Scholar]. The growth of this "intercalated" zone transports the newly induced nephron towards the future cortex. An examination of the length of the stems of the collecting system shows that the earlier generations have shorter stems than the later ones. Oliver performed a detailed analysis of the rate of growth of these branches as a function of time and found that after the fourth month of embryonic life, the length of the 10th to 15th generation increase dramatically and continues to increase even after birth. The length of the earlier generation increases at a much slower rate. However, dissection of individual kidneys at different dates showed that the length of any single generation is not predictable. In some kidneys one generation (such as the 11th) has a very high rate while another generation (such as the 14th) was much slower; in another kidney the reverse may be true. This randomness was ascribed by Oliver to be largely due to the location of the segment in the kidney. Segments that happened to be present in the central medulla (that is, between future cortex and future papilla) grew very rapidly, while those that were located in the inner medulla lengthened very little. After birth, the rates of growth of the central medulla continued to be very high, but now branches present in the cortex also lengthened at fast rates. These findings demonstrate the highly regulated nature of differential growth of the branching tree and raise important questions regarding its mechanism. The observed pattern of lengthening implies the existence of gradients of external growth promoting activity that are low in the inner medulla and high in the central medulla and eventually cortex. During elongation, branching must be inhibited but proliferation between branch points must increase. What is the growth factor that causes proliferation and what inhibits branching in the lengthening branch? Is it merely the absence of GDNF/c-ret, to mention only one set of candidates, or is there a specific inhibitor of a signaling molecul
Referência(s)