&p.1:Abstract This review highlights the important roles themesonephros may play in development. In the ovine fe-tus it is an excretory and endocrine organ and may con-tribute to the formation of normal gonads and adrenals.The metanephros of the ovine fetus has the importantfunction of providing large quantities of dilute urine forthe maintenance of amniotic and allantoic fluid volumes,essential for normal placentation and development.
&kwd:Key words Mesonephros · Metanephros ·Nephrogenesis · Renin-angiotensin&bdy:
Introduction
Functioning kidneys are essential for life after birth and,although the fetus can survive without kidneys (renalagenesis), development is compromised. There are manyreasons why the study of kidney development is impor-tant. Recently it has been proposed that many diseasestates of the adult, including hypertension and kidneydisease, may be determined by events that occurred dur-ing fetal development [1]. Maternal treatment with somedrugs (such as angiotensin converting enzyme inhibitors)can alter fetal kidney function and cause developmentalabnormalities. Babies that are growth retarded may havesmall kidneys, even when corrected for body weight,which may in later life compromise renal function [2].This makes it very important to understand normal kid-ney development and identify factors that may influencerenal growth and functioning.
The fetal metanephric kidney produces a large volumeof dilute urine which is a major input into the amnioticfluid [3]. The amniotic fluid is essential as an aqueousenvironment, for the symmetrical growth of the fetus,and correct lung development. Any factor that preventsurine being produced by the kidneys could result in lack
of amniotic fluid and subsequent fetal growth deformi-ties. In addition, the kidney produces hormones, some ofwhich may act locally and some that affect other devel-oping organs or systems.
This review examines the normal growth and functionof the mammalian meso- and metanephros. There havebeen a number of recent reviews describing the morpho-logical development of the metanephros and factors im-portant for correct structural formation [3–6]. This re-view shall focus more on mesonephric development andfunctions of the mesonephros and metanephros in the fe-tus. Information from the human is described whereavailable, but most experimental studies examining func-tion have been conducted on the fetal lamb or the fetaland neonatal mouse. The fetal sheep provides a goodmodel for many aspects of nephrogenesis, because thetiming and development of each set of kidneys is similarto the human (Table 1). Thus, much of the work de-scribed in this review is from studies in the chronicallycannulated ovine fetus.
Nephrogenesis – an overview
During embryonic and fetal development in mammals,there are three different pairs of renal organs. The firsttwo, the pronephros and the mesonephros, exist, in mostcases, for defined periods of intrauterine development.These organs regress and the third organ, the metaneph-ros, becomes the permanent adult kidney [3]. The pro-nephros is the most primitive and is not thought to befunctional in mammals during embryogenesis. The me-sonephros is a functional kidney in some species, but thetubules lack a loop of Henle.
The metanephros, which becomes the permanentadult kidney, first appears some time after the mesoneph-ros, but for a period of gestation the two sets of kidneysco-exist. Development of each set of kidneys is consid-ered below.K.M. Moritz · E.M. Wintour (✉)
Howard Florey Institute of Experimental Physiology and Medicine, University of Melbourne, Parkville, 3052, AustraliaTel.: +61-3-93444664, Fax: +61-3-93481707&/fn-block:
Pediatr Nephrol (1999) 13:171–178 © IPNA 1999
D E V E L O P M E N TA L B I O L O G Y R E V I E W
&roles:Karen M. Moritz · E. Marelyn Wintour
Functional development of the meso- and metanephros
&misc:Received: 16 March 1998 / Revised: 7 May 1998 / Accepted: 7 May 1998
The pronephros
In mammals, the pronephros is quite rudimentary andnon-functional, but in many lower vertebrates, such asamphibia and fish, the pronephros acts as the embryonickidney and is essential for survival [7]. In the proneph-ros, after filtration through the glomus, fluid enters thecoelomic cavity also known as the nephrocoel. Fromthere the filtrate is collected via ciliated funnels (neph-rostomes) that are connected to the pronephric tubules.Surrounding these tubules is a blood sinus into which re-absorbed fluid passes, whilst unabsorbed fluid is excret-ed via the pronephric (Wolffian) ducts. The sheep doesnot develop a pronephros as such, but does form a giantglomerulus at the cranial end of the future mesonephros[8]. Many genes known to be critical for meso- and/ormetanephric induction are also expressed by the pro-nephros. The expression patterns of these genes in thepronephric kidney is similar to that observed in the me-so- and metanephros, suggesting a role for these genes inthe growth and development of all three kidney types [7].
The mesonephros
In the human embryo at day 24–26, mesonephric ductsform on the lateral and ventral sides of the nephrogenicridge and induce formation of the mesonephros [9]. Byday 28, these ducts join the cloaca and by 8 weeks postconception, the human mesonephros has reached maxi-mal size and starts to regress. Complete regression oc-curs by week 16 [10]. Although present in most other an-imals, it is interesting to note that mesonephric develop-ment and complexity varies significantly between spe-cies (Table 1). In some species the mesonephros appearsto have an excretory capacity, whilst in others is clearlynon-functional. The mesonephros of the sheep [11] ex-ists for a very similar period of gestation to that of thehuman, whereas the rodent mesonephros is present muchlater and for a shorter period of gestation [12]. Generally,the glomeruli of the mesonephros are relatively largecompared with those of the metanephros (Fig. 1B, C) butthere are many fewer of them (between 10 and 50 perkidney). Several marsupial species have been well stud-ied as the young are born with only a mesonephros pres-ent, and in these species the mesonephros is present for asignificant time after birth [13]. The elephant fetus,which has a gestation of 22 months, has a mesonephros
containing nephrostomes. As described above, these areducts that connect the coelomic cavity with the renal glo-meruli and were thought to exist only transiently in thecranial portion of the mesonephros before glomeruli for-mation. Their presence throughout mesonephric growthand degeneration may reflect the elephant’s aquatic an-cestry [14]. In both the mouse and rat [15], as well aschick [16], the cranial and caudal nephrons of the meso-nephros have been shown to vary in both morphologyand gene expression, suggesting that regulation of devel-opment may differ from the two parts of the mesoneph-ros. An example of this is seen in mice that are null mu-tant for the Wilms tumor gene product (WT-1). Thesemice develop cranial mesonephric tubules but not caudaltubules [15].
The metanephros
The metanephros, which is the permanent kidney inmammals, begins formation as a budding from the cau-dal end of the mesonephric duct (for detailed descriptionof human nephrogenesis see [10]). This first event re-quires an induction signal from the surrounding meta-nephric mesenchyme. Although the signal(s) have notbeen clearly elucidated, it has been shown that uretericbud cells contain receptors for a number of growth fac-tors. After the ureteric bud has formed, reciprocal induc-tive interactions occur. The ureteric bud causes the meta-nephric mesenchyme to differentiate and form nephrons,whilst the metanephric mesenchyme causes the uretericbud to grow and bifurcate to form collecting ducts. Thedifferentiation of the metanephric mesenchyme involvesat least two components: apoptosis is stopped and themetanephric mesenchyme differentiates into stem cells atthe periphery of the kidney, which forms the nephrogeniczone and which can then epithelialize [4].
The ureteric bud dilates at its growing tip and formsan ampulla. Cells from the nephrogenic ridge clusteraround this ampulla and are known as the metanephriccap or blastema. Induction signals from the ampullacause the metanephric blastema to condense and form aclosed tube of epithelial cells, the nephrogenic vesicle.This becomes a rounded structure and the cells prolifer-ate to form a comma- and then S-shaped body. The ure-teric bud undergoes a series of divisions in which theampulla divides with one part going to induce a nephro-genic vesicle and the other to divide again. By this pro-
172
Table 1 Meso- and metaneph-ric development in differentspecies. Post-partum (pp)&/tbl.c:&tbl.b:
Species Length of Mesonephric Percentage Metanephric Percentagegestation development gestation development of gestation
Human 40 weeks 4–16 10–40 5–36.5 12–90Sheep 145–152 days 17–57 11–38 27–135 18–90Pig 112–114 days 15–24 9–21 20-pp3 19–118Mouse 20 days 10–13 53–70 11-pp3 55–113Rat 21 days 12–17 57–81 12-pp8 57–138Elephant 92 weeks <13–>24 <14–>26 14–? 15–?Native cat 21 days –pp30 –240 pp2–pp89 110–520
&/tbl.b:
cess, many hundreds of thousands of nephrons may beformed [4]. Nephrogenesis is complete before birth inhuman and sheep, but not in many other species (Table1).
Many gene products have been implicated in meta-nephric development and well reviewed in recent years[3–6]. These include a wide variety of growth factors(GF) and their receptors (insulin-like, epidermal, trans-forming, nerve, hepatocyte, fibroblast, platelet-derived,and vascular-endothelial GFs, bone morphogenetic pro-teins, leukocyte inhibitory factor), proto-oncogenes, tran-scription factors, and suppressor genes (n-myc, PAX2and8, WT1, Lim 1, Wnt 4, bcl 2, hepatocyte nuclear factor),extracellular matrix (ECM) components and degradingenzymes (collagen, fibronectin, tenascin, proteoglycans),and ECM ligands and receptors (integrins, cell adhesionmolecules). There appears to be a certain degree of re-dundancy in that some factors identified during in vitrostudies as being of crucial importance in normal devel-opment do not cause abnormalities when the gene isknocked-out in vivo [17].
As the ureteric bud branches from the Wolffian ductand is the source of the inducers of metanephric mesen-chyme condensation, no metanephros forms without me-sonephros and Wolffian duct formation first. However, sofar only a few genes have been studied specifically inmesonephric development. Abnormalities occur whenthe PAX-2 or Lim 1 gene has been “knocked-out” (nomesonephros formation) or the WT1gene is deleted (ab-normal mesonephros forms [7]). Metanephric kidneyswith serious morphological and functional deficits occurin the absence of a normal functioning renin-angiotensinsystem (RAS) [12], or one prostaglandin-synthesizingenzyme – cyclo-oxygenase 2 [18].
Possible functions of the mesonephros (Table 2)
Excretion
The ability of the mesonephros to function as an excreto-ry organ in some species may be related to the type ofplacentation. Those species with a more-primitive pla-centa (such as the pig and sheep) have the most highlydeveloped mesonephros [19]. Ovine fetuses can producea hypotonic urine from day 18 of gestation. This urinepasses via the urachus into a second fluid-filled sac, theallantois. The allantoic fluid in ovine fetuses at 20 daysof gestation is approximately 3 ml, but by day 24 there is20 ml of fluid (E.M. Wintour, unpublished observations),suggesting the mesonephros has a well-developed excre-tory capacity by 3 weeks of gestation. Expansion of the
173
Fig. 1 A The ovine fetus at 27 days’ in the region of the dorsalaorta (da), gonad (go), and mesonephros (ms). Note the associa-tion between the gonad and mesonephros, ×40, bar=250 µM,B Ovine mesonephros and C metanephros at 41 days’, ×100,bar=100 µM. Note the difference in size of glomeruli (g). pt, Prox-imal tubule; cd collecting duct; nz, nephrogenic zone&/fig.c:
allantoic membrane is important for establishing contactbetween fetal blood vessels and the maternal uterine cir-culation at specific sites, known as caruncles. Thus me-sonephric function is important for initial placentation insheep.
Erythropoiesis
In the primitive embryo there is a region termed theAGM – the region of the dorsal aorta, gonad, and meso-nephros (Fig. 1). An interesting study has recently iden-tified this area as the source of definitive hematopoiesis,i.e., the pre-liver site of hematopoietic activity [20]. Themesonephros is known to be the site of erythropoiesis inthe fish (where it is the permanent kidney) and the genefor erythropoietin (EPO) is expressed in ovine meso-nephros [21, 22].
Contribution to gonad and adrenal
In the development of the ovary and testis, cells of themesonephros clearly play a part [19]. The gonad beginsformation as a thickening in the middle of the nephro-genic ridge. This “genital ridge” becomes elongated andcovered with coelomic epithelium and soon develops alongitudinal duct lateral to the mesonephric duct. Nearly20 years ago Zamboni et al. [23] observed in the sheepfetus that at day 24–29 post conception cells from thecranial third of the mesonephros migrated outside theglomerulus and formed clusters. These clusters were ob-served all the way into the genital ridge and became as-sociated with primordial germ cells. After sexual differ-entiation at 31 days’ in the female fetus, the movementof cells became more organized and pronounced, suchthat a compact mass of cells extended from the meso-nephros into the ovary. The cell mass then breaks up andassociates with germinal cells.
In the male, although development of the testis canoccur in the absence of the mesonephros, many cords failto differentiate correctly [24]. These authors showed inisolated tissue that from day E11.5 cells migrate into thedifferentiating testis from the mesonephric region. If themesonephros was removed or separated by a filter fromthe testes, then normal cords did not form.
Cells of the mesonephros may also be involved ingrowth of the adrenal gland. Morphological studies indi-
cate that cells from the mesonephros migrate into the ad-renal cortex [25]. It has also been proposed that a uniquepopulation of mesenchymal cells gives rise to both thegonad and adrenal [26].
Role in limb development
It has been proposed that the mesonephros is importantfor limb development in the chick [27]. In particular, itwas thought that fibroblast-growth factor 8 (FGF-8) pro-duced in the mesonephros induced wing budding [28].However, this has now been disproved by elegant studiesfrom Fernandez et al. [29]. They were able to block de-velopment of the mesonephros by mechanical arrest ofthe caudal extension of the Wolffian duct. The mesodermin the mesonephric area did not express the FGF-8 gene,but limb development was completely normal. This is agood example of how careful one must be in the inter-pretation of the functional consequences of gene local-ization studies.
Endocrine function
The mesonephros expresses the gene for erythropoietinand all components of the RAS, as detailed later.
Special features of metanephric function in utero
The urine produced by the fetus is an important compo-nent of amniotic fluid and any factor that alters fetalurine production in the long term can seriously affectamniotic fluid volume. Recent studies show that bilateralnephrectomy in the ovine fetus at 100 days’ results insome changes in fetal plasma composition over the 2weeks following nephrectomy. Fetal plasma chloride de-creased, whilst phosphate, magnesium, and creatinine in-creased (Fig. 2).
Metanephric function in the fetus differs from that inthe adult in that there is a lower glomerular filtration rate(GFR) (approximately half that of the adult) and renalblood flow is only 3% of cardiac output, compared with25% in the adult [30]. By mid-gestation in the sheep, it ispossible to chronically cannulate the fetal bladder. Basalurine flow rate at this age (75 days’) is 5–6 ml/h, even
174
Table 2 Features of the meso-nephros&/tbl.c:&tbl.b: Structural features No loops of Henle
Small number of relatively large glomeruliSpecies variations Permanent kidney of fish
Variable complexity: pig>sheep>human>ratPresent at different gestational periods (see Table 1)
Excretory function Present in some species – sheep, pig – where fluid contributesto allantoic fluid
Genes expressed Erythropoietin, Na/K ATPase, calbindin D28K, growth hormone receptor, renin, angiotensin converting enzyme, angiotensinogen, AT1 and AT2 receptors, WT-1, PAX-2 and 8
Other possible functions Role in gonad and adrenal development
&/tbl.b:
though the fetus only weighs 150–250 g. This equates tonearly 0.5 l urine/kg body weight per day, which in anadult would clearly represent a case of diabetes insip-idus. The urine during development is always hypotonicunless the fetus is severely stressed [3]. The productionof a relatively large volume of hypotonic urine is essen-tial for normal fetal fluid maintenance. It is possible forthe fetus to maintain this urine flow because: (1) the kid-ney is relatively insensitive to arginine vasopressin(AVP) and (2) AVP release is relatively insensitive to os-motic stimulation [31]. The relative insensitivity to AVPreflects to some extent low expression of the gene encod-
ing the water channel, aquaporin 2 (AQP 2), rather than alack of functioning V2 receptors.
AQPs are a recently cloned family of specific waterchannels. AQP 1 is the water channel found in the proxi-mal tubules and in the thin descending limb of the loopof Henle [3]. The expression of the gene for AQP 1 canbe detected in the metanephric kidney of ovine fetusesfrom 41 days of gestation, but not in the mesonephros atthis time [32]. Gene expression increased sevenfold from60 to 140 days of gestation and reached adult levels by 6weeks after birth [33]. Dexamethasone treatment for sev-eral days at mid-gestation, previously shown to acceler-ate morphological development [34], also increased AQP1 expression at this time. In the human fetus, AQP 1 isseen in the metanephros from week 15 of gestation andAQP 2 from as early as week 12 [35].
Urine cannot be concentrated unless the specific wa-ter channels, encoded by the AQP 2gene, are present onthe apical surface of the principal cells of the corticaland medullary collecting ducts. The gene for AQP 2 isregulated by AVP and is expressed in the ovine fetal kid-ney, where mRNA can be detected by polymerase chainreaction from about 40 days of gestation in the meta-nephros. Levels increase over gestation, however theyare only 5% of adult levels at mid-gestation and 41% ofadult levels by term (E.M. Wintour, unpublished obser-vations). The slow acquisition of these water channelsaccounts for some of the insensitivity to AVP.
In the ovine fetus during the last third of gestation,stress, such as hemorrhage or hypoxia, will cause releaseof AVP, which will decrease urine flow rate and cancause the urine to become hypertonic. The maximal uri-nary osmolality, however, is much less than in the adult[36]. Many other factors have been identified to increaseurine flow. Despite the very high urine flow, flow ratecan actually be increased without affecting osmolality,three- to fivefold by high levels of glucocorticoids, an-giotensin II, or atrial natriuretic peptide during the lastthird of gestation [4, 36].
Development of metanephric endocrine function
The kidney, in addition to its excretory function, is animportant endocrine organ in utero. Major hormonesproduced by the kidney include EPO and the metaneph-ros also expresses components of the RAS. Thus angio-tensin II can be produced locally and may have effectson development and function of the metanephros.
Erythropoietin
Although the kidney has long been recognized as the ma-jor site of EPO production in the adult, for many yearsthe liver was thought to be the predominant site of pro-duction in the fetus, at least until close to term. However,recent molecular and hybridization histochemistry stud-ies have demonstrated that the EPO gene is expressed
175
Cre
atin
ine
(mm
ol/l
)
∗∗∗ ∗∗∗ ∗∗∗ ∗∗∗∗∗∗
∗∗∗
∗
70
Cl
(mm
ol/l
)
∗∗∗∗∗∗ ∗∗∗ ∗∗∗
Mg
(mm
ol/l
)
∗∗∗∗
∗∗∗ ∗∗∗
PO
4 (m
mol
/l)
∗ ∗∗∗∗ ∗∗∗ ∗∗∗ ∗∗∗
80
90
100
110
120
∗∗
∗∗∗
0.5
1.0
1.5
2.0
1
2
3
4
A
B
0.10
0.20
0.05
0.15
∗
Sample
0 1 2 3 4 5 6 7 8
C
D
Fig. 2 Changes in: A chloride (CI), B creatinine, C phosphate(PO4), and D magnesium (Mg) in ovine fetal plasma for 2 weeksfollowing nephrectomy. Numbers represent time after surgery:1=4 h, 2=48 h, 3=4 days, 4=7 days, 5=9 days, 6=11 days, and7=14 days. Squaresindicate control (intact) fetuses, triangles re-present fetuses which had undergone bilateral nephrectomy.*P<0.05, **P<0.01, ***P<0.001
strongly in the meso- and metanephros from as early as40 days of gestation in the sheep [21]. In fact, metaneph-ric expression of the EPO gene was highest from 60–100days of gestation and declined markedly close to term.At all ages, in both the meso- and metanephros, EPOmRNA was expressed in interstitial cells in the vicinityof the proximal tubules. Metanephric expression of theEPO gene can be altered by changes in glucocorticoidstatus, with dexamethasone or cortical infusions causinga significant decrease in expression in the first two-thirdsof gestation, whilst adrenalectomy caused an increaseclose to term [21]. It is interesting that neonates withsome congenital kidney diseases have normal EPO pro-duction and some infants with renal agenesis actuallyhave elevated serum EPO concentrations. This may indi-cate that the liver is able to compensate for the lack ofkidney EPO production under normal circ*mstances. Inovine fetuses that had been nephrectomized at 100 daysof gestation (term=150 days), basal EPO concentrationswere maintained for 2 weeks, although levels did not in-crease with the onset of hypoxia (K.M. Moritz, E.M.Wintour, unpublished observations). When nephrectomi-zed fetuses hemorrhaged, liver expression of the EPOgene only increased to the same degree as in intact hem-orrhaged fetuses, suggesting the liver is unable to com-pensate in times of hemorrhagic stress.
Renin-angiotensin system
The RAS is present and active during fetal life. It isthought that the major role of this system in the fetus isto maintain fetal GFR and ensure that large volumes ofurine are produced [30]. Evidence for this comes from awide range of animal experiments and clinical studies[12]. Much of the work in this area has been done in thedeveloping rat kidney [37], but there is now recent evi-dence from the human and sheep. Knock-out studies inmice have also implicated the RAS as being necessaryfor the normal growth and development of the metaneph-ric kidney.
Although the RAS operates systemically, the kidneyis able to produce all components of the system and thusthe local (intra-renal) production of angiotensin II maybe very important. Renin mRNA can be detected in thehuman mesonephros at about 30 days of gestation [38]and in the sheep from at least 40 days’. Expression isfirst observed in blood vessels outside the kidney andthen becomes more widespread throughout the renal ar-terial vasculature. In the metanephros, renin is first re-ported in the human at 56 days’ and the sheep at 41days’ [38, 39]. A similar profile of expression is seen forangiotensinogen, which is first observed in proximal tu-bules of the human mesonephros at 25–30 days of gesta-tion and in the metanephros from 56 days’. The mRNAfor angiotensin converting enzyme is also detected in themeso- and metanephros of the human and sheep fromvery early in gestation in proximal tubules and collectingducts [39]. Finally, the developing meso- and metaneph-
ros express the genes for both the angiotensin type 1 and2 receptors. In the ovine fetus, both receptor types arepresent in the mesonephros at 27 days of gestation and inthe meso- and metanephros at 41 days’ [40]. Throughoutgestation, the angiotensin 1 (AT1) receptor was locatedin developing glomeruli as well as in the medulla andmedullary rays (when present). At 41 days of gestation,the AT2 receptor was expressed in interstitial cells of themetanephros around the comma- and S-shaped nephronstructures. By 75 days’ it was possible to clearly identifythe AT2 mRNA in the macula densa, a specialized partof the distal tubule close to the glomerulus. Expressionof the AT2 receptor declined towards term and was ab-sent in the 2-day lamb kidney.
Expression of all components of the system from veryearly in gestation means angiotensin II, produced locallyin the kidney, could influence renal development andfunction from this very early stage. What may be the roleof angiotensin II in early kidney development? Some in-formation can be gained from studies where the systemhas been altered. Interesting abnormalities have been ob-served in mice in which the gene for one part of thissystem has been knocked-out. Mice that lack the genefor angiotensinogen show delayed glomerular maturationand develop lesions in the renal cortex [41], whilst an-giotensin converting enzyme knock-out mice have dis-torted renal vasculature along with greatly increased lev-els of renin gene expression, and both have hypoplasticpapillae [42]. In the mouse there are two subtypes of theAT1 receptor, namely the AT1a and the AT1b, and if ei-ther subtype is knocked-out individually, there do not ap-pear to be any kidney abnormalities [43, 44]. However,in double-mutant mice (i.e., those null mutant for boththe 1a and 1b receptors), there are grossly abnormal kid-neys, similar to those seen in the angiotensinogen knock-out mice [45].
The level of expression of the RAS may also be cru-cial for normal development. Babies that have high cordrenin concentrations have significantly smaller kidneys,as determined by ultrasound [46], and infants with intra-uterine growth retardation have high blood angiotensin IIlevels. Chronic (3-day) infusion of angiotensin to theovine fetus at mid-gestation causes a diuresis and an in-crease in blood pressure and causes a decrease in kidneyrenin gene expression (K.M. Moritz and E.M. Wintour,unpublished observations). It does not, however, affectexpression of the angiotensin receptors. Later in gesta-tion (120 days of gestation), a 3-day infusion of angio-tensin results in the complete abolition of renin gene ex-pression and causes expression of the AT1 receptor todecrease by about 50%. This indicates the kidney is sen-sitive to changes in the RAS from at least mid-gestation.Increased levels of angiotensin II may also have other ef-fects, including causing an increase in levels of AQP 1and 2 gene expression in the kidney [33]. In addition, an-giotensin II has been proposed as a renal growth factor[47] for both mesangial and tubular cells. These effectsare mediated by the AT1 receptor and may involve onco-gene activation and collagen synthesis [48].
176
Concluding remarks
There is increasing interest in the role of the mesoneph-ros as an important organ during fetal development, notmerely as a source for the formation of the metanephros.The metanephros also has a number of functions, butwhilst in utero its major role is to ensure the productionof large volumes of urine for adequate fetal fluids. Hor-monal systems (AVP, angiotensin II) are also modifiedduring development to allow the metanephros to functionin this manner.
&p.2:Acknowledgements Work of the authors is supported by a blockgrant to the Howard Florey Institute from the National Health andMedical Research Council of Australia. The authors would like tothank Dr. Donna Butkus and Ms. Nora Tennis and Hayley John-ston for the preparation of Fig. 1.
References
1. Law CM, Sheill AW (1996) Is blood pressure inversely relatedto birth weight? The strength of evidence from systematic re-view of the literature. J Hypertens 14:935–941
2. Hinchcliffe SA, Lynch MRJ, Sargent PH, Howard CV, DelzenD van (1992) The effect of intrauterine growth retardation onthe development of renal nephrons. Br J Obstet Gynaecol 99:296–301
3. Wintour EM, Alcorn D, Rockell MD (1998) Development andfunction of the fetal kidney. In: Brace RA, Hanson MA,Rodeck CH (eds) Fetus and neonate-physiology and clinicalapplications. Cambridge University Press, Cambridge, pp3–56
4. Bard JBL, Davies JA, Karanova I, Lehtonen E, Sariola H, Va-inio S (1996) Kidney development: the inductive interactions.Cell Dev Biol 7:195–202
5. Horster M, Huber S, Tschop J, Dittrich G, Braun G (1997)Epithelial nephrogenesis. Pflugers Arch 434:647–660
6. Kanwar YS, Frank AC, Kumar A, Wada J, Ota K, Wallner EI(1997) Role of extracellular matrix, growth factors and proto-oncogenes in metanephric development. Kidney Int 52:589–606
7. Vize PD, Seufert DW, Carroll TJ, Wallingford JB (1997) Mod-el systems for the study of kidney development: analysis of or-gan induction and patterning. Dev Biol 188:189–204
8. Davies J (1951) Nephric development in the sheep with refer-ence to the problem of the ruminant pronephros. J Anat 85:6–11
9. De Martino C, Zamboni L (1966) A morphologic study of themesonephros of the human embryo. J Ultrastruct Res 16:399–427
10. Beck F, Moffat DB, Davies DP (1985) Human embryology.Blackwell Scientific Press, Oxford, pp 246–276
11. Tiedemann K (1976) The mesonephros of cat and sheep. Com-parative morphological and histochemical studies. Adv AnatEmbryol 52:1–119
12. Wintour EM (1997) The renin angiotensin system and the de-velopment of the kidney. Trends Endocrinol Metab 8:199–202
13. Wilkes GE, Parsons PA (1988) The development of renal func-tion. In: Tyndale-Biscoe CH, Janssens PA (eds) The develop-ing marsupial. Springer, Berlin Heidelberg New York, pp176–189
14. Gaeth A, Short RV, Renfree MB (1997) The development ofthe urogenital system of the African elephant (abstract). Pro-ceedings of the Australian Society for Reproductive Biology,28th Annual Conference, 28 Sept–1 Oct 1997, Canberra
15. Saino K, Hellstedt P, Kreidberg JA, Saxen L, Sariola H (1997)Differential regulation of two sets of mesonephric tubules byWT-1. Development 124:1293–1299
16. Friebova-Zemanova Z (1981) Formation of the chick meso-nephros. Anat Embryol (Berl) 161:341–354
17. Sariola H (1996) Does the kidney express redundant or impor-tant molecules during nephrogenesis? Exp Nephrol 4:70–76
18. Dinchuk JE, Car BD, Focht RJ, Johnston JJ, Jaffee RD, Cov-ington MB, Contel NR, Eng VM, Collins RJ, Czerniak PM,Gorry SA, Trzaskos JM (1995) Renal abnormalities and an al-tered inflammatory response in mice lacking cyclooxygenaseII. Nature 378:406–409
19. Grinstead J, Aagesen L (1984) Mesonephric excretory func-tion related to its influence on differentiation of fetal gonads.Anat Rec 210:552–556
20. Medvinsky A, Dzierzak E (1996) Definitive hematopoiesis isautonomously initiated by the AGM region. Cell 86:897–906
21. Moritz KM, Lim GB, Wintour EM (1997) Developmental reg-ulation of erythropoietin and erythropoiesis. Am J Physiol273:R1829–R1844
22. Wintour EM, Butkus A, Earnest L, Pompolo S (1996) Theerythropoietin gene is expressed strongly in the mammalianmesonephric kidney. Blood 88:3349–3353
23. Zamboni L, Bezard J, Mauleon P (1979) The role of the meso-nephros in the development of the fetal sheep ovary. Ann BiolAnim Biochem Biophys 19:1153–1178
24. Buehr M, Gu S, McLaren A (1993) Mesonephric contributionto testis differentiation in the fetal mouse. Development 117:273–281
25. Uphadyay S, Zamboni L (1982) Preliminary observation onthe role of the mesonephros in the development of the adrenalcortex. Anat Rec 202:105–111
26. Hatano O, Takakusu A, Mormura M, Morohashi K (1996)Identical origin of adrenal cortex and gonad revealed by ex-pression profiles of Ad4BP(SF-1). Genes Cell 1:663–671
27. Stephens TD, McNulty TR (1981) Evidence for a metamericpattern in the development of the chick humerus. J EmbryolExp Morphol 61:191–205
28. Crossley HP, Minowada G, MacArtur CA, Martin GR (1996)Roles for FGF8 in the induction, initiation and maintenance ofchick limb development. Cell 84:127–136
29. Fernandez-Teran M, Piedra ME, Simandl BK, Fallon JF, RosMA (1997) Limb initiation and development is normal in theabsence of the mesonephros. Dev Biol 189:246–255
30. Lumbers ER (1995) Functions of the renin-angiotensin systemduring development. Clin Exp Pharmacol Physiol 22:499–505
31. Ervin MG, Kullama LK, Ross MG, Leake RD, Fisher DA(1993) Vasopressin receptors and effects during fetal develop-ment. Regul Pept 45:203–208
32. Butkus A, Alcorn D, Earnest L, Moritz K, Giles M, WintourEM (1997) Expression of aquaporin-1 (AQP1) in the adult anddeveloping sheep kidney. Biol Cell 89:313–320
33. Wintour EM, Earnest L, Alcorn D, Butkus A, Shandley L, Jey-aseelan K (1998) Ovine AQP1: cDNA cloning, ontogeny andregulation of renal expression. Pediatr Nephrol 12:545–553
34. Wintour EM, Alcorn D, McFarlane AC, Moritz K, PotocnikSJ, Tangalakis K (1994) Effect of maternal glucocorticoidtreatment on fetal fluids at 0.4 gestation. Am J Physiol 266:R1174–R1181
35. Devuyst O, Burrow CR, Smith BL, Agre P, Knepper MA, Wil-son PD (1996) Expression of aquaporins-1 and -2 duringnephrogenesis and in autosomal dominant polycystic kidneydisease. Am J Physiol 271:F169–F183
36. Lumbers ER (1995) Development of renal function in the fe-tus: a review. Reprod Fertil Dev 7:415–426
37. Gomez RA, Pupilli C, Everett AD (1991) Molecular and cellu-lar aspects of renin during kidney ontogeny. Pediatr Nephrol5:80–87
38. Schultz S, Le Moullec JM, Corvol P, Gasc JM (1996) Earlyexpression of all the components of the renin-angiotensin-system in human development. Am J Pathol 149:2067–2079
39. Wintour EM, Alcorn D, Butkus A, Congiu M, Earnest L, Pom-polo S, Potocnik SJ (1996) Ontogeny of hormonal and excre-tory function of the meso- and metanephros in the ovine fetus.Kidney Int 50:1624–1633
177
40. Butkus A, Albiston A, Alcorn A, Giles M, McCausland J,Moritz K, Zhuo J, Wintour EM (1997) Ontogeny of angioten-sin II receptors, types 1 and 2, in ovine mesonephros andmetanephros. Kidney Int 51:628–636
41. Niimura F, Labosky PA, Kakuchi J, Okubo S, Yoshida H,Oikawa T, Ichiki T, Naftilan AJ, Fogo A, Inagami T, HoganNLM, Ichikawa I (1995) Gene targeting reveals a requirementfor angiotensin in the development and maintenance of kidneymorphology and growth factor regulation. J Clin Invest 96:2947–2954
42. Hilgers KF, Reddi V, Krege JH, Smithies O, Gomez RA(1997) Aberrant renal vascular morphology and renin expres-sion in mutant mice lacking angiotensin-converting enzyme.Hypertension 29:216–221
43. Ito M, Oliverio MI, Mannon PJ, Best CF, Maeda N, SmithiesO, Coffman TM (1995) Regulation of blood pressure by thetype 1A angiotensin receptor gene. Proc Natl Acad Sci USA92:3521–3525
44. Chen X, Li W, Yoshida H, Tsuchida S, Nishimura H, Take-moto F, Okubo S, Fogo A, Matsusaka T, Ichikawa I (1997)
Targeting deletion of angiotensin type 1B receptor gene in themouse. Am J Physiol 272:F299–F304
45. Tsuchida S, Matsuska T, Chen X, Okubo S, Niimura F, Ni-shimura H, Fogo A, Utsunomiya H, Inagami T, Ichikawa I(1998) Murine double nullizygotes of the angiotensin type 1Aand 1B receptor genes duplicate severe abnormal phenotypesof angiotensinogen nullizygotes. J Clin Invest 101:755–760
46. Konje JC, Bell SC, Morton JJ, Chazel R de, Taylor DJ (1996)Human fetal kidney morphometry during gestation and the re-lationship between weight, kidney morphometry and plasmaactive renin concentrations at birth. Clin Sci (Colch) 91:169–175
47. Wolf G, Nielson EG (1993) Angiotensin II as a renal growthfactor. J Am Soc Nephrol 3:1531–1540
48. Maric C, Aldred GP, Antoine AM, Dean RG, Eitle E, Mendel-sohn FAO, Williams DA, Harris PJ, Alcorn D (1996) Effectsof angiotensin II on cultured rat renomedullary interstitial cellsare mediated by AT1A receptors. Am J Physiol 271:F1020–F1028
178
L I T E R AT U R E A B S T R AC T S
A.E. Heuvelink · F.L. van den Biggelaar · E. de BoerR.G. Herbes · W.J. Melchers · J.H. Huis in ’t VeldL.A. Monnens&bdy:
Isolation and characterizationof verocytotoxin-producing Escherichiacoli O157 strains from Dutch cattleand sheep
&misc:J Clin Microbiol (1998) 36:878–882
In the periods from July to November 1995 and 1996, fecal sam-ples from Dutch cattle and sheep were collected at the mainslaughterhouses of The Netherlands, located at diffeent geographicsites. The samples were examined for the presence of verocytotox-in (VT)-producing Escherichia coli(VTEC) of serogroup 0157. E.coli O157 strains could be isolated from 57 (10.6%) of 540 adultcattle, 2 (0.5%) of 397 veal calves, 2 (3.8%) of 52 ewes, and 2(4.1%) of 49 lambs. Immunomagnetic separation with O157-spe-cific-antibody-coated beads appeared to be significantly more sen-sitive than conventional plating for detection of the organism in fe-ces. With the exception of two isolates from adult cattle which ap-peared to be negative for VT genes, all animal isolates were posi-tive for both VT (VT1 and/or VT2) and E. coli attaching-and-ef-facing gene sequences, and therefore, they were regarded as poten-tial human pathogens. Although genomic typing by pulsed-fieldgel electrophoresis revealed a wide variety of distinct restrictionpatterns, comparison of the 63 animal isolates with 33 fecal O157VTEC strains previously isolated from humans with the diarrhea-associated form of the hemolytic-uremic syndrome by their phagetypes and VT genotypes showed a marked similarity between ani-mal and human isolates: 30 (90.9%) of the 33 human isolates ap-peared to be of E. coli O157 strain types also isolated from cattleand sheep. It was concluded that Dutch cattle and sheep are an im-portant reservoir of E. coli O 157 strains that are potentially patho-genic for humans.
M.R. DeBaun · M.J. Siegel · P.L. Choyke&bdy:
Nephromegaly in infancy and early childhood:a risk factor for Wilms tumorin Beckwith-Wiedemann syndrome
&misc:J Pediatr (1998) 132:401–404
Objective Beckwith-Wiedemann Syndrome (BWS) is an over-growth syndrome associated with macrosomia, omphalocele, mac-roglossia, visceromegaly and Wilms tumor (WT). We conducted acase-control study in children with BWS to examine whethernephromegaly increases the risk of WT.Methods The BWS Registry was used to identify control and casepatients. Control patients were defined as children with BWS whowere older than 6 years and had no imaging evidence of renal dis-ease or previous WT and for whom complete records were avail-able; 31 patients met these criteria. Case patiens were defined aschildren with BWS who had WT and screening renal imaging be-fore the diagnosis of WT; 12 of these patients had serial screeningimages before diagnosis of WT and comprised the study popula-tion. Only renal images obtained before the diagnosis of WT wasmade were used to assess renal length.Results All 12 patients with WT had nephromegaly (> or =95thpercentile of age adjusted renal length) on serial screening studies.Only four of 31 control patients (specificity=86%) had nephro-megaly resulting in an odds ratio of 72 (95% confidence inter-val=13–391) for the risk of WT with nephromegaly.Conclusions In patients with BWS, persistent nephromegaly is astrong risk factor for the development of WT. If screening for WTis done in this population, infants with nephromegaly should beconsidered those at greatest risk for WT, and screening may bebest targeted at this group.
