MOLECULAR CONTROL OF ORGANOGENESISRole of laminin γ2 and γ2*, type XVIII collagen and Wnt2b

YANFENGLIN

Biocenter Oulu,University of Oulu

Department of Biochemistry,University of Oulu

OULU 2001

YANFENG LIN

MOLECULAR CONTROL OF ORGANOGENESISRole of laminin γ2 and γ2*, type XVIII collagen and Wnt2b

Academic Dissertation to be presented with the assent ofthe Faculty of Medicine, University of Oulu, for publicdiscussion in Kuusamonsal i (Auditorium YB210),Linnanmaa, on December 7th, 2001, at 12 noon.

OULUN YLIOPISTO, OULU 2001

Copyright © 2001University of Oulu, 2001

Manuscript received 9 November 2001Manuscript accepted 15 November 2001

Communicated byDoctor Päivi MiettinenDoctor Veli-Matti Kähäri

ISBN 951-42-6566-1 (URL: http://herkules.oulu.fi/isbn9514265661/)

ALSO AVAILABLE IN PRINTED FORMATISBN 951-42-6565-3ISSN 0355-3221 (URL: http://herkules.oulu.fi/issn03553221/)

OULU UNIVERSITY PRESSOULU 2001

Lin, Yanfeng, Molecular control of organogenesis. Role of laminin γ2 and γ2*, type XVIII collagen and Wnt2b Biocenter Oulu and Department of Medical Biochemistry, University of Oulu, P.O.Box 5000, FIN-90014 University of Oulu, Finland; Department of Biochemistry, University of Oulu, P.O.Box 3000,FIN-90014 University of Oulu, Finland2001Oulu, Finland(Manuscript received 9 November 2001)

Abstract

How cell and tissue interactions lead to complex structures and differentiated cell types during organogenesis isstill one of the most fundermental questions in modern molecular biology. Laminin appears to have a role inbranching morphogenesis during organ development. Laminin5 (α3ß3γ2) is an epithelium-specific isoform oflaminin and previous report has shown that two alternative transcripts for the γ2 chain, the longer γ2 and theshorter γ2*, result from alternative use of the last exon in the human LAMC2 gene. But the transcription of murinelaminin γ2 and γ2* and their biological significance have remained unclear. Type XVIII collagen is a newlyidentified member of the collagen family. It may be involved in the Wnt signaling pathway, since its longest N-terminal variant contains a frizzled domain, which is part of the Wnt receptor and could antagonize Wnt signalingwhen secreted. Wnt2b is a new member of the Wnt family. Also its function in organogenesis is unknown. In thisstudy, we have investigated the expressions of laminin γ2 and γ2*, type XVIII collagen and Wnt2b during mouseorganogenesis. The function of type XVIII collagen in developing lung, kidney and a recombination of uretericbud and lung mesenchyme tissue and the function of the Wnt-2b gene during kidney organogenesis were studiedby using the combined methods of traditional experimental embryology and modern molecular biology.

Two alternative laminin γ2 transcripts were demonstrated in mouse. In the developing kidney, the shorter γ2*form was localized in the mesenchyme, whereas the longer γ2 form was only present in the epithelium of theWolffian duct and in the ureteric bud, indicating different functions for the γ2 variants. Type XVIII collagen wasexpressed throughout the respective epithelial bud at the initiation of lung and kidney organogenesis. It becomedlocalized to the epithelial tips in the early-stage lung, while it was confined to the epithelial stalk region and wasabsent from the nearly formed ureteric tips in the kidney. In recombinants of ureteric bud and lung mesenchyme,the type XVIII collagen expression pattern in the ureteric bud shifted from the kidney to the lung type,accompanied by a shift in epithelial Sonic Hedgehog expression. The lung mesenchyme was also sufficient toinduce ectopic lung Surfactant Protein C expression in the ureteric bud. A blocking antibody for the type XVIIIcollagen reduced the number of epithelial tips in the lung and completely blocked ureteric development with lungmesenchyme, which was associated with a notable reduction in the expression of Wnt2. The shift in type XVIIIcollagen expression in ureteric bud and lung mesenchyme tissue recombinant was also accompanied by thesignificant morphological changes in the branching pattern in ureteric bud development. Wnt2b was expressed innumerous developing organs in the mouse embryo, but it was typically localized in the perinephric mesenchymalcells in the region that partly overlaps the presumptive renal stroma at E11.5. Functional studies of the kidneydemonstrated that 3T3 cells expressing Wnt2b were not capable of inducing tubule formation but rather stimulatedureteric development. Recombination of ureteric bud treated with cells expressing Wnt2b and isolated kidneymesenchyme resulted in recovery of the expression of epithelial marker genes and better reconstitutedorganogenesis. Lithium, a known activator of Wnt signaling, was also sufficient to promote ureteric branching inreconstituted kidney in a manner comparable to Wnt2b signaling.

Our data suggest that different organ morphogenesis is regulated by an intraorgan patterning process thatinvolves coordination between inductive signals and matrix molecules, such as type XVIII collagen. In the mousekidney, Wnt2b may act as an early mesenchymal signal controlling morphogenesis of epithelial tissue, and theWnt pathway may regulate ureteric branching directly.

Keywords: patterning, organogenesis, Wnt signaling, epithelial mesenchymal interactions

To Shaobing and Lizhong

Acknowledgements

This study was carried out at Biocenter Oulu and the Department of Biochemistry,University of Oulu, Finland, between the years 1997 and 2001.

I wish to express my first and deepest gratitude to my supervisor, Professor SeppoVainio, for introducing me to the world of science, for promoting independent scientificthinking and for his constant support and encouragement during all these years. I alsowish to acknowledge Professor Taina Pihlajaniemi, Ph.D., scientific director of BiocenterOulu, and Professor Kalervo Hiltunen, Ph.D., M.D., Chairman of the Department ofBiochemistry, and all the other group leaders at the Department of Biochemistry forproviding such an inspiring working atmosphere and excellent research facilities.

I am grateful to Dr. Päivi Miettinen and Dr. Veli-Matti Kähäri for their valuablecomments on the manuscript of this thesis. Dr. Sirkka-Liisa Leinonen is gratefullyacknowledged for the careful and rapid revision of the language.

I wish to thank all my co-authors for their fruitful collaboration and contributions tothis work. I also wish to thank the senior scientists in Vainio’s team, Hellevi Peltoketo,Reetta Vuolteenaho, Paula Reponen and Sirpa Kontusaari for their kind advice. I own mythanks to my colleagues, Satu Kuure, Marika Uusitalo, Juha Peräsaari, Minna Heikkilä,Petri Itäranta, Mikael Niku, Tarja Ruusunen, Tiina Pihkala, Tiia Rinta-Rahko, IldikoLoikkanen, Maritta Ilmonen, Maria Kuure, Hannele Härkman, Johanna Kekolahti-Liias,Sanna Kalmukoski, Soili Miettunen and all the other former and present members of ourgroup and department for their friendship, inspiring discussions, useful advice, andassistance with many aspects of this work. I would also like to express my warmestthanks to Professor Karl Tryggvason, Professor Iain Drummond and Professor JordonKreidberg. Their advice, support and encouragement have meant a lot to me. I also wantto present my warmest thanks to Dr. Juha Tuukkanen, Dr. Ahmed Yagi, Dr. Tomi Airenne,Dr. Marko Rehn and Dr. Ritva Heljasvaara for their support, insights and inspiration. Iwould also like to thank the Chinese community in Oulu, especially the scientists inBiocenter Oulu, Deqi Huang, Yongmei Qin, Chunguang Wang, Weidong Xin, ChunleiHan, Hongmin Tu and Jingdong Shan, for their friendship and helpful advice. I wouldalso like to thank all the scientists and the skilful staff of the Department of Biochemistryfor helping me in many ways. In particular, I wish to thank Seppo Kilpeläinen, Terttu

Keskitalo, Irja Leinonen, Maire Jarva, Tuula Koret, Anneli Kaattari, Jaakko Keskitali,Kyösti Keränen and Virpi Hannus for their friendly help.

My biggest and warmest thanks belong to my dear wife Shaobing for herunderstanding and support in all matters of life and to our lovely son Lizhong, who bringspleasure into our life. Finally, my most sincere thanks go to my parents and other familymembers, especially my uncle, Dr. Chunrong Lin, for his never-ending encouragementand persistent support for my academic career, his thorough understanding of traditionalChinese philosophy has become deeply etched in my mind and will continue to have afar-reaching influence in my future life.

This study was financially supported by grants from the Academy of Finland, theSigrid Juselius Foundation and the Finnish Center for International Mobility (CIMO).

Oulu, Finland, August 2001 Yanfeng Lin

Abbreviations

BM basement membraneBmp bone morphogenetic proteinCDNA complementary DNACRD cysteine-rich domainDvl dishevelledECM extracellular matrixEGF epidermal growth factorFGF fibroblast growth factorFz frizzledGDNF glial cell line-derived neurotrophic factorHGF hepatocyte growth factorLB/KM lung bud and kidney mesenchymeLB/LM lung bud and lung mesenchymeLB/TM lung bud and tracheal mesenchymeLIF leukemia inhibitory factorLRP low-density lipoprotein receptor-related proteinMMP matrix metalloproteinasePBS phosphate-buffered salinePFA paraformaldehydePG proteoglycanPIFB position index of the first branchPMT percentage of the module typeRT-PCR reverse transcriptase PCRSFRP secreted frizzled-related proteinShh sonic hedgehogSP-C surfactant protein CSP/KM spinal cord and kidney mesenchymeTB/LM tracheal bud and lung mesenhcymeTGFβ transforming growth factor betaUB/KM ureteric bud and kidney mesenchymeUB/LM ureteric bud and lung mesenchyme

UB/TM ureteric bud and tracheal mesenchymeVEGF vascular endothelial growth factor

List of original articles

This thesis is based on the following articles, which are referred to in the text by theirRoman numerals:

I Airenne T, Lin Y, Olsson M, Ekblom P, Vainio S & Tryggvason K (2000)Differential expression of mouse laminin γ2 and γ2* chain transcripts. Cell &Tissue Research 300: 129-37.

II Lin Y, Zhang S, Rehn M, Itäranta P, Tuukkanen J, Heljäsvaara R, Peltoketo H,Pihlajaniemi T & Vainio S (2001) Induced repatterning of type XVIII collagenexpression in ureter bud from kidney to lung type: association with Sonichedgehog and ectopic surfactant protein C. Development 128: 1573-85.

III Lin Y, Zhang S, Tuukkanen J, Peltoketo H, Pihlajaniemi T & Vainio S (2001)Morphological characterization of lung and kidney development: reprogrammingureter bud branching with lung mesenchyme towards early lung typemorphogenesis. Submitted to International Journal of Developmental Biology.

IV Lin Y, Liu A, Zhang S, Ruusunen T, Kreidberg J, Drummond I & Vainio S (2001)

Induction of ureter branching as a response to Wnt2b signaling during early kidneyorganogenesis. Developmental Dynamics 222: 26-39.

Contents

Abstract Acknowledgements Abbreviations List of original articles Contents 1 Introduction....................................................................................................................152 Review of the literature..................................................................................................17

2.1 General mechanisms of development.....................................................................172.1.1 Epithelial-mesenchymal interactions ...........................................................172.1.2 Permissive and instructive inductions ..........................................................182.1.3 Organ-specific branching morphogenesis ....................................................182.1.4 Condensation of organ-specific mesenchymal cells.....................................19

2.2 Lung organogenesis................................................................................................192.3 Kidney organogenesis ............................................................................................212.4 Matrix molecules and their roles in organogenesis ................................................22

2.4.1 Collagen .......................................................................................................232.4.2 Proteoglycans ...............................................................................................262.4.3 Laminin ........................................................................................................27

2.5 Growth factors and their roles in organogenesis ....................................................272.5.1 Wnts and organogenesis ...............................................................................272.5.2 FGFs and roles of FGFs pathway during organogenesis..............................312.5.3 Transforming growth factor β superfamily...................................................322.5.4 Hedgehog family..........................................................................................332.5.5 Other growth factors.....................................................................................34

2.6 Growth factors and matrix molecules in lung organogenesis .................................342.7 Growth factors and matrix molecules in kidney development ...............................38

3 Outlines of the present research .....................................................................................444 Material and methods.....................................................................................................46

4.1 Mouse strains, organs and tissues (I-IV) ................................................................464.2 Organ culture methods (I-IV).................................................................................464.3 Tissue recombination (II-IV)..................................................................................46

4.4 Hanging drop assays and agarose bead experiments (II, IV) .................................474.5 Ureteric bud cultured with functional cell lines and kidney reconstitution assays

(IV)........................................................................................................................474.6 Antibodies, whole-mount and section immunostaining (II-IV)..............................484.7 Inhibition of Type XVIII collagen function by a specific antibody (II) .................484.8 Probes, whole-mount and section in situ hybridization and staining of explants (I,

II, IV) ....................................................................................................................484.9 RNA isolation and RT-PCR (I, IV).........................................................................494.10 Photography and Image Analysis (I-IV)...............................................................494.11 Cell proliferation assay (II) ..................................................................................49

5 Results............................................................................................................................505.1 Laminin γ2 and γ2* expression in developing kidney rudiments (I) ......................505.2 Expression of type XVIII collagen in embryonic lung and kidney (II) ..................505.3 Specific branching pattern in embryonic lung and kidney (II, III) .........................515.4 Lung mesenchyme is sufficient to completely respecify type XVIII collagen

expression in the ureteric bud towards the lung type (II, III) ................................525.5 Respecification of type XVIII collagen was associated with early lung type

ureteric bud branching morphogenesis and induction of SP-C (II, III) .................525.6 Type XVIII collagen is important for lung development and ureteric bud patterning

(II) .........................................................................................................................535.7 Search for the mesenchymal factor and the epithelial mediator involved in the

ureteric bud and type XVIII collagen repatterning process (II) ............................545.8 Wnt2b gene is expressed in several epithelial-mesenchymal derived organs (IV) .555.9 Wnt2b fails to induce tubulogenesis in co-cultured kidney mesenchyme but

promotes ureteric bud survival and growth (IV) ...................................................555.10 Recombination of Wnt2b-pre-treated ureteric bud tissue with isolated

nephrogenic mesenchyme results in a recovery of organogenesis (IV) ................565.11 Lithium is also sufficient to promote ureteric branching in the reconstituted

kidney in a manner comparable to Wnt2b signaling (IV) .....................................575.12 Reconstitution was significantly better in Wnt2b or lithium-treated explants than

control or Wnt4-treated samples (IV) ...................................................................576 Discussion......................................................................................................................58

6.1 Role of type XVIII collagen in epithelial development..........................................586.2 Type XVIII collagen may regulate modes of epithelial morphogenesis in kidney

and lung and Wnt2 could be the lung specific mesenchymal factor......................596.3 Reprogramming of ureteric bud branching with lung mesenchyme favors early

lung type branching morphogenesis......................................................................636.4 Involvement of Wnt2b in early kidney development ..............................................646.5 Wnt2b regulates ureteric bud branching and kidney organogenesis .......................64

7 Conclusions and future perspectives ..............................................................................67ReferencesOriginal articles

1 Introduction

The mechanisms by which epithelial and mesenchymal tissue interactions lead to tissue-specific morphogenesis and the factor(s) that determine the complex organ-specific formare poorly known. The epithelial branching morphogenesis in different organs is crucialfor the establishment of organ-specific forms and structures. Embryonic lung and kidneyhave been proposed as useful models for studying the molecular mechanisms ofbranching morphogenesis (Saxén, 1987; Hogan, 1999; Metzger & Krasnow, 1999).Studies on the inducers involved in the model systems have implicated only a smallnumber of signaling molecules responsible for the control of organogenesis, such asfibroblast growth factor (FGF), hedgehog, transforming growth factor beta (TGFβ), Wntand epidermal growth factor receptor (EGF-R) binding ligands (for a review, see Hogan,1999). All these encode secreted factors and can mediate autocrine or paracrine signalingover short- or long-range distances and regulate cell behavior. The extracellular matrix(ECM) is also an important component in morphogenesis. It does not only provide aphysical substratum for the spatial organization of cells, but may also play a more activerole in inductive tissue interactions by controlling the activities of growth factors (for areview, see Hilfer, 1996). Wnt signal transduction via cell surface Fz receptors and low-density lipoprotein receptor-related protein (LRP) co-receptors is linked to a variety ofcellular processes, including changes in intracellular calcium levels, control of themolecular machinery associated with cell migration, polarity, adhesion, and cellproliferation (Semenov et al, 2001; Mao, et al, 2001). Despite the advances in ourunderstanding of Wnt signaling in cellular function and the finding that several Wntgenes are necessary for embryogenesis, the mechanisms by which Wnt genes regulatemorphogenesis are not well understood.

In this work, we carried out studies of laminin γ2 and γ2* transcription in the earlykidney development and the expression of type XVIII collagen during lung and kidneyorganogenesis. We also investigated the patterning process that involves type XVIIIcollagen and its correlation with the change in the epigenesis of the ureteric bud from thekidney type towards the early lung type. The specific instructive factors in lungmesenchyme were investigated, and it turned out that the factor(s) in lung mesenchyme(the most obvious candidate is Wnt2) may interact with Shh and/or type XVIII collagento re-specify the ureteric bud branching morphogenesis. The shift in type XVIII collagen

16

expression in ureteric bud and lung mesenchyme tissue recombinant is also accompaniedby changes in the branching pattern in ureteric bud development, which was shown bythe significant changes of several morphological parameters. This work also studied theexpression of Wnt2b in developing mouse embryo. By using cells expressing Wnt2b, itsfunctions during the processes of ureteric bud growth and branching and kidneyreconstitution were further studied. We demonstrated that Wnt2b supports ureteric budgrowth and branching and promotes kidney reconstitution, whereas Wnt4 signaling,which is sufficient to induce kidney tubules, does not have such functions.

2 Review of the literature

2.1 General mechanisms of development

Development has ensured the continuity of life from one generation to the next. How afertilized egg, a single cell, divides to produce all the cells of the body has been a mysterythroughout human history (Gilbert, 2000). The advances and applications of molecularbiology have made a breakthrough in the understanding of the mechanisms ofdevelopment at the molecular level. There are four events involved in this development:1) regional specification, 2) cell differentiation, 3) morphogenesis and 4) cellproliferation (Slack, 2001). At the very early stage of embryogenesis, cells are featurelessand need to become programmed. This process involves intracellular signaling, cell-cellinteractions, cell-matrix interactions and different inducing factors (growth factors,cytokines or hormones) as well as cytoskeleton, cell adhesion molecules and ECMcomponents. Genetic signals control all of these events and lead to a complex and uniquetype of organ.

2.1.1 Epithelial-mesenchymal interactions

Organogenesis is characterized by coordinated growth and differentiation of cells inepithelial and mesenchymal cells lineages. Classical tissue separation and recombinationexperiments and recent molecular biological proceedings have demonstrated that theinteractions between epithelial and mesenchymal tissues are crucial for organdevelopment (for a review, see Gurdon, 1992). The interactions between epithelial andmesenchymal tissues are sequential and reciprocal. The mesenchyme influences theepithelium, epithelial tissue, once changed by the mesenchyme, can secrete factors thatchange the mesenchyme. Such interactions continue until an organ is formaed withorgan-specific mesenchymal cells and organ specific epithelia. In the developing tooth,for example, the epithelial cells differentiate into enamel-producing ameloblasts and the

18

neural-crest-derived mesenchyme into bone-forming odentoblasts via a set of reciprocaltissue interactions. Epithelial branching does not occur in the absence of mensechymaltissue, and mesenchymal cells do not differentiate without signals from epithelial tissue(Wessells, 1970; Saxén, 1987; for a comprehensive review, see Thesleff et al, 1995).

The change of one differentiated cell type into another is known as transdifferentiation, and a number of cell types have been found to undergo such a change in tissue culture, partucularly when the culture conditions are altered by addition of chemical agents. One typical example can be seen in the pigmented epithelial cell culture. A single pigmented epithelial cell from the embryonic chick retina can be growth in culture to produce a monolayer of pigmented cells. On further culture in the presence of hyaluronidase, serum, and phenylthiourea, they lose their pigment and retinal cell characteristic. If cultured at a high density with ascorbic acid, they differentiate as lens cells and produce the lens-specific protein crystalline. (Wolpert, et al, 1998).

2.1.2 Permissive and instructive inductions

The sophisticated inductive events of epithelial-mesenchymal inductive tissueinteractions have been simply divided into ´permissive` and ´instructive` categories(Saxén, 1977; for a review, see Lehtonen & Saxén, 1986). In instructive induction, asignal from the inducing cell is necessary for initiating gene expression in the respondingcell. Without the inducing cell, the responding cell would not be capable of differentiatingin that particular way. This means that the target cells possess more than onedevelopmental option, and the inductive effect leads to the selection of one of them. Forexample, when mouse urinary bladder epithelium is combined with urogenital sinusmesenchyme, it is converted into prostatic epithelium (Cunha et al, 1983). In permissiveinduction, the responding tissue contains all the potential needed for expression and onlyrequires an environment that permits the expression of certain traits. It can also be saidthat the final fate of the responding tissue has already been determined, but an exogenousstimulus is still required for the expression of its phenotype. A typical example is that themesenchyme of metanephric blastema is converted into kidney tubules only whenexposed to an inducer, such as ureteric bud. An intermediary form of inductive tissueinteraction has, however, been observed in the skin. Heterotypic and interspeciescombinations between dermal mesenchyme and epidermis have demonstrated that thetype of appendix (feather, hair, scale) is determined by the genetic constitution of theepidermis, but their size and distribution are directed by the dermis (for a review, seeLehtonen & Saxén, 1986).

2.1.3 Organ-specific branching morphogenesis

At the early stage of embryogenesis, organs are composed of small epithelial rudimentssurrounded by mesenchymal cells. The interactions between the epithelial andmesenchymal tissues induce proliferation and branching of the epithelium into themesenchyme. It is likely that similar molecular mechanisms exist for the epithelial

19

branching program, and different branching organs may share common features both atthe morphological and the molecular levels. But epithelial branching morphogenesis fordifferent organs is crucial for the establishment of organ-specific forms and structures. Inmost organs, however, the patterns of branching are seldom random but rather underfixed development control, at least in the early branching generations (for reviews, seeKrasnow, 1997; Metzger & Krasnow, 1999). For instance, in early mammalian lung, theepithelial bud sprouts and sends out lateral branches always in precise, invariantpositions. Whether the epithelium or the mesenchyme possesses the decisive patterningprogram is not clear. In most organ systems, specific mesenchymal factors instructepithelial morphogenesis and determine the form of the particular organ. Themesenchyme-common factor and the mesenchyme-specific factor have been postulated toexplain the varying specificity of mesenchymal induction, and the latter may containdevelopmentally significant information (Grobstein, 1967; Saxén, 1987). Epithelial cells,however, seem to differentiate according to their own origin. Recent tissue separation andrecombination experiments have provided further evidence to suggest that the programmefor epithelial branching may be in the epithelium itself and not regulated bymesenchymal cells (Qiao et al, 1999b). Therefore, either the epithelium or themesenchyme may posses the information necessary for organ-specific morphogenesis anddifferentiation (for a review, see Thesleff et al, 1995).

2.1.4 Condensation of organ-specific mesenchymal cells

The condensation, or aggregation, of mesenchymal cells has been recognized as oneimportant feature in morphogenesis. The formation of an aggregate is often a prelude tothe formation of structures in the early stage of developing organs, such as themesenchymal aggregates during kidney tubulogenesis. The condensate is localized to amorphogenetic field consisting of a limited number of cell layers. The condensationprocess is connected to the determination of the fate of cells in a morphological field. It isnot known, however, to what extent the mechanisms of condensation are shared in thevarious organ systems (for a review, see Thesleff et al, 1995).

2.2 Lung organogenesis

Lung primordium evaginates as an epithelial bud from the endoderm of the primitiveforegut in the laryngotracheal groove into the surrounding splanchnic mesenchyme inmouse embryos at the developmental age of about E9.5 (Ten Have-Opbroek, 1991) (Fig.1A-D). During this process, the embryonic foregut is divided by a longitudinal septuminto two tubes, the esophagus (dorsal) and the trachea (ventral). Shortly after itsappearance, the primordium develops into the prospective trachea and the left and rightprimary bronchi. After the conducting airways have been formed, lung growth continuesby sacculation and septation of the peripheral saccules to form alveoli. The alveoli arelined by specialized epithelial cells (Type I and Type II cells) and vascularized by anextensive capillary bed to facilitate gas exchange.

20

The histological development of mouse lung has been divided into four stages (for areview, see Kaplan, 2000) (Fig. 1E): (1) pseudoglandular stage (E9.5–16.5): developmentof the bronchial and respiratory tree, formation of the undifferentiated primordial system;(2) canalicular stage (E16.5–17.5): development of terminal sacs and earlyvascularization; (3) terminal sac stage (E17.5 to postnatal day 5 (P5)): increase in thenumber of terminal sacs and vascularization, differentiation of type I and II alveolar cells;and (4) alveolar stage (P5–30), when the terminal sacs develop into mature alveolar ductsand alveoli. In the developing airway, the upper airways are lined with ciliated columnarcells and mucus secreting cells. The lower airways are lined with Clara cells. The alveoliare lined with alveolar type I and II epithelial cells. The primitive epithelium of earlymurine lung co-expresses a number of lineage markers, including surfactant-associatedproteins.

The lung mesenchyme has been demonstrated, based on its ability to induce earlytracheal epithelium that has been denuded of its own mesenchyme, to branch in a lung-like pattern. These observations had been repeated and extended by showing that the lungbranching morphogenesis induced in the tracheal epithelium by lung mesenchyme isaccompanied by re-programming of the tracheal epithelial cells to express an alveolartype II cell phenotype (Shannon, 1994; for a review, see Hogan, 1999), including theexpression of the type II cell-specific marker surfactant protein C (SP-C). But when thetip of the left bronchus was denuded and covered with its own tracheal mesenchyme, nobranching occurred at the tips of the left bronchus (Wessells, 1970). Therefore, all of thepatterning information is committed to the pulmonary epithelium and the surroundingmesenchyme.

21

Fig. 1. Diagrams showing key events in mouse lung organogenesis. (A) The primitive lunganalage emerging from the ventral surface of the primitive foregut at E9.5 in mouse. (B) Thetwo primary bronchial branches arise from the lateral aspects of the laryngo-tracheal grooveat E10. (C) The embryonic larynx and trachea with the two primary bronchial branchesseparated dorso-ventrally from the esophagus at E10.5. (D) The primitive lobar bronchibranching from the primary bronchi at E11.5. (E) A schematic rendering of the fourdeveloping stages of mouse embryonic lung (Modified from Warburton et al, 2000).

2.3 Kidney organogenesis

Development of the metanephric kidney starts with the formation of a Wolffian duct-derived ureteric bud, which invades the nephric mesenchyme around day 11 of mousedevelopment in response to a signal from the mesenchyme (Saxén, 1987). Mesenchymeinduction is necessary for the process of further growth and branching of the ureteric bud.The ureteric bud enters the metanephrogenic mesenchyme and induces the surroundingmesenchymal cells to condense (Fig. 2). Condensing mesenchymal cells aggregate andepithelialize to form renal vesicles, which are first comma-shaped and finally S-shapedbodies that fuse with the branching ureteric epithelium. Tissue interactions play animportant role in kidney development. Without the ureteric bud, there would be no tubuleformation, which shows that the bud emits an inductive signal necessary for tubuleformation. The induction is a reciprocal process because there is also an effect of themesenchyme on the ureteric bud. Without the presence of the mesenchyme, the budwould not arise from the nephric duct and would not continue to grow and branch (forreviews, see Sariola & Sainio, 1997; Vainio et al, 1999). In a developing kidney, theureteric bud mostly carried out dichotomous branching (Saxén, 1987) or rapid bifidbranching (the main stem divides into two) (for reviews, see Sariola & Sainio, 1997; Al-Awqati & Goldberg 1998; Davies & Davey, 1999). Recent data, however, suggest thatthere is also a subtle control mechanism that can add new branches at sites other than thetermini to the developing ureteric bud (for a review, see Davies & Davey, 1999). Themain dichotomous branching pattern of the ureteric bud in an early stage kidney isdifferent from the lateral branches often present in developing lobule organs. But thebranching pattern of the ureteric bud becomes irregular at the later developing stage andin the renal cortex area (for a review, see Al-Awqati & Goldberg 1998). The metanephricmesenchyme has been thought to be very unique in its capacity to support theproliferation and branching of the ureteric bud (Saxén, 1987). However, recent data havesuggested that the ureteric bud is competent to branch with heterologous mesenchyme,and that this ability depends on the developmental stage. After the formation of the firstepithelial branch, lung mesenchyme is sufficient to support the early growth andbranching of the ureteric bud (Kispert et al, 1996; Sainio et al, 1997).

22

Fig. 2. The early developmental stages of kidney morphogenesis A. Metanephric mesenchymecondenses around the tip of the epithelial ureteric bud. B. Mesenchyme is induced and formsa comma-shaped body. C. and D. The epithelialized body elongates and forms an S-shapedbody. Simultaneously, endothelial cells invade the lower crevice of the body. E. and F. The S-shaped body fuses with the tip of the ureteric bud and the glomerulus expands (modifiedfrom Saxén, 1987).

2.4 Matrix molecules and their roles in organogenesis

ECM consists of collagens, elastin, proteoglycans (PG), and a variety of specializedglycoproteins, such as fibronectin and laminin. It is now firmly established that ECM is asource of essential developmental signals, and cell-ECM interactions are critical in theregulation of gene expression and morphogenesis. For instance, gene-targeting micereveal a contribution of tenascin-C to proliferation and migration in oligodendrocyteprecursors during development (Garcion et al, 2001). Matrix metalloproteinases (MMPs)are a family of zinc-dependent metalloendopeptidases collectively capable of degradingessentially all extracellular matrix components. The activity of MMPs is specificallyinhibited by tissue inhibitors of metalloproteinases (TIMPs). Controlled degradation ofextracellular matrix is an essential feature in physiological situations, such as fetal tissuedevelopment, angiogenesis, and tissue repair (for reviews, see Werb et al, 1999;Johansson et al, 2000). In embryogenesis, the ECM remodelling processes are importantfor the morphogenesis of all the fetal organs as well (Canete-Soler et al, 1995; for areview, see Lelongt et al, 2001).

23

2.4.1 Collagen

The collagen superfamily includes 19 proteins formally defined as collagen and severalless well-characterized collagens (for a review, see Myllyharju & Kivirikko, 2001). Allcollagen molecules consist of three polypeptide chains, called α chains, that are wrappedaround each other into a triple helix. In each of the polypeptide chains, every third aminoacid is glycine, and thus the sequence of an α chain in a protein can be expressed as (Gly-X-Y)n, where X and Y represent amino acids other than glycine and n varies inaccordance with the collagen type and domain. All collagens contain a non-collagenousdomain in addition to the actual collagen domains. Most collagens form polymericassemblies, and the superfamily can be divided into several families based on thesepolymeric structures and other features: (A) collagens that form fibrils containing a long,uninterrupted stretch of Gly-X- Y repeats (types I, II, III, V, and XI); (B) collagens thatare located on the surface of fibrils and are called fibril-associated collagens withinterrupted triple helices (FACIT) and structurally related collagens (types IX, XII, XIV,XVI and XIX). Type XII collagen is a representative of FACIT collagens. (C) collagensthat form hexagonal networks (types VIII and X); (D) the family of type IV collagensfound in BMs; (E) type VI collagen, which forms beaded filaments; (F) type VIIcollagen, which forms anchoring fibrils for BMs; (G) collagens with transmembranedomains (types XIII and XVII); and (H) the family of type XV and XVIII collagens. TypeXV and XVIII collagens constitute the newest subgroup, also called MULTIPLEXINS(proteins with multiple triple helix domains and interruptions) (Oh et al, 1994, for areview, see Myllyharju & Kivirikko, 2001).

Collagen plays a dominant role in maintaining the structural integrity of varioustissues and organs and also has a number of other important newly identified functions.Type IIA procollagen functions in the ECM distribution of bone morphogenetic proteins(Bmps) in chondrogenetic tissue (Zhu et al, 1999). Type XIII collagen is expressed in theneuronal structures and can enhance neurite outgrowth (Sund et al, 2001). Collagens arealso thought to mediate epithelial-mesenchymal interactions during organogenesis, suchas branching morphogenesis. Still, in collagen I-deficient mouse, the development of allexplants, such as lung and kidney, was normal. Immunostaining of mutant mice revealedsubstantial production of collagens III and V, indicating either that collagen I has no rolein the morphogenesis of these organs, or that its function is shared or can be substitutedfor by other collagens (Kratochwil, et al, 1986).

2.4.1.1 Type XVIII collagen

Type XVIII collagen was discovered simultaneously by three laboratories in 1994 (Abe etal. 1993; Oh et al. 1994; Rehn & Pihlajaniemi 1994). It is a component of the basementmembrane (BM) zone characterized by the occurrence of short and long N-terminalvariants (Rehn et al, 1994; Saarela et al, 1998; Saarela, 1998). The variant polypeptideforms are transcribed from two widely separated promoters (Rehn et al, 1996). TypeXVIII collagen may play roles in morphogenesis and tumorigenesis since its longestvariant contains a Fz domain, homologous to the ligand-binding domain of frizzled

24

molecules which act as part of the Wnt receptors (Rehn & Pihlajaniemi, 1995; Rehn et al,1998; for a review, Wodarz & Nusse, 1998), and its C terminus contains a proteolyticanti-angiogenic peptide with tumor-suppressing activity, known as endostatin (O’Reillyet al. 1997; Dhanabal et al. 1999a,b; for reviews, see Cao. 2001; Myllyharju & Kivirikko,2001). The function of type XVIII collagen, however, is unknown.

The cDNA-derived structure of type XVIII collagen has been determined for mouseand human (Rehn et al. 1994; Saarela et al.1998). The collagen α1 (XVIII) cDNAsequence contains an open reading frame of 3420 bp, including ten triple-helical (COL)domains, varying in length, separated by 11 non-collagenous (NC) regions. Type XVIIIcollagen has a region of homology with a large heparin-binding amino-terminal portionof thrombospondin- 1 (TSP-1). TSP-1 contains adhesive domains, which are required forcell attachment, and the soluble TSP-1 is a potent inhibitor of angiogenesis (Volpert et al.1998; Davies et al. 2001). But the significance of the thrombospondin homology of typeXVIII collagen is unknown (Rehn and Pihlajaniemi 1994). Type XVIII collagen also hascharacteristics of an heparan sulfate proteoglycans (HSPG), such as long heparitinase-sensitive carbohydrate chains and a highly negative net charge. In situ hybridizationshowed that the main site of expression of collagen XVIII HSPG in the chick embryo isin the kidney and the peripheral nervous system. As a substrate, collagen XVIIImoderately promoted the adhesion of Schwann cells but had no such activity onperipheral nervous system neurons and axons (Halfter et al. 1998). The mouse type XVIIIcollagen differs from type XV by the presence of variant polypeptide forms characterizedby three N-terminal noncollagenous domains of differing lengths (Fig. 3). The longestNC1 domain was strikingly characterized by a 110-residue sequence with 10 cysteines,which was found to be homologous with the previously identified Fz proteins belongingto the family of G-protein-coupled membrane receptors (Rehn & Pihlajaniemi, 1995).

25

Fig. 3. Schematic structures of the full-length variant polypeptides of the mouse α1 (XVIII)collagen chains. The collagenous sequences are shown in white, the non-collagenous domainscommon to all variants are shown in black, the non-collagenous sequences common to bothlong variant NC1 domain portions are shown in gray, and the non-collagenous sequenceunique to NC1-764 variant is shown by cross-hatching. The putative signal peptide 1 isindicated with brisk hatching, and the putative signal peptides 2 are indicated with righthatching. The lengths of the amino acid sequences (aa) specific to each variant are given, as isalso the length of the common regions. C, cysteine residue; 10C, cluster of 10 cysteineresidues; N, potential N-glycosylation site; 2N, two adjacent N-glycosylation sites; O, potentialO-linked glycosylation site; Fz and tsp, Fz and thrombospondin sequence motifs, respectively;ac, acidic domain. The cDNA fragments, named ELQ-2.9, used for the production andpurification of the anti-all mo XVIII antibody, and Q36.4, used for the production andpurification of the anti-long moXVIII, are shown above in the schematic structure. (AfterPihlajaniemi & Rehn, 1995; Saarela, 1998).

The Fz domain is an interesting motif. It shows 26-27% identity and 50-51% homologywith the cysteine-rich domain (CRD) found in each of the three previously characterized´frizzled` proteins, namely, the rat Fz proteins 1 and 2 and the Drosophila Fz protein (fora review, see Pihlajaniemi & Rehn, 1995). In Drosophila, the extracellular part of the Fzprotein is a tissue polarity receptor for the wingless family of signaling molecules.Therefore, the corresponding Fz domain in type XVIII collagen may also possess ligand-binding properties (Pihlajaniemi & Rehn, 1995; Rehn et al, 1998). In addition to beingpresent in type XVIII collagen, the frizzled domain is a widely used building block thatoccurs in a diverse set of proteins, including seven-span transmembrane receptors, themetallocarboxypeptidase enzyme carboxypeptidase Z, a family of secreted Fz-likeproteins, two subfamilies of receptor protein tyrosine kinases, e.g. the receptor tyrosinekinase MuSK, and a protein that contains an LDL-receptor class-A domain andSmoothened (Smo). The occurrence of Fz in this diverse set of proteins suggests thatnovel biological functions of this protein family will be discovered in the near future.(Rehn et al, 1998).

26

Type XVIII collagen is abundant in the retina, epidermis, pia, cardiac and striatedmuscle, kidney, blood vessels, and lung. Positive staining was notably seen in associationwith the blood vessels, but also in the BM zone of the skin and faintly in Bowman’scapsule in the kidney (Muragaki et al. 1995). In both human and mouse tissue, typeXVIII collagen resides in the majority of the BM zones of many tissue and organs(Saarela et al, 1998; Saarela, 1998). Type XVIII collagen protein was already visible atE9 in some BM-like structures, such as liver sinusoids. The staining pattern in embryoswas more restricted by the antibody recognizing the two longer variants, since thesevariants were found to reside only in a limited number of BM zones, such as the kidney,skin and liver sinusoids. The antibody recognizing all type XVIII collagen variantsstained most of the BM zones of the central and peripheral nervous system, thegastrointestinal system and the cardiovascular system and some of those in the respiratorysystem. The short variant is significantly expressed in the tissues where the long variantis also present, except in liver sinusoids. There are also variant-specific differences in thedistribution of type XVIII collagen during mouse development. The short variant wasfound to be expressed evenly throughout the course of development, while the expressionof the long variant became stronger during E11-15 (Saarela, 1998).

2.4.2 Proteoglycans

Proteoglycans (PG) are very diverse molecules, and various combinations of bothdifferent types of proteins and classes of glycoaminoglycan (GAG) chains are found invertebrates (for a review, see Perrimon & Bernfield, 2000, 1). The classes ofglycoaminoglucan chains include chondroitin sulfate, keratan sulfate, dermatan sulfateand heparan sulfate. The structural diversities of PG underlie their various functions. Forexample, HSPG bind a number of growth factors, e.g., FGFs, VEGF and epidermalgrowth factor (EGF). Interaction with heparan sulfate has been shown to be critical forgrowth factor signaling (Bernfield et al, 1999). Two major families of cell surfaceHSPGs, namely syndecans and glypicans, have been identified. Syndecans have theirindividual expression patterns and functions based on the variable features of theircytoplasmic and extracellular domains (for a review, see Rapraeger, 2001). Glypicans,defined as glycosyl phosphotidyl inositol (GPI)-linked HSPG, play essential roles duringembryogenesis (Perrimon N & Bernfield M, 2000). Most likely, glypicans could functionas co-receptors or by determining the activity ranges of morphogens and growth factorsin the control of cell growth and differentiation (for a review, see De Cat & David,2001). Recent biochemical and genetic studies indicated that glypicans' heparan sulfateglycosaminoglycans were also critical for endostatin binding, and glypicans serve as low-affinity receptors for endostatin in renal tubular cells, such as endothelial cells. Thisdemonstrated a role for endostatin in branching morphogenesis and suggested the criticalimportance of glypicans in mediating endostatin activities (Karumanchi et al, 2001).

27

2.4.3 Laminin

Laminin is one of the important multifunctional ECM glycoproteins and appears to havea role in branching morphogenesis (Wright et al, 1999). It is composed of threepolypeptide chains designated α, β, γ. These laminin chains are assembled to form severalheterotrimeric isoforms containing one α, one β and one γ chain that twist around eachother. 12 laminin isoforms have been characterized so far (for reviews, see Tryggvason,1993; Tunggal et al, 2000). Laminin is an important BM component. It interacts withspecific membrane receptors, such as the integrin molecules. During embryonicdevelopment, laminin is the first synthesized ECM molecule and has been implicated toplay a significant role in the morphogenesis of organs where epithelial-mesenchymalinteractions and branching take place (for a review, see Durbeej & Ekblom, 1997).Laminin5 (α3β3γ2) is an epithelium-specific laminin isoform, which has been implicatedin junctional epidermolysis bullosa (Fine et al, 1991; Epstein, et al, 1992; Pulkkinen et al,1994). There are two alternative transcripts for the γ2 chain in human, which is resultedfrom alternative use of the last exon in the 3’-end of human LAMC2 gene (Airenne et al,1996). The longer γ2 transcript contains all the 23 exons of the LAMC2 gene and wasfound to be a general product of epithelial cells. In contrast, the shorter γ2* transcript,which lacks the last exon of LAMC2 gene, had a highly restricted tissue distribution. Itwas observed only in 17-week-old cerebral cortex, in lung, and in the distal tubules of thekidney. The putative γ2* polypeptide differs from the γ2 polypeptide in that it contains 82residues less at the carboxy terminal (C-terminal) domain I than the γ2 chain. But thetranscription of laminin γ2 and γ2* in mouse and their biological significance hasremained unclear.

2.5 Growth factors and their roles in organogenesis

Growth factors are mainly proteins, and most of them are secreted from signaling cells.They bind to specific receptors and activate a signal transduction pathway which maylead to the activation or repression of specific genes or change cell behaviour and cellularmetabolism. Cumulative data from studies on organ culture, gene targeting experimentsand tissue separation and recombination experiments have suggested that a variety ofsoluble specific growth factors, including Wnts, FGFs, TGFβs, EGF-R binding ligandsand hedgehogs, are involved in specific tissue interactions and branching morphogenesis.(Slack, 2001).

2.5.1 Wnts and organogenesis

Close to 20 Wnt genes have been identified in mouse so far and the Wnt signalingtransduction pathway functions in various developmental processes, including body axisformation, development of the central nervous system, axial specification in limbdevelopment, and mouse mammary gland and kidney development (Table 1). Wnts

28

encode secreted glycoproteins closely associated with the cell surface and extracellularmatrix and stimulate cells in an autocrine and/or paracrine manner (for reviews, seeCadigan & Nusse 1997; Hlsken & Behrens, 2000; Pandur & Kuhl 2001; for web site, seehttp://www.stanford.edu/~rnusse/Wntwindow.html). When Wnts are secreted from a cell,they can accumulate in the surface the recipient or donor cell by binding to theirreceptors, and ECM molecules, such as PG, appear to be required in these processes (Lin& Perrimon, 1999).

Table 1. Wnt gene family and the gene inactivated phenotype (modified from Uusitalo etal. 1999).

Mouse Wnt gene -/- phenotype Reference McMahon 1990 Wnt-1 Loss midbrain, loss cerebellum Thomas 1990

Wnt-2 Placental defects Monkley 1996 Wnt-2b/13 - Wnt-3 Defects in axis formation, early gastrulation

defect Liu 1999

Takada 1994, Somites, tailbud defects Yoshikawa 1997

Wnt-3a

Loss hippocampus Lee 2000 Failure in kidney tubule formation Stark 1994 Wnt-4 Defects in female development; failure in Müllerian duct formation

Vainio 1999

Truncated A-P axis, reduction of proliferating cells and outgrowth

Wnt-5a

Defects in developing ears, face and genitals

Yamaguchi 1999

Wnt-5b - Wnt-6 -

Limb polarity defects, A-P patterning Parr 1995 Female infertility Parr 1998 Abnormal development of oviduct and uretus

Miller 1998

Wnt-7a

Delayed maturation synapses in cerebellum Hall 2000 Wnt-7b Placental developmental defects Parr 2001 Wnt-8a - Wnt-8b - Wnt - Wnt-10a - Wnt-10b Inhibition adipogenesis Ross 2000 Wnt-11 - Wnt-15 - Wnt-16 -

29

2.5.1.1 The canonical Wnt/β-catenin signaling pathway and the factors

involved

A series of genetic, cell biology and molecular biology studies have identified the natureand the relative order of the components with the Wnt/wingless signaling pathway(Cadigan & Nusse, 1997; Miller & Moon, 1996). The receptors of Wnts are members ofthe Fz family containing a CRD (Vinson et al, 1989; Leyns, et al, 1997). After binding toFz, the Wnt signal is transduced to a cytoplasmic protein, Dishevelled (Dvl). Uponactivation, Dvl then inhibits the activity of glycogen synthase kinase-3β (GSK-3β) bymodulating its enzymatic activity. In the absence of Wnt signals, a complex containingGSK-3β, Axin, and the tumor suppressor protein adenomatous polyposis coli (APC)promotes the phosphorylation of β-catenin and results in a reduction of cytoplasmic β-catenin levels (Fig. 4A). In the presence of Wnt signals, Wnts inactivate GSK-3β and thusinhibit the degradation of β-catenin. The stabilization of β-catenin results in itsaccumulation in cytoplasm. It either enters the nucleus, where it binds to Tcf/Lef, atranscription factor, and modulates specific gene expression, or it can bind to cadherins atthe plasma membrane, where it stabilizes cell-cell adhesive complexes (Fig. 4B) (forreviews see Miller et al, 1999; Thorpe et al, 2000).

Fig. 4. Simplified Wnt signaling pathway: (A) In the absence of Wnt, GSK-3β phosphorylatesβ-catenin, resulting in its degradation. (B) Conversely, Wnt signaling results in the inhibitionof GSK-3 β, the stabilization of β -catenin and the subsequent translocation of β-catenin to thenucleus. (After Nusse, 2001).

30

Fzs are seven-transmembrane proteins, and ten Fz members have been cloned. The Fzprotein was initially identified by mutation analyses in Drosophila melanogaster, wherethe mutations cause abnormal tissue polarity in the epidermis of the wings (Vinson &Adler, 1987). Recently, Fz5, which specifically synergized with Wnt2 and Wnt5a, wasfound to be essential for yolk sac and placental angiogenesis (Ishikawa et al, 2001). Dvlis a cytoplasmic protein and a positive regulator of the Wnt signaling pathway (Sokol etal, 1995). Dishevelled proteins possess three conserved domains: a DIX domain, presentin the Wnt antagonizing protein Axin (Zeng et al, 1997), a PDZ domain involved inprotein-protein interactions (Ponting et al, 1997), and a DEP domain found in proteinsthat regulate Rho GTPases. In the canonical Wnt pathway, the PDZ domain is necessary.In the mouse, three Dvl genes have been identified. Studies on transgenic mice suggestthat Dvl-2 is part of the Wnt signaling pathway involved in hair development (Millar etal, 1999), and Dvl-1 is required for the formation and/or function of specific neuronalpathways (Lijam et al, 1997). In the complex of Axin, GSK-3β, β-catenin and APC,GSK-3β efficiently phosphorylates β-catenin, APC and Axin, while Dvl preventsphosphorylation (Behrens et al, 1998). Axin binds GSK-3β and potently inhibits itsactivity in vivo. More interestingly, axin, in addition to facilitating β-cateninphosphorylation by GSK-3β, can also mediate the inhibition of GSK-3β in response toextracellular signals, such as Wnts (Hedgepeth et al, 1999). GSK-3β (zeste white-3/shaggy in Drosophila) was first identified as an inhibitor of glycogen synthase andsubsequently as a negative regulator of Wnt signaling. GSK-3β and its Drosophilahomologue, ZW-3/shaggy, are required for most Wnt functions in differentdevelopmental systems (for a review, see Ferkey & Kimelman, 2000).

Lithium has marked effects on embryonic development and patterning in differentorganisms. It has been demonstrated that lithium may mimic Wnt signaling by directinhibition of GSK-3β, leading to the accumulation of cytoplasmic β–catenin (Klein &Melton, 1996; Hedgepeth et al, 1997). One of the Wnt co-receptors is the cell-surfaceproteoglycan encoded by the division abnormally delayed gene (Dally). Dally may beinvolved in the formation of a multiprotein complex with Fz to regulate both short- andlong-range Wg signaling (Tsuda et al, 1999; Lin et al, 2000). Another Drosophilaglypican gene, dally-like (dly), is also involved in Wg signaling (Baeg et al, 2001).Overexpression of dly leads to an accumulation of extracellular Wg, which indicates thatdly plays a role in the extracellular distribution of Wg. The family of LDL-receptor-related proteins (LRPs) was recently found to be required for the canonical Wnt signalingpathway. A Drosophila mutant, arrow, which encodes a LRP, shows phenotypes similarto the wg mutant (Wehrli et al, 2000). LRP6 was found to be required for Wnt/β-cateninsignaling act as a co-receptor for Wnt (Pinson et al, 2000; Mao et al, 2001). Mice lackingLRP-6 exhibited developmental defects similar to those caused by deficiencies in variousWnt proteins. LRP-5, a close homolog of LRP-6, also functions as a co-receptor for Wntproteins in mammalian cells and can transduce the canonical Wnt signal by binding andrecruiting Axin to membranes (Mao et al, 2001). This establishes the physical linkbetween a Wnt receptor and one of the Wnt intracellular signaling mediators and providesinsights into the mechanism by which a Wnt receptor transduces its signals in thecanonical pathway. Several secreted frizzled-related proteins (sFRPs) that also contain aCRD have been identified, and some of these can bind and antagonize Wnt proteins(Lescher, et al, 1998; Ladher et al, 2000). The ECM components, such as type XVIII

31

collagen, were shown to contain frizzled-like motifs as well (Rehn & Pihlajaniemi, 1995;for a review, see Rehn & Pihlajaniemi, 1996). These findings suggest a possible linkbetween Wnt signaling and ECM. Other Wnt antagonists, such as Wnt inhibitory factor-1, Cerberus, FrzB, Dickopfs, and Drosophila Naked cuticle, bind to Wnts and block theinteraction with Fz proteins or LRPs (Glinka et al, 1998; Hsieh et al, 1999; Zeng, et al,2000; for a review, see Niehrs et al, 2001).

2.5.1.2 The Wnt/Ca2+ pathway, the non-canonical Wnt signaling pathway

More evidence has been reported to suggest that Wnts may be qualitatively different andpartly regulated at the level of receptors, and members of the Fz receptor family can alsosignal through other intracellular pathways distinct from β-catenin/TCF transcriptionalregulation (Slusarski, et al, 1997). This is called non-canonical Wnt signaling andcharacterized by an intracellular Ca2+-release initiated by some Wnts and Fzs in Xenopusand zebra fish embryos. The cellular response results from interactions of Ca2+-sensitiveenzymes, such as protein kinase C and Ca2+/calmodulin-dependent kinase II (CamKII),with their targets. Therefore, the Wnt/Ca2+ signaling pathway leads to a release ofintracellular calcium and an activation of protein kinase C (for reviews, see Miller et al,1999; Kühl et al, 2000). In Drosophila, the Frizzled-regulated polarity pathway of planarcells involves the small GTPases Rho and the Jun N-terminal kinase. A similar Wnt/JNKpathway exists in vertebrates to regulate morphogenetic movements, such as convergentextension during gastrulation. Therefore, Fz receptors can signal through β-catenin-dependent and -independent pathways.

2.5.2 FGFs and roles of FGFs pathway during organogenesis

FGFs are a family of potent heparin-binding polypeptides that determine the developmentof certain cells into mesoderm for the production of blood vessels, limb outgrowth, andthe growth and differentiation of numerous cell types (Gilbert, 2000). In mouse, there areat least 22 FGF genes, and most of them (FGFs 3-8, 10, 15, 17-19, and 21-23) haveamino-terminal signal peptides and are readily secreted from cells. The FGFs 9, 16 and20 lack an obvious amino-terminal signal peptide. FGF1 and FGF2 also lack signalsequences, but they can be found on the cell surface and within the extracellular matrix.FGF11-14 lack signal sequences and are thought to remain intracellular (for a review, seePowers et al, 2000). The phenotypes of the inactivation of FGF member in mice rangefrom very early embryonic lethality to subtle phenotypes in adult mice. FGF10inactivation causes defects in multiple organs, including limb, lung, thymus, pituitary,and ear (Min et al, 1998; Sekine et al, 1999; Ohuchi, et al, 2000). FGF9 mutant mice notonly exhibit lung hypoplasia and neonatal death but also show male-to-female sexreversal, which demonstrates FGF signaling also plays a novel role in testicularembryogenesis (Colvin et al, 2001a, b). The FGFs within a subfamily have similarreceptor-binding properties and overlapping expression patterns. Some functionalredundancy of FGFs is likely to occur. For example, FGF17 and FGF8 cooperate to

32

regulate neuroepithelial proliferation at the midbrain-hindbrain junction. Disruption ofFGF17 in mouse decreases precursor cell proliferation in the medial cerebellar anlageafter E11.5. The loss of an additional copy of FGF8 enhanced the phenotype andaccelerated its onset (Xu et al, 2000). FGFs function by activating a set of receptortyrosine kinases called FGF receptor tyrosine kinase receptors (FGFRs). FGFRs containtwo or three immunoglobulin-like domains and a heparin-binding sequence. The bindingof FGF to their receptors is accomplished with the aid of PG, namely syndecan (Jaakkola& Jalkanen, 1999). Alternative mRNA splicing of the FGF receptor gene specifies thesequence of the carboxy terminal half of the immunoglobulin domain III, resulting ineither the IIIb or the IIIc isoform of the FGF receptor. This alternative splicing isregulated in a tissue-specific manner and dramatically affects the ligand-receptor bindingspecificity (for a review, see Ornitz & Itoh, 2001). Exon IIIb is expressed in epitheliallineages, and exon IIIc tends to be expressed in mesenchymal lineages. Ligands specificfor these receptor splice forms are expressed in adjacent tissues, resulting in directionalepithelial-mesenchymal signaling.

2.5.3 Transforming growth factor superfamily

The transforming growth factor β (TGFβ) superfamily contains over 30 members ofstructurally related polypeptides, including the TGFβ family, the activin family, Bmps,growth differentiation factors (GDFs) and other proteins, such as GDNF (Gilbert, 2000)(Fig.4). The TGFβ family (TGFβ1, 2 and 3) is important in regulating the formation ofthe ECM and has been implicated in epithelial and mesenchymal interactions, e.g., in thebranching morphogenesis of the kidney and the mammary gland and in the inductiveevents of the transdifferentiation of mammary epithelial cells into mesenchymal cells(Miettinen et al, 1994; for a review, see Mummery, 2001). Mice lacking TGFβ1 exhibittwo different phenotypes. Approximately 50% of them die at mid-gestation due to defectsin yolk sac vasculogenesis and hematopoiesis (Dickson et al, 1995). The other embryossurvive beyond birth (Letterio et al, 1994), but then develop a severe multifocalinflammatory disorder and die at 3-4 weeks of age (Shull et al, 1992; Kulkarni et al,1993). TGFβ2-null mice have many developmental defects and die perinatally due tocongenital cyanosis (Sanford et al, 1997). TGFβ3-null mice have delayed lungdevelopment and die shortly after birth (Kaartinen et al, 1995). The activin-like factorsinclude vg-1, nodal, and nodal-related factors, which are all involved in the induction andpatterning of mesoderm in vertebrate embryos. Bmps were discovered as factorspromoting ectopic formation of cartilage and bone in rodents. The induction of boneformation is, however, only one of their many functions. Bmps have been found toregulate cell division, apoptosis, cell migration, and differentiation (Hogan 1996). Theyare also involved in the early body plan. There are a number of receptors for the TGFβsuperfamily. TGFβ family members elicit their cellular responses through cell surfacetype I and type II receptors. Five type II receptors and seven type I receptors, also termedactivin receptor-like kinases (Alk), have thus far been identified. The type II receptor is aconstitutively active kinase. In all cases, the ligand binds first to a type II receptor andenables it to form a complex with the type I receptor. The type I serine/threonine kinase

33

thereby becomes activated and transduces signals downstream (for reviews, see Derynck& Feng, 1997). Gene targeting experiments demonstrated that the TGFβ type I receptor iscrucial for the function of TGFβ during vascular development (Larsson et al, 2001).Activation of this receptor causes phosphorylation of Smad proteins in the cytoplasm.Smads1, 5 and 8 are targets for Bmp receptors. Smads 2 and 3 for activin receptors. Smad4 is required by both pathways, and Smad 6 is inhibitory to both by displacing thebinding of Smad4. The phosphorylated Smads then form complexes and translocate intothe nucleus to regulate gene expression.

Fig. 5. Members of the TGFβ superfamily. (After Gilbert, 2000)

2.5.4 Hedgehog family

Vertebrates have at least three homologues of the Drosophila hedgehog gene: sonichedgehog (Shh), desert hedgehog (Dhh), and Indian hedgehog (Ihh). Dhh is expressed inthe Sertoli cells of the testis, and mice homozygous for a null allele of Dhh exhibitdefective spermatogenesis (Clark et al, 2000). Ihh is expressed in the gut and in cartilage

34

and is important in postnatal development (Vortkamp et al, 1996). Shh is a prototypicalmorphogen known to regulate epithelial/mesenchymal interactions and embryonicpatterning during embryogenesis (Bitgood & McMahon, 1995; Gritli-Linde et al, 2001).Hedge peptide binds to a membrane receptor called patched. It is constitutively active andis repressed by ligand binding. When active, it represses the activity of another cellmembrane protein, Smo, which in turn activates a Gli-type transcription factor in such away that it can move to the nucleus and turn on target genes. In the absence of hedgehog,patched is active, Smo inactive, and gli inactive. In the presence of hedgehog, patched isinhibited, Smo is active, and gli is active. Mice lacking Ptc1 activity display constitutiveactivation of the hedgehog response genes in target tissues (Goodrich et al, 1997). Smomutants reveal redundant roles for Shh and Ihh signaling in the regulation of L/Rasymmetry and demonstrate an absolute requirement for hedgehog signaling insclerotomal development and a role in cardiac morphogenesis (Zhang et al, 2001). Shhcould also be an indirect angiogenic factor, since Shh is able to induce robustangiogenesis by inducing all the three isoforms of the vascular endothelial growth factorand angiopoietins (VEGF) from interstitial mesenchymal cells (Pola et al, 2001). Ihh isan endogenous signal that plays a key role in the development of the earliesthematovascular system, since it can re-specify the prospective neural ectoderm (anteriorepiblast) along the hematopoietic and endothelial (posterior) lineages (Dyer et al, 2001).The Shh partner, Smo, also contains the Fz domain. The fact that Wnts can specificallydown-regulate Shh and Ptc was shown recently in presumptive dental ectoderm (Sarkar etal, 2000), but there is no direct evidence of interactions between Wnt and Smo so far(Nusse, 1996).

2.5.5 Other growth factors

Some other factors, such as EGF-R ligands, platelet-derived growth factors (PDGF),hepatocyte growth factor (HGF) and insulin-like growth factors (IGFs) may each play animportant role during development. By acting through a common EGF-R the EGF-Rligands, which include EGF, TGFα, betacellulin, amphiregulin (AR) and heparin-bindingEGF (HB-EGF), are related to several organ developments, such as tooth, lung andkidney, (Miettinen, et al, 1995, 7, 9; Partanen et al, 1998; Okada et al, 2001; for a review,see Miettinen, 1997). HGF may function via the HGF/c-met pathway and collaboratewith FGFs in lung branching morphogenesis (Ohmichi et al, 1998).

2.6 Growth factors and matrix molecules in lung organogenesis

Lung development takes place within a complex milieu of cytokines, peptide growthfactors and matrix molecules (Hilfer, 1996; Warburton et al, 1998, 9; 2000) (Fig. 6A).These factors drive the lung bud through pseudoglandular stage, canalicular stage,terminal sac stage and alveolar stage. For details of the factors implicated in lungdevelopment see, http://www.ana.ed.ac.uk/anatomy/database/lungbase/lunghome.html.

35

Among these growth factors, Shh is the key factor in the control of branchingmorphogenesis (Pepipecelli, et al, 1998). It is widely expressed in the distal lungepithelium. Ectopic expression of Shh disrupts branching in the transgenic lungs, but nochange could be observed in Bmp4, Wnt2 and FGF7 gene expression (Bellusci et al,1997b). In the Shh deficient mice, Bmp-4 was distributed normally but at higher levels.Wnt7b expression was not altered, but Wnt2 was down-regulated. This supports theassumption that the primary target of Shh signaling is the lung mesenchyme and indicatesthat mesenchymal signaling is abnormal in Shh deficient mice (Pepipecelli, et al, 1998).

In the developing embryonic lung, expressions of Bmp5 and Bmp7 are detected in themesenchyme and the endoderm, respectively, whereas Bmp4 expression is restricted tothe distal epithelial cells and the adjacent mesenchyme. Constitutive expression of Bmp4in transgenic mice results in small lungs with grossly distended terminal buds andinhibition of epithelial proliferation (Bellusci et al, 1996). Therefore, Bmps, particularlyBmp4, are considered to be among the key growth factors essential for lung development.The pathways of Shh and Bmp4 were initially thought to be independent of each other inembryonic lung development, since Bmp4 does not change the following overexpressionof Shh (Bellusci et al, 1997b). The fact that the phenotypes of SP-C-Shh and SP-C-Bmp4transgenic lungs are very different suggests this hypothesis as well. Nevertheless, it hadbeen speculated that Shh secreted by the distal tip epithelium may regulate the expressionof Bmp4 and/or Wnt2 in the adjacent mesenchyme (Bellusci et al,1996). In line with thisspeculation, Bmp4 was found to be expressed at the normal sites but at higher levels inShh mutants, suggesting that enhanced Bmp4 signaling could contribute to the blockingof Shh signaling. Therefore, Bmp4 may coordinate with Shh and be involved in lungmorphogenesis by locally inhibiting epithelial proliferation at the tips of the end buds(Bellusci et al, 1996; Pepipecelli, et al, 1998).

Among the growth factors originating from lung mesenchyme, FGFs play distinctroles in regulating lung epithelial branching. A breakthrough in identifying themesodermal factors controlling lung morphogenesis came with the targeted expression ofa dominant negative FGF receptor (FGFR2IIIb) in the endoderm of transgenic mouseembryos (Peters, et al, 1994). FGFR2 is normally expressed in airway epithelium fromthe early embryonic lung bud stage through late fetal lung development (Peters et al,1992; Orr-Urtreger et al, 1993). In dominant negative FGFR2IIIb transgenic embryos, thetrachea and the primary bronchi developed normally, but no secondary or terminal budswere present. More evidence of the role of FGF in epithelial/mesenchymal signaling hasbeen obtained both in vitro and in vivo. FGF7 (KGF) and FGF10 were expressed in thelung mesenchyme. Ectopic FGF7 caused cystadenomatoid malformation in the fetal lung,associated with disrupted branching morphogenesis in vivo (Simonet, et al, 1995).Component deletion experiments by applying neutralized KGF antibody or antisenseKGF oligonucleotides in a mesenchyme-free tracheal bud matrigel culture showed thatFGF7 is necessary for lung branching morphogenesis as well. But FGF7-deficient micedevelop normally, indicating that FGF7 is not required for embryonic lung developmentor other factors may compensate to maintain normal development in its absence (Post etal, 1996).

FGF10, which has a higher affinity for FGFR2 (Arman et al, 1999), is expressed in acomplex and dynamic pattern in the mesenchyme near the positions where new buds areemerging. FGF10 appears to direct early bronchial branching and seems to be used

36

repeatedly to pattern successive rounds of branching. The buds grow toward areas ofFGF10-soaked, implanted bead, and ectopic branches grow out and target the bead(Bellusci et al, 1997a). Interestingly, FGF10 deficient mice show a striking phenotype -the absence of lungs with just a blind-ended trachea remaining. Therefore, FGF10 isabsolutely required for the normal patterning and development of lung epithelium (Min,1998; Sekine et al, 1999). FGF and Bmp signals may form a regulatory network and havebeen proposed as prime candidates for mediating the epithelial-mesenchymal interactionsin many organ systems (Thesleff et al, 1995; Martin, 1998; Tickle, 1999). It has beendemonstrated that Bmp4 and FGF10 play opposite roles during lung bud morphogenesis(Weaver et al, 2000). FGF10 and Bmp4 are expressed in developing lung in dynamic andcomplementary expression patterns (Bitgood & McMahon, 1995; Bellusci et al, 1996,1997a; Park et al, 1998). FGF10 upregulates Bmp4-lacZ expression in the endodermclosest to the bead. Exogenous purified Bmp4 added to the culture medium inhibitsFGF10 induction. However, the Bmp-binding protein Noggin enhances FGF-inducedmorphogenesis. FGF10 can induce Shh and mouse Sprouty in the same way as it inducesBmp4. Bmp4 inhibits epithelial proliferation and may hence limit branch growth. HGFshave also been implicated in normal and abnormal lung morphogenesis (Ohmichi et al,1998).

Besides Wnt2, other Wnts are also expressed in embryonic lung, such as Wnt2b,Wnt5a and Wnt7b (Zakin et al, 1998; Clark, et al, 1993). However, their functions duringlung organogenesis are not known. Members of the Sprouty family are inducible negativeregulators of growth factors that act through tyrosine kinase receptors and directlyantagonize FGF and EGF signals. In vitro and in vivo investigations of the mouse Sproutygene have demonstrated that Sprouty2 functions as a negative regulator of embryoniclung branching morphogenesis by inhibiting FGF10 (Mailleux et al, 2001). TGFβ1, 2, 3and TGFβR1 and TGFβR2 are differently distributed in embryonic lung. Both TGF-β1and TGF-β2 inhibit pulmonary branching morphogenesis. The TGFβ3-null mutationleads to an immature neonatal lung phenotype, which is rapidly fatal. Therefore,activation of TGFβ/TGFβR signaling negatively regulates the lung branchingmorphogenesis. (For a review, see Kaplan, 2000). EGF signaling is also important forlung branching morphogenesis. In EGF receptor mutant mice, the lung had impairedbranching and deficient alveolization, which resulted in a respiratory distress-likesyndrome (Miettinen, et al, 1995, 7; for a review, see Kaplan, 2000). Fetal rat pulmonaryepithelium undergoes proliferation and differentiation in the absence of mesenchyme,when cultured in the presence of various growth factors, such as EGF (Deterding &Shannon, 1995).

In embryonic lung, VEGF is expressed in the epithelium and its receptor, Flk1, isdetected in the mesenchyme. VEGF is required for lung mesenchyme vasculogenesis andit is suggested to be involved in lung morphogenesis as well (Breier et al, 1992, 5). SPC-VEGF transgenic mice die after birth and disrupt the branching morphogenesis (Zeng etal, 1998). TGFβ1 may perturb embryonic lung epithelial differentiation and inhibitvascular development in the lung by cross-talking with the VEGFs-Flk pathway (Zeng etal, 2001).

The morphogenesis and differentiation of the embryonic lung do not only dependupon reciprocal interactions between mesenchymal and epithelial cells, but also dependon cell-extracellular matrix interaction. A variety of ECM proteins, including collagen,

37

nidogen, laminins, fibronectin, HSPG, and chondroitin sulfate proteoglycan, are known tobe important in embryonic lung development (for a review, see Durbeej & Ekblom,1997). Type IV collagen and elastin co-distribute in the lungs during development and areconcentrated at the epithelial-mesenchymal interface (Mercer and Crapo, 1990).Incubation of lung explants in the presence of collagen synthesis inhibitors alters fetallung branching. Type VI collagen has been predicted to have an important structural rolein matrix organization and a biological function in mediating cell-matrix interactions. It isexpressed notably in mouse embryonic lung, and the staining was concentratedunderneath the epithelium (Marvulli et al, 1996).

Nidogen is an integral BM component produced by mesenchymal cells in the lung(Senior et al, 1996). It contains EGF domains, which interact with laminin and mayregulate lung development (Ekblom, 1996; Schuger, 1997). Laminin-1 is expressed in thedeveloping mouse lung by epithelial and mesenchymal cells and plays a role in branchingmorphogenesis. Different laminin sites are engaged in promoting lung organogenesis byserving different functions during development. The cross region of laminin-1 selectivelypromotes epithelial cell proliferation. The outer globular region of α1 and β1 chainsmediates BM formation and epithelial cell polarization. The inner globular region oflaminin β1 chain stimulates lumen formation (Schuger, 1997).

The laminin α2 chain is expressed in developing lung (Flores-Delgado et al, 1998). Itis deposited in the bronchial epithelial BM and adjacent to the peribronchialmesenchymal cells. Functional studies of laminin α2 chain have shown that its expressioncoincides with the period of active bronchial myogenesis (Virtanen et al, 1996), andlaminin α2 expression is reversible and can be switched on and off by altering the cell’sshape in culture in smooth muscle myogenesis of developing lung (Relan et al, 1999).Anti-laminin antibodies block epithelial polarization and inhibit airway development(Schuger et al, 1995). This provided powerful evidence of the functional importance ofthe specific BM-related molecules. Accordingly, it has also been shown thatundersulfated ECMs potentiate type II cell responses to certain growth factors, whileheavily sulfated ECMs retard the same response (Sannes et al, 1993).

Adhesion of cells to the ECM is critical for lung branching. The null mutation ofintegrin α1β1 leads to embryonic death prior to lung formation, while the null a3β1mutation results in reduced bronchial branching (Kreidberg et al. 1996). The expressionof fibronectin (FN) was most intense in the BMs of vessels and airways, suggesting thatFN may affect vessel formation, alveolar epithelial cell differentiation and lung growthand maturation (Roman et al, 1992). Elastin is developmentally regulated in developinglung and seems to influence the lung organogenesis (Pierce et al, 1995). Recently, studieson the morphological changes and the distribution of MMPs and TIMP2 in developinglung demonstrated that MMPs and their inhibitors might contribute to the formation ofairways and alveoli in fetal lung development (Fukuda et al, 2000).

Likewise, the key transcriptional factors, including the thyroid transcription factor-1(TTF-1), HNF-3β, and the hepatocyte nuclear factor family of forkhead homologues, theGATA family of zinc finger factors (such as Gata-6), Gli genes, homeodomain proteins,as well as the basic helix-loop-helix factors, have been implicated in lung branchingmorphogenesis, cell differentiation, and lung-selective gene expression (for reviews, seePerl and whitsett, 1999; Minoo, 2000; Costa et al, 2001).

38

Fig. 6. A schematic presentation of the expression patterns of different transcription factors,proto-oncogenes, growth factors, cell adhesion and ECM molecules in the early developinglung (A) and kidney (B). The dotted lines delineate induced mesenchymal cells around thetips of the ureteric bud. For abbreviations, see page 2 and the text.

2.7 Growth factors and matrix molecules in kidney development

The complex development of the kidney is governed by sequential cell and tissueinteractions, and multiple signaling systems have to be operative to coordinate thedevelopment between the epithelial compartment of the ureter and the metanephricmesenchyme (Saxén et al, 1986). The molecular basis of how the ureteric bud developsinto the complex tree-shaped structure of the renal collecting duct system remains unclear(Pohl et al, 2000a) This multiple complex process apparently involves the consequentlyactivated genes and both contact-mediated and secreted signals, such as growth factors,matrix molecules, and transcription factors (Saxén, 1987; Vainio & Müller, 1997) (Fig.6B). For details and updated information, see the web site of kidney developmentdatabase: http://golgi.ana.ed.ac.uk/kidhome.html.

Among these growth factors, GDNF plays a pivotal role in the process of ureteric budbranching morphogenesis (Sainio et al, 1997; for reviews see Sariola & Sainio, 1997;Schedl and Hastie, 2000; Davies, 2001). GDNF is a mesenchyme-derived signal forureter budding and branching. GDNF is expressed at high levels in the metanephricblastema prior to the invasion of the ureteric bud, and the inactivation of GDNF functionperturbs the initiation of ureteric bud and kidney development (Sanchez, et al, 1996;Pichel et al, 1996; Moore et al, 1996). Moreover, local application of the GDNF beadinduces an ectopic ureteric bud from the Wolffian duct (Sainio et al, 1997; Pepicelli et al,1997). The expression of c-ret is visible in the mesonephric duct and becomes restrictedto the growing ureteric tip as soon as induction has started (Pachnis et al, 1993). C-ret is

39

expressed in a region adjacent to GDNF expression, and in vitro studies have shown thatc-ret binds GDNF (Durbec et al, 1996). Mice lacking c-ret or its co-receptor GDNF-α1have the same phenotype as GDNF deficient mice (Jing et al, 1996). The importance ofGDNF/c-ret signaling has also been shown by ectopic expression experiments intransgenic mice (Srinivas et al, 1999a). Recently, it has been shown that impressivebranching of an isolated ureteric bud occurs in the absence of kidney mesenchyme, whenit is cultured in an appropriate matrix context with GDNF and soluble factors secreted bya metanephric mesenchyme cell line (BSN cells) (Qiao et al, 1999b). This suggests thatGDNF is the key soluble factor obtainable from the metanephric mesenchyme. TheGDNF function via c-ret contributes to the growth and proliferation of the ureteric budand it may be needed to regulate the expression of the other genes involved in renaldevelopment, such as PG. By binding to the cadherins present in the kidney mesenchyme,GDNF could contribute to the regulation of adhesion between the mesenchyme andepithelial cells (Sariola and Sainio, 1997).

Eya1, a transcription factor belonging to the homeobox gene family, and WT1, whichis encoded by the Wilms’ tumor suppressor gene, may be the upstream genes that regulateGDNF expression (Kreidberg et al, 1993; Xu et al, 1999). Mice mutated for Eya1 lackexpression of GDNF. However, in WT1 deficient, GDNF remains in the mutantmesenchyme, although the ureteric bud is not induced (Donovan et al, 1999). GDNFsignaling can be inhibited by Bmp4 (Miyazaki et al, 2000). Bmp4 is expressed in themesenchymal cells surrounding the Wolffian duct and the ureteric stalk, whereas thecandidate receptors for Bmp4, Alk3 and Alk6, are expressed in the Wolffian duct and theureteric epithelium. Embryos homozygous for null mutations of Bmp4 die between E6.5-10.0. But several defects, including cystic kidneys accompanied by urinary tractanomalies, were found in Bmp4 +/- embryo (Dunn, et al, 1997). Wnt 11 is known to be atarget molecule of GDNF-c-ret signaling (Pepicelli et al, 1997). The inhibition of GDNFsignaling by Bmp4 has also demonstrated the downregulation of Wnt11 expression at thetip of the ureteric bud. And furthermore, Bmp4 determines the site of initial ureteralbudding on the Wolffian duct and promotes the growth and elongation of the ureter.Human recombinant Bmp4 severely distorts the branching and growth of the ureteric budin vitro (Raatikainen-Ahokas, et al, 2000). In cultured kidney with Bmp4 present, theureteric tips displayed normal expression of c-ret, but WT1 and Pax2 expression in thenephrogenic mesenchyme was inhibited. Therefore, Bmp4 might distort uretericbranching primarily via molecular changes in the metanephric mesenchyme rather than inthe ureteric epithelium.

Several other Bmps, including Bmp2, 5 and 7, are also expressed during renaldevelopment (for reviews, see Hogan, 1996; Kuure et al, 2000). Bmp7 is highlyexpressed in the Wolffian duct, the ureteric bud and its branches and also in thepretubular aggregates. In Bmp7 deficient mice, few or no nephrons are formed, whichshows that a loss of Bmp7 function leads to renal defects likely to result in a progressiveloss of nephrogenic mesenchyme by apoptosis. The mutant mesenchyme progressivelyundergoes apoptosis, and the expression of WT1, Pax2 and Wnt4 genes is reduced(Dudley et al, 1995; Luo et al, 1995). Bmp-7 induces tubulogenesis when added touninduced metanephric mesenchyme (Vukicevic et al, 1996). Recent studies, however,demonstrated that Bmp7 signaling prevented apoptosis of the metanephric mesenchyme,but the metanephric mesenchyme is not competent to respond to signals promoting

40

tubulogenesis. In conjunction with FGF2, Bmp7 promotes growth and maintainsmesenchymal competence in vitro (Dudley et al, 1999). Bmp7 may cooperate with Wnt4to regulate the specification of tubular structures as well (Godin et al, 1998; for a reviewsee Kuure et al, 2000). Further insight into the role of Bmp7 has demonstrated that Bmp7exerts dose-dependent and opposite effects on the ureteric bud and the collecting ductmorphogenesis in vitro. At low doses (0.5 nM), Bmp7 stimulates tubular number, length,and branching, while at higher doses (10 nM), it generates shorter, unbranched tubules(Piscione et al, 1997, 2001). Bmp2 is expressed in the metanephric mesenchymal cellsadjacent to the ureteric bud branches and collecting ducts. Early embryonic lethality ofthe Bmp2 deficient mice has precluded the study of Bmp2 functions during renaldevelopment. Nevertheless, treatment of embryonic kidney explants with Bmp2 has beenfound to inhibit branching morphogenesis (Zhang and Bradley, 1996; Piscione et al,1997, 2001). Thus, Bmp2, 4 and 7 all have a survival effect on isolated metanephricmesenchyme, but they simultaneously have distinct effects on kidney development(Raatikainen-Ahokas et al, 2000).

Wnt proteins may function as the classic metanephric inducer because Wnt1-expressing cells are able to induce tubulogenesis in the metanephric mesenchyme(Herzlinger et al, 1994). Wnt1 may mimic another Wnt in the experiments, since Wnt1 isnot present in the developing kidney. Several Wnt proteins are expressed in the non-endogenous inducer spinal cord, suggesting that Wnts may be involved in metanephricinduction. At least 5 Wnt genes that are expressed in the developing kidney have beenidentified so far, including Wnt4, 6, 7b, 11 and 15. Wnt6, Wnt7b and Wnt15 expressed inthe ureteric bud and the collecting duct, Wnt11 at the ureteric tips, and Wnt4 in themetanephric mesenchyme and the tubules (Stark et al, 1994; Kispert et al, 1996, 8;Carroll et al, 2001; for a review, see Carroll and McMahon, 2000). Wnt4 mutants haveseverely hypoplastic kidneys, which implies that Wnt4 is necessary for tubulogenesis(Stark et al, 1994). Wnt4 is also a sufficient signal for nephrogenesis in vitro (Kispert etal, 1998).

In the absence of Wnt4, the metanephric mesenchyme condenses normally, and theearly induction markers WT-1, Pax-2 and N-myc are induced but it fails to aggregate intopretubule clusters and to undergo tubulogenesis. Wnt4 is a mesenchymal factor that isinvolved in the transition of tubular mesenchyme to epithelium and acts downstream ofthe initial inductive events by controlling cell adhesion possibly via cadherins expressedbetween induced pretubular cells (for reviews, see Vainio et al, 1999; Uusitalo et al,1999; Kuure et al, 2000). Interestingly, the ureteric epithelium of a mutant kidneyundergoes several rounds of branching, and the expression of ureteric gene markers, suchas c-ret and Wnt11, remain unchanged, suggesting that Wnt4 does not play a role in theureteric growth and branching process. Wnt7b may not be involved in the process oftubular induction, since it is expressed in the stalk rather than at the tips of the uretericbud. But the facts that Wnt7b is persistently present in the ureteric bud and the branchesderived from it, and that its amount increases during development toward adulthoodsuggest that Wnt7b may play a role in duct maintenance or growth rather than induction(Nguyen et al, 1999; for a review, see Carroll and McMahon, 2000). It has been shownthat Wnt expression and Wnt signaling are modulated by PG (Davies et al, 1995). Theinterference of PG synthesis leads to a loss of Wnt-11 expression at the ureteric tips(Kispert et al, 1996).

41

Some Wnt receptors, Fzs, are present at the early stages of nephrogenesis as well. Forexample, chick Frizzled-4 (cFz-4) was expressed in a similar pattern as Wnt4 in the chickmetanephron (Stark et al, 2000). The sFRPs are also normally expressed and servefunctions in the developing metanephros (Leimeister et al, 1998). sFRP2 is a target of theWnt4 signaling pathway in the metanephric kidney and may modulate Wnt4 signaling(Lescher, et al, 1998). Its expression is overlapped by Wnt4 during metanephricdevelopment, but is lost in the Wnt4 mutant metanephric mesenchyme and co-inducedwith Wnt4 in isolated metanephric mesenchyme by cells expressing Wnt4. sFRP1 isdistributed in the stroma in the metanephros. sFRP1 and sFRP2 may compete locally toregulate Wnt signaling during renal organogenesis. Both tubule formation and budbranching were markedly inhibited by sFRP1, but concurrent sFRP2 treatment restoredsome tubular differentiation and bud branching (Yoshino et al, 2001). Recently, oneelegant study has shown that the ureteric bud cell line secreted factors LIF, FGF2 andTGFβ2 are able to induce metanephric mesenchymal tubulogenesis (Barasch, et al, 1999;Plisov et al, 2001). LIF is expressed in the ureteric bud at E12.5, which is consistent withits role as an inducer. LIF deficient mice, however, do not show kidney abnormalities(Escary et al, 1993), suggesting that additional factors are required for the inducetion ofnephrogenesis in vivo. These may include members of the Wnt family, such as Wnt11, orcell surface proteins, such as integrins. The effect of LIF was abrogated by neutralizationof Wnt ligands with sFRP1, and the combined growth factor stimulation of explantsaugmented TCF1/Lef1 activation, suggesting that LIF and TGFβ/FGF2 function througha common Wnt-dependent mechanism.

It has been proposed that renal stroma is an important signaling center fornephrogenesis. Inactivation of the stromal-specific winged-helix transcription factor BF-2results in a defect in the ductal system and in nephrons (Hatini et al, 1996). Two otherstromal components, RARα and RARβ2, are involved in the controlling of the uretericbranching system. Retinoic acid signaling may be essential for maintaining stromalfunction. In mice carrying heterozygous mutations for stromally expressed RARα2 andRARβ2 receptors, defects in ureteric bud growth were observed (Mendelsohn et al,1999), including the downregulation of ureteric bud specific markers, such as c-ret andWnt11. Both in vivo and in vitro experiments suggest a role for FGFs in kidneydevelopment (Celli et al, 1998; Karavanova et al, 1996; Qiao et al, 1999a). Furthermore,FGF2 is one of a limited number of factors that induce ureteric bud cells to undergomorphogenesis (Sakurai et al, 1997). In transgenic mice, the ectopic expression of adominant negative form of FGFR causes kidney agenesis and a loss of Pax-2 expression.KGF is expressed in stromal cells adjacent to the truncal portions of the ureteric bud andthe collecting duct branches (Qiao et al, 1999a). KGF mutation results in decreasedgrowth of the ureteric bud and collecting ducts, while exogenous administration of FGF7augments ureteric bud growth (Qiao et al, 1999a). Therefore, FGF7 could be astimulating growth factor expressed by renal stromal cells. A recent analysis of Glypican3(Gpc3) mutant mice showed that Gpc3 modulates the actions of stimulatory andinhibitory growth factors, such as Bmp and KGF, during branching morphogenesis inembryonic kidney explants (Grisaru et al, 2001). Gpc3 is expressed in the ureteric budand the collecting duct cells. Gpc3 deficiency abrogated the inhibitory activity of Bmp2on branch formation in embryonic kidney explants, converted Bmp7-dependent inhibitioninto stimulation and enhanced the stimulatory effects of KGF.

42

The composition of the ECM and the cell matrix-interaction apparently influenceureteric bud branching morphogenesis and kidney organogenesis (Ekblom et al, 1981;Sariola et al, 1984; for a review, see Davies and Davey, 1999). The interactions betweenlaminin-1 and other BM components are important for epithelial morphogenesis in thekidney (for a review see, Durbeej & Ekblom, 1997). In the embryonic kidney, lamininhas been found in the induced mesenchymal cell condensates that later form kidneytubules, and in the BMs of vessels, nephric tubules and glomeruli (Ekblom et al, 1980a).Triggered by the branching ureteric bud kidney morphogenesis is preceded by thedisappearance of certain interstitial proteins from the induced areas, namely fibronectin,collagen type I and type III, which are replaced by a set of basal membrane proteins,collagen type IV, laminin, and a heparan sulfate rich proteoglycan . Experiments withchimeric kidneys have shown that in the tubular BM laminin is derived from theepithelial cells but, in the glomerular BM, it has a dual origin (Sariola et al, 1984). Onlylaminin B chain, but not the A chain, is expressed in the undifferentiated nephrogenicmesenchyme. Further investigations also show that the laminin A and B chains aresynthesized differentially in the embryonic nephrogenic tissue. The BM around theureteric bud is labeled by the antibodies against laminin A and B chains. In themesenchyme, laminin A chain appears when the mesenchyme converts into tubules(Holm et al, 1988). The laminin A chain is fundamental for the initiation of cell polarityduring kidney tubulogenesis, since antisera against the carboxy terminal of laminininhibited polarization, but antisera against the N-terminal of laminin failed to inhibitkidney morphogenesis (Klein et al, 1988). The laminin α1 chain is expressed in ratglomerular BMs and some other types of epithelial BM. But, β2 chains are only detectedduring advanced stages of glomerulogenesis and vascular development. The α5 chaincould be a major α chain of the glomerular BM (Durbeej et al, 1996). The distributionssuggest a key role for these molecules in the epithelial-mesenchymal interactionsnecessary for kidney development.

Transgenic mice lacking either integrin α3β1 or integrin α8β1 chains show reducedcollecting duct development, the α8β1 -/- phenotype being more dramatic than the α3β1 -/- (Kreidberg et al, 1996; Müller et al, 1997). Both of these deficients are interesting, asneither molecule is expressed in the duct itself, suggesting that the known ECM ligandsfor integrin α8β1 are either dispensable for or not involved in α8β1 signaling duringkidney development, or that the collecting duct phenotype is merely secondary tochanges in the mesenchyme. The phenotype also suggests the presence of an unknownligand. To solve the mystery, one extracellular matrix protein, nephronectin, wasidentified recently. Nephronectin, which contains five EGF-like repeats is expressed inthe ureteric bud epithelium and is believed to be the ligand mediating α8β1 function inthe kidney (Brandenberger et al, 2001; Miner 2001).

PG of the cell surface and the extracellular matrix, such as syndecan-1, are alsoimportant for ureteric growth and branching (Vainio et al, 1989; Pohl et al, 2000b). Iftheir glycoaminoglucan chains are removed or altered, developing ureteric buds fail togrow or branch (for a review, see Davies & Davey, 1999). No mesenchymal condensationoccurs if the heparan sulfate 2- sulfotransferase gene is inactivated (Bullock et al, 1998).In this mutant, Wnt11 is undetectable, and both c-ret and GDNF decrease rapidly. It hasbeen proposed that PG are involved in the branching and growth processes by attractinggrowth factors near the cell membrane and by making them readily avaible for

43

interactions with local receptors with higher affinity or to release them. Without normalglycoaminoglucans, for instance, fibroblasts become very much less responsive to FGF2(Rapraeger et al, 1991).

A growing number of reports suggest that MMPs and its inhibitor TIMPs play a role inrenal development. MT1-MMP (MMP-14), an activator of pro-MMP-2, is expressed in asimilar pattern to MMP-2 within immature nephron structures, while MMP-9 localizesonly to the invading vascular structures (Tanney et al, 1998). Both MT-1-MMP andMMP-2 antisense oligodeoxynucleotide (ODN) block the development of embryonicmetanephros in the culture. On the other hand TIMP-2 antisense ODN induced a mildincrease in the size of explants. Concomitant exposure of MT-1-MMP and MMP-2antisense ODNs induced profound alterations in the metanephroi. Treatment of TIMP-2antisense ODN to metanephroi exposed to MT-1-MMP/MMP-2 antisense notablyrestored the morphology of the explants. The MT-1-MMP antisense ODN also resulted ina failure in the activation of pro-MMP-2 to MMP-2. Therefore, MT-1-MMP, MMP2 andMMP9 modulate the organogenesis of the metanephros by mediating paracrine/juxtacrineepithelial-mesenchymal interactions (Kanwar et al, 1999). It has been reported that anti-MMP9 IgGs and the natural inhibitor of MMP9, TIMP1, impair renal morphogenesis(Lelongt et al, 1997). Strangely kidney development is normal in the MMP9-deficientmice (Andrew et al, 2000). To explain the discrepancies, it has been proposed that MMP9activity is possibly required for kidney development in vitro but not in vivo. There mayalso be another enzyme able to compensate MMP9 activity in vivo. Another explanationis that the antibodies and TIMP-1 may somehow, either directly or indirectly, disrupt thecell-cell or cell-matrix interactions that are necessary for branching morphogenesis andthis cause the phenotype in vitro (for a review, see Lelongt, et al, 2001).

3 Outlines of the present research

There is evidence to suggest that matrix molecules, such as laminin and collagen, interactwith other factors and play a role in the branching morphogenesis of developing organs.However, the transcription of lamininγ2 and γ2* in mouse and their biologicalsignificance has remained unclear. The newly identified type XVIII collagen contains aFz motif, which is implicated in the Wnt signaling pathway. The CRD is involved inligand binding as part of the Wnt receptor and could function as an antagonist for Wntsignaling when secreted. Powerful evidence is accumulating to show that Wnts and Fzplay very important roles during vertebrate development and mammalian organogenesis(Cadigan & Nusse 1997; Mao et al 2001). As type XVIII collagen has been discoveredrecently, its function during organogenesis and its possible correlations with signalingmolecules, such as Wnts are still being elucidated.

Wnt2b, initially called Wnt13, is a new member of the Wnt gene family. Wnt2b wascloned by PCR, and its cDNA fragment was found to share 66.9% amino-acid homologywith human and mouse Wnt2 (Katoh et al 1996). Its wide expression in many developinghuman fetal organs suggests that it may be involved in normal human development ordifferentiation. Recently, Wnt2b has been shown to function as an ectopic limb inducer(Kawakami et al, 2001). However, no other functional study on Wnt2b has been reportedso far.

Therefore, the aims for this research were:

1. to study the expression of laminin γ2 and γ2* in the developing kidney and theexpression of type XVIII collagen in the developing lung and kidney

2. to explore the functions of type XVIII collagen in the branching morphogenesis of lung and kidney and its possible roles in correlation with other signaling molecules during the patterning process of ureteric bud

3. to study the possible differences of the lung and kidney types of branchingmorphogenesis by using computerized morphological analysis and to provide someapplicable parameters for studying epithelial morphogenesis

4. to determine the mRNA transcription of Wnt2b gene during the differentdevelopmental stages of mouse embryo

45

5. to explore the functions of Wnt2b in kidney tubulogenesis and ureteric budbranching morphogenesis

4 Material and methods

More detailed descriptions of the materials and methods are presented in the originalarticles I-IV.

4.1 Mouse strains, organs and tissues (I-IV)

The CD-1, Rosa-26 (Soriano, 1999) and green fluorescent protein (GFP) (Hadjantonakis,et al, 1998) mouse lines were used for the experiments. The vaginal plug was used as acriterion for mating (0.5 days of gestation). Organs or tissues were isolated from embryosof different stages in Dulbecco’s PBS and/or cultured for different periods, depending onthe purpose of the experiments. The lungs were isolated from E9.5-14.5 embryos, and theurogenital blocks or kidneys were isolated from E10.5-15.5 embryos.

4.2 Organ culture methods (I-IV)

The organ cultures were performed according to Vainio et al. (1993). The dissected organrudiments were placed directly on slices of Nuclepore filter (pore size 0.1 mm, Costar)supported by stainless steel grids. The culture medium, i.e. Dulbecco´s modified Eagle´smedium (MEM, Gibco), was supplemented with 20% foetal bovine serum (FBS) (Gibco)and penicillin/streptomycin (GibcoBRL). The samples were incubated in 5% CO2 inhumidified air at 37 °C, and the medium was changed every other day.

4.3 Tissue recombination (II-IV)

The tissue separation and recombination experiments were mainly made as described inprevious publications (Vainio et al, 1993; Shannon et al, 1994). Most of the tissueseparation and recombination experiments were carried out on E11.5 mouse embryos ifnot otherwise indicated. The organ rudiments were incubated for two minutes in 3%

47

pancreatin/trypsin (GibcoBRL, UK) in Tyrode´s solution. The epithelium andmesenchyme of organ rudiments were separated microsurgically in the culture medium.Several different types of heterotypic and homotypic recombination were carried out,such as ureteric bud and lung mesenchyme (UB/LM), lung bud and kidney mesenchyme(LB/KM), ureteric bud and kidney mesenchyme (UB/KM), and lung bud and lungmesenchyme (LB/LM). Additional combinations included tracheal bud and lungmesenchyme (TB/LM), lung bud and tracheal mesenchyme (LB/TM), and spinal cordand kidney mesenchyme (SP/KM). GDNF (PeproTech, USA) was always added to theculture medium in the UB/KM recombination.

4.4 Hanging drop assays and agarose bead experiments (II, IV)

The hanging drop assay was conducted according to Vainio et al. (1992). The isolatedureter buds were washed before and after incubation in a hanging drop with 30-35 µl ofculture medium. When we tested the function of GDNF in triggering the ureteric budcompetence in response to lung mesenchyme, the medium in the hanging drop contained100ng/ml of human recombinant GDNF. Bovine serum albumin (BSA) (Sigma, USA)was used as control. When we studied the function of Wnt2b in a kidney reconstitutionassay, Wnt2b expressing cells dissociated by trypsin EDTA were used in the hanging dropculture. Cells expressing Wnt4 or Wnt6 and untransfected NIH 3T3 cells served ascontrols. In the kidney reconstitution assay, the subculture time in the hanging drop was 2to 4 hours. The bud was washed extensively to remove the cells and recombined withkidney mesenchyme that had been separated. We generated the agarose beads containingGDNF, FGF2, FGF7, FGF10 and BSA respectively by using affi-gel blue agarose beads(100-200 mesh, 75-150 nm diameter; Bio-Rad, UK) (Vainio et al, 1993).

4.5 Ureteric bud cultured with functional cell lines and kidney

reconstitution assays (IV)

To study the functions of Wnt2b signaling, two general approaches were applied. Onewas the culturing of the ureteric bud with 3T3 cells expressing Wnt2b. The isolatedureteric bud was placed on top of a monolayer of 3T3 cells expressing Wnt2b. 3T3 cellsexpressing Wnt6 or Wnt4, untransfected cells and normal culture media were used ascontrols. The other assay was the kidney reconstitution assay. Ureteric buds preincubatedwith 3T3 cells expressing Wnt2b or 3T3 cells expressing Wnt4 in the hanging drop assaywere recombined with kidney mesenchyme. Double staining of the brush-border antigenand cytokeratin Endo-A was used to visualize the tubules and epithelial structures,respectively. The numbers of branching tips and mesenchymal aggregates were used toestimate the extent of kidney reconstitution. 3T3 cells expressing Wnt2b were tested fortheir ability to induce kidney tubules similarly to Wnt4, and this was done according toKispert et al. (1998).

48

4.6 Antibodies, whole-mount and section immunostaining (II-IV)

The ELQ and Q36.4 antibodies against type XVIII collagen have been described bySaarela et al. (1998). ELQ recognizes all the three isoforms, whereas Q36.4 is directedagainst the two longest isoform (Fig. 3). The concentration of ELQ and Q36.4 antibodiesused for immunostaining was 20 µg/ml. The TROMA-I antibody against cytokeratinEndo A was from the Developmental Studies Hybridoma Bank (USA), and the antibodyagainst mouse pulmonary surfactant protein C (SP-C), m-20, was from Santa CruzBiotechnology, Inc. (USA). The antibody against the brush-border antigen was from Dr.Ekblom (Ekblom et al, 1980b). The secondary antibodies used were peroxidase-conjugated donkey anti-rabbit IgG and donkey anti-rat IgG, TRITC or FITC-conjugateddonkey anti-rabbit IgG, TRITC or FITC-conjugated donkey anti-rabbit IgG and BiotinSP-conjugated mouse anti-goat IgG (Jackson Immuno Research Laboratories, Inc, USA).The normal bovine serum was also from Jackson Immuno Research Laboratories, Inc,and the normal goat serum from DAKO (Danmark).

Samples cultured on a nuclepore filter were fixed in methanol-dimethyl sulfoxide(DMSO) (4:1) overnight at +4°C. The whole-mount immunostaining was done accordingto Marti et al, (1995). Diaminobenzidine (DAB) and aminoethylacarbazole (AEC)(ZYMED, USA) were used as substrates for the peroxidase-conjugated secondaryantibodies. The single or double labeling for whole-mount immunofluorescence wasaccording to Sainio et al. (1997). The β galactosidase / antibody double staining wasaccomplished by performing the β-galactosidase staining first, followed by the whole-mount immunostaining. For immunohistochemistry, the samples were fixed overnight in4% paraformaldehyde (PFA), after which 8 µm paraffin-embedded serial sections werecut. The SP-C antigen was localized according to Hsu et al. (1981).

4.7 Inhibition of Type XVIII collagen function by a specific antibody

(II)

Tests were performed to determine the active concentrations of the ELQ antibodyrequired for the blocking experiments in organ culture, and thereafter a concentration of120 µg/ml was used. For the UB/LM recombinants, isolated lung mesenchymes werepreincubated for one hour after separation in a medium containing 120 µg/ml of ELQ,recombined and subcultured in the presence of the antibody. Antibody binding wasvisualized by adding peroxidase-coupled anti-rabbit IgG as a secondary antibody.

4.8 Probes, whole-mount and section in situ hybridization and

staining of explants (I, II, IV)

The details for preparing the laminin γ2 and γ2* probes and the Wnt-2b probe is describedin the articles I and IV, respectively. Probes specific for all mRNAs encoding the mouse

49

a1l (XVIII) collagen chain (Rehn and Pihlajaniemi, 1995) was used. The other probeswere: Shh (Bellusci, et al, 1997b), Wnt11 (Kispert, et al, 1996), Wnt2 (McMahon andMcMahon, 1989), Wnt7b (Parr, et al, 1993), Sox-9 (Wright, et al, 1995), Fgf10 (Bellucsiet al, 1997a), Bmp4 (Bellusci et al, 1996), Gli 1-3 (Motoyama et al, 1998) and CC10(Whitsett, 1998). All of these were obtained as gifts. A probe for mouse SP-C wasobtained as an expressed sequence tag cDNA (Gbacc AA511605, Genome Systems,USA) and was confirmed by sequencing.

Whole-mount and radioactive in situ hybridization, the synthesis of non-radioactiveand 35S-labelled RNA probes were described in the articles II & IV. The double stainingfor mRNA and protein was also based on the protocol described by Parr et al. (1993),with some modifications. The β-galactosidase stainings of cultured whole-mount sampleswere performed according to Nonchev et al. (1996) and those for section samplesaccording to Bobola et al. (1995). For histological analysis, the whole-mount stainedexplants were embedded in glycolmethacrylate (JB-4; Polysciences Inc, USA) andsectioned at 8-10 µm.

4.9 RNA isolation and RT-PCR (I, IV)

RT-PCR was used to verify laminin γ2, γ2* and Wnt2b expression. For Wnt-2b RT-PCR,embryonic lung and kidney at E11.5, E12.5 and E13.5 were dissected in Dulbecco’s PBSand stored in RNA later (Ambion) until used for RNA isolation. Total RNA was isolatedwith the RNAeasy Mini Isolation Kit (Qiagen), and RT-PCR was performed with theRobust RT-PCR kit (Finnzymes) according to the manufacturers’ instructions.

4.10 Photography and Image Analysis (I-IV)

The photography of the whole-mount stained samples and the radioactive in situ isdescribed in the articles II, III and IV. A Leitz confocal microscope (Leica, Germany) wasused for monitoring fluorescence in the explants. For the time lapsed studies, therecombinants of GFP-UB/KM and GFP-UB/LM samples were photographed every 24hours. Digital photomicrography was used to acquire the images. Image analysis wasperformed using the Scion Image analysis software (version beta 3b, Scion corporation,USA), which was also used to generate the skeletons and to measure the parameters. Thecomposites were generated using the Adobe Photoshop v5.0 program.

4.11 Cell proliferation assay (II)

The apoptotic cells were visualized by using an in situ cell detection kit (BoehringerMannheim). See article II. Bromodeoxyuridine labeling and detection of the explants wascarried out using a cell proliferation kit (RPN 20, Amersham Life Science, UK)according to Sainio et al, (1997).

5 Results

5.1 Laminin γ2 and γ2* expression in developing kidney rudiments (I)

Kidney is one of the most extensively studied tissues regarding laminin expression. InPCR analysis, both the longer γ2 and the shorter γ2* transcripts were found to beexpressed at the E11, 12, 13, 15, 16 and 17, newborn and adult stages. However, theamount of the γ2 transcript was constantly higher than that of γ2* transcription (Fig. 3, I).To further elucidate the biological function of γ2 and γ2* transcription, a whole-mount insitu analysis was carried out. The laminin γ2 transcript was strongly expressed in theepithelial tissue of the ureteric bud and its branches. After branching, gene expressionwas intensely localized to the growing tips. Expression was also seen in the Wolffianduct. In contrast to the γ2 chain transcript, the γ2* transcript was only localized to themeso- and meta-nephrogenic mesenchymal cells adjacent to the inductive epithelium.The short γ2* transcript was spatially adjacent to the γ2 transcript. However, itsexpression was weak. No expression was detected in the perinephric region of themetanephros (Fig. 4, I).

5.2 Expression of type XVIII collagen in embryonic lung and kidney

(II)

At the initiation of lung and kidney organogenesis (E10.5), when the epithelial bud growsinto mesenchyme, type XVIII collagen mRNA is uniformly present in the epithelial bud inboth of these organs. Type XVIII collagen is seen on the basal side of the epithelium andin some endothelial cells of the lung mesenchyme. During the formation of the firstbranches, type XVIII collagen expression localized to the lung epithelial tips and was lostfrom the stalk area, while an opposite pattern was observed in the kidney, where bothtype XVIII collagen protein and mRNA were lost from the branching ureter tips. Thisdifferential expression pattern persisted during the subsequent developmental stages (Fig.

51

1A-J, II). The different expression patterns of type XVIII collagen were compared tothose of type IV collagen, which showed no difference in the epithelial expressionpatterns between the lung and the kidney. Therefore, the differences in type XVIIIcollagen expression pattern between the lung and the kidney correlate with the type ofearly epithelial development.

5.3 Specific branching pattern in embryonic lung and kidney (II, III)

The spatial and temporal appearance of early epithelial buds was followed in vivo and invitro in mouse lung and kidney. Image analyses of the organ rudiments revealed specificbranching patterns in them. In the cultured lung, several secondary lateral epithelial sidebuds grow out of the two primary branches, initially in an unbranched state. Since at leastthree side branches usually grow, the branching is called the 1-3 type, i.e. three sidebranches develop from one previous one. During the later stages of development (E11.5 +72 hours) of subculture, the branches of the lung develop according to the 1-3 type andinvariantly in view of position, whereas a few other types of branching also take place atthe later stages (Fig. 2B, 2C, arrows, III). Therefore, studies of the time course of the lungtype of branching and epithelial skeletonization were performed. Statistical analyses ofsamples cultured for 72 hours lung revealed that over 80% of the branches were of the 1-3 type, i.e. at least three lateral epithelial buds were growing out of a primary branch.This type of branching, in which the majority of branches are of the 1-3 type, is referredto as "lung-type" branching. The percentage for the module is also termed as thepercentage of the module type (PMT).

In the kidney the secondary branching is mainly dichotomous after the formation ofthe first primary branch and tends to follow the 1-2 type, in which each tip generates twonew ones (Fig. 1L-V, II; Fig. 1-3, III) (also see Saxén, 1987). After culturing for 72 hoursfrom E11.5, side branches also appear in the kidney sample (Fig. 3B, 3C, arrows, III). Bycontrast to lung, an average of 88% of the ureteric branches in the kidney weredichotomous, i.e. each epithelial tip had generated two new tips. In other words, the PMTin developing kidney is 88%. Therefore, the 1-2 type was defined as "kidney-type"epithelial development (Table 3A, III).

In addition to calculating the PMT, other parameters were further developed tocharacterize the lung or kidney type of epithelial development, e.g. the number ofbranched tips, the length of the left and right main branches, the first angle of thebranched epithelium, and the position index of the first branch (PIFB). All of theparameters demonstrated the differences between the lung and kidney types of epithelialbranching morphogenesis (Table 1, 2, III; Fig. 5A, 5B, III). Therefore, embryonic lungand kidney possess their own distinct branching patterns during the process of epithelialbranching morphogenesis.

52

5.4 Lung mesenchyme is sufficient to completely respecify type XVIII

collagen expression in the ureteric bud towards the lung type (II, III)

The different expression of type XVIII collagen during epithelial morphogenesis suggestsa role for epithelial-mesenchymal interactions in the process of type XVIII collagenpatterning, and this was tested by the tissue recombination approach (Fig. 2A-C, II).Control experiments showed that separation and recombination of the respectiveepithelial bud and mesenchyme restored the lung specific (tip type) or kidney specific(stalk type) of type XVIII collagen expression. In heterotypic recombinants between thelung mesenchyme and the ureteric bud, type XVIII collagen expression disappeared fromthe stalk and was instead localized in the epithelial tip cells, constituting the early lungtype. This was observed both with the antibody ELQ reconizing all the type XVIIIcollagen variants and with the Q36.4 antibody (Fig. 2D-G, II). The anti-long Q36.4specifically recognizes only the two longest type XVIII collagen variants (Saarela, et al,1998). Interestingly, even the long variants were respecified to a lung-type location in theepithelial tip cells, even though these two isoforms are not expressed normally in earlyembryonic lung (data not shown). Hence, repatterning also occurred with the variantsdirected by separate promoters, suggesting that the signals involved in the patterningprocess also integrate variant-specific mechanisms of type XVIII collagen generegulation.

5.5 Respecification of type XVIII collagen was associated with early

lung type ureteric bud branching morphogenesis and induction of SP-

C (II, III)

To test whether the respecification of type XVIII collagen expression in the ureteric budwas accompanied by other properties associated with the lung, we first monitoredepithelial morphogenesis by image analysis. As expected, the early lung type of epithelialbranching (mostly 1-3 type branching) and the kidney type (typically 1-2 type branching)were observed in homotypic lung and kidney recombinants, while all the UB/LMrecombinants in which morphogenesis had begun (n = 38) showed ureteric branching thatresembled that of early invariant lung more than the kidney type. More specifically, 84%of the branches in the recombinant samples cultured for 120 hours (n = 23, 235 branchescalculated) showed the 1-3 type branching, while only 16% showed the 1-2 typebranching. Calculations of the average tip number and PIFB values provided furtherevidence of the changes in morphogenesis and the reprogramming of the ureteric bud. Allthe values of a heterotypic recombinant were significantly different from those of kidneysamples but similar to those of lung samples (Fig. 2H-M; Table 3B, Fig. 6, III).

SP-C, a marker for type II pneumocytes, is expressed in the distal lung bud epithelium.The SP-C mRNA and protein were expressed in the embryonic lung but not in the kidney,whereas other differentiation markers, TTF-1 and CC-10 were present in both of theseorgans and are not organ-specific in this context. SP-C gene expression was also induced

53

in the ureteric bud when it was grown together with lung mesenchyme (Fig. 6A-G, II),whereas no significant changes were recorded in the expression of the HNF-3β, or CC-10genes (data not shown). Therefore, the respecification of type XVIII collagen inheterotypic tissue recombinants was accompanied by changes in the branching pattern,and suggest that lung mesenchyme may function as an instructive inducer tissue andchange the status of cell differentiation in the ureteric bud.

5.6 Type XVIII collagen is important for lung development and

ureteric bud patterning (II)

Anti-all type XVIII collagen (ELQ) was used to test the role of the type XVIII collagenrepatterning process in reprogramming the ureteric branching profile. Lungorganogenesis was noemal after addition of 120 µg/ml of preimmune IgG to the culturemedium (data not shown), while the same concentration of anti-all ELQ reduced lungbranching (Fig. 3A-C, II), but had no clear effect on kidney development (data notshown). During the three-day culture period the number of lung epithelial tips increasedin the preimmune IgG-treated explants, while the treatment with anti-all ELQ resulted inan average of 34% decrease in the tip number (Table 1, II). To monitor for signals thatcould be involved in the functioning of type XVIII collagen, we screened for theexpression of three genes implicated in lung development, Wnt2, FGF10, and Shh, aftertype XVIII collagen antibody blocking. Wnt2 expression was markedly reduced in thelung mesenchyme, whereas no difference was detected in the expression of FGF10 orShh (Fig. 3D-N, II). The smaller size of the lung samples cultured in the presence of typeXVIII collagen antibody suggested changes in cell proliferation and/or apoptosis. IgG-treated control lungs contained apoptotic cells in the distal tip mesenchyme but not in theepithelium, whereas the whole mesenchyme and epithelium were undergoing apoptosis inthe ELQ-treated lung. Consistent with the increase in apoptosis, also cell proliferation inthe epithelium and mesenchyme was reduced in the ELQ-treated explants (Fig. 3O-R, II).Furhtermore, ColXVIII antibody treatment also impaired bung branching. Hence, typeXVIII collagen is important for lung development.

Analysis of the UB/LM recombinants indicated that branching of the ureteric budremained unchanged within lung mesenchyme in the presence of preimmune IgG. The 21out of 36 explants that had initiated growth represented the average ureteric growthwithout any IgG treatment. Unexpectedly, the type XVIII collagen antibody almostcompletely blocked morphogenesis in the heterotypic tissue explants: only one out of the37 explants initiated development (Fig. 3S-U, II; Table 2, II). This suggests that typeXVIII collagen itself is required for the ureteric bud to initiate epithelial development inheterotypic recombinants and probably it has a major role in the epithelial patterning.

54

5.7 Search for the mesenchymal factor and the epithelial mediator

involved in the ureteric bud and type XVIII collagen repatterning

process (II)

Our attempts to identify the mesenchymal signals involved in the repatterning of typeXVIII collagen led to the conclusion that FGF10 is essential for lung organogenesis andis expressed specifically in mesenchymal cells, but does not appear to be lung-specific. Itwas also expressed weakly in the kidney mesenchyme, and its expression persistedadjacent to the ureter bud in the recombinants (Fig. 5P-R, II). Wnt2b was also present inmesenchymal cells in both organs and remained expressed in the recombinants, whereasWnt2 (McMahon and McMahon, 1989) was expressed only in the lung mesenchyme inboth homotypic and heterotypic recombinants (Fig. 5S-U, II). The role of the distal tipmesenchyme and certain growth factors in bud formation and in promoting type XVIIIcollagen expression in the lung epithelium was further tested using the tracheal tube as amodel system (Shannon, 1994). The distal tip mesenchyme induced ectopic trachealepithelial budding and type XVIII collagen expression, suggesting a role for type XVIIIcollagen in epithelial budding, whereas neither FGF2, FGF7, FGF10 beads nor Wnt2bcells alone could replace the distal tip mesenchyme in these functions. These beads werealso unable to reprogram the expression of type XVIII collagen in the ureteric bud ofrecombinants when implanted in the lung mesenchyme (Fig. 5V-Z´, II). Therespecification of type XVIII collagen expression apparently involves factors that act inthe ureteric bud to mediate responses to mesenchymal repatterning signals. Screeningwas therefore initiated for the secreted factors implicated in lung and kidneydevelopment. Of these factors, the expression of a morphogen, Shh, correlated with typeXVIII collagen expression in lung and kidney. Respecification of type XVIII collagen inthe tissue recombinants was associated with a change in Shh expression in the distal tipcells, and thus Shh in the epithelial bud was repatterned completely by the lungmesenchyme in a manner similar to type XVIII collagen (Fig. 5A-C, II). Several otherepithelial markers, including Wnt7b, Wnt11 and Sox9, and also Gli1-3 and Bmp4, failed todemonstrate such a repatterning response in the recombinants (Fig. 5D-O, II). Lamininγ2, which was one of the matrix molecules under study in this work, was also tested inthe heterotypic tissue recombinants. It is expressed in the ureteric bud in normaldeveloping kidney, and no significant changes were observed in the UB/LM heterotypictissue recombinants (data not shown). Hence, the factor involved in patterning type XVIIIcollagen may be different or combinationally signaling, and Shh is a candidate signal tobe involved in the control of tissue-specific morphogenesis and type XVIII collagenexpression.

55

5.8 Wnt2b gene is expressed in several epithelial-mesenchymal

derived organs (IV)

To explore the potential role of Wnt2b during organogenesis, its expression in mouseembryos at E11.5-15.5 was examined. Wnt2b is expressed in several organs or tissues(Fig. 1, IV; Fig.2, IV). The most interesting fact to us is that the Wnt2b gene is expressedin several epithelial-mesenchymal derived organs, such as limb, lung and kidney. In thekidney, weak expression is seen in the perinephric cells that encapsulate thedifferentiating mesenchyme at E11.5 and the mesenchymal cells in the region close to thestalk of the ureter. The expression area of Wnt2b is close to site of BF-2, a winged-helixtranscription factor and also a marker of presumptive stromal cells (Fig. 2D-F, IV). Itsexpression in kidney was undetectable by whole-mount in situ hybridization at E13.5, butRT-PCR analysis confirmed the presence of transcripts at E11.5, E12.5 and E13.5 kidney.Wnt2b expression is most closely associated with cells that either encapsulate developingorgans or lie a few cell layers away from the epithelial tissue in the mesenchyme.Therefore, the expression pattern of Wnt2b suggests that it may play a role in the controlof epithelial or mesenchymal morphogenesis during organogenesis.

5.9 Wnt2b fails to induce tubulogenesis in co-cultured kidney

mesenchyme but promotes ureteric bud survival and growth (IV)

The 3T3 cellsexpressing Wnt2b protein were generated to assess whether Wnt2b couldhave paracrine effects on neighboring tissues. As Wnt2b is expressed in the kidneymesenchyme, and as Wnt signaling has previously been implicated in the induction ofkidney tubules (Stark et al, 1994; Kispert et al, 1998), we were interested to test if Wnt2bcould act as an inducer of tubulogenesis. The kidney mesenchyme was separated from theureteric bud and placed on a monolayer of NIH 3T3 cells expressing Wnt2b or Wnt4 orin contact with a piece of dorsal spinal cord. Co-culturing of the mesenchyme sampleswith cells expressing Wnt2b did not induce tubules (Fig. 3, IV). The expression of Wnt2bin the mesenchyme and its lack of activity in kidney tubule induction assays suggestedthat it might alternatively function as a reciprocal signal to control the growth andbranching of the epithelial bud. To test this option, we analyzed the influence of Wnt2bon isolated E11.5 kidney ureteric buds by culturing them on a monolayer of cellsexpressing Wnt2b. Neither the control NIH 3T3 cells nor the 3T3 cells that expressedWnt6, another Wnt protein, stimulated ureteric growth. In contrast, the fibroblasts thatexpressed Wnt2b protein had a significant effect on epithelial survival and growth (Fig.4A-C, IV). The effect of 3T3 cells expressing Wnt4 on ureteric branching was alsoinvestigated. A degenerating epithelial remnant was detected in 24% of the 58 samples,and when detected, the bud was more rudimentary than that grown on control 3T3 cells.The rest of the samples completely lacked the ureteric bud (Fig. 4D, IV). Therefore,Wnt2b signaling is operational during kidney organogenesis. It does not play a direct rolein kidney tubule induction, but may specifically play a role in the control of epithelialgrowth and branching.

56

5.10 Recombination of Wnt2b-pre-treated ureteric bud tissue with

isolated nephrogenic mesenchyme results in a recovery of

organogenesis (IV)

The ureteric bud epithelium has been thought to be the primary inducer of nephrogenesisin the developing kidney (Saxén, 1987). However, when separated from the mesenchymeand cultured for any length of time, the ureter is a relatively poor inducer oftubulogenesis in reconstituted kidney explant cultures. Having observed that Wnt2bsignaling may promote epithelial proliferation and branching, we developed a hangingdrop assay to test more directly its role in kidney organ culture (Fig. 5, IV). The separatedureteric bud was preincubated in a hanging drop with normal medium or mediacontaining lithium, normal 3T3 cells, or 3T3 cells expressing Wnt2b or Wnt4 for variouslengths of time before washing and recombination with kidney mesenchyme. Whenisolated ureteric buds were incubated in normal medium for 4-6 hours and thenrecombined with freshly isolated kidney mesenchyme, no reconstitution of kidneyorganogenesis occurred in most cases, which was consistent with the previouslypublished reports (Sariola et al, 1989). Similarly, preincubation of isolated ureteric budswith NIH 3T3 cells was not sufficient to promote the recovery of good kidneyreconstitution. On the other hand, preincubation of the ureter with cells expressing Wnt2bfollowed by recombination with kidney mesenchyme led to a remarkable recovery andreconstitution of in vitro kidney development. The Wnt2b-treated ureteric bud epitheliainitiated branching morphogenesis and tubules were induced, as judged by doublestaining of the reconstituted explants with antiserum against the brush border andcytokeratin Endo-A (Table 1, IV; Fig. 6A-C, IV). The timing of the reconstitutionresponse of the ureteric bud was also monitored. Two out of six samples (33%) initiateddevelopment and poor reconstitution after a 2-hour preincubation with cells expressingWnt2b, whereas 12/13 (92%) initiated organogenesis and good reconstitution after a 4-hour preincubation of the isolated ureteric bud. We conclude that preincubation of theureteric bud with 3T3 cells expressing Wnt2b not Wnt4 promotes ureteric branching andsubsequent tubule induction in kidney mesenchyme in vitro (Fig. 6D, IV; Fig. 7, IV).

Having demonstrated that Wnt2b acts to induce ureteric development, we wereinterested to find out whether the ureteric bud would also be patterned normally. C-ret,Wnt11 and Sox9, which are expressed at the tips of the ureteric bud in normallydeveloping kidney explants, were located similarly in the reconstituted organs. The datasuggest that patterning of the epithelium in the reconstituted explants occurs normally,confirming that the fetal kidney mesenchyme plays an important role in maintaining thecapacity of the ureteric bud to undergo branching morphogenesis, and furtherdemonstrating that the effect of the mesenchyme can be mimicked by Wnt2b (Fig. 8A-F,IV).

57

5.11 Lithium is also sufficient to promote ureteric branching in the

reconstituted kidney in a manner comparable to Wnt2b signaling (IV)

To test whether Wnt signaling could act directly on the ureteric bud epithelium toregulate bud development, we applied lithium treatment, which has been reported toactivate the Wnt signaling pathway by inactivating glycogen synthase kinase 3β (GSK-3β) (Hedgepeth et al, 1997). A separated ureteric bud was incubated for 30 minutes in asolution containing 10 mM lithium chloride, after which the lithium was washed awayand the ureteric bud was recombined with freshly isolated kidney mesenchyme, as in thecase with the cells expressing Wnt2b. Similarly to the effects of Wnt2b, transientincubation of a ureteric bud with lithium was sufficient to promote good reconstitution ofkidney organogenesis. The explants treated with lithium initiated organogenesissuccessfully with isolated mesenchyme. Staining of the reconstituted explants withantiserum against the brush border and ureteric tip marker Sox9 confirmed that tubuleshad been induced and the ureteric bud epithelia was patterned normally in thereconstituted explants (Fig. 6E-G, IV; Fig. 7, IV).

5.12 Reconstitution was significantly better in Wnt2b or lithium-

treated explants than control or Wnt4-treated samples (IV)

The development of explants preincubated with normal NIH 3T3 cells, media control,cells expressing Wnt2b, lithium or cells expressing Wnt4 was monitored after 96 hours inculture, and the success rate of the initiated morphogenesis, i.e. tubulogenesis andureteric branching, was scored. Explants with Wnt2b or lithium-pre-treated epithelia werereconstituted, both quantitatively and qualitatively, significantly better than explantstreated with normal NIH 3T3 cells, normal media or cells expressing Wnt4. Somereconstitution, albeit poor, occurred in the media control samples and those grown withNIH 3T3 cells or cells expressing Wnt4. The first two cases resulted in over 30% poorerreconstitution, as only 17% of the samples in which the ureteric bud had been pre-treatedwith Wnt4 initiated branching and showed induction of tubules, while the rest eitherfailed to branch at all or branched poorly, so that none of these cases showed expressionof Brush Border Antigen. It appears that the isolated ureteric bud is still sufficient toinitiate some tissue interactions if it is completely separated from the mesenchyme priorto recombination (see also Grobstein, 1955). However, reconstitution takes place at amuch higher frequency and more completely when the bud is treated with cellsexpressing Wnt2b or lithium. Wnt4 signaling does not support reconstitution of kidneyorganogenesis via signaling to the ureteric bud, but rather restrains it (Fig. 7, IV).

6 Discussion

6.1 Role of type XVIII collagen in epithelial development

The mechanisms by which the rather limited number of embryonic signaling moleculesform networks and regulate the spatial organization of cells during morphogenesis remainpoorly understood (Hogan, 1999). To address these questions, two embryonic modelsystems were used, the lung and the kidney. These organs undergo characteristicepithelial branching morphogenesis. The present results show that the expression ofcollagen XVIII correlates with the type of epithelial development. At the initiation ofkidney and lung organogenesis, type XVIII collagen is expressed throughout theepithelial bud. After this common stage, its pattern in the lung is restricted to theepithelial tips, whereas in the kidney it is localized in the epithelial stalk region. Hence,collagen XVIII is distributed in opposite epithelial regions after the initial stages oforganogenesis and its localization correlates with differences in epithelial development.

The characteristic expression of type XVIII collagen in the lung and kidney epitheliumprovided an opportunity to study the correlations of signaling molecules and the organ-specific form and patterning of type XVIII collagen in the tissue recombinants ofUB/LM. We propose that the lung mesenchyme is able to reprogram the development ofureteric bud epigenesis towards the early lung type. We support our argument by thefollowing findings: 1) the embryonic lung mesenchyme completely respecified typeXVIII collagen expression in the ureteric bud from the epithelial stalk to the distal tip, 2)the shift in type XVIII collagen expression correlated with that seen in the expression ofShh and SP-C, 3) the ureteric bud combined with lung mesenchyme lost its potential topromote endothelial growth into the mesenchyme on a chorioallantois membrane (datanot shown), which has been reported to be a typical feature of angiogenesis in the kidney(Saxén, 1987), 4) epithelial morphogenesis changed from mostly dichotomous branchingof kidney towards the lung type, in which several side branches grow out of the primaryepithelial buds. The values for the applicable parameters, such as the average tip numberand the PIFB and PMT values of the heterotypic recombinant, are significantly differentfrom those for UB/KM homotypic recombinants, but similar to LB/LM homotypic

59

recombined samples, which further confirms the change in morphogenesis and thereprogramming of the ureteric bud.

Thus, the lung mesenchyme seems to act as an instructive inducer of the ureteric budand to be sufficient to respecify and reverse these marker gene patterns of the ureteric budin recombinants. The response occurs throughout the epithelial bud and appears to beaccompanied by a change in cell differentiation, as suggested by the expression of themarker genes, SP-C, collagen XVIII and Shh, and the epigenesis of the ureteric budtowards the early lung type. The data suggest a novel intra-organ patterning process thatoccurs at the protein and mRNA levels, and the different pieces of morphogeneticinformation during organogenesis may hence be linked to the use of similar genes butthey may be patterned differently. To our knowledge this is the first report where such astriking repatterning process is demonstrated during mouse organogenesis. The assaysystem could serve as an experimental model of genetic and morphogenetic research onhow the differences in form appear during organogenesis.

6.2 Type XVIII collagen may regulate modes of epithelial

morphogenesis in kidney and lung and Wnt2 could be the lung

specific mesenchymal factor

During renal development, the GDNF/c-ret signaling pathway plays a dominant role, andWnt11 is the target signal of GDNF. In this work, we demonstrated that type XVIIIcollagen and Shh were lost from the tips of the ureteric bud during the initiation of thefirst branching of the ureteric bud in developing kidney. This process is associated withupregulation of Wnt11 and Wnt6 (also see Itäranta et al, 2001). The maintenance ofWnt11 at the branching tips of the ureteric bud is associated with the loss of type XVIIIcollagen. Hence, their expressions appear to be mutually exclusive. This may be due tothe fact that the Fz domain containing type XVIII collagen antagonizes Wnt signaling andplays an opposite role compared to the epithelial Wnts during the ureteric bud branchingmorphogenesis. The downregulation of type XVIII collagen would help the tip Wnts todiffuse into the adjacent mesenchyme and to contribute to the induction of tubules (Starket al, 1994; Kispert et al, 1998; Vainio et al, 1999). Upregulation of Wnt11 and Wnt6 atthe tips may cause downregulation of Shh, since there is evidence that Wnts canspecifically downregulate Shh and Ptc (Sarkar et al, 2000). We do not know how the lossof type XVIII collagen from the ureteric tips occurs and how it is maintained in the stalk.It is likely that constant remodeling of the ECM is needed at the growing tips of theinvading ureteric bud for branching to take place. MMPs and TIMPs are expressed in themesenchyme and epithelium, respectively, and their balance is regulated by manycytokines and growth factors, e.g. TGFβ and EGF (Johansson, et al, 1997; Miettinen etal, 1999; for a review, see Lelongt et al, 2001). MT-1 MMP and MMP2 expresseddifferentially in the undifferentiated mesenchyme and the epithelial structures,respectively, suggests that these proteases play central roles in the development of thisstructurally complex organ (Tanney et al, 1998). MMP2 and MMP9 are expressed in thekidney mesenchyme adjacent to the branching ureteric bud at E11.5, has been implicated

60

in Wnt signaling (Roth et al, 2000). Therefore, MMPs are candidates to be involved inthe degradation process. Since type XVIII collagen is also lost at the mRNA level,transcription level regulation may also play a role.

During lung organogenesis, Shh, FGFs and EGF-R binding ligands play importantroles (Pepipecelli et al, 1998; Min et al, 1998; Sekine et al, 1999; Miettinen et al, 1997).In this work, we also demonstrated that type XVIII collagen is functional during the lungbranching morphogenesis. Loss of type XVIII collagen from the branching tips byantibody blocking causes a dramatic reduction of Wnt2 gene expression. But FGF10 andShh do not show a similar decrease. It has been reported that a Shh mutant causes Wnt2downregulation but not Bmp4 and Wnt7b. This indicates that type XVIII collagen maycooperate with Shh in the epithelium and interact with mesenchymal Wnts, such as Wnt2and Wnt2b, during lung organogenesis. Type XVIII collagen, which locates at the tip ofthe lung bud, could bind Wnts to the bronchial epithelial tips and thereby concentratethem. This could promote the characteristic morphogenesis of the epithelial type. Thishypothesis is based on the fact that type XVIII collagen may function as a proteoglycan(Halfter et al, 1998) and PG are necessary for Wnt signaling (Lin & Perrimon, 1999). Thehigh level expression of Wnt2/2b in the lung mesenchyme is very close to the tips wheretype XVIII collagen is expressed and Wnt7b is expressed in the epithelium would suggestinteractions between type XVIII collagen and Wnts and their roles in lung branchingmorphogenesis. Together with some other growth factors, such as FGFs and EGF-Rbinding ligands, Wnt2/2b may transmit inductive signals to the lung epithelium toactivate the cell cycle and in that way to promote epithelial growth and branching. Themesenchymal inductive signals mediated by Shh or Bmp4 in the epithelium may inducetype XVIII collagen at the branching tips and repress its expression in the stalk. At thesame time, feedback inhibitors, e.g. TGFβ1, Shh, Bmp4 and Sprouty, may send signals tothe mesenchymal cells to regulate the secretion of mesenchymal growth factors (for areview, see Kaplan, 2000). The notion that Shh secreted by the distal tip epithelium mayregulate the expression of Bmp4 and/or Wnt2 in the adjacent mesenchyme is alsosupported by a previous report (Bellusci et al, 1996). The localization of type XVIIIcollagen expression in the branching tips and their loss from the stalk may be important.This may ensure that the important signals, such as Wnts, are located at the right site andexert their function.

61

Fig. 7. Model for the signaling molecule cascades involved in the experimental repatterningprocess of ureteric bud branching morphogenesis and the respecification of type XVIIIcollagen. Organogenesis is regulated by sequential and reciprocal morphogenetic signalingmolecules. During kidney development, the GDNF/c-ret signaling pathway maintains Wnt11at the branching tips of the ureteric bud. The initiation of Wnt11 at the branching tip isassociated with the loss of type XVIII collagen. PG and MMPs may be involved in thisprocess. The signaling loop is necessary for normal ureteric bud branching in kidneydevelopment and also for the ureteric bud to become competent to be reprogrammed by lungmesenchyme. In the lung, the high level expression of Wnt2 and Wnt2b suggests their roles inthe lung branching morphogenesis. Together with FGFs, they may transmit inductive signalsto the lung epithelium and activate the cell cycle. These inductive signals mediated by Shhand Bmp4 induce type XVIII collagen to the bronchial branching tips, and the Shh and Bmp4may send inhibitory feedback to regulate the release of the mesenchymal growth factors aswell. The induction of type XVIII collagen to the branching tips in the developing lung canfurther attract more Wnts and initiate the next round of reciprocal signals for branching. InUB/LM heterotypic tissue recombinants, a specific mesenchymal factor, such as Wnt2,functions independently or together with other growth factors, interacts with an epithelialfactor, e.g. Shh, and repatterns the branching morphogenesis of the ureteric bud towardsearly lung type branching. Mediated by Shh, type XVIII collagen is upregulated, which helpsto attract more mesenchymal Wnts close to the branching ureteric tips. The more Wnts, e.g.Wnt-2, are recruited to the cascade, the better it is for the ureteric bud to be repatternedtoward early lung type branching. Therefore, type XVIII collagen may play a dual functionduring organogenesis. In the kidney, it functions via frizzled domain antagonized Wnts and isinvolved in the upregulation of Wnt11 and the release of more Wnts to the mesenchyme. Inthe lung and UB/LM tissue recombinant type XVIII collagen functions as a proteoglycan,attracting more Wnts to the branching center and promoting lung type branching.Abbreviations, see page 2 and the text.

In UB/LM heterotypic tissue recombinants, when the ureteric bud becomes competent,the initial signals that change the quality of the bud epithelium are apparently derivedfrom the lung mesenchyme (Fig. 7 C, D). The distal tip mesenchyme was found to induce

62

tracheal epithelial budding and ectopic type XVIII collagen expression, also indicatingthat the distal tip mesenchyme contains factors sufficient to promote the repatterningprocess. However, such factors as Wnt-2b, FGF2, 7 and 10 did not induce budding ortype XVIII collagen expression, which suggests a role for other factors or that they areneeded as a combination in the process. Culturing lungs in the presence of a type XVIIIcollagen antibody leads to a reduction in epithelial branching. The expression of Wnt-2,which remains expressed in heterotypic recombinants, is dramatically reduced in lungstreated with the antibody, while the expression of Shh and FGF-10 is not so clearlyreduced. Therefore, specific mesenchymal factor(s), e.g. Wnt2, may functionindependently of or in combination with other growth factors, interacting with epithelialfactors, e.g. Shh and type XVIII collagen, and repattern the branching morphogenesis ofthe ureteric bud towards early lung type branching. The mechanisms and significance ofthe regulatory interaction between Wnt-2 and type XVIII collagen remain to bedemonstrated, as no lung phenotype has been detected in the Wnt-2 deficient mice(Monkley et al, 1996).

The facts that the type XVIII collagen antibody reduced the number of epithelial tipsin lung explants and almost completely blocked epithelial morphogenesis in heterotypicrecombinants suggests an important role for type XVIII collagen in the chain ofmolecular events associated with the reprogramming of ureteric bud development. Themarked decrease in Wnt-2 expression in antibody-blocked explants supports a role fortype XVIII collagen in epithelial development as well. The respecification of type XVIIIcollagen at the tip in UB/LM, which was possibly mediated by Shh and Bmp4, could beinvolved in binding Wnts to the branching epithelial tips and concentrating them there inorder to promote typical morphogenesis. Type XVIII collagen may also function as aproteoglycan, and PG are necessary for Wnt signaling. The more Wnts, e.g. Wnt2/2b,there are in the UB/LM recombinants recruited to the molecular cascade, the better theureteric bud may be repatterned toward the early lung type branching. The laminin γ2 andγ2* were expressed in the different compartments of the developing kidney at differentlevels. This may suggest different functions for the two laminin γ2 chain variants inmediating epithelial-mesenchymal interactions during organogenesis. Laminin γ2* maybe an unstable laminin molecule and important for rapid remodeling of the matrix duringearly morphogenesis. But there is also the possibility that laminin γ2 and γ2* have someregulatory functions, such as signaling during differentiation. Thus, laminin γ2 expressionwas tested in UB/LM heterotypic recombinants as well. No significant changes wereobserved. Wnt2b, which is very much homologous to the Wnt2 gene, is expressed in bothlung and kidney and could be a common factor for type XVIII collagen and ureteric budrepatterning in UB/LM. But when we compared its expressions in lung and kidney, wefound that Wnt2b was expressed at a relatively low level and relatively far away from thebranching epithelium in a way very much different from its expression in lung. Itsexpression remained in the UB/LM recombinants. Thus, Wnt2b may play a moresignificant role in lung and may also contribute to the repatterning process together withWnt2.

To summarize, we have demonstrated an intra-organ patterning process that involvestype XVIII collagen, an extracellular matrix protein containing endostatin and frizzleddomains. The repatterning of type XVIII collagen expression was accompanied byectopic expression of lung SP-C and the morphogen Shh, and this was associated with

63

changes in the early epigenesis of the ureteric bud toward the early lung type. Our datasupport a model according to which differential forms may be regulated by patterningcues exchanged during morphogenetic inductive tissue interactions, and these localizedsignaling activities of growth factors involve extracellular matrix molecules, such as typeXVIII collagen.

6.3 Reprogramming of ureteric bud branching with lung

mesenchyme favors early lung type branching morphogenesis

After having demonstrated that lung mesenchyme is able to alter ureteric bud function,we were interested to know if the morphogenesis of the ureteric bud also resembles thelung type of epithelium branching in UB/LM recombination. Several factors have beenproposed as critical for establishing the pattern of branching during organogenesis,including the stereotypy of the branching pattern, the number of generations, branchpoints and stems, the branching angles, the pace of development and fasciculation (for areview, see Al-Awqati and Goldberg 1998). To study further the extent of repatternedbranching morphogenesis of the ureteric bud in lung mesenchyme, we first explored thedifference in epithelial branching between the lung and the kidney. We compared thenumbers of epithelial tips, left and right or anterior and posterior, the lengths of the firstbranches, and the angle of the first epithelial branch. In addition, we developed valuesreferred to as PIFB and PMT to represent characteristics of the branches. Among theseparameters we found that the numbers of epithelial tips, PIFB and PMT, weresignificantly different between kidney and lung as well as in UB/KM and LB/LMhomotypic recombinants. Therefore, these applicable parameters were used to monitorhow lung mesenchyme modifies the early growth and branching of the ureteric bud. Wefound that all the three applicable parameters for recombinants of lung mesenchyme andureteric bud were shifted towards the lung type. The facts demonstrated that lungmesenchyme also re-directs the morphogenesis of renal epithelium. All the three valuesare similar to those obtained for homotypic lung recombinants, and significantly differentfrom the values for homotypic kidney recombinants. The results indicate that thedevelopment of a ureteric bud can really shift from the dichotomous kidney type with abranching pattern to include aspects of early lung-type epithelial morphogenesis.However, the capacity of lung mesenchyme to promote ureteric bud branching andreprogramming or the competence of ureteric bud epithelium to follow the cues fromlung mesenchymal cells maybe limited. In heterotypic UB/LM tissue recombinants, theureteric bud does not branch as extensively with lung mesenchyme as the lung bud does,and especially the ureteric bud expresses genes such as c-ret, Wnt11 and Sox9 inheterotypic recombinants, suggesting that the epithelial bud still retains some of itsureteric bud characteristics. This may contribute to a restriction of the potential ofepithelium to interact with lung mesenchyme during subculture, and the ureteric bud mayalso lack the expression of certain receptors for the signaling molecules expressed bylung mesenchyme. The culture conditions may also limit the repatterning process and abetter branching process. We have tried to culture UB/LM heterotypic recombinants inchick chorioallantoic membrane (CAM) (data not shown). It seems that chorioallantoic

64

membrane was not a good system for UB/LM culture, since it did not promote anyfurther growth and branching of the ureteric bud because neither lung nor UB/LMsurvived in it.

6.4 Involvement of Wnt2b in early kidney development

It has also been reported that Wnt2b is expressed in many different organ and tissues inmouse and human (Katoh et al, 1996; Zakin et al, 1998). Our results are consistent withthese reports, but we especially found Wnt2b to be widely expressed in mouse organs ofepithelial mesenchymal origin, such as kidney, lung, and salivary gland. This expressionpattern suggests that Wnt2b may play certain roles in development and organogenesis.Wnt2b is expressed as early as E11.5 in cells located in the perinephric region of thedeveloping mouse kidney, i.e. in close proximity to the presumptive stroma marked byBF-2 expression. Analyses of the function of the stromal genes BF-2 (Hatini et al, 1996),RARα and RARβ2 (Mendelsohn et al, 1999) have suggested that the kidney stromal cellsmay synthesize factors that influence kidney tubulogenesis. Such stromal signals are alsofound to regulate ureteric development. In BF-2 deficient mice the rate of growth andbranching of the ureter and collecting system was reduced. In the RARα/RARβ2 doublemutant c-ret expression is downregulated in the ureteric bud and the ureteric branching isimpaired (Mendelsohn et al, 1999). More interestingly, forced expression of Ret in Rarα/Rarβ2 double mutant rescues renal development, restoring ureteric bud growth andstromal cell patterning (Batourina et al, 2001). Furthermore, retinoids reduce the activityof the β-catenin-lymphoid enhancer factor / T-cell factor (LEF/TCF) signaling pathway,and β-catenin interacts directly with RAR in a retinoid-dependent manner (Easwaran etal, 1999). Kidney organogenesis is also stimulated by derivatives of vitamin A (Vilar etal, 1996). Hence, there may be a regulatory link between Wnt signaling and retinoic acidin the kidney as well. The pattern of Wnt2b expression and data showing that thisexpression can be regulated by retinoic acid in cultured human cells (Wakeman et al,1998) suggests that Wnt2b may be one of the functional signals during early kidneydevelopment.

6.5 Wnt2b regulates ureteric bud branching and kidney

organogenesis

Wnt2b may be one of the factors contributing to perinephric signaling during earlykidney development. Unlike Wnt4, it does not induce tubulogenesis, but it supportsureteric bud epithelial development. It promotes kidney reconstitution, while Wnt4 doesnot. Wnt2b, however, is expressed some distance away from the ureteric bud epithelium.It may diffuse and move closer to the ureteric bud. Or its signaling may be redundant andmimic the activity of the other Wnts present in mesenchymal cells. Of the Wnts identifiedso far, only Wnt4 has been localized to the kidney mesenchyme. Ureteric branching isinitially unaffected in Wnt4 deficient embryos (Stark et al, 1994), and Wnt4 and Wnt2b

65

signaling cause two distinct phenomena (present data). Thus, Wnt2b and Wnt4 are notredundant. Hence, specific Wnts, such as Wnt4, regulate mesenchymal differentiation andtubule induction (Kispert et al, 1998), while another member of the family, Wnt2b, actsreciprocally to regulate ureteric branching.

Wnt6, Wnt11, and Wnt7b are expressed in the ureteric bud (Kispert et al, 1996, 1998)and could thus be involved in mediating the effect of Wnt2b on the ureteric bud orphenocopy activities of Wnt2b signaling. Wnt6 (Itäranta et al, 2001) and Wnt11 areexpressed at the branching ureteric tip. Cells expressing Wnt6, however, are not capableof promoting branching of the isolated ureter, which excludes its role in the process.Wnt11 gene expression is initiated in reconstituted explants, but its expression is lessintense than in the control samples, which suggests that Wnt11 is not involved inmediating Wnt2b signaling. Wnt7b signaling can induce tubulogenesis in vitro (Kispert etal, 1998), but it is not expressed at the ureteric tips. Hence, Wnt7b is also unlikely tophenocopy the activity of Wnt2b. Lithium reproduces the effect of cells expressingWnt2b, suggesting that Wnt signaling acts directly on the ureteric bud to regulate itsbranching. Interestingly, both epithelial and mesenchymal responses can be induced bylithium (Davies et al, 1995, present data), suggesting that both tubule induction andmaintenance of the bud’s inductive activity and branching are mediated by activation ofthe β-catenin signaling pathway rather than alternative signal transduction pathways.

The facts that Wnt2b induces growth and development of the ureteric bud, but does not induce kidney tubules in the kidney mesenchyme in vitro, whereas Wnt4 induces tubules but not epithelial branching suggest that the signal transduction cascade upstream of GSK3 capable of responding to Wnt4 and Wnt2b may not be shared by the ureteric bud epithelium and kidney mesenchyme. Recently, Wnt2b was shown to be expressed transiently at the initiation of organogenesis in the developing chick limb (Kawakami et al, 2001), and expression also takes place in the developing mouse limb, at least at E10.5 to E11.5 (data not shown). Interestingly, Wnt2b can induce ectopic limb development in the chick, and this induction occurs via sequential activation of FGF10 and FGF8, which can also induce ectopic limb development. FGF10 is expressed in the developing lung and kidney as well (Min et al, 1998; Sekine et al, 1999; Clark, et al, 2001) and can promote branching of the lung epithelium (Park et al, 1998). Hence, Wnt2b function may be conserved and operate in the lung and kidney to regulate epithelial branching via control over FGF signaling. FGF7 and GDNF (Sainio et al, 1997; for a review, see Sariola and Sainio, 1997) are additional mesenchymal signals that have been shown to regulate ureteric development. Receptors for GDNF, c-ret and FGF7 are present at the ureteric epithelial tips, further supporting a role for these in reciprocal tissue interactions (Qiao et al, 1999a). Wnt2b expression in the kidney is most intense at E11.5 and E12.5 and diminishes after that, which suggests that Wnt2b may be functional in promoting the initiation of epithelial branching at the early stages of kidney development, when GDNF has also been shown to be active. Taken together, the expression pattern of the Wnt2b gene in mouse embryonic tissues suggests a role for Wnt2b in epithelial-mesenchymal interactions. Wnt2b signaling is capable of inducing ureteric bud growth and branching and indirectly, via the epithelium, tubulogenesis. Lithium reproduces the effects of Wnt2b signaling, suggesting that signaling to the ureteric bud is direct. Wnt2b may thus act as a mesenchymal signal, mediating reciprocal signaling from the mesenchyme to the epithelium to coordinate early kidney organogenesis. Finally, the data currently available

66

indicate that Wnt signaling in the kidney is specific to either the ureteric bud or the kidney mesenchyme, depending on the Wnt family member that is active.

7 Conclusions and future perspectives

Two alternative laminin γ2 transcripts were demonstrated in the developing mousekidney. The shorter γ2* form was localized in the mesenchyme, whereas the longer γ2form was only present in the ureteric bud, which indicates different functions for the γ2variants in mediating the epithelial and mesenchymal interactions during organogenesis.

The role of type XVIII collagen during organogenesis was investigated. This proteinhas a Fz domain and an endostatin domain. It has characteristic of proteoglycan as well.We demonstrated that type XVIII collagen is expressed uniformly in the early lung andkidney epithelial buds, but it becomes localized to different epithelial regions in theseorgans. This is not the case with type IV collagen. We have also found that type XVIIIcollagen is repatterned in experimental situations where the ureteric bud is co-culturedwith lung mesenchyme. Its expression changes from the kidney-type to the lung-type. Inassociation with the repatterned type XVIII collagen expression, ureteric budmorphogenesis also changed towards that of the early lung type, which ischaracteristically different from the kidney type branching. The repatterned expression oftype XVIII collagen appears to be essential for the respecification of ureter towards lungphenotype, since the type XVIII functional antibody blocks epithelial proliferation. Inepithelium, an important morphogen, Shh, accompanies the repatterning of type XVIIIcollagen. We do not yet know the lung mesenchyme-derived signals that are involved inthe type XVIII collagen repatterning process, but Wnt2 is a candidate factor. In theantibody-blocked lung, Wnt2 expression is dramatically reduced in the mesenchyme. Ourdata also demonstrate that the distal tip lung mesenchyme, which expresses the Wnt2gene, is sufficient to induce ectopic epithelial budding and type XVIII collagenexpression, providing more evidence of Wnt2 signal functioning in this repatterningprocess. The respecified expressions of type XVIII collagen and Shh were accompaniedby ectopic surfactant protein C gene expression at the tips of the ureteric bud. Therefore,type XVIII collagen may be involved in the complicated process of establishing organ-specific structure and have important morphogenetic roles during organogenesis. Thelaminin γ2 expression was also studied in heterotypic recombinants. No significantcorresponding change was found. We have also demonstrated that embryonic lung andkidney posses their own specific branching patterns and lung mesenchyme is able to alterthe morphogenesis of the ureteric bud towards the lung type of epithelium. Our findings

68

provide morphological parameters for monitoring epithelial branching duringorganogenesis. We are currently analyzing the function of type XVIII collagen inmetanephric mesenchyme tubulogenesis. The main focus will be on how type XVIIIcollagen correlates with the other important signaling molecules, such as Wnt, Hedgehogand VEGF, in the process of tubulogenesis. If type XVIII collagen is involved in uretericbranching in a specific way, misexpression of type XVIII collagen may respecify uretericgrowth. If the frizzled domain is functional, ectopic expression of this collagen mayblock the induction. To understand fully the repatterning process of the ureteric bud, weshould concentrate on the functional mesenchymal factor(s) from the lung mesenchyme.Microarray will be one of the tools applied to a further analysis of genes specific for lungmesenchyme. Overexpression of type XVIII collagen in the lung endoderm or theureteric bud by applying the SP-C or Pax2 promoters, respectively, will be accomplishedto further investigate the functions of type XVIII collagen during lung and kidneydevelopment.

The expression and function of Wnt2b during mouse development were alsoinvestigated in our work. The Wnt2b gene is expressed in the mesenchymal cells adjacentto the epithelium in many organs, including the early lung and kidney. As Wnt signalingis sufficient to trigger kidney tubule development, Wnt2b was also a candidate signal fortubulogenesis of metanephric kidney. However, Wnt2b did not induce tubules. Instead, itwas shown to function as a reciprocal signal to promote ureteric growth and branchingand also indirectly, via the epithelium, to promote tubulogenesis. Wnt2b is the first Wntthat has been shown to have such activity. This data also provides more evidence to implythat Wnt secreted factors are not all functionally equivalent. Further cocultureexperiments will be required to determine whether Wnt2b may have synergistic effectsalong with GDNF, neurturin or persephin and in controlling kidney development.

Taken together, our data point out that intraorgan patterning process revealed by theexpression of type XVIII collagen may be regulated by the position of inductive factorsand matrix molecules. Wnt 2b may act as an early mesenchymal signal controllingmorphogenesis of epithelial tissue, and the Wnt pathway may regulate ureteric branchingdirectly. In this work also an experimental model was developed that may be useful tostudy molecular machenism of the tissue specific morphogenesis.

References

Abe N, Muragaki Y, Yoshioka H, Inoue H & Ninomiya Y (1993) Identification of a novel collagenchain represented by extensive interruptions in the triple-helical region. Biochem Biophys ResCommun196: 576-582.

Ackley BD, Crew JR, Elamaa H, Pihlajaniemi T, Kuo CJ, & Kramer JM (2001) TheNC1/Endostatin Domain of Caenorhabditis elegans Type XVIII Collagen Affects CellMigration and Axon Guidance. J Cell Biol 152: 1219-1232.

Airenne T, Haakana H, Sainio K, Kallunki T, Kallunki P, Sariola H & Tryggvason K (1996)Structure of the human laminin gamma 2 chain gene (LAMC2): alternative splicing withdifferent tissue distribution of two transcripts. Genomics 32: 54-64.

Al-Awqati Q & Goldberg MR (1998) Architectural patterns in branching morphogenesis in thekidney. Kidney Int 54: 1832-1842.

Andrews KL, Betsuyaku T, Rogers S, Shipley JM, Senior RM & Miner JH (2000) Gelatinase B(MMP-9) is not essential in the normal kidney and does not influence progression of renaldisease in a mouse model of Alport syndrome. Am J Pathol 157: 303-311.

Arman E, Haffner-Krausz R, Gorivodsky M & Lonai P (1999) Fgfr2 is required for limboutgrowth and lung-branching morphogenesis. Proc Natl Acad Sci USA 96: 11895-11899.

Baeg GH, Lin X, Khare N, Baumgartner S & Perrimon N (2001) Heparan sulfate proteoglycans arecritical for the organization of the extracellular distribution of Wingless. Development 128: 87-94.

Barasch J, Yang J, Ware CB, Taga T, Yoshida K, Erdjument-Bromage H, Tempst P, Parravicini E,Malach S, Aranoff T & Oliver JA (1999). Mesenchymal to epithelial conversion in ratmetanephros is induced by LIF. Cell 99: 377-386.

Behrens J, Jerchow BA, Wurtele M, Grimm J, Asbrand C, Wirtz R, Kuhl M, Wedlich D &Birchmeier W (1998) Functional interaction of an axin homolog, conductin, with beta-catenin,APC, and GSK3beta. Science 280: 596-599.

Bellusci S, Henderson R, Winnier G, Oikawa T & Hogan BL (1996) Evidence from normalexpression and targeted misexpression that bone morphogenetic protein (Bmp-4) plays a role inmouse embryonic lung morphogenesis. Development 122: 1693-1702.

Bellusci S, Grindley J, Emoto H, Itoh N & Hogan BL (1997a) Fibroblast growth factor 10(FGF10) and branching morphogenesis in the embryonic mouse lung. Development 124: 4867-4878.

Bellusci S, Furuta Y, Rush MG, Henderson R, Winnier G & Hogan BL (1997b) Involvement ofSonic hedgehog (Shh) in mouse embryonic lung growth and morphogenesis. Development124:53-63.

Bernfield M, Gotte M, Park PW, Reizes O, Fitzgerald ML, Lincecum J & Zako M (1999)Functions of cell surface heparan sulfate proteoglycans. Annu Rev Biochem 68: 729-777.

70

Bhanot P, Brink M, Samos CH, Hsieh JC, Wang Y, Macke JP, Andrew D, Nathans J & Nusse R(1996) A new member of the frizzled family from Drosophila functions as a Wingless receptor.Nature 382: 225-230.

Bitgood MJ & McMahon AP (1995) Hedgehog and Bmp genes are coexpressed at many diversesites of cell-cell interaction in the mouse embryo. Dev Biol 172: 126-138.

Bobola N, Hirsch E, Albini A, Altruda F, Noonan, D & Ravazzolo R (1995) A single cis-actingelement in a short promoter segment of the gene encoding the interphotoreceptor retinoid-binding protein confers tissue-specific expression. J Biol Chem 270: 1289-1294.

Batourina E, Gim S, Bello N, Shy M, Clagett-Dame M, Srinivas S, Costantini F & Mendelsohn C(2001) Vitamin A controls epithelial/mesenchymal interactions through Ret expression. NatGenet 27: 74-78.

Brandenberger R, Schmidt A, Linton J, Wang D, Backus C, Denda S, Muller U & Reichardt LF(2001) Identification and characterization of a novel extracellular matrix protein nephronectinthat is associated with integrin alpha8beta1 in the embryonic kidney. J Cell Biol 154: 447-458.

Breier G, Albrecht U, Sterrer S & Risau W (1992) Expression of vascular endothelial growth factorduring embryonic angiogenesis and endothelial cell differentiation. Development 114: 521-532.

Breier G, Clauss M & Risau W (1995) Coordinate expression of vascular endothelial growth factorreceptor-1 (flt-1) and its ligand suggests a paracrine regulation of murine vascular development.Dev Dyn 204: 228-239.

Bullock SL (1998) Renal agenesis in mice homozygous for a gene trap mutation in the geneencoding heparan sulfate 2-sulfotransferase 1. Genes Dev 12: 1894-1906.

Cadigan KM & Nusse R (1997) Wnt signaling: a common theme in animal development. GenesDev 11: 3286-3305.

Canete-Soler R, Gui YH, Linask KK & Muschel RJ (1995) Developmental expression of MMP-9(gelatinase B) mRNA in mouse embryos. Dev Dyn 204: 30-40.

Cao Y (2001) Endogenous angiogenesis inhibitors and their therapeutic implications. Int J BiochemCell Biol 33: 357-369.

Carroll TJ & McMahon AP (2000) Secreted molecules in metanephric induction. J Am SocNephrol 16: S116-119.

Carroll TJ, Ishibashi M, Parr B & McMahon AP (2001) The role of Wnts in the development of thekidney collecting ducts (abstract). 60th Annual Meeting of Society for Developmental Biology,Seattle, USA, 205-206.

Celli G, LaRochelle WJ, Mackem S, Sharp R & Merlino G (1998) Soluble dominant-negativereceptor uncovers essential roles for fibroblast growth factors in multiorgan induction andpatterning. EMBO J 17: 1642-1655.

Clark AM, Garland KK & Russell LD (2000) Desert hedgehog (Dhh) gene is required in the mousetestis for formation of adult-type Leydig cells and normal development of peritubular cells andseminiferous tubules. Biol Reprod 63: 1825-1838.

Clark CC, Cohen I, Eichstetter I, Cannizzaro LA, McPherson JD, Wasmuth JJ & Iozzo RV (1993)Molecular cloning of the human proto-oncogene Wnt-5A and mapping of the gene (WNT5A) tochromosome 3p14-p21. Genomics 18: 249-260.

Clark JC, Tichelaar JW, Wert SE, Itoh N, Perl AK, Stahlman MT & Whitsett JA (2001) FGF-10disrupts lung morphogenesis and causes pulmonary adenomas in vivo. Am J Physiol Lung CellMol Physiol 280: L705–715.

Colvin JS, White AC, Pratt SJ & Ornitz DM (2001a) Lung hypoplasia and neonatal death in Fgf9-null mice identify this gene as an essential regulator of lung mesenchyme. Development 128:2095-2106.

Colvin JS, Green RP, Schmahl J, Capel B & Ornitz DM (2001b) Male-to-female sex reversal inmice lacking fibroblast growth factor 9. Cell 104: 875-889.

Costa RH, Kalinichenko VV & Lim L (2001) Transcription factors in mouse lung development andfunction. Am J Physiol Lung Cell Mol Physiol 280: L823-838.

Cunha GR, Fujii H, Neubauer BL, Shannon JM, Sawyer L & Reese BA (1983) Epithelial-mesenchymal interactions in prostatic development. I. morphological observations of prostaticinduction by urogenital sinus mesenchyme in epithelium of the adult rodent urinary bladder. JCell Biol 96: 1662-1670.

71

Davies CL, Melder RJ, Munn LL, Mouta-Carreira C, Jain RK & Boucher Y (2001) Decorininhibits endothelial migration and tube-like structure formation: role of thrombospondin-1.Microvasc Res 62: 26-42.

Davies JA, Lyon M, Gallagher J & Garrod DR (1995a) Sulphated proteoglycan is required forcollecting duct growth and branching but not nephron formation during kidney development.Development 121: 1507–1517.

Davies JA & Garrod DR (1995b) Induction of early stages of kidney tubule differentiation bylithium ions. Dev Biol 167: 50-60.

Davies JA & Davey MG (1999) Collecting duct morphogenesis. Pediatr Nephrol 13: 535-541.Davies J (2001) Intracellular and extracellular regulation of ureteric bud morphogenesis. J Anat

198(Pt 3): 257-264.De Cat B & David G (2001) Developmental roles of the glypicans. Semin Cell Dev Biol 12: 117-

225.Derynck R & Feng XH (1997) TGF-beta receptor signaling. Biochim Biophys Acta 1333: F105-

150.Deterding RR & Shannon JM (1995) Proliferation and differentiation of fetal rat pulmonary

epithelium in the absence of mesenchyme. J Clin Invest 95: 2963-2972.Dhanabal M, Volk R, Ramchandran R, Simons M & Sukhatme VP (1999a) Cloning, expression,

and in vitro activity of human endostatin. Biochem Biophys Res Commun 258: 345-352.Dhanabal M, Ramchandran R, Waterman MJ, Lu H, Knebelmann B, Segal M & Sukhatme VP

(1999b) Endostatin induces endothelial cell apoptosis. J Biol Chem 274: 11721-11726.Dickson MC, Martin JS, Cousins FM, Kulkarni AB, Karlsson S & Akhurst RJ (1995) Defective

haematopoiesis and vasculogenesis in transforming growth factor-beta 1 knock out mice.Development 121: 1845-1854.

Donovan MJ, Natoli TA, Sainio K, Amstutz A, Jaenisch R, Sariola H & Kreidberg JA (1999)Initial differentiation of the metanephric mesenchyme is independent of WT1 and the uretericbud. Dev Genet 24: 252-262.

Dudley AT, Lyons KM & Robertson EJ (1995) A requirement for bone morphogenetic protein-7during development of the mammalian kidney and eye. Genes Dev 9: 2795–2807.

Dudley AT, Godin RE, Robertson EJ. Interaction between FGF and BMP signaling pathwaysregulates development of metanephric mesenchyme. Genes Dev 1999 13: 1601-1613.

Dunn NR, Winnier GE, Hargett LK, Schrick JJ, Fogo AB & Hogan BL (1997) Haploinsufficientphenotypes in Bmp4 heterozygous null mice and modification by mutations in Gli3 and Alx4.Dev Biol 188: 235-247.

Durbec P, Marcos-Gutierrez CV, Kilkenny C, Grigoriou M, Wartiowaara K, Suvanto P, Smith D,Ponder B, Costantini F & Saarma M (1996) GDNF signalling through the Ret receptor tyrosinekinase. Nature 381: 789-793.

Durbeej M, Fecker L, Hjalt T, Zhang HY, Salmivirta K, Klein G, Timpl R, Sorokin L, Ebendal T,Ekblom P & Ekblom M (1996) Expression of laminin alpha 1, alpha 5 and beta 2 chains duringembryogenesis of the kidney and vasculature. Matrix Biol 15: 397-413.

Durbeej M & Ekblom P (1997) Dystroglycan and laminins: glycoconjugates involved in branchingepithelial morphogenesis. Exp Lung Res 23: 109-118.

Dyer MA, Farrington SM, Mohn D, Munday JR & Baron MH (2001) Indian hedgehog activateshematopoiesis and vasculogenesis and can respecify prospective neurectodermal cell fate in themouse embryo. Development 128: 1717-1730.

Easwaran V, Pishvaian M, Salimuddin & Byers S (1999) Cross-regulation of beta-catenin-LEF/TCF and retinoid signaling pathways. Curr Biol 9: 1415-1418.

Ekblom P, Alitalo K, Vaheri A, Timpl R & Saxen L (1980a) Induction of a basement membraneglycoprotein in embryonic kidney: possible role of laminin in morphogenesis. Proc Natl AcadSci USA 77: 485-489.

Ekblom P, Miettinen A & Saxen L (1980b) Induction of brush border antigens of the proximaltubule in the developing kidney. Dev Biol 74: 263–274.

Ekblom P (1981) Formation of basement membranes in the embryonic kidney: animmunohistological study. J Cell Biol 91: 1-10.

72

Ekblom P (1992) Renal development. In: Seldin DW, Giebisch G (eds) The kidney: physiology andpathophysiology. Raven, New York, pp 475–501.

Ekblom P, Ekblom M, Fecker L, Klein G, Zhang HY, Kadoya Y, Chu ML, Mayer U & Timpl R(1994) Role of mesenchymal nidogen for epithelial morphogenesis in vitro. Development 120:2003-2014.

Ekblom P (1996) Receptors for laminins during epithelial morphogenesis. Curr Opin Cell Biol 8:700-706.

Epstein EH Jr (1992) Molecular genetics of epidermolysis bullosa. Science 256: 799-804.Escary JL, Perreau J, Dumenil D, Ezine S & Brulet P (1993) Leukaemia inhibitory factor is

necessary for maintenance of haematopoietic stem cells and thymocyte stimulation. Nature 363:361-364.

Ferkey DM & Kimelman D (2000) GSK-3: new thoughts on an old enzyme. Dev Biol 225: 471-479.

Finch PW, Cunha GR, Rubin JS, Wong J & Ron D (1995) Pattern of keratinocyte growth factorand keratinocyte growth factor receptor expression during mouse fetal development suggests arole in mediating morphogenetic mesenchymal-epithelial interactions. Dev Dyn 203: 223-240.

Fine JD, Bauer EA, Briggaman RA, Carter DM, Eady RA, Esterly NB, Holbrook KA, Hurwitz S,Johnson L, Lin A, Pearson R & Sybert VP (1991) Revised clinical and laboratory criteria forsubtypes of inherited epidermolysis bullosa. A consensus report by the Subcommittee onDiagnosis and Classification of the National Epidermolysis Bullosa Registry. J Am AcadDermatol 24: 119-135.

Flores-Delgado G, Bringas P & Warburton D (1998) Laminin 2 attachment selects myofibroblastsfrom fetal mouse lung. Am J Physiol 275(3 Pt 1): L622-630.

Fukuda Y, Ishizaki M, Okada Y, Seiki M & Yamanaka N (2000) Matrix metalloproteinases andtissue inhibitor of metalloproteinase-2 in fetal rabbit lung. Am J Physiol Lung Cell Mol Physiol279: L555-561.

Garcion E, Faissner A & ffrench-Constant C (2001) Knockout mice reveal a contribution of theextracellular matrix molecule tenascin-C to neural precursor proliferation and migration.Development 128: 2485-2496.

Gilbert SF. 2000. Developmental Biology. 6th Edition. Sinauer Associates, Inc. pp 4-149.Glinka A, Wu W, Delius H, Monaghan AP, Blumenstock C & Niehrs C (1998) Dickkopf-1 is a

member of a new family of secreted proteins and functions in head induction. Nature391(6665): 357-362.

Godin RE, Takaesu NT, Robertson EJ & Dudley AT (1998) Regulation of BMP7 expression duringkidney development. Development 125: 3473-3482.

Goodrich LV, Milenkovic L, Higgins KM & Scott MP (1997) Altered neural cell fates andmedulloblastoma in mouse patched mutants. Science 277: 1109-1113.

Grisaru S, Cano-Gauci D, Tee J, Filmus J & Rosenblum ND (2001) Glypican-3 modulates BMP-and FGF-mediated effects during renal branching morphogenesis. Dev Biol 231: 31-46.

Gritli-Linde A, Lewis P, McMahon AP & Linde A (2001) The whereabouts of a morphogen: directevidence for short- and graded long-range activity of hedgehog signaling peptides. Dev Biol236: 364-386.

Grobstein, G (1955) Inductive interaction in the development of the mouse metanephros. J ExpZool 130: 319-340.

Grobstein, C (1967) Mechanisms of organogenetic tissue interaction. Natl Cancer Inst Monogr 26:279-299.

Grove EA, Tole S, Limon J, Yip L & Ragsdale CW (1998) The hem of the embryonic cerebralcortex is defined by the expression of multiple Wnt genes and is compromised in Gli3-deficientmice. Development 125: 2315-2325.

Gurdon JB (1992) The generation of diversity and pattern in animal development. Cell 68: 185-99.Hadjantonakis AK, Gertsenstein M, Ikawa M, Okabe M & Nagy A (1998) Generating green

fluorescent mice by germline transmission of green fluorescent ES cells. Mech Dev 76: 79-90.Halfter W, Dong S, Schurer B & Cole GJ (1998) Collagen XVIII is a basement membrane heparan

sulfate proteoglycan. J Biol Chem 273: 25404-25412.

73

Hall AC, Lucas FR & Salinas PC (2000) Axonal remodeling and synaptic differentiation in thecerebellum is regulated by WNT-7a signaling. Cell 100: 525-535.

Hammerschmidt M, Brook A & McMahon AP (1997) The world according to hedgehog. TrendsGenet 13: 14-21.

Hatini V, Huh SO, Herzlinger D, Soares VC & Lai E (1996) Essential role of stromal mesenchymein kidney morphogenesis revealed by targeted disruption of Winged Helix transcription factorBF-2. Genes Dev 10: 1467-1478.

Hedgepeth CM, Conrad LJ, Zhang J, Huang HC, Lee VM & Klein PS (1997) Activation of the Wntsignaling pathway: a molecular mechanism for lithium action. Dev Biol 185: 82-91.

Hedgepeth CM, Deardorff MA, Rankin K & Klein PS (1999) Regulation of glycogen synthasekinase 3beta and downstream Wnt signaling by axin. Mol Cell Biol 19: 7147-7157.

Heine UI, Munoz EF, Flanders KC, Roberts AB & Sporn MB (1990) Colocalization of TGF-beta 1and collagen I and III, fibronectin and glycosaminoglycans during lung branchingmorphogenesis. Development 109: 29-36.

Herzlinger D, Qiao J, Cohen D, Ramakrishna N & Brown AM (1994) Induction of kidneyepithelial morphogenesis by cells expressing Wnt-1. Dev Biol 166: 815-818.

Hilfer SR (1996) Morphogenesis of the lung: control of embryonic and fetal branching. Annu RevPhysiol 58: 93-113.

Hlsken J & Behrens J (2000) The Wnt signalling pathway. J Cell Sci 113 (Pt 20): 3545.Holm K, Risteli L & Sariola H (1988) Differential expression of the laminin A and B chains in

chimeric kidneys. Cell Differ 24: 223-227.Hogan, BL (1996) Bone morphogenetic proteins: multifunctional regulators of vertebrate

development. Genes Dev 10: 1580-1594.Hogan BL (1999) Morphogenesis. Cell 96: 225-233.Hsieh JC, Kodjabachian L, Rebbert ML, Rattner A, Smallwood PM, Samos CH, Nusse R, Dawid

IB & Nathans J (1999) A new secreted protein that binds to Wnt proteins and inhibits theiractivities. Nature 398: 431-436.

Hsu SM, Raine L & Fanger H (1981) Use of avidin-biotin-peroxidase complex (ABC inimmunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP)procedures. J Histochem Cytochem 29: 577-580.

Ishikawa T, Tamai Y, Zorn AM, Yoshida H, Seldin MF, Nishikawa S & Taketo MM (2001) MouseWnt receptor gene Fzd5 is essential for yolk sac and placental angiogenesis. Development 128:25-33.

Itäranta P, Lin Y, Peräsaari J & Vainio S (2001) Wnt-6 is expressed in the ureter bud and inducestubule development in vitro. Genesis In Press.

Jaakkola P & Jalkanen M (1999) Transcriptional regulation of Syndecan-1 expression by growthfactors. Prog Nucleic Acid Res Mol Biol 63: 109-138.

Jaskoll, TF, Don-Wheeler G, Johnson R & Slavkin HC (1988) Embryonic mouse lungmorphogenesis and type II cytodifferentiation in serumless, chemically defined medium usingprolonged in vitro cultures. Cell Differ 24: 105-117.

Jing S, Wen D, Yu Y, Holst PL, Luo Y, Fang M, Tamir R, Antonio L, Hu Z, Cupples R, Louis JC,Hu S, Altrock BW & Fox GM (1996) GDNF-induced activation of the ret protein tyrosinekinase is mediated by GDNFR-alpha, a novel receptor for GDNF. Cell 85: 1113-1124.

Johansson N, Westermarck J, Leppa S, Hakkinen L, Koivisto L, Lopez-Otin C, Peltonen J, Heino J& Kahari VM (1997) Collagenase 3 (matrix metalloproteinase 13) gene expression by HaCaTkeratinocytes is enhanced by tumor necrosis factor alpha and transforming growth factor beta.Cell Growth Differ 8: 243-250.

Johansson N, Ahonen M & Kahari VM (2000) Matrix metalloproteinases in tumor invasion. CellMol Life Sci 57: 5-15.

Kaartinen V, Voncken JW, Shuler C, Warburton D, Bu D, Heisterkamp N & Groffen J (1995)Abnormal lung development and cleft palate in mice lacking TGF-beta 3 indicates defects ofepithelial-mesenchymal interaction. Nat Genet 11: 415-421.

Kanwar YS, Ota K, Yang Q, Wada J, Kashihara N, Tian Y & Wallner EI (1999) Role ofmembrane-type matrix metalloproteinase 1 (MT-1-MMP), MMP-2, and its inhibitor innephrogenesis. Am J Physiol 277(6 Pt 2): F934-947.

74

Karavanova ID, Dove LF, Resau JH & Perantoni AO (1996) Conditioned medium from a ratureteric bud cell line in combination with bFGF induces complete differentiation of isolatedmetanephric mesenchyme. Development 122: 4159-4167.

Kaplan F (2000) Molecular determinants of fetal lung organogenesis. Mol Genet Metab 71: 321-41.Karumanchi SA, Jha V, Ramchandran R, Karihaloo A, Tsiokas L, Chan B, Dhanabal M, Hanai J,

Venkataraman G, Shriver Z, Keiser N, Kalluri R, Zeng H, Mukhopadhyay D, Chen RL, LanderAD, Hagihara K, Yamaguchi Y, Sasisekharan R, Cantley L & Sukhatme VP (2001) Cell surfaceglypicans are low-affinity endostatin receptors. Mol Cell 7: 811-822.

Katoh M, Hirai M, Sugimura T & Terada M (1996) Cloning, expression and chromosomallocalization of Wnt-13, a novel member of the Wnt gene family. Oncogene 13: 873-876.

Kawakami Y, Capdevila J, Buscher D, Itoh T, Rodriguez Esteban C & Izpisua Belmonte JC (2001)WNT signals control FGF-dependent limb initiation and AER induction in the chick embryo.Cell 104: 891-900.

Kispert A, Vainio S, Shen L, Rowitch DH & McMahon AP (1996) Proteoglycans are required formaintenance of Wnt-11 expression in the ureter tips. Development 122: 3627–3637.

Kispert A, Vainio S, & McMahon AP (1998) Wnt-4 is a mesenchymal signal for epithelialtransformation of metanephric mesenchyme in the developing kidney. Development 125: 4225-4234.

Klein G, Langegger M, Timpl R & Ekblom P (1988) Role of laminin A chain in the developmentof epithelial cell polarity. Cell 55: 331-341.

Klein PS & Melton DA (1996) A molecular mechanism for the effect of lithium on development.Proc Natl Acad Sci USA 93: 8455-8459.

Krasnow MA (1997) Genes that control organ form: lessons from bone and branchingmorphogenesis. Cold Spring Harb Symp Quant Biol 62:235-240.

Kratochwil K, Dziadek M, Lohler J, Harbers K & Jaenisch R (1986) Normal epithelial branchingmorphogenesis in the absence of collagen I. Dev Biol 117: 596-606.

Kreidberg JA, Sariola H, Loring JM, Maeda M, Pelletier J, Housman D & Jaenisch R (1993) WT-1is required for early kidney development. Cell 74: 679-691.

Kreidberg JA, Donovan MJ, Goldstein SL, Rennke H, Shepherd K, Jones RC & Jaenisch R (1996)Alpha 3 beta 1 integrin has a crucial role in kidney and lung organogenesis. Development 122:3537–3547.

Kühl M, Sheldahl LC, Park M, Miller JR & Moon RT (2000) The Wnt/Ca2+ pathway: a newvertebrate Wnt signaling pathway takes shape. Trends Genet 16: 279-283.

Kulkarni AB, Huh CG, Becker D, Geiser A, Lyght M, Flanders KC, Roberts AB, Sporn MB, WardJM & Karlsson S (1993) Transforming growth factor beta 1 null mutation in mice causesexcessive inflammatory response and early death. Proc Natl Acad Sci USA 90: 770-774.

Kuure S, Vuolteenaho R & Vainio S (2000) Kidney morphogenesis: cellular and molecularregulation. Mech Dev 92: 31-45.

Ladher RK, Church VL, Allen S, Robson L, Abdelfattah A, Brown NA, Hattersley G, Rosen V,Luyten FP, Dale L & Francis-West PH (2000) Cloning and expression of the Wnt antagonistsSfrp-2 and Frzb during chick development. Dev Biol 218: 183-198.

Larsson J, Goumans MJ, Sjostrand LJ, van Rooijen MA, Ward D, Leveen P, Xu X, ten Dijke P,Mummery CL & Karlsson S (2001) Abnormal angiogenesis but intact hematopoietic potentialin TGF-beta type I receptor-deficient mice. EMBO J 20: 1663-1673.

Lee SM, Tole S, Grove E & McMahon AP (2000) A local Wnt-3a signal is required fordevelopment of the mammalian hippocampus. Development 127: 457-467.

Lehtonen E & Saxen L (1986) Cytodifferentiation vs. organogenesis in kidney development. ProgClin Biol Res 217A: 411-418.

Leimeister C, Bach A & Gessler M (1998) Developmental expression patterns of mouse sFRPgenes encoding members of the secreted frizzled related protein family. Mech Dev 75: 29-42.

Lelongt B, Trugnan G, Murphy G & Ronco PM (1997) Matrix metalloproteinases MMP2 andMMP9 are produced in early stages of kidney morphogenesis but only MMP9 is required forrenal organogenesis in vitro. J Cell Biol 136: 1363-1373.

75

Lelongt B, Legallicier B, Piedagnel R & Ronco PM (2001) Do matrix metalloproteinases MMP-2and MMP-9 (gelatinases) play a role in renal development, physiology and glomerular diseases?Curr Opin Nephrol Hypertens 10: 7-12.

Lescher B, Haenig B & Kispert A (1998) sFRP-2 is a target of the Wnt-4 signaling pathway in thedeveloping metanephric kidney. Dev Dyn 213: 440-451.

Letterio JJ, Geiser AG, Kulkarni AB, Roche NS, Sporn MB & Roberts AB (1994) Maternal rescueof transforming growth factor-beta 1 null mice. Science 264: 1936-1938.

Leyns L, Bouwmeester T, Kim SH, Piccolo S & De Robertis EM (1997) Frzb-1 is a secretedantagonist of Wnt signaling expressed in the Spemann organizer. Cell 88: 747-756.

Lijam N, Paylor R, McDonald MP, Crawley JN, Deng CX, Herrup K, Stevens KE, Maccaferri G,McBain CJ, Sussman DJ & Wynshaw-Boris A (1997) Social interaction and sensorimotorgating abnormalities in mice lacking Dvl1. Cell 90: 895-905.

Lin X & Perrimon N (1999) Dally cooperates with Drosophila Frizzled 2 to transduce Winglesssignalling. Nature 400: 281-284.

Liu P, Wakamiya M, Shea MJ, Albrecht U, Behringer RR & Bradley A (1999) Requirement forWnt3 in vertebrate axis formation. Nat Genet 22: 361-365.

Luo G, Hofmann C, Bronckers AL, Sohocki M, Bradley A & Karsenty G (1995) BMP-7 is aninducer of nephrogenesis, and is also required for eye development and skeletal patterning.Genes Dev 9: 2808-2820

Mailleux AA, Tefft D, Ndiaye D, Itoh N, Thiery JP, Warburton D & Bellusci S (2001) Evidencethat SPROUTY2 functions as an inhibitor of mouse embryonic lung growth and morphogenesis.Mech Dev 102: 81-94.

Mao B, Wu W, Li Y, Hoppe D, Stannek P, Glinka A & Niehrs C (2001) LDL- receptor-relatedprotein 6 is a receptor for Dickkopf proteins. Nature 411: 321-325.

Mao J, Wang J, Liu B, Pan W, Farr GH, Flynn C, Yuan H, Takada S, Kimelman D, Li L & Wu D(2001) Low-density lipoprotein receptor-related protein-5 binds to axin and regulates thecanonical wnt signaling pathway. Mol Cell 7: 801-809.

Marti E, Takada R, Bumcrot DA, Sasaki H & McMahon AP (1995) Distribution of Sonic hedgehogpeptides in the developing chick and mouse embryo. Development 121: 2537-2547.

Martin GR (1998) The roles of FGFs in the early development of vertebrate limbs. Genes Dev 12:1571-1586.

Marvulli D, Volpin D & Bressan GM (1996) Spatial and temporal changes of type VI collagenexpression during mouse development. Dev Dyn 206: 447-454.

McMahon JA & McMahon AP (1989) Nucleotide sequence, chromosomal localization anddevelopmental expression of the mouse int-1-related gene. Development 107: 643-650.

McMahon AP& Bradley A (1990) The Wnt-1 (int-1) proto-oncogene is required for developmentof a large region of the mouse brain. Cell 62: 1073-1085.

Mendelsohn C, Batourina E, Fung S, Gilbert T & Dodd J (1999) Stromal cells mediate retinoid-dependent functions essential for renal development. Development 126: 1139–1148.

Mercer RR & Crapo JD (1990) Spatial distribution of collagen and elastin fibers in the lungs. JAppl Physiol 69: 756-765.

Metzger RJ & Krasnow MA (1999) Genetic control of branching morphogenesis. Science 284:1635-1639.

Miettinen PJ, Ebner R, Lopez AR & Derynck R (1994) TGF-beta induced transdifferentiation ofmammary epithelial cells to mesenchymal cells: involvement of type I receptors. J Cell Biol127: 2021-2036.

Miettinen PJ, Berger JE, Meneses J, Phung Y, Pedersen RA, Werb Z & Derynck R (1995)Epithelial immaturity and multiorgan failure in mice lacking epidermal growth factor receptor.Nature 376: 337-341.

Miettinen PJ (1997) Epidermal growth factor receptor in mice and men--any applications to clinicalpractice? Ann Med 29: 31-34.

Miettinen PJ, Warburton D, Bu D, Zhao JS, Berger JE, Minoo P, Koivisto T, Allen L, Dobbs L,Werb Z & Derynck R (1997) Impaired lung branching morphogenesis in the absence offunctional EGF receptor. Dev Biol. 186: 224-236.

76

Miettinen PJ, Chin JR, Shum L, Slavkin HC, Shuler CF, Derynck R & Werb Z (1999) Epidermalgrowth factor receptor function is necessary for normal craniofacial development and palateclosure. Nat Genet 22: 69-73.

Miyazaki Y, Oshima K, Fogo A, Hogan BL & Ichikawa I (2000) Bone morphogenetic protein 4regulates the budding site and elongation of the mouse ureter. J Clin Invest 105: 863-873.

Millar SE, Willert K, Salinas PC, Roelink H, Nusse R, Sussman DJ & Barsh GS (1999) WNTsignaling in the control of hair growth and structure. Dev Biol 207: 133-149.

Miller C & Sassoon DA (1998) Wnt-7a maintains appropriate uterine patterning during thedevelopment of the mouse female reproductive tract. Development 125: 3201-3211.

Miller JR, Hocking AM, Brown JD & Moon RT (1999) Mechanism and function of signaltransduction by the Wnt/beta-catenin and Wnt/Ca2+ pathways. Oncogene 18: 7860-7872.

Miller JR & Moon RT (1996) Signal transduction through beta-catenin and specification of cell fateduring embryogenesis. Genes Dev 10: 2527-2539.

Miner JH (2001) Mystery solved: discovery of a novel integrin ligand in the developing kidney. JCell Biol 154: 257-260.

Min H, Danilenko DM, Scully SA, Bolon B, Ring BD, Tarpley JE, DeRose M & Simonet WS(1998) Fgf-10 is required for both limb and lung development and exhibits striking functionalsimilarity to Drosophila branchless. Genes Dev 12: 3156-3161.

Minoo P (2000) Transcriptional regulation of lung development: emergence of specificity. RespirRes 1: 109-115.

Monkley SJ, Delaney SJ, Pennisi DJ, Christiansen JH & Wainwright BJ (1996) Targeted disruptionof the Wnt2 gene results in placentation defects. Development 122: 3343-3353.

Miyazaki Y, Oshima K, Fogo A, Hogan BL & Ichikawa I (2000) Bone morphogenetic protein 4regulates the budding site and elongation of the mouse ureter. J Clin Invest 105: 863-873.

Moon RT, Brown JD & Torres M (1997) WNTs modulate cell fate and behavior during vertebratedevelopment. Trends Genet 13: 157-162.

Moore MW, Klein RD, Farinas I, Sauer H, Armanini M, Phillips H, Reichardt LF, Ryan AM,Carver-Moore K & Rosenthal A (1996) Renal and neuronal abnormalities in mice lackingGDNF. Nature 382: 76-79.

Motoyama J, Liu J, Mo R, Ding Q, Post M & Hui CC (1998) Essential function of Gli2 and Gli3 inthe formation of lung, trachea and oesophagus. Nat Genet 20: 54-57.

Muragaki Y, Timmons S, Griffith CM, Oh SP, Fadel B, Quertermous T & Olsen BR (1995) MouseCol18a1 is expressed in a tissue-specific manner as three alternative variants and is localized inbasement membrane zones. Proc Natl Acad Sci USA 92: 8763-8767.

Mummery C (2001) Transforming growth factor beta and mouse development. Microsc Res Tech52: 374-386.

Myllyharju J & Kivirikko KI (2001) Collagens and collagen-related diseases. Ann Med 33: 7-21.Müller U, Wang D, Denda S, Meneses J, Pedersen RA & Reichardt LF (1997) Integrin alpha8beta1

is critically important for epithelial-mesenchymal interactions during kidney morphogenesis.Cell 88: 603–613.

Nguyen HT, Thomson AA, Kogan BA, Baskin LS & Cunha GR (1999) Expression of the Wntgene family during late nephrogenesis and complete ureteral obstruction. Lab Invest 79: 647-658.

Niehrs C, Kazanskaya O, Wu W & Glinka A (2001) Dickkopf1 and the Spemann-Mangold headorganizer. Int J Dev Biol 45(1 Spec No): 237-240.

Nonchev S, Maconochie M, Vesque C, Aparicio S, Ariza-McNaughton L, Manzanares M,Maruthainar K, Kuroiwa A, Brenner S, Charnay P & Krumlauf R (1996) The conserved role ofKrox-20 in directing Hox gene expression during vertebrate hindbrain segmentation. Proc NatlAcad Sci USA 93: 9339-9345.

Nusse R (1996) Patching up Hedgehog. Nature 384: 119-120.Nusse R (1997) A versatile transcriptional effector of Wingless signaling. Cell 89: 321-323.Ohmichi H, Koshimizu U, Matsumoto K & Nakamura T (1998) Hepatocyte growth factor (HGF)

acts as a mesenchyme-derived morphogenic factor during fetal lung development. Development125: 1315-1324.

77

Oh SP, Warman ML, Seldin MF, Cheng SD, Knoll JH, Timmons S & Olsen BR (1994) Cloning ofcDNA and genomic DNA encoding human type XVIII collagen and localization of the alpha1(XVIII) collagen gene to mouse chromosome 10 and human chromosome 21. Genomics 19:494-499.

Ohuchi H, Hori Y, Yamasaki M, Harada H, Sekine K, Kato S & Itoh N (2000) FGF10 acts as amajor ligand for FGF receptor 2 IIIb in mouse multi-organ development. Biochem Biophys ResCommun 277: 643-649.

Okada T, Iwamoto A, Kusakabe K, Mukamoto M, Kiso Y, Morioka H, Kodama H, Sasaki F &Morikawa Y (2001) Perinatal development of the rat kidney: proliferative activity andepidermal growth factor. Biol Neonate 79: 46-53.

Ornitz DM & Itoh N (2001) Fibroblast growth factors. Genome Biol 2: REVIEWS3005.O'Reilly MS, Boehm T, Shing Y, Fukai N, Vasios G, Lane WS, Flynn E, Birkhead JR, Olsen BR &

Folkman J (1997) Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell88: 277-285.

Orr-Urtreger A, Bedford MT, Burakova T, Arman E, Zimmer Y, Yayon A, Givol D & Lonai P(1993) Developmental localization of the splicing alternatives of fibroblast growth factorreceptor-2 (FGFR2). Dev Biol 158: 475-486.

Pachnis V, Mankoo B & Costantini F (1993) Expression of the c-ret proto-oncogene during mouseembryogenesis. Development 119: 1005-1017.

Pandur P & Kuhl M (2001) An arrow for wingless to take-off. Bioessays 23: 207-210.Park WY, Miranda B, Lebeche D, Hashimoto G & Cardoso WV (1998) FGF-10 is a chemotactic

factor for distal epithelial buds during lung development. Dev Biol 201: 125-134.Parr BA, Shea MJ, Vassileva G & McMahon AP (1993) Mouse Wnt genes exhibit discrete

domains of expression in the early embryonic CNS and limb buds. Development 119: 247-261.Parr BA & McMahon AP (1995) Dorsalizing signal Wnt-7a required for normal polarity of D-V

and A-P axes of mouse limb. Nature 374: 350-353.Parr BA & McMahon AP (1998) Sexually dimorphic development of the mammalian reproductive

tract requires Wnt-7a. Nature 395: 707-710.Parr BA, Cornish VA, Cybulsky MI & McMahon AP (2001) Wnt7b regulates placental

development in mice. Dev Biol 237: 324-332.Partanen AM, Alaluusua S, Miettinen PJ, Thesleff I, Tuomisto J, Pohjanvirta R & Lukinmaa PL

(1998) Epidermal growth factor receptor as a mediator of developmental toxicity of dioxin inmouse embryonic teeth. Lab Invest 78: 1473-1481.

Perl AK & Whitsett JA (1999) Molecular mechanisms controlling lung morphogenesis. Clin Genet56: 14-27.

Pepicelli CV, Kispert A, Rowitch DH & McMahon AP (1997) GDNF induces branching andincreased cell proliferation in the ureter of the mouse. Dev Biol 192: 193-198.

Pepicelli CV, Lewis PM & McMahon AP (1998) Sonic hedgehog regulates branchingmorphogenesis in the mammalian lung. Curr Biol 8: 1083-1086.

Perrimon N & Bernfield M (2000) Specificities of heparan sulphate proteoglycans indevelopmental processes. Nature 404: 725-728.

Perrimon N & Bernfield M (2001) Cellular functions of proteoglycans-an overview. Semin CellDev Biol 12: 65-67.

Peters KG, Werner S, Chen G & Williams LT (1992) Two FGF receptor genes are differentiallyexpressed in epithelial and mesenchymal tissues during limb formation and organogenesis in themouse. Development 114: 233-243.

Peters K, Werner S, Liao X, Wert S, Whitsett J & Williams L (1994) Targeted expression of adominant negative FGF receptor blocks branching morphogenesis and epithelial differentiationof the mouse lung. EMBO J. 13: 3296-3301.

Pierce RA, Mariencheck WI, Sandefur S, Crouch EC & Parks WC (1995) Glucocorticoidsupregulate tropoelastin expression during late stages of fetal lung development. Am J Physiol268(3 Pt 1): L491-500.

Pichel JG, Shen L, Sheng HZ, Granholm AC, Drago J, Grinberg A, Lee EJ, Huang SP, Saarma M,Hoffer BJ, Sariola H & Westphal H (1996) Defects in enteric innervation and kidneydevelopment in mice lacking GDNF. Nature 382: 73-76.

78

Pihlajaniemi T & Rehn M (1995) Two new collagen subgroups: membrane-associated collagensand types XV and XVII. Prog Nucleic Acid Res Mol Biol 50: 225-262.

Pinson KI, Brennan J, Monkley S, Avery BJ & Skarnes WC (2000) An LDL-receptor-relatedprotein mediates Wnt signalling in mice. Nature 407: 535-538.

Piscione TD, Yager TD, Gupta IR, Grinfeld B, Pei Y, Attisano L, Wrana JL & Rosenblum ND(1997) BMP-2 and OP-1 exert direct and opposite effects on renal branching morphogenesis.Am J Physiol 273(6 Pt 2): F961-975.

Piscione TD, Phan T & Rosenblum ND (2001) BMP7 controls collecting tubule cell proliferationand apoptosis via Smad1-dependent and -independent pathways. Am J Physiol Renal Physiol280: F19-33.

Plisov SY, Yoshino K, Dove LF, Higinbotham KG, Rubin JS & Perantoni AO (2001) TGF beta 2,LIF and FGF2 cooperate to induce nephrogenesis. Development 128: 1045-1057.

Pola R, Ling LE, Silver M, Corbley MJ, Kearney M, Blake Pepinsky R, Shapiro R, Taylor FR,Baker DP, Asahara T & Isner JM (2001) The morphogen Sonic hedgehog is an indirectangiogenic agent upregulating two families of angiogenic growth factors. Nat Med 7: 706-711.

Proetzel G, Pawlowski SA, Wiles MV, Yin M, Boivin GP, Howles PN, Ding J, Ferguson MW &Doetschman T (1995) Transforming growth factor-beta 3 is required for secondary palatefusion. Nat Genet 11: 409-414.

Pohl M, Sakurai H, Stuart RO & Nigam SK (2000a) Branching morphogenesis during kidneydevelopment. Annu Rev Physiol 62: 595-620.

Pohl M, Sakurai H, Stuart RO & Nigam SK (2000b) Role of hyaluronan and CD44 in in vitrobranching morphogenesis of ureteric bud cells. Dev Biol 224: 312-325.

Ponting CP, Phillips C, Davies KE & Blake DJ (1997) PDZ domains: targeting signallingmolecules to sub-membranous sites. Bioessays 19: 469-479.

Post M, Souza P, Liu J, Tseu I, Wang J, Kuliszewski M & Tanswell AK (1996) Keratinocytegrowth factor and its receptor are involved in regulating early lung branching. Development122: 3107-3115.

Powers CJ, McLeskey SW & Wellstein A (2000) Fibroblast growth factors, their receptors andsignaling. Endocr Relat Cancer 7: 165-197.

Pulkkinen L, Christiano AM, Airenne T, Haakana H, Tryggvason K & Uitto J (1994) Mutations inthe gamma 2 chain gene (LAMC2) of kalinin/laminin 5 in the junctional forms of epidermolysisbullosa. Nat Genet 6: 293-297.

Qiao J, Uzzo R, Obara-Ishihara T, Degenstein L, Fuchs E & Herzlinger D (1999a) FGF-7modulates ureteric bud growth and nephron number in the developing kidney Development.126:547-554.

Qiao J, Sakurai H & Nigam SK (1999b) Branching morphogenesis independent of mesenchymal-epithelial contact in the developing kidney. Proc Natl Acad Sci USA 96: 7330-7335.

Raatikainen-Ahokas A, Hytonen M, Tenhunen A, Sainio K & Sariola H (2000) BMP-4 affects thedifferentiation of metanephric mesenchyme and reveals an early anterior-posterior axis of theembryonic kidney. Dev Dyn 217: 146-158.

Rapraeger AC, Krufka A & Olwin BB (1991) Requirement of heparan sulphate for bFGF-mediatedfibroblast growth and myoblast differentiation. Science 252: 1705-1708.

Rapraeger AC (2001) Molecular interactions of syndecans during development. Semin Cell DevBiol 12: 107-116.

Rattner A, Hsieh JC, Smallwood PM, Gilbert DJ, Copeland NG, Jenkins NA & Nathans J (1997) Afamily of secreted proteins contains homology to the cysteine-rich ligand-binding domain offrizzled receptors. Proc Natl Acad Sci USA 94: 2859-2863.

Rehn M, Hintikka E & Pihlajaniemi T (1994) Primary structure of the alpha 1 chain of mouse typeXVIII collagen, partial structure of the corresponding gene, and comparison of the alpha1(XVIII) chain with its homologue, the alpha 1(XV) collagen chain. J Biol Chem 269: 13929-13935.

Rehn M & Pihlajaniemi T (1994) Alpha 1(XVIII), a collagen chain with frequent interruptions inthe collagenous sequence, a distinct tissue distribution, and homology with type XV collagen.Proc Natl Acad Sci USA 91: 4234-4238.

79

Rehn M, & Pihlajaniemi T (1995) Identification of three N-terminal ends of type XVIII collagenchains and tissue-specific differences in the expression of the corresponding transcripts. Thelongest form contains a novel motif homologous to rat and Drosophila frizzled proteins. J BiolChem 270: 4705-4711.

Rehn M, Hintikka E & Pihlajaniemi T (1996) Characterization of the mouse gene for the alpha 1chain of type XVIII collagen (Col18a1) reveals that the three variant N-terminal polypeptideforms are transcribed from two widely separated promoters. Genomics 32: 436-446.

Rehn M & Pihlajaniemi T (1996) Type XV and XVIII collagens, a new subgroup within the familyof collagens. Seminar in Cell & Development Biology 7: 673-679.

Rehn M, Pihlajaniemi T, Hofmann K & Bucher P (1998) The frizzled motif: in how many differentprotein families does it occur? Trends Biochem Sci 23: 415-417.

Relan NK, Yang Y, Beqaj S, Miner JH & Schuger L (1999) Cell elongation induces laminin alpha2chain expression in mouse embryonic mesenchymal cells: role in visceral myogenesis. J CellBiol 147: 1341-1350.

Roman J & McDonald JA (1992) Expression of fibronectin, the integrin alpha 5, and alpha-smoothmuscle actin in heart and lung development. Am J Respir Cell Mol Biol 6: 472-480.

Ross SE, Hemati N, Longo KA, Bennett CN, Lucas PC, Erickson RL & MacDougald OA (2000)Inhibition of adipogenesis by Wnt signaling. Science 289: 950-953.

Roth W, Wild-Bode C, Platten M, Grimmel C, Melkonyan HS, Dichgans J & Weller M (2000)Secreted Frizzled-related proteins inhibit motility and promote growth of human malignantglioma cells. Oncogene 19: 4210-4220.

Saarela J, Rehn M, Oikarinen A, Autio-Harmainen H & Pihlajaniemi T (1998) The short and longforms of type XVIII collagen show clear tissue specificities in their expression and location inbasement membrane zones in humans. Am J Pathol 153: 6116-6126.

Saarela J (1998) Type XV and XVIII collagen: The primary structure of variant human α1 (XVIII)chains, comparison of the expression of type XV and XVIII collagen mRNA transcripts, andlocation of type XVIII collagen protein in mature and developing human and mouse tissue. ActaUniv Oul D 449: 41-43.

Sainio K, Suvanto P, Davies J, Wartiovaara J, Wartiovaara K, Saarma M, Arumae U, Meng X,Lindahl M, Pachnis V & Sariola H (1997) Glial-cell-line-derived neurotrophic factor is requiredfor bud initiation from ureteric epithelium. Development 124: 4077-4087.

Sakurai H, Barros EJ, Tsukamoto T, Barasch J & Nigam SK. An in vitro tubulogenesis systemusing cell lines derived from the embryonic kidney shows dependence on multiple solublegrowth factors (1997) Proc Natl Acad Sci USA 94: 6279-6284.

Sanchez MP, Silos-Santiago I, Frisen J, He B, Lira SA & Barbacid M (1996) Renal agenesis andthe absence of enteric neurons in mice lacking GDNF. Nature 382: 70-73.

Sanford LP, Ormsby I, Gittenberger-de Groot AC, Sariola H, Friedman R, Boivin GP, Cardell EL& Doetschman T (1997) TGFbeta2 knockout mice have multiple developmental defects that arenon-overlapping with other TGFbeta knockout phenotypes. Development. 124: 2659-2670.

Sannes PL, Burch KK, Khosla J, McCarthy KJ & Couchman JR (1993) Immunohistochemicallocalization of chondroitin sulfate, chondroitin sulfate proteoglycan, heparan sulfateproteoglycan, entactin, and laminin in basement membranes of postnatal developing and adultrat lungs. Am J Respir Cell Mol Biol 8: 245-251.

Sariola H, Timpl R, von der Mark K, Mayne R, Fitch JM, Linsenmayer TF & Ekblom P (1984)Dual origin of glomerular basement membrane. Dev Biol 101: 86-96.

Sariola H, Ekblom P & Henke-Fahle S (1989) Embryonic neurons as in vitro inducers ofdifferentiation of nephrogenic mesenchyme. Dev Biol 132: 271-281.

Sariola H & Sainio K (1997) The tip-top branching ureter. Curr Opin Cell Biol 9: 877-884.Sarkar L, Cobourne M, Naylor S, Smalley M, Dale T & Sharpe PT (2000) Wnt/Shh interactions

regulate ectodermal boundary formation during mammalian tooth development. Proc Natl AcadSci USA 97: 4520-4524.

Saxén L (1977) Directive versus permissive induction: a working hypothesis. Soc Gen Physiol Ser32: 1-9.

Saxén L, Sariola H & Lehtonen E (1986) Sequential cell and tissue interactions governingorganogenesis of the kidney. Anat Embryol (Berl) 175: 1-6.

80

Saxén L (1987) Organogenesis of the kidney. Cambridge: Univ. Press. pp1-173.Schedl A & Hastie ND (2000) Cross-talk in kidney development. Curr Opin Genet Dev 10: 543-

549.Schuger L, Skubitz AP, de las Morenas A & Gilbride K (1995) Two separate domains of laminin

promote lung organogenesis by different mechanisms of action. Dev Biol 169: 520-532.Schuger L (1997) Laminins in lung development. Exp Lung Res 23: 119-129.Sekine K, Ohuchi H, Fujiwara M, Yamasaki M, Yoshizawa T, Sato T, Yagishita N, Matsui D,

Koga Y, Itoh N & Kato S (1999) Fgf10 is essential for limb and lung formation. Nat Genet 21:138-141.

Semenov MV, Tamai K, Brott BK, Kuhl M, Sokol S & He X (2001) Head inducer Dickkopf-1 is aligand for Wnt coreceptor LRP6. Curr Biol 11: 951-961.

Senior RM, Griffin GL, Mudd MS, Moxley MA, Longmore WJ & Pierce RA (1996) Entactinexpression by rat lung and rat alveolar epithelial cells. Am J Respir Cell Mol Biol 14: 239-247.

Shannon JM (1994) Induction of alveolar type II cell differentiation in fetal tracheal epithelium bygrafted distal lung mesenchyme. Dev Biol 166: 600-614.

Shannon JM, Gebb SA & Nielsen LD (1999) Induction of alveolar type II cell differentiation inembryonic tracheal epithelium in mesenchyme-free culture. Development 126: 1675-1688.

Simonet WS, DeRose ML, Bucay N, Nguyen HQ, Wert SE, Zhou L, Ulich TR, Thomason A,Danilenko DM & Whitsett JA (1995) Pulmonary malformation in transgenic mice expressinghuman keratinocyte growth factor in the lung. Proc Natl Acad Sci USA 92: 12461-12465.

Slack JMW (2001) Essential Developmental Biology. Blackwell Science Ltd. P 6-47.Slusarski DC, Yang-Snyder J, Busa WB & Moon RT (1997) Modulation of embryonic intracellular

Ca2+ signaling by Wnt-5A. Dev Biol 182: 114-120.Sokol SY, Klingensmith J, Perrimon N & Itoh K (1995) Dorsalizing and neuralizing properties of

Xdsh, a maternally expressed Xenopus homolog of dishevelled. Development 121: 1637-1647.Soriano P (1999) Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 21:

70-71.Srinivas S, Wu Z, Chen CM, D'Agati V & Costantini F (1999) Dominant effects of RET receptor

misexpression and ligand-independent RET signaling on ureteric bud development.Development 126: 375-386.

Stark K, Vainio S, Vassileva G & McMahon AP (1994) Epithelial transformation of metanephricmesenchyme in the developing kidney regulated by Wnt-4. Nature 372: 679-683.

Stark MR, Rao MS, Schoenwolf GC, Yang G, Smith D & Mauch TJ (2000) Frizzled-4 expressionduring chick kidney development. Mech Dev 98: 121-125.

Sund M, Vaisanen T, Kaukinen S, Ilves M, Tu H, Autio-Harmainen H, Rauvala H & PihlajaniemiT (2001) Distinct expression of type XIII collagen in neuronal structures and other tissuesduring mouse development. Matrix Biol 20: 215-231.

Takada S, Stark KL, Shea MJ, Vassileva G, McMahon JA & McMahon AP (1994) Wnt-3aregulates somite and tailbud formation in the mouse embryo. Genes Dev 8: 174-189.

Tanney DC, Feng L, Pollock AS & Lovett DH (1998) Regulated expression of matrixmetalloproteinases and TIMP in nephrogenesis. Dev Dyn 213: 121-129.

Ten Have-Opbroek A (1991) Lung development in the mouse embryo. Exp Lung Res 17: 111-130.Thesleff I, Vaahtokari A & Partanen AM (1995) Regulation of organogenesis. Common molecular

mechanisms regulating the development of teeth and other organs. Int J Dev Biol 39: 35-50.Thomas KR & Capecchi MR (1990) Targeted disruption of the murine int-1 proto-oncogene

resulting in severe abnormalities in midbrain and cerebellar development. Nature 346: 847-850.Thorpe CJ, Schlesinger A & Bowerman B (2000) Wnt signalling in Caenorhabditis elegans:

regulating repressors and polarizing the cytoskeleton. Trends Cell Biol 10: 10-7.Tickle C (1999) Morphogen gradients in vertebrate limb development. Semin Cell Dev Biol 10:

345-351.Tryggvason K (1993). The laminin family. Curr Opin Cell Biol 5: 877-882.Tsuda M, Kamimura K, Nakato H, Archer M, Staatz W, Fox B, Humphrey M, Olson S, Futch T,

Kaluza V, Siegfried E, Stam L & Selleck SB (1999) The cell-surface proteoglycan Dallyregulates Wingless signalling in Drosophila. Nature 400: 276-280.

81

Tunggal P, Smyth N, Paulsson M & Ott MC (2000) Laminins: structure and genetic regulation.Microsc Res Tech 51: 214-227.

Uusitalo M, Heikkila M & Vainio S (1999) Molecular genetic studies of Wnt signaling in themouse. Exp Cell Res 253: 336-348.

Vainio S, Lehtonen E, Jalkanen M, Bernfield M, Saxen L (1989) Epithelial mesenchymalinteractions regulate the stage-specific expression of a cell surface proteoglycan, syndecan, inthe developing kidney. Dev Biol 134: 382-391.

Vainio S, Jalkanen M, Bernfield M & Saxen L (1992) Transient expression of syndecan inmesenchymal cell aggregates of the embryonic kidney. Dev Biol 152: 221-232.

Vainio S, Karavanova I, Jowett A & Thesleff I (1993) Identification of BMP-4 as a signalmediating secondary induction between epithelial and mesenchymal tissues during early toothdevelopment. Cell 75: 45-58.

Vainio S & Müller U (1997) Inductive tissue interactions, cell signaling, and the control of kidneyorganogenesis. Cell 90: 975-978.

Vainio SJ, Itaranta PV, Perasaari JP & Uusitalo MS (1999) Wnts as kidney tubule inducing factors.Int J Dev Biol 43: 419-423.

Vilar J, Gilbert T, Moreau E & Merlet-Benichou C (1996) Metanephros organogenesis is highlystimulated by vitamin A derivatives in organ culture. Kidney Int 49: 1478-1487.

Vinson CR & Adler PN (1987) Directional non-cell autonomy and the transmission of polarityinformation by the frizzled gene of Drosophila. Nature 329: 549-551.

Vinson CR, Conover S & Adler PN (1989) A Drosophila tissue polarity locus encodes a proteincontaining seven potential transmembrane domains. Nature 338: 263-264.

Virtanen I, Laitinen A, Tani T, Paakko P, Laitinen LA, Burgeson RE & Lehto VP (1996)Differential expression of laminins and their integrin receptors in developing and adult humanlung. Am J Respir Cell Mol Biol 15: 184-196.

Volpert OV, Lawler J & Bouck NP (1998) A human fibrosarcoma inhibits systemic angiogenesisand the growth of experimental metastases via thrombospondin-1. Proc Natl Acad Sci USA 95:6343-6348.

Vortkamp A, Lee K, Lanske B, Segre GV, Kronenberg HM & Tabin CJ (1996) Regulation of rateof cartilage differentiation by Indian hedgehog and PTH-related protein. Science 273: 613-622.

Vukicevic S, Kopp JB, Luyten FP & Sampath TK (1996) Induction of nephrogenic mesenchyme byosteogenic protein 1 (bone morphogenetic protein 7). Proc Natl Acad Sci USA 93: 9021-9026.

Wakeman JA, Walsh J & Andrews PW (1998) Human Wnt-13 is developmentally regulated duringthe differentiation of NTERA-2 pluripotent human embryonal carcinoma cells. Oncogene 17:179-186.

Warburton D, Wuenschell C, Flores-Delgado G & Anderson K (1998) Commitment anddifferentiation of lung cell lineages. Biochem Cell Biol 76: 971-995.

Warburton D & Lee MK (1999) Current concepts on lung development. Curr Opin Pediatr 11: 188-192.

Warburton D, Schwarz M, Tefft D, Flores-Delgado G, Anderson KD & Cardoso WV (2000) Themolecular basis of lung morphogenesis. Mech Dev 92: 55-81.

Weaver M, Dunn NR & Hogan BL (2000) Bmp4 and Fgf10 play opposing roles during lung budmorphogenesis. Development 127: 2695-2704.

Wehrli M, Dougan ST, Caldwell K, O'Keefe L, Schwartz S, Vaizel-Ohayon D, Schejter E,Tomlinson A & DiNardo S (2000) arrow encodes an LDL-receptor-related protein essential forWingless signalling. Nature 407: 527-530.

Wessells NK (1970) Mammalian lung development: interactions in formation and morphogenesisof tracheal buds. J Exp Zool 175: 455-466.

Werb Z, Vu TH, Rinkenberger JL & Coussens LM (1999) Matrix-degrading proteases andangiogenesis during development and tumor formation. APMIS 107: 11-8.

Whitsett J (1998) A lungful of transcription factors. Nat Genet 20: 7-9.Wodarz A & Nusse R (1998) Mechanisms of Wnt signaling in development. Annu Rev Cell Dev

Biol 14: 59-88.Wolpert L, Beddington R, Brockes J, Jessell T, Lawrence P & Meyerowitz (1998) Peinciples of

development. Oxford University Press, pp274-275.

82

Wright E, Hargrave MR, Christiansen J, Cooper L, Kun J, Evans T, Gangadharan, U, Greenfield A& Koopman P (1995) The Sry-related gene Sox9 is expressed during chondrogenesis in mouseembryos. Nat Genet 9: 15-20.

Wright C, Strauss S, Toole K, Burt AD & Robson SC (1999) Composition of the pulmonaryinterstitium during normal development of the human fetus. Pediatr Dev Pathol 2: 424-431.

Xu J, Liu Z & Ornitz DM (2000). Temporal and spatial gradients of Fgf8 and Fgf17 regulateproliferation and differentiation of midline cerebellar structures. Development 127: 1833-1843.

Xu PX, Adams J, Peters H, Brown MC, Heaney S & Maas R (1999) Eya1-deficient mice lack earsand kidneys and show abnormal apoptosis of organ primordia. Nat Genet 23: 113-117.

Yamaguchi TP, Bradley A, McMahon AP & Jones S (1999) A Wnt5a pathway underlies outgrowthof multiple structures in the vertebrate embryo. Development 126: 1211-1223.

Yoshikawa Y, Fujimori T, McMahon AP & Takada S (1997) Evidence that absence of Wnt-3asignaling promotes neuralization instead of paraxial mesoderm development in the mouse. DevBiol 183: 234-242.

Yoshino K, Rubin JS, Higinbotham KG, Uren A, Anest V, Plisov SY & Perantoni AO 2001)Secreted Frizzled-related proteins can regulate metanephric development. Mech Dev 102: 45-55.

Zakin LD, Mazan S, Maury M, Martin N, Guenet JL & Brulet P (1998) Structure and expression ofWnt13, a novel mouse Wnt2 related gene. Mech Dev 73: 107-116.

Zeng L, Fagotto F, Zhang T, Hsu W, Vasicek TJ, Perry WL 3rd, Lee JJ, Tilghman SM, GumbinerBM & Costantini F (1997) The mouse Fused locus encodes Axin, an inhibitor of the Wntsignaling pathway that regulates embryonic axis formation. Cell 90: 181-192.

Zeng W, Wharton KA Jr, Mack JA, Wang K, Gadbaw M, Suyama K, Klein PS & Scott MP (2000)naked cuticle encodes an inducible antagonist of Wnt signalling. Nature 403: 789-795.

Zeng X, Wert SE, Federici R, Peters KG & Whitsett JA (1998) Enhances pulmonaryvasculogenesis and disrupts lung morphogenesis in vivo. Dev Dyn 211: 215-227.

Zeng X, Gray M, Stahlman MT & Whitsett JA (2001) TGF-beta1 perturbs vascular developmentand inhibits epithelial differentiation in fetal lung in vivo. Dev Dyn 221: 289-301.

Zhang H & Bradley A (1996) Mice deficient for BMP2 are nonviable and have defects inamnion/chorion and cardiac development. Development 122: 2977-2986.

Zhang XM, Ramalho-Santos M & McMahon AP (2001) Smoothened mutants reveal redundantroles for Shh and Ihh signaling including regulation of L/R asymmetry by the mouse node. Cell105: 781-792.

Zhu Y, Oganesian A, Keene DR & Sandell LJ (1999) Type IIA procollagen containing thecysteine-rich amino propeptide is deposited in the extracellular matrix of prechondrogenic tissueand binds to TGF-beta1 and BMP-2. J Cell Biol 144: 1069-1080.