REVIEW

Growth factor–heparan sulfate “switches” regulating stagesof branching morphogenesis

Sanjay K. Nigam & Kevin T. Bush

Received: 12 June 2013 /Revised: 28 November 2013 /Accepted: 4 December 2013 /Published online: 2 February 2014# IPNA 2014

Abstract The development of branched epithelial organs, suchas the kidney, mammary gland, lung, pancreas, and salivarygland, is dependent upon the involvement and interaction ofmultiple regulatory/modulatory molecules, including solublegrowth factors, extracellular matrix components, and their re-ceptors. How the function of these molecules is coordinated tobring about the morphogenetic events that regulate iterative tip-stalk generation (ITSG) during organ development remains tobe fully elucidated. A common link to many growth factor-dependent morphogenetic pathways is the involvement of var-iably sulfated heparan sulfates (HS), the glycosaminoglycanbackbone of heparan sulfate proteoglycans (HSPG) on extracel-lular surfaces. Genetic deletions of HS biosynthetic enzymes(e.g.,C5-epimerase, Hs2st), as well as considerable in vitro data,indicate that variably sulfated HS are essential for kidney devel-opment, particularly in Wolffian duct budding and early uretericbud (UB) branching. A role for selective HS modifications byenzymes (e.g., Ext, Ndst, Hs2st) in stages of branching morpho-genesis is also strongly supported for mammary gland ductalbranching, which is dependent upon a set of growth factorssimilar to those involved in UB branching. Taken together, these

studies provide support for the notion that the specific spatio-temporal HS binding of growth factors during the developmentof branched epithelial organs (such as the kidney, mammarygland, lung and salivary gland) regulates these complex pro-cesses by potentially acting as “morphogenetic switches” duringthe various stages of budding, branching, and other develop-mental events central to epithelial organogenesis. It may be thattwo or more growth factor-selective HS interactions constitutea functionally equivalent morphogenetic switch; this may helpto explain the paucity of severe branching phenotypes withindividual growth factor knockouts.

Keywords Branchingmorphogenesis . Heparan sulfate .

Heparan sulfate proteoglycan . Kidney . Ureteric bud .

Mammary gland . Breast . Growth factors .

2-O-sulfotransferase . 6-O-sulfotransferase

Introduction

Many growth factors have been identified, both in vivo andin vitro, as playing key roles in morphogenetic processesessential for kidney development, including ureteric bud(UB) emergence from the Wolffian duct (WD), UB branchingmorphogenesis and the metanephric mesenchyme (MM)-derived nephron formation. These growth factors includeglial-derived neurotrophic factor (GDNF) [1], pleiotrophin[2], hepatocyte growth factor (HGF) [3], epidermal growthfactor (EGF) receptor ligands (e.g., EGF, HBEGF, TGFα)[4–6], WNT family members [7], bone morphogenetic pro-teins (BMPs) [8–10], TGFβ [10, 11], fibroblast growth factors(FGFs) [12, 13], leukemia inhibitory factor (LIF) [14, 15] andothers. One of the striking features of these growth factors istheir ability to bind heparan-sulfate [HS; the glycosaminogly-can backbone of heparan sulfate proteoglycans (HSPGs)] and,indeed, their differential binding to selectively sulfated

S. K. NigamDepartment of Medicine, University of California, La Jolla,San Diego, CA 92093-0693, USA

S. K. Nigam :K. T. BushDepartment of Pediatrics, University of California, La Jolla,San Diego, CA 92093-0693, USA

S. K. NigamDepartment of Cellular and Molecular Medicine, University ofCalifornia, La Jolla, San Diego, CA 92093-0693, USA

S. K. Nigam (*)Department of Cellular & Molecular Medicine, Pediatrics, Medicineand Bioengineering, University of California, 9500 Gilman Drive,La Jolla, CA, USAe-mail: nigamlab@ucsd.edu

Pediatr Nephrol (2014) 29:727–735DOI 10.1007/s00467-013-2725-z

heparan moieties (e.g., 2-OS versus 6-OS) [16–21] (Table 1).HS, which is abundantly present within the developing kidneyand can be associated with the cell surface, basement mem-brane, or the extracellular matrix, is essential for renal devel-opment and most soluble factors known to induce key devel-opmental stages bind heparin (a highly sulfated form of hep-aran sulfate). In fact, many of the growth factors above wereinitially purified based on elution from heparin columns withvarious concentrations of salt; in some cases, they were puri-fied in this manner using the isolated UB culture system or theisolated MM culture system as an assay [2, 10, 14, 22, 23].

During its biosynthesis, HS is selectively modified andvariably sulfated by a series of specialized enzymes locatedin the Golgi compartment, including polymerases [i.e.,exostosin glycosyltransferases (Ext) family], an epimerase (i.e.,C5-epimerase) and various sulfotransferases [i.e.,N-deacetylase/N-sulfotransferase (Ndst) family, 2-O-sulfotransferase (Hs2st),the glucosaminyl 6-O-sulfotransferases (Hs6st) and 3-O-sulfotransferases (Hs3st)] [24–26] (Fig. 1). The first step in HSsynthesis is linear polymerization of glucuronic acid-N-acetylglucosamine disaccharide units mediated by the EXT enzymecomplex (Fig. 1). The growing chain then undergoes modifica-tion by the coordinated action of specific sulfotransferases and aC5 epimerase. The first sulfation step is mediated by the N-deacetylase N-sulfotransferase (NDST) family of enzymes. In

mammals, there are four isoforms of the NDSTenzyme, each ofwhich appears to have a specific spatiotemporal distribution[27]. For example, NDST1 and 2 are expressed in all embryonicand adult tissues, but at different levels [28–30], while tran-scripts of NDST3 and 4 are predominantly expressed duringembryogenesis [28, 31]. Further modifications to the HS chainoccur after the introduction of N-sulfate groups; these includeC5 epimerization of glucuronic acid to iduronic acid and O-sulfation (i.e., 2-O-sulfation, 6-O-sulfation and 3-O-sulfation) atvarious positions. These sulfation modifications occur only on asubset of the residues creating an almost endless variety ofdifferent HS-GAG chains [25, 32].

Renal development: growth factor–heparan sulfateinteractions

The development of the mammalian kidney is dependent uponthe coordinated and reciprocal interactions of two primordialtissues, the UB and the MM. Based on the sequence ofbranching events and known molecular pathways involved inthese events (from in vitro and in vivo data), a three-stage modelof development can be considered: (1) UB outgrowth from theWolffian duct (initiation of urinary tract development), (2) UBgrowth into the T-shape and branching through iterative tip-stalkgeneration (ITSG) (development of the upper collecting system)and (3) UB and MM interactions leading to the induction anddifferentiation of the MM (initiation and development of thenephron) as well as maturation of the UB into the collectingducts. A common feature among developing branched epithelialtissues, such as the kidney, appears to be a critical role for HS asa regulatory molecule for many key morphogenetic growthfactors (Table 1). We have previously proposed that thesegrowth factors, either alone or in combination, not only bindto selectively sulfated heparan, but in some cases, it is thiscombination (of the growth factor bound to selectively sulfatedheparan) that acts as a “morphogenetic switch” in WD budding,UB branching, and nephron development [10, 18, 19, 33–35](Fig. 2). Furthermore, it may be that these different combina-tions of growth factors with selectively sulfated heparan arefunctionally comparable as “morphogenetic switches”, particu-larly for the iterative tip-stalk generating mechanisms responsi-ble for UB branching [36, 37] (Fig. 3). For example, the asso-ciation of growth factor “X” with HS “A” may be functionallycomparable, from the viewpoint of the ability to induce UBiterative tip-stalk generation, to that of growth factor “Y” withHS “B.” Given that many of the growth factors (capable ofinteracting with HS) mentioned above have been implicated inUB branching, the combinatorial possibilities of “morphogenet-ically equivalent switches” are quite large.

This may help explain the apparent “redundancy” ofgrowth factors in branching based on single-gene knockoutresults. Generally, growth factor knockouts in the kidney

Table 1 Morphogen and heparan sulfate proteoglycan interactions

Factors Characteristic of factor–HS interaction

GDNF 2-O-sulfation required for biding to GFRα1 [49]

BMP4 Activity modulate by glypican-3 [50]

BMP4 antagonist, Noggin, binds HS [51]

BMP2 Activity modulated by glypican-3 [52]

Wnt Activity stimulated by desulfation at 6-O-position [53]

Pleiotrophin Binds to syndecan [54]

FGF1 2-O- and 6-O-sulfation required for binding [55]

Cell bound HS mediates receptor activity [56]

Highly sulfated HS potentiates mitogenesis [57]

FGF2 Pentasaccharide with N-sulfation and 2-O-sulfationsufficient for binding [55]

6-O-sulfation required for activation of FGFR1 [55]

FGF7 Heparin abrogates activation of FGFR2IIIb [58]

FGF 6-O-sulfation important for binding [59]

TGFβ1 HS potentiates dimerization

Requires high level of N-sulfation for binding [60]

HB-EGF Cell surface HSPG optimizes activity [61]

Endostatin Proteolytic cleavage product of collagen XVII

Requires interaction with glypican for activity [62]

Adapted from: Shah MM, Sampogna RV, Sakurai H, Bush KT,Nigam SK. Branching morphogenesis and kidney disease.Development. 2004; 131:1449-1462 (used with permission)

728 Pediatr Nephrol (2014) 29:727–735

show a mild or no effect on branching, despite the factthat they have a profound effect in in vitro systems such asthe isolated UB culture or UB cell culture in 3D matrices.However, these are often “minimal systems” designed toisolate a particular effect for an experimental purpose. In vivo,nearly all of the combinations of “morphogenetically equiva-lent switches”—of different growth factors and selectivelysulfated heparans—would be expected to remain active inthe knockout. Thus, even if one growth factor (growth factorX) is knocked out, thereby inactivating the switch involvinggrowth factor X and HS A, a different switch, involvinggrowth factor Y and HS B, might be able to take over, thusexplaining the apparent redundancy.

Moreover, we have argued that different stages inbranching—from WD budding to early UB branchingto late UB branching—are regulated by these growthfactor-selectively sulfated HS morphogenetic switches[10, 18, 19, 33–35]. According to this model, in thekidney, it is unlikely that single knockouts of heparin-binding growth factors will have a major effect on UBbranching. It may be necessary to knockout more thanone growth factor gene or to knock out the selectivesulfation mechanism (thereby destroying the ability ofselectively sulfated HS to bind multiple growth factors)in order to obtain clear budding and branchingphenotypes.

Fig. 1 Schematic diagram of heparan sulfate synthesis. The heparansulfate (HS) chain is linked via xylose (orange star) to a serine residueon a core protein (light blue line). Ext(s) then catalyze HS chain polymer-ization with units of D-glucuronic acid (GlcA) and N-acetyl glucosamine(GlcNac). The growing chain undergoes modification by the coordinatedaction of specific biosynthetic enzymes; the initial modification is medi-ated by the N-deacetylase N-sulfotransferase (NDST) family of enzymes,which sulfate the GlcNAc residues at the N-position. Further modifica-tions to the HS chain occur after N-sulfation. First, the C5 epimerase actson GlcA residues immediately adjacent to the N-sulfated-GlcNAcs tocreate L-iduronic acid (IdoA) residues, which can be sulfated at the 2-O

position by the 2-O-sulfotransferase. Next, the 6-O-sulfotransferase addssulfate groups to the 6-OH of the GlcNAc residues adjacent to the uronicacid. Finally, certain arrangements of sulfated sugar and IdoA residues actas a target for the 3-O-sulfotransferase. In general, the reactions proceedsequentially, and the specific arrangement of sulfated residues and iduronicacid epimers in the HSPG gives rise to the specific binding sequences forgrowth factors. (Adapted from Crawford BE, Garner OB, Bishop JR,Zhang DY, Bush KT, Nigam SK, Esko JD. Loss of the heparan sulfatesulfotransferase, Ndst1, in mammary epithelial cells selectively blockslobuloalveolar development in mice. PLoS One 2010; 5:e10691, usedwith permission)

Pediatr Nephrol (2014) 29:727–735 729

In the kidney, considerable in vitro and in vivo data is, atleast, consistent with this model. Moreover, as we discussbelow, in the mammary gland, a system in which the mesen-chyme plays a less obvious morphogenetic role (in that, while

secreting factors, it does not itself give rise to epithelial cellsthat become a major part of the tubular-ductal system asoccurs in the kidney) the data appears even more consistentwith this model.

Gene deletions and kidney development

Gene deletion studies have been performed and defects inkidney development have been described for some of theHS biosynthetic enzymes. For example, deletion of Hs2stresults in an apparent defect in UB branching morphogenesisleading to renal agenesis and perinatal lethality [37]. However,we have recently shown that the UB branching defect isprobably largely secondary to a disruption in the inductionand differentiation of the metanephric mesenchyme (MM) dueto alterations in the HS-mediated binding of growth factors bythe MM cells [19]. A similar knockout phenotype was report-ed in mice deficient in C5 epimerase [39], however since C5epimerase and Hs2st appear to interact as a functional com-plex in vivo [40], the similarity among these mutants is notsurprising.

In vitro analysis

However, for many of the other biosynthetic enzymes, multi-ple isoforms are found in vertebrates and knockouts of indi-vidual isoforms have not been found to display overt kidneydefects. Nevertheless, the importance of the HS to kidneydevelopment has been investigated in vitro using whole em-bryonic kidney [41–45] as well as culture of isolated kidneyprogenitor tissues [18, 19, 35]. For example, the addition ofsodium chlorate, a competitive inhibitor of sulfate for thevarious sulfotransferase enzymes, to culture media of thewhole kidney, as well as to the isolated UB results in perturbedbranching morphogenesis [35, 41] (Fig. 4). In addition, treat-ment with heparin lyase (a glycanase that specifically de-grades HS) but not chondroitinase ABC [which degradeschondroitin sulfate (CS) and dermatan sulfate but not HS]markedly impaired branching of the UB [35, 41] (Fig. 4).Furthermore, perturbation of both CS and HS synthesis, butnot perturbation of CS synthesis alone, significantly affectedUB branching morphogenesis [35] (Fig. 4). Taken together,these results strongly support the notion that HS plays acritical role in UB branching morphogenesis.

However, these studies did not examine the role of variablesulfation in kidney development. Using heparin (a maximallysulfated form of HS) and chemically modified heparin, it hasbeen shown that 2-O sulfation of HS is an important mediatorof GDNF signaling [45]. More recently, the effect of variablysulfated heparinoids (i.e., 2-de-O-sulfated heparin and 6-de-O-sulfated heparin) on the growth and development of therodent kidney was investigated in in vitro models of kidneydevelopment (i.e., isolated UB, isolated MM, and whole

Fig. 2 Diagram illustrating the relative importance of 2-O-sulfated HSversus 6-O-sulfated HS in the major morphogenetic processes critical tokidney development. Together with our earlier study [19], data suggestthat 2-O-sulfated heparan sulfate (HS) has greater importance in MM-derived nephron formation, while 6-O-sulfated HS appear to have greaterimportance in UB branching morphogenesis. (With permission fromShah MM, Sakurai H, Gallegos TF, Sweeney DE, Bush KT, Esko JD,Nigam SK. Growth factor-dependent branching of the ureteric bud ismodulated by selective 6-O sulfation of heparan sulfate. DevelopmentalBiology 2011; 356:19-27)

Fig. 3 Proposed model for the role of variably sulfated HS in uretericbud (UB) branching morphogenesis. Heparan sulfate (HS) is postulatedto mediate the transitions from stages of development through the differ-ential binding of stimulatory and inhibitory growth factors (GF), effec-tively creating a “switch” between different stages of branching. Forexample, near the tips of the UB, HS may modulate growth and branchpromoting signals, while at the stalks, HS may modulate inhibitorysignals to prevent further budding and maintain tubule caliber. (Withpermission from Steer DL, Shah MM, Bush KT, Stuart RO, SampognaRV, Meyer TN, Schwesinger C, Bai X, Esko JD, Nigam SK. Regulationof ureteric bud branchingmorphogenesis by sulfated proteoglycans in thedeveloping kidney. Developmental Biology 2004; 272:310-327)

730 Pediatr Nephrol (2014) 29:727–735

embryonic kidney culture) [18, 19]. In support of the notionthat specifically sulfated HS can potentially control/modulategrowth factor activity (either positively or negatively) andthus may act as a regulatory mechanism for specific morpho-genetic events during the stages of kidney development, it wasfound that these variably sulfated heparinoids have differentialeffects on UB branching morphogenesis. For example, it wasfound that factors that bind to 2-O-sulfated HS are involvedmore in MM induction while growth factors that bind to 6-Osulfated HS are more involved in UB growth and branchingmorphogenesis [18, 19] (Fig. 2). Moreover, column fraction-ation of a MM-cell-conditioned media showed that UBbranch-stimulating factors bind with higher affinity to the 6-O sulfated moiety of HS than they do to the 2-O moiety [18](Fig. 5). Although these findings do not address the mecha-nism of action of these variably sulfated heparinoids, this datastrongly supports the view that different sulfation patterns arelikely to have distinct effects on different growth factor-mediated stages of nephrogenesis. Furthermore, the data raisethe possibility that differential binding of growth factors tospecifically sulfated HS chains likely modulate these highlyregulated morphogenetic events.

Heparan sulfate in developing mammary gland and otherbranched organs

Similar studies have been performed in other branching or-gans. For example, treatment of isolated embryonic lung budswith either heparin lyase or sodium chlorate has a markedlyinhibitory effect, even in the presence of FGF10, suggesting

that essential FGF signaling is disrupted in the absence ofsulfated heparan [46]. In cultured lung explants, treatmentwith sodium chlorate was found to hinder branching morpho-genesis [46, 47], while, in contrast, treatment with low dosesof heparin induced an enhancement of epithelial branching[48]. Interestingly, as with the kidney, the addition of variablysulfated forms of heparin had differing effects on lungbranching morphogenesis. For example, while treatment ofcultured lung explants with de-O-sulfated heparin had littleor no effect on branching morphogenesis, the addition of over-O-sulfated heparin disrupted branching and resulted in mark-edly enlarged ducts [46, 47]. Moreover, although the additionof N-sulfated heparin did not rescue branching in sodiumchlorate-treated lung cultures, the addition of either over-sulfated, 2-de-O-sulfated, or 6-de-O-sulfated heparin did[47]. This indicates a requirement for O-sulfation. InDrosophila, it was found that mutation of both hs6st andhs2st was required to disrupt tracheal branching morphogen-esis, indicating that compensation for the loss of one biosyn-thetic enzyme can occur [49]. Finally, in lung, not only is therea great diversity of HS structure, but there is a clear spatial andtemporal alteration of HS structure during lung development[46, 50, 51]. However, to date only mice lacking Hs6st1 (oneof threeHs6st isoforms found inmammals [52]) were found tohave defects in lung development—these mice have perturbedalveolarization, a late stage of lung morphogenesis [53].

In the salivary gland, early in vitro studies found thattreatment of cultured salivary glands with β-xyloside(a competitive inhibitor of the synthesis of the glycosamino-glycan chains onHSPGs), inhibited branchingmorphogenesis

Fig. 4 Perturbation of heparan sulfate (HS) synthesis alters the growth ofcultured embryonic kidney. (a–j) Fluorescent photomicrographs of E13rat kidneys cultured for 3 days in the absence (a, control) or presence ofeither varying concentrations of sodium chlorate (b–d), chlorate (30 mM)plus sulfate (30 mM) (e), 100 mU/ml heparitinase (f), 100 mU/mlchondroitinase ABC (g), boiled heparitinase, and chondroitinase ABC(h), 250 μ5 naphthalene xyloside (inhibits native synthesis of both HS

and CS (i), or 250 μ5 decalin xyloside (inhibits native synthesis of CSalone) (j). k Bar graph showing the effects of the various treatments onureteric bud branching morphogenesis. (With permission from Steer DL,Shah MM, Bush KT, Stuart RO, Sampogna RV, Meyer TN, SchwesingerC, Bai X, Esko JD, Nigam SK. Regulation of ureteric bud branchingmorphogenesis by sulfated proteoglycans in the developing kidney.Developmental Biology 2004; 272:310-327)

Pediatr Nephrol (2014) 29:727–735 731

Fig. 5 Morphogenetic activity of heparin and de2OS and de6OS columnfractions. Phase contrast photomicrographs of isolated UB cultured ineither whole conditioned media (BSN-CM) (a), DMEM/F12 (b), orpurified fractions (c–t) eluted with various NaCl concentrations from aheparin column (c–h), a 2-OS-depleted heparin column (i–n) or a 6-OS-depleted heparin column (o–t) various fractionation columns. A progres-sive downward shift of binding affinity of the branch stimulating activityis clearly seen, suggesting that 6-O sulfation is necessary for high affinity

binding. Of note, robust branch stimulatory activity is detected in theflow-through fraction of the de6OS column (o), but not in the heparin (c)or de2OS (i) columns, indicating a lack of binding of branch stimulatingfactors to the de6OS column despite the presence of other sulfationmodifications. (With permission from Shah MM, Sakurai H, GallegosTF, Sweeney DE, Bush KT, Esko JD, Nigam SK. Growth factor-depen-dent branching of the ureteric bud is modulated by selective 6-O sulfationof heparan sulfate. Developmental Biology 2011; 356:19-27)

Table 2 Effects of deletions of genes involved in HS initiation, polymerization and sulfation on mammary gland branching and differentiation

Stage of mammary gland development affected

Deleted gene Phenotype Primarybranching

Secondarybranching

Lobuloalveolardevelopment

Reference

Ext1 Lack of branching ductal epithelium X [45]

Ndst1 Lack of lobuloalveolar expansion X [44]

Ndst2 Mild increase in branching X X [43]

Ndst1/Ndst2 Hyperbranching X X [43]

Hs2st Decrease in secondary/ductal side-branching X [45]

Genes are listed in the order of heparan sulfatemodifications catalyzed duringGAG assembly and sulfation: Ext→Ndst→Hsglce→Hs2st→Hs6st→Hs3st

Adapted from: Bush KT, Crawford BE, Garner OB, Nigam KB, Esko JD, Nigam SK. N-sulfation of heparan sulfate regulates early branching events inthe developing mammary gland. J Biol Chem 2012; 287:42064-42070, used with permission

732 Pediatr Nephrol (2014) 29:727–735

in cultured salivary glands [54]. More recently, it was foundthat differences in the sulfation patterns of the HS can influ-ence both duct elongation and end bud expansion duringsubmandibular gland development [55]. It has also beenshown that deletion of Hs6st disrupts lacrimal gland buddingand formation, while it is completely abolished in micelacking both Hs2st and Hs6st [21]. In addition, we haveconditionally deleted various HS biosynthetic enzymes in

mammary epithelial cells [56–58] (Table 2). Not surprisingly,deletion of Ext1, a subunit of the co-polymerase complex thatcatalyzes the initial formation of the HS GAG chain,completely blocked primary ductal branching morphogenesis[58]. In contrast, primary ductal epithelial branching was notinhibited by deletion of Hs2st; however there was a markeddecrease in secondary and ductal side-branching as well as areduction in the number of bifurcated terminal end buds [58](Fig. 6; Table 2). Interestingly, the phenotype was similar tothe knockout of the Hgf-receptor (cmet) [58]. Ndst1 apparent-ly affects a different stage of mammary gland development asbranching morphogenesis appeared normal in the absence ofthis biosynthetic enzyme [57]. However, deletion of Ndst1inhibits the formation of lobuloalveoli from the ductal epithe-lia (a late stage of ductal epithelial development), resulting ininsufficient milk production [57] (Table 2). On the other hand,while deletion of Ndst2, the other member of this enzymefamily expressed in mammary epithelia, resulted in a mildincrease in ductal branched structures, deletion of both ofthese HS biosynthetic enzymes resulted in a dramatic andsignificant increase in both primary and secondary ductalbranching (i.e., a “hyperbranching” phenotype) [56](Table 2). Taken together, the data demonstrate regulation ofindividual stages of epithelial branching morphogenesis byvariably sulfated HS and are consistent with the notion thatselective HS-growth factor interactions help “switch” thestages of branching in the organogenesis of epithelial tissues.

Conclusions

A variety of in vitro cell culture, ex vivo organ culture, andin vivo knockout studies indicate that similar sets of heparin-binding growth factors, such as those that activate met (Hgf)and Egf receptor, are involved in UB and mammary ductbranching morphogenesis during organ development [5,59–61]. Likewise, current evidence suggests a key role forvarious HSmodifications for stages in early UB formation andbranching; as we discuss, this is now very clear in the mam-mary gland [56–58]. Collectively, the data suggests an inti-mate relationship between growth factors and selective hepa-ran sulfation in various stages of budding and branchingduring epithelial organogenesis. Nevertheless, much morebiochemical, genetic, and structural work remains to be doneto define the extent to which growth factor-selective HSinteractions act, as we propose, as the equivalent of “switches”regulating the stages of branching, as well as other stages oforganogenesis.

Acknowledgments The authors wish to thank Dr. Wei Wu for hisassistance and helpful comments on the manuscript.

Fig. 6 Deletion of Hs2st perturbs mammary ductal epithelialbranching morphogenesis. a, b Whole mounts of 5-week fourthinguinal glands from either (a) Hs2stf/fMMTVCre − (wild type) orb Hs2stf/fMMTVCre + knockout mice. c Bar graph showing theeffect of Hs2st deletion on ductal branching as determined by thenumber of branch points per 1 mm2 in a 10-week gland. (Withpermission adapted from Garner OB, Bush KT, Nigam KB,Yamaguchi Y, Xu D, Esko JD, Nigam SK. Stage-dependent reg-ulation of mammary ductal branching by heparan sulfate andHGF-cMet signaling. Developmental Biology 2011; 355:394-403)

Pediatr Nephrol (2014) 29:727–735 733

References

1. Costantini F, Kopan R (2010) Patterning a complex organ: branchingmorphogenesis and nephron segmentation in kidney development.Dev Cell 18:698–712

2. Sakurai H, Bush KT, Nigam SK (2001) Identification of pleiotrophinas a mesenchymal factor involved in ureteric bud branching morpho-genesis. Development 128:3283–3293

3. Santos OF, Moura LA, Rosen EM, Nigam SK (1993) Modulation ofHGF-induced tubulogenesis and branching by multiple phosphory-lation mechanisms. Dev Biol 159:535–548

4. Barros EJ, Santos OF, Matsumoto K, Nakamura T, Nigam SK (1995)Differential tubulogenic and branching morphogenetic activities ofgrowth factors: implications for epithelial tissue development. ProcNatl Acad Sci USA 92:4412–4416

5. Sakurai H, Tsukamoto T, Kjelsberg CA, Cantley LG, Nigam SK(1997) EGF receptor ligands are a large fraction of in vitro branchingmorphogens secreted by embryonic kidney. Am J Physiol 273:F463–472

6. Ishibe S, Karihaloo A,Ma H, Zhang J,Marlier A, MitobeM, TogawaA, Schmitt R, Czyczk J, Kashgarian M, Geller DS, Thorgeirsson SS,Cantley LG (2009) Met and the epidermal growth factor receptor actcooperatively to regulate final nephron number and maintaincollecting duct morphology. Development 136:337–345

7. Carroll TJ, McMahon AP (2000) Secreted molecules in metanephricinduction. J Am Soc Nephrol 11(Suppl 16):S116–119

8. Piscione TD, Phan T, Rosenblum ND (2001) BMP7 controlscollecting tubule cell proliferation and apoptosis via Smad1-dependent and -independent pathways. Am J Physiol Renal Physiol280:F19–33

9. Piscione TD, Yager TD, Gupta IR, Grinfeld B, Pei Y, Attisano L,Wrana JL, Rosenblum ND (1997) BMP-2 and OP-1 exert direct andopposite effects on renal branching morphogenesis. Am J Physiol273:F961–975

10. Bush KT, Sakurai H, Steer DL, Leonard MO, Sampogna RV, MeyerTN, Schwesinger C, Qiao J, Nigam SK (2004) TGF-beta superfamilymembers modulate growth, branching, shaping, and patterning of theureteric bud. Dev Biol 266:285–298

11. Sakurai H, Nigam SK (1997) Transforming growth factor-beta selec-tively inhibits branching morphogenesis but not tubulogenesis. Am JPhysiol 272:F139–146

12. Qiao J, Uzzo R, Obara-Ishihara T, Degenstein L, Fuchs E, HerzlingerD (1999) FGF-7 modulates ureteric bud growth and nephron numberin the developing kidney. Development 126:547–554

13. Bates CM (2007) Role of fibroblast growth factor receptor signalingin kidney development. Pediatr Nephrol 22:343–349

14. Barasch J, Yang J,Ware CB, Taga T, Yoshida K, Erdjument-BromageH, Tempst P, Parravicini E, Malach S, Aranoff T, Oliver JA (1999)Mesenchymal to epithelial conversion in rat metanephros is inducedby LIF. Cell 99:377–386

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

16. Deakin JA, Blaum BS, Gallagher JT, Uhrin D, Lyon M (2009) Thebinding properties of minimal oligosaccharides reveal a commonheparan sulfate/dermatan sulfate-binding site in hepatocyte growthfactor/scatter factor that can accommodate a wide variety of sulfationpatterns. J Biol Chem 284:6311–6321

17. Kreuger J, SpillmannD, Li JP, Lindahl U (2006) Interactions betweenheparan sulfate and proteins: the concept of specificity. J Cell Biol174:323–327

18. ShahMM, Sakurai H, Gallegos TF, SweeneyDE, Bush KT, Esko JD,Nigam SK (2011) Growth factor-dependent branching of the uretericbud is modulated by selective 6-O sulfation of heparan sulfate. DevBiol 356:19–27

19. ShahMM, Sakurai H, Sweeney DE, Gallegos TF, Bush KT, Esko JD,Nigam SK (2010) Hs2st mediated kidney mesenchyme inductionregulates early ureteric bud branching. Dev Biol 339:354–365

20. Sugaya N, Habuchi H, Nagai N, Ashikari-Hada S, Kimata K (2008)6-O-sulfation of heparan sulfate differentially regulates various fibro-blast growth factor-dependent signalings in culture. J Biol Chem 283:10366–10376

21. Qu X, Carbe C, Tao C, Powers A, Lawrence R, van Kuppevelt TH,Cardoso WV, Grobe K, Esko JD, Zhang X (2011) Lacrimal glanddevelopment and Fgf10-Fgfr2b signaling are controlled by 2-O- and6-O-sulfated heparan sulfate. J Biol Chem 286:14435–14444

22. Barasch J, Qiao J, McWilliams G, Chen D, Oliver JA, Herzlinger D(1997) Ureteric bud cells secrete multiple factors, including bFGF,which rescue renal progenitors from apoptosis. Am J Physiol 273:F757–767

23. Sakurai H, Bush KT, Nigam SK (2005) Heregulin induces glial cellline-derived neurotrophic growth factor-independent, non-branchinggrowth and differentiation of ureteric bud epithelia. J Biol Chem 280:42181–42187

24. Esko JD (2001) Special considerations for proteoglycans and glycos-aminoglycans and their purification. Curr ProtocMol Biol 22:17.2.1–17.2.9

25. Esko JD, Selleck SB (2002) Order out of chaos: assembly of ligandbinding sites in heparan sulfate. Annu Rev Biochem 71:435–471

26. Rapraeger AC (2002) Heparan sulfate-growth factor interactions.Methods Cell Biol 69:83–109

27. Grobe K, Ledin J, Ringvall M, Holmborn K, Forsberg E, Esko JD,Kjellen L (2002) Heparan sulfate and development: differential rolesof the N-acetylglucosamine N-deacetylase/N-sulfotransferase iso-zymes. Biochim Biophys Acta 1573:209–215

28. Aikawa J, Grobe K, TsujimotoM, Esko JD (2001)Multiple isozymesof heparan sulfate/heparin GlcNAc N-deacetylase/GlcN N-sulfotransferase. Structure and activity of the fourth member,NDST4. J Biol Chem 276:5876–5882

29. Humphries DE, Lanciotti J, Karlinsky JB (1998) cDNA cloning,genomic organization and chromosomal localization of human hep-aran glucosaminyl N-deacetylase/N-sulphotransferase-2. Biochem J332:303–307

30. Toma L, Berninsone P, Hirschberg CB (1998) The putative heparin-specific N-acetylglucosaminyl N-Deacetylase/N-sulfotransferase alsooccurs in non-heparin-producing cells. J Biol Chem 273:22458–22465

31. Aikawa J, Esko JD (1999) Molecular cloning and expression of athird member of the heparan sulfate/heparin GlcNAc N-deacetylase/N-sulfotransferase family. J Biol Chem 274:2690–2695

32. Hacker U, Nybakken K, Perrimon N (2005) Heparan sulphate pro-teoglycans: the sweet side of development. Nat Rev Mol Cell Biol 6:530–541

33. Shah MM, Sampogna RV, Sakurai H, Bush KT, Nigam SK (2004)Branching morphogenesis and kidney disease. Development 131:1449–1462

34. Shah MM, Tee JB, Meyer T, Meyer-Schwesinger C, Choi Y,Sweeney DE, Gallegos TF, Johkura K, Rosines E, Kouznetsova V,Rose DW, Bush KT, Sakurai H, Nigam SK (2009) The instructiverole of metanephric mesenchyme in ureteric bud patterning,sculpting, and maturation and its potential ability to buffer uretericbud branching defects. Am J Physiol Renal Physiol 297:F1330–1341

35. Steer DL, ShahMM,BushKT, Stuart RO, Sampogna RV,Meyer TN,Schwesinger C, Bai X, Esko JD, Nigam SK (2004) Regulation ofureteric bud branching morphogenesis by sulfated proteoglycans inthe developing kidney. Dev Biol 272:310–327

36. Nigam SK, Shah MM (2009) How does the ureteric bud branch? JAm Soc Nephrol 20:1465–1469

37. Nigam SK (2013) Concise review: can the intrinsic power ofbranching morphogenesis be used for engineering epithelial tissuesand organs? Stem Cells Transl Med 2:993–1000

734 Pediatr Nephrol (2014) 29:727–735

38. Bullock SL, Fletcher JM, Beddington RS, Wilson VA (1998) Renalagenesis in mice homozygous for a gene trap mutation in the geneencoding heparan sulfate 2-sulfotransferase. GenesDev 12:1894–1906

39. Li JP, Gong F, Hagner-McWhirter A, Forsberg E, Abrink M,Kisilevsky R, Zhang X, Lindahl U (2003) Targeted disruption of amurine glucuronyl C5-epimerase gene results in heparan sulfatelacking L-iduronic acid and in neonatal lethality. J Biol Chem 278:28363–28366

40. Pinhal MA, Smith B, Olson S, Aikawa J, Kimata K, Esko JD (2001)Enzyme interactions in heparan sulfate biosynthesis: uronosyl 5-epimerase and 2-O-sulfotransferase interact in vivo. Proc Natl AcadSci USA 98:12984–12989

41. Davies J, Lyon M, Gallagher J, Garrod D (1995) Sulphated proteo-glycan is required for collecting duct growth and branching but notnephron formation during kidney development. Development 121:1507–1517

42. Lelongt B, Makino H, Dalecki TM, Kanwar YS (1988) Role ofproteoglycans in renal development. Dev Biol 128:256–276

43. Platt JL, Brown DM, Granlund K, Oegema TR, Klein DJ (1987)Proteoglycan metabolism associated with mouse metanephric devel-opment: morphologic and biochemical effects of beta-D-xyloside.Dev Biol 123:293–306

44. Platt JL, Trescony P, Lindman B, Oegema TR (1990) Heparin andheparan sulfate delimit nephron formation in fetal metanephric kid-neys. Dev Biol 139:338–348

45. Davies JA, Yates EA, Turnbull JE (2003) Structural determinants ofheparan sulphate modulation of GDNF signalling. Growth Factors21:109–119

46. Izvolsky KI, Zhong L, Wei L, Yu Q, Nugent MA, Cardoso WV(2003) Heparan sulfates expressed in the distal lung are required forFgf10 binding to the epithelium and for airway branching. Am JPhysiol Lung Cell Mol Physiol 285:L838–846

47. Izvolsky KI, Shoykhet D, Yang Y, Yu Q, Nugent MA, Cardoso WV(2003) Heparan sulfate-FGF10 interactions during lung morphogen-esis. Dev Biol 258:185–200

48. Jesudason EC, Connell MG, Fernig DG, Lloyd DA, Losty PD (2000)Heparin and in-vitro experimental lung hypoplasia. Pediatr Surg Int16:247–251

49. Kamimura K, Koyama T, Habuchi H, Ueda R, Masu M, Kimata K,Nakato H (2006) Specific and flexible roles of heparan sulfatemodifications in DrosophilaFGF signaling. J Cell Biol 174:773–778

50. Thompson SM, Connell MG, Fernig DG, Ten Dam GB, vanKuppevelt TH, Turnbull JE, Jesudason EC, Losty PD (2007) Novel'phage display antibodies identify distinct heparan sulfate domains indeveloping mammalian lung. Pediatr Surg Int 23:411–417

51. Thompson SM, Fernig DG, Jesudason EC, Losty PD, van deWesterlo EM, van Kuppevelt TH, Turnbull JE (2009) Heparan

sulfate phage display antibodies identify distinct epitopes with com-plex binding characteristics: insights into protein binding specific-ities. J Biol Chem 284:35621–35631

52. Habuchi H, Tanaka M, Habuchi O, Yoshida K, Suzuki H, Ban K,Kimata K (2000) The occurrence of three isoforms of heparan sulfate6-O-sulfotransferase having different specificities for hexuronic acidadjacent to the targeted N-sulfoglucosamine. J Biol Chem 275:2859–2868

53. Habuchi H, Nagai N, Sugaya N, Atsumi F, Stevens RL, Kimata K(2007) Mice deficient in heparan sulfate 6-O-sulfotransferase-1 ex-hibit defective heparan sulfate biosynthesis, abnormal placentation,and late embryonic lethality. J Biol Chem 282:15578–15588

54. Thompson HA, Spooner BS (1983) Proteoglycan and glycosamino-glycan synthesis in embryonic mouse salivary glands: effects of beta-D-xyloside, an inhibitor of branching morphogenesis. J Cell Biol 96:1443–1450

55. Patel VN, Likar KM, Zisman-Rozen S, Cowherd SN, Lassiter KS,Sher I, Yates EA, Turnbull JE, Ron D, Hoffman MP (2008) Specificheparan sulfate structures modulate FGF10-mediated submandibulargland epithelial morphogenesis and differentiation. J Biol Chem 283:9308–9317

56. Bush KT, Crawford BE, Garner OB, Nigam KB, Esko JD, Nigam SK(2012) N-sulfation of heparan sulfate regulates early branching eventsin the developing mammary gland. J Biol Chem 287:42064–42070

57. Crawford BE, Garner OB, Bishop JR, Zhang DY, Bush KT, NigamSK, Esko JD (2010) Loss of the heparan sulfate sulfotransferase,Ndst1, in mammary epithelial cells selectively blocks lobuloalveolardevelopment in mice. PLoS One 5:e10691

58. Garner OB, Bush KT, Nigam KB, Yamaguchi Y, Xu D, Esko JD,Nigam SK (2011) Stage-dependent regulation of mammary ductalbranching by heparan sulfate and HGF-cMet signaling. Dev Biol355:394–403

59. Sternlicht MD, Kouros-Mehr H, Lu P, Werb Z (2006) Hormonal andlocal control of mammary branching morphogenesis. Differentiation74:365–381

60. Sakurai H, Barros EJ, Tsukamoto T, Barasch J, Nigam SK (1997) Anin vitro tubulogenesis system using cell lines derived from the em-bryonic kidney shows dependence on multiple soluble growth fac-tors. Proc Natl Acad Sci USA 94:6279–6284

61. Santos OF, Barros EJ, Yang XM, Matsumoto K, Nakamura T, ParkM, Nigam SK (1994) Involvement of hepatocyte growth factor inkidney development. Dev Biol 163:525–529

62. Karihaloo A, Karumanchi SA, Barasch J, Jha V, Nickel CH, Yang J,Grisaru S, Bush KT, Nigam S, Rosenblum ND, Sukhatame VP,Cantley LG (2001) Endostatin regulates branching morphogenesisof renal epithelial cells and ureteric bud. Proc Natl Acad Sci USA 98:12509–12514

Pediatr Nephrol (2014) 29:727–735 735