Partial Rescue of Glomerular Laminin 5 Mutations by Wild-Type Endothelia Produce Hybrid Glomeruli
Dale R. Abrahamson*,
Patricia L. St. John*,
Kathryn Isom*,
Barry Robert and
Jeffrey H. Miner
* Department of Anatomy and Cell Biology, University of Kansas Medical Center, Kansas City, Kansas; Comparative Biology Core, Pennington Biomedical Research Center, Baton Rouge, Louisiana; and Renal Division, Department of Internal Medicine, Washington University, St. Louis, Missouri
Correspondence: Dr. Dale R. Abrahamson, Department of Anatomy and Cell Biology, University of Kansas Medical Center, MS 3038, 3901 Rainbow Boulevard, Kansas City, KS 66160. Phone: 913-588-7000; Fax: 913-588-2710; E-mail: dabrahamson{at}kumc.edu
Received for publication February 16, 2007.
Accepted for publication April 9, 2007.
Both endothelial cells and podocytes are sources for laminin1 at the inception of glomerulogenesis and then for laminin5 during glomerular maturation. Why glomerular basement membranes(GBM) undergo laminin transitions is unknown, but this may dictateglomerular morphogenesis. In mice that genetically lack laminin5, laminin 521 is not assembled, vascularized glomeruli failto form, and animals die at midgestation with neural tube closureand placental deficits. It was previously shown that renal corticesof newborn mice contain endothelial progenitors (angioblasts)and that when embryonic day 12 kidneys are transplanted intonewborn kidney, hybrid glomeruli (host-derived endothelium anddonor-derived podocytes) result. Reasoning that host endotheliummay correct the glomerular phenotype that is seen in laminin5 mutants, 5 null embryonic day 12 metanephroi were graftedinto wild-type newborn kidney. Hybrid glomeruli were identifiedin grafts by expression of a host-specific LacZ lineage marker.Labeling of glomerular hybrid GBM with chain-specific antibodiesshowed a markedly stratified distribution of laminins: 5 wasfound only on the inner endothelial half of GBM, whereas 1 locatedto outer layers beneath mutant podocytes. For measurement ofthe contribution of host endothelium to hybrid GBM, immunofluorescentsignals for laminin 5 were quantified: Hybrid GBM containedapproximately 50% the normal 5 complement as wild-type GBM.Electron microscopy of glomerular hybrids showed vascularization,but podocyte foot processes were absent. It was concluded that(1) endothelial and podocyte-derived laminins remain tetheredto their cellular origin, (2) developing endothelial cells contributelarge amounts of GBM laminins, and (3) podocyte foot processdifferentiation may require direct exposure to laminin 5.
Developing glomeruli undergo major transitions in glomerularbasement membrane (GBM) protein composition. Specifically, GBMof the most immature nephrons (comma- and S-shaped) containcollagen 1,2 (IV) and laminin 111 (Ln 111). As glomeruli mature,these proteins disappear and are replaced by collagen 3,4,5(IV)and laminin 521 (Ln 521), and this composition persists in GBMthroughout life.1,2 Although mechanisms for the collagen (IV)and laminin isoform substitutions are not known, these transitionsare believed to be necessary for acquisition of the highly differentiatedstate that is exhibited by glomerular endothelial cells andpodocytes. Indeed, both endothelial cells and podocytes synthesizethe different laminin chains at appropriate stages,3 and thismay be true for some or all of the different collagen (IV) isoformsas well.
The functional importance of GBM protein isoform substitutionshas been examined in several gene deletion studies in mice.For example, when collagen (IV) 3 is deleted, the collagen 345(IV)network fails to form, GBM become multilaminated and mice becomeproteinuric at approximately 4 wk of age, and they begin dyingat 6 to 8 wk with kidney failure.4–6 Indeed, the progressiveglomerular disease seen in these mice closely resembles whatoccurs in many patients with Alport syndrome, which is causedby point mutations in the 3(IV), 4(IV), or 5(IV) collagen genes.7When laminin 2 chain is deleted in mice, the GBM appears ultrastructurallynormal, but widespread podocyte foot process broadening occursand mice die at approximately 3 wk with neuromuscular deficitsand proteinuria.8 Recently, a laminin 2 chain mutation has beenlinked to human Pierson syndrome, which is marked by fatal neuromusculardefects in infancy and proteinuria.9–11 The most dramaticglomerular phenotype in mice occurs in laminin 5 mutants. Targeteddisruption of laminin 5 results in embryonic lethality (embryonicdays 12 through 15 [E12 through 15]) with multiple phenotypes,including incomplete neural tube closure and placental dysmorphogenesis.12In kidney, early steps in nephrogenesis appear normal, includingpodocyte synthesis and then elimination of laminin 1. However,endothelial cells fail to form glomerular capillaries, podocytesdo not mature, and normal GBM never develop.13
We previously showed in embryonic mouse kidneys that endothelialprogenitors (angioblasts) that reside among nephrogenic mesenchymalcells can form the renal microvasculature, including the glomerularendothelium.14 Furthermore, when embryonic kidneys are graftedinto newborn host kidney cortices, angioblasts within the nephrogeniczone of host kidneys migrate into grafts and, together withangioblasts of graft origin, establish the microvasculature.14In engrafted kidneys, this results in glomerular hybrids thatcontain host-derived endothelial cells and graft-derived podocytes.14Here, we sought to determine whether embryonic kidneys derivedfrom laminin 5 knockouts would develop when grafted into kidneysof genetically tagged newborn hosts that were wild-type forlaminin. Specifically, we examined whether host endothelialcells would populate glomeruli within grafts and whether laminin5 would be deposited in GBM of these hybrid glomeruli. We alsoinvestigated the ultrastructure of capillary walls in glomerularhybrids and measured the relative contribution of endothelialcells to GBM laminin.
Distribution of Laminin 5 in Developing Kidney
When cryostat sections from E16 laminin 5 knockout kidneys weredoubly immunolabeled with anti–laminin 5 chain and anti–vascularendothelial growth factor receptor 2 (anti-VEGFR2) IgG, VEGFR2-positiveendothelial cells were seen within a few rudimentary glomerularstructures, but laminin 5 was absent (Figure 1A). Sections fromwild-type siblings, by contrast, contained early capillary loop–stageglomeruli with endothelial cells closely apposed to laminin5-positive GBM (Figure 1B). In addition, Bowman's capsule, alongwith some tubular basement membranes of wild-type kidneys, werepositive for laminin 5 (Figure 1B).
Figure 1. (A) Cryostat section from a laminin 5 knockout embryo double labeled with anti–vascular endothelial growth factor receptor 2 (anti-VEGFR2; to label endothelial cells: green) and anti–laminin 5 (red). A few primitive glomerular capillaries are present (arrows), but laminin 5 is absent. (B) Section from wild-type sibling doubly labeled with anti-VEGFR2 (green) and anti–laminin 5 (red). Note early capillary loop–stage glomerulus containing both endothelial cells as well as laminin 5 in developing glomerular basement membrane (GBM; arrows).
Expression of Laminin 5 in Hybrid Glomeruli
Sections of host kidneys that had been processed for -galactosidasehistochemistry were examined at low magnification to identifythose that contained grafts of laminin 5 null metanephroi. Graftswere distinguished as pale, unstained areas surrounded by deeplystained host tissue as a result of the -galactosidase reactionproduct. These sections were then examined at higher magnificationto determine whether glomeruli within grafts contained host-derivedblue cells, signifying their hybridity (shown diagrammaticallyin Figure 2). When sections bearing hybrid glomeruli were found,the serial sections before and after these slides were thenprocessed for immunofluorescence labeling with anti-lamininantibodies. The following results are representative of sixseparate grafts that contained proven hybrid glomeruli. Figure 3shows two serial pairs from different grafts, each with onesection processed for -galactosidase histochemistry and viewedwith differential interference contrast optics, and the adjacentsection was doubly immunolabeled for laminin 1 and 5 chainsand viewed by epifluorescence. Note that laminin 5 protein waspresent in GBM of hybrid glomeruli but not in Bowman's capsuleor in tubular basement membranes within the grafts.
Figure 2. Diagram illustrating how migration of host-derived angioblasts can result in hybrid glomeruli within grafts. In this case, the host bears the lineage marker, LacZ, which provides the blue label.
Figure 3. Sections showing hybrid glomeruli. Top panels are from separate samples processed for LacZ; bottom panels show corresponding serial sections immunolabeled for laminin. (A and C) Host tissue is intensely blue and can easily be distinguished from graft (dashed black line demarcates margin between host and graft tissue). Note ingress of a number of host-derived cells into graft and the formation of hybrid glomeruli (arrows) that contain host (blue) endothelial cells. (B and D) Immunofluorescence images of serial sections doubly labeled for laminin 1 (green) and 5 (red) chains. Laminin 5 protein is present in GBM of hybrid glomeruli (*same tubule in serial sections).
To examine GBM of hybrid glomeruli more closely, we carriedout confocal microscopy of sections that were dually immunolabeledfor laminin 1 and 5. As shown in Figure 4, laminin 5 was presentin the vascular stalk as well as the GBM. However, laminin 1,which ordinarily is no longer present in capillary loop–stageglomeruli, persisted in GBM of glomerular hybrids (Figure 4,B through D). Furthermore, the hybrid GBM were clearly stratified,with laminin 5 found predominantly in the inner, subendotheliallayer, and laminin 1 occupying the outer, subepithelial layer(Figure 4, C and D). In addition, the amount of laminin 5 thatwas present within a given glomerulus seemed to correlate positivelywith the degree of hybridity. For example, Figure 5A shows asection through a graft that contained one glomerulus with alarge number of host-derived endothelial cells, a second glomeruluswith a few host cells, and a third glomerulus with no host cellsevident. The serial, immunolabeled section shows abundant expressionof laminin 5 in the glomerulus with the largest amount of hostendothelium, weak expression of 5 in the glomerulus populatedby a few host cells, and no 5 apparent in glomeruli that lackedhost cells (Figure 5B). Similarly, the same glomerulus thatcontained the most numerous host endothelial cells also exhibitedthe largest amount of laminin 1 in a GBM pattern (Figure 5,C and D).
Figure 4. Separate confocal images of the same hybrid glomerulus, dually labeled for laminin 1 (green) and laminin 5 (red). Note laminin 5 presence in vascular stalk (VS) as well as GBM. A higher power view of merged image C is shown in D. GBM in glomerular hybrid is stratified; laminin 5 is on endothelial surface, whereas laminin 1 (which ordinarily is not present in capillary loop–stage glomeruli) occupies podocyte surface of GBM. Central areas of signal overlap appear yellow.
Figure 5. Images relating degree of glomerular hybridity with laminin 5 expression. (A) Section processed for LacZ, showing a hybrid glomerulus that contains many host-derived endothelial cells (1), a hybrid with a few host cells (2), and a nonhybrid glomerulus with no host cells (3). (B through D) Confocal images from serial section showing distribution of laminin 5 (B), laminin 1 (C), and merged image (D). Maximal laminin 5 immunolabeling occurs in hybrid glomerulus with most host cells (1).
Ultrastructure of Glomerular Hybrids
To investigate the ultrastructure of glomerular hybrids, wecarried out electron microscopy of three separate grafts thatcontained hybrid glomeruli and neighboring host tissue usingthe Bluo-gal reagent to identify cells that expressed -galactosidase.Glomeruli that developed within ROSA26 host kidneys had -galactosidasereaction product appearing as electron-dense precipitates onplasma membranes of both endothelial cells and podocytes (Figure 6).GBM condensation and podocyte foot process formation, althoughstill incomplete in the host kidney at this age (postnatal day9), appeared entirely normal (Figure 6).
Figure 6. For investigation of the ultrastructure of glomeruli developing within grafts, tissues were developed with Bluo-gal, an alternative substrate for -galactosidase that forms an insoluble, electron-dense precipitate. In glomeruli developing within ROSA26 host tissue, both endothelial cells and podocytes express the transgenic enzyme (black precipitates, arrows). GBM condensation and podocyte foot process formation, although incomplete at this stage of development, appear normal.
Two classes of glomeruli were identified within grafts: Thosethat lacked host-derived endothelium (i.e., nonhybrid, laminin5 knockout glomeruli), and those that contained host-derivedendothelium (glomerular hybrids). Laminin 5 knockout glomeruliwere nonvascularized and unperfused by blood cells (Figure 7A).In addition, their GBM were loosely organized, and podocytefoot processes were absent (Figure 7B). Hybrid glomeruli withhost-derived (Bluo-gal positive) endothelium had GBM that appearedmore condensed and better organized than the nonhybrids (Figure 8,A and B), and capillaries perfused by erythrocytes were readilyidentified (Figure 8B). However, like nonhybrid glomeruli, formationof podocyte foot processes were not observed in glomerular hybrids(Figure 8).
Figure 7. Glomeruli developing within grafts that lack host-derived endothelium (i.e., nonhybrid, laminin 5 knockout glomeruli) do not contain Bluo-gal reaction product and are nonvascularized. GBM are loosely organized (*), and podocyte (Po) foot processes are absent. En, endothelium.
Figure 8. Ultrastructural examination of hybrid glomeruli show host-derived, Bluo-gal–positive endothelial cells (En; arrows) and Bluo-gal–negative podocytes (Po). GBM are better organized than in nonhybrid glomeruli, but podocyte foot processes are absent. RBC, erythrocyte, signifying perfusion.
GBM Laminin Quantification
To estimate the relative contribution of endothelial cells todeveloping GBM, we quantified anti–laminin 5 immunofluorescencesignal strengths across GBM from host glomeruli and comparedthem with those from hybrid glomeruli. Figure 9 (left panel)shows confocal images of a host glomerulus and regions whereGBM intensities were measured (blue line). Histogram plots atbisected regions of the GBM showed peak intensity values foranti–laminin 5 chain (red) that measured approximately3000 luminosity units. Only background levels for anti–laminin1 chain (green) were observed at these same points in normal,host glomeruli (Figure 9). Using this technique, anti–laminin5 intensity values averaged 2889 ± 134 (n = 26) for normal,host GBM. As described previously, a range of intensities foranti–laminin 5 (and anti-1) were observed in hybrid glomeruli,and the intensities seemed to be linked directly with the extentof vascularization by wild-type host endothelial cells. Figure 9(right panel) shows a hybrid glomerular capillary with a relativelylarge amount of laminin 5 (as well as 1) immunoreactivity. Owingto the range of anti–laminin 5 fluorescence intensitiesseen in hybrid GBM, we selected for measurement only hybridglomeruli that contained significant immunolabeling for laminin5. Using this criterion, maximal anti–laminin 5 intensitiesfor hybrid GBM averaged 1675 ± 109 (n = 23), or approximatelyhalf that seen in normal GBM of wild-type host glomeruli.
Figure 9. For evaluation of the relative contribution of endothelial cells to developing GBM, immunofluorescence signal strengths across GBM were quantified from host glomeruli (left) and compared with those from hybrid glomeruli (right) in adjacent areas that contained metanephric grafts. Histogram plots show peak intensity values for laminin 5 chain (red) at bisected regions of the GBM. Only background levels for laminin 1 chain (green) are observed at these same points in normal glomeruli (left), whereas hybrids show abnormally high levels of 1 in outer layer of GBM (right).
Four new observations stem from these experiments. First, whenlaminin 5 mutant metanephroi were grafted into ROSA26 transgenichosts, numerous hybrid glomeruli within grafts that containedhost-derived endothelial cells were identified and were perfusedwith blood. These glomerular hybrids thereby indicate at leastpartial correction of the aborted glomerulogenesis seen in laminin5 null kidneys, which is the most prominent kidney phenotypefor this mutation. Second, most hybrid glomeruli contained distinctlystratified GBM with the unusual composition of both laminin5 and 1. Laminin 5 derived from host endothelial cells distributedto the subendothelial GBM surface, whereas laminin 1 occupiedonly the subepithelial GBM layer beneath mutant podocytes. Third,despite the formation of perfused glomerular capillaries, podocytefoot process formation failed to occur in hybrid glomeruli.This may have been due to the unusual laminin composition inhybrid GBM and specifically the retention of laminin 1 and/orthe absence of laminin 5 in the subepithelial layer. Finally,endothelial cells in hybrid glomeruli contributed importantlyto the GBM, at least with respect to laminin 5. Perhaps thesecells are also sources of significant GBM protein in glomerularfibrotic disorders.
Initial studies on laminin 5 null kidneys showed that endothelialcells fail to adhere within developing glomeruli and true capillaryloops never form.13 In addition, the podocytes in these mutantsare disordered and cluster as a multilayer of cells in the defectiveglomeruli. The normal disappearance of laminin 1 still occurs,but because of an absence of laminin 5, an intact basement membranestructure is not maintained in laminin 5 mutants.13 Why endothelialcells are excluded from laminin 5 null glomeruli is unclear,but this may be due to inadequate adhesive properties and thefailure of a definitive GBM to form. Although laminin 4 is present,this particular chain is truncated and may not be able to createa stable laminin network by itself in the absence of full-lengthlaminin chains.13 As reported here, hybrid glomeruli achievedthrough grafting laminin 5 null metanephroi into laminin wild-typehosts clearly contained host-derived endothelial cells and wereblood-perfused. These findings provide further evidence indicatingthat the initial signals for glomerular vascularization by endothelialcells are operative in laminin 5 mutants. In addition, onlyhybrid glomeruli within grafts contained laminin 5 in GBM; glomerulithat were not vascularized by host-derived endothelial cellswere entirely negative for 5.
One of the more striking findings from our studies was thatGBM in glomerular hybrids displayed a distinctly stratifieddistribution pattern for laminins. Specifically, laminin 5 wasfound on the inner, endothelial layer of the GBM. This evidenceconfirms previous findings in normal mice showing an endothelial(as well as podocyte) origin of laminin 5.3 Because laminin1 is also expressed by endothelial cells of the most immatureglomeruli,3 our results also suggest that the laminin 1:5 transitionoccurred normally in the host-derived endothelial cells thatoccupied hybrid glomeruli. Quite surprising, however, and unlikethe previous findings in laminin 5 mutant embryos,13 our studiesof glomerular hybrids showed the retention of seemingly largeamounts of laminin 1 in hybrid GBM. Like laminin 5, the 1 distributionwas distinctly stratified but in this case was found exclusivelyin the outer, subepithelial layer of the GBM beneath the mutantpodocytes. These results indicate that the normal process fordownregulation of laminin 1 synthesis and/or its removal fromthe GBM of hybrid glomeruli had not transpired. However, becausethe aberrant presence of laminin 1 occurred most abundantlyin glomerular hybrids, we speculate that the presence of wild-typeendothelium (and/or the presence of a vascular supply) enhancedmutant podocyte survival and promoted their synthesis and depositionof laminin 1.
Another intriguing finding regarding the stratified GBM in hybridglomeruli was that the laminin that originated from the endothelialcells did not extend across the full GBM width and contact podocytes.Similarly, the laminin 1 that was derived from mutant podocytesdid not project across the GBM and contact the endothelium.These results suggest that laminins, once secreted, remain closelyassociated with their respective cell of origin, presumablytethered to the integrins, dystroglycan, and other basementmembrane–binding proteins on the basal surfaces of glomerularendothelial cells and podocytes, respectively. Furthermore,because we found laminin 5 only in serial sections that cameimmediately before or after sections that contained proven hybridglomeruli, the secreted laminin apparently did not diffuse largedistances from its cell of origin in the plane of the GBM. Again,this may reflect close association of GBM laminin with plasmalemmalbasement membrane receptors. An important caveat is that theGBM assembled in these glomerular hybrids created through graftingmay not accurately represent what occurs during normal glomerulardevelopment. As discussed previously, the GBM originates initiallyby the fusion of a dual basement membrane structure assembledjointly by endothelial cells and podocytes.3,15 Perhaps thisfusion was incomplete (or failed entirely) in hybrid glomeruli,possibly because of the atypical laminin composition in thepodocyte layer. Alternatively, basement membrane proteins otherthan laminin may behave differently in hybrid GBM, and we arecurrently using this same grafting technology with differentbasement membrane mutants to address this possibility. In addition,new experiments that make use of conditional or cell-specificexpression of reporter-tagged basement membrane proteins mayhelp to clarify this issue further. Nevertheless, our findingsclearly suggest that the GBM may indeed contain layers or strataof compositionally distinct matrix. If true, then delaminatingdisorders such as Alport disease may involve events that disruptadhesive interactions between these GBM strata.
Although endothelial cells were recruited successfully to formhybrid glomeruli in grafts of laminin 5 mutant metanephroi,the GBM and podocytes both were abnormal. Electron microscopyof glomerular hybrids showed lengths of uncondensed GBM anda complete absence of foot process formation by podocytes. Asmentioned previously, laminin 1 is present in the vascular cleftregion of comma- and S-shaped nephrons. The visceral epithelialcells, which are the forerunners of podocytes, are cuboidalor columnar in shape at this stage, and foot processes havenot yet formed.15 As glomeruli progress into the early capillaryloop stage, however, laminin 1 is eliminated and quickly replacedby laminin 5.1–3 The molecular and cellular mechanismsresponsible for this laminin transitioning are unknown. Nevertheless,coincident with laminin isoform substitution, podocytes begindeveloping their characteristic interdigitating foot processes,and this continues into maturing glomeruli stages, when podocytedifferentiation concludes. Perhaps the abnormally prolongedpresence of laminin 1 in hybrid GBM blocked the maturation ofpodocytes and inhibited foot process formation. Along theselines, we previously showed in Alport mice that there is ectopicexpression of laminin 1 in the subepithelial GBM outpockets,which is precisely where podocyte foot processes are also broadlyeffaced.16
Instead of inhibition of foot process formation by the prolongedpresence of laminin 1 in hybrid GBM, perhaps process formationfailed because of an absence of laminin 5 in the immediate,subjacent matrix beneath podocytes. In other words, laminin5 may normally provide an inductive cue for foot process formation,which was lacking in hybrid GBM. As indicated previously, laminin5 deletion results in nonvascularized glomerular epithelialtufts that lack GBM.13 In addition to kidney, laminin 5 waspreviously shown to be required for normal organogenesis inseveral other systems, including vascularization of the placenta,neural tube closure, and septation of the digits.12 In addition,laminin 5 deletion results in defective cranial sensory andtrunk sympathetic ganglia and abnormal neural crest cell migration.17In lung, laminin 5 mutants exhibit abnormal lobar septation,absence of alveolar type I and reduction in alveolar type IIcells, and diminished VEGF production.18,19 In small intestine,laminin 5 mutation causes excessive folding of intestinal loopsand decreased expression of desmin and actin in smooth musclecell layers.20 Laminin 5 null mice also demonstrate defectivetooth germ and dental cusp formation and reduced proliferationof dental epithelium with significant decreases of Shh and Fgfrin this cell layer.21
Although mechanisms that account for the multiple phenotypesin laminin 5 mutants are not fully understood, at least someof the effects may be due to an absence of the 5, C-terminallaminin type globular (LG) domain. This domain consists of fivehomologous structural subunits (LG1 through 5)22 and mediatesbinding to integrins, dystroglycan, Lutheran, and/or other lamininreceptors on a variety of cells.23,24 For examination of LGfunction in detail, transgenic mice in which all or part ofthe 5 LG domain was replaced with corresponding regions of thehuman 1 LG domain were created.25 These laminin variants werethen mated onto the laminin 5 knockout background. Like laminin5 mutants, mice that express the entire 1 LG domain in placeof native 5 die before birth. However, their glomerular capillarywalls develop normally and contain endothelial cells, intactGBM, and differentiated podocytes with foot processes.25 However,glomerular capillary diameters are unusually large, apparentlybecause the intercapillary mesangial cells fail to adhere properlyto a GBM that lacks the 5 LG domain.25 This issue was investigatedfurther by creating mutants that expressed 1 LG3 through 5 inplace of 5 LG3 through 5.26 In this case, glomeruli developnormally initially and mice survive for several months. Theyeventually die of nephrotic syndrome, however, with greatlythickened GBM, particularly in the subendothelial layer, indicatingthat the 5 LG 3 through 5 domain is necessary for maintenanceof normal glomerular structure and function.26 Perhaps futurestudies on glomerular hybrids that express various laminin domaincombinations will shed additional light on specific regionsof the laminin molecule that are important for normal glomerulogenesis.
Previously, we showed that both glomerular endothelial cellsand podocytes synthesize laminins, but until now, it has notbeen possible to measure the relative contribution of each celllayer, respectively, to the GBM. Through the creation of glomerularhybrids after grafting, we were able to document for the firsttime that endothelial cells contribute significant amounts oflaminin 5 to the GBM. Indeed, our immunofluorescence quantificationof hybrids showed that up to approximately half of the fullGBM complement of laminin 5 could be derived from the endothelium.Whether the same is true for normal, wild-type glomeruli isnot yet known, however. In addition, whether endothelial cellsparticipate heavily in synthesis of other important constituentsof the GBM, such as collagen type IV, is unclear. Nevertheless,our findings in developing glomeruli raise the possibility thatglomerular endothelial cells of mature kidney may also be asource of significant amounts of extracellular matrix in fibroticdisorders, and we are testing this hypothesis. If this provesto be true, then therapies that specifically inhibit glomerularendothelial matrix deposition and could effectively slow progressiveglomerulosclerosis might be developed.
Animals and Antibodies
Laminin 5 mutant mice were created and genotyped as describedpreviously.12,13 ROSA26 transgenic mice, which express the bacterialtransgene LacZ ubiquitously but are otherwise normal,27 wereobtained from The Jackson Laboratory [Bar Harbor, ME; strainB6;129S-Gt(ROSA)26Sor/J]. All procedures with animals were approvedby the Institutional Animal Care and Use Committee to ensurecompliance with the Health Research Extension Act, Animal WelfareAct, and the Public Health Service Policy on Humane Care andUse of Laboratory Animals.
Monoclonal rat anti-mouse laminin 1 chain (mAb 8B3) and rabbitanti-mouse laminin 5 chain antibodies were prepared and characterizedas described previously.28,29 Anti–rat fluorescein andanti–rabbit rhodamine IgG were obtained from MP Biomedicals(Irvine, CA). Anti-VEGFR2 was purchased from BD-PharMingen (SanDiego, CA).
Metanephric Grafts and Tissue Processing
For creation of hybrid glomeruli, metanephroi from E12 laminin5 null mice were grafted into kidney cortices of 1- to 2-d-oldROSA26 hosts using procedures that were previously developed.14In most cases, kidneys bearing grafts were harvested 6 to 13d after implantation and fixed overnight in 0.2% paraformaldehydein PBS that contained 20 mM magnesium chloride (pH 7.3). Afterequilibration in 30% sucrose, tissues were embedded in OCT andrapidly frozen in isopentane chilled in a dry ice/acetone bath.Kidneys were then serially sectioned in a cryostat, and alternateserial sections, 8 µm thick, were developed for -galactosidasehistochemistry as described previously.14,30 The nondevelopedsections were refixed with 4% paraformaldehyde for 10 min, incubatedwith 0.1 M glycine in PBS for 10 min, treated with 0.1% SDSin PBS for 1 h at 50°C, and stored at –80°C forpossible immunolabeling later. For double immunolabeling, slideswere brought to room temperature and incubated with rat anti-laminin1 IgG (50 µg/ml) or with anti-VEGFR2 (50 µg/ml)and rabbit anti-laminin 5 antisera (1:100). After washing inPBS and labeling with anti–rat fluorescein and anti–rabbitrhodamine, sections were coverslipped with ProLong Gold antifadereagent (Molecular Probes, Eugene, OR).
Microscopy
Sections were viewed by standard epifluorescence and differentialinterference contrast with a Leica DM5000B microscope (Bannockburn,IL). Slides were also examined with a Zeiss LSM 510 scanninglaser confocal microscope (Thornwood, NY), and images were capturedat an optical section thickness of 0.2 µm. GBM fluorescenceintensities were measured on slides that were processed on thesame day and viewed under identical conditions using the quantitativeco-localization tool in the LSM 510 software package (Zeiss).
For investigation of the ultrastructure of glomeruli developingwithin grafts, kidney tissues were fixed for 2 h on ice using2% paraformaldehyde and 0.4% glutaraldehyde in PBS plus 2 mMmagnesium chloride. Fixed tissues were then washed with bufferand sectioned to a thickness of 200 µm using a Vibratome(St. Louis, MO). Sections were then developed with Bluo-gal,which, in the presence of the LacZ gene product, -galactosidase,yields an insoluble, electron-dense precipitate.31 Specifically,sections were developed in 20 mM potassium ferrocyanide, 20mM potassium ferricyanide, 2 mM magnesium chloride, and 2 mMBluo-gal (Invitrogen, Carlsbad, CA) in 20 mM Tris (pH 7.3) overnightat 37°C with gentle mixing in a rotator. After washing withbuffer, samples were refixed with Karnovsky's fixative for 30min, washed three times with 0.1 M sodium cacodylate plus 3.5%sucrose (pH 7.3), and then postfixed for 1.5 h with Palade'sosmium tetroxide on ice. Samples were routinely dehydrated throughethanol and propylene oxide and embedded in Poly/Bed 812 (Polysciences,Warrington, PA).
Funding for this study came from National Institutes of Healthgrants DK052483 and DK065123. Confocal images were acquiredat the KUMC Confocal Imaging Facility, supported in part bythe Kansas IDeA Network of Biomedical Research Excellence (RR016475).
Portions of this work were presented previously in abstractform (J Am Soc Nephrol 13: 101A, 2002, and J Am Soc Nephrol16: 427A–428A, 2005).
We thank Dr. Elizabeth Petroske and Eileen Roach for technicalassistance.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
See the related editorial, "Chimerism of the Renal GlomerulusRevisited," on pages 2215–2217.
Miner JH: Building the glomerulus: A matricentric view.
J Am Soc Nephrol 16
: 857
–861, 2005[Free Full Text]
St. John PL, Abrahamson DR: Glomerular endothelial cells and podocytes jointly synthesize laminin-1 and -11 chains.
Kidney Int 60
: 1037
–1046, 2001[CrossRef][Medline]
Miner JH, Sanes JR: Molecular and functional defects in kidneys of mice lacking collagen alpha3(IV): Implications for Alport syndrome.
J Cell Biol 135
: 1403
–1413, 1996[Abstract/Free Full Text]
Cosgrove D, Meehan DT, Grunkemeyer JA, Kornak JM, Sayers R, Hunter WJ, Samuelson GC: Collagen COL4A3 knockout: A mouse model for autosomal Alport syndrome.
Genes Dev 10
: 2981
–2992, 1996[Abstract/Free Full Text]
Lu W, Phillips CL, Killen PD, Hlaing T, Harrison WR, Elder FF, Miner JH, Overbeek PA, Meisler LH: Insertional mutation of the collagen genes Col4a3 and Col4a4 in a mouse model of Alport syndrome.
Genomics 61
: 113
–124, 1999[CrossRef][Medline]
Hudson BG: The molecular basis of Goodpasture and Alport syndromes: Beacons for the discovery of the collagen IV family.
J Am Soc Nephrol 15
: 2514
–2527, 2004[Free Full Text]
Noakes PG, Miner JH, Gautam M, Cunningham JM, Sanes JR, Merlie JP: The renal glomerulus of mice lacking s-laminin/laminin beta2: Nephrosis despite molecular compensation by laminin beta1.
Nat Genet 10
: 400
–406, 1995[CrossRef][Medline]
Zenker M, Aigner T, Wendler O, Tralau T, Muntefering H, Fenski R, Pitz S, Schumacher V, Royer-Pokora B, Wuhl E, Cochat P, Bouvier R, Kraus C, Mark K, Madlon H, Dotsch J, Reacher W, Maruniak-Chudek I, Lennert T, Neumann LM, Reis A: Human laminin beta 2 deficiency causes congenital nephrosis with mesangial sclerosis and distinct eye abnormalities.
Hum Mol Genet 13
: 2625
–2632, 2004[Abstract/Free Full Text]
Zenker M, Pierson M, Jonveaux P, Reis A: Demonstration of two novel LAMB2 mutations in the original Pierson syndrome family reported 42 years ago.
Am J Med Genet A 138
: 73
–74, 2005[Medline]
Jarad G, Cunningham J, Shaw AS, Miner JH: Proteinuria precedes podocyte abnormalities in Lamb2–/– mice, implicating the glomerular basement membrane as an albumin barrier.
J Clin Invest 116
: 2272
–2279, 2006[CrossRef][Medline]
Miner JH, Cunningham J, Sanes JR: Roles for laminin in embryogenesis: Exencephaly, syndactyly, and placentopathy in mice lacking the laminin alpha5 chain.
J Cell Biol 143
: 1713
–1723, 1998[Abstract/Free Full Text]
Miner JH, Li C: Defective glomerulogenesis in the absence of laminin alpha5 demonstrates a developmental role for the kidney glomerular basement membrane.
Dev Biol 217
: 278
–289, 2000[CrossRef][Medline]
Robert B, St. John PL, Abrahamson DR: Evidence that embryonic kidney cells expressing flk1–1 are intrinsic, vasculogenic angioblasts.
Am J Physiol 271
: F744
–F753, 1996[Medline]
Abrahamson DR: Structure and development of the glomerular capillary wall and basement membrane.
Am J Physiol 253
: F783
–F794, 1987[Medline]
Abrahamson DR, Prettyman AC, Robert B, St. John PL: Laminin-1 reexpression in Alport mouse glomerular basement membranes.
Kidney Int 63
: 826
–834, 2003[CrossRef][Medline]
Coles EG, Gammill LS, Miner JH, Bronner-Fraser M: Abnormalities in neural crest cell migration in laminin alpha5 mutant mice.
Dev Biol 289
: 218
–228, 2006[CrossRef][Medline]
Nguyen NM, Miner JH, Pierce RA, Senior RM: Laminin alpha5 is required for lobar septation and visceral pleural basement membrane formation in the developing mouse lung.
Dev Biol 246
: 231
–244, 2002[CrossRef][Medline]
Nguyen NM, Kelley DG, Schlueter JA, Meyer MJ, Senior RM, Miner JH: Epithelial laminin alpha5 is necessary for distal epithelial cell maturation, VEGF production, and alveolization in the developing murine lung.
Dev Biol 282
: 111
–125, 2005[CrossRef][Medline]
Bolcato-Bellemin AL, Lefebvre O, Arnold C, Sorokin L, Miner JH, Kedinger M, Simon-Assman P: Laminin alpha5 chain is required for intestinal smooth muscle development.
Dev Biol 260
: 376
–390, 2003[CrossRef][Medline]
Fukumoto S, Miner JH, Ida H, Fukumoto E, Yuasa K, Miyazaki H, Hoffman MP, Yamada Y: Laminin alpha5 is required for dental epithelium growth and polarity and the development of tooth bud and shape.
J Biol Chem 281
: 5008
–5016, 2006[Abstract/Free Full Text]
Timpl R, Tisi D, Talts FJ, Andac Z, Sasaki T, Hohenester E: Structure and function of laminin LG modules.
Matrix Biol 19
: 309
–317, 2000[CrossRef][Medline]
Talts JF, Timpl R: Mutation of the basic sequence in the laminin alpha2LG3 module leads to a lack of proteolytic processing and has different effects on beta1 integrin-mediated cell adhesion and alpha-dystroglycan binding.
FEBS Lett 458
: 319
–323, 1999[CrossRef][Medline]
Kikkawa Y, Moulson CL, Virtanen I, Miner JH: Identification of the binding site for the Lutheran blood group glycoprotein on laminin alpha5 through expression of chimeric laminin alpha chains in vivo.
J Biol Chem 277
: 44864
–44869, 2002[Abstract/Free Full Text]
Kikkawa Y, Virtanen I, Miner JH: Mesangial cells organize the glomerular capillaries by adhering to the G domain of laminin alpha5 in the glomerular basement membrane.
J Cell Biol 161
: 187
–196, 2003[Abstract/Free Full Text]
Kikkawa Y, Miner JH: Molecular dissection of laminin alpha5 in vivo reveals separable domain-specific roles in embryonic development and kidney function.
Dev Biol 296
: 265
–277, 2006[CrossRef][Medline]
Friedrich G, Soriano P: Promoter traps in embryonic stem cells: A genetic screen to identify and mutate developmental genes in mice.
Genes Dev 5
: 1513
–1523, 1991[Abstract/Free Full Text]
Abrahamson DR, Irwin MH, St John PL, Perry EW, Accavitti MA, Heck LW, Couchman JR: Selective immunoreactivities of kidney basement membranes to monoclonal antibodies against laminin: Localization of the ends of the long arm and short arms to discrete microdomains.
J Cell Biol 109
: 3477
–3491, 1989[Abstract/Free Full Text]
Miner JH, Patton BL, Lentz SI, Gilbert DJ, Snider WD, Jenkins NA, Copeland NG, Sanes JR: The laminin alpha chains: Expression, developmental transitions, and chromosomal locations of alpha1–5, identification of heterotrimeric laminins 8–11, and cloning of a novel alpha3 isoform.
J Cell Biol 137
: 685
–701, 1997[Abstract/Free Full Text]
Robert B, St John PL, Abrahamson DR: Direct visualization of renal vascular morphogenesis in Flk1 heterozygous mutant mice.
Am J Physiol 275
: F164
–F172, 1998[Medline]
Weis J, Fine SM, David C, Savarirayan S, Sanes JR: Integration site-dependent expression of a transgene reveals specialized features of cells associated with neuromuscular junctions.
J Cell Biol 113
: 1385
–1397, 1991[Abstract/Free Full Text]
Related Articles
This Month's Highlights
J. Am. Soc. Nephrol. 2007 18: A13.
[Full Text][PDF]
Chimerism of the Renal Glomerulus Revisited
Yashpal S. Kanwar, Farhad R. Danesh, and Sumant S. Chugh
J. Am. Soc. Nephrol. 2007 18: 2215-2217.
[Full Text][PDF]
This article has been cited by other articles:
Y. S. Kanwar, F. R. Danesh, and S. S. Chugh Chimerism of the Renal Glomerulus Revisited
J. Am. Soc. Nephrol.,
August 1, 2007;
18(8):
2215 - 2217.
[Full Text][PDF]
5 Mutations by Wild-Type Endothelia Produce Hybrid Glomeruli
Top
Abstract
Introduction
1 at the inception of glomerulogenesis and then for laminin 
2
1 is not assembled, vascularized glomeruli fail to form, and animals die at midgestation with neural tube closure and placental deficits. It was previously shown that renal cortices of newborn mice contain endothelial progenitors (angioblasts) and that when embryonic day 12 kidneys are transplanted into newborn kidney, hybrid glomeruli (host-derived endothelium and donor-derived podocytes) result. Reasoning that host endothelium may correct the glomerular phenotype that is seen in laminin 
