Glomerular morphogenesis is marked by the progressive appearance and disappearance of different type IV collagen and laminin isotypes from developing glomerular basement membrane (GBM) (reviewed in1). Specifically, vascular cleft basement membranes of the earliest, comma- and S-shaped nephric figures contain 
1(IV) and 2(IV) collagen and laminin-1 (1
1
1 heterotrimers). In intermediate, capillary loop stage glomeruli, GBMs contain 1(IV), 2(IV), 3(IV), 4(IV), and 5(IV) collagen chains, and laminin-11 (521) makes its appearance. Maturing-stage glomeruli GBMs contain predominantly 3(IV), 4(IV), 5(IV) collagen and laminin-11. Laminin-1 is apparently completely absent from mature glomeruli in humans, and in mice this isotype, along with laminin-2 (211), is confined to mesangial matrices1.
Although the reasons for these temporally controlled isoform substitutions in the GBM are not known, striking glomerular phenotypes are observed when collagen type IV or laminin isoforms are mutated. Humans with mutations in the 3, 4, or 5(IV) chains suffer from Alport syndrome, which in kidney is characterized by an absence of the 3(IV), 4(IV), 5(IV) collagen network, the protracted presence of 1(IV) and 2(IV) collagen chains in the GBM, a progressive delamination and thickening of the GBM, proteinuria, and renal insufficiency (reviewed in2). At least some of these changes are believed to be due to the susceptibility of 1(IV) and 2(IV) collagen chains to proteolytic degradation, a process that occurs much less readily in the cysteine-enriched 3(IV), 4(IV), and 5(IV) collagen network normally found in mature glomeruli3. Autosomal-recessive and X-linked models in dogs display very similar histopathologic features to human Alport syndrome4,5,6. Three different mouse models of Alport disease have been created through the deletion of the noncollagenous 1 (NC1) domain of 3(IV) collagen7,8 or an insertional mutation knocking out both the 3(IV) and 4(IV) collagen genes9. In all three cases, mice develop renal phenotypes that closely parallel those seen in humans, indicating that the formation and maintenance of an 3(IV), 4(IV), 5(IV) collagen network is crucial for normal GBM structure and function. Deletions of the laminin 5 or 2 chains (both found in laminin-11) in mice also result in prominent glomerular defects, including nonvascular glomeruli (5)10, and massive proteinuria accompanied by podocyte foot process effacement (2)11.
A number of studies have examined GBM structure and composition in human Alport patients and in canine and mouse models (reviewed in12). Although early reports did not detect major errors in glomerular laminin isoform distribution, a comprehensive immunofluorescence study using well-characterized, chain-specific antibodies was published recently showing potentially important changes in GBM laminin composition12. This analysis revealed that Alport GBMs abnormally contain laminin 2 chain [probably as part of the laminin-2 and/or -4 isotype (211 and 221, respectively)] and, in both mice and dogs, laminin 1 and 1 chains, probably representing laminin-112. In the study presented here, we concentrated on the temporal expression pattern and distribution of laminin 1 and 1 in the immature Alport mouse. We were interested in whether the laminin isoform switching process was operative in developing Alport kidney and specifically if laminin-1 was expressed continuously from the vascular cleft stages of early glomeruli or reexpressed at a later stage during the disease process. We also used immunoelectron microscopy to identify the ultrastructural distribution and cellular origins of these aberrant laminin chains in the GBM. Our findings show that laminin-1 down-regulation occurred normally in the Alport mouse but that derepression occurred shortly thereafter. Further, we show that laminin-1 originated in the diseased glomerulus from both endothelial cells and podocytes.
METHODS
Animals
Mice containing a targeted deletion of the NC1 domain of the 3(IV) collagen chain8 were obtained from Jackson Laboratory (Jax mice strain 129-Col4a3tm1Dec, Bar Harbor, ME, USA). Our colony was maintained by heterozygote interbreeding, and progeny were genotyped by polymerase chain reaction (PCR). Heterozygous and wild-type littermates served as controls for observations made in 3(IV) collagen null mice.
Antibodies
Monoclonal rat antimouse laminin 1 and 1 chain IgGs (designated 8B3 and 5A2, respectively) were prepared and characterized as previously described13,14. For some experiments, these immunoglogulins (IgGs) were directly conjugated to activated horseradish peroxidase (HRP)13. Polyclonal antibody against laminin 515 chain was kindly provided by Dr. Jeffrey Miner (Washington University, St. Louis, MO, USA). Anticollagen 3(IV) antibody (MAB3) was purchased from Wieslab, AB (Lund, Sweden). Fluorescein and rhodamine-conjugated secondary antibodies came from ICN Biomedicals (Costa Mesa, CA, USA).
Immunolocalization
Deeply anesthetized mice underwent nephrectomy. For immunofluorescence microscopy, kidneys were frozen in isopentane chilled in a dry ice–acetone bath. For immunofluorescence labeling with chain-specific antibodies against laminin 1, 5, and 1 chains, as well as the 3 chain of collagen type IV, cryostat sections were fixed for 10 minutes in 100% methanol at -20°C. After labeling with primary antibodies, sections were washed with phosphate-buffered saline (PBS), incubated with appropriate fluorochrome-conjugated secondary antibodies, and permanently coverslipped with Prolong mounting media (Molecular Probes, Inc., Eugene, OR, USA).
For immunoelectron microscopy, anesthetized mice received intravenous injections of 0.1 mL 1.0 mg/mL antilaminin 1 chain monoclonal antibody (mAb) 8B3-HRP or antilaminin 1 chain mAb 5A2-HRP and allowed to recover. The next day, mice were reanesthetized, and kidneys were fixed in situ by the subcapsular injection of Karnovsky's fixative. Vibratome sections, 50 
m thick, were then processed for peroxidase histochemistry as before16, and postfixed with 2% osmium tetroxide. In other cases, postfixation immunolabeling of kidney tissue from uninjected animals was processed exactly as described earlier to identify cells engaged in laminin biosynthesis17. Tissues were embedded in Polybed 812 (Polysciences, Warrington, PA, USA) and ultrathin sections were stained for 2 minutes with Reynold's lead citrate.
RESULTS
Dysregulation of the laminin isoform switch in Alport kidneys
As indicated earlier, abnormal laminin isoforms have been documented in human Alport disease, as well as in dog and mouse models12. Specifically, laminin-1 (111), which normally occurs within nascent GBM found in vascular clefts of early nephric figures, is ordinarily replaced by laminin-11 (5 21) in mature GBM. However, in mouse and dog Alport models, laminin-1 and -11 chains are both observed in diseased GBMs12. To evaluate whether the abnormal presence of laminin-1 in mouse Alport GBM reflects its persistent expression throughout glomerular development, we examined kidneys from infant 3(IV) collagen knockouts and compared them with wild-type and heterozygous littermates. In wild type and heterozygous mice, laminin 1 chain was observed in vascular cleft basement membranes of the earliest nephric figures, and almost entirely disappeared from GBMs of capillary loop stage glomeruli Figure 1a. In maturing stage glomeruli from the same kidney section, laminin 1 chain was seen in mesangial matrices but was absent from peripheral loop GBMs Figure 1b. Similar to wild-type and heterozygous mice, laminin 1 was expressed in vascular cleft basement membranes of 3(IV) collagen knockout mice and quickly disappeared in capillary loop stage GBMs Figure 1c. In maturing stage GBMs of knockouts, however, laminin 1 chain reappeared and was clearly observed within GBMs of several peripheral capillary loops Figures 1d and Figure 1e. This evidence indicates, therefore, that laminin 1 down-regulation proceeded normally in early capillary loop stage glomeruli in the knockouts, but that synthesis of this laminin chain then abnormally resumed in maturing-stage glomeruli.
Figure 1.
Immunofluorescence micrographs showing the distribution of laminin 
1 chain in 3-day-old wild-type (WT) mice and 3(IV) collagen knockout (KO) littermates. (a) In WT mice, laminin 1 chain is observed in vascular cleft basement membrane (arrow) but has disappeared from capillary loop stage glomerular basement membrane (GBM) (*). (b) In maturing stage glomeruli, laminin 1 chain is restricted to mesangial matrices; peripheral loop GBMs are negative. (c) Similar to WT mice, laminin 1 chain is present in vascular clefts of KO mice (arrows), but quickly disappears in capillary loop stage glomeruli (*). (d and e) Maturing stage GBMs of KOs contain laminin 1 chain in peripheral loop GBMs (arrows) as well as mesangial matrices.
Immunolocalization of laminin 1 and 
1 chains within Alport mouse GBMs
Glomerular morphogenesis concludes by 2 weeks of age in the mouse. Figure 2 shows the glomerular distribution of collagen 3(IV), and laminin 1 and 1 chains of laminin-1 in 2-week-old wild-type Figure 2a, c, and e and 3(IV) collagen knockouts Figure 2b, d, and f. At this age, laminin-1 is normally confined in glomeruli to the mesangial matrices, whereas laminin-11 is found in GBMs. As shown here, laminin 1 and 1 chains were confined to mesangial areas in wild type glomeruli Figure 2 c and e, but these chains were also present in GBMs as well as mesangia of Alport mice at this age Figures 2 d and f. To determine the ultrastructural location of laminin 1 and 1 chains, we carried out immunoelectron microscopy on 4-week-old Alport mice and hetereozygous and wild type littermates as controls. Consistent with the immunofluorescence findings shown in Figures 2c and 2e, laminin 1 and 1 chains were found exclusively in mesangial matrices of mature glomeruli at this age and were absent from GBMs in heterozygous and wild-type mice. Figure 3 shows localization for laminin 1 chain; an identical localization pattern was seen for laminin 1 chain in wild-type and heterozygous mice (not shown). In 3(IV) collagen knockout mouse GBMs, by contrast, antilaminin 1 chain antibody bound abundantly to many of the irregular, supepithelial GBM out pockets commonly seen Figure 4. Not all of these GBM projections contained the laminin 1 chain, however Figure 4. On the other hand, laminin 1 chain was detected in essentially all of the GBM irregularities of Alport mice Figure 5, although there were variations in peroxidase reaction product intensity between separate out pockets. We suspect that labeled lengths of GBM that were negative for laminin 1 chain but positive for 1 chain probably contained laminin 2 chain [constituting laminin-2 (1, 12)], but we did not pursue this possibility further. Remarkably, despite the complete absence of 3(IV) collagen, areas within many glomerular capillary loops appeared to be ultrastructurally normal Figure 5. Importantly, however, regions of the glomerular capillary containing lengths of GBM with normal ultrastructure covered by typical, regularly interdigitating foot processes were not labeled with either antilaminin 1 or 1 antibodies. In other words, laminin-1 chains were observed only in those areas of peripheral capillary loops containing GBM irregularities and podocyte foot process effacement.
Figure 2.
Distribution of 3(IV) collagen, and laminin 1 and 1 chains in glomeruli of 2-week-old wild-type (WT) and 3(IV) collagen knockout (KO) littermates. Collagen 3(IV) is strongly positive in glomerular basement membranes (GBMs) of WT (a), but absent in KOs (b). Laminin 1 and 1 chains are prominent in mesangial matrices of WT mice (c) and (e) but also seen in GBMs of KOs [arrows in (d) and (f)].
Figure 3.
Electron micrograph from a 4-week-old heterozygous (Het) mouse showing normal distribution of laminin 1 chain [using monoclonal antibody (mAb) 5A2-horseradish peroxide (HRP) histochemistry as described in the Methods section]. Peroxidase reaction product is found exclusively in mesangial matrix (MM); glomerular basement membranes (GBMs) are negative for laminin 1 chain at this age. Abbreviations are: CL, capillary lumen; En, endothelium; M, mesangial cell; Po, podocyte; US, urinary space.
Figure 4.
Electron micrographs of glomerular capillaries from 4-week-old 3(IV) collagen knockout (KO) (A) mouse labeled with antilaminin 1 chain [monoclonal antibody (mAb 8B3)-horseradish peroxide (HRP) (B) ] conjugate. Several subepithelial GBM projections contain laminin 1 chain (arrows), but some do not (arrowheads).
Figure 5.
Glomerular capillary loops from 4-week-old 3(IV) collagen knockout (KO) (A) mouse labeled with antilaminin 1 chain [monoclonal antibody (mAb 5A2)-horseradish peroxide (HRP) (B)] conjugate. Most of the abnormal glomerular basement membrane (GBM) segments contain laminin 1 chain (arrows). Note that capillary loops with normal morphology (*) do not contain aberrant laminin 1 chain expression.
Origin of laminin-1 in Alport GBMs
To examine the cellular sources for these abnormally expressed laminin chains, we carried out postfixation immunocytochemistry. When sections of lightly fixed 4-week-old 3(IV) collagen knockout mice were incubated with antilaminin 1- or antilaminin 1-HRP conjugates, peroxidase reaction product was identified within biosynthetic pathways of both endothelial cells and podocytes Figure 6. Intracellular labeling in these cells for laminin-1 chains was never observed in kidneys of wild-type or heterozygous littermates processed at this age.
Figure 6.
Lightly fixed sections of 4-week-old 3(IV) collagen knockout mouse kidney incubated with antilaminin 1- and 1-chain [monoclonal antibody (mAbs)-horseradish peroxide (HRP), as indicated. Intracellular labeling (arrows) for both chains is observed within glomerular endothelial cells (En) and podocytes (Po).
Colocalization of laminin 1, 1, and 5 chains
To compare the distribution of laminin 1 and 1 (laminin-1) with laminin 5 (presumably laminin-11), we carried out double-label immunofluorescence in 4-week-old Alport mice and normal littermates. As shown in the top panel of Figure 7, which illustrates distributions in heterozygous mice, laminin 1 was found in mesangial areas whereas laminin 5 was contained within its usual, GBM, location. In knockouts, by contrast, laminin 1 and 1 chains were commonly found together with laminin 5 chain in GBM out pockets, indicating the presence of both laminin-1 and -11 isoforms in the same GBM segments Figure 7.
Figure 7.
Double-label immunofluorescence of 4-week-old heterozygous (Het) and 3(IV) collagen knockout (KO) kidney sections labeled with anti-laminin 1, 5, or 1 antibodies, as indicated. As shown in the heterozygous (Het) panel, laminin 1 is found in mesangial matrices, whereas laminin 5 is contained within GBMs. In KOs, by contrast, laminin 1 and 1 chains are often found together with laminin 5 chain in peripheral capillary loop glomerular basement membranes (GBMs) (arrows in merged fields).
DISCUSSION
Earlier studies have provided evidence showing abnormal laminin deposition in GBMs in humans with Alport syndrome, as well as in mouse and dog models of the disease8,12. In an effort to understand better the progression of this disorder, we examined immature Alport mice to explore the ultrastructural distribution and cellular sources for some of these abnormally expressed laminin chains. Our results show that laminin-1 down-regulation occurred normally and on schedule in early glomerular development of Alport mice. However, and presumably because of the absence of 3(IV) collagen (as well as absence of 4 and 5(IV) chains), the synthesis of laminin-1 abruptly resumed. Second, abnormal laminin-1 deposition was sustained in Alport GBMs of mature mice and this isoform was derived, at least in part, from both glomerular endothelial cells and podocytes. Finally, many of the GBM irregularities typical of Alport glomeruli contained both laminin-1 and -11 chains, whereas laminin-1 was generally absent from lengths of morphologically normal GBM. These findings suggest that the GBM and podocyte structural irregularities seen in Alport mice may result from the focal presence of both laminin-1 and -11 isoforms, together with the absence of the 3, 4, 5(IV) collagen network.
We previously carried out an immunoelectron microscopic examination of the temporal expression patterns for laminin-1 and -11 chains in wild-type mice17. Normally, laminin 1 chain is abruptly down-regulated after vascular cleft stages (comma- and S-shaped nephric figures) and is undetectable in GBMs of capillary loop stage and maturing glomeruli17. Normal down-regulation was also observed in the 3-day-old Alport kidneys examined here, but laminin 1 chain was then found abundantly in GBMs of late capillary loop and maturing stage glomeruli in the same samples.
What might stimulate the renewed synthesis of laminin-1 in Alport mice? We propose that developing glomerular endothelial cells and podocytes continually monitor their evolving GBM substrates and this surveillance may, in part, dictate cell differentiation programs. If the GBM is assembled incorrectly, perhaps the adherent cells attempt to compensate for such defects by reversing/altering their normal isoform substitution patterns. In turn, these temporally abnormal GBMs may thwart normal endothelial and podocyte differentiation and maturation. In Alport mice, the failure of appearance of a collagen 3, 4, 5(IV) network in capillary loop stage glomeruli may have been sensed by endothelial cells and podocytes, which then responded by resuming synthesis of laminin-1. Support for such an outside-in signaling role by developing GBM comes from studies on 3 integrin knockout mice. Mutations affecting 31 integrin, which is the principal integrin in the peripheral glomerular wall and located specifically at the foot process-GBM interface, result in failure of foot process development and perinatal death18. Importantly, the GBM is also disorganized in these integrin mutants, and probably reflects reciprocal interactions occurring dynamically between adherent cells and the GBM18. Similar conclusions were reached on analysis of laminin 5 and 2 mutant mice. When laminin 5 chain is deleted, an intact GBM is not assembled (or cannot be maintained) and avascular glomeruli develop, apparently because of an inability of endothelial cells to bind to the disrupted/absent GBM10. On the other hand, when laminin 2 chain is deleted, a recognizable GBM is assembled but podocyte foot processes are abnormally broadened, suggesting that laminin-11 is important for normal foot process registration and/or maintenance11.
Although dynamic reciprocities between the endothelium-GBM-podocytes surely occur, exactly how laminin and collagen IV isoform substitutions are regulated at either the gene or protein level are very poorly understood. Patients with nail patella syndrome have mutations encoding a LIM sequence motifs, first identified in homeodomain proteins Lin-11, Isl-1, and Mec-3, Lmx1b, and some of these patients have GBM abnormalities as well19. Recently, Lmx1b has been shown to regulate expression of COL4A3 and COL4A4, as well as the NPHS2 (podocin) and CD2AP genes20,21,22. Although these findings show that expression of certain basement membrane and slit diaphragm complex proteins may be coordinated in podocytes, there is little understanding of how this may affect regulation of the other collagen type IV chains or any of the laminin isoforms. Indeed, laminin 2 null mice, which do not assemble functional laminin-11, still undergo the normal collagen IV isoform switching program, despite compensatory synthesis of laminin-111, suggesting that collagen IV and laminin genes are controlled independently. In contrast, the absence of collagen 3(IV) chain in the Alport mouse clearly affects laminin isoform expression, but mechanisms underlying this laminin dysregulation are utterly unknown.
How does GBM laminin dysregulation, together with the diffuse absence of collagen 3, 4, and 5(IV), lead to the glomerular structural and functional abnormalities of Alport disease? This issue raises several difficult questions. First, we considered the possibility that the laminin-1 chains seen in Alport GBMs were derived by errant translocation out of the mesangial matrix, a site where laminin-1 normally resides in the mature mouse glomerulus. However, we showed by immunoelectron microscopy that laminin-1 chains were found within the intracellular biosynthetic pathways in both endothelial cells and podocytes, indicating that some if not all of Alport GBM laminin-1 was derived from glomerular capillary loop, not mesangial, cells. Another intriguing observation is that laminin-1 was not uniformly distributed throughout Alport GBM, but concentrated specifically and only within the GBM projections and other irregularities. This suggests that, despite the diffuse absence of 3, 4, 5(IV) collagen network throughout the entire Alport GBM, the resumption of laminin-1 synthesis may be carried out only by some, and not all, endothelial cells and podocytes. If true, a second and focal pathogenic event ("second hit") may induce this regional resynthesis of laminin-1. Alternatively, laminin-1 may be secreted by most cells but concentrated within certain GBM microdomains by podocytes. Because these microdomains are always located beneath areas of podocyte foot process effacement, perhaps laminin-1 is towed from a more diffuse GBM distribution and accumulated into foci by effacing foot processes. This remodeling of the GBM and mobility of foot processes would conceivably occur more readily in the absence of the protease-resistant 3, 4, 5(IV) collagen network. Although we cannot currently distinguish between these alternatives, we reemphasize that GBM laminin-1 is normally expressed only in the earliest stages of glomerular development, when podocytes are tall columnar cells that have not yet formed elaborate foot process interdigitations17. Therefore, the reexpression of laminin-1 in Alport glomeruli may be an important pathogenic event that, together with an absence of the 3, 4, 5(IV) collagen network, and maintenance of 1 and 2(IV) collagen, promotes podocyte injury by the recapitulation of an earlier developmental program. The unusual occurrence of both laminin-1 and -11 isoforms within the same GBM segments in the Alport mouse may further disrupt normal GBM function and architecture, leading to the abnormal morphologic features unique to this disease.
CONCLUSION
In conclusion, we have shown that laminin-1, which is normally down-regulated in early stages of glomerular development, is reexpressed in a murine model of Alport disease and becomes concentrated in the abnormal GBM projections typical of this disorder. We speculate that this laminin reexpression is due to a surveillance process conducted by glomerular endothelial cells and podocytes which, in the absence of 3, 4, 5(IV) collagen deposition into the GBM, reactivates laminin-1 genes in both cell types, and maintains expression of collagen 1 and 2(IV) Figure 8. Clearly, much more work is required before we begin to understand how basement membrane protein genes are activated and silenced under normal and abnormal conditions. Perhaps continued studies on laminin dysregulation in the Alport mouse will shed more light on this problem.
Figure 8.
Diagram summarizing temporal changes in laminin and collagen IV chain expression during glomerular capillary wall development as shown in this and previous studies, using chain specific antibodies1,7,15,17. (A) Vascular cleft basement membranes of early nephric figures contain laminin 1 and 1 chains (laminin-1) and 1(IV) and 2(IV) chains of collagen type IV. (B) Beginning with the glomerular capillary loop stage of development, laminin 5 and 2 chains (laminin-11) and 3(IV), 4(IV), and 5(IV) collagen chains are normally observed. Alport GBMs, however, lack 3(IV), 4(IV), and 5(IV) collagen. (C) GBMs of maturing stage glomeruli normally contain laminin 5 and 2 (laminin-11) and 3(IV), 4(IV), and 5(IV) collagen. Laminin-1 and collagen 1(IV) and 2(IV) chains have disappeared. In Alport kidneys, laminin-1 is reexpressed and found together with laminin-11, as well as 1(IV) and 2(IV) collagen. (Laminin 2 chain is also present in Alport glomerular basement membranes (GBMs)12, but the timing of its appearance during glomerular development is not yet known). (Redrawn from reference17).
References
| 1. | Miner JH. Developmental biology of glomerular basement membrane components. Curr Opin Nephrol Hypertens 1998; 7: 13–19. | PubMed | ISI | ChemPort | |
| 2. | Kashtan CE. Alport syndrome: An inherited disorder of renal, ocular and cochlear basement membranes. Medicine (Baltimore) 1999; 78: 338–360. | Article | PubMed | ChemPort | |
| 3. | Kalluri R, Shield CF & Todd P et al. Isoform switching of type IV collagen is developmentally arrested in X-linked Alport syndrome leading to increased susceptibility of renal basement membranes to endoproteolysis. J Clin Invest 1997; 99: 2470–2478. | PubMed | ISI | ChemPort | |
| 4. | Harvey SJ, Zheng K & Sado Y et al. Role of distinct type IV collagen networks in glomerular development and function. Kidney Int 1998; 54: 1857–1866. | Article | PubMed | ISI | ChemPort | |
| 5. | Lees G, Helman RG & Kashtan CE et al. A new form of X-linked dominant hereditary nephritis in dogs. Am J Vet Res 1999; 60: 373–383. | PubMed | ISI | ChemPort | |
| 6. | Lees G, Helman RG & Kashtan CE et al. A model of autosomal recessive Alport syndrome in English cocker spaniel dogs. Kidney Int 1998; 54: 706–719. | Article | PubMed | ISI | ChemPort | |
| 7. | Miner JH & Sanes JR. Molecular and functional defects in kidneys of mice lacking alphacollagen3(IV): Implications for Alport syndrome. J Cell Biol 1996; 135: 1403–1413. | Article | PubMed | ISI | ChemPort | |
| 8. | Cosgrove D, Meehan DT & Grunkemeyer JA et al. Collagen COL4A3 knockout: A mouse model for autosomal Alport syndrome. Genes Dev 1996; 10: 2981–2992. | PubMed | ISI | ChemPort | |
| 9. | Lu W, Phillips CL & Killen PD et al. Insertional mutation of the collagen genes Col4a3 and Col4a4 in a mouse model of Alport syndrome. Genomics 1999; 61: 113–124. | Article | PubMed | ISI | ChemPort | |
| 10. | Miner JH & Li C. Defective glomerulogenesis in the absence of laminin  5 demonstrates a developmental role for the kidney glomerular basement membrane. Dev Biol 2000; 217: 278–289. | Article | PubMed | ISI | ChemPort | |
| 11. | Noakes PG, Miner JH & Gautam M et al. The renal glomerulus of mice lacking s-laminin/laminin  2: Nephrosis despite molecular compensation by laminin 1. Nat Genet 1995; 10: 400–406. | Article | PubMed | ISI | ChemPort | |
| 12. | Kashtan CE, Kim Y & Lees GE et al. Abnormal glomerular basement membrane laminins in murine, canine, and human Alport syndrome: Aberrant laminin 2 deposition is species independent. J Am Soc Nephrol 2001; 12: 252–260. | PubMed | ISI | ChemPort | |
| 13. | Abrahamson DR, Irwin MH & St. John PL et al. Selective immunoreactivities of kidney basement membranes to monoclonal antibodies against laminin: Localization of the end of the long arm and the short arms to discrete microdomains. J Cell Biol 1989; 109: 3477–3491. | Article | PubMed | ISI | ChemPort | |
| 14. | St. John PL, Wang R & Yin Y et al. Glomerular laminin isoform transitions: Errors in metanephric culture are corrected by grafting. Am J Physiol 2001; 280: F695–F705. | ISI | ChemPort | |
| 15. | Miner JH, Patton BL & Lentz SI et al. The laminin chains: Expression, developmental transitions, and chromosomal locations of 1-5, identification of heterotrimeric laminins 8-11, and cloning of a novel 3 isoform. J Cell Biol 1997; 137: 685–701. | Article | PubMed | ISI | ChemPort | |
| 16. | Abrahamson DR & St. John PL. Loss of laminin epitopes during glomerular basement membrane assembly in developing mouse kidneys. J Histochem Cytochem 1992; 40: 1943–1953. | PubMed | ISI | ChemPort | |
| 17. | St. John PL & Abrahamson DR. Glomerular endothelial cells and podocytes jointly synthesize laminin-1 and -11 chains. Kidney Int 2000; 60: 1037–1046. | ISI | |
| 18. | Kreidberg JA, Donovan MJ & Goldstein SL et al. Alpha 3 beta 1 integrin has a crucial role in kidney and lung organogenesis. Development 1996; 122: 3537–3547. | PubMed | ISI | ChemPort | |
| 19. | Dreyer SD, Zhou G & Baldini A et al. Mutations in LMX1B cause abnormal skeletal patterning and renal dysplasia in nail patella syndrome. Nat Genet 1998; 19: 47–50. | Article | PubMed | ISI | ChemPort | |
| 20. | Morello R, Zhou G & Dreyer SD et al. Regulation of glomerular basement membrane collagen expression by LMX1B contributes to renal disease in nail patella syndrome. Nat Genet 2001; 27: 205–208. | Article | PubMed | ISI | ChemPort | |
| 21. | Miner JH, Morello R & Andrews KL et al. Transcriptional induction of slit diaphragm genes by Lmx1b is required in podocyte differentiation. J Clin Invest 2002; 109: 1065–1072. | Article | PubMed | ISI | ChemPort | |
| 22. | Rohr C, Prestel J & Heidet L et al. The LIM-homeodomain transcription factor Lmx1b plays a crucial role in podocytes. J Clin Invest 2002; 109: 1073–1082. | Article | PubMed | ISI | ChemPort | |
Acknowledgments
This work was supported by National Institutes of Health grants DK-34972 and DK-52483 (D.R.A.) and the American Heart Association (B.R.). We thank Dr. Jeffrey Miner for antilaminin 5 antibody and Eileen Roach for technical assistance. Preliminary findings from this study were presented at the 34th Annual Meeting of the American Society of Nephrology, San Francisco, CA, USA, October 13-17, 2001. and are published in abstract form (J Am Soc Nephrol 12:560A, 2001).
