Chimerism of the renal glomerulus revisited.

Since the early descriptions of Alport syndrome, when the glomerular basement membrane (GBM) was found split into multiple laminar strands, investigators have wondered about the contribution of various cell types in the differential synthesis of these strands1; podocytes and endothelia flanking the GBM were always potential candidates. To address this issue, investigators began to examine various developmental events during glomerulogenesis using interspecies grafting, transplantation, and the generation of mutant mice with the ultimate goal of producing hybrid glomeruli showing chimerism among the laminin and collagen chains of the GBM. The article by Abrahamson et al.2 in this issue highlights the value of generating hybrid glomeruli during mouse embryonic development to tease out the cells synthesizing various laminin isoforms. Morphogenesis in mammals is governed by various homotypic and heterotypic cell interactions leading to differentiation of epithelia or endothelia followed by the directional migration of endothelial cells to vascularize various local compartments.3 Nephrogenesis ensues through similar cell interactions during embryonic life, when the glomerulus passes through a series of developmental stages: Vesicle, commaand S-shaped body, precapillary, and maturing capillary stages.4 Accompanying these stages are changes in the basal laminae lining epithelial and endothelial surfaces. Initially the two basal laminae form a loosely organized matrix in S-shaped cleft and precapillary stages that ultimately assembles into a compact GBM with maturity. This process suggests dual cellular origin of the GBM.5 The GBM is an amorphous scaffold of matrix stratified into a central compact lamina densa that is flanked by relatively loose lamina rara interna and externa. These regions are composed of high molecular weight proteins, including type IV collagen, laminins, entactin/nidogen, and sulfated proteoglycans.6 The last are known modulators of morphogenesis with a wide variety of potential functional domains.6 Their glycosaminoglycan chains are made up of either heparan or chondroitin sulfate, which is attached to its respective core peptides. In the early 1980s, Farquhar and colleagues5 suggested the occurrence of a switch in the expression of these glycosaminoglycan chains during development. In this switch, chondroitin sulfate, distributed randomly in the matrix but relatively closer to the endothelia, is substituted by heparan sulfate in the mature GBM and equally distributed in the lamina rarae, thus raising the possibility of the dual cellular source for their synthesis by cells endowed with differential capabilities for posttranslational modification. Similarly, a dual origin of various peptide chains of laminin and collagen, albeit at the translational level or by alternative splicing, and their substitution during metanephric development would be possible. Both laminin and collagen have a number of peptide chains most likely synthesized by different cells but assembled in various combinations in the matrix to yield differential properties while retaining the characteristic functional domains of each specific isoform.4,6 The cellular source of individual isoforms would dictate function, and the ideal model for delineating such functional effects would be the generation of hybrid glomeruli producing a chimeric GBM—an idea of Lauri Saxen that needs revisiting. In the mid-1980s, Saxen and colleagues7,8 designed a series of interspecies (mouse/quail or quail/chick) grafting experiments to generate hybrid glomeruli to study the origin of endothelium. Eleven-day mouse, avascular kidney explants were transplanted onto quail chorioallantoic membranes. The grafts were vascularized by invading avian-derived endothelial cells, as assessed by the presence of deeply stained nucleoli, whereas podocytes lacking prominent nucleoli were of murine origin.7 Similar results were observed in quail/chick transplantation experiments. No vascularization was seen when uninduced mouse metanephric mesenchyme was transplanted onto chorioallantoic membrane, suggesting that early progenitor epithelial cells in the induced kidney produce a chemoattractant for migrating endothelial cells that guide them to vascularize these hybrid glomeruli.8 The GBM that formed in these hybrid glomeruli demonstrated co-reactivity with species-specific antibodies to both type IV collagen and laminin, suggesting dual origin.9 The other basement membranes in these interspecies metanephric explants were exclusively of either avian or murine origin. Co-reactivity of the GBM with species-specific antibodies led to studies of the cellular source of GBM proteins, and having an incomplete fusion of the GBM in these interspecies hyPublished online ahead of print. Publication date available at www.jasn.org.