While transmission EM imaging of the glomerulus with OsO4 demonstrates an electron dense layer in the GBM, the significance of this finding has been unclear

While transmission EM imaging of the glomerulus with OsO4 demonstrates an electron dense layer in the GBM, the significance of this finding has been unclear. basement membrane (GBM), an essential mediator of glomerular ultrafiltration. Using multichannel STORM and STORM-electron microscopy correlation, we constructed a molecular reference frame that revealed a laminar organization of ECM proteins within the GBM. GDC-0032 (Taselisib) Separate analyses of domains near the N- and C-termini of agrin, laminin, and collagen IV in mouse and human GBM revealed a highly oriented macromolecular organization. Our analysis also revealed disruptions in this GBM architecture in a mouse model of Alport syndrome. These results provide the first nanoscopic glimpse into the organization of a complex ECM. DOI: http://dx.doi.org/10.7554/eLife.01149.001 null mutation (Miner and Sanes, 1996). In both humans and mice, the lack of this network results in a compensatory increase in the expression of collagen 112(IV), which is normally found at low levels in adult GBM (Kashtan and Kim, 1992; Miner and Sanes, 1996). Despite this compensation, the GBM becomes segmentally split and thickened, and this is associated with hematuria, proteinuria, and progressive renal failure (Hudson et al., 2003). To investigate the architecture of the GBM in this GDC-0032 (Taselisib) disease model, we examined the organization of agrinC and collagen 112(IV). In contrast to the wild-type GBM, the Alport GBM showed segments of capillary loops where the two layered organization of agrinC was disrupted, while in some segments it was intact (Figure 6A,B). The localization of collagen 112(IV) was also altered, with its distribution spread across the width of the GBM and no longer restricted to the endothelial side. This altered collagen 112(IV) distribution was evident even in capillary loop segments that showed a relatively preserved agrinC organization (Figure 6C,D). Thus, we conclude that an intact collagen 345(IV) network helps to maintain agrin and collagen 112(IV) organization in the healthy GBM. Open in a separate window Figure 6. Breakdown of the GBM molecular architecture in a mouse model of Alport syndrome.(A and B) STORM and EM images of a capillary loop from a collagen Rabbit Polyclonal to TSPO 3(IV) knockout (KO) mouse kidney labeled with agrinC shows a thin GBM with 2-layered agrin (single arrow) as well as a breakdown of the 2-layered agrin labeling pattern (double arrow) and a thick irregular GBM stretch showing disorganized, diffuse agrinC labeling (arrowheads). (C and D) Images and quantification of capillary loops selected from collagen 3(IV) KO and heterozygous (HET) kidney that show GDC-0032 (Taselisib) two layers of agrinC. Despite the intact agrin layers, collagen 112(IV) shows an atypical distribution spread across the GBM in the KO vs a single peak in the HET littermate control. GDC-0032 (Taselisib) DOI: http://dx.doi.org/10.7554/eLife.01149.018 Discussion Determining the structure, molecular interactions and spatial organization of ECM proteins are important steps towards understanding their roles in tissue function, morphogenesis and disease. Our study describes a novel super resolution fluorescence microscopy approach to reconstruct the molecular architecture of ECM proteins within the GBM. We developed a method to achieve systematic nanoscale molecule mapping in dense tissue sections using STORM and ultrastructural imaging on the same samples using EM. The robust and relatively rapid throughput GDC-0032 (Taselisib) of this methodology allowed us to reconstruct the molecular architecture of the mouse as well as human GBMs. Nanoscopic imaging in tissues Light microscopic imaging shows the components of the GBM as diffusely distributed, suggesting either an amorphous structure or an organization below the resolution of conventional light microscopy. While transmission EM imaging of the glomerulus with OsO4 demonstrates an electron dense layer in the GBM, the significance of this finding has been unclear. Immunogold-EM has been deployed to study ECM protein organization (Miosge et al., 1999), but constraints of sample preparation, antibody accessibility, and quantitation prevent its extensive application. In our study, we performed STORM as well as EM and immunogold-EM using the same tissue and antibody labeling procedures. This allowed us to validate data using both methods and also illustrated the challenges associated with immunogold EM. Super resolution fluorescence microscopic methods have now been widely used to resolve subcellular structures in cultured cells, but the use of these methods to analyze tissue has been challenging. Single molecule probe based methods (F)PALM and STORM were originally developed using photoswitchable fluorescent proteins (Betzig et al., 2006; Hess et al., 2006) or photoswitching properties of fluorescent dyes (Rust et al., 2006). Exogenous expression of multiple photoswitchable fusion proteins and their functional validation is not.