Figure 1.
Venn diagram of the identified proteins by the IP-MS studies.
Figure 2.
Statistical characterization of MYH9-interacting proteins.
We compiled a database of known human protein-protein interactions (PPI) and used this as a basis from which to evaluate the density of interactions between the human and murine proteins identified by IP-MS. There are 1269 human proteins identified by the IP-MS and in the database there are 5605 interactions between them. In order to ascertain if this number of interactions is larger than what might be expected by chance we selected (n = 105) random sets of 1269 proteins from the database and counted the number of interactions between them, the probability density is shown in (A). Obviously the observed number of 5505 proteins occupies an extreme (large) position in this distribution, having a p value p<10−5. In order to determine if an overrepresentation of “hub” proteins is responsible for this extreme density we compared the degree distribution of the subgraph of the PPI network induced by the identified human proteins (C) to the degree distribution of the whole PPI network (D); the similarity between the two distributions indicates that there is no overrepresentation of hub proteins. This analysis was repeated for the (N = 128) identified murine proteins where the distribution of the number of interactions is shown in (B) and the actual number of interactions is 50 (p<10−5), and the corresponding degree distributions of the induced subgraph and whole PPI network are shown in (E) and (F) respectively. These figures reveal similar results as for the human proteins.
Table 1.
OMIM disease networks identified in MYH9 interactions.
Figure 3.
Enrichment analysis with Network2Canvas of identified proteins interacting with MYH9.
Eight different gene set libraries: WikiPathways, Reactome, BioCarta pathways, PPI hubs, Kinase Enrichment Analysis (KEA), GO biological process (BP), GO cellular component, GO molecular function, VirusMINT, and protein domains from PFAM and InterPro. On each grid all the terms from the gene set libraries are arranged based on their gene content similarity. The highlighted terms are enriched terms where the brighter colors denote higher significance. Some relevant terms are annotated. A) Analysis of mouse MYH9 interacting proteins; B) Analysis of human MYH9 interacting proteins.
Figure 4.
The adjacency matrix of the network of known interactions between the set of MYH9 interacting human proteins, and 3000 random proteins for comparison.
The 623 proteins identified as interacting with MYH9 (in the upper left) along with random set of 3000 random human proteins are plotted along both the x and y axes. Previously described interactions amongst these proteins are depicted as a node at the x, y intersection of a given pair of proteins (A). The increased density of interactions (visually evident as a higher density of nodes in the upper left) within the set of MYH9-interacting proteins indicates that these proteins also belong to previously described complexes. A close-up view is provided which also displays the community structure, with discrete clusters boundaries indicated with orange lines (B). This community structure indicated that nine distinct clusters exist, representing nine distinct groups of proteins with multiple previously described interactions.
Figure 5.
MYH9 is expressed within podocyte processes.
Human renal tissue was stained by immunofluorescence (red) as well as phalloidin (green) and DAPI counterstains. MYH9 expression was detected in podocytes (arrows) as well as mesangial cells and parietal epithelial cells. MYH9 appeared to be expressed within the major processes of podocytes overlaying glomerular capillary tufts.
Figure 6.
Validation of shRNA constructs against MYH9.
MYH9 expression was assayed in total protein collected from 293T cells, four days following transfection with shRNA constructs targeting MYH9 mRNA (shRNAs 1–4). A construct bearing a random shRNA served as negative control (shRNA-NC). Construct 2 achieved a dramatic reduction in expression, while the other constructs did not reduce protein levels.
Figure 7.
Effect of knockdown of MYH9 expression on podocytes.
Human conditionally immortalized podocytes were transfected with shRNA-2 to reduce MYH9 expression or shRNA-NC as negative control. Following transfection, podocytes were grown at GR conditions to allow differentiation. Analysis by Real Time-PCR (A) and western blot (B) confirmed knockdown of expression. Podocytes were stained with phalloidin (red) to visualize the actin cytoskeleton and DAPI counterstain. Control podocytes exhibited large cell bodies, with stress fibers typical of differentiated cells (C). Podocytes with reduced MYH9 expression demonstrated smaller cell size with rarefied actin cytoskeleton, and lacking stress fibers typical of differentiation (D).