Pathogenic INF2 Variants Causing Podocytopathy and Neuropathy Disarrange Cytoskeletal and Organelle Interactions in Living Cells

Mutations in inverted formin 2 (INF2), an actin assembly factor, are a major cause of focal segmental glomerulosclerosis (single FSGS phenotype). Rare INF2 variants affecting an extreme N-termini (up to residue 184) cause Charcot-Marie-Tooth (CMT) in addition to FSGS (dual FSGS/CMT phenotype). All variants reside within the DID domain, which impairs the intramolecular, auto-inhibition mechanism. The kidney resident, INF2 CAAX isoform localizes predominantly in the endoplasmic reticulum (ER), a dynamic and largest network generating a multiple contact site with other organelles. However, the biological effects of pathogenic INF2 variants on cytoskeletal-organelle interaction remains unclear. This study aimed to clarify the cellular effects of INF2 variants by comparing those causing single FSGS vs dual FSGS/CMT in a living cell.

We transfected HeLa and COS-7 cells with GFP-tagged INF2 wild-type or pathogenic variants, causing either single FSGS (T161N, N202S) or dual FSGS/CMT disorder (G73D, V108D). High-resolution (100 nm) live-cell imaging using a spinning-disk, confocal microscopy (DragonFly) was performed to visualize a dynami ER morphology, undergoing a fast conformational transition from tubule to sheet. vice versa.  Dynamic behaviors of actin stress fibers and organelles (mitochondria, lysosomes) were assessed by transfecting Lifeact construct and loading the fluorescent tracking dyes. Vesicle motility was quantified by the trajectory analysis. Additionally, scanning electron microscopy (SEM) was used to evaluate ultrastructure of cytoskeleton-organelle interface. 

Wild-type INF2 cells showed a tubular pattern in peripheral ER, which uniformly spread along the microtubule arrays, while leaving some sheet-like appearance in perinuclear regions. FSGS-single disease variants retained the ER pattern but displayed more sheet-like appearance. In contrast, FSGS/CMT variant cells had a marked tubule-to-sheet transition in the peripheral ER, as well as retraction form cell edges with deposition of INF2 aggregates, resulting from disarranged actin and microtubule networks. Mitochondrial fragmentation and peripheral mis-distribution were more severe in FSGS/CMT variants than in FSGS variants. Live imaging revealed a reduced lysosomal vesicle motility in INF2 variant expressing cells (n=100, p<0.0001) and increased the ER-mitochondria contact sites. SEM analysis with sections of INF2 variant cells revealed a prevalent sheet-like ER pattern with and disrupted actin-microtubule network, indicating impaired ER interfaces with other organelles.    

Pathogenic INF2 variants impair cytoskeletal architecture and ER-organelle contact sites, generating a spectrum of pathogenic effect from FSGS (mild) to FSGS/CMT variants (severe). High resolution, live-cell imaging revealed previously unrecognize alterations, including shortened actin bundles, elongated filopodia, and dysmorphic mitochondria. These findings support a shared mechanism involving cytoskeletal collapse and organelle dysfunction among INF2-disorders. Further study is needed to define the tissue-specific, cofactors that modulate deleterious effects of variants in podocyte or neuron.