Parkinsonism Sac domain mutation in Synaptojanin-1 affects ciliary properties in iPSC-derived dopaminergic neurons.

Synaptojanin-1 (SJ1) is a major neuronal-enriched PI(4, 5)P2 4- and 5-phosphatase implicated in the shedding of endocytic factors during endocytosis. A mutation (R258Q) that impairs selectively its 4-phosphatase activity causes Parkinsonism in humans and neurological defects in mice (SJ1RQKI mice). Studies of these mice showed, besides an abnormal assembly state of endocytic factors at synapses, the presence of dystrophic nerve terminals selectively in a subset of nigro-striatal dopamine (DA)-ergic axons, suggesting a special lability of DA neurons to the impairment of SJ1 function. Here we have further investigated the impact of SJ1 on DA neurons using iPSC-derived SJ1 KO and SJ1RQKI DA neurons and their isogenic controls. In addition to the expected enhanced clustering of endocytic factors in nerve terminals, we observed in both SJ1 mutant neuronal lines increased cilia length. Further analysis of cilia of SJ1RQDA neurons revealed abnormal accumulation of the Ca2+ channel Cav1.3 and of ubiquitin chains, suggesting a defect in the clearing of ubiquitinated proteins at the ciliary base, where a focal concentration of SJ1 was observed. We suggest that SJ1 may contribute to the control of ciliary protein dynamics in DA neurons, with implications on cilia-mediated signaling.

Parkinsonism Sac domain mutation in Synaptojanin-1 affects ciliary properties in iPSC-derived dopaminergic neurons.

Synaptojanin-1 (SJ1) is a major neuronal-enriched PI(4,5)P2 4- and 5-phosphatase implicated in the shedding of endocytic factors during endocytosis. A mutation (R258Q) that impairs selectively its 4-phosphatase activity causes Parkinsonism in humans and neurological defects in mice (SJ1RQKI mice). Studies of these mice showed, besides an abnormal assembly state of endocytic factors at synapses, the presence of dystrophic nerve terminals selectively in a subset of nigro-striatal dopamine (DA)-ergic axons, suggesting a special lability of DA neurons to the impairment of SJ1 function. Here we have further investigated the impact of SJ1 on DA neurons using iPSC-derived SJ1 KO and SJ1RQKI DA neurons and their isogenic controls. In addition to the expected enhanced clustering of endocytic factors in nerve terminals, we observed in both SJ1 mutant neuronal lines increased cilia length. Further analysis of cilia of SJ1RQDA neurons revealed abnormal accumulation of the Ca2+ channel Cav1.3 and of ubiquitin chains, suggesting an impaired clearing of proteins from cilia which may result from an endocytic defect at the ciliary base, where a focal concentration of SJ1 was observed. We suggest that SJ1 may contribute to the control of ciliary protein dynamics in DA neurons, with implications on cilia-mediated signaling.

Centronuclear Myopathy Caused by Defective Membrane Remodelling of Dynamin 2 and BIN1 Variants.

Centronuclear myopathy (CNM) is a congenital myopathy characterised by centralised nuclei in skeletal myofibers. T-tubules, sarcolemmal invaginations required for excitation-contraction coupling, are disorganised in the skeletal muscles of CNM patients. Previous studies showed that various endocytic proteins are involved in T-tubule biogenesis and their dysfunction is tightly associated with CNM pathogenesis. DNM2 and BIN1 are two causative genes for CNM that encode essential membrane remodelling proteins in endocytosis, dynamin 2 and BIN1, respectively. In this review, we overview the functions of dynamin 2 and BIN1 in T-tubule biogenesis and discuss how their dysfunction in membrane remodelling leads to CNM pathogenesis.

Imaging-based evaluation of pathogenicity by novel DNM2 variants associated with centronuclear myopathy.

A centronuclear myopathy (CNM) is a group of inherited congenital diseases showing clinically progressive muscle weakness associated with the presence of centralized myonuclei, diagnosed by genetic testing and muscle biopsy. The gene encoding dynamin 2, DNM2, has been identified as a causative gene for an autosomal dominant form of CNM. However, the information of a DNM2 variant alone is not always sufficient to gain a definitive diagnosis as the pathogenicity of many gene variants is currently unknown. In this study, we identified five novel DNM2 variants in our cohort. To establish the pathogenicity of these variants without using clinicopathological information, we used a simple in cellulo imaging-based assay for T-tubule-like structures to provide quantitative data that enable objective determination of pathogenicity by novel DNM2 variants. With this assay, we demonstrated that the phenotypes induced by mutant dynamin 2 in cellulo are well correlated with biochemical gain-of-function features of mutant dynamin 2 as well as the clinicopathological phenotypes of each patient. Our approach of combining an in cellulo assay with clinical information of the patients also explains the course of a disease progression by the pathogenesis of each variant in DNM2-associated CNM.

Dynamin 2 and BAR domain protein pacsin 2 cooperatively regulate formation and maturation of podosomes.

Podosomes are actin-rich adhesion structures formed in a variety of cell types, such as monocytic cells or cancer cells, to facilitate attachment to and degradation of the extracellular matrix (ECM). Previous studies showed that dynamin 2, a large GTPase involved in membrane remodeling and actin organization, is required for podosome function. However, precise roles of dynamin 2 at the podosomes remain to be elucidated. In this study, we identified a BAR (Bin-Amphiphysin-Rvs167) domain protein pacsin 2 as a functional partner of dynamin 2 at podosomes. Dynamin 2 and pacsin 2 interact and co-localize to podosomes in Src-transformed NIH 3T3 (NIH-Src) cells. RNAi of either dynamin 2 or pacsin 2 in NIH-Src cells inhibited podosome formation and maturation, suggesting essential and related roles at podosomes. Consistently, RNAi of pacsin 2 prevented dynamin 2 localization to podosomes, and reciprocal RNAi of dynamin 2 prevented pacsin 2 localization to podosomes. Taking these results together, we conclude that dynamin 2 and pacsin 2 co-operatively regulate organization of podosomes in NIH-Src cells.

Mutant BIN1-Dynamin 2 complexes dysregulate membrane remodeling in the pathogenesis of centronuclear myopathy.

Membrane remodeling is required for dynamic cellular processes such as cell division, polarization, and motility. BAR domain proteins and dynamins are key molecules in membrane remodeling that work together for membrane deformation and fission. In striated muscles, sarcolemmal invaginations termed T-tubules are required for excitation-contraction coupling. BIN1 and DNM2, which encode a BAR domain protein BIN1 and dynamin 2, respectively, have been reported to be causative genes of centronuclear myopathy (CNM), a hereditary degenerative disease of skeletal muscle, and deformation of T-tubules is often observed in the CNM patients. However, it remains unclear how BIN1 and dynamin 2 are implicated in T-tubule biogenesis and how mutations in these molecules cause CNM to develop. Here, using an in cellulo reconstitution assay, we demonstrate that dynamin 2 is required for stabilization of membranous structures equivalent to T-tubules. GTPase activity of wild-type dynamin 2 is suppressed through interaction with BIN1, whereas that of the disease-associated mutant dynamin 2 remains active due to lack of the BIN1-mediated regulation, thus causing aberrant membrane remodeling. Finally, we show that in cellulo aberrant membrane remodeling by mutant dynamin 2 variants is correlated with their enhanced membrane fission activities, and the results can explain severity of the symptoms in patients. Thus, this study provides molecular insights into dysregulated membrane remodeling triggering the pathogenesis of DNM2-related CNM.

Internalization of AMPA-type Glutamate Receptor in the MIN6 Pancreatic ß-cell Line.

The activity of AMPA-type glutamate receptor is involved in insulin release from pancreatic ß-cells. However, the mechanism and dynamics that underlie AMPA receptor-mediated insulin release in ß-cells is largely unknown. Here, we show that AMPA induces internalization of glutamate receptor 2/3 (GluR2/3), AMPA receptor subtype, in the mouse ß-cell line MIN6. Immunofluorescence experiments showed that GluR2/3 appeared as fine dots that were distributed throughout MIN6 cells. Intracellular GluR2/3 co-localized with AP2 and clathrin, markers for clathrin-coated pits and vesicles. Immunoelectron microscopy revealed that GluR2/3 was also localized at plasma membrane. Surface biotinylation and immunofluorescence measurements showed that addition of AMPA caused an approximate 1.8-fold increase in GluR2/3 internalization under low-glucose conditions. Furthermore, internalized GluR2 largely co-localized with EEA1, an early endosome marker. In addition, GluR2/3 co-immunoprecipitated with cortactin, a F-actin binding protein. Depletion of cortactin by RNAi in MIN6 cells altered the intracellular distribution of GluR2/3, suggesting that cortactin is involved in internalization of GluR2/3 in MIN6 cells. Taken together, our results suggest that pancreatic ß-cells adjust the amount of AMPA-type GluR2/3 on the cell surface to regulate the receptive capability of the cell for glutamate.Key words: endocytosis, GluR2, AMPA, cortactin, MIN6.

[Corrigendum]Phosphorylation of cortactin by cyclin-dependent kinase 5 modulates actin bundling by the dynamin 1-cortactin ring-like complex and formation of filopodia and lamellipodia in NG108-15 glioma-derived cells.

Subsequently to the publication of the above article, the authors have realized that the second­listed author, The Mon La, had not been properly credited as one of the co­writers of the paper. Therefore, the Authors' Contributions of the Declarations section of the article should have read as follows: Authors' contributions HY, KTa and TML designed the research and wrote the paper. HY, TA, YM, EO and TT performed mutant protein construction, protein purification and actin bundling experiments. TA and YM performed electron microscopy. EO, TML, KS and KF performed immunofluorescent microscopy, cell migration assay and analyzed data. FYW and KTo identified phosphorylation sites by MALDI­MS. All authors read and approved the final manuscript. The authors apologize to the readership of the Journal for the misinformation in this regard, and for any inconvenience caused. [the original article was published in International Journal of Oncology 54: 550­558, 2019; DOI: 10.3892/ijo.2018.4663].

Phosphorylation of cortactin by cyclin-dependent kinase 5 modulates actin bundling by the dynamin 1-cortactin ring-like complex and formation of filopodia and lamellipodia in NG108-15 glioma-derived cells.

Dynamin copolymerizes with cortactin to form a ring­like complex that bundles and stabilizes actin filaments. Actin bundle formation is crucial for generation of filopodia and lamellipodia, which guide migration, invasion, and metastasis of cancer cells. However, it is unknown how the dynamin­cortactin complex regulates actin bundle formation. The present study investigated phosphorylation of cortactin by cyclin­dependent kinase 5 (CDK5) and its effect on actin bundle formation by the dynamin­cortactin complex. CDK5 directly phosphorylated cortactin at T145/T219 in vitro. Phosphomimetic mutants in which one or both of these threonine residues was substituted by aspartate were used. The three phosphomimetic mutants (T145D, T219D and T145DT219D) had a decreased affinity for F­actin. Furthermore, electron microscopy demonstrated that these phosphomimetic mutants could not form a ring­like complex with dynamin 1. Consistently, the dynamin 1­phosphomimetic cortactin complexes exhibited decreased actin­bundling activity. Expression of the phosphomimetic mutants resulted in not only aberrant lamellipodia and short filopodia but also cell migration in NG108­15 glioma­derived cells. These results indicate that phosphorylation of cortactin by CDK5 regulates formation of lamellipodia and filopodia by modulating dynamin 1/cortactin­dependent actin bundling. Taken together, these findings suggest that CDK5 is a potential molecular target for anticancer therapy.