Optic Nerve Hypoplasia (ONH) is the leading cause of childhood blindness in developed nations and its prevalence has been rising. Yet, we know little about the genetic, molecular, or cellular mechanisms underlying ONH. A previous study described ONH in a cohort of patients with mutations in CASK, an X-linked gene with established roles in neural development and synaptic function. I have demonstrated that heterozygous deletion of CASK in mice (Cask+/-) recapitulates many of the phenotypes observed in patients with CASK mutations, including ONH. This includes reduced optic nerve size, reduced numbers of retinal ganglion cells (RGCs), reduced RGC axonal diameter, and deficits in vision-related tasks. Further analysis on a homozygous partial loss of function variant (Caskfl/fl) also displayed ONH with reduced numbers of RGCs. In order to understand the mechanisms underlying CASK-associated ONH, I explored whether RGCs, the projection neurons of the retina and the cells whose axons comprise the optic nerve, generate CASK. Indeed, mRNA analysis revealed expression of CASK by a large cohort of RGCs. In order to assess whether loss of CASK from a majority of RGCs leads to ONH, I crossed a conditional allele of CASK (CASKfl/fl) with transgenic mice that express Cre Recombinase (Cre) in RGCs. Deletion of CASK from RGCs did not further alter ONH size nor RGC survival. These results demonstrate that loss of CASK signaling in this discrete neuronal populations is not sufficient to lead to further disruption in the assembly of the subcortical visual circuit, suggesting a non-cell autonomous mechanism for loss of CASK in ONH. / Doctor of Philosophy / The connection between the eye and the brain is crucial for successful vision. Impairment of this connection by either loss of the retinal neurons that project axons to the brain or damage to the nerve (optic nerve) lead to blindness. This occurs in a disease called Optic Nerve Hypoplasia (ONH), which is the leading cause of childhood blindness in developed countries. Discovering the risk factors associated with this disease and mechanisms underlying the disease can help us build tools to treat and repair the optic nerve. Previously, mutations in the CASK gene were found in patients with ONH. Here, I developed a mouse model of CASK mutations to phenocopy the human patients, and used this model to explore the development of ONH. For example, with this mouse model I described for the first time, the timeline of disease progression. Surprisingly, I also showed that loss of CASK specifically from the neurons whose axons generate the optic nerve did not lead to ONH, suggesting that ONH may develop from a failure of a network of cells, rather than just one population of cells.

No description available.

No description available.

Understanding how different regions of the central nervous system (CNS) are affected by genetic insults is critical to advancing the study of CNS pathologies. The cerebellum is one such region which is disproportionately hypoplastic in the majority of cases of CASK gene mutation in humans. CASK is an enigmatic multi-domain scaffolding protein which plays a vital role in organizing protein complexes at the pre-synapse through interactions with both active zone proteins and trans-synaptic adhesion molecules such as liprins-α and neurexins. Mutations in the X-linked CASK gene in humans are largely post-natally lethal in the hemizygous condition and result in microcephaly with pontine and cerebellar hypoplasia (PCH) and also optic nerve hypoplasia (ONH) in heterozygous mutations. Herein, I used various molecular and genetic strategies to uncover the role of the CASK protein in brain function and pathogenesis of cerebellar hypoplasia associated with CASK mutations/deletions. First, using the face- and construct-validated heterozygous CASK knockout (Cask+/-) murine model, I conducted bulk RNA-sequencing and proteomics experiments from whole brain lysates to uncover changes in the Cask+/- brain. RNA-sequencing revealed the majority of changes to be broadly categorized into metabolic, nuclear, synaptic, and extracellular-matrix associated transcripts. Proteomics revealed the majority of changes cluster as synaptic proteins, metabolic proteins, and ribosomal subunits. Thus, absence of CASK in half of brain cells seems to affect synaptic protein content, cell metabolism, and protein homeostasis. Extending these observations, I conducted GFP-trap immunoprecipitation followed by tandem mass spectroscopy to reveal protein complexes in which CASK participates. Commensurate with proteomic changes, CASK was found to complex with synaptic proteins, metabolic proteins, cytoskeletal elements, ribosomal subunits, and protein folding machinery. Next, in order to investigate the pathogenesis of CASK-linked cerebellar hypoplasia, I utilized a human case of early truncation wherein the 27th arginine of CASK is converted to a stop codon. Immunohistochemical analysis of this brain revealed an upregulation of glial fibrillary acidic protein, a common marker for degenerative cell death. To mechanistically test the hypothesis that cerebellar hypoplasia results from cell death rather than developmental failure, I created a murine model wherein CASK is deleted from the majority of cerebellar cells post-development using Cre recombinase driven by the Calb2 promoter. Deleting CASK from all cerebellar granule neurons post-migration indeed leads to degeneration of the cerebellum via massive depletion of granule cells while sparing Purkinje cells. Overall, the cerebellum shrinks by approximately half in cross-sectional area and degeneration is accompanied by a collapsing of the molecular layer and of Purkinje cell dendrites. In addition, cerebellar degeneration presents with a profound locomotor ataxia. In conclusion, CASK seems to be affecting brain energy homeostasis and synaptic connections via interactions with metabolic proteins, synaptic proteins, and protein homeostatic elements. Further, alterations in brain volume associated with CASK-linked disorders is the result of degenerative cell death rather than developmental failure as previously posited. / Doctor of Philosophy / One of the main challenges facing modern neuroscience is the question of how constitutive mutations in genes present in every cell can cause different effects on different parts of the brain. CASK is one such gene which is expressed in every cell of the brain and, when mutated, typically results in an overall smaller brain volume. However, the cerebellum is one region of the brain involved in motor coordination which is disproportionately smaller than the rest of the brain. Through this gene, I investigate here two questions principally: (1) what is the role of the CASK protein in cells? And (2) how is the cerebellum differentially affected? Firstly, I conduct a molecular investigation into what changes in the brain of a mouse model of CASK deletion which recapitulates the majority of human cases found in girls. This genetic model results in half of cells in the body lacking CASK and leads to smaller brain volume with disproportionate reduction in cerebellum size, as in the human subjects. Using a variety of molecular and biochemical methods, I uncover that several classes of proteins are changed in this brain, primarily those associated metabolism and cell-to-cell communication. Further, my experiments indicate that CASK interacts with many of these proteins. Next, I use human cases as well as a novel mouse model to uncover the trajectory of CASK-linked reduction in cerebellar size. The human case indicates molecular signatures of cell death, a surprising finding given that CASK-linked disorders are thought to result from developmental failure. Investigating this mechanistically in a mouse model, I uncover that when CASK is deleted after development, cerebellar cells still die and the cerebellum actually shrinks. Thus, my work herein elucidates potential roles for the CASK molecule in cells and shows, for the first time, that CASK-linked cerebellar size diminishment is degenerative in nature rather than developmental. This degeneration of the cerebellum occurs very early on in infancy and so was missed until now. The most important implication is that a degenerative process could be halted with therapies other than relying exclusively on genetic therapies.

Th The kinetics of exocytotic secretion depend not only on the spatial relationship between calcium channels and the exocytotic apparatus, but also on the total amount of Ca2+ influx through Ca2+ channels, the free Ca2+ around the release site and the filling state of the release-ready vesicles. These factors may differ between neurons and endocrine cells. Bovine chromaffin cells (BCCs) are neuroendocrine cells responsible for catecholamine release from the adrenal glands. Ca2+ imaging experiments have shown that localized zones of Ca2+ influx exist on BCC membranes, but how different Ca2+ channel subtypes are distributed, and the mechanisms by which they are targeted, remain to be elucidated. CASK (calcium, calmodulin associated serine kinase) and Mint1 (Munc-18-interacting protein 1), which are modular adaptor proteins involved in synaptic targeting, have recently been found to function in targeting of á1B Ca2+ channels in hippocampal neurons. These data led to the proposal that Ca2+ channels are clustered in BCCs and that CASK and Mint1 play important roles in targeting and/or anchoring channels to their proper location. p*Using RT-PCR and Western blotting, CASK is demonstrated present in isolated BCCs. Mint1 is shown to be present by Western blotting as well. Immunocytochemical experiments and experiments in which BCCs were transfected with plasmids expressing á1A, á1B, and á1C subunits labeled with green fluorescent protein, have shown that á1A and á1B subunits are clustered on the plasma membranes of BCCs, while the á1C subunit is distributed in diffuse patches. With immunoprecipitation, it was determined that CASK interacts biochemically with á1A and á1B Ca2+ channels. Transfection of BCCs with NC3-GFP, which codes for the sequence of the á1B Ca2+ channel that interacts with CASK and Mint1, results in a punctate pattern of fluorescence, which is consistent with the binding of GFP labeled peptide to complexes of CASK and Mint1 at sites of release. Furthermore, immunocytochemical analysis of cells transfected with NC3-GFP showed that á1B Ca2+ channels have a dispersed distribution suggesting that they have been displaced from the binding sites. These data suggest that CASK and Mint1 are important in clustering and targeting Ca2+ channels in the BCC plasma membrane. This study is the first to show the existence and function of CASK and Mint1 in BCCs, and may contribute to our understanding of the exocytotic process in neuroendocrine cells

Th The kinetics of exocytotic secretion depend not only on the spatial relationship between calcium channels and the exocytotic apparatus, but also on the total amount of Ca2+ influx through Ca2+ channels, the free Ca2+ around the release site and the filling state of the release-ready vesicles. These factors may differ between neurons and endocrine cells. Bovine chromaffin cells (BCCs) are neuroendocrine cells responsible for catecholamine release from the adrenal glands. Ca2+ imaging experiments have shown that localized zones of Ca2+ influx exist on BCC membranes, but how different Ca2+ channel subtypes are distributed, and the mechanisms by which they are targeted, remain to be elucidated. CASK (calcium, calmodulin associated serine kinase) and Mint1 (Munc-18-interacting protein 1), which are modular adaptor proteins involved in synaptic targeting, have recently been found to function in targeting of á1B Ca2+ channels in hippocampal neurons. These data led to the proposal that Ca2+ channels are clustered in BCCs and that CASK and Mint1 play important roles in targeting and/or anchoring channels to their proper location. p*Using RT-PCR and Western blotting, CASK is demonstrated present in isolated BCCs. Mint1 is shown to be present by Western blotting as well. Immunocytochemical experiments and experiments in which BCCs were transfected with plasmids expressing á1A, á1B, and á1C subunits labeled with green fluorescent protein, have shown that á1A and á1B subunits are clustered on the plasma membranes of BCCs, while the á1C subunit is distributed in diffuse patches. With immunoprecipitation, it was determined that CASK interacts biochemically with á1A and á1B Ca2+ channels. Transfection of BCCs with NC3-GFP, which codes for the sequence of the á1B Ca2+ channel that interacts with CASK and Mint1, results in a punctate pattern of fluorescence, which is consistent with the binding of GFP labeled peptide to complexes of CASK and Mint1 at sites of release. Furthermore, immunocytochemical analysis of cells transfected with NC3-GFP showed that á1B Ca2+ channels have a dispersed distribution suggesting that they have been displaced from the binding sites. These data suggest that CASK and Mint1 are important in clustering and targeting Ca2+ channels in the BCC plasma membrane. This study is the first to show the existence and function of CASK and Mint1 in BCCs, and may contribute to our understanding of the exocytotic process in neuroendocrine cells

Calcium/calmodulin-dependent serine protein kinase (CASK) is a membrane-associated guanylate kinase (MAGUK) family protein, which is encoded by a gene of identical name present on the X chromosome. CASK may participate in presynaptic scaffolding, gene expression regulation, and cell junction formation. CASK is essential for survival in mammals. Heterozygous mutations in the CASK gene (in females) produce X-linked intellectual disability (XLID) and mental retardation and microcephaly with pontine and cerebellar hypoplasia (MICPCH, OMIM# 300749). CASK mutations are also frequently associated with optic nerve hypoplasia (ONH) which is the most common cause of childhood blindness in developed countries. Some patients with mutations in CASK have been also diagnosed with optic nerve atrophy (ONA) and glaucoma. We have used floxed CASK (CASKfloxed), CASK heterozygous knockout (CASK(+/-)), CASK neuronal knockout (CASKNKO) and tamoxifen inducible CASK knockout (CASKiKO) mouse models to investigate the mechanism and pathology of CASK-linked ONH. Our observations indicate that ONH occurs with 100% penetrance in CASK(+/-) mice, which also displayed microcephaly and disproportionate cerebellar hypoplasia. Further, we found that CASK-linked ONH is a complex developmental neuropathology with some degenerative components. Cellular pathologies include loss of retinal ganglion cells (RGC), astrogliosis, axonopathy, and synaptopathy. The onset of ONH is late in development, observed only around the early postnatal stage in mice reaching the plateau phase by three weeks of birth. The developmental nature of the disorder is confirmed by deleting CASK after maturity since CASKiKO mice did not produce any obvious optic nerve pathology. Strikingly the CASKfloxed mice expressing ~49% level of CASK did not manifest ONH despite displaying a slightly smaller brain and cerebellar hypoplasia indicating that ONH may not simply be an extension of microcephaly. We discovered that deleting CASK in neurons produced lethality before the onset of adulthood. The CASKNKO mice exhibited delayed myelination of the optic nerve. Overall this work suggests that CASK is critical for neuronal maturation and CASK-linked ONH is a pervasive developmental disorder of the subcortical visual pathway. Finally, in a side project, I also described a new methodology of targeting neurons using receptor-mediated endocytosis which would help target retinal neurons for therapeutic purposes in the future. / Ph. D. / 7 in 10,000 children suffer from childhood blindness, for whom all the visual information from the outside world is completely blocked. Although classified as a rare disease, optic nerve hypoplasia (ONH), or the underdevelopment of optic nerve, is the leading cause of childhood blindness in developed countries, accounting for 15% of childhood blindness. Only a handful of genes have been shown to associate with ONH. The CASK gene, whose protein product calcium/calmodulin-dependent serine protein kinase (CASK) plays a role in presynaptic scaffolding, is one of them. Mutations in the CASK gene not only produce ONH, but also microcephaly and intellectual disability. Investigating the mechanism of CASK-linked ONH will provide critical data to understand the molecular basis of optic nerve formation and maturation. Here we have used the CASK heterozygous knockout mouse model to replicate the ONH and microcephaly seen in female human patients. We discovered that the onset of CASK-linked ONH corresponded to the late third trimester developmental stage in humans, thus ONH is developmental in nature. ONH pathologies include thinning of optic nerves, axonal atrophy, and synaptopathy. In contrast to the postnatal death of constitutive CASK loss of function in mice, CASK ablation in adult mice did not lead to lethality. CASK deletion also delays neuronal myelination. Overall, our results indicate that CASK is critical for postnatal maturation of the central nervous system and mutations of the CASK gene is sufficient to lead to ONH. Early intervention and proper gene therapy may treat CASK-linked ONH.

L’implication d’un facteur circulant et des dysfonctions du système immunitaire entrainant les altérations de la barrière de filtration glomérulaire a été suggérée dans la pathogénèse de la hyalinose segmentaire et focal récidivante. Nous avons identifié par spectrométrie de masse la présence de la protéine CASK dans des sérums de patients après immunoadsoroption sur une colonne de protéine A. CASK recombinante est capable d'induire des modifications de l’architecture des podocytes in vitro, tels qu’une redistribution de la protéine de diaphragme de fente ZO-1 et de la protéine régulatrice d’actine synaptopodine, et une perte de fibres de stress d’actine. Ces podocytes acquièrent ainsi un phénotype motile et une perméabilité accrue à l’albumine en présence de CASK recombinante in vitro. L’injection de CASK chez des souris entraine une protéinurie et l’effacement des pédicelles de podocytes. L’interaction entre CASK et son récepteur CD98 dans les podocytes a été mise en évidence par l’expérience de pontage covalent et co-immunoprécipitation. L’inhibition de l’expression de CD98 par ARNi a permis de préserver l’architecture des podocytes en présence de CASK. Nous avons remarqué la surexpression de CASK dans les monocytes chez les patients atteints de la HSF récidivante par rapport aux témoins. In vitro, CASK est surexprimée dans les macrophages ayant une polarité M2 et est retrouvée dans le surnageant de la culture de ces cellules. La sécrétion de CASK est associée aux exosomes qui sont des microvésicules d’origine endosomale. Dans les cellules, CASK est partiellement co-distribuée avec ALIX, un marqueur exosomal, et leur interaction a été mise en évidence par co-immunoprécipitation. CASK est fortement exprimée dans les exosomes de patients atteints de HSF récidivante comparé aux donneurs sains. Le traitement des podocytes par des exosomes issus des macrophages de type M2 induit des altérations du cytosquelette et augmente la motilité des podocytes comme cela avait été observé en présence de CASK recombinante. Pour conclure, nous avons identifié CASK comme nouveau facteur soluble qui pourrait jouer un rôle au cours de la HSF récidivante après transplantation rénale. Ces découvertes ouvrent de nouvelles orientations pour le traitement des malades atteints de SNI récidivant. / Focal segmental glomerulosclerosis (FSGS) is often associated with a high rate of progression to end-stage renal disease. The idiopathic form has a high recurrence rate (rFSGS) after transplantation suggesting the presence of a systemic circulating factor that causes the glomerular permeability. This factor can be removed by plasmapheresis or immunoadsorption using protein-A columns. We used mass spectrometry to analyze the proteins eluted from protein-A columns, taken from patients with rFSGS after immunoadsorption. A serum form of calcium/calmodulin-dependent serine/threonine kinase (CASK) was identified in rFSGS patients but not in controls. In cultured podocytes, recombinant CASK induced reorganization of the actin cytoskeleton. We also demonstrated the interaction of CASK with CD98 at the cell surface. Injection of recombinant CASK in mice induced proteinuria and foot process effacement on podocytes. We identified that CASK is produced by monocytes in patients with rFSGS. CASK is also expressed and secreted by M2 polarized macrophages but not by M1 subset. CASK was associated with exosomes produced by these cells. CASK has a partial codistribution with ALIX, an exosomal component involved in their development. We’ve also demonstrated that CASK interacts with ALIX in M2 macrophages. Moreover exosomes derived from M2 macrophages cause podocytes cytoskeleton alterations and increase of podocyte motility as observed previously with recombinant CASK. In conclusion, a serum form of CASK secreted by macrophages acts as a permeability factor in patients with rFSGS suggesting its involvement in the physiopathology of rFSGS.

The continual increase in electricity dependence for the advancement of society has led to increased demand in electricity globally. This increased demand, among other things such as global warming interventions and energy security have encouraged the need to diversify electricity generation sources. Civilian use of nuclear power dates back to the 1950s. The United States of America and France are currently leading with the highest nuclear power generation in the world, generating 101 GWe and 63 GWe, respectively. Several countries such as China and the United Arab Emirates have committed to new nuclear build in order to increase their nuclear power generation capacities. Standing against the prospects of growth of the nuclear power industry are technical and nontechnical challenges. These include proliferation risk, safety, high capital costs and high-level waste management. Most spent nuclear fuel from power reactors is currently stored in the spent fuel pools on reactor sites, and some have been reprocessed. It is estimated that about 32% (370 000 tons of Heavy Metal) of the total spent fuel generated from power reactors have been reprocessed up to date. With most of the spent fuel pools filling up, alternative interim and long term disposal of spent nuclear fuel solutions have been under investigation from as early as the 1970s. South Africa has planned an interim dry storage facility for the spent nuclear fuel to be established at the existing Koeberg power station. The interim dry storage facility will make use of HI-STAR 100 multi-purpose casks to store spent nuclear fuel until the country decides on final disposal solution. There are many aspects that are critical to safe, efficient and cost-effective long term storage of spent nuclear fuel. Some of the physics and engineering aspects concerning dry storage facilities are briefly discussed. The aspects presented here are: radiation containment, spent fuel, sub-criticality, decay heat removal, site location aspects, response to seismic events, cask corrosion, transportation infrastructure, operability and monitoring. The study of the three existing dry cask storages from the USA, Hungary and Belgium gives an overview of the dry cask technology in use today. These presentations are based on publicly available reliable information. The proposed dry storage facility at Koeberg will be in the existing power station footprint using the HI-STAR 100 casks. The decision to have the proposed dry storage facility at Koeberg will minimise related licence applications and part of security installations as the site already has some security. The location of the facility in the power station’s footprint also allows for cost-effective and safe transportation of casks from the reactor building to the proposed facility. The modularity aspect of the dry cask storage facility at MV Paks in Hungary should also be employed at Koeberg to allow for more storage. This will cater for additional casks that may need to be stored if more nuclear power plants are procured in the future. South Africa’s air traffic around the Western Cape is not as congested as Belgium’s. There is, therefore, no need for the casks to be housed in concrete buildings like Doel’s. Most of Koeberg’s high-level waste would have had a longer cooling time in the pools compared to the minimum cooling time required for the chosen cask technology. This will provide a conservative, safe approach for Koeberg’s facility. Dry cask storage technology has provided a reliable interim dry storage solution for several countries. Despite uncertainties for long term disposal options, the proposed dry cask storage facility at Koeberg is a suitable interim storage alternative for South Africa to allow continuous operation of the plant. This conclusion is based on the physics and engineering aspects that have been presented in this minor dissertation.

La hyalinose segmentaire et focale (HSF) est une maladie rénale sévère dont la physiopathologie est complexe. La récidive de la maladie après transplantation rénale et l’obtention de sa rémission après un traitement par immunoadsorption (IA) illustre l’implication d’un facteur circulant dans sa physiopathologie, capable de se fixer à la protéine A. Récemment, suPAR a été rapporté comme agent causal et marqueur de la HSF. Le premier objectif de notre travail a été de vérifier si suPAR se fixe à la protéine A. Le deuxième objectif a été d’identifier le facteur circulant responsable de la récidive de la HSF après transplantation rénale, à partir de l’analyse par spectrométrie de masse des protéines liées à la colonne de protéine A après (IA). Premièrement, nous avons mesuré la concentration de suPAR par un test ELISA parmi les protéines fixées à la colonne de protéine A après IA chez 7 patients atteints de HSF récidivantes et dans le sérum de 13 patients atteints de HSF récidivantes et de 11 contrôles sains. Le sérum des patients a été immunoadsorbé in vitro sur bille de protéine A sépharose. Nous avons quantifié suPAR avant et après la procédure et dans l’éluat des protéines fixées à la protéine A. La concentration de suPAR est plus élevée chez les patients atteints de HSF récidivantes par rapport aux groupes contrôles. La concentration de suPAR est très faible dans les proteines éluées à partir de la colonne de protéine A, indiquant que suPAR ne se lie pas à la protéine A et n’est pas le facteur circulant élué par les colonnes de protéines A. Deuxièmement, nous avons identifié le FC à partir des protéines fixées à la colonne de protéine A par une caractérisation des protéines par spectrométrie de masse chez des patients traités pour récidive de HSF et chez un patient contrôle. Nous avons recherché le FC dans le sérum de patient atteint de HSF, de patient ayant une néphropathie diabétique et chez des contrôles sains. L’effet de la protéine recombinante du FC a été testé in vitro sur une culture de podocytes et in vivo chez la souris. Nous avons identifié une forme sérique de CASK (calcium calmoduline sérine thréonine kinase), à partir des protéines fixées à la colonne de protéine A après IA. CASK est présente uniquement dans le sérum de patients atteints de HSF et non dans les groupes contrôles. In vitro, la protéine recombinante de CASK (CASKr) induit une redistribution de l’actine du cytosquelette des podocytes en culture par une interaction avec CD98. CASKr altére la perméabilité des podocytes à l’abumine et induit in vivo une protéinurie chez la souris associé à un effacement des pédicelles.En conclusion, suPAR ne se fixe pas à la protéine A ni in vivo ni in vitro. Une forme sérique de CASK est impliqué dans la récidive de la HSF avec comme cible potentiel CD98 sur le podocyte. / Focal and segmental glomerulosclerosis (FSGS) is a serious disease, the pathogenesis of which is unknown. Its recurrence after transplantation (Tx) and its partial remission after treatment with immunoadsorption (IA) on a protein A column indicate the existence of a circulating factor (CF) responsible for the disease that is able to bind to a protein A column. Recently, the soluble receptor of urokinase (suPAR) was described as the factor responsible for FSGS. The first aim of my work was to test the capacity of suPAR to bind to protein A and to be eliminated by IA. The second aim was to identify the CF responsible of the recurrence of the disease after renal transplantation from the analysis of proteins eluted from protein-A columns from patients with rFSGS who had undergone therapeutic (IA). First, we measured suPAR in eluates of protein A columns from 7 patients with recurrent FSGS after Tx (rFSGS) treated with IA, and in the serum of 13 patients with rFSGS and 11 healthy donors (HD). Additionally, the plasma of these patients was immunoadsorbed in vitro on a protein A Sepharose column and we quantified suPAR in the eluates and in pre- and post-column samples. The concentration of suPAR was higher in the plasma of patients with rFSGS than in the plasma of HD patients. However, the concentration of suPAR was similar before and after IA on protein A for the rFSGS and HD samples. The suPAR concentration was very low in the eluates from protein A columns incubated with plasma from HD or rFSGS patients. However, 85% of rFSGS patients showed a decrease in immunoglobulin G and proteinuria. Secondly, we analyzed proteins eluted from protein-A columns from patients with rFSGS who had undergone therapeutic immunoadsorption. Compared to control a differential band was identified by mass spectrometry. The expression of this protein was tested by immunochemical methods in sera from healthy controls, from patients with proteinuria caused by diabetic nephropathy, and from rFSGS patients. The effect of the recombinant protein was evaluated in vitro (podocytes) and in vivo experiments (mice). A soluble form of calcium/calmodulin-dependent serine/threonine kinase (CASK) eluted from protein-A columns was identified by mass spectrometry. CASK was immunoprecipitated only in the sera from patients with rFSGS. Recombinant CASK induced reorganization of the actin cytoskeleton of cultured podocytes through an interaction with CD98 at the cell surface. In vitro, CASK increased the permeability of podocyte monolayers, and induced proteinuria and foot-process effacement in miceIn conclusion, suPAR does not significantly bind to protein A in vitro or in vivo. Soluble CASK acts as a permeability factor in patients with rFSGS bindinding CD98 on podocytes.