Final Report Summary - NGIDD (Neuron-glia interactions in nerve development and disease)
Executive summary:
Myelin disorders present a significant health burden and are an important cause of disability. Myelin is laid down in segments around a majority of axons and contributes to the normal physiology of the nervous system. Schwann cells and oligodendroglia cells are the myelin-forming cells in the peripheral and central nervous system respectively and the formation and maintenance of myelin results from a continuous molecular dialogue with the neuronal axon. Despite the importance of myelin for the normal function of the nervous system, relatively little is known about the molecular mechanisms that underlie axon-glia communication in health and disease. Consequently, the therapeutic options for these disorders are very limited. The NGIDD project aimed to bridge the gap in our knowledge and identify new drug targets to stimulate the regenerative capacity of the nervous system.
To realize this overall aim, the members of the NGIDD consortium addressed the following questions related to; 1 the initial engagement of glial cells with axons, 2 the control of myelin formation and maintenance, 3 the functional organization of the axo-glial unit and 4, the pathology of de- and dismyelination.
Over the last four years, the NGIDD consortium has made major steps towards answering these questions. NGIDD has identified a completely novel molecular interaction consisting of the Schwann cell derived protein Lgi4 and its axonal receptor Adam22. NGIDD has elucidated the role of integrins, focal adhesion kinase and N-WASP in myelin formation and axonal sorting. Furthermore NGIDD has generated a range of mouse models of peripheral neuropathies and has provided evidence that the drug rapamycin is a beneficial modulator of myelin abnormalities in these mouse models. Additionally, NGIDD has generated and validated a cell-based assay to screen drugs that ameliorate the pathogenic effects of maladaptive unfolding protein responses that plays a role in a number of inherited neuropathies.
Thus, scientific and technological progress made by the NGIDD consortium has opened new avenues of inquiry to further our understanding of the molecular and cellular interactions that shape our myelinated nerves and govern their physiology. Importantly, this research has already yielded new lead compounds and preclinical models to explore new therapeutic approaches to combat the devastating neurological effects in de- and dismyelinating diseases in humans.
The NGIDD consortium united scientists from five different countries with different scientific expertise into a common cause. The consortium has contributed to the exchange of ideas, reagents and personnel, has trained a large number of scientists and has participated in scientific and civic conferences to disseminate the novel findings of the consortium, thus successfully contributing to the larger European Research Area.
Project Context and Objectives:
Although for many years it has been appreciated that myelinating glial cells and their underlying axons are engaged in a complex dialogue which is crucial for the normal function of both cell types, the identity of the molecules that mediate this interaction remained almost completely unknown. Studies by members of the consortium contributed significantly to our understanding how myelinating glial cells communicate with the axons they ensheath. The studies carried out in this collaborative effort provide important new insights on the molecular mechanisms that control myelination and the functional organization of the axon. Uncovering the mechanisms by which myelinating glial cells recognize and ensheath their associated axons is highly significant given the high prevalence of human diseases resulting from demyelination or alteration in myelin, such as multiple sclerosis and a number of inherited neuropathies.
Many developmental mechanisms are recapitulated after injury and pathology in these diseases, and can be exploited to promote regeneration, remyelination and re-establishment of functional axo-glial units. Furthermore, understanding the molecular basis for axo-glial contact is of particular interest as it is becoming increasingly evident that although in many of this disease the first pathological event is the disruption of myelin, secondary changes in the axon, which eventually lead to axonal loss is the main cause of clinical impairment. In summary, the present collaboration aimed to uncover the molecular mechanisms that govern the Schwann cell-axon engagement, myelination, and the establishment of distinct structural domains, all of which are processes that are crucial for the generation of a fully functional myelinated nerve. Increased knowledge about these mechanisms will prove useful in the design of therapeutic strategies for many neuropathies and demyelinating disorders.
The collaborative project NGIDD consisted of seven European research groups and one biotech company Axxam. The specific expertise of the individual members was mobilized to address four key objectives. These key objectives aligned with the four workpackages in this collaborative effort under one management (workpackage 5) to realize the main objective of this project; to push our understanding of myelination and myelination-related diseases beyond the state of the art and to identify new targets for pharmaceutical intervention to aid the regenerative capacity of the nervous system.
Initial events leading up to myelination
The glial cells and neurons of the peripheral nervous system are in close apposition and interdependent from the moment they are generated from the neural crest during embryonic development. Schwann cell precursors proliferate and migrate along the outgrowing axonal bundles to give rise to the mature axon-glia unit that characterize the fully developed and functional nervous system.
One crucial step in achieving correct axo-glia relationships in peripheral nerve development is radial axonal sorting. This developmental step allows Schwann cells to achieve a 1:1 relationship with large axons destined to be myelinated. Radial sorting is a pre-requisite for myelination, and is impaired in a common form of hereditary human neuropathy that accompanies congenital muscular dystrophy 1A. Sorting processes are also recapitulated during nerve repair. Thus, to clarify the mechanisms governing axonal sorting is of utmost importance not only to understand normal nerve development, but also to promote therapeutic intervention in the context of hereditary diseases and nerve repair. Therefore, an important question addressed in this context is how molecules such as extracellular matrix laminins and their integrin receptors, together with Focal Adhesion Kinase and the small GTPase Rac1 enable radial sorting of axons.
Many other players in radial sorting remain to be identified. Work on the natural mouse mutant Claw paw (clp) has identified one such player. Delayed axonal sorting and abnormal myelination of peripheral nerve fibers characterize Claw paw mice. It was found that a mutation in the Lgi4 gene causes the defects in nerve development in these mice. Lgi4 is a member of a small family of genes that encode secreted Leucine-rich proteins of unknown function. Lgi1, the founding member of this family, is frequently mutated in autosomal dominant partial epilepsy with auditory features (ADPEAF). How Lgi4 contributes to sorting and myelination is not known. An important insight into the possible mechanism of action of Lgi4 was obtained during this project through the identification of the receptor for Lgi4.
Control of myelin formation
It has been known for some time that, in postnatal development, axons themselves induce the associated Schwann cells to make myelin. However, the nature of this axonal signal was unclear until recently when a member lab of the consortium showed that neuregulin-1 (NRG1) type III is a critical growth factor of axons regulating myelination by Schwann cells in a quantitative fashion, thereby optimizing myelin sheath thickness. Binding of NRG1 to glial ErbB2 and ErbB3 receptors (via the EGF-like signaling domain common to all NRG1 isoforms) is likely to activate multiple signaling pathways, but it is unknown how these are tied into the myelination program. One attractive second messenger for myelination, which is activated by NRG1-ErbB signaling, is the PI3K pathway with the generation of phosphatidyl-inositol (3,4,5)-triphosphate (?PIP3?) as a signaling lipid. Activation of PI3K has been shown to modulate Schwann cell growth and myelination in vitro, but the role of PIP3 in myelination in vivo was still unclear. PIP3 is an activator of Akt1 and the mTOR pathway in RNA transcription and translation, but is also a subcellular activator of rac/cdc42-dependent process formation. To address the function of glial PI3K and mTOR signalling in vivo conditional mouse mutants of the Pten and mTOR gene will be generated.
The requirement for membrane-bound NRG1 in the PNS indicates that Schwann cell-axon contact is a prerequisite for myelination. A participant of this consortium has identified potential candidate cell adhesion molecules that mediate such axon-Schwann cell contact. These Nectin-like (Necl)/SynCAM cell adhesion molecules are located at the axon-glia interface along the internodes where they mediate Schwann cell-nerve interaction necessary for myelination. The role of these molecules and their interaction with NRG1 signals will be investigated.
The functional organization of myelinated axons
Myelinating glial cells and their underlying axons are engaged in a complex dialogue that is crucial for the normal physiological function of both cell types. An important consequence of this interaction is the accumulation of voltage-gated Na+ channels at short, regularly spaced interruptions in the myelin sheath known as the nodes of Ranvier. At the nodes, these channels are found in a multiprotein complex that includes cell adhesion molecules and cytoplasmic adaptor proteins. In peripheral myelinated axons Na+ channel clustering is controlled by a direct contact with Schwann cell microvilli that emanate from the outer collar of the cell. As demonstrated by participants 3 and 4, Schwann cell-specific ablation of dystroglycan and to a lesser extent of laminin ?1 cause disruption of microvillar organization and reduction in nodal Na+ channel clustering, suggesting that the microvilli play a direct role in node assembly. As shown by participants 2 and 4, Na+ channels are recruited to clusters of two cell adhesion molecules Nfasc186 and NrCAM that were first positioned by binding to gliomedin present on the glial. These observations led us to propose that the focal presentation of gliomedin to the axon during myelination causes the initial clustering of the axonodal CAMs into higher-ordered oligomers, which facilitates the recruitment of ankyrin G and Na+ channels. The participation of the groups of participant 2, 3 and 4 in this framework allows a unique opportunity to examine and advance our current knowledge of how Na+ channels are clustered in PNS nodes.
Multiple animal models of CMT neuropathies have been produced and validated as good functional and pathological models of the corresponding human diseases Pathogenetic mechanisms are beginning to be identified in these models, and preliminary studies suggest treatment strategies for CMT1A, due to peripheral myelin protein 22kD (PMP22) overexpression.
Whereas CMT1A accounts for approximately one half of hereditary neuropathies, other demyelinating neuropathies have been defined by mutations in 23 distinct genes. Many are dominantly inherited, and produce mutant proteins, suggesting toxic mechanisms. There are two fundamental questions: What range of toxic mechanisms produce such widely variable phenotypes and how do they interfere with myelination? Second, presuming a broad array of mutation-specific mechanisms, are there common downstream mechanisms amenable to therapeutic intervention?
We hypothesize that in neuropathy, toxic pathogenetic mechanisms in Schwann cells may target the axonal signals necessary for formation or maintenance of myelin. More specifically, activated protein quality control in Schwann cells, or inefficient proteasome function, may perturb the cell intrinsic pathways normally activated by axonal signals during development, or inappropriately reactivate them in adult myelin forming Schwann cells. We will test this hypothesis by:
1) providing further expression analysis of activation of these signals in our collective animal models;
2) genetically or pharmacologically reversing superactivation of these signals in CMT1A and CMT1B animals to provide direct evidence for their pathogenetic role;
3) cooperating with our SME partner 8 to develop a high content cell-based assay to identify small molecules that can reverse the inappropriate trafficking and activated protein quality control associated with CMT mutant proteins.
These aims represent an emerging direction in pathogenesis of hereditary neuropathy that exploits our collective animal models and complementary expertise. We expect a major advance beyond the current knowledge of pathogenesis in neuropathy and we expect to produce new animal models, information and small molecule tools that will inform pharmaceutical strategies to intervene where toxicity perturbs axoglial signals in neuropathy.
The S&T results of the NGIDD project has been described in period reports, scientific publications, conference contributions, contributions in civic debate, through our website and through a brochure.
What follows here is a description of the major achievements per workpackage.
WP1: Engagement: Glial cells and neurons
i. To identify the role of the newly identified Lgi4 in early glia-axonal contact.
ii. To determine the molecular mechanisms underlying radial axonal sorting.
WP2: Control of myelin formation and maintenance
i. Dissecting axon-glia interactions essential for myelination.
ii. Dissecting regulatory mechanisms downstream of Neuregulin-1/ErbB signaling.
iii. Understanding myelin outfoldings in mouse models of neurological disease.
WP3: Functional organization of the axo-glial unit
i. To determine how myelinating glial cells organize the axonal membrane they ensheath.
ii. To determine the molecular mechanisms underlying Na+ channel clustering at nodes.
iii. To identify novel paranodal axoglial junction proteins.
WP4: Disturbed axo-glial signaling and nerve pathology
i. Study axo-glia signals in our collective animal models of neuropathy.
ii. Establish a pathogenic role for these signals in our neuropathy mouse models.
iii. Develop a cell-based assay to identify small molecules that reverse or ameliorate the pathogenic mechanism of CMT mutant proteins.
Project Results:
WP1: Engagement: Glial cells and neurons
Activity 1.1
Previous studies suggest that proteins of the Adams family are receptors for Lgi proteins: Lgi1 and Adam22 co-precipitate from brain membrane preparations using PSD95 or Kv1.1 antibodies. Lgi1/Lgi4 bind to cells that ectopically express Adam22/23 and deletion of Adam22 presents a similar peripheral nerve phenotype to that of claw paw animals. We analysed expression of Lgi and Adam family members in the developing nerve and found that all members of the Lgi family are expressed in both neurons and Schwann cells. Out of the wide range of Adam family members that are expressed in peripheral nerve, we focused on the Adams closely related to Adam22: Adam23 and Adam11. All three genes are expressed in Schwann cells and neurons. To test whether binding of Lgi4 to Adam22 is direct and/or requires the presence of additional molecules at the cell membrane, we utilized fusion proteins that consist of the ectodomain of Adam22 (either normal Adam22 or Adam22 with a D509N mutation in the disintegrin-like domain that is analogous to the ju160 allele of C.elegans Adm1/Unc-71, which causes axon guidance and sex myoblast migration defects) and the human IgG Fc moiety. Adam22-Fc protein was found to efficiently precipitate Lgi4 from conditioned medium. In contrast, no Lgi4 protein was found precipitated with the Adam22-Fc protein containing the D509N mutation. These results indicate that binding of Lgi4 to Adam22 does not require additional membrane associated receptors and that binding depends on the integrin-binding domain of Adam22.
-Role of Lgi4 Adam interactions in nerve development
Adam22 is expressed by neurons as well as SCs raising the question whether Lgi4 functions in an autocrine or/and paracrine fashion. To answer this question, we generated primary sensory neuron cultures from Adam22+/neo and Adam22neo/neo embryos and seeded these cultures with rat SCs. Rat SCs readily ensheath and myelinate Adam22+/neo neurons, whereas Adam22neo/neo neurons are not myelinated. These results indicate that Adam22 function is required in the neuron, but do not exclude an additional function in SCs.
To further explore a possible SC-autonomous function of Adam22 and to corroborate its neuronal function in nerve development, we generated a conditional Adam22 allele and bred it to homozygosity in the presence of the Wnt1Cre transgene, thereby deleting Adam22 in all neural crest derivatives, including sensory neurons, endoneurial fibroblasts and SCs, but not in spinal motor neurons. The sciatic nerve is a mixed nerve containing both primary sensory and motor fibers. Analysis of myelin formation in the sciatic nerve of these mice revealed a mosaic pattern of normally myelinated fibers and promyelin arrested fibers. Most of the myelinated axons express Choline acetyltransferase (ChAT), indicating they are motor fibers. Therefore motor neurons are myelinated normally by Adam22-deficient SCs. As in Lgi4-deficient claw paw mice at P12, non-myelinated fibers in Wnt1:Adam22cKO/cKO sciatic nerves are associated with predominantly Oct6 positive promyelinating cells, whereas Krox20 positive SCs are invariably myelinating cells. Examination of the ventral motor and dorsal sensory roots revealed normal myelinated motor fibers and severely hypomyelinated sensory fibers respectively. These data demonstrate that Adam22 is required in neurons, whereas deleting Adam22 in SCs has no effect on myelin formation. Thus, axonally expressed Adam22 serves as the major receptor for SC-derived Lgi4 in PNS myelination.
Adam22 proteins have been shown to interact with integrins. As Lgi4 binds to the disintegrin-like domain of Adam22 it is possible that Lgi4 binding modulates (either positively or negatively) Adam22 integrin interactions. Indeed, Adam22 interacts with a number of integrins that are expressed on Schwann cells and are involved in axonal sorting and myelin stabilization. However, the actions of these integrins have been interpreted in terms of their interaction with the Schwann cell extracellular matrix and not with the axon.
In a collaborative effort between partners 2 and 1 and Dr. Rasband from Baylor College of Medicine, we found that ADAM22 is a component of the Kv1 channel complex and is found together with Caspr2 at the juxtaparanodal region in myelinated axons. Furthermore, we found that among all the known Kv1 channel interacting proteins, only ADAM22 is found at every site where Kv1 channels are clustered. Analysis of Caspr2, PSD-93, PSD-95, and double PSD-93/PSD-95 null mice showed ADAM22 clustering at BCTs requires PSD-95, but ADAM22 clustering at juxtaparanodes requires neither PSD-93 nor PSD-95. In direct contrast, analysis of ADAM22 null mice demonstrated juxtaparanodal clustering of PSD-93 and PSD-95 requires ADAM22, whereas Kv1.2 and Caspr2 clustering is normal in ADAM22 null mice. Thus, ADAM22 is an axonal component of the Kv1 K+ channel complex that recruits MAGUKs (i.e. PSD-93 and PSD-95) to juxtaparanodes. These findings are of particular importance as the excitable properties of neurons depend not only on the kinds of ion channels expressed in the plasma membrane, but also on the location of these channels. Among the many different ion channels expressed in the nervous system, the Kv1 channels are an excellent example of channels that are restricted to distinct subcellular locations. Mutations or diseases that disrupt clustering, localization, or composition of Kv1 channels severely compromise nervous system function and lead to conduction block, episodic ataxia, and/or epilepsies. A major question is whether the clustering of Kv1 and Adam receptors is dependent on binding of an Lgi ligand.
Lgi proteins are characterized by a typical extra-cellular Leucine-Rich Repeat (LRR) and a WDR5 related seven-bladed propeller structure called the epitempin (EPTP) domain. Both these structural motifs provide a platform for specific protein-protein interactions. As described above, all members of the Lgi family bind to a subfamily of Adam proteins eg Adam11, Adam22 and Adam23. These Adam proteins lack metalloproteinase activity and are predominantly expressed in the nervous system. How binding of Lgi4 to Adam22 results in proper axonal ensheathment and myelination is a major unresolved question in this field of research. We began to address this question by asking whether binding of an Lgi ligand to Adam22 is sufficient to promote myelination by Schwann cells. Using a well-defined embryonic dorsal root ganglion neuron Schwann cell co-culture system we demonstrated that Schwann cells will readily myelinate associated axons in wildtype or clp heterozygous cultures but that no myelin is formed in cultures derived from claw paw embryos. Viral transduction of Schwann cells with vectors expressing Lgi4, but not Lgi1, Lgi2 or Lgi3, restores myelin formation in claw paw cultures. Thus, despite the fact that all these proteins bind to Adam22 - although probably with different affinities - only Lgi4 elicits a myelination response. We next performed a number of experiments in which the LRR domain and the EPTP domain were exchanged between Lgi4 and Lgi3 or Lgi1. Again the chimaeric molecules were tested in the co-culture system and it was found that functional specificity resides in the EPTP domain of Lgi4 in particular in the last four blades of the beta propeller structure. Within this domain there is a short amino acid sequence that connects blade 5 and 6 and that is fully conserved in all Lgi proteins suggesting it might be important for Lgi function in general. Indeed, mutations in this region, especially the S440L mutation, in Lgi1 causes epilepsie and psychotic disorders in a Japanese family. To test whether this Serine residue is also essential for Lgi4 function we introduced this mutation in Lgi4 and tested the mutant protein in the co-culture system. We found that this mutation did not affect Lgi4 function. Thus, a crucial amino acid residue in Lgi1 function is not relevant for Lgi4 function suggesting they engage in different interactions. To further identify functionally relevant sequences within the Lgi4 protein we took advantage of the more than 900 million years of independent evolutionary history between modern day fishes and mammals. Curiously, comparative genomics had failed to identify an Lgi4 homologue among the five Lgi genes in the genomes of fishes (species examined; sea bass, salmon, pufferfish, medaka and zebrafish). We cloned the zebrafish Lgi cDNAs and expressed them in the claw paw co-culture system and found that only drLgi1A, and to a lesser extent drLgi1B, but not drLgi2A, drLgi2B and drLgi3, promotes myelination. Examination of a sequence alignment of the primary amino acid sequences of drLgi1A, drLgiB and Lgi4 revealed two short patches that are uniquely conserved between Lgi4 and drLgi1A and drLgi1B but not Lgi1. These patches map next to each other on the upper rim of the EPTP domain of Lgi4. We hypothesize that these short sequence patches represent unique interaction interfaces that distinguish Lgi4 from Lgi1. Whether this patch is involved in Adam22 interactions or perhaps with an accessory molecule is at present unclear and provides an important challenge for the future.
Activity 1.2
During peripheral nerve development Schwann cells (SCs) surround large bundles of mixed-calibre axons send cytoplasmic processes into the bundles, thereby pulling large calibre axons to the edge of the bundles (radial sorting). The sorted large axons are then surrounded in a 1:1 relationship with SCs, and are subsequently myelinated. The mechanisms by which molecules promote sorting are incompletely understood. Axonal sorting requires multiple events: the formation of families of SC with multiple axons contained in a common basal lamina, the matching of the number of axons and SC, the insertion of Schwann cell processes around axons to recognize and segregate large ones, and the defasciculation of single axons with their own daughter Schwann cell and basal lamina. It is unknown which laminin-receptor pairs contribute to each step. For example laminins promote the formation of Schwann cell processes and the interaction with axons, via activation of Rac1 by a beta1 integrin receptor, but laminins also promote Schwann cell proliferation and survival, via activation of PI3K, FAK and Cdc42 by an unknown laminin receptor. Thus which laminin receptors control matching of axons and SC number is unclear. B1 integrin receptors, but not dystroglycan, transmit survival and proliferation laminin signals. Laminins promote SC proliferation and survival and dystroglycan is required in vitro for laminin-mediated protection of SC from apoptosis. To determine which laminin receptors promote proliferation and survival in vivo, we measured the rate of apoptotic and proliferating cells in laminin receptors mutant nerves by TUNEL and anti-histone H3 staining. Proliferation was significantly decreased in beta1 integrins and double mutants at P3 and P5. Dystroglycan mutants showed increased proliferation at P15. Decreased proliferation rates for b1 integrin and double mutants at P3 were confirmed using BrdU incorporation. Apoptosis was increased in double mutants (beta1-integrin/dystroglycan) at P3 and P5. Staining with the SC marker periaxin showed that 80% of the TUNEL positive cells were SC. Single b1 integrin, but not single dystroglycan mutants also showed increased apoptosis. Thus only b1 integrin receptor(s) mediate the survival and proliferation signal from laminins during axonal sorting.
Radial sorting is also impaired in models of human diseases, congenital muscular dystrophy (MDC) 1A, MDC1D and Fukuyama, owing to loss-of-function mutations in the genes coding for laminin b2, Large or fukutin glycosyltransferases, respectively. It is not clear which receptor(s) are activated by laminin 211, or glycosylated by Large and fukutin during sorting. We identified the relevant receptor as dystroglycan and reported these findings in a paper in Development. We showed that the absence of dystroglycan in a specific genetic background causes sorting defects with topography identical to that of laminin 211 mutants, and recapitulating the MDC1A, MDC1D and Fukuyama phenotypes. By epistasis studies in mice lacking both integrin beta1 and dystroglycan in SCs, we show that absence of beta1 integrins impairs proliferation and survival, and arrests radial sorting at early stages, that beta1 integrins and dystroglycan activate different pathways, and that the absence of both molecules is synergistic. Thus, the function of dystroglycan and ?eta1 integrins is not redundant but is sequential. These data identify dystroglycan as a functional laminin 211-receptor during axonal sorting and the key substrate relevant to the pathogenesis of glycosyltransferase congenital muscular dystrophies.
How are these integrin signals transduced in sorting Schwann cells? FAK associates with beta1 integrin in Schwann cells and the presence of basal lamina activates FAK. Focal adhesion kinase (FAK) is a non-receptor tyrosine kinase central to several signal transduction pathways, including those that stimulate proliferation. Here we have investigated the role of Schwann cell FAK by targeted deletion using the Cre-loxP system.
During the sorting process, SCs undergo a burst of proliferation, ensuring that an adequate number of SCs are available to sort all the required axons, and we have shown that Focal Adhesion Kinase (FAK) is partially required for this proliferation. The onset of sorting coincides with the formation by SCs of a shared basal lamina, a major constituent of which is the extracellular matrix (ECM) protein laminin. Indeed, in the absence of laminin, the basal lamina is not formed, SC proliferation is reduced, and process extension and thereby radial sorting does not occur. Interestingly, at the onset of sorting, the SC basal lamina forms in a fragmentary manner; thus, by EM, at E18.5 in mice, in some areas of peripheral nerve the basal lamina appears to be fully-formed, whereas in others it is virtually absent. As ECM, via FAK signaling, can regulate cell migration and proliferation, and can also regulate differentiation, we have been trying to determine whether the SC phenotype is affected by large differences in ECM concentration, and whether FAK plays a role.
In summary, we have found that wildtype SC proliferation and differentiation is virtually unaffected over a relevant range of Laminin1 concentrations. In contrast, though FAK mutant SCs proliferated and differentiated normally at the highest Laminin1 concentration, at lower concentrations FAK mutant SCs failed to spread, did not proliferate and were hyper-sensitive to and differentiated prematurely in response to dbcAMP. These experiments suggest that the FAK mutant SCs are sensitive to cAMP because they do not spread on low Laminin1. Interestingly, in vivo around the time of radial sorting, FAK KO SCs prematurely upregulate the differentiation markers Oct6 and Krox20. This suggests that FAK-mediated SC spreading in vivo may regulate both proliferation and differentiation of SCs.
WP2: Control of myelin formation and maintenance
Activity 2.1
Nectin-like (Necl)/SynCAM cell adhesion molecules are located at the axon-glia interface along the internodes where they mediate Schwann cell-nerve interaction necessary for myelination. The role of these molecules and their interaction with NRG1 signals will be investigated. We took two approaches to study the role of Necl4 in myelination in vivo. In the first, we have generated and begun analyzing transgenic mice expressing a normal and a deletion mutant of Necl4 in Schwann cells and oligodendrocytes using an MBP promoter. Analysis of the transgenes showed that expressing the deletion mutant of Necl4 resulted in profound neurological defects and abnormal myelin formation. These are encouraging results, which support a role for Necl proteins in myelination. In the second approach, we used gene targeting to eliminate the expression of Necl4 in myelinating glial cells. Necl4 was completely absent from homozygous animals as was evident by Western-blot analysis of spinal cord and sciatic nerve lysates, as well as by immunofluorescence analysis of teased sciatic nerves and brain sections. Homozygous Necl4-/- mice were viable and fertile, but displayed mild motor abnormalities, enhanced limb-clasping reflexes, and reduced conduction velocity of peripheral nerves. Histological examination of sciatic nerve cross-sections prepared from homozygous animals at different ages revealed a marked delay in myelination and formation of abnormal myelin membrane detachments and outgrowth. Furthermore, we show that deletion of Necl4 resulted in a profound effect on the organization of myelinated axons - generation of split paranodal junction that was reflected by abnormal distribution of Caspr, aberrant clustering of Kv1.2 channels away from the juxtaparanodal region, and absence of defined juxtamessaxonal lines along the internodes. Our data suggest that axoglial contact, mediated by Necl4, is required for proper myelination and the molecular organization of myelinated axons.
How does Necl4 affect Schwann cell behavior? We have identified protein 4.1G as a cytoplasmic protein that interact with Necl4 in Schwann cells. Extensive immunolabeling experiments showed that these two molecules co-localized along the myelin internode. Furthermore, clustering of Necl4 on the surface of Schwann cells recruits protein 4.1G into the clusters, indicating that these proteins are associated at the plasma membrane interface. Analysis of mice lacking protein 4.1G revealed that these mice exhibit abnormal distribution of Necl4 and several other non-compact myelin proteins, further supporting a functional connection between these two proteins. Similar to ablating Necl4, deletion of 4.1G in mice resulted in aberrant distribution of both glial adhesion molecules and axonal proteins that were present along the internodes. In wild-type nerves, juxtaparanodal proteins (i.e. Kv1 channels, Caspr2, and TAG-1) were concentrated throughout the internodes in a double strand that flanked paranodal junction components (i.e. Caspr, contactin and NF155), and opposes the inner mesaxon of the myelin sheath. In contrast, in 4.1G-/- mice, these proteins ?piled up? at the juxtaparanodal region or aggregated along the internodes. These findings suggest that protein 4.1G contributes to the organization of the internodal axolemma by targeting and/or maintaining glial transmembrane proteins along the axoglial interface.
In our search for proteins that control myelin formation and maintenance, we concentrated on proteins that would translate extracellular signals to the movement of the actin cytoskeleton that, as we suspected, will be required for myelination. We have shown that N-WASP is required for membrane wrapping and myelination by Schwann cells. During peripheral nerve myelination, Schwann cells sort larger axons, ensheath them and eventually wrap their membrane to form the myelin sheath. These processes involve extensive changes in cell shape, but the exact mechanisms involved are still unknown. Neural Wiskott-Aldrich syndrome protein (N-WASP) integrates various extracellular signals to control actin dynamics and cytoskeletal reorganization through activation of the Arp2/3 complex. By generating mice lacking N-WASP in myelinating Schwann cells, we show that N-WASP is crucial for myelination. In N-WASP deficient nerves, Schwann cells sort and ensheath axons, but most of them fail to myelinate and arrest at the promyelinating stage. Yet, a limited number of Schwann cells form unusually short internodes, containing thin myelin sheaths, with the occasional appearance of myelin misfoldings. These data suggest that regulation of actin filament nucleation in Schwann cells by N-WASP is crucial for membrane wrapping, longitudinal extension and myelination.
Activity 2.2 and 2.3
Binding of NRG1 typeIII to glial ErbB2 and ErbB3 receptors is likely to activate multiple signaling pathways, but it is unknown how they are tied into the myelination program. One attractive second messenger is the PI3K pathway with the generation of phosphatidyl-inositol (3,4,5)-triphosphate ('PIP3') as a signaling lipid. To determine the role of PI3Kinase signaling in PNS myelination in vivo we analyzed a mouse line were the Pten gene (encoding a D3-lipid phosphatase that antagonizes PI3K signaling) was specifically ablated from the Schwann cell lineage by using a CnpCre driver line. We could demonstrate by PCR amplification of sciatic nerve genomic DNA that the Cre-mediated inactivation of Pten in Schwann cells resulted in efficient deletion of Pten exon 5 in mutant sciatic nerves. Western blot analysis confirmed a significant downregulation of PTEN protein and enhanced phosphorylation of known PI3 Kinase effector proteins, i.e. AKT and S6 in mutant sciatic nerves. When the levels of PtdIns(3,4,5)P3 and PtdIns(4,5)P2 were analyzed in acidic lipid extracts of sciatic nerves by gas-chromatographic quantification of their associated fatty acids we found, as predicted, an increase in PtdIns(3,4,5)P3 levels and a decrease in PtdIns(4,5)P2.
When small caliber axons (below 2 µm in diameter) of sciatic nerves were quantified by g-ratio analysis using ultra-thin cross sections, loss of PTEN caused a significant increase of myelin sheath thickness. However, when the degree of myelination was quantified for axons larger than 2µm (in the same cohort of mice), g-ratios of Pten mutants were unaltered compared to controls. Instead, we noticed that one very prominent feature was focal hypermyelination (Tomacula) and myelin outfoldings reminiscent of the neuropathological features in a diverse group of demyelinating neuropathies, including hereditary neuropathy with liability to pressure palsies (HNPP). Teased fiber preparations and the electron microscopic analysis of longitudinal sciatic nerve sections at postnatal day 30 (P30) revealed that these myelin abnormalities predominantly originated from paranodal regions and Schmidt-Lantermann incisures. These two regions of non-compacted myelin might represent zones of controlled membrane incorporation and myelin sheath growth. We therefore investigated whether the lack of Pten would mimic abnormal axon-glia signaling and PI3K activation in these regions. Indeed, when p-AKT was used as a sensor of PtdIns(3,4,5)P3, and co-immunostained with MPZ (a marker of compact myelin) and myelin-associated glycoprotein (MAG; a marker of non-compacted myelin), we found a striking colocalization of p-AKT in MAG-positive Schmidt-Lantermann incisures and paranodes. This indicates that in those regions PTEN could be involved in regulating PtdIns(3,4,5)P3 levels. Levels of the two cell polarity proteins Par3 and Dlg1, normally located at regions of non-compact myelin, and suggested to play a role in the control of myelination, were unchanged when tested by immunohistochemistry on teased fibers.
WP3: Functional organization of the axo-glial unit
Activity 3.1
A major consequence of myelination is the establishment of axonal domains that restrict sodium and potassium channels distribution. We investigated the mechanism through which these domains are established and what molecules play a role in this process.
Heminodal clustering of Na+ channels requires Gliomedin - To examine the role of gliomedin in the assembly of the nodes of Ranvier, we generated gliomedin null mice. Homozygous gldn-/- mice exhibited no overt neurological abnormalities, formed compact PNS myelin that was indistinguishable from their wild type littermates, but displayed disorganization and impaired attachment of Schwann cell microvilli to the nodal axolemma. As Schwann cell microvilli were implicated in node formation, we examined the distribution of Na+ channels in the developing sciatic nerves of wild type and gldn-/- at P6, when both nodes and heminodes can be detected. We found that while in wild type nerves Na+ channels were clustered in mature nodes (i.e. nodes that are flanked by two adjacent Caspr-labeled PNJs), as well as in heminodes, in gldn-/- nerves these channels were absent from heminodes and only detected in mature nodes. In adult sciatic nerves of gldn-/- mice, Na+ channels were colocalized at the nodes of Ranvier with NF186, NrCAM, ankyrin G and ?IV spectrin, Caspr was present at the PNJ and Kv1 channels were localized at the juxtaparanodal region similar to wild type nerves. As an additional experimental approach, we used myelinating Schwann-dorsal root ganglion (DRG) neurons cultures, which enable monitoring the detailed molecular assembly of the nodes in PNS axons. In contrast to wild type cultures, Na+ channels, NF186, as well as ankyrin G and IV spectrin were not clustered at heminodes in gldn-/- myelinating cultures. Notably, while both in vivo and in myelinating cultures of wild type nerves, heminodes appeared adjacent to Caspr-labeled PNJs, in gldn-/- nerves, heminodes were absent near Caspr-positive sites, indicating that heminodal clustering is not the result of PNJ formation. We then counted the number of sites in which Na+ channels were clustered near the edge of a myelin segment. In gldn-/- the majority of Na+ channels were either absent or diffused (gldn-/- culture 92%, nerve 81%), compared to wild type nerves where they were clustered in more than 90% of the sites. These results demonstrate that gliomedin is required for clustering of the nodal complex at heminodes during myelination.
NF186 and NrCAM are required for clustering of Na+ channels at heminodes but not for their accumulation in mature nodes - The observation that clustering of the nodal complex requires gliomedin prompted us to examine whether this activity is mediated by its two axonodal receptors, NrCAM and NF186. We first examined sciatic nerves of nrcam-/- mice, in which Na+ channel clustering in PNS nodes is delayed. Immunolabeling of teased sciatic nerves isolated from 6 day-old (P6) mice using antibodies to Na+ channels and Caspr (to label the PNJs) showed that Na+ channels were not clustered at heminodes that were already bordered by Caspr-labeled PNJs. In addition, Na+ channels, NF186 and gliomedin were all detected at heminodes in myelinating DRG cultures derived from wild type, but not from nrcam-/- mice. In both sciatic nerves and myelinated cultures, Na+ channels, NF186, gliomedin, ankyrin G and ?IV spectrin were accumulated at mature nodes. Thus, similar to gliomedin, NrCAM is required for heminodal clustering but not for the accumulation of Na+ channels at mature nodes of Ranvier. Previous observations by UEDIN showed that genetic ablation of the neuronal (NF186) and glial (NF155) isoforms of neurofascin leads to the disruption of both PNJs and nodes. In line with these findings, nodes fail to form in myelinating co-cultures of DRG neurons and Schwann cells isolated from nfasc-/- mice. To examine directly the role of NF186 in the assembly of PNS nodes, we determined whether Na+ channels accumulate at nodes in the absence of NF186 by co-culturing DRG neurons isolated from nfasc-/- mice with wild type Schwann cells. The expression of NF155 in myelinating Schwann cells promoted the full recovery of the PNJs as demonstrated by immunolabeling for Caspr. PNJ formation in these axons was accompanied by the accumulation of Na+ channels at nodes together with all components of the nodal complex (i.e. ankyrin G, ?IV spectrin and NrCAM) except NF186. However, as observed in gliomedin and NrCAM-deficient DRG cultures, in the absence of axonal NF186, Na+ channels were not clustered at heminodes. To further determine the contribution of the axonodal CAMs to the formation of mature nodes, we established myelinating cultures using DRG neurons isolated from double nrcam-/-/nfasc-/- mutant mice and wild type Schwann cells (to recover the PNJs). Analysis of these cultures showed that Na+ channels, as well as ankyrin G and ?IV spectrin, accumulated at nodes that were flanked by PNJs, but not at heminodes. Taken together, these results demonstrate that gliomedin and its axonodal receptors NrCAM and NF186, are all required for developmental clustering of Na+ channel at heminodes. They further reveal the existence of an additional mechanism that enables the accumulation of Na+ channels in mature nodes independently of heminodal clustering and the axonodal CAMs.
Involvement of the PNJ in node formation - One possible mechanism that could compensate for the lack of heminodal clustering might be provided by the PNJ, which serves as a membrane barrier separating nodal Na+ channels and juxtaparanodal K+ channels. However, the role of the PNJ in node formation remains unclear since several studies using paranodal mutant mice have indicated that these structures are dispensable for PNS node formation. Myelinating DRG cultures lacking gliomedin, NrCAM or NF186, in which heminodes fail to form, allowed us to directly examine the role of the PNJ in the assembly of PNS nodes. We found that in wild type cultures Na+ channels were clustered at heminodes, which eventually merged, forming a mature focal node of Ranvier. In contrast, in gldn-/- and nrcam-/- cultures, as well as in cultures containing nfasc-/- neurons and wild type Schwann cells (i.e. thus considered as nf186-/-) Na+ channels were spread between the two growing myelin segments. A similar distribution of Na+ channels was detected in developing sciatic nerves of gldn-/- and nrcam-/- nerves. Interestingly, although Na+ channels co-clustered with NF186 during all developmental stages in wt cultures, NF186 was not detected in gldn-/- and nrcam-/- cultures until these channels were already accumulated at focal nodes, further indicating that accumulation of Na+ channels in mature nodes occurs independently of the axonodal CAMs. Immunolabeling of gldn-/-, nrcam-/- and nf186-/- cultures using antibodies to Caspr, Na+ channels and MBP, showed that Na+ channels were eliminated from the area occupied by the PNJ, but invaded the axolemma underneath the myelin sheath in its absence. These observations suggest that in addition to being actively clustered at heminodes, Na+ channels are constrained between two neighboring myelin segments by the PNJ. Furthermore, in the absence of the PNJ, as found in mice lacking Caspr for example, the presence of gliomedin and the axonodal CAMs enable the clustering of Na+ channels at heminodes and consequently in mature nodes.
Activity 3.3
The functional organization of the axo-glial unit has been remarkably well conserved during evolution. The main molecules involved and the underlying molecular mechanisms are even conserved in simple model animals such as Drosophila. Drosophila offers a large number of experimental advantages that make this system very attractive to identify additional gene functions mediating or controlling neuron-glia interaction in the nervous system. One of the key approaches towards this aim resides in cell-type specific gene silencing. We have generated a number of Gal4 driver lines, which now allow activating gene expression in almost every glial cell type of Drosophila. In combination with large collections of transgenic lines established in Vienna, Austria, Harvard, USA, or Kyoto, Japan, that contain constructs which allow a Gal4 dependent expression of double stranded RNA, cell-type specific RNA interference (RNAi) can be induced. Such genome wide RNAi screens will eventually unravel most of the genes required for glial differentiation.
One of the main aims of the proposal was to identify novel genes acting in Drosophila glial cells. In flies just as in primitive vertebrate fish the blood brain barrier is established by glial cells, which form extensive cellular junctions characterized by septate junctions. In the mammalian nervous system, very similar junctions are established at the paranodal junctions. In both systems, the core components of these junctions are NeurexinIV, (the mammalian homolog is the Caspr protein) and Neuroglian (the mammalian homolog is the Neurofascin155 protein). This and other cases of functional homology suggest that Drosophila can serve as a model system to identify novel components underlying neuron-glial interaction.
In the funding period we have followed two different experimental avenues to reach these goals. On the one hand we continued our efforts with direct genetic screens for mutants affecting the integrity of the blood brain barrier. On the other hand we silenced individual genes specifically in glial cells in a RNA-interference based approach and thus determined genes required for glial cell development. Whereas the first approach directly pointed towards a gene function, the second approach just gave an indication whether or not a particular gene has any role in glial cell development and thus requires additional genetic work.
WP4: Disturbed axo-glial signaling and nerve pathology
Activity 4.1
Tomacula and myelin outfoldings are a striking neuropathological feature of a diverse group of demyelinating neuropathies, including Charcot-Marie-Tooth disease 4B (CMT4B) and hereditary neuropathy with liability to pressure palsies (HNPP). Despite major progress in defining the genetic basis of these diseases, the molecular mechanisms that cause focal myelin growth and disrupt Schwann cell-axon interaction in some forms of CMT, but not in others, have remained obscure. Partner 5 has hypothesized that myelin outfoldings and "tomacula" result from uncontrolled myelin membrane growth, a developmental program that is at least in part controlled by neuregulin-1/ErbB2 and PI3K signaling. Partner 5 found in mice that lack Pten expression in Schwann cells, that glia cell-specific hyperactivation of the endogenous PI3K pathway causes focal hypermyelination, myelin outfoldings, and tomacula, associated with a progressive peripheral neuropathy. In previous experiments (see WP 2) Partner 5 found activated AKT kinase associated with phosphatidylinositol-3,4,5-trisphosphate (PtdIns(3,4,5)P3) at paranodal loops and Schmidt-Lantermann incisures of Pten mutants. This and the focal hypermyelination pathology indicate that regions of non-compact myelin permit membrane incorporation and myelin growth. We suggest a model that elevated PtdIns(3,4,5)P3 levels contribute to the pathology of tomaculous neuropathies at least partially by activation of the AKT/mTOR pathway. Indeed, a pharmacological reduction of aberrant mTOR activation by rapamycin, a specific mTOR inhibitor, prevented aberrant myelin growth in Pten conditional mutant nerves to a significant extent.
To analyze whether inhibition of mTOR signaling could prevent the formation of myelin abnormalities Partner 5 treated Pten mutants and littermate control mice with rapamycin. We selected 3 groups of abnormal myelin formations by histological criteria (tomacula, comma-shaped outfoldings, recurrent loops and quantified them in sciatic nerve cross sections of controls and conditional Pten mutants that were treated with vehicle (-Rap) or rapamycin (+Rap) from postnatal day (P) 12 to P43 (Del 4.8). In the vehicle treated groups, as expected, all myelin abnormalities were higher in Pten mutants than in controls. After 31 days of rapamycin treatment tomacula were 75% less frequent in the drug treated mutants (52 ± 12.87 n=8) when compared to vehicle treated mutants (209 ± 20.76 n=6) (p<0.0001). Similarly, recurrent loops were reduced by 75 % in the rapamycin treated mutants (17.88 ± 3.393 n=8) when compared to vehicle treated mutants (72.17 ± 7.897 in Pten mutants, n=6) (p<0.0001). Comma-like outfoldings were reduced by 49% (60.13 ± 13.37 n=8 in treated Pten mutants vs. 117.2 ± 6.882 n=6 in Pten mutants) (p=0.0051). Rapamycin treatment had no visible effect on control mice. Thus, although the growth abnormalities of the myelin sheaths were not completely corrected, the results indicate that enhanced mTOR signaling plays a role in the formation of redundant myelin in the Pten mutants.
Activity 4.2
The proper cellular responses following nerve trauma, in particular demyelination, require the activity of the transcription factor c-jun. We asked whether the removal of c-jun in a neuropathy model could ameliorate the clinical symptoms in this model. To this end P0CreXcJunflox/C3 mice were created together with suitable P0CreXcJunflox and C3 control mice. This was done to investigate whether the elevated c-Jun expression detected in C3 mice (which over express PMP22 and serve as a model of CMT1A neuropathy) was deleterious for C3 pathology (as predicted by the ability of c-Jun to suppress myelin differentiation) or helpful (as predicted by the ability of c-Jun to activate a neuron-supportive Schwann cell program during mechanical nerve injury).
The P0CreXcJunflox/C3 mice (that over express PMP22 but lack c-Jun) were subjected to morphometric nerve analysis and an extensive behavioral testing of sensory and motor performance. The results were compared to the same tests carried out on P0CreXcJunflox and C3 control mice.
The behavioral tests included, beam walking, grid walking, hanging wire grip strength test, Rotarod performance, von Frey hair sensory test, hotplate test and footprint analysis. It was found that deletion of c-Jun in PMP22 over expressing mice resulted in a clear worsening of motor and sensory motor performance. The outcome of pure sensory tests (von Frey hair sensory test, hotplate test) did not change. It should be noted that most/all motor tasks involve the function of sensory neurons, in particular sensory afferents from muscle spindles.
Electrophysiological measurements showed a strong reduction in F-wave persistence in P0CreXcJunflox/C3 mice compared to C3 controls. This indicates reduced excitability of spinal motoneurons, which in turn could be caused by reduced input from sensory neurons innervating muscle spindles.
Finally, a correlated study had provided similar evidence that dedifferentiation may be beneficial, not detrimental, to myelinating S63del (CMT1B) Schwann cells, in which the unfolded protein response is pathogenic. Two independent approaches had suggested that persistent expression of genes that limit differentiation, in this case Sox2 and Id2, in Schwann cells improves both plasticity after injury and limits proteo-toxic stress in neuropathy. To eliminate the possibility that compensation by either factor in the absence of the other produced this paradoxical conclusion, we examined S63del nerves in which both Sox2 and Id2 were ablated in Schwann cells. Again, the ablated nerves were more pathological than S63del with Id2 and Sox2 intact, demonstrating that the two factors are beneficial in S63del neuropathy. It is likely that these factors promote plasticity of Schwann cells in injured nerves, and this plasticity is advantageous in neuropathy. A similar conclusion was drawn based on studies of the role of the dedifferentiation factors cJun and Notch after nerve injury and in neuropathy.
Activity 4.3
Develop a cell-based assay to identify small molecules that reverse or ameliorate the pathogenic mechanism of CMT mutant proteins.
NGIDD has constructed the Xbp-1unspliced-EGFP reporter construct and performed titration assays with G418 in transfected Cos-7 cells. We have established that there is not obvious cell death in Cos-7 cells due to transient transfection of the P0S63del-dsRED or P0wt-dsRED. Based on these collective data, we performed a permanent transfection in Cos-7 cells with the Xbp-1unspliced-EGFP reporter construct, in order to generate the UPR reporter cell line (Del 4.11). After two weeks of selection in 700 μ g/ml G418, a limiting dilution, at 1, 5 or 10 cells/well, in ten 96 MTPs, was performed. The pool of G418 resistant cells was analyzed at fluorescence microscope, upon 18 hrs. stimulation with 10µM Tunicamycin (to induce a UPR and Xbp1 splicing, and hence GFP expression).
When almost all of the wells were confluent, plates were replicated into 384 MTPs, in order to obtain a quadruplicate for each single clone. The first so called clone pool analysis was carried out on these plates, on uncounted cells. Cells were incubated in complete medium with or without 10 μ M Tunicamycin for 18 hrs. Plates were analysed at FLIPRTETRA (MDS) and emitted fluorescence recorded. Positive clones were selected on the basis of the highest EGFP dependent fluorescence signal detected. A total of 23 clones, deriving from limiting dilution at 10 cells/well, were selected, picked up and seeded in 6 wells plate for cell amplification. After clone selection, Cos-7 reporter cells were maintained in complete medium supplemented with 500 μ g/ml G418. Unfortunately, as these clones were expanded and carried longer in culture, most of them died. We suspect toxicity in the cells permanently transfected with the Xbp-1unspliced-GFP reporter construct. They die perhaps due to chronic, low-level, leaky expression of GFP in the permanently transfected Cos-7 cells (a basal level of Xbp-1 splicing in Cos-7 cells is observed and may activate GFP expression even in the absence of ER stress). Currently 4 remaining reporter clones are in culture, to be tested in further experiments.
To avoid the problem of Xbp1-EGFP fusion protein toxicity, we have engineered two alternative plasmids based on firefly luciferase reporter genes. First, using the single cistron with IRES strategy, we have completed construction of the alternative Xbp-1unspliced-firefly luciferase-IRES-renilla luciferase plasmid and will perform permanent transfections in Cos-7 cells as before (Del 4.11). Performing the same analysis as outlined above for Xbp-1EGFP cells, we found only very low level activation of luciferase after tunicamycin treatment to induce UPR in Xbp1-luciferaseIRESrenilla cells.
Based on the results obtained, two future directions can be envisaged:
Results obtained upon screening the LOPAC compound collection on the UPR cell lines allow the identification of some promising hits. Therefore next steps would be the further validation of these hits and the interpretation by a biological proof of concept, using proof-of-principle in vitro assays and in vivo models. Anyway it has to be underlined that these hits won't represent per se lead compounds for med-chem programs, but the investigation on their activity on secondary assays might additionally confirm the validity of the overall system. Treatment of myelinating explant cultures from S63del mice with these hit compounds is underway to provide just this validation.
The development of a stable cell line expressing P0S63Del together with an appropriate reporter gene, suitable for the analysis of compounds interfering with UPR mechanisms and signaling represents a big success, since so far only transiently transfected cells have been produced. The screening of the LOPAC collection proved that the assay is very robust and suitable for running true screening campaigns. Therefore the next steps would imply the screening of larger compound collections, in order to identify hits that might indeed represent lead compounds for future med-chem programs.
Potential Impact:
NGIDD has contributed to a better understanding of nervous system function and dysfunction in disease. Furthermore, our work on patho-mechanism of neuropathy directly feeds into industrial development and new therapies and contributes to improved diagnosis and management of diseases of the nervous system.
Primary neurological diseases of peripheral nerves cause significant disability. In addition, the developmental and pathogenic processes in the PNS are of direct relevance for our understanding of CNS processes. Due to the simpler cell structure and accessibility of the PNS, many of the molecular mechanisms governing CNS development and pathology have been first elucidated in the PNS. Our work encompasses aspects of both normal development and pathology to find relevant pathogenetic mechanisms accessible to therapeutic intervention. On the one hand, several of the developmental mechanisms that we have elucidated are arrested or impaired in genetic diseases. Examples are lack of axonal sorting and Na+ channel clustering in laminin deficiencies, myelin protein over-expression in CMT1A, and myelin outfolding in CMT4B and HNPP. On the other hand, most mechanisms used during development to form functional myelinated fibers are recapitulated during repair after common multifactorial or acquired neurological diseases, such as multiple sclerosis, diabetes, spinal cord and nerve injury and inflammatory neuropathies. Finally, it is becoming clear that many dominant diseases in the nervous system act through toxic mechanisms, such as protein misfolding and dysregulation of quality control. Our studies of these processes in peripheral neuropathies ha?e identified pathogenetic mechanisms and therapeutic targets relevant to other common diseases of the nervous system, such as neurodegenerative diseases.
We have developed a high throughput cell-based assay that reports the early events of maladaptive stress response. We have validated this approach and we will use this cell based assay to identify small molecules useful as tools in proof of therapeutic principle studies in our animal models of neuropathy, and form the basis for medicinal chemistry to develop therapeutic agents that can reduce the toxicity of the integrated stress response. Our data indicate that modulation of such pathways in neuropathy (and other misfolded protein disease) patients may diminish both the severity of their disease, and the resulting disability. Since conformational or protein misfolding diseases include for example the more common Alzheimer and Parkinson diseases or Diabetes, the information produced by our experiments may find broader application in other diseases. (Ron, 2002; Lindsten et al., 2003)
Finally, one of the central objectives of this project was to identify common downstream pathogenetic mechanisms and their targets in order to find a common therapeutic strategy for the wide array of neuropathies resulting from mutations in 23 genes. Altered trafficking of mutant proteins and activation of protein quality control mechanisms (unfolded protein response, proteasome degradation, lysosomal-based autophagy) are common features of hereditary neuropathies. We have focused our attention on how these features may perturb signal pathways in glia downstream of developmentally important axonal to glia interactions. Thus, what we have learned about this toxicity in our CMT animal models will likely be applicable to many neuropathies. Indeed, we have shown altered PIP3/AKT signaling and neuropathy in PTEN mutant mice resembling myelin outfoldings as in the rare CMT4B neuropathy. We demonstrated that rapamycin treatment meliorates myelin outfoldings in our animal model. Such myelin abnormalities are present in a much broader range of neuropathies, including the more common HNPP. Thus our results suggest that rapamycin treatment might be beneficial in a range of neuropathies.
Project website: http://www.ngidd.eu(s’ouvre dans une nouvelle fenêtre)
Contact details coordinator:
dies Meijer, PhD
professor in Developmental Neurobiology
department of Genetics
erasmus University Medical Center
street address; Dr Molewaterplein 50
3015GE Rotterdam
postal address: PO box 2040
3000CA Rotterdam
tel +31-(0)10-7044250
d.meijer@erasmusmc.nl
contact detail project manager:
rini de Crom, PhD
managing Director
FP7-CP NGIDD
erasmus MC - Dept. Cell Biology and Genetics
s-Gravendijkwal 230
3015 CE Rotterdam
the Netherlands
phone: +31-107-043459
fax: +31-107-044743
ngidd@erasmusmc.nl
Myelin disorders present a significant health burden and are an important cause of disability. Myelin is laid down in segments around a majority of axons and contributes to the normal physiology of the nervous system. Schwann cells and oligodendroglia cells are the myelin-forming cells in the peripheral and central nervous system respectively and the formation and maintenance of myelin results from a continuous molecular dialogue with the neuronal axon. Despite the importance of myelin for the normal function of the nervous system, relatively little is known about the molecular mechanisms that underlie axon-glia communication in health and disease. Consequently, the therapeutic options for these disorders are very limited. The NGIDD project aimed to bridge the gap in our knowledge and identify new drug targets to stimulate the regenerative capacity of the nervous system.
To realize this overall aim, the members of the NGIDD consortium addressed the following questions related to; 1 the initial engagement of glial cells with axons, 2 the control of myelin formation and maintenance, 3 the functional organization of the axo-glial unit and 4, the pathology of de- and dismyelination.
Over the last four years, the NGIDD consortium has made major steps towards answering these questions. NGIDD has identified a completely novel molecular interaction consisting of the Schwann cell derived protein Lgi4 and its axonal receptor Adam22. NGIDD has elucidated the role of integrins, focal adhesion kinase and N-WASP in myelin formation and axonal sorting. Furthermore NGIDD has generated a range of mouse models of peripheral neuropathies and has provided evidence that the drug rapamycin is a beneficial modulator of myelin abnormalities in these mouse models. Additionally, NGIDD has generated and validated a cell-based assay to screen drugs that ameliorate the pathogenic effects of maladaptive unfolding protein responses that plays a role in a number of inherited neuropathies.
Thus, scientific and technological progress made by the NGIDD consortium has opened new avenues of inquiry to further our understanding of the molecular and cellular interactions that shape our myelinated nerves and govern their physiology. Importantly, this research has already yielded new lead compounds and preclinical models to explore new therapeutic approaches to combat the devastating neurological effects in de- and dismyelinating diseases in humans.
The NGIDD consortium united scientists from five different countries with different scientific expertise into a common cause. The consortium has contributed to the exchange of ideas, reagents and personnel, has trained a large number of scientists and has participated in scientific and civic conferences to disseminate the novel findings of the consortium, thus successfully contributing to the larger European Research Area.
Project Context and Objectives:
Although for many years it has been appreciated that myelinating glial cells and their underlying axons are engaged in a complex dialogue which is crucial for the normal function of both cell types, the identity of the molecules that mediate this interaction remained almost completely unknown. Studies by members of the consortium contributed significantly to our understanding how myelinating glial cells communicate with the axons they ensheath. The studies carried out in this collaborative effort provide important new insights on the molecular mechanisms that control myelination and the functional organization of the axon. Uncovering the mechanisms by which myelinating glial cells recognize and ensheath their associated axons is highly significant given the high prevalence of human diseases resulting from demyelination or alteration in myelin, such as multiple sclerosis and a number of inherited neuropathies.
Many developmental mechanisms are recapitulated after injury and pathology in these diseases, and can be exploited to promote regeneration, remyelination and re-establishment of functional axo-glial units. Furthermore, understanding the molecular basis for axo-glial contact is of particular interest as it is becoming increasingly evident that although in many of this disease the first pathological event is the disruption of myelin, secondary changes in the axon, which eventually lead to axonal loss is the main cause of clinical impairment. In summary, the present collaboration aimed to uncover the molecular mechanisms that govern the Schwann cell-axon engagement, myelination, and the establishment of distinct structural domains, all of which are processes that are crucial for the generation of a fully functional myelinated nerve. Increased knowledge about these mechanisms will prove useful in the design of therapeutic strategies for many neuropathies and demyelinating disorders.
The collaborative project NGIDD consisted of seven European research groups and one biotech company Axxam. The specific expertise of the individual members was mobilized to address four key objectives. These key objectives aligned with the four workpackages in this collaborative effort under one management (workpackage 5) to realize the main objective of this project; to push our understanding of myelination and myelination-related diseases beyond the state of the art and to identify new targets for pharmaceutical intervention to aid the regenerative capacity of the nervous system.
Initial events leading up to myelination
The glial cells and neurons of the peripheral nervous system are in close apposition and interdependent from the moment they are generated from the neural crest during embryonic development. Schwann cell precursors proliferate and migrate along the outgrowing axonal bundles to give rise to the mature axon-glia unit that characterize the fully developed and functional nervous system.
One crucial step in achieving correct axo-glia relationships in peripheral nerve development is radial axonal sorting. This developmental step allows Schwann cells to achieve a 1:1 relationship with large axons destined to be myelinated. Radial sorting is a pre-requisite for myelination, and is impaired in a common form of hereditary human neuropathy that accompanies congenital muscular dystrophy 1A. Sorting processes are also recapitulated during nerve repair. Thus, to clarify the mechanisms governing axonal sorting is of utmost importance not only to understand normal nerve development, but also to promote therapeutic intervention in the context of hereditary diseases and nerve repair. Therefore, an important question addressed in this context is how molecules such as extracellular matrix laminins and their integrin receptors, together with Focal Adhesion Kinase and the small GTPase Rac1 enable radial sorting of axons.
Many other players in radial sorting remain to be identified. Work on the natural mouse mutant Claw paw (clp) has identified one such player. Delayed axonal sorting and abnormal myelination of peripheral nerve fibers characterize Claw paw mice. It was found that a mutation in the Lgi4 gene causes the defects in nerve development in these mice. Lgi4 is a member of a small family of genes that encode secreted Leucine-rich proteins of unknown function. Lgi1, the founding member of this family, is frequently mutated in autosomal dominant partial epilepsy with auditory features (ADPEAF). How Lgi4 contributes to sorting and myelination is not known. An important insight into the possible mechanism of action of Lgi4 was obtained during this project through the identification of the receptor for Lgi4.
Control of myelin formation
It has been known for some time that, in postnatal development, axons themselves induce the associated Schwann cells to make myelin. However, the nature of this axonal signal was unclear until recently when a member lab of the consortium showed that neuregulin-1 (NRG1) type III is a critical growth factor of axons regulating myelination by Schwann cells in a quantitative fashion, thereby optimizing myelin sheath thickness. Binding of NRG1 to glial ErbB2 and ErbB3 receptors (via the EGF-like signaling domain common to all NRG1 isoforms) is likely to activate multiple signaling pathways, but it is unknown how these are tied into the myelination program. One attractive second messenger for myelination, which is activated by NRG1-ErbB signaling, is the PI3K pathway with the generation of phosphatidyl-inositol (3,4,5)-triphosphate (?PIP3?) as a signaling lipid. Activation of PI3K has been shown to modulate Schwann cell growth and myelination in vitro, but the role of PIP3 in myelination in vivo was still unclear. PIP3 is an activator of Akt1 and the mTOR pathway in RNA transcription and translation, but is also a subcellular activator of rac/cdc42-dependent process formation. To address the function of glial PI3K and mTOR signalling in vivo conditional mouse mutants of the Pten and mTOR gene will be generated.
The requirement for membrane-bound NRG1 in the PNS indicates that Schwann cell-axon contact is a prerequisite for myelination. A participant of this consortium has identified potential candidate cell adhesion molecules that mediate such axon-Schwann cell contact. These Nectin-like (Necl)/SynCAM cell adhesion molecules are located at the axon-glia interface along the internodes where they mediate Schwann cell-nerve interaction necessary for myelination. The role of these molecules and their interaction with NRG1 signals will be investigated.
The functional organization of myelinated axons
Myelinating glial cells and their underlying axons are engaged in a complex dialogue that is crucial for the normal physiological function of both cell types. An important consequence of this interaction is the accumulation of voltage-gated Na+ channels at short, regularly spaced interruptions in the myelin sheath known as the nodes of Ranvier. At the nodes, these channels are found in a multiprotein complex that includes cell adhesion molecules and cytoplasmic adaptor proteins. In peripheral myelinated axons Na+ channel clustering is controlled by a direct contact with Schwann cell microvilli that emanate from the outer collar of the cell. As demonstrated by participants 3 and 4, Schwann cell-specific ablation of dystroglycan and to a lesser extent of laminin ?1 cause disruption of microvillar organization and reduction in nodal Na+ channel clustering, suggesting that the microvilli play a direct role in node assembly. As shown by participants 2 and 4, Na+ channels are recruited to clusters of two cell adhesion molecules Nfasc186 and NrCAM that were first positioned by binding to gliomedin present on the glial. These observations led us to propose that the focal presentation of gliomedin to the axon during myelination causes the initial clustering of the axonodal CAMs into higher-ordered oligomers, which facilitates the recruitment of ankyrin G and Na+ channels. The participation of the groups of participant 2, 3 and 4 in this framework allows a unique opportunity to examine and advance our current knowledge of how Na+ channels are clustered in PNS nodes.
Multiple animal models of CMT neuropathies have been produced and validated as good functional and pathological models of the corresponding human diseases Pathogenetic mechanisms are beginning to be identified in these models, and preliminary studies suggest treatment strategies for CMT1A, due to peripheral myelin protein 22kD (PMP22) overexpression.
Whereas CMT1A accounts for approximately one half of hereditary neuropathies, other demyelinating neuropathies have been defined by mutations in 23 distinct genes. Many are dominantly inherited, and produce mutant proteins, suggesting toxic mechanisms. There are two fundamental questions: What range of toxic mechanisms produce such widely variable phenotypes and how do they interfere with myelination? Second, presuming a broad array of mutation-specific mechanisms, are there common downstream mechanisms amenable to therapeutic intervention?
We hypothesize that in neuropathy, toxic pathogenetic mechanisms in Schwann cells may target the axonal signals necessary for formation or maintenance of myelin. More specifically, activated protein quality control in Schwann cells, or inefficient proteasome function, may perturb the cell intrinsic pathways normally activated by axonal signals during development, or inappropriately reactivate them in adult myelin forming Schwann cells. We will test this hypothesis by:
1) providing further expression analysis of activation of these signals in our collective animal models;
2) genetically or pharmacologically reversing superactivation of these signals in CMT1A and CMT1B animals to provide direct evidence for their pathogenetic role;
3) cooperating with our SME partner 8 to develop a high content cell-based assay to identify small molecules that can reverse the inappropriate trafficking and activated protein quality control associated with CMT mutant proteins.
These aims represent an emerging direction in pathogenesis of hereditary neuropathy that exploits our collective animal models and complementary expertise. We expect a major advance beyond the current knowledge of pathogenesis in neuropathy and we expect to produce new animal models, information and small molecule tools that will inform pharmaceutical strategies to intervene where toxicity perturbs axoglial signals in neuropathy.
The S&T results of the NGIDD project has been described in period reports, scientific publications, conference contributions, contributions in civic debate, through our website and through a brochure.
What follows here is a description of the major achievements per workpackage.
WP1: Engagement: Glial cells and neurons
i. To identify the role of the newly identified Lgi4 in early glia-axonal contact.
ii. To determine the molecular mechanisms underlying radial axonal sorting.
WP2: Control of myelin formation and maintenance
i. Dissecting axon-glia interactions essential for myelination.
ii. Dissecting regulatory mechanisms downstream of Neuregulin-1/ErbB signaling.
iii. Understanding myelin outfoldings in mouse models of neurological disease.
WP3: Functional organization of the axo-glial unit
i. To determine how myelinating glial cells organize the axonal membrane they ensheath.
ii. To determine the molecular mechanisms underlying Na+ channel clustering at nodes.
iii. To identify novel paranodal axoglial junction proteins.
WP4: Disturbed axo-glial signaling and nerve pathology
i. Study axo-glia signals in our collective animal models of neuropathy.
ii. Establish a pathogenic role for these signals in our neuropathy mouse models.
iii. Develop a cell-based assay to identify small molecules that reverse or ameliorate the pathogenic mechanism of CMT mutant proteins.
Project Results:
WP1: Engagement: Glial cells and neurons
Activity 1.1
Previous studies suggest that proteins of the Adams family are receptors for Lgi proteins: Lgi1 and Adam22 co-precipitate from brain membrane preparations using PSD95 or Kv1.1 antibodies. Lgi1/Lgi4 bind to cells that ectopically express Adam22/23 and deletion of Adam22 presents a similar peripheral nerve phenotype to that of claw paw animals. We analysed expression of Lgi and Adam family members in the developing nerve and found that all members of the Lgi family are expressed in both neurons and Schwann cells. Out of the wide range of Adam family members that are expressed in peripheral nerve, we focused on the Adams closely related to Adam22: Adam23 and Adam11. All three genes are expressed in Schwann cells and neurons. To test whether binding of Lgi4 to Adam22 is direct and/or requires the presence of additional molecules at the cell membrane, we utilized fusion proteins that consist of the ectodomain of Adam22 (either normal Adam22 or Adam22 with a D509N mutation in the disintegrin-like domain that is analogous to the ju160 allele of C.elegans Adm1/Unc-71, which causes axon guidance and sex myoblast migration defects) and the human IgG Fc moiety. Adam22-Fc protein was found to efficiently precipitate Lgi4 from conditioned medium. In contrast, no Lgi4 protein was found precipitated with the Adam22-Fc protein containing the D509N mutation. These results indicate that binding of Lgi4 to Adam22 does not require additional membrane associated receptors and that binding depends on the integrin-binding domain of Adam22.
-Role of Lgi4 Adam interactions in nerve development
Adam22 is expressed by neurons as well as SCs raising the question whether Lgi4 functions in an autocrine or/and paracrine fashion. To answer this question, we generated primary sensory neuron cultures from Adam22+/neo and Adam22neo/neo embryos and seeded these cultures with rat SCs. Rat SCs readily ensheath and myelinate Adam22+/neo neurons, whereas Adam22neo/neo neurons are not myelinated. These results indicate that Adam22 function is required in the neuron, but do not exclude an additional function in SCs.
To further explore a possible SC-autonomous function of Adam22 and to corroborate its neuronal function in nerve development, we generated a conditional Adam22 allele and bred it to homozygosity in the presence of the Wnt1Cre transgene, thereby deleting Adam22 in all neural crest derivatives, including sensory neurons, endoneurial fibroblasts and SCs, but not in spinal motor neurons. The sciatic nerve is a mixed nerve containing both primary sensory and motor fibers. Analysis of myelin formation in the sciatic nerve of these mice revealed a mosaic pattern of normally myelinated fibers and promyelin arrested fibers. Most of the myelinated axons express Choline acetyltransferase (ChAT), indicating they are motor fibers. Therefore motor neurons are myelinated normally by Adam22-deficient SCs. As in Lgi4-deficient claw paw mice at P12, non-myelinated fibers in Wnt1:Adam22cKO/cKO sciatic nerves are associated with predominantly Oct6 positive promyelinating cells, whereas Krox20 positive SCs are invariably myelinating cells. Examination of the ventral motor and dorsal sensory roots revealed normal myelinated motor fibers and severely hypomyelinated sensory fibers respectively. These data demonstrate that Adam22 is required in neurons, whereas deleting Adam22 in SCs has no effect on myelin formation. Thus, axonally expressed Adam22 serves as the major receptor for SC-derived Lgi4 in PNS myelination.
Adam22 proteins have been shown to interact with integrins. As Lgi4 binds to the disintegrin-like domain of Adam22 it is possible that Lgi4 binding modulates (either positively or negatively) Adam22 integrin interactions. Indeed, Adam22 interacts with a number of integrins that are expressed on Schwann cells and are involved in axonal sorting and myelin stabilization. However, the actions of these integrins have been interpreted in terms of their interaction with the Schwann cell extracellular matrix and not with the axon.
In a collaborative effort between partners 2 and 1 and Dr. Rasband from Baylor College of Medicine, we found that ADAM22 is a component of the Kv1 channel complex and is found together with Caspr2 at the juxtaparanodal region in myelinated axons. Furthermore, we found that among all the known Kv1 channel interacting proteins, only ADAM22 is found at every site where Kv1 channels are clustered. Analysis of Caspr2, PSD-93, PSD-95, and double PSD-93/PSD-95 null mice showed ADAM22 clustering at BCTs requires PSD-95, but ADAM22 clustering at juxtaparanodes requires neither PSD-93 nor PSD-95. In direct contrast, analysis of ADAM22 null mice demonstrated juxtaparanodal clustering of PSD-93 and PSD-95 requires ADAM22, whereas Kv1.2 and Caspr2 clustering is normal in ADAM22 null mice. Thus, ADAM22 is an axonal component of the Kv1 K+ channel complex that recruits MAGUKs (i.e. PSD-93 and PSD-95) to juxtaparanodes. These findings are of particular importance as the excitable properties of neurons depend not only on the kinds of ion channels expressed in the plasma membrane, but also on the location of these channels. Among the many different ion channels expressed in the nervous system, the Kv1 channels are an excellent example of channels that are restricted to distinct subcellular locations. Mutations or diseases that disrupt clustering, localization, or composition of Kv1 channels severely compromise nervous system function and lead to conduction block, episodic ataxia, and/or epilepsies. A major question is whether the clustering of Kv1 and Adam receptors is dependent on binding of an Lgi ligand.
Lgi proteins are characterized by a typical extra-cellular Leucine-Rich Repeat (LRR) and a WDR5 related seven-bladed propeller structure called the epitempin (EPTP) domain. Both these structural motifs provide a platform for specific protein-protein interactions. As described above, all members of the Lgi family bind to a subfamily of Adam proteins eg Adam11, Adam22 and Adam23. These Adam proteins lack metalloproteinase activity and are predominantly expressed in the nervous system. How binding of Lgi4 to Adam22 results in proper axonal ensheathment and myelination is a major unresolved question in this field of research. We began to address this question by asking whether binding of an Lgi ligand to Adam22 is sufficient to promote myelination by Schwann cells. Using a well-defined embryonic dorsal root ganglion neuron Schwann cell co-culture system we demonstrated that Schwann cells will readily myelinate associated axons in wildtype or clp heterozygous cultures but that no myelin is formed in cultures derived from claw paw embryos. Viral transduction of Schwann cells with vectors expressing Lgi4, but not Lgi1, Lgi2 or Lgi3, restores myelin formation in claw paw cultures. Thus, despite the fact that all these proteins bind to Adam22 - although probably with different affinities - only Lgi4 elicits a myelination response. We next performed a number of experiments in which the LRR domain and the EPTP domain were exchanged between Lgi4 and Lgi3 or Lgi1. Again the chimaeric molecules were tested in the co-culture system and it was found that functional specificity resides in the EPTP domain of Lgi4 in particular in the last four blades of the beta propeller structure. Within this domain there is a short amino acid sequence that connects blade 5 and 6 and that is fully conserved in all Lgi proteins suggesting it might be important for Lgi function in general. Indeed, mutations in this region, especially the S440L mutation, in Lgi1 causes epilepsie and psychotic disorders in a Japanese family. To test whether this Serine residue is also essential for Lgi4 function we introduced this mutation in Lgi4 and tested the mutant protein in the co-culture system. We found that this mutation did not affect Lgi4 function. Thus, a crucial amino acid residue in Lgi1 function is not relevant for Lgi4 function suggesting they engage in different interactions. To further identify functionally relevant sequences within the Lgi4 protein we took advantage of the more than 900 million years of independent evolutionary history between modern day fishes and mammals. Curiously, comparative genomics had failed to identify an Lgi4 homologue among the five Lgi genes in the genomes of fishes (species examined; sea bass, salmon, pufferfish, medaka and zebrafish). We cloned the zebrafish Lgi cDNAs and expressed them in the claw paw co-culture system and found that only drLgi1A, and to a lesser extent drLgi1B, but not drLgi2A, drLgi2B and drLgi3, promotes myelination. Examination of a sequence alignment of the primary amino acid sequences of drLgi1A, drLgiB and Lgi4 revealed two short patches that are uniquely conserved between Lgi4 and drLgi1A and drLgi1B but not Lgi1. These patches map next to each other on the upper rim of the EPTP domain of Lgi4. We hypothesize that these short sequence patches represent unique interaction interfaces that distinguish Lgi4 from Lgi1. Whether this patch is involved in Adam22 interactions or perhaps with an accessory molecule is at present unclear and provides an important challenge for the future.
Activity 1.2
During peripheral nerve development Schwann cells (SCs) surround large bundles of mixed-calibre axons send cytoplasmic processes into the bundles, thereby pulling large calibre axons to the edge of the bundles (radial sorting). The sorted large axons are then surrounded in a 1:1 relationship with SCs, and are subsequently myelinated. The mechanisms by which molecules promote sorting are incompletely understood. Axonal sorting requires multiple events: the formation of families of SC with multiple axons contained in a common basal lamina, the matching of the number of axons and SC, the insertion of Schwann cell processes around axons to recognize and segregate large ones, and the defasciculation of single axons with their own daughter Schwann cell and basal lamina. It is unknown which laminin-receptor pairs contribute to each step. For example laminins promote the formation of Schwann cell processes and the interaction with axons, via activation of Rac1 by a beta1 integrin receptor, but laminins also promote Schwann cell proliferation and survival, via activation of PI3K, FAK and Cdc42 by an unknown laminin receptor. Thus which laminin receptors control matching of axons and SC number is unclear. B1 integrin receptors, but not dystroglycan, transmit survival and proliferation laminin signals. Laminins promote SC proliferation and survival and dystroglycan is required in vitro for laminin-mediated protection of SC from apoptosis. To determine which laminin receptors promote proliferation and survival in vivo, we measured the rate of apoptotic and proliferating cells in laminin receptors mutant nerves by TUNEL and anti-histone H3 staining. Proliferation was significantly decreased in beta1 integrins and double mutants at P3 and P5. Dystroglycan mutants showed increased proliferation at P15. Decreased proliferation rates for b1 integrin and double mutants at P3 were confirmed using BrdU incorporation. Apoptosis was increased in double mutants (beta1-integrin/dystroglycan) at P3 and P5. Staining with the SC marker periaxin showed that 80% of the TUNEL positive cells were SC. Single b1 integrin, but not single dystroglycan mutants also showed increased apoptosis. Thus only b1 integrin receptor(s) mediate the survival and proliferation signal from laminins during axonal sorting.
Radial sorting is also impaired in models of human diseases, congenital muscular dystrophy (MDC) 1A, MDC1D and Fukuyama, owing to loss-of-function mutations in the genes coding for laminin b2, Large or fukutin glycosyltransferases, respectively. It is not clear which receptor(s) are activated by laminin 211, or glycosylated by Large and fukutin during sorting. We identified the relevant receptor as dystroglycan and reported these findings in a paper in Development. We showed that the absence of dystroglycan in a specific genetic background causes sorting defects with topography identical to that of laminin 211 mutants, and recapitulating the MDC1A, MDC1D and Fukuyama phenotypes. By epistasis studies in mice lacking both integrin beta1 and dystroglycan in SCs, we show that absence of beta1 integrins impairs proliferation and survival, and arrests radial sorting at early stages, that beta1 integrins and dystroglycan activate different pathways, and that the absence of both molecules is synergistic. Thus, the function of dystroglycan and ?eta1 integrins is not redundant but is sequential. These data identify dystroglycan as a functional laminin 211-receptor during axonal sorting and the key substrate relevant to the pathogenesis of glycosyltransferase congenital muscular dystrophies.
How are these integrin signals transduced in sorting Schwann cells? FAK associates with beta1 integrin in Schwann cells and the presence of basal lamina activates FAK. Focal adhesion kinase (FAK) is a non-receptor tyrosine kinase central to several signal transduction pathways, including those that stimulate proliferation. Here we have investigated the role of Schwann cell FAK by targeted deletion using the Cre-loxP system.
During the sorting process, SCs undergo a burst of proliferation, ensuring that an adequate number of SCs are available to sort all the required axons, and we have shown that Focal Adhesion Kinase (FAK) is partially required for this proliferation. The onset of sorting coincides with the formation by SCs of a shared basal lamina, a major constituent of which is the extracellular matrix (ECM) protein laminin. Indeed, in the absence of laminin, the basal lamina is not formed, SC proliferation is reduced, and process extension and thereby radial sorting does not occur. Interestingly, at the onset of sorting, the SC basal lamina forms in a fragmentary manner; thus, by EM, at E18.5 in mice, in some areas of peripheral nerve the basal lamina appears to be fully-formed, whereas in others it is virtually absent. As ECM, via FAK signaling, can regulate cell migration and proliferation, and can also regulate differentiation, we have been trying to determine whether the SC phenotype is affected by large differences in ECM concentration, and whether FAK plays a role.
In summary, we have found that wildtype SC proliferation and differentiation is virtually unaffected over a relevant range of Laminin1 concentrations. In contrast, though FAK mutant SCs proliferated and differentiated normally at the highest Laminin1 concentration, at lower concentrations FAK mutant SCs failed to spread, did not proliferate and were hyper-sensitive to and differentiated prematurely in response to dbcAMP. These experiments suggest that the FAK mutant SCs are sensitive to cAMP because they do not spread on low Laminin1. Interestingly, in vivo around the time of radial sorting, FAK KO SCs prematurely upregulate the differentiation markers Oct6 and Krox20. This suggests that FAK-mediated SC spreading in vivo may regulate both proliferation and differentiation of SCs.
WP2: Control of myelin formation and maintenance
Activity 2.1
Nectin-like (Necl)/SynCAM cell adhesion molecules are located at the axon-glia interface along the internodes where they mediate Schwann cell-nerve interaction necessary for myelination. The role of these molecules and their interaction with NRG1 signals will be investigated. We took two approaches to study the role of Necl4 in myelination in vivo. In the first, we have generated and begun analyzing transgenic mice expressing a normal and a deletion mutant of Necl4 in Schwann cells and oligodendrocytes using an MBP promoter. Analysis of the transgenes showed that expressing the deletion mutant of Necl4 resulted in profound neurological defects and abnormal myelin formation. These are encouraging results, which support a role for Necl proteins in myelination. In the second approach, we used gene targeting to eliminate the expression of Necl4 in myelinating glial cells. Necl4 was completely absent from homozygous animals as was evident by Western-blot analysis of spinal cord and sciatic nerve lysates, as well as by immunofluorescence analysis of teased sciatic nerves and brain sections. Homozygous Necl4-/- mice were viable and fertile, but displayed mild motor abnormalities, enhanced limb-clasping reflexes, and reduced conduction velocity of peripheral nerves. Histological examination of sciatic nerve cross-sections prepared from homozygous animals at different ages revealed a marked delay in myelination and formation of abnormal myelin membrane detachments and outgrowth. Furthermore, we show that deletion of Necl4 resulted in a profound effect on the organization of myelinated axons - generation of split paranodal junction that was reflected by abnormal distribution of Caspr, aberrant clustering of Kv1.2 channels away from the juxtaparanodal region, and absence of defined juxtamessaxonal lines along the internodes. Our data suggest that axoglial contact, mediated by Necl4, is required for proper myelination and the molecular organization of myelinated axons.
How does Necl4 affect Schwann cell behavior? We have identified protein 4.1G as a cytoplasmic protein that interact with Necl4 in Schwann cells. Extensive immunolabeling experiments showed that these two molecules co-localized along the myelin internode. Furthermore, clustering of Necl4 on the surface of Schwann cells recruits protein 4.1G into the clusters, indicating that these proteins are associated at the plasma membrane interface. Analysis of mice lacking protein 4.1G revealed that these mice exhibit abnormal distribution of Necl4 and several other non-compact myelin proteins, further supporting a functional connection between these two proteins. Similar to ablating Necl4, deletion of 4.1G in mice resulted in aberrant distribution of both glial adhesion molecules and axonal proteins that were present along the internodes. In wild-type nerves, juxtaparanodal proteins (i.e. Kv1 channels, Caspr2, and TAG-1) were concentrated throughout the internodes in a double strand that flanked paranodal junction components (i.e. Caspr, contactin and NF155), and opposes the inner mesaxon of the myelin sheath. In contrast, in 4.1G-/- mice, these proteins ?piled up? at the juxtaparanodal region or aggregated along the internodes. These findings suggest that protein 4.1G contributes to the organization of the internodal axolemma by targeting and/or maintaining glial transmembrane proteins along the axoglial interface.
In our search for proteins that control myelin formation and maintenance, we concentrated on proteins that would translate extracellular signals to the movement of the actin cytoskeleton that, as we suspected, will be required for myelination. We have shown that N-WASP is required for membrane wrapping and myelination by Schwann cells. During peripheral nerve myelination, Schwann cells sort larger axons, ensheath them and eventually wrap their membrane to form the myelin sheath. These processes involve extensive changes in cell shape, but the exact mechanisms involved are still unknown. Neural Wiskott-Aldrich syndrome protein (N-WASP) integrates various extracellular signals to control actin dynamics and cytoskeletal reorganization through activation of the Arp2/3 complex. By generating mice lacking N-WASP in myelinating Schwann cells, we show that N-WASP is crucial for myelination. In N-WASP deficient nerves, Schwann cells sort and ensheath axons, but most of them fail to myelinate and arrest at the promyelinating stage. Yet, a limited number of Schwann cells form unusually short internodes, containing thin myelin sheaths, with the occasional appearance of myelin misfoldings. These data suggest that regulation of actin filament nucleation in Schwann cells by N-WASP is crucial for membrane wrapping, longitudinal extension and myelination.
Activity 2.2 and 2.3
Binding of NRG1 typeIII to glial ErbB2 and ErbB3 receptors is likely to activate multiple signaling pathways, but it is unknown how they are tied into the myelination program. One attractive second messenger is the PI3K pathway with the generation of phosphatidyl-inositol (3,4,5)-triphosphate ('PIP3') as a signaling lipid. To determine the role of PI3Kinase signaling in PNS myelination in vivo we analyzed a mouse line were the Pten gene (encoding a D3-lipid phosphatase that antagonizes PI3K signaling) was specifically ablated from the Schwann cell lineage by using a CnpCre driver line. We could demonstrate by PCR amplification of sciatic nerve genomic DNA that the Cre-mediated inactivation of Pten in Schwann cells resulted in efficient deletion of Pten exon 5 in mutant sciatic nerves. Western blot analysis confirmed a significant downregulation of PTEN protein and enhanced phosphorylation of known PI3 Kinase effector proteins, i.e. AKT and S6 in mutant sciatic nerves. When the levels of PtdIns(3,4,5)P3 and PtdIns(4,5)P2 were analyzed in acidic lipid extracts of sciatic nerves by gas-chromatographic quantification of their associated fatty acids we found, as predicted, an increase in PtdIns(3,4,5)P3 levels and a decrease in PtdIns(4,5)P2.
When small caliber axons (below 2 µm in diameter) of sciatic nerves were quantified by g-ratio analysis using ultra-thin cross sections, loss of PTEN caused a significant increase of myelin sheath thickness. However, when the degree of myelination was quantified for axons larger than 2µm (in the same cohort of mice), g-ratios of Pten mutants were unaltered compared to controls. Instead, we noticed that one very prominent feature was focal hypermyelination (Tomacula) and myelin outfoldings reminiscent of the neuropathological features in a diverse group of demyelinating neuropathies, including hereditary neuropathy with liability to pressure palsies (HNPP). Teased fiber preparations and the electron microscopic analysis of longitudinal sciatic nerve sections at postnatal day 30 (P30) revealed that these myelin abnormalities predominantly originated from paranodal regions and Schmidt-Lantermann incisures. These two regions of non-compacted myelin might represent zones of controlled membrane incorporation and myelin sheath growth. We therefore investigated whether the lack of Pten would mimic abnormal axon-glia signaling and PI3K activation in these regions. Indeed, when p-AKT was used as a sensor of PtdIns(3,4,5)P3, and co-immunostained with MPZ (a marker of compact myelin) and myelin-associated glycoprotein (MAG; a marker of non-compacted myelin), we found a striking colocalization of p-AKT in MAG-positive Schmidt-Lantermann incisures and paranodes. This indicates that in those regions PTEN could be involved in regulating PtdIns(3,4,5)P3 levels. Levels of the two cell polarity proteins Par3 and Dlg1, normally located at regions of non-compact myelin, and suggested to play a role in the control of myelination, were unchanged when tested by immunohistochemistry on teased fibers.
WP3: Functional organization of the axo-glial unit
Activity 3.1
A major consequence of myelination is the establishment of axonal domains that restrict sodium and potassium channels distribution. We investigated the mechanism through which these domains are established and what molecules play a role in this process.
Heminodal clustering of Na+ channels requires Gliomedin - To examine the role of gliomedin in the assembly of the nodes of Ranvier, we generated gliomedin null mice. Homozygous gldn-/- mice exhibited no overt neurological abnormalities, formed compact PNS myelin that was indistinguishable from their wild type littermates, but displayed disorganization and impaired attachment of Schwann cell microvilli to the nodal axolemma. As Schwann cell microvilli were implicated in node formation, we examined the distribution of Na+ channels in the developing sciatic nerves of wild type and gldn-/- at P6, when both nodes and heminodes can be detected. We found that while in wild type nerves Na+ channels were clustered in mature nodes (i.e. nodes that are flanked by two adjacent Caspr-labeled PNJs), as well as in heminodes, in gldn-/- nerves these channels were absent from heminodes and only detected in mature nodes. In adult sciatic nerves of gldn-/- mice, Na+ channels were colocalized at the nodes of Ranvier with NF186, NrCAM, ankyrin G and ?IV spectrin, Caspr was present at the PNJ and Kv1 channels were localized at the juxtaparanodal region similar to wild type nerves. As an additional experimental approach, we used myelinating Schwann-dorsal root ganglion (DRG) neurons cultures, which enable monitoring the detailed molecular assembly of the nodes in PNS axons. In contrast to wild type cultures, Na+ channels, NF186, as well as ankyrin G and IV spectrin were not clustered at heminodes in gldn-/- myelinating cultures. Notably, while both in vivo and in myelinating cultures of wild type nerves, heminodes appeared adjacent to Caspr-labeled PNJs, in gldn-/- nerves, heminodes were absent near Caspr-positive sites, indicating that heminodal clustering is not the result of PNJ formation. We then counted the number of sites in which Na+ channels were clustered near the edge of a myelin segment. In gldn-/- the majority of Na+ channels were either absent or diffused (gldn-/- culture 92%, nerve 81%), compared to wild type nerves where they were clustered in more than 90% of the sites. These results demonstrate that gliomedin is required for clustering of the nodal complex at heminodes during myelination.
NF186 and NrCAM are required for clustering of Na+ channels at heminodes but not for their accumulation in mature nodes - The observation that clustering of the nodal complex requires gliomedin prompted us to examine whether this activity is mediated by its two axonodal receptors, NrCAM and NF186. We first examined sciatic nerves of nrcam-/- mice, in which Na+ channel clustering in PNS nodes is delayed. Immunolabeling of teased sciatic nerves isolated from 6 day-old (P6) mice using antibodies to Na+ channels and Caspr (to label the PNJs) showed that Na+ channels were not clustered at heminodes that were already bordered by Caspr-labeled PNJs. In addition, Na+ channels, NF186 and gliomedin were all detected at heminodes in myelinating DRG cultures derived from wild type, but not from nrcam-/- mice. In both sciatic nerves and myelinated cultures, Na+ channels, NF186, gliomedin, ankyrin G and ?IV spectrin were accumulated at mature nodes. Thus, similar to gliomedin, NrCAM is required for heminodal clustering but not for the accumulation of Na+ channels at mature nodes of Ranvier. Previous observations by UEDIN showed that genetic ablation of the neuronal (NF186) and glial (NF155) isoforms of neurofascin leads to the disruption of both PNJs and nodes. In line with these findings, nodes fail to form in myelinating co-cultures of DRG neurons and Schwann cells isolated from nfasc-/- mice. To examine directly the role of NF186 in the assembly of PNS nodes, we determined whether Na+ channels accumulate at nodes in the absence of NF186 by co-culturing DRG neurons isolated from nfasc-/- mice with wild type Schwann cells. The expression of NF155 in myelinating Schwann cells promoted the full recovery of the PNJs as demonstrated by immunolabeling for Caspr. PNJ formation in these axons was accompanied by the accumulation of Na+ channels at nodes together with all components of the nodal complex (i.e. ankyrin G, ?IV spectrin and NrCAM) except NF186. However, as observed in gliomedin and NrCAM-deficient DRG cultures, in the absence of axonal NF186, Na+ channels were not clustered at heminodes. To further determine the contribution of the axonodal CAMs to the formation of mature nodes, we established myelinating cultures using DRG neurons isolated from double nrcam-/-/nfasc-/- mutant mice and wild type Schwann cells (to recover the PNJs). Analysis of these cultures showed that Na+ channels, as well as ankyrin G and ?IV spectrin, accumulated at nodes that were flanked by PNJs, but not at heminodes. Taken together, these results demonstrate that gliomedin and its axonodal receptors NrCAM and NF186, are all required for developmental clustering of Na+ channel at heminodes. They further reveal the existence of an additional mechanism that enables the accumulation of Na+ channels in mature nodes independently of heminodal clustering and the axonodal CAMs.
Involvement of the PNJ in node formation - One possible mechanism that could compensate for the lack of heminodal clustering might be provided by the PNJ, which serves as a membrane barrier separating nodal Na+ channels and juxtaparanodal K+ channels. However, the role of the PNJ in node formation remains unclear since several studies using paranodal mutant mice have indicated that these structures are dispensable for PNS node formation. Myelinating DRG cultures lacking gliomedin, NrCAM or NF186, in which heminodes fail to form, allowed us to directly examine the role of the PNJ in the assembly of PNS nodes. We found that in wild type cultures Na+ channels were clustered at heminodes, which eventually merged, forming a mature focal node of Ranvier. In contrast, in gldn-/- and nrcam-/- cultures, as well as in cultures containing nfasc-/- neurons and wild type Schwann cells (i.e. thus considered as nf186-/-) Na+ channels were spread between the two growing myelin segments. A similar distribution of Na+ channels was detected in developing sciatic nerves of gldn-/- and nrcam-/- nerves. Interestingly, although Na+ channels co-clustered with NF186 during all developmental stages in wt cultures, NF186 was not detected in gldn-/- and nrcam-/- cultures until these channels were already accumulated at focal nodes, further indicating that accumulation of Na+ channels in mature nodes occurs independently of the axonodal CAMs. Immunolabeling of gldn-/-, nrcam-/- and nf186-/- cultures using antibodies to Caspr, Na+ channels and MBP, showed that Na+ channels were eliminated from the area occupied by the PNJ, but invaded the axolemma underneath the myelin sheath in its absence. These observations suggest that in addition to being actively clustered at heminodes, Na+ channels are constrained between two neighboring myelin segments by the PNJ. Furthermore, in the absence of the PNJ, as found in mice lacking Caspr for example, the presence of gliomedin and the axonodal CAMs enable the clustering of Na+ channels at heminodes and consequently in mature nodes.
Activity 3.3
The functional organization of the axo-glial unit has been remarkably well conserved during evolution. The main molecules involved and the underlying molecular mechanisms are even conserved in simple model animals such as Drosophila. Drosophila offers a large number of experimental advantages that make this system very attractive to identify additional gene functions mediating or controlling neuron-glia interaction in the nervous system. One of the key approaches towards this aim resides in cell-type specific gene silencing. We have generated a number of Gal4 driver lines, which now allow activating gene expression in almost every glial cell type of Drosophila. In combination with large collections of transgenic lines established in Vienna, Austria, Harvard, USA, or Kyoto, Japan, that contain constructs which allow a Gal4 dependent expression of double stranded RNA, cell-type specific RNA interference (RNAi) can be induced. Such genome wide RNAi screens will eventually unravel most of the genes required for glial differentiation.
One of the main aims of the proposal was to identify novel genes acting in Drosophila glial cells. In flies just as in primitive vertebrate fish the blood brain barrier is established by glial cells, which form extensive cellular junctions characterized by septate junctions. In the mammalian nervous system, very similar junctions are established at the paranodal junctions. In both systems, the core components of these junctions are NeurexinIV, (the mammalian homolog is the Caspr protein) and Neuroglian (the mammalian homolog is the Neurofascin155 protein). This and other cases of functional homology suggest that Drosophila can serve as a model system to identify novel components underlying neuron-glial interaction.
In the funding period we have followed two different experimental avenues to reach these goals. On the one hand we continued our efforts with direct genetic screens for mutants affecting the integrity of the blood brain barrier. On the other hand we silenced individual genes specifically in glial cells in a RNA-interference based approach and thus determined genes required for glial cell development. Whereas the first approach directly pointed towards a gene function, the second approach just gave an indication whether or not a particular gene has any role in glial cell development and thus requires additional genetic work.
WP4: Disturbed axo-glial signaling and nerve pathology
Activity 4.1
Tomacula and myelin outfoldings are a striking neuropathological feature of a diverse group of demyelinating neuropathies, including Charcot-Marie-Tooth disease 4B (CMT4B) and hereditary neuropathy with liability to pressure palsies (HNPP). Despite major progress in defining the genetic basis of these diseases, the molecular mechanisms that cause focal myelin growth and disrupt Schwann cell-axon interaction in some forms of CMT, but not in others, have remained obscure. Partner 5 has hypothesized that myelin outfoldings and "tomacula" result from uncontrolled myelin membrane growth, a developmental program that is at least in part controlled by neuregulin-1/ErbB2 and PI3K signaling. Partner 5 found in mice that lack Pten expression in Schwann cells, that glia cell-specific hyperactivation of the endogenous PI3K pathway causes focal hypermyelination, myelin outfoldings, and tomacula, associated with a progressive peripheral neuropathy. In previous experiments (see WP 2) Partner 5 found activated AKT kinase associated with phosphatidylinositol-3,4,5-trisphosphate (PtdIns(3,4,5)P3) at paranodal loops and Schmidt-Lantermann incisures of Pten mutants. This and the focal hypermyelination pathology indicate that regions of non-compact myelin permit membrane incorporation and myelin growth. We suggest a model that elevated PtdIns(3,4,5)P3 levels contribute to the pathology of tomaculous neuropathies at least partially by activation of the AKT/mTOR pathway. Indeed, a pharmacological reduction of aberrant mTOR activation by rapamycin, a specific mTOR inhibitor, prevented aberrant myelin growth in Pten conditional mutant nerves to a significant extent.
To analyze whether inhibition of mTOR signaling could prevent the formation of myelin abnormalities Partner 5 treated Pten mutants and littermate control mice with rapamycin. We selected 3 groups of abnormal myelin formations by histological criteria (tomacula, comma-shaped outfoldings, recurrent loops and quantified them in sciatic nerve cross sections of controls and conditional Pten mutants that were treated with vehicle (-Rap) or rapamycin (+Rap) from postnatal day (P) 12 to P43 (Del 4.8). In the vehicle treated groups, as expected, all myelin abnormalities were higher in Pten mutants than in controls. After 31 days of rapamycin treatment tomacula were 75% less frequent in the drug treated mutants (52 ± 12.87 n=8) when compared to vehicle treated mutants (209 ± 20.76 n=6) (p<0.0001). Similarly, recurrent loops were reduced by 75 % in the rapamycin treated mutants (17.88 ± 3.393 n=8) when compared to vehicle treated mutants (72.17 ± 7.897 in Pten mutants, n=6) (p<0.0001). Comma-like outfoldings were reduced by 49% (60.13 ± 13.37 n=8 in treated Pten mutants vs. 117.2 ± 6.882 n=6 in Pten mutants) (p=0.0051). Rapamycin treatment had no visible effect on control mice. Thus, although the growth abnormalities of the myelin sheaths were not completely corrected, the results indicate that enhanced mTOR signaling plays a role in the formation of redundant myelin in the Pten mutants.
Activity 4.2
The proper cellular responses following nerve trauma, in particular demyelination, require the activity of the transcription factor c-jun. We asked whether the removal of c-jun in a neuropathy model could ameliorate the clinical symptoms in this model. To this end P0CreXcJunflox/C3 mice were created together with suitable P0CreXcJunflox and C3 control mice. This was done to investigate whether the elevated c-Jun expression detected in C3 mice (which over express PMP22 and serve as a model of CMT1A neuropathy) was deleterious for C3 pathology (as predicted by the ability of c-Jun to suppress myelin differentiation) or helpful (as predicted by the ability of c-Jun to activate a neuron-supportive Schwann cell program during mechanical nerve injury).
The P0CreXcJunflox/C3 mice (that over express PMP22 but lack c-Jun) were subjected to morphometric nerve analysis and an extensive behavioral testing of sensory and motor performance. The results were compared to the same tests carried out on P0CreXcJunflox and C3 control mice.
The behavioral tests included, beam walking, grid walking, hanging wire grip strength test, Rotarod performance, von Frey hair sensory test, hotplate test and footprint analysis. It was found that deletion of c-Jun in PMP22 over expressing mice resulted in a clear worsening of motor and sensory motor performance. The outcome of pure sensory tests (von Frey hair sensory test, hotplate test) did not change. It should be noted that most/all motor tasks involve the function of sensory neurons, in particular sensory afferents from muscle spindles.
Electrophysiological measurements showed a strong reduction in F-wave persistence in P0CreXcJunflox/C3 mice compared to C3 controls. This indicates reduced excitability of spinal motoneurons, which in turn could be caused by reduced input from sensory neurons innervating muscle spindles.
Finally, a correlated study had provided similar evidence that dedifferentiation may be beneficial, not detrimental, to myelinating S63del (CMT1B) Schwann cells, in which the unfolded protein response is pathogenic. Two independent approaches had suggested that persistent expression of genes that limit differentiation, in this case Sox2 and Id2, in Schwann cells improves both plasticity after injury and limits proteo-toxic stress in neuropathy. To eliminate the possibility that compensation by either factor in the absence of the other produced this paradoxical conclusion, we examined S63del nerves in which both Sox2 and Id2 were ablated in Schwann cells. Again, the ablated nerves were more pathological than S63del with Id2 and Sox2 intact, demonstrating that the two factors are beneficial in S63del neuropathy. It is likely that these factors promote plasticity of Schwann cells in injured nerves, and this plasticity is advantageous in neuropathy. A similar conclusion was drawn based on studies of the role of the dedifferentiation factors cJun and Notch after nerve injury and in neuropathy.
Activity 4.3
Develop a cell-based assay to identify small molecules that reverse or ameliorate the pathogenic mechanism of CMT mutant proteins.
NGIDD has constructed the Xbp-1unspliced-EGFP reporter construct and performed titration assays with G418 in transfected Cos-7 cells. We have established that there is not obvious cell death in Cos-7 cells due to transient transfection of the P0S63del-dsRED or P0wt-dsRED. Based on these collective data, we performed a permanent transfection in Cos-7 cells with the Xbp-1unspliced-EGFP reporter construct, in order to generate the UPR reporter cell line (Del 4.11). After two weeks of selection in 700 μ g/ml G418, a limiting dilution, at 1, 5 or 10 cells/well, in ten 96 MTPs, was performed. The pool of G418 resistant cells was analyzed at fluorescence microscope, upon 18 hrs. stimulation with 10µM Tunicamycin (to induce a UPR and Xbp1 splicing, and hence GFP expression).
When almost all of the wells were confluent, plates were replicated into 384 MTPs, in order to obtain a quadruplicate for each single clone. The first so called clone pool analysis was carried out on these plates, on uncounted cells. Cells were incubated in complete medium with or without 10 μ M Tunicamycin for 18 hrs. Plates were analysed at FLIPRTETRA (MDS) and emitted fluorescence recorded. Positive clones were selected on the basis of the highest EGFP dependent fluorescence signal detected. A total of 23 clones, deriving from limiting dilution at 10 cells/well, were selected, picked up and seeded in 6 wells plate for cell amplification. After clone selection, Cos-7 reporter cells were maintained in complete medium supplemented with 500 μ g/ml G418. Unfortunately, as these clones were expanded and carried longer in culture, most of them died. We suspect toxicity in the cells permanently transfected with the Xbp-1unspliced-GFP reporter construct. They die perhaps due to chronic, low-level, leaky expression of GFP in the permanently transfected Cos-7 cells (a basal level of Xbp-1 splicing in Cos-7 cells is observed and may activate GFP expression even in the absence of ER stress). Currently 4 remaining reporter clones are in culture, to be tested in further experiments.
To avoid the problem of Xbp1-EGFP fusion protein toxicity, we have engineered two alternative plasmids based on firefly luciferase reporter genes. First, using the single cistron with IRES strategy, we have completed construction of the alternative Xbp-1unspliced-firefly luciferase-IRES-renilla luciferase plasmid and will perform permanent transfections in Cos-7 cells as before (Del 4.11). Performing the same analysis as outlined above for Xbp-1EGFP cells, we found only very low level activation of luciferase after tunicamycin treatment to induce UPR in Xbp1-luciferaseIRESrenilla cells.
Based on the results obtained, two future directions can be envisaged:
Results obtained upon screening the LOPAC compound collection on the UPR cell lines allow the identification of some promising hits. Therefore next steps would be the further validation of these hits and the interpretation by a biological proof of concept, using proof-of-principle in vitro assays and in vivo models. Anyway it has to be underlined that these hits won't represent per se lead compounds for med-chem programs, but the investigation on their activity on secondary assays might additionally confirm the validity of the overall system. Treatment of myelinating explant cultures from S63del mice with these hit compounds is underway to provide just this validation.
The development of a stable cell line expressing P0S63Del together with an appropriate reporter gene, suitable for the analysis of compounds interfering with UPR mechanisms and signaling represents a big success, since so far only transiently transfected cells have been produced. The screening of the LOPAC collection proved that the assay is very robust and suitable for running true screening campaigns. Therefore the next steps would imply the screening of larger compound collections, in order to identify hits that might indeed represent lead compounds for future med-chem programs.
Potential Impact:
NGIDD has contributed to a better understanding of nervous system function and dysfunction in disease. Furthermore, our work on patho-mechanism of neuropathy directly feeds into industrial development and new therapies and contributes to improved diagnosis and management of diseases of the nervous system.
Primary neurological diseases of peripheral nerves cause significant disability. In addition, the developmental and pathogenic processes in the PNS are of direct relevance for our understanding of CNS processes. Due to the simpler cell structure and accessibility of the PNS, many of the molecular mechanisms governing CNS development and pathology have been first elucidated in the PNS. Our work encompasses aspects of both normal development and pathology to find relevant pathogenetic mechanisms accessible to therapeutic intervention. On the one hand, several of the developmental mechanisms that we have elucidated are arrested or impaired in genetic diseases. Examples are lack of axonal sorting and Na+ channel clustering in laminin deficiencies, myelin protein over-expression in CMT1A, and myelin outfolding in CMT4B and HNPP. On the other hand, most mechanisms used during development to form functional myelinated fibers are recapitulated during repair after common multifactorial or acquired neurological diseases, such as multiple sclerosis, diabetes, spinal cord and nerve injury and inflammatory neuropathies. Finally, it is becoming clear that many dominant diseases in the nervous system act through toxic mechanisms, such as protein misfolding and dysregulation of quality control. Our studies of these processes in peripheral neuropathies ha?e identified pathogenetic mechanisms and therapeutic targets relevant to other common diseases of the nervous system, such as neurodegenerative diseases.
We have developed a high throughput cell-based assay that reports the early events of maladaptive stress response. We have validated this approach and we will use this cell based assay to identify small molecules useful as tools in proof of therapeutic principle studies in our animal models of neuropathy, and form the basis for medicinal chemistry to develop therapeutic agents that can reduce the toxicity of the integrated stress response. Our data indicate that modulation of such pathways in neuropathy (and other misfolded protein disease) patients may diminish both the severity of their disease, and the resulting disability. Since conformational or protein misfolding diseases include for example the more common Alzheimer and Parkinson diseases or Diabetes, the information produced by our experiments may find broader application in other diseases. (Ron, 2002; Lindsten et al., 2003)
Finally, one of the central objectives of this project was to identify common downstream pathogenetic mechanisms and their targets in order to find a common therapeutic strategy for the wide array of neuropathies resulting from mutations in 23 genes. Altered trafficking of mutant proteins and activation of protein quality control mechanisms (unfolded protein response, proteasome degradation, lysosomal-based autophagy) are common features of hereditary neuropathies. We have focused our attention on how these features may perturb signal pathways in glia downstream of developmentally important axonal to glia interactions. Thus, what we have learned about this toxicity in our CMT animal models will likely be applicable to many neuropathies. Indeed, we have shown altered PIP3/AKT signaling and neuropathy in PTEN mutant mice resembling myelin outfoldings as in the rare CMT4B neuropathy. We demonstrated that rapamycin treatment meliorates myelin outfoldings in our animal model. Such myelin abnormalities are present in a much broader range of neuropathies, including the more common HNPP. Thus our results suggest that rapamycin treatment might be beneficial in a range of neuropathies.
Project website: http://www.ngidd.eu(s’ouvre dans une nouvelle fenêtre)
Contact details coordinator:
dies Meijer, PhD
professor in Developmental Neurobiology
department of Genetics
erasmus University Medical Center
street address; Dr Molewaterplein 50
3015GE Rotterdam
postal address: PO box 2040
3000CA Rotterdam
tel +31-(0)10-7044250
d.meijer@erasmusmc.nl
contact detail project manager:
rini de Crom, PhD
managing Director
FP7-CP NGIDD
erasmus MC - Dept. Cell Biology and Genetics
s-Gravendijkwal 230
3015 CE Rotterdam
the Netherlands
phone: +31-107-043459
fax: +31-107-044743
ngidd@erasmusmc.nl
