Steven S. Scherer, MD PhD - US grants

Peripheral neuropathies and nerve injuries are common clinical problems, and the only medical therapy is to identify and eliminate their cause. Since axonal regeneration occurs in peripheral nerves, and is a major mechanism by which recovery occurs in patients who have a neuropathy or nerve injury, we need to understand axonal regeneration at a molecular level in order to create new therapies. Schwann cells play a central role in axonal regeneration. When Schwann cells are 'denervated' by the loss of axonal contact, they undergo profound changes that are part of what is termed Wallerian degeneration. In neuropathies and nerve injuries, these denervated Schwann cells and their basal laminae provide the pathway through which axons regenerate. To understand how Schwann cells promote axonal regeneration at a molecular level, we need to learn what genes are up- and down-regulated, what roles these genes play in Wallerian degeneration, and how they are regulated. The approach will be to make a cDNA library from degenerating rat sciatic nerve, and screen this library with cDNA probes from normal sciatic nerve and from degenerating sciatic nerve, thereby identifying clones that are differentially expressed in degenerating nerve. Clones that are up- regulated and down-regulated during Wallerian degeneration will be collected, and the former will be analysed first, as these seem more likely to play a role in axonal regeneration. Clones representing less abundant cDNAs that are up-regulating during Wallerian degeneration will be isolated with a subtractive probe. Clones will be characterized by (a) cross- hybridization to find the ones that are copies of the same gene, (b) by performing Northern blots to confirm that the clones are truly up-regulated during Wallerian degeneration and to determine the size of their mRNA, and (c) by partially sequencing clones to see if they are novel or have previously been described. Novel clones that are up-regulated will be completely sequenced, their amino acid sequence will be deduced, and their cellular localization will be demonstrated by in situ hybridization & immunohistochemistry.

Inherited demyelinating neuropathies, also known as Charcot-Marie-Tooth (CMT) disease type 1, are characterized by demyelination and remyelination of peripheral nerve axons. We were the first to show that mutations in connexin32 (Cx32), a gap junction protein, cause X-linked dominant CMT (CMTX), and that Cx32 is localized to the paranodes and incisures of the Schwann cell myelin sheath. We hypothesize that there are both loss-of-function and toxic gain-of-function effects of CMTX mutations, and that both may lead to demyelination because they interfere with the formation or function of gap junctions in the myelin sheath, although other explanations are possible. In this proposal, we will examine the function of Cx32 in the myelin sheath as well as the effects of CMTX mutations in vitro and in vivo, to determine the normal function of Cx32 as well as how CMTX mutations cause demyelination. Our Specific Aims are (l) to determine whether there are functional gap junctions in myelinating Schwann cells, by demonstrating that small, but not large, molecular weight, markers can diffuse radially across the myelin sheath through incisures; (2) to study the effects of CMTX mutations on the synthesis and function of Cx32 in cultured mammalian cells, by examining the intracellular processing of different Cx32 mutants and their electrophysiological properties; (3) to determine whether different CMTX mutations can cause dominantly inherited demyelinating neuropathy in transgenic mice. By systematically evaluating the function of different CNN mutations in vitro and in vivo, we will determine whether different CMTX mutations cause a loss-of-function or a toxic gain-of-function. In addition, we will determine whether CMTX mutations disrupt the function of the gap junctions that are found at incisures and paranodes in the myelin sheath, and explore the relationship between the disruption of these gap junctions to the pathogenesis of demyelination. The other members of this Program Project will enable us to accomplish our goals. Dr. Rutkowski will establish myelinating co-cultures, so that we can determine whether Cx32 mutations affect the ability of Schwann cells to form and maintain myelin sheaths. Drs. Barchi and Kallen will help us to develop a spatial model of the Cx32 protein, so that we can better predict the consequence of CMTX mutations.

DESCRIPTION (Taken from Abstract): Myelination is one of the fundamental adaptations of the vertebrate nervous system, and its normal functioning is essential for health. Genetic manipulation of mice has revealed that two, unrelated transcription factors, SCIP (also known as tst-1 or Oct-6) and Krox-20, are essential for myelination in the peripheral nervous system. These transcription factors are expressed by Schwann cells, and myelinating Schwann cells do not develop normally in mice that are genetically null for either SCIP or Krox-20. In both SCIP and Krox-20-null mice, the development of myelinating Schwann cells appears to be arrested at the promyelinating stage, when myelinating Schwann cells have ensheathed axons in a 1:1 manner but have not yet elaborated a myelin sheath. In SCIP-null mice, Schwann cells are transiently arrested at this stage, whereas in Krox-20-null mice, Schwann cells are permanently arrested at the promyelinating stage. It is unclear how the lack of either SCIP or Krox-20 results in dysmyelination. The observation that SCIP and Krox-20 both appear to be expressed at relatively high levels in promyelinating Schwann cells, which is exactly when the development of myelinating Schwann cells is arrested in null mice, leads them to hypothesize that these two transcription factors interact in promyelinating Schwann cells to regulate the expression of genes that are essential for the formation of the myelin sheath. In this grant, the principal investigator will evaluate various aspects of this hypothesis, by determining (1) at what stage Schwann cells express SCIP, (2) whether SCIP and Krox-20 interact in the regulation of Schwann cell development, (3) the effects of SCIP and Krox-20 on the phenotype of Schwann cells in vitro, and (4) the target genes of SCIP and Krox-20.

DESCRIPTION (provided by applicant): Dominantly inherited demyelinating neuropathies, known as Charcot-Marie-Tooth disease type 1 (CMTI), are among the most common inherited neurological diseases. CMTI is caused by mutations in one of several genes that are expressed by myelinating Schwann cells, including PMP22, MPZ, and GJBJ. Although demyelination is the first pathological consequence, axonal loss rather than demyelination per se, is the main cause of neurologic disability. In this grant, we will evaluate the consequences of disrupted axon-Schwann cell interactions in demyelinating neuropathies, by determining whether demyelination and remyelination reorganize the axonal membrane in (i) several animal models of inherited demyelinating neuropathy, and in (ii) an animal model of acute demyelination (after lysolecithin injection into the sciatic nerve). In this way, we will determine whether different genetic causes of demyelination have similar effects on the molecular organization of axonal membranes, and the temporal order in which the molecular components of nodes, paranodes, and juxtaparanodes are disassembled by demyelination and reassembled during remyelination. We will also determine whether axonal loss in demyelinating neuropathies can be ameliorated in two ways-by breeding Mpz/Po-null and Pmp22-null mice with either Wids or transgenic mice in which glial-derived neurotrophic factor (GDNF) is overexpressed in muscle. If expressing the Wlds gene or the GDNF-transgene preserves axons Mpz/Po-null and Pmp22-null mice, this will provide proof of the concept that therapies directed at neurons rather than myelinating Schwann cells can be effective treatments for inherited demyelinating neuropathies.

Mutations in GJB1, GJA1, and GJA12, the genes that encode human connexin32 (hCx32), hCx43, and hGx47, cause the X-linked form of Charcot-Marie-Tooth disease (CMT1X), oculodehtodigital. dysplasia (ODDD), and PeHzaeus-Merzbacher-ijke disease (PMLD), respectively, all of which cause important CNS abnormalities that appear to be related to abnormal functioning of oligodendrocytes. The central theme of this grant is that Cx30:Cx32 and Cx43:Cx47 heterptypic channels mediate astrocyte/oligodendrocyte (A/O) coupling, which is disrupted by mutations of GJB1/Cx32,GJA1/Cx43, or GJA12/Cx47. 1.Investigate the molecular defects of hCx47 mutants causingi PMLD. We will characterize further the nature.ofthese defects, .arid determine whether wild type (WT) hCx47 or these hOx47 mutants can form functional channels with hCx43 by dye transfer and electrophysiblogy. 2. Investigate the molecular defects of hCx43 jmytants causing ODDD. We will investigate the molecular nature of the mutaiit proteins, determine whether cells expressing an ODDD mutant can form functional channels by themselves, or with celJs expressing either WT hCx43 or WT iGx47, and determine.1whether ODDD mutants'have dominant-negative effects on-WT hCx43. 3-Determine whether hCx32 "CNS mutants" have dominant effects on hCx47. [unreadable] ,.'..' n HeLa cells these "CNS mutants" accumulatefiithef in.the endoplasniic reticujum (ER) or in the Gblgi. Our Dreliminary evidenceindicates that co-expression of these."CNS mutants" with WT hCx47 results in partial retention of Wt hCx47 in the ER or Golgi, indicating .tn.atthese Cx32 mutants exert a dominant effect on WT hCx47. We will characterize further the nature of these defects. . . 4. Determine the role of Cx32 and Cx47 in astrocyte/oligodendrocyte coupling. We will immunostain the brains of mice that lack Cx32 and/or Cx47, and determine whether the localization of their proposed partners is altered. We will also investigate A/O coupling by injecting astrpcytes genetically abeled with green fluorescent protein (GFP) in acute spinal cord slices from Gjb1/cx32 and Gja12/cx47 double hull" mice with small molecules that can cross GJs. In this way, we will determine the relative mportance of the two kinds of heterotypic channels in A/0 coupling.

We hypothesize that in normal myelinated PNS axons, the combination of Kv1.1, Kv1.2, KCNQ2, and KCNQ3 is necessary for repolarization, and that the misexpression of Kv3.1 b has a deleterious effect on axonal conduction of de/remyelinated axons. In unpublished work, we have found the alphas isoform of Na,K-ATPase is the only one that is clearly localized to axons, that it is (surprisingly) excluded from nodes, and appears to be focally diminished by demyelination. Thus, we also hypothesize that the misexpression of alphas may contribute to depolarization of de/remyelinated axons. The proposed experiments build on these findings, with the central theme of illuminating how K+ homeostasis works in normal and de/remyelinated axons. Aim #1: Do Kv3.1b channels contribute to conduction failure in demyelinating diseases? We will investigate whether KvS.lb is the 4-AP-sensitive channel by comparing the effects of DTX-I and 4- AP on axonal conduction in TremblerJ mice on a KvS.lb -null (Kcncl-/-) versus Kcnc1+/+ background. Aim #2: What is the role of KCNQ2 in myelinated axons? Because Kcnq2-null mice die at birth, before myelination and the formation of nodes, we will generate a conditional Kcnq2-null mouse and analyze the structure and function of myelinated axons, including the expression of KCNQ3. Aim #3: What Na,K-ATPase isoforms are expressed by myelinated axons and demyelinated axons? we will localize alpha1-3 by immunoelectron microscopy in CMSand PNS myelinated axons, along with their beta subunits.

Charcot-Marie-Tooth disease (CMT) is the eponym for all inherited neuropathies that are not part of a syndrome. CMT is a relatively common disease, affecting ~1:2500 people (Skre, 1974), but the genetic causes of CMT are increasingly diverse, with more than 30 genes identified to date. CMT1, CMT2, and CMT4 collectively constitute the majority of CMT cases, and are the focus of this grant. Mutations that primarily affect myelinating Schwann cells cause demyelinating forms of CMT;mutations that primarily affect neurons cause the "axonal" forms of CMT. The pace of progress is truly amazing, but this has created a void between what is known and what most health care professionals know about CMT. To address this problem, I will develop and maintain a website that incorporates the following features:

DESCRIPTION (provided by applicant): Recessive mutations in GJC2, the human gene that encodes Cx47, cause Pelizaeus-Merzbacher-like disease (PMLD) and hereditary spastic paraparesis (HSP), presumably related their lack of Cx47 function in oligodendrocytes. How the loss of Cx47 function results in the clinical picture of PMLD is unknown. MRI paints the picture of profound dysmyelination, but this has yet to be confirmed in an autopsy or biopsy. In cellular and animal models, the mutants associated with PMLD impair GJ communication, but we do not yet know why GJ communication is essential for the proper functioning of oligodendrocytes. In this competing renewal, we will determine whether novel Cx47 mutants affect GJ coupling, explore whether O: O coupling a general feature of oligodendrocytes, and the functional significance of O: O coupling. 1. Investigate the molecular defects of hCx47 mutants causing PMLD or familial lymphedema. In the last grant cycle of this grant, we showed that 3 recessive Cx47 mutants associated with PMLD have defective trafficking and do not form functional channels with either Cx47 or Cx43. We will build on this observation by investigating 2 additional missense mutations, 2 frameshift mutations that affect the C-terminus of Cx47, deletion of the PDZ-binding domain (del433-437; a relevant, but not a naturally-occurring mutation), and 6 dominant mutations that cause a completely different phenotype, Familial Lymphedema. The ability of each mutant to form functional GJ plaques will be investigated - by immunostaining, scrape loading, fluorescence recovery after photobleaching (FRAP), and electrophysiology. Whether the dominant mutants have dominant-negative effects on WT Cx47 will also be investigated (by co-staining, co-immunoprecipitation, and electrophysiology). We will also generate lines of transgenic mice that express the Cx47del433-437 or R257C, one of the dominant mutations causing Familial Lymphedema. We will assess the ability of these mutants to rescue the phenotype of Gjc2-null (Gjc2-/-) mice, and of the R257C mutant to worsen the phenotype of Gjc2-heterozygous (Gjc2+/-) mice. 2. Is O: O coupling a general feature of oligodendrocytes? In the last grant cycle of this grant, we showed that O: A coupling in lost in mice lacking both Cx32 and Cx47. We also showed, expectedly, O: O coupling was prominent in the corpus callosum, that O: O coupling was also lost in mice lacking both Cx32 and Cx47, that the GJs directly join intrafascicular oligodendrocytes. The goal of this aim is to determine whether O: O coupling is found in other white matter tracts, and hence is a general feature of intrafascicular oligodendrocytes. To that end, we will measure the diffusion of sulforhodamine-B (SR- B), examine the ultrastructure of O: O junctions, and examine the expression of glial connexins (Cx30, Cx32, Cx43, Cx47) in two tracts - the optic nerve and the ventral funiculus of the cervical spinal cord. 3. The functional significance of O: O coupling. The function of O: A and O: O coupling is uncertain. The experimental results to date support two, non-mutually exclusive, functions - spatial buffering of K+ and metabolic cooperation. K+ clearance is modestly decreased in acute brain slices from mice in which astrocytes lack both Cx43 and Cx30. These mice also provide the best evidence for metabolic coupling, demonstrating that Cx30 and Cx43 are required for the intracellular diffusion of a fluorescent glucose analogue (2-NDBG) to areas of neuropil with increased neuronal activity. We reasoned that O: O coupling might serve a similar purpose for myelinated axons, and propose an experimental analysis of this possibility in this Aim. We will determine whether (a) activity increases the extent of 2-NDBG diffusion, and (b) whether glucose infused into a single oligodendrocyte can rescue axonal conduction in glucose-deprived conditions.

Peripheral neuropathy is one of the most common neurodegenerative diseases in America, with an overall prevalence of 1.66%, and 6.6% in persons older than 60 years. In spite of its prevalence, a cause cannot be identified in a substantial fraction of patients, and unless a treatable cause is identified, treatment is limited to symptoms. These sobering statistics have hung over the field for decades. However, if one accepts the idea that the genetic causes of neuropathy (which are being found at an astounding rate) will likely be informative, then exploiting this rich source of information should provide the keys to unlocking the causes and thus finding new treatments of neuropathies, not only the specific genetic causes, but also the pathways involved in the more common cause, such as diabetic neuropathy. In this view, each new cause of CMT helps to complete the ?puzzle? of how molecular defects cause neuropathy, with many unanticipated surprises. We recently described a new genetic cause of dominantly inherited demyelinating neuropathy in humans ? de novo, dominant mutations in PMP2. PMP2 encodes the myelin protein P2, which is a cytosolic protein that binds to fatty acids and cholesterol. It is minor component of compact myelin; more abundant in the PNS myelin than in CNS myelin. The disease-associated mutations described to date - p.Ile43Asn, p.Thr51Pro, and p.Ile52Thr ? are found in a similar region of the protein, giving rise to the idea that the mutants share a common toxic gain of function in myelinating Schwann cells that results in demyelination. We have made one mutation in mice, p.Ile52Thr, using CRISPR-Cas9, and propose to analyze this mouse model in the following specific aims. Aim 1: Do p.Ile52Thr mutant mice develop a demyelinating neuropathy? We generated four, independent lines of mice harboring the p.Ile52Thr mutation, and have germ line transmission in three of these lines. We will perform pathological and electrophysiological analysis on each of the three lines, comparing homozygous (p.Ile52Thr/p.Ile52Thr), heterozygous (p.Ile52Thr/+), and wild type (WT) (+/+) mice. Although heterozygous mice more accurately model the human disease, we included homozygous mice because they might have a more pronounced phenotype. Aim 2. Are PNS myelin sheaths altered in p.Ile52Thr mutant mice? Because P2 binds to fatty acids and cholesterol, we will investigate the possibility that the p.Ile43Asn mutation alters the lipid composition of myelin, by analyzing nerves by lipidomics and by performing electron microscopy of compact myelin.

Abstract Chronic Inflammatory Demyelinating Polyneuropathy (CIDP) is the most common acquired chronic autoimmune neuropathy. Current treatments for CIDP are non-specific, ineffective in one-third of patients, and do not result in complete remission in most patients. Thus, more effective, mechanism-based therapies are needed, and understanding the immune tolerance defects that result in PNS autoimmunity will enable their development. Studies to date suggest a model in which CD4+ T cells, macrophages, and complement lead to the autoimmune destruction of Schwann cells in the PNS. Our data indicate that Schwann cells unexpectedly undergo changes during autoimmune attack that may expand the inflammatory response. Schwann cells turn on expression of Periostin, a secreted extracellular matrix protein important in chemotaxis of pathogenic macrophages; increase expression of CD49b, an integrin important in binding complement protein C1q; and induce expression of MHC Class II, a molecule required for antigen-presentation to CD4+ T cells. Thus, we hypothesize that Schwann cell-associated changes may promote autoimmunity through increased macrophage recruitment, complement deposition, and CD4+ T cell activation. To test this, we propose to determine whether: i) Schwann cell-specific Periostin expression is sufficient to drive macrophage recruitment and neuropathy development; ii) loss of CD49b in Schwann cells and/or C1q prevents complement activation and protects from neuropathy; and iii) Schwann cell-specific MHCII deficiency dampens CD4+ T cell stimulation and protects against PNS autoimmunity. These Aims will be tested in CIDP mouse models and patient nerve biopsies. This project takes advantage of complementary expertise of the multiple PI's (Dr. Su in PNS autoimmunity and Dr. Scherer in Schwann cell biology) to elucidate the role of Schwann cells in promoting demyelinating neuropathy. Successful completion of the Aims of this project will pave the way to identifying new targets for mechanism-based immunotherapeutic interventions for CIDP. Additionally, findings from these studies will contribute to a broader understanding of how Schwann cells may amplify inflammation in immune- mediated diseases of the PNS.