Skin biopsies demonstrate MPZ splicing abnormalities in Charcot-Marie-Tooth neuropathy 1B

OBJECTIVE

To demonstrate that intronic mutations in the myelin protein zero (MPZ) cause Charcot-Marie-Tooth neuropathy 1B (CMT1B) by disrupting MPZ splicing.

METHODS

We report a family with a T>G transversion at the invariant + 2 position in intron 4 of MPZ (c.614 + 2T>G) that abolishes 5' donor site recognition and is predicted to alter MPZ splicing. We obtained detailed clinical and neurophysiologic analysis of the family. We performed skin biopsies to investigate splicing abnormalities, MPZ protein levels, and localization in myelinated nerves.

RESULTS

Patients developed a late onset neuropathy with minimally slow nerve conduction velocities. Skin biopsies confirmed the predicted skipping of exon 4 and downstream frameshift of the mutant MPZ. Quantitative immuno-EM demonstrated normal nerve MPZ levels, suggesting that the mutant MPZ was transported to compact myelin.

CONCLUSIONS

Intronic mutations cause CMT1B by disrupting splicing and certain MPZ mutations may cause neuropathy by interacting with the wild type MPZ in the extracellular space of compact myelin.

Protein zero is necessary for E-cadherin-mediated adherens junction formation in Schwann cells

Protein Zero (P0), the major structural protein in the peripheral nervous system (PNS) myelin, acts as a homotypic adhesion molecule and is thought to mediate compaction of adjacent wraps of myelin membrane. E-Cadherin, a calcium-dependent adhesion molecule, is also expressed in myelinating Schwann cells in the PNS and is involved in forming adherens junctions between adjacent loops of membrane at the paranode. To determine the relationship, if any, between P0-mediated and cadherin-mediated adhesion during myelination, we investigated the expression of E-cadherin and its binding partner, beta-catenin, in sciatic nerve of mice lacking P0 (P0(-/-)). We find that in P0(-/-) peripheral myelin neither E-cadherin nor beta-catenin are localized to paranodes, but are instead found in small puncta throughout the Schwann cell. In addition, only occasional, often rudimentary, adherens junctions are formed. Analysis of E-cadherin and beta-catenin expression during nerve development demonstrates that E-cadherin and beta-catenin are localized to the paranodal region after the onset of myelin compaction. Interestingly, axoglial junction formation is normal in P0(-/-) nerve. Taken together, these data demonstrate that P0 is necessary for the formation of adherens junctions but not axoglial junctions in myelinating Schwann cells.

Mutations in the cytoplasmic domain of P0 reveal a role for PKC-mediated phosphorylation in adhesion and myelination

Mutations in P0 (MPZ), the major myelin protein of the peripheral nervous system, cause the inherited demyelinating neuropathy Charcot-Marie-Tooth disease type 1B. P0 is a member of the immunoglobulin superfamily and functions as a homophilic adhesion molecule. We now show that point mutations in the cytoplasmic domain that modify a PKC target motif (RSTK) or an adjacent serine residue abolish P0 adhesion function and can cause peripheral neuropathy in humans. Consistent with these data, PKCalpha along with the PKC binding protein RACK1 are immunoprecipitated with wild-type P0, and inhibition of PKC activity abolishes P0-mediated adhesion. Point mutations in the RSTK target site that abolish adhesion do not alter the association of PKC with P0; however, deletion of a 14 amino acid region, which includes the RSTK motif, does abolish the association. Thus, the interaction of PKCalpha with the cytoplasmic domain of P0 is independent of specific target residues but is dependent on a nearby sequence. We conclude that PKC-mediated phosphorylation of specific residues within the cytoplasmic domain of P0 is necessary for P0-mediated adhesion, and alteration of this process can cause demyelinating neuropathy in humans.

A molecular basis for hereditary motor and sensory neuropathy disorders

Charcot-Marie-Tooth disease (CMT), or inherited peripheral neuropathies, is one of the most frequent genetically inherited neurologic disorders, with a prevalence of approximately one in 2500 people. CMT is usually inherited in an autosomal dominant fashion, although X-linked and recessive forms of CMT also exist. Over the past several years, considerable progress has been made toward understanding the genetic causes of many of the most frequent forms of CMT, particularly those caused by mutations in Schwann cell genes inducing the demyelinating forms of CMT, also known as CMT1. Because the genetic cause of these disorders is known, it is now possible to study how mutations in genes encoding myelin proteins cause neuropathy. Identifying these mechanisms will be important both for understanding demyelination and for developing future treatments for CMT.

Gtx, an oligodendrocyte-specific homeodomain protein, has repressor activity

Myelin, a multilamellar membrane structure that facilitates nerve conduction, is synthesized in the central nervous system (CNS) by oligodendrocytes. Gtx, a member of the homeodomain family of transcriptional factors, is a candidate regulator of myelin gene expression, because it is uniquely expressed in myelinating oligodendrocytes in postnatal rodent brain. To analyze the regulatory activity of Gtx, we first identified the optimal Gtx-binding sequence using an in vitro DNA-binding assay. This sequence, (A/T)TTAATGA, contains a TAAT core and is similar, but not identical, to that of other homeodomain protein binding sites. When coexpressed in cultured cells along with a minimal promoter containing five tandem repeats of this optimal Gtx-binding sequence, Gtx demonstrated repressor activity, which was also present when Gtx was tethered to DNA by way of the strong GAL4 DNA-binding domain. Truncations of the GAL4-Gtx fusion identified a portable repressor domain within a relatively proline/alanine-rich region N-terminal to the Gtx homeodomain. Cotransfection of a Gtx expression vector into a variety of cell lines, including oligodendrocytes, along with constructs containing portions of the PLP, MBP, or Gtx promoters fused to a reporter gene, however, did not modulate transcription from any of these promoter constructs. These data support the notion that the oligodendrocyte-specific homeodomain protein Gtx can act as a transcriptional repressor. In addition, they suggest that interaction of Gtx with other, as yet undefined, transcriptional regulators modifies Gtx activity in oligodendrocytes.

Neurological dysfunction and axonal degeneration in Charcot-Marie-Tooth disease type 1A

Charcot-Marie-Tooth disease type 1A (CMT1A), the most frequent form of CMT, is caused by a 1.5 Mb duplication on the short arm of chromosome 17. Patients with CMT1A typically have slowed nerve conduction velocities (NCVs), reduced compound motor and sensory nerve action potentials (CMAPs and SNAPs), distal weakness, sensory loss and decreased reflexes. In order to understand further the molecular pathogenesis of CMT1A, as well as to determine which features correlate with neurological dysfunction and might thus be amenable to treatment, we evaluated the clinical and electrophysiological phenotype in 42 patients with CMT1A. In these patients, muscle weakness, CMAP amplitudes and motor unit number estimates correlated with clinical disability, while motor NCV did not. In addition, loss of joint position sense and reduction in SNAP amplitudes also correlated with clinical disability, while sensory NCV did not. Taken together, these data strongly support the hypothesis that neurological dysfunction and clinical disability in CMT1A are caused by loss or damage to large calibre motor and sensory axons. Therapeutic approaches to ameliorate disability in CMT1A, as in amyotrophic lateral sclerosis and other neurodegenerative diseases, should thus be directed towards preventing axonal degeneration and/or promoting axonal regeneration.

Proteolipid protein mRNA stability is regulated by axonal contact in the rodent peripheral nervous system

Proteolipid protein (PLP) and its alternatively spliced isoform, DM20, are the main intrinsic membrane proteins of compact myelin in the CNS. PLP and DM20 are also expressed by Schwann cells, the myelin-forming cells in the PNS, and are necessary for normal PNS function in humans. We have investigated the expression of PLP in the PNS by examining transgenic mice expressing a LacZ transgene under the control of the PLP promoter. In these animals, myelinating Schwann cells expressed beta-galactosidase more prominently than nonmyelinating Schwann cells. PLP/DM20 mRNA levels, but not those of LacZ mRNA, increased during sciatic nerve development and decreased after axotomy, with resultant Wallerian degeneration. PLP/DM20 transcription rates, in nuclear run off experiments, however, did not increase in developing rat sciatic nerve despite robust increases in PLP/DM20 mRNA levels during the same period. In RNAse protection studies, PLP mRNA levels fell to undetectable levels following nerve transection whereas levels of DM20 were essentially unchanged despite both being transcribed from the same promoter. Finally, cotransfection studies demonstrated that PLP-GFP, but not DM20-GFP mRNA is down-regulated in Schwann cells cultured in the absence of forskolin. Taken together these data demonstrate that steady state levels of PLP mRNA are regulated at a posttranscriptional level in Schwann cells, and that this regulation is mediated by Schwann cell-axonal contact. Since the difference between these two mRNAs is a 105-bp sequence in PLP and not in DM20, this sequence is likely to play a role in the regulation of PLP mRNA.

Additional copies of the proteolipid protein gene causing Pelizaeus-Merzbacher disease arise by separate integration into the X chromosome

The proteolipid protein gene (PLP) is normally present at chromosome Xq22. Mutations and duplications of this gene are associated with Pelizaeus-Merzbacher disease (PMD). Here we describe two new families in which males affected with PMD were found to have a copy of PLP on the short arm of the X chromosome, in addition to a normal copy on Xq22. In the first family, the extra copy was first detected by the presence of heterozygosity of the AhaII dimorphism within the PLP gene. The results of FISH analysis showed an additional copy of PLP in Xp22.1, although no chromosomal rearrangements could be detected by standard karyotype analysis. Another three affected males from the family had similar findings. In a second unrelated family with signs of PMD, cytogenetic analysis showed a pericentric inversion of the X chromosome. In the inv(X) carried by several affected family members, FISH showed PLP signals at Xp11.4 and Xq22. A third family has previously been reported, in which affected members had an extra copy of the PLP gene detected at Xq26 in a chromosome with an otherwise normal banding pattern. The identification of three separate families in which PLP is duplicated at a noncontiguous site suggests that such duplications could be a relatively common but previously undetected cause of genetic disorders.

Absence of P0 leads to the dysregulation of myelin gene expression and myelin morphogenesis

P0, the major peripheral nervous system (PNS) myelin protein, is a member of the immunoglobulin supergene family of membrane proteins and can mediate homotypic adhesion. P0 is an essential structural component of PNS myelin; mice in which P0 expression has been eliminated by homologous recombination (P0-/-) develop a severe dysmyelinating neuropathy with predominantly uncompacted myelin. Although P0 is thought to play a role in myelin compaction by promoting adhesion between adjacent extracellular myelin wraps, as an adhesion molecule it could also have a regulatory function. Consistent with this hypothesis, Schwann cells in adult P0-/- mice display a novel molecular phenotype: PMP22 expression is down-regulated, MAG and PLP expression are up-regulated, and MBP expression is unchanged. As in quaking viable mutant mice (qk(v)), which have uncompacted myelin morphologically similar to that found in P0-/- mice, neither the qKI-6 or qKI-7 proteins are expressed in P0-/- peripheral nerve. In addition to these changes in gene expression in the P0 knockout, PLP/DM-20 accumulates in the endoplasmic reticulum of P0-/- Schwann cells, whereas MAG accumulates in redundant loops of uncompacted myelin, not at nodes of Ranvier or Schmidt-Lantermann incisures. Taken together, these results demonstrate that P0 is involved, either directly or indirectly, in the regulation of both myelin gene expression and myelin morphogenesis.

GFAP-positive and myelin marker-positive glia in normal and pathologic environments

The data herein demonstrate that in addition to the well-characterized myelin marker-positive, glial fibrillary acidic protein (GFAP)-negative, membrane sheet-bearing oligodendrocytes, another type of myelin marker-positive, process-bearing glia exists in normal and pathologic conditions. This second type of myelin marker-positive glia expresses GFAP, and therefore these cells have been referred to as mixed phenotype glia. Although mixed phenotype glia have been documented previously, their identity and function have remained a mystery. The goal of this immunocytochemical study was to further characterize these cells. Using the MBPlacZ transgenic mouse in which beta-galactosidase is under the control of the myelin basic protein (MBP) gene promoter, GFAP-positive/beta-galactosidase-positive and myelin/oligodendrocyte-specific protein (MOSP)-positive/beta-galactosidase-positive cells were detected in subcortical white matter and in perivascular locations within cerebral white and gray matter. In cultures prepared from highly enriched myelin marker-positive immature glia, mixed phenotype glia were detected that were GFAP-positive and either MOSP-, MBP-, O1-, and O4-positive. The expression of multiple myelin markers by mixed phenotype glia may suggest that these cells are of oligodendrocyte origin. Increased numbers of MOSP-positive/GFAP-positive mixed phenotype glia were detected in sections from adult hypomyelinated brain from shiverer, quaking, and PKU mice compared to myelinated control adult mouse brain. Similarly, cultures from control brain exposed to elevated pH for 2-3 weeks showed dramatically increased numbers of mixed phenotype glia (80%) compared to control (<10%). Increased numbers of mixed phenotype glia also were detected in shiverer cultures (40%). Since increases in the number of mixed phenotype glia occur in shiverer, quaking, and PKU mouse brain, these data suggest that mixed phenotype glia contribute to gliosis in pathologic white matter.

Charcot-Marie-Tooth disease type 1: molecular pathogenesis to gene therapy

Charcot-Marie-Tooth disease type 1 (CMT1) is caused by mutations in the peripheral myelin protein, 22 kDa (PMP22) gene, protein zero (P0) gene, early growth response gene 2 (EGR-2) and connexin-32 gene, which are expressed in Schwann cells, the myelinating cells of the peripheral nervous system. Although the clinical and pathological phenotypes of the various forms of CMT1 are similar, including distal muscle weakness and sensory loss, their molecular pathogenesis is likely to be quite distinct. In addition, while demyelination is the hallmark of CMT1, the clinical signs and symptoms of the disease are probably produced by axonal degeneration, not demyelination itself. In this review we discuss the molecular pathogenesis of CMT1, as well as approaches to an effective gene therapy for this disease.

Cell cycle arrest induced by ectopic expression of p27 is not sufficient to promote oligodendrocyte differentiation

Oligodendrocyte differentiation is accompanied by dramatic changes in gene expression as well as cell cycle arrest. To determine whether cell cycle arrest is sufficient to induce the changes in cell phenotype associated with differentiation, we inhibited oligodendrocyte precursor proliferation in vitro by overexpressing p27, a cyclin kinase inhibitor, using a recombinant adenovirus. Ectopic expression of p27 efficiently inhibited oligodendrocyte precursor cell division, even in the presence of exogenous mitogens, by blocking the activity of the cyclin-dependent kinase, cdk2. Although the cells had stopped dividing, they did not express galactocerebroside (GalC) or myelin basic protein (MBP), changes associated with oligodendrocyte differentiation, suggesting that they had not differentiated. After removal of exogenous mitogens, however, adenovirus-expressing oligodendrocyte precursors differentiated with a temporal profile similar to that of control, uninfected oligodendrocytes, as indicated by expression of GalC and MBP. We conclude that cell cycle arrest is not sufficient to induce differentiation of dividing oligodendrocyte precursors, and that modulation of additional, as yet unknown, signaling pathways is required for this to occur.

The molecular pathogenesis of Pelizaeus-Merzbacher disease

In 1885, Pelizaeus described 5 boys in a single family with nystagmus, spastic quadriparesis, ataxia, and delay in cognitive development. In 1910, Merzbacher reexamined this family, which then included 14 affected individuals, including 2 girls, and found that all affected family members shared a common female ancestor. Also, he noted that the disease was passed exclusively through the female line without male-to-male transmission. Pathological analysis of brain tissue from one affected individual showed that most of the central white matter lacked histochemical staining for myelin, although there were occasional small regions of preserved myelin, giving the sections a "tigroid" appearance. The description of this family provides the clinical, genetic, and pathological basis for Pelizaeus-Merzbacher disease (PMD): an X-linked disorder of myelination classically characterized by nystagmus, spastic quadriparesis, ataxia, and cognitive delay in early childhood.

Regulation of myelin-specific gene expression. Relevance to CMT1

Schwann cells, the myelinating cells of the peripheral nervous system, are derived from the neural crest. Once neural crest cells are committed to the Schwann cell fate, they can take on one of two phenotypes to become myelinating or nonmyelinating Schwann cells, a decision that is determined by interactions with axons. The critical step in the differentiation of myelinating Schwann cells is the establishment of a one-to-one relationship with axons, the so-called "promyelinating" stage of Schwann cell development. The transition from the promyelinating to the myelinating stage of development is then accompanied by a number of significant changes in the pattern of gene expression, including the activation of a set of genes encoding myelin structural proteins and lipid biosynthetic enzymes, and the inactivation of a set of genes expressed only in immature or nonmyelinating Schwann cells. These changes are regulated mainly at the transcriptional level and also require continuous interaction between Schwann cells and their axons. Two transcription factors, Krox 20 (EGR2) and Oct 6 (SCIP/Tst1), are necessary for the transition from the promyelinating to the myelinating stage of Schwann cell development. Krox 20, expressed in myelinating but not promyelinating Schwann cells, is absolutely required for this transition, and myelination cannot occur in its absence. Oct 6, expressed mainly in promyelinating Schwann cells and then down-regulated before myelination, is necessary for the correct timing of this transition, since myelination is delayed in its absence. Neither Krox 20 nor Oct 6, however, is required for the initial activation of myelin gene expression. Although the mechanisms of Krox 20 and Oct 6 action during myelination are not known, mutation in Krox 20 has been shown to cause CMT1, further implicating this protein in the pathogenesis of this disease. Identifying the molecular mechanisms of Krox 20 and Oct 6 action will thus be important both for understanding myelination and for designing future treatments for CMT1. Point mutlations in the genes encoding the myelin proteins PMP22 and P0 cause CMT1A without a gene duplication and CMT1B, respectively. Although the clinical and pathological phenotypes of CMT1A and CMT1B are similar, their molecular pathogenesis is quite different. Point mutations in PMP22 alter the trafficking of the protein, so that it accumulates in the endoplasmic reticulum (ER) and intermediate compartment (IC). Mutant PMP22 also sequesters its normal counterpart in the ER, further reducing the amount of PMP22 available for myelin synthesis at the membrane, and accounting, at least in part, for its severe effect on myelination. Mutant PMP22 probably also activates an ER-to-nucleus signal transduction pathway associated with misfolded proteins, which may account for the decrease of myelin gene expression in Schwann cells in Trembler mutant mice. In contrast, absence of expression of the homotypic adhesion molecule, P0, in mice in which the gene has been inactivated, produces a unique pattern of Schwann cell gene expression, demonstrating that P0 plays a regulatory as well as a structural role in myelination. Whether this role is direct, through a P0-mediated adhesion pathway, or indirect, through adhesion pathways mediated by cadherins or integrins, however, remains to be determined. The molecular mechanisms underlying dysmyelination in CMT1 are thus complex, with pleitropic effects on Schwann cell physiology that are determined both by the type of mutation and the protein mutated. Identifying these molecular mechanisms, however, are important both for understanding myelination and for designing future treatments for CMT1. Although demyelination is the hallmark of CMT1, the clinical signs and symptoms of this disease are probably produced by axonal degeneration, not demyelination. (ABSTRACT TRUNCATED)

Overcoming cellular immunity to prolong adenoviral-mediated gene expression in sciatic nerve

In a previous report, we demonstrated that a first generation (E1- and E3-deleted) recombinant adenovirus can transduce expression of the E. coli lacZ gene into Schwann cells, both in vitro and in vivo, suggesting that this method might be useful for future therapy of peripheral neuropathy, including CMT1. Adenoviral-mediated gene transfer was limited, however, by demyelination and Wallerian degeneration at the site of virus injection, as well as by attenuation of viral gene expression over time. In our current work we have optimized adenoviral-mediated gene expression after intraneural injection into sciatic nerve. Using an improved injection protocol, peak expression of lacZ occurs between 10 and 14 days after injection of 2-week-old animals, decreases thereafter, and there is minimal associated tissue injury. In contrast, very few adenoviral-infected Schwann cells are found in nerves of adult animals 10 days after injection, probably due to immune clearance of viral-infected cells. Consistent with this notion, high levels of lacZ are found in sciatic nerve 30 days after injection of adult SCOD mice, which have a genetic defect in both cellular and humoral immunity, of adult beta 2 microglobulin-deficient mice (beta 2 M-/-), which have a genetic defect in cellular immunity, or of adult mice treated with the immunosuppressing agent FK506. In addition, adenoviral-infected Schwann cells co-cultured with axons in vitro, in the absence of a host immune response, ensheath axons and express lacZ for at least 8 weeks. These data thus demonstrate that expression of first generation recombinant adenovirus in sciatic nerve in adult mice, as in other tissues, is limited mainly by the host cellular immune response to the virus, which can be overcome by attenuation of host cell-mediated immunity. Adenoviral vectors might thus be used to modulate Schwann cell gene expression in patients with peripheral neuropathy after appropriate immunosuppression.

The absence of myelin P0 protein produces a novel molecular phenotype in Schwann cells

In order to better understand the pathogenesis of demyelination in P0 knockout (P0-/-) mice, we analyzed the myelin gene expression and the localization of myelin proteins in P0 null mouse sciatic nerve. We have demonstrated that the severe demyelinating neuropathy of P0-knockout mouse is associated with changes in the program of myelin gene expression. Some changes in myelin gene expression occur early, others occur during adulthood. We also provide evidence that the absence of P0 is associated with changes in the localization of specific paranodal proteins in the peripheral nerve. These data suggest that P0 plays an important role, either directly or indirectly, in the program of Schwann cell gene expression and in the specific distribution of peripheral myelin proteins. Furthermore, myelin gene dysregulation and improper localization of paranodal proteins may account, in part, for the pathogenesis of demyelination in P0-knockout mice, as well as in human demyelinating peripheral neuropathy associated with mutations in the P0 gene.

Peripheral neuropathy caused by proteolipid protein gene mutations

Pelizaeus-Merzbacher disease (PMD) is a dysmyelinating disorder of the central nervous system typically caused by duplications or missense mutations of the proteolipid protein (PLP) gene. Most investigators have found that peripheral nerve function and structure is normal in PMD patients. We have found that null mutations of the PLP gene cause demyelinating peripheral neuropathy, whereas duplications and a proline 14 to leucine mutation do not affect nerve function. A family with a nonsense mutation at position 144, which affects only PLP but not the alternatively spliced gene product DM20, has a very mild syndrome, including normal peripheral nerve function. Our findings suggest that DM20 alone is sufficient to maintain normal nerve function and that there may be domains of PLP/DM20 that have a relatively more active role in the peripheral nervous system compared with that in the central nervous system.

Correlation between weakness and axonal loss in patients with CMT1A

We have developed a protocol to measure the progression of disability in patients with Charcot Marie Tooth (CMT) disease, particularly CMT1 over a several year period. Because CMT1 is a chronic disease, the natural history of changes occurring in such a brief period are not well understood, making clinical trials for CMT1 patients difficult to evaluate. We hypothesize that weakness in CMT1 correlates with axonal loss secondary to the abnormalities in Schwann cell myelin gene expression, which cause the disease. To test this hypothesis, we elected to carefully evaluate CMT patients by various modalities to measure strength, sensory loss, and axonal loss and demyelination and to compare these modalities to determine whether they correlated with findings on clinical examination. As suspected, patient weakness correlates more with secondary axonal loss than with demyelination, even though the primary abnormality in CMT1 is demyelination.

Modulation of cell-mediated immunity prolongs adenovirus-mediated transgene expression in sciatic nerve

In a previous report, we demonstrated that a first-generation (E1- and E3-deleted) recombinant adenovirus can transduce expression of the E. coli lacZ gene into Schwann cells, both in vitro and in vivo, suggesting that this method might be useful for future therapy of peripheral neuropathy, including CMT1. Adenovirus-mediated gene transfer was limited, however, by demyelination and Wallerian degeneration at the site of virus injection, as well as by attenuation of viral transgene expression over time. In our current work we have optimized adenoviral vector-mediated transgene expression after intraneural injection into sciatic nerve. Using an improved injection protocol, peak expression of lacZ occurs between 10 and 14 days after injection of 2-week-old rats, decreases thereafter, and there is minimal associated tissue injury. In contrast, few lacZ-expressing Schwann cells are found in nerve of adult animals 10 days after injection, probably owing to immune clearance of virus-infected cells. Consistent with this notion, high levels of LacZ are found in sciatic nerve 30 days after injection of adult SCID mice, which have a genetic defect in both cellular and humoral immunity, of adult beta2-microglobulin-deficient mice (beta2M4-/-), which have a genetic defect in cellular immunity, or of adult mice treated with the immunosuppressing agent FK506. In addition, adenovirus-infected Schwann cells cocultured with axons in vitro, in the absence of a host immune response, ensheathe axons and express lacZ for at least 8 weeks. These data thus demonstrate that lacZ transgene expression of first-generation recombinant adenovirus in sciatic nerve in adult mice, as in other tissues, is limited mainly by the host cellular immune response to the virus, which can be overcome by attenuation of host cell-mediated immunity. Adenoviral vectors might thus be used to modulate Schwann cell gene expression in patients with peripheral neuropathy after appropriate immunosuppression.

MEBA derepresses the proximal myelin basic protein promoter in oligodendrocytes

The central nervous system expression of myelin basic protein (MBP) is restricted to oligodendrocytes and is developmentally regulated; these regulatory features are transcriptionally mediated. We have previously shown that the proximal 149 nucleotides of the MBP promoter were both necessary and sufficient to activate the transcription of MBP in cultured oligodendrocytes, but not in other cell types. Sequences within the distal portion of this promoter, which contains a nuclear factor 1 (NF1) binding site, repressed activation of the MBP promoter in Cos-7 cells, but not in oligodendrocytes. We now describe a sequence upstream of and partially overlapping the NF1 site that activates the MBP promoter in oligodendrocytes, but not in Cos-7 cells. A protein complex binds to this site, designated MEBA (myelinating glia-enriched DNA binding activity), and is enriched in nuclear extracts prepared from the brain, oligodendrocytes, and Schwann cells. The amount of MEBA parallels MBP expression and myelinogenesis in the developing brain and parallels new MBP expression as purified oligodendrocytes differentiate. Mutational analyses of binding and function distinguish MEBA, an activator, from NF1, a repressor of MBP transcription, and suggest that MEBA consists of at least two proteins. Because the binding sites of MEBA and NF1 overlap, we suggest that MEBA may either compete with or modify NF1 binding, thereby activating the MBP promoter in oligodendrocytes.

Regulation of oleoyl-CoA synthesis in the peripheral nervous system: demonstration of a link with myelin synthesis

We studied the regulation of oleic acid synthesis in the PNS. During mouse postnatal development, the proportion of 18:1 rises in the sciatic nerve from 17% at 5 days of age to 33% at 25 days. However, this rise does not occur in the dysmyelinating mutant mouse trembler. In normal mouse development, the total stearoyl-CoA desaturase (SCD) activity measured in sciatic nerve homogenates is high during the first 3 weeks. Yet in trembler nerves, this SCD activity represents only 15% of normal values. Using the RT-PCR technique, we demonstrate that the SCD2 isoform is predominantly expressed in the PNS. Northern blot analysis showed that the mRNA levels for SCD2 parallel those of other specific myelin proteins in both normal mouse and trembler mutant development. Similar experiments in a rat demyelination-remyelination model confirmed that SCD2 mRNA levels are regulated in the PNS in a similar manner to myelin-specific proteins.

A minimal human MBP promoter-lacZ transgene is appropriately regulated in developing brain and after optic enucleation, but not in shiverer mutant mice

Previous studies, both in vitro and in vivo, suggest that small portions of the mouse myelin basic protein (MBP) promoter are sufficient to activate regulated expression of MBP. To confirm our previous in vitro studies, we prepared transgenic mice with short regions of the human MBP promoter fused to the lacZ reporter gene. We found that 750 nucleotides of the proximal human MBP promoter is sufficient to activate oligodendrocyte-specific, developmentally regulated expression of lacZ in three independent lines. This promoter, however, does not activate expression of lacZ in Schwann cells in peripheral nerve or in adult mouse brain. The relative levels of beta-galactosidase specific activity, mRNA, and transcription parallel those of MBP mRNA during myelinogenesis. Thus, we exploited this transgene as a quantitative tool to evaluate the response to stimuli known to affect myelination. Transgene expression is reduced 75 % after optic enucleation, as previously reported for levels of MBP mRNA, indicating that axons signal to this portion of the proximal MBP promoter to fully activate MBP expression during myelinogenesis. Instead, in adult shiverer mice, another setting in which MBP transcription is modulated, transgene expression is not increased, in contrast to the increased transcriptional activation of MBP previously reported in these mice. These data suggest that the regulatory region that mediates transcriptional activation of the MBP gene is modular, since discrete subregions are required for activation in Schwann cells, during myelinogenesis in oligodendrocytes, during maintenance myelination in adult brain, and in the dysmyelinating mutant shiverer mouse.

Evidence that the homeodomain protein Gtx is involved in the regulation of oligodendrocyte myelination

We have investigated the patterns of postnatal brain expression and DNA binding of Gtx, a homeodomain transcription factor. Gtx mRNA accumulates in parallel with the RNAs encoding the major structural proteins of myelin, myelin basic protein (MBP), and proteolipid protein (PLP) during postnatal brain development; Gtx mRNA decreases in parallel with MBP and PLP mRNAs in the brains of myelin-deficient rats, which have a point mutation in the PLP gene. Gtx mRNA is expressed in differentiated, postmitotic oligodendrocytes but is not found in oligodendrocyte precursors or astrocytes. These data thus demonstrate that Gtx is expressed uniquely in differentiated oligodendrocytes in postnatal rodent brain and that its expression is regulated in parallel with the major myelin protein mRNAs, encoding MBP and PLP, under a variety of physiologically relevant circumstances. Using a Gtx fusion protein produced in bacteria, we have confirmed that Gtx is a sequence-specific DNA-binding protein, which binds DNA sequences containing a core AT-rich homeodomain binding site. Immunoprecipitation of labeled DNA fragments encoding either the MBP or PLP promoter regions with this fusion protein has identified several Gtx-binding fragments, and we have confirmed these data using an electrophoretic mobility shift assay. In this way we have identified four Gtx binding sites within the first 750 bp of the MBP promoter and four Gtx binding sites within the first 1. 3 kb of the PLP promoter. In addition, inspection of the PLP promoter sequence demonstrates the presence of six additional Gtx binding sites. These data, taken together, strongly suggest that Gtx is important for the function of differentiated oligodendrocytes and may be involved in the regulation of myelin-specific gene expression.

Proteolipid protein is necessary in peripheral as well as central myelin

Alternative products of the proteolipid protein gene (PLP), proteolipid protein (PLP) and DM20, are major components of compact myelin in the central nervous system, but quantitatively minor constituents of Schwann cells. A family with a null allele of PLP has a less severe CNS phenotype than those with other types of PLP mutations. Moreover, individuals with PLP null mutations have a demyelinating peripheral neuropathy, not seen with other PLP mutations of humans or animals. Direct analysis of normal peripheral nerve demonstrates that PLP is localized to compact myelin. This and the clinical and pathologic observations of the PLP null phenotype indicate that PLP/DM20 is necessary for proper myelin function both in the central and peripheral nervous systems.

Heterozygous P0 knockout mice develop a peripheral neuropathy that resembles chronic inflammatory demyelinating polyneuropathy (CIDP)

Demyelinating peripheral neuropathies are clinically divided into inherited and acquired types. Inherited demyelinating neuropathies are caused by mutations in genes expressed by myelinating Schwann cells, whereas acquired ones, including chronic inflammatory demyelinating polyneuropathy (CIDP), are probably caused by autoimmune mechanisms. We find that heterozygous P0 knockout (P0+/-) mice develop a neuropathy that resembles CIDP. By one year of age, P0+/- mice develop severe, asymmetric slowing of motor nerves, with temporal dispersion or conduction block, which are features of acquired demyelinating neuropathies including CIDP. Moreover, morphological analysis of affected nerves reveals severe and selective demyelination of motor fibers, focal regions of demyelination, and inflammatory cells. These data suggest that immune-mediated mechanisms may contribute to the pathogenesis of the neuropathy in P0+/- mice.

Toward a transgenic mouse model of remyelination

The molecular mechanisms necessary for remyelination by oligodendrocytes remain unexplored. We previously characterized a myelin basic protein promoter-lacZ (MBP-lacZ) transgene whose expression is regulated uniquely during development, and also in pathological situations, suggesting that it may be a useful reporter of molecular mechanisms during remyelination. As a first step toward creating a transgenic mouse model of remyelination, we cultured oligodendrocytes from these transgenic mice and showed that expression of MBP-lacZ appeared in parallel with a marker of oligodendrocyte maturation, galactocerebroside (GC). In addition, basic fibroblast growth factor blocked the expression of both MBP-lacZ and GC in these cells. Therefore, expression of MBP-lacZ reflects not only the developmental stage of oligodendrocytes, but also extrinsic influences on oligodendrocytes. These data suggest that MBP-lacZ may be a useful marker in transgenic mouse models of remyelination.

Neuropsychiatry of 18q- syndrome

Our understanding of neuropsychiatric abnormalities in patients with deletions of the long arm of chromosome 18 (18q- syndrome) is based mainly on sporadic case reports. We characterized the neuropsychiatric phenotype in 27 patients across a wide age range (2-47 years) with breakpoints ranging from 18q22.3-18q21.2. Adaptive behavior scores (Vineland Composite) were significantly higher in females than in males (62 +/- 5 vs. 43 +/- 3). Intelligence ranged from borderline to severely deficient (IQ, 73- < 40), with academic achievement similarly impaired. Performance in specific neuropsychological functions, including attention, novel problem solving, memory, language, visuomotor integration, and fine motor dexterity, was consistently in the moderately-to-severely impaired range. Behavioral problems were common in both sexes, including aggressivity, hyperactivity, and temper tantrums. Contrary to the few previous reports, we found no evidence of psychosis in any patients. In a subset of patients selected on the basis of no prior knowledge of behavioral problems, 1 of 16 patients (6%) had autism, as defined by the Autistic Diagnostic Interview--Revised (ADI-R) [Lord et al., 1994: J Autism Dev Disord 24:659-685]. Thus, the prevalence of autism in 18q- syndrome is probably no greater than that in other developmental disabilities with a similar level of cognitive impairment. In contrast to what has been believed since 18q- was first described 30 years ago, we found no relationship between chromosome deletion size and any measure of cognition or behavior; nor were there any correlations between any of these measures with the presence or absence of abnormalities on MRI or somatosensory-evoked potentials.

Axon-Schwann cell interactions regulate the expression of c-jun in Schwann cells

The transcription factor c-jun is selectively expressed by non-myelinating Schwann cells in normal peripheral nerve, and be "denervated," previously myelinatng Schwann cells, after axotomy. When axons regenerate into the distal nerve-stump, the expression of c-jun declines as Schwann cells remyelinate axons. Treating cultured Schwann cells with forskolin, a drug that mimics many of the effects of axon-Schwann cell interactions, decreases the expression of myelin-specific genes. Overexpressing c-jun in cultured Schwann cells, however, does not decrease the expression of a myelin basic protein promoter-reporter construct, indicating that c-jun expression may not directly regulate myelin-specific gene expression. These data suggest that c-jun is invloved in regulating the phenotype of non-myelinating and denervated Schwann cells.

White matter changes associated with deletions of the long arm of chromosome 18 (18q- syndrome): a dysmyelinating disorder?

PURPOSE

To evaluate the MR findings in the central nervous systems of patients with deletions of the long arm of chromosome 18 (18q- syndrome).

METHODS

Sixteen patients with 18q- syndrome ranging in age from 3 to 46 years (mean, 17 years) were studied with high-field-strength MR imaging. Images were analyzed for abnormal T2 hyperintensity in the white matter, abnormal T2 hypointensity in the deep gray matter, and atrophy.

RESULTS

Ten of 16 patients had abnormal white matter. Diffuse, bilaterally symmetric deep white matter T2 hyperintensity, most pronounced in the periventricular regions, was most common, noted in eight cases. Focal deep white matter lesions and/or abnormalities involving the subcortical white matter were also noted in four cases. The cerebellum, brain stem, and corpus callosum were spared. Ventriculomegally associated with volume loss, and abnormal T2 hypointensity in the basal ganglia and/or thalami were each present in 11 patients.

CONCLUSIONS

The 18q- syndrome is associated with white matter disease and abnormal T2 hypointensity in the deep gray matter. The basis for the white matter abnormalities is unknown, but may be related to one of the two genes for myelin basic protein included in the deleted segment of chromosome 18.

Analysis of clinical variation seen in patients with 18q terminal deletions

Twenty-six patients with deletions of 18q were analyzed at the clinical and molecular levels in an attempt to delineate regions of chromosome 18 important to the 18q- syndrome phenotype. Molecular cytogenetic analysis was carried out using fluorescence in situ hybridization (FISH), and deletions ranging from 18q21.1-qter to 18q22.3-qter were detected. The parental origin of the deletions was determined by the analysis of inheritance of microsatellite markers. No correlation between size, parental origin, or severity of the resulting phenotype was found. The results suggest that a critical region for 18q- syndrome lies in the most distal portion of 18q and that it confers susceptibility for the various clinical manifestations of the 18q- syndrome when present in one copy.

An adenoviral vector can transfer lacZ expression into Schwann cells in culture and in sciatic nerve

Although a number of genetic defects in the P0, peripheral myelin protein-22, and connexin-32 genes recently were shown to cause the demyelinating forms of Charcot-Marie-Tooth disease, there is yet no effective treatment for these patients. Recent studies showed that replication defective adenoviral vectors can efficiently introduce genes into muscle, brain, lung, and other tissues, suggesting that this vector system may be useful for the treatment of a number of genetic diseases. In this work, we demonstrated that a replication deficient adenovirus expressing the Escherichia coli beta-galactosidase gene (AdCMVLacZ) can introduce genes into Schwann cells, in culture as well as in sciatic nerve. Schwann cells cultured at a multiplicity of infection of 250:1 did not demonstrate cytopathic effects. Following injection of AdCMVLacZ into sciatic nerve of rats, lacZ-expressing, myelinating Schwann cells could be detected for at least 45 days. These data suggest that in the future, these vectors may be useful both in perturbing Schwann cell gene expression and in designing therapies for the treatment of Charcot-Marie-Tooth disease.

Regulation of neurofibromin expression in rat sciatic nerve and cultured Schwann cells

Loss of function mutations at the NF1 locus may act intrinsically in Schwann cells to cause the formation of benign Schwann cell tumors (neurofibromas) in patients with type 1 neurofibromatosis. To identify contexts in Schwann cells in which such mutations may play an important role, we measured the levels of NF1 mRNA and neurofibromin in rat sciatic nerve during development, after axotomy, and in cultured rat Schwann cells. NF1 mRNA was present in developing sciatic nerve throughout the period of active Schwann cell proliferation and myelination. After nerve transection, no alteration in NF1 message level was detected, but neurofibromin levels increased, as assessed by immunohistochemistry and Western blotting, suggesting that, in vivo, neurofibromin expression in Schwann cells is post-transcriptionally induced during Wallerian degeneration. Cultured rat Schwann cells constitutively expressed NF1 mRNA and neurofibromin. Schwann cell proliferation induced by exposure to serum and forskolin was not associated with changes in NF1 mRNA or neurofibromin expression, whereas Schwann cell proliferation induced by extracts of embryonic brain membranes was associated with increased NF1 message and neurofibromin expression. Thus, Schwann cells, both in vivo and in vitro, express NF1 mRNA constitutively; the expression of NF1 mRNA and neurofibromin is modulated by only some mitogenic stimuli in Schwann cells.

The single lamprey neurofilament subunit (NF-180) lacks multiphosphorylation repeats and is expressed selectively in projection neurons

The lamprey is considered the most primitive living vertebrate and its neurofilaments (NFs) are unique in being homopolymers of a single 180 kDa subunit (NF-180). Previous immunologic studies have suggested that the sidearm of NF-180 is highly phosphorylated selectively in the largest diameter axons. We report in this study the isolation and characterization of cDNA clones encoding the NF-180 lamprey protein. In situ hybridization with digoxigenin-labeled cRNA revealed NF-180 message exclusively in neurons with long axons, such as reticulospinal neurons and cranial motor neurons. The core of NF-180 was similar in structure to those of mammalian neurofilaments, but surprisingly, the carboxy sidearm lacked the multiphosphorylation repeats characteristic of higher vertebrate and invertebrate neurofilaments. Overall there was a paucity of potential phosphorylation sites in the NF-180 carboxy-terminus compared to NF-M and NF-H of mammals, fish and squid. This, along with the highly acidic nature of the NF-180 sidearm, makes it unlikely that phosphorylation of sidearm residues regulates interfilament spacing and axon diameter through global electrostatic repulsion of the carboxy-terminus away from the filament backbone. Furthermore, the expression of a single neurofilament subunit in the lamprey that is most similar to the NF-M of higher vertebrates suggests that all three mammalian neurofilament subunits evolved from a single NF-M-like precursor.

Expression of P0 protein mRNA along rat sciatic nerve during development

The temporal and spatial expression of P0 protein mRNA in the spinal nerve roots and in distal (popliteal trifurcation) segments of the developing rat sciatic nerve was examined using in situ hybridization. P0 mRNA was undetectable in the PNS portions at day 16 post-fertilization (E16) but present in the roots as well as in the sciatic nerve by E18. While Schwann cells were labelled by in situ hybridization at E18, it was not until birth (P1) that cells revealed the typical clusters of silver grains in a perinuclear distribution. Hybridization signal rose between P5 and P10, than decreased at P40, with no significant difference of P0 mRNA levels between the different PNS portions at each age studied. These findings demonstrate the lack of proximo-distal gradient in abundance of P0 mRNA during rat PNS development and add further insight in understanding the dynamics of axonal regulation of myelination in the PNS.

Transcriptional regulation of the rat PLP promoter in primary cultures of oligodendrocytes

Proteolipid protein (PLP) is the major intrinsic membrane protein of CNS myelin and is expressed in oligodendrocytes as part of a coordinate program of myelin-specific gene activation. In order to identify the DNA sequences and proteins involved in the regulation of PLP transcription, we have analyzed the 5' flanking sequences of the rat PLP gene by transient transfections into primary cultures of developing oligodendrocytes, the glial tumor line, C6, and L cells. High levels of expression of the CAT reporter gene in oligodendrocytes and C6 cells were obtained with constructs containing both 4270 and 225 bp of PLP promoter. A fusion construct containing 1061 bp of the PLP promoter, however, showed two-fold lower CAT expression. In addition, the activity of these promoter fusion constructs in oligodendrocytes was 2.5-4.6 higher than that observed in C6 cells, while very little expression was found in L cells. These data suggest that 225 bp of PLP promoter is sufficient for oligodendrocyte-specific regulation of PLP expression. Furthermore, both positive and negative elements within the PLP promoter are involved in this process. Finally, primary cultures of developing oligodendrocytes are a useful model system for the analysis of myelin-specific gene activation.

Expression of growth-associated protein-43 kD in Schwann cells is regulated by axon-Schwann cell interactions and cAMP

We have examined the regulation of growth-associated protein 43 kD (GAP-43) in rat Schwann cells. In unlesioned adult nerves, GAP-43-immunoreactivity was restricted to non-myelinating Schwann cells and unmyelinated axons. When adult nerves were transected to cause permanent axotomy, previously myelinating Schwann cells expressed progressively more GAP-43-immunoreactivity over 3 weeks, and GAP-43 mRNA levels increased over a similar time course. The peak level of GAP-43 mRNA occurred at least 2 weeks later than that of nerve growth factor receptor, another marker of denervated Schwann cells. In contrast, after nerve-crush, which allows axonal regeneration, many fewer Schwann cells had GAP-43-immunoreactivity, and the amount of GAP-43 mRNA was markedly lower than in transected nerves. Forskolin, a drug that activates adenylate cyclase and mimics many effects of axon-Schwann cell interactions, markedly reduced GAP-43-immunoreactivity and mRNA expression in cultured Schwann cells, whereas interleukin-1 had no effect. These data demonstrate that axon-Schwann cell interactions inhibit the expression of GAP-43 in Schwann cells and that this effect is mimicked by forskolin.

A novel cyclic AMP response element, CACTTGATC, mediates forskolin induction of the myelin basic protein promoter in the rat Schwannoma line, D6P2T

The rat Schwannoma cell line D6P2T constitutively expresses the mRNA encoding the major myelin protein, P0, but only expresses the mRNA encoding myelin basic protein (MBP) after exposure to forskolin or other substances that raise the levels of intracellular cyclic AMP. In this study we have investigated the molecular basis for forskolin induction of MBP transcription in D6P2T cells. We have found that a 9-bp sequence element, CACTTGATC, located between nucleotides -85 and -77 in the MBP promoter, is necessary for forskolin induction of chloramphenicol acetyltransferase (CAT) expression after transient transfection of MBP promoter-CAT fusion constructs into D6P2T cells. Although similar DNase I footprints, one of which is located within the above 9-bp sequence element, are produced by nuclear extracts prepared from both forskolin-treated and untreated cells, this same sequence can be shown to interact with a forskolin-inducible protein complex using an electrophoretic mobility shift assay. In addition, mutation of this 9-bp sequence abolishes both formation of this new protein--DNA complex and forskolin-inducible CAT expression from the heterologous SV40 promoter. Finally, we have shown that the appearance of this forskolin-inducible protein--DNA complex precedes that of MBP mRNA. Taken together, these data strongly support the notion that the induction of MBP transcription by forskolin in D6P2T cells is mediated by the binding of a forskolin-inducible protein complex to the MBP promoter sequence CACTTGATC.

Differential regulation of the 2',3'-cyclic nucleotide 3'-phosphodiesterase gene during oligodendrocyte development

The two major isoforms of 2',3'-cyclic nucleotide phosphodiesterase (CNP), 48 and 46 kDa, have recently been shown to be produced from a single gene by alternative splicing. In addition, messenger RNA encoding the larger isoform is transcribed from a separate promoter, approximately 1 kb upstream from that encoding the smaller isoform. We have investigated the expression of these two CNP isoforms and have found that they are differentially expressed during the process of oligodendrocyte maturation. In oligodendrocyte precursors, only the mRNA encoding the larger protein is found. At the time of oligodendrocyte differentiation, however, both CNP mRNAs are induced. These patterns of CNP expression are likely due to stage-specific transcriptional regulation of the two CNP promoters during the process of oligodendrocyte differentiation.

Beta 4 integrin expression in myelinating Schwann cells is polarized, developmentally regulated and axonally dependent

In developing and regenerating peripheral nerve, Schwann cells interact with axons and extracellular matrix in order to ensheath and myelinate axons. Both of these interactions are likely to be mediated by adhesion molecules, including integrins, which mediate cell-cell and cell-extracellular matrix interactions. Recently, the beta 4 integrin subunit was reported to be expressed by Schwann cells in peripheral nerve. We have examined the expression of beta 4, beta 1 and their common heterodimeric partner, the alpha 6 integrin subunit, in developing and regenerating rat peripheral nerve. beta 4 and alpha 6 are enriched in peripheral nerve and they co-localize at the abaxonal surface of myelinating Schwann cells, opposite the Schwann cell basal lamina, which contains possible ligands of alpha 6 beta 4. In contrast, beta 4 and alpha 6 are expressed in a different pattern in non-myelinating Schwann cells. The level of beta 4, but not alpha 6 or beta 1 mRNAs, increases progressively in developing nerves, reaching a peak in adult nerves well after the peak of the myelin-specific mRNAs. After axotomy, the expression of beta 4 mRNA and protein, but not alpha 6 or beta 1 mRNAs, fall rapidly but subsequently are reinduced by regenerating axons. Similarly, in cultured Schwann cells, the expression of beta 4 mRNA, but not alpha 6 mRNA, is significantly modulated by forskolin, a drug that elevates cAMP and mimics some of the effects of axonal contact. beta 4 integrin expression in Schwann cells, therefore, is regulated by Schwann cell-axon interactions, which are known to be critical in determining the Schwann cell phenotype. Furthermore, the polarized expression of alpha 6 beta 4 to the abaxonal surface of myelinating Schwann cells suggests that alpha 6 beta 4 may mediate in part the morphological changes required of Schwann cells in the process of myelination in the peripheral nervous system.

Organization of the mouse GP42/Basigin gene: a member of the Ig superfamily

We have mapped and sequenced the GP42/Basigin gene isolated from a Balb/C mouse genomic library. The genomic organization and upstream, putative regulatory, regions of this gene have not been previously reported. Our data show that exon 5 of the GP42/Basigin gene encodes the carboxy proximal half of the second Ig-like domain, the highly conserved transmembrane region and a portion of the cytoplasmic tail. This inclusion of Ig-like and other functional domains in a single exon is unusual. Splice junction analysis indicates that two reported alternate GP42/Basigin cDNA isoforms are likely due to cloning artifacts. In addition, we find that GP42/Basigin is polymorphic in mice. Our data also support the proposal that the transmembrane domain and portions of the cytoplasmic region of GP42/Basigin have been evolutionarily conserved.

Axons regulate Schwann cell expression of the POU transcription factor SCIP

SCIP (suppressed cAMP-inducible POU) is a POU domain transcription factor expressed by Schwann cells. Drugs that elevate intracellular cAMP, such as forskolin, increase the expression of SCIP and partially mimic the inductive effects of axons on Schwann cell gene expression. Thus, SCIP may be involved in a differentiation pathway in Schwann cells that is activated by axons. We have examined this issue by studying SCIP expression in developing, degenerating, and regenerating rat peripheral nerves, and in Schwann cell-neuron cocultures. High levels of SCIP mRNA were detected in developing and regenerating nerves, and axotomy at these times caused the level of SCIP mRNA to plummet. Similarly, there were many SCIP-immunoreactive Schwann cell nuclei in developing and regenerating nerves, and their number fell sharply after axotomy. SCIP-immunoreactive Schwann cells were associated with axons in developing and regenerating nerves, and in Schwann cell-neuron cocultures. These data demonstrate that axons upregulate the expression of SCIP in Schwann cells, and that SCIP is expressed in Schwann cells that ensheathe axons. Thus, SCIP may mediate some of the changes in Schwann cell gene expression that accompany axonal ensheathment.