Abstract
Background: Charcot-Marie-Tooth (CMT) neuropathy is a heterogeneous group of inherited disorders of the peripheral nervous system. The authors recently mapped an autosomal dominant demyelinating form of CMT type 1 (CMT1C) to chromosome 16p13.1-p12.3.
Objective: To find the gene mutations underlying CMT1C.
Methods: The authors used a combination of standard positional cloning and candidate gene approaches to identify the causal gene for CMT1C. Western blot analysis was used to determine relative protein levels in patient and control lymphocyte extracts. Northern blotting was used to characterize gene expression in 1) multiple tissues; 2) developing sciatic nerve; and 3) nerve-crush and nerve-transection experiments.
Results: The authors identified missense mutations (G112S, T115N, W116G) in the LITAFgene (lipopolysaccharide-induced tumor necrosis factor-α factor) in three CMT1C pedigrees. LITAF, which is also referred to as SIMPLE, is a widely expressed gene encoding a 161-amino acid protein that may play a role in protein degradation pathways. The mutations associated with CMT1C were found to cluster, defining a domain of the LITAF protein having a critical role in peripheral nerve function. Western blot analysis suggested that the T115N and W116G mutations do not alter the level of LITAF protein in peripheral blood lymphocytes. The LITAF transcript is expressed in sciatic nerve, but its level of expression is not altered during development or in response to nerve injury. This finding is in stark contrast to that seen for other known genes that cause CMT1.
Conclusions: Mutations in LITAF may account for a significant proportion of CMT1 patients with previously unknown molecular diagnosis and may define a new mechanism of peripheral nerve perturbation leading to demyelinating neuropathy.
Charcot-Marie-Tooth neuropathy (CMT; also called hereditary motor and sensory neuropathy [HMSN]) is a group of disorders characterized by degenerative changes in peripheral nerves leading to progressive distal muscle weakness, atrophy, and sensory loss.1 CMT affects approximately 1 in 2,000 individuals,2 making it one of the most common inherited neurologic diseases. Although multiple modes of inheritance can be encountered, CMT is most frequently transmitted in an autosomal dominant manner. CMT Type I (CMT1) is characterized by demyelination and reduced nerve conduction velocities (NCVs) (typically <38 m/s), whereas CMT Type II (CMT2) denotes patients with predominantly axonal disease and preserved or only mildly reduced NCVs.1
CMT1 has been divided into four subtypes based on genetic mapping studies. CMT1A is associated with a 1.4 megabase (Mb) duplication on chromosome 17p11.2-p123,4⇓ and a gene dosage effect for peripheral myelin protein (PMP22).5-9⇓⇓⇓⇓ PMP22 point mutations have been found in rare CMT1A patients who lack the duplication.10-12⇓⇓ CMT1B results from mutations in the myelin protein zero gene (MPZ) on chromosome 1, which encodes Po, the major structural protein of peripheral myelin (reviewed in Kamholz et al.).13 The CMT1D gene maps to chromosome 10 and is associated with mutations in the early growth response gene-2 (EGR2), also known as Krox-20.14 There remain CMT1 patients lacking mutations in these known genes, and these patients are designated as CMT1C.15
We recently mapped a gene for CMT1C in two pedigrees to chromosome 16p within a 9-cM interval flanked by markers D16S519 and D16S764. The disease-linked haplotypes in these two pedigrees were not conserved, suggesting that the mutation in each family arose independently.16 By inspecting the public human genome database sequence in the CMT1C gene region, we identified more than 20 candidates including the LITAF/SIMPLEgene.17,18⇓ Studies on LITAF in a human colorectal cell line indicated LITAFexpression was increased 10-fold in the presence of the tumor suppressor p53,19 which is known to regulate pathways leading to cellular growth arrest or apoptosis. Given that proliferation and apoptosis are changed in Schwann cells carrying PMP22 alterations,20 we considered LITAF an attractive candidate for CMT1C.
Subjects and methods.
Subjects.
CMT1C pedigrees K1550, K1551, and K2900 are of Irish, English, and Dutch descent, and are depicted in figure 1, A through C. Affected individuals met widely accepted criteria for CMT1 including distal muscle weakness and atrophy, depressed deep tendon reflexes, and sensory impairment.1 The mean ulnar (16.7 m/s [n = 3], 25.3 [n = 8]), median (23 m/s [n = 5], 25.8 m/s [n = 12], and peroneal (20.4 m/s [n = 4], 21 m/s [n = 6]) motor nerve conduction velocities (MNCV) of affected K1550 and K1551 patients are consistent with CMT1. MNCVs for the propositus in K2900 were also reduced (ulnar, 19.2 m/s; peroneal, 15.0 m/s; tibial, 29.2 m/s). There was prominent temporal dispersion and evidence for conduction block, especially in the tibial nerves. One affected individual (2.7) in pedigree K1550 had a sural nerve biopsy taken during reconstructive foot surgery that demonstrated “onion-bulb hypertrophy” typical of demyelinating CMT (figure 2). Control DNA samples for mutational analyses were taken from a collection of predominantly Caucasians of European descent.
Figure 1. CMT1 pedigrees K1551 (A), K1550 (B), and K2900 (C). Affected individuals are denoted by blackened symbols, males are denoted by squares, females are denoted by circles, and deceased persons are indicated by a diagonal line. Underlined numbers indicate individuals screened by restriction endonuclease analysis.
Figure 2. Nerve biopsy from affected K1550 family member. Characteristic onion-bulb formations are evident. Multiple Schwann cell nuclei are indicated by arrows. Original magnification ×295.
Molecular analysis.
Under a protocol of informed consent approved by the institutional review board (IRB) of the University of Washington, Seattle, 15 to 20 mL of blood were obtained by venipuncture for high-molecular-weight DNA as described previously21 and used as a template for PCR. The following PCR primers were employed to amplify exons 2 through 4 of the LITAF (SIMPLE) gene: exon 2, forward 5′-caactgaatttcttatctgg-3′, reverse 5′-gtaaaactggaacgtactgg-3′, anneal 55 °C, 387-bp product; exon 3, forward 5′-atagccagacgatgaacg-3′, reverse 5′-atggtgcagttgagaacc-3′, anneal 53 °C, 385-bp product; exon 4, forward 5′-gaacattttggcagc-3′, reverse 5′-TAATGGTAG-GCACTAAAGG-3′, anneal 59 °C, 636-bp product. The BsrI and NciI restriction analyses were performed directly on the exon 3 amplification product according to the manufacture’s specification (New England BioLabs, Beverly, MA) followed by electrophoresis on 2% agarose gels.
Gene expression.
Young adult Sprague-Dawley rats were anesthetized and the sciatic nerves were transected at the sciatic notch, and both cut ends were ligated and pulled apart to prevent axonal regeneration into the distal stump. The entire distal nerve stump, about 4 cm in length, was harvested during the next 1 to 58 days. Alternatively, sciatic nerves were crushed at the sciatic notch, and the segment distal to the crush was isolated 4 to 58 days later and divided into two 2-cm segments, termed P (the segment immediately adjacent to the crush) and D (the more distal segment). These experiments are more extensively described in the article by Scarlato et al.22 The animal protocols were approved by the Institutional Animal Care and Use Committee at the University of Pennsylvania.
RNA was isolated as previously described by CsCl2 gradient centrifugation.23 For the lesioned adult rat sciatic nerves, total RNA was isolated from the distal stumps of sciatic nerves that were transected or crushed. All lanes contain an equal amount (10 μg) of total RNA. Blots were prehybridized, hybridized, and washed under standard conditions, with a final stringency of 0.2× SSC and 0.1% sodium dodecyl sulfate (SDS) at 65 °C for 30 minutes. 32P-labeled cDNA probes with specific activities of 2 × 109 cpm/μg were prepared using the High Prime system (Roche) according to the manufacturer’s specifications. The following complementary DNAs (cDNAs) were used as a probe: a 1.4-kb fragment of LITAF (BF5666818: cDNA clone courtesy of Dr. Bento Soares); a full-length cDNA of rat myelin protein zero (P0); and a full-length cDNA of rat glyceraldehyde 3-phosphate dehydrogenase (GAPDH).24
Protein extracts and immunoblots.
Blood samples were cleared of red blood cells by lysis in osmotic buffer (PureGene, Minneapolis, MN). Intact lymphocytes remaining in the lysate were then pelleted by centrifugation, washed in phosphate-buffered saline, and lysed in boiling SDS-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer. Fifty micrograms of each extract was resolved on an SDS-PAGE gel and transferred to polyvinylidene fluoride (PVDF) membrane. Blots were then incubated with anti-LITAF monoclonal antibodies (Transduction Labs, San Diego, CA; 1:5000), followed by horseradish peroxidase-conjugated goat antimouse antibodies (Sigma, St. Louis, MO; 1:20,000). Detection was performed with the ECL Plus system (Amersham, Piscataway, NJ).
Results.
Analysis of the three LITAF coding exons and flanking intron nucleotide sequence in pedigrees K1550 and K1551 revealed mutations in exon 3. In K1550, a C-to-A transversion was detected at nt 344 leading to a T115N substitution, and in K1551, a G-to-A-transition at nt 334 resulting in a G112S change (figure 3, A and B). Each of these mutations introduced a novel BsrI restriction endonuclease site, which was utilized to confirm that the mutation cosegregated with disease in both families (see figure 1, A and B). Neither mutation was detected in 200 unrelated control chromosomes. The LITAFsequence was also evaluated in a third CMT1 family (K2900) previously shown not to have alterations in the PMP22, MPZ, or EGR2 genes. The proband in this family (3.2) has decreased NCVs ranging from 15 to 30 m/s. In K2900, a T-to-G transversion was detected at nt 346, causing a W116G substitution (see figure 3, A and B). This mutation introduced a novel NciI site that was used to confirm that the mutation faithfully cosegregated with CMT1C (see figure 1C) and was absent in 100 unrelated control chromosomes. These amino acid residues (G112, T115, and W116) are conserved across human, mouse, rat, and chicken (see figure 3C), suggesting that they play a critical role in LITAF protein function.
Figure 3. LITAF mutations in CMT1 pedigrees. (A) The nucleotide and deduced amino acid sequence of human LITAF. Nucleotide numbering starts with the ORF. The vertical bars correspond to the exon/intron boundaries. The arrow, circle, and triangle indicate residues that are mutated in the K1551, K1550, and K2900 pedigrees. (B) Electropherograms showing heterozygous mutated genomic nucleotide sequences from an affected individual in the K1551, K1550, and K2900 pedigree. Protein alignment shows conservation of the glycine, threonine, and tryptophan residues during evolution. (C) Altered amino acid residues in each of the families is boxed. (D) The LITAF protein is present in peripheral blood lymphocytes from a patient carrying PMP22 duplication (1), unaffected control (2), affected K1550 (3), and affected K2900 (4).
Western blot analysis of peripheral blood lymphocytes indicated that the T115N and W116G substitutions do not appear to alter the LITAF protein level compared to a control individual and an individual carrying the PMP22 duplication (see figure 3D).
Northern blot analysis indicates that a 2.4-kb Litaf message is present at moderate levels in rat sciatic nerve (figure 4A), with expression remaining constant during sciatic nerve development (see figure 4B). Following axotomy of transected sciatic nerve, Litaf expression remained essentially constant for a 48-day time course (see figure 4C). Following crush injury, we observed a general increase in Litaf expression over a 58-day time course, more pronounced proximal vs distal to the injury site (see figure 4D).
Figure 4. Northern blot analysis of Litaf in multiple rat tissue (a), developing sciatic nerve (b), transected adult rat sciatic nerve (c), and crushed adult rat sciatic nerve (d). The number of days after each of these lesions is indicated; the 0 time point is from unlesioned nerves. In crushed nerves, the distal nerve-stumps were divided into proximal (P) and distal (D) segments of equal lengths. The blots were successively hybridized with radiolabeled complementary DNA probes for LITAF, MPZ, and GAPDH.
Discussion.
We have provided evidence that mutations in the LITAF (SIMPLE) gene are the molecular basis of CMT1C. We predict that such mutations may account for a significant proportion of CMT1 patients lacking a mutation in previously known genes (e.g., PMP22, MPZ, Cx32).
Previous studies have shown that the LITAF gene is induced by a variety of cellular signals. For example, expression of human LITAF is stimulated by bacterial outer membrane components18 in a monocyte cell line and by the tumor suppressor p53 in a colorectal line cell.19 The rat orthologue, EET-1 (estrogen-enhanced transcript-1),25 is regulated by estrogen. Several studies suggest that the LITAF transcript is ubiquitous18,19,25,26⇓⇓⇓; however, expression in the sciatic nerve had not been characterized until this report. The fact that Litaf expression was unchanged as a result of nerve injury stands in distinct contrast to other CMT1 genes such as MPZ (see figure 4), PMP22, connexin-32 (GJB1, Cx32), and EGR2, all of which have been found to demonstrate altered expression as a result of nerve injury.24,27,28⇓⇓
The cellular role of LITAFremains to be elucidated. The LITAF gene was originally cloned as a putative nuclear transcription factor involved in tumor necrosis factor-α gene regulation.17,29⇓ However, subsequent studies indicated that the LITAF gene encodes a lysosomal protein, and was referred to as SIMPLE for small integral membrane protein of the lysosome/late endosome.18 Moriwaki et al. isolated LITAF from human monocytes as a Mycobacterium bovis Bacillus Calmette-Guerin cell wall skeleton inducible cDNA displaying significant homology (>85% amino acid) to a previously identified gene in both rat (acc. no. U53184) and mouse (acc. no. NM-019980). LITAF colocalized with LAMP-1 (lysosome-associated membrane protein-1) to perinuclear lysosomes and late endosomes.18
The pathogenic role of LITAF mutations in CMT1C is unknown, but aberrant protein degradation may be involved. The mouse orthologue, Litaf, was isolated as a Nedd4 WW domain-binding protein-3 (N4WBP3).26 The Litaf amino acid residues critical for interaction with Nedd4 are conserved between mouse and human. Nedd4 is an ubiquitin-protein ligase that is involved in coupling ubiquitin to substrate proteins. Following ubiquitination, proteins are targeted for proteosomal or lysosomal degradation or both. Improper protein degradation has been implicated in the pathogenesis of CMT1A. In Schwann cells, PMP22 shows a rapid turnover rate with only a fraction of the protein synthesized actually being targeted to the plasma membrane.30 Overexpression of the PMP22 gene in CMT1A is associated with demyelination and formation of perinuclear protein aggregates.31 Furthermore, inhibition of the ubiquitin-pathway exacerbates aggresome formation in Schwann cells overexpressing PMP22.31 Subcellular localization of LITAF to the lysosomes raises the intriguing possibility that LITAF may play a primary role in the degradation of proteins critical to peripheral nerve function. It is possible that the amino acid substitutions seen in CMT1C patients may reduce binding efficiency with protein interactors, potentially affecting protein degradation
Identification of LITAF (SIMPLE) mutations in CMT1C provides a molecular marker for this disorder and will help bring insights into the function of LITAF (SIMPLE).
Acknowledgments
Postdoctoral funding was provided to Dr. Street during this study by the Charcot-Marie-Tooth Association, the Muscular Dystrophy Association, and the Genetic Approaches to Aging Research Training Grant (NIH AG00057). Postdoctoral funding was provided to Dr. Bennett by the Muscular Dystrophy Association and Dr. Kleopa by the Multiple Sclerosis Society. This study was funded by a Muscular Dystrophy Association Research Grant (P. F. C.); NIH grants NS38181 (P. F. C.), DC02739 (B. L T.), NS2878 (S.S.S.), and HD02274 (CHDD Genetics Core); and research funds from the V.A. Puget Sound Health Care System (T. D. B. and H. P. L).
Acknowledgment
The authors thank the participating families for their cooperation throughout this study.
- Received October 13, 2002.
- Accepted November 4, 2002.
References
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within the last 8 weeks.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.