HMG Advance Access published February 7, 2013

Human Molecular Genetics, 2013 doi:10.1093/hmg/ddt022 116

Motor and sensory neuropathy due to myelin infolding and paranodal damage in a transgenic mouse model of Charcot MarieTooth disease type 1C
Samuel M. Lee1,{, Di Sha1,{, Anum A. Mohammed1, Seneshaw Asress2, Jonathan D. Glass2,3, Lih-Shen Chin1, and Lian Li1,
Department of Pharmacology, 2Department of Neurology and 3Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA 30322, USA
Received July 7, 2012; Revised January 15, 2013; Accepted January 22, 2013
1

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Charcot Marie Tooth disease type 1C (CMT1C) is a dominantly inherited motor and sensory neuropathy. Despite human genetic evidence linking missense mutations in SIMPLE to CMT1C, the in vivo role of CMT1C-linked SIMPLE mutations remains undetermined. To investigate the molecular mechanism underlying CMT1C pathogenesis, we generated transgenic mice expressing either wild-type or CMT1C-linked W116G human SIMPLE. Mice expressing mutant, but not wild type, SIMPLE develop a late-onset motor and sensory neuropathy that recapitulates key clinical features of CMT1C disease. SIMPLE mutant mice exhibit motor and sensory behavioral impairments accompanied by decreased motor and sensory nerve conduction velocity and reduced compound muscle action potential amplitude. This neuropathy phenotype is associated with focally infolded myelin loops that protrude into the axons at paranodal regions and near Schmidt Lanterman incisures of peripheral nerves. We nd that myelin infolding is often linked to constricted axons with signs of impaired axonal transport and to paranodal defects and abnormal organization of the node of Ranvier. Our ndings support that SIMPLE mutation disrupts myelin homeostasis and causes peripheral neuropathy via a combination of toxic gain-of-function and dominant-negative mechanisms. The results from this study suggest that myelin infolding and paranodal damage may represent pathogenic precursors preceding demyelination and axonal degeneration in CMT1C patients.

INTRODUCTION
Charcot Marie Tooth disease (CMT), also known as hereditary motor and sensory neuropathy, encompasses a genetically heterogeneous group of inherited disorders of the peripheral nervous system (PNS) (1,2). CMT is categorized into the demyelinating type, which accounts for 80% of CMT cases and the axonal degeneration type, which accounts for 20% of CMT cases (3). CMT type 1C (CMT1C) is a dominantly inherited, demyelinating type of peripheral neuropathy characterized by slowed motor and sensory nerve conduction velocity with typical CMT clinical symptoms, including progressive motor weakness and sensory loss (4 8). Human

genetic studies have revealed that CMT1C is linked to missense mutations in small integral membrane protein of lysosome/late endosome [SIMPLE; also known as lipopolysaccharide-induced TNF-a factor (LITAF)], a ubiquitously expressed protein of unknown function (4 8). Our recent study indicates that endogenous SIMPLE is an early endosomal membrane protein (9) rather than a lysosomal/ late endosomal protein as previously suggested (10). We found that CMT1C-linked mutations map in and around the transmembrane domain of SIMPLE and that these mutations cause mislocalization of SIMPLE protein from the early endosome to the cytosol in cultured cells (9). The in vivo role of CMT1C-linked SIMPLE mutations remains undetermined.

To whom correspondence should be addressed at: 5109 Rollins Research Center, 1510 Clifton Road, Atlanta, GA 30322-3090, USA. Tel: + 1 4047270361; Fax: + 1 4047270365; Email: chinl@pharm.emory.edu; lianli@pharm.emory.edu S.M.L. and D.S. authors contributed equally to this work.

# The Author 2013. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Human Molecular Genetics, 2013

Molecular analysis of the genotype phenotype relationship in hereditary peripheral neuropathy is essential for a mechanistic understanding of CMT pathogenesis. The identied linkage of heterozygous SIMPLE missense mutations to autosomal dominant CMT1C (4 8) raises two possibilities: CMT1C may be due to haplo-insufciency of SIMPLE or dominant effects of the SIMPLE mutants. To distinguish these possibilities and determine the in vivo role of the identied human SIMPLE mutations, we generated transgenic mice expressing CMT1C-linked human SIMPLE W116G mutant or human SIMPLE wild-type (WT) protein. Characterization of these transgenic mice reveals that expression of human SIMPLE W116G mutant, but not human SIMPLE WT, causes a late-onset motor and sensory neuropathy in mice that recapitulates key clinical features of CMT1C disease. SIMPLE W116G mutant mice exhibit motor and sensory nerve conduction defects and impaired motor and sensory performance. The observed motor and sensory impairments are associated with focally infolded myelin loops that protrude into the axons at paranodal regions and near Schmidt Lanterman incisures of motor and sensory nerves. By immunohistochemical analysis, we show that the SIMPLE W116G mutant protein, like endogenous SIMPLE protein, is localized in noncompact myelin cytoplasmic regions of myelinating Schwann cells but is absent in axons. Our ndings support that SIMPLE mutation causes CMT1C pathogenesis via a combination of toxic gain-of-function and dominant-negative mechanisms.

receptor subfamily 1, group D, member 2) (Fig. 1C). This integration site was conrmed by PCR reactions using primers located in the anking sequences of the integration site at both 5 -end and 3 -end of the SIMPLE W116G transgene (Fig. 1C and D). Quantitative immunoblot analysis indicated that the expression level of human SIMPLE W116G mutant protein relative to the level of endogenous mouse SIMPLE protein in sciatic nerves is 22% for homozygous (SIMPLEW116G/W116G) mice and 10% for heterozygous (SIMPLEW116G/+) mice at 3 months of age (Fig. 1E and F). We also established four transgenic lines expressing human SIMPLE WT protein and selected the line with an expression level similar to that in the SIMPLE W116G mutant mice (Fig. 1E; Supplementary Material, Fig. S1) for the analyses described below. The expression level of human SIMPLE WT protein relative to the level of endogenous mouse SIMPLE protein in sciatic nerves is 22% for homozygous (SIMPLEWT/WT) mice and 11% for heterozygous (SIMPLEWT/+) mice at 3 months of age (Fig. 1E and F). We found that the expression levels of human SIMPLE WT and SIMPLE W116G mutant proteins in sciatic nerves of the transgenic mice did not change signicantly between 3 months and 1 year of age (Fig. 1E G). Endogenous SIMPLE is localized to early endosomes in myelinating Schwann cells but is absent in myelin sheath or axons The cellular and subcellular localization of endogenous SIMPLE protein in myelinated peripheral nerves remains poorly characterized. Therefore, we performed immunouorescence confocal microscopic analyses to determine endogenous SIMPLE protein distribution in sciatic nerves from adult non-transgenic mice and rat. We found that endogenous SIMPLE exhibits a punctate staining pattern, and double immunostaining analyses revealed that SIMPLE is localized in myelinating Schwann cells labeled by various Schwann cell markers (Fig. 2A and C E) but not in axons labeled by neurolament H (Fig. 2B). We observed no colocalization of SIMPLE with the compact myelin marker myelin basic protein (MBP) (Fig. 2A), indicating that SIMPLE is not a structural component of myelin sheath. We found that SIMPLE is localized in the Dlg1-positive abaxonal cytoplasmic region (Fig. 2C) at the outer side of the compact myelin (Fig. 2A) and in the adaxonal cytoplasmic region next to myelin-associated glycoprotein (MAG)-positive membrane (Fig. 2D) at the inner side of the compact myelin facing the axon (Fig. 2A and B). Our results also showed that SIMPLE is enriched in the cytoplasmic regions at Schmidt-Lanterman incisures (Fig. 2C and D) and paranodal domains (Fig. 2E). Consistent with our previous report that SIMPLE is an early endosomal membrane protein (9), we observed a substantial colocalization of SIMPLE with the early endosome marker Rab5 (Fig. 2F). Triple immunostaining analyses conrmed the colocalization of SIMPLE with Rab5 in S100-positive myelinating Schwann cells (Fig. 2G). To further characterize the membrane localization of SIMPLE in sciatic nerves, we performed density gradient fractionation analysis and found that SIMPLE co-fractionated with Rab5-positive early endosomal membranes, but not with MAG-positive plasma

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

RESULTS
Generation of transgenic mice expressing human SIMPLE WT or SIMPLE W116G mutant To investigate the pathogenic role of CMT1C-linked SIMPLE mutation in vivo, we generated transgenic mice expressing either N-terminal hemagglutinin (HA)-tagged human SIMPLE W116G mutant or SIMPLE WT protein under the control of the human cytomegalovirus (CMV) promoter (Fig. 1A). The CMV promoter was used to drive ubiquitous expression of SIMPLE transgenes because endogenous SIMPLE is a ubiquitously expressed protein (4,9). Transgenic founder mice were identied by PCR analysis of tail genomic DNAs (Fig. 1B) and bred with FVB/N mice to establish transgenic lines on a pure FVB/N background. The expression of HA-tagged human SIMPLE W116G mutant or SIMPLE WT protein was conrmed by immunoblot analysis of sciatic nerve (Fig. 1E) and Schwann cell lysates (Supplementary Material, Fig. S1) from transgenic mice using our anti-SIMPLE antibody that recognizes both mouse and human SIMPLE proteins (9). We established three transgenic lines expressing human SIMPLE W116G mutant and selected the line with the highest expression of SIMPLE mutant protein (Fig. 1E) for the experiments described below. To exclude the possibility of integration site effects, we performed genome walking PCR and sequencing analyses to identify the transgene integration site in this SIMPLE W116G transgenic line. We found that the SIMPLE W116G transgene inserted into a non-coding region on mouse chromosome 14 between genes Thrb (thyroid hormone receptor beta) and Nr1d2 (nuclear

Human Molecular Genetics, 2013

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 1. Generation of SIMPLE WT and SIMPLE W116G transgenic mice. (A) Schematic of transgenic constructs encoding HA-tagged human SIMPLE W116G or SIMPLE WT under the control of the CMV promoter. The positions of the PCR primer pairs used for amplication of the transgene are indicated. (B) Identication of transgenic mice by PCR analysis of tail genomic DNAs using transgene-specic primer pairs (a) and (b), yielding 196-bp and 244-bp PCR products, respectively. The endogenous (endo.) mouse SIMPLE gene was detected using a pair of primers targeting a mouse SIMPLE-specic genomic region, yielding a 236-bp PCR product. (C) Schematic illustrating the integration of the SIMPLE W116G transgene into a non-coding region on mouse chromosome 14 between genes Thrb and Nr1d2. The genes are depicted with 5 3 orientation by lled arrows. Although the transgene is depicted by a single lled arrow, there may be multiple tandem repeats. The location and orientation of the PCR primers used for conrmation of the transgene integration site are indicated by open arrows. (D) The identied SIMPLE W116G transgene integration site was independently conrmed by PCR analysis of genomic DNAs from the indicated heterozygous and homozygous transgenic mice or non-transgenic control (CTRL) mice. PCR amplication using primers A and D yields the non-transgenic genomic PCR product AD with the expected size of 849 bp. Primers A and B were used in the PCR reaction that anks the integration site at the 5 -end of the transgene, yielding the 550-bp PCR product AB, and primers C and D were used for conrming the integration site at the 3 -end of the transgene, yielding the 1050-bp PCR product CD. The asterisk indicates a non-specic PCR product. (E) Analysis of SIMPLE expression in sciatic nerves from the indicated 3-month-old (3 months) or 1-year-old (1 year) heterozygous and homozygous transgenic mice or non-transgenic control (CTRL) mice by immunoblotting with antibodies against SIMPLE and actin. Endo., endogenous. (F and G) Quantitative analysis of the expression level of HA-tagged human SIMPLE WT or SIMPLE W116G proteins relative to the level of endogenous mouse SIMPLE protein (F) or actin (G). Data represent mean + SEM. (n 3 mice per genotype/age group).

membranes derived from the inner surface of myelin sheath, Schmidt-Lanterman incisures and paranodal loops (Fig. 2H). Together, these results indicate that SIMPLE is localized to the early endosomes in non-compact myelin cytoplasmic regions of myelinating Schwann cells but is absent in myelin membranes or axons. SIMPLE W116G mutant, but not SIMPLE WT, is partially mislocalized from the membrane to the cytosol in myelinated peripheral nerves of transgenic mice Next, we performed double labeling immunouorescence confocal microscopic analyses to assess the localization of human

SIMPLE WT and SIMPLE W116G proteins in sciatic nerves from SIMPLEWT/WT and SIMPLEW116G/W116G transgenic mice. We found that HA-tagged SIMPLE WT and SIMPLE W116G mutant proteins, like the endogenous mouse SIMPLE protein, are localized in non-compact myelin cytoplasmic regions of myelinating Schwann cells such as Schmidt-Lanterman incisures (Fig. 3A) but not in axons (Fig. 3B). These results conrm that the human SIMPLE WT and SIMPLE W116G proteins are both targeted in a similar manner as the endogenous SIMPLE protein to the cytoplasm of myelinating Schwann cells in our transgenic mice. Despite similar staining patterns, the punctate staining of HA-SIMPLE W116G mutant in SIMPLEW116G/W116G mice

Human Molecular Genetics, 2013

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 2. Localization of endogenous SIMPLE on early endosomes in non-compact myelin cytoplasmic regions of myelinating Schwann cells. (AG) Teased sciatic nerves from 3-month-old non-transgenic mice (A, B and D F) or rats (C and G) were stained with anti-SIMPLE antibodies (green) in combination with anti-MBP (A), NF-H (B), Dlg1 (C), MAG (D and E) or Rab5 (F and G) antibodies (red) and anti-S100 (G) antibodies (blue). For each panel, 510 sciatic nerve bers from at least three non-transgenic mice or rats were analyzed, and the staining of a representative ber was shown. The locations of Schmidt-Lanterman incisures (arrows), nodes of Ranvier (arrows with double arrowhead), and colocalization of SIMPLE with Rab5 (arrowheads) were indicated. The insets (A D) are 2-fold magnications of the original image. Ab, abaxonal domain of Schwann cells; Ad, adaxonal domain of Schwann cells. Scale bar, 5 mm. (H) Membrane fractions of sciatic nerves from 3-month-old non-transgenic mice were separated on a 1030% linear Optiprep gradient into 36 fractions, with fraction 1 corresponding to the top of the gradient. Equal volumes of each fraction were subjected to SDSPAGE followed by immunoblot analysis. The experiment was repeated three times, and the result of a representative experiment was shown.

was weaker than that of HA-SIMPLE WT in SIMPLEWT/WT mice (Fig. 3A). Because there was no signicant difference between HA-SIMPLE WT and HA-SIMPLE W116G protein levels in these mice (Fig. 1C E), the weaker staining raised the possibility that SIMPLE W116G mutation may cause SIMPLE mislocalization to the cytosol in transgenic mice as we have previously observed in transfected cells (9). To test this possibility, we performed subcellular fractionation analyses to compare the localization of SIMPLE WT and mutant proteins in sciatic nerves from SIMPLEWT/WT and SIMPLEW116G/W116G mice. We found that human SIMPLE WT, like endogenous mouse SIMPLE, was exclusively localized to the membrane fraction in SIMPLEWT/WT mice. In contrast, human SIMPLE W116G protein was partially mislocalized to the cytosol in the SIMPLEW116G/W116G mice (Fig. 3C). These results, together with the immunostaining data (Figs 2 and 3A and B), support that CMT1C-linked SIMPLE W116G mutation causes partial mislocalization of human SIMPLE protein from the early endosomal membrane to the cytosol in myelinating Schwann cells. Our subcellular

fractionation analyses showed that endogenous mouse SIMPLE protein remained fully associated with the membranes in SIMPLEW116G/W116G mice (Fig. 3C), indicating that the membrane localization of endogenous mouse SIMPLE protein in myelinated peripheral nerves was not affected by the expression of human SIMPLE W116G mutant protein. SIMPLE W116G mutant mice, but not SIMPLE WT transgenic mice, exhibit motor and sensory impairments SIMPLE W116G mutant and SIMPLE WT transgenic mice were viable, fertile and born according to expected Mendelian ratios. They developed normally, and necropsy of these mice showed normally formed organs without any abnormality at 3, 12 and 15 months of age. Histological analysis revealed abnormalities in peripheral nerves (described later) but not in brain or spinal cord of SIMPLE W116G mutant mice (Supplementary Material, Fig. S2 and data not shown). Although no observable behavioral abnormality was observed at the age of 3 and 6 months, SIMPLEW116G/W116G mice showed

Human Molecular Genetics, 2013

In addition to motor abnormalities, we observed that SIMPLEW116G/W116G mice occasionally self-mutilated their tails (Fig. 4D), which could be a consequence of paresthesia reported in patients with CMT1C (5,8). Analysis of nociceptive sensory function by the tail-ick test revealed that SIMPLEW116G/W116G mice (Fig. 4E), but not SIMPLEWT/WT mice (Supplementary Material, Fig. S3D), performed signicantly worse than their non-transgenic littermates, suggesting impaired sensory function. Together, these results indicate that SIMPLEW116G/W116G mice have motor and sensory impairments that are consistent with the clinical features of human patients with CMT1C motor and sensory neuropathy (4 8). SIMPLE W116G mutant mice, but not SIMPLE WT transgenic mice, have motor and sensory nerve conduction defects The pathophysiological hallmark of demyelinating CMT is slowed motor nerve conduction velocity (MNCV , 38 m/s) (2,19 21). Human CMT1C patients carrying SIMPLE W116G or other mutations show reduced MNCVs with 100% penetrance (4 8). To determine whether SIMPLE W116G mutant mice recapitulate this CMT1C phenotype, we performed electrophysiological analyses of motor nerve conduction in mutant mice and controls at 3 months and 1 year of age. In 3-month-old animals, the MNCVs and compound muscle action potential (CMAP) amplitudes were not signicantly different in SIMPLEW116G/W116G and SIMPLEW116G/+ mice compared with their non-transgenic littermates (Fig. 5B and C). At 1 year of age, however, SIMPLEW116G/W116G mice showed signicantly reduced MNCVs (Fig. 5A and B) and CMAP amplitudes (Fig. 5A and C) compared with their nontransgenic littermates. The 1-year-old SIMPLEW116G/+ mice also had signicantly lower MNCVs than those of the nontransgenic controls (Fig. 5B), although their CMAP amplitudes were not signicantly altered (Fig. 5C). In contrast, SIMPLEWT/ WT and SIMPLEWT/+ mice showed no signicant difference in the MNCV or the CMAP amplitude compared with the nontransgenic controls at 1 year of age (Supplementary Material, Fig. S4A and B). In addition to motor nerve conduction defects, electrophysiological analyses of tail sensory nerve conduction also revealed a signicant decrease in the sensory nerve conduction velocity (SNCV) in the SIMPLEW116G/W116G and SIMPLEW116G/+ mice compared with their non-transgenic littermates at the age of 1 year but not 3 months (Fig. 6A and B). Although the sensory nerve action potential (SNAP) amplitudes of the SIMPLEW116G/W116G and SIMPLEW116G/+ mice did not signicantly differ from those of the nontransgenic controls (Fig. 6C), we found that sensory nerve action potentials could not be evoked in 2 of 11 SIMPLEW116G/W116G mice tested (Fig. 6A and D). The inability to evoke sensory nerve action potentials in a small percentage of CMT1C patients has been previously reported (4) and is correlated with a loss of sensory function. Interestingly, one of the mice with no evoked sensory nerve action potentials exhibited self-injurious behavior shown in Figure 4D, suggesting that the self-mutilation behavior may be resulted from a loss of sensory nerve conduction. In contrast to the SIMPLE W116G mutant mice, SIMPLEWT/WT and SIMPLEWT/+ mice

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 3. Localization of human SIMPLE WT and SIMPLE W116G proteins in myelinating Schwann cells of transgenic mice. (A and B) Teased sciatic nerves from the indicated 3-month-old transgenic mouse or non-transgenic control (CTRL) mouse were stained with anti-HA (green) and anti-MAG (red) antibodies (A) or with anti-HA (green) and anti-NF-H (red) antibodies (B). For each panel, 5 10 teased sciatic nerve bers per mouse from three mice per genotype were analyzed, and the staining of a representative ber was shown. Insets are 2-fold magnications of the original image. Arrows indicate the locations of Schmidt-Lanterman incisures. Scale bar, 5 mm. (C) Post-nuclear supernatants (T) of sciatic nerves from the indicated 1-year-old transgenic mice were separated into cytosol (C) and membrane (M) fractions. Aliquots representing an equal percentage of each fraction were analyzed by immunoblotting with antibodies against HA, SIMPLE, and the cytosolic protein DJ-1. The experiment was repeated three times, and the result of a representative experiment was shown.

abnormal clenching of toes and clasping of hind limbs upon tail suspension at 1 year of age, with some of them exing all four limbs and even their whole body (Fig. 4A and B; Supplementary Material, Video S1). This phenotype is also seen in other mouse models of CMT (11 15) and is suggestive of abnormal motor function. In contrast, non-transgenic control mice and SIMPLEWT/WT mice extended their hind limbs and paws outward upon tail suspension (Fig. 4A and B; Supplementary Material, Fig. S3A; Supplementary Material, Video S2). To further evaluate the motor function of transgenic mice, we performed the rotarod test and found that SIMPLEW116G/W116G mice (Fig. 4C and Supplementary Material, Fig. S3B), but not SIMPLEWT/WT mice (Supplementary Material, Fig. S3C), showed impaired motor performance after the fourth trial when compared with their non-transgenic littermates. The phenotype of impaired motor performance in the later trials of rotarod test is also seen in other mouse models of CMT (16,17), and this phenotype may reect reduced neuromuscular function associated with muscle fatigue which has been observed in patients suffering from CMT and other types of peripheral neuropathies (18). We also found that SIMPLEW116G/+ mice exhibited a small but signicant reduction in the rotarod performance at the last trial (Fig. 4C), indicating that these mice may also have a mild motor decit.

Human Molecular Genetics, 2013

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 4. Impaired motor and sensory performance in SIMPLE W116G mutant mice. (A) When suspended by the tail, an abnormal phenotype of hind-limb clasping and occasional limb/body exing was seen in SIMPLEW116G/W116G mice but not in non-transgenic control (CTRL) or SIMPLEW116G/+ mice at 1 year of age. (B) Quantitative analysis of the percentage of mice with the hind-limb clasping phenotype in 1-year-old SIMPLEW116G/W116G, SIMPLEW116G/+ and control (CTRL) mice (n 7 13 mice per genotype). (C) Impaired rotarod performance in SIMPLEW116G/W116G and SIMPLEW116G/+mice compared with the non-transgenic control (CTRL) mice. The latency was plotted versus the trial number. Data represent mean + SEM. (n 6 10 mice per genotype). P , 0.05 versus the control. (D) Self-mutilation of the tail in a SIMPLEW116G/W116G mouse was shown, which was suggestive of paresthesia. (E) Impaired tail-ick response in SIMPLEW116G/W116G mice compared with the non-transgenic control (CTRL) mice at 1 year of age. Data represent mean + SEM. (n 6 10 mice per genotype). P , 0.05 versus the control.

did not show any difference in the SNCV or the SNAP amplitude compared with the non-transgenic controls at 1 year of age (Supplementary Material, Fig. S4C and D). Together, these results indicate that SIMPLE W116G mutant mice, but not SIMPLE WT transgenic mice, display motor and sensory nerve conduction defects that are consistent with the electrophysiological ndings from human CMT1C patients (4 8).

SIMPLE W116G mutation causes peripheral nerve dysmyelination with myelin infolding and reduced axon caliber To investigate the pathological changes underlying the observed motor and sensory neuropathy phenotype in SIMPLEW116G/ W116G mice, we performed histological analyses of sciatic

nerves, ventral roots and dorsal roots from the mutant mice and the control mice. At 3 months of age, SIMPLEW116G/ W116G mice showed no obvious abnormality in the myelin or axonal structure of peripheral nerves (Supplementary Material, Fig. S5AC). However, analysis of semithin cross sections of peripheral nerves from 1-year-old SIMPLEW116G/W116G mice revealed abnormal Schwann cellaxon units with focally infolded myelin sheaths that appeared as single, double or triple internal myelin rings within a myelinated axon (Fig. 7AC). Myelin infolding was observed in both motor and sensory nerves, and the infolding was more prominent in sciatic nerves which are more distally located with respect to the neuronal cell bodies (Fig. 7A) compared with ventral and dorsal roots which are more proximally located (Fig. 7B and C). The myelin infolding phenotype appeared to be specic to the peripheral nerves, as no myelin infolding was observed in myelinated nerves from the central nervous system

Human Molecular Genetics, 2013

Myelin infoldings originate from the paranodal regions and near Schmidt-Lanterman incisures To further characterize the pathological changes in SIMPLEW116G/W116G mice, we performed electron microscopic analyses of sciatic nerves from the mutant mice and control mice at 12 15 months of age. The results revealed that, unlike non-transgenic controls (Fig. 8A), SIMPLEW116G/ W116G nerves showed abundant myelin abnormalities with focally folded structures (Fig. 8B I). Interestingly, we found only myelin infoldings but no myelin outfoldings. Some myelin infoldings were observed in the orientation perpendicular to the nerve axis with myelin sheath that protrudes into a compressed axon (Fig. 8B). More frequently, myelin infoldings were observed in the orientation parallel to the nerve axis, which appeared in cross sections as one or more internal myelin rings inside the myelinated axons of both large and small caliber nerve bers (Fig. 8C F). The internal myelin rings had the same number of myelin lamellae and periodicity as the myelin sheath surrounding the axon (Fig. 8F H), suggesting that the internal myelin rings are invaginated loops of the myelin sheath. Light microscopic and electron microscopic analyses of longitudinal sciatic nerve sections revealed that the infolded myelin loops predominantly originated from the myelin sheath at the paranodal regions (Fig. 9C H) and the internodal regions adjacent to Schmidt-Lanterman incisures (Figs 8I and 9B) in SIMPLEW116G/W116G mice. In contrast, myelin infolding was virtually absent in the longitudinal sciatic nerve sections from the control mice (Figs 8J and 9A).
Figure 5. Motor nerve conduction defects in SIMPLE W116G mutant mice. (A) Representative traces of compound muscle action potential (CMAP) recorded from 1-year-old SIMPLEW116G/W116G mice and the non-transgenic littermate control (CTRL) mice after stimulation at the sciatic notch (proximal) and at the ankle (distal). The stimulus artifact (closed arrowhead) and the onset of CMAP (open arrowhead) were indicated. (B and C) Quantitative analysis of motor nerve conduction velocities (MNCVs) (B) and CMAP amplitudes (C) in SIMPLEW116G/W116G, SIMPLEW116G/+ and control (CTRL) mice at the age of 3 months (n 10 19 mice per genotype) and 1 year (n 13 mice per genotype). Data represent mean + SEM. P , 0.05 versus the control.

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

SIMPLE W116G mutation disrupts the integrity of Schwann cell axon units and nodes of Ranvier Ultrastructural analysis by electron microscopy revealed that SIMPLEW116G/W116G axons with myelin infoldings were often displaced, deformed or constricted (Fig. 8B F), consistent with the reduced axon caliber and axon area (Fig. 7F and G) from the morphometric analysis of semithin sections. Moreover, loss of myelin compaction (Fig. 8M) and widened spacing of the Schmidt-Lanterman incisures (Fig. 8K and L) were occasionally observed. We also observed signs of axonal damage and axonal degeneration (Fig. 8MP). Quantitative analysis revealed a small but signicant increase in the percentage of sciatic nerves undergoing axonal degeneration in SIMPLEW116G/W116G mice compared with the control mice at 1215 months of age (Fig. 8S). In addition, a small but signicant increase in the percentage of sciatic nerves with demyelinated axons was also observed in the SIMPLEW116G/W116G mice compared with the control mice (Fig. 8Q, R and T). Given the prominent presence of myelin infoldings at the paranodal regions (Fig. 9C and D), we analyzed SIMPLEW116G/W116G nodes of Ranvier and paranodal regions in more detail. Analysis of longitudinal sciatic nerve sections revealed that myelin infolding did not affect the two sides of the node symmetrically in most cases (Fig. 9C, D and G), although occasionally myelin infolding can be observed in both sides of the node (Fig. 9E). Paranodal regions with myelin infolding often exhibited signs of dys/demyelination and

of SIMPLEW116G/W116G mice even up to 15 months of age (Supplementary Material, Fig. S2). Histological analyses did not reveal obvious signs of demyelination and remyelination, such as thinly myelinated large axons or onion-bulb formation in peripheral nerves of 1-year-old SIMPLEW116G/W116G mice. Morphometric analysis indicated that, correlated with signicantly increased myelin infoldings (Fig. 7D), the axon contours of SIMPLEW116G/W116G sciatic nerves were signicantly less circular (Fig. 7E), and the axon caliber (or axon diameter) and axon area of the mutant nerves were signicantly smaller than those of the non-transgenic controls (Fig. 7F and G). However, myelin thickness of SIMPLEW116G/W116G nerves was unchanged (Fig. 7H). In addition, the G-ratio (as calculated by dividing the axon diameter by the ber diameter) was signicantly reduced (Fig. 7I) as a direct result of decreased axon caliber (Fig. 7F). There was no loss of axons in the SIMPLEW116G/W116G mice compared with the nontransgenic controls (Fig. 7J).

Human Molecular Genetics, 2013

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 6. Sensory nerve conduction defects in SIMPLE W116G mutant mice. (A) Representative traces of sensory nerve action potential (SNAP) recorded from 1-year-old SIMPLEW116G/W116G mice and the non-transgenic control (CTRL) mice after stimulation at the tail. The stimulus artifact (closed arrowhead) and the onset of SNAP (open arrowhead) were indicated. (B D) Quantitative analysis of sensory nerve conduction velocities (SNCVs) (B), SNAP amplitudes (C) and percentage of mice with no evoked SNAP (D) in SIMPLEW116G/W116G, SIMPLEW116G/+ and control (CTRL) mice at the age of 3 months (n 1019 mice per genotype) and 1 year (n 1113 mice per genotype). Data represent mean + SEM. P , 0.05 versus the control.

axonal damage (Fig. 9C and E G). Non-compacted myelin whorls were found to extend from the compact myelin into the axon (Fig. 9C, E and F), suggesting a loss of myelin compaction at the paranodal region that may impair axonal function and integrity. We observed the accumulation of electron-dense organelles, mainly mitochondria, in the axoplasm of the paranodal region next to the infolded myelin, likely as a result of impaired axonal transport (Fig. 9E and F). Our data suggest that the focally infolded myelin loops, particularly those protruding deep into the axons perpendicularly (Fig. 9C, E and F) may physically block axonal transport, leading to axonal degeneration observed in SIMPLEW116G/ W116G sciatic nerves (Fig. 8M P and S). Electron microscopic analysis showed that, despite myelin infolding, axoglial junction and Schwann cell microvilli appeared largely intact and the myelin sheath is properly lined by Schwann cell basal lamina (Fig. 9E H), indicating that the overall nodal architecture is intact in the SIMPLEW116G/W116G nerves. However, we observed that the nodes of Ranvier with myelin infolding often showed paranodal retraction leading to a substantially larger nodal gap (Fig. 9G and H). The enlarged nodal gaps were also observed in the semithin longitudinal sciatic nerve sections (Fig. 9D) of 1-year-old SIMPLEW116G/W116G mice. Quantitative analysis indicated that the nodal gap length of SIMPLEW116G/W116G nerves is signicantly longer than that of the control nerves (Fig. 9I).

To further characterize SIMPLE W116G mutation-induced changes at nodes of Ranvier, we performed immunouorescence confocal microscopic analyses of sciatic nerves from 1-year-old SIMPLEW116G/W116G and control mice. As expected, the control nerves showed normal nodal organization with three clearly segregated, axonal compartments: the node, marked by the clustering of sodium (Nav) channels; the paranode, labeled by the staining of the adhesion molecule Caspr and the juxtaparanode, dened by the presence of potassium (Kv1.2) channels (Fig. 9J and K). In SIMPLEW116G/W116G mutant nerves, although Nav channels remained tightly clustered at the nodes, the distribution of the paranodal marker Caspr was altered, with elongated Caspr-positive structures that extended into the juxtaparanodal region (Fig. 9J). The distribution of the juxtaparanodal marker Kv1.2 channels was also altered, with elongated Kv1.2-positive structures that extended into the paranodal region and the internodal region (Fig. 9K). The elongation of Caspr and Kv1.2 clusters affected the two sides of the node asymmetrically (Fig. 9J and K). Double immunostaining with anti-Caspr and anti-Kv1.2 antibodies revealed a substantial overlap of the paranodal and juxtaparanodal markers (Fig. 9K), indicating the segregation of the paranodal and juxtaparanodal compartments is disrupted in the mutant axons. Quantitative analysis indicated that 73% of nodes of Ranvier in SIMPLEW116G/ W116G sciatic nerves had the paranode-juxtaparanode overlap phenotype.

Human Molecular Genetics, 2013

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 7. Histological analysis reveals myelin infoldings in SIMPLE W116G mutant mice. Semithin cross sections of sciatic nerves (A), dorsal roots (B), ventral roots (C) from 1-year-old SIMPLEW116G/W116G and non-transgenic control (CTRL) mice. For each panel, 10 randomly selected elds (150400 nerve bers per eld) from 3 to 5 semithin cross sections of sciatic nerve per mouse were examined, and 3 mice were analyzed per genotype. Representative images were shown. Arrowheads indicate myelin infoldings. Scale bar, 10 mm. (D) Quantication of the percentage of myelinated nerves with focally infolded myelin showed increased myelin infolding in SIMPLEW116G/W116G nerves compared with the control (CTRL). (E) Quantication of the percentage of axons with different ranges of roundness showed that axon bers in SIMPLEW116G/W116G nerves are signicantly less circular compared with the control (CTRL). (F) Quantication of the percentage of axons with different ranges of axon diameter showed a signicantly smaller axon diameter in SIMPLEW116G/W116G nerves compared with the control (CTRL) at 12 15 months of age. (G) Quantication of axon area showed a signicant reduction in SIMPLEW116G/W116G sciatic nerves compared with the control (CTRL). (H and I) Quantications of myelin thickness (H) and G-ratio (I) showed a signicantly reduced G-ratio with unaltered myelin thickness in SIMPLEW116G/W116G nerves compared with the control (CTRL) at 12 15 months of age. (J) Quantication of axons per mm2 showed no difference in axon count between SIMPLEW116G/W116G and control (CTRL) nerves. For morphometric analyses in (DG and J), 10 randomly selected elds (150400 nerve bers per eld) from 3 to 5 semithin cross sections of sciatic nerve per mouse were analyzed, and 3 mice were analyzed per genotype. For morphometric analyses in (H and I), 10 randomly selected elds (50100 nerve bers per eld) from 3 to 5 ultrathin cross sections of sciatic nerve per mouse were analyzed, and 3 mice were analyzed per genotype. Data represent mean + SEM. P , 0.05 versus the control.

Myelin infoldings precede axonal defects in the peripheral nerves of SIMPLE W116G mutant mice To determine whether myelin infoldings or axonal defects appeared rst in SIMPLEW116G/W116G mice, we stained the sciatic nerves from 8-month-old and 1-year-old SIMPLEW116G/W116G and control mice with Nile red, a uorescent lipophilic dye that strongly labels lipid-rich structures such as myelin. Fluorescence confocal microscopic analysis of Nile red-labeled nerves showed myelin infoldings at paranodal regions of sciatic nerves in both ages of SIMPLEW116G/W116G mice that are largely absent in the age-matched control mice (Fig. 10A). Quantitative analysis revealed that the percentage of paranodes with myelin infolding is signicantly increased in the 1-year-old mutant mice compared with the 8-month-old mutant mice (Fig. 10B). In addition, Nile red staining analysis revealed a signicant increase in the percentage of nodes with enlarged nodal gaps in SIMPLEW116G/W116G mice compared with the control mice at 1 year but not at 8 months of age (Fig. 10A and C). Next, we performed immunostaining analyses with antiCaspr antibodies to assess age-dependent changes in the paranodal axonal compartments of sciatic nerves from

8-month-old and 1-year-old SIMPLEW116G/W116G and control mice. We found that, consistent with the enlarged nodal gap (Figs 9D, G, I and 10A, C), the gap between Caspr-positive paranodes was signicantly enlarged in 1-year-old, but not in 8-month-old, SIMPLEW116G/W116G mice compared with the age-matched control mice (Fig. 10D and E). Furthermore, the axonal Caspr-positive structures were signicantly elongated in 1-year-old, but not in 8-month-old, mutant mice compared with the control mice (Fig. 10D, F and G). Together, these results provide evidence supporting that myelin infolding is a pathogenic event that precedes axonal defects and nodal disorganization.

DISCUSSION
Despite human genetic evidence linking heterozygous SIMPLE missense mutations to dominantly inherited CMT1C (4 8), it is unclear whether these mutations cause peripheral neuropathy via haplo-insufciency of SIMPLE or dominant effects of the SIMPLE mutants. Our nding of the motor and sensory neuropathy phenotype by transgenic expression of human SIMPLE W116G mutant protein on a

10

Human Molecular Genetics, 2013

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 8. Electron microscopic analysis of myelin abnormalities and axonal degeneration in SIMPLE W116G mutant mice. Ultrastructural analysis of sciatic nerves from 12- to 15-month-old mice showed myelin and axon abnormalities of SIMPLEW116G/W116G mice (B I and K R) that are virtually absent in the control (CTRL) mice (A and J). The most frequent type of myelin abnormality in the SIMPLEW116G/W116G nerves is the focally infolded myelin that protrudes into the axon (B) and the presence of one or more infolded myelin loops within myelinated axons of both large and small calibers (CF). The infolded myelin loops (H) and the myelin sheath (G) of the axon have the identical periodicity and number of lamellae. Schwann cell cytoplasmic vesicles (G) were indicated by an arrow. Longitudinal section showed infolded myelin loop originating from the myelin sheath near the Schmidt-Lanterman incisures (I). Loss of myelin compaction and widened spacing of the Schmidt-Lanterman incisures were occasionally observed in SIMPLEW116G/W116G nerves in longitudinal (K) and cross sections (L) but were absent in the control (J). The Schmidt-Lanterman incisures (I L) were indicated by arrows, and myelin infolding (I) was indicated by arrowhead. SIMPLEW116G/W116G nerve bers with various stages of axonal degeneration (M P), intramyelin edema (Q) and demyelinated axons (R) were also occasionally observed. Scale bar, 2 mm (A-F, I-R) or 200 nm (G and H). (S and T) Quantication of the percentage of sciatic nerves undergoing axonal degeneration (S) and demyelination (T) showed a signicantly increased axonal degeneration and demyelination in SIMPLEW116G/W116G mice compared with the control (CTRL) at 1215 months of age. For each quantication, 10 randomly selected elds (50100 nerve bers per eld) from 3 to 5 ultrathin cross sections of sciatic nerve per mouse were analyzed, and 3 mice were analyzed per genotype. Data represent mean + SEM. P , 0.05 versus the control.

WT SIMPLE background in mice supports that human CMT1C disease is caused by dominant effects of the SIMPLE mutant rather than haplo-insufciency of WT SIMPLE protein. Consistent with this view, a recent study reported that SIMPLE knockout mice developed normally and did not show a neuropathy phenotype (22). The lack of a neuropathy phenotype in our control transgenic mice, which express human SIMPLE WT protein at an equivalent level to that of human SIMPLE W116G protein in the mutant transgenic mice, indicates that the neuropathy phenotype of the SIMPLE W116G mutant mice is the consequence of the SIMPLE mutation rather than an overexpression of SIMPLE protein. Our nding of the SIMPLE W116G

transgene integration into a non-coding genomic region (Fig. 1C and D) supports that the neuropathy phenotype in the SIMPLE mutant mice is not due to an integration site effect. The observed motor and sensory neuropathy phenotype in SIMPLEW116G/W116G transgenic mice recapitulates key clinical features of human CMT1C disease. CMT1C patients carrying SIMPLE W116G mutation have an average age of onset at 41 years old (4). Consistent with this late onset, SIMPLEW116G/W116G mice exhibited motor and sensory nerve conduction defects and impaired motor and sensory performance at 1 year but not 3 months of age. Like CMT1C patients (4 8), SIMPLEW116G/W116G mice displayed

Human Molecular Genetics, 2013

11

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 9. Paranodal myelin infoldings, axonal defects and widened nodes of Ranvier in SIMPLE W116G mutant mice. (A D) Analysis of longitudinal semithin sections of sciatic nerves from 1-year-old SIMPLEW116G/W116G mice showed myelin infolding at the paranodal (C and D) and internodal (B) regions, loss of myelin compaction at the paranodal regions (C and D) and the widening of node of Ranvier (D) that are absent in the control nerves (A). Scale bar, 20 mm. (E H) Ultrastructural analysis of longitudinal sciatic nerve sections from SIMPLEW116G/W116G mice at 12 15 months of age revealed myelin infolding, noncompacted myelin whorls and accumulations of axoplasm (ap) containing electron-dense organelles at the paranodal regions (E and F). Paranodal retraction of myelin and widening of the node of Ranvier were also observed (G and H). The compact myelin sheath is lined by basal lamina, and microvilli (mv) and axoglial junctions appeared intact. Myelin infolding (arrowhead), Schmidt-Lanterman incisures (arrow), and nodal gap length (bracket) were indicated. Scale bar, 10 mm. (I) Quantication of the nodal gap length showed a signicant widening of the nodes of Ranvier in 1-year-old SIMPLEW116G/W116G mice compared the control (CTRL). For quantication, 12 15 sciatic nerve bers from three mice per genotype were analyzed. Data represent mean + SEM. P , 0.05 versus the control. (J and K) Teased sciatic nerves from the indicated 1-year-old SIMPLEW116G/W116G or non-transgenic control (CTRL) mice were stained with anti-pan-Nav (red) and anti-Caspr (green) antibodies (J) or with anti-Kv1.2 (red) and anti-Caspr (green) antibodies (K). For each panel, 10 randomly selected nodes per mouse were analyzed, and 3 mice were analyzed per genotype. Representative images were shown. Arrowhead indicated the location of myelin infolding. Scale bar, 5 mm.

motor impairments accompanied not only by a reduced MNCV (which is characteristic of demyelinating CMT) but also by a reduced CMAP amplitude (which is indicative of axonal degeneration). Furthermore, as reported in CMT1C patients (4 8), SIMPLEW116G/W116G mice also exhibited sensory impairments accompanied by a reduced SNCV. In addition, similar to CMT1C where a subgroup of patients with the SIMPLE W116G mutation have no evoked SNAP and show paresthesia (4), we found that a small percentage of SIMPLEW116G/W116G mice lost the ability to evoke SNAP and showed self-mutilating behavior which could be caused by the loss of sensory nerve conduction. CMT1C is a rare form of hereditary motor and sensory neuropathy and biopsy material is scarce. Therefore, our

knowledge of the pathology of human CMT1C is very limited. The only published pathological result is a semithin cross section showing the presence of onion-bulb formations in the sural nerve biopsy material from a CMT1C patient (6). We did not nd obvious evidence for onion-bulb formations in the peripheral nerves of SIMPLEW116G/W116G mice. A similar lack of onion-bulb formations has been reported in several other mouse models of human CMT (23 25), which could be due to the shorter life span of mice, shorter lengths of nerves in mice and/or some other species difference between mice and humans (26). Despite the lack of onion-bulb formations, we found that the motor and sensory functional impairments in SIMPLEW116G/ W116G mice are associated with focally infolded myelin loops that protrude into the axons at paranodal regions and near

12

Human Molecular Genetics, 2013

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Figure 10. Myelin infolding precedes axonal damage in SIMPLE W116G mutant mice. (A) Nile red-stained teased sciatic nerve bers from 8-month-old (8 months) and 1-year-old (1 year) SIMPLEW116G/W116G mice showed myelin infolding at the paranodal regions and near Schmidt-Lanterman incisures and occasional widening of the node of Ranvier that are absent in the control (CTRL) mice. Myelin infolding (arrowhead), Schmidt-Lanterman incisures (arrow) and nodal gap (bracket) were indicated. Scale bar, 20 mm. (B and C) Quantication of the percentage of paranodal regions with myelin infolding (B) and the percentage of nodes with enlarged nodal gap (C) showed age-dependent increases in myelin infolding and nodal disorganization in SIMPLEW116G/W116G mice compared with the CTRL mice. For each analysis in (A C), 2025 randomly selected nodes of Ranvier per mouse were analyzed, and three mice were analyzed per genotype/age group. Representative images were shown. Data represent mean + SEM. P , 0.05 versus the control, #P , 0.05 versus the 8-month-old mice for the corresponding genotype. (D) Teased sciatic nerves from the indicated 8-month-old (8 months) and 1-year-old (1 year) SIMPLEW116G/W116G and control (CTRL) mice were stained with anti-Caspr (green) antibody. Caspr lengths (red brackets), Caspr gap (yellow brackets) and myelin infolding (arrowhead) were indicated. Scale bar, 5 mm. (E G) Quantication of the Caspr gap distance (E), Caspr length (F) and the percentage of paranodes with elongated Caspr staining (G) showed signicantly enlarged Caspr gap (E) and longer Caspr length (F and G) in 1-year-old mutant nerves, but not in 8-month-old mutant nerves, compared with the CTRL mice. For each analysis in (D G), 1020 randomly selected nodes of Ranvier per mouse were analyzed, and three mice were analyzed per genotype/age group. Representative images were shown. Data represent mean + SEM. P , 0.05 versus the control, #P , 0.05 versus the 8-month-old mice for the corresponding genotype.

Schmidt Lanterman incisures of motor and sensory nerves. Our nding that myelin infolding is often linked to constricted axons with signs of impaired axonal transport suggests that focally infolded myelin may physically block axonal transport. Our analyses revealed that myelin infolding is also associated with altered distribution of the paranodal and juxtaparanodal axonal molecules and a widened nodal gap at the nodes of Ranvier. In addition to myelin infolding and paranodal damage, we have identied a small but signicant percentage of peripheral nerves with signs of demyelination and axonal degeneration. These pathological changes provide a structural basis for the observed motor and sensory nerve conduction defects, including reduced MNCV and SNCV, decreased CMAP amplitude and a loss of evoked SNAP in SIMPLEW116G/W116G mice. The nding of myelin infolding and paranodal damage in 1-year-old but not 3-month-old SIMPLEW116G/W116G mice indicates that these pathological changes occur after the myelination has completed, and the architecture of Schwann cell axon units has been established. Therefore, myelin infolding and paranodal damage are likely caused by defects in the maintenance rather than the formation of myelin sheath and Schwann cell axon units.

Our data show that myelin infolding appears in peripheral nerves of SIMPLEW116G/W116G mice at 8 months of age, prior to the development of paranodal axonal defects and nodal disorganization. The percentage of paranodal regions with myelin infolding is signicantly increased as the mutant mice get older to the 1 year of age and develop the motor and sensory neuropathy with paranodal axonal defects and nodal disorganization. Based on our results, we propose that myelin infolding and paranodal damage may represent pathogenic precursors leading to demyelination and axonal degeneration in CMT1C patients, and these pathological changes should be considered as major factors contributing to the CMT1C motor and sensory neuropathy. Focally folded myelin, mostly myelin outfolding but also myelin infolding, has been found in several types of inherited demyelinating neuropathies, such as CMT1B (27 29), CMT4A (30) and CMT4B (31,32), and chronic inammatory demyelinating neuropathy (33) as well as in various animal models of demyelinating CMT (16,23,34,35) and other peripheral nerve disorders (36,37). In addition, widened nodal gap has been reported in CMT1A (38) and CMT4C (13). These shared pathological features in this diverse group of disorders

Human Molecular Genetics, 2013

13

suggest that abnormal myelin folding and nodal disorganization may represent convergent pathogenic mechanisms leading to demyelinating peripheral neuropathy. Our immunohistochemical analysis reveals that human SIMPLE W116G transgene, like endogenous SIMPLE, is expressed in myelinating Schwann cells but not in axons. Furthermore, the SIMPLE W116G mutant protein is partially mislocalized from the early endosomal membrane to the cytosol in the SIMPLEW116G/W116G mice. These results, together with our published data showing that the W116G mutation promotes SIMPLE protein misfolding (9,39), support the hypothesis that SIMPLE mutation causes peripheral neuropathy via a combination of toxic gain-of-function and dominant-negative pathogenic mechanisms. We recently reported that SIMPLE interacts and colocalizes with endosomal sorting complex required for transport (ESCRT) components STAM1, Hrs and TSG101 on early endosomes and functions with the ESCRT machinery in the control of endosomal cargo sorting to the lysosome for degradation (40). We showed that CMT-linked SIMPLE mutants, including SIMPLE W116G mutant, are loss-of-function mutants that act in a dominant-negative manner to impair endosome-to-lysosome trafcking in the presence of WT SIMPLE (40). Consistent with the critical role of endosometo-lysosome trafcking in signaling attenuation (41), we found that SIMPLE mutant-induced trafcking impairment leads to prolonged signaling of neuregulin-1 (NRG1)-activated ErbB2/ErbB3 receptors to downstream pathways in Schwann cells (40). A NRG1-ErbB downstream pathway that functions in the regulation of myelin membrane growth is the phosphotidylinositol 3 kinase (PI3K)-Akt-mTORsignaling pathway, and hyperactivation of this signaling pathway in Schwann cells has been shown to cause the formation of focally folded myelin loops and progressive peripheral neuropathy (42). These results, together with the data from this study, support a model in which SIMPLE W116G mutant disrupts endosome-to-lysosome trafcking and signaling attenuation of NRG1-activated ErbB2/ErbB3 receptors, causing prolonged activation of downstream signaling pathways such as the PI3K-Akt-mTOR pathway in Schwann cells, thereby leading to myelin infolding and paranodal damage, ultimately resulting in demyelinating peripheral neuropathy. In conclusion, our ndings obtained from this work provide new insights into the pathogenic mechanisms of CMT1C-linked SIMPLE mutations. The transgenic mouse model of CMT1C generated in this study should facilitate the investigation of CMT pathogenesis and therapeutics.

in pseudo-pregnant female FVB/N mice. Transgenic founder mice were identied by the PCR analyses of genomic DNAs isolated mouse tails using two independent sets of transgenespecic primer pairs and then bred with FVB/N mice to established SIMPLE WT and SIMPLE W116G transgenic lines on a pure FVB/N background. Zygosity of SIMPLE WT and SIMPLE W116G transgenic mice was determined by a wellestablished protocol using SYBR Green real-time quantitative PCR (Q-PCR) of tail genomic DNAs and comparative DDC CT (22 ) analysis as described (44 46). Expression of T HA-tagged SIMPLE WT and SIMPLE W116G proteins in transgenic mice was determined by immunoblot and immunostaining analyses using anti-SIMPLE antibody and antiHA antibody. Determination of transgene integration site For identication of the transgene integration site in the SIMPLE W116G transgenic line used in this study, genomic DNAs were isolated from liver tissues of SIMPLE W116G transgenic mice and then subjected to genome walking PCR analyses using a series of transgene-specic primers and the degenerate random tagging primers provided by APAgeneTM Gold-RT Genome Walking Kits (Bio S&T, Inc., Montreal, Canada) in accordance to the manufacturers instructions. The PCR products were either cloned or directly sequenced to identify the transgene integration site. The identied integration site was conrmed by PCR analyses shown in Figure 1D using the following primers: A (5 -CATTTATC AGGGTTATTGTCTC), B (5 -GTCCCTATTGGCGTTACT ATGG), C (5 -TTGCATCGCATTGTCTGAGTAG) and D (5 -GGTAATAAAGAGGAGGGTGAGGAG). Antibodies The generation and characterization of rabbit polyclonal anti-SIMPLE antibody was described in our previous study (9). Other antibodies used in this study include the following: anti-HA (12CA5), anti-actin (Millipore), anti-Rab5 (BD Transduction), anti-S100 (Sigma Aldrich), anti-Dlg1 (Enzo), anti-MBP (Millipore), anti-MAG (Millipore), anti-NF-H (Millipore), anti-Caspr (Abcam), anti-pan-Nav (Sigma) and anti-Kv1.2 (NeuroMab). All secondary antibodies were purchased from Jackson ImmunoResearch Laboratories, Inc. Teased nerve bers, Nile red staining and immunouorescence confocal microscopy Mice were perfused with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Sciatic nerves were dissected and postxed in 4% paraformaldehyde overnight at 48C and washed in 0.1 M phosphate buffer. Individual bers were separated in glycerol on glass slides and were immersed in cold acetone for 10 min. Teased bers were then rehydrated in PBS and incubated at room temperature for 30 min in a blocking solution containing 10% horse serum, 0.5% Triton X-100 and PBS. For Nile red staining, teased bers were mounted in Nile red solution (0.5 mg/ml in acetone) diluted 1000 in 75% glycerol as described (13). For immunostaining, primary antibodies were diluted in the blocking solution and

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

MATERIALS AND METHODS

Generation and genotyping of SIMPLE WT and SIMPLE W116G transgenic mice Conventional molecular biological techniques were used to generate the transgenic constructs encoding HA-tagged human SIMPLE WT and SIMPLE W116G mutant under the control of the human CMV promoter in the pCHA vector (43). After digestion with Mlu I and Nae I, the linearized SIMPLE transgene DNAs were microinjected into pronuclei of fertilized embryos from the FVB mice and then implanted

14

Human Molecular Genetics, 2013

were incubated with the teased bers overnight. After washing, teased bers were incubated with the appropriate secondary antibodies conjugated to FITC, Texas Red or Cy5, and were mounted with ProLong Gold (Invitrogen). Confocal images were acquired in room temperature with the Nikon Eclipse Ti confocal microscope equipped with 60 /1.4 oil immersion objectives and the Nikon EZ-C1 software. Western blot analysis Primary Schwann cells were isolated from postnatal day 2 (P2) to P4 non-transgenic, SIMPLE WT and SIMPLE W116G mutant mice and cultured using established protocols as described (9,47,48). For western blot analysis, sciatic nerves or primary Schwann cells were homogenized in 1% SDS and then subjected to SDS PAGE. The proteins were transferred onto nitrocellulose membranes and probed with the indicated antibodies. Antibody binding was detected by using the enhanced chemiluminescence (ECL) system (Amersham Biosciences). The expression levels of SIMPLE proteins were quantied by measuring the intensity of SIMPLE and actin bands from the immunoblot images using the ImageJ software as described previously (9). Subcellular fractionation analyses Subcellular fractionation of sciatic nerve homogenates into cytosol and membrane fractions were performed as described (9). For density gradient fractionation analysis, the membrane fraction from sciatic nerves was resuspended in fractionation buffer and was then placed on a 10 30% linear Optiprep (Sigma) gradient and centrifuged for 20 h at 125 000 g in a SW41 rotor (Beckman Coulter) as described previously (9). After centrifugation, the gradient was harvested into 250 ml fractions by an Auto Densi-Flow gradient harvester (Labconco). Behavioral tests For the rotarod test, mice were placed on a standard rotarod apparatus (Columbus instruments) and acclimated to a nonaccelerating rotarod at an initial speed of 1.25 rpm for 30 min. At the beginning of each trial, rotarod speed was accelerated at the rate of 1 rpm/min, and the latency for a mouse to fall off the accelerating rotarod was recorded. Each animal was tested six times with a 5 min of rest between each trial. For the tail-ick test, mouse tails were immersed in a water bath containing 528C water as the nociceptive stimulus, and the latency between tail immersion and tail-ick response was measured blindly from recorded videos. The test was repeated three times with a 15 min of rest between each trial, and the mean latency of each mouse over three trials was used for statistical analysis. Electrophysiology SIMPLE WT and W116G mutant mice and their nontransgenic littermates were analyzed at 3 months and 1 year of age. Mice were anesthetized with chloral hydrate (400 mg/kg of body weight, i.p.) and placed under a heat

lamp to avoid hypothermia. Nerve conduction studies were performed using standard equipment (Nicolet Viking Quest). For sciatic nerve motor conduction assays, stimuli were given at the sciatic notch and at the ankle, and recording electrodes were inserted into interosseous muscles of the left foot, whereas a ground electrode was inserted subcutaneously at the tail. For tail nerve sensory conduction assays, the recording electrodes were placed at the base of the tail, keeping the anode and the cathode about 5 mm apart, and with a ground electrode placed subcutaneously 2 cm distal. Stimuli were given 4 cm distal and sensory nerve conduction was averaged over 20 stimuli. Histological analysis and electron microscopy Mice were anesthetized followed by perfusion with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Dorsal roots, ventral roots and sciatic nerves were harvested and postxed by immersion in 4% paraformaldehyde and 2% glutaraldehyde overnight at 48C. Nerves were then washed in 0.1 M phosphate buffer and embedded in parafn. Cross and longitudinal semithin (0.5 mm) sections were stained with toluidine blue and were analyzed by light microscopy (Olympus). For ultrastructural studies, nerves were prepared for electron microscopy by postxation with OsO4 followed by en bloc staining with uranyl acetate. Cross and longitudinal ultrathin (100 nm) sections were cut and stained with uranyl acetate and lead nitrate. Grids containing the ultrathin nerve sections were examined by a Hitachi H-7500 transmission electron microscope. Morphometric analyses Quantitative analyses of focally folded myelin, number of axons, axon roundness, axon diameter and axon area were performed from 10 randomly selected elds of semithin sciatic nerve cross sections (150 400 nerve bers per eld, 40 magnication) per mouse with three mice analyzed per genotype. The number of nerve bers with focally infolded myelin (internal myelin loops) and the total number of bers on each eld were counted, and the percentage of nerve bers with myelin infoldings was calculated. Axon roundness, axon diameter and axon area were measured by using the ImagePro software (Media Cybermetics). Axon roundness, an index which measures the circularity of axons, was calculated according to the formula [roundness perimeter2/(4 p area)]. A roundness index of 1 represents a perfect circle, and the roundness index increases as the shape of an axon in the cross section becomes less circular. Myelin thickness and G-ratio were determined as previously described (49) from electron microscopic images (3000 magnication) for 50 randomly selected sciatic nerve bers from three mice per genotype by using the ImagePro software (Media Cybermetics). Quantitative analyses of axonal degeneration and demyelination were performed from 10 randomly selected elds of sciatic nerve electron microscopic images (50 100 nerve bers per eld, 3000 magnication) per mouse with three mice analyzed per genotype. The number of nerve bers with degenerating axons or demyelinated axons and the total number of bers on each eld were
Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

Human Molecular Genetics, 2013

15

counted, and the percentage of nerve bers with degenerating axons or demyelinated axons was calculated. Quantitative analysis of the nodal gap length was performed from longitudinal sciatic nerve sections (40 magnication) for 12 15 nodes of Ranvier from three mice per genotype, and the nodal gap length was measured by using the ImageJ software. Quantitative analysis of the percentage of paranodal regions with myelin infolding and the percentage of nodes with enlarged nodal gap were performed as described previously (16) from Nile red uorescence confocal microscopic images (60 magnication) for 20 25 randomly selected nodes of Ranvier per mouse with three mice analyzed per genotype/age group. Quantitative analysis of the Caspr gap distance, Caspr length and the percentage of paranodes with elongated Caspr cluster were performed as described previously (16) from immunouorescence confocal microscopic images (60 magnication) for 10 20 randomly selected nodes of Ranvier per mouse with three mice analyzed per genotype/ age group. All quantitative analyses were performed in a blinded manner. Statistical analysis Data were subjected to statistical analyses by Students t-tests or one-way analysis of variance using the SigmaPlot software (Systat Software, Inc.). Results are expressed as mean + SEM. A P-value of , 0.05 was considered statistically signicant.

SUPPLEMENTARY MATERIAL
Supplementary Material is available at HMG online.

ACKNOWLEDGEMENTS
The authors thank Emory Transgenic Mice Core Facility for the microinjection and implantation service in generation of transgenic mice, Emory Electron Microscopy Core Facility for processing semithin and thin sections of peripheral nerve samples and Emory Neuroscience Pathology Core Facility (supported in part by National Institutes of Health Grant NS055077) for processing mouse brain samples. This work was supported by the National Institutes of Health (NS063501 to S.M.L., AG034126 to L.S.C. and ES015813 and GM103613 to L.L.); and the Emory University Research Committee (SK38167 to L.S.C.). Conict of Interest statement. None declared.

REFERENCES
1. Pareyson, D. and Marchesi, C. (2009) Diagnosis, natural history, and management of Charcot-Marie-Tooth disease. Lancet Neurol., 8, 654667. 2. Patzko, A. and Shy, M.E. (2011) Update on Charcot-Marie-Tooth disease. Curr. Neurol. Nuerosci. Rep., 11, 78 88. 3. Nelis, E., Van Broeckhoven, C., De Jonghe, P., Lofgren, A., Vandenberghe, A., Latour, P., Le Guern, E., Brice, A., Mostacciuolo, M.L., Schiavon, F. et al. (1996) Estimation of the mutation frequencies in Charcot-Marie-Tooth disease type 1 and hereditary neuropathy with liability to pressure palsies: a European collaborative study. Eur. J. Hum. Genet., 4, 25 33.

4. Bennett, C.L., Shirk, A.J., Huynh, H.M., Street, V.A., Nelis, E., Van Maldergem, L., De Jonghe, P., Jordanova, A., Guergueltcheva, V., Tournev, I. et al. (2004) SIMPLE mutation in demyelinating neuropathy and distribution in sciatic nerve. Ann. Neurol., 55, 713 720. 5. Sai, G.M., Szigeti, K., Wiszniewski, W., Shy, M.E., Krajewski, K., Hausmanowa-Petrusewicz, I., Kochanski, A., Reeser, S., Mancias, P., Butler, I. et al. (2005) SIMPLE mutations in Charcot-Marie-Tooth disease and the potential role of its protein product in protein degradation. Hum. Mutat., 25, 372383. 6. Street, V.A., Bennett, C.L., Goldy, J.D., Shirk, A.J., Kleopa, K.A., Tempel, B.L., Lipe, H.P., Scherer, S.S., Bird, T.D. and Chance, P.F. (2003) Mutation of a putative protein degradation gene LITAF/SIMPLE in Charcot-Marie-Tooth disease 1C. Neurology, 60, 22 26. 7. Latour, P., Gonnaud, P.M., Ollagnon, E., Chan, V., Perelman, S., Stojkovic, T., Stoll, C., Vial, C., Ziegler, F., Vandenberghe, A. et al. (2006) SIMPLE mutation analysis in dominant demyelinating Charcot-Marie-Tooth disease: three novel mutations. J. Peripher. Nerv. Syst., 11, 148155. 8. Gerding, W.M., Koetting, J., Epplen, J.T. and Neusch, C. (2009) Hereditary motor and sensory neuropathy caused by a novel mutation in LITAF. Neuromuscul. Disord., 19, 701 703. 9. Lee, S.M., Olzmann, J.A., Chin, L.S. and Li, L. (2011) Mutations associated with Charcot-Marie-Tooth disease cause SIMPLE protein mislocalization and degradation by the proteasome and aggresome-autophagy pathways. J. Cell Sci., 124, 33193331. 10. Moriwaki, Y., Begum, N.A., Kobayashi, M., Matsumoto, M., Toyoshima, K. and Seya, T. (2001) Mycobacterium bovis Bacillus Calmette-Guerin and its cell wall complex induce a novel lysosomal membrane protein, SIMPLE, that bridges the missing link between lipopolysaccharide and p53-inducible gene, LITAF(PIG7), and estrogen-inducible gene, EET-1. J. Biol. Chem., 276, 23065 23076. 11. Dequen, F., Filali, M., Lariviere, R.C., Perrot, R., Hisanaga, S. and Julien, J.P. (2010) Reversal of neuropathy phenotypes in conditional mouse model of Charcot-Marie-Tooth disease type 2E. Hum. Mol. Genet., 19, 26162629. 12. Filali, M., Dequen, F., Lalonde, R. and Julien, J.P. (2011) Sensorimotor and cognitive function of a NEFL(P22S) mutant model of Charcot-Marie-Tooth disease type 2E. Behav. Brain Res., 219, 175 180. 13. Arnaud, E., Zenker, J., de Preux Charles, A.S., Stendel, C., Roos, A., Medard, J.J., Tricaud, N., Kleine, H., Luscher, B., Weis, J. et al. (2009) SH3TC2/KIAA1985 protein is required for proper myelination and the integrity of the node of Ranvier in the peripheral nervous system. Proc. Natl Acad. Sci. U.S.A., 106, 1752817533. 14. Rosso, G., Cal, K., Canclini, L., Damian, J.P., Ruiz, P., Rodriguez, H., Sotelo, J.R., Vazquez, C. and Kun, A. (2010) Early phenotypical diagnoses in Trembler-J mice model. J. Neurosci. Methods, 190, 1419. 15. King, R.H., Chandler, D., Lopaticki, S., Huang, D., Blake, J., Muddle, J.R., Kilpatrick, T., Nourallah, M., Miyata, T., Okuda, T. et al. (2011) Ndrg1 in development and maintenance of the myelin sheath. Neurobiol. Dis., 42, 368 380. 16. Bolis, A., Coviello, S., Bussini, S., Dina, G., Pardini, C., Previtali, S.C., Malaguti, M., Morana, P., Del Carro, U., Feltri, M.L. et al. (2005) Loss of Mtmr2 phosphatase in Schwann cells but not in motor neurons causes Charcot-Marie-Tooth type 4B1 neuropathy with myelin outfoldings. J. Neurosci., 25, 85678577. 17. Wrabetz, L., DAntonio, M., Pennuto, M., Dati, G., Tinelli, E., Fratta, P., Previtali, S., Imperiale, D., Zielasek, J., Toyka, K. et al. (2006) Different intracellular pathomechanisms produce diverse Myelin Protein Zero neuropathies in transgenic mice. J. Neurosci., 26, 2358 2368. 18. Menotti, F., Bazzucchi, I., Felici, F., Damiani, A., Gori, M.C. and Macaluso, A. (2012) Neuromuscular function after muscle fatigue in Charcot-Marie-Tooth type 1A patients. Muscle Nerve, 46, 434 439. 19. Nave, K.A., Sereda, M.W. and Ehrenreich, H. (2007) Mechanisms of disease: inherited demyelinating neuropathiesfrom basic to clinical research. Nat. Clin. Pract. Neurol., 3, 453 464. 20. Scherer, S.S. and Wrabetz, L. (2008) Molecular mechanisms of inherited demyelinating neuropathies. Glia, 56, 1578 1589. 21. Berger, P., Niemann, A. and Suter, U. (2006) Schwann cells and the pathogenesis of inherited motor and sensory neuropathies (Charcot-Marie-Tooth disease). Glia, 54, 243 257. 22. Somandin, C., Gerber, D., Pereira, J.A., Horn, M. and Suter, U. (2012) LITAF (SIMPLE) regulates Wallerian degeneration after injury but is not

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013

16

Human Molecular Genetics, 2013

23. 24.

25.

26. 27.

28.

29.

30.

31.

32.

33.

34.

35.

essential for peripheral nerve development and maintenance: implications for Charcot-Marie-Tooth disease. Glia., 60, 15181528. Tersar, K., Boentert, M., Berger, P., Bonneick, S., Wessig, C., Toyka, K.V., Young, P. and Suter, U. (2007) Mtmr13/Sbf2-decient mice: an animal model for CMT4B2. Hum. Mol. Genet., 16, 29913001. Verhamme, C., King, R.H., ten Asbroek, A.L., Muddle, J.R., Nourallah, M., Wolterman, R., Baas, F. and van Schaik, I.N. (2011) Myelin and axon pathology in a long-term study of PMP22-overexpressing mice. J. Neuropathol. Exp. Neurol., 70, 386 398. Bonneick, S., Boentert, M., Berger, P., Atanasoski, S., Mantei, N., Wessig, C., Toyka, K.V., Young, P. and Suter, U. (2005) An animal model for Charcot-Marie-Tooth disease type 4B1. Hum. Mol. Genet., 14, 3685 3695. Suter, U. and Scherer, S.S. (2003) Disease mechanisms in inherited neuropathies. Nat. Rev. Neurosci., 4, 714 726. Fabrizi, G.M., Taioli, F., Cavallaro, T., Rigatelli, F., Simonati, A., Mariani, G., Perrone, P. and Rizzuto, N. (2000) Focally folded myelin in Charcot-Marie-Tooth neuropathy type 1B with Ser49Leu in the myelin protein zero. Acta Neuropath., 100, 299304. Kochanski, A., Drac, H., Jedrzejowska, H. and Hausmanowa-Petrusewicz, I. (2003) Focally folded myelin in Charcot-Marie-Tooth type 1B disease is associated with Asn131Lys mutation in myelin protein zero gene: short report. Eur. J. Neurol., 10, 547 549. Iida, M., Koike, H., Ando, T., Sugiura, M., Yamamoto, M., Tanaka, F. and Sobue, G. (2012) A novel MPZ mutation in Charcot-Marie-Tooth disease type 1B with focally folded myelin and multiple entrapment neuropathies. Neuromuscul. Disord., 22, 166 169. Chung, K.W., Hyun, Y.S., Lee, H.J., Jung, H.K., Koo, H., Yoo, J.H., Kim, S.B., Park, C.I., Kim, H.N. and Choi, B.O. (2011) Two recessive intermediate Charcot-Marie-Tooth patients with GDAP1 mutations. J. Peripher. Nerv. Syst., 16, 143 146. Quattrone, A., Gambardella, A., Bono, F., Aguglia, U., Bolino, A., Bruni, A.C., Montesi, M.P., Oliveri, R.L., Sabatelli, M., Tamburrini, O. et al. (1996) Autosomal recessive hereditary motor and sensory neuropathy with focally folded myelin sheaths: clinical, electrophysiologic, and genetic aspects of a large family. Neurology, 46, 1318 1324. Bird, T.D., Kraft, G.H., Lipe, H.P., Kenney, K.L. and Sumi, S.M. (1997) Clinical and pathological phenotype of the original family with Charcot-Marie-Tooth type 1B: a 20-year study. Ann. Neurol., 41, 463 469. Sander, S., Ouvrier, R.A., McLeod, J.G., Nicholson, G.A. and Pollard, J.D. (2000) Clinical syndromes associated with tomacula or myelin swellings in sural nerve biopsies. J. Neurol. Neurosurg. Psychiat., 68, 483 488. Bolino, A., Bolis, A., Previtali, S.C., Dina, G., Bussini, S., Dati, G., Amadio, S., Del Carro, U., Mruk, D.D., Feltri, M.L. et al. (2004) Disruption of Mtmr2 produces CMT4B1-like neuropathy with myelin outfolding and impaired spermatogenesis. J. Cell Biol., 167, 711721. Robinson, F.L., Niesman, I.R., Beiswenger, K.K. and Dixon, J.E. (2008) Loss of the inactive myotubularin-related phosphatase Mtmr13 leads to a

36.

37.

38.

39.

40.

41. 42.

43.

44.

45.

46.

47.

48.

49.

Charcot-Marie-Tooth 4B2-like peripheral neuropathy in mice. Proc. Natl Acad. Sci. U.S.A., 105, 4916 4921. Cai, Z., Finnie, J.W., Blumbergs, P.C., Manavis, J., Ghabriel, M.N. and Thompson, P.D. (2006) Early paranodal myelin swellings (tomacula) in an avian riboavin deciency model of demyelinating neuropathy. Exp. Neurol., 198, 65 71. Cai, Z., Sutton-Smith, P., Swift, J., Cash, K., Finnie, J., Turnley, A., Thompson, P.D. and Blumbergs, P.C. (2002) Tomacula in MAG-decient mice. J. Peripher. Nerv. Syst., 7, 181 189. Yoshikawa, H., Nishimura, T., Kaido, M., Toyooka, K., Fujimura, H., Sakoda, S. and Yanagihara, T. (1996) Cation binding at the node of Ranvier in biopsied peripheral nerves of patients with Charcot-Marie-Tooth disease type 1A and hereditary neuropathy with liability to pressure palsies. Acta. Neuropathol., 91, 587594. Lee, S.M., Chin, L.S. and Li, L. (2012) Protein misfolding and clearance in demyelinating peripheral neuropathies: therapeutic implications. Commun. Integr. Biol., 5, 107 110. Lee, S.M., Chin, L.-S. and Li, L. (2012) Charcot-Marie-Tooth disease-linked protein SIMPLE functions with the ESCRT machinery in endosomal trafcking. J. Cell Biol., 199, 799 816. Wegner, C.S., Rodahl, L.M.W. and Stenmark, H. (2011) ESCRT proteins and cell signalling. Trafc, 12, 1291 1297. Goebbels, S., Oltrogge, J.H., Wolfer, S., Wieser, G.L., Nientiedt, T., Pieper, A., Ruhwedel, T., Groszer, M., Sereda, M.W. and Nave, K.-A. (2012) Genetic disruption of Pten in a novel mouse model of tomaculous neuropathy. EMBO Mol. Med., 4, 486 499. Chin, L.S., Raynor, M.C., Wei, X., Chen, H.Q. and Li, L. (2001) Hrs interacts with sorting nexin 1 and regulates degradation of epidermal growth factor receptor. J. Biol. Chem., 276, 70697078. Livak, K.J. and Schmittgen, T.D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 25, 402 408. Haurogne, K., Bach, J.M. and Lieubeau, B. (2007) Easy and rapid method of zygosity determination in transgenic mice by SYBR Green real-time quantitative PCR with a simple data analysis. Transgenic Res., 16, 127131. Sakurai, T., Kamiyoshi, A., Watanabe, S., Sato, M. and Shindo, T. (2008) Rapid zygosity determination in mice by SYBR Green real-time genomic PCR of a crude DNA solution. Transgenic Res., 17, 149155. Pannunzio, M.E., Jou, I.M., Long, A., Wind, T.C., Beck, G. and Balian, G. (2005) A new method of selecting Schwann cells from adult mouse sciatic nerve. J. Neurosci. Methods, 149, 74 81. Haastert, K., Mauritz, C., Chaturvedi, S. and Grothe, C. (2007) Human and rat adult Schwann cell cultures: fast and efcient enrichment and highly effective non-viral transfection protocol. Nat. Protoc., 2, 99104. Kobsar, I., Berghoff, M., Samsam, M., Wessig, C., Maurer, M., Toyka, K.V. and Martini, R. (2003) Preserved myelin integrity and reduced axonopathy in connexin32-decient mice lacking the recombination activating gene-1. Brain, 126, 804 813.

Downloaded from http://hmg.oxfordjournals.org/ at Emory University on March 11, 2013