Gracile axonal dystrophy (GAD), a new neurological mutant in the mouse

A new neurological mutant has been found in the F2 offspring of CBA/Nga and RFM/Nga mice. Affected mice exhibited ataxia beginning at about 80 days of age, followed by tremor, difficulty in moving, and muscular atrophy of the hind limbs. The neurological signs became progressively severe, and death occurred by 5 to 6 months of age. Since the animals could be distinguished from normal mice by the abnormal positions of the hind limbs when the mouse was hung by the tail after 1 month of age, they could be bred until onset of the signs. Pathological examination revealed neuroaxonal dystrophy and degeneration in the gracile nucleus of the medulla oblongata and the gracile fascicules of the spinal cord, which could be the main cause of the clinical signs. The mutation is inherited as an autosomal recessive trait. It was, therefore, named gracile axonal dystrophy (GAD) with the gene symbol gad. The mice could be a new pathological model for the study of neuroaxonal dystrophy.     

Reduced fertility in gracile axonal dystrophy (gad) mice

Mean litter size in gad/gad females was significantly lower than in normal females (+/+ and gad/+) in intra- and inter-strain crosses. The reduction in litter size was not dependent on the genotypes of the males, but could be attributed to the gad/gad females themselves. The numbers of corpora lutea and implants in gad/gad females were slightly reduced as compared with those in the controls, but the number of live fetuses was significantly lower than that in normal females 14 days after copulation (P less than 0.02). Hence, reduced litter size in gad/gad females was accounted for mostly by embryonic and fetal death after implantation, which was inferred to be due to impaired uterine function.     

Proteomic and histochemical analysis of proteins involved in the dying-back-type of axonal degeneration in the gracile axonal dystrophy (gad) mouse

Local axonal degeneration is a common pathological feature of peripheral neuropathies and neurodegenerative disorders of the central nervous system, including Alzheimer's disease, Parkinson's disease, and stroke; however, the underlying molecular mechanism is not known. Here, we analyzed the gracile axonal dystrophy (gad) mouse, which displays the dying-back-type of axonal degeneration in sensory neurons, to find the molecules involved in the mechanism of axonal degeneration. The gad mouse is analogous to a null mutant of ubiquitin carboxyl-terminal hydrolase L1 (UCH-L1). UCH-L1 is a deubiquitinating enzyme expressed at high levels in neurons, as well as testis and ovary. In addition, we recently discovered a new function of UCH-L1—namely to bind to and stabilize mono-ubiquitin in neurons, and found that the level of mono-ubiquitin was decreased in neurons, especially in axons of the sciatic nerve, in gad mice. The low level of ubiquitin suggests that the target proteins of the ubiquitin proteasome system are not sufficiently ubiquitinated and thus degraded in the gad mouse; therefore, these proteins may be the key molecules involved in axonal degeneration. To identify molecules involved in axonal degeneration in gad mice, we compared protein expression in sciatic nerves between gad and wild-type mice at 2 and 12 weeks old, using two-dimensional difference gel electrophoresis. As a result, we found age-dependent accumulation of several proteins, including glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and 14-3-3, in gad mice compared with wild-type mice. Histochemical analyses demonstrated that GAPDH and 14-3-3 were localized throughout axons in both gad and wild-type mice, but GAPDH accumulated in the axons of gad mice. Recently, it has been suggested that a wide range of neurodegenerative diseases are characterized by the accumulation of intracellular and extracellular protein aggregates, and it has been reported that oxidative stress causes the aggregation of GAPDH. Furthermore, histochemical analysis demonstrated that sulfonated GAPDH, a sensor of oxidative stress that elicits cellular dysfunction, was expressed in the axons of gad mice, and 4-hydroxy-2-nonenal, a major marker of oxidative stress, was also only detected in gad mice. Our findings suggest that GAPDH may participate in a process of the dying-back-type of axonal degeneration in gad mice and may provide valuable insight into the mechanisms of axonal degeneration.     

Effects of dietary vitamin E supplement on gracile axonal dystrophy (gad) mice

We have studied the effects of dietary vitamin E supplement on the clinical signs and pathological changes in GAD (gracile axonal dystrophy) mice. The control diet contained 2 mg of dl-alpha-tocopheryl acetate (2 I.U.) and vitamin E-supplemented diet contained 58.5 mg of dl-alpha-tocopheryl nicotinate (50 I.U.), per 100 mg of feed. The diet was given to normal (gad/+) and GAD (gad/gad) mice from 21 to 130 days of age. During the feeding, there was no improvement in clinical signs in the GAD mice fed the vitamin E-supplemented diet. The gracile nucleus of the medulla oblongata and the gracile fascicules of the spinal cord were investigated for pathology at 130 days of age, and alpha-tocopherol was also assayed in the serum, liver, brain and spinal cord at that time. There were no pathological differences in the gracile nucleus and fascicules between the GAD mice fed the control and vitamin E-supplemented diet. The alpha-tocopherol levels in the serum and target organs in the control GAD mice were not significantly different from those in control normal mice, showing that GAD mice could absorb and transport alpha-tocopherol. In the supplemented GAD mice, no significant increases in alpha-tocopherol levels were observed in the liver, brain or spinal cord. Particularly, the percentage increase of alpha-tocopherol level in the liver of GAD mice was very low in comparison with that in normal mice, even though the liver can store vitamin E. Thus it may be that the capacity to store vitamin E is lowered in GAD mice. Further studies are needed to investigate in detail the vitamin E metabolism in the mutant mice.     

Proteomic analysis of brain proteins in the gracile axonal dystrophy (gad) mouse, a syndrome that emanates from dysfunctional ubiquitin carboxyl-terminal hydrolase L-1, reveals oxidation of key proteins

Ubiquitin carboxyl-terminal hydrolase L-1 (UCH L-1) is a crucial enzyme for proteasomal protein degradation that generates free monomeric ubiquitin. Our previous proteomic study identified UCH L-1 as one specific target of protein oxidation in Alzheimer's disease (AD) brain, establishing a link between the effect of oxidative stress on protein and the proteasomal dysfunction in AD. However, it is unclear how protein oxidation affects function, owing to the different responses of proteins to oxidation. Analysis of systems in which the oxidized protein displays lowered or null activity might be an excellent model for investigating the effect of the protein of interest in cellular metabolism and evaluating how the cell responds to the stress caused by oxidation of a specific protein. The gracile axonal dystrophy (gad) mouse is an autosomal recessive spontaneous mutant with a deletion on chromosome 5 within the gene encoding UCH L-1. The mouse displays axonal degeneration of the gracile tract. The aim of this proteomic study on gad mouse brain, with dysfunctional UCH L-1, was to determine differences in brain protein oxidation levels between control and gad samples. The results showed increased protein oxidation in thioredoxin peroxidase (peroxiredoxin), phosphoglycerate mutase, Rab GDP dissociation inhibitor alpha/ATP synthase and neurofilament-L in the gad mouse brain. These findings are discussed with reference to the effect of specific protein oxidation on potential mechanisms of neurodegeneration that pertain to the gad mouse.     

Altered retinoid distribution in the repeated epilation (Er) mutant mouse.

The repeated epilation (Er) mutation in mice causes successive loss and regrowth of hair in heterozygotes (Er/+), and blocks orofacial development and epidermal differentiation in lethal homozygotes (Er/Er). Because the mutation affects a systemic factor, because the Er phenotypes resemble exposure to excess retinoids and because retinoids are critical regulators of epithelial differentiation, we examined whether systemic retinoid levels are altered by the Er defect. Liver retinoic acid and retinol were elevated 1.5- and 3.5-fold, respectively, in adult heterozygotes (Er/+) compared to normal (+/+) animals. Retinyl palmitate was increased 2-fold in heterozygous skin and 3-fold in kidney, but the retinol level in plasma was only half that of normal animals. Newborn heterozygous liver also had nearly 2-fold increased retinoids compared to normal. In contrast, Er/Er newborns had reduced retinoid levels in liver, two-thirds the retinol and 15% the retinyl palmitate compared to normal, but greater than 4-fold elevated levels of retinyl palmitate in the extrahepatic body. Tissue contents of retinol binding protein (RBP), which transports vitamin A from the liver to the remainder of the body, were determined by immunoblotting with anti-mouse RBP. Newborn normal and mutant animals had similar liver microsomal RBP contents. RBP contents in plasma and in liver microsomes were also similar in normal and Er/+ adults despite different retinol contents in the Er/+ tissues. Hair follicles of the Er/+, but not the normal adult, were stained with this antiserum to RPB in the outer root sheath layer. These results strongly suggest that altered retinoid distribution is associated with, and may be responsible for, the altered epithelial differentiation in the Er mutant.     

Neuroinflammation by cytotoxic T-lymphocytes impairs retrograde axonal transport in an oligodendrocyte mutant mouse

Mice overexpressing proteolipid protein (PLP) develop a leukodystrophy-like disease involving cytotoxic, CD8+ T-lymphocytes. Here we show that these cytotoxic T-lymphocytes perturb retrograde axonal transport. Using fluorogold stereotactically injected into the colliculus superior, we found that PLP overexpression in oligodendrocytes led to significantly reduced retrograde axonal transport in retina ganglion cell axons. We also observed an accumulation of mitochondria in the juxtaparanodal axonal swellings, indicative for a disturbed axonal transport. PLP overexpression in the absence of T-lymphocytes rescued retrograde axonal transport defects and abolished axonal swellings. Bone marrow transfer from wildtype mice, but not from perforin- or granzyme B-deficient mutants, into lymphocyte-deficient PLP mutant mice led again to impaired axonal transport and the formation of axonal swellings, which are predominantly located at the juxtaparanodal region. This demonstrates that the adaptive immune system, including cytotoxic T-lymphocytes which release perforin and granzyme B, are necessary to perturb axonal integrity in the PLP-transgenic disease model. Based on our observations, so far not attended molecular and cellular players belonging to the immune system should be considered to understand pathogenesis in inherited myelin disorders with progressive axonal damage.     

Behavioral study in the gray mouse lemur (Microcebus murinus) using compounds considered sweet by humans

This study presents the results from two-bottle preference (TBP) tests performed on the gray mouse lemur (Microcebus murinus), a small Malagasy primate. We found that of 18 compounds considered sweet by humans, M. murinus preferred only six: D-tryptophan, dulcin, fructose, sucrose, SC45647, and xylitol. The animals neither preferred nor rejected acesulfame-K, alitame, aspartame, N-4-cyanophenyl-N'-cyanoguanidineacetate (CCGA), cyanosuosan, cyclamate, monellin, saccharin, suosan, super-aspartame, N-trifluoroacetyl-L-glutamyl-4-aminophenylcarbonitrile (TGC), and thaumatin. Together with previously recorded taste-nerve responses in M. murinus to acesulfame-K, alitame, aspartame, cyclamate, monellin, saccharin, and suosan [Hellekant et al., Chem Senses 18:307-320, 1993b], the current results suggest that these compounds either do not taste sweet to M. murinus or they have an aversive taste component. In this work we also relate these findings to phylogeny.     

Gap junctional vesicles in the neural tube of the splotch (Sp) mutant mouse.

The lumbosacral region of the neural tube was studied by means of transmission electron microscopy in retrospectively confirmed normal (+/+; Sp/+) and abnormal (Sp/Sp) embryos of the splotch mutant mouse early on the ninth day of gestation when the caudal neural groove is normally in the process of closing to form the neural tube. In abnormal embryos, a consistent feature is the presence of gap junctional vesicles, particularly in the region of the neural groove which subsequently fails to close, whereas these structures are rarely observed in similar areas of normal embryos. The possible significance of gap junctional vesicles is discussed in terms of cellular adhesion during early neurogenesis.     

Histological studies of normal and mutant mouse embryo hearts (a glycogen content study)

Mutant tail-short (Ts/+) embryos are developmentally retarded compared with normal +/+ litter mates. The development of the heart of Ts/+ embryos is severely affected if the tail-short gene is transferred to a new genetic (50% A/Gr) background. The aim of the present study was to investigate the glycogen content of the sinus muscle, the cushion and the atrial and ventricular walls of the heart. In normal embryos the sinus muscle is well developed by the 15th day post coitum (d.p.c.) and is crowded with glycogen granules. In Ts/+ mutant embryos, on the other hand, the development of this muscle is retarded and it contains only a little, diffusely distributed glycogen. The atrial and ventricular walls of embryos with a normal heart are well trabeculated and contain a large quantity of glycogen granules, while in mutant embryos they are less well trabeculated and contain only a little glycogen in a diffuse of finely granular form.     

Overexpression of the CT GalNAc transferase inhibits muscular dystrophy in a cleavage-resistant dystroglycan mutant mouse

Transgenic mice that express dystroglycan containing a serine to alanine point mutation at the normal site of cleavage (DG(S654A)) in their skeletal muscles fail to express endogenously cleaved dystroglycan and have muscular dystrophy [Neuromusc. Disord., in press]. Dystrophic DG(S654A) muscles have reduced binding of antibodies, including VIA4-1, that recognize carbohydrate antigens on alpha dystroglycan, a finding similar to muscles in some forms of congenital muscular dystrophy. Here we describe one DG(S654A) transgenic line where VIA4-1 antibody binding is absent in skeletal muscle. In theory, the absence of this carbohydrate antigen should inhibit later glycosylation events that would occur on the structure or structures this antibody binds to. One such modification is likely to be the CT carbohydrate antigen, which is present on alpha dystroglycan in muscles overexpressing the CT GalNAc transferase [Dev. Biol. 242 (2002) 58]. To test the relationship between the VIA4-1 and CT carbohydrate antigens, we made DG(S654A)/CT GalNAc transferase (DG(S654A)/CT) transgenic mice. Surprisingly, dystroglycan was cleaved, and alpha dystroglycan was glycosylated with the VIA4-1 antigen, in DG(S654A)/CT muscles. In addition, muscles in DG(S654A)/CT transgenic mice had little or no evidence of muscular dystrophy when compared to DG(S654A) littermates. These experiments demonstrate that the CT GalNAc transferase can affect the post-translational processing of dystroglycan and the extent of muscular dystrophy even in muscles where the VIA4-1 antigen is not present.     

Thymic dysfunction in the mutant diabetic (db/db) mouse

Thymic function has been explored in genetically diabetic homozygous C57BL/KsJ (db/db) mice by evaluating their serum thymic factor (FTS) levels with a rosette assay. As previously reported for other autoimmune mice (NZB or MRL/I mice), the age-dependent decline of FTS levels was significantly accelerated in diabetic mice when compared to heterozygous littermates. Furthermore, FTS inhibitory molecules were detected in db/db mouse sera (as early as 10 wk of age) as evaluated by their ability to absorb in vitro the activity of synthetic FTS in the rosette assay, and in vivo for their capacity to induce the disappearance of endogenous FTS when injected into normal mice. These inhibitors were shown to be immunoglobulins. Histologically, the thymus presented an accelerated involution starting with a cortical lymphocytic depletion and an increased number of Hassall's corpuscles. Ultrastructural studies showed alterations in thymic epithelial cells, mainly represented by an increasing number of cytoplasmic vacuoles. By means of indirect immunofluorescence with anti-FTS monoclonal antibodies, it was shown that the number of FTS+ cells was reduced in db/db mouse thymuses: at the age of 22 wk, diabetic mice had 10 times fewer FTS+ cells than heterozygotes of the same age. Taken together, these results indicate important abnormalities in the thymus of diabetic mice. It is possible that the associated lymphocyte dysfunction plays a role in the pathogenesis of the autoimmune disease presented by db/db mice.     

A rat homolog of the mouse deafness mutant jerker (je)

An autosomal recessive deafness mutant was discovered in our colony of Zucker (ZUC) rats. These mutants behave like shaker-waltzer deafness mutants, and their inner ear pathology classifies them among neuroepithelial degeneration type of deafness mutants. To determine whether this rat deafness mutation (−) defines a unique locus or one that has been previously described, we mapped its chromosomal location. F₂ progeny of (Pbrc:ZUC × BN/Crl) A/a B/b H/h+/− F₁ rats were scored for coat color and behavioral phenotypes. Segregation analysis indicated that the deafness locus might be loosely linked with B on rat Chromosome (Chr) 5 (RNO5). Therefore, 40 −/− rats were scored for BN and ZUC alleles at four additional loci, D5Mit11, D5Mit13, Oprd1, and Gnb1, known to map to RNO5 or its homolog, mouse Chr 4 (MMU4). Linkage analysis established the gene order (cM distance) as D5Mit11–(19.3)–B–(17.9)–D5Mit13–(19.2)–Oprd1–(21.5) − (1.2) Gnb1, placing the deafness locus on distal RNO5. The position of the deafness locus on RNO5 is similar to that ofjerker (je) on MMU4; the phenotypes and patterns of inheritance of the deafness mutation and je are also similar. It seems likely that the mutation affects the rat homolog of je. The rat deafness locus should, therefore, be named jerker and assigned the gene symbol Je. Received: 13 June 1995 / Accepted: 4 January 1996     

Direct evidence for axonal transport defects in a novel mouse model of mutant spastin-induced hereditary spastic paraplegia (HSP) and human HSP patients

Mutations in spastin are the most common cause of hereditary spastic paraplegia (HSP) but the mechanisms by which mutant spastin induces disease are not clear. Spastin functions to regulate microtubule organisation, and because of the essential role of microtubules in axonal transport, this has led to the suggestion that defects in axonal transport may underlie at least part of the disease process in HSP. However, as yet there is no direct evidence to support this notion. Here we analysed axonal transport in a novel mouse model of spastin-induced HSP that involves a pathogenic splice site mutation, which leads to a loss of spastin protein. A mutation located within the same splice site has been previously described in HSP. Spastin mice develop gait abnormalities that correlate with phenotypes seen in HSP patients and also axonal swellings containing cytoskeletal proteins, mitochondria and the amyloid precursor protein (APP). Pathological analyses of human HSP cases caused by spastin mutations revealed the presence of similar axonal swellings. To determine whether mutant spastin influenced axonal transport we quantified transport of two cargoes, mitochondria and APP-containing membrane bound organelles, in neurons from mutant spastin and control mice, using time-lapse microscopy. We found that mutant spastin perturbs anterograde transport of both cargoes. In neurons with axonal swellings we found that the mitochondrial axonal transport defects were exacerbated; distal to axonal swellings both anterograde and retrograde transport were severely reduced. These results strongly support a direct role for defective axonal transport in the pathogenesis of HSP because of spastin mutation.     

Poly(ADP ribose) polymerase-defective mutant cell clone of mouse L1210 cells.

By a sequential mutation and selection utilizing N-methyl-N'-nitro-N-nitrosoguanidine as a mutagen, we succeeded in separating a poly(ADP ribose) polymerase-defective mutant clone (Cl-3527) from a mouse L1210 cell clone (Cl-3). The enzyme activity per cell in Cl-3527 cells was only 8% of that in wild type L1210 (CCL 219) cells. Immunoblot analysis of the enzyme protein in crude extracts of the mutant and wild type cells revealed that the enzyme defect was manifested as the loss of a 113-kDa wild type enzyme band in Cl-3527. Further analysis of partially purified enzyme from Cl-3527 by immunoblotting revealed that the molecular size of the enzyme in Cl-3527 was 108 kDa and that the amount of the mutant enzyme protein was markedly decreased in Cl-3527. The mutant enzyme was much more heat-labile than the wild type enzyme but the Km for NAD+, requirements for Mg2+ and nicked DNA, and the inhibition by 3-aminobenzamide, a potent inhibitor of the enzyme, however, were not so different from those of wild type enzyme. The mutant cells showed prolonged doubling time, increased temperature-sensitivity, increased percentage of active enzyme on a treatment of cells at high temperature, and increased expression of plasma membrane NADase, compared to wild type cells. Introduction of wild type ADPR pol gene into Cl-3527 cells partially restored the ADPR pol activity and the heat-resistance.