- Email: [email protected]
Selective impairment of fast anterograde axonal transport in the peripheral nerves of asymptomatic transgenic mice with a G93A mutant SOD1 gene
Brain Research 819 Ž1999. 120–131
Research report
Selective impairment of fast anterograde axonal transport in the peripheral nerves of asymptomatic transgenic mice with a G93A mutant SOD1 gene Hitoshi Warita a , Yasuto Itoyama a , Koji Abe b
b, )
a Department of Neurology, Tohoku UniÕersity School of Medicine, Sendai, Japan Department of Neurology, Okayama UniÕersity Medical School 2-5-1 Shika-machi, Okayama 700-8558, Japan
Accepted 8 December 1998
Abstract Transgenic mice that express a mutant CurZn superoxide dismutase ŽSOD1. gene have been provided a valuable model for human amyotrophic lateral sclerosis ŽALS.. We studied a possible impairment of fast axonal transport in transgenic mice carrying a Gly 93 ™ Ala ŽG93A. mutant SOD1 gene found in human familial ALS ŽFALS.. Left sciatic nerve was ligated for 6 h in transgenic ŽTg. and age-matched wild-type ŽWT. mice. Immunohistochemical analyses were performed for accumulations of kinesin and cytoplasmic dynein on both sides of the ligation site. Clinical function and histology in the spinal cords, sciatic nerves and gastrocnemius muscles were also assessed. The mice were examined at an early asymptomatic stage Žaged 19 weeks. and a late stage Ž30 weeks. just before the development of the symptoms. WT mice showed an apparent increase in immunoreactivities for kinesin and cytoplasmic dynein at proximal and distal of the ligation, respectively. In contrast, the young Tg mice showed a selective decrease of kinesin accumulation in the proximal of the ligation. The mice were asymptomatic with a mild histological change only in muscles. The old Tg mice showed a marked reduction of the immunoreactivity for kinesin and cytoplasmic dynein on both sides of the ligation. They had a significant loss of spinal motor neurons, relatively small myelinated fiber densities of sciatic nerves, and severe muscular changes. These results provide direct evidence that the SOD1 mutation leads to impaired fast axonal transport, particularly in the anterograde direction at an early, asymptomatic stage preceding loss of spinal motor neurons and peripheral axons. This impairment may contribute to subsequent selective motor neuron death in the present model implicated for human FALS. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Axonal transport; SOD1; Transgenic mouse; Kinesin; Dynein; ALS
1. Introduction Amyotrophic lateral sclerosis ŽALS. is a progressive and fatal neurodegenerative disorder characterized by loss of motor neurons in the cortex, brainstem and spinal cord, which causes skeletal muscle wasting, weakness and ultiAbbreviations: AH, anterior horn of spinal cord; ALS, amyotrophic lateral sclerosis; ANOVA, one-way analysis of variance; D-LI, cytoplasmic dynein-like immunoreactivity; FALS, familial ALS; G93A, mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine; HE, hematoxylin and eosin; K-LI, kinesin-like immunoreactivity; MFD, myelinated fiber density; MND, motor neuron diseases; NF-H, neurofilament heavy subunit; NF-M, neurofilament intermediate subunit; NF-L, neurofilament light subunit; O, with older age; PB, phosphate buffer; PBS, phosphate-buffered saline; SALS, sporadic ALS; S.D., standard deviation; SOD1, CurZn superoxide dismutase; TB, toluidine blue; Tg, transgenic; Y, with younger age; WT, non-transgenic wild-type ) Corresponding author. Fax: q 81-86-235-7368; E-mail: [email protected]
mately complete paralysis. About 5 to 10% of all ALS patients are familial and inherited mainly with an autosomal dominant manner. Despite clinical and genetic heterogeneity, familial ALS ŽFALS. and sporadic ALS ŽSALS. have a similar pathology, suggesting a common pathogenesis. In approximately 15 to 20% of FALS cases, a variety of dominant, missense mutations and a small deletion in the CurZn superoxide dismutase ŽSOD1; EC 1.15.1.1. gene, have been identified w19,35,39x. Several lines of transgenic mouse models that express a mutant SOD1 gene have been established w23,37,47x. These transgenic mice develop phenotypes closely similar to human FALS in accordance with elevated levels of mutant SOD1 gene expression, although pathological findings in the central nervous system ŽCNS. are different from those of human FALS to some extent w15,16x. The SOD activities are normal or increased in these transgenic models and the mice with overexpression of wild-type SOD1 gene or a lower expression level of mutant SOD1 gene do not show
0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 1 3 5 1 - 1
H. Warita et al.r Brain Research 819 (1999) 120–131
ALS phenotypes. Knockout mice disrupted one or both alleles of SOD1 gene do not result in motor neuron degeneration w36x. Furthermore, no correlation is observed between the degree of reduced SOD activity and disease severity in human FALS with SOD1 mutations w6x. These observations support a hypothesis that mutant SOD1 protein has a novel toxic gain of function rather than loss of function. However, the mechanisms of how this ubiquitously expressed mutant protein leads to selective vulnerability of motor neurons are not clearly understood. Neurons are highly polarized and bidirectional axonal transport between somata and axon terminal is indispensable for their morphological and functional integrity. The axonal transport system is classified into two rate components: fast Ž100–400 mmrday. and slow Ž0.1–3 mmrday. components. The fast components consist of various membrane-bound organelles and synaptic membrane precursors, while the slow one includes cytoskeletal proteins and soluble enzymes w10,25,45x. Although the molecular mechanisms for slow axonal transport are not well characterized, recent studies revealed two types of microtubuleactivated ATPases, kinesin and cytoplasmic dynein, as fast axonal motors w9,32,34,44x. The anterograde motor kinesin primarily carries cargo toward the axon terminal w27,44x, whereas cytoplasmic dynein is passively associated with anterograde movement in an inactive form and then is actively involved in the retrograde transport toward the cell body from the axon terminal w28x. The aberrant accumulation of neurofilaments in perikarya and swollen proximal axons of spinal motor neurons is an early neuropathological hallmark in human ALS and other motor neuron diseases ŽMND. w18,24,42x. Overexpression of wild-type murine neurofilament light subunit ŽNF-L. w48x, human neurofilament heavy subunit ŽNF-H. gene w14x and a mutant NF-L gene w31x in transgenic mice exhibit perikaryal and axonal accumulation of neurofilaments leads to MND phenotypes. Although these data suggest a key role for these cytoskeletal proteins associated with reduced slow axonal transport in the pathogenesis of ALS w13,33,49x, NF-H mutations are identified
121
in few cases of human SALS w22x and not linked to human FALS cases w38,46x. Thus, it is likely that additional factors underlie the disease mechanisms. Besides these neurofilament pathology, impairment of fast axonal transport is also considered to be associated with the neuronal dysfunction in human ALS w12,41x. However, it has been controversial how the fast axonal transport is altered, and a possible change of fast axonal transport in peripheral nerves has not been well investigated in association with the FALS-linked SOD1 mutations w49x. To test whether the fast axonal components are modified and how they are altered developmentally from an early preclinical stage, we examined accumulations of kinesin and cytoplasmic dynein in ligated sciatic nerves of transgenic mice with FALS-linked mutant human SOD1, in addition to clinical and histological evaluation of spinal cord, peripheral nerves and skeletal muscles. 2. Materials and methods 2.1. Animal model Transgenic mice carrying a mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine ŽG93A. were used in this study. Their strain designation is B6SJL-TgN ŽSOD1-G93A. 1Gur dl , that were obtained from the Jackson Laboratory ŽBar Harbor, ME, USA. w23x. At around 30 weeks of age, the G93A transgenic mice developed progressive muscle weakness and spasticity in one or more limbs beginning with a posterior limb. One to two weeks later, they could not feed for themselves due to severe paralysis with hyperextended posture of hindlimbs. The G93A transgenic mice ŽTg. and age-matched non-transgenic wild-type B6SJL littermates ŽWT. were examined simultaneously. They were divided into four groups: Tg with younger age ŽTgrY; aged 18.6 " 0.5 weeks Žmean " S.D.., n s 5., Tg with older ŽTgrO; 30.2 " 1.1, n s 5., WT with younger ŽWTrY; 19.0 " 0.0, n s 4., and WT with older ŽWTrO; 30.2 " 1.1, n s 5. ŽTable 1.. Genomic DNA was extracted from tail
Table 1 Clinical settings, SOD activity and neurological evaluations
Number Sex Age Žw. Body weight Žg. SOD activity ŽUrmg. Circular cage rotation Žnumber. Transverse bar rotation Žnumber. Angle of slippage Ždegree of slope.
WTrY
WTrO
TgrY
TgrO
ns4 M s 2, F s 2 19.0 " 0.0 27.5 " 5.5 184.0 " 13.4 253.0 " 72.5 99.3 " 32.5 39.3 " 1.3
ns5 M s 2, F s 3 30.2 " 1.1 30.6 " 4.6 182.2 " 34.9 254.2 " 131.1 141.4 " 85.1 36.2 " 3.3
ns5 M s 4, F s 1 18.6 " 0.5 26.8 " 2.6 391.8 " 47.1a 262.2 " 127.6 86.6 " 22.1 36.2 " 3.3
ns5 M s 3, F s 2 30.2 " 1.1 28.8 " 5.9 236.8 " 16.3 b 100.4 " 82.7 86.0 " 64.7 34.6 " 4.2
Each value represents means" S.D. WT s non-transgenic wild-type mice, Tg s transgenic mice carrying mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine, Y s with younger age, O s with older age, M s Male, and F s Female. a p - 0.01 relative to the WTrY, b p - 0.05 to the TgrY group.
122
H. Warita et al.r Brain Research 819 (1999) 120–131
biopsy. The transgene of the G93A mutant human SOD1 was confirmed by specific primers for exon 4 of the gene w39x. The enzyme activity of SOD1 protein Žunits per milligram of protein, Urmg. was measured by our previ-
ous method w4x. In brief, hemoglobin was extracted from peripheral erythrocytes by ethanol and chloroform, and the aqueous phase was recovered for measurement of the SOD activity. For clinical and neurological evaluation of the
Fig. 1. Histology in lumber cord anterior horns and gastrocnemius muscles. High power view of HE-stained transverse sections of L4–L5 spinal anterior horns Ža, c, e and g. and gastrocnemius muscles Žb, d, f and h.. In contrast to normal anterior horns of WTrY, WTrO and TgrY mice Ža, c and e., moderate loss of motor neurons and mild gliosis in TgrO mice are noted Žg.. The gastrocnemius muscles of TgrY mice show small group atrophy Žf, arrows.. Whereas, the muscles of TgrO present severe changes such as small group atrophy Žh, arrows., centrally-placed nuclei Žh, arrowheads., chromatic nuclear clump Žh, an open arrow., small angulated fibers Žh, open arrowheads. and hypertrophic fibers Žh, asterisk.. Centrally-placed nuclei are observed up to 1% of muscle fibers of normal control mice Žd, arrowheads.. Magnification: =100. Scale bar: 100 mm.
H. Warita et al.r Brain Research 819 (1999) 120–131
four groups of mice, body weight, rolling number of circular cage, grasping on a rolling column, and angle of slipping down from woody slope were measured as described previously before the ligation of their peripheral nerves w2x. 2.2. Ligation of mouse peripheral nerÕes Under anesthesia with inhalation of a nitrous oxider oxygenrhalothane Ž68%:30%:2%. mixture, the skin of the left lower limb was incised. The left sciatic nerve was then ligated very tightly at mid-thigh level with surgical thread ŽProlene Blue Monofilament Ž6-0., Ethicon, NJ, USA. w28x, followed by suture of the skin to prevent infection. In addition, the skin of the contralateral side was incised and sutured as a sham-operation and the mice were kept in a restricted cage. Immediately after ligation of sciatic nerves, the animals presented severe paralysis of left lower limbs. After 6 h of the sciatic nerve ligation, the mice were decapitated under deep anesthesia with ether. Pieces of the left sciatic nerve both proximal and distal Žat least 7 mm of each. to the ligated portion, right sciatic nerves, lumber spinal cord and bilateral gastrocnemius muscles were quickly removed. The nerves and spinal cord were rapidly frozen in powdered dry ice and the muscles were in 2-methylbutane ŽWako, Osaka, Japan. with liquid nitrogen. The samples were stocked at y808C. During the experiment, the mice were treated in accordance with the declaration of Helsinki and the guiding principles in the care and use of animals. All experimental protocols and procedures were approved by the Animal Committee of the Tohoku University School of Medicine, Japan.
123
2.4. QuantitatiÕe morphology Spinal cord neuronal counts were performed manually by an examiner blinded to the mouse groups. For each mouse, at least two or more Nissl-stained sections were analyzed in the spinal segments from L4 through L5. For each section, the number of motor neurons in both sides of the anterior horns was measured under light microscopy at a 100 = magnification. The average number of motor neurons in each section was added and divided by the number of sections counted in order to provide the number of motor neurons in unilateral anterior horn per section. Only cell profiles containing a distinct nucleus with nucleolus were counted in order to avoid double counting. The entire sciatic nerve specimen with TB staining was measured using computerized software ŽNIH Image Ver. 1.61, National Institutes of Health, USA. in contiguous non-
2.3. Histological study Transverse sections of the lumber cord ŽL4–L5. and the muscle samples were cut Ž10 mm of thickness. on a cryostat at y208C. A set of the sections of the spinal cord and muscle was stained with hematoxylin and eosin ŽHE.. Another cross-sections of spinal cord were stained for Nissl substance using Cresyl violet for neuronal counting. Small pieces of the sciatic nerves away from the ligated portion were fixed with 2% glutaraldehyde in 0.1 M phosphate buffer ŽPB, pH 7.4. for 2 h at room temperature, followed by 1% osmium tetroxide in 0.1 M PB at 48C, dehydrated in alcohols and embedded in Epon; Epon 812 ŽOuken, Tokyo, Japan.r1-Dodecenylsuccinic Anhydride ŽDDSA; Ouken.rMethylnadic Anhydride ŽMethyl-5norbornene-2,3-dicarboxylic Anhydride ŽMNA; Nissin EM, Tokyo, Japan.r2,4,6-Tris Ždimethylaminomethyl. phenol ŽDMP-30; TAAB Laboratories Equipment, Reading, UK. in the ratio of 45.4%:25.2%:28.0%:1.4%. Transverse sections with 0.5 mm thickness were stained with toluidine blue ŽTB..
Fig. 2. ŽA. Number of motor neurons in the anterior horn of lumber cord. Neuronal counting with Nissl-stained specimen shows significant decrease in the number of TgrO mice Ž19.2"3.3, ns 5. compared to WTrY Ž28.0"3.4, ns 4, ) p- 0.01., WTrO Ž28.2"2.4, ns 5, ) p0.01. and TgrY mice Ž27.8"3.2, ns 5, ) p- 0.01.. In contrast, the number of TgrY mice is not reduced significantly. ŽB. Myelinated fiber density ŽMFD. of sciatic nerve. The sciatic nerve sections with TB staining were measured in order to provide MFD Žrmm2 .. Although TgrO mice show relatively smaller MFD Ž20,075"1,714.3, ns 5., there are no significant differences among the four category of mice.
124
H. Warita et al.r Brain Research 819 (1999) 120–131
overlapping field under microscope at a 200 = magnification. The myelinated axon caliber Žmm. and the myelinated fiber density Žnumber of myelinated fibers per mm2 , MFD. of each transverse section was calculated. Histogram of the
myelinated axon caliber for each mouse category was constructed with statistical analyses. The measurement of myelinated axons in a fixed area Ž8.8 = 10y2 mm2 . for each mouse was submitted for the histogram.
Fig. 3. Representative microphotographs of the sciatic nerves with TB staining Ža, c, e and g. and histograms of myelinated axon caliber Žb, d, f and h. are exhibited. No histological changes are shown in the sciatic nerves Ža, c, e and g.. The histogram of sciatic nerve axon caliber showed a similar pattern with a peak in the class of 2.0 to 3.0 mm in all the four groups of mice. The histogram shows a similar pattern with a peak in the class of 2.0 to 3.0 mm in all the four groups of mice Žb, d, f and h.. There are developmental differences in axon caliber classes of 2.0–3.0 Ž)) p - 0.01., 3.0–4.0 Ž)) p - 0.01., 4.0–5.0 Ž) p - 0.05., 5.0–6.0 Ž) p - 0.05. and 6.0–7.0 mm Ž)) p - 0.01. between WTrY and WTrO Žb and d. or TgrY and TgrO mice Žf and h.. However, no differences are presented between age-matched groups such as WTrY and TgrY Žb and f. or WTrO and TgrO Žd and h.. Original magnification in the left panels Ža, c, e and g.: 100 = , Scale bar: 100 mm.
H. Warita et al.r Brain Research 819 (1999) 120–131
2.5. Immunohistochemistry The longitudinal sections Ž10 mm of thickness. of the ligated sciatic nerves were carefully obtained on a cryostat at y228C, and stained with an antibody for kinesin and cytoplasmic dynein by employing an immunohistochemical technique according to our previous report w1x. Briefly, after the treatment with 10% normal horse serum in phosphate-buffered saline ŽPBS, pH 7.4. for 2 h to avoid non-specific stain, the slides were washed and incubated with the primary antibody in a buffer containing 10% normal horse serum and 0.3% Triton X-100 ŽWako. at 48C for 16 h. The primary antibodies used in the present study were as follows: mouse monoclonal anti-bovine brain kinesin heavy chain diluted at 1:500 ŽChemicon International, Temecula, CA, USA. and mouse monoclonal antibovine brain cytoplasmic dynein 74 kD intermediate chain ŽChemicon International. at 1:500. Endogenous peroxidase was blocked with 0.3% H 2 O 2 and 10% methanol for 20 min. The sections were washed and incubated with biotinylated anti-mouse IgG Ž1r200. in the buffer for 3 h, and then incubated with avidin–biotin–horseradish peroxidase complex solution using a kit ŽVECTASTAIN Elite ABC kit, Vector Laboratories, Burlingame, CA, USA. for 30 min. Finally, immunoreactivity was revealed with a minute exposure with 3,3X-diaminobenzidine tetrahydrochloride Ž0.5 mgrml in 100 mM Tris–HCl buffer ŽpH 7.2., Wako. and 0.02% H 2 O 2 . The degree of immunohistochemically reactive product was rated for semiquantitation as follows: null Žy., small number of reactive product or weak staining Ž"., moderate number of Žq., large number of Žqq ., and large number of densely reactive product Žqqq .. In addition, optical density of kinesin and cytoplasmic dynein immunoreactivity in the longitudinal sections of sciatic nerve was quantified in arbitrary units by using the NIH Image. For each mouse, at least five immunostained sections were measured to calculate the average density for a defined area Ž4.4 = 10y2 mm2 . just adjacent to the ligation site.
Service, Germany. and the null hypothesis was rejected at the 0.05 level.
3. Results 3.1. Clinical settings, SOD actiÕity, and neurological eÕaluation Clinical settings, the SOD activity, and neurological estimates are summarized in Table 1. All the G93A mutant SOD1 Tg mice ŽTgrY and TgrO. and WT littermates ŽWTrY and WTrO. were clinically free from any symptoms such as muscle weakness, reduced motility or abnormal posture. No significant differences are observed in body weight, number of circular cage rotation, transverse bar rotation and angle of slippage ŽTable 1.. Although the difference was not detected by one-way ANOVA, TgrO mice showed relatively smaller number of circular cage rotation Ž100.4 " 82.7, mean " S.D., n s 5. compared to TgrY group Ž262.2 " 127.6, n s 5.. The SOD activity of TgrY mice was approximately two folds higher Ž391.8 " 47.1 Urmg, n s 5. than WTrY group Ž184.0 " 13.4, n s 4, p - 0.01. ŽTable 1.. However, TgrO mice had approximately 40% lower SOD activity Ž236.8 " 16.3, n s 5. than that of TgrY mice Ž391.8 " 47.1, n s 5, p 0.05..
Table 2 Immunoreactivity for kinesin and cytoplasmic dynein Groups
Case no.
WTrY
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
WTrO
2.6. Statistics TgrY
Results are expressed as mean " standard deviation ŽS.D... Statistical analyses were performed using one-way analysis of variance ŽANOVA. among means of value with the category of mice as the independent factor. Multiple pair-wise comparisons between means were tested by Tukey–Kramer post-hoc test when ANOVA showed significant differences Ž p - 0.05.. In regard to the histograms of myelinated axon caliber, the frequency in each class, skewness and kurtosis were analyzed by one-way ANOVA as mentioned above. In addition, two-way ANOVA was used for testing the histograms with the different types of mice and classes of axon caliber as the independent factors. All statistical analyses were performed using computerized software ŽWinSTAT w Ver. 3.1, Scientific Software
125
TgrO
Kinesin
Dynein
Proximal
Distal
Proximal
Distal
qqq qqq qqq qqqq qq qqq qq qqq qq qq q q qq q q " " q "
q q q q q qq q q q q q q q " q " " " "
qq qq qq qqq qqq qq qq qq qq qq q qq qq q q q q q "
qqqq qqqq qqq qqqq qqqq qqq qqq qqq qqqq qqq qq qqq qqq qqq q q qq qq "
The degree of immunohistochemically reactive product was rated for semiquantitation as follows: null Žy., small number of reactive product or weak staining Ž"., moderate number of Žq., large number of Žqq., and large number of densely reactive product Žqqq.. WT s non-transgenic wild-type mice, Tg s transgenic mice carrying mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine, Ys with younger age, Os with older age.
126
H. Warita et al.r Brain Research 819 (1999) 120–131
ŽFig. 1g., while the other mice ŽWTrY, WTrO and TgrY. showed no histological change ŽFig. 1a,c,e.. Neuronal counting with Nissl-stained specimen revealed that
Fig. 4. Immunohistochemistry for kinesin on both sides just proximal Žleft panels; a, c, e and g. and distal Žright panels; b, d, f and h. to the ligated portion Žarrowheads. of sciatic nerves. The immunoreactivity was decreasing in order of WTrY and WTrO, TgrY, and TgrO Ža, c, e and g. mice on proximal side of the ligation. TgrY mice show apparently weaker K-LI on proximal side Že. than those of WTrY Ža. and WTrO Žc.. TgrO mice show only slight staining on both sides of the ligation Žg and h.. Original magnification: 100=, Scale bar: 100 mm.
3.2. Histological study In lumber spinal cord ŽL4–L5. anterior horns ŽAH., histological study with HE staining presented moderate loss of motor neurons and reactive gliosis in TgrO mice
Fig. 5. Immunohistochemistry for cytoplasmic dynein on both sides just proximal Žleft panels; a, c, e and g. and distal Žright panels; b, d, f and h. to the ligated site Žarrowheads. of sciatic nerves. In contrast to the K-LI, the reduction of the immunoreactivity is mild in TgrY mice Žf. on distal side of the ligation compared to those of WTrY Žb. and WTrO Žd. mice. Note the staining proximal to the ligated site of TgrY is weaker Že. than those of WTrY and WTrO Ža and c. as well as immunostaining of kinesin. Moreover, TgrO mice show obvious depletion of the immunoreactivity on both sides of the ligated site Žg and h.. Original magnification: 100=, Scale bar: 100 mm.
H. Warita et al.r Brain Research 819 (1999) 120–131
the number of motor neurons in AH of TgrO mice was significantly smaller Ž19.2 " 3.3, n s 5. than those of WTrY Ž28.0 " 3.4, n s 4, p - 0.01., WTrO Ž28.2 " 2.4, n s 5, p - 0.01., and TgrY mice Ž27.8 " 3.2, n s 5, p 0.01. ŽFig. 2A.. Whereas, the number of TgrY motor neurons in AH Ž27.8 " 3.2. did not show significant decrease compared to the other three groups. Otherwise, the sections stained with HE or Cresyl violet showed no apparent histological change or neuronal loss in spinal cord of the four groups of mice. The sciatic nerve sections with TB staining are depicted in Fig. 3a,c,e,g. Computerized measurement of MFD showed no significant differences, although TgrO mice had relatively smaller MFD Ž20,075 " 1,714.3rmm 2 , n s 5. compared to WTrO Ž22,104 " 1,671.9, n s 5. and TgrY Ž22,982 " 1,613.4, n s 5. mice ŽFig. 2B.. The histogram of sciatic nerve axon caliber showed a similar pattern with a peak in the class of 2.0 to 3.0 mm in all the four groups of mice. There were developmental differences in axon caliber classes of 2.0–3.0 Ž p - 0.01., 3.0–4.0
127
Ž p - 0.01., 4.0–5.0 Ž p - 0.05., 5.0–6.0 Ž p - 0.05., and 6.0–7.0 mm Ž p - 0.01. between WTrY and WTrO, or TgrY and TgrO mice ŽFig. 3b,d,f,h.. These differences were, however, not observed between age-matched groups as WTrY and TgrY, or WTrO and TgrO. Testing by two-way ANOVA also showed no significant differences in the frequency distribution of the myelinated axon caliber among the four groups. The skewness and kurtosis of the histograms did not show significant differences among the four groups Žone-way ANOVA, data not shown.. Bilateral gastrocnemius muscles of both WTrY and WTrO mice did not show any histological abnormalities ŽFig. 1b and d.. On the other hand, the muscles of TgrY mice are mildly involved in which small group atrophy was modestly observed ŽFig. 1f, arrows.. In contrast, TgrO muscles showed a severe histological change such as small group atrophy ŽFig. 1h, arrows., centrally-placed nuclei ŽFig. 1h, arrowheads., chromatic nuclear clump ŽFig. 1h, an open arrow., small angulated fibers ŽFig. 1h, open arrowheads. and hypertrophic fibers ŽFig. 1h, asterisk..
Fig. 6. Optical density for kinesin- and cytoplasmic dynein-like immunoreactivity ŽK-LI and D-LI. on both sides of the ligated site of sciatic nerve sections. In asymptomatic TgrY mice, the densities of both K-LI and D-LI are significantly decreased only on proximal side of the ligated site compared to those of WTrY and WTrO Ž p - 0.01.. In contrast, TgrO mice show significantly reduced densities of K-LI and D-LI on both sides of the ligation compared to those of WTrY and WTrO Ž p - 0.01..
128
H. Warita et al.r Brain Research 819 (1999) 120–131
3.3. Immunohistochemical comparison of the kinesin and cytoplasmic dynein accumulations Immunohistochemistry of both just proximal and distal to the ligated site of sciatic nerves with anti-kinesin and anti-cytoplasmic dynein antibodies are summarized with semiquantitation in Table 2 and Fig. 6. In all the cases, kinesin-like immunoreactivity ŽK-LI. and cytoplasmic dynein-like one ŽD-LI. in the sciatic nerves was observed on both sides of the ligated portion ŽFigs. 4–6.. While K-LI was predominant on the proximal side, D-LI was denser on the distal side ŽFigs. 4–6.. Staining was restricted to both sides of the ligation and presented a gradational pattern, in which the immunoreactivity was more intense and larger number as the region was closer to the ligated portion. The degree of K-LI proximal to the ligature Žproximal K-LI. was clearly different among the four groups of mice. It was decreasing in order of WTrY, WTrO, TgrY and TgrO ŽTable 2; Fig. 4a,c,e,g and Fig. 6.. The proximal K-LI of TgrY was apparently weaker Žq; qq; optical densitys 2.51 " 0.49, p - 0.01, one-way ANOVA. than those of WTrY Žqqq; qqqq; 3.82 " 0.61. and WTrO Žqq ; qqq; 3.77 " 0.65. ŽTable 2; Fig. 4a,c,e,g and Fig. 6.. TgrO had only slight K-LI on both proximal Ž"; q; 0.53 " 0.09. and distal Ž"; q; 0.26 " 0.10. to the ligated site ŽTable 2; Fig. 4g,h and Fig. 6.. The K-LI distal to the ligation Ždistal K-LI. was generally weak and not clearly different among WTrY, WTrO and TgrY, although optical density in TgrO mice showed significantly weak K-LI Ž"; q; 0.26 " 0.10, p - 0.01. ŽTable 2; Fig. 4b,d,f and Fig. 6.. In contrast to K-LI, the reduction of distal D-LI was mild in TgrY mice Žqq ; qqq; 3.98 " 0.28. compared to WTrY Žqqq; qqqq; 4.17 " 0.34. and WTrO mice Žqqq; qqqq; 4.08 " 0.33. ŽTable 2; Fig. 5b,d,f and Fig. 6.. As well as the proximal K-LI, the proximal D-LI of TgrY was also weaker Žq; qq; 0.99 " 0.22, p - 0.01. than those of WTrY and WTrO Žqq ; qqq; 1.71 " 0.22 and 1.40 " 0.22. ŽTable 2; Fig. 5a,c,e and Fig. 6.. TgrO mice showed weak D-LI Ž"; qq; proximal, 0.44 " 0.15, p - 0.01 and distal, 2.86 " 0.30, p - 0.01. on both sides of the ligation ŽTable 2; Fig. 5g,h and Fig. 6.. When the primary antibody was omitted or replaced with normal IgG, neither K-LI nor D-LI was detected, and every specimen of contralateral non-ligated sciatic nerves showed only slight number of or faint staining Ždata not shown..
4. Discussion Although body weight and clinical scores reflecting motility, motor activity and limb muscle strength were not significantly different among the four groups of mice ŽTable 1., histological data in muscle ŽFig. 1b,d,f,h. and circular cage score ŽTable 1. suggest that TgrY mice were
at an early, asymptomatic stage and TgrO mice were just before the onset of the disease. These clinical observations of the G93A transgenic mice were different from the previous report in the age of onset. It was about 3 months later in the present Tg mice compared to the original G93A transgenic line w15,23x. Histopathological study revealed that TgrY mice showed only mild denervated change in gastrocnemius muscles without loss of spinal motor neurons in lumber cord ŽL4–L5., while TgrO showed approximately 30% loss Žcompared to WTrO. of spinal motor neurons with severe neurogenic muscular changes ŽFig. 1e–h and Fig. 2A.. Although TgrO mice had relatively smaller MFD ŽFig. 2B and Fig. 3g., TgrY mice were exempt from apparent histological changes of sciatic nerves ŽFig. 3e.. These observations suggest that histological alterations begin with muscular changes followed by the spinal cord and peripheral nerve pathology in the present model. In addition, the developmental change between the younger and older WT mice was shown not in the spinal cord and skeletal muscles but in the size distribution of myelinated axons of peripheral nerves ŽFig. 3.. Although centrally-placed nuclei are found in a small fraction of normal muscles Žup to 1% in adult human, also found similarly in the present WT mice, Fig. 1d, an arrowhead., it is generally interpreted as a myogenic change with regeneration of muscle fibers. Thus, the observation that the present G93A transgenic mice had increasing number of centrally-placed nuclei in common Ž10–20% in TgrO, Fig. 1h, arrowheads. suggests a distinct process different from a simple denervation. In the previous report of the G93A transgenic mice, the vacuolar change of spinal motor neurons was observed in the latest stage w16x. In lumber cord of the present G93A transgenic mice, however, any vacuolar change was not obviously found under light microscopy ŽFig. 1e and g.. The elevation of SOD activity in the blood samples was smaller in the present Tg mice Žapproximately 2.1 times higher in TgrY than that of WTrY mice, Table 1. than the previous report Žfour folds in brain. w23x. Therefore, the later onset and the lack of vacuolar change in the present mouse model may be related to the relatively mild level of overexpression of the mutant SOD1 gene probably because of decreased copy number with repeated mating. Indeed, the transgenic lines of lower copy number of the G93A mutant SOD1 gene have been reported to show delayed onset and fewer vacuoles with more neurofilamentous inclusions in spinal motor neurons at later stage, that resemble more closely to human ALS pathology w17,49x. The 6-h ligation of sciatic nerve resulted in accumulative expression of fast axonal motors, kinesin and cytoplasmic dynein on both sides of the ligated portion. The degree of proximal K-LI and distal D-LI to the ligation may represent the amount or the rates of anterograde and retrograde fast axonal traffic, respectively. The TgrY mice showed a marked decrease of proximal K-LI with only a minimal decline of distal D-LI in the ligated sciatic nerves
H. Warita et al.r Brain Research 819 (1999) 120–131
ŽTable 2; Fig. 4e, Fig. 5f and Fig. 6.. These observations may reflect a possible impairment of fast axonal transport particularly in anterograde direction at an early, asymptomatic stage of the disease. In TgrY mice, not only the K-LI but also the D-LI was reduced on proximal side of the ligated site, although both K-LI and D-LI on distal side of the ligated site were not significantly decreased compared to WTrY mice ŽFig. 6.. These data support a defective fast anterograde transport machinery rather than declines in protein levels of these motors in the transgenic mice. Marked depletion of K-LI and D-LI in TgrO mice suggests more severe dysfunction of bidirectional fast axonal transport ŽFigs. 4–6.. This is the first report that suggests an early but selective impairment of fast anterograde axonal transport in the peripheral nerves of FALSlinked SOD1 transgenic mice. The weak immunostaining of proximal K-LI and distal D-LI in WTrO mice ŽFig. 4c, Fig. 5d and Fig. 6. compared to WTrY ŽFig. 4a, Fig. 5b and Fig. 6. suggests an association between reduced fast axonal transport and aging process. Since the number of motor neurons and MFD of sciatic nerves were not reduced and histological change was restricted to the skeletal muscles in TgrY mice ŽFig. 1f., it is suggested the selective reduction of K-LI was not caused by loss of spinal motor neurons or peripheral nerve lesion. The present data suggest that a functional impairment of selective anterograde fast axonal transport may precede motor neuron loss and axonal degeneration. The sciatic nerve contains not only motor but sensory axons. No histological changes were, however, observed in posterior horns of spinal cord even in TgrO mice that had severely reduced K-LI and D-LI accompanied with significant loss of spinal motor neurons. Thus, the reduction of K-LI and D-LI in the sciatic nerves seems to represent their ‘motor’ axonal dysfunction. There is no certain methods to distinguish motor from sensory axons by morphologic or immunocytochemical determination in peripheral nerves at this moment. Further study to discriminatively detect a change in motor and sensory nerves may elucidate the mechanism more clearly. Recent studies revealed an association of kinesin and cytoplasmic dynein superfamily proteins with a various membranous organelles w26x. Fast anterograde traffic involves precursors of synaptic vesicles, plasma membrane constituents and mitochondria, while fast retrograde traffic concerns prelysosomal vesicles, growth factors, mitochondria and recycled proteins from anterograde transport. Thus, a disruption of fast anterograde transport could cause a reduced action potential propagation, defective neurotransmitter release and a metabolic dysfunction at the nerve terminals. In motor neurons, therefore, impaired fast anterograde transport would primarily result in neuromuscular dysfunction. In fact, an electromyographic impairment of neuromuscular function was recently observed precedent to the beginning of clinical signs and the spinal motor neuron loss in the G93A mutant SOD1 transgenic
129
mice w30x. The precedence of neurogenic muscular change in the present TgrY mice is compatible with the reported data. Although previous works of fast axonal transport in human ALS have shown controversial results w7,11,12x, more recent studies have suggested reduction of fast axonal transport in the proximal axons of spinal motor neurons in human ALS w41x and also in the transgenic mice carrying the G93A mutant SOD1 w49x. The latter report showed an impairment of both slow and fast anterograde transport in the ventral root of the transgenic mice that express a lower copy number of G93A mutant SOD1 similar to the present mice. In the report, slow components including neurofilaments and tubulin decline as early as 26 weeks of age without loss of spinal motor neurons and ventral root axons before the retardation of a part of fast anterograde components at the age presenting clinical symptoms Žabout 29 weeks of age. w49x. By contrast, the present results suggest an much earlier alteration Žabout 19 weeks of age. of the fast anterograde transport in the peripheral nerve. The pathological findings reflecting an excessive accumulation of neurofilaments are reported in human FALS with SOD1 mutations w40x and in the transgenic mice harboring the G93A mutant SOD1 gene w15,16,23,43x. Mutant SOD1 has been proposed to allow increased access of peroxynitrite and result in nitration of tyrosine residues on proteins w5x. Recently, increased nitration of protein– tyrosine residue was reported in spinal motor neurons of the human ALS w3x and in spinal cord and cerebral cortex of transgenic mice with the G93A mutation w21x. Since human NF-L is rich in tyrosine residue, the tyrosine nitration may lead to neurofilament disruption, suggesting an association between SOD1 mutations and the neurofilament pathology. In the masses of accumulated neurofilaments in neuronal perikarya and swollen axons, membrane-bound cytoplasmic organelles were also accumulated. It has been suggested, therefore, that a secondary inhibition of fast axonal transport may also contribute to the MND phenotypes w8,13,49x associated with neurofilament pathology and also with SOD1 mutations. The recent work has suggested that axonal neurofilaments does not play essential role in axonal degeneration mediated by Gly 37 ™ Arg ŽG37R. mutant SOD1 w20x. These reported data and the present results raise a possibility that the mutations in SOD1 gene directly or indirectly lead to impaired fast axonal transport in the peripheral nerves. Furthermore, mutations of the kinesin heavy chain gene in Drosophila cause a specific disruption of fast axonal transport leading to progressive distal paralysis w29x, suggesting an inhibition of fast axonal transport causes a MND phenotype by itself. Acknowledgements The authors would like to thank Mr. Y. Onodera, A. Ito and Ms. M. Takahashi for technical assistance and are
130
H. Warita et al.r Brain Research 819 (1999) 120–131
grateful to Drs. T. Hayashi, H. Kitagawa, M. Nagai, I. Nagano, M. Tateyama and M. Sakurai for helpful discussions and generous support. This work was partly supported by Grant-in-Aid for Scientific Research ŽB. 09470151 from the Ministry of Education, Science and Culture of Japan, and by grants ŽTashiro K and Itoyama Y. from the Ministry of Health and Welfare of Japan.
w18x w19x
w20x
References w21x w1x K. Abe, M. Aoki, J. Kawagoe, T. Yoshida, A. Hattori, K. Kogure, Y. Itoyama, Ischemic delayed neuronal death. A mitochondrial hypothesis, Stroke 26 Ž1995. 1478–1489. w2x K. Abe, S. Morita, T. Kikuchi, Y. Itoyama, Protective effect of a novel free radical scavenger, OPC-14117, on wobbler mouse motor neuron disease, J. Neurosci. Res. 48 Ž1997. 63–70. w3x K. Abe, L.H. Pan, M. Watanabe, T. Kato, Y. Itoyama, Induction of nitrotyrosine-like immunoreactivity in the lower motor neuron of amyotrophic lateral sclerosis, Neurosci. Lett. 199 Ž1995. 152–154. w4x M. Aoki, M. Ogasawara, Y. Matsubara, K. Narisawa, S. Nakamura, Y. Itoyama, K. Abe, Familial amyotrophic lateral sclerosis ŽALS. in Japan associated with H46R mutation in CurZn superoxide dismutase gene: a possible new subtype of familial ALS, J. Neurol. Sci. 126 Ž1994. 77–83. w5x J.S. Beckman, M. Carson, C.D. Smith, W.H. Koppenol, ALS, SOD and peroxynitrite, Nature 364 Ž1993. 584. w6x A.C. Bowling, E.E. Barkowski, D. McKenna-Yasek, P. Sapp, H.R. Horvitz, M.F. Beal, R.H. Brown Jr., Superoxide dismutase concentration and activity in familial amyotrophic lateral sclerosis, J. Neurochem. 64 Ž1995. 2366–2369. w7x W.G. Bradley, P. Good, C.G. Rasool, L.S. Adelman, Morphometric and biochemical studies of peripheral nerves in amyotrophic lateral sclerosis, Ann. Neurol. 14 Ž1983. 267–277. w8x S. Brady, Motor-neuron diseases. Interfering with the runners, Nature 375 Ž1995. 12–13. w9x S.T. Brady, A novel brain ATPase with properties expected for the fast axonal transport motor, Nature 317 Ž1985. 73–75. w10x S.T. Brady, Molecular motors in the nervous system, Neuron 7 Ž1991. 521–533. w11x A.C. Breuer, M.B. Atkinson, Fast axonal transport alterations in amyotrophic lateral sclerosis ŽALS. and in parathyroid hormone ŽPTH.-treated axons, Cell Motil. Cytoskeleton 10 Ž1988. 321–330. w12x A.C. Breuer, M.P. Lynn, M.B. Atkinson, S.M. Chou, A.J. Wilbourn, K.E. Marks, J.E. Culver, E.J. Fleegler, Fast axonal transport in amyotrophic lateral sclerosis: an intra-axonal organelle traffic analysis, Neurology 37 Ž1987. 738–748. w13x J.F. Collard, F. Cote, ˆ ´ J.P. Julien, Defective axonal transport in a transgenic mouse model of amyotrophic lateral sclerosis, Nature 375 Ž1995. 61–64. w14x F. Cote, ˆ ´ J.F. Collard, J.P. Julien, Progressive neuronopathy in transgenic mice expressing the human neurofilament heavy gene: a mouse model of amyotrophic lateral sclerosis, Cell 73 Ž1993. 35–46. w15x M.C. Dal Canto, M.E. Gurney, Development of central nervous system pathology in a murine transgenic model of human amyotrophic lateral sclerosis, Am. J. Pathol. 145 Ž1994. 1271–1279. w16x M.C. Dal Canto, M.E. Gurney, Neuropathological changes in two lines of mice carrying a transgene for mutant human Cu,Zn SOD, and in mice overexpressing wild type human SOD: a model of familial amyotrophic lateral sclerosis ŽFALS., Brain Res. 676 Ž1995. 25–40. w17x M.C. Dal Canto, M.E. Gurney, A low expressor line of transgenic mice carrying a mutant human Cu,Zn superoxide dismutase ŽSOD1. gene develops pathological changes that most closely resemble those
w22x
w23x
w24x
w25x
w26x w27x
w28x
w29x
w30x
w31x
w32x
w33x
w34x
w35x
in human amyotrophic lateral sclerosis, Acta Neuropathol. ŽBerlin. 93 Ž1997. 537–550. M.B. Delisle, S. Carpenter, Neurofibrillary axonal swellings and amyotrophic lateral sclerosis, J. Neurol. Sci. 63 Ž1984. 241–250. H.X. Deng, A. Hentati, J.A. Tainer, Z. Iqbal, A. Cayabyab, W.Y. Hung, E.D. Getzoff, P. Hu, B. Herzfeldt, R.P. Roos, C. Warner, G. Deng, E. Soriano, C. Smyth, H.E. Parge, A. Ahmed, A.D. Roses, R.A. Halliwell, M.A. Periak-Vance, T. Siddique, Amyotrophic lateral sclerosis and structural defects in Cu, Zn superoxide dismutase, Science 261 Ž1993. 1047–1051. J. Eyer, D.W. Cleveland, P.C. Wong, A.C. Peterson, Pathogenesis of two axonopathies does not require axonal neurofilaments, Nature 391 Ž1998. 584–587. R.J. Ferrante, L.A. Shinobu, J.B. Schulz, R.T. Matthews, C.E. Thomas, N.W. Kowall, M.E. Gurney, M.F. Beal, Increased 3-nitrotyrosine and oxidative damage in mice with a human copperrzinc superoxide dismutase mutation, Ann. Neurol. 42 Ž1997. 326–334. D.A. Figlewicz, A. Krizus, M.G. Martinoli, V. Meininger, M. Dib, G.A. Rouleau, J.P. Julien, Variants of the heavy neurofilament subunit are associated with the development of amyotrophic lateral sclerosis, Hum. Mol. Genet. 3 Ž1994. 1757–1761. M.E. Gurney, H. Pu, A.Y. Chiu, M.C. Dal Canto, C.Y. Polchow, D.D. Alexander, J. Caliendo, A. Hentati, Y.W. Kwon, H.X. Deng, W. Chen, P. Zhai, R.L. Sufit, T. Siddique, Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation, Science 264 Ž1994. 1772–1775. A. Hirano, H. Donnenfeld, S. Sasaki, I. Nakano, Fine structural observations of neurofilamentous changes in amyotrophic lateral sclerosis, J. Neuropathol. Exp. Neurol. 43 Ž1984. 461–470. N. Hirokawa, The molecular mechanism of organelle transport along microtubules: the identification and characterization of KIFs Žkinesin superfamily proteins., Cell Struct. Funct. 21 Ž1996. 357–367. N. Hirokawa, Kinesin and dynein superfamily proteins and the mechanism of organelle transport, Science 279 Ž1998. 519–526. N. Hirokawa, R. Sato-Yoshitake, N. Kobayashi, K.K. Pfister, G.S. Bloom, S.T. Brady, Kinesin associates with anterogradely transported membranous organelles in vivo, J. Cell Biol. 114 Ž1991. 295–302. N. Hirokawa, R. Sato-Yoshitake, T. Yoshida, T. Kawashima, Brain dynein ŽMAP1C. localizes on both anterogradely and retrogradely transported membranous organelles in vivo, J. Cell Biol. 111 Ž1990. 1027–1037. D.D. Hurd, W.M. Saxton, Kinesin mutations cause motor neuron disease phenotypes by disrupting fast axonal transport in Drosophila, Genetics 144 Ž1996. 1075–1085. P.F. Kennel, F. Finiels, F. Revah, J. Mallet, Neuromuscular function impairment is not caused by motor neurone loss in FALS mice: an electromyographic study, Neuroreport 7 Ž1996. 1427–1431. M.K. Lee, J.R. Marszalek, D.W. Cleveland, A mutant neurofilament subunit causes massive, selective motor neuron death: implications for the pathogenesis of human motor neuron disease, Neuron 13 Ž1994. 975–988. R.J. Lye, M.E. Porter, J.M. Scholey, J.R. McIntosh, Identification of a microtubule-based cytoplasmic motor in the nematode C. elegans, Cell 51 Ž1987. 309–318. J.R. Marszalek, T.L. Williamson, M.K. Lee, Z. Xu, P.N. Hoffman, M.W. Becher, T.O. Crawford, D.W. Cleveland, Neurofilament subunit NF-H modulates axonal diameter by selectively slowing neurofilament transport, J. Cell Biol. 135 Ž1996. 711–724. B.M. Paschal, R.B. Vallee, Retrograde transport by the microtubule-associated protein MAP 1C, Nature 330 Ž1987. 181– 183. A. Pramatarova, D.A. Figlewicz, A. Krizus, F.Y. Han, I. CeballosPicot, A. Nicole, M. Dib, V. Meininger, R.H. Brown, G.A. Rouleau, Identification of new mutations in the CurZn superoxide dismutase gene of patients with familial amyotrophic lateral sclerosis, Am. J. Hum. Genet. 56 Ž1995. 592–596.
H. Warita et al.r Brain Research 819 (1999) 120–131 w36x A.G. Reaume, J.L. Elliott, E.K. Hoffman, N.W. Kowall, R.J. Ferrante, D.F. Siwek, H.M. Wilcox, D.G. Flood, M.F. Beal, R.H. Brown Jr., R.W. Scott, W.D. Snider, Motor neurons in CurZn superoxide dismutase-deficient mice develop normally but exhibit enhanced cell death after axonal injury, Nat. Genet. 13 Ž1996. 43–47. w37x M.E. Ripps, G.W. Huntley, P.R. Hof, J.H. Morrison, J.W. Gordon, Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis, Proc. Natl. Acad. Sci. U.S.A. 92 Ž1995. 689–693. w38x K. Rooke, D.A. Figlewicz, F.Y. Han, G.A. Rouleau, Analysis of the KSP repeat of the neurofilament heavy subunit in familial amyotrophic lateral sclerosis, Neurology 46 Ž1996. 789–790. w39x D.R. Rosen, T. Siddique, D. Patterson, D.A. Figlewicz, P. Sapp, A. Hentati, D. Donaldson, J. Goto, J.P. O’Regan, H.X. Deng, Mutations in CurZn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis, Nature 364 Ž1993. 59–62. w40x G.A. Rouleau, A.W. Clark, K. Rooke, A. Pramatarova, A. Krizus, O. Suchowersky, J.P. Julien, D. Figlewicz, SOD1 mutation is associated with accumulation of neurofilaments in amyotrophic lateral sclerosis, Ann. Neurol. 39 Ž1996. 128–131. w41x S. Sasaki, M. Iwata, Impairment of fast axonal transport in the proximal axons of anterior horn neurons in amyotrophic lateral sclerosis, Neurology 47 Ž1996. 535–540. w42x M.L. Schmidt, M.J. Carden, V.M. Lee, J.Q. Trojanowski, Phosphate dependent and independent neurofilament epitopes in the axonal swellings of patients with motor neuron disease and controls, Lab. Invest. 56 Ž1987. 282–294.
131
w43x P.H. Tu, P. Raju, K.A. Robinson, M.E. Gurney, J.Q. Trojanowski, V.M. Lee, Transgenic mice carrying a human mutant superoxide dismutase transgene develop neuronal cytoskeletal pathology resembling human amyotrophic lateral sclerosis lesions, Proc. Natl. Acad. Sci. U.S.A. 93 Ž1996. 3155–3160. w44x R.D. Vale, T.S. Reese, M.P. Sheetz, Identification of a novel force-generating protein, kinesin, involved in microtubule-based motility, Cell 42 Ž1985. 39–50. w45x R.B. Vallee, G.S. Bloom, Mechanisms of fast and slow axonal transport, Annu. Rev. Neurosci. 14 Ž1991. 59–92. w46x J.D. Vechio, L.I. Bruijn, Z. Xu, R.H. Brown Jr., D.W. Cleveland, Sequence variants in human neurofilament proteins: absence of linkage to familial amyotrophic lateral sclerosis, Ann. Neurol. 40 Ž1996. 603–610. w47x P.C. Wong, C.A. Pardo, D.R. Borchelt, M.K. Lee, N.G. Copeland, N.A. Jenkins, S.S. Sisodia, D.W. Cleveland, D.L. Price, An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria, Neuron 14 Ž1995. 1105–1116. w48x Z. Xu, L.C. Cork, J.W. Griffin, D.W. Cleveland, Increased expression of neurofilament subunit NF-L produces morphological alterations that resemble the pathology of human motor neuron disease, Cell 73 Ž1993. 23–33. w49x B. Zhang, P. Tu, F. Abtahian, J.Q. Trojanowski, V.M.Y. Lee, Neurofilaments and orthograde transport are reduced in ventral root axons of transgenic mice that express human SOD1 with a G93A mutation, J. Cell Biol. 139 Ž1997. 1307–1315.
Research report
Selective impairment of fast anterograde axonal transport in the peripheral nerves of asymptomatic transgenic mice with a G93A mutant SOD1 gene Hitoshi Warita a , Yasuto Itoyama a , Koji Abe b
b, )
a Department of Neurology, Tohoku UniÕersity School of Medicine, Sendai, Japan Department of Neurology, Okayama UniÕersity Medical School 2-5-1 Shika-machi, Okayama 700-8558, Japan
Accepted 8 December 1998
Abstract Transgenic mice that express a mutant CurZn superoxide dismutase ŽSOD1. gene have been provided a valuable model for human amyotrophic lateral sclerosis ŽALS.. We studied a possible impairment of fast axonal transport in transgenic mice carrying a Gly 93 ™ Ala ŽG93A. mutant SOD1 gene found in human familial ALS ŽFALS.. Left sciatic nerve was ligated for 6 h in transgenic ŽTg. and age-matched wild-type ŽWT. mice. Immunohistochemical analyses were performed for accumulations of kinesin and cytoplasmic dynein on both sides of the ligation site. Clinical function and histology in the spinal cords, sciatic nerves and gastrocnemius muscles were also assessed. The mice were examined at an early asymptomatic stage Žaged 19 weeks. and a late stage Ž30 weeks. just before the development of the symptoms. WT mice showed an apparent increase in immunoreactivities for kinesin and cytoplasmic dynein at proximal and distal of the ligation, respectively. In contrast, the young Tg mice showed a selective decrease of kinesin accumulation in the proximal of the ligation. The mice were asymptomatic with a mild histological change only in muscles. The old Tg mice showed a marked reduction of the immunoreactivity for kinesin and cytoplasmic dynein on both sides of the ligation. They had a significant loss of spinal motor neurons, relatively small myelinated fiber densities of sciatic nerves, and severe muscular changes. These results provide direct evidence that the SOD1 mutation leads to impaired fast axonal transport, particularly in the anterograde direction at an early, asymptomatic stage preceding loss of spinal motor neurons and peripheral axons. This impairment may contribute to subsequent selective motor neuron death in the present model implicated for human FALS. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Axonal transport; SOD1; Transgenic mouse; Kinesin; Dynein; ALS
1. Introduction Amyotrophic lateral sclerosis ŽALS. is a progressive and fatal neurodegenerative disorder characterized by loss of motor neurons in the cortex, brainstem and spinal cord, which causes skeletal muscle wasting, weakness and ultiAbbreviations: AH, anterior horn of spinal cord; ALS, amyotrophic lateral sclerosis; ANOVA, one-way analysis of variance; D-LI, cytoplasmic dynein-like immunoreactivity; FALS, familial ALS; G93A, mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine; HE, hematoxylin and eosin; K-LI, kinesin-like immunoreactivity; MFD, myelinated fiber density; MND, motor neuron diseases; NF-H, neurofilament heavy subunit; NF-M, neurofilament intermediate subunit; NF-L, neurofilament light subunit; O, with older age; PB, phosphate buffer; PBS, phosphate-buffered saline; SALS, sporadic ALS; S.D., standard deviation; SOD1, CurZn superoxide dismutase; TB, toluidine blue; Tg, transgenic; Y, with younger age; WT, non-transgenic wild-type ) Corresponding author. Fax: q 81-86-235-7368; E-mail: [email protected]
mately complete paralysis. About 5 to 10% of all ALS patients are familial and inherited mainly with an autosomal dominant manner. Despite clinical and genetic heterogeneity, familial ALS ŽFALS. and sporadic ALS ŽSALS. have a similar pathology, suggesting a common pathogenesis. In approximately 15 to 20% of FALS cases, a variety of dominant, missense mutations and a small deletion in the CurZn superoxide dismutase ŽSOD1; EC 1.15.1.1. gene, have been identified w19,35,39x. Several lines of transgenic mouse models that express a mutant SOD1 gene have been established w23,37,47x. These transgenic mice develop phenotypes closely similar to human FALS in accordance with elevated levels of mutant SOD1 gene expression, although pathological findings in the central nervous system ŽCNS. are different from those of human FALS to some extent w15,16x. The SOD activities are normal or increased in these transgenic models and the mice with overexpression of wild-type SOD1 gene or a lower expression level of mutant SOD1 gene do not show
0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 1 3 5 1 - 1
H. Warita et al.r Brain Research 819 (1999) 120–131
ALS phenotypes. Knockout mice disrupted one or both alleles of SOD1 gene do not result in motor neuron degeneration w36x. Furthermore, no correlation is observed between the degree of reduced SOD activity and disease severity in human FALS with SOD1 mutations w6x. These observations support a hypothesis that mutant SOD1 protein has a novel toxic gain of function rather than loss of function. However, the mechanisms of how this ubiquitously expressed mutant protein leads to selective vulnerability of motor neurons are not clearly understood. Neurons are highly polarized and bidirectional axonal transport between somata and axon terminal is indispensable for their morphological and functional integrity. The axonal transport system is classified into two rate components: fast Ž100–400 mmrday. and slow Ž0.1–3 mmrday. components. The fast components consist of various membrane-bound organelles and synaptic membrane precursors, while the slow one includes cytoskeletal proteins and soluble enzymes w10,25,45x. Although the molecular mechanisms for slow axonal transport are not well characterized, recent studies revealed two types of microtubuleactivated ATPases, kinesin and cytoplasmic dynein, as fast axonal motors w9,32,34,44x. The anterograde motor kinesin primarily carries cargo toward the axon terminal w27,44x, whereas cytoplasmic dynein is passively associated with anterograde movement in an inactive form and then is actively involved in the retrograde transport toward the cell body from the axon terminal w28x. The aberrant accumulation of neurofilaments in perikarya and swollen proximal axons of spinal motor neurons is an early neuropathological hallmark in human ALS and other motor neuron diseases ŽMND. w18,24,42x. Overexpression of wild-type murine neurofilament light subunit ŽNF-L. w48x, human neurofilament heavy subunit ŽNF-H. gene w14x and a mutant NF-L gene w31x in transgenic mice exhibit perikaryal and axonal accumulation of neurofilaments leads to MND phenotypes. Although these data suggest a key role for these cytoskeletal proteins associated with reduced slow axonal transport in the pathogenesis of ALS w13,33,49x, NF-H mutations are identified
121
in few cases of human SALS w22x and not linked to human FALS cases w38,46x. Thus, it is likely that additional factors underlie the disease mechanisms. Besides these neurofilament pathology, impairment of fast axonal transport is also considered to be associated with the neuronal dysfunction in human ALS w12,41x. However, it has been controversial how the fast axonal transport is altered, and a possible change of fast axonal transport in peripheral nerves has not been well investigated in association with the FALS-linked SOD1 mutations w49x. To test whether the fast axonal components are modified and how they are altered developmentally from an early preclinical stage, we examined accumulations of kinesin and cytoplasmic dynein in ligated sciatic nerves of transgenic mice with FALS-linked mutant human SOD1, in addition to clinical and histological evaluation of spinal cord, peripheral nerves and skeletal muscles. 2. Materials and methods 2.1. Animal model Transgenic mice carrying a mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine ŽG93A. were used in this study. Their strain designation is B6SJL-TgN ŽSOD1-G93A. 1Gur dl , that were obtained from the Jackson Laboratory ŽBar Harbor, ME, USA. w23x. At around 30 weeks of age, the G93A transgenic mice developed progressive muscle weakness and spasticity in one or more limbs beginning with a posterior limb. One to two weeks later, they could not feed for themselves due to severe paralysis with hyperextended posture of hindlimbs. The G93A transgenic mice ŽTg. and age-matched non-transgenic wild-type B6SJL littermates ŽWT. were examined simultaneously. They were divided into four groups: Tg with younger age ŽTgrY; aged 18.6 " 0.5 weeks Žmean " S.D.., n s 5., Tg with older ŽTgrO; 30.2 " 1.1, n s 5., WT with younger ŽWTrY; 19.0 " 0.0, n s 4., and WT with older ŽWTrO; 30.2 " 1.1, n s 5. ŽTable 1.. Genomic DNA was extracted from tail
Table 1 Clinical settings, SOD activity and neurological evaluations
Number Sex Age Žw. Body weight Žg. SOD activity ŽUrmg. Circular cage rotation Žnumber. Transverse bar rotation Žnumber. Angle of slippage Ždegree of slope.
WTrY
WTrO
TgrY
TgrO
ns4 M s 2, F s 2 19.0 " 0.0 27.5 " 5.5 184.0 " 13.4 253.0 " 72.5 99.3 " 32.5 39.3 " 1.3
ns5 M s 2, F s 3 30.2 " 1.1 30.6 " 4.6 182.2 " 34.9 254.2 " 131.1 141.4 " 85.1 36.2 " 3.3
ns5 M s 4, F s 1 18.6 " 0.5 26.8 " 2.6 391.8 " 47.1a 262.2 " 127.6 86.6 " 22.1 36.2 " 3.3
ns5 M s 3, F s 2 30.2 " 1.1 28.8 " 5.9 236.8 " 16.3 b 100.4 " 82.7 86.0 " 64.7 34.6 " 4.2
Each value represents means" S.D. WT s non-transgenic wild-type mice, Tg s transgenic mice carrying mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine, Y s with younger age, O s with older age, M s Male, and F s Female. a p - 0.01 relative to the WTrY, b p - 0.05 to the TgrY group.
122
H. Warita et al.r Brain Research 819 (1999) 120–131
biopsy. The transgene of the G93A mutant human SOD1 was confirmed by specific primers for exon 4 of the gene w39x. The enzyme activity of SOD1 protein Žunits per milligram of protein, Urmg. was measured by our previ-
ous method w4x. In brief, hemoglobin was extracted from peripheral erythrocytes by ethanol and chloroform, and the aqueous phase was recovered for measurement of the SOD activity. For clinical and neurological evaluation of the
Fig. 1. Histology in lumber cord anterior horns and gastrocnemius muscles. High power view of HE-stained transverse sections of L4–L5 spinal anterior horns Ža, c, e and g. and gastrocnemius muscles Žb, d, f and h.. In contrast to normal anterior horns of WTrY, WTrO and TgrY mice Ža, c and e., moderate loss of motor neurons and mild gliosis in TgrO mice are noted Žg.. The gastrocnemius muscles of TgrY mice show small group atrophy Žf, arrows.. Whereas, the muscles of TgrO present severe changes such as small group atrophy Žh, arrows., centrally-placed nuclei Žh, arrowheads., chromatic nuclear clump Žh, an open arrow., small angulated fibers Žh, open arrowheads. and hypertrophic fibers Žh, asterisk.. Centrally-placed nuclei are observed up to 1% of muscle fibers of normal control mice Žd, arrowheads.. Magnification: =100. Scale bar: 100 mm.
H. Warita et al.r Brain Research 819 (1999) 120–131
four groups of mice, body weight, rolling number of circular cage, grasping on a rolling column, and angle of slipping down from woody slope were measured as described previously before the ligation of their peripheral nerves w2x. 2.2. Ligation of mouse peripheral nerÕes Under anesthesia with inhalation of a nitrous oxider oxygenrhalothane Ž68%:30%:2%. mixture, the skin of the left lower limb was incised. The left sciatic nerve was then ligated very tightly at mid-thigh level with surgical thread ŽProlene Blue Monofilament Ž6-0., Ethicon, NJ, USA. w28x, followed by suture of the skin to prevent infection. In addition, the skin of the contralateral side was incised and sutured as a sham-operation and the mice were kept in a restricted cage. Immediately after ligation of sciatic nerves, the animals presented severe paralysis of left lower limbs. After 6 h of the sciatic nerve ligation, the mice were decapitated under deep anesthesia with ether. Pieces of the left sciatic nerve both proximal and distal Žat least 7 mm of each. to the ligated portion, right sciatic nerves, lumber spinal cord and bilateral gastrocnemius muscles were quickly removed. The nerves and spinal cord were rapidly frozen in powdered dry ice and the muscles were in 2-methylbutane ŽWako, Osaka, Japan. with liquid nitrogen. The samples were stocked at y808C. During the experiment, the mice were treated in accordance with the declaration of Helsinki and the guiding principles in the care and use of animals. All experimental protocols and procedures were approved by the Animal Committee of the Tohoku University School of Medicine, Japan.
123
2.4. QuantitatiÕe morphology Spinal cord neuronal counts were performed manually by an examiner blinded to the mouse groups. For each mouse, at least two or more Nissl-stained sections were analyzed in the spinal segments from L4 through L5. For each section, the number of motor neurons in both sides of the anterior horns was measured under light microscopy at a 100 = magnification. The average number of motor neurons in each section was added and divided by the number of sections counted in order to provide the number of motor neurons in unilateral anterior horn per section. Only cell profiles containing a distinct nucleus with nucleolus were counted in order to avoid double counting. The entire sciatic nerve specimen with TB staining was measured using computerized software ŽNIH Image Ver. 1.61, National Institutes of Health, USA. in contiguous non-
2.3. Histological study Transverse sections of the lumber cord ŽL4–L5. and the muscle samples were cut Ž10 mm of thickness. on a cryostat at y208C. A set of the sections of the spinal cord and muscle was stained with hematoxylin and eosin ŽHE.. Another cross-sections of spinal cord were stained for Nissl substance using Cresyl violet for neuronal counting. Small pieces of the sciatic nerves away from the ligated portion were fixed with 2% glutaraldehyde in 0.1 M phosphate buffer ŽPB, pH 7.4. for 2 h at room temperature, followed by 1% osmium tetroxide in 0.1 M PB at 48C, dehydrated in alcohols and embedded in Epon; Epon 812 ŽOuken, Tokyo, Japan.r1-Dodecenylsuccinic Anhydride ŽDDSA; Ouken.rMethylnadic Anhydride ŽMethyl-5norbornene-2,3-dicarboxylic Anhydride ŽMNA; Nissin EM, Tokyo, Japan.r2,4,6-Tris Ždimethylaminomethyl. phenol ŽDMP-30; TAAB Laboratories Equipment, Reading, UK. in the ratio of 45.4%:25.2%:28.0%:1.4%. Transverse sections with 0.5 mm thickness were stained with toluidine blue ŽTB..
Fig. 2. ŽA. Number of motor neurons in the anterior horn of lumber cord. Neuronal counting with Nissl-stained specimen shows significant decrease in the number of TgrO mice Ž19.2"3.3, ns 5. compared to WTrY Ž28.0"3.4, ns 4, ) p- 0.01., WTrO Ž28.2"2.4, ns 5, ) p0.01. and TgrY mice Ž27.8"3.2, ns 5, ) p- 0.01.. In contrast, the number of TgrY mice is not reduced significantly. ŽB. Myelinated fiber density ŽMFD. of sciatic nerve. The sciatic nerve sections with TB staining were measured in order to provide MFD Žrmm2 .. Although TgrO mice show relatively smaller MFD Ž20,075"1,714.3, ns 5., there are no significant differences among the four category of mice.
124
H. Warita et al.r Brain Research 819 (1999) 120–131
overlapping field under microscope at a 200 = magnification. The myelinated axon caliber Žmm. and the myelinated fiber density Žnumber of myelinated fibers per mm2 , MFD. of each transverse section was calculated. Histogram of the
myelinated axon caliber for each mouse category was constructed with statistical analyses. The measurement of myelinated axons in a fixed area Ž8.8 = 10y2 mm2 . for each mouse was submitted for the histogram.
Fig. 3. Representative microphotographs of the sciatic nerves with TB staining Ža, c, e and g. and histograms of myelinated axon caliber Žb, d, f and h. are exhibited. No histological changes are shown in the sciatic nerves Ža, c, e and g.. The histogram of sciatic nerve axon caliber showed a similar pattern with a peak in the class of 2.0 to 3.0 mm in all the four groups of mice. The histogram shows a similar pattern with a peak in the class of 2.0 to 3.0 mm in all the four groups of mice Žb, d, f and h.. There are developmental differences in axon caliber classes of 2.0–3.0 Ž)) p - 0.01., 3.0–4.0 Ž)) p - 0.01., 4.0–5.0 Ž) p - 0.05., 5.0–6.0 Ž) p - 0.05. and 6.0–7.0 mm Ž)) p - 0.01. between WTrY and WTrO Žb and d. or TgrY and TgrO mice Žf and h.. However, no differences are presented between age-matched groups such as WTrY and TgrY Žb and f. or WTrO and TgrO Žd and h.. Original magnification in the left panels Ža, c, e and g.: 100 = , Scale bar: 100 mm.
H. Warita et al.r Brain Research 819 (1999) 120–131
2.5. Immunohistochemistry The longitudinal sections Ž10 mm of thickness. of the ligated sciatic nerves were carefully obtained on a cryostat at y228C, and stained with an antibody for kinesin and cytoplasmic dynein by employing an immunohistochemical technique according to our previous report w1x. Briefly, after the treatment with 10% normal horse serum in phosphate-buffered saline ŽPBS, pH 7.4. for 2 h to avoid non-specific stain, the slides were washed and incubated with the primary antibody in a buffer containing 10% normal horse serum and 0.3% Triton X-100 ŽWako. at 48C for 16 h. The primary antibodies used in the present study were as follows: mouse monoclonal anti-bovine brain kinesin heavy chain diluted at 1:500 ŽChemicon International, Temecula, CA, USA. and mouse monoclonal antibovine brain cytoplasmic dynein 74 kD intermediate chain ŽChemicon International. at 1:500. Endogenous peroxidase was blocked with 0.3% H 2 O 2 and 10% methanol for 20 min. The sections were washed and incubated with biotinylated anti-mouse IgG Ž1r200. in the buffer for 3 h, and then incubated with avidin–biotin–horseradish peroxidase complex solution using a kit ŽVECTASTAIN Elite ABC kit, Vector Laboratories, Burlingame, CA, USA. for 30 min. Finally, immunoreactivity was revealed with a minute exposure with 3,3X-diaminobenzidine tetrahydrochloride Ž0.5 mgrml in 100 mM Tris–HCl buffer ŽpH 7.2., Wako. and 0.02% H 2 O 2 . The degree of immunohistochemically reactive product was rated for semiquantitation as follows: null Žy., small number of reactive product or weak staining Ž"., moderate number of Žq., large number of Žqq ., and large number of densely reactive product Žqqq .. In addition, optical density of kinesin and cytoplasmic dynein immunoreactivity in the longitudinal sections of sciatic nerve was quantified in arbitrary units by using the NIH Image. For each mouse, at least five immunostained sections were measured to calculate the average density for a defined area Ž4.4 = 10y2 mm2 . just adjacent to the ligation site.
Service, Germany. and the null hypothesis was rejected at the 0.05 level.
3. Results 3.1. Clinical settings, SOD actiÕity, and neurological eÕaluation Clinical settings, the SOD activity, and neurological estimates are summarized in Table 1. All the G93A mutant SOD1 Tg mice ŽTgrY and TgrO. and WT littermates ŽWTrY and WTrO. were clinically free from any symptoms such as muscle weakness, reduced motility or abnormal posture. No significant differences are observed in body weight, number of circular cage rotation, transverse bar rotation and angle of slippage ŽTable 1.. Although the difference was not detected by one-way ANOVA, TgrO mice showed relatively smaller number of circular cage rotation Ž100.4 " 82.7, mean " S.D., n s 5. compared to TgrY group Ž262.2 " 127.6, n s 5.. The SOD activity of TgrY mice was approximately two folds higher Ž391.8 " 47.1 Urmg, n s 5. than WTrY group Ž184.0 " 13.4, n s 4, p - 0.01. ŽTable 1.. However, TgrO mice had approximately 40% lower SOD activity Ž236.8 " 16.3, n s 5. than that of TgrY mice Ž391.8 " 47.1, n s 5, p 0.05..
Table 2 Immunoreactivity for kinesin and cytoplasmic dynein Groups
Case no.
WTrY
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
WTrO
2.6. Statistics TgrY
Results are expressed as mean " standard deviation ŽS.D... Statistical analyses were performed using one-way analysis of variance ŽANOVA. among means of value with the category of mice as the independent factor. Multiple pair-wise comparisons between means were tested by Tukey–Kramer post-hoc test when ANOVA showed significant differences Ž p - 0.05.. In regard to the histograms of myelinated axon caliber, the frequency in each class, skewness and kurtosis were analyzed by one-way ANOVA as mentioned above. In addition, two-way ANOVA was used for testing the histograms with the different types of mice and classes of axon caliber as the independent factors. All statistical analyses were performed using computerized software ŽWinSTAT w Ver. 3.1, Scientific Software
125
TgrO
Kinesin
Dynein
Proximal
Distal
Proximal
Distal
qqq qqq qqq qqqq qq qqq qq qqq qq qq q q qq q q " " q "
q q q q q qq q q q q q q q " q " " " "
qq qq qq qqq qqq qq qq qq qq qq q qq qq q q q q q "
qqqq qqqq qqq qqqq qqqq qqq qqq qqq qqqq qqq qq qqq qqq qqq q q qq qq "
The degree of immunohistochemically reactive product was rated for semiquantitation as follows: null Žy., small number of reactive product or weak staining Ž"., moderate number of Žq., large number of Žqq., and large number of densely reactive product Žqqq.. WT s non-transgenic wild-type mice, Tg s transgenic mice carrying mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine, Ys with younger age, Os with older age.
126
H. Warita et al.r Brain Research 819 (1999) 120–131
ŽFig. 1g., while the other mice ŽWTrY, WTrO and TgrY. showed no histological change ŽFig. 1a,c,e.. Neuronal counting with Nissl-stained specimen revealed that
Fig. 4. Immunohistochemistry for kinesin on both sides just proximal Žleft panels; a, c, e and g. and distal Žright panels; b, d, f and h. to the ligated portion Žarrowheads. of sciatic nerves. The immunoreactivity was decreasing in order of WTrY and WTrO, TgrY, and TgrO Ža, c, e and g. mice on proximal side of the ligation. TgrY mice show apparently weaker K-LI on proximal side Že. than those of WTrY Ža. and WTrO Žc.. TgrO mice show only slight staining on both sides of the ligation Žg and h.. Original magnification: 100=, Scale bar: 100 mm.
3.2. Histological study In lumber spinal cord ŽL4–L5. anterior horns ŽAH., histological study with HE staining presented moderate loss of motor neurons and reactive gliosis in TgrO mice
Fig. 5. Immunohistochemistry for cytoplasmic dynein on both sides just proximal Žleft panels; a, c, e and g. and distal Žright panels; b, d, f and h. to the ligated site Žarrowheads. of sciatic nerves. In contrast to the K-LI, the reduction of the immunoreactivity is mild in TgrY mice Žf. on distal side of the ligation compared to those of WTrY Žb. and WTrO Žd. mice. Note the staining proximal to the ligated site of TgrY is weaker Že. than those of WTrY and WTrO Ža and c. as well as immunostaining of kinesin. Moreover, TgrO mice show obvious depletion of the immunoreactivity on both sides of the ligated site Žg and h.. Original magnification: 100=, Scale bar: 100 mm.
H. Warita et al.r Brain Research 819 (1999) 120–131
the number of motor neurons in AH of TgrO mice was significantly smaller Ž19.2 " 3.3, n s 5. than those of WTrY Ž28.0 " 3.4, n s 4, p - 0.01., WTrO Ž28.2 " 2.4, n s 5, p - 0.01., and TgrY mice Ž27.8 " 3.2, n s 5, p 0.01. ŽFig. 2A.. Whereas, the number of TgrY motor neurons in AH Ž27.8 " 3.2. did not show significant decrease compared to the other three groups. Otherwise, the sections stained with HE or Cresyl violet showed no apparent histological change or neuronal loss in spinal cord of the four groups of mice. The sciatic nerve sections with TB staining are depicted in Fig. 3a,c,e,g. Computerized measurement of MFD showed no significant differences, although TgrO mice had relatively smaller MFD Ž20,075 " 1,714.3rmm 2 , n s 5. compared to WTrO Ž22,104 " 1,671.9, n s 5. and TgrY Ž22,982 " 1,613.4, n s 5. mice ŽFig. 2B.. The histogram of sciatic nerve axon caliber showed a similar pattern with a peak in the class of 2.0 to 3.0 mm in all the four groups of mice. There were developmental differences in axon caliber classes of 2.0–3.0 Ž p - 0.01., 3.0–4.0
127
Ž p - 0.01., 4.0–5.0 Ž p - 0.05., 5.0–6.0 Ž p - 0.05., and 6.0–7.0 mm Ž p - 0.01. between WTrY and WTrO, or TgrY and TgrO mice ŽFig. 3b,d,f,h.. These differences were, however, not observed between age-matched groups as WTrY and TgrY, or WTrO and TgrO. Testing by two-way ANOVA also showed no significant differences in the frequency distribution of the myelinated axon caliber among the four groups. The skewness and kurtosis of the histograms did not show significant differences among the four groups Žone-way ANOVA, data not shown.. Bilateral gastrocnemius muscles of both WTrY and WTrO mice did not show any histological abnormalities ŽFig. 1b and d.. On the other hand, the muscles of TgrY mice are mildly involved in which small group atrophy was modestly observed ŽFig. 1f, arrows.. In contrast, TgrO muscles showed a severe histological change such as small group atrophy ŽFig. 1h, arrows., centrally-placed nuclei ŽFig. 1h, arrowheads., chromatic nuclear clump ŽFig. 1h, an open arrow., small angulated fibers ŽFig. 1h, open arrowheads. and hypertrophic fibers ŽFig. 1h, asterisk..
Fig. 6. Optical density for kinesin- and cytoplasmic dynein-like immunoreactivity ŽK-LI and D-LI. on both sides of the ligated site of sciatic nerve sections. In asymptomatic TgrY mice, the densities of both K-LI and D-LI are significantly decreased only on proximal side of the ligated site compared to those of WTrY and WTrO Ž p - 0.01.. In contrast, TgrO mice show significantly reduced densities of K-LI and D-LI on both sides of the ligation compared to those of WTrY and WTrO Ž p - 0.01..
128
H. Warita et al.r Brain Research 819 (1999) 120–131
3.3. Immunohistochemical comparison of the kinesin and cytoplasmic dynein accumulations Immunohistochemistry of both just proximal and distal to the ligated site of sciatic nerves with anti-kinesin and anti-cytoplasmic dynein antibodies are summarized with semiquantitation in Table 2 and Fig. 6. In all the cases, kinesin-like immunoreactivity ŽK-LI. and cytoplasmic dynein-like one ŽD-LI. in the sciatic nerves was observed on both sides of the ligated portion ŽFigs. 4–6.. While K-LI was predominant on the proximal side, D-LI was denser on the distal side ŽFigs. 4–6.. Staining was restricted to both sides of the ligation and presented a gradational pattern, in which the immunoreactivity was more intense and larger number as the region was closer to the ligated portion. The degree of K-LI proximal to the ligature Žproximal K-LI. was clearly different among the four groups of mice. It was decreasing in order of WTrY, WTrO, TgrY and TgrO ŽTable 2; Fig. 4a,c,e,g and Fig. 6.. The proximal K-LI of TgrY was apparently weaker Žq; qq; optical densitys 2.51 " 0.49, p - 0.01, one-way ANOVA. than those of WTrY Žqqq; qqqq; 3.82 " 0.61. and WTrO Žqq ; qqq; 3.77 " 0.65. ŽTable 2; Fig. 4a,c,e,g and Fig. 6.. TgrO had only slight K-LI on both proximal Ž"; q; 0.53 " 0.09. and distal Ž"; q; 0.26 " 0.10. to the ligated site ŽTable 2; Fig. 4g,h and Fig. 6.. The K-LI distal to the ligation Ždistal K-LI. was generally weak and not clearly different among WTrY, WTrO and TgrY, although optical density in TgrO mice showed significantly weak K-LI Ž"; q; 0.26 " 0.10, p - 0.01. ŽTable 2; Fig. 4b,d,f and Fig. 6.. In contrast to K-LI, the reduction of distal D-LI was mild in TgrY mice Žqq ; qqq; 3.98 " 0.28. compared to WTrY Žqqq; qqqq; 4.17 " 0.34. and WTrO mice Žqqq; qqqq; 4.08 " 0.33. ŽTable 2; Fig. 5b,d,f and Fig. 6.. As well as the proximal K-LI, the proximal D-LI of TgrY was also weaker Žq; qq; 0.99 " 0.22, p - 0.01. than those of WTrY and WTrO Žqq ; qqq; 1.71 " 0.22 and 1.40 " 0.22. ŽTable 2; Fig. 5a,c,e and Fig. 6.. TgrO mice showed weak D-LI Ž"; qq; proximal, 0.44 " 0.15, p - 0.01 and distal, 2.86 " 0.30, p - 0.01. on both sides of the ligation ŽTable 2; Fig. 5g,h and Fig. 6.. When the primary antibody was omitted or replaced with normal IgG, neither K-LI nor D-LI was detected, and every specimen of contralateral non-ligated sciatic nerves showed only slight number of or faint staining Ždata not shown..
4. Discussion Although body weight and clinical scores reflecting motility, motor activity and limb muscle strength were not significantly different among the four groups of mice ŽTable 1., histological data in muscle ŽFig. 1b,d,f,h. and circular cage score ŽTable 1. suggest that TgrY mice were
at an early, asymptomatic stage and TgrO mice were just before the onset of the disease. These clinical observations of the G93A transgenic mice were different from the previous report in the age of onset. It was about 3 months later in the present Tg mice compared to the original G93A transgenic line w15,23x. Histopathological study revealed that TgrY mice showed only mild denervated change in gastrocnemius muscles without loss of spinal motor neurons in lumber cord ŽL4–L5., while TgrO showed approximately 30% loss Žcompared to WTrO. of spinal motor neurons with severe neurogenic muscular changes ŽFig. 1e–h and Fig. 2A.. Although TgrO mice had relatively smaller MFD ŽFig. 2B and Fig. 3g., TgrY mice were exempt from apparent histological changes of sciatic nerves ŽFig. 3e.. These observations suggest that histological alterations begin with muscular changes followed by the spinal cord and peripheral nerve pathology in the present model. In addition, the developmental change between the younger and older WT mice was shown not in the spinal cord and skeletal muscles but in the size distribution of myelinated axons of peripheral nerves ŽFig. 3.. Although centrally-placed nuclei are found in a small fraction of normal muscles Žup to 1% in adult human, also found similarly in the present WT mice, Fig. 1d, an arrowhead., it is generally interpreted as a myogenic change with regeneration of muscle fibers. Thus, the observation that the present G93A transgenic mice had increasing number of centrally-placed nuclei in common Ž10–20% in TgrO, Fig. 1h, arrowheads. suggests a distinct process different from a simple denervation. In the previous report of the G93A transgenic mice, the vacuolar change of spinal motor neurons was observed in the latest stage w16x. In lumber cord of the present G93A transgenic mice, however, any vacuolar change was not obviously found under light microscopy ŽFig. 1e and g.. The elevation of SOD activity in the blood samples was smaller in the present Tg mice Žapproximately 2.1 times higher in TgrY than that of WTrY mice, Table 1. than the previous report Žfour folds in brain. w23x. Therefore, the later onset and the lack of vacuolar change in the present mouse model may be related to the relatively mild level of overexpression of the mutant SOD1 gene probably because of decreased copy number with repeated mating. Indeed, the transgenic lines of lower copy number of the G93A mutant SOD1 gene have been reported to show delayed onset and fewer vacuoles with more neurofilamentous inclusions in spinal motor neurons at later stage, that resemble more closely to human ALS pathology w17,49x. The 6-h ligation of sciatic nerve resulted in accumulative expression of fast axonal motors, kinesin and cytoplasmic dynein on both sides of the ligated portion. The degree of proximal K-LI and distal D-LI to the ligation may represent the amount or the rates of anterograde and retrograde fast axonal traffic, respectively. The TgrY mice showed a marked decrease of proximal K-LI with only a minimal decline of distal D-LI in the ligated sciatic nerves
H. Warita et al.r Brain Research 819 (1999) 120–131
ŽTable 2; Fig. 4e, Fig. 5f and Fig. 6.. These observations may reflect a possible impairment of fast axonal transport particularly in anterograde direction at an early, asymptomatic stage of the disease. In TgrY mice, not only the K-LI but also the D-LI was reduced on proximal side of the ligated site, although both K-LI and D-LI on distal side of the ligated site were not significantly decreased compared to WTrY mice ŽFig. 6.. These data support a defective fast anterograde transport machinery rather than declines in protein levels of these motors in the transgenic mice. Marked depletion of K-LI and D-LI in TgrO mice suggests more severe dysfunction of bidirectional fast axonal transport ŽFigs. 4–6.. This is the first report that suggests an early but selective impairment of fast anterograde axonal transport in the peripheral nerves of FALSlinked SOD1 transgenic mice. The weak immunostaining of proximal K-LI and distal D-LI in WTrO mice ŽFig. 4c, Fig. 5d and Fig. 6. compared to WTrY ŽFig. 4a, Fig. 5b and Fig. 6. suggests an association between reduced fast axonal transport and aging process. Since the number of motor neurons and MFD of sciatic nerves were not reduced and histological change was restricted to the skeletal muscles in TgrY mice ŽFig. 1f., it is suggested the selective reduction of K-LI was not caused by loss of spinal motor neurons or peripheral nerve lesion. The present data suggest that a functional impairment of selective anterograde fast axonal transport may precede motor neuron loss and axonal degeneration. The sciatic nerve contains not only motor but sensory axons. No histological changes were, however, observed in posterior horns of spinal cord even in TgrO mice that had severely reduced K-LI and D-LI accompanied with significant loss of spinal motor neurons. Thus, the reduction of K-LI and D-LI in the sciatic nerves seems to represent their ‘motor’ axonal dysfunction. There is no certain methods to distinguish motor from sensory axons by morphologic or immunocytochemical determination in peripheral nerves at this moment. Further study to discriminatively detect a change in motor and sensory nerves may elucidate the mechanism more clearly. Recent studies revealed an association of kinesin and cytoplasmic dynein superfamily proteins with a various membranous organelles w26x. Fast anterograde traffic involves precursors of synaptic vesicles, plasma membrane constituents and mitochondria, while fast retrograde traffic concerns prelysosomal vesicles, growth factors, mitochondria and recycled proteins from anterograde transport. Thus, a disruption of fast anterograde transport could cause a reduced action potential propagation, defective neurotransmitter release and a metabolic dysfunction at the nerve terminals. In motor neurons, therefore, impaired fast anterograde transport would primarily result in neuromuscular dysfunction. In fact, an electromyographic impairment of neuromuscular function was recently observed precedent to the beginning of clinical signs and the spinal motor neuron loss in the G93A mutant SOD1 transgenic
129
mice w30x. The precedence of neurogenic muscular change in the present TgrY mice is compatible with the reported data. Although previous works of fast axonal transport in human ALS have shown controversial results w7,11,12x, more recent studies have suggested reduction of fast axonal transport in the proximal axons of spinal motor neurons in human ALS w41x and also in the transgenic mice carrying the G93A mutant SOD1 w49x. The latter report showed an impairment of both slow and fast anterograde transport in the ventral root of the transgenic mice that express a lower copy number of G93A mutant SOD1 similar to the present mice. In the report, slow components including neurofilaments and tubulin decline as early as 26 weeks of age without loss of spinal motor neurons and ventral root axons before the retardation of a part of fast anterograde components at the age presenting clinical symptoms Žabout 29 weeks of age. w49x. By contrast, the present results suggest an much earlier alteration Žabout 19 weeks of age. of the fast anterograde transport in the peripheral nerve. The pathological findings reflecting an excessive accumulation of neurofilaments are reported in human FALS with SOD1 mutations w40x and in the transgenic mice harboring the G93A mutant SOD1 gene w15,16,23,43x. Mutant SOD1 has been proposed to allow increased access of peroxynitrite and result in nitration of tyrosine residues on proteins w5x. Recently, increased nitration of protein– tyrosine residue was reported in spinal motor neurons of the human ALS w3x and in spinal cord and cerebral cortex of transgenic mice with the G93A mutation w21x. Since human NF-L is rich in tyrosine residue, the tyrosine nitration may lead to neurofilament disruption, suggesting an association between SOD1 mutations and the neurofilament pathology. In the masses of accumulated neurofilaments in neuronal perikarya and swollen axons, membrane-bound cytoplasmic organelles were also accumulated. It has been suggested, therefore, that a secondary inhibition of fast axonal transport may also contribute to the MND phenotypes w8,13,49x associated with neurofilament pathology and also with SOD1 mutations. The recent work has suggested that axonal neurofilaments does not play essential role in axonal degeneration mediated by Gly 37 ™ Arg ŽG37R. mutant SOD1 w20x. These reported data and the present results raise a possibility that the mutations in SOD1 gene directly or indirectly lead to impaired fast axonal transport in the peripheral nerves. Furthermore, mutations of the kinesin heavy chain gene in Drosophila cause a specific disruption of fast axonal transport leading to progressive distal paralysis w29x, suggesting an inhibition of fast axonal transport causes a MND phenotype by itself. Acknowledgements The authors would like to thank Mr. Y. Onodera, A. Ito and Ms. M. Takahashi for technical assistance and are
130
H. Warita et al.r Brain Research 819 (1999) 120–131
grateful to Drs. T. Hayashi, H. Kitagawa, M. Nagai, I. Nagano, M. Tateyama and M. Sakurai for helpful discussions and generous support. This work was partly supported by Grant-in-Aid for Scientific Research ŽB. 09470151 from the Ministry of Education, Science and Culture of Japan, and by grants ŽTashiro K and Itoyama Y. from the Ministry of Health and Welfare of Japan.
w18x w19x
w20x
References w21x w1x K. Abe, M. Aoki, J. Kawagoe, T. Yoshida, A. Hattori, K. Kogure, Y. Itoyama, Ischemic delayed neuronal death. A mitochondrial hypothesis, Stroke 26 Ž1995. 1478–1489. w2x K. Abe, S. Morita, T. Kikuchi, Y. Itoyama, Protective effect of a novel free radical scavenger, OPC-14117, on wobbler mouse motor neuron disease, J. Neurosci. Res. 48 Ž1997. 63–70. w3x K. Abe, L.H. Pan, M. Watanabe, T. Kato, Y. Itoyama, Induction of nitrotyrosine-like immunoreactivity in the lower motor neuron of amyotrophic lateral sclerosis, Neurosci. Lett. 199 Ž1995. 152–154. w4x M. Aoki, M. Ogasawara, Y. Matsubara, K. Narisawa, S. Nakamura, Y. Itoyama, K. Abe, Familial amyotrophic lateral sclerosis ŽALS. in Japan associated with H46R mutation in CurZn superoxide dismutase gene: a possible new subtype of familial ALS, J. Neurol. Sci. 126 Ž1994. 77–83. w5x J.S. Beckman, M. Carson, C.D. Smith, W.H. Koppenol, ALS, SOD and peroxynitrite, Nature 364 Ž1993. 584. w6x A.C. Bowling, E.E. Barkowski, D. McKenna-Yasek, P. Sapp, H.R. Horvitz, M.F. Beal, R.H. Brown Jr., Superoxide dismutase concentration and activity in familial amyotrophic lateral sclerosis, J. Neurochem. 64 Ž1995. 2366–2369. w7x W.G. Bradley, P. Good, C.G. Rasool, L.S. Adelman, Morphometric and biochemical studies of peripheral nerves in amyotrophic lateral sclerosis, Ann. Neurol. 14 Ž1983. 267–277. w8x S. Brady, Motor-neuron diseases. Interfering with the runners, Nature 375 Ž1995. 12–13. w9x S.T. Brady, A novel brain ATPase with properties expected for the fast axonal transport motor, Nature 317 Ž1985. 73–75. w10x S.T. Brady, Molecular motors in the nervous system, Neuron 7 Ž1991. 521–533. w11x A.C. Breuer, M.B. Atkinson, Fast axonal transport alterations in amyotrophic lateral sclerosis ŽALS. and in parathyroid hormone ŽPTH.-treated axons, Cell Motil. Cytoskeleton 10 Ž1988. 321–330. w12x A.C. Breuer, M.P. Lynn, M.B. Atkinson, S.M. Chou, A.J. Wilbourn, K.E. Marks, J.E. Culver, E.J. Fleegler, Fast axonal transport in amyotrophic lateral sclerosis: an intra-axonal organelle traffic analysis, Neurology 37 Ž1987. 738–748. w13x J.F. Collard, F. Cote, ˆ ´ J.P. Julien, Defective axonal transport in a transgenic mouse model of amyotrophic lateral sclerosis, Nature 375 Ž1995. 61–64. w14x F. Cote, ˆ ´ J.F. Collard, J.P. Julien, Progressive neuronopathy in transgenic mice expressing the human neurofilament heavy gene: a mouse model of amyotrophic lateral sclerosis, Cell 73 Ž1993. 35–46. w15x M.C. Dal Canto, M.E. Gurney, Development of central nervous system pathology in a murine transgenic model of human amyotrophic lateral sclerosis, Am. J. Pathol. 145 Ž1994. 1271–1279. w16x M.C. Dal Canto, M.E. Gurney, Neuropathological changes in two lines of mice carrying a transgene for mutant human Cu,Zn SOD, and in mice overexpressing wild type human SOD: a model of familial amyotrophic lateral sclerosis ŽFALS., Brain Res. 676 Ž1995. 25–40. w17x M.C. Dal Canto, M.E. Gurney, A low expressor line of transgenic mice carrying a mutant human Cu,Zn superoxide dismutase ŽSOD1. gene develops pathological changes that most closely resemble those
w22x
w23x
w24x
w25x
w26x w27x
w28x
w29x
w30x
w31x
w32x
w33x
w34x
w35x
in human amyotrophic lateral sclerosis, Acta Neuropathol. ŽBerlin. 93 Ž1997. 537–550. M.B. Delisle, S. Carpenter, Neurofibrillary axonal swellings and amyotrophic lateral sclerosis, J. Neurol. Sci. 63 Ž1984. 241–250. H.X. Deng, A. Hentati, J.A. Tainer, Z. Iqbal, A. Cayabyab, W.Y. Hung, E.D. Getzoff, P. Hu, B. Herzfeldt, R.P. Roos, C. Warner, G. Deng, E. Soriano, C. Smyth, H.E. Parge, A. Ahmed, A.D. Roses, R.A. Halliwell, M.A. Periak-Vance, T. Siddique, Amyotrophic lateral sclerosis and structural defects in Cu, Zn superoxide dismutase, Science 261 Ž1993. 1047–1051. J. Eyer, D.W. Cleveland, P.C. Wong, A.C. Peterson, Pathogenesis of two axonopathies does not require axonal neurofilaments, Nature 391 Ž1998. 584–587. R.J. Ferrante, L.A. Shinobu, J.B. Schulz, R.T. Matthews, C.E. Thomas, N.W. Kowall, M.E. Gurney, M.F. Beal, Increased 3-nitrotyrosine and oxidative damage in mice with a human copperrzinc superoxide dismutase mutation, Ann. Neurol. 42 Ž1997. 326–334. D.A. Figlewicz, A. Krizus, M.G. Martinoli, V. Meininger, M. Dib, G.A. Rouleau, J.P. Julien, Variants of the heavy neurofilament subunit are associated with the development of amyotrophic lateral sclerosis, Hum. Mol. Genet. 3 Ž1994. 1757–1761. M.E. Gurney, H. Pu, A.Y. Chiu, M.C. Dal Canto, C.Y. Polchow, D.D. Alexander, J. Caliendo, A. Hentati, Y.W. Kwon, H.X. Deng, W. Chen, P. Zhai, R.L. Sufit, T. Siddique, Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation, Science 264 Ž1994. 1772–1775. A. Hirano, H. Donnenfeld, S. Sasaki, I. Nakano, Fine structural observations of neurofilamentous changes in amyotrophic lateral sclerosis, J. Neuropathol. Exp. Neurol. 43 Ž1984. 461–470. N. Hirokawa, The molecular mechanism of organelle transport along microtubules: the identification and characterization of KIFs Žkinesin superfamily proteins., Cell Struct. Funct. 21 Ž1996. 357–367. N. Hirokawa, Kinesin and dynein superfamily proteins and the mechanism of organelle transport, Science 279 Ž1998. 519–526. N. Hirokawa, R. Sato-Yoshitake, N. Kobayashi, K.K. Pfister, G.S. Bloom, S.T. Brady, Kinesin associates with anterogradely transported membranous organelles in vivo, J. Cell Biol. 114 Ž1991. 295–302. N. Hirokawa, R. Sato-Yoshitake, T. Yoshida, T. Kawashima, Brain dynein ŽMAP1C. localizes on both anterogradely and retrogradely transported membranous organelles in vivo, J. Cell Biol. 111 Ž1990. 1027–1037. D.D. Hurd, W.M. Saxton, Kinesin mutations cause motor neuron disease phenotypes by disrupting fast axonal transport in Drosophila, Genetics 144 Ž1996. 1075–1085. P.F. Kennel, F. Finiels, F. Revah, J. Mallet, Neuromuscular function impairment is not caused by motor neurone loss in FALS mice: an electromyographic study, Neuroreport 7 Ž1996. 1427–1431. M.K. Lee, J.R. Marszalek, D.W. Cleveland, A mutant neurofilament subunit causes massive, selective motor neuron death: implications for the pathogenesis of human motor neuron disease, Neuron 13 Ž1994. 975–988. R.J. Lye, M.E. Porter, J.M. Scholey, J.R. McIntosh, Identification of a microtubule-based cytoplasmic motor in the nematode C. elegans, Cell 51 Ž1987. 309–318. J.R. Marszalek, T.L. Williamson, M.K. Lee, Z. Xu, P.N. Hoffman, M.W. Becher, T.O. Crawford, D.W. Cleveland, Neurofilament subunit NF-H modulates axonal diameter by selectively slowing neurofilament transport, J. Cell Biol. 135 Ž1996. 711–724. B.M. Paschal, R.B. Vallee, Retrograde transport by the microtubule-associated protein MAP 1C, Nature 330 Ž1987. 181– 183. A. Pramatarova, D.A. Figlewicz, A. Krizus, F.Y. Han, I. CeballosPicot, A. Nicole, M. Dib, V. Meininger, R.H. Brown, G.A. Rouleau, Identification of new mutations in the CurZn superoxide dismutase gene of patients with familial amyotrophic lateral sclerosis, Am. J. Hum. Genet. 56 Ž1995. 592–596.
H. Warita et al.r Brain Research 819 (1999) 120–131 w36x A.G. Reaume, J.L. Elliott, E.K. Hoffman, N.W. Kowall, R.J. Ferrante, D.F. Siwek, H.M. Wilcox, D.G. Flood, M.F. Beal, R.H. Brown Jr., R.W. Scott, W.D. Snider, Motor neurons in CurZn superoxide dismutase-deficient mice develop normally but exhibit enhanced cell death after axonal injury, Nat. Genet. 13 Ž1996. 43–47. w37x M.E. Ripps, G.W. Huntley, P.R. Hof, J.H. Morrison, J.W. Gordon, Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis, Proc. Natl. Acad. Sci. U.S.A. 92 Ž1995. 689–693. w38x K. Rooke, D.A. Figlewicz, F.Y. Han, G.A. Rouleau, Analysis of the KSP repeat of the neurofilament heavy subunit in familial amyotrophic lateral sclerosis, Neurology 46 Ž1996. 789–790. w39x D.R. Rosen, T. Siddique, D. Patterson, D.A. Figlewicz, P. Sapp, A. Hentati, D. Donaldson, J. Goto, J.P. O’Regan, H.X. Deng, Mutations in CurZn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis, Nature 364 Ž1993. 59–62. w40x G.A. Rouleau, A.W. Clark, K. Rooke, A. Pramatarova, A. Krizus, O. Suchowersky, J.P. Julien, D. Figlewicz, SOD1 mutation is associated with accumulation of neurofilaments in amyotrophic lateral sclerosis, Ann. Neurol. 39 Ž1996. 128–131. w41x S. Sasaki, M. Iwata, Impairment of fast axonal transport in the proximal axons of anterior horn neurons in amyotrophic lateral sclerosis, Neurology 47 Ž1996. 535–540. w42x M.L. Schmidt, M.J. Carden, V.M. Lee, J.Q. Trojanowski, Phosphate dependent and independent neurofilament epitopes in the axonal swellings of patients with motor neuron disease and controls, Lab. Invest. 56 Ž1987. 282–294.
131
w43x P.H. Tu, P. Raju, K.A. Robinson, M.E. Gurney, J.Q. Trojanowski, V.M. Lee, Transgenic mice carrying a human mutant superoxide dismutase transgene develop neuronal cytoskeletal pathology resembling human amyotrophic lateral sclerosis lesions, Proc. Natl. Acad. Sci. U.S.A. 93 Ž1996. 3155–3160. w44x R.D. Vale, T.S. Reese, M.P. Sheetz, Identification of a novel force-generating protein, kinesin, involved in microtubule-based motility, Cell 42 Ž1985. 39–50. w45x R.B. Vallee, G.S. Bloom, Mechanisms of fast and slow axonal transport, Annu. Rev. Neurosci. 14 Ž1991. 59–92. w46x J.D. Vechio, L.I. Bruijn, Z. Xu, R.H. Brown Jr., D.W. Cleveland, Sequence variants in human neurofilament proteins: absence of linkage to familial amyotrophic lateral sclerosis, Ann. Neurol. 40 Ž1996. 603–610. w47x P.C. Wong, C.A. Pardo, D.R. Borchelt, M.K. Lee, N.G. Copeland, N.A. Jenkins, S.S. Sisodia, D.W. Cleveland, D.L. Price, An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria, Neuron 14 Ž1995. 1105–1116. w48x Z. Xu, L.C. Cork, J.W. Griffin, D.W. Cleveland, Increased expression of neurofilament subunit NF-L produces morphological alterations that resemble the pathology of human motor neuron disease, Cell 73 Ž1993. 23–33. w49x B. Zhang, P. Tu, F. Abtahian, J.Q. Trojanowski, V.M.Y. Lee, Neurofilaments and orthograde transport are reduced in ventral root axons of transgenic mice that express human SOD1 with a G93A mutation, J. Cell Biol. 139 Ž1997. 1307–1315.