- Email: [email protected]
Expression of adenovirus-mediated E. coli lacZ gene in skeletal muscles and spinal motor neurons of transgenic mice with a mutant superoxide dismutase gene
Neuroscience Letters 246 (1998) 153–156
Expression of adenovirus-mediated E. coli lacZ gene in skeletal muscles and spinal motor neurons of transgenic mice with a mutant superoxide dismutase gene Hitoshi Warita a, Koji Abe a ,*, Yasuhiro Setoguchi b, Yasuto Itoyama a a Department of Neurology, Tohoku University School of Medicine, Sendai, Japan Department of Respiratory Medicine, Juntendo University School of Medicine, Tokyo, Japan
b
Received 2 March 1998; received in revised form 13 March 1998; accepted 13 March 1998
Abstract A replication-defective recombinant adenoviral vector containing E. coli lacZ gene was injected into the right biceps brachii muscles of transgenic mice carrying mutant human Cu/Zn superoxide dismutase (SOD1) gene and non-transgenic wild-type mice at 27 weeks of age. Although the transgenic mice showed remarkable neurogenic muscular changes and a marked motor neuron loss in the anterior horn of spinal cord, the lacZ gene was widely expressed in all the injected muscles of transgenic mice as well as of wild-type mice at 7 days after the injection. In one transgenic and two wild-type mice, the lacZ gene expression was first detected in a few motor neurons of right lower cervical cord (C5–C6). These results demonstrate that an adenovirusmediated foreign gene is transferred and expressed in skeletal muscles both of normal and transgenic mice model for familial amyotrophic lateral sclerosis (FALS), and also, in the spinal motor neurons, may be transferred by retrograde transport from innervated muscles. 1998 Elsevier Science Ireland Ltd.
Keywords: Gene transfer; lacZ gene; Familial amyotrophic lateral sclerosis; Superoxide dismutase; Transgenic mouse
Amyotrophic lateral sclerosis (ALS) causes progressive skeletal muscle weakness and ultimately complete paralysis due to loss of motor neurons. There is currently no effective therapy to prevent or cure. About 5–10% of ALS cases is familial (FALS), and approximately 20% of FALS kindreds carry missense mutations in the Cu/Zn superoxide dismutase gene (SOD1; EC 1.15.1.1). Several lines of transgenic mice that express a mutant human gene encoding SOD1 develop phenotypes closely similar to those of FALS, providing a valuable model for ALS [9,10,13] in order to establish a curative therapy for ALS patients. The primary target of therapy is motor neurons in cerebral cortex, brainstem, and spinal cord. Viral vectors including adenoviruses are attractive candidates for direct gene delivery in vivo. Replication-deficient recombinant adenoviral vectors are capable
* Corresponding author. 1-1 Seiryo-machi, Aoba-ku, Sendai 9808574, Japan. Tel.: +81 22 7177189; fax: +81 22 7177192; e-mail: [email protected]
of infecting not only immature or dividing cell types but a variety of terminally differentiated cell types such as neurons and myocytes. In the present study, a replication-defective adenoviral vector containing the E. coli lacZ gene was injected into the skeletal muscles of symptomatic transgenic mice with a mutant human SOD1 and of non-transgenic wild-type (WT) mice, and the expression of lacZ gene was examined in skeletal muscles and spinal motor neurons. Transgenic mice expressing mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine (G93A) are obtained from the Jackson Laboratory (Bar Harbor, ME, USA) [9]. Their strain designation is B6SJL-TgN (SOD1-G93A) 1Gurdl. Symptomatic G93A transgenic mice (n = 4) at the age of 27 weeks and age-matched WT littermates (SJL/J; n = 4) were used in this study. The transgene of G93A mutant human SOD1 was confirmed by specific primers for exon 4 of the SOD1 gene [11]. The enzyme activity of SOD1 protein was measured by our previous method [4]. Clinical assessments of both G93A and WT mice were also measured before the
0304-3940/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0304- 3940(98) 00245- 6
154
H. Warita et al. / Neuroscience Letters 246 (1998) 153–156
Table 1
LacZ gene expression in muscles and spinal cords G93A Case no. Muscle Spinal cord
1 Biceps Triceps Right Left
+ − − −
WT 2
3
2+ − + −
2+ − − −
4 + − − −
5 + − − −
6 + − − −
7
8
2+ − + −
2+ − + −
G93A, transgenic mice carrying the G93A mutant human SOD1 gene; WT, non-transgenic wild-type mice. The degree of staining was categorized into three grades: minus, no staining; + , small (1–10); 2 + , large (10–100) number of stained cells, respectively.
vector injection as previously reported [1]. Experimental protocols and procedures were approved by the Animal Committee of the Tohoku University School of Medicine, Japan. The same recombinant adenoviral vector was used in this study as in our previous reports [2,3,12], which expresses the E. coli lacZ gene under the control of the human cytomegalovirus (CMV) promoter with the nuclear localizing signal of simian virus 40 (SV40). Under anesthesia with inhalation of a nitrous oxide/oxygen/halothane (68%:30%:2%) mixture, the skin of the right arm was incised, and Ad-CMVn-lacZ (2.5 × 109 plaque forming units (p.f.u.) in 5 ml of vector vehicle consisting of 10 mM Tris–HCl (pH 7.4), 1 mM MgCl2, and 10% glycerol) was slowly injected into the right biceps brachii muscle via a 10 ml sterile Hamilton syringe (the Gastight Highperformance Syringe; Hamilton, Reno, NV, USA), followed by suture of the skin to prevent infection. For histological evaluation of in vivo gene transfer and expression, the animals were decapitated at 7 days after the administration of AdCMVn-lacZ [7,8]. The bilateral biceps and triceps muscles and cervical cords were quickly removed and frozen in powdered dry ice. Transverse sections of 10 mm thickness were cut on a cryostat at −20°C. A set of the sections was stained with hematoxylin and eosin (HE). Other sections were reacted with X-Gal (5-Bromo-4-chloro-3-indolyl-bd-galactopyranoside; Boehringer Mannheim Biochemica, Mannheim, Germany) [2,3]. After the reaction with XGal, the sections were counterstained with hematoxylin. The number of blue color-stained cells was counted in sections of skeletal muscle and cervical cord. The degree of XGal staining was categorized into three grades for semiquantitation as follows: no staining, −; a small (1–10), +; or a large (10–100), 2+ number of stained cells, respectively. A set of spinal cord sections was immunohistochemically stained with a rabbit polyclonal antibody against bovine glial fibrillary acidic protein (GFAP; 1/100, Dako; Carpinter, CA, USA) after the reaction with X-Gal [3]. The G93A group at 27 weeks of age demonstrates a significant decrease of the body weight (26.8 ± 1.9 g, n = 4, mean ± SD, P , 0.01, Student’s t-test) as compared to WT mice (33.8 ± 1.1, n = 4) that showed no clinical symptoms. Rolling number of circular cage (16.0 ± 13.3, P , 0.01, Welch’s test), transverse bar rotations (17.8 ± 12.0, P ,
0.05, t-test), and angle of slope holding (degree of slope, 31.5 ± 8.4, P , 0.01, t-test) were all greatly reduced in the transgenic mice compared to WT group (128.5 ± 55.5, 108.3 ± 22.1, 47.8 ± 5.9, respectively). The SOD activity in peripheral erythrocyte lysate was higher in the transgenic mice (241.4 ± 14.7 unit/mg protein, P , 0.01, t-test) than that in WT mice (172.8 ± 32.1). As compared to normal findings in the WT group (Fig. 1a), histological study showed a diffuse neurogenic muscular atrophy with small group atrophy and small angular fibers in the biceps and triceps brachii muscles of the G93A group (Fig. 1b, arrows). Although central nuclei were barely found in WT mice (,1%, Fig. 1a), they were observed in 10–20% of muscle fibers in the G93A mice (Fig. 1b, arrowheads). In contrast to normal finding in the WT group (Fig. 2a,c), examination of the spinal cord revealed a marked loss of motor neurons and mild gliosis in the anterior horn of G93A mice (Fig. 2b,d). The lacZ gene expression in the muscles and spinal cords are summarized in Table 1, and the examples in Figs. 1c,d and 2e–h. The adenoviral vector was transferred, and the lacZ gene was expressed in the skeletal muscles and the spinal cords. In biceps brachii muscles, all cases in both G93A and WT groups expressed the lacZ gene with nuclear and perinuclear cytoplasmic pattern in 5–100 of muscle fibers per transverse section. In muscles of the transgenic mice, both atrophic and preserved muscle fibers expressed the lacZ gene (Fig. 1c,d). The lacZ positive blue color of muscle cells was localized in the injected muscle, while those of contralateral side or triceps brachii muscles did not show any expression (data not shown). In cervical cord, the lacZ gene expression was detected in two of WT (Fig. 2e,g) and one case of G93A (Fig. 2f,h) mice, respectively. The expression was confined to the large cells of anterior horn in the ipsilateral side of injection and the spinal level (C5–C6) corresponding to biceps brachii muscle inoculated (Fig. 2e,f). The number of motor neurons expressing lacZ gene was only one in the G93A mice (Fig. 2f), and one (Fig. 2g) and two (Fig. 2e) in the two WT mice, respectively. Double staining with lacZ and GFAP demonstrated the blue-stained cells in the spinal cord were not GFAP-positive (\Fig. 2g,h, arrows). In the present study, the adenovirus-mediated lacZ gene was transferred and expressed in injected skeletal muscles and also in the spinal motor neurons not only in asympto-
H. Warita et al. / Neuroscience Letters 246 (1998) 153–156
Fig. 1. Representative microphotographs of the biceps brachii muscles of WT and G93A mice. HE staining of WT (a) and G93A (b) mouse. Small grouping atrophy (arrows) and central nuclei (arrowheads) are noted in the transgenic muscle (b). (c,d) Show the muscle of a G93A mouse with lacZ gene expression as blue colors counterstained with hematoxylin. Original magnification: (a–c) ×50, (d) × 100. Scale bars, (a–c) 200 mm, (d) 100 mm, respectively.
matic WT mice but also in symptomatic transgenic mice carrying the G93A mutant human SOD1 gene. Although the lacZ gene expression in injected muscle was not different between the G93A and WT groups, the number of spinal motor neurons expressing the lacZ gene was relatively smaller in cases with G93A than WT mice. Body weight and clinical scores in such parameters as motility, motor activity, and forelimb muscle strength were significantly decreased in G93A mice, suggesting that they were at the progressed stage of the disease. The present results suggest that adenoviral vector could infect and invade into denervated muscle fibers, and the muscle fibers could express the lacZ gene sufficiently as well as intact muscle fibers. Infection of adenovirus and invasion into cells require a specific receptor [5]. Our results suggest that denervation of muscles by overexpression of the G93A SOD1 gene would not affect such a receptor for adenoviral attachment and internalization into skeletal muscle fibers. The number of infected muscle cells (average 5.5% per each muscle section) is relatively smaller than a previous report (17.6%) [14], probably because the previous authors used b-actin promoter without nuclear transfer signal. The lining of b-gal positive muscle fibers just along the perimysium suggests a limitation of infection beyond the juxtaposing fasciculi (Fig. 1c) [3]. LacZ staining in cervical cord suggests that the delivery of adenoviral vector into the spinal cord was mainly by retrograde axonal transport from innervated muscles, but not through systemic circulation. This is the first report that demonstrates in vivo transgene expression in the spinal cord by intramuscular injection of vector even in symptomatic transgenic mice, although it is already reported in
155
normal rodents [7,8]. The confinement of the lacZ gene expression in the normal spinal motor neurons in the present study (Fig. 2g,h) is similar to previous reports with normal spinal cord [7,8]. The number of b-gal positive motor neurons in the spinal cord was much smaller (only one and two motor neurons in the WT mice) than in skeletal muscle (5– 100 muscle fibers; Table 1 and Fig. 2e). This may be because of multiple innervation of muscle cells by a single motor neuron and of limited infectivity of adenovirus at the axon terminal. Only one motor neuron in four of the G93A mice expressed the lacZ transgene in the spinal cord, and the number is smaller than the WT (Table 1 and Fig. 2f). Recent reports suggest an impairment of axonal transport in animal models for ALS such as the G93A transgenic mice [15,16] and human neurofilament heavy-subunit gene transgenic mice [6]. Thus, the limited vector delivery may be related to an impaired axonal transport. Together with the previous report on gene transfer to pmn mice [10], the present experimental study raises a great possibility that in vivo gene delivery by retrograde axonal transport could be applied for ALS or other motor neuron diseases, even at the progressed stage as well as the presymptomatic state.
Fig. 2. Histology of the anterior horn of the spinal cord of WT (a,c,e,g) and transgenic (b,d,f,h) mice. (a–d) Show HE staining, (e,f) lacZ expression as blue colors counterstained with hematoxylin, and (g,h) also lacZ expression doublestained with GFAP. In contrast to normal anterior horn (a,c), loss of motor neurons and mild gliosis in G93A sample (b,d) are noted. X-Gal positive cells are large pyramidal neurons, and are not GFAP positive (g,h, arrows). Magnification: (a,b,e,f) ×20, (c,d) ×50, (g,h) ×100. Scale bars, (a,b,e,f) 500 mm, (c,d) 200 mm, (g,h) 100 mm, respectively. An arrow in (a) shows the central canal of the spinal cord. Dotted lines in (e) and (f) show the margin of the anterior horn and the white matter of the spinal cord.
156
H. Warita et al. / Neuroscience Letters 246 (1998) 153–156
[1] Abe, K., Morita, S., Kikuchi, T. and Itoyama, Y., Protective effect of a novel free radical scavenger, OPC-14117, on wobbler mouse motor neuron disease, J. Neurosci. Res., 48 (1997) 63–70. [2] Abe, K., Setoguchi, Y., Hayashi, T. and Itoyama, Y., Dissociative expression of adenoviral mediated E. coli LacZ gene between ischemic and reperfused rat brains, Neurosci. Lett., 226 (1997) 53–56. [3] Abe, K., Setoguchi, Y., Hayashi, T. and Itoyama, Y., In vivo adenovirus-mediated gene transfer and the expression in ischemic and reperfused rat brain, Brain Res., 763 (1997) 191–201. [4] Aoki, M., Ogasawara, M., Matsubara, Y., Narisawa, K., Nakamura, S., Itoyama, Y. and Abe, K., Familial amyotrophic lateral sclerosis (ALS) in Japan associated with H46R mutation in Cu/Zn superoxide dismutase gene: a possible new subtype of familial ALS, J. Neurol. Sci., 126 (1994) 77–83. [5] Bergelson, J.M., Cunningham, J.A., Droguett, G.G., KurtJones, E.A., Krithivas, A., Hong, J.S., Horwitz, M.S., Crowell, R.L. and Finberg, R.W., Isolation of a common receptor for coxsackie B viruses and adenoviruses 2 and 5, Science, 275 (1997) 1320–1323. [6] Collard, J.F., Coˆte´, F. and Julien, J.P., Defective axonal transport in a transgenic mouse model of amyotrophic lateral sclerosis, Nature, 375 (1995) 61–64. [7] Finiels, F., Ribotta, M.G.Y., Barkats, M., Samolvk, M.L., Robert, J.J., Privat, A., Revah, F. and Mallet, J., Specific and efficient gene transfer strategy offers new potentialities for the treatment of motor neuron diseases, NeuroReport, 7 (1995) 373–378. [8] Ghadge, G.D., Roos, R.P., Kang, U.J., Wollmann, R., Fishman, P.S., Kalynych, A.M., Barr, E. and Leiden, J.M., CNS gene delivery by retrograde transport of recombinant replicationdefective adenoviruses, Gene Ther., 2 (1995) 132–137. [9] Gurney, M.E., Pu, H., Chiu, A.Y., Dal Canto, M.C., Polchow, C.Y., Alexander, D.D., Caliendo, J., Hentati, A., Kwon, Y.W.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
and Deng, H.X., Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation, Science, 264 (1994) 1772–1775. Haase, G., Kennel, P., Pettmann, B., Vigne, E., Akli, S., Reyah, F., Schmalbruch, H. and Kahn, A., Gene therapy of murine motor neuron disease using adenoviral vectors for neurotrophic factors, Nature Med., 3 (1997) 429–436. Ripps, M.E., Huntley, G.W., Hof, P.R., Morrison, J.H. and Gordon, J.W., Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis, Proc. Natl. Acad. Sci. USA, 92 (1995) 689–693. Rosen, D.R., Siddique, T., Patterson, D., Figlewicz, D.A., Sapp, P., Hentati, A., Donaldson, D., Goto, J., O’Regan, J.P. and Deng, H.X., Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis, Nature, 364 (1993) 59–62. Setoguchi, Y., Jaffe, H.A., Danel, C. and Crystal, R.G., Ex vivo and in vivo gene transfer to the skin using replication-deficient recombinant adenovirus vectors, J. Invest. Dermatol., 102 (1994) 415–421. Wong, P.C., Pardo, C.A., Borchelt, D.R., Lee, M.K., Copeland, N.G., Jenkins, N.A., Sisodia, S.S., Cleveland, D.W. and Price, D.L., An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria, Neuron, 14 (1995) 1105–1116. Yang, Y., Haecker, S.E., Su, Q. and Wilson, J.M., Immunology of gene therapy with adenoviral vectors in mouse skeletal muscle, Hum. Mol. Genet., 5 (1996) 1703–1712. Zhang, B., Tu, P., Abtahian, F., Trojanowski, J.Q. and Lee, V.M.Y., 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 (1997) 1307– 1315.
Expression of adenovirus-mediated E. coli lacZ gene in skeletal muscles and spinal motor neurons of transgenic mice with a mutant superoxide dismutase gene Hitoshi Warita a, Koji Abe a ,*, Yasuhiro Setoguchi b, Yasuto Itoyama a a Department of Neurology, Tohoku University School of Medicine, Sendai, Japan Department of Respiratory Medicine, Juntendo University School of Medicine, Tokyo, Japan
b
Received 2 March 1998; received in revised form 13 March 1998; accepted 13 March 1998
Abstract A replication-defective recombinant adenoviral vector containing E. coli lacZ gene was injected into the right biceps brachii muscles of transgenic mice carrying mutant human Cu/Zn superoxide dismutase (SOD1) gene and non-transgenic wild-type mice at 27 weeks of age. Although the transgenic mice showed remarkable neurogenic muscular changes and a marked motor neuron loss in the anterior horn of spinal cord, the lacZ gene was widely expressed in all the injected muscles of transgenic mice as well as of wild-type mice at 7 days after the injection. In one transgenic and two wild-type mice, the lacZ gene expression was first detected in a few motor neurons of right lower cervical cord (C5–C6). These results demonstrate that an adenovirusmediated foreign gene is transferred and expressed in skeletal muscles both of normal and transgenic mice model for familial amyotrophic lateral sclerosis (FALS), and also, in the spinal motor neurons, may be transferred by retrograde transport from innervated muscles. 1998 Elsevier Science Ireland Ltd.
Keywords: Gene transfer; lacZ gene; Familial amyotrophic lateral sclerosis; Superoxide dismutase; Transgenic mouse
Amyotrophic lateral sclerosis (ALS) causes progressive skeletal muscle weakness and ultimately complete paralysis due to loss of motor neurons. There is currently no effective therapy to prevent or cure. About 5–10% of ALS cases is familial (FALS), and approximately 20% of FALS kindreds carry missense mutations in the Cu/Zn superoxide dismutase gene (SOD1; EC 1.15.1.1). Several lines of transgenic mice that express a mutant human gene encoding SOD1 develop phenotypes closely similar to those of FALS, providing a valuable model for ALS [9,10,13] in order to establish a curative therapy for ALS patients. The primary target of therapy is motor neurons in cerebral cortex, brainstem, and spinal cord. Viral vectors including adenoviruses are attractive candidates for direct gene delivery in vivo. Replication-deficient recombinant adenoviral vectors are capable
* Corresponding author. 1-1 Seiryo-machi, Aoba-ku, Sendai 9808574, Japan. Tel.: +81 22 7177189; fax: +81 22 7177192; e-mail: [email protected]
of infecting not only immature or dividing cell types but a variety of terminally differentiated cell types such as neurons and myocytes. In the present study, a replication-defective adenoviral vector containing the E. coli lacZ gene was injected into the skeletal muscles of symptomatic transgenic mice with a mutant human SOD1 and of non-transgenic wild-type (WT) mice, and the expression of lacZ gene was examined in skeletal muscles and spinal motor neurons. Transgenic mice expressing mutant human SOD1 gene containing an amino acid substitution of glycine at position 93 by alanine (G93A) are obtained from the Jackson Laboratory (Bar Harbor, ME, USA) [9]. Their strain designation is B6SJL-TgN (SOD1-G93A) 1Gurdl. Symptomatic G93A transgenic mice (n = 4) at the age of 27 weeks and age-matched WT littermates (SJL/J; n = 4) were used in this study. The transgene of G93A mutant human SOD1 was confirmed by specific primers for exon 4 of the SOD1 gene [11]. The enzyme activity of SOD1 protein was measured by our previous method [4]. Clinical assessments of both G93A and WT mice were also measured before the
0304-3940/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0304- 3940(98) 00245- 6
154
H. Warita et al. / Neuroscience Letters 246 (1998) 153–156
Table 1
LacZ gene expression in muscles and spinal cords G93A Case no. Muscle Spinal cord
1 Biceps Triceps Right Left
+ − − −
WT 2
3
2+ − + −
2+ − − −
4 + − − −
5 + − − −
6 + − − −
7
8
2+ − + −
2+ − + −
G93A, transgenic mice carrying the G93A mutant human SOD1 gene; WT, non-transgenic wild-type mice. The degree of staining was categorized into three grades: minus, no staining; + , small (1–10); 2 + , large (10–100) number of stained cells, respectively.
vector injection as previously reported [1]. Experimental protocols and procedures were approved by the Animal Committee of the Tohoku University School of Medicine, Japan. The same recombinant adenoviral vector was used in this study as in our previous reports [2,3,12], which expresses the E. coli lacZ gene under the control of the human cytomegalovirus (CMV) promoter with the nuclear localizing signal of simian virus 40 (SV40). Under anesthesia with inhalation of a nitrous oxide/oxygen/halothane (68%:30%:2%) mixture, the skin of the right arm was incised, and Ad-CMVn-lacZ (2.5 × 109 plaque forming units (p.f.u.) in 5 ml of vector vehicle consisting of 10 mM Tris–HCl (pH 7.4), 1 mM MgCl2, and 10% glycerol) was slowly injected into the right biceps brachii muscle via a 10 ml sterile Hamilton syringe (the Gastight Highperformance Syringe; Hamilton, Reno, NV, USA), followed by suture of the skin to prevent infection. For histological evaluation of in vivo gene transfer and expression, the animals were decapitated at 7 days after the administration of AdCMVn-lacZ [7,8]. The bilateral biceps and triceps muscles and cervical cords were quickly removed and frozen in powdered dry ice. Transverse sections of 10 mm thickness were cut on a cryostat at −20°C. A set of the sections was stained with hematoxylin and eosin (HE). Other sections were reacted with X-Gal (5-Bromo-4-chloro-3-indolyl-bd-galactopyranoside; Boehringer Mannheim Biochemica, Mannheim, Germany) [2,3]. After the reaction with XGal, the sections were counterstained with hematoxylin. The number of blue color-stained cells was counted in sections of skeletal muscle and cervical cord. The degree of XGal staining was categorized into three grades for semiquantitation as follows: no staining, −; a small (1–10), +; or a large (10–100), 2+ number of stained cells, respectively. A set of spinal cord sections was immunohistochemically stained with a rabbit polyclonal antibody against bovine glial fibrillary acidic protein (GFAP; 1/100, Dako; Carpinter, CA, USA) after the reaction with X-Gal [3]. The G93A group at 27 weeks of age demonstrates a significant decrease of the body weight (26.8 ± 1.9 g, n = 4, mean ± SD, P , 0.01, Student’s t-test) as compared to WT mice (33.8 ± 1.1, n = 4) that showed no clinical symptoms. Rolling number of circular cage (16.0 ± 13.3, P , 0.01, Welch’s test), transverse bar rotations (17.8 ± 12.0, P ,
0.05, t-test), and angle of slope holding (degree of slope, 31.5 ± 8.4, P , 0.01, t-test) were all greatly reduced in the transgenic mice compared to WT group (128.5 ± 55.5, 108.3 ± 22.1, 47.8 ± 5.9, respectively). The SOD activity in peripheral erythrocyte lysate was higher in the transgenic mice (241.4 ± 14.7 unit/mg protein, P , 0.01, t-test) than that in WT mice (172.8 ± 32.1). As compared to normal findings in the WT group (Fig. 1a), histological study showed a diffuse neurogenic muscular atrophy with small group atrophy and small angular fibers in the biceps and triceps brachii muscles of the G93A group (Fig. 1b, arrows). Although central nuclei were barely found in WT mice (,1%, Fig. 1a), they were observed in 10–20% of muscle fibers in the G93A mice (Fig. 1b, arrowheads). In contrast to normal finding in the WT group (Fig. 2a,c), examination of the spinal cord revealed a marked loss of motor neurons and mild gliosis in the anterior horn of G93A mice (Fig. 2b,d). The lacZ gene expression in the muscles and spinal cords are summarized in Table 1, and the examples in Figs. 1c,d and 2e–h. The adenoviral vector was transferred, and the lacZ gene was expressed in the skeletal muscles and the spinal cords. In biceps brachii muscles, all cases in both G93A and WT groups expressed the lacZ gene with nuclear and perinuclear cytoplasmic pattern in 5–100 of muscle fibers per transverse section. In muscles of the transgenic mice, both atrophic and preserved muscle fibers expressed the lacZ gene (Fig. 1c,d). The lacZ positive blue color of muscle cells was localized in the injected muscle, while those of contralateral side or triceps brachii muscles did not show any expression (data not shown). In cervical cord, the lacZ gene expression was detected in two of WT (Fig. 2e,g) and one case of G93A (Fig. 2f,h) mice, respectively. The expression was confined to the large cells of anterior horn in the ipsilateral side of injection and the spinal level (C5–C6) corresponding to biceps brachii muscle inoculated (Fig. 2e,f). The number of motor neurons expressing lacZ gene was only one in the G93A mice (Fig. 2f), and one (Fig. 2g) and two (Fig. 2e) in the two WT mice, respectively. Double staining with lacZ and GFAP demonstrated the blue-stained cells in the spinal cord were not GFAP-positive (\Fig. 2g,h, arrows). In the present study, the adenovirus-mediated lacZ gene was transferred and expressed in injected skeletal muscles and also in the spinal motor neurons not only in asympto-
H. Warita et al. / Neuroscience Letters 246 (1998) 153–156
Fig. 1. Representative microphotographs of the biceps brachii muscles of WT and G93A mice. HE staining of WT (a) and G93A (b) mouse. Small grouping atrophy (arrows) and central nuclei (arrowheads) are noted in the transgenic muscle (b). (c,d) Show the muscle of a G93A mouse with lacZ gene expression as blue colors counterstained with hematoxylin. Original magnification: (a–c) ×50, (d) × 100. Scale bars, (a–c) 200 mm, (d) 100 mm, respectively.
matic WT mice but also in symptomatic transgenic mice carrying the G93A mutant human SOD1 gene. Although the lacZ gene expression in injected muscle was not different between the G93A and WT groups, the number of spinal motor neurons expressing the lacZ gene was relatively smaller in cases with G93A than WT mice. Body weight and clinical scores in such parameters as motility, motor activity, and forelimb muscle strength were significantly decreased in G93A mice, suggesting that they were at the progressed stage of the disease. The present results suggest that adenoviral vector could infect and invade into denervated muscle fibers, and the muscle fibers could express the lacZ gene sufficiently as well as intact muscle fibers. Infection of adenovirus and invasion into cells require a specific receptor [5]. Our results suggest that denervation of muscles by overexpression of the G93A SOD1 gene would not affect such a receptor for adenoviral attachment and internalization into skeletal muscle fibers. The number of infected muscle cells (average 5.5% per each muscle section) is relatively smaller than a previous report (17.6%) [14], probably because the previous authors used b-actin promoter without nuclear transfer signal. The lining of b-gal positive muscle fibers just along the perimysium suggests a limitation of infection beyond the juxtaposing fasciculi (Fig. 1c) [3]. LacZ staining in cervical cord suggests that the delivery of adenoviral vector into the spinal cord was mainly by retrograde axonal transport from innervated muscles, but not through systemic circulation. This is the first report that demonstrates in vivo transgene expression in the spinal cord by intramuscular injection of vector even in symptomatic transgenic mice, although it is already reported in
155
normal rodents [7,8]. The confinement of the lacZ gene expression in the normal spinal motor neurons in the present study (Fig. 2g,h) is similar to previous reports with normal spinal cord [7,8]. The number of b-gal positive motor neurons in the spinal cord was much smaller (only one and two motor neurons in the WT mice) than in skeletal muscle (5– 100 muscle fibers; Table 1 and Fig. 2e). This may be because of multiple innervation of muscle cells by a single motor neuron and of limited infectivity of adenovirus at the axon terminal. Only one motor neuron in four of the G93A mice expressed the lacZ transgene in the spinal cord, and the number is smaller than the WT (Table 1 and Fig. 2f). Recent reports suggest an impairment of axonal transport in animal models for ALS such as the G93A transgenic mice [15,16] and human neurofilament heavy-subunit gene transgenic mice [6]. Thus, the limited vector delivery may be related to an impaired axonal transport. Together with the previous report on gene transfer to pmn mice [10], the present experimental study raises a great possibility that in vivo gene delivery by retrograde axonal transport could be applied for ALS or other motor neuron diseases, even at the progressed stage as well as the presymptomatic state.
Fig. 2. Histology of the anterior horn of the spinal cord of WT (a,c,e,g) and transgenic (b,d,f,h) mice. (a–d) Show HE staining, (e,f) lacZ expression as blue colors counterstained with hematoxylin, and (g,h) also lacZ expression doublestained with GFAP. In contrast to normal anterior horn (a,c), loss of motor neurons and mild gliosis in G93A sample (b,d) are noted. X-Gal positive cells are large pyramidal neurons, and are not GFAP positive (g,h, arrows). Magnification: (a,b,e,f) ×20, (c,d) ×50, (g,h) ×100. Scale bars, (a,b,e,f) 500 mm, (c,d) 200 mm, (g,h) 100 mm, respectively. An arrow in (a) shows the central canal of the spinal cord. Dotted lines in (e) and (f) show the margin of the anterior horn and the white matter of the spinal cord.
156
H. Warita et al. / Neuroscience Letters 246 (1998) 153–156
[1] Abe, K., Morita, S., Kikuchi, T. and Itoyama, Y., Protective effect of a novel free radical scavenger, OPC-14117, on wobbler mouse motor neuron disease, J. Neurosci. Res., 48 (1997) 63–70. [2] Abe, K., Setoguchi, Y., Hayashi, T. and Itoyama, Y., Dissociative expression of adenoviral mediated E. coli LacZ gene between ischemic and reperfused rat brains, Neurosci. Lett., 226 (1997) 53–56. [3] Abe, K., Setoguchi, Y., Hayashi, T. and Itoyama, Y., In vivo adenovirus-mediated gene transfer and the expression in ischemic and reperfused rat brain, Brain Res., 763 (1997) 191–201. [4] Aoki, M., Ogasawara, M., Matsubara, Y., Narisawa, K., Nakamura, S., Itoyama, Y. and Abe, K., Familial amyotrophic lateral sclerosis (ALS) in Japan associated with H46R mutation in Cu/Zn superoxide dismutase gene: a possible new subtype of familial ALS, J. Neurol. Sci., 126 (1994) 77–83. [5] Bergelson, J.M., Cunningham, J.A., Droguett, G.G., KurtJones, E.A., Krithivas, A., Hong, J.S., Horwitz, M.S., Crowell, R.L. and Finberg, R.W., Isolation of a common receptor for coxsackie B viruses and adenoviruses 2 and 5, Science, 275 (1997) 1320–1323. [6] Collard, J.F., Coˆte´, F. and Julien, J.P., Defective axonal transport in a transgenic mouse model of amyotrophic lateral sclerosis, Nature, 375 (1995) 61–64. [7] Finiels, F., Ribotta, M.G.Y., Barkats, M., Samolvk, M.L., Robert, J.J., Privat, A., Revah, F. and Mallet, J., Specific and efficient gene transfer strategy offers new potentialities for the treatment of motor neuron diseases, NeuroReport, 7 (1995) 373–378. [8] Ghadge, G.D., Roos, R.P., Kang, U.J., Wollmann, R., Fishman, P.S., Kalynych, A.M., Barr, E. and Leiden, J.M., CNS gene delivery by retrograde transport of recombinant replicationdefective adenoviruses, Gene Ther., 2 (1995) 132–137. [9] Gurney, M.E., Pu, H., Chiu, A.Y., Dal Canto, M.C., Polchow, C.Y., Alexander, D.D., Caliendo, J., Hentati, A., Kwon, Y.W.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
and Deng, H.X., Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation, Science, 264 (1994) 1772–1775. Haase, G., Kennel, P., Pettmann, B., Vigne, E., Akli, S., Reyah, F., Schmalbruch, H. and Kahn, A., Gene therapy of murine motor neuron disease using adenoviral vectors for neurotrophic factors, Nature Med., 3 (1997) 429–436. Ripps, M.E., Huntley, G.W., Hof, P.R., Morrison, J.H. and Gordon, J.W., Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis, Proc. Natl. Acad. Sci. USA, 92 (1995) 689–693. Rosen, D.R., Siddique, T., Patterson, D., Figlewicz, D.A., Sapp, P., Hentati, A., Donaldson, D., Goto, J., O’Regan, J.P. and Deng, H.X., Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis, Nature, 364 (1993) 59–62. Setoguchi, Y., Jaffe, H.A., Danel, C. and Crystal, R.G., Ex vivo and in vivo gene transfer to the skin using replication-deficient recombinant adenovirus vectors, J. Invest. Dermatol., 102 (1994) 415–421. Wong, P.C., Pardo, C.A., Borchelt, D.R., Lee, M.K., Copeland, N.G., Jenkins, N.A., Sisodia, S.S., Cleveland, D.W. and Price, D.L., An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria, Neuron, 14 (1995) 1105–1116. Yang, Y., Haecker, S.E., Su, Q. and Wilson, J.M., Immunology of gene therapy with adenoviral vectors in mouse skeletal muscle, Hum. Mol. Genet., 5 (1996) 1703–1712. Zhang, B., Tu, P., Abtahian, F., Trojanowski, J.Q. and Lee, V.M.Y., 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 (1997) 1307– 1315.