Efficient three-drug cocktail for disease induced by mutant superoxide dismutase.код для вставкиСкачать
ORIGINAL ARTICLES Efficient Three-Drug Cocktail for Disease Induced by Mutant Superoxide Dismutase Jasna Kriz, MD, PhD, Geneviève Gowing, BSc, and Jean-Pierre Julien, PhD There is currently no effective pharmacological treatment for amyotrophic lateral sclerosis (ALS). Because evidence suggests that multiple pathways may contribute to ALS pathogenesis, we tested in a mouse model of ALS (SOD1G37R mice) a combination approach consisting of three drugs for distinct targets in the complex pathway to neuronal death: minocycline, an antimicrobial agent that inhibits microglial activation, riluzole, a glutamate antagonist, and nimodipine, a voltage-gated calcium channel blocker. The efficacy of this three-drug cocktail was remarkable when administered in the diet from late presymptomatic stage (8 –9 months). It delayed the onset of disease, slowed the loss of muscle strength, and increased the average longevity of SOD1G37R mice by 6 weeks. The protective effect of the treatment was corroborated by the reduced immunodetection signals for markers of gliosis and neurodegeneration in the spinal cord of SOD1G37R mice. These results indicate that such three-drug combination may represent an effective strategy for ALS treatment. Ann Neurol 2003;53:429 – 436 Amyotrophic lateral sclerosis (ALS) is a late-onset neurodegenerative disease characterized by progressive muscle weakness and atrophy followed by eventual paralysis, leading to death within 2 to 5 years after diagnosis.1–3 The disease occurs in both sporadic and familial forms with highly similar clinical courses. In the past decade, much effort has been focused to elucidate the mechanisms of disease induced by missense mutations in the gene coding for superoxide dismutase 1 (SOD1) that are responsible for approximately 25% of familial ALS.4,5 Yet, the toxicity of SOD1 mutants is not fully understood. Transgenic mice overexpressing various SOD1 mutants linked to familial ALS develop progressive motor neuron disease with many pathological features found in both familial and sporadic ALS cases.6 – 8 The current view is that the motor neuron death induced by mutant SOD1 is complex and may involve multiple pathways including formation of protein aggregates, proteosome dysfunction, axonal transport defects, oxidative damage, mitochondrial defects, alterations in calcium homeostasis, caspase activation, and changes in levels of Bcl-2 members.1–3,9 Moreover, excitotoxicity due to astrocyte dysfunction and proinflammatory molecules released from activated microglia are other factors that might contribute to propagation of the neurodegenerative process.10 –13 So far, the various pharmacological agents that have From the Centre for Research in Neurosciences, McGill University, Research Institute of the McGill University Health Centre, Montréal, Québec H3G 1A4, Canada. Received Aug 21, 2002, and in revised form Nov 20. Accepted for publication Nov 21, 2002. been tested in ALS mice have produced only modest benefits.1–3 Recently, three different groups including ours reported the beneficial effects of minocycline in ALS mice expressing mutant SOD1.14 –16 Because much evidence suggests that multiple pathways may contribute to ALS pathogenesis, in addition to minocycline, we designed a drug combination approach aimed at different targets in the complex pathway to neuronal death. This drug cocktail consisted of minocycline, an antibiotic that also inhibits microglial activation, riluzole, a glutamate antagonist, and nimodipine, a voltage-gated calcium channel blocker. Here, we report that such three-drug cocktail is remarkably effective to delay onset of disease and mortality in a mouse model of ALS. Materials and Methods Generation of SOD1G37R Mice Transgenic mice overexpressing SOD1G37R by approximately fivefold (line 29)7,17 were enriched in C57BL/6 background. Only mice heterozygous for the SOD1G37R transgene were used for our study. All mice were genotyped by Southern blotting. The use of animals and all surgical procedures were conducted according to the Guide of Care and Use of Experimental Animals of the Canadian Council of Animal Care. Address correspondence to Dr Julien, Research Institute of the McGill University Health Centre, 1650 Cedar Avenue, Montréal, Québec H3G 1A4, Canada. E-mail: email@example.com © 2003 Wiley-Liss, Inc. 429 Three-drug Therapy Protocol The SOD1G37R mice were housed at the standard temperature (21°C) and in a light-controlled environment with ad libitum access to the food and water. The study was conducted using littermates. The mouse littermates were fed a regular rodent food (Harlan Teklad, Houston, TX) and were randomly divided into three-drug therapy–treated and control groups, including wild-type littermates. At the age of 8 to 9 months, SOD1G37R mice from the experimental groups were administered the triple medicated diet TD 01146 (Harlan, Teklad), containing 1,000mg/kg of minocycline, 500mg/kg of riluzole, and 500mg/kg of nimodipine. All three compounds were purchased from Sigma (Oakville, Ontario, Canada). Daily intake of food in mice was 12 to 18gm/100gm of body weight. Calculated loss of compounds during the food processing was 10 to 15%. Thus, we estimate that the mice received daily approximately 80 to 100mg/kg of minocycline and 30 to 40mg/kg for each nimodipine and riluzole. Onset of the clinical disease was determined by measurement of motor strength and by the hind limb contraction when mice were suspended by their tail.14 They were killed when they started to lie on the side and to express difficulties in grooming, as described previously.14,18 To confirm the effects of the three-drug therapy, we conducted two independent experiments with different sets of SOD1G37R mouse littermates. The therapy was initiated at the late presymptomatic stage of disease (8 –9-month-old mice). Muscle Strength Test With minor adjustments, the muscle strength test was conducted as previously described.14 The mice were allowed to grab vertically oriented wire (approximately 2mm in diameter) with the loop at the lower end. For more consistent measurements, the wire was maintained in the vertically oriented circular motion (circle radius, approximately 15cm at 35r.p.m.). Immunohistochemistry Mice were killed by intraperitoneal injection of chlorale hydrate, perfused with 16gm/L sodium cacodylate buffer (pH 7.4) followed by fixative (3% glutaraldehyde in sodium cacodylate buffer). Immunohistochemical studies were performed as previously described.14,17 Incubation with the primary antibodies anti-Cdk5 (C-8, 1:1,000; Santa Cruz Biotechnology, Santa Cruz, CA), anti–glial fibrillary acidic protein monoclonal antibody (anti-GFAP 1:200 dilution; Sigma), anti–mouse Mac2 rat monoclonal antibody (TIB166) distributed by ATCC (1:500 dilution; Manassas, VA), and anti–p-p38 and anti-cleaved caspase-3 rabbit polyclonal antibody (1:500 dilution; New England Biolab, Missisagua, Ontario, Canada) was performed overnight at room temperature in phosphate-buffered saline/bovine serum albumin. The labeling was developed using a vector ABC kit (Vector Laboratories, Burlington, Ontario, Canada) and Sigma-fast tablets (Sigma). Tissue sections for the axonal counting were prepared for embedding in Epon as described previously.17 Western Blots The mice were killed by overdose of chloral hydrate (intraperitoneally). Immediately after, total protein extracts were 430 Annals of Neurology Vol 53 No 4 April 2003 obtained from L4 –L5 spinal cord sections by homogenization in sodium dodecyl sulfate–urea (0.5% sodium dodecyl sulfate, 8M urea in 7.4 phosphate buffer) with a cocktail of protease inhibitors (phenylmethyl sulfonyl fluoride 2mM, leupeptin 2mg/ml, pepstatin 1mg/ml, and 1 mg/ml aprotinin). The protein was measured using a detergent compatible (DC) protein assay (BioRad, Hercules, CA). The proteins were separated on 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes, and detected using monoclonal primary antibodies against anti-GFAP (1:2,000; Sigma), anti-actin (C-4; 1:5,000; Boehringer-Manheim, Indianapolis, IN), and anti– p-p38, rabbit polyclonal antibody (anti–p-p38, Thr180/ Tyr182 1:500 dilution; New England Biolab). The Western blots were revealed using the Renaissance chemiluminescence kit (NEN Life Science, Boston, MA). Data Analysis Data are expressed as a mean ⫾ standard error. Statistical significance was assessed by two-tailed Student’s t test ( p ⱕ 0.05). Results Three-drug Cocktail Increases the Life Span of SOD1G37R Mice Mouse littermates heterozygous for the SOD1G37R transgene (line 29) were fed a regular rodent food (Harlan Teklad). At late presymptomatic stage (8 and 9 months), the mouse littermates were randomly divided into two groups, the three-drug therapy–treated and control groups. The three drugs (minocycline, riluzole, and nimodipine) were delivered as a dietary supplement in a special custom-made rodent diet. Figure 1A shows the survival curves of the drug cocktail– treated SOD1G37R mice and control SOD1G37R mice fed on regular diet. As compared with control littermates, the average life span of the drug-treated SOD1G37R mice was increased by 6 weeks (54.1 ⫾ 0.9; n ⫽ 10 vs 48.0 ⫾ 0.6; n ⫽ 10, p ⱕ 0.05; Table). As shown in Figure 1B, the distribution of mortality for the tested mice showed almost no overlapping between the two tested groups. For control SOD1G37R mice, the mortality peak was at 47 to 48 weeks, whereas the mortality distribution for the drug-treated mice was more equally spread between 52 to 58 weeks (see Fig 1B). Three-drug Cocktail Delays the Onset of Disease and Muscle Strength Decline in SOD1G37R Mice To determine the effects of the three-drug therapy on disease onset and progression in SOD1G37R mice, we conducted a muscle strength assay. This assay is based on the time that mice are able to grip a vertical circulating wire. Unlike normal mice, the SOD1G37R mice exhibited an age-dependent decline in hanging time (Fig 2). The onset of disease in SOD1G37R mice was characterized by a rapid decline in muscle strength (at treatment significantly delayed the onset of muscle weakness and improved the motor performance of the SOD1G37R mice throughout the tested period (see Fig 2 and Table). Effective Protection against the Loss of Motor Axons in SOD1G37R Mice To assess whether the three-drug therapy delayed degeneration of motor neurons, we counted the total number of motor axons in L4 and L5 ventral roots from treated SOD1G37R mice (n ⫽ 3 or 4) and control SOD1G37R littermates (n ⫽ 4) at early symptomatic stage of disease (44 weeks) and at late stage of disease (48 weeks) (Fig 3A and B). At early stage of disease, motor axons from treated SOD1G37R mice were mostly spared unlike axons from control SOD1G37R littermates (see Fig 3A). For instance, at 44 weeks, the L5 ventral roots from control SOD1G37R mice had 390 ⫾ 23 remaining axons, whereas those from drugtreated mice had 823 ⫾ 41 axons, which is not significantly different from control values (911 ⫾ 36). A similar pattern was observed at the level of L4 ventral roots (713 ⫾ 20 for drug-treated vs 352 ⫾ 20 for control SOD1G37R mice). At 48 weeks, the numbers of remaining axons in the L4 and L5 ventral roots from drug-treated SOD1G37R mice were 637 ⫾ 112 and 685 ⫾ 115, respectively. Thus, whereas some axonal loss was evident at 48 weeks, most motor axons were still present in the drug-treated SOD1G37R mice. Fig 1. The three-drug therapy increases the life span of SOD1G37R mice. (A) Survival curves of control SOD1G37R mice and drug-treated SOD1G37R mice (n ⫽ 10). The survival probability of transgenic mice is plotted as a function of the age in weeks. Three-drug treatment starting at late presymptomatic stage of disease increased the average life span of SOD1G37R mice by 6 weeks. (triangles) Three-drug therapy; (squares) SOD1G37R. (B) Mortality distribution of drugtreated versus control SOD1G37R mice according to their age in weeks. (striped bars) SOD1G37R; (filled bars) three-drug therapy. the age of 43– 44 weeks), followed by a slower declining stage (46 – 47 weeks) that progressed to a stage of complete hind limb paralysis. Note that the three-drug Reduced Cdk5 Mislocalization and Capsase-3 Activation Recent studies demonstrated the involvement of caspase-3 activation in ALS pathogenesis.1,9,19 Activation of caspase-3 occurs late in the course of disease and it is associated with the loss of large motor neurons.9 Two other pathological hallmarks of degenerating neurons in SOD1G37R mice are the mislocalization of Cdk517 and the nuclear localization of Cdk420). Cdk5 is normally targeted to the cell membrane by its activator p35. However, in SOD1G37R mice, Cdk5 is mostly detected in the cytoplasm of motor neurons.17 To examine whether the three-drug therapy attenuated the signals for markers of neurodegeneration in the spinal cord sections of SOD1G37R mice, we conducted immunohistochemistry with anti–Cdk5, anti–Cdk4, and anti–caspase-3 antibodies. Table. Three-Drug Therapy Delays the Onset of Disease and Increase Longevity of SOD1G37R Mice (mean ⫾ SEM) SOD1G37R Mice Three-drug therapy (n ⫽ 10) Control (n ⫽ 10) a Muscle Weakness (weeks) Onset of Paralysis (weeks) End-Stage Paralysis (weeks) 47.8 ⫾ 0.95a 43.0 ⫾ 0.92 49.6 ⫾ 1.06a 45.4 ⫾ 0.59 54.1 ⫾ 0.98* 48.0 ⫾ 0.62 Significantly different from control ( p ⱕ 0.05). SEM ⫽ standard error of the mean. Kriz et al: Three-Drug Therapy for ALS 431 nocycline exerted additional effects on glial cell activation in SOD1G37R, we examined by immunohistochemistry and Western blotting expression of Mac-2 and phosphorylated form of p38 MAPK (p-p38), which are markers of microglial activation, and GFAP as a marker for astrogliosis. At early symptomatic stage (44 weeks), the spinal cord sections of age-matched normal mice and drug-treated SOD1G37R mice were almost completely devoid of Mac-2 immunoreactivity (Fig 5). In contrast, the spinal cord of control SOD1G37R mice showed a robust Mac-2 immunoreactivity. The Mac2–immunoreactive cells showed morphology typical of activated microglia/macrophages (irregular shape, short processes; see Fig 5a, panel C). A similar pattern of immunoreactivity was observed with antibodies Fig 2. The three-drug combination improved muscle strength and delayed disease onset. Muscle strength was assessed by the time that mice were able to grip a circulating wire. Each point represents mean ⫾ SEM, Significant difference in comparison of drug-treated versus control SOD1G37R mice (p ⱕ 0.05 by two-tailed Student’s t test). The number of mice in each group was as follows: normal mice, n ⫽ 6; drug-treated SOD1G37R mice and control SOD1G37R mice, n ⫽ 8. Whereas the spinal motor neurons of control SOD1G37R mice exhibited robust cytoplasmic immunoreactivities for Cdk5 and Cdk4 at 10 months (Fig 4G, H), very low levels of cytoplasmic Cdk5 and Cdk4 immunoreactivity was detected in spinal cord sections of 10-month-old SOD1G37R mice under drug treatment (see Fig 4D, E). At the age of 11 months, cytoplasmic immunoreactivities for Cdk5 and Cdk4 were detected in spinal motor neurons of drug-treated SOD1G37R mice but at reduced levels as compared with control SOD1G37R mice (see Fig 4J, K, M, N). Antibodies against activated form of caspase-3 yielded a weak cytoplasmic immunostaining in several spinal motor neurons of drug-treated SOD1G37R mice at 10 months (see Fig 4F), indicating that caspase-3 activation preceded axonal degeneration. Again, much stronger caspase-3 immunoreactivity was detected in motor neurons of 10-month-old control SOD1G37R littermates. We conclude that beneficial effects of the three-drug treatment were associated with reduced signals for markers of neurodegeneration. Three-Drug Therapy Attenuates Astrocytosis and Microglial Activation Astrocytosis and microgliosis are nonneuronal events that are likely to contribute to the neurodegenerative processes in ALS.11–13 Recently, we showed that minocycline alone attenuates microglial activation but not astrocytosis in SOD1G37R mice.14 To determine whether inclusion of nimodipine and riluzole together with mi- 432 Annals of Neurology Vol 53 No 4 April 2003 Fig 3. The three-drug therapy delayed axonal loss in L4 and L5 ventral roots of SOD1G37R mice. The histograms show the total number of axons in L4 and L5 ventral roots of normal mice (wt), drug-treated (G37R-tr), and control SOD1G37R (G37R) mice at the age of 10 months (A) and 11 months (B). The counts were determined using ventral root sections of three to four animals for each group. (asterisks) Significant difference for the comparison of drug-treated or control SOD1G37R littermates versus wild-type mice/ p ⱕ 0.05 by two-tailed Student’s t test. Fig 4. The three-drug treatment attenuates deregulation of Cdks (Cdk5 and Cdk4) and caspase-3 activation. The micrographs show the immunoreactivities of Cdk5, Cdk4, and activated caspase-3 in the spinal cord of WT mice (A–C), of drug-treated SOD1 G37R mice (D–F) and of control SOD1G37R littermates (G–I) at 10 months as well as of drug-treated SOD1G37R mice (J–L) and of control SOD1G37R littermates (M–O) at 11 months. Bar ⫽ 100m. against p-p38. Control SOD1G37R mice yielded a strong p-p38 immunoreactivity in the white and gray matter (predominantly ventral horns) of the spinal cord (see Fig. 5a, panel F). The p-p38 signal was considerably attenuated by the three-drug treatment showing the similar level of immunoreactivity as in normal mice (see Fig 5a, panels D, E). This was further confirmed by Western blotting. At age of 10 months, spinal cord extracts from normal mice or from drug-treated SOD1G37R mice showed very low levels of p-p38 as compared with control SOD1G37R mice (see Fig 5b). During disease progression, the lev- els of p-p38 in spinal cord extracts gradually increased in drug-treated SOD1G37R mice. Unlike minocycline alone,14 the presence of riluzole and nimodipine in our three-drug cocktail markedly attenuated GFAP immunoreactivity in spinal cord sections of SOD1G37R mice at 44 week (see Fig 5a, panels G, H). This was further confirmed by the weak GFAP immunostaining on Western blot of spinal cord extracts from drug-treated SOD1G37R mice as compared with control SOD1G37R littermates (see Fig 5c). Note, however, that the three-drug therapy was unable to completely abolish astrogliosis as Kriz et al: Three-Drug Therapy for ALS 433 Fig 5. (a) The three-drug therapy attenuates microglial activation and astrogliosis in the spinal cord of SOD1G37R mice. Micrographs show the Mac-2, p-p38, and GFAP immunoreactivities in spinal cords of 10-month-old wild-type mice (A, D, G), drugtreated SOD1G37R mice (B, E, H), and control SOD1G37R littermates (C, F, I). Bar ⫽ 100m. At 10 months of age, lower levels of p-p38 (b) and GFAP (c) were detected by immunoblotting of spinal cord extracts from drug-treated SOD1G37R mice as compared with control SOD1G37R mice. WT ⫽ wild type; GFAP ⫽ glial fibrillary acidic protein. reflected by increased GFAP levels at later stages of disease (see Fig 5c). Discussion Here, we report for the first time to our knowledge a new and effective pharmacological approach to slow ALS pathogenesis caused by mutant SOD1. This treatment is based on the combination of three drugs, minocycline, riluzole, and nimodipine, aiming at distinct targets in the disease process. Remarkably, the three- 434 Annals of Neurology Vol 53 No 4 April 2003 drug cocktail delayed the onset of disease by 4 weeks and increased the average longevity of SOD1G37R mice by 6 weeks. Our analysis confirmed the striking neuroprotective effect of the three-drug cocktail. At 11 months, which corresponds to end-stage paralysis for control SOD1G37R mice, the motor axons in drugtreated SOD1G37R littermates were still largely spared (see Fig 3). In addition, the three-drug therapy attenuated microgliosis and reactive astrocytosis and reduced markers of neurodegeneration such as deregulation of Cdk5 and Cdk4 as well as caspase-3 activation (see Fig 4). Minocycline, a semisynthetic tetracycline derivative that effectively crosses the blood–brain barrier, is extensively used in humans with relatively few side effects.21 Several reports showed that minocycline can exhibit biological effects completely distinct from its antimicrobial action.22,23 Minocycline exerted neuroprotective effects in experimental brain ischemia and brain injury and also in mice models of Huntington’s and Parkinson’s disease.24 –28 It has been suggested that minocycline exerts neuroprotective effects by preventing microglial activation, reducing the induction of caspase-1 and by inhibiting cytochrome c release from mitochondria.14 –16,22–28 In addition, minocycline inhibits matrix metalloproteases, nitric oxide synthases, protein tyrosin nitration, cyclooxygenase-2, and prostaglandin E2 production and also may confer neuroprotection through inhibition of excitotoxin-induced microglial activation.22,23,29 We showed previously that minocycline attenuated microglial activation and extended the longevity of SOD1G37R mice by 3 weeks. Similar results were obtained by other groups in testing the effects of minocycline in ALS mice expressing another SOD1 mutant (G93A).14 –16 Although minocycline can reduce microgliosis, it had no effect on astrogliosis in animal models of stroke24,25 or of ALS.14 The efficacy of the cocktail described here may be related in part to synergistic effects of the three drugs and to the ability to attenuate both microgliosis and astrogliosis. Yet, it remains unclear how the addition of riluzole and nimodipine to minocycline may alter astrogliosis, and certainly further studies with combinations of two drugs may be needed to elucidate the separate beneficial effects of nimodipine and riluzole. The precise mechanism of action of riluzole has not been fully elucidated. It appears to involve interference with excitatory amino acid in the central nervous system, possibly through inhibition of glutamic acid release, blockade or inactivation of sodium channels, and/or activation of G-protein coupled transduction pathways.3,18,30,31 In mice models of ALS, treatment with riluzole alone extended survival for only 13 to 15 days.18 Nimodipine is the L-type voltage-gated calcium channel blocker with preferential effects on the central nervous system.32 Although, in a small clinical trial nimodipine alone was not effective in slowing the progression of disease, its pharmacological properties may be of potential benefit in a drug cocktail.33 Some previous studies suggested altered calcium currents and increased intracellular calcium in ALS pathogenesis.34 –36 In PC12 cells, treatment with nimodipine prevented massive Ca2⫹ influx through voltage-gated calcium channels induced by membrane depolarization, a phenomenon associated with mitochondrial disruption and followed by cell apoptosis and/or necrosis.37 It also exhibits neuroprotective effects in ischemia/hypoxiainduced neuronal damage, and it posses certain anticonvulsant effects against kainic acid–induced seizures.38,39 Recently, nimodipine was shown to promote regeneration and functional recovery after intracranial facial nerve crush.40 Previous attempts to alleviate SOD1-mediated disease with pharmacological approaches in transgenic mouse models have produced only modest benefits. Vitamin E, gabapentin, and salicylate had no effect on survival of SOD1G93A mice.18,41 More benefits were provided in SOD1G93A mice by the intracerebroventricular administration of zVAD-fmk, a broad caspase inhibitor,19 janus 3 inhibitor,42 RPR119990,43 and, with creatine, a compound believed to improve mitochondrial function.44 The three-drug cocktail described here represents one of the most efficient pharmacological treatment in a mouse model of ALS. By conferring neuroprotection together with attenuation of microgliosis and reactive astrocytosis, this three-drug therapy may represent a novel and effective treatment for ALS and possibly other neurodegenerative diseases. This work was supported by the Canadian Institutes of Health Research (CIHR; 42 503), the Neuromuscular Research Partnership (ALS Canada, Muscular Dystrophy Association of Canada, and CIHR), the Center for ALS Research at Johns Hopkins, and a CIHR Senior Investigator Award (J.-P.J.). The technical help of P. Hince and D. Houle is gratefully acknowledged. We are grateful to Drs D. L. Price and D. W. Cleveland for the kind gift of SOD1G37R mice. References 1. Cleveland DW, Rothstein JD. From Charcot to Lou Gehrig: deciphering selective motor neuron death in ALS. Nat Rev Neurosci 2001;2:806 – 819. 2. Julien J-P. Amyotrophic lateral sclerosis: unfolding the toxicity of the misfolded. Cell 2001;104:581–591. 3. Rowland LP, Shneider NA. Amyotrophic lateral sclerosis. N Eng J Med 2001;344:1688 –1699. 4. Rosen DR, Siddique T, Patterson D, et al. Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature 1993;362:59 – 62. 5. Cudkowicz ME, McKenna-Yasek D, Sapp PE, et al. Epidemiology of mutations in superoxide dismutase in amyotrophic lateral sclerosis. Ann Neurol 1997;41:210 –221. 6. Gurney ME, Pu H, Chiu AY, et al. Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation. Science 1994;264:1772–1775. 7. Wong PC, Pardo CA, Borchelt DR, et al. An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria. Neuron 1995;14:1105–1116. 8. Brujin LI, Houseweart MK, Kato S, et al. Aggregation and motor neuron toxicity of an ALS-linked SOD1 mutant independent from wild-type SOD1. Science 1998;281:1851–1854. Kriz et al: Three-Drug Therapy for ALS 435 9. Passinelli P, Houseweart MK, Brown RH, Cleveland DW. Caspase-1 and-3 are sequentially activated in a motor neuron death in Cu,Zn superoxidase dismutase-mediated familial amyotrophic lateral sclerosis. Proc Natl Acad Sci USA 2000;97: 13901–13906. 10. Rothstein JD, Van Kammen M, Levey AI, et al. Selective loss of glial glutamate transporter GLT-1 in amyotrophic lateral sclerosis. Ann Neurol 1995;38:73– 84. 11. Kawamata T, Akiyama H, Yamada T, McGeer PL. Immunologic reactions in amyotrophic lateral sclerosis brain and spinal cord tissue. Am J Pathol 1992;140:691–707. 12. Schiffer D, Cordera S, Cavalla P, Mighelli A. Reactive astrogliosis of the spinal cord in amyotrophic lateral sclerosis. J Neurol Sci 1996;139:27–33. 13. Hall ED, Oostveen JA, Gurney ME. Relationship of microglial and astrocytic activation to disease onset and progression in a transgenic model of familial ALS. Glia 1998;23:249 –256. 14. Kriz J, Nguyen MD, Julien J-P. Minocycline slows disease progression in a mouse model of amyotrophic lateral sclerosis. Neurobiol Dis 2002;10:268 –278. 15. Van Den Bosch L, Tilkin P, Lemmens G, Robberecht W. Minocycline delays disease onset and mortality in a transgenic model of ALS. Neuroreport 2002;13:1– 4. 16. Zhu S, Stavrovskaya IG, Drozda M, et al. Minocycline inhibits cytochrome c release and delays progression of amyotrophic lateral sclerosis in mice. Nature 2002;417:74 –78. 17. Nguyen MD, Larivière R, Julien J-P. Deregulation of Cdk5 in a mouse model of ALS: toxicity alleviated by perikaryal neurofilament inclusions. Neuron 2001;30:135–147. 18. Gurney ME, Cutting FB, Zhai P, et al. Benefit of vitamin E, riluzole, and gabapentin in a transgenic model of familial amyotrophic lateral sclerosis. Ann Neurol 1996;39:147–157. 19. Li M, Ona VO, Guegan C, et al. Functional role of caspase-1 and caspase-3 in an ALS transgenic mouse model. Science 2000;288:335–339. 20. Nguyen MD, Boudreau M, Kriz J et al: Cell cycle regulators in neuronal death pathway of amytrophic lateral sclerosis caused by mutant superoxide dismutase 1. J Neurosc: 2003 (in press). 21. Goulden V, Glass D, Cunliffe WJ. Safety of long-term highdose minocycline in the treatment of acne. Br J Dermatol 1996;134:693– 695. 22. Amin AR, Attur MG, Thakker GD, et al. A novel mechanism of action of tetracyclines: effects on nitric oxide synthases. Proc Natl Acad Sci USA 1996;93:14014 –14019. 23. Golub LM, Lee HM, Ryan ME, et al. Tetracyclines inhibit connective tissue breakdown by multiple non-antimicrobial mechanisms. Adv Dent Res 1998;12:12–26. 24. Yrjänheikki J, Keinänen R, Pellikka M, et al. Tetracyclines inhibit microglial activation and are neuroprotective in global brain ischemia. Proc Natl Acad Sci USA 1998;95: 15769 –15774. 25. Yrjänheikki J, Tikka T, Keinanen R, et al. A tetracycline derivative, minocycline, reduces inflammation and protects against focal cerebral ischemia with a wide therapeutic window. Proc Natl Acad Sci USA 1999;96:13496 –13500. 26. Sanchez Mejia RO, Ona VO, Li M, et al. Minocycline reduces traumatic brain injury-mediated caspase-1 activation, tissue damage, and neurological dysfunction. Neurosurgery 2001;48: 1393–1399. 27. Chen M, Ona VO, Li M, et al. Minocycline inhibits caspase-1 and caspase-3 expression and delays mortality in a transgenic mouse model of Huntington disease. Nat Med 2000;6: 797– 801. 436 Annals of Neurology Vol 53 No 4 April 2003 28. Du Y, Ma Z, Lin S, et al. Minocycline prevents nigrostriatal dopaminergic neurodegeneration in the MPTP model of Parkinson’s disease. Proc Natl Acad Sci USA 2001;98: 14669 –14674. 29. Tikka T, Fiebich BL, Goldsteins G, et al. Minocycline, a tetracycline derivative, is neuroprotective against excitotoxicity by inhibiting activation and proliferation of microglia. J Neurosci 2001;21:2580 –2588. 30. Bensimon G, Lacomblez L, Meininger VA. Controlled trial of riluzole in amyotrophic lateral sclerosis. ALS/Riluzole Study Group. N Eng J Med 1994;330:585–591. 31. Martin D, Thompson MA, Nadler JV. The neuroprotective agent riluzole inhibits release of glutamate and aspartate from slices of hippocampal area CA1. Eur J Pharmacol 1993;250: 473– 476. 32. Singh BN. The mechanism of action of calcium antagonists relative to their clinical applications. Br J Clin Pharmacol 1986; 21:109 –121. 33. Miller RG, Shepherd R, Dao H, et al. Controlled trial of nimodipine in amyotrophic lateral sclerosis. Neuromuscul Disord 1996;6:101–104. 34. Siklos L, Engelhardt J, Harati Y, et al. Ultrastructural evidence for altered calcium in motor nerve terminals in amyotrophic lateral sclerosis. Ann Neurol 1996;39:209 –216. 35. Mosier DR, Baldelli P, Delbono O, et al. Amyotrophic lateral sclerosis immunoglobulins increase Ca2⫹ currents in a motoneuron cell line. Ann Neurol 1995;37:102–109. 36. Engelhardt JI, Siklos L, Komuves L, et al. Antibodies to calcium channels from ALS patients passively transferred to mice selectively increase intracellular calcium and induce ultrastructural changes in motoneurons. Synapse 1995;20:185–199. 37. Cano-Abad MF, Villaroya M, Garcia AG, et al. Calcium entry through L-type calcium channels causes mitochondrial disruption and chromaffin cell death. J Biol Chem 2001;43: 39695–39704. 38. Korenkov AI, Pahnke J, Frei K, et al. Treatment with nimodipine or mannitol reduces programmed cell death and infarct size following focal cerebral ischemia. Neurosurg Rev 2000;23: 145–150. 39. Kriz J, Ž upan G, Simonić A. Differential effects of dihydropyridine calcium channel blockers in kainic acid-induced seizures in rats. Epilepsy Res 2002 (in press). 40. Mattsson P, Janson AM, Aldskogius H, Svensson M. Nimodipine promotes regeneration and functional recovery after intracranial facial nerve crush. J Comp Neurol 2001;437:106 –117. 41. Barneoud P, Curet O. Beneficial effects of lysine acetylsalicylate, a soluble salt of aspirin, on motor performance in a transgenic model of amyotrophic lateral sclerosis. Exp Neurol 1999; 155:243–251. 42. Trieu VN, Liu R, Liu X-P, Uckun FM. A specific inhibitor of Janus kinase-3 increases survival in a transgenic mouse model of amyotrophic lateral sclerosis. Biochem Biophys Res Commun 2000;267:22–25. 43. Canton T, Bohme GA, Boireau A, et al. RPR119990, a novel ␣-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid antagonist: synthesis, pharmacological properties, and activity in an animal model of amyotrophic lateral sclerosis. J Pharmacol Exp Ther 2001;299:314 –322. 44. Klivenyi P, Ferrante RJ, Matthews RT, et al. Neuroprotective effects of creatine in a transgenic animal model of amyotrophic lateral sclerosis. Nat Med 1999;5:347–350.