- Email: [email protected]
Protection by dietary zinc in ALS mutant G93A SOD transgenic mice
Neuroscience Letters 379 (2005) 42–46
Protection by dietary zinc in ALS mutant G93A SOD transgenic mice Irina P. Ermilovaa , Vladimir B. Ermilova , Mark Levya , Emily Hob , Cliff Pereirac , Joseph S. Beckmana,∗ a
b
Linus Pauling Institute, Environmental Health Sciences Center, Department of Biochemistry and Biophysics, Oregon State University, Corvallis, OR 97331, USA Environmental Health Sciences Center, Department of Nutrition and Exercise Sciences, Oregon State University, Corvallis, OR 97331, USA c Environmental Health Sciences Center, Department of Statistics, Oregon State University, Corvallis, OR 97331, USA Received 7 November 2004; received in revised form 19 December 2004; accepted 20 December 2004
Abstract Mutations to the copper, zinc superoxide dismutase (SOD) gene are responsible for 2–3% of amyotrophic lateral sclerosis (ALS) cases. These mutations result in the protein having a reduced affinity for zinc. SOD becomes toxic to motor neurons when zinc is missing from its active site. Recently, high dosages of zinc (75 and 375 mg/kg/day) have been paradoxically reported to increase the death of G93A-mutant SOD transgenic mice [G.J. Groeneveld, J. de Leeuw van Weenen, F.L. van Muiswinkel, H. Veldman, J.H. Veldink, J.H. Wokke, P.R. Bar, L.H. van den Berg, Zinc amplifies mSOD1-mediated toxicity in a transgenic mouse model of amyotrophic lateral sclerosis, Neurosci. Lett. 352 (2003) 175–178]. In contrast, we have found that moderate supplementation of zinc (∼12 mg/kg/day) delayed death in G93A-mutant SOD mice by 11 days compared to mice on a zinc-deficient diet. Supplementing zinc with even 18 mg/kg/day resulted in a more rapid death of some mice, consistent with the results of Groenevelt et al. However, large amounts of zinc competitively inhibit copper absorption, which inhibits the copper-dependent ceruloplasmin, and can cause a lethal anemia. We found that supplementing the 18 mg/kg/day dosage of zinc with 0.3 mg/kg/day of copper prevented the early death from zinc treatment alone. These data support a role for moderate levels of dietary zinc potentially protecting against the toxicity of ALS-associated SOD and the protection does not result from depleting copper. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Zinc deficiency; Amyotrophic lateral sclerosis; G93A human SODI; Transgenic mice; Lou Gehrig’s disease; Superoxide dismutase
The discovery of mutations to the antioxidant enzyme, copper, zinc superoxide dismutase (SOD) over a decade ago has propelled enormous research into ALS, but the basis for the disease remains enigmatic. Only about 2–3% of all ALS cases are associated with SOD [22]. Further screening has now identified over 105 different missense mutations occurring at approximately 40 different positions that are linked to ALS (http://www.alsod.org). SOD is a dimeric protein that contains one copper and one zinc per subunit. Its primary activity is to catalyze sequentially the conversion of two superoxide anions (O2 •− ) to molecular oxygen and hydrogen peroxide [16]. Copper is essential for the catalytic activity for facilitating electron transfer. ∗
Corresponding author. Fax: 1 541 738 6626. E-mail address: [email protected] (J.S. Beckman).
0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.12.045
Zinc also plays an important but subtle role in the function of Cu, Zn SOD [5]. Our previous studies have shown that the absence of zinc in SOD is sufficient to change SOD from being protective to being pro-apoptotic to motor neurons [11,12]. Zinc-deficient SOD is toxic because it subverts low molecular weight antioxidants like ascorbate to transfer electrons to oxygen to make superoxide. In the presence of nitric oxide, the superoxide reacts to form peroxynitrite [3,5]. Previously, we have shown that peroxynitrite triggers apoptotic cascades in motor neurons. In addition, peroxynitrite activates astrocytes to become toxic to motor neurons [7,18]. Furthermore, mutations to SOD associated with ALS reduce the protein’s affinity for zinc slightly [8]. Neurofilaments have an exceptionally strong affinity for zinc and are particularly abundant in motor neurons [9,19]. Because neurofilaments can be a major binding site for zinc in motor neurons, they
I.P. Ermilova et al. / Neuroscience Letters 379 (2005) 42–46
may limit the availability of zinc for SOD and might contribute to the selective vulnerability of motor neurons in ALS [5]. Zinc is a ubiquitous mineral that plays a critical role in the cellular growth, differentiation, and metabolism of both plants and animals. It is required for the processes of gene transcription, the formation of zinc finger proteins that act as DNA binding transcription factors, and the activity of more than 300 different enzymes involved in every aspect of cellular metabolism [1]. Zinc is an essential trace element for humans, and zinc deficiency can lead to a broad spectrum of deficiency symptoms, including alopecia, immune suppression, dermatitis, night blindness, delayed wound healing, and growth retardation [1]. Although overt human zinc deficiency is rare in North America, recent data indicate that mild zinc deficiency may be common [2,23]. Indeed, it is estimated that as much as 10% of the population may be ingesting less than 50% of the current Recommended Dietary Allowance (i.e., 8 mg/day for women, 11 mg/day for men) [26]. Over-expression of ALS mutant SOD in transgenic mice or in rats leads to a progressive motor neuron loss resulting in the death of the animals as rapidly as 130 days. SOD is a constitutively highly expressed protein in vivo and is about 0.5% of soluble protein expressed in brain and spinal cord. In the most highly used transgenic mouse model developed by Gurney et al. [15] that express the G93A SOD mutation, SOD protein is about 13% of total cell protein [4]. This high level of expression must place an enormous demand for zinc in these mice. If the loss of zinc from the active site of SOD is responsible for the toxic gain-of-function in SOD, a zinc-deficient diet should accelerate disease by favoring the accumulation of zinc-deficient SOD. Some evidence in support of this premise exists. Metallothioneins are major intracellular zinc-binding proteins that are important for protecting cells when zinc is low [24]. Genetic deletion of the three major isoforms of metallothionein in brain accelerates the development of ALS in G93A transgenic mice [17,20]. The two astrocytic isozymes for metallothionein were found to be as important as the neuronal isozyme for delaying the disease. Conversely, zinc supplementation should be protective if dietary zinc is limiting in the SOD transgenic mice. In apparent opposition of this premise, high dosages of zinc given in the drinking water have been reported to paradoxically accelerate the death of ALS-SOD transgenic mice [14]. The dosages used in this study were exceptionally high, estimated to be 75 and 375 mg/kg/day of zinc. Notably, the levels of zinc supplementation used in this study were approximately 15 and 75 times greater than current recommendations established for rodents by the American Institute of Nutrition [21], which in turn are five-fold higher than the minimum requirements to maintain optimal health in mice. Excessive zinc blocks the absorption of copper. Elevated zinc is a standard treatment for reducing the copper overload in Wilson’s disease. However, copper is necessary for ceruloplasmin to insert iron into heme and its deficiency can lead to a fatal
43
anemia [13]. In the present study, we report that zinc deficiency accelerates the progression of ALS in G93A SOD mice, while moderate supplementation of zinc provided significant protection. Addition of a low dosage of copper to the drinking water prevented the premature death of transgenic mice treated with high doses of zinc. Transgenic mice expressing G93A SOD were maintained by crossing B6SJL-TgN(SOD1-G93A)1Gur males with matched B6SJLF1/J female mice obtained from Jackson Lab. Mice were bred and maintained in microisolator cages with daily observation. Hemizygous mice for the G93A SOD gene were bred to nontransgenic matched controls from Jackson lab. Transgenic SOD mice were identified by resolving human from mouse SOD from 2 l of tail blood using 12% native polyacrylamide gels [6,20]. As litters are weaned, the SOD positive mice were randomly assigned to treatment groups. Zinc-deficient rodent diet (<2 ppm of Zn) was purchased from Purina Mills and stored at 4 ◦ C. Mice were placed on the same zinc-deficient diet but supplemented with different amounts of zinc in the drinking water. Drinking water was deionized water (18 M) with the addition of Zn2+ of 0, 30, 60 or 90 ppm, corresponding to 0, 0.46, 0.92 and 1.38 mM Zn2+ . The zinc was supplied as zinc sulfate (Sigma). Assuming that mice drink 4–6 ml of water per day and have an average body mass of 25 g, drinking water supplemented with 30 ppm Zn2+ corresponds to an uptake of 6–8 mg/kg/day. Groenevelt et al. [14] used substantially higher amounts of zinc in the drinking water to give effective estimated dosages of 75 and 375 mg/kg/day. The amounts added to the drinking water were not specified in the paper, but were estimated to be 500 and 2500 ppm Zn2+ in the drinking water based upon a paper cited as the source of their method.
Fig. 1. Effect of zinc-deficient diet on males (n = 20) vs. females (n = 9). Zinc deficiency is significantly different from the 30 ppm zinc treatment for both sexes as tested by a factorial design ANOVA. The male group was twice as large because a second trial using only males was included that was conducted a year after the first trial and closely reproduced the original data (p > 0.5 for the trial main effect and interaction). Circles indicate individual data located above the 75th percentile a distance greater than 1.5 times the interquartile range.
44
I.P. Ermilova et al. / Neuroscience Letters 379 (2005) 42–46
Fig. 2. Loss of hind limb grasping reflex for the male mice shown in Fig. 1 (n = 20, p ≤ 0.02). Loss of hind limb function was assessed by the ability of mice to grasp a pencil when suspended by the tail, which is an early overt sign of disease onset.
Mice were sacrificed under anesthesia when they lost the ability to right themselves in 5 s when placed on their side. Because all SOD expressing mice died within a well-defined period with no survivors (i.e. no censoring), the means for the time to death were compared between treatments using methods for uncensored data (e.g., ANOVA). Without censoring, data could be plotted as box plots, where a central line represents the median, the boxes contain the interquartiles (the middle 50% of the data), and vertical lines describe the top and bottom 25% of the data. This type of plot describes the distribution of the data more clearly than traditional survivorship curves, particularly when multiple groups are compared. Residuals were found to be adequate for standard linear models, allowing the data to be analyzed by ANOVA followed by Tukey’s adjustment for post hoc tests. Transgenic mice placed on a zinc-deficient diet at 50 days of age survived for 7 days less than littermates supplied with standard amounts of zinc in the drinking water (30 ppm) on the same diet (p = 0.0004, main effect). All of the G93A SOD positive mice died of motor neuron disease with severe hind limb wasting. During this study, females consistently survived 6 days longer than males (p = 0.0034). The mean reduction in survival due to zinc deficiency was quite similar between males and females (7.0 and 7.4 days, respectively, p > 0.8 interaction). To reduce variability, subsequent studies were conducted only with male mice while females were used for breeding. Mice supplemented at 30 ppm zinc died at the same time as that of G93A SOD mice maintained on standard rodent chow (not shown). G93A SOD mice developed hind limb weakness, a consistent early sign of motor neuron disease, 6 days earlier than G93A SOD on the same diet
Fig. 3. Weights of the male G93A SOD mice on zinc-deficient diet (circles) or with zinc in the drinking water (squares) (mean ± S.D. with n = 20).
supplemented with 30 ppm zinc (Fig. 2). Weight gain for the zinc-deficient mice was essentially the same as with the zincsupplemented group until ALS symptoms appeared (Fig. 3). At that time, mice on the zinc-deficient diet began to lose more weight relative to the zinc-supplemented G93A mice. The zinc-supplemented G93A mice also began to lose weight a week later when they developed ALS-like symptoms. Mice are known to be difficult to make zinc-deficient compared to other species, but over-expression of SOD made them more susceptible to zinc deficiency (Fig. 4). Under the same conditions, nontransgenic littermates survived on a zinc-deficient diet for approximately 8 months before developing overt skin rashes and losing weight while the G93A SOD mice died of ALS within two months. Transgenic mice over-expressing wild type SOD developed zinc-
Fig. 4. Effect of zinc-deficient diet on G93A SOD transgenic mice (n = 20), wild-type SOD transgenic mice (n = 6) and non-transgenic littermates (n = 7). All mice were placed on the zinc-deficient diet at age 50 days.
I.P. Ermilova et al. / Neuroscience Letters 379 (2005) 42–46
Fig. 5. Effects of zinc deficiency vs. increasing doses of zinc supplementation. In the −Zn (30 days), mice were placed on the zinc-deficient diet at an age of 30 days vs. at 50 days. All other groups were given zinc supplements in the drinking water at 50 days. The sample size in all groups was nine male mice assigned randomly to each group.
deficient symptoms more rapidly than nontransgenic mice, losing weight and developing skin rashes, but the wild-type SOD mice did not develop symptoms of overt motor neuron disease. These results suggest that the burden of expressing large amounts of SOD may lead to a generalized zincdeficiency. Having established that zinc deficiency can accelerate ALS, we then tested the effects of increasing the dosage of zinc well beyond that supplied by normal rodent chow (Fig. 5). Mean survival on the 60 ppm diet was 4.5 days longer than that on the 30 ppm treatment and all the mice on 60 ppm survived longer than the median survival of the mice on 30 ppm zinc. However, this difference was not significant by post hoc testing (p = 0.15). Placing the G93A SOD mice on the zinc-deficient diet at 30 days of age resulted in the mice dying of ALS-like symptoms 4 days sooner than when started on the diet at 50 days. Treatment with 90 ppm zinc resulted in an early death of 70% of mice, but three lived longer than the median survival of the 60 ppm zinc group. Higher doses of zinc interfere with copper uptake and are known to cause a fatal anemia. Supplementing the intake of 90 ppm zinc with 2 ppm copper prevented the early mortality of the mice. The addition of copper provided similar protection as observed with 30 ppm zinc with two mice surviving longer than in the 30 or 60 ppm group. The first day for onset of symptoms as assessed by the loss of the ability of the hind legs to grasp a glass rod (mean ± S.D.) progressively increased from 96 ± 7 (0 ppm), 108 ± 12 (30 ppm), 108 ± 13 (60 ppm), 113 ± 19 (90 ppm) to 118 ± 10 (90 ppm + 2 ppm Cu). The overall trend was significant but not the differences between groups by post hoc tests. Dietary zinc substantially modulated death due to ALS of G93A SOD mice, changing survival by two weeks of the zincdeficient group begun at 30 days compared to treatment with elevated dosages of zinc. In contrast to the previous results
45
of Groenevelt et al. [14], we found that moderate elevation of dietary zinc did not accelerate death in the G93A SOD mice. As with any drug, zinc exhibits a bell-shaped curve on survival. The daily intake of zinc to the lowest dosage (75 mg/kg/day) used by Groenevelt et al. [14] caused a more rapid death of G93A SOD transgenic mice. Zinc in the Groenevelt et al. study was given on top of the normal levels of zinc present in rodent chow and was 100–500 times greater than the daily requirements for zinc in mice. We found that as little as ∼18 mg/kg/day (90 ppm), given with a zinc-free diet, was sufficient to enhance death of G93A SOD mice. We also found that the decreased survival of G93A mice on 90 ppm zinc could be mitigated by adding 2 ppm copper to the drinking water, consistent with copper-induced anemia causing the enhanced toxicity [13]. These results also show that the protection provided by zinc in the G93A SOD mice did not result simply from depletion of copper. Certainly zinc deficiency affects a wide range of cellular processes and the present study does not prove that the zincdeficient diet increased the amount of zinc-deficient SOD. However, zinc deficiency accelerated the onset of ALS symptoms in G93A mice SOD and the mice clearly died with hind limb paralysis (Figs. 1 and 2). The same zinc-deficient treatment in nontransgenic littermates resulted in the classical symptoms of zinc-deficiency such as the development of skin rashes only much later. The high expression of SOD in transgenic mice, which can be 13% of total soluble protein [15], creates a large demand for extra zinc and copper. This may explain why the G93A SOD mice are more prone to zincdeficiency than non-transgenic mice. Wild-type SOD overexpressing mice on a zinc-deficient diet developed symptoms of zinc deficiency earlier than transgenic mice, likely from the additional demand of zinc needed to supply SOD. However, these mice did not develop symptoms of motor neuron disease. Variations in dietary zinc content in standard animal chow and drinking water might be an unrecognized source of variability between experiments and laboratories. Our results could hasten the testing of therapeutics by shortening the time needed for disease development and reducing the variability. We also noted a distinct decrease in the survival of male versus female G93A SOD mice, which has been previously reported [25]. However, the effects of zinc deficiency on accelerating death were about the same in both sexes. At present, nothing can be done to treat carriers of SOD mutations to reduce their risk of developing ALS. Approximately 10% of the US population consumes less that 50% of the minimum recommended daily allowance of zinc. We have previously suggested taking moderate zinc supplements might be helpful in families known to be potential carriers of the SOD gene [10]. Our results suggest that zinc supplements might be helpful to avoid deficiency, but should not substantially exceed the Recommended dietary allowance (8–11 mg per day for adults). If anemia results from overdosing with zinc, it can be quickly reversed with small amounts of copper.
46
I.P. Ermilova et al. / Neuroscience Letters 379 (2005) 42–46
Acknowledgments This work was supported by the Center of Excellence P01 AT002034 from the National Center for Complementary and Alternative Medicine, from NIH grant R01 NS033291, and the Linus Pauling Institute. It was made possible, in part, through the support from the Oregon State Environmental Health Sciences Center (ES00210 from the National Institute of Environmental Health Sciences, NIH) and the statistical core in particular. We thank Dr. Mark Leid for his advice on the manuscript.
References [1] P.J. Aggett, J.G. Comerford, Zinc and human health, Nutr. Rev. 53 (1995) S16–S22. [2] K. Alaimo, M.A. McDowell, R.R. Briefel, A.M. Bischof, C.R. Caughman, C.M. Loria, C.L. Johnson, Dietary intake of vitamins, minerals, and fiber of persons ages 2 months and over in the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988–91, Adv. Data (1994) 1–28. [3] J.S. Beckman, T.W. Beckman, J. Chen, P.M. Marshall, B.A Freeman, Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury by nitric oxide and superoxide, Proc. Natl. Acad. Sci. U.S.A. 87 (1990) 1620–1624. [4] J.S. Beckman, A.G. Esetvez, L. Barbeito, J.P. Crow, CCS knockout mice establish an alternative source of copper for SOD in ALS, Free Radic. Biol. Med. 33 (2002) 1433–1435. [5] J.S. Beckman, A.G. Estevez, J.P. Crow, L. Barbeito, Superoxide dismutase and the death of motoneurons in ALS, Trends Neurosci. 24 (2001) S15–S20. [6] W.F. Beyer Jr., I. Fridovich, Assaying for superoxide dismutase activity: some large consequences of minor changes in conditions, Anal. Biochem. 161 (1987) 559–566. [7] P. Cassina, H. Peluffo, M. Pehar, L. Martinez-Palma, A. Ressia, J.S. Beckman, A.G. Estevez, L. Barbeito, Peroxynitrite triggers a phenotypic transformation in spinal cord astrocytes that induces motor neuron apoptosis, J. Neurosci. Res. 67 (2002) 21–29. [8] J.P. Crow, J.B. Sampson, Y. Zhuang, J.A. Thompson, J.S. Beckman, Decreased zinc affinity of amyotrophic lateral sclerosis-associated superoxide dismutase mutants leads to enhanced catalysis of tyrosine nitration by peroxynitrite, J. Neurochem. 69 (1997) 1936– 1944. [9] J.P. Crow, Y.Z. Ye, M. Strong, M. Kirk, S. Barnes, J.S. Beckman, Superoxide dismutase catalyzes nitration of tyrosines by peroxynitrite in the rod and head domains of neurofilament-L, J. Neurochem. 69 (1997) 1945–1953. [10] A.G. Est´evez, J.P. Crow, J.B. Sampson, C. Reiter, Y.-X. Zhuang, G.J. Richardson, M.M. Tarpey, L. Barbeito, J.S. Beckman, Induction of nitric oxide-dependent apoptosis in motor neurons by zinc-deficient superoxide dismutase, Science 286 (1999) 2498–2500. [11] A.G. Est´evez, J.B. Sampson, Y.-X. Zhuang, N. Spear, G.J. Richardson, J.P. Crow, M.M. Tarpey, L. Barbeito, J.S. Beckman, Liposomedelivered superoxide dismutase prevents nitric oxide-dependent motor neuron death induced by trophic factor withdrawal, Free Radic. Biol. Med. 28 (2000) 437–446.
[12] A.G. Estevez, N. Spear, S.M. Manuel, R. Radi, C.E. Henderson, L. Barbeito, J.S. Beckman, Nitric oxide and superoxide contribute to motor neuron apoptosis induced by trophic factor deprivation, J. Neurosci. 18 (1998) 923–931. [13] P.L. Fox, The copper-iron chronicles: the story of an intimate relationship, Biometals 16 (2003) 9–40. [14] G.J. Groeneveld, J. de Leeuw van Weenen, F.L. van Muiswinkel, H. Veldman, J.H. Veldink, J.H. Wokke, P.R. Bar, L.H. van den Berg, Zinc amplifies mSOD1-mediated toxicity in a transgenic mouse model of amyotrophic lateral sclerosis, Neurosci. Lett. 352 (2003) 175–178. [15] M.E. Gurney, H. Pu, A.Y. Chiu, M.C. Dal Corto, 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 (1994) 1772–1775. [16] J.M. McCord, I. Fridovich, Superoxide dismutase: an enzymic function for erythrocuprein (hemocuprein), J. Biol. Chem. 244 (1969) 6049–6055. [17] S. Nagano, M. Satoh, H. Sumi, H. Fujimura, C. Tohyama, T. Yanagihara, S. Sakoda, Reduction of metallothioneins promotes the disease expression of familial amyotrophic lateral sclerosis mice in a dosedependent manner, Eur. J. Neurosci. 13 (2001) 1363–1370. [18] M. Pehar, P. Cassina, M.R. Vargas, R. Castellanos, L. Viera, J.S. Beckman, A.G. Estevez, L. Barbeito, Astrocytic production of nerve growth factor in motor neuron apoptosis: implications for amyotrophic lateral sclerosis, J. Neurochem. 89 (2004) 464–473. [19] K.B. Pierson, M.A. Evenson, 200 KD neurofilament protein binds Al, Cu and Zn, Biochem. Biophys. Res. Commun. 152 (1988) 598–604. [20] K. Puttaparthi, W.L. Gitomer, U. Krishnan, M. Son, B. Rajendran, J.L. Elliott, Disease progression in a transgenic model of familial amyotrophic lateral sclerosis is dependent on both neuronal and nonneuronal zinc binding proteins, J. Neurosci. 22 (2002) 8790–8796. [21] P.G. Reeves, F.H. Nielsen, G.C. Fahey Jr., AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet, J. Nutr. 123 (1993) 1939–1951. [22] 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, Z. Rahmani, A. Krizus, D. McKenna-Yasek, A. Cayabyab, S.M. Gaston, R. Berger, R.E. Tanszi, J.J. Halperin, B. Herzfeldt, R. Van den Bergh, W.-Y. Hung, T. Bird, G. Deng, D.W. Mulder, C. Smyth, N.G. Lang, E. Soriana, M.A. Pericak-Vance, J. Haines, G.A. Rouleau, J.S. Gusella, H.R. Horvitz, R.H. Brown Jr., Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis, Nature 362 (1993) 59–62. [23] H.H. Sandstead, J.C. Smith Jr., Deliberations and evaluations of approaches, endpoints and paradigms for determining zinc dietary recommendations, J. Nutr. 126 (1996) 2410S–2418S. [24] D.A. Suhy, K.D. Simon, D.I. Linzer, T.V. O’Halloran, Metallothionein is part of a zinc-scavenging mechanism for cell survival under conditions of extreme zinc deprivation, J. Biol. Chem. 274 (1999) 9183–9192. [25] J.H. Veldink, P.R. Bar, E.A. Joosten, M. Otten, J.H. Wokke, L.H. van den Berg, Sexual differences in onset of disease and response to exercise in a transgenic model of ALS, Neuromuscul. Disord. 13 (2003) 737–743. [26] P. Wakimoto, G. Block, Dietary intake, dietary patterns, and changes with age: an epidemiological perspective, J. Gerontol. A: Biol. Sci. Med. Sci. 56 (Spec. No. 2) (2001) 65–80.
Protection by dietary zinc in ALS mutant G93A SOD transgenic mice Irina P. Ermilovaa , Vladimir B. Ermilova , Mark Levya , Emily Hob , Cliff Pereirac , Joseph S. Beckmana,∗ a
b
Linus Pauling Institute, Environmental Health Sciences Center, Department of Biochemistry and Biophysics, Oregon State University, Corvallis, OR 97331, USA Environmental Health Sciences Center, Department of Nutrition and Exercise Sciences, Oregon State University, Corvallis, OR 97331, USA c Environmental Health Sciences Center, Department of Statistics, Oregon State University, Corvallis, OR 97331, USA Received 7 November 2004; received in revised form 19 December 2004; accepted 20 December 2004
Abstract Mutations to the copper, zinc superoxide dismutase (SOD) gene are responsible for 2–3% of amyotrophic lateral sclerosis (ALS) cases. These mutations result in the protein having a reduced affinity for zinc. SOD becomes toxic to motor neurons when zinc is missing from its active site. Recently, high dosages of zinc (75 and 375 mg/kg/day) have been paradoxically reported to increase the death of G93A-mutant SOD transgenic mice [G.J. Groeneveld, J. de Leeuw van Weenen, F.L. van Muiswinkel, H. Veldman, J.H. Veldink, J.H. Wokke, P.R. Bar, L.H. van den Berg, Zinc amplifies mSOD1-mediated toxicity in a transgenic mouse model of amyotrophic lateral sclerosis, Neurosci. Lett. 352 (2003) 175–178]. In contrast, we have found that moderate supplementation of zinc (∼12 mg/kg/day) delayed death in G93A-mutant SOD mice by 11 days compared to mice on a zinc-deficient diet. Supplementing zinc with even 18 mg/kg/day resulted in a more rapid death of some mice, consistent with the results of Groenevelt et al. However, large amounts of zinc competitively inhibit copper absorption, which inhibits the copper-dependent ceruloplasmin, and can cause a lethal anemia. We found that supplementing the 18 mg/kg/day dosage of zinc with 0.3 mg/kg/day of copper prevented the early death from zinc treatment alone. These data support a role for moderate levels of dietary zinc potentially protecting against the toxicity of ALS-associated SOD and the protection does not result from depleting copper. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Zinc deficiency; Amyotrophic lateral sclerosis; G93A human SODI; Transgenic mice; Lou Gehrig’s disease; Superoxide dismutase
The discovery of mutations to the antioxidant enzyme, copper, zinc superoxide dismutase (SOD) over a decade ago has propelled enormous research into ALS, but the basis for the disease remains enigmatic. Only about 2–3% of all ALS cases are associated with SOD [22]. Further screening has now identified over 105 different missense mutations occurring at approximately 40 different positions that are linked to ALS (http://www.alsod.org). SOD is a dimeric protein that contains one copper and one zinc per subunit. Its primary activity is to catalyze sequentially the conversion of two superoxide anions (O2 •− ) to molecular oxygen and hydrogen peroxide [16]. Copper is essential for the catalytic activity for facilitating electron transfer. ∗
Corresponding author. Fax: 1 541 738 6626. E-mail address: [email protected] (J.S. Beckman).
0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.12.045
Zinc also plays an important but subtle role in the function of Cu, Zn SOD [5]. Our previous studies have shown that the absence of zinc in SOD is sufficient to change SOD from being protective to being pro-apoptotic to motor neurons [11,12]. Zinc-deficient SOD is toxic because it subverts low molecular weight antioxidants like ascorbate to transfer electrons to oxygen to make superoxide. In the presence of nitric oxide, the superoxide reacts to form peroxynitrite [3,5]. Previously, we have shown that peroxynitrite triggers apoptotic cascades in motor neurons. In addition, peroxynitrite activates astrocytes to become toxic to motor neurons [7,18]. Furthermore, mutations to SOD associated with ALS reduce the protein’s affinity for zinc slightly [8]. Neurofilaments have an exceptionally strong affinity for zinc and are particularly abundant in motor neurons [9,19]. Because neurofilaments can be a major binding site for zinc in motor neurons, they
I.P. Ermilova et al. / Neuroscience Letters 379 (2005) 42–46
may limit the availability of zinc for SOD and might contribute to the selective vulnerability of motor neurons in ALS [5]. Zinc is a ubiquitous mineral that plays a critical role in the cellular growth, differentiation, and metabolism of both plants and animals. It is required for the processes of gene transcription, the formation of zinc finger proteins that act as DNA binding transcription factors, and the activity of more than 300 different enzymes involved in every aspect of cellular metabolism [1]. Zinc is an essential trace element for humans, and zinc deficiency can lead to a broad spectrum of deficiency symptoms, including alopecia, immune suppression, dermatitis, night blindness, delayed wound healing, and growth retardation [1]. Although overt human zinc deficiency is rare in North America, recent data indicate that mild zinc deficiency may be common [2,23]. Indeed, it is estimated that as much as 10% of the population may be ingesting less than 50% of the current Recommended Dietary Allowance (i.e., 8 mg/day for women, 11 mg/day for men) [26]. Over-expression of ALS mutant SOD in transgenic mice or in rats leads to a progressive motor neuron loss resulting in the death of the animals as rapidly as 130 days. SOD is a constitutively highly expressed protein in vivo and is about 0.5% of soluble protein expressed in brain and spinal cord. In the most highly used transgenic mouse model developed by Gurney et al. [15] that express the G93A SOD mutation, SOD protein is about 13% of total cell protein [4]. This high level of expression must place an enormous demand for zinc in these mice. If the loss of zinc from the active site of SOD is responsible for the toxic gain-of-function in SOD, a zinc-deficient diet should accelerate disease by favoring the accumulation of zinc-deficient SOD. Some evidence in support of this premise exists. Metallothioneins are major intracellular zinc-binding proteins that are important for protecting cells when zinc is low [24]. Genetic deletion of the three major isoforms of metallothionein in brain accelerates the development of ALS in G93A transgenic mice [17,20]. The two astrocytic isozymes for metallothionein were found to be as important as the neuronal isozyme for delaying the disease. Conversely, zinc supplementation should be protective if dietary zinc is limiting in the SOD transgenic mice. In apparent opposition of this premise, high dosages of zinc given in the drinking water have been reported to paradoxically accelerate the death of ALS-SOD transgenic mice [14]. The dosages used in this study were exceptionally high, estimated to be 75 and 375 mg/kg/day of zinc. Notably, the levels of zinc supplementation used in this study were approximately 15 and 75 times greater than current recommendations established for rodents by the American Institute of Nutrition [21], which in turn are five-fold higher than the minimum requirements to maintain optimal health in mice. Excessive zinc blocks the absorption of copper. Elevated zinc is a standard treatment for reducing the copper overload in Wilson’s disease. However, copper is necessary for ceruloplasmin to insert iron into heme and its deficiency can lead to a fatal
43
anemia [13]. In the present study, we report that zinc deficiency accelerates the progression of ALS in G93A SOD mice, while moderate supplementation of zinc provided significant protection. Addition of a low dosage of copper to the drinking water prevented the premature death of transgenic mice treated with high doses of zinc. Transgenic mice expressing G93A SOD were maintained by crossing B6SJL-TgN(SOD1-G93A)1Gur males with matched B6SJLF1/J female mice obtained from Jackson Lab. Mice were bred and maintained in microisolator cages with daily observation. Hemizygous mice for the G93A SOD gene were bred to nontransgenic matched controls from Jackson lab. Transgenic SOD mice were identified by resolving human from mouse SOD from 2 l of tail blood using 12% native polyacrylamide gels [6,20]. As litters are weaned, the SOD positive mice were randomly assigned to treatment groups. Zinc-deficient rodent diet (<2 ppm of Zn) was purchased from Purina Mills and stored at 4 ◦ C. Mice were placed on the same zinc-deficient diet but supplemented with different amounts of zinc in the drinking water. Drinking water was deionized water (18 M) with the addition of Zn2+ of 0, 30, 60 or 90 ppm, corresponding to 0, 0.46, 0.92 and 1.38 mM Zn2+ . The zinc was supplied as zinc sulfate (Sigma). Assuming that mice drink 4–6 ml of water per day and have an average body mass of 25 g, drinking water supplemented with 30 ppm Zn2+ corresponds to an uptake of 6–8 mg/kg/day. Groenevelt et al. [14] used substantially higher amounts of zinc in the drinking water to give effective estimated dosages of 75 and 375 mg/kg/day. The amounts added to the drinking water were not specified in the paper, but were estimated to be 500 and 2500 ppm Zn2+ in the drinking water based upon a paper cited as the source of their method.
Fig. 1. Effect of zinc-deficient diet on males (n = 20) vs. females (n = 9). Zinc deficiency is significantly different from the 30 ppm zinc treatment for both sexes as tested by a factorial design ANOVA. The male group was twice as large because a second trial using only males was included that was conducted a year after the first trial and closely reproduced the original data (p > 0.5 for the trial main effect and interaction). Circles indicate individual data located above the 75th percentile a distance greater than 1.5 times the interquartile range.
44
I.P. Ermilova et al. / Neuroscience Letters 379 (2005) 42–46
Fig. 2. Loss of hind limb grasping reflex for the male mice shown in Fig. 1 (n = 20, p ≤ 0.02). Loss of hind limb function was assessed by the ability of mice to grasp a pencil when suspended by the tail, which is an early overt sign of disease onset.
Mice were sacrificed under anesthesia when they lost the ability to right themselves in 5 s when placed on their side. Because all SOD expressing mice died within a well-defined period with no survivors (i.e. no censoring), the means for the time to death were compared between treatments using methods for uncensored data (e.g., ANOVA). Without censoring, data could be plotted as box plots, where a central line represents the median, the boxes contain the interquartiles (the middle 50% of the data), and vertical lines describe the top and bottom 25% of the data. This type of plot describes the distribution of the data more clearly than traditional survivorship curves, particularly when multiple groups are compared. Residuals were found to be adequate for standard linear models, allowing the data to be analyzed by ANOVA followed by Tukey’s adjustment for post hoc tests. Transgenic mice placed on a zinc-deficient diet at 50 days of age survived for 7 days less than littermates supplied with standard amounts of zinc in the drinking water (30 ppm) on the same diet (p = 0.0004, main effect). All of the G93A SOD positive mice died of motor neuron disease with severe hind limb wasting. During this study, females consistently survived 6 days longer than males (p = 0.0034). The mean reduction in survival due to zinc deficiency was quite similar between males and females (7.0 and 7.4 days, respectively, p > 0.8 interaction). To reduce variability, subsequent studies were conducted only with male mice while females were used for breeding. Mice supplemented at 30 ppm zinc died at the same time as that of G93A SOD mice maintained on standard rodent chow (not shown). G93A SOD mice developed hind limb weakness, a consistent early sign of motor neuron disease, 6 days earlier than G93A SOD on the same diet
Fig. 3. Weights of the male G93A SOD mice on zinc-deficient diet (circles) or with zinc in the drinking water (squares) (mean ± S.D. with n = 20).
supplemented with 30 ppm zinc (Fig. 2). Weight gain for the zinc-deficient mice was essentially the same as with the zincsupplemented group until ALS symptoms appeared (Fig. 3). At that time, mice on the zinc-deficient diet began to lose more weight relative to the zinc-supplemented G93A mice. The zinc-supplemented G93A mice also began to lose weight a week later when they developed ALS-like symptoms. Mice are known to be difficult to make zinc-deficient compared to other species, but over-expression of SOD made them more susceptible to zinc deficiency (Fig. 4). Under the same conditions, nontransgenic littermates survived on a zinc-deficient diet for approximately 8 months before developing overt skin rashes and losing weight while the G93A SOD mice died of ALS within two months. Transgenic mice over-expressing wild type SOD developed zinc-
Fig. 4. Effect of zinc-deficient diet on G93A SOD transgenic mice (n = 20), wild-type SOD transgenic mice (n = 6) and non-transgenic littermates (n = 7). All mice were placed on the zinc-deficient diet at age 50 days.
I.P. Ermilova et al. / Neuroscience Letters 379 (2005) 42–46
Fig. 5. Effects of zinc deficiency vs. increasing doses of zinc supplementation. In the −Zn (30 days), mice were placed on the zinc-deficient diet at an age of 30 days vs. at 50 days. All other groups were given zinc supplements in the drinking water at 50 days. The sample size in all groups was nine male mice assigned randomly to each group.
deficient symptoms more rapidly than nontransgenic mice, losing weight and developing skin rashes, but the wild-type SOD mice did not develop symptoms of overt motor neuron disease. These results suggest that the burden of expressing large amounts of SOD may lead to a generalized zincdeficiency. Having established that zinc deficiency can accelerate ALS, we then tested the effects of increasing the dosage of zinc well beyond that supplied by normal rodent chow (Fig. 5). Mean survival on the 60 ppm diet was 4.5 days longer than that on the 30 ppm treatment and all the mice on 60 ppm survived longer than the median survival of the mice on 30 ppm zinc. However, this difference was not significant by post hoc testing (p = 0.15). Placing the G93A SOD mice on the zinc-deficient diet at 30 days of age resulted in the mice dying of ALS-like symptoms 4 days sooner than when started on the diet at 50 days. Treatment with 90 ppm zinc resulted in an early death of 70% of mice, but three lived longer than the median survival of the 60 ppm zinc group. Higher doses of zinc interfere with copper uptake and are known to cause a fatal anemia. Supplementing the intake of 90 ppm zinc with 2 ppm copper prevented the early mortality of the mice. The addition of copper provided similar protection as observed with 30 ppm zinc with two mice surviving longer than in the 30 or 60 ppm group. The first day for onset of symptoms as assessed by the loss of the ability of the hind legs to grasp a glass rod (mean ± S.D.) progressively increased from 96 ± 7 (0 ppm), 108 ± 12 (30 ppm), 108 ± 13 (60 ppm), 113 ± 19 (90 ppm) to 118 ± 10 (90 ppm + 2 ppm Cu). The overall trend was significant but not the differences between groups by post hoc tests. Dietary zinc substantially modulated death due to ALS of G93A SOD mice, changing survival by two weeks of the zincdeficient group begun at 30 days compared to treatment with elevated dosages of zinc. In contrast to the previous results
45
of Groenevelt et al. [14], we found that moderate elevation of dietary zinc did not accelerate death in the G93A SOD mice. As with any drug, zinc exhibits a bell-shaped curve on survival. The daily intake of zinc to the lowest dosage (75 mg/kg/day) used by Groenevelt et al. [14] caused a more rapid death of G93A SOD transgenic mice. Zinc in the Groenevelt et al. study was given on top of the normal levels of zinc present in rodent chow and was 100–500 times greater than the daily requirements for zinc in mice. We found that as little as ∼18 mg/kg/day (90 ppm), given with a zinc-free diet, was sufficient to enhance death of G93A SOD mice. We also found that the decreased survival of G93A mice on 90 ppm zinc could be mitigated by adding 2 ppm copper to the drinking water, consistent with copper-induced anemia causing the enhanced toxicity [13]. These results also show that the protection provided by zinc in the G93A SOD mice did not result simply from depletion of copper. Certainly zinc deficiency affects a wide range of cellular processes and the present study does not prove that the zincdeficient diet increased the amount of zinc-deficient SOD. However, zinc deficiency accelerated the onset of ALS symptoms in G93A mice SOD and the mice clearly died with hind limb paralysis (Figs. 1 and 2). The same zinc-deficient treatment in nontransgenic littermates resulted in the classical symptoms of zinc-deficiency such as the development of skin rashes only much later. The high expression of SOD in transgenic mice, which can be 13% of total soluble protein [15], creates a large demand for extra zinc and copper. This may explain why the G93A SOD mice are more prone to zincdeficiency than non-transgenic mice. Wild-type SOD overexpressing mice on a zinc-deficient diet developed symptoms of zinc deficiency earlier than transgenic mice, likely from the additional demand of zinc needed to supply SOD. However, these mice did not develop symptoms of motor neuron disease. Variations in dietary zinc content in standard animal chow and drinking water might be an unrecognized source of variability between experiments and laboratories. Our results could hasten the testing of therapeutics by shortening the time needed for disease development and reducing the variability. We also noted a distinct decrease in the survival of male versus female G93A SOD mice, which has been previously reported [25]. However, the effects of zinc deficiency on accelerating death were about the same in both sexes. At present, nothing can be done to treat carriers of SOD mutations to reduce their risk of developing ALS. Approximately 10% of the US population consumes less that 50% of the minimum recommended daily allowance of zinc. We have previously suggested taking moderate zinc supplements might be helpful in families known to be potential carriers of the SOD gene [10]. Our results suggest that zinc supplements might be helpful to avoid deficiency, but should not substantially exceed the Recommended dietary allowance (8–11 mg per day for adults). If anemia results from overdosing with zinc, it can be quickly reversed with small amounts of copper.
46
I.P. Ermilova et al. / Neuroscience Letters 379 (2005) 42–46
Acknowledgments This work was supported by the Center of Excellence P01 AT002034 from the National Center for Complementary and Alternative Medicine, from NIH grant R01 NS033291, and the Linus Pauling Institute. It was made possible, in part, through the support from the Oregon State Environmental Health Sciences Center (ES00210 from the National Institute of Environmental Health Sciences, NIH) and the statistical core in particular. We thank Dr. Mark Leid for his advice on the manuscript.
References [1] P.J. Aggett, J.G. Comerford, Zinc and human health, Nutr. Rev. 53 (1995) S16–S22. [2] K. Alaimo, M.A. McDowell, R.R. Briefel, A.M. Bischof, C.R. Caughman, C.M. Loria, C.L. Johnson, Dietary intake of vitamins, minerals, and fiber of persons ages 2 months and over in the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988–91, Adv. Data (1994) 1–28. [3] J.S. Beckman, T.W. Beckman, J. Chen, P.M. Marshall, B.A Freeman, Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury by nitric oxide and superoxide, Proc. Natl. Acad. Sci. U.S.A. 87 (1990) 1620–1624. [4] J.S. Beckman, A.G. Esetvez, L. Barbeito, J.P. Crow, CCS knockout mice establish an alternative source of copper for SOD in ALS, Free Radic. Biol. Med. 33 (2002) 1433–1435. [5] J.S. Beckman, A.G. Estevez, J.P. Crow, L. Barbeito, Superoxide dismutase and the death of motoneurons in ALS, Trends Neurosci. 24 (2001) S15–S20. [6] W.F. Beyer Jr., I. Fridovich, Assaying for superoxide dismutase activity: some large consequences of minor changes in conditions, Anal. Biochem. 161 (1987) 559–566. [7] P. Cassina, H. Peluffo, M. Pehar, L. Martinez-Palma, A. Ressia, J.S. Beckman, A.G. Estevez, L. Barbeito, Peroxynitrite triggers a phenotypic transformation in spinal cord astrocytes that induces motor neuron apoptosis, J. Neurosci. Res. 67 (2002) 21–29. [8] J.P. Crow, J.B. Sampson, Y. Zhuang, J.A. Thompson, J.S. Beckman, Decreased zinc affinity of amyotrophic lateral sclerosis-associated superoxide dismutase mutants leads to enhanced catalysis of tyrosine nitration by peroxynitrite, J. Neurochem. 69 (1997) 1936– 1944. [9] J.P. Crow, Y.Z. Ye, M. Strong, M. Kirk, S. Barnes, J.S. Beckman, Superoxide dismutase catalyzes nitration of tyrosines by peroxynitrite in the rod and head domains of neurofilament-L, J. Neurochem. 69 (1997) 1945–1953. [10] A.G. Est´evez, J.P. Crow, J.B. Sampson, C. Reiter, Y.-X. Zhuang, G.J. Richardson, M.M. Tarpey, L. Barbeito, J.S. Beckman, Induction of nitric oxide-dependent apoptosis in motor neurons by zinc-deficient superoxide dismutase, Science 286 (1999) 2498–2500. [11] A.G. Est´evez, J.B. Sampson, Y.-X. Zhuang, N. Spear, G.J. Richardson, J.P. Crow, M.M. Tarpey, L. Barbeito, J.S. Beckman, Liposomedelivered superoxide dismutase prevents nitric oxide-dependent motor neuron death induced by trophic factor withdrawal, Free Radic. Biol. Med. 28 (2000) 437–446.
[12] A.G. Estevez, N. Spear, S.M. Manuel, R. Radi, C.E. Henderson, L. Barbeito, J.S. Beckman, Nitric oxide and superoxide contribute to motor neuron apoptosis induced by trophic factor deprivation, J. Neurosci. 18 (1998) 923–931. [13] P.L. Fox, The copper-iron chronicles: the story of an intimate relationship, Biometals 16 (2003) 9–40. [14] G.J. Groeneveld, J. de Leeuw van Weenen, F.L. van Muiswinkel, H. Veldman, J.H. Veldink, J.H. Wokke, P.R. Bar, L.H. van den Berg, Zinc amplifies mSOD1-mediated toxicity in a transgenic mouse model of amyotrophic lateral sclerosis, Neurosci. Lett. 352 (2003) 175–178. [15] M.E. Gurney, H. Pu, A.Y. Chiu, M.C. Dal Corto, 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 (1994) 1772–1775. [16] J.M. McCord, I. Fridovich, Superoxide dismutase: an enzymic function for erythrocuprein (hemocuprein), J. Biol. Chem. 244 (1969) 6049–6055. [17] S. Nagano, M. Satoh, H. Sumi, H. Fujimura, C. Tohyama, T. Yanagihara, S. Sakoda, Reduction of metallothioneins promotes the disease expression of familial amyotrophic lateral sclerosis mice in a dosedependent manner, Eur. J. Neurosci. 13 (2001) 1363–1370. [18] M. Pehar, P. Cassina, M.R. Vargas, R. Castellanos, L. Viera, J.S. Beckman, A.G. Estevez, L. Barbeito, Astrocytic production of nerve growth factor in motor neuron apoptosis: implications for amyotrophic lateral sclerosis, J. Neurochem. 89 (2004) 464–473. [19] K.B. Pierson, M.A. Evenson, 200 KD neurofilament protein binds Al, Cu and Zn, Biochem. Biophys. Res. Commun. 152 (1988) 598–604. [20] K. Puttaparthi, W.L. Gitomer, U. Krishnan, M. Son, B. Rajendran, J.L. Elliott, Disease progression in a transgenic model of familial amyotrophic lateral sclerosis is dependent on both neuronal and nonneuronal zinc binding proteins, J. Neurosci. 22 (2002) 8790–8796. [21] P.G. Reeves, F.H. Nielsen, G.C. Fahey Jr., AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet, J. Nutr. 123 (1993) 1939–1951. [22] 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, Z. Rahmani, A. Krizus, D. McKenna-Yasek, A. Cayabyab, S.M. Gaston, R. Berger, R.E. Tanszi, J.J. Halperin, B. Herzfeldt, R. Van den Bergh, W.-Y. Hung, T. Bird, G. Deng, D.W. Mulder, C. Smyth, N.G. Lang, E. Soriana, M.A. Pericak-Vance, J. Haines, G.A. Rouleau, J.S. Gusella, H.R. Horvitz, R.H. Brown Jr., Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis, Nature 362 (1993) 59–62. [23] H.H. Sandstead, J.C. Smith Jr., Deliberations and evaluations of approaches, endpoints and paradigms for determining zinc dietary recommendations, J. Nutr. 126 (1996) 2410S–2418S. [24] D.A. Suhy, K.D. Simon, D.I. Linzer, T.V. O’Halloran, Metallothionein is part of a zinc-scavenging mechanism for cell survival under conditions of extreme zinc deprivation, J. Biol. Chem. 274 (1999) 9183–9192. [25] J.H. Veldink, P.R. Bar, E.A. Joosten, M. Otten, J.H. Wokke, L.H. van den Berg, Sexual differences in onset of disease and response to exercise in a transgenic model of ALS, Neuromuscul. Disord. 13 (2003) 737–743. [26] P. Wakimoto, G. Block, Dietary intake, dietary patterns, and changes with age: an epidemiological perspective, J. Gerontol. A: Biol. Sci. Med. Sci. 56 (Spec. No. 2) (2001) 65–80.