Biometals (2016) 29:827–839 DOI 10.1007/s10534-016-9959-8 Treatment strategies in Alzheimer’s disease: a review with focus on selenium supplementation Jan Aaseth . Jan Alexander . Geir Bjørklund . Knut Hestad . Petr Dusek . Per M. Roos . Urban Alehagen Received: 24 July 2016 / Accepted: 25 July 2016 / Published online: 16 August 2016 Ó The Author(s) 2016. This article is published with open access at Springerlink.com Abstract Alzheimer’s disease (AD) is a neurode- inhibition of enzymes responsible for its formation, or generative disorder presenting one of the biggest to promote resolution of existing cerebral Ab plaques. healthcare challenges in developed countries. No However, these approaches have failed to demonstrate effective treatment exists. In recent years the main significant cognitive improvements. Intracellular focus of AD research has been on the amyloid rather than extracellular events may be fundamental hypothesis, which postulates that extracellular precip- in AD pathogenesis. Selenate is a potent inhibitor of itates of beta amyloid (Ab) derived from amyloid tau hyperphosphorylation, a critical step in the precursor protein (APP) are responsible for the formation of neurofibrillary tangles. Some selenium cognitive impairment seen in AD. Treatment (Se) compounds e.g. selenoprotein P also appear to strategies have been to reduce Ab production through protect APP against excessive copper and iron J. Aaseth Á K. Hestad P. M. Roos (&) Department of Research, Innlandet Hospital Trust, Institute of Environmental Medicine, IMM, Karolinska Brumunddal, Norway Institutet, Nobels va¨g 13, Box 210, 17177 Stockholm, Sweden J. Aaseth Á K. Hestad e-mail: [email protected] Department of Public Health, Hedmark University of Applied Sciences, Elverum, Norway P. M. Roos Department of Clinical Physiology, St.Goran Hospital, J. Alexander Stockholm, Sweden Norwegian Institute of Public Health, Oslo, Norway U. Alehagen J. Alexander Division of Cardiovascular Medicine, Department of Norwegian University of Life Sciences (NMBU), Medical and Health Sciences, Linko¨ping University, A˚ s, Norway Linko¨ping, Sweden G. Bjørklund Council for Nutritional and Environmental Medicine, Mo i Rana, Norway P. Dusek Department of Neurology and Center of Clinical Neuroscience, Charles University in Prague, 1st Faculty of Medicine and General University Hospital in Prague, Prague, Czech Republic 123 828 Biometals (2016) 29:827–839 deposition. Selenoproteins show anti-inflammatory In the present paper, we first discuss treatment properties, and protect microtubules in the neuronal strategies based on traditional hypotheses of AD cytoskeleton. Optimal function of these selenoenzymes pathogenesis: (a) the transmitter hypothesis, (b) the requires higher Se intake than what is common in metabolic hypothesis, (c) the tau protein hypothesis, Europe and also higher intake than traditionally and (d) the amyloid cascade hypothesis. Then we recommended. Supplementary treatment with address the metal-based hypothesis of neuroinflam- N-acetylcysteine increases levels of the antioxidative mation which opens new therapeutic possibilities cofactor glutathione and can mediate adjuvant protec- (Ward et al. 2014). Oxidative stress from copper tion. The present review discusses the role of Se in AD (Cu) and iron (Fe) toxicity is implicated in the metal treatment and suggests strategies for AD prevention by hypothesis of AD pathogenesis. In this context we optimizing selenium intake, in accordance with the discuss a putative therapeutic or preventive role of metal dysregulation hypothesis. This includes in par- selenium (Se) supplementation, evoked by a Swedish ticular secondary prevention by selenium supplemen- study reporting improved vitality and reduced signs of tation to elderly with mild cognitive impairment. inflammation and oxidative stress after Se yeast and coenzyme Q10 intervention in an elderly population Keywords Alzheimer’s disease Á Copper Á Iron Á (Alehagen et al. 2013). Neuroinflammation Á Transmitters Á Selenium Traditional hypotheses Introduction The transmitter hypothesis Alzheimer’s disease (AD) is a neurodegenerative dis- Cholinergic neurons projecting to the hippocampus in order prevalent in old age. In developed countries 13 % the temporal lobe are affected early in AD. Deficient of people over 65 suffer from AD according to the spatial memory in rodents has been mapped to grid Alzheimer’s Association. Projected AD prevalence cells that collect sensory signals in the entorhinal indicate 100 million patients globally by 2050 (Alzhei- cortex (Hafting et al. 2005). The language problems mer’s 2015), leading to considerable economic burden and declining verbal recall characterizing AD patients for society and suffering for patients and caregivers. AD are presumed to arise from dysfunction in hippocam- is classified as genetic or sporadic. Genetic AD is an pal regions (Lim et al. 2012), and these cells are early-onset hereditary disease representing 1–2 % of influenced by cholinergic modulation (Konishi et al. diagnosed cases (Campion et al. 1999). In genetic AD 2015). Loss of cholinergic inputs to the hippocampus mutated genes coding for amyloid precursor protein is a well characterized abnormality in AD. Decreased (APP; chromosome 21) are found, and presenilin 1 acetylcholine release combined with reductions in (PS1; chromosome 14) and presenilin 2 (PS2; chromo- nicotinic and muscarinic receptors in the cortex and some 1), promote amyloid beta (Ab) formation. The hippocampus of AD brains examined post-mortem has vast majority of patients suffer from sporadic AD. been seen (Tata et al. 2014). Acetylcholinesterase Many sporadic AD patients are carriers of the e4 inhibitors used in AD treatment act by increasing allele of the ApoE gene (apolipoprotein E; chromo- acetylcholine bioavailability at synaptic loci. Unfor- some 19). The mechanism whereby ApoE e4 allele tunately, these enzyme inhibitors are not capable of increases AD risk is largely unknown (Hardy and reversing AD, nor slowing disease progression (Wal- Selkoe 2002). Recent research has unmasked minor lace and Bertrand 2013). Memantine acts on the mutations which mediate an intermediate AD risk. N-methyl-D-aspartate (NMDA) receptors blocking Most genes associated with AD roughly cluster within glutamate activity (Parsons et al. 2007). A dysfunction three metabolic pathways: lipid metabolism, inflam- of glutamatergic transmission has been hypothesized matory response, and endocytosis (Giri et al. 2016). to be involved in the neurodegeneration in AD. Aging is considered the principal risk factor for Memantine appears to improve this dysfunction, and sporadic AD, followed by hypertension, dyslipidemia, has been associated with a moderate decrease in metabolic syndrome and diabetes (Drachman 2014). clinical deterioration, with a small positive effect on 123 Biometals (2016) 29:827–839 829 cognition (Areosa et al. 2005). However, although (Yuraszeck et al. 2010). Derangements of microtubuli transmitter dysfunction is seen in AD, it can be and of the neuronal cytoskeleton provide clues to the suspected that the initial biochemical lesions involve understanding of AD pathogenesis. Intact microtubuli structural and functional impairment of vital proteins are involved in transport of essential substances from responsible for transmitter transport and neuronal neuronal bodies to synaptic structures. Phosphorylation integrity. regulates tau protein binding to microtubuli. Under physiological conditions the tau protein remains sol- The metabolic hypothesis uble, but hyperphosphorylation compromises its normal functions (Mehta et al. 2015), and leads to formation of Clinical studies suggest that the metabolic syndrome, insoluble neurofibrillary tangles, which are bundles of including hypertension, obesity, and insulin resistance paired helical protein filaments. Such excessive phos- or type 2 diabetes (T2DM), is a significant risk factor phorylation in AD must result from an imbalance for AD development (Kivipelto et al. 2005). Disturbed between phosphorylating kinases and de-phosphorylat- hippocampal insulin signalling is likely present in AD ing phosphatases. Increased expression of active (Hokama et al. 2014). Increased insulin resistance and kinases adjacent to neurofibrillary tangles has been oxidative stress with elevated levels of advanced described in AD (Hochgrafe et al. 2015). One of these glycation end products (AGE) are proposed mecha- kinases and a potential drug target is cyclin-dependent nisms by which metabolic syndrome may increase the kinase 5 (CDK5). Increased intracellular calcium in AD risk of AD (Li et al. 2012). A recent study in mice brains is associated with CDK5 activation (Shukla et al. indicated an association between amount of hypotha- 2012). CDK5 inhibitors have demonstrated neuropro- lamic beta-amyloid fragments, neuroinflammation tective properties in in vitro and in vivo AD models and peripheral glucose intolerance (Clarke et al. (Zimmer et al. 2012). Sodium selenate also reduces tau 2015). Reactive oxygen species (ROS) and tumor phosphorylation, both in cell cultures and in AD mouse necrosis factor alpha also contribute to this intriguing models (Corcoran et al. 2010b). Administration of syndrome combination (Lourenco et al. 2013). As the selenate to rodents produces cognitive improvements molecular mechanisms in AD and in insulin resistance and reduced neurodegeneration (van Eersel et al. 2010). seem related, it is tempting to assume that drugs used In these models selenate is presumably