An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Noebels JL, Avoli M, Rogawski MA, et al., editors. Jasper's Basic Mechanisms of the Epilepsies. 5th edition. New York: Oxford University Press; 2024. doi: 10.1093/med/9780197549469.003.0054
Metformin is a biguanide drug that is commonly used as a first-line treatment for type 2 diabetes. This chemical works by activating the energy sensor AMPK that inhibits hepatic gluconeogenesis. Metformin’s therapeutic properties go beyond its use as an antidiabetic compound. As a result, a large number of publications have recently reported promising results in a variety of neurological diseases. Although several hypotheses have been formulated regarding the specific mechanisms of action of metformin in the central nervous system, such as a decrease of neuroinflammation or a reduction of oxidative stress, the underlying pathways of these positive effects remain obscure. In this chapter, we review the newly discovered benefits of metformin as a neuroprotective agent for Lafora disease, a rare form of progressive myoclonus epilepsy caused by mutations in the EPM2A or EPM2B genes. The aggregation of Lafora bodies, abnormal glycogen inclusions that accumulate in the brain and other tissues, is the major histopathological characteristic. Treatment with metformin in Lafora disease knockout mice reduced the number of Lafora bodies, decreased neuronal loss and reactive astrogliosis, and improved the functional and epileptic symptoms.
Metformin (N,N-Dimethylimidodicarbonimidic diamide) belongs to the family of biguanide compounds that have glucose-lowering effects (Grytsai et al., 2021). Among the biguanide family of compounds, metformin was initially of little clinical interest due to its low potency, requiring high doses to be effective. However, metformin showed a higher safety profile than its counterparts, such as phenformin or buformin, which were discarded for clinical use because they induced lactic acidosis (LaMoia and Shulman, 2021; Sanz et al., 2021). Metformin is currently the most commonly prescribed drug for type 2 diabetes (T2D) and is taken by an estimated 150 million people worldwide. It has the advantage over other non-insulin-based diabetes therapies of reducing blood glucose levels without inducing hypoglycemia. Due to its superior safety profile, it has become the first-line treatment for T2D and is now featured on the World Health Organization’s essential medicines list (LaMoia and Shulman, 2021; Foretz et al., 2014; Pryor and Cabreiro, 2015). However, although metformin is usually well tolerated, it does have some side effects. In some patients, it can produce lactic acidosis, gastrointestinal discomfort, and vitamin B12 deficiency. For this reason, metformin should be administered initially at low doses that are increased if side effects do not appear (Demare et al., 2021).
The regular dose of metformin used in diabetic patients is 1–2 g/day, leading to a plasma metformin concentration of 50–100 μM. Numerous reports on the possible mechanisms of action have been published and some contradictory results have been shown due to differences in the analytical cellular system or in the doses used (LaMoia and Shulman, 2021; Sanz et al., 2021).
Due to its unusually hydrophilic nature, metformin cannot passively diffuse through cell membranes and must rely on members of organic cation transporters (OCTs) (LaMoia and Shulman, 2021; Sanz et al., 2021). Therefore, only tissues that express members of the OCT family (e.g., liver, kidney, and small intestine) are targets for the action of metformin (Wang et al., 2002; Alnouti et al., 2006; Ursini et al., 2018). Metformin can also act on neuronal cells as they express two members of the OCT family, namely OCT1 and plasma membrane monoamine transporter (PMAT) (Li and Barres, 2018). Once metformin enters the cells, it accumulates mainly within mitochondria (LaMoia and Shulman, 2021) (Fig. 54–1).
Metabolic pathways affected by metformin. A diagram of the main pathways related to glycolysis and gluconeogenesis is depicted. Gluconeogenic substrates are highlighted in gray, and enzymes directly affected by metformin are highlighted in orange. Specific (more...)
The therapeutic potential of metformin expands its prescribed use as an antidiabetic drug. Thus, metformin has been shown to be effective in the treatment of multiple diseases, including polycystic ovary syndrome (Velazquez et al., 1994), cardiovascular diseases (Lexis et al., 2015), and cancer (Wang et al., 2017; Evans et al.; 2005; Li, 2011). In addition, it delays the aging process (Wang et al., 2017; Barzilai et al., 2016; Piskovatska et al., 2020), alleviating the associated inflammation (Bharath et al., 2020) and modulating the microbiota to promote health (Pollak, 2017). Furthermore, metformin may have additional beneficial effects yet to be discovered. In this work, we will review the different mechanisms of action of metformin that have been proposed and the beneficial effect as a neuroprotective agent with special attention to epileptic disorders and Lafora disease, a form of progressive myoclonus epilepsy.
It is now widely accepted that the antihyperglycemic effect of metformin is mainly due to the suppression of hepatic glucose production (HGP) (LaMoia and Shulman, 2021; Foretz et al., 2014; Pryor and Cabreiro, 2015). Hepatic glucose production is the result of the balance between glucose forming pathways (gluconeogenesis and glycogenolysis) and glucose consuming pathways (glycogen synthesis, glycolysis, and the pentose phosphate pathway) (Fig. 54–1). Among them, hepatic gluconeogenesis contributes more than 50% of HGP and is considered the main pathway regulated by metformin. Metformin negatively regulates gluconeogenesis at different levels:
At the transcriptional level: metformin prevents cAMP responsive binding (CREB)-mediated transcription of the gluconeogenic glucose 6-phosphatase (G6PC) and phosphoenolpyruvate carboxykinase 1 (PEPCK1) genes. This is an indirect effect due to the inhibition of mitochondrial complex I by metformin (see below), leading to an increase in AMP levels, which inhibits adenylate cyclase and thus leads to a decrease in levels of cAMP, a mediator of CREB-dependent transcription. In addition, metformin has been proposed to activate AMP-activated protein kinase (AMPK), which has a negative effect on the transcriptional regulation of gluconeogenesis genes, among others (Demare et al., 2021) (see below).
Reducing the availability of gluconeogenic substrates: hepatic gluconeogenesis depends on the availability of appropriate substrates, such as glycerol, lactate, pyruvate, alanine, and dihydroxyacetone phosphate (DHAP), in order to convert them to glucose (Fig. 54–1, gray boxes). Glycerol and DHAP are mutually interconnected, since glycerol is converted to glycerol 3-P (G3P) by glycerate kinase (GLCTK) and then G3P is converted to DHAP by mitochondrial glycerol-3-phosphate dehydrogenase (GPD2). Metformin has been shown to inhibit mitochondrial GPD2 at regular concentrations (50–100 μM) (Fig. 54–1, orange box) (LaMoia and Shulman, 2021; Foretz et al., 2014; Pryor and Cabreiro, 2015). Therefore, after treatment with metformin, the levels of DHAP are reduced, and this leads to a decrease in the flux of gluconeogenesis. As a consequence of this inhibition, G3P and glycerol accumulate in hepatocytes (Fig. 54–1).
The function of GPD2 is coupled to one of the major NADH/NAD+ shuttles, the alpha-glycerophosphate shuttle, which consumes NADH and transforms DHAP to glycerol-3P through the action of cytosolic glycerol-3P dehydrogenase 1 (GPD1). Inhibition of GPD2 by metformin alters the cytosolic redox balance and leads to a higher NADH/NAD+ ratio, due to low levels of DHAP. This high NADH/NAD+ ratio also prevents the conversion of lactate into pyruvate by lactate dehydrogenase (LDH). Therefore, inhibition of GPD2 by metformin decreases the levels of two of the main gluconeogenic substrates, DHAP and pyruvate, and leads to the accumulation of glycerol and lactate (Fig. 54–1). This accumulation of lactate is probably the cause of the appearance of lactic acidosis in some patients treated with metformin (LaMoia and Shulman, 2021; Foretz et al., 2014; Pryor and Cabreiro, 2015).
However, metformin does not affect the gluconeogenic use of pyruvate and alanine as substrates, since their entry into the gluconeogenesis pathway does not involve a redox-dependent mechanism (Fig. 54–1). This would explain why hypoglycemia is rarely observed in patients treated with metformin or in healthy individuals, since part of the gluconeogenesis pathway is still active (LaMoia and Shulman, 2021).
Recently it has been reported that metformin interacts with PEN2, a subunit of the γ-secretase. Metformin-bound PEN2 forms a complex with ATP6AP1, a subunit of the lisosomal v-ATPase, which leads to the inhibition of v-ATPase. This leads to the activation of AMP-activated protein kinase (AMPK), a master regulator of energy homeostasis (Foretz et al., 2014; Steinberg and Carling, 2019) (see below), by the glucose-sensing pathway, without affecting cellular AMP levels (Fig. 54–1, left side) (Ma et al., 2022).
An additional mechanism to explain the hypoglycemic effect of metformin on hepatocytes is its inhibitory action on mitochondrial complex I (NADH/ubiquinone oxidoreductase) (El-Mir et al., 2000). Mitochondrial complex I of the respiratory chain is the site of the contribution of NADH to the proton gradient of OXPHOS (oxidative phosphorylation) (Fig. 54–1, orange box). Inhibition of complex I by metformin reduces the mitochondria’s ability to consume NADH and the production of ATP. High levels of NADH and low levels of ATP have a crucial negative impact on the gluconeogenesis pathway, as this process requires a large amount of energy and depends on a correct NADH/NAD+ balance. Furthermore, as ATP production is reduced, the AMP/ATP ratio increases, and this leads to the activation of AMP-activated protein kinase (AMPK). This activation leads to the activation of catabolic pathways (e.g., glycolysis through the activation of Pfkfb3, an enzyme involved in the formation of 2,6-fructose bisphosphate, an allosteric activator of phosphofructokinase 1 [Pfk1]) and the inhibition of anabolic pathways (e.g., glycogen synthesis by inhibiting glycogen synthase [Gs]), to restore energy balance (Fig. 54–1) (Demare et al., 2021). AMPK exerts this function both at the transcriptional level, regulating the activity of different transcriptional factors by phosphorylation (e.g., the downregulation of CREB, carbohydrate-responsive element binding protein [ChREBP], and sterol regulatory element binding protein [SREBP-1]), which are involved in the expression of genes related to gluconeogenesis, the carbohydrate metabolism, and sterol biosynthesis, respectively; on the other hand, AMPK upregulates peroxisome proliferator-activated receptor γ co-activator 1 alpha (PGC1alpha), involved in mitochondrial biogenesis, and activates pro-health span molecules, such as the forkhead box O3 (FOXO3) transcription factor and sirtuin 1 (SIRT1) deacetylase, which in turn induce the expression of protective molecules (Foretz et al., 2014; Demare et al., 2021; Steinberg and Carling, 2019). AMPK also operates at the level of key metabolic enzyme activity (e.g., inhibition of acetyl-Co carboxylase [Acc1/2], an enzyme involved in the synthesis of malonyl-CoA, an intermediate in fatty acid synthesis, and an inhibitor of fatty acid oxidation; therefore, AMPK activation inhibits the synthesis of fatty acids and promotes their degradation) (Steinberg and Carling, 2019). Therefore, the increase in the AMP/ATP ratio caused by metformin activates AMPK indirectly, but this effect is only obtained when metformin is administered at supra-pharmacological concentrations (>1 mM) (LaMoia and Shulman, 2021). Recent results support the indirect effect of metformin on the activation of AMPK, as they show that metformin does not affect AMPK directly but acts on the upstream liver kinase (LKB1), which participates in the phosphorylation and activation of the catalytic alpha subunit of AMPK (Shaw et al., 2005) (Fig. 54–1, orange box).
Substantial evidence shows that metformin exerts antioxidant effects. Some of these effects can be attributed to the inhibition of mitochondrial complex I which reduces reactive oxygen species (ROS) production by the OXPHOS respiratory chain (Demare et al., 2021). In addition, metformin has other functions related to the activation of the AMPK pathway: (1) reduction of ROS levels by upregulating the expression of antioxidant enzymes, such as thioredoxin, through the AMPK–FOXO3 pathway; (2) modulation of the expression of sirtuin 3 (SIRT3) deacetylase, whose activity promotes antioxidant effects in the cell; (3) downregulation of NADPH oxidase, one of the main producers of cellular ROS; and (4) enhancement of mitochondrial biogenesis by enhancing the function of PGC1alpha transcription factor (Apostolova et al., 2020).
Following brain injury, neuroinflammation is initially neuroprotective, but when it becomes chronic or excessive, it eventually causes damage (Vezzani et al., 2019). It is now accepted that sustained brain inflammation promotes neuronal hyperexcitability and seizures, and that dysregulation in the immunoinflammatory function of the glia is a common factor that predisposes or contributes to the generation of seizures. At the same time, acute seizures upregulate the production of pro-inflammatory cytokines in microglia and astrocytes, triggering a cascade of inflammatory mediators. Thus, epileptic seizures and inflammatory mediators form a positive feedback loop, reinforcing each other (Vezzani et al., 2019). For this reason, it has recently been proposed that the treatment of inflammation with specific anti-inflammatory drugs may be beneficial in the treatment of refractory epilepsies (Vezzani et al., 2019). However, since the chronic use of general anti-inflammatory drugs is not recommended due to their detrimental performance in long-term treatments (Andreasson et al., 2016), only specific anti-inflammatory compounds, selected after a thorough understanding of the main related inflammatory pathways, should be used with each particular type of epilepsy.
Activation of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB) is a hallmark of neuroinflammation and is present in most neurological diseases. The Toll-like receptor 4 (TLR4) signaling pathway induces NF-kB activation through myeloid differentiation primary response 88 (MyD88) and tumor necrosis receptor-associated factor 6 (TRAF6), leading to the expression of pro-inflammatory mediators: cytokines, chemokines, cyclooxygenase 2 (COX2), and inducible nitric oxide synthase (iNOS) (Sanz and Garcia-Gimeno, 2020). It has been described that the activation of AMPK by metformin reduces general inflammatory conditions since it inhibits the signaling of NF-kB, as well as the expression of pro-inflammatory cytokines (interleukin 1-beta [IL-1beta], interleukin 6 [IL-6], tumor necrosis factor alpha [TNFalpha], C–C motif chemokine ligand 2 [CCL2], etc.) in different cell types (Gu et al., 2014), suggesting that AMPK activation could protect against neuroinflammation (Rotermund et al., 2018; Abdi et al., 2021; Khodadadi et al., 2022). Similarly, activation of AMPK by berberine reduces activated microglia; neutrophil infiltration; and IL-1beta, IL-6, CCL2, and CXCL2 production, which occur after traumatic brain injury (Chen et al., 2014), and activation of AMPK with AICAR attenuates inflammation through the inhibition of NF-kB activation and IL-1b expression (Xiang et al., 2019). In all cases, AMPK prevented the activation of the TLR4/NF-kB signaling pathway (Gu et al., 2014; Chen et al., 2014; Qin et al., 2020). AMPK also inhibits lipopolysaccharide (LPS)-induced expression of pro-inflammatory cytokines (TNF-alpha, IL-1beta, and IL-6) by attenuating LPS-induced, TLR4-mediated NF-kB activation (Giri et al., 2004). Similarly, AMPK prevented the advanced glycation end-product (AGE)-mediated signaling pathway, which ends with an increase in NF-kB expression and reduced iNOS and COX2 levels in AGE-treated human neural stem cells (hNSCs) (Chung et al., 2017). The anti-inflammatory action of AMPK was also associated with the inhibition of LPS-induced activation of the phosphatidyl inositol 3 kinase (PI3-kinase)/RAC-alpha serine/threonine-protein kinase (Akt) pathway (Jhun et al., 2004). Downregulation of NF-kB levels inhibits the activation of the nucleotide-binding oligomerization domain and leucine-rich repeat and pyrin domain 3 (NLRP3) inflammasomes, while decreasing the activation of caspase1 and reducing the production of IL-1beta (Salminen et al., 2011). These mechanisms are particularly important in microglia, where AMPK inhibits the release of pro-inflammatory markers, decreasing neuroinflammation (Bayliss et al., 2016), and in astrocytes, where AMPK inhibits elevated ER stress and hyperglycemia-induced inflammation (Wang et al., 2021). Because neuroinflammation is linked to a variety of neurological disorders, including epilepsy, metformin may have beneficial effects in these conditions.
Metformin’s potential use in central nervous system disorder is gaining interest. Although the nature of specific neurological disorders varies, they all share common pathogenic pathways. The AMPK and mechanistic target of rapamycin (mTOR) kinase pathways are examples of these. AMPK is a master regulator of energy homeostasis, as previously stated. It is active in periods of energy deprivation, when it restores energy balance by activating catabolic and suppressing anabolic pathways. The mTOR pathway, on the other hand, is activated in high-energy environments and works by activating anabolic while suppressing catabolic pathways. The AMPK and mTOR pathways are linked, and activation of the AMPK network causes inhibition of the mTOR system, either by directly inactivating components of the mTOR complex (e.g., raptor, tuberous sclerosis complex 2 [TSC2]) or indirectly reversing the action of mTOR on common substrates (e.g., ULK1 in the autophagy process) (Van Nostrand et al., 2020). Metformin, in particular, increases AMPK signaling and suppresses mTOR signaling through both AMPK-dependent and AMPK-independent mechanisms (LaMoia and Shulman, 2021).
Different treatment techniques to block the mTOR pathway have been proposed since various forms of epilepsy are linked to the overexpression of this system (Citraro et al, 2016). Microglia produce less pro-inflammatory cytokines and chemokines (IL-1beta, TNFalpha, CCL2, iNOS, and others) when mTOR signaling is inhibited. mTOR inhibition improved the motor impairment in a middle cerebral artery occlusion (MCAO) model of cerebral stroke (Li et al., 2016). Remarkably, metformin-induced AMPK activation reduced mTOR signaling, resulting in enhanced seizure control in mTOR overactivation mice (LaMoia and Shulman, 2021; Pryor and Cabreiro, 2015; Kalender et al., 2010; Mehrabi et al., 2018). The beneficial effects of inhibiting the mTOR pathway in epilepsy were supported by a recent study that found that blocking the mTOR pathway by activating AMPK with metformin improved lithium- and pilocarpine-induced status epilepticus in rats (Bojja et al., 2021) and by others in which metformin plus caloric restriction decreased generalized convulsive seizures in the electrical kindling epilepsy model (Rubio Osornio et al., 2018).
AMPK activation has also been shown to be beneficial in other epilepsy models. In the pilocarpine-induced status epilepticus rat model, AMPK activation may be neuroprotective against status epilepticus-induced brain injury by modulating PGC-1 and UCP2 levels, which can counteract intracellular oxidative stress following status epilepticus, suggesting the AMPK/PGC-1 pathway as a target for developing novel therapeutic approaches for treating epileptic seizures (Han et al., 2011). The activation of AMPK by metformin was found to improve the inflammatory state and histological abnormalities, and to ameliorate cognitive deterioration in an epileptic diabetic rat model (Mohamed et al., 2020). Metformin lowered the levels of IL-1beta, TNF-alpha, NF-kB, and caspase 3 and normalized glutamate and γ-aminobutyric acid (GABA) levels. These effects were thought to be the result of both AMPK activation and mTOR inhibition (Mohamed et al., 2020).
Metformin also displays antiseizure properties in mice models of induced epilepsy using the chemoconvulsant pentylenetetrazol (PTZ), an antagonist of the GABAA receptor (Vernadakis and Woodbury, 1969; Reinhard and Reinhard, 1977; Swinyard et al., 1989). Its antioxidant and anti-apoptotic properties, as well as the overexpression of heat shock protein 70 (Hsp70), are thought to be responsible for these benefits (Mehrabi et al., 2018; Zhao et al., 2014; Hussein et al., 2019). In a PTZ-induced epilepsy model and a kainic acid–induced chronic seizure model, AMPK activation by metformin accelerated seizure termination (Yang et al., 2017).
Metformin has also been shown to produce a downregulation of the RNA-like endoplasmic reticulum kinase (PERK) pro-apoptotic pathway, which prevents the activation of the eukaryotic initiation factor 2 alpha (eIF2alpha), and the expression of the activating transcription factor 4 (ATF4), and the C/EBP homologous protein (CHOP) (Chung et al., 2017). In this regard, metformin inhibited CHOP expression and apoptosis induced by status epilepticus in rats (Chen et al., 2018), although metformin’s positive effects in this model could potentially be attributed to an increase in autophagy (Zeyghami et al., 2020).
In the kainate-induced status epilepticus model, metformin-activated AMPK exhibits an anti-inflammatory effect by inhibiting IL-1beta production and reducing the expression of glial fibrillary acidic protein (GFAP) and S100beta markers of astrogliosis, as well as enhancing the secretion of the anti-inflammatory cytokine IL-10 and progranulin levels (Vazifehkhah et al., 2020b). Metformin prevented neuronal cell death, abnormal neurogenesis, and mossy fiber sprouting in this model, indicating its neuroprotective role against the kainate-induced epileptogenic process (Vazifehkhah et al., 2020a).
The regulatory effect of AMPK on the glucose transporter GLUT1 is another link between AMPK and epilepsy. In endothelial cells and astrocytes, GLUT1 is the primary glucose transporter. In patients with GLUT1 deficiency syndrome (OMIM 606777), GLUT1 defects affect glucose transport in the brain, resulting in epileptic seizures (Koepsell, 2020). Activation of AMPK has been found to regulate the translocation of GLUT1 from internal storage to the plasma membrane by destabilizing the thioredoxin-interacting protein (TXNIP). This leads to an improvement of astrocytic glucose uptake and glycolysis, preserving neuronal metabolic functionality (Muraleedharan et al., 2020). This could explain the favorable effects of metformin-induced AMPK activation on PTZ-induced seizures in mice on a high-fat diet, as metformin treatment resulted in the restoration of GLUT1 and GLUT3 expression levels (Nesci et al., 2020).
In conclusion, in both acute and chronic epilepsy models, AMPK activation reduced epileptic activity by delaying the onset of epilepsy, decreased the frequency of epileptic seizures, accelerated seizure termination, and prevented cognitive deficits by minimizing neuronal death in the hippocampus. Metformin’s antiepileptic benefits could be attributed to reduced oxidative brain damage, activation of the AMPK pathway, suppression of mTOR signaling, decreased levels of brain-derived neurotrophic factor (BDNF) and its tropomyosin tyrosine B (TrkB) receptor, and/or improved proteostasis (H S et al., 2019).
As a result, the AMPK signaling route can be thought of as a new anti-inflammatory and antiseizure signaling mechanism. Its activation inhibits the mTOR and TLR4/NF-kB signaling pathways, making it a promising therapeutic target for immunoinflammatory diseases such as epilepsy (Pilon et al., 2004). Indeed, AMPK activators, such as metformin, appeared as attractive candidates with therapeutic potential as antiseizure medicines in a recent screening of repurposing medications with anticonvulsive effects (Brueggeman et al., 2019).
Lafora disease (OMIM 254780) is an ultra-rare progressive myoclonic epilepsy (PME) characterized by the accumulation of insoluble glycogen-like inclusions (Lafora bodies) in the brain and other tissues (Lafora and Glueck, 1911). It is a neurodegenerative epileptic disorder that may also be considered a glycogen storage disease (Cavanagh, 1999). Various genetically modified animal models with symptoms similar to those seen in patients have been used to investigate the molecular basis of the disease and search for potential therapies (Ganesh et al., 2002; Criado et al., 2012). Metformin is one of the chemicals that has been shown to reduce symptoms in animal models (Berthier et al., 2016; Sánchez-Elexpuru et al., 2017b) and is already being used in clinical practice (Bisulli et al., 2019).
Lafora disease is a fatal neurological disorder that commonly presents in previously normal children between the ages of 10 and 15 years. Epileptic seizures and/or cognitive problems leading to difficulties in school are the first symptoms (Berkovic et al., 1986). The cognitive and neurological degeneration is rapid, and patients quickly develop motor and intellectual impairments that progress to severe dementia, ataxia, dysarthria, amaurosis, and respiratory failure. There is no specific treatment for the disease and only antiseizure medications can be used just to treat seizures. Patients rapidly acquire drug resistance and seizures become frequent and the myoclonus continuous. In most cases, status epilepticus or aspiration pneumonia lead to death, usually within 5 to 10 years after diagnosis (Berkovic et al., 1986; Serratosa, 1999; Minassian, 2001).
Lafora disease is caused by mutations in the EPM2A (epilepsy of progressive myoclonus type 2 gene A) (Minassian et al., 1998; Serratosa et al., 1999) or NHLRC1/EPM2B (NHL repeat-containing protein 1/epilepsy of progressive myoclonus type 2 gene B) (Chan et al., 2003) genes, which encode the laforin and malin proteins, respectively.
In patients with Lafora disease, mutations in the EPM2A gene account for about 60% of the cases, whereas mutations in the EPM2B gene account for 35% of cases. As a group, patients with EPM2B mutations tend to have a significantly milder phenotype than those with EPM2A mutations (Gómez-Abad et al., 2005). However, mutations in the same gene, and even the same mutation, may lead to significant different progression rates due to genetic or epigenetic factors (Jara-Prado et al., 2014).
Several mouse models of Lafora disease have been generated by disrupting the Epm2a or Epm2b genes in mice (Ganesh et al., 2002; Chan et al., 2004; DePaoli-Roach et al., 2010; Turnbull et al., 2010; Criado et al., 2012). We and others have reported the neurological abnormalities in Epm2a–/– (Ganesh et al., 2002) and Epm2b–/– knockout mice (Criado et al., 2012). Both mouse models develop Lafora bodies and neurological alterations that resemble those present in patients with Lafora disease. Both Epm2a–/– and Epm2b–/– mice display deficits in episodic memory, impaired spontaneous motor activity and coordination, ataxia, muscular weakness, dyskinesias and abnormal hindlimb clasping, and spontaneous epileptic activity (Ganesh et al., 2002; Criado et al., 2012; García-Cabrero et al., 2012). The Epm2a null mouse model already exhibited significant neuronal cell loss at 2 months of age, with the majority of degeneration occurring in the absence of Lafora bodies (Ganesh et al., 2002). Dying neurons commonly showed swelling in the endoplasmic reticulum, Golgi networks, and mitochondria in the absence of apoptotic bodies or DNA fragmentation. Neuronal degeneration and Lafora inclusion bodies in Epm2a–/– mice appeared before impaired behavioral responses, ataxia, and electroencephalogram (EEG) epileptiform activity (Ganesh et al., 2002). Epm2b–/– knockout mice show neurological and behavioral impairments by 3–6 months of age, which are associated with a large number of LBs in the cortex, hippocampus, and cerebellum. Without detectable LBs, 16-day-old Epm2b–/– mice display impairment of autophagy, which persists in adult animals (Criado et al., 2012).
Sporadically, Epm2a–/– mice show spontaneous tonic-clonic seizures recorded by video-EEG analysis (García-Cabrero et al., 2012), while both Lafora disease mouse models show spontaneous myoclonic seizures, single spikes, polyspikes, and spike-wave and polyspike-wave complexes correlating with myoclonic jerks (Ganesh et al., 2002; Criado et al., 2012; García-Cabrero et al., 2012).
Neuronal excitability is additionally altered in Epm2a–/– and Epm2b–/– animals, as evidenced by greater susceptibility to the chemoconvulsant PTZ (García-Cabrero et al., 2014). Following injection of PTZ, control mice showed periods of immobility and overt indications of convulsive activity linked with movements of the face, mouth, and forelimbs. Following that, mice develop clonic convulsions, which eventually progress to generalized tonic-clonic seizures. We looked for the presence of both myoclonic jerks, which commonly started within minutes following PTZ injection, and tonic-clonic seizures. Isolated spikes and spike-wave correlations were detected in EEG recordings of operated mice (García-Cabrero et al., 2014). We observed that the percentages of Epm2a–/– and Epm2b–/– mice experiencing myoclonic jerks and tonic-clonic seizures were higher than that of control mice at both subconvulsive and convulsive dosages of PTZ. When compared to age-matched control mice, mutant animals had a shorter latency to tonic-clonic seizures and the length of PTZ-induced seizures was longer after convulsive dosages of PTZ (García-Cabrero et al., 2014).
Functional abnormalities in Epm2a–/– and Epm2b–/– mice correlate with neuronal degeneration, reactive astrogliosis, increased oxidative stress, altered proteostasis, and impaired autophagy (Aguado et al., 2010; Knecht et al., 2010, 2012; Criado et al., 2012; Puri et al., 2012; Romá-Mateo et al., 2015; Burgos et al., 2020; Lahuerta et al., 2018, 2020; Sanz and Garcia-Gimeno, 2020). Moreover, the aggregation of numerous Lafora bodies in the cerebral cortex, hippocampus, basal ganglia, cerebellum, and brainstem was comparable and associated with the neurologic abnormalities reported in the Epm2a–/– and Epm2b–/– mice, suggesting that these inclusions could also cause cognitive and behavioral impairment (Criado et al., 2012; García-Cabrero et al., 2012).
We have studied the effect of various pharmacological treatments in the Epm2b–/– mouse model (Berthier et al., 2016; Sánchez-Elexpuru et al., 2017a, 2017b). To improve proteostasis, we used 4-phenylbutyric acid (4-PBA), a chemical chaperone that reverses the misfolding and aggregation of proteins associated with various neurodegenerative diseases (Ricobaraza et al., 2009; Zhou et al., 2011); trehalose, another chemical chaperone that prevents protein denaturation and protects cellular integrity against stress phenomena (Tanaka et al., 2004; Sinha et al., 2021); and sodium selenate, an agent that reduces oxidative stress and the appearance of epileptic seizures in other animal models (Brenneisen et al., 2005; Rehni and Singh, 2013). Our results indicate that 4-PBA and sodium selenate considerably improve memory impairment, motor activity, abnormal hindlimb clasping, and sensitivity to PTZ (Berthier et al., 2016; Sánchez-Elexpuru et al., 2017a, 2017b). These compounds reduced the number of Lafora bodies, neuronal degeneration and gliosis, and the percentage of mice experiencing myoclonic jerks after subconvulsive and convulsive doses of PTZ. 4-PBA also reduced the percentage of animals with PTZ-induced generalized seizures and lethality, while treatment with selenate completely eradicated generalized seizures and mortality (Berthier et al., 2016; Sánchez-Elexpuru et al., 2017a, 2017b). Because neuroinflammation is a critical hallmark in Lafora disease (Lahuerta et al., 2020), we recently treated Epm2b–/– mice with propranolol and epigallocatechin gallate, two anti-inflammatory compounds, and found that propranolol improved several behavioral tests while also lowered the levels of reactive glia in the brain of Epm2b–/– mice (Mollá et al., 2021).
We also investigated whether metformin had a favorable effect on Lafora disease, given that AMPK activation has been shown to play a neuroprotective role in a variety of neurodegenerative diseases (Poels et al., 2009; Han et al., 2011; Dulovic et al., 2014; Ashabi et al., 2014). Treatment with metformin in adult mice for 2 months clearly reduced the number of PAS-positive and polyubiquitin aggregates in the brain, and decreased neuronal loss and reactive astrogliosis (Berthier et al., 2016). Metformin treatment also improved the functional symptoms in Epm2b–/– mice. In the tail suspension test, adult Lafora disease mice improved their hindlimb clasping performance following 2 months of metformin treatment. Metformin improved spontaneous motor activity and motor coordination abnormalities in the open field and decreased the time to run downward and increased locomotion activity in the vertical pole test (Berthier et al., 2016).
Metformin also reduced hypersensibility to PTZ, thereby minimizing neuronal hyperexcitability in the Epm2b–/– mouse model of Lafora disease (Sánchez-Elexpuru et al., 2017b). Following injection of a convulsive dosage of PTZ, the percentage of animals with generalized seizures, mortality, seizure latency, and seizure length were measured. Control mice after PTZ administration at convulsive dosages showed freezing and convulsive activity, which later progressed to generalized tonic-clonic seizures, which were sometimes associated with death. PTZ induced seizures in 50% of wild-type mice, but in more than 70% of mice lacking malin. Metformin treatment reduced the number of Epm2b–/– mice that developed seizures, lowering it below wild-type values (Sánchez-Elexpuru et al., 2017b). In addition, PTZ-induced mortality was prevented after metformin treatment. As we indicated above, when compared to wild-type mice, Epm2b–/– mice had lower seizure latency values and increased seizure lengths. Metformin treatment enhanced the delay of PTZ-induced seizures in malin knockout mice, whereas seizure lengths were even shorter than in wild-type mice. Thus, metformin reduced PTZ-induced seizures, mortality, and seizure length in Epm2b–/– mice, as well as increased the latency for PTZ-induced seizure onset, alleviating the hyperexcitability seen in mice lacking the malin protein (Sánchez-Elexpuru et al., 2017b).
Based on these findings, metformin was recognized as an orphan drug for the treatment of Lafora disease by the European Medicines Agency and the U.S. Food and Drug Administration in 2016 and 2017, respectively. The outcomes of metformin administration were unclear in a report of 10 patients since the patients in the study were severely affected in advanced stages of the disease. In any case, the authors recommended that treatment should begin as soon as possible after the onset of Lafora disease (Bisulli et al., 2019). Future studies and trials should determine the value of metformin for the treatment Lafora disease.
This work was supported by grants from the Spanish Ministry of Science and Innovation (SAF2017-83151-R to PS and Rti2018-095784b-100 SAF to MPS and JMS) and a grant from the National Institutes of Health (P01 NS097197), which established the Lafora Epilepsy Cure Initiative (LECI), to PS, JMS and MPS.
The authors declare no conflict of interest related to the content of this manuscript.
This is an open access publication, available online and distributed under the terms of a Creative Commons Attribution-Non Commercial-No Derivatives 4.0 International licence (CC BY-NC-ND 4.0), a copy of which is available at https://creativecommons.org/licenses/by-nc-nd/4.0/. Subject to this license, all rights are reserved.
Your browsing activity is empty.
Activity recording is turned off.
See more...
