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/9780197549469.003.0051
Unverricht-Lundborg disease (ULD; EPM1) is an inherited neurodegenerative disorder characterized by onset at 6–15 years, stimulus-sensitive, action-activated myoclonus, epilepsy, and progressive neurological deterioration. It is caused by biallelic pathogenic variants in the CSTB gene, encoding a cystatin B. The most common of these is an unstable expansion of a dodecamer repeat element in the promoter region of the gene, leading to marked downregulation of CSTB expression. Total loss of CSTB is associated with severe neonatal-onset encephalopathy. A cystatin B–deficient mouse models the EPM1 disease relatively well. Myoclonic seizures, preceded by microglial activation, develop at one month of age and are followed by progressive gray and white matter degeneration and motor problems. CSTB is an inhibitor of cysteine proteases of the cathepsin family showing both nuclear and cytoplasmic localization with partial co-localization with lysosomal markers. CSTB function has been linked to protecting neurons from oxidative damage through an oxidative stress-responsive cystatin B-cathepsin B signaling pathway. In the nucleus CSTB has been shown to participate in regulation of cell cycle, and histone H3 tail proteolytic cleavage. Downstream effects of CSTB dysfunction have also been implicated in apopotosis, microglial dysfunction, inflammation, neurogenesis and synapse physiology. Despite the progress made, the exact disease mechanisms in EPM1 remain elusive. This chapter discusses the clinical features of EPM1 and recent advances in understanding its pathophysiology.
Progressive myoclonus epilepsy of Unverricht-Lundborg type (Unverricht-Lundborg disease; ULD; EPM1) is an autosomal recessive neurodegenerative disorder and the most common single cause of progressive myoclonus epilepsy (Franceschetti et al., 2014). It occurs worldwide and is more common in Finland than anywhere else in the world (Sipilä et al., 2020). EPM1 was previously called Baltic myoclonus, Baltic myoclonic epilepsy, and Mediterranean myoclonus. Following identification of the underlying gene, CSTB (Pennacchio et al., 1996), and advances in genetic testing, these disorders are now collectively classified as EPM1.
EPM1 is characterized by onset at 6–15 years, progressive stimulus-sensitive, action-activated myoclonus, and generalized tonic-clonic epileptic seizures (Lehesjoki and Kälviäinen, 2020). As EPM1 progresses, patients develop ataxia, incoordination, intentional tremor, and dysarthria, which reflect widespread neuronal degeneration in the brain. Cognitive performance of patients with EPM1, especially verbal memory, is mostly within the normal range (Äikiä et al., 2021). However, patients show impaired performance in time-limited tests dependent on motor functions. The electroencephalogram (EEG) in EPM1 patients is abnormal, with spontaneous spike-wave discharges, photosensitivity, polyspike discharges during rapid eye movement (REM) sleep, and background slowing (Ferlazzo et al., 2007; Kälviäinen et al., 2008). The EEG abnormalities are more pronounced at initial diagnosis, but in general, they diminish as the disease stabilizes. At the time of diagnosis, the magnetic resonance imaging (MRI) scan is typically normal, with atrophic changes appearing later. These affect cortical motor, sensorimotor, visual, and auditory areas, as well as pons, medulla, and cerebellar hemispheres (Mascalchi et al., 2002; Koskenkorva et al., 2009, 2012; Nigri et al., 2017). The white matter tracts are also widely affected (Manninen et al., 2012). MRI-navigated transcranial magnetic stimulation studies of EPM1 patients have shown changes in cortical responses, which include increased prevailed inhibition in the primary motor cortex and impaired intracortical interactions and coherence (Danner at al., 2009; Julkunen et al., 2013). The limited histopathological findings consist of widespread degenerative changes with no constant evidence of storage material (Haltia et al., 1969; Koskiniemi et al., 1974; Eldridge et al., 1983; Cohen et al., 2011).
Symptomatic pharmacological and rehabilitative management, including psychosocial support, is the mainstay of EPM1 patients’ care (Lehesjoki and Kälviäinen, 2020). Generalized tonic-clonic seizures usually respond well to therapy, while myoclonus is relatively resistant to known therapies. Valproic acid is the first drug of choice and clonazepam, piracetam, levetiracetam, topiramate, and zonisamide are used as add-on therapy for myoclonus. Sodium channel blockers and GABAergic drugs should be avoided. With modern antiepileptic medication, the life expectancy of EPM1 patients has gradually increased, but survival after 40 years of age in Finnish patients has been reported to be poorer than in the population controls (Sipilä et al., 2020).
EPM1 is caused by biallelic partial loss-of-function mutations in the CSTB gene encoding cystatin B. In approximately 90% of cases EPM1 is caused by homozygosity for an expansion of the dodecamer repeat in the promoter region of CSTB (see Table 51–1) resulting in a 90%–95% decrease in CSTB mRNA expression (Lalioti et al., 1997a; Joensuu et al., 2007). The remaining patients are compound heterozygous for the repeat expansion and a mutation affecting the CSTB coding sequence, apart from two reported patients (see below). There is evidence that correlation exists between the expansion length and the age of onset or disease severity (Hyppönen et al., 2015). However, disease severity also varies among affected individuals with apparently similar repeat-size expansions. Individuals compound heterozygous for the repeat expansion and a null mutation develop a more severe form of EPM1, with an earlier age of onset, more severe myoclonus, and poorer cognitive performance (Koskenkorva et al., 2011; Canafoglia et al., 2012). Biallelic null mutations of CSTB cause a neonatal-onset encephalopathy manifesting as severe developmental delay, progressive microcephaly, and hypomyelination on brain imaging (Mancini et al., 2016; O’Brien at al., 2017). Available data on the consequences of CSTB mutations on gene and/or protein expression suggest that the severity of the human disorders correlates with the degree of residual CSTB expression (Fig. 51–1).
Disease-Associated CSTB Mutations.
Genotype-phenotype correlations of CSTB mutations in human and phenotype of the Cstb knockout mouse model (Cstb–/–). While the severity of phenotype in “classical” EPM1 varies between patients with different genotypes, (more...)
At the onset of EPM1, when symptoms are mild, the differential diagnoses of EPM1 include juvenile myoclonic epilepsy, which has a more favorable outcome. In patients presenting with symptoms closely resembling EPM1, but who do not harbor mutations in CSTB, other progressive myoclonus epilepsy (PME) diseases to consider include action myoclonus-renal failure syndrome (SCARB2-associated PME, EPM4) (Dibbens et al., 2009), North Sea progressive myoclonus epilepsy (GOSR2-associated PME, EPM6) (Corbett et al., 2011), and myoclonus epilepsy and ataxia due to mutation in the potassium channel (MEAK, EPM7) (Muona et al., 2015).
The CSTB gene underlying EPM1 was identified by positional cloning (Pennacchio et al., 1996). CSTB is transcribed as a 0.8 kb mRNA using one of the two potential transcription start sites located 97 and 108 nucleotides upstream from the translation initiation codon (Lalioti et al., 1997a). It is alternatively spliced with at least five isoforms of unknown physiological significance, some of which show tissue specificity (Joensuu et al., 2007). CSTB encodes a ubiquitously expressed 98 amino acid protein of approximately 11 kDa. The subcellular localization of CSTB is dependent on the cell cycle and developmental stage of the cell. In immature proliferating cells, CSTB is found mainly in nucleus, whereas in differentiated cells, it is also found in the cytosol, where it is partially associated with lysosomal membranes (Riccio et al., 2001; Brännval et al., 2003; Alakurtti et al., 2005; Čeru at al., 2010; Daura et al., 2021).
CSTB is an inhibitor of papain-like cysteine proteases of the cathepsin superfamily (Turk and Bode, 1991). The crystal structure of recombinant human CSTB protein with the carboxymethylated cysteine protease papain has been determined and consists of a five-stranded antiparallel β-sheet wrapped around a five-turn α-helix and an extension of the C-terminus running along the convex side of the β-sheet (Stubbs et al., 1990). CSTB has been shown to interact in vitro with the active site of several cysteine proteases, cathepsins (e.g., B, H, L, S), by tight, reversible binding (Green et al., 1984; Abrahamson et al., 1986; Brömme et al., 1991; Machleidt et al., 1991; Lenarčič et al., 1996). The main interactions are provided through a tripartite wedge formed by conserved residues in the N-terminal trunk, the first β-hairpin loop of sequence Gln-Val-Val-Ala-Gly (QVVAG), the second hairpin loop, and the C-terminus of CSTB (Stubbs et al., 1990). Cysteine cathepsins have a broad spectrum of physiological functions, many of these occurring in endo/lysosomal compartments and cytosol, where they are involved for example in maintaining cellular homeostasis through nonselective protein degradation, in antigen processing during immune response, in lipid metabolism and in apoptosis (Yadati et al., 2020). They also have functions in nucleus and extracellular space (Yadati et al., 2020). In nucleus, the developmentally regulated cysteine protease, cathepsin L, is linked to cell proliferation and differentiation by cleavage of the transcription factor CUX1 (Goulet et al., 2004) and to regulation of gene expression through cleavage of histone tails (Goulet et al., 2004; Bulynko et al., 2006; Duncan et al., 2008; Adams-Cioaba et al., 2011; Morin et al., 2012; Daura et al., 2021). As CSTB, an inhibitor of several cysteine cathepsins, has been localized both to nucleus and cytosol, it can thus be anticipated to be involved in several cellular processes.
A total of 15 disease-associated mutations have been reported in CSTB (Table 51–1). Most EPM1 patients are homozygous for at least 30 copies of the normally polymorphic (with two or three copies) and unstable expansion of the 12-nucleotide dodecamer (5′-CCCCGCCCCGCG-3′) repeat element located upstream of the transcription start sites in the promoter region of CSTB. The dodecamer expansion mutation causes significant downregulation of CSTB mRNA and protein expression with less than 10% of expression from that seen in controls (Alakurtti et al., 2005; Joensuu et al., 2007). It is found in approximately 90% of the disease alleles worldwide especially in populations with a founder effect (Lehesjoki and Kälviäinen, 2020). In cells of patients homozygous for this mutation, the inhibitory activity of CSTB is significantly reduced with enhanced activity of cathepsins B, L, and S (Rinne et al., 2002). The other EPM1-associated mutations affect the CSTB coding sequence and mainly occur in compound heterozygosity with the repeat expansion. Of these, mutations leading to a frameshift, to aberrant splicing, or to premature termination of CSTB translation (Table 51–1) result in reduced mRNA and protein expression, or no detectable protein (Pennacchio et al., 1996; Lafrenière et al., 1997; Bespalova et al., 1997b; Kagitani-Shimono et al., 2002; Alakurtti et al., 2005; Joensuu et al., 2007; Canafoglia et al., 2012). The amino acid substitution mutations p.Gly4Arg, p.Gly50Glu, and p.Gln71Pro in the N-terminal region, in the highly conserved pentapeptide of QVVAG-motif of the first β-hairpin loop or close to the second hairpin loop of CSTB, respectively, likely affect the stability and interaction of CSTB with its target cysteine cathepsins. In cellular transfection experiments, the p.Gly4Arg, p.Gly50Glu, and p.Gln71Pro missense mutant CSTB proteins fail to associate with lysosome, implying an essential role of lysosomal association for the physiological function of CSTB (Alakurtti et al., 2005; Joensuu et al., 2007). Only two EPM1 patients without the expansion mutation have been reported. One of them is homozygous for the N-terminal p.Gly4Arg missense substitution predicting impaired inhibitory activity of the mutant protein (Lalioti et al., 1997b). The other one is homozygous for the c.66G>A variant, which results in abnormal splicing and apparent absence of CSTB protein (Pinto et al., 2012). Biallelic CSTB mutations p.Arg68* and p.His75Serfs*2, with a predicted total loss of CSTB protein, are associated with neonatal-onset encephalopathy (Fig. 51–1) (Mancini et al., 2016; O’Brien et al., 2017).
A mouse model for EPM1 has been generated by targeted disruption of the mouse Cstb gene, resulting in a predicted in-frame stop codon and a loss of the functional CSTB protein in tissues (Pennacchio et al., 1998). The mutant Cstb–/– mice in an isogenic 129Sv background recapitulate the key symptoms of EPM1 reasonably well, including development of myoclonic seizures by 1 month of age and progressive ataxia by 6 months of age (Figs. 51–1 and 51–2). The frequent myoclonic seizures occur during sleep, lasting from a few seconds to several minutes, and progress from twitching of isolated muscles to spasms, which affect the entire body. Electrocorticogram recordings reveal bilaterally synchronous 4–6 Hz repetitive spiking commencing with the myoclonus (Pennacchio et al., 1998). Contrary to EPM1 patients, no tonic-clonic seizures, photosensitivity, or spike-wave complexes on EEG have been reported. In certain mouse colonies, about one-third of Cstb–/– mice develop corneal inflammation and ocular opacity, while in others the eye phenotype is not seen at all (Pennacchio et al., 1998; personal observation). This is most likely due to differences in the mouse background strain, as the genetic background has a strong impact on the pathophysiology, implying that genetic factors influence the phenotype (Pennacchio et al., 1998). Interestingly, 17-month-old heterozygous Cstb+/– mice exhibit a mild phenotype, including seizures, ataxia, and neuron loss in cortex and cerebellum (Kaasik et al., 2007).
Pathology of Cstb–/– mouse and proposed disease mechanisms. Clinical and pathological findings in Cstb–/– mouse in relation to mouse age (E: embryonal; P: postnatal) are shown above the age axis. Below the axis are shown (more...)
The neuropathological hallmark of Cstb–/– mice is an early and progressive apoptotic loss of cerebellar granule neurons from 1 month of age (Fig. 51–2) Pennacchio et al., 1998). In addition to cerebellum, the first signs of neuronal death can also be seen in primary somatosensory cortex of 1-month-old Cstb–/– mice, followed by other cortical as well as thalamic areas, and leading to progressive volume loss from 2 months of age (Shannon et al., 2002; Tegelberg et al., 2012). In line with these findings, longitudinal MRI analysis has revealed early-onset progressive atrophy in cerebellum and cortex of Cstb–/– mice, while in striatum and hippocampus this becomes evident first at 6 months of age (Manninen et al., 2014). Diffusion tensor imaging (DTI) shows white matter changes that emerge early in cerebellum and thalamus, progress to affect most white matter tracts by 6 months of age and correspond to reduced myelin content and degenerating axons in the affected areas (Manninen et al., 2013, 2014). This correlates well with MRI and DTI findings of EPM1 patients’ brains showing loss of grey matter volume and widespread white matter degeneration (Koskenkorva et al., 2009; Manninen et al., 2013).
Brain atrophy and neuronal death is preceded by early and progressive microglial activation emerging in 2-week-old Cstb–/– mice (Fig. 51–2) (Tegelberg et al., 2012). Microglial cells show a typical activated phenotype with retracted, less ramified processes, enlarged soma, and bushy appearance in 1- to 4-month-old animals, and rod-like phenotype with thick processes from 6 months of age (Franceschetti et al., 2007; Tegelberg et al., 2012). Widespread microglial activation is most prominent in the middle and posterior thalamic nuclei, substantia nigra, claustrum, amygdaloid nuclei, and cingulate cortex (Tegelberg et al., 2012). Hippocampus, however, is largely spared from gliosis in animals up to 6 months of age (Tegelberg et al., 2012), but it shows glial activation later in the disease course (Franceschetti et al., 2007). Microgliosis is followed by astroglial activation at 1 month of age, coinciding with the first signs of neuron loss and appearance of myoclonus (Tegelberg et al., 2012). The pattern of astroglial activation mirrors that of microgliosis and progresses, especially in areas of more pronounced neuron loss.
Glial activation is reflected in upregulated expression of genes related to inflammatory processes, especially those of complement cascade, of antigen processing and presentation, and of cytokine biosynthesis, in 1-month-old Cstb–/– cerebella (Joensuu et al., 2014). Moreover, expression of inflammatory marker proteins is increased in the cortex of Cstb–/– mice (Okuneva et al., 2015). Inflammation in the central nervous system is accompanied with peripheral inflammation, manifesting as elevated levels of chemokines and pro-inflammatory cytokines in serum, as well as increased number of splenic macrophages with pro-inflammatory phenotype (Okuneva et al., 2016). Chemokines are involved in the regulation of angiogenesis (Romagnani et al., 2004). Indeed, increased brain vascularization is seen in 1-month-old Cstb–/– mice (Okuneva et al., 2016). However, although elevated numbers of peripheral immune cells have been detected in 1-month-old Cstb–/– brain, there are no signs of blood–brain barrier disruption (Okuneva et al., 2015, 2016).
Cstb–/– mice also exhibit a bone phenotype characterized by thickening and elevated bone mineral density of trabecular bone, possibly due to impaired osteoclast homeostasis (Manninen et al., 2015). This is in line with previous reports showing that CSTB is capable of inhibiting osteoclast apoptosis and modulating bone resorption in vitro through its inhibitory activity on cathepsin K (Laitala-Leinonen et al., 2006). The exact mechanisms of the role of CSTB in bone metabolism, however, are not well characterized and may extend beyond direct regulation of cathepsin K.
Proteolytic processing of histone tails is an evolutionarily conserved epigenetic mechanism that influences multiple biological processes in several organs and organisms through the regulation of gene expression and chromatin structure (Yi and Kim, 2018). A cysteine protease, cathepsin L, a known target for CSTB (Turk and Bode, 1991), has been reported to stabilize the histone modification landscape (Bulynko et al., 2006), and it has emerged as one of the key enzymes responsible for histone cleavage (Adams-Cioba et al., 2011), particularly in cell-state transitions (Duncan et al., 2008). An interaction of CSTB with cathepsin L and with histones H2A.Z, H2B, and H3 was reported in nuclei of T98G astrocytoma cells (Čeru et al., 2010). CSTB binds to histones preferentially in the G1 phase of cell cycle and delays the cell cycle progression, potentially through the regulation of the transcription factor CUX1 (Čeru et al., 2010). In line with this, Cstb–/– mouse embryonal fibroblasts (MEFs) reach the S-phase faster than wild-type MEFs, while inhibition of cathepsin activity delays the transition (Čeru et al., 2010).
CSTB has a crucial function during brain development as indicated by the observation that complete loss of CSTB causes severe, neonatal encephalopathy with progressive microcephaly in humans (Mancini et al., 2016; O’Brien et al., 2017). The key mechanisms for the developmental function of CSTB are likely to be mediated through its role as an inhibitor of cysteine cathepsins involved in regulation of cell cycle and of chromatin structure during neural stem cell renewal and differentiation (Čeru at al., 2010; Daura et al., 2021). Available evidence suggests that in mouse neural stem cells, CSTB is involved in defining the temporal window for initiation of neuronal differentiation through regulating proteolytic cleavage of histone H3 tail by cathepsins B and L (Daura et al., 2021). In undifferentiated Cstb–/– mouse self-renewing neural progenitor cells (NPCs), CSTB deficiency triggers premature histone H3 tail cleavage, which is sustained in differentiating neural cells (Daura et al., 2021). This leads to decreased self-renewing capacity of Cstb–/– NPCs (Daura et al., 2021), which is in line with the data on EPM1 patient-derived brain organoids that show decreased progenitor proliferation, increased number of immature neurons, smaller organoid size, and impaired migration of interneurons (Di Matteo et al., 2020).
Neuronal differentiation involves a metabolic transition from aerobic glycolysis used by NPCs to mitochondrial oxidative phosphorylation preferred by mature neurons. This is seen as a robust upregulation of mitochondrial biogenesis and respiratory capacity in differentiating NPCs (Agostini et al., 2016; Zheng et al., 2016). In differentiating Cstb–/– NPCs, however, the upregulation of mitochondrial respiration is delayed, together with downregulated expression of electron transport chain genes (Daura et al., 2021).
In addition to spontaneous myoclonic seizure activity (Pennacchio et al., 1998), Cstb–/– mice are more susceptible to kainate-induced epileptic seizures and seizure-induced neuronal damage compared to wild-type animals (Franceschetti et al., 2007). Also, in vitro slice preparations of Cstb–/– hippocampi show latent hyperexcitability, and when perfused with kainate, produce more pronounced epileptiform discharges (Franceschetti et al., 2007).
Several lines of evidence suggest that hyperexcitability in CSTB-deficient brain is caused by changes in GABAergic inhibition. Cstb–/– mice show altered expression of genes connected to GABAergic pathway (e.g., increased expression of genes encoding GABAA receptor subunits Gabrd and Gabra6) as early as postnatal day (P) 7. Electrophysiological recordings of P7 cerebellar Cstb–/– Purkinje cells show a shift of the balance toward decreased inhibition, as well as almost complete lack of synchronous inhibitory bursts important for the synapse development (Joensuu et al., 2014). Recent data, both in the developing mouse brain and in human cerebral organoids, suggest that CSTB is important for interneuron migration during cortical development, this function being possibly mediated through cell-non-autonomous mechanisms (Di Matteo et al., 2020).
Symptom onset in 1-month-old Cstb–/– mice coincides with a diminished number of GABAergic synaptic terminals in cerebellum and reduced ligand binding to cerebellar GABAA receptors (Joensuu et al., 2014). No change in the number of inhibitory interneurons can be seen in cerebellum at this age, and interneuron loss manifests only with more progressed pathology (Joensuu et al., 2014; personal observation). Interneuron loss is evident in 8-month-old Cstb–/– hippocampus (Franceschetti et al., 2007) and cortex, where it is accompanied with decreased GABAergic inhibition and GABA release from cortical synaptosomes (Buzzi et al., 2012). Loss of GABAergic terminals in motor cortex has also been reported in postmortem brain of a single EPM1 patient (Buzzi et al., 2012).
Neurons are highly dependent on efficient cargo transport over long distances, as well as local protein synthesis and energy production in synapses (Rossi et al., 2019). CSTB protein is detected in synaptosomal fractions from rodent brain tissue and in synaptic regions of human cerebral organoids (Penna et al., 2019; Gorski et al., 2020). Moreover, CSTB is shown to be locally translated in synaptosomes of rat cerebral cortex (Penna et al., 2019).
Gene expression profiles on cultured Cstb–/– cerebellar granule neurons suggest alterations in intracellular transport and in cytoskeleton, while in P7 Cstb–/– cerebellum changes are connected to synaptic function and plasticity (Joensuu et al., 2014). This is corroborated by quantitative mass-spectrometry-based proteomics analysis on cerebellar synaptosome preparations of presymptomatic P14 Cstb–/– mice, which show altered abundance of proteins associated with intracellular transport of proteins, organelles, and other cellular substances (Gorski et al., 2020). Moreover, a substantial number of the differentially abundant proteins in the synaptosomes are related to mRNA translation or to the ribonucleoprotein complex. As intracellular trafficking is tied to local protein synthesis and is critical for synaptic plasticity, these data suggest that in addition to being important to formation of synaptic networks, CSTB might also play a role in synaptic physiology and plasticity.
Microglia are resident immune cells of the brain that colonize the developing brain early in embryonal life. In addition to their role in immune surveillance and as resident macrophages, microglia are intimately involved in brain development and maturation as well as in maintaining homeostasis in the healthy central nervous system (Nayak et al., 2014). Microglia also directly sense and suppress neuronal activity, thus protecting brain from excessive neuronal activity (Badimon et al., 2020).
Microglial activation in Cstb–/– mice precedes astroglial activation, neuron loss, and onset of clinical symptoms (Tegelberg et al., 2012). In presymptomatic P14 animals, the activation of microglia is skewed toward neuroprotective, anti-inflammatory M2 type (Okuneva et al., 2015), probably as a reaction to altered neuronal activity in developing Cstb–/– brain. Moreover, the phagocytotic capacity of microglia is impaired in dentate gyrus of developing, P14 Cstb–/– mice, leading to accumulation of dead neurons in hippocampal formation (Sierra-Torre et al., 2020). While seizures have been shown to interfere with the microglial recognition of dead cells (Abiega et al., 2016), in Cstb–/– mice the impairment of phagocytosis precedes seizure activity. Neuronal apoptosis is, nevertheless, closely connected to local neuronal activity in Cstb–/– hippocampus, suggesting that altered local activity may change apoptotic dynamics and contribute to the impairment of microglial phagocytosis (Sierra-Torre et al., 2020).
At the symptom onset at approximately 1 month of age, microglial activation in Cstb–/– mice is polarized toward the pro-inflammatory M1 type (Okuneva et al., 2015). Similar switch is seen in other neurodegenerative diseases where persistence of inflammatory triggers leads to M1 activation (Tang and Le, 2016). Major histocompatibility complex class II (MHCII) is considered one of the markers of M1 activation and a key protein in antigen presentation. The surface expression of MHCII is, however, suppressed in Cstb–/– microglia at 1 month of age in both M1 and M2 type cells (Okuneva et al., 2015), suggesting that CSTB deficiency may affect the antigen presentation capacity of microglia. This may be due to increased cathepsin activity in cells lacking CSTB, as cathepsins are known to be involved in antigen processing, and their increased activity leads to decreased surface expression of MHCII (Kitamura et al., 2005).
In vitro studies on cultured microglia and on bone marrow–derived macrophages (BMDMs) further show that CSTB deficiency alters immune cell functions, including increased nitric oxide (NO) release, decreased anti-inflammatory IL-10 production (Maher et al., 2014a; Okuneva et al., 2015), antigen presentation, and chemotaxis (Maher et al., 2014b; Körber et al., 2016). These changes may be associated with the downregulation of interferon signaling and alterations in JAK-STAT pathway in CSTB-deficient microglia (Maher et al., 2014a, 2014b; Körber et al. 2016). Interferon signaling is an important regulator of innate immune response and autophagy, and its dysregulation is linked to a multitude of pathological conditions (McGlasson et al., 2015). In addition, interferon signaling has been reported to play a role in neuronal development, suppressing differentiation of GABAergic and promoting differentiation of glutamatergic neurons from neuronal progenitor cells (Ahn et al., 2015).
In conclusion, although microglial activation is a normal reaction to altered neuronal activity and cell death, CSTB-deficient microglia seem to be inherently dysfunctional and fail in their normal functions during development and maturity. Early changes in microglia are expected to contribute to the initiation and disease progression, especially during the important time frame between 2 weeks and 1 month of Cstb–/– mouse disease pathogenesis.
Neurons in Cstb–/– mouse brain die gradually by apoptosis during brain maturation and disease progression (Pennacchio et al., 1998; Shannon et al., 2002; Sierra-Torre et al., 2020). This leads to a loss of almost half of the cerebellar volume by 6 months of age and significant atrophy of other brain areas (Tegelberg et al., 2012). Neurons are especially sensitive to oxidative stress as brain has a high aerobic metabolic demand and relatively low antioxidant levels (Cobley et al., 2018). Oxidative stress has been linked to the pathophysiology of epilepsies and neurodegenerative diseases (Kim et al., 2015). Furthermore, increased activity during epileptic seizures exacerbates the imbalance in production and scavenging of reactive oxygen species (ROS). Epileptic activity also leads to rapid increase of Cstb expression in wide area of forebrain in the rat kindling model (D’Amato et al., 2000).
Impaired redox homeostasis is a key mechanism by which CSTB deficiency triggers neurodegeneration (Lehtinen et al., 2009). Mitochondrial respiratory chain function is the major source of ROS in the cell. In BMDMs under endotoxin-induced inflammation, CSTB translocates to mitochondria, where it protects mitochondrial membrane integrity and prevents excessive ROS production (Maher et al., 2014b). CSTB deficiency is associated with changes in mitochondrial respiratory chain function in vitro, in a model of neuronal stem cell renewal and differentiation (Daura et al., 2021). This is in line with proteome analysis of synaptosome preparations from P14 Cstb–/– mouse cerebellum that shows alterations in abundance of mitochondrial proteins with highest enrichment of proteins associated to respiratory chain (Gorski et al., 2020). The proteome analysis also shows changes in the abundance of ROS-associated proteins, for example, reduced abundance of antioxidant superoxide dismutase 1 (SOD1). Cerebella of 6-month-old Cstb–/– mice show diminished antioxidant capacity (e.g., reduced SOD activity and glutathione levels) and elevated lipid peroxidation (Lehtinen et al., 2009).
CSTB expression in cultured cerebellar granule neurons increases upon oxidative insult, suggesting a protective role for CSTB in response to oxidative stress (Lehtinen et al., 2009). CSTB deficiency leads to sensitivity to oxidative stress-induced cell death, mediated by cathepsin B, the expression of which is also increased in response to oxidative stress. Consistent with these data, Cstb-Cathepsin B double-knockout mice present with reduced granule cell death (Houseweart et al., 2003). These mice, however, still display myoclonic seizures and ataxia, suggesting that CSTB controls neuronal excitability independently of its inhibitory activity on cathepsin B (Houseweart et al., 2003).
The authors declare no relevant conflicts.
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...
