Gain in toxic function of ste
fin B EPM1 mutants aggregates: Correlation
between cell death, aggregate number/size and oxidative stress
Mira Polajnar
a,b,1, Tina Zava
šnik-Bergant
a, Nata
ša Kopitar-Jerala
a, Magda Tu
šek-Žnidarič
c, Eva
Žerovnik
a,b,⁎
aDepartment of Biochemistry, Molecular and Structural Biology, Jožef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia bJožef Stefan International Postgraduate School, Jamova 39, 1000 Ljubljana, Slovenia
c
National Institute of Biology, Večna pot 111, 1000 Ljubljana, Slovenia
a b s t r a c t
a r t i c l e i n f o
Article history:
Received 6 November 2013 Received in revised form 28 May 2014 Accepted 29 May 2014
Available online 5 June 2014 Keywords: Protein aggregate Amyloid Toxic oligomer Oxidative stress Cell death Cystatin B
EPM1 is a rare progressive myoclonus epilepsy accompanied by apoptosis in the cerebellum of patients. Muta-tions in the gene of stefin B (cystatin B) are responsible for the primary defect underlying EPM1. Taking stefin B aggregates as a model we asked what comesfirst, protein aggregation or oxidative stress, and how these two processes correlate with cell death.
We studied the aggregation in cells of the stefin B wild type, G4R mutant, and R68X fragment before (Ceru et al., 2010, Biol. Cell). The present study was performed on two more missense mutants of human stefin B, G50E and Q71P, and they similarly showed numerous aggregates upon overexpression. Mutant- and oligomer-dependent increase in oxidative stress and cell death in cells bearing aggregates was shown. On the other hand, there was no correlation between the size and number of the aggregates and cell death. We suggest that differences in toxicity of the aggregates depend on whether they are in oligomeric/protofibrillar or fibrillar form. This in turn likely depends on the mutant's 3D structure where unfolded proteins show lower toxicity. Imaging by transmission electron microscopy showed that the aggregates in cells are of different types: bigger perinuclear, surrounded by membranes and sometimes showing vesicle-like invaginations, or smaller, punctual and dispersed through-out the cytoplasm. All EPM1 mutants studied were inactive as cysteine proteases inhibitors and in this way con-tribute to loss of stefin B functions. Relevance to EPM1 disease by gain in toxic function is discussed.
© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
1. Introduction
Epilepsy is as a common and diverse set of chronic neurological dis-orders characterized by enduring propensity to generate epileptic sei-zures. Idiopathic epilepsy syndrome is defined as purely epilepsy with no clear underlying structural brain lesions. It is therefore suspected that multiple gene abnormalities may be the route of such epilepsies together with metabolic and environmental influences. Genetic epilepsies account for only about 1% of all epilepsies but they present themselves with severe symptoms and poor outcome. The most commonly mutated genes in genetic epilepsies are for protein subunits of voltage-gated (e.g. sodium-gated or calcium-gated) or ligand-gated ion channels[1]. These
mutations can cause impaired excitability of inhibitoryγ-aminobutyric acid-ergic (GABAergic) interneurons[2]or hyperexcitability offiring neu-rons[3]. In both cases, currents in the brain are disrupted.
Other mutations rather than channel-related ones have been found for certain rare epilepsies. Progressive myoclonus epilepsies (PMEs) are a group of different genetic generalized epilepsies with symptoms such as myoclonus and tonic-clonic seizures, dementia and progressive neurodegeneration of grey matter in the cerebrum and cerebellum (for reviews see[4,5]). PMEs can be divided pathogenically into two groups: non-lysosome related, such as Lafora disease, and lysosome related, such as Unverricht–Lundborg disease (progressive myoclonus epilepsy of type 1, EPM1), most common of all PMEs and in the focus of our re-search. For all types of PMEs, accumulation of proteinaceous inclusions or aggregates has been discovered[6–8].
EPM1 or Unvericht–Lundborg disease is caused by mutations in the gene encoding stefin B (cystatin B) (CSTB), primarily known as a cyste-ine protease inhibitor[9,10]. The most common change found in the stefin B gene is the dodecamer repeat expansion in the promoter region
[11]which leads to reduced mRNA and consequently reduced protein levels. The CSTB gene is located in chromosome 21q22.3 and altogether 14 different mutations of this gene have been reported to underlie EPM1[12–15]. Pathology results in a marked loss of Purkinje cells in Abbreviations: EPM1, progressive myoclonus epilepsy of type 1; TEM, transmission
electron microscopy
⁎ Corresponding author at: Department of Biochemistry, Molecular and Structural Biology, Jožef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia. Tel.: +386 1 477 3753; fax: +386 1 477 3984.
E-mail address:[email protected](E.Žerovnik).
1
Present address: Autophagy Signaling Group, Institute of Biochemistry II, Goethe University Medical School, University Hospital Building 75, Theodor-Stern-Kai 7, 60590 Frankfurt am Main, Germany and DKFZ, German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany.
http://dx.doi.org/10.1016/j.bbamcr.2014.05.018
0167-4889/© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/). Contents lists available atScienceDirect
Biochimica et Biophysica Acta
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / b b a m c rthe Purkinje layer of the cerebellum[16], cerebral atrophy, reduced cortical thickness[17]and white matter loss throughout[18].
Mutations in two genes have been reported in CSTB mutation nega-tive patients presenting with symptoms similar to EPM1. Berkovic and co-workers found a locus on the human chromosome 12, termed EPM1B. Further studies by Bassuk et al.[19]showed that mutations in the prickle-like protein 1 precursor PRICKLE1, a gene encoding a protein involved in the non-canonical Wnt signaling pathway, are the cause of the progressive myoclonus epilepsy of type 1B (EPM1B)[19]. Mutations in the lysosome membrane protein 2, SCARB2[20], have been reported by the same authors. Deciphering the signaling pathways of these proteins will likely intersect with those of stefin B and will add to our understanding of EPM1[19,21].
In most cases the EPM1 mutants produce lower expression of the stefin B protein, whose most well-known function is inhibition of the cysteine proteases—cathepsins B, H, K, L and S[22–24]. The physio-logical function of stefin B in the molecular pathogenesis of the dis-ease is not known. Loss of function signs are observed in patients, which are relatively well recapitulated in stefin B knockout (KO) mice. Although the protective role of stefin B against apoptosis was suggested[25–27], it is still unclear how the reduced protein levels could lead to myoclonus and/or the epileptic phenotype with tonic-clonic light sensitive seizures. It was suggested that the primary stefin B deficit could trigger several secondary processes, such as overstimulation of serotoninergic transmission[28]or a defect in dopamine transmission[29], which would in turn cause the seizure phenotype. It was recently shown in stefin B KO mice that there is lo-calized glial activation in brain regions with significant neural loss that precedes the appearance of myoclonus which confirms the pa-thology of glial excessive activation[30]. Genes involved in increased apoptosis and glial activation were shown to be overexpressed[31]. In accordance to other neurodegenerative conditions, abundant gliosis appears in these regions[32].
It is therefore important to clarify protein's alternative function(s). It was reported that stefin B is part of the multiprotein complex regulating cytoskeleton[33]and that the protein is involved in protection against oxidative stress[34,35]and possesses putative chaperone-like function
[36,37]. Stefin B is also known to bind copper and could thus have a role in preventing oxidative stress[38]and is proposed to play a role in au-tophagy ([7]and submitted manuscript). Its neuroprotective function was suggested due to increased expression in the forebrain neurons after seizure activity in rats[39]. Stefin B also binds to histones and indirectly regulates the cell cycle through inhibition of cathepsin L in the nucleus[40]. If such functions get lost in EPM1, the consequences could well lead to neurodegeneration.
Additionally to the loss of function that occurs, gain in toxic function of some EPM1 stefin B missense mutant aggregates may play a role, as proposed by us before[41]. Stefin B in its wild type (wt) form is an amyloidogenic protein and it has served as an appropriate model for studies of amyloid formation[42,43]. Protein misfolding and aggrega-tion are at the core of the so called conformaaggrega-tional disorders[44,45]. The conformational disorders involve protein conformational changes leading to alternatively folded (i.e. misfolded and intermediate) states, rich inβ secondary structure that have a high tendency to aggregate or forms amyloids. Neurodegenerative diseases such as Parkinson's (PD), Alzheimer's (AD), Huntington's disease (HD) and various other dementias, like amyotrophic lateral sclerosis (ALS) and prion diseases, all belong to the conformational disorders.
Intracelullar and cytoplasmic inclusions were found in the brain of one EPM1 patient with the promote extension mutation[46]. The authors reported that the inclusions consisted of the lysosomal protein cathepsin B, transmembrane protein CD68 (cluster of differ-entiation 68) or RNA-binding FUS. Related proteins were found in mass spectormetry experiments of insoluble cell fractions from stefin B knockout mice, performed in our department (submitted manuscript).
A consensus prevails that amyloidfibril formation is a generic prop-erty of proteins[47]. It was shown that even someα helical proteins such as myoglobin can form amyloid-likefibrils, undergoing α to β tran-sition. Fibril formation occurs also with proteins which are natively un-folded, intrinsically disordered[48], or are predominantly inβ sheet. Different proteins probably follow different pathways towards amyloid fibrils, depending on the structural class[44,49].
Size and morphology of the prefibrillar wt stefin B aggregates have been determined and are in agreement with other amyloid forming pro-teins[50]. Endogenous protein can form oligomers intracellularly[51]
but these prefibrillar oligomers can also be toxic to cells[50,52]. A con-sensus is reached that the mechanism of amyloid cytotoxicity most like-ly involves membrane interaction and perforation[53]. As observed for a number of amyloid forming proteins, stefin B binds to acidic phospho-lipid bilayers and is internalized through the plasma membrane where it lowers cell viability[50,52,54].
We have shown that wt stefin B, missense G4R and fragmented R68X EPM1 mutants aggregate upon overexpression in cells[55]. In this paper, additional experiments on two missense EPM1 mutants G50E and Q71P are presented, which confirm that they also aggregate to a much greater extent than the wt. By correlating size and number of the aggregates with cell death and oxidative stress, we here present data that show that smaller, punctual aggregates are more toxic to cells and that oxidative stress and cell death correlate for those EPM1 missense mutants, which increase oxidative stress most. We think that aggregate toxicity may contribute to EPM1 disease development and neurotoxicity and discuss this.
2. Materials and methods 2.1. Materials
Primers, all horseradish peroxidase-conjugated secondary antibod-ies and propidium iodide (PI) were purchased from Sigma Aldrich (St. Louis, MO, USA). LeuleuOMe was supplied by Bachem (Bubendorf, Switzerland). Lipofectamine® 2000 was from Invitrogen (Carlsbad, CA, USA). Pfu DNA polymerase for the polymerase chain reaction (PCR) was purchased from Fermentas (Vilnius, Lithuania). AnnexinV– PE (phycoerythrin) was purchased from BD Biosciences (Franklin Lakes, NJ, USA). Gelatine and sucrose were from Merck (Whitehouse Station, NJ, USA). Uranyl acetate was purchased at SPI Supplies (West Chester, PA, USA). Protease inhibitor cocktail tablets (25×) were provid-ed by Roche (Basel, Switzerland). Pierce® BCA protein assay kit was purchased from Thermo Fisher Scientific (Waltham, MA, USA).
For cell culture experiments, Dulbecco's modified Eagle's medium withL-glutamine, Ham's F12, fetal bovine serum (FBS), penicillin/
streptomycin (100 ×), 1 × Dulbecco's phosphate buffer saline (PBS), glutaMAXTM (all PAA Laboratories GmbH, Pasching, Austria) and TrypLE™ Select (Gibco, Invitrogen, Carlsbad, CA, USA) were used. Bodipy 581/591 C11, MitoSOXTMRed, and Prolong Antifade reagent were from
Molecular Probes (Invitrogen, Carlsbad, CA, USA). Polyclonal antibodies for stefin B and PCNA were purchased from Abcam (Cambridge, UK) and Santa Cruz (Dallas, TX, USA), respectively.
2.2. Cloning
Mutagenesis was performed by PCR on a 2720 Thermo Cycler PCR System (Applied Biosystems, Foster City, CA, USA) using a QuickChange® II Site-Directed Mutagenesis kit (Stratagene, Agilent Technologies, Inc., Santa Clara, CA, USA) and Pfu DNA polymerase. Mutagenesis was performed using a template with complementary DNA gene for human stefin B (C3E31) (NM_000100.2) with or without a T-Sapphire tag on the C terminus[55]. For the R68X mutant chemically synthesized gene was inserted via BamHI and XhoI restriction sites in a pcDNA3 vector (Invitrogen, Carlsbad, CA, USA).
The following primers were used for the construction of mutants: 5′-CGAGGTCATGATGTGCCGGGCGCCCTCCGCCAC-3′ and
5′-GTGGCGGAGGGCGCCCGGCACATCATGACCTCG-3′ (G4R mutant); 5′-AGCCAGGTGGTCGCGGAGACAAACTACTTCATC-3′ and
5′-GATGAAGTAGTTTGTCTCCGCGACCACCTGGCT-3′ (G50E mutant); 5′-CACCTGCGAGTGTTCCCATCTCTCCCTCATGAA-3′ and
5′-TTCATGAGGGAGAGATGGGAACACTCGCAGGTG-3′ (Q71P mutant). DNA sequences were confirmed using Sanger sequencing (BigDyeTM
terminator kit) and resolved with Automatic Sequencer 3730XL (Applied Biosystems, Foster City, CA, USA) in Macrogen (Rockville, MD, USA).
2.3. Expression in Escherichia coli (E. coli) and purification procedure Recombinant stefin B wt and mutants were produced in E. coli and purified according to published procedures[56,57].
2.4. BANA test
To evaluate the inhibitory activity of stefin B, BANA test was per-formed. Stefin B monomers, dimers, tetramers and oligomers were di-luted in BANA buffer (0.1 M phosphate buffer, 1.5 mM EDTA, pH 6.0) so A280was 0.5. Papain was diluted in the same buffer to 0.02 mg/ml
(0.5μM). Eight different molar ratios [E]:[I] were prepared: 1:45, 1:22, 1:11, 1:4, 1:2, 1:0.2. Firstly, papain was activated with 5 mM cysteine for 5 min at 37 °C. Next, BANA substrate was added to the reaction mixture in 2 mMfinal concentration and incubated for 10 min at 37 °C. The reaction was stopped with a stop reagent (1 volume of re-agent III:1 volume of color rere-agent) and incubated at room temperature. Absorbance was measured at 520 nm on a Lambda 18 UV/VIS spectrom-eter (Perkin-Elmer, Waltham, MA, USA). Reagent III consisted of 10 mM p-chloromercurybenzoic acid and 50 mM EDTA at pH 6.0. Color reagent consisted of 3 mM Fast Garnet GBC salt in 4% Brij 35, pH 6.0.
2.5. Fibrillation kinetics
ThT dye was used to determine the presence of amyloid-likefibrils. Fluorescence at 482 nm (upon excitation at 440 nm) was measured using a LS 50 B luminescence spectrometer (Perkin-Elmer, Waltham, MA, USA). Samples were prepared as described in[57].
2.6. Cell culture and stable and transient transfections
HEK293T (human embryonic kidney) cells were grown in DMEM medium, SH-SY5Y (neuroblastoma) cells in DMEM/HAM'S F12 medium (1:1) supplemented with 1% (v/v) of non-essential amino acids (NEAA) and CHO (Chinese hamster ovarian) in HAM'S F12 medium. All media were additionally supplemented with 1% (v/v) GlutaMAXTM, 10% (v/v)
FBS and 1% (v/v) penicillin/streptomycin. Cells were grown in Petri dishes at 37 °C in 5% CO2. Cells grown on 15 mm coverslips (SH-SY5Y)
or in 10 cm petri dishes (CHO) were transiently transfected using Lipo-fectamine® 2000, according to the manufacturer's instructions. For sta-ble transfection, transfected SH-SY5Y cells were transferred to selective medium containing 0.2μg/ml geneticin 48 h after transfection. Cells were fed with selective medium every 3–4 days until geneticin-resistant foci could be identified. Afterwards, cells were diluted at least twice (1:3) and grown from a heterogeneous population. 2.7. Cell lysates
For protein expression analysis, HEK293T were lysed after 24 h after transfection and overexpression. Radioimmunoprecipitation assay (RIPA) buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 0,1% SDS, 1%
deoxycholate, 1% Triton X-100, 5 mM EDTA, and 1× protease inhibitor cocktail) was used. Cells were scrapped and left on ice for 30 min. Cell lysates were clarified by centrifugation at 10000 × g for 20 min at 4 °C. Protein concentrations of the supernatants were determined by BCA assay and the absorbance was measured at 562 nm using a spectrophotometer.
2.8. Fluorescence microscopy
Samples forfluorescent imaging of lipid peroxidation were observed with an Olympus IX81 microscope (Olympus, Center Valley, PA, USA) and 60× oil objective. Images were recorded by ORCA-R2 camera (Ha-mamatsu Photonics, Ha(Ha-mamatsu City, Japan), using Cell F software. Green and red emission werefiltered using U-N49002 and U-N41002 filter cubes (Chroma technology) and fluorescence signals were separately recorded with the ORCA-R2 camera. Confocal images were taken at the Center of Excellence in Nanoscience and Nanotechnology – CO Nanocenter, Ljubljana, Slovenia. Leica TCS SP5 X confocal micro-scope (Leica Microsystems GmbH, Wetzlar, Germany) was used for op-tical slicing (60× oil objective, NA = 1.4). T-Sapphire was excited at 488 nm with white light laser (470–670 nm) and the emission signal was re-corded using a photomultiplier detector (PMT). At the same time differ-ential interference contrast (DIC) images were taken. Leica LAS AF software (Leica Microsystems GmbH, Wetzlar, Germany) was used to process the images.
2.9. Transmission electron microscopy (TEM)
TEM images of protein samples were taken at the National Institute of Biology, Ljubljana, Slovenia. Samples were observed with a Philips CM 100 transmission electron microscope (Philips/FEI Corporation, Eindhoven, Holland) operating at 80 kV and prepared as described pre-viously[57]. The thickness offibrils was measured in Digital Micrograph software (Gatan, Pleasanton, CA, USA) that was beforehand calibrated according to manufacturer's instructions.
Cell samples werefixed with 4% paraformaldehyde at room temper-ature for 2 h. Mixture of growth medium andfixative was removed and the pellet embedded in 10% gelatin. Gelatine was solidified on ice and cells excised into small blocks, later on immersed in 2.6 M sucrose in 0.1 M phosphate buffer and infiltrated at 4 ºC overnight. Cryoprotected cells were mounted on specimen carriers and stored in liquid nitrogen. Trimming (at−90 °C) and sectioning (−120 °C) of frozen blocks were performed with Leica EM FC6 ultramicrotome (Leica Microsystems GmbH, Wetzlar, Germany) for cryosectioning. Ultrathin cryosections (70 nm) were cut using diamond cryo-immuno knife (Diatome, Biel, Switzerland). Sections were collected and thawed on droplets of pick-up solution (1:1 mixture of 2.3 M sucrose and 2% methyl cellulose). Re-trieved ultrathin sections of cells were put on carbon-coated Formvar film spread over copper grids (Agar Scientific, Elektron Technology UK Ltd., Essex, UK). For immunogold labeling sections were incubated with polyclonal rabbit anti-human stefin B antibody in blocking solution (2% BSA in PBS) for 30 min. PBS was used for extensive rinsing after each incubation step. Sections were further incubated with Protein A-gold (10 nm) for 20 min according to the producer's instructions (Cell Mi-croscopy Centre, University Medical Centre Utrecht). Controls omitting the primary antibody were also performed. All immunoreagents were centrifuged at 16,000 × g (at 4 °C) prior to their use with cell sections. Sections were contrasted with 0.4% uranyl acetate in solution with 1.8% methyl cellulose (pH 4). Uranyl contrasted the cell proteins but not the lipids so the membranes appeared white on TEM micrographs. Contrasted cells were observed at 150 kV with JEM-2100 LaB6 transmis-sion electron microscope (Jeol Ltd., Tokyo, Japan) at the Centre for Elec-tron Microscopy, Jožef Stefan Institute, Ljubljana, Slovenia. Micrographs were taken using Gatan digital camera and Digital Micrograph software.
2.10. Flow cytometry
Stably transfected SH-SY5Y cell were analyzed byflow cytometry on FACSCaliburTM (BD Biosciences, Franklin Lakes, NJ, USA) with CellQuest software (version 3.3). At least 10000 cells were acquired for each measurement with three or four parallels and experiments were inde-pendently performed three times. The rest of the method was the same as described[55].
2.11. Caspase-3-like activity
Caspase-3-like activity was measured by the DEVDase assay, in which Ac-DEVD-AFC (Bachem, Bubendorf, Switzerland) is afluorogenic substrate cleaved by caspase-3 but also by caspases-6, 7, 8 and 10[58]. Cell lysates were prepared in caspase lysis buffer (50 mM HEPES, 200 mM NaCl, 10% saccharose, 0.1% CHAPS, 5 mM MgCl2, 0.02% BSA,
1% NP-40, 0.5% Triton X-100). Cell lysates and DEVD substrate were di-luted in caspase buffer (same as caspase lysis buffer but without deter-gents NP-40 and Triton X-100) and separately incubated at 37 °C for 10 min. After the incubation time DEVD was added to the cell lysates. Fluorescence was measured in an automatic plate reader with excita-tion wavelength at 400 nm and emission wavelength at 505 nm. For positive control, cells were incubated with 5 mM LeuLeuOMe for 12 h. 2.12. Oxidative stress
Oxidative stress was determined by MitoSOX Red (5μM, 30 min) and Bodipy 581/591 C11(2μM, 30 min) in stably transfected cells.
MitoSOX is oxidized by superoxide anion and exhibits redfluorescence. Bodipy 581/591 C11exhibits a shift influorescence emission peak from
~590 to ~510 nm after interaction with peroxyl radicals. This increase or shift in fluorescence can be detected by flow cytometry. Stably transfected SHSY5Y cells were washed with warm PBS, loaded with Bodipy 581/591 C11or MitoSOX in PBS for 30 min at 37 °C. Cells were
harvested, pelleted and resuspended in PBS and subjected to FACS anal-ysis. Dyes were excited using a 488 nm Ar laser and detected with the FL2 (564–606 nm) or FL1 (515–545 nm) detector, respectively. At least 7000 cells were acquired for each measurement with three or four parallels and experiments were independently performed three times. Bodipy 581/591 C11was additionally tested under the
fluores-cence microscope. 2.13. Statistical analyses
For statistical analyses Office Analysis ToolPak was used employing Student's t-test: two sample assuming unequal variances. Graphs repre-sent the mean and s.e.m. of at least three independent experiments. 3. Results
3.1. Inhibitory activity and kinetics of amyloidfibril formation
During isolation and purification, the G50E and Q71P mutant were tested for inhibitory activity with a BANA assay where they exhibited approximately 10-fold lower activity compared to the wt (Supplemen-tary Fig. 1).
The time course of amyloidfibril formation was studied by measur-ing ThTfluorescence versus time. In the presence of 9% of TFE, pH 4.8 and at room temperature, the lag phase for the wt stefin B was at around 50 h; meanwhile, both G50E and Q71P mutants did not show any lag phase and started to aggregate swiftly (Fig. 1). Soon after the lag phase,fibrils prevailed for the wt, while for the mutants, the fibrils were short (Q71P) or not yet present (G50E). Thefinal yield of fibrils (judged by ThTfluorescence intensity) was also lower for the mutants than for the wt stefin B.
Fibril formation was followed in parallel by TEM. Four samples of wt stefin B and the G50E and Q71P mutants were taken for TEM analysis within approximately thefirst 24 h and one after 2 days. Three samples were taken on the 5th, 6th and on the 9th day when the mutants already reached the plateau phase, while the wt stefin B was still increasing in ThTfluorescence. The last sample was taken after more than 2 weeks (16 days) offibrillation when the reactions reached the plateau phase (Fig. 1). Only prefibrillar aggregates were observed in the first 24 h in the case of G50E mutant (Fig. 2B), whereas in the case of wt stefin B and Q71P mutant (Fig. 2A and C), somefibrils had already grown. All samples had similar ThTfluorescence at this point. The difference be-tween these proteins persisted throughout the plateau phase where only maturefibrils were present in wt stefin B, whereas in the G50E and Q71P mutant samples, some aggregates remained in addition to short, somewhat stumpy lookingfibrils (protofibrils). The protofibrils of mutants were thicker in diameter compared to the wt. For both mutants the thickness of protofibrils ranged between 7 and 11 nm. The average value of G50E protofibrils was 8.68 ± 1.38 nm and of the Q71P mutant 9.35 ± 0.82 nm. On the other hand, the average value of wtfibrils was 6.0 ± 0.78 nm ranging between 4.5 and 7.5 nm.
3.2. Formation of stefin B aggregates in cells
Stefin B and chosen EPM1 mutants G50E, Q71P, G4R and R68X were tagged with T-Sapphire and overexpressed in HEK293T cells for 24 h. At least 600 (3 × 200 cells) cells with confirmed protein expression were counted and checked for the aggregate formation. Wt protein formed very small aggregates/oligomers in only about 5% of all counted cells (Figs. 3 and 4A). Similar small aggregates were formed by the G4R mu-tant, which were however more abundant and found in about 20% of all counted cells. The biggest aggregates were found in cells transfected with R68X, G50E and Q71P mutants. The latter two mutants formed the highest amount of aggregates (in 65–70% of cells), while R68X formed aggregates in half of the cells (50%).Table 1gathers these results as the amount (%) of cells with the aggregates.
Cells that overexpressed the three most aggregate prone mutants, i.e. G50E, Q71P and R68X, were also observed to be of round shape and with fragmented nucleus that indicated toxicity of these aggregates. InFig. 4A, confocalfluorescence images of the wt and EPM1 mutants G50E and Q71P are shown. No clear expression of G50E and Q71P in Fig. 1. ThTfluorescence emission intensity at 482 nm plotted as a function of time. Fibrilla-tion reacFibrilla-tion of the wt stefin B (stars), G50E (blank squares) and Q71P (black diamonds) mutant. In all cases, protein concentration was 34μM. Fibrils were grown in the pH 4.8 ac-etate buffer in the presence of 9% TFE and were diluted into ThT buffer at different times of fibril growth. Arrows mark different time points when samples were taken for TEM analy-sis. Results of one representative experiment are shown.
the nucleus can be detected, in contrast to wt stefin B. Wt protein is uni-formly distributed through the cytoplasm and the nucleus. Cells remained sedentary with preserved extended shape. No large aggre-gates were present. Expression of the G50E and Q71P mutant was ob-served mostly around the nucleus, overtaking a large part of the
cytoplasm, especially towards the plasma membrane. Expression levels of the wt and the G50E and Q71P mutants upon overexpression (tagged and untagged forms) were monitored by immunoblotting (Fig. 4B). En-dogenous stefin B levels were not changed upon overexpression and the level of tagged stefin B forms was lower compared to the endogenous Fig. 2. TEM images in the lag and plateau phases of thefibrillation of the wt stefin B (A), G50E mutant (B), and Q71P mutant (C). Samples were taken from the fibrillation reaction at different time points as marked on the arrows. Protein concentration was 34μM. The fibrils differed by length and thickness. The average value of wt fibrils was 6 ± 0.78 nm (ranging between 4.5 and 7.5 nm). The G50E and the Q71P formed shorter protofibrils with a diameter ranging between 7 and 11 nm, namely 8.68 ± 1.38 and 9.35 ± 0.82 nm, respectively.
protein. Due to polyclonal antibodies, some unspecific binding can be observed with mock and untagged stefin B samples.
3.3. Cell viability
Cell viability of stably transfected SH-SY5Y cells was tested byflow cytometry (Figs. 5 and 6). Interestingly, cell viability was the lowest for two EPM1 mutants that form differently sized aggregates. Even though we noticed round shape of cells with abundant aggregates of the G50E, Q71P and R68X mutants,flow cytometry showed that only the G50E mutant, which produced the highest amount of big aggre-gates, lowered cell viability to approximately 50% as well as the G4R mutant, which produced the lowest amount of scattered and diffuse ag-gregates (Table 1).
This observation correlates with the fact that smaller aggregates of G4R mutant may be more toxic to cells. The G4R isolated oligomers in the range of octamers to 12-mers were already shown to be toxic and to perforate artificial membranes[54]. In the same paper, wt stefin B oligomers were also shown to be toxic. Our results confirm that although wt does not produce large aggregates, overexpression of the Fig. 3. Number of cells with stefin B aggregates. HEK293T cells were transfected with wt
and mutated variants of stefin B. After 24 h of overexpression, at least 600 cells with visible protein expression were counted. Estimation of cells with aggregates was made only by visual inspection. No distinction between large or small aggregates was made.
Fig. 4. (A) Confocalfluorescence microscopy and differential interference contrast (DIC) microscopy of aggregates in HEK293T cells. All proteins were tagged with T-Sapphire at the C-terminus. Samples were observed with a TCS SP5 X confocal microscope and 60 × oil objective. (B) Protein levels of wt and the G50E and Q71P mutant upon overexpression in HEK293T were monitored by immunoblotting.
wt protein causes its oligomerization that leads to gain in toxic function. On average, only 65% of cells survived the overexpression of the wt protein. The percentage of live cells was similar for the cells with overexpressed mutated proteins R68X and Q71P, which produced big perinuclear aggregates.
Cells were PI and AnnexinV positive or only PI positive, meaning they could be either in late apoptotic or in necrotic stage, respectively (Fig. 6). However, no caspase-3-like activity was detected (Supplemen-tary Fig. 2) meaning that cells were dying by caspase-independent cell death, most likely by necrosis. Our results thus show that the amount of aggregates does not correlate with cell death but it is most probably the characteristics of the oligomers/aggregates that define the cell fate and direct it to the cell death pathway. We later discuss the morphology of the aggregates as observed by TEM.
3.4. Measuring oxidative stress
Oxidative stress was measured by Bodipy 581/591 C11and MitoSOX
Red in SH-SY5Y cells. With Bodipy 581/591 C11oxidative stress can be
measured indirectly by determining peroxidation of lipids. When re-duced, this dye exhibits redfluorescence and when oxidized, its emission peak shifts to the green spectrum. MitoSOX specifically targets superoxide anions, which are the predominant ROS (reactive oxygen species) gener-ated in the mitochondria, and exhibits redfluorescence when oxidized.
Firstly, SH-SY5Y cell were transiently transfected with different con-structs and dyed with Bodipy 581/591 C11after 24 h of overexpression.
Microscopy imaging showed whole cell or only localized sites of lipid peroxidation (Fig. 7). Secondly, the results were validated with stably
transfected SH-SY5Y cell. Thefluorescence intensity of Bodipy 581/591 C11was measured byflow cytometry. Flow cytometry measurements
showed a correlation between cell viability (Figs. 5 and 6) and oxidative stress (Figs. 8 and 9) for certain EPM1 missense mutants. As gathered in
Fig. 9, surprisingly, overexpression of the wt protein caused one of the highest increases in lipid peroxidation but this did not result in massive cell death—at least not compared to the G50E and G4R mutants who showed the highest amount of lipid peroxidation (Fig. 9) but also the highest cell death (Fig. 5). Q71P and R68X mutants did not exhibit rele-vant shifts influorescence (Fig. 8), and thus did not produce additional ROS formation. With MitoSOX Red, no significant increase in fluores-cence was detected, meaning that stefin B aggregates do not significantly influence or damage mitochondrial membranes (Supplementary Figs. 3 and 4).
3.5. TEM images of stefin B aggregates in transfected cells
CHO cells were used to study the characteristics of aggregates be-cause they do not contain endogenous stefin B protein. No long fibrils of transfected stefin B mutants, resembling the fibrils formed during ki-netic studies in vitro in mildly acidic environment (Fig. 2), were found in transfected CHO cells (Fig. 10). Nevertheless, upon overexpression of Q71P (Fig. 10A) and G50E (Fig. 10B), a noticeable protein aggregation that was morphologically distinguished from adjacent cytoplasm due to heavy uranyl acetate binding to the aggregated protein was observed compared to non-transfected control sample (not shown).
The spherical aggregate of Q71P (Fig. 10A) shows more than 500 nm in diameter whereas the unevenly shaped aggregates of G50E (Fig. 10B) are of smaller size (200–400 nm). Since uranyl acetate contrasts all teins (but not the lipid membranes) we cannot exclude that the pro-teins within the aggregates were only stefin B mutants or there might be some other proteins in addition. Still, immunogold labeling with anti-human stefin B antibody clearly confirmed the localization and high content of O71P (Fig. 10A) as well as G50E (Fig. 10B) within the immunogold labeled aggregates.
The amorphous structure of the labeled aggregates (Fig. 10A and B) is not homogenous and it does not look that the aggregates would be surrounded by a membrane. Another type of aggregates formed by the R68X mutant was found to be surrounded by membranes and to partially colocalize with microtubule-associated protein light chain 3 (LC3). LC3 is an autophagy marker indicated that these aggregates are targeted for autophagic degradation[55].
With these data (Fig. 10) we can further confirm that the aggrega-tion of stefin B mutants Q71P and G50E (Fig. 4A) was not associated with thefluorophore tag but rather resulted from the stefin B EPM1 mu-tants' proneness to form oligomers/aggregates.
4. Discussion
In vitro characterization of EPM1 mutants G50E and Q71P has par-tially already been studied and it has been shown that both mutants possess similar secondary structure than the wt, whereas the tertiary structure is partially unfolded. This especially holds true for the Q71P mutant that has increased hydrophobic surfaces[56]. The G4R mutant, albeit inactive like the G50E and Q71P mutants (Supplementary Fig. 1), has the same secondary and tertiary structure as the wt and is of similar stability than the wt[57]. The R68X fragment is on the other hand completely unfolded and R68X forms amyloid-fibrils very fast even at neutral pH and under mild conditions[57]. The G4R mutant exhibits a prolonged lag phase compared to the wt, during which prefibrillar ag-gregates accumulate[57].
Both of the newly studied missense mutants G50E and Q71P fibril-lated in vitro in the slightly acidic environment with the addition of 9% TFE (Figs. 1 and 2B and C), as was previously shown for the wt
[57]. Thefibrillation kinetics was similar for both mutants, without a lag phase. The reaction tofibrils was not complete; it ended as a mixture Table 1
Characteristics of protein aggregates by expressed proteins. Characteristics/
Protein
Number of cells with aggregates
Size of aggregates Cell death Oxidative stress WT * * ** *** G4R * * *** *** G50E *** *** *** *** R68X ** *** ** – Q71P *** *** ** *
Table 1 sums up counted number of cells bearing labeled protein aggregates, an estimate of their size and their effects on cell viability and oxidative stress, respectively; comparison is done for the wild type (WT) stefin B and its three missense and one truncated EPM1 mu-tants. Protein aggregates tagged with greenfluorescent protein (GFP) were observed in HEK293T cells. Cell death and oxidative stress, i.e., lipid peroxidation, were measured in stably transfected SH-SY5Y cells. Cell death was estimated byflow cytometry. Legend: *low/small, **medium, ***big/high.
Fig. 5. Bar graph presentation of cell viability of stably transfected SH-SY5Y cells. Cell viability was measured onflow cytometer using annexin V and propidium iodide (PI) dyes. Control = untransfected cells, plasmid = cells transfected with an empty pcDNA3 plasmid. Student's t-test, ***pb 0.01.
of granular aggregates, from which growth of newly formedfibrils could be observed. This also could explain the low ThTfluorescence compared to the wt in thefinal plateau phase (Fig. 1). The reaction ended with protofibrils, which were thicker than the wt fibrils. In more details: the protofibrils of the mutants were shorter, thicker and stumpy looking
(around 9 nm in diameter) (Fig. 2B and C) compared to the wt that formed thinner (around 6 nm), but denser and longerfibrils (Fig. 2A).
Evidence exists that soluble, prefibrillar aggregates are more cytotoxic than mature and insoluble amyloidfibrils. Moreover, fibrils (stuck in the aggresomes) or amyloid plaques are a protective Fig. 6. Annexin V and PIfluorescence measurements of cell viability by flow cytometry on stably transfected SH-SY5Y cells. Live cells are annexin V and PI negative, thus in lower left quadrant. Necrotic and/or late apoptotic cells appear in the upper two quadrants. All experiments were repeated three times. The representative results of one experiment are shown. Control = untransfected cells; plasmid control = cells transfected with an empty pcDNA3 plasmid.
Fig. 7. Fluorescence images of SH-SY5Y cells stained with Bodipy 581/591 C11. Cells were transiently transfected with different stefin B mutants and after 24 h stained with Bodipy 581/591
C11(2μM, 30 min). A: control, non-transfected cells, B: cells transfected with an empty pcDNA3 plasmid, C: wt, D: G4R mutant, E: G50E mutant, F: R68X mutant, G: Q71P mutant, H:
positive control (1 mM H2O2, 30 min). Green and red emission werefiltered using U-N49002 and U-N41002 filter cubes and merged images of two fluorescence signals are shown.
mechanism[59]. The incomplete in vitrofibrillation of the G50E and Q71P mutants could thus indicate potential cytotoxic effects. Cellular studies indeed confirmed a higher cytotoxicity, especially so of the G50E (see, three paragraphs below).
In our previous cellular studies, G4R formed small and diffuse aggre-gates while aggreaggre-gates of the R68X mutant were bigger, vesicular and perinuclear[55]. In order to study G50E and Q71P mutants, HEK293T and SH-SY5Y cells were used for transfection and the effects of protein aggregation on cells were observed. Both G50E and Q71P mutants formed big perinuclear aggregates compared to the wt which formed few and very little aggregates upon overexpression in HEK293T cells (Figs. 3 and 4A andTable 1). The R68X mutant was expressed alongside for comparison, knowing that it is inactive, without any tertiary struc-ture and that it undergoesfibril formation very quickly. The fourth EPM1 mutant, G4R, was used as another comparison. It also loses its in-hibitory activity but it keeps its tertiary structure complete[57].
Imaging by confocal microscopy of EPM1 mutants G50E and Q71P in comparison to the wt (Fig. 4A) showed that the aggregates in cells are of two types. Bigger and perinuclear aggregates of G50E, Q71P (previously also observed for R68X [55]), or smaller, punctual and dispersed throughout the cytoplasm similar as for wt and G4R proteins[55].
TEM images (Fig. 10) of the aggregates immunogold labeled with specific anti-stefin B antibody and protein A-gold confirmed that non-tagged Q71P and G50E mutants form rather big cytoplasmic aggregates; however, the shown examples are likely not aggresomes as they do not localize in vicinity of the nucleus and lipid membranes cannot be clearly
observed. Interestingly, in addition to gold labeled aggregates of the Q71P mutant in the cytoplasm, no significant number of gold particles was observed inside large vesicular organelles with abundant internal membranes (Supplementary Fig. 5), most probably representing a population of lysosome-like vesicles.
For the purpose of measuring cell viability, SH-SY5Y cells were stably transfected with non-tagged wt stefin B and its four EPM1 mutants. Flow cytometry results showed that cell viability decreased in all cells compared to the controls (Figs. 5 and 6) and was the lowest (~ 50%) for the G50E and G4R mutants. The Q71P and R68X mutants reduced the cell viability to about 70%. Thus, cell viability did not correlate with the size and number of the aggregates in cells. We hypothesize that this is due to the exposed hydrophobic surfaces (Q71P) and the complete loss of the tertiary structure (R68X), which implies that these mutants aggregate very quickly and cause less damage than the aggregates that aggregate slowly (G50E) or form small cytotoxic oligomers (G4R)[54]. Interestingly, overexpression of the wt protein also reduced cell viability. This could be due to the formation of small oligomers (but higher than the tetramers which are not toxic) that were shown to be cytotoxic and to perforate membranes[50,54].
Cells were either only PI positive or Annexin V and PI positive, mean-ing that caspase-independent cell death (e.g. necrosis) was the predom-inant cell death pathway. This was confirmed by testing DEVDase activity, a marker for apoptotic cell death (Supplementary Fig. 2). We think that necrotic cell death could be the consequence of increased ox-idative stress. There is no consensus on what is the predominant cell death pathway in neurodegeneration; however, calpain and cathepsin activation may have an important role in necrotic neuronal cell death
[6].
Oxidative stress has long been linked to neuronal cell death in neu-rodegenerative diseases. For some time it was unclear whether it is a primary cause or merely a downstream consequence of the neurode-generative process (for review see[60]); however, a consensus was reached in the last years that excess ROS generation appears in most neurodegenerative diseases before cell death[61,62]. Furthermore, sen-sitivity to oxidative stress has been connected to stefin B deficiency in a neuronal[34]and breast cancer cell model[35].
To detect oxidative stress, stefin B mutants and the wt were overexpressed in SH-SY5Y cells and Bodipy 581/591 C11dye and MitSOX
Red were used as probes. MitoSOX Red specifically targets mitochondrial superoxide anions; however, no ROS originating from the mitochondria were detected (Supplementary Figs. 3 and 4). Aggregates and oligomers thus did not affect or perforate mitochondrial membranes.
Bodipy 581/591 C11is afluorescent probe for labeling lipid
peroxida-tion. SH-SY5Y cells with transiently overexpressed stefin B mutants were observed byfluorescence microscope. Images showed mostly lo-calized increase in ROS generation (Fig. 7). This increase happened within thefirst 24 h of protein overexpression and aggregate formation. Further,flow cytometry measurements on cells with stable protein Fig. 8. Shift in Bodipy 581/591 C11fluorescence measured by flow cytometer. Bodipy 581/591 C11was excited using a 488 nm Ar laser and detected on a FL1 (515–545 nm) photodetector
onflow cytometer. A: negative versus positive control; B: negative versus plasmid control; C: plasmid versus positive control; D: plasmid control versus wt; E: plasmid control versus G4R; F: plasmid versus G50E; G: plasmid versus R68X; H: plasmid versus Q71P. Negative control = untransfected cells; plasmid = cells transfected with an empty pcDNA3 plasmid; positive control = untransfected cell treated with 1 mM H2O2for 30 min. All experiments were repeated three times. The representative results of one experiment are shown.
Fig. 9. Fluorescence intensity of Bodipy 581/591 C11(mean) measured byflow cytometry
on stably transfected SH-SY5Y cells. Bodipy 581/591 C11was excited using a 488 nm Ar
laser and detected on a FL1 photodetector (515–545 nm) on flow cytometer. Plasmid = cells transfected with an empty pcDNA3 plasmid, negative = untransfected cells, positive = untransfected cells treated with 1 mM H2O2for 30 min. Student's t-test,
overexpression showed increasedfluorescence intensity for G50E and G4R mutants (Figs. 8 and 9) that also gave the lowest cell viability (Fig. 5). The amount of lipid peroxidation in R68X and Q71P expressing cells statistically did not differ from the control group, although Q71P mutant had a high statistical error of the measurement and nofirm con-clusion could be met. Interestingly, overexpression of the wt stefin B protein caused the highest amount of lipid peroxidation (Fig. 9) that also reduced cell viability but not as much as the expression of the G4R and G50E mutants (Fig. 5). The results show that even though wt stefin B oligomers have been shown to be cytotoxic[50]and apparently increase oxidative stress, its oligomers are not as cytotoxic as the ones from the G4R and G50E mutants. Also, the protective mechanisms that normally rescue cells from cell death comprising the wt stefin B function must be still in place, thus preventing cell death.
One should be aware that overexpression of stefin B gene and pro-tein happens in vivo by various insults, among them epileptic seizures. That stefin B may be protective in those conditions is possible; however, it may also contribute to more damage by increasing oxidative stress. On the other hand, stefin B KO animals have increased neuronal cell death of granule cells in the cerebellum[63]and KO cells show in-creased sensitivity to ROS[34,35]. The balanced expression of this pro-tein and its proper localization and function in signaling must all be understood to explain the beneficial effects of stefin B. The still unre-solved alternative function in the nucleus will also help to understand its cellular role in health and disease.
When mutations of the transcript region appear in EPM1 it is usually only on one allele while on the other allele the most common dodecamer repeat expansion in the promoter region is found. Even though the pathological mutants get expressed to only about 15–35%
[12]we think that the formation of the aggregates can still modulate the pathology. Specifically for the G50E protein, it was shown to have a relatively high expression in EPM1 patients[64]. However, it is of note that patients bearing the R68X mutation on one of the alleles show more severe pathology and clinical symptoms[65]. Thesefindings thus leave an open question, whether the bigger aggregates that can se-quester other proteins from the cytoplasm or oligomers/smaller aggre-gates scattered in the cytoplasm are more damaging.
In conclusion: 1. The protein aggregates of the EPM1 mutants of human stefin B can be taken as a model for protein aggregation in cells and their degradation and toxic effects may be of general interest. 2. No clear correlation with the size or number of aggregates and cell death was shown, whereas correlation to oxidative stress was found. We propose that the smaller and more diffuse, scattered aggregates throughout the cytoplasm of the wt protein, G50E and G4R mutant (which are all folded proteins and oligomer-prone) may bind to mem-branes and cause increase in ROS that could lead to necrotic cell death. 3. Although the expression of the missense EPM1 mutants in patients
is not high (up to 40% of the normal protein level), the aggregates cyto-toxicity may contribute to EPM1 disease. We propose that in addition to loss of stefin B normal function, gain in toxic function of the aggregates may occur.
Acknowledgements
The work was supported by the project J7-4050 (led by E.Ž.) and by the program P1-0140 (led by B. Turk)financed by the Ministry for Science and Technology of the Republic of Slovenia via the Slovenian Research Agency (ARRS). We thank Dr Oliver Griesbeck (Max Planck Institute of Neurobiology, Martinsried, Germany) for providing the T-Sapphire variant GFP plasmid.
Appendix A. Supplementary data
Supplementary data to this article can be found online athttp://dx. doi.org/10.1016/j.bbamcr.2014.05.018.
References
[1] M.H. Meisler, J.A. Kearney, Sodium channel mutations in epilepsy and other neuro-logical disorders, J. Clin. Invest. 115 (2005) 2010–2017.
[2] T.H. Rhodes, C. Lossin, C.G. Vanoye, D.W. Wang, A.L. George Jr., Noninactivating voltage-gated sodium channels in severe myoclonic epilepsy of infancy, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 11147–11152.
[3] A. Escayg, M. De Waard, D.D. Lee, D. Bichet, P. Wolf, T. Mayer, J. Johnston, R. Baloh, T. Sander, M.H. Meisler, Coding and noncoding variation of the human calcium-channel beta4-subunit gene CACNB4 in patients with idiopathic generalized epilepsy and episodic ataxia, Am. J. Hum. Genet. 66 (2000) 1531–1539.
[4] N. Ramachandran, J.M. Girard, J. Turnbull, B.A. Minassian, The autosomal recessively inherited progressive myoclonus epilepsies and their genes, Epilepsia 50 (Suppl. 5) (2009) 29–36.
[5] A. Shahwan, M. Farrell, N. Delanty, Progressive myoclonic epilepsies: a review of genetic and therapeutic aspects, Lancet Neurol. 4 (2005) 239–248.
[6] M. Artal-Sanz, N. Tavernarakis, Proteolytic mechanisms in necrotic cell death and neurodegeneration, FEBS Lett. 579 (2005) 3287–3296.
[7] M. Polajnar, E. Zerovnik, Impaired autophagy: a link between neurodegenerative diseases and progressive myoclonus epilepsies, Trends Mol. Med. 17 (2011) 293–300.
[8] P. Genton, Unverricht-Lundborg disease (EPM1). EpilepsiaSpecial Issue: Epilepsy at the Cutting Edge: A Symposium to Honor Fred and Eva Andermann, 51 (Issue Supplement s1) (2010) 37–39,http://dx.doi.org/10.1111/j.1528-1167.2009.02441.x. [9] L.A. Pennacchio, A.E. Lehesjoki, N.E. Stone, V.L. Willour, K. Virtaneva, J. Miao, E.
D'Amato, L. Ramirez, M. Faham, M. Koskiniemi, J.A. Warrington, R. Norio, A. de la Chapelle, D.R. Cox, R.M. Myers, Mutations in the gene encoding cystatin B in pro-gressive myoclonus epilepsy (EPM1), Science 271 (1996) 1731–1734.
[10]M.D. Lalioti, M. Mirotsou, C. Buresi, M.C. Peitsch, C. Rossier, R. Ouazzani, M. Baldy-Moulinier, A. Bottani, A. Malafosse, S.E. Antonarakis, Identification of mutations in cystatin B, the gene responsible for the Unverricht-Lundborg type of progressive myoclonus epilepsy (EPM1), Am. J. Hum. Genet. 60 (1997) 342–351.
[11] M.D. Lalioti, H.S. Scott, C. Buresi, C. Rossier, A. Bottani, M.A. Morris, A. Malafosse, S.E. Antonarakis, Dodecamer repeat expansion in cystatin B gene in progressive myoclo-nus epilepsy, Nature 386 (1997) 847–851.
[12]T. Joensuu, A.E. Lehesjoki, O. Kopra, Molecular background of EPM1-Unverricht-Lundborg disease, Epilepsia 49 (2008) 557–563.
Fig. 10. Transmission electron microscopy (TEM) of stefin B mutants Q71P (A) and G50E (B), both transfected in CHO cells. Ultra-thin sections (70 nm) were immunogold labeled with rabbit anti-stefin B antibody and Protein A-gold (10 nm). Cell proteins were contrasted with uranyl acetate. Aggregated proteins (*) and the position of gold particles (arrows) are denoted. Bars represent 200 nm.
[13]E. Pinto, J. Freitas, A.J. Duarte, I. Ribeiro, D. Ribeiro, J.L. Lima, J. Chaves, O. Amaral, Unverricht-Lundborg disease: homozygosity for a new splicing mutation in the cystatin B gene, Epilepsy Res. 99 (2012) 187–190.
[14] L. Canafoglia, E. Gennaro, G. Capovilla, G. Gobbi, A. Boni, F. Beccaria, M. Viri, R. Michelucci, P. Agazzi, S. Assereto, D.A. Coviello, M. Di Stefano, D. Rossi Sebastiano, S. Franceschetti, F. Zara, Electroclinical presentation and genotype-phenotype rela-tionships in patients with Unverricht-Lundborg disease carrying compound hetero-zygous CSTB point and indel mutations, Epilepsia 53 (2012) 2120–2127.
[15] O.O. Erdinc, T. Joensuu, F. Ilgen-Uslu, N. Bebek, C. Ozkara, K. Tutkavul, A. Gunduz, A.-E. Lehesjoki, B. Baykan, Unverricht-Lundborg disease in Turkey: delineating the phe-notype between cystatin B mutation positive and negative Cases, J. Neurol. Sci. 27 (2010) 001–011.
[16] R. Eldridge, M. Iivanainen, R. Stern, T. Koerber, B.J. Wilder, "Baltic" myoclonus epilepsy: hereditary disorder of childhood made worse by phenytoin, Lancet 2 (1983) 838–842.
[17] P. Koskenkorva, J. Khyuppenen, E. Niskanen, M. Kononen, P. Bendel, E. Mervaala, A.E. Lehesjoki, R. Kalviainen, R. Vanninen, Motor cortex and thalamic atrophy in Unverricht-Lundborg disease: voxel-based morphometric study, Neurology 73 (2009) 606–611.
[18] O. Manninen, P. Koskenkorva, K.K. Lehtimaki, J. Hypponen, M. Kononen, T. Laitinen, H. Kalimo, O. Kopra, R. Kalviainen, O. Grohn, A.E. Lehesjoki, R. Vanninen, White mat-ter degeneration with Unverricht–Lundborg progressive myoclonus epilepsy: a translational diffusion-tensor imaging study in patients and cystatin B-deficient mice, Radiology 269 (2013) 232–239,http://dx.doi.org/10.1148/radiol.13122458. [19] A.G. Bassuk, R.H. Wallace, A. Buhr, A.R. Buller, Z. Afawi, M. Shimojo, S. Miyata, S.
Chen, P. Gonzalez-Alegre, H.L. Griesbach, S. Wu, M. Nashelsky, E.K. Vladar, D. Antic, P.J. Ferguson, S. Cirak, T. Voit, M.P. Scott, J.D. Axelrod, C. Gurnett, A.S. Daoud, S. Kivity, M.Y. Neufeld, A. Mazarib, R. Straussberg, S. Walid, A.D. Korczyn, D.C. Slusarski, S.F. Berkovic, H.I. El-Shanti, A homozygous mutation in human PRICKLE1 causes an autosomal-recessive progressive myoclonus epilepsy-ataxia syndrome, Am. J. Hum. Genet. 83 (2008) 572–581.
[20] L.M. Dibbens, R. Michelucci, A. Gambardella, F. Andermann, G. Rubboli, M.A. Bayly, T. Joensuu, D.F. Vears, S. Franceschetti, L. Canafoglia, R. Wallace, A.G. Bassuk, D.A. Power, C.A. Tassinari, E. Andermann, A.E. Lehesjoki, S.F. Berkovic, SCARB2 mutations in progressive myoclonus epilepsy (PME) without renal failure, Ann. Neurol. 66 (2009) 532–536.
[21] P. Genton, Unverricht-Lundborg disease (EPM1), Epilepsia 51 (Suppl. 1) (2010) 37–39.
[22] V. Turk, J. Brzin, M. Kotnik, B. Lenarcic, T. Popovic, A. Ritonja, M. Trstenjak, L. Begic-Odobasic, W. Machleidt, Human cysteine proteinases and their protein inhibitors stefins, cystatins and kininogens, Biomed. Biochim. Acta 45 (1986) 1375–1384.
[23] B. Lenarcic, I. Krizaj, P. Zunec, V. Turk, Differences in specificity for the interactions of stefins A, B and D with cysteine proteinases, FEBS Lett. 395 (1996) 113–118.
[24] S.W. Dahl, T. Halkier, C. Lauritzen, I. Dolenc, J. Pedersen, V. Turk, B. Turk, Human re-combinant pro-dipeptidyl peptidase I (cathepsin C) can be activated by cathepsins L and S but not by autocatalytic processing, Biochemistry 40 (2001) 1671–1678.
[25] N. Kopitar-Jerala, A. Schweiger, R.M. Myers, V. Turk, B. Turk, Sensitization of stefin B-deficient thymocytes towards staurosporin-induced apoptosis is independent of cysteine cathepsins, FEBS Lett. 579 (2005) 2149–2155.
[26] F. Yang, K.H. Tay, L. Dong, R.F. Thorne, C.C. Jiang, E. Yang, H.Y. Tseng, H. Liu, R. Christopherson, P. Hersey, X.D. Zhang, Cystatin B inhibition of TRAIL-induced apo-ptosis is associated with the protection of FLIP(L) from degradation by the E3 ligase itch in human melanoma cells, Cell Death Differ. 17 (2010) 1354–1367.
[27] T. Sun, V. Turk, B. Turk, N. Kopitar-Jerala, Increased expression of stefin B in the nucleus of T98G astrocytoma cells delays caspase activation, Front. Mol. Neurosci. 5 (2012) 93.
[28] A. Vaarmann, A. Kaasik, A. Zharkovsky, Altered tryptophan metabolism in the brain of cystatin B-deficient mice: a model system for progressive myoclonus epilepsy, Epilepsia 47 (2006) 1650–1654.
[29]M. Korja, V. Kaasinen, S. Lamusuo, R. Parkkola, K. Nagren, R.J. Marttila, Substantial thalamostriatal dopaminergic defect in Unverricht-Lundborg disease, Epilepsia 48 (2007) 1768–1773.
[30] S. Tegelberg, O. Kopra, T. Joensuu, J.D. Cooper, A.E. Lehesjoki, Early microglial activa-tion precedes neuronal loss in the brain of the Cstb−/− mouse model of progres-sive myoclonus epilepsy, EPM1, J. Neuropathol. Exp. Neurol. 71 (2012) 40–53.
[31] K. Lieuallen, L.A. Pennacchio, M. Park, R.M. Myers, G.G. Lennon, Cystatin B-deficient mice have increased expression of apoptosis and glial activation genes, Hum. Mol. Genet. 10 (2001) 1867–1871.
[32]P. Shannon, L.A. Pennacchio, M.K. Houseweart, B.A. Minassian, R.M. Myers, Neuro-pathological changes in a mouse model of progressive myoclonus epilepsy: cystatin B deficiency and Unverricht-Lundborg disease, J. Neuropathol. Exp. Neurol. 61 (2002) 1085–1091.
[33] R. Di Giaimo, M. Riccio, S. Santi, C. Galeotti, D.C. Ambrosetti, M. Melli, New insights into the molecular basis of progressive myoclonus epilepsy: a multiprotein complex with cystatin B, Hum. Mol. Genet. 11 (2002) 2941–2950.
[34] M.K. Lehtinen, S. Tegelberg, H. Schipper, H. Su, H. Zukor, O. Manninen, O. Kopra, T. Joensuu, P. Hakala, A. Bonni, A.E. Lehesjoki, Cystatin B deficiency sensitizes neurons to oxidative stress in progressive myoclonus epilepsy, EPM1, J. Neurosci. 29 (2009) 5910–5915.
[35] M. Butinar, M.T. Prebanda, J. Rajkovic, B. Jeric, V. Stoka, C. Peters, T. Reinheckel, A. Kruger, V. Turk, B. Turk, O. Vasiljeva, Stefin B deficiency reduces tumor growth via sensitization of tumor cells to oxidative stress in a breast cancer model, Oncogene (2013),http://dx.doi.org/10.1038/onc.2013.314(Aug 19).
[36] K. Skerget, A. Taler-Vercic, A. Bavdek, V. Hodnik, S. Ceru, M. Tusek-Znidaric, T. Kumm, D. Pitsi, M. Pompe-Novak, P. Palumaa, S. Soriano, N. Kopitar-Jerala, V. Turk, G. Anderluh, E. Zerovnik, Interaction between oligomers of stefin B and amyloid-beta in vitro and in cells, J. Biol. Chem. 285 (2010) 3201–3210.
[37] A. Taler-Vercic, E. Zerovnik, Binding of amyloid peptides to domain-swapped dimers of other amyloid-forming proteins may prevent their neurotoxicity, Bioessays 32 (2010) 1020–1024.
[38]E. Zerovnik, K. Skerget, M. Tusek-Znidaric, C. Loeschner, M.W. Brazier, D.R. Brown, High affinity copper binding by stefin B (cystatin B) and its role in the inhibition of amyloidfibrillation, FEBS J. 273 (2006) 4250–4263.
[39] E. D'Amato, Z. Kokaia, A. Nanobashvili, M. Reeben, A.E. Lehesjoki, M. Saarma, O. Lindvall, Seizures induce widespread upregulation of cystatin B, the gene mutated in progressive myoclonus epilepsy, in rat forebrain neurons, Eur. J. Neurosci. 12 (2000) 1687–1695.
[40] S. Ceru, S. Konjar, K. Maher, U. Repnik, I. Krizaj, M. Bencina, M. Renko, A. Nepveu, E. Zerovnik, B. Turk, N. Kopitar-Jerala, Stefin B interacts with histones and cathepsin L in the nucleus, J. Biol. Chem. 285 (2010) 10078–10086.
[41]S. Ceru, S. Rabzelj, N. Kopitar-Jerala, V. Turk, E. Zerovnik, Protein aggregation as a possible cause for pathology in a subset of familial Unverricht-Lundborg disease, Med. Hypotheses 64 (2005) 955–959.
[42]E. Zerovnik, M. Pompe-Novak, M. Skarabot, M. Ravnikar, I. Musevic, V. Turk, Human stefin B readily forms amyloid fibrils in vitro, Biochim. Biophys. Acta 1594 (2002) 1–5.
[43] E. Zerovnik, M. Skarabot, K. Skerget, S. Giannini, V. Stoka, S. Jenko-Kokalj, R.A. Staniforth, Amyloidfibril formation by human stefin B: influence of pH and TFE onfibril growth and morphology, Amyloid 14 (2007) 237–247.
[44] E. Zerovnik, Amyloid-fibril formation. Proposed mechanisms and relevance to con-formational disease, Eur. J. Biochem. 269 (2002) 3362–3371.
[45] G.B. Irvine, O.M. El-Agnaf, G.M. Shankar, D.M. Walsh, Protein aggregation in the brain: the molecular basis for Alzheimer's and Parkinson's diseases, Mol. Med. 14 (2008) 451–464.
[46]N.R. Cohen, S.R. Hammans, J. Macpherson, J.A. Nicoll, New neuropathological find-ings in Unverricht-Lundborg disease: neuronal intranuclear and cytoplasmic inclu-sions, Acta Neuropathol. 121 (2011) 421–427.
[47] C.M. Dobson, Getting out of shape, Nature 418 (2002) 729–730.
[48] V.N. Uversky, Amyloidogenesis of natively unfolded proteins, Curr. Alzheimer Res. 5 (2008) 260–287.
[49]E. Zerovnik, V. Stoka, A. Mirtic, G. Guncar, J. Grdadolnik, R.A. Staniforth, D. Turk, V. Turk, Mechanisms of amyloidfibril formation–focus on domain-swapping, FEBS J. 278 (2011) 2263–2282.
[50] S. Ceru, S.J. Kokalj, S. Rabzelj, M. Skarabot, I. Gutierrez-Aguirre, N. Kopitar-Jerala, G. Anderluh, D. Turk, V. Turk, E. Zerovnik, Size and morphology of toxic oligomers of amyloidogenic proteins: a case study of human stefin B, Amyloid 15 (2008) 147–159.
[51] E. Cipollini, M. Riccio, R. Di Giaimo, F. Dal Piaz, G. Pulice, S. Catania, I. Caldarelli, M. Dembic, S. Santi, M. Melli, Cystatin B and its EPM1 mutants are polymeric and aggre-gate prone in vivo, Biochim. Biophys. Acta 1783 (2008) 312–322.
[52] S. Ceru, E. Zerovnik, Similar toxicity of the oligomeric molten globule state and the prefibrillar oligomers, FEBS Lett. 582 (2008) 203–209.
[53]M. Stefani, Protein aggregation diseases: toxicity of soluble prefibrillar aggregates and their clinical significance, Methods Mol. Biol. 648 (2010) 25–41.
[54] S. Rabzelj, G. Viero, I. Gutierrez-Aguirre, V. Turk, M. Dalla Serra, G. Anderluh, E. Zerovnik, Interaction with model membranes and pore formation by human stefin B: studying the native and prefibrillar states, FEBS J. 275 (2008) 2455–2466.
[55] S. Ceru, R. Layfield, T. Zavasnik-Bergant, U. Repnik, N. Kopitar-Jerala, V. Turk, E. Zerovnik, Intracellular aggregation of human stefin B: confocal and electron micros-copy study, Biol. Cell. 102 (2010) 319–334.
[56] M. Polajnar, R. Vidmar, M. Vizovisek, M. Fonovic, N. Kopitar-Jerala, E. Zerovnik, Influ-ence of partial unfolding and aggregation of human stefin B (cystatin B) EPM1 mu-tants G50E and Q71P on selective cleavages by cathepsins B and S, Biol. Chem. 394 (2013) 783–790.
[57] S. Rabzelj, V. Turk, E. Zerovnik, In vitro study of stability and amyloid-fibril forma-tion of two mutants of human stefin B (cystatin B) occurring in patients with EPM1, Protein Sci. 14 (2005) 2713–2722.
[58]T. Cirman, K. Oresic, G.D. Mazovec, V. Turk, J.C. Reed, R.M. Myers, G.S. Salvesen, B. Turk, Selective disruption of lysosomes in HeLa cells triggers apoptosis mediated by cleavage of Bid by multiple papain-like lysosomal cathepsins, J. Biol. Chem. 279 (2004) 3578–3587.
[59] R.R. Kopito, Aggresomes, inclusion bodies and protein aggregation, Trends Cell Biol. 10 (2000) 524–530.
[60]J.K. Andersen, Oxidative stress in neurodegeneration: cause or consequence? Nat. Med. 10 (2004) S18–S25 (Suppl.).
[61] A. Federico, E. Cardaioli, P. Da Pozzo, P. Formichi, G.N. Gallus, E. Radi, Mitochondria, oxidative stress and neurodegeneration, J. Neurol. Sci. 322 (2012) 254–262.
[62] H. Xie, S. Hou, J. Jiang, M. Sekutowicz, J. Kelly, B.J. Bacskai, Rapid cell death is preceded by amyloid plaque-mediated oxidative stress, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 7904–7909,http://dx.doi.org/10.1073/pnas.1217938110(Epub 2013 Apr 22). [63] L.A. Pennacchio, D.M. Bouley, K.M. Higgins, M.P. Scott, J.L. Noebels, R.M. Myers,
Pro-gressive ataxia, myoclonic epilepsy and cerebellar apoptosis in cystatin B-deficient mice, Nat. Genet. 20 (1998) 251–258.
[64] T. Joensuu, M. Kuronen, K. Alakurtti, S. Tegelberg, P. Hakala, A. Aalto, L. Huopaniemi, N. Aula, R. Michellucci, K. Eriksson, A.E. Lehesjoki, Cystatin B: mutation detection, al-ternative splicing and expression in progressive myclonus epilepsy of Unverricht-Lundborg type (EPM1) patients, Eur. J. Hum. Genet. 15 (2007) 185–193.
[65]P. Koskenkorva, J. Hypponen, M. Aikia, E. Mervaala, T. Kiviranta, K. Eriksson, A.E. Lehesjoki, R. Vanninen, R. Kalviainen, Severer phenotype in Unverricht-Lundborg disease (EPM1) patients compound heterozygous for the dodecamer repeat expan-sion and the c.202CNT mutation in the CSTB gene, Neurodegener. Dis. 8 (2011) 515–522.
