HSPB1 loss of chaperone-like activity in the pathogenesis of Amyotrophic Lateral Sclerosis. Simona Capponi1,2, Thomas Geuens2, Alessandro Geroldi1, Elena Cichero3, Paola Origone1,4, Simonetta Verdiani5, Elias Adriaenssens2, Vicky De Winter2, Monica Bandettini di Poggio6, Paola Fossa3, Paola Mandich1,4, Emilia Bellone1,4 and Vincent Timmerman2
1 Dept. of Neuroscience, Rehabilitation, Ophthalmology, Genetics and Maternal and Child Health, Section of Medical Genetics, University of Genoa, 16132 Genoa, Italy;
2 Peripheral Neuropathy Group, VIB Department of Molecular Genetics, University of Antwerp, B26-10 Antwerp, Belgium;
3 Dept. of Pharmacy, Section of Medicinal Chemistry, School of Medical and Pharmaceutical Sciences, University of Genoa, 16132 Genoa, Italy;
4 COU Medical Genetics, IRCCS AOU San Martino IST - Istituto Nazionale per la Ricerca sul Cancro, Genoa, Italy;
5 Dept. of Health Science, University of Genoa, 16132 Genoa, Italy;
6 Dept. of Neuroscience, Santa Corona Hospital, Pietra Ligure, 17027 Savona, Italy.
Corresponding author:
Prof. Dr. Vincent Timmerman, PhD Peripheral Neuropathy Group
VIB Department of Molecular Genetics University of Antwerp - CDE
Universiteitsplein 1, Building V B-2610 Antwerp, Belgium Tel: +32-3-265.10.24
Tel: +32-3-265.11.02 (Secretary VIB-DMG: Mrs. Gisèle Smeyers) Fax: +32-3-265.11.12
ABSTRACT
Genetic discoveries in Amyotrophic Lateral Sclerosis (ALS) have a significant impact on deciphering molecular mechanisms of motor neuron degeneration but, despite recent advances, the aetiology of most sporadic cases remains elusive. Several cellular mechanisms contribute to the motor neuron degeneration in ALS, including RNA metabolism, cellular interactions between neurons and non-neuronal cells and seeding of misfolded protein with prion-like propagation. In this scenario, the importance of protein turnover and degradation in motor neuron homeostasis gained increased recognition. In this study, we evaluated the role of the candidate gene HSPB1, a molecular chaperone involved in several proteome-maintenance functions. In a cohort of 153 unrelated Italian ALS patients, we identified the novel extended frameshift p.Ala204Glyfs*6 in a sporadic ALS case. Functional characterization demonstrated that the mutant protein alters the dynamic equilibrium of HSPB1, sequestering the wild-type protein in a stable dimer and resulting in a loss of chaperone-like activity. Our results underline the relevance of identifying rare but pathogenic variations in sporadic neurodegenerative diseases, speculating a possible correlation between specific pathomechanisms linked to HSPB1 mutations and the associated neurological phenotype. Our study confirms previous data and provides additional lines of evidence to propose
INTRODUCTION
Amyotrophic Lateral Sclerosis (ALS) is a devastating, adult onset neurodegenerative disorder characterized by the progressive loss of upper and lower motor neurons. The disease is mainly sporadic (sALS), but 5-10% of patients have a positive family history (fALS). Starting from the hypothesis of a sporadic disorder as the result of complex genome-environment interactions, the inheritability of sALS has been largely demonstrated in recent years. Although significant advances have been made in the molecular genetics of ALS, the screening of the currently known large- or small-effect genes only solves a small fraction of cases and thus the aetiology of sALS remains mostly unexplained (1). In this scenario, there is a growing interest in identifying additional genetic causes and modifying or risk factors for sALS.
In this study we evaluated the involvement of the small Heat-Shock Protein B1 (HSPB1) in sALS. The HSPB1 gene is an interesting functional candidate as it encodes an inducible ATP-independent molecular chaperone able to reduce the aggregation of misfolded proteins, either promoting their refolding or their degradation (2). This function is crucial in neurodegenerative conditions, including ALS, since one of the earliest neuropathological hallmarks of the disease is the formation of ubiquitinated aggregates in motor neurons and astrocytes (3). The cytoprotective properties of HSPB1 have been demonstrated in cultured neuronal cells and in vivo in models of ALS (4). Additionally, the expression of HSPB1, together with the ATP-dependent chaperones HSP70 and HSP90, has been found to be upregulated in neurodegenerative conditions (5). Recently, its upregulation has been demonstrated in the lumbar spinal cord of ALS patients as well (6). The finding that mutations in HSPB1 result in autosomal dominant or recessive axonal peripheral neuropathies (distal Hereditary Motor Neuropathy and Charcot-Marie-Tooth -CMT- type 2F) highlights the relevance of this gene to maintain motor neuron integrity (7,8). Interestingly, a heterozygous variant in the promoter region of HSPB1 was identified in a Belgian sALS patient, drastically impairing the heat-shock protein responses (9).
Here, we report the identification of a novel extended frameshift mutation in the HSPB1 gene (c.610_611insG; p.Ala204Glyfs*6) in a sporadic ALS patient. We demonstrate that the mutant protein results in a loss of chaperone-like activity, describing, for the first time, a novel pathomechanism linked to HSPB1 mutations.
RESULTS Genetic findings
Molecular analysis of HSPB1 revealed the presence of several sequence variants, described in Supplementary Table 2. Among these, we identified, the heterozygous frameshift c.610_611insG, p.Ala204Glyfs*6 in a 73 year old sALS patient (ALS160) originating from Puglia (Italy). The guanine insertion affects the third to last codon of HSPB1 and leads to the translation of eight additional base pairs in the 3’UTR of the gene, resulting in an extended mutant protein (Fig. 1A-1B). The alignment of HSPB1 C-terminal domain across different species showed the conservation of the p.Ala204 amino acid and none of the selected HSPB1 orthologues displayed an additional C-terminal segment similar to the p.Ala204Glyfs*6 variant (Fig. 1C).
The sequence variant c.610_611insG was absent from public databases for human variations (dbSNP138, 1000 Genomes and Exome Variant Server), and was also not found in 500 ethnically matched control chromosomes. Moreover, the variant was absent from the additional screening of 660 chromosomes of Charcot-Marie-Tooth patients. In addition, the variant was not identified in a large international ALS genetic study (L. van den Berg and J. Landers, personal communication), confirming its extremely low allele frequency and suggesting a possible regional effect. To better frame this variation in the context of ALS genetics, we additionally excluded the presence of other possible pathogenic mutations in the patient ALS160 analyzing the full list of the ALS-associated genes to date reported, through whole-exome sequencing (see Supplementary material).
Clinical description of the patient ALS160
The patient ALS160 has a negative family history for neurological disorders up to the second degree of relatives. She experienced the first symptoms at the age of 73 with balance disturbances, frequent falls and dysarthria. Two years after the initial presentation, a diagnosis of Multiple System Atrophy with cerebellar predominance was made. At the age of 77, due to the onset of bulbar symptoms and diffuse signs of spinal amyotrophy, she underwent neurophysiological studies with evidence of axonal motor neuropathy in the lower limbs and signs of denervation and reinnervation at four limbs and paravertebral muscles. A diagnosis of motor neuron disease was made. At age 78, five years after symptom onset, neurological examination showed severe cranial involvement with paralysis of vertical gaze, hypotrophic tongue with lateral fasiculations, scanning dysarthria, mixed dysphagia and respiratory insufficiency leading to nocturnal non-invasive ventilation. Mild hyposthenia with muscular hypotrophy, fasciculations and brisk deep tendon reflexes were evident at four limbs. Signs of involvement of extrapyramidal system were consistent with hypomimia,
resting, postural and kinetic tremor in both hands and head. Camptocormia and Pisa Syndrome were also evident. Postural change were performed with assistance and the patient was able to ambulate for small distances with walker with short stepped gait and therefore mostly wheelchair bounded. Signs of cognitive impairment were not evident. To date, at the age of 82, the patient has no major worsening in the clinical phenotype.
Functional analysis of p.Ala204Glyfs*6
HSPB1 expression and induction in ALS160 EBV-transformed LCL
To explore the effect of p.Ala204Glyfs*6 on cellular homeostasis, we first investigated if the presence of the mutation could impair HSPB1 expression. We found that both wild-type and mutant alleles are expressed on mRNA and protein level in patient-derived EBV-transformed lymphoblastoid cell lines (LCLs) (Fig. 2B and 2C). Interestingly, in these cell lines, the relative mRNA expression level of HSPB1 showed a marked variability among patient and control samples (Fig. 2B). On protein level, HSPB1 seemed to be noticeably reduced in patient ALS160, compared to controls (Fig. 2C) but the inhibition of the proteasomal- or lysosomal-dependent protein degradation (through MG132 and Leupeptin, respectively) excluded that the observed reduction resulted from the activation of these pathways (Supplementary Fig. 1).
Although it is expressed in basal conditions, HSPB1 is striking upregulated under stress conditions, in which the concentration of aggregation-prone folding intermediates increases and this protein mainly exerts its chaperone and cytoprotective functions (10). Through heat-shock experiments, we demonstrated that HSPB1 is still inducible in patient-derived LCLs. Interestingly, 24h after heat-shock (HS R24h), the expression level of HSPB1 returned to the basal condition, in which the comparison between patient and control still revealed a marked reduction (Fig. 3A).
Together, these results suggest that there is not specific degradation or a defect in the expression/induction of HSPB1 in ALS160, compared to controls. We therefore decided to evaluate whether HSPB1 was still functional upon homeostasis perturbation.
HSPB1 p.Ala204Glyfs*6 sequesters the wild-type protein in a stable dimer
Small heat-shock proteins exhibit a pronounced dynamic equilibrium between monomeric, dimeric and oligomeric state. The oligomers are believed to be reservoirs which dissociate upon stress into the active dimer and monomer forms. The HSPB1 dimer interface is located within the alpha-crystallin domain and is further strengthened and stabilized by the C-terminus (11). Therefore, we evaluated whether the additional C-terminal extension of the p.Ala204Glyfs*6 mutation affected HSPB1 dimerization, by analysing the protein migration pattern in non-reducing conditions. In the
presence of two different proteins, we expected the formation of three dimer combinations (wild-type/wild-type, wild-type/mutant and mutant/mutant). Interestingly, the p.Ala204Glyfs*6 LCLs showed only one structure, with a higher molecular weight than controls. In order to better understand mutant HSPB1 dimerization properties and composition, we used CHO-K1 cell lines, stably expressing HSPB1 wild-type, HSPB1 p.Ala204Glyfs*6 or both. Intriguingly, while both the wild-type and the mutant protein are able to form homodimers, the double stable CHO-K1 cell line confirmed the pattern observed in LCLs, showing only one dimer and apparently corresponding to the wild-type/mutant dimer (Fig. 3B). This data suggests that when the two proteins are expressed together, the dimerization pattern shifts predominantly to the wild-type/mutant form.
In the context of HSPB1 dynamic equilibrium, we previously demonstrated that, during post-stress recovery, the monomeric form is upregulated and the dimers dissociates into monomers (12). Interestingly, we demonstrated that, upon heat-shock, in patient-derived LCLs, the increased expression of monomeric HSPB1 was not associated with a dimer-monomer shuttling (Fig. 3C). This result suggests that the wild-type/mutant dimer forms a stable structure, not able to dissociate and therefore perturbing the dynamic equilibrium of HSPB1.
To explain the different dimerization properties observed, we evaluated the surface steric and electrostatic profile of the mutant C-terminus extension, building a specific 3D-model (Fig. 3D). In the native HSPB1, molecular modelling data suggests that the formation of oligomers and dimers is promoted by the exposure of a number of weak non-covalent contacts sites in specific surface patches. While the wild-type HSPB1 displays a precise charge distribution across the C-terminal domain, the mutant protein shows marked electrostatic alterations, with a smaller highly hydrophobic region (HY1), only composed of the N-terminus, a reduced negative charge area (NC) and an additional positive patch (PC) resulting from the mutated residues 204-208 (Supplementary Table 3). Taken together, these data suggest that a difference between native and mutant HSPB1 dimerization kinetics can be partially explained by different hydrophobic and polar contacts at the protein surface.
HSPB1 p.Ala204Glyfs*6 results in a loss of chaperone-like activity
Previous studies have shown that shuttling between monomers, dimers and oligomers is required for the proper substrate binding and chaperone-like activity of HSPB1 (13). Non-native substrates are maintained in a folding-competent state, promoting their refolding or, when this is not possible, their degradation (14). Based on the different dimerization profile observed for HSPB1 p.Ala204Glyfs*6, we investigated whether this mutant HSPB1 still preserves its chaperone activity. We therefore performed an indirect clearance assay using the aggregation prone Superoxide
Dismutase 1 (SOD1) mutation p.Ala4Val as a substrate. The formation of aggregates was quantified
by separating the aggregates-containing pellet fraction from the supernatant. As shown in Fig. 4, native HSPB1 assists in refolding/degradation of SOD1 p.Ala4Val, reducing the amount of insoluble SOD1 p.Ala4Val in the pellet fraction. However, the presence of the HSPB1 p.Ala204Glyfs*6 increases the amount of SOD1 p.Ala4Val aggregates in the pellet fraction, when compared to the wild-type HSPB1, suggesting that the mutant protein results in a loss of chaperone-like activity.
Validation cohort
Validation cohort included 94 ALS patients consecutively seen at the Turin ALS centre, 51 males and 41 females, with a mean age at onset of 65.7 (SD10.1) years. Thirty-seven patients had a bulbar onset and 57 a spinal onset. Six of them were familial. We found one missense mutations, a c.570 G>C causing a substitution of a glutamine with a histidine at codon 190 (p.Gln190His); this variation is present in ExAc Version0.3 with a very low frequency (G allele n=121308, C allele n=10). In silico predictions using SIFT and PolyPhen2 show missense variant to be deleterious (p.Gln190His: SIFT score 0.00, PolyPhen2 score 0.974).
Clinical description of the patient
Case SLA2014-469. This is a female patient who developed dysphagia and dysarthria at the age of
58 years. Few months later she also showed weakness and atrophy at the muscles of the right hand. One year after the onset of the first symptoms, she underwent a neurophysiological examination, which showed active and chronic denervation at upper and lower limbs at and at genioglossus muscle. Spirometry showed a marked restrictive pattern (FVC 56%). Creatine kinase was increased. Her family history was negative for ALS, but her mother was affected by Alzheimer’s diseases. Neuropsychological examination showed a mild cognitive impairment, mainly behavioural, according to Strong et al classification (2008). She was diagnosed with definite ALS. She died 24 months after the onset of symptoms due to respiratory failure. She carried p.Gln190His mutation of the HSPB1 gene.
DISCUSSION
Several neurodegenerative disorders, as Huntington, Alzheimer, Parkinson and ALS are characterized by the presence of neuronal intracellular aggregates. Misfolded proteins accumulate in the cytoplasm and form compact, insoluble structures prone to precipitate. The accumulation of aggregates is caused by the incapacity of the molecular chaperones to overcome the instable folding intermediates. A number of studies demonstrated that the overexpression of the chaperones HSP70 and HSP90 could suppress protein aggregation, hence influencing the phenotype (15). In addition, it has already been shown that mutated SOD1 aggregates sequester a number of heat-shock proteins, including HSPB1. In this scenario, the availability of HSPs to undertake their normal housekeeping, as well as stress-induced, functions is reduced, confirming their crucial involvement in the motor neuron response to cellular stress (16).
In this report we describe the first HSPB1 coding variant identified in a sALS patient. We demonstrated that the extended frameshift p.Ala204Glyfs*6 sequesters the wild-type protein in a stable dimer, severely affecting HSPB1 dynamic equilibrium. Ultimately, the mutation results in a loss of chaperone-like activity, being unable to clear protein aggregates. While some of the CMT-associated missense mutations characterized so far displayed hyperactivity (12, 17), this is the first
HSPB1 mutation which leads to a loss of chaperone-like activity. Interestingly, Dierick and
collaborators previously reported the SNP c.-217C>T in the HSPB1 promoter in a Belgian sALS patient. This variant leads to the impairment of HSPB1 upregulation in stress conditions, thus resulting in a severely affected heat-shock response (9). Taken this data and our findings together, it is therefore tempting to speculate a correlation between different pathomechanisms of HSPB1 mutations and the resulting neurodegenerative phenotype. The CMT-associated HSPB1 mutations result in a hyperactive protein, enhancing the binding to their client proteins (12). This leads to several downstream cellular effects, including the microtubules stabilization (17). On the contrary, an impaired heat-shock response, either due to the lack of HSPB1 upregulation during stress conditions (9) or resulting from a loss of chaperone-like activity, might contribute to the pathogenesis of ALS making the motor neurons more vulnerable to homeostasis perturbation. It is well documented that the aging-related accumulation of misfolded/oxidized proteins is a challenge to the proteostasis system, in particular in long-living cells like motor neurons. This is due to the fact that the age-related decline in the protestasis machinery determines the inability to upregulate chaperones in response to conformational stress (14). In this scenario, the presence of HSPB1 mutations would trigger the disease manifestation, perturbing central proteostasis, including folding and clearance mechanisms.
Interestingly, the impairment of protein degradation pathways have already been linked to the pathogenesis of ALS through the identification of mutations in the ubiquitin-like protein
Ubiquilin-2 (UBQLNUbiquilin-2) (18) and the autophagy-related Sequestosome 1 (SQSTM1 - p6Ubiquilin-2) (19) in familial and
isolated cases.
The ALS is a complex disease and the interactions among several genes influence its pathogenesis, providing with a different effect on the phenotypic presentation (20). In regard to the complexity to decipher the genetic landscape of sALS, the investigation of the cellular implications of rare but pathogenic mutations can provide a novel approach to evaluate relevant molecular pathways in sporadic diseases. The investigation of the HSPB1 gene in broader ALS cohorts might therefore offer additional insights in dissecting the molecular basis of this motor neuron disorder.
MATERIAL AND METHODS
Cohort enrolment and Genetic Studies
The studied cohort was composed of 153 unrelated Italian patients presenting a clinical diagnosis of ALS. Patients were enrolled in frame of the multicentre-multisource prospective population based registry LIGALS (Liguria Amyotrophic Lateral Sclerosis Registry) (21), and through other Italian centres. After informed consent, patients underwent neurological examination and were diagnosed according to the El-Escorial revised criteria (22). A description of the enrolled cohort is provided in Supplementary Table 1.
Differential diagnosis with spinobulbar muscular atrophy was considered in male ALS patients. The cohort was previously analysed for the major genes associated with ALS pathogenesis (SOD1,
TDP43, FUS, FIG4, OPTN, VCP and MATR3, and C9orf72 and ATXN2 expansions - data not
shown).
The coding regions of HSPB1 (Ensembl reference sequence: ENST00000248553) was analysed by direct sequencing. The SNP c.-217T>C (rs545738637) previously identified in HSPB1 promoter (9) was tested by direct sequencing at the genomic position 7:76302496 (GRCh38). Sequencing products were run on a 3130xl Genetic Analyzer (Applied Biosystems - Life Technologies) and analysed with SeqScape Software v.2.7 (Applied Biosystems - Life Technologies). The variant c.-217T>C (rs545738637) was not identified in our cohort.
The studied control cohort consisted in 250 ethnically matched individuals. The control screening was performed by Denaturing HighPerformance Liquid Chromatography (DHPLC) (Wave MD -Transgenomics). The optimal melting temperatures for the analysis were defined as the best chromatographic resolution of the different heteroduplex in the heterozygous DNAs (data not shown).
Whole-exome sequencing was performed by deCODE Genetics. Exome enrichment was performed using Nextera and run on HiSeq2500. Reads alignment and variants calling was performed with BWA and GATK.
Functional studies
Cell culture, treatments and heat-shock experiments - All cell culture media and supplements were
purchased from Life Technologies. EBV-transformed Lymphoblastoid Cell Lines (LCLs) were cultivated at 37 °C and 6% CO2 in Gibco® RPMI 1640 supplemented with 15% foetal calf serum, 1% Sodium Pyruvate and 1% Penicillin-Streptomycin. The Chinese Hamster Ovary cell line (CHO-K1) was cultivated at 37 °C and 5% CO2 in GIBCO® Dulbecco's Modified Eagle Medium:
Nutrient Mixture F-12 (DMEM/F-12) supplemented with 10% fetal calf serum and 1% Penicillin-Streptomycin.
To evaluate the possible degradation pathways at the basis of the reduced HSPB1 expression in patient LCLs, we treated the cells inhibiting proteasomal and lysosomal pathways. The proteasome was inhibited with MG132 (Sigma-Aldrich) at the final concentration of 25 µM for 2, 4, 6, 8 and 24h. A DMSO control was used for the latest time point (24h) (data not shown). The lysosomal dependent degradation was inhibited using Leupeptin (Enzo Life Sciences) at the final concentration of 40 µM for 24, 48 and 72h, according to literature data (23).
The heat-shock experiments were performed by shocking the LCLs at 42 °C for 1h. The cells were placed back in the incubator to allow recovery at different time points (2, 4, 6, 8 12 and 24h) for the induction experiments. For the dimerization experiments the recovery time was set at 4h.
RNA isolation, cDNA sequencing and RT-qPCR - LCLs pellets were washed in ice-cold PBS and
centrifuged at 800 rpm for 8 min. RNA isolation was performed with the RNeasy Lipid Tissue Mini kit (Qiagen) and concentration and purity were evaluated with Nanodrop measurement. DNAse treatment was performed with the TURBO™ DNAse treatment kit (Life Technologies) and the cDNA conversion was performed with the SuperScript® III First-Strand Synthesis System (Life Technologies). To evaluate the relative expression level of HSPB1, we perfomed a RT-qPCR experiment on ViiA™ 7 Real-Time PCR System (Life Technologies). The reaction was performed using Power SYBR® Master Mix (Life Technologies) according to manufacturer’s instructions. The reactions were performed in triplicate, using four independent replicates. The expression level of HSPB1 was evaluated with qbase+ software (Biogazelle), using the three most stable reference genes among GAPDH, HMBS, TBP, HPRT1 and SDHA. Primer sequences are available upon request.
Plasmids and stable cell lines preparation - To generate stable CHO-K1 cell lines, we designed
specific constructs encoding HSPB1 wild type or mutant using the Gateway recombination system (Life Technologies). Since the mutant protein determines the additional translation of 8 base pairs in the 3’UTR of the gene, we decided to generate plasmids without tag, in order to preserve the structure of the C-terminal domain. Both wild-type and mutant cDNA, including the 3’UTR, were purchased from GenArt (Life Technologies). HSPB1 open reading frame (ORF) was amplified by PCR using specific primers flanked by attB recombination sites to allow the insertion of the product in the pDONR221 vector. The pDONRs were transferred by recombination to the pLenti6/V5 destination vector (Life Technologies). To evaluate the ability of HSPB1 to clear SOD1 p.A4V aggregates, we generated double stable cell lines expressing both SOD1 p.A4V and HSPB1 wild-type or p.Ala204Glyfs*6, alternatively. The wild-wild-type SOD1 plasmid was purchased from the
plasmid repository DNASU (Biodesign Institute at Arizona State University, http://dnasu.asu.edu). The c.14C>T substitution was inserted with site direct mutagenesis on wild-type SOD1, resulting in the p.A4V mutation according to the current amino acid nomenclature. The plasmid was first integrated in the pDONR221 and subsequently recombined in the pLenti6 destination vector, generating a GFP-fused protein (pLenti/GFP backbone available in house). All the plasmids were verified by Sanger sequencing. Stable cell lines were generated by Lentiviral transduction (24).
Western Blotting - Cells were lysed in E1a lysis buffer (1% Nonidet P-40, 20 mM Hepes pH8, 250
mM NaCl, 50 mM Tris-HCl, 1 mM EDTA, 2 mM DTT, 4 mM Sodium orthovanadate, 20 mM Glycerol-2-Phosphate, 10 mM Sodium Fluoride, 1 mM Sodium Pyrophosphate, together with the complete protease and the Phospho-STOP inhibitor mixtures - Roche Applied Science) for 30 min on ice and cleared by centrifugation (14000 rpm for 10 min). For the SOD1 clearance assay, the supernatant was collected after the first centrifugation. The pellet fraction was lysed in E1a and sonicated for 15 sec. After Bradford quantification, the lysates were boiled for 5 min at 95 °C in SDS loading buffer (LB) (Life Technologies) supplemented with 100 mM DTT (reducing condition). To detect HSPB1 dimers on the Western blots, cell lysates were boiled in non-reducing SB. Mouse monoclonal anti-HSPB1 antibody was purchased from Stressgen, mouse monoclonal anti-HSF1 was purchased from Santa Cruz Biotechnology, mouse monoclonal anti-ubiquitin (ubiquitinated proteins) was purchased from Cell Signaling, rabbit polyclonal anti-LC3 was purchased from Cell Signaling, mouse monoclonal anti-GFP (Jl-8) was purchased from Clontech, mouse monoclonal actin was purchased from Sigma-Aldrich and rabbit polyclonal anti-GAPDH was purchased from GeneTex.
Molecular modelling - Up to now, various crystallographic data of human heat-shock proteins
(HSPs) have been obtained and released on protein data bank, one of the latest being human HSPB1 α-crystallin domain (ACD; 84-176 residues) (pdb code: 4MJH; resolution = 2.60 Å), reported by Hochberg and collaborators (25). Since small HSPs share a highly conserved ACD, a whole human B1 theoretical model was performed by homology modelling techniques, on the basis of a multiple template alignment strategy focused on the X-ray data of the available HSPs.
Thus, HSPB1 model was obtained using the X-ray coordinates of human HSP40 (pdb code: 2QLD, resolution = 2.70 Å) (26), HSP70 (pdb code: 3JXU, resolution = 2.14 Å) (27) and HSP90 (pdb code: 3K97, resolution = 1.95 Å) (28), aligned together with that of HSP27 ACD (pdb code:4MJH; resolution = 2.60 Å). The amino acid sequence of HSPB1 (P04792) was retrieved from the SWISSPROT database (29) while the three-dimensional structure co-ordinates file of 27 ACD (pdb code: 4MJH) and 40-90 (pdb code: 2QLD, 3JXU and 3K97) were obtained from the Protein Data Bank (30).
The amino acid sequence of HSPB1 was aligned with the corresponding residues of the heat-shock proteins 40-90 by a multiple alignment procedure using MOE software (Chemical Computing Group Inc. Montreal. H3A2R7 Canada. http://www.chemcomp.comp). A refined alignment was determined on the basis of hydrophobicity similarity criteria, using the sequence editor tool in MOE software. The protein structure was minimized with MOE using the AMBER94 force field. The stereochemistry of the model has been validated through the analysis of Ramachandran plot, chi plot and clash contacts reports (Molecular Operating Environment suite). The HSPB1 mutant model was obtained by the residue rotamer explorer implemented in MOE sequence editor, followed by residue minimisation using AMBER99 force field. Aggregation-prone regions in the two models were calculated using the Protein Patch Analyzer module implemented in MOE software.
ACKNOWLEDGEMENTS
We sincerely appreciated the commitment of the patient described in this work and her family. We are also thankful to the Galliera Genetic Bank, member of “Network Telethon of Genetic Biobanks” (project no. GTB12001A) and of the EuroBioBank network, funded by Telethon Italy, for providing us with the LCL specimens.
This work was supported by PRA2014, University of Genoa (Italy) and by the ACMT-rete association to PM. VT is supported in part by the Association Belge contre les Maladies Neuromusculaires (ABMM), the Fund for Scientific Research Flanders (FWO) and the EC 7th Framework Programme under grant agreement number 2012-305121, ‘Integrated European–omics research project for diagnosis and therapy in rare neuromuscular and neurodegenerative diseases (NEUROMICS)’. SC and AG received a postdoctoral fellowship from the University of Genoa (Italy). TG received a PhD fellowship from the FWO.
CONFLICT OF INTEREST STATEMENT The authors declare no conflict of interest.
REFERENCES
1. Finsterer, J. and Burgunder, J.M. (2014) Recent progress in the genetics of motor neuron disease. Eur. J. Med. Genet., 57, 103-112.
2. Haslbeck, M. and Vierling, E. (2015) A first line of stress defense: small heat shock proteins and their function in protein homeostasis. J. Mol. Biol., 427, 1537-1548.
3. Cheroni, C., Peviani, M., Cascio, P., Debiasi, S., Monti, C. and Bendotti, C. (2005) Accumulation of human SOD1 and ubiquitinated deposits in the spinal cord of SOD1G93A mice during motor neuron disease progression correlates with a decrease of proteasome.
Neurobiol. Dis., 18, 509-522.
4. Sharp, P.S., Akbar, M.T., Bouri, S., Senda, A., Joshi, K., Chen, H.J., Latchman, D.S, Wells, D.J. and de Belleroche, J. (2008) Protective effects of heat shock protein 27 in a model of ALS occur in the early stages of disease progression. Neurobiol. Dis., 30, 42-55.
5. Vleminckx, V., Van Damme, P., Goffin, K., Delye, H., Van Den Bosch, L. and Robberecht, W. (2002) Upregulation of HSP27 in a transgenic model of ALS. J. Neuropathol. Exp.
Neurol., 61, 968-974.
6. Anagnostou, G., Akbar, M.T., Paul, P., Angelinetta, C., Steiner, T.J. and de Belleroche, J. (2010) Vesicle associated membrane protein B (VAPB) is decreased in ALS spinal cord.
Neurobiol. Aging, 31: 969-385.
7. Evgrafov, O.V., Mersiyanova, I., Irobi, J., Van Den Bosch, L., Dierick, I., Leung, C.L., Schagina, O., Verpoorten, N., Van Impe, K., Fedotov, V. et al. (2004) Mutant small heat-shock protein 27 causes axonal Charcot-Marie-Tooth disease and distal hereditary motor neuropathy. Nat. Genet., 36, 602-606.
8. Houlden, H., Laura, M., Wavrant-De Vrièze, F., Blake, J., Wood, N. and Reilly M.M. (2008) Mutations in the HSP27 (HSPB1) gene cause dominant, recessive, and sporadic distal HMN/CMT type 2. Neurology, 71, 1660-1668.
9. Dierick, I., Irobi, J., Janssens, S., Theuns, J., Lemmens, R., Jacobs, A., Corsmit, E., Hersmus, N., Van Den Bosch, L., Robberecht, W. et al. (2007) Genetic variant in the HSPB1 promoter region impairs the HSP27 stress response. Hum. Mutat., 28, 830.
10. Oesterreich, S., Hickey, E., Weber, L.A. and Fuqua, S.A. (1996) Basal regulatory promoter elements of the hsp27 gene in human breast cancer cells. Biochem. Biophys. Res. Commun., 222, 155-163.
11. Vos, M.J., Hageman, J., Carra, S. and Kampinga, H.H. (2008) Structural and functional diversities between members of the human HSPB, HSPH, HSPA, and DNAJ Chaperone Families. Biochemistry, 47, 7001-7011.
12. Almeida-Souza, L., Goethals, S., De Winter, V., Dierick, I., Gallardo, R., Van Durme, J., Irobi, J., Gettemans, J., Rousseau, F., Schymkowitz, J. et al. (2012) Increased monomerization of mutant HSPB1 leads to protein hyperactivity in Charcot-Marie-Tooth neuropathy. J. Biol. Chem., 285, 12778-12786.
13. Giese, K.C., and Vierling, E. (2002) Changes in oligomerization are essential for the chaperone activity of a small heat shock protein in vivo and in vitro. J. Biol. Chem., 277, 46310-46318.
14. Hartl, F.U., Bracher, A. and Hayer-Hartl, M. (2011) Molecular chaperones in protein folding and proteostasis. Nature, 475, 324-432.
15. Muchowski, P.J. (2002) Protein misfolding, amyloid formation, and neurodegeneration: a critical role for molecular chaperones? Neuron, 35, 9-12.
16. Kalmar, B., Lu, C.H. and Greensmith, L. (2014) The role of heat shock proteins in Amyotrophic Lateral Sclerosis: The therapeutic potential of Arimoclomol. Pharmacol. Ther., 141, 40-54.
17. Almeida-Souza L, Asselbergh B, d'Ydewalle C, Moonens K, Goethals S, De Winter V, et al. Small heat-shock protein HSPB1 mutants stabilize microtubules in Charcot-Marie-Tooth neuropathy. J. Neurosci. 2011; 31: 15320-8.
18. Deng, H.X., Chen, W., Hong, S.T., Boycott, K.M., Gorrie, G.H., Siddique, N., Yang, Y., Fecto, F., Shi, Y., Zhai, H. et al. (2011) Mutations in UBQLN2 cause dominant X-linked juvenile and adult-onset ALS and ALS/dementia. Nature, 477, 211-215.
19. Fecto, F., Yan, J., Vemula, S.P., Liu, E., Yang, Y., Chen, W., Zheng, J.G., Shi, Y., Siddique, N., Arrat, H. et al. (2011) SQSTM1 mutations in familial and sporadic amyotrophic lateral sclerosis. Arch. Neurol., 68, 1440-1446.
20. Leblond, C.S., Kaneb, H.M., Dion, P.A. and Rouleau, G.A. (2014) Dissection of genetic factors associated with amyotrophic lateral sclerosis. Exp. Neurol., 262 Pt B, 91-101.
21. Bandettini di Poggio, M., Sormani, M.P., Truffelli, R., Mandich, P., Origone, P., Verdiani, S., Mantero, V., Scialò, C., Schenone, A., Mancardi, G.L. et al. (2013) Clinical epidemiology of ALS in Liguria, Italy. Amyotroph. Lateral Scler. Frontotemporal Degener., 14, 52-57. 22. Brooks, B.R., Miller, R.G., Swash, M., Munsat, T.L., for the World Federation of Neurology
the diagnosis of amyotrophic lateral sclerosis. ALS and other motor neuron disorders, 1, 293-299.
23. Chen, C.Y., Chen, H.F., Gi, S.J., Chi, T.H., Cheng, C.K., Hsu, C.F., Ma, Y.S., Wei, Y.H., Liu, C.S. and Hsieh, M. (2011) Decreased heat shock protein 27 expression and altered autophagy in human cells harboring A8344G mitochondrial DNA mutation. Mitochondrion, 11, 739-749.
24. Salmon, P. and Trono, D. (2006) Production and titration of lentiviral vectors. Curr. Protoc.
Neurosci. Chapter 4: Unit 4.21.
25. Hochberg, G.K., Ecroyd, H., Liu, C., Cox, D., Cascio, D., Sawaya, M.R., Collier, M.P., Stroud, J., Carver, J.A., Baldwin, A.J. et al. (2014) The structured core domain of αB-crystallin can prevent amyloid fibrillation and associated toxicity. Proc. Nat. Acad. Sci. USA, 111, E1562-1570.
26. Hu, J., Wu, Y., Li, J., Qian, X., Fu, Z. and Sha, B. (2008) The crystal structure of the putative peptide-binding fragment from the human HSP40 protein Hdj1. BMC Struct. Biol., 8, 3. 27. Wisniewska, M., Karlberg, T., Lehtiö, L., Johansson, I., Kotenyova, T., Moche, M. and
Schüler, H. (2010) Crystal structures of the ATPase domains of four human HSP70 isoforms: HSPA1L/HSP70-hom, HSPA2/HSP70-2, HSPA6/HSP70B', and HSPA5/BiP/GRP78. PLoS
One, 5, e862511.
28. Kung, P.P., Huang, B., Zhang, G., Zhou, J.Z., Wang, J., Digits, J.A., Skaptason, J., Yamazaki, S., Neul, D., Zientek, M. et al. (2010) Dihydroxyphenylisoindoline amides as orally bioavailable inhibitors of the heat shock protein 90 (HSP90) molecular chaperone. J.
Med. Chem., 53, 499-503.
29. Bairoch, A. and Apweiler, R. (2000) The SWISS-PROT protein sequence database and its supplement TrEMBL in 2000. Nucleic Acids Res., 28, 45-48.
30. Berman, H.M., Westbrook, J., Feng, Z., Gilliland, G., Beth, T.N., Weissig, H., Feng, Z., Gilliland, G.L., Iype, L., Jain, S, et al. (2000) The protein data bank. Nucleic Acids Res., 28, 235-242.
LEGENDS TO FIGURES
Fig. 1: Molecular analysis of HSPB1
The pedigree of the sporadic patient ALS160 shows no affected members to the second degree of relatives (panel A). The electropherogram of the heterozygous frame shift mutation on genomic DNA is shown compared to a wild-type sequence. The shift in the HSPB1 open reading frame is depicted with the corresponding nucleotide difference between wild-type and mutant allele in the electropherogram data. The insertion of a guanine in position c.610_611 determines the translation of 8 additional base pairs in the 3’UTR of the gene, resulting in an extended protein (panel B). Multiple alignment of HSPB1 C-terminus in different orthologues shows that none of the selected species presents with an additional C-terminal segment, as is the case for the mutant p.Ala204Glyfs*6 (panel C).
Fig. 2: HSPB1 expression in LCLs
The expression of HSPB1 was evaluated at both mRNA and protein level in patient’s derived EBV-transformed lymphoblastoid cell lines (LCLs), compared to controls. Panel A shows the expression of the mutant allele on cDNA level, compared to a wild-type sequence. Panel B shows the relative expression of HSPB1 at the mRNA level in patients LCLs (ALS160) compared to controls. Control set A is composed of two ethnicity matched samples (Control 1 and Control 3, panel C), while control set B includes two age and gender matched samples (ceph884.16 and ceph1416.12). The relative mRNA quantification revealed a reduction in expression of the HSPB1 transcript although this is not statistically significant due to the high variability observed. According to this result, we decided to use Control 1 from control set A for the further experiments. Panel C shows the HSPB1 protein expression in patient compared to controls. While Control 1 to 4 show only one band, in patient ALS160 LCLs we could detect two bands (lane 5), resulting from the expression of the two alleles, encoding respectively the wild-type protein of 27 kDa (lower band) and the mutant protein predicted 27.3 kDa (upper band).
Fig. 3: HSPB1 p.Ala204Glyfs*6 sequesters the wild-type protein in a stable dimer
The upregulation of HSPB1 in both control and patient LCLs upon heat-shock is depicted in panel A. The untreated condition (NT) is compared to the heat-shock condition without recovery (HS noR) and to different recovery time points, respectively 2, 4, 6, 8, 12 and 24h (HS R2h to HS R24h). As a positive control, the membranes were probed for Heat-Shock Factor 1 (HSF1), known
molecular weight as depicted in panel A lanes 3 and 4. To evaluate the molecular weight of the dimeric structures observed in patient-derived LCLs, we compared this cell line with control LCLs and with three CHO-K1 stable cell lines expressing the constructs encoding for HSPB1 wild-type, mutant or a combination of both (panel B). The naïve CHO-K1 cell line does not express HSPB1, thus allowing to evaluate its dimerization in the absence of the endogenous protein. The dimer-monomer shuttling has been evaluated in patient-derived LCLs compared to control (panel C). Basal conditions are compared to post-stress conditions (HS R4h) in both non-reducing (nred) and reducing (red) settings. Panel D shows the patch surfaces of HSPB1 wild-type (a) and mutant (b) protein. Hydrophobic (HY1 and HY2), positive (PC) and negative-charged (NC) exposed areas are depicted in green, blue and red, respectively.
Fig. 4: HSPB1 p.Ala204Glyfs*6 results in a loss of chaperone-like activity
The chaperone-like activity of HSPB1 p.Ala204Glyfs*6 has been evaluated as its ability to clear SOD1 p.Ala4Val-GFP aggregates in double stable CHO-K1 cell lines. The clearance of SOD1 aggregates determines a different solubility of the protein, reducing the signal in the pellet fraction. HSPB1 wild-type (wt) is able to clear the SOD1 p.Ala4Val aggregates in double stable CHO-K1 cell lines as shown in lane 7. When comparing the CHO-K1 lines expressing HSPB1 wt (lane 7) to the one expressing p.Ala204Glyfs*6 (mut) (lane 8), we showed that the presence of the mutation determines an increase of the GFP signal in the pellet fraction, suggesting that the mutant protein is not able to clear the SOD1 aggregates.
Supplementary Figure 1: HSPB1 p.Ala204Glyfs*6 degradation
The possible degradation of HSPB1 p.Als204Glyfs*6 has been evaluated by inhibiting the proteasomal or lysosomal pathways. Panel A shows the proteasomal inhibition with MG132 (25 µM). The untreated condition (-, lanes 1 and 7) is compared to 2, 4, 6, 8, 24h MG132 treatment. The efficacy of the inhibition has been confirmed by the increase in ubiquitinated proteins. The lysosomal degradation has been inhibited using Leupeptin (40 µM) (panel B), comparing the untreated condition (-, lanes 1 and 5) to 24, 48 and 72h Leupeptin treatment. The inhibition of lysosomal-dependent degradation pathway results in the accumulation of LC3-II (lower band, LC3 probing), used as treatment control.
ABBREVIATIONS
ALS Amyotrophic Lateral Sclerosis
sALS sporadic Amyotrophic Lateral Sclerosis fALS familial Amyotrophic Lateral Sclerosis HSPB1 small Heat-Shock Protein B1 CMT Charcot-Marie-Tooth disease
LCL EBV-transformed Lymphoblastoid Cell Line HS Heat-Shock
HY1 highly HYdrophobic region 1 NC Negative Charged area PC Positive Charged area SOD1 SuperOxide Dismutase 1
