The role of survival motor neuron protein (SMN) in protein homeostasis

The role of survival motor neuron protein (SMN) in protein homeostasis Ever since loss of survival motor neuron (SMN) protein was identified as the direct cause of the childhood inherited neuro - degenerative disorder spinal muscular atrophy, significant efforts have been made to reveal the molecular functions of this ubiquitously expressed protein. Resulting research demonstrated that SMN plays important roles in multiple fundamental cellular homeostatic pathways, including a well-characterised role in the assembly of the spliceosome and biogenesis of ribonucleoproteins. More recent studies have shown that SMN is also involved in other housekeeping processes, including mRNA trafficking and local translation, cytoskeletal dynamics, endocytosis and autophagy. Moreover, SMN has been shown to influence mitochondria and bioenergetic pathways as well as regulate function of the ubiquitin–proteasome system. In this review, we summarise these diverse functions of SMN, confirming its key role in maintenance of the homeostatic environ - ment of the cell. Keywords Spinal muscular atrophy · Ribonucleoprotein · Translation · Cytoskeleton · Ubiquitin · Bioenergetic pathway Introduction “survival motor neuron”, despite subsequent research show- ing that it is a ubiquitously expressed protein, required by The survival motor neuron (SMN) protein was first high - all cells and tissue types, not just neurons [3]. Over the past lighted as a protein of interest when mutations in its cod- three decades, significant research efforts have sought to bet - ing gene, SMN1, were linked to the neuromuscular disease ter understand the mechanisms through which SMN acts [3]. spinal muscular atrophy (SMA) [1], a leading genetic cause Most of the resulting knowledge has been generated from of infant mortality. SMA presents in a range of severities animal models of SMA, where reduced expression of SMN with the most severe form, Type 1, being fatal within the reveals its role in several important intracellular processes, first 2 years of life. Patients show degeneration of α-motor which we will discuss in this review. neurons in the lower spinal cord leading to progressive mus- The full-length—294 amino acids, 38 kDa—human iso- cle weakness. The clear importance of SMN protein to the form of SMN (also known as FL-SMN, referred to as SMN motor system, alongside findings that knockout of Smn in hereafter) is mainly transcribed from the telomeric SMN1 mice was embryonically lethal [2], led to it being named gene, located on chromosome 5q13. SMN1 contains nine exons, 1, 2a, 2b, 3, 4, 5, 6, 7 and 8, with exon 8 remain- ing untranslated (Table 1). An inverted duplication in the Helena Chaytow, Yu-Ting Huang and Kiterie M. E. Faller region of SMN1 resulted in a second centromeric copy of the contributed equally. gene, termed SMN2, an evolutionarily recent event unique to Homo sapiens [4]. SMN2 differs from SMN1 at 5 bases, * Thomas H. Gillingwater and a C-to-T transition in exon 7 of SMN2 favours skipping t.gillingwater@ed.ac.uk of exon 7 during splicing, resulting in the majority of SMN2 Euan MacDonald Centre for Motor Neurone Disease products being a truncated isoform referred as SMNΔ7 [1, Research, University of Edinburgh, Edinburgh, UK 5]. However, limited amounts of SMN can still be produced Edinburgh Medical School: Biomedical Sciences, University from the SMN2 gene and it is known that the copy number of of Edinburgh, Edinburgh, UK SMN2 is inversely correlated with SMA disease severity [6]. Royal (Dick) School of Veterinary Studies, University Patients with homozygous null mutations of SMN1 carrying of Edinburgh, Edinburgh, UK Vol.:(0123456789) 1 3 H. Chaytow et al. Table 1 Main isoforms of SMN, their composition, expression and localisation SMN isoform Splicing Protein isoform Expression Localisation References Full-length SMN (FL- Exons 1, 2a, 2b, 3, 4, 5, Functional SMN protein High expression during Nuclear gems and cyto- [102, 205] SMN) 6, 7, 8 development, decreas- solic, including axons, ing into the adult CNS dendrites and synapses SMNΔ7 Exons 1, 2a, 2b, 3, 4, Degradation signal intro- High expression during Nuclear accumulation [104, 206] 5, 6, 8 duced by the change in development, decreas- C-terminal ing into the adult CNS Axonal-SMN (a-SMN) Inclusion of intron 3 Truncated protein due to Expressed during devel- Motor neuron axons [12] premature stop codon opment, not detected on the boundary of in the adult CNS exon 3/intron 3 SMN6B Inclusion of an Alu ele- Truncated protein due to Unknown Nuclear and cytosolic [10] ment forming exon 6B premature stop codon after exon 6B SMNΔ5 Exclusion of exon 5 Unknown Expressed in the mature Unknown [11] CNS Other splicing isoforms of SMN have also been discovered in cell cultures, although their role in  vivo is yet to be determined. These include isoforms excluding exons 3, 4 and 5 or multiple exons both in stressed and normal conditions [207, 208]. Skipping of any internal exons of SMN maintains the reading frame four or more copies of SMN2 show a less severe phenotype, highly conserved motif with a function in protein–protein later age onset, and can have a normal lifespan [7]. SMNΔ7 interactions [17, 18]. The SMN Tudor domain binds to the is highly unstable and quickly subjected to the ubiquitin–pro- C-terminal arginine- and glycine-rich tails of Sm proteins teasome pathway for degradation [8, 9]. Phylogenetic studies which contain symmetrical dimethylated arginine residues, further highlighted the importance of SMN, since it is highly thereby facilitating the assembly of spliceosomes as dis- conserved throughout evolution and there are even multiple cussed later [18–24]. Mutations in this domain, which dis- copies of SMN1 in the chimpanzee genome [4]. Other SMN rupt the interaction of SMN and Smith core (Sm) proteins, isoforms have been found in various tissues (Table 1). An are often found in SMA patients [18, 25–27]. “SMN6B” protein can be translated from both the SMN1 and The Tudor domain of SMN is also responsible for an SMN2 genes by the inclusion of an intronic Alu sequence interaction with coilin, a marker of Cajal bodies (CBs) [28]. as an alternative exon following exon 6 [10]. SMN6B is Liu and Dreyfuss originally described the localisation of present both in the nucleus and cytosol, and is twofold more SMN to nuclear bodies which they termed “gems” [29], stable than SMNΔ7 but twofold less stable than full-length and which are coilin negative as opposed to CBs. Gems are SMN. SMN6B is able to interact with Gemin2, although its composed of SMN complex proteins, whereas CBs are more physiological function is not fully understood [10]. mRNA complex nuclear structures to which SMN also localises, transcripts of another isoform, SMNΔ5 (with the exclusion and the presence of SMN in these nuclear bodies increases of exon 5), can be found in muscle and the central nervous during neuronal differentiation [30]. CBs are enriched with system (CNS) although, again, whether it has a physiologi- small nuclear RNAs (snRNAs) and small nucleolar RNAs cal function is not clear [11]. An axonal-SMN (a-SMN) has (snoRNAs), and are essential for the biogenesis of the small also been proposed, being produced from intron 3 retention nuclear ribonucleoproteins (snRNP) complex [31]. Interest- during splicing. Due to an in-frame stop codon in intron 3, ingly, motor neurons from a Type I SMA patient showed a a-SMN mRNA is much shorter and encodes a protein of reduced number of CBs and defects in recruitment of SMN 19 kDa. a-SMN is reported to be localised to the axon, and and snRNP for spliceosomal maturation [32]. Tapia et al. its expression is enhanced in the spinal cord and the brain [33] also demonstrated that SMN has a SUMO-interacting during development and declines in the adulthood, with a motif (SIM) in the Tudor domain, which is required for Sm hypothesised role in axonogenesis [12]. protein binding and the assembly of CBs. Three polypro- The SMN protein contains several functional motifs line stretches encoded by exons 4–6 are responsible for the including, moving from N-terminus to C-terminus, a basic/ binding of profilins, key proteins in the regulation of actin lysine-rich domain, a Tudor domain, a proline-rich domain dynamics [34, 35]. A tyrosine/glycine-rich region in the and a YG box. The basic/lysine-rich region is encoded by C-terminus of SMN, termed the YG box, is found to facili- exon 2 and has been found to interact with Gemin2 and tate oligomerisation of SMN by formation of the glycine- RNA in vitro and in vivo [13–16]. The Tudor domain is a zipper structure [36]. Mutations found in the YG box count 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis for nearly half of missense mutations in SMA patients and a cDNA copy of SMN2 lacking exon 7 or other variants of this motif was shown to be essential for survival in chick the SMN genes [43–45] (reviewed in [46]). Another widely cells under conditions of SMN depletion [37, 38]. The YG used animal model is the zebrafish (Danio rerio). Having the box is required for SMN self-oligomerisation and proteins advantages of well-characterised motor neurons and neuro- with mutations found in this motif, as seen in the Type I muscular junctions, easy manipulation of gene overexpres- SMA patients, severely impair this association [39]. A recent sion or knockdown by injection of in vitro-transcribed RNA study demonstrated that a phosphor degron within the YG or antisense oligonucleotides, respectively, and a transpar- Slmb box is exposed to SCF ubiquitin E3 ligase when SMN is ent body for imaging applications, zebrafish are becoming monomeric, implying that the YG box plays a role in SMN increasingly popular for translational SMA research pro- stability and indicating the importance of SMN oligomerisa- jects [47–50]. Fruit flies (Drosophila melanogaster) have tion [9]. Post-translational modic fi ations to the SMN protein also been used to study SMN biology. They possess one discern its localisation and function. As well as SUMOyla- copy of an Smn gene ortholog or DmSMN and can also be tion via the SIM regulating Sm protein binding, and ubiq- easily genetically manipulated. Each of the models referred uitination of SMN (discussed later), the protein may also to in this review are summarised in Table 2, with more com- be acetylated, which promotes a cytoplasmic localisation prehensive reviews of SMA models available elsewhere and increases its half-life [40], or phosphorylated on certain [51–54]. serine/threonine residues to promote its localisation to CBs. The selective cell death of motor neurons is a key feature Mutations of three tyrosine residues in the Tudor domain of the disease, but the reasons behind this selectivity are greatly affect its enrichment in CBs via disrupted interaction still poorly understood. A recent study demonstrated that with coilin [41, 42]. there were heterogeneous and surprisingly diverse expres- Several model organisms have been utilised to study sion levels of SMN in SMA-patient-derived iPSC motor the SMN protein and its role in SMA (Table 2). Although neurons. Moreover, motor neurons with lower levels of the protein product of SMN2 is truncated and unstable, its SMN protein were more susceptible to cell death from toxic expression is crucial for survival once SMN1 expression compounds, whilst overexpression of SMN in motor neu- has been lost. As SMN2 is specific to humans, most of the rons was protective [55]. SMN, therefore, clearly plays a commonly used mouse models have undergone genetic major role in SMA pathology and the specific vulnerability manipulation to generate an endogenous Smn null mutation to motor neurons in this disease. To understand why SMN with concurrent overexpression of the human SMN2 gene, is so vital for healthy cell maintenance, we must understand Table 2 Overview of animal models referred to in this review Species Endogenous Modelling strategy and/or genotype References SMN ortho- logue Caenorhabditis elegans CeSMN Knockdown of expression through RNAi [209] Drosophila melanogaster DmSMN Point mutations or transposon insertions for knockout or knockdown studies [210, 211] Danio rerio Smn Knockdown of expression through antisense oligonucleotides [47] Mus musculus Smn −/−  Smn knockout Smn Smn null mutation by targeted insertion of [2] β-galactosidase in Smn exon 2A H7/H7 tg/−  Taiwanese mice Smn ; SMN2Hung Two copies of the SMN2 transgene, Smn [43] exon 7 is replaced with hypoxanthine phosphoribosyl transferase (HPRT) but transcripts without exon 7 are produced. 2A/2A tg/tg tg/tg  SMNΔ7 Smn ; SMN2 ; SMN∆7 One copy of the SMN2 transgene and one [45] SMNΔ7 transgene on Smn null back- ground 2B/−  Smn2B Smn Mutation within the splicing enhancer of [212] Smn exon 7 producing SMN2-like tran- scripts and reduced FL-SMN protein −/− tg/tg  Burghes severe model Smn ; SMN2 Smn null mutation by target replace of [44] β-galactosidase in Smn exon 2A; with one copy of SMN2 For a comprehensive review of animal models of SMA, see Edens et al. [51] 1 3 H. Chaytow et al. its role under normal, as well as disease, conditions. In this to the TMG cap [72, 74–77]. In that process, SMN has been review, we describe the role of the SMN protein in regulat- shown to have a direct interaction with importin-β facilitated ing protein homeostasis. Protein homeostasis within cells by WRAP 53 [78, 79]. WRAP 53 also plays a fundamental can be regulated by two major pathways, production and role in the trafficking of SMN towards CBs by facilitating clearance, which reach a dynamic balance to support and the interaction between SMN and coilin [80]. The snRNA maintain physiological status. Production pathways incorpo- then dissociates from the SMN complex and undergoes its rate protein translation, folding, modification and assembly, final maturation steps within the CB. Studies of splicing while protein clearance pathways are centred around protein activity in cells from SMA patients or mouse models con- disassembly and degradation. We discuss known functions firm the fundamental role of SMN in snRNP assembly with of the SMN protein, starting with its first-described role in a correlation between the reduction in snRNPs levels and RNA splicing and spliceosomal assembly, followed by more disease severity [81–83]. A recent study has also identified recently discovered functions in regulating mitochondrial an alternative assembly pathway, whereby the U1-specific homeostasis, endocytosis and the cytoskeleton, ubiquitina- RNA-binding protein (RBP) U1-70K can directly interact tion and autophagy, and RNA transportation, thus giving a with the SMN-Gemin2 complex, independently of Gemin5. broad picture of the many ways in which SMN plays a key This U1-specific Sm core-assembly pathway not only con- role in regulating protein homeostasis. tributes to U1 overabundance, but it was also proposed that SMN-Gemin2 could play a role as a hub, where various RBPs and their RNA cargos congregate, hence promoting SMN and ribonucleoprotein assembly ribonucleoprotein exchange [84]. The SMN complex is also involved in the biogenesis Although it is now clear that the SMN protein contrib- of U7 snRNPs, a specific subgroup of snRNPs which are utes to numerous cellular processes and pathways, the first involved in processing the 3′ stem loop of histone mRNAs identified and most studied function of SMN is its role in by endonucleolytic cleavage of the pre-mRNA sequence snRNP assembly. The spliceosome is a complex machine, which immediately follows the hairpin [85]. The assembly which catalyses the removal of introns from pre-mRNA of U7 snRNPs is overall analogous to that of the spliceo- transcripts (see [56] for a detailed review). The biogene- some snRNPs, with the exception of the slightly degenerate sis of an snRNP starts in the nucleus by the transcription Sm-binding site of the U7 snRNA and the replacement of of uridine-rich snRNAs (U1, U2, U4, U5, U6, U11, U12, two of the Sm proteins in the Sm core (SmD1 and SmD2) U4atac, U6atac), which are then exported to the cytoplasm. by two U7-specific Sm-like proteins (Lsm10 and Lsm11). Each snRNA is then bound to accompanying proteins and— Similar to spliceosome assembly, the SMN complex is a with the exception of U6 and U6atac—to a set of seven Sm specificity chaperone that is necessary to precisely recognise proteins. Whilst Sm proteins can spontaneously associate and combine U7 snRNA with the Sm heptamers containing in vitro with almost any single short-stranded uridine-rich Lsm10 and Lsm11, without which the U7 snRNPs cannot RNA forming a thermodynamically stable heptamer [57], function in histone RNA processing [86–88]. this process lacks specificity regarding RNA substrates. Although less extensively studied, the SMN complex is The role of the SMN protein, tightly associated with eight suspected to be involved in the assembly and metabolism of other proteins (Gemin 2–8 and Unrip) in a large macromo- other ribonucleotide complexes, including small nucleolar lecular SMN complex, is to chaperone the biogenesis of ribonucleoproteins (snoRNPs), associated with the post- snRNPs from snRNAs and Sm proteins in the cytoplasm, transcriptional processing and modification of ribosomal and subsequent snRNP trafficking to the nucleus [14, 29, RNA in the nucleolus (methylation and pseudouridylation) 58–66]. First, the SMN complex enforces specificity during [89]. Indeed, SMN has been shown to directly interact with the snRNPs’ assembly with the direct and specific binding fibrillarin and GAR1, two markers of snoRNPs, and expres- of Gemin5 to the cytoplasmic snRNAs [67–71]. The Sm sion of a dominant-negative mutant of SMN results in abnor- core is then loaded onto the snRNA in an ATP-dependent mal accumulation of snoRNPs in large aggregates outside process with Gemin2 playing a key architectural role in this of the nucleolus [90]. Furthermore, in SMA-patient-derived assembly. Finally, the snRNA undergoes hypermethylation cells, a decreased localisation in CBs of the snoRNP chaper- of its m7G-cap by TGS1 (trimethylguanosine synthetase one Nopp140 was observed, which correlated with disease 1), leading to the formation of a trimethylguanosine (TMG) severity [91]. In addition, SMN may be involved with tel- cap and trimming of its 3′ end before it can be imported omerase, a large RNP complex that adds repeat sequences back into the nucleus. The TMG cap and Sm core operate at the chromosomal ends. It comprises a telomerase reverse as nuclear localisation signals [72, 73]. Importation to the transcriptase (TERT), the telomerase RNA and other associ- nucleus also necessitates the binding of the nuclear import ated proteins (for a review on telomerase RNA, see [92]). complex (snurportin and nuclear import receptor importin-β) The telomerase RNP belongs to the H/ACA snoRNP class 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis and it is suspected that SMN plays a role in telomerase bio- Recent studies have identified that SMN can bind to the genesis, ensuring specificity of assembly and correct traf- α-COP subunit of the COPI vesicle [107]. The COPI system, ficking [93, 94]. a Golgi-derived vesicular transport system, is involved in As snRNP assembly and splicing occurs in all cells, intracellular trafficking in neurites, necessary for the mat- including neurons, why do low levels of SMN in SMA par- uration of neuronal cell processes [108]. Knocking down ticularly affect motor neurons [95]? This remains a major α-COP was found to disrupt SMN localisation within growth question challenging the SMN and SMA research field. As cones, resulting in its accumulation within the trans-golgi previously noted, the reduction in snRNP biogenesis cor- network [109]. Depletion of α-COP reduced neurite forma- relates with the degree of clinical severity in SMA [81]. tion in NSC-34 cells and primary cortical neurons, with However, SMN deficiency seems to preferentially disrupt shortening of both map2-positive dendrites and tau-positive the formation of minor snRNPs, such as those responsible axons [110], and both α-COP and SMN are required for cor- for the removal of U12-containing intron genes [81, 83, rect neurite formation [111]. This indicates a role for SMN 96]. Amongst these, the gene coding for a transmembrane in trafficking for the purposes of neuronal outgrowth and for - protein, stasimon, has been identified as being aberrantly mation of the axonal and synaptic cytoskeleton (see below). spliced in a Drosophila model of SMA [97]. Upregula- In keeping with this potential role for SMN, Rossoll and tion of stasimon rescued deficient neuromuscular junc- colleagues discovered an interaction between SMN and tion (NMJ) transmission in SMN-deficient Drosophila and the RBP hnRNP-R [112]. SMN and hnRNP-R were found improved neuronal development in SMN-deficient zebrafish to co-localise in the cytoplasm of primary cultured motor −/− [97]. Other genes, not all containing U12-introns, such as neurons, and in motor neurons cultures from Smn mice, chondrolectin, agrin and neurexin2 have also been identi- there was a large reduction in β-actin mRNA localisation in fied as being abnormally spliced and could, therefore, play axons and growth cones. Primary motor neurons cultured a role in the pathophysiology of the disease [98–100]. It from Taiwanese SMA mice showed growth cones with a could consequently be the case that defects in splicing have threefold reduction in size compared to healthy controls, as a larger effect on a specific subset of neuronal genes, thereby well as reduced staining for β-actin mRNA with no overall rendering motor neurons particularly vulnerable. However, change in protein expression [112]. Since these initial find- despite this evidence for mis-splicing in the SMA disease ings, fluorescence in situ hybridisation experiments against pathway, other studies have suggested that wide-spread the polyA tails of mRNA revealed a more than 50% reduc- splicing defects mainly occur during the late stage of the tion in localisation of mRNA transcripts along the axon of disease [101], supporting the theory that alternative roles primary motor neurons following SMN knockdown [113]. of SMN may play an equally important part in cell function. In addition, further co-localisation studies have shown SMN to associate with a number of RBPs via its Tudor domain, including KSRP [114], FMRP [115], HuD [113, 116], FUS SMN and trafficking [117] and IMP1 [118]. The association between SMN and other RBPs has linked it to another motor neuron disease, The first indications that SMN played a role aside from its amyotrophic lateral sclerosis (ALS). RBPs associated with canonical functions in the spliceosome came when electron ALS, FUS and TDP-43 have been shown to co-localise in microscopy revealed localisation of SMN in the dendrites nuclear gems with SMN and mutations in either of their and axons of motor neurons in the developing rat spinal cord genes in ALS patient fibroblasts show reduced gem forma- [102]. It has been suggested that there is a progressive shift tion leading to abnormal accumulation of snRNAs in the in SMN protein localisation from mainly nuclear during nucleus [119, 120]. This highlights an interesting mecha- development to a more cytoplasmic and axonal localisa- nistic link between ALS and SMA. tion in the mature neuron [103]. SMN was also found to be SMN acts as a molecular chaperone for the binding of present at the growth cones of cultured motor neurons, and RBPs to mRNA transcripts as well the RBPs’ binding to the live cell imaging showed puncta positive for SMN being cytoskeleton and subsequent localisation, as evidenced by actively transported bi-directionally along axons [104]. disruption of these processes in SMN deficiency [121]. Both SMN co-localises with some elements of the SMN complex IMP1 and HuD have been shown to influence the locali- in the axon, such as Gemin2, but Sm proteins show very sation and translation of β-actin and GAP-43 mRNA tran- low abundance in distal neurites, and most axonally located scripts, which are in turn both necessary for correct axonal SMN granules lack Sm proteins [105]. The neuron-specific growth [113, 118, 122]. Indeed, SMN knockdown leads to protein neurochondrin is required for the correct localisation a reduction of HuD in the axonal compartment [113], while of SMN in the cytoplasm, and neurochondrin was found not knockdown of HuD in zebrafish leads to a similar phenotype to co-localise with snRNPs, further indicating that SMN is to SMN knockdown of shorter axons [122]. Further experi- involved in activities other than splicing [106]. ments using FLAG-tagged SMN in NSC-34 cells [123] 1 3 H. Chaytow et al. identified SMN as a binding partner for several species of transcribed in a cluster and are associated with increased non-coding RNAs, including snRNAs, snoRNAs and ribo- cell proliferation via the mTOR pathway [130]. Primary somal RNAs, which were expected due to SMN’s known motor neurons with a 50% knockdown of SMN pro- role in RNA processing, as well as miRNAs and tRNAs. tein showed increased expression of miR-183 in neu- The majority of RNAs identified were mRNA, with many rites whereas there was no change in expression in the being part of ribosomal and/or metabolic pathways. When cell body, along with downregulation of proteins in the compared to mRNAs known to localise to neuronal axons, mTOR pathway [131]. It is possible that SMN regulates the study’s authors identified 75 axonally localised SMN- the mTOR pathway and, therefore, protein translation associated mRNAs, including RNA transcripts of several through miR-183, since motor neurons in the Taiwanese ribosomal proteins and Ubb, the transcript of the protein mouse model and SMA-patient-derived fibroblasts both ubiquitin [123]. However, it should be noted that this group showed reduced levels of de novo protein synthesis, and of mRNAs is unlikely to be comprehensive, as it does not a knockdown of miR-183 in Taiwanese mice produced include the known RNA transcripts regulated by SMN a mild rescue of the phenotype with improved survival including β-actin, GAP43, tau and neuritin [112, 116, 124]. and increased body weight [131]. Another indication that SMN interacts with the mTOR pathway came from studies examining the effect of manipulating the PTEN pathway SMN and translation on primary motor neurons of SMNΔ7 mice. PTEN is a negative regulator of the mTOR pathway, and in SMN- While SMN plays an integral role in the transport of RNA depleted primary motor neurons where axonal growth was transcripts along axons and dendrites, it also appears to be defective, decrease of PTEN/activation of mTOR rescued involved in the local translation of proteins. The transporta- the SMA phenotype [132]. tion of mRNA transcripts along the axon allows for rapid Most of the work detailed above was performed in vitro, protein turnover in distal regions of the neuron in response where primary cultures allow analysis of changes in axonal to, for example, activity [125]. Dysregulation of local trans- growth and the ability to isolate axonal compartments lation has been associated with several other neurodegen- relatively easily. However, in vivo evidence is important erative disorders, including Alzheimer’s disease and amyo- to determine the role of SMN in these mechanisms. A trophic lateral sclerosis (reviewed in [126]). recent study examined translational pathways in vivo and Recent evidence has pointed to a role for SMN in the local found SMN associating with ribosomes in control tissue, translation of mRNA transcripts, as well as their localisa- as well as a shift in residual SMN levels to non-ribosomal tion. Early on, it was reported that loss of SMN changed the fractions and an overall reduction in the number of ribo- expression of plastin-3 in a zebrafish model of SMA [127]. somes associated with polysomes in Taiwanese SMA mice Although, at the time of this discovery, the mechanisms [133]. This study also compared the whole transcriptome behind changes in protein expression were not clear, more to the translatome using next-generation sequencing, and recent studies suggest that SMN affects the local translatome confirmed significant deficiencies in translation following through several mechanisms. Translation within the axon reduced SMN expression in vivo [133], including results was found to be dysregulated in primary neurons derived that point to defects in the biogenesis of ribosomes, sug- −/− from SMN mice in vitro, with no corresponding change in gesting a possible explanation for translational defects that somatic translation, and axonal defects could be rescued by occur upon SMN depletion. overexpression of the RNA binding proteins HuD and IMP1 Taken together, the studies detailed above strongly sug- [128], suggesting a link between the role of SMN in mRNA gest that SMN plays an important role in regulating pro- trafficking and translation. Furthermore, ultrafractionation of tein translation through several mechanisms. First, through cell extracts from a motor neuron-like cell line revealed an subcellular localisation of mRNAs along the axon; second, association of the SMN protein with polyribosomes, whilst through association with ribosomes themselves regulating treatment with RNase displaced RBPs associated with the the availability of ribosomal units for local translation and polyribosomes such as SMN, but also other known binding finally, through regulation of the mTOR pathway. In this proteins such as FMRP [129]. When SMN was introduced way, SMN is well positioned to play a role in the develop- to an in vitro translation system, there was a dose-dependent mental polarisation of motor neurons, as well as control reduction in translation efficiency, with no change in transla - their growth, maturation and proper function. This crucial tion when incubated with the SMNΔ7 fragment [129]. role, alongside the expansive physical size of motor neu- Alongside its direct interaction with ribosomes, sug- rons, may partly explain why motor neurons are particu- gesting a possible direct role in translational control, SMN larly susceptible to loss of SMN, as opposed to other more may also influence protein translation through micro- ubiquitous roles that the protein plays in the cell. RNAs. MicroRNAs miR-183, miR-96 and miR-182 are 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis its functional change from being an attractant to a repel- SMN and the cytoskeleton lent signalling factor, thereby contributing to growth cone collapse [145]. In the same study, cleaved PlexinD1 was The cytoskeleton—incorporating key components such as found to be enriched in actin rods, a pathological struc- microtubules, neurofilaments and actin protein—plays a ture of elongated actin aggregates also found in some age- fundamental role in regulating neuronal architecture and related neurodegenerative diseases but not in control cells. function. It is crucial for signalling and trafficking of vari- Interestingly, by studying discordant families where sib- ous molecules, but also for the formation of growth cones lings of SMA patients were asymptomatic despite carrying during neuronal development. Therefore, it is perhaps not the same SMN1 and SMN2 alleles as their affected siblings, surprising that defects in the cytoskeleton have been linked Oprea et al. [146] were able to identify the first protective to several neurodegenerative disorders, including SMA genetic modifier of SMA: plastin 3 (PLS3). PLS3 is involved (for reviews on the neuronal cytoskeleton see [134, 135]). in axonogenesis by bundling F-actin and stabilising growth The observation that SMN localised to neurites and cones. Its overexpression was able to rescue the axon out- growth cones [102, 105, 136–138] and that SMN modu- growth defects in SMN-deficient zebrafish and increase lated the localisation of β-actin within growth cones [112] the life span of the intermediate Smn2B model [146, 147]. provided the first hints of a possible role for SMN in regu- Other studies have suggested additional roles for SMN in lating cytoskeletal dynamics. At the same time, knock- regulating microtubule formation, required for transporting ing down SMN in zebrafish was found to result in motor mRNAs, proteins and organelles to or from the nucleus to axon pathfinding deficits [47], whilst SMN-deficient cell distal regions of the neuron (reviewed in [148]). Stathmin, cultures showed neurite extension defects [104, 112, 139, a protein known to promote microtubule depolymerisation 140]. The fact that selective overexpression of the SMN [149] was found to be upregulated in the spinal cord in Tai- C-terminal domain could rescue these neurite deficits in wanese mice and also in SMN-depleted NSC-34 cells lead- SMN-deficient PC12 cells argued in favour of a role of ing to defects in the structure of axons and reduced mito- SMN in microfilament metabolism independent of snRNP chondrial transport along the axons [150]. In SMN-deficient biogenesis, as the Tudor domain where Sm proteins binds cells, microtubules failed to re-polymerise following treat- was not present in this C-terminal construct [139]. ment with the microtubule-depolymerisation agent nocoda- Growth cone formation and neurite extension are medi- zole, an effect which could be rescued by knocking down ated by actin dynamics, and SMN has been found to colo- stathmin [150]. However, the detailed mechanisms linking calize with profilin 2a during neurite cell extension [141] SMN to these pathways, and whether or not they are indeed and in nuclear gems [34]. Profilin 2a is an actin-binding separate from SMN’s involvement in the mRNA traffick - protein primarily expressed in the nervous system where ing of components of the cytoskeleton such as GAP-43 and it is involved in the regulation of actin turnover by pro- β-actin, remains to be determined. moting actin polymerisation. Profilin 2a binds to a stretch of proline residue within the SMN protein [35], and this interaction with SMN modulates the activity of profilin. SMN and endocytosis Profilin 2a is also a known downstream target of the rho- kinase (ROCK) pathway, a key regulator of actin dynamics Endocytosis is a basic cellular process, essential for neuronal (reviewed in [142]). Knocking down SMN in PC12 cells signalling, axonal and dendritic growth (reviewed in [151]). resulted in an upregulation of profilin 2a, which, combined It plays a particularly important role at synapses (including with its increased availability due to decreased interaction at neuromuscular junction synapses formed by motor neu- with SMN, lead to an upregulation of the ROCK pathway rons), facilitating synaptic vesicle recycling necessary for with subsequent inhibition of neuronal outgrowth [143]. repeated rounds of neurotransmitter release. A bioinformat- ROCK pathway inhibition in an intermediate SMA mouse ics analysis carried out on two different species (Caenorhab- model (Smn2B) also resulted in increased life span and ditis elegans and D. melanogaster) identified the endocytic amelioration of muscle pathology [144] (see Fig.  1 for pathway, along with mRNA regulation, as potential modi- a summary of the role of SMN in cytoskeleton dynam- fiers of SMN loss [152], with numerous individual genes ics). Moreover, it was recently suggested that SMN loss being highlighted. In another study, SMN depletion resulted resulted in the dysregulation of the actin cytoskeleton in a marked impairment of endocytic function in multiple by interfering with PlexinD1. PlexinD1 is a receptor for tissues of C. elegans [153]. The neuromuscular junction was class 3 semaphorins and acts as a signalling factor to guide particularly affected, with structural and functional changes axonal growth. In the Taiwanese mouse model and in being reported. A reduction in the number of pre-synaptic iPSC-derived motor neurons from SMA patients, PlexinD1 docked vesicles was observed, accompanied by unusually was shown to be cleaved by metalloproteases, resulting in large cisternae suggestive of arrested endocytic vesicle 1 3 P lin 2a H. Chaytow et al. Kinesin RNA granule (Pre)synaptic vesicle Clathrin TRANSPORT Plastin 3 Microtubules F-actin G-actin Mitochondrion Ion channel Direct inhibition by SMN In dark blue: changes associated with SMN loss Microtubule destabilisation SMN Actin depolymerisation ENDOCYTOSIS 2+ Ca Fig. 1 Schematic overview of the alterations in cytoskeletal dynamics transport (e.g. synaptic vesicle, RNA granules and mitochondria) and and endocytosis observed following SMN deficiency. The diagram endocytosis. In the absence of SMN, not only a destabilisation of the highlights these changes at the level of the motor neuron and neuro- microtubules is observed, but also a depolymerisation of the actin muscular junction. All changes associated with SMN loss are repre- cytoskeleton, which has been linked to the activation of the RhoA/ sented in dark blue. SMN deficiency results in a decrease in cellular ROCK pathway maturation [153]. This was associated with a decreased In this model, endosomal defects were noted not only at the activity of, and disruption to, the NMJ (a key feature of level of the NMJ, but also in non-neural tissue as endocytic SMA [154–156]): synaptic transmission was reduced, likely activity in coelomocyte cells was lower. The importance of secondary to an impairment in synaptic vesicle recycling. SMN for NMJ homeostasis was further demonstrated in the 1 3 RhoA/ROCK LIMK The role of survival motor neuron protein (SMN) in protein homeostasis Taiwanese model of SMA, where pre-synaptic uptake of [161]). Autophagy involves a double-membrane bound FM1-43 dye by endocytosis was significantly reduced upon structure engulfing target proteins and organelles to form electrical stimulation. Interestingly, this disturbance was an autophagosome. The autophagosome later fuses with lys- restored by PLS3 overexpression. The fact that PLS3 overex- osomes to become an autolysosome, in which the proteins pression could improve the endocytic defect was perhaps not and organelles are degraded (reviewed in [162]). Autophagy surprising, however, as the actin cytoskeleton is required for is a finely balanced mechanism: a decrease in expression this process [157] and yeast cells lacking Sac6p, the PLS3 of autophagy-related genes may lead to the accumulation ortholog, are defective for the internalisation step of endocy- of unwanted proteins whereas over-active autophagy leads tosis [158]. Moreover, another F-actin binding and bundling to increased numbers of autophagosomes, possibly leading protein, coronin 1C (CORO1C), has been shown to interact to cell death [163, 164]. Both of these outcomes have been with PLS3 and its overexpression rescued endocytosis in described in various models of SMN depletion, indicating a SMN-deficient cells and improved the axonal phenotype role for SMN in the regulation of autophagy. in Smn-deficient zebrafish [159]. The importance of SMN It is debatable whether an increase in amount of for endocytic processes has also been confirmed in SMA- autophagosomes is protective or deleterious to the cell. patient-derived cells, which proved resistant to infection by Through measuring expression of LC3-II, a marker of a clathrin endocytosis-dependent virus [153] (see Fig. 1 for autophagosomes, it has been shown that autophagosome a summary of the role of SMN in endocytosis). number is increased in primary motor neurons following Using the same approach that led to the discovery of lentiviral SMN knockdown [165] and in spinal cords of PLS3 as a modifier of SMA, a second modifier, neuronal cal- the Taiwanese mouse model [166] and the SMNΔ7 mouse cium sense protein neurocalcin delta (NCALD) was recently model [167]. Another way of measuring autophagic activ- reported [160]. Contrary to PLS3 which acts as a positive ity is through autophagic flux indicated by the level of p62/ regulator of endocytosis, NCALD is a negative regulator SQSTM1 protein [168–170]. Again, the p62 protein level of endocytosis and axonal growth. Knockdown of NCALD was found to be upregulated in the spinal cord of Burghes restored neurite outgrowth in SMN-deficient cells and severe SMA mice compared to their control littermates improved axonal growth and NMJ function in a zebrafish [171], as well as in an NSC-34 cell line following lentivi- model of SMA. An enhanced neuromuscular function in C. ral SMN knockdown, and in the spinal cord of Taiwanese elegans and murine models of SMA was also observed fol- SMA mice [166], indicating a reduction in autophagic flux. lowing NCALD depletion [160]. In the absence of calcium Inconsistent with data from the Taiwanese mouse model, or at low calcium levels, NCALD, which localises to growth autophagic flux did not appear to increase in the spinal cord cones and pre-synaptic sites at the NMJ, interacts with clath- of SMNΔ7 mice [166, 167]. Inhibition of lysosomal prote- rin, which mediates the endocytosis needed for fast recycling olysis with Bafilomycin A1 (BafA1) resulted in an accumu- at axon terminals. Low SMN levels have been shown to lead lation of LC3-II in cultured motor neurons from the Burghes 2+ to a reduction of voltage-activated Ca influx [98, 160], severe model, suggesting that SMN deficiency can activate and it is possibly through this mechanism that endocytosis autophagy [171]. and vesicle recycling was impaired. It was postulated that, Conversely, autophagy modulators can alter SMN pro- 2+ in normal motor neurons, the high local Ca concentra- tein levels. Treating cultured motor neurons isolated from tion observed following neurotransmitter release led to the wild-type mice with mTORC1 inhibitor rapamycin, which dissociation of NCALD from clathrin, therefore “freeing” is believed to enhance the activity of autophagy [172, 173], clathrin to perform its endocytic function. In SMA, due to showed increased SMN levels, whilst in BafA1-treated 2+ low Ca concentrations, dissociation did not occur and the motor neurons SMN levels were decreased [171]. A recent clathrin was, therefore, not available for coating of the vesi- study has indicated that SMN may be partially degraded cles. Moreover, disturbed calcium homeostasis would also through the autophagy pathway, since a knockdown of p62 be predicted to affect the function of actin-bundling proteins in stem cell-derived motor neurons from SMNΔ7 mice PLS3 and CORO1C, giving further strength to the hypoth- increased SMN protein levels [174]. A role for SMN in esis that low calcium levels secondary to SMN deficiency autophagy is also supported by the finding that overexpres- play an important role in endocytosis impairment [158]. sion of the SMN-binding partner α-COP, normally involved in cytoskeletal growth [110], partially restored autophagic flux in SMN-depleted cells [166], although the mecha- SMN and autophagy nism involved remains unclear. Moreover, injection of the autophagy inhibitor 3-methyladenine (3-MA) into SMNΔ7 Autophagy is a highly conserved catabolic process utilised mice at P3 greatly reduced autophagic activity and protected by cells to break down unwanted macromolecules such motor neurons from degeneration, possibly via inhibition as aggregated proteins or cellular organelles (reviewed in of the apoptotic pathway as shown by reduced expression 1 3 H. Chaytow et al. of apoptotic markers [167]. On the other hand, rapamycin Mitochondrial oxidative phosphorylation is a core part failed to influence the loss of motor neurons, but reduced of bioenergetic pathways. Mitochondrial electron trans- survival significantly in SMNΔ7 mice [167]. These con- port chain function relies on a supply of electrons from flicting findings show that further work is still required to the carriers NADH and FADH through upstream reac- fully elucidate the interaction between SMN and autophagy tions (mainly glycolysis and TCA cycle). Proteomics stud- pathways. ies identified that bioenergetics pathways were affected by SMN deficiency, more specifically GAPDH, an enzyme of the glycolysis pathway, was downregulated in SMA models SMN, mitochondrial homeostasis [181]. Interestingly, gene expression studies of affected and and bioenergetics pathways disease-resistant motor neuron pools in mice revealed that susceptible neurons had lower basal expression not only of SMN deficiency has been linked to changes in oxidative specifically mitochondria-related genes but also of genes stress, mitochondrial dysfunction and impairment of bioen- involved in more generic bioenergetic pathways. Specifi- ergetic pathways. Acsadi et al. [175] showed that knocking cally, the expression of PGK1, a key enzyme of the glyco- down SMN levels by ~ 66% in NSC-34 cells resulted in a lytic pathway, was significantly elevated in motor neurons marked reduction in ATP levels. This was associated with an that are intrinsically resistant to low levels of SMN, with increase in cytochrome c oxidase activity and mitochondrial experimental elevation/activation of PGK1 sufficient to res- membrane potential, resulting in increased free radical pro- cue motor axon defects and loss of neuromuscular function duction. This increase in oxidative stress in SMN-deficient in a zebrafish model of SMA [182]. cells was further confirmed in spinal motor neurons derived Taken together, these studies highlight that SMN defi- from human embryonic stem cells (hESCs). Interestingly, ciency leads to impairment in mitochondria and bioenerget- mitochondrial superoxide production was only increased in ics pathways. However, the precise mechanisms involved the SMN-knockdown hESCs which were made to differenti- in these interactions remain unclear. Studies in various cell ate into spinal motor neurons, but not in the cells differenti- types have shown that SMN does not localise to mitochon- ated into forebrain neurons [176]. dria [175, 183]. Therefore, it has been postulated that the Further analysis of mitochondrial dysfunction was per- effects of SMN on mitochondrial function could be indirect, formed by the same group using two models of SMN-defi- possibly by affecting preferentially the splicing, translation cient cells, SMA Type 1 patient-specific-induced pluripo- or mRNA transport of genes fundamental to mitochondrial tent stem cells (iPSCs) and SMN-knockdown hESCs, both homeostasis [175, 177]. As previously mentioned, cytoskel- differentiated into spinal motor neurons [177]. Impaired etal changes can also lead to decreased mitochondrial trans- mitochondrial axonal transport and a reduction in axonal port, particularly within long axons [150]. Therefore, fur- mitochondrial number and area were noted at early stages ther studies are required to better understand how SMN of cell culture. Partial rescue by the anti-oxidant N-acetyl- affects these energetic pathways, fundamental for cellular cysteine provides evidence to support the hypothesis that homeostasis. oxidative stress plays an important role in neuronal degen- eration in SMN-deficient motor neurons. However, experi- ments on SMA patient iPSCs led to conflicting results as, SMN and ubiquitin pathways in this model, no oxidative stress was detected [178]. These inconsistencies could be secondary to differences in the way Another key mechanism required for protein homeostasis is the stem cells were differentiated and highlight the limita- the protein degradation pathway. There are two major routes tions of studying cell type-specific pathological processes of protein degradation in eukaryotes: the ubiquitin–protea- in cell cultures. More recently, studies in SMNΔ7 and Tai- some system (UPS) and lysosomal proteolysis, or autophagy wanese mouse models confirmed marked mitochondrial (see above). The mammalian ubiquitin pathway is initiated dysfunction in spinal motor neurons, with decreased basal by activation of the E1 ubiquitin-activating enzyme UBA1, and maximal mitochondrial respiration, impaired mitochon- which then transfers ubiquitin onto one of around 40 E2 drial membrane potential, impaired mitochondrial mobility, conjugating enzymes. E2 ligases control whether a substrate increased oxidative stress level and increased fragmentation is mono- or polyubiquitinated [184]. E3 ligases (of which [179]. Interestingly, mitochondrial defects in SMA are not there are several hundred) collect the substrate protein and thought to be limited to motor neurons in vivo, as they have form a complex between it and the ubiquitinated E2 ligase, also been identified in SMA patient muscle associated with where the ubiquitin is transferred onto the protein substrate. a downregulation of mitochondrial biogenesis regulatory Ubiquitination is a dynamic process, and proteins can be factors [180]. stripped of their ubiquitin by deubiquitinating enzymes. 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis SMN has been shown to be ubiquitinated and ultimately patients treated with salbutamol also showed an increase degraded via the ubiquitin–proteasome system, with a pro- in SMN levels in the blood [192]. tein half-life of between 6 and 10 h depending on the cell Through proteomic analysis, SMN has been found to line analysed [185, 186]. Inhibition of the proteasome in interact with several components of the ubiquitin pathway, SMA-patient-derived fibroblasts increased the intracel- including UBA1 and several E3 ligases, as summarised lular abundance of SMN, both in terms of the amount of in Fig. 2 [9, 193]. Mutations in the UBA1 gene cause the SMN protein and the number of nuclear gems [187]. Mon- disease X-linked SMA [194], a rare condition with simi- oubiquitination, as opposed to polyubiquitination, serves lar symptoms to classical SMA but with no mutations in other functions in the cell instead of degradation including the SMN1 gene, suggesting a link between UBA1 and SMN protein trafficking and intracellular localisation (reviewed which, when lost, leads to SMA-like phenotypes. Mutations in [188]) and SMN is known to be monoubiquitinated in the Drosophila homologue of UBA1 cause motor defects, [186]. Indeed, preventing the monoubiquitination of SMN indicating that the motor system is particularly suscepti- changed the localisation of the protein from the cytoplasm ble to the loss of UBA1 despite its ubiquitous expression to the nucleus, and also prevented its co-localisation with [195]. Proteomic analysis of hippocampal synaptosomes Sm proteins [189]. Meanwhile, the SMNΔ7 fragment is from Burghes severe SMA mice showed decreased levels polyubiquitinated and quickly degraded [186]. Pharmaco- of UBA1 compared to controls, with decreased expression logical inhibition of ubiquitination of SMN, such as with also reported in spinal cord and skeletal muscle [196]. The the small molecule ML372, increased SMN protein levels Taiwanese SMA mouse model similarly showed tissue-wide and slowed disease progression of SMNΔ7 mice leading to lower levels of UBA1, along with changes in splicing of the longer survival, increased motor neuron size and less mus- UBA1 transcript, which may account (at least in part) for cle atrophy [190]. When SMA-patient-derived fibroblasts the altered protein expression. Experimental suppression of were treated with salbutamol, the β -adrenergic recep- UBA1 in wild-type zebrafish was sufficient to phenocopy tor agonist, there was also an increase in levels of SMN SMA-like motor axon defects. Likewise, in the zebrafish protein, possibly acting via activation of protein kinase SMA model UBA1 expression was reduced by 70%, whilst A, thereby preventing SMN ubiquitination [191]. SMA increasing UBA1 expression rescued the SMN-knockdown SMN E3 Ligases Protein Substrate E.g. SMN E2 Ligases E3 Ligases ATP AMP Protein Substrate E3 Ligases E.g. SMN Protein E2 Ligases Intracellular E2 Ligas UBA1 UBA1 B1 Substrate Ub processes Ub Ub E.g. S N SMN Ub Ub Protein Substrate Proteasome E.g. SMN or or degradation Protein SMN 7 Substrate Ub E.g. SMN Ub Deubiquitinating Ub Enzymes Ub SMN Ub Ub Ub Ub Fig. 2 Diagrammatic representation of the ubiquitin pathway and the 1, Itch and TRAF6. Ubiquitin is then transferred to the protein sub- components, where SMN interacts. SMN is both ubiquitinated via the strate and the complex dissociates. Monoubiquitinated substrates con- UPS pathway and an interacting protein influencing several steps of tinue on to other intracellular processes, whereas polyubiquitinated the process. SMN directly interacts with the UBA1 enzyme, which substrates are targeted for proteasome degradation. SMN has also transfers ubiquitin to the E2 ligases. Ubiquitinated E2 ligases then been shown to interact with deubiquitinating enzymes, which remove form a complex with E3 ligases bound to protein substrates. SMN has ubiquitin from protein substrates been shown to interact with several E3 ligases, including Mindbomb 1 3 H. Chaytow et al. phenotype [197]. Finally, treating Taiwanese mice with homeostasis in which SMN has been shown to interact: its an AAV9-UBA1 expression vector improved the survival, well-known role as part of the ribonucleoprotein complex, weight gain and motor performance of the mice as well as but also other stages of RNA processing such as transport rescuing motor neuron cell number in the spinal cord and and local translation, important neuronal functions such as neuromuscular junction pathology [197]. cytoskeletal dynamics and endocytosis, protein turnover As well as UBA1, SMN is known to interact with sev- processes of autophagy and ubiquitin–proteasome pathway eral other ubiquitin-associated enzymes (Fig.  2). Several and regulation of mitochondrial activity. Through tradition E3 ubiquitin ligases have been shown to interact with SMN and necessity, the majority of current research into the func- and so may be involved in its degradation through recruit- tion of SMN comes from SMA models of SMN deficiency. ment into the UPS. For example, Mindbomb 1 directly However, as this review has highlighted, SMN function is interacts with SMN [190]. Overexpression of Mindbomb involved in so many aspects of normal intracellular activity 1 was shown to increase the amount of ubiquitinated SMN that future SMN research should move beyond its associa- protein in cell culture, while a knockdown of Mindbomb 1 in tion with disease to better understand its role in maintain- the C. elegans model of SMA improves the SMN-deficient ing the homeostatic environment of the cell. Two major phenotype of defects in pharyngeal pumping [198]. Other questions need answering in terms of the function of SMN. E3 ligases known to interact with SMN include Itch [189], First, to what extent is SMN involved in the regulation of TRAF6 [199] and the Drosophila E3 ligase SCFslmb [9]. processes discussed in this review. While some areas have Monoubiquitination following interaction with Itch was been researched extensively, such as ribonucleoprotein pro- shown to regulate SMN’s intracellular localisation [189]. duction, other areas of SMN involvement are a relatively TRAF6 activity is apparently inhibited by SMN, and so new discovery, such as the association of SMN with mito- SMN binding may be involved in the activation of NF-κB chondrial function and ubiquitin degradation, and so further signalling further downstream [199]. exploration is needed. Secondly, the particular vulnerabil- Ubiquitin carboxy-terminal hydrolase L1 (UCHL1) is a ity of motor neurons in SMA patients cannot be ignored. deubiquitinating enzyme specifically expressed in neuronal Although the idea that SMA is in fact a systemic disease, tissue, and its downregulation has been associated with with defects seen across tissue types, is gaining acceptance Parkinson’s and Alzheimer’s diseases [200–202]. Follow- in the research community, a better understanding of the ing knockdown of UCHL1 in cell culture, there was a con- multiplicity of SMN functions could serve to highlight areas cordant increase in SMN expression [203]. Conversely, in of particular susceptibility in motor neurons which lead to Taiwanese mice, there was an increase in UCHL1 expres- their cell death in SMA. As SMN is at the cornerstone of so sion. However, inhibition of UCHL1 expression in Taiwan- many molecular pathways, fundamental research into these ese mice failed to increase SMN levels and did not have an cellular homeostasis processes is crucial to the better under- effect on survival or phenotype of the SMA model, with standing of cellular biology. evidence suggesting that an increase in UCHL1 levels in the Acknowledgements We would like to thank Dr. Ewout Groen for help- absence of SMN may be a compensatory response to restore ful comments on the manuscript, and all the members of the lab for levels of ubiquitination [204]. Usp9x, another deubiquitinat- ongoing discussions and advice. Research in the Gillingwater labora- ing enzyme known to interact with SMN, also influences its tory relevant to this review is funded by the UK SMA Research Con- ubiquitination levels, where a loss of Usp9x impairs SMN sortium (SMA Trust), MND Scotland and SMA Europe. nuclear gem formation while overexpression leads to an increase in ubiquitinated SMN [186]. It, therefore, appears Compliance with ethical standards that SMN is regulated at several levels of the UPS, which Conflict of interest THG is Chair of the Scientific and Clinical Advi- may have an effect on cell-wide ubiquitination as well as sory Board of the SMA Trust. regulation of the SMN protein itself. Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creat iveco Concluding remarks and future perspectivesmmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the SMN, originally discovered due to its association with the Creative Commons license, and indicate if changes were made. neurodegenerative disorder spinal muscular atrophy, is in fact a ubiquitous protein with numerous roles within the cell. Although its first-identified and most-described function is in the biogenesis of ribonucleoproteins, it is now evident that SMN plays a more general housekeeping role. With this in mind, here we have discussed various areas of intracellular 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis 19. Meister G, Fischer U (2002) Assisted RNP assembly: SMN and References PRMT5 complexes cooperate in the formation of spliceosomal UsnRNPs. EMBO J 21:5853–5863 1. Lefebvre S et al (1995) Identification and characterization of a 20. Paushkin S, Gubitz AK, Massenet S, Dreyfuss G (2002) The spinal muscular atrophy-determining gene. Cell 80:155–165 SMN complex, an assemblyosome of ribonucleoproteins. Curr 2. Schrank B, Götz R, Gunnersen JM, Ure JM, Toyka KV, Smith Opin Cell Biol 14:305–312 AG, Sendtner M (1997) Inactivation of the survival motor neu- 21. Hebert MD, Shpargel KB, Ospina JK, Tucker KE, Matera AG ron gene, a candidate gene for human spinal muscular atrophy, (2002) Coilin methylation regulates nuclear body formation. Dev leads to massive cell death in early mouse embryos. Proc Natl Cell 3:329–337 Acad Sci USA 94:9920–9925 22. Sprangers R, Groves MR, Sinning I, Sattler M (2003) High-res- 3. Groen EJN, Talbot K, Gillingwater TH (2018) Advances in olution X-ray and NMR structures of the SMN Tudor domain: therapy for spinal muscular atrophy: promises and challenges. conformational variation in the binding site for symmetrically Nat Rev Neurol 14:214–224 dimethylated arginine residues. J Mol Biol 327:507–520 4. Rochette CF, Gilbert N, Simard LR (2001) SMN gene dupli- 23. Côté J, Richard S (2005) Tudor domains bind symmetrical cation and the emergence of the SMN2 gene occurred in dis- dimethylated arginines. J Biol Chem 280:28476–28483 tinct hominids: SMN2 is unique to Homo sapiens. Hum Genet 24. Meister G, Eggert C, Fischer U (2002) SMN-mediated assem- 108:255–266 bly of RNPs: a complex story. Trends Cell Biol 12:472–478 5. Lorson CL, Hahnen E, Androphy EJ, Wirth B (1999) A single 25. Bühler D, Raker V, Lührmann R, Fischer U (1999) Essential nucleotide in the SMN gene regulates splicing and is respon- role for the tudor domain of SMN in spliceosomal U snRNP sible for spinal muscular atrophy. Proc Natl Acad Sci USA assembly: implications for spinal muscular atrophy. Hum Mol 96:6307–6311 Genet 8:2351–2357 6. Feldkötter M, Schwarzer V, Wirth R, Wienker TF, Wirth B 26. Kotani T et al (2007) A novel mutation at the N-terminal of (2002) Quantitative analyses of SMN1 and SMN2 based on real- SMN Tudor domain inhibits its interaction with target proteins. time lightCycler PCR: fast and highly reliable carrier testing and J Neurol 254:624–630 prediction of severity of spinal muscular atrophy. Am J Hum 27. Cuscó I, Barceló MJ, del Río E, Baiget M, Tizzano EF (2004) Genet 70:358–368 Detection of novel mutations in the SMN Tudor domain in type 7. Talbot K, Tizzano EF (2017) The clinical landscape for SMA in I SMA patients. Neurology 63:146–149 a new therapeutic era. Gene Ther 24:529–533 28. Hebert MD, Szymczyk PW, Shpargel KB, Matera AG (2001) 8. Vitte J et al (2007) Refined characterization of the expression and Coilin forms the bridge between Cajal bodies and SMN, the stability of the SMN gene products. Am J Pathol 171:1269–1280 spinal muscular atrophy protein. Genes Dev 15:2720–2729 9. Gray KM et al (2018) Self-oligomerization regulates stability 29. Liu Q, Dreyfuss G (1996) A novel nuclear structure containing of survival motor neuron protein isoforms by sequestering an the survival of motor neurons protein. EMBO J 15:3555–3565 SCF(Slmb) degron. Mol Biol Cell 29:96–110 30. Navascues J, Berciano MT, Tucker KE, Lafarga M, Matera 10. Seo J, Singh NN, Ottesen EW, Lee BM, Singh RN (2016) A AG (2004) Targeting SMN to Cajal bodies and nuclear gems novel human-specific splice isoform alters the critical C-terminus during neuritogenesis. Chromosoma 112:398–409 of Survival Motor Neuron protein. Sci Rep 6:30778 31. Machyna M, Heyn P, Neugebauer KM (2013) Cajal bod- 11. Gennarelli M, Lucarelli M, Capon F, Pizzuti A, Merlini L, Ange- ies: where form meets function. Wiley Interdiscip Rev RNA lini C, Novelli G, Dallapiccola B (1995) Survival motor neuron 4:17–34 gene transcript analysis in muscles from spinal muscular atrophy 32. Tapia O, Bengoechea R, Palanca A, Arteaga R, Val-Bernal patients. Biochem Biophys Res Commun 213:342–348 JF, Tizzano EF, Berciano MT, Lafarga M (2012) Reorganiza- 12. Setola V, Terao M, Locatelli D, Bassanini S, Garattini E, Batt- tion of Cajal bodies and nucleolar targeting of coilin in motor aglia G (2007) Axonal-SMN (a-SMN), a protein isoform of the neurons of type I spinal muscular atrophy. Histochem Cell Biol survival motor neuron gene, is specifically involved in axonogen- 137:657–667 esis. Proc Natl Acad Sci USA 104:1959–1964 33. Tapia O, Lafarga V, Bengoechea R, Palanca A, Lafarga M, Ber- 13. Young PJ, Jensen KT, Burger LR, Pintel DJ, Lorson CL (2002) ciano MT (2014) The SMN Tudor SIM-like domain is key to Minute virus of mice NS1 interacts with the SMN protein, and SmD1 and coilin interactions and to Cajal body biogenesis. J they colocalize in novel nuclear bodies induced by parvovirus Cell Sci 127:939–946 infection. J Virol 76:3892–3904 34. Giesemann T, Rathke-Hartlieb S, Rothkegel M, Bartsch JW, 14. Liu Q, Fischer U, Wang F, Dreyfuss G (1997) The spinal muscu- Buchmeier S, Jockusch BM, Jockusch H (1999) A role for poly- lar atrophy disease gene product, SMN, and its associated protein proline motifs in the spinal muscular atrophy protein SMN. Pro- SIP1 are in a complex with spliceosomal snRNP proteins. Cell filins bind to and colocalize with smn in nuclear gems. J Biol 90:1013–1021 Chem 274:37908–37914 15. Lorson CL, Androphy EJ (1998) The domain encoded by exon 2 35. Nölle A et al (2011) The spinal muscular atrophy disease protein of the survival motor neuron protein mediates nucleic acid bind- SMN is linked to the Rho-kinase pathway via profilin. Hum Mol ing. Hum Mol Genet 7:1269–1275 Genet 20:4865–4878 16. Bertrandy S, Burlet P, Clermont O, Huber C, Fondrat C, Thierry- 36. Martin R, Gupta K, Ninan NS, Perry K, Van Duyne GD (2012) Mieg D, Munnich A, Lefebvre S (1999) The RNA-binding The survival motor neuron protein forms soluble glycine zipper properties of SMN: deletion analysis of the zebrafish ortho- oligomers. Structure 20:1929–1939 logue defines domains conserved in evolution. Hum Mol Genet 37. Burghes AH, Beattie CE (2009) Spinal muscular atrophy: why do 8:775–782 low levels of survival motor neuron protein make motor neurons 17. Ponting CP (1997) Tudor domains in proteins that interact with sick? Nat Rev Neurosci 10:597–609 RNA. Trends Biochem Sci 22:51–52 38. Wang J, Dreyfuss G (2001) A cell system with targeted dis- 18. Selenko P, Sprangers R, Stier G, Bühler D, Fischer U, Sattler M ruption of the SMN gene: functional conservation of the SMN (2001) SMN tudor domain structure and its interaction with the protein and dependence of Gemin2 on SMN. J Biol Chem Sm proteins. Nat Struct Biol 8:27–31 276:9599–9605 1 3 H. Chaytow et al. 39. Lorson CL et al (1998) SMN oligomerization defect correlates 60. Carissimi C, Saieva L, Baccon J, Chiarella P, Maiolica A, Sawyer with spinal muscular atrophy severity. Nat Genet 19:63–66 A, Rappsilber J, Pellizzoni L (2006) Gemin8 is a novel compo- 40. Lafarga V, Tapia O, Sharma S, Bengoechea R, Stoecklin G, nent of the survival motor neuron complex and functions in small Lafarga M, Berciano MT (2018) CBP-mediated SMN acetylation nuclear ribonucleoprotein assembly. J Biol Chem 281:8126–8134 modulates Cajal body biogenesis and the cytoplasmic targeting 61. Charroux B, Pellizzoni L, Perkinson RA, Shevchenko A, Mann of SMN. Cell Mol Life Sci 75:527–546 M, Dreyfuss G (1999) Gemin3: a novel DEAD box protein that 41. Husedzinovic A, Oppermann F, Draeger-Meurer S, Chari A, interacts with SMN, the spinal muscular atrophy gene product, Fischer U, Daub H, Gruss OJ (2014) Phosphoregulation of the and is a component of gems. J Cell Biol 147:1181–1194 human SMN complex. Eur J Cell Biol 93:106–117 62. Gubitz AK, Feng W, Dreyfuss G (2004) The SMN complex. Exp 42. Hebert MD, Poole AR (2017) Towards an understanding of regu- Cell Res 296:51–56 lating Cajal body activity by protein modification. RNA Biol 63. Gubitz AK, Mourelatos Z, Abel L, Rappsilber J, Mann M, 14:761–778 Dreyfuss G (2002) Gemin5, a novel WD repeat protein compo- 43. Hsieh-Li HM, Chang JG, Jong YJ, Wu MH, Wang NM, Tsai CH, nent of the SMN complex that binds Sm proteins. J Biol Chem Li H (2000) A mouse model for spinal muscular atrophy. Nat 277:5631–5636 Genet 24:66–70 64. Li DK, Tisdale S, Lotti F, Pellizzoni L (2014) SMN control of 44. Monani UR et al (2000) The human centromeric survival motor RNP assembly: from post-transcriptional gene regulation to (−/−) neuron gene (SMN2) rescues embryonic lethality in Smn motor neuron disease. Semin Cell Dev Biol 32:22–29 mice and results in a mouse with spinal muscular atrophy. Hum 65. Pellizzoni L (2007) Chaperoning ribonucleoprotein biogenesis Mol Genet 9:333–339 in health and disease. EMBO Rep 8:340–345 45. Le TT et al (2005) SMNDelta7, the major product of the cen- 66. Pellizzoni L, Baccon J, Rappsilber J, Mann M, Dreyfuss G tromeric survival motor neuron (SMN2) gene, extends survival (2002) Purification of native survival of motor neurons com- in mice with spinal muscular atrophy and associates with full- plexes and identification of Gemin6 as a novel component. J length SMN. Hum Mol Genet 14:845–857 Biol Chem 277:7540–7545 46. Bebee TW, Dominguez CE, Chandler DS (2012) Mouse models 67. Battle DJ, Lau CK, Wan L, Deng H, Lotti F, Dreyfuss G (2006) of SMA: tools for disease characterization and therapeutic devel- The Gemin5 protein of the SMN complex identifies snRNAs. opment. Hum Genet 131:1277–1293 Mol Cell 23:273–279 47. McWhorter ML, Monani UR, Burghes AH, Beattie CE (2003) 68. Pellizzoni L, Yong J, Dreyfuss G (2002) Essential role for the Knockdown of the survival motor neuron (Smn) protein in SMN complex in the specificity of snRNP assembly. Science zebrafish causes defects in motor axon outgrowth and pathfind- 298:1775–1779 ing. J Cell Biol 162:919–931 69. Yong J, Golembe TJ, Battle DJ, Pellizzoni L, Dreyfuss G 48. Ciura S, Lattante S, Le Ber I, Latouche M, Tostivint H, Brice (2004) snRNAs contain specific SMN-binding domains that A, Kabashi E (2013) Loss of function of C9orf72 causes motor are essential for snRNP assembly. Mol Cell Biol 24:2747–2756 deficits in a zebrafish model of amyotrophic lateral sclerosis. Ann 70. Yong J, Kasim M, Bachorik JL, Wan L, Dreyfuss G (2010) Neurol 74:180–187 Gemin5 delivers snRNA precursors to the SMN complex for 49. McLean DL, Fetcho JR (2011) Movement, technology and dis- snRNP biogenesis. Mol Cell 38:551–562 covery in the zebrafish. Curr Opin Neurobiol 21:110–115 71. Yong J, Pellizzoni L, Dreyfuss G (2002) Sequence-specific 50. Babin PJ, Goizet C, Raldúa D (2014) Zebrafish models of human interaction of U1 snRNA with the SMN complex. EMBO J motor neuron diseases: advantages and limitations. Prog Neuro- 21:1188–1196 biol 118:36–58 72. Fischer U, Luhrmann R (1990) An essential signaling role for 51. Edens BM, Ajroud-Driss S, Ma L, Ma YC (2015) Molecular the m3G cap in the transport of U1 snRNP to the nucleus. Sci- mechanisms and animal models of spinal muscular atrophy. Bio- ence 249:786–790 chim Biophys Acta 1852:685–692 73. Mouaikel J, Narayanan U, Verheggen C, Matera AG, Bertrand 52. Frattini E et al (2015) Pluripotent stem cell-based models of spi- E, Tazi J, Bordonne R (2003) Interaction between the small- nal muscular atrophy. Mol Cell Neurosci 64:44–50 nuclear-RNA cap hypermethylase and the spinal muscular atro- 53. Sleigh JN, Gillingwater TH, Talbot K (2011) The contribution phy protein, survival of motor neuron. EMBO Rep 4:616–622 of mouse models to understanding the pathogenesis of spinal 74. Narayanan U, Ospina JK, Frey MR, Hebert MD, Matera AG muscular atrophy. Dis Model Mech 4:457–467 (2002) SMN, the spinal muscular atrophy protein, forms a pre- 54. Van Alstyne M, Pellizzoni L (2016) Advances in modeling and import snRNP complex with snurportin1 and importin beta. treating spinal muscular atrophy. Curr Opin Neurol 29:549–556 Hum Mol Genet 11:1785–1795 55. Rodriguez-Muela N et al (2017) Single-cell analysis of SMN 75. Fischer U, Sumpter V, Sekine M, Satoh T, Luhrmann R (1993) reveals its broader role in neuromuscular disease. Cell Rep Nucleo-cytoplasmic transport of U snRNPs: definition of a 18:1484–1498 nuclear location signal in the Sm core domain that binds a 56. Matera AG, Wang Z (2014) A day in the life of the spliceosome. transport receptor independently of the m3G cap. EMBO J Nat Rev Mol Cell Biol 15:108–121 12:573–583 57. Raker VA, Hartmuth K, Kastner B, Luhrmann R (1999) Spliceo- 76. Huber J, Cronshagen U, Kadokura M, Marshallsay C, Wada somal U snRNP core assembly: Sm proteins assemble onto an T, Sekine M, Luhrmann R (1998) Snurportin1, an m3G-cap- Sm site RNA nonanucleotide in a specific and thermodynami- specific nuclear import receptor with a novel domain structure. cally stable manner. Mol Cell Biol 19:6554–6565 EMBO J 17:4114–4126 58. Baccon J, Pellizzoni L, Rappsilber J, Mann M, Dreyfuss G (2002) 77. Palacios I, Hetzer M, Adam SA, Mattaj IW (1997) Nuclear Identification and characterization of Gemin7, a novel compo- import of U snRNPs requires importin beta. EMBO J nent of the survival of motor neuron complex. J Biol Chem 16:6783–6792 277:31957–31962 78. Mahmoudi S, Henriksson S, Weibrecht I, Smith S, Soderberg 59. Carissimi C, Baccon J, Straccia M, Chiarella P, Maiolica A, O, Stromblad S, Wiman KG, Farnebo M (2010) WRAP53 is Sawyer A, Rappsilber J, Pellizzoni L (2005) Unrip is a compo- essential for Cajal body formation and for targeting the survival nent of SMN complexes active in snRNP assembly. FEBS Lett of motor neuron complex to Cajal bodies. PLoS Biol 8:e1000521 579:2348–2354 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis 79. Narayanan U, Achsel T, Luhrmann R, Matera AG (2004) Cou- 98. See K et al (2014) SMN deficiency alters Nrxn2 expression and pled in vitro import of U snRNPs and SMN, the spinal muscular splicing in zebrafish and mouse models of spinal muscular atro- atrophy protein. Mol Cell 16:223–234 phy. Hum Mol Genet 23:1754–1770 80. Henriksson S, Farnebo M (2015) On the road with WRAP53beta: 99. Sleigh JN, Barreiro-Iglesias A, Oliver PL, Biba A, Becker T, guardian of Cajal bodies and genome integrity. Front Genet 6:91 Davies KE, Becker CG, Talbot K (2014) Chondrolectin affects 81. Gabanella F, Butchbach ME, Saieva L, Carissimi C, Burghes AH, cell survival and neuronal outgrowth in in vitro and in vivo mod- Pellizzoni L (2007) Ribonucleoprotein assembly defects correlate els of spinal muscular atrophy. Hum Mol Genet 23:855–869 with spinal muscular atrophy severity and preferentially affect a 100. Zhang Z et al (2013) Dysregulation of synaptogenesis genes ante- subset of spliceosomal snRNPs. PLoS One 2:e921 cedes motor neuron pathology in spinal muscular atrophy. Proc 82. Wan L, Battle DJ, Yong J, Gubitz AK, Kolb SJ, Wang J, Dreyfuss Natl Acad Sci USA 110:19348–19353 G (2005) The survival of motor neurons protein determines the 101. Baumer D et al (2009) Alternative splicing events are a late fea- capacity for snRNP assembly: biochemical deficiency in spinal ture of pathology in a mouse model of spinal muscular atrophy. muscular atrophy. Mol Cell Biol 25:5543–5551 PLoS Genet 5:e1000773 83. Zhang Z, Lotti F, Dittmar K, Younis I, Wan L, Kasim M, Drey- 102. Pagliardini S, Giavazzi A, Setola V, Lizier C, Di Luca M, DeBi- fuss G (2008) SMN deficiency causes tissue-specific perturba - asi S, Battaglia G (2000) Subcellular localization and axonal tions in the repertoire of snRNAs and widespread defects in transport of the survival motor neuron (SMN) protein in the splicing. Cell 133:585–600 developing rat spinal cord. Hum Mol Genet 9:47–56 84. So BR, Wan L, Zhang Z, Li P, Babiash E, Duan J, Younis I, 103. Giavazzi A, Setola V, Simonati A, Battaglia G (2006) Neuronal- Dreyfuss G (2016) A U1 snRNP-specific assembly pathway specific roles of the survival motor neuron protein: evidence reveals the SMN complex as a versatile hub for RNP exchange. from survival motor neuron expression patterns in the develop- Nat Struct Mol Biol 23:225–230 ing human central nervous system. J Neuropathol Exp Neurol 85. Marzluff WF, Wagner EJ, Duronio RJ (2008) Metabolism and 65:267–277 regulation of canonical histone mRNAs: life without a poly(A) 104. Zhang HL, Pan F, Hong D, Shenoy SM, Singer RH, Bassell GJ tail. Nat Rev Genet 9:843–854 (2003) Active transport of the survival motor neuron protein 86. Pillai RS, Grimmler M, Meister G, Will CL, Luhrmann R, Fis- and the role of exon-7 in cytoplasmic localization. J Neurosci cher U, Schumperli D (2003) Unique Sm core structure of U7 23:6627–6637 snRNPs: assembly by a specialized SMN complex and the role 105. Zhang H, Xing L, Rossoll W, Wichterle H, Singer RH, Bassell of a new component, Lsm11, in histone RNA processing. Genes GJ (2006) Multiprotein complexes of the survival of motor neu- Dev 17:2321–2333 ron protein SMN with Gemins traffic to neuronal processes and 87. Pillai RS, Will CL, Luhrmann R, Schumperli D, Muller B growth cones of motor neurons. J Neurosci 26:8622–8632 (2001) Purified U7 snRNPs lack the Sm proteins D1 and D2 106. Thompson LW, Morrison KD, Shirran SL, Groen EJN, Gill- but contain Lsm10, a new 14 kDa Sm D1-like protein. EMBO J ingwater TH, Botting CH, Sleeman JE (2018) Neurochondrin 20:5470–5479 interacts with the SMN protein suggesting a novel mechanism 88. Schumperli D, Pillai RS (2004) The special Sm core structure for spinal muscular atrophy pathology. J Cell Sci. https ://doi. of the U7 snRNP: far-reaching significance of a small nuclear org/10.1242/jcs.21148 2 ribonucleoprotein. Cell Mol Life Sci 61:2560–2570 107. Peter CJ et al (2011) The COPI vesicle complex binds and moves 89. Reichow SL, Hamma T, Ferre-D’Amare AR, Varani G (2007) with survival motor neuron within axons. Hum Mol Genet The structure and function of small nucleolar ribonucleoproteins. 20:1701–1711 Nucleic Acids Res 35:1452–1464 108. Todd AG, Lin H, Ebert AD, Liu Y, Androphy EJ (2013) COPI 90. Pellizzoni L, Baccon J, Charroux B, Dreyfuss G (2001) The sur- transport complexes bind to specific RNAs in neuronal cells. vival of motor neurons (SMN) protein interacts with the snoRNP Hum Mol Genet 22:729–736 proteins fibrillarin and GAR1. Curr Biol 11:1079–1088 109. Ting CH, Wen HL, Liu HC, Hsieh-Li HM, Li H, Lin-Chao S 91. Renvoise B, Colasse S, Burlet P, Viollet L, Meier UT, Lefebvre (2012) The spinal muscular atrophy disease protein SMN is S (2009) The loss of the snoRNP chaperone Nopp140 from Cajal linked to the Golgi network. PLoS One 7:e51826 bodies of patient b fi roblasts correlates with the severity of spinal 110. Li H, Custer SK, Gilson T, Hao LT, Beattie CE, Androphy EJ muscular atrophy. Hum Mol Genet 18:1181–1189 (2015) α-COP binding to the survival motor neuron protein SMN 92. Zhang Q, Kim NK, Feigon J (2011) Architecture of human tel- is required for neuronal process outgrowth. Hum Mol Genet omerase RNA. Proc Natl Acad Sci USA 108:20325–20332 24:7295–7307 93. Bachand F, Boisvert FM, Cote J, Richard S, Autexier C 111. Custer SK, Todd AG, Singh NN, Androphy EJ (2013) Dily- (2002) The product of the survival of motor neuron (SMN) sine motifs in exon 2b of SMN protein mediate binding to the gene is a human telomerase-associated protein. Mol Biol Cell COPI vesicle protein α-COP and neurite outgrowth in a cell 13:3192–3202 culture model of spinal muscular atrophy. Hum Mol Genet 94. Poole AR, Hebert MD (2016) SMN and coilin negatively regulate 22:4043–4052 dyskerin association with telomerase RNA. Biol Open 5:726–735 112. Rossoll W, Jablonka S, Andreassi C, Kroning AK, Karle K, 95. Powis RA, Gillingwater TH (2016) Selective loss of alpha motor Monani UR, Sendtner M (2003) Smn, the spinal muscular neurons with sparing of gamma motor neurons and spinal cord atrophy-determining gene product, modulates axon growth and cholinergic neurons in a mouse model of spinal muscular atro- localization of beta-actin mRNA in growth cones of motoneu- phy. J Anat 228:443–451 rons. J Cell Biol 163:801–812 96. Winkler C, Eggert C, Gradl D, Meister G, Giegerich M, Wedlich 113. Fallini C, Zhang H, Su Y, Silani V, Singer RH, Rossoll W, Bassell D, Laggerbauer B, Fischer U (2005) Reduced U snRNP assembly GJ (2011) The survival of motor neuron (SMN) protein interacts causes motor axon degeneration in an animal model for spinal with the mRNA-binding protein HuD and regulates localization muscular atrophy. Genes Dev 19:2320–2330 of poly(A) mRNA in primary motor neuron axons. J Neurosci 97. Lotti F et  al (2012) An SMN-dependent U12 splicing event 31:3914–3925 essential for motor circuit function. Cell 151:440–454 114. Tadesse H, Deschenes-Furry J, Boisvenue S, Cote J (2008) KH- type splicing regulatory protein interacts with survival motor 1 3 H. Chaytow et al. neuron protein and is misregulated in spinal muscular atrophy. 134. Leterrier C, Dubey P, Roy S (2017) The nano-architecture of the Hum Mol Genet 17:506–524 axonal cytoskeleton. Nat Rev Neurosci 18:713–726 115. Piazzon N, Rage F, Schlotter F, Moine H, Branlant C, Mass- 135. Kevenaar JT, Hoogenraad CC (2015) The axonal cytoskeleton: enet S (2008) In vitro and in cellulo evidences for association of from organization to function. Front Mol Neurosci 8:44 the survival of motor neuron complex with the fragile X mental 136. Béchade C, Rostaing P, Cisterni C, Kalisch R, La Bella V, Pett- retardation protein. J Biol Chem 283:5598–5610 mann B, Triller A (1999) Subcellular distribution of survival 116. Hubers L, Valderrama-Carvajal H, Laframboise J, Timbers J, motor neuron (SMN) protein: possible involvement in nucleo- Sanchez G, Cote J (2011) HuD interacts with survival motor neu- cytoplasmic and dendritic transport. Eur J Neurosci 11:293–304 ron protein and can rescue spinal muscular atrophy-like neuronal 137. Francis JW, Sandrock AW, Bhide PG, Vonsattel JP, Brown RH defects. Hum Mol Genet 20:553–579 (1998) Heterogeneity of subcellular localization and electro- 117. Groen EJ et al (2013) ALS-associated mutations in FUS disrupt phoretic mobility of survival motor neuron (SMN) protein in the axonal distribution and function of SMN. Hum Mol Genet mammalian neural cells and tissues. Proc Natl Acad Sci USA 22:3690–3704 95:6492–6497 118. Fallini C, Rouanet JP, Donlin-Asp PG, Guo P, Zhang H, Singer 138. Fan L, Simard LR (2002) Survival motor neuron (SMN) protein: RH, Rossoll W, Bassell GJ (2014) Dynamics of survival of motor role in neurite outgrowth and neuromuscular maturation dur- neuron (SMN) protein interaction with the mRNA-binding pro- ing neuronal differentiation and development. Hum Mol Genet tein IMP1 facilitates its trafficking into motor neuron axons. Dev 11:1605–1614 Neurobiol 74:319–332 139. van Bergeijk J, Rydel-Könecke K, Grothe C, Claus P (2007) 119. Yamazaki T et al (2012) FUS-SMN protein interactions link the The spinal muscular atrophy gene product regulates neu- motor neuron diseases ALS and SMA. Cell Rep 2:799–806 rite outgrowth: importance of the C terminus. FASEB J 120. Tsuiji H et  al (2013) Spliceosome integrity is defective in 21:1492–1502 the motor neuron diseases ALS and SMA. EMBO Mol Med 140. Locatelli D et al (2012) Human axonal survival of motor neuron 5:221–234 (a-SMN) protein stimulates axon growth, cell motility, C-C motif 121. Donlin-Asp PG, Fallini C, Campos J, Chou CC, Merritt ME, ligand 2 (CCL2), and insulin-like growth factor-1 (IGF1) produc- Phan HC, Bassell GJ, Rossoll W (2017) The survival of motor tion. J Biol Chem 287:25782–25794 neuron protein acts as a molecular chaperone for mRNP assem- 141. Sharma A, Lambrechts A, Hao LT, Le TT, Sewry CA, Ampe C, bly. Cell Rep 18:1660–1673 Burghes AH, Morris GE (2005) A role for complexes of survival 122. le Hao T, Duy PQ, An M, Talbot J, Iyer CC, Wolman M, Beattie of motor neurons (SMN) protein with gemins and profilin in CE (2017) HuD and the survival motor neuron protein interact neurite-like cytoplasmic extensions of cultured nerve cells. Exp in motoneurons and are essential for motoneuron development, Cell Res 309:185–197 function, and mRNA regulation. J Neurosci 37:11559–11571 142. Schmandke A, Strittmatter SM (2007) ROCK and Rho: biochem- 123. Rage F, Boulisfane N, Rihan K, Neel H, Gostan T, Bertrand istry and neuronal functions of Rho-associated protein kinases. E, Bordonne R, Soret J (2013) Genome-wide identification Neuroscientist 13:454–469 of mRNAs associated with the protein SMN whose depletion 143. Bowerman M, Shafey D, Kothary R (2007) Smn depletion alters decreases their axonal localization. RNA 19:1755–1766 profilin II expression and leads to upregulation of the RhoA/ 124. Akten B et al (2011) Interaction of survival of motor neuron ROCK pathway and defects in neuronal integrity. J Mol Neurosci (SMN) and HuD proteins with mRNA cpg15 rescues motor neu- 32:120–131 ron axonal deficits. Proc Natl Acad Sci USA 108:10337–10342 144. Bowerman M, Beauvais A, Anderson CL, Kothary R (2010) 125. Glock C, Heumüller M, Schuman EM (2017) mRNA trans- Rho-kinase inactivation prolongs survival of an intermediate port and local translation in neurons. Curr Opin Neurobiol SMA mouse model. Hum Mol Genet 19:1468–1478 45:169–177 145. Rademacher S et al (2017) Metalloprotease-mediated cleavage 126. Donlin-Asp PG, Rossoll W, Bassell GJ (2017) Spatially and of PlexinD1 and its sequestration to actin rods in the motoneu- temporally regulating translation via mRNA-binding proteins in ron disease spinal muscular atrophy (SMA). Hum Mol Genet cellular and neuronal function. FEBS Lett 591:1508–1525 26:3946–3959 127. le Hao T, Wolman M, Granato M, Beattie CE (2012) Survival 146. Oprea GE et al (2008) Plastin 3 is a protective modifier of auto- motor neuron affects plastin 3 protein levels leading to motor somal recessive spinal muscular atrophy. Science 320:524–527 defects. J Neurosci 32:5074–5084 147. Kaifer KA, Villalón E, Osman EY, Glascock JJ, Arnold LL, Cor- 128. Fallini C, Donlin-Asp PG, Rouanet JP, Bassell GJ, Rossoll nelison DD, Lorson CL (2017) Plastin-3 extends survival and W (2016) Deficiency of the survival of motor neuron protein reduces severity in mouse models of spinal muscular atrophy. impairs mRNA localization and local translation in the growth JCI Insight 2:e89970 cone of motor neurons. J Neurosci 36:3811–3820 148. Prokop A (2013) The intricate relationship between microtu- 129. Sanchez G et al (2013) A novel function for the survival moto- bules and their associated motor proteins during axon growth neuron protein as a translational regulator. Hum Mol Genet and maintenance. Neural Dev 8:17 22:668–684 149. Belmont LD, Mitchison TJ (1996) Identification of a protein that 130. Weeraratne SD et  al (2012) Pleiotropic effects of miR- interacts with tubulin dimers and increases the catastrophe rate 183~96~182 converge to regulate cell survival, proliferation and of microtubules. Cell 84:623–631 migration in medulloblastoma. Acta Neuropathol 123:539–552 150. Wen HL, Lin YT, Ting CH, Lin-Chao S, Li H, Hsieh-Li HM 131. Kye MJ et al (2014) SMN regulates axonal local translation via (2010) Stathmin, a microtubule-destabilizing protein, is dysregu- miR-183/mTOR pathway. Hum Mol Genet 23:6318–6331 lated in spinal muscular atrophy. Hum Mol Genet 19:1766–1778 132. Ning K et al (2010) PTEN depletion rescues axonal growth defect 151. Deinhardt K, Schiavo G (2005) Endocytosis and retrograde and improves survival in SMN-deficient motor neurons. Hum axonal traffic in motor neurons. Biochem Soc Symp 72:139–150 Mol Genet 19:3159–3168 152. Dimitriadi M et  al (2010) Conserved genes act as modifiers 133. Bernabo P et al (2017) In vivo translatome profiling in spinal of invertebrate SMN loss of function defects. PLoS Genet muscular atrophy reveals a role for SMN protein in ribosome 6:e1001172 biology. Cell Rep 21:953–965 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis 153. Dimitriadi M et al (2016) Decreased function of survival motor 172. Kamada Y, Funakoshi T, Shintani T, Nagano K, Ohsumi M, neuron protein impairs endocytic pathways. Proc Natl Acad Sci Ohsumi Y (2000) Tor-mediated induction of autophagy via an USA 113:E4377–E4386 Apg1 protein kinase complex. J Cell Biol 150:1507–1513 154. Dachs E, Hereu M, Piedrafita L, Casanovas A, Caldero J, 173. Wu L et al (2013) Rapamycin upregulates autophagy by inhib- Esquerda JE (2011) Defective neuromuscular junction organi- iting the mTOR-ULK1 pathway, resulting in reduced podocyte zation and postnatal myogenesis in mice with severe spinal mus- injury. PLoS One 8:e63799 cular atrophy. J Neuropathol Exp Neurol 70:444–461 174. Rodriguez-Muela N, Parkhitko A, Grass T, Gibbs RM, Nor- 155. Martinez-Hernandez R, Bernal S, Also-Rallo E, Alias L, Barcelo abuena EM, Perrimon N, Singh R, Rubin LL (2018) Blocking MJ, Hereu M, Esquerda JE, Tizzano EF (2013) Synaptic defects p62/SQSTM1-dependent SMN degradation ameliorates Spinal in type I spinal muscular atrophy in human development. J Pathol Muscular Atrophy disease phenotypes. J Clin Invest. https :// 229:49–61doi.org/10.1172/JCI95 231 156. Murray LM, Comley LH, Thomson D, Parkinson N, Talbot K, 175. Acsadi G, Lee I, Li X, Khaidakov M, Pecinova A, Parker Gillingwater TH (2008) Selective vulnerability of motor neurons GC, Huttemann M (2009) Mitochondrial dysfunction in a and dissociation of pre- and post-synaptic pathology at the neuro- neural cell model of spinal muscular atrophy. J Neurosci Res muscular junction in mouse models of spinal muscular atrophy. 87:2748–2756 Hum Mol Genet 17:949–962 176. Wang ZB, Zhang X, Li XJ (2013) Recapitulation of spinal motor 157. Engqvist-Goldstein AE, Drubin DG (2003) Actin assembly and neuron-specific disease phenotypes in a human cell model of endocytosis: from yeast to mammals. Annu Rev Cell Dev Biol spinal muscular atrophy. Cell Res 23:378–393 19:287–332 177. Xu CC, Denton KR, Wang ZB, Zhang X, Li XJ (2016) Abnormal 158. Kubler E, Riezman H (1993) Actin and fimbrin are required mitochondrial transport and morphology as early pathological for the internalization step of endocytosis in yeast. EMBO J changes in human models of spinal muscular atrophy. Dis Model 12:2855–2862 Mech 9:39–49 159. Hosseinibarkooie S et al (2016) The power of human protective 178. Patitucci TN, Ebert AD (2016) SMN deficiency does not modifiers: PLS3 and CORO1C unravel impaired endocytosis induce oxidative stress in SMA iPSC-derived astrocytes or in spinal muscular atrophy and rescue SMA phenotype. Am J motor neurons. Hum Mol Genet 25:514–523 Hum Genet 99:647–665 179. Miller N, Shi H, Zelikovich AS, Ma YC (2016) Motor neuron 160. Riessland M et al (2017) Neurocalcin delta suppression pro- mitochondrial dysfunction in spinal muscular atrophy. Hum tects against spinal muscular atrophy in humans and across Mol Genet 25:3395–3406 species by restoring impaired endocytosis. Am J Hum Genet 180. Ripolone M et al (2015) Impaired muscle mitochondrial bio- 100:297–315 genesis and myogenesis in spinal muscular atrophy. JAMA 161. Kaur J, Debnath J (2015) Autophagy at the crossroads of catab- Neurol 72:666–675 olism and anabolism. Nat Rev Mol Cell Biol 16:461–472 181. Fuller HR, Gillingwater TH, Wishart TM (2016) Common- 162. Lamb CA, Yoshimori T, Tooze SA (2013) The autophagosome: ality amid diversity: multi-study proteomic identification of origins unknown, biogenesis complex. Nat Rev Mol Cell Biol conserved disease mechanisms in spinal muscular atrophy. 14:759–774 Neuromuscul Disord 26:560–569 163. Martin S, Harper CB, May LM, Coulson EJ, Meunier FA, 182. Boyd PJ et  al (2017) Bioenergetic status modulates motor Osborne SL (2013) Inhibition of PIKfyve by YM-201636 neuron vulnerability and pathogenesis in a zebrafish model of dysregulates autophagy and leads to apoptosis-independent spinal muscular atrophy. PLoS Genet 13:e1006744 neuronal cell death. PLoS One 8:e60152 183. La Bella V, Kallenbach S, Pettmann B (2000) Expression 164. Mariño G, Niso-Santano M, Baehrecke EH, Kroemer G (2014) and subcellular localization of two isoforms of the survival Self-consumption: the interplay of autophagy and apoptosis. motor neuron protein in different cell types. J Neurosci Res Nat Rev Mol Cell Biol 15:81–94 62:346–356 165. Garcera A, Bahi N, Periyakaruppiah A, Arumugam S, Soler 184. van Wijk SJ, Timmers HT (2010) The family of ubiquitin-con- RM (2013) Survival motor neuron protein reduction deregu- jugating enzymes (E2s): deciding between life and death of pro- lates autophagy in spinal cord motoneurons in vitro. Cell Death teins. FASEB J 24:981–993 Dis 4:e686 185. Burnett BG, Munoz E, Tandon A, Kwon DY, Sumner CJ, Fisch- 166. Custer SK, Androphy EJ (2014) Autophagy dysregulation in beck KH (2009) Regulation of SMN protein stability. Mol Cell cell culture and animals models of spinal muscular atrophy. Biol 29:1107–1115 Mol Cell Neurosci 61:133–140 186. Han KJ et al (2012) Ubiquitin-specific protease 9x deubiquit- 167. Piras A et al (2017) Inhibition of autophagy delays motoneuron inates and stabilizes the spinal muscular atrophy protein-survival degeneration and extends lifespan in a mouse model of spinal motor neuron. J Biol Chem 287:43741–43752 muscular atrophy. Cell Death Dis 8:3223 187. Chang HC, Hung WC, Chuang YJ, Jong YJ (2004) Degrada- 168. Ichimura Y, Kominami E, Tanaka K, Komatsu M (2008) Selec- tion of survival motor neuron (SMN) protein is mediated via the tive turnover of p62/A170/SQSTM1 by autophagy. Autophagy ubiquitin/proteasome pathway. Neurochem Int 45:1107–1112 4:1063–1066 188. Hicke L (2001) Protein regulation by monoubiquitin. Nat Rev 169. Loos B, du Toit A, Hofmeyr JH (2014) Defining and meas- Mol Cell Biol 2:195–201 uring autophagosome flux—concept and reality. Autophagy 189. Han KJ, Foster D, Harhaj EW, Dzieciatkowska M, Hansen K, 10:2087–2096 Liu CW (2016) Monoubiquitination of survival motor neuron 170. Yoshii SR, Mizushima N (2017) Monitoring and measuring regulates its cellular localization and Cajal body integrity. Hum autophagy. Int J Mol Sci 18:E1685 Mol Genet 25:1392–1405 171. Periyakaruppiah A, de la Fuente S, Arumugam S, Bahí N, 190. Abera MB et al (2016) ML372 blocks SMN ubiquitination and Garcera A, Soler RM (2016) Autophagy modulators regulate improves spinal muscular atrophy pathology in mice. JCI Insight survival motor neuron protein stability in motoneurons. Exp 1:e88427 Neurol 283:287–297 191. Harahap NIF et al (2015) Salbutamol inhibits ubiquitin-mediated survival motor neuron protein degradation in spinal muscular atrophy cells. Biochem Biophys Rep 4:351–356 1 3 H. Chaytow et al. 192. Tiziano FD et al (2010) Salbutamol increases survival motor ubiquitination in primary spinal muscular atrophy fibroblasts. neuron (SMN) transcript levels in leucocytes of spinal muscular Clin Chim Acta 411:1920–1928 atrophy (SMA) patients: relevance for clinical trial design. J Med 204. Powis RA, Mutsaers CA, Wishart TM, Hunter G, Wirth B, Genet 47:856–858 Gillingwater TH (2014) Increased levels of UCHL1 are a com- 193. Fuller HR, Mandefro B, Shirran SL, Gross AR, Kaus AS, Bot- pensatory response to disrupted ubiquitin homeostasis in spinal ting CH, Morris GE, Sareen D (2015) Spinal muscular atrophy muscular atrophy and do not represent a viable therapeutic target. patient iPSC-derived motor neurons have reduced expression of Neuropathol Appl Neurobiol 40:873–887 proteins important in neuronal development. Front Cell Neurosci 205. Jablonka S, Sendtner M (2017) Developmental regulation of 9:506 SMN expression: pathophysiological implications and perspec- 194. Ramser J et al (2008) Rare missense and synonymous variants tives for therapy development in spinal muscular atrophy. Gene in UBE1 are associated with X-linked infantile spinal muscular Ther 24:506–513 atrophy. Am J Hum Genet 82:188–193 206. Cho S, Dreyfuss G (2010) A degron created by SMN2 exon 7 195. Liu HY, Pfleger CM (2013) Mutation in E1, the ubiquitin acti- skipping is a principal contributor to spinal muscular atrophy vating enzyme, reduces Drosophila lifespan and results in motor severity. Genes Dev 24:438–442 impairment. PLoS One 8:e32835 207. Seo J, Singh NN, Ottesen EW, Sivanesan S, Shishimorova M, 196. Wishart TM et al (2014) Dysregulation of ubiquitin homeostasis Singh RN (2016) Oxidative stress triggers body-wide skipping and beta-catenin signaling promote spinal muscular atrophy. J of multiple exons of the spinal muscular atrophy gene. PLoS One Clin Investig 124:1821–1834 11:e0154390 197. Powis RA et al (2016) Systemic restoration of UBA1 ameliorates 208. Singh NN, Seo J, Rahn SJ, Singh RN (2012) A multi-exon- disease in spinal muscular atrophy. JCI Insight 1:e87908 skipping detection assay reveals surprising diversity of splice 198. Kwon DY, Dimitriadi M, Terzic B, Cable C, Hart AC, Chitnis A, isoforms of spinal muscular atrophy genes. PLoS One 7:e49595 Fischbeck KH, Burnett BG (2013) The E3 ubiquitin ligase mind 209. Miguel-Aliaga I, Culetto E, Walker DS, Baylis HA, Sattelle DB, bomb 1 ubiquitinates and promotes the degradation of survival Davies KE (1999) The Caenorhabditis elegans orthologue of the of motor neuron protein. Mol Biol Cell 24:1863–1871 human gene responsible for spinal muscular atrophy is a maternal 199. Kim EK, Choi EJ (2017) SMN1 functions as a novel inhibitor for product critical for germline maturation and embryonic viability. TRAF6-mediated NF-kappaB signaling. Biochim Biophys Acta Hum Mol Genet 8:2133–2143 1864:760–770 210. Chan YB, Miguel-Aliaga I, Franks C, Thomas N, Trülzsch B, 200. Setsuie R, Wada K (2007) The functions of UCH-L1 and its rela- Sattelle DB, Davies KE, van den Heuvel M (2003) Neuromus- tion to neurodegenerative diseases. Neurochem Int 51:105–111 cular defects in a Drosophila survival motor neuron gene mutant. 201. Choi J, Levey AI, Weintraub ST, Rees HD, Gearing M, Chin Hum Mol Genet 12:1367–1376 LS, Li L (2004) Oxidative modifications and down-regulation 211. Rajendra TK, Gonsalvez GB, Walker MP, Shpargel KB, Salz HK, of ubiquitin carboxyl-terminal hydrolase L1 associated with Matera AG (2007) A Drosophila melanogaster model of spinal idiopathic Parkinson’s and Alzheimer’s diseases. J Biol Chem muscular atrophy reveals a function for SMN in striated muscle. 279:13256–13264 J Cell Biol 176:831–841 202. Castegna A et al (2002) Proteomic identification of oxidatively 212. Bowerman M, Murray LM, Beauvais A, Pinheiro B, Kothary modified proteins in Alzheimer’s disease brain. Part I: creatine R (2012) A critical smn threshold in mice dictates onset of an kinase BB, glutamine synthase, and ubiquitin carboxy-terminal intermediate spinal muscular atrophy phenotype associated with hydrolase L-1. Free Radic Biol Med 33:562–571 a distinct neuromuscular junction pathology. Neuromuscul Dis- 203. Hsu SH et  al (2010) Ubiquitin carboxyl-terminal hydrolase ord 22:263–276 L1 (UCHL1) regulates the level of SMN expression through 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cellular and Molecular Life Sciences Springer Journals

The role of survival motor neuron protein (SMN) in protein homeostasis

Free
18 pages
Loading next page...
 
/lp/springer_journal/the-role-of-survival-motor-neuron-protein-smn-in-protein-homeostasis-TVDo4ksLXW
Publisher
Springer International Publishing
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Cell Biology; Biomedicine, general; Life Sciences, general; Biochemistry, general
ISSN
1420-682X
eISSN
1420-9071
D.O.I.
10.1007/s00018-018-2849-1
Publisher site
See Article on Publisher Site

Abstract

Ever since loss of survival motor neuron (SMN) protein was identified as the direct cause of the childhood inherited neuro - degenerative disorder spinal muscular atrophy, significant efforts have been made to reveal the molecular functions of this ubiquitously expressed protein. Resulting research demonstrated that SMN plays important roles in multiple fundamental cellular homeostatic pathways, including a well-characterised role in the assembly of the spliceosome and biogenesis of ribonucleoproteins. More recent studies have shown that SMN is also involved in other housekeeping processes, including mRNA trafficking and local translation, cytoskeletal dynamics, endocytosis and autophagy. Moreover, SMN has been shown to influence mitochondria and bioenergetic pathways as well as regulate function of the ubiquitin–proteasome system. In this review, we summarise these diverse functions of SMN, confirming its key role in maintenance of the homeostatic environ - ment of the cell. Keywords Spinal muscular atrophy · Ribonucleoprotein · Translation · Cytoskeleton · Ubiquitin · Bioenergetic pathway Introduction “survival motor neuron”, despite subsequent research show- ing that it is a ubiquitously expressed protein, required by The survival motor neuron (SMN) protein was first high - all cells and tissue types, not just neurons [3]. Over the past lighted as a protein of interest when mutations in its cod- three decades, significant research efforts have sought to bet - ing gene, SMN1, were linked to the neuromuscular disease ter understand the mechanisms through which SMN acts [3]. spinal muscular atrophy (SMA) [1], a leading genetic cause Most of the resulting knowledge has been generated from of infant mortality. SMA presents in a range of severities animal models of SMA, where reduced expression of SMN with the most severe form, Type 1, being fatal within the reveals its role in several important intracellular processes, first 2 years of life. Patients show degeneration of α-motor which we will discuss in this review. neurons in the lower spinal cord leading to progressive mus- The full-length—294 amino acids, 38 kDa—human iso- cle weakness. The clear importance of SMN protein to the form of SMN (also known as FL-SMN, referred to as SMN motor system, alongside findings that knockout of Smn in hereafter) is mainly transcribed from the telomeric SMN1 mice was embryonically lethal [2], led to it being named gene, located on chromosome 5q13. SMN1 contains nine exons, 1, 2a, 2b, 3, 4, 5, 6, 7 and 8, with exon 8 remain- ing untranslated (Table 1). An inverted duplication in the Helena Chaytow, Yu-Ting Huang and Kiterie M. E. Faller region of SMN1 resulted in a second centromeric copy of the contributed equally. gene, termed SMN2, an evolutionarily recent event unique to Homo sapiens [4]. SMN2 differs from SMN1 at 5 bases, * Thomas H. Gillingwater and a C-to-T transition in exon 7 of SMN2 favours skipping t.gillingwater@ed.ac.uk of exon 7 during splicing, resulting in the majority of SMN2 Euan MacDonald Centre for Motor Neurone Disease products being a truncated isoform referred as SMNΔ7 [1, Research, University of Edinburgh, Edinburgh, UK 5]. However, limited amounts of SMN can still be produced Edinburgh Medical School: Biomedical Sciences, University from the SMN2 gene and it is known that the copy number of of Edinburgh, Edinburgh, UK SMN2 is inversely correlated with SMA disease severity [6]. Royal (Dick) School of Veterinary Studies, University Patients with homozygous null mutations of SMN1 carrying of Edinburgh, Edinburgh, UK Vol.:(0123456789) 1 3 H. Chaytow et al. Table 1 Main isoforms of SMN, their composition, expression and localisation SMN isoform Splicing Protein isoform Expression Localisation References Full-length SMN (FL- Exons 1, 2a, 2b, 3, 4, 5, Functional SMN protein High expression during Nuclear gems and cyto- [102, 205] SMN) 6, 7, 8 development, decreas- solic, including axons, ing into the adult CNS dendrites and synapses SMNΔ7 Exons 1, 2a, 2b, 3, 4, Degradation signal intro- High expression during Nuclear accumulation [104, 206] 5, 6, 8 duced by the change in development, decreas- C-terminal ing into the adult CNS Axonal-SMN (a-SMN) Inclusion of intron 3 Truncated protein due to Expressed during devel- Motor neuron axons [12] premature stop codon opment, not detected on the boundary of in the adult CNS exon 3/intron 3 SMN6B Inclusion of an Alu ele- Truncated protein due to Unknown Nuclear and cytosolic [10] ment forming exon 6B premature stop codon after exon 6B SMNΔ5 Exclusion of exon 5 Unknown Expressed in the mature Unknown [11] CNS Other splicing isoforms of SMN have also been discovered in cell cultures, although their role in  vivo is yet to be determined. These include isoforms excluding exons 3, 4 and 5 or multiple exons both in stressed and normal conditions [207, 208]. Skipping of any internal exons of SMN maintains the reading frame four or more copies of SMN2 show a less severe phenotype, highly conserved motif with a function in protein–protein later age onset, and can have a normal lifespan [7]. SMNΔ7 interactions [17, 18]. The SMN Tudor domain binds to the is highly unstable and quickly subjected to the ubiquitin–pro- C-terminal arginine- and glycine-rich tails of Sm proteins teasome pathway for degradation [8, 9]. Phylogenetic studies which contain symmetrical dimethylated arginine residues, further highlighted the importance of SMN, since it is highly thereby facilitating the assembly of spliceosomes as dis- conserved throughout evolution and there are even multiple cussed later [18–24]. Mutations in this domain, which dis- copies of SMN1 in the chimpanzee genome [4]. Other SMN rupt the interaction of SMN and Smith core (Sm) proteins, isoforms have been found in various tissues (Table 1). An are often found in SMA patients [18, 25–27]. “SMN6B” protein can be translated from both the SMN1 and The Tudor domain of SMN is also responsible for an SMN2 genes by the inclusion of an intronic Alu sequence interaction with coilin, a marker of Cajal bodies (CBs) [28]. as an alternative exon following exon 6 [10]. SMN6B is Liu and Dreyfuss originally described the localisation of present both in the nucleus and cytosol, and is twofold more SMN to nuclear bodies which they termed “gems” [29], stable than SMNΔ7 but twofold less stable than full-length and which are coilin negative as opposed to CBs. Gems are SMN. SMN6B is able to interact with Gemin2, although its composed of SMN complex proteins, whereas CBs are more physiological function is not fully understood [10]. mRNA complex nuclear structures to which SMN also localises, transcripts of another isoform, SMNΔ5 (with the exclusion and the presence of SMN in these nuclear bodies increases of exon 5), can be found in muscle and the central nervous during neuronal differentiation [30]. CBs are enriched with system (CNS) although, again, whether it has a physiologi- small nuclear RNAs (snRNAs) and small nucleolar RNAs cal function is not clear [11]. An axonal-SMN (a-SMN) has (snoRNAs), and are essential for the biogenesis of the small also been proposed, being produced from intron 3 retention nuclear ribonucleoproteins (snRNP) complex [31]. Interest- during splicing. Due to an in-frame stop codon in intron 3, ingly, motor neurons from a Type I SMA patient showed a a-SMN mRNA is much shorter and encodes a protein of reduced number of CBs and defects in recruitment of SMN 19 kDa. a-SMN is reported to be localised to the axon, and and snRNP for spliceosomal maturation [32]. Tapia et al. its expression is enhanced in the spinal cord and the brain [33] also demonstrated that SMN has a SUMO-interacting during development and declines in the adulthood, with a motif (SIM) in the Tudor domain, which is required for Sm hypothesised role in axonogenesis [12]. protein binding and the assembly of CBs. Three polypro- The SMN protein contains several functional motifs line stretches encoded by exons 4–6 are responsible for the including, moving from N-terminus to C-terminus, a basic/ binding of profilins, key proteins in the regulation of actin lysine-rich domain, a Tudor domain, a proline-rich domain dynamics [34, 35]. A tyrosine/glycine-rich region in the and a YG box. The basic/lysine-rich region is encoded by C-terminus of SMN, termed the YG box, is found to facili- exon 2 and has been found to interact with Gemin2 and tate oligomerisation of SMN by formation of the glycine- RNA in vitro and in vivo [13–16]. The Tudor domain is a zipper structure [36]. Mutations found in the YG box count 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis for nearly half of missense mutations in SMA patients and a cDNA copy of SMN2 lacking exon 7 or other variants of this motif was shown to be essential for survival in chick the SMN genes [43–45] (reviewed in [46]). Another widely cells under conditions of SMN depletion [37, 38]. The YG used animal model is the zebrafish (Danio rerio). Having the box is required for SMN self-oligomerisation and proteins advantages of well-characterised motor neurons and neuro- with mutations found in this motif, as seen in the Type I muscular junctions, easy manipulation of gene overexpres- SMA patients, severely impair this association [39]. A recent sion or knockdown by injection of in vitro-transcribed RNA study demonstrated that a phosphor degron within the YG or antisense oligonucleotides, respectively, and a transpar- Slmb box is exposed to SCF ubiquitin E3 ligase when SMN is ent body for imaging applications, zebrafish are becoming monomeric, implying that the YG box plays a role in SMN increasingly popular for translational SMA research pro- stability and indicating the importance of SMN oligomerisa- jects [47–50]. Fruit flies (Drosophila melanogaster) have tion [9]. Post-translational modic fi ations to the SMN protein also been used to study SMN biology. They possess one discern its localisation and function. As well as SUMOyla- copy of an Smn gene ortholog or DmSMN and can also be tion via the SIM regulating Sm protein binding, and ubiq- easily genetically manipulated. Each of the models referred uitination of SMN (discussed later), the protein may also to in this review are summarised in Table 2, with more com- be acetylated, which promotes a cytoplasmic localisation prehensive reviews of SMA models available elsewhere and increases its half-life [40], or phosphorylated on certain [51–54]. serine/threonine residues to promote its localisation to CBs. The selective cell death of motor neurons is a key feature Mutations of three tyrosine residues in the Tudor domain of the disease, but the reasons behind this selectivity are greatly affect its enrichment in CBs via disrupted interaction still poorly understood. A recent study demonstrated that with coilin [41, 42]. there were heterogeneous and surprisingly diverse expres- Several model organisms have been utilised to study sion levels of SMN in SMA-patient-derived iPSC motor the SMN protein and its role in SMA (Table 2). Although neurons. Moreover, motor neurons with lower levels of the protein product of SMN2 is truncated and unstable, its SMN protein were more susceptible to cell death from toxic expression is crucial for survival once SMN1 expression compounds, whilst overexpression of SMN in motor neu- has been lost. As SMN2 is specific to humans, most of the rons was protective [55]. SMN, therefore, clearly plays a commonly used mouse models have undergone genetic major role in SMA pathology and the specific vulnerability manipulation to generate an endogenous Smn null mutation to motor neurons in this disease. To understand why SMN with concurrent overexpression of the human SMN2 gene, is so vital for healthy cell maintenance, we must understand Table 2 Overview of animal models referred to in this review Species Endogenous Modelling strategy and/or genotype References SMN ortho- logue Caenorhabditis elegans CeSMN Knockdown of expression through RNAi [209] Drosophila melanogaster DmSMN Point mutations or transposon insertions for knockout or knockdown studies [210, 211] Danio rerio Smn Knockdown of expression through antisense oligonucleotides [47] Mus musculus Smn −/−  Smn knockout Smn Smn null mutation by targeted insertion of [2] β-galactosidase in Smn exon 2A H7/H7 tg/−  Taiwanese mice Smn ; SMN2Hung Two copies of the SMN2 transgene, Smn [43] exon 7 is replaced with hypoxanthine phosphoribosyl transferase (HPRT) but transcripts without exon 7 are produced. 2A/2A tg/tg tg/tg  SMNΔ7 Smn ; SMN2 ; SMN∆7 One copy of the SMN2 transgene and one [45] SMNΔ7 transgene on Smn null back- ground 2B/−  Smn2B Smn Mutation within the splicing enhancer of [212] Smn exon 7 producing SMN2-like tran- scripts and reduced FL-SMN protein −/− tg/tg  Burghes severe model Smn ; SMN2 Smn null mutation by target replace of [44] β-galactosidase in Smn exon 2A; with one copy of SMN2 For a comprehensive review of animal models of SMA, see Edens et al. [51] 1 3 H. Chaytow et al. its role under normal, as well as disease, conditions. In this to the TMG cap [72, 74–77]. In that process, SMN has been review, we describe the role of the SMN protein in regulat- shown to have a direct interaction with importin-β facilitated ing protein homeostasis. Protein homeostasis within cells by WRAP 53 [78, 79]. WRAP 53 also plays a fundamental can be regulated by two major pathways, production and role in the trafficking of SMN towards CBs by facilitating clearance, which reach a dynamic balance to support and the interaction between SMN and coilin [80]. The snRNA maintain physiological status. Production pathways incorpo- then dissociates from the SMN complex and undergoes its rate protein translation, folding, modification and assembly, final maturation steps within the CB. Studies of splicing while protein clearance pathways are centred around protein activity in cells from SMA patients or mouse models con- disassembly and degradation. We discuss known functions firm the fundamental role of SMN in snRNP assembly with of the SMN protein, starting with its first-described role in a correlation between the reduction in snRNPs levels and RNA splicing and spliceosomal assembly, followed by more disease severity [81–83]. A recent study has also identified recently discovered functions in regulating mitochondrial an alternative assembly pathway, whereby the U1-specific homeostasis, endocytosis and the cytoskeleton, ubiquitina- RNA-binding protein (RBP) U1-70K can directly interact tion and autophagy, and RNA transportation, thus giving a with the SMN-Gemin2 complex, independently of Gemin5. broad picture of the many ways in which SMN plays a key This U1-specific Sm core-assembly pathway not only con- role in regulating protein homeostasis. tributes to U1 overabundance, but it was also proposed that SMN-Gemin2 could play a role as a hub, where various RBPs and their RNA cargos congregate, hence promoting SMN and ribonucleoprotein assembly ribonucleoprotein exchange [84]. The SMN complex is also involved in the biogenesis Although it is now clear that the SMN protein contrib- of U7 snRNPs, a specific subgroup of snRNPs which are utes to numerous cellular processes and pathways, the first involved in processing the 3′ stem loop of histone mRNAs identified and most studied function of SMN is its role in by endonucleolytic cleavage of the pre-mRNA sequence snRNP assembly. The spliceosome is a complex machine, which immediately follows the hairpin [85]. The assembly which catalyses the removal of introns from pre-mRNA of U7 snRNPs is overall analogous to that of the spliceo- transcripts (see [56] for a detailed review). The biogene- some snRNPs, with the exception of the slightly degenerate sis of an snRNP starts in the nucleus by the transcription Sm-binding site of the U7 snRNA and the replacement of of uridine-rich snRNAs (U1, U2, U4, U5, U6, U11, U12, two of the Sm proteins in the Sm core (SmD1 and SmD2) U4atac, U6atac), which are then exported to the cytoplasm. by two U7-specific Sm-like proteins (Lsm10 and Lsm11). Each snRNA is then bound to accompanying proteins and— Similar to spliceosome assembly, the SMN complex is a with the exception of U6 and U6atac—to a set of seven Sm specificity chaperone that is necessary to precisely recognise proteins. Whilst Sm proteins can spontaneously associate and combine U7 snRNA with the Sm heptamers containing in vitro with almost any single short-stranded uridine-rich Lsm10 and Lsm11, without which the U7 snRNPs cannot RNA forming a thermodynamically stable heptamer [57], function in histone RNA processing [86–88]. this process lacks specificity regarding RNA substrates. Although less extensively studied, the SMN complex is The role of the SMN protein, tightly associated with eight suspected to be involved in the assembly and metabolism of other proteins (Gemin 2–8 and Unrip) in a large macromo- other ribonucleotide complexes, including small nucleolar lecular SMN complex, is to chaperone the biogenesis of ribonucleoproteins (snoRNPs), associated with the post- snRNPs from snRNAs and Sm proteins in the cytoplasm, transcriptional processing and modification of ribosomal and subsequent snRNP trafficking to the nucleus [14, 29, RNA in the nucleolus (methylation and pseudouridylation) 58–66]. First, the SMN complex enforces specificity during [89]. Indeed, SMN has been shown to directly interact with the snRNPs’ assembly with the direct and specific binding fibrillarin and GAR1, two markers of snoRNPs, and expres- of Gemin5 to the cytoplasmic snRNAs [67–71]. The Sm sion of a dominant-negative mutant of SMN results in abnor- core is then loaded onto the snRNA in an ATP-dependent mal accumulation of snoRNPs in large aggregates outside process with Gemin2 playing a key architectural role in this of the nucleolus [90]. Furthermore, in SMA-patient-derived assembly. Finally, the snRNA undergoes hypermethylation cells, a decreased localisation in CBs of the snoRNP chaper- of its m7G-cap by TGS1 (trimethylguanosine synthetase one Nopp140 was observed, which correlated with disease 1), leading to the formation of a trimethylguanosine (TMG) severity [91]. In addition, SMN may be involved with tel- cap and trimming of its 3′ end before it can be imported omerase, a large RNP complex that adds repeat sequences back into the nucleus. The TMG cap and Sm core operate at the chromosomal ends. It comprises a telomerase reverse as nuclear localisation signals [72, 73]. Importation to the transcriptase (TERT), the telomerase RNA and other associ- nucleus also necessitates the binding of the nuclear import ated proteins (for a review on telomerase RNA, see [92]). complex (snurportin and nuclear import receptor importin-β) The telomerase RNP belongs to the H/ACA snoRNP class 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis and it is suspected that SMN plays a role in telomerase bio- Recent studies have identified that SMN can bind to the genesis, ensuring specificity of assembly and correct traf- α-COP subunit of the COPI vesicle [107]. The COPI system, ficking [93, 94]. a Golgi-derived vesicular transport system, is involved in As snRNP assembly and splicing occurs in all cells, intracellular trafficking in neurites, necessary for the mat- including neurons, why do low levels of SMN in SMA par- uration of neuronal cell processes [108]. Knocking down ticularly affect motor neurons [95]? This remains a major α-COP was found to disrupt SMN localisation within growth question challenging the SMN and SMA research field. As cones, resulting in its accumulation within the trans-golgi previously noted, the reduction in snRNP biogenesis cor- network [109]. Depletion of α-COP reduced neurite forma- relates with the degree of clinical severity in SMA [81]. tion in NSC-34 cells and primary cortical neurons, with However, SMN deficiency seems to preferentially disrupt shortening of both map2-positive dendrites and tau-positive the formation of minor snRNPs, such as those responsible axons [110], and both α-COP and SMN are required for cor- for the removal of U12-containing intron genes [81, 83, rect neurite formation [111]. This indicates a role for SMN 96]. Amongst these, the gene coding for a transmembrane in trafficking for the purposes of neuronal outgrowth and for - protein, stasimon, has been identified as being aberrantly mation of the axonal and synaptic cytoskeleton (see below). spliced in a Drosophila model of SMA [97]. Upregula- In keeping with this potential role for SMN, Rossoll and tion of stasimon rescued deficient neuromuscular junc- colleagues discovered an interaction between SMN and tion (NMJ) transmission in SMN-deficient Drosophila and the RBP hnRNP-R [112]. SMN and hnRNP-R were found improved neuronal development in SMN-deficient zebrafish to co-localise in the cytoplasm of primary cultured motor −/− [97]. Other genes, not all containing U12-introns, such as neurons, and in motor neurons cultures from Smn mice, chondrolectin, agrin and neurexin2 have also been identi- there was a large reduction in β-actin mRNA localisation in fied as being abnormally spliced and could, therefore, play axons and growth cones. Primary motor neurons cultured a role in the pathophysiology of the disease [98–100]. It from Taiwanese SMA mice showed growth cones with a could consequently be the case that defects in splicing have threefold reduction in size compared to healthy controls, as a larger effect on a specific subset of neuronal genes, thereby well as reduced staining for β-actin mRNA with no overall rendering motor neurons particularly vulnerable. However, change in protein expression [112]. Since these initial find- despite this evidence for mis-splicing in the SMA disease ings, fluorescence in situ hybridisation experiments against pathway, other studies have suggested that wide-spread the polyA tails of mRNA revealed a more than 50% reduc- splicing defects mainly occur during the late stage of the tion in localisation of mRNA transcripts along the axon of disease [101], supporting the theory that alternative roles primary motor neurons following SMN knockdown [113]. of SMN may play an equally important part in cell function. In addition, further co-localisation studies have shown SMN to associate with a number of RBPs via its Tudor domain, including KSRP [114], FMRP [115], HuD [113, 116], FUS SMN and trafficking [117] and IMP1 [118]. The association between SMN and other RBPs has linked it to another motor neuron disease, The first indications that SMN played a role aside from its amyotrophic lateral sclerosis (ALS). RBPs associated with canonical functions in the spliceosome came when electron ALS, FUS and TDP-43 have been shown to co-localise in microscopy revealed localisation of SMN in the dendrites nuclear gems with SMN and mutations in either of their and axons of motor neurons in the developing rat spinal cord genes in ALS patient fibroblasts show reduced gem forma- [102]. It has been suggested that there is a progressive shift tion leading to abnormal accumulation of snRNAs in the in SMN protein localisation from mainly nuclear during nucleus [119, 120]. This highlights an interesting mecha- development to a more cytoplasmic and axonal localisa- nistic link between ALS and SMA. tion in the mature neuron [103]. SMN was also found to be SMN acts as a molecular chaperone for the binding of present at the growth cones of cultured motor neurons, and RBPs to mRNA transcripts as well the RBPs’ binding to the live cell imaging showed puncta positive for SMN being cytoskeleton and subsequent localisation, as evidenced by actively transported bi-directionally along axons [104]. disruption of these processes in SMN deficiency [121]. Both SMN co-localises with some elements of the SMN complex IMP1 and HuD have been shown to influence the locali- in the axon, such as Gemin2, but Sm proteins show very sation and translation of β-actin and GAP-43 mRNA tran- low abundance in distal neurites, and most axonally located scripts, which are in turn both necessary for correct axonal SMN granules lack Sm proteins [105]. The neuron-specific growth [113, 118, 122]. Indeed, SMN knockdown leads to protein neurochondrin is required for the correct localisation a reduction of HuD in the axonal compartment [113], while of SMN in the cytoplasm, and neurochondrin was found not knockdown of HuD in zebrafish leads to a similar phenotype to co-localise with snRNPs, further indicating that SMN is to SMN knockdown of shorter axons [122]. Further experi- involved in activities other than splicing [106]. ments using FLAG-tagged SMN in NSC-34 cells [123] 1 3 H. Chaytow et al. identified SMN as a binding partner for several species of transcribed in a cluster and are associated with increased non-coding RNAs, including snRNAs, snoRNAs and ribo- cell proliferation via the mTOR pathway [130]. Primary somal RNAs, which were expected due to SMN’s known motor neurons with a 50% knockdown of SMN pro- role in RNA processing, as well as miRNAs and tRNAs. tein showed increased expression of miR-183 in neu- The majority of RNAs identified were mRNA, with many rites whereas there was no change in expression in the being part of ribosomal and/or metabolic pathways. When cell body, along with downregulation of proteins in the compared to mRNAs known to localise to neuronal axons, mTOR pathway [131]. It is possible that SMN regulates the study’s authors identified 75 axonally localised SMN- the mTOR pathway and, therefore, protein translation associated mRNAs, including RNA transcripts of several through miR-183, since motor neurons in the Taiwanese ribosomal proteins and Ubb, the transcript of the protein mouse model and SMA-patient-derived fibroblasts both ubiquitin [123]. However, it should be noted that this group showed reduced levels of de novo protein synthesis, and of mRNAs is unlikely to be comprehensive, as it does not a knockdown of miR-183 in Taiwanese mice produced include the known RNA transcripts regulated by SMN a mild rescue of the phenotype with improved survival including β-actin, GAP43, tau and neuritin [112, 116, 124]. and increased body weight [131]. Another indication that SMN interacts with the mTOR pathway came from studies examining the effect of manipulating the PTEN pathway SMN and translation on primary motor neurons of SMNΔ7 mice. PTEN is a negative regulator of the mTOR pathway, and in SMN- While SMN plays an integral role in the transport of RNA depleted primary motor neurons where axonal growth was transcripts along axons and dendrites, it also appears to be defective, decrease of PTEN/activation of mTOR rescued involved in the local translation of proteins. The transporta- the SMA phenotype [132]. tion of mRNA transcripts along the axon allows for rapid Most of the work detailed above was performed in vitro, protein turnover in distal regions of the neuron in response where primary cultures allow analysis of changes in axonal to, for example, activity [125]. Dysregulation of local trans- growth and the ability to isolate axonal compartments lation has been associated with several other neurodegen- relatively easily. However, in vivo evidence is important erative disorders, including Alzheimer’s disease and amyo- to determine the role of SMN in these mechanisms. A trophic lateral sclerosis (reviewed in [126]). recent study examined translational pathways in vivo and Recent evidence has pointed to a role for SMN in the local found SMN associating with ribosomes in control tissue, translation of mRNA transcripts, as well as their localisa- as well as a shift in residual SMN levels to non-ribosomal tion. Early on, it was reported that loss of SMN changed the fractions and an overall reduction in the number of ribo- expression of plastin-3 in a zebrafish model of SMA [127]. somes associated with polysomes in Taiwanese SMA mice Although, at the time of this discovery, the mechanisms [133]. This study also compared the whole transcriptome behind changes in protein expression were not clear, more to the translatome using next-generation sequencing, and recent studies suggest that SMN affects the local translatome confirmed significant deficiencies in translation following through several mechanisms. Translation within the axon reduced SMN expression in vivo [133], including results was found to be dysregulated in primary neurons derived that point to defects in the biogenesis of ribosomes, sug- −/− from SMN mice in vitro, with no corresponding change in gesting a possible explanation for translational defects that somatic translation, and axonal defects could be rescued by occur upon SMN depletion. overexpression of the RNA binding proteins HuD and IMP1 Taken together, the studies detailed above strongly sug- [128], suggesting a link between the role of SMN in mRNA gest that SMN plays an important role in regulating pro- trafficking and translation. Furthermore, ultrafractionation of tein translation through several mechanisms. First, through cell extracts from a motor neuron-like cell line revealed an subcellular localisation of mRNAs along the axon; second, association of the SMN protein with polyribosomes, whilst through association with ribosomes themselves regulating treatment with RNase displaced RBPs associated with the the availability of ribosomal units for local translation and polyribosomes such as SMN, but also other known binding finally, through regulation of the mTOR pathway. In this proteins such as FMRP [129]. When SMN was introduced way, SMN is well positioned to play a role in the develop- to an in vitro translation system, there was a dose-dependent mental polarisation of motor neurons, as well as control reduction in translation efficiency, with no change in transla - their growth, maturation and proper function. This crucial tion when incubated with the SMNΔ7 fragment [129]. role, alongside the expansive physical size of motor neu- Alongside its direct interaction with ribosomes, sug- rons, may partly explain why motor neurons are particu- gesting a possible direct role in translational control, SMN larly susceptible to loss of SMN, as opposed to other more may also influence protein translation through micro- ubiquitous roles that the protein plays in the cell. RNAs. MicroRNAs miR-183, miR-96 and miR-182 are 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis its functional change from being an attractant to a repel- SMN and the cytoskeleton lent signalling factor, thereby contributing to growth cone collapse [145]. In the same study, cleaved PlexinD1 was The cytoskeleton—incorporating key components such as found to be enriched in actin rods, a pathological struc- microtubules, neurofilaments and actin protein—plays a ture of elongated actin aggregates also found in some age- fundamental role in regulating neuronal architecture and related neurodegenerative diseases but not in control cells. function. It is crucial for signalling and trafficking of vari- Interestingly, by studying discordant families where sib- ous molecules, but also for the formation of growth cones lings of SMA patients were asymptomatic despite carrying during neuronal development. Therefore, it is perhaps not the same SMN1 and SMN2 alleles as their affected siblings, surprising that defects in the cytoskeleton have been linked Oprea et al. [146] were able to identify the first protective to several neurodegenerative disorders, including SMA genetic modifier of SMA: plastin 3 (PLS3). PLS3 is involved (for reviews on the neuronal cytoskeleton see [134, 135]). in axonogenesis by bundling F-actin and stabilising growth The observation that SMN localised to neurites and cones. Its overexpression was able to rescue the axon out- growth cones [102, 105, 136–138] and that SMN modu- growth defects in SMN-deficient zebrafish and increase lated the localisation of β-actin within growth cones [112] the life span of the intermediate Smn2B model [146, 147]. provided the first hints of a possible role for SMN in regu- Other studies have suggested additional roles for SMN in lating cytoskeletal dynamics. At the same time, knock- regulating microtubule formation, required for transporting ing down SMN in zebrafish was found to result in motor mRNAs, proteins and organelles to or from the nucleus to axon pathfinding deficits [47], whilst SMN-deficient cell distal regions of the neuron (reviewed in [148]). Stathmin, cultures showed neurite extension defects [104, 112, 139, a protein known to promote microtubule depolymerisation 140]. The fact that selective overexpression of the SMN [149] was found to be upregulated in the spinal cord in Tai- C-terminal domain could rescue these neurite deficits in wanese mice and also in SMN-depleted NSC-34 cells lead- SMN-deficient PC12 cells argued in favour of a role of ing to defects in the structure of axons and reduced mito- SMN in microfilament metabolism independent of snRNP chondrial transport along the axons [150]. In SMN-deficient biogenesis, as the Tudor domain where Sm proteins binds cells, microtubules failed to re-polymerise following treat- was not present in this C-terminal construct [139]. ment with the microtubule-depolymerisation agent nocoda- Growth cone formation and neurite extension are medi- zole, an effect which could be rescued by knocking down ated by actin dynamics, and SMN has been found to colo- stathmin [150]. However, the detailed mechanisms linking calize with profilin 2a during neurite cell extension [141] SMN to these pathways, and whether or not they are indeed and in nuclear gems [34]. Profilin 2a is an actin-binding separate from SMN’s involvement in the mRNA traffick - protein primarily expressed in the nervous system where ing of components of the cytoskeleton such as GAP-43 and it is involved in the regulation of actin turnover by pro- β-actin, remains to be determined. moting actin polymerisation. Profilin 2a binds to a stretch of proline residue within the SMN protein [35], and this interaction with SMN modulates the activity of profilin. SMN and endocytosis Profilin 2a is also a known downstream target of the rho- kinase (ROCK) pathway, a key regulator of actin dynamics Endocytosis is a basic cellular process, essential for neuronal (reviewed in [142]). Knocking down SMN in PC12 cells signalling, axonal and dendritic growth (reviewed in [151]). resulted in an upregulation of profilin 2a, which, combined It plays a particularly important role at synapses (including with its increased availability due to decreased interaction at neuromuscular junction synapses formed by motor neu- with SMN, lead to an upregulation of the ROCK pathway rons), facilitating synaptic vesicle recycling necessary for with subsequent inhibition of neuronal outgrowth [143]. repeated rounds of neurotransmitter release. A bioinformat- ROCK pathway inhibition in an intermediate SMA mouse ics analysis carried out on two different species (Caenorhab- model (Smn2B) also resulted in increased life span and ditis elegans and D. melanogaster) identified the endocytic amelioration of muscle pathology [144] (see Fig.  1 for pathway, along with mRNA regulation, as potential modi- a summary of the role of SMN in cytoskeleton dynam- fiers of SMN loss [152], with numerous individual genes ics). Moreover, it was recently suggested that SMN loss being highlighted. In another study, SMN depletion resulted resulted in the dysregulation of the actin cytoskeleton in a marked impairment of endocytic function in multiple by interfering with PlexinD1. PlexinD1 is a receptor for tissues of C. elegans [153]. The neuromuscular junction was class 3 semaphorins and acts as a signalling factor to guide particularly affected, with structural and functional changes axonal growth. In the Taiwanese mouse model and in being reported. A reduction in the number of pre-synaptic iPSC-derived motor neurons from SMA patients, PlexinD1 docked vesicles was observed, accompanied by unusually was shown to be cleaved by metalloproteases, resulting in large cisternae suggestive of arrested endocytic vesicle 1 3 P lin 2a H. Chaytow et al. Kinesin RNA granule (Pre)synaptic vesicle Clathrin TRANSPORT Plastin 3 Microtubules F-actin G-actin Mitochondrion Ion channel Direct inhibition by SMN In dark blue: changes associated with SMN loss Microtubule destabilisation SMN Actin depolymerisation ENDOCYTOSIS 2+ Ca Fig. 1 Schematic overview of the alterations in cytoskeletal dynamics transport (e.g. synaptic vesicle, RNA granules and mitochondria) and and endocytosis observed following SMN deficiency. The diagram endocytosis. In the absence of SMN, not only a destabilisation of the highlights these changes at the level of the motor neuron and neuro- microtubules is observed, but also a depolymerisation of the actin muscular junction. All changes associated with SMN loss are repre- cytoskeleton, which has been linked to the activation of the RhoA/ sented in dark blue. SMN deficiency results in a decrease in cellular ROCK pathway maturation [153]. This was associated with a decreased In this model, endosomal defects were noted not only at the activity of, and disruption to, the NMJ (a key feature of level of the NMJ, but also in non-neural tissue as endocytic SMA [154–156]): synaptic transmission was reduced, likely activity in coelomocyte cells was lower. The importance of secondary to an impairment in synaptic vesicle recycling. SMN for NMJ homeostasis was further demonstrated in the 1 3 RhoA/ROCK LIMK The role of survival motor neuron protein (SMN) in protein homeostasis Taiwanese model of SMA, where pre-synaptic uptake of [161]). Autophagy involves a double-membrane bound FM1-43 dye by endocytosis was significantly reduced upon structure engulfing target proteins and organelles to form electrical stimulation. Interestingly, this disturbance was an autophagosome. The autophagosome later fuses with lys- restored by PLS3 overexpression. The fact that PLS3 overex- osomes to become an autolysosome, in which the proteins pression could improve the endocytic defect was perhaps not and organelles are degraded (reviewed in [162]). Autophagy surprising, however, as the actin cytoskeleton is required for is a finely balanced mechanism: a decrease in expression this process [157] and yeast cells lacking Sac6p, the PLS3 of autophagy-related genes may lead to the accumulation ortholog, are defective for the internalisation step of endocy- of unwanted proteins whereas over-active autophagy leads tosis [158]. Moreover, another F-actin binding and bundling to increased numbers of autophagosomes, possibly leading protein, coronin 1C (CORO1C), has been shown to interact to cell death [163, 164]. Both of these outcomes have been with PLS3 and its overexpression rescued endocytosis in described in various models of SMN depletion, indicating a SMN-deficient cells and improved the axonal phenotype role for SMN in the regulation of autophagy. in Smn-deficient zebrafish [159]. The importance of SMN It is debatable whether an increase in amount of for endocytic processes has also been confirmed in SMA- autophagosomes is protective or deleterious to the cell. patient-derived cells, which proved resistant to infection by Through measuring expression of LC3-II, a marker of a clathrin endocytosis-dependent virus [153] (see Fig. 1 for autophagosomes, it has been shown that autophagosome a summary of the role of SMN in endocytosis). number is increased in primary motor neurons following Using the same approach that led to the discovery of lentiviral SMN knockdown [165] and in spinal cords of PLS3 as a modifier of SMA, a second modifier, neuronal cal- the Taiwanese mouse model [166] and the SMNΔ7 mouse cium sense protein neurocalcin delta (NCALD) was recently model [167]. Another way of measuring autophagic activ- reported [160]. Contrary to PLS3 which acts as a positive ity is through autophagic flux indicated by the level of p62/ regulator of endocytosis, NCALD is a negative regulator SQSTM1 protein [168–170]. Again, the p62 protein level of endocytosis and axonal growth. Knockdown of NCALD was found to be upregulated in the spinal cord of Burghes restored neurite outgrowth in SMN-deficient cells and severe SMA mice compared to their control littermates improved axonal growth and NMJ function in a zebrafish [171], as well as in an NSC-34 cell line following lentivi- model of SMA. An enhanced neuromuscular function in C. ral SMN knockdown, and in the spinal cord of Taiwanese elegans and murine models of SMA was also observed fol- SMA mice [166], indicating a reduction in autophagic flux. lowing NCALD depletion [160]. In the absence of calcium Inconsistent with data from the Taiwanese mouse model, or at low calcium levels, NCALD, which localises to growth autophagic flux did not appear to increase in the spinal cord cones and pre-synaptic sites at the NMJ, interacts with clath- of SMNΔ7 mice [166, 167]. Inhibition of lysosomal prote- rin, which mediates the endocytosis needed for fast recycling olysis with Bafilomycin A1 (BafA1) resulted in an accumu- at axon terminals. Low SMN levels have been shown to lead lation of LC3-II in cultured motor neurons from the Burghes 2+ to a reduction of voltage-activated Ca influx [98, 160], severe model, suggesting that SMN deficiency can activate and it is possibly through this mechanism that endocytosis autophagy [171]. and vesicle recycling was impaired. It was postulated that, Conversely, autophagy modulators can alter SMN pro- 2+ in normal motor neurons, the high local Ca concentra- tein levels. Treating cultured motor neurons isolated from tion observed following neurotransmitter release led to the wild-type mice with mTORC1 inhibitor rapamycin, which dissociation of NCALD from clathrin, therefore “freeing” is believed to enhance the activity of autophagy [172, 173], clathrin to perform its endocytic function. In SMA, due to showed increased SMN levels, whilst in BafA1-treated 2+ low Ca concentrations, dissociation did not occur and the motor neurons SMN levels were decreased [171]. A recent clathrin was, therefore, not available for coating of the vesi- study has indicated that SMN may be partially degraded cles. Moreover, disturbed calcium homeostasis would also through the autophagy pathway, since a knockdown of p62 be predicted to affect the function of actin-bundling proteins in stem cell-derived motor neurons from SMNΔ7 mice PLS3 and CORO1C, giving further strength to the hypoth- increased SMN protein levels [174]. A role for SMN in esis that low calcium levels secondary to SMN deficiency autophagy is also supported by the finding that overexpres- play an important role in endocytosis impairment [158]. sion of the SMN-binding partner α-COP, normally involved in cytoskeletal growth [110], partially restored autophagic flux in SMN-depleted cells [166], although the mecha- SMN and autophagy nism involved remains unclear. Moreover, injection of the autophagy inhibitor 3-methyladenine (3-MA) into SMNΔ7 Autophagy is a highly conserved catabolic process utilised mice at P3 greatly reduced autophagic activity and protected by cells to break down unwanted macromolecules such motor neurons from degeneration, possibly via inhibition as aggregated proteins or cellular organelles (reviewed in of the apoptotic pathway as shown by reduced expression 1 3 H. Chaytow et al. of apoptotic markers [167]. On the other hand, rapamycin Mitochondrial oxidative phosphorylation is a core part failed to influence the loss of motor neurons, but reduced of bioenergetic pathways. Mitochondrial electron trans- survival significantly in SMNΔ7 mice [167]. These con- port chain function relies on a supply of electrons from flicting findings show that further work is still required to the carriers NADH and FADH through upstream reac- fully elucidate the interaction between SMN and autophagy tions (mainly glycolysis and TCA cycle). Proteomics stud- pathways. ies identified that bioenergetics pathways were affected by SMN deficiency, more specifically GAPDH, an enzyme of the glycolysis pathway, was downregulated in SMA models SMN, mitochondrial homeostasis [181]. Interestingly, gene expression studies of affected and and bioenergetics pathways disease-resistant motor neuron pools in mice revealed that susceptible neurons had lower basal expression not only of SMN deficiency has been linked to changes in oxidative specifically mitochondria-related genes but also of genes stress, mitochondrial dysfunction and impairment of bioen- involved in more generic bioenergetic pathways. Specifi- ergetic pathways. Acsadi et al. [175] showed that knocking cally, the expression of PGK1, a key enzyme of the glyco- down SMN levels by ~ 66% in NSC-34 cells resulted in a lytic pathway, was significantly elevated in motor neurons marked reduction in ATP levels. This was associated with an that are intrinsically resistant to low levels of SMN, with increase in cytochrome c oxidase activity and mitochondrial experimental elevation/activation of PGK1 sufficient to res- membrane potential, resulting in increased free radical pro- cue motor axon defects and loss of neuromuscular function duction. This increase in oxidative stress in SMN-deficient in a zebrafish model of SMA [182]. cells was further confirmed in spinal motor neurons derived Taken together, these studies highlight that SMN defi- from human embryonic stem cells (hESCs). Interestingly, ciency leads to impairment in mitochondria and bioenerget- mitochondrial superoxide production was only increased in ics pathways. However, the precise mechanisms involved the SMN-knockdown hESCs which were made to differenti- in these interactions remain unclear. Studies in various cell ate into spinal motor neurons, but not in the cells differenti- types have shown that SMN does not localise to mitochon- ated into forebrain neurons [176]. dria [175, 183]. Therefore, it has been postulated that the Further analysis of mitochondrial dysfunction was per- effects of SMN on mitochondrial function could be indirect, formed by the same group using two models of SMN-defi- possibly by affecting preferentially the splicing, translation cient cells, SMA Type 1 patient-specific-induced pluripo- or mRNA transport of genes fundamental to mitochondrial tent stem cells (iPSCs) and SMN-knockdown hESCs, both homeostasis [175, 177]. As previously mentioned, cytoskel- differentiated into spinal motor neurons [177]. Impaired etal changes can also lead to decreased mitochondrial trans- mitochondrial axonal transport and a reduction in axonal port, particularly within long axons [150]. Therefore, fur- mitochondrial number and area were noted at early stages ther studies are required to better understand how SMN of cell culture. Partial rescue by the anti-oxidant N-acetyl- affects these energetic pathways, fundamental for cellular cysteine provides evidence to support the hypothesis that homeostasis. oxidative stress plays an important role in neuronal degen- eration in SMN-deficient motor neurons. However, experi- ments on SMA patient iPSCs led to conflicting results as, SMN and ubiquitin pathways in this model, no oxidative stress was detected [178]. These inconsistencies could be secondary to differences in the way Another key mechanism required for protein homeostasis is the stem cells were differentiated and highlight the limita- the protein degradation pathway. There are two major routes tions of studying cell type-specific pathological processes of protein degradation in eukaryotes: the ubiquitin–protea- in cell cultures. More recently, studies in SMNΔ7 and Tai- some system (UPS) and lysosomal proteolysis, or autophagy wanese mouse models confirmed marked mitochondrial (see above). The mammalian ubiquitin pathway is initiated dysfunction in spinal motor neurons, with decreased basal by activation of the E1 ubiquitin-activating enzyme UBA1, and maximal mitochondrial respiration, impaired mitochon- which then transfers ubiquitin onto one of around 40 E2 drial membrane potential, impaired mitochondrial mobility, conjugating enzymes. E2 ligases control whether a substrate increased oxidative stress level and increased fragmentation is mono- or polyubiquitinated [184]. E3 ligases (of which [179]. Interestingly, mitochondrial defects in SMA are not there are several hundred) collect the substrate protein and thought to be limited to motor neurons in vivo, as they have form a complex between it and the ubiquitinated E2 ligase, also been identified in SMA patient muscle associated with where the ubiquitin is transferred onto the protein substrate. a downregulation of mitochondrial biogenesis regulatory Ubiquitination is a dynamic process, and proteins can be factors [180]. stripped of their ubiquitin by deubiquitinating enzymes. 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis SMN has been shown to be ubiquitinated and ultimately patients treated with salbutamol also showed an increase degraded via the ubiquitin–proteasome system, with a pro- in SMN levels in the blood [192]. tein half-life of between 6 and 10 h depending on the cell Through proteomic analysis, SMN has been found to line analysed [185, 186]. Inhibition of the proteasome in interact with several components of the ubiquitin pathway, SMA-patient-derived fibroblasts increased the intracel- including UBA1 and several E3 ligases, as summarised lular abundance of SMN, both in terms of the amount of in Fig. 2 [9, 193]. Mutations in the UBA1 gene cause the SMN protein and the number of nuclear gems [187]. Mon- disease X-linked SMA [194], a rare condition with simi- oubiquitination, as opposed to polyubiquitination, serves lar symptoms to classical SMA but with no mutations in other functions in the cell instead of degradation including the SMN1 gene, suggesting a link between UBA1 and SMN protein trafficking and intracellular localisation (reviewed which, when lost, leads to SMA-like phenotypes. Mutations in [188]) and SMN is known to be monoubiquitinated in the Drosophila homologue of UBA1 cause motor defects, [186]. Indeed, preventing the monoubiquitination of SMN indicating that the motor system is particularly suscepti- changed the localisation of the protein from the cytoplasm ble to the loss of UBA1 despite its ubiquitous expression to the nucleus, and also prevented its co-localisation with [195]. Proteomic analysis of hippocampal synaptosomes Sm proteins [189]. Meanwhile, the SMNΔ7 fragment is from Burghes severe SMA mice showed decreased levels polyubiquitinated and quickly degraded [186]. Pharmaco- of UBA1 compared to controls, with decreased expression logical inhibition of ubiquitination of SMN, such as with also reported in spinal cord and skeletal muscle [196]. The the small molecule ML372, increased SMN protein levels Taiwanese SMA mouse model similarly showed tissue-wide and slowed disease progression of SMNΔ7 mice leading to lower levels of UBA1, along with changes in splicing of the longer survival, increased motor neuron size and less mus- UBA1 transcript, which may account (at least in part) for cle atrophy [190]. When SMA-patient-derived fibroblasts the altered protein expression. Experimental suppression of were treated with salbutamol, the β -adrenergic recep- UBA1 in wild-type zebrafish was sufficient to phenocopy tor agonist, there was also an increase in levels of SMN SMA-like motor axon defects. Likewise, in the zebrafish protein, possibly acting via activation of protein kinase SMA model UBA1 expression was reduced by 70%, whilst A, thereby preventing SMN ubiquitination [191]. SMA increasing UBA1 expression rescued the SMN-knockdown SMN E3 Ligases Protein Substrate E.g. SMN E2 Ligases E3 Ligases ATP AMP Protein Substrate E3 Ligases E.g. SMN Protein E2 Ligases Intracellular E2 Ligas UBA1 UBA1 B1 Substrate Ub processes Ub Ub E.g. S N SMN Ub Ub Protein Substrate Proteasome E.g. SMN or or degradation Protein SMN 7 Substrate Ub E.g. SMN Ub Deubiquitinating Ub Enzymes Ub SMN Ub Ub Ub Ub Fig. 2 Diagrammatic representation of the ubiquitin pathway and the 1, Itch and TRAF6. Ubiquitin is then transferred to the protein sub- components, where SMN interacts. SMN is both ubiquitinated via the strate and the complex dissociates. Monoubiquitinated substrates con- UPS pathway and an interacting protein influencing several steps of tinue on to other intracellular processes, whereas polyubiquitinated the process. SMN directly interacts with the UBA1 enzyme, which substrates are targeted for proteasome degradation. SMN has also transfers ubiquitin to the E2 ligases. Ubiquitinated E2 ligases then been shown to interact with deubiquitinating enzymes, which remove form a complex with E3 ligases bound to protein substrates. SMN has ubiquitin from protein substrates been shown to interact with several E3 ligases, including Mindbomb 1 3 H. Chaytow et al. phenotype [197]. Finally, treating Taiwanese mice with homeostasis in which SMN has been shown to interact: its an AAV9-UBA1 expression vector improved the survival, well-known role as part of the ribonucleoprotein complex, weight gain and motor performance of the mice as well as but also other stages of RNA processing such as transport rescuing motor neuron cell number in the spinal cord and and local translation, important neuronal functions such as neuromuscular junction pathology [197]. cytoskeletal dynamics and endocytosis, protein turnover As well as UBA1, SMN is known to interact with sev- processes of autophagy and ubiquitin–proteasome pathway eral other ubiquitin-associated enzymes (Fig.  2). Several and regulation of mitochondrial activity. Through tradition E3 ubiquitin ligases have been shown to interact with SMN and necessity, the majority of current research into the func- and so may be involved in its degradation through recruit- tion of SMN comes from SMA models of SMN deficiency. ment into the UPS. For example, Mindbomb 1 directly However, as this review has highlighted, SMN function is interacts with SMN [190]. Overexpression of Mindbomb involved in so many aspects of normal intracellular activity 1 was shown to increase the amount of ubiquitinated SMN that future SMN research should move beyond its associa- protein in cell culture, while a knockdown of Mindbomb 1 in tion with disease to better understand its role in maintain- the C. elegans model of SMA improves the SMN-deficient ing the homeostatic environment of the cell. Two major phenotype of defects in pharyngeal pumping [198]. Other questions need answering in terms of the function of SMN. E3 ligases known to interact with SMN include Itch [189], First, to what extent is SMN involved in the regulation of TRAF6 [199] and the Drosophila E3 ligase SCFslmb [9]. processes discussed in this review. While some areas have Monoubiquitination following interaction with Itch was been researched extensively, such as ribonucleoprotein pro- shown to regulate SMN’s intracellular localisation [189]. duction, other areas of SMN involvement are a relatively TRAF6 activity is apparently inhibited by SMN, and so new discovery, such as the association of SMN with mito- SMN binding may be involved in the activation of NF-κB chondrial function and ubiquitin degradation, and so further signalling further downstream [199]. exploration is needed. Secondly, the particular vulnerabil- Ubiquitin carboxy-terminal hydrolase L1 (UCHL1) is a ity of motor neurons in SMA patients cannot be ignored. deubiquitinating enzyme specifically expressed in neuronal Although the idea that SMA is in fact a systemic disease, tissue, and its downregulation has been associated with with defects seen across tissue types, is gaining acceptance Parkinson’s and Alzheimer’s diseases [200–202]. Follow- in the research community, a better understanding of the ing knockdown of UCHL1 in cell culture, there was a con- multiplicity of SMN functions could serve to highlight areas cordant increase in SMN expression [203]. Conversely, in of particular susceptibility in motor neurons which lead to Taiwanese mice, there was an increase in UCHL1 expres- their cell death in SMA. As SMN is at the cornerstone of so sion. However, inhibition of UCHL1 expression in Taiwan- many molecular pathways, fundamental research into these ese mice failed to increase SMN levels and did not have an cellular homeostasis processes is crucial to the better under- effect on survival or phenotype of the SMA model, with standing of cellular biology. evidence suggesting that an increase in UCHL1 levels in the Acknowledgements We would like to thank Dr. Ewout Groen for help- absence of SMN may be a compensatory response to restore ful comments on the manuscript, and all the members of the lab for levels of ubiquitination [204]. Usp9x, another deubiquitinat- ongoing discussions and advice. Research in the Gillingwater labora- ing enzyme known to interact with SMN, also influences its tory relevant to this review is funded by the UK SMA Research Con- ubiquitination levels, where a loss of Usp9x impairs SMN sortium (SMA Trust), MND Scotland and SMA Europe. nuclear gem formation while overexpression leads to an increase in ubiquitinated SMN [186]. It, therefore, appears Compliance with ethical standards that SMN is regulated at several levels of the UPS, which Conflict of interest THG is Chair of the Scientific and Clinical Advi- may have an effect on cell-wide ubiquitination as well as sory Board of the SMA Trust. regulation of the SMN protein itself. Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creat iveco Concluding remarks and future perspectivesmmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the SMN, originally discovered due to its association with the Creative Commons license, and indicate if changes were made. neurodegenerative disorder spinal muscular atrophy, is in fact a ubiquitous protein with numerous roles within the cell. Although its first-identified and most-described function is in the biogenesis of ribonucleoproteins, it is now evident that SMN plays a more general housekeeping role. With this in mind, here we have discussed various areas of intracellular 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis 19. Meister G, Fischer U (2002) Assisted RNP assembly: SMN and References PRMT5 complexes cooperate in the formation of spliceosomal UsnRNPs. EMBO J 21:5853–5863 1. Lefebvre S et al (1995) Identification and characterization of a 20. Paushkin S, Gubitz AK, Massenet S, Dreyfuss G (2002) The spinal muscular atrophy-determining gene. Cell 80:155–165 SMN complex, an assemblyosome of ribonucleoproteins. Curr 2. Schrank B, Götz R, Gunnersen JM, Ure JM, Toyka KV, Smith Opin Cell Biol 14:305–312 AG, Sendtner M (1997) Inactivation of the survival motor neu- 21. Hebert MD, Shpargel KB, Ospina JK, Tucker KE, Matera AG ron gene, a candidate gene for human spinal muscular atrophy, (2002) Coilin methylation regulates nuclear body formation. Dev leads to massive cell death in early mouse embryos. Proc Natl Cell 3:329–337 Acad Sci USA 94:9920–9925 22. Sprangers R, Groves MR, Sinning I, Sattler M (2003) High-res- 3. Groen EJN, Talbot K, Gillingwater TH (2018) Advances in olution X-ray and NMR structures of the SMN Tudor domain: therapy for spinal muscular atrophy: promises and challenges. conformational variation in the binding site for symmetrically Nat Rev Neurol 14:214–224 dimethylated arginine residues. J Mol Biol 327:507–520 4. Rochette CF, Gilbert N, Simard LR (2001) SMN gene dupli- 23. Côté J, Richard S (2005) Tudor domains bind symmetrical cation and the emergence of the SMN2 gene occurred in dis- dimethylated arginines. J Biol Chem 280:28476–28483 tinct hominids: SMN2 is unique to Homo sapiens. Hum Genet 24. Meister G, Eggert C, Fischer U (2002) SMN-mediated assem- 108:255–266 bly of RNPs: a complex story. Trends Cell Biol 12:472–478 5. Lorson CL, Hahnen E, Androphy EJ, Wirth B (1999) A single 25. Bühler D, Raker V, Lührmann R, Fischer U (1999) Essential nucleotide in the SMN gene regulates splicing and is respon- role for the tudor domain of SMN in spliceosomal U snRNP sible for spinal muscular atrophy. Proc Natl Acad Sci USA assembly: implications for spinal muscular atrophy. Hum Mol 96:6307–6311 Genet 8:2351–2357 6. Feldkötter M, Schwarzer V, Wirth R, Wienker TF, Wirth B 26. Kotani T et al (2007) A novel mutation at the N-terminal of (2002) Quantitative analyses of SMN1 and SMN2 based on real- SMN Tudor domain inhibits its interaction with target proteins. time lightCycler PCR: fast and highly reliable carrier testing and J Neurol 254:624–630 prediction of severity of spinal muscular atrophy. Am J Hum 27. Cuscó I, Barceló MJ, del Río E, Baiget M, Tizzano EF (2004) Genet 70:358–368 Detection of novel mutations in the SMN Tudor domain in type 7. Talbot K, Tizzano EF (2017) The clinical landscape for SMA in I SMA patients. Neurology 63:146–149 a new therapeutic era. Gene Ther 24:529–533 28. Hebert MD, Szymczyk PW, Shpargel KB, Matera AG (2001) 8. Vitte J et al (2007) Refined characterization of the expression and Coilin forms the bridge between Cajal bodies and SMN, the stability of the SMN gene products. Am J Pathol 171:1269–1280 spinal muscular atrophy protein. Genes Dev 15:2720–2729 9. Gray KM et al (2018) Self-oligomerization regulates stability 29. Liu Q, Dreyfuss G (1996) A novel nuclear structure containing of survival motor neuron protein isoforms by sequestering an the survival of motor neurons protein. EMBO J 15:3555–3565 SCF(Slmb) degron. Mol Biol Cell 29:96–110 30. Navascues J, Berciano MT, Tucker KE, Lafarga M, Matera 10. Seo J, Singh NN, Ottesen EW, Lee BM, Singh RN (2016) A AG (2004) Targeting SMN to Cajal bodies and nuclear gems novel human-specific splice isoform alters the critical C-terminus during neuritogenesis. Chromosoma 112:398–409 of Survival Motor Neuron protein. Sci Rep 6:30778 31. Machyna M, Heyn P, Neugebauer KM (2013) Cajal bod- 11. Gennarelli M, Lucarelli M, Capon F, Pizzuti A, Merlini L, Ange- ies: where form meets function. Wiley Interdiscip Rev RNA lini C, Novelli G, Dallapiccola B (1995) Survival motor neuron 4:17–34 gene transcript analysis in muscles from spinal muscular atrophy 32. Tapia O, Bengoechea R, Palanca A, Arteaga R, Val-Bernal patients. Biochem Biophys Res Commun 213:342–348 JF, Tizzano EF, Berciano MT, Lafarga M (2012) Reorganiza- 12. Setola V, Terao M, Locatelli D, Bassanini S, Garattini E, Batt- tion of Cajal bodies and nucleolar targeting of coilin in motor aglia G (2007) Axonal-SMN (a-SMN), a protein isoform of the neurons of type I spinal muscular atrophy. Histochem Cell Biol survival motor neuron gene, is specifically involved in axonogen- 137:657–667 esis. Proc Natl Acad Sci USA 104:1959–1964 33. Tapia O, Lafarga V, Bengoechea R, Palanca A, Lafarga M, Ber- 13. Young PJ, Jensen KT, Burger LR, Pintel DJ, Lorson CL (2002) ciano MT (2014) The SMN Tudor SIM-like domain is key to Minute virus of mice NS1 interacts with the SMN protein, and SmD1 and coilin interactions and to Cajal body biogenesis. J they colocalize in novel nuclear bodies induced by parvovirus Cell Sci 127:939–946 infection. J Virol 76:3892–3904 34. Giesemann T, Rathke-Hartlieb S, Rothkegel M, Bartsch JW, 14. Liu Q, Fischer U, Wang F, Dreyfuss G (1997) The spinal muscu- Buchmeier S, Jockusch BM, Jockusch H (1999) A role for poly- lar atrophy disease gene product, SMN, and its associated protein proline motifs in the spinal muscular atrophy protein SMN. Pro- SIP1 are in a complex with spliceosomal snRNP proteins. Cell filins bind to and colocalize with smn in nuclear gems. J Biol 90:1013–1021 Chem 274:37908–37914 15. Lorson CL, Androphy EJ (1998) The domain encoded by exon 2 35. Nölle A et al (2011) The spinal muscular atrophy disease protein of the survival motor neuron protein mediates nucleic acid bind- SMN is linked to the Rho-kinase pathway via profilin. Hum Mol ing. Hum Mol Genet 7:1269–1275 Genet 20:4865–4878 16. Bertrandy S, Burlet P, Clermont O, Huber C, Fondrat C, Thierry- 36. Martin R, Gupta K, Ninan NS, Perry K, Van Duyne GD (2012) Mieg D, Munnich A, Lefebvre S (1999) The RNA-binding The survival motor neuron protein forms soluble glycine zipper properties of SMN: deletion analysis of the zebrafish ortho- oligomers. Structure 20:1929–1939 logue defines domains conserved in evolution. Hum Mol Genet 37. Burghes AH, Beattie CE (2009) Spinal muscular atrophy: why do 8:775–782 low levels of survival motor neuron protein make motor neurons 17. Ponting CP (1997) Tudor domains in proteins that interact with sick? Nat Rev Neurosci 10:597–609 RNA. Trends Biochem Sci 22:51–52 38. Wang J, Dreyfuss G (2001) A cell system with targeted dis- 18. Selenko P, Sprangers R, Stier G, Bühler D, Fischer U, Sattler M ruption of the SMN gene: functional conservation of the SMN (2001) SMN tudor domain structure and its interaction with the protein and dependence of Gemin2 on SMN. J Biol Chem Sm proteins. Nat Struct Biol 8:27–31 276:9599–9605 1 3 H. Chaytow et al. 39. Lorson CL et al (1998) SMN oligomerization defect correlates 60. Carissimi C, Saieva L, Baccon J, Chiarella P, Maiolica A, Sawyer with spinal muscular atrophy severity. Nat Genet 19:63–66 A, Rappsilber J, Pellizzoni L (2006) Gemin8 is a novel compo- 40. Lafarga V, Tapia O, Sharma S, Bengoechea R, Stoecklin G, nent of the survival motor neuron complex and functions in small Lafarga M, Berciano MT (2018) CBP-mediated SMN acetylation nuclear ribonucleoprotein assembly. J Biol Chem 281:8126–8134 modulates Cajal body biogenesis and the cytoplasmic targeting 61. Charroux B, Pellizzoni L, Perkinson RA, Shevchenko A, Mann of SMN. Cell Mol Life Sci 75:527–546 M, Dreyfuss G (1999) Gemin3: a novel DEAD box protein that 41. Husedzinovic A, Oppermann F, Draeger-Meurer S, Chari A, interacts with SMN, the spinal muscular atrophy gene product, Fischer U, Daub H, Gruss OJ (2014) Phosphoregulation of the and is a component of gems. J Cell Biol 147:1181–1194 human SMN complex. Eur J Cell Biol 93:106–117 62. Gubitz AK, Feng W, Dreyfuss G (2004) The SMN complex. Exp 42. Hebert MD, Poole AR (2017) Towards an understanding of regu- Cell Res 296:51–56 lating Cajal body activity by protein modification. RNA Biol 63. Gubitz AK, Mourelatos Z, Abel L, Rappsilber J, Mann M, 14:761–778 Dreyfuss G (2002) Gemin5, a novel WD repeat protein compo- 43. Hsieh-Li HM, Chang JG, Jong YJ, Wu MH, Wang NM, Tsai CH, nent of the SMN complex that binds Sm proteins. J Biol Chem Li H (2000) A mouse model for spinal muscular atrophy. Nat 277:5631–5636 Genet 24:66–70 64. Li DK, Tisdale S, Lotti F, Pellizzoni L (2014) SMN control of 44. Monani UR et al (2000) The human centromeric survival motor RNP assembly: from post-transcriptional gene regulation to (−/−) neuron gene (SMN2) rescues embryonic lethality in Smn motor neuron disease. Semin Cell Dev Biol 32:22–29 mice and results in a mouse with spinal muscular atrophy. Hum 65. Pellizzoni L (2007) Chaperoning ribonucleoprotein biogenesis Mol Genet 9:333–339 in health and disease. EMBO Rep 8:340–345 45. Le TT et al (2005) SMNDelta7, the major product of the cen- 66. Pellizzoni L, Baccon J, Rappsilber J, Mann M, Dreyfuss G tromeric survival motor neuron (SMN2) gene, extends survival (2002) Purification of native survival of motor neurons com- in mice with spinal muscular atrophy and associates with full- plexes and identification of Gemin6 as a novel component. J length SMN. Hum Mol Genet 14:845–857 Biol Chem 277:7540–7545 46. Bebee TW, Dominguez CE, Chandler DS (2012) Mouse models 67. Battle DJ, Lau CK, Wan L, Deng H, Lotti F, Dreyfuss G (2006) of SMA: tools for disease characterization and therapeutic devel- The Gemin5 protein of the SMN complex identifies snRNAs. opment. Hum Genet 131:1277–1293 Mol Cell 23:273–279 47. McWhorter ML, Monani UR, Burghes AH, Beattie CE (2003) 68. Pellizzoni L, Yong J, Dreyfuss G (2002) Essential role for the Knockdown of the survival motor neuron (Smn) protein in SMN complex in the specificity of snRNP assembly. Science zebrafish causes defects in motor axon outgrowth and pathfind- 298:1775–1779 ing. J Cell Biol 162:919–931 69. Yong J, Golembe TJ, Battle DJ, Pellizzoni L, Dreyfuss G 48. Ciura S, Lattante S, Le Ber I, Latouche M, Tostivint H, Brice (2004) snRNAs contain specific SMN-binding domains that A, Kabashi E (2013) Loss of function of C9orf72 causes motor are essential for snRNP assembly. Mol Cell Biol 24:2747–2756 deficits in a zebrafish model of amyotrophic lateral sclerosis. Ann 70. Yong J, Kasim M, Bachorik JL, Wan L, Dreyfuss G (2010) Neurol 74:180–187 Gemin5 delivers snRNA precursors to the SMN complex for 49. McLean DL, Fetcho JR (2011) Movement, technology and dis- snRNP biogenesis. Mol Cell 38:551–562 covery in the zebrafish. Curr Opin Neurobiol 21:110–115 71. Yong J, Pellizzoni L, Dreyfuss G (2002) Sequence-specific 50. Babin PJ, Goizet C, Raldúa D (2014) Zebrafish models of human interaction of U1 snRNA with the SMN complex. EMBO J motor neuron diseases: advantages and limitations. Prog Neuro- 21:1188–1196 biol 118:36–58 72. Fischer U, Luhrmann R (1990) An essential signaling role for 51. Edens BM, Ajroud-Driss S, Ma L, Ma YC (2015) Molecular the m3G cap in the transport of U1 snRNP to the nucleus. Sci- mechanisms and animal models of spinal muscular atrophy. Bio- ence 249:786–790 chim Biophys Acta 1852:685–692 73. Mouaikel J, Narayanan U, Verheggen C, Matera AG, Bertrand 52. Frattini E et al (2015) Pluripotent stem cell-based models of spi- E, Tazi J, Bordonne R (2003) Interaction between the small- nal muscular atrophy. Mol Cell Neurosci 64:44–50 nuclear-RNA cap hypermethylase and the spinal muscular atro- 53. Sleigh JN, Gillingwater TH, Talbot K (2011) The contribution phy protein, survival of motor neuron. EMBO Rep 4:616–622 of mouse models to understanding the pathogenesis of spinal 74. Narayanan U, Ospina JK, Frey MR, Hebert MD, Matera AG muscular atrophy. Dis Model Mech 4:457–467 (2002) SMN, the spinal muscular atrophy protein, forms a pre- 54. Van Alstyne M, Pellizzoni L (2016) Advances in modeling and import snRNP complex with snurportin1 and importin beta. treating spinal muscular atrophy. Curr Opin Neurol 29:549–556 Hum Mol Genet 11:1785–1795 55. Rodriguez-Muela N et al (2017) Single-cell analysis of SMN 75. Fischer U, Sumpter V, Sekine M, Satoh T, Luhrmann R (1993) reveals its broader role in neuromuscular disease. Cell Rep Nucleo-cytoplasmic transport of U snRNPs: definition of a 18:1484–1498 nuclear location signal in the Sm core domain that binds a 56. Matera AG, Wang Z (2014) A day in the life of the spliceosome. transport receptor independently of the m3G cap. EMBO J Nat Rev Mol Cell Biol 15:108–121 12:573–583 57. Raker VA, Hartmuth K, Kastner B, Luhrmann R (1999) Spliceo- 76. Huber J, Cronshagen U, Kadokura M, Marshallsay C, Wada somal U snRNP core assembly: Sm proteins assemble onto an T, Sekine M, Luhrmann R (1998) Snurportin1, an m3G-cap- Sm site RNA nonanucleotide in a specific and thermodynami- specific nuclear import receptor with a novel domain structure. cally stable manner. Mol Cell Biol 19:6554–6565 EMBO J 17:4114–4126 58. Baccon J, Pellizzoni L, Rappsilber J, Mann M, Dreyfuss G (2002) 77. Palacios I, Hetzer M, Adam SA, Mattaj IW (1997) Nuclear Identification and characterization of Gemin7, a novel compo- import of U snRNPs requires importin beta. EMBO J nent of the survival of motor neuron complex. J Biol Chem 16:6783–6792 277:31957–31962 78. Mahmoudi S, Henriksson S, Weibrecht I, Smith S, Soderberg 59. Carissimi C, Baccon J, Straccia M, Chiarella P, Maiolica A, O, Stromblad S, Wiman KG, Farnebo M (2010) WRAP53 is Sawyer A, Rappsilber J, Pellizzoni L (2005) Unrip is a compo- essential for Cajal body formation and for targeting the survival nent of SMN complexes active in snRNP assembly. FEBS Lett of motor neuron complex to Cajal bodies. PLoS Biol 8:e1000521 579:2348–2354 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis 79. Narayanan U, Achsel T, Luhrmann R, Matera AG (2004) Cou- 98. See K et al (2014) SMN deficiency alters Nrxn2 expression and pled in vitro import of U snRNPs and SMN, the spinal muscular splicing in zebrafish and mouse models of spinal muscular atro- atrophy protein. Mol Cell 16:223–234 phy. Hum Mol Genet 23:1754–1770 80. Henriksson S, Farnebo M (2015) On the road with WRAP53beta: 99. Sleigh JN, Barreiro-Iglesias A, Oliver PL, Biba A, Becker T, guardian of Cajal bodies and genome integrity. Front Genet 6:91 Davies KE, Becker CG, Talbot K (2014) Chondrolectin affects 81. Gabanella F, Butchbach ME, Saieva L, Carissimi C, Burghes AH, cell survival and neuronal outgrowth in in vitro and in vivo mod- Pellizzoni L (2007) Ribonucleoprotein assembly defects correlate els of spinal muscular atrophy. Hum Mol Genet 23:855–869 with spinal muscular atrophy severity and preferentially affect a 100. Zhang Z et al (2013) Dysregulation of synaptogenesis genes ante- subset of spliceosomal snRNPs. PLoS One 2:e921 cedes motor neuron pathology in spinal muscular atrophy. Proc 82. Wan L, Battle DJ, Yong J, Gubitz AK, Kolb SJ, Wang J, Dreyfuss Natl Acad Sci USA 110:19348–19353 G (2005) The survival of motor neurons protein determines the 101. Baumer D et al (2009) Alternative splicing events are a late fea- capacity for snRNP assembly: biochemical deficiency in spinal ture of pathology in a mouse model of spinal muscular atrophy. muscular atrophy. Mol Cell Biol 25:5543–5551 PLoS Genet 5:e1000773 83. Zhang Z, Lotti F, Dittmar K, Younis I, Wan L, Kasim M, Drey- 102. Pagliardini S, Giavazzi A, Setola V, Lizier C, Di Luca M, DeBi- fuss G (2008) SMN deficiency causes tissue-specific perturba - asi S, Battaglia G (2000) Subcellular localization and axonal tions in the repertoire of snRNAs and widespread defects in transport of the survival motor neuron (SMN) protein in the splicing. Cell 133:585–600 developing rat spinal cord. Hum Mol Genet 9:47–56 84. So BR, Wan L, Zhang Z, Li P, Babiash E, Duan J, Younis I, 103. Giavazzi A, Setola V, Simonati A, Battaglia G (2006) Neuronal- Dreyfuss G (2016) A U1 snRNP-specific assembly pathway specific roles of the survival motor neuron protein: evidence reveals the SMN complex as a versatile hub for RNP exchange. from survival motor neuron expression patterns in the develop- Nat Struct Mol Biol 23:225–230 ing human central nervous system. J Neuropathol Exp Neurol 85. Marzluff WF, Wagner EJ, Duronio RJ (2008) Metabolism and 65:267–277 regulation of canonical histone mRNAs: life without a poly(A) 104. Zhang HL, Pan F, Hong D, Shenoy SM, Singer RH, Bassell GJ tail. Nat Rev Genet 9:843–854 (2003) Active transport of the survival motor neuron protein 86. Pillai RS, Grimmler M, Meister G, Will CL, Luhrmann R, Fis- and the role of exon-7 in cytoplasmic localization. J Neurosci cher U, Schumperli D (2003) Unique Sm core structure of U7 23:6627–6637 snRNPs: assembly by a specialized SMN complex and the role 105. Zhang H, Xing L, Rossoll W, Wichterle H, Singer RH, Bassell of a new component, Lsm11, in histone RNA processing. Genes GJ (2006) Multiprotein complexes of the survival of motor neu- Dev 17:2321–2333 ron protein SMN with Gemins traffic to neuronal processes and 87. Pillai RS, Will CL, Luhrmann R, Schumperli D, Muller B growth cones of motor neurons. J Neurosci 26:8622–8632 (2001) Purified U7 snRNPs lack the Sm proteins D1 and D2 106. Thompson LW, Morrison KD, Shirran SL, Groen EJN, Gill- but contain Lsm10, a new 14 kDa Sm D1-like protein. EMBO J ingwater TH, Botting CH, Sleeman JE (2018) Neurochondrin 20:5470–5479 interacts with the SMN protein suggesting a novel mechanism 88. Schumperli D, Pillai RS (2004) The special Sm core structure for spinal muscular atrophy pathology. J Cell Sci. https ://doi. of the U7 snRNP: far-reaching significance of a small nuclear org/10.1242/jcs.21148 2 ribonucleoprotein. Cell Mol Life Sci 61:2560–2570 107. Peter CJ et al (2011) The COPI vesicle complex binds and moves 89. Reichow SL, Hamma T, Ferre-D’Amare AR, Varani G (2007) with survival motor neuron within axons. Hum Mol Genet The structure and function of small nucleolar ribonucleoproteins. 20:1701–1711 Nucleic Acids Res 35:1452–1464 108. Todd AG, Lin H, Ebert AD, Liu Y, Androphy EJ (2013) COPI 90. Pellizzoni L, Baccon J, Charroux B, Dreyfuss G (2001) The sur- transport complexes bind to specific RNAs in neuronal cells. vival of motor neurons (SMN) protein interacts with the snoRNP Hum Mol Genet 22:729–736 proteins fibrillarin and GAR1. Curr Biol 11:1079–1088 109. Ting CH, Wen HL, Liu HC, Hsieh-Li HM, Li H, Lin-Chao S 91. Renvoise B, Colasse S, Burlet P, Viollet L, Meier UT, Lefebvre (2012) The spinal muscular atrophy disease protein SMN is S (2009) The loss of the snoRNP chaperone Nopp140 from Cajal linked to the Golgi network. PLoS One 7:e51826 bodies of patient b fi roblasts correlates with the severity of spinal 110. Li H, Custer SK, Gilson T, Hao LT, Beattie CE, Androphy EJ muscular atrophy. Hum Mol Genet 18:1181–1189 (2015) α-COP binding to the survival motor neuron protein SMN 92. Zhang Q, Kim NK, Feigon J (2011) Architecture of human tel- is required for neuronal process outgrowth. Hum Mol Genet omerase RNA. Proc Natl Acad Sci USA 108:20325–20332 24:7295–7307 93. Bachand F, Boisvert FM, Cote J, Richard S, Autexier C 111. Custer SK, Todd AG, Singh NN, Androphy EJ (2013) Dily- (2002) The product of the survival of motor neuron (SMN) sine motifs in exon 2b of SMN protein mediate binding to the gene is a human telomerase-associated protein. Mol Biol Cell COPI vesicle protein α-COP and neurite outgrowth in a cell 13:3192–3202 culture model of spinal muscular atrophy. Hum Mol Genet 94. Poole AR, Hebert MD (2016) SMN and coilin negatively regulate 22:4043–4052 dyskerin association with telomerase RNA. Biol Open 5:726–735 112. Rossoll W, Jablonka S, Andreassi C, Kroning AK, Karle K, 95. Powis RA, Gillingwater TH (2016) Selective loss of alpha motor Monani UR, Sendtner M (2003) Smn, the spinal muscular neurons with sparing of gamma motor neurons and spinal cord atrophy-determining gene product, modulates axon growth and cholinergic neurons in a mouse model of spinal muscular atro- localization of beta-actin mRNA in growth cones of motoneu- phy. J Anat 228:443–451 rons. J Cell Biol 163:801–812 96. Winkler C, Eggert C, Gradl D, Meister G, Giegerich M, Wedlich 113. Fallini C, Zhang H, Su Y, Silani V, Singer RH, Rossoll W, Bassell D, Laggerbauer B, Fischer U (2005) Reduced U snRNP assembly GJ (2011) The survival of motor neuron (SMN) protein interacts causes motor axon degeneration in an animal model for spinal with the mRNA-binding protein HuD and regulates localization muscular atrophy. Genes Dev 19:2320–2330 of poly(A) mRNA in primary motor neuron axons. J Neurosci 97. Lotti F et  al (2012) An SMN-dependent U12 splicing event 31:3914–3925 essential for motor circuit function. Cell 151:440–454 114. Tadesse H, Deschenes-Furry J, Boisvenue S, Cote J (2008) KH- type splicing regulatory protein interacts with survival motor 1 3 H. Chaytow et al. neuron protein and is misregulated in spinal muscular atrophy. 134. Leterrier C, Dubey P, Roy S (2017) The nano-architecture of the Hum Mol Genet 17:506–524 axonal cytoskeleton. Nat Rev Neurosci 18:713–726 115. Piazzon N, Rage F, Schlotter F, Moine H, Branlant C, Mass- 135. Kevenaar JT, Hoogenraad CC (2015) The axonal cytoskeleton: enet S (2008) In vitro and in cellulo evidences for association of from organization to function. Front Mol Neurosci 8:44 the survival of motor neuron complex with the fragile X mental 136. Béchade C, Rostaing P, Cisterni C, Kalisch R, La Bella V, Pett- retardation protein. J Biol Chem 283:5598–5610 mann B, Triller A (1999) Subcellular distribution of survival 116. Hubers L, Valderrama-Carvajal H, Laframboise J, Timbers J, motor neuron (SMN) protein: possible involvement in nucleo- Sanchez G, Cote J (2011) HuD interacts with survival motor neu- cytoplasmic and dendritic transport. Eur J Neurosci 11:293–304 ron protein and can rescue spinal muscular atrophy-like neuronal 137. Francis JW, Sandrock AW, Bhide PG, Vonsattel JP, Brown RH defects. Hum Mol Genet 20:553–579 (1998) Heterogeneity of subcellular localization and electro- 117. Groen EJ et al (2013) ALS-associated mutations in FUS disrupt phoretic mobility of survival motor neuron (SMN) protein in the axonal distribution and function of SMN. Hum Mol Genet mammalian neural cells and tissues. Proc Natl Acad Sci USA 22:3690–3704 95:6492–6497 118. Fallini C, Rouanet JP, Donlin-Asp PG, Guo P, Zhang H, Singer 138. Fan L, Simard LR (2002) Survival motor neuron (SMN) protein: RH, Rossoll W, Bassell GJ (2014) Dynamics of survival of motor role in neurite outgrowth and neuromuscular maturation dur- neuron (SMN) protein interaction with the mRNA-binding pro- ing neuronal differentiation and development. Hum Mol Genet tein IMP1 facilitates its trafficking into motor neuron axons. Dev 11:1605–1614 Neurobiol 74:319–332 139. van Bergeijk J, Rydel-Könecke K, Grothe C, Claus P (2007) 119. Yamazaki T et al (2012) FUS-SMN protein interactions link the The spinal muscular atrophy gene product regulates neu- motor neuron diseases ALS and SMA. Cell Rep 2:799–806 rite outgrowth: importance of the C terminus. FASEB J 120. Tsuiji H et  al (2013) Spliceosome integrity is defective in 21:1492–1502 the motor neuron diseases ALS and SMA. EMBO Mol Med 140. Locatelli D et al (2012) Human axonal survival of motor neuron 5:221–234 (a-SMN) protein stimulates axon growth, cell motility, C-C motif 121. Donlin-Asp PG, Fallini C, Campos J, Chou CC, Merritt ME, ligand 2 (CCL2), and insulin-like growth factor-1 (IGF1) produc- Phan HC, Bassell GJ, Rossoll W (2017) The survival of motor tion. J Biol Chem 287:25782–25794 neuron protein acts as a molecular chaperone for mRNP assem- 141. Sharma A, Lambrechts A, Hao LT, Le TT, Sewry CA, Ampe C, bly. Cell Rep 18:1660–1673 Burghes AH, Morris GE (2005) A role for complexes of survival 122. le Hao T, Duy PQ, An M, Talbot J, Iyer CC, Wolman M, Beattie of motor neurons (SMN) protein with gemins and profilin in CE (2017) HuD and the survival motor neuron protein interact neurite-like cytoplasmic extensions of cultured nerve cells. Exp in motoneurons and are essential for motoneuron development, Cell Res 309:185–197 function, and mRNA regulation. J Neurosci 37:11559–11571 142. Schmandke A, Strittmatter SM (2007) ROCK and Rho: biochem- 123. Rage F, Boulisfane N, Rihan K, Neel H, Gostan T, Bertrand istry and neuronal functions of Rho-associated protein kinases. E, Bordonne R, Soret J (2013) Genome-wide identification Neuroscientist 13:454–469 of mRNAs associated with the protein SMN whose depletion 143. Bowerman M, Shafey D, Kothary R (2007) Smn depletion alters decreases their axonal localization. RNA 19:1755–1766 profilin II expression and leads to upregulation of the RhoA/ 124. Akten B et al (2011) Interaction of survival of motor neuron ROCK pathway and defects in neuronal integrity. J Mol Neurosci (SMN) and HuD proteins with mRNA cpg15 rescues motor neu- 32:120–131 ron axonal deficits. Proc Natl Acad Sci USA 108:10337–10342 144. Bowerman M, Beauvais A, Anderson CL, Kothary R (2010) 125. Glock C, Heumüller M, Schuman EM (2017) mRNA trans- Rho-kinase inactivation prolongs survival of an intermediate port and local translation in neurons. Curr Opin Neurobiol SMA mouse model. Hum Mol Genet 19:1468–1478 45:169–177 145. Rademacher S et al (2017) Metalloprotease-mediated cleavage 126. Donlin-Asp PG, Rossoll W, Bassell GJ (2017) Spatially and of PlexinD1 and its sequestration to actin rods in the motoneu- temporally regulating translation via mRNA-binding proteins in ron disease spinal muscular atrophy (SMA). Hum Mol Genet cellular and neuronal function. FEBS Lett 591:1508–1525 26:3946–3959 127. le Hao T, Wolman M, Granato M, Beattie CE (2012) Survival 146. Oprea GE et al (2008) Plastin 3 is a protective modifier of auto- motor neuron affects plastin 3 protein levels leading to motor somal recessive spinal muscular atrophy. Science 320:524–527 defects. J Neurosci 32:5074–5084 147. Kaifer KA, Villalón E, Osman EY, Glascock JJ, Arnold LL, Cor- 128. Fallini C, Donlin-Asp PG, Rouanet JP, Bassell GJ, Rossoll nelison DD, Lorson CL (2017) Plastin-3 extends survival and W (2016) Deficiency of the survival of motor neuron protein reduces severity in mouse models of spinal muscular atrophy. impairs mRNA localization and local translation in the growth JCI Insight 2:e89970 cone of motor neurons. J Neurosci 36:3811–3820 148. Prokop A (2013) The intricate relationship between microtu- 129. Sanchez G et al (2013) A novel function for the survival moto- bules and their associated motor proteins during axon growth neuron protein as a translational regulator. Hum Mol Genet and maintenance. Neural Dev 8:17 22:668–684 149. Belmont LD, Mitchison TJ (1996) Identification of a protein that 130. Weeraratne SD et  al (2012) Pleiotropic effects of miR- interacts with tubulin dimers and increases the catastrophe rate 183~96~182 converge to regulate cell survival, proliferation and of microtubules. Cell 84:623–631 migration in medulloblastoma. Acta Neuropathol 123:539–552 150. Wen HL, Lin YT, Ting CH, Lin-Chao S, Li H, Hsieh-Li HM 131. Kye MJ et al (2014) SMN regulates axonal local translation via (2010) Stathmin, a microtubule-destabilizing protein, is dysregu- miR-183/mTOR pathway. Hum Mol Genet 23:6318–6331 lated in spinal muscular atrophy. Hum Mol Genet 19:1766–1778 132. Ning K et al (2010) PTEN depletion rescues axonal growth defect 151. Deinhardt K, Schiavo G (2005) Endocytosis and retrograde and improves survival in SMN-deficient motor neurons. Hum axonal traffic in motor neurons. Biochem Soc Symp 72:139–150 Mol Genet 19:3159–3168 152. Dimitriadi M et  al (2010) Conserved genes act as modifiers 133. Bernabo P et al (2017) In vivo translatome profiling in spinal of invertebrate SMN loss of function defects. PLoS Genet muscular atrophy reveals a role for SMN protein in ribosome 6:e1001172 biology. Cell Rep 21:953–965 1 3 The role of survival motor neuron protein (SMN) in protein homeostasis 153. Dimitriadi M et al (2016) Decreased function of survival motor 172. Kamada Y, Funakoshi T, Shintani T, Nagano K, Ohsumi M, neuron protein impairs endocytic pathways. Proc Natl Acad Sci Ohsumi Y (2000) Tor-mediated induction of autophagy via an USA 113:E4377–E4386 Apg1 protein kinase complex. J Cell Biol 150:1507–1513 154. Dachs E, Hereu M, Piedrafita L, Casanovas A, Caldero J, 173. Wu L et al (2013) Rapamycin upregulates autophagy by inhib- Esquerda JE (2011) Defective neuromuscular junction organi- iting the mTOR-ULK1 pathway, resulting in reduced podocyte zation and postnatal myogenesis in mice with severe spinal mus- injury. PLoS One 8:e63799 cular atrophy. J Neuropathol Exp Neurol 70:444–461 174. Rodriguez-Muela N, Parkhitko A, Grass T, Gibbs RM, Nor- 155. Martinez-Hernandez R, Bernal S, Also-Rallo E, Alias L, Barcelo abuena EM, Perrimon N, Singh R, Rubin LL (2018) Blocking MJ, Hereu M, Esquerda JE, Tizzano EF (2013) Synaptic defects p62/SQSTM1-dependent SMN degradation ameliorates Spinal in type I spinal muscular atrophy in human development. J Pathol Muscular Atrophy disease phenotypes. J Clin Invest. https :// 229:49–61doi.org/10.1172/JCI95 231 156. Murray LM, Comley LH, Thomson D, Parkinson N, Talbot K, 175. Acsadi G, Lee I, Li X, Khaidakov M, Pecinova A, Parker Gillingwater TH (2008) Selective vulnerability of motor neurons GC, Huttemann M (2009) Mitochondrial dysfunction in a and dissociation of pre- and post-synaptic pathology at the neuro- neural cell model of spinal muscular atrophy. J Neurosci Res muscular junction in mouse models of spinal muscular atrophy. 87:2748–2756 Hum Mol Genet 17:949–962 176. Wang ZB, Zhang X, Li XJ (2013) Recapitulation of spinal motor 157. Engqvist-Goldstein AE, Drubin DG (2003) Actin assembly and neuron-specific disease phenotypes in a human cell model of endocytosis: from yeast to mammals. Annu Rev Cell Dev Biol spinal muscular atrophy. Cell Res 23:378–393 19:287–332 177. Xu CC, Denton KR, Wang ZB, Zhang X, Li XJ (2016) Abnormal 158. Kubler E, Riezman H (1993) Actin and fimbrin are required mitochondrial transport and morphology as early pathological for the internalization step of endocytosis in yeast. EMBO J changes in human models of spinal muscular atrophy. Dis Model 12:2855–2862 Mech 9:39–49 159. Hosseinibarkooie S et al (2016) The power of human protective 178. Patitucci TN, Ebert AD (2016) SMN deficiency does not modifiers: PLS3 and CORO1C unravel impaired endocytosis induce oxidative stress in SMA iPSC-derived astrocytes or in spinal muscular atrophy and rescue SMA phenotype. Am J motor neurons. Hum Mol Genet 25:514–523 Hum Genet 99:647–665 179. Miller N, Shi H, Zelikovich AS, Ma YC (2016) Motor neuron 160. Riessland M et al (2017) Neurocalcin delta suppression pro- mitochondrial dysfunction in spinal muscular atrophy. Hum tects against spinal muscular atrophy in humans and across Mol Genet 25:3395–3406 species by restoring impaired endocytosis. Am J Hum Genet 180. Ripolone M et al (2015) Impaired muscle mitochondrial bio- 100:297–315 genesis and myogenesis in spinal muscular atrophy. JAMA 161. Kaur J, Debnath J (2015) Autophagy at the crossroads of catab- Neurol 72:666–675 olism and anabolism. Nat Rev Mol Cell Biol 16:461–472 181. Fuller HR, Gillingwater TH, Wishart TM (2016) Common- 162. Lamb CA, Yoshimori T, Tooze SA (2013) The autophagosome: ality amid diversity: multi-study proteomic identification of origins unknown, biogenesis complex. Nat Rev Mol Cell Biol conserved disease mechanisms in spinal muscular atrophy. 14:759–774 Neuromuscul Disord 26:560–569 163. Martin S, Harper CB, May LM, Coulson EJ, Meunier FA, 182. Boyd PJ et  al (2017) Bioenergetic status modulates motor Osborne SL (2013) Inhibition of PIKfyve by YM-201636 neuron vulnerability and pathogenesis in a zebrafish model of dysregulates autophagy and leads to apoptosis-independent spinal muscular atrophy. PLoS Genet 13:e1006744 neuronal cell death. PLoS One 8:e60152 183. La Bella V, Kallenbach S, Pettmann B (2000) Expression 164. Mariño G, Niso-Santano M, Baehrecke EH, Kroemer G (2014) and subcellular localization of two isoforms of the survival Self-consumption: the interplay of autophagy and apoptosis. motor neuron protein in different cell types. J Neurosci Res Nat Rev Mol Cell Biol 15:81–94 62:346–356 165. Garcera A, Bahi N, Periyakaruppiah A, Arumugam S, Soler 184. van Wijk SJ, Timmers HT (2010) The family of ubiquitin-con- RM (2013) Survival motor neuron protein reduction deregu- jugating enzymes (E2s): deciding between life and death of pro- lates autophagy in spinal cord motoneurons in vitro. Cell Death teins. FASEB J 24:981–993 Dis 4:e686 185. Burnett BG, Munoz E, Tandon A, Kwon DY, Sumner CJ, Fisch- 166. Custer SK, Androphy EJ (2014) Autophagy dysregulation in beck KH (2009) Regulation of SMN protein stability. Mol Cell cell culture and animals models of spinal muscular atrophy. Biol 29:1107–1115 Mol Cell Neurosci 61:133–140 186. Han KJ et al (2012) Ubiquitin-specific protease 9x deubiquit- 167. Piras A et al (2017) Inhibition of autophagy delays motoneuron inates and stabilizes the spinal muscular atrophy protein-survival degeneration and extends lifespan in a mouse model of spinal motor neuron. J Biol Chem 287:43741–43752 muscular atrophy. Cell Death Dis 8:3223 187. Chang HC, Hung WC, Chuang YJ, Jong YJ (2004) Degrada- 168. Ichimura Y, Kominami E, Tanaka K, Komatsu M (2008) Selec- tion of survival motor neuron (SMN) protein is mediated via the tive turnover of p62/A170/SQSTM1 by autophagy. Autophagy ubiquitin/proteasome pathway. Neurochem Int 45:1107–1112 4:1063–1066 188. Hicke L (2001) Protein regulation by monoubiquitin. Nat Rev 169. Loos B, du Toit A, Hofmeyr JH (2014) Defining and meas- Mol Cell Biol 2:195–201 uring autophagosome flux—concept and reality. Autophagy 189. Han KJ, Foster D, Harhaj EW, Dzieciatkowska M, Hansen K, 10:2087–2096 Liu CW (2016) Monoubiquitination of survival motor neuron 170. Yoshii SR, Mizushima N (2017) Monitoring and measuring regulates its cellular localization and Cajal body integrity. Hum autophagy. Int J Mol Sci 18:E1685 Mol Genet 25:1392–1405 171. Periyakaruppiah A, de la Fuente S, Arumugam S, Bahí N, 190. Abera MB et al (2016) ML372 blocks SMN ubiquitination and Garcera A, Soler RM (2016) Autophagy modulators regulate improves spinal muscular atrophy pathology in mice. JCI Insight survival motor neuron protein stability in motoneurons. Exp 1:e88427 Neurol 283:287–297 191. Harahap NIF et al (2015) Salbutamol inhibits ubiquitin-mediated survival motor neuron protein degradation in spinal muscular atrophy cells. Biochem Biophys Rep 4:351–356 1 3 H. Chaytow et al. 192. Tiziano FD et al (2010) Salbutamol increases survival motor ubiquitination in primary spinal muscular atrophy fibroblasts. neuron (SMN) transcript levels in leucocytes of spinal muscular Clin Chim Acta 411:1920–1928 atrophy (SMA) patients: relevance for clinical trial design. J Med 204. Powis RA, Mutsaers CA, Wishart TM, Hunter G, Wirth B, Genet 47:856–858 Gillingwater TH (2014) Increased levels of UCHL1 are a com- 193. Fuller HR, Mandefro B, Shirran SL, Gross AR, Kaus AS, Bot- pensatory response to disrupted ubiquitin homeostasis in spinal ting CH, Morris GE, Sareen D (2015) Spinal muscular atrophy muscular atrophy and do not represent a viable therapeutic target. patient iPSC-derived motor neurons have reduced expression of Neuropathol Appl Neurobiol 40:873–887 proteins important in neuronal development. Front Cell Neurosci 205. Jablonka S, Sendtner M (2017) Developmental regulation of 9:506 SMN expression: pathophysiological implications and perspec- 194. Ramser J et al (2008) Rare missense and synonymous variants tives for therapy development in spinal muscular atrophy. Gene in UBE1 are associated with X-linked infantile spinal muscular Ther 24:506–513 atrophy. Am J Hum Genet 82:188–193 206. Cho S, Dreyfuss G (2010) A degron created by SMN2 exon 7 195. Liu HY, Pfleger CM (2013) Mutation in E1, the ubiquitin acti- skipping is a principal contributor to spinal muscular atrophy vating enzyme, reduces Drosophila lifespan and results in motor severity. Genes Dev 24:438–442 impairment. PLoS One 8:e32835 207. Seo J, Singh NN, Ottesen EW, Sivanesan S, Shishimorova M, 196. Wishart TM et al (2014) Dysregulation of ubiquitin homeostasis Singh RN (2016) Oxidative stress triggers body-wide skipping and beta-catenin signaling promote spinal muscular atrophy. J of multiple exons of the spinal muscular atrophy gene. PLoS One Clin Investig 124:1821–1834 11:e0154390 197. Powis RA et al (2016) Systemic restoration of UBA1 ameliorates 208. Singh NN, Seo J, Rahn SJ, Singh RN (2012) A multi-exon- disease in spinal muscular atrophy. JCI Insight 1:e87908 skipping detection assay reveals surprising diversity of splice 198. Kwon DY, Dimitriadi M, Terzic B, Cable C, Hart AC, Chitnis A, isoforms of spinal muscular atrophy genes. PLoS One 7:e49595 Fischbeck KH, Burnett BG (2013) The E3 ubiquitin ligase mind 209. Miguel-Aliaga I, Culetto E, Walker DS, Baylis HA, Sattelle DB, bomb 1 ubiquitinates and promotes the degradation of survival Davies KE (1999) The Caenorhabditis elegans orthologue of the of motor neuron protein. Mol Biol Cell 24:1863–1871 human gene responsible for spinal muscular atrophy is a maternal 199. Kim EK, Choi EJ (2017) SMN1 functions as a novel inhibitor for product critical for germline maturation and embryonic viability. TRAF6-mediated NF-kappaB signaling. Biochim Biophys Acta Hum Mol Genet 8:2133–2143 1864:760–770 210. Chan YB, Miguel-Aliaga I, Franks C, Thomas N, Trülzsch B, 200. Setsuie R, Wada K (2007) The functions of UCH-L1 and its rela- Sattelle DB, Davies KE, van den Heuvel M (2003) Neuromus- tion to neurodegenerative diseases. Neurochem Int 51:105–111 cular defects in a Drosophila survival motor neuron gene mutant. 201. Choi J, Levey AI, Weintraub ST, Rees HD, Gearing M, Chin Hum Mol Genet 12:1367–1376 LS, Li L (2004) Oxidative modifications and down-regulation 211. Rajendra TK, Gonsalvez GB, Walker MP, Shpargel KB, Salz HK, of ubiquitin carboxyl-terminal hydrolase L1 associated with Matera AG (2007) A Drosophila melanogaster model of spinal idiopathic Parkinson’s and Alzheimer’s diseases. J Biol Chem muscular atrophy reveals a function for SMN in striated muscle. 279:13256–13264 J Cell Biol 176:831–841 202. Castegna A et al (2002) Proteomic identification of oxidatively 212. Bowerman M, Murray LM, Beauvais A, Pinheiro B, Kothary modified proteins in Alzheimer’s disease brain. Part I: creatine R (2012) A critical smn threshold in mice dictates onset of an kinase BB, glutamine synthase, and ubiquitin carboxy-terminal intermediate spinal muscular atrophy phenotype associated with hydrolase L-1. Free Radic Biol Med 33:562–571 a distinct neuromuscular junction pathology. Neuromuscul Dis- 203. Hsu SH et  al (2010) Ubiquitin carboxyl-terminal hydrolase ord 22:263–276 L1 (UCHL1) regulates the level of SMN expression through 1 3

Journal

Cellular and Molecular Life SciencesSpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off