U1 snRNP Biogenesis Defects in Neurodegenerative Diseases

The U1 small ribonucleoprotein (U1 snRNP) plays a pivotal role in the intricate process of gene expression, specifically within nuclear RNA processing. By initiating the splicing reaction and modulating 3’‐end processing, U1 snRNP exerts precise control over RNA metabolism and gene expression. This ribonucleoparticle is abundantly present, and its complex biogenesis necessitates shuttling between the nuclear and cytoplasmic compartments. Over the past three decades, extensive research has illuminated the crucial connection between disrupted U snRNP biogenesis and several prominent human diseases, notably various neurodegenerative conditions. The perturbation of U1 snRNP homeostasis has been firmly established in diseases such as Spinal Muscular Atrophy, Pontocerebellar hypoplasia, and FUS‐mediated Amyotrophic Lateral Sclerosis. Intriguingly, compelling evidence suggests a potential correlation in Fronto‐temporal dementia and Alzheimer's disease as well. Although the U snRNP biogenesis pathway is conserved across all eukaryotic cells, neurons, in particular, appear to be highly susceptible to alterations in spliceosome homeostasis. In contrast, other cell types exhibit a greater resilience to such disturbances. This vulnerability underscores the intricate relationship between U1 snRNP dynamics and the health of neuronal cells, shedding light on potential avenues for understanding and addressing neurodegenerative disorders.


Introduction
Nuclear RNA processing constitutes a pivotal phase in gene expression, molding the transcriptional output and finely regulating the localization, translation, and decay of mRNA.Within this intricate process, nascent pre-mRNA undergoes 5'capping, splicing, and 3'-end processing, each step meticulously orchestrated to contribute to cellular homeostasis and fitness maintenance. [1]omprising the U1 snRNA, a heptameric ring of Sm proteins, and three U1-specific proteins (U1-A, U1-C, and U1-70K), the three-dimensional organization of U1 snRNP, as revealed by its crystal structure, [2] underscores its functional prowess in mRNA splicing.U1 snRNP serves as a seeding particle during spliceosome assembly, binding to and defining the 5'-splice site by directly hybridizing the 5'-end of the U1 snRNA with the site, [3] a pivotal and limiting step in the splicing reaction, critical for splicing regulation.Moreover, 5'-splice site recognition emerged as a strategic hotspot for targeted splicing corrections using synthetic splicing switches, as demonstrated previously. [4,5]The abundance of U1 snRNP within the nucleus becomes a decisive factor in maintaining physiological splicing patterns and an accurate gene expression program.Beyond its role in pre-mRNA splicing, U1 snRNP assumes a crucial role in another facet of mRNA metabolism, specifically 3'-end processing. [6]The binding of U1 snRNP to nascent pre-mRNA not only stimulates mRNA transcription but also prevents premature cleavage and polyadenylation.This interplay involving RNA polymerase II, U1 snRNP, and polyadenylation proves vital for curbing productive transcription outside of proteincoding genes and ensuring the correct length of transcripts. [7]ndeed, investigations have revealed that the intracellular quantity of U1 snRNP modulates transcript length, influencing the encoded genetic information.Termed "telescripting", the molecular intricacies governing this process remain a subject of ongoing elucidation. [8]Altogether, as a main player in spliceosome assembly and a modulator of transcription and 3'-end processing, U1 snRNP is pivotal for gene expression and its regulation.
Over the past two decades, a myriad of human diseases has been associated with aberrant splicing and 3'-end processing patterns.Given the pivotal role of U1 snRNP in both of these processes, previous investigations have illuminated two mechanisms underlying non-physiological expression patterns involving U1 snRNP.In the realm of cancer, recurrent mutations have been pinpointed at the 5'-end of the U1 snRNA, the functional region governing 5'-splice site selectivity. [9]These mutations offer insights into how cancer cells may accrue non-physiological splicing events, unraveling a facet of the complex landscape of cancer-associated molecular alterations.Another dimension emerges in the context of various neurodegenerative diseases, where disruptions in RNA metabolism have been frequently observed. [10]In this context, the reduction in intracellular concentrations of spliceosome components like U1 snRNP emerged as a potential contributor to non-physiological splicing patterns and aberrant transcript lengths.
Within this manuscript, diverse strategies that induce disturbances in the U1 snRNP biogenesis pathway, specifically those linked to neurodegenerative diseases, are reviewed.Moreover, we present perspectives on future research endeavors aimed at enhancing our understanding of the intricate relationships between neuronal fitness and the dysregulation of mRNA metabolism.This exploration not only delves into the molecular underpinnings of neurodegenerative disorders but also paves the way for potential therapeutic interventions addressing the intricate interplay between U1 snRNP dynamics and neuronal health.

U1 snRNP Biogenesis
The biogenesis of U1 snRNP is a multifaceted process that commences in the nucleus with the synthesis of the U1 snRNA precursor by RNA polymerase II. [11]While unicellular organisms like S. cerevisiae possess a single copy of spliceosomal snRNAs, humans encode more than 200 genes for U1 snRNA variants.The expression of these diverse isoforms is intricately linked to cell state and differentiation status, yet the precise roles of these isoforms remain elusive. [12]Co-transcriptionally, the U1 snRNA precursor undergoes capping with an m7G-cap and is cleaved by Integrator at the 3'-box, a cis-RNA element situated approximately 200 nucleotides downstream of the coding sequence's end. [13]Following transcription, all Pol II-transcribed snRNAs, including U1 snRNA, undergo cytoplasmic maturation and assembly in higher eukaryotes (reviewed in, [14] Figure 1).The initial signal for nuclear export is the cotranscriptionally acquired m 7 G-cap, recognized by the heterodimeric cap-binding complex (CBC) comprising CBP20 (CBP stands for Cap Binding Protein) and CBP80.Upon binding, CBC recruits phosphorylated adaptor of export protein (PHAX), which harbors a nuclear export signal (NES) binding to the export receptor Exportin 1.Following PHAX dephosphorylation, the export complex dissociates, releasing U1 snRNA into the cytoplasm.
In the cytoplasm, three pivotal steps unfold: the assembly of the heptameric Sm core, 3'-trimming of U1 snRNA, and cap hypermethylation.The Sm core assembles on the Sm site, a short pyrimidine-rich region of U1 snRNA, occurring at the SMN (Survival of Motor Neuron) complex, comprising SMN and eight other proteins (Gemins 2-8 and unrip).Arginine methylation by PTRM5 (Protein arginine methyltransferase 5) of Sm B/B', D3, and D1, followed by loading into the SMN complex, leads to assembly on the U1 snRNA precursor.U1-70K also appears to play a role, bridging pre-U1 to SMN-Gemin2-Sm in a Gemin5independent manner, establishing an additional, U1-exclusive Sm core-assembly pathway. [15]This mechanism was proposed to contribute to U1 snRNP overabundance and regulates snRNP repertoire.Prior to nuclear import of the assembled Sm core, two RNA processing steps occur: hypermethylation of the cap structure and trimming of the 3'-extension.The DEDD (containing a motif Asp-Glu-Asp-Asp) family deadenylase member TOE1 (Target of EGR1) accomplishes 3'-end trimming, [16] while cap hypermethylation transforms the m 7 G cap into an m 3 2,2,7 G-cap, catalyzed by the methyltransferase Tgs1/PIMT [17] (Trimethylguanosine synthase 1/ L-isoaspartyl methyltransferase).Tgs1/PIMT binds to the Sm core, catalyzing this conversion and allowing nuclear import of the partially assembled U1 snRNP.Importin β facilitates the nuclear import of U1 snRNP pre-particles, recognizing a bipartite nuclear localization signal (NLS) comprising the m 3 2,2,7 G-cap and a region of the Sm core domain.Importin β associates directly with the Sm core NLS, while  The U1 snRNP biogenesis pathway.The U1 snRNA is produced as a precursor that contains a 3'-extension and is capped by a m7G before being exported.In the cytoplasm, the U1 precursor is bound by U1-70 K before the Sm heptamer is assembled at the SMN complex on the Sm site.U1-70 K was shown to be involved at this stage [15] and is depicted in orange.The U1 snRNA is subsequently trimmed in 3' by TOE1 while the cap is hyperphosphorylated by Tgs1/PIMT.The trimethylated cap (TmG) is then bound by the snuportin 1 followed by the importin β.The importin β also binds to the Sm core to promote the nuclear import of the pre-particles.In the nucleus, additional U1-specific proteins are added and RNA modifications occur (pseudouridine and 2'-O-methylation are represented by the symbol Ψ and red stars, respectively).At this end of the cycle, the mature U1 snRNP can act during splicing and modulate 3'-end processing.
snuportin 1 associates with the hypermethylated cap before recruiting importin β.Once inside the nucleus, U1 snRNP preparticles reside in Cajal bodies, where U1-specific proteins and RNA modifications, including pseudouridylations and 2'-Oribose methylations, are added. [18]This marks the completion of U1 snRNP maturation, rendering it competent to participate in RNA processing, including splicing and the regulation of 3'-end processing.

Injuries of U1 snRNP Biogenesis in Neurodegenerative Diseases
Neurons are particularly sensitive to non-physiological levels of spliceosome components and injuries in the U1 snRNP biogenesis pathway were linked to several neurodegenerative diseases.
Spinal Muscular Atrophy (SMA) is a devastating neuromuscular syndrome that causes a rapid loss of function of motor neurons until paralysis and death. [19]This syndrome is a major genetic cause of infantile death and is caused by the homozygous inactivation of the SMN1 gene.In the human genome, the SMN1 gene encodes for the SMN protein whose main role consists in chaperoning the assembly of the Sm core on U snRNA precursors. [20]This protein is essential and the loss of SMN1 is partially complemented by the SMN2 paralog gene.The severity of SMA symptoms negatively correlates with the SMN2 copy number.SMA represents a first example of neurodegenerative disease in which the U1 snRNP biogenesis pathway is affected by the reduction of available SMN complex (Figure 2).In line with a reduced intracellular level of U1 snRNP, SMA is associated with a large number of widespread perturbations in RNA metabolism, including alterations in pre-mRNA splicing [21] and mRNA length. [22]ontocerebellar hypoplasia (PCH) is a group of related conditions that affect the development of the brain. [23]The term pontocerebellar refers to the pons and the cerebellum, which are the brain structures that are most severely affected in many forms of this disorder.The pons is located at the base of the brain in an area called the brainstem, where it transmits signals between the cerebellum and the rest of the brain.The cerebellum, which is located at the back of the brain, normally coordinates movement.The term hypoplasia refers to the underdevelopment of these brain regions.Pontocerebellar hypoplasia also causes impaired growth of other parts of the brain, leading to an unusually small head size.Among this group of rare syndromes, Pontocerebellar hypoplasia type 7 (PCH7) is a unique recessive syndrome characterized by neurodegeneration with ambiguous genitalia.In 2017, Lardelli et al. identified the genetic causes of PCH7. [16]They uncovered in 12 human families with PCH7 that the biallelic loss of function of the TOE1 gene causes this syndrome.In line with their finding, TOE1-morphant zebrafish displayed mid-and hind-brain degeneration, modeling PCH-like structural defects in vivo.As mentioned above, the DEDD deadenylase TOE1 is responsive for the 3'-end trimming of the pre-U1 snRNA in the cytoplasm during U1 snRNP biogenesis. [24]Loss of TOE1 activity perturbs U1 snRNP biogenesis (Figure 2) and was associated with the disease.Together with SMA, PCH7 represents another example of neurodegenerative disease due to injuries in U1 snRNP biogenesis.
Mutations in the RNA-binding protein Fused in Sarcoma (FUS) cause early-onset amyotrophic lateral sclerosis (ALS), a relentless adult-onset disease characterized by loss of motor neurons in the motor cortex and spinal cord, leading to muscle weakness and eventually paralysis and death. [25,26]The FUS protein is composed of a N-terminal prion like domain responsive for the auto-assembly of the protein and its ability to form liquid droplets. [27]The C-terminal part of the protein is composed of two globular RNA binding domains, a RNA Recognition Motif (RRM) and a zinc finger, both of them being flanked by RGG repeat-containing regions.In healthy cells, FUS is nuclear and binds bipartite RNA targets. [28]FUS RRM recognizes RNA hairpin shape and is specific for YNY motif located in the 3'-part of the loop.In contrast, the zinc finger domain is highly specific for GGU motif embedded in a single stranded RNA.In healthy cells, FUS is nuclear and its main partner is the U1 snRNA.Cross-linking immunoprecipitation (CLIP) experiments coupled with structure determination by Nuclear Magnetic Resonance revealed that FUS binds U1 snRNP on the stem loop 3 of U1 snRNA which harbors a UGU motif at the 3'-part of the loop 3. [29] Even if it was not demonstrated, we proposed that FUS could act as a splicing factor involved in the recruitment of U1 snRNP on weak 5'-splice sites by targeting the stem loop 3 of the U1 snRNA using its RRM domain and a GGU motif located on the target pre-mRNA via its zinc finger domain.In FUS-linked ALS conditions, FUS accumulates in the cytoplasm and forms solid aggregates in the motor neurons.Even after this change in subcellular localization, in FUS-linked ALS, the main RNA partner of FUS remains the U1 snRNA. [29]nalysis of the CLIP data performed in FUS-linked ALS conditions revealed that the FUSÀ U1 snRNA crosslinked clustered in the stem loop 3 region but also in the Sm site, region of the U1 snRNA where the Sm core is assembled.The presence of FUS was later shown to impair the spontaneous assembly of the Sm core on the U1 snRNA in vitro.The colocalization between the U1 snRNA and FUS aggregates was observed in iPS-cell derived motor neurons and in the spinal cord of FUSlinked ALS mouse model.These results suggest that during FUS-linked ALS, FUS achieves a toxic gain of function in the cytoplasm that allows it to trap U1 snRNP biogenesis intermediates in liquid droplets that will maturate into solid aggregates (Figure 2).FUS-induced ALS represents another neurodegenerative disease caused by an injury in the U1 snRNP homeostasis.In line with this statement, aberrant alternative splicing programs and disturbed transcript lengths were previously pinpointed in FUS-linked ALS patients. [30]ltogether, SMA, PCH7 and FUS-linked ALS represent three neurodegenerative conditions that were linked to perturbations in U1 snRNP homeostasis.Interestingly, despite sharing several common symptoms, the identified injuries in the U1 snRNP biogenesis pathway are independent in the three conditions, suggesting that U1 snRNP homeostasis is essential for the maintenance of neuronal fitness.

What about other Neurodegenerative Syndromes?
Spinal Muscular Atrophy, Pontocerebellar hypoplasia type 7, and FUS-induced ALS are neurodegenerative diseases that cause injuries to the U1 snRNP biogenesis pathway, there are experimental proofs to support these mechanisms.However, there are other neurodegenerative diseases for which lines of evidences have been presented in the literature but the molecular mechanisms inducing perturbations in U1 snRNP homeostasis have not yet been demonstrated experimentally.
Mutations in FUS are not the only causes of Amyotrophic Lateral Sclerosis and, among others, another RNA binding protein called TDP-43 (TAR DNA-binding protein 43) can also accumulate in the cytoplasm upon specific mutation and induce Amyotrophic Lateral Sclerosis. [31,32]Previous studies have shown that U snRNA homeostasis is dysregulated in the case of TDP-43-induced ALS. [33]However, it remained to be determined whether the U1 snRNA is also trapped by cytoplasmic TDP-43.The RNA binding region of TDP-43 is composed of two RRMs which binds to GU rich sequences.The solution structure of TDP43 RRMs bound to 5'-GUGUGAAUGAAU-3' has revealed how the protein contacts specifically GU-rich sequences. [34]gure 2. Injuries in U1 snRNP biogenesis and neurodegenerative diseases.On the top right corner, a simplified scheme illustrates U1 snRNP biogenesis pathway.In Spinal Muscular Atrophy (SMA), the SMN complex is deficient, Sm core assembly efficiency is decreased and the production of functional U1 snRNP is reduced.In Pontocerebellar hypoplasia type 7 (PCH7), the TOE1 enzyme is mutated and cannot efficiently trim the 3'-end of U1 snRNA precursors.Consequently, the neurons cannot produce U1 snRNP efficiently.In FUS-induced ALS, mutated FUS accumulates in the cytoplasm and traps U1 snRNP biogenesis intermediates in liquid droplets that maturate in solid aggregates.Consequently, this sequestration of pre-U1 snRNP particles reduces the pool of nuclear and functional U1 snRNP.SMA, PCH7 and FUS-induced ALS are labelled in red since experimental evidences and mechanistic studies have demonstrated how injuries in U1 snRNP homeostasis trigger the neurodegenerative diseases.There are growing evidences in the literature that TDP-43 induced ALS, Frontotemporal dementia and Alzheimer's disease could also originate from injuries in U1 snRNP biogenesis pathway.However, since it was not demonstrated experimentally, these neurodegenerative diseases are labelled in light grey.
Interestingly, a similar GU-rich sequence is found on the Sm site of U1 snRNA and its flaking region: 5'-UUGUGGUAGU-3', suggesting that TDP-43 could bind to this sequence and disturb U1 snRNP homeostasis (Figure 2).This represents an attractive hypothesis that would require experimental evidence in order to be validated.In line with this hypothesis, dysregulated alternative splicing program was previously observed in TDP-43-induced ALS. [35]However, further investigation are still required to determine if these non-physiological splicing patterns are caused by the loss of nuclear TDP-43 or a decrease in U1 snRNP concentration.
The protein FUS has two closed homologs in the human genome: TAF15 (TATA-box binding protein associated factor 15) and EWS (Ewing Sarcoma); altogether, these three RNA binding proteins define the FET family. [36]While mutations in FUS cause ALS, mutations of TAF15 or EWS are responsive for another neurodegenerative disease coined Frontotemporal lobar dementia (FTLD). [37]FTLD is a clinically and pathologically heterogeneous syndrome, characterized by a progressive decline in behaviour or language associated with degeneration of the frontal and anterior temporal lobes.According to the high homology of sequence and structure between FUS, EWS and TAF15, one can also speculate that the three FET could act similarly on U1 snRNP homeostasis and explain the pathomechanism of FTLD (Figure 2).However, even if the hypothesis is attractive, experimental proofs supporting the proposed scenario are required.
In 2013, a first link between Alzheimer's disease (AD) and U1 snRNP has been discovered. [38]It was established that AD patients have widespread alterations in RNA metabolism, including RNA splicing and transcript lengths alterations, both processes in which U1 snRNP plays a pivotal role.Interestingly, U1 snRNP splicing disfunction was recently linked to neuronal hyperexcitability and cognitive impairment observed during AD. [39]In Alzheimer's disease (AD), the protein tau (tubulinassociated unit) forms cytoplasmic aggregates in neurons and causes neurodegeneration. [40]Using immunostaining, colocalization between cytoplasmic tau and hypermethylated cap was firstly established, suggesting that tau aggregates contains U snRNAs. [41]Purification of the sarkosyl-insoluble tau aggregates and analysis of the U snRNA content using RT-PCR revealed that there is an enrichment for the U1 snRNA in the tau aggregates.This statement was further supported by showing the colocalization of U1-70K (a U1-specific protein) and the cytoplasmic tau aggregates. [42]Later, it was shown that the C-terminal part of U1-70K, a region with highly repetitive basic (Arg/Lys) and acidic (Asp/Glu) residues is sufficient to interact with tau from AD patients but not from other tauopathies. [43]Another intriguing link between Alzheimer's disease and U1 snRNP biogenesis resides in the role of Tgs1/PIMT in neurodegeneration. [44]This enzyme colocalizes with neurofibrillary tangles in AD patient brains, [45] suggesting that U1 snRNP biogenesis intermediates could also be trapped in cytoplasmic tau aggregates (Figure 2).The Tgs1/PIMT enzyme is the methyltransferase involved in the hypermethylation of the cap of pre-U1 snRNP particles, suggesting that its deregulation or sequestration during AD could have an influence on U1 snRNP homeostasis.This second link between U1 snRNP and Alzheimer's disease suggests that another injury in the U1 snRNP biogenesis pathway may occur during AD, however, this hypothesis requires additional experimental proofs to be validated.

Summary and Outlook
Experimental evidences from the literature support that injuries in the U1 snRNP biogenesis pathway are found in several neurodegenerative diseases including Spinal Muscular Atrophy, Pontocerebellar Hypoplasia 7 and FUS-linked ALS.There is a clear relationship between neurodegeneration and altered U1 snRNP homeostasis.During the last decade, it was shown that it is possible to correct the SMN deficiency of SMA patients by multiple ways (splicing correction of the paralog gene SMN2 [4,5,46] or gene therapy that brings a novel copy of the SMN1 gene [47] ) and thus restore physiological U snRNP biogenesis pathway.Similar gene therapy approaches may also be applied to complement deficient TOE1 activity in the context of PCH7.However, getting rid of problematic mutated RBP that accumulates in the cytoplasm in the context of ALS, like FUS, may require more elaborated strategies.Since the U1 snRNP biogenesis pathway is shared between all the cells of an organism, why do these injuries affect only neurons and why others cells are not sensitive to this change?Literature has shown that oxidative stress is also a major driver of neurodegeneration. [48]ould it be possible that neurons accumulate more oxidative stress than other cells and are more sensitive to impairment in spliceosome homeostasis?Are there neuron specific splicing changes that are induced by the U1 snRNP homeostasis injury?Are the gene expression changes detrimental for neuronal glutamate transmission which is a hallmark of neurodegenerative diseases [49,50] ?Another possibility could come from noncanonical functions of U1 snRNP components in the axon.Indeed, it was recently proposed that cytoplasmic U1-70K participates in axonal mRNA transport and is essential for neuromuscular junctions in zebrafish. [51]Could the injury in U1 snRNP homeostasis disturb non-canonical cytoplasmic functions that are essential for neurons?Future investigations should bring answers to these questions.Furthermore, the molecular mechanisms triggering altered U1 snRNP homeostasis have been identified for SMA, PCH7 and FUS-ALS.However, they are growing evidences that such a phenomena may also occur in other neurodegenerative syndromes, like FTLD or AD.It remains now to validate or refute these hypotheses.To conclude, cell fitness is associated with the abundance of U1 snRNP.It was described in a single manuscript that during apoptosis, cleavage of the 5'-end of U1 snRNA can occur. [52]Could it be possible to promote specifically the degradation of U1 snRNP to induce cell death in a pathological context as cancer?Future investigations should consider this scenario and maybe explore how cells respond upon U1 snRNA-specific cleavage.
Biographical Sketch.Sebastien Campagne studied biology at the University of Toulouse and received his Ph.D. in 2010.After a postdoc at ETH Zurich in Switzerland, he was recruited as INSERM researcher in the ARNA unit located at the University of Bordeaux (2021).Since 2023, he was appointed group leader in the European Institute of Chemistry and Biology (IECB) in Bordeaux.His research interests include the comprehension of RNA metabolism in health and diseases and the rational development of synthetic splicing switches to correct specifically splicing and gene expression for therapeutic outcomes.

Figure 1 .
Figure1.The U1 snRNP biogenesis pathway.The U1 snRNA is produced as a precursor that contains a 3'-extension and is capped by a m7G before being exported.In the cytoplasm, the U1 precursor is bound by U1-70 K before the Sm heptamer is assembled at the SMN complex on the Sm site.U1-70 K was shown to be involved at this stage[15] and is depicted in orange.The U1 snRNA is subsequently trimmed in 3' by TOE1 while the cap is hyperphosphorylated by Tgs1/PIMT.The trimethylated cap (TmG) is then bound by the snuportin 1 followed by the importin β.The importin β also binds to the Sm core to promote the nuclear import of the pre-particles.In the nucleus, additional U1-specific proteins are added and RNA modifications occur (pseudouridine and 2'-O-methylation are represented by the symbol Ψ and red stars, respectively).At this end of the cycle, the mature U1 snRNP can act during splicing and modulate 3'-end processing.