A nop56 Zebrafish Loss-of-Function Model Exhibits a Severe Neurodegenerative Phenotype

NOP56 belongs to a C/D box small nucleolar ribonucleoprotein complex that is in charge of cleavage and modification of precursor ribosomal RNAs and assembly of the 60S ribosomal subunit. An intronic expansion in NOP56 gene causes Spinocerebellar Ataxia type 36, a typical late-onset autosomal dominant ataxia. Although vertebrate animal models were created for the intronic expansion, none was studied for the loss of function of NOP56. We studied a zebrafish loss-of-function model of the nop56 gene which shows 70% homology with the human gene. We observed a severe neurodegenerative phenotype in nop56 mutants, characterized mainly by absence of cerebellum, reduced numbers of spinal cord neurons, high levels of apoptosis in the central nervous system (CNS) and impaired movement, resulting in death before 7 days post-fertilization. Gene expression of genes related to C/D box complex, balance and CNS development was impaired in nop56 mutants. In our study, we characterized the first NOP56 loss-of-function vertebrate model, which is important to further understand the role of NOP56 in CNS function and development.


Introduction
NOP56 is part of a core protein complex, along with NOP58, fibrillarin (FBL) and a 15.5 kD RNA-binding protein, of the C/D box small nucleolar ribonucleoprotein (snoRNP), which is in charge of cleavage and modification of precursor ribosomal RNAs (pre-rRNAs) and assembly of the 60S ribosomal subunit [1][2][3] Specifically, this C/D box snoRNP introduces a direct 2 -O-ribose methylation in specific residues in pre-rRNA and is involved in the endonucleolytic cleavages of the 35S rRNA primary transcript [4]. This C/D complex is highly conserved throughout evolution from archaea to humans [3,[5][6][7][8][9].
NOP56 is decisive for the final maturation and activation of a catalytically active C/D box snoRNP. Small nucleolar RNAs (snoRNAs) control protein levels of NOP56 depending on the availability of C/D box snoRNP assembly factors [10]. Mutations in NOP56 in yeast and human cells lead to altered ribosomal biogenesis, notably on the pathway of 25S/5.8S rRNA synthesis, produce defects in 60S subunit assembly and disrupt cell cycle progression [7,10].
C/D complex dysregulation has been associated with cancer, mainly by effects on snoRNAs, which show differential expression patterns in diverse types of human cancers provides the first NOP56 loss-of-function vertebrate model, which will be an important tool to further understand the role of NOP56 in CNS development and function.

Zebrafish Care and Maintenance
Zebrafish carrying the wild-type and mutant nop56 sa12582 alleles [42] were obtained from the European Zebrafish Resource Center (EZRC). The Nop56 sa12582 line was generated by the ENU method (N-ethyl-N-nitrosourea) [43], which caused a C > T change that resulted in a premature stop codon.
Zebrafish individuals were maintained in the fish facilities of the Department of Genetics of the University of Santiago de Compostela (code of ethical approval committee and for animal experiments: AE-LU-003, ES270280346401) at 28 • C with a photoperiod of 14 h of light and 10 h of darkness according to standardized protocols [44,45]. All experiments involving animals followed the guidelines of the European Community and Spanish Government on animal care and experimentation (Directive 2012-63-UE and RD 53/2013).

Genotyping
DNA was extracted with Chelex 100 Resin (1422822, Bio-Rad; Hercules, CA, USA) and amplified by PCR with AmpliTaq Gold™ DNA Polymerase (Thermo Fisher Scientific; Waltham, MA, USA) in a thermal cycler with the primers specific for the mutation in nop56 sa12582 : F: 5 TGGCGGAAGATTTGATTCTG 3 and R: 5 TTTCCACTCGA-CATTCATCG 3 . PCR products were detected using 1% agarose gels. Sequencing of the PCR products were performed using the capillary sequencer 3730xl DNA Analyzer (Thermo Fisher Scientific) by Sanger sequencing. The results were analyzed with GeneMapper ® Software (Thermo Fisher Scientific; Waltham, MA, USA) and aligned using the CodonCode Aligner software (Codon Code Corporation; Centerville, USA). The results indicated if the analyzed individuals possess the mutation of interest and whether they are homozygous or heterozygous. Finally, we analyzed if the segregation of the genotypes was Mendelian by using a chi-square test.

Characterization of General Morphology
Fish carrying wild-type and mutant nop56 sa12582 alleles were used for a phenotypic study between 24 and 120 h post-fertilization (hpf). For this, 446 embryos and larvae were dechorionated and anesthetized with 0.002% tricaine methanesulfonate (Ms-222, Sigma-Aldrich; Saint Louis, MO, USA). Photographs were taken every 24 h with a Nikon Ds-Ri1 camera coupled to an inverted fluorescence microscope (AZ100 Multizoom Nikon; Tokyo, Japan). Images were analyzed with ImageJ software (National Institutes of Health; Bethesda, MD, USA), annotating alterations in general morphology and measuring parameters such as total length of the body and size of the head, eye and otoliths. All fish were genotyped by the end of the experiments. For more detailed analysis, a few selected embryos and larvae were mounted in lateral and ventral views in 1% agarose in phosphate-buffered saline pH 7.4 (PBS) and imaged likewise at higher magnification. In this case, we performed a detailed analysis of malformations in the jaw, body midline, somites and notochord.
In addition, brains and eyes of adult wild-type and heterozygous fish were analyzed. Adult fish were euthanized by tricaine methanesulfonate overdose and fixed overnight in 4% paraformaldehyde (PFA) in PBS at 4 • C. After dissection of the brain and eyes, these were imaged under a stereomicroscope. ImageJ software was then used to measure the area of the brain and optic tectum.

Survival Analysis
Survival rate of each group (wild types, heterozygous and homozygous) was calculated. A Kaplan-Meier graph of survival between 1 and 7 days post-fertilization was generated and the average days of survival were calculated. All fish were genotyped after the experiments.
Each sample was run in triplicate and we used primers for beta-actin 2 as housekeeping gene, F: 5 ACTTCACGCCGACTCAAACT 3 R: 5 ATCCTGAGTCAAGCGCCAAA 3 . Relative mRNA levels of each gene were normalized to the expression of the housekeeping gene through the 2 ∆∆CT calculation method.

Acridine Orange
Apoptotic cells of wild-type, nop56 +/− and nop56 −/− embryos at 24-30 hpf were stained with a 3 µg/mL solution of acridine orange in E3 medium for 30 min. After that, the embryos were washed twice in E3 medium. Confocal photomicrographs were taken with a Leica TCS SPE confocal laser microscope (Leica Microsystems, Wetzlar, Germany) with the GFP filter set (excitation 473, emission 520) at 5 × magnification and always using the same parameters of laser intensity and photomultiplier gain. The images were analyzed with ImageJ software (National Institutes of Health; Bethesda, MD, USA). Pixels of the embryo were selected using the tracing tool and the area; the mean and the integrated density of each fish were measured. These measurements were combined in the formula of corrected total cell fluorescence (CTCF): CTCF = integrated density − (area of selected cell × mean fluorescence of background readings).

Immunofluorescence
Fish were euthanized by tricaine methanesulfonate overdose and then fixed with 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS; pH 7.4) for at least 2 h at room temperature.
For serotonin (5-HT) and zebrin immunofluorescence experiments, after washes in PBS, larvae were incubated in collagenase (2 mg/mL in PBS) for 25 min at room temperature and then in glycine (50 mM in PBS with 1% Triton X-100) for 10 min at room temperature. Then, the animals were incubated with rabbit anti-5-HT (Immunostar, Still Water, MN, USA; Cat#: 20080; dilution 1:2500) or Zebrin II (mouse anti-Aldolase C; kindly provided by Prof. Richard Hawkes: dilution 1:100) antibodies overnight at 4 • C. The larvae were rinsed in PBS with 1% Triton X-100 (1% PBST) and incubated overnight at 4 • C with a Cy3-conjugated goat anti-rabbit antibody (Millipore; Burlington, MA, USA; dilution 1:200) or a FITC-conjugated goat anti-mouse antibody (Millipore; dilution 1:100). Antibodies were always diluted in 1% PBST, 1% DMSO, 1% normal goat serum and 1% bovine serum albumin. Larvae were mounted with 70% glycerol in PBS. Confocal photomicrographs were taken with TCS-SP2 spectral confocal laser microscope (Leica Microsystems, Wetzlar, Germany). For the quantification of serotonergic neurons, the total number of 5-HT-ir interneurons located in the 4 spinal cord segments located at the level of the caudal fin was quantified manually using the cell counter plugin of the Fiji software (Image J; National Institutes of Health; Bethesda, MD, USA) going through the stack of confocal optical sections.

Behavioral Analysis
Embryos between 2 and 3 dpf were dechorionated with a pair of forceps and were subjected to touch-evoked escape response analysis. Behavior was recorded with a Nikon Ds-Ri1 camera coupled to an inverted fluorescence microscope (AZ100 Multizoom Nikon; Tokyo, Japan). Distance moved was measured under quantification mode of Zebralab software (Viewpoint).
Movement of 4 dpf zebrafish larvae was quantified with a Zebralab system composed of a Zebrabox (Viewpoint; Civrieux, France). The 4 dpf larvae were introduced in 96-well plates and total movement was measured in periods of 10 min during an hour alternating light and dark.

Morpholino Injection
p53 morpholino injection was used to assess that the observed phenotype is not a consequence of upregulation of p53-mediated apoptosis caused by off-target effects. p53 morpholino was designed and synthesized by Gene Tools (Philomath, OR, USA). Its sequence is: 5 GCGCCATTGCTTTGCAAGAATTG3 . It was injected into one-cell-stage embryos in a volume of 3-5 nL/embryo and at a concentration of 1 mM.

Statistical Analysis
Data analysis of the three groups (wild type, nop56 +/− and nop56 −/− ) and graphs were generated with GraphPad Prism (GraphPad; San Diego, CA, USA), using one-way ANOVA with Bonferroni test for multiple comparisons after performing the D'Agostino-Pearson normality test. Statistical significance was established at p-value < 0.05. Statistically significant tests are indicated with asterisks in the graphs.
Nop56 −/− individuals have a characteristic abnormal phenotype, showing a reduced body length (differences are statistically significant from 48 hpf onwards, p-value < 0.001) and a smaller head, otoliths and eyes (these 3 characteristics being statistically significant from 24 hpf onwards: p-value < 0.001) ( Figure 1). In addition, after detailed analysis, we also observed abnormal eye morphology in the homozygous fish compared to wild type, including coloboma and midline malformations ( Figure 2). We also observed other characteristics such as swollen yolk in 100% of the nop56 −/− embryos since 48 hpf (less than 2% of the wild type and nop56 +/− had swollen yolk). Cardiac edema was detected in 77% of the nop56 −/− embryos at 72 hpf, 88% at 96 hpf and in 100% at 120 hpf (less than 3% of the wild type and nop56 +/− had cardiac edema). Lordosis was first observed at 48 hpf in 40% of nop56 −/− , at 120 hpf the number rises to 50% of the larvae (less than 5% of the wild type and nop56 +/− had this malformation at this developmental stage). Nop56 −/− embryos fail to hatch from the chorion, but this is not the cause of the curved body shape because when manually removed from the chorion, they keep their curved body shape. From 96 hpf, we observed absence of the swimming bladder in 100% of the nop56 −/− embryos (100% of the wild-type and nop56 +/− embryos had swimming bladder).
In addition, a more detailed analysis under a bright-field microscope showed additional malformations in homozygous fish compared to wild types by 3.5 dpf (Figure 2), mainly affecting the jaw, eye and brain. Homozygous fish had much smaller jaw and seemed to show abnormal patterning of the brain, with much smaller or absent cerebellum and smaller midbrain.
In order to analyze possible macroscopic differences in the brain of heterozygous individuals for the nop56 gene in adult zebrafish, the dimensions of different regions of fixed brains were measured. Data analysis between wild-type individuals and heterozygotes carrying the mutation at six months of age, resulted in no significant differences observed in the sizes corresponding to different areas of the brain, including the cerebellum (not shown).
Finally, we analyzed survival of wild-type (n = 43), nop56 +/− (n = 75) and nop56 −/− (n = 44) zebrafish during their first week of life and plotted their survival curves using a Kaplan-Meier survival graph ( Figure 1B). A log-rank statistical test comparing the individual group survival curves revealed that the nop56 −/− had a significantly shorter survival with decease start as soon as 4 dpf as compared to nop56 +/− and wild-type (p < 0.0001) animals during early development. Our results showed that 80% of the nop56 −/− larvae died before 6 dpf and all the nop56 −/− animals died by 7 dpf. In addition, at 96 and 120 hpf, there is total absence of swimming bladder in all nop56 −/− embryos, curved body shape (in 50% of nop56 −/− larvae) and a cardiac edema (in 88% of the nop56 −/− larvae at 96 hpf and 100% at 120 hpf). For these parameters, no statistically significant differences were observed between wild type and nop56 +/− . Statistically significant data in the graphs are indicated with a *. Scale bar: 1000 µm. (n = 44) zebrafish, revealing significantly different survivals for the nop56 −/− embryos p < 0.0001.
In addition, a more detailed analysis under a bright-field microscope showed additional malformations in homozygous fish compared to wild types by 3.5 dpf (Figure 2), mainly affecting the jaw, eye and brain. Homozygous fish had much smaller jaw and seemed to show abnormal patterning of the brain, with much smaller or absent cerebellum and smaller midbrain. In order to analyze possible macroscopic differences in the brain of heterozygous individuals for the nop56 gene in adult zebrafish, the dimensions of different regions of fixed brains were measured. Data analysis between wild-type individuals and heterozygotes carrying the mutation at six months of age, resulted in no significant differences observed in the sizes corresponding to different areas of the brain, including the cerebellum (not shown).

Malformations in the Central and Peripheral Nervous System of nop56 −/− Larvae
We quantified mRNA levels by RT-qPCR of three genes that are expressed in specific cell types of the zebrafish cerebellum: ptf1a and grid2 (Purkinje cells) and cbln12 (granule cells) [47,48]. We observed a significant reduction in ptf1a, grid2 and cbln12 in nop56 −/− (p-value < 0.0001) (Figure 3). grid2 and cbln12 expression was observed to be also reduced in nop56 +/− between 72 and 120 hpf in comparison with wild-type embryos (p-value < 0.0001) but not as strong as nop56 −/− larvae.
We analyzed in more detail the central and peripheral nervous system in 3.5 dpf wild type, nop56 +/− and nop56 −/− by using alpha-tubulin and synaptic vesicles 2 (SV2). In the head, we observed abnormal innervation of the jaw (Figure 4). In the brain, we observed smaller optic tectum and smaller or absent cerebellum ( Figure 4). Zebrin II immunofluorescence experiments revealed a complete lack of cerebellar Purkinje cells in 4 dpf nop56 −/− animals as compared to wild-type or heterozygous fish ( Figure 5).

Malformations in the Central and Peripheral Nervous System of nop56 Larvae
We quantified mRNA levels by RT-qPCR of three genes that are expressed in specific cell types of the zebrafish cerebellum: ptf1a and grid2 (Purkinje cells) and cbln12 (granule cells) [47,48]. We observed a significant reduction in ptf1a, grid2 and cbln12 in nop56 −/− (p-value < 0.0001) (Figure 3). grid2 and cbln12 expression was observed to be also reduced in nop56 +/− between 72 and 120 hpf in comparison with wild-type embryos (p-value < 0.0001) but not as strong as nop56 −/− larvae. We analyzed in more detail the central and peripheral nervous system in 3.5 dpf wild type, nop56 +/− and nop56 −/− by using alpha-tubulin and synaptic vesicles 2 (SV2). In the head, we observed abnormal innervation of the jaw (Figure 4). In the brain, we observed smaller optic tectum and smaller or absent cerebellum ( Figure 4). Zebrin II immunofluorescence experiments revealed a complete lack of cerebellar Purkinje cells in 4 dpf nop56 −/− animals as compared to wild-type or heterozygous fish ( Figure 5).

Neuromuscular Junction and 5-HT Spinal Cord Interneurons
We next analyzed neuromuscular junction integrity. For this, we used alpha-tubulin to label the whole trunk innervation and SV2 as a presynaptic marker. Both markers revealed abnormal innervation of the trunk myomeres in nop56 −/− compared with wild types by 3.5 dpf (Figure 6). Alpha-tubulin shows that axons of primary motor neurons are thicker and less branched in nop56 −/− than in wild types ( Figure 6 and Supplementary Materials: Videos S1 and S2). SV2 also shows less branching of the axons innervating the myomeres, as well as differences in myosepta staining.

Neuromuscular Junction and 5-HT Spinal Cord Interneurons
We next analyzed neuromuscular junction integrity. For this, we used alpha-tubulin to label the whole trunk innervation and SV2 as a presynaptic marker. Both markers revealed abnormal innervation of the trunk myomeres in nop56 −/− compared with wild types by 3.5 dpf ( Figure 6). Alpha-tubulin shows that axons of primary motor neurons are thicker and less branched in nop56 −/− than in wild types ( Figure 6 and Supplementary Materials: Videos S1 and S2). SV2 also shows less branching of the axons innervating the myomeres, as well as differences in myosepta staining.

Neuromuscular Junction and 5-HT Spinal Cord Interneurons
We next analyzed neuromuscular junction integrity. For this, we used alpha-tubulin to label the whole trunk innervation and SV2 as a presynaptic marker. Both markers revealed abnormal innervation of the trunk myomeres in nop56 −/− compared with wild types by 3.5 dpf ( Figure 6). Alpha-tubulin shows that axons of primary motor neurons are thicker and less branched in nop56 −/− than in wild types ( Figure 6 and Supplementary Materials: Videos S1 and S2). SV2 also shows less branching of the axons innervating the myomeres, as well as differences in myosepta staining.  We also observed that the number of 5-HT-ir neurons was significantly reduced in the spinal cord of 4 dpf nop56 −/− larvae (one-way ANOVA, p < 0.0001) as compared to control wild-type animals (Figure 7). No significant differences were observed between control wild-type and nop56 +/− animals. We also observed that the number of 5-HT-ir neurons was significantly reduced in the spinal cord of 4 dpf nop56 −/− larvae (one-way ANOVA, p < 0.0001) as compared to control wild-type animals (Figure 7). No significant differences were observed between control wild-type and nop56 +/− animals.

nop56 −/− Larvae Have Increased Apoptosis Mainly in the CNS
Acridine orange was used for in vivo staining of cellular apoptosis. An increase in apoptosis was seen in the nop56 −/− embryos (p-value< 0.0001), mainly in the eye, brain and spinal cord (Figure 8).

nop56 −/− Larvae Have Increased Apoptosis Mainly in the CNS
Acridine orange was used for in vivo staining of cellular apoptosis. An increase in apoptosis was seen in the nop56 −/− embryos (p-value < 0.0001), mainly in the eye, brain and spinal cord (Figure 8). We also observed that the number of 5-HT-ir neurons was significantly reduced in the spinal cord of 4 dpf nop56 −/− larvae (one-way ANOVA, p < 0.0001) as compared to control wild-type animals (Figure 7). No significant differences were observed between control wild-type and nop56 +/− animals.

nop56 −/− Larvae Have Increased Apoptosis Mainly in the CNS
Acridine orange was used for in vivo staining of cellular apoptosis. An increase in apoptosis was seen in the nop56 −/− embryos (p-value< 0.0001), mainly in the eye, brain and spinal cord (Figure 8). We examined mRNA expression by RT-qPCR of p21 and p53 genes that are related to apoptosis and cell cycle control [49]. We found both were overexpressed in nop56 −/− larvae (p-value < 0.0001). The microinjection of a p53 translation block morfolino at one-cell-stage did not rescue the nop56 −/− malformations (not shown) even when the expression of p53 was similar to wild type when measured at 72 hpf ( Figure 9). The reduction in p53 mRNA concentrations can be explained by the fact that it was observed that sometimes when a translation block morpholino binds to an mRNA, its secondary structure suffer changes, altering the availability of mRNA for nucleotyc degradation [50]. We examined mRNA expression by RT-qPCR of p21 and p53 genes that are related to apoptosis and cell cycle control [49]. We found both were overexpressed in nop56 −/− larvae (p-value < 0.0001). The microinjection of a p53 translation block morfolino at one-cell-stage did not rescue the nop56 −/− malformations (not shown) even when the expression of p53 was similar to wild type when measured at 72 hpf ( Figure 9). The reduction in p53 mRNA concentrations can be explained by the fact that it was observed that sometimes when a translation block morpholino binds to an mRNA, its secondary structure suffer changes, altering the availability of mRNA for nucleotyc degradation [50].  Touch-evoked response at 48 hpf was measured and observed to be impaired in nop56 −/− embryos. Moreover, when locomotion was measured at 96 hpf, that nop56 −/− larvae did not have the ability to swim was observed (Figure 10), which is consistent with our observation of abnormalities in neuromuscular junctions and spinal cord interneurons. Touch-evoked response at 48 hpf was measured and observed to be impaired in nop56 −/− embryos. Moreover, when locomotion was measured at 96 hpf, that nop56 −/− larvae did not have the ability to swim was observed (Figure 10), which is consistent with our observation of abnormalities in neuromuscular junctions and spinal cord interneurons. 3.6. Reduced Expression of the zpld1 Genes from the Cupula of the Inner Ear ZPLD1 (zona pellucida-like domain 1) is a gene expressed in the cupula, a gelatinous membrane overlying the crista ampullaris of the semicircular canal in the inner ear, which is important for sensing rotation of the head and critical for normal balance [51]. In Figure 10. Touch-evoked escape was reduced at 48 hpf in nop56 −/− embryos and locomotion was absent in nop56 −/− larvae at 96 hpf (p-value < 0.0001). Statistically significant data in the graphs are indicated with a *.

Reduced Expression of the zpld1 Genes from the Cupula of the Inner Ear
ZPLD1 (zona pellucida-like domain 1) is a gene expressed in the cupula, a gelatinous membrane overlying the crista ampullaris of the semicircular canal in the inner ear, which is important for sensing rotation of the head and critical for normal balance [51]. In zebrafish, there are two paralogous genes: zpld1a and zpld1b. Since SCA36 patients have hearing loss, and nop56 −/− zebrafish embryos developed a balance defect, we performed an expression analysis of these genes in our loss-of-function model. The expression of zpld1a and zpld1b in nop56 −/− was significantly reduced in homozygous mutants as compared to wild-type and heterozygous larvae (p-value < 0.0001) (Figure 11). zpld1b was also significantly reduced in nop56 +/− embryos at 120 hpf, but not as strong as npc1 −/− expression. Figure 10. Touch-evoked escape was reduced at 48 hpf in nop56 −/− embryos and locomotio absent in nop56 −/− larvae at 96 hpf (p-value < 0.0001). Statistically significant data in the grap indicated with a *.

Reduced Expression of the zpld1 Genes from the Cupula of the Inner Ear
ZPLD1 (zona pellucida-like domain 1) is a gene expressed in the cupula, a gelat membrane overlying the crista ampullaris of the semicircular canal in the inne which is important for sensing rotation of the head and critical for normal balance [5 zebrafish, there are two paralogous genes: zpld1a and zpld1b. Since SCA36 patients hearing loss, and nop56 −/− zebrafish embryos developed a balance defect, we perfo an expression analysis of these genes in our loss-of-function model. The express zpld1a and zpld1b in nop56 −/− was significantly reduced in homozygous mutants as pared to wild-type and heterozygous larvae (p-value < 0.0001) (Figure 11). zpld1b wa significantly reduced in nop56 +/− embryos at 120 hpf, but not as strong as npc1 −/− ex sion. Figure 11. Graphic representation of expression analysis by RT-qPCR. (A) Zpld1a expression is significantly reduced in nop56 −/− larvae between 72 and 120 hpf (p-value < 0.0001). (B) Zpld1a expression is significantly reduced in nop56 −/− larvae between 72 and 120 hpf (p-value < 0.0001) and in nop56 +/− at 120 hpf. Statistically significant data in the graphs are indicated with a *.

nop56 Disruption Causes Overexpression in nop58 and fbl Related Genes and Impaired rRNA Processing
NOP56 is a protein required for the assembly of the 60S ribosomal subunit. Together with NOP58 and FBL, they function as core proteins to form the C/D box small nucleolar ribonucleoprotein that modifies and processes ribosomal RNAs (see introduction). For this reason, we examined the mRNA expression of nop56 gene and associated genes nop58 and fbl by RT-qPCR ( Figure 12). We observed significant differences in nop56 expression between wild type, nop56 +/− and nop56 −/− embryos (p-value < 0.0001). nop56 −/− embryos have a reduced expression of nop56 and nop56 +/− embryos have an intermediate expression between nop56 −/− and wild type. Nop58 expression levels were increased in nop56 −/− larvae respect to wild type and nop56 +/− (p-value < 0.0001). fbl expression levels were also increased at 120 hpf in nop56 +/− and nop56 −/− embryos respect to wild-type animals (p-value < 0.0001).
In addition, as NOP56 participates in rRNAs processing, we examined by RT-qPCR the rRNA expression at 120 hpf of the 5 externally transcribed sequence (5ETS) and internally transcribed sequences (ITS1 and ITS2) of the 47S intermediate rRNA, whose transcription is implicated in the first step of ribosome biogenesis. Moreover, we quantified by RT-qPCR the expression at 120 hpf of 18S rRNA, which is generated after processing of the 47S one. We obtained similar results as in a zebrafish knockout model of fbl gene [46]. 18S rRNA levels were significantly lower in nop56 −/− in comparison with nop56 +/− and wild type (p-value < 0.0001), which means that rRNA processing is impaired. 5 ETS and ITS2 expression did not differ between the three groups, although we detected a surprising overexpression of ITS1 (p-value < 0.0001) (Figure 13). This indicates that 47S rRNA transcription was not affected but subsequent processing was indeed affected. nop58 and fbl by RT-qPCR ( Figure 12). We observed significant differences in nop56 expression between wild type, nop56 +/− and nop56 −/− embryos (p-value < 0.0001). nop56 −/− embryos have a reduced expression of nop56 and nop56 +/− embryos have an intermediate expression between nop56 −/− and wild type. Nop58 expression levels were increased in nop56 −/− larvae respect to wild type and nop56 +/− (p-value < 0.0001). fbl expression levels were also increased at 120 hpf in nop56 +/− and nop56 −/− embryos respect to wild-type animals (p-value < 0.0001). Figure 12. Graphic representation of mRNA expression analysis by RT-qPCR. (A) nop56 expression is significantly different in wild-type, nop56 +/− and nop56 −/− larvae between 72 and 120 hpf (p-value < 0.0001). (B) nop58 expression is significantly higher in nop56 −/− and nop56 +/− larvae between 72 and 120 hpf (p-value < 0.0001.) (C) fbl expression is significantly higher in nop56 −/− larvae between 72 and 120 hpf (p-value < 0.0001) and in nop56 +/− at 120 hpf respect to wild-type larvae (p-value < 0.0001). Statistically significant data in the graphs are indicated with a *.
In addition, as NOP56 participates in rRNAs processing, we examined by RT-qPCR the rRNA expression at 120 hpf of the 5′ externally transcribed sequence (5ETS) and internally transcribed sequences (ITS1 and ITS2) of the 47S intermediate rRNA, whose transcription is implicated in the first step of ribosome biogenesis. Moreover, we quantified by RT-qPCR the expression at 120 hpf of 18S rRNA, which is generated after processing of the 47S one. We obtained similar results as in a zebrafish knockout model of fbl gene [46]. 18S rRNA levels were significantly lower in nop56 −/− in comparison with nop56 +/− and wild type (p-value < 0.0001), which means that rRNA processing is impaired. 5′ETS and ITS2 expression did not differ between the three groups, although we detected a surprising overexpression of ITS1 (p-value < 0.0001) ( Figure 13). This indicates that 47S rRNA transcription was not affected but subsequent processing was indeed affected. Figure 12. Graphic representation of mRNA expression analysis by RT-qPCR. (A) nop56 expression is significantly different in wild-type, nop56 +/− and nop56 −/− larvae between 72 and 120 hpf (p-value < 0.0001). (B) nop58 expression is significantly higher in nop56 −/− and nop56 +/− larvae between 72 and 120 hpf (p-value < 0.0001.) (C) fbl expression is significantly higher in nop56 −/− larvae between 72 and 120 hpf (p-value < 0.0001) and in nop56 +/− at 120 hpf respect to wild-type larvae (p-value < 0.0001). Statistically significant data in the graphs are indicated with a *.

Expression of Genes Related to ALS Is also Affected in the nop56 Knockout Embryos
Mutations in autophagy regulator protein C9ORF72 and in DNA/RNA binding proteins TDP-43 and FUS are associated with Amyotrophic Lateral Sclerosis (ALS) disease. A hexanucleotide (GGGGCC) expansion in a noncoding region of C9ORF72 gene is the main cause of ALS and Frontotemportal Dementia (FTD). This expansion generates RNA foci, RNA/DNA G-quadruplexes and dipeptide repeat proteins (DPRs) that cause neurotoxicity [52]. TARDBP encodes the protein TDP-43, which regulates transcription, alternative splicing of mRNA and non-homologous end joining (NHEJ) repair of DNA in motor neurons [53]. FUS (Fused in Sarcoma) is involved in RNA metabolism (transcription, splicing and export to cytoplasm) and DNA repair of double strand breaks through DNA damage response [54]. It has been shown that FUS loss of function causes impairment of proper DNA damage response leading to neurodegeneration and formation of Figure 13. Graphic representation of expression analysis by RT-qPCR. (A) 18S rRNA expression is significantly reduced in nop56 −/− larvae at 120 hpf (p-value < 0.0001). (B) 5ETS expression does not differ between wild type, nop56 −/− and nop56 +/− embryos at 120 hpf (C) ITS1 expression is significantly higher in nop56 −/− larvae between at 120 hpf (p-value < 0.0001). (D) ITS2 expression does not differ between wild type, nop56 −/− and nop56 +/− larvae at 120 hpf. Statistically significant data in the graphs are indicated with a *.

Expression of Genes Related to ALS Is Also Affected in the nop56 Knockout Embryos
Mutations in autophagy regulator protein C9ORF72 and in DNA/RNA binding proteins TDP-43 and FUS are associated with Amyotrophic Lateral Sclerosis (ALS) disease. A hexanucleotide (GGGGCC) expansion in a noncoding region of C9ORF72 gene is the main cause of ALS and Frontotemportal Dementia (FTD). This expansion generates RNA foci, RNA/DNA G-quadruplexes and dipeptide repeat proteins (DPRs) that cause neurotoxicity [52]. TARDBP encodes the protein TDP-43, which regulates transcription, alternative splicing of mRNA and non-homologous end joining (NHEJ) repair of DNA in motor neu-rons [53]. FUS (Fused in Sarcoma) is involved in RNA metabolism (transcription, splicing and export to cytoplasm) and DNA repair of double strand breaks through DNA damage response [54]. It has been shown that FUS loss of function causes impairment of proper DNA damage response leading to neurodegeneration and formation of aggregates [55]. Expression assay of c9orf72, tardbp and fus ( Figure 14) revealed a significant reduction in the expression of these genes in the nop56 −/− embryos (c9orf72 p-value < 0.001, tardbp p-value < 0.001, fus p-value < 0.0001).

Discussion
In this work, we characterized a zebrafish loss-of-function model of the nop56 gene. Homozygous mutants nop56 +/− showed a severe neurodegenerative phenotype characterized by, among other features, absence of cerebellum, reduced numbers of spinal cord neurons and impaired movement. Apoptosis pathway was increased in the CNS causing death before 7 days post-fertilization (dpf). Nop58 and fbl genes were overexpressed, while rRNA processing seems to be impaired. Other genes with reduced expression in nop56 −/− were c9orf72, fus, tardbp (genes related to ALS), zpld1a, zpld1b (genes involved in the development of the cupula of the inner ear, which is responsible for balance) and ptf1a, grid2 and cbln12 (expressed in the zebrafish cerebellum).
Previous studies in cells demonstrated that reduced expression of NOP56 gene leads to decreased pre-rRNA biogenesis and an increase in apoptotic cells [10]. When NOP58 expression is reduced in cells, increased expression of NOP56 and FBL can be observed [10]. Similar to these results, our data also indicate impaired rRNA processing and increased apoptosis (mainly in the CNS), together with mRNA overexpression of nop58 and fbl, which seems to be a compensatory mechanism for reduced nop56 expression. A zebrafish fbl mutant develops a similar phenotype with severe neurodegeneration accompanied by eye abnormalities, massive apoptosis, defects in ribosome biogenesis and activity, and impaired S-phase progression [46].
Strong nop56 mRNA expression was previously localized in the retina, posterior midbrain lamina and the cerebellum of zebrafish embryos [56]. In mice Nop56 protein expression was also found mainly in CNS in Purkinje cells of the cerebellum, motor neurons of the hypoglossal nucleus and spinal cord anterior horn [19]. Increased apoptosis restricted to the CNS was observed, which seems to be in concordance with nop56 expression mainly in the CNS.
Locomotor defects in nop56 −/− could be caused by absence of cerebellum, reduced numbers of 5-HT-ir spinal cord neurons and defected body innervation. Previous studies linked cerebellar ataxia with the impairment of the serotonergic cerebellar system, where it was used as a target for some experimental treatments [57][58][59]. More recently, it has been seen in other ataxia animal models that treatments with serotonergic agents im- Figure 14. Graphic representation of mRNA expression analysis by RT-qPCR. (A) c9of72 expression is significantly reduced in nop56 −/− embryos between 72 and 120 hpf (p-value < 0.001) and in nop56 +/− at 120 hpf (p-value=0.0015). (B) tardbp expression is significantly reduced in nop56 −/− embryos between 72 and 120 hpf (p-value < 0.001). (C) fus expression is significantly reduced in nop56 −/− embryos between 72 and 120 hpf (p-value < 0.0001) and in nop56 +/− embryos at 120 hpf (p-value < 0.0001). Statistically significant data in the graphs are indicated with a *.

Discussion
In this work, we characterized a zebrafish loss-of-function model of the nop56 gene. Homozygous mutants nop56 +/− showed a severe neurodegenerative phenotype characterized by, among other features, absence of cerebellum, reduced numbers of spinal cord neurons and impaired movement. Apoptosis pathway was increased in the CNS causing death before 7 days post-fertilization (dpf). Nop58 and fbl genes were overexpressed, while rRNA processing seems to be impaired. Other genes with reduced expression in nop56 −/− were c9orf72, fus, tardbp (genes related to ALS), zpld1a, zpld1b (genes involved in the development of the cupula of the inner ear, which is responsible for balance) and ptf1a, grid2 and cbln12 (expressed in the zebrafish cerebellum).
Previous studies in cells demonstrated that reduced expression of NOP56 gene leads to decreased pre-rRNA biogenesis and an increase in apoptotic cells [10]. When NOP58 expression is reduced in cells, increased expression of NOP56 and FBL can be observed [10]. Similar to these results, our data also indicate impaired rRNA processing and increased apoptosis (mainly in the CNS), together with mRNA overexpression of nop58 and fbl, which seems to be a compensatory mechanism for reduced nop56 expression. A zebrafish fbl mutant develops a similar phenotype with severe neurodegeneration accompanied by eye abnormalities, massive apoptosis, defects in ribosome biogenesis and activity, and impaired S-phase progression [46]. Strong nop56 mRNA expression was previously localized in the retina, posterior midbrain lamina and the cerebellum of zebrafish embryos [56]. In mice Nop56 protein expression was also found mainly in CNS in Purkinje cells of the cerebellum, motor neurons of the hypoglossal nucleus and spinal cord anterior horn [19]. Increased apoptosis restricted to the CNS was observed, which seems to be in concordance with nop56 expression mainly in the CNS.
Locomotor defects in nop56 −/− could be caused by absence of cerebellum, reduced numbers of 5-HT-ir spinal cord neurons and defected body innervation. Previous studies linked cerebellar ataxia with the impairment of the serotonergic cerebellar system, where it was used as a target for some experimental treatments [57][58][59]. More recently, it has been seen in other ataxia animal models that treatments with serotonergic agents improved significantly the locomotion [60]. In addition, we observed reduced mRNA expression of zpld1 genes in nop56 −/− , this reduced expression of ZPLD1 genes in other species was related to balance dysfunction. In mouse, two spontaneous mutations in Zpld1 gene resulted in vestibular dysfunction but not auditory dysfunction [61]. In humans, a mutation in ZPLD1 was found in a patient with balanced translocation and cerebral cavernous malformations [62].
Our results show altered mRNA levels of tdp-43 and fus (which are related to ALS) in nop56 −/− . The relationship between TDP-43, FUS and NOP56 was previously signaled by Miyazaki and colleagues [63] in an ALS mouse model, which shows a progressive reduction in the mRNA levels of Nop56, Tdp-43 and Fus in large motor neurons. TDP-43 and FUS participate, as NOP56, in RNA processing pathway. We also analyzed mRNA expression of another gene related to ALS, c9orf72. This gene not only is an autophagy regulator, but also participates in the regulation of actin dynamics and axon extension in motor neurons [64]. We found a reduced mRNA expression of c9orf72 in nop56 a12582 mutants, which correlated with reduced numbers of spinal cord neurons.
Nop56 sa12582 homozygous mutant had a severe cerebellar defect; in fact, they did not develop a cerebellum and had a premature death. In humans, the only known illness that affected NOP56 gene is SCA36. There are no homozygous patients for SCA36. However, the heterozygous intronic repeated expansion of GGCCTG in NOP56 causes SCA36, which presented less severe phenotype than our nop56 −/− zebrafish with reduced locomotion, sensory hearing loss and discrete motor neuron impairment. Heterozygous nop56 fishes had reduced nop56 mRNA expression in comparison with wild types, but not as dramatically as homozygous nop56 −/− fish. Nop58 and fbl mRNA were also overexpressed in nop56 +/− compared with wild type, while c9orf72, fus, cbln12, grid2 and zpld1b had reduced mRNA expression starting mainly at 5 dpf. All these statistically significant differences of expression compared to the wild type are not as strong as those of the nop56 −/− Our studies in the brain of nop56 +/− adult fishes of 6 months did not show any macroscopic differences and also we did not observe any alteration in nop56 +/− Purkinje cells at 4 dpf. This does not mean that there are no neuronal differences between nop56 +/− and adult wild type, especially seeing that genes related to Purkinje cells (grid2), granular cells (cbln12), to balance (zpld1b) and genes that have been related to ALS (c9orf72 and fus) had reduced expression compared to wild type. Future research is necessary to see if there really are differences at the CNS level between adult nop56 +/− and adult wild type. Although SCA36 illness is thought to be a gain-of-function disease as ALS, recent studies in ALS signaled a causative combination of factors in which haploinsufficiency had an important role [22,24,25,52,65]. For these reasons, future studies on nop56 +/− older adult fishes are necessary.
Mice injected with a construct containing the SCA36 expansion showed not only RNA foci and DPR inclusions in the brain, but also locomotor defects and loss of Purkinje cells [30]. The generation of overexpression of SCA36 expansion in the zebrafish model facilitates the search for candidate therapies through drug or genetic screens to ameliorate the SCA36 symptoms.

Conclusions
We provided the first vertebrate NOP56 loss of function model in which we reported a severe neurodegenerative phenotype mainly characterized by early death, increase of apoptosis, absence of cerebellum, reduced numbers of spinal cord neurons and impaired movement.