The SON RNA splicing factor is required for intracellular trafficking that promotes centriole assembly

Control of centrosome assembly is critical for cell division, intracellular trafficking and cilia. Regulation of centrosome number occurs through the precise duplication of centrioles that reside in centrosomes. Here we explored transcriptional control of centriole assembly and find that the RNA splicing factor SON is specifically required for completing procentriole assembly. Whole genome mRNA sequencing identified genes whose splicing and expression are affected by the reduction of SON, with an enrichment in genes involved in the microtubule cytoskeleton, centrosome and centriolar satellites. SON is required for the proper splicing and expression of CEP131 which encodes a major centriolar satellite protein and is required to organize the trafficking and microtubule network around the centrosomes. This study highlights the importance of the distinct microtubule trafficking network that is intimately associated with nascent centrioles and is responsible for procentriole development.

To determine whether the initiating events of centriole assembly are affected when 165 SON is depleted, we examined the localization of core centriole assembly factors. CEP152 and 166 CEP192 are centriole scaffold proteins that interact with and are required for PLK4 recruitment 167 to the centriole (T.-S. Kim et al., 2013;Sonnen et al., 2013). Total CEP152 and CEP192 are 168 reduced at the centrosome in siSON (59% of siControl for both), but the mother centriole levels 169 are only reduced to 75% of controls, suggesting that some of the reduction in protein levels is 170 due to a lack of recruitment to procentrioles (Figure 2A suggesting that the defect in centriole assembly is not due to a failure to recruit PLK4 ( Figure  173  does not correlate with more centrioles, these data suggest that their overall decrease in 190 abundance is not responsible for the centriole assembly phenotype. 191 To ascertain whether centriole assembly initiates when SON is depleted, we examined 192 SAS-6 levels. Because SAS-6 is lost from mature centrioles during exit from mitosis, SAS-6 signal 193 exclusively marks assembling procentrioles (Puklowski et  CPAP and CEP135 were detected around mother centrioles, even though no Centrin labeled 202 procentriolar foci were observed ( Figure 3A). This indicates that centriole assembly initiates in 203 SON knockdown cells but is unable to proceed to completion. 204 To determine the structural events associated with these new procentriole assemblies, 205 we used electron tomography. Consistent with the immunofluorescence data, when SON is 206 depleted, three out of four cells examined exhibited incomplete procentriole rosettes with 207 procentrioles that were on average shorter than those observed in control cells ( Figure 3B, C). 208 Moreover, the microtubules in these procentrioles lack the C-tubule which is present in control 209 conditions ( Figure 3B insets, Movie S1, Movie S2). We therefore conclude that SON depletion 210 prevents daughter centriole assembly at a stage after early procentriole assembly and initial 211 microtubule nucleation events, but prior to complete triplet microtubule formation, Centrin 212 deposition and centriole elongation. 213

SON impacts splicing of genes encoding centriolar satellite and microtubule proteins. 214
To determine SON splicing targets responsible for the centriole assembly defect, we 215 utilized global mRNA sequencing of RNA isolated from cells under conditions of S phase arrest 216 and PLK4 induction. From these data we identified 4,413 genes downregulated in SON depleted 217 cells (Table S1). To establish whether SON has a greater effect on specific cellular structures, we 218 identified Cellular Component (CC) Gene Ontology (GO) terms most effected by SON depletion. 219 The most highly enriched GO terms include the microtubule cytoskeleton and centrosome 220 ( Figure 4A, Table S1), which is consistent with the microtubule disruption previously observed 221 after SON knockdown (Ahn et al., 2011;Sharma et al., 2011). Along with changes to expression 222 levels, we also examined changes in alternative splicing using the software packages MAJIQ 223 (Table S2) (Table S3) (Shen et al., 2014). From this 224 analysis we found 2,996 genes by MAJIQ and 3,531 genes by rMATS that showed significant 225 changes in alternative splicing ( Figure 4B). These two software packages identified genes with 226 considerable overlap (74.3% of MAJIQ identified genes in the rMATS set and 63.1% rMATS 227 identified genes in the MAJIQ set, Figure Supplement 4A). These genes also showed strong 228 enrichment for microtubule-associated GO terms including microtubule cytoskeleton, 229 microtubule organizing center, centrosome and centriole ( Figure 4A). To investigate additional 230 targets of SON that could contribute to the centriole assembly defect, we also tested for 231 overlap with genes included in the Centrosome Database (CD, portion of the Centrosome and 232 Cilia Database (Gupta et al., 2015)) and genes encoding proteins identified in an extensive 233 proximity mapping of centriolar satellites (Gheiratmand et al., 2019). Interestingly, differentially 234 spliced genes strongly overlapped with both of these additional lists, indicating that these 235 proteins are affected at a high rate by SON knockdown (Figure 4A, B). 236 Genes from the centrosome database and satellite proximity lists whose splicing is 237 affected by SON were then sorted by their differential expression (Table S4) additional targets that could impact centriole assembly. Therefore, the SON splicing factor is 250 SON is required for centriole assembly.
13 required for the homeostatic control of mRNAs encoding centrosomal, centriolar satellite and 251 microtubule cytoskeletal proteins. 252

SON depletion reduces centriolar satellites and trafficking structures. 253
We examined the distributions of centrosomal and centriolar satellite proteins whose 254 splicing is affected by SON using a semi-automated radial intensity analysis algorithm in control 255 and SON knockdown cells ( Figure 5A, B) (Sankaran et al., 2020). -tubulin, together with -256 tubulin complex members, acts to nucleate microtubules and is concentrated at the 257 centrosome (Stearns et al., 1991;Tovey & Conduit, 2018). SON depletion did not alter -tubulin 258 fluorescence intensity at the centrosome itself. However, a significant reduction in -tubulin 259 was observed in the region 1.5 to 2.5 µm from the centrosome. SON knockdown had a similar 260 effect on the distribution of Pericentrin, a large scaffolding protein that interacts with -tubulin 261 (Dictenberg et al., 1998) and localizes both to the centrosome and to mobile cytoplasmic 262 granules that transit to and from the centrosome (Galati et al., 2018;Young et al., 2000). The 263 Pericentrin in cytoplasmic granules is nearly eliminated, suggesting that Pericentrin-dependent 264 trafficking to and from centrosomes is severely affected when SON is depleted ( Figure 5A, B). 265 Centrobin is associated with daughter centrioles and important for their elongation (Gudi et al., 266 2011;Zou et al., 2005). Centrobin also limits the recruitment of additional PCM proteins (Jeffery 267 et al., 2013), stabilizes CPAP (Gudi et al., 2014(Gudi et al., , 2015 and promotes C-tubule formation and 268 maintenance (Reina et al., 2018). This makes Centrobin an intriguing candidate for the SON 269 knockdown centriole assembly phenotype as cells with reduced SON initiate but do not 270 complete centriole assembly ( Figure 3A, B), have a larger decrease in centrosomal CPAP than 271 other early centriole assembly factors ( Figure 2B), and produce procentrioles without the C-272 tubule ( Figure 3B). We observed that Centrobin fluorescence intensity is reduced at the 273 centrosome in the SON knockdown ( Figure 5A   To assess whether the reduction in centriolar satellite fluorescence intensities at or 301 around the centrosomes was due to mis-localization or total cellular loss of protein, total 302 fluorescence levels inside cells were quantified in both siSON and siControl cells ( Figure  303 Supplement 5D). Some proteins whose genes depend upon SON for correct splicing had 304 significant reductions in overall protein levels (83% for -tubulin, 53% for Pericentrin and 44% 305 for CEP131). PCM1 was also reduced in total protein levels (78%) but alternative splicing was 306 not significantly affected by SON depletion, suggesting that when satellites are disrupted, PCM1 307 protein stability is affected. We also detected an elevation in total cellular SAS-6 protein levels, 308 even though there was not a large transcriptional change as determined by RNA-sequencing 309 (1.06 fold change relative to siControl). In summary, SON activity is required for the normal 310 accumulation and distribution of centriolar satellites and Pericentrin trafficking structures. It 311 may do so through its robust effect on CEP131 protein levels. 312 SIM imaging revealed localization of PCM1 and CEP131 directly adjacent to nascent 313 procentrioles in siControl cells ( Figure 5A . To assess whether the centrosomal CEP152 reduction observed with SON depletion is 339 due to loss of CEP131, we quantified centrosomal CEP152 fluorescence in both SON and 340 CEP131 depletion conditions. Reductions in CEP152 were observed in both conditions ( Figure  341 Supplement 6C, D). Treatment with siCEP131 was more effective at reducing CEP131 levels than 342 treatment with siSON, but CEP152 levels in siSON were reduced by slightly more than by 343 siCEP131 alone. This suggests that promotion of correct CEP131 splicing by SON is partially 344 responsible for the observed reduction in CEP152. 345 We next examined how Pericentrin depletion in siSON contributes to the centriole 346 assembly defect. Depletion of Pericentrin was efficient, and reduced centriole assembly to a 347 greater degree than depletion of CEP131, although not to the level of SON depletion ( Figure 6F Because SON depletion is permissive for early steps of centriole assembly but prevents 363 advancement of the assembly process, we examined CEP131 and Pericentrin depletion for 364 procentriole intermediates as found in the SON knockdown. We utilized SAS-6 fluorescence as a 365 proxy for these procentriole intermediates. In SON depleted cells, we once again observed SAS-366 6 signal in a rosette formation without Centrin foci. Early centriole assembly intermediates 367 were never observed when Pericentrin was depleted, which is consistent with reports that SAS-368 6 requires Pericentrin for its localization (Ito et al., 2019), and only rarely observed when 369 CEP131 was depleted ( Figure Supplement 6E, F, G). We therefore conclude that when SAS-6 is 370 present at procentrioles in Pericentrin depleted cells, centriole assembly is able to proceed. 371 When SAS-6 is present at procentrioles in CEP131 depleted cells, centriole assembly usually 372 proceeds, but occasionally is slowed or arrested, as is the case for most cells depleted of SON 373 In total, CEP131 and Pericentrin depletion can recapitulate some of the centriole 378 assembly defects observed when SON is depleted, but depletion of these proteins individually 379 or in combination, nor can depletion of Centrobin, explain the severe centriole assembly defect 380 observed upon SON depletion, suggesting the mechanism of SON-based regulation of centriole 381 assembly is multifactorial. 382

SON depletion alters the microtubule landscape near the centrosome. 383
Because centriolar satellites utilize microtubule-dependent trafficking (Conkar et al.,384 2019; Kubo et al., 1999), and because we observed close association between centriolar 385 satellites and the assembling procentrioles, we examined the microtubule network near 386 centrosomes. We observed a 2-fold increase in microtubule density in SON depleted cells, 387 consistent with previous reports ( Figure 7A, B, Figure  Using SIM, we resolved the relationship between assembling centrioles, microtubules and 389 centriolar satellites ( Figure 7C). Microtubules originated from nascent procentrioles. 390 Furthermore, the centriolar satellite protein, PCM1, is positioned at the ends of these 391 microtubules, suggesting that assembling centrioles have microtubules and centriolar satellites 392 in close proximity to provide components required for their elongation and maturation. When 393 SON is depleted, the microtubule arrangement is distinctly different, with microtubules 394 crowded around mother centrioles. To quantify this rearrangement, we examined the 395 microtubule intensity around mother centrioles to determine the average distance at which 396 microtubules begin. Because the average intensity of microtubules dissipates as a function of 397 distance from the centrosome ( Figure 7B), we examined the shift from increasing microtubule 398 intensity to decreasing intensity as measurements are quantified radially from the mother 399 centriole as a way to establish the boundary of microtubule minus-ends close to the 400 centrosome. This occurred on average 480 nm from the mother centrioles in siControl cells and 401 340 nm in siSON cells ( Figure 7C, D) indicating that not only are there more microtubules 402 around centrosomes when SON is depleted but that the network encroaches closer to the 403 mother centrioles. 404 SON is required for centriole assembly. procentrioles. This localization is reduced when SON is depleted ( Figure 7G, H). We suggest that 422 microtubules are actively remodeled by centrosome localized Katanin p80 to ensure a 423 transition from the mother centriole to nascent procentrioles. When SON is depleted, Katanin is 424 removed from the centrosome thereby increasing microtubule nucleation directly from the 425 SON is required for centriole assembly.  Kodani et al., 2015). In this study, using RPE-1 cells, we observed close association of centriolar satellites with assembling procentrioles, but depletion of satellites 450 (through CEP131 depletion) and Pericentrin associated trafficking particles caused reductions in 451 centriole assembly, but did not eliminate it. The resilience in centriole assembly observed in this 452 study could be due to the S phase arrest and PLK4 induction used in these experiments, which 453 may make the process unnaturally robust. Alternatively, it could reflect the lack of reliance on 454 centriolar satellites for centriole assembly in RPE-1 cells. 455 There is, however, increasingly strong evidence that centriolar satellites are required for To assess total fluorescence within a cell, background was subtracted as described in the 569 previous paragraph. The Cetn channel was thresholded to generate a binary image 570 encompassing all the cell boundaries. This was subjected to the erode, dilate and fill holes 571 binary functions followed by applying the create selection function. The selection was added to 572 the region of interest manager and then transferred to the channel to be quantified. 573 To measure the proximity of microtubules to centrosomes in structured illumination 574 microscopy reconstructions, line scan measurements of fluorescence intensity were employed 575 beginning at the center of the centriole signal. The line was rotated 45 , generating eight 576 measurements for each centriole. Each line scan measurement was then normalized to one at 577 the initiating measurement at the centriole center prior to averaging all the line scans for each 578 condition. 579

Flow cytometry 580
Cells were grown, treated with si RNAs and S phase arrested as described above in 6 581 well plates (inoculum of 64,000 cells/well). To collect cells, media, PBS wash and trypsin 582 dissociated cells were combined into one tube per condition, spun and washed with PBS and 583 resuspended in Krishan stain (3.8mM Sodium citrate, 69nM Propidium iodide, 0.01% NP40, 584 0.01mg/mL RNaseA) (Krishan, 1975) and incubated at 4°C until run and analyzed by the 585

University of Colorado Cancer Center Flow Cytometry Shared Resource on a Beckman Coulter 586
FC500 flow cytometer. Cell cycle analysis was performed using ModFit LT software (Verity 587 Software House, Topsham, ME). 588

RNA isolation, RT-PCR, and library preparation 589
RNA was isolated using the manufacturers recommended protocols in three ways for 590 the three replicates averaged in Figure 4C. Cytoplasmic RNA was isolated using the RNeasy 591 Minikit (Qiagen) and the supplementary protocol provided by the manufacturer that includes a 592 cell lysis step using buffer RLN and a step to separate nuclei from the cytoplasm by gentle 593 centrifugation. RNA was also isolated using the Monarch Total RNA isolation miniprep kit (New 594 England Biolabs). RNA was also extracted using the TRI Reagent (Thermo Fisher Scientific). 595 cDNA was generated using 1 g RNA, random hexamers (Thermo Fisher Scientific SO142) and 596 SuperScript III Reverse Transcriptase (Thermo Fisher Scientific 18080044). Primers for PCR were 597 SON is required for centriole assembly.   The last tab shows the 30 CC GO terms from this dataset when the genes from the centrosome 949 portion of the CCDB and genes from the centriolar satellite proximity screen are included as 950 their own terms as determined by p values < 0.05. 951 952   Table S4. The intersection between centrosomal components or satellite components and 953 genes alternatively spliced when SON is depleted. Genes from the centrosome portion of the 954 CCDB (tab 1) and genes from the centriolar satellite proximity screen (tab 2) were cross 955 referenced with genes requiring SON for proper splicing and ordered based on differential 956 expression. Differential expression was determined by the DESeq2 software package, and 957 results for alternative splicing as determined by the MAJIQ software package or the rMATS 958 software package are included. 959 960 Movie S1. Serial tomographic slices through a volume containing an siControl rosette. The 961 tomographic volume as built from three, serial 250 thick sections. This centrosome is displayed 962 in Figure 3B and Figure Supplement 7A (A'). The centrosome is close to the nucleus and 963 numerous nuclear pore complexes can be observed in the serial tomographic slices. The model