Human Rhinovirus 16 impacts cilia structure in 3D cultured primary bronchial epithelial tissue through alternative splicing of host cilia RNAs

24 Human rhinoviruses (HRV) are a leading cause of the common cold but can often lead to 25 respiratory complications such as wheeze in young children. In a transcriptomic study of 26 respiratory nasal swab specimens from children hospitalised with acute wheeze, a significant 27 alteration was found in the expression of the serine/arginine rich splicing factor (SRSF) 28 kinase, SRPK1, between HRV positive children with acute exacerbations and HRV negative 29 controls. As this kinase can regulate host RN A splicing, we hypothesised that HRV infection 30 could dysregulate the expression of host mRNAs to affect antiviral functions or to alter the 31 morphological features of the infected resp iratory epithelium. Here, we show that 32 pharmacological inhibition of SRPK1 in prim ary bronchial epithelial cells resulted in 33 increased HRV16 replication while overexpressi on of SRPK1 reduced viral replication. In a 34 primary bronchial epithelial 3D model infe cted with HRV16 decreased phosphorylation of 35 SRSF1, 3 and 6 was observed. Furthermore, transcriptomic and alternative splicing (AS) 36 bioinformatic analysis revealed the significantly altered AS of 1228 host genes during 37 infection. Subsequent pathway analysis revealed the enrichment of most of these genes in 38 networks related to cilia development and function. HRV16 infection led to significantly 39 decreased cilia length and total cilia numbers in the primary bronchial epithelial 3D model 40 together with changes to selected cilia proteins. Overall, this investigation has unravelled 41 novel cellular networks implemented during HRV infection that may lead to acute 42 exacerbations of respiratory infections. 43 44 Author summary 45 Human rhinoviruses cause the common cold. In immunocompetent individuals this is usually 46 a self-limiting infection. However, in young children and the elderly, infection can lead to 47 complications such as bronchiolitis, croup, and wheezing. Rhinovirus infection can 48 exacerbate chronic conditions such as cystic fi brosis, chronic obstructive pulmonary disease 49 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 3 and asthma. Understanding the molecular pathol ogy of this exacerbation could lead to new 50 avenues for therapy. In this study, we discovered that a multifunctional cellular enzyme 51 called serine arginine protein kinase 1 (SRPK1) is a restriction factor for human rhinovirus 16 52 (HRV16) infection. One key cellular function of SRPK1 is to regulate RNA splicing through 53 modifying the SR proteins that normally enhanc e splicing. In three dimensional tissues 54 grown from human bronchial epithelial cells, we found that HRV16 infection led to decreased 55 levels of modified SR proteins. This change resulted in significant alterations in RNA 56 expression in the infected cell. Most of these alterations affected production of the correct 57 versions of cilia proteins resulting in reduced cilia numbers and cilia blunting. This type of 58 damage due to HRV infection would result in inefficient clearance of subsequent viral 59 infections prolonging the viral infection leading to lower respiratory tract infection and to 60 exacerbations of existing chronic disease. 61 62 63 64 65 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 4

66 Human rhinoviruses (HRVs) are positive-sense single stranded RNA enteroviruses within 67 the Picornaviridae family, collectively comprising more than 160 genotypes grouped into 3 68 species (A, B and C) [1]. HRVs are placed amongst the most frequent human infectious 69 agents, causing more than half of upper respiratory tract infections globally, though evidence 70 now includes the lower respiratory tract in t heir niche [2]. They are causative agents of the 71 common cold, a self-limiting infection with mild symptoms in immunocompetent individuals. 72 However, HRV infection can lead to severe respiratory complications in 73 immunocompromised groups such as pre- school children and the elderly. These 74 complications are usually associated with the migration of the infection into the lower 75 respiratory tract and include bronchiolitis, croup, wheezing as well as the exacerbation of 76 chronic conditions such as cystic fibrosis, chronic obstructive pulmonary disease and asthma 77 [2, 3]. 78 79 Strong evidence associating HRV and asthma exacerbations has been accumulating, with 80 HRV shown to account for more than 50% of total exacerbations and common cold 81 complications in asthmatics costing about 60 billion USD annually [1]. Additionally, HRV-82 induced wheezing in early age is linked with asthma development in adulthood, while 83 offspring of atopic (hyperallergic) mothers are more susceptible to HRV infections [4]. 84 Despite all the evidence associating HRV with asthma, molecular mechanisms underlying 85 this pathophysiology remain unclear. 86 87 The Mechanisms of Acute Viral Respiratory Infections in Children (MAVRIC) study 88 conducted in Perth, Australia, aimed to investigate further the molecular mechanisms 89 underlying asthma pathogenesis in the context of HRV infections. In this study, nasal swabs 90 containing nasal epithelial cells along with a variety of immune cells including neutrophils 91 and peripheral blood mononuclear cells were coll ected from pre-school children hospitalised 92 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 5 with acute wheezing; these were subjected to microarray analysis comparing expression 93 profiles between uninfected controls and HRV infected patients [5-8]. Surprisingly, for an 94 RNA virus infection operating in the cytoplasm, genes involved in the pre-mRNA splicing 95 process showed altered RNA expression. 96 97 It is well-established that HRV proteases 2A pro and 3C pro can degrade nuclear proteins and 98 that proteases from the three different HR V species degrade different substrates [9]. Key 99 substrates for HRV proteases are the nucleoporins that make up the nuclear pore complex 100 (NPC) [9-11]. Degradation of nucleoporins re sults in mis-localisation of heterogenous 101 ribonucleoprotein particles (hnRNPs) and SR sp licing factors from the nucleus to the 102 cytoplasm [12, 13]. It has been suggested that this virus-mediated relocation of splicing 103 factors aids viral replication by supplying RNA genome-binding proteins [14] or by regulating 104 virus mRNA translation [13, 15]. 105 106 There are nine classical SR proteins (serine-arginine-rich splicing factors (SRSFs) 1-9). 107 They all contain an N-terminal RNA rec ognition motif (RRM) and a C-terminal serine-108 arginine-rich (RS) domain [16]. Some SR proteins such as SRSF1 and SRSF6 possess an 109 additional pseudo-RRM [17]. SRSFs are found predom inantly in the nucleus but some (e.g. 110 SRSF1) can dynamically shuttle to and from the cytoplasm [16]. SRSFs are essential 111 regulators of constitutive and alternative sp licing. They bind exonic or intronic sequence 112 enhancers to define exon-intron boundaries and stabi lise formation of the spliceosome at 113 these boundaries to enhance splicing [18, 19]. A role for virus-associated splicing regulation 114 in respiratory disease was strengthened when the SRSF6 gene was found to be upregulated 115 in equine airway smooth muscle cells from asthmatic horses [20]. Several viruses of different 116 Baltimore classification groups, for example, human papillomavirus [21], human 117 immunodeficiency virus [22], influenza A viru s [23], and alphavirus [24] have evolved to 118 utilise or control SR protein family members and the host splicing machinery. SR proteins 119 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 6 are also known to exert regulatory functions beyond splicing in the nucleus, including nuclear 120 export, cytoplasmic stability and translation [25, 26]. 121 122 SRPK1 is a moonlighting protein involved in nu merous intracellular signalling pathways [27]. 123 It has been shown to regulate innate immunity to viral infection [28]. However, it is a key 124 regulator of cellular splicing. The RS domain of SR proteins is subject to phosphorylation by 125 kinases including serine/arginine protein kinases (SRPK) and CDK-like kinases (Clk) [29, 126 30]. This post-translational modification regulates both the function and subcellular 127 localisation of SRSFs [31, 32]. SRPK is normally present in the cytoplasm of cells where it 128 phosphorylates newly synthesised SRSFs, to licenc e their entry into the nucleus [33]. There 129 are three SRPKs in human cells (SRPK1, 2 and 3). Only SRPK1 and SRPK2 are expressed 130 in epithelial cells. In the nucleus, SRPK1 interacts with Clk1 to promote splicing [30]. The 131 dynamic phosphorylation of the SRSF RS domains by SRPK1 governs their levels, activity 132 and cellular localisation [31, 34-37]. Not all SR proteins are equally affected by SRPK1 133 activity. For example, SR proteins e.g. SRSF1, with two RRMs may be phosphorylated to 134 control splicing in a different manner from t heir single RRM–containing counterparts e.g. 135 SRSF3 [38]. 136 137 Here we show SRPK1 activity on SR proteins is reduced during HRV16 (a variant of HRV-A) 138 infection of primary epithelial cells and that SRPK1 is a restriction factor for HRV16 infection. 139 RNA-Seq analysis revealed significant host transcriptome changes between HRV16-infected 140 and mock-infected 3D cultured primary bronchi al epithelial cells similar to those found 141 previously [39-43]. However, alternative spli cing analysis of the RNA-Seq data revealed that 142 splicing of RNAs involved in cilia structure and function was significantly altered. HRV16 143 infection impacted expression of key cilia co mponents. This may indicate a sophisticated 144 viral mechanism of host cell disruption duri ng infection. The data suggest that HRV16 may 145 regulate splicing of cilia-related RNAs leading to altered mucociliary clearance and ultimately 146 prolonging productive viral replication. 147 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 7 148 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 8

149 Viruses stock generation 150 A variant of HRV-A, HRV16, was used for this study. Viral stocks obtained from ATCC 151 (ATCC VR-283), were propagated in HeLa Ohio cells at 33°C, seeded at 70-80% confluence 152 and grown in Dulbecco's modified Eagle medium (DMEM) with 10% foetal bovine serum and 153 1% penicillin/streptomycin (Thermo Fisher Sc ientific, UK). Infected cell lysates were 154 centrifuged at 120 x g for 7 min at 4°C. The supernatant was filtered through a 0.22 μ M pore 155 and stored at -80°C. Viral titres were determined via TCID50 assays using HeLa Ohio cells in 156 96-well plates. Cells were inoculated for 4 hours wi th serially diluted virus. Inoculates were 157 washed off and cells were incubated for 5 days at 33°C. Cell death was recorded, with each 158 dilution tested in quadruplicate and viral titre calculated using the Spearman & Karber 159

[44]. 160 161 Cell growth 162 Normal Human Bronchial Epithelial cells (HBECs) were purchased from Lonza (Basel, 163 Switzerland # CC-2540S). The cell donor was a female Caucasian aged 16 years old, with 164 no known disease or smoking history and a BMI of 22. Cells were grown in PneumaCult-Ex 165 plus medium (Stem Cell Technologies, Cambridge, UK) in T75 flasks for 2D culture. For air-166 liquid interface 3D culture, cells were seeded on 6.5 mm transwells with 0.4 µm pore 167 polyester membranes (Corning, Berlin, Germany). When cells reached full confluence, airlift 168 was performed by aspirating the apical medium and replacing the basal medium with 169 PneumaCult ALI maintenance medium (Stem Cell Technologies, Cambridge, UK). Medium 170 was replaced 3 times per week, with initial mucus production at approximately 2 weeks post-171 airlift. Mucus was washed off once per week and tissues reached full differentiation by 4 172 weeks post-airlift. 173 174 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 9 175 HBEC infections 176 A mucus wash and basolateral medium change were performed. Cultures were incubated 177 for one hour at 33 oC while virus stocks were thawed on ice. Based on calculations of an in 178 vivo infectious dose [40] the inoculum (3x10 6 pfu) was added in 100 µl medium apically. The 179 same volume of medium was added to the top of mock-infected control cultures. Tissues 180 were incubated for 3.5 hours at 33 oC then the inoculum was aspirated off and tissues were 181 washed apically three times with PBS without Ca 2+ and Mg 2+ to remove non-internalised 182 virus. Tissues were then incubated 33oC for the times stated in the experiments. 183 184 SRPK1 overexpression, depletion and SRPIN340 inhibition 185 The SRPK1 (transcript variant 1) human cDNA clone (untagged) (OriGene,Herford, 186 Germany, #RC205315) was transfected using Lipofectamine2000 (Thermo Fisher Scientific, 187 UK) at a concentration of 200ng/ml for 48 hours. SRPK1 was depleted by transfecting 188 Dharmacon SMART-Pool siRNAs in RNAiM ax transfection reagent (Thermo Fisher 189 Scientific, UK) into bronchial epithelial cells. siGLO (Dharmacon, # D-001630-01) was used 190 as a non-target siRNA control and to monitor transfection efficiency. SRPIN340 (Sigma 191 Aldrich, UK, #5.04293) was dissolved in DMSO to 20mM stocks and was administered at 192 20μ M for 48 hours. Both treatments were adminis tered to primary bronchial epithelial cells 193 seeded at 60% confluence. 194 195 RNA extraction and RT-qPCR 196 2D cultures: Cells grown in 6-well plates were lysed in 500 μ l of Trizol reagent (Thermo 197 Fisher Scientific, UK) and stored at -20°C. Upon thawing cells were scraped into Trizol and 198 RNA was isolated according to the manufacturer’s instructions. 3D cultures: Tissues were 199 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 10 flash frozen in liquid nitrogen and stored at - 80°C. Upon thawing cells were vortexed to 200 detach from the transwell membrane then tissu es were ground to a fine powder under liquid 201 nitrogen using a mortar and pestle and RNA wa s isolated using the RNeasy extraction kit 202 (Qiagen, Germany) according to the manufacturer’s instructions. RNA was quantified using a 203 NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific, UK, #ND-2000). 204 cDNAs were synthesised using the Maxima First Strand cDNA Synthesis kit with DNase 205 digestion according to the manufacturer’s instru ctions (Thermo Fisher Scientific, UK). 20 μ l 206 reactions were prepared using the Takyon™ ROX Probe 2X MasterMix dTTP blue 207 (Eurogentec, Camberly, UK), primers and probes at 300 and 100 nM respectively and 208 nuclease-free water. Primer/probe sets used are listed in Table 1. 209 Reactions were run on an ABI 7500 thermocycler (Thermo Fisher Scientific, UK) at this 210 profile: 95°C (5 min), 60°C (15 sec), 72°C (3 min), 40 cycles. Data analysis was performed 211 using the 7500 v2.3 (Thermo Fisher Scientific, UK) software. Ct values were determined 212 relative to GAPDH as the reference target gene. 213 214 Protein extraction and western blotting 215 Cells grown in 6-well plates were scraped in 400 μ l 2X Bolt LDS buffer (Thermo Fisher 216 Scientific) containing PhosphoSTOP (Merck, UK catalogue # 04693116001) and complete 217 miniprotease inhibitor cocktail (Merck, UK, catalogue # 200-664-3) in PBS. 20 µl of sample 218 was loaded per lane on Bolt 4-12% Bis-Tris polyacrylamide gels (Thermo Fisher Scientific, 219 UK) and electrophoresed at 150 V for 60 minutes. Prot eins were transferred to nitrocellulose 220 membranes using the iBlot2 Dry Blotting System (Thermo Fisher Scientific, UK). Membranes 221 were blocked in 5% (w/v) milk powder in PBS containing 0.01% (v/v) Tween. PBST at room 222 temperature for 1 hour, then washed in PBST (3 x 7 minutes) and incubated with primary 223 antibody in 5% milk powder in PBST for 1 hour at room temperature or overnight at 4°C with 224 rotation. Following incubation, membranes were washed in PBST (3 x 7 minutes) and 225 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 11 incubated with secondary antibody in 5% mi lk powder in PBST for 1 hour at room 226 temperature in the dark. A final round of washes was performed in PBST (2 x 7 minutes) and 227 PBS (2 x 7 minutes), and bands were visualised on the LI-COR Odyssey CLx Infrared 228 imaging system. Primary antibodies were SRPK1 1:300 (1:500, clone G211-637 BD 229 Transduction Laboratories, catalogue #611072), SRSF1 1:2500 1:1000, Mab96, Thermo 230 Fisher Scientific, catalogue # 32-4500), SRSF 3 (1:300, Life Technologies, UK, catalogue 231 #334200 ), SRSF6 (1:300 Abcam, UK, catalogue #ab140623, GAPDH (1:1000, Meridian Life 232 Sciences, UK, catalogue #H86504M, clone 6C5), HRV16 VP0/VP2 (1:250, QED 233 Biosciences, Aachen, Germany, catalogue # 18758). mAb104-detecting phosphorylated 234 SRSFs was prepared from hybridoma (AT CC CRL-2067) supernatants and was used neat 235 with 5% milk powder and 0.01% Tween. Secondary antibodies were goat anti-rabbit Dylight 236 800 conjugate (1:2000, Thermo-Fisher, UK, catalogue #SA5-35571), goat anti-mouse 237 Dylight 800 conjugate (1:2000, Thermo-Fisher, UK, catalogue #SA5-35521) and IRDye anti-238 mouse 800CW (1:2000, IRDye Licor Biosciences Ltd, UK, catalogue #926-32210). 239 Membranes were imaged on an Odyssey Infrared Im ager (LiCOR). The intensity of protein 240 bands was quantified using Odyssey Image Studio software. Protein levels were determined 241 and normalised to the level of the endogenous control (GAPDH). 242 243 Formalin fixed paraffin embedded (FFPE) sample preparation 244 Primary bronchial epithelial 3D cultures were fixed by fully submerging in 10% (v/v) buffered 245 formaldehyde (BNF) at room temperature overni ght. The cultures were submitted to the 246 Veterinary Diagnostic Services (University of Glasgow) for paraffin embedding, and 247 haematoxylin and eosin (H&E) staining. 248 249 Immunofluorescence microscopy 250 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 12 Antigen retrieval for 4µm sections from formalin-fixed, paraffin-embedded samples was 251 carried out in 10 mM sodium citrate buffer, pH 6.0 using a Menarini Access Retrieval Unit, at 252 1100C on full pressure for 10min. (Veterinary Diagnostic Services, University of Glasgow). 253 Microscope slides were washed sequentially six times in PBST. Slides were blocked in 10% 254 (v/v) filtered donkey serum in PBS for 1 hour at room temperature. Sections were incubated 255 with primary antibody diluted in 5% donkey se rum in PBS for 2 hours at room temperature. 256 Slides were washed and sections were incubated with secondary antibody diluted in 5% 257 donkey serum in PBS for 2 hours at room tem perature under dark conditions. Slides were 258 mounted in ProLong Gold Antifade Mountant with DAPI (Thermo Fisher, UK, catalogue # 259 P36931) and visualised on a ZEISS LSM 710 conf ocal microscope. Images were acquired 260 using the ZEN Blue software. Primary antibodies were HRV16 VP0/VP2 (1:100, QED 261 Biosciences, Aachen, Germany, catalogue # 18758), β -tubulin (1:250,Merck, UK, catalogue 262 #AB9354) and TMEM67 (1:250, Proteintech, UK, catalogue #Ag5009). Secondary antibody: 263 donkey anti-mouse Alexa-fluor 555-labelled anti body (1:1000 Thermo Fisher Scientific, UK, 264 catalogue #A-31570). 265 266 RNA sequencing, differential expression and pathway analysis 267 3D bronchial epithelial cultures were infected (3x10 6 pfu) at four weeks post-airlift and 268 harvested 48 hours post-infection. Total RNA wa s prepared from 3 biological replicates per 269 condition and sequenced in-house. Eluted RNA was quantified using a NanoDrop 2000 270 Spectrophotometer (Thermo Fisher Scient ific, ND-2000) and quality controlled on a 271 TapeStation (Agilent Technologies, G2991AA). All samples had a RIN score of ≥ 9. One 272 microgram of total RNA was used to prepare libraries for sequencing using an Illumina 273 TruSeq Stranded mRNA HT kit (Illumina, #20020594) and SuperScript2 Reverse 274 Transcriptase (Thermo Fisher Scientific, #18064014) according to the manufacturer’s 275 instructions. Libraries were pooled in equi molar concentrations and sequenced using an 276 Illumina NextSeq 500 sequencer (Illumina, #FC-404). RNA-Seq reads were analysed for 277 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 13 quality using FastQC (version 0.11.9) and reads were trimmed of adaptor sequences and 278 low-quality bases using Trimgalore 279 (https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/). The trimmed reads were 280 aligned to the human genome GRCh38 ( Ensembl) using Hisat2 (version 2.2.0) [45]. 281 FeatureCounts (Version 2.0.1) [46] was used to quantify reads mapping to gene annotation 282 files. Reads counts were normalized to counts per million (CPM). The edgeR package was 283 used to calculate the gene expression level an d to analyse differentially expressed genes 284 between sample groups. RNA-Seq data sets are freely available from the European 285 Nucleotide Archive accession number PRJEB88791, and heat maps were generated in 286 GraphPad Prism (version 9). Pathway analysis was performed using the Ingenuity Pathway 287 Analysis (IPA) tool using the rat, mouse, human and undefined species data, in all cells and 288 tissues. Only experimentally observed data were selected. 289 290 Over representation analysis of alternative splicing events. 291 Bam files from the RNA paired-end sequencing were sorted by co-ordinate, indexed and 292 subject to SplAdder analysis [47] to measure and quantify alternative splicing events. 293 Percentage spliced in (PSI) values were quantified for each splicing event, and a two-tailed 294 student's t-test was performed for values fr om mock-infected and HRV16-infected primary 295 bronchial epithelial cells to determine the most significantly differentially spliced genes. 296 Pathway analysis was performed using Webgestalt ( http://www.webgestalt.org/) [48]. Over 297 representation analysis was carried out test for biological processes using Benjamini-298 Hochberg multiple testing adjustment. Sashimi plots were generated using MISO 299 (https://pypi.org/project/misopy/0.5.4/) [49]. 300 301 Cilia count 302 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 14 Five sections from three biological replicates per condition were randomly selected and cilia 303 were manually counted from images obtained at x20 magnification from two technical 304 replicates per condition using a tally meter from FFPE H&E-stained samples. Image J ( 305 https://imagej.net/software/fiji) was used to quantify the cilia length by measuring the 306 proportion of distance in pixels. Ten cilia were measured in each section for each condition. 307 Statistical analysis 308 Statistical analyses were carried out using a students’ t-test. P-values of < 0.05 were 309 considered statistically significant. 310 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 15

311 SRPK1 restricts HRV16 infection in primary epithelial cell culture. 312 The kinase SRPK1 was found to be up-regulated at the mRNA level in the MAVRIC study 313 which compared nasal swab samples from HRV-infected young children with wheezing 314 exacerbations to HRV-negative controls to [7 , 50]. However, when the study population was 315 divided into two different phenotypes based on a Th1/type 1 interferon response versus a 316 Th2/IFNγ response, SRPK1 mRNA was significantly downregulated in the second population. 317 The two phenotypes had quite different clinical characteristics. In the case of a Th2/IFN γ 318 response, illness progressed more slowly but t here was a greater chance of hospitalisation 319 and repeat infections/exacerbation of diseas e [7] suggesting that SRPK1 downregulation 320 was associated with more severe disease. 321 However, SRPK1 activity, as opposed to SRPK1 levels, is regulated by key cell signalling 322 pathways such as CK2 and Akt, which can be impacted upon virus infection [51]. To find out 323 more about the relationship between HRV infection and SRPK1, we overexpressed the 324 kinase in human bronchial epithelial cells (HBECs). We also inhibited the kinase by treating 325 cells with 20µM SRPIN340, a specific inhibitor of the kinase activity of SRPK1 [52]. Each 326 treatment was carried out 48 hours prior to HRV16 infection. We examined changes due to 327 the treatments in the logarithmic phase of vi ral production (MOI=3, 16 hours post infection, 328 Supplementary Fig. 1A). Supplementary Fig. 1B shows a significant increase in SRPK1 329 mRNA levels in overexpressing cells relative to mock-transfected cells. The effectiveness of 330 the SRPIN340 treatment was shown through decreased phosphorylation of SRSF1 331 (Supplementary Fig. 1C) and SRSF6 (Supplementary Fig. 1E) compared to mock-treated 332 cells. Levels of phosphorylated SRSF3 were not significantly decreased by SRPIN340 333 treatment (Supplementary Fig. 1D). No significant change in levels of phosphorylated 334 SRSFs was detected when SRPK1 was overexpressed (SRPK1 OE) (Supplementary Fig. 335 1C-E). 336 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 16 We assessed how inhibition and overexpression of SRPK1 impacted HRV16 infection. The 337 levels of viral RNA, tissue-released virus parti cles and expression of viral capsid proteins 338 VP0 and VP2 were compared between mock-treated, SRPIN340-treated and SRPK1-339 overexpressed HBECs infected with HRV16 for 16 hours at an MOI=3 (Fig. 2). SRPK1 340 inhibition led to increased HRV16 replication, as reflected by the increased viral RNA 341 production (Fig. 2A, SRPIN340) although this wa s not statistically significant (p=0.07). 342 However, a statistically significant increase was observed upon kinase inhibition in tissue-343 released virus particles (p<0.05) (Fig. 2B, SRPIN340) and virus capsid VP2 protein 344 production (Fig. 2C, D, SRPIN340). In cont rast, SRPK1 overexpression did not have a 345 significant effect on viral genome replication when compared to untreated controls (Fig. 2A, 346 SRPK1 OE). SRPK1-overexpressing HBECs showed significantly decreased virus shedding 347 (Fig. 2B, SRPK1 OE) and viral capsid protein VP0 production (Fig 2C, D, SRPK1 OE) 348 compared to untreated samples, thus displaying an opposite effect from respective SRPK1-349 inhibited samples. Taken together, these data s uggest that SRPK1 is a restriction factor for 350 HRV16 infection in primary epithelial cells, wi th the activity of SRPK1 repressing viral 351 replication, assembly and release, at 16 hours post-infection. 352 353 HRV16 infection alters phosphorylation of SR proteins in primary human bronchial 354 epithelial cells. 355 Next, to assess activity of SRPK1 during HRV16 infection we quantified changes in 356 expression of selected substrates of SRPK1 in HBECs. SRSF1, SRSF3 and SRSF6 were 357 selected for further study. SRSF1 is the prototypical SRSF protein [37, 53]. SRSF3 was 358 previously shown to be involved with the internal ribosome entry site (IRES)-dependent 359 translation of picornavirus mRNAs [54] and SRSF6 is known to be upregulated in equine 360 airway smooth muscle cells from asthmatic horses [20]. There was no significant change in 361 total levels of SRSF1 (Fig. 2A, Supplementary Fig. 2A) or SRSF6 (Fig 2C, Supplementary 362 Fig. 2A) during a 48 hour time course of infection (MOI=3). However, SRSF3 showed 363 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 17 significantly decreased expression at 16, 24 and 48 hours of infection (Fig. 2B, 364 Supplementary Fig. 2A). Next, we measured le vels of phosphorylated SR proteins using the 365 SR protein phosphor-specific antibody Mab104 [38]. Levels of phosphorylated SRSF1 were 366 significantly decreased at 4 and 48 hours post infection in HRV16-infected cells compared to 367 mock-infected cells (Fig. 2D, Supplementary Fig. 2B). There was no significant change in 368 levels of phosphorylated SRSF3 at any time point in HRV16-infected cells compared to 369 mock infected cells (Fig. 2E, Supplementary Fig. 2B). Finally, phosphorylated SRSF6 levels 370 were significantly decreased in HRV16-infected cells at 4 and 24 hours post infection (Fig. 371 2F, Supplementary Fig. 2B). Taken together, these data indicate that HRV16 infection 372 represses the expression of SRSF3 and inhibits phosphorylation of SRSF1 and SRSF6. The 373 downregulated phosphorylation seen for SRSF1 and SRSF6 suggests that the activity of 374 SRPK1 is repressed at early and late times of infection during HRV16 infection. 375 376 HRV16 infection in 3D cultured primary bronchial epithelial tissues inhibits SRPK1 377 protein levels and phosphorylation of SR proteins. 378 Next, we evaluated levels and activity of SRPK1 in HRV16 infection of air-liquid interface 3D 379 tissue cultures of HBECs. 3D cultures enable the differentiation of these cells, providing 380 increased physiological relevance in parameters such as cell polarisation, cilia formation, cell 381 to cell interactions and nutrient access [55]. HBECs were grown at the air-liquid interface in 382 transwell cultures over a four-week period (Supplementary Fig. 3 compare A to B). After this 383 time, the tissues were fully differentiated as indicated by cilia development (Supplementary 384 Fig. 3B, arrowhead) and a multi-layered pseudostratified epithelium was developed. The 385 tissues also produced mucus as indicated by periodic acid-Schiff (PAS) staining of the 386 tissues which detects polysaccharides (Suppl ementary Fig. 3, compare C to D). This 387 suggests the presence of goblet cells (S upplementary Fig. 3D, arrowheads) further 388 confirming tissue differentiation. HRV16 (3x10 6 pfu) was applied to the surface of the 389 epithelium and infection was allowed to proceed for 48 hours. Infection induced evident 390 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 18 mechanical damage to the structure of the epithelium (Supplementary Fig. 3 compare E and 391 F, arrows in F shows regions of damage). However, the extent of damage may not be due 392 only to HRV16 infection but also due to tissue handling for preparation of formalin-fixed 393 paraffin-embedded blocks. Immunofluorescence staining to detect the virus capsid protein 394 VP2 indicated that infection was established in all the layers of the epithelium 395 (Supplementary Fig. 3 compare G and H). West ern blot analysis indicated a significant 396 decrease in SRPK1 and phosphorylated SRSF1, SRSF3, and SRSF6 protein levels in 3D 397 HBEC tissues infected with HRV16 (3x10 6 pfu) for 48 hours compare to mock-infected 398 tissues (Fig. 3). 399 400 Transcriptomic analysis confirms increased expression of innate immune genes 401 during HRV16 infection of 3D cultured primary bronchial epithelial tissues. 402 RNA sequencing was performed on HRV16 and mock infected primary bronchial epithelial 403 3D cultured tissues. Tissues were mock-infected or infected (3x10 6 pfu) for 48 hours at 4 404 weeks post-airlifting. Total RN A extracts were prepared from triplicate HRV16-infected and 405 mock-infected HBEC 3D air-liquid interface cultures. The counts per million (CPM) of 406 individual transcripts were similar between infected (median=33.1 CPM per transcript) and 407 mock infected tissues (median=30.2 CPM per tr anscript). 82.2% of bases achieved a quality 408 score of Q30 and Illumina software was used to assign sequencing reads to their 409 corresponding samples. Differentially expressed genes (DEGs) between infected and mock-410 infected samples were identified by mappi ng sequencing reads to the reference human 411 genome and quantifying the mapped reads of individual transcripts. 412 Analysis indicated that there were 4034 DEGs due to HRV16 infection (false discovery rate 413 (FDR)<0.05, Benjamini-Hochberg correction). 2143 genes were up-regulated, and 1891 414 genes were down-regulated in infected when compared to mock-infected tissues 415 (Supplementary Table 1). The most statistically significant differences in gene expression 416 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 19 during infection were seen in up-regulated genes as illustrated in the volcano plot in Fig. 4A 417 (red dots). Key innate immune gene changes are indicated similar to those found by Ong et 418 al [43]. The violin plot in Fig. 4B shows t hat the distribution of gene expression was similar 419 between infected and mock-infected samples. However, more genes showed a higher level 420 of expression (increased reads per million) in control mock-infected (Ctrs) compared to 421 infected (HRV16) samples (Fig. 4B). 422 Fig. 4C shows a heat map of the top 20 genes that were up- or down-regulated due to 423 HRV16 infection. The most up-regulated genes were involved in anti-viral defence, primarily 424 the type I interferon signalling pathway (Fig. 2A arrows, Fig. 4C, Table 2), demonstrated by 425 the strong up-regulation of these genes in HRV 16-infected samples, across all biological 426 replicates as previously reported [39, 43, 56, 57]. On the other hand, the functions of the 427 most down-regulated genes during infection were very variable, but included genes involved 428 in epithelial homeostasis, metabolic processes, and cell signalling (Fig. 4C). Although 429 SRPK1 was significantly downregulated at the protein level in HRV16 infected primary 430 bronchial epithelial 3D cultured tissues (Fig . 3), SRPK1 mRNA expression was found to be 431 significantly upregulated by 1.3-fold. Protein- protein interaction networks were constructed 432 using Gene Ontology (GO) enrichment analysis of up- (240) or down-regulated (9) DEGs 433 with log2 fold change >2 or <-2 (Fig. 4D). The antigen presentation pathway was the most 434 enriched biological process. ISG15 protein conjugation, apoptosis and type 1 interferon 435 signalling were also identified as biological pathways induced by HRV16 infection. Molecular 436 function enrichment indicated HRV16-inducti on of antiviral OAS activity, chemokine 437 receptors, and antigen processing (Fig. 4D). 438 Over-representation analysis was performed to identify cellular pathways affected by HRV16 439 infection. A subset of DEGs was selected for this pathway analysis, with log2-fold change 440 values within the range of 1 to -1. This cut-off point included 655 DEGs of which 466 were 441 up-regulated and 189 down-regulated during infection (Supplementary Table 2). Pathway 442 analysis mapped the vast majority of the DEGs in innate immune pathways, with the 443 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 20 interferon signalling pathway scoring the highes t enrichment value (Supplementary Fig. 4). 444 However, genes involved in epithelial cell junctions (claudins and cadherins) and the 445 epithelial barrier (repetin, involucrin) were also downregulated as expected [58] 446 (Supplementary Table 2). Table 2 lists the to p thirty upregulated protein-coding genes of 447 known antiviral functions. 448 449 HRV16 infection of 3D cultured primary bronchial epithelial tissues leads to changes 450 in the splicing of cilia-related genes 451 Since we had shown that HRV16 infection alters splicing factors, we wanted to assess 452 whether the splicing of host pre-mRNA transcripts was altered during infection. We used the 453 alternative splicing toolbox SplAdder [47] to perform further analysis on our RNA-seq data. 454 Using SplAdder, we identified the total number of single splicing events in our dataset, along 455 with the alternative splicing (AS) mechanism of each particular event. We identified 10952 456 genes with alternative splicing events, with each event categorized by its splicing 457 mechanism (e.g., exon skipping, intron retention etc.). The differential splicing between 458 infected and mock-infected samples was determined by comparing the percentage spliced-in 459 (PSI) output values given for each transcript. PSI indicates the efficiency of splicing of a 460 particular exon into the transcript population of a gene, with values ranging from 0 to 1. 461 Therefore, different PSI values for a give n transcript between infected and mock-infected 462 samples, indicate a different AS pattern for that respective gene during infection. 10514 out 463 of 10952 genes (96%) had a different PSI value in infected compared to mock-infected 464 samples. Following a 2-tailed student’s t-test analysis, this number was limited to 1228 465 genes that were significantly differentially spliced due to HRV16 infection (11% of all the 466 alternatively spliced genes in the dataset). 467 1469 single splicing events specifically associated with HRV16 infection were identified as 468 being involved in producing mRNAs as some were alternatively spliced via more than a 469 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 21 single mechanism, of which about 40% were exon skipping events (Fig. 5A). The rest of the 470 splicing events identified included alternative 3’ sp lice sites (23%), alternative 5’ splice sites 471 (18%), intron retention (10%), multiple exon skipping (8%) and mutually exclusive exons 472 (2%) (Fig. 5A). To interpret the biological significance of this HRV16-induced effect on 473 splicing, we performed a pathway analysis using WebGestalt on the 1228 differentially 474 spliced genes. While response to interferon was identified as a major pathway altered by 475 HRV16 infection through alternative splicing, the functional enrichment analysis indicated 476 that most differentially spliced genes during HRV16 infection were involved in the 477 microtubule cytoskeleton required for cilia structure, ciliogenesis and cilia function (Fig. 5B). 478 To verify specific alternative splicing changes due to HRV16 infection seen in the functional 479 enrichment analysis we analysed three cilia-related genes whose alternative splicing is 480 known to be related to ciliopathies [59]. T hese were radial spoke head protein 1 (RSPH1), 481 which is located in the spokes of cilia and controls cilia motility [60]; intraflagellar transport 482 protein 74 (IFT74), which is involved in protein transport within cilia and is required for 483 ciliogenesis [61]; and transmembrane protein 67 (TMEM67), a ciliary transition zone protein 484 required for cilia structure [62]. Figure 6A indicates the positions of these proteins in cilia. 485 Sashimi plots showing the exon coverage from the RNA-Seq data were generated for the 486 three selected genes using MISO (Fig. 6B-D). There was a small decrease in gene coverage 487 for RSPH1 and TMEM67 indicating reduced transcription of the genes upon HRV16 488 infection. More importantly, clear differenc es in exon inclusion can be seen for each gene 489 (see arrowheads indicated on the maps below the gene coverage profiles) comparing mock-490 infected to HRV16-infected cells suggesting that HRV16 infection alters mRNA splicing of 491 these RNAs leading to mutations in the encoded proteins. 492 493 HRV16 infection leads to impaired cilia in 3D cultured primary bronchial epithelial 494 tissues 495 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 22 Based on the finding that HRV16 infection induced changes in the alternative splicing of 496 genes involved in cilia development and function, we wanted to assess whether infection 497 affected the numbers and morphology of cilia in our primary bronchial epithelial 3D model 498 (Fig. 7). Figure 7A shows a representative image of an H&E stained, mock-infected 3D 499 culture of HBECs. Figure 7B shows a repres entative image of an H&E stained HBEC tissue 500 infected with HRV16 (3x106 pfu) for up to 48 hours. HRV16-infected tissues showed reduced 501 cilia density, and the cilia appeared shorter than those on mock-infected tissues (enlarged 502 images in A and B). Cilia numbers were manually counted and, compared to mock infected 503 tissues treated in exactly the same manner, in fected tissues showing significantly decreased 504 cilia numbers at all infection timepoints investigated (Fig. 7C). The largest decrease was 505 recorded at 48 hours post infection, where infe cted samples had 4.5-fold less cilia compared 506 to mock-infected samples. Additionally, the average cilium length in these respective 507 samples was quantified (Fig. 7D). This analysis indicated that the average cilium length was 508 significantly decreased in infected compared to mock-infected samples at 24 and 48 hours 509 post infection. As with cilia numbers, the average cilium length showed the largest difference 510 at 48 hpi, where the average cilium length in infected samples was decreased by 1.5-fold 511 compared to mock-infected samples (Fig. 7D). 512 To confirm these data, we carried out immunofluorescence staining of sections of 3D 513 cultures of HBECs either mock-infected or HRV16-infected. We chose to use antibodies 514 against β -tubulin, a core cilium structural protein and TMEM67, a ciliary transition zone 515 protein located at the junctions of cilia to the plasma membrane. β -tubulin was observed 516 along the lengths of the cilia in mock-infected cells (Fig. 8 A, B) but levels of the protein were 517 reduced in HRV16-infected cells and cilia were not clearly seen (Fig. 8 C, D). Quantification 518 of β -tubulin staining showed a significant decrease at all time points, and this was especially 519 significant at later times of infection (Fig. 8 I). TMEM67 staining for mock-infected tissues 520 was found in cells in the mid layers and on the outer surface of the tissue, in discrete sub-521 cilia regions (Fig. 8 E, F). The location of staining was similar for HRV16-infected tissues, but 522 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 23 the staining was more diffuse (Fig. 8 G, H). Q uantification of staining showed a statistically 523 significant decrease in TMEM67 levels in infected tissues at all time points (Fig. 8 J). Taken 524 together these data suggest that HRV16 infection causes a reduction in cilia density and 525 changes in cilia structure in the respiratory epithelium. 526 527 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 24

528 Human rhinovirus (HRV) is the most common respiratory virus infecting humans. Repeat 529 infections in childhood can lead to wheezing and asthma, and these symptoms are 530 exacerbated by subsequent HRV infections. In addition, in individuals with chronic 531 obstructive pulmonary disease (COPD), HRV infe ction can lead to severe disease requiring 532 hospitalisation. Discovering cellular factors that lead to exacerbations of HRV infection is key 533 to designing novel therapies against these respiratory infections and conditions. This study 534 has revealed two novel aspects of the cellular response to HRV16 infection. First, we have 535 shown that the cellular splicing kinase SRPK1 is a restriction factor for HRV16 (a variant of 536 HRV-A) infection in primary bronchial epithelial cells. Second, changes in alternative splicing 537 cause by an HRV16 infection-induced down regulation of SRPK1 impact cilia structure and 538 function in 3D cultured primary bronchial epithelial tissues. 539 Transcriptomics studies in nasal scrapings [7, 39], bronchial scrapings [63] and nasal and 540 lung epithelial air-liquid interface 3D cultures [41, 43, 57] have shown that HRV16 infection 541 alters expression of immunity-related genes. Our transcriptomic analysis of HBEC air-liquid 542 interface cultures agrees with the conclusions of these studies. There was significant up-543 regulation of genes encoding proteins involved in anti-viral defence (e.g. OASL, MX2), and 544 the interferon signalling pathway (e.g. IFNL1, INFL2) and several chemokines (e.g. CXCL10, 545 CXCL11) known to be induced by HRV infection. Although SRPK1 is well known as a kinase 546 that controls splicing via phosphorylation of SR proteins [29], it also positively regulates 547 innate immunity to viral infections via NF- ĸ B and interferon gamma [28]. Therefore, SRPK1 548 down-regulation during HRV16 infection could be beneficial to the infectious process, as we 549 have demonstrated. 550 In an in vivo study of a mixed nasal cell population from children with wheeze infected with 551 HRV, overall SRPK1 mRNA expression was upregulated compared to HRV-negative 552 controls [7]. SRPK1 mRNA was also upregulated 1.3-fold after a 48 hour HRV16 infection of 553 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 25 3D HBEC tissues in this study. However, increased SRPK1 expression at the RNA level did 554 not lead to increased levels of SRPK1 protein. In fact, in 3D HBEC tissues infected with 555 HRV16 there was a significant decrease in SRPK1 levels and in phosphorylation of selected 556 SR proteins compared to mock infected tissues. This result agrees with the decreased 557 phosphorylation of SR proteins observed in a study where rhabdomyosarcoma cells were 558 infected with another enterovirus, enterovirus A71 [64]. Our data suggest that HRV16 559 represses the activity of SRPK1 during infect ion. This could be either through decreased 560 protein levels, as we have observed, or due to changes in cell signalling caused by HRV16 561 infection impacting SRPK1 kinase activity, which is itself controlled by Ck2/Akt-mediated 562 phosphorylation [27]. 563 However, inhibition of SRPK1 kinase activity has been shown to inhibit replication of viruses 564 such as HIV, hepatitis C and Sindbis virus [65-68] and it has recently been reported that 565 SRPK1 inhibition can reduced IRE-dependent trans lation and replication of enterovirus A71 566 [64]. These findings clearly suggest that SRPK1 is not a restriction factor of these viruses as 567 it is required for their life cycles. In contra st, both inhibition and overexpression of SRPK1 568 has been shown to reduced Ebola viral replication [69], while SRPK1 is a known restriction 569 factor for hepatitis B virus infection [70]. Thes e diverse findings indicate that viruses may 570 utilise SRPK1 in different ways for replication. Further, balancing activity of SRPK1 571 throughout virus infection may be key. For example, repression of SRPK1 stimulation of 572 innate immunity early in infection could facilitate viral infection while upregulation later in 573 infection could be required for viruses such as HIV and that require either cellular splicing 574 [22] or enteroviruses that require SRPK1 activi ty for stimulation of viral RNA translation [64] 575 to complete their life cycle. 576 Changes in expression of genes involved in cilia formation and function were found 577 previously in a study comparing the transcriptome of air-liquid interface cultures established 578 from the lungs of non-asthmatic versus asthmati c individuals [41] and changes were greater 579 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 26 following HRV infection. Gene ontology analysis of RNA sequencing data showed that cilia 580 function was found to be altered in human nasal epithelial cells infected with HRV16 [43]. 581 Stimulation of the antiviral immune res ponse by addition of poly(I:C) in human nasal 582 epithelial stem/progenitor cells also revealed c hanges to genes involved in ciliogenesis and 583 function [71]. In agreement with these studies, inspection of our DEG data from RNA 584 sequencing showed reduced expr ession of genes encoding dynein axonemal assembly 585 proteins, Bardet-Bield syndrome (BBS) genes important for cilia development and function, 586 intraflagellar transport proteins, and tubulins, structural components of cilia. 587 The hypophosphorylation of SR proteins due to HRV16 infection in 3D HBEC tissues that we 588 have observed would be expect ed to cause changes to cellular splicing [30, 72]. We 589 observed a major impact on alternative spli cing in HBEC tissues. Remarkably, when we 590 analysed changes to gene expression due to alte red splicing events, we discovered that 591 most of the changes affected splicing of mRNAs encoding proteins involved in cilia structure 592 and function. Interestingly, tissue-specific alter native splicing of cilia-related RNAs underlies 593 a wide range of ciliopathies [59]. This suggests that correct alternative splicing is key to 594 production of normal cilia proteins. Indeed, when we examined the morphology of the air-595 liquid interface cultures following HRV16 infection we found a blunting and loss of cilia 596 compared to mock infected tissues. We propose that a combination of transcriptional down-597 regulation of cilia-related genes together with specific changes in splicing of cilia-related 598 mRNAs could lead to malformed or damaged cili a. These morphological changes could lead 599 to increased susceptibility to recurrent HRV infections and/or the development of chronic 600 inflammatory disease. 601 There are several limitations to this study. Most importantly, the analyses were carried out 602 using HBECs from a single donor. We cannot rule out the possibility that data from this 603 donor may not reflect the general population. We studied virus infection of HBEC tissues at 604 48 hours post infection, and transcriptomic changes could be significantly different at other 605 time points. As we only analysed transcriptomic changes in bronchial epithelial cells, a 606 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 27 comparison of these changes between bronchial and nasal epithelial cells, grown as fully 607 differentiated 3D tissues would have been interesting to perform. It is known that HRV 608 infection causes redistribution to the cytoplasm of hnRNP proteins, known antagonists of SR 609 proteins in alternative splicing [73], and ec topic expression of the HRV16 3C protease has 610 also been shown to mislocate SRSF2, a splicing factor which we have not studied, in the 611 nucleus of transfected cells. This means that changes in SR protein phosphorylation may 612 only contribute to, rather than directly caus e, the changes in splicing induced by HRV16 613 infection. Finally, examination of changes in expression of SRPK1, and its effects on its 614 substrates, and any impact on ciliated epithelial cell function in HRV-C infected cells would 615 be important to consider in the future since HRV-C infection can have more serious 616 consequences, clinically. 617 618

619 This study has revealed the cellular kinase SRPK1 as a new restriction factor for HRV16 620 infection in epithelial cells. The mechanism of restriction is not known. However, since 621 SRPK1 can activate production of interfer on, interferon response factors and certain 622 chemokines [28], induction of innate immuni ty to viral infection seems likely. Indeed, 623 transcriptomic analysis showed up-regulation by HRV16 of many innate immune factors. 624 However, a key finding of this study is that HRV16 infection alters expression of splicing 625 factors and their phosphorylation. Importantly, we found major changes in alternative splicing 626 of RNAs encoding structural components of cilia. This suggests a molecular mechanism by 627 which HRV infection might result in damage to the mucociliary compartment which would 628

in inefficient clearance of subsequent viral infection enabling infection to move to the 629 lower airways, which would lead to exacerbations. 630 631

632 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 28 This work was funded by a BBSRC Industrial Case CSV Training Award no. BB/R505341/1. 633 Quan Gu is supported by an MRC award #MC_UU_00034/5. We are grateful to Chris 634 McRae at Astra Zeneca for facilitating agreem ents between Astra Zeneca and the University 635 of Glasgow and for arranging the industrial pl acement for student Chris Rozario (C.R.) in 636 Astra Zeneca. We thank especially Jenny Horndahl, at AstraZeneca, Sweden for hosting 637 C.R. during his industrial placement. We would like to thank the staff at the University of 638 Glasgow’s Veterinary Diagnostics service for carrying out paraffin embedding, sectioning 639 and staining of tissues. Prof Carl Goodyear, Director, the GLAZgo Discovery Centre helped 640 arrange the studentship and provided critical feedback throughout the study. 641 642 Declaration of Interests 643 Prof Maciewicz is retired from Astra Zeneca but at the start of the project she owned shared 644 in Astra Zeneca. Astra Zeneca had no input into the design of the study or interpretation of 645 the data. 646 647 Author contributions 648 C.R. Investigation, formal analysis. writing, original draft 649 Q.G. Investigation, formal analysis, data curation, visualisation. 650 A.S. Data curation, resources. 651 R.M. Conceptualisation, supervision, funding acquisition, writing review and editing. 652 S.V.G. Conceptualisation, supervision, funding ac quisition, visualisation, writing review and 653 editing. 654 655 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 29 656 Primer name Sequence GAPDH Forward 5’-GAAGGTGAAGGTCGGAGT–3’ GAPDH Reverse 5’-GAAGATGGTGATGGGATTTC-3’ GAPDH Probe 5’-FAM-CAAGCTTCCCGTTCTCAGCC–TAMRA–3’ HRV16 Forward 5’-CCCTGAATGTGGCTAACCTT–3’ HRV16 Reverse 5’-ACGGACACCCAAAGTAGTTG–3’ HRV16 Probe 5’-FAM-ACAATCCAGTGTGTAGCTGGTCGT-TAMRA–3’ SRPK1 Forward 5’-ACCCTCCAGGAATCTCTACTT-3’ SRPK1 Reverse 5’-CCATGCTTTGTTCATGCCTAT-3’ SRPK1 Probe 5’-FAM-ACTTCACCCTCTTGGGCCTTTCAT-BHQ-3’ 657 Table 1. List of primers and probes used in RT-qPCR experiments. 658 659 660 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 30 Log2-fold change Gene name Gene function 7.434 CXCL11 Chemokine 7.434 ZBP1 Innate sensor 7.223 IFNL1 Antiviral response 6.268 CXCL10 Chemokine 6.101 IFI6 Induced by interferon 6.085 OASL Innate immunity to viral infection 5.971 IFNL3 Antiviral response 5.912 IFNL2 Antiviral response 5.705 ISG15 Innate immunity to viral infection 5.631 MX2 Innate immunity to viral infection 5.559 IFITM1 Induced by interferon 5.423 IFIT1 Induced by interferon 5.239 IFI27 Induced by interferon 5.229 AIM2 Innate immunity to viral infection 5.049 IFIT3 Induced by interferon 5.043 EPSTI1 Induced by interferon 5.004 CXCL9 Chemokine 4.810 HERC5 Innate immunity to viral infection 4.672 DUOXA2 Antiviral protein in respiratory epithelial cells 4.447 CMPK2 Innate immunity to viral infection 4.370 CSF3 Controls production of granuolcytes 4.303 RSAD2 Antiviral protein 4.257 OAS3 Innate immunity to viral infection 4.238 XAF1 Induced by interferon 4.141 MX1 Innate immunity to viral infection 4.121 BST2 Innate immunity to viral infection .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 31 4.007 HERC6 Innate immunity to viral infection 3.938 OAS2 Innate immunity to viral infection 3.796 HLA-F Major histocompatability protein 3.768 IFI44 Induced by interferon 3.404 IFITM3 Induced by interferon 661 Table 2. List of top 30 immune-related genes up-regulated by HRV16 infection. 662 663 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 32 Figure legends 664 Figure 1. Changes in SRPK1 levels and activity alter HRV16 infection in primary 665 human bronchial epithelial cells. A. RT-qPCR analysis of tissue-associated HRV16 RNA 666 levels in untreated, SRPIN340-treated and (SRPIN340) SRPK1 overexpressing (SRPK1 OE) 667 HBECs infected with HRV16 (MOI=3) for 16 hours. The Log2 ΔΔ Ct values were calculated 668 using the values of the housekeeping gene GAPDH as the control target. B. TCID 50 669 quantification of tissue-released infective virus using supernatants from untreated, 670 SRPIN340-treated (SRPIN340) and SRPK1 overexpressing (SRPK1 OE) HBECs infected 671 with HRV16 (MOI=3) for 48 hours. C. Western blot analysis of the HRV16 capsid proteins 672 VP0/VP2 expression in mock-treated and mock-treated SRPIN340-treated or SRPK1 673 overexpressed (OE) cells infected with HRV16 (MOI=3) for 48 hours. A GAPDH loading 674 control is shown under the virus capsid protein blot. D. Quantification of western blot data 675 from three separate experiments. All data show the mean and standard error from three 676 separate experiments. ns=non-significant stat istical difference, *=p<0.05, **=p<0.005, 677 ***=p<0.0005 (student’s t-test, 2-tailed). 678 679 Figure 2. HRV16 infection of primary huma n bronchial epithelial cells downregulates 680 expression of SRSF3 and decreases phosphorylation of SRSF1 and SRSF6. 681 Quantification of western blot expression data relative to GAPDH expression of A. total 682 SRSF1, B. total SRSF3, C. total SRSF6, D. phosphorylated SRSF1, E. phosphorylated 683 SRSF3, and F. phosphorylated SRSF6 in mock-infected (Control) and HRV16-infected 684 (HRV16: MOI=3) HBECs at 4, 16, 24 and 48 hours post-infection. The data show the mean 685 and standard error of the mean from three separate experiments. Significant statistical 686 difference, *=p<0.05, **=p<0.005, ***=p<0.0005 (student’s t-test, 2-tailed). 687 688 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 33 Figure 3. HRV16 infection in 3D cultured primary bronchial epithelial tissues 689 significantly decreases the phosphorylation of SRSF proteins. A. Western blot analysis 690 of levels of SRPK1 and phosphorylated SRSF1, SRSF3, and SRSF6 (detected with 691 Mab104) in mock-infected (Mock) and HRV16-infected (HRV16) HBEC air-liquid interface 692 cultures at 48 hours post-infection (hpi) with 3x10 6 pfu HRV16. GAPDH is shown as a 693 loading control and was used to determine relative proteins levels. B. Quantification of data 694 from three separate experiments. The data s how the mean and standard error of the mean 695 from three separate experiments. ns=no signifi cant statistical difference. Significant 696 statistical difference, *p<0.05, **p<0.005 (student’s t-test, 2-tailed). 697 698 Figure 4. HRV16 induced changes in gene expression during infection in 3D cultured 699 primary bronchial epithelial tissues. 3D tissues were cultured at the air-liquid interface for 700 4 weeks, then mock infected or infected apically with 3x10 6 pfu HRV16 for 48 hours. RNA 701 sequencing was performed on three replicates for each condition. A. Volcano plot of 702 differentially expressed genes (DEGs) (scatte red dots, n=4034) between infected and mock-703 infected samples (FDR<0.05). The x-axis is the log2 fold change (infected/controls) while the 704 y-axis is the -log(FDR) calculated value. Black dots indicate the DEGs with log2 fold change 705 within the range of -1 to 1. Red dots indicate DEGs with changes less than -1 or greater than 706 1. B. Violin plot comparing the distributi on of gene expression between mock-infected (Ctrs) 707 and infected (HRV16) tissues. C. Heatmap of the top 20 upregulated and downregulated 708 genes in infected (HRV) and mock-infected (CTR) samples. Each row represents 1 of 3 709 replicates per condition. The color scale ranges from deep purple (no expression) to yellow 710 (high level expression). D. Gene networks induced during HRV16 infection of HBECs in air-711 liquid interface culture. PPI network functional enrichment analysis (generated using 712 STRING) using DEGs with log2-fold changes of greater than 2 or less than -2. Nodes 713 represent the interacting proteins with lines r epresenting direct links. A colour key shows the 714 different networks. Up to four colours per node are shown. 715 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 34 716 Figure 5. HRV16 infection in 3D cultured primary bronchial epithelial tissues leads to 717 changes in the splicing of cilia-related genes. A. Pie chart of the relative proportions of 718 the various splicing events identified in 1469 single alternative splicing events identified 719 using SplAdder. B. Pathway analysis of the si gnificantly differentially spliced genes during 720 HRV16 infection in primary bronchial epithelial 3D cultures (p<0.05, student’s t-test, 2-tailed). 721 722 Figure 6. Representative changes in cilia-related RNA splicing using Sashimi plots. A. 723 Diagram of the structure of a cilium showing the positions of the proteins encoded by the 724 RNAs shown in B-D (generated using biorender: https://biorender.com). B-D. Sashimi plots 725 of read coverage across genes encoding B. RSPH1, C. IFT74, and D. TMEM67. Dark blue 726 lines, read coverage from mock-infected air- liquid interface cultures. Green lines, read 727 coverage from HRV16-infected air-liquid interface cultures. Blue lines and black arrows 728 beneath the Sashimi plots indicate skipped exons. 729 730 Figure 7. Cilia are reduced in number and in length due to HRV16 infection in 3D 731 cultured primary bronchial epithelial tissues. A. H&E-stained 3D air-liquid interface 732 culture grown for 4 weeks of mock-infected HBECs. A magnification of the upper surface of 733 the tissue is shown above the main image. B. H&E-stained 3D air-liquid interface culture of 734 HBECs grown for 4 weeks then infected with 3x10 6 pfu HRV16 for 48 hours. Scale 735 bars=50µm. A magnification of the upper surface of the tissue is shown above the main 736 image. C. Graph of quantification of cilia numbers in mock-infected and HRV16-infected 3D 737 cultures. D. Average cilium length on mock-infected and HRV16-infected 3D cutlures. 738 Counts were taken from five sections from each of three replicate tissues at 16, 24 and 48 739 hours post-infection. The data show the mean and standard error of the mean. Significant 740 statistical difference, *p=<0.05, **p,<0.005, ***p<0.0005 (students t-test, 2-tailed). 741 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 35 742 Figure 8. HRV16 infection of 3D cultured primary bronchial epithelial tissues leads to 743 decreased expression of the cilia proteins β -tubulin and TMEM67. Fluorescence 744 confocal microscopy image of HBEC 3D air-liqui d interface cultures grown for 4 weeks using 745 an antibody to detect β -tubulin in A. mock-infected (Control) and C. HRV16-infected 746 (HRV16: 3x10 6 pfu) tissues at 48 hours post-infection. B & D. The same images but with 747 DAPI staining added to show the cell nuclei in the tissues. Fluorescence confocal 748 microscopy image of an HBEC 3D air-liquid interface culture grown for 4 weeks using an 749 antibody to detect TMEM67 in E. mock-infected (Control) and G. HRV16-infected (HRV16: 750 3x106 pfu) tissues at 48 hours post-infection. F & H. The same images but with DAPI staining 751 added to show the cell nuclei in the tissues. Scale bars=20µm. I. Graph of the quantification 752 of staining intensity of the β -tubulin antibody over time in mock-infected (Control) and 753 HRV16-infected (HRV16) 3D HBEC cultures. J. Graph of the quantification of staining 754 intensity of the TMEM67 antibody over time in mock-infected (Control) and HRV16-infected 755 (HRV16) 3D HBEC cultures. The data show the mean and standard error of the mean from 5 756 sections from three separate air-liquid interf ace cultures. *p<0.01, **p<0.001, ***p<0.0001 757 (student’s t-test, 2-tailed). 758 759 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 36

760 761 1. Palmenberg AC, Spiro D, Kuzmickas R, Wang S, Djikeng A, Rathe JA, et al. 762 Sequencing and analyses of all known human rhinovirus genomes reveal structure and 763 evolution. Science. 2009;324(5923):55-9. Epub 20090212. doi: 10.1126/science.1165557. 764 PubMed PMID: 19213880; PubMed Central PMCID: PMCPMC3923423. 765 2. Royston L, Tapparel C. Rhinoviruses and Respiratory Enteroviruses: Not as Simple 766 as ABC. Viruses. 2016;8(1). Epub 20160111. doi: 10.3390/v8010016. PubMed PMID: 767 26761027; PubMed Central PMCID: PMCPMC4728576. 768 3. Jacobs SE, Lamson DM, St George K, Wals h TJ. Human rhinoviruses. Clin Microbiol 769 Rev. 2013;26(1):135-62. doi: 10.1128/cmr.00077-12. PubMed PMID: 23297263; PubMed 770 Central PMCID: PMCPMC3553670. 771 4. Drysdale SB, Mejias A, Ramilo O. Rhinovirus - not just the common cold. J Infect. 772 2017;74 Suppl 1:S41-s6. doi: 10.1016/s0163-4453(17)30190-1. PubMed PMID: 28646961. 773 5. Bizzintino JA, Bosco A, Currie A, Khoo S-K, Franks K, Oo S, et al. Gene 774 Coexpression Networks Underlying Acute Wheezing Illnesses in Young Children. B27 Viral 775 infection and pediatric asthma. 2015. p. A2638-A. 776 6. Oberg L, Bosco A, Israelsson E, Bizzintino J, Currie A, Khoo S-K, et al. Interferon 777 Type-I Gene Signature is Muted in Children with Human Rhinovirus-C Induced-Wheeze. 778 C32 Clinical Asthma II2015. p. A4180-A. 779 7. Khoo SK, Read J, Franks K, Zhang G, Bizzintino J, Coleman L, et al. Upper Airway 780 Cell Transcriptomics Identify a Major New Immunological Phenotype with Strong Clinical 781 Correlates in Young Children with Acute Wheezing. J Immunol. 2019;202(6):1845-58. Epub 782 2019/02/13. doi: 10.4049/jimmunol.1800178. PubMed PMID: 30745463. 783 8. Coleman LA, Khoo SK, Franks K, Prastanti F, Le Souëf P, Karpievitch YV, et al. 784 Personal Network Inference Unveils Heterogeneous Immune Response Patterns to Viral 785 Infection in Children with Acute Wheezing. J Pers Med. 2021;11(12). Epub 20211203. doi: 786 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 37 10.3390/jpm11121293. PubMed PMID: 34945765; PubMed Central PMCID: 787 PMCPMC8706513. 788 9. Watters K, Inankur B, Gardiner JC, Warrick J, Sherer NM, Yin J, Palmenberg AC. 789 Differential Disruption of Nucleocytoplasmic Trafficking Pathways by Rhinovirus 2A 790 Proteases. J Virol. 2017;91(8). Epub 2017/02/10. doi: 10.1128/jvi.02472-16. PubMed PMID: 791 28179529; PubMed Central PMCID: PMCPMC5375692. 792 10. Walker E, Jensen L, Croft S, Wei K, Fulcher AJ, Jans DA, Ghildyal R. Rhinovirus 16 793 2A Protease Affects Nuclear Localization of 3CD during Infection. J Virol. 794 2016;90(24):11032-42. Epub 2016/09/30. doi: 10.1128/jvi.00974-16. PubMed PMID: 795 27681132; PubMed Central PMCID: PMCPMC5126362. 796 11. Fitzgerald KD, Chase AJ, Cathcart AL, Tran GP, Semler BL. Viral proteinase 797 requirements for the nucleocytoplasmic relocalization of cellular splicing factor SRp20 during 798 picornavirus infections. J Virol. 2013;87(5):2390-400. Epub 2012/12/21. doi: 799 10.1128/jvi.02396-12. PubMed PMID: 23255796; PubMed Central PMCID: 800 PMCPMC3571363. 801 12. Gustin KE, Sarnow P. Effects of poliovirus infection on nucleo-cytoplasmic trafficking 802 and nuclear pore complex composition. EMBO J.. 2001;20(1-2):240-9. doi: 803 10.1093/emboj/20.1.240. PubMed PMID: 11226174; PubMed Central PMCID: 804 PMCPMC140206. 805 13. Cammas A, Pileur F, Bonnal S, Lewis SM, Lévêque N, Holcik M, Vagner S. 806 Cytoplasmic Relocalization of Heterogeneous Nuclear Ribonucleoprotein A1 Controls 807 Translation Initiation of Specific mRNAs. Mol Biol Cell. 2007;18(12):5048-59. doi: 808 10.1091/mbc.e07-06-0603. PubMed PMID: 17898077. 809 14. Flather D, Nguyen JHC, Semler BL, Gers hon PD. Exploitation of nuclear functions by 810 human rhinovirus, a cytoplasmic RNA virus. PLoS Pathog. 2018;14(8):e1007277. Epub 811 2018/08/25. doi: 10.1371/journal.ppat.1007277. PubMed PMID: 30142213; PubMed Central 812 PMCID: PMCPMC6126879. 813 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 38 15. Fitzgerald KD, Semler BL. Re-localizati on of cellular protein SRp20 during poliovirus 814 infection: bridging a viral IRES to the host cell translation apparatus. PLos Pathog. 815 2011;7:e1002127. 816 16. Long JC, Caceres JF. The SR protein family of splicing factors: master regulators of 817 gene expression. Biochem J. 2009;417(1):15-27. 818 17. Manley JL, Krainer AR. A rational nomenclature for serine/arginine-rich protein 819 splicing factors (SR proteins). Genes Dev. 2010;24:1073-4. 820 18. Ohno K, Takeda JI, Masuda A. Rules and tools to predict the splicing effects of 821 exonic and intronic mutations. Wiley Interdiscip Rev RNA. 2018;9(1). Epub 20170926. doi: 822 10.1002/wrna.1451. PubMed PMID: 28949076. 823 19. Black DL. Mechanisms of alternative pre-messenger RNA splicing. Annu Rev 824 Biochem. 2003;72:291-336. 825 20. Issouf M, Vargas A, Boivin R, Lavoie JP. SRSF6 is upregulated in asthmatic horses 826 and involved in the MYH11 SMB expression. Physiol Rep. 2018;6(20):e13896. doi: 827 10.14814/phy2.13896. PubMed PMID: 30350466; PubMed Central PMCID: 828 PMCPMC6198134. 829 21. Graham SV, Faizo AAA. C ontrol of human papillomavirus gene expression by 830 alternative splicing. Virus Res. 2017;231:83-95. Epub 20161117. doi: 831 10.1016/j.virusres.2016.11.016. PubMed PMID: 27867028; PubMed Central PMCID: 832 PMCPMC5335905. 833 22. Tazi J, Bakkour N, Marchand V, Ayadi L, Aboufirassi A, Branlant C. Alternative 834 splicing: regulation of HIV-1 multiplication as a target for therapeutic action. FEBS J. 835 2010;277:867-76. 836 23. Esparza M, Bhat P, Fontoura BM. Viral- host interactions during splicing and nuclear 837 export of influenza virus mRNAs. Curr Opin Virol. 2022;55:101254. Epub 20220729. doi: 838 10.1016/j.coviro.2022.101254. PubMed PMID: 35908311; PubMed Central PMCID: 839 PMCPMC9945342. 840 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 39 24. Kamel W, Ruscica V, Embarc-Buh A, de Laurent ZR, Garcia-Moreno M, 841 Demyanenko Y, et al. Alphavirus infection tri ggers selective cytoplasmic translocation of 842 nuclear RBPs with moonlighting antiviral roles. Mol Cell. 2024;84(24):4896-911.e7. Epub 843 20241211. doi: 10.1016/j.molcel.2024.11.015. PubMed PMID: 39642884. 844 25. Kim J, Park RY, Chen JK, Kim J, Jeong S, Ohn T. Splicing factor SRSF3 represses 845 the translation of programmed cell death 4 mRNA by associating with the 5'-UTR region. 846 Cell Death Differ. 2014;21(3):481-90. Epub 20131129. doi: 10.1038/cdd.2013.171. PubMed 847 PMID: 24292556; PubMed Central PMCID: PMCPMC3921596. 848 26. Sanford JR, Gray NK, Beckman K, Càceres JF. A novel role for shuttling SR proteins 849 in mRNA translation. Genes Dev. 2004;18:755-68. 850 27. Patel M, Sachidanandan M, Adnan M. Serine arginine protein kinase 1 (SRPK1): a 851 moonlighting protein with theranostic ability in cancer prevention. Mol Biol Rep. 852 2019;46(1):1487-97. Epub 20181208. doi: 10.1007/s11033-018-4545-5. PubMed PMID: 853 30535769. 854 28. Nousiainen L, Sillanpää M, Jiang M, Thompson J, Taipale J, Julkunen I. Human 855 kinome analysis reveals novel kinases contri buting to virus infection and retinoic-acid 856 inducible gene I-induced type I and type III IFN gene expression. Innate Immun. 857 2013;19(5):516-30. Epub 20130212. doi: 10.1177/1753425912473345. PubMed PMID: 858 23405030. 859 29. Giannakouros T, Nikolakaki E, Mylonis I, Georgatsou E. Serine-arginine protein 860 kinases: a small protein kinase family with a large cellular presence. FEBS J. 861 2011;278(4):570-86. doi: 10.1111/j.1742-4658.2010.07987.x. 862 30. Aubol BE, Wu G, Keshwani MM, Movassat M, Fattet L, Hertel KJ, et al. Release of 863 SR Proteins from CLK1 by SRPK1: A Symbiotic Kinase System for Phosphorylation Control 864 of Pre-mRNA Splicing. Mol Cell. 2016;63(2):218-28. Epub 20160707. doi: 865 10.1016/j.molcel.2016.05.034. PubMed PMID: 27397683; PubMed Central PMCID: 866 PMCPMC4941815. 867 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 40 31. Yun CY, Fu XD. Conserved SR protein ki nase functions in nuclear import and its 868 action is counteracted by arginine methylati on in Saccharomyces cerevisiae. J Cell Biol. 869 2000;150(4):707-18. doi: 10.1083/jcb.150.4.707. PubMed PMID: 10952997; PubMed Central 870 PMCID: PMCPMC2175287. 871 32. Lai MC, Lin RI, Huang SY, Tsai CW, Tarn WY. A human importin-beta family protein, 872 transportin-SR2, interacts with the phosphorylat ed RS domain of SR proteins. J Biol Chem. 873 2000;275(11):7950-7. doi: 10.1074/jbc.275.11.7950. PubMed PMID: 10713112. 874 33. Nikolakaki E, Sigala I, Giannakouros T. Good Cop, Bad Cop: The Different Roles of 875 SRPKs. Front Genet. 2022;13:902718. Epub 20220602. doi: 10.3389/fgene.2022.902718. 876 PubMed PMID: 35719374; PubMed Central PMCID: PMCPMC9202992. 877 34. Zhou Z, Qiu J, Liu W, Zhou Y, Plocinik Ryan M, Li H, et al. The Akt-SRPK-SR Axis 878 Constitutes a Major Pathway in Transducing EGF Signaling to Regulate Alternative Splicing 879 in the Nucleus. Mol Cell. 2012;47(3):422-33. doi: 880 https://doi.org/10.1016/j.molcel.2012.05.014. 881 35. Lai M-C, Lin R-I, Huang S-Y, Tsai C-W, Tarn W-Y. A Human Importin- β Family 882 Protein, Transportin-SR2, Interacts with the Phosphorylated RS Domain of SR Proteins. J 883 Biol Chem. 2000;275(11):7950-7. doi: 10.1074/jbc.275.11.7950. 884 36. Michlewski G, Sanford JR, Cbceres JF. The Splicing Factor SF2/ASF Regulates 885 Translation Initiation by Enhancing Phosphorylation of 4E-BP1. Mol Cell. 2008;30(2):179-89. 886 37. Howard JM, Sanford JR. The RNAissance family: SR proteins as multifaceted 887 regulators of gene expression. Wiley Interdis ciplinary Rev: RNA. 2015;6(1):93-110. doi: 888 10.1002/wrna.1260. 889 38. Hanamura A, Càceres JF, Mayeda A, Franza BR, Krainer AR. Regulated tissue-890 specific expression of antagonistic pre-mRNA splicing factors. RNA. 1998;4:430-44. 891 39. Proud D, Turner RB, Winther B, Wiehler S, Tiesman JP, Reichling TD, et al. Gene 892 expression profiles during in vivo human rhinovir us infection: insights into the host response. 893 Am J Respir Crit Care Med. 2008;178(9):962-8. 894 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 41 40. Chen Y, Hamati E, Lee PK, Lee WM, Wachi S, Schnurr D, et al. Rhinovirus induces 895 airway epithelial gene expression through double-stranded RNA and IFN-dependent 896 pathways. Am J Respir Cell Mol Biol. 2006;34(2):192-203. Epub 20051006. doi: 897 10.1165/rcmb.2004-0417OC. PubMed PMID: 16210696; PubMed Central PMCID: 898 PMCPMC2644182. 899 41. Bai J, Smock SL, Jackson GR, Jr., MacIsaac KD, Huang Y, Mankus C, et al. 900 Phenotypic Responses of Differentiated Asthmatic Human Airway Epithelial Cultures to 901 Rhinovirus. PLOS ONE. 2015;10(2):e0118286. doi: 10.1371/journal.pone.0118286. 902 42. Kim TK, Bheda-Malge A, Lin Y, Sreekrishna K, Adams R, Robinson MK, et al. A 903 systems approach to understanding human rhinovirus and influenza virus infection. Virology. 904 2015;486:146-57. Epub 20151006. doi: 10.1016/j.virol.2015.08.014. PubMed PMID: 905 26437235; PubMed Central PMCID: PMCPMC7111289. 906 43. Ong HH, Andiappan AK, Duan K, Lum J, Liu J, Tan KS, et al. Transcriptomics of 907 rhinovirus persistence reveals sustained expression of RIG-I and interferon-stimulated genes 908 in nasal epithelial cells in vitro. Allergy. 2022;n/a(n/a). doi: https://doi.org/10.1111/all.15280. 909 44. Lei C, Yang J, Hu J, Sun X. On the Calculation of TCID(50) for Quantitation of Virus 910 Infectivity. Virol Sin. 2021;36(1):141-4. Epub 20200526. doi: 10.1007/s12250-020-00230-5. 911 PubMed PMID: 32458296; PubMed Central PMCID: PMCPMC7973348. 912 45. Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. Graph-based genome alignment 913 and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol. 2019;37(8):907-15. Epub 914 20190802. doi: 10.1038/s41587-019-0201-4. PubMed PMID: 31375807; PubMed Central 915 PMCID: PMCPMC7605509. 916 46. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for 917 assigning sequence reads to genomic features. Bioinformatics. 2014;30(7):923-30. Epub 918 20131113. doi: 10.1093/bioinformatics/btt656. PubMed PMID: 24227677. 919 47. Kahles A, Ong CS, Zhong Y, Rätsch G. SplAdder: identification, quantification and 920 testing of alternative splicing events from RNA-Seq data. Bioinformatics. 2016;32(12):1840-921 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 42 7. Epub 2016/02/14. doi: 10.1093/bioinformatics/btw076. PubMed PMID: 26873928; 922 PubMed Central PMCID: PMCPMC4908322. 923 48. Liao Y, Wang J, Jaehnig EJ, Shi Z, Zhang B. WebGestalt 2019: gene set analysis 924 toolkit with revamped UIs and APIs. Nucleic Acids Res. 2019;47(W1):W199-w205. doi: 925 10.1093/nar/gkz401. PubMed PMID: 31114916; PubMed Central PMCID: 926 PMCPMC6602449. 927 49. Katz Y, Wang ET, Silterra J, Schwartz S, Wong B, Thorvaldsdóttir H, et al. 928 Quantitative visualization of alternative exon expression from RNA-seq data. Bioinformatics. 929 2015;31(14):2400-2. Epub 20150122. doi: 10.1093/bioinformatics/btv034. PubMed PMID: 930 25617416; PubMed Central PMCID: PMCPMC4542614. 931 50. Khoo S-K, Read J, Franks K, Zhang G, Bizzintino J, Coleman L, et al. Airway IRF7 hi 932 versus IRF7 lo molecular response patterns determine clinical phenotypes in children with 933 acute wheezing. J Immunol. 2019; 202(6):1845-1858. doi: 10.4049/jimmunol.1800178. 934 PubMed PMID: 30745463 935 51. Dunn EF, Connor JH. HijAkt: The PI3K/Akt pathway in virus replication and 936 pathogenesis. Prog Mol Biol Transl Sci. 2012;106:223-50. doi: 10.1016/b978-0-12-396456-937 4.00002-x. PubMed PMID: 22340720; PubMed Central PMCID: PMCPMC7149925. 938 52. Bates DO, Morris JC, Oltean S, Donaldson LF. Pharmacology of Modulators of 939 Alternative Splicing. Pharmacol Rev. 2017;69(1):63-79. doi: 10.1124/pr.115.011239. 940 PubMed PMID: 28034912; PubMed Central PMCID: PMCPMC5226212. 941 53. Änkö M-L, Müller-McNicoll M, Brandl H, Curk T, Gorup C, Henry I, et al. The RNA-942 binding landscape of two SR proteins reveal unique function s and bidning to diverse RNA 943 classes. Genome Biol. 2012;13:R17. 944 54. Bedard KM, Daijogo S, Semler BL. A nucleo-cytoplasmic SR protein functions in viral 945 IRES-mediated translation initiation. EMBO J. 2007;26(2):459-67. Epub 20061221. doi: 946 10.1038/sj.emboj.7601494. PubMed PMID: 17183366; PubMed Central PMCID: 947 PMCPMC1783453. 948 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 43 55. Duval K, Grover H, Han LH, Mou Y, Pegoraro AF, Fredberg J, Chen Z. Modeling 949 Physiological Events in 2D vs. 3D Cell Cultur e. Physiology (Bethesda). 2017;32(4):266-77. 950 doi: 10.1152/physiol.00036.2016. PubMed PMID: 28615311; PubMed Central PMCID: 951 PMCPMC5545611. 952 56. Murray LM, Thillaiyampalam G, Xi Y, Cristino AS, Upham JW. Whole transcriptome 953 analysis of high and low IFN- α producers reveals differential response patterns following 954 rhinovirus stimulation. Clin Transl Im munol. 2021;10(11):e1356. Epub 2021/12/07. doi: 955 10.1002/cti2.1356. PubMed PMID: 34868584; PubMed Central PMCID: PMCPMC8599968. 956 57. Chang EH, Pouladi N, Guerra S, Jandova J, Kim A, Li H, et al. Epithelial cell 957 responses to rhinovirus identify an early-life-onset asthma phenotype in adults. J Allergy Clin 958 Immunol. 2022;150(3):604-11. Epub 20220331. doi: 10.1016/j.jaci.2022.03.020. PubMed 959 PMID: 35367470; PubMed Central PMCID: PMCPMC9463086. 960 58. Michi AN, Love ME, Proud D. Rhinovirus-Induced Modulation of Epithelial 961 Phenotype: Role in Asthma. Viruses. 2020;12(11). Epub 20201119. doi: 962 10.3390/v12111328. PubMed PMID: 33227953; PubMed Central PMCID: 963 PMCPMC7699223. 964 59. Wheway G, Lord J, Baralle D. Splicing in the pathogenesis, diagnosis and treatment 965 of ciliopathies. Biochim Biophys Acta Gene Regul Mech. 2019;1862(11-12):194433. Epub 966 20191104. doi: 10.1016/j.bbagrm.2019.194433. PubMed PMID: 31698098. 967 60. Onoufriadis A, Shoemark A, Schmidts M, Patel M, Jimenez G, Liu H, et al. Targeted 968 NGS gene panel identifies mutations in RSPH1 causing primary ciliary dyskinesia and a 969 common mechanism for ciliary central pair agene sis due to radial spoke defects. Hum Mol 970 Genet. 2014;23(13):3362-74. Epub 20140211. doi: 10.1093/hmg/ddu046. PubMed PMID: 971 24518672; PubMed Central PMCID: PMCPMC4049301. 972 61. Bakey Z, Cabrera OA, Hoefele J, Antony D, Wu K, Stuck MW, et al. IFT74 variants 973 cause skeletal ciliopathy and motile cilia defects in mice and humans. PLoS Genet. 974 2023;19(6):e1010796. Epub 20230614. doi: 10.1371/journal.pgen.1010796. PubMed PMID: 975 37315079; PubMed Central PMCID: PMCPMC10298753. 976 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 44 62. Collin GB, Won J, Hicks WL, Cook SA, Nishina PM, Naggert JK. Meckelin is 977 necessary for photoreceptor intraciliary transport and outer segment morphogenesis. Invest 978 Ophthalmol Vis Sci. 2012;53(2):967-74. Epub 20120223. doi: 10.1167/iovs.11-8766. 979 PubMed PMID: 22247471; PubMed Central PMCID: PMCPMC3317434. 980 63. Singhania A, Rupani H, Jayasekera N, Lumb S, Hales P, Gozzard N, et al. Altered 981 Epithelial Gene Expression in Peripheral Airways of Severe Asthma. PLoS One. 982 2017;12(1):e0168680. Epub 20170103. doi: 10.1371/journal.pone.0168680. PubMed PMID: 983 28045928; PubMed Central PMCID: PMCPMC5207492. 984 64. Lee KM, Wu CC, Fan YT, Chiang HJ, Lien PY, Wang JP, et al. Subversion of 985 phosphorylated SR proteins by enterovirus A71 in IRES-dependent translation revealed by 986 RNA-interactome analysis. PLoS Pathog. 2025;21(6):e1013242. Epub 20250616. doi: 987 10.1371/journal.ppat.1013242. PubMed PMID: 40522993; PubMed Central PMCID: 988 PMCPMC12193706. 989 65. Fukuhara T, Hosoya T, Shimizu S, Sumi K, Oshiro T, Yoshinaka Y, et al. Utilization 990 of host SR protein kinases and RNA-splicing mach inery during viral replication. Proc Natl 991 Acad Sci U S A. 2006;103(30):11329-33. Epub 20060713. doi: 10.1073/pnas.0604616103. 992 PubMed PMID: 16840555; PubMed Central PMCID: PMCPMC1544086. 993 66. Karakama Y, Sakamoto N, Itsui Y, Nakagawa M, Tasaka-Fujita M, Nishimura-994 Sakurai Y, et al. Inhibition of hepatitis C virus replication by a specific inhibitor of serine-995 arginine-rich protein kinase. Antimicr ob Agents Chemother. 2010;54(8):3179-86. Epub 996 20100524. doi: 10.1128/aac.00113-10. PubMed PMID: 20498328; PubMed Central PMCID: 997 PMCPMC2916360. 998 67. Yaron TM, Heaton BE, Levy TM, Johnson JL, Jordan TX, Cohen BM, et al. Host 999 protein kinases required for SARS-CoV-2 nucleocapsid phosphorylation and viral replication. 1000 Science Signaling. 2022;15(757):eabm0808. doi: doi:10.1126/scisignal.abm0808. 1001 68. Faizo AAA, Bellward C, Hernandez-Lopez HR, Stevenson A, Gu Q, Graham SV. The 1002 splicing factor kinase, SR protein kinase 1 (SRPK1) is essential for late events in the human 1003 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint 45 papillomavirus life cycle. PLoS Pathog. 2025;21(4):e1012697. Epub 20250409. doi: 1004 10.1371/journal.ppat.1012697. PubMed PMID: 40203066. 1005 69. Takamatsu Y, Krähling V, Kolesnikova L, Halwe S, Lier C, Baumeister S, et al. 1006 Serine-Arginine Protein Kinase 1 Regulates Ebola Virus Transcription. mBio. 2020;11(1). 1007 Epub 20200225. doi: 10.1128/mBio.02565-19. PubMed PMID: 32098814; PubMed Central 1008 PMCID: PMCPMC7042693. 1009 70. Zheng Y, Fu XD, Ou JH. Suppression of hepatitis B virus replication by SRPK1 and 1010 SRPK2 via a pathway independent of the phosphorylation of the viral core protein. Virology. 1011 2005;342(1):150-8. Epub 20050824. doi: 10.1016/j.virol.2005.07.030. PubMed PMID: 1012 16122776. 1013 71. Chen Q, Tan KS, Liu J, Ong HH, Zhou S, Huang H, et al. Host Antiviral Response 1014 Suppresses Ciliogenesis and Motile Ciliary Functions in the Nasal Epithelium. Front Cell Dev 1015 Biol. 2020;Volume 8 - 2020. doi: 10.3389/fcell.2020.581340. 1016 72. Zhou Z, Fu X-D. Regulation of splicing by SR proteins and SR protein-specific 1017 kinases. Chromosoma. 2013;122(3):191-207. doi: 10.1007/s00412-013-0407-z. 1018 73. Eperon IC, Makarova OV, Mayeda A, Munroe SH, Caceres JF, Hayward DG, Krainer 1019 AR. Selection of alternative 5' splice sites: role of U1 snRNP and models for the antagonistic 1020 effects of SF2/ASF and hnRNP A1. Mol Cell Biol. 2000;20(2):8303-18. 1021 1022 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 26, 2025. ; https://doi.org/10.1101/2025.08.26.672366doi: bioRxiv preprint