{"paper_id":"278d7cec-fa30-4776-bf1a-c0b5c00873c7","body_text":"Title \nInteraction between  ciliary component proteins  from Chlamydomonas revealed by \nCRISPR/CAS9, cryo-electron tomography and mass spectrometry \n \n \nLeo Luo 1,2, Noemi Zimmermann 1,2, Akira Noga 1,3, Alexander Leitner 4 and Takashi \nIshikawa1,2,* \n1. Department of Biology and Chemistry, Paul Scherrer Institute \n2. Department of Biology, ETH Zurich \n3. Tokyo Institute of Technology, Japan \n4. Institute of Molecular Systems Biology, Department of Biology, ETH Zurich \n* Correspondence to takashi.ishikawa@psi.ch \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nAbstract \nTo understand molecular mechanism of ciliary beating motion, knowledge of location, \ninteraction and dynamics of >400 component proteins are indispensable. While recent progress \nof structural biology revealed conformation and localization of >100 proteins, we still need to \ninvestigate their networking, art of their interaction and assembly mechanism. We applied \nCRISPR/CAS9 genome editing technique to the green algae Chlamydomonas to engineer a \ndeletion mutant of a ciliary component, FAP263, located at the distal protrusion, and examined \nit structurally by cryo-electron tomography (cryo-ET) and mass spectrometry (MS). Cryo-ET \nand atomic model fitting demonstrated that  the FAP263 deletion mutan t lacks  additional \ncomponents, FAP78, and FAP184. Unassigned density near FAP263 in the cryo -ET map of \nWT cilia is likely FAP151, a s suggested by  cross-linking mass spectrometry. Based on the \nstructure, we modeled how these four proteins  might form a complex.  Furthermore, it was \nshown that dynein f phosphorylation is inhibited in the FAP263 mutant, indicating an important \nrole of this protein complex for dynein f phosphorylation.  Our study demonstrates a novel \napproach to investigate protein networking inside cilia. \n \n.CC-BY 4.0 International licensemade available under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nIntroduction \nMotile cilia, beating organelle to cause either swimming of the cells or extracellular fluid flow, are \ncomposed of more than 400 proteins (Pazour et al, 2005). Ciliary motion is considered to be product of \ndynamic interactions and networking of these proteins. There has been wide variety of attempts to reveal \nthe protein interactions, 3D arrangements and their dynamic changes. Recent progress of single particle \ncryo-EM analysis and folding prediction by  Alphafold2 allowed visualization of >100 components \nwhich exist in the peripheral microtubule doublet with 96nm periodicity (Walton et al , 2023) , \ncomponents of the central pair apparatus with 32nm periodicity (Gui et al, 2022), and intraflagellar \ntransport complexes (Hesketh et al, 2022). Cryo-electron tomography of intact cilia, combined with \nhigh resolution obtained by single particle cryo -EM, enabled modeling of conformational change of \ndynein and associated proteins during motion (Zimmermann et al, 2023). Further expansion of our \nknowledge to cover more component proteins at various states of beating motion and at various stages \nof ciliogenesis is awaited both for basic understanding of ciliary movement and medical investigation \nof ciliopathy (Wallmeier et al, 2020).  \nBoth structural and functional studies utilize deletion mutants. By comparing the wild type and deletion \nmutants, or by comparing the wild type and a strain with a deletion mutant resc ued by a tagged gene, \nlocation of the target protein can be detected based on decrease or increase of density in the 3D map, \nrespectively. In cilia research, combination of deletion mutants and cryo -ET revealed functional roles \nand locations of a number of  component proteins in intact cilia (Ishikawa, 2016). In the case target \nproteins are small and hard to detect as a loss of density in deletion mutants or in case deletion could \ncause collapse of other proteins in the complex, genetic tag to increase density helps efficiently. \nComponents of the radial spoke (Oda et al, 2014c), the dynein regulatory complex (Oda et al, 2014b), \nruler proteins for the doublet microtubule (Oda et al , 2014a)  and an inner dynein scaffold protein  \n(Kutomi et al, 2021), were located by cryo-ET in this way.   \nHowever, availability of deletion mutants for motile cilia research was severely limited. \nChlamydomonas reinhardtii has been the most popular model organism for motile cilia research \nbecause of abundance of mutants of ciliary components, which were isolated based on motility defect. \nActive mutagenesis of Chlamydomonas by targeting gene of interest, however, is complicated because \nhomologous recombination of this species is not established. While there are wide variety of deletion \nmutants, first induced chemically or by radiation and later by random insertion (Li et al , 2016) , \nsystematic methods to delete targeted gene have been unavailable for years. Meanwhile Tetrahymena \nthermophila, another popular model organism for motile cilia research can be mutated in targeting genes \nusing homologous recombination, which has been used for biochemical research by purifying mutated \nproteins (Ichikawa et al , 2015) . For cellular studies  such as cellular cryo -ET of cilia , mutation of \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nTetrahymena cannot be used easily, since there are 50 copies of each gene in the small nucleus and \ncomplete exclusion of the intrinsic sequence is not straightforward. \nIn this study, we generated deletion mutant of  a ciliary component FAP263 from Chlamydomonas, \nwhich is located near the outer surface of the A -tubule and dynein f, using CRISPR/CAS9. We \ncharacterized these deletion mutants biochemically and structurally. Our study demonstrates proof-of-\nprinciple of combination of CRISPR/CAS9 and cryo-ET for cilia research, as well as its advantage to \nstudy influence of gene deletion to other component proteins. \n \nResults and discussion \nWe made deletion mutant of FAP263  by inserting stop codon to Chlamydomonas genome by \nCRISPR/CAS9 following the pr otocol of (Shin et al , 2016) . Mutation was confirmed by PCR and \nsequencing (Fig.1). \nAfter back-crossing the mutant to WT to minimize a risk of off -target effects, we further structurally \nanalyzed FAP263 deletion mutant by cryo-ET and subtomogram averaging (Figs.2, 3A, Supplementary \nFig.1). Protein components can be located in the cryo-ET map by fitting high-resolution single particle \ncryo-EM structure of split doublet microtubule from Chlamydomonas cilia (Walton et al, 2023). There \nis an area, where density exists in WT, but not in the FAP263 deletion mutant (Fig.2 ; Supplementary \nFig.1) which likely corresponds to proteins lost by FAP263 deletion . In the fitted atomic model from \nsingle particle cryo-EM, this lost area corresponds to FAP78/FAP184 and FAP263 (Fig.3B). Therefore \nthis density can be explained as a complex of FAP263, FAP78 and FAP184.  \nHowever we found a small unassigned region as well (Fig.3B). This unassigned density is positioned \nbetween FAP78 and FAP263. To find candidates for an additional protein located in this region, we \nperformed cross -linking mass spectrometry (Leitner et al , 2014)  and looked for proteins that were \nidentified in earlier proteomics work on Chlamydomonas cilia (Pazour et al , 2005) , but were not \nlocalized in the cryo-EM structure. The cross-link data points to a putative interaction between FAP78 \nand the protein FAP151 (Fig.4), which was not involved in the list of single particle cryo-EM analysis. \nIt should be pointed out that the confidence of this identification is relatively low, which may be \nattributed to the low abundance of the protein complex in cilia, and/or specific mass spectrometric \nproperties of the cross-linkned peptides. Nevertheless, we modeled FAP151 by Alphafold2 and fitted \nto the unassigned density (Fig.3B). FAP151 fits well to other components and is likely a piece of this \ncomplex (Fig.3CD). Since FAP78 is solved in full -length by single particle cryo -EM (Walton et al, \n2023), this density may contain a part of FAP78 as well.  \nFurthermore, we performed phosphoproteomics  experiments to study the role of protein \nphosphorylation on putative functions of the subcomplex involving FAP263. Comparative mass \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nspectrometric analysis of phosphopeptides from wild type and FAP263 mutant samples after enrichment \nwith titanium dioxide (Leitner et al , 2010)  resulted in a high coverage of the Chlamydomonas \nphosphoproteome. Four proteins in the distal protrusion  have more than one assigned spectrum for \npeptides that are detected phosphorylated in WT but not in the FAP263 mutant (Table 2). Among dynein \nisoforms, dynein f (DHC10) is located close to FAP263 and likely influenced by its deletion for \nphosphorylation. Indeed dynein f shows phosphorylation in WT, but not upon FAP263 deletion (Table \n1). This suggests that the FAP78/FAP151/FAP184/FAP263 complex is responsible for dynein f \nphosphorylation. Among these four proteins, FAP78 is likely to have kinase activity, according to \nclassification by Panther (https://phytozome-\nnext.jgi.doe.gov/report/gene/Creinhardtii_v5_6/Cre12.g536600). However, phosphorylation by \nanother protein located near this complex, for example a Nima kinase CNK4, cannot be excluded. \nAlthough we have not seen visual difference of swimming property between WT and the FAP263 \ndeletion mutant, change of beating frequency and amplitude caused by mutation of Ccdc113/Ccdc96 \n(corresponding to FAP263/184) in Tetrahymena was reported (Bazan et al , 2021) . While \nphosphorylation of IC138, associated with dynein f, was reported (Bower et al, 2009), phosphorylation \nof dynein f itself has not been studied. Functional roles of this complex is still to be investigated. \nIn this study, we demonstrated CRISPR/CAS9 of Chlamydomonas is useful for structural research of \ncilia. The FAP263 deletion causes loss of FAP78, FAP151, and FAP184. This approach can be applied \nto solve various unanswered questions, such as precise location of various species of dyneins, role of \nindividual dyneins and regulatory proteins in ciliary motion, as well as ciliogenesis.  \n \n \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nFigure caption \nFig.1 \nSequence of the FAP263 deletion mutant by CRISPR/CAS9. \n \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nFig. 2 \nCryo-ET structure of FAP263 deletion mutant. The maps by subtomogram averaging from cryo-ET of \nWT (A) and the FAP263 deletion mutant  (B). The distal protrusion (in WT) and the corresponding \nplace (in the deletion mutant) are indicated by red arrows. Other major components (ODA: outer dynein \narms; IDA: inner dynein arms; RS: radial spoke) are also indicated. \n \n \n \n \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nFig. 3 \nEnlarged view of the distal protrusion including FAP263 and associated proteins in cilia from averaged \ncryo-ET map of WT (grey) and the FAP263 deletion mutant (yellow). In (A) the density difference, \ncorresponding to the distal protrusion, is indicated by a red arrow. In (B), atomic models from (Walton \net al, 2023) are fitted to the cryo-ET map and superimposed. FAP78 (green), FAP184 (blue), FAP283 \n(red). (C) FAP151, modeled by Alphafold2 is added in purple. (D) Only atomic models are presented \nfrom (C). The color code is the same as (C). \n \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nFig. 4 Crosslinking MS sites, 366K of FAP78 (green) and 355K of FAP151 (purple), are shown in white.  \n \n \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nSupplementary Fig.1 Cross sections from averaged subtomograms from cryo -ET of WT  (left) and \nFAP263 deletion mutant (right). In the top row, presence and absence of the distal protrusion is \nindicated by red arrows. In the middle and bottom rows, the paralle l sections including doublet \nmicrotubules (DMT) shows outer dynein arms (ODA) and inner dynein arms (IDA) in the same \nstructure between WT and the mutant. \n \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nTable 1. Phosphorylation of dyneins \nGene symbol / uniprot \naccession number \nAverage number of peptide \nspectrum matches for \nphosphopeptides in control \nAverage number of peptide \nspectrum matches for \nphosphopeptides in FAP263 \nmutant \nDHC1 A0A2K3D1N8 0 0.3 \nDHC2 A0A2K3DE97 2.7 2.3 \nDHC3 A0A2K3DN16 23.7 28 \nDHC4 A0A2K3E2Q0 21.7 31.3 \nDHC5 A0A2K3E2P7 4.7 0.6 \nDHC6 A0A2K3DSC5 2.3 0 \nDHC7 A0A2K3CYE9 6.3 9.7 \nDHC8 A0A2K3CUR8 11.7 14 \nDHC9 A0A2K3E486 2 2.3 \nDHC10 A0A2K3CY97 1.3 0 \nDHC11 A0A2K3D5V4 13 19.3 \nDHC12 A0A2K3DQQ4 25.7 43.3 \nDHC13 A0A2K3DV97 41.3 31 \nDHC14 A8J1M5 1.3 0.3 \nDHC15 A0A2K3D8D3 0 0 \nDHC16 A0A2K3DLV6 0 0 \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nTable 2. Phosphorylation of possible distal protrusion proteins. \nGene symbol / uniprot \naccession number \nAverage number of identified \nphosphopeptides in control \nAverage number of identified \nphosphopeptides in FAP 263 \nmutant \nFAP263 A0A2K3CQ22 5.7 0 \nFAP184 A0A2K3DWC0 48.7 2.7 \nFAP78 A0A2K3D574 11.3 2.7 \nFAP151 A0A2K3D802 1 0 \nCNK4 Q6UPR2 1.7 1 \n \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nMaterials and methods \nStrains and media \nC. reinhardtii  strain CC -503 cw92 mt+ (Chlamydomonas Resource Center, \nhttps://www.chlamycollection.org/) were used in this study. Cells were grown in Tris-acetate-\nphosphate (TAP) medium in a constant light/dark cycle (light cycle: 21:00 -11:00, dark cycle \n11:00-21:00). \n \ncrRNA design \ncrRNAs were chosen using Benchling (Hirano et al, 2019). On-target and off -target scores \ncalculated by Benchling against Chlamydomonas r. genome were aimed to be above 60 and 90, \nrespectively. crRNA s were purchased from Integrated DNA Technology’s (IDT) online \ncustom Alt-R® CRISPR-Cas9 guide RNA tool.  \n \nDesign of repair template and preparation \nThe repair template was designed to have two homology arms upstream and downstream of \nthe cutting side, each with a length of 25 bp. In the middle of the homology arms is a FLAG -\ntag with two stop codons. The FLAG-tag serves the detection of mutant colonies by polymerase \nchain reaction (PCR). The repair template was ordered as a forward and reverse oligonucleotide \non Microsynth AG. The oligonucleotides were ordered with 3 phosphorothioate bonds on both \n5’ and 3’. To anneal the oligonucleotides, 1 µL of forward and 1 µL of reverse oligonucleotides \nwere mixed in 18 µL IDT RNA duplex buffer. Afterwards the mix was heated to 95 °C for 2 \nminutes and cooled down at 0.1 °C/sec to room temperature. \n \nParomomycin cassette for selection \nThe plasmid pSI103 -1, which confers paromomycin resistance, was ordered from \nChlamydomonas Resource Center. The resistance cassette was amplified using  chemically \ncompetent E. coli. The plasmid was  extracted with a plasmid extraction kit from QIAGEN. \nAfterwards, the DNA is linearized by KpnI -HF from Biolabs . The linearized DNA was \nconcentrated to 1 µg/µL by ethanol precipitation. \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\n \nPreparation of Cas9/gRNA RNP \ngRNA was assembled by mixing 5 μL of 100  μM IDT Alt-R® crRNA with 5 μL of 100 μM \nIDT Alt-R® tracrRNA. The mix was incubated at  95 °C for 2 minutes  and cooled down at \n0.1 °C/sec to room temperature.  4.5 μL of 10 μM gRNA was mixed with 4.5 μL of 10 μM IDT \nAlt-R® Cas9, 1.5 µL of 10x NEB 3.1 and 4. 5 µL of ddH 2O. The RNP system was incubated \nat 37 °C for 15 minutes. \n \nTransformation by electroporation \n150 mL of cw92 cells were grown for 3 days in TAP medium. Cells were centrifuged at 800 g \nfor 5 minutes. The cells were washed once with electroporation  buffer (30 mM Hepes, 5 mM \nMgSO4, 50  mM K -acetate, 1  mM Ca -acetate, 60  mM Sucrose ). Afterwards the cells were \npelleted again and resuspended in 2-3 times the volume of cells in electroporation buffer. Cells \nwere diluted to a concentration of 3 * 108  cells/mL. Then they were incubated at 40 °C for 30 \nminutes at 120 rpm. 100 µL of heat shocked cells, 15 µL of RNP, 4.5 µL of repair template \nand 1 µg of pSI103 -1 were mixed in a cuvette from Bio -Rad (Catalog No. 165 -2086). The \nmixture was electroporated using ECM® 630  from BTX – Harvard Apparatus with the \nfollowing conditions: 410 V low vol tage, Resistor 25 Ohm, Capacitance 600 uF. After \nelectroporation cells were incubated for 1 hour at 15 °C. The cells were transferred into 10 mL \nof 60 mM TAP sucrose for recovery overnight under constant light and shaking.  \n \nTransfer on TAP agar plates \nAfter recovery we centrifuged the cells at 800 g for 5 minutes and plated them on 1.5% TAP \nagar plates with 10 µg/mL paromomycin. Colonies can be picked after 5-7 days and transferred \nto 96 well plate with TAP. Confirmation of mutant colonies was done by PCR using Phusion™ \nHigh-Fidelity DNA Polymerase. Screening for mutants were done by gel electrophoresis. To \nconfirm mutants, they were sent for sequencing at Microsynth AG. \n \nCell culture and harvesting cilia \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nChlamydomonas cell culture and cilia isolation were followed by Witman’s protocol (Witman, \n1986). \n \nCryo-ET \nCryo-ET grid preparation and data acquisition were after our previous work (Zimmermann et \nal, 2023), using one-sided manual blotting and freezing by Cryo-plunge (Gatan, USA) and the \nTitan Krios G 3 transmission electron microscope (TFS, USA) with Quantum energy filter \n(Gatan). Subtomogram average using pseudo nine-fold symmetry and 96nm periodicity of cilia \nwas carried on using our algorithm published previously (Bui & Ishikawa, 2013; Zimmermann \net al, 2023).  \n \nCross-linking mass spectrometry \nC. reinhardtii strain cc124- was cultured for 3 days. Cilia were isolated by dibucaine. Isolated \ncilia by dibucaine were treated with 1% OGP in equal volume to remove cell membrane. \ncOmplete™ Proteasehemmer-Cocktail by Roche were used to stop protein degradation. Protein \nconcentration was measured using BCA assay and adjusted to 0.5-2 mg/ml with a total amount \nof 50-100 µg protein. Cross-linking experiments were performed at room temperature with the \namine-reactive disuccinimidyl suberate in isotopically light  and heavy form for 1 hour (DSS -\nd0/d12, Creative Molecules). Afterwards the protocol by (Leitner et al., 2014) was used to \nprocess the samples further. Samples were analyzed by liquid chromatography -tandem mass \nspectrometry on an Orbitrap Fusion Lumos instrument (ThermoFisher Scientific), and MS data \nwas analyzed by xQuest. \n \nPhosphoproteomics \nC. reinhardtii strain cc124- was used as wild type control for this experiment. Three replicates \nof the cc124- strain and three replicates of the FAP263 mutant strain were cultured in 300 ml \nof TAP medium. Each culture was maintained for a duration of three days.  Isolation of cilia \nand sample preparation were done as described above. To prevent dephosphorylation, \nPhosSTOP™ from Sigma Aldrich was added to sample. Amount of protein was adjusted to \n150 µg per sample by BCA assay. Mass spectrometric analysis of protein phosphorylation was \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nperformed following (Leitner et al., 2010) using Titansphere TiO material (GL Sciences) for \nenrichment. Samples were analyzed by liquid chromatography -tandem mass spectrometry on \nan Orbitrap Fusion Lumos instrument (ThermoFisher Scientific), and MS data was analyzed \nby FragPipe/MSFragger. For data analysis a cutoff of Peptide Prophet Probability > 0.95 was \nchosen. For comparison of phosphorylation states between wild type and FAP263 mutant the \ntotal amount of peptide spectrum matches of phospho peptides were counted. \n \n \nAuthor contribution \nLL conducted experiments, including CRISPR/CAS9, cryo-ET of the FAP263 deletion mutant, \nMS. NZ did cryo-ET of WT. NA backcrossed mutant cells to WT. AL supervised and designed \nMS. TI designed the entire project. The manuscript was prepared by LL and TI.  \n \nAcknowledgement \nWe appreciate Prof. Kenichi Wakabayashi for advices at start-up of CRISPR/CAS9, ScopeM \nand CEMK (ETHZ) for cryo-EM support and Prof. Paula Picotti (ETH Zurich) for access to \nthe laboratory infrastructure and instrumentation. This research was funded by grants from \nSwiss National Science Foundation (IZLIZ3_200294, 310030_192644), NanoArgovia \n(FunkEM project) and Novartis Biomedical Foundation (to TI). \n \n \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nReferences \nBazan R, Schröfel A, Joachimiak E, Poprzeczko M, Pigino G & Wloga D (2021) \nCcdc113/Ccdc96 complex, a novel regulator of ciliary beating that connects radial \nspoke 3 to dynein g and the nexin link. PLoS Genet 17: e1009388 \nBower R, VanderWaal K, O’Toole E, Fox L, Perrone C, Mueller J, Wirschell M, Kamiya R, \nSale WS & Porter ME (2009) IC138 defines a subdomain at the base of the I1 dynein \nthat regulates microtubule sliding and flagellar motility. Mol Biol Cell 20: 3055–3063 \nBui KH & Ishikawa T (2013) 3D structural analysis of flagella/cilia by cryo-electron \ntomography. Meth Enzymol 524: 305–323 \nGui M, Wang X, Dutcher SK, Brown A & Zhang R (2022) Ciliary central apparatus structure \nreveals mechanisms of microtubule patterning. Nat Struct Mol Biol 29: 483–492 \nHesketh SJ, Mukhopadhyay AG, Nakamura D, Toropova K & Roberts AJ (2022) IFT-A \nstructure reveals carriages for membrane protein transport into cilia. Cell 185: 4971-\n4985.e16 \nHirano S, Abudayyeh OO, Gootenberg JS, Horii T, Ishitani R, Hatada I, Zhang F, Nishimasu \nH & Nureki O (2019) Structural basis for the promiscuous PAM recognition by \nCorynebacterium diphtheriae Cas9. Nat Commun 10: 1968 \nIchikawa M, Saito K, Yanagisawa H -a., Yagi T, Kamiya R, Yamaguchi S, Yajima J, \nKushida Y, Nakano K, Numata O, et al (2015) Axonemal dynein light chain-1 locates \nat the microtubule-binding domain of the heavy chain. Molecular Biology of the Cell \n26: 4236–4247 \nIshikawa T (2016) Axoneme Structure from Motile Cilia. Cold Spring Harb Perspect Biol \nKutomi O, Yamamoto R, Hirose K, Mizuno K, Nakagiri Y, Imai H, Noga A, Obbineni JM, \nZimmermann N, Nakajima M, et al (2021) A dynein-associated photoreceptor protein \nprevents ciliary acclimation to blue light. Sci Adv 7: eabf3621 \nLeitner A, Sturm M, Hudecz O, Mazanek M, Smått J-H, Lindén M, Lindner W & Mechtler K \n(2010) Probing the phosphoproteome of HeLa cells using nanocast metal oxide \nmicrospheres for phosphopeptide enrichment. Anal Chem 82: 2726–2733 \nLeitner A, Walzthoeni T & Aebersold R (2014) Lysine-specific chemical cross-linking of \nprotein complexes and identification of cross-linking sites using LC-MS/MS and the \nxQuest/xProphet software pipeline. Nat Protoc 9: 120–137 \nLi X, Zhang R, Patena W, Gang SS, Blum SR, Ivanova N, Yue R, Robertson JM, Lefebvre \nPA, Fitz-Gibbon ST, et al (2016) An Indexed, Mapped Mutant Library Enables \nReverse Genetics Studies of Biological Processes in Chlamydomonas reinhardtii. \nPlant Cell 28: 367–387 \nOda T, Yanagisawa H, Kamiya R & Kikkawa M (2014a) Cilia and flagella. A molecular ruler \ndetermines the repeat length in eukaryotic cilia and flagella. Science 346: 857–860 \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint \n\nOda T, Yanagisawa H & Kikkawa M (2014b) Detailed Structural and biochemical \ncharacterization of the nexin-dynein regulatory complex. Mol Biol Cell \nOda T, Yanagisawa H, Yagi T & Kikkawa M (2014c) Mechanosignaling between central \napparatus and radial spokes controls axonemal dynein activity. J Cell Biol 204: 807–\n819 \nPazour GJ, Agrin N, Leszyk J & Witman GB (2005) Proteomic analysis of a eukaryotic \ncilium. J Cell Biol 170: 103–113 \nShin S-E, Lim J-M, Koh HG, Kim EK, Kang NK, Jeon S, Kwon S, Shin W-S, Lee B, \nHwangbo K, et al (2016) CRISPR/Cas9-induced knockout and knock-in mutations in \nChlamydomonas reinhardtii. Sci Rep 6: 27810 \nWallmeier J, Nielsen KG, Kuehni CE, Lucas JS, Leigh MW, Zariwala MA & Omran H \n(2020) Motile ciliopathies. Nat Rev Dis Primers 6: 77 \nWalton T, Gui M, Velkova S, Fassad MR, Hirst RA, Haarman E, O’Callaghan C, Bottier M, \nBurgoyne T, Mitchison HM, et al (2023) Axonemal structures reveal \nmechanoregulatory and disease mechanisms. Nature 618: 625–633 \nWitman GB (1986) Isolation of Chlamydomonas flagella and flagellar axonemes. Meth \nEnzymol 134: 280–290 \nZimmermann N, Noga A, Obbineni JM & Ishikawa T (2023) ATP-induced conformational \nchange of axonemal outer dynein arms revealed by cryo-electron tomography. EMBO \nJ 42: e112466 \n \n.CC-BY 4.0 International licensemade available under a \n(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 \nThe copyright holder for this preprintthis version posted April 3, 2024. ; https://doi.org/10.1101/2024.04.02.587733doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}