Primary cilia are microtubule structures that extend from the distal end of the mature, mother centriole. CEP164 is a component of the distal appendages carried by the mother centriole that is required for primary cilium formation. Recent data have implicated CEP164 as a ciliopathy gene and suggest that CEP164 plays some roles in the DNA damage response (DDR). We used reverse genetics to test the role of CEP164 in the DDR. We found that conditional depletion of CEP164 in chicken DT40 cells using an auxin-inducible degron led to no increase in sensitivity to DNA damage induced by ionising or ultraviolet irradiation. Disruption of CEP164 in human retinal pigmented epithelial cells blocked primary cilium formation but did not affect cellular proliferation or cellular responses to ionising or ultraviolet irradiation. Furthermore, we observed no localisation of CEP164 to the nucleus using immunofluorescence microscopy and analysis of multiple tagged forms of CEP164. Our data suggest that CEP164 is not required in the DDR.
Primary cilia are membrane-enclosed, microtubule-based organelles that extend like antennae from the surface of most mammalian cell types to sense and transduce various extracellular signals. They arise from the basal body, a template provided when the mature, mother centriole docks to the plasma membrane (Goetz and Anderson, 2010). Centrioles display structural polarity, with the proximal ends containing microtubule triplets that taper to doublets at the distal ends. The distal ends of mature centrioles carry two sets of appendages, which anchor cytoplasmic microtubules and which allow the docking of the mother centriole to the cell membrane during the formation of the primary cilium (Goetz and Anderson, 2010). The cilium core, the axoneme, consists of nine microtubule doublets that extend from the basal body.
In mammalian cells, cilium formation is closely regulated and linked to the cell cycle, as cilia must be resorbed to allow the basal body to act as a centrosome and to organise the mitotic spindle. Cellular quiescence, a temporary exit from the cell cycle that can be induced by the removal of growth factors, facilitates ciliogenesis (Kobayashi and Dynlacht, 2011). Current models associate primary cilia with cell cycle exit and reduced proliferation, although the underlying mechanisms of such a link are not well defined (Goto et al., 2013).
CEP164 encodes a centriolar appendage protein that is required for ciliogenesis (Graser et al., 2007; Schmidt et al., 2012). It has also been implicated in modulating the DNA damage response (DDR), particularly CHK1 (Sivasubramaniam et al., 2008). CEP164 was initially identified in a proteomic analysis of the centrosome and, later, as a component of the distal appendages whose depletion by small interfering RNA (siRNA) treatment caused a marked reduction in primary cilium formation (Andersen et al., 2003; Graser et al., 2007; Schmidt et al., 2012). Immunoelectron microscopy demonstrated the localisation of CEP164 to the distal end of the mother centriole (Graser et al., 2007). Dual photoactivated localization microscopy (PALM) and stochastic optical reconstruction microscopy (STORM) imaging has localised CEP164 in a ring around the centriole barrel with a periodic enrichment of the signal within the ring (Sillibourne et al., 2011), and stimulated emission depletion microscopy has found that the enriched CEP164 signal corresponds to nine symmetrically-arranged clusters around the centriole, indicative of its association with each of the nine distal appendages (Lau et al., 2012).
Recent data have indicated CEP164 mutations play a role in nephronophthisis-related ciliopathy, a rare recessive degenerative disease of the kidney, retina and brain, suggesting a link between ciliopathy and a DDR role for CEP164 (Chaki et al., 2012). We set out to explore the mechanisms that link ciliary dysfunction with DDR defects, using gene targeting to ablate CEP164 function.
RESULTS AND DISCUSSION
To analyse the roles of CEP164 in DNA repair, we used gene targeting in chicken DT40 cells to insert a tag that combined GFP with an auxin-inducible degron (AID; Nishimura et al., 2009) into the CEP164 locus of cells that stably expressed the TIR1 E3 ligase component (Fig. S1A,B). As shown in Fig. 1A, AID-GFP-tagged CEP164 localised to the centrosome, although we observed no localisation of CEP164 to the nucleus, even after UV irradiation of cells to levels that induced a substantial formation of γ-H2AX foci (Fig. 1B). Upon addition of auxin, AID-GFP-tagged CEP164 was depleted within 1 h (Fig. 1C; Fig. S1C,D). CEP164-deficient cells showed doubling times of 8.3 h (clone 1) and 8.3 h (clone 2), compared with control times of 8.4 h and 8.4 h for each clone, respectively, and 8.3 h for wild-type cells. We observed no difference in sensitivity to ionising radiation or UV treatment between CEP164-deficient and wild-type cells (Fig. 1D,E). In keeping with this observation, ionising-radiation-induced centrosome amplification, a potential readout for the DDR (Bourke et al., 2007), occurred to the same levels in both CEP164-deficient and wild-type cells (Fig. S1E). These data show that CEP164 plays a limited role, if any, in nuclear responses to ionising radiation or UV-induced DNA damage in DT40 cells.
Next, we cloned human CEP164 and expressed N- and C-terminally GFP- and FLAG-tagged versions in human cell lines. As shown in Fig. 2A–D, we consistently observed a centrosomal localisation for recombinant overexpressed CEP164, but saw no nuclear signal. Immunofluorescence microscopy with previously published anti-CEP164 antibodies (Graser et al., 2007) also detected centrosomal, but not nuclear signals (Fig. 2E,F). Next, we generated a new monoclonal antibody to CEP164. As shown in Fig. S2A, monoclonal antibody 1F3G10 generated against amino acids 6–296 of CEP164 recognised a protein of ∼200 kDa in three human cell lines. We next confirmed 1F3G10 specificity by depleting CEP164 from RPE1 cells using siRNA. After CEP164 depletion, the 1F3G10 signal disappeared, with no effect on a GAPDH control (Fig. 2B). In immunofluorescence microscopy experiments, 1F3G10 detected a signal that partly colocalised with ninein, a component of the subdistal appendages, but localised adjacent to CEP135, a centriole proximal end component (Fig. S2C), consistent with the known localisation of CEP164 at the distal appendages (Graser et al., 2007). Although we conclude from these experiments that the 1F3G10 monoclonal antibody is specific for CEP164, there was no signal seen in the nucleus of U2OS or RPE1 cells.
Previous data have indicated a role for CEP164 in primary ciliogenesis (Čajanek and Nigg, 2014; Graser et al., 2007; Schmidt et al., 2012; Chaki et al., 2012). Despite the feasibility of inducing ciliogenesis in DT40 cells (Prosser and Morrison, 2015), we prefer to examine the roles of CEP164 in a cell line with high levels of primary ciliation. Thus, we used CRISPR-Cas9 technology to disrupt CEP164 in hTERT-RPE1 cells, which show high levels of primary cilium formation upon serum starvation. We used a guide RNA designed to direct DNA double-strand breaks in exon 9 (the 7th coding exon) of the human CEP164 locus, and selected clones that had lost CEP164 expression by immunoblot analysis (Fig. 3A). Sequence analysis demonstrated that CEP164-deficient clones had incurred mutations in the CEP164 locus that led to premature stop codons being transcribed in-frame with the gene (Fig. S3A). Immunofluorescence microscopy confirmed that these clones no longer expressed CEP164, although they still carried intact centrioles (Fig. 3B). These cells proliferated as rapidly as wild-type cells, with doubling times of 24.1 h (clone 1) and 23.6 h (clone 2), compared with 23.5 h for wild-type cells. We saw no alteration in cell cycle distribution in the absence of CEP164 (Fig. S3B). Strikingly, CEP164-deficient cells showed a complete absence of primary ciliation capacity that was rescued by transgenic expression of CEP164 (Fig. 3C,D). Transmission electron microscopy analysis of the CEP164-null cells revealed no obvious structural defects in centriole structures, based on the dimensions of the centriole barrels (Fig. S3C). A total of 16 vesicles were identified in proximity to the centrioles in seven CEP164-null cells, but no docking was observed, consistent with a defect at the vesicle-docking stage in cilium formation seen in siRNA knockdown experiments (Schmidt et al., 2012) (Fig. 3E). Thus, our disruption of the CEP164 locus confirms the findings made on the roles of CEP164 in primary ciliogenesis with siRNA experiments (Čajanek and Nigg, 2014; Graser et al., 2007; Schmidt et al., 2012).
We next tested whether CEP164 deficiency impacted on the ability of cells to withstand UV-induced DNA damage. A clonogenic survival assay showed that CEP164-deficient RPE1 cells were no more sensitive than wild-type cells (Fig. 4A). In a positive control experiment, CETN2-null RPE1 cells (Prosser and Morrison, 2015) showed an increased UV sensitivity, as had centrin-deficient chicken DT40 cells (Dantas et al., 2011). Furthermore, we observed no localisation of CEP164 to nuclear DNA damage foci marked by γ-H2AX after ionising radiation or UV treatment in RPE1 or HeLa cells with either of the two antibodies in immunofluorescence experiments (Fig. 4B,C). Taken together, these data indicate no defect in the response to DNA damage in CEP164-deficient hTERT-RPE1 cells.
The results in the two models we have explored do not support a role for CEP164 in the DDR. We did not see a proliferative decline, such as that described in IMCD3 cells after siRNA knockdown of CEP164 (Chaki et al., 2012) or an acceleration of cell cycle progression, as has been described after siRNA knockdown of CEP164 in RPE-FUCCI cells (Slaats et al., 2014). In our null lines, we observed no elevated sensitivity to ionising radiation or UV irradiation in the absence of CEP164, which contrasts with the phenotypes of UV sensitivity and loss of the G2-to-M checkpoint reported with siRNA knockdown of CEP164 in HeLa cells (Sivasubramaniam et al., 2008; Pan and Lee, 2009). These discrepancies have potential implications for understanding how CEP164 mutations cause disease.
There are clear technical differences in the approaches that we have used and those previously reported. An obvious possibility is that the gradual or partial depletion imposed by siRNA treatment might lead to cellular responses different to those seen with the loss of a protein, although our degron-mediated experiment might have been expected to address this. Another possibility is that off-target effects of the siRNA treatments resulted in more marked phenotypes. Although the proliferative decline and cell cycle defects in IMCD3 cells were rescued by transgenic expression of human CEP164 (Chaki et al., 2012; Slaats et al., 2014), it is worth noting that rescues for the UV sensitivity and checkpoint defects seen in CEP164-knockdown cells were not performed (Sivasubramaniam et al., 2008; Pan and Lee, 2009), so that the specificity of these RNAi phenotypes cannot be assessed.
We have not seen any significant nuclear localisation of CEP164 during the normal cell cycle or after DNA damage in one chicken and three human cell lines, using three different antibodies and multiple differently-tagged versions of transgenically expressed CEP164. Similarly to published results (Graser et al., 2007; Schmidt et al., 2012), our experiments have detected only cytosolic or centrosomal signals, in contrast to the predominantly nuclear signals reported with those antibodies generated in the original study that implicated CEP164 in the DDR (Sivasubramaniam et al., 2008). Tagging experiments and several antibodies used in a recently published study have shown predominantly cytosolic or centrosomal CEP164 signals, although these authors also observed nuclear signals using the original CEP164 antibodies (Chaki et al., 2012). Controls for the specificity of the nuclear immunofluorescence signals seen with these reagents after CEP164 knockdown or depletion have not been detailed (Sivasubramaniam et al., 2008; Pan and Lee, 2009).
We have performed our DNA damage sensitivity and localisation analyses in cell lines from different tissues. Thus, although we cannot exclude the possibility that CEP164 contributes to the DDR in certain cell types, this does not appear to be a general activity of the protein. Our data, which support a marked defect in primary cilium formation, but normal levels of DNA repair capacity in the absence of CEP164, suggest that the principal cellular defect associated with CEP164 deficiency is the inability to undertake primary ciliogenesis.
MATERIALS AND METHODS
Chicken DT40 cells were cultured as previously described (Takata et al., 1998). hTERT-RPE1 cells were cultured as previously described (Prosser and Morrison, 2015). HeLa and U2OS cells were obtained from ATCC and cultured in Dulbecco's modified Eagle's medium (DMEM; Lonza or PAA/GE Healthcare), supplemented with 10% fetal calf serum (FCS; Lonza or Biochrom). Jurkat cells were from the European Collection of Animal Cell Cultures and were grown in RPMI with 10% FCS (Lonza). Auxin (Sigma-Aldrich) was prepared at 0.5 M in ethanol. Ionising radiation treatments used a 137Cs source (Mainance Engineering). For UV-C irradiation, cells were irradiated using an NU-6 254-nm UV-C lamp at 23 J/m2/min (Benda). DT40 clonogenic survival assays were performed as previously described (Takata et al., 1998), with 500 µM auxin added to the medium of cells 24 h prior to irradiation where a degron-tagged protein was to be depleted, and retained in the methylcellulose medium used for clonogenesis. For UV clonogenic survival assays in hTERT-RPE1, cells were counted before being serially diluted and plated in 10-cm dishes. The cells in each dish were allowed to adhere for 6 h before the medium was siphoned off and they were irradiated. Conditioned medium (filtered medium taken from 50% confluent cells) was used to replenish the dishes before incubation.
For targeting the chicken CEP164 locus, 5′ and 3′ homology arms and probe sequence were amplified from DT40 genomic DNA with KOD polymerase (Novagen/Merck) using the following primers: 5′ arm, 5′-gacgtcCAGACAACAAGCTAGGATATGTACCT-3′ and 5′-ccgcggGTACCGGTACACTTTAATTTGTCTGT-3′; 3′ arm, 5′-agatctAAGGTGGGACTTGGTGTT-TTCAGCC-3′ and 5′-cctaggTTTGGGTTTCAGTGCCATCCCGTG-3′; 5′ probe, 5′-CTTCTGATTTCAGTCCTGCGTGTT-3′ and 5′-CAGACATTAAATACAAGTCCCCTCC-3′ (lowercase letters represent non-genomic sequences used for cloning).
The probe for Southern analysis was labelled with digoxigenin using the PCR DIG Probe Synthesis Kit (Roche). AID-encoding sequence (Eykelenboom et al., 2013) was subcloned into pEGFP-N1 (BD Biosciences/Clontech) and a TIR1-9myc plasmid (pJE108; Eykelenboom et al., 2013) was stably cloned into DT40 cells to control the degron.
For cloning human CEP164 cDNA, hTERT-RPE1 RNA was extracted using TRI reagent (Invitrogen). Reverse transcription was performed using SuperScript First-Strand (Invitrogen) and PCR with KOD Hot Start. cDNAs were cloned into pGEM-T Easy (Promega), sequenced and then subcloned into pEGFP-N1, pEGFP-C1 (BD Biosciences/Clontech) or pCMV8 Tag 4A (Agilent Technologies, Santa Clara, CA, USA). The primers used to amplify human CEP164 cDNA (isoform 1, NP_055771.4) were as follows: 5′-aagcttATGGCTGGACGACCCCTCCGCA-3′ and 5′-gtcgacCAGAAGCGATACACCYYCACTC-3′ (lowercase letters represent sequence added for cloning). Isoform 2 (UniProt Q9UPV0, CE164_HUMAN) was cloned by mutating CEP164 cDNA isoform 1 using the QuikChange Lightning Site-Directed Mutagenesis Kit (#210518, Agilent) with the following primers: deletion of GGAG, 5′-AGCAGTCCAAAGGCCTGGAAGGTTATCTCCTC-3′ and 5′-GAGGAGATAACCTTCCAGGCCTTTGGACTGCT-3′; deletion of GTGAGTGGTGGCGGCAGCAGAGGATCGACTCAA, 5′-CCCCGCCTCACCCCCCGAGTC-TCA-3′ and 5′-TGAGACTCGGGGGGTGAGGCGGGG-3′; insertion of GGAGAGGTACCAT, 5′-AGCAGTCCAAAGGCCTGGAGGAGAGGTACCATAGGTTATCTCCTC-3′ and 5′-GAGGAGATAACCTATGGTACCTCTCCTCCAGGCCTTTGGACTGCT-3′; insertion of TCGACTCAA, 5′-CCCCGCCTCACCTCGACTCAACCCCGAGTCTCA-3′ and 5′-TGAGACTCGGGGTTGAGTCGAGGTGAGGCGGGG-3′.
CRISPR/Cas9 targeting of CEP164 in hTERT-RPE1 cells
Primers targeting exon 9 (Mali et al., 2013) were cloned into pX330-U6-Chimeric_BB-CBh-hSpCas9 (plasmid 43330; Addgene; Cong et al., 2013): 5′-CACCGCTGTTGGCAAAGGGCGACA-3′ and 5′-AAACTGTCGCCCTTTGCCCACAGC-3′. Transfections used Lipofectamine 2000 (Invitrogen). Genomic PCR products obtained with the diagnostic primer pair, 5′-CTGGGTGATTGATAACCATTGGG-3′ and 5′-CGCAAATGAAGCTCCTGACTCAGT-3′ were cloned into pGEM-T-Easy and sequenced.
Monoclonal antibody generation
cDNA sequence encoding CEP164 amino acids 6–296 was cloned into pGEX-4T1 (GE Healthcare) and the bacterially expressed GST fusion product was purified over a glutathione column prior to thrombin cleavage. Purified CEP164 protein fragment was used for hybridoma generation (Dundee Cell Products). Individual supernatants were screened by immunoblotting and microscopy and then concentrated antibody was purified from the best-performing 1F3G10 supernatant (Proteogenix).
Cells were transfected with 50 nmol custom siRNA targeting CEP164 from Qiagen (5′-CAGGUGACAUUUACUAUUUCA-3′) or Silencer Select siRNA targeting GAPDH (5′-UGGUUUACAUGUUCCAAUATT-3′) using Oligofectamine (Invitrogen).
Cells were fixed for analysis as previously described (Prosser and Morrison, 2015). Donkey and goat secondary antibodies were labelled with Cy3, Alexa Fluor 488 or Alexa Fluor 594 (Jackson ImmunoResearch or Molecular Probes). Rabbit polyclonal antibodies were against the following proteins: γ-tubulin (1:1000, T3559, Sigma) γ-H2AX (1:1000, Ab2893, Abcam), CEP135 (1:5000, 1420 739, Bird and Hyman, 2008), CEP164 (1:1000, HPA037606, Sigma), CEP164 (1:1000, R171, Graser et al., 2007), detyrosinated α-tubulin (1:1000, Ab48389, Abcam) and ninein (1:200, ab4447, Abcam). Mouse monoclonal antibodies used were against γ-tubulin (1:1000, GTU88, Sigma or TU-30/11–465-C100, Exbio), γ-H2AX (1:1000, JBW301, Upstate), CEP164 (1:100,000, 1F3G10) and centrin (1:1000, 20H5, Millipore). Images of DT40 and hTERT-RPE1 cells were captured on an IX71 microscope (Olympus) with a 100× oil objective, NA 1.35, using Volocity software (PerkinElmer), and are presented as maximum intensity projections of z-stacks after deconvolution. Alternatively, images of U2OS cells were captured and processed using an Axiovert 200 M microscope equipped with a Plan-Apochromat 63×, NA 1.4 objective and AxioVision software (Carl Zeiss Microscopy) and are presented as single sections.
hTERT-RPE1 cells were serum-starved in 0.1% FCS for 24 h prior to harvest. Cell pellets were prepared for transmission electron microscopy and imaged with an H-7000 Electron Microscope (Hitachi) as described (Prosser and Morrison, 2015).
Whole-cell extracts were prepared using RIPA buffer (50 mM Tris-HCl pH 7.4, 1% NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA and protease inhibitor cocktail). Immunoblot analyses used primary antibodies against the following proteins: α-tubulin (1:10,000, B512, Sigma), CEP164 (1:10,000, IF3G10), GFP (1:1000, 11814460001, Roche) and GAPDH (1:1000, 14C10, Cell Signaling).
Cells were fixed in 70% ice-cold ethanol overnight at 4°C, washed twice in PBS, and incubated in 40 µg/ml propidium iodide and 200 µg/ml RNase A in PBS for 1 h. Cytometry was performed on a FACSCanto (BD).
We acknowledge the National Biophotonics and Imaging Platform Ireland and the NCBES Flow Cytometry core facility, which were supported by Irish Government Programme for Research in Third-Level Institutions cycles 4 and 5.
O.M.D., K.K., A.K. and C.G.M. were responsible for project conception and direction. O.M.D., D.G., K.K., S.K., T.J.D., P.D., A.K. and C.G.M. undertook data analysis. O.M.D., T.J.D., K.K. (cell biology); D.G., S.K. (monoclonal antibody) and P.L. (electron micscopy) performed experimental work. O.M.D. and C.G.M. wrote the paper.
This work was funded by Science Foundation Ireland [Principal Investigator award 10/IN.1/B2972]; and the European Commission [grant number SEC-2009-4.3-02, project 242361 ‘BOOSTER’].
The authors declare no competing or financial interests.