Microtubule-bundling protein Spef1 enables mammalian ciliary central apparatus formation

Microtubule-bundling protein Spef1 enables mammalian ciliary central apparatus formation Abstract Cilia are cellular protrusions containing nine microtubule (MT) doublets and function to propel cell movement or extracellular liquid flow through beating or sense environmental stimuli through signal transductions. Cilia require the central pair (CP) apparatus, consisting of two CP MTs covered with projections of CP proteins, for planar strokes. How the CP MTs of such ‘9 + 2’ cilia are constructed, however, remains unknown. Here we identify Spef1, an evolutionarily conserved microtubule-bundling protein, as a core CP MT regulator in mammalian cilia. Spef1 was selectively expressed in mammalian cells with 9 + 2 cilia and specifically localized along the CP. Its depletion in multiciliated mouse ependymal cells by RNAi completely abolished the CP MTs and markedly attenuated ciliary localizations of CP proteins such as Hydin and Spag6, resulting in rotational beat of the ependymal cilia. Spef1, which binds to MTs through its N-terminal calponin-homologous domain, formed homodimers through its C-terminal coiled coil region to bundle and stabilize MTs. Disruption of either the MT-binding or the dimerization activity abolished the ability of exogenous Spef1 to restore the structure and functions of the CP apparatus. We propose that Spef1 bundles and stabilizes central MTs to enable the assembly and functions of the CP apparatus. central pair apparatus, cilium, microtubule, mouse ependymal cell, mouse tracheal epithelial cell, Spef1 Introduction Cilia are microtubule (MT)-based hair-like organelles built on specialized centrioles, or basal bodies. They initially emerge as motile cilia or flagella in protozoa to propel cell movement or extracellular liquid flow and are preserved in animal evolution. In mammals, motile cilia in epithelial cells of the trachea, oviduct, and ependyma play pivotal roles in mucus clearance, ovum transport, and cerebrospinal fluid circulation, respectively. Their defects in human cause primary ciliary dyskinesia (PCD), whose clinical symptoms include chronic rhinosinusitis, chronic bronchitis, infertility, and hydrocephalus (Fliegauf et al., 2007; Brooks and Wallingford, 2014; Praveen et al., 2015). The majority of motile cilia, including those of the aforementioned epithelial cells, contain a ‘9 + 2’ axoneme, which consists of two central (C1 and C2) MTs surrounded by nine peripheral MT doublets. Multiple proteinous projections associate asymmetrically with the CP MTs to constitute the CP apparatus, which contacts with radial spokes protruded from the peripheral doublets and forms a machinery to orchestrate ciliary motility by generating and transmitting mechanochemical signals to achieve spatiotemporal regulation of axonemal dynein activities (Carbajal-Gonzalez et al., 2013; Oda et al., 2014; Teves et al., 2016). Accordingly, the 9 + 2 type of motile cilia stroke in a back-and-forth manner with the beat directions vertical to the plane of the C1 and C2 MTs (Satir et al., 2014). Their CP-less mutants are paralyzed in the unicellular green algae Chlamydomonas reinhardtii (Witman et al., 1978; Smith and Lefebvre, 1997b; Smith and Yang, 2004) or become rotatory as seen in some PCD patients (Praveen et al., 2015). How the CP MTs are formed is poorly understood. MT formation usually requires generation of a short template (nucleation) followed by the addition of free tubulin preferentially to the plus end (elongation). Unlike the peripheral MT doublets, which are extended from the triplets of the basal body, the CP MTs are devoid of clearly defined templates. Although distinct pathways such as self-assembly and γ-tubulin-mediated nucleation are proposed in protozoa (Silflow et al., 1999; McKean et al., 2003; Dymek and Smith, 2012; Lechtreck et al., 2013), validations are still required. On the other hand, as MTs are dynamic but the central MTs are stable and usually exist in pair(s) (Gerdes et al., 2009), it is feasible to speculate that certain CP proteins may function to ensure proper number, elongation, and maintenance of the CP MTs. Indeed, in C. reinhardtii, mutations of the C1–C2 bridge-localized PF20p (Spag16) (Teves et al., 2016) have been found to cause up to 73% and 27% of 9 + 0 and 9 + 1 axonemes, respectively, when flagellar cross-sections were examined by electron microscopy (EM) (Adams et al., 1981; Smith and Lefebvre, 1997a). In addition, mutations of the C1 MT-associated PF16p (orthologue of vertebrate sperm-associated antigen 6, Spag6) (Teves et al., 2016) result in up to 55% and 60% of 9 + 0 and 9 + 1 axonemes, respectively (Dutcher et al., 1984; Smith and Lefebvre, 1996). Nevertheless, Spag6- or Spag16-deficient or even the double knockout mice display normal 9 + 2 axonemes (Sapiro et al., 2002; Zhang et al., 2006, 2007). Therefore, whether mammals, even metazoa, use completely different set of CP proteins to control their central MT formation or possess any common key regulator(s) are pressing issues to be clarified. In this study, we identified a novel CP protein, Spef1, as a pivotal regulator for the mammalian CP MT formation. Results Mammalian Spef1 is highly expressed in cells with 9 + 2 cilia To identify proteins essential for mammalian CP MTs assembly, we searched our cDNA array results (Xu et al., 2015) for MT-related genes that were upregulated during multicilia differentiation of mouse tracheal epithelial cells (mTECs) after being cultured at an air–liquid interface (ALI) (You et al., 2002; Zhao et al., 2013). Spef1 (also called CLAMP) (Figure 1A), an evolutionarily conserved microtubule stabilizer (Chan et al., 2005; Dougherty et al., 2005), was found to be highly expressed from ALI d3 in cultured mTECs and coincident with multicilia formation (Figure 1B and C). Similar to the end-binding (EB) family proteins, Spef1 contains an MT-binding calponin homology (CH) domain at its N-terminus and a coil-coiled (CC) region at its C-terminus (Figure 1A) (Bu and Su, 2003; Dougherty et al., 2005; Maurer et al., 2012). Exogenously expressed Spef1 has been shown to localize to motile cilia in Xenopus embryonic epidermis (Gray et al., 2009). Its ciliary function, however, was not documented. We thus performed detailed examination on this protein. Figure 1 View largeDownload slide Spef1 is highly expressed in multiciliated cells and tissues. (A) Diagram for mouse Spef1. CH, calponin homology; CC, coiled coil. (B) cDNA microarray results obtained using mRNAs isolated from mTECs, which were cultured at an air–liquid interface (ALI) for the indicated periods to induce multicilia formation, as probes. Bbs7, Ift52, and Spag6 are known cilia-related proteins and thus served as markers for multiciliogenesis. (C and D) Expression patterns of Spef1 in cultured mTECs (C) and 2-month-old mouse tissues (D). Gapdh or α-tubulin served as protein loading control. The increased levels of AC-tub in C indicate multicilia formation. (E) Experimental scheme for generation of mEPCs, through serum starvation (SS)-induced differentiation of glial cells from dissected telencephalon tissues. (F) Expression levels of Spef1 in cultured mEPCs and cell lines. NIH3T3, IMCD3, and hTERT-RPE1 cells that were serum starved for 48 h to induce primary cilia were also examined. α-tubulin served as loading control. Figure 1 View largeDownload slide Spef1 is highly expressed in multiciliated cells and tissues. (A) Diagram for mouse Spef1. CH, calponin homology; CC, coiled coil. (B) cDNA microarray results obtained using mRNAs isolated from mTECs, which were cultured at an air–liquid interface (ALI) for the indicated periods to induce multicilia formation, as probes. Bbs7, Ift52, and Spag6 are known cilia-related proteins and thus served as markers for multiciliogenesis. (C and D) Expression patterns of Spef1 in cultured mTECs (C) and 2-month-old mouse tissues (D). Gapdh or α-tubulin served as protein loading control. The increased levels of AC-tub in C indicate multicilia formation. (E) Experimental scheme for generation of mEPCs, through serum starvation (SS)-induced differentiation of glial cells from dissected telencephalon tissues. (F) Expression levels of Spef1 in cultured mEPCs and cell lines. NIH3T3, IMCD3, and hTERT-RPE1 cells that were serum starved for 48 h to induce primary cilia were also examined. α-tubulin served as loading control. In mouse tissues, Spef1 was highly abundant in the trachea, lung, oviduct, and testis (Figure 1D). When radial glia were isolated from P0 mouse brains and induced to differentiate into multiciliated ependymal cells (mEPCs) by serum starvation (Figure 1E) (Spassky et al., 2005; Delgehyr et al., 2015), Spef1 was also highly expressed in the mEPCs (Figure 1F). By contrast, it was only weakly expressed or undetectable in non-ciliated cycling cells or cells that were induced by serum deprivation to generate 9 + 0 primary cilia (Figure 1F). The expression of Spef1 is thus tightly correlated with the presence of 9 + 2 motile cilia. Spef1 is a CP-specific protein Super-resolution 3D-structured illumination microscopy (3D-SIM) using mEPCs collected at Day 7 post serum starvation (SS d7) revealed that Spef1 specifically localized to the central region of the 9 + 2 ependymal cilia, similar to Hydin, a C2 MT-associated protein (Figure 2A) (Lechtreck and Witman, 2007; Lechtreck et al., 2008; Teves et al., 2016). Both Spef1 and Hydin showed punctate distributions along the axonemes marked by acetylated tubulin (Ac-tubulin) (Figure 2A). By contrast, Spef1 was not detected in primary cilium (Figure 2B). Therefore, Spef1 is a novel CP-specific protein. Figure 2 View largeDownload slide Spef1 localizes to the CP in mEPCs. (A) Localization of Spef1 along the CP of 9 + 2 motile cilia. Cultured mEPCs at Day 7 post serum starvation (SS d7) were imaged with 3D-SIM. AC-tub and Hydin served as axonemal and CP markers, respectively. The framed region was magnified to show details. (B) Spef1 was absent in 9 + 0 primary cilia. hTERT-RPE1 cells were serum starved to induce primary cilia and subjected to immunostaining. AC-tub marked the ciliary axoneme. DAPI labeled the nucleus. (C–E) Expression and distribution of Spef1 during multiciliogenesis in mEPCs. Cultured mEPCs at SS d3 were immunostained and imaged with 3D-SIM. Cep152, Centrin, and AC-tub served as markers for the deuterosome, centriole, and ciliary axoneme, respectively, though Cep152 also localizes to the proximal side of the centrioles when released from deuterosomes. Schematic illustrations are provided in E to aid understanding. Figure 2 View largeDownload slide Spef1 localizes to the CP in mEPCs. (A) Localization of Spef1 along the CP of 9 + 2 motile cilia. Cultured mEPCs at Day 7 post serum starvation (SS d7) were imaged with 3D-SIM. AC-tub and Hydin served as axonemal and CP markers, respectively. The framed region was magnified to show details. (B) Spef1 was absent in 9 + 0 primary cilia. hTERT-RPE1 cells were serum starved to induce primary cilia and subjected to immunostaining. AC-tub marked the ciliary axoneme. DAPI labeled the nucleus. (C–E) Expression and distribution of Spef1 during multiciliogenesis in mEPCs. Cultured mEPCs at SS d3 were immunostained and imaged with 3D-SIM. Cep152, Centrin, and AC-tub served as markers for the deuterosome, centriole, and ciliary axoneme, respectively, though Cep152 also localizes to the proximal side of the centrioles when released from deuterosomes. Schematic illustrations are provided in E to aid understanding. We next examined expression and localization of Spef1 during basal body amplification and multiciliogenesis using mEPCs collected at SS d3. In mTECs, the process of basal body amplification can be divided into six stages (Tang, 2013; Zhao et al., 2013). mEPCs displayed a similar process, in which deuterosomes with 0–2 nascent centrioles initially appeared in stage II (Figure 2C and E), grew in size with increased numbers of centrioles in stage III (Yan et al., 2016), exhibited Cep152-positive protrusions in stage IV (Figure 2C and E), and then released their associated centrioles for ciliogenesis in stages V and VI (Figure 2C and E). Only in mEPCs, the stages V and VI were not clearly distinguishable because the characteristic basal body clustering in mTECs of stage VI (Supplementary Figure S1) was not observed in mEPCs. Immunofluorescent signals of Spef1 were only observed over the distal end of some centrioles in stages V and VI (Figure 2C and E). Immunostaining with Ac-tubulin indicated that the basal bodies in mEPCs of stages V and VI did not initiate ciliary assembly simultaneously (Figure 2C–E), similar to those in mTECs (Supplementary Figure S1). Spef1 emerged in growing cilia and frequently enriched at the ciliary tip (Figure 2D and E), potentially correlated with the assembly of CP MTs. Spef1 is required for the planar beating of ependymal cilia To understand whether Spef1 has a role in the CP, we depleted Spef1 in mEPCs using two different siRNAs (Spef-i1 and Spef-i2) (Figure 3A and B). High-speed live imaging indicated that Spef1 deficiency altered the back-and-forth stroke pattern of ependymal cilia to rotation (Figure 3C and Supplementary Movie S1). More than 84% of the Spef1-depleted cells displayed rotatory cilia, compared to a score of 3% in mEPCs treated with the control siRNA Ctrl-i (Figure 3D). To exclude off-target effect, we performed rescue experiments by ectopically expressing a siRNA-insensitive GFP-Spef1 in the Spef-i2-treated mEPCs through lentiviral infection (Figure 3A) and found that GFP-Spef1 restored the ciliary beat pattern. As a control, Centrin1-GFP had no rescue effect (Figure 3E and F; Supplementary Movie S2). These results further implicate a function of Spef1 in the CP formation because rotatory cilia are often found to lack the CP in PCD patients (Praveen et al., 2015). Figure 3 View largeDownload slide Depletion of Spef1 in mEPCs alters ciliary beat pattern from planar to rotational. (A) Experimental scheme for RNAi and rescue in mEPCs. Lentiviral infection was used to stably express GFP-tagged Spef1 or Centrin1. (B) Efficient depletion of Spef1 in mEPCs by RNAi. (C) Representative ciliary beat patterns. Trajectories of four cilia during the first 56 ms of live imaging are shown for each EPC, together with a diagram depicting either the planar or rotational beat pattern (also refer to Supplementary Movie S1). (D) Quantification results from three independent experiments. At least 40 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. (E and F) Expression of an RNAi-insensitive GFP-Spef1 in Spef-i2-treated mEPCs restored the planar ciliary beat pattern (refer to Supplementary Movie S2). Centrin1-GFP served as negative control. The quantification results were from three independent experiments. At least 33 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. Figure 3 View largeDownload slide Depletion of Spef1 in mEPCs alters ciliary beat pattern from planar to rotational. (A) Experimental scheme for RNAi and rescue in mEPCs. Lentiviral infection was used to stably express GFP-tagged Spef1 or Centrin1. (B) Efficient depletion of Spef1 in mEPCs by RNAi. (C) Representative ciliary beat patterns. Trajectories of four cilia during the first 56 ms of live imaging are shown for each EPC, together with a diagram depicting either the planar or rotational beat pattern (also refer to Supplementary Movie S1). (D) Quantification results from three independent experiments. At least 40 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. (E and F) Expression of an RNAi-insensitive GFP-Spef1 in Spef-i2-treated mEPCs restored the planar ciliary beat pattern (refer to Supplementary Movie S2). Centrin1-GFP served as negative control. The quantification results were from three independent experiments. At least 33 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. Spef1 is essential for the CP formation We then examined the cilia structures by transmission EM. While 95.6% ± 0.6% of the axonemal cross-sections from the control cilia displayed normal 9 + 2 configuration, 83.5% ± 4.0% of the ciliary cross-sections from mEPCs treated with Spef-i2 completely lacked both CP MTs (Figure 4A). The remaining 16.5% of the cilia had mostly two central MTs (15.5%) and more rarely one central MT (1.0%), possibly due to inefficient knockdown of Spef1 in these cilia. Therefore, we conclude that Spef1 is required for the formation of both CP MTs. Figure 4 View largeDownload slide Spef1 is essential for CP MT formation and CP protein retention in ependymal cilia. mEPCs were treated as in Figure 3A with Ctrl-i or Spef-i2 and analyzed. (A) Transmission EM revealed loss of CP MTs. The arrowheads indicate CP MTs in the control cilium. The quantification results were from two independent experiments. At least 50 cilia were scored in each experiment and condition. Error bars represent SD. (B) CP proteins Hydin and Spag6 were markedly reduced in the Spef-i2-treated mEPCs. Rsph4a, a radial spoke protein, served as negative control. The quantification results are presented as box plots. At least 100 cilia were quantified in each condition. The fluorescent intensity of the indicated proteins was normalized to that of AC-tub. Student’s t-test: ***P < 0.001; n.s., not significant. (C) Experimental scheme for cilia purification. (D) Spag6 levels declined in Spef1-depleted cilia but not total mEPCs. Lysates of total cells or ependymal cilia purified as in C were subjected to immunoblotting. Lamin B1 and Gapdh served as negative controls for the quality of cilia purification, whereas AC-tub served as a cilia marker. Note that the levels of Rsph4a and IFT52, a protein involved in intraflagellar transport, were not affected. Figure 4 View largeDownload slide Spef1 is essential for CP MT formation and CP protein retention in ependymal cilia. mEPCs were treated as in Figure 3A with Ctrl-i or Spef-i2 and analyzed. (A) Transmission EM revealed loss of CP MTs. The arrowheads indicate CP MTs in the control cilium. The quantification results were from two independent experiments. At least 50 cilia were scored in each experiment and condition. Error bars represent SD. (B) CP proteins Hydin and Spag6 were markedly reduced in the Spef-i2-treated mEPCs. Rsph4a, a radial spoke protein, served as negative control. The quantification results are presented as box plots. At least 100 cilia were quantified in each condition. The fluorescent intensity of the indicated proteins was normalized to that of AC-tub. Student’s t-test: ***P < 0.001; n.s., not significant. (C) Experimental scheme for cilia purification. (D) Spag6 levels declined in Spef1-depleted cilia but not total mEPCs. Lysates of total cells or ependymal cilia purified as in C were subjected to immunoblotting. Lamin B1 and Gapdh served as negative controls for the quality of cilia purification, whereas AC-tub served as a cilia marker. Note that the levels of Rsph4a and IFT52, a protein involved in intraflagellar transport, were not affected. To investigate whether CP proteins such as Hydin and Spag6 were affected, we examined their immunofluorescent signals by confocal microscopy. Compared to the bright ciliary signals of Spef1 in ctrl-i-treated mEPCs, only a few weak puncta were observed in the Spef-i2-treated mEPCs (Figure 4B). Quantifications showed that the ciliary Spef1 reduced by 4-fold upon RNAi as compared to the control cilia (Figure 4B). Apparently, such remaining amounts of Spef1 were not sufficient to support the CP formation. Accordingly, ciliary signals of Hydin and Spag6 also reduced by 5.4- and 2.5-fold, respectively, in the Spef-i2-treated mEPCs, whereas Rsph4a, a radial spoke protein (Pigino et al., 2011), was not affected (Figure 4B). To clarify whether the depletion of Spef1 caused down-regulation or reduced CP retention of the other CP proteins, we established a protocol to purify the ependymal cilia (Figure 4C). Immunoblotting revealed that the cilia preparations were abundant in ciliary markers such as Ac-tubulin, Rsph4a, and IFT52 but devoid of Gapdh and Lamin B1 (Figure 4D), suggesting the lack of cytoplasmic and nuclear contaminations. Although we were unable to detect Hydin due to its huge molecular mass (>500 kDa), both Spef1 and Spag6 were markedly downregulated in the Spef-i2-treated cilia as compared to the Ctrl-i-treated cilia (Figure 4D). Unlike Spef1, however, the protein levels of Spag6 in the total cell lysates were not altered (Figure 4D). Taken together, these IF and WB results suggest a reduced ciliary retention of CP proteins such as Hydin and Spag6 upon the depletion of Spef1. Spef1 bundles and stabilizes MTs in vivo and in vitro As the CP MTs naturally appear in pair, we speculated that Spef1 may bundle and stabilize them. Consistent with the previous reports (Dougherty et al., 2005; Werner et al., 2014), transiently expressed GFP-Spef1, but not GFP, bundled MTs in U2OS cells (Figure 5A). Furthermore, the MTs in the GFP-Spef1-expressing cells were highly acetylated (Figure 5B), indicating that they were stabilized MTs (Janke and Bulinski, 2011; Portran et al., 2017). Figure 5 View largeDownload slide Spef1 bundles and stabilizes MTs both in vivo and in vitro. (A) GFP-Spef1 bundled MTs in U2OS cells. GFP served as negative control. (B) MTs bundles in GFP-Spef1-expressing cells were highly acetylated. (C) Proteins purified from E. coli for in vitro MT bundling assays. The lanes were cropped from the same SDS-PAGE gel. (D) Spef1 alone was sufficient to bundle MTs. MTs polymerized in vitro were incubated with the indicated concentrations of purified GST, GST-Spef1, or His-Spef1 at 37°C for 10 min and then imaged. Figure 5 View largeDownload slide Spef1 bundles and stabilizes MTs both in vivo and in vitro. (A) GFP-Spef1 bundled MTs in U2OS cells. GFP served as negative control. (B) MTs bundles in GFP-Spef1-expressing cells were highly acetylated. (C) Proteins purified from E. coli for in vitro MT bundling assays. The lanes were cropped from the same SDS-PAGE gel. (D) Spef1 alone was sufficient to bundle MTs. MTs polymerized in vitro were incubated with the indicated concentrations of purified GST, GST-Spef1, or His-Spef1 at 37°C for 10 min and then imaged. To clarify whether Spef1 alone was sufficient to induce MT bundling, we performed in vitro MT bundling assays. We purified GST-or polyhistidine (His)-tagged Spef1 from Escherichia coli (Figure 5C) and assembled rhodamine-labeled and Taxol-stabilized MTs in vitro using purified bovine tubulin. When GST, GST-Spef1, and His-Spef1 of various concentrations were incubated with the stabilized MTs, both GST-Spef1 and His-Spef1 induced MT bundling in a concentration-dependent manner, resulting in massive MT bundles at 4 μM (Figure 5D). By contrast, GST was not effective (Figure 5D). Thus, Spef1 alone is sufficient to bundle MTs. Both the MT-binding and bundling activities of Spef1 are required for CP formation Next we analyzed how different regions of Spef1 contribute to its functions. Arg31 in the CH domain of Spef1 is an evolutionarily conserved amino acid residue from Chlamydomonas to human (Figure 6A; Chan et al., 2005). We found that mutating Arg31 into Ala disrupted both the MT-binding and bundling activities of Spef1 in cells (Figure 6A and B). Consistent with the previous report (Dougherty et al., 2005), Spef1ΔCC, a mutant lacking its C-terminal CC region, bound to MTs but failed to bundle them (Figure 6A and B). Co-immunoprecipitation (co-IP) experiments showed that FLAG-Spef1 complexed with GFP-Spef1, but not GFP-ΔCC (Figure 6C), suggesting that the CC domain enables the MT bundling activity of Spef1 by mediating homodimerization. Therefore, the CH domain is responsible for the MT-binding activity of Spef1, while both the CH and CC domains are required for bundling MTs. Figure 6 View largeDownload slide Spef1 bundles MTs through homodimerization. (A) Diagrams of Spef1 and its mutants and summary of their properties based on the results in B and C. The conserved ‘R’ in the sequence alignments, which was mutated into ‘A’ in the mouse Spef1R31A mutant, is shown in red. (B) Spef1R31A failed to bind to MTs, whereas and Spef1ΔCC failed to bundle MTs when transiently expressed in U2OS cells. (C) Spef1 homodimerized through its CC region. The indicated proteins were transiently expressed in HEK293T cells. Co-IP was then performed using anti-FLAG resin. The GFP-containing proteins were visualized by using anti-GFP antibody. Figure 6 View largeDownload slide Spef1 bundles MTs through homodimerization. (A) Diagrams of Spef1 and its mutants and summary of their properties based on the results in B and C. The conserved ‘R’ in the sequence alignments, which was mutated into ‘A’ in the mouse Spef1R31A mutant, is shown in red. (B) Spef1R31A failed to bind to MTs, whereas and Spef1ΔCC failed to bundle MTs when transiently expressed in U2OS cells. (C) Spef1 homodimerized through its CC region. The indicated proteins were transiently expressed in HEK293T cells. Co-IP was then performed using anti-FLAG resin. The GFP-containing proteins were visualized by using anti-GFP antibody. Next we performed rescue experiments to examine functions of different mutants in motile cilia by ectopically expressing GFP-Spef1, GFP-Spef1R31A, and GFP-Spef1ΔCC in Spef-i2-treated mEPCs (Figure 7A; see Figure 3A for experimental scheme). Despite their similar expression levels (Figure 7A), confocal microscopy and quantifications indicated that the ciliary immunofluorescent signals of GFP-Spef1 were 2- and 8.4-fold over those of GFP-Spef1R31A and GFP-Spef1ΔCC, respectively (Figure 7B and C). Accordingly, the ciliary signals of endogenous Hydin and Spag6 showed similar tendencies, whereas those of Rsph4a were not altered (Figure 7B and C). Figure 7 View largeDownload slide Spef1R31A or Spef1ΔCC failed to rescue the CP formation in Spef1-depleted ependymal cilia. (A) Expression of RNAi-insensitive GFP-Spef1 and mutants in Spef1-depleted mEPCs (refer to Figure 3A for the experimental procedure). The samples were immunoblotted with anti-Spef1 antibody. Gapdh served as loading control. (B and C) Ciliary localizations of Spag6, Hydin, and Rsph4a in Spef-i2-treated mEPCs expressing GFP-Spef1 or its mutants. AC-tub labels ciliary axonemes. The quantification results are presented as box plots. At least 100 cilia were quantified in each condition. The fluorescent intensity of the indicated proteins was normalized to that of AC-tub. Student’s t-test, **P < 0.01; ***P < 0.001; n.s., not significant. (D and E) GFP-Spef1, but not Spef1R31A, restored the CP MTs (arrowheads). The statistical results were from three independent experiments. At least 101 cilia were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. (F and G) GFP-Spef1, but not its mutants, rescued the ciliary beat defect. Trajectories of four cilia in the first 56 ms of live imaging are shown for each EPC, together with a diagram depicting either the planar or rotational beat pattern (also refer to Supplementary Movie S3). The quantification results were from three independent experiments. At least 30 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. Figure 7 View largeDownload slide Spef1R31A or Spef1ΔCC failed to rescue the CP formation in Spef1-depleted ependymal cilia. (A) Expression of RNAi-insensitive GFP-Spef1 and mutants in Spef1-depleted mEPCs (refer to Figure 3A for the experimental procedure). The samples were immunoblotted with anti-Spef1 antibody. Gapdh served as loading control. (B and C) Ciliary localizations of Spag6, Hydin, and Rsph4a in Spef-i2-treated mEPCs expressing GFP-Spef1 or its mutants. AC-tub labels ciliary axonemes. The quantification results are presented as box plots. At least 100 cilia were quantified in each condition. The fluorescent intensity of the indicated proteins was normalized to that of AC-tub. Student’s t-test, **P < 0.01; ***P < 0.001; n.s., not significant. (D and E) GFP-Spef1, but not Spef1R31A, restored the CP MTs (arrowheads). The statistical results were from three independent experiments. At least 101 cilia were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. (F and G) GFP-Spef1, but not its mutants, rescued the ciliary beat defect. Trajectories of four cilia in the first 56 ms of live imaging are shown for each EPC, together with a diagram depicting either the planar or rotational beat pattern (also refer to Supplementary Movie S3). The quantification results were from three independent experiments. At least 30 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. As GFP-Spef1R31A was able to mildly both enter the cilia and rescue the ciliary signals of Hydin and Spag6, we investigated whether the GFP-Spef1R31A-expressing cilia contain CP MTs by using transmission EM. To ensure that the GFP-tagged proteins were expressed in the majority of the Spef1-depleted mEPCs, we infected the cells with the protein-expression lentivirus for three times and found that >86% of the cells readily became GFP-positive at SS d7 (Supplementary Figure S2). We found that 93.2% ± 4.4% of the cilia in the GFP-Spef1 samples contained two central MTs (Figure 7D and E). The remaining 6.8% had no central MT(s). By contrast, only 16.5% ± 7.3% of the cilia in the GFP-Spef1R31A samples contained 1 (2.1%) or 2 (14.4%) central MTs, whereas the remaining 83.5% had no central MT (Figure 7D and E), similar to the samples treated with Spef-i2 alone (Figure 4A). Thus, unlike Spef1, Spef1R31A is unable to rescue CP MTs. Finally, we examined ciliary motilities. We found that neither GFP-Spef1ΔCC nor GFP-Spef1R31A restored the ciliary beat pattern from rotational to planar as did GFP-Spef1 (Figure 7F and G; Supplementary Movie S3). Taken together, we conclude that both the MT-binding and bundling activities of Spef1 are required for its functions in the formation of CP MTs and CP apparatus. Discussion In this study, we demonstrated that Spef1 is a key regulator for the construction of the 9 + 2 axoneme. The incidence of CP-less cilia (83.5% ± 4.0%) in the Spef1 RNAi samples (Figure 4A) was similar to those of rotational ones (Figure 3D), suggesting that the remaining CP-containing cilia can be attributed to inefficient RNAi or failure in transfection (Figures 3B and 4B). Thus, we conclude that Spef1 is essential for the CP MT formation. Spef1 is structurally similar to EB proteins in that both contain an N-terminal MT-binding CH domain and a C-terminal CC domain for homodimerization and protein interactions (Figure 6) (Bu and Su, 2003; Dougherty et al., 2005; Sen et al., 2013). When purified Spef1 on beads was incubated with free tubulin, it did not exhibit MT-nucleation activities (data not shown). On the other hand, both EB proteins and Spef1 distribute evenly along MTs to bundle and stabilize MTs upon overexpression (Figures 5 and 6) (Bu and Su, 2003; Ligon et al., 2003; Dougherty et al., 2005). EB proteins also promote MT growth by binding between protofilaments at the plus end of MTs (Ligon et al., 2003; Maurer et al., 2012; Zhang et al., 2015). Endogenous Spef1 distributed rather evenly as puncta along the CP in mature cilia (e.g. those at SS d7) (Figure 2A) but preferably at ciliary tips in growing cilia at SS d3 (Figure 2D). The plus ends of the CP MTs are oriented toward the ciliary tip (Euteneuer and McIntosh, 1981). Therefore, the tip-specific enrichment of Spef1 suggests that Spef1 may exhibit plus-end binding in growing cilia. As both the MT-binding and bundling activities of Spef1 are required for CP MT formation (Figures 6 and 7), Spef1 may bundle the CP MTs nucleated by unknown factor(s) to keep the MTs as a pair, facilitate the CP MT elongation, and prevent them from disassembly during the growth of cilia. We found that Spef1 is also important for ciliary retentions of other CP proteins such as Spag6 and Hydin (Figures 4B, D and 7B, C). By contrast, non-CP ciliary proteins such as Rsph4a and IFT52 were not affected (Figure 4B and D). The MT binding-defective mutant Spef1R31A failed to rescue the CP MTs but partly rescued the ciliary localizations of Spag6 and Hydin as compared to the wild-type Spef1 and the non-effective Spef1ΔCC (Figure 7B and C). GFP-Spef1R31A itself also exhibited moderate ciliary localization as compared to GFP-Spef1 (positive control) and GFP-Spef1ΔCC (negative control) (Figure 7B and C). These results imply that Spef1 might facilitate ciliary entry or retention of Spag6 and Hydin in an MT binding-independent manner and the CP MTs further immobilize these proteins for the assembly of the CP apparatus. Apparently, future studies are required to clarify the detailed mechanisms. To our knowledge, Spef1 is so far the first documented CP protein essential for the CP MT formation. As both the CH and CC domains of Spef1 are highly conserved from protozoans to mammals (Chan et al., 2005), the CP-related functions of Spef1 are possibly evolutionarily conserved in protozoan. By searching for Spef1-associated proteins, additional regulators for the CP MT formation may be unraveled in both protozoan and mammals. In C. reinhardtii, three mutants (pf15, pf18, and pf19) have been found to completely lack the CP MTs (Adams et al., 1981). While the mutated gene of pf18 is still unknown, those of pf15 and pf19 encode PF15p and PF19p, which are orthologues of the regulatory and catalytic subunits of Katanin, an MT-severing enzyme. PF15p and PF19p, however, are not CP-localized and have been speculated to provide the initial templates or free tubulin for CP MTs (Dymek et al., 2004; Dymek and Smith, 2012; Lechtreck et al., 2013). It will be interesting to understand how these different pathways cooperate to achieve proper CP MT formation. Materials and methods Plasmids The full-length or partial cDNAs for Spef1 (NM_027641) were amplified by PCR from total cDNAs from mouse testis and constructed into pEGFP-C1, pcDNA3.1-NFLAG, pET28a, or pGEX4T-1 to express GFP-, FLAG-, His-, or GST-tagged fusion proteins. pLV-EGFP-C1 was used for lentiviral expression of GFP-fusion proteins. The cDNAs for the R31A mutant (228 CGG → GCG) and the RNAi-insensitive constructs of Spef1 (776 ACTTAAGAACGTGCGC → GCTCAAAAATGTACGT) were generated by PCR. The full-length cDNAs of mouse Rsph4a (NM_001162957), Spef1, and mouse cDNA fragments coding for amino acids 2312–2602 of Hydin (NM_172916) were constructed into pET28a to express antigens for antibody production. All the constructs were verified by sequencing. Cell culture, transfection, and viral infection Cells were maintained at 37°C in an atmosphere containing 5% CO2. Unless otherwise indicated, the culture medium was Dulbecco’s Modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (Biochrom), 0.3 mg/ml glutamine (Sigma), 100 U/ml penicillin (Invitrogen), and 100 U/ml streptomycin (Invitrogen). mTECs were isolated and cultured as described previously (You et al., 2002; Zhao et al., 2013). mEPCs were obtained as described (Delgehyr et al., 2015), with some modifications. The telencephala of P0 C57BL/6 J mice was dissected and digested with freshly prepared Papain solution (10 U/ml; Worthington Biochemical Corporation) for 30 min at 37°C. After gentle pipetting with a P1000 tip and centrifugation at 1400 rpm for 5 min at room temperature, the pelleted cells were resuspended in the culture medium and seeded into 25 cm2 laminin-coated flasks (1 brain/flask). After culturing for 1 day, neurons were shaken off and removed. The remaining radial glia-enriched cells were further cultured to ~90% confluency (usually 4 days) and then transferred into the wells of laminin-coated 29-mm glass-bottom dishes (Cellvis, D29-14-1.5-N) at a density of 2 × 105 cells per well. The cells were then maintained in serum-free medium to induce differentiation into EPCs. Generally, 30%–60% of the cells became multiciliated at SS d5 or later. To express exogenous proteins, HEK 293 T and U2OS cells were transfected using the calcium phosphate method and Lipofectamine 2000 (Life Technologies), respectively, for 48 h. Lentiviral production and infection were performed as described previously (Zhao et al., 2013). Cultured mouse radial glial cells were infected at SS d−1 unless otherwise stated. For RNAi, siRNA were transfected into radial glia/EPCs using Lipofectamine RNAiMAX (Life Technologies) at SS d−1 and d3, respectively. A total of 4 μl oligonucleotides (20 μM) and 6 μl Lipofectamine were used for a well of cells in a 29-mm glass-bottom dish. For rescue experiments, radial glia were infected with lentivirus at SS d−2 to express RNAi-insensitive GFP-Spef1 and its mutants or Centrin1-GFP, in addition to the transfections with siRNA. Sequences of siRNAs are listed as follows: Spef-i1: 5′-GAUGCACAAUUAUGUCCCUtt-3′; Spef-i2: 5′-CAACUUAAGAACGUGCGCAtt-3′; Ctrl-i: 5′-UUCUCCGAACGUGUCACGUtt-3′. Purification of ependymal cilia Cells from the telencephala of 10 P0 mice were seeded into two 75 cm2 laminin-coated flasks. After removing neurons, the cells were further cultured to near confluency (usually 5 days), followed by serum starvation to induce differentiation into mEPCs. The EPCs were harvested at SS d7 by centrifugation at 2000 rpm for 5 min at 4°C. The cells were resuspended in 3 ml of deciliation buffer (20 mM PIPES, pH 5.5, 250 mM sucrose, 20 mM CaCl2, 0.05% Triton X-100) and agitated vigorously for 10 min on a vortex mixer (G560OE, Scientific Industries). After centrifugation at 3000 rpm for 5 min at 4°C, the supernatant, which contained cilia, was collected. This step was repeated once. The cilia were then pelleted by centrifugation at 20000× g for 30 min, washed twice with 1 ml PBS, and resuspended in 100 μl of lysis buffer (20 mM Tris-HCl, pH 7.5, 100 mM KCl, 0.1% NP-40, 1 mM EDTA, 10 mM Na4O7P2, and protease inhibitors). The samples were used for mass spectrometric or immunoblotting analysis. Light microscopy mEPCs were pre-extracted with 0.5% Triton X-100 in PBS for 30 sec, followed by fixation with 4% paraformaldehyde in PBS for 10 min at room temperature and permeabilization with 0.5% Triton X-100 for 15 min. U2OS cells were fixed with −20°C methanol for 3 min and then 75% ethanol at 4°C for 10 min. Immunofluorescent staining was carried out as described (Zhao et al., 2013). GFP signals in mEPCs were visualized by immunostaining using anti-GFP antibody. Antibodies used are listed in Supplementary Table S1. Confocal images were captured by using Leica TCS SP8 system with a 63×/1.40 oil immersion objective and processed with maximum intensity projections. 3D-SIM super-resolution images were taken on DeltaVision OMX SR system (GE Healthcare) with a 100×/1.40 oil immersion objective (Olympus). Immersion oils with refractive indices of 1.516 were used to obtain optimal images and serial z-stack sectioning was set at 125 nm intervals. Images were reconstructed by softWoRx 5.0 software (GE Healthcare). Fluorescence intensity of ciliary proteins was measured using Adobe Photoshop Software. After the background subtraction, the mean fluorescent intensity of the interested proteins was normalized to the fluorescent intensity of Ac-tubulin. Ciliary motilities were recorded at 140 fps (frames per second) by using an Andor Neo sCMOS camera on Olympus IX71 microscope with a 63×/1.40 oil immersion objective (Xu et al., 2015). Cilia trajectories were illustrated with the manual tracking plugin in the Image J software. Four traceable cilia in each cell were tracked over a period of 56 ms. Electron microscopy For transmission EM, cultured EPCs were fixed in 2.5% glutaraldehyde overnight at 4°C, washed with PBS, and treated with 1% OsO4 for 30 min at room temperature. The samples were dehydrated with graded ethanol series and embedded in Epon 812 resin. The 70-nm ultrathin sections were stained with 1% lead citrate and 2% uranyl acetate. Images were captured at 80 kV using a Tecnai G2 Spirit transmission electron microscope (FEI). Immunoprecipitation Co-IP was performed as described previously (Zhao et al., 2013). Briefly, HEK293T cells transfected for 48 h were lysed with the lysis buffer containing 10% Glycerol. Pre-cleared cell lysates were incubated with anti-FLAG beads (Sigma, A2220) or anti-GFP beads (Allele Biotechnology) for 3 h at 4°C. After three times of wash with the lysis buffer and three times with wash buffer (20 mM Tris-HCl, pH 7.5, 150 mM KCl, 0.5% NP-40, 1 mM EDTA, 10 mM Na4P2O7, 10% glycerol), proteins on the anti-FLAG beads were eluted with 30 μl of 1 mg/ml FLAG peptide, while those on the GFP beads were directly denatured in 30 μl of SDS loading buffer. In vitro MT bundling MTs were polymerized in vitro as described (Jiang et al., 2015). Briefly, 2 μl of bovine tubulin stock (tubulin: rhodamine-labeled tubulin = 5:1) (Cytoskeleton, Inc.) were diluted in 2 μl of reaction buffer (80 mM PIPES, pH 6.9, 1 mM GTP, 2 mM MgCl2, 0.5 mM EGTA, 40% glycerol) to yield a final concentration of 20 μM tubulin. The mixture was incubated at 35°C for 15 min, diluted with 40 μl of pre-warmed (25°C) stabilization buffer (80 mM PIPES, pH 7.0, 20 μM Taxol, 2 mM MgCl2, 0.5 mM EGTA), and further incubated at 25°C for 30 min. To assay for MT bundling ability, 2 μl of the Taxol-stabilized MTs was mixed with 6 μl purified GST, GST-Spef1, or His-Spef1 of varying concentrations. After incubation at 37°C for 10 min, 3 μl of each mixture were gently squashed under an 18-mm circular coverslip. Microscopic examination and imaging were performed using Olympus BX51 microscope with a 100×/1.35 oil immersion objective and QImaging Retiga EXi CCD. Statistics Microscopic and biochemical results were repeated at least twice. Quantification results are presented as mean ± standard deviation (SD) unless otherwise stated. Two-tailed Student’s t-test (GraphPad Prism software) was used to calculate P-values between unpaired samples. Differences were considered significant when P < 0.05. Only results from three or more independent experiments were applied to the t-tests. In the box plots, the bottom and the top of the box represent the 25th and 75th percentiles, respectively. The band is the median. The ends of the whiskers indicate the maximum and minimum of the data. Supplementary material Supplementary material is available at Journal of Molecular Cell Biology online. Acknowledgements The authors thank Xingping Zhu and Tong Yin for technical support on 3D-SIM and Fengling Qin and Fangjie Qi for instrument support and technical assistance. Funding This work was equally supported by the National Natural Science Foundation of China (31330045), National Key R&D Program of China (2017YFA0503500), and Chinese Academy of Sciences (XDB19020000). It was also supported by the National Natural Science Foundation of China (31601092 to L.Z. and 31771495 to X.Y.). Conflict of interest: none declared. References Adams , G.M. , Huang , B. , Piperno , G. , et al. . ( 1981 ). Central-pair microtubular complex of Chlamydomonas flagella: polypeptide composition as revealed by analysis of mutants . J. Cell Biol. 91 , 69 – 76 . Google Scholar CrossRef Search ADS PubMed Brooks , E.R. , and Wallingford , J.B. ( 2014 ). Multiciliated cells . Curr. Biol. 24 , R973 – R982 . Google Scholar CrossRef Search ADS PubMed Bu , W. , and Su , L.K. ( 2003 ). Characterization of functional domains of human EB1 family proteins . J. Biol. Chem. 278 , 49721 – 49731 . Google Scholar CrossRef Search ADS PubMed Carbajal-Gonzalez , B.I. , Heuser , T. , Fu , X. , et al. . ( 2013 ). Conserved structural motifs in the central pair complex of eukaryotic flagella . Cytoskeleton 70 , 101 – 120 . Google Scholar CrossRef Search ADS PubMed Chan , S.W. , Fowler , K.J. , Choo , K.H. , et al. . ( 2005 ). Spef1, a conserved novel testis protein found in mouse sperm flagella . Gene 353 , 189 – 199 . Google Scholar CrossRef Search ADS PubMed Delgehyr , N. , Meunier , A. , Faucourt , M. , et al. . ( 2015 ). Ependymal cell differentiation, from monociliated to multiciliated cells . Methods Cell Biol. 127 , 19 – 35 . Google Scholar CrossRef Search ADS PubMed Dougherty , G.W. , Adler , H.J. , Rzadzinska , A. , et al. . ( 2005 ). CLAMP, a novel microtubule-associated protein with EB-type calponin homology . Cell Motil. Cytoskeleton 62 , 141 – 156 . Google Scholar CrossRef Search ADS PubMed Dutcher , S.K. , Huang , B. , and Luck , D.J. ( 1984 ). Genetic dissection of the central pair microtubules of the flagella of Chlamydomonas reinhardtii . J. Cell Biol. 98 , 229 – 236 . Google Scholar CrossRef Search ADS PubMed Dymek , E.E. , Lefebvre , P.A. , and Smith , E.F. ( 2004 ). PF15p is the chlamydomonas homologue of the Katanin p80 subunit and is required for assembly of flagellar central microtubules . Eukaryot. Cell 3 , 870 – 879 . Google Scholar CrossRef Search ADS PubMed Dymek , E.E. , and Smith , E.F. ( 2012 ). PF19 encodes the p60 catalytic subunit of katanin and is required for assembly of the flagellar central apparatus in Chlamydomonas . J. Cell Sci. 125 , 3357 – 3366 . Google Scholar CrossRef Search ADS PubMed Euteneuer , U. , and McIntosh , J.R. ( 1981 ). Polarity of some motility-related microtubules . Proc. Natl Acad. Sci. USA 78 , 372 – 376 . Google Scholar CrossRef Search ADS Fliegauf , M. , Benzing , T. , and Omran , H. ( 2007 ). When cilia go bad: cilia defects and ciliopathies . Nat. Rev. Mol. Cell Biol. 8 , 880 – 893 . Google Scholar CrossRef Search ADS PubMed Gerdes , J.M. , Davis , E.E. , and Katsanis , N. ( 2009 ). The vertebrate primary cilium in development, homeostasis, and disease . Cell 137 , 32 – 45 . Google Scholar CrossRef Search ADS PubMed Gray , R.S. , Abitua , P.B. , Wlodarczyk , B.J. , et al. . ( 2009 ). The planar cell polarity effector Fuz is essential for targeted membrane trafficking, ciliogenesis and mouse embryonic development . Nat. Cell Biol. 11 , 1225 – 1232 . Google Scholar CrossRef Search ADS PubMed Janke , C. , and Bulinski , J.C. ( 2011 ). Post-translational regulation of the microtubule cytoskeleton: mechanisms and functions . Nat. Rev. Mol. Cell Biol. 12 , 773 – 786 . Google Scholar CrossRef Search ADS PubMed Jiang , H. , Wang , S. , Huang , Y. , et al. . ( 2015 ). Phase transition of spindle-associated protein regulate spindle apparatus assembly . Cell 163 , 108 – 122 . Google Scholar CrossRef Search ADS PubMed Lechtreck , K.F. , Delmotte , P. , Robinson , M.L. , et al. . ( 2008 ). Mutations in Hydin impair ciliary motility in mice . J. Cell Biol. 180 , 633 – 643 . Google Scholar CrossRef Search ADS PubMed Lechtreck , K.F. , Gould , T.J. , and Witman , G.B. ( 2013 ). Flagellar central pair assembly in Chlamydomonas reinhardtii . Cilia 2 , 15 . Google Scholar CrossRef Search ADS PubMed Lechtreck , K.F. , and Witman , G.B. ( 2007 ). Chlamydomonas reinhardtii hydin is a central pair protein required for flagellar motility . J. Cell Biol. 176 , 473 – 482 . Google Scholar CrossRef Search ADS PubMed Ligon , L.A. , Shelly , S.S. , Tokito , M. , et al. . ( 2003 ). The microtubule plus-end proteins EB1 and dynactin have differential effects on microtubule polymerization . Mol. Biol. Cell 14 , 1405 – 1417 . Google Scholar CrossRef Search ADS PubMed Maurer , S.P. , Fourniol , F.J. , Bohner , G. , et al. . ( 2012 ). EBs recognize a nucleotide-dependent structural cap at growing microtubule ends . Cell 149 , 371 – 382 . Google Scholar CrossRef Search ADS PubMed McKean , P.G. , Baines , A. , Vaughan , S. , et al. . ( 2003 ). Gamma-tubulin functions in the nucleation of a discrete subset of microtubules in the eukaryotic flagellum . Curr. Biol. 13 , 598 – 602 . Google Scholar CrossRef Search ADS PubMed Oda , T. , Yanagisawa , H. , Yagi , T. , et al. . ( 2014 ). Mechanosignaling between central apparatus and radial spokes controls axonemal dynein activity . J. Cell Biol. 204 , 807 – 819 . Google Scholar CrossRef Search ADS PubMed Pigino , G. , Bui , K.H. , Maheshwari , A. , et al. . ( 2011 ). Cryoelectron tomography of radial spokes in cilia and flagella . J. Cell Biol. 195 , 673 – 687 . Google Scholar CrossRef Search ADS PubMed Portran , D. , Schaedel , L. , Xu , Z.J. , et al. . ( 2017 ). Tubulin acetylation protects long-lived microtubules against mechanical ageing . Nat. Cell Biol. 19 , 391 – 398 . Google Scholar CrossRef Search ADS PubMed Praveen , K. , Davis , E.E. , and Katsanis , N. ( 2015 ). Unique among ciliopathies: primary ciliary dyskinesia, a motile cilia disorder . F1000Prime Rep. 7 , 36 . Google Scholar CrossRef Search ADS PubMed Sapiro , R. , Kostetskii , I. , Olds-Clarke , P. , et al. . ( 2002 ). Male infertility, impaired sperm motility, and hydrocephalus in mice deficient in sperm-associated antigen 6 . Mol. Cell. Biol. 22 , 6298 – 6305 . Google Scholar CrossRef Search ADS PubMed Satir , P. , Heuser , T. , and Sale , W.S. ( 2014 ). A structural basis for how motile cilia beat . Bioscience 64 , 1073 – 1083 . Google Scholar CrossRef Search ADS PubMed Sen , I. , Veprintsev , D. , Akhmanova , A. , et al. . ( 2013 ). End binding proteins are obligatory dimers . PLoS One 8 , e74448 . Google Scholar CrossRef Search ADS PubMed Silflow , C.D. , Liu , B. , LaVoie , M. , et al. . ( 1999 ). Gamma-tubulin in Chlamydomonas: characterization of the gene and localization of the gene product in cells . Cell Motil. Cytoskeleton 42 , 285 – 297 . Google Scholar CrossRef Search ADS PubMed Smith , E.F. , and Lefebvre , P.A. ( 1996 ). PF16 encodes a protein with armadillo repeats and localizes to a single microtubule of the central apparatus in Chlamydomonas flagella . J. Cell Biol. 132 , 359 – 370 . Google Scholar CrossRef Search ADS PubMed Smith , E.F. , and Lefebvre , P.A. ( 1997 a). PF20 gene product contains WD repeats and localizes to the intermicrotubule bridges in Chlamydomonas flagella . Mol. Biol. Cell 8 , 455 – 467 . Google Scholar CrossRef Search ADS PubMed Smith , E.F. , and Lefebvre , P.A. ( 1997 b). The role of central apparatus components in flagellar motility and microtubule assembly . Cell Motil. Cytoskeleton 38 , 1 – 8 . Google Scholar CrossRef Search ADS PubMed Smith , E.F. , and Yang , P. ( 2004 ). The radial spokes and central apparatus: mechano-chemical transducers that regulate flagellar motility . Cell Motil. Cytoskeleton 57 , 8 – 17 . Google Scholar CrossRef Search ADS PubMed Spassky , N. , Merkle , F.T. , Flames , N. , et al. . ( 2005 ). Adult ependymal cells are postmitotic and are derived from radial glial cells during embryogenesis . J. Neurosci. 25 , 10 – 18 . Google Scholar CrossRef Search ADS PubMed Tang , T.K. ( 2013 ). Centriole biogenesis in multiciliated cells . Nat. Cell Biol. 15 , 1400 – 1402 . Google Scholar CrossRef Search ADS PubMed Teves , M.E. , Nagarkatti-Gude , D.R. , Zhang , Z. , et al. . ( 2016 ). Mammalian axoneme central pair complex proteins: broader roles revealed by gene knockout phenotypes . Cytoskeleton 73 , 3 – 22 . Google Scholar CrossRef Search ADS PubMed Werner , M.E. , Mitchell , J.W. , Putzbach , W. , et al. . ( 2014 ). Radial intercalation is regulated by the Par complex and the microtubule-stabilizing protein CLAMP/Spef1 . J. Cell Biol. 206 , 367 – 376 . Google Scholar CrossRef Search ADS PubMed Witman , G.B. , Plummer , J. , and Sander , G. ( 1978 ). Chlamydomonas flagellar mutants lacking radial spokes and central tubules. Structure, composition, and function of specific axonemal components . J Cell Biol . 76 , 729 – 747 . Google Scholar CrossRef Search ADS PubMed Xu , Y. , Cao , J. , Huang , S. , et al. . ( 2015 ). Characterization of tetratricopeptide repeat-containing proteins critical for cilia formation and function . PLoS One 10 , e0124378 . Google Scholar CrossRef Search ADS PubMed Yan , X. , Zhao , H. , and Zhu , X. ( 2016 ). Production of basal bodies in bulk for dense multicilia formation . F1000Res. 5 , pii: F1000 Faculty Rev-1533. You , Y. , Richer , E.J. , Huang , T. , et al. . ( 2002 ). Growth and differentiation of mouse tracheal epithelial cells: selection of a proliferative population . Am. J. Physiol. Lung Cell. Mol. Physiol. 283 , L1315 – L1321 . Google Scholar CrossRef Search ADS PubMed Zhang , R. , Alushin , G.M. , Brown , A. , et al. . ( 2015 ). Mechanistic origin of microtubule dynamic instability and its modulation by EB proteins . Cell 162 , 849 – 859 . Google Scholar CrossRef Search ADS PubMed Zhang , Z. , Kostetskii , I. , Tang , W. , et al. . ( 2006 ). Deficiency of SPAG16L causes male infertility associated with impaired sperm motility . Biol. Reprod. 74 , 751 – 759 . Google Scholar CrossRef Search ADS PubMed Zhang , Z. , Tang , W. , Zhou , R. , et al. . ( 2007 ). Accelerated mortality from hydrocephalus and pneumonia in mice with a combined deficiency of SPAG6 and SPAG16L reveals a functional interrelationship between the two central apparatus proteins . Cell Motil. Cytoskeleton 64 , 360 – 376 . Google Scholar CrossRef Search ADS PubMed Zhao , H. , Zhu , L. , Zhu , Y. , et al. . ( 2013 ). The Cep63 paralogue Deup1 enables massive de novo centriole biogenesis for vertebrate multiciliogenesis . Nat. Cell Biol. 15 , 1434 – 1444 . Google Scholar CrossRef Search ADS PubMed © The Author(s) (2018). Published by Oxford University Press on behalf of Journal of Molecular Cell Biology, IBCB, SIBS, CAS. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Molecular Cell Biology Oxford University Press

Microtubule-bundling protein Spef1 enables mammalian ciliary central apparatus formation

Loading next page...
 
/lp/ou_press/microtubule-bundling-protein-spef1-enables-mammalian-ciliary-central-X2X90ADaZw
Publisher
Oxford University Press
Copyright
© The Author(s) (2018). Published by Oxford University Press on behalf of Journal of Molecular Cell Biology, IBCB, SIBS, CAS. All rights reserved.
ISSN
1674-2788
eISSN
1759-4685
D.O.I.
10.1093/jmcb/mjy014
Publisher site
See Article on Publisher Site

Abstract

Abstract Cilia are cellular protrusions containing nine microtubule (MT) doublets and function to propel cell movement or extracellular liquid flow through beating or sense environmental stimuli through signal transductions. Cilia require the central pair (CP) apparatus, consisting of two CP MTs covered with projections of CP proteins, for planar strokes. How the CP MTs of such ‘9 + 2’ cilia are constructed, however, remains unknown. Here we identify Spef1, an evolutionarily conserved microtubule-bundling protein, as a core CP MT regulator in mammalian cilia. Spef1 was selectively expressed in mammalian cells with 9 + 2 cilia and specifically localized along the CP. Its depletion in multiciliated mouse ependymal cells by RNAi completely abolished the CP MTs and markedly attenuated ciliary localizations of CP proteins such as Hydin and Spag6, resulting in rotational beat of the ependymal cilia. Spef1, which binds to MTs through its N-terminal calponin-homologous domain, formed homodimers through its C-terminal coiled coil region to bundle and stabilize MTs. Disruption of either the MT-binding or the dimerization activity abolished the ability of exogenous Spef1 to restore the structure and functions of the CP apparatus. We propose that Spef1 bundles and stabilizes central MTs to enable the assembly and functions of the CP apparatus. central pair apparatus, cilium, microtubule, mouse ependymal cell, mouse tracheal epithelial cell, Spef1 Introduction Cilia are microtubule (MT)-based hair-like organelles built on specialized centrioles, or basal bodies. They initially emerge as motile cilia or flagella in protozoa to propel cell movement or extracellular liquid flow and are preserved in animal evolution. In mammals, motile cilia in epithelial cells of the trachea, oviduct, and ependyma play pivotal roles in mucus clearance, ovum transport, and cerebrospinal fluid circulation, respectively. Their defects in human cause primary ciliary dyskinesia (PCD), whose clinical symptoms include chronic rhinosinusitis, chronic bronchitis, infertility, and hydrocephalus (Fliegauf et al., 2007; Brooks and Wallingford, 2014; Praveen et al., 2015). The majority of motile cilia, including those of the aforementioned epithelial cells, contain a ‘9 + 2’ axoneme, which consists of two central (C1 and C2) MTs surrounded by nine peripheral MT doublets. Multiple proteinous projections associate asymmetrically with the CP MTs to constitute the CP apparatus, which contacts with radial spokes protruded from the peripheral doublets and forms a machinery to orchestrate ciliary motility by generating and transmitting mechanochemical signals to achieve spatiotemporal regulation of axonemal dynein activities (Carbajal-Gonzalez et al., 2013; Oda et al., 2014; Teves et al., 2016). Accordingly, the 9 + 2 type of motile cilia stroke in a back-and-forth manner with the beat directions vertical to the plane of the C1 and C2 MTs (Satir et al., 2014). Their CP-less mutants are paralyzed in the unicellular green algae Chlamydomonas reinhardtii (Witman et al., 1978; Smith and Lefebvre, 1997b; Smith and Yang, 2004) or become rotatory as seen in some PCD patients (Praveen et al., 2015). How the CP MTs are formed is poorly understood. MT formation usually requires generation of a short template (nucleation) followed by the addition of free tubulin preferentially to the plus end (elongation). Unlike the peripheral MT doublets, which are extended from the triplets of the basal body, the CP MTs are devoid of clearly defined templates. Although distinct pathways such as self-assembly and γ-tubulin-mediated nucleation are proposed in protozoa (Silflow et al., 1999; McKean et al., 2003; Dymek and Smith, 2012; Lechtreck et al., 2013), validations are still required. On the other hand, as MTs are dynamic but the central MTs are stable and usually exist in pair(s) (Gerdes et al., 2009), it is feasible to speculate that certain CP proteins may function to ensure proper number, elongation, and maintenance of the CP MTs. Indeed, in C. reinhardtii, mutations of the C1–C2 bridge-localized PF20p (Spag16) (Teves et al., 2016) have been found to cause up to 73% and 27% of 9 + 0 and 9 + 1 axonemes, respectively, when flagellar cross-sections were examined by electron microscopy (EM) (Adams et al., 1981; Smith and Lefebvre, 1997a). In addition, mutations of the C1 MT-associated PF16p (orthologue of vertebrate sperm-associated antigen 6, Spag6) (Teves et al., 2016) result in up to 55% and 60% of 9 + 0 and 9 + 1 axonemes, respectively (Dutcher et al., 1984; Smith and Lefebvre, 1996). Nevertheless, Spag6- or Spag16-deficient or even the double knockout mice display normal 9 + 2 axonemes (Sapiro et al., 2002; Zhang et al., 2006, 2007). Therefore, whether mammals, even metazoa, use completely different set of CP proteins to control their central MT formation or possess any common key regulator(s) are pressing issues to be clarified. In this study, we identified a novel CP protein, Spef1, as a pivotal regulator for the mammalian CP MT formation. Results Mammalian Spef1 is highly expressed in cells with 9 + 2 cilia To identify proteins essential for mammalian CP MTs assembly, we searched our cDNA array results (Xu et al., 2015) for MT-related genes that were upregulated during multicilia differentiation of mouse tracheal epithelial cells (mTECs) after being cultured at an air–liquid interface (ALI) (You et al., 2002; Zhao et al., 2013). Spef1 (also called CLAMP) (Figure 1A), an evolutionarily conserved microtubule stabilizer (Chan et al., 2005; Dougherty et al., 2005), was found to be highly expressed from ALI d3 in cultured mTECs and coincident with multicilia formation (Figure 1B and C). Similar to the end-binding (EB) family proteins, Spef1 contains an MT-binding calponin homology (CH) domain at its N-terminus and a coil-coiled (CC) region at its C-terminus (Figure 1A) (Bu and Su, 2003; Dougherty et al., 2005; Maurer et al., 2012). Exogenously expressed Spef1 has been shown to localize to motile cilia in Xenopus embryonic epidermis (Gray et al., 2009). Its ciliary function, however, was not documented. We thus performed detailed examination on this protein. Figure 1 View largeDownload slide Spef1 is highly expressed in multiciliated cells and tissues. (A) Diagram for mouse Spef1. CH, calponin homology; CC, coiled coil. (B) cDNA microarray results obtained using mRNAs isolated from mTECs, which were cultured at an air–liquid interface (ALI) for the indicated periods to induce multicilia formation, as probes. Bbs7, Ift52, and Spag6 are known cilia-related proteins and thus served as markers for multiciliogenesis. (C and D) Expression patterns of Spef1 in cultured mTECs (C) and 2-month-old mouse tissues (D). Gapdh or α-tubulin served as protein loading control. The increased levels of AC-tub in C indicate multicilia formation. (E) Experimental scheme for generation of mEPCs, through serum starvation (SS)-induced differentiation of glial cells from dissected telencephalon tissues. (F) Expression levels of Spef1 in cultured mEPCs and cell lines. NIH3T3, IMCD3, and hTERT-RPE1 cells that were serum starved for 48 h to induce primary cilia were also examined. α-tubulin served as loading control. Figure 1 View largeDownload slide Spef1 is highly expressed in multiciliated cells and tissues. (A) Diagram for mouse Spef1. CH, calponin homology; CC, coiled coil. (B) cDNA microarray results obtained using mRNAs isolated from mTECs, which were cultured at an air–liquid interface (ALI) for the indicated periods to induce multicilia formation, as probes. Bbs7, Ift52, and Spag6 are known cilia-related proteins and thus served as markers for multiciliogenesis. (C and D) Expression patterns of Spef1 in cultured mTECs (C) and 2-month-old mouse tissues (D). Gapdh or α-tubulin served as protein loading control. The increased levels of AC-tub in C indicate multicilia formation. (E) Experimental scheme for generation of mEPCs, through serum starvation (SS)-induced differentiation of glial cells from dissected telencephalon tissues. (F) Expression levels of Spef1 in cultured mEPCs and cell lines. NIH3T3, IMCD3, and hTERT-RPE1 cells that were serum starved for 48 h to induce primary cilia were also examined. α-tubulin served as loading control. In mouse tissues, Spef1 was highly abundant in the trachea, lung, oviduct, and testis (Figure 1D). When radial glia were isolated from P0 mouse brains and induced to differentiate into multiciliated ependymal cells (mEPCs) by serum starvation (Figure 1E) (Spassky et al., 2005; Delgehyr et al., 2015), Spef1 was also highly expressed in the mEPCs (Figure 1F). By contrast, it was only weakly expressed or undetectable in non-ciliated cycling cells or cells that were induced by serum deprivation to generate 9 + 0 primary cilia (Figure 1F). The expression of Spef1 is thus tightly correlated with the presence of 9 + 2 motile cilia. Spef1 is a CP-specific protein Super-resolution 3D-structured illumination microscopy (3D-SIM) using mEPCs collected at Day 7 post serum starvation (SS d7) revealed that Spef1 specifically localized to the central region of the 9 + 2 ependymal cilia, similar to Hydin, a C2 MT-associated protein (Figure 2A) (Lechtreck and Witman, 2007; Lechtreck et al., 2008; Teves et al., 2016). Both Spef1 and Hydin showed punctate distributions along the axonemes marked by acetylated tubulin (Ac-tubulin) (Figure 2A). By contrast, Spef1 was not detected in primary cilium (Figure 2B). Therefore, Spef1 is a novel CP-specific protein. Figure 2 View largeDownload slide Spef1 localizes to the CP in mEPCs. (A) Localization of Spef1 along the CP of 9 + 2 motile cilia. Cultured mEPCs at Day 7 post serum starvation (SS d7) were imaged with 3D-SIM. AC-tub and Hydin served as axonemal and CP markers, respectively. The framed region was magnified to show details. (B) Spef1 was absent in 9 + 0 primary cilia. hTERT-RPE1 cells were serum starved to induce primary cilia and subjected to immunostaining. AC-tub marked the ciliary axoneme. DAPI labeled the nucleus. (C–E) Expression and distribution of Spef1 during multiciliogenesis in mEPCs. Cultured mEPCs at SS d3 were immunostained and imaged with 3D-SIM. Cep152, Centrin, and AC-tub served as markers for the deuterosome, centriole, and ciliary axoneme, respectively, though Cep152 also localizes to the proximal side of the centrioles when released from deuterosomes. Schematic illustrations are provided in E to aid understanding. Figure 2 View largeDownload slide Spef1 localizes to the CP in mEPCs. (A) Localization of Spef1 along the CP of 9 + 2 motile cilia. Cultured mEPCs at Day 7 post serum starvation (SS d7) were imaged with 3D-SIM. AC-tub and Hydin served as axonemal and CP markers, respectively. The framed region was magnified to show details. (B) Spef1 was absent in 9 + 0 primary cilia. hTERT-RPE1 cells were serum starved to induce primary cilia and subjected to immunostaining. AC-tub marked the ciliary axoneme. DAPI labeled the nucleus. (C–E) Expression and distribution of Spef1 during multiciliogenesis in mEPCs. Cultured mEPCs at SS d3 were immunostained and imaged with 3D-SIM. Cep152, Centrin, and AC-tub served as markers for the deuterosome, centriole, and ciliary axoneme, respectively, though Cep152 also localizes to the proximal side of the centrioles when released from deuterosomes. Schematic illustrations are provided in E to aid understanding. We next examined expression and localization of Spef1 during basal body amplification and multiciliogenesis using mEPCs collected at SS d3. In mTECs, the process of basal body amplification can be divided into six stages (Tang, 2013; Zhao et al., 2013). mEPCs displayed a similar process, in which deuterosomes with 0–2 nascent centrioles initially appeared in stage II (Figure 2C and E), grew in size with increased numbers of centrioles in stage III (Yan et al., 2016), exhibited Cep152-positive protrusions in stage IV (Figure 2C and E), and then released their associated centrioles for ciliogenesis in stages V and VI (Figure 2C and E). Only in mEPCs, the stages V and VI were not clearly distinguishable because the characteristic basal body clustering in mTECs of stage VI (Supplementary Figure S1) was not observed in mEPCs. Immunofluorescent signals of Spef1 were only observed over the distal end of some centrioles in stages V and VI (Figure 2C and E). Immunostaining with Ac-tubulin indicated that the basal bodies in mEPCs of stages V and VI did not initiate ciliary assembly simultaneously (Figure 2C–E), similar to those in mTECs (Supplementary Figure S1). Spef1 emerged in growing cilia and frequently enriched at the ciliary tip (Figure 2D and E), potentially correlated with the assembly of CP MTs. Spef1 is required for the planar beating of ependymal cilia To understand whether Spef1 has a role in the CP, we depleted Spef1 in mEPCs using two different siRNAs (Spef-i1 and Spef-i2) (Figure 3A and B). High-speed live imaging indicated that Spef1 deficiency altered the back-and-forth stroke pattern of ependymal cilia to rotation (Figure 3C and Supplementary Movie S1). More than 84% of the Spef1-depleted cells displayed rotatory cilia, compared to a score of 3% in mEPCs treated with the control siRNA Ctrl-i (Figure 3D). To exclude off-target effect, we performed rescue experiments by ectopically expressing a siRNA-insensitive GFP-Spef1 in the Spef-i2-treated mEPCs through lentiviral infection (Figure 3A) and found that GFP-Spef1 restored the ciliary beat pattern. As a control, Centrin1-GFP had no rescue effect (Figure 3E and F; Supplementary Movie S2). These results further implicate a function of Spef1 in the CP formation because rotatory cilia are often found to lack the CP in PCD patients (Praveen et al., 2015). Figure 3 View largeDownload slide Depletion of Spef1 in mEPCs alters ciliary beat pattern from planar to rotational. (A) Experimental scheme for RNAi and rescue in mEPCs. Lentiviral infection was used to stably express GFP-tagged Spef1 or Centrin1. (B) Efficient depletion of Spef1 in mEPCs by RNAi. (C) Representative ciliary beat patterns. Trajectories of four cilia during the first 56 ms of live imaging are shown for each EPC, together with a diagram depicting either the planar or rotational beat pattern (also refer to Supplementary Movie S1). (D) Quantification results from three independent experiments. At least 40 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. (E and F) Expression of an RNAi-insensitive GFP-Spef1 in Spef-i2-treated mEPCs restored the planar ciliary beat pattern (refer to Supplementary Movie S2). Centrin1-GFP served as negative control. The quantification results were from three independent experiments. At least 33 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. Figure 3 View largeDownload slide Depletion of Spef1 in mEPCs alters ciliary beat pattern from planar to rotational. (A) Experimental scheme for RNAi and rescue in mEPCs. Lentiviral infection was used to stably express GFP-tagged Spef1 or Centrin1. (B) Efficient depletion of Spef1 in mEPCs by RNAi. (C) Representative ciliary beat patterns. Trajectories of four cilia during the first 56 ms of live imaging are shown for each EPC, together with a diagram depicting either the planar or rotational beat pattern (also refer to Supplementary Movie S1). (D) Quantification results from three independent experiments. At least 40 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. (E and F) Expression of an RNAi-insensitive GFP-Spef1 in Spef-i2-treated mEPCs restored the planar ciliary beat pattern (refer to Supplementary Movie S2). Centrin1-GFP served as negative control. The quantification results were from three independent experiments. At least 33 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. Spef1 is essential for the CP formation We then examined the cilia structures by transmission EM. While 95.6% ± 0.6% of the axonemal cross-sections from the control cilia displayed normal 9 + 2 configuration, 83.5% ± 4.0% of the ciliary cross-sections from mEPCs treated with Spef-i2 completely lacked both CP MTs (Figure 4A). The remaining 16.5% of the cilia had mostly two central MTs (15.5%) and more rarely one central MT (1.0%), possibly due to inefficient knockdown of Spef1 in these cilia. Therefore, we conclude that Spef1 is required for the formation of both CP MTs. Figure 4 View largeDownload slide Spef1 is essential for CP MT formation and CP protein retention in ependymal cilia. mEPCs were treated as in Figure 3A with Ctrl-i or Spef-i2 and analyzed. (A) Transmission EM revealed loss of CP MTs. The arrowheads indicate CP MTs in the control cilium. The quantification results were from two independent experiments. At least 50 cilia were scored in each experiment and condition. Error bars represent SD. (B) CP proteins Hydin and Spag6 were markedly reduced in the Spef-i2-treated mEPCs. Rsph4a, a radial spoke protein, served as negative control. The quantification results are presented as box plots. At least 100 cilia were quantified in each condition. The fluorescent intensity of the indicated proteins was normalized to that of AC-tub. Student’s t-test: ***P < 0.001; n.s., not significant. (C) Experimental scheme for cilia purification. (D) Spag6 levels declined in Spef1-depleted cilia but not total mEPCs. Lysates of total cells or ependymal cilia purified as in C were subjected to immunoblotting. Lamin B1 and Gapdh served as negative controls for the quality of cilia purification, whereas AC-tub served as a cilia marker. Note that the levels of Rsph4a and IFT52, a protein involved in intraflagellar transport, were not affected. Figure 4 View largeDownload slide Spef1 is essential for CP MT formation and CP protein retention in ependymal cilia. mEPCs were treated as in Figure 3A with Ctrl-i or Spef-i2 and analyzed. (A) Transmission EM revealed loss of CP MTs. The arrowheads indicate CP MTs in the control cilium. The quantification results were from two independent experiments. At least 50 cilia were scored in each experiment and condition. Error bars represent SD. (B) CP proteins Hydin and Spag6 were markedly reduced in the Spef-i2-treated mEPCs. Rsph4a, a radial spoke protein, served as negative control. The quantification results are presented as box plots. At least 100 cilia were quantified in each condition. The fluorescent intensity of the indicated proteins was normalized to that of AC-tub. Student’s t-test: ***P < 0.001; n.s., not significant. (C) Experimental scheme for cilia purification. (D) Spag6 levels declined in Spef1-depleted cilia but not total mEPCs. Lysates of total cells or ependymal cilia purified as in C were subjected to immunoblotting. Lamin B1 and Gapdh served as negative controls for the quality of cilia purification, whereas AC-tub served as a cilia marker. Note that the levels of Rsph4a and IFT52, a protein involved in intraflagellar transport, were not affected. To investigate whether CP proteins such as Hydin and Spag6 were affected, we examined their immunofluorescent signals by confocal microscopy. Compared to the bright ciliary signals of Spef1 in ctrl-i-treated mEPCs, only a few weak puncta were observed in the Spef-i2-treated mEPCs (Figure 4B). Quantifications showed that the ciliary Spef1 reduced by 4-fold upon RNAi as compared to the control cilia (Figure 4B). Apparently, such remaining amounts of Spef1 were not sufficient to support the CP formation. Accordingly, ciliary signals of Hydin and Spag6 also reduced by 5.4- and 2.5-fold, respectively, in the Spef-i2-treated mEPCs, whereas Rsph4a, a radial spoke protein (Pigino et al., 2011), was not affected (Figure 4B). To clarify whether the depletion of Spef1 caused down-regulation or reduced CP retention of the other CP proteins, we established a protocol to purify the ependymal cilia (Figure 4C). Immunoblotting revealed that the cilia preparations were abundant in ciliary markers such as Ac-tubulin, Rsph4a, and IFT52 but devoid of Gapdh and Lamin B1 (Figure 4D), suggesting the lack of cytoplasmic and nuclear contaminations. Although we were unable to detect Hydin due to its huge molecular mass (>500 kDa), both Spef1 and Spag6 were markedly downregulated in the Spef-i2-treated cilia as compared to the Ctrl-i-treated cilia (Figure 4D). Unlike Spef1, however, the protein levels of Spag6 in the total cell lysates were not altered (Figure 4D). Taken together, these IF and WB results suggest a reduced ciliary retention of CP proteins such as Hydin and Spag6 upon the depletion of Spef1. Spef1 bundles and stabilizes MTs in vivo and in vitro As the CP MTs naturally appear in pair, we speculated that Spef1 may bundle and stabilize them. Consistent with the previous reports (Dougherty et al., 2005; Werner et al., 2014), transiently expressed GFP-Spef1, but not GFP, bundled MTs in U2OS cells (Figure 5A). Furthermore, the MTs in the GFP-Spef1-expressing cells were highly acetylated (Figure 5B), indicating that they were stabilized MTs (Janke and Bulinski, 2011; Portran et al., 2017). Figure 5 View largeDownload slide Spef1 bundles and stabilizes MTs both in vivo and in vitro. (A) GFP-Spef1 bundled MTs in U2OS cells. GFP served as negative control. (B) MTs bundles in GFP-Spef1-expressing cells were highly acetylated. (C) Proteins purified from E. coli for in vitro MT bundling assays. The lanes were cropped from the same SDS-PAGE gel. (D) Spef1 alone was sufficient to bundle MTs. MTs polymerized in vitro were incubated with the indicated concentrations of purified GST, GST-Spef1, or His-Spef1 at 37°C for 10 min and then imaged. Figure 5 View largeDownload slide Spef1 bundles and stabilizes MTs both in vivo and in vitro. (A) GFP-Spef1 bundled MTs in U2OS cells. GFP served as negative control. (B) MTs bundles in GFP-Spef1-expressing cells were highly acetylated. (C) Proteins purified from E. coli for in vitro MT bundling assays. The lanes were cropped from the same SDS-PAGE gel. (D) Spef1 alone was sufficient to bundle MTs. MTs polymerized in vitro were incubated with the indicated concentrations of purified GST, GST-Spef1, or His-Spef1 at 37°C for 10 min and then imaged. To clarify whether Spef1 alone was sufficient to induce MT bundling, we performed in vitro MT bundling assays. We purified GST-or polyhistidine (His)-tagged Spef1 from Escherichia coli (Figure 5C) and assembled rhodamine-labeled and Taxol-stabilized MTs in vitro using purified bovine tubulin. When GST, GST-Spef1, and His-Spef1 of various concentrations were incubated with the stabilized MTs, both GST-Spef1 and His-Spef1 induced MT bundling in a concentration-dependent manner, resulting in massive MT bundles at 4 μM (Figure 5D). By contrast, GST was not effective (Figure 5D). Thus, Spef1 alone is sufficient to bundle MTs. Both the MT-binding and bundling activities of Spef1 are required for CP formation Next we analyzed how different regions of Spef1 contribute to its functions. Arg31 in the CH domain of Spef1 is an evolutionarily conserved amino acid residue from Chlamydomonas to human (Figure 6A; Chan et al., 2005). We found that mutating Arg31 into Ala disrupted both the MT-binding and bundling activities of Spef1 in cells (Figure 6A and B). Consistent with the previous report (Dougherty et al., 2005), Spef1ΔCC, a mutant lacking its C-terminal CC region, bound to MTs but failed to bundle them (Figure 6A and B). Co-immunoprecipitation (co-IP) experiments showed that FLAG-Spef1 complexed with GFP-Spef1, but not GFP-ΔCC (Figure 6C), suggesting that the CC domain enables the MT bundling activity of Spef1 by mediating homodimerization. Therefore, the CH domain is responsible for the MT-binding activity of Spef1, while both the CH and CC domains are required for bundling MTs. Figure 6 View largeDownload slide Spef1 bundles MTs through homodimerization. (A) Diagrams of Spef1 and its mutants and summary of their properties based on the results in B and C. The conserved ‘R’ in the sequence alignments, which was mutated into ‘A’ in the mouse Spef1R31A mutant, is shown in red. (B) Spef1R31A failed to bind to MTs, whereas and Spef1ΔCC failed to bundle MTs when transiently expressed in U2OS cells. (C) Spef1 homodimerized through its CC region. The indicated proteins were transiently expressed in HEK293T cells. Co-IP was then performed using anti-FLAG resin. The GFP-containing proteins were visualized by using anti-GFP antibody. Figure 6 View largeDownload slide Spef1 bundles MTs through homodimerization. (A) Diagrams of Spef1 and its mutants and summary of their properties based on the results in B and C. The conserved ‘R’ in the sequence alignments, which was mutated into ‘A’ in the mouse Spef1R31A mutant, is shown in red. (B) Spef1R31A failed to bind to MTs, whereas and Spef1ΔCC failed to bundle MTs when transiently expressed in U2OS cells. (C) Spef1 homodimerized through its CC region. The indicated proteins were transiently expressed in HEK293T cells. Co-IP was then performed using anti-FLAG resin. The GFP-containing proteins were visualized by using anti-GFP antibody. Next we performed rescue experiments to examine functions of different mutants in motile cilia by ectopically expressing GFP-Spef1, GFP-Spef1R31A, and GFP-Spef1ΔCC in Spef-i2-treated mEPCs (Figure 7A; see Figure 3A for experimental scheme). Despite their similar expression levels (Figure 7A), confocal microscopy and quantifications indicated that the ciliary immunofluorescent signals of GFP-Spef1 were 2- and 8.4-fold over those of GFP-Spef1R31A and GFP-Spef1ΔCC, respectively (Figure 7B and C). Accordingly, the ciliary signals of endogenous Hydin and Spag6 showed similar tendencies, whereas those of Rsph4a were not altered (Figure 7B and C). Figure 7 View largeDownload slide Spef1R31A or Spef1ΔCC failed to rescue the CP formation in Spef1-depleted ependymal cilia. (A) Expression of RNAi-insensitive GFP-Spef1 and mutants in Spef1-depleted mEPCs (refer to Figure 3A for the experimental procedure). The samples were immunoblotted with anti-Spef1 antibody. Gapdh served as loading control. (B and C) Ciliary localizations of Spag6, Hydin, and Rsph4a in Spef-i2-treated mEPCs expressing GFP-Spef1 or its mutants. AC-tub labels ciliary axonemes. The quantification results are presented as box plots. At least 100 cilia were quantified in each condition. The fluorescent intensity of the indicated proteins was normalized to that of AC-tub. Student’s t-test, **P < 0.01; ***P < 0.001; n.s., not significant. (D and E) GFP-Spef1, but not Spef1R31A, restored the CP MTs (arrowheads). The statistical results were from three independent experiments. At least 101 cilia were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. (F and G) GFP-Spef1, but not its mutants, rescued the ciliary beat defect. Trajectories of four cilia in the first 56 ms of live imaging are shown for each EPC, together with a diagram depicting either the planar or rotational beat pattern (also refer to Supplementary Movie S3). The quantification results were from three independent experiments. At least 30 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. Figure 7 View largeDownload slide Spef1R31A or Spef1ΔCC failed to rescue the CP formation in Spef1-depleted ependymal cilia. (A) Expression of RNAi-insensitive GFP-Spef1 and mutants in Spef1-depleted mEPCs (refer to Figure 3A for the experimental procedure). The samples were immunoblotted with anti-Spef1 antibody. Gapdh served as loading control. (B and C) Ciliary localizations of Spag6, Hydin, and Rsph4a in Spef-i2-treated mEPCs expressing GFP-Spef1 or its mutants. AC-tub labels ciliary axonemes. The quantification results are presented as box plots. At least 100 cilia were quantified in each condition. The fluorescent intensity of the indicated proteins was normalized to that of AC-tub. Student’s t-test, **P < 0.01; ***P < 0.001; n.s., not significant. (D and E) GFP-Spef1, but not Spef1R31A, restored the CP MTs (arrowheads). The statistical results were from three independent experiments. At least 101 cilia were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. (F and G) GFP-Spef1, but not its mutants, rescued the ciliary beat defect. Trajectories of four cilia in the first 56 ms of live imaging are shown for each EPC, together with a diagram depicting either the planar or rotational beat pattern (also refer to Supplementary Movie S3). The quantification results were from three independent experiments. At least 30 cells were scored in each experiment and condition. Error bars represent SD. Student’s t-test, ***P < 0.001. As GFP-Spef1R31A was able to mildly both enter the cilia and rescue the ciliary signals of Hydin and Spag6, we investigated whether the GFP-Spef1R31A-expressing cilia contain CP MTs by using transmission EM. To ensure that the GFP-tagged proteins were expressed in the majority of the Spef1-depleted mEPCs, we infected the cells with the protein-expression lentivirus for three times and found that >86% of the cells readily became GFP-positive at SS d7 (Supplementary Figure S2). We found that 93.2% ± 4.4% of the cilia in the GFP-Spef1 samples contained two central MTs (Figure 7D and E). The remaining 6.8% had no central MT(s). By contrast, only 16.5% ± 7.3% of the cilia in the GFP-Spef1R31A samples contained 1 (2.1%) or 2 (14.4%) central MTs, whereas the remaining 83.5% had no central MT (Figure 7D and E), similar to the samples treated with Spef-i2 alone (Figure 4A). Thus, unlike Spef1, Spef1R31A is unable to rescue CP MTs. Finally, we examined ciliary motilities. We found that neither GFP-Spef1ΔCC nor GFP-Spef1R31A restored the ciliary beat pattern from rotational to planar as did GFP-Spef1 (Figure 7F and G; Supplementary Movie S3). Taken together, we conclude that both the MT-binding and bundling activities of Spef1 are required for its functions in the formation of CP MTs and CP apparatus. Discussion In this study, we demonstrated that Spef1 is a key regulator for the construction of the 9 + 2 axoneme. The incidence of CP-less cilia (83.5% ± 4.0%) in the Spef1 RNAi samples (Figure 4A) was similar to those of rotational ones (Figure 3D), suggesting that the remaining CP-containing cilia can be attributed to inefficient RNAi or failure in transfection (Figures 3B and 4B). Thus, we conclude that Spef1 is essential for the CP MT formation. Spef1 is structurally similar to EB proteins in that both contain an N-terminal MT-binding CH domain and a C-terminal CC domain for homodimerization and protein interactions (Figure 6) (Bu and Su, 2003; Dougherty et al., 2005; Sen et al., 2013). When purified Spef1 on beads was incubated with free tubulin, it did not exhibit MT-nucleation activities (data not shown). On the other hand, both EB proteins and Spef1 distribute evenly along MTs to bundle and stabilize MTs upon overexpression (Figures 5 and 6) (Bu and Su, 2003; Ligon et al., 2003; Dougherty et al., 2005). EB proteins also promote MT growth by binding between protofilaments at the plus end of MTs (Ligon et al., 2003; Maurer et al., 2012; Zhang et al., 2015). Endogenous Spef1 distributed rather evenly as puncta along the CP in mature cilia (e.g. those at SS d7) (Figure 2A) but preferably at ciliary tips in growing cilia at SS d3 (Figure 2D). The plus ends of the CP MTs are oriented toward the ciliary tip (Euteneuer and McIntosh, 1981). Therefore, the tip-specific enrichment of Spef1 suggests that Spef1 may exhibit plus-end binding in growing cilia. As both the MT-binding and bundling activities of Spef1 are required for CP MT formation (Figures 6 and 7), Spef1 may bundle the CP MTs nucleated by unknown factor(s) to keep the MTs as a pair, facilitate the CP MT elongation, and prevent them from disassembly during the growth of cilia. We found that Spef1 is also important for ciliary retentions of other CP proteins such as Spag6 and Hydin (Figures 4B, D and 7B, C). By contrast, non-CP ciliary proteins such as Rsph4a and IFT52 were not affected (Figure 4B and D). The MT binding-defective mutant Spef1R31A failed to rescue the CP MTs but partly rescued the ciliary localizations of Spag6 and Hydin as compared to the wild-type Spef1 and the non-effective Spef1ΔCC (Figure 7B and C). GFP-Spef1R31A itself also exhibited moderate ciliary localization as compared to GFP-Spef1 (positive control) and GFP-Spef1ΔCC (negative control) (Figure 7B and C). These results imply that Spef1 might facilitate ciliary entry or retention of Spag6 and Hydin in an MT binding-independent manner and the CP MTs further immobilize these proteins for the assembly of the CP apparatus. Apparently, future studies are required to clarify the detailed mechanisms. To our knowledge, Spef1 is so far the first documented CP protein essential for the CP MT formation. As both the CH and CC domains of Spef1 are highly conserved from protozoans to mammals (Chan et al., 2005), the CP-related functions of Spef1 are possibly evolutionarily conserved in protozoan. By searching for Spef1-associated proteins, additional regulators for the CP MT formation may be unraveled in both protozoan and mammals. In C. reinhardtii, three mutants (pf15, pf18, and pf19) have been found to completely lack the CP MTs (Adams et al., 1981). While the mutated gene of pf18 is still unknown, those of pf15 and pf19 encode PF15p and PF19p, which are orthologues of the regulatory and catalytic subunits of Katanin, an MT-severing enzyme. PF15p and PF19p, however, are not CP-localized and have been speculated to provide the initial templates or free tubulin for CP MTs (Dymek et al., 2004; Dymek and Smith, 2012; Lechtreck et al., 2013). It will be interesting to understand how these different pathways cooperate to achieve proper CP MT formation. Materials and methods Plasmids The full-length or partial cDNAs for Spef1 (NM_027641) were amplified by PCR from total cDNAs from mouse testis and constructed into pEGFP-C1, pcDNA3.1-NFLAG, pET28a, or pGEX4T-1 to express GFP-, FLAG-, His-, or GST-tagged fusion proteins. pLV-EGFP-C1 was used for lentiviral expression of GFP-fusion proteins. The cDNAs for the R31A mutant (228 CGG → GCG) and the RNAi-insensitive constructs of Spef1 (776 ACTTAAGAACGTGCGC → GCTCAAAAATGTACGT) were generated by PCR. The full-length cDNAs of mouse Rsph4a (NM_001162957), Spef1, and mouse cDNA fragments coding for amino acids 2312–2602 of Hydin (NM_172916) were constructed into pET28a to express antigens for antibody production. All the constructs were verified by sequencing. Cell culture, transfection, and viral infection Cells were maintained at 37°C in an atmosphere containing 5% CO2. Unless otherwise indicated, the culture medium was Dulbecco’s Modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (Biochrom), 0.3 mg/ml glutamine (Sigma), 100 U/ml penicillin (Invitrogen), and 100 U/ml streptomycin (Invitrogen). mTECs were isolated and cultured as described previously (You et al., 2002; Zhao et al., 2013). mEPCs were obtained as described (Delgehyr et al., 2015), with some modifications. The telencephala of P0 C57BL/6 J mice was dissected and digested with freshly prepared Papain solution (10 U/ml; Worthington Biochemical Corporation) for 30 min at 37°C. After gentle pipetting with a P1000 tip and centrifugation at 1400 rpm for 5 min at room temperature, the pelleted cells were resuspended in the culture medium and seeded into 25 cm2 laminin-coated flasks (1 brain/flask). After culturing for 1 day, neurons were shaken off and removed. The remaining radial glia-enriched cells were further cultured to ~90% confluency (usually 4 days) and then transferred into the wells of laminin-coated 29-mm glass-bottom dishes (Cellvis, D29-14-1.5-N) at a density of 2 × 105 cells per well. The cells were then maintained in serum-free medium to induce differentiation into EPCs. Generally, 30%–60% of the cells became multiciliated at SS d5 or later. To express exogenous proteins, HEK 293 T and U2OS cells were transfected using the calcium phosphate method and Lipofectamine 2000 (Life Technologies), respectively, for 48 h. Lentiviral production and infection were performed as described previously (Zhao et al., 2013). Cultured mouse radial glial cells were infected at SS d−1 unless otherwise stated. For RNAi, siRNA were transfected into radial glia/EPCs using Lipofectamine RNAiMAX (Life Technologies) at SS d−1 and d3, respectively. A total of 4 μl oligonucleotides (20 μM) and 6 μl Lipofectamine were used for a well of cells in a 29-mm glass-bottom dish. For rescue experiments, radial glia were infected with lentivirus at SS d−2 to express RNAi-insensitive GFP-Spef1 and its mutants or Centrin1-GFP, in addition to the transfections with siRNA. Sequences of siRNAs are listed as follows: Spef-i1: 5′-GAUGCACAAUUAUGUCCCUtt-3′; Spef-i2: 5′-CAACUUAAGAACGUGCGCAtt-3′; Ctrl-i: 5′-UUCUCCGAACGUGUCACGUtt-3′. Purification of ependymal cilia Cells from the telencephala of 10 P0 mice were seeded into two 75 cm2 laminin-coated flasks. After removing neurons, the cells were further cultured to near confluency (usually 5 days), followed by serum starvation to induce differentiation into mEPCs. The EPCs were harvested at SS d7 by centrifugation at 2000 rpm for 5 min at 4°C. The cells were resuspended in 3 ml of deciliation buffer (20 mM PIPES, pH 5.5, 250 mM sucrose, 20 mM CaCl2, 0.05% Triton X-100) and agitated vigorously for 10 min on a vortex mixer (G560OE, Scientific Industries). After centrifugation at 3000 rpm for 5 min at 4°C, the supernatant, which contained cilia, was collected. This step was repeated once. The cilia were then pelleted by centrifugation at 20000× g for 30 min, washed twice with 1 ml PBS, and resuspended in 100 μl of lysis buffer (20 mM Tris-HCl, pH 7.5, 100 mM KCl, 0.1% NP-40, 1 mM EDTA, 10 mM Na4O7P2, and protease inhibitors). The samples were used for mass spectrometric or immunoblotting analysis. Light microscopy mEPCs were pre-extracted with 0.5% Triton X-100 in PBS for 30 sec, followed by fixation with 4% paraformaldehyde in PBS for 10 min at room temperature and permeabilization with 0.5% Triton X-100 for 15 min. U2OS cells were fixed with −20°C methanol for 3 min and then 75% ethanol at 4°C for 10 min. Immunofluorescent staining was carried out as described (Zhao et al., 2013). GFP signals in mEPCs were visualized by immunostaining using anti-GFP antibody. Antibodies used are listed in Supplementary Table S1. Confocal images were captured by using Leica TCS SP8 system with a 63×/1.40 oil immersion objective and processed with maximum intensity projections. 3D-SIM super-resolution images were taken on DeltaVision OMX SR system (GE Healthcare) with a 100×/1.40 oil immersion objective (Olympus). Immersion oils with refractive indices of 1.516 were used to obtain optimal images and serial z-stack sectioning was set at 125 nm intervals. Images were reconstructed by softWoRx 5.0 software (GE Healthcare). Fluorescence intensity of ciliary proteins was measured using Adobe Photoshop Software. After the background subtraction, the mean fluorescent intensity of the interested proteins was normalized to the fluorescent intensity of Ac-tubulin. Ciliary motilities were recorded at 140 fps (frames per second) by using an Andor Neo sCMOS camera on Olympus IX71 microscope with a 63×/1.40 oil immersion objective (Xu et al., 2015). Cilia trajectories were illustrated with the manual tracking plugin in the Image J software. Four traceable cilia in each cell were tracked over a period of 56 ms. Electron microscopy For transmission EM, cultured EPCs were fixed in 2.5% glutaraldehyde overnight at 4°C, washed with PBS, and treated with 1% OsO4 for 30 min at room temperature. The samples were dehydrated with graded ethanol series and embedded in Epon 812 resin. The 70-nm ultrathin sections were stained with 1% lead citrate and 2% uranyl acetate. Images were captured at 80 kV using a Tecnai G2 Spirit transmission electron microscope (FEI). Immunoprecipitation Co-IP was performed as described previously (Zhao et al., 2013). Briefly, HEK293T cells transfected for 48 h were lysed with the lysis buffer containing 10% Glycerol. Pre-cleared cell lysates were incubated with anti-FLAG beads (Sigma, A2220) or anti-GFP beads (Allele Biotechnology) for 3 h at 4°C. After three times of wash with the lysis buffer and three times with wash buffer (20 mM Tris-HCl, pH 7.5, 150 mM KCl, 0.5% NP-40, 1 mM EDTA, 10 mM Na4P2O7, 10% glycerol), proteins on the anti-FLAG beads were eluted with 30 μl of 1 mg/ml FLAG peptide, while those on the GFP beads were directly denatured in 30 μl of SDS loading buffer. In vitro MT bundling MTs were polymerized in vitro as described (Jiang et al., 2015). Briefly, 2 μl of bovine tubulin stock (tubulin: rhodamine-labeled tubulin = 5:1) (Cytoskeleton, Inc.) were diluted in 2 μl of reaction buffer (80 mM PIPES, pH 6.9, 1 mM GTP, 2 mM MgCl2, 0.5 mM EGTA, 40% glycerol) to yield a final concentration of 20 μM tubulin. The mixture was incubated at 35°C for 15 min, diluted with 40 μl of pre-warmed (25°C) stabilization buffer (80 mM PIPES, pH 7.0, 20 μM Taxol, 2 mM MgCl2, 0.5 mM EGTA), and further incubated at 25°C for 30 min. To assay for MT bundling ability, 2 μl of the Taxol-stabilized MTs was mixed with 6 μl purified GST, GST-Spef1, or His-Spef1 of varying concentrations. After incubation at 37°C for 10 min, 3 μl of each mixture were gently squashed under an 18-mm circular coverslip. Microscopic examination and imaging were performed using Olympus BX51 microscope with a 100×/1.35 oil immersion objective and QImaging Retiga EXi CCD. Statistics Microscopic and biochemical results were repeated at least twice. Quantification results are presented as mean ± standard deviation (SD) unless otherwise stated. Two-tailed Student’s t-test (GraphPad Prism software) was used to calculate P-values between unpaired samples. Differences were considered significant when P < 0.05. Only results from three or more independent experiments were applied to the t-tests. In the box plots, the bottom and the top of the box represent the 25th and 75th percentiles, respectively. The band is the median. The ends of the whiskers indicate the maximum and minimum of the data. Supplementary material Supplementary material is available at Journal of Molecular Cell Biology online. Acknowledgements The authors thank Xingping Zhu and Tong Yin for technical support on 3D-SIM and Fengling Qin and Fangjie Qi for instrument support and technical assistance. Funding This work was equally supported by the National Natural Science Foundation of China (31330045), National Key R&D Program of China (2017YFA0503500), and Chinese Academy of Sciences (XDB19020000). It was also supported by the National Natural Science Foundation of China (31601092 to L.Z. and 31771495 to X.Y.). Conflict of interest: none declared. References Adams , G.M. , Huang , B. , Piperno , G. , et al. . ( 1981 ). Central-pair microtubular complex of Chlamydomonas flagella: polypeptide composition as revealed by analysis of mutants . J. Cell Biol. 91 , 69 – 76 . Google Scholar CrossRef Search ADS PubMed Brooks , E.R. , and Wallingford , J.B. ( 2014 ). Multiciliated cells . Curr. Biol. 24 , R973 – R982 . Google Scholar CrossRef Search ADS PubMed Bu , W. , and Su , L.K. ( 2003 ). Characterization of functional domains of human EB1 family proteins . J. Biol. Chem. 278 , 49721 – 49731 . Google Scholar CrossRef Search ADS PubMed Carbajal-Gonzalez , B.I. , Heuser , T. , Fu , X. , et al. . ( 2013 ). Conserved structural motifs in the central pair complex of eukaryotic flagella . Cytoskeleton 70 , 101 – 120 . Google Scholar CrossRef Search ADS PubMed Chan , S.W. , Fowler , K.J. , Choo , K.H. , et al. . ( 2005 ). Spef1, a conserved novel testis protein found in mouse sperm flagella . Gene 353 , 189 – 199 . Google Scholar CrossRef Search ADS PubMed Delgehyr , N. , Meunier , A. , Faucourt , M. , et al. . ( 2015 ). Ependymal cell differentiation, from monociliated to multiciliated cells . Methods Cell Biol. 127 , 19 – 35 . Google Scholar CrossRef Search ADS PubMed Dougherty , G.W. , Adler , H.J. , Rzadzinska , A. , et al. . ( 2005 ). CLAMP, a novel microtubule-associated protein with EB-type calponin homology . Cell Motil. Cytoskeleton 62 , 141 – 156 . Google Scholar CrossRef Search ADS PubMed Dutcher , S.K. , Huang , B. , and Luck , D.J. ( 1984 ). Genetic dissection of the central pair microtubules of the flagella of Chlamydomonas reinhardtii . J. Cell Biol. 98 , 229 – 236 . Google Scholar CrossRef Search ADS PubMed Dymek , E.E. , Lefebvre , P.A. , and Smith , E.F. ( 2004 ). PF15p is the chlamydomonas homologue of the Katanin p80 subunit and is required for assembly of flagellar central microtubules . Eukaryot. Cell 3 , 870 – 879 . Google Scholar CrossRef Search ADS PubMed Dymek , E.E. , and Smith , E.F. ( 2012 ). PF19 encodes the p60 catalytic subunit of katanin and is required for assembly of the flagellar central apparatus in Chlamydomonas . J. Cell Sci. 125 , 3357 – 3366 . Google Scholar CrossRef Search ADS PubMed Euteneuer , U. , and McIntosh , J.R. ( 1981 ). Polarity of some motility-related microtubules . Proc. Natl Acad. Sci. USA 78 , 372 – 376 . Google Scholar CrossRef Search ADS Fliegauf , M. , Benzing , T. , and Omran , H. ( 2007 ). When cilia go bad: cilia defects and ciliopathies . Nat. Rev. Mol. Cell Biol. 8 , 880 – 893 . Google Scholar CrossRef Search ADS PubMed Gerdes , J.M. , Davis , E.E. , and Katsanis , N. ( 2009 ). The vertebrate primary cilium in development, homeostasis, and disease . Cell 137 , 32 – 45 . Google Scholar CrossRef Search ADS PubMed Gray , R.S. , Abitua , P.B. , Wlodarczyk , B.J. , et al. . ( 2009 ). The planar cell polarity effector Fuz is essential for targeted membrane trafficking, ciliogenesis and mouse embryonic development . Nat. Cell Biol. 11 , 1225 – 1232 . Google Scholar CrossRef Search ADS PubMed Janke , C. , and Bulinski , J.C. ( 2011 ). Post-translational regulation of the microtubule cytoskeleton: mechanisms and functions . Nat. Rev. Mol. Cell Biol. 12 , 773 – 786 . Google Scholar CrossRef Search ADS PubMed Jiang , H. , Wang , S. , Huang , Y. , et al. . ( 2015 ). Phase transition of spindle-associated protein regulate spindle apparatus assembly . Cell 163 , 108 – 122 . Google Scholar CrossRef Search ADS PubMed Lechtreck , K.F. , Delmotte , P. , Robinson , M.L. , et al. . ( 2008 ). Mutations in Hydin impair ciliary motility in mice . J. Cell Biol. 180 , 633 – 643 . Google Scholar CrossRef Search ADS PubMed Lechtreck , K.F. , Gould , T.J. , and Witman , G.B. ( 2013 ). Flagellar central pair assembly in Chlamydomonas reinhardtii . Cilia 2 , 15 . Google Scholar CrossRef Search ADS PubMed Lechtreck , K.F. , and Witman , G.B. ( 2007 ). Chlamydomonas reinhardtii hydin is a central pair protein required for flagellar motility . J. Cell Biol. 176 , 473 – 482 . Google Scholar CrossRef Search ADS PubMed Ligon , L.A. , Shelly , S.S. , Tokito , M. , et al. . ( 2003 ). The microtubule plus-end proteins EB1 and dynactin have differential effects on microtubule polymerization . Mol. Biol. Cell 14 , 1405 – 1417 . Google Scholar CrossRef Search ADS PubMed Maurer , S.P. , Fourniol , F.J. , Bohner , G. , et al. . ( 2012 ). EBs recognize a nucleotide-dependent structural cap at growing microtubule ends . Cell 149 , 371 – 382 . Google Scholar CrossRef Search ADS PubMed McKean , P.G. , Baines , A. , Vaughan , S. , et al. . ( 2003 ). Gamma-tubulin functions in the nucleation of a discrete subset of microtubules in the eukaryotic flagellum . Curr. Biol. 13 , 598 – 602 . Google Scholar CrossRef Search ADS PubMed Oda , T. , Yanagisawa , H. , Yagi , T. , et al. . ( 2014 ). Mechanosignaling between central apparatus and radial spokes controls axonemal dynein activity . J. Cell Biol. 204 , 807 – 819 . Google Scholar CrossRef Search ADS PubMed Pigino , G. , Bui , K.H. , Maheshwari , A. , et al. . ( 2011 ). Cryoelectron tomography of radial spokes in cilia and flagella . J. Cell Biol. 195 , 673 – 687 . Google Scholar CrossRef Search ADS PubMed Portran , D. , Schaedel , L. , Xu , Z.J. , et al. . ( 2017 ). Tubulin acetylation protects long-lived microtubules against mechanical ageing . Nat. Cell Biol. 19 , 391 – 398 . Google Scholar CrossRef Search ADS PubMed Praveen , K. , Davis , E.E. , and Katsanis , N. ( 2015 ). Unique among ciliopathies: primary ciliary dyskinesia, a motile cilia disorder . F1000Prime Rep. 7 , 36 . Google Scholar CrossRef Search ADS PubMed Sapiro , R. , Kostetskii , I. , Olds-Clarke , P. , et al. . ( 2002 ). Male infertility, impaired sperm motility, and hydrocephalus in mice deficient in sperm-associated antigen 6 . Mol. Cell. Biol. 22 , 6298 – 6305 . Google Scholar CrossRef Search ADS PubMed Satir , P. , Heuser , T. , and Sale , W.S. ( 2014 ). A structural basis for how motile cilia beat . Bioscience 64 , 1073 – 1083 . Google Scholar CrossRef Search ADS PubMed Sen , I. , Veprintsev , D. , Akhmanova , A. , et al. . ( 2013 ). End binding proteins are obligatory dimers . PLoS One 8 , e74448 . Google Scholar CrossRef Search ADS PubMed Silflow , C.D. , Liu , B. , LaVoie , M. , et al. . ( 1999 ). Gamma-tubulin in Chlamydomonas: characterization of the gene and localization of the gene product in cells . Cell Motil. Cytoskeleton 42 , 285 – 297 . Google Scholar CrossRef Search ADS PubMed Smith , E.F. , and Lefebvre , P.A. ( 1996 ). PF16 encodes a protein with armadillo repeats and localizes to a single microtubule of the central apparatus in Chlamydomonas flagella . J. Cell Biol. 132 , 359 – 370 . Google Scholar CrossRef Search ADS PubMed Smith , E.F. , and Lefebvre , P.A. ( 1997 a). PF20 gene product contains WD repeats and localizes to the intermicrotubule bridges in Chlamydomonas flagella . Mol. Biol. Cell 8 , 455 – 467 . Google Scholar CrossRef Search ADS PubMed Smith , E.F. , and Lefebvre , P.A. ( 1997 b). The role of central apparatus components in flagellar motility and microtubule assembly . Cell Motil. Cytoskeleton 38 , 1 – 8 . Google Scholar CrossRef Search ADS PubMed Smith , E.F. , and Yang , P. ( 2004 ). The radial spokes and central apparatus: mechano-chemical transducers that regulate flagellar motility . Cell Motil. Cytoskeleton 57 , 8 – 17 . Google Scholar CrossRef Search ADS PubMed Spassky , N. , Merkle , F.T. , Flames , N. , et al. . ( 2005 ). Adult ependymal cells are postmitotic and are derived from radial glial cells during embryogenesis . J. Neurosci. 25 , 10 – 18 . Google Scholar CrossRef Search ADS PubMed Tang , T.K. ( 2013 ). Centriole biogenesis in multiciliated cells . Nat. Cell Biol. 15 , 1400 – 1402 . Google Scholar CrossRef Search ADS PubMed Teves , M.E. , Nagarkatti-Gude , D.R. , Zhang , Z. , et al. . ( 2016 ). Mammalian axoneme central pair complex proteins: broader roles revealed by gene knockout phenotypes . Cytoskeleton 73 , 3 – 22 . Google Scholar CrossRef Search ADS PubMed Werner , M.E. , Mitchell , J.W. , Putzbach , W. , et al. . ( 2014 ). Radial intercalation is regulated by the Par complex and the microtubule-stabilizing protein CLAMP/Spef1 . J. Cell Biol. 206 , 367 – 376 . Google Scholar CrossRef Search ADS PubMed Witman , G.B. , Plummer , J. , and Sander , G. ( 1978 ). Chlamydomonas flagellar mutants lacking radial spokes and central tubules. Structure, composition, and function of specific axonemal components . J Cell Biol . 76 , 729 – 747 . Google Scholar CrossRef Search ADS PubMed Xu , Y. , Cao , J. , Huang , S. , et al. . ( 2015 ). Characterization of tetratricopeptide repeat-containing proteins critical for cilia formation and function . PLoS One 10 , e0124378 . Google Scholar CrossRef Search ADS PubMed Yan , X. , Zhao , H. , and Zhu , X. ( 2016 ). Production of basal bodies in bulk for dense multicilia formation . F1000Res. 5 , pii: F1000 Faculty Rev-1533. You , Y. , Richer , E.J. , Huang , T. , et al. . ( 2002 ). Growth and differentiation of mouse tracheal epithelial cells: selection of a proliferative population . Am. J. Physiol. Lung Cell. Mol. Physiol. 283 , L1315 – L1321 . Google Scholar CrossRef Search ADS PubMed Zhang , R. , Alushin , G.M. , Brown , A. , et al. . ( 2015 ). Mechanistic origin of microtubule dynamic instability and its modulation by EB proteins . Cell 162 , 849 – 859 . Google Scholar CrossRef Search ADS PubMed Zhang , Z. , Kostetskii , I. , Tang , W. , et al. . ( 2006 ). Deficiency of SPAG16L causes male infertility associated with impaired sperm motility . Biol. Reprod. 74 , 751 – 759 . Google Scholar CrossRef Search ADS PubMed Zhang , Z. , Tang , W. , Zhou , R. , et al. . ( 2007 ). Accelerated mortality from hydrocephalus and pneumonia in mice with a combined deficiency of SPAG6 and SPAG16L reveals a functional interrelationship between the two central apparatus proteins . Cell Motil. Cytoskeleton 64 , 360 – 376 . Google Scholar CrossRef Search ADS PubMed Zhao , H. , Zhu , L. , Zhu , Y. , et al. . ( 2013 ). The Cep63 paralogue Deup1 enables massive de novo centriole biogenesis for vertebrate multiciliogenesis . Nat. Cell Biol. 15 , 1434 – 1444 . Google Scholar CrossRef Search ADS PubMed © The Author(s) (2018). Published by Oxford University Press on behalf of Journal of Molecular Cell Biology, IBCB, SIBS, CAS. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Molecular Cell BiologyOxford University Press

Published: Feb 23, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off