LZTS2 Negatively Regulates Centrosomal CEP135 Levels and Microtubule Nucleation
Catarina Peneda, Joana N. Bugalhao, Marco Antonio Dias Louro, Andreia Henriques‐Soares, Monica Bettencourt‐Dias
TL;DR
This study shows that LZTS2, a tumor suppressor, controls microtubule formation at the centrosome by reducing levels of CEP135, a key microtubule nucleation factor.
Contribution
The paper identifies a new role for LZTS2 as a negative regulator of centrosomal microtubule nucleation through CEP135.
Findings
LZTS2 depletion increases microtubule nucleation at the centrosome.
LZTS2 negatively regulates centrosomal levels of CEP135.
LZTS2 depletion can partially rescue impaired microtubule nucleation caused by CEP135 knockdown.
Abstract
The microtubule cytoskeleton is a fundamental functional component of the cell. In vertebrate proliferating cells, centrosomes are the primary microtubule organizing center (MTOC), and their dysregulation has been linked to genomic instability and cancer. LZTS2, a known tumor suppressor, localizes to centrosomes and regulates microtubule severing. However, whether LZTS2 regulates centrosome structure and/or its function in microtubule organization or ciliation remains unknown. Here, we investigate the function of LZTS2 at the centrosome. Through fluorescence and electron microscopy assays, we observed that LZTS2 knockdown does not affect centriole biogenesis or structure, nor ciliation. Importantly, we show that LZTS2 depletion increases microtubule nucleation at the centrosome. Moreover, LZTS2 negatively regulates centrosomal levels of CEP135. Notably, depletion of LZTS2 can partially…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
FIGURE 1
FIGURE 2
FIGURE 3
FIGURE 4
FIGURE 5- —European Research Council10.13039/501100000781
- —Fundação para a Ciência e a Tecnologia10.13039/501100001871
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMicrotubule and mitosis dynamics · Protist diversity and phylogeny · Epigenetics and DNA Methylation
Introduction
1
The microtubule cytoskeleton is a dynamic network that plays crucial roles in several cellular processes such as migration, cell polarity, intracellular transport, and cell division (Bodakuntla et al. 2019; Janke and Magiera 2020; Lara Ordóñez et al. 2022).
The centrosome is the main microtubule organizing center (MTOC) in proliferating animal cells. The canonical human centrosome consists of two centrioles, which are barrel‐shaped microtubule structures, surrounded by a complex and highly organized matrix of proteins called the pericentriolar material (PCM) (Vasquez‐Limeta and Loncarek 2021). Centrioles can convert to basal bodies, docking to the cell membrane and supporting the assembly of cilia (Anderson 1972; Tanos et al. 2013). As such, they play major roles in sensing, cell motility, and signaling (Amack 2022; Hilgendorf et al. 2024). Numerical and structural aberrations in centrosomes and cilia have been associated with several human diseases, such as microcephaly, retinopathies, and cancer (Barbelanne and Tsang 2014; Delaval and Doxsey 2010; Marteil et al. 2018; Purkerson et al. 2024). Proper regulation of centrosome microtubule nucleation is important during cell division, where it ensures accurate chromosome segregation. Dysregulation of this process is frequently observed in cancer cells, contributing to genomic instability (Maiato and Logarinho 2014; Godinho et al. 2014; Milunović‐Jevtić et al. 2016). Moreover, dysregulation of microtubule dynamics during interphase, in particular, through increased levels of nucleators or centriole amplification, leads to higher capacity to invade, metastasize, and change the tumor microenvironment (Molina et al. 2013; Byrne et al. 2014; Adams et al. 2021; Prakash et al. 2023; Mangaonkar et al. 2024; Sun et al. 2024). While several proteins are known to affect microtubule nucleation, such as the key regulators γ‐tubulin and CDK5RAP2 (Dictenberg et al. 1998; Dammermann and Merdes 2002; Takahashi et al. 2002; Casenghi et al. 2003; Choi et al. 2010), other centrosomal factors, including CEP135, might influence microtubule organization beyond their role in centriole integrity and ability to bind microtubules (Uetake et al. 2004; Carvalho‐Santos et al. 2012; Chu and Gruss 2022). How all these factors are normally controlled and might be deregulated in the context of cancer is not clear.
Leucine‐Zipper putative Tumor Suppressor 2 (LZTS2) is altered in several cancers, including hepatocellular carcinoma, clear cell renal cell carcinoma, and colon cancer (Johnson et al. 2013; Xu et al. 2025; Tang et al. 2024; Dong et al. 2023; Peng et al. 2022). Its downregulation is also associated with increased malignancy in different cancers (Yu et al. 2017; Xu et al. 2018; Lu et al. 2021). While LZTS2 has a function in signaling by negatively regulating β‐catenin‐mediated transcription and WNT signaling (Thyssen et al. 2006; Li et al. 2011; Dong et al. 2023; Liu et al. 2024), it is also likely to be an important regulator of the microtubule cytoskeleton. It localizes to the centrosome, basal bodies of cilia, and midbody, and has been shown to interact with key factors associated with microtubule dynamics, such as γ‐tubulin in vitro (Sudo and Maru 2007), and to inhibit the p80 subunit of the katanin complex, hindering the translocation of microtubules from the centrosome to the midbody during cytokinesis (Sudo and Maru 2008). However, whether LZTS2 affects centrosome biogenesis, structure, and other aspects of its function is poorly characterized.
Here, we investigated the unexplored role of LZTS2 in centrosome biogenesis, structure, and function. Using a combination of fluorescence imaging and electron microscopy techniques, we identified a novel function of LZTS2 in microtubule nucleation and uncovered an interaction with CEP135, a known regulator of microtubule nucleation through yet uncharacterized mechanisms (Uetake et al. 2004; Chu and Gruss 2022).
Results and Discussion
2
LZTS2 Depletion Does Not Affect the Number or Structure of Centrioles or Ciliation Capacity
2.1
Given LZTS2 localization to centrosomes and its role as a tumor suppressor, we first investigated whether it affects fundamental aspects of centrosome structure and function. First, we asked if LZTS2 has a role in cilia assembly. To address this, we performed a ciliation assay using the hTERT RPE‐1 human cell line. This cell line is known to assemble primary cilia upon cell cycle exit and thus provide a suitable model for addressing this question. In short, cells were transfected with LZTS2‐specific or control siRNAs and after 24 h the media was replaced by serum‐free media. Cells were serum‐starved for 24 h, after which they were fixed. LZTS2 knockdown efficiency was validated by qRT‐PCR (Figure S1A,B). Cilia were identified by polyglutamylated tubulin and CEP135 staining and imaged using spinning disk microscopy (Figure 1A). We then scored the percentage of cells with a detectable cilium (Figure 1B). Upon LZTS2 knockdown, 72.49% of the cells were ciliated compared with 78.1% of control cells. Thus, we concluded that LZTS2 depletion did not significantly impair cilia formation in hTERT‐RPE‐1 cells.
LZTS2 knockdown does not affect ciliation efficiency nor centriole number. (A, B) hTERT RPE‐1 cells were transfected with control or LZTS2 siRNA. After 24 h, cells were serum starved (to induce ciliation) for another 24 h and analyzed by immunofluorescence. (A) Cilia were identified by anti‐polyglutamylated tubulin (GT335) in green and anti‐CEP135 in magenta. DNA was stained with DAPI, in blue. (B) The percentage of ciliated cells was quantified. Note that no significant differences were observed (two tailed Student's t test performed on the log‐odds ratio of cells with visible cilia). (C, D) U2OS cells were transfected with control or LZTS2 siRNA for 48 h, and centriole number was analyzed by immunofluorescence. (C) Bona fide centrioles were identified by co‐localization of centrin (green) and CEP135 (magenta). DNA was stained with DAPI (blue). (D) Centriole number per cell was quantified and the percentage of cells with fewer centrioles (< 2 centrioles), wild type‐like (2–4), or extra centrioles (> 4) are represented in black, dark green or light green, respectively. No significant differences were observed between the control and LZTS2 knockdown conditions (Fisher's exact test, Bonferroni correction). Plot bars correspond to mean ± SD. At least 100 cells were analyzed in three independent experiments. Scale bar corresponds to 10 μm.
Second, we tested if LZTS2 plays a role in centriole biogenesis. We resorted to human osteosarcoma U2OS cells, a cell line often used in centriole structural studies, having been used to identify many players in centriole number and structure regulation (Kleylein‐Sohn et al. 2007; Tang et al. 2009; Iyer et al. 2025). U2OS cells were transfected with control or specific siRNAs to knock down LZTS2 and fixed after 48 h. LZTS2 knockdown in U2OS was validated by qRT‐PCR and western blotting (Figure S1C,D). Centrioles were immunolabeled using antibodies against CEP135 and Centrin, and interphase cells were imaged using spinning disk microscopy for centriole number quantification (Figure 1C,D).
We observed no significant differences in centriole numbers between cells treated with control siRNA and cells treated with LZTS2 siRNA (Figure 1C,D). In addition, the percentage of cells with amplified (more than 4) or abnormally low (less than 2) centrioles was not significantly affected.
Several proteins recruited to centrioles have been reported to participate in centriole elongation or in ensuring its structural integrity. For instance, CPAP and CP110 are known to regulate centriole length in an antagonistic fashion (Schmidt et al. 2009). Therefore, we asked if LZTS2 could also have an impact on centriolar architecture. We knocked down LZTS2 for 48 h in U2OS cells and assessed centriole structure by measuring its diameter and length by electron microscopy. Our results showed that both centriole length (Figure 2A,C) and diameter (Figure 2B,D) were conserved in cells depleted of LZTS2 when compared with the control. Centriolar subdistal and distal appendages were observed in both conditions (Figure 2E), although we could not analyze their structure in detail with this technique.
Centriole structure was not affected by LZTS2 knockdown in U2OS cells. U2OS cells were transfected with control or LZTS2 siRNA. Forty‐eight hours after transfection cells were fixed and processed for electron microscopy (A–E) or immunofluorescence (F). Longitudinal (A) and transversal (B) sections were analyzed and centriole length (C) and diameter (D) were measured. Scale bars correspond to 200 nm. No significant differences were detected (two‐tailed Student's t test). (E) Distal and subdistal appendages were observed in longitudinal and transversal sections in both control and LZTS2 depletion conditions. At least 25 cells were analyzed from three independent experiments. Scale bars correspond to 200 nm. (F) Distal (CEP164, in magenta) and subdistal (Ninein, in yellow) appendages components were observed in their expected localization in both control and LZTS2‐depleted conditions, with CEP164 localizing at the distal region of the mother centriole and Ninein localizing both at the proximal and distal regions of the mother centriole and in the proximal region of the daughter centriole, as depicted in the scheme (right). CEP135 (in cyan) was used to label centrioles and to serve as a reference for their proximal region. Scale bar corresponds to 500 nm.
To further investigate the presence of appendages, we stained cells for a distal appendage protein, CEP164, and for a subdistal appendage protein, Ninein, and analyzed them by super‐resolution microscopy (three‐dimensional structured illumination microscopy—3D‐SIM). We observed the same localization pattern for CEP164 and for Ninein in both control and LZTS2‐depleted cells (Figure 2F). Thus, while we cannot rule out any effects of LZTS2 in the shape and number of centriolar appendages, LZTS2‐depleted cells display wild‐type‐like centriole structure. In summary, we found no evidence for a role of LZTS2 in centriole and cilia biogenesis.
LZTS2 Depletion Increases Centrosomal Microtubule Nucleation
2.2
Changes in centrosomal microtubule nucleation capacity can profoundly affect cellular function, leading to chromosome segregation defects and potentially leading to altered signaling pathways, such as altered Rac1 activity (Maiato and Logarinho 2014; Godinho et al. 2014; Paim et al. 2024; Bertalan et al. 2015; Rincón and Monje‐Casas 2020; Purkerson et al. 2024). Given LZTS2 localization to centrosomes (Sudo and Maru 2008) we hypothesized that this tumor suppressor might regulate centrosomal microtubule organization. This could be particularly relevant in cancer, where LZTS2 is frequently downregulated (Yu et al. 2017; Xu et al. 2018; Lu et al. 2021).
To specifically examine LZTS2's role in microtubule nucleation, we performed microtubule regrowth assays in cells transfected with control or LZTS2 siRNAs. We used very short recovery periods (1 min), allowing us to distinguish nucleation effects from potential changes in microtubule severing or anchoring, as reported before for LZTS2, through inhibition of the p80 subunit of the katanin complex (Sudo and Maru 2008). Knockdown efficiency was validated by western blotting (Figure S2). Strikingly, LZTS2 depletion led to increased microtubule nucleation compared to control cells (Figure 3A,B). To understand the mechanism behind this increased nucleation, we examined potential interactions with known regulators of this process. We focused on CEP135, which was shown to play a role in the regulation of the PCM and microtubule nucleation from the centrosome in interphase cells (Uetake et al. 2004; Chu and Gruss 2022). While CEP135 knockdown decreased microtubule nucleation as previously reported (Uetake et al. 2004), co‐depletion of LZTS2 and CEP135 partially rescued this defect (Figure 3A,B). This genetic interaction suggests that LZTS2 might negatively regulate CEP135‐mediated microtubule nucleation.
*LZTS2 knockdown partially rescues the decrease in microtubule nucleation induced by CEP135 knockdown and CEP135 levels at the centrosome without affecting centriole number. Microtubule regrowth assays were performed in U2OS cells 48 h after transfection with control, LZTS2, CEP135 or LZTS2 and CEP135 combined siRNAs. (A) Microtubule asters were immunolabeled with ɑ‐tubulin (in gray), and Pericentrin (in magenta) was used to label centrosomes. The DNA was stained with DAPI (in blue). Scale bar corresponds to 10 μm. (B) ɑ‐tubulin intensity per cell was quantified (see Section 4) and the mean of each independent experiment is represented (normalized by the control). At least 50 cells were analyzed in six independent experiments. Significant differences are represented (****p‐value < 0.0001; **p‐values < 0.001, see Section 4 for statistical analysis). (C) Centriole number per cell was quantified and the percentage of cells with lower number of centrioles (< 2), wild type‐like (2–4) or with extra centrioles (> 4) is represented. Note that we could not detect significant differences between the experimental conditions (Fisher's exact test, Bonferroni's correction). One hundred cells were quantified in three independent experiments. Plot bars correspond to mean ± SD.
CEP135 is a core structural protein of the centriole. It bridges the interaction between centriolar microtubule triplets and a scaffolding structure called the cartwheel (Lin et al. 2013). Previous studies showed that CEP135 depletion impairs centriole biogenesis induced by Plk4 overexpression (Lin et al. 2013). This raises the question of whether the microtubule nucleation defects observed upon CEP135 knockdown are due to faulty centriole duplication, which could then be rescued by LZTS2 knockdown. As described above, we quantified the centriole number in samples fixed 48 h after transfection with the control or specific siRNAs for LZTS2, CEP135, or the combination of both LZTS2 and CEP135 (Figure 3C). Coherently with our previous results, LZTS2 knockdown did not significantly affect centriole numbers compared with the control cells, and neither did CEP135 knockdown. Codepletion of LZTS2 and CEP135 also did not yield centriole number alterations (Figure 3C). These results indicate that the rescue of microtubule nucleation defects imparted by CEP135 knockdown following LZTS2 depletion cannot be explained by defects in centriole number.
LZTS2 Negatively Regulates CEP135 Levels at the Centrosome
2.3
Given the absence of overt centriole number alterations upon knockdown of LZTS2 and/or CEP135, we asked if LZTS2 could regulate CEP135. We quantified the levels of centrosomal CEP135 in fixed U2OS cells following treatment with control, LZTS2, and/or CEP135‐specific siRNAs for 48 h (Figure 4A,B). As expected, depletion of CEP135 significantly reduced the centrosomal levels of the protein (Figure 4B). Interestingly, we observed that CEP135 levels at the centrosome were significantly increased upon LZTS2 knockdown. Moreover, we observed a partial rescue of CEP135 centrosomal levels upon co‐depletion of LZTS2 and CEP135 (Figure 4B). These results suggest that LZTS2 negatively regulates the levels of CEP135 at the centrosome, although its knockdown is insufficient to completely rescue CEP135 depletion. Thus, the effect of LZTS2 in controlling microtubule nucleation from the centrosome could be explained in part by its role in inhibiting CEP135 centrosomal recruitment or stability.
*LZTS2 knockdown partially rescues CEP135 levels at the centrosome and CEP135 co‐immunoprecipitates with LZTS2. (A, B) U2OS cells transfected with control, LZTS2, CEP135, or LZTS2 and CEP135 combined siRNAs. (A) After fixation, samples were immunolabeled for CEP135 (green) and Ninein (magenta). DNA was stained with DAPI and the scale bar corresponds to 10 μm. (B) CEP135 intensity at the centrosome was measured and the mean of each independent experiment is represented. At least 50 cells were analyzed in each of the four independent experiments. Plot bars correspond to mean ± SD. Significant differences are represented (****p‐value < 0.0001; **p‐values < 0.001, see Section 4 for statistical analysis). (C) U2OS cell extracts were incubated with Dynabeads coupled with anti‐LZTS2 antibody or IgG (as control). After washing, the bound proteins were eluted, separated by SDS‐PAGE and analyzed by western blotting. LZTS2 was successfully immunoprecipitated in the LZTS2 IP and CEP135 co‐immunoprecipitated with LZTS2, while it was not detected in the IgG control. At least three independent experiments were performed.
These results led us to investigate if LZTS2 and CEP135 could interact. To address this, we performed co‐immunoprecipitation (co‐IP) (Figure 4C) assays using U2OS cell extracts to immunoprecipitate LZTS2. We observed that CEP135 coimmunoprecipitates with LZTS2 and was not detected in the IgG control (Figure 4C). Whereas we cannot rule out that this interaction may be indirect, it is possible that these proteins are part of a complex or pathway that regulates centrosomal microtubule nucleation.
LZTS2 has been shown to interact with γ‐tubulin in mitotic cells (Sudo and Maru 2008). Due to the central role of this molecule in microtubule nucleation, we asked whether the observed LZTS2 microtubule nucleation effects could also be due to alterations in γ‐tubulin. In order to address this question, we repeated the knockdown experiments in U2OS cells and measured the centrosomal levels of γ‐tubulin. LZTS2 knockdown did not affect the centrosomal levels of γ‐tubulin (Figure 5A,D). Conversely, both CEP135 knockdown and co‐depletion of LZTS2 and CEP135 resulted in a comparable reduction of centrosomal γ‐tubulin. Thus, increased microtubule nucleation from interphase centrosomes upon LZTS2 knockdown cannot be explained by changes in γ‐tubulin centrosomal levels. Other PCM components are known to be important for the regulation of microtubule dynamics, such as CDK5RAP2 and Pericentrin (Choi et al. 2010; Serna et al. 2024; Dictenberg et al. 1998), whose centrosomal levels were previously shown to be dependent on CEP135 (Uetake et al. 2004; Chu and Gruss 2022). Thuhypothesizedesised that the differences in microtubule nucleation we observed could be associated with changes in those proteins. To address this, we quantified the centrosomal levels of CDK5RAP2 and Pericentrin in U2OS cells depleted of LZTS2, ,CEP135, or the combination of LZTS2 and CEP135. We analyzed the centrosomal levels of Centriolin, a centriole‐appendage protein, to include a centrosomal, but non‐PCM protein, in our analysis. We did not detect significant changes in Centriolin centrosomal levels in any of the experimental conditions (Figure 5G,H). In contrast, CEP135, but not LZTS2 depletion, led to a decrease in centrosomal CDK5RAP2 and Pericentrin (Figure 5B,C,E,F). Interestingly, LZTS2 knockdown partially rescued the effect of CEP135 knockdown on these proteins (Figure 5E,F). Thus, our results support the hypothesis that LZTS2 regulates centrosomal microtubule nucleation partly by regulating CEP135 levels at the centrosome, which in turn affects PCM accumulation.
*Centrosomal levels of γ‐tubulin, CDK5RAP2, Pericentrinm, and Centriolin upon LZTS2, CEP135, or double knockdown. U2OS cells were transfected with control, LZTS2, CEP135, or LZTS2 and CEP135 combined siRNAs. After fixation, samples were immunolabeled for (A) γ‐tubulin (green) and CEP135 (magenta), (B) γ‐tubulin (green) and CDK5RAP2 (magenta), or (C) Centriolin (green) and Pericentrin (magenta). DNA was stained with DAPI and the scale bars correspond to 10 μm. (D–G) Intensity of (D) γ‐tubulin, (E) CDK5RAP2, (F) Pericentrin, and (G) Centriolin at the centrosome was measured and the mean of each independent experiment is represented. At least 80 cells were analyzed in each of the three independent experiments. Plot bars correspond to mean ± SD. Significant differences are represented (****p‐value < 0.0001; ***p‐values < 0.001; **p‐values < 0.01; p‐values < 0.05, see Section 4 for statistical analysis).
Conclusion
3
In this study, we uncovered a novel mechanism by which the tumor suppressor LZTS2 regulates centrosome function. Whereas LZTS2 does not seem to affect overall centrosome structure or cilia formation, it acts as a negative regulator of CEP135 and centrosomal microtubule nucleation. This regulation reveals a previously unknown layer of control of microtubule nucleation that may be relevant for LZTS2 tumor suppressor function. Our findings show that LZTS2 knockdown leads to increased microtubule nucleation, which has been associated with chromosomal instability and increased invasion (Ertych et al. 2014; Godinho et al. 2014).
Our data show that CEP135‐mediated microtubule nucleation is negatively regulated by LZTS2. It has been described that LZTS2 interacts with γ‐tubulin in vitro, a central player in centrosomal microtubule nucleation (Sudo and Maru 2007; Zheng et al. 1995; Sulimenko et al. 2022). Whereas the consequences of this interaction remain uncharacterized in cells, we were not able to detect any significant difference in γ‐tubulin centrosomal levels upon LZTS2 knockdown. Thus, our results do not evidently depend on γ‐tubulin levels. On the other hand, CEP135 promotes the centrosomal recruitment of PCM proteins, possibly via dynactin (Uetake et al. 2004; Chu and Gruss 2022), which could be related to its role in microtubule nucleation. Future studies should determine if the PCM changes observed explain the differences in microtubule nucleation registeed, and explore what other factors are involved in LZTS2 function at the centrosome.
Whereas LZTS2 was identified as a tumor suppressor gene (Kim et al. 2011; Johnson et al. 2013; Xu et al. 2018), its role in tumorigenesis is still elusive. One possibility is that LZTS2 has an anti‐proliferative role that is suppressed upon downregulation or loss‐of‐heterozygosity (Lu et al. 2021; Yu et al. 2017). Another possibility, in line with our data and that of others, is that the loss of LZTS2 function can impact microtubule nucleation, which could potentially lead to increased migration, invasiveness, or to chromosome segregation errors, leading to aneuploidy. Exploring how LZTS2 is involved in these processes would be crucial for understanding its potential role in cancer development. Future work should focus on better characterizing these functions in suitable cancer models.
Materials and Methods
4
Cell Lines and Culture Conditions
4.1
hTERT RPE‐1 cells were cultured in DMEM/F‐12 (Dulbecco's modified Eagle medium/Nutrient Mixture F‐12) (Gibco) supplemented with 10% fetal bovine serum (Biowest), 0.348% sodium bicarbonate (Gibco), 2 mM L‐glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in a humidified atmosphere with 5% CO_2_.
U2OS cells were cultured in DMEM (Gibco) supplemented with 10% fetal bovine serum (Biowest), 2 mM L‐glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in a humidified atmosphere with 5% CO_2_.
Cell Transfection and Ciliation Assays
4.2
For transfection with siRNAs, 5 × 10^4^ cells were seeded in 24‐well plates containing glass coverslips. On the following day, cells were transfected using Lipofectamin RNAiMAX Transfection Reagent (Thermo Fisher Scientific) following the manufacturer's protocol (Peneda et al. 2020). Forty‐eight hours after transfection, the cells were fixed with 100% methanol at −20°C for 10 min and processed for immunofluorescence. For sample preparation for western blotting, RNA extraction, or electron microscopy, the same protocol was followed, but the cells were seeded in six‐well plates (2.5 × 10^5^ cells/well) and the amount of transfection reagents adjusted accordingly.
For the ciliation assays, the media was replaced by serum‐free media 24 h (to induce starvation and cilia assembly) after transfection. Cells were fixed with methanol 24 h after media replacement and processed for immunofluorescence.
LZTS2 was depleted using Silencer Select siRNA assay ID s39014 (Thermo Fisher Scientific), and CEP135 was depleted using Silencer Select siRNA assay ID s18589 (Thermo Fisher Scientific). As a control, cells were transfected using Silencer Select siRNA Negative Control No. 1 (Thermo Fisher Scientific) or GL2 (5′CGUACGCGGAAUACUUCG3’) (Dharmacon) (used in two electron microscopy experiments).
qRT‐PCR
4.3
For RNA extraction, cells transfected in six‐well plates were collected and centrifuged at 400g for 5 min, the supernatant was discarded, and the cell pellet was snap‐frozen. RNA was extracted using the RNeasy Mini Kit (Qiagen) and the optional on‐column DNAse I treatment steps were always performed. RNA quantity and purity were assessed using NanoDrop 2000 (Thermo Scientific). cDNA synthesis was carried out using the High‐Capacity RNA‐to‐cDNA kit (Applied Biosystems).
For the expression analysis by RT‐qPCR, iTaq Universal SYBR Green (BioRad) was used and the mixes were prepared according to the manufacturer's protocol. Samples were prepared in triplicate in 384‐well plates. The qPCR run and data analysis were performed using QuantStudio (Applied Biosystems). The following primers were used for quantifying the mRNA levels of GAPDH (forward: 5′ACATCGCTCAGACACCATG 3′; reverse: 5′TGTAGTTGAGGTCAATGAAGGG3′), LZTS2 (forward: 5′TGGAAAGCTGGAGAAGAA CATGG3′; reverse: 5′GACCCATGAAGCTAGGCAGG3′).
Co‐IP Assays
4.4
For co‐IP assays, Dynabeads Protein G (Invitrogen) were washed three times with lysis buffer (10 mM Tris pH 7.4, 5 mM EDTA, 100 mM NaCl, 1% Triton X‐100, 0.2 mM Na3VO4, 50 mM NaF, 1 mM DTT and protease inhibitor cocktail (Roche)) using a magnetic rack and incubated with 25 μL of anti‐LZTS2 (rabbit, Proteintech, 15677‐1‐AP) or anti‐IgG (Thermo Fisher Scientific, 31243) antibodies diluted in 500 μL lysis buffer for an hour with agitation at room temperature. Beads coupled with antibodies were then washed three times with lysis buffer and finally resuspended in 500 μL lysis buffer. In parallel, 2.5 × 10^6^ cells per condition were centrifuged at 200g for 5 min, and pellets were resuspended in 100 μL ice‐cold lysis buffer and incubated on ice for 30 min. The lysates were then centrifuged at 14,500g at 4°C for 15 min, and the supernatants were transferred to a new tube. Twenty microliter of each supernatant was stored (input of co‐IP) and the remaining supernatant was added to the previously prepared beads coupled with antibodies and incubated with agitation at 4°C overnight. Beads were washed three times with lysis buffer using a magnetic rack and pelleted beads (bound fraction of co‐IP) and the input fraction was resuspended in previously heated 25 and 20 μL of 2× loading buffer (0.02 M Tris pH 6.8, 4% (w/v) SDS, 16% (v/v) glycerol, 3% (w/v) DTT and 0.02% bromophenol blue), respectively, boiled at 95°C for 5 min, and analyzed by western blotting.
Western Blotting
4.5
For total protein cell extracts preparation, transfected cells were detached with TrypLE Express (Gibco) for 5 min at 37°C. Cells were collected, centrifuged, and pellets were resuspended in previously heated 2× loading buffer (see composition in Section 4.4), boiled for 5 min at 95°C, and incubated with 1 μL of 25–29 U benzonase (Merck Millipore) for 30 min at room temperature. Samples were boiled for an additional 5 min at 95°C. For both total cell extracts and co‐IP samples, proteins were separated by 4%–15% Tris‐Glycine gradient SDS‐PAGE (BioRad) and transferred onto 0.22 μm nitrocellulose membranes (LI‐COR). Membranes were blocked with 5% (w/v) milk in tris‐buffered saline (TBS) for 1 h at room temperature, followed by incubation with the primary antibodies anti‐LZTS2 (rabbit, 1:500, Proteintech,15677‐1‐AP), anti‐LZTS2 (mouse, 1:500, Santa Cruz Biotechnology, sc‐271958), anti‐CEP135 (rabbit, 1:500, Abcam, ab75005), anti‐β‐Actin (mouse, 1:1000, Abcam, ab6276) at 4°C overnight. Membranes were then washed with 1× TBS and incubated with the secondary antibodies IRDye 800RD Goat anti‐Rabbit (1:10 000, #926‐32211, LI‐COR) or IRDye 680RD Goat anti‐Mouse (1:10 000, #926‐32220, LI‐COR) for 1 h at room temperature. Membranes were washed and developed using an Odyssey imager (LI‐COR).
Electron Microscopy
4.6
For electron microscopy processing, cells were seeded in six‐well plates and transfected following the protocol described above. In this case, four 13 mm coverslips were added to each well to increase the number of cells to be analyzed. U2OS cells were transfected with Control (GL2 or Negative Control No. 1) or LZTS2 siRNA as described in 4.1 and then processed for electron microscopy analysis. First, the fixative solution (2% formaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate buffer [PB], pH 7.4) was added to an equal volume of culture media and incubated for 15 min at room temperature. The solution was removed and samples were incubated in fixative solution for 1 h at room temperature. Samples were then washed three times in 0.1 M PB (pH 7.4) and postfixed in 1% osmium in 0.1 M PB for 1 h on ice in the dark. Samples were washed with 0.1 M of PB (pH 7.4) for 5 min, twice. Then, they were incubated with dH_2_O for 5 min, twice. Samples were stained with 1% of tannic acid in H_2_O for 20 min on ice and washed for 5 min with dH_2_O five times. Samples were treated with 0.5% Uranyl Acetate in H_2_O for 1 h at room temperature and in the dark. Then, the samples were dehydrated in 30% ethanol for 10 min, next in 50% ethanol for 10 min, and finally in 75% ethanol overnight at 4°C. Samples were dehydrated in 90% ethanol for 10 min and in 100% ethanol for 10 min, three times. For embedding, the processed coverslips were added directly on top of a beam capsule filled with Embed‐812 epoxy resin with cells facing down. Resin was hardened overnight at 60°C. Sections of the blocks were cut using a Leica UC7 ultramicrotome with a thickness of 70 nm and picked in slot grids coated with 1% formvar in chloroform. The sections were stained with 2% uranyl acetate in 70% methanol for 5 min, followed by Reynold's Lead Citrate for 5 min. Image acquisition was performed in a Hitachi H‐7650 Transmission Electron Microscope (100 kV) with an XR41M mid mount AMT digital camera. The images captured were analyzed using ImageJ. For measuring length, 27 centrioles were analyzed in the control and 41 in the LZTS2 siRNA condition. For the diameter analyses, 44 centrioles were analyzed in the control and 55 in the LZTS2 siRNA condition.
Microtubule Regrowth Assays
4.7
For microtubule regrowth assays, cells were incubated on ice for 40 min. After that, the media was replaced by warmed media (37°C) and the plate was transferred to a 37°C incubator to allow microtubule nucleation to occur for 1 min. Cells were washed with a pre‐extraction buffer (0.5% Triton X‐100 in PBS) for 20 s (before fixation time) and immediately fixed with 100% methanol at −20°C for 10 min. Samples were then processed for immunofluorescence as described below.
Immunofluorescence and Imaging
4.8
After fixation, samples were blocked with 10% FBS in PBS 1× for 30 min at room temperature. Samples were incubated with primary antibodies for 1 h at room temperature. The primary antibodies used in this study are: anti‐Centrin (mouse, 1:500, Milipore, 04‐1624), anti‐CEP135 (rabbit, 1:500, Abcam, 75005), anti‐α‐tubulin (mouse, 1:200, Sigma‐Aldrich, DM1A), anti‐Pericentrin (rabbit, 1:1000, Abcam, AB4448), anti‐polyglutamylated tubulin (mouse, 1:500, AdipoGen Life Sciences, AG‐20B‐0020 clone GT335), anti‐Ninein (mouse, 1:100, Santa Cruz Biotechnology, sc‐376420), anti‐CEP164 (goat, 1:500, Santa Cruz Biotechnology, sc‐240226), anti‐CDK5RAP2 (rabbit, 1:5000, Millipore, 06‐1398), anti‐γ‐tubulin (mouse, 1:200, Sigma‐Aldrich, T6557, clone GTU88) anti‐Centriolin (mouse, 1:100, Santa Cruz Biotechnology, sc‐365521).
Samples were then washed three times with PBS. After that, cells were incubated with secondary antibodies and DAPI (to stain DNA) at room temperature for 1 h. The secondary antibodies used in this study were donkey anti‐mouse Alexa Fluor 555 or 568 (1:500, Molecular probes), donkey anti‐mouse Alexa Fluor 488 (1:500, Molecular probes), donkey anti‐rabbit Alexa Fluor 488 (1:500, Molecular probes), and donkey anti‐rabbit Alexa Fluor 555 or 568 (1:500, Molecular probes). Samples were washed with PBS, and each coverslip was mounted in Vectashield (Vector) mounting media (microtubule regrowth assays) and sealed with nail polish or in Dako Fluorescence mounting medium (Dako). Image acquisition was performed using an Andor Dragonfly 200 Spinning Disk Confocal microscope, Fusion Software, and Z‐stack images were acquired using a 100× oil objective. SIM images were acquired using a GE HealthCare Deltavision OMX system, equipped with two PCO Edge 5.5 sCMOS cameras, using a 60 × 1.42NA Oil immersion objective. Images were deconvolved and reconstructed with Applied Precision's softWorx software.
Image Analysis and Fluorescence Quantification
4.9
Images were analyzed using the Fiji/ImageJ software. For the analysis of CEP135 intensity at the centrosome, MAX Z‐Projections were obtained to define the ROIs (regions of interest) to be analyzed. The ROIs were defined using the “Triangle” threshold, and to guarantee that only bona fide centrosomes were analyzed, only the ones that showed a co‐localization of both markers were considered (centrin and CEP135) The measurement of the signal intensity was then performed in the SUM Z‐projection of the image, by extracting the Raw Integrated Density for each defined ROI. The same process was used to measure the intensity of microtubule asters, using the “Triangle” threshold to identify the ROIs corresponding to the asters, and for the centrosomal levels of γ‐tubulin and CDK5RAP2. For Pericentrin and Centriolin, the thresholds “Moments” and “MaxEntropy” were used, respectively.
Statistical Analysis and Data Visualization
4.10
All plots were generated using GraphPad Prism software. For all statistical analyses, the significance level was set to 0.05. Statistical analyses of the ciliation assays and the length and diameter analysis by electron microscopy were performed using GraphPad Prism (two tailed Student's t test). For ciliation assays, we first calculated the log‐odds ratio of cells with cilia. Statistical analyses of centriole number quantifications, microtubule regrowth assays, and centrosomal intensities of CEP135 were performed in R (v. 4.4.1), as described below.
For centriole number analyses, we first binned the data into three distinct categories: cells with fewer centrioles than expected (< 2), wild‐type‐like centriole numbers (2–4), and bona fide extra centrioles (> 4). First, we performed Fisher's exact tests (using the built‐in function in base R, fisher.test) between all pairs of independent experiments for each condition (control and LZTS2 siRNA in Figure 1D, and control siRNA, LZTS2 siRNA, CEP135 siRNA, and both LZTS2 and CEP135 siRNAs in Figure 3C) and confirmed that there were no significant differences across different independent experiments. Then, we pooled the data across independent experiments and performed Fisher's exact test for all pairs of conditions. In all these analyses, *p‐*values were corrected for multiple testing according to the Bonferroni method.
For microtubule regrowth assays and centrosomal CEP135, γ‐tubulin, CDK5RAP2, Pericentri,n and Centriolin intensity measurements, the setup we designed included two independent variables/predictors (LZTS2 siRNA and CEP135 siRNA) and we further tested their interaction (co‐depletion of LZTS2 and CEP135). We employed a linear mixed model using the lmer function (from the package lme4) taking the log‐transformed raw integrated densities as a normally distributed response variable, LZTS2 and CEP135 siRNAs as independent predictors, and their interaction, and experimental batch as a randomeffectst. To perform multiple comparisons between pairs of conditions, we estimated marginal means for each combination of predictors using the emmeans function (from package emmeans, v. 1.10.6) and determined significance by comparing pairwise means and correcting for multiple testing using the Tukey method.
Author Contributions
C.P. performed experiments for centriole number quantification, qRT‐PCRs, and analysis of centrosomal CEP135. Centriole counting was performed by J.N.B. and M.A.D.L. C.P. and J.N.B. performed ciliation and regrowth assays. C.P. and A.H.S. performed electron microscopy experiments and analysis. C.P. and J.N.B. performed immunoprecipitations and western blotting, analysis of centrosomal levels of PCM proteins and Centriolin, and designed the experiments. M.A.D.L. performed all statistical analyses. C.P. conceptualized the study. C.P., J.N.B., and M.B.D. critically discussed the results. C.P. and M.A.D.L. wrote the manuscript. J.N.B. and M.B.D. edited the manuscript.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Figure S1. LZTS2 knockdown validation. (A,B) qRT‐PCR analysis of LZTS2 relative expression in hTERT RPE‐1 cells 48 h after transfection with control or LZTS2 siRNA (A) and after 24 h of serum starvation (B). (C,D) Analysis of LZTS2 relative expression (C) or protein levels (D) in U2OS cells 48 h after transfection with control or LZTS2 siRNA. β‐actin was used as loading control. Figure S2. Knockdown validation in U2OS cells. U2OS cells were transfected with control, LZTS2, CEP135 or a combination of LZTS2 and CEP135 siRNAs. The efficiency of the knockdown was confirmed by western blotting for LZTS2 and CEP135 proteins. β‐actin was used as loading control.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Adams, S. D. , J. Csere , G. D'angelo , et al. 2021. “Centrosome Amplification Mediates Small Extracellular Vesicle Secretion via Lysosome Disruption.” Current Biology: CB 31, no. 7: 1403–1416 e 7. 10.1016/j.cub.2021.01.028.33592190 PMC 8047808 · doi ↗ · pubmed ↗
- 2Amack, J. D. 2022. “Structures and Functions of Cilia During Vertebrate Embryo Development.” Molecular Reproduction and Development 89, no. 12: 579–596. 10.1002/mrd.23650.36367893 PMC 9805515 · doi ↗ · pubmed ↗
- 3Anderson, R. G. 1972. “The Three‐Dimensional Structure of the Basal Body From the Rhesus Monkey Oviduct.” Journal of Cell Biology 54, no. 2: 246–265. 10.1083/jcb.54.2.246.5064817 PMC 2108883 · doi ↗ · pubmed ↗
- 4Barbelanne, M. , and W. Y. Tsang . 2014. “Molecular and Cellular Basis of Autosomal Recessive Primary Microcephaly.” Bio Med Research International 2014: 547986. 10.1155/2014/547986.25548773 PMC 4274849 · doi ↗ · pubmed ↗
- 5Bertalan, Z. , Z. Budrikis , C. A. La Porta , and S. Zapperi . 2015. “Role of the Number of Microtubules in Chromosome Segregation During Cell Division.” P Lo S One 10, no. 10: e 0141305. 10.1371/journal.pone.0141305.26506005 PMC 4624697 · doi ↗ · pubmed ↗
- 6Bodakuntla, S. , A. S. Jijumon , C. Villablanca , C. Gonzalez‐Billault , and C. Janke . 2019. “Microtubule‐Associated Proteins: Structuring the Cytoskeleton.” Trends in Cell Biology 29, no. 10: 804–819. 10.1016/j.tcb.2019.07.004.31416684 · doi ↗ · pubmed ↗
- 7Byrne, F. L. , L. Yang , P. A. Phillips , et al. 2014. “RN Ai‐Mediated Stathmin Suppression Reduces Lung Metastasis in an Orthotopic Neuroblastoma Mouse Model.” Oncogene 33, no. 7: 882–890. 10.1038/onc.2013.11.23396365 · doi ↗ · pubmed ↗
- 8Carvalho‐Santos, Z. , P. Machado , I. Alvarez‐Martins , et al. 2012. “BLD 10/CEP 135 Is a Microtubule‐Associated Protein That Controls the Formation of the Flagellum Central Microtubule Pair.” Developmental Cell 23, no. 2: 412–424. 10.1016/j.devcel.2012.06.001.22898782 · doi ↗ · pubmed ↗