The development of a methodology to measure primary ciliary length in neural progenitor cells with a target molecule knocked down

Abstract

The primary cilium is an immotile cellular antenna that extends from the basal body and protrudes from the cell surface during the G₀/G₁ phases. The cilium is resorbed when it receives growth factor stimuli, and the ciliary resorption triggers the cell cycle re-entry into the G₁/S phases. The dysregulation of ciliary dynamics during embryonic development can lead to various hereditary organ dysplasias, including microcephaly. Neural progenitor cells display primary cilia on their apical surfaces during the embryonic stage. Ciliary resorption in the cells is responsible for cell proliferation and corticogenesis. However, the molecular mechanisms underlying ciliary resorption in the neural progenitor cells in vivo are poorly understood. Mapping cilia to knockdown cells on a one-to-one basis is technically challenging and represents the biggest barrier to solving these mechanisms. In this study, we developed a short hairpin RNA (shRNA)-based pCAGI-Arl13b-tdTomato plasmid to label cilia in knockdown cells. This plasmid contains a shRNA sequence and a cilium marker, Arl13b-tdTomato. Using in utero electroporation, we transfected the plasmid into embryonic neural progenitor cells and found that the primary cilia of the transfected cells specifically expressed Arl13b-tdTomato. We also found that Arl13b-tdTomato expression did not alter the ciliary length. Microtubule-associated serine/threonine kinase and t-complex testis expressed-1 are major regulators of ciliary resorption. Knocking down each of these molecules resulted in longer cilia. These results suggest that the pCAGI-Arl13b-tdTomato plasmid is useful for measuring ciliary length in the shRNA-transfected developing cortical neural progenitor cells in vivo.

Highlights

• •Cell culture studies have shown that MAST4 and Tctex-1 regulate ciliary resorption.
• •Ciliary resorption mechanisms in neural progenitor cells in vivo remain unclear.
• •ShRNA in the pCAGI-Arl13b-tdTomato plasmid was used to knockdown target genes.
• •Arl13b-tdTomato⁺ cilia were selectively observed only in the knockdown cells.
• •The pCAGI-Arl13b-tdTomato plasmid is useful for measuring ciliary length.

1. Introduction

Primary cilia originate from the centrosome-derived basal body, and appear on the cell surface during the G₀/G₁ phases of the cell cycle in nearly all cell types. The cilia contain selective membrane receptors and ion channels that function as cellular antennae [1]. When the cilium receives selective growth factors, they are resorbed, allowing the cells to re-enter the cell cycle from a quiescent state [2,3]. Disorders of cilium formation and/or function during embryonic development lead to systemic organ malformations known as ciliopathies. These include microcephaly, polycystic kidney disease, retinopathy, and bone malformations [4].

During the G₀/G₁ phase of the embryonic stage, neural progenitor cells of the cerebral cortex expand from the cortical to the ventricular surfaces, with their cell bodies located in the intermediate zone. Each cell expresses a primary cilium on the ventricular surface. When insulin-like growth factor-1 (IGF-1) in the cerebrospinal fluid acts on the cilia, the cells re-enter the cell cycle in the G₁/S phases and move to the ventricular zone during the M phase [5]. Thus, ciliary resorption is considered a trigger for cell cycle re-entry, cell proliferation, and tissue formation.

Studies using cell lines, including fibroblasts (e.g. MEFs and NIH3T3) and retinal pigment epithelial cells (RPE-1), have revealed the physiological functions and molecular mechanisms of ciliary resorption [2,3,6]. These studies demonstrate that pathways of both microtubule-associated serine/threonine protein kinase 4 (MAST4)/t-complex testis expressed-1 (Tctex-1) and Aurora kinase A (AurkA)/histone deacetylase 6 (HDAC6) are key regulatory mechanisms of ciliary resorption [2,6]. Tctex-1 was originally identified as a light chain in the cytoplasmic dynein complex [7]. It carries various intracellular components, including G protein-coupled receptors [8,9]. By contrast, it is released from the dynein complex when phosphorylated at threonine 94 (T94), allowing it to play a dynein-independent role [10]. MAST4 is a serine/threonine kinase that is involved in neurodevelopment and spermatogenesis, showing its roles in cell proliferation in vivo [[11], [12], [13]]. Recent studies have shown that ciliary IGF-1 receptor (IGF-1R)-derived signaling induces Tctex-1 phosphorylation at residue T94 in a ciliary basal structure called the transition zone via the action of MAST4 [5,6,14]. These molecular events are essential for ciliary resorption and cell cycle reentry. These studies demonstrated that MAST4 and Tctex-1 plays a critical role in ciliary resorption and subsequent cell proliferation. Short hairpin RNA (shRNA)-mediated knockdown (KD) of target molecules is a convenient and powerful strategy that can facilitate the study of the functions of these molecules.

Cilia-mediated cell cycle progression regulates neocortex maturation [15]. Knockout (KO) of the centrosomal proteins Cenpj, KIF2A, and WDR62 in mice, as well as mutation of the centrosomal protein RRP7A in humans, is linked to a decrease in the reentry of neural progenitor cells into the cell cycle, the generation of premature neurons, and a small brain size [[16], [17], [18]]. Furthermore, longer cilia have been observed in the brain and organoids prepared of KO mice [17,18]. These studies suggested a link between ciliary resorption and neocortical maturation. However, there have been less information about molecular mechanisms underlying ciliary resorption in neural progenitor cells in vivo. To unveil molecules that regulates ciliary resorption in the cells, measuring ciliary length is critical experiment. KO mice is one of the reliable tools for this purpose, however, as generating KO mice generally requires a longer time and effort and as more space is required to maintain them in a breeding facility. shRNA-mediated knockdown of target molecules is considered an effective strategy to elucidate the roles of the molecules in ciliary resorption in neural progenitor cells in vivo.

In utero electroporation (IUE) is an efficient gene transfer system for neural progenitor cells in the developing neocortex [19]. As wild-type animals can be transfected with genes of interest, this technique allows researchers to conveniently and quickly address the functions of target molecules in vivo. The pCAG-IRES-GFP (pCAGIG) plasmid contains a chicken β-actin promoter with a cytomegalovirus enhancer (CAG) promoter, as well as an internal ribosome entry site (IRES)-regulated green fluorescent protein (GFP) sequence. This allows the identification of GFP⁺ cells as the transfected cells [20]. Transfection of the pCAGIG plasmid carrying the Tctex-1 shRNA sequence (pCAG-Tctex-1-shRNA-IRES-GFP) into neural progenitor cells of the developing neocortex decelerates cell cycle reentry and proliferation [6]. However, it is unclear whether Tctex-1 contributes to ciliary resorption in neural progenitor cells, because it is extremely difficult to clearly discriminate between the cilia of GFP⁺ transfected cells and those of GFP⁻ untransfected cells. That is, it was not possible to measure ciliary length in Tctex-1-KD cells owing to technical issues. Furthermore, the roles of MAST4 in ciliary length control and neural progenitor cell proliferation has not yet been elucidated.

The aim of this study was to establish a method for measuring ciliary length in the developing neural progenitor cells in which a target molecule was knocked down using shRNA, especially by focusing on MAST4 and Tctex-1. We created a pCAG-shRNA-IRES-Arl13b-tdTomato plasmid (hereafter referred to as pCAGI-Arl13b-tdTomato), containing shRNA and an IRES-regulated Arl13b-tdTomato sequences that acted as a cilia marker. Transfection of the pCAGI-Arl13b-tdTomato plasmid into neural progenitor cells by IUE enabled us to measure ciliary length in cells with the target molecule knocked down. Using this tool, we successfully demonstrated that downregulation of MAST4 and Tctex-1 provided primary cilia longer in neural progenitor cells in vivo during the embryonic stage.

2. Materials and methods

2.1. Antibodies

The following primary antibodies were used: rabbit anti-Arl13b (Proteintech, 17711-1-AP; IB: 1:2,000, IF: 1:1000), mouse anti-acetylated α-tubulin (Ac-Tub) (Sigma-Aldrich, T6793; IF: 1:2000), mouse anti-γ-tubulin (γ-Tub) (Sigma-Aldrich, T6557; IF: 1:1000), mouse Tubulin β3 (TUJ1) (BioLegend [Covance], MMS-435P; IF: 1:500), chicken anti-GFP (Abcam, ab13970; IF: 1:1000), chicken anti-mCherry (Abcam, ab205402; IF: 1:1000), rabbit anti-RFP (Medical & Biological Laboratories, PM005; IB: 1:2,500, IF: 1:1000), and mouse anti-β-actin (Sigma-Aldrich, A5441; IB: 1:5000). The following secondary antibodies were used: CF dye-conjugated secondary antibodies (Biotium; IF: 1:400) and horse radish peroxidase (HRP)-conjugated secondary antibodies (Cell Signaling Technology; IB: 1:5000).

2.2. Plasmid preparation

The pArl13b-GFP expression plasmid [21] and pCAGI-mGFP plasmid, in which the membrane GFP (mGFP) sequence was driven by a CAG promoter, were kindly provided by Dr. Kenji Kontani (Mejji Pharmaceutical University) and Dr. Ching-Hwa Sung (Weill Cornell Medical College), respectively.

To generate pCAG-mtdTomato (membrane bound form of tdTomato), growth associated protein-43 palmitoylation signal (Rat, XM_032900206.1, amino acid sequence; MLCCMRRTKQVEKNDEDQK) [22] was inserted at the N-terminus of tdTomato in pCAG-tdTomato, which was a gift from Dr. Angelique Bordey (Addgene plasmid # 83029; http://n2t.net/addgene:83029; RRID:Addgene_83029) [23]. The mGFP sequence in the pCAGI-mGFP plasmid was replaced with the mtdTomato sequence, which was digested from the pCAG-mtdTomato plasmid, to generate the pCAGI-mtdTomato plasmid. To insert a ciliary targeting signal (CTS) sequence (amino acid sequence; LVCCWFKKSKTRKIKPE) [24] at the C-terminus of tdTomato, we first generated the pCAGI-GFP-CTS plasmid. The pCAGIG plasmid was digested with BsrG-I and Bgl-II and oligonucleotides encoding the CTS sequence and a stop codon with BsrG-I and Bgl-II sites was inserted into the digested pCAGIG plasmid to produce pCAGI-GFP-CTS. Next, we replaced the GFP sequence in the pCAGI-GFP-CTS plasmid with the tdTomato sequence using an in-fusion technique to produce the pCAGI-tdTomato-CTS plasmid. To produce the pCAGI-Arl13b-tdTomato plasmid, we first created the pCAGI-Arl13b-GFP plasmid. The Arl13b sequence was amplified by polymerase chain reaction (PCR), excluding the stop codon, using the pArl13b-GFP expression plasmid as a template. The PCR products were then inserted between the IRES and GFP sequences of the pCAGIG plasmid, and the frames of the Arl13b and GFP sequences were aligned to produce the pCAGI-Arl13b-GFP plasmid. The GFP sequence in pCAGI-Arl13b-GFP was replaced with a tdTomato sequence to produce the pCAGI-Arl13b-tdTomato plasmid. shRNA sequences targeting mouse MAST4 (GGG AAC AGG TGG AGC CTA TGA A) and mouse Tctex-1 (GGG TTA CAC TCC GCA AGT TCC) were inserted behind the U6 promoter to generate the pCAG-MAST4-sh-I-Arl13b-tdTomato and pCAG-Tctex-1-sh-I-Arl13b-tdTomato as previously described [6,14].

2.3. IUE

IUE procedures were performed in Slc:ICR mouse (Japan SLC, Hamamatsu, Japan) brains at embryonic day 13.5 using plasmids driven by either the CAG or U6 promoter, as previously described with minor modifications [6]. Briefly, pregnant mice were anesthetized by an intraperitoneal injection of ketamine (120 mg/kg) and xylazine (6 mg/kg). Following laparotomy, the plasmid solution was microinjected into the fetal cerebral ventricles and in utero electroporation was performed using forceps-type electrodes (4 pulses, 45 V, 50 ms, and 950 ms intervals; CUY21 electroporator, NEPA GENE Co., Ltd., Chiba, Japan). The abdominal wall was closed with sutures, and the animals were maintained until tissue collection. All procedures were approved and monitored by the Institutional Animal Care and Use Committee of Gifu Pharmaceutical University. Electroporated brains were harvested 40 h after IUE, fixed in 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer for 3 h at room temperature, immersed in 30% sucrose at 4 °C for 24 h, and embedded in optimal cutting temperature compound (Sakura Finetek Co., Ltd., Tokyo, Japan). Cryosections (30 μm) were prepared using a Leica CM1520 cryostat (Leica Biosystems, Nussloch, Germany), mounted on MAS-coated glass slides (Matsunami Glass Ind., Ltd., Osaka, Japan), and processed for immunostaining.

2.4. Cell culture and transfection

Human embryonic kidney 293 (HEK293) and NIH3T3 cells were obtained from the Cell Resource Center for Biomedical Research, Tohoku University (Miyagi, Japan). C3H10T1/2 cells were provided by the Japanese Collection of Research Bioresources (Osaka, Japan). RPE-1 cells were purchased from the American Type Culture Collection (cat. #CRL-400).

C3H10T1/2, HEK293, and NIH3T3 cells were cultured in Dulbecco's modified Eagle medium (DMEM) (Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich, St. Louis, MO, USA). RPE-1 cells were cultured in DMEM/F12 (FujiFilm-Wako, Osaka, Japan) supplemented with 10% FBS. All cells were maintained in a humidified atmosphere at 37 °C and 5% CO₂. The cells were transfected with plasmid DNA using an ECM 830 Square Wave Electroporation System (BTX Harvard Apparatus, Holliston, MA, USA). The pulse conditions were as follows: 100 V, 10 ms, 5 pulses, and a 100 ms interval for NIH3T3 cells; 175 V, 10 ms, 2 pulses, and a 100 ms interval for C3H10T1/2 cells; and 115 V, 30 ms, 2 pulses, and a 100 ms interval for RPE-1 cells. Two days after transfection, the cells were fixed with 4% PFA in PBS for 10 min at room temperature or harvested for western blotting.

2.5. Immunofluorescence

The fixed brain cryosections were treated with HistoVT One (Nacalai Tesque Inc., Kyoto, Japan) at 70 °C for 20 min for antigen retrieval. Cryosections or fixed cells were subjected to a standard immunofluorescence protocol [6]. The samples were treated with 50 mM NH₄Cl in PBSc/m (PBS supplemented with 0.2 mM Ca²⁺ and 2 mM Mg²⁺) for 10 min to quench PFA. They were then blocked with BTPA (PBSc/m supplemented with 0.25% Triton X-100, 0.5% BSA, and 0.02% NaN₃) for 30 min and incubated with primary antibodies in BTPA for 60 min. Finally, cells were incubated with secondary antibodies in BTPA for 45 min. Images were acquired using a LSM880 confocal microscope (Zeiss, Oberkochen, Germany) with a 10 × or 63 × objective lens and a z-interval of 0.25 μm. The length of the cilia was measured using the acquired confocal images and NIH ImageJ software. For presentation purposes, the brightness of the images was slightly modified using Adobe Photoshop Elements 2021 Editor software (Adobe Systems, San Jose, CA, USA).

2.6. Immunoblotting

Cell lysates were prepared as described previously [14]. Briefly, cells were washed twice with STE buffer (150 mM NaCl, 50 mM Tris-HCl, and 2 mM EDTA, pH 7.4) on ice and lysed in STET buffer (STE buffer containing 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, and 1 × protease inhibitor cocktail). The lysates were rotated for 30 min at 4 °C, and centrifuged at 13,000×g for 10 min at 4 °C. The resulting supernatant, which contained an equal amount of protein, was mixed with 5 × Leammli sample buffer (final concentrations of 62.5 mM Tris-HCl, 10% glycerol, 2% SDS, 5% β-mercaptoethanol, and 0.5 mg/mL bromophenol blue; pH 6.8) and boiled at 75 °C for 5 min. Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes. The membrane was then incubated in TBST (50 mM Tris-HCl, 150 mM NaCl, and 0.02% Tween 20; pH 7.4) containing 4% nonfat dry milk (NFDM) for 1 h at room temperature. Next, the membrane was probed with primary antibodies in TBST containing 1% NFDM overnight at 4 °C, followed by incubation with HRP-conjugated secondary antibodies in TBST containing 1% NFDM for 1 h at room temperature. Finally, the membrane was treated with a chemiluminescent HRP substrate (ImmunoStar LD, Fujifilm-Wako) and digitally recorded using ChemiDoc XRS1/Image Laboratory software (Bio-Rad, Hercules, CA, USA).

2.7. Statistics

Graph drawing and statistical analyses were performed using GraphPad Prism software (ver. 8.4). The following tests were performed: Student's t-test, one-way analysis of variance (ANOVA) followed by Tukey's test, and two-way ANOVA followed by Tukey's test. A p-value less than 0.05 was considered as statistically significant. The number of samples is indicated in the figure legends.

3. Results

3.1. GFP and membrane-tdTomato are not suitable for labeling cilia in cortical neural progenitor cells

Neural progenitor cells of the cerebral cortex expand from the cortical to the ventricular surfaces and express primary cilia on the ventricular surface (Fig. 1A). shRNA-based knockdown using the pCAGIG plasmid is one of the most efficient methods for downregulating target molecules in neural progenitor cells of the developing cortex [6] (Fig. 1A and B). As the pCAGIG plasmid encodes GFP, GFP⁺ cells were identified as transfected cells in this system. Cortical neural progenitor cells project Arl13b⁺ primary cilia, formed from γ-tubulin⁺ basal bodies on the ventricular surface (Supplementary Fig. 1). First, we investigated whether the cilia in GFP⁺ neural progenitor cells appear GFP. Two days after transfection of the pCAGIG plasmid into an E13.5 embryonic mouse brain, the entire brain was harvested, fixed, and immunostained with anti-Arl13b (a primary cilium marker) and anti-GFP antibodies. As previously reported [6], the majority of the GFP signal was detected in the intermediate zone, with some detected between the subventricular and ventricular zones (Fig. 1C and D). In addition, Arl13b-labeled primary cilia were abundant on the ventricular surface (Fig. 1C–E). Three examples of GFP signals localized in the ventricular zone are illustrated in Fig. 1D and E. Contrary to our expectations, multiple Arl13b⁺ cilia were associated with the GFP signal (Fig. 1E, arrowheads), indicating the difficulty in mapping cilia to GFP⁺ cells on a one-to-one basis. Furthermore, although some cilia were expected to organize into GFP⁺ cells, none were labeled with GFP. We expect that these GFP signals in the ventricular zone are the fiber ends of the G₁/S-phase neural progenitor cells from intermediate zone-localizing cells and not the cell bodies of M-phase cells, because the GFP signals lack M-phase nuclei (Fig. 1D and E).

Fig. 1.

Cilia are not labeled by cytosolic expression of green fluorescent protein (GFP). (A) A schematic diagram of the cell cycle-dependent migration of neural progenitor cells (NPC) and generation of neurons. NPC contact both the cortical and ventricular surfaces and project primary cilia into the ventricular spaces. The cell bodies of NPC are located in the intermediate zone (IZ) during the G₀ phase, and migrates to the ventricular zone (VZ) during the M phase. NPC generate multipolar post-mitotic neurons. The neuron migrates in the IZ. CP: cortical plate. (B) A schematic diagram of a pCAGIG plasmid. CAG: chicken β-actin promoter; IRES: internal ribosome entry site. (C) Low-power field images of the pCAGIG-transfected cortex. The expression of GFP and primary cilia in the cryosections was immunolabeled with anti-GFP (green) and anti-Arl13b (red) antibodies, respectively. The nuclei were stained with DAPI (blue). Arl13b-labeled cilia projected significantly onto the ventricular surface. (D and E) Enlarged views of the boxed area in C and D, respectively. The numbers in D and E correspond to each other. The arrowheads indicate cilia associated with GFP⁺ fiber end of NPC. Scale bars: 50 μm (C), 20 μm (D), and 1 μm (E).

mGFP was effectively distributed across the cell membrane. We hypothesized that the cilia of neural progenitor cells can be labeled with mGFP. In contrast, there is a technical issue with the use of GFP as a marker for transfected cells. Typically, we select cryosections containing transfected cells by observing GFP fluorescence under a fluorescence microscope before labeling them with an anti-GFP antibody [6]. However, in this study, the GFP-derived fluorescence was too weak to identify the transfected cells in cryosections. Therefore, we switched from GFP to tdTomato. For this reason, we replaced the GFP sequence in the pCAGIG plasmid not only with the mGFP and but also with membrane-bound form of tdTomato (mtdTomato) sequences to generate the pCAGI-mG and pCAGI-mtdTomato plasmids, respectively (Fig. 2A, Supplementary Fig. 2A–B). We transfected these plasmids into the E13.5 embryonic mouse brain. First, we confirmed that tdTomato was visible under a fluorescence microscope before labeling the cryosections with an anti-mCherry antibody. We found that the fiber end membranes of neural progenitor cells appeared to be labeled better by mGFP and mtdTomato than by GFP at the γ-tubulin-labeled ventricular surface (Fig. 1C–E vs. Supplementary Fig. 2C–E). We also found that mtdTomato was distributed at the Arl13b-labeled cilia; however, the amount of mtdTomato in the cilia was very limited (Fig. 2B–D; arrowhead). Hence, mtdTomato is not suitable for labeling cilia in transfected neural progenitor cells for our purpose.

Fig. 2.

Only a small amount of membrane-tdTomato is distributed at cilia. (A) A schematic diagram of a pCAGI-membrane-tdTomato (pCAGI-mtdTomato) plasmid. CAG: chicken β-actin promoter; IRES: internal ribosome entry site; mem: membrane-targeted site. (B) Low-power field images of the cortex transfected with pCAGI-mtdTomato. The expression of mtdTomato, primary cilia, and basal bodies in the cryosections was immunolabeled with anti-mCherry (red), anti-Arl13b (green), and γ-tubulin (γ-Tub) (yellow) antibodies, respectively. The nuclei were stained with DAPI (blue). (C and D) Enlarged views of the boxed area in B and C, respectively. Scale bars: 100 μm (C), 20 μm (D), and 2 μm (E).

3.2. Arl13b-tdTomato selectively localizes in the cilia

As mtdTomato is not suitable for labeling cilia, we fused a peptide or protein that targeted cilia with tdTomato. First, we inserted a cilia-targeting signal (CTS; LVCCWFKKSKTRKIKPE) sequence [24] at the 3′ end of the tdTomato sequence to create the pCAGI-tdTomato-CTS plasmid (Fig. 3A). However, even when the tdTomato-CTS protein was transiently expressed in RPE-1 cells, only a small portion of the tdTomato-CTS protein was localized to the cilia (Fig. 3B and C). Next, we inserted the Arl13b sequence to the 3’ end of the tdTomato sequence to create the pCAGI-Arl13b-tdTomato plasmid (Fig. 4A). The molecular weight of Arl13b-tdTomato (104 kDa) was validated by western blotting using anti-Arl13b and anti-RFP antibodies (Fig. 4B, Supplementary Fig. 3). Immunofluorescence of serum-starved NIH3T3 and C3H10T1/2 cells revealed that the Arl13b-tdTomato signal was selectively detected in Ac-Tub-labeled cilia of tdTomato⁺ cells but not in tdTomato⁻ cells, suggesting that Arl13b-tdTomato selectively localized to cilia (Fig. 4C and D, Supplementary Fig. 4).

Fig. 3.

Ciliary distribution of tdTomato-CTS. (A) A schematic diagram of a pCAGI-tdTomato-CTS plasmid. CAG: chicken β-actin promoter; IRES: internal ribosome entry site; CTS: cilia-targeting signal. (B) RPE-1 cells were transfected with the pCAGI-tdTomato-CTS plasmid and starved of serum for 40 h. The expression of tdTomato-CTS and primary cilia in the cells was immunolabeled with anti-mCherry (red) and anti-acetylated α-tubulin (Ac-Tub) (green) antibodies, respectively. The nuclei were stained with DAPI (blue). (C) Enlarged views of the boxed areas of (1) and (2) in B. The red signal of the boxed area of (1) was both weakly and strongly exposed. Scale bars: 20 μm (A) and 2 μm (B).

Fig. 4.

Arl13b-tdTomato is selectively localized in cilia. (A) A schematic diagram of the pCAGI-Arl13b-tdTomato plasmid. CAG: chicken β-actin promoter; IRES: internal ribosome entry site. (B) Expression of Arl13b-tdTomato (Arl13b-tdT) in HEK293 cells. The cells were harvested and separated by SDS-PAGE. Then, they were immunoblotted with anti-Arl13b, anti-RFP, or anti-β-actin antibodies, respectively. The molecular sizes of tdTomato, Arl13b-tdTomato, and β-actin were 54, 104, and 42 kDa, respectively. (C) The ciliary localization of Arl13b-tdTomato in NIH3T3 cells. NIH3T3 cells were transfected with the pCAGI-Arl13b-tdTomato plasmid and starved of serum for 40 h. The expression of Arl13b-tdTomato, primary cilia, and basal bodies in the cells was immunolabeled with anti-mCherry (red), anti-acetylated α-tubulin (Ac-Tub) (green), and anti-γ-tubulin (γ-Tub) (cyan) antibodies, respectively. The nuclei were stained with DAPI (blue). (D) Enlarged views of the boxed area in C. The left and right panels represent Arl13b-tdTomato⁺ and Arl13b-tdTomato^- cilia, respectively. (E) Suppression of ciliary resorption by knockdown of MAST4 or Tctex-1. NIH3T3 cells were transfected with the pCAGI-Arl13b-tdTomato plasmid that harbors a MAST4-shRNA (MAST4-sh) or a Tctex-1-shRNA (Tctex-1-sh) sequence. The cells were serum-starved for 40 h and then serum-retreated for 2 or 24 h. The rate of cells with cilia was quantified. ^#p < 0.05, ^##p < 0.01, ^###p < 0.001; two-way ANOVA followed by Tukey's test. In the ciliary resorption experiment, 100 cells were counted in each experiment (three independent experiments). Scale bars: 20 μm (C), 2 μm (D).

3.3. Knockdown of MAST4 and Tctex-1 suppress ciliary resorption in cortical neural progenitor cells

To evaluate the knockdown efficiency, mouse MAST4-shRNA (MAST4-sh) and Tctex-1-shRNA (Tctex-1-sh) sequences were inserted into the pCAGI-Arl13b-tdTomato plasmid and subjected to a ciliary resorption assay. NIH3T3 cells were transfected with plasmids and serum starved for 40 h to induce ciliogenesis. The cells were then re-treated with serum for 2 and 24 h to induce ciliary resorption. As demonstrated in our previous studies on RPE-1 cells [14,25], we found that both MAST4-sh and Tctex-1-sh effectively suppressed serum-induced ciliary resorption in these cells (Fig. 4E), indicating that the pCAG-MAST4-sh-I-Arl13b-tdTomato and pCAG-Tctex-1-sh-I-Arl13b-tdTomat plasmids are useful for suppressing ciliary resorption.

The pCAG-MAST4-sh-I-Arl13b-tdTomato and pCAG-Tctex-1-sh-I-Arl13b-tdTomato plasmids were electroporated in utero on one side of the E13.5 mouse brain. The prepared cryosections were subjected to immunostaining with an anti-Arl13b antibody to visualize the entire cilia and an anti-mCherry antibody to label exogenous Arl13b-tdTomato. As the result, entire cilia were positive for Arl13b signaling, whereas some were positive for tdTomato signaling, suggesting that cilia of transfected neural progenitor cells harbor Arl13b-tdTomato (Fig. 5A and B). We note that some of the cilia were confirmed to be positive for Arl13b-tdTomato in the absence of anti-mCherry immunostaining. Additionally, the anti-mCherry antibody enhanced the Arl13b-tdTomato signal, enabling the detection of weakly expressed Arl13b-tdTomato which was not detected when the anti-mCherry antibody was omitted.

Fig. 5.

Knockdown of either MAST4 or Tctex-1 results in elongated primary cilia in developing cortical neural progenitor cells. (A) Ciliary localization of Arl13b-tdTomato in developing cortical neural progenitor cells. E15.5 mouse brains were harvested 2 days after the transfection with pCAGI-Arl13b-tdTomato (control), pCAG-MAST4-sh-I-Arl13b-tdTomato (MAST4-sh), or pCAG-Tctex-1-sh-I-Arl13b-tdTomato (Tctex-1-sh). The expression of Arl13b-tdTomato and primary cilia in the cryosections was immunolabeled with anti-mCherry (red) and anti-Arl13b (green) antibodies, respectively. The nuclei were stained with DAPI (blue). (B) Enlarged views of the boxed area in A. (C) Quantification of the Arl13b-tdTomato⁺ ciliary length. ∗∗p < 0.01, ∗∗∗p < 0.001, One-way ANOVA followed by Tukey's test. In the ciliary length experiment, 134–144 cilia were measured. Scale bars: 5 μm (A), 2 μm (B).

Subsequently, the length of Arl13b-tdTomato⁺ cilia was measured using the obtained confocal images and the NIH image software. First, we ascertained whether Arl13b-tdTomato protein influences ciliary length in neural progenitor cells. To this end, a comparison was made between the ciliary length of control-transfected cells and the naïve cilia of non-transfected cells on the opposite side of the same individual. Consequently, the ciliary length of the control-transfected cells was comparable to that of the naïve cilia (Supplementary Fig. 5A), suggesting that Arl13b-tdTomato did not affect ciliary length in neural progenitor cells. Next, to evaluate the roles of MAST4 and Tctex-1 in ciliary length in cells, we quantified the ciliary length between MAST4-shRNA- or Tctex-1-shRNA-transfected cells and control-transfected cells. We found that the cilia exhibited a significantly increased length in MAST4-shRNA- and Tctex-1-shRNA-transfected cells (Fig. 5B and C). In addition, the ciliary length of the MAST4-shRNA- and Tctex-1-shRNA-transfected cells exceeded that of the naïve cilia of each individual (Supplementary Fig. 5B–C).

A previous work showed that phospho-(T94)Tctex-1 shortens duration of the G₁ phase and accelerates S phase entry [6]. In this study, to investigate roles of MAST4 and Tctex-1 in proliferation of neural progenitor cells, the cells were knocked down with MAST4 or Tctex-1 and immunostained with anti-tubulin β3 antibody, a post-mitotic neuronal marker, and ant-mCherry antibody. Tubulin β3 was mainly distributed in cell body in neural progenitor cells. Although Arl13b-tdTomato strongly appeared in primary cilia, they are also distributed in cell body (Supplementary Fig. S2). Thus, we focused on tubulin β3⁺/tdToamto⁺ cells in the intermediate zone of the cortex. Consisting with the previous work 6, MAST4-KD, as well as Tctex-1-KD, significantly increased the rate of tubulin β3⁺ cells (Supplementary Fig. 6). This data emphasis the relevance of ciliary resorption and proliferation of neural progenitor cells.

Collectively, we develop a novel technique enabling the selective labeling of primary cilia and measuring their length in target molecule-knocked down neural progenitor cells.

4. Discussion

Measuring ciliary length is a critical experiment to elucidate the molecular mechanisms underlying ciliary resorption in cortical neural progenitor cells during embryonic development. In this study, we constructed a pCAGI-Arl13b-tdTomato plasmid that can downregulate target molecules using shRNA and label the cilia of successfully transfected cells. Using this plasmid, we showed that knockdown of both MAST4 and Tctex-1 elongated ciliary length in cortical neural progenitor cells.

This study shows that the pCAGI-Arl13b-tdTomato plasmid is a powerful tool for analyzing the role of target molecules in ciliary length in vivo, in which cells are densely populated, making it difficult to map cells to cilia in a one-to-one ratio. This plasmid possesses both shRNA and Arl13b-tdTomato sequences; thus, cells that have Arl13b-tdTomato⁺ cilia were considered as knockdown cells. One of the major advantages of this plasmid is that Arl13b-tdTomato itself does not disturb the ciliary length in cortical neural progenitor cells, as shown in Supplementary Fig. 5. In contrast, several studies have suggested that Arl13b can modify ciliary length in some cells [26,27], therefore, careful consideration is required when this plasmid is used in other cells. We speculate that Arl13b-tdTomato expression is controlled by the IRES in the pCAGI-Arl13b-tdTomato plasmid, and its expression level is not high enough to modify the ciliary length in neural progenitor cells. The second advantage is that shRNA sequences are easily inserted into the plasmid, and the efficiency of shRNAs has been validated in vivo using the pCAGIG plasmid in previous reports [5,6]. Hence, it is possible to prepare shRNAs against many different target molecules in a short period. Third, wild-type pregnant mice were subjected to IUE. DNA plasmids were injected into the lateral ventricle of the embryonic mouse brain and were transfected into cortical neural progenitor cells. Although specific equipment and mastery of its operation are required, this technique enables researchers to analyze the roles of many different target molecules in a short amount of time.

Suppression of primary ciliary resorption, leading to elongation, has also been observed in studies of some resorption-regulating molecules. Downregulation of MAST4, Tctex-1, and the molecules involved in the AurkA-mediated pathway, such as HEF1 and polo-like kinase 1, reversed serum-induced ciliary shortening in cultured cells [2,14,25,28]. The cerebral cortex from E12.5 of WDR62^−/− mice possesses longer cilia than that of wild-type mice, as well as serum-re-added mouse embryonic fibroblasts prepared from WDR62^−/− mice [17]. The results of the present study are consistent with those of previous studies, suggesting that MAST4 and Tctex-1 are responsible for ciliary resorption in cortical neural progenitor cells. Moreover, this study showed a possibility that they regulate cell proliferation and corticogenesis through primary ciliary resorption in neural progenitor cells in vivo during the embryonic stage.

The mechanisms of ciliary resorption can be investigated through use of the pCAGI-Arl13b-tdTomato plasmid in the developing cortical neural progenitor cells in the future. Although MAST4 and Tctex-1 were focused in this study, roles of other ciliary resorption-regulating molecules, including HEF1, AurkA, and HDAC6, should also be investigated. Since it is necessary to measure length of as many cilia as possible over time, visualizing a large number of cilia in each experiment is a requisite technique. Wholemount immunostaining of the embryonic brain and en-face view of apical cilia would be one of the efficient method [29].

In summary, the pCAGI-Arl13b-tdTomato plasmid is valuable for measuring ciliary length in cells in which target molecules are successfully knocked down, particularly in the developing cortical neural progenitor cells. Using this plasmid, we will be able to elucidate the molecular mechanisms underlying ciliary resorption in the cortical neural progenitor cells.

CRediT authorship contribution statement

Masaki Saito: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Visualization, Writing – original draft. Wataru Otsu: Investigation, Methodology, Writing – original draft. Kenichi Ishibashi: Funding acquisition, Validation, Writing – review and editing. Gen-ichi Atumi: Supervision: Funding acquisition, Writing – review and editing.

Funding

This research was funded by Grants-in-Aid for Scientific Research from the Japan Society for Promotion of Science (Nos. 21K06059 and 24K09369 to Masaki Saito; No. 23K10944 to Kenichi Ishibashi; No. 24K08814 to Gen-ichi Atsumi; No. 21K14999 to Wataru Otsu), ACRO Incubation Grants of Teikyo University (to Masaki Saito), and ACRO Team Research Grants of Teikyo University (to Gen-ichi Atsumi).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Glossary

ANOVA: analysis of variance; AurkA: Aurora kinase A; CAG: cytomegalovirus enhancer; CTS: ciliary targeting signal; DMEM: Dulbecco's modified Eagle medium; GFP: green fluorescent protein; HDAC6: histone deacetylase 6; HRP: horseradish peroxidase; IGF-1: insulin-like growth factor-1; IGF-1R: IGF-1 receptor; IUE: in utero electroporation; KO: knockout; MAST4: microtubule-associated serine/threonine protein kinase 4; NFDM: nonfat dry milk; mGFP: membrane GFP; mtdTomato: membrane-bound form of tdTomato; pCAGIG: pCAG-IRES-GFP; PCR: polymerase chain reaction; PFA: paraformaldehyde; RPE-1: retinal pigment epithelial cells; shRNA: short hairpin RNA; T94: threonine 94; Tctex-1: t-complex testis expressed-1.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Fig. S1

Primary cilia and basal body structure in neural progenitor cells. (A) A low-power field image of the E13.5 cortex. (B, C) Enlarged views of the boxed area of A and B, respectively. Primary cilia and basal bodies in the cryosection were immunolabeled with anti-Arl13b (red) and γ-tubulin (γ-Tub) (cyan) antibodies, respectively. The nuclei were stained with DAPI. Scale bars: 100 μm (A), 10 μm (B), and 2 μm (C).

Fig. S2

The ciliary membrane is not visualized by membrane-green fluorescent protein (GFP) or membrane-tdTomato. (A) A schematic diagram of a pCAGI-membrane-GFP (pCAGI-mGFP) plasmid. (B) A schematic diagram of a pCAGI-membrane-tdTomato (pCAGI-mtdTomato) plasmid. CAG: chicken β-actin promoter; IRES: internal ribosome entry site; mem: membrane-targeted site. The expression of mGFP, mtdTomato, and basal bodies in the cryosections was immunolabeled with anti-GFP (green), anti-RFP (red), and anti-γ-tubulin (γ-Tub) (cyan) antibodies, respectively. The nuclei were stained with DAPI (blue). (C) Low-power field images of the cortex co-transfected with pCAGI-mGFP and pCAGI-mtdTomato. (D and E) Enlarged views of the boxed area in C and D, respectively. Scale bars: 100 μm (C), 20 μm (D), and 2 μm (E).

Fig. S3

The original western blot images demonstrated in Fig. 4B. Whole blots probed with anti-Arl13b, anti-RFP, or anti-β-actin antibody was shown.

Fig. S4

Ciliary distribution of Arl13b-tdTomato in C3H10T1/2 cells. (A) C3H10T1/2 cells that were transfected with the pCAGI-Arl13b-tdTomato plasmid were starved of serum for 40 h. The expression of Arl13b-tdTomato, primary cilia, and basal bodies in the cells was immunolabeled with anti-mCherry (red), anti-acetylated α-tubulin (Ac-Tub) (green), and anti-γ-tubulin (γ-Tub) (cyan) antibodies, respectively. The nuclei were stained with DAPI (blue). (B) Enlarged views of (1) and (2) in A. Scale bars: 20 μm (A) and 2 μm (B).

Fig. S5

Comparison of ciliary length between transfected and non-transfected cells. The cortical neural progenitor cells were transfected with pCAGI-Arl13b-tdTomato (control) (A), pCAG-MAST4-sh-I-Arl13b-tdTomato (MAST4-sh) (B), or pCAG-Tctex-1-sh-I-Arl13b-tdTomato (Tctex-1-sh) (C) at E13.5. The cryosections were immunolabeled with anti-Arl13b and anti-mCherry antibodies. The nuclei were stained with DAPI. The ciliary length of both the tdTomato⁺/Arl13b⁺ cilia formed in the transfected cerebral cortex and the tdTomato⁻/Arl13b⁺ cilia formed in the non-transfected cerebral cortex were measured. ∗∗∗p < 0.001, Student's t-test. N.S.: not significant. The numbers of cilia measured were 134–294 (A), 137–140 (B), and 105–144 (C).

Fig. S6

Tubulin β3-positive cells in transfected cells. (A) Representative images of tubulin β3⁺ cells in transfected mouse brain slices. The cortical neural progenitor cells were transfected with pCAGI-Arl13b-tdTomato (control), pCAG-MAST4-sh-I-Arl13b-tdTomato (MAST4-sh), or pCAG-Tctex-1-sh-I-Arl13b-tdTomato (Tctex-1-sh) at E13.5. The cryosections were immunolabeled with anti-tubulin β3 (green) and anti-mCherry (red) antibodies. The nuclei were stained with DAPI (blue). Cell bodies of the tdTomato⁺ transfected cells were shown. (B) Quantification of the tubulin β3⁺/tdTomato⁺ cells. n = 3 (Control), 4 (MAST4-sh), 2 (Tctex-1-sh). Scale bars: 10 μm.

Data availability

Data will be made available on request.