Transforming Growth Factor-β1 and Bone Morphogenetic Protein-2 Inhibit Differentiation into Mature Ependymal Multiciliated Cells

Abstract

Ependymal cilia play pivotal roles in cerebrospinal fluid flow. In the primary culture system, undifferentiated glial cells differentiate well into ependymal multiciliated cells (MCCs) in the absence of fetal bovine serum (FBS). However, the substances included in FBS which inhibit this differentiation process have not been clarified yet. Here, we constructed the polarized primary culture system of ependymal cells using a permeable filter in which they retained ciliary movement. We found that transforming growth factor-β1 (TGF-β1) as well as Bone morphogenetic protein (BMP)-2 inhibited the differentiation with ciliary movement. The inhibition on the differentiation by FBS was recovered by the TGF-β1 and BMP-2 inhibitors in combination.

INTRODUCTION

Motile cilia in ependymal cells (on the ependyma lining of the brain ventricles) play pivotal roles in cerebrospinal fluid (CSF) flow in well-defined directions.^1–3) They have a so-called 9 + 2 structure in which two central microtubular singlets are surrounded by nine microtubular doublets with two rows of dynein arms containing the motor protein dynein.⁴⁾ The dynein arms walk on the back of the neighboring doublet, which enable the cilium to beat. Dysfunction of dynein heavy chain Mdnah5 in mice inhibits CSF flow, which results in closure of aqueduct and hydrocephalus during early postnatal brain development.⁵⁾ The defects of functional cilia in human cause primary ciliary dyskinesia (PCD), whose clinical symptoms include chronic rhinosinusitis, chronic bronchitis, infertility and hydrocephalus.⁶⁾ Beating of ependymal motile cilia is required not only for normal CSF flow but also directional migration of neuroblasts.^7–9)

Physiological regulation of mammalian ependymal motile cilia was studied in several experimental models in vitro; the brain slice preparation, isolated ependymal cells, and primary cultured cells.^10–14) Ciliary movement has been well studied in the non-polarized primary culture system using culture dishes or wells. However, there has been no reports for the ciliary movement in the polarized primary culture system.^14–16) Here we newly measured the ciliary movement in the development of ependymal multiciliated cells (MCCs) in the polarized primary culture system. It is widely accepted that fetal bovine serum (FBS) inhibits differentiation from undifferentiated glial cells into MCCs.^15,16) However, the substance(s) which inhibit this differentiation process into MCCs have not been fully understood yet. Here, we developed polarized primary culture system of ependymal MCCs using the Transwell permeable filter as described previously^17,18) and found that FBS in the upper chamber (in the apical medium) interferes with differentiation into ependymal MCCs. We narrowed down the candidates and finally identified the substance(s) in FBS which inhibit this differentiation process. Among many substances, transforming growth factor-β1 (TGF-β1), which is one of the major growth factors and cytokines with pleiotropic effects¹⁹⁾ found in FBS inhibited the differentiation step. Bone morphogenetic protein (BMP)-2, which was originally found as a regulator of cartilage and bone formation²⁰⁾ also inhibited the differentiation step at the higher concentration than that found in FBS. The inhibitory effect of FBS was not abolished by the addition of TGF-β1 inhibitor alone nor BMP-2 inhibitor alone. However, the TGF-β1 inhibitor in combination with BMP-2 inhibitor abolished the inhibitory effect of FBS, suggesting that TGF-β1 and BMP-2 are two major substances which inhibit the differentiation into MCCs.

MATERIALS AND METHODS

Animals

All animal experiments were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and the protocols were approved by the Animal Ethics Committees of Ritsumeikan University (BKC-2019-003, BKC-2021-022). The animals were cared for and the experiments were carried out according to the guidelines of this committee. Pregnant female (C57BL/6J) mice were purchased from Shimizu Experimental Animals (Kyoto, Japan). Mice were fed standard pellet food and water ad libitum.

Antibodies and Reagents

Anti-α-tubulin (acetyl K40) (AcTub) antibody (rabbit) (EPR16772) was purchased from Abcam (Cambridge, U.K.). Anti-FoxJ1 antibody (mouse) (2A5) and Alexa Fluor 594-conjugated goat anti-rabbit immunoglobulin G (IgG) secondary antibody (A21207) were from Thermo Fischer Scientific (Waltham, MA, U.S.A.).

Human recombinant TGF-β1, BMP-2, and LY364947 were purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). LDN-193189, a selective BMP signaling inhibitor was purchased from Selleck Biotech (Houston, TX, U.S.A.).

Culture

Primary cultured ependymal cells were prepared from newborn mice. Briefly, mice at postnatal day 0–1 were anesthetized by hypothermia, and the whole brains were removed from the skull and minced with scissors. Tissue fragments suspended in 1 mL of Hanks’ Balanced Salt Solution were incubated at 37 °C for 15 min in 0.25% trypsin (Nacalai Tesque, Kyoto, Japan) containing 0.1 mg/mL bovine pancreatic DNaseI (Sigma-Aldrich, St. Louis, MO, U.S.A.). Following centrifugation at 1000 rpm for 10 min, the supernatant was discarded, and the pellet was re-suspended in the proliferation medium; Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS, 5 µg/mL human insulin (FUJIFILM Wako Pure Chemical Corporation), 100 units/mL penicillin G and 100 μg/mL streptomycin (FUJIFILM Wako Pure Chemical Corporation). The cells were seeded on 75 cm² flasks and maintained at 37 °C in a humidified 5% CO₂ atmosphere. After reaching confluent, the cells were harvested and seeded in fibronectin-coated 24 well Transwell^® chamber insert. The upper and lower chambers were filled with the proliferation medium. After 3 d culture in the proliferation phase, the medium in the upper chamber was changed to differentiation medium; FBS-free MEM supplemented with 5 μg/mL fatty acid-free bovine serum albumin (BSA) (Sigma-Aldrich), 10 µg/mL transferrin (Sigma-Aldrich), 5 µg/mL insulin, 2 mM L-glutamine (FUJIFILM Wako Pure Chemical Corporation), 0.5 U/mL Thrombin (Sigma-Aldrich), 100 units/mL Penicillin G and 100 µg/mL Streptomycin. The media were changed twice a week.

Immunofluorescence Analysis and Cell Counting

Immunofluorescence analysis of the ependymal cells on the permeable filter was performed as described previously.²¹⁾ The cells on the filter were fixed in methanol at −30 °C for 10 min and washed three times with phosphate buffered saline (PBS) containing 10 mM glycine. The filter was blocked by pre-incubating in PBS containing 3% BSA at room temperature for 60 min. All antibody incubations were carried out using Can Get Signal Immunostain Solution A (Toyobo, Osaka, Japan). The filter was incubated overnight at 4 °C with anti-AcTub (1 : 100 dilution) antibody, followed by three washes with PBS containing 0.1% BSA. Alexa Fluor 594-conjugated goat anti-rabbit IgG secondary antibody (1 : 100 dilution) were incubated with the cells on the filter for 30 min at room temperature. After a final wash with PBS containing 0.1% BSA, the filters were mounted with fluorescent mounting medium (Funakoshi, H-1000, Japan). Nuclei were stained with 4′,6-diamidino-2-phenylindol (DAPI). The filters were examined using a confocal laser scanning microscope (FV-10i, Olympus, Tokyo, Japan). The number of MCCs was determined by counting the number of cells in the window (0.01 mm²) stained with the anti-AcTub antibody. Total number of cells was determined by counting the number of cells stained with DAPI.

Ultrafiltration of Proliferation Medium

The proliferation medium was fractionated by using ultrafiltration spin columns Vivaspin Turbo (Sartorius, Göttingen, Germany) with the threshold molecular mass being 10 kDa.

Measurements of Ciliary Beat Frequency (CBF)

The CBF values were measured as described previously.^22,23) The permeable support filter with the ependymal cells was cut into small pieces (squares of sides 2–5 mm). A cutted filter with the ependymal cells was placed on a coverslip pre-coated with Cell-Tak (Becton Dickinson Labware, Bedford, MA, U.S.A.), and was set in a micro-perfusion chamber (20 µL) mounted on an inverted light microscope (Eclipse Ti, Nikon, Tokyo, Japan) with a high-speed camera (FASTCAM-1024PCI, Photron Ltd., Tokyo, Japan). The stage of microscope was heated to 37 °C, and the video images were recorded for 2 s at 500 fps. Image J was used to measure CBF. The values of CBF were counted from the number of peaks of light intensity in 1 s.

Western Blot

Cells on the permeable filter of Transwell insert were washed with PBS, homogenized with the radio immunoprecipitation assay (RIPA) buffer (50 mM Tris–HCl, pH 7.6, 150 mM NaCl, 1% Nonidet-P40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS) with protease inhibitor cocktail), incubated at 4 °C for 20 min, and centrifuged at 16000 × g for 20 min at 4 °C. The supernatant was used as cell lysate. Proteins were separated by Laemmli’s SDS–polyacrylamide gel electrophoresis, and then transferred to a polyvinylidene difluoride membrane. The membrane was blocked for 1 h with 5% milk powder in TBST (10 mM Tris–HCl, pH 8.5, 150 mM NaCl, and 0.1% Tween 20), and exposed to primary antibodies diluted with Solution 1 (Can Get Signal^® Immunoreaction Enhancer Solution, TOYOBO) overnight at 4 °C. After rinsing with TBST, the membrane was incubated with secondary antibody diluted with Solution 2 (Can Get Signal^® Immunoreaction Enhancer Solution, TOYOBO) for 1 h at room temperature. After washing, antigen–antibody complexes on the membrane were visualized with a chemiluminescence system (ECL plus; GE Healthcare, Waukesha, WI, U.S.A.).

Statistical Analysis

Statistical analysis was performed by two tailed t test, chi-square test, and one-way ANOVA with Tukey’s post-hoc test. The confidence limit of p < 0.05 was considered to be statistically significant. The results are expressed as the means ± standard deviation (S.D.).

RESULTS

The Effect of FBS on Differentiation into Ependymal MCCs in the Polarized Primary Culture

In the present study we cultured proliferated progenitor cells on the Transwell permeable filter, providing different media to the apical and basal sides. It has been generally accepted that FBS inhibits differentiation from undifferentiated glial cells into ependymal MCCs in non-polarized culture conditions.^15,16) Here, we studied the effects of FBS on differentiation into ependymal MCCs in the polarized culture conditions using the permeable filter. We used two kinds of media in the differentiation phase as reported in Materials and Methods. One is the proliferation medium containing 10% FBS, and the other is the FBS-free differentiation medium. Culture was performed in three different conditions as shown in Fig. 1. One is the condition in which the differentiation medium (without FBS) was used as both the apical and basal media (termed −/− culture). Second is the condition in which the differentiation medium (without FBS) was used as the apical medium whereas the proliferation medium (with FBS) was used as the basal medium (termed −/+ culture). Third is the condition in which the proliferation medium (with FBS) was used as both the apical and basal media (termed +/+ culture).

Fig. 1. Schematic Pattern of Culture Conditions

Figure 2 shows the immunofluorescence patterns of AcTub in ependymal MCCs cultured in the different conditions (−/+ and +/+). The number of MCCs which were stained with the anti-AcTub antibody was counted, and the percentage of MCCs to the total number of cells was shown. In the −/− culture condition, the cells cannot be maintained stably until three weeks (data not shown). Therefore, the −/− culture condition was not considered hereafter. In the −/+ culture condition, the percentage of MCCs was progressively increased and reached 60 ± 14% on day 28. Unexpectedly, in the +/+ culture condition, the percentage of MCCs was also increased, and reached 29 ± 14% on day 28. The number of MCCs in the −/+ culture condition was significantly higher than that in the +/+ culture condition. It should be noted that the total number of cells in the window stained with DAPI was not different between the −/+ and +/+ culture conditions (Supplementary Table S1). These results suggest that FBS in the upper chamber inhibited the differentiation into ependymal MCCs although undifferentiated glial cells can differentiate into ependymal MCCs to a certain extent even in the presence of apical FBS.

Fig. 2. Immunofluorescence of Ependymal MCCs Cultured in the −/+ and +/+ Conditions with an Anti-AcTub Antibody

(Left) Cells were cultured in the −/+ and +/+ conditions for indicated period (7–28 d). Cilia were stained with the anti-AcTub antibody (red). Nuclei were stained with DAPI (blue). Scale bars represent 10 μm. Representative patterns of 3 experiments (culture) are shown. (Right) The percentages of MCCs to the total cell number in the −/+ (○) and +/+ culture (●) conditions are shown. Data are presented as means ± S.D. (n = 50 windows in 3 cultures). * p < 0.05 vs. −/+ culture.

The Effect of FBS on Ciliary Movement

In order to study the effects of FBS on ciliary movement, at first, we counted the number of MCCs with beating cilia under microscope, and their percentages to the total number of MCCs were calculated. The percentage values of MCCs with beating cilia in the −/+ and +/+ culture conditions on day 28 were 87 and 72%, respectively (Fig. 3A). Next, we compared the ciliary movement and measured their CBF values in the −/+ and +/+ culture conditions. Figure 3B shows the distribution of CBF values in the −/+ and +/+ culture conditions. The peak CBF values of MCCs in the −/+ and +/+ culture conditions on day 28 were 15 and 9 Hz, respectively. In the previous studies, the mean CBF values were reported to range between 10.7 and 20.1 Hz.^10,11) Here, we defined the cells showing their CBF values higher than 15 Hz as mature MCCs. The percentage of mature MCCs on day 28 in the −/+ and +/+ culture conditions was 55 and 9%, respectively. Average values of CBF in the −/+ culture conditions (14.3 ± 4.0 Hz) was significantly higher than that in the +/+ culture conditions (9.1 ± 3.9 Hz) (Fig. 3C). These results suggest that FBS in the upper chamber inhibited the differentiation into mature ependymal MCCs which show ciliary movement.

Fig. 3. Effects of FBS on Ciliary Movement

(A) The percentages of MCCs showing ciliary beating to the total number of MCC (shown by red parts) were compared on day 28 between the −/+ and +/+ culture conditions. Total numbers of cells are 511 and 481 in the −/+ and +/+ conditions, respectively (4 independent culture). (B) The distribution of CBF values of MCCs on day 28 was compared between the −/+ and +/+ culture conditions. (C) The CBF values of MCCs on day 28 was compared between the −/+ and +/+ culture conditions. Total numbers of cells are 256 and 302 in the −/+ and +/+ conditions, respectively (4 independent culture). Data are presented as means ± S.D. * p < 0.05 vs. −/+ culture.

Identification of the Substances in FBS Which Inhibit Differentiation into Ependymal MCCs

From the above-mentioned results, it is likely that FBS contains substances which inhibit differentiation into mature ependymal MCCs. Therefore, we tried to identify such substance in FBS. At first, to narrow down the candidates, the proliferation medium was divided into two preparations with the molecular mass lower and higher than 10 kDa (termed Low-10 and High-10 fractions, respectively) by ultrafiltration. Then, the Low-10 or High-10 fraction was added to the apical chamber of the −/+ culture condition to study the effects of these fractions on the differentiation into ependymal MCCs. The percentage of MCCs in the presence of Low-10 fraction on day 28 was 46 ± 21% (Fig. 4). On the other hand, the percentage of MCCs in the presence of High-10 fraction on day 28 was 33 ± 14%, which is almost comparable to that cultured in the +/+ condition (29 ± 14%). These results suggest that substances with the molecular mass higher than 10 kDa are mainly involved in the inhibition of differentiation into MCCs.

Fig. 4. Immunofluorescence of Ependymal MCCs Cultured in the Presence of Low-10 or High-10 Fraction with an Anti-AcTub Antibody

(Left) Cells were cultured in the presence of Low-10 or High-10 fraction in the −/+ condition for indicated period (7–28 d). Cilia were stained with the anti-AcTub antibody (red). Nuclei were stained with DAPI (blue). Scale bars represent 10 μm. Representative patterns of 2 experiments are shown. Immunofluorescence patterns of MCCs in the −/+ and +/+ culture conditions shown in Fig. 2 were presented for comparison. (Right) The percentages of MCCs to the total cell number in the −/+ culture condition in the presence of Low-10 (□) or High-10 (■) are shown. The percentages of MCCs to the total cell number in the −/+ (○) and +/+ culture (●) conditions shown in Fig. 2 were presented for comparison. Data are presented as means ± S.D. (n = 50 windows in 2 culture). * p < 0.05 vs. −/+ culture. ^† p < 0.05 vs. +/+ culture.

The Effects of TGF-β1 and BMP-2 on the Differentiation into Ependymal MCCs

Based on the molecular mass of target compounds shown above, in combination with the proteomics data about growth factors included in FBS reported previously,²⁴⁾ we preliminarily checked several growth factors whether they inhibit the differentiation into MCCs from the apical side of −/+ culture. Among them, we focused on TGF-β1 with a molecular mass of 25 kDa (homodimer as an active form) as one of the candidates of such inhibitory compounds, and studied its effects on differentiation into ependymal MCCs. The concentration of TGF-β1 in FBS was reported to be 10–20 ng/mL,²⁵⁾ and its concentration in human serum is around 10 ng/mL.²⁶⁾ Here, 1 ng/mL TGF-β1 (estimated concentration included in 10% FBS) was added to the apical chamber of the −/+ culture condition. The percentage value of MCCs in the presence of 1 ng/mL TGF-β1 on day 28 was 29 ± 16%, which is almost comparable to that cultured in the +/+ condition (29 ± 14%) (Fig. 5). We also studied the effects of LY364947, a potent ATP-competitive inhibitor of TGF-β receptor,²⁷⁾ on the differentiation into ependymal MCCs. LY364947 (10 μM) almost completely abolished the inhibitory effect of TGF-β1 on the differentiation into ependymal MCCs with the percentage value of MCC being 57 ± 15% as shown in Fig. 5. These results suggest that TGF-β1 inhibits differentiation from undifferentiated glial cells into ependymal MCCs.

Fig. 5. Immunofluorescence of Ependymal MCCs Cultured in the Presence of TGF-β1 with an Anti-AcTub Antibody

(Left) Cells were cultured in the presence of 1 ng/mL TGF-β1 and in the presence or absence of 10 μM LY364947 for indicated period (7–28 d). Cilia were stained with the anti-AcTub antibody (red). Nuclei were stained with DAPI (blue). Scale bars represent 10 μm. Representative patterns of 3 experiments are shown. (Right) The percentages of MCCs to the total cell number in the presence of 1 ng/mL TGF-β1 and in the presence (○) or absence of 10 μM LY364947 (●) in the −/+ culture condition are shown. Data are presented as means ± S.D. (n = 40 windows in 3 culture). * p < 0.05 vs. (in the absence LY364947).

Recently, Hiraoka et al. reported that BMP-2 and BMP-4 with molecular mass of 26 and 34 kDa, respectively (homodimer as active forms) at the concentration of 50 ng/mL or higher inhibit ciliogenesis of ependymal MCCs.²⁸⁾ Omiya et al. reported that BMP signaling including BMP-2 and BMP-6 suppressed the differentiation of neural stem-progenitor cells into ependymal cells through down-regulation of GemC1, Mcidas and Foxj1 which are key transcriptional regulators of ependymal specification and differentiation.²⁹⁾ We also tried to study the effect of BMP-2 on differentiation into ependymal MCCs. BMP-2 was added to the apical chamber of the −/+ culture condition. The concentrations of BMP-2 in FBS and human serum were reported to be 617 and 125 pg/mL, respectively.³⁰⁾ Therefore, we studied the effect of BMP-2 at the concentration of 60 pg/mL (as estimated concentration included in 10% FBS) on differentiation into ependymal MCCs in the −/+ condition. The percentage of MCCs in the presence of 60 pg/mL BMP-2 on day 28 was 46 ± 11% (Fig. 6A), which was slightly (but not significantly) lower than that cultured in the −/+ culture condition (60 ± 14%), and higher than that cultured in the +/+ condition (29 ± 14%). The percentages of MCCs in the presence of 1, 10, and 100 ng/mL BMP-2 on day 28 were 27 ± 17, 20 ± 10, and 19 ± 10%, respectively (Fig. 6A), which were significantly lower than that cultured in the −/+ condition (60 ± 14%), and similar or lower than that cultured in the +/+ condition (29 ± 14%). These results suggest that BMP-2 inhibits the differentiation in a concentration-dependent manner and that BMP-2 at the concentration included in 10% FBS slightly inhibits differentiation into ependymal MCCs (Fig. 6A). We also studied the effects of LDN-193189, a selective BMP signaling inhibitor on the differentiation into ependymal MCC.^31,32) LDN-193189 (100 nM) almost completely abolished the inhibitory effect of 100 ng/mL BMP-2 on the differentiation into ependymal MCCs as shown in Fig. 6B.

Fig. 6. Immunofluorescence of Ependymal MCCs Cultured in the Presence of BMP-2 with an Anti-AcTub Antibody

(A) (Left) Cells were cultured in the absence or presence of 60 pg/mL, 1 or 100 ng/mL BMP-2 in the −/+ culture condition for indicated period (7–28 d). Cilia were stained with the anti-AcTub antibody (red). Nuclei were stained with DAPI (blue). Scale bars represent 10 μm. Representative patterns of 3 experiments (culture) are shown. (Right) The percentages of MCCs to the total cell number in the absence (○) or presence of 60 pg/mL (□), 1 ng/mL (■) or 100 ng/mL BMP-2 (●) in the −/+ culture condition are shown. Data are presented as means ± S.D. (n = 40 windows in 3 culture). * p < 0.05 vs. −/+ culture. (B) (Left) Cells were cultured in the presence of 100 ng/mL BMP-2 and in the presence or absence of 100 nM LDN-193189 in the −/+ culture condition for indicated period (7–28 d). Cilia were stained with the anti-AcTub antibody (red). Nuclei were stained with DAPI (blue). Scale bars represent 10 μm. Representative patterns of 2 experiments (culture) are shown. (Right) The percentages of MCCs to the total cell number in the presence of 100 ng/mL BMP-2 and in the presence (○) or absence of 100 nM LDN-193189 (●) in the −/+ culture condition are shown. Data are presented as means ± S.D. (n = 40 windows in 3 culture). * p < 0.05 vs. (in the absence LDN-193189).

Next, we studied the effects of LY364947 and LDN-193189 on the inhibition of differentiation by FBS. The percentage of MCCs in the presence of LY364947 in the +/+ culture condition on day 28 was 34 ± 16%, which is almost comparable to that cultured in the +/+ condition (29 ± 14%) (Fig. 7). Therefore, LY364947 alone did not recover from the inhibitory effect of FBS on the differentiation into ependymal MCCs. The percentage of MCCs in the presence of LDN-193189 in the +/+ culture condition on day 28 was 24 ± 16%, which is also almost comparable to that cultured in the +/+ condition (29 ± 14%) (Fig. 7). Therefore, LDN-193189 alone did not recover from the inhibitory effect of FBS on the differentiation into ependymal MCCs.

Fig. 7. Immunofluorescence of Ependymal MCCs Cultured in the Presence of LY364947 and LDN-193189 with an Anti-AcTub Antibody

(Left) Cells were cultured in the +/+ culture condition in the presence of 10 μM LY364947, 100 nM LDN-193189 or their combination for indicated period (7–28 d). Cilia were stained with the anti-AcTub antibody (red). Nuclei were stained with DAPI (blue). Scale bars represent 10 μm. Representative patterns of 2 experiments (culture) are shown. (Right) The percentages of MCCs to total cell number in the +/+ culture condition in the absence (●) or presence of 10 μM LY364947 (■), 100 nM LDN-193189 (□) or their combination (○) are shown. Data are presented as means ± S.D. (n = 40 windows in 2 culture). * p < 0.05 vs. +/+ culture.

Finally, we studied the effects of LY364947 and LDN-193189 in combination on the inhibition of differentiation by FBS. The percentage of MCCs in the presence of LY364947 and LDN-193189 in the +/+ culture condition on day 28 was 51 ± 16% (Fig. 7), which is significantly higher than that cultured in the +/+ condition (29 ± 14%), and close to that cultured in the −/+ condition (60 ± 14%). These results altogether suggest that TGF-β1 as well as BMP-2 are two major molecules which inhibit differentiation into ependymal MCCs.

The Effects of TGF-β1 and BMP-2 on Ciliary Movement

In order to study the effects of TGF-β1 and BMP-2 on ciliary movement, we counted the number of MCCs with beating cilia under microscope. The percentages of MCCs with beating cilia in the absence or presence of 1 ng/mL TGF-β1 and 100 ng/mL BMP-2 on day 28 were 87, 65, and 16%, respectively (Fig. 8A). Next, we compared the ciliary movement and measured their CBF values in the presence of 1 ng/mL TGF-β1 or 100 ng/mL BMP-2 on day 28, respectively. The percentage of mature MCCs showing CBF values higher than 15 Hz on day 28 was 55% in the −/+ culture condition. Conversely, in the presence of TGF-β1 or BMP-2, mature MCCs were not observed (Fig. 8B). Average values of CBF in the presence of TGF-β1 or BMP-2 (4.3 ± 2.3 and 7.9 ± 3.0 Hz, respectively) were significantly lower than that in the −/+ culture conditions (14.3 ± 4.03 Hz) (Fig. 8C). Therefore, TGF-β1 and BMP-2 seem to inhibit the development of functional cilia although the cells contain cilia stained with an anti-AcTub antibody. It should be mentioned that TGF-β1 and BMP-2 showed stronger inhibitory effects on ciliary movement than FBS.

Fig. 8. Effects of TGF-β1 and BMP-2 on Ciliary Movement

(A) The percentages of MCCs showing ciliary beating to the total number of MCC (shown by red parts) cultured in the absence and presence of 1 ng/mL TGF-β1 or 100 ng/mL BMP-2 for 28 d are presented. Total numbers of cells are 375, 403, and 511 in the culture with TGF-β1, BMP-2 and in the control (−/+), respectively (4 independent culture). (B) The distribution of CBF values of MCCs cultured in the absence or presence of 1 ng/mL TGF-β1 or 100 ng/mL BMP-2 on day 28. (C) The CBF values of MCCs on day 28 were compared between the culture in TGF-β1 and 100 ng/mL BMP-2 and in the control (−/+) conditions. Data are presented as means ± S.D. Total numbers of cells cultured in TGF-β1 and BMP-2 and in the control (−/+) are 266, 62, and 256, respectively (2 independent culture). * p < 0.05 vs. −/+ condition.

Finally, we studied the effects of LY364947 and LDN-193189 on the inhibition of ciliary movement by FBS. Figure 9A shows the distribution of CBF values of ciliary movement in the presence or absence of LY364947 or LDN-193189 alone, and their combination in the +/+ culture conditions. The percentage of mature MCCs in the presence of LY364947 or LDN-193189 in the +/+ culture conditions was 0 and 4%, respectively. Conversely, the percentage of mature MCCs in the presence of LY364947 and LDN-193189 in combination was recovered to 41% (Fig. 9A), which was higher than the percentage in the absence of these inhibitors. Average values of CBF in the presence of LY364947 or LDN-193189 (6.5 ± 2.6 and 8.4 ± 2.4 Hz, respectively) were significantly lower than that in the −/+ culture conditions (14.3 ± 4.0 Hz) (Fig. 9B). However, average value of CBF in the presence of LY364947 and LDN-193189 in combination (13.3 ± 4.5 Hz) was comparable with that in the −/+ culture conditions (Fig. 9B). These results suggest that TGF-β1 and BMP-2 in combination inhibit the development of mature ependymal cilia.

Fig. 9. (A) The Distribution of CBF Values Was Studied in the +/+ Culture Conditions in the Absence (+/+) or Presence of 10 μM LY364947, 100 nM LDN-193189 Alone or Their Combination

The distribution of CBF values in the −/+ culture conditions was presented for comparison. (B) The CBF values of MCCs on day 28 were compared between the culture in TGF-β1, 100 ng/mL BMP-2 alone or in combination and in the control (−/+) and (+/+) conditions. Data are presented as means ± S.D. Total numbers of cells cultured in the −/+ and +/+ conditions, and in the presence of LY364947, LDN-193189 alone or their combination are 256, 302, 100, 100, and 203 cells, respectively in (A) and (B). * p < 0.05 vs. −/+ condition, and ^† p < 0.05 vs. +/+ condition.

It should be noted that TGF-β1 and BMP-2 also have acute effects. They decreased the CBF values of ependymal MCCs cultured in the −/+ condition by 37 and 33% within 20 min incubation as shown in Fig. 10. These results suggest that the inhibitory effects of TGF-β1 and BMP-2 on ciliary movement shown in Figs. 8A–C are partly due to their acute inhibitory effects.

Fig. 10. Acute Effects of TGF-β1 or BMP-2 on the CBF Values of Ciliary Cells Ependymal MCCs Cultured in the −/+ Culture Conditions for 28 d Were Incubated with 1 ng/mL TGF-β1 or 100 ng/mL BMP-2 for 5, 10, 15, and 20 min

The CBF values were measured, and presented as the percentage of control value of CBF measured before the treatment. Control CBF value in the absence of TGF-β1 or BMP-2 was 18.6 ± 1.7 Hz (n = 3). Data are presented as means ± S.D.

The Expression of Foxj1

Foxj1 is a master transcriptional factor which regulates differentiation program of motile cilium.^33,34) The expression of Foxj1 mRNA was decreased by BMP-2 through down-regulation of its upstream regulator, GemC1,²⁹⁾ whereas the effect of TGF-β1 on the expression of Foxj1 is controversial. Here, we studied the expression of Foxj1 protein during the development into MCCs in the Transwell culture by Western blot (Fig. 11). The expression of Foxj1 protein was observed before switching to the differentiation medium (on day 0). The expression level of Foxj1 increased on days 7 and 14 in the −/+ culture condition. Conversely, the expression level of Foxj1 was not up-regulated in the +/+ culture condition. TGF-β1 and BMP-2 also down-regulated Foxj1 expression. These results suggest that the inhibitory effects of FBS, TGF-β1 and BMP-2 on the differentiation into MCCs are partly due to the decreased expression of Foxj1.

Fig. 11. (A) Western Blots of Lysate of Ependymal MCCs with an Anti-foxj1 Antibody

Cells were cultured in the −/+ and conditions. Cells were also cultured in the −/+ condition in the presence of 1 ng/mL TGF-β1 or 100 ng/mL BMP-2 for 7 or 14 d. These cell lysates were blotted with the anti-Foxj1 antibody. Western blot of GAPDH as a housekeeping protein was also shown. Representative patterns of 3 experiments are shown. (B) Densitometric analysis of Western blot was performed. Data are presented as means ± S.D. (n = 3). * p < 0.05 vs. −/+ condition on Day 14.

DISCUSSION

Here, we cultured undifferentiated glial cells prepared from a newborn mouse brain on the permeable filter (Transwell) as reported previously¹⁷⁾ which can provide different media to the apical and basal sides of cell monolayer. In this study, two kinds of media were used as shown in Materials and Methods in detail. In short, one is the proliferation medium with 10% FBS, and the other is the differentiation medium containing 5 μg/mL albumin instead of FBS. In fact, under physiological conditions, protein concentration of normal CSF is very low (100–400 μg/mL), and the main protein in the CSF is albumin, which occupies 50–70% of the total protein.³⁵⁾ Embryonic CSF contains many kinds of growth factors involved in brain development.³⁶⁾ In non-polarized primary culture system of undifferentiated glial cells, FBS-free medium is strongly recommended to successfully generate high percentages of ependymal MCCs.^15,16) In the present study, two types of culture systems were designed which were termed −/+ and +/+ cultures as described in Results (Fig. 1). We found that the differentiation medium without FBS is suitable as the apical medium for the differentiation into MCCs, and that FBS in the apical medium inhibits the differentiation into ependymal MCCs (Fig. 2). Next, we tried to identify the substances in FBS which inhibit this differentiation process, and found TGF-β1 as one of such molecules. TGF-β1 inhibited the differentiation into mature MCCs at the concentrations comparable with those found in 10% FBS and human serum (1 ng/mL) (Fig. 5). The concentration of TGF-β1 in CSF under the physiological conditions was reported to be 40 pg/mL,³⁷⁾ which is lower than that used in this study (1 ng/mL). However, TGF-β1 concentration in CSF is possibly increased when ependymal ventricular wall (accompanying bleeding) is impaired by herpes simplex virus (HSV) infection.¹⁶⁾ It is also interesting that TGF-β1 in the CNS and brain was up-regulated following brain injury.³⁸⁾ However, the effects of TGF-β1 on the differentiation into MCCs have not been reported. It should be noted that many neuropathological alterations including severe hydrocephalus were observed in transgenic mice overexpressing TGF-β1³⁹⁾ because dysfunction of ependymal cilia is directly linked to hydrocephalus.

BMP-2 at the concentration of 50 to 100 ng/mL inhibited ciliogenesis during ependymal cell differentiation in non-polarized primary cell culture as reported previously.²⁸⁾ We also found that BMP-2 (100 ng/mL) fully inhibited the differentiation into ependymal MCCs as shown in Fig. 6A. However, it did not well inhibit the differentiation at the concentrations comparable with those in 10% FBS and human serum (60 pg/mL) as shown in Fig. 6A. BMP-2 was reported to be undetectable in CSF.⁴⁰⁾ Therefore, BMP-2 concentration used in this study (60 pg/mL to 100 ng/mL) are much higher than that observed in CSF under the physiological conditions. However, the concentration of BMP-2 is possibly increased to a certain extent when ependymal ventricular wall (accompanying bleeding) is impaired by HSV infection. It should be noted that BMP-2 decreased the number of MCCs with ciliary movement (Figs. 8A–C) and that BMP-2 down-regulated the expression of Foxj1 (Fig. 11).

Moreover, here we newly reported that LY364947 (a potent ATP-competitive inhibitor of TGF-β receptor)²⁷⁾ or LDN-193189 (a selective BMP signaling inhibitor)^31,32) alone did not recover the inhibitory effect of FBS (Fig. 7). However, these two inhibitors in combination recovered the inhibitory effect of FBS (Fig. 7). These results further suggest that TGF-β1 and BMP-2 may synergically inhibit the differentiation into MCCs. We tentatively hypothesize that some substance in FBS (except for TGF-β1) may stimulate BMP-2 secretion from neighboring glial cells or ependymal cells (by paracrine or autocrine manner) in the culture. In fact, BMP-2 is widely expressed in central nervous system including neurons and astrocytes.⁴¹⁾ As a consequence, BMP-2 secreted in this way as well as TGF-β1 included in FBS may inhibit differentiation into mature ependymal MCCs as shown in Fig. 12. Therefore, inhibitors of BMP-2 and TGF-β1 in combination are necessary to recover from the inhibitory effects of FBS. It should be necessary to measure the BMP-2 concentrations in the medium in the presence of TGF-β1. Both TGF-β1 and BMP-2 down-regulated the expression of Foxj1 protein. However, the signaling pathway of TGF-β1 between Smad activation and Foxj1 down-regulation is unclear at present. It cannot be completely excluded that other molecules in FBS also inhibit the differentiation.

Fig. 12. Working Hypothesis about the Reaction Process of TGF-β1 and BMP-2 Leading to Inhibition of Differentiation into Mature Ependymal MCCs

The signaling mechanisms of the TGF-β1 was reviewed by Tzavlaki and Moustakas previously.⁴³⁾ TGF-β1 included in FBS binds to the TGFβR, leading to activation of Smad2/3. The activated Smad2/3 associates with Smad4, and translocates to the nucleus to inhibit the expression of transcriptional factors involved in ciliogenesis. BMP-2 seems to be secreted from the neighboring cells, and binds to the BMPR, leading to activation of Smad1/5/8. The activated Smad1/5/8 associates with Smad4, and translocates to the nucleus to inhibit the expression of GemC1, which is a transcriptional factor located upstream of Mcidas and Foxj1 as reported previously.²⁹⁾

Here, we found that TGF-β1 and BMP-2 decreased the number of mature MCCs with ciliary movement. Tözser et al. reported that TGF-β signaling regulates the function of the transition zone (a specialized ciliary domain present at the base of the cilium) which affects the cilia length in Xenopus neural and epidermal cilia,⁴²⁾ rather than down-regulation of Foxj1. We need to study the ultrastructure of cilia of TGF-β1 and BMP-2-treated cells in the future.

In conclusion, we identified TGF-β1 as one of the major inhibitory factors involved in FBS for the differentiation into mature ependymal MCCs. This is the first report that TGF-β1 and BMP-2 inhibit the differentiation into mature ependymal MCCs by down-regulation of Foxj1.

Acknowledgments

This research was supported in part by Fellowship from Takeda Science Foundation (to SA) and MEXT-supported Program for the Strategic Research Foundation at Private Universities to SA.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1) Del Bigio MR. Ependymal cells: biology and pathology. Acta Neuropathol., 119, 55–73 (2010).
• 2) Lee L. Riding the wave of ependymal cilia: genetic susceptibility to hydrocephalus in primary ciliary dyskinesia. J. Neurosci. Res., 91, 1117–1132 (2013).
• 3) Kumar V, Umair Z, Kumar S, Goutam RS, Park S, Kim J. The regulatory roles of motile cilia in CSF circulation and hydrocephalus. Fluids Barriers CNS., 18, 31 (2021).
• 4) Choksi SP, Lauter G, Swoboda P, Roy S. Switching on cilia: transcriptional networks regulating ciliogenesis. Development, 141, 1427–1441 (2014).
• 5) Ibañez-Tallon I, Pagenstecher A, Fliegauf M, Olbrich H, Kispert A, Ketelsen U-P, North A, Heintz N, Omran H. Dysfunction of axonemal dynein heavy chain Mdnah5 inhibits ependymal flow and reveals a novel mechanism for hydrocephalus formation. Hum. Mol. Genet., 13, 2133–2141 (2004).
• 6) Afzelius BA. Cilia-related diseases. J. Pathol., 204, 470–477 (2004).
• 7) Sawamoto K, Wichterle H, Gonzalez-Perez O, Cholfin JA, Yamada M, Spassky N, Murcia NS, Garcia-Verdugo JM, Marin O, Rubenstein JLR, Tessier-Lavigne M, Okano H, Alvarez-Buylla A. New neurons follow the flow of cerebrospinal fluid in the adult brain. Science, 311, 629–632 (2006).
• 8) Huang S, Hirota Y, Sawamoto K. Various facets of vertebrate cilia: motility, signaling, and role in adult neurogenesis. Proc. Jpn. Acad., Ser. B, Phys. Biol. Sci., 85, 324–336 (2009).
• 9) Lehtinen MK, Zappaterra MW, Chen X, Yang YJ, Hill AD, Lun M, Maynard T, Gonzalez D, Kim S, Ye P, D’Ercole AJ, Wong ET, LaMantia AS, Walsh CA. The cerebrospinal fluid provides a proliferative niche for neural progenitor cells. Neuron, 69, 893–905 (2011).
• 10) Lechtreck K-F, Delmotte P, Robinson ML, Sanderson MJ, Witman GB. Mutations in Hydin impair ciliary motility in mice. J. Cell Biol., 180, 633–643 (2008).
• 11) Lechtreck K-F, Sanderson MJ, Witman GB. High-speed digital imaging of ependymal cilia in the murine brain. Methods Cell Biol., 91, 255–264 (2009).
• 12) Hirst RA, Sikand KS, Rutman A, Mitchell TJ, Andrew PW, O’Callaghan C. Relative roles of pneumolysin and hydrogen peroxide from Streptococcus pnenmoniae in inhibition of ependymal ciliary beat frequency. Infect. Immun., 68, 1557–1562 (2000).
• 13) Wilson GR, Wang HX, Egan GF, Robinson PJ, Delatycki MB, O’Bryan MK, Lockhart PJ. Deletion of the Parkin co-regulated gene causes defects in ependymal ciliary motility and hydrocephalus in the quaking^viable mutant mouse. Hum. Mol. Genet., 19, 1593–1602 (2010).
• 14) El Zein L, Ait-Lounis A, Morlé L, Thomas J, Chhin B, Spassky N, Reith W, Durand B. RFX3 governs growth and beating efficiency of motile cilia in mouse and controls the expression of genes involved in human ciliopathies. J. Cell Sci., 122, 3180–3189 (2009).
• 15) Weibel M, Pettmann B, Artault JC, Sensenbrenner M, Labourdette G. Primary culture of rat ependymal cells in serum-free defined medium. Brain Res., 390, 199–209 (1986). https://pubmed.ncbi.nlm.nih.gov/3955370/
• 16) Abdi K, Lai C-H, Paez-Gonzalez P, Lay M, Pyun J, Kuo CT. Uncovering inherent cellular plasticity of multiciliated ependyma leading to ventricular wall transformation and hydrocephalus. Nat. Commun., 9, 1655 (2018).
• 17) Laabich A, Sensenbrenner M, Delaunoy JP. Monolayer cultures of ependymal cells on porous bottom dishes. A tool for transport studies across the brain cerebrospinal barrier. Neurosci. Lett., 103, 157–161 (1989).
• 18) Gabrion JB, Herbuté S, Bouillé C, Maurel D, Kuchler-Bopp S, Laabich A, Delaunoy J-P. Ependymal and choroidal cells in culture: characterization and functional differentiation. Microsc. Res. Tech., 41, 124–157 (1998).
• 19) Morikawa M, Derynck R, Miyazono K. TGF-β and the TGF-β family: context-dependent roles in cell and tissue physiology. Cold Spring Harb. Perspect. Biol., 8, a021873 (2016).
• 20) Wozney JM, Rosen V, Celeste AJ, Mitsock LM, Whitters MJ, Kriz RW, Hewick RM, Wang EA. Novel regulators of bone formation: molecular clones and activities. Science, 242, 1528–1534 (1988).
• 21) Gillespie JL, Anyah A, Taylor JM, Marlin JW, Taylor TAH. A Versatile method for immunofluorescent staining of cells cultured on permeable membrane inserts. Med. Sci. Monit. Basic Res., 22, 91–94 (2016). https://basic.medscimonit.com/abstract/index/idArt/900656
• 22) Komatani−Tamiya N, Daikoku E, Takemura Y, Shimamoto C, Nakano T, Iwasaki Y, Kohda Y, Matsumura H, Marunaka Y, Nakahari T. Procaterol-stimulated increases in ciliary bend amplitude and ciliary beat frequency in mouse bronchioles. Cell. Physiol. Biochem., 29, 511–522 (2012).
• 23) Inui T, Murakami K, Yasuda M, Hirano S, Ikeuchi Y, Kogiso H, Hosogi S, Inui T, Marunaka Y, Nakahari T. Ciliary beating amplitude controlled by intracellular Cl⁻ and a high rate of CO₂ production in ciliated human nasal epithelial cells. Pflügers Arch., 471, 1127–1142 (2019).
• 24) Zheng X, Baker H, Hancock WS, Fawaz F, McCaman M, Pungor E Jr. Proteomics analysis for the assessment of different lots of fetal bovine serum as a raw material for cell culture. Part IV. Application of proteomics to the manufacture of biological drugs. Biotechnol. Prog., 22, 1294–1300 (2006).
• 25) Oida T, Weiner HL. Depletion of TGF-β from fetal bovine serum. J. Immunol. Methods, 362, 195–198 (2010).
• 26) Ciprandi G, De Amici M, Negrini S, Marseglia G, Tosca MA. TGF-β and IL-17 serum levels and specific immunotherapy. Int. Immunopharmacol., 9, 1247–1249 (2009).
• 27) Peng S-B, Yan L, Xia X, Watkins SA, Brooks HB, Beight D, Herron DK, Jones ML, Lampe JW, McMillen WT, Mort N, Sawyer JS, Yingling JM. Kinetic characterization of novel pyrazole TGF-β receptor I kinase inhibitors and their blockade of the epithelial mesenchymal transition. Biochemistry, 44, 2293–2304 (2005).
• 28) Hiraoka K, Inada H, Yanai K, Osumi N. Bone morphogenetic proteins inhibit ciliogenesis of ependymal cells in vitro. Tohoku J. Exp. Med., 252, 199–208 (2020).
• 29) Omiya H, Yamaguchi S, Watanabe T, Kuniya T, Harada Y, Kawaguchi D, Gotoh Y. BMP signaling suppresses Gemc1 expression and ependymal differentiation of mouse telencephalic progenitors. Sci. Rep., 11, 613 (2021).
• 30) Hu J-G, Lü H-Z, Wang Y-X, Bao M-S, Zhao B-M, Zhou J-S. BMP Signaling mediates astrocyte differentiation of oligodendrocyte progenitor cells. Tohoku J. Exp. Med., 222, 195–200 (2010).
• 31) Cuny GD, Yu PB, Laha JK, Xing X, Liu J-F, Lai CS, Deng DY, Sachidanandan C, Bloch KD, Peterson RT. Structure–activity relationship study of bone morphogenetic protein (BMP) signaling inhibitors. Bioorg. Med. Chem. Lett., 18, 4388–4392 (2008).
• 32) Yu PB, Deng DY, Lai CS, Hong CC, Cuny GD, Bouxsein ML, Hong DW, McManus PM, Katagiri T, Sachidanandan C, Kamiya N, Fukuda T, Mishina Y, Peterson RT, Bloch KD. BMP type I receptor inhibition reduces heterotopic ossification. Nat. Med., 14, 1363–1369 (2008).
• 33) Yu X, Ng CP, Habacher H, Roy S. Foxj1 transcription factors are master regulators of the motile ciliogenic program. Nat. Genet., 40, 1445–1453 (2008).
• 34) Spassky N, Meunier A. The development and functions of multiciliated epithelia. Nat. Rev. Mol. Cell Biol., 18, 423–436 (2017).
• 35) Di Terlizzi R, Platt S. The function, composition and analysis of cerebrospinal fluid in companion animals: part I—function and composition. Vet. J., 172, 422–431 (2006).
• 36) Zappaterra MW, Lehtinen MK. The cerebrospinal fluid: regulator of neurogenesis, behavior, and beyond. Cell. Mol. Life Sci., 69, 2863–2878 (2012).
• 37) Liu Y-C, Hsiao H-T, Wang JC, Wen T, Chen S. TGF-β1 in plasma and cerebrospinal fluid can be used as a biological indicator of chronic pain in patients with osteoarthritis. PLOS ONE, 17, e0262074 (2022).
• 38) Morganti-Kossmann MC, Hans VH, Lenzlinger PM, Dubs R, Ludwig E, Trentz O, Kossmann T. TGF-β is elevated in the CSF of patients with severe traumatic brain injuries and parallels blood-brain barrier function. J. Neurotrauma, 16, 617–628 (1999).
• 39) Wyss-Coray T, Feng L, Masliah E, Ruppe MD, Lee HS, Toggas SM, Rockenstein EM, Mucke L. Increased central nervous system production of extracellular matrix components and development of hydrocephalus in transgenic mice overexpressing transforming growth factor-β1. Am. J. Pathol., 147, 53–67 (1995). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1869892/
• 40) Gautschi OP, Toffoli AM, Joesbury KA, Skirving AP, Filgueira L, Zellweger R. Osteoinductive effect of cerebrospinal fluid from brain-injured patients. J. Neurotrauma, 24, 154–162 (2007).
• 41) Sato T, Mikawa S, Sato K. BMP2 expression in the adult rat brain. J. Comp. Neurol., 518, 4513–4530 (2010).
• 42) Tözser J, Earwood R, Kato A, Brown J, Tanaka K, Didier R, Megraw TL, Blum M, Kato Y. TGF-β signaling regulates the differentiation of motile cilia. Cell Rep., 11, 1000–1007 (2015).
• 43) Tzavlaki K, Moustakas A. TGF-β signaling. Biomolecules, 10, 487 (2020).