Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Aug 4.
Published in final edited form as: Exp Cell Res. 2026 May 12;460(1):115064. doi: 10.1016/j.yexcr.2026.115064

GSK3β functions as a mechanosensitive regulator that links flow stress to Shh signaling

Yu Dai 1,2,*, Ying Liu 1,3,*, Hao Xu 1,4, Yueyang Chi 1, Jiayang Chen 1, Vivien Li 1, Yu Jiang 1,#
PMCID: PMC13431002  NIHMSID: NIHMS2189525  PMID: 42128195

Abstract

Primary cilia sense laminar flow through physical deflection, yet how this mechanical cue is translated into signaling regulation remains unclear. Here, we demonstrate that flow-induced ciliary bending promotes the accumulation of GSK3β within cilia, leading to repression of Sonic Hedgehog (Shh) signaling. This response requires the tumor suppressor folliculin (FLCN). Loss of GSK3β results in ligand-independent ciliary accumulation of Smoothened (Smo) and aberrant Shh signaling. Mechanistically, GSK3β phosphorylates Smo at two sites in its C-terminal intracellular domain, and mutations at these sites enhance Smo retention within cilia. Shh stimulation induces inhibitory phosphorylation of GSK3β at T390, thereby relieving its negative regulation of Shh signaling. Together, these findings identify GSK3β as a mechanosensitive regulator that couples flow-induced ciliary bending to Shh pathway activity through control of Smo ciliary localization.

Keywords: Sonic Hedgehog Signaling, GSK3β, Primary cilium, Mechanotransduction

Introduction

Most non-dividing, quiescent cells possess a solitary, microtubule-based, membrane-enclosed surface projection known as the primary cilium [1,2]. This unique, cell state–dependent organelle houses various receptors, ion channels, and key components of multiple signaling pathways, making it essential for sensing and transmitting chemical cues from the environment [3]. Beyond chemical signaling, the primary cilium also functions as a critical mechanosensor [4]. In response to mechanical stimuli such as fluid flow, ciliary bending modulates downstream signaling cascades involving calcium influx, planar cell polarity, and mechanistic target of rapamycin (mTOR) signaling [5-7]. This mechanosensory function is crucial for maintaining tissue homeostasis and regulating cell proliferation [8]. Disruption of ciliary mechanotransduction has been implicated in a range of diseases, including polycystic kidney disease and certain cancers [9-11].

Among the many signaling pathways coordinated through the cilium, the Sonic Hedgehog (Shh) pathway is particularly reliant on ciliary function for proper activation and signal transduction [12]. Core Shh signaling components, including Patched1 (Ptch1), Smoothened (Smo), and the Gli transcription factors, undergo regulated trafficking to and from the cilium upon Shh ligand stimulation [13-15]. In the absence of ligand, Ptch1 resides in the cilium and inhibits Smo, thus preventing downstream signaling. Shh binding leads to Ptch1 exit from the cilium and subsequent Smo accumulation, which activates Gli transcription factors (Gli1, Gli2, and Gli3). These factors then translocate to the nucleus to regulate Shh target gene expression [16]. Smo is thought to activate the Gli proteins by interfering with their interaction with Suppressor of Fused (Sufu), which localizes to the base and shaft of the cilium in unstimulated cells but accumulates at the ciliary tip upon ligand stimulation [17].

Gli3 activity is further regulated by glycogen synthase kinase 3 beta (GSK3β). In the absence of Shh ligand, GSK3β, together with protein kinase A (PKA) and casein kinase 1 (CK1), phosphorylates Gli3, promoting its partial proteolysis into a transcriptional repressor form. Upon Shh activation, phosphorylation by GSK3β is diminished, and full-length Gli activators are stabilized. Thus, GSK3β serves as a key negative regulator of Shh signaling by controlling Gli3 activation [18].

In ciliated cells, the major components of the Shh pathway operate in the primary cilium in response to ligand stimulation. However, it remains unclear whether flow-induced ciliary bending intersects with the signaling and how the mechanical cues are converted into chemical signals. In the present study, we show that flow-induced bending of cilia leads to the ciliary accumulation of GSK3β and that this accumulation plays a pivotal role in regulating Shh signaling.

Results

Flow-induced ciliary accumulation of GSK3β

It has been previously shown that flow stress controls mTORC1 signaling through regulation of ciliary accumulation of LKB1, suggesting that mechanosignals, like chemical signals, can modulate ciliary signaling by altering the localization of signaling proteins within the cilium [19]. To explore this concept further, we analyzed several kinases known to associate with ciliary pathways for their distribution under flow stress. Among these, GSK3β showed a clear flow-dependent change in localization. In mouse embryonic fibroblasts (MEF) cells grown in the absence of flow, only a small fraction of cilia exhibited detectable GSK3β staining (Fig. 1A). Upon exposure to flow stress, however, the proportion of cilia with detectable GSK3β significantly increased, and the staining intensity became readily discernible (Fig. 1A and 1B). GSK3β was not detected in GSK3β-null MEF cells, confirming the specificity of the antibody used for GSK3β detection (Fig. S1). These results indicate that ciliary accumulation of GSK3β is dynamically regulated by flow stress.

Figure 1. Flow stress promotes ciliary accumulation of GSK3β and Gli3 processing.

Figure 1.

Open in a new tab

Ciliated MEF cells were exposed to flow stress (flow) or mock treatment (ctrl) for 16 h. A. Confocal imaging of cilia stained with anti–acetylated tubulin (Tub, red), anti-GSK3β (green), and DAPI (blue). Scale bar, 10 μm. B. Quantification of GSK3β accumulation. Left: percentage of GSK3β-positive cilia (n = 3, total 150 cilia/condition,). Right: relative levels of GSK3β in cilia (n = 3, total 60 cilia/condition). C, D. Western blot and quantification of full-length (FL) and repressor (R) forms of Gli3 in GSK3β wild-type (+) and null (−) MEF cells under flow (+) or mock (−) condition (n = 3). E, F. Western blot and quantification of full-length (FL) and repressor (R) forms of Gli3 in MEF cells treated with LiCl (+) or vehicle (−), with (+) or without (−) flow (n = 3) for 16 h. Data are presented as mean ± SD. ***P < 0.001, **P < 0.01, *P < 0.05, ns = not significant.

GSK3β is a well-established negative regulator of the Shh pathway, where it phosphorylates full-length Gli3 (Gli3FL), promoting its partial proteasomal processing into the truncated repressor form (Gli3R) that suppresses Shh target gene expression [20]. To assess whether flow-induced ciliary accumulation of GSK3β influences Shh signaling, we examined Gli3 degradation in ciliated MEF cells subjected to flow stress. Compared with mock-treated controls, flow-stressed cells showed a marked decrease in the Gli3FL/R ratio (Fig. 1C, 1D), indicating an enhanced conversion of Gli3 into its repressor form. Consistent with the increased level of the Gli3 repressor, the expression levels of Gli1 and Ptch1, two major targets of the Shh pathway, were reduced in flow-stressed cells (Fig. S2).

This increase in Gli3R levels correlated with flow-induced ciliary accumulation of GSK3β, suggesting a causal link between the two events. To test this, we assessed Gli3 processing in GSK3β-deficient MEF cells. While the basal Gli3FL/R ratio in these cells was slightly elevated compared to wild type, flow stress failed to induce the reduction observed in controls (Fig. 1C, 1D). The absence of GSK3β also nullified the negative effect of flow stress on Gli1 and Ptch1 expression (Fig. S2). Similarly, inhibition of GSK3β with LiCl abolished the flow-induced increase in Gli3R levels (Fig. 1E, 1F). Together, these findings demonstrate that GSK3β is essential for flow-induced Gli3 degradation and suppression of Shh signaling.

To further investigate the effect of flow stress on ciliary GSK3β accumulation, we analyzed the dynamics of this process under various flow conditions. As shown in Fig. 2, flow stress induced ciliary accumulation of GSK3β in a time- and shear force–dependent manner. When ciliated MEF cells were subjected to a constant shear force of 1.0 dyn/cm2, significant increases in both the number of GSK3β-positive cilia and the level of ciliary GSK3β accumulation were observed within 4 hours, reaching a maximum at 8 hours (Fig. 2A, 2B). Similarly, significant increases in ciliary GSK3β accumulation were detected when cells were exposed to a shear force of 0.2 dyn/cm2 for 8 hours, with maximal accumulation occurring at 0.5 dyn/cm2 (Fig. 2C, 2D). These results demonstrate that GSK3β ciliary accumulation dynamically responds to changes in flow stress conditions.

Figure 2. Time and shear force dependent ciliary accumulation of GSK3β.

Figure 2.

Open in a new tab

Ciliated MEF cells were exposed to flow stress at a shear force of 1.0 dyn/cm2. The ciliary presentation of GSK3β at the indicated time points was analyzed by confocal imaging. A. Percentage of GSK3β-positive cilia (n = 3, total 150 cilia/time point). B. Relative levels of GSK3β in cilia (n = 3, total 60 cilia/time point). Ciliated MEF cells were exposed to flow stress at different shear forces for 8 h. The ciliary presentation of GSK3β at the indicated shear force was analyzed by confocal imaging. C. Percentage of GSK3β-positive cilia (n = 3, total 150 cilia/time point). D. Relative levels GSK3β in cilia (n = 3, total 60 cilia/condition). Quantified data are presented as mean ± SD. **P < 0.01, *P < 0.05, ns = not significant.

Flow-induced ciliary accumulation of GSK3β requires FLCN

We previously demonstrated that flow-induced accumulation of LKB1 within primary cilia depends on folliculin (FLCN), a tumor suppressor protein that interacts with kinesin motor proteins of the intraflagellar transport (IFT) machinery [19]. To determine whether a similar mechanism governs the ciliary localization of GSK3β under flow stress, we examined the role of FLCN in this process. As shown in Fig. 3, under no-flow conditions, GSK3β was largely absent from cilia in FLCN-deficient UOK257 cells, whereas a small subset of cilia in FLCN-proficient cells displayed faint GSK3β staining. Upon exposure to flow stress, GSK3β remained undetectable in the cilia of FLCN-deficient cells. By contrast, FLCN-proficient UOK257-2 cells exhibited a robust increase in the percentage of cilia positive for GSK3β (Fig. 3A), consistent with the response observed in wild-type MEF cells (Fig. 1A). Quantitative analysis confirmed a significant reduction in flow-induced ciliary GSK3β localization in the absence of FLCN (Fig. 3B). These findings demonstrate a requirement for FLCN in this process.

Figure 3. Flow-induced ciliary accumulation of GSK3β requires FLCN.

Figure 3.

Open in a new tab

A. Ciliated FLCN-deficient UOK 257 (FLCN−/−) and proficient UOK257-2 (FLCN+/+) cells were subjected to flow stress (flow) or mock treatment (ctrl) for 16 h. Ciliary localization of GSK3β was examined by confocal imaging. Cilia were stained with anti–acetylated tubulin (Tub, red), GSK3β with anti-GSK3β antibody (green), and nuclei with DAPI (blue). Scale bar, 10 μm. B. Quantification of GSK3β-positive cilia (n = 3, total 150 cilia/condition). Data are presented as mean ± SD. ***P < 0.001, ns = not significant.

GSK3β regulates ciliary distribution of Sufu and Smo

Dynamic redistributions of Sufu and Smo within the cilium are key events in Shh ligand stimulated Gli activation [21]. To determine whether GSK3β plays a role in these events, we analyzed the localization of Sufu and Smo in GSK3β-deficient cells. In serum-starved wild-type MEFs, a majority of cilia displayed Sufu distributed along the shaft of cilia, with only a subset exhibiting Sufu concentrated at the tip. By contrast, in MEF cells lacking GSK3β, either by genetic deletion (Fig. 4A, 4B) or inhibition with LiCl (Fig. 4C, 4D), this distribution was reversed, with tip-localized Sufu becoming predominant. These results suggest that GSK3β restricts the tip accumulation of Sufu.

Figure 4. Loss of GSK3β activity enhances Sufu accumulation at the ciliary tip.

Figure 4.

Open in a new tab

A. Confocal images of Sufu distribution in ciliated GSK3β wild-type (+/+) and null (−/−) MEF cells. Cilia were stained with anti–acetylated tubulin (Tub, green), Sufu with anti-Sufu antibody (red), and nuclei with DAPI (blue). B. Quantification of Sufu tip vs. full-length (FL) distribution in GSK3β wild-type (+/+) and null (−/−) MEF cells (n = 3, total 150 cilia/condition). C. Confocal images of Sufu localization in ciliated wild-type MEF cells treated with 10 mM LiCl or vehicle for 16 h. D. Quantification of Sufu tip vs. full-length (FL) distribution in control- or LiCl-treated cells (n = 3, total 150 cilia/condition). Data are presented as mean ± SD.

We next examined the effect of GSK3β deletion on Smo localization. In serum-starved wild-type MEF cells, Smo was absent from most cilia. In contrast, a significant fraction of cilia in GSK3β-deficient MEF cells exhibited detectable Smo (Fig. 5A, 5B). The absence of GSK3β also increased the ciliary levels of Smo (Fig. 5C). Shh ligand stimulation increased the proportion of Smo-positive cilia in both wild-type and GSK3β-deficient cells. Under this condition, no significant difference in ciliary accumulation of Smo was observed between the two types of cells (Fig. 5B and 5C), suggesting that GSK3β regulates basal but not ligand-induced Smo accumulation. Consistent with this finding, ectopically expressed GFP-tagged Smo showed significantly greater ciliary localization in serum-starved and unstimulated GSK3β-deficient cells compared with wild-type controls (Fig. S3). Together, these observations indicate that GSK3β negatively regulates the basal ciliary accumulation of Smo, while Shh stimulation overrides this inhibition.

Figure 5. Loss of GSK3β promotes ciliary accumulation of Smo.

Figure 5.

Open in a new tab

A. Confocal images of ciliary Smo localization in serum-starved GSK3β wild type (+/+) and null (−/−) MEF cells treated with Shh or vehicle control (ctrl) for 6 h. Cilia were stained with anti–acetylated tubulin (Tub, red), Smo with anti-Smo antibody (green), and nuclei with DAPI (blue). B. Quantification of Smo-positive cilia shown in A (n = 3, total 150 cilia/condition). C. Relative levels of Smo in cilia shown in A (n = 3, total 60 cilia/condition). Data are presented as mean ± SD. ***P < 0.001, **P < 0.01.

GSK3β regulates Smo through phosphorylation at its C-terminal domain

The finding that GSK3β deletion promoted ciliary accumulation of Smo revealed a role of GSK3β in regulating Smo. Sequence analysis of the C-terminal intracellular domain (CID) of Smo identified two putative GSK3β phosphorylation sites at positions S615 and T644, which conform to the consensus motif T/SXXXT/S (Fig. 6A). In this motif, the first serine/threonine serves as the target of GSK3β phosphorylation, while the C-terminal serine/threonine functions as a priming site, typically phosphorylated by another kinase [22]. This observation suggests that GSK3β may control Smo through direct phosphorylation. Consistent with the notion, we found that GSK3β was able to phosphorylate bacterially expressed recombinant CID of Smo in vitro. Importantly, the phosphorylation was markedly enhanced by priming with casein kinase 1 delta (CK1δ), but not with protein kinase A (PKA) (Fig. 6B). Mutation of both sites (S615A and T644A) substantially abolished GSK3β-mediated phosphorylation (Fig. 6C), confirming that these residues are major GSK3β target sites within CID of Smo. These results establish that GSK3β is capable of phosphorylating Smo directly.

Figure 6. GSK3β-directed phosphorylation of Smo regulates its ciliary accumulation.

Figure 6.

Open in a new tab

A. Predicted GSK3β phosphorylation sites (S615 and T644) within the C-terminal intracellular domain (CID) of Smo. B. In vitro phosphorylation of GST-SmoCID by GSK3β after primed with CK1δ or PKA. C. In vitro phosphorylation of wild-type and mutant GST-SmoCID proteins (S615A, T644A, S615A/T644A) by GSK3β after primed with CK1δ. Phosphorylation was detected by autoradiography, quantified using ImageJ software, and normalized to protein levels. Data are expressed relative to the level of the wild type protein and presented as mean ± SD (n = 3). D. Ciliary localization of GFP-tagged wild-type and mutant Smo proteins after treating with Shh or vehicle control for 6 h. E. Quantification of GFP-positive cilia in cells expressing different Smo-GFP proteins (n = 3, total 150 cilia/condition). Data are presented as mean ± SD. ***P < 0.001, **P < 0.01.

To assess the functional impact of these phosphorylation events on Smo ciliary localization, we generated GFP-tagged Smo mutants carrying either single (S615A or T644A) or double (S615A/T644A) substitutions and expressed them in Smo-null MEF cells. Compared with cells expressing wild-type Smo-GFP, those expressing the single-point mutants showed a significantly higher proportion of GFP-positive cilia. Cells expressing the double mutant exhibited an even greater increase in ciliary Smo accumulation (Fig. 6D, 6E). Shh stimulation further enhanced ciliary Smo accumulation in cells expressing single-point mutants, but not in those expressing the double mutant (Fig. 6D, 6E). The percentage of GFP-positive cilia in cells expressing the double mutant was comparable to that observed in GSK3β-null MEF cells expressing wild-type Smo-GFP (Fig. S3). Together, these findings strongly suggest that GSK3β negatively regulates ciliary accumulation of Smo by phosphorylating its C-terminal intracellular domain at residues S615 and T644.

Shh regulates GSK3β activity

Having established a role for GSK3β in regulating ciliary accumulation of Smo, we next investigated whether GSK3β activity itself is modulated by Shh stimulation. To this end, we examined phosphorylation of GSK3β at three sites known to regulate its activity, including S9, Y216, and T390 [23], in response to Shh treatment. As shown in Fig. 7A, Shh stimulation selectively increased phosphorylation at T390, while phosphorylation at S9 and Y216 remained unchanged. Since phosphorylation at T390 has previously been shown to inhibit GSK3β activity [24], these results suggest that Shh stimulation is able to downregulate GSK3β activity through T390 phosphorylation.

Figure 7. Shh induces GSK3β phosphorylation at T390.

Figure 7.

Open in a new tab

A. HEK293 cells were starved for serum (0.5%) for 24 h followed by treatment with Shh (+) or vehicle control (−) for 6 h. GSK3β phosphorylation at S9, Y216, and T390 in the treated cells were analyzed by western blot. B. Ciliated GSK3β-null (−/−) MEF cells stably expressing HA-tagged wild-type, T390A, or T390D mutants of GSK3β, along with control GSK3β-null (−/−) MEF cells, were serum-starved (0.5% serum) for 24 hours, followed by Shh and mock treatment for 6 hr. The expression of full-length (FL) and repressor (R) forms of Gli3 in the treated cells was analyzed by western blot. C. Quantification of Gli3 FL/R ratios (n = 3) shown in B. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ns = not significant.

To assess the functional significance of this regulation, we examined Shh-induced Gli3 processing in GSK3β-null MEF cells reconstituted with wild-type GSK3β, a non-phosphorylatable mutant (T390A), or a phosphomimetic mutant (T390D). In cells expressing either wild-type GSK3β or the phosphomimetic T390D mutant, Shh stimulation significantly increased the FL/R ratio of Gli3, suggesting an enhanced stabilization of the full-length activator form. In contrast, in cells expressing the T390A mutant, Shh treatment failed to significantly alter the FL/R ratio, indicating that Gli3 activation was impaired (Fig. 7B, 7C). Consistent with the diminished activation of Gli3, the Shh-induced expression of Gli1 and Ptch1 was significantly reduced in cells expressing the T390A mutant (Fig. S4). These findings demonstrate that Shh signaling attenuates GSK3β activity through phosphorylation at T390, and this inhibitory modification is important for Shh-induced activation of Gli3.

Although the phosphomimetic T390D mutant was expected to be inactive, we did not observe a significant increase in the FL/R ratio in cells expressing this mutant in the absence of Shh stimulation, compared with the control GSK3 null cells (Fig. 7B, 7C). One possible explanation is that the mutant does not fully recapitulate the phosphorylated state and retains partial activity, a limitation commonly associated with phosphomimetic substitutions [25].

The findings that GSK3β is regulated both by flow stress and Shh signaling suggest the kinase may function as converge point to integrate mechanic and Shh signals for regulation of Smo. Accordingly, we examined whether the Shh stimulated ciliary accumulation of Smo was affected by flow stress. In serum-starved MEF cells without Shh stimulation, assessment of ciliary Smo accumulation in response to flow stress was challenging due to the low number of Smo-positive cilia and low levels of ciliary Smo (data not shown), precluding a definitive conclusion. In serum starved and Shh stimulated GSK3β-null MEF cells expressing wild type GSK3β, flow stress induced a slight decrease in both the percentage of Smo-positive cilia and the level of ciliary Smo, however, these changes were not statistically significant. Similarly, flow stress did not to cause any detectable changes in the ciliary Smo accumulation in GSK3β-null MEF cells treated with Shh (Fig. 8A, 8B). These observations suggest that flow stress does not affect ciliary Smo accumulation under Shh treatment. One possible explanation for the lack of an effect of flow stress on Smo is that flow-induced ciliary GSK3β is inhibited by Shh-stimulated phosphorylation at T390. To test this hypothesis, we examined the effect of flow stress on GSK3β-null MEF cells expressing the nonphosphorylatable T390A mutant of GSK3β under Shh stimulation. We found that flow stress induced a significant decrease in both the percentage of Smo-positive cilia and the level of ciliary Smo (Figs. 8A, 8B). These results indicate that in the absence of Shh-mediated inhibition of GSK3β, flow stress can regulate ciliary Smo accumulation, supporting a coordinated interaction between mechanical cues and Shh signaling.

Figure 8. Flow stress regulates ciliary Smo accumulation.

Figure 8.

Open in a new tab

Ciliated GSK3β-null (−/−) MEF cells expressing wild type HA-tagged wild-type, T390A, along with control GSK3β-null (−/−) MEF cells were serum starved (0.5% serum) for 24 h. The starved cells were treated with Shh and exposed to flow stress at 1.0 dyn/cm2 or mock treatment for 8 h. Ciliary presentation of Smo in the treated cells was analyzed by confocal imaging and quantified. A. Percentage of Smo-positive cilia (n = 3, total 150 cilia/condition). B. Relative levels of Smo in cilia (n = 3, total 60 cilia/condition). Data are presented as mean ± SD. *P < 0.05, ns = not significant.

Discussion

As a mechanosensory organelle, the primary cilium plays a central role in converting mechanical cues into biochemical signals, yet the underlying mechanisms remain poorly understood. In this study, we demonstrate that flow-induced bending of primary cilia represses Shh pathway activity by promoting ciliary accumulation of GSK3β, a key negative regulator of Shh signaling. These findings uncover a mechanism by which mechanical stimuli are translated into modulation of Shh pathway activity.

A previous study has demonstrated that flow stress enhances ciliary retention of the tumor suppressor LKB1, leading to inhibition of mTORC1 signaling [19]. Our current data reveal that GSK3β distribution within the cilium is also responsive to flow stress (Fig. 1). These findings suggest the existence of a common mechanism that regulates ciliary accumulation of soluble protein kinases. In support of the view, the flow-induced accumulation of both LKB1 and GSK3β requires FLCN (Fig. 3), a tumor suppressor known to function as a GAP for Rag GTPases [26]. In the primary cilium, FLCN interacts directly with kinesin-2 motor proteins, but is not required for ciliogenesis [19,27]. These findings rule out FLCN as an essential component of intraflagellar transport. FLCN is likely a cargo protein that is transported into the cilium, where it facilitates retention of specific signaling proteins in response to mechanical stress.

GSK3β is classically known to promote proteasomal processing of full-length Gli3 into its truncated repressor form (Gli3R) through phosphorylation events coordinated with PKA and CK1 [18]. Our results extend this role, showing that GSK3β also limits the ciliary accumulation of Smo. In the absence of GSK3β, a greater proportion of cilia exhibits Smo localization, even without Shh ligand stimulation. However, Shh ligand can further increase Smo-positive cilia in GSK3β-deficient cells (Fig. 5), indicating that GSK3β acts as an auxiliary regulator rather than the primary determinant of Smo localization. Mechanistically, our data suggest that GSK3β phosphorylates the Smo C-terminal intracellular domain at two sites that match the consensus sequence of GSK3β phosphorylation (Fig. 6). Smo mutants lacking these sites exhibit an increased ciliary retention compared to wild-type Smo, consistent with the idea that GSK3β-dependent phosphorylation facilitates Smo internalization and removal from the cilium. Reduced phosphorylation in the absence of GSK3β would thus facilitate Smo accumulation in the ciliary membrane.

We also show that Shh ligand stimulation induces phosphorylation of GSK3β at T390, a modification previously linked to p38 MAPK–dependent inhibition of GSK3β activity [24]. This suggests that Shh receptor engagement may either activate p38 or facilitate its recruitment to GSK3β, resulting in a reduced kinase activity. Inhibition of GSK3β via T390 phosphorylation would enhance ciliary Smo accumulation and relieve GSK3β-mediated suppression of Gli3 activation. Supporting this model, we show that Shh-induced Gli3 activation is diminished in cells expressing a non-phosphorylatable GSK3β T390A mutant (Fig. 7). Thus, Shh not only relieves Ptch1-mediated inhibition of Smo but also increases the pathway output by downregulating GSK3β activity, thereby promoting efficient Gli3 activation.

While flow stress strongly promotes ciliary accumulation of GSK3β (Fig. 1), its effect on ciliary localization of Smo is insignificant in the presence of Shh (Fig. 8). A likely explanation is that Shh-induced phosphorylation of GSK3β at T390 inhibits its kinase activity, thereby limiting its ability to regulate Smo. Consistent with the view, we found that flow stress significantly decreased ciliary Smo accumulation when Shh-mediated inhibition of GSK3β in blocked (Fig. 8). This observation reveals a mechanism whereby flow-derived mechanical signals act in coordination with Shh signaling to modulate ciliary localization and activation of Smo. In the presence of Shh, reduced GSK3β activity offsets the effect of its increased ciliary accumulation, thereby attenuating the impact of flow stress on Smo and overall Shh signaling. Conversely, when Shh signaling is low, flow-induced enrichment of GSK3β in cilia may suppress Smo function by inhibiting its activity and/or reducing its ciliary accumulation. This coordinated regulation enables mechanical cues to integrate with chemical signals, thereby fine-tuning Shh pathway activity within primary cilia. Further studies are required to validate this model and elucidate the underlying molecular mechanisms.

In summary, our results reveal a novel mechanotransduction mechanism in which flow-induced bending of primary cilia promotes ciliary accumulation of GSK3β. This localized enrichment of GSK3β facilitates Gli3 processing into its repressor form, thereby attenuating Shh signaling. Furthermore, our data show that Shh ligand promotes pathway activation in part by downregulating GSK3β activity through T390 phosphorylation. Together, these findings highlight GSK3β as a central node linking mechanical cues and ligand-induced modulation of Shh signaling at the primary cilium.

Materials and Methods

Cell Lines, cultures and plasmids

GSK3β knockout (GSK3β −/−) and wild-type (GSK3β +/+) MEFs were provided by James Woodgett (Mount Sinai Hospital, Toronto) [28]. Smoothened knockout (Smo −/−) and wild-type (Smo +/+) MEFs were obtained from Philip Beachy (Stanford University) [29]. The FLCN-deficient UOK257 renal carcinoma cell line and its cognate wild-type control line, UOK257–2, were provided by Laura Schmidt and Marston Linehan (NCI) [30]. HEK293 cells were from ATCC.

MEF cells were maintained in DMEM supplemented with 10% fetal bovine serum (FBS). UOK cells were cultured in DMEM supplemented with 10% heat-inactivated FBS and 100 U/ml penicillin/streptomycin. Ciliogenesis was induced by allowing cells to grow to overconfluence. Maximal ciliation was observed 3 days after full confluence in MEF cells and 5 days in UOK cells, at which point the cells were analyzed.

Antibodies for FLCN (Cat. # 3697), GSK3β (Cat #12456), phospho-GSK3β (Ser9) (Cat# 9323), phospho-GSK3β(T390) (Cat#3548), and Sufu (Cat#2522) were purchased from Cell Signaling Technology (Danvers, MA). Antibodies for phospho-GSK3β(Y216) (Cat# SAB4300237), acetylated-tubulin (Cat. # T7451), and GFP (Cat# 11814460001) were from Sigma-Aldrich (St. Louis, MO). Antibodies for β-actin (Cat# BDB612657) and Gli3 (Cat# AF3690) were from Fisher Scientific (Pittsburgh, PA). Anti-Smo antibody for immunofluorescent imaging was a gift from Rajat Rohatgi at Stanford University [13]. Recombinant mouse Shh ligand (Cat#464SH025) was purchased from Fisher Scientific.

The pGFP-Smo plasmid, which expresses a C-terminal GFP-tagged mouse Smo gene (Cat. #25395), was obtained from Addgene (Watertown, MA). This plasmid was originally generated in Philip Beachy’s laboratory [31]. The GFP-Smo fusion gene was subsequently subcloned into the pLenti6 vector. Site-directed mutagenesis was performed to generate point mutants, including GFP-Smo (S615A), GFP-Smo (T644A), and the double mutant GFP-Smo (S615A/T644A). The pLenti6-GFP-Smo constructs were used for transient transfection of MEF cells.

To study Smo phosphorylation, plasmids (pGEX6P1-SmoCID) encoding N-terminally GST-tagged Smo C-terminal intracellular domain (CID, aa550-793) were generated by PCR amplification of the CID sequence from the Lenti6-GFP-Smo constructs and cloned into the pGEX6P1 vector. The pcDNA3-HA-GSK3β plasmid (Cat. #14753) was obtained from Addgene [32]. The HA-tagged GSK3β gene was subcloned into the pLenti6 vector, and site-directed mutagenesis was used to generate phosphorylation mutants of GSK3β (T390A) and GSK3β (T390D).

Shh treatment

MEF cells were grown into confluence to induce ciliation. The ciliated cells were then serum-starvation (0.5% serum) for 24 h followed by treatment with recombinant Shh at a concentration of 0.5 μg / ml of culture medium.

In vitro kinase assay

Recombinant Protein Kinase A (PKA) catalytic subunit (Cat. #P6000) was purchased from New England Biolabs (Ipswich, MA), recombinant casein kinase 1 delta (CK1δ) (Cat. # NBP25170301) from Fisher Scientific, and recombinant GSK3β (Cat. #ab63193) from Abcam (Waltham, MA). [γ-32P] ATP (Cat. #BLU002250UC) was from Perkin Elmer (Shelton, CT). GST-tagged CID of Smo and its mutant forms were produced in E. coli by expressing pGEX6P1-SmoCID plasmids. Recombinant GST-SmoCID proteins were purified using glutathione-conjugated beads.

For priming reactions, purified GST-SmoCID proteins (100 μg) were incubated with either PKA catalytic subunit (10,000 U) or CK1δ (2 μg) for 30 min at 30 °C in 300 μl kinase buffer containing 50 mM Tris-Cl (pH 7.4), 50 mM NaCl, 10 mM MgCl2, 1 mM DTT, supplemented with 1 mM ATP. The primed GST-SmoCID proteins were re-purified using glutathione-conjugated beads (30 μl) and eluted with 100 μl of 10 mM reduced glutathione.

For GSK3β phosphorylation assays, the primed GST-SmoCID proteins (1 μg) were incubated with recombinant GSK3β (100 ng) for 30 min at 30 °C in 50 μl of kinase buffer supplemented with 200 μM ATP and 1 μCi [γ-32P] ATP. Reactions were terminated by adding 50 μl of 2x SDS sample buffer, followed by incubation at 95 °C for 5 min. Samples were resolved by SDS-PAGE, and phosphorylation was assessed by autoradiography to detect 32P incorporation into GST-SmoCID proteins.

Flow assay

Flow assays were performed as previously described [19]. Briefly, cells were seeded into μ-slide I 0.8 luer chambers (L50 × W5 × H0.8 mm; ibidi) pre-coated with collagen. Chambers were placed between two reservoirs containing 40 ml of medium, connected via a peristaltic pump. The pump was activated by an electronic controller for 5 seconds every hour, allowing approximately 0.2 ml of medium to pass through the chamber. Cells were maintained in the chambers for 3–5 days until maximal ciliogenesis was achieved.

Ciliated cells were exposed to flow at a shear stress of 1.0 dyn/cm2 for 16 h unless otherwise indicated. The flow system setup was as previously described [19], consisting of an air pressure pump and a two-way switch valve that drove 40 ml of culture medium unidirectionally between two reservoirs through the flow chamber. Except for the air pump, the entire system was housed inside a CO2 incubator.

Immunofluorescence staining and image analysis

Cells were fixed with 4% paraformaldehyde at room temperature and permeabilized with 0.02% saponin in PBS containing 2% BSA and 1% fish skin gelatin, as previously described [19]. Fixed cells were incubated overnight at 4 °C with primary antibodies, followed by 1 h incubation at room temperature with Alexa Fluor 488– or Cy3-conjugated secondary antibodies (Fisher Scientific). Primary cilia were labeled with anti–acetylated tubulin antibody, and nuclei were counterstained with DAPI.

Fluorescent images were acquired using an Olympus BX61WI Fluoview™ FV1000 confocal microscope equipped with an Olympus PlanApo 60×/1.45 NA oil objective. Image analysis was performed using ImageJ software. Ciliary GSK3β and Smo levels were quantified based on average pixel intensity of fluorescence signals colocalized with acetylated-tubulin–marked cilia, as previously described [33]. All images from a given experiment were acquired with identical acquisition settings.

qPCR

RNAs were purified using Qiagen RNeasy Kit. qPCR was performed using Bio-Rad iTaq™ Universal SYBR Green One-Step Kit with Bio-Rad CFX96 C1000 thermal cycler. The primers for Gli1 are: 5’-ACTCGGACTCGCAGGAG-3’ (forward) and 5’-ACTCGGTCATTCTCACACTTG-3’ (reverse). The primers for Ptch1 are: 5’-ACAAGCCCATCGACATTAGTC-3’(forward) and 5’-CAAGCGGTCAGGTAGATGTAG-3’ (reverse). The expression levels were normalized against those of β-actin.

Data Analysis

All data were from three independent replicate experiments and quantitative values were expressed as mean ± SD. Pairwise data comparison was analyzed with Student’s t-test; ***P < 0.001, **P < 0.01, *P < 0.05, ns = not significant.

Supplementary Material

1

Highlights:

  • Flow stress promotes ciliary accumulation of GSK3β.

  • GSK3β represses Shh signaling by phosphorylating Smo and limiting its levels within cilia.

  • Shh ligand stimulation inhibits GSK3β

Acknowledgement

The authors thank Drs. James Woodgett, Philip Beachy, Laura Schmidt and Marston Linehan for providing cell lines used in the study and Dr. Rajat Rohatgi for anti-Smo antibody. This work was supported by a grant from NIH (GM132127) to YJ.

Footnotes

Declaration of interests

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Yu Jiang reports financial support was provided by National Institutes of Health. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • [1].Goetz SC, and Anderson KV (2010) The primary cilium: a signalling centre during vertebrate development. Nature reviews genetics 11, 331–344 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [2].Satir P, and Christensen ST (2007) Overview of structure and function of mammalian cilia. Annu. Rev. Physiol 69, 377–400. [DOI] [PubMed] [Google Scholar]
  • [3].Nachury MV, and Mick DU (2019) Establishing and regulating the composition of cilia for signal transduction. Nature reviews Molecular cell biology 20, 389–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Berbari NF, O'Connor AK, Haycraft CJ, and Yoder BK (2009) The primary cilium as a complex signaling center. Current biology 19, R526–R535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Nauli SM, Alenghat FJ, Luo Y, Williams E, Vassilev P, Li X, Elia AEH, Lu W, Brown EM, and Quinn SJ (2003) Polycystins 1 and 2 mediate mechanosensation in the primary cilium of kidney cells. Nature genetics 33, 129–137. [DOI] [PubMed] [Google Scholar]
  • [6].Nonaka S, Tanaka Y, Okada Y, Takeda S, Harada A, Kanai Y, Kido M, and Hirokawa N (1998) Randomization of left–right asymmetry due to loss of nodal cilia generating leftward flow of extraembryonic fluid in mice lacking KIF3B motor protein. Cell 95, 829–837. [DOI] [PubMed] [Google Scholar]
  • [7].Boehlke C, Kotsis F, Patel V, Braeg S, Voelker H, Bredt S, Beyer T, Janusch H, Hamann C, and Gödel M (2010) Primary cilia regulate mTORC1 activity and cell size through Lkb1. Nature cell biology 12, 1115–1122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [8].Pan J, Seeger-Nukpezah T, and Golemis EA (2013) The role of the cilium in normal and abnormal cell cycles: emphasis on renal cystic pathologies. Cellular and Molecular Life Sciences 70, 1849–1874. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [9].Michaud EJ, and Yoder BK (2006) The primary cilium in cell signaling and cancer. Cancer research 66, 6463–6467. [DOI] [PubMed] [Google Scholar]
  • [10].Hildebrandt F, Benzing T, and Katsanis N (2011) Ciliopathies. New England Journal of Medicine 364, 1533–1543. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Reiter JF, and Leroux MR (2017) Genes and molecular pathways underpinning ciliopathies. Nat Rev Mol Cell Biol 18, 533–547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Bangs F, and Anderson KV (2017) Primary cilia and mammalian hedgehog signaling. Cold Spring Harbor perspectives in biology 9, a028175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Rohatgi R, Milenkovic L, and Scott MP (2007) Patched1 regulates hedgehog signaling at the primary cilium. Science 317, 372–376. [DOI] [PubMed] [Google Scholar]
  • [14].Corbit KC, Aanstad P, Singla V, Norman AR, Stainier DYR, and Reiter JF (2005) Vertebrate Smoothened functions at the primary cilium. Nature 437, 1018–1021. [DOI] [PubMed] [Google Scholar]
  • [15].Hui C.-c., and Angers S (2011) Gli proteins in development and disease. Annual review of cell and developmental biology 27, 513–537. [DOI] [PubMed] [Google Scholar]
  • [16].Zhang Y, and Beachy PA (2023) Cellular and molecular mechanisms of Hedgehog signalling. Nature Reviews Molecular Cell Biology 24, 668–687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Chen Y, Yue S, Xie L, Pu XH, Jin T, and Cheng SY (2011) Dual Phosphorylation of suppressor of fused (Sufu) by PKA and GSK3beta regulates its stability and localization in the primary cilium. J Biol Chem 286, 13502–13511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Zhou M, and Jiang J (2022) Gli Phosphorylation Code in Hedgehog Signal Transduction. Front Cell Dev Biol 10, 846927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Zhong M, Zhao X, Li J, Yuan W, Yan G, Tong M, Guo S, Zhu Y, Jiang Y, Liu Y, and Jiang Y (2016) Tumor Suppressor Folliculin Regulates mTORC1 through Primary Cilia. J Biol Chem 291, 11689–11697. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Østerlund T, and Kogerman P (2006) Hedgehog signalling: how to get from Smo to Ci and Gli. Trends in cell biology 16, 176–180. [DOI] [PubMed] [Google Scholar]
  • [21].Tukachinsky H, Lopez LV, and Salic A (2010) A mechanism for vertebrate Hedgehog signaling: recruitment to cilia and dissociation of SuFu–Gli protein complexes. Journal of Cell Biology 191, 415–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Doble BW, and Woodgett JR (2003) GSK-3: tricks of the trade for a multitasking kinase. Journal of cell science 116, 1175–1186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].McCubrey JA, Steelman LS, Bertrand FE, Davis NM, Sokolosky M, Abrams SL, Montalto G, D'Assoro AB, Libra M, and Nicoletti F (2014) GSK-3 as potential target for therapeutic intervention in cancer. Oncotarget 5, 2881. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Thornton TM, Pedraza-Alva G, Deng B, Wood CD, Aronshtam A, Clements JL, Sabio G, Davis RJ, Matthews DE, and Doble B (2008) Phosphorylation by p38 MAPK as an alternative pathway for GSK3β inactivation. Science 320, 667–670. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Bressler SG, Grunhaus D, Hurevich M, and Friedler A (2026) Methods for studying the effects of phosphorylation patterns in proteins. Biochemical Society Transactions 54, BST20250137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Tsun Z-Y, Bar-Peled L, Chantranupong L, Zoncu R, Wang T, Kim C, Spooner E, and Sabatini DM (2013) The folliculin tumor suppressor is a GAP for the RagC/D GTPases that signal amino acid levels to mTORC1. Molecular cell 52, 495–505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Luijten MNH, Basten SG, Claessens T, Vernooij M, Scott CL, Janssen R, Easton JA, Kamps MAF, Vreeburg M, and Broers JLV (2013) Birt–Hogg–Dubé syndrome is a novel ciliopathy. Human molecular genetics 22, 4383–4397 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].Hoeflich KP, Luo J, Rubie EA, Tsao M-S, Jin OU, and Woodgett JR (2000) Requirement for glycogen synthase kinase-3β in cell survival and NF-κB activation. Nature 406, 86–90. [DOI] [PubMed] [Google Scholar]
  • [29].Ma Y, Erkner A, Gong R, Yao S, Taipale J, Basler K, and Beachy PA (2002) Hedgehog-mediated patterning of the mammalian embryo requires transporter-like function of dispatched. Cell 111, 63–75. [DOI] [PubMed] [Google Scholar]
  • [30].Yang Y, Padilla-Nash HM, Vira MA, Abu-Asab MS, Val D, Worrell R, Tsokos M, Merino MJ, Pavlovich CP, and Ried T (2008) The UOK 257 cell line: a novel model for studies of the human Birt–Hogg–Dubé gene pathway. Cancer genetics and cytogenetics 180, 100–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Chen JK, Taipale J, Cooper MK, and Beachy PA (2002) Inhibition of Hedgehog signaling by direct binding of cyclopamine to Smoothened. Genes Dev 16, 2743–2748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].He X, Saint-Jeannet JP, Woodgett JR, Varmus HE, and Dawid IB (1995) Glycogen synthase kinase-3 and dorsoventral patterning in Xenopus embryos. Nature 374, 617–622. [DOI] [PubMed] [Google Scholar]
  • [33].Rohatgi R, Milenkovic L, Corcoran RB, and Scott MP (2009) Hedgehog signal transduction by Smoothened: pharmacologic evidence for a 2-step activation process. Proceedings of the National Academy of Sciences 106, 3196–3201. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2189525-supplement-1.pdf (5.7MB, pdf)

RESOURCES