The transcription factor CREB regulates epithelial-mesenchymal transition of lens epithelial cells by phosphorylation-dependent and phosphorylation-independent mechanisms

Abstract

Epithelial mesenchymal transition (EMT) of lens epithelial cells (LECs) is one of the most important pathogenic mechanisms in lens fibrotic disorders, and the regulatory mechanisms of EMT have not been fully understood. Here, we demonstrate that the cAMP-response element binding protein (CREB) can regulate lens EMT in a phosphorylation-dependent and phosphorylation-independent manners with dual mechanisms. First, CREB-S133 phosphorylation is implicated in TGFβ-induced EMT of mouse LECs and also in injury-induced mouse anterior subcapsular cataract model. The interaction between CREB and p300 is necessary for CREB regulation of TGFβ-induced EMT, since inhibition of CREB–p300 interaction and p300 knockdown led to attenuated expression of mesenchymal genes. Second, S133A-CREB, a mutant mimicking constant dephosphorylation at S133, exhibits notable occupancy in the enhancers of mesenchymal genes and confers robust transcription activity on EMT genes. Introduction of R314A mutation in S133A-CREB, which abolishes the interaction between S133A-CREB and its co-activator, cAMP-regulated transcriptional co-activators led to substantial suppression of mesenchymal gene expression in mouse LECs. Taken together, our results showed that CREB regulates lens EMT in dual mechanisms and that the S133A-CREB acts as a novel transcription factor. Mechanistically, CREB interacts with p300 in a S133 phosphorylation-dependent manner to positively regulate lens EMT genes. In contrast, S133A-CREB interacts with cAMP-regulated transcriptional co-activators to confer a robust activation of lens EMT genes.

The cAMP response element-binding protein (CREB) belongs to the basic region/leucine zipper domain (bZIP) transcription factor superfamily and mediates the regulation of the cAMP response genes by binding as a dimer to a conserved cAMP response element (CRE): TGACGTCA in the promoter/enhancer regions (1, 2). Structurally, CREB contains a kinase inducible domain (KID) phosphorylated by various kinases in the N-terminus and a bZIP domain responsible for binding to DNA in the C-terminus. After stimulation by the external stimuli, the intracellular signaling cascades including cAMP/PKA and Ca²⁺/CaM/CaMKII pathways are activated and converged to promote the phosphorylation of CREB at S133 residue (3, 4). Upon phosphorylation at S133, KID–KIX domain interaction facilitates the recruitment of the transcription co-activator CREB-binding protein (CBP) and its paralog p300 (5), thereby activates the transcription of CREB target genes. In contrast to the CBP/p300 co-activators, the cAMP-regulated transcriptional co-activators (CRTCs) are recently identified as transcriptional co-activators of CREB, and their interactions with CREB are independent of phosphorylation of the later at serine 133 (6). In addition, the CRTC binds to the bZIP domain of CREB through the conserved N-terminal CREB-binding domain (CBD). Upon binding, the CRTC dramatically potentiates CREB transcriptional activity (7). In response to cAMP stimulus, the CRTCs:CBP/p300 complex has been reported to have reciprocal effects on their recruitment to the promoter of CREB target genes (8). The CREB:CRTC axis has been shown to modulate oxidative and proteotoxic stresses (9), adipose tissue function (10), neuronal activity (11), and metabolic programming in insulin-sensitive tissues (6).

The vertebrate lens is a transparent organ responsible for focusing light onto the photoreceptors in the retina. It contains a single layer of epithelial cells in the anterior area, and the remaining lens consists of fiber cells at different stages of differentiation (12). The transparent lens becomes opaque (cataract) due to genetic mutations, stress insult or aging (12). Such opacity is corrected through surgical operation to remove the opaque lens followed by implantation of a transparent artificial lens (13). During this process, one of the most encountered syndromes is development of the posterior subcapsular cataract (PCO) where the remaining epithelial cells of the operated lens are activated to migrate towards the posterior end and then become mesenchymal cells. Such epithelial-mesenchymal transition (EMT) of lens epithelial cells not only leads to secondary pathogenesis of PCO but also causes the anterior subcapsular cataract (ASC) (13, 14).

As a general transcription factor, CREB plays a critical role in the synaptic plasticity of the nervous system. It is also implicated in neurodegenerative disorders, diabetic complications, and tumorigenesis (15, 16, 17). In the ocular lens, we have recently demonstrated that CREB plays an important role in stress- or age-induced cataractogenesis. By regulating αB-crystallin and p300/p53 pathway, CREB promotes stress-induced apoptosis followed by cataractogenesis (18).

In the present study, we present first evidence that CREB regulates lens EMT in both S133 phosphorylation-dependent and phosphorylation-independent ways through its interaction with different co-activators. The wild-type CREB (WT-CREB) acts through S133 phosphorylation to recruit CBP/p300 co-activators and regulates TGFβ-induced EMT of lens epithelial cells. Interestingly, the mutant S133A-CREB can also regulate lens EMT in a robust way. However, the introduction of R314A-CREB mutation, which abolishes the interaction between S133A-CREB and CRTC co-activators, in S133A-CREB, suppresses its function in regulating lens EMT. Thus, our results reveal that CREB regulates lens EMT in dual mechanisms: S133 phosphorylation-dependent and phosphorylation-independent manners through its interactions with different co-activators.

Results

CREB regulates expression of mesenchymal genes in lens epithelial cells

To understand the physiological role of CREB in the ocular lens, we first examined the protein expression level of CREB in 2-month-old mouse ocular tissues including retina, cornea, and lens epithelium, as well as the major organ tissues including heart, liver, lung, brain, and muscle. Western blot analysis showed that CREB was expressed at a moderate level in lens epithelium compared with its high expression levels in cornea and lung (Fig. S1, A and B). Howbeit, CREB expression was kept at comparable levels in lens epithelium in postnatal 1-, 2-, and 6-month-old mouse. (Fig. S1, C and D).

Previous study (19) has revealed that there are well-conserved full CREB-binding sites in the core promoter regions of mesenchymal marker genes including Fn1 and Snai2. We therefore employed EMSA to test if CREB can bind to the putative CREB-binding sites in the core promoters of Fn1 and Snai2 in lens. As shown in Fig. S2A, the nuclear extracts from mouse lens epithelial cells, αTN4-1 exhibited strong binding signal to the labeled Fn1 oligo probes, which can only be competed off in the presence of 50-fold of excessive cold WT but not mutant oligos. Moreover, addition of CREB antibody to the nuclear extract-oligo probe complex gave rise to a supershifting band, confirming the specificity and authenticity of the CREB binding to the Fn1 core promoter sequence. Similar binding signals were observed when labeled Snai2 probes were applied (Fig. S2B). Together, these results suggested that CREB may act as an important regulator of mesenchymal genes.

It is well established that TGFβ-induced EMT is significantly involved in posterior capsular opacification (13, 14), in which mesenchymal gene activation occurs. To characterize the function of CREB on mesenchymal gene expression, we conducted CREB knockdown with shRNA in mouse lens epithelial cells. qRT-PCR and Western blot confirmed that CREB was knockdown about 50% with either of two independent shRNAs (Fig. S3, A–C). As a result of CREB silencing, TGFβ-induced expression of Fibronectin and SNAI2 was significantly attenuated (Fig. 1, A–C); IF staining of Fibronectin further confirmed the Western blot results (Fig. 1, D and E). Next, we employed CREB knockout (CREB-KO) in mouse lens epithelial cells (18) to test whether CREB has an indispensable role in mesenchymal gene activation. Western blot analysis showed that TGFβ-induced activation of EMT markers including both Fibronectin and SNAI2 at the protein levels in the mock knockout (mock-KO) cells (Left two lanes of panels of Figure 1, F–H). However, expression of both Fibronectin and SNAI2 was abolished in CREB-KO cells even after TGFβ treatment (Right two lanes of panels of Figure 1, F–H), and this result was further confirmed with IF staining (Fig. 1, I and J). In addition, Western blot analysis showed that transfection of WT-CREB plasmid in CREB-KO cells could partially rescue the expression of Fibronectin and SNAI2 (Fig. S4, A–D). Taken together, these results showed that CREB is an important transcription factor controlling EMT of lens epithelial cells.

Figure 1.

CREB loss of function leads to attenuated mesenchymal gene expression in mouse lens epithelial cells.A, cells were treated with 20 ng/ml TGFβ for 24 h, and Western blot analysis was used to examine the protein expression levels of Fibronectin and SNAI2 in mock or TGFβ-treated-αTN4-1 cells overexpressing NC-shRNA, CREB 1#shRNA, and CREB 2#shRNA. B and C, quantification results of Fibronectin and SNAI2 protein levels in (A). D, immunofluorescence staining for Fibronectin proteins in mock or TGFβ-treated-Mock-sh and CREB-sh-αTN4-1 cells. Scale bar represents 20 μm. E, quantification results of mean fluorescence of Fibronectin protein in (D). F, cells were treated with 20 ng/ml TGFβ for 24 h, and Western blot analysis was used to examine the protein expression levels of Fibronectin and SNAI2. G and H, quantification results of Fibronectin and SNAI2 protein levels in F. I, immunofluorescence staining for Fibronectin proteins in mock or TGFβ-treated-Mock-KO and CREB-KO-αTN4-1 cells. Scale bar represents 20 μm. J, quantification results of mean fluorescence of Fibronectin protein in (I). In B, C, E, G, H, and J, the data shown are the averages of three independent biological experiments, and error bars represent the SD of mean. Statistical analysis: Two-way ANOVA followed by Tukey’s post hoc test. ∗∗p < 0.01, ∗p < 0.05, n.s, statistically not significant.

CREB phosphorylation at S133 is necessary for its interaction with p300 and its control of expression of mesenchymal genes in lens epithelial cells

In response to stimuli, CREB phosphorylation at S133 provides a docking site for its interaction with the co-activator CBP/p300 and subsequently activates expression of the target genes (5). During TGFβ-induced EMT of mouse lens epithelial cell line αTN4-1, Western blot analysis showed that CREB phosphorylation level at S133 (p-CREB S133) was enhanced at 4 h and 8 h after TGFβ treatment (Fig. S5, A and B) and was maintained at 24 h (Fig. 2, A and B), when the mesenchymal markers including both Fibronectin and SNAI2 were activated (Fig. 2, A and B). To investigate whether CREB phosphorylation at S133 is implicated in the control of EMT during lens fibrosis in vivo, we conducted immunofluorescence staining and Western blot analysis in the injury-induced mouse model of ASC. Consistently, Western blot analysis showed that p-CREB S133 level was enhanced during ASC development (Fig. 2, C and D). Meanwhile, the immunofluorescence intensity of p-CREB S133 is much more enhanced, consistent with increased expression of the EMT marker, α-smooth muscle actin, in ASC plaques than that in normal control lens epithelium (Fig. 2E). These results showed that CREB phosphorylation at S133 might has a regulatory role in the EMT of lens epithelial cells.

Figure 2.

CREB S133 phosphorylation and CBP interaction mediate mesenchymal genes expression in lens epithelial cells.A, αTN4-1 cells were treated with 20 ng/ml TGFβ for 24 h, and Western blot analysis was used to examine the protein expression levels of p-CREB S133, Fibronectin, and SNAI2 in mock or TGFβ-treated αTN4-1 cells. B, quantification results of Fibronectin, SNAI2, and p-CREB S133 protein levels in A. C, the anterior capsules of one lens of the mouse was punctured with a needle to induce ASC, and the other lens remains as untreated normal. After 3 days of healing, the protein expression levels of lens capsular epithelium from normal and injured-induced ASC mice were examined. Western blot analysis was used to examine the protein expression levels of p-CREB S133 and Fibronectin in the lens epithelium of the normal and injured-induced ASC mice. D, quantification results of Fibronectin and p-CREB S133 protein levels in (C). E, immunofluorescence were used to analyze p-CREB S133 (green), α-SMA (red), and lens epithelial cell nuclei in the lens capsule of the normal and injured-induced ASC mice (blue) (n = 3 lenses per group). two-tailed Student’s t test was used in (B) and (D), The data shown are the averages of three independent biological experiments, and error bars represent the SD of mean. ∗∗p < 0.01, ∗p < 0.05, n.s, statistically not significant.

The p300 functions as a CREB co-factor to regulate lens EMT

To further confirm that CREB phosphorylation at S133 modulates TGFβ-induced EMT in αTN4-1 cells, we treated cells with CREB:CBP interaction inhibitor, which in essence blocks the interaction of KIX-KID domains (20, 21). To verify the blocking effect, we cloned the core promoter fragment of Fn1 gene (Fn1-P1) where CREB-binding sites reside into the luciferase reporter plasmid (Fig. S6A) and transfected αTN4-1 cells with pGL3-Fn1-P1 construct and WT-CREB overexpressing plasmid with or without CREB:CBP interaction inhibitor. Luciferase assay showed the CREB:CBP interaction inhibitor led to reduced transcriptional activity of pGL3-Fn1-P1 in a dosage-dependent manner (Fig. S6B). Meanwhile, chromatin immunoprecipitation (ChIP)-qPCR showed that 50 μM CREB:CBP interaction inhibitor reduced p300 enrichment on the core promoter of Fn1 to approximately 30% in αTN4-1 cells (Fig. S6C). Consistently, compared with DMSO treatment, CREB:CBP interaction inhibitor suppressed TGFβ-induced expression of the EMT markers in a dosage-dependent manner (Fig. 3, A–C). Specifically, treatment of the αTN4-1 cells with 10 μM inhibitor slightly affected the expression of Fibronectin and SNAI2 (Fig. 3, A–C), whereas the inhibitor at concentrations of 25 μM and 50 μM significantly suppressed the TGFβ-induced expression of Fibronectin and SNAI2 (Fig. 3, A–C). This result led us to speculate that CBP/p300 loss of function may inhibit TGFβ-induced EMT of mouse lens epithelial cells. Subsequently, we tested this hypothesis and silenced CBP expression with two independent CBP shRNAs. As shown in Figure 3, D and E, both CBP shRNA markedly inhibited the expression of CBP at mRNA and protein levels in αTN4-1 cells. Unexpectedly, CBP knockdown did not affect TGFβ-induced expression of Fibronectin and SNAI2 (Fig. 3, F–J). A possible explanation to this observation is that the paralog of CBP, p300 functionally compensated for the silencing of CBP. Indeed, Western blot analysis showed that p300 was significantly upregulated at protein level in the αTN4-1 cells overexpressing CBP shRNAs (Fig. S7, A and B).

Figure 3.

CBP gene silencing did not attenuate mesenchymal gene expression in mouse lens epithelial cells.A, mock and TGFβ-treated αTN4-1 cells were cotreated with indicated concentration of CREB-CBP inhibitor or DMSO for 24 h. Western blot analysis of the protein levels of Fibronectin and SNAI2. B and C, quantification results of Fibronectin and SNAI2 protein levels in (A). qRT-PCR (D) and Western blot analysis (E) was used to examine the mRNA and protein expression levels, respectively, of CBP in αTN4-1 cells overexpressing NC-shRNA, CBP 1#shRNA, and CBP 2#shRNA. F, Western blot analysis was used to examine the protein expression levels of Fn1 and SNAI2 in mock or TGFβ-treated-αTN4-1 cells overexpressing NC-shRNA, CBP 1#shRNA, and CBP 2#shRNA. G and H, quantification results of Fibronectin and SNAI2 protein levels in (F). I, immunofluorescence staining for Fibronectin proteins in the mock or TGFβ-treated-αTN4-1 cells overexpressing NC-shRNA, CBP 1#shRNA, and CBP 2#shRNA. Scale bar represents 20 μm. J, quantification results of mean fluorescence of Fibronectin protein in I. The data shown are the averages of three independent biological experiments, and error bars represent the SD of mean. Statistical analysis: One-way ANOVA followed by Tukey’s correction was used in (B), (C), and (D). Two-way ANOVA followed by Tukey’s post hoc test was used in (G), (H), and (J). ∗∗p < 0.01, ∗p < 0.05. n.s, statistically not significant.

To further identify the role of p300 functioning as a CREB co-activator in the EMT of mouse lens epithelial cells, we conducted ChIP-qPCR to show that p300 was still bound to the promoter regions of the Fn1 and Snai2 with comparable affinity in both CBP knockdown cells and mock knockdown cells (Fig. S7, C and D). Meanwhile, ChIP-qPCR showed that 50 μM CREB:CBP interaction inhibitor led to reduced·approximately 50% p300-binding affinity to the core promoter of Fn1 in αTN4-1 cells under TGFβ treatment (Fig. S7E). Next, we conducted p300 gene silencing using two independent shRNAs to confirm the contribution of p300 on mesenchymal gene expression. As shown in Figure 4, A and B, p300 shRNAs markedly inhibited the expression of p300 at mRNA and protein levels in αTN4-1 cells. Meanwhile, p300 knockdown significantly attenuated the TGFβ-induced expression of Fibronectin and SNAI2 (Fig. 4, C–I). These results showed that p300 acts as an important CREB co-activator controlling EMT gene expression in lens.

Figure 4.

p300 gene silencing led to attenuated mesenchymal gene expression in mouse lens epithelial cells. qRT-PCR(A) and Western blot analysis (B) was used to examine the mRNA and protein expression levels, respectively, of p300 in αTN4-1 cells overexpressing NC-shRNA, p300 1#shRNA, and p300 2#shRNA. C and D, qRT-PCR was used to examine the mRNA expression levels of Fn1 (C) and Snai2 (D) in mock and TGFβ-treated αTN4-1 cells overexpressing NC-shRNA, p300 1#shRNA, and p300 2#shRNA. E, Western blot analysis was used to examine the protein expression levels of Fibronectin and SNAI2 in mock or TGFβ-treated αTN4-1 cells overexpressing NC-shRNA, p300 1#shRNA, and p300 2#shRNA. F and G, quantification results of Fibronectin and SNAI2 protein levels in (E). H, immunofluorescence staining for Fibronectin proteins in the mock or TGFβ-treated αTN4-1 cells overexpressing NC-shRNA, p300 1#shRNA, and p300 2#shRNA. Scale bar represents 20 μm. I, quantification results of mean fluorescence of Fibronectin protein in (H). The data shown are the averages of three independent biological experiments, and error bars represent the SD of mean. Statistical analysis: One-way ANOVA followed by Tukey’s correction was used in (A). Two-way ANOVA followed by Tukey’s post hoc test was used in (C), (D), (F), (G), and (I). ∗∗p < 0.01, ∗p < 0.05. n.s, statistically not significant.

S133A-CREB acts as a novel transcription factor to regulate lens EMT

Next, we determined whether CREB S133 phosphorylation is indispensable for mesenchymal gene expression of lens epithelial cells. As expected, qPCR analysis show that WT-CREB significantly increased the mRNA levels of mesenchymal genes Fn1 and Snai2 at basal condition (Fig. 5A). Meanwhile, we also established αTN4-1 cells overexpressing S133D-CREB mutant mimicking constant phosphorylation at S133 (Fig. S8A). As shown in Fig. S8, B–G, compared with WT-CREB, S133D-CREB demonstrated much enhanced expression of mesenchymal markers Fibronectin and SNAI2 in the TGFβ-induced EMT of lens epithelial cells, which is consistent with its constant activation. However, surprisingly, S133A-CREB mutant mimicking constant dephosphorylation at S133 also led to enhanced transactivation of Fn1 and Snai2 as WT-CREB did (Fig. 5A). Western blot analysis confirmed that S133A-CREB overexpression enhanced Fibronectin and SNAI2 expression at protein level (Fig. 5B). Next, we treated αTN4-1 cells overexpressing the empty vector (Vector), WT-CREB, and S133A-CREB with TGFβ. Western blot showed that overexpression of S133A-CREB led to much more enhanced expression of mesenchymal markers Fibronectin and SNAI2 (Fig. 5, C–E). Moreover, compared with WT mouse lens, S133A-CREB heterozygous mouse lens displayed significantly higher level of Fibronectin protein in lens epithelium (Fig. 5, F–G).

Figure 5.

S133A-CREB serves as a novel transcription factor.A, qRT-PCR was used to examine the mRNA expression levels of Fn1 and Snai2 in αTN4-1 cells overexpressing pCI-neo-vector(Vector), pCI-neo-WT-CREB (WT-CREB), and pCI-neo-S133A-CREB (S133A-CREB). B, Western blot analysis was used to examine the protein expression levels of Fibronectin and SNAI2 in αTN4-1 cells overexpressing Vector, WT-CREB and S133A-CREB. C, αTN4-1 cells overexpressing Vector, WT-CREB or S133A-CREB were treated with 20 ng/ml TGFβ for 24 h. Western blot analysis was used to examine the protein expression levels of Fibronectin and SNAI2. D and E, quantification results of Fibronectin and SNAI2 protein levels in (C). F, Western blot analysis was used to examine the protein expression levels of Fibronectin in S133A-CREB heterozygote and the WT littermate mice. G, quantification results of Fibronectin protein levels in (F). H and I, ChIP-qPCR assays to demonstrate that S133A-CREB mutant proteins bind to the promoter region of Fn1 and Snai2 in vivo. The data shown are the averages of three independent biological experiments, and error bars represent the SD of mean. Statistical analysis: One-way ANOVA in (A). Two-way ANOVA followed by Tukey’s post hoc test was used in (D) and (E). Two-tailed Student’s t test in (G), (H), and (I). ∗∗p < 0.01, ∗p < 0.05, n.s, statistically not significant.

To further understand the differential regulation of EMT genes by WT-CREB and S133A-CREB, αTN4-1 cells overexpressing the empty Vector, WT-CREB, and S133A-CREB were subjected to ChIP assay using anti-CREB antibody. As shown in Figure 5, H and I, ChIP-qPCR showed that both S133A-CREB and WT-CREB bound to the promoter regions of the mesenchymal genes Fn1 and Snai2. These data suggested that S133A-CREB acts as a novel transcription factor to activate mesenchymal gene expression.

R314A mutation leads to attenuated mesenchymal gene expression

CRTC co-activators have been shown to dramatically increase CREB-mediated transcriptional activity independent of CREB S133 phosphorylation, and arginine 314 (R314) in the bZIP domain of CREB contributes to the recruitment of CRTC co-activators (6). To test whether the interaction between CREB and CRTCs contributes to the regulation of mesenchymal gene expression, we introduced R314A mutation, which specifically abolishes CRTC binding, but does not affect CREB dimerization nor DNA-binding properties (6) in both WT-CREB and S133A-CREB proteins (Fig. S9). Meanwhile, we made a dominant negative CREB, KCREB, in which arginine 301 was mutated to leucine (R301L). KCREB mutant protein loses the ability to bind CRE either as a homodimer or as a heterodimer with WT-CREB protein (22) (Fig. S9), Accordingly, we established stable αTN4-1 cell lines overexpressing empty vector (Vector), WT-CREB, KCREB, R314A-CREB, S133A-CREB, S133A-KCREB, and S133A/R314A-CREB. Western blot showed that S133A-CREB, S133A-KCREB, and S133A/R314A-CREB led to declined global p-CREB S133 levels in αTN4-1 cells compared with the overexpression of WT-CREB, KCREB, and R314A-CREB (Fig. S10, A and B). Nevertheless, overexpression of either WT-CREB or S133A-CREB enhanced the expression of mesenchymal marker Fibronectin (Fig. S10, A, C and D), while KCREB mutation in S133A-CREB led to substantially reduced expression of Fibronectin and SNAI2 (Fig. S10, A, C and D). On the other hand, compared with WT-CREB, R314A-CREB mutant overexpression reduced Fibronectin expression moderately but hardly affected SNAI2 expression (Fig. S10, A, C and D). In contrast, introduction of R314A mutation in S133A-CREB mutant caused substantially reduced expression of Fibronectin and SNAI2 (Fig. S10, A, C and D).

To understand the molecular basis of R314A and S133A mutations affect CREB target gene expression, we conduct CREB ChIP-seq. As shown in Fig. S11, A and B, both αTN4-1 Vector cells and αTN4-1 WT-CREB cells demonstrated predominant CREB-binding motif, while αTN4-1 R314A-CREB shifted binding preference to ATF1 (Fig. S11C); αTN4-1 S133A-CREB still retained CREB-binding motif (Fig. S11D); αTN4-1 S133A/R314A-CREB showed both CREB- and ATF1-binding predominance (Fig. S11E). In addition, ChIP assay against CREB antibody followed by qPCR using primers targeting Fn1 and Snai2 promoters showed that R314A-CREB bound to the promoter regions of the mesenchymal genes, with comparable affinity to WT-CREB (Fig. S12, A and B).

Next, we conduct RNA-seq analysis to identify the global effects of R314A and S133A mutations on mesenchymal gene expression in lens epithelial cells. WT-CREB and S133A-CREB overexpression led to Gene Set Enrichment Analysis (GSEA) of EMT among differentially expressed genes (Figs. S13, A and B, S14, A and B). As to the upregulated genes compared with Vector cells, WT-CREB and S133A-CREB showed extracellular matrix organization among the top ten Gene Ontology (GO) terms (Figs. S13C, S14C). On the other hand, compared with WT-CREB, R314A-CREB mutant led to GSEA enrichment of EMT among 233 differentially expressed genes (Fig. S15, A and B). Moreover, the downregulated genes in R314A-CREB mutant showed extracellular matrix organization among the top ten GO terms (Fig. S15C). Similarly, introduction of R314A mutation in S133A-CREB caused GSEA enrichment of EMT among differentially expressed genes (Fig. S16, A and B), and downregulated genes enriched included the extracellular matrix-receptor interaction among the top ten Kyoto Encyclopedia of Genes and Genomes pathways (Fig. S16C). These results further confirmed that S133A-CREB acts a transcriptional factor controlling extracellular matrix gene expression, and moreover, R314A mutation in WT-CREB and S133A-CREB attenuated the transcriptional activity on these EMT genes in lens epithelial cells.

S133A-CREB control of mesenchymal gene expression requires another co-activator, CRTC

The effect of R314A-CREB mutant on extracellular matrix gene expression led us to further explore the role of CRTC co-activators in the EMT of lens epithelial cells. We first transiently overexpressed Flag-CRTC2 fusion protein in αTN4-1 cells (Fig. S17A); IF staining of Flag-CRTC2 fusion protein showed that the Pearson R’s value of CRTC2 nuclear localization increased to 0.54 in response to TGFβ stimuli (Fig. S17B).

CRTC-DN dominant-interfering peptide is the N-terminal 1 to 44 amino acids of CRTC1, which constitutes the highly conserved CREB-binding domain, and it has been shown to block CREB activation by all identified three CRTC proteins (23, 24). We established a stable αTN4-1 cell line overexpressing HA-GFP-CRTC-DN fusion protein (Fig. S17C). We then transiently transfected Vector, WT-CREB, and S133A-CREB into the αTN4-1 HA-GFP control cells and αTN4-1-HA-GFP-CRTC-DN cells (Fig. S17D). Western blot showed that, in HA-GFP cells, WT-CREB and S133A-CREB induced much enhanced activation of Fibronectin in response to TGFβ than Vector did (Fig. S17, E and F). On the contrary, in the presence of CRTC-DN, compared to Vector, WT-CREB and S133A-CREB hardly induced the further activation of Fibronectin in response to TGFβ (Fig. S17, G and H). Together, these data suggested that the interaction with CRTC co-activators is essential for the transactivity of WT-CREB and S133A-CREB.

To further address the role of CRTC co-activator in the S133A-CREB control of EMT of lens epithelial cells, we examined the mRNA levels of Fn1 at 24 h, 36 h, and 48 h (Fig. S18A), and then we analyzed mRNA levels of Fn1 in TGFβ-treated αTN4-1 cell lines overexpressing Vector, WT-CREB, KCREB, R314A-CREB, S133A-CREB, S133A-KCREB, and S133A/R314A-CREB. S133A-CREB led remarkably enhanced activation of Fn1 upon TGFβ treatment at 24 h, 36 h, and 48 h (Fig. S18, B–D). Similarly, TGFβ-induced activation of Snai2 at mRNA level at 36 h and 48 h (Fig. S18E). S133A-CREB led remarkably enhanced activation of Snai2 upon TGFβ treatment at 48 h (Fig. S18F). RNA-seq on the mock and TGFβ-treated cells overexpressing Vector, WT-CREB, KCREB, R314A-CREB, S133A-CREB, S133A-KCREB, and S133A/R314A-CREB showed that EMT-associated genes such as Fn1, Snai2, Col1a1, Col1a3, Col5a2, and Postn were highly activated in S133A-CREB cells (Fig. S18G). In contrast, R314A mutation in either WT-CREB or S133A-CREB caused substantial reduction in the expression levels of these genes (Fig. S18G). CREB ChIP-seq demonstrated that S133A-CREB and R314A-CREB proteins as well as WT-CREB bind to the proximal promoters of Fn1, Snai2, Col1a1, Col1a3, Col5a2 (Fig. S19, A–F). Though R314A introduction into S133A-CREB failed to attenuate binding enrichment on the promoter regions of the above genes (Fig. S19, A–F), we speculated that CRTC co-activators of S133A-CREB determine its transcription activity.

Finally, we conducted Western blot to confirm that the R314A mutation in S133A-CREB led to remarkable suppression of TGFβ-induced expression of Fibronectin and SNAI2 upon TGFβ treatment (Fig. 6, A–C). R314A mutation alone barely affected TGFβ-induced Fibronectin expression but impaired the activation of SNAI2 (Fig. 6, A–E). In contrast, R314A mutation in S133A-CREB (S133A/R314A-CREB) significantly attenuated the expression of both Fibronectin and SNAI2. IF staining of Fibronectin further validated the Western blot results (Fig. 6, D and E).

Figure 6.

CRTC co-activators are essential for the transactivation of S133A-CREB. αTN4-1 cells overexpressing pCI-neo-vector(Vector), WT-CREB, KCREB, R314A-CREB, S133A-CREB, S133A-KCREB, and S133A/R314A-CREB were treated with 20 ng/ml TGFβ for 24 h. A, Western blot analysis was used to examine the protein expression levels of Fibronectin and SNAI2. B and C, quantification results of Fibronectin and SNAI2 protein levels in (A). Error bars represent the SD of the mean (n = 3). Statistical analysis: Two-way ANOVA followed by Tukey’s correction. ∗∗p < 0.01, ∗p < 0.05. D, immunofluorescence staining for Fibronectin proteins in mock or TGFβ-treated-Mock-KO and CREB-KO-αTN4-1 cells. Scale bar represents 20 μm. E, quantification results of mean fluorescence of Fibronectin protein in (D). The data shown were derived from two different clones, and three independent Western blot experiments were conducted for each clone. The error bars represent theSD of mean. Statistical analysis: Two-way ANOVA followed by Tukey’s post hoc test. ∗∗p < 0.01, ∗p < 0.05, n.s, statistically not significant.

Discussion

In this study, we have demonstrated that CREB plays an essential role in regulating EMT of lens epithelial cells. CREB regulates lens EMT in two distinct mechanisms. First, for WT CREB, the S133 phosphorylation helps to recruit the co-activators, p300 to regulate the expression of the EMT marker genes including Fn1 and Snai2. Second, for S133A-CREB, through interaction with another group of co-activators, CRTCs, it can robustly regulate expression of the same group of EMT genes. Together, our results demonstrate that CREB regulates the EMT of lens epithelial cells in both phosphorylation (at S133)-dependent and phosphorylation-independent manners in which distinct co-activators are recruited.

CREB controls gene expression in dual mechanisms

Protein phosphorylation is one of the most important mechanisms regulating activities and functions of around one third of total eukaryote proteins in response to extracellular stimuli (25, 26, 27, 28). In most cases, protein phosphorylation activates or suppresses activities of the target proteins in an on/off manner (27). For example, Pax6, a key regulator of early eye development (29, 30), is activated upon the phosphorylation of the multiple serine/threonine residues of the C-terminal PST activation domain by numerous kinases (31, 32). Dephosphorylation of these residues by protein serine/threonine phosphatase-1 inactivates Pax6 functions (33). Different from this classic on/off model after phosphorylation/dephosphorylation, here we observed that CREB regulates the expression of Fn1 and Snai2, marker genes of EMT in a different way. WT CREB can positively regulate the expression of both Fn1 and Snai2 genes after its phosphorylation at S133. More strikingly, S133A-CREB, a mutant mimicking constant dephosphorylation, presumably having impaired function in regulating EMT genes, was found to promote the expression of the EMT genes, Fn1 and Snai2 in a robust way. Thus, CREB regulates the expression of EMT genes in both phosphorylation-dependent and phosphorylation-independent manners.

To elucidate the underlying phosphorylation-independent mechanism, we created a panel of new mutants including KCREB, R314A-CREB, S133A-KCREB, and S133A/R314A-CREB where S133A-CREB cannot be phosphorylated at S133, and R314A renders the interaction between CREB with CRTC co-activators. Our results reveal while KCREB inhibits the expression of EMT genes in lens epithelial cells, R314A attenuates EMT gene expression, implying that R314A mutation interferes with CREB function in recruiting the CBP/p300 co-activators. However, for S133A-CREB, introduction of R314A mutation greatly attenuated its transcriptional activity on expression of the EMT genes, implying that S133A-CREB regulation of EMT genes requires the participation of the co-activators, CRTCs while R314A mutation prevents the recruitment of CRTCs into S133A-CREB–CRTC complex.

In human and mouse, the CRTC family consists of CRTC1, CRTC2, and CRTC3, which share similar modular structures: N-terminal CBD, central regulatory (REG) domain, splicing domain, and C-terminal transactivation domain. Upon the activation by PKA, CRTC co-activators are phosphorylated, transported to nucleus, and bound to CREB bZIP domain in an S133 phosphorylation-independent manner, thereby contributing significantly to S133A-CREB target gene activation (6). CRTC recruitment does not appear to modulate CREB DNA-binding activity but rather enhances the interaction of CREB glutamine-rich Q2 activation domain with the TFIID following its recruitment to the promoter (7). CRTCs were regarded to be important additional cofactors that discriminates the differential CREB target genes in response to non-cAMP signals (6, 34, 35, 36). Our results demonstrate that neither S133A-CREB nor R314A-CREB abolishes mesenchymal gene expression, suggesting that in lens epithelial cells, CREB interactions with CBP/p300 and CRTCs are alternative mechanisms mediating mesenchymal gene expression, which is in sharp contrast to the canonical cAMP-dependent gene activation scenario that CBP/p300 has a reciprocal effect on TORC2 recruitment (36). Furthermore, we found that the R314A mutation in S133A-CREB significantly attenuated the activation of Fibronectin and SNAI2, and the mesenchymal gene expression levels in S133A/R314A-CREB overexpressing cells are comparable to that in S133A-KCREB dominant negative mutant overexpressing cells. It has been proposed that CRTC–DNA interaction contributes the selective transcription activation of the CRE genes (37). CREB ChIP-seq in this study showed that, in spite of R314A-CREB–binding preference on ATF1, both R314A-CREB and S133A-CREB bind to the proximal promoter in the EMT genes such as Fn1, Snai2, Col1a1, Col3a1, and Postn. Attenuated expression of these EMT genes in αTN4-1-S133A/R314A-CREB cells implied that CRTCs co-activator binding to CREB is essential for the outcome of target gene expression. In this regard, our results favor that the interactions between CRTC and S133A-CREB may help to bind to CRE sites more tightly so that the later can robustly activate mesenchymal gene expression.

CREB, a multiple function transcription factor, plays an important part in lens cataractogenesis

CREB is a general transcription factor that plays fundamental roles in different organ systems and its knockout leads to perinatal lethality with respiratory distress and impaired fetal T cell development of the immune system (38). In the nervous system, CREB is implicated in the synaptic plasticity associated with long-term memory (36, 39, 40). Disruption of CREB (both α- and δ-isoforms) in mice causes defects in long-term potentiation and long-term memory (41). On the other hand, expression of the dominant-active CREB polypeptide accelerates the learning process (42, 43). CREB is also found to mediate growth factor–dependent survival of both sympathetic and cerebellar neurons (44, 45). Both NGF and BDNF activate the RSK90 kinase to phosphorylate CREB at S133 and thus promote expression of the anti-apoptotic gene Bcl-2 to promote neuron survival (46).

In contrast to these earlier studies in the nervous system, our recent studies have shown that CREB plays an important role in promoting lens pathogenesis through several mechanisms. First, under stress conditions, CREB directly binds to the promoter of the αB-crystallin gene to negatively regulate its expression in lens epithelial cells and promotes stress-induced apoptosis followed by cataractogenesis (47). During aging process, from transparent human lens to cataract patients, we observed that CREB is significantly upregulated, and accordingly, αB-crystallin becomes significantly downregulated (18). Moreover, CREB positively regulates p300 to increase p53 acetylation and its stability which promotes upregulation of the proapoptotic regulators, Bak/Bax to enhance apoptosis of lens epithelial cells (18). In the present study, we demonstrated that CREB can directly bind to the promoters of the mesenchymal genes to promote their expressions. As a result, CREB promotes EMT which is a basis for the development of ASC and PCO. Together, our results reveal that CREB is an important mediator of lens cataractogenesis induced by stress factors and aging.

PCO is one of the most prevalent complications after cataract surgery, causing vision impairment or even blindness (48). Up to now, there is no drug with low toxicity to intraocular tissues but can effectively inhibit EMT of lens epithelial cells simultaneously (49). Recent studies showed that artepillin C, which is a natural phenolic compound with a wide range of pharmacological benefits including antioxidant, antimicrobial, anti-inflammatory (50, 51), functions as an inhibitor of CREB-CRTC2 (51) and enhances insulin sensitivity by suppressing CREB/CRTC2-mediated gluconeogenic transcription (51). Our work revealed CREB/CRTCs contributes significantly to EMT of lens epithelial cells when CREB is in the dephosphorylation status (S133A-CREB), and we proposed that the natural compound artepillin C disrupting CREB–CRTC2 axis could be a promising candidate for PCO drug prophylaxis, in view of the relatively low stoichiometry of phosphorylation in essence. Future work will be conducted to evaluate the differential contribution of CRTC1, CRTC2, and CRTC3 activating mesenchymal gene expression, which will shed new light on the therapeutic drug development of clinic lens fibrotic diseases.

In summary, as a general transcription factor, CREB occupies approximately 4000 promoter sites in human cells in vivo (19). Previous studies have showed that CREB can regulate EMT gene expression in a phosphorylation-dependent manner in non-lens cells (52, 53). In the present studies, we demonstrate that both phosphorylated CREB (at S133) and nonphosphorylated S133A-CREB can bind to the target gene promoters to regulate lens EMT genes, both Fn1 and Snail as well as others. While S133 phosphorylation of CREB by various kinases promotes recruitment of the co-activators, p300 as partners to reside on the promoters of the target genes and thus enhance their expression, S133A-CREB (independent of S133 phosphorylation) can recruit CRTC co-activators as partners to reside in the same promoters of the EMT genes and promotes expression of these genes in a more robust way. Thus, CREB regulates lens EMT with dual mechanisms.

Experimental procedures

Animals

The mice used in this study were handled in compliance with the Guide for the Care and Use of Laboratory Animals (National Academy Press). 4-week-old and adult mice were obtained from the animal facility of Sun Yat-sen University. Lens epithelium was collected from these mice and used to extract total RNA and proteins.

Mouse ASC injury model

All animal studies conformed to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research. Animal care and experimental procedures were carried out in accordance with the approved guidelines of the Ethics Committee in Animal and Human Experimentation of Sun Yat-sen University. The injury-induced ASC mouse model was established as previously described (54, 55). Briefly, 4- to 6-week-old C57BL/6J mice were subjected to general anesthesia with an intraperitoneal injection of pentobarbital sodium (70 mg/kg) and topical anesthesia with dicaine eyedrops. After the topical application of mydriatic eyedrops, a small incision was made in the central anterior capsule of one eye through the cornea with the blade part of a 26-gauge hypodermic needle. The other eye of the mouse was used as a normal control without any treatment. The depth of injury was approximately 300 μm or one-fourth of the length of the blade part of the needle, which has been reported previously to lead to the formation of fibrotic tissue around the capsular wound (54). The mice were allowed to heal for 2 days. Then, the mice were sacrificed, and the lens capsules were harvested for Western blot analysis and whole-mount staining.

Immunofluorescent staining for lens anterior capsule whole mount

For whole-mount lens anterior capsules, injured mice were sacrificed, and eyes were enucleated and immediately transferred to a culture plate filled with PBS. To isolate the lens anterior capsule, the lens was dissected carefully under a dissecting microscope. Immediately, the anterior capsules of the lenses were isolated under a dissecting microscope and transferred to 4% paraformaldehyde and allowed to fix for 15 min on ice. After fixation, the capsular samples were washed three times with PBS and permeabilized with 0.3% Triton X-100. The permeabilized samples were blocked in normal goat serum for 1 h followed by overnight incubation at 4 °C with primary antibodies against p-CREB S133 (1:200, Cell Signaling Technology) and α-SMA (1: 1:100, Cell Signaling Technology). On the following day, the capsules were washed with PBS three times for 30 min and thereafter incubated with incubated with Alexa Fluor 488 goat anti-rabbit IgG (1:500, Cell Signaling Technology), Alexa Fluor 594 goat anti-mouse IgG (1:500, Cell Signaling Technology), and DAPI for 1 h at room temperature. After washing with PBS for 30 min, the whole anterior capsule was mounted flat on a microscope slide and covered with a coverslip after adding a drop of anti-fade mounting medium. The slides were stored at 4 °C in the dark for later examination using a confocal microscope. Images were acquired with a Zeiss LSM confocal laser scanning microscope (CLSM, Carl Zeiss), and average fluorescence pixel intensity of Fibronectin protein was analyzed with ZEN 2.3 lite software, according to the published measurement protocol (56).

Cell culture and treatment

Mouse lens epithelial cells (αTN4-1) (18) were cultured in monolayer in a 37 °C incubator with 5% CO₂ in Dulbecco’s modified Eagle’s medium (Sigma) supplemented with 10% fetal bovine serum and 50 Units/ml penicillin and streptomycin. CREB KO mouse epithelial cell line was described in previous study (18), in which CREB knockout was mediated by CRISPR-gRNA technology. Cells were passaged with 0.25% trypsin, seeded in six-well plates at a density of 1 × 10⁵ cells/well, and treated with 20 ng/ml TGFβ (Syd Labs, Sigma) in Dulbecco’s modified Eagle’s medium without fetal bovine serum for 24 h.

Western blot analysis

For total protein extraction, the cells were lysed in RIPA buffer (150 mM NaCl, 50 mM Tris–Cl pH 8.0, 1% NP-40, 1% sodium deoxycholate, 0.5% SDS) with protease inhibitor cocktail and protein phosphatase inhibitor cocktail (Roche). The denatured protein samples were separated by 8∼12% SDS-PAGE and transferred to nitrocellulose membranes (Bio-Rad). The membrane was blocked with 5% nonfat milk, incubated with primary antibodies at 4 °C overnight, and washed with TBST (10 mM Tris pH 8.0, 150 mM NaCl, 0.1% Tween-20) three times. Then, the membrane was incubated with horseradish peroxidase–conjugated secondary antibodies for 1 h. Protein bands were detected with chemiluminescence detection reagents. β-Tubulin or α-Actinin was used as a loading control. Quantification analysis was performed using ImageJ 1.41 (National Institutes of Health). Statistical analysis was conducted using Prism 9 software (GraphPad). The sources and dilutions of antibodies were as follows: rabbit anti-Fibronectin (1:1000, Abcam #23750), rabbit anti-SNAI2 (1:1000, Cell Signaling Technology #9585), rabbit anti-CREB (1:1000, Cell Signaling Technology #9197), rabbit anti-p-CREB S133 (1:1000, Cell Signaling Technology #9198), rabbit anti-CBP (1:1000, Cell Signaling Technology #7389), rabbit anti-p300 (1:1000, Cell Signaling Technology #70088), mouse anti-β-Tubulin (1:5000, proteintech, 66240-1-Ig), rabbit anti-α-Actinin(1:5000, proteintech, 11313-2-AP), horseradish peroxidase–conjugated horse anti-rabbit/mouse IgG (1:5000, Cell Signaling Technology). The specificities of above-mentioned primary antibodies have been validated by the corresponding manufacturer.

RNA isolation and real-time PCR and RNA-seq analysis

Total RNA was isolated using TRIzol reagent (Invitrogen) and then processed with an RQ1 RNase- Free DNase kit (Promega) to eliminate genomic DNA contamination. First-strand cDNA was synthesized with a First Strand cDNA Synthesis Kit (Fermentas). Real-time qPCR was performed in triplicate using SYBR Green I Master mix (Vazyme) with a Roche LightCycler 480II real-time qPCR system. The qRT-PCR primers are listed in Table S1. β-actin was used as internal control. All primers were synthesized by TsingKe Biotech. Statistical analysis was performed using Prism 9 software (GraphPad).

For the RNA-seq analyses, total RNA was extracted from mock or TGFβ-treated αTN4-1 cell lines overexpressing Vector, WT-CREB, KCREB, R314A-CREB, S133A-CREB, S133A-KCREB, and S133A/R314A-CREB using TRIzol reagent according to the manufacturer’s instructions. RNA-seq library preparation and subsequent sequencing were conducted by the Geneplus (ShenZhen. RNA samples were sequenced on DNBSEQ-T7 platform. The obtained sequence reads were cleaned and mapped to GRCm39 using Tophat. Gene expression and changes were analyzed using Bowtie2 and RSEM. The relative abundance of mRNAs was normalized and presented as fragments per kilobase of transcript per million mapped reads. Hierarchical cluster and scatter plot analyses of gene expression levels were performed using R software (http://www.r-project.org/). GO and Kyoto Encyclopedia of Genes and Genomes analysis were carried out by ClusterProfiler(V3.14.0), and GSEA was conducted with GSEA software (V4.1.0). The RNA-seq data were deposited in the NCBI Gene Expression Omnibus under accession number GSE276922.

ChIP-qPCR and ChIP-seq assays

ChIP was performed using Simple ChIP Enzymatic Chromatin IP Kit (Cell Signaling Technology#9003). Briefly, for CREB ChIP, 5 × 10⁶ αTN4-Vector, αTN4-WT-CREB, αTN4- S133A-CREB cells were cross-linked with 1% formaldehyde at room temperature for 10 min; for p300 ChIP, 5 × 10⁶ αTN4- NC-sh, CBP-sh 1#, CBP-sh 2# cells, and αTN4 cells that were mock-treated or treated with 20 ng/ml TGFβ for 24 h with or without 50 μM CREB:CBP interaction inhibitor were cross-linked with 1% formaldehyde at room temperature for 15 min. After micrococcal nuclease digestion, chromatin was sonicated into 150 to 900 bp DNA fragments. Ten micrograms of CREB antibody or p300 antibody or normal IgG were incubated with 10 μg chromatin at 4 °C overnight, then magnetic Protein A/G beads were added in the immunoprecipitation complex for incubation for at least 2 h. CREB ChIP complex was washed with low salt buffer and high salt buffer and subjected to chromatin elution, reversal of cross-linking, and DNA purification. CREB or p300 bound DNA fragments were amplified using qPCR ChIP-qPCR primers are listed in Table S1. Meanwhile, CREB ChIP DNA was subjected to ChIP-seq analysis by Geneplus. The identification of ChIP-seq peaks was performed using MACS 2.2. Enriched transcriptional factor motif analysis on the ChIP-seq peaks was carried out using HOMER. The specificities of CREB and p300 primary antibodies in ChIP assay have been validated by the corresponding manufacturer. The ChIP-seq data were deposited in the NCBI Gene Expression Omnibus under accession number GSE276462.

Gel mobility shifting assays

The gel mobility shifting assay (EMSA) was conducted as we described before (33). The oligos used were listed in Supporting Fig. S2, A and B. For the binding assays, 2 μg of nuclear extracts from αTN4-1 cells were incubated with biotin-labeled double-stranded synthetic oligos for 20 min on ice. For competition experiments, 50-fold of the unlabeled WT or mutant (MT) oligos were pre-incubated with the nuclear extracts for 20 min before addition of the labeled probe. For supershifting assays, the nuclear extracts were incubated with anti-CREB or normal IgG for 20 min on ice, then the reaction mixture was incubated with the labeled oligos for 20 min at room temperature, allowing formation of the supershifting complex. The reaction mixture was separated with 4% native gel.

Expression constructs

cDNA obtained from αTN4-1 cells was used as PCR templates to amplify the full length of the coding sequences of mouse CREB and CRTC2 gene. pCI-neo-WT-CREB plasmid was constructed as previous study described (18). pCI-neo-S133A-CREB, pCI-neo-KCREB, pCI-neo-R314A-CREB, pCI-neo-S133A-KCREB, and pCI-neo-S133A/R314A-CREB mutation constructs were obtained using the QuikChange Site-Directed Mutagenesis Kit according to the manual (Agilent Technologies). CRTC2 was cloned into the 3 × Flag pCMV10 vector at EcoRI (5′) and KpnI (3′) sites to generate Flag-CRTC2 fusion protein. CRTC2-DN peptide was clone into pEGFP-N3 vector at EcoRI (5′) and BamHI (3′) sites to generate GFP-CRTC2-DN fusion fragment. Then GFP-CRTC2-DN or GFP fragments were subcloned into pLVX-IRES-Puro-3 × HA-N vector at the EcoRI (5′) and BamHI (3′) sites to generate pLVX-IRES-Puro-3 × HA-CRTC-DN-GFP and pLVX-IRES-Puro-3 × HA-GFP constructs, respectively. The primers used to conduct the mutagenesis of S133A-CREB, KCREB, and R314A-CREB, and cloning of CRTC, CRTC2-DN-GFP are listed in Table S2. All constructs are confirmed by sequencing.

RNA interference and CRTC-DN overexpression

For shRNA-mediated gene knockdown, a set of single-stranded oligonucleotides encoding the mouse CREBCBP and p300 target shRNA and its complement were synthesized. CREB #1: sense, 5′-CCGGCAGCAGCTCATGCAACATCATCTCGAGATGATGTTGCATGAGCTGCTGTTTTTG-3′, CREB #2 sense: CCGGACTGATGGACAGCAGATTCTA CTCGAGTAGAATCTGCTGTCCATCAGTTTTTTG-3′. CBP #1: sense, 5′- CCGGCCAACCTCAGACGACAATTTCCTCGAGGAAATTGTCGTCTGAGGTTGGTTTTTG-3′, CBP #2: sense, 5′ CCGGGAGGATCATTAACGACTATAACTCGAGTTATAGTCGTTAATGATCCTCTTTTTG-3′. p300 #1: sense, 5′- CCGGCCAACAGGAATGACTACCAATCTCGAGATTGGTAGTCATTCCTGTTGGTTTTTG-3′, p300 #2: sense, 5′ CCGGGCTAGTCCTATGGGTGTAAATCTCGAGATTTACACCCATAGGACTAGCTTTTTG-3′. The oligonucleotide sense and antisense pair were annealed and inserted into pLKO lentiviral expression plasmid. The psPAX2 packaging plasmid, pMD2.G envelope plasmid, and pLKO-NC shRNA, pLKO-CREB 1# shRNA, pLKO-CREB 2# shRNA, pLKO-CBP 1# shRNA, pLKO-CBP 2# shRNA, pLKO-p300 1# shRNA, pLKO-p300 2# shRNA, pLVX-IRES-Puro-3 × HA -CRTC-DN-GFP, and pLVX-IRES-Puro-3 × HA-GFP constructs were used to co-transfect HEK-293FT cells. Virus supernatant was collected 48 h post-transfection, filtered through a 0.45 μm polyethersulfone filter. The virus suspension was mixed with 8 μg/ml polybrene to infect αTN4-1 cells. Puromycin screening was conducted 48 h post-infection. Western blot and qRT-PCR analyses were used to identify gene knockdown and CRTC-DN overexpression.

Luciferase assays

Genomic DNA obtained from αTN4-1 cells was used as PCR templates to amplify the Fn1 P1 fragment, which was cloned into the pGL3 basic vector at KpnI (5′) and XhoI (3′) sites to generate the firefly luciferase reporter plasmid. Dual-luciferase reporter kit (Beyotime) was used in our analysis. Briefly, 50 ng of internal control pRL Renilla Luciferase plasmid, 0.3 μg of pCI-neo-WT-CREB or Vector, together with pGL3-Fn1 P1 plasmid were introduced into α-TN4-1 cells using Lipofectamine 2000 (Invitrogen). CREB-CBP interaction inhibitor or DMSO were added in the cell culture medium 6 h post-transfection. After 24 h, the transfected cells harvested for luciferase assays according to the company instruction manual.

Immunofluorescence staining

Cells were seeded on Millicell EZ 24-well glass slides (Millipore). After treatment for the indicated times, the cells were fixed with 4% paraformaldehyde, permeabilized with 0.3% Triton X-100, and blocked with normal goat serum. Then, the sections were incubated with anti-Fibronectin (1:200, Abcam) or anti-Flag (1:200, Proteintech) prepared with goat serum. After washing with PBS for three times, the sections were incubated with Alexa Fluor 488 goat anti-rabbit/Mouse IgG (1:500, Cell Signaling Technology) and DAPI for 1 h at room temperature. Slides were mounted with anti-fade fluorescent mounting medium (Applygen). Images were acquired by a Zeiss LSM confocal laser scanning microscope (CLSM, Carl Zeiss) and analyzed with ZEN2.1 software.

Statistical analysis

Quantitative data are presented as the mean ± SD (standard deviation of the mean) and analyzed by one-way ANOVA or two-way ANOVA, and multiple comparisons between the groups were corrected using Tukey's post hoc test. Two-tailed Student’s t test was performed when two groups were compared. Statistical significance was set at a level of p < 0.05. In the figures, ∗ and ∗∗ refer to p < 0.05 and p < 0.01, respectively. The data shown are the averages and SD of at least three biological independent experiments. Statistical analysis was performed with GraphPad Prism 9.5.

Data availability

All data presented in the figures and supplementary information of this paper are available. The RNA-seq and ChIP-seq data have been deposited in the NCBI Gene Expression Omnibus (GEO) under accession number GSE276922 and GSE276922, respectively.

Supporting information

This article contains supporting information.

Conflicts of interest

The authors declare that they have no conflicts of interest with the contents of this article.

Reviewed by members of the JBC Editorial Board. Edited by Kirill Martemyanov