Skip to main content
NCBI home page
iScience logoLink to iScience
. 2020 Jul 15;23(8):101369. doi: 10.1016/j.isci.2020.101369

Evolutionarily Conserved Roles for Apontic in Induction and Subsequent Decline of Cyclin E Expression

Xian-Feng Wang 1,2, Jin-Xiao Liu 2, Zhi-Yuan Ma 2, Yang Shen 2, Hao-Ran Zhang 2, Zi-Zhang Zhou 2, Emiko Suzuki 1, Qing-Xin Liu 2,∗, Susumu Hirose 1,3,∗∗
PMCID: PMC7394757  PMID: 32736066

Summary

Cyclin E is a key factor for S phase entry, and deregulation of Cyclin E results in developmental defects and tumors. Therefore, proper cycling of Cyclin E is crucial for normal growth. Here we found that transcription factors Apontic (Apt) and E2f1 cooperate to induce cyclin E in Drosophila. Functional binding motifs of Apt and E2f1 are clustered in the first intron of Drosophila cyclin E and directly contribute to the cyclin E transcription. Knockout of apt and e2f1 together abolished Cyclin E expression. Furthermore, Apt up-regulates Retinoblastoma family protein 1 (Rbf1) for proper chromatin compaction, which is known to repress cyclin E. Notably, Apt-dependent up-regulation of Cyclin E and Rbf1 is evolutionarily conserved in mammalian cells. Our findings reveal a unique mechanism underlying the induction and subsequent decline of Cyclin E expression.

Subject Areas: Biological Sciences, Molecular Biology, Cell Biology

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    Mutual activation of apt and e2f1 promotes rapid induction of CycE at S phase entry

  • •

    Apt also up-regulates Rbf1, but Rbf1 is inactivated through phosphorylation by Cdk2

  • •

    After initiation of S phase, Rbf1 becomes active and represses cycE

  • •

    Apt governs both induction and subsequent repression of cycE

Biological Sciences; Molecular Biology; Cell Biology

Introduction

As a key factor for S phase entry, Cyclin E (CycE) is crucial for both mitotic and endocycling cells (Dulic et al., 1992; Knoblich et al., 1994; Lilly and Spradling, 1996). Previous studies have shown that transcription factor E2f, a heterodimer of E2f1 and Dp, plays an important role in cycE expression in normal cell cycle and endocycle (De Veylder et al., 2002; Duronio et al., 1995; van den Heuvel and Dyson, 2008; Zielke et al., 2011). However, Drosophila e2f1 null mutant can survive to the third-instar larval stage and residual S phase occurs in the mutant cells (Duronio et al., 1998; Royzman et al., 1997), suggesting an involvement of another transcription factor in S phase entry.

One candidate for such factor is Apt. Drosophila Apt (also termed Trachea defective, Tdf) is a DNA-binding transcription factor that is involved in the development of multiple organs and tissues, such as tracheae, head, heart, ovary, stem cell, nervous system, and imaginal discs (Eulenberg and Schuh, 1997; Gellon et al., 1997; Lie and Macdonald, 1999; Liu et al., 2003, 2014; Monahan and Starz-Gaiano, 2016; Shen et al., 2018; Starz-Gaiano et al., 2008; Su et al., 1999; Wang et al., 2017). We have found that Apt directly regulates the expression of cycE during the development of imaginal discs (Liu et al., 2014; Wang et al., 2017). Therefore, Apt might participate in the expression of cycE in other tissues also. Besides, Apt can suppress tumor metastasis, and the human homolog of Apt, FSBP, is a cancer-related factor (Lau et al., 2010; Woodhouse et al., 2003).

Although CycE is crucial for S phase entry, it should decrease subsequently for progression of the cell cycle. Rbf1 is a key player in the decline of CycE expression (Cayirlioglu et al., 2003; Korenjak et al., 2012; van den Heuvel and Dyson, 2008; Weng et al., 2003). During S phase entry, Rbf1 is inactivated by phosphorylation with Cyclin-dependent kinase 2 (Cdk2) but becomes active by de-phosphorylation after initiation of S phase (Du et al., 1996; Edgar and Orr-Weaver, 2001; van den Heuvel and Dyson, 2008). The activated Rbf1 binds to E2f1 and also forms another complex containing E2f2 and Dp to repress E2f1-target genes including cycE and many other genes (Cayirlioglu et al., 2003; Korenjak et al., 2012; van den Heuvel and Dyson, 2008; Weng et al., 2003). Retinoblastoma protein (Rb), a mammalian counterpart of Rbf1, promotes chromatin compaction for transcriptional silencing by interaction with chromatin regulators such as histone deacetylases and histone methyltransferases (Brehm et al., 1998; Nielsen et al., 2001; Talluri and Dick, 2012). Therefore, Rbf1 is also expected to participate in chromatin compaction for silencing.

Here we provide evidence that both Drosophila Apt and mouse FSBP play important roles in the induction of CycE and up-regulation of Rbf1 for proper chromatin compaction. Mechanistically, we showed that Apt and E2f1 mutually activate the expression of each other to induce cycE for S phase entry in the salivary gland. Furthermore, we observed that the binding motifs of Apt and E2f are clustered in the first intron of cycE. Based on the results of chromatin immunoprecipitation (ChIP) and transgenic reporter assays, we found direct contribution of the Apt-binding sites and the E2f1-binding sites to the cycE transcription in the salivary gland. Moreover, we also found that Apt up-regulates Rbf1 to direct proper chromatin compaction for transcriptional silencing. Finally, we demonstrated evolutionary conservation of these mechanisms in mammalian cells.

Results

Apt and E2f1 Activate Expression of Each Other

To investigate the function of Apt in endoreplication of the salivary gland, we first compared the expression of Apt and E2f1 proteins by immunostaining (Figure 1A). Endocycle in the salivary gland proceeds asynchronously, and hence each cell resides in various phases of endocycle. E2f1 peaks at S phase entry and declines after initiation of S phase (Zielke et al., 2011). According to the oscillation of E2f1 during endocycle, some cells expressed E2f1 strongly, whereas other cells expressed weakly. Intriguingly, we noticed that the expression of Apt exhibits a similar pattern as that of E2f1. The observed tight correlation between the levels of Apt and E2f1 proteins suggests almost-synchronous oscillation of Apt and E2f1 during endocycle.

Figure 1.

Figure 1

Open in a new tab

Apt and E2f1 Up-regulate Each Other in the Salivary Gland

(A) Expression of Apt and E2f1 at 72–84 h after egg laying (AEL). Each picture is the same single focal plane image of the salivary gland obtained with a confocal microscope. n = 16 with all samples showing the represented phenotype. Scale bar, 20 μm. The graph shows the correlation between Apt and E2f1 protein levels in each cell. n = 177. R2 (coefficient of determination) = 0.89.

(B) RT-qPCR assays for expression of e2f1, apt, and cycE mRNAs. The glands were prepared from early third-instar larvae. Data were average ± SD relative to the dpp > GFP mRNA level. dpp > GFP samples were normalized to 1. ∗∗∗p < 0.001 (Student's t test).

(C) Transgenic reporter assays for apt transcription. The reporter AptPlacZ (apt promoter region with wild-type E2f1-binding motifs) showed an expression pattern similar to the endogenous E2f1 expression. n = 8 with all samples showing the represented phenotype. AptMPlacZ (base substitutions in the E2f1-binding motifs in the apt promoter region) showed significantly decreased LacZ expression. n = 5 with all samples showing the represented phenotype. Scale bars, 20 μm.

(D) RT-qPCR assays for lacZ mRNA levels from AptPlacZ or AptMPlacZ. Data were average ± SD relative to the mRNA level of AptPlacZ. ∗∗∗p < 0.001 (Student's t test).

See also Figures S1 and S2.

As Apt and E2f1 are transcription factors, the strong correlation between the levels of Apt and E2f1 could be due to the interdependence of the apt and e2f1 expression. To test the possibility, we analyzed mRNA levels of apt and e2f1 in the salivary gland by RT-qPCR. To compare the mRNA levels among samples with different genome dosages, each mRNA level was normalized to that of β-tubulin mRNA. RNAi knockdown of e2f1 using a dpp-GAL4 driver (Figure S1A) decreased the expression of apt, and vice versa (Figure 1B). Furthermore, the expression of Apt and E2f1 proteins were dependent on each other (Figure S1B). These data demonstrate mutual activation of apt and e2f1. The positive feedback between apt and e2f1 would support their rapid and robust transaction. To examine whether these activations are direct or not, we searched for E2f1- and Apt-binding motifs in the apt or e2f1 promoter. E2f1-binding sites were found in the apt promoter region, suggesting that E2f1 might directly activate apt transcription. We used the 1.5-kb promoter region containing the E2f1-binding sites (Figure S2) to verify this possibility through transgenic reporter assays. As we expected, wild-type reporter gene was expressed in the salivary gland (Figure 1C), whereas the expression level of the reporter gene significantly decreased in the E2f1-binding site mutant line (Figures 1C and 1D). These results demonstrate that E2f1 can directly activate apt transcription in the salivary gland. Because the Apt-binding site was not found in the e2f1 promoter region, the apt-mediated activation of e2f1 might be indirect.

Both Apt and E2f1 Are Required for CycE Expression and Endoreplication

Considering the tight correlation between the expression of Apt and E2f1, it is most likely that Apt is involved in endocycle together with E2f1. To test the possibility, we induced apt null mutant clones in embryonic salivary glands where cells still undergo mitosis and observed the glands at the third-instar larval stage after many rounds of endoreplication in control cells. Compared with control cells, apt-knockout cells showed obvious decrease in DAPI fluorescence (Figures 2A and 2B). The decreased DAPI fluorescence indicates the role for Apt in endoreplication. In addition to the decreased DAPI staining, loose chromatin appearance was observed in apt-mutant clone cells (Figure 2A). We will refer to the phenotype in the later section.

Figure 2.

Figure 2

Open in a new tab

Apt Is Required for Endoreplication and CycE Expression in the Salivary Gland

(A) An apt-mutant clone (marked by white dotted lines) was stained with DAPI. n = 12 with all samples showing the represented phenotype. Scale bar, 20 μm.

(B) Quantification of DAPI fluorescence from apt-mutant clone cells and neighboring control cells. Data were presented as mean ± SD. n = 4 cells for apt-mutant clones and 10 cells for control. ∗∗∗p < 0.001 versus the control (Student's t test).

(C) Immunofluorescence staining with anti-CycE antibody and DAPI in apt-mutant clones. Arrows show reduced expression of CycE in apt-mutant clones. n = 11 with all samples showing the represented phenotype. Scale bar, 20 μm.

(D) e2f1-knockout cells (marked by red arrowheads) showed lower ploidy. n = 7 with all samples showing the represented phenotype. Scale bar, 20 μm.

(E) An apt- and e2f1-double knockout cell (marked by yellow arrowheads) showed almost no endoreplication. n = 4 with all samples showing the represented phenotype. Scale bar, 20 μm.

(E′) Close-up image around the yellow arrowhead in (E). Scale bar, 20 μm.

(F) CycE was decreased but still detectable in an e2f1-mutant cell (marked by red arrowheads). n = 4 with all samples showing the represented phenotype. Scale bar, 20 μm.

(G) apt- and e2f1-double mutant cell (marked by yellow arrowheads) abolished CycE expression. n = 3 with all samples showing the represented phenotype. Scale bar, 20 μm.

(H) Quantification of DAPI fluorescence in e2f1-mutant clone cells, apt-e2f1-double mutant clone cells, and neighboring control cells. Data were presented as mean ± SD. n = 4 cells for control, 3 cells for e2f1-mutant clone, and 3 cells for apt-e2f1-double mutant clone. ∗∗∗p < 0.001 versus the control (Student's t test). See also Table S4.

In agreement with the previous studies (De Veylder et al., 2002; Duronio et al., 1995; van den Heuvel and Dyson, 2008; Zielke et al., 2011), knockdown of e2f1 decreased the expression of cycE in the salivary gland (Figure 1B). Furthermore, knockdown of apt also reduced the expression of cycE. Having established that both E2f1 and Apt are cycE activators, we compared the DNA content and CycE protein level among control cells, apt-knockout cells, e2f1-knockout cells, and apt- and e2f1-double knockout cells. The DNA content decreased clearly in apt- or e2f1-knockout clone cells (Figures 2A, 2B, 2D and 2H). Double knockout of apt and e2f1 completely blocked endoreplication (Figures 2E, 2E′ and 2H). Knockout of either apt or e2f1 significantly decreased the CycE expression, but residual CycE protein was still detectable (Figures 2C and 2F). Upon double knockout of apt and e2f1, the expression level of CycE reduced below the detection limit (Figure 2G). These data collectively demonstrate that both Apt and E2f1 are required for proper CycE expression and endoreplication in the salivary gland.

Apt and E2f1 Can Directly Activate cycE Transcription

As Apt and E2f1 activate the expression of each other and both Apt and E2f1 are required for the CycE expression, effect of e2f1 null mutation on the CycE expression is a combination of a direct effect due to the absence of E2f1 and an indirect effect due to the reduced Apt level. To address the direct contribution of Apt or E2f1 to cycE transcription, we focused on cis-regulatory elements of cycE. Expression of cycE is regulated by complex tissue-specific cis-elements (Jones et al., 2000). Although cis-elements for the expression in the salivary gland have not been reported, we found a clustering of two adjacent E2f1-binding motifs and four Apt-binding motifs in the first intron of cycE (Figure 3A). To test whether Apt or E2f1 binds to these motifs, we carried out ChIP experiments using Apt or E2f1 antibodies. The ChIP data clearly showed that Apt and E2f1 bind to the corresponding motifs (Figures 3B and 3C). Consistently, RNAi knockdown of apt prominently reduced the occupancies of Apt on the Apt motifs (Figure 3B). We then assessed the functional activities of these sites by transgenic reporter assays (Figures 3D and S3A). Control reporter carrying the 3-kb wild-type cycE region (cycEPlacZ) expressed LacZ in the salivary gland. Compared with the control, the LacZ expression decreased but was still detectable in E2f1-binding site mutation line (E2f1BSMPlacZ). Simultaneous disruption of the E2f1- and Apt-binding sites (E2f1BSM + AptBSMPlacZ) virtually abolished the reporter gene expression. Furthermore, the LacZ expression in E2f1BSMPlacZ reduced significantly upon RNAi knockdown of apt compared with non-RNAi control (Figure S3B). These data demonstrate that Apt and E2f1 can directly activate the cycE transcription. These data also indicate that both the Apt- and E2f1-binding sites are required for the normal level of cycE transcription.

Figure 3.

Figure 3

Open in a new tab

Both Apt- and E2f1-Binding Sites Are Required for Normal Level of cycE Transcription

(A) Clustering of Apt- and E2f1-binding motifs in the first intron of cycE. Apt-binding motifs and E2f1-binding motifs are indicated in red and blue, respectively.

(B and C) ChIP assays revealed occupancy of Apt (B) or E2f1 (C) at each binding motif. ChIP-qPCR was performed using antibodies against Apt (B) or E2f1 (C). Control regions of the anti-Apt antibody ChIP were set around 1.2 kb upstream of the Apt motif 1 (Ctr-1) and around 1.2 kb downstream of the Apt motif 4 (Ctr-2). Control regions of the anti-E2f1 antibody ChIP were set around 1.2 kb upstream of the E2f1 motif 1 (Ctr-1) and around 1.4 kb downstream of the E2f1 motif 2 (Ctr-2). Data were presented as mean ± SD. n = three biological replicates. ∗∗∗p < 0.001 (Student's t test) between ptc-GAL4 and control IgG or between aptRNAi and control ptc-GAL4 in (B) and versus control IgG in (C).

(D) Reporter assays for contribution of Apt- and E2f1-binding motifs to cycE transcription. The left panel shows strategy of the transgenic reporter assays. Middle panel: The wild-type reporter (cycEPlacZ) displayed clear LacZ expression. n = 16 with all samples showing the represented phenotype. The E2f1-binding motif's mutant reporter (E2f1BSMPlacZ) exhibited reduced LacZ expression. n = 11 with all samples showing the represented phenotype. The reporter of Apt- and E2f1-binding motif's mutant (E2f1BSM + AptBSMPlacZ) almost abolished the LacZ expression. n = 8 with all samples showing the represented phenotype. Scale bars, 100 μm. Right panel: Quantification of the LacZ expression.

Data were mean ± SD relative to the level of cycEPlacZ. n = 23 for cycEPlacZ, 20 for E2f1BSMPlacZ, and 29 for E2f1BSM + AptBSMPlacZ. ∗∗∗p < 0.001 (Student's t test). See also Figures S3 and S4.

The 3-kb regulatory element of cycE (termed I element here) is different from the 4-kb region (termed U element) that governs the cycE expression in the eye and wing discs (Liu et al., 2014; Wang et al., 2017). The I element is within the first intron, whereas the U element is immediately upstream of the start site of the cycE transcript A (Thurmond et al., 2019) (Figure S4A). E2f1-binding sites are present in the I element but not in the U element. To test the specificity of the cis-regulatory region, we checked the reporter activity of the I element in the eye disc. The I element failed to reproduce the complete expression pattern of CycE in the eye disc (Figure S4B). These results further support the tissue-specific regulation of cycE (Jones et al., 2000).

Apt Up-regulates Rbf1 and Directs Proper Chromatin Compaction

e2f1 mutant cells induced in the salivary gland displayed small nuclei with low DNA content (Figure 2D). This is exactly expected from the reduced endoreplication. By contrast, nuclei of apt mutant cells were abnormal. Despite the reduced level of endoreplication, the size of nuclei in the apt mutant clone cells was comparable to that of control cells (Figure 2A). The ratio of nuclear size to DNA content was ~6.5 times higher in apt-mutant cells than that in control cells (Figure 4A). A higher-magnification image of apt mutant nuclei exhibited intra-chromosomal chromatin de-compaction and large inter-chromosomal spaces compared with control nuclei (Figure S5). As the loose chromatin is associated with increased transcription activity, loss of apt function would induce de-repression of multiple genes. Indeed, apt-knockout cells exhibited many signals of transcribing RNA polymerase II, under the conditions where the signals were barely detectable in control cells (Figure 4B). Then, what is a global repressor that governs the silencing of many genes in the downstream of Apt? One candidate is Rbf1, because it forms a complex with E2f1 and another complex including E2f2 and Dp to repress E2f-target genes and many other genes (Cayirlioglu et al., 2003; Korenjak et al., 2012; van den Heuvel and Dyson, 2008; Weng et al., 2003). Consistent with this idea, we observed large nuclei with de-compacted chromatin upon RNAi knockdown of rbf1 (Figure 4C). Therefore, it is possible that Apt up-regulates rbf1, and hence the apt mutant cells exhibit large nuclei with de-compacted chromatin. To test the possibility, we analyzed the expression of rbf1 in the salivary gland by RT-qPCR. As shown in Figure 4D, we observed a significant reduction in the expression of rbf1 and e2f2 upon RNAi knockdown of apt leaving the dp expression unchanged. We also examined the expression of Rbf1 protein in apt-knockout or apt-overexpressing cells. Compared with control cells, apt-mutant clone cells showed decreased expression of Rbf1 protein, whereas apt-overexpressing cells exhibited highly compact chromatin with increased Rbf1 protein levels (Figures 4E and 4F). Consistently, we observed de-repression of Rbf1-target genes, such as CG4679, gigas, diap3, and Ipp upon knockdown of apt (Figure 4D). Importantly, overexpression of Rbf1 suppressed chromatin de-compaction upon RNAi knockdown of apt (Figure 4G). Based on these results, we reasoned that the nuclear defects in the apt-knockout cells are due to release from Rbf1-mediated chromatin compaction and de-repression of many Rbf1-target genes.

Figure 4.

Figure 4

Open in a new tab

Apt Regulates Chromatin Compaction through rbf1 in the Salivary Gland

(A) The ratio of nuclear size to DNA content in apt-mutant clone cells, e2f1-mutant clone cells, or control cells. Data were presented as mean ± SD. n = 10 for control, 4 for apt-mutant clone, and 3 for e2f1-mutant clone. ∗∗∗p < 0.001 versus the control (Student's t test).

(B) Immunostaining of control or apt-mutant clone cells with the antibody against transcribing RNA polymerase II (H14). An apt-mutant clone cell (marked by yellow dotted line) showed prominent signals of transcribing RNA polymerase II compared with control cells. n = 3 with all samples showing the represented phenotype. Scale bar, 20 μm.

(C) ptc-GAL4-driven RNAi knockdown of rbf1 induces de-compaction of chromatin. Compared with control cells, rbf1-knockdown cells exhibited larger nuclei with de-compacted chromatin. n = 10 with all samples showing the represented phenotype. Scale bars, 20 μm. Graph shows the ratio of nuclear size to DNA content. Data were mean ± SD relative to control. n = 100 for control and 100 for rbf1 RNAi. The control samples were normalized to 1. ∗∗∗p < 0.001 (Student's t test).

(D) RT-qPCR assays for the expression of dp, rbf, e2f2, and Rbf1-target genes (CG4679, gigas, diap3, Ipp) in dpp > GFP control and dpp > GAL4-driven apt-knockdown salivary glands. Data were average ±SD relative to the dpp > GFP mRNA level. dpp > GFP samples were normalized to 1. ∗∗p < 0.01, ∗∗∗p < 0.001 (Student's t test).

(E) The expression of Rbf1 in apt-mutant clone cells. Arrows indicate the decreased expression of Rbf1 in the clone. n = 4 with all samples showing the represented phenotype. Scale bar, 20μm.

(F) Up-regulation of Rbf1 in apt-overexpression cells. Yellow arrowheads indicate y+-flipped out cells expressing actin-GAL4 that drives overexpression of apt. n = 4 with all samples showing the represented phenotype. Scale bar, 20 μm. fb, fat body.

(G and H) (G) Chromatin de-compaction upon ptc-Gal4-driven RNAi of apt was suppressed by simultaneous overexpression of Rbf1. Upper panels are low-magnification images. Lower panels show higher-magnification images of the nuclei. n = 10 with all samples showing the represented phenotype. Scale bars, 20 μm. Graph shows the ratio of nuclear size to DNA content. Data were mean ± SD relative to control. n = 100 for control, 100 for apt RNAi, and 100 for apt RNAi + Rbf1 overexpression. The control samples were normalized to 1. ∗∗∗p < 0.001 (Student's t-test). (H) ChIP assays showed occupancy of Apt on its binding motif at 156 nucleotides upstream of the transcription start site of rbf1. A control region was set around 500 nucleotides downstream of the Apt-binding motif. Data were presented as mean ± SD. n = three biological replicates. ∗∗∗p < 0.001 versus the control (Student's t test). See also Figure S5.

There exists a single Apt-binding motif at 156 nucleotides upstream of the transcription start site of rbf1. ChIP assays showed occupancy of Apt on the motif (Figure 4H). These data suggest that rbf1 is a direct target of Apt.

Apt Up-regulates Rbf1 Also in Mitotic Cycling Cells

So far, we demonstrate the roles for Apt in the induction of CycE and chromatin compaction for silencing in endocycling salivary gland cells. Then, how is the situation in mitotic cycling cells? We have shown that Apt activates the cycE expression for S phase entry in imaginal disc cells undergoing mitotic cycles (Liu et al., 2014; Wang et al., 2017). This led us to examine whether Apt up-regulates Rbf1 also in the wing disc. Strong knockdown and overexpression of apt induced cell migration and apoptosis, respectively, in the wing disc, which hampered inspection of the nuclear defects. Therefore, we employed mild knockdown or overexpression of apt that was triggered by heat shock-induced flipping out of y+ from actin > y+>GAL4. RNAi knockdown of apt reduced the Rbf1 expression (Figure 5A) and slightly increased the nuclear size/DNA ratio in wing disc cells (Figure 5B). Conversely, overexpression of Apt enhanced the Rbf1 expression (Figure 5C) and reduced the nuclear size/DNA ratio (Figure 5D). These results indicate that Apt up-regulates Rbf1 to direct proper chromatin compaction in wing disc cells also. Collectively, these data suggest that Apt-mediated CycE induction and chromatin compaction are general mechanisms common to both mitotic cycling and endocycling cells.

Figure 5.

Figure 5

Open in a new tab

Apt Regulates the Expression of Rbf1 and Chromatin Compaction in the Wing Disc

(A) Immunostaining of a wing disc harboring apt-knockdown clones with anti-Rbf1 antibody (red). GFP (green) represents the region of y+-flipped out apt-knockdown cells. n = 20 with all samples showing the represented phenotype. Lower panels are close-up images around the apt-knockdown clone. Scale bars, 20 μm.

(B) The wing disc nuclei of control cells (without GFP) and apt-knockdown cells (with GFP) were stained with DAPI. n = 3 with all samples showing the represented phenotype. Red dotted line indicates the boundary between control cells and apt-knockdown cells. Scale bar, 20 μm. Graph shows the ratio of nuclear size to DNA content in the control or apt-RNAi cells. Data were mean ± SD relative to control. n = 55 for control and 32 for apt-RNAi. The control samples were normalized to 1. ∗∗p < 0.01 (Student's t test).

(C) Immunostaining of a wing disc harboring apt-overexpression cells with anti-Rbf1 antibody (red). GFP (green) implies the y+-flipped out apt-overexpressing region. n = 25 with all samples showing the represented phenotype. Scale bar, 20 μm.

(D) The wing disc nuclei of control cells (without GFP) and apt-overexpression cells (with GFP) were stained with DAPI. n = 3 with all samples showing the represented phenotype. Red dotted line indicates the boundary between control cells and apt-overexpression cells. Scale bar, 20 μm. Graph shows the ratio of nuclear size to DNA content. Data were mean ± SD relative to control. n = 40 for control and 52 for apt-overexpression. The control samples were normalized to 1. ∗∗∗p < 0.001 (Student's t test).

Mutual Activation of FSBP and E2f1, and FSBP-Mediated Chromatin Compaction in Mammalian Cells

As Apt, E2f1, and Rbf1 are evolutionarily conserved transcription factors, the aforementioned mechanisms could be also conserved in mammalian cells. To test the possibility, we focused on the mammalian homologs of these factors, FSBP, E2f1, and Rb. In mouse NIH3T3 cells, RNAi knockdown of FSBP significantly decreased the expression of E2f1 and CycE homologs (CCNE1 and CCNE2) (Figure 6A). Upon E2f1 knockdown, the expression of FSBP, CCNE1, and CCNE2 were attenuated (Figure 6A). These data indicate mutual activation between FSBP and E2f1 and requirement of FSBP and E2f1 for the expression of CCNEs. In addition, knockdown of FSBP reduced the expression of Rb1, whereas overexpression of FSBP increased the Rb1 expression (Figure 6B). Consistently, the target genes of Rb (CDC6 and DHFR) also showed up- and down-regulation in FSBP-knockdown and FSBP-overexpressing cells, respectively (Figure 6B). These data indicate FSBP-mediated up-regulation of Rb. Furthermore, each FSBP-knockdown cell exhibited a large nucleus with less compact chromatin and a lower Rb protein level than the control cell did (Figures 6C–6E). Taken together, these results demonstrate Apt- and FSBP-mediated conserved mechanisms for CycE induction and chromatin compaction.

Figure 6.

Figure 6

Open in a new tab

Mutual Activation between FSBP and E2f1, and FSBP-Mediated Chromatin Compaction in Mouse NIH3T3 Cells

(A) RT-qPCR assays for relative mRNA levels of FSBP, E2f1, CCNE1, and CCNE2 in FSBP-knockdown cells or E2f1-knockdown cells. Data were mean ± SD relative to Mock. ∗p < 0.05, ∗∗p < 0.01 (Student's t test).

(B) Relative mRNA levels of FSBP, Rb1, CDC6, and DHFR from FSBP-knockdown cells or ectopic FSBP-expressing cells. Data were mean ± SD relative to Mock. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 (Student's t test).

(C) Control and FSBP-knockdown cells were stained with DAPI (blue) and anti-FSBP antibody (green). n = 81 cells for control and 25 cells for FSBP knockdown with all samples showing the represented phenotype. Scale bar, 20 μm.

(D) Control and FSBP-knockdown cells were stained with DAPI (blue) and anti-Rb1 antibody (red). n = 85 cells for control and 26 cells for FSBP knockdown with all samples showing the represented phenotype. Scale bar, 20 μm.

(E) Quantification of the nuclear size. Data were mean ± SD relative to Mock. n = 16 for control and 6 for FSBP RNAi. ∗∗∗p < 0.001 (Student's t test).

Discussion

This study revealed Apt-mediated up-regulation of two key players in the cell cycle, CycE and Rbf1. What is the significance of this finding? The positive feedback between apt and e2f1 ensures rapid and robust induction of CycE at S phase entry. Apt also up-regulates Rbf1, but Rbf1 is inactivated through phosphorylation with Cdk2. After initiation of S phase, Rbf1 becomes active through de-phosphorylation and represses cycE (Du et al., 1996; Edgar and Orr-Weaver, 2001; van den Heuvel and Dyson, 2008). Together with Crl4Cdt2-mediated degradation of E2f1 (Zielke et al., 2011), this leads to a rapid decline of CycE. Therefore, Apt governs both induction and subsequent repression of cycE with the aid of the periodic phosphorylation and de-phosphorylation of Rbf1.

E2f, a heterodimer of E2f1 and Dp, has been studied for many years, and it is a key regulator of CycE expression for S phase entry (Duronio et al., 1995). However, residual S phase takes place in a null mutant of Drosophila e2f1 or dp (Duronio et al., 1998; Royzman et al., 1997). Here, we solved the discrepancy: another factor Apt also participates in the activation of cycE. Until this study, contribution of “another factor” if any was thought to be rather trivial compared with that of E2f, because e2f1 or dp mutation severely reduced the CycE expression. Our study revealed that the notion is not correct. As Apt and E2f1 up-regulate each other and both Apt and E2f1 are required for the cycE expression, disruption of e2f1 or dp function leads to depletion of both E2f and Apt, which in turn causes a severe defect in cycE expression. This masked the contribution of “another factor” Apt. Actually, transgenic reporter assays indicated that both the Apt-binding sites and the E2f1-binding sites in the regulatory region of cycE are necessary for the normal level of cycE transcription.

apt-Mutant cells induced in the salivary gland exhibited abnormal nuclei. The size of nucleus/DNA content was ~6.5 times higher than that of control cells, which resulted in de-compacted chromatin. Our study suggests that the unusual phenotype is due to release from Rbf1-mediated chromatin compaction and de-repression of Rbf1-target genes that occupy many loci throughout the genome (Korenjak et al., 2012). The nuclear size/DNA content of e2f1-mutant cells was also higher than that of control cells. However, the difference was less prominent than that between apt-mutant cells and control cells. We surmise the following explanation for it. Within a cell, there might exist a balance between the amounts of the E2f1/Dp complex and those of the Rbf1/E2f2/Dp complex. In e2f1-mutant cells, the level of the latter complex would increase in the absence of the former complex. This would compensate the decrease in Rbf1 and E2f2 due to reduced Apt, and would direct toward chromatin compaction.

This study underscores the importance of FSBP, a hitherto not-well-characterized transcription factor. Here we found FSBP- and Apt-mediated up-regulation of Rb and Rbf1, respectively. This raises an intriguing possibility that FSBP (Apt) suppresses tumor metastasis through up-regulation of Rb (Rbf1). Future studies should address the issue experimentally.

Limitation of the Study

We demonstrate here Apt-dependent up-regulation of rbf1. There exists a single Apt-binding motif at 156 nucleotides upstream of the transcription start site of rbf1. ChIP assays showed occupancy of Apt on the motif. Therefore, it is most likely that Apt directly activates rbf1 transcription through the binding site. However, further functional analyses including disruption of the Apt-binding site are necessary to verify the possibility.

Resource Availability

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Susumu Hirose (shirose@nig.ac.jp).

Material Availability

This study did not generate new materials.

Data and Code Availability

This study did not generate new datasets.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Acknowledgments

We are grateful to Yash Hiromi for critical reading of the manuscript. We thank Yoichiro Tamori, Nicholas Dyson, Denise Montell, Vienna Drosophila Resource Center, Fly Stocks of National Institute of Genetics, and Bloomington Stock Center for kindly providing fly lines and reagents. We also thank Shu Kondo and the people in E.S.'s lab for discussions and suggestions. This work was supported by grants from the Japan Society for the Promotion of Science to E.S. (JSPS KAKENHI JP16F16703), the National Key Research and Development Program of China (2017YFE0129800), the National Natural Science Foundation of China (31872971 and 31571502), Funds of “Shandong Double Tops” Program (SYL2017YSTD09), and the Construction Engineering Special Fund of “Taishan Scholars” (no. ts201712022) to Q.-X.L. X.-F.W. was a postdoctoral fellow of JSPS.

Author Contributions

X.-F.W., Q.-X.L., S.H., and E.S. designed the project. S.H. and Q.-X.L. supervised the study. X.-F.W. performed the majority of the experiments and analyzed the data with the following exceptions. J.-X.L. and Z.-Z.Z. did the in vitro cell culture experiments and data analysis. Z.-Y.M. helped with the injection to prepare the transgenic flies. Y.S. and H.-R.Z. helped with a part of the genetic and staining experiments. X.-F.W. wrote the original draft. S.H. and Q.-X.L. revised the manuscript.

Declaration of Interests

The authors declare no conflict of interest.

Published: August 21, 2020

Footnotes

Supplemental Information can be found online at https://doi.org/10.1016/j.isci.2020.101369.

Contributor Information

Qing-Xin Liu, Email: liuqingxin@sdau.edu.cn.

Susumu Hirose, Email: shirose@nig.ac.jp.

Supplemental Information

Document S1. Transparent Methods, Figures S1–S5, and Tables S1–S4
mmc1.pdf (6.7MB, pdf)

References

  1. Brehm A., Miska E.A., McCance D.J., Reid J.L., Bannister A.J., Kouzarides T. Retinoblastoma protein recruits histone deacetylase to repress transcription. Nature. 1998;391:597–601. doi: 10.1038/35404. [DOI] [PubMed] [Google Scholar]
  2. Cayirlioglu P., Ward W.O., Silver Key S.C., Duronio R.J. Transcriptional repressor functions of Drosophila E2F1 and E2F2 cooperate to inhibit genomic DNA synthesis in ovarian follicle cells. Mol. Cell Biol. 2003;23:2123–2134. doi: 10.1128/MCB.23.6.2123-2134.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. De Veylder L., Beeckman T., Beemster G.T., de Almeida Engler J., Ormenese S., Maes S., Naudts M., Van Der Schueren E., Jacqmard A., Engler G. Control of proliferation, endoreduplication and differentiation by the Arabidopsis E2Fa-DPa transcription factor. EMBO J. 2002;21:1360–1368. doi: 10.1093/emboj/21.6.1360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Du W., Vidal M., Xie J.E., Dyson N. RBF, a novel RB-related gene that regulates E2F activity and interacts with cyclin E in Drosophila. Genes Dev. 1996;10:1206–1218. doi: 10.1101/gad.10.10.1206. [DOI] [PubMed] [Google Scholar]
  5. Dulic V., Lees E., Reed S.I. Association of human cyclin E with a periodic G1-S phase protein kinase. Science. 1992;257:1958–1961. doi: 10.1126/science.1329201. [DOI] [PubMed] [Google Scholar]
  6. Duronio R.J., Bonnette P.C., O'Farrell P.H. Mutations of the Drosophila dDP, dE2F, and cyclin E genes reveal distinct roles for the E2F-DP transcription factor and cyclin E during the G1-S transition. Mol. Cell Biol. 1998;18:141–151. doi: 10.1128/mcb.18.1.141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Duronio R.J., O'Farrell P.H., Xie J.E., Brook A., Dyson N. The transcription factor E2F is required for S phase during Drosophila embryogenesis. Genes Dev. 1995;9:1445–1455. doi: 10.1101/gad.9.12.1445. [DOI] [PubMed] [Google Scholar]
  8. Edgar B.A., Orr-Weaver T.L. Endoreplication cell cycles: more for less. Cell. 2001;105:297–306. doi: 10.1016/s0092-8674(01)00334-8. [DOI] [PubMed] [Google Scholar]
  9. Eulenberg K.G., Schuh R. The tracheae defective gene encodes a bZIP protein that controls tracheal cell movement during Drosophila embryogenesis. EMBO J. 1997;16:7156–7165. doi: 10.1093/emboj/16.23.7156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gellon G., Harding K.W., McGinnis N., Martin M.M., McGinnis W. A genetic screen for modifiers of Deformed homeotic function identifies novel genes required for head development. Development. 1997;124:3321–3331. doi: 10.1242/dev.124.17.3321. [DOI] [PubMed] [Google Scholar]
  11. Jones L., Richardson H., Saint R. Tissue-specific regulation of cyclin E transcription during Drosophila melanogaster embryogenesis. Development. 2000;127:4619–4630. doi: 10.1242/dev.127.21.4619. [DOI] [PubMed] [Google Scholar]
  12. Knoblich J.A., Sauer K., Jones L., Richardson H., Saint R., Lehner C.F. Cyclin E controls S phase progression and its down-regulation during Drosophila embryogenesis is required for the arrest of cell proliferation. Cell. 1994;77:107–120. doi: 10.1016/0092-8674(94)90239-9. [DOI] [PubMed] [Google Scholar]
  13. Korenjak M., Anderssen E., Ramaswamy S., Whetstine J.R., Dyson N.J. RBF binding to both canonical E2F targets and noncanonical targets depends on functional dE2F/dDP complexes. Mol. Cell Biol. 2012;32:4375–4387. doi: 10.1128/MCB.00536-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Lau K.F., Perkinton M.S., Rodriguez L., McLoughlin D.M., Miller C.C. An X11alpha/FSBP complex represses transcription of the GSK3beta gene promoter. Neuroreport. 2010;21:761–766. doi: 10.1097/WNR.0b013e32833bfca0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Lie Y.S., Macdonald P.M. Apontic binds the translational repressor Bruno and is implicated in regulation of oskar mRNA translation. Development. 1999;126:1129–1138. doi: 10.1242/dev.126.6.1129. [DOI] [PubMed] [Google Scholar]
  16. Lilly M.A., Spradling A.C. The Drosophila endocycle is controlled by Cyclin E and lacks a checkpoint ensuring S-phase completion. Genes Dev. 1996;10:2514–2526. doi: 10.1101/gad.10.19.2514. [DOI] [PubMed] [Google Scholar]
  17. Liu Q.X., Jindra M., Ueda H., Hiromi Y., Hirose S. Drosophila MBF1 is a co-activator for Tracheae Defective and contributes to the formation of tracheal and nervous systems. Development. 2003;130:719–728. doi: 10.1242/dev.00297. [DOI] [PubMed] [Google Scholar]
  18. Liu Q.X., Wang X.F., Ikeo K., Hirose S., Gehring W.J., Gojobori T. Evolutionarily conserved transcription factor Apontic controls the G1/S progression by inducing cyclin E during eye development. Proc. Natl. Acad. Sci. U S A. 2014;111:9497–9502. doi: 10.1073/pnas.1407145111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Monahan A.J., Starz-Gaiano M. Apontic regulates somatic stem cell numbers in Drosophila testes. BMC Dev. Biol. 2016;16:5. doi: 10.1186/s12861-016-0103-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Nielsen S.J., Schneider R., Bauer U.M., Bannister A.J., Morrison A., O'Carroll D., Firestein R., Clearly M., Jenuwein T., Herrera R.E., Kouzarides T. Rb targets histone H3 methylation and HP1 to promoters. Nature. 2001;412:561–565. doi: 10.1038/35087620. [DOI] [PubMed] [Google Scholar]
  21. Royzman I., Whittaker A.J., Orr-Weaver T.L. Mutations in Drosophila DP and E2F distinguish G1-S progression from an associated transcriptional program. Genes Dev. 1997;11:1999–2011. doi: 10.1101/gad.11.15.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Shen Y., Wang L., Hirose S., Zhou Z., Liu Q. The transcriptional factor Apt regulates neuroblast differentiation through activating CycE expression. Biochem. Biophys. Res. Commun. 2018;499:889–894. doi: 10.1016/j.bbrc.2018.04.012. [DOI] [PubMed] [Google Scholar]
  23. Starz-Gaiano M., Melani M., Wang X., Meinhardt H., Montell D.J. Feedback inhibition of Jak/STAT signaling by apontic is required to limit an invasive cell population. Dev. Cell. 2008;14:726–738. doi: 10.1016/j.devcel.2008.03.005. [DOI] [PubMed] [Google Scholar]
  24. Su M.T., Venkatesh T.V., Wu X., Golden K., Bodmer R. The pioneer gene, apontic, is required for morphogenesis and function of the Drosophila heart. Mech. Dev. 1999;80:125–132. doi: 10.1016/s0925-4773(98)00197-x. [DOI] [PubMed] [Google Scholar]
  25. Talluri S., Dick F.A. Regulation of transcription and chromatin structure by pRB: here, there and everywhere. Cell Cycle. 2012;11:3189–3198. doi: 10.4161/cc.21263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Thurmond J., Goodman J.L., Strelets V.B., Attrill H., Gramates L.S., Marygold S.J., Matthews B.B., Millburn G., Antonazzo G., Trovisco V. FlyBase 2.0: the next generation. Nucleic Acids Res. 2019;47:D759–D765. doi: 10.1093/nar/gky1003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. van den Heuvel S., Dyson N.J. Conserved functions of the pRB and E2F families. Nat. Rev. Mol. Cell Biol. 2008;9:713–724. doi: 10.1038/nrm2469. [DOI] [PubMed] [Google Scholar]
  28. Wang X.F., Shen Y., Cheng Q., Fu C.L., Zhou Z.Z., Hirose S., Liu Q.X. Apontic directly activates hedgehog and cyclin E for proper organ growth and patterning. Sci. Rep. 2017;7:12470. doi: 10.1038/s41598-017-12766-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Weng L., Zhu C., Xu J., Du W. Critical role of active repression by E2F and Rb proteins in endoreplication during Drosophila development. EMBO J. 2003;22:3865–3875. doi: 10.1093/emboj/cdg373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Woodhouse E.C., Fisher A., Bandle R.W., Bryant-Greenwood B., Charboneau L., Petricoin E.F., 3rd, Liotta L.A. Drosophila screening model for metastasis: semaphorin 5c is required for l(2)gl cancer phenotype. Proc. Natl. Acad. Sci. U S A. 2003;100:11463–11468. doi: 10.1073/pnas.2031202100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Zielke N., Kim K.J., Tran V., Shibutani S.T., Bravo M.J., Nagarajan S., van Straaten M., Woods B., von Dassow G., Rottig C. Control of Drosophila endocycles by E2F and CRL4(CDT2) Nature. 2011;480:123–127. doi: 10.1038/nature10579. [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

Document S1. Transparent Methods, Figures S1–S5, and Tables S1–S4
mmc1.pdf (6.7MB, pdf)

Data Availability Statement

This study did not generate new datasets.

Articles from iScience are provided here courtesy of Elsevier

RESOURCES