Juvenile Hormone Activates the Transcription of Cell-division-cycle 6 (Cdc6) for Polyploidy-dependent Insect Vitellogenesis and Oogenesis

Abstract

Although juvenile hormone (JH) is known to prevent insect larval metamorphosis and stimulate adult reproduction, the molecular mechanisms of JH action in insect reproduction remain largely unknown. Earlier, we reported that the JH-receptor complex, composed of methoprene-tolerant and steroid receptor co-activator, acts on mini-chromosome maintenance (Mcm) genes Mcm4 and Mcm7 to promote DNA replication and polyploidy for the massive vitellogenin (Vg) synthesis required for egg production in the migratory locust (Guo, W., Wu, Z., Song, J., Jiang, F., Wang, Z., Deng, S., Walker, V. K., and Zhou, S. (2014) PLoS Genet. 10, e1004702). In this study we have investigated the involvement of cell-division-cycle 6 (Cdc6) in JH-dependent vitellogenesis and oogenesis, as Cdc6 is essential for the formation of prereplication complex. We demonstrate here that Cdc6 is expressed in response to JH and methoprene-tolerant, and Cdc6 transcription is directly regulated by the JH-receptor complex. Knockdown of Cdc6 inhibits polyploidization of fat body and follicle cells, resulting in the substantial reduction of Vg expression in the fat body as well as severely impaired oocyte maturation and ovarian growth. Our data indicate the involvement of Cdc6 in JH pathway and a pivotal role of Cdc6 in JH-mediated polyploidization, vitellogenesis, and oogenesis.

Introduction

In addition to repressing insect larval metamorphosis, juvenile hormone (JH)² has an essential role in stimulating adult reproduction (2, 3). During the larval stages, JH maintains the larval characteristics of insects by modulating the cellular responses to 20-hydroxyecdysone (20E) during each molting. In the final-instar larvae, the very low titer or the absence of JH leads to 20E-induced metamorphosis (2, 4). In adult insects, newly synthesized JH stimulates many aspects of reproduction, including the previtellogenic development, vitellogenesis, and oogenesis (3, 5). Cumulative studies have demonstrated that JH exerts the genomic actions through its receptor, methoprene-tolerant (Met) (6). JH induces the heterodimerization of Met with steroid receptor co-activator (SRC) (also known as Taiman in Drosophila or FISC in the mosquito, Aedes aegypti) to form a transcriptionally active complex to regulate the transcription of target genes in several insect systems, including the beetle Tribolium castaneum, silkworm Bombyx mori, migratory locust Locusta migratoria, and mosquito A. aegypti(1, 7–11).

Vitellogenesis, vitellogenin (Vg) synthesized in the fat body of many insects and taken up by maturing oocytes, plays a critical role in egg production. In Drosophila melanogaster both JH and 20E are involved in vitellogenesis, although 20E is responsible for the high rate of Vg synthesis in the fat body (4, 12, 13). In the mosquito A. aegypti, JH controls the previtellogenic development of fat body competence for Vg synthesis (14, 15). In the red flour beetle Tribolium castaneum, JH regulates Vg synthesis in the fat body, whereas 20E affects Vg synthesis through oocyte maturation (16–18). In many other insect species including the linden bug Pyrrhocoris apterus and the German cockroach Blattella germanica as well as Locusta migratoria, JH acts independently of 20E to stimulate vitellogenesis and oocyte maturation (3, 5, 19–21). Knockdown of Met results in dramatic reduction of Vg expression and blocks oocyte maturation in T. castaneum, P. apterus, and L. migratoria (11, 17, 21). Despite this understanding of JH regulation, the molecular mechanisms of JH action in insect vitellogenesis and oocyte maturation remain poorly understood.

Polyploidy, the existence of more than two genome copies in a cell, is found in highly metabolically active cells and tissues like fat body, follicular epithelium, nurse cells, midgut, salivary gland, and wing imaginal discs of insects (22–25). Polyploidy is generated by repeated G/S cycles and enhanced DNA replication (26, 27). It has been reported that 20E regulates DNA replication and polyploidy during insect metamorphosis, but the underlying mechanisms have yet been defined (3, 28–31). In locusts, polyploidization in the fat body and follicle cells of adult females during vitellogenesis and oocyte maturation is dependent on JH (1, 32, 33). In a previous report we show that JH acts through its receptor complex Met-SRC on two mini-chromosome maintenance (Mcm) genes, Mcm4 and Mcm7, to promote DNA replication and polyploidization for the massive Vg synthesis required for egg production in locusts (1).

At the onset of the G₁ phase, recruitment of the replicative helicase Mcm2–7 onto origins of DNA replication to form the prereplication complex requires the loading factor cell-division-cycle 6 (Cdc6), a member of AAA⁺ ATPase family (34–38). Along with the origin recognition complex and Cdt1 (Cdc10 protein-dependent transcript 1), Cdc6 loads Mcm proteins onto origins of replication to facilitate the formation of stable prereplication complex in G₁ phase, thereby licensing these sites to initiate DNA replication in S phase (39–41). The loading of Mcm helicase complex onto DNA at the origins of replication is central to DNA replication (42–44). In our RNA-seq-based gene expression profiling, Cdc6 together with Mcm2–7 were identified as the up-regulated genes in the fat body of JH-deprived adult female locusts further treated with methoprene (1). Given the functional importance of Cdc6 in loading Mcm2–7, we speculated that Cdc6, like Mcm, is a regulatory target of the JH pathway. We, therefore, wondered if JH and its receptor complex target Cdc6 for transcriptional regulation, which consequently modulates locust polyploidy, vitellogenesis, and oogenesis. We found that JH stimulates the expression of Cdc6, and that JH-induced Met-SRC complex directly activates Cdc6 transcription by binding to the upstream consensus sequence with the E-box-like motif. We observed that depletion of Cdc6 results in substantial reduction of Vg expression, arrested oocyte maturation, and blocked ovarian growth. This work implicates a crucial role of Cdc6 in JH-dependent polyploidy, vitellogenesis, and oogenesis, which provides new insight into the mechanisms of JH regulation in insect reproduction.

Experimental Procedures

Insects

The gregarious colony of migratory locust L. migratoria was maintained at a density of ∼300 locusts per cage (25 cm × 25 cm × 25 cm) under a photoperiod of 14 h light:10 h dark and at 30 ± 2 °C. Locusts were fed with wheat bran supplied continuously and wheat seedlings provided once daily. The wheat seedlings were grown 5–7 days on the soil-less culture under a photoperiod of 14 h light:10 h dark and at 22 ± 2 °C (45).

Hormone Treatment

JH-deprived female adult locusts were obtained by inactivation of the corpora allata via the topical application of 500 μg (100 μg/μl dissolved in acetone) precocene III (Sigma) to the dorsal neck membrane of locust within 12 h after eclosion (46). To restore the JH activity, an active JH analog, S-(+)-methoprene was topically applied at 150 μg (30 μg/μl dissolved in acetone) per locust 10 days post precocene treatment (46). Topical application of acetone (5 μl per locust) alone was used as the solvent control.

RNA Isolation and qRT-PCR

Total RNA from locust tissues and Drosophila Schneider 2 cells (S2 cells) was extracted using TRNzol reagent (Tiangen) following the manufacturer's instruction. First-strand cDNA was reverse-transcribed with 2 μg of total RNA using FastQuant RT kit with gDNase (Tiangen). qRT-PCR was performed using the Mx3005P detection system (Agilent) and RealMasterMix SYBR Green kit (Tiangen) initiated at 95 °C for 15 min, then 40 cycles at 95 °C for 10 s followed by 58 °C for 20 s and 72 °C for 30 s. Melting curve analysis was conducted to confirm the specificity of amplification. The 2^−ΔΔCt method was used for calculating the relative gene expression levels, normalized by β-actin. Primers for qRT-PCR are listed in Table 1. The specificity of primers was confirmed by BLAST in the NCBI database and the locust genome (47). The qRT-PCR products were sequenced for further confirmation of specificity.

TABLE 1.

Primers used for qRT-PCR and RNAi

Dm, D. melanogaster.

Gene	Forward primer	Reverse primer	Product
			bp
qRT-PCR
Cdc6	CGTGCGTTAGACATTGGA	GGTTGGCTGGATTAGATTCA	109
VgA	CCCACAAGAAGCACAGAACG	TTGGTCGCCATCAACAGAAG	99
Met	CCACTTACAGGCTTGCTA	GCCCTTCTTCACCTTCTT	144
β-actin	AATTACCATTGGTAACGAGCGATT	TGCTTCCATACCCAGGAATGA	73
DmMet	CGTCCTTAGATTCGCCACCC	GAGGCAGACATACCCGTTCC	108
DmGce	CTCAGTCCCTTCACCTTCAT	ACCTTGTTCGTCTCCTTGTC	182
DmTaiman	AGCGATGTAAAGCCCGAGA	AAAGCAGCATTCCACCCAC	127
Dmβ-actin	ACTTCTGCTGGAAGGTGGAC	ATCCGCAAGGATCTGTATGC	138
RNAi
Cdc6	CATTGGACGCCGTGTTCTTGA	TGCAGCAACTTCTTCCTCATCC	466
GFP	CACAAGTTCAGCGTGTCCG	GTTCACCTTGATGCCGTTC	527
Met	TTAGGGCAGCATCAGAAAG	TCGTCGGGAGGAAGTGTAT	421
DmMet	CTGCCAACTATCCGATTGTCTC	CTCTCGCCGTAGTCACTGTT	454
DmTaiman	AGCATCAGCACCAGCATCA	GTCGTTGTCGTAGAGTTGTTGT	508

RNA Interference (RNAi)

cDNA templates were amplified by PCR, cloned into pGM-T easy vector (Tiangen), and confirmed by sequencing. Double-stranded RNA (dsRNA) was then synthesized by in vitro transcription with T7 RiboMAX Express RNAi system (Promega) according to the manufacturer's manual. For RNAi in locusts, female adults within 12 h after eclosion were intra-abdominally injected with 8 μg (in a volume of 8 μl) of Met (GenBank^TM accession number KF471131) or Cdc6 (GenBank^TM accession number KT692979) dsRNA dissolved in a mixture of acetone and H₂O (2:1 ratio) and boosted on day 5. In JH rescue experiments, methoprene or acetone (solvent control) was applied on day 6, and the effects were examined on day 8. For RNAi in S2 cells, Drosophila Met (FlyBase: FBpp0073368) and Taiman (FlyBase: FBpp0292873) dsRNA (38 nm) were transfected into S2 cells using Lipofectamine 2000 (Invitrogen) for 48 h before transfection of the recombinant vectors. In all RNAi experiments, dsGFP was used as the control. Primers used for dsRNA synthesis are included in Table 1. The specificity of primers was confirmed by BLAST in the NCBI database and locust genome (47).

Tissue Imaging and Confocal Microscopy

Ovaries and ovarioles were imaged with a Nikon D7000 camera and an Olympus CKX41 microscope, respectively. The length and width of primary oocytes were measured using Leica M205C microscope. For cell staining, fat bodies and ovarioles were fixed in 4% paraformaldehyde for 15 min and permeabilized in 0.1% Triton X-100 for additional 15 min. F-actin was stained with 0.165 μm Alexa-Fluor 488 phalloidin (excitation wavelength 488 nm) (Invitrogen). Nuclei were stained with 5 μm Hoechst 33342 (excitation wavelength 350 nm) (Sigma). The images were captured by ZEISS LSM 710 confocal microscopy and analyzed with ZEN2010 software (Carl Zeiss).

Flow Cytometry

The fat body, follicle epithelium, and brain were separately homogenized in a Dounce homogenizer. Cells were collected by centrifugation (800 × g), fixed in 70% ethanol overnight, and further incubated with PBS buffer containing 50 μg/ml propidium iodide (Sigma), 100 μg/ml RNaseA (Promega), and 0.2% Triton X-100 for 2 h at 4 °C. The cells were then filtered by 300-mesh cell strainers (BD Falcon) and analyzed using a BD FACSCalibur Flow Cytometry System with Flowjo 7.6.1 software (BD Biosciences). Brain nuclei were used as a diploid control.

Western Blot and Immunoprecipitation

The procedures were described previously (1). Briefly, locust Met (nt 1–3108) and SRC (GenBank^TM accession number KF471132; nt 1–1786) cDNA were cloned into pAc5.1/FLAG and pAc5.1/V5 vectors (Invitrogen), respectively. The constructs of pAc5.1/FLAG-Met (N-terminally tagged) and/or pAc5.1/SRC-V5 (C-terminally tagged) were transfected into Drosophila S2 cells using Lipofectamine 2000 (Invitrogen). 10 μm JH III or methoprene was used for cell treatment. Cells were lysed with 50 mm Tris-HCl, pH 7.5, 150 mm NaCl, 2 mm EDTA, 1 mm DTT, 1% Nonidet P-40, 1 mm PMSF, 1 mm NaF, and a protease inhibitor mixture (Roche Applied Science). After centrifugation at 14,000 × g for 10 min, lysates were fractionated on 8% SDS-PAGE and transferred to PVDF membranes (Millipore). Western blots were carried out using anti-FLAG and anti-V5 antibodies (Medical & Biological Laboratories, Co.), with anti-actin antibody (Abmart) as the loading control. For immunoprecipitation, the precleared lysates were incubated with anti-V5 antibody for 60 min at 4 °C, and the immunocomplexes were captured with protein A-agarose (Sigma) and eluted in Laemmli sample buffer followed by Western blotting with anti-FLAG antibody.

Luciferase Reporter Assay

The upstream promoter region (nt −1253 to −60) of the Cdc6 gene was cloned into pGL4.10 vector (Promega) and confirmed by sequencing. After pretreatment with Drosophila Met and Taiman dsRNA for 48 h, S2 cells were transfected with pAc5.1/FLAG-Met and/or pAc5.1/SRC-V5 plus pGL4.10-Cdc6^{−1253 to −60} using Lipofectamine 2000 (Invitrogen). After 48 h, 10 μm methoprene was applied for 6 h. Luciferase activity was measured using Dual-Luciferase Reporter Assay System and GloMax 96 Microplate Luminometer (Promega).

Electrophoresis Mobility Shift Assay (EMSA)

Nuclear extracts from locust fat bodies and S2 cells were isolated using NE-PER Nuclear and Cytoplasmic Extraction Reagents kit (Thermo Scientific). Cdc6 probe (5′-CGAGAAACACGCGAAAAATA-3′) was end-labeled with [γ-³²P]ATP by T4 DNA kinase (New England BioLabs), purified by Sephadex G-25 column (GE Healthcare), and incubated with nuclear protein extracts in the binding buffer containing 10 mm Tris-HCl, pH 7.5, 50 mm NaCl, 1 mm MgCl₂, 1 mm DTT, 1 mm EDTA, 10% glycerol, and 50 ng/μl poly(dI/dC). In the competition assays, a 50× molar excess of unlabeled Cdc6 probe or nonspecific AP2 oligonucleotide (Promega) (48) was added into the binding reaction. In the supershift assays, anti-FLAG antibody (Sigma), anti-V5 (Invitrogen) antibody, or the control IgG (Sigma) was preincubated with the nuclear extracts at 4 °C for 1 h before the addition of labeled probes. The DNA-protein complex was resolved in 5% native polyacrylamide gels and visualized using x-ray film (Eastman Kodak Co). The band intensity was quantified by ImageJ.

Statistical Analysis

Student's t test by SPSS 20.0 software was used for statistical analyses. Significant difference was considered at p < 0.05. Values were reported as the mean ± S.E.

Results

Cdc6 Expression Responds to JH during Locust Vitellogenesis and Oogenesis

A single Cdc6 gene without isoforms was identified in the locust genome and transcriptome (47, 49). In RNA-seq-based gene expression profiling, the expression level of Cdc6 was increased by 9.8-fold in JH-deprived locust fat bodies further treated with methoprene for 24 h (1). Under our rearing conditions the first gonadotropic cycle of female adult locusts was ∼10–12 days, and vitellogenesis started from ∼5 days post adult eclosion (PAE). We first investigated the spatiotemporal expression patterns of locust Cdc6. qRT-PCR using total RNA from adult females collected at 8 days PAE showed that Cdc6 was highly expressed in the ovary (Ov) followed by the fat body (Fb), whereas Cdc6 mRNA levels were much lower in the head (He), cuticle (Cu), midgut (Mg), thorax muscle (Tm), and back leg (Bl) (Fig. 1A). Knowing that Cdc6 was more abundant in the ovary and fat body, two key tissues involved in the reproduction of adult female locusts, we next analyzed the expression profiles of Cdc6 in the ovary and fat body of female adult locusts from 0 to 8 days PAE. Compared with that at the day of adult eclosion (0 day PAE), Cdc6 mRNA levels in the fat body were sharply increased by ∼7-fold at 2–4 days PAE, declined but remained significantly elevated (1.9-fold) on day 6, and then reached a peak (9.7-fold) on day 8. Similarly, Cdc6 mRNA levels in the ovary were increased by 4.2-fold on day 2, remained high on days 4 and 6 (3.1-fold and 1.9-fold, respectively), and further increased by 6.1-fold on day 8 (Fig. 1B). As locust hemolymph JH titers are undetectable at eclosion but increase significantly in the previtellogenic stage and rise to a peak in the vitellogenic phase (50, 51), an increase of Cdc6 mRNA levels in both fat body and ovary appeared to correlate with the phase of elevated JH titers.

FIGURE 1.

Cdc6 expression and response to juvenile hormone in adult female locusts. A, relative mRNA levels of Cdc6 in seven selected tissues from adult females at 8 days PAE. Cdc6 mRNA levels in the fat body were used as the calibrator. Fb, fat body; Ov, ovary; Cu, cuticle; Mg, midgut; Th, thorax muscle; Bl, back leg; He, head. Different letters indicate significant difference at p < 0.05. n = 12–16. B, developmental profiles of Cdc6 abundance in the fat body and ovary of adult females from the day of eclosion (0 PAE) to 8 days PAE. Cdc6 mRNA levels at 0 PAE were arbitrarily set to 1.0. *, p < 0.05; **, p < 0.01; ***, p < 0.001. n = 12–16. C and D, relative mRNA levels of Cdc6 in the fat body (C) and ovary (D) of adult females treated with precocene for 10 days (P) and those further treated with methoprene or acetone (solvent control) for 6–48 h. PAE10, 10-day-old adult females as the positive control. *, p < 0.05 and **, p < 0.01. n = 6–8.

To explore the dynamics of JH-stimulated Cdc6 expression, qRT-PCR was conducted using total RNA from the fat body and ovary of precocene-treated female adults for 10 days as well as those further treated with methoprene for 6, 12, 24, and 48 h. As shown in Fig. 1, C and D, chemical allatectomy by precocene treatment resulted in 59 and 91% reduction of Cdc6 mRNA levels in the fat body and ovary, respectively. We next assessed the effects of JH application on Cdc6 expression using JH-deprived adults further treated with methoprene for 6–48 h. Compared with JH-deprived fat bodies, Cdc6 mRNA levels were significantly increased by 2.1-fold at 6 h and then continually elevated by 6.1–13.8-fold at 12–48 h post methoprene treatment (Fig. 1C). In the ovary, Cdc6 mRNA levels were increased by 2.3–2.8-fold at 6–24 h post methoprene application but declined at 48 h (Fig. 1D). In the parallel experiment of solvent controls, acetone treatment had no significant effect on Cdc6 expression in fat bodies or ovaries (Fig. 1, C and D). The data indicate that JH stimulates the expression of Cdc6 in both the fat body and ovary of locusts.

Cdc6 Is Transcriptionally Regulated by the JH-receptor Complex

As an initial step, Met RNAi was carried out to determine the requirement of Met for JH-dependent Cdc6 expression using fat bodies as the representative tissue. qRT-PCR demonstrated that 61–85% of Met RNAi efficiency was obtained in the fat body of dsMet-injected adult females at 4–8 days PAE (Fig. 2A). Correspondingly, Cdc6 mRNA levels were reduced by 61, 46, and 36% at 4, 6, and 8 days PAE, respectively (Fig. 2B), indicating the dependence of Cdc6 expression on Met. Analysis of the upstream sequence of locust Cdc6 gene revealed an E-box-like motif (CACGCG, nt −1063 to −1058) that has been previously reported for Met binding as a JH-response element (1, 8, 48, 52, 53). To test the recognition of Cdc6 DNA element by Met, EMSA was conducted using nuclear extracts from dsMet- versus dsGFP-treated fat bodies. A 20-mer nucleotide probe corresponding to the sequence containing E-box-like motif in the upstream of locust Cdc6 was used (Fig. 2C). Two bands were visualized when the [³²P]Cdc6 probe and nuclear extracts derived from dsGFP-treated fat bodies were incubated, but only the faster moving band was abolished with 50× molar excess of unlabeled Cdc6 probe (Fig. 2D). This faster moving band showed 34% reduction in intensity when nuclear extracts from Met-depleted fat bodies were used (Fig. 2, D and E). It suggests the possible involvement of endogenous Met in the Cdc6 probe binding complex.

FIGURE 2.

Responsiveness of Cdc6 expression to Met. A, Met knockdown efficiency in the fat body on day 4–8. **, p < 0.01 and ***, p < 0.001 compared with the respective dsGFP controls. n = 8. B, effect of Met knockdown on Cdc6 expression in the fat body on day 4–8. *, p < 0.05 compared with the respective dsGFP controls. n = 8. C, alignment of DNA sequences containing the E-box-like motif in the upstream of Kr-h1, Rps28, and Early trypsin (ET) from the mosquito A. aegypti (Aa) (8, 48, 53), Kr-h1 from Drosophila melanogaster (Dm) (52), Mcm4 and Cdc6 from the locust Locusta migratoria (Lm) (1). D, EMSA using the [³²P]Cdc6 probe and fat body nuclear protein extracts from dsMet- or dsGFP-treated adult females on day 8. A representative experiment is shown. The arrow indicates the specific band. FP, free probe. E, quantitative analysis of the intensity of specific band by ImageJ. *, p < 0.05. n = 3.

As locust cell line and Met antibody are unavailable, we performed luciferase reporter assays and EMSA using S2 cells to further confirm Met binding to the Cdc6 promoter. We cloned cDNAs of locust Met and SRC into pAc5.1/FLAG and pAc5.1/V5 vectors, respectively, to express the FLAG-Met and SRC-V5 fusion proteins in S2 cells. To diminish the effect of endogenous Met, Gce (germ cell-expressed; the paralog of Met) and Taiman (Tai), S2 cells were subjected to RNAi using Drosophila Met and Tai dsRNA (1) before transfection of the recombinant vectors. It is noted that the sequence of Drosophila Met dsRNA shares ∼40% identity to that of Drosophila Gce (FlyBase: FBpp0292296). Pretreatment of S2 cells by Drosophila Met and Tai dsRNA resulted in an ∼80% reduction of endogenous Met, Gce, and Tai expression (Fig. 3A) but had no significant effect on transfected FLAG-Met and SRC-V5 (Fig. 3B). Immunoprecipitation and Western blot demonstrated that the addition of JH III or methoprene induced the interaction of expressed FLAG-Met and SRC-V5 (Fig. 3C), indicating the dependence of JH on the formation of locust Met and SRC heterodimer. It has been previously reported that Cdc6 is transcriptionally regulated in an E2F-dependent manner (54, 55). Two putative E2F-binding sites were found in the proximal region (nt −59 to −1) of locust Cdc6 gene. To eliminate the possible interference by endogenous E2F, the Cdc6 upstream sequence from nt −1253 to −60 containing the E-box-like motif was cloned into the pGL4.10 vector and co-transfected with pAc5.1/FLAG-Met and/or pAc5.1/SRC-V5 for luciferase assays. In the absence of methoprene, the co-expression of FLAG-Met and SRC-V5 slightly induced Cdc6 reporter activity, similar to that of FLAG-Met or SRC-V5 expression alone (Fig. 3D). However, after methoprene treatment, FLAG-Met plus SRC-V5 led to 5.2-fold increase in Cdc6 reporter activity compared with the control (Fig. 3D). The data indicate that JH-induced Met-SRC complex activates Cdc6 transcription.

FIGURE 3.

Transcriptional regulation of Cdc6 by the JH-receptor complex. A, RNAi efficiency of Drosophila Met, Gce, and Tai in S2 cells treated with Drosophila Met and Tai dsRNA (dsDmMet+dsDmTai), compared with the dsGFP controls. *, p < 0.05 and **, p < 0.01. n = 3. B, effect of dsDmMet+dsDmTai treatment on the expression of FLAG-Met (Met) and SRC-V5 (SRC) in S2 cells. n = 3. C, Western blot (WB) and immunoprecipitation (IP) showing the expression of FLAG-Met (second panel from the top) and SRC-V5 (third panel from the top) in S2 cells, and the interaction of FLAG-Met and SRC-V5 in the presence of JH III or methoprene (upper panel). α-FLAG, anti-FLAG antibody; α-V5, anti-V5 antibody; α-Actin, anti-actin antibody. D, luciferase reporter assay using S2 cells transfected with pGL4.10/Cdc6^{−1253 to −60} alone (Control), pGL4.10/Cdc6^{−1253 to −60} + pAc5.1/FLAG-Met (Met), pGL4.10/Cdc6^{−1253 to −60} + pAc5.1/SRC-V5 (SRC), pGL4.10/Cdc6^{−1253 to −60} + pAc5.1/FLAG-Met + pAc5.1/SRC-V5 (Met+SRC). Me, methoprene (10 μm). E, EMSA using the [³²P]Cdc6 probe and nuclear protein extracts from S2 cells with expressed FLAG-Met and SRC-V5 and treated with 10 μm methoprene. AP2, a nonspecific oligonucleotide. The arrow indicates the specific band. FP, free probe.

In EMSA using methoprene-treated nuclear extracts from S2 cells with expressed FLAG-Met and SRC-V5, a band composed with the [³²P]Cdc6 probe was eliminated by a 50× molar excess of the unlabeled Cdc6 probe (Fig. 3E). When a 50× molar excess of unlabeled AP2 oligonucleotide was used, the specific binding was not competed (Fig. 3E). This specific band was diminished when cell nuclear extracts were preincubated with anti-FLAG or anti-V5 antibody (Fig. 3E). When IgG was preincubated with the cell nuclear extracts, this band was reduced in intensity but not abolished (Fig. 3E). These results confirm that the JH-receptor complex binds to the 20-mer DNA sequence of Cdc6 promoter with specificity.

Cdc6 Knockdown Inhibits Polyploidization in the Fat Body and Follicle Cells

In the fat body of adult female locusts, 76% knockdown efficiency of Cdc6 RNAi was obtained at 4 days PAE (Fig. 4A). On days 6 and 8, Cdc6 mRNA levels in the fat body were reduced by 74 and 62%, respectively (Fig. 4A). It has been previously reported that efficient gene knockdown in locust ovary is unachievable via intra-abdominal injection of dsRNA dissolved in H₂O due to inefficient dsRNA uptake into follicle cells and oocytes (56). In the present study, dsRNA was dissolved in a mixture of acetone and H₂O at the ratio of 2:1 to facilitate dsRNA uptake. As shown in Fig. 4B, Cdc6 mRNA levels in the ovary were significantly reduced to 57 and 44% that of its normal levels at 6 and 8 days PAE, respectively. However, Cdc6 expression was not significantly altered on day 4 (Fig. 4B).

FIGURE 4.

Cdc6 RNAi efficiency in the fat body and ovary. A, Cdc6 RNAi efficiency in the fat body (Fb) of adult females at 4–8 days post adult eclosion. B, Cdc6 RNAi efficiency in the ovary (Ov) of adult females at 4–8 days post adult eclosion. *, p < 0.05 compared with the respective dsGFP controls; n.s., no significant difference. n = 8.

To visualize the morphological change of nuclei after Cdc6 RNAi, F-actin and nuclei were stained with fluorescence-labeled phalloidin and Hoechst 33342, respectively, followed by confocal microscopy. Knockdown of Cdc6 resulted in smaller nuclei in both fat body and follicle cells (Fig. 5, A and B). Notably, when Cdc6 was depleted by RNAi, ∼15% of fat body cells and ∼21% of follicle cells were seen with double nuclei accompanied with insignificant change of cell numbers on day 6–8. The observation suggests that locust Cdc6 has a role in controlling the G₂-M transition in addition to the initiation of DNA replication.

FIGURE 5.

Cdc6 knockdown reduces ploidy in the fat body and follicle cells. A and B, morphology of fat body (A) and follicle cells (B) of dsCdc6- versus dsGFP-treated adult females at 4–8 days post adult eclosion. Fb, fat body; Fc, follicle cells; Blue, nuclei; green, F-actin. Yellow arrows indicate cells with double nuclei. White bar, 20 μm. C and D, flow cytometry analysis showing the DNA contents in the fat body (C) and follicle cells (D) of dsCdc6- versus dsGFP-treated adult females at 4–8 days post adult eclosion. Six locusts were used in each analysis.

Quantitative analysis of ploidy by flow cytometry showed that Cdc6-depleted fat body and follicle cells had markedly lower DNA contents compared with the dsGFP controls (Fig. 5, C and D). At 4 days PAE, Cdc6-knockdown fat bodies showed 2C and 4C peaks as well as 8C populations, whereas dsGFP-treated fat bodies had 2C and 8C peaks and 4C populations. Distinct from peaks at 8C and 16C in dsGFP controls, only 2C and 4C peaks were observed in Cdc6-depleted fat bodies on day 6. At 8 days PAE, dsCdc6-treated fat bodies had dominantly 4C contents compared with the dsGFP control with 8C and 16C peaks (Fig. 5C). With respect to follicle cells, Cdc6-knockdown samples were chiefly at 4C on days 6–8, whereas the dsGFP controls had 8C populations on day 6 and 8C plus 16C populations or peak on day 8 (Fig. 5D).

Cdc6 RNAi Blocks Locust Vitellogenesis and Oogenesis

The migratory locust has two Vg genes, VgA and VgB, that are coordinately induced by JH and expressed in similar patterns (57). VgA (GenBank^TM accession number KF171066) was selected as the representative to evaluate the effect of Cdc6 knockdown on Vg expression in the fat body (1). Knockdown of Cdc6 reduced VgA mRNA levels to 47, 34, and 4% that of its control levels on days 4, 6, and 8, respectively (Fig. 6A). We next examined the effect of Cdc6 knockdown on oocyte maturation and ovarian growth. Depletion of Cdc6 via RNAi resulted in blocked oocyte maturation and arrested ovarian development (Fig. 6, B and C). Consequently, the primary oocytes and ovaries of Cdc6-depleted locusts remained small on days 6–8. Conversely, the primary oocytes and ovaries of dsGFP controls markedly enlarged (Fig. 6B). Statistically, the length × width index of primary oocytes of dsGFP control locusts were increased from 0.9 to 6.3 on days 4–8, whereas that of Cdc6 RNAi locusts had no significant change (Fig. 6C). It must be noted that the impaired oocyte maturation and ovarian growth were also seen in Cdc6 RNAi locusts at 4 days PAE, which was presumably due to declined synthesis of Vg and other forms of yolk proteins as the efficient Cdc6 knockdown was achieved in the fat body on day 4 (Fig. 4A).

FIGURE 6.

Cdc6 knockdown blocks vitellogenesis and oogenesis. A, VgA expression levels in fat bodies of dsCdc6-treated adult females at 4–8 days post adult eclosion. *, p < 0.05 compared with the respective dsGFP controls. n = 8. B, comparison of primary oocytes (Po), ovarioles (Ol), and ovaries (Ov) of dsCdc6- and dsGFP-treated adult females at 4–8 days post adult eclosion. Scale bars: white, 5 mm; black, 0.5 mm. C, statistical analysis for length × width index of primary oocytes of dsCdc6- and dsGFP-treated adult females at 4–8 days post adult eclosion. ***, p < 0.001 compared with the respective dsGFP controls. n = 25–30.

Next, we treated Cdc6 RNAi locusts with methoprene and examined the effect on ploidy, Vg expression, oocyte maturation, and ovarian growth. As shown in Fig. 7, A and B, further application of methoprene on Cdc6-depleted locusts did not restore the defective nuclei and decreased ploidy of either fat body or follicle cells to their normal levels. However, after methoprene treatment, Cdc6-RNAi locusts showed a decrease of the 2C population and an increase of 8C contents in the fat body and follicle cells (Fig. 7B). These observations suggest that JH regulates other factors involved in the stimulation of polyploidization and that the increase in ploidy may be time-sensitive (Fig. 6). The percentage of cells with double nuclei did not change significantly in either fat body or follicle cells of methoprene-treated locusts that were previously subjected to Cdc6 RNAi. As shown in Fig. 7C, the capacity of methoprene to induce VgA expression in the fat body was abrogated by Cdc6 knockdown. Similarly, methoprene application on dsCdc6-treated female locusts was unable to restore the defective oocyte maturation and ovarian growth to the normal levels (Fig. 7, D and E). Taken together, these data indicate a crucial role of Cdc6 in polyploidization of fat body and follicle cells as well as vitellogenesis, oocyte maturation, and successful egg production in locusts.

FIGURE 7.

Methoprene treatment is unable to rescue the defective phenotypes resulted from Cdc6 RNAi. Cdc6 dsRNA was injected within 12 h post adult eclosion and boosted on day 5. Methoprene or acetone was applied on day 6, and the effects were examined on day 8. A, morphology of fat body (Fb) and follicle cells (Fc) of adult females injected with dsGFP further treated with acetone (dsGFP), dsCdc6 further treated with acetone (dsCdc6), or dsCdc6 further treated with methoprene (dsCdc6+Me). Blue, nuclei; green, F-actin. Yellow arrows indicate cells with double nuclei. Scale bar, 20 μm. B, flow cytometry analysis showing DNA contents in the fat body (Fb) and follicle cells (Fc) of three experimental groups. Six locusts were used in each analysis. C, relative VgA expression in the fat body of three groups. *, p < 0.05 compared with the dsGFP+acetone control (dsGFP). n.s., no significant difference. n = 8. D, comparison of primary oocytes (Po), ovarioles (Ol), and ovaries (Ov) of three groups. Scale bars: white, 5 mm; black, 0.5 mm. E, statistical analysis for length × width index of primary oocytes of three groups. ***, p < 0.001 compared with the dsGFP+acetone control (dsGFP). n = 25–30.

Discussion

Cdc6 and JH-dependent Vitellogenesis and Oogenesis

In the present study we have demonstrated that depletion of Cdc6 resulted in lower ploidy and significantly reduced Vg expression in the fat body as well as blocked oocyte maturation and arrested ovarian growth, similar to that caused by knockdown of Mcm4 or Mcm7 (1). Moreover, JH treatment of Cdc6-depleted locusts did not restore the defective phenotypes to the normal levels, resembling the failure of JH rescue on Mcm4 or Mcm7 RNAi (1). These data provide the evidence that Cdc6, like Mcm4 and Mcm7, is essential for JH-dependent polyploidization, vitellogenesis, and egg production. It has been well established that Cdc6 is a key factor for replication origin licensing late in G₁ phase by regulating the formation of prereplication complexes that initiate DNA synthesis at S phase (37, 43, 58, 59). Polyploidy is known to promote transcriptomic and metabolomic outputs (27, 60). It is likely that JH stimulates the expression of Cdc6, along with Mcm and genes coding for other DNA replication factors, to replicate multiple copies of the genome for massive yolk protein synthesis in the fat body required for the production of a number of matured eggs. In humans, Cdc6 has been linked to oncogenesis through its role in DNA replication initiation and its interference with the tumor suppressor genes INK4/ARF (61–63). Overexpression of Cdc6 in human primary cells promotes DNA hyperreplication related to oncogene activation, whereas Cdc6 knockdown prevents cell proliferation and promotes apoptosis (62, 64).

Follicle cells, constituting the follicular epithelium that surrounds the oocyte, play important functions in oocyte development (3, 5, 20). The follicular epithelium not only determines the size and shape of ovarian follicle but also initiates the intercellular spaces, known as patency, to facilitate the transport of Vg to the oocyte membrane where Vg is internalized into maturing oocytes by receptor-mediated endocytosis (3, 14, 65). In this study, efficient RNAi of Cdc6 in locust ovary was achieved using an alternative approach via dissolving dsRNA in the mixture of acetone and H₂O. We observed that disruption of Cdc6 in the ovary led to decreased ploidy in follicle cells as well. The impaired polyploidization in follicle cells is likely to restrain the development of follicular epithelium, which in turn blocks patency initiation and consequently suppresses Vg transportation and ovarian growth.

Intriguingly, in synchrony with declined ploidy, ∼15% of fat body cells and 21% of follicle cells had double nuclei after Cdc6 knockdown, whereas the cell numbers were not significantly changed. These phenotypes suggest that Cdc6-depleted fat body and follicle cells are missing checkpoints to prevent mitotic entry in addition to decreased DNA replication. Studies in the yeast, Drosophila, Xenopus, and mammalian cells have shown that Cdc6 activates and maintains the checkpoint response to prevent mitosis before DNA replication is completed (34, 62, 66). Cdc6 RNAi in synchronized Drosophila S2 cells results in the entrance of cells into mitosis with incompletely replicated DNA (67). Our data, therefore, support the dual functions of Cdc6 in initiation of DNA synthesis and in S-M phase transition.

Regulation of Cdc6 Expression by JH

Cdc6 expression is tightly regulated at the level of transcription in a cell cycle-dependent manner (43, 62, 63). Androgen can act through its receptor binding to a 15-bp palindromic androgen response element in the Cdc6 promoter on the transcription of Cdc6 gene in prostate cancer cells (55, 68). By qRT-PCR, we showed that Cdc6 expression was significantly increased in the fat body and ovary post adult female eclosion, which are in conformity with the elevated JH titers (50, 51). Methoprene treatment of JH-deprived locusts induced Cdc6 expression, whereas Met RNAi resulted in a remarkable decrease of Cdc6 abundance. Because Cdc6 was expressed in response to JH and is Met-dependent in S2 cells and an E-box-like motif was identified in the upstream promoter region of Cdc6 gene, we reasoned that Met could directly target Cdc6 for transcriptional regulation. Our luciferase reporter assays demonstrated that Cdc6 transcription was activated by the JH-receptor complex comprised of Met and SRC in the presence of methoprene. EMSA using nuclear extracts from locust fat bodies and Drosophila S2 cells with expressed FLAG-Met and SRC-V5 documented the specific binding of JH-receptor complex to the DNA sequence containing E-box-like motif in the upstream of Cdc6 gene. Collectively, JH and its receptor appear to directly target the components of DNA replication machinery, including Cdc6, Mcm4, and Mcm7 (1), to coordinately initiate DNA replication for the multiple sets of chromosomes.

In summary, our present study has demonstrated that JH acts through its receptor complex Met-SRC to regulate the transcription of Cdc6. Loss of Cdc6 function can result in significantly decreased ploidy and precocious mitotic entry in both fat body and follicle cells accompanied by substantial reduction of Vg expression and arrested development of follicular epithelium. Consequently, oocyte maturation and ovarian growth are blocked. Upon JH induction, Cdc6 along with Mcm and possibly other factors involved in DNA replication and cell cycle coordinate to replicate the genome for multiple copies in the fat body and follicle cells for the massive synthesis and efficient uptake of Vg and possibly other macromolecules required for producing a large number of eggs. Further identification and characterization of genes responsible for JH-dependent regulation of polyploidy during vitellogenesis and oogenesis should help decipher the mechanisms of JH-modulated reproduction and high fecundity in insects.

Author Contributions

S. Z. conceived the study and wrote the paper. Z. W. performed and analyzed the experiments and wrote the paper. W. G. designed and performed the experiments. Y. X. performed the experiments. All authors reviewed the results and approved the final version of the manuscript.