Steroid hormone ecdysone deficiency stimulates preparation for photoperiodic reproductive diapause

Abstract

Diapause, a programmed developmental arrest primarily induced by seasonal environmental changes, is very common in the animal kingdom, and found in vertebrates and invertebrates alike. Diapause provides an adaptive advantage to animals, as it increases the odds of surviving adverse conditions. In insects, individuals perceive photoperiodic cues and modify endocrine signaling to direct reproductive diapause traits, such as ovary arrest and increased fat accumulation. However, it remains unclear as to which endocrine factors are involved in this process and how they regulate the onset of reproductive diapause. Here, we found that the long day-mediated drop in the concentration of the steroid hormone ecdysone is essential for the preparation of photoperiodic reproductive diapause in Colaphellus bowringi, an economically important cabbage beetle. The diapause-inducing long-day condition reduced the expression of ecdysone biosynthetic genes, explaining the drop in the titer of 20-hydroxyecdysone (20E, the active form of ecdysone) in female adults. Application of exogenous 20E induced vitellogenesis and ovarian development but reduced fat accumulation in the diapause-destined females. Knocking down the ecdysone receptor (EcR) in females destined for reproduction blocked reproductive development and induced diapause traits. RNA-seq and hormone measurements indicated that 20E stimulates the production of juvenile hormone (JH), a key endocrine factor in reproductive diapause. To verify this, we depleted three ecdysone biosynthetic enzymes via RNAi, which confirmed that 20E is critical for JH biosynthesis and reproductive diapause. Importantly, impairing Met function, a component of the JH intracellular receptor, partially blocked the 20E-regulated reproductive diapause preparation, indicating that 20E regulates reproductive diapause in both JH-dependent and -independent manners. Finally, we found that 20E deficiency decreased ecdysis-triggering hormone signaling and reduced JH production, thereby inducing diapause. Together, these results suggest that 20E signaling is a pivotal regulator that coordinates reproductive plasticity in response to environmental inputs.

Fig 1. Changes in 20E production in the female adults of C. bowringi under photoperiodic conditions.

(A) Determination of 20E titer in the hemolymph of the LD- and SD-induced C. bowringi at 4 days post eclosion (PE). (B) 20E biosynthesis pathway in insects. (C) The heat map of expression patterns of 20E biosynthetic genes at 0, 2 and 4 days PE in the LD- and SD-induced females. (D) Tissue expression patterns of Spo, Phm, Sad and Shd in SD-induced females at 4 days PE. HE, head; MG, midgut; FB, fat body; OV, ovary. The relative gene expression levels in tissues are presented as fold changes compared to the heads. (E) Determination of mRNA levels of Spo, Phm, Sad and Shd in the ovaries of the LD- and SD-induced females from 0 to 4 days PE. The relative expressions of genes at different time points are presented as fold changes compared to the LD females at 0 days PE. Different letters above bars indicate significant between-group differences determined by one-way ANOVA followed by Tukey’s LSD test (α = 0.05). Error bars represent the sd. Asterisks indicate significant differences determined by an Independent-Samples t-test. *P < 0.05, **P < 0.01. Nvd, neverland; Dib, disembodied.

Fig 2. 20E signaling promotes ovarian development under SD condition.

(A) The LD-induced females were injected with 1μg of 20E at 0 days PE, and the induction of reproduction in response to 20E signaling was tested after 4 days of injection. (B) The development grades and (C) sizes of ovaries were analyzed after treated with 20E. (D) Effect of 20E injection on the expression of Vg1 and Vg2 mRNA in the fat bodies. The relative gene expression levels in LD+20E sample are presented as fold changes compared to LD+Control. (E) The reproductive females were microinjected with dsGFP or dsEcR at 0 days PE, and representative phenotypes of ovaries were imaged by a stereomicroscope at 4 days PE. (F) The development grades and (G) sizes of ovaries were analyzed after treated with dsEcR. (H) Relative abundance of EcR, Vg1 and Vg2 mRNA in the fat bodies after treated with dsEcR. The relative gene expression levels in SD+dsEcR sample are presented as fold changes compared to SD+dsGFP. (I) Representative examples of ovarian development were determined on the 4^th day after 20E biosynthetic genes’ RNAi in SD-induced females. Error bars represent the sd. *P < 0.05, **P < 0.01.

Fig 3. 20E signaling suppresses lipid accumulation of diapause preparation.

(A) Nile red staining was performed in the fat bodies of the LD-induced female adults after 4 days of 20E injection and (B) relative intensity of fluorescence of lipid droplets quantitatively was analyzed by using ImageJ. Meanwhile, the (C) relative triglyceride (TG) content and (D) total lipid content were detected in the whole body at 4 days PE. At 0 days PE, the reproductive females were microinjected with dsGFP or dsEcR to study the lipid accumulation in the fat body at 4 days PE using (E) Nile red staining and (F) relative intensity of fluorescence of lipid droplets quantitatively analyzed by using ImageJ. Statistical analysis for (G) relative TG content and (H) total lipid content after EcR RNAi knockdown. (I) The numbers of DEGs, expression patterns of DEGs and top 20 KEGG pathways were displayed. (J) Heat map representing transcripts of genes involved in lipolysis and lipogenesis at the 2^nd day after 20E injection in the LD-induced females. Orange dots represent the DEGs. Transcripts of lipogenesis and lipolysis genes were determined in the fat bodies at the 4^th days after (K) 20E injection and (L) EcR knockdown. Relative gene expression levels represent the fold changes between (K) LD+Control and LD+20E samples and (L) SD+dsGFP and SD+dsEcR treatments. (M) Nile red staining of fat bodies and (N-P) quantitative analysis of fat contents after TGL1 RNAi. Error bars represent the sd. *P < 0.05, **P < 0.01.

Fig 4. Effects of 20E signaling on the JH signaling in the female adults of C. bowringi under photoperiodic conditions.

(A) Top 20 pathways identified in KEGG pathway analysis in DEGs. (B) JH biosynthesis pathway in insects, which is divided into early (mevalonate pathway) and late (JH branch) steps. (C) The heat map represented transcripts of JH biosynthesis and JH-inducible genes at the 2^nd days after 20E injection in the LD-induced females. Orange dots represent the DEGs. (D) JH titer measurement in the hemolymph of the LD-induced female adults at the 4^th days after 20E injection, as determined by LC-MS/MS. N.D., not detected. (E) Expression levels of JH biosynthetic genes were detected in the heads at 4 days PE after injection with 20E. (F) The relative abundance of JH-inducible genes mRNA in the fat bodies after treated with 20E injection. Relative gene expression values in the LD+20E sample are presented as fold changes compared to LD+Control. (G) Detection of JH titer in the hemolymph of SD-female 4 days after injection with either dsGFP or dsEcR on the day of eclosion. (H) The expression of JH biosynthetic genes in the heads were analyzed by qRT-PCR after EcR RNAi. (I) The differential expression of JH-inducible genes was determined in the fat bodies after dsEcR injection. JHAMT1, JHAMT2 and Kr-h1 transcript levels in the heads after silencing with (G) Spo, (K) Sad and (L) Shd. Relative gene expression values in the RNAi samples are presented as fold changes compared to the SD+dsGFP control. Error bars represent the sd. *P < 0.05, **P < 0.01.

Fig 5. 20E promotes SD-induced ovarian development by targeting JH signaling.

(A) The LD-induced females were treated with dsMet following a 24 h induction with 1μg 20E to investigate the induction of reproduction in response to 20E-JH signaling. (B) The development grades and (C) sizes of ovaries were analyzed following dual-injection with dsMet and 20E. (D) The expression of Vg1 and Vg2 were detected in the fat bodies after injection with dsMet and 20E. Relative gene expression values represent the fold changes between LD+dsMet+Control and LD+dsMet+20E. (E) Representative phenotypes of ovaries from SD females injected with dsEcR and JHA. (F) The development grades and (G) sizes of ovaries were analyzed after treated with dsEcR and JHA. (H) Relative abundance of Vg1 and Vg2 mRNA in the fat bodies after injection with dsEcR and JHA. Relative gene expression values represent the fold changes between SD+dsEcR+acetone and SD+dsEcR+JHA. Error bars represent the sd. *P < 0.05, **P < 0.01.

Fig 6. Knocking down ETH arrests ovarian development and decreases JH signaling under SD condition.

(A) Tissue distribution of ETH in SD-induced females at 4 days PE. HE, head; MG, midgut; AFBT, abdominal fat bodies attached with trachea; OV, ovary. Relative gene expression values in the tissue samples are presented as fold changes compared to the ovary samples. (B) Patterns of ETH expression detected by qRT-PCR in the in the AFBT tissues on 0–4 days after eclosion. Relative expression levels of genes at different time points are presented as fold changes compared to the LD females at 4 days PE. (C) The transcriptional expression of ETH was analyzed after 20E injection for 6, 48 and 96 h in the AFBT tissues of the LD-induced females. (D) qRT-PCR of ETH in the AFBT tissues of SD-females injected with dsEcR, analyzed at 48 and 96 h. (E-G) Effect of knockdown of ETH in female on ovarian development and ovary length and width. (H) Relative expression levels of Vg1 and Vg2 after dsETH injection measured by qRT-PCR in the fat bodies. (I) Reduction of JH titer in SD-females following ETH silencing. qRT-PCR measurement of (J) ETH, (K) JH biosynthetic genes and Kr-h1 expression levels were analyzed in the heads after dsETH treatment. The relative gene expression levels in the treatments are presented as fold changes compared to the control groups. Different letters above bars indicate significant between-group differences determined by one-way ANOVA followed by Tukey’s LSD test (α = 0.05). Error bars represent the sd. *P < 0.05, **P < 0.01.

Fig 7. The regulation of lipid accumulation by 20E and JH signaling pathways under photoperiodic conditions.

(A) Nile red staining was performed in the fat bodies of the LD-induced female adults after dual-injection with dsMet and 20E. (B) Relative intensity of fluorescence of lipid droplets quantitatively analyzed by using ImageJ. Meanwhile, (C) the relative TG content and (D) total lipid content were detected in the whole body at 4 days PE. (E) At 0 days PE, the reproductive females were microinjected with dsEcR following a 24 h induction with 30 μg of JHA to study the lipid accumulation in the fat body at 4 days PE using Nile red staining. (F) Relative intensity of fluorescence of lipid droplets quantitatively analyzed by using ImageJ. Statistical analysis for (G) relative TG content and (H) total lipid content after dual-injection with dsEcR and JHA. Relative levels of key genes related to lipolysis and lipogenesis transcripts in the fat bodies after (I) dual-injection with dsMet and 20E, as well as (J) dsEcR and JHA. Relative gene expression values in the treatment group are presented as fold changes compared to the control groups. Error bars represent the sd. *P < 0.05, **P < 0.01.

Fig 8. Model for 20E signaling in photoperiodic reproduction and diapause of C. bowringi females.

20E signaling under the short-day (SD; 12 h light, 12 h dark) condition, results in ovarian development via ETH-JH signaling. 20E-regulated JH signaling also inhibits lipid storage to block diapause preparation. Under the long-day (LD; 16 h light, 8 h dark) condition, 20E and ETH signaling is shut down and JH biosynthesis is inactivated, resulting in accumulation of lipid stores. In addition, 20E could suppress lipid accumulation and might be independent of JH signaling. Arrows represent promotion and T-bars represent inhibition. Gray and black represent the OFF and ON activity states of the genes or physiological processes, respectively.