N-acetyltransferase (nat) is a critical conjunct of photoperiodism between the circadian system and endocrine axis in Antheraea pernyi

Abstract

Since its discovery in 1923, the biology of photoperiodism remains a mystery in many ways. We sought the link connecting the circadian system to an endocrine switch, using Antheraea pernyi. PER-, CLK- and CYC-ir were co-expressed in two pairs of dorsolateral neurons of the protocerebrum, suggesting that these are the circadian neurons that also express melatonin-, NAT- and HIOMT-ir. The results suggest that a melatonin pathway is present in the circadian neurons. Melatonin receptor (MT2 or MEL-1B-R)-ir in PTTH-ir neurons juxtaposing clock neurons suggests that melatonin gates PTTH release. RIA showed a melatonin rhythm with a peak four hours after lights off in adult brain both under LD16:8 (LD) and LD12:12 (SD), and both the peak and the baseline levels were higher under LD than SD, suggesting a photoperiodic influence. When pupae in diapause were exposed to 10 cycles of LD, or stored at 4 °C for 4 months under constant darkness, an increase of NAT activity was observed when PTTH released ecdysone. DNA sequence upstream of nat contained E-boxes to which CYC/CLK could bind, and nat transcription was turned off by clk or cyc dsRNA. dsRNA(NAT) caused dysfunction of photoperiodism. dsRNA(PER) upregulated nat transcription as anticipated, based on findings in the Drosophila melanogaster circadian system. Transcription of nat, cyc and clk peaked at ZT12. RIA showed that dsRNA(NAT) decreased melatonin while dsRNA(PER) increased melatonin. Thus nat, a clock controlled gene, is the critical link between the circadian clock and endocrine switch. MT-binding may release PTTH, resulting in termination of diapause. This study thus examined all of the basic functional units from the clock: a photoperiodic counter as an accumulator of mRNA(NAT), to endocrine switch for photoperiodism in A. pernyi showing this system is self-complete without additional device especially for photoperiodism.

Figure 1. NAT-ir, PER-ir, CYC-ir and CLK-ir in the brain of A. pernyi, early pupal stage.

(A) The topography of immunoreactive cells. Lower-case letters in the map indicate regions shown in the photographs (e.g., b is the site of B). (B) Two PER-ir neurons in the dorsolateral protocerebrum (DL). (C) Two NAT-ir neurons in the adjacent section to B in the DL. (D) CYC-ir in the DL region. (E) Two large CLK-ir neurons in the adjacent section to D in DL region. (F, I, L) CYC/CLK/NAT-ir in the DL region, respectively. (G, J, M) PER-ir in the same sections as (F, I, L) in the DL region, respectively. (H, K, N) Merged image of PER-ir and CYC/CLK/NAT-ir respectively in the DL region. (O) hMT2-ir in the DL regio. (P) PTTH-ir in the DL region. (Q) Merged image of hMT2- and PTTH-ir in the DL region. (L) MT-ir in the DL region. (M) PTTH- ir in the DL region. (N) Merged image of PTTH-ir and MT-ir in the DL region. Scale bars = 100 μm (B–E) and 100 μm (F–N).

Figure 2. Changes in melatonin content and aaNAT activity in the brain-SOG in A. pernyi.

Data are presented as mean ± SEM, n = 4–6 for each time point and treatment. A) Melatonin content at different Zeitgeiber times (ZTs), 2–3 days adults were kept under LD 16∶8 and 12∶12 at 25°C. Melatonin content was measured by RIA. B) Activity of aaNAT after diapause pupae were stored for designated months at 4°C. C) aaNAT activity after diapause pupae were kept for designated cycles of LD 16∶8 The activity was measured by radioenzymatic assay at pH 7.6 and 8.6. *P<0.05, **P<0.01.

Figure 3. A diagramatic figure showing cis-elements in the promotor region of nat.

Two perfect E-boxes (CACGTG) in orange, 4 canonical E-boxes (CANNTG) in pink, 2 CREs (NNNCGTCA) in red and 2 TATA motifs (TATAAA or TATAAAA) in blue were recognized.

Figure 4. Real time PCR analyses of Ap nat, Ap clk and Ap cyc transcriptional levels.

Transcriptional rhythms of the three genes have the same acrophase at Zt 12. *P<0.05, **P<0.01, ***P<0.001.

Figure 5. Photoperiodic terminations of diapause puape after RNAi against nat.

dsRNAs against Ap nat (dsRNA^NAT) or GFP (dsRNA^GFP) were injected into the pupae. Nuclease-free water (NFW) is used as negative control (Cont.). The control and treated pupae were placed under LD 16∶8 or 12∶12. A) NFW- injected group and dsRNA^GFP- injected group responded to long-day photoperiod producing moth emergence at 100% after 40 days under long-day condition but dsRNA^NAT- injected group stayed in diapause for 40 days even under LD 16∶8. B) On the contrary, under LD 12∶12 all groups produced moth emergence not more than 20% after 40 days. n = 25–30 for each treatment.

Figure 6. Effect of dsRNAs injections against some circadian clock genes on nat transcription.

Data are presented as mean ± SEM, n = 5–8 for each time point and treatment. Pupae were maintained under LD 12∶12. A) dsRNAs of cyc, clk, and GFP were synthesized and injected to diapause pupae. NFW is used as negative control (Cont.). dsRNA^CYC and dsRNA^CLK suppressed nat transcription. B) dsRNA^PER up-regulated nat transcriptional levels after 24 hours from injection of dsRNA^PER. *P<0.05, **P<0.01, ***P<0.001.

Figure 7. Effect of dsRNA against per on adult emergence under LD and SD.

Pupae were injected with dsRNA^PER, dsRNA^GFP and NFW (-negative control) (Cont.). n = 25–30 for each treatment. A) Diapause was terminated after injection of dsRNA^PER even under short-day (LD 12∶12). B) Injection of dsRNA^PER showed no difference from the controls under long-day (LD 16∶8).

Figure 8. Effect of injection of dsRNAs against nat and per on melatonin level.

Data are presented as mean ± SEM, n = 15 for each time point and treatment. Silencing nat decreased melatonin content at 48 hours after injection dsRNA^NAT, while knocking down per upregulated melatonin level at 24 hours after dsRNA^PER injection. *P<0.05, ***P<0.001.