Eggs of M. brassicae were obtained from a laboratory colony maintained at the National Institute of Agrobiological Sciences, Japan. The larvae were reared on an artificial diet of “Insecta LFS” (Nihon Nosan Kogyo, Yokohama, Japan) at 25°C under a 14 hr light:10 hr dark photoperiod (long-day conditions) or at 23°C under a 10 hr light:14 hr dark photoperiod (short-day conditions). The animals under the long-day and short-day conditions entered pupal diapause at rates of 0% (0/145) and 98.6% (138/140), respectively.

Total RNA was extracted from 10 pupal brains using TRIzol reagent (Invitrogen). Single-stranded cDNA for PCR was synthesized from the total RNA with Superscript III reverse transcriptase (Invitrogen). PCR was performed using the following degenerate primers:

sense primer, 5′-(A/G)T(G/T)GATTA(C/T)G(A/C)(A/T)AA(C/T)ATGA-3′;

antisense primer, 5′-TA(A/G)TC(C/T)CT(C/G)GT(A/G)CA(C/G/T)A(C/T)ACA-3′.

These degenerate primers were designed from consensus sequences obtained after the alignment of previously determined PTTH cDNA sequences for H. virescens (GenBank accession number: AY172671), H. armigera (AY286543), Helicoverpa zea (AY172670) and Helicoverpa assulta (AY780526). The PCR was performed under the following conditions: 94°C for 5 min and 30 cycles of 94°C for 30 s, 45°C for 1 min and 72°C for 1 min, with a final extension step of 7 min at 72°C. The amplified approximately 400-bp fragments were gel purified, cloned into a pGEM-T vector (Promega) and sequenced. The entire sequence of MabPTTH was determined by 5′- and 3′-RACE followed by nested PCR, using the SMART RACE cDNA Amplification Kit (Clontech). The products of the 5′ and 3′ nested PCR were sequenced, and the obtained sequence and its deduced amino acid sequence were analyzed and aligned using GENETIX ver. 5.2 (GENETYX). The primers used were as follows:

5′ RACE primer, 5′-GCCAACGCTGATGGGACGATATTCACC-3′;

3′ RACE primer, 5′-CCGGATCCTTCCTGCGCTTGTAACCC-3′;

5′ nested primer, 5′-CCATGCACTGGTATGGCCAGGGGCATGAC-3′;

3′ nested primer, 5′-CTGGGACGAAACACGTTCCCGCAACGCG-3′.

The sequence reported in this paper has been deposited in the GenBank database (accession no. AB748456).

Whole-mount in situ hybridization was performed as previously described [21], with DIG-labeled RNA probes synthesized using the obtained 400-bp DNA fragment as a template.

The recombinant MabPTTH was expressed as previously described [22], with some modifications. The isolated MabPTTH cDNA was amplified with restriction enzyme sites, Nde I and BamH I, by PCR and then inserted into pET3b vector. The initial methionine residue is involved in the Nde I restriction site and is fused with the predicted first amino acid residue of the mature MabPTTH, glycine. The construct was then transformed into the E. coli strain BL21, and recombinant MabPTTH was induced by 1 mM IPTG for 7 h.

Mouse monoclonal antibodies (D17 and D18) and a rabbit polyclonal antibody were generated against the recombinant MabPTTH monomer using standard methods. These antibodies were purified using Protein A affinity chromatography and stored at a concentration of 1 mg/ml. The same rabbit anti-ecdysone antiserum has been used previously [16].

Whole-mount immunohistochemistry was performed as described [21] with the following modifications: collagenase treatment was omitted, the incubation times with the primary and secondary antibody were 16 hr and 4 hr, respectively, and the anti-MabPTTH mouse monoclonal antibody (D17) was used as the primary antibody.

Hemolymph was sampled at 12-hr intervals, immediately after lights-on and 12 hr after that point, and it was then processed as described previously [16].

To immobilize the 20-hydroxyecdysone (20E) on an EIA plate (Costar, 3590), its conjugate with ovalbumin was produced. Twenty-hydroxyecdysone was conjugated with ovalbumin using the ketone group at C-6 and carboxymethoxylamine, a bifunctional amino acid spacer, as described [23]. The wells of the EIA plates were incubated with 50 µl of 20E-ovalbumin conjugate solution (0.1 µg/ml in 50 mM Tris-HCl buffered saline (TBS)) overnight at 4°C, followed by blocking with 100 µl of 4% skimmed milk in TBS for 1 h at 25°C. After washing the wells with TBS containing 0.05% Tween-20, 50 µl of serially diluted 20E or the test samples were distributed to the wells, followed by the addition of 50 µl of 1∶100,000 diluted anti-ecdysone rabbit antiserum. The test samples, 20E and the antiserum were diluted with 0.5% bovine serum albumin in TBS. After overnight incubation at 4°C, the wells were washed, and the rabbit antibody bound to the immobilized 20E was quantified using the DELFIA system as described [16]. Because 20E was used as the standard hormone, the titers were expressed as 20E equivalents.

TR-FIA for MabPTTH was essentially the same as that for BomPTTH [16]. In the present assay, an anti-MabPTTH mouse monoclonal antibody (D18) was immobilized on the EIA plate, and an anti-MabPTTH rabbit antibody was used as the detection antibody. The brain extract was used as the standard hormone, as we had no purified native PTTH or correctly folded recombinant PTTH dimer. Brains of ND day-1 wandering larvae were dissected in insect saline and homogenized in 100 µl cold TBS containing 4-amidinophenylmethansulfonyl fluoride hydrochloride and using Bioruptor UCD-250, an ultrasonic homogenizer (Cosmo Bio, Tokyo, Japan). The homogenate was heated at 70°C for 5 min, cooled in an ice-water bath, and centrifuged at 15,000×g for 10 min. The supernatant was diluted with dilution buffer and used as the PTTH standard for TR-FIA, with one brain equivalent of PTTH defined as 1 unit.

Quantitative RT-PCR was performed as described previously [24]. For the absolute quantification of mRNAs, serial dilutions of plasmids containing the cDNAs of MabPTTH and RpL8 [25] were used for standards. After the molar amounts were calculated, the transcript levels of the MabPTTH were normalized with the RpL8 levels in the same samples. The primers used in this analysis were as follows:

MabPTTH sense primer, 5′-GACGAGTACATGCTAGAAGACCAG-3′;

MabPTTH antisense primer, 5′-TCAGGTTGAATGGAATCCGTGTCG-3′;

RpL8 sense primer, 5′-ATCAAGGGTGTCGTGAAGGACATC-3′;

RpL8 antisense primer, 5′-CAGTAGACAAACTGGCCAGTGTAC -3′.

The brain extract was prepared as described above and concentrated by ultrafiltration over a 10 kDa cut-off membrane (Ultrafree, Millipore). The PTTH titer of the condensed brain extract was determined by TR-FIA, and the extract was then stored at −80°C. For PTTH absorption, the condensed brain extract was incubated with 1.4 µg per 10 units PTTH of anti-MabPTTH or control rabbit antibody and 3.5 µl of Protein G-coupled agarose gel (Roche) for 1 hr. The supernatant after centrifugation was used as the PTTH-absorbed brain extract.

Prior to injection, the pupae were anesthetized by submersion in water for 1 hr. The brain extract was then injected into the vertex of the head. The wounds were sealed with melted paraffin.

A small piece of thoracic integument to which the PG is attached (henceforth called the PG) was cut out in insect saline as previously described [26] and preincubated for 30 min in Grace’s Insect Medium. The PGs were cultured for 3 hr at 25°C in the medium (200 µl) containing varying doses of PTTH (the >10 kD brain extract prepared as above). The ecdysteroids released into the medium were measured by TR-FIA.

We isolated the entire sequence of the MabPTTH cDNA by homology-based degenerate PCR and RACE strategies (see the Materials and Methods section for details). The predicted open reading frame encodes a 223-amino acid protein representing preproPTTH (Fig. 1A). This precursor protein exhibits a high homology to the corresponding proteins of other noctuid moths (Fig. 1B), with 69% identity to the H. armigera PTTH precursor [20]. The protein consists of a putative signal peptide (28 a.a.), a precursor domain (87 a.a.) and a PTTH monomer peptide (108 a.a.). Seven cysteine residues are present within the PTTH monomer at conserved locations, as in other known PTTHs [20]. In situ hybridization revealed that MabPTTH is expressed in two dorsolateral neurosecretory cells in each brain hemisphere (Fig. 1C), as are the PTTH genes of other lepidopteran insects [15], [22], [27].

The MabPTTH cDNA was expressed in E. coli, but the product was a mixture of the misfolded monomer or oligomer of PTTH. Despite much effort with various methods, we were unable to obtain a correctly folded PTTH dimer, which is the biologically active form of PTTH. Therefore, the mixture was reduced and alkylated, then purified for use as an immunogen to produce the MabPTTH antibodies that were necessary for developing the MabPTTH immunoassay. All of the mouse monoclonal antibodies (D17 and D18) and the rabbit polyclonal antibody generated against this peptide were found to immunostain the same neurosecretory cells in the brain that were labeled by in situ hybridization (Fig. 1D). These neurosecretory cells project their axons to the corpora allata. These results indicate that in M. brassicae, PTTH is produced in two pairs of lateral brain neurosecretory cells and is released from the corpora allata, as in other lepidopteran insects [28].

The larvae were reared under either long-day (nondiapause-inductive) or short-day (diapause-inductive) conditions. In this study, the animals reared under the former and the latter conditions are designated as nondiapause-destined (ND) and diapause-destined (DD) animals, respectively.

The shutdown of ecdysteroid secretion in diapausing pupae has been demonstrated in many insects. To confirm this shutdown for M. brassicae, hemolymph ecdysteroid titers were measured in both photoperiodic conditions (Fig. 2A,B). The overall pattern of the changes in the hemolymph ecdysteroid titer before pupal ecdysis was similar between the two types of animals, although the peak titer in the DD larvae was less than half the peak value in the ND larvae: this titer was slightly increased around the onset of wandering and peaked 2 days before pupation. In contrast, the titers after pupation differed remarkably between the ND and DD animals: the titer in the ND animals steeply increased 2 days after pupation, forming a large peak 3 days later, whereas no increase in the ecdysteroid titer was observed in the DD animals.

A very sensitive assay for MabPTTH based on a TR-FIA protocol was developed, and the PTTH titers in the hemolymph of both ND and DD animals were determined (Fig. 2C,D). In the ND animals, the titer increased remarkably 2 days before pupal ecdysis. Although it dropped soon after that, the titer increased again at the time of pupal ecdysis, then fluctuated at high levels over the following week. The PTTH titer in the DD animals also increased before pupation, but the peak titer was only half the peak value of the ND animals. Following this peak, the titer gradually decreased and became undetectable 1 day after pupation. Interestingly, the first notable difference in the PTTH titer between the ND and the DD animals was observed during the early wandering stages, when the titers were significantly lower in the DD animals (two tailed t-test, p<0.01).

We next measured the PTTH mRNA and protein levels in the brain to determine what is responsible for the cessation of PTTH secretion in the DD animals. The mRNA level determined by quantitative RT-PCR was relatively high until the early wandering stage, when it decreased in both types of animals (Fig. 3A,B). Although the overall pattern of changes was similar, a notable difference was found between the ND and the DD animals after the decline of the mRNA level: a moderate level of PTTH mRNA was maintained until the early pupal stages in the ND animals, whereas the level was low in the DD animals, especially after pupation. This result indicates that PTTH gene expression is attenuated in the DD animals after the wandering stage.

However, the measurement of the brain PTTH content revealed that the reduced PTTH gene expression is not responsible for the cessation of PTTH secretion in the DD animals, because the PTTH content was actually higher in the DD animals than in the ND animals during the period from the early last instar to the second pupal day (Fig. 3C,D). This result clearly indicates that the cessation of PTTH secretion in the DD pupae is due not to a paucity of PTTH but rather to an attenuation of the secretory activity of the PTTH neurons.

The cessation of PTTH secretion observed after pupation in the DD animals strongly suggested that PTTH serves as a switch between continuous development and pupal diapause. To test this interpretation, we injected a brain extract into the DD pupae immediately after ecdysis. The brain extract did induce adult development in a dose-dependent manner (Fig. 4A), demonstrating that the absence of some brain-derived factor in the hemolymph is the only cause of pupal diapause. To identify this brain factor, we removed PTTH from the brain extract by immunoprecipitation, then injected the extract into DD pupae of the same stage. The PTTH-absorbed brain extract failed to induce adult development (Fig. 4B), leading us to conclude that the diapause-averting factor in the brain is indeed PTTH.

The successful induction of adult development in the DD pupae by PTTH injection suggested that the DD pupal PGs are as sensitive to PTTH as the ND pupal PGs, at least immediately after pupal ecdysis. To confirm this prediction, we investigated the dose-dependent effects of PTTH on the ecdysteroid secretion of day-0 pupal PGs in vitro (Fig. 4C). The basal activity of the PG, measured in the absence of PTTH, tended to be lower in the DD pupae than in the ND pupae. However, ecdysteroid secretion by the DD pupal PGs significantly increased in the presence of PTTH at 1 unit/ml or higher doses. By contrast, the ND pupal PGs did not significantly respond to PTTH at a concentration of 2 units/ml or less. Interestingly, at 4 units/ml of PTTH, the secretion was three times higher in the ND pupal PGs than in the DD pupal PGs. These results suggest that the DD pupal PG is sensitive enough to respond to PTTH, but the maximal rate of ecdysteroid secretion of this PG is less than that of the ND pupal PG.

Because a single injection of PTTH (brain extract) could block the initiation of diapause, we asked if the long-lasting release of PTTH observed in the ND pupae is essential for pupa-adult development. It was conceivable that the injected PTTH might have directly or indirectly stimulated the brain to release PTTH through a positive-feedback mechanism. Therefore, the changes in the PTTH and ecdysteroid titers in the hemolymph after the PTTH injection were monitored for 4 days (Fig. 4D). The PTTH titer increased immediately after injection but then declined rapidly, with no signs of re-elevation by day 4. In contrast, the ecdysteroid titer increased steadily following the injection. In the saline-injected control, no significant changes in either the PTTH or the ecdysteroid titer were observed.