Biosynthetic Activity of Corpora Allata, Growth of Sex Accessory Glands and Mating in the Male Moth Agrotis Ipsilon (Hufnagel)

In male moths, sexual attraction and mating occur only when the female sexual pheromone is detected by the antennae and the information is subsequently integrated in the central nervous system (for a review, see Hansson, 1995). During mating, a spermatophore originating in the sex accessory glands (SAGs) of the male is transferred into the female. The SAGs of insects grow and produce proteinaceous material that is associated with sperm production and transfer (see Happ, 1992) and has various modulatory functions on female behaviour (Leopold, 1976). In most insect species, growth and protein synthesis in the SAGs seem to be regulated by both ecdysteroids and juvenile hormone (JH), with development and differentiation being under the control of ecdysteroids and protein secretion being regulated by JH (see Gillott and Gaines, 1992).

This general scheme is well documented in many insect orders including the butterflies Danaus plexippus, Vanessa cardui and Nymphalis antiopa among the Lepidoptera (Herman, 1973; Herman and Bennett, 1975; Herman and Dallmann, 1981). In moths, nothing is known about the role of JH in SAG maturation. The corpora allata of all male moths studied so far are known to produce JH acids (JHAs) instead of JHs (Peter et al. 1981; Bhaskaran et al. 1988; Cusson et al. 1993; Ho et al. 1995). In a few cases, such as in Hyalophora cecropia and Manduca sexta, SAGs have been shown to contain JHA methyltransferase (JHAMT) activity which enables the conversion of JHAs into JH (Bhaskaran et al. 1988). However, little is known of the role of either JHA or JH in male reproduction in moths.

In the noctuid moth Agrotis ipsilon, mating does not occur until a few days after eclosion (Swier et al. 1976, 1977; Gadenne, 1993). Moreover, the presence of the CA is required for the male to respond to the sex pheromone (Gadenne et al. 1993). Allatectomy suppressed male responsiveness, and both JHA and JH were able to restore the sexual behaviour of the operated males (Duportets et al .1996). The CA of male A. ipsilon produce JHAs (Duportets et al. 1996), and no JHAMT has been detected at any site, including the SAGs (L. Duportets, unpublished observation).

To clarify the role of JHAs in males of the moth A. ipsilon, we performed electrophoretic studies of SAG contents as a function of age and JHA biosynthesis. In this paper, we studied SAG maturation, the role of JHAs in this maturation and the possible links between SAG maturation and male sexual responsiveness.

The colony of Agrotis ipsilon (Hufnagel) originated from moths caught in southern France. Adults from field catches are introduced into the colony each spring. Larvae were reared on an artificial diet and maintained in individual plastic cups until pupation under a 16 h:8 h light:dark cycle at 21±1 °C (modified from Poitout and Buès, 1974). Pupae were observed each day to identify day 0 (newly emerged) adults. Adults were held in plastic boxes and had access to 20 % sucrose solution. Males and females were kept in different chambers.

The following symbols were used to identify males according to their age following emergence and their mating status: A+x denoted x-day-old adults and M+y denoted the number of days after mating. For example, A+6, M+2 would mean 6-day-old adults, 2 days after mating. M₁ represents the first mating and M₂ the second mating. Pairs were formed between day 4 males (A+4 males) and females before the onset of scotophase. Mating always occurs at mid-scotophase (Gadenne, 1993). Males were removed from the boxes just after the termination of mating (A+4, M₁+0) and 24 h (A+5, M₁+1) or 48 h (A+6, M₁+2) later for extirpation of the sex accessory glands (SAGs) and corpora allata (CA). At the end of scotophase, successful mating was always checked by tactile location of the spermatophore by gently pressing the abdomen of the live female between two fingers. At the end of the experiments, the females were dissected and checked again for the presence of the spermatophore.

To analyse the effect of a second mating (M₂) of males on both JH biosynthetic activity and protein synthesis by the sex glands, mated females were removed and replaced by new virgin day 4 females that were allowed to mate with males that had mated the preceding night. Mating was observed, and the SAGs and CA were removed either immediately after the second mating of the males (A+5, M₂+0) or the next day (A+6, M₂+1).

Removal of the CA was performed according to Gadenne (1993). Newly emerged day 0 males (A+0 males) were allowed free access to drink and were allatectomized less than 3 h after emergence. Care was taken to ensure that only the CA and not the corpora cardiaca were removed. Sham operations were performed by opening the head and touching, but not removing, the CA.

[¹⁴C]Sodium acetate was purchased from Amersham International (Buckinghamshire, UK). JH-III and Ficoll were purchased from Sigma (Taufkirschen, Germany). JH-II was purchased from SciTech (Prague, Czech Republic). Fluvastatin (Sandoz compound XU 62-320) was a gift from Dr F. Kathawala (Sandoz Research Institute, Hanover, NJ, USA). JHA was obtained by an enzymatic procedure, using pig liver carboxyesterase (Sigma) (Duportets et al. 1996), in which three enzyme units (1 unit will hydrolyse 1.0 μmol of ethylbutyrate to butyric acid and ethanol per minute at pH 8.0 at 25 °C, Sigma) converted 1 μmol of JH into JHA in 1 h at 40 °C in a phosphate buffer (50 mmol l⁻¹, pH 7.4). The JHA formed was subsequently purified by reverse-phase liquid chromatography (RPLC).

Injections were performed using a 10 μl Hamilton syringe. Allatectomized 0-day-old (A+0) males of A. ipsilon were injected on day 1 with either 2 μl of olive oil or 2 μl of olive oil containing 5 μg of synthetic JH-III (Sigma) or JH-III acid.

In our rearing conditions, scotophase starts at 09:00 h and ends at 17:00 h, and mating always occurs between 12:00 h and 15:00 h. Fluvastatin, a hydroxymethylglutaryl-CoA reductase inhibitor, inhibits juvenile hormone biosynthesis in insects (Debernard et al. 1994) including A. ipsilon (Duportets et al. 1996). Fluvastatin (30 μg in 2 μl of water) was injected into the mated males just after lights on (i.e. 17:00 h). Twenty-four hours later, both SAGs and the CA from mated/injected males were removed, and the SAGs were analysed for protein content and the CA for JH biosynthesis. Water injections were used as control experiments.

A second experiment was conducted with fluvastatin using the same injection procedures as described above. The following day, however, mated/injected males were placed with virgin females to monitor their ability to re-mate. After observing re-mating of these males, the mated females were checked for the presence of a spermatophore. Fertility was measured by counting the number of females laying viable eggs.

The SAGs of A. ipsilon males are long paired glands of a pink colour originating from the ductus ejaculatorius duplex (Gemeno et al. 1998). The glands were dissected under methanol (100 %) and stored in Eppendorf tubes at −20 °C under methanol. Proteins were precipitated, sonicated and centrifuged at 24 000 g in pure methanol at 4 °C for 15 min. Pellets were homogenized in ammonium formate buffer, pH 4.8. After protein extraction and centrifugation, supernatants were collected and stored at −80 °C. Total protein contents were measured following the method of Bradford (1976).

Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS–PAGE) of freeze-dried sex accessory gland extracts was performed using the method of Laemmli (1970) with a Bio-Rad Mini Protean II apparatus (Richmond, CA, USA) and 7.5 % polyacrylamide slab gels. Each well was loaded with 10–15 μg of total proteins. Electrophoresis was run at 120 V for concentration and at constant voltage (150 V) for separation. Pre-stained molecular markers were obtained from Bio-Rad (high range, molecular mass 52–197 kDa). Gels were stained with Coomassie Blue for 10–15 min. Protein levels in the gels were estimated by densitometry using a BioProfil Gel analysis system (Vilber Lourmat, France). At least three replicates of each electrophoresis experiment were performed.

Rates of JH biosynthesis were determined in vitro using a radiochemical assay (Pratt and Tobe, 1974). Corpora allata were dissected out and incubated for 4 h at 28 °C in medium TC 199 (75 ml, Flow Laboratories, Irvine, UK) supplemented with Ficoll (20 mg ml⁻¹) and [2-¹⁴C]sodium acetate (final concentration, 943 mmol l⁻¹; specific activity, 1.08 Bq pmol⁻¹). Each incubation was performed with three pairs of CA (pCA). Rates of JHA biosynthesis were calculated according to a dual-labelling procedure. Female CA incubated in medium TC 199 containing [2-¹⁴C]sodium acetate and L-[methyl-³H]methionine (final concentration, 101 mmol l⁻¹; specific activity, 17.27 Bq pmol⁻¹) produce dually labelled JH-III and JH-II. One methyl group from methionine was incorporated into each JH molecule and, on the basis of this, acetate incorporation was found to be 0.9 per JH-III molecule and 0.8 per JH-II molecule. We assumed an identical level of acetate incorporation into JHA in males.

Analytical conditions were essentially those described by Halarnkar and Schooley (1990). The HPLC system (Beckman System Gold HPLC, Palo Alto, CA, USA) with ultraviolet detector (Beckmann 166) was connected to a radioactivity monitor (Berthold LB 506, Postfach, Germany). The column was a 50 mm×4.6 mm polymer column (PLRP-5, 5 mm, 100 A) protected with a 5 mm×3 mm PLRP-5 guard cartridge (Polymer Laboratories, Amherst, MA, USA). Buffer A was 5 mmol l⁻¹ Hepes adjusted to pH 6.2. Buffer B was 5 mmol l⁻¹ Hepes buffer (final concentration) in 80 % acetonitrile. Juvenile hormone and its catabolites were separated at a flow rate of 1 ml min⁻¹ using a linear gradient from 5 % to 100 % buffer B over 20 min. Ultraviolet emission was monitored at 245 nm.

Statistical analyses of the effects of fluvastatin injection into males were performed by the comparison of proportions using the χ²-test, P⩽ 0.05.

To examine the statistical significance of differences between mean rates of JHA biosynthesis and of protein content, two-sample t-tests were used, P⩽ 0.05. Values are presented as means ± S.E.M.

In male A. ipsilon, a post-emergence enlargement of the SAGs occurs, and the change from thin and rather white structures to thick and reddish structures is easily detectable under the dissecting microscope. Sex gland maturation was investigated by measuring SAG protein content in males between the ages of 0 and 6 days. As shown in Fig. 1, the protein content of the SAGs increased with age to a maximum level in 4-day-old males. These showed a twenty-fold increase in protein content over 0-day-old males, and higher levels than those found in 5- or 6-day-old males.

Fig. 1 also shows that the rate of JHA biosynthesis (JHA-II and JHA-III + unknown compounds) increased gradually with age, starting at a low level (15.44±2.43 pmol pCA⁻¹ h⁻¹, N=5) and reaching a level of 63.53±15.45 pmol pCA⁻¹ h⁻¹ (N=8) at day 4. No difference was found in the JHA-II/JHA-III ratio, and the rate of the biosynthesis of the unknown compounds increased proportionally to that of JHA-II and JHA-III.

The protein content of the SAGs in allatectomized 4-day-old males was significantly different from that of untreated 4-day-old males (P<0.001), although it was not as low as that of 0-day-old males (P<0.001) (Fig. 2). The protein level of sham-operated males did not differ significantly from that of 4-day-old males.

Injection of JH-III or JH-III acid (5 μg on day 1) increased the level of protein in allatectomized males, but it remained significantly lower (P=0.0035 for JH-III and P=0.0036 for JH-III acid) than those of untreated or sham-operated 4-day-old males (Fig. 2). Injection of olive oil did not induce a significant increase in protein levels in the glands of allatectomized males. Electrophoretic studies revealed a similar number of proteins present in both control and operated males (Fig. 3). Allatectomy did not alter the pattern of proteins extracted from the SAGs, although the protein content was very much reduced. The major protein at 90 kDa, which appeared soon after emergence (see Fig. 5), continued to be present after allatectomy.

The protein content of SAGs from newly mated males was very low compared with that of control unmated males (P<0.001) (Fig. 4A). Sex accessory glands dissected out 24 h after mating showed an increase in the level of protein present (P<0.001), which was maintained for a further 24 h and reached levels comparable with those found in mature non-mated 6-day-old males (A+6 males; see Fig. 1). The synthesis of protein 24 h after mating was significantly inhibited (P<0.001) when fluvastatin was injected after males had mated (Fig. 4B). When males were allowed to mate for a second time during the following night, the protein content of the SAGs was very low just after the second mating, not significantly different from that of A+4, M₁+0 males, and the recovery of protein synthesis was lower 24 h after this second mating than after the first mating (Fig. 4C).

Electrophoresis studies of the protein content of SAGs revealed that the same proteins were present in mated and non-mated males (Fig. 5), even though the protein content was lower in mated males. In particular, the most abundant protein band of 90 kDa, which was hardly visible from SAGs of 0-day-old males, was present, although at low level, in SAG extracts from once-mated (A+4, M₁+0) and twice-mated (A+5, M₂+0) males.

Biosynthesis of JHAs showed a sharp increase 24 h after mating (P<0.001) and levels remained high 48 h post-mating (Fig. 4A). When males were allowed to mate a second time, the level of JHA synthesis remained high after the second mating, although no further significant increase was induced by this second mating (Fig. 4C). Injection of fluvastatin into males just after mating prevented an increase in JHA production by the CA on the following day (P=0.0011) (Fig. 4B).

The ability of mated males injected with fluvastatin to re-mate on the following night with virgin females was assayed. Table 1 shows that fluvastatin-injected mated males were able to perform a second mating (53 % re-matings) as well as control males (62 % re-matings), but these matings were significantly less successful (56 % mated females with a spermatophore) than those of control males (100 % mated females with a spermatophore). While most females (82 %) that mated with control males laid fertile eggs, only 24 % of females that mated with fluvastatin-injected males laid fertile eggs (Table 1).

In this study, we show that the CA regulate the development of the SAGs in male A. ipsilon. The pattern of development of SAGs in Lepidoptera is the same as that identified in several other orders of insects including Diptera, Dictyoptera and Orthoptera (for a review, see Gillott and Gaines, 1992). Just after emergence, the glands are undifferentiated and have a low protein content which subsequently increases until day 4. In A. ipsilon, the main protein band detected in the SAGs had a molecular mass of approximately 90 kDa. In Blattella germanica, four main bands have been detected ranging in mass from 15 to 65 kDa (Vilaplana et al. 1996). When determined in vitro, the rate of JHA biosynthesis showed a gradual increase with age, which is similar to observations made in another moth, Pseudaletia unipuncta (Cusson et al. 1993). As already demonstrated in male moths, the CA of A. ipsilon do not produce JH but release JHA (Duportets et al. 1996) and other unknown compounds that may correspond to acid-conjugated forms of JH that are still unidentified, as reported in M. sexta (Granger et al. 1995). The rate of total isoprenoid biosynthesis (JHA-II and its conjugated forms) in A. ipsilon showed a gradual age-related increase. A similar increase with age in the rates of JH and total protein synthesis by the glands has been demonstrated in Blattella germanica (Belles and Piulachs, 1992; Piulachs et al. 1992).

Allatectomy performed on day 0 prevented the normal development of the sex glands. The total protein content remained at a low level, but no single proteins seemed to be specifically affected. Both JH and JHA partially restored the protein content of the SAGs of allatectomized males, as has been shown in monarch butterflies using JH alone (Herman, 1975). This report is the first to show that JHA has a biological effect on the glands of male moths. In B. germanica, allatectomy inhibited the development of the sex glands, and no particular protein could be detected whose concentration was affected by the treatment, although a qualitative difference in the appearence of protein bands was seen between experiments performed at day 5 and day 10 (Vilaplana et al. 1996). Our experiments show that allatectomy in A. ipsilon did not inhibit the synthesis of the main protein band. It therefore appears that JHA acts on the synthesis of the total protein content of the glands. It is still not clear whether it is JHA and/or JH that controls the development of the SAGs. Juvenile hormone injected in vivo can readily be converted into JHA by JHA methyltransferase, but we have been able to show no such activity in the SAGs or in any other tissue (L. Duportets, unpublished observations).

The mating of males induced changes in both the protein content of the SAGs and the rate of JHA biosynthesis by the CA. Just after mating, the protein content of the SAGs dropped to a low level but was restored on the following day concomitant with a sharp increase in the level of JHA biosynthesis, which remained at this high level on the following day. Since the production of JHA and the replenishment of the protein content of the SAGs are synchronous, we postulate that these events are linked. Further evidence to support this view was provided by injections of fluvastatin, which inhibited JHA production in vivo (Duportets et al. 1996). Complete inhibition lasted for 3 h, and JHA production was then slowly restored (Duportets et al. 1996). In the present study, we show that an injection of fluvastatin prevented the normal increase in the protein content of the SAGs in mated males, indicating that a reduction in JHA production could reduce the replenishment of protein in the SAGs.

Although it is possible to find mated females of A. ipsilon with more than one spermatophore in field-trap experiments, this happens very rarely under our rearing conditions. However, if males were allowed to mate with a different female each night, multiple mating of the males readily occurred, leading to many inseminated females. Once the males had mated, they were unable to mate a second time in the same scotophase (M. C. Dufour, unpublished observation). The protein content of the SAGs of mated males was very low, although their rate of JHA biosynthesis did not show any significant decrease. It appears that the male of A. ipsilon is not equipped for two matings in a single night.

In the cockroach B. germanica, JH biosynthesis showed a dissimilar pattern of activity so that, although it decreased just after mating, it did not increase 24 h post-mating (Vilaplana et al. 1996). In A. ipsilon, a signal must be produced by the male during mating inducing the CA to produce more JHA. It is probable that this mating stimulus may act on the brain to produce and release allatotropin factors and/or to remove allatostatic factors. Work in progress in our laboratory shows that M. sexta allatotropin factors can indeed induce a marked increase in the rate of JHA production by males to levels identical with those found in mated males (L. Duportets and F. Couillaud, unpublished observation). When mated males were injected with fluvastatin, there was no increase in the rate of JHA biosynthesis and the recovery of protein synthesis was lower than that of controls (Fig. 4B). Moreover, although the fluvastatin-injected males were able to re-mate, their ability to transfer a spermatophore was greatly decreased (Table 1). JH and/or JHA have previously been shown to control the processing of the olfactory stimulus leading to pheromone responsiveness (Gadenne et al. 1993; Duportets et al. 1996). Through the pheromone response, JHA is able separately to control both the production of sex accessory gland proteins and the mating process. Fluvastatin-injected males were able to mate (to respond to the pheromone), although their SAGs were not ready for a successful mating. In a previous paper, we showed that fluvastatin temporarily inhibited the response of the males to the sex pheromone (Duportets et al. 1996). The strongest effect was obtained by injecting the JH biosynthesis inhibitor 3 h before the behavioural test. Here, we show that fluvastatin disturbed the re-mating behaviour and induced numerous unsuccessful spermatophore transfers (Table 1). The number of re-matings could be explained by the fact that fluvastatin was injected just after the first mating, much more than 3 h before the CA were removed for the analysis of JHA biosynthesis. The recovery of JHA production may have been sufficient to allow a pheromone response, but the restoration of the protein content of the SAGs was delayed so that protein levels were too low at the time of mating. This is confirmed by re-mating experiments showing that, when CA activity was not inhibited, males were able to re-mate successfully, even though the protein content of the SAGs had been lowered by the first mating (Fig. 4C).

To our knowledge, we have shown for the first time that JHA in a male moth acts both on the level of protein storage in the SAGs and on the responsiveness to a female-produced sex pheromone. In addition, JHA seems to act separately on these systems so that the hormone is necessary (i) for the olfactory signal to be processed by the brain, thus allowing mating behaviour (i.e. copulation), and (ii) for the sex accessory glands to develop and store proteins, thus allowing successful mating (i.e. transfer of the spermatophore).

We thank Dr M. D. Piulachs and two anonymous reviewers for their helpful comments on an earlier version of the manuscript. We also thank S. Madi for her help in the RPLC analysis. This work was supported in part by Conseil Régional Aquitaine (F.C. and C.G.).