Involvement of cAMP and cGMP in the mode of action of molt ...

ELSEVIER

Molecular and Cehlar

Endocrinology 102 (1994) 53-61

Involvement of CAMP and cGMP in the mode of action of molt-inhibiting hormone (MIH) a neuropeptide which inhibits steroidogenesis in a crab B. Sai’di *la, N. de Be&

a, S.G. Webster b, D. Sedlmeier ‘, F. Lachaise a

’ Laboratoire de Bbchimie et Physiologic du D&eloppment,

ENS, URA 686, 46 rue d’Ulm, 75230 Paris Ceder 05, France b School of Biological Sciences, University of Wales, Bangor, Gwynea’d LL57 2UW, UK ’ Universitiit Bonn, Institut fir Zoophysiologie, 5300 Bonn 1, Germany

(Received 13 September 1993; accepted 1 February 1994)

Abstract In crustaceans, production of molting hormones (or ecdysteroids) by the molting glands (Y-organs; YO), is under negative control exerted by a neuropeptide, the molt-inhibiting hormone (MIH). MIH of the crab Carcinus maenas inhibits in vitro steroidogenesis of basal (intermolt crab) or activated (premolt crab) YO. MIH inhibits secretion of the two ecdysteroids synthesized by crab YO, ecdysone (E) secreted throughout the molting cycle, and 25-deoxyecdysone (25dE), secreted during the premolt period. At a MIH concentration of 10m8 M, E is reduced about 50% and 25dE 94%. Regardless of the molting stage, this inhibition of steroidogenesis is reversible, dose dependent and measurable after 5 min. On intermolt YO, MIH induced cGMP increase and 8BrcGMP mimics the effect of MIH: at this stage cGMP seems to be involved with MIH inhibition of steroidogenesis. On premolt YO MIH induced a transient increase of CAMP (Zfold) and a long-lasting enhancement of cGMP @O-fold). On active YO, we demonstrated that a low concentration (10m5 M) of dbcAMP, 8BrcAMP, 8BrcGMP, or agents increasing intracellular CAMP, mimic MIH effects and inhibit steroidogenesis. From these observations it is concluded that both cyclic nucleotides are involved in the mode of action of MIH on activated YO. At this premolt period, MIH/cAMP may act cooperatively with MIH/cGMP in the inhibitory control of steroidogenesis by crab YO. Key words: Cyclic AMP; Cyclic GMP; Neuropeptide;

Steroidogenesis;

1. Introduction

Classically, it is well known in vertebrates (Baulieu and Kelly, 1990) and insects (Smith and Sedlmeier, 1990) that steroidogenesis is globally under the positive control of peptidic factors. In contrast, in crustaceans, synthesis of ecdysteroids or molting hormones by Yorgans (YO) is negatively regulated by a factor secreted from the sinus glands, a neurohemal structure located in the eyestalks (Lachaise et al., 1993). In arthropods, molting is a cyclical process, expressed externally by periodic shedding of the exoskeleton and physiologically triggered by progressive increases in titres of circulating ecdysteroids or molting hormones. The crustacean molting gland (YO), a

* Corresponding author. Elsevier Science lreland Ltd. SSDI 0303-7207(94)00038-B

Crab

steroidogenic organ of one cellular type (Spaziani, 1990), secretes ecdysteroids (invertebrate steroids) which are converted by peripheral tissues into molting hormones (Lachaise, 1990). Therefore, in arthropods, regulation of steroidogenesis is of major importance in controlling molting events in these animals. Unlike most insects many crustaceans continue to molt as adults. Blockage of molting events may occur via various processes including vitellogenesis, temperature decrease, parasitism or the action of a molt-inhibiting hormone (MIH), a neuropeptide of a new family of peptidic hormones isolated from the sinus glands of the eyestalks (Smith and Sedlmeier, 1990). This peptide has been purified from several decapod crustaceans (review in Webster and Keller, 1989) and in particular from the crab Curcinus muenas by Webster and Keller (1986). Recently, Webster (1991) detailed the complete sequence of Curcinus maenus MIH which consists of exactly 78 amino acids including six cysteine

54

B. Saiifi ef al. /Molecular

and Cellular ~n~~~~~io~

residues. Soumoff and O’Connor (1982) were the first to demonstrate that ecdysteroid secretion by YO is inhibited by sinus gland extracts, a result that has been repeatedly confirmed (review Lachaise et al., 1993). Our understanding of neuroendocrine regulation of the molt-controlling eyestalk-Y-organ system in decapod crustaceans needs a detailed knowledge of the molecular mechanisms underlying neuropeptide actions upon YO. Interestingly, previous studies performed with sinus gland extracts in the crab Cancer antennarius led to the conclusion that CAMP is involved as second messenger in the inhibition of crab steroidogenesis (review Mattson, 1986). In crayfishes, cGMP (instead of CAMP) is involved in inhibition of steroidogenesis (Sedlmeier and Fenrich, 1993). The present study on the role of cyclic nucleotides in the mode of action of MIH in the shore crab Carcinus maena~ was designed for two reasons. Firstly, in the crab Carcinus maenas steroidogenesis has been well explored: YO of the crab synthesize two ecdysteroids, ecdysone (E) and 25 deoxyecdysone (25dE), which are secreted into the medium and converted by peripheral tissues into 20-hydroxyecdysone and ponasterone A respectively @chaise et al., 1986, 1989). Secondly, since pure Curcinus MIH was available, it was possible to investigate direct responses to MIH rather than to crude sinus gland extracts with respect to the MIH signal transduction pathway.

2. Materials and methods 2.1. Animals

The crabs Carcinur maenas were provided by the Biological Station of Roscoff (Brittany, France) and maintained in a compartmentalized aquarium after determination of the molting stages according to Drach and Tchernigovtzeff (1967); reconstituted seawater circulated continuously through charcoal and crushed oystershell filters. Seawater was kept at 17°C and crabs were maintained under natural daylight. The crabs were fed with mussels every 3 days. For C~rcin~, eyestalk ablation (which suppresses MIH source> has no consequence on molting events in wintering adult crabs and old crabs of both sexes as well as in female crabs during vitellogenesis. Wintering crabs, old crabs and female crabs undergoing vitellogenesis which are blocked in the intermolt state have a basal steroidogenic activity of YO and do not synthesize 25dE. In ‘premolt’ crabs which are at D, to D, stages with intense steroidogenic activity of YO, MIH is able to inhibit both molt and steroidogenesis of YO (Mattson and Spaziani, 1985a). After stage D,, MIH has no effect on moit (Lachaise et al., 1993). Consequently, we performed our experiments with crabs of

102 (19941 53-61

two distinct sizes: crabs of large size which molt in spring and summer, and small crabs which molt during all seasons. Intermolt crabs (C,) may be blocked in several ways as indicated above, we therefore made our experiments essentially on premolt crabs CD,-D,), using for comparison, intermolt animals. Large crabs with a cephalothoracic width from 4 to 5 cm and YO with a protein content of 49.6 f 3 pg (n = 60) and small ones with a cephalothoracic width from 1 to 2 cm and protein content of which is 13.68 + 0.97 fig (n = 60) were used. Our protein content assays revealed that among a population of crabs of the same size there is no significant difference between intermolt and premolt animals. 2.2. Culture med~m Medium 199 devoid of phenol red and with Hank’s salts (Sigma) was adjusted to Curcinus maenas hemolymph osmotic pressure (1035 mOsm) to give the following final concentrations: 477 mM NaCl, 12 mM KCl, 64 mM MgCl,, 15 mM Na,SO,, were added NaHCO, (4.2 mM), Hepes (10 mM), 0.16% penicillin, 0.1% streptomycin and proline (0.006%), pH 7.4. Culture medium was filtered through a 0.22 pm polyacetate membrane before use. 2.3. Preparation of MIH MIH extraction and purification from sinus gland of crabs was performed as described in Webster (1986). Purified MIH was quantified either by amino acid analysis or by RIA or Elisa as detailed in Webster (1991, 1993). 2.4. Preparation of YO Crabs were anesthetized in ice before dissection. The paired YO were excised and placed in cooled wells of 96-well culture plates after removal of adhering fatty tissues. Wells had previously been filled with control or treated medium. YO were incubated at room temperature for time course studies as indicated in the appropriate figure legends. In~bation experiments were performed with either single YO or quarters of YO (Mattson and Spaziani, 1985b) or with a pool of 4-40 whole YO. One organ was used as control and the other as experimental. The culture medium and/or the YO were collected and stored at -20°C until assayed for ecdysteroid, CAMP or cGMP contents. Whilst it is generally accepted that cyclic nucleotides are never secreted in vivo, observations on the accumulation of cyclic nucleotides in the medium exposed to tissues cultured in vitro when stimulated by hormones (accumulation of CAMP: Browne et al., 1990, Kato et al., 1992; accumulation of cGMP: Billiar et al.,

B. Saiiii

et al. /Molecular

and

Cellular

1992) suggest that such measurements may be a useful tool in elucidating signal transduction mechanisms. Every 2 min half the medium was removed and separated into 3 equal volumes, one of which served for CAMP assay, another for cGMP and the third for ecdysteroid measurements. The half removed was replaced by an equivalent volume of new medium and the rate of secretion per 2 min was calculated as the total amount measured minus half amount determined at the previous time. Statistical significance was established by Student’s r-test. Differences are accepted as significant at p < 0.05 or 0.01. 2.5. Chemicals Ecdysone was obtained from Simes (Milan, Italy). 25dE was synthesized in our laboratory (Lachaise et al., 1988). Forskolin, 1,9-dideoxyforskolin, 3-isobutyl1-methylxanthine (or IBMX) and analogs of cyclic nucleotides (&bromo-CAMP: 8BrcAMP, 8-bromo-cGMP: 8BrcGMP, dibutyryl-CAMP: dbcAMP, dibutyryl-cGMP: dbcGMP) were all purchased from Sigma Chemical, Saint Louis, MO, USA. Drug solutions were made by dissolving the agents in either alcohol or DMSO (dimethylsulfoxide). Carriers for stock solution were: 95%

Endoctilo~

102

(1994)

53-61

55

ethyl alcohol. for forskolin and 1,9-dideoxyforskolin, 2% DMSO for cyclic nucleotides analogs and 100% DMSO for IBMX. Control experiments indicated that these carriers did not affect YO activity. Reagents for CAMP measurements were purchased from Cayman Chemical @PI-BIO, Saclay, France). 2.6. Ecdysteroid assays The effects of MIH on ecdysteroid production by YO were first determined by measuring contents of ecdysone and 25dE in the medium by RIA (25dE) or EIA (E) following an HPLC separation of the two steroids. Dry samples of medium were analyzed on a reverse-phase Cl8 Radial Pak liquid chromatography cartridge (Water Associates; USA). The different products were separated by an isocratic solvent 1 (methanol/H,0 1: 1 v/v) during 10 min followed by a linear gradient (20-80% of acetonitrile in solvent 1 for 30 min) (1 ml/min). Retention time for E was 10 min and for 25dE 24 min. The fractions corresponding to E and 25dE were collected and analysed quantitatively by RIA with DUL2 antibody (a gift from Professor J. Koolman (Marburg, Germany) which recognizes E 4fold better than it does 25dE (Lachaise et al., 1992)

summer

Fig. 1. Maximal production (sprina) and minimal production (winter) of ecdysone (E) and 25deoxyecdysone (25dE) by Y-organ (YO) from small crabs at two periods of the molting cycle. Ecdysteroid production was measured within 30 min. The bars represent ecdysone (black columns) and 25dE (hatched columns) levels expressed in pmol of e&steroid per YO. Values are means f SE of four independent measurements. I: intermolt crabs (C, stage), basal secretion of e&steroids, inactive YO. P: premolt crabs (Do-D, stages), high secretion of ecdysteroids, active YO.

and by EIA with 4919 antibody (a gift from Professor P. Porcheron) directed against 20-hydroxyecdysone and that well recognizes ecdysone, but lo-fold less 25dE, using the method described by Porcheron et al. (1989). For subsequent testing of the effects of the various pharmacological agents and MIH on inhibition of steroid production, secreted ecdysteroids were measured only with EIA method. 2.7. EL4 for CAMP and R.lA for cGMP

12 0

10 1

b

8i

*

-P -1

/’

ti

W b E 0. MIN

in vitro Fig. 2. Time-course studies of ecdysone (E) p~u~ion during 6 h by one YO from intermolt (1) and premoit (PI small crabs. At various times, 5% sample of the medium were analyzed to determine the E concentration by EIA, and 5% of fresh medium were added. Active premolt (PI YO secrete approximately twice as much ecdysteroid compared to intermolt YO (I). Values are meansf SE of 10 independent experiments. p < 0.01 (Student’s t-test).

Incubated YO and/or aliquots of the medium were placed in ice-cold ethanol to stop reactions, sonicated (10 min), boiled (6 min), and centrifugated (12,000 X g, 10 min). These operations were repeated three times. The supematants were pooled and evaporated to dryness. The assay used for CAMP determination is a competitive enzyme immunoassay with acetylcholinesterase-linked CAMP as label (CAMP Kit, Spi-Bio, France). Acetylation of standards and samples by acetic anhydride was performed in order to enhance the sensitivity, the minimum detectable concentration of the cyclic nucleotides being close to 0.04 pmol/ml (Pradelles et al., 1989). cGMP was measured by a 3H-cGMP kit (Amersham). Cyclic nucleotide contents are expressed as pmol per mg protein or per YO.

secretion rate of ecdysteroids increases at the beginning of the incubation period (Fig. 2) and decreases after 6 h of incubation regardless the molting stage. ~nsequently, all incubations were for 6 h. As has already been established by Webster (19861, we systematically ascertained that right and left YO of Carcinus secreted similar quantities of ecdysteroids hence permitting the use of pairs of YO for experiments (one experimental, one control).

2.8. Protein content assay

3.2. MIH effects

After extraction of ecdysteroids and cyclic nucleotides, the dry pellets were dissolved in a solution of 0.5 M NaOH, that contained 5% SDS. Proteins were quantified by Lowry’s method as modified by Peterson (1977) using bovine serum albumin as reference.

Preliminary studies were performed to establish the characteristics of MIH effects before considering the effects of the various analogs listed above. Inhibition performed by the inhibiting ~l~eptide occurs at any time of the year and of the molting stage, either in intermolt or in premolt. The % inhibition exerted on 25dE synthesis is much more pronounced than that for E (Fig. 3). After a 6 h incubation with MIH (1O-8 M), followed by extraction of ecdysteroids, HPLC and EIA of the E and RIA for the 25dE zones, we observed that the synthesis of both E and 25dE is consistently reduced in premolt crabs: E is reduced nearly 50% (Fig. 3A) and 25dE 94% (Fig. 3B). A concentration of lOma M MIH reduces steroidogenie activity of YO of premolt crabs to that of the intermolt level. For intermolt crabs, also, a 50% decrease in E secretion has been demonstrated with the same experimental conditions. Considering that E is secreted in a quite similar amount at any molting stage and time of the year, we discuss below the effects of MIH essentially on E synthesis as measured by ELA, however, when the 25dE secretory level was very high it cross react with 4919 antibody (see Materials and methods) and we measured the inhibition of the synthesis of the two ecdysteroids. Fig. 4 shows the dose-dependency effect of MIH on YO incubated for 1 h. The observed inhibition is

3. Results 3.1. In vitro ec~ysiero~ secretion of Y-organ YO of the crab Carcinus maenas secrete two ecdysteroids, ecdysone (El and 25-deoxyecdysone (25dE), the ratio of which varies during the molting cycle as well as during the year. At the intermolt stage, in vitro production of 25dE is low and less than that of E (Lachaise et al., 1986). The synthesis of the former becomes very significant in premolt and can be lo- to 20-fold as much as E produced during spring, and is approximately 2-4 times higher in autumn or in winter for small crabs (Fig. 1). In vitro secretory activity of E measured in the YO incubation medium varies as a function of the molting stage. This secretion is higher during premolt and displays a 100% increase compared to intermolt crabs (Fig. 2). YO secretory activity fluctuates also during timecourse incubation: in our e~erimental conditions, the

0A

n 8

EC EMIH

4

5

20

30

60

120

130 240 300

360

incubation time (min)

Fig. 3. Inhibition of ecdysone (E) and 25-deoxyecdysone (25dE) synthesis exerted by MIH (lo-* M), measured after 6 h of incubation time on premolt YO. E and 25dE were extracted and analyzed by HPLC followed by EL4 for ecdysone and RIA for 25dE (see Materials and methods). A: MIH inhibition of ecdysone synthesis; B: MIH inhibition of 25dE synthesis. E: ecdysone; 25dE: 25deoxyecdysone; EC: E secretion from control YO, E MIH: E secretion from MIH-treated YO; 25dE C: 25dE secretion from control YO; 25dE MIH: 25dE secretion from MIH-treated YO. Values are means rt SE of four independent measurements; p < 0.05 (Student’s f-test).

dose-related up to lo-’ M and statistically significant; beyond this concentration there is no further increase; inhibition of steroidogenesis is never complete. Further experiments were performed at an MIH concentration

801

I

Fig. 5. Inhibition of ecdysone production by premolt crabs: timecourse study of ecdysone synthesis in the absence (C: black columns) and presence of lo-’ M MIH (hatched columns); E secretion was measured as in Fig. 2. n = 10, values are means It SE of ten experiments. p < 0.01 (Student’s r-test). Data are also expressed as a percentage of relevant control: % inhibition are calculated with the means of these ten values for control and experimental (curve). At S-360 min, inhibition was significantly exerted and 50% inhibition was obtained after 20 min until the end of the culture period.

co~es~nding to an EC,, of 10m8 M, bringing the rate of secretion of ecdysone by premolt YO to the level observed in intermolt animals. We showed that significant inhibition of steroidogenesis is detectable within 5 min, when single or quarter YO are incubated (Fig. 5, curve). Moreover, at this concentration (lo-’ M) MIH induces a 50% inhibition of ecdysone synthesis throughout the time-course study (6H, Fig. 5, columns). It was also demonstrated that the reversibility of MIH action at lo-* M, is dependent upon the duration of the incubation. After 2 h, MIH inhibition of ecdysteroid production is still reversible. When YO incubated for less than 2 h in the presence of the neurohormone are transferred to a medium devoid of MIH, the control and experimental organs secrete equivalent amounts of ecdysteroids. However, if the incubation period lasts 3 h or more, the secretion rate of treated YO never reaches that of controls, and there is a continuing and significant inhibition of steroidogenesis (l&40%). 3.3. CAMP-cGMP assays with 10 - ’ M MIH

0,Ol

0,i

IO 1 MIH (nM)

1001000

Fig. 4. Dose-response curve of % inhibition of steroidogenesis by MIH from YO of premolt crabs incubated for 1 h. E secretion was measured as in Fig. 2. Results are expressed as means of percentage of inhibition over control If:SE of fwe experiments. Pairwise comparison indicate that results are signifi~ntly different (p < 0.05; Student’s t-test) except IO8 versus 1000 nM.

cAMP and cGMP basal Ieveis pranced within the YO and their rate of accumulation. We first evaluated CAMP

and cGMP content in right and left YO immediately after their removal, and after various incubation times, both for intermolt and premolt organs. The concentration was similar for right and Ieft YO, between animals, and during time-course studies. CAMP values

B. Sai’di et al. /Molecular

58

n n

Z

and Cellular Endocrinology 102 (1994) 53-61

P-C P-MIH

ra

I-C

q

I-MIH

20

g 0 1

2

3

4

5

min

Fig. 6. YO CAMP content, incubation of YO quarters. Comparison between controls YO (Q and MIH-treated YO (MIH, 10-s M). Solid columns: Y-organs from premolt crabs (PI. Hatched columns: YO from intermolt crabs (I). In premolt crabs, statistically significant difference was measured from l-4 min (n = 10; f SE, p < 0.01; Student’s t-test).

were estimated to be about 34 + 4 pmol/mg protein and 58 f 8 pmol/mg protein, respectively, for premolt and intermolt YO (Fig. 6). cGMP contents for intermolt and premolt YO were 6 &-2 pmol/mg protein (Fig. 7). We investigated the accumulation of cyclic nucleotides in the incubation medium as a mean of monitoring MIH effect. Cyclic nucleotides are accumulated in the incubation medium of crab YO, the accumulation rate for CAMP is 15 * 10 fmol/YO/2 min and 5 f 1 fmol/YO/2 min for cGMP. MIH action on intracellular cyclic nucleotide content within YO. We explored the variations of production of

both cyclic nucleotides within the YO, during shortand long-term studies, in the presence or absence of MIH in the incubation medium. A time-course of cyclic nucleotide accumulation was also established, in the absence or presence of MIH every 2 min, the first 0.5 h of culture. CAMP. To reduce interindividual variability, 4 identical pools of YO were used which consisted of using quarters of YO in the incubations, a pool of 4 x 0.25 right YO on the one side (control), and a pool of 4 X 0.25 left YO on the other (experimental, with MIH). At the appropriate time, the quarters of YO are removed for CAMP (or cGMP) assay, and the medium stored for ecdysteroid measurements. We demonstrated that quarters of YO of premolt crabs exposed to MIH resulted in a 2-fold rise in CAMP, 1 min after its addition. This effect is maintained for 1 to 4 min (Fig. 61, but is no longer detectable after 5 min to 6 h of incubation, a time corresponding to the detection of steroidogenesis inhibition (Fig. 5). For intermolt crabs, CAMP content is approximately 1.5 to 2 times as much compared to premolt YO and this concentration is not increased significantly after MIH exposure (Fig. 61, whatever the incubation time.

cGMP. MIH also elicits a cGMP rise in premolt crabs, but to a greater extent: the quantity of cGMP measured at 3 min within the YO in the presence of MIH is lo-fold as much (compared to non-treated YO). At 30 min, this accumulation is enhanced 60-fold (6 pmol/mg protein for the control and 360 pmol for treated YO). After 40 min no further rise in cGMP was detected (Fig. 7). For intermolt crabs, cGMP amounts of control are identical to those found for premolt animals; this content is enhanced in the presence of the neuropeptide, as it is the case for premolt crabs. However, the enhancement of cGMP by MIH (20-fold) is not pronounced as in premolt YO after MIH addition. MIH effect on cyclic nucleotides released into the incubation medium (premolt stage). A few min before

the establishment of ecdysteroidogenesis alteration, a transient rise in CAMP secreted into the medium is triggered: this rate is enhanced by up 2-3-fold, when compared to untreated YO. cGMP profile analysis shows a progressive accumulation in the experimental pool reaching a 20-fold increase at 2 min and a lOO-fold increase at 20 min. From 20 min, the rise in cGMP accumulated remains stable and the concentration is maintained lOO-fold higher for MIH-treated YO (data not shown). 3.4. Effects of pharmacological agents

We investigated whether or not pharmacological agents known to increase intracellular nucleotide concentrations are able to mimic the MIH effect on E secretion. Do these products induce inhibition of the synthesis at any stage of the molting cycle? Does this effect occur rapidly? Is this action reversible and doserelated?

0

10

20

30

40

50

60

70

min

Fig. 7. YO cyclic cGMP content without IBMX, incubation of YO quarters of premolt crabs: comparison between controls YO (0 and MIH-treated YO (MIH, 1Om8 M). Significant statistical difference was measured from l-60 min (n = 3; *SE, p < 0.01; (Student’s t-test)).

B. Saiiii et al. /Molecular

0-i

0

,

,

20

40

(

60

.

,

.

60

,

100

I

120

and Cellular Endocrinology 102 (1994) 53-61

1 140

min Fig. 8. Time-course of the inhibitory effect on steroidogenesis of pharmacological agents that increase CAMP level or cyclic nucleotide analogs in YO of premolt crabs. Treatments are 10m4 M IBMX, lo-’ M dbcAMP, 10m5 M 8BrcGMP and 1OW.6forskolin. E secretion was measured as in Fig. 2. Significant statistical difference was measured from 5-120 min tn = 3; +SE, p < 0.05; (Student’s t-test)). Data represent the % obtained by the means of the relevant control.

Analogs of CAMP. Fig. 8 shows the time-course of the effect of lo-’ M of dbcAMP on ecdysteroid production by premolt crab YO. The decrease in steroid production was mimicked when dbcAMP was substituted for MIH, this inhibition being detectable within 5 min. The addition of 8BrcAMP, at the same concentration, also markedly decreased ecdysteroid production within as little as 5 min. The observed inhibition is of approximately 30% and is identical to that obtained with 10m9 M MIH, and is reversible (results not shown). Surprisingly, opposing effects are observed, regarding ecdysteroid synthesis depending on whether premolt or intermolt animals are used: with premolt YO, at 10m5 M dbcAMP, there is a significant inhibition of steroidogenesis, whereas at 5 x 1O-s-1O-” M, the analog increases E synthesis in a significant manner. The effect is opposite with intermolt YO, namely, at 5 x 10-5-10-3M, the analog partially suppresses E secretion and at 10e5 M this agent induces stimulation of E production. Analogs of cGMP. On YO from premolt animals, dbcGMP, at low3 M, seems to stimulate E secretion as does dbcAMP at the same concentration; at lower concentrations, 10-4-10-7 M, dbcGMP does not display any significant action. With YO originating from intermolt crabs, dbcGMP does not exhibit any significant effect regardless of the concentration used. However, in premolt crabs, 8BrcGMP strongly inhibits steroidogenesis at lo-’ M (90% after 1 h of incubation, Fig. 8); this inhibition is lower at lo-” M (20% after 1 h of incubation). In intermolt crabs 8BrcGMP inhibits steroidogenesis but the maximum inhibition is 50%. Effects of IBMX. On premolt crabs, IBMX, an inhibitor of cyclic nucleotide-phosphodiesterase, at 10e4 M, exerts an inhibitory effect on ecdysteroid synthesis by YO (60%) (Fig. 81, the maximum observed in the

59

presence of MIH (lo-’ M). However, 10m4 M IBMX also maintains its inhibitory action upon YO of intermolt crabs. The inhibitory effect occurring under the influence of IBMX was observed with concentrations between 10m4 and lo-’ M regardless of the molting stage. Lower concentrations (1O-5-1O-7 Ml, were uneffective. Effects of forskolin and I, 9-dideoxyforskolin . Forskolin, a diterpene activator of adenylate cyclase, at lo-’ M exerts an inhibitory effect (20%) on ecdysteroid synthesis by YO of premolt crabs (Fig. 8). YO from intermolt animals subjected to lo-” M forskolin promote steroidogenesis. At higher concentrations, forskolin has no significant effect regardless the molting stage of animals. In premolt crabs, it was of interest to ascertain that the forskolin effect was only a result of adenylate cyclase activation, and not due to alteration of the activity of any enzyme of the steroidogenesis pathway. This was tested with an inactive derivative of the diterpene, 1,9-didcoxyforskolin, that does not stimulate adenylate cyclase activity (Keogh et al., 1992). This analog failed to inhibit steroidogenesis. These findings clearly indicate that forskolin acted specifically on CAMP generation.

4. Discussion 4.1. MIH action on ecdysteroidogenesis premolt and intermolt crabs

by YO from

Our studies have shown that MIH inhibits steroidogenesis of YO both from premolt and intermolt crabs. At intermolt, a period during which the animals do not exhibit any phenomenon related to molting and YO are slightly active secreting a minimal rate of ecdysteroids where E is the major steroid, MIH at a concentration of 10-s M induces a 50% decrease in E production. This basal rate of secretion in intermolt is not sufficient to trigger molting; this rate is maintained at a very low level either through the continuing presence of MIH (in the case of crabs that undergo molting processes after eyestalk removal), or through a longlasting and irreversible effect of MIH and/or via other processes such as vitellogenesis, parasitism or decline of temperature and photoperiod (in the case of crabs that do not trigger molting after eyestalk removal) (Lachaise et al., 1993). At premolt, the period during which the animal has triggered processes of molting and during which YO are very active and secrete a high level of 25dE, MIH reduces as E secretion rate by half, but inhibits preferentially and almost completely 25dE production (as radiolabelled ecdysteroid precursor conversion; Lachaise et al., 19881, thereby bringing the activity of active

YO close to that of inactive ones. Our results suggest that MIH acts during premolt turning the active YO into the inactive state characteristic of intermolt, this reduction of activity being sufficient to inhibit the molting events. With the crab Cancer antennatius, Watson and Spaziani (1985) have shown that in destalked crabs (artificial premolt) YO are more active, secrete an unknown product later identified as 3-dehydroecdysone (Spaziani et al., 19891, in a quantity five-fold greater than ecdysone. Synthesis of 3-dehydroecdysone is preferentially inhibited by eyestalk extracts (Watson and Spaziani, 1985). We also demonstrated that inhibition of steroid synthesis exerted by MIH is- detectable within 5 min, which suggests a direct short-term effect upon enzymes of steroidogenesis or on steroid precursor uptake, and that MIH action is irreversible after 3 h suggesting a long-term effect of MIH, perhaps by inhibiting protein synthesis. Mattson and Spaziani (1985b) observed an inhibition of steroidogenesis from 2 h onwards and demonstrated that this action proceeds via the inhibition of protein synthesis (Mattson and Spaziani, 1986). 4.2. Are CAMP and /or action of MZH?

cGMP involved in the mode of

With the crab Curcinus begs, during premolt, it was shown that pharmacological products (analogs: dbCAMP, 8BrcAMP, 8BrcGMP lo-’ M) or agents that induce the enhancement of intracellular CAMP, either via the activation of adenylyl cyclase (forskolin 10s6 M) or the inhibition of CAMP hydrolysis by phosphodiesterase (IBMX, 10e3 and 10e4 MI, seem to mimic the inhibitory effect of MIH on ecdysteroidogenesis. However, their effects are dose-dependent: in premolt YO high concentrations (10e3 M) of CAMP analogs induce a stimulatory instead of an inhibitory effect on ecdysteroidogenesis. In the cockroach Corpora allata low con~ntrations of 8BrcAMP stimulate juvenile hormone biosynthesis but high concentrations have the inverse effect (Meller et al., 1985). Moreover, analog effects depend on the molting stage: in intermolt crabs, a high concentration of CAMP analog (10m3 M) leads to inhibition of steroidogenesis, whereas a 1O-5 M concentration stimulates E synthesis, Similar opposite results of dbcAMP on steroidogenic tissues (depending on the developmental stage or on the steroidogenic activity of the tissue) have been observed in the gypsy moth, Lymantria dyspar (Loeb et al., 1993). In premolt crabs, MIH elicits a 2-fold rise in intracellular and secreted CAMP, but this rise is transient and precedes the observed inhibition of steroidogenesis. At intermolt, MIH does not seem to alter CAMP content. In marine arthropods, a 2-fold rise in CAMP has been reported, either under the effects of neuropeptides originating from pericardial organs or tho-

racic ganglion extracts on isolated perfused gills (Kamemoto, 19911, or under the action of forskolin (4 x lop6 M) or IBMX (10B3 M) on cardiac muscle (Groome and Watson, 1989). In vertebrates, CAMP concentration is raised Zfold in porcine luteal cells in the presence of lo-’ M of LH (Rajkumar et al., 1991) or rat Leydig cells with 2.4 X 10-r’ M of hCG (Browne et al., 1990). In the crab Cancer antenna~~ the inhibitory effect of eyestalk extracts on steroidogenesis in active YO acts via CAMP as second messenger (Mattson and Spaziani, 1985c). These authors also noticed that CAMP rise (~4) induced by eyestalk extracts decreases abruptly at the moment when inhibition of steroid synthesis is detected, namely, after 2 h of incubation time. However, compared to the results of Mattson and Spaziani (1985c), the CAMP peak found in our studies is smaller (2-fold compared to control) and is ephemeral (l-3 min). Sinus glands extracts from Orconectes limosus induced a rise in cGMP (X 6 from 7-43 pmol within 2 h) within YO from an a~i~cially promoted premolt stage (Sedlmeier and Fenrich, 1993). A major difference between Cancer or Orconectes and Carcinus lies in the fact that in our model, besides this transient peak of CAMP, MIH is capable of eliciting a significant and long-lasting rise in cGMP, measurable within the YO as well as in the medium. Thus in our system MIH exhibits an unusual phenomenon since this hormone elicits the enhancement of both cyclic nucleotides. In crustaceans, other reports have displayed the same intriguing results, namely a simultaneous increase of both cyclic nucleotide levels. Injection of eyestalk extracts into the crayfish ~ro~arn~a~ clarkii promotes a 2-fold increase in CAMP and a 50-fold increase in cGMP in the antenna1 glands, within 5 min of injection (Kamemoto and Gyama, 1985). The crustacean hyperglycemic hormone, (CHH) from the crayfish Orconectes limosw, elicits a rise in CAMP (lo-fold) and cGMP (20-fold) (Sedlmeier and Keller, 1981; Sedlmeier, 1985). Increases in cyclic nucleotide concentration in MIH treated premolt YO could act via two or only one type of cyclic nucleotide dependent protein kinases. Indeed, in vertebrates, two distinct signal transduction pathways (via CAMP and cGMP1 can increase testosterone production. In mouse Leydig cells classically CAMP increase is concomitant with steroidogenesis increase. However, ANP is able to stimulate steroidogenesis via cGMP and the authors hypothesize that this cyclic nucleotide can bind to and activate CAMP-dependent protein kinase (Hipkin and Moger, 1991; Schumacher et al., 1992). In conclusion, in YO of intermolt crabs (basal steroidogenesis activity), MIH inhibits (50%) ecdysone synthesis and induced cGMP increase: at this stage cGMP seems to be involved with MIH inhibition of stero~ogenesis. In YO of premolt crabs (intense

B. Saidi et al./Molecular and CellularEndocrinology102 (1994) 53-61

steroidogenesis activity) MIH inhibits ecdysone (SO%1 and 25dE (94%) synthesis and induced cGMP increase and a transient peak of CAMP. In premolt crabs we found evidence which suggests that MIH/cAMP may act cooperatively with MIH/cGMP in the inhibitory control of steroid production by crab YO. Which kinase(s) is (are) activated remains under investigation.

Acknowledgements We are grateful to Prof. R. Lafont for valuable discussion and suggestions and the Biological Station of Roscoff for providing the crab Carcinus maenas. We thank G. Carpentier for Student’s f-test analysis, Z. Melion for expert technical assistance.

References Baulieu, E.E. and Kelly, P.A. (1990) in Hormones, from Molecules to Disease (Baulieu, E.E. and Kelly, P.A., eds.), pp. 445-456, Hermann, France. Billiar, T.R., Curran, R.D., Harbrecht, B.G., Stadler, J., Williams, D.L., Ochoa, J.B., Di Silvio, M., Simmons, R.L. and Murray, S.A. (1992) Am. Physiol. Sot. C1077-C1082. Browne, ES., Flasch, M.V., Sohal, G.S. and Bhalla, V.K. (1990) Mol. Cell. Endocrinol. 70, 49-63. Drach, P. and Tchernigovtxeff, C. (1967) Vie Milieu 18, 595-610. Groome, J.R. and Watson, W.H. (1989) J. Exp. Biol. 145, 419-437. Hipkin, R.W. and Moger, W.H. (1991) Mol. Cell. Endocrinol. 82, 251-257. Kamemoto, F.I. (1991) Zool. Sci. 8, 827-833. Kamemoto, F.I. and Gyama, S.N. (1985) in Current Trends in Comparative Endocrinology (Lofts, B. and Holmes, W.N., eds.), pp. 883-886, Hong Kong University Press, Hong Kong. Kato, T., Kumai, A. and Okamoto, R. (1992) Toxicol. Lett. 62, 19-24. Keogh, D.P., Mitchell, M.J., Crooks, J.R. and Smith, S.L. (1992) Experientia 48, 39-41. Lachaise, F. (1990) in Morphogenetic Hormones of Arthropods I (Gupta, A.P. ed.), part 1, pp. 275-323, Rutgers University Press, New Brunswick, New Jersey. Lachaise, F., Carpentier, G., Sommi, G., Colardeau, J. and Beydon, P. (1989) J. Exp. Zool. 252, 283-292.

61

Lachaise, F., Goudeau, M., Carpentier, G., Sa’idi, B. and Goudeau, H. (19921 J. Eq. Zool. 261, 886-96. Lachaise, F., Hubert, M., Webster, S. and Lafont, R. (1988) J. Insect Physiol. 34, 557-562. Lachaise, F., Leroux, A., Hubert, M. and Lafont, R. (1993) J. Crust. Biol. 13, 198-234. Lachaise, F., Meister, M.F., HCtru, C. and Iafont, R. (1986) Mol. Cell. Endocrinol. 45, 253-262. Loeb, M.J., Gelman, D.B. and Bell, R.A. (19931 Arch. Insect Biochem. Physiol. 23, 13-28. Mattson, M.P. (1986) Zool. Sci. 3, 733-744. Mattson, M.P. and Spaziani, E. (1985al J. Exp. Zool. 234, 319-323. Mattson, M.P. and Spaxiani, E. (1985b) J. Exp. Zool. 236, 93-101. Mattson, M.P. and Spaziani, E. (198%) Mol. Cell. Endocrinol. 42, 185-189. Mattson, M.P. and Spaziani, E. (1986) Gen. Camp. Endocrinol. 63, 414-423. Meller, V.H., Aucoin, R.R., Tobe, S.S. and Feyereisen R. (1985) Mol. Cell. Endocrinol. 43, 155-163. Peterson, G.L. (1977) Anal. B&hem. 83, 340-356. Porcheron, P., Moriniere, M., Grassi, J. and Pradelles, P. (1989) Insect. Biochem. 19, 117-122. Pradelles, P., Grassi, J., Chabardes, D. and Guiso, N. (1989) Anal. Chem. 61,447-453. Rajkumar, E, Chadrese, P.J., Ly, H. and Murphy, B.D. (1991) J. Endocrinol. 130, 273-280. Schumacher, H., Miiller, D. and Mukhopadhyay, AK. (1992) Mol. Cell. Endocrinol. 90, 47-52. Sedlmeier, D. 11985) Am. Zool. 25,223X32. Sedlmeier, D. and Fenrich, R. (1993) J. Exp. Zool. 265,448-453. Sedlmeier, D. and Keller, R. 0981) Gen. Camp. Endocrinol. 45, 82-90. Smith, W. and Sedlmeier, D. (1990) Invert. Reprod. Dev. 18,33-37. Soumoff, C. and O’Connor, J.D. (1982) Gen. Camp. Endocrinol. 48, 432-439. Spaxiani, E. (19901 in Mo~hogenetic Hormones of Arthropods (Gupta, A.P., eds.), Vol. I, part 2, pp. 233-267, Rutgers University Press, New Brunswick, New Jersey. Spaxiani, E., Rees, H.H., Wang, W.L. and Watson, R.D. (1989) Mol. Cell. Endocrinol. 66, 17-25. Watson, R.D. and Spaziani, E. (1985) Gen. Comp. Endocrinol. 59, 140-148. Webster, S.G. (1986) Gen. Camp. Endocrinol. 61, 237-247. Webster, S.G. (19911 Proc. R. Sot. London B 244, 247-252. Webster, S.G. (19931 Proc. R. Sot. London B 251, 53-59. Webster, S.G. and Keller, R. (19861 J. Comp. Physiol. B 156,617-624. Webster, S.G. and Keller, R. (1989) in Ecdysone: from Chemistry to Mode of Action (J. Koolman, ed.), pp. 211-216, Thieme, Stuttgart.