DEVELOPMENTAL

BIOLOGY

144,%%378

(1991)

Hormonal Regulation of Epidermal Metamorphosis in Vitro: Control of Expression of a Larval-Specific Cuticle Gene KIYOSHI HIRUMA, JIM HARDIE,~ AND LYNN M. RIDDIFORD Department of Zoology,

University

of Washington, Seattle, Washington 98195

Accepted December 28, 1990 Fourth (penultimate) instar larval epidermis of the tobacco hornworm, Manduca se&a, was used to develop an in vitro culture system to study the hormonal control of metamorphosis at both the cellular and the molecular level. Immediate exposure to 4 X 10m6M 20-hydroxyecdysone (20-HE) for more than 8 hr, followed by hormone-free medium for 24 hr, caused the formation of a new larval cuticle. By contrast, incubation in hormone-free medium for more than 24 hr prior to exposure to 20-HE allowed pupal cuticle synthesis. The cessation of expression of the larval-specific cuticular gene LCP-14 occurred rapidly in response to 20-HE during the larval molt in vitro (half-life: ea. 6 hr), even in the presence of 3 x 10-s h4 JH I. This suppression by 20-HE was prevented by cycloheximide, indicating that 20-HE does not act directly on this gene. Incubation with ol-amanitin showed that the half-life of LCP-14 was more than 10 hr. Thus, 20-HE must both suppress gene transcription and destabilize the mRNA. LCP-14 mRNA subsequently reappeared 24 hr after exposure to hormone-free medium, indicating that suppression was temporary. By contrast, when JH and its effects were absent after preincubation in hormone-free medium for 48 hr, 20-HE caused permanent suppression of LCP-14 mRNA, since the mRNA did not reappear after removal of 20-HE. Exposure of Day 2 fifth instar larval epidermis to 3 X lo-’ M 20-HE, which causes pupal commitment in the absence of JH I, also permanently suppressed LCP-14 gene expresSiOn.

0 1991 Academic

Press, Inc.

INTRODUCTION

Metamorphosis is one of the most fascinatingphenomena in animal development. How do differentiated cells switch their ongoing gene expression to make the new products that will contribute to the new form? In insects metamorphosis is governed primarily by two hormones, juvenile hormone (JH) and ecdysone. Ecdysone and its active metabolite, 20-hydroxyecdysone (20-HE), initiate molting, whereas JH determines the character of the molt with high amounts allowing a larval molt and its absence allowing metamorphosis (Riddiford, 1985). Thus, the larval epidermal cells produce a larval cuticle in response to ecdysteroids in the presence of JH. When JH disappears in the final larval instar, the lepidopteran epidermis responds to a low level of ecdysteroids by becoming pupally committed; these cells then form only a pupal cuticle when next exposed to a molting concentration of ecdysteroids, even when JH is present (Riddiford, 1976,1978; Tsutsumiuchi et aZ., 1989). In Lepidoptera, many, but not all, of the cuticular proteins are stage-specific as the larval, pupal, and adult cuticles differ considerably in structure (Kiely and Riddiford, 1985; Cox and Willis, 1985, 1987; Wolfgang and Riddiford, 1986). Therefore, one can study the hormonal ’ Present address: Agricultural and Food Research Council Link Research Group in Aphid Biology, Department of Biology, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, England.

regulation of a stage-specific cuticle gene as a model for metamorphosis of the insect epidermis. Using the penultimate (fourth) instar larval epidermis, we now define the hormonal conditions necessary to obtain either a larval or a pupal molt in vitro. We then show that under these conditions the gene encoding a 14-kDa larval endocuticular protein (LCP-14) (Rebers and Riddiford, 1988) is always repressed by a molting concentration of 20-HE. When JH is present during this exposure, the gene is reexpressed upon removal of 20HE. By contrast, exposure to 20-HE in the absence of JH causes the permanent suppression of this gene. MATERIALS

Experimental

AND METHODS

Animals

Larvae of the tobacco hornworm, Manduca sexta, were reared individually on an artificial diet at 25.5”C in a 12L:12D photoperiod as described by Truman (1972) and Bell and Joachim (1976). Lights off, the beginning of a new day, was set at 0O:OO Arbitrary Zeitgeber Time (AZT) (Pittendrigh, 1965). Allatectomy (removal of the corpora allata, the source of JH) was performed as described by Hiruma (1980) after the larvae were anesthetized by submersion in water for 15 to 30 min. Gate II fourth instar larvae were allatectomized at 18:00-19:00 AZT on Day 2 (before spiracle apolysis). 369

0012-1606/91$3.00 Capyright All rights

0 1991 by Academic Press. Inc. of reproduction in any form reserved.

370

DEVELOPMENTALBIOLOGY vOLUME144,1991

Culture Procedure The dorsal abdominal integument was dissected free of most muscle and fat body under Manduca saline (Riddiford et al., 1979). Pieces (2 X 5 mm on Day 0,2.5 X 6 mm on Day 1,3 x ‘7 mm on Day 2 fourth larvae, and 7 x 15 mm on Day 2 fifth larvae) of freshly dissected integument were cultured in 0.5 ml of Grace’s medium (GIBCO) per culture well (Linbro Disposotrays, Flow Laboratories) on a glass wool support and incubated at 25.5”C in a 95% 02-5% CO, atmosphere on a slow rotary shaker (Hiruma and Riddiford, 1984). After culture, the epidermis was removed and stored at -70°C until further processing. Dissections and cultures were initiated between 18:00-20:00 AZT for the fourth (penultimate) instar and at 13:00-15:00 AZT for the fifth (last) instar larvae. To ensure the absence of Gate I Day 2 fourth instar larvae that would have already initiated the molt, we selected only larvae which showed no spiracle apolysis. 20-HE (Rohto Pharmaceutical Co.) was dissolved in 10% isopropanol and the exact concentration measured spectrophotometrically at 240 nm (t = 12,670) (Meltzer, 1971). This stock solution was added to the medium to give the desired concentration and then sterilized through a Millipore filter (0.22 pm). JH I (Sigma) was dissolved in Grace’s medium by sonication as described by Riddiford et aZ. (1979). All glassware, glass wool, and culture wells for JH experiments were coated with polyethylene glycol20,OOO (J. T. Baker Chemical Co.). Cycloheximide (Sigma) (a protein synthesis inhibitor) and aamanitin (Sigma) (a mRNA synthesis inhibitor) were dissolved in Grace’s medium shortly before the culture and then sterilized through a Millipore filter. For determination of inhibition of protein synthesis, integument of Day 2 fourth instar larvae was cultured in 0.5 ml Grace’s medium with various concentrations of cycloheximide for 2 hr, and then for 1 hr in 50 ~1 medium containing the same concentrations of cycloheximide without leucine to which 50 PCi [3H]leucine (L-[3,4,5‘H(N)]-, 146.7 Ci /mmole, New England Nuclear) was added. Radioactivity of TCA-precipitable 3H-labeled protein was counted per microgram of total protein. Protein concentrations were determined by BCA Protein Assay Reagent (Pierce Chemical Co.). Bovine serum albumin (Sigma) was used as a standard. Scoring System for Newly Synthesized

Cuticle in Vitro

After the required culture period, newly synthesized cuticles were scored as “larval” if they had larval setae, surface papillae, and a thick rubbery texture (Fig. 1A) or as “pupal” if they had no setae and a smooth surface containing pockmark structures (Fig. 1C) that were often tanned. Scoring was done under either the dissecting or the compound microscope.

The newly synthesized larval cuticle of each piece was scored as shown in Table 1. RNA Extraction

and Hybridizations

RNA was extracted from the epidermis in 7.6 M guanidine-HCl in 0.1 M potassium acetate buffer, pH 5, as described by Cheley and Anderson (1984). The RNA pellet was suspended in TE [lo m2M Tris-HCl (pH S), 1 mM EDTA (pH S)], reprecipitated, and then resuspended in TE. Integrity of the RNA was confirmed by nondenaturing 1% agarose gel electrophoresis. Total RNA was electrophoresed on 1.5% agaroseformaldehyde gels (Lehrach et aL, 1977) and transferred to nylon membranes (Hybond-N, Amersham) for hybridization. Dot blot analysis of 1 pg total RNA was performed after denaturation by formaldehyde on nylon membranes (Kafatos et aL, 1979). The 4.6-kb BgZII/BamHI fragment of the genomic LCP-14 gene (26-10) (Rebers and Riddiford, 1988), which contains three of the four exons (exon 2 to 4), was labeled with [32P]dATP (600 Ci/mmole, New England Nuclear) by hexanucleotide priming of DNA fragments directly in low melting agarose (Feinberg and Vogelstein, 1984). Specific activity of the probe was about 10’ dpm/ fig DNA. Hybridizations (5 X lo6 dpm/ml 32P-labeled probe) and washing conditions were as described by Horodyski et al. (1989). After hybridization, the filters were exposed to Kodak XAR or XRP film with an intensifying screen at -70°C. Quantification of dot-blot hybridization was performed by densitometry using a Quick Scan R & D (Helena Laboratories). To standardize the quantification, 1, 0.5,0.25, and 0.125 pg of a stock RNA derived from Day 2 fourth instar epidermis were loaded on each dot-blot. RESULTS

Induction

of Larval

Cuticle Forwmtion

in Vitro

When integuments of Day 2 fourth (penultimate) instar larvae were cultured in the presence of 2 pg/ml 20-HE (the peak concentration of ecdysteroid during a larval molt (Curtis et aZ., 1984)), apolysis began 6 hr after the onset of incubation. Although apolysis occurred, no new cuticle observable under the light microscope was found when epidermis was cultured with 2 pg/ml20-HE for 6 to 24 hr (Table 2). When cultured for 38 hr or longer with 20-HE, the epidermis synthesized a new larval cuticle (Table 2). The cuticles synthesized during the 38- to 73-hr cultures with 20-HE were thin, had setae that were often very short, and occasionally some papillae that are normally seen in the integument of fifth instar larvae (score 2). None formed a normal

HIRUMA, HARDIE, AND RIDDIFORD

Metamwphosis

of

an Insect Integument

in Vitro

FIG. 1. Typical newly synthesized larval and pupal cuticle. (A) Newly synthesized larval cuticle (score 3) produced in vitro. Integument of Day 2 fourth (penultimate) instar larvae was cultured in 2 pg/ml20-HE for 17 hr followed by hormone-free medium for 24 hr. (B) Cuticle of freshly ecdysed Day 0 fifth instar larva. (C) Pupal cuticle produced in vitro. Integument of Day 2 fourth instar larvae was cultured in hormone-free medium for 48 hr and then exposed to 3 rg/ml20-HE for ‘72 hr followed by culture in hormone-free medium for ‘72 hr. (D) Cuticle of freshly pupated animal. S, seta; M, melanized dot; P, pock marks. Bar = 300 pm.

thick cuticle (score 3). When the integument was cultured in the absence of 20-HE, apolysis never occurred and no new cuticle was formed after ‘73 hr (Table 2), or even after 168 hr (N = 36). TABLE 1 SCOR~G SYSTEM FOR NEWLY SYNTHESIZED LARVAL CUTICLE

Since Day 2 fourth epidermis failed to synthesize a normal thick larval cuticle in the continuous presence of 20-HE, we cultured the integument with 2 pg/ml20-HE TABLE 2 EFFECT OF CONTINUOUS20-HE ON APOLYSIS OF THE INTEGUMENTS OF DAY 2 FOURTH INSTAR LARVAE IN VITRO

IN VITRO

Score 0

1

2 3

Description No newly synthesized cuticle; in some cases, old cuticle is apolysed. Very thin new cuticle that is impossible to separate intact from epidermis; equivalent to the newly forming cuticle 12-13 hr after head capsule slippage (HCS). Thin cuticle that is difficult to separate intact from epidermis; larval setae (sometimes fairly short) and occasionally papillae are formed. Fairly thick cuticle (5-7 pm) that is easily separable from epidermis; both larval setae (similar length to that of the original explanted cuticle) and papillae are formed (see Fig. 1A); equivalent to the new fifth instar cuticle 17-19 hr after HCS (lo-12 hr before ecdysis).

Apolysis Culture time (hr) 0

3 6 14 17 24

38 41 53 73

No. 32 16 16 16 42 42 16 16 16 16

2 pg/ml 20-HE

(new cuticle, scorea)

No 20-HE

-

(No) (No)

-

f +

(NoI (No)

-

+ + + + +

(No) (No) (Yes, (Yes, (Yes,

-

+

(Yes,

1)

2) 2) 2)

-, no apolysis; f, partial apolysis; +, apolysis. a See Table 1 for description of the scores. Cuticle formation scored only with the light microscope.

-

was

372

DEVELOPMENTAL BIOLOGY

VOLUME 144, I991 31

f

/

2 I,

.-.-.

/

6

J,-,/ 0

,

6

12

,

IS

,

24

r”

I

36

48

FIG. 2. Duration of exposure to hormone-free medium required for new cuticle synthesis after exposure to 2 pg/ml20-HE for 17 hr. Integuments were explanted from Day 2 fourth (penultimate) instar larvae (before the ecdysteroid rise for larval molting). Points represent averages + SD (within the symbol, if not shown) (N = 10-45).

for 17 hr and then transferred it to hormone-free medium for various periods. Figure 2 shows that thick new larval cuticles, having both setae and papillae (score 3), were synthesized after 17 hr or more in hormone-free medium. To determine the duration of exposure to ZO-HE required for synthesis of new cuticle, we incubated integuments of fourth instar larvae with 2 pg/ml 20-HE for varying times and subsequently in hormone-free medium for 24 hr. Figure 3 shows that 20-HE stimulated

Day I

5 w s

2

2

I

,

,

(

I 2C

8 24

Day 2

OJ.’ 0

4

I 0 Hours

I 12

I 16

1, 0.3

I.0 20HE

Hours after removal of 2OHE

,

o1 . .-. 0 0.1

in 2OHE (2pg/ml)

FIG. 3. Synthesis of new larval cuticle by epidermal tissue from Day 0, 1, and 2 fourth (penultimate) instar larvae after exposure to 2 pg/ml 20-HE for the designated time followed by culture in hormone-free medium for 24 hr. New cuticle was scored as outlined in Table 1. Points represent averages f SD (within the symbol, if not shown) (N = 12-39).

II 2.0

3.0

(&ml)

FIG. 4. Dose-response curve for 20-HE induction of larval cuticle synthesis in vitro. Pieces of integument of Day 2 fourth (penultimate) instar larvae were cultured with various concentrations of 20-HE for 17 hr and then transferred to hormone-free medium for 24 hr. New cuticle was scored as outlined in Table 1. Doses of 0.1 and 0.3 pg/ml 20-HE induced only apolysis. Points represent averages f SD (within the symbol, if not shown) (N = 10-24).

new larval cuticle synthesis by the explants taken from Day 0 fourth instar larvae after exposure for 12 hr or more. In contrast, Day 1 or Day 2 explants required only an 8-hr exposure. In addition, all Day 2 explants produced a new cuticle after only a 4-hr exposure to 20-HE, although the newly synthesized cuticles were fairly thin and most of them lacked surface papillae. Thus, the epidermal cells become more responsive to 20-HE during the first 3 days of the fourth instar and in all cases the newly synthesized cuticle was larval. Even when the hormone-free period was extended for 96 hr after incubation with a higher concentration of 20-HE (3 pg/ml) for 24 hr, the newly synthesized cuticles were all larval with setae and papillae (N = 12-39). Although the cuticles resulting from in vitro culture were larval, even those described as thick (score 3) were thinner and had shorter setae than newly ecdysed fifth larval cuticle (cf. Figs. 1A and 1B). The thickness (5-7 pm) and the surface structures observed by light microscopy were almost identical to fifth instar larval cuticle lo-12 hr before ecdysis. The lengths of the setae produced in vitro were essentially the same as those of Day 0, 1, and 2 fourth instar cuticles, respectively. Melanization did not occur in the newly synthesized cuticle, even where melanized dots are normally seen in intact larvae (Fig. 1B). To determine the concentration of 20-HE required for the synthesis of the larval cuticle, we incubated Day 2 fourth instar larval integument in varying concentrations of 20-HE for 17 hr followed by hormone-free medium for 24 hr. Figure 4 shows that 1 pg/ml20-HE was sufficient to induce a thick new larval cuticle. This ecdysteroid concentration is about half the maximal concentration of hemolymph ecdysteroid (2-3 pg/ml20-HE equivalents) during a molt (Curtis et aL, 1984; Bollenbather et al., 1987). Lower concentrations of 0.1 and 0.3 pg/ml 20-HE induced only apolysis, and less than 0.1 pg/ml20-HE had no visible effect on the explants (data not shown).

HIRUMA,

HARDIE,

AND RIDDIFORD

Metamorphosis

56 loo

0 hr (22)

50

0 L--l P F7LPCL

1001

L

36hr (24)

PP>LWL 1

L

46hrt76)

P RLPCL ]

L

72hr (36)

50

0

P PaLPCL

L

P P>LWL

P P>LPL PLP L (>50% pupal) or P < L (>50% larval) as determined by the proportion of the surface area with larval or pupal characters. FIG.

Pupal Cuticle Formation in Vitro

Since the JH titer is high throughout the fourth (penultimate) larval instar in Manduca (Fain and Riddiford, 1975), it is possible that retention of JH itself or its long-lasting effects on the epidermal cells (Mitsui and Riddiford, 1978; Nijhout, 1975) allowed the formation of a new larval cuticle in response to 20-HE in the absence of JH in the culture medium. To test this hypothesis, we incubated integument of Day 2 fourth instar larvae in hormone-free medium for varying periods followed by exposure to 3 fig/ml 20-HE for 72 hr and then to hormone-free medium for 10 days. Figure 5 shows that immediate exposure to 20-HE induced only larval cuticle, but after only 12 hr in hormone-free medium prior to 20-HE some pupal characters appeared. As the hormone-free period was extended, an increased proportion of new cuticles showed pupal characters. After 96 hr of preincubation, all formed pupal cuticle. Pupal cuticles were synthesized during the 72-hr exposure to 20-HE, but their thickness increased during the subsequent exposure to hormone-free medium. Most of the pupal cuticle produced in vitro had pock structures (Fig. lC), which are normally present in pupal cuticle (Roseland and Riddiford, 1980) (Fig. 1D) and also in precocious pupal cuticle that is formed after allatectomy of fourth instar larvae (data not shown).

of an In-sect Integument

in Vitro

373

When either 1 or 0.1 pg/ml JH I was present in the medium during the initial incubation, the subsequent exposure to 20-HE elicited only larval cuticle without any pupal structures (Fig. 5). By contrast, exposure to 0.01 pg/ml JH I [3 X lo-* iV, approximately the concentration present in vivo at the outset of the molt (Fain and Riddiford, 1975)] for 48 hr was only partly effective in preventing the subsequent production of pupal cuticle in response to 20-HE (Fig. 5). A lo-fold lower concentration of JH I had no effect (data not shown). Apparently, during the initial incubation in hormone-free medium, JH in the epidermis was metabolized and/or its effects were decayed so that in response to 20-HE the epidermis produced pupal cuticle. In many cases, the cultured pupal cuticle tanned. Tanning began in the hormone-free medium 4 days after withdrawal of 20-HE, and the proportion of tanning increased to a maximum 2 to 5 days later (data not shown). Tanned pupal cuticle was not formed in the continuous presence of 20-HE for 14 days after incubation in hormone-free medium for 36 hr, although untanned, thin transparent pupal cuticles were synthesized (N = 12). Hormonal Regulation of the Larval Cuticle Protein (LCP-14) Gene Expression in Vitro Larval molt. Figure 6 shows that the amount of LCP14 mRNA declines as the ecdysteroid titer rises for the fourth instar molt. The level remains low until 16 hr after head capsule slippage (HCS) at which time it begins to rise to a plateau near the time of ecdysis. This increase coincides with the major decline in ecdysteroid titer. Interestingly, this timing is similar to that of the onset of lamellate endocuticle deposition, which occurs about 14 hr after HCS (Wolfgang and Riddiford, 1986).

FIG. 6. Expression of LCP-14 gene in the epidermis during the last larval molt of Manducu. Expression was determined by dot-blot analysis of total epidermal RNA using the 4.6-kb Bg1WBamHI fragment of the genomie clone (26-10) (Rebers and Riddiford, 1988) as a probe. Relative expression was referred to the expression in the epidermis of Day 2 fourth instar larvae (18:OO AZT) as 10. Points represent averages k SD (N = 4-5). Cellular events occurring in the integument are from Wolfgang and Riddiford (1986). Ecdysteroid titer is from Curtis et al. (1984) and Bollenbacher et al. (1987). HCS, head capsule slippage.

374

DEVELOPMENTALBIOLOGY

Culture

VOLUME144,1991

Time (hr)

Culture

Time (hr)

FIG. 7. Hormonal regulation of LCP-14 gene expression in fourth instar larval epidermis during a larval molt in vitro. Integuments of Day 2 fourth (penultimate) instar larvae (l&O0 AZT) were cultured. (A) Effects of 20-HE (2 pg/ml) and JH I (0.01, 0.1, or 1 pgg/ml as designated) on the expression of the mRNA. (B) Effects of 10 pg/ml cycloheximide (CHX) on the suppressive action of 20-HE (2 pg/ml) after a 6-hr culture. 4c md “dNot significantly different at 0.1 > P > 0.05 and 0.5 > P > 0.1, respectively. b%ignificantly different at P < 0.001. (C) Stability of the LCP-14 mRNA in vitro. Transcription of new mRNA was blocked by 3 rig/ml cu-amanitin, and the mRNA levels were determined in the presence or absence of 10 rg/ml cycloheximide (CHX). The 4.6-kb BglII/BamHI fragment of the genomic clone of LCP-14 (26-10) (Rebers and Riddiford, 1988) was used as a probe. Relative expression was determined as described in Fig. 6. Bars represent averages + SD (N = 4-12).

When Day 2 fourth instar epidermis was incubated in hormone-free medium for 48 hr, LCP-14 mRNA levels remained high (Fig. 7A). When the incubation was extended to 120 or 168 hr, the epidermis lost 50-60s of this mRNA. The presence of 1 pg/ml JH I had no effect on this loss. By contrast, the addition of 2 pg/ml 20-HE caused a rapid loss of the mRNA (half-life = about 6 hr) so that by 14 hr only trace amounts were detectable. Again 0.01 to 1 pug/ml JH I had no effect on this action of 20-HE (Fig. 7A). To determine whether 20-HE was acting directly on the LCP-14 gene or whether protein synthesis was required, we incubated epidermis with 20-HE in the presence of cycloheximide, a protein synthesis inhibitor. Preliminary experiments showed that exposure of this Day 2 fourth epidermis to concentrations of greater than 10 yg/ml cycloheximide for longer than 6 hr caused extensive cell damage. Exposure to 10 pg/ml cycloheximide for 6 hr caused 81% inhibition of protein synthesis (N = 5), but the cells remained viable as [3H]leucine was incorporated into proteins at the same rate as in controls 24 hr after removal of the cycloheximide (N = 5). When Day 2 fourth instar epidermis was exposed to 20HE in the presence of 10 pg/ml cycloheximide or to cycloheximide alone, LCP-14 mRNA remained high (Fig.

7B). Thus, the suppressive effect of 20-HE on LCP-14 expression appears to be indirect via a protein induced by 20-HE. To determine the normal half-life of LCP-14 mRNA, we incubated fourth instar larval epidermis with 3 pg/ ml cw-amanitin and then assessedLCP-14 mRNA levels at various times. This concentration of a-amanitin is sufficient to block the 20-HE-regulated onset of transcription of dopa decarboxylase (DDC) in this epidermis (Hiruma and Riddiford, 1986, and unpublished). Figure 7C shows that even after a lo-hr exposure LCP-14 mRNA remained high, indicating a half-life of much longer than 10 hr. The addition of cycloheximide had no effect. By the end of the lo-hr exposure to cz-amanitin, the epidermis was fragile and was showing signs of cell death. Therefore, longer exposures to determine the half-life were not done. In view of the prolonged stability of LCP-14 mRNA under these conditions, its rapid disappearance (half-life = cit. 6 hr) in the presence of 20-HE (Fig. 7A) indicates that 20-HE must cause destabilization of the existing mRNA as well as a suppression of transcription. To mimic the molting rise and then the decline of ecdysteroid during the larval molt, we cultured Day 2 fourth instar larval integument with 2 pg/ml20-HE for

HIRUMA,

HARDIE,

AND

RIDDIFORD

Metamorphosis

LCP-14

/

:

/

j

i-l

CtNA

Hours after removal of 2OHE

FIG. 8. Induction of LCP-14 mRNA during a larval molt in vitro as a function of time in hormone-free medium after exposure of Day 2 fourth instar integument to 2 fig/ml 20-HE for 1’7 hr. The 4.6-kb &$II/BamHI fragment of the genomic clone (26-10) (Rebers and Riddiford, 1988) was used as a probe. (A) Quantitative analysis by dotblot experiments. Relative expression was determined as described in Fig. 6. Bars represent averages f SD (N = 4-8). (B) Northern hybridization showing that the same RNA species appeared 24 and 48 hr after removal of 20-HE (top). The same blot was hybridized with DrosophilurDNA (bottom). Three micrograms of total RNA were loaded in each lane.

17 hr followed by incubation in hormone-free medium. Figure 8A shows that incubation with 20-HE for 17 hr caused a loss of LCP-14 mRNA; this RNA increased about 24 hr after removal from 20-HE. By 72 hr in hormone-free medium, the level of expression was similar to that in epidermis of Day 2 fourth instar larvae. The size (640 bp) of this new mRNA was the same as that of the RNA seen in intact larvae (Fig. 8B), indicating that the same mRNA species was induced. Reprobing with Drosophila ribosomal DNA shows that the variation in LCP-14 mRNA levels was not an artifact of unequal loading. Thus, the increase in ecdysteroid titer during a larval molt causes the loss of LCP-14 mRNA, and the subsequent decline of ecdysteroid is required for its reexpression. Pupal molt. Figure 5 showed that when Day 2 fourth (penultimate) larval integuments were cultured in hormone-free medium for 48 hr and then in 3 pg/ml20-HE for ‘72 hr, followed by hormone-free medium, most of the explants produced pupal cuticle. Analysis of the LCP-14 mRNA level under these experimental conditions showed that the mRNA level after 48 hr in hnrmnne- _

of an Insect Integument in Vi’itro

375

free medium was 80% of the level at the time of explantation (Fig. 9). Subsequent incubation with 3 pg/ml20HE for 72 hr caused an almost complete loss of this mRNA (Fig. 9), whereas there was only a slow decline to 40% of the initial level in hormone-free medium. When these pieces were then transferred to hormone-free medium, they formed pupal cuticle (Fig. 5) and no LCP-14 reappeared (Fig. 9). During the fifth instar LCP-14 mRNA increases during the first 3 days, then begins to fall during the final day of feeding, and disappears after wandering (Rebers and Riddiford, 1988). This decline coincides with the two peaks of ecdysteroid (about 30 and 75 rig/ml 20-HE equivalents) on Days 2 and 3 (Wolfgang and Riddiford, 1986) in the absence of JH (Baker et ab, 1987). To determine whether the fall of JH and/or these low levels of ecdysteroid caused this decline, we cultured Day 2 epidermis under various hormonal conditions for 24 hr. In this case, after 24 hr in hormone-free medium LCP-14 mRNA was reduced to 57% of its initial level (Fig. 10). Exposure to 50 or 100 rig/ml 20-HE caused a further decrease to about 25%, whereas 500 rig/ml 20-HE caused a decline to 12% of the initial level, similar to the level seen in wandering stage epidermis (Fig. 10). The presence of 1 pug/ml JH I, which prevents the 20-HE-induced pupal commitment of this epidermis (Riddiford, 1976), prevented the decline in response to 20-HE, but had no effect on the loss of this RNA in hormone-free medium, indicating that this loss is not due to the lack of JH but likely to either some aspect of the culture conditions or exposure to low 20-HE in the absence of JH in (Wolfgang and Riddivivo on Day 2 before explantation

IO-

a b

c

c

O-

NO culture

NH -~ 48hr

S-t-l NH 48hNH 72hE 120hr

Culture

NH 48hNH 72hE 48hNH 168hr-

48ht.W 72hE 72hNH 192hr

Time

FIG. 9. LCP-14 gene expression during a pupal molt in vitro. Pieces of integument of Day 2 fourth (penultimate) instar larvae were explanted into hormone-free medium for 48 hr and then cultured with 3 rg/ml20-HE for 72 hr followed by hormone-free medium for 48 or 72 hr. NH, no hormone; E, 20-HE. The mRNA level of LCP-14 was determined by a dot-blot analysis using the 4.6-kb BglII/BumHI fragment of the genomic clone (26-10) (Rebers and Riddiford, 1988) as a probe. Relative expression was determined as described in Fig. 6. Bars represent averages f SD (N = 4-12). ‘sband‘-‘Significantly different at 0.02 > P > 0.01 and 0.001 > P, respectively.

376

DEVELOPMENTAL

Day 2 5th

!OHE (rig/ml)

20HE (q/ml) JH I (I&ml)

BIOLOGY

wo

FIG. 10. Regulation of LCP-14 mRNA by 20-HE and JH I in vitro in fifth instar larval epidermis. Pieces of integument of Day 2 fifth instar larvae (5.3-6.7 g, male) at 13:00-15:00 AZT were cultured under various hormonal conditions for 24 hr followed by RNA extraction. Quantitative analysis of the mRNA level was performed by dot-blot hybridization using the 4.6-kb &$II/BamHI fragment of the genomic clone (26-10) (Rebers and Riddiford, 1988) as a probe. WO, Day 4 fifth instar larvae (2200 AZT) that had entered a wandering stage. Relative expression was determined as described in Fig. 6. Bars represent averages + SD (N = 4). *b Md “‘Significantly different at 0.005 > P > 0.002 and 0.02 z P > 0.01, respectively. P*dend d*eNot significantly different at 0.5 > P > 0.2 and 0.1 > P > 0.05, respectively.

ford, 1986). These results indicate that the loss of LCP14 mRNA by the time of wandering is due to the small peaks of ecdysteroid in the absence of JH. DISCUSSION

Munduca fourth instar epidermis normally synthesizes a new larval cuticle in response to an ecdysteroid rise at the time of molting. The determination to produce larval rather than pupal cuticle is controlled by JH, which is high when ecdysone is released at the end of the fourth instar (Fain and Riddiford, 1975, 1976). The current studies show that this epidermis can synthesize either larval or pupal cuticle in response to ZO-HE in vitro, depending on the hormonal milieu. Moreover, the gene encoding the larval cuticle protein LCP-14 proved to be a good molecular marker for studying this switch as it was expressed only during larval cuticle production. New larval cuticle formation by Day 2 fourth instar epidermis in vitro required only 8 hr of exposure to 4.2 x lop6 M 20-HE [nearly the peak of molting concentration (Curtis et aZ., 1984)], followed by the absence of 20HE, just as has been found for pupal cuticle deposition by Drosophila imaginal discs (Fristrom et cd, 1982; Doctor et al, 1985). The presence of JH in the culture medium was not required when the tissue had been freshly explanted, presumably because JH and/or the factors it induces were still present in the cells when 20-HE induced the molt (Nijhout, 1975; Fain and Riddiford, 1977). Under these conditions the cuticle formed in vitro

VOLUME

16&1991

was very similar to that formed in viva except for the length of the setae and a lack of melanized dots. Failure of growth of setae induced in vitro has also been seen in Drosophila imaginal discs (Fristrom and Fristrom, 1975; Milner, 1977). Presumably a factor(s) necessary for growth is unavailable in these defined culture media. The lack of cuticular melanization is due to the presence of P-alanine coupled with the lack of either dopa or dopamine, a necessary precursor for melanin, in Grace’s medium (Hiruma and Riddiford, 1984; Hiruma et al., 1985). The abundance of LCP-14 mRNA during a larval molt in vitro mimicked that seen in viva (Rebers and Riddiford, 1988; Fig. 6). Its high levels in intermolt epidermis were maintained during culture in hormone-free medium and under these conditions the mRNA was quite stable with a half-life longer than 10 hr. When 20-HE was added, LCP-14 mRNA disappeared much more rapidly (half-life: ca. 6 hr). This destabilization of the mRNA was dependent on continued protein synthesis as has been found in several eukaryotic systems where regulated mRNA destabilization occurs, such as that of cmyc or histone mRNA (see reviews, Cleveland and Yen, 1989; Nielsen and Shapiro, 1990). Whether this requirement is for the synthesis of a short-lived specific RNase or for a direct coupling of degradation to translation is not clear for any of these systems. Presumably, 20-HE also rapidly suppressed transcription of LCP-14 mRNA as it does in the case of the salivary gland glue protein mRNA (Crowley and Meyerowitz, 1984). In the case of LCP-14, this suppressive effect on transcription persisted as long as 20-HE was present. Upon removal, the mRNA reappeared although after a 24-hr delay. The decline of ecdysteroid is responsible for coordinating many developmental events at the end of a molt, such as chitin deposition (Fristrom et aL, 1982), procuticular protein synthesis (Doctor et ah, 1985), DDC synthesis (Hiruma and Riddiford, 1985, 1986) and transcription (Clark et uZ., 1986; Hiruma and Riddiford, 1990), preecdysial pigmentation (Schwartz and Truman, 1983; Hiruma et aL, 1985), muscle degeneration (Schwartz and Truman, 1983), neuronal death (Bennett and Truman, 1985), release of eclosion hormone (Truman et ah, 1983), and appearance of the eclosion hormone and cyclic GMP-regulated phosphoproteins (Morton and Truman, 1988). Our results with DDC mRNA (Hiruma and Riddiford, 1990 and unpublished; Riddiford et aL, 1990) show that this continued suppression by 20-HE is also dependent on protein synthesis. After the critical time of induction of DDC mRNA by 20-HE either in viva or in vitro, continued exposure to 20-HE prevented DDC transcription, but addition of cycloheximide allowed accumulation of this mRNA within 3 hr (Hiruma and Riddiford, unpublished). Thus, 20-HE can -induce negative as well as positive regulatory fac-

HIRUMA, HARDIE, AND RIDDIFORD

Metamorphosis

tors (Ashburner et al., 1974; Thummel et al, 1990; Urness and Thummel, 1990; Segraves and Hogness, 1990). The half-lives of these factors in the absence of 20-HE can then determine the timing of expression of these 20HE-dependent genes (Riddiford et al, 1990). In order to produce pupal cuticle in Guo, larval epidermis first becomes pupally committed in response to low levels of 20-HE in the absence of JH and/or factors induced by JH, then it responds to higher concentrations of 20-HE with production of pupal cuticle (Riddiford, 1978; Mitsui and Riddiford, 1978). In vitro, pupal cuticle was only formed after fourth instar larval epidermis had been preincubated in hormone-free medium for 24 hr or more prior to exposure to 20-HE. Similarly, Day 2 fifth larval epidermis required 2 to 3 days of incubation in hormone-free medium before exposure to 20HE in order to form pupal cuticle (Mitsui and Riddiford, 1978). In both cases, the presence of JH or a JH analog during the preincubation period allowed only larval cuticle formation in response to 20-HE. Thus, during the initial period in hormone-free medium, the JH present in the cells at the time of explantation must be metabolized and any factors that it may have induced necessary for the production of larval cuticle in response to 20-HE also must disappear. Then the epidermis can respond to 20-HE by changing commitment and producing a pupal cuticle. The cessation of LCP-14 gene expression normally occurs at the time of wandering (Rebers and Riddiford, 1988), and our studies show that this cessation is due to exposure to a low level of (l-2 X lop7 M) 20-HE in the absence of JH, similar to that seen in viva on Day 2 and Day 3 of the fifth instar (Wolfgang and Riddiford, 1986). At low levels of 20-HE, the presence of 3 X 1O-6 M JH I prevents the ecdysteroid-induced suppression of this mRNA. Even as the concentration of 20-HE rises to lop6 &l, the mRNA did not decline. During preincubation of Day 2 fourth epidermis in hormone-free medium, LCP-14 mRNA remains quite high as the epidermis continues to produce endocuticle in vitro (Wolfgang and Riddiford, 1986). Once 20-HE was added, this RNA disappeared and did not reappear during pupal cuticle formation after the removal of 20HE. These in vitro results thus support the findings in vivo that LCP-14 mRNA is not expressed during formation of the pupal cuticle (Rebers and Riddiford, 1988). How 20-HE and JH interact to regulate the expression of such a larval intermolt gene is not known. Palli et al. (1990) have shown that a 29-kDa nuclear protein binds to the LCP-14 gene and that this protein shows the same developmental expression pattern as that of a 29kDa nuclear JH binding protein, but whether these are one and the same protein awaits the purification of the JH binding protein. If they are, then the binding of a JH-receptor complex directly to regulatory region(s) of

of an Insect

Integument

in Vitro

377

this gene may be necessary to prevent the permanent suppression by 20-HE. The present studies have shown that the penultimate instar lepidopteran epidermis is an excellent tissue in which to study the hormonal regulation of metamorphosis in vitro. Although the presence of JH does not alter the 20-HE-induced suppression of ongoing cuticular gene expression, it prevents the permanent cessation of that expression. Whether it alters the type of transcription factors induced by 20-HE to regulate these genes or acts directly on the larval genes awaits further study. We thank Dr. John Rebers for a critical reading of the manuscript. This work was supported by grants from NSF (DCB 85-10875), NIH (AI-12459), and USDA (89-37263-4350). REFERENCES ASHBURNER, M., CHIHARA, C., MELTZER, P., and RICHARDS, G. (1974). Temporal control of puffing activity in polytene chromosomes. Cold Spring Harbor Symp. &ant. BioL 38,655-662. BAKER, F. C., TSAI, L. W., REUTER, C. C., and SCHOOLEY,D. A. (1987). In viva fluctuation of JH, JH acid, and ecdysteroid titer, and JH esterase activity, during development of fifth stadium Manduca sexta. Insect B&hem. 17,989-996. BELL, R. A., and JOACHIM, F. G. (1976). Techniques for rearing laboratory colonies of tobacco hornworms and pink bollworms. Ann. EntcmoL Sot. Am.

69,365-373.

BENNE”IT, K. L., and TRUMAN, J. W. (1985). Steroid-dependent survival of indentifiable neurons in cultured ganglia of the moth Man&co sexta.

Science

229,58-60.

BOLLENBACHER, W. E., GRANGER, N. A., KATAHIRA, E. J., and O’BRIEN, M. A. (1987). Developmental endocrinology of larval moulting in the tobacco hornworm, Mar&case&a. J. Exp. Biol. 128, 175-192. CHELEY, S., and ANDERSON, R. (1984). A reproducible microanalytical method for the detection of specific RNA sequences by dot-blot hybridization. AnaL B&hem. 137,15-19. CLARK, W. C., DOCTOR, J., FRISTROM, J. W., and HODGETTS, R. B. (1986). Differential responses of the dopa decarboxylase gene to 20OH-ecdysone in Drosophila melanogaster. Dev. BioL 114,141-150. CLEVELAND, D. W., and YEN, T. J. (1989). Multiple determinants of eukaryotic mRNA stability. New Biologist 1,121-126. COX, D. L., and WILLIS, J. H. (1985). The cuticular proteins of Hydophora cecropia from different anatomical regions and metamorphic stages. Insect B&hem. 15,349-362. COX, D. L., and WILLIS, J. H. (1987). Analysis of the cuticular proteins of Hyalophora cecropia with two-dimensional electrophoresis. Insect Biochem. 17,457-468. CROWLEY,T. E., and MEYEROWITZ, E. M. (1984). Steroid regulation of RNAs transcribed from the Drosophila 68C polytene chromosome puff. Dev. BioL 102,110-121. CURTIS, A. T., HORI, M., GREEN, J. M., WOLFGANG, W. J., HIRUMA, K., and RIDDIFORD, L. M. (1984). Ecdysteroid regulation of the onset of cuticular melanization in allatectomized and black mutant Manduca sexta larvae. J. Insect PhysioL 30,597-606. DOCTOR,J., FRISTROM, D., and FRISTROM, J. W. (1985). The pupal cuticle of Drosophila: Biphasic synthesis of pupal cuticle proteins in vivo and in vitro in response to 20-hydroxyecdysone. J. Cell BioL 101, 189-200. FAIN, M. J., and RIDDIFORD, L. M. (1975). Juvenile hormone titers in the hemolymph during late larval development of the tobacco hornworm, Manduca sexta. BioL Bull. 149,506-521.

378

DEVE&OPMENTAL BIOLOGY

FAIN, M. J., and RIDDIFORD, L. M. (1976). Reassessment of the critical periods for prothoracicotropic hormone and juvenile hormone secretion in the larval molt of the tobacco hornworm Manduca sexta. Gen. Camp. E?ubcrinoL 30,131-141. FEINBERG, A. P., and VOGELSTEIN, B. (1984). Addendum. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biodmn. 137,266-267. FRISTROM, D., and FRISTROM, J. W. (1975). The mechanism of evagination of imaginal discs of Drosqphilu melanogaster. I. General considerations. Dev. BioL 43, l-23. FRISTROM, J. W., DOCTOR,J., FRISTROM, D. K., LOGAN, W. R., and SILVERT, D. J. (1982). The formation of the pupal cuticle by Drosophila imaginal discs in vitro. Dev. BioL 91, 337-350. HIRUMA, K. (1980). Possible roles of juvenile hormone in the prepupal stage of Mamestra bra&cue. Gen. Cwmp. Endocrinol. 41,392-399. HIRUMA, K., and RIDDIFORD, L. M. (1984). Regulation of melanization of tobacco hornworm larval cuticle in vitro. J. Exp. ZooL 230,393403. HIRUMA, K., and RIDDIFORD, L. M. (1985). Hormonal regulation of dopa decarboxylase during a larval molt. Dev. BioL 110,509~513. HIRUMA, K., and RIDDIFORD, L. M. (1986). Inhibition of dopa decarboxylase synthesis by 20-hydroxyecdysone during the last larval moult of Man&urn sexta. Insect Biochem. 16,225~231. HIRUMA, K., and RIDDIFORD, L. M. (1990). Regulation of dopa decarboxylase gene expression in the larval epidermis of the tobacco hornworm by 20-hydroxyecdysone and juvenile hormone. Dew. BioL 138, 214-224. HIRUMA, K., RIDDIFORD, L. M., HOPKINS, T. L., and MORGAN, T. D. (1985). Roles of dopa decarboxylase and phenoloxidase in the melanization of the tobacco hornworm and their control by 20-hydroxyecdysone. J. Camp. PhysioL 155B, 659-669. HORODYSKI, F. M., and RIDDIFORD, L. M. (1989). Expression and hormonal control of a new larval cuticular multigene family at the onset of metamorphosis of the tobacco hornworm. Dev. BioL 132, 292-303. HORODYSKI, F. M., RIDDIFORD, L. M., and TRUMAN, J. W. (1989). Isolation and expression of the eclosion hormone gene from the tobacco hornworm, Manduca sexta. Proc. NatL Acad Sci. USA 86,8123-812’7. KAFATOS, F. C., JONES, C. W., and EFSTRATIADIS, A. (1979). Determination of nucleic acid sequence homologies and relative concentrations by a dot hybridization procedure. Nucleic Acid Res. 7,1541-1552. KIELY, M. L., and RIDDIFORD, L. M. (1985). Temporal programming of epidermal cell protein synthesis during the larval-pupal transformation of Manduca sexta. Wilhelm Roux’s Arch. Dev. Biol. 194,325335. LEHRACH, H., DIAMOND, D., WOZNEY, J. M., and BOEDTKER, H. (1977). RNA molecular weight determination by gel electrophoresis under denaturing conditions, a critical reexamination. Biochemistry 16, 4743-4751. MELTZER, Y. L. (1971). “Hormonal and Attractant Pesticide Technology.” Noyes Data Corp., Parke Ridge, NJ. MILNER, M. J. (1977). The eversion and differentiation of Drosophila melanogaster leg and wing imaginal discs cultured in vitro with an optimal concentration of @-ecdysone. J. EmlnyoL Exp. MwphoL 37, 105-117. MITSUI, T., and RIDDIFORD, L. M. (1978). Hormonal requirements for the larval-pupal transformation of the epidermis of Manduca sexta in vitro. De-u. BioL 62, 193-205. MORTON, D. B., and TRUMAN, J. W. (1988). The EGPs: The eclosion hormone and cyclic GMP-regulated phosphoproteins. II. Regulation of appearance by the steroid hormone 20-hydroxyecdysone in Manduca se&a. J. Neurosci 8,1338-1345. NIELSEN, D. A., and SHAPIRO, D. J. (1990). Insights into hormonal control of messenger RNA stability. Mol. EndowinoL 4,953-957.

VOLUME 144,199I

NIJHOUT, H. F. (1975). Dynamics of juvenile hormone action in larvae of the tobacco hornworm, Manduca sexta BioL Bull. 149,568-579. PALLI, S. R., OSIR, E. O., ENG, W.-S., BOEHM, M. F., EDWARDS, M., KUJXSAR, P., UJVARY, I., HIRUMA, K., PRESTWICH, G. D., and RIDDIFORD, L. M. (1990). Juvenile hormone receptors in larval insect epidermis: Identification by photoaffinity labeling. Proc. NatL Acad 5’1% USA 87,796-800.

PI’ITENDRIGH, C. S. (1965). On the mechanism of entrainment of a circadian rhythm by light cycle. In “Circadian Clocks” (J. Aschoff, Ed.), pp. 276-291. North-Holland, Amsterdam. REBERS, J. E., and RIDDIFORD, L. M. (1988). Structure and expression of a Manduca sexta larval cuticle gene homologous to Drosophila cuticle genes. J. Mol. BioL 203,411-423. RIDDIFORD, L. M. (1976). Hormonal control of insect epidermal cell commitment in vitro. Nature 259,115-117. RIDDIFORD, L. M. (1978). Ecdysone-induced change in cellular commitment of the epidermis of the tobacco hornworm, Manduca sexta, at the initiation of metamorphosis. Gen. Comp. EndocrinoL 34, 438446. RIDDIFORD, L. M. (1985). Hormone action at the cellular level. In “Comprehensive Insect Physiology, Biochemistry, and Pharmacology” (G. A. Kerkut and L. I. Gilbert, Eds.), Vol. 8, pp. 37-84. Pergamon, Oxford. RIDDIFORD, L. M., BAECKMANN, A., HICE, R. H., and REBERS, J. (1986). Developmental expression of three genes for larval cuticular proteins of the tobacco hornworm, Manduca sexta. Dev. Biol. 118,82-94. RIDDIFORD,L. M., CURTIS, A. T., and KIGUCHI, K. (1979). Culture of the epidermis of the tobacco hornworm Manduca sexta. Tissue Cult Assoc. Man.

5,975-985.

RIDDIFORD, L. M., PALLI, S. R., and HIRUMA, K. (1990). Hormonal control of sequential gene expression in Manduca epidermis. In “Progress in Comparative Endocrinology” (A. Epple, C. G. Scanes, and M. H. Stetson, Eds.), pp. 226-231. Wiley-Liss, New York. ROSELAND, C. R., and RIDDIFORD, L. M. (1980). Analysis of a cuticular spacing pattern after metamorphosis In vitro of larval integument. In “Invertebrate Systems In V&-o” (E. Kurstak, K. Maramorosch, and A. Diibendorfer, Eds.), pp. 117-123. Elsevier/North Holland, Amsterdam. SCHWARTZ, L. M., and TRUMAN, J. W. (1983). Hormonal control of rates of metamorphic development in the tobacco hornworm Manduca sexta. Dev. Biol. 99,103-114. SEGRAVES, W. A., and HOGNESS, D. S. (1990). The E75 ecdysone-inducible gene responsible for the 75B early puff in Drosophila encodes two new members of the steroid receptor superfamily. GenRs Dev. 4, 204-219. THUMMEL, C. S., BURTIS, K. C., and HOGNESS, D. S. (1990). Spatial and temporal patterns of E7.4A transcription during Drosophila development. Cell 61,101-111. TRUMAN, J. W. (1972). Physiology of insect rhythms. I. Circadian organization of the endocrine events underlying the moulting cycle of larval tobacco hornworms. .J. Exp. BioL 57,805-832. TRUMAN, J. W., ROUNTREE, D. B., REISS, S. E., and SCHWARTZ, L. M. (1983). Ecdysteroids regulate the release and action of eclosion hormone in the tobacco hornworm, Manduca sexta (L). J. Insect PhysioL 29,895-900. TSUTSUMIUCHI, K., MIYADAI, T., and MITSUI, T. (1989). Action of ecdysone and juvenile hormone on larval-pupal transformation of the epidermis of the cabbage armyworm, Mamestra brassicae in vitro. J. Pesticide Sci. 14,345-350. URNESS, L. D., and THUMMEL, C. S. (1990). Molecular interactions within the ecdysone regulatory hierarchy: DNA binding properties of the Drosophila ecdysone-inducible E7&4 protein. Cell 63, 47-61. WOLFGANG, W. J., and RIDDIFORD, L. M. (1986). Larval cuticular morphogenesis in the tobacco hornworm, Manduca s&a, and its hormonal regulation. Dev. Biol. 113.305-316.