- Email: [email protected]
Synthesis of the major adult cuticle proteins of Drosophila melanogaster during hypoderm differentiation
DEVELOPMENTAL
BIOLOGY
107,420-431
(19%)
Synthesis of the Major Adult Cuticle Proteins of Drosophila melanogasfer during Hypoderm Differentiation ALAN H. ROTER,’ JANICE B. SPOFFORD, AND HEWSON The Committee
wn Developmental
Biology
Received
December
and Department 17, 1983;
of Biology,
accepted
in revised
University form
August
SWIFT
of Chicago,
Chicago,
Illinois
61.1637
27, 1984
Differentiating imaginal hypodermal cells of Drosophila melanogaster form adult cuticle during the second half of the pupal stage (about 40 to 93 hr postpupariation). A group of proteins with molecular weights of 23,000, 20,000, and 14,000 is identified as putative major wing cuticle proteins with the following biological properties: These proteins are abundant components of cuticle and are major synthetic products of cuticle-secreting hypodermal cells. They are leucine-rich and methionine-free and are the most prominent proteins of this type synthesized by wing hypoderm at 65 hr, during the period of procuticle formation. Electron microscopic autoradiography shows that leucine-rich, methionine-free proteins specifically localize to the apical cell surface and newly secreted cuticle of 65-hr wing cells. This strongly suggests the export of these proteins to the cuticle. Lastly, these proteins undergo a reduction in extractability just after eclosion, during the period of cuticle protein crosslinking (sclerotization). The synthesis of these major hypoderm proteins is temporally regulated in development. In wing cells, the ll-kDa proteins are synthesized first, from 53 to 78 hr, and the 20- and 23-kDa proteins are synthesized from 63 to 93 hr. The pattern of synthesis for these proteins is similar in abdominal cells but delayed by 6 to 10 hr. Two-dimensional gel electrophoresis shows that each of the 23-, 20-, and 14-kDa size classes contains at least two component polypeptides. Patterns of protein synthesis in cells of the imaginal hypodermis are regulated in a precise temporal sequence during the production of adult cuticle. Their study yields a useful system for the analysis of molecular events in gene control and cell differentiation. 0 1985 Academic Press, Inc.
INTRODUCTION
At puparium formation in Drosophila, the imaginal disks undergo extensive morphological and biochemical changes to form the adult hypodermis. These cells follow a strict developmental program of synthetic events depending on their predetermined states. The terminal event in hypodermal cell differentiation is the synthesis of the adult cuticle. In this paper, we identify the major protein components of the adult cuticle and characterize their synthesis in differentiating hypodermal cells. Insect cuticles, in general, are composed of chitin (a polymer of iV-acetyl-D-glucosamine), lipids, waxes, and an assortment of proteins (reviewed by Hackman, 1974; Andersen, 1979). Histological and histochemical studies have shown that cuticle has several layers, each with a different chemical composition (reviewed by Wigglesworth, 1948; Neville, 1975; Hepburn, 1976). Two of the major subdivisions, epicuticle and procuticle, are composed of “lipids and proteins” and “chitin and proteins,” respectively, but very little is known about the protein components of these layers. Some of the larval cuticle proteins of Drosophila melanogaster have ’ Current address: Room 16-717, nology, Cambridge, Mass. 02139. 0012-16O6/85 Copyright All rights
$3.00
Q 1985 by Academic Press, Inc. of reproduction in any form reserved.
Massachusetts
Institute
of Tech-
420
been purified and sequenced (Fristrom et aL, 19’78; for these proteins have Snyder et a& 1982). The genes been cloned, and their DNA sequences have been determined (Snyder, et aZ., 1981; Snyder et aL, 1982). In addition, cuticle proteins from various stages of the life cycle of Drosophila have been shown to possess characteristic differences, but their synthesis has so far not been studied (Chihara et aL, 1982). Mitchell and collaborators (Mitchell and Peterson, 1981; Mitchell et aZ., 1983) have described the morphology and some molecular events during differentiation and cuticle formation in wing hypodermal cells, but evidence for the identity and patterns of synthesis of adult cuticle proteins was not provided. The purpose of the present study has been to identify a subset of the adult cuticle proteins so that the major molecular events of hypodermal cell differentiation can be studied. We define the major wing cuticle proteins of D. melanogaster and describe their temporally regulated synthesis during hypoderm differentiation. We provide evidence that proteins with molecular weights of 23,000, 20,000, and 14,000 are major components of wing cuticle based on their abundance in cuticle-containing tissues, their lack of methionine, their autoradiographic localization to the apical cell surface and cuticle, and the decrease in their extractability from cuticle during the
ROTER, SPOFFORD, AND SWIFT
posteclosion crosslinking processes. In addition, we show that these proteins are synthesized in a strict developmentally regulated manner. The timing of this program of synthetic events is delayed in abdominal hypoderm relative to hypodermal cells of wing. We hope that the description of these developmentally regulated synthetic events may form the basis for further studies on regulated gene expression during cell differentiation in Drosophila. MATERIALS
AND METHODS
Culture and staging of pupae. Oregon R (Chicago strain f) D. melanogaster were grown in a population cage by standard methods as described by Elgin and Miller (1978). Embryos were collected on grape juice agar plates for 4 to 8 hr and inoculated into quart containers with cornmeal medium. After 5 days incubation at 25°C pupae were stage-selected by flotation (Mitchell and Mitchell, 1964). Pupae with a synchrony of &l hr were grown at 25 + 1°C until dissected. The time of puparium formation is taken as 5 hr before the onset of flotation. All pupal stages are expressed in terms of hours post-puparium formation. Synchrony of populations of older pupae (greater than 48 hr) was verified by eye and bristle pigmentation states (Nash, 1976). For time points after eclosion, synchronous late third instar larvae were grown at 25°C in cornmeal vials which were cleared at lo-min intervals as soon as the first eclosed flies were observed. The adults were aged in cornmeal vials at 25°C. Dissection of integument and preparation of cuticle. Puparia were attached to double-surfaced cellophane tape and slit lengthwise to remove the pharate adults. These were dissected in MOPS-buffered Ringer’s solution containing 80 mM NaCl, 10 mM KCl, 1 mM CaClz, 0.2 mM MgCl,, 10 mM MOPS buffer, pH 7.0 (Mitchell et al, 1977). The wing sacs were carefully extended away from the body and were severed at their bases with a fine scalpel. The abdomens were severed from the thoraxes, and the genitalia and the last abdominal segment were removed with a fine-point scalpel. The remaining tissue was slit along the ventral midline and folded open so that the digestive and reproductive organs could be carefully removed using No. 5 watchmakers’ forceps. Tissues from pupae older than 65 hr were removed from their pupal cuticle sheaths, but younger tissues were too fragile for removal and, therefore, were left attached for the described treatments. Wing and abdominal tissues were pulse-labeled in vitro and extracted as described below. Abdominal cuticle was prepared by scraping the integument clean of all visible muscle and adipose
Synthesis
of Cuticle
Proteins
421
tissue in preextraction solution using a curved tungsten needle under a dissecting microscope. The preparations were blotted dry and extracted as described below. The preextraction solution contained 0.5 MNaCl, 0.5% Nonidet P-40 (Particle Data Laboratories, Inc.), 1% 2mercaptoethanol, 50 mM Tris-HCl, pH 7.0, 1 mM phenylmethylsulfonyl fluoride (PMSF)’ and was saturated for phenylthiourea. Adults were etherized and the wings were removed with No. 5 watchmakers’ forceps. Wings were rinsed in water and removed to a clean microscope slide on which each wing was torn into at least three fragments to aid in extraction of the inner tissues. The fragments were transferred to SDS extraction solution and extracted as described below. Pulse-labeling of integument. Dissected tissues from two or three pupae were incubated in a 2-~1 drop of culture medium M3 (Shields and Sang, 1977) lacking yeastolate, serum, and leucine, but containing 10 &i of rH]leucine (130-170 CVmmole) or [35S]methionine (900-1000 Ci/mmole). Tissues were incubated for 45 min at 25°C in an organ culture dish with watersaturated atmosphere. Under these conditions, explants incorporated label linearly with time for at least 60 min (data not shown). The tissues experienced little physiological stress as evidenced by only minor amounts of synthesis of the 70-kDa heat shock protein without translational shutdown of normal protein synthesis. Following incubation, explants were extracted as described below. Extraction of proteins. For one-dimensional sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, cuticle or integument preparations were incubated in 50 ~1 of SDS extraction solution (2% SDS, electrophoresis grade, 5% 2-mercaptoethanol, 10% glycerol, 1 mM PMSF, 65 mM Tris-HCl, pH 6.8) for l-2 hr at 65°C. For two-dimensional nonequilibrium pH gradient electrophoresis (NEPHGE), tissues were homogenized in 100 ~1 of lysis solution (0.5% Nonidet P-40, 1% 2mercaptoethanol, 2 mM PMSF, 50 mM Tris-HCl, pH 7.0) in a miniature Dounce homogenizer. Ten microliters of a solution containing 1 mg/ml of DNase I (Sigma D4763), 0.5 mg/ml of RNase A (Sigma R5000), 5 mM CaC12, and 5 mM MgClz was added, and nucleic acids were digested for 5 min at 25°C. An additional 10 ~1 of DNase/RNase solution was added and the incubation was repeated. The lysate was precipitated with 10 vol of acetone at -20°C overnight and centrifuged. The * Abbreviations used: PMSF, phenylmethylsulfonyl fluoride; PTU, phenylthiourea; SDS, sodium dodecyl sulfate; NEPHGE, nonequilibrium pH gradient electrophoresis; TCA, trichloroaeetic acid; V. hr, volt-hours, EM, electron microscope.
422
DEVELOPMENTALBIOLOGY
pellet was dissolved in 25 ~1 of NEPHGE loading solution (2% LKB ampholines, pH 3.5-10, 6 M urea, 5% 2-mercaptoethanol, 4% Nonidet P-40). Proteins in samples prepared for NEPHGE were identical to those prepared for SDS-gels as shown by diluting NEPHGE samples with SDS extraction solution and analyzing them on SDS-gels (data not shown). Samples were analyzed for hot TCA-insoluble cpm by spotting 2-~1 aliquots onto l-cm squares of Whatman 3MM paper which were processed in bulk through 10% TCA at 0°C for 10 min, 5% TCA at 0°C for 10 min, 5% TCA at 100°C for 10 min, two washes of 95% ethanol, and an acetone wash. The paper squares were dried and counted in 4 ml of Omnifluor (New England Nuclear) in a Packard scintillation counter. The cpm loaded in each gel lane are specified in the figure legends. All protein samples were stored at -70°C for no more than a week prior to electrophoresis. Gel electrophwesis. One-dimensional SDS-polyacrylamide electrophoresis was by the method of Laemmli (1970), except that twice the concentrations of Tris and SDS were used in both the stacking and separating gels. The separating gels (165 mm long X 185 mm wide X 1.5 mm thick) were poured as slabs containing 15 to 20% acrylamide throughout or 7.5 to 25% acrylamide in a slightly concave exponential gradient (N,N’-methylenebisacrylamide = 2.7% of total acrylamide). Samples were heated to 65°C for 10 min just prior to loading. Electrophoresis was carried out at 30-50 mA constant current for 6-7 hr and was stopped when the bromophenol blue dye reached the gel bottom. Molecular size protein standards were obtained from Bio-Rad and BDH and included phosphorylase b (92.5 kDa), bovine serum albumin (66.2 kDa), ovalbumin (45 kDa), carbonic anhydrase (31 kDa), soybean trypsin inhibitor (21.5 kDa) myoglobin (16.9 kDa), and lysozyme (14.4 kDa). Two-dimensional nonequilibrium pH gradient electrophoresis was by the method of O’Farrell et al. (1977). Samples were heated to 37°C for 5 min just prior to loading them at the anodal ends of the firstdimension gels. The first dimension was run in tubes, 13 cm long, 2.2 mm inside diameter, for 2.5 hr at 400 V (1000 V - hr). Tube gels were equilibrated in SDS extraction solution and loaded onto the top of slab gels before the second dimension was run as described for SDS-gels. Radioactively labeled wing proteins were run in lanes at the side of each slab to help identify spots which comigrated with the major low molecular weight wing proteins. Gel staining and jluorography. Gels were pre-fixed in 45% methanol, 10% acetic acid for 30 min, rinsed overnight in 5% methanol, 7% acetic acid and silverstained by the method of Oakley et al. (1980). Gels
VOLUME 10’7, 1985
were photographed and completely destained in Kodak Rapid Fixer Solution A (concentrated sodium thiosulfate) to minimize quenching due to silver deposits (Van Keuren et aL, 1981). Gels were then embedded with 2,5-diphenyloxalone by the method of Bonner and Laskey (1974), dried, and placed against preflashed Kodak XAR-5 film (Laskey and Mills, 1975). Films were exposed for 1 to 7 days, depending on the cpm loaded, at -70°C. Electron microscopk autoradiography. Wing integument explants from 65-hr pupae were pulse-labeled in rH]leucine or [35S]methionine as described above and chased for 15 min in complete M3 medium (Shields and Sang, 1977) containing leucine and methionine. Tissues were fixed in half-strength Karnovsky’s fixative as described by Poodry (1980), postfixed in 1% 0~0~ in 0.1 M cacodylate buffer, pH 7.6, dehydrated in ethanol, embedded in Epon, and cut as gold sections (loo-120 nm). Autoradiography was performed by the method of Salpeter and Bachman (1972) using Ilford L4 emulsion. Specimens were exposed at 4°C and developed in Kodak Microdol-X developer. Sections were viewed and photographed in a Siemens 101 electron microscope. EM autoradiographs were analyzed by counting the number of silver grains within each 0.5-cm length of a l-cm-wide template aligned along an axis perpendicular to the cuticle using a Commodore 2001 computer and a summagraphics image analyzer. The total number of grains falling within each box gave a relative value for grain density at specific distances from the cuticle. The grain density values at each distance were summed and individually expressed as a percentage of the total number of grains counted.
RESULTS
Identification
of the Major Wing Cuticle Proteins
One purpose of the present work has been to define the biological properties of a group of proteins which we believe to be adult cuticle proteins. In pilot studies, whole pharate adults were removed from their pupal cases, homogenized in preextraction solution and rinsed. Then the cuticle proteins were extracted with SDS extraction solution. These methods were based on those described by Fristrom et al. (1978). The complexity of our results (three bands at greater than 100 kb, bands at approximately 73, 60, 53, 23.5, 23, 20, 14, 10, and 9 kb, as well as several minor bands) suggested that we select a subset of these components. We have chosen the low molecular weight, major wing cuticle proteins in order to focus our studies of this highly complex system on a few tractable components. Proteins from dissected tissues. In contrast to the
ROTER, SPOFFORD,AND SWIFT
uniformly pliable larval cuticles, the adult cuticle is more complex in its structure and regional variation. Wing blade tissue is unique in that it contains primarily homogeneous cuticle and hypodermal cells (Figs. lA, D, and E). The muscles which move the wings form attachments only at the wing hinge and within the thorax and, therefore, do not contaminate wing blade preparations. On the other hand, isolated abdominal integument contains large amounts of attached cellular material, especially muscles of the abdominal wall (Figs. lB, F, and G). It is possible to manually scrape the abdominal cuticle free of most of the attached tissues (Fig. lC), but even these preparations contain some cellular debris. As a compromise between biochemically purified mass cuticle (Fristrom et al, 1978) and pure cuticle from specific body regions, we have analyzed proteins from hand-dissected tissues, which inevitably contain some cellular contamination. Since hand-dissected tissues yield very small amounts
Synthesis
of Cuticle
Proteins
423
of SDS-soluble proteins, exact protein quantitation is difficult. Figure 2 shows various loads of proteins extracted from the tissues analyzed. Wing integument (e.g., lane 10) contains abundant proteins at 43 (actin), 25-35,23,20, and 14 kDa. Upon lesser loadings of these proteins (lanes 2 and 6), it is apparent that the most abundant is the 20-kDa band, followed in order of decreasing amounts by the bands at 23,14,43, and 2535 kDa. It is significant that the 23-, 20-, and 14-kDa bands are more prominent than that of actin, a generally abundant cellular protein. Abdominal integument and abdominal cuticle (Fig. 2, lanes 3, 4, 7, and 8) contain more complex mixtures of proteins than wing integument contains. This is expected from abdominal integument preparations since the proteins of several cell types are represented. Abdominal cuticle preparations, after removal of most of the contaminating tissues, have fewer proteins than abdominal integument, but they still contain at least
FIG. 1. Light microscopy of late pupal wing (A, D, and E), abdominal integument (B, F, and G), and abdominal cuticle (C) from 90- to 96-hr pupae. Epon-embedded tissues were sectioned at 1 nm and stained with toluidine blue for light microscopy. Panel A is a lowmagnification (460X) micrograph of a transverse section through a folded pupal wing, while panels D and E are high magnifications (1560x) of this tissue. The arrowhead denotes a medial member of the anterior triple row of bristles. Panel B is a low-magnification (460X) micrograph of a section through abdominal integument and underlying tissues, whereas panels F and G are high magnifications (1560X) of selected regions. Panel C is a low magnification (460X) micrograph of a section through a typical abdominal cuticle preparation after most adhering tissue was scraped away. Note the tissue complexity of the abdominal integument preparations relative to wing integument and abdominal cuticle. c = cuticle (ec = epicuticle, pc = procuticle). H = hypodermis. M = muscle. T = tracheole. Bar for panels A, B, and C = 30 pm. Bar for panels D, E, F, and G = 10 pm.
424
DEVELOPMENTALBIOLOGY
12345678910 M WAIACM
WAI
ACMW
-my0
c-
d-
-act
-3Okd -23kd
ef-
I:ss2bd - l6kd - I4kd
FIG. 2. Silver-stained acrylamide gradient (7.5 to 25%) SDS gel showing extractable proteins of wing (W), abdominal integument (AI), abdominal cuticle (AC) and indirect flight muscle (M) from 90to 96-hr pupae. Multiple lanes of single samples (e.g., muscle proteins in lanes 1, 5, and 9) represent relative protein amounts of 1:2:4 from left to right. The letters at the left indicate mohilities of molecular weight marker proteins (a = 92.5, h = 66.2, c = 45, d = 31, e = 21.5, and f = 14.4 kDa). The 14-, 20-, and 23-kDa ACPs and the 16-, 18-, and 30-kDa abdominal proteins as well as actin (act) and myosin heavy chain protein (myo) are indicated.
10 abundant proteins below 45 kDa. We can assume that actin is not a secreted cuticle protein but represents a protein of the contaminating cellular material in these preparations. We observe a dramatic enrichment for nine proteins relative to actin when contaminating tissues are scraped from the cuticle. Specifically, compare lane 3 (integument) with lane 8 (cuticle) in Fig. 2. The intensity of the actin band is reduced severalfold in proteins extracted from scraped cuticle as compared to crude integument. In contrast, the intensities of bands at 40, 35, 30, 28, 23, 20, 18, 16, and 14 kDa are increased between two- and tenfold. The degree of enrichment probably depends on how closely the protein is associated with the cuticle. Qualitatively, all nine of these components are more closely associated with abdominal cuticle than is actin. Of the proteins extracted from wing and abdominal preparations, those represented by the bands at 23,20, and 14 kDa are among the most abundant components common to both tissues. This suggests that they serve
VOLUME 107, 1985
a general role in cuticle structure, although the stoichiometric ratios of these proteins do vary between tissues (e.g., the 20-kDa protein is more abundant in wing than in abdominal cuticle). Preparations of indirect flight muscle, a tissue completely devoid of cuticle, lack any proteins at 23, 20, and 14 kDa, further supporting the identification of these components as cuticle specific. The abdominal bands at 40, 35, 30, 28, 18, and 16 kDa are possibly abdomen-specific cuticle proteins, but we have not characterized them further at this time. Cuticle proteins luck methionine. Since insect cuticle lacks sulfur (Neville, 19’75), we have compared the incorporation of leucine and methionine into proteins synthesized during wing cuticle formation. Figure 3 shows the patterns of incorporation of labeled leucine and methionine into proteins synthesized by 65-hr wing hypodermal cells. Three points are pertinent: (1) The 23-, 20-, and 14-kDa proteins are all synthesized by this tissue at 65 hr, and these three bands account for greater than 40% of the leucine incorporation. (2) The 23-, 20-, and 14-kDa proteins completely lack methionine as shown by their absence of labeling (we know that they are being actively synthesized at this stage). (3) The 23-, 20-, and 1QkDa proteins are the
M LM
C
t
C
FIG. 3. Comparison of leucine and methionine incorporation into cuticle proteins. Wing integument from 65-hr pupae was dissected and pulse labeled for 45 min with [‘Hlleucine or [%S]methionine such that wings from the same animal were being compared. Proteins were analyzed on a 17.5% acrylamide SDS gel. Gel lanes were loaded with either 10,000 *H cpm or 4000 to 5000 ?S cpm. The fluorograph is shown (3-day exposure). L indicates leucine-labeled and M indicates methionine-labeled proteins. The 14-, 20-, and 23-kDa cuticle protein hands (arrows) represent a substantial portion of the leucine but no detectable methionine incorporation into newly synthesized proteins.
ROTER, SPOFFORD, AND SWIFT
most prominent bands which are both leucine-rich and methionine-free and, therefore, account for the major difference obtained when these two labeled amino acids are compared. Cellular localization of the leucine-rich, methioninefree proteins. The striking difference in incorporation patterns of leucine and methionine into wing-synthesized proteins provides a basis for the cellular localization of these components. This is accomplished through EM autoradiography of tissues which have been pulselabeled with either [3H]leucine or r5S]methionine for the period during which the 23-, 20-, and 14-kDa proteins are the prominent leucine-rich, methioninefree synthetic products. These autoradiographs (Fig. 4) show much of the newly synthesized protein over the hypoderm cytoplasm when either labeled amino acid is used (also see Fig. 5). These represent the proteins which contain both leucine and methionine. In contrast, there is a marked concentration of label over the apical cell surface and cuticle in leucine-labeled cells (34% of the silver grains are in the region -0.35 to +0.‘7 pm) but not in those labeled with methionine (12% of the grains are in the region -0.35 to +0.‘7 pm). These data strongly suggest that one or more of the 23-, 20-, or 14-kDa proteins is transported to the cuticle. Because we were not anxious to incubate tissue explants for longer than 1 hr, it was not possible to chase more leucine counts into the cuticle. The trend of accumulation of leucine-rich, methionine-free proteins at the cell apex and cuticle, however, is clear. Loss of extractability of cuticle proteins after eclosion. To see if the extractability of the 23-, 20-, and 14-kDa proteins is affected during the presumed period of cuticle crosslinking, we have analyzed the extractable proteins from wing and abdominal tissues at several time points after eclosion. These data (Fig. 6) show that there is a significant decrease in the amounts of extractable 23-, 20-, and 14-kDa proteins within 60 min of eclosion. The amounts of actin extractable from these tissues remain constant during the posteclosion period. Actin can be considered representative of cellular contamination in these preparations. Therefore, a cellular protein is not affected by the cuticular crosslinking whereas the putative cuticle proteins are dramatically affected. Temporally Regulated Synthesis of Cuticle Proteins Synthesis and accumulation of cuticle proteins in wing integument. To study the synthetic events occurring during cuticle formation, we explanted wing tissue and pulse labeled it for 45 min in vitro. The patterns of proteins synthesized in explanted tissues correlate with proteins synthesized in vivo when labeled amino
Synthesis
of Cuticle
Proteins
425
acids were injected into live pupae but, for in vivo labeling, longer incubation periods were necessary (Roter, 1983). The advantage of the explant pulse-labeling technique is that large amounts of amino acid are incorporated in a short time, allowing for precise staging of animals. Figure 7A shows the pattern of protein accumulation in wing explants. The 23-, 20-, and 14-kDa proteins are detected as accumulated wing proteins only after 58 hr. This corresponds with the pattern of synthesis in that none of these proteins is synthesized at detectable levels before 53 hr (see below). Therefore, if cuticle synthesis begins at about 40 hr (Mitchell et aL, 1983), these components must relate to later events in cuticle synthesis. The synthesis of proteins in wing integument is regulated in a temporally specific manner (Fig. ‘7B). A group of proteins in the range 25 to 35 kDa is rapidly synthesized from as early as 48 to 58 hr with low-level synthesis continuing until 73 hr. The 23-, 20-, and 14kDa proteins, on which we will focus, are the major synthetic products following this early group. The 14kDa protein is synthesized first, beginning at 53 hr, reaching peak synthesis at 63 to 68 hr and decreasing to barely detectable levels by 78 hr. Synthesis of the ZO- and 23-kDa proteins is detectable by 63 hr, peaking synchronously near 73 hr and dropping to a low level that continues through 93 hr. Changes in the levels of actin synthesis have been reported (Mitchell et al, 1983); thus actin cannot be used as a standard for comparing rates of synthesis. Instead the above discussion is based on a comparison of the band intensities of single proteins at the different time points when equal amounts of total counts were loaded per gel lane. Comparison of temporal synthetic patterns in wing and abdomen. Since the 23-, 20-, and 14-kDa proteins are present in cuticle of both wing and abdomen, we have investigated their patterns of synthesis in both tissues. All three proteins are synthesized in 64- to 70hr wing integument and in 70-hr abdominal integument (Fig. 8). Although many proteins are synthesized in the abdominal integument preparations, the 23-, 20-, and 14-kDa bands represent a significant fraction of the leucine incorporation. In agreement with the result that abdominal cuticle contains less of the 20-kDa protein than does wing cuticle (see Fig. 2), the rate of synthesis of the 20-kDa band is lower in abdominal than in wing integument. Other differences include the apparent asynchrony of development between wing and abdomen. Synthesis of the 14-kDa protein in wing begins before 57 hr but is not detected in abdomen at this time (Fig. 8). Abdominal synthesis of the 14-kDa protein begins by 64 hr, a time at which wing is synthesizing all three
426
DEVELOPMENTAL BIOLOGY
VOLUME 107, 1985
FIG. 4. Cuticle morphogenesis and cellular localization of newly synthesized proteins by EM autoradiography. Panels A-C are electron micrographs (19,000X magnification) of abdominal tissues showing the cellular and cuticular appearances at 53, 65, and 80 hr, respectively. Panel D is a typical EM autoradiograph (12,100X) of a 65-hr wing integument cell pulse labeled for 45 min in [3H]leueine and chased with unlabeled amino acids for 15 min. Panel E is similar to D except that the cell has been labeled with [%]methionine. Note the abundance of apical silver grains in leucine-labeled but not in methionine-labeled hypodermis. Bar for panels A, B, and C = 0.5 pm. Bar for panels D and E = 1 @cm.G = Golgi apparatus. EC = epicuticle. PC = procuticle.
components. In addition, although the 20- and 23-kDa proteins are clearly synthesized in 64-hr wing cells, they are only barely detectable in abdominal cells.
These data suggest a 6- to lo-hr difference in timing of developmental events between hypodermal cells of wing and abdomen.
ROTER, SPOFFORD, AND SWIFT
2
‘-
”
0
427
of Cuticle Proteins
other is not clear. A detailed investigation of the posttranslational processing events involved in cuticle protein synthesis is currently underway. Figures 9C, D, and E show the two-dimensional analysis of wing synthesized proteins at 58, 65, and 78 hr. The temporal pattern of synthetic events generally follows that described in Fig. 7B. The synthesis of components within each size class is synchronous, with two exceptions: (1) The 23b spot is synthesized early at 58 hr, while synthesis of the other 23-kDa proteins does not begin until 63 to 65 hr. (2) Synthesis of the 14a protein persists longer (until 78 hr) than synthesis of the other spots near 14 kDa and for a longer period than is detected on one-dimensional gels (see Figs. 7 and 8), probably due to the higher sensitivity obtained on two-dimensional gels.
METHIONINE
-4.2-3.5-2.8-2.1
Synthesis
-1.4-0.7 DISTANCE
0 FROM
0.7
1.4 2.1
CUTICLE
2.8
3.5 4.2
SURFACE
4.9
DISCUSSION
5.6 6.3
(pm1
FIG. 5. Analysis of autoradiography. EM autoradiographs similar to those shown in Fig. 4 were quantitated as described under Materials and Methods. The distance from the outer surface of the cuticle is expressed on the ordinate with negative numbers representing the outside and positive numbers representing the cellular side of the cuticle. The shaded bars represent the grain density directly over the cuticle. For leucine-labeled tissue, N = 249; for methionine-labeled tissue, N = 369, where N = the total number of silver grains analyzed.
Two-dimensional gel analysis of synthesized proteins. Although the 23-, 20-, and 14-kDa bands are considered above as single proteins, analysis of 20% acrylamideSDS gels reveals two bands at 23 kDa (Figs. 6 and 7) and a group of components near 14 kDa (Fig. 7B). Repeated attempts to analyze 65-hr wing-synthesized proteins on standard two-dimensional gels (first dimension: pH 3.5 to 10 isoelectric focusing; second dimension: SDS-slab gel) result in the resolution of many protein spots, but no major synthetic products in the 23-, 20-, or 14-kDa size classes (data not shown). In contrast, the use of a nonequilibrium pH gradient electrophoresis (NEPHGE) as a first dimension, which allows resolution of proteins with high isoelectric points (O’Farrell et ab, 1977), results in the resolution of major synthetic products in the size classes of interest (Fig. 9). Wing (64 hr, Fig. 9A) and abdominal (70 hr, Fig. 9B) tissues synthesizing all three of the 23-, 20-, and 14kDa proteins were analyzed on NEPHGE two-dimensional gels. The 23- and 20-kDa bands resolve into 4 and 2 proteins, respectively, in both tissues. Wing tissue synthesizes many components at 14 kDa but abdominal cells appear to make only some of these proteins. Exactly how these components relate to each
The Major Wing Cuticle Proteins of D. melanogaster The classical isolation and identification of cuticle proteins has usually depended on differences in solu-
PUPA
10
30
60
ADULT
WAWAWAWAWA
-act
-23kd -2Okd :;gj -14kd
FIG. 6. Decrease in extractability of cuticle proteins after eclosion. Wing integument and abdominal cuticle were dissected from a late pupa (90-96 hr), flies at 10, 30, and 60 min after eclosion and older adults (12 hr after eclosion). Gel lanes for each time point contained proteins from the tissues of single animals except for the 12-hr point which required tissues of two animals. The silver stained 20% acrylamide SDS gel is shown. The 23-, 20-, and ll-kDa cuticle proteins are indicated as well as the 16- and 18-kDa abdomenspecific proteins and actin (act). The flies at 30 min and older had fully expanded wings prior to their dissection. Note the decreased amounts of cuticle proteins relative to cellular proteins (e.g., actin) as the post-eclosion crosslinking process proceeded.
428
DEVELOPMENTAL BIOLOGY 48 53 58 63 68 73 70 83 a8 93 .---.---m*
VOLUME 107. 1985 48 53 58 63 68 73 78 83 88 93
- act
- 16.9kd
- 14.4kd
B FIG. 7. Synthesis and accumulation of cuticle proteins in wing integument. Wing integument was dissected from pupae at various stages of cuticle formation (indicated above each lane in hr post-pupariation) and pulse labeled for 45 min with [‘Hlleucine. Panel A is the silver stained 20% acrylamide SDS gel, showing total wing protein, and panel B is a fluorograph of the same gel, showing newly synthesized protein. 25,000 cpm were loaded per lane (d-day fluorographic exposure). The mobilities of molecular weight marker proteins are shown (16.9 kDa = myoglohin, 14.4 kDa = lysozyme). The 14-, 20-, and 23-kDa cuticle protein bands as well as actin (act) are indicated. Note the rapid synthesis of cuticle proteins from 58 to 33 hr, and the continued low-level synthesis of the 20- and 23-kDa proteins through 93 hr.
bility properties in cuticle preparations. Extraction of cuticle with boiling water yields two fractions, the soluble “arthropodin” proteins and those remaining insoluble in the cuticle, the “sclerotin” (Richards, 1951). Later investigators have used sequential extractions with water, NaCl or KC1 solutions, aqueous urea, dilute alkali, and boiling HCI to remove different fractions of proteins, peptides, or amino acids with increasingly stronger denaturing conditions (reviewed by Hackman, 1974; Andersen, 1979). Sequential extractions remove groups of proteins which are presumably held in the cuticle by bonds of varying strength, from ionic and H bonds to covalent linkages. Fristrom et al. (1978) used variable solubilities to purify a small group of saltsolution-insoluble, aqueous-urea-soluble, proteins from This success third instar cuticles of D. wwlanoga&w. with larval cuticle led to the use of these methods to
define cuticle proteins from other stages of the L?r+ sophila life cycle, including the adult (Chihara et cd, 1982), but the large number of resulting proteins has led us, in the present work, to define a subset of the adult cuticle proteins of D. melanogaster and to present evidence, using several biologically related techniques, that these are indeed major components of the cuticle. We present evidence that the 23-, 20-, and 14-kDa proteins are the major cuticle proteins in adult wing tissue. These proteins are abundant in cuticle-containing tissue. Like other cuticle proteins (Neville, 19’75), they lack methionine. Their extractability decreases during the period of cuticle hardening, a property expected of proteins being crosslinked covalently into the cuticular matrix. Radioautographic localization of silver grains to the cuticle in leucine but not methionine labeled tissues is primarily due to these proteins. There
ROTER,
57
64
---WAWAWAWA
70
SPOFFORD,
AND
SWIFT
78
FIG. 8. Differential temporally regulated synthesis of cuticle proteins in abdominal (A) versus wing (W) integument. Wing and abdominal integument explants for each time point were dissected from three staged pupae such that the different tissues came from the same animals. The tissues were pulse labeled for 45 min with [3H]leucine and analyzed on a 15% acrylamide-SDS gel fluorograph. Lanes were loaded with approximately 10,000 cpm of wing extract or 20,000 cpm of abdominal extract. As indicated at the top of the lanes, wing and abdominal tissues were analyzed at 57, 64, 70, and 78 hr postpupariation. The 14-, 20-, and 23-kDa major cuticle proteins (as marked) show different stage specificities of synthesis in wing versus abdomen.
are also indications (Roter, 1983), to be published elsewhere, that these proteins undergo the processing expected of secreted proteins, reinforcing their eventual extracellular localization. Temporally Regulated Synthesis
Early light microscopic (reviewed by Wigglesworth, 1948; Richards, 1951) and more recent electron microscopic studies (reviewed by Locke, 1976) demonstrate the sequential deposition of cuticle layers. The sequential pattern of synthesis of the cuticle proteins seen in the present work (i.e., synthesis of the 14-kDa protein begins and peaks before the 20- and 23-kDa proteins) is doubtless another indication that these components are synthesized as they are needed for construction of different cuticle layers. The precise location of these proteins in the cuticle is still unclear, since the timing of their synthesis does not strictly correlate with cuticle thickening. According to Mitchell et al. (1983),
Synthesis
of Cuticle Proteins
429
synthesis of D. melanogaster wing cuticle begins with secretion of the outermost cuticulin layer at about 40 hr past puparium formation and wing cuticle reaches maximum thickness by about 60 hr. Our fluorographs demonstrate a group of proteins in the size range 25 to 35 kDa that are rapidly synthesized during the later part of this period (see Fig. 7B). They are probably also cuticle proteins, but we have not focused on them in these initial studies because (1) they were not detected as the most abundant proteins in late pupal cuticle preparations, and (2) they appear to become unextractable shortly after synthesis, a property which makes them difficult to study. The synthesis of the 23-, 20-, and 14-kDa major wing cuticle proteins occurs primarily after the cuticle has reached its presumed full thickness. In contrast, abdominal cuticle thickens dramatically during the 65- to 80-hr period (see Figs. 4B and 4C). In addition, the striking chitinous lamellar pattern seen in the abdominal procuticle layer (Fig. 4C) and seen generally in insect procuticle (Neville, 1975) appears to be either highly condensed or completely absent in wing cuticle (Mitchell et al, 1983; our unpublished observations). Since regional variation of cuticle structure lies principally in the procuticle layer, it is not surprising that wing procuticle differs greatly from abdominal procuticle. With these differences in mind, we believe there are several possible explanations for the synthesis of what we contend to be major wing cuticle proteins after the presumed completion of cuticle thickening: (1) These proteins may be synthesized late yet enter a preformed chitin matrix to “fill-in” and strengthen the procuticle. (2) These proteins may construct an inner cuticular layer which bonds the cell and the procuticle. (3) These proteins may be components of thin procuticle layers which are not obvious on electron micrographs. Because of the folded and convoluted nature of preeclosion wing cuticle, the visualizations and precise thickness measurements of cuticle layers are difficult. The exact localization of the major wing cuticle proteins will have to await specific cytochemical analysis, for example, by using antibodies to individual cuticle proteins. The difference in patterns of cuticle protein synthesis between wing and abdominal hypodermis reported here is indicative of a temporal offset in the states of differentiation of these cells. The value of about 6 to 10 hr difference between the developmental stages of wing and abdomen agrees with the value of 9 to 10 hr determined by Mitchell and Petersen (1981). As demonstrated using bithorax mutants, the timing of differentiation in hypoderm cells depends on the determined fate of the cells, not their position in the fly (Mitchell and Petersen, 1983). The characterization of molecular events in the regulation of specific synthetic products,
430
DEVELOPMENTAL BIOLOGY OH-
VOLUME 107, 1985
NE PHGE
H+
A
FIG. 9. Two-dimensional gel analyses of stage-specific protein synthesis in wing (A, C, D, and E) and abdominal (B) integument. Wing and abdominal integument explants were pulse-labeled with [3H]leucine, and the newly synthesized proteins were analyzed by twodimensional electrophoresis and fluorography. Panel A shows proteins synthesized by 65-hr wing and panel B shows those made by 70-hr abdominal tissues. Spots corresponding in molecular weight to the major cuticle proteins are enclosed in circles and their size classes are indicated. The actin spots are enclosed in squares for reference. These two stages are chosen because they represented periods when all of the major cuticle proteins are synthesized. Panels C through E show the low molecular weight region at the basic end of gels containing newly synthesized proteins from 58, 65-, and ‘7%hr wing integument explants, respectively. Each member of each size class is designated by a letter, starting with the most acidic protein and lettering each spot respectively. Note that most of the proteins within each size class of cuticle proteins are synthesized coordinately, with the exception of spot 23b which is synthesized before the other 23-kDa proteins. Also, spot 14a shows incorporation at ‘78 hr when other ll-kDa spots (14b-d) no longer label.
such as the major wing cuticle proteins, may provide a significant step toward an understanding of the processes responsible for this temporal control of cell differentiation. We thank Ms. Sagami Paul for her excellent assistance with histological preparations and thin sectioning. This work was supported in part by NIH Grant CA-14599. A.H.R. was supported by NIH Training Grant HD-07136 to The Committee on Developmental Biology. REFERENCES ANDERSEN, S. 0. (1979). Biochemistry of insect cuticle. Annu. Rev. Entomol. 24, 29-61.
BONNER, W. M., and LASKEY, R. A. (1974). A film detection method for tritium-labeled proteins and nucleic acids in polyacrylamide gels. Eur.
J. Biochem.
46, 83-88.
CHIHARA, C. J., SILVERT, D. J., and FRISTROM, J. W. (1982). The cuticle proteins of Drosophila melanogaster: Stage specificity. Dev. Biol
89, 379-388.
ELGIN, S. C. R., and MILLER, D. W. (1978). Mass rearing of flies and mass production and harvesting of embryos. In “The Genetics and (M. Ashburner and T. R. F. Wright, eds.), Biology of Drosophila” Vol. 2b, pp. 112-120. Academic Press, New York. FRISTROM, J. W., HILL, R. J., and WATT, F. (1978). The procuticle of Drosophila: Heterogeneity of urea-soluble proteins. Biochemistry 17, 3917-3924. HACKMAN, R. H. (1974). Chemistry of insect cuticle. In “The Physiology
ROTER,
SPOFFORD,
AND
SWIFT
of Insecta” (M. Rockstein, ed.), 2nd ed., Vol. VI, pp. 237-258. Academic Press, New York. HEPBURN, H. R. (1976). “The Insect Integument.” Elsevier, Amsterdam. HILLMAN, R., and LESNIK, L. H. (1970). Cuticle formation in the embryo of Drosophila melanogaster. J. Mwphd 131,383-396. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature &n&m) 227, 680-685. LASKEY, R. A., and MILLS, A. D. (1975). Quantitative film detection of $H and ‘“C in polyacrylamide gels by fluorography. Eur. J. B&hem 56,335-341. LOCKE, M. (1976). The role of plasma membrane plaques and golgi complex vessicles in cuticle deposition during the molt/intermoult cycle. In “The Insect Integument” (H. R. Hepburn, ed.), pp. 237258. Elsevier, Amsterdam. MITCHELL, H. K., LIPPS, L. S., and TRACY, U. M. (1977). Transcriptional changes in pupal hypoderm in Drosophila melanogaster. B&hem. Genet. 15, 575-587. MITCHELL, H. K., and MITCHELL, A. (1964). Mass culture and age selection in Drosophila Drosophila Ir&rm. Serv. 39, 135-137. MITCHELL, H. K., and PETERSEN, N. S. (1981). Rapid changes in gene expression in differentiating tissues of Drosophila. Dev. Biol 85, 233-242. MITCHELL, H. K., and PETERSEN, N. S. (1983). Gradients of differentiation in wild-type and bithorax mutants of Drosophila Dev. BioL 95, 459-467. MITCHELL, H. K., ROACH, J., and PETERSEN, N. S. (1983). The morphogenesis of cell hairs on Drosophila wings. Deu. Biol 95, 387-398. NASH, W. G. (1976). Patterns of pigmentation color states regulated by the y locus in Drowphila melanogaster. Dew. Biol 48, 336-343. NEVILLE, A. C. (1975). “Biology of the Arthropod Cuticle.” SpringerVerlag, New York.
Synthesis
of Cuticle
Proteins
431
OAKLEY, B. R., KIRSCH, D. R., and MORRIS, N. R. (1980). A simplified silver stain for detecting proteins in polyacrylamide gels. Anal. B&hem. 105,361-363. O’FARRELL, P. Z., GOODMAN, H. M., and O’FARRELL, P. H. (1977). High resolution two-dimensional electrophoresis of basic as well as acidic proteins. Cell 12, 1133-1142. POODRY, C. A. (1980). Epidermis: Morphology and development. In. “The Genetics and Biology of Drosophila” (M. Ashburner and T. R. F. Wright, eds.), Vol. 2d, pp. 443-497. Academic Press, New York. RICHARDS, A. G. (1951). “The Integument of Arthropods.” Minnesota Univ. Press, Minneapolis. ROTER, A. H. (1983). “The Adult Cuticle Proteins of Drosophila melanogaster: A Model System for Studying Developmentally Regulated Gene Expression.” Ph. D. thesis, University of Chicago. SALPETER, M. M., and BACHMANN, L. (1972). Autoradiography. In “Principles and Techniques of Electron Microscopy” (M. A. Hayat, ed.), Vol. 2, pp. 221-278. Van Nostrand-Reinhold, New York. SHIELDS, G., and SANG, J. H. (1977). Improved medium for culture of Drosophila Embryonic Cells. Drosophila I@rm Serv. 52, 161. SYNDER, M., HIRSH, J., and DAVIDSON, N. (1981). The cuticle genes of Drosophila: A developmentally regulated gene cluster. Cell 25,165177. SNYDER, M., HUNKAPILLER, M., YUEN, D., SILVERT, D., FRISTROM, J., and DAVIDSON, N. (1982). Cuticle protein genes of Drosophila: Structure, organization and evolution of four clustered genes. Cell 29, 1027-1040. VAN KEUREN, M. L., GOLDMAN, D., and MERRJL, C. R. (1981). Detection of radioactively labeled proteins is quenched by staining methods: Quenching is minimal for “C and partially reversible for ‘H with a photochemical stain. Anal. B&hem 116, 248-255. WIGGLESWORTH, V. B. (1948). The insect cuticle. Biol. Rev. 23, 408451.
BIOLOGY
107,420-431
(19%)
Synthesis of the Major Adult Cuticle Proteins of Drosophila melanogasfer during Hypoderm Differentiation ALAN H. ROTER,’ JANICE B. SPOFFORD, AND HEWSON The Committee
wn Developmental
Biology
Received
December
and Department 17, 1983;
of Biology,
accepted
in revised
University form
August
SWIFT
of Chicago,
Chicago,
Illinois
61.1637
27, 1984
Differentiating imaginal hypodermal cells of Drosophila melanogaster form adult cuticle during the second half of the pupal stage (about 40 to 93 hr postpupariation). A group of proteins with molecular weights of 23,000, 20,000, and 14,000 is identified as putative major wing cuticle proteins with the following biological properties: These proteins are abundant components of cuticle and are major synthetic products of cuticle-secreting hypodermal cells. They are leucine-rich and methionine-free and are the most prominent proteins of this type synthesized by wing hypoderm at 65 hr, during the period of procuticle formation. Electron microscopic autoradiography shows that leucine-rich, methionine-free proteins specifically localize to the apical cell surface and newly secreted cuticle of 65-hr wing cells. This strongly suggests the export of these proteins to the cuticle. Lastly, these proteins undergo a reduction in extractability just after eclosion, during the period of cuticle protein crosslinking (sclerotization). The synthesis of these major hypoderm proteins is temporally regulated in development. In wing cells, the ll-kDa proteins are synthesized first, from 53 to 78 hr, and the 20- and 23-kDa proteins are synthesized from 63 to 93 hr. The pattern of synthesis for these proteins is similar in abdominal cells but delayed by 6 to 10 hr. Two-dimensional gel electrophoresis shows that each of the 23-, 20-, and 14-kDa size classes contains at least two component polypeptides. Patterns of protein synthesis in cells of the imaginal hypodermis are regulated in a precise temporal sequence during the production of adult cuticle. Their study yields a useful system for the analysis of molecular events in gene control and cell differentiation. 0 1985 Academic Press, Inc.
INTRODUCTION
At puparium formation in Drosophila, the imaginal disks undergo extensive morphological and biochemical changes to form the adult hypodermis. These cells follow a strict developmental program of synthetic events depending on their predetermined states. The terminal event in hypodermal cell differentiation is the synthesis of the adult cuticle. In this paper, we identify the major protein components of the adult cuticle and characterize their synthesis in differentiating hypodermal cells. Insect cuticles, in general, are composed of chitin (a polymer of iV-acetyl-D-glucosamine), lipids, waxes, and an assortment of proteins (reviewed by Hackman, 1974; Andersen, 1979). Histological and histochemical studies have shown that cuticle has several layers, each with a different chemical composition (reviewed by Wigglesworth, 1948; Neville, 1975; Hepburn, 1976). Two of the major subdivisions, epicuticle and procuticle, are composed of “lipids and proteins” and “chitin and proteins,” respectively, but very little is known about the protein components of these layers. Some of the larval cuticle proteins of Drosophila melanogaster have ’ Current address: Room 16-717, nology, Cambridge, Mass. 02139. 0012-16O6/85 Copyright All rights
$3.00
Q 1985 by Academic Press, Inc. of reproduction in any form reserved.
Massachusetts
Institute
of Tech-
420
been purified and sequenced (Fristrom et aL, 19’78; for these proteins have Snyder et a& 1982). The genes been cloned, and their DNA sequences have been determined (Snyder, et aZ., 1981; Snyder et aL, 1982). In addition, cuticle proteins from various stages of the life cycle of Drosophila have been shown to possess characteristic differences, but their synthesis has so far not been studied (Chihara et aL, 1982). Mitchell and collaborators (Mitchell and Peterson, 1981; Mitchell et aZ., 1983) have described the morphology and some molecular events during differentiation and cuticle formation in wing hypodermal cells, but evidence for the identity and patterns of synthesis of adult cuticle proteins was not provided. The purpose of the present study has been to identify a subset of the adult cuticle proteins so that the major molecular events of hypodermal cell differentiation can be studied. We define the major wing cuticle proteins of D. melanogaster and describe their temporally regulated synthesis during hypoderm differentiation. We provide evidence that proteins with molecular weights of 23,000, 20,000, and 14,000 are major components of wing cuticle based on their abundance in cuticle-containing tissues, their lack of methionine, their autoradiographic localization to the apical cell surface and cuticle, and the decrease in their extractability from cuticle during the
ROTER, SPOFFORD, AND SWIFT
posteclosion crosslinking processes. In addition, we show that these proteins are synthesized in a strict developmentally regulated manner. The timing of this program of synthetic events is delayed in abdominal hypoderm relative to hypodermal cells of wing. We hope that the description of these developmentally regulated synthetic events may form the basis for further studies on regulated gene expression during cell differentiation in Drosophila. MATERIALS
AND METHODS
Culture and staging of pupae. Oregon R (Chicago strain f) D. melanogaster were grown in a population cage by standard methods as described by Elgin and Miller (1978). Embryos were collected on grape juice agar plates for 4 to 8 hr and inoculated into quart containers with cornmeal medium. After 5 days incubation at 25°C pupae were stage-selected by flotation (Mitchell and Mitchell, 1964). Pupae with a synchrony of &l hr were grown at 25 + 1°C until dissected. The time of puparium formation is taken as 5 hr before the onset of flotation. All pupal stages are expressed in terms of hours post-puparium formation. Synchrony of populations of older pupae (greater than 48 hr) was verified by eye and bristle pigmentation states (Nash, 1976). For time points after eclosion, synchronous late third instar larvae were grown at 25°C in cornmeal vials which were cleared at lo-min intervals as soon as the first eclosed flies were observed. The adults were aged in cornmeal vials at 25°C. Dissection of integument and preparation of cuticle. Puparia were attached to double-surfaced cellophane tape and slit lengthwise to remove the pharate adults. These were dissected in MOPS-buffered Ringer’s solution containing 80 mM NaCl, 10 mM KCl, 1 mM CaClz, 0.2 mM MgCl,, 10 mM MOPS buffer, pH 7.0 (Mitchell et al, 1977). The wing sacs were carefully extended away from the body and were severed at their bases with a fine scalpel. The abdomens were severed from the thoraxes, and the genitalia and the last abdominal segment were removed with a fine-point scalpel. The remaining tissue was slit along the ventral midline and folded open so that the digestive and reproductive organs could be carefully removed using No. 5 watchmakers’ forceps. Tissues from pupae older than 65 hr were removed from their pupal cuticle sheaths, but younger tissues were too fragile for removal and, therefore, were left attached for the described treatments. Wing and abdominal tissues were pulse-labeled in vitro and extracted as described below. Abdominal cuticle was prepared by scraping the integument clean of all visible muscle and adipose
Synthesis
of Cuticle
Proteins
421
tissue in preextraction solution using a curved tungsten needle under a dissecting microscope. The preparations were blotted dry and extracted as described below. The preextraction solution contained 0.5 MNaCl, 0.5% Nonidet P-40 (Particle Data Laboratories, Inc.), 1% 2mercaptoethanol, 50 mM Tris-HCl, pH 7.0, 1 mM phenylmethylsulfonyl fluoride (PMSF)’ and was saturated for phenylthiourea. Adults were etherized and the wings were removed with No. 5 watchmakers’ forceps. Wings were rinsed in water and removed to a clean microscope slide on which each wing was torn into at least three fragments to aid in extraction of the inner tissues. The fragments were transferred to SDS extraction solution and extracted as described below. Pulse-labeling of integument. Dissected tissues from two or three pupae were incubated in a 2-~1 drop of culture medium M3 (Shields and Sang, 1977) lacking yeastolate, serum, and leucine, but containing 10 &i of rH]leucine (130-170 CVmmole) or [35S]methionine (900-1000 Ci/mmole). Tissues were incubated for 45 min at 25°C in an organ culture dish with watersaturated atmosphere. Under these conditions, explants incorporated label linearly with time for at least 60 min (data not shown). The tissues experienced little physiological stress as evidenced by only minor amounts of synthesis of the 70-kDa heat shock protein without translational shutdown of normal protein synthesis. Following incubation, explants were extracted as described below. Extraction of proteins. For one-dimensional sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, cuticle or integument preparations were incubated in 50 ~1 of SDS extraction solution (2% SDS, electrophoresis grade, 5% 2-mercaptoethanol, 10% glycerol, 1 mM PMSF, 65 mM Tris-HCl, pH 6.8) for l-2 hr at 65°C. For two-dimensional nonequilibrium pH gradient electrophoresis (NEPHGE), tissues were homogenized in 100 ~1 of lysis solution (0.5% Nonidet P-40, 1% 2mercaptoethanol, 2 mM PMSF, 50 mM Tris-HCl, pH 7.0) in a miniature Dounce homogenizer. Ten microliters of a solution containing 1 mg/ml of DNase I (Sigma D4763), 0.5 mg/ml of RNase A (Sigma R5000), 5 mM CaC12, and 5 mM MgClz was added, and nucleic acids were digested for 5 min at 25°C. An additional 10 ~1 of DNase/RNase solution was added and the incubation was repeated. The lysate was precipitated with 10 vol of acetone at -20°C overnight and centrifuged. The * Abbreviations used: PMSF, phenylmethylsulfonyl fluoride; PTU, phenylthiourea; SDS, sodium dodecyl sulfate; NEPHGE, nonequilibrium pH gradient electrophoresis; TCA, trichloroaeetic acid; V. hr, volt-hours, EM, electron microscope.
422
DEVELOPMENTALBIOLOGY
pellet was dissolved in 25 ~1 of NEPHGE loading solution (2% LKB ampholines, pH 3.5-10, 6 M urea, 5% 2-mercaptoethanol, 4% Nonidet P-40). Proteins in samples prepared for NEPHGE were identical to those prepared for SDS-gels as shown by diluting NEPHGE samples with SDS extraction solution and analyzing them on SDS-gels (data not shown). Samples were analyzed for hot TCA-insoluble cpm by spotting 2-~1 aliquots onto l-cm squares of Whatman 3MM paper which were processed in bulk through 10% TCA at 0°C for 10 min, 5% TCA at 0°C for 10 min, 5% TCA at 100°C for 10 min, two washes of 95% ethanol, and an acetone wash. The paper squares were dried and counted in 4 ml of Omnifluor (New England Nuclear) in a Packard scintillation counter. The cpm loaded in each gel lane are specified in the figure legends. All protein samples were stored at -70°C for no more than a week prior to electrophoresis. Gel electrophwesis. One-dimensional SDS-polyacrylamide electrophoresis was by the method of Laemmli (1970), except that twice the concentrations of Tris and SDS were used in both the stacking and separating gels. The separating gels (165 mm long X 185 mm wide X 1.5 mm thick) were poured as slabs containing 15 to 20% acrylamide throughout or 7.5 to 25% acrylamide in a slightly concave exponential gradient (N,N’-methylenebisacrylamide = 2.7% of total acrylamide). Samples were heated to 65°C for 10 min just prior to loading. Electrophoresis was carried out at 30-50 mA constant current for 6-7 hr and was stopped when the bromophenol blue dye reached the gel bottom. Molecular size protein standards were obtained from Bio-Rad and BDH and included phosphorylase b (92.5 kDa), bovine serum albumin (66.2 kDa), ovalbumin (45 kDa), carbonic anhydrase (31 kDa), soybean trypsin inhibitor (21.5 kDa) myoglobin (16.9 kDa), and lysozyme (14.4 kDa). Two-dimensional nonequilibrium pH gradient electrophoresis was by the method of O’Farrell et al. (1977). Samples were heated to 37°C for 5 min just prior to loading them at the anodal ends of the firstdimension gels. The first dimension was run in tubes, 13 cm long, 2.2 mm inside diameter, for 2.5 hr at 400 V (1000 V - hr). Tube gels were equilibrated in SDS extraction solution and loaded onto the top of slab gels before the second dimension was run as described for SDS-gels. Radioactively labeled wing proteins were run in lanes at the side of each slab to help identify spots which comigrated with the major low molecular weight wing proteins. Gel staining and jluorography. Gels were pre-fixed in 45% methanol, 10% acetic acid for 30 min, rinsed overnight in 5% methanol, 7% acetic acid and silverstained by the method of Oakley et al. (1980). Gels
VOLUME 10’7, 1985
were photographed and completely destained in Kodak Rapid Fixer Solution A (concentrated sodium thiosulfate) to minimize quenching due to silver deposits (Van Keuren et aL, 1981). Gels were then embedded with 2,5-diphenyloxalone by the method of Bonner and Laskey (1974), dried, and placed against preflashed Kodak XAR-5 film (Laskey and Mills, 1975). Films were exposed for 1 to 7 days, depending on the cpm loaded, at -70°C. Electron microscopk autoradiography. Wing integument explants from 65-hr pupae were pulse-labeled in rH]leucine or [35S]methionine as described above and chased for 15 min in complete M3 medium (Shields and Sang, 1977) containing leucine and methionine. Tissues were fixed in half-strength Karnovsky’s fixative as described by Poodry (1980), postfixed in 1% 0~0~ in 0.1 M cacodylate buffer, pH 7.6, dehydrated in ethanol, embedded in Epon, and cut as gold sections (loo-120 nm). Autoradiography was performed by the method of Salpeter and Bachman (1972) using Ilford L4 emulsion. Specimens were exposed at 4°C and developed in Kodak Microdol-X developer. Sections were viewed and photographed in a Siemens 101 electron microscope. EM autoradiographs were analyzed by counting the number of silver grains within each 0.5-cm length of a l-cm-wide template aligned along an axis perpendicular to the cuticle using a Commodore 2001 computer and a summagraphics image analyzer. The total number of grains falling within each box gave a relative value for grain density at specific distances from the cuticle. The grain density values at each distance were summed and individually expressed as a percentage of the total number of grains counted.
RESULTS
Identification
of the Major Wing Cuticle Proteins
One purpose of the present work has been to define the biological properties of a group of proteins which we believe to be adult cuticle proteins. In pilot studies, whole pharate adults were removed from their pupal cases, homogenized in preextraction solution and rinsed. Then the cuticle proteins were extracted with SDS extraction solution. These methods were based on those described by Fristrom et al. (1978). The complexity of our results (three bands at greater than 100 kb, bands at approximately 73, 60, 53, 23.5, 23, 20, 14, 10, and 9 kb, as well as several minor bands) suggested that we select a subset of these components. We have chosen the low molecular weight, major wing cuticle proteins in order to focus our studies of this highly complex system on a few tractable components. Proteins from dissected tissues. In contrast to the
ROTER, SPOFFORD,AND SWIFT
uniformly pliable larval cuticles, the adult cuticle is more complex in its structure and regional variation. Wing blade tissue is unique in that it contains primarily homogeneous cuticle and hypodermal cells (Figs. lA, D, and E). The muscles which move the wings form attachments only at the wing hinge and within the thorax and, therefore, do not contaminate wing blade preparations. On the other hand, isolated abdominal integument contains large amounts of attached cellular material, especially muscles of the abdominal wall (Figs. lB, F, and G). It is possible to manually scrape the abdominal cuticle free of most of the attached tissues (Fig. lC), but even these preparations contain some cellular debris. As a compromise between biochemically purified mass cuticle (Fristrom et al, 1978) and pure cuticle from specific body regions, we have analyzed proteins from hand-dissected tissues, which inevitably contain some cellular contamination. Since hand-dissected tissues yield very small amounts
Synthesis
of Cuticle
Proteins
423
of SDS-soluble proteins, exact protein quantitation is difficult. Figure 2 shows various loads of proteins extracted from the tissues analyzed. Wing integument (e.g., lane 10) contains abundant proteins at 43 (actin), 25-35,23,20, and 14 kDa. Upon lesser loadings of these proteins (lanes 2 and 6), it is apparent that the most abundant is the 20-kDa band, followed in order of decreasing amounts by the bands at 23,14,43, and 2535 kDa. It is significant that the 23-, 20-, and 14-kDa bands are more prominent than that of actin, a generally abundant cellular protein. Abdominal integument and abdominal cuticle (Fig. 2, lanes 3, 4, 7, and 8) contain more complex mixtures of proteins than wing integument contains. This is expected from abdominal integument preparations since the proteins of several cell types are represented. Abdominal cuticle preparations, after removal of most of the contaminating tissues, have fewer proteins than abdominal integument, but they still contain at least
FIG. 1. Light microscopy of late pupal wing (A, D, and E), abdominal integument (B, F, and G), and abdominal cuticle (C) from 90- to 96-hr pupae. Epon-embedded tissues were sectioned at 1 nm and stained with toluidine blue for light microscopy. Panel A is a lowmagnification (460X) micrograph of a transverse section through a folded pupal wing, while panels D and E are high magnifications (1560x) of this tissue. The arrowhead denotes a medial member of the anterior triple row of bristles. Panel B is a low-magnification (460X) micrograph of a section through abdominal integument and underlying tissues, whereas panels F and G are high magnifications (1560X) of selected regions. Panel C is a low magnification (460X) micrograph of a section through a typical abdominal cuticle preparation after most adhering tissue was scraped away. Note the tissue complexity of the abdominal integument preparations relative to wing integument and abdominal cuticle. c = cuticle (ec = epicuticle, pc = procuticle). H = hypodermis. M = muscle. T = tracheole. Bar for panels A, B, and C = 30 pm. Bar for panels D, E, F, and G = 10 pm.
424
DEVELOPMENTALBIOLOGY
12345678910 M WAIACM
WAI
ACMW
-my0
c-
d-
-act
-3Okd -23kd
ef-
I:ss2bd - l6kd - I4kd
FIG. 2. Silver-stained acrylamide gradient (7.5 to 25%) SDS gel showing extractable proteins of wing (W), abdominal integument (AI), abdominal cuticle (AC) and indirect flight muscle (M) from 90to 96-hr pupae. Multiple lanes of single samples (e.g., muscle proteins in lanes 1, 5, and 9) represent relative protein amounts of 1:2:4 from left to right. The letters at the left indicate mohilities of molecular weight marker proteins (a = 92.5, h = 66.2, c = 45, d = 31, e = 21.5, and f = 14.4 kDa). The 14-, 20-, and 23-kDa ACPs and the 16-, 18-, and 30-kDa abdominal proteins as well as actin (act) and myosin heavy chain protein (myo) are indicated.
10 abundant proteins below 45 kDa. We can assume that actin is not a secreted cuticle protein but represents a protein of the contaminating cellular material in these preparations. We observe a dramatic enrichment for nine proteins relative to actin when contaminating tissues are scraped from the cuticle. Specifically, compare lane 3 (integument) with lane 8 (cuticle) in Fig. 2. The intensity of the actin band is reduced severalfold in proteins extracted from scraped cuticle as compared to crude integument. In contrast, the intensities of bands at 40, 35, 30, 28, 23, 20, 18, 16, and 14 kDa are increased between two- and tenfold. The degree of enrichment probably depends on how closely the protein is associated with the cuticle. Qualitatively, all nine of these components are more closely associated with abdominal cuticle than is actin. Of the proteins extracted from wing and abdominal preparations, those represented by the bands at 23,20, and 14 kDa are among the most abundant components common to both tissues. This suggests that they serve
VOLUME 107, 1985
a general role in cuticle structure, although the stoichiometric ratios of these proteins do vary between tissues (e.g., the 20-kDa protein is more abundant in wing than in abdominal cuticle). Preparations of indirect flight muscle, a tissue completely devoid of cuticle, lack any proteins at 23, 20, and 14 kDa, further supporting the identification of these components as cuticle specific. The abdominal bands at 40, 35, 30, 28, 18, and 16 kDa are possibly abdomen-specific cuticle proteins, but we have not characterized them further at this time. Cuticle proteins luck methionine. Since insect cuticle lacks sulfur (Neville, 19’75), we have compared the incorporation of leucine and methionine into proteins synthesized during wing cuticle formation. Figure 3 shows the patterns of incorporation of labeled leucine and methionine into proteins synthesized by 65-hr wing hypodermal cells. Three points are pertinent: (1) The 23-, 20-, and 14-kDa proteins are all synthesized by this tissue at 65 hr, and these three bands account for greater than 40% of the leucine incorporation. (2) The 23-, 20-, and 14-kDa proteins completely lack methionine as shown by their absence of labeling (we know that they are being actively synthesized at this stage). (3) The 23-, 20-, and 1QkDa proteins are the
M LM
C
t
C
FIG. 3. Comparison of leucine and methionine incorporation into cuticle proteins. Wing integument from 65-hr pupae was dissected and pulse labeled for 45 min with [‘Hlleucine or [%S]methionine such that wings from the same animal were being compared. Proteins were analyzed on a 17.5% acrylamide SDS gel. Gel lanes were loaded with either 10,000 *H cpm or 4000 to 5000 ?S cpm. The fluorograph is shown (3-day exposure). L indicates leucine-labeled and M indicates methionine-labeled proteins. The 14-, 20-, and 23-kDa cuticle protein hands (arrows) represent a substantial portion of the leucine but no detectable methionine incorporation into newly synthesized proteins.
ROTER, SPOFFORD, AND SWIFT
most prominent bands which are both leucine-rich and methionine-free and, therefore, account for the major difference obtained when these two labeled amino acids are compared. Cellular localization of the leucine-rich, methioninefree proteins. The striking difference in incorporation patterns of leucine and methionine into wing-synthesized proteins provides a basis for the cellular localization of these components. This is accomplished through EM autoradiography of tissues which have been pulselabeled with either [3H]leucine or r5S]methionine for the period during which the 23-, 20-, and 14-kDa proteins are the prominent leucine-rich, methioninefree synthetic products. These autoradiographs (Fig. 4) show much of the newly synthesized protein over the hypoderm cytoplasm when either labeled amino acid is used (also see Fig. 5). These represent the proteins which contain both leucine and methionine. In contrast, there is a marked concentration of label over the apical cell surface and cuticle in leucine-labeled cells (34% of the silver grains are in the region -0.35 to +0.‘7 pm) but not in those labeled with methionine (12% of the grains are in the region -0.35 to +0.‘7 pm). These data strongly suggest that one or more of the 23-, 20-, or 14-kDa proteins is transported to the cuticle. Because we were not anxious to incubate tissue explants for longer than 1 hr, it was not possible to chase more leucine counts into the cuticle. The trend of accumulation of leucine-rich, methionine-free proteins at the cell apex and cuticle, however, is clear. Loss of extractability of cuticle proteins after eclosion. To see if the extractability of the 23-, 20-, and 14-kDa proteins is affected during the presumed period of cuticle crosslinking, we have analyzed the extractable proteins from wing and abdominal tissues at several time points after eclosion. These data (Fig. 6) show that there is a significant decrease in the amounts of extractable 23-, 20-, and 14-kDa proteins within 60 min of eclosion. The amounts of actin extractable from these tissues remain constant during the posteclosion period. Actin can be considered representative of cellular contamination in these preparations. Therefore, a cellular protein is not affected by the cuticular crosslinking whereas the putative cuticle proteins are dramatically affected. Temporally Regulated Synthesis of Cuticle Proteins Synthesis and accumulation of cuticle proteins in wing integument. To study the synthetic events occurring during cuticle formation, we explanted wing tissue and pulse labeled it for 45 min in vitro. The patterns of proteins synthesized in explanted tissues correlate with proteins synthesized in vivo when labeled amino
Synthesis
of Cuticle
Proteins
425
acids were injected into live pupae but, for in vivo labeling, longer incubation periods were necessary (Roter, 1983). The advantage of the explant pulse-labeling technique is that large amounts of amino acid are incorporated in a short time, allowing for precise staging of animals. Figure 7A shows the pattern of protein accumulation in wing explants. The 23-, 20-, and 14-kDa proteins are detected as accumulated wing proteins only after 58 hr. This corresponds with the pattern of synthesis in that none of these proteins is synthesized at detectable levels before 53 hr (see below). Therefore, if cuticle synthesis begins at about 40 hr (Mitchell et aL, 1983), these components must relate to later events in cuticle synthesis. The synthesis of proteins in wing integument is regulated in a temporally specific manner (Fig. ‘7B). A group of proteins in the range 25 to 35 kDa is rapidly synthesized from as early as 48 to 58 hr with low-level synthesis continuing until 73 hr. The 23-, 20-, and 14kDa proteins, on which we will focus, are the major synthetic products following this early group. The 14kDa protein is synthesized first, beginning at 53 hr, reaching peak synthesis at 63 to 68 hr and decreasing to barely detectable levels by 78 hr. Synthesis of the ZO- and 23-kDa proteins is detectable by 63 hr, peaking synchronously near 73 hr and dropping to a low level that continues through 93 hr. Changes in the levels of actin synthesis have been reported (Mitchell et al, 1983); thus actin cannot be used as a standard for comparing rates of synthesis. Instead the above discussion is based on a comparison of the band intensities of single proteins at the different time points when equal amounts of total counts were loaded per gel lane. Comparison of temporal synthetic patterns in wing and abdomen. Since the 23-, 20-, and 14-kDa proteins are present in cuticle of both wing and abdomen, we have investigated their patterns of synthesis in both tissues. All three proteins are synthesized in 64- to 70hr wing integument and in 70-hr abdominal integument (Fig. 8). Although many proteins are synthesized in the abdominal integument preparations, the 23-, 20-, and 14-kDa bands represent a significant fraction of the leucine incorporation. In agreement with the result that abdominal cuticle contains less of the 20-kDa protein than does wing cuticle (see Fig. 2), the rate of synthesis of the 20-kDa band is lower in abdominal than in wing integument. Other differences include the apparent asynchrony of development between wing and abdomen. Synthesis of the 14-kDa protein in wing begins before 57 hr but is not detected in abdomen at this time (Fig. 8). Abdominal synthesis of the 14-kDa protein begins by 64 hr, a time at which wing is synthesizing all three
426
DEVELOPMENTAL BIOLOGY
VOLUME 107, 1985
FIG. 4. Cuticle morphogenesis and cellular localization of newly synthesized proteins by EM autoradiography. Panels A-C are electron micrographs (19,000X magnification) of abdominal tissues showing the cellular and cuticular appearances at 53, 65, and 80 hr, respectively. Panel D is a typical EM autoradiograph (12,100X) of a 65-hr wing integument cell pulse labeled for 45 min in [3H]leueine and chased with unlabeled amino acids for 15 min. Panel E is similar to D except that the cell has been labeled with [%]methionine. Note the abundance of apical silver grains in leucine-labeled but not in methionine-labeled hypodermis. Bar for panels A, B, and C = 0.5 pm. Bar for panels D and E = 1 @cm.G = Golgi apparatus. EC = epicuticle. PC = procuticle.
components. In addition, although the 20- and 23-kDa proteins are clearly synthesized in 64-hr wing cells, they are only barely detectable in abdominal cells.
These data suggest a 6- to lo-hr difference in timing of developmental events between hypodermal cells of wing and abdomen.
ROTER, SPOFFORD, AND SWIFT
2
‘-
”
0
427
of Cuticle Proteins
other is not clear. A detailed investigation of the posttranslational processing events involved in cuticle protein synthesis is currently underway. Figures 9C, D, and E show the two-dimensional analysis of wing synthesized proteins at 58, 65, and 78 hr. The temporal pattern of synthetic events generally follows that described in Fig. 7B. The synthesis of components within each size class is synchronous, with two exceptions: (1) The 23b spot is synthesized early at 58 hr, while synthesis of the other 23-kDa proteins does not begin until 63 to 65 hr. (2) Synthesis of the 14a protein persists longer (until 78 hr) than synthesis of the other spots near 14 kDa and for a longer period than is detected on one-dimensional gels (see Figs. 7 and 8), probably due to the higher sensitivity obtained on two-dimensional gels.
METHIONINE
-4.2-3.5-2.8-2.1
Synthesis
-1.4-0.7 DISTANCE
0 FROM
0.7
1.4 2.1
CUTICLE
2.8
3.5 4.2
SURFACE
4.9
DISCUSSION
5.6 6.3
(pm1
FIG. 5. Analysis of autoradiography. EM autoradiographs similar to those shown in Fig. 4 were quantitated as described under Materials and Methods. The distance from the outer surface of the cuticle is expressed on the ordinate with negative numbers representing the outside and positive numbers representing the cellular side of the cuticle. The shaded bars represent the grain density directly over the cuticle. For leucine-labeled tissue, N = 249; for methionine-labeled tissue, N = 369, where N = the total number of silver grains analyzed.
Two-dimensional gel analysis of synthesized proteins. Although the 23-, 20-, and 14-kDa bands are considered above as single proteins, analysis of 20% acrylamideSDS gels reveals two bands at 23 kDa (Figs. 6 and 7) and a group of components near 14 kDa (Fig. 7B). Repeated attempts to analyze 65-hr wing-synthesized proteins on standard two-dimensional gels (first dimension: pH 3.5 to 10 isoelectric focusing; second dimension: SDS-slab gel) result in the resolution of many protein spots, but no major synthetic products in the 23-, 20-, or 14-kDa size classes (data not shown). In contrast, the use of a nonequilibrium pH gradient electrophoresis (NEPHGE) as a first dimension, which allows resolution of proteins with high isoelectric points (O’Farrell et ab, 1977), results in the resolution of major synthetic products in the size classes of interest (Fig. 9). Wing (64 hr, Fig. 9A) and abdominal (70 hr, Fig. 9B) tissues synthesizing all three of the 23-, 20-, and 14kDa proteins were analyzed on NEPHGE two-dimensional gels. The 23- and 20-kDa bands resolve into 4 and 2 proteins, respectively, in both tissues. Wing tissue synthesizes many components at 14 kDa but abdominal cells appear to make only some of these proteins. Exactly how these components relate to each
The Major Wing Cuticle Proteins of D. melanogaster The classical isolation and identification of cuticle proteins has usually depended on differences in solu-
PUPA
10
30
60
ADULT
WAWAWAWAWA
-act
-23kd -2Okd :;gj -14kd
FIG. 6. Decrease in extractability of cuticle proteins after eclosion. Wing integument and abdominal cuticle were dissected from a late pupa (90-96 hr), flies at 10, 30, and 60 min after eclosion and older adults (12 hr after eclosion). Gel lanes for each time point contained proteins from the tissues of single animals except for the 12-hr point which required tissues of two animals. The silver stained 20% acrylamide SDS gel is shown. The 23-, 20-, and ll-kDa cuticle proteins are indicated as well as the 16- and 18-kDa abdomenspecific proteins and actin (act). The flies at 30 min and older had fully expanded wings prior to their dissection. Note the decreased amounts of cuticle proteins relative to cellular proteins (e.g., actin) as the post-eclosion crosslinking process proceeded.
428
DEVELOPMENTAL BIOLOGY 48 53 58 63 68 73 70 83 a8 93 .---.---m*
VOLUME 107. 1985 48 53 58 63 68 73 78 83 88 93
- act
- 16.9kd
- 14.4kd
B FIG. 7. Synthesis and accumulation of cuticle proteins in wing integument. Wing integument was dissected from pupae at various stages of cuticle formation (indicated above each lane in hr post-pupariation) and pulse labeled for 45 min with [‘Hlleucine. Panel A is the silver stained 20% acrylamide SDS gel, showing total wing protein, and panel B is a fluorograph of the same gel, showing newly synthesized protein. 25,000 cpm were loaded per lane (d-day fluorographic exposure). The mobilities of molecular weight marker proteins are shown (16.9 kDa = myoglohin, 14.4 kDa = lysozyme). The 14-, 20-, and 23-kDa cuticle protein bands as well as actin (act) are indicated. Note the rapid synthesis of cuticle proteins from 58 to 33 hr, and the continued low-level synthesis of the 20- and 23-kDa proteins through 93 hr.
bility properties in cuticle preparations. Extraction of cuticle with boiling water yields two fractions, the soluble “arthropodin” proteins and those remaining insoluble in the cuticle, the “sclerotin” (Richards, 1951). Later investigators have used sequential extractions with water, NaCl or KC1 solutions, aqueous urea, dilute alkali, and boiling HCI to remove different fractions of proteins, peptides, or amino acids with increasingly stronger denaturing conditions (reviewed by Hackman, 1974; Andersen, 1979). Sequential extractions remove groups of proteins which are presumably held in the cuticle by bonds of varying strength, from ionic and H bonds to covalent linkages. Fristrom et al. (1978) used variable solubilities to purify a small group of saltsolution-insoluble, aqueous-urea-soluble, proteins from This success third instar cuticles of D. wwlanoga&w. with larval cuticle led to the use of these methods to
define cuticle proteins from other stages of the L?r+ sophila life cycle, including the adult (Chihara et cd, 1982), but the large number of resulting proteins has led us, in the present work, to define a subset of the adult cuticle proteins of D. melanogaster and to present evidence, using several biologically related techniques, that these are indeed major components of the cuticle. We present evidence that the 23-, 20-, and 14-kDa proteins are the major cuticle proteins in adult wing tissue. These proteins are abundant in cuticle-containing tissue. Like other cuticle proteins (Neville, 19’75), they lack methionine. Their extractability decreases during the period of cuticle hardening, a property expected of proteins being crosslinked covalently into the cuticular matrix. Radioautographic localization of silver grains to the cuticle in leucine but not methionine labeled tissues is primarily due to these proteins. There
ROTER,
57
64
---WAWAWAWA
70
SPOFFORD,
AND
SWIFT
78
FIG. 8. Differential temporally regulated synthesis of cuticle proteins in abdominal (A) versus wing (W) integument. Wing and abdominal integument explants for each time point were dissected from three staged pupae such that the different tissues came from the same animals. The tissues were pulse labeled for 45 min with [3H]leucine and analyzed on a 15% acrylamide-SDS gel fluorograph. Lanes were loaded with approximately 10,000 cpm of wing extract or 20,000 cpm of abdominal extract. As indicated at the top of the lanes, wing and abdominal tissues were analyzed at 57, 64, 70, and 78 hr postpupariation. The 14-, 20-, and 23-kDa major cuticle proteins (as marked) show different stage specificities of synthesis in wing versus abdomen.
are also indications (Roter, 1983), to be published elsewhere, that these proteins undergo the processing expected of secreted proteins, reinforcing their eventual extracellular localization. Temporally Regulated Synthesis
Early light microscopic (reviewed by Wigglesworth, 1948; Richards, 1951) and more recent electron microscopic studies (reviewed by Locke, 1976) demonstrate the sequential deposition of cuticle layers. The sequential pattern of synthesis of the cuticle proteins seen in the present work (i.e., synthesis of the 14-kDa protein begins and peaks before the 20- and 23-kDa proteins) is doubtless another indication that these components are synthesized as they are needed for construction of different cuticle layers. The precise location of these proteins in the cuticle is still unclear, since the timing of their synthesis does not strictly correlate with cuticle thickening. According to Mitchell et al. (1983),
Synthesis
of Cuticle Proteins
429
synthesis of D. melanogaster wing cuticle begins with secretion of the outermost cuticulin layer at about 40 hr past puparium formation and wing cuticle reaches maximum thickness by about 60 hr. Our fluorographs demonstrate a group of proteins in the size range 25 to 35 kDa that are rapidly synthesized during the later part of this period (see Fig. 7B). They are probably also cuticle proteins, but we have not focused on them in these initial studies because (1) they were not detected as the most abundant proteins in late pupal cuticle preparations, and (2) they appear to become unextractable shortly after synthesis, a property which makes them difficult to study. The synthesis of the 23-, 20-, and 14-kDa major wing cuticle proteins occurs primarily after the cuticle has reached its presumed full thickness. In contrast, abdominal cuticle thickens dramatically during the 65- to 80-hr period (see Figs. 4B and 4C). In addition, the striking chitinous lamellar pattern seen in the abdominal procuticle layer (Fig. 4C) and seen generally in insect procuticle (Neville, 1975) appears to be either highly condensed or completely absent in wing cuticle (Mitchell et al, 1983; our unpublished observations). Since regional variation of cuticle structure lies principally in the procuticle layer, it is not surprising that wing procuticle differs greatly from abdominal procuticle. With these differences in mind, we believe there are several possible explanations for the synthesis of what we contend to be major wing cuticle proteins after the presumed completion of cuticle thickening: (1) These proteins may be synthesized late yet enter a preformed chitin matrix to “fill-in” and strengthen the procuticle. (2) These proteins may construct an inner cuticular layer which bonds the cell and the procuticle. (3) These proteins may be components of thin procuticle layers which are not obvious on electron micrographs. Because of the folded and convoluted nature of preeclosion wing cuticle, the visualizations and precise thickness measurements of cuticle layers are difficult. The exact localization of the major wing cuticle proteins will have to await specific cytochemical analysis, for example, by using antibodies to individual cuticle proteins. The difference in patterns of cuticle protein synthesis between wing and abdominal hypodermis reported here is indicative of a temporal offset in the states of differentiation of these cells. The value of about 6 to 10 hr difference between the developmental stages of wing and abdomen agrees with the value of 9 to 10 hr determined by Mitchell and Petersen (1981). As demonstrated using bithorax mutants, the timing of differentiation in hypoderm cells depends on the determined fate of the cells, not their position in the fly (Mitchell and Petersen, 1983). The characterization of molecular events in the regulation of specific synthetic products,
430
DEVELOPMENTAL BIOLOGY OH-
VOLUME 107, 1985
NE PHGE
H+
A
FIG. 9. Two-dimensional gel analyses of stage-specific protein synthesis in wing (A, C, D, and E) and abdominal (B) integument. Wing and abdominal integument explants were pulse-labeled with [3H]leucine, and the newly synthesized proteins were analyzed by twodimensional electrophoresis and fluorography. Panel A shows proteins synthesized by 65-hr wing and panel B shows those made by 70-hr abdominal tissues. Spots corresponding in molecular weight to the major cuticle proteins are enclosed in circles and their size classes are indicated. The actin spots are enclosed in squares for reference. These two stages are chosen because they represented periods when all of the major cuticle proteins are synthesized. Panels C through E show the low molecular weight region at the basic end of gels containing newly synthesized proteins from 58, 65-, and ‘7%hr wing integument explants, respectively. Each member of each size class is designated by a letter, starting with the most acidic protein and lettering each spot respectively. Note that most of the proteins within each size class of cuticle proteins are synthesized coordinately, with the exception of spot 23b which is synthesized before the other 23-kDa proteins. Also, spot 14a shows incorporation at ‘78 hr when other ll-kDa spots (14b-d) no longer label.
such as the major wing cuticle proteins, may provide a significant step toward an understanding of the processes responsible for this temporal control of cell differentiation. We thank Ms. Sagami Paul for her excellent assistance with histological preparations and thin sectioning. This work was supported in part by NIH Grant CA-14599. A.H.R. was supported by NIH Training Grant HD-07136 to The Committee on Developmental Biology. REFERENCES ANDERSEN, S. 0. (1979). Biochemistry of insect cuticle. Annu. Rev. Entomol. 24, 29-61.
BONNER, W. M., and LASKEY, R. A. (1974). A film detection method for tritium-labeled proteins and nucleic acids in polyacrylamide gels. Eur.
J. Biochem.
46, 83-88.
CHIHARA, C. J., SILVERT, D. J., and FRISTROM, J. W. (1982). The cuticle proteins of Drosophila melanogaster: Stage specificity. Dev. Biol
89, 379-388.
ELGIN, S. C. R., and MILLER, D. W. (1978). Mass rearing of flies and mass production and harvesting of embryos. In “The Genetics and (M. Ashburner and T. R. F. Wright, eds.), Biology of Drosophila” Vol. 2b, pp. 112-120. Academic Press, New York. FRISTROM, J. W., HILL, R. J., and WATT, F. (1978). The procuticle of Drosophila: Heterogeneity of urea-soluble proteins. Biochemistry 17, 3917-3924. HACKMAN, R. H. (1974). Chemistry of insect cuticle. In “The Physiology
ROTER,
SPOFFORD,
AND
SWIFT
of Insecta” (M. Rockstein, ed.), 2nd ed., Vol. VI, pp. 237-258. Academic Press, New York. HEPBURN, H. R. (1976). “The Insect Integument.” Elsevier, Amsterdam. HILLMAN, R., and LESNIK, L. H. (1970). Cuticle formation in the embryo of Drosophila melanogaster. J. Mwphd 131,383-396. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature &n&m) 227, 680-685. LASKEY, R. A., and MILLS, A. D. (1975). Quantitative film detection of $H and ‘“C in polyacrylamide gels by fluorography. Eur. J. B&hem 56,335-341. LOCKE, M. (1976). The role of plasma membrane plaques and golgi complex vessicles in cuticle deposition during the molt/intermoult cycle. In “The Insect Integument” (H. R. Hepburn, ed.), pp. 237258. Elsevier, Amsterdam. MITCHELL, H. K., LIPPS, L. S., and TRACY, U. M. (1977). Transcriptional changes in pupal hypoderm in Drosophila melanogaster. B&hem. Genet. 15, 575-587. MITCHELL, H. K., and MITCHELL, A. (1964). Mass culture and age selection in Drosophila Drosophila Ir&rm. Serv. 39, 135-137. MITCHELL, H. K., and PETERSEN, N. S. (1981). Rapid changes in gene expression in differentiating tissues of Drosophila. Dev. Biol 85, 233-242. MITCHELL, H. K., and PETERSEN, N. S. (1983). Gradients of differentiation in wild-type and bithorax mutants of Drosophila Dev. BioL 95, 459-467. MITCHELL, H. K., ROACH, J., and PETERSEN, N. S. (1983). The morphogenesis of cell hairs on Drosophila wings. Deu. Biol 95, 387-398. NASH, W. G. (1976). Patterns of pigmentation color states regulated by the y locus in Drowphila melanogaster. Dew. Biol 48, 336-343. NEVILLE, A. C. (1975). “Biology of the Arthropod Cuticle.” SpringerVerlag, New York.
Synthesis
of Cuticle
Proteins
431
OAKLEY, B. R., KIRSCH, D. R., and MORRIS, N. R. (1980). A simplified silver stain for detecting proteins in polyacrylamide gels. Anal. B&hem. 105,361-363. O’FARRELL, P. Z., GOODMAN, H. M., and O’FARRELL, P. H. (1977). High resolution two-dimensional electrophoresis of basic as well as acidic proteins. Cell 12, 1133-1142. POODRY, C. A. (1980). Epidermis: Morphology and development. In. “The Genetics and Biology of Drosophila” (M. Ashburner and T. R. F. Wright, eds.), Vol. 2d, pp. 443-497. Academic Press, New York. RICHARDS, A. G. (1951). “The Integument of Arthropods.” Minnesota Univ. Press, Minneapolis. ROTER, A. H. (1983). “The Adult Cuticle Proteins of Drosophila melanogaster: A Model System for Studying Developmentally Regulated Gene Expression.” Ph. D. thesis, University of Chicago. SALPETER, M. M., and BACHMANN, L. (1972). Autoradiography. In “Principles and Techniques of Electron Microscopy” (M. A. Hayat, ed.), Vol. 2, pp. 221-278. Van Nostrand-Reinhold, New York. SHIELDS, G., and SANG, J. H. (1977). Improved medium for culture of Drosophila Embryonic Cells. Drosophila I@rm Serv. 52, 161. SYNDER, M., HIRSH, J., and DAVIDSON, N. (1981). The cuticle genes of Drosophila: A developmentally regulated gene cluster. Cell 25,165177. SNYDER, M., HUNKAPILLER, M., YUEN, D., SILVERT, D., FRISTROM, J., and DAVIDSON, N. (1982). Cuticle protein genes of Drosophila: Structure, organization and evolution of four clustered genes. Cell 29, 1027-1040. VAN KEUREN, M. L., GOLDMAN, D., and MERRJL, C. R. (1981). Detection of radioactively labeled proteins is quenched by staining methods: Quenching is minimal for “C and partially reversible for ‘H with a photochemical stain. Anal. B&hem 116, 248-255. WIGGLESWORTH, V. B. (1948). The insect cuticle. Biol. Rev. 23, 408451.