- Email: [email protected]
Developmental variations of a nonfibroin mRNA of Bombyx mori silkgland, encoding for a low-molecular-weight silk protein
DEVELOPMENTAL
BIOLOGY
97, 398-407
(1983)
Developmental Variations of a Nonfibroin mRNA of Bombyx mori Silkgland, Encoding for a Low-Molecular-Weight Silk Protein PIERRE D&oar&ment
de Biologie
COUBLE,
AGNBS
MOINE,
ANNIE
GAREL,
AND
G&&ale et Appliqutk, Labwratoire Associi au CNRS boulevard du 11 Novembre 1918 69622, Villeurbanne Received
January
8, 1992; accepted
in revised
fm
JEAN-CLAUDE
PRUDHOMME
No. 92, Universith Cedex, France January
Claude
Bernard
Lyon-I
.&‘s’,
7, 198.3
The characterization of a new silk protein mRNA (P25 mRNA) in posterior silkgland cells (PSG) and the developmental variations of its cell molecular concentration versus that of fibroin mRNA are described. A 80% pure P25 cDNA was obtained by class separation of total nonfibroin cDNA from PSG and used to identify the mRNA in blotted PSG mRNA as a single 1100 nucleotide long species. When purified from agarose gel and translated in a reticulocyte cell-free system, P25 mRNA yielded a 25-kD polypeptide (P25), identical to a 25-kD protein of the cocoon in terms of pIvalue and partial peptide mapping pattern. Moreover, this protein comigrated with an abundant polypeptide of the posterior silkgland (PSG) and of the middle silkgland (MSG). When tritiated leucine was injected in vivo, labeled P25 showed up in the PSG after a 2-hr pulse but appeared in the MSG only after 24 hr of labeling. Since MSG cells were found to be devoid of P25 mRNA, we concluded that P25 is exclusively synthesized in the PSG, that it accumulates in the MSG lumen and that it is spun out in the same way as fibroin. Specific probes were used to measure the concentrations of P25 mRNA and also fibroin mRNA in PSG total RNA by hybridization with an excess of cDNA. Both species are highly degraded in the few hours following the physiological arrest of feeding which precedes the fourth molting period. Their subsequent accumulation during the fifth intermolt is triggered by food uptake and proceeds in such a way that a constant 1:l molar ratio is maintained during the period of silk secretion. INTRODUCTION
contribute to the production of two silk proteins in the same cell, we studied the stoichiometric relationships between their cellular content at periods of both intense and inactive silk secretion. Comparative titrations during development showed that their cell molecular concentrations are equal, suggesting that the transcription of their genes as well as their stability are probably dependent on the same regulatory mechanisms.
B&X mori posterior silkgland (PSG) is highly specialized in the synthesis and the secretion of fibroin, the major protein of the cocoon. PSG mRNA complexity analysis has shown that the pool of polyadenylated mRNA varies drastically between the stage of maximum fibroin synthesis (mid-fifth intermolt) and the stage of no production of fibroin (larval molting) (Couble et a,!., 1981a,b). This concerns, in particular, two mRNA: the fibroin mRNA and another mRNA, the function of which is unknown. Interestingly, kinetics analysis indicated that both species are the most abundant sequences at mid-intermolt, whereas at ecdysis, their representation is lowered 100 to lO,OOO-fold. Variation of cellular fibroin mRNA concentration reflects the existence of a cyclical turn-on and turn-off of the transcription of the gene and a degradation, occurring at molting, of the mRNA accumulated during the preceding intermolt (Suzuki and Suzuki, 1974; Maekawa and Suzuki, 1980). However, no data were available to interpret the existence of the other abundant mRNA that could account for the synthesis of a well-known PSG protein. In this paper, we report the characterization of this mRNA (P25 mRNA) and its identification as a species encoding for a low-molecular-weight silk protein (P25) produced by the posterior (PSG) but not in the middle silkgland (MSG). As fibroin and P25 mRNA 0612-1606183 Copyright All rights
$3.06
0 1983 by Academic Press. Inc. of reproduction in any form reserved.
MATERIALS
AND
METHODS
Silkworms Bombgx mori larvae, hybrids of European strains 200 and 300, were reared in the laboratory with fresh mulberry leaves provided four times daily. Silkglands were studied at various stages of development from the end of the fourth intermolt up to Day 7 of the fifth and last larval intermolt. RNA Extraction
and Liquid
H@idixation
Total RNA was extracted from freshly dissected silkglands according to the LiCl-urea method of Auffray and Rougeon (1930). The purification of polyadenylated mRNA by oligo-dT cellulose chromatography, the separation of fibroin mRNA from bulk mRNA (nonfibroin mRNA) by sucrose gradient centrifugation, the synthesis of complementary cDNA, and the liquid hy398
COUBLE
ET AL.
Low-Molecular-
Weight
of Speci&
P25
cDNA
A purified P25 cDNA was obtained by fractionation of cDNA reverse transcribed from PSG nonfibroin mRNA extracted at Day 6 of the fifth intermolt. For this purpose, an excess of nonfibroin mRNA was incubated with homologous 3H- or 32P-labeled cDNA up to a R,t of 5 X 10m3. Hybridization was then stopped by adding 30 ml of 0.01 M phosphate buffer (pH 7.0) and the mixture was chromatographed on hydroxylapatite at 6’7°C. Single-stranded cDNA was eluted with 0.12 M phosphate buffer whereas the double-stranded P25 cDNA was recovered by washing with 0.5 M phosphate buffer. The procedure was repeated twice and yielded 80 to 85% pure P25 cDNA as tested by hybridization with an excess of nonfibroin mRNA (see Fig. 2). P25
mRNA
and Fibroin
mRNA
I a
bridization of cDNA with an excess of polyadenylated mRNA were carried out as previously described (Couble et aZ., 1981a,b). Purification
Silk
mRNA
Titrations
Titration of specific sequences was performed by hybridizing PSG total RNA with an excess of tritiated cDNA. For determination of the content of fibroin mRNA, we used cDNA from sucrose gradient-purified mRNA: 0.5 ng of cDNA was incubated with increasing amounts of heat-denatured total RNA for 5 hr at 65°C in 50 ~1 containing 10 mM Tris-Cl (pH 7.5), 0.2 mM EDTA, and 0.4 M NaCl in order to attain saturation. The mixture was then diluted 10 times with 30 mM sodium acetate (pH 4.5), 0.1 M NaCl, and 0.12 mM ZnSO(. Half of the volume was TCA precipitated and the other half treated with 1000 units/ml of Sl nuclease (Miles laboratories) for 1 hr at 3’7°C before precipitation. Precipitates were collected on nitrocellulose filters (Millipore) and the percentage of hybrids resistant to nuclease treatment was determined. The fibroin mRNA content was obtained by comparing the slope of each experimental curve with that of a reference curve obtained by annealing, under the same conditions, known amounts of purified fibroin mRNA with an excess of specific cDNA (Fig. lb). Titration of P25 mRNA was carried out similarly. However, 0.6 ng of class-purified P25 cDNA, 600 nucleotides long, was used in a final volume of 20 ~1; also, we found it necessary to continue the incubation for 18 hr in order to complete the reaction. The standard used to quantify P25 mRNA in the RNA to be tested was nonfibroin mRNA extracted at Day 6 of the fifth intermolt and containing 29% of P25 mRNA as determined by complexity analysis (see Fig. la).
bp
0:2 ng
of
P25
pg
of
V6
pg
I
in
total
of
10hr
0:s
non-fibroin
RNA
mRNA
”
5
(a)
(0) 0:l
0:05 0.5
h
d
014
mRNA
1
1.5
post
ecdyiis
total
RNA
IA)
b
0 a ,N .E502 c” ;; 258
ng
of
fibroin
ng
of
V6
pg
of’lOhr
mRNA total
RNA
post
(0)
to?al
RNA
(3
FIG. 1. Titration of P25 and fibroin mRNA. Standard curves of cDNA excess hybridization experiments are represented together with two experimental curves, given as examples. a Shows the hybridization of various amounts of nonfibroin mRNA known to contain 29% of P25 mRNA (O), total RNA from 10 hr postapolysis (A), and from Day 6 of the fifth intermolt (0) with 0.6 ng of purified tritiated P25 cDNA. b Shows the hybridization of 0.5 ng of tritiated fibroin cDNA when incubated with increasing amounts of pure fibroin mRNA (O), 10 hr postapolysis (A) and 6 days ecdysis (0) total RNA. Background values were not subtracted.
mRNA
Electrophoresis
and Hybridizations
“Northern” blots were performed according to the method of Thomas (1980). Briefly, mRNA were denatured in 1 M glyoxal and 50% dimethyl sulfoxide for 1 hr at 5O”C, then electrophoresed on horizontal 1.5% agarose gel and transferred by blotting onto 0.45-pm nitrocellulose filters (Schleicher and Schull). Hybridizations were carried out with 32P-labeled cDNA in 50% formamide, 0.75 M NaCl, 0.075 M sodium citrate, 50 mM
DEVELOPMENTALBIOLOGY VOLUME 97,1983
400
phosphate, 0.02% each of polyvinylpyrolidone, ficoll, and bovine serum albumin, 200 pg/ml of herring sperm DNA, for 20 hr at 42°C. After washing, the blots were exposed to Kodak X-ray films. P25 mRNA
PuriJicaticm
Urea-denatured PSG nonfibroin mRNA was electrophoresed in 1.5% agarose and the position of P25 mRNA located with reference markers (N&d111 restricted fragments of X-phage DNA). P25 mRNA as well as mRNA of lower and higher molecular weight were then eluted in wells by means of a second electrophoresis, perpendicularly to the first direction, and ethanol precipitated. Analysis
of Cell-Free
Translation
Products
Micrococcal nuclease-treated rabbit reticulocyte lysate (Amersham) was supplemented with 1 pCi/pl of tritiated leucine (123 Ci/mmole) or serine (28 Ci/mmole) and used at the rate of 10 ~1 per assay: 0.5 to 1 pg of various mRNA was tested and incubation was carried out at 37°C for 30 min. The mixture was then brought to 4% SDS and 1% @mercaptoethanol, heated at 100°C for 3 min, and electrophoresed in a linear 3.75-15% gradient of acrylamide with a 3% acrylamide stacking gel, according to Laemmli (1970). 14C-labeled protein markers (Amersham) were coelectrophoresed for calibration of the migration. Gels were treated for fluorography according to Bonner and Laskey (1977) and autoradiographed on X-ray Kodak films. In Vivo Labeling
and Analysis
of Silk Proteins
Silkworms received 20 &i of [3H]leucine (123 Ci/ mmole) and the total proteins extracted from PSG and MSG were analyzed after 2 hr a labeling pulse that allows silk proteins to be secreted into the lumen of the organ (Couble et aL, 1977), and after a 24-hr chase so as to study PSG silk protein accumulation in the MSG. Soluble proteins were recovered from posterior and middle sections of freshly dissected silkglands, after rinsing the glands several times in distilled water to remove hemolymph contaminants, and incubating small pieces in 200 to 500 ~1 of distilled water. Silk proteins were extracted from the cocoon by either 60% (w/w) neutral LiSCN or a 4% SDS and 7 M urea mixture. ,. Proteins were treated with fl-mercaptoethanol prior to electrophoresis on a 3.75-15% linear gradient of acrylamide as described for cell-free synthesized proteins.
Peptide
Mapping
Proteins were mapped by limited proteolysis, in the presence of SDS, with V8 protease from Staphylococcus aureus, according to Cleveland et al. (1977) and to Bordier and Crettol-Jarvinen (1979). LiSCN-extracted cocoon proteins (100-200 pg) and [3H]leucine-labeled P25 from in vitro translation of purified P25 mRNA were electrophoresed on a 15% acrylamide, 2-mm-thick slab gel according to Laemmli (1970). Gel regions with 25kD proteins were excised and incubated in 125 mM TrisCl (pH 6.8) and 0.1% SDS for 1 hr at room temperature. Gel pieces were then mounted between two casting glass plates prior to pouring a 17.5% acrylamide, 2-mm-thick resolving gel, with a 3% stacking gel. Stacking gel covered the first dimension gel lanes by 1 mm. The gel was placed in the electrophoresis apparatus and overlayed with 1.5 ml of sample buffer containing 4 to 16 pg of V8 protease (Miles laboratories). Following costacking of protease and silk proteins, the electrophoresis was interrupted for 40 min, and then migration was completed. Cocoon proteins and 3H-labeled P25 maps were revealed, respectively, by Coomassie brilliant blue staining and fluorography. RESULTS
P25 mRNA
Probe
PSG cells of Bom&r mori contain a major mRNA species, besides fibroin mRNA, the existence of which was deduced from hybridization kinetics of an excess of nonfibroin mRNA with homologous cDNA (Couble et al, 1981a,b). Since this mRNA was estimated to represent 30% of nonfibroin mRNA and defined a major class of complexity, the purification of the corresponding cDNA was carried out through class separation of total nonfibroin cDNA (see Materials and Methods). The resulting probe, back hybridized with an excess of nonfibroin mRNA, exhibited 75 to 85% of class-specific cDNA as shown on Fig. 2. This probe was used for sizing and titrating the mRNA under study that we named P25 mRNA according to the size of the encoded protein (see below). P25 mRNA
Sizing
P25 mRNA was localized on Northern blots of total RNA (Fig. 3a) and nonfibroin mRNA (Fig. 3b) by hybridization with purified 32P-labeled P25 cDNA. In both cases, P25 cDNA annealed to a single species; its size was estimated at 1100 nucleotides using glyoxaldenatured Hind111 restricted fragments of X-phage DNA as markers according to MC Master and Carmichael (1977). As a control, total cDNA recovered as single-stranded
COUBLE ET
Low-Molecular-
AL.
Silk
401
mRNA
sized in the cell-free system driven by the agarose purified mRNA of lower and higher molecular weight than that of P25 mRNA (Figs. 4b and c). As expected, P25 was found as the most labeled polypeptide when nonfibroin mRNA from PSG were translated in vitro (Fig. 4d). It was also detected, but at much lower relative intensity when polyadenylated mRNA from molting stage were cell-free translated (compare Figs. 4d and e), as predicted from the lower relative abundance of P25 mRNA at this period (Couble et ah, 1981b). Similar experiments carried out with tritiated serine instead of leucine led to the same conclusions (data not shown), suggesting that the relative P25 signal intensity in both cases accounted for the abundance of the protein.
80
E o .c ro .-N ‘*-0 i
Weight
60
= 4c s
Identi,fication of P25 as a Silk Protein
2c
The existence of a major polypeptide band of 25 kD in the total protein extracts of the posterior (Fig. 5a) and middle (Fig. 5d) silkgland and in the cocoon silk (Figs. 5g and h) suggested that P25 could be a silk pro-
1I lo-’
lo-’
1o-z
R,
lo-’
10”
10’
x t
b
C
FIG. 2. Control of P25 cDNA and P25 mRNA purification procedures. Class-purified P25 cDNA (0) and cDNA reverse transcribed from agarose gel purified P25 mRNA (0) were hybridized with an excess of polyadenylated non-fibroin mRNA. The reference curve (- - -) corresponds to the kinetics of hybridization of an excess of nonfibroin mRNA with homologous cDNA. The percentage of hybridization is plotted against log R& the product of the concentration of mRNA in moles of nucleotides z 1-l z time (set). Symbols represent experimental data without self-annealing corrections.
species after class separation of P25 cDNA was hybridized with nonfibroin mRNA (Fig. 3~). As expected, a number of mRNA showed up, including P25 mRNA, since the probe was not completely devoid of P25 cDNA. Characterization
P25
m RNA
of P25 mRNA Translation Product
A rabbit reticulocyte lysate was used as a cell-free translation system to study the proteins encoded by various preparations of mRNA. The denatured leucineor serine-labeled proteins were analyzed by electrophoresis on a 3.75-1570 linear gradient of acrylamide. P25 mRNA was purified from agarose gel taking advantage of its abundance in polyadenylated nonfibroin mRNA. Hybridization kinetics of an excess of nonfibroin mRNA with the cDNA transcribed from purified P25 mRNA showed that the preparation contained a minimum of 55% of the specific sequence (Fig. 2). When polypeptide of 25 kD translated in vitro, a predominant (=P25) was obtained (Fig. 4a). Such a protein was found in reduced amounts among the polypeptides synthe-
FIG. 3. Characterization of P25 mRNA in PSG RNA separated on agarose gel. P25 mRNA was revealed as a single species on the following Northern blots of RNA: lanes a and b are, respectively, total RNA (10 pg) and nonfibroin mRNA (1 pg) separated on a 1.5% agarose gel and hybridized with class purified 3zP-P25 cDNA; lane c corresponds to nonfibroin mRNA, as in lane b, hybridized with BP total nonfibroin cDNA recovered as single-stranded cDNA after class separation of P25 cDNA.
DEVELOPMENTAL BIOLOGY
402
a
b
c
VOLUME 97,1983
d
e
f
FIG. 4. In vitro characterization of the P25 mRNA encoded protein in cell-free products of various mRNA preparations. Autoradiograms correspond to the cell-free [BH]Ieucine-labeIed proteins translated from agarose purified P25 mRNA (a), PSG nonfibroin mRNA of higher (b) and lower (c) molecular weight than P25 mRNA, PSG nontibroin mRNA from Day 6 of the fifth intermolt (d), total mRNA from PSG at exuviation (e) and from MSG at mid-last intermolt (f). Proteins were denatured in SDS and @-mercaptoethanol prior to electrophoresis in a 3.75-15% linear gradient of acrylamide. Size of marker proteins are indicated in kD.
tein assembled and secreted in the F’SG, stored in the MSG lumen, and spun, like fibroin itself. Since these proteins detected in viva could not be identified as P25 solely by virtue of their apparent size, we further compared them in terms of isoelectric point and partial peptide mapping pattern. First, double-dimension electrophoresis according to O’Farrell (1975) indicated a p1 value of 4.8-4.9 for both the in vitro synthesized P25 and a 25-kD protein of the cocoon (data not shown). In a second experiment, the in vitro and in vivo 25kD proteins, initially separated on acrylamide gel, were digested by Staphylococcus V8 protease and the resulting peptides were analyzed by second electrophoresis. The results presented on Fig. 6 show that the partial digestion of labeled P25 generated two peptides that comigrated with two proteolytic products arising from the cleavage of one 25-kD silk protein of the cocoon. In addition, a second, slower migrating 25-kD silk protein was detected as an uncleaved polypeptide after V8 protease digestion. The characteristics of size, pIvalue, and peptide mapping pattern that P25 has in common with a cocoon
protein led to the conclusion that P25 mRNA does indeed encode for a silk protein. Some data indicated that P25 production is specific of PSG cells. Interestingly, the presence of an abundant 25-kD protein in MSG protein extracts could not be explained by the presence of a corresponding mRNA since polyadenylated mRNA from this region of the organ failed to produce any 25-kD polypeptides when translated in vitro (Fig. 4f). After 2 hr of in vivo labeling with tritiated leucine, P25 was intensely labeled in the PSG (Fig. 5b) but remained unlabeled in the MSG (Fig. 5e). Following a 24-hr chase, the relative radioactivity of P25 was decreased in the PSG (Fig. 5c), whereas tritiated 25-kD protein from MSG showed up as an intense band (Fig. 5f). We therefore conclude that P25 is probably synthesized exclusively in the PSG cells and that the 25-kD protein in the MSG represents the luminal P25 accumulated in the middle silkgland before spinning. P25 mRiVA Variations Throughout Development
The cellular concentration of P25 mRNA in the PSG was determined and compared to the variations of fi-
COUBLE
a
b
ET AL.
i&w-Molecular-
d
c
Weight
e
Silk
mRNA
403
f
r S L
P25-
FIG. 5. In viva detection of cold and labeled P25 in total protein extracts from PSG, MSG, and cocoon. A 25-kD abundant protein is present in the Coomassie brilliant blue-stained proteins of PSG (a) and MSG (d) and in the silk proteins of the cocoon extracted either by neutral LiSCN (g) or SDS-urea (h). Lanes b and c are autoradiograms of in viva [3H]leucine-labeled proteins from PSG at, respectively, 2 hr (b) and 24 hr (c) after injection of the precursor. Similarly, short-and long-term labelings of MSG proteins are, respectively, shown in lanes e and f. In all cases, proteins were treated with SDS and P-mercaptoethanol and analyzed on a 3.75-15% linear gradient of acrylamide. Marker sizes are indicated in kD. f, fibroin; s, sericin.
broin mRNA content. This was done at different stages occurring at the end of the larval life, including the last day of the fourth intermolt followed by the fourth molting stage and the first ‘7 days of the fifth and last intermolt (cocoon spinning started on Day 8). For this purpose, cDNA excess experiments were performed using class-purified [3H]P25 cDNA and [3H]fibroin cDNA obtained by reverse transcription of the specific mRNA prepared by sucrose gradient centrifugation (Couble et aL, 1981). Standard amounts of the cDNA were incubated with increasing amounts of total RNA from PSG cells at different developmental stages, under experimental conditions that allowed hybridization saturation at each RNA concentration. The concentration of each sequence was deduced from the comparison of the slope of the experimental curves with that of reference curves (Fig. 1). We emphasize the high degree of sensitivity of this method, since it allows detection of sequences no more abundant in the cell than the gene itself. In this regard, cDNA excess titration of fibroin mRNA is more sensitive than titration procedures based on the analysis of the specific pattern of RNase Tl digestion of fibroin mRNA (Suzuki and Brown, 1972; Suzuki and Suzuki, 1974).
Variations of P25 mRNA and fibroin mRNA content throughout development are represented on Figs. 7 and 8, respectively. Results revealed that the concentration of both mRNA followed a similar pattern. Particularly remarkable is the dramatic exponential-like loss of these sequences observed within the few hours following the physiological arrest of feeding. Thus, from 6 hr before to 10 hr after apolysis (when epidermal cells are freed from the old exoskeleton) 85% of P25 mRNA sequences disappeared from the cell. During the same time interval, fibroin mRNA content dropped by a factor of 400. Thirty-six hours later, at ecdysis, P25 mRNA concentration was 330 molecules per haploid genome. At this time, fibroin mRNA was not more abundant in the cell than the gene itself, since 0.0015% of total RNA corresponds to 1 to 2 molecules per haploid genome. As shown on Figs. 7 and 8, P25 mRNA and fibroin mRNA concentrations rose rapidly in the PSG cells after the animals had been refed following ecdysis. If the animals were not provided with food after molting, the low level of mRNA of both species was maintained, at least 22 hr postexuviation, emphasizing the importance of alimentary factors in the recovery of both mRNA. Accumulation of P25 and fibroin mRNA proceeded in such
404
DEVELOPMENTAL BIOLOGY
ID
P25
VOLUME 9’7, 1983
calculated above is significantly higher than the range of 18,000-36,000 molecules per haploid genome that we previously estimated at the same stage by means of complexity analysis of polyadenylated mRNA (Couble et al., 1981a,b). This discrepancy may be explained by the imprecision of the estimation based on kinetics analysis, but may also lie in the fact that part of the pool of P25 mRNA consists of nonadenylated sequences. As a matter of fact, since only 41% of fibroin mRNA does contain a poly(A) tail at the 3’ end at Day 6 of the fifth intermolt (Lizardi et ah, 1975) and given our observation that the relative poly(A) content of PSG cells decrease along the last intermolt from 0.01% at exuviation of 0.005% at Day 7 (data not shown), it may be suggested that the proportion of adenylated sequences decreases as development proceeds. DISCUSSION
4
In a previous paper, we demonstrated the existence of a major mRNA species in the PSG cells of bombys 4
a
b
FIG. 6. Peptide mapping analysis of P25 mRNA cell-free product and 25-kD silk proteins of the cocoon. 25-kD polypeptides were initially separated from bulk proteins by electrophoresis on a 15% acrylamide gel (1D). The selected proteins were then digested with V8 protease in a 3% stacking gel of a second electrophoresis and the cleaved peptides were separated in a 17.5% acrylamide resolving gel (2D). Lane a shows a partial peptide mapping of the 25-kD silk proteins of the cocoon extracted with neutral LiSCN (Coomassie brilliant blue staining). [sH&eucine-labeled P25 was obtained by cell-free translation of agarose gel purified P25 mRNA. P25 was run on the same gels as cocoon proteins and cleaved peptides were revealed by fluorography (lane b). 25-kD silk proteins of the cocoon were resolved as two major bands in the first dimension (a, top). The slower migrating component remained uncleaved after V8 protease treatment whereas the other generated two peptides (arrows) with migration and relative intensity identical to the spots obtained after proteolysis of labeled P25.
a way that their respective relative increment curves were almost strictly superimposable and that an equimolarity relationship was nearly maintained throughout the last intermolt. This increase culminated on Day 7 of the intermolt in a maximal concentration of about 50,000 molecules per haploid genome for both P25 and fibroin mRNA. The variations of fibroin mRNA content during the fifth intermolt are consistent with the ones reported by Suzuki and Suzuki (1974), using a chemical titration procedure. However, the concentration of P25 mRNA
. ./ L-4
-0 .’ /’ _____----
L ”
.
f
1
2
.
: /I ,I’ ,: : ,I’
,/ /’ I’/’ 3
4
5
6
7
7 IV’h molt
Vth
intermolt
{days
1
FIG. ‘7. Concentration changes of P25 mRNA in posterior silkgland cells at the end of the larval life. Experimental data are expressed in relative (0) and absolute (m) values. Each point resulted from the exploitation of the hybridization of an excess of purified P25 cDNA with increasing amounts of PSG total RNA of a given stage, under conditions detailed in Materials and Methods. The open sign corresponds to the titration in total RNA from animals that were not refed after exuviation. (fl) Physiological arrest of feeding, (A) apolysis,
(+I exuviation.
COUBLE
GA41234567 lVth
ET AL.
Low-Molecular-Weight
\
Silk mRNA
4
7
molt
Vth
intermolt
405
( days
1
.-c
a E .-C 0 ; c
- 0.0025
hours
FIG. 8. Variations of the concentration Fig. 7. Inset represents the fluctuations apolysis, (+) exuviation.
of fibroin of fibroin
mRNA mRNA
post
apolysis
4
in posterior silkgland cells at the end of the larval life. during molting on an enlarged scale. (0) physiological
mori. Its variable abundance in polyadenylated mRNA, studied at exuviation and Day 6 of the last intermolt, suggested a production coordinated with fibroin secretion (Couble et aZ., 1981a,b). However, the function of the corresponding protein was unknown. The present paper identifies this mRNA as encoding for a low-molecular-weight silk protein and describes the stoichiometric relationship between this mRNA and fibroin mRNA, during development. We found that purified cDNA of the first abundance class of nonfibroin mRNA hybridize to a single 1100 nucleotides long species on northern blots. This species,
Data are expressed arrest of feeding,
as in (A)
after extraction from agarose gel, yields a predominant peptide of 25 kD when translated in cell-free system. The demonstration that this peptide does arise from the mRNA detected previously by complexity analysis is further supported by the following results: (i) complexity measurements revealed that P25 mRNA is more abundant than any other mRNA, except fibroin mRNA, at the end of the fifth intermolt (Couble et ah, 1981a,b); thus, when translated in vitro, nonfibroin mRNA from this stage gave P25 as the major polypeptide. (ii) In contrast, we found that, at ecdysis, P25 mRNA is no more abundant than an estimated 38 other major mRNA
406
DEVELOPMENTAL
BIOLOGY
species; in this case, when the cell-free system was driven by molting stage mRNA, P25 was not more represented than a number of other proteins. The identity of the size, isoelectric point, and peptide mapping pattern of the cell-free translated product of P25 mRNA with that of a major polypeptide of the cocoon led to the conclusion that P25 is a silk protein. In agreement with the abundance of P25 mRNA at the stage of massive silk production, P25 was found as an abundant and rapidly in viva labelled protein in the PSG. The same polypeptide was also revealed as a substantial band in the MSG. However the cellular mRNA of the MSG yielded no detectable P25 when translated in vitro. Moreover, in the MSG, P25 is not labeled in vivo by short-term pulses but only after a 24-hr chase. This strongly suggests that P25 observed in the MSG is a luminal protein and supports the conclusion that P25 is specifically synthesized and secreted by PSG cells and that it is transported to the MSG lumen where it accumulates before spinning. The presence, besides P25, of a slower migrating component of 25 kD in the cocoon extracts has been revealed by peptide mapping. According to our results, no abundant mRNA, except P25 mRNA has so far been characterized that could be responsible for the synthesis of such a silk protein, in either the PSG or the MSG. We assume that it corresponds to a modified form of P25 in which the susceptibility of the V8 cleavage site is altered. A similar interpretation has been proposed to explain discrepancies in the peptide mapping pattern of two forms of myosin encoded by the same mRNA (Bandman et aL, 1982). The existence of a low-molecular-weight silk protein has already been recognized by Japanese authors. Shimura et al. (1976) and Gamo et al. (1977) described a socalled “small fibroin” of 25-kD molecular weight, as an abundant protein in the lumen of both the PSG and the MSG. Confirmation of the presence of such a protein in the cocoon was brought recently by Tokutake (1980) and Shimura et al. (1981), who recovered a 24- to 25-kD polypeptide with similar composition to the “small fibroin.” Thus, the size of this molecule, its supposed site of production, and its abundance relative to fibroin suggest that it indeed corresponds to the silk protein encoded by P25 mRNA. Furthermore, the linkage of this protein to fibroin by disulfide bonds (Shimura et aL, 1981) argues for an equimolar relationship between both proteins which would fit the 1:l ratio of P25 and fibroin mRNA that we observed during the last larval intermolt. The narrow parallelism of the cellular pools of P25 and fibroin mRNA implies coordinated regulatory mechanisms. Maekawa and Suzuki (1980) have shown that the cyclical variations of fibroin mRNA content depend on both variations of transcription of the gene
VOLUME
97, 1983
and variations of the stability of the mRNA. Transcription is stopped after apolysis and preexisting molecules are degraded whereas, during intermolt, transcription is active and the mRNA is stable. The breakdown of P25 mRNA at apolysis showing similar kinetics to that of fibroin mRNA suggests that the corresponding gene is no longer transcribed during molting. It remains, however, that the residual cell concentration of P25 mRNA is 300 times the number of copies of fibroin mRNA, at ecdysis. This may result from a higher intrinsic stability of P25 mRNA versus fibroin mRNA, but may be also the consequence of a residual transcription of P25 gene. Only in vivo measurements of P25 gene transcription would allow to decide. The extensive accumulation of P25 mRNA during the fifth intermolt necessitates an intense transcription of the gene and a high stability of the mRNA itself. If P25 mRNA is as stable as fibroin mRNA, the observed equimolarity relationship would imply a strict adjustment of the transcription rate of their gene. The control of expression of silk encoding genes is not yet understood. However, the variations of P25 and fibroin mRNA during the molting cycle suggest that the hormones-ecdysone (Calvez et ak, 1976) and juvenile hormone (Daillie, 1979)-controlling larval development, regulate the transcription of silk protein genes and the stability of their mRNA. Our observation that, after ecdysis, the accumulation of both mRNA is triggered by food uptake argues also for the existence of an alimentary signal. The interaction of these physiological factors in the synchronization of gene expression remains to be determined. For this purpose, we have cloned the P25 gene, the structural analysis of which is under progress. We would like to thank Dr. G. Bosquet and Dr. B. Calvez for their generous help concerning pl determination; Dr. R. Ouazana and Dr. J. J. Madjar for their advice and discussion on peptide mapping; Dr. P. Ledger for critically reading the manuscript; Dr. J. W. Beard for the gift of reverse transcriptase; Mrs. L. Fourets for typing; and Mrs. J. Gaillard for preparing the drawings. This work has been supported by CNRS Grant 8002. REFERENCES AUFFRAY, C., and ROUGEON, T. (1980). Purification of mouse immunoglobulin heavy chain messenger RNAs from total myeloma tumor RNA. Eur. J. B&hem. 107,303-314. BANDMAN, E., MATSUDA, R., and STROHMAN, R. C. (1982). Myosin heavy chains from two different adult fast-twitch muscles have different peptide maps but identical mRNAs. Cell 29, 645-650. BONNER, W. M., and LASKEY, K. A. (1974). A film detection method for tritium labelled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem 46, 83-88. BORDIER, C., and CRETTOL-JKRVINEN, A. (1979). Peptide mapping of heterologous protein samples. J. Biol. Chem 8, 2565-2567. CALVEZ, B., HIRN, M., and DE REGGI, M. (1976). Ecdysone changes in the haemolymph of the silkworms (Bombyx mori and Phylosamia
COUBLE ET AL.
Low-Molecular-
Cynthia) during larval and pupal development. FEBS I&t. 71, 5761. CLEVELAND, D. W., FISCHER, S. G., KIRSCHNER, M. W., and LAEMMLI, U. K. (197’7). Peptide mapping by limited proteolysis in sodium dodecyl sulfate and analysis by gel electrophoresis. J. BioL Chem. 252, 1102-1106. COUBLE,P., PRUDHOMME, J. C., and DAILLIE, J. (19’77). The biosynthesis of fibroin. III: Involvement of the Golgi apparatus in transport and secretion. Exp. Cell Res. 109, 139-150. COUBLE, P., GAREL, A., and PRUDHOMME, J. C. (1981a). Complexity and diversity of polyadenylated mRNA in the silkgland of Bombgx mori: Changes related to fibroin production. Dev. BioL 82,139-149. COUBLE, P., GAREL, A., and PRUDHOMME, J. C. (1981b). Expression du genbme de la cellule sericigene de Born&c mori au dernier stade larvaire. Reprod Nutr. D&velop. 21, 257-264. DAILLIE, J. (1979). Juvenile hormone modifies larvae and silkgland development in Born&x mori. Biochimie 61, 275-281. GAMO, T., INOKUCHI,T., and LAUFER, H. (1977). Polypeptides of fibroin and sericin secreted from the different sections of the silkgland in Bombyx
mwi.
Insect
Biochem.
7,285-295.
LAEMMLI, U. K., (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T1. Nature (l&&m) 227, 680-685. LIZARDI, P., WILLIAMSON, R., and BROWN, D. D. (1975). The size of fibroin messenger RNA and its polyadenylic acid content. Cell 4, 199-205. MAEKAWA, II., and SUZUKI, Y. (1980). Repeated turn-off and turn-on
Weight
Silk
of fibroin gene transcription byx
mori.
Den
407
mRNA
BioL
during silk gland development of Born-
78, 394-406.
MCMASTER, G. K., and CARMICHAEL, G. G. (1977). Analysis of single and double stranded nucleic acids on polyacrylamide and agarose gels by using glyoxal and a&dine orange. Proc. Nat. Acad Sci. USA 74, 4835-4838. O’FARRELL, P. (1975). High resolution two-dimensional electrophoresis of proteins J. BioL Chem. 250,4007-4021. SHIMURA, K., KIKUCHI, A., OHTOMO, K., KATAGATA, Y., and HYODO, A. (1976). Studies on silk fibroin on Bombyx mori. I. Fractionation of fibroin prepared from the posterior silkgland. .I. B&hem. 80, 693-702.
SHIMURA, K., KIKUCHI, A., KATAGATA, Y., and OHOTOMO, K. (1981). The occurrence of small component proteins in the cocoon fibroin of Bomtyx mori. J. Ser. Sci. Japan 51, 20-26. SUZUKI, Y., and BROWN, D. D. (1972). Isolation and identification of the messenger for silk fibroin from Born&x mwi. J. Mol. BioL 63, 409-429.
SUZUKI, Y., and SUZUKI, E. (1974). Quantitative measurements of fibroin messenger RNA synthesis in the posterior silkgland of normal and mutant Bombyx mcwi. J. Mol. BioL 88,393-407. THOMAS, P. S. (1980). Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc. Nat. Acad Sci. USA 779, 5201-5205. TOKUTAKE, S. (1980). Isolation of the smallest component of silk protein. Biochem. J. 187, 413-41’7.
BIOLOGY
97, 398-407
(1983)
Developmental Variations of a Nonfibroin mRNA of Bombyx mori Silkgland, Encoding for a Low-Molecular-Weight Silk Protein PIERRE D&oar&ment
de Biologie
COUBLE,
AGNBS
MOINE,
ANNIE
GAREL,
AND
G&&ale et Appliqutk, Labwratoire Associi au CNRS boulevard du 11 Novembre 1918 69622, Villeurbanne Received
January
8, 1992; accepted
in revised
fm
JEAN-CLAUDE
PRUDHOMME
No. 92, Universith Cedex, France January
Claude
Bernard
Lyon-I
.&‘s’,
7, 198.3
The characterization of a new silk protein mRNA (P25 mRNA) in posterior silkgland cells (PSG) and the developmental variations of its cell molecular concentration versus that of fibroin mRNA are described. A 80% pure P25 cDNA was obtained by class separation of total nonfibroin cDNA from PSG and used to identify the mRNA in blotted PSG mRNA as a single 1100 nucleotide long species. When purified from agarose gel and translated in a reticulocyte cell-free system, P25 mRNA yielded a 25-kD polypeptide (P25), identical to a 25-kD protein of the cocoon in terms of pIvalue and partial peptide mapping pattern. Moreover, this protein comigrated with an abundant polypeptide of the posterior silkgland (PSG) and of the middle silkgland (MSG). When tritiated leucine was injected in vivo, labeled P25 showed up in the PSG after a 2-hr pulse but appeared in the MSG only after 24 hr of labeling. Since MSG cells were found to be devoid of P25 mRNA, we concluded that P25 is exclusively synthesized in the PSG, that it accumulates in the MSG lumen and that it is spun out in the same way as fibroin. Specific probes were used to measure the concentrations of P25 mRNA and also fibroin mRNA in PSG total RNA by hybridization with an excess of cDNA. Both species are highly degraded in the few hours following the physiological arrest of feeding which precedes the fourth molting period. Their subsequent accumulation during the fifth intermolt is triggered by food uptake and proceeds in such a way that a constant 1:l molar ratio is maintained during the period of silk secretion. INTRODUCTION
contribute to the production of two silk proteins in the same cell, we studied the stoichiometric relationships between their cellular content at periods of both intense and inactive silk secretion. Comparative titrations during development showed that their cell molecular concentrations are equal, suggesting that the transcription of their genes as well as their stability are probably dependent on the same regulatory mechanisms.
B&X mori posterior silkgland (PSG) is highly specialized in the synthesis and the secretion of fibroin, the major protein of the cocoon. PSG mRNA complexity analysis has shown that the pool of polyadenylated mRNA varies drastically between the stage of maximum fibroin synthesis (mid-fifth intermolt) and the stage of no production of fibroin (larval molting) (Couble et a,!., 1981a,b). This concerns, in particular, two mRNA: the fibroin mRNA and another mRNA, the function of which is unknown. Interestingly, kinetics analysis indicated that both species are the most abundant sequences at mid-intermolt, whereas at ecdysis, their representation is lowered 100 to lO,OOO-fold. Variation of cellular fibroin mRNA concentration reflects the existence of a cyclical turn-on and turn-off of the transcription of the gene and a degradation, occurring at molting, of the mRNA accumulated during the preceding intermolt (Suzuki and Suzuki, 1974; Maekawa and Suzuki, 1980). However, no data were available to interpret the existence of the other abundant mRNA that could account for the synthesis of a well-known PSG protein. In this paper, we report the characterization of this mRNA (P25 mRNA) and its identification as a species encoding for a low-molecular-weight silk protein (P25) produced by the posterior (PSG) but not in the middle silkgland (MSG). As fibroin and P25 mRNA 0612-1606183 Copyright All rights
$3.06
0 1983 by Academic Press. Inc. of reproduction in any form reserved.
MATERIALS
AND
METHODS
Silkworms Bombgx mori larvae, hybrids of European strains 200 and 300, were reared in the laboratory with fresh mulberry leaves provided four times daily. Silkglands were studied at various stages of development from the end of the fourth intermolt up to Day 7 of the fifth and last larval intermolt. RNA Extraction
and Liquid
H@idixation
Total RNA was extracted from freshly dissected silkglands according to the LiCl-urea method of Auffray and Rougeon (1930). The purification of polyadenylated mRNA by oligo-dT cellulose chromatography, the separation of fibroin mRNA from bulk mRNA (nonfibroin mRNA) by sucrose gradient centrifugation, the synthesis of complementary cDNA, and the liquid hy398
COUBLE
ET AL.
Low-Molecular-
Weight
of Speci&
P25
cDNA
A purified P25 cDNA was obtained by fractionation of cDNA reverse transcribed from PSG nonfibroin mRNA extracted at Day 6 of the fifth intermolt. For this purpose, an excess of nonfibroin mRNA was incubated with homologous 3H- or 32P-labeled cDNA up to a R,t of 5 X 10m3. Hybridization was then stopped by adding 30 ml of 0.01 M phosphate buffer (pH 7.0) and the mixture was chromatographed on hydroxylapatite at 6’7°C. Single-stranded cDNA was eluted with 0.12 M phosphate buffer whereas the double-stranded P25 cDNA was recovered by washing with 0.5 M phosphate buffer. The procedure was repeated twice and yielded 80 to 85% pure P25 cDNA as tested by hybridization with an excess of nonfibroin mRNA (see Fig. 2). P25
mRNA
and Fibroin
mRNA
I a
bridization of cDNA with an excess of polyadenylated mRNA were carried out as previously described (Couble et aZ., 1981a,b). Purification
Silk
mRNA
Titrations
Titration of specific sequences was performed by hybridizing PSG total RNA with an excess of tritiated cDNA. For determination of the content of fibroin mRNA, we used cDNA from sucrose gradient-purified mRNA: 0.5 ng of cDNA was incubated with increasing amounts of heat-denatured total RNA for 5 hr at 65°C in 50 ~1 containing 10 mM Tris-Cl (pH 7.5), 0.2 mM EDTA, and 0.4 M NaCl in order to attain saturation. The mixture was then diluted 10 times with 30 mM sodium acetate (pH 4.5), 0.1 M NaCl, and 0.12 mM ZnSO(. Half of the volume was TCA precipitated and the other half treated with 1000 units/ml of Sl nuclease (Miles laboratories) for 1 hr at 3’7°C before precipitation. Precipitates were collected on nitrocellulose filters (Millipore) and the percentage of hybrids resistant to nuclease treatment was determined. The fibroin mRNA content was obtained by comparing the slope of each experimental curve with that of a reference curve obtained by annealing, under the same conditions, known amounts of purified fibroin mRNA with an excess of specific cDNA (Fig. lb). Titration of P25 mRNA was carried out similarly. However, 0.6 ng of class-purified P25 cDNA, 600 nucleotides long, was used in a final volume of 20 ~1; also, we found it necessary to continue the incubation for 18 hr in order to complete the reaction. The standard used to quantify P25 mRNA in the RNA to be tested was nonfibroin mRNA extracted at Day 6 of the fifth intermolt and containing 29% of P25 mRNA as determined by complexity analysis (see Fig. la).
bp
0:2 ng
of
P25
pg
of
V6
pg
I
in
total
of
10hr
0:s
non-fibroin
RNA
mRNA
”
5
(a)
(0) 0:l
0:05 0.5
h
d
014
mRNA
1
1.5
post
ecdyiis
total
RNA
IA)
b
0 a ,N .E502 c” ;; 258
ng
of
fibroin
ng
of
V6
pg
of’lOhr
mRNA total
RNA
post
(0)
to?al
RNA
(3
FIG. 1. Titration of P25 and fibroin mRNA. Standard curves of cDNA excess hybridization experiments are represented together with two experimental curves, given as examples. a Shows the hybridization of various amounts of nonfibroin mRNA known to contain 29% of P25 mRNA (O), total RNA from 10 hr postapolysis (A), and from Day 6 of the fifth intermolt (0) with 0.6 ng of purified tritiated P25 cDNA. b Shows the hybridization of 0.5 ng of tritiated fibroin cDNA when incubated with increasing amounts of pure fibroin mRNA (O), 10 hr postapolysis (A) and 6 days ecdysis (0) total RNA. Background values were not subtracted.
mRNA
Electrophoresis
and Hybridizations
“Northern” blots were performed according to the method of Thomas (1980). Briefly, mRNA were denatured in 1 M glyoxal and 50% dimethyl sulfoxide for 1 hr at 5O”C, then electrophoresed on horizontal 1.5% agarose gel and transferred by blotting onto 0.45-pm nitrocellulose filters (Schleicher and Schull). Hybridizations were carried out with 32P-labeled cDNA in 50% formamide, 0.75 M NaCl, 0.075 M sodium citrate, 50 mM
DEVELOPMENTALBIOLOGY VOLUME 97,1983
400
phosphate, 0.02% each of polyvinylpyrolidone, ficoll, and bovine serum albumin, 200 pg/ml of herring sperm DNA, for 20 hr at 42°C. After washing, the blots were exposed to Kodak X-ray films. P25 mRNA
PuriJicaticm
Urea-denatured PSG nonfibroin mRNA was electrophoresed in 1.5% agarose and the position of P25 mRNA located with reference markers (N&d111 restricted fragments of X-phage DNA). P25 mRNA as well as mRNA of lower and higher molecular weight were then eluted in wells by means of a second electrophoresis, perpendicularly to the first direction, and ethanol precipitated. Analysis
of Cell-Free
Translation
Products
Micrococcal nuclease-treated rabbit reticulocyte lysate (Amersham) was supplemented with 1 pCi/pl of tritiated leucine (123 Ci/mmole) or serine (28 Ci/mmole) and used at the rate of 10 ~1 per assay: 0.5 to 1 pg of various mRNA was tested and incubation was carried out at 37°C for 30 min. The mixture was then brought to 4% SDS and 1% @mercaptoethanol, heated at 100°C for 3 min, and electrophoresed in a linear 3.75-15% gradient of acrylamide with a 3% acrylamide stacking gel, according to Laemmli (1970). 14C-labeled protein markers (Amersham) were coelectrophoresed for calibration of the migration. Gels were treated for fluorography according to Bonner and Laskey (1977) and autoradiographed on X-ray Kodak films. In Vivo Labeling
and Analysis
of Silk Proteins
Silkworms received 20 &i of [3H]leucine (123 Ci/ mmole) and the total proteins extracted from PSG and MSG were analyzed after 2 hr a labeling pulse that allows silk proteins to be secreted into the lumen of the organ (Couble et aL, 1977), and after a 24-hr chase so as to study PSG silk protein accumulation in the MSG. Soluble proteins were recovered from posterior and middle sections of freshly dissected silkglands, after rinsing the glands several times in distilled water to remove hemolymph contaminants, and incubating small pieces in 200 to 500 ~1 of distilled water. Silk proteins were extracted from the cocoon by either 60% (w/w) neutral LiSCN or a 4% SDS and 7 M urea mixture. ,. Proteins were treated with fl-mercaptoethanol prior to electrophoresis on a 3.75-15% linear gradient of acrylamide as described for cell-free synthesized proteins.
Peptide
Mapping
Proteins were mapped by limited proteolysis, in the presence of SDS, with V8 protease from Staphylococcus aureus, according to Cleveland et al. (1977) and to Bordier and Crettol-Jarvinen (1979). LiSCN-extracted cocoon proteins (100-200 pg) and [3H]leucine-labeled P25 from in vitro translation of purified P25 mRNA were electrophoresed on a 15% acrylamide, 2-mm-thick slab gel according to Laemmli (1970). Gel regions with 25kD proteins were excised and incubated in 125 mM TrisCl (pH 6.8) and 0.1% SDS for 1 hr at room temperature. Gel pieces were then mounted between two casting glass plates prior to pouring a 17.5% acrylamide, 2-mm-thick resolving gel, with a 3% stacking gel. Stacking gel covered the first dimension gel lanes by 1 mm. The gel was placed in the electrophoresis apparatus and overlayed with 1.5 ml of sample buffer containing 4 to 16 pg of V8 protease (Miles laboratories). Following costacking of protease and silk proteins, the electrophoresis was interrupted for 40 min, and then migration was completed. Cocoon proteins and 3H-labeled P25 maps were revealed, respectively, by Coomassie brilliant blue staining and fluorography. RESULTS
P25 mRNA
Probe
PSG cells of Bom&r mori contain a major mRNA species, besides fibroin mRNA, the existence of which was deduced from hybridization kinetics of an excess of nonfibroin mRNA with homologous cDNA (Couble et al, 1981a,b). Since this mRNA was estimated to represent 30% of nonfibroin mRNA and defined a major class of complexity, the purification of the corresponding cDNA was carried out through class separation of total nonfibroin cDNA (see Materials and Methods). The resulting probe, back hybridized with an excess of nonfibroin mRNA, exhibited 75 to 85% of class-specific cDNA as shown on Fig. 2. This probe was used for sizing and titrating the mRNA under study that we named P25 mRNA according to the size of the encoded protein (see below). P25 mRNA
Sizing
P25 mRNA was localized on Northern blots of total RNA (Fig. 3a) and nonfibroin mRNA (Fig. 3b) by hybridization with purified 32P-labeled P25 cDNA. In both cases, P25 cDNA annealed to a single species; its size was estimated at 1100 nucleotides using glyoxaldenatured Hind111 restricted fragments of X-phage DNA as markers according to MC Master and Carmichael (1977). As a control, total cDNA recovered as single-stranded
COUBLE ET
Low-Molecular-
AL.
Silk
401
mRNA
sized in the cell-free system driven by the agarose purified mRNA of lower and higher molecular weight than that of P25 mRNA (Figs. 4b and c). As expected, P25 was found as the most labeled polypeptide when nonfibroin mRNA from PSG were translated in vitro (Fig. 4d). It was also detected, but at much lower relative intensity when polyadenylated mRNA from molting stage were cell-free translated (compare Figs. 4d and e), as predicted from the lower relative abundance of P25 mRNA at this period (Couble et ah, 1981b). Similar experiments carried out with tritiated serine instead of leucine led to the same conclusions (data not shown), suggesting that the relative P25 signal intensity in both cases accounted for the abundance of the protein.
80
E o .c ro .-N ‘*-0 i
Weight
60
= 4c s
Identi,fication of P25 as a Silk Protein
2c
The existence of a major polypeptide band of 25 kD in the total protein extracts of the posterior (Fig. 5a) and middle (Fig. 5d) silkgland and in the cocoon silk (Figs. 5g and h) suggested that P25 could be a silk pro-
1I lo-’
lo-’
1o-z
R,
lo-’
10”
10’
x t
b
C
FIG. 2. Control of P25 cDNA and P25 mRNA purification procedures. Class-purified P25 cDNA (0) and cDNA reverse transcribed from agarose gel purified P25 mRNA (0) were hybridized with an excess of polyadenylated non-fibroin mRNA. The reference curve (- - -) corresponds to the kinetics of hybridization of an excess of nonfibroin mRNA with homologous cDNA. The percentage of hybridization is plotted against log R& the product of the concentration of mRNA in moles of nucleotides z 1-l z time (set). Symbols represent experimental data without self-annealing corrections.
species after class separation of P25 cDNA was hybridized with nonfibroin mRNA (Fig. 3~). As expected, a number of mRNA showed up, including P25 mRNA, since the probe was not completely devoid of P25 cDNA. Characterization
P25
m RNA
of P25 mRNA Translation Product
A rabbit reticulocyte lysate was used as a cell-free translation system to study the proteins encoded by various preparations of mRNA. The denatured leucineor serine-labeled proteins were analyzed by electrophoresis on a 3.75-1570 linear gradient of acrylamide. P25 mRNA was purified from agarose gel taking advantage of its abundance in polyadenylated nonfibroin mRNA. Hybridization kinetics of an excess of nonfibroin mRNA with the cDNA transcribed from purified P25 mRNA showed that the preparation contained a minimum of 55% of the specific sequence (Fig. 2). When polypeptide of 25 kD translated in vitro, a predominant (=P25) was obtained (Fig. 4a). Such a protein was found in reduced amounts among the polypeptides synthe-
FIG. 3. Characterization of P25 mRNA in PSG RNA separated on agarose gel. P25 mRNA was revealed as a single species on the following Northern blots of RNA: lanes a and b are, respectively, total RNA (10 pg) and nonfibroin mRNA (1 pg) separated on a 1.5% agarose gel and hybridized with class purified 3zP-P25 cDNA; lane c corresponds to nonfibroin mRNA, as in lane b, hybridized with BP total nonfibroin cDNA recovered as single-stranded cDNA after class separation of P25 cDNA.
DEVELOPMENTAL BIOLOGY
402
a
b
c
VOLUME 97,1983
d
e
f
FIG. 4. In vitro characterization of the P25 mRNA encoded protein in cell-free products of various mRNA preparations. Autoradiograms correspond to the cell-free [BH]Ieucine-labeIed proteins translated from agarose purified P25 mRNA (a), PSG nonfibroin mRNA of higher (b) and lower (c) molecular weight than P25 mRNA, PSG nontibroin mRNA from Day 6 of the fifth intermolt (d), total mRNA from PSG at exuviation (e) and from MSG at mid-last intermolt (f). Proteins were denatured in SDS and @-mercaptoethanol prior to electrophoresis in a 3.75-15% linear gradient of acrylamide. Size of marker proteins are indicated in kD.
tein assembled and secreted in the F’SG, stored in the MSG lumen, and spun, like fibroin itself. Since these proteins detected in viva could not be identified as P25 solely by virtue of their apparent size, we further compared them in terms of isoelectric point and partial peptide mapping pattern. First, double-dimension electrophoresis according to O’Farrell (1975) indicated a p1 value of 4.8-4.9 for both the in vitro synthesized P25 and a 25-kD protein of the cocoon (data not shown). In a second experiment, the in vitro and in vivo 25kD proteins, initially separated on acrylamide gel, were digested by Staphylococcus V8 protease and the resulting peptides were analyzed by second electrophoresis. The results presented on Fig. 6 show that the partial digestion of labeled P25 generated two peptides that comigrated with two proteolytic products arising from the cleavage of one 25-kD silk protein of the cocoon. In addition, a second, slower migrating 25-kD silk protein was detected as an uncleaved polypeptide after V8 protease digestion. The characteristics of size, pIvalue, and peptide mapping pattern that P25 has in common with a cocoon
protein led to the conclusion that P25 mRNA does indeed encode for a silk protein. Some data indicated that P25 production is specific of PSG cells. Interestingly, the presence of an abundant 25-kD protein in MSG protein extracts could not be explained by the presence of a corresponding mRNA since polyadenylated mRNA from this region of the organ failed to produce any 25-kD polypeptides when translated in vitro (Fig. 4f). After 2 hr of in vivo labeling with tritiated leucine, P25 was intensely labeled in the PSG (Fig. 5b) but remained unlabeled in the MSG (Fig. 5e). Following a 24-hr chase, the relative radioactivity of P25 was decreased in the PSG (Fig. 5c), whereas tritiated 25-kD protein from MSG showed up as an intense band (Fig. 5f). We therefore conclude that P25 is probably synthesized exclusively in the PSG cells and that the 25-kD protein in the MSG represents the luminal P25 accumulated in the middle silkgland before spinning. P25 mRiVA Variations Throughout Development
The cellular concentration of P25 mRNA in the PSG was determined and compared to the variations of fi-
COUBLE
a
b
ET AL.
i&w-Molecular-
d
c
Weight
e
Silk
mRNA
403
f
r S L
P25-
FIG. 5. In viva detection of cold and labeled P25 in total protein extracts from PSG, MSG, and cocoon. A 25-kD abundant protein is present in the Coomassie brilliant blue-stained proteins of PSG (a) and MSG (d) and in the silk proteins of the cocoon extracted either by neutral LiSCN (g) or SDS-urea (h). Lanes b and c are autoradiograms of in viva [3H]leucine-labeled proteins from PSG at, respectively, 2 hr (b) and 24 hr (c) after injection of the precursor. Similarly, short-and long-term labelings of MSG proteins are, respectively, shown in lanes e and f. In all cases, proteins were treated with SDS and P-mercaptoethanol and analyzed on a 3.75-15% linear gradient of acrylamide. Marker sizes are indicated in kD. f, fibroin; s, sericin.
broin mRNA content. This was done at different stages occurring at the end of the larval life, including the last day of the fourth intermolt followed by the fourth molting stage and the first ‘7 days of the fifth and last intermolt (cocoon spinning started on Day 8). For this purpose, cDNA excess experiments were performed using class-purified [3H]P25 cDNA and [3H]fibroin cDNA obtained by reverse transcription of the specific mRNA prepared by sucrose gradient centrifugation (Couble et aL, 1981). Standard amounts of the cDNA were incubated with increasing amounts of total RNA from PSG cells at different developmental stages, under experimental conditions that allowed hybridization saturation at each RNA concentration. The concentration of each sequence was deduced from the comparison of the slope of the experimental curves with that of reference curves (Fig. 1). We emphasize the high degree of sensitivity of this method, since it allows detection of sequences no more abundant in the cell than the gene itself. In this regard, cDNA excess titration of fibroin mRNA is more sensitive than titration procedures based on the analysis of the specific pattern of RNase Tl digestion of fibroin mRNA (Suzuki and Brown, 1972; Suzuki and Suzuki, 1974).
Variations of P25 mRNA and fibroin mRNA content throughout development are represented on Figs. 7 and 8, respectively. Results revealed that the concentration of both mRNA followed a similar pattern. Particularly remarkable is the dramatic exponential-like loss of these sequences observed within the few hours following the physiological arrest of feeding. Thus, from 6 hr before to 10 hr after apolysis (when epidermal cells are freed from the old exoskeleton) 85% of P25 mRNA sequences disappeared from the cell. During the same time interval, fibroin mRNA content dropped by a factor of 400. Thirty-six hours later, at ecdysis, P25 mRNA concentration was 330 molecules per haploid genome. At this time, fibroin mRNA was not more abundant in the cell than the gene itself, since 0.0015% of total RNA corresponds to 1 to 2 molecules per haploid genome. As shown on Figs. 7 and 8, P25 mRNA and fibroin mRNA concentrations rose rapidly in the PSG cells after the animals had been refed following ecdysis. If the animals were not provided with food after molting, the low level of mRNA of both species was maintained, at least 22 hr postexuviation, emphasizing the importance of alimentary factors in the recovery of both mRNA. Accumulation of P25 and fibroin mRNA proceeded in such
404
DEVELOPMENTAL BIOLOGY
ID
P25
VOLUME 9’7, 1983
calculated above is significantly higher than the range of 18,000-36,000 molecules per haploid genome that we previously estimated at the same stage by means of complexity analysis of polyadenylated mRNA (Couble et al., 1981a,b). This discrepancy may be explained by the imprecision of the estimation based on kinetics analysis, but may also lie in the fact that part of the pool of P25 mRNA consists of nonadenylated sequences. As a matter of fact, since only 41% of fibroin mRNA does contain a poly(A) tail at the 3’ end at Day 6 of the fifth intermolt (Lizardi et ah, 1975) and given our observation that the relative poly(A) content of PSG cells decrease along the last intermolt from 0.01% at exuviation of 0.005% at Day 7 (data not shown), it may be suggested that the proportion of adenylated sequences decreases as development proceeds. DISCUSSION
4
In a previous paper, we demonstrated the existence of a major mRNA species in the PSG cells of bombys 4
a
b
FIG. 6. Peptide mapping analysis of P25 mRNA cell-free product and 25-kD silk proteins of the cocoon. 25-kD polypeptides were initially separated from bulk proteins by electrophoresis on a 15% acrylamide gel (1D). The selected proteins were then digested with V8 protease in a 3% stacking gel of a second electrophoresis and the cleaved peptides were separated in a 17.5% acrylamide resolving gel (2D). Lane a shows a partial peptide mapping of the 25-kD silk proteins of the cocoon extracted with neutral LiSCN (Coomassie brilliant blue staining). [sH&eucine-labeled P25 was obtained by cell-free translation of agarose gel purified P25 mRNA. P25 was run on the same gels as cocoon proteins and cleaved peptides were revealed by fluorography (lane b). 25-kD silk proteins of the cocoon were resolved as two major bands in the first dimension (a, top). The slower migrating component remained uncleaved after V8 protease treatment whereas the other generated two peptides (arrows) with migration and relative intensity identical to the spots obtained after proteolysis of labeled P25.
a way that their respective relative increment curves were almost strictly superimposable and that an equimolarity relationship was nearly maintained throughout the last intermolt. This increase culminated on Day 7 of the intermolt in a maximal concentration of about 50,000 molecules per haploid genome for both P25 and fibroin mRNA. The variations of fibroin mRNA content during the fifth intermolt are consistent with the ones reported by Suzuki and Suzuki (1974), using a chemical titration procedure. However, the concentration of P25 mRNA
. ./ L-4
-0 .’ /’ _____----
L ”
.
f
1
2
.
: /I ,I’ ,: : ,I’
,/ /’ I’/’ 3
4
5
6
7
7 IV’h molt
Vth
intermolt
{days
1
FIG. ‘7. Concentration changes of P25 mRNA in posterior silkgland cells at the end of the larval life. Experimental data are expressed in relative (0) and absolute (m) values. Each point resulted from the exploitation of the hybridization of an excess of purified P25 cDNA with increasing amounts of PSG total RNA of a given stage, under conditions detailed in Materials and Methods. The open sign corresponds to the titration in total RNA from animals that were not refed after exuviation. (fl) Physiological arrest of feeding, (A) apolysis,
(+I exuviation.
COUBLE
GA41234567 lVth
ET AL.
Low-Molecular-Weight
\
Silk mRNA
4
7
molt
Vth
intermolt
405
( days
1
.-c
a E .-C 0 ; c
- 0.0025
hours
FIG. 8. Variations of the concentration Fig. 7. Inset represents the fluctuations apolysis, (+) exuviation.
of fibroin of fibroin
mRNA mRNA
post
apolysis
4
in posterior silkgland cells at the end of the larval life. during molting on an enlarged scale. (0) physiological
mori. Its variable abundance in polyadenylated mRNA, studied at exuviation and Day 6 of the last intermolt, suggested a production coordinated with fibroin secretion (Couble et aZ., 1981a,b). However, the function of the corresponding protein was unknown. The present paper identifies this mRNA as encoding for a low-molecular-weight silk protein and describes the stoichiometric relationship between this mRNA and fibroin mRNA, during development. We found that purified cDNA of the first abundance class of nonfibroin mRNA hybridize to a single 1100 nucleotides long species on northern blots. This species,
Data are expressed arrest of feeding,
as in (A)
after extraction from agarose gel, yields a predominant peptide of 25 kD when translated in cell-free system. The demonstration that this peptide does arise from the mRNA detected previously by complexity analysis is further supported by the following results: (i) complexity measurements revealed that P25 mRNA is more abundant than any other mRNA, except fibroin mRNA, at the end of the fifth intermolt (Couble et ah, 1981a,b); thus, when translated in vitro, nonfibroin mRNA from this stage gave P25 as the major polypeptide. (ii) In contrast, we found that, at ecdysis, P25 mRNA is no more abundant than an estimated 38 other major mRNA
406
DEVELOPMENTAL
BIOLOGY
species; in this case, when the cell-free system was driven by molting stage mRNA, P25 was not more represented than a number of other proteins. The identity of the size, isoelectric point, and peptide mapping pattern of the cell-free translated product of P25 mRNA with that of a major polypeptide of the cocoon led to the conclusion that P25 is a silk protein. In agreement with the abundance of P25 mRNA at the stage of massive silk production, P25 was found as an abundant and rapidly in viva labelled protein in the PSG. The same polypeptide was also revealed as a substantial band in the MSG. However the cellular mRNA of the MSG yielded no detectable P25 when translated in vitro. Moreover, in the MSG, P25 is not labeled in vivo by short-term pulses but only after a 24-hr chase. This strongly suggests that P25 observed in the MSG is a luminal protein and supports the conclusion that P25 is specifically synthesized and secreted by PSG cells and that it is transported to the MSG lumen where it accumulates before spinning. The presence, besides P25, of a slower migrating component of 25 kD in the cocoon extracts has been revealed by peptide mapping. According to our results, no abundant mRNA, except P25 mRNA has so far been characterized that could be responsible for the synthesis of such a silk protein, in either the PSG or the MSG. We assume that it corresponds to a modified form of P25 in which the susceptibility of the V8 cleavage site is altered. A similar interpretation has been proposed to explain discrepancies in the peptide mapping pattern of two forms of myosin encoded by the same mRNA (Bandman et aL, 1982). The existence of a low-molecular-weight silk protein has already been recognized by Japanese authors. Shimura et al. (1976) and Gamo et al. (1977) described a socalled “small fibroin” of 25-kD molecular weight, as an abundant protein in the lumen of both the PSG and the MSG. Confirmation of the presence of such a protein in the cocoon was brought recently by Tokutake (1980) and Shimura et al. (1981), who recovered a 24- to 25-kD polypeptide with similar composition to the “small fibroin.” Thus, the size of this molecule, its supposed site of production, and its abundance relative to fibroin suggest that it indeed corresponds to the silk protein encoded by P25 mRNA. Furthermore, the linkage of this protein to fibroin by disulfide bonds (Shimura et aL, 1981) argues for an equimolar relationship between both proteins which would fit the 1:l ratio of P25 and fibroin mRNA that we observed during the last larval intermolt. The narrow parallelism of the cellular pools of P25 and fibroin mRNA implies coordinated regulatory mechanisms. Maekawa and Suzuki (1980) have shown that the cyclical variations of fibroin mRNA content depend on both variations of transcription of the gene
VOLUME
97, 1983
and variations of the stability of the mRNA. Transcription is stopped after apolysis and preexisting molecules are degraded whereas, during intermolt, transcription is active and the mRNA is stable. The breakdown of P25 mRNA at apolysis showing similar kinetics to that of fibroin mRNA suggests that the corresponding gene is no longer transcribed during molting. It remains, however, that the residual cell concentration of P25 mRNA is 300 times the number of copies of fibroin mRNA, at ecdysis. This may result from a higher intrinsic stability of P25 mRNA versus fibroin mRNA, but may be also the consequence of a residual transcription of P25 gene. Only in vivo measurements of P25 gene transcription would allow to decide. The extensive accumulation of P25 mRNA during the fifth intermolt necessitates an intense transcription of the gene and a high stability of the mRNA itself. If P25 mRNA is as stable as fibroin mRNA, the observed equimolarity relationship would imply a strict adjustment of the transcription rate of their gene. The control of expression of silk encoding genes is not yet understood. However, the variations of P25 and fibroin mRNA during the molting cycle suggest that the hormones-ecdysone (Calvez et ak, 1976) and juvenile hormone (Daillie, 1979)-controlling larval development, regulate the transcription of silk protein genes and the stability of their mRNA. Our observation that, after ecdysis, the accumulation of both mRNA is triggered by food uptake argues also for the existence of an alimentary signal. The interaction of these physiological factors in the synchronization of gene expression remains to be determined. For this purpose, we have cloned the P25 gene, the structural analysis of which is under progress. We would like to thank Dr. G. Bosquet and Dr. B. Calvez for their generous help concerning pl determination; Dr. R. Ouazana and Dr. J. J. Madjar for their advice and discussion on peptide mapping; Dr. P. Ledger for critically reading the manuscript; Dr. J. W. Beard for the gift of reverse transcriptase; Mrs. L. Fourets for typing; and Mrs. J. Gaillard for preparing the drawings. This work has been supported by CNRS Grant 8002. REFERENCES AUFFRAY, C., and ROUGEON, T. (1980). Purification of mouse immunoglobulin heavy chain messenger RNAs from total myeloma tumor RNA. Eur. J. B&hem. 107,303-314. BANDMAN, E., MATSUDA, R., and STROHMAN, R. C. (1982). Myosin heavy chains from two different adult fast-twitch muscles have different peptide maps but identical mRNAs. Cell 29, 645-650. BONNER, W. M., and LASKEY, K. A. (1974). A film detection method for tritium labelled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem 46, 83-88. BORDIER, C., and CRETTOL-JKRVINEN, A. (1979). Peptide mapping of heterologous protein samples. J. Biol. Chem 8, 2565-2567. CALVEZ, B., HIRN, M., and DE REGGI, M. (1976). Ecdysone changes in the haemolymph of the silkworms (Bombyx mori and Phylosamia
COUBLE ET AL.
Low-Molecular-
Cynthia) during larval and pupal development. FEBS I&t. 71, 5761. CLEVELAND, D. W., FISCHER, S. G., KIRSCHNER, M. W., and LAEMMLI, U. K. (197’7). Peptide mapping by limited proteolysis in sodium dodecyl sulfate and analysis by gel electrophoresis. J. BioL Chem. 252, 1102-1106. COUBLE,P., PRUDHOMME, J. C., and DAILLIE, J. (19’77). The biosynthesis of fibroin. III: Involvement of the Golgi apparatus in transport and secretion. Exp. Cell Res. 109, 139-150. COUBLE, P., GAREL, A., and PRUDHOMME, J. C. (1981a). Complexity and diversity of polyadenylated mRNA in the silkgland of Bombgx mori: Changes related to fibroin production. Dev. BioL 82,139-149. COUBLE, P., GAREL, A., and PRUDHOMME, J. C. (1981b). Expression du genbme de la cellule sericigene de Born&c mori au dernier stade larvaire. Reprod Nutr. D&velop. 21, 257-264. DAILLIE, J. (1979). Juvenile hormone modifies larvae and silkgland development in Born&x mori. Biochimie 61, 275-281. GAMO, T., INOKUCHI,T., and LAUFER, H. (1977). Polypeptides of fibroin and sericin secreted from the different sections of the silkgland in Bombyx
mwi.
Insect
Biochem.
7,285-295.
LAEMMLI, U. K., (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T1. Nature (l&&m) 227, 680-685. LIZARDI, P., WILLIAMSON, R., and BROWN, D. D. (1975). The size of fibroin messenger RNA and its polyadenylic acid content. Cell 4, 199-205. MAEKAWA, II., and SUZUKI, Y. (1980). Repeated turn-off and turn-on
Weight
Silk
of fibroin gene transcription byx
mori.
Den
407
mRNA
BioL
during silk gland development of Born-
78, 394-406.
MCMASTER, G. K., and CARMICHAEL, G. G. (1977). Analysis of single and double stranded nucleic acids on polyacrylamide and agarose gels by using glyoxal and a&dine orange. Proc. Nat. Acad Sci. USA 74, 4835-4838. O’FARRELL, P. (1975). High resolution two-dimensional electrophoresis of proteins J. BioL Chem. 250,4007-4021. SHIMURA, K., KIKUCHI, A., OHTOMO, K., KATAGATA, Y., and HYODO, A. (1976). Studies on silk fibroin on Bombyx mori. I. Fractionation of fibroin prepared from the posterior silkgland. .I. B&hem. 80, 693-702.
SHIMURA, K., KIKUCHI, A., KATAGATA, Y., and OHOTOMO, K. (1981). The occurrence of small component proteins in the cocoon fibroin of Bomtyx mori. J. Ser. Sci. Japan 51, 20-26. SUZUKI, Y., and BROWN, D. D. (1972). Isolation and identification of the messenger for silk fibroin from Born&x mwi. J. Mol. BioL 63, 409-429.
SUZUKI, Y., and SUZUKI, E. (1974). Quantitative measurements of fibroin messenger RNA synthesis in the posterior silkgland of normal and mutant Bombyx mcwi. J. Mol. BioL 88,393-407. THOMAS, P. S. (1980). Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc. Nat. Acad Sci. USA 779, 5201-5205. TOKUTAKE, S. (1980). Isolation of the smallest component of silk protein. Biochem. J. 187, 413-41’7.