Downloaded from symposium.cshlp.

org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

The Effect of Heat Shock on Gene Expression in

Drosophila melanogaster
M.-E. MIRAULT, M. GOLDSCHMIDT-CLERMONT,L. MORAN, A. P. ARRIGO, AND A. TISSI~RES
Ddpartement de Biologic Mol~culaire, Universit~ de Gen~ve, 1211 Geneva, Switzerland

When Drosophila melanogasteris exposed to 37~ RESULTS

a series of specific genes is activated, whereas most Synthesis of Heat-shock-induced Proteins
of the other genes, active at 25~ before this heat
shock, appear to be repressed. Tissue-culture cells or salivary glands were pulse-
The induction of about eight or nine new puffs labeled with [ssS]methionine, beth at 25~ and after
on the salivary gland chromosomes occurs very a heat shock at 37~ the left panel in Figure 1 shows
shortly after the temperature shift to 37~ and at the electrophoretic pattern of the labeled proteins.
the same time most of the puffs active at 25~ before A small number of new protein bands appear after
the heat shock rapidly regress at 37~ (Ritossa 1962; heat shock, but the synthesis of most of the proteins
Ashburner 1970). The same new puffs are induced synthesized at 25~ is strongly reduced, as observed
under a variety of other stress conditions unrelated earlier (Tissi~res et al. 1974; Lewis et al. 1975;
to temperature (Berendes 1972; Lewis et al. 1975). McKenzie et al. 1975). The overall rate of incorpora-
The heat shock also induces the rapid synthesis of tion of [35S]methionine is roughly the same at the
a small number of proteins, whereas the rate of syn- two temperatures. Salivary gland and tissue-culture
thesis of most cellular proteins, normally made at cell labeling patterns after heat shock are quali-
25~ is strongly reduced (Tissi~res et al. 1974; Lewis tatively similar (see Fig. 1), and similar patterns
et al. 1975; McKenzie et al. 1975). This phenomenon are observed if [SH]leucine is substituted for
is observed in different tissues (Tissi~res et al. 1974), [35S]methionine (data not shown).
in m a n y different wild-type strains (L. Moran and The apparent molecular weights of the heat-shock
A. P. Arrigo, unpubl.), and in different Drosophila pelypeptides, as given in Figure 1, were determined
rnelanogaster tissue-culture cell lines (McKenzie et from their migration in sodium dodecyl sulfate
al. 1975; Moran et al. 1977). These observations sug- (SDS)-polyacrylamide gels of different concentra-
gest that new messenger RNAs, ceding for the heat- tions, calibrated with protein standards of known
shock proteins, are synthesized at the high tempera- molecular weight. The molecular weights of several
ture, whereas most preexisting messenger RNAs, heat-shock polypeptides were dependent on gel con-
made at 25~ are no longer translated. This is fur- centrations, which could be due to chemical modifi-
ther suggested by the rapid disappearance of preex- cations. Thus the molecular weights should be con-
isting polyribosomes in cells shifted to high tempera- sidered as nominal values.
ture, followed by the appearance of new and larger Kinetics of synthesis of heat-shock proteins. A pulse-
heat-shock polyribosomes (McKenzie et al. 1975). labeling with [35S]methionine of about 10 minutes
New species of RNA are synthesized at the high from the start of the heat shock was necessary to
temperature and they were shown to hybridize in detect the heat-shock proteins in tissue-culture cells.
situ at heat-shock puff sites (McKenzie et al. 1975; Within this time, synthesis of all heat-shock proteins
Spradling et al. 1975, 1977). is initiated. We could not observe a clear-cut se-
Here we report the analysis and characterization quence of appearance in contrast to the observation
of the heat-shock-induced polypeptides and messen- of Lewis et al. (1975) with Drosophila hydei, al-
ger RNAs from tissue-culture cells. The heat-shock- though our experiments cannot rule out an asyn-
induced messenger RNAs are shown to direct the chrony of a few minutes.
synthesis of the heat-shock-induced polypeptides. The m a x i m u m rate of synthesis of the heat-shock
Moreover, evidence is presented to show that mes- proteins in tissue-culture cells is reached within 90-
senger RNAs, normally synthesized and translated 120 minutes and subsequently declines to about 50%
at 25~ are still present in the cytoplasm after heat of its initial rate after 6 - 8 hours of prolonged incu-
shock, although not efficiently translated in the bation at 37~ In contrast, when the tissue-culture
cells. cells are brought back to 25~ after a heat shock
819
 Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

820 MIRAULT ET AL.

ment. We conclude that, under these conditions, the

heat-shock proteins are stable.
When tissue-culture cells are shifted to 37~ and
incubated at this temperature, the heat-shock prote-
ins are synthesized at high rates, and after 1 to 2
hours they can already be detected on gels stained
with Coomassie blue. They accumulate at 37~ until,
by 8 hours, they represent about 10% of the total
protein of the cells, as estimated from the densitome-
ter tracing of the stained gels (Fig. 1, right panel).

Fingerprint Analysis of Heat-shock-induced

Polypeptides
H o w m a n y distinct heat-shock-induced polypep-
tides are there? In an attempt to answer this ques-
tion, [ssS]methionine-labeled protein bands from
heat-shocked salivary glands and tissue-culturecells
were eluted from the gels, oxidized with performic
Figure 1. The effect of heat shock on protein synthesis. acid, and digested with trypsin. After separation by
(a) Salivary glands (SG) dissected from wild-type two-dimensional chromatography, the methionine-
Drosophila melanogaster (Kolmar) raised at 25~ were la- labeled peptides were detected by autoradiography.
beled with [35S]methionine (40 #Ci/ml, New England Nu- The results are shown in Figure 2. The 84,000-,
clear, 300-500 Ci/mM) for 30 rain at 25~ (control, C) and
for the same time at 25~ following a heat shock of 20 70,000-, 68,000-, 26,000-, 23,000-, and 22,000-dalton
min (HS) in a medium consisting of one part 10% ethanol bands (see Fig. 1) all give clearly different finger-
and five parts of Grace's (1962) medium without methio- prints. The fingerprint of the 27,000-dalton band dis-
nine. A tissue-culture cell line KC161 from Dr. Echalier plays all the spots found in the fingerprint of the
(Echalier and Ohanessian 1969) was adapted to grow in 26,000-dalton band, plus additional spots. Due to
suspension in the D22 medium of Echalier and Ohanessian
(1970), supplemented with 2% fetal calf serum; the cells technical difficulties in separating these two protein
were grown at 25~ to concentrations between 2 and 8 X 10e bands completely, we cannot say whether these pro-
cells per ml. Tissue-culture cells (TC) were labeled for 1 teins share some common sequences. The results in-
hr at 25~ (C) or for the same time at 37aC following a dicate that each of the other heat-shock protein
2-hr preincubation at 37~ (HS) in medium without cold
methionine, yeast extract, and serum. The proteins were bands is likely to represent a distinct polypeptide.
electrophoresed in SDS-12.5% pelyacrylamide gels accord- On the basis of this analysis, we conclude that the
ing to the method of Laemmli (1970) and Studier (1973), synthesis of at least six distinct polypeptides is in-
and the labeled proteins were detected by autoradiography. duced by heat shock.
(b) Coomassie blue staining patterns of proteins from
tissue-culture cells grown at 25~C (C) and from cells incu- Two-dimensional gel analysis of the heat-shock
bated for 1 or 8 hr at 37~ (HS) showing the accumulation
of heat-shock proteins with time. polypeptides. Since each fingerprint of Figure 2
could result from a possible mixture of different
polypeptides of the same or similar molecular
of 1 hour and the rate of synthesis of heat-shock weights, we carried out a two-dimensional gel analy-
proteins is examined by pulse-labeling at various sis according to the method of O'Farrell (1975). This
times thereafter, we find a gradual decrease of the method combines isoelectric focusing in 8 M urea
rate of synthesis of these proteins until,by 8 hours, in the first dimension and SDS-polyacrylamide
no synthesis of heat-shock protein is detected~ and gel electrophoresis in the second. The effect of
normal protein synthesis has resumed; the cellscon- heat shock on the two-dimensional gel patterns of
tinue to grow normally and can respond to a second [85S]methionine-labeled polypeptides is striking, and
heat shock. the data confirm the observations already made in
The stabilityof the heat-shock proteins was inves- the one-dimensional gel analysis. The major new
tigated in pulse-chase experiments with salivary finding, however, is the complexity of the 70,000-
glands and tissue-culture cells. The proteins were dalton heat-shock band, which now displays five dis-
labeled with [3"S]methionine for i hour at 37~ and tinct spots on the two-dimensional gel (Fig. 3, HS).
chased for various times at 25~ with 200 vg/ml Although the last 70,000-dalton spot on the right
unlabeled methionine. The intensity of each of the appears in the 25~ two-dimensional gel pattern
labeled heat-shock protein bands seen on autoradio- (Fig. 3, C) at roughly equal intensity, the four other
graphs of SDS-pelyacrylamide gels did not show any 70,000-dalton protein spots are strongly induced by
significant variation for periods up to 20 hours, and heat shock. The 27,000-dalton band also appears as
the total amount of radioactivityin acid-precipitable multiple spots. On the other hand, the single spots
material remained constant throughout the experi- corresponding to the 22,000-, 23,000-, 26,000-, 68,-
 Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

HEAT SHOCK OF DROSOPHILA MELANOGASTER 821

Figure 2. Tryptic fingerprints of the heat-

shock-induced polypeptides. Tissue-culture
cells were labeled with [85S]methionine for 1
hr at 37~ and the total proteins were frac-
tionated by SDS-polyacrylamide gel electro-
phoresis as described in the legend to Fig. I,
except that the electrophoreticmigration was
increased in order to improve the separation
of the 68,000- and 70,000-dalton polypeptides.
The radioactive heat-shock-induced polypep-
tides were eluted from the gel by electropho-
resis, oxidized with performic acid, and di-
gestod with trypsin as described by Allet et
al.(1973).The trypticpeptides were separated
by two-dimensional, thin-layer chromatogra-
phy (firstdimension, pyridine/isoamyl alco-
hol/water: 7/7/6; second dimension, butanol/
pyridine/acetic acid/water:5/4/1/4), and the
labeled spots were detected by autoradiog-
raphy.

000-, and 84,000-dalton protein bands are likely to rect comparison of spot intensities in Figure 3 might
originate from unique polypeptides, the synthesis lead to an overestimation of minor spots, since the
of which is strongly induced by heat shock. Tryptic major spots have been largely overexposed.
fingerprint analysis indicates that the two major In several instances when the proteins were la-
spots in the 70,000 molecular weight range repre- beled at 25~ we could detect weak spots in the
sent two polypeptides with very similar, ifnot identi- two-dimensional gel pattern, corresponding pre-
cal, primary structures. These two spots account for cisely to heat-shock spot positions, e.g., (6.7; 1.85),
over 9 0 % of the protein synthesis at 70,000 daltons. (7.0; 1.85), (9.75; 1.35). They represent probably a
The differences in isoelectric points m a y result low level of heat-shock protein synthesis at 25~
from posttranslational modification. Note that a di- In conclusion, the evidence obtained so far sug-
 Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

822 M I R A U L T ET AL.

Heat-shock-induced Messenger RNAs

Polyribosomes from heat-shocked cells consist of
a major fraction composed of large polyribosomes
with 20 to 30 ribosomes and a minor fraction of
smaller polyribosemes (McKenzie et al. 1975). The
newly synthesized poly(A)-containing R N A found in
these polyribosomes was analyzed by sedimentation
in sucrose gradients and compared with the R N A
from polyribosomes of the corresponding size from
non-heat-shocked cells (Fig. 4). Polysomal RNA, la-
beled with [SH]uridine at 37~ sediments essentially
as two peaks: a 20S RNA fraction predominant in
the large polyribosomes and a 12S R N A fraction
found mostly in the small polyribosomes. In con-
trast, the poly(A)-containing R N A labeled at 25~C
sediments quite heterogeneously t h r o u g h o u t the
gradient, and very little of this labeled R N A is seen
in the heat-shock RNA sedimentation profiles.
Several R N A fractions isolated from a preparative
sucrose gradient and sedimenting at 20S, 18S, 15S,
and 12S were assayed for messenger R N A activity
in vitro, using the messenger-dependent reticulocyte
lysate described by Pelham and Jackson (1976). The
analysis of the products synthesized in vitro are
shown in Figure 5. The 20S heat-shock R N A directed
Figure 3. Two-dimensional gel analysis of[ssS]methionine- the synthesis of at least three polypeptides of 84,000,
labeled heat-shock polypeptides. A cell culture was divided
into two portions and each portion labeled with 50/~Ci/ 70,000 and 68,000 daltons, plus traces of smaller
ml [ssS]methionine. One portion was labeled for 2 hr at products. The 12S heat-shock R N A directed the syn-
37~ following a 1-hr incubation at 37~ in medium with- thesis of at least three polypeptides comigrating
out methionine and yeast extract (HS); the other was la- with the small heat-shock polypeptides labeled in
beled for 2 hr at 25cC following a 1-hr incubation at 25~C
in the same medium (C). About 2.5 • 107 cells of each por- vivo. Electrophoretic migration, fingerprint analysis
tion were washed in ice-cold isotonic saline and lysed in of each individual protein band labeled in vivo and
0.1 ml of a solution containing 10 mM HEPES, pH 7.4, in vitro, respectively, and two-dimensional gel anal-
0.5% NP-40, 50 pg/ml pancreatic ribonuclease A, 50 pg/ ysis in several cases suggest t h a t probably all of
ml pancreatic deoxyribenuclease I, and 20 mM ~-mercap- the heat-shock polypeptides have been faithfully
toethanol. Each lysate was incubated under nitrogen for
10 min at 30cC followed by a further incubation of 5 rain synthesized in vitro. RNA fractions sedimenting at
in the presence of 80 mM EDTA, pH 7.4. The samples were 18S and 15S directed the synthesis of small amounts
then treated according to the method of O'Farrell (1975), of products comigrating with polypeptides synthe-
and an aliquot corresponding to about 2.5 • 105 cells was sized in vivo at 25~ The products synthesized
analyzed by isoelectric focusing in the first dimension fol-
lowed by electrophoresis in 12.5% polyacrylamide-SDS gels in vitro by the 20S, 18S, 15S, and 12S R N A frac-
in the second dimension (O'Farrell 1975). Fluorographic tions from cells kept at 25~ (Fig. 5, left panel) cor-
exposure was for 30 hr at --70~C with preflashed Kodak respond to those observed after in vivo labeling at
XR-5 films, as described by Laskey and Mills (1975). The 25~
heat-shock (HS) and control (C) patterns are shown in the F u r t h e r purification of individual heat-shock mes-
figure. An autoradiogram of the heat~hock proteins ana-
lyzed by electrophoresis in the SDS dimension only is senger RNAs was attempted by polyacrylamide gel
shown on the right as reference. The 70,000, 68,000, 27,000, electrophoresis. Figure 6 shows the electrophoretic
and 26,000 heat-shock protein bands display spots with pattern of [3I-I]uridine-labeled 20S heat-shock R N A
the following coordinates: (5.2; 1.8), (5.7; 1.8), (6.65; 1.85), in a 7 M urea-polyacrylamide gel. Three R N A peaks
(7.0; 1.85), (8.0; 1.85), (6.05; 2.0), (1.3; 6.35), (1.5; 6.35), (0.6;
6.5). Evidence suggests that the reproducible streaking- have been partially resolved. W h e n assayed in vitro,
like pattern seen in the SDS dimension of the figure (HS) fraction HI directed the synthesis of a 68,000-dalton
results from some partial proteolytic digestion of the ma- polypeptide; fraction II, the most a b u n d a n t heat-
jor 70,0(O-dalton heat-shock protein, possibly occurring shock mRNA, directed predominantly the synthesis
within the cells at 37~C. of a 70,0004alton polypeptide; and fraction I di-
rected the synthesis of an 84,000- as well as a 70,000-
dalton polypeptide. This last fraction was still con-
gests t h a t heat shock strongly induces the synthesis taminated by messengers present in fraction II.
of at least six distinct polypeptides, and possibly as Simila~ electrophoretic and translation patterns
m a n y as nine. were obtained with 20S heat-shock R N A fraction-
 Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

Figure 4. Sucrose gradient sedimentation "20 S" "12 s" "20 $" "12 S"
8
of polysomal poly(A)-containing RNA syn- t b
thesized at 37~ (a,b)and at 25~ (c,d).
Tissue-culture cells were labeled with
[SH]uridine (100 #Ci/ml) for 60 rain at 37~
3000
I
23`= 165
~ 1500
following a 15-rain incubation at this tom- l 1
perature, and for 60 rain at 25~ in a paral-
lel incubation. The cells were quickly 2000 I000
chilled to 0~C and lysed according to the
method of McKenzie et al. (1975). Frac-
tions of large and small polyribesomes ,ooo 5O0
were pooled from sucrose gradients, as de- (J
scribed previously, and precipitated with w w
1 volume of ethanol (Moran et al. 1977). Z

The pellets (100-500 A26ounits) were first 0 "~

E: I0 20 30 ~b 2'o 3o
dissolved in 10 ml of a solution containing m
,f' C d J
0.02 M NaCI, 0.04 M EDTA, 0.02 M Trls- I

HCI, pH 7.5, 1% SDS, and 1 mg proteinase " 1500 1500 "

K (Merck, Germany), incubated 90 sec at
100~ treated again with 1 mg proteinase
K for 15 rain at 25~ and phenol-extracted Ioo0 1000
according to the method of Aviv and Leder
(1972). The RNA was fractionated on
oligo(dT)-cellulose, and the poly(A)-con-
taining RNA samples were analyzed by 500 50O
sedimentation in linear 5-20% sucrose
gradients according to the method of
Haines et al. (1974). Centrifugation was 0
carried out for 150 rain in the SW56 rotor Io 20 30 ro 20 30
FRACTION NUMBER
of the Beckman centrifuge at 56,000 rpm
and 20~ (a) RNA from large, heat-shock
polyribosomes; (b) RNA from small, heat-shock polyribosomes; (c) RNA from large polyribesomes of cells labeled at
25~ (d) RNA from small polyribosomes of cells labeled at 25~

Figure 5. Fluorogram of [3sS]methionine-

labeled polypeptides synthesized in vitro
by poly(A)-containing R N A fractions iso-
lated from preparative sucrose gradients.
The R N A was translated in vitro in the
messenger RNA-dependent reticulocyte
lysate (Pelham and Jackson 1976). Incuba-
tion was carried out for 60 rain at 30~
at rate-limiting R N A concentrations, and
the ["S]methionineqabeled polypeptides
were analyzed by SDS-polyacrylamide gel
electrophoresis and detected by fluorogra-
phy. (Center panel) The polypeptides la-
beled in vivo for 1 hr at 25~ (C) or during
a 1-hr heat shock at 37~ (HS) are shown
at two different exposures. (Right and left
pane/s)The polypeptides synthesized in vi-
tro by 20S, 18S, 15S, and 12S poly(A)-con-
taining RNA fractions from heat-shock
cells (HS) and from 25~ control cells (C),
respectively. Fluorographic exposure was
at --70~ for 5 hr.

823
 Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

824 MIRAULT ET AL.

gether, the results indicate that three heat-shock-

induced messenger RNA species synthesized at 37~
each code for one of the three major heat-shock poly-
peptides of large molecular weight.
The electrophoretic pattern of the poly(A)-contain-
ing RNA from small heat-shock polyribosomes, la-
beled with [3H]uridine at 37~ is shown in Figure
8. The 12S heat-shock RNA has been partially re-
solved into two peaks. The RNA from the minor
peak directed predominantly the in vitro synthesis
o f a 23,0(X)-dalton polypeptide. RNA from the major
peak gave rise to three protein bands corresponding
to the small heat-shock polypeptides labeled in vivo.
Similar results were obtained with nondenaturing
gels (Moran et al. 1977). The partial resolution of
the 12S heat-shock RNA suggests the existence of
at least three distinct messenger RNAs coding for
the small heat-shock polypeptides.
Estimates of the quantity and specific activity of
beth the 20S and 12S RNAs clearly indicate that
the major part of these heat-shock messenger RNAs
is synthesized after transfer of the cells to the high
temperature. Thus, the synthesis of these RNAs is
induced by heat shock.
In summary, we have so far identified six heat-
shock-induced messenger RNAs, each of which codes
for a distinct heat-shock polypeptide.

Figure 6. Electrophoretic pattern of 208 heat-shock Many messenger RNAs normally translated at 25~
poly(A)-containingRNA labeled with [SH]uridine. The 20S are not e~ciently translated after heat shock. The
heat-shock poly(A)-containingRNA was isolated from suc- rapid shut-off in the synthesis of many proteins
rose gradients as in Fig. 4 and electrophoresed in 3.3%
polyacrylamide gels in 7 M urea, according to the method (Figs. 1, 3, and 9) and the concomitant disappear-
of Reijnders et al. (1973) as modified by Spradiing et al. ance of most 25~ polyribosomes after heat shock
(1977), at 14 V/cm for 22 hr. The cylindrical gels (14 cm (McKenzie et al. 1975) suggest that most of the mes-
long, 0.6 cm diameter) were cut frozen in 2-ram slices, and senger RNA present at 25~ is either quickly de-
the RNA was eluted by sbsklng each slice in 0.5 ml 0.01 graded or not efficiently translated into proteins at
M Trls-HC1, pH 7.4 and 0.1% SDS at 4~ for 48 hr. This
elution step was repeated once and the total TCA-precipita- high temperature.
ble radioactivity was determined. RNA fractions to be When poly(A)-containing RNA prepared from the
translated in vitro were repurified on oligo(dT)-celluiose cytoplasm of heat-shocked tissue-culture cells is
prior to in vitro translation. More than 90% of the radioac- translated in vitro, the labeled products include not
tivity was then recovered as poly(A)-containingRNA. After
several ethanol precipitations to remove salt and SDS, suit- only the heat-shock polypeptides, but also those nor-
able amounts of the RNA were dissolved directly into 15 really synthesized in vivo at 25~ (Fig. 9). Essentially
/~1messenger RNA-dependent reticulocyte lysate (Pelham the same results were obtained when in vitro trans-
and Jackson 1976) supplemented with [ssS]methionine. lation was done under saturating, rather than rate-
The lysates were incubated for 90 min at 30r Samples limiting, RNA concentrations, whether at 25~ or
(2/xl) were analyzed by SDS-pelyacrylamide gel electropho-
resis. The fluorograms of the polypeptides synthesized in 37cC (R. J. Jackson, pers. comm.). In contrast, we
vitro by the 20S RNA peak fractions I, II, and III are shown have seen that purified poly(A)-containing RNA pre-
next to the heat-shock polypeptides labeled in vivo. pared from heat-shock polyribosomes directs mainly
the synthesis of the heat-shock proteins.
ated in nondenaturing gels (Moran et al. 1977). The We conclude that much of the messenger RNA
molecular weight of the major component in 20S made and normally translated at 25~C is still present
RNA was estimated to be about 9 • 10~ by electro- in the cytoplasm of heat-shocked cells but not effi-
phoresis in 98% formamide gels under fully denatur- ciently translated. The active polyribosomes contain
ing conditions according to the method of Spehr et mainly heat-shock-induced messenger RNAs. We
al. (1975). The fingerprint of the 70,000-dalton poly- have observed that at least some histones and sev-
peptides synthesized in vitro by fraction II is shown eral other proteins continue to be synthesized effi-
in Figure 7, together with that of the 70,000-dalton ciently after heat shock, suggesting that some mes-
heat-shock polypeptides synthesized in vivo. Both senger RNAs are more resistant to the translation
fingerprint patterns are identical. Thus, taken to- shut-off caused by heat shock.
 Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

Figure 7. Tryptic fingerprints of the

70,000-dalton heat-shock polypeptides
synthesized in vivo or in vitro as di-
rected by the 20S fraction-II heat-
shock mRNA (see Fig. 6). The [35S]
methionine-labeled polypeptides were
eluted from the gels, digested with
trypsin, and analyzed as described in
the legend to Fig. 2.

Figure 9. In vitro translation products of cytoplasmic

poly(A)-containing RNA from the cytoplasm of heat-
shocked cells and of cells kept at 25~. A culture of cells
grown at 25~C was divided into two equal parts of about
2.5 X 109 cells each. One part was incubated for 80 rain
at 37~ (HS) while the other part was kept at 25~ (C).
The cells were lysed according to the method of McKenzie
et al. (1975) and the postmitochondrial supernatants (2
ml) were phenol-extracted at room temperature (Aviv and
Leder 1972) in 12 m] of a solution containing 0.02 M NaC1,
0.04 M EDTA, 0.02 M Tris-HCl, pH 7.5, 1% SDS, and 2
mg proteinase K (Merck, Germany). Phenol extraction and
proteinase-K treatment were repeated until the interphase
precipitates completely disappeared. After ethanol precipi-
tation, the RNA was fractionated on oligo(dT)-celinlose and
the poly(A)-containing RNA was reprecipitated twice with
Figure 8. Electrophoretic pattern of poly(A)-containing 3 volumes of ethanol to remove traces of SDS before final
RNA from small heat-shock polyribosemes. Electruphore- dissolution in sterile bidistilled water. The poly(A)-contain-
sis was carried out in 3.5% polyacrylamide gels in 7 M ing RNA was translated in vitro and the products displayed
urea and the RNA was recovered and translated in vitro as in Fig. 5. (Left)Patterns of proteins synthesized in vivo:
as described in the legend to Fig. 6. The fluorograms of heat-shock proteins at two dilutions (HS) and proteins syn-
the pelypeptides synthesized in vitro by discrete 12S RNA thesized at 25cC (C). (Right) Patterns of proteins synthe-
fractions and separated by electrophoresls are displayed sized in vitro: B, blank with no RNA added; HS, with cyto-
on top of the corresponding gel slices. In vivo labeled heat- plasmic pely(A)-containing RNA from heat-shocked cells;
shock polypeptides analyzed in the same gel are shown C, with cytoplasmic pely(A)-containing RNA from cells
as references. kept at 25~
825
 Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

826 MIRAULT ET AL.

DISCUSSION be outcompeted by the heat-shock-induced messen-

ger RNAs. This has not been observed in vitro in
Gene expression in heat-shocked Drosophila mela- the reticulocyte lysate system, even under heat-
nogaster appears to be controlled beth at the level shock RNA saturating conditions. We do not know,
of transcription and at that of translation. At the however, whether such mRNA competition does oc-
level of transcription, a specific induction of RNA cur within the cells, accounting for the shut off ob-
synthesis is observed. The messenger RNAs coding served in vivo.
for the heat-shock proteins are produced at high
rates, and the synthesis of the heat-shock proteins
correlates with the accumulation of their corre- Acknowledgments
sponding messenger RNAs. Six of these messenger We wish to thank Mrs. Jacqueline Molliet, for ex-
species have been identified and partially purified cellent technical assistance, Dr. R. J. Jackson for
by gel electrophoresis. A similar electrophoretic pat- helpful comments on the manuscript, and Dr. G.
tern of heat-shock RNA has been obtained by Echalier for a strain of Drosophila melanogaster
Spradling et al. (1977), who have hybridized in situ tissue-culture cells. This work was supported by
specific RNA fractions to salivary gland chromo- grant No. 3.491.075 from the Swiss National Foun-
somes. Combining the results of in situ hybridization dation for Scientific Research.
and in vitro translation reported here, a tentative
correlation between puffs and polypeptides can be
drawn in several cases. Assuming that the A1, A2, REFERENCES
and A3 RNA species of Spradling et al. (1977) are
AwJ.~r,B., K. J. KATAGIRI, and R. F. GZSTZLAND. 1973. Char-
the 20S-I, -II, and -III species described here, the acterization ofpolypeptides made in vitrofrom bacterio-
corresponding heat-shock polypeptides of 84,000, phage lambda DNA. J. Mol. Biol. 78: 589.
70,000, and 68,000 apparent molecular weight origi- ASHRU~, M. 1970. Patterns of puffing activity in the
nate from sequences at puff sites 63C, 87A-87C1, salivary gland chromosomes of Drosophila. V. Re-
and 95D, respectively. No correlation can be drawn sponses to environmental treatments. Chromosoma
31: 356.
as yet for the small heat-shock polypeptides as their AvIv, H. and P. LEDER.1972. Purification of biologically
messengers have not yet been adequately resolved. active globin messenger RNA by chromatography on
Our results (Fig. 3 and fingerprint analysis of the oligothymidylic acid-cellulose. Proc. Natl. Acad. Sci. 69:
two major 70,000-dalton spots shown in Fig. 3) are 1408.
B ~ D z s , H. D. 1972. The control of puling in Drosophila
not inconsistent with the presence of two major hydei. In Developmental studies on giant chromosomes
70,000-dalton polypeptides, with slight differences (ed. W. Beermann et al.), p. 186. Springer Verlag, New
in primary structures, arising from two different York.
genes. ECHALIER,G. and A. 0HANESSIAN.1969. Isolement, en cul-
In parallel with the induction of specific RNAs, ture in vitro, de lign~.s cellulaires diplo'fdes de
Drosophila melanogaster. C. R. Acad. Sci. D 268: 1771.
shut off in the synthesis of many messenger RNA __. 1970. In vitro culture of Drosophila melanogaster
species normally synthesized at 25~ is observed. embryonic cells. In Vitro 6: 162.
Normal developmental puffs regress after heat GRACE,T. D. C. 1962. Establishment of four strains of cells
shock (Ashburner 1970), as does the synthesis of the from insect tissues grown in vitro. Nature 195: 788.
I-I~N~s, M. E., N. H. CAREY,and R. D. PALMrrF~. 1974.
corresponding messengers (see Fig. 4 together with Purification of ovalbumin messenger RNA. Eur. J. Bio-
the results of in situ hybridizations by McKenzie chem. 43: 549.
et al. [1975] and Spradling et al. [1977]). LAF~MLLU. K. 1970. Cleavage of structural proteins dur-
Finally, the finding of messenger RNA not trans- ing the assembly of the head of bacteriophage T4.
lated in heat-shocked cells but nevertheless present Nature 227: 680.
LASKEY,R. A. and A. D. MmLS.1975. Quantitative fluorog-
in the cytoplasm and translatable in vitro suggests raphy of acrylamide gels. Eur. J. Biocher~ 56: 335.
an additional type of control at the level of transla- Lzwm, M., P. J. HELMSING,and M: ASHBURNER.1975. Par-
tion. Although a more thorough analysis of this phe- allel changes in puffing activity and patterns of protein
nomenon remains to be done, we can conclude that synthesis in salivary glands of Drosophila. Proc. Natl.
Acad. Sci. 72: 3604.
an appreciable fraction of the messenger RNAs McKzNzxe, S. L., S. H~.mKOFF, and M. MF~E~ON. 1975.
made at 25~C is still present but not translated effi- Locallsation of heat-induced polyribosomal RNA. Proc.
ciently in vivo at high temperature. The degree of Natl. Acad. Sci. 72: 1117.
translation shut-off varies from experiment to ex- MORAN,L., M.-E. MXRAULT,A. P. ARmGO,M. GOLaSCHMI~r-
periment, but we have observed a consistent de- CLERMONT, and A. TISSXkRES. 1977. Heat shock of
Drosophila melanogaster induces the synthesis of new
crease in the rate of synthesis of most cellular messenger RNAs and proteins. Proc. R. Soc. Lond. B
proteins in heat-shocked cells. Although the corre- (in press).
sponding messenger RNAs may undergo some deg- O'FAR~TJ.~ P. H. 1975. High resolution two-dimensional
radation, our in vitro translation results (Fig. 9) electrophoresis of proteins. J. Biol. Chem. 250: 4007.
~ , H. R. B. and R. J. JACKSON.1976. mRNA-depend-
show that degradation alone cannot account for the ent translation system from reticulocytes. Eur. J. Bio-
translation shut-off observed in vivo at high tem- chem. 67: 247.
perature. The untranslated messengers could simply RF~ND~.RS,P., P. SLOOF,J. SXVAL,and P. BORST.1973. Gel
 Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

H E A T SHOCK OF DROSOPHILA M E L A N O G A S T E R 827

electrophoresis of RNA under denaturing conditions. SPRADLING,A., S. PENMAN,and M. L. PARVUE.1975. mRNA

Biochim. Biophys. Acta 324: 320. in Drosophilacultured cells. Cell 4: 395.
RrrossA, F. 1962. A new putfmg pattern induced by tem- STUDIER,F. W. 1973. Analysis of bacteriophage T7 early
perature shock and DNP in Drosophila. Experientla 18: RNAs and proteins on slab gels. J. Mol. Biol. 79: 237.
571. TISSI~.RES, A., H. K. MITCHELL, and U. M. TRACY. 1974.
SPOHR, G., M.-E. MIRAULT,T. IMAIZUMI,and K. SCHERRER. Protein synthesis in salivary glands of Drosophila mela-
1975. RNA electrophoresis on gels in formamide. Eur. nogaster: Relation to chromosome puffs. J. Mol. Biol.
J. Biochem. 62: 313. 84: 389.
SPRAVUNG,A., M. L. PARDUE,and S. PENMAN. 1977. Mes-
senger RNA in heat-shocked Drosophila cells. J. Mol.
Biol. 109: 559.
Downloaded from symposium.cshlp.org on June 29, 2016 - Published by Cold Spring Harbor Laboratory Press

The Effect of Heat Shock on Gene Expression in

Drosophila melanogaster
M.-E. Mirault, M. Goldschmidt-Clermont, L. Moran, et al.

Cold Spring Harb Symp Quant Biol 1978 42: 819-827

Access the most recent version at doi:10.1101/SQB.1978.042.01.082

References This article cites 20 articles, 4 of which can be accessed free

at:
http://symposium.cshlp.org/content/42/819.refs.html

Article cited in:

http://symposium.cshlp.org/content/42/819#related-urls
Email alerting Receive free email alerts when new articles cite this article -
service sign up in the box at the top right corner of the article or click
here

To subscribe to Cold Spring Harbor Symposia on Quantitative Biology go to:

http://symposium.cshlp.org/subscriptions

Copyright © 1978 Cold Spring Harbor Laboratory Press