DEVELOPMENTAL
125,200-207
BIOLOGY
(1988)
Developmental Regulation of a Constitutively Expressed Mouse mRNA Encoding a 72-kDa Heat Shock-like Protein LUTZ B. GIEBEL, BERND P. DWORNICZAK, Zentrum
fiir
Molekulare
Biologic der Universit&t Heidelberg D-6.900 Heidelhnrg, Federal Republic
AND
E. K. F.
BAUTZ’
(ZMBH), Im Neuenheimer oj’ Germnn y
Feld
~82,
Multiple heat shock cognate (hsc70) cDNA clones were isolated from the mouse embryonal carcinoma cell line F9. They all encode a single 72-kDa protein, which is constitutively expressed in all mouse cell lines and tissues tested, and which is only slightly induced by hyperthermia. hsc70 RNA is very abundant in F9 stem cells and brain, but very little is found in ll-day-old embryos. Upon differentiation of F9 stem cells induced by retinoic acid and cyclic AMP, expression of the hsc70 gene decreases only slightly, suggesting that hsc70 is highly expressed in early mouse development and is then down-regulated towards the end of embryogenesis. In adult tissues only the brain retains the high level of hsc70 gene expression found in F9 stem cells. We also show that expression of hsc70 protein and clathrin is uncoupled in F9 cells, indicating that the uncoating activity of coated vesicles may not be the only function of hsc70 protein. o 1988 Academic
Press, Inc.
INTRODUCTION
Cells of all organisms, even phylogenetically as distant as bacteria and man, respond to an increase in temperature or other environmental stresses such as drugs, recovery from anoxia, and arsenite poisoning by transient expression of a small set of proteins, while the expression of most other cellular proteins declines concomitantly. The most prominent member of these socalled heat shock proteins (hsps) is a 70-kDa gene product that has been highly conserved during evolution of pro- and eukaryotes. In eukaryotes the hsp70 gene belongs to a multigene family encoding a number of proteins of similar primary structure (for review, see Craig, 1985). In Drosophila five members of this family are induced by hyperthermia, whereas at least three additional genes (heat shock cognates) are expressed constitutively and are not induced by heat shock (Craig et al., 1983). These so-called heat shock “cognate” genes (hsc) encode two proteins that arc expressed at all dcvelopmental stages and one protein that is abundant in embryos. In the yeast Saccharorruyces cerevisiae a hsp70 multigene family of eight related genes was described (Craig and Jacobsen, 1985). This family also consists of both heat-inducible and constitutively expressed genes. In mouse there are at least three heat-inducible genes in the hsp70 gene family (Lowe and Moran, 1986). To understand the function and regulation of heat shock “cognate” genes in the mammalian hsp70 multi’ To whom 0012-1606/88 Copyright All rights
correspondence
should
$3.OC
0 1988 by Academia Press, Inc. of reproduction in any form reserved.
be addressed. 200
gene family clones. We cDNA clone hsc70 gene regulation.
we isolated multiple mouse hsc70 cDNA here report the nucleotide sequence of a containing the entire coding region of the and investigate hsc70 gene expression and
MATERIALS
AND
METHODS
Construction of a mouse hscro cDNA library. A synthetic 20mer oligonucleotide complementary to the mouse hsc70 mRNA (Fig. 1, position 162-181) was prepared on a DNA synthesizer (Applied Biosystems) by the phosphoamidit method. Purified primer (100 fmole) was hybridized to 10 pg of poly(A+)-selected F9 RNA in 0.4 MNaCl, 10 mM Pipes, pH 6.4, containing 12 units of ribonuclease inhibitor for 2 hr at 50°C. hsc70 cDNA was prepared using the Amersham cDNA synthesis kit, ligated into EcoRI “arms” of bacteriophage XgtlO (Maniatis et al., 1985), and packaged using the Promega Biotee in vitro packaging system. Recombinant phages were propagated on Escherichia coli hosts C600 HFL and C600. Nucbeotide sequence. Convenient restriction fragments of various mouse hsc70 cDNA clones were cloned into either M13mp18 or M13mp19 and the nucleotide sequences were determined by the Sanger dideoxy chain termination procedure using either [o(-~‘P]- or [cx-~~S]dCTP (Sanger et al., 1977). Cell culture, heat shock, dgerentiation, and in vivo labeling of proteins. F9 cells, kindly provided by S. Strickland, were grown as a monolayer on gelatinized
GIEBEL, DWORNICZAK, AND BAUTZ
Developmental
Regulation qf hsc70
gaggcagctgccgggcattagtgtggtctcgtcgtcagcgcagctgggcctacacacaagcaaccATGTCT~GGGACCTGCAGTTGGCATTGATCTCGG
201 100
HetSerLysGlyProAlaValGlyIleAspLeuGl CACCACCTACTCCTGTGTGGGTGTCTTCCAGCATGGAAAGGTGGAAATTA~GCCAATGACCAGGGTMCCGCACCACGCCMGCTATG~GC~TCACG yThrThrTyrSerCysValGlyValPheGlnHisGlyLysValGluIleIleAlaAsnAspGlnGlyAsnArgThrThrProSerTyrValAlaPheThr
200 300
ATGATGCTGTTGTTCAC-TCTGATATGAAGCACTGGCCCTTCATGGTGGTGMTGATGCAGGCAGGCCCMGGTCCMGTCGMTACAMGGGGAGACMA spAspAlaValValGlnSerAspMetLysHisTrpProPheMetValValAsnAsnAlaGlyArgProLysValGlnValGluTyrLysGlyGluThrLy
400
AAGTTTCTACCCAGAGGAAGTGTCCTCCATGGTTCTGACAAAGATGAAGGMATTGCAGMGCATACCTCGGMAGACTGTTACCAACGCTGTGGTCACA sSerPheTyrProGluGluValSerSerMe~ValLeuThrLysMetLysGluIleAlaGluAlaTyrLeuGlyLysThrValThrAsnAlaValValThr
500
GTGCCCGCTTACTTCAATGACTCTCAGCGACAGGCAACAAAAGATGCTGGMCTA~GCTGGCCTC~TGTAC~CG~TCATC~TGAACCAACTGCTG ValProAlaTyrPheAsnAspSerGlnArgGlnAlaThrLysAspAlaGlyThrIleAlaGlyLeuAsnValLeuArgIleIleAsnGluProThrAlaA
600
CTGCTATTGCTTATGGC~AGATAAGAAGGTCGGAGCTGMAGGAATGTGCTCATTTTPGACTTGGGAGCTGGCACTTTTGATGTGTCAATCCTCACTAT laAlaIleAlaTyrGlyLeuAspLysLysValGlyAlaGluArgAsnValLeuIlePheAspLeuGlyGlyGlyThrPheAspValSerIleLeuThrIl
AAGCGAAAGCACAAGAMGACATCAGTGAGAACAAGAGAGCTGTCCGCCGTCTCCGCACGGCCTGCGAGCGGGCCAAGCGCACCCTCTCCTCCAGCACCC LysArgLysHisLysLysAspIleSerGluAsnLysArgAlaValArgArgLeuArgThrAlaCysGluArgAlaLysArgThrLeuSerSerSerThrG ACGCCAGTATTGAGATTGATTCTCTCTATGAGGGAA~GACTTCCCTCCATTACCCGTGCTCGA~GAGGAG~GMTGCTGACCTGTTCCGTGG lnAlaSerIleGluIleAspSerLeuTyrGluGlyIleAspPheTyrThrSerIleThrArgAlaArgPheGluGluLeuAsnAlaAspLeuPheArgGl 1100 ATCCAGAAACTTCTGCAAGACTTCTTCAATGGMMGAGCTGAACAAGAGCA,TTMCCCCGATGMGCTG~GCCTATCGTGCAGCTGTCCAGGCAGCCA IleClnLysLeuLeuGlnAspPhePheAsnGlyLysGluLeuAsnLysSerIleAsnProAspGluAlaValAlaTyrGlyAlaAlaValGlnAlaAlaI
1200
TTCTATCTGGAGACAAGTCTGAGAACGTTCAGGATTTGCTGCTCTTGGATGTCACTCCTCTTTCCCTTGGTATTGMACTGCTGGCGGAGTCATGACTGT lrLeuSerGlyAspLysSerGluAsnValGlnAspLeuLeuLeuLeuAspValThrProLeuSerLeuGlyIleGluThrAlaGlyGlyVal~e~ThrVa
1300
CCTCATCAAGCGCAATACCACCATCCCCACCAAGCAGACACAGACTCTCACCACCTACTCTGACMCCAGCCTGG~TACTCATTCAGGTGTATGAAGGT
1400
lLeuIleLysArgAsnThrThrIleProThrLysGlnThrGlnThrLeuThrThrTyrSerAspAsnGlnProGlyValLeuIleGlnValTyrGluGly 1500 ACATCGATGCCAATGGCATCCTCAATGTTTCTGCTGTAGATMGAGCACAGG~GGAGMCMGATCACCATCACCMn;ACMGGGCCGC~GAGTM spIleAspAlaAsnGlyIleLeuAsnValSerAlaValAspLysSerThrGlyLysGluAsnLysIleThrIleThrAsnAspLysGlyArgLeuSerLy
1600
CGAAGATATTGAGCGCATGGTCCMGAAGCTGAGMGTACMGGCTGAGGATGAGMGCAGAGAGATAAGG~CCTCCMGMCTCACTGGAGTCCTAT
1700
sGluAspIleGluArgMetValGlnGluAlaGluLysTyrLysAlaGluAspGluLysGlnArgAspLysValSerSerLysAsnSerLeuGluSerTyr GCCTTCAACATGAAAGCAACTGTGGMGATGAGAAACTTCAAGGCMGATCAATGATGAGGACMACAGMGATTCTTCACMCTCCAATGAMTCATCA
1800
AlaPheAsnMetLysAlaThrValGluAspGluLysLeuGlnGlyLysIleAsnAspGluAspLysGlnLysIleLeuAspLysCysAsnGluIleIleS GCTGGCTGGATAAGMCCAGACTGCAGAGAAGGAAGAATTTGAGCATCAGCAGAMGAACTGGAG~GTCTGCAACCCTAT
1900
erTrpLeuAspLysAsnGlnThrAlaGluLysGluGluPheGluHisGlnGlnLysGluLeuGluLysValCysAsnProIleIleThrLysLeuTYrGl CAGTGCAGGTGGCATGCCTGGAGGGATGCCTGGTGGCTTCCCAGGTGGAGGAGCTCCCCCATCTGGTGGTGCTTC~CAGGCCCCACCATTGAAGAGGTG nSerAlaGlyGlyMetProGlyGlyMetProGlyGlyPheProGlyGlyGlyAlaProProSerGlyGlyAlaSerSerGlyProThrIleGlUGlUVal
2000 2089
FIG. 1. Nucleotide sequence of the mouse hsc70 cDNA clone. Small letters denote the 5’ and 3’ untranslated regions and capital letters denote the coding regions. The deducted amino acid sequence is shown. Termination codon and polyadenylation signal are indicated.
175-cm2 tissue culture flasks in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal calf serum. Differentiation was induced by 0.5 WM retinoic acid and 1 mM dibutyryladenosine 5’-3’ cyclic monophosphate as described by Strickland et al. (1980). Heat shock was performed at 42°C for 4 hr. Proteins of heat-shocked and non-heat-shocked F9 cells were labeled in viva with [35S]methionine. Cells were grown to
near confluence, washed with prewarmed methioninefree medium, and proteins were labeled with 20-30 pCi/ml medium of [35S]methionine (HO0 CVmmole, Amersham) in methionine-free medium for 1 hr at 37°C (non-heat-shocked cells) or 42°C (heat-shocked cells). Cells were then washed in ice-cold phosphate-buffered saline and lysed as described by Laemmli (1970). Southern analysis. Genomic F9 DNA was digested
202
DEVELOPMENTAL
BIOLOGY
with various restriction enzymes and analyzed by the method of Southern (1975) using as a probe a nicktranslated 500-bp cDNA fragment derived from the coding region of the mouse hsc70 gene. Hybridization was performed at 68°C for 20 hr. Filters were washed twice in 3X SSC, 0.1% SDS and twice in 0.1X SSC, 0.1% SDS each for 30 min at 68°C. Northern and RNA dot-blot analysis. Total cellular RNA was isolated by the guanidium thiocyanide method of Chirgwin et al. (1979). Polyadenylated RNA was selected by chromatography on oligo(dT)-cellulose (Aviv and Lederer, 1972). For Northern analysis polyadenylated RNA was fractionated on 1% formaldehyde agarose gels and transferred to nitrocellulose. For RNA dot-blot analysis total cellular RNA was transferred to nitrocellulose using a microsample filtration manifold device. Hybridization was performed in 50% formamide, 20 mM phosphate buffer, 5X Denhardt’s, 5~ SSC, 0.1% SDS, and 100 pg/ml yeast RNA for 20 hr at 42°C using as a probe a nick-translated 1.2-kb cDNA fragment containing the 3’half of the gene. Filters were washed twice in 3x SSC, 0.1% SDS and twice in 0.5~ SSC, 0.1% SDS for 30 min each at 60°C. Western blot analysis. Total cellular proteins of F9 cells, heat-shocked F9 cells, mouse brain, and mouse liver were separated on a 7.5% SDS-polyacrylamide gel, and transferred to nitrocellulose by electrophoresis. Filters were incubated in 0.14 M NaCl, 10 mM Tris-Cl, pH 7.4, supplemented with 50% horse serum for 60 min at room temperature. Filters were then incubated with either a polyclonal rabbit anti-human hsp70 antibody which cross-reacts with both mouse hsc70 and hsp70 protein or a monoclonal mouse antibovine a-clathrin antibody which cross-reacts with mouse cu-clathrin protein in 1 M NaCl, 10 mM Tris-Cl, pH 7.4, supplemented with 10% horse serum for 90 min at room temperature. Filters were washed for 60 min in 1 M NaCl, 10 mM Tris-Cl, pH 7.4, 1% Triton, rinsed in H20, and incubated with either peroxidase-conjugated goat antirabbit IgG or goat anti-mouse IgG antibodies in 1 M NaCl, 10 mM Tris-Cl, pH 7.4, for 10 min. Peroxidase reaction was performed in 0.14 M NaCl, 10 mM Tris-Cl, pH 7.4, 0.02% 4-chloro-l-naphtol, 5% methanol, 0.05% HeOz. In vitro translation of hybrid-selected RNA. Denatured plasmid hsc70 cDNA (10 pg each) was immobilized on nitrocellulose using a microsample filtration manifold device. Five filters each were placed into a plastic vial and hybridized to 20 pg of poly(A+)-selected RNA isolated from F9 stem cells and heat-shocked F9 stem cells in 50% formamide, 0.9 M NaCl, 0.2% SDS, 1 mM EDTA, and 0.02 M Pipes, pH 6.4, for 6 hr at 37°C. Filters were washed 10 times in 1X SSC, 0.5% SDS for 5 min at 6O”C, 3 times in 2 mM EDTA, pH 8, for 5 min at
VOLUME
125, 1988
room temperature, and once in 2 mM EDTA, pH 8, for 1 min at 60°C. Hybrid-arrested RNA was eluted by boiling in 300 ~1 of 1 mM EDTA, pH 8, for 90 set and ethanol precipitated with 10 pug of yeast tRNA. Hybrid-selected RNA was translated in vitro using a rabbit reticulocyte system (Amersham) in the presence of [35S]methionine. Radiolabeled products were analyzed by one-dimensional electrophoresis through 7.5% SDS acrylamide gels. Nuclear run-on transcription assay. F9 nuclei were prepared for in vitro transcription as described by Camuro and Schibler (1984) with few modifications. Three confluent tissue culture flasks each of F9 stem cells and F9 cells induced by retinoic acid and dibutyryl cyclic AMP were trypsinized, suspended in phosphate-buffered saline, and pelleted at 1000~ for 5 min. Cells were resuspended in 10 ml lysis buffer (Schibler et al., 1983) containing 0.05% NP40 and lysed by repipetting the cell suspensions. Nuclei were pelleted at 1000~ for 4 min and resuspended in 10 ml lysis buffer. Nuclei were then repelleted through a 3.5-ml cushion of 30% (w/w) sucrose in lysis buffer. The pellet was resuspended in 0.2 ml nuclei storage buffer (Schibler et al., 1983) at a concentration of 2 X 10’ nuclei/ml and stored at -70°C until use. All steps were performed at 4°C. Nuclei (2 X 107) were thawed for the run-on transcription assay and incubated for 20 min at 26°C in 0.2 ml of 70 mM KCl, 5 mM MgClz, 2.5 mM DTT, 0.1 mM EDTA, 0.4 mM ATP, GTP, CTP, and 200 &i [cz-~‘P]UTP (600 Ci/mmol, NEN). Radiolabeled RNA was isolated as described by Groudine et al. (1981). Labeled RNA (6 X lo6 cpm) was hybridized to cDNA clones (10 pg per slot) immobilized on nitrocellulose for 4 days at 42°C in 50% formamide, 20 mM phosphate buffer, 5X SSC, 5X Denhardt’s, 0.1% SDS, and 100 cl.g/ml yeast RNA. Filters were washed twice in 2~ SSC at 60°C treated with RNase A (10 pg/ml in 2~ SSC) for 30 min at room temperature, and washed again in 0.1X SSC, 0.1% SDS at room temperature for 30 min. RESULTS
Isolation
and Cloning of the Mouse hsc70 Gene
A F9 stem cell XgtlO cDNA library, kindly provided by M. Kessel, was screened by plaque hybridization (Benton and Davis, 1977) using a probe derived from the human hsc70 coding region (Dworniczak and Mirault, 1987). Twenty-five recombinant phages were isolated, their cDNA inserts were subcloned into bacteriophage vectors Ml3 mp18 or Ml3 mp19, and the nucleotide sequences were determined by the Sanger dideoxy chain termination method (Sanger et al., 1977). None of the 25 cDNA clones contained the 5’ end of the hsc70 mRNA.
GIEBEL,
DWORNICZAK,
Extensive overlaps showed that all cDNA clones were derived from a single hsc70 mRNA species. To isolate and clone the 5’ end of the mature RNA we constructed a F9 cDNA library primed with a synthetic mouse hsc70 oligonucleotide (Fig. 1, position 162-181) and screened with a probe derived from the 5’end of the human hsc70 gene. Six recombinant phages were isolated, each containing cDNA inserts representing the 5’ end of the mRNA. The nucleotide sequence of the hsc70 cDNA is shown in Fig. 1. The gene encodes 646 amino acids as do the human (,Dworniczak and Mirault, 1987) and rat (O’Malley et al., 1985; Sorger and Pelham, 1987) hsc70 genes. Nucleotide sequence comparisons of these genes revealed a homology of 96% between mouse and rat and 89% between mouse and human in the coding region. At the protein level there is one amino acid exchange between mouse and rat at amino acid position 428 and one between mouse and human at position 578. Obviously the hsc70 amino acid sequence has been highly conserved during mammalian evolution. In contrast there is no homology in the 3’ untranslated regions of the human and mouse hsc70 genes except the polyadenylation signal AATAAA. Southern Blot Analysis of the Mouse
Genome
To assess the number of hsc70-related DNA sequences in the mouse genome, DNA isolated from F9 cells was digested with various restriction enzymes, and fragments were separated by agarose gel electrophoresis, transferred to nitrocellulose, and hybridized to a 500-bp nick-translated probe derived from the hsc’70
Rl
H3
PI
-9.Okb
- 3.6 k b -2.5kb
-1.7kb - 1.5k
203
AND BAIJTZ
b
FIG. 2. Southern blot analysis of F9 genomic DNA. Genomic DNA isolated from F9 stem cells was digested with various restriction enzymes, fractionated on an agarose gel, transferred to nitrocellulose by the Southern blot technique, and hybridized to a radiolabeled probe derived from the coding region of a hsc70 cDNA clone (Rl = EcoRI, H3 = HindIII, Pl = PstI).
12
3
/ -2.7kb
FIG. 3. Northern blot analysis of F9 RNA. Poly(A’)-selected RNA (5 pg) isolated from heat-shocked F9 stem cells (lane l), F9 stem cells (lane 2), and F9 cells induced with retinoic acid and dibutyryladenosine 3’4 cyclic monophosphate for 5 days (lane 3) were transferred to nitrocellulose by the Northern blot technique and hybridized to a radiolabeled probe derived from the coding region of a hsc70 cDNA clone.
coding region. The probe did not contain any restriction sites of the enzymes used. Figure 2 shows at least eight distinguishable bands for all three restriction enzymes. No cross hybridization to hsp70 genes could be observed under very stringent conditions (data not shown) using a probe derived from the coding region of a mouse hsp70 gene (kindly provided by R. I. Morimoto). This suggests at least eight hsc70-related sequences in the mouse hsp70 multigene family. Heat Shock Inducibility
qf the hsc70 Gene
We studied the heat inducibility of hsc70 transcription and translation products to ascertain that the gene we isolated is not a bona fide heat shock gene. Northern analysis (Fig. 3) revealed a 2.7-kb transcript which is constitutively expressed in F9 cells and induced only about twofold by heat shock. This was confirmed by Western blot analysis of proteins from F9 and heatshocked F9 cells using a polyclonal anti-human hsp70 antibody. Lanes 1 and 2 of Fig. 4 show a constitutively expressed 72- and a 70-kDa protein which is greatly induced by heat shock. Hybrid Selection and in Vitro Translation of hsc70 mRNA
In order to demonstrate that the DNA clones we isolated encode the 72-kDa protein observed in the Western blot, we selected hsc70 mRNA by hybridization of both F9 and heat-shocked F9 poly(A+) RNA to hsc70 cDNA clones immobilized on nitrocellulose. The [““S]methionine-labeled translation products of this selected RNA species were analyzed by SDS-PAGE (Fig.
204
DEVELOPMENTAL 1
BIOLOGY
"
_--
--
FIG. 4. Western blot analysis of hsc70 and clathrin protein in different mouse tissues and F9 stem cells. Proteins were isolated from F9 stem cells (lane l), heat-shocked F9 stem cells (lane 2), brain (lane 3), and liver (lane 4), separated on SDS-polyacrylamide gels (200 pg of each brain and liver), transferred to nitrocellulose, and analyzed using a polyclonal anti-human hsp70 antibody that cross-reacts with both mouse hsc and mouse hsp’70 protein and a monoclonal antibovine 180-kDa clathrin antibody that cross-reacts with mouse 180-kDa clathrin. The 180-kDa band represents clathrin, the 72-kDa band hsc70, and the ‘IO-kDa band hsp70 protein.
5). Hybridization to RNA of both F9 and heat-shocked F9 cells selected a RNA species, whose translation product comigrates exclusively with the F9 72-kDa protein labeled in ~$0. This result demonstrates that the cDNA clones we isolated in F9 cells encode a constitutively expressed hsc70 protein which is only slightly induced by heat shock. 123
125,198s
Tissue-Specific
234
18Okd-
72kdm 7Okd-
VOLUME
4
hsc70hsp70-
Expression
of hsc70
To examine whether there is a tissue-specific expression of hsc70, we isolated total cellular RNA from various tissues, and equal amounts of each were transferred to nitrocellulose filters by the dot-blot technique. A 500-bp nick-translated probe derived from the coding region was hybridized to the filters, and the relative amounts of hsc70 RNA were determined by densitometric scanning of the autoradiographs. Hybridization conditions were such that cross hybridization to hsp70 RNA was prevented. Figure 6 shows hsc70 constitutively expressed in all tissues tested. The amounts of RNA relative to that in brain were brain, 100%; spleen, 30%; kidney, 70%; testes, 30%; ovaries, 2%; adrenal gland, 90%; intestine, 37%; liver, 12%; tongue, 35%; heart, 8% ; lung, 5% ; pancreas, 50% ; 14-day-old embryo, 6%; and F9 stem cells, 100%. The difference of expression in brain and liver was confirmed at the protein level (Fig. 4). Thus brain and F9 stem cells had higher levels of hsc70 RNA than all other tissues tested. Hsc70 Expression during Differentiation of F9 Stem Cells
Kothary et al. (1987) recently showed that hsp70 is expressed in different extraembryonic tissues from 10.5- to 18.5day-old mouse embryos. They also found hsc70 to be constitutively expressed at all stages during mouse embryogenesis and in all tissues examined with no significant variation. However the earliest stage analyzed was an 8.5-day-old embryo. We used the F9 teratocarcinoma cell model to study the expression of hsc70 during early murine development. Embryonal carcinoma cells, the stem cells of teratocarcinomas, provide a model for studying mamma-
TOllg"e Heart
-
SPlC?Wl
-
Kidney
Brain
Testes Ovaries
LUW Pancreas ElUbL-yCl 14 days old
-
Adrenal --
-
Intestine Liver F9 Stem
FIG. 5. Autoradiogram of [3SS]methionine-labeled in. vitro translation products of RNAs selected by hybrid arrest with hsc70 cDNA clones, Lanes 1 and 2 represent [%]methionine-labeled proteins isolated from F9 stem cells and heat-shocked F9 stem cells, respectively. Lanes 3 and 4 represent [35S]methionine-labeled in vitro translation products of RNA from F9 stem cells and heat-shocked F9 stem cells, respectively, selected by hybrid arrest with hsc70 cDNA clones. The lower highly labeled band, according to the manufacturer’s report, represents a protein or another macromolecule labeled by a mRNAindependent process.
Gland
Cells
Brain Liver
FIG. 6. Abundance of mouse hsc70 RNA in F9 stem cells, mouse embryos, and different adult tissues. Total cellular RNA isolated from F9 stem cells, ll-day-old embryos, and different adult tissues (10 kg each) were transferred to nitrocellulose by the dot-blot technique and hybridized to a radiolabeled probe derived from the hsc70 coding region.
GIEBEL, DWORNICZAK, AND BAUTZ
lian development and differentiation (Martin, 1975, 1980; Bernstine et al., 1973). One of the best-characterized cell lines, the embryonal carcinoma stem cell line F9, differentiates after treatment with retinoic acid and dibutyryl cyclic AMP into parietal endoderm (Strickland and Mahdavi, 1978), an extra embryonic cell, which is generated early in mouse embryogenesis. The phenotype of those parietal endodermal cells is characterized by induction of several genes, e.g., those encoding collagen type IV, plasminogen activator, and laminin (Strickland, 1980; Kurkinen et al., 1983). Concomitantly there is a marked reduction of cellular oncogens ~53 and c-myc mRNAs (Campisi et al., 1984), which reflects an irreversible transition from a tumorigenic state to nondividing, nontumorigenic endodermal cells (Strickland, 1980). As a measure of differentiation we looked for induction of collagen type IV and downregulation of p53 and c-myc expression. Northern analysis (Fig. 3) of poly(A+)-selected F9 RNA before and after induction with retinoic acid and dibutyryl cyclic AMP for 120 hr revealed a slight decrease in the amount of stable hsc70 mRNA. As an internal standard we used the mouse P-actin gene, since its expression remains constant during F9 differentiation (Dony et al., 1985). The relative amount of hsc70 messenger was determined by quantitative densitometric scanning of autoradiographs of RNA dot blots. The amount of stable hsc70 messenger decreases twofold during F9 differentiation. To investigate whether this down-regulation occurs at the transcriptional or post-transcriptional level, we performed a nuclear run-on transcription assay before and after induction of differentiation for 120 hr. In this assay transcripts that are initiated in vivo are faithfully elongated in vitro using 32P-labeled ribonucleoside triphosphates. The hybridization of labeled RNA to complementary DNA immobilized on nitrocellulose is a direct measure of the level of transcription at the time of cell lysis, since new RNA transcripts are not initiated and Palmiter, in vitro (Groudine et al, 1981; McKnight 1979; Marzluff, 1978). Figure 7 shows an autoradiograph of the nuclear run-on assay. To quantitate the amount of labeled run-on transcripts before and after induction of F9 differentiation we performed densitometric scanning of different exposures of these autoradiographs. @Actin was used as the denominator, since Dony et al. (1985) had demonstrated that the transcription rate of fl-actin remains constant throughout F9 differentiation induced by retinoic acid and cyclic AMP. a-Fetoprotein was used as a negative control, because its expression is typical for visceral endoderm and remains inactive throughout parietal endoderm differentiation (Ruoslathi and Sepp%l%, 1979). The results clearly demonstrate a twofold decrease in the rate of hsc70 transcrip-
205
Developmental Regulation of hsc70
12
34
A
1
2
3
4
B
FIG. ‘7. Nuclear run-on analysis of the transcription rate of the hsc70 gene. Autoradiogram of 32P-labeled nuclear run-on transcripts hybridized to cDNA clones of P-a&in (lane l), a-fetoprotein (lane 2), plasmids pBR322 and pUC18 (lane 3), and hsc70 (lane 4) immobilized on nitrocellulose filters. Nuclei were isolated from F9 cells induced with retinoic acid and dihutyryl cyclic AMP for 120 hr (A) and from F9 stem cells (B).
tion after 120 hr of differentiation. This is consistent with the Northern blot data, demonstrating that the down regulation of hsc70 expression during F9 differentiation occurs at the transcriptional level. RNA and protein data of F9 cell differentiation, murine brain, and of 14-day-old embryos suggest that hsc70 is very actively transcribed in early mouse development. Very late in embryogenesis hsc70 is down-regulated in most tissues but brain, which retains the F9 stem cell hsc70 gene activity. Clathrin and hsc70 Expression in Tissues and F’Y Cells
A clue to the physiological role of the hsc70 protein came from the work of Ungewickell (1985) and of Chapell et al. (1986; for review, see Rothman and Schmid, 1986). They demonstrated that bovine and yeast hsp and hsc70 proteins have an ATPase activity which catalyzes the release of clathrin triskelia from coated vesicles. The formation of coated pits and coated vesicles plays an essential role in various endo- and exocytic processes, and in intra- and transcellular transport phenomena between membrane-limited compartments of eukaryotic cells (Goldstein et al., 1979; Pearse and Bretscher, 1981; for review, see Steinman et al., 1983). The coats consist of a polyhedral framework containing polymers of a 180-kDa protein named clathrin (Pearse, 1975). According to these findings we would expect a direct relationship between hsc70 and clathrin protein expression. Figure 4 shows a Western blot analysis of F9 cells and different tissues. In the brain both hsc70 and clathrin proteins are expressed abundantly, while in liver both proteins are expressed at a much lower level. Thus the expression of hsc70 is proportional to that of clathrin. In contrast F9 cells and heat-shocked
206
DEVELOPMENTAL
BIOLOGY
F9 cells express hsc?O at a very high level and clathrin at a level not detectable by this assay. This suggests that at least in F9 cells, which are a model for early mouse development, the uncoating activity of coated vesicles may not be the only function of hsc70 protein. DISCUSSION
To eludicate the function of “cognates” in the mouse hsp70 multigene family we isolated hsc70 cDNA clones, determined their nucleotide sequences, and studied expression and regulation at both the RNA and protein level. In mouse the hsc70 mRNA encodes a 72-kDa protein which has been strongly conserved during mammalian evolution. Compared with human and rat hsc70 genes there is only one amino acid difference at the protein level, and only 11% differences with respect to the human and 4% in respect to the rat coding region at the nucleotide level. Southern blot analysis of the hsp70 multigene family in F9 cells identified at least eight sequences closely related to the hsc’70 gene and clearly distinguishable from their heat inducible counterparts. In human we isolated two hsc70-related sequences (L. Giebel and B. Dworniczak, unpublished results). Analysis at the nucleotide level demonstrated great homology to the human hsc70 gene, but they were pseudogenes, since they contained multiple termination codons in the deduced reading frame. Additionally, these sequences appeared to be perfect DNA copies of mature hsc70 RNA. This fact and the isolation of about 30 identical cDNA clones derived from a single hse70 mRNA species suggest that at least in F9 cells, there may be only one actively transcribed hsc70 gene and all other seven or more related sequences may be pseudogenes arisen by reverse transcription of the mature RNA (this conclusion is supported by the rat hsc70 data of Sorger and Pelham (1987)). The mouse hsc70 gene encodes a 2.7-kb messenger species which is translated into a ‘72-kDa protein. The gene is constitutively expressed in all mouse tissues and mouse cell lines examined and is only induced twofold by heat shock. These results are consistent with recent data of Kothary et al. (1987) who using a rat hsc70 cDNA (O’Malley et ah, 1985) and a mouse hsp70 cDNA (Lowe and Moran, 1986) as probes found constitutive h&i’0 expression in all mouse cells and tissues examined. They also found that hse70 is induced by heat shock in F9 cells. hsc70 RNA exhibits a tissue-specific expression. While expression is extremely high in brain, comparable to that in F9 stem cells, all other tissues express lower amounts of the gene product. To study the expression of hsc70 during early murine development we used the F9 teratocarcinoma cell
VOLUME
125,1988
model. Differentiation of F9 stem cells into parietal endoderm by retinoic acid and dibutyryl cyclic AMP revealed an about twofold decrease in the amount of stable hsc70 mRNA. This down-regulation was due to a reduced transcription rate. On the other hand 14-dayold embryos showed very little hsc70 transcripts, suggesting that, while hsc70 is very active in early mouse development, towards the end of embryogenesis it is down-regulated to various degrees in almost all tissues except for brain which fully retains the F9 stem cell gene activity. In a recent review on the functions of the major heat shock and glucose-regulated proteins (Pelham, 1986) Pelham hypothesized that the function of hsp and hsc70 protein is to disrupt hydrophobic protein aggregates formed by interaction of denatured proteins after heat shock or to prevent false hydrophobic interactions of nascent protein chains to assure proper folding. They would do so by binding tightly to the hydrophobic surfaces, and the energy of ATP hydrolysis would be required to release themselves from their substrates. Support for this idea came from studies of the uneoating of coated vesicles in vitro by an uncoating ATPase, which was shown to be identical wit,h hsp70 and hsc70 protein. Our Western blot data of hsc70 and 180-kDa clathrin (the major coat protein of coated vesicles) expression in mouse tissues demonstrate that tissues with vast numbers of coated vesicles like brain also express hsc70 protein at a very high level, while tissues with small numbers of coated vesicles like liver also exhibit low hsc70 expression. Un~oating of coated vesicles, however, may not be the only function in mouse embryogenesis, since F9 stem cells express hsc?O at a very high level but clathrin at a hardly detectable level. Also, in different tissues there is a 30-fold molar excess of hsc70 protein over clathrin; this is far more than would appear to be necessary to disrupt the clathrin coats. Pelham’s hypothesis that hsc70 recognizes nascent proteins and prevents aggregation problems during protein folding implies that all rapidly growing cells or all cells which secrete large amounts of proteins express hsc7’0 abundantly. Rapidly growing LTK-‘- or SV40-transformed NIH3T3 cells and human Hela or SV80 cells (L. Giebel and B. Dworniczak, unpublished results) express hsc70 at a low level. Moreover, F9 stem cells, after differentiation into nondividing parietal endoderm cells, still express hsc70 protein abundantly. Thus, the function of hsc70 protein, especially in early mouse development, remains rather obscure, and much is to be learned about its mechanism of action and precise physiological role. We thank Dr. M. Kessel Bruder for the monoclonal ported by Grant dw 312-I
for providing a F9 cDNA library, and Dr. G. anticlathrin antibody. This work was supof the Deutsche Forsehungsgemeinschaft.
GIEBEL,
D~ORNICZAK,
AND BAUTZ
REFERENCES W. D., and DAVIS, clones by hybridization 180-182.
BENTON,
R. W. (1977). Screening lgt recombinant to single plaques in situ. Science 196,
E. G., HOOPER, M. L., GRANDCHAMP, S., and EPHRUSSI, B. (1973). Alkaline phosphatase activity in mouse teratocarcinoma. Proc. N&l. Acad. Sci. USA 70, 3899-3903. CAMPISI, J., GRAY, H. E., PARDEE, A. B., DEAN, M., and SONNENSHEIN, G. E. (1984). Cell-cycle control of c-myc but not c-ras expression is lost following chemical transformation. Cell 36, 241-247. CAMURO, M., and SCHIBLER, U. (1984). Accumulation of rare and moderately abundant mRNAs in mouse L-cells is mainly post-transcriptionally regulated. J. Mol. Biol. 178, 869-880. CHAPELL, T. G., WELCH, W. J., SCHLOSSMAN, D. M., PALTER, K. B., SCHLESSINGER, M. J., and ROTHMAN, J. E. (1986). Uncoating ATPase is a member of the 70 kilodalton family of stress proteins. Ce12 45, 3-13. CHIRGWIN, J. M., PRYZYBYLA, A. E., MACDONALD, R. J., and RUTTER, W. J. (1979). Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18,5294-5295. CRAIG, E. A., INGOLIA, T. D., and MANSEAU, L. J. (1983). Expression of Drosophila heat shock cognate genes during heat shock and development. Dec. Biol. 99, 418-426. CRAIG, E. A., and JACOBSEN, K. (1985). Mutations in cognate genes of Saccharomyces cerecisiae hsp70 result in reduced growth rates at low temperatures. Mol. Cell. Bid. 5, 3517-3524. CRAIG, E. A. (1985). The heat shock response. CRCCrit. Re?). Biochem l&239-280. DONY, C., KESSEL, M., and GRUSS, P. (1985). Post-transcriptional control of mgc and p53 expression during differentiation of the embryonal carcinoma cell line F9. Nature (London) 317,636-639. D~ORNICZAK, B. P., and MIRAULT, M. E. (1987). Structure and expression of a human gene coding for a 71 kd heat shock ‘cognate’ protein. h’uc/eir Acids Res. 15, 5181-5197. GOLDSTEIN, J. L., ANDERSON, R. G. W., and BROU’N, M. S. (1979). Coated pits, coated vesicles, and receptor mediated endocytosis.
BERSTINE,
Nature
(LondoxJ
279,
679-685.
GROUDINE, M., PERETZ, M., and WEINTRAUB, H. (1981). Transcriptional regulation of hemoglobin switching in chicken embryos. Mol. Cell. Biol. 1, 281-288. KOTHARY, R., PERRY, M. D., MORAN, L. A., and ROSSANT, J. (1987). Cell lineage-specific expression of the mouse hsp68 gene during embryogenesis. De??. Biol. 121, 342-348. KURKINEN, M., BARLOW, D. P., HELFMAN, D. M., WILLIAMS, J. G., and HOGAN, B. L. M. (1983). Isolation of cDNA clones for basal lamina components: Type IV procollagen. Nucleic Acids Res. 11, 6199-6209. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lnndcmj 227, 680-685. LOWE, D. J., and MORAN, L. A. (1986). Molecular cloning and analysis of DNA complementary to three mouse Mr-68000 heat shock protein mRNAs. J. Biol. Chem. 261, 2102-2112.
207 MANIATIS, T., FRITSCH, E. F., and SAMBROOK, J. (1985). “Molecular Cloning: A Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. MARTIN, G. R. (1975). Teratocarcinamas as a model system for the study of embryogenesis and neoplasia. Cell 5,229-243. MARTIN, G. R. (1980). Teratocarcinomas and mammalian embryogenesis. Science 209, 768-776. MARZLUFF, W. F. (1978). Transcription of RNA in isolated nuclei. Methods Cell Biol. 19, 317-331. MCKNIGHT, G. S., and PALMITER, R. J. (1979). Transcriptional regulation of the ovalalbumin and conalbumin genes by steroid hormones in chick oviduct. J. Biol. Chem. 254, 9050-9058. O’MALLEY, K., MAURON, A., BARCHAS, J. D., and KEDES, L. (1985). Constitutively expressed rat mRNA encoding a 70-kilodalton heatshock-like protein. Mol. Cell. Biol. 5, 3476-3483. PALTER, K. B., WATANABE, M., STINSON, L., MOHAWALD, A. P., and CRAIG, E. A. (1986). Expression and localization of Droso<r melanoyaster hsp70 cognate proteins. Mol. Cell Biol. 6, 1187-1203. PEARSE, B. M. F. (1975). Coated vesicles from pig brain: Purification and biochemical characterization. J. Mol. Biol. 97, 93-98. PEARSE, B. M. F., and BRETSCHER, M. S. (1981). Membrane recycling by coated vesicles. Annu. Rw. Bioch,em. 50, 85-101. PELHAM, H. R. B. (1986). Speculations on the functions of the major heat shock and glucose-regulated proteins. Cell 46,959-961. ROTHMAN, J. E., and SCHMID, S. L. (1986). Enzymatic recycling of clathrin from coated vesicles. Ce11 46, 5-9. RUOSLATHI, E., and SEPP;~L& M. (1979). cu-Fetoprotein in cancer and fetal development. Adv. Cancer Res. 29, 275-346. SANGER, F., NICKLEN, S., and COULSON, A. R. (1977). DNA sequencing with chain termination inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467. SCHIBLER, U., HAGENB~CHLE, O., WELLAUER, P. K., and PITTET, A. C. (1983). Two promoters of different strength control the transcription of the mouse alpha-amylase gene Amy-la in the parotid gland and the liver. Cell 33, 501-508. SORGER, P. K., and PELHAM, H. R. B. (1987). Cloning and expression of a gene encoding hsc73, the major hsp70-like protein in unstressed rat cells. EMBO J. 6, 993-998. SOUTHERN, E. M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98, 503-517. STRICKLAND, S., and MAHDAVI, V. (1978). The induction of differentiation in teratocarcinoma stem cells by retinoic acid. Cell 15, 393-403. STRICKLAND, S., SMITH, K. K., and MAROTTI, K. R. (1980). Hormonal induction of differentiation in teratocarcinoma stem cells: Generation of parietal endoderm by retinoic acid and dibutyryl CAMP. Ceil 21,347-355. STEINMAN, R. M., MELLMAN, I. S., MULLER, W. A., and COHN, Z. A. (1983). Endocytosis and the recycling of plasma membrane. J. Cc2( Biol. 96, l-7. UNGEWICKELL, E. (1985). The 70 kd heat shock proteins are structurally and functionally related to the uncoating protein that releases clathrin triskelia from coated vesicles. EMBO J. 4, 3385-3391.