Isolation, Characterization, and Sequence of an Escherichia coli Heat ...

JOURNAL

OF

BACTERIOLOGY, May 1991,

p.

Vol. 173, No. 9

2944-2953

0021-9193/91/092944-10$02.00/0 Copyright © 1991, American Society for Microbiology

Isolation, Characterization, and Sequence of an Escherichia coli Heat Shock Gene, htpX DANIEL KORNITZER,t* DINA TEFF, SHOSHY ALTUVIA,t AND AMOS B. OPPENHEIM Department of Molecular Genetics, The Hebrew University-Hadassah Medical School, Jerusalem, Israel 91010 Received 30 October 1990/Accepted 27 February 1991

We isolated and characterized a new Escherichia coli gene, htpX. The htpX gene has been localized at min 40.3 on the chromosome. We determined its transcription and translation start site. htpX expresses a 32-kDa protein from a monocistronic transcript; expression of this protein is induced by temperature upshift. htpX is expressed from a or32-dependent promoter and is thus part of the heat shock regulon. Cells carrying a htpX gene disruption grow well at all temperatures and under all conditions tested and have no apparent phenotype. However, cells which overexpress a truncated form of the protein display a higher rate of degradation of puromycyl peptides.

The heat shock response consists in a transient increase, after temperature upshift, in the synthesis of a small subset of proteins, the heat shock proteins (hsps). This response has been observed in all cells so far examined (reviewed in reference 31). In Escherichia coli, temperature upshift, as well as other treatments such as UV irradiation, bacteriophage infection, and overproduction of abnormal proteins, leads to the preferential synthesis of at least 17 hsps (reviewed in reference 38). The central regulator of the heat shock response in E. coli is the heat shock gene-specific RNA polymerase subunit, cr32 (22, 23). Some hsps have been highly conserved throughout evolution. Three E. coli hsps, GroEL, DnaK, and C62.5, are homologous to the mitochondrial HSP58 protein and to the eukaryotic HSP70 and HSP90 protein families, respectively (5, 6, 35). In contrast, the small hsps, although often conserved within the same organism, display little similarity between different organisms (32). GroE is involved in the assembly of multiprotein complexes (15, 20), and DnaK, DnaJ, and GrpE have been shown to participate in the refolding of denatured proteins (14). In addition, strains carrying mutations in these genes are defective in the proteolysis of abnormal proteins (48). The Lon protease, which degrades abnormal proteins in the cell (21, 34), is a heat shock protein as well (16). A common denominator of most conditions that induce the synthesis of the hsps is thought to be the appearance of unfolded proteins in the cell (4, 17, 41). It is possible that the primary role of the heat shock response is to remove unfolded proteins from the cell, either by refolding or by degrading them. The lysogenization process of bacteriophage A involves several phage proteins, such as the CII transcriptional activator, which is required for the initiation of transcription of the repressor and integrase genes (45), and the CIII protein, whose role is to stabilize CII (3, 24). Several host genes, e.g., rnc (2) and hfl (7), are involved in this process as well. We were interested in isolating additional host genes participating in the regulation of phage lysogenization. A screening procedure aimed at isolating such genes from *

Corresponding author.

t Present address: Whitehead Institute, Nine Cambridge Center, Cambridge, MA 02142. t Present address: Laboratory of Molecular Biology, National Cancer Institute, Bethesda, MD 20892. 2944

an E. coli phage library yielded several clones; one of these clones was characterized further and found to carry a new E. coli heat shock gene, designated htpX. Overexpression of a truncated form of the HtpX protein does lead to an increase in the degradation of abnormal proteins. However, disruption of the chromosomal copy of this gene does not affect cell viability, nor does it affect the plaque morphology of an infecting phage. The function of HtpX in the cell therefore remains unknown.

MATERIALS AND METHODS

Strains. The E. coli and phage strains used are described in Table 1. Media. Bacteria were propagated in LB medium containing 40 jig of ampicillin per ml. LBMM is LB plus 0.2% (wt/vol) maltose and 10 mM MgSO4. TM is 10 mM Tris (pH 7.4) plus 10 mM MgSO4. Phages were plated on TB plates. For protein labeling, the cells were grown in M56 minimal medium supplemented with 0.5% (vol/vol) glycerol and 20 ,ug of ampicillin per ml. Plasmid and phage constructions. Standard cloning methods were as described previously (33). XPR6 was isolated from an E. coli library, constructed by introducing fragments from a partial Sau3A digest of the E. coli genome into the BamHI site of the XD69 vector (37). pPR6L contains the 3.5-kb BglII-HindIII fragment of the XPR6 insert cloned in HindIII-BamHI-digested pGEM-3 (Promega Biotec); pPR6M contains the 1.5-kb AccI-HindIII fragment (AccI end filled in with Klenow DNA polymerase) of the XPR6 insert cloned in pGEM-3 digested with HindIII and SmaI; pPR6S contains the 1-kb HindIII-EcoRI fragment of the XPR6 insert cloned in pGEM-3 digested with the same enzymes. Plasmid pPR6S' was built by introducing the BamHI-EcoRI fragment of the XPR6 insert into pGEM-3 digested with the same enzymes; this construct was used to sequence the 70 bp between the BamHI site and the HindIII site upstream of the gene. pPR6Cm was built by introducing the cat gene carried on a PstI fragment into the unique PstI site of pPR6S. The disrupted htpX gene was transferred to phage XPR6 by recombination in the following way. XPR6 was grown on A5039 cells carrying pPR6Cm. The resulting lysate was used to lysogenize a strain carrying pintC57 (12), which expresses the X int gene (this is necessary since the bacterial insert of XPR6 disrupts the phage int gene), and the lysogens were

E. COLI htpX

VOL. 173, 1991 TABLE 1. Bacteria and phage strains Strain

E. coli W3350 JK20000 SG12049

SG20322 A1964 A4832

A4833 A5039 A6006 A6007 A6512 A6515 A6516 A6517 A6518 A6519 A6520 A6535 A6536

Genotype

Source or reference

SuMC4100 slp-l::TnJO C600 cipA82::ATniO

Our collection S. Gottesman S. Gottesman

MC4100 lon-146::ATnJO cps-ll:: Lac-Mu dl SA500 hflAl:TnJO

S. Gottesman

argE his Y thr leu mtl xyl ara lac Y str-31 galK2 htpR+:TnlO valS:TnS supF(Ts) A4832 htpRJ65(Am) K37 lacZ::TnS lacIql W3350 htpXVCm A5039 htpXVCm SG20322 htpXVCm A5039 cipA82::ATnJO A5039 1on-146::A&TniO A5039 slp-l::TnJO A6007 clpA82::A&TniO A6007 lon-146: :ATnJO A6007 slp-i::TnJO A5039 hflAl:TnJO A6007 hflAJ:TnlO

Our collection C. Gross

C. Gross D. Friedman This work This work This work This work This work This work This work This work This work This work This work

Phage A

XAO3 AAO5 XAO7 AAO749 XAO960 XAO965 XD69

c+ papa cI60

c1857 c1857 can-1 cI857 cIIItor862

c1857 cIIIam6ll

XBaml° srI(1-2)A att+ imm2' nin5

Our collection 26 Our collection I. Herskowitz 3 3 37

shn6° XPR6

XPR6Cm

AD69 derivative with E. coli partial Sau3A DNA insert cloned within the int gene XPR6 with chloramphenicol resistance insert within htpX

This work This work

selected for chloramphenicol resistance. These lysogens then pooled and induced with mitomycin C. Phages screened for chloramphenicol resistance and ampicillin sensitivity. The DNA of one Cmr Aps phage, XPR6Cm, was further analyzed with a set of restriction enzymes to confirm that it carries the cat fragment within the htpX coding were were

sequence.

Strain construction. The conjugation and P1 transduction procedures were performed as described by Miller (36). (i) Transfer of the disrupted htpX gene to the bacterial chromosome. In a first stage, the immunity region of XPR6Cm (imm21) was exchanged with the XcI857 immunity region by a regular phage cross (36). W3350 cells were coinfected by both XcI857 and XPR6Cm and UV irradiated for 20 sec. The resulting lysate was then propagated on a strain lysogenic for Ximm21 to select against the XPR6Cm phage. This lysate was used to lysogenize strain A5039/ pintC57 in order to select for XcI857 phages carrying the chloramphenicol resistance marker; chloramphenicol-resistant lysogens were heat induced at 42°C to yield a lysate of the recombinant phage, XcI857PR6Cm. In the second stage, this phage was used to lysogenize strain W3350. In the absence of functional int gene product, the phage is unable to integrate in the specific X attachment site within the bacterial chromosome; stable lysogens are obtained only by homolo-

2945

gous recombination between the phage insert and the bacterial chromosome. These stable lysogens were selected by virtue of the chloramphenicol resistance conferred by the phage. Such cells carry two copies of the htpX gene region, one with the original bacterial htpX gene and one with the phage-carried disrupted gene, and the phage genome inserted between these two copies. The lysogens were grown at 42°C to select for loss of the XcI857 phage, which carries a thermolabile repressor. Cells able to grow at 42°C would have lost, by homologous recombination, one of the htpX copies along with the phage genome. Some 10% of these cells were Cmr, indicating that they were left with the disrupted htpX copy in the cell. This was confirmed by Southern blot analysis (see Fig. 9). The resulting strain,

W3350 htpXVCm, was designated A6006. (ii) Construction of A6007 and A6512. The htpXVCm gene disruption was transferred to A5039 and SG20322, respectively, by P1 transduction from A6006, using the chloramphenicol resistance marker. (iii) Construction of A5039 and A6007 derivatives. A6515 and A6518 were constructed by transduction of clpA82::ATnlO from SG12049; A6516 and A6519 were constructed by transduction of lon-146: :ATnJO (34) from SG20322; A6517 and A6520 were constructed by transduction of slp-i::TnJO from JK20000. A6535 and A6536 were constructed by transduction of hflAI :TnJO from A1964, using the tetracycline resistance marker of TnWO. The strains were tested for transduction of the hflAl mutation linked to the TnWO marker by observing the plaque morphology of a A c+ phage. Analysis of labeled proteins. (i) In vivo. Cells were grown in supplemented M56 minimal medium at 30°C to an optical density of 0.4 at 600 nm. At the indicated times after the temperature shift to 42°C, 0.2-ml aliquots were pulse-labeled for 1 min with 10 ,uCi of [35S]methionine, precipitated with 10% (wt/vol) trichloroacetic acid (TCA), and subjected to gel electrophoresis on a 10 to 26% polyacrylamide gradient gel. The amount of protein synthesis was measured by scanning densitometry of the autoradiogram using a Molecular Dynamics 300A densitometer. (ii) In vitro. DNA of plasmid pPR6L was added to an S30 protein synthesis extract supplied by Amersham as specified by the manufacturer. DNA sequencing. Exonuclease III in conjunction with mung bean nuclease (Stratagene) and restriction enzymes were used to generate a set of nested deletions in the insert of plasmids pPR6S and pPR6M. Primers complementary to the vector were initially used, followed by a set of htpXspecific primers to complete the sequence of both strands. DNA sequence analysis was performed by the dideoxynucleotide chain termination method (44) on supercoiled plasmid DNA prepared as described by Chen and Seeburg (8). mRNA analyses. Total cellular RNA was extracted as described by Salser et al. (43). This method yields DNA-free RNA. (i) Dot blot. The total RNA concentration was determined by measuring the optical density at 260 nm, and decreasing amounts of RNA were bound to nitrocellulose filters (50). As a control, some samples were treated with DNase-free RNase before being loaded on the filters. The filters were probed with the labeled antisense htpX mRNA synthesized from the T7 promoter of HindlIl-digested plasmid pPR6S. (ii) Northern (RNA) blot. The RNA was extracted as described above, glyoxal treated, electrophoresed, and transferred to nitrocellulose as described by Maniatis et al. (33). The antisense probe described above was used.

2946

J. BACTERIOL.

KORNITZER ET AL.

(iii) Primer elongation. An end-labeled oligonucleotide primer complementary to nucleotides 81 to 98 of the htpX sequence (see Fig. 3) was annealed to 5 ,g of the RNA isolated as described above and then elongated with avian myeloblastosis virus reverse transcriptase (Promega Biotec) as described before (1). The extension products were analyzed on a sequencing gel. Extension inhibition by the 30S ribosomal subunit. The toeprint assay was done as described previously (1). Briefly, annealing mixtures contained RNA synthesized in vitro from plasmid pPR6S by SP6 RNA polymerase and an oligonucleotide complementary to nucleotides 81 to 98 of the htpX coding sequence. After annealing the 32P-end-labeled oligonucleotide to the RNA, 1 to 2 ,uM purified E. coli 30S ribosomal subunits (kindly provided by R. Traut) was added to the mixture, followed by fMet-tRNA and 0.5 U of avian myeloblastosis virus reverse transcriptase. The reaction was terminated by the addition of loading dye; the reaction products were analyzed by electrophoresis on a sequencing gel. In vivo protein degradation experiments. The assay was performed as described previously (19). Cells were grown in M56 minimal medium-0.5% glycerol-0.05% yeast extract at 37°C to an optical density at 600 nm of 0.3 to 0.4. Puromycin (Sigma) was added to 80 pug/ml; after 10 min, [3H]leucine (Amersham) was added to 0.5 ,uCi/ml. After an additional 10 min, the cells were collected by filtration on a GF/C filter and washed twice with and resuspended in chase medium (growth medium plus leucine [75 ,ug/ml]). At the indicated times, 0.5-ml aliquots were removed; bovine serum albumin was added to 0.4 mg/ml, and TCA was added to 5% (wt/vol). The aliquots were left on ice overnight and then centrifuged. A 0.25-ml sample of the supernatant was added to 5 ml of Insta-Gel scintillation liquid (Packard) and counted. The amount of radioactivity at time zero was regarded as background and substracted from all other values. The total incorporation of [3H]leucine was measured by counting 0.25 ml of the culture before TCA precipitation. Sequence analyses. The protein hydropathy profile, data base searches, and sequence comparisons were performed by using the University of Wisconsin sequence analysis software package (11). Nucleotide sequence accession number. The GenBank accession number for the htpX sequence is M58470.

RESULTS Isolation of the E. coli DNA fragment carrying htpX. To identify bacterial genes involved in the regulation of the lysogenization process of bacteriophage X, E. coli DNA fragments were cloned into the phage vector XD69 (37). The rationale was that a host gene carried by the phage would be overexpressed upon infection, both because of the higher copy number and because of the fact that a gene cloned in the BamHI site of XD69 would be transcribed from the strong PL promoter. Thus, a gene interfering with the lysis/ lysogeny decision, if located on the phage, could change the plaque morphology due to its overexpression. The E. coli XD69 library was grown on the bacterial strain W3350, and a number of phages forming clear plaques were isolated. One of these isolates, XPR6, was investigated further by subcloning several overlapping fragments of its bacterial insert into a plasmid vector (pGEM-3) to yield plasmids pPR6L, pPR6M, and pPR6S (Fig. lb to d). htpX is a heat shock gene. We found that cells carrying these plasmids form normal-sized colonies at 30 and 37°C. At

a.

x

htpX

PR6

coding

XD69

sequence

K

B H

vector

sequence

E. coli insert

cat gene

R

Bil

H

B 1kb

BiI

b.

pPR6L

c.

pPR6M

d.

pPR6S

RV A

P

R e.

f.

pPR6Cm

P BI R P

IlI2.tILl

v

P,, o

'

R

H -

P

DLMilMMlldl!E

X PR6Cm

FIG. 1. Restriction maps of the various constructs (see text for description). Open boxes indicate XD69 vector sequences; light gray boxes indicate the htpX coding sequence; dark gray boxes indicate the cat gene sequences; heavy lines indicate E. coli chromosomal sequences around the htpX gene. Restriction enzymes: A, AccI; B, BamHI; H, Hindlll; K, KpnI; P, PstI; R, EcoRI; BI, BglI; HI, HpaI; RV, EcoRV.

42°C, however, cells carrying pPR6S formed very small colonies, and no colonies were visible with cells carrying pPR6M or pPR6L. Since at least some of the heat shock genes are required for cell survival at high temperatures (40, 42, 55), we tested whether the heat shock response, i.e., the pattern of synthesis of hsps upon temperature upshift (38), was affected in cells carrying pPR6S or pPR6L. The results of this analysis (Fig. 2) showed that whereas the pattern of synthesis of the major hsps remains essentially unchanged, two new protein bands appear: a 22 kDa band in cells carrying pPR6S, and a 33-kDa band in cells carrying pPR6L. Each of these bands became the major synthesized protein after heat shock: over 30% of the methionine is incorporated in the 33-kDa band at 42°C. It is possible that the unregulated high expression of these proteins is responsible for the loss of cell viability at high temperatures. To test whether the 33-kDa protein band represented a plasmid-encoded protein, an in vitro protein synthesis assay was performed. pPR6L DNA yielded a protein which migrated with the 33-kDa protein band obtained in vivo, supporting this hypothesis (not shown). The insert of pPR6L includes that of pPR6S (Fig. 1); however, pPR6L does not direct the synthesis of the 22-kDa protein synthesized in cells carrying pPR6S. We inferred that this protein is a truncated form of the 33-kDa protein and that only part of the 33-kDa coding sequence is present in pPR6S. We called the protein HtpX and the gene encoding it htpX (for high-temperature production X). Sequence analysis of the htpX gene. The insert of plasmid pPR6M is smaller than that of plasmid pPR6L (Fig. 1), but it directs the synthesis of the same 33-kDa protein (not shown). We sequenced both strands of the insert of pPR6M. The DNA sequence (Fig. 3) reveals only one open reading frame (ORF) long enough to encode a -33-kDa protein. The predicted molecular weight of the derived protein sequence is 31,923, in good agreement with the molecular weight

r,

kD 200-

30 2

42' ,1

2947

E. COLI htpX

VOL. 173, 1991

1

6 10 2030- 303

r

2- 6

,

_-,

10 20 30 30 2

-,

,

GATCCTTATCTGGATGAGACGGTGAATATCGCACTCGATCTGGC

-284

42-

42r1__.

1,

6 10 20 30

97-

68

-240

Hindill GAAGCTTGAAAAAGCCAGACCCGCGGAACAACCCGCTCCCGTCAAGTAATATCAATCG

-180

CACGAAATTGTGCCTGATTTTTTAACAGCGACAAGATGCCGTAAATCAGATGCTACAA

-120

AATGTAAAGTTGTGTCTTTCTGGTGACTTACGCACTATCCAGA=ITMAATAGTCGCGT

-60

AACCCATACGATGTGGGTATCGCATATTGCGTTTTGTTAAACTGAGGTAAAMQAAAATT

0

-10SD ATGATGCGAATCGCGCTCTTCCTGCTAACGAACCTGGCCGTAATGGTCGTTTTCGGGCTG

M

M

R

I

A

L

F

L

L

T

N

L

A

V

M

V

V

F

G

L

43 26

60

GTACTGAGCCTGACAGGGATACAGTCGAGCAGCGTTCAGGGGCTGATGATCATGGCCTTG V L S L T G I O S S S V 0 G L M I M A L

120

CTGTTCGGTTTTGGTGGTTCCTTCGTTTCGCTTCTGATGTCCAAATGGATGGCATTACGA L

F

G

F

G

G

S

F

V

S

L

L

M

S

K

W

M

A

L

R

180

TCTGTTGGCGGGGAAGTGATCGAGCAACCGCGTAACGAAAGGGAACGTTGGCTGGTCAAT S V G G E V I E Q P R N E R E R W L V N

240

ACTGTAGCAACCCAGGCTCGTCAGGCGGGGATCGCTATGCCGCAAGTGGCTATCTACCAT O V A I Y H T V A T O A R 0 A G I A M P

18-

300

GCGCCGGACATCAACGCTTTTGCAACCGGTGCGCGCCGTGATGCCTCTCTGGTTGCTGTC P D I N A F A T G A R R D A S L V A V

A

pGEM-3

pPR6S

FIG. 2. Synthesis of proteins following temperature shift in cells carrying various htpX plasmids. Cells carrying the plasmids indicated at the bottom were labeled at 30°C or at various times (in minutes) after transfer to 42°C (indicated above the gel). The proteins were run on a 10 to 26% polyacrylamide gradient gel. The relative migration of proteins of known molecular size is indicated on the left. The two arrows on the right indicate the 33-kDa protein synthesized by cells carrying pPR6L (upper arrow) and the 22-kDa protein synthesized by cells carrying pPR6S (lower arrow).

Pat

360

AGCACCGGTTTGCTGAAACATGAGCCCGGATGAAGCCGAGGCGGTAATTGCTCACGAA S T G L L 0 N M S P D E A E A V I A H E

420

ATCAGCCACATCGCCAATGGTGATATGGTCACCATGACGCTGATTCAGGGCGTGGTGAAC I S H I A N G D M V T M T L I 0 G V V N

480

ACCTTCGTTATCTTTATTTCCCGTATTCTGGCGCAGCTTGCCGCGGGTTTTATGGGCGGA L A A G F M G G T F V I F I S R I L A O

540

AATCGTGATGAAGGTGAAGAGAGCAACGGCAACCCGCTGATCTACTTTGCGGTTGCAACG N R D E G E E S N G N P L I Y F A V A T

pPR6L

600

GTTCTGGAACTGGTGTTTGGTATTCTGGCGAGCATTATCACCATGTGGTTCTCGCGTCAT L E L V F G I L A S I I T M W F S R H _ Bgl I EcoRI CGT CATGCTGATGCCGGTTCGGCAAAACTGGTTGGTCGCGAGP AATGATTGCC R E F H A D A G S A K L V G R E K M I A

V

660

720

PstI

GCGCTGCACGCCTGAAAACCAGCTATGAACCGCAAGAAGCAACCAGCATGATGGCTCTC A

estimated from gel electrophoresis (Fig. 2). Approximately 60 nucleotides upstream to the ORF, there is a sequence similar to the Cu32-specific promoters consensus sequence (38; Table 2): CTTGAA at -35 (identical to the consensus sequence) and ACCCAT at -10 (versus CCCCAT for the consensus sequence). The spacing between these two elements in the htpX promoter, 12 nucleotides, is slightly shorter than the spacing in the other heat shock promoters (which ranges from 13 to 15 nucleotides [Table 2]), whereas the distance between the -10 region and the transcription start site of htpX is 1 to 2 nucleotides longer than this distance in the other heat shock promoters. The DNA sequence suggests that HtpX is expressed from a monocistronic transcript. Two inverted repeats followed by a row of thymidines are present, one before and one after the htpX ORF. These sequences could represent rho-independent transcription terminators, the first one of the gene preceding htpX and the second one of htpX itself. A weak ribosome-binding site consensus sequence (AAG) is present before the ORF. A primer extension inhibition (toeprint) experiment demonstrated that this sequence effectively binds ribosomes. In this experiment, purified 30S ribosomal subunit, bound in vitro to the mRNA, inhibits primer elongation with reverse transcriptase; the primer elongation is usually found to terminate 15 nucleotides downstream of the initiation codon (1, 53). 30S ribosomes were found to bind effectively to in vitro-synthesized htpX mRNA, as indicated by the ribosome-induced transcription termination (Fig. 4). The site of termination is in agreement with the assignment of the ribosome-binding sequence (Fig. 3) and allows us to determine that the first of the two consecutive AUG codons at the beginning of the htpX ORF serves as the initiation codon (Fig. 4). In plasmid pPR6S the ORF is truncated at the EcoRI site, yielding an ORF of 236 amino acids (223 from the insert and 13 from the vector sequence). The predicted molecular

L

OR

L

K

T

S

Y

E

P

O

E

A

T

S

M

M

A

L

780

TGCATTAACGGTAAGTCGAAATCGCTCAGTGAGTTGTTCATGACCCACCCGCCGCTGGAT C I N G K S K S L S E L F M T H P P L D

840

AAACGAATTGAAGCTCTGCGTACGGGTGAATACCTGAAGTAATATGAAGTAAATACTA= K

900

R

I

E

A

L

R

T

G

E

Y

L

K

end

AAGC CGTCCTGATCGAmTTTTTATGCCCTGGATCG

939

FIG. 3. Sequence of htpX. The derived protein sequence is indicated under the nucleotide sequence. Numbering of the sequence begins at the initiation codon. Inverted repeats before and after the htpX coding sequence are indicated by head-to-head arrows. The -35 and -10 consensus sequences are indicated, as well as the Shine-Dalgarno sequence (SD). The transcription initiation site as determined in Fig. 8 is indicated with an arrow. The recognition sites of the restriction enzymes appearing in Fig. 1 are indicated above the sequence.

weight of the resulting protein is 25,656, which accounts for the 22-kDa protein band visible in cells carrying this plasmid (Fig. 2). htpX mRNA size analysis. The distance between the putative promoter and terminator sequences is about 970 nucleotides. The mRNA from cells carrying the vector plasmid, pPR6S or pPR6L, was electrophoresed on an agarose gel and transferred to a nitrocellulose filter. The htpX mRNA was then visualized with an RNA antisense probe (Fig. 5). Cells carrying the control plasmid (lane 1) or pPR6L (lane 3) display a single band of about 1 kb, i.e., in good agreement with the predicted size of 970 nucleotides. As expected, the amount of htpX mRNA is much lower in cells carrying only the chromosomal copy of the gene (lane 1) than in cells carrying the gene on a multicopy plasmid (lane 3). In cells carrying pPR6S, a set of bands of 1 kb and higher are visible (lane 2). This is probably due to the fact that pPR6S lacks the natural terminator of htpX; instead, the transcription appears to terminate at various, probably weaker, transcription terminators within the vector. htpX mRNA synthesis is enhanced after heat shock induc-

2948

KORNITZER ET AL.

J. BACTERIOL.

TABLE 2. Comparison of the htpX promoter region sequence with sequences of other known heat shock promotersa Gene

groE

dnaKpl

dnaKp2

htpGpl rpoDphS lonb htpX Consensus

-35 region

13- to 15-bp region

TTTCCCCCTTGAA TCTCCCCCTTGAT TTGGGCAGTTGAA GCTCTCGCTTGAA TGCCACCCTTGAA TCTCGGCGTTGAA ATCCAGACTTGAA T tC CcCTTGAA

GGGGCGAAGCCTCAT GACGTGGTTTACGA ACCAGACGTTTCG ATTATTCTCCCTTGT AAACTGTCGATGTGG TGTGGGGGAAACAT AATAGTCGCGTA

-10 region

+1

CCCCATTTCTCTGG TCAC CCCCATTTAGTAG TCAA CCCCTATTACAGA CTCAC CCCCATCTCTCCC ACATC GACGATATAGCAG ATAA CCCCATATACTGACG TAC ACCCATACGATGTGGGTAT CCCCATtTA

a The promoter sequences and the derived consensus sequence were taken from Neidhardt and VanBogelen (38) except for the groE sequence, which was taken from Cowing and Gross (10), and the htpX sequence (this work). b The lon transcription start site has not yet been determined.

tion. The expression of the heat shock genes is only transiently induced after a temperature shift; in E. coli, heat shock mRNA synthesis reaches a peak approximately 5 min after the temperature shift and decreases gradually afterwards until it again reaches pre-heat shock levels or a new plateau specific for the new growth temperature (reviewed in t RNA - + ACG

T

reference 38). The observation that the level of HtpX synthesis does not decrease after the initial rise subsequent to the temperature shift (Fig. 2) prompted us to assay the mRNA levels at several times after the temperature shift. The htpX mRNA was quantitated by dot blot analysis and scintillation counting of the filter. The results (Fig. 6A) show that the chromosomal copy of the gene, after an initial 10-fold induction, undergoes a shutoff and reaches a level of only 1.5 times the pre-heat shock level within 30 min. In contrast, when the gene is on a multicopy plasmid, pPR6L (Fig. 6B), the mRNA level barely decreases after the initial rise, and even 30 min after the temperature shift it is still sixfold higher than the pre-heat shock level. This observation is consistent with the prolonged high level of HtpX protein synthesis observed in Fig. 2. We do not know whether this phenomenon is general to plasmid-encoded heat shock genes or whether it is particular to htpX. htpX is a32 regulated. The gel depicted in Fig. 2 demonstrated that the expression of htpX is heat induced. Since, besides &32, another heat-shock-specific a factor has recently been isolated in E. coli (U24; 13, 52), we set out to determine whether htpX is under the control of x32. We used a strain which carries an amber mutation within the gene encoding a32, htpRJ65(Am), and a temperature-sensitive amber suppressor, supF(Ts) (54). In this strain, the expression of u32-regulated genes is expected to be about normal at 30°C but strongly reduced at 42°C. The total cellular RNA from this strain and from the isogenic litpR+ strain was isolated before and 6 min after temperature upshift, and the htpX-specific mRNA was quantitated by dot blot analysis

let

1.8

1 2 3 -

1.3-

0.805

FIG. 4. Toeprint assay. In vitro-synthesized htpX mRNA was bound to purified 30S ribosomal subunits in the presence (+) or absence (-) of fMet-tRNA, annealed to a specific primer, and used as template for primer elongation. The reaction products were run on a sequencing gel next to a sequencing reaction performed with the same oligonucleotide primer. A, C, G, and T indicate the nucleotides complementary to the sequenced strand; i.e., they indicate the sequence of the sense strand depicted in Fig. 3. The arrow indicates the site of primer extension termination induced by the presence of the bound ribosomal subunit.

wI

-

FIG. 5. Northern blot of htpX mRNA. RNA was extracted from cells grown at 37°C and carrying the vector plasmid pGEM-3 alone (lane 1), pPR6S (lane 2), or pPR6L (lane 3). The migration of RNAs of known size is indicated on the left.

VOL. 173, 1991

E. COLI htpX

2949

A

86

z

m

62

xa.

|

100

\

20

.

0

0

*

10

30

0 10 20 30 Time after heat-shock induction (min) FIG. 6. htpX mRNA levels after heat shock induction. RNA from E. coli cells carrying no plasmid (A) or carrying pPR6L (B) was extracted and subjected to dot blot analysis as for Fig. 7. The radioactivity was counted and plotted as a function of the time after shift from 20 Time after heat-shock induction (min)

30 to 42°C. The results are normalized to the amounts found in cells carrying no plasmid, at 30°C (=1).

(Fig. 7) and by Northern blot analysis (not shown). The dot blot experiment shows that heat shock induction of the mRNA is strongly reduced in the htpRJ65(Am) supF(Ts) strain (2-fold induction versus 15-fold in the control strain), as would be expected from an HtpR-regulated gene. The residual induction observed in the mutant strain could be due to stabilization of u32 present in the cells before the heat induction; indeed, the stabilization of the very labile or32 protein, and not necessarily its de novo synthesis, is thought to be a major mechanism of the induction of the heat shock proteins (47, 51). In summary, this experiment demonstrates that htpX expression is induced by the u32 heat shock transcription factor.

b

a

A _

RNA

2p9g

0.5pg

+

*

_

+

*

i

0.25pg 151B

[

-

+

4)

(30C)

(42'C)

10 z E

5-

CC 0O

a b FIG. 7. Quantitation of htpX mRNA in cells defective in the htpR gene. (A) Dot blot analysis of RNA extracted before (-) and 6 min after (+) temperature shift from 30 to 42°C from htpR+ cells (lanes a) and from htpR165(Am) supF(Ts) cells (lanes b), using an htpX-specific probe. (B) Radioactivity of dots was counted by scintillation and plotted. The results are normalized to the amount found in htpR+ cells at 30°C (=1).

We assigned the position of the htpX promoter on the basis of the similarity to known cr32-regulated promoters (Fig. 3; Table 2). To confirm this assignment, we determined the position of the 5' end of the htpX mRNA by a primer extension experiment. Total cellular RNA isolated from pPR6L-carrying htpR+ and htpRl65(Am) supF(Ts) cells, before and after heat shock induction, was used as a template for primer elongation with reverse transcriptase, using an htpX-specific primer. The results (Fig. 8) show that the position of the 5' end of the mRNA is as expected from the position of the consensus a32 promoter sequence. This 5' end therefore probably represents the site of transcription initiation. The relative band intensities before and after heat induction in the wild-type and mutant cells were as observed with the mRNA from the chromosomal copy of the gene (Fig. 7). No other bands, which could have resulted from the existence of additional promoters or from processing of the mRNA, could be detected on the gel. Disruption of the chromosomal htpX gene. We wanted to (i) determine whether htpX is responsible for the clear-plaque phenotype of XPR6 and (ii) gain insight int6 the function of the HtpX protein in the cell. Therefore, the coding region of htpX was disrupted by a DNA fragment carrying the chloramphenicol acetyltransferase gene, cat. A PstI fragment carrying the cat gene was first inserted into the PstI site of plasmid pPR6S, yielding pPR6Cm (Fig. le). This gene disruption was transferred to the original phage clone (XPR6) by phagemid recombination (Materials and Methods) to yield XPR6Cm (Fig. lf). Phage XPR6Cm, carrying the disrupted copy of htpX, forms clear plaques, indicating that the phenotype of XPR6 is due not to the htpX gene but probably is due to the presence of another gene on the 8-kb-long insert. The gene disruption was then transferred from the phage to the bacterial chromosome by recombination to yield strain htpXVCm as described in Materials and Methods. The structure of the disrupted gene was confirmed by Southern blot analysis. Using the insert of plasmid pPR6S as a probe, a change in the restriction pattern of strain htpXVCm (Fig. 9, lanes b) in comparison with the parental strain (lanes a) could be observed. An increase of about 2 kb was observed in the BamHI and BglI restriction fragment size, as expected from the size of the cat fragment. Digestion with EcoRI yields two fragments in the htpXVCm DNA (R+H and R,

2950

J. BACTERIOL.

KORNITZER ET AL.

a b

R+H R

ACG T

B

BI

a b a b a b a b kb -9.4 -6.6 -4.4

-2.0 -1.4 -0.9

FIG. 9. Southern blot analysis of the htpX gene disruption. Lanes: a, DNA from the parental strain, A5039; b, DNA from strain htpXVCm. Enzymes used: B, BamHI; BI, BglI; H, HindIII; R, EcoRI. The relative migration of some of the X DNA HindlIl fragments used as markers is indicated on the right.

FIG. 8. Determination of htpX transcript start point. Primer elongation using RNA extracted before (-) and 6 min after (+) temperature upshift from pPR6L-carrying htpR+ cells (lanes a) and htpRJ65(Am) supF(Ts) cells (lanes b) as the template and an htpX-specific end-labeled primer. The reaction products were run on a sequencing gel next to a sequencing reaction performed with the same end-labeled primer and using pPR6S DNA as the template. A, C, G, and T indicate the nucleotides of the sequenced strand, i.e., the antisense strand.

lanes b), as expected from the fact that the cat sequence carries an EcoRI recognition site. The htpXVCm strain, carrying the disrupted htpX, has no observable phenotype. It grows as well as the parental strain at all temperatures (i.e., up to 47°C); it grows on minimal medium; no restrictive carbon source could be found; it is not UV hypersensitive; lambdoid phages grow well on htpXVCm and lysogenize it efficiently (data not shown). Mapping of htpX on the E. coli chromosome. We used the Kohara sequential phage library (29) to determine the location of htpX, using a probe made from the insert of plasmid pPR6S. We found a single clone, clone 335 of the miniset, that hybridized to our probe. This clone corresponds to the region around nucleotide 1,930,000 of the E. coli map published by Kohara et al. (29), or min 40.3 on the genetic map. The restriction map derived from the analysis of the insert of XPR6 (Fig. 1) and the restriction map derived from the Southern blot analysis (Fig. 9) are in good agreement with this part of the map of Kohara et al. We also determined the map position of htpX by means of Hfr recombination and P1 transduction, using the htpXVCm

strain and a sequentially ordered antibiotic resistance marker strain library (46). Hfr mapping confirmed the location of htpX between min 36 and 45 of the genetic map. We then transduced htpXVCm to strain CAG12122 (zea-3125:: TnJOKan; 46) using phage P1, thus obtaining a strain (CG12122 htpXVCm) which carries a TnJOKan marker at min 40.25 and a chloramphenicol resistance marker within the htpX gene. A P1 lysate was grown on this strain and used to transduce a kanamycin-sensitive, chloramphenicol-sensitive strain. The transductants were selected for kanamycin resistance and then screened for chloramphenicol resistance. All of the seven transductants carried both resistances, demonstrating that the two markers are closely linked. From all of these experiments, we concluded that htpX maps close to min 40.25 on the E. coli genetic map. Effect of the truncated HtpX protein on the rate of lysogenization of A. We found that X forms clear plaques on a strain carrying pPR6S but not on a strain carrying the larger insert, pPR6L. We quantitated the effect of pPR6S on phage growth by measuring the frequency of lysogenization of cells carrying this plasmid. The cells were infected with a wildtype phage, a clll-overexpressing mutant (tor862), and a mutant defective in clII (c111611). The frequency of lysogenization of all three phages was much lower in cells carrying pPR6S than in the control strain carrying the vector plasmid pGEM-3 (Table 3), indicating that the presence of pPR6S in the cells interferes with the lysogenization process. We screened additional known phage mutants and found that the plaque morphology of a A can-1 phage is unchanged in cells carrying pPR6S. X can-1 carries a mutation that stabilizes the CII protein from proteolytic decay (25), suggesting that the presence of pPR6S in the cells may affect lysogenization through an effect on the stability of the CII protein. Presence of the truncated HtpX protein in the cell enhances proteolysis. The fact that the truncated htpX present on pPR6S reduced the rate of lysogenization of several phages

VOL. 173,

1991

E. COLI htpX

TABLE 3. Frequency of lysogenization of XcIII wild-type and mutants on W3350 cells carrying pPR6Sa Plasmid

Phage

% Survivors

% Nonimmune

% Lysogens

pGEM-3

XcI857 XcI857tor862 XcI857cIII611 XcI857 XcI857tor862 XcI857cIII611

59 84 22 13 15 12

3.5 1.7 0.5 3.9 3.5 6.5

55 82 22 9 12 6

pPR6S

a The cells were grown in LBMM medium at 30'C at an optical density of 0.3 at 600 nm, centrifuged, resuspended in TM, and infected at a multiplicity of infection of 3.

but did not affect mutant can-1, which expresses a more stable CII protein, led us to test whether the cloned htpX would affect the rate of general proteolysis. We compared the rate of proteolysis of puromycyl peptides (18) in cells carrying either pPR6S or pPR6L with the rate in cells carrying the vector plasmid pGEM-3. The antibiotic puromycin is incorporated in the growing polypeptide chain and induces its premature release from the ribosome, yielding short, unstable polypeptides. To measure the rate of puromycyl peptide degradation, cells were exposed to puromycin and [3H]leucine; the label was then chased by washing the cells and resuspending them in medium containing unlabeled leucine (Materials and Methods). The results of these experiments (Fig. 1OA) show that the rate of proteolysis is indeed strongly enhanced in cells carrying pPR6S, which express the truncated form of HtpX: 24% of the labeled amino acid incorporated into the proteins in the presence of puromycin became TCA soluble after 90 min, versus only 14% in the control cells. The presence of pPR6L, directing the synthesis of the full-length HtpX in the cells, increased proteolysis only slightly, to 16% TCA-soluble radioactivity after 90 min. In the same set of experiments, we tested whether cells carrying the htpXVCm gene disruption display reduced proteolysis. As can be seen in Fig. 10, the htpXVCm strain displays the same kinetics of puromycyl peptide degradation as the parental strain, indicating that the absence of HtpX did not affect the rate of general proteolysis. We further asked whether the enhanced proteolysis in0

A

pPR6S -0-

pPR6L pGEM-3

-0-

htpX+ htpX-Cm

-0-

0-

I.0 -o 0CZ

0

c) B

a))

D6

-_

pPR6S,

I

0

20

40

60

puromycin puromycin

80

Time (min.) FIG. 10. Rate of protein degradation in cells carrying various htpX plasmids or the htpXVCm gene disruption. Cells were pulselabeled with radioactive leucine either in the presence (curves A) or in the absence (curves B) of puromycin. The cells were washed free of the radioactive amino acid, and the kinetics of reappearance of TCA-soluble radioactivity was measured. Each curve represents the average of at least two independent experiments.

2951

duced by pPR6S is limited to the puromycyl peptides or whether normal proteins are also more rapidly degraded in these cells. The same experiment was performed as before, but puromycin was omitted from the medium. The results (Fig. 10B) show that the rate of degradation is similar in cells carrying pPR6S and in cells carrying the control plasmid, pGEM-3. Therefore, the proteolytic activity associated with the truncated form of HtpX seems to be specific for abnormal proteins. DISCUSSION We have isolated and characterized an E. coli heat shock gene, htpX. This gene directs the synthesis of a 32-kDa protein, HtpX; the synthesis of the protein is greatly enhanced upon shift to a higher temperature. The size of the mRNA as determined by Northern blot indicates that HtpX is expressed from a monocistronic transcript. By measuring the htpX mRNA levels in an htpR amber mutant, we determined that htpX is under the regulation of the cr32 factor. The site of transcription initiation was determined by primer extension; the htpX promoter shows high similarity to known heat shock promoters. A strain in which the htpX coding sequence was disrupted is viable under all conditions tested, including high temperature; htpX therefore belongs to the group of the nonessential E. coli heat shock genes, such as Ion (34). Comparison of the htpX DNA and derived amino acid sequence with a sequence data base revealed no significant homologies. According to its size, HtpX is part of the small hsp subgroup, of which the eukaryotic representatives show very little conservation (32). Determination of the hydropathy profile, which reveals some similarities among the eukaryotic small hsps, allowing them to be classified into one superfamily (32, 49), failed to uncover similarities with HtpX. Analysis of the hydropathy profile of the HtpX sequence according to Kyte and Doolittle (30), however, suggests that it is an integral membrane protein with four transmembrane segments, including a putative signal sequence at the amino terminus. Indeed, in this analysis, when a window of 20 residues is used, a segment is predicted to cross the membrane whenever the hydrophobicity value reaches 1.6 or higher (28, 30); four segments of HtpX, around residues 13, 43, 167, and 205, fulfill this condition. It should further be noted that the amino end of the protein resembles the signal peptide consensus sequence, with a basic residue at the amino end (Arg-3) and a hydrophobic core of 15 residues, followed by the sequence Val-17-X-gly19; if indeed this sequence functions as a signal peptide, the cleavage by the signal peptidase would be expected to occur after this last residue (39). In fact, upon HtpX overexpression, a protein band does appear which, according to its relative migration in the gel, may represent the processed form of the protein (Fig. 2, pPR6L). Further experiments will be needed to prove this hypothesis. Overexpression of the truncated HtpX enhances the rate of degradation of puromycyl peptides. However, neither overexpression of the full-length HtpX nor insertional inactivation of htpX was found to have a measurable effect on the rate of puromycyl peptide degradation. In addition, we found that the viability of cells carrying an insertional inactivation of htpX is unaffected even when the htpXVCm allele is combined with the mutant allele of a set of known protease genes: Ion (34), cipA (27), slp (20a), or hflA (7, 9) (not shown). Since the truncated HtpX is extremely unstable, with a half-life of less than 1 min (not shown), the

2952

KORNITZER ET AL.

simplest hypothesis is that the mere overproduction of this rapidly degraded protein results in the induction of cellular proteolytic systems and that this effect is unrelated to the normal function of HtpX. Although htpX is part of the heat shock regulon, cells in which the chromosomal copy of htpX was inactivated by a gene insertion do not display any particular phenotype, even at an elevated temperature. One possible explanation for this lack of phenotype is that the effects of the inactivation of HtpX are too subtle to be detected in our experiments. Another possibility is that HtpX is required for cell viability, but only under a particular set of conditions, yet to be discovered. Indeed, a bacterial cell can be exposed in nature to a wide range of stress conditions, only part of which are usually being reproduced in laboratory experiments. ACKNOWLEDGMENTS

We thank Hilla Giladi, Max Gottesman, and Ariella Oppenheim for critical reading of the manuscript, R. Traut for supplying purified 30S ribosomal subunits, Susan Gottesman for sending bacterial strains, Carol Gross for sending the strain library, and Y. Kohara for sending the sequential phage library. This work was supported by a grant from the National Council for Research and Development, Israel, the Gesellschaft fur Biotechnologische Forschung, mbH, Braunschweig, Germany, and grant GM38694 from the National Institutes of Health. Part of this work was performed in the Irene and Davide Sala Laboratory for Molecular Genetics. REFERENCES 1. Altuvia, S., D. Kornitzer, D. Teff, and A. B. Oppenheim. 1989. Alternative mRNA structures of the cIII gene of bacteriophage determine the rate of its translation initiation. J. Mol. Biol. 210:265-280. 2. Altuvia, S., H. Locker-Giladi, S. Koby, 0. Ben-Nun, and A. B. Oppenheim. 1987. RNaseIII stimulates the translation of the cIII gene of bacteriophage X. Proc. Natl. Acad. Sci. USA 84:65116515. 3. Altuvia, S., and A. B. Oppenheim. 1986. Translational regulatory signals within the coding region of the bacteriophage X cIII gene. J. Bacteriol. 167:415-419. 4. Ananthan, J., A. L. Goldberg, and R. Voelimy. 1986. Abnormal proteins serve as eukaryotic stress signals and trigger the activation of heat-shock genes. Science 232:522-524. 5. Bardwell, J. C. A., and E. A. Craig. 1984. Major heat shock gene of Drosophila and the Escherichia coli heat-inducible dnaK gene are homologous. Proc. Natl. Acad. Sci. USA 81:848-852. 6. Bardwell, J. C. A., and E. A. Craig. 1987. Eukaryotic Mr 83,000 heat shock protein has a homologue in Escherichia coli. Proc. Natl. Acad. Sci. USA 84:5177-5181. 7. Belfort, M., and D. L. Wulff. 1973. An analysis of the processes of infection and induction of E. coli mutant hfl-1 by bacteriophage lambda. Virology 55:183-192. 8. Chen, E. Y., and P. H. Seeburg. 1985. Supercoil sequencing: a fast and simple method for sequencing plasmid DNA. DNA 4:165-170. 9. Cheng, H. H., P. J. Muhlrad, M. A. Hoyt, and H. Echols. 1988. Cleavage of the cII protein of phage X by purified HflA protease: control of the switch between lysis and lysogeny. Proc. Natl. Acad. Sci. USA 85:7882-7886. 10. Cowing, D. A., and C. A. Gross. 1989. Interaction of Escherichia coli RNA polymerase holoenzyme containing -32 with heat shock promoters. J. Mol. Biol. 210:513-520. 11. Devereux, J., P. Haeberli, and 0. Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395. 12. Enquist, L., A. Honigman, S.-L. Hu, and W. Szybalski. 1979. Expression of X int gene function in ColEl hybrid plasmids carrying the C fragment of bacteriophage X. Virology 92:557560.

J. BACTERIOL.

13. Erickson, J. W., and C. A. Gross. 1989. Identification of the orE subunit of Escherichia coli RNA polymerase: a second alternate r factor involved in high-temperature gene expression. Genes Dev. 3:1462-1471. 14. Gaitanaris, G. A., A. G. Papavassiliou, P. Rubock, S. J. Silverstein, and M. E. Gottesman. 1990. Renaturation of denatured X repressor requires heat shock proteins. Cell 61:1013-1020. 15. Georgopoulos, C. P., R. W. Hendrix, A. D. Kaiser, and W. B. Wood. 1972. Role of the host cell in bacteriophage morphogenesis: effects of a bacterial mutation on T4 head assembly. Nature (London) New Biol. 239:38-41. 16. Goff, S. A., L. P. Casson, and A. L. Goldberg. 1984. The heat shock regulatory gene, htpR, influences rates of protein degradation and expression of the lon gene in Escherichia coli. Proc. Natl. Acad. Sci. USA 81:6647-6651. 17. Goff, S. A., and A. L. Goldberg. 1985. Production of abnormal proteins in E. coli stimulates transcription of lon and other heat shock genes. Cell 41:587-595. 18. Goldberg, A. L. 1972. Degradation of abnormal proteins in Escherichia coli. Proc. Natl. Acad. Sci. USA 69:422-426. 19. Goldberg, A. L., K. H. Sreedhara Swamy, C. H. Chung, and F. S. Larimore. 1983. Proteases of Escherichia coli. Methods Enzymol. 80:680-702. 20. Goloubinoff, P., A. A. Gatenby, and G. E. Lorimer. 1989. GroE heat-shock proteins promote assembly of foreign prokaryotic ribulose biphosphate carboxylase oligomers in Escherichia coli. Nature (London) 337:4 -47. 20a.Gottesman, S. Personal communication. 21. Gottesman, S., M. E. Gottesman, J. E. Shaw, and M. L. Pearson. 1981. Protein degradation in E. coli: the Ion mutation and bacteriophage X N and clI protein stability. Cell 24:225-235. 22. Grossman, A. D., J. W. Erickson, and C. A. Gross. 1984. The htpR gene product of E. coli is a sigma factor for heat-shock promoters. Cell 38:383-390. 23. Grossman, A. D., D. B. Straus, W. A. Walter, and C. A. Gross. 1987. o32 synthesis can regulate the synthesis of heat shock proteins in Escherichia coli. Genes Dev. 1:179-184. 24. Hoyt, M. A., D. M. Knight, A. Das, H. I. Mifler, and H. Echols. 1982. Control of phage X development by stability and synthesis of cII protein: role of the viral cIII and host hflA, himA and himD genes. Cell 31:565-573. 25. Jones, M. O., and I. Herskowitz. 1978. Mutants of bacteriophage X which do not require the cIII gene for efficient lysogenization. Virology 88:199-212. 26. Kaiser, A. D. 1957. Mutations in a temperate bacteriophage affecting its ability to lysogenize Escherichia coli. Virology 3:42-61. 27. Katayama, Y., S. Gottesman, J. Pumphrey, S. Rudikoff, W. P. Clark, and M. R. Maurizi. 1988. The two-component, ATPdependent Clp protease of Escherichia coli. J. Biol. Chem. 263:15226-15236. 28. Klein, P., M. Kanehisa, and C. DeLisi. 1985. The detection and classification of membrane-spanning proteins. Biochim. Biophys. Acta 815:468-476. 29. Kohara, Y., K. Akiyama, and K. Isono. 1987. The physical map of the whole E. coli chromosome: application of a new strategy for rapid analysis and sorting of a large genomic library. Cell 50:495-508. 30. Kyte, J., and R. F. Doolittle. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132. 31. Lindquist, S. 1986. The heat shock response. Annu. Rev. Biochem. 55:1151-1191. 32. Lindquist, S., and E. A. Craig. 1988. The heat-shock proteins. Annu. Rev. Genet. 22:631-677. 33. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 34. Maurizi, M. R., P. Trisler, and S. Gottesman. 1985. Insertional mutagenesis of the lon gene in Escherichia coli: lon is dispensable. J. Bacteriol. 164:1124-1135. 35. McMullin, T. W., and R. L. Hallberg. 1988. A highly evolutionary conserved mitochondrial protein is structurally related to

E. COLI htpX

VOL. 173, 1991

36. 37. 38.

39.

40. 41.

42. 43. 44. 45.

46.

the protein encoded by the Escherichia coli groEL gene. Mol. Cell. Biol. 8:371-380. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Mizusawa, S., and D. F. Ward. 1982. A bacteriophage lambda vector for cloning with BamHI and Sau3A. Gene 20:317-322. Neidhardt, F. C., and R. A. VanBogelen. 1987. Heat shock response, p. 1334-1345. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. American Society for Microbiology, Washington, D.C. Oliver, D. 1985. Protein secretion in Escherichia coli. Annu. Rev. Microbiol. 39:615-648. Paek, K. H., and G. Walker. 1987. Escherichia coli dnaK null mutants are inviable at high temperature. J. Bacteriol. 169:283290. Parsell, D., and R. T. Sauer. 1989. Induction of a heat shock-like response by unfolded proteins in Escherichia coli: dependence on protein level not protein degradation. Genes Dev. 3:12261332. Raina, S., and C. Georgopoulos. 1990. A new Escherichia coli heat shock gene, htrC, whose product is essential for viability only at high temperatures. J. Bacteriol. 172:3417-3426. Salser, W., R. F. Gesteland, and A. Bolle. 1967. In vitro synthesis of bacteriophage lysozyme. Nature (London) 215: 588-591. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad Sci. USA 74:5463-5467. Shimatake, H., and M. Rosenberg. 1981. Purified X regulatory protein cll positively activates promoters for lysogenic development. Nature (London) 292:128-132. Singer, M., T. A. Baker, G. Schnitzler, S. W. Deischel, M. Goel, W. Dove, K. J. Jaacks, A. D. Grossman, J. W. Erickson, and

47.

48. 49.

50. 51. 52. 53.

54.

55.

2953

C. A. Gross. 1989. A collection of strains containing genetically linked alternating antibiotic resistance elements for genetic mapping of Escherichia coli. Microbiol. Rev. 53:1-24. Straus, D. B., W. A. Walter, and C. A. Gross. 1987. The heat shock response of E. coli is regulated by changes in the concentration of 32. Nature (London) 329:348-351. Straus, D. B., W. A. Walter, and C. A. Gross. 1988. Escherichia coli heat shock gene mutants are defective in proteolysis. Genes Dev. 2:1851-1858. Susek, R. E., and S. Lindquist. 1989. hsp26 of Saccharomyces cerevisiae is related to the superfamily of small heat shock proteins but is without a demonstrable function. Mol. Cell. Biol. 9:5265-5271. Thomas, P. S. 1983. Hybridization of denatured RNA transferred or dotted to nitrocellulose paper. Methods Enzymol. 100:255-266. Tilly, K., J. Spence, and C. Georgopoulos. 1989. Modulation of stability of the Escherichia coli heat-shock regulatory factor c2. J. Bacteriol. 171:1585-1589. Wang, Q., and J. M. Kaguni. 1989. A novel sigma factor is involved in expression of the rpoH gene of Escherichia coli. J. Bacteriology 171:4248-4253. Winter, R. B., L. Morrissey, P. Gauss, L. Gold, T. Hsu, and J. Karam. 1987. Bacteriophage T4 regA protein binds to mRNAs and prevents translation initiation. Proc. Natl. Acad. Sci. USA 84:7822-7826. Yamamori, T., and T. Yura. 1982. Genetic control of the heat-shock protein synthesis and its bearing on growth and thermal resistance in Escherichia coli K-12. Proc. Natl. Acad. Sci. USA 79:860-864. Zhou, Y.-N., N. Kusukawa, J. W. Erickson, C. A. Gross, and T. Yura. 1988. Isolation and characterization of Escherichia coli mutants that lack the heat shock sigma factor a32. J. Bacteriol. 170:3640-3649.