JOURNAL

OF

BACTERIOLOGY, Dec. 1991,

p.

Vol. 173, No. 24

7982-7987

0021-9193/91/247982-06$02.00/0 Copyright © 1991, American Society for Microbiology

Characterization of the Heat Shock Response in Mycobacterium bovis BCG BHARVIN K. R. PATEL,t DILIP K. BANERJEE, AND PHILIP D. BUTCHER* Department of Medical Microbiology, St. George's Hospital Medical School, London SW17 ORE, United Kingdom Received 13 December 1990/Accepted 3 October 1991

We have for the first time characterized the heat shock response in mycobacteria both at the level of transcription, by RNA extraction, Northern (RNA) blotting, and hybridization with gene-specific probes for the Mycobacterium tuberculosis 65- and 71-kDa heat shock proteins (HSPs), and at the level of translation, by [35SJmethionine labelling, sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and autoradiography. We observed increased synthesis of 40-, 65-, 71-, and 90-kDa proteins, which appear to be major HSPs in mycobacteria. The 40-, 71-, and 90-kDa HSPs are coordinately regulated in terms of temperature requirements and kinetics of induction but differ in the levels of expression. The 65- and 71-kDa HSPs are differentially regulated in response to temperature, with different kinetics and levels of induction. mRNA transcript sizes for the 71-, 65-, 40-, and 30-kDa proteins were found to be broadly consistent with DNA sequence open reading frames. A maximum increase of about 69-fold in the levels of mRNA for the 71-kDa HSP after 45 min of heat shock at 45C was observed, whereas the 65-kDa HSP mRNA increased only 5-fold. It was also found that in M. bovis BCG, as in Escherichia coli, a major control mechanism of the heat shock response is operative at the level of transcription. An ability to characterize the heat shock response in mycobacteria provides an experimental model with which to study environmentally regulated gene expression and an opportunity to identify virulence genes, which may coregulate as part of the heat shock regulon.

The heat shock response is a cellular response to stress that is characterized by an increased synthesis of a class of proteins called heat shock proteins (HSPs). These proteins are also synthesized constitutively and perform essential functions during normal cell growth (10, 17). The response is a highly conserved genetic system (11) and can be induced, in part at least, in vitro by many other stimuli, including nutrient starvation, viral infection, contact with heavy metals, and exposure to ethanol, oxidants, UV radiation, amino acid analogs, or DNA-damaging agents (10, 17). HSPs are biologically important because they have been implicated in thermotolerance (28), immunodominance (27), and autoimmunity (24). Indeed, an increasing interest in HSPs as factors determining host-pathogen interaction has led to the suggestion that HSPs of pathogenic organisms may play a role in infection and may be virulence factors (9). Thus, HSPs may provide a link between the stress response, pathogen survival, pathogenesis, and immunity. However, almost all of our knowledge about the regulation of the prokaryotic heat shock response is derived from Escherichia coli (10, 11, 17, 18, 23, 26). On the basis of gene sequence homology (27), several mycobacterial antigens have been identified as homologs of major HSPs in E. coli. Recently, it has been shown by radiolabelling and immunoblotting that the 65-, 71-, and 18-kDa antigens function as HSPs in Mycobacterium smegmatis (22), M. bovis BCG (14), and M. habana (8), respectively, as has also been shown for the 90-, 71-, 65-, 45-, 19-, and 15-kDa proteins of M. tuberculosis (9). No information about the regulation of the heat shock response in mycobacteria at the level of transcription is available. We have recently developed a * Corresponding author. t Present address: Department of Radiation Medicine, Vincent T. Lombardi Cancer Research Center, Georgetown University, Washington, DC 20007-2197. 7982

method for the extraction and characterization of intact mRNA from slowly growing mycobacteria (19). Here we characterize the heat shock response in M. bovis BCG both at the transcriptional level, by Northern (RNA) blot analysis of HSP mRNA transcripts, and at the translational level, by protein radiolabelling and sodium dodecyl sulfate (SDS)10% polyacrylamide gel electrophoresis (PAGE) analysis. MATERIALS AND METHODS

Mycobacterial culture. M. bovis BCG (Pasteur strain) was grown at 37°C in Dubos broth base supplemented with 10% Dubos medium albumin (Difco) and harvested in mid-exponential-phase growth at a density of 2 x 108/ml. Mycobacteria were counted by the method previously described (6). Heat shock. For RNA extraction, mycobacterial cells were recovered by centrifugation at 2,500 x g for 10 min at room temperature. Bacteria were resuspended in fresh Dubos broth at 1 x 109 to 2 x 109/mi. Portions (1-mI volumes) were incubated in sterile microfuge tubes at 37°C for 1 h to allow recovery before initiating heat shock. For protein labelling, cells were not concentrated; instead, portions (10-ml volumes) of culture in 30-mi sterile plastic universals were used. Heat shock was carried out by incubation at 42, 45, and 48°C for the required length of time, ranging from 10 to 90 min. [35S]methionine labelling of proteins and SDS-PAGE. De novo protein synthesis in mycobacteria at 37°C and during heat shock was examined by adding L-[35S]methionine (1 p.Ci/ml; Amersham) at time intervals specified in Results. Radioactive labelling was terminated after a chase with 10 mM L-methionine at 37°C for 10 min by chilling the samples on ice for 10 min. The cells were collected and washed twice with phosphate-buffered saline (PBS) containing 1 mM L-methionine. Cells were lysed by homogenization for 5 to 10 min with an equal amount of glass beads (75 to 150 jum; Sigma) in 25 ,ul of PBS. Proteins were solubilized by boiling the sample

VOL. 173, 1991

for 15 to 20 min in SDS sample buffer and separated by one-dimensional SDS-PAGE (7). The percentage of [35S]methionine incorporated into BCG was determined by trichloroacetic acid precipitation (20). After fixing, staining, and equilibration with Amplify (Amersham), gels were fluorographed for 3 to 7 days with preflashed Kodak X-Omat RP X-ray film at -70°C. RNA extraction, electrophoresis, and Northern hybridization. Total RNA from slowly growing mycobacteria may be extracted by the method described in detail previously (19). BCG cells were pelleted by centrifugation at 10,000 x g for 30 s. The bacterial pellet was immediately resuspended in 1.0 ml of 4 M guanidinium thiocyanate lysis solution (containing 0.5% sodium N-lauroylsarcosine, 25 mM sodium citrate [pH 7.0], and 0.1 M 2-mercaptoethanol) by vigorous vortexing followed by immediate sonication (Braun Labsonic 1510) for 45 s at 50 W, using an ultrathin probe (4 mm in diameter). Immediately, each tube was subjected to continuous vortexing for 10 to 15 min. The lysate was then allowed to remain at room temperature for 3 to 4 h, with intermittent vortexing every 30 min for 2 to 3 min. mRNA was purified after bacterial lysis by centrifugation through a 5.7 M CsCl cushion in 0.1 M EDTA (pH 7.5) for 18 h at 36,000 x g and 20°C, essentially as described previously (2). After centrifugation, RNA was redissolved in 360 ,ul of TE (10 mM Tris-HCl [pH 7.4], 1 mM EDTA [pH 8.0]), ethanol precipitated from 0.3 M sodium acetate (pH 5.2), and finally dissolved in 5 ,ul of TE for immediate use or storage at -70°C. RNA was separated by formaldehyde (2.2 M)-denaturing agarose (1.4%) gel electrophoresis in 0.1 M 3-(Nmorpholino)propanesulfonic acid (MOPS; pH 7.0), 40 mM sodium acetate, and 5 mM EDTA (pH 8.0) (20). After electrophoresis, ethidium bromide staining, and photography, RNA was transferred by Northern blotting to nylon membranes (Hybond-N; Amersham) by using 20x SSPE (3.6 M NaCl, 0.2 M sodium phosphate, 0.02 M EDTA [pH 7.3]) and fixed by baking for 2 h at 80°C. DNA-RNA hybridization. The following plasmid clones were used: clone Y3111a, a 2.3-kb KpnI-EcoRI fragment of the 71-kDa protein of M. tuberculosis in pUC19; clone Y3111b, a 4.8-kb EcoRI fragment of the 71-, 40-, and 30-kDa proteins of M. tuberculosis in pUC8; and clone B3115, a 1.9-kb fragment of M. tuberculosis 65-kDa antigen in pUC8. These clones were a kind gift from R. Lathigra and D. Young, MRC Tuberculosis Unit, Hammersmith Hospital, London. Clone pMBr340 was a 0.47-kb BamHI fragment in pGEM1 of the M. paratuberculosis 16S rRNA. Insert DNA was radiolabelled with [32P]dCTP by random priming (4) and used as probes. The specific activity of the probes was approximately 2 x 108 to 4 x 108 cpm/,ug of DNA. Membranes were prehybridized at 42°C in hybridization solution (50% formamide, 5x SSPE, 5x Denhardt's solution, 0.1% SDS, 40 ,ug of denatured salmon sperm DNA per ml) for 1 to 2 h. Denatured radiolabelled probe (purified on Nick-column [Pharmacia]) was added directly to the hybridization solution and incubated at 42°C for 14 to 16 h. The membranes were washed twice for 15 min each time at 65°C in 3 x SSPE-0.1% SDS and then twice for 30 min each time at 65°C in lx SSPE-0.1% SDS. For autoradiography, the membranes were exposed for 24 to 72 h to preflashed X-ray film (Kodak X-Omat RP) at -70°C with an intensifying screen. Filters were deprobed for rehybridization by heating to 65°C in 0.005 M Tris-HCI (pH 8.0)-0.002 M EDTA-0.1x Denhardt's solution for 2 h. Stripping of probe was confirmed by autoradiography. The autoradiographs were scanned with a

HEAT SHOCK RESPONSE IN M. BOVIS BCG

7983

2202 Ultroscan laser densitometer and quantitated by using LKB-2190 GelScan software (LKB-Pharmacia). RESULTS Induction and analysis of HSPs. Initial experiments to define the conditions for heat shock in terms of temperature

requirement were performed by radiolabelling mid-log-phase cultures of M. bovis BCG grown at 37°C with [35`]methionine during a 2-h heat shock at 42, 45, and 48°C. Analysis of accumulated newly synthesized proteins by SDS-PAGE and autoradiography revealed major differences between heatshocked and nonshocked BCG at all temperatures, with the increased synthesis of four major proteins of 40, 65, 71, and 90 kDa. At 48°C, protein synthesis was markedly reduced, with the 71- and 90-kDa proteins remaining dominant (results not shown). Similar results have recently been obtained by others using M. tuberculosis (9). It was therefore of interest to determine the kinetics of induction of these proteins at different temperatures rather than observing total accumulated newly synthesized protein, so as to investigate the coordinate or independent nature of control of HSP expression and to correlate protein synthesis with mRNA levels. Accordingly, mid-log-phase M. bovis BCG grown at 37°C was heat shocked at 42, 45, and 48°C for time intervals ranging from 15 to 90 min and then metabolically radiolabelled with [35S]methionine for the last 15 min of the heat pulse. Labelling was continued for a further 105 min during recovery at 37°C to allow a total labelling period of 2 h. In this way, the time point at which particular proteins start to be synthesized or synthesis terminated could be determined and correlated with mRNA levels measured by Northern blotting (see Fig. 2). Thus, the protein profile would reflect such changes in gene expression that would not be shown by accumulation labelling at the heat shock temperature. Control cells were labelled at 37°C for 2 h. This experimental protocol was adopted because a labelling period of 2 h was required to incorporate sufficient radioactivity for subsequent autoradiography, since less than 0.1% of the total radioactivity was incorporated during this time interval (5 x 103 cpm/109 BCG/h at 37°C). The amount of de novo protein synthesis at each temperature was measured by trichloroacetic acid precipitation; incorporation of [35`]methionine per 109 BCG increased at all temperatures compared with incorporation at 37°C after 15 min, with a maximum threefold increase at 45°C, and thereafter started to decline to normal levels or to below-normal levels at 48°C. This finding demonstrates that protein synthesis was stimulated by a short pulse of heat shock but that continued heat shock caused a decrease and confirms the general phenomenon of a heat shock response in BCG. The [35S]methionine-labelled proteins from 1.4 x 109 BCG so treated were separated by SDS-PAGE to analyze the effect of temperature and time of heat shock on the protein profile. Coomassie blue staining of the gel revealed that equal amounts of protein were loaded in each lane; the autoradiograph is shown in Fig. 1. The amount of background protein synthesis observed on the autoradiograph was transiently increased for 15 min when the temperature was shifted from 37 to 42°C and then returned to preshift levels. At 45°C, background protein synthesis gradually decreased with time, while most of the proteins except the major HSPs ceased immediately after the temperature shift at 48°C. Induced synthesis of the 40-, 71-, and 90-kDa proteins after the temperature shift from 37 to 42°C was a gradual process and reached the maximum level after about 60 min. The rate of induction of these proteins

7984

J. BACTERIOL.

PATEL ET AL.

'li.~

'-, .1---.

CO *W

1.4

:.,...i:, AM.'

4~

.:.. w

w

lo

-_ _

pF....

_

.... .

....

_,W

FIG. 1. Electrophoretic profile of heat shock-induced proteins. The autoradiograph of an SDS-PAGE shows the protein profile of [35S]methionine-labelled M. bovis BCG in response to heat at three temperatures, 42, 45, and 48°C, over 90 min. The lanes labelled 0 min indicate non-heat-shocked controls maintained at 37°C. Each lane represents protein extracted from 1.4 x 109 M. bovis BCG. Relative positions of the standard molecular weight markers are marked. The four major HSPs discussed in the text are marked with bold arrows and their molecular sizes.

rapidly increased with temperature shifts to 45 or 48°C. A transient induction of 65-kDa protein synthesis was observed only when the temperature was shifted from 37 to 42°C, while shifts to 45 and 48°C resulted in reduction of this protein (Fig. 1). Densitometry of the autoradiograph revealed an approximately 40-fold maximum increase in the level of the 71-kDa protein after heat shock at 45°C for 30 min, which then gradually diminished. At 42°C, the level of induction was lower, whereas at 48°C, rapid induction followed by complete cessation of protein synthesis was observed. Similar kinetic profiles and levels of expression were observed for the 40- and 90-kDa proteins, although the 40-kDa protein increased only up to 12-fold. The kinetics and levels of expression for the 65-kDa protein were different from those of the other heat-inducible proteins; a maximum sevenfold increase after 15 min was observed after a temperature shift from 37 to 42°C, which returned rapidly to preshift levels. At temperatures above 42°C, synthesis of the 65-kDa protein gradually decreased with time. These data demonstrate that the 65- and 71-kDa proteins are independently regulated. Despite the quantitative limitations of densitometry of nonuniformly labelled proteins (i.e., methionine labelling) separated only in one dimension, it was of interest to compare the semiquantitative data on the levels of the major HSPs of M. bovis BCG with those reported for M. tuberculosis and E. coli with use of similar methods. At 37°C, the 65-kDa heat shock protein of BCG formed approximately 11% of the [35S]methionine-labelled total cellular proteins, which compares with 10% for M. tuberculosis (9), while its E. coli homolog HSP-60 (GroEL) accounts for only 1.6% of total cell proteins under normal growth conditions (17). The levels of 40-, 71-, and 90-kDa proteins revealed them to be only minor components (less than 0.5%) of the total protein

synthesis at 37°C; after heat shock at 45°C, the 71-kDa protein represented >20% of newly synthesized protein. The finding of no changes induced by heat shock in the stained SDS-PAGE protein profile indicates that these increases in de novo synthesis of heat-inducible proteins represent only a fraction of the total amount of each protein normally present. Characterization of the heat shock response at the level of transcription. A method for the extraction of structurally intact mycobacterial total RNA has been developed recently in our laboratory, and criteria for the presence of mRNA transcripts within this population have been met (19). This has allowed us to study the heat shock response in slowly growing mycobacteria at the level of transcription by monitoring the expression of specific mRNA transcripts by Northern blotting, hybridization with radiolabelled genespecific probes, autoradiography, and densitometry. We concentrated our study on the mRNA for the 65- and 71-kDa HSPs because they show differential regulation at the protein level in terms of kinetics of induction and levels of expression. To determine the kinetics of induction and levels of expression of the 65- and 71-kDa HSP mRNAs, total RNA was extracted from 109 BCG at different time intervals after a temperature shift from 37°C to either 42, 45, or 48°C. As a control, total RNA was extracted from BCG maintained at 37°C. Formaldehyde gel electrophoresis of the total RNA fractions revealed discrete bands for the 23S and 16S rRNAs (Fig. 2A). Electrophoretically separated total RNA was transferred to nylon filters by Northern blotting and hybridized with radiolabelled DNA probes for the M. tuberculosis 65-, 30-, 40-, and 71-kDa proteins. To control for the variable yield and loading of RNA in each lane, Northern blots were hybridized with a mycobacterial 16S rRNA gene-specific

HEAT SHOCK RESPONSE IN M. BOVIS BCG

VOL. 173, 1991

7985

C 42 C: Y3111b(30,40 & 71kD)

B 42C: 65kD

-238

a^ Z

-23S

-2600

1-2

*

6

3.400

W

0 D 45 C:C5kD

10

20

30

45

60--Min

E45 C: 71kD

0

10

20

30

45

600

60--Min

F

451C: 16S rRNA -23S

-23S

-23S

-168

-16S

'

,_ -

0

15

G 48C:65kD

30

40

50

60

-Min 0

15

30

40

50

48'C: 7lkD

-Min

0

15

30

40

50

60-Min

48C:16S rRNA

.2&.

4

I 50 30 40 15 0 15 30 40 50 so0 h 0 50 60 M 15 30 FIG. 2. Effect of heat shock on mRNA transcript levels. (A) Photograph of an ethidium bromide-stained gel of total RNA extracted from M. bovis BCG after different time intervals of heat shock at 420C. Duration of the heat shock is indicated in minutes below each lane. Each lane represents total RNA extracted from 109 BCG. Lane E shows the relative positions of the 23S and 16S rRNAs of E. coli. (B and C) Autoradiographs of the Northern blot of the gel shown in panel A, hybridized with the gene-specific probes for 65-kDa (clone B3115) protein (B) and the 30-, 40-, and 71-kDa (clone Y3111b) proteins (C) of M. tuberculosis. (D to I) Autoradiographs of Northern blots of RNA isolated from BCG after heat shock at 45°C (D to F) and at 48°C (G to I), hybridized with the 65-kDa (B3115) protein gene-specific probe (D and G), the 71-kDa (Y3111a) protein gene-specific probe (E and H), and pMBr340, a mycobacterial 16S rRNA probe (F and I). Positions of the 16S and 23S rRNAs of E. coli are marked.

probe (pMBr340), the levels of which did not alter in heat (Fig. 2F and I). Analysis of the autoradiographs presented in Fig. 2 revealed discrete size classes of mRNA transcripts. Hybridization with the 65-kDa probe (B3115) showed a single band of approximately 2,600 nucleotides (Fig. 2B), larger than the expected transcript size from this gene sequence (21). However, variations in transcript size for the 65-kDa mRNA at 45°C (Fig. 2D) and 48°C (Fig. 2G) were observed between experiments. Probing with Y3111b (consisting of the genes for 71-, 40-, and 30-kDa proteins) revealed two bands of response to

approximately 2,000 and 400 to 600 nucleotides and a faint area of hybridization at about 2,900 nucleotides (Fig. 2C). Probing the Northern blots for heat shock at 45 and 48°C with Y3111a (which contains the gene for the 71-kDa protein only) gave a discrete band of approximately 2,000 nucleotides at 48°C (Fig. 2H) and two bands of approximately 2,000 and 1,100 nucleotides at 45°C (Fig. 2E). The band of 2,000 nucleotides would roughly correspond to a gene sequence for the 71-kDa protein, while the lower band of 1,100 nucleotides could be a breakdown product. However, the significance of these mRNA transcript sizes in relation to

7986

PATEL ET AL.

gene sequences remains to be confirmed. Nevertheless, mRNA can clearly be detected by Northern hybridization, and increased levels of 71-kDa mRNA in response to heat shock can clearly be demonstrated. Despite mRNA transcript size variation, quantitation by densitometry of autoradiographs revealed major differences between the levels of mRNA for the 71- and 65-kDa proteins. The optical density of the total area of hybridization in each lane (representing the heterogeneity of transcripts due to partial degradation) was corrected for variation in total RNA loaded (hybridization with the 16S rRNA probe) and expressed as fold increases over values for non-heat-shocked controls. Maximum levels of 71-kDa mRNA were induced after 45 min of heat shock at 45°C, with an average increase of 69-fold (n = 4; range, 42- to 85-fold); the levels rapidly rose within 15 min and decreased to a lower level after 1 h. Levels of 65-kDa mRNA increased only to a maximum of fivefold at 45°C before returning to preshift levels, in contrast to the 71-kDa mRNA; the 65-kDa mRNA also showed differences in temperature and kinetic profiles. The 65-kDa mRNA therefore appears to be differentially regulated compared with 71-kDa mRNA in response to heat, a phenomenon also observed at the protein level. Pretreatment of BCG cultures with rifampin at 1 ,ug/ml (=2 x MIC) for 2 h prior to heat shock reduced heat shockinduced 71-kDa mRNA transcript levels by 90% compared with untreated cultures. Exposure to rifampin for 15 h abolished the response completely at the levels of both protein synthesis and mRNA transcription, indicating a transcriptional level of control for the heat shock response.

DISCUSSION We have presented experiments that for the first time attempt to characterize the heat shock response in mycobacteria at the levels of both transcription and translation, using M. bovis BCG. We found that the 40-, 65-, 71-, and 90-kDa proteins are the major mycobacterial proteins which are induced in response to heat. Some minor HSPs may exist but cannot be resolved by one-dimensional SDS-PAGE. It may be possible to identify other mycobacterial HSPs by twodimensional gel electrophoresis, as has been demonstrated with Neisseria gonorrhoeae (25). It should be considered that quantitative data from nonuniform labelling with [35S]methionine and separation on one-dimensional gels impose certain limitations on interpretation, but comparison with results obtained for E. coli (10, 11, 17, 18, 23, 26) by similar methods shows interesting differences in the kinetics of the heat shock response and the levels of expression of HSPs and their respective mRNA transcripts. The levels of dnaK dnaJ mRNA in E. coli reach a maximum of about 11-fold after 5 min of temperature shift (23). The levels of 64-kDa protein reached a maximum of about fivefold in 5 min, while its mRNA levels reached a maximum of about threefold in 3 min, after a temperature shift from 30 to 42°C (26). In contrast, we found that in BCG a maximum increase in the levels of 71-kDa mRNA reached an average of 69-fold in 45 min after a temperature shift from 37 to 45°C. The levels of 65-kDa protein reached a maximum of about sixfold in 15 min, while its mRNA levels reached a maximum of about threefold in 30 min, after a temperature shift from 37 to 42°C. Thus, it is clear that the kinetics of induction for 65- and 71-kDa HSPs and corresponding mRNAs are slower in mycobacteria than in E. coli and that the levels of 71-kDa HSP are considerably higher. Whether such differences are associated with the intracellular and

J. BACTERIOL.

pathogenic life-style of M. tuberculosis or simply reflect differences in generation times needs to be determined. On the basis of gene sequence homologies with other prokaryotic HSPs, it had been speculated that these proteins in mycobacteria are heat inducible. Our data present strong evidence in favor of these E. coli HSP homologs in mycobacteria functioning as HSPs. Kinetic data indicate that the 71-kDa (DnaK homolog) and 90-kDa proteins are coordinately regulated in response to heat but that the 40-kDa (DnaJ) protein shows lower levels of expression and may be regulated separately (Fig. 1). The 65-kDa (GroEL homolog) protein, which accounts for a major proportion (11%) of the mycobacterial de novo-synthesized proteins, is regulated independently at higher temperatures compared with the 71-kDa protein in terms of temperature requirements, kinetics of induction, and levels of expression at the levels of both transcription and translation. This finding provides experimental evidence for the suggestion that at least two forms of HSP gene regulation in mycobacteria are operative and that these forms function differently in response to heat. These functional differences may be explained by the presence of consensus sequences for (rE and J32 promoters in dnaK but only a Cu32 promoter sequence in groEL and groES in M. tuberculosis (9). Hybridization of Northern blots with DNA probes for the 65-, 71-, 40-, and 30-kDa proteins revealed sizes of mRNA broadly consistent with DNA sequence open reading frames. Although some variation in the size allocation of mRNA transcripts was observed, the important point in the context of this work is that specific mRNA can in fact be analyzed and the increased levels of HSP mRNA induced in response to heat could be detected. Further studies such as S1 mapping or primer extension analysis will clarify the relationship between mRNA transcript sizes and gene sequences. Nevertheless, the appearance of relatively discrete zones of hybridization on the Northern blots that varied in both size and quantity, depending on which gene probe was used, indicates that mRNA transcripts were in fact being extracted and could be quantitated. An inherent difficulty with mRNA extraction from mycobacteria is the short half-life of prokaryotic mRNA combined with a complex cell wall that is difficult to lyse efficiently. Immediate sonication and breakage of the cell wall allows rapid equilibration with the guanidium thiocyanate solution. Since this takes a minimum of 3 min from harvest, it is not surprising that partial degradation of mRNA, which is reflected in the size heterogeneity of mRNA transcripts and broad bands on the Northern blots, occurs. A significant proportion of the mRNA appears to be of the correct size and can be quantitated. We have previously demonstrated that mRNA extracted in this way can be further analyzed by cDNA synthesis and polymerase chain reaction amplification (19), indicating its structural integrity. We have demonstrated that heat induction of the 71-kDa HSP mRNA and protein synthesis in mycobacteria are abolished by rifampin. Thus, we conclude that in mycobacteria, as in E. coli (26), a major control mechanism of the heat shock response is operative at the level of transcription. Because of the low level of incorporation of [35S]methionine into slowly growing mycobacteria such as BCG, which necessitates long radiolabelling times, it is somewhat difficult to compare the kinetics observed at the levels of transcription and translation. Thus, it cannot be precisely demonstrated that mRNA synthesis exactly precedes protein synthesis. The levels of accumulated de novo-synthesized protein are a reflection of protein stability and of the effi-

VOL. 173, 1991

ciency of translation and stability of mRNA. Nevertheless, the results show that 71-kDa HSP synthesis broadly reflects the levels of mRNA after heat shock, and the kinetics of HSP and mRNA induction are closely mirrored. However, the results also show that although the 71-kDa HSP mRNA is present at 48°C for at least 60 min, 71-kDa HSP synthesis ceased after 15 min of heat shock at this temperature. A most likely explanation is that translation is inhibited by prolonged exposure to such high temperatures. It was observed that the 65-kDa HSP mRNA was present at both 45 and 48°C, but that protein synthesis ceased immediately after heat shock at these temperatures. The reason is unknown but may reflect some posttranscriptional regulatory mechanism for the 65-kDa protein. Whether such levels of control operate in mycobacteria remains to be determined. In terms of the biological significance of the heat shock response in pathogenic microorganisms, it has been shown for a wide variety of bacterial pathogens that virulence factors and other factors which play a role in host-parasite interactions are coordinately regulated with the HSPs. These stress proteins are expressed as overlapping subsets of proteins in response to environmental stimuli likely to be encountered within the infected host (3, 12, 13, 15, 16). Recently, it has been observed that syntheses of over 30 Salmonella proteins, including GroEL and DnaK HSPs, are selectively induced during infection of macrophages by this organism (1). Differential expression of a selective class of proteins by Shigellaflexneri in intracellular and extracellular environments has also been reported (5). Therefore, an ability to characterize the heat shock response in mycobacteria provides an important model for studying the environmental regulation of genes which may have a role in virulence and pathogenicity of mycobacteria in such diseases as tuberculosis and leprosy. ACKNOWLEDGMENTS This investigation received support from the UNDP/World Bank/ WHO Special Programme for Research and Training in Tropical Diseases. B.K.R.P. was a recipient of an Overseas Research Student Award (1989-1990) from the Secretary for Education and Science, United Kingdom. We are grateful to D. B. Young and R. B. Lathigra, MRC Tuberculosis and Related Infections Unit, Hammersmith Hospital, London, for providing M. tuberculosis clones and helpful discussions. REFERENCES 1. Buchmeier, N. A., and F. Heifron. 1990. Induction of Salmonella stress proteins upon infection of macrophages. Science 248:730732. 2. Chirgwin, J. M., A. E. Przybyla, R. J. MacDonald, and W. J. Rutter. 1979. Isolation of biologically active RNA from sources enriched in RNase. Biochemistry 18:5294-5299. 3. Dorman, C. J., N. N. Bhriain, and C. F. Higgins. 1990. DNA

supercoiling and environmental regulation of virulence gene expression in Shigellaflexneri. Nature (London) 344:789-792. 4. Feinberg, A. P., and B. Vogelstein. 1984. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 137:266-267. 5. Headley, V. L., and S. M. Payne. 1990. Differential protein expression by Shigellaflexneri in intracellular and extracellular environments. Proc. Natl. Acad. Sci. USA 87:4179-4183. 6. Holmes, I. B., and G. R. F. Hilson. 1972. The effect of rifampin and dapsone on experimental Mycobacterium leprae infections: minimum inhibitory concentrations and bactericidal action. J. Med. Microbiol. 5:251-261. 7. Laemmli, U. K. 1970. Cleavage of structural proteins during the

HEAT SHOCK RESPONSE IN M. BOVIS BCG

7987

assembly of the head of bacteriophage T4. Nature (London) 227:680-685. 8. Lamb, F. I., N. B. Singh, and M. J. Colston. 1990. The specific 18-kilodalton antigen of Mycobacterium leprae is present in Mycobacterium habana and functions as a heat-shock protein. J. Immunol. 144:1922-1925. 9. Lathigra, R. B., P. D. Butcher, T. R. Garbe, and D. B. Young. 1991. Heat shock proteins as virulence factors of pathogens. Curr. Microbiol. Immunol. 167:125-143. 10. Lindquist, S. 1986. The heat shock response. Annu. Rev. Biochem. 55:1151-1191. 11. Lindquist, S., and E. A. Craig. 1988. The heat-shock proteins. Annu. Rev. Genet. 22:631-677. 12. Maurelli, A. T. 1989. Temperature regulation of virulence genes in pathogenic bacteria: a general strategy for human pathogens? Microb. Pathog. 7:1-10. 13. Maurelli, A. T., B. Blackmon, and R. Curtiss III. 1984. Temperature-dependent expression of virulence genes in Shigella species. Infect. Immun. 43:195-201. 14. Mehlert, A., and D. B. Young. 1989. Biochemical and antigenic characterization of the M. tuberculosis 7lkD antigen, a member of the 7OkD heat shock protein family. Mol. Microbiol. 3:125130. 15. Miller, J. F., J. J. Mekalanos, and S. Falkow. 1989. Coordinate regulation and sensory transduction in the control of bacterial virulence. Science 243:916-922. 16. Morgan, R. W.,- M. F. Christman, F. C. Jacobson, G. Storz, and B. N. Ames. 1986. Hydrogen peroxide-inducible proteins in Salmonella typhimurium overlap with heat shock and other stress proteins. Proc. Natl. Acad. Sci. USA 83:8059-8063. 17. Neidhardt, F. C., and R. A. V-anBogelen. 1987. Heat shock response, p. 1334-1345. In F. C. Neidhardt, J. L. Ingraham, B. Magasanik, K. B. Low, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology, vol. 2. American Society for Microbiology, Washington, D.C. 18. Neidhardt, F. C., R. A. VanBogelen, and V. Vaughn. 1984. The genetics and regulation of heat shock proteins. Annu. Rev. Genet. 18:295-329. 19. Patel, B. K. R., D. K. Banerjee, and P. D. Butcher. 1991. Extraction and characterisation of mRNA from mycobacteria: implication for virulence gene identification. J. Microbiol. Methods 13:99-111. 20. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 21. Shinnick, T. M. 1987. The 65-kilodalton antigen of Mycobacterium tuberculosis. J. Bacteriol. 169:1080-1088. 22. Shinnick, T. M., M. H. Vodkin, and J. C. Williams. 1988. The Mycobacterium tuberculosis 65-kilodalton antigen is a heat shock protein which corresponds to common antigen and to the Escherichia coli GroEL protein. Infect. Immun. 56:446-451. 23. 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 cr32. Nature (London) 329:348-351. 24. Van Eden, W., J. E. R. Thole, R. Van der Zee, A. NoordziJ, J. D. A. van Embden, E. J. Hensen, and I. R. Cohen. 1988. Cloning of the mycobacterial epitope recognized by T lymphocytes in adjuvant arthritis. Nature (London) 331:171-173. 25. Woods, M. L., II, R. Bonfiglioli, Z. A. McGee, and C. Georgopoulos. 1990. Synthesis of a selective group of proteins by Neisseria gonorrhoeae in response to thermal stress. Infect. Immun. 58:719-725. 26. Yamamori, T., and T. Yura. 1980. Temperature-induced synthesis of specific proteins in Escherichia coli: evidence for transcriptional control. J. Bacteriol. 142:843-851. 27. Young, D. B., R. B. Lathigra, R. Hendrix, D. Sweetser, and R. A. Young. 1988. Stress proteins are immune targets in leprosy and tuberculosis. Proc. Natl. Acad. Sci. USA 85:42674270. 28. Zeuthen, M. L., and D. H. Howard. 1989. Thermotolerance and the heat-shock response in Candida albicans. J. Gen. Microbiol. 135:2509-2518.

Characterization of the heat shock response in Mycobacterium bovis BCG.

We have for the first time characterized the heat shock response in mycobacteria both at the level of transcription, by RNA extraction, Northern (RNA)...
2MB Sizes 0 Downloads 0 Views