Previous Article | Next Article ![]()
Journal of Bacteriology, January 2009, p. 673-677, Vol. 191, No. 2
0021-9193/09/$08.00+0 doi:10.1128/JB.01256-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Department of Microbiology and Immunology, Emory University School of Medicine, Atlanta, Georgia 30322
Received 8 September 2008/ Accepted 30 October 2008
|
|
|---|
GAS causes many types of disease. Streptococcus pyogenes, the group A streptococcus (GAS), is a common and serious human pathogen that causes many different types of disease (12, 38). Among these diseases are streptococcal toxic shock syndrome and necrotizing fasciitis (2, 40, 52). Serious nonsuppurative sequelae can occur in some individuals following GAS infections, including acute rheumatic fever infections, chronic rheumatic heart disease infections (41), and acute glomerulonephritis (38).
To combat GAS infections, an understanding of the mode by which this bacterium initiates infection and of the ways in which it overcomes host defenses is important. GAS may enter through the oral mucosa to produce pharyngitis ("strep throat"), one of the most common childhood diseases. Although pharyngitis is usually a self-limited infection, annual direct costs associated with this disease are thought to be in the range of $1 billion in the United States (41). Another major route for GAS entry is the skin, usually following abrasions or insect bites. Local superficial infections may result in impetigo, deeper tissue involvement may manifest as cellulitis, and GAS may also become extremely invasive and lead to severe tissue destruction (necrotizing fasciitis) (38). Thus, this important human pathogen is capable of initiating infection in very different environments in its host, indicating its ability to grow in the face of very different stress conditions.
Although there appears to be some association of specific strains with particular diseases (5), some strains of GAS seem to be capable of causing most or all of the different types of disease discussed above. Therefore, we can conclude that interactions between the host and the pathogen are critically important in determining the outcome of infection. As an infection develops, GAS expresses different gene products in response to changes in the host microenvironment, which, in turn, is altered by the infection (43). For example, colonization of the skin (pH 5 [31, 57], usually at temperatures below 37°C, and a high concentration of salt if sweating occurs) probably requires different GAS factors than colonization of the oral mucosa (pH 6.7 and the presence of saliva) (1). If GAS invades different tissues, it may be faced with lower oxygen tension, different pH levels (pH 7.4 in blood), and a temperature of 37°C, or in the case of fever, even higher temperatures. The destruction of host tissue will alter the pH and other environmental conditions as well, and the production of an abscess results in a pH below 6.5 (7). Therefore, it seems likely that the regulation of expression of GAS virulence factors and even of "housekeeping" genes in response to changes in the host microenvironment plays a critical role in determining the specific type of disease that develops.
Importance of CovRS in pathogenesis. Although our understanding of GAS pathogenesis is not nearly as complete as that of the pathogenesis of many gram-negative bacteria, several transcriptional regulators have been identified in GAS, and their study is under way. Characterization has begun for several regulators specific for one or a few promoters (e.g., Nra and RofA, etc.) (6, 29, 33, 36), as well as more general ones, including the activator Mga (multiple gene regulator of GAS) (28). Among the 13 two-component signal-transducing systems (TCS) encoded in GAS genomes, the TCS CovRS (CsrRS) seems to play a central role in growth and pathogenesis (41, 43). In addition to regulating GAS gene expression in response to stress in laboratory settings (14), CovRS is required for biofilm formation (11). It also modulates the transcriptome during growth in human blood (22) and has been shown to be important in the development of GAS disease in animal models, including in cynomologus macaques (56), and probably also in humans (53, 54). CovRS controls expression of about 15% of the GAS genome either directly or indirectly (21). Unlike the DNA binding protein (response regulator) of most TCS, the response regulator CovR represses most of the genes it controls (18, 20, 35, 48). It also represses its own transcription (23).
The GAS CovRS homologs in group B streptococcus, Streptococcus mutans, and Enterococcus faecalis have also been shown to regulate many virulence genes and to be important for pathogenesis in these organisms (6, 27, 32, 33, 55). Therefore, the development of a greater understanding of GAS CovRS might be helpful in elucidating the virulence of these additional human pathogens.
CovRS mediates the stress response of GAS. In most bacteria, response to environmental stress is mediated by secondary sigma factors, which are required for transcription of genes needed for growth in a stress environment. However, GAS does not encode homologs to stress response sigma factors. The only alternative sigma factor encoded in GAS genomes is homologous to the competence sigma factor of Streptococcus pneumoniae, and it is not expressed in laboratory growth conditions (44). Instead, we have found that CovS is required for GAS growth under general stress conditions (39°C, a pH of 6, or a high concentration of salt) (14). Under these stress conditions, genetic evidence indicates that CovS inactivates CovR and thus derepresses genes required for growth of the bacteria (13, 14). This work suggests that the environmental signal for the CovRS system may be envelope stress, as was found for the CpxRA TCS of Escherichia coli (15, 50).
Iron starvation stress. One of the important defenses of the host against infection is its ability to sequester iron. Like all bacteria, GAS requires iron for growth and has several mechanisms of competing with the host for iron acquisition. Because these mechanisms are energetically expensive and because a high iron concentration is toxic, the acquisition of iron by bacteria is regulated by several systems. These have only been partially characterized in GAS. Most intracellular iron in the host is stored as ferritin, or in heme complexes like hemoglobin and myoglobin, and by transferrin and lactoferrin. GAS can use heme as well as hemoglobin, myoglobin, heme-albumin, and catalase as sources of iron in laboratory media (39). However, it cannot use transferrin or lactoferrin. To transport iron into the cell, GAS uses two iron-specific transport systems encoded by sia (streptococcal iron acquisition) (3), also called hts (34), and a fhu homolog called siu (streptococcal iron uptake) (39). The multimetal transporter encoded by the mts operon genes is also used by GAS to obtain iron (26).
To determine whether CovRS plays a role in iron starvation stress, as it does in other stresses it is likely to encounter during infection, we grew the GAS serotype M6 strain JRS4 and its
covS and
covS
covR derivatives at 37°C overnight in modified Z-THY (Todd-Hewitt broth supplemented with 2% yeast extract, 0.825 mM MgCl2, 0.825 mM MnCl2, 0.825 mM CaCl2, and 0.825 mM ZnCl2 and buffered with 100 mM HEPES to pH 7.4) (17). For iron starvation, 15 mM of the iron chelator nitrilotriacetic acid (NTA) was added to the medium at the time of inoculation. Growth of the overnight cultures was measured as the optical density at 600 nm (OD600) and expressed relative to the OD of a control culture of the same strain grown in the absence of NTA. The experiment was repeated at least five times for each strain, and the results are presented in Fig. 1. We found that the
covS mutant is significantly more sensitive to iron starvation than its wild-type parent, and this phenotype is complemented by the presence of the wild-type covS gene on a plasmid. In addition, we found that the plasmid with a covS H280A mutation, in which the phosphorylatable histidine has been replaced with alanine, is unable to complement the
covS mutant in this experiment. The role of CovR in the GAS iron starvation stress response is indicated by the finding that the
covR
covS double mutant grows almost as well as its wild-type parent. Thus, as we previously found for GAS growth at a high temperature, low pH, and a high concentration of salt (14), it appears that CovS is needed to inactivate CovR for growth under iron starvation conditions. This suggests that one or more genes required for growth in low iron conditions are repressed by CovR.
![]() View larger version (9K): [in a new window] |
FIG. 1. Growth of GAS under iron starvation. GAS was grown at 37°C overnight in THY under static conditions. Turbidity was measured as the absorbance at 600 nm. Growth of each strain was normalized to growth of the same strain in the absence of NTA. The wild-type strain is JRS4 (w.t.); the covS strain is JRS331 (14); the strain indicated as covS/covS+ is JRS331, with covS expressed from the constitutive lactococcal promoter P23 on plasmid pJRS325 (14); and the covS covR strain is JRS392 (14). The strain indicated as covS(H-A) is strain JRS331/pJRS9010. The plasmid is a derivative of pJRS325, in which histidine 280 of CovS was replaced with alanine.
|
covR strain and that of its parent for mtsR, mtsA, perR, fhu/siu, and shr, the first gene in the sia operon. Cells grown in Todd-Hewitt broth supplemented with 0.2% yeast extract (THY) at 37°C were harvested in the log phase and in transition to the stationary phase. The RNA was pelleted by sedimentation through 5.7 M CsCl (37), and transcripts were quantified using the iScript one-step RT-PCR kit with SYBR green (Bio-Rad) as described previously (49). The transcript levels were expressed relative to the amount of the proS transcript. Each experiment was repeated with at least two different RNA samples. We saw no difference between the wild-type strain and the
covR strain for shr, siu, perR, or mtsR. Because the amounts of the transcript for shr and siu were very low, as might be expected in the presence of iron, we are unable to determine whether they are regulated by CovR under these growth conditions. However, there was two- to threefold more mtsA transcript (relative to proS) in the
covR mutant than in its wild-type parent strain in the transition phase of growth. In the log phase, this difference was slightly higher. The
covS mutant showed no significant difference from its wild-type parent in the amounts of any of these transcripts, and the
covR
covS double mutant had two- to fourfold more mtsA transcript than its wild-type parent. Thus, it appears that CovR represses mtsA and that increased expression of mtsA may be required for growth of GAS in laboratory iron starvation conditions. It also suggests that iron starvation can be added to the list of stresses that cause CovS to inactivate CovR, presumably by favoring the phosphatase activity of CovS. This is the first indication that the global CovRS TCS has significant input to the iron uptake system of GAS and reemphasizes the pivotal role that CovRS plays in GAS growth and pathogenesis.
![]() View larger version (9K): [in a new window] |
FIG. 2. Summary of the regulation of iron uptake genes. We show here that CovR represses mtsA, and this repression is relieved during iron starvation by CovS inactivation of CovR. This suggests that iron starvation stress causes CovS to be converted to CovS*, which acts as a phosphatase instead of a kinase. PerR activates the mtsA operon either directly or indirectly (47), but it is not regulated by CovR. The mtsA operon is also repressed by MtsR in the presence of iron (Fe) or manganese (Mn) in some GAS strains (24). In the presence of iron or other appropriate metal ions, MtsR also represses transcription of the sia operon directly (4).
|
Under several stress conditions, a
covS mutant is unable to grow while a
covS
covR mutant grows well (14). To determine whether the same was true for the stress provided by LL-37, we compared the growth of the GAS serotype M6 strain JRS4 and its
covS,
covS
covR, and
covR derivatives. The GAS strains to be tested were grown overnight at 37°C in THY, diluted 1/20 in prewarmed (37°C) THY, and grown at 37°C until mid-log phase. The culture was then diluted 1/10 into prewarmed THY for growth in the presence of mature human cathelicidin LL-37 in 96-well plates (polystyrene with flat bottom; Costar 3598). To each well containing 1 volume of the diluted LL-37 in a final concentration of 0.2% bovine serum albumin, 4 volumes of the diluted bacterial culture was added, and the liquid was mixed well by pipetting it up and down. Serial twofold dilutions of LL-37 at concentrations from 250 µg/ml to 3.9 µg/ml were used. Each LL-37 concentration was tested at least in duplicate for every strain. Controls for each strain without LL-37 were always included to be sure the cells were growing in the plates. Growth at 37°C was monitored using a BioTek Synergy HT plate reader. The OD600 was measured every 20 min for at least 6 h. Just before each reading, the plates were shaken for 10 s at the lowest intensity on the plate reader.
The results, presented in Fig. 3, show that CovS is required for GAS growth in the presence of LL-37. Thus, LL-37 is similar to other stresses for GAS. To determine whether the requirement for CovS in the presence of LL-37 results from the CovS inactivation of CovR, we investigated the mutant that was deleted from both genes. In the absence of CovR, the CovS deletion mutant is less sensitive to LL-37 than the
covS strain. This implies that, as is the case for other stresses, one or more genes required for growth in the presence of LL-37 are repressed by CovR.
![]() View larger version (15K): [in a new window] |
FIG. 3. Growth of GAS in the presence of the human cathelicidin LL-37. The MIC was defined as the lowest LL-37 concentration at which no growth of cells was detectable. The MIC was calculated relative to the MIC for the wild-type strain (JRS4) in each independent experiment. The calculated average MIC for JRS4 was 198 µg/ml (standard deviation = 63). For each strain, the relative MIC was measured in at least five independent experiments, and the average MIC of all experiments is presented, with the standard deviation indicated by a vertical bar. The strains tested were JRS4 (w.t.), JRS948 ( covR) (18), JRS331 ( covS) (14), and JRS392 ( covR covS) (14).
|
covS
covR mutant is more sensitive to LL-37 than its wild-type parent, which differs from the findings above for iron starvation and all other stresses studied. Therefore, we investigated the effect of LL-37 on a
covR mutant. We found that the single
covR mutant grows as poorly as the
covS
covR double mutant in the presence of LL-37. The simplest interpretation of this is that CovR is needed for the activation (either directly or indirectly) of a gene(s) required for growth of GAS in the presence of the cationic antimicrobial peptide LL-37. We do not favor the alternative interpretation that CovS regulates a gene required for growth in the presence of LL-37 independently of CovR, i.e., by cross talking with another TCS response regulator, because CovS has not been demonstrated to have a CovR-independent function. An operon activated by CovR (13; B. J. Froehlich and J. R. Scott, unpublished data), whose homolog in other gram-positive bacteria increases the resistance to cationic antimicrobial peptides (including the mouse analog of LL-37), is dlt (8, 19, 30, 45, 46). The dlt operon is responsible for D-alanylation of the anionic polymers lipoteichoic acid and wall lipoteichoic acid. A mutant deficient in dlt, therefore, has a higher negative charge density in its cell wall, which may allow better binding of cationic antimicrobial peptides. Additionally, a dlt mutation in Bacillus subtilis alters expression of the surface stress-responsive two-component system liaRS and leads to the decreased production of the surface quality control proteases HtrA and HrtB (homologs of DegP of Escherichia coli) (25). Therefore, in GAS, the CovR-activated operon required for growth in the presence of LL-37 might be dlt.
Growth of GAS under stress conditions (a low concentration of salt, pH 6, and 39°C) requires the inactivation of CovR by CovS. Expression of the CovR-repressed gene rsc is required for growth at 39°C, and rsc is derepressed by CovS when the strain is subjected to temperature stress (13). Since the gntR family gene crgR was found to be required for growth of GAS in the presence of LL-37 (42), it seemed possible that, like rsc, crgR might be repressed by CovR and derepressed in the presence of CovS under stress conditions (growth in LL-37). To test this possibility, we asked whether CovR represses crgR. RNA was collected from JRS4 and its
covR mutant JRS948 at the exponential and transitional phases of growth, and quantitative RT-PCR was used to assay the crgR message. The results were normalized to the amount of proS mRNA in the same RNA preparation. From two separate RNA preparations in each growth phase, no difference in the amount of the crgR message was detectable (data not shown). We conclude, therefore, that the simple hypothesis is incorrect. Instead, it appears that one or more genes other than crgR, which are required for growth in the presence of the antimicrobial cationic peptide, are repressed by CovR. Because crgR is predicted to be a transcriptional regulator whose regulon has not yet been defined, further work is required to identify the Cov regulon gene(s) needed for GAS growth in the presence of LL-37.
Conclusions. Other bacteria that lack stress sigma factors, including the important human pathogen Streptococcus pneumoniae and the dental pathogens Streptococcus mutans and Streptococcus sanguinis, encode CovRS homologs. In the case of S. mutans, as well as in GAS (which has a CovRS homologous system as well as a stress sigma factor), this TCS has also been implicated in response to stress (51). Thus, an increased understanding of the CovS mediation of the stress response may also provide a model for the mechanism used in these other pathogens as well. The studies reported above were undertaken to determine whether CovRS is required for GAS growth under the additional stress conditions encountered during human infection. We found that in terms of stress provided by a low concentration of iron and by the presence of the human antimicrobial peptide LL-37, CovS is required for growth of GAS and that the deletion of CovR overcomes this effect. Thus, in addition to 39°C, pH 6, and a low concentration of salt, a low concentration of iron and the presence of LL-37 appear to stimulate CovS phosphatase activity so that CovS inactivates the CovR repression of genes required for growth. The above work further supports the pivotal role CovRS plays in enabling GAS to grow under the many stress conditions that it encounters as it infects its host. A greater understanding of the CovRS system may suggest an effective approach to prevent this pathogen from causing serious human disease.
ADDENDUM IN PROOF While this paper was being reviewed, Wessels' group published experiments extending our findings on LL37 to additional strains of GAS (I. Gryllos, H. J. Tran-Winkler, M. F. Cheng, H. Chung, R. Bolcome, W. Lu, R. I. Lehrer, and M. R. Wessels, Proc. Natl. Acad. Sci. 105:16755-16760, 2008). They found that addition of LL-37 to the medium increased the expression of several Cov-regulon virulence genes in a CovS-dependent manner. Although they did not address the role of CovR in this derepression, their results are completely consistent with our conclusion that growth with LL-37 produces membrane stress, leading to activation of the phosphatase activity of CovS, which results in derepression of Cov-regulon genes.
This work was supported in part by NIH grant AI20723, and C.B. was supported in part by grant K12 GM000680 from Emory University and the Atlanta University Center.
Published ahead of print on 7 November 2008. ![]()
|
|
|---|
X factor regulates modification of the cell envelope and resistance to cationic antimicrobial peptides. J. Bacteriol. 186:1136-1146.
X in Streptococcus pyogenes is restricted at two levels. J. Bacteriol. 185:4291-4297.This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»