Regulation of Staphylococcus aureus Type 5 and Type 8 Capsular Polysaccharides by CO₂

Abstract

Staphylococcus aureus expression of capsular polysaccharide type 5 (CP5) has been shown to be downregulated by CO₂. Here we show that CO₂ reduces CP5 expression at the transcriptional level and that CO₂ regulates CP8 expression depending on the genetic background of the strains. Growth in the presence of air supplemented with 5% CO₂ caused a significant decrease in CP8 expression in four S. aureus strains, a marginal effect in four strains, and higher CP8 expression in strain Becker. Absolute CP8 expression in the nine S. aureus strains differed largely from strain to strain. Four groups of strains were established due to sequence variations in the promoter region of cap5 and cap8. To test whether these sequence variations are responsible for the different responses to CO₂, promoter regions from selected strains were fused to the reporter gene xylE in pLC4, and the plasmids were electrotransformed into strains Becker and Newman. XylE activity was negatively regulated by CO₂ in all derivatives of strain Newman and was always positively regulated by CO₂ in all derivatives of strain Becker. Differences in promoter sequences did not influence the pattern of CP8 expression. Therefore, the genetic background of the strains rather than differences in the promoter sequence determines the CO₂ response. trans-acting regulatory molecules may be differentially expressed in strain Becker versus strain Newman. The strain dependency of the CP8 expression established in vitro was also seen in lung tissue sections of patients with cystic fibrosis infected with CP8-positive S. aureus strains.

Staphylococcus aureus is a pathogen which causes a number of serious human diseases, such as endocarditis, osteomyelitis, skin abcesses, and chronic endobronchial infections in patients with cystic fibrosis (CF) (15, 20). Several extracellular and cell surface-bound components act as virulence factors in S. aureus, including capsular polysaccharides (CPs) (23, 30). Although S. aureus strains can produce 11 serologically distinct CPs (16, 29), the majority of clinical isolates of this pathogen have been described as CP5 or CP8 positive (1, 3, 4, 14). Previously, however, we showed that S. aureus strains producing CP5 (12) or CP8 (22) in vitro lack these polysaccharides when directly examined by immunofluorescence microscopy of thin airway sections from CF patients. CP5 was reexpressed when the isolates were grown under normal air conditions, whereas the addition of 1% CO₂ rendered the strains CP5 negative. Since the mean value of the inspiratory and expiratory CO₂ in the bronchioli is about 4% (11), CP5 expression in vivo may be inhibited due to the elevated pCO₂ compared to the pCO₂ in normal air (0.03%). In contrast to the negative effect of CO₂ on CP5 expression, it was previously shown that α- and β-hemolysin expression, as well as expression of S. aureus toxic shock syndrome toxin 1 (17), is increased in the presence of elevated CO₂ concentrations (6, 7, 24).

CO₂ also regulates the expression of surface components in other bacteria. For example, the expression of the fibrillar surface M protein of Streptococcus pyogenes (5) and the capsule of Bacillus anthracis (18, 21) is increased in the presence of elevated CO₂ concentrations. Additionally, capsule synthesis in Cryptococcus neoformans is positively regulated by CO₂ (10). In contrast, the production of slime by Staphylococcus epidermidis is decreased when the bacteria are incubated with 5% CO₂ (8, 25). These observations show that CO₂ is an important environmental signal for many bacteria and appears to be involved in regulating virulence factors in more than one manner.

The molecular basis for the regulation of S. aureus CP5 by CO₂ is still unresolved. We wanted to clarify our observations at a transcriptional level and to extend our previous observations to CP8-expressing S. aureus strains. The CP5-producing strains S. aureus Reynolds and Newman, the CP8-positive strain Becker, and several clinical S. aureus isolates from CF patients were used.

MATERIALS AND METHODS

Sequencing of cap promoter.

The cap5 and -8 promoter regions from 11 S. aureus strains were amplified by PCR (Advantage kit; Clontech) using two primers, GAATTCGGTCAATCAGTCGGAATT (EcoRI site is underlined) and AAGCTTCAAGTTTTTTTGTAATA (HindIII site is underlined). The amplified fragments correspond to −512 to +71 of the published cap8 sequence of strain Becker, with +1 being the transcriptional start (27, 28). The PCR-amplified DNA fragments were cloned into the pGEM-T vector (Promega) and sequenced. To avoid possible errors generated by PCR, two independent PCR amplifications from each strain were performed. No mismatch was found between the two independent PCR amplifications from each strain. Sequences from all 11 strains are 583 bp in length, except that from strain Becker, which had a 1-bp deletion. The sequences were then compared by the Clustal method (13) of the MegAlign program (Dnastar Inc.).

cap promoter fusion.

The lipase-negative S. aureus strain Reynolds was supplemented with the promoter test plasmid pPS11cap5 containing a 424-bp fragment of the cap5 promoter fused to the lipase gene of Staphylococcus hyicus. A cap5 promoter fragment (424 bp) was generated by PCR using DNA from S. aureus strain Reynolds. The cap8 primer pairs (upper primer, TTTGGATCCAACTAATCCTAAAGAAGCACTAA; lower primer, CTCCATTTATAACCTTTCATGAACCTAGGTTT) were selected to span nucleotides 32 to 456 of the published sequence (27), with artificial BamHI cleavage sites (underlined) at the 5′ ends. The cap8 primer pairs were selected for the cap5 promoter PCR due to the availability of only the cap8 sequence data at that time and the expected high homology between the cap5 and cap8 sequences (27). Additionally, the cap start codon (in bold) was changed in the lower primer. The PCR fragment was digested with BamHI and ligated into the BamHI site of pPS11 (31) containing a promoterless lipase gene (lip) of S. hyicus and a chloramphenicol resistance gene. The ligation mixture was used for protoplast transformation into Staphylococcus carnosus (9). Single, chloramphenicol-resistant colonies were grown on lipase test plates (Tributyrin agar base supplemented with 1% Tween 20; Merck, Darmstadt, Germany). Insertion of the cap5 promoter region in pPS11cap5 was confirmed by sequencing (LI-COR, model 40002; WG-Biotech, Ebersberg, Germany). The plasmid pPS11cap5 was electrotransformed into S. aureus strain Reynolds as previously described (2). Thus, strain Reynolds contains, besides the natural chromosomal cap5 promoter, additional multicopy promoters on the extrachromosomal plasmid. For the measurement of the CP5 promoter activity, strain Reynolds containing pPS11cap5 was grown to an optical density at 600 nm (OD₆₀₀) of 7.4 in air and in air supplemented with 5% CO₂, and lipase activity in the culture supernatant fluid was determined as previously described (31).

The promoter regions from strains CF1, CF4, CF6, CF12, and Becker were fused to the promoterless reporter gene xylE in pLC4 (26), which resulted in plasmids pCL8388, -8389, -8390, -8396, and -8420, respectively. The cap promoter fragments were obtained by PCR (for primers, see above). The resultant plasmids were electrotransformed into strains Becker (a type 8 capsule strain) and Newman (a type 5 capsule strain) and were incubated in Luria-Bertani broth for 18 h at 37°C with air or air supplemented with 5% CO₂. The XylE activities were assayed to measure the promoter activities (32).

ELISA for detection of CP8.

Bacterial cells were diluted from an overnight culture to an OD₆₀₀ of 0.05 in 30 ml of tryptic soy broth (Oxoid, Hampshire, United Kingdom). The bacteria were incubated with shaking at 37°C under air or air with 5% CO₂ (Aerotron incubator; Infors, Einsbach, Germany) for 16 h. CP8 expression of clinical isolates of S. aureus was assessed by a two-step inhibition enzyme-linked immunosorbent assay (ELISA) (3). Microtiter wells were coated with 5% gelatin in phosphate-buffered saline (PBS) at 37°C for 1 h. After washing with PBS-Tween 20, the wells were incubated with 100-μl volumes of washed S. aureus cells and 100 μl of an anti-CP8 monoclonal immunoglobulin G3 (IgG3) antibody diluted in PBS-Tween supplemented with 0.5% gelatin at a concentration giving an OD₄₉₂ of 0.2 to 0.5, which was determined by preliminary titration. For CP8 antibody production, BALB/c mice were immunized with 5 × 10⁷ cells of S. aureus strain Becker as previously described (3). After incubation at 37°C for 1 h and then overnight at 4°C, 100-μl samples from each well were transferred to another plate which had previously been coated with purified CP8 and blocked with gelatin. This plate was incubated at 37°C for 1 h, and after washing with PBS-Tween, an anti-mouse peroxidase-conjugated IgG (Diagnostics Pasteur) was added to the wells and the plate was incubated at 37°C for 45 min. After washing, enzyme substrate (o-phenylenediamine dihydrochloride; Dako, Copenhagen, Denmark) was added, and after 10 min at room temperature, the reaction was stopped and the OD₄₉₂ was read. For each ELISA run, negative controls were used (wells not receiving test samples but receiving PBS-Tween supplemented with 0.5% gelatin) and titration of purified CP8 was performed to determine the assay sensitivity. The amount of CP8 in the samples was determined from standard titration curves of purified CP8 and expressed in nanograms per milliliter. The lower limit of the assay is 1 ng of CP8/ml.

Detection of CP8 and teichoic acid by immunofluorescence.

CP8 production was assessed by indirect immunofluorescence using monoclonal antibodies (IgG3; Institut Pasteur, Paris, France) and fluorescein isothiocyanate (FITC)-conjugated IgG rabbit antibodies against mouse IgG (Dako). Cryostat thin sections (5 to 10 μm) were prepared (Kryostat 2800 Frigocut E; Reichert-Jung, Heidelberg, Germany) from shock-frozen lung tissue material from two CF patients. The thin sections were fixed on slides with acetone for 10 min, incubated for 20 min with normal rabbit serum (Dako) diluted 1:5, and incubated with anti-CP8 antibody (final dilution, 33 μg/ml) for 1 h at room temperature. After washing, slides were incubated with FITC-conjugated antibodies and diluted 1:40 for 30 min at room temperature. After washing, slides were mounted with Permafluor (Dako) for 24 h and visualized using a fluorescence microscope (Axioplan; Zeiss, Oberkochen, Germany). Teichoic acid expression on S. aureus strains was determined using a rabbit antiserum. Slides were preincubated with swine antiserum and diluted 1:5, and a FITC-conjugated swine antibody against rabbit IgG, diluted 1:40, was used (Dako). The rabbit serum against teichoic acid (SL-39) was prepared by immunizing the animals with a killed S. aureus strain completely lacking CP5 and CP8. The serum did not contain antibodies against CP5 or CP8 as demonstrated by ELISA.

RESULTS AND DISCUSSION

Regulation of CP5 by CO₂.

Previously, we have shown that the CP5 expression of strain Reynolds is inhibited by the growth of S. aureus strains in air supplemented with 5% CO₂. This was further confirmed using two other CP5 strains (CF1 and Newman) which both showed a significant decrease of capsular material on the surface after growth with 5% CO₂ (Table 1). To analyze whether this is mediated by downregulation of the promoter activity, we cloned the cap5 promoter in front of the lipase gene and compared the lipase activity in strain Reynolds(pPS11cap5). After growth in the presence of air supplemented with 5% CO₂, the lipase activity was 60% lower compared to that for growth under normal air conditions (P < 0.001; Mann-Whitney test) (Fig. 1). Thus, CO₂ affects cap5 gene expression at the transcriptional level. The ratio obtained from the promoter fusion is lower than those previously obtained by ELISA, possibly due to the presence of the cap5 promoter in multiple copies in the promoter test assay.

TABLE 1.

Quantitation of CP on the surface of S. aureus strains

Strain	CP production (ng of CP/109 CFU)a in:		Ratio of CP production (air/air plus CO2)	Pb value
	Air	Air plus CO2
Becker	1,550 ± 580	2,250 ± 866	0.7	0.228
CF4	113 ± 64	5 ± 5	22.6	0.015
CF6	15.3 ± 12.5	1 ± 0	15.3	0.063
CF7	1,867 ± 611	1,333 ± 569	1.4	0.330
CF8	1,225 ± 556	628 ± 126	1.95	0.081
CF9	373 ± 424	93 ± 40	4.0	0.237
CF10	953 ± 832	323 ± 267	2.95	0.199
CF11	5,150 ± 3,553	7.8 ± 8.7	660	0.028
CF12	2,125 ± 2,069	50 ± 52	42.5	0.022
Newman	1,080 ± 170	180 ± 85	6
CF1	2,200 ± 283	240 ± 0	9.17

^aStrains were grown in tryptic soy broth under normal air conditions or in air supplemented with 5% CO₂ with shaking, and aliquots were taken after 15 h. Aliquots of 10⁹ cells were assayed for cell-bound CP antigen by ELISA (3). Values represent the means ± standard deviations of two independent growth cultures. 

^bA two-tailed unpaired t test was used to calculate P values. 

FIG. 1.

Downregulation of the cap5 promoter under air plus 5% CO₂ growth conditions. The lipase-negative S. aureus strain Reynolds was supplemented with the promoter test plasmid pPS11cap5 containing a 424-bp fragment of the cap5 promoter fused to the lipase gene of S. hyicus. For details, see Materials and Methods.

Regulation of CP8 by CO₂.

Next, we wanted to know whether CP8 is influenced in the same manner by CO₂. In contrast to CP5, quantitative detection of the CP8 antigen by ELISA on type 8 bacterial cells gave conflicting results with respect to CO₂ regulation when several strains were tested (Table 1). In only four of nine CP8-positive S. aureus strains examined (CF4, CF6, CF11, and CF12), a significant decrease in CP8 expression was found when CP8-positive S. aureus strains were grown in the presence of air supplemented with 5% CO₂ compared to cells grown under normal air conditions. In other strains, the effect of CO₂ was less pronounced, and in one case, the type 8 prototypic strain Becker, CP8 expression was higher under supplemented CO₂ growth conditions. The results also show that absolute CP8 expression in S. aureus strains as well as the regulatory CO₂ effect may vary considerably from one strain to another. For example, strain CF7 produces about 1 order of magnitude more CP8 than strain CF4 does. The reason for these effects may be related to sequence variations in the promoter region of cap or in other genes which mediate cap transcription.

cap promoter sequence.

To analyze whether differences in the promoter region account for differences in the CO₂ effect, the upstream sequences of eight CP8 strains and two CP5 strains (CF1 and Reynolds) were sequenced. As shown in Fig. 2, most of the mismatches were found between nucleotides −231 and −89. Based on the sequence comparison, the strains can be grouped into the following four groups: group 1, Reynolds, CF1 (type 5 strain), and CF11, with identical sequences; group 2, CF4, CF8, CF9, CF10, and CF12, with identical sequences except CF12 has one mismatch; group 3, CF6 and CF7, with three mismatches; and group 4, Becker (Fig. 2). Interestingly, the promoter sequence of one CP8 strain (CF11) was identical to the sequence derived from the CP5 strains Reynolds and CF1 but different from that of the other CP8 strains.

FIG. 2.

DNA sequence alignment of the cap5 and -8 promoter regions from various S. aureus strains. A region of 583 bp from each strain was compared, but only sequences corresponding to positions −56 to −455 (indicated by arrowheads) with respect to the transcriptional start site of the cap8 sequence of strain Becker are shown. Mismatched sequences are boxed. We found no mismatches between strains in the sequences that were not shown. Note that CF1 and strain Reynolds are type 5 strains.

Activity of different cap promoters in strains Newman and Becker.

To test whether differences in the cap upstream sequences are responsible for the different responses to CO₂, we fused each of the promoter regions from strains CF1, CF4, CF6, CF12, and Becker to the promoterless reporter gene xylE, which resulted in plasmids pCL8388, -8389, -8390, -8396, and -8420, respectively. These strains were selected as representatives of the four groups defined by sequencing. Strain Becker (a type 8 capsule strain) and strain Newman (a type 5 capsule strain) containing the cap promoter fusion plasmids were incubated with air or air supplemented with 5% CO₂, and the XylE activities were measured (Table 2). Interestingly, XylE activity was negatively regulated by CO₂ in all derivatives of strain Newman containing the various promoter sequences in front of xylE. In contrast, with strain Becker as the genetic background, XylE activity was always positively correlated with CO₂ pressure. The difference in the promoter sequences used in the different constructs did not influence the pattern of CO₂ regulation. For instance, the promoter fusions derived from strain Becker resulted in enhanced XylE activity in strain Becker(pSN8420) but in decreased activity in strain Newman(pSN8420) after growth of the strains with 5% CO₂. Therefore, the genetic background of the strains rather than differences in the promoter sequence determines the CO₂ response. trans-acting regulatory molecules such as transcriptional activators or sigma factors may be differentially expressed in strain Becker versus strain Newman.

TABLE 2.

Promoter activities as measured by the XylE assay

Groupa	Strain	XylE activity (mU/mg of protein)b in:		Ratio of XylE activity (air/air plus CO2)
		Air	Air plus CO2
1 (CF1)	Becker(pSN8388)	3.83 ± 0.68	5.91 ± 1.45	0.65
2 (CF4)	Becker(pSN8389)	3.66 ± 1.46	6.51 ± 1.10	0.56
3 (CF6)	Becker(pSN8390)	2.44 ± 0.44	3.12 ± 0.43	0.78
2 (CF12)	Becker(pSN8396)	5.34 ± 0.72	6.81 ± 0.84	0.78
4 (Becker)	Becker(pSN8420)	7.70 ± 1.13	10.21 ± 0.88	0.75
1 (CF1)	Newman(pSN8388)	5.52 ± 0.29	3.57 ± 0.37	1.55
2 (CF4)	Newman(pSN8389)	6.13 ± 0.96	3.72 ± 0.66	1.55
3 (CF6)	Newman(pSN8390)	6.66 ± 1.10	3.28 ± 0.27	2.03
2 (CF12)	Newman(pSN8396)	6.41 ± 0.99	3.63 ± 0.30	1.77
4 (Becker)	Newman(pSN8420)	1.17 ± 0.24	0.67 ± 0.08	1.76

^aFor group definitions, see the text. The strain origin of the cap fragment is indicated in parentheses. 

^bXylE activities of the fusion plasmids are expressed as means ± standard deviations of at least three independent tests. 

CP8 expression in lung tissue sections of CF patients.

Previously, we postulated that elevated CO₂ during lung infections in patients with CF may account for the downregulation of CP5 during infection (12). Since CP8 expression is only marginally affected (CF7, CF8, CF9, and CF10) or even enhanced (strain Becker) by CO₂, it may be assumed that CP8-producing S. aureus strains are CP8 positive during infection. Indeed, CP8-positive S. aureus has been detected in experimental endocarditis (19) and in our own investigations (1). Here we demonstrate that the strain dependency of CP8 expression established by in vitro tests is also seen in lung tissue sections of CF patients infected with CP8-positive S. aureus strains. As shown in Fig. 3, strain CF7, which was only marginally affected by CO₂ in regards to CP8, also expressed CP8 in vitro in the airway lumen of the patient. In contrast, strain CF12, which had significantly reduced CP8 expression by CO₂ in vitro (Table 1) did not express CP8 in vivo. In summary, the regulation of CP8 seems to be more complex than that of CP5 in S. aureus.

FIG. 3.

Expression of S. aureus CP8 in lung tissue sections of two CF patients. Lung tissue sections filled with inflammatory plaques from CF patients 7 (A and C) and 12 (B and D) infected with S. aureus strains CF7 and CF12, respectively, were stained with a monoclonal antibody against CP8 (A and B) and a polyclonal rabbit antibody against teichoic acid (C and D) followed by FITC-conjugated anti-mouse (A and B) and anti-rabbit (C and D) antibodies. Note the absence of CP8 in panel B and the presence of CP8 in panel D. Magnification, ×1,000; bars = 10 μm.