Search tips
Search criteria 


Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. 2012 May; 194(10): 2479–2490.
PMCID: PMC3347183

Regulation and Function of Pilus Island 1 in Group B Streptococcus


Group B streptococcus (GBS) pili may enhance colonization and infection by mediating bacterial adhesion to host cells, invasion across endothelial and epithelial barriers, and resistance to bacterial ingestion and killing by host phagocytes. However, it remains unclear how pilus expression is regulated and how modulation of pilus production affects GBS interactions with the human host. We investigated the regulation and function of pilus island 1 (PI-1) pili in GBS strain 2603. We found that PI-1 gene expression was controlled by the CsrRS two-component system, by Ape1, an AraC-type regulator encoded by a divergently transcribed gene immediately upstream of PI-1, and by environmental pH. The response regulator CsrR repressed expression of Ape1, which is an activator of PI-1 gene expression. In addition, CsrR repressed PI-1 gene expression directly, independent of its regulation of Ape1. In vitro assays demonstrated specific binding of both CsrR and Ape1 to chromosomal DNA sequences upstream of PI-1. Pilus gene expression was activated by acidic pH, and this effect was independent of CsrRS and Ape1. Unexpectedly, characterization of PI-1 deletion mutants revealed that PI-1 pili do not mediate adhesion of strain 2603 to A549 respiratory epithelial cells, ME180 cervical cells, or VK2 vaginal cells in vitro. PI-1 pili reduced internalization and intracellular killing of GBS by human monocyte-derived macrophages, by approximately 50%, but did not influence complement-mediated opsonophagocytic killing by human neutrophils. These findings shed new light on the complex nature of pilus regulation and function in modulating GBS interactions with the human host.


Pilus-like appendages on Gram-positive bacteria were first noted in the 1960s and 1970s, in morphological studies of corynebacteria (33, 34). More recent reports identified pili on other Gram-positive bacteria, notably the pathogenic streptococci group B streptococcus (GBS), Streptococcus pyogenes, and Streptococcus pneumoniae (2, 16, 21). After the first report of GBS pili in 2005, initial studies seemed to confirm expectations that these structures participated in bacterial attachment to host cells and contributed to the organism's ability to colonize and invade the infected host (3, 18). In addition, GBS pilus proteins were found to be immunogenic in animals, evoking opsonic antibodies that conferred protection against experimental infection (17, 28). The latter studies have generated great interest in pilus proteins as potential vaccine antigens (20).

Two distinct genomic loci have been identified in GBS, one or both of which can harbor a pilus gene cluster or island (3, 28). Pilus island 1 (PI-1), present in 72% of strains, contains genes encoding three structural pilus proteins, two sortases, and a divergently transcribed putative AraC-type transcriptional regulator (Fig. 1). The second pilus locus is located at a separate chromosomal site and is present in one of two variant forms in all GBS strains: PI-2a in 73% of strains and PI-2b in 27% of strains (20). PI-2a includes genes encoding three pilus proteins, two sortases, and a RogB family transcriptional regulator (Fig. 1). The last gene is absent in PI-2b, which instead encodes a putative LepA-type signal peptidase. The structural pilus proteins encoded by each island include a major or backbone component that is polymerized to form the shaft of the pilus fiber and two minor or accessory proteins that are associated with the polymer and/or the cell surface. In this paper, we use the pilus protein nomenclature suggested by Pezzicoli et al. (26): “BP” for backbone protein and “AP1” or “AP2” for ancillary protein 1 or 2, followed by a hyphen and “1,” “2a,” or “2b” to designate the pilus island (e.g., BP-1 for the backbone protein of pilus island 1 and AP1-2a for ancillary protein 1 of pilus island 2a).

Fig 1
Schematic of the chromosomal regions of GBS strain 2603 that contain the pilus loci (PI-1 and PI-2a). The striped arrows represent genes coding for LPXTG-containing cell surface anchor family proteins. The checked arrows represent genes coding for sortase ...

Recent reports have suggested that the role of GBS pili in pathogenesis may not be as clear-cut as implied by earlier studies. A detailed investigation of the contribution of pili to biofilm formation concluded that PI-2a pili, but not PI-1 or PI-2b pili, mediate formation of biofilms on both abiotic and biotic surfaces (27). Another study, using the background of type III GBS strain NEM316, found no role for BP-2a (PilB) in resistance to phagocytosis and intracellular killing by macrophages, in contrast to earlier work (19, 23). The same study concluded that PI-2a pili contributed to virulence in a neonatal infection model but not in adult mice. These reports reveal that the specific functions of GBS pili may depend on pilus type, level of expression, and particular host and environmental factors.

Our goals in the present investigation were to examine the functions of PI-1 pili and to characterize the regulatory mechanisms that control pilus expression, an area little explored up to the present. We chose to work with type V GBS strain 2603V/R (referred to henceforth as strain 2603) because it harbors inactivating mutations in RogB, the transcriptional activator associated with PI-2a, and in Rga, another activator of PI-2a gene expression (3, 4, 29). Presumably as a consequence of one or both mutations, strain 2603 produces undetectable amounts of PI-2a pili. Thus, strain 2603 effectively produces only PI-1 pili, a feature that simplifies analysis of pilus function. We identified three factors as critical regulators of PI-1 pilus gene expression: the CsrRS two-component system, environmental pH, and Ape1, the AraC family regulator associated with PI-1. Our data suggest that PI-1 pili inhibit both uptake and intracellular killing of GBS by macrophages. However, in contrast to a previous report, we found no evidence that PI-1 pili of strain 2603 mediate GBS attachment to epithelial cells.


Bacterial strains and growth conditions.

Bacterial strains and plasmids are listed in Table 1. GBS type V strain 2603V/R and mutant strains 2603 ΔcsrR and 2603 ΔcsrS have been described previously (8). Escherichia coli M15(pREP4) and Lactococcus lactis NZ9000 (12) were used for recombinant protein expression. Unless otherwise specified, GBS was grown in Todd-Hewitt broth (THB; Difco) or on Trypticase soy agar (TSA) supplemented with 5% defibrinated sheep blood (PML Microbiologicals). For experiments testing the effects of pH, GBS was cultured at 37°C in complex medium (CM; 10 g/liter proteose peptone, 5 g/liter Trypticase peptone, 5 g/liter yeast extract, 2.5 g/liter KCl, 1 mM urea, 1 mM arginine) (30). E. coli was grown in Luria-Bertani broth, and L. lactis was grown in GM17 broth (M17 broth [Oxoid] supplemented with 0.5% glucose) (1). When appropriate, antibiotics were used at the following concentrations: for E. coli, erythromycin (erm) at 200 μg/ml, kanamycin at 25 μg/ml, and ampicillin at 100 μg/ml; for L. lactis, chloramphenicol at 10 μg/ml; and for GBS, erm at 1 μg/ml and chloramphenicol at 10 μg/ml.

Table 1
Bacterial strains and plasmids used in this study

Construction of ape1 mutagenesis plasmid.

To introduce an internal deletion into ape1, primers 1197 and 1215 were used to PCR amplify the first 117 bp of ape1 and 947 bp of adjacent upstream flanking sequence, using GBS strain 2603 chromosomal DNA as the template. (Primer sequences are listed in Table S1 in the supplemental material.) Primers 1196 and 1214 were used to amplify the last 129 bp of ape1 and 841 bp of downstream flanking DNA. Primer 1196 contains 18 bp of DNA that is complementary to primer 1197. The two gel-purified PCR products containing complementary ends were mixed and amplified with primers 1214 and 1215 to create a 963-bp internal deletion of ape1 by overlap PCR (5, 7). The 2,034-bp overlap PCR product was digested with BamHI and KpnI and ligated into BamHI/KpnI-digested pJRS233.

Construction of mutagenesis plasmid for deletion of sag0649, encoding AP1-1.

To introduce an internal deletion into sag0649, encoding AP1-1, primers 1278 and 1279 were used to PCR amplify the first 117 bp of sag0649 and 814 bp of adjacent upstream flanking sequence, using GBS strain 2603 chromosomal DNA as the template. Primers 1280 and 1281 were used to amplify the last 138 bp of sag0649 and 785 bp of downstream flanking DNA. Primer 1280 contains 19 bp of DNA that is complementary to primer 1279. The two gel-purified PCR products containing complementary ends were mixed and amplified with primers 1278 and 1281 to create a 2,418-bp internal deletion of sag0649 by overlap PCR. The 1,854-bp overlap PCR product was digested with KpnI and PstI and ligated into KpnI/PstI-digested pJRS233.

Construction of mutagenesis plasmid for deletion of PI-1 gene cluster.

For construction of a plasmid to delete the entire PI-1 gene cluster (sag0645 to -0649), primers 1286 and 1295 were used to PCR amplify the first 309 bp of sag0645 and 788 bp of adjacent upstream flanking sequence, using GBS strain 2603 chromosomal DNA as the template. Primers 1296 and 1294 were used to amplify the last 134 bp of sag0649 and 937 bp of downstream flanking DNA. Primer 1296 contains 18 bp of DNA that is complementary to primer 1295. The two gel-purified PCR products containing complementary ends were mixed and amplified with primers 1286 and 1294 to create a 6,711-bp internal deletion of the pilus gene cluster by overlap PCR. The 2,168-bp overlap PCR product was digested with BamHI and KpnI and ligated into BamHI/KpnI-digested pJRS233.

Construction of GBS mutants 2603 Δape1, 2603 ΔcsrR Δape1, 2603 ΔAP1-1, 2603 ΔcsrR ΔAP1-1, 2603 ΔPI-1, and 2603 ΔcsrR ΔPI-1.

The deletion construct in plasmid pJRS233 was introduced into GBS strain 2603 or 2603 ΔcsrR by electroporation, and transformants were selected by growth at 30°C in the presence of erm. A single erm-resistant colony was used to inoculate a liquid culture supplemented with erm. After overnight incubation at 30°C, the culture was diluted 10-fold with fresh broth containing erm and incubated at 37°C to select organisms in which the recombinant plasmid had integrated into the GBS chromosome by homologous recombination. Dilutions of each culture were plated on medium containing erm and incubated overnight; erm-resistant colonies representing plasmid integrants were serially passaged twice on solid medium at 37°C. Integrant strains were serially passaged at least five times in broth at 30°C in the absence of erm; excision of the plasmid from the chromosome via a second recombination event either completed the allelic exchange or reconstituted the wild-type genotype. erm-sensitive colonies were screened for the expected deletion mutation by PCR amplification using primer pairs that flanked the target gene(s).

Quantitative reverse transcription-PCR (RT-PCR).

GBS strains grown overnight on TSA blood agar were inoculated into 10 ml THB. The GBS cells were harvested when the culture density reached an A600 of 0.3 and were collected by centrifugation (3,200 × g, 5 min). For experiments testing the effects of pH, GBS cells were inoculated into 10 ml CM (pH 7.4) and grown to an A600 of 0.35. The bacterial culture was then divided into two tubes. GBS cells were collected by centrifugation, resuspended with CM at either pH 7.4 or pH 5.0, and incubated for 30 min at 37°C.

For RNA isolation, the GBS cell pellet was resuspended in 0.5 ml 0.9% NaCl and 1 ml RNA Protect buffer (Qiagen) and kept at room temperature for 5 min. After centrifugation, the bacterial pellet was treated with 100 U mutanolysin (Sigma) and 15 mg/ml lysozyme (Sigma) in Tris-EDTA (TE) buffer, pH 8.0, in a final volume of 100 μl. Total bacterial RNA was then isolated using an RNeasy miniprep kit (Qiagen) according to the manufacturer's instructions. RNA samples were treated with DNase I (Invitrogen) for 60 min at 37°C to remove contaminating DNA. The RNA concentration was adjusted to 100 ng/μl, and samples were stored at −80°C until use (9).

Purified RNA and gene-specific primer pairs were used for real-time PCR assays using SYBR green reagents (Qiagen) and a model 7300 RealTime PCR system (Applied Biosystems) according to the manufacturers' instructions. Duplicate reaction mixtures containing 50 ng of template RNA were used for the genes of interest. To exclude DNA contamination of RNA samples, replicate control samples were assayed in the absence of reverse transcriptase. Standard curves were obtained by performing PCR with SYBR green detection on serial dilutions of spectrophotometrically quantified genomic DNA. The housekeeping gene recA was used as a reference control gene to normalize experimental results. The following conditions were used for real-time PCR: 1 cycle of 30 min at 50°C, 1 cycle of activation at 95°C for 15 min, and 36 cycles of PCR amplification, with denaturation at 95°C for 15 s, annealing at 55°C for 30 s, and extension at 72°C for 30 s. Relative gene expression was quantified by the comparative threshold cycle (CT) method (ΔΔCT method) according to the manufacturer's guidelines (

Whole-cell immunoblotting.

GBS strains were grown in THB to late exponential phase (A600 = 0.5). GBS cells collected from 1 ml of culture were washed once in phosphate-buffered saline (PBS) and resuspended in 200 μl PBS. Samples of 5 μl were spotted on nitrocellulose membranes and then fixed by heating for 1 h at 60°C. Membranes were blocked in TBS (1× PBS with 0.05% Tween 20) containing 5% skim milk for 1 h, followed by washing three times in TBS. Membranes were then incubated for 1.5 h at room temperature with primary antibody (rabbit antiserum to either recombinant BP-1 [GBS80] or recombinant AP1-2a [GBS67]) diluted 1:10,000. Membranes were washed three times in TBS and then incubated for 1 h with goat anti-rabbit IgG conjugated to alkaline phosphatase or horseradish peroxidase and diluted 1:10,000. Membranes were washed three times in TBS. Positive bands were visualized with the addition of 5-bromo-4-chloro-3-indolylphosphate–nitroblue tetrazolium (BCIP/NBT; Sigma) or peroxidase substrate (Pierce).

Cell wall protein extracts.

GBS cells were grown in THB at 37°C and harvested for protein analysis at late exponential phase (A600 = 0.5). Bacterial cells were washed twice with 40 mM sodium phosphate buffer (pH 7.0) and then treated with protoplast solution (20% sucrose, 10 mM MgCl2, 20 mM Tris, pH 7.0, 0.05% Triton X-100, protease inhibitor cocktail [Roche Diagnostics], 500 U/ml mutanolysin). The digestion was carried out overnight under gentle agitation. After centrifugation at 13,000 × g for 15 min at 4°C, the supernatant, containing proteins released from the cell wall, was collected and stored at 4°C.

Western immunoblotting.

For immunoblotting, protein preparations were fractionated by SDS-PAGE under reducing conditions, using a NuPAGE 12% Bis-Tris gel (Invitrogen), and then transferred onto a nitrocellulose membrane. The membrane was incubated in TBS-5% milk to block nonspecific binding, followed by washing three times in TBS. Primary antibody (rabbit antiserum to either BP-1 or AP1-2a; courtesy of Immaculada Margarit y Ros) was added at 1:10,000 in TBS for 1 h at room temperature. After being washed three times in TBS, membranes were incubated for 1 h with goat anti-rabbit IgG conjugated to alkaline phosphatase or horseradish peroxidase and diluted 1:5,000. Membranes were washed three times in TBS. Positive bands were visualized with the addition of BCIP/NBT or peroxidase substrate.

Expression of His6-CsrR in pQE-30.

The plasmid vector pQE-30 was used for GBS CsrR protein overexpression in E. coli. A 687-bp PCR-generated BamHI/HindIII DNA fragment corresponding to the csrR coding sequence was amplified using the specific primers 880 and 881 and then cloned between the BamHI and HindIII sites of pQE-30 to give pQE30-csrR. The resulting clone was first transformed into E. coli DH5α for amplification and then isolated and introduced into E. coli M15 bearing a plasmid carrying the lac repressor (pREP-4). The transformants were inoculated into LB medium containing both ampicillin (100 μg/ml) and kanamycin (25 μg/ml) and grown at 37°C until an A600 of 0.5 to 0.7. Expression of His6-CsrR was induced by the addition of isopropyl-β-d-thiogalactopyranoside (IPTG) to 1 mM, followed by 3 h of incubation at 37°C.

Expression of Ape1-His6 fusion protein in PNZ8048.

The ape1 gene was PCR amplified using primers 1276 and 1277, with the latter including a sequence coding for six histidine residues at the 5′ terminus. The resulting PCR product was digested by NcoI and SacI and then ligated to similarly digested PNZ8048, using T4 DNA ligase. Plasmid pNZ8048 contains the nisin-inducible PnisA promoter with a downstream start codon (ATG) within an NcoI restriction site (CCATGG). The ligation reaction mixture was transformed into Z-Competent E. coli cells (Zymo Research) following the manufacturer's instructions, and transformants were selected on LB agar containing 10 μg/ml chloramphenicol. After incubation for 24 to 48 h at 30°C, positive colonies were identified by PCR, and the plasmid was extracted from E. coli by use of a Qiagen miniprep kit (Qiagen). The resulting plasmid (pNZ-ape1) had a C-terminally six-histidine-tagged ape1 gene, and the correct nucleotide sequence was confirmed by DNA sequencing (Genewiz). Plasmid pNZ-ape1 was transformed into electrocompetent L. lactis NZ9000, prepared as previously described (6), using a Gene Pulser apparatus (Bio-Rad). Transformants were selected on GM17 agar containing 10 μg/ml chloramphenicol. After 24 h of incubation at 30°C, colonies were analyzed by PCR for the presence of the desired ape1 sequence. An overnight culture of L. lactis NZ9000 (pNZ-ape1) was subcultured into fresh GM17 broth containing 10 μg/ml chloramphenicol and incubated statically at 30°C. Filter-sterilized culture supernatant of the nisin-secreting strain L. lactis NZ9700 was used as a source of nisin (1, 14). Nisin was added when the culture density reached an A600 of 0.5. Induction was conducted at 16°C overnight without shaking, after which cells were pelleted (3,200 × g for 10 min) and washed once with buffer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, pH 8.0) for isolation of the recombinant protein.

Purification of recombinant His6-tagged proteins.

The induced bacterial cells were collected, washed, and then lysed by sonication. Insoluble cell debris was removed by centrifugation (20,800 × g, 20 min). The cell lysate was then passed over a Ni2+-nitrilotriacetic acid (Ni2+-NTA) agarose column for purification according to the manufacturer's instructions (Qiagen). Purity of the affinity-purified proteins was demonstrated by SDS-PAGE as described above and by Coomassie blue staining (see Fig. S1 in the supplemental material). The purified protein was suspended in 50% glycerol and stored at −20°C. The concentration of purified protein was estimated by comparison with a bovine serum albumin standard, using a bicinchoninic acid (BCA) kit (Pierce).

Expression of Ape1 in strain 2603 Δape1.

To overexpress the Ape1 protein in strain 2603 Δape1, the plasmid construct pNZ-ape1 (described above) was introduced by electroporation into electrocompetent GBS 2603 Δape1. Transformants were identified by PCR and confirmed by DNA sequencing of the cloned ape1 gene.

Expression of AP1-1 in Lactococcus lactis NZ9000.

To express the AP1-1 adhesin protein in Lactococcus lactis NZ9000, the gene encoding the protein, sag0649, was amplified by PCR using GBS 2603 chromosomal DNA as the template and using primers 1461 and 1450. The PCR product was digested with BamHI and XbaI and then ligated to similarly digested pNZ-SEC vector, which harbors the P44 promoter from plasmid pNZ44 and the SEC signal sequence from the L. lactis MG1363 chromosome (1). The recombinant plasmid pNZ-SEC-AP-1 was transformed into E. coli DH5α chemically competent cells (Zymo Research). After verification of DNA sequences, the plasmid construct was subsequently transformed into electrocompetent L. lactis NZ9000 cells. Colonies were screened by PCR after 24 h of incubation at 30°C. For identification of the recombinant protein on the lactococcal cell surface, cell wall-anchored proteins were released by the following procedure. L. lactis cells were collected from 20 ml of overnight culture, washed twice with PBS, pH 7.4, resuspended in 0.5 ml of protoplast buffer (50 mM HEPES, 0.01 M MgCl2, 0.5 M sucrose, pH 7.0), and then treated with lysozyme (2 mg/ml) and mutanolysin (60 U/ml) at 37°C for 1 h. Supernatant was collected after centrifugation of protoplasts at 5,000 × g for 45 min at 4°C, and supernatant proteins were analyzed by SDS-PAGE as described above and by Coomassie blue staining. AP1-1 expression in the same protein preparation was also analyzed by Western immunoblotting using a 1:5,000 dilution of rabbit antiserum to AP1-1 (GBS104; courtesy of Hung Ton-That) as described above.

Electrophoretic mobility shift assays (EMSAs).

A 402-bp DNA fragment representing the putative promoter region upstream of sag0645 was amplified by PCR from 2603 chromosomal DNA by use of primers 1245 and 1246. The purified PCR product was end labeled with [γ-32P]dATP in the presence of T4 polynucleotide kinase, using a gel shift assay system kit (Promega). The labeled probe was purified from free nucleotides on a G-25 spin column (Amersham). As a negative control, a similar-size fragment was amplified from the promoter region of the capsule synthesis (cps) locus (9). Purified recombinant His6-CsrR or Ape1-His6 (3.8 μM) was mixed with probe DNA (2.3 nM) in binding buffer [20 mM Tris, pH 7.5, 1 mM CaCl2, 1 mM dithiothreitol (DTT), 10 μg/ml poly(dI-dC), and 100 μg/ml bovine serum albumin] in a total volume of 10 μl. DNA and protein were incubated for 15 min at room temperature. The reaction products were mixed with 2 μl of 50% glycerol and loaded onto a 5% Tris-borate-EDTA (TBE) polyacrylamide gel (Bio-Rad; TBE is 90 mM Tris-borate, 2 mM EDTA, pH 8.0). After loading of the samples, the gel was run at room temperature in 0.5× TBE buffer at 350 V for 15 to 20 min and then dried and exposed to X-ray film.

Cell culture.

The human alveolar basal epithelial carcinoma cell line A549 was propagated in F-12K medium with 10% heat-inactivated fetal bovine serum (FBS). The human cervical carcinoma cell line ME180 was maintained in RPMI medium with 10% heat-inactivated FBS. VK2 cells, derived from human vaginal epithelial cells, were cultured in keratinocyte-SF medium with 0.1 ng/ml of epithelial growth factor, 0.05 mg/ml of bovine pituitary extract, and 44.1 mg/liter of calcium chloride. Each cell culture was incubated at 37°C in 5% CO2. The culture medium was changed every 2 to 3 days.

Adherence assays.

Cells were seeded into 24-well tissue culture plates at a density of 5 × 105 to 8 × 105 cells per well and cultured at 37°C in 5% CO2 for 3 to 5 days to establish a monolayer. GBS cells were grown in liquid culture to an A600 of 0.35 in THB and then washed once with PBS. Washed GBS cells were resuspended in infection medium: RPMI medium with 2% heat-inactivated FBS for the adhesion assay with ME180 and A549 cells or keratinocyte-SF medium with 0.4 mM CaCl2 for the assay with VK2 cells. The cell monolayer was infected with 5 × 106 to 8 × 106 GBS cells per well (multiplicity of infection [MOI] = 10) in infection medium for 1 h at 37°C in 5% CO2. The monolayer was washed five times with PBS, detached with 0.2 ml of 0.25% trypsin-EDTA for 10 min, and then lysed with 0.8 ml of 0.025% Triton X-100. The lysate was vigorously pipetted to liberate intracellular bacteria. The serially diluted lysate was cultured on blood agar to enumerate viable bacteria. Assays were repeated at least three times in triplicate. The percentage of adherent GBS cells was calculated as follows: % adherent cells = (number of CFU of adherent GBS/number of CFU in initial inoculum) × 100%.

Isolation of human peripheral blood monocytes and preparation of monocyte-derived macrophages.

Peripheral blood mononuclear cells (PBMCs) from healthy donors were isolated by Ficoll-Paque gradient centrifugation. Monocytes were purified from PBMCs by depletion of nonmonocytes (negative selection) by use of a monocyte isolation kit II according to the protocol provided by the manufacturer (Miltenyi Biotec). Monocytes (106 cells per well) were seeded in 12-well plates containing Dulbecco's modified Eagle medium (DMEM; Gibco) with 10% FBS. Recombinant human macrophage colony-stimulating factor (M-CSF) (150 ng/ml; R&D Systems, Inc.) was added to the culture for about 10 days to promote differentiation of monocytes to macrophages.

Macrophage internalization and intracellular survival assay.

Macrophages were washed 3 times with PBS to remove nonadherent cells. Bacteria were grown to mid-exponential phase (A600 = 0.35) in THB, washed once with PBS, and resuspended in DMEM containing 2% FBS. GBS cells were added to 2.5 × 105 macrophages per well at an MOI of 5. Following 1 h of incubation, monolayers were washed with PBS three times, and tissue culture medium containing penicillin (5 μg/ml) and gentamicin (100 μg/ml) was added and incubated for 1 h to kill extracellular bacteria. Cells were then detached and lysed with 0.25% trypsin and 0.025% Triton X-100, and the lysate was cultured on blood agar for enumeration of viable intracellular bacteria. The internalization rate was calculated as the ratio of CFU recovered immediately after antibiotic treatment (T0, representing internalized bacteria) to CFU in the initial inoculum. The intracellular survival rate was calculated as the ratio of CFU recovered from replicate wells 3 h after the addition of antibiotics to CFU recovered at T0. Assays were performed in triplicate and were repeated at least 3 times.

Opsonophagocytic killing assay.

Peripheral blood leukocytes (PBLs) were isolated from healthy human volunteers by dextran sedimentation. GBS cells were grown in THB to mid-exponential phase (A600 = 0.35) and then washed once with PBS. A total of 3 × 106 GBS cells were mixed with an equal number of PBLs (MOI = 1) in Eagle's minimum essential medium (EMEM) supplemented with 10% human serum preabsorbed with GBS cells. Control assays were performed using the same but heat-inactivated (56°C, 30 min) human serum. The reaction mixtures were incubated for 1 h at 37°C with end-over-end rotation, and then a 100-μl aliquot was removed from each assay mix and cells were lysed with 0.025% Triton X-100. Lysates were serially diluted and cultured on blood agar to determine the number of recovered bacteria. The results are expressed as survival indexes, calculated as the CFU recovered from the assay mixture/CFU recovered from the control.

Statistical analysis.

Data are presented as means ± standard deviations (SD) unless otherwise stated. Statistical analysis was performed using Prism 5.0 (Graphpad Software Inc.). A two-tailed t test was used to analyze the differences between groups. Statistical significance was defined for differences with P values of <0.05. Asterisks in the figures represent significant differences between groups (NS, P > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).


Pilus expression in GBS strain 2603.

Analysis of the genome sequence of GBS strain 2603 revealed two gene clusters (nucleotide positions 631950 to 639103 and 1419717 to 1411965) that represent PI-1 and PI-2a, respectively (Fig. 1) (31). Upstream of PI-2a is a truncated form of rogB, which in strain 2603 contains a frameshift mutation that is predicted to result in premature termination of translation after amino acid 326 (3). As described above, the intact rogB gene encodes a RofA family protein that activates PI-2a pilus expression in GBS strain NEM316 (3, 4). We evaluated the expression of the PI-2a pilus by whole-cell immunoblotting with antiserum directed against the PI-2a accessory protein (AP1-2a or GBS67) encoded by sag1408 (17). Our results showed that AP1-2a was undetectable in GBS 2603, although a strong signal was detected for a control strain, 515 (see Fig. S2 in the supplemental material). Poor expression of PI-2a in strain 2603 is likely due, at least in part, to one or both of the frameshift mutations that abrogate expression of the two regulatory proteins, RogB and Rga, that normally activate PI-2a gene transcription (3, 4, 29). These results indicated that pilus 2a is expressed poorly or not at all in strain 2603, so pilus 1 is the only potentially functional pilus in this strain (see below).

Expression of PI-1 pilus is repressed by CsrR in GBS 2603.

The CsrRS (or CovRS) two-component regulatory system controls the expression of multiple virulence factors in GBS (8, 9, 15). To test whether pilus expression is modulated by the CsrRS system, we compared pilus production in wild-type 2603 to that in an isogenic csrR deletion mutant, 2603 ΔcsrR, by whole-cell immunoblot analysis using a specific antiserum directed against BP-1, the pilus backbone protein (GBS80) encoded by sag0645 (16, 17). The result demonstrated a marked increase of BP-1 expression in strain 2603 ΔcsrR relative to that in wild-type strain 2603 (Fig. 2A). To confirm that the alteration was due to deletion of csrR and not to an unknown second mutation, we also performed the same assay using a repaired 2603 ΔcsrR strain in which the mutant csrR allele was replaced by a copy of the wild-type gene (8). The repaired strain restored BP-1 expression to a level similar to that of the original wild type (Fig. 2A). This result provided evidence that the observed increase in BP-1 expression was a result of csrR inactivation.

Fig 2
Increased expression of PI-1 pilus in mutant strains 2603 ΔcsrR and 2603 ΔcsrS. (A) Whole-cell immunoblot. Suspensions of GBS cells were spotted on a nitrocellulose membrane and probed with antiserum directed against BP-1 (GBS80). Strains ...

Pilus expression was also characterized by Western blotting of cell wall protein extracts. Proteins released from GBS cells by digestion of the cell wall were separated by SDS-PAGE, transferred to a nitrocellulose membrane, and detected with antiserum raised against recombinant BP-1. For wild-type 2603, the antiserum recognized a single band corresponding to the predicted molecular weight of monomeric BP-1 (Fig. 2B). Inactivation of csrR, however, resulted in an increase in the amount of BP-1 and, more dramatically, in the formation of a ladder-like pattern of higher-molecular-weight forms of the protein. The apparent increase in polymer formation in 2603 ΔcsrR is consistent with increased expression of the sortase proteins encoded by sag0647 and sag0648 as well as of the BP-1 subunit, all of which are essential for the production of BP-1 polymers (28). Again, we observed that the strain with repaired CsrR restored BP-1 expression to a monomeric pattern similar to that of the wild type (Fig. 2B). Thus, immunoblots of both whole cells and separated cell wall proteins indicate that CsrR represses the expression of the PI-1 pilus.

We also tested whether the observed changes in BP-1 expression in 2603 ΔcsrR reflected an altered transcription of sag0645, which encodes BP-1. We quantified the relative expression of sag0645 by quantitative RT-PCR (qRT-PCR) using RNA samples from wild-type 2603 and 2603 ΔcsrR. The results demonstrated that inactivation of csrR was associated with a marked increase in transcription of sag0645 (Fig. 2C). In addition, the transcription of sag0647, which encodes a sortase, was similarly increased in 2603 ΔcsrR relative to wild-type 2603 (Fig. 2C), consistent with the predicted transcriptional linkage of the PI-1 genes as an operon. These results suggest that increased production of pilus in the csrR mutant is attributable to increased gene transcription of the PI-1 pilus operon.

Finally, we investigated whether CsrS, the sensor component of the CsrRS two-component regulatory system, is also involved in the regulation of pilus gene expression. Relative transcription levels of sag0645 and sag0647 were compared in the wild type and a csrS deletion mutant, 2603 ΔcsrS. Expression of both genes was increased in 2603 ΔcsrS, although less so than in 2603 ΔcsrR (Fig. 2C). Together, these data indicate that CsrRS represses PI-1 pilus expression.

The AraC family protein Ape1 activates expression of the PI-1 pilus.

The open reading frame sag0644 is located immediately upstream of the PI-1 pilus operon and is transcribed in the opposite orientation (Fig. 1). The Sag0644 sequence includes an AraC-like helix-turn-helix motif in the C-terminal region that suggests it may function as a transcriptional regulator. Sag0644 has 72% identity in predicted amino acid sequence to MsmR, a positive regulator of the AraC/XylS family located in an analogous location in the Streptococcus pyogenes FCT or pilus region (22). Thus, it seems likely that sag0644 encodes an AraC-type regulator based on the genetic arrangement, sequence similarity, and presence of a characteristic DNA-binding motif. On the basis of these observations and data presented below, we have named the sag0644 gene ape1, for activator of pilus expression 1.

To determine whether ape1 is also regulated by the CsrRS system, we compared expression of ape1 in wild-type 2603 and 2603 ΔcsrR. As shown in Fig. 2C, inactivation of the csrR gene resulted in markedly increased expression of ape1 (sag0644); deletion of csrS led to a similar but less extreme change (Fig. 2C). The results indicate that CsrRS regulates ape1 gene expression in the same manner as it does the divergently transcribed PI-1 pilus operon in GBS 2603.

To test if Ape1 plays a role in the regulation of pilus expression, the ape1 gene was inactivated by allelic exchange. Expression of the pilus genes sag0645 and sag0647 was analyzed by qRT-PCR using RNA samples from wild-type 2603 and the 2603 Δape1 mutant, which were grown at pH 7.4. Under these conditions, there was no significant difference in sag0645 gene expression between the two strains and a decrease of approximately 25% in expression of sag0647 (Fig. 3A). Since the ape1 gene is tightly repressed by CsrR in strain 2603 (Fig. 2C), we reasoned that the relatively minor effect of Ape1 on regulation of the pilus genes might reflect the low level of Ape1 expression. If that were the case, we would expect to observe an increased regulatory effect of Ape1 on pilus expression in the 2603 ΔcsrR mutant background, in which ape1 expression is derepressed. To test that possibility, we inactivated the ape1 gene in the 2603 ΔcsrR mutant strain and compared pilus gene expression in the double mutant with that of 2603 ΔcsrR. Inactivation of ape1 expression reduced expression of the PI-1 genes sag0645 and sag0647 by approximately 50% in the 2603 ΔcsrR strain background (Fig. 3C). These data suggest that Ape1 expressed at a higher level in the CsrR mutant can activate expression of the PI-1 genes. The results also imply the presence of a regulatory cascade in which CsrRS controls expression of Ape1, which in turn regulates PI-1 gene expression.

Fig 3
Ape1 regulator activates the expression of PI-1 pilus genes. (A to D) Quantitative RT-PCR analysis of the effect of ape1 inactivation. The expression of pilus genes was compared between wild-type 2603 and the 2603 Δape1ape1) strain ...

In order to further evaluate the role of Ape1 in pilus gene regulation, we overexpressed ape1 from plasmid pNZ-ape1 in strain 2603 Δape1. The 12-fold increase in ape1 expression (compared to that in wild-type 2603) (see Fig. S3 in the supplemental material) was associated with a roughly 2-fold increase in PI-1 gene transcription (Fig. 3G and H). This result confirms the role of Ape1 as a transcriptional activator of the PI-1 operon. However, the extent of upregulation did not reach that observed in the setting of CsrR deletion (Fig. 2C), so it appears that the two regulators play nonredundant roles in controlling pilus gene expression.

To test if the regulatory function of Ape1 is affected by environmental factors such as pH, we studied the regulation of pilus expression by Ape1 when GBS strains were exposed to different pH environments. A greater activation effect of Ape1 was observed in GBS 2603 grown at pH 5.0 (Fig. 3B) than in that grown at pH 7.4. The activation effect of Ape1 was even more pronounced in the 2603 ΔcsrR background when the strain was grown at pH 5.0 (Fig. 3D). The latter result was also supported by similar findings in whole-cell immunoblot assays of BP-1 protein expression (Fig. 3E and F). An earlier study found that exposure of strain 2603 to pH 5.5 resulted in a 2- to 3-fold increase in ape1 expression as assessed in transcriptional profiling experiments using genomic microarrays (30). Increased expression of Ape1 could account for the observed upregulation of the PI-1 genes sag0645 and sag0647 at pH 5.0. To reexamine this question, we compared the transcriptional expression of ape1 at pH 5.0 and pH 7.4 by qRT-PCR. In contrast to the previous study, we found that exposure to pH 5.0 resulted in a decrease in expression of ape1 of approximately 80% compared to exposure to pH 7.4 for wild-type 2603, with no change in ape1 expression in 2603 ΔcsrR (see Fig. S4 in the supplemental material). The reason for the discrepancy with the earlier study is not clear but may reflect differences in the experimental growth conditions used in the two studies. Alternative explanations for the stronger activity of Ape1 as a transcriptional activator at acidic pH could be that the translation or stability of Ape1 is greater or the binding affinity of Ape1 for the target promoter is stronger at low pH.

CsrRS and Ape1 regulate expression of PI-1 pilus independently.

The results presented above indicate that both CsrR and Ape1 regulate the expression of PI-1 genes (Fig. 2 and and3).3). Since Ape1 expression is regulated by CsrR (Fig. 2C), it was possible that CsrRS regulation of PI-1 gene expression was mediated entirely through CsrRS regulation of Ape1, which in turn would regulate PI-1 gene expression. If that were the case, CsrR regulation should be abolished in the absence of Ape1. We therefore compared PI-1 gene expression in 2603 ΔcsrR Δape1 with that in 2603 Δape1. Contrary to the hypothesis proposed above, increased expression of the pilus genes was observed in 2603 ΔcsrR Δape1 (Fig. 4), indicating that CsrR retains its function as a repressor of pilus expression even in the absence of Ape1. These data imply that the negative regulation of PI-1 gene expression by CsrRS is not mediated exclusively through Ape1 but rather that CsrR also represses PI-1 gene transcription directly.

Fig 4
CsrR regulation of pilus genes in the absence of Ape1. The expression of PI-1 pilus genes in strain 2603 ΔcsrR Δape1csrRΔape1) was compared to that in 2603 Δape1ape1) by RT-PCR with bacteria grown ...

Ape1 and CsrR both bind directly to the promoter region of the PI-1 pilus operon.

We performed EMSAs to further characterize the interaction of CsrR and Ape1 with the promoter region of the PI-1 pilus operon. We expressed CsrR as a His6 fusion in E. coli and purified the recombinant protein by Ni2+-affinity chromatography. His6-CsrR was used in an EMSA with a DNA probe from strain 2603 that corresponded to the promoter region of the PI-1 pilus operon. A band shift indicating formation of a higher-molecular-weight complex was observed in the presence of His6-CsrR, indicating direct binding of CsrR to DNA sequences upstream of the PI-1 operon (Fig. 5A). Specificity of the binding interaction was supported by competition with excess unlabeled probe but not with an excess of unlabeled cps promoter sequence that served as a negative control. These results indicate that CsrR binds directly to the promoter region of the PI-1 pilus operon.

Fig 5
Binding of purified recombinant His6-CsrR or Ape1-His6 to the promoter region of the PI-1 pilus operon. EMSAs were performed using 32P-labeled DNA fragments corresponding to the promoter region of the PI-1 pilus operon and purified His6-CsrR (A) or Ape1-His ...

Investigation of recombinant Ape1 protein was hampered by poor solubility and instability of His6-Ape1 expressed from the pQE30 vector in E. coli. To overcome these problems, we utilized an inducible expression system, the nisin-controlled expression system in Lactococcus lactis (1, 13). The ape1 gene was amplified and cloned as an ape1-His6 fusion in the plasmid vector pNZ8048 under the control of a nisin-inducible promoter (12). The expression of Ape1-His6 was induced with nisin at a low temperature (16°C) to enhance the solubility and stability of the protein. The recombinant protein was purified following the same protocol as that used for the His6-CsrR protein. A distinct band shift to a higher molecular weight was observed in EMSAs in the presence of Ape1-His6, indicating direct binding of Ape1 to DNA sequences upstream of the PI-1 operon (Fig. 5B). The specificity of the binding interaction was demonstrated by competition assays, as shown in Fig. 5B. These results indicate that Ape1 and CsrR both bind directly to the promoter region of the PI-1 pilus operon; they provide a molecular basis for the observed independent regulation of PI-1 gene transcription by the two regulator proteins.

pH regulates PI-1 pilus gene expression.

GBS encounters various pH environments during colonization and infection of the human host, from the acidic milieu of the vagina to the near-neutral pH in the neonatal lung or bloodstream. It is possible that environmental pH modulates pilus expression and therefore affects colonization and infection. We tested the effect on pilus expression of GBS exposure to either pH 7.4 (to represent the bloodstream or neonatal lung) or pH 5.0 (to mimic the vaginal milieu). After 30 min of exposure, the amounts of transcript from PI-1 pilus genes were compared by qRT-PCR. The data showed a striking increase in expression of sag0645 and sag0647 at pH 5.0 compared with that at pH 7.4 in GBS 2603 (Fig. 6A). Similar results were observed in GBS mutant strains 2603 ΔcsrR, 2603 Δape1, and 2603 ΔcsrR Δape1. Whole-cell immunoblotting also demonstrated the increased expression of pilus at pH 5.0 in 2603 ΔcsrR (Fig. 6B). Together, these results suggest that an acidic environmental pH activates the expression of PI-1 pilus genes and that this effect is at least partly independent of the regulatory effects of CsrR and Ape1.

Fig 6
pH effect on PI-1 pilus expression. (A) The expression of PI-1 pilus genes of each GBS strain was compared by qRT-PCR after exposure to different pHs. Values represent gene expression relative to that of strain 2603 at pH 7.4. Strains tested were 2603, ...

PI-1 pilus does not contribute to GBS adherence to epithelial cells.

As shown above, the expression of the PI-1 pilus is repressed by CsrR, and only a very limited amount of pilus protein was detected on the GBS surface by immunoblotting. Since pili in multiple other species and GBS have been shown to mediate bacterial attachment to host cells, we tested whether the PI-1 pilus is involved in GBS adherence to human epithelial cells under either repressed or derepressed control of CsrR. To test the role in adherence of the PI-1 pilus, we first generated a nonpolar in-frame deletion mutant of GBS 2603 that lacked accessory protein 1 (AP1-1 or GBS104), encoded by sag0649, homologs of which have been found to represent the pilus adhesin in GBS strains 2603 and COH1 (11, 26). We then examined the relative association of GBS wild-type and mutant strains with three types of human epithelial cells, namely, ME180 (cervical carcinoma cell line), A549 (adenocarcinoma alveolar basal epithelial cell line), and VK2 (vaginal epithelial cell line) cells. However, the adherence of the AP1-1 mutant strain 2603 ΔAP1-1 was not significantly different from that of wild-type 2603 (Fig. 7A). One possibility is that AP1-1 is not the only component of the pilus that contributes to adherence of GBS 2603 to human epithelial cells (11), so deletion of this single gene did not impede GBS adherence. To evaluate the potential role of other PI-1 pilus proteins in adherence, we constructed a nonpolar in-frame deletion mutant of GBS 2603 that lacked the entire PI-1 pilus operon (sag0645 to -0649). Expression of the downstream genes sag0650 and sag0651 in the PI-1 mutant 2603 ΔPI-1 was determined by qRT-PCR and was found to be unchanged from that in wild-type 2603 (see Fig. S5 in the supplemental material). We found that deletion of the entire pilus operon did not affect GBS adherence to any of the three types of epithelial cells tested (Fig. 7B), in agreement with the results for deletion of the single gene (sag0649) encoding the presumed pilus adhesin, AP1-1.

Fig 7
Adherence of GBS strains to human epithelial cells. A549, ME180, or VK2 cells were infected with GBS strains at an MOI of 10 for 1 h. (A) Effect of AP1-1 inactivation on GBS adherence. Adherence rates were compared among strains 2603, 2603 ΔAP1-1 ...

Since the adherence of wild-type strain 2603 to epithelial cells is relatively weak, especially with ME180 and A549 cells, it might be difficult to observe a further reduction when the pilus genes are inactivated. Furthermore, the contribution of pili to adhesion might be more apparent if the pili were more highly expressed. To address these considerations, we also tested the adherence of strain 2603 ΔcsrR, in which PI-1 pilus production is derepressed. We found that GBS adherence to host cells was greatly increased in the 2603 ΔcsrR mutant (Fig. 7), in agreement with previous results for strain NEM316 (15). To test whether the increased adherence of 2603 ΔcsrR was due to increased pilus expression, we constructed a double mutant, 2603 ΔcsrR ΔPI-1, and tested its adherence. Figure 7 shows that there was no significant difference in the adherence of 2603 ΔcsrR ΔPI-1 and 2603 ΔcsrR, indicating that the increased adherence of 2603 ΔcsrR was not due to derepressed PI-1 pilus expression. To further investigate the potential role of PI-1 pili in adherence, we expressed the adhesion protein AP1-1 on the surface of Lactococcus lactis cells. Despite a high level of expression, we observed only a modest increase in adherence to VK2 cells (about a 1.4-fold increase) and no change in adherence to ME180 cells (see Fig. S6 in the supplemental material). Using a similar approach, Maisey et al. found that expression of the PI-2a adhesin protein (PilA or AP1-2a) conferred a marked increased in adherence of L. lactis to brain microvascular endothelial cells, highlighting the differences in function of PI-1 pili compared to PI-2a pili (18). Therefore, it appears that the lack of PI-1 contribution to adherence of strain 2603 reflects not only a low level of expression but also relatively weak intrinsic binding of the AP1-1 protein to host cells. Taken together, these data indicate that the PI-1 pilus does not contribute to GBS 2603 adherence to the three types of epithelial cells tested.

PI-1 pilus contributes to GBS resistance to phagocytic killing by macrophages but not neutrophils.

Conflicting results have been reported on the function of GBS pili in resistance to phagocytic killing (19, 23). To evaluate whether the PI-1 pilus plays a role in phagocytic resistance, we first assessed GBS internalization and intracellular survival in human monocyte-derived macrophages. During a 1-h incubation of GBS with macrophages, we observed 2-fold more internalization of 2603 ΔPI-1 than of wild-type 2603 (Fig. 8A). Furthermore, inactivation of csrR (strain 2603 ΔcsrR) resulted in a significant increase in GBS phagocytosis. Further comparison showed a 50% increase in the internalization of 2603 ΔcsrR ΔPI relative to that of 2603 ΔcsrR. In studies of intracellular survival, 2603 ΔPI showed a 50% decrease in survival rate compared to the wild type, whereas 2603 ΔcsrR showed a 1.8-fold increase in intracellular survival compared to the wild type (Fig. 8B). The double mutant 2603 ΔcsrR ΔPI-1, however, exhibited a similar survival rate in human macrophages to that of 2603 ΔcsrR. Thus, our data suggest that the PI-1 pilus of GBS 2603 enhances resistance to macrophage phagocytosis by inhibiting both GBS uptake and intracellular killing.

Fig 8
Internalization and intracellular survival of GBS in human monocyte-derived macrophages. Human macrophages were infected with GBS 2603, 2603 ΔPI-1 (ΔPI-1), 2603 ΔcsrRcsrR), or 2603 ΔcsrR ΔPI-1 (Δ ...

To further test the role of the PI-1 pilus in resistance to phagocytosis, opsonophagocytic killing assays were performed using the same four GBS strains and human peripheral blood leukocytes consisting primarily of neutrophils. Each strain was incubated with peripheral blood leukocytes in the presence of 10% human serum as a complement source. Negative-control assays were performed using heat-inactivated serum. The survival index was calculated as the ratio of the number of GBS cells recovered in the assay to the number of GBS cells recovered from the negative control. As shown in Fig. 9, the survival indexes of all four GBS strains were similar, suggesting that the PI-1 pilus does not promote resistance to complement-mediated opsonophagocytosis. Taken together, our results indicate that the PI-1 pilus of GBS 2603 contributes to resistance to phagocytic killing by macrophages but not neutrophils. Our results do not exclude the possibility that PI-1 pili play a larger role in opsonophagocytic resistance for strains in which they are more highly expressed.

Fig 9
Opsonophagocytic killing of GBS by human peripheral blood leukocytes. GBS cells were mixed with an equal number of peripheral blood leukocytes in the presence of 10% human serum for 1 h. Strains tested were GBS 2603, 2603 ΔPI-1 (ΔPI-1), ...


The present investigation characterizes the regulation and functional importance of the PI-1 pilus in GBS. The PI-1 pilus is less well studied than the PI-2a or PI-2b pilus, in part because while 72% of GBS isolates have PI-1, all GBS isolates also have one of the PI-2 loci (20). We took advantage of strain 2603, which has inactivating mutations in the transcriptional activators RogB and Rga that regulate PI-2a pilus gene expression. Since PI-2a pili are not expressed at a detectable level in GBS 2603, this strain represents a useful platform for characterizing the regulation and function of the PI-1 pilus without confounding effects of a second pilus type.

We found that PI-1 pilus gene expression was regulated by at least three factors, namely, the CsrRS two-component system; Ape1, an AraC-like transcriptional activator located upstream of the PI-1 operon; and environmental pH. Deletion mutants of csrR and, to a lesser extent, csrS exhibited increased transcript levels of the PI-1 genes and a corresponding increase in the immunoreactive PI-1 pilus backbone protein, indicating that CsrRS represses PI-1 expression. We observed a similar effect of CsrR and CsrS on expression of the divergently transcribed ape1 gene, i.e., upregulation of ape1 in csrR and csrS mutants. We also investigated the potential role of Ape1 in regulating pilus expression. Deletion of ape1 resulted in little or no change in PI-1 gene expression in strain 2603 but was associated with a 2-fold reduction in PI-1 gene expression in a csrR deletion mutant background. Since ape1 is upregulated in the csrR mutant, these results are consistent with Ape1 acting as an activator of PI-1 gene expression, an effect that becomes apparent with the higher expression of ape1 in the absence of CsrR. This model was further supported by the finding that overexpression of ape1 in trans was associated with increased PI-1 gene expression. However, CsrR regulation of PI-1 gene expression is not mediated exclusively through regulation of ape1 by CsrR, since CsrR regulation of the pilus genes was retained in an ape1 mutant. CsrR and Ape1 were shown to bind independently to the putative promoter region upstream of PI-1 in EMSAs. Thus, CsrRS and Ape1 have both independent and interconnected roles in regulating PI-1 expression.

We found that acidic pH is a third factor that activates PI-1 pilus gene expression, independent of CsrRS and Ape1. Environmental pH has been shown to regulate gene expression in many bacteria, and in most cases the mechanism is unknown. Since intracellular pH is controlled within narrow limits, it is presumed that responses to mild acid or alkali stress are mediated through a signaling mechanism transduced at the cell surface (24, 32). We speculate that such a mechanism operates in the Ape1- and CsrR-independent transcriptional regulation observed in our studies of pilus gene expression. That is, another two-component system or standalone transcriptional regulator may influence pilus gene expression in response to changes in pH.

While CsrRS is an important regulator of PI-1 pilus expression, it appears to have much less importance for the PI-2a pilus. Samen et al. found that inactivation of CsrRS did not change PI-2a expression in GBS strain NEM316 (29). We found a modest increase (about 2-fold) in PI-2a gene expression in 515 ΔcsrR, a csrR deletion mutant in the type Ia GBS strain 515 (unpublished data). Two RofA family proteins, RogB and Rga, were reported to be activators of PI-2a pilus expression in NEM316 (3, 4, 29). However, frameshift mutations are found in both rogB and rga in GBS 2603, consistent with the poor expression of the PI-2a pilus.

GBS pili have been shown to promote adherence to human epithelial cells in vitro. In the type V GBS strain 10/84 background, a mutant defective for expression of PilA (AP1-2a) was found to be 60% less adherent to human brain microvascular endothelial cells (18). PilA (AP1-2a) was also shown to contribute to adherence of the type III strain NEM316 to A549 respiratory epithelial cells (10). One study found that deletion of AP1-1 in type III strain COH1 reduced adherence to ME180 cells by 40% (26). In general, these studies have identified AP1-1 as the major pilus adhesin, although the backbone protein (BP-2a or PilB) was shown to affect the adherence of NEM316 to A549 cells under flow conditions, perhaps by enhancing display of the adhesin protein on the bacterial surface (10). In contrast to our results, an earlier study found that deletion of the AP1-1 adhesin protein (Sag0649 or GBS104) in GBS 2603 reduced adherence to A549 cells by approximately 70% (11). Our data showed that knocking out either AP1-1 or the entire pilus operon (sag0645 to -0649) did not affect the adherence of GBS 2603 to A549 or the other two cell lines studied. The reason for these conflicting results is not clear; possibilities include differences in methods used to assess bacterial adherence or the presence of an unknown second mutation that affected adherence of the AP1-1 deletion strain in the earlier study. We also tested the effect on adherence of expressing AP1-1 on the surface of L. lactis. Although AP1-1 was highly expressed, it conferred only a modest increase in adherence of L. lactis to VK2 cells and had no effect on adherence to ME180 cells.

On the other hand, we observed a marked increase in adherence in the csrR deletion mutant. The increased adherence, however, was not correlated with enhanced expression of the pilus. On the contrary, we consistently observed a slightly higher adherence of 2603 ΔcsrR ΔPI-1 to epithelial cells than that of 2603 ΔcsrR (especially to ME180 and VK2 cells). It is likely that CsrR regulates GBS adherence through other surface factors and that the elimination of pili from the ΔcsrR mutant enhanced exposure of these factors on the bacterial cell surface, thereby promoting GBS adherence.

Conflicting results have been reported on the role of pili in GBS resistance to phagocytosis. Maisey et al. found that the PI-2a pilus backbone protein BP-2a (PilB) promoted phagocyte resistance and systemic virulence in type V GBS strain 10/84 (19), whereas Papasergi et al. reported that the PI-2a pilus did not enhance GBS NEM316 survival in murine macrophages (23). To further investigate the possible role of pilus in GBS virulence, we tested GBS resistance to phagocytic killing by human macrophages and neutrophils by using wild-type 2603 and pilus mutants. Our data suggest that the PI-1 pilus enhances GBS resistance to phagocytic uptake by macrophages and inhibits intracellular killing. We did not detect a significant effect on complement-mediated opsonophagocytic killing by neutrophils.

In summary, our results identify the CsrRS two-component system, the AraC-like regulator Ape1, and environmental pH as major factors that regulate GBS PI-1 pilus expression. In strain 2603, PI-1 pili contribute to resistance to macrophage phagocytosis, but they do not appear to mediate GBS adherence to epithelial cells. These observations provide new insight into the regulation of GBS pilus expression and, in the context of earlier studies, highlight the complexity of how pili modulate interactions of GBS with the human host.

Supplementary Material

Supplemental material:


This work was supported in part by Public Health Service grants R01 AI59502 (M.R.W.) and U01 AI-060603 (L.C.P.), both from the National Institute of Allergy and Infectious Disease.

We thank Samantha J. Mascuch for technical assistance, Maghnus O'Seaghdha for helpful advice, Immaculada Margarit y Ros (Novartis Vaccines, Siena, Italy) and Hung Ton-That (University of Texas, Houston, TX) for providing antisera to recombinant GBS proteins, and Cormac Gahan (University College Cork) for providing L. lactis strains and plasmids.


Published ahead of print 9 March 2012

Supplemental material for this article may be found at


1. Bahey-El-Din M, Griffin BT, Gahan CG. 2008. Nisin inducible production of listeriolysin O in Lactococcus lactis NZ9000. Microb. Cell Fact. 7:24. [PMC free article] [PubMed]
2. Barocchi MA, et al. 2006. A pneumococcal pilus influences virulence and host inflammatory responses. Proc. Natl. Acad. Sci. U. S. A. 103:2857–2862. [PubMed]
3. Dramsi S, et al. 2006. Assembly and role of pili in group B streptococci. Mol. Microbiol. 60:1401–1413. [PubMed]
4. Gutekunst H, Eikmanns BJ, Reinscheid DJ. 2003. Analysis of RogB-controlled virulence mechanisms and gene repression in Streptococcus agalactiae. Infect. Immun. 71:5056–5064. [PMC free article] [PubMed]
5. Ho SN, Hunt HD, Horton RM, Pullen JK, Pease LR. 1989. Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77:51–59. [PubMed]
6. Holo H, Nes IF. 1989. High-frequency transformation, by electroporation, of Lactococcus lactis subsp. cremoris grown with glycine in osmotically stabilized media. Appl. Environ. Microbiol. 55:3119–3123. [PMC free article] [PubMed]
7. Horton RM, Hunt HD, Ho SN, Pullen JK, Pease LR. 1989. Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77:61–68. [PubMed]
8. Jiang SM, Cieslewicz MJ, Kasper DL, Wessels MR. 2005. Regulation of virulence by a two-component system in group B streptococcus. J. Bacteriol. 187:1105–1113. [PMC free article] [PubMed]
9. Jiang SM, et al. 2008. Variation in the group B Streptococcus CsrRS regulon and effects on pathogenicity. J. Bacteriol. 190:1956–1965. [PMC free article] [PubMed]
10. Konto-Ghiorghi Y, et al. 2009. Dual role for pilus in adherence to epithelial cells and biofilm formation in Streptococcus agalactiae. PLoS Pathog. 5:e1000422. [PMC free article] [PubMed]
11. Krishnan V, et al. 2007. An IgG-like domain in the minor pilin GBS52 of Streptococcus agalactiae mediates lung epithelial cell adhesion. Structure 15:893–903. [PMC free article] [PubMed]
12. Kuipers OP, de Ruyter PGGA, Kleerebezem M, De Vos WM. 1998. Quorum sensing-controlled gene expression in lactic acid bacteria. J. Biotechnol. 64:15–21.
13. Kuipers OP, Rollema HS, Siezen RJ, De Vos WM. 1995. Lactococcal expression systems for protein engineering of nisin. Dev. Biol. Stand. 85:605–613. [PubMed]
14. Kunji ER, Slotboom DJ, Poolman B. 2003. Lactococcus lactis as host for overproduction of functional membrane proteins. Biochim. Biophys. Acta 1610:97–108. [PubMed]
15. Lamy MC, et al. 2004. CovS/CovR of group B streptococcus: a two-component global regulatory system involved in virulence. Mol. Microbiol. 54:1250–1268. [PubMed]
16. Lauer P, et al. 2005. Genome analysis reveals pili in group B Streptococcus. Science 309:105. [PubMed]
17. Maione D, et al. 2005. Identification of a universal group B streptococcus vaccine by multiple genome screen. Science 309:148–150. [PMC free article] [PubMed]
18. Maisey HC, Hensler M, Nizet V, Doran KS. 2007. Group B streptococcal pilus proteins contribute to adherence to and invasion of brain microvascular endothelial cells. J. Bacteriol. 189:1464–1467. [PMC free article] [PubMed]
19. Maisey HC, et al. 2008. A group B streptococcal pilus protein promotes phagocyte resistance and systemic virulence. FASEB J. 22:1715–1724. [PMC free article] [PubMed]
20. Margarit I, et al. 2009. Preventing bacterial infections with pilus-based vaccines: the group B streptococcus paradigm. J. Infect. Dis. 199:108–115. [PubMed]
21. Mora M, et al. 2005. Group A Streptococcus produce pilus-like structures containing protective antigens and Lancefield T antigens. Proc. Natl. Acad. Sci. U. S. A. 102:15641–15646. [PubMed]
22. Nakata M, Podbielski A, Kreikemeyer B. 2005. MsmR, a specific positive regulator of the Streptococcus pyogenes FCT pathogenicity region and cytolysin-mediated translocation system genes. Mol. Microbiol. 57:786–803. [PubMed]
23. Papasergi S, et al. 2011. The GBS PI-2a pilus is required for virulence in mice neonates. PLoS One 6:e18747. [PMC free article] [PubMed]
24. Perez JC, Groisman EA. 2007. Acid pH activation of the PmrA/PmrB two-component regulatory system of Salmonella enterica. Mol. Microbiol. 63:283–293. [PMC free article] [PubMed]
25. Perez-Casal J, Price JA, Maguin E, Scott JR. 1993. An M protein with a single C repeat prevents phagocytosis of Streptococcus pyogenes: use of a temperature-sensitive shuttle vector to deliver homologous sequences to the chromosome of S. pyogenes. Mol. Microbiol. 8:809–819. [PubMed]
26. Pezzicoli A, et al. 2008. Pilus backbone contributes to group B Streptococcus paracellular translocation through epithelial cells. J. Infect. Dis. 198:890–898. [PubMed]
27. Rinaudo CD, et al. 2010. Specific involvement of pilus type 2a in biofilm formation in group B Streptococcus. PLoS One 5:e9216. [PMC free article] [PubMed]
28. Rosini R, et al. 2006. Identification of novel genomic islands coding for antigenic pilus-like structures in Streptococcus agalactiae. Mol. Microbiol. 61:126–141. [PubMed]
29. Samen U, et al. 2011. Rga is a regulator of adherence and pili formation in Streptococcus agalactiae. Microbiology 157:2319–2327. [PubMed]
30. Santi I, et al. 2009. CsrRS regulates group B Streptococcus virulence gene expression in response to environmental pH: a new perspective on vaccine development. J. Bacteriol. 191:5387–5397. [PMC free article] [PubMed]
31. Tettelin H, et al. 2002. Complete genome sequence and comparative genomic analysis of an emerging human pathogen, serotype V Streptococcus agalactiae. Proc. Natl. Acad. Sci. U. S. A. 99:12391–12396. [PubMed]
32. Weinrick B, et al. 2004. Effect of mild acid on gene expression in Staphylococcus aureus. J. Bacteriol. 186:8407–8423. [PMC free article] [PubMed]
33. Yanagawa R, Honda E. 1976. Presence of pili in species of human and animal parasites and pathogens of the genus Corynebacterium. Infect. Immun. 13:1293–1295. [PMC free article] [PubMed]
34. Yanagawa R, Otsuki K, Tokui T. 1968. Electron microscopy of fine structure of Corynebacterium renale with special reference to pili. Jpn. J. Vet. Res. 16:31–37. [PubMed]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)