OUP user menu

Nasal immunization of mice with Lactobacillus casei expressing the pneumococcal surface protein C primes the immune system and decreases pneumococcal nasopharyngeal colonization in mice

Marília de Lúcia Hernani, Patrícia Cristina Duarte Ferreira, Daniela Mulari Ferreira, Eliane Namie Miyaji, Paulo Lee Ho, Maria Leonor Sarno Oliveira
DOI: http://dx.doi.org/10.1111/j.1574-695X.2011.00809.x 263-272 First published online: 1 August 2011

Abstract

Streptococcus pneumoniae colonizes the upper respiratory tract of healthy individuals, from where it can be transmitted to the community. Occasionally, bacteria invade sterile niches, causing diseases. The pneumococcal surface protein C (PspC) is a virulence factor that is important during colonization and the systemic phases of the diseases. Here, we have evaluated the effect of nasal or sublingual immunization of mice with Lactobacillus casei expressing PspC, as well as prime-boosting protocols using recombinant PspC, on nasopharyngeal pneumococcal colonization. None of the protocols tested was able to elicit significant levels of anti-PspC antibodies before challenge. However, a significant decrease in pneumococcal recovery from the nasopharynx was observed in animals immunized through the nasal route with L. casei-PspC. Immune responses evaluated after colonization challenge in this group of mice were characterized by an increase in mucosal anti-PspC immunoglobulin A (IgA) 5 days later, a time point in which the pneumococcal loads were already low. A negative correlation between the concentrations of anti-PspC IgA and pneumococcal recovery from the nasopharynx was observed, with animals with the lowest colonization levels having higher IgA concentrations. These results show that nasal immunization with L. casei-PspC primes the immune system of mice, prompting faster immune responses that result in a decrease in pneumococcal colonization.

Keywords
  • Lactobacillus casei
  • PspC
  • pneumococcal colonization

Introduction

Streptococcus pneumoniae is commonly present in the nasopharyngeal human tract. Colonization is a dynamic process that usually occurs early in life and bacteria can be cleared from the respiratory mucosa in a few weeks, without causing much harm (Weiser, 2010). However, under some conditions that depend on the age and immune status of the individual, the pneumococcus may invade and cause diseases such as otitis media, pneumonia, meningitis and sepsis. Around 1 million children die every year from pneumococcal diseases, mainly in developing countries (de Quadros, 2009; O'Brien et al., 2009). It is well accepted that colonization precedes invasive diseases and therefore prophylactic strategies for its prevention are interesting to diminish both transmission and diseases. Conjugated vaccines composed of pneumococcal polysaccharides and protein carriers (PCV7, PCV10 or PCV13, according to the number of polysaccharides included) have been adopted for vaccination of children in several countries. After some years of PCV7 use, the incidence of pneumococcal diseases caused by serotypes included in the vaccine has drastically reduced and, in addition, the transmission of these serotypes has also decreased (Dagan, 2009; Pilishvili et al., 2010). Nevertheless, a substitution of the prevalent disease-causing serotypes was observed, making necessary the adoption of new formulations such as PCV10 or PCV13, to adequate the vaccine to the new circulating strains (Bettinger et al., 2010; Hsu et al., 2010; Techasaensiri et al., 2010). Therefore, broad coverage of conjugated vaccines is attained by the inclusion of several polysaccharides in the formulation, a practice that increases the production costs. In this context, protein antigens emerge as possible alternatives for effective broad-coverage vaccines at lower costs. The pneumococcal surface protein C (PspC) is a virulence factor exposed on the pneumococcal surface that binds to different host components, modulating the interaction of the bacteria with the immune system. Two domains present in the N-terminal region of PspC have been shown to interact with factor H, an inhibitor of the complement system (Dave et al., 2001; Hammerschmidt et al., 2007). This interaction was shown to mediate the adhesion of the pneumococci to epithelial cells in vitro and to increase the invasion of the bacteria in mouse lungs (Hammerschmidt et al., 2007; Quin et al., 2007). In cooperation with pneumococcal surface protein A (PspA), PspC has been shown to inhibit bacterial immune adherence to host erythrocytes and transferring to macrophages, which would ultimately impair phagocytosis. This effect was related to the reduction of complement deposition on the bacterial surface in the presence of both virulence factors (Li et al., 2007). PspC also interacts with human secretory immunoglobulin A (sIgA) and this has been shown to increase pneumococcal adherence to epithelial cells (Zhang et al., 2000; Dave et al., 2004). Different studies described the protective effects of immunization with PspC, alone or in combination with other antigens, in animal models of colonization and bacteremia (Brooks-Walter et al., 1999; Ogunniyi et al., 2001; Balachandran et al., 2002; Cao et al., 2007; Ogunniyi et al., 2007). Our group has evaluated the protective efficacy of nasal vaccines composed of PspC or Lactobacillus casei expressing PspC (L. casei-PspC) against an invasive intranasal challenge with S. pneumoniae ATCC6303 (Ferreira et al., 2009). Although no protection was observed in this model, immunization with L. casei-PspC elicited an increase in the infiltration of neutrophils in lungs, 13 h after challenge. The secretion of interferon-γ (IFN-γ) and interleukin-17 (IL-17) by the lungs and spleen cells were also induced in mice immunized with L. casei-PspC at this time point. A question was raised about the low cross-reactivity between the PspC used as an immunogen and the one expressed by the ATCC6303, a serotype 3 strain, which is in fact a variant also named as Hic (H-binding inhibitor of complement), commonly expressed by serotype 3 strains (Iannelli et al., 2002). In the present work, we have evaluated the immunization of mice with L. casei-PspC by both nasal and sublingual mucosa, as well as a prime-boost protocol using recombinant PspC. The protective efficacy was analyzed using a mice model of pneumococcal colonization, using the 0603 strain (serotype 6B), which expresses a PspC with high identity at the N-terminal region (81% of identical amino acids in the region analyzed) to the one expressed in L. casei.

Materials and methods

Bacterial strains and growth conditions

Lactococcus lactis MG1363 was grown in M17 (Difco) supplemented with 0.5% glucose, L. casei CECT5275 was grown in Man, Rogosa and Sharpe medium (Difco) and S. pneumoniae 0603 (serotype 6B, kindly provided by Dr Richard Malley, Children's Hospital, Boston) was grown in Todd–Hewitt broth (Difco) supplemented with 0.5% yeast extract (THY). All bacteria were grown at 37 °C without shaking and 5 µg mL−1 of erythromycin was added to the growth media of recombinant L. lactis and L. casei. Pneumococcal strain 0603 was always plated in blood agar and grown overnight at 37 °C before inoculation in THY. All bacterial stocks were maintained at −80 °C in their respective media containing 20% glycerol.

Recombinant protein and plasmids

The construction of the pAE-PspC vector as well as the procedures for the expression and purification of the PspC fragment have been described previously (Ferreira et al., 2009). The fragment expressed comprises the N-terminal plus the proline-rich region of PspC from the S. pneumoniae strain 491/00 (serotype 6B from Instituto Adolfo Lutz, São Paulo, Brazil), GenBank accession number EF424119. The same PspC fragment was expressed in L. casei, as described below.

Cloning and recombinant procedures in lactic acid bacteria

The pT1NX vector (Campos et al., 2008) was used for the constitutive intracellular expression of PspC in L. casei (Ferreira et al., 2009). For the expression attached to the cell wall, the pspC DNA fragment was cloned into the pT1NXssAnch vector (Steidler et al., 1995) in a way to produce a fusion with the usp-45 signal sequence and the anchoring sequence from L. casei peptidase. The following primers were used for amplification: PspCwlacF: 5′-ACCGGTGATATCCCATGCGACAGAGAACGAG-3′, and PspCwlacR: 5′-GGATCCTTAAGATCTTTGTGGTTGTTCAGC-3′. After sequencing confirmation, the fragments were cloned by digestion with PinAI and BglII (restriction sites underlined), which produce compatible ends with the NgoMIV and the BamHI sites present in the vector. Ligation products were used to transform competent L. lactis as described previously (Oliveira et al., 2006). Plasmids isolated from L. lactis were then used for electroporation of L. casei (Oliveira et al., 2006). Lactococcus lactis and L. casei transformants were selected by plating on the respective media containing 1.8% agar and 5 µg mL−1 of erythromycin.

Analysis of PspC expression

For Western blot analysis, recombinant L. casei were grown until the late stationary phase (OD600 nm≥2). Bacteria were collected by centrifugation and suspended in 100 mM Tris-HCl, pH 8. Mechanical lysis was performed in a Beads-beater equipment (Biospec) with three cycles of 4600 r.p.m. for 30 s, using glass beads. Samples were transferred to a clean tube and centrifuged at 15 000 g for 5 min for the analysis of intracellular expression. The centrifugation step was excluded when analyzing the expression of PspC attached to the cell wall. Protein extracts were fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and electrotransferred to nitrocellulose membranes using the Mini Protean II equipment (Bio-Rad). Membranes were probed with mouse polyclonal anti-PspC antiserum and horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (Sigma). Detection was performed using the chemiluminescent ECL kit (GE Healthcare). The exposition of PspC on L. casei surface was analyzed by flow cytometry using FACSCalibur (Benton and Dickinson). Briefly, bacteria were grown until the late stationary phase (OD600 nm≥2); a volume corresponding to 109 CFU was centrifuged, washed in phosphate-buffered saline and incubated with polyclonal anti-PspC developed in mice, for 30 min, in ice. After washing, samples were incubated with anti-mouse IgG-fluorescein isothiocyanate (Sigma) for 30 min, in ice, washed twice and suspended in 2% formaldehyde for analysis. Flow cytometry was performed with 10 000 events recorded and the median of fluorescent bacteria was used to compare the strains.

Immunization of mice and detection of anti-PspC antibodies by enzyme-linked immunosorbent assay (ELISA)

Five- to seven-week-old female C57BL/6 mice (produced by the Central Animal facility from Instituto Butantan) were supplied with food and water ad libitum. Experimental protocols were approved previously by the Ethical Committee for Animal Research of Instituto Butantan, under the license number 587/09. Lactobacillus casei expressing PspC in the intracellular compartment (L.c.-PspC), PspC attached to the cell wall (L.c.-wPspC) or carrying the empty vector (L.c.) were grown until OD600 nm≥2, collected by centrifugation at 3300 g, for 5 min, washed with saline and then suspended according to the use. Groups of six animals were anesthetized through the intraperitoneal route with 200 µL of a 0.2% xilazine and 0.5% ketamine mixture and inoculated intranasally with six doses of 109L. casei CFU in 10 µL of saline, on days 0, 1, 14, 15, 28 and 29. For sublingual immunization, mice were subjected to deep anesthesia by an intraperitoneal inoculation of a 0.2% xilazine and 1.0% ketamine solution. Mice received 109L. casei CFU, in 7 µL of saline, on days 0, 14 and 28. For prime-boost protocols, mice were inoculated intranasally with recombinant L. casei on days 0, 1, 14 and 15, followed by a subcutaneous injection, on day 28, of 5 µg of recombinant purified PspC, previously treated with Triton X-114 to remove LPS (Aida & Pabst, 1990) and without the use of adjuvants. Mice were bled through the retrorbital plexus 20 days after the last immunization. Vaginal washes were collected from days 15 to 19 after the last immunization by gentle pipetting 25 µL of saline, twice, and samples were pooled for each animal. Antibody levels were evaluated in sera, vaginal washes and nasal washes (collected 21 days after the last immunization or at different periods after the pneumococcal challenge, as described below) by ELISA in plates coated with PspC (10 µg mL−1). The assay was performed using goat anti-mouse IgG, IgA and rabbit anti-goat conjugated with HRP (Southern Biotech). Antibody concentrations were determined using IgA and IgG concentration curves (Southern Biotech).

Intranasal challenge with S. pneumoniae 0603

Streptococcus pneumoniae 0603 (serotype 6B) was used for pneumococcal colonization challenge. DNA sequencing of the N-terminal region of the PspC expressed by this strain revealed that the protein shares 81% of identical amino acids with the PspC fragment expressed in L. casei (GenBank accession number JF340458). The 0603 (serotype 6B) strain was grown in THY medium until OD600 nm=0.4, aliquoted and kept frozen at −80 °C. A suspension containing 5 × 106 bacteria in 10 µL was gently inoculated into both nostrils of mice previously anesthetized through the intraperitoneal route with 200 µL of a 0.2% xilazine and 0.5% ketamine mixture, 21 days after the last immunization. Mice were sacrificed 24 h or 5 days after the intranasal pneumococcal challenge to collect nasal washes and the spleen. For nasal washes, a catheter was inserted into the trachea of the mice and the upper respiratory tract was rinsed with 200 µL of saline. An additional rinse with 200 µL of saline was performed in order to remove most of the bacteria attached to the nasopharynx. The volumes were pooled and serially diluted for plating in blood agar. CFU were counted after incubation at 37 °C for 24 h. The remaining volumes were maintained at −20°C until use.

Detection of antigen-specific cytokine secretion

Single-cell suspensions of spleens were obtained as described previously (Ferreira DM et al., 2008; Ferreira PCD et al., 2008). Viable cell counts were determined by trypan blue exclusion. For antigen stimulation, 5 × 106 cells mL−1 from individual animals were cultivated in the presence of 5 µg mL−1 PspC for 72 h. Supernatants were assayed using sandwich ELISA (Peprotech) for the presence of tumor necrosis factor-α (TNF-α), gamma interferon (IFN-γ) or Interleukin-17 (IL-17), and the results were subtracted from the cytokine levels obtained in nonstimulated cultures.

Statistical analysis

Differences in colonization rates or antibody concentrations were analyzed using a two-tailed Mann–Whitney U-test. Correlation analysis between IgA levels and pneumococcal recovery from nasal washes was performed using Spearman's correlation test. In all cases, P≤0.05 was considered significantly different.

Results

Expression of PspC by L. casei

Extracts from clones transformed with the recombinant plasmids, pT1NX-PspC (for intracellular expression) and pT1NXssAnch-PspC (for expression attached to the cell wall) were analyzed by a Western blot (Fig. 1a). The amount of protein loaded corresponds to 109 CFU, which is the dose inoculated in mice. A band of 50 kDa, corresponding to the PspC fragment, was observed in the extract from L. casei-PspC. Fusion with the anchoring peptide produces a band of 70 kDa, which could be observed in the extract from L. casei-wPspC. No reactivity was observed in the extracts from L. casei (Fig. 1a). Comparison with a concentration curve has shown that a dose of L. casei-PspC would contain approximately 120 ng of PspC, whereas a dose of L. casei-wPspC would contain approximately 30 ng of the protein. Correct localization of PspC was confirmed by flow cytometry (Fig. 1b). After incubation with anti-PspC antisera and the secondary fluorescent antibody, the curve observed for L. casei-wPspC presented a median value of 52, whereas L. casei-PspC and L. casei displayed very similar curves, with median values of 13.5 and 13.1, respectively. Similar reactivity was also observed when L. casei-wPspC was incubated only with the secondary antibody (median value of 14.8).

Figure 1

Expression of PspC by Lactobacillus casei. (a) Lysates from L. casei-PspC (L.c.-PspC) or L. casei-wPspC (L.c.-wPspC) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes that were incubated with anti-PspC antibodies generated in mice. Lysates from L. casei were used as a negative control. PspC (filled arrow) and PspC fused with the anchoring peptide (dashed arrow) are pointed. A concentration curve of recombinant PspC was used as a reference for the estimation of PspC expression by L. casei. (b) Lactobacillus casei (L.c.), L. casei-PspC (L.c.-PspC) or L. casei-wPspC (L.c.-wPspC) were incubated with the anti-PspC antiserum, washed and further incubated with anti-mouse IgG-fluorescein isothiocyanate. L.c.-wPspC incubated only with the secondary antibody (filled area) was used as a negative control. Flow cytometry analyses were conducted with 10 000 gated events recorded. Medians of the curves are indicated for each sample.

Induction of anti-PspC antibodies and inhibition of pneumococcal colonization

Before the immunization experiments, the permanency of the different L. casei strains in the nasal mucosa was determined, in groups of four animals. After inoculation of one dose containing 109 cells (day 1), the three strains, L. casei, L. casei-PspC and L. casei-wPspC, could be recovered from nasal washes on days 1, 3 and 5, with no differences observed among the groups. No erythromycin-resistant bacteria could be recovered from nasal washes after day 5 (data not shown).

Immunization of mice was conducted according to the different protocols. In the first experiments, L. casei-PspC was inoculated with six doses through the nasal or three doses through the sublingual routes. Control groups received either L. casei or were non-immunized (Non). Prime-boost protocols were also tested in which animals received the first four doses of L. casei or L. casei-PspC through the nasal route and 15 days later, one dose of the recombinant protein (L. casei+rPspC and L. casei-PspC+rPspC, respectively). As a control, one group of mice received only the last dose of recombinant protein (Non+rPspC). Analysis of anti-PspC IgG in sera or IgA in vaginal and nasal washes revealed that the L. casei-PspC was not able to induce significant levels of antibodies by both nasal and sublingual routes (data not shown). In addition, prime-boost protocols in which four nasal doses of L. casei-PspC were followed by one subcutaneous dose of rPspC were also tested, but no induction of anti-PspC IgG or IgA was observed (data not shown). Experiments were then conducted to evaluate the effect of vaccination on pneumococcal colonization. Thus, 21 days after immunization, the same groups of mice were challenged with the 0603 pneumococcal strain (serotype 6B) through the nasal route and pneumococcal loads in nasal washes were evaluated 5 days later (Fig. 2a). A slight, although not significant, decrease in pneumococcal recovery from nasal washes was observed in mice immunized with L. casei through the nasal route, when compared with non-immunized mice. On the other hand, nasal immunization of mice with L. casei-PspC significantly reduced the pneumococcal nasopharyngeal loads when compared with nonimmunized mice or mice immunized with L. casei. No reduction on pneumococcal colonization was observed in mice inoculated with L. casei or L. casei-PspC through the sublingual route, when compared with non-immunized mice. Therefore, pneumococcal colonization in mice immunized with L. casei-PspC through the nasal route was also significantly lower when compared with mice immunized with the same formulation through the sublingual route. A significant reduction was also observed when this group was compared with the group that was subjected to the prime-boost protocol (L. casei-PspC+rPspC). None of the other groups displayed a significant reduction in pneumococcal colonization (Fig. 2a).

Figure 2

Pneumococcal colonization in immunized mice. (a) Mice were immunized with six doses of L. casei (L.c.) or L. casei-PspC (L.c.-PspC) by the nasal route or three doses of the same formulations by the sublingual (s.l.) route. Non-immunized mice were used as controls. Additional groups received four doses of the L. casei formulations, followed by one subcutaneous dose of PspC (L.c.+rPspC or L.c. PspC+rPspC). Controls received only the subcutaneous dose of PspC (Non+PspC). (b) Mice were immunized with six doses of L. casei (L.c.), L. casei-PspC (L.c.-PspC) or L. casei-wPspC (L.c.-wPspC) by the nasal route. Five days after pneumococcal challenge, dilutions of individual nasal washes were plated on blood agar and α-hemolytic colonies were counted after a 24-h incubation. Log10 of the total CFU and the median of each group are shown. One colony was considered when no colonies were observed on the plates. The limit of detection of the assay was 40 colonies per total sample. *P≤0.05, **P<0.01, Mann–Whitney U-test. #P=0.07 and P=0.03, by two-tailed and one-tailed Mann–Whitney U-test, respectively. Results are representative of two independent experiments.

The effect of nasal immunization was also evaluated for the L casei strain expressing PspC attached to the cell wall (L. casei-wPspC). As observed in Fig. 2b, immunization of mice with both strains expressing PspC led to a decrease in pneumococcal colonization. However, significant results from the two-tailed Mann–Whitney U-test were only observed when we compared the non-immunized group with the group immunized with L. casei-PspC. Once again, a reduction in pneumococcal colonization (albeit not significant) was also observed for the group that received L. casei.

Induction of immune responses after challenge

To further investigate the effect on pneumococcal colonization, we decided to analyze the immune responses elicited by the L. casei-PspC vaccine after challenge. When we tested the immune responses 24 h after challenge, we observed that the levels of anti-PspC IgA in nasal washes were very similar among the groups (Fig. 3a). No significant reduction in pneumococcal colonization could be observed at this point (Fig. 3b). A slightly higher secretion of TNF-α by spleen cells was observed for the group immunized with L. casei-PspC when compared with non-immunized mice (Fig. 4a), while the secretion of IFN-γ was similar for all the groups (Fig. 4b). The levels of IL-17 were below the limit of detection (data not shown).

Figure 3

Immune responses and pneumococcal colonization 24 h after challenge. (a) Anti-PspC IgA was determined by ELISA in nasal washes collected 24 h after pneumococcal challenge. Concentrations were determined using an IgA concentration curve. Individual antibody levels and the median of each group are shown. (b) Twenty-four hours after pneumococcal challenge, dilutions of individual nasal washes were plated on blood agar and α-hemolytic colonies were counted after a 24-h incubation. Log10 of total CFU and the median of each group are shown. One colony was considered when no colonies were observed on the plates. The limit of detection of the assay was 40 colonies per total sample. Results are representative of two independent experiments.

Figure 4

Cytokine secretion by spleen cells 24 h after pneumococcal challenge. Spleen cells were isolated from immunized mice 24 h after colonization challenge and incubated with PspC for 72 h. Secretion of TNF-α (a) and IFN-γ (b) was detected in the supernatants through sandwich ELISA. The levels observed in the supernatants of nonstimulated cells were discounted. Bars represent the mean of six animals per group with the SDs. *P=0.05, Mann–Whitney U-test. Results are representative of two independent experiments.

At a later point, 5 days after challenge, the anti-PspC IgA levels were already increased in nasal washes from mice immunized with L. casei-PspC by the nasal route. This was not observed in the group immunized by the sublingual route or the group that was subjected to the prime-boost protocol (Fig. 5a). Moreover, anti-PspC IgA levels were also significantly higher in the group inoculated with L. casei-wPspC by the nasal route, 5 days after challenge (Fig. 5b). On the other hand, no differences in the mucosal levels of anti-PspC IgG were observed and, at this point, the secretion of TNF-α, IFN-γ and IL-17 by spleen cells was very low and similar in all the groups tested (data not shown).

Figure 5

Induction of anti-PspC IgA in immunized mice, 5 days after challenge. (a, b) Anti-PspC IgA was evaluated in nasal washes of mice immunized with the different protocols and formulations. Samples were collected 5 days after pneumococcal challenge and were analyzed by ELISA. Concentrations were determined using an IgA concentration curve. Individual antibody levels and the median of each group are shown. *P<0.05, **P<0.01, Mann–Whitney U-test. Results are representative of two independent experiments.

Induction of specific IgA correlates with reduction in pneumococcal colonization

Animals immunized with L. casei-PspC through the nasal route were used for the analysis of correlation between antibody production and colonization by S. pneumoniae. For this, the data shown in Figs 2 and 5 were grouped. As observed in Fig. 6a, the most pronounced reduction in colonization was observed for the group immunized with L. casei-PspC, with significantly lower levels when compared with non-immunized animals or animals inoculated with L. casei. In addition, a reduction of pneumococcal colonization was also observed in the group immunized with L. casei, in comparison with the non-immunized group (Fig. 6a). Significantly higher concentrations of anti-PspC IgA were observed only in animals immunized with L. casei-PspC (Fig. 6b). Moreover, a significant negative correlation between the levels of anti-PspC-IgA and bacterial loads in nasal washes was observed (Fig. 6c).

Figure 6

Correlation between anti-PspC IgA levels and pneumococcal recovery from nasal washes, 5 days after challenge. Pooled data of pneumococcal colonization (a) and induction of anti-PspC IgA (b) from two independent experiments are shown. *P<0.05, ***P<0.001, Mann–Whitney U-test. One colony was considered when no colonies were observed on the plates. The limit of detection of the assay was 40 colonies per total sample. The data were used for the analysis of correlation using Spearman's correlation test (c).

Discussion

Lactic acid bacteria modulate the immune system. These properties are well characterized for several strains of Lactobacillus (Cross, 2002; Plant & Conway, 2002; Mohamadzadeh et al., 2005), which have also been proposed and tested as live vaccine vectors carrying heterologous antigens (Ferreira DM et al., 2008; Ferreira PCD et al., 2008; Wells & Mercenier, 2008; Mohamadzadeh et al., 2009). The first report of benefits against pneumococcal respiratory infection was shown by an increased protective response induced by nasal inoculation of mice with an isolate of Lactobacillus fermentum (Cangemi de Gutierrez et al., 2001). Resistance to pneumococcal infection was also shown in both normal (Racedo et al., 2006) and malnourished mice fed with L. casei (Villena et al., 2005). As a live vaccine vector, different strains of Lactobacillus expressing the pneumococcal surface antigen A (PsaA) were shown to reduce pneumococcal colonization upon nasal immunization of mice (Oliveira et al., 2006). Moreover, nasal immunization of mice with L. casei expressing PspA-induced antibodies that increased complement deposition on the pneumococcal surface and conferred significant protection against lethal challenges (Campos et al., 2008; Ferreira et al., 2009).

In a previous work, our group has shown that immunization with L. casei-PspC did not protect mice against a lethal challenge with the pneumococcal strain ATCC6303 (serotype 3), even though specific immune responses were detected after challenge (Ferreira et al., 2009). Here, we have shown that nasal immunization of mice with L. casei-PspC significantly reduced pneumococcal colonization after challenge with strain 0603 (serotype 6B). This effect was dependent on the route of administration, since sublingual immunization did not exert any effect on pneumococcal colonization. Moreover, the number of L. casei-PspC doses was important, since a prime-boost protocol, using rPspC through the subcutaneous route after nasal immunization with L. casei-PspC, was not protective. Although very low levels of PspC were produced by the construction that directs the protein to the L. casei cell wall (L. casei-wPspC), the protein was available at the surface and well recognized by the polyclonal anti-PspC antiserum. Nasal immunization of mice with L. casei-wPspC also reduced pneumococcal colonization, but the results were only marginally significant, since the two-tailed Mann–Whitney U-test has shown a P=0.07, when compared with non-immunized mice. However, if we assume that the vaccination effect would be beneficial against pneumococcal colonization (considering the previously described effects of L. casei on animal models of pneumococcal infection) and apply the one-tailed Mann–Whitney U-test, pneumococcal loads in this group are significantly lower (P=0.03, compared with the non-immunized group). In addition, the simple nasal administration of L. casei had an effect on pneumococcal colonization as described previously by our group (Oliveira et al., 2006). No differences in the permanency of the Lactobacillus strains (L. casei, L. casei-PspC or L. casei-wPspC) were observed after nasal inoculation in mice, and the maximum period of recovery from nasal washes (up to 5 days after inoculation) was very similar to that observed before for the strains expressing PsaA (Oliveira et al., 2006) and PspA (Campos et al., 2008) using the same expression vector. It is possible that the expression vector remains stable in the bacteria for only a few days after the inoculation, since we are measuring the recovery of erythromycin-resistant L. casei. In this case, L. casei could remain longer, but were not detected by this protocol.

Although we have observed a reduction in colonization, no anti-PspC antibodies were observed in vaginal or nasal mucosa as well as in the sera of mice immunized with L. casei-PspC, before the challenge. The failure in inducing specific antibodies to PspC has already been shown by our group in immunized BALB/c mice (Ferreira et al., 2009). Twenty-four hours after the colonization challenge, the levels of both mucosal and systemic anti-PspC antibodies were similar in all the groups. On the other hand, a significant increase in anti-PspC IgA in nasal washes was observed 5 days after challenge in the group immunized with L. casei-PspC. This result suggests that the vaccination with L. casei-PspC primes the immune system, resulting in an earlier induction of mucosal anti-PspC antibodies upon contact with the pathogen. In a mice model of pneumococcal carriage, antibodies' responses to PspC were not detected until 28 days after challenge (Palaniappan et al., 2005). Nevertheless, nasal immunization of mice with PspC in combination with CTB (cholera toxin B subunit) significantly reduced pneumococcal colonization (Balachandran et al., 2002). Interestingly, in the same study, the authors could not detect anti-PspC IgA in nasal washes before the challenge and the levels of these antibodies in the sera were only slightly higher. In our model, anti-PspC IgG were not detected under all the conditions tested. In a previous work we have shown that immunization of mice with L. casei-PsaA induced very low levels of anti-PsaA antibodies, but also led to the inhibition of pneumococcal colonization. However, immune responses after challenge and possible correlations of antibody production and colonization were not analyzed in those experiments (Oliveira et al., 2006). In the present work, when we pooled the results from two experiments, we could clearly observe the induction of anti-PspC IgA, 5 days after challenge, in the group immunized with L. casei-PspC and a significant negative correlation between the levels of these antibodies and pneumococcal recovery from nasal washes. Using a model of pneumococcal infection, Trzcinski and colleagues have also observed significant correlations between antibody production against pneumococcal antigens such as PsaA and PspA and protection against pneumococcal colonization. The authors further demonstrate that the antibodies were not the effectors of protection in their model, but rather reflect the intensity of the immune response elicited (Trzcinski et al., 2005). This possibility cannot be excluded here. Pneumococcal colonization is also reduced in mice immunized with L. casei and no induction of anti-PspC antibodies was observed in this group, indicating that this may be an effect of unspecific stimulation of the immune system as already described for Lactobacillus strains (Villena et al., 2005; Racedo et al., 2006).

The activation of CD4+ T lymphocytes and the secretion of IL-17 have been implicated in the protection against pneumococcal colonization induced by a whole-cell-inactivated vaccine and a vaccine composed of PsaA, PspC and a pneumolysin mutant (Malley et al., 2006; Basset et al., 2007; Lu et al., 2008). Such responses were also implicated in the clearance of colonization in naïve mice (Zhang et al., 2009). In our work, only a slight induction in TNF-α secretion was observed in mice immunized with L. casei-PspC, 24 h after challenge. No differences in IFN-γ secretion were observed, whereas the levels of IL-17 were below the limit of detection. Five days after challenge, the levels of all these cytokines were low (data not shown). Recently, a negative correlation between the induction of IgG2a in the sera of mice and pneumococcal nasopharyngeal colonization was established for a DNA vaccine expressing PspA (Ferreira et al., 2010). In addition, the induction of secretory IgA by a nasal vaccine composed of PspA and a DNA encoding an Flt3 ligand as an adjuvant was shown to be essential for protection of mice against pneumococcal nasal colonization (Fukuyama et al., 2010). Thus, at least two mechanisms seem to be implicated in protection against pneumococcal colonization and may be preferentially induced by different vaccines or conditions. In the case of L. casei-PspC, the induction of IgA after challenge may contribute towards pneumococcal clearance. Finally, an important consideration is related to the sequence variability of PspC. Currently, we have only been able to shown protection against a pneumococcal strain that expresses a PspC with high identity at the N-terminal region to the one expressed by L. casei-PspC. Further studies on coverage among pneumococcal strains expressing different PspC molecules would be necessary.

Acknowledgements

This work was supported by FAPESP and CNPq. We thank Dr Jorge M.C. Ferreira Jr for the help with FACS analysis and Fabiana Barros Rodrigues for technical support.

Footnotes

  • Editor: Willem van Eden

References

View Abstract