Streptococcus pneumoniae Transmission Is Blocked by Type-Specific Immunity in an Infant Mouse Model

ABSTRACT Epidemiological studies on Streptococcus pneumoniae show that rates of carriage are highest in early childhood and that the major benefit of the pneumococcal conjugate vaccine (PCV) is a reduction in the incidence of nasopharyngeal colonization through decreased transmission within a population. In this study, we sought to understand how anti-S. pneumoniae immunity affects nasal shedding of bacteria, the limiting step in experimental pneumococcal transmission. Using an infant mouse model, we examined the role of immunity (passed from mother to pup) on shedding and within-litter transmission of S. pneumoniae by pups infected at 4 days of life. Pups from both previously colonized immune and PCV-vaccinated mothers had higher levels of anti-S. pneumoniae IgG than pups from non-immune or non-vaccinated mothers and shed significantly fewer S. pneumoniae over the first 5 days of infection. By setting up cross-foster experiments, we demonstrated that maternal passage of antibody to pups either in utero or post-natally decreases S. pneumoniae shedding. Passive immunization experiments showed that type-specific antibody to capsular polysaccharide is sufficient to decrease shedding and that the agglutinating function of immunoglobulin is required for this effect. Finally, we established that anti-pneumococcal immunity and anti-PCV vaccination block host-to-host transmission of S. pneumoniae. Moreover, immunity in either the donor or recipient pups alone was sufficient to reduce rates of transmission, indicating that decreased shedding and protection from acquisition of colonization are both contributing factors. Our findings provide a mechanistic explanation for the reduced levels of S. pneumoniae transmission between hosts immune from prior exposure and among vaccinated children.

A key step in the life cycle of pathogens is transmission from one host to another.
In general, this process involves colonization of host surfaces, exit ("shedding"), and acquisition and establishment of the organism by a new, susceptible host. An example of this paradigm is the Gram-positive bacterial pathogen Streptococcus pneumoniae, which colonizes the mucosal surfaces of the human nasopharynx and is shed in nasal and oral secretions. Pneumococcal transmission among the human population occurs through close contact with upper respiratory tract (URT) secretions of colonized individuals (1).
The estimated global burden of serious pneumococcal disease is 14.5 million cases (2), and the WHO estimates that 476,000 deaths of children (Ͻ5 years old) occurred in 2008 due to pneumococcal infections (WHO/IVB 2012). Young children are the main reservoir of S. pneumoniae (3), although healthy adult carriers also serve as a reservoir and as a source for transmission to other hosts (4).
In 2000, the first pneumococcal conjugate vaccine (PCV) was licensed for use in the United States and included the seven most common capsular polysaccharide serotypes responsible for invasive pneumococcal disease in children. Extended versions now include up to 13 of the more than 90 known serotypes (5). These vaccines are approved for use in children as young as 6 weeks old to protect against pneumococcal disease in early infancy (6). Numerous epidemiological studies on the impact of PCVs have indicated that their major contribution to public health is due to the indirect effect of vaccination (herd immunity), by which reduced carriage of S. pneumoniae in vaccinated children decreases pneumococcal transmission to vulnerable unvaccinated groups, especially the elderly (7)(8)(9)(10). However, protection by PCV is incomplete as most pneumococcal serotypes are not included in the current vaccine formulations.
Although pneumococcal disease and colonization have been extensively studied in animal models, there is still little known about transmission of this common pathogen. In this study, we sought to understand how anti-S. pneumoniae immunity affects transmission between hosts. Our prior studies have shown that a critical factor in transmission is the number of pneumococci shed from the nasopharynx of the infected host: those strains that shed at high levels also have higher intralitter transmission rates in an infant mouse model (11,12). Factors that increase S. pneumoniae shedding, such as high-density colonization, influenza A virus (IAV) coinfection (11), and Toll-like receptor 2 (TLR2) deficiency in the context of IAV coinfection (13), also increase transmission rates. High shedding is required for the organism to prevail over a tight population bottleneck that exists in transmission between hosts (13). Similarly, pneumococcal strains that shed at high rates in the absence of IAV transmit at detectable levels in an infant mouse model (12). However, in this model, no host factors have yet been identified that decrease shedding and transmission without impacting colonization density. With the concept of herd immunity in mind, we hypothesize that anti-pneumococcal immunity, induced by prior colonization or vaccination, would decrease the number of pneumococci shed from an infected pup and thereby decrease transmission of the pathogen to littermates.

RESULTS
S. pneumoniae infection induced antibody that reduces bacterial shedding. We utilized an infant mouse model (11,12) to assess the effect of anti-pneumococcal immunity on colonization and shedding of S. pneumoniae. The pups were infected on day 4 of life, as we previously established that pups infected on later days do not shed S. pneumoniae as highly as those infected on day 4. However, as there was not sufficient time to develop an adaptive immune response before infection in 4-day-old pups, it was necessary to induce an anti-pneumococcal immune response in adult female mice that would pass antibodies to their pups. Female mice were immunized by intranasal (i.n.) infection with an S. pneumoniae isolate of serotype 4 (T4) or 23F (T23F). At least 4 weeks post-infection (p.i.), their litters were utilized in shedding and colonization studies. At the time that the pups are infected, the dams were no longer colonized, as previous data show that adult mice fully clear nasopharyngeal colonization of S. pneu-moniae by 21 days p.i. (14). Furthermore, preliminary assessment indicated that pups were not infected from their mothers before receiving the inoculum on day 4 of life.
Pups were infected on day 4 of life with T4 or T23F, and sera were collected 5 days p.i., to assess anti-pneumococcal IgG by an enzyme-linked immunosorbent assay (ELISA) using whole bacteria (in which the plates were coated with the same strain). Prior maternal infection induced an anti-pneumococcal response in mothers that is passed to their litters, as immune pups had significantly higher anti-T4 (Fig. 1A) and anti-T23F (Fig. 1B) serum IgG titers than non-immune pups. Levels of anti-S. pneumoniae IgG were similar in dams and their pups. We were unable to detect an anti-S. pneumoniae serum IgA titer in these pups. Colonization was measured on day 5 p.i.; samples were collected by upper respiratory tract (URT) lavage and plated to determine colonization density. Anti-S. pneumoniae titers did not impact pneumococcal colonization, as T4 and T23F colonization of pups from immune mothers did not differ from S. pneumoniae colonization of pups from unimmunized mothers ( Fig. 1C and D). Although the airway surface fluid is heavily diluted while obtaining URT lavage samples, anti-S. pneumoniae IgG could be detected in nasal lavage samples from immune mice by whole-cell ELISA. S. pneumoniae shedding from the nares of pups was enumerated on days 1 to 5 p.i. Despite similar colonization levels, pups from immunized mothers showed significantly lower T4 (Fig. 1E) and T23F (Fig. 1F) shedding compared to pups from nonimmunized mothers, indicating a role for anti-pneumococcal antibody in decreased S. pneumoniae shedding.
PCV13 induced antibody that reduces bacterial shedding. As pneumococcal infection induces an immune response sufficient to decrease S. pneumoniae shedding upon subsequent exposure to the same serotype, we next assessed the impact of the pneumococcal conjugate vaccine PCV13 (Prevnar13; Pfizer) on S. pneumoniae shedding and colonization. Throughout this study, the term "vaccinated" will refer to mothers that were treated with PCV13 and their pups, whereas "immunized" or "immune" refers to mothers that were infected with S. pneumoniae prior to breeding and their litters. Serum IgG titers, colonization, and shedding were assessed in pups from vaccinated (PCV13) mothers and compared to those of pups from non-vaccinated mothers.
Maternal vaccination induced an anti-pneumococcal polysaccharide (PPS) antibody response in dams that is passed to their pups. Vaccinated pups had significantly higher serum IgG anti-T4 PPS ( Fig. 2A) and anti-T23F PPS (Fig. 2B) titers compared to pups from non-vaccinated mothers, as evaluated by ELISA, using plates coated with purified capsular polysaccharide. This result was expected as both T4 and T23F PPS are included in PCV13. These anti-PPS titers were not sufficient to impact S. pneumoniae colonization in pups: T4 and T23F colonized vaccinated and non-vaccinated pups at similar levels ( Fig. 2C and D). However, vaccinated pups shed significantly less T4 and T23F S. pneumoniae than pups from unvaccinated mothers ( Fig. 2E and F). These data indicate that anti-PPS antibodies are sufficient to reduce S. pneumoniae shedding.
Type-specific antibody is sufficient to reduce S. pneumoniae T23F shedding. To test whether anti-S. pneumoniae antibody can impact shedding, we infected (non-immune) pups on day 4 of life with S. pneumoniae T23F and then 24 or 48 h p.i. treated the pups by i.n. instillation with sera from a T23F-immunized rabbit (15). These experiments were done only with the T23F strain due to the availability of hyperimmune sera to the T23F strain. The pneumococcal nasal shedding sample was collected just prior to and 6 h after treatment with sera (Fig. 3), a time point chosen due to the short half-life of antibody delivered i.n. The average ratio of pre-to post-treatment shedding values was compared to T23F shedding from pups that received pre-immune rabbit sera. Importantly, in pilot studies we established that immune sera did not affect CFU quantification or URT colonization. Treatment with anti-T23F sera significantly reduced S. pneumoniae shedding (Fig. 3A) compared to shedding from littermate controls that received pre-immune control sera (Fig. 3B), which supports the hypothesis that antipneumococcal antibody is sufficient to decrease S. pneumoniae shedding. Interestingly, this effect of anti-T23F sera was most noticeable in pups with high shedding values Similarly, pups from T23F-infected mothers had higher anti-T23F S. pneumoniae serum IgG titers than pups from non-immune mothers. All pups were infected on day 4 of life, and serum was collected 5 days p.i. All ELISA data are represented as mean values with standard errors of the means (SEM). Pups were infected on day 4 of life, and colonization was assessed by culture of URT lavages on day 5 p.i. S. pneumoniae T4 (C) and T23F (D) colonization of pups from immune mothers did not differ from S. pneumoniae colonization of pups from non-immune mothers. Pneumococcal shedding samples, measured by culturing nasal secretions, were collected on days 1 to 5 p.i., and results are shown as median shedding, with each symbol representing the value from an individual pup on a single day. Pups from immune mothers shed significantly less T4 (E) and T23F (F) than pups from non-immune mothers. n Ն 14 pups/group. ns, not significant; **, P Ͻ 0.01; ****, P Ͻ 0.0001 (Mann-Whitney test). Similarly, pups from PCV13-vaccinated mothers had higher anti-T23F PPS serum IgG titers than pups from non-vaccinated mothers. ELISA data are shown as mean titer ϩ SEM. All pups were infected on day 4 of life, and serum was collected 5 days p.i. Pups were infected on day 4 of life, and colonization was assessed by URT lavage on day 5 p.i. (C) T4 and (D) T23F colonization of pups from vaccinated mothers did not differ from S. pneumoniae colonization of pups from non-vaccinated mothers. Pneumococcal shedding was collected on days 1 to 5 p.i. and is shown as median shedding, with each symbol representing the value from an individual pup on a single day. Pups from vaccinated mothers shed significantly less T4 (E) and T23F (F) than pups from non-vaccinated mothers. n Ն 11 pups/group. **, P ϭ 0.01; ***, P Ͻ 0.001 (Mann-Whitney test).
pre-treatment (Fig. 3A). We further explored the role of type-specific antibody by comparing shedding from anti-T23F serum-treated pups to shedding from littermate controls that were treated with serum raised to an unencapsulated derivative of the T23F strain (Fig. 3C). Treatment with anti-T23F sera significantly decreased the shedding ratio compared to treatment with the sera raised against the unencapsulated strain ( Fig. 3), highlighting the role of type-specific anti-PPS antibody in reduced shedding.
Anti-PPS IgG is sufficient to reduce shedding and requires the agglutinating function of immunoglobulin. To test if IgG was responsible for mediating the effect on shedding, we carried out similar experiments to those described above, where pups were infected on day 4 of life with S. pneumoniae T23F; 24 or 48 h p.i. they were treated with IgG purified from the anti-T23F rabbit serum. The IgG was sufficient to reduce shedding (Fig. 4A), similar to whole anti-T23F sera (Fig. 4B). The average ratios of preto post-treatment shedding values were not different between the two treatment groups (Fig. 4).
We next sought to define the mechanism by which anti-S. pneumoniae IgG decreases shedding. We recently reported that anti-S. pneumoniae IgG can block the establishment of pneumococcal colonization when adult mice were passively immunized prior to infection; this effect is independent of Fc-mediated opsonization and requires the agglutinating function of the antibody (16). To address whether agglutination is the mechanism by which IgG impacts shedding, purified anti-S. pneumoniae IgG was digested with pepsin to generate Fab fragments, which no longer have the agglutinating capacity of antibodies and lack the Fc region (16). As a control, the purified anti-S. pneumoniae IgG was also digested with papain to produce F(ab=) 2 fragments in which the Fab fragments are still joined by a disulfide bond and are thus able to agglutinate but also lack the Fc region (Fig. 4). We previously confirmed that the F(ab=) 2 fragments could agglutinate S. pneumoniae in vitro, but Fab fragments did not (16). S. pneumoniae T23F-infected pups were treated with equivalent amounts of Treatment with anti-T23F rabbit sera decreased shedding values, while treatment of littermates with pre-immune rabbit sera (B) or sera of a similar titer raised to the unencapsulated derivative strain (C) did not reduce shedding. Mean pre-treatment/post-treatment shedding ratios are significantly decreased in pups that received anti-T23F sera compared to pups that received pre-immune sera or sera raised to an isogenic unencapsulated strain. Each line represents one pup. Mean shedding ratios were compared by Mann-Whitney test.
Fab and F(ab=) 2 fragments, and we compared the change in shedding pre-to posttreatment. Shedding either did not change or increased upon Fab treatment (Fig. 4C) while treatment of littermate controls with an equivalent quantity of F(ab=) 2 resulted in decreased shedding values (Fig. 4D). The ratio of pre-to post-treatment shedding was significantly lower in F(ab=) 2 -treated pups compared to Fab-treated pups (Fig. 4). Collectively, these data indicate that anti-pneumococcal antibodies, and specifically anti-PPS antibody, is sufficient to decrease S. pneumoniae shedding from infected pups; furthermore, the data suggest that it is the agglutinating function of the IgG to type-specific PPS that is necessary to reduce shedding.
Anti-S. pneumoniae antibodies are passed from mother to pup both in utero (systemic immunity) and post-natally (mucosal immunity). To determine how and when the pups acquired anti-S. pneumoniae antibodies from their mothers, we set up a cross-foster experiment in which pups from an immune or vaccinated mother were placed with a non-immune mother and pups from a non-immune mother were placed with an immune or vaccinated mother within hours after birth. Pups remained with the foster mother for the duration of the experiment. On day 4 of life, the pups were infected with T4 or T23F, and shedding was assessed daily for 5 days p.i. Shedding values were significantly decreased from both immune and vaccinated pups with a non-immune mother and from non-immune pups with an immune or vaccinated mother compared to controls lacking anti-S. pneumoniae antibodies (Table 1), indicating that pups receive anti-S. pneumoniae antibodies from their mother both in utero and post-natally and at levels sufficient to impact pneumococcal shedding.
Anti-pneumococcal immunity and anti-PPS vaccination significantly reduce transmission of S. pneumoniae T4. Finally, we addressed the role of anti-pneumococcal immunity on close-contact transmission of S. pneumoniae from one infected pup to another susceptible littermate. One-half of a litter was infected with T4 ("index" mice), and the other half was not infected ("contact" mice). After a 10-day exposure period, all FIG 4 Anti-PPS IgG is sufficient to reduce pneumococcal shedding, and the agglutinating function of IgG is necessary to impact shedding. Pups were infected on day 4 of life with S. pneumoniae T23F and treated by i.n. instillation 24 to 48 h p.i. with anti-T23F rabbit sera or IgG purified from anti-23F rabbit sera. Shedding was assessed before and 6 h post-treatment. (A) Treatment with anti-23F rabbit sera decreased shedding, as did treatment of littermates with IgG from anti-T23F rabbit sera (B), and these pre-treatment/post-treatment shedding ratios are not different. However, treatment with Fab fragments did not impact shedding (C), whereas treatment of littermates with an equivalent quantity of F(ab=) 2 (D) was sufficient to decrease shedding. S. pneumoniae T23F pre-treatment/post-treatment shedding ratios are significantly decreased in pups that received F(ab=) 2 fragments generated from anti-T23F IgG compared to pups that received Fab fragments of anti-23F IgG. Each line represents one pup. Mean shedding ratios were compared by Mann-Whitney test.
Immunity Blocks Pneumococcal Transmission ® pups were euthanized and assessed for colonization. Each colonized contact pup was considered a transmission event. Only transmission of S. pneumoniae T4 was assessed, as T4 transmission rates, but not T23F rates, were high enough to allow for meaningful comparsion. Immunity to T4 significantly decreased transmission within the litters (Table 2), and transmission was also significantly reduced among pups from PCV13vaccinated mothers (Table 2), as no transmission events occurred in either immune or vaccinated litters. These data indicate that anti-PPS immunity is sufficient to decrease intralitter transmission. To address whether it is the anti-pneumococcal immunity of the index pup (exit stage) or the contact pup (entry state) that is blocking transmission, we set up mixed-litter transmission studies wherein only the index or contact mice were T4 immune (and all dams were non-immune). There were no S. pneumoniae transmission events in either scenario ( Table 2), indicating that both decreased pneumococcal shedding and prevention of S. pneumoniae acquisition are key factors in reducing the transmission of S. pneumoniae.

DISCUSSION
The aim of this study was to determine how immunity to S. pneumoniae impacts host-to-host transmission. We focused on this aspect of the infectious process because extensive worldwide experience with PCV has demonstrated that its main contribution to lowering the burden of pneumococcal disease is by reduced transmission (for S. pneumoniae serotypes included in the vaccine formulation [9,10]). Any novel and potentially more broadly acting pneumococcal vaccine would likewise need to impact transmission to be as effective as the PCV.
Our findings in an infant mouse model showed that anti-S. pneumoniae immunity decreases pneumococcal shedding from infected pups. Since URT colonization densi- a "Immune" refers to mothers immunized by colonization and "Vaccinated" refers to mothers immunized by PCV administration. b Compared to median shedding of non-immune/non-vaccinated pups with a non-immune/non-vaccinated mother (Mann-Whitney test). Non-immune 0 13 0 (0.041) Non-immune Immune 0 12 0 (0.044) a "Immune" refers to mothers immunized by colonization and "Vaccinated" refers to mothers immunized by PCV administration. b Compared to non-immune contact mice with non-immune index mice condition (Fisher's exact test). c See reference 12.
ties were similar among all pups in this study, we eliminated the lower density of bacteria in the URT as a cause of decreased nasal shedding of S. pneumoniae in immune mice. Shedding is variable from pup to pup and time point to time point, as previously documented (12); our comparisons are therefore based on the analysis of many pups over multiple days. The decrease in mean bacterial shedding was correlated with a complete loss of within-litter S. pneumoniae transmission and was consistent with prior observations on the association of high shedding values and intralitter transmission (11)(12)(13). Here we report decreased S. pneumoniae shedding and transmission in pups from mothers that were immune due to either prior colonization or PCV vaccination, making it unlikely that indirect effects (such as a maternally acquired change in the URT flora of pups from previously colonized mothers)-rather than immunity-account for our findings.
Passive immunization with type-specific antisera, IgG, and IgG fragments showed that antibody was sufficient to decrease pneumococcal shedding and that this effect could be attributed to the agglutinating function of IgG. In a study in which the nasal spaces of recently colonized mice were viewed in tissue sections, pneumococci are seen in much larger clusters in immunized compared to nonimmunized animals (16). Agglutination may facilitate more rapid clearance of bacteria that detach from mucosal surfaces of the upper airways and thereby decrease bacterial counts in a shedding assay. These bacterial aggregates might be more easily removed by mucociliary flow rather than expelled from nares into the environment where they can access a new host. Anti-S. pneumoniae IgG is able to block the acquisition of colonization in a dose-dependent manner (if present in sufficient amounts at the time of challenge), and this effect is dependent upon its agglutinating function (16). Observations in the present report suggest that in addition to its ability to block acquisition, the agglutinating effect of mucosal anti-S. pneumoniae IgG decreases the number of shed bacteria below a threshold critical for transmission.
The anti-S. pneumoniae antibody measured in the pups was primarily IgG, and therefore, our data support the role of IgG in decreasing shedding and blocking transmission-an observation consistent with the immune response to PCV, which is administered parenterally and stimulates a robust IgG response (17). Accordingly, a human challenge study showed that anti-capsular antibody generated after vaccination with PCV is sufficient to protect adults against challenge with an S. pneumoniae strain of a serotype in the vaccine (15). Of note, in that study, anti-PPS IgG induced by parenteral vaccination with PCV was measureable in nasal wash samples from human volunteers (15), which correlates with our findings of anti-S. pneumoniae IgG detected in URT lavage samples from immune mice. In the experimental human pneumococcal challenge study, protection from acquisition correlated with the agglutinating activity of nasal wash samples (15). Anti-pneumococcal mucosal IgG is also present following URT S. pneumoniae carriage (18), and our present report demonstrated that anti-S. pneumoniae IgG is induced by colonization in adult mice, mucosal IgG was passed from dam to pup, and IgG was capable of decreasing pneumococcal shedding and blocking transmission.
In this study, the presence of maternally acquired anti-S. pneumoniae antibody did not affect the colonization density of S. pneumoniae in the URT of infant mice, nor did it prevent the acquisition of colonization following i.n. inoculation with~10 3 organisms. Humoral immunity has minimal impact on the numbers of mucosal pneumococci once colonization is established (19,20). We have previously reported that passive i.n. immunization with type-specific antibody could inhibit acquisition and transmission in adult mice (16); infant mice, however, are more susceptible to colonization with S. pneumoniae and can be colonized by a dose of fewer than 10 pneumococci (13)-an amount that is approximately 100-fold fewer than what is required to establish similar levels of colonization in adults. It appears that in infant mice, the amount of mucosal IgG was insufficient to prevent pneumococcal acquisition by i.n. inoculation with the inoculum size tested. In contrast, evidence suggests that the levels of antibody present were adequate to block natural acquisition within the litter, as there was no transmis-sion from non-immune index to immune contact pups. One explanation for the difference in acquisition by inoculation but not transmission might be a matter of the challenge dose: our recent study on the population bottleneck in transmission shows that colonization is most often established only by a single pneumococcal cell/chain (13).
The primary goal of this study was not to test protection via maternal immunity or immunization-a topic that has been the focus of much clinical investigation (reviewed in reference 21). Pneumococcal carriage is common by 2 months of age in the developing world and, in some locations, this pathogen is second only to group B Streptococcus as a cause of invasive bacterial infection in infants (22). Although vaccination during pregnancy leads to high anti-PPS levels in infants, the efficacy of maternal immunization for the prevention of carriage or disease early in childhood is still debated and under investigation. Studies in mouse models demonstrate the protection of newborns from pneumococcal disease through maternal immunization (23) and support a role for the transfer of protective antibody from breast milk or from the serum pool in utero (24). The cross-foster experiments in our study, in which immune pups were raised by a non-immune dam or non-immune pups by an immune dam, both showed decreased shedding compared to non-immune pups raised by a non-immune dam, confirming that protective antibody can be transferred both in utero and post-natally, in accordance with published observations (25,26).
A better understanding of factors that impact S. pneumoniae transmission, including those that affect bacterial colonization, shedding, and acquisition, has implications for future vaccine development. In this study, we provide a mechanistic explanation for the "herd immunity" effect, seen widely after the introduction of PCV. By highlighting the importance of reduced shedding and bacterial acquisition in the S. pneumoniae transmission cycle, we have identified vulnerable points for intervention. This study provides a more comprehensive analysis of how active immunity to S. pneumoniae and, in particular, antibody to PPS, impact transmission and promote herd immunity. PCVinduced type-specific immunity was shown here to be sufficient to reduce shedding and prevent transmission for isolates of two serotypes included in the vaccine formulation. Moreover, in passive immunization assays, high-titer serum raised against whole pneumococci, but not its unencapsulated derivative, decreased pneumococcal shedding. These results point to a requirement for type-specific immunity in blocking pneumococcal transmission. We cannot, however, exclude the possibility that natural infection presents a distinct set of antigens not targeted by the immune serum tested herein. PPS is an immunodominant and abundant antigen, and physiological levels of type-specific antibody readily agglutinate encapsulated pneumococci (15). It remains unclear, therefore, whether there are other serotype-independent pneumococcal targets of agglutinating antibody that could similarly reduce shedding, block acquisition, and prevent transmission.
Finally, there are few tractable models available for examining host factors in transmission, especially involving infections of the respiratory tract. The mechanistic insight from our investigation of S. pneumoniae may be relevant to the role of immunity in the transmission of other infectious agents.

MATERIALS AND METHODS
Bacterial strains and culture. Streptococcus pneumoniae isolate P2406 (streptomycin resistant [Str r ]) is a capsule type 4 (T4) derivative of TIGR4 (12). Strain P1397 is an Str r derivative of P1121, a capsule type 23F (T23F) strain isolated from the nasopharynx in a human experimental carriage study (27). S. pneumoniae strains were grown in tryptic soy (TS) broth (BD, Franklin Lakes, NJ) at 37°C without aeration to an optical density at 620 nm (OD 620 ) of 1.0 or incubated on TS agar plates supplemented with 100 l of catalase (30,000 U/ml; Worthington Biochemical) and Str (200 g/ml) at 37°C in 5% CO 2 , overnight.
Mouse studies. Wild-type C57BL/6J mice (Jackson Laboratories, Bar Harbor, ME) were used in all experiments. Mice were maintained and bred in a conventional animal facility. Pups were housed with a dam for the course of the studies. Following infection, all mice appeared healthy, showed normal activity, and gained weight similar to uninfected controls. These studies were conducted in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals. All mouse studies were approved by the Institutional Animal Care and Use Committee of the New York University Medical Center. All procedures were in compliance with Biosafety in Microbiological and Biomedical Laboratories.