Present address: Courtney Shires, Department of Otolaryngology, University of Tennessee Health Science Center, Memphis, TN 38163, USA.
Characterization of Streptococcus pneumoniae isolated from children with otitis media
Version of Record online: 23 APR 2007
FEMS Immunology & Medical Microbiology
Volume 50, Issue 1, pages 119–125, June 2007
How to Cite
Onwubiko, C., Shires, C., Quin, L. R., Swiatlo, E. and McDaniel, L. S. (2007), Characterization of Streptococcus pneumoniae isolated from children with otitis media. FEMS Immunology & Medical Microbiology, 50: 119–125. doi: 10.1111/j.1574-695X.2007.00245.x
Editor: Jenelle Kyd
- Issue online: 23 APR 2007
- Version of Record online: 23 APR 2007
- Received 28 December 2006; accepted 23 February 2007.First published online April 2007.
- otitis media;
- capsular serotypes;
- BOX PCR
Streptococcus pneumoniae is the main causative agent of acute otitis media in children. Serotype-based vaccines have provided some protection against otitis media, but not as much as anticipated, demonstrating the need for alternative vaccine options. Pneumococcal otitis media isolates were obtained from children 5 years old or younger from hospitals around Mississippi in the prevaccine era (1999–2000). These isolates were compared by capsular typing, pneumococcal surface protein A (PspA) family typing, antibiotic susceptibility, and DNA fingerprinting. Our study shows that there is great genetic variability among pneumococcal clinical isolates of otitis media, except with regard to PspA. Therefore, efforts focused on the development of a PspA-based pneumococcal vaccine would be well placed.
Streptococcus pneumoniae is a major cause of pneumonia, meningitis, and otitis media worldwide (Tan, 2002). Pneumococcal infection occurs after a carriage strain moves from the nasopharynx to invade another body site (Weiser, 2004). As the nasopharynx of children is the main reservoir for pneumococci, children are most susceptible to pneumococcal diseases. Streptococcus pneumoniae is the most common cause of acute otitis media in children, resulting in an estimated 7 million cases and 5 billion dollars in health-care costs per year in the United States alone (Bondy et al., 2000; Tan, 2002). Pneumococcal otitis media results in more damage to the ear and is more likely to lead to the reoccurrence of infection than other causes of otitis media (Fireman et al., 2003). The current mode of treatment often includes a course of antibiotics, but, with increasing antimicrobial resistance, other forms of treatment and/or prevention are needed.
Currently, a 7-valent polysaccharide conjugate vaccine is available that has been shown to cause a slight (6–7.8%) decrease in the overall number of cases of otitis media, with a larger decrease (57%) in otitis media caused by serotypes covered by the vaccine (Eskola et al., 2001; Fireman et al., 2003). However, the vaccine does not provide protection for all common serotypes responsible for causing otitis media. Furthermore, it has been shown that those vaccine serotypes are now being replaced by nonvaccine serotypes as major causes of disease (Eskola et al., 2001). While this vaccine is a positive step towards prevention in pneumococcal disease, more needs to be done to increase the coverage of vaccines and to make them more widely available.
To accomplish these goals, work is being carried out to develop vaccines based on pneumococcal proteins that are associated with virulence and are therefore found in all disease-causing strains. These virulence factors include pneumococcal surface protein A (PspA) and C, and the primary pneumococcal toxin pneumolysin (McDaniel & Swiatlo, 2004).
PspA, a choline-binding protein, is required for full pneumococcal virulence (McDaniel et al., 1987). One of the main ways PspA aids the pneumococcus in escaping host defences is by binding apolactoferrin in the blood and preventing conversion to its bactericidal form, lactoferrin (Håkansson et al., 2001). Host evasion is also accomplished by inhibiting the activation of the complement system (Tu et al., 1999). PspA has been shown to elicit protection against pneumococcal disease and is a potential vaccine candidate because it is expressed by all pneumococci and because PspA from one serotype can provide cross-protection against PspA expressed by other serotypes (McDaniel et al., 1991; Moore et al., 2006). PspA proteins have been divided into six clades that have been grouped into three families based on DNA sequence similarities (Hollingshead et al., 2000), with most pneumococcal strains belonging to family 1 or 2.
The purpose of this study was to examine pneumococcal otitis media isolates from a limited geographical area to determine the level of genetic relatedness between them. By examining these strains by capsule typing, PspA family typing, BOX PCR, and antibiotic resistance patterns, we aimed to highlight similarities in the strains that cause otitis media in order to provide a basis for the future development of vaccination or treatment options.
Materials and methods
Bacterial strains and growth conditions
The Mississippi Pneumococcal Collection comprises a group of clinical isolates obtained from clinical laboratories in hospitals throughout Mississippi during 1999–2000; these isolates were grouped by anatomical source. From this collection, we chose ear isolates obtained from children ≤5 years of age presenting to Mississippi hospitals with earache. These isolates were obtained by means of tympanocentesis, myringotomy, or from drainage after rupture of the eardrum. These isolates were screened by the production of α-hemolytic colonies on blood agar and by optochin sensitivity to confirm identity as S. pneumoniae. Bacteria were grown to mid-log phase in Todd–Hewitt broth containing 0.5% yeast extract and stored in 10% glycerol at −80°C.
Capsular typing was carried out by multiplex PCR as described by Pai et al. (2006). Briefly, liquid cultures of pneumococci were grown to mid-log phase and collected by centrifugation. Pellets were suspended in pneumococcal lysis buffer (0.1% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 0.15 M sodium citrate) and incubated at 37°C for 1 h. Lysates were diluted with phosphate-buffered saline and stored at −20°C. PCR reagents were as follows: GoTaq Green Master Mix (Promega, Madison, WI), primers as described (Pai et al., 2006), lysate as template, and H2O in a total volume of 25 μL. Thermal cycling was carried out under the following conditions: 94°C for 2 min; 30 cycles of 94°C for 45 s, 54°C for 45 s, and 65°C for 2 min 30 s; and 65°C for 5 min. Gel electrophoresis was used to compare the band sizes of the PCR products with a 100-bp ladder (Promega) in order to determine the capsule type (Pai et al., 2006). Isolates that we were unable to type or that were typed as 6A/B were further analysed by the multiplex bead assay (Yu et al., 2005; Lin et al., 2006).
DNA was isolated from pneumococci according to the genomic DNA miniprep protocol (Ausubel et al., 1997) with slight modifications. Pneumococci were grown to mid-log phase and collected by centrifugation. The pellet was lysed at 37°C for 30 min in Tris-EDTA and sodium dodecyl sulfate. DNA was treated with cetyltrimethylammonium bromide/sodium chloride and extracted with chloroform/isoamyl alcohol. DNA was precipitated by isopropanol, washed with ethanol, suspended in Tris-EDTA, and stored at −20°C until further use.
PspA family typing
PspA typing was performed by PCR using the family 1 and 2 primers as previously described (Payne et al., 2005), DNA template as isolated above, and GoTaq Green Master Mix (Promega) in a total volume of 25 μL. For positive controls, we used DNA isolated from WU2, a PspA family type 1 pneumococcal strain (McDaniel et al., 1987), and the family type 2 pneumococcal strain EF5668 (McDaniel et al., 1998). Cycling conditions were as follows: 94°C for 10 min; 30 cycles of 94°C for 1 min, 60°C for 1 min, 72°C for 3 min; and 72°C for 10 min. The samples were run in two different reactions: one reaction with primers designed to detect PspA family type 1, and another reaction to detect family type 2. Each set of PCR samples was analysed separately by gel electrophoresis: those samples that produced a similar banding pattern to WU2 were classified as family type 1, and those with the same banding pattern as EF5668 were classified as family type 2; some strains exhibited the banding pattern of both controls, and these were considered as both family type 1 and type 2 (Payne et al., 2005).
BOX PCR was performed with the BOXA1R primer using thermal cycling conditions as described previously (Payne et al., 2005). PCR was carried out under the following conditions: DNA template as isolated above, 200 μM dNTP, Taq DNA polymerase (New England Biolabs, Beverly, MA), 1 × ThermoPol buffer (New England Biolabs), BOXA1R primer and H2O in a total volume of 50 μL. Products were visualized on agarose gel to confirm success of reaction. PCR products were then analysed using the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA), which calculates band sizes of fragments based on an internal ladder. The computer-generated patterns and band sizes were used to examine genetic variability. Band sizes differing by ≤2% were considered equivalent, and strains that differed by three bands were classified as separate BOX types (Payne et al., 2005).
Pneumococcal strains were tested for antibiotic susceptibility by microbroth dilution according to the National Committee for Clinical Laboratory Standards (NCCLS) (Gillis et al., 2005). Strains were tested against the following antibiotics: penicillin, amoxicillin, ceftriaxone, cefuroxime axetil, azithromycin, clindamycin, tetracycline, trimethoprim-sulfamethoxazole (TMP-SMX), levofloxacin, gatifloxacin, moxifloxacin, and vancomycin. According to NCCLS guidelines, strains were classified as susceptible, intermediate, or resistant; for simplicity, intermediate strains were included in the resistant group.
Of the ≥300 strains in the Mississippi Pneumococcal Collection, 29 isolates fell within the parameters we set for the study. Twelve of the isolates were obtained from males, and 17 were from females. The ages of the children ranged from 6 months old to 5 years old. These isolates came from all geographical regions of the state.
Capsule and PspA typing
To determine similarities in capsular serotype of the clinical isolates, we serotyped them using multiplex PCR. Of the 29 strains examined, the most common serotypes were 19F and 23F, with six strains (20.7%) falling within each group (Table 1). Other common capsular types were 14 and 19A, containing five and four strains, respectively. Twenty-one (72.4%) of the isolates belonged to serotypes covered by the currently licensed heptavalent conjugate vaccine (4, 6B, 9V, 14, 18C, 19F, 23F). Of the remaining isolates, six were typed either 6A or 19A, which are considered vaccine-related serotypes (McEllistrem et al., 2005). One strain (AW124) was nontypeable by the methods used in this study; this strain was retested and confirmed to be pneumococcus by optochin sensitivity and bile solubility.
|Capsule type||Strain||PspA type||BOX type|
The majority (55.2%) of the isolates that we studied belonged to PspA family 1. The remaining strains fell almost equally into family 2 (24.1%) or into both family 1 and 2 (20.7%).
We measured the susceptibility of the clinical isolates to a number of antibiotics that are used to treat otitis media. Sixty-five percent of strains were resistant to penicillin as defined by a minimum inhibitory concentration of ≥0.12 (Fig. 1), and of these penicillin-resistant strains, only one had not developed resistance to any of the other antibiotics tested in this study. However, the antibiotic to which most strains were resistant was TMP-SMX (69%). All isolates were susceptible to the fluoroquinolones (levofloxacin, gatifloxacin, and moxifloxacin) and to vancomycin. Other antibiotics that had good activity against many strains were ceftriaxone and clindamycin. Fourteen (48.3%) of the strains had developed resistance to four or more antibiotics, while only five (17.2%) were susceptible to all antibiotics tested.
BOX PCR has previously been shown to be a useful tool for typing pneumococci on the basis of genetic differences (Hermans et al., 1995; van Belkum et al., 1996). Therefore, we used this method to examine pneumococcal otitis media isolates for genetic relatedness. We performed BOX PCR and grouped strains according to their PCR banding pattern, allowing for a difference of three bands between strains and a tolerance of ≤2% between bands. By these criteria, we found that the 29 isolates examined fell into 24 BOX types (Fig. 2, Table 1). Only three of these groups contained more than one strain: group G (three strains), group V (two strains), and group W (three strains). Group V isolates were similar in every aspect of comparison, even to the point of resistance to the same antibiotics. The only point of difference was that while one strain was PspA family type 1, the other expressed both PspA family type 1 and family type 2. Group W isolates were not as comparable: two of the strains were PspA type 1, and two were capsule serogroup 19, and all three strains were resistant to penicillin, azithromycin, and TMP-SMX (two of the three strains had additional antibiotic resistances). In Group G, however, even though all isolates produced similar banding patterns, one isolate was noticeably different from the other two isolates within the group. This isolate was susceptible to all antibiotics, was nontypeable by the methods used in this study, and was both PspA family type 1 and family type 2; however, the other two strains were both PspA type 1 and had resistance to penicillin, cefuroxime, and azithromycin. So we observed that, even within genetically related groups, there was variation in antibiotic susceptibility and capsule serotypes among isolates. This method of analysis demonstrates that there was great variability among these pneumococcal isolates that produced the same clinical manifestation.
Our analysis illustrated the degree of variability in pneumococcal otitis media isolates. The BOX typing showed that there is great genetic variability among these isolates, even though they were isolated from patients presenting with similar complaints. BOX type is not in itself necessary for vaccine development, but shows that, because there is so much genetic variability among strains, it is important to target some aspect of the pneumococcus in which there is less variability. The groups identified by BOX PCR indicate that similarity in capsule serotype does not directly correspond to genetic relatedness. This finding agrees with the conclusion reached by others using methods of genotyping other than BOX PCR (Hall et al., 1996; Enright & Spratt, 1998; Givon-Lavi et al., 1999). However, our results indicated that capsule type is associated with antibiotic resistance to an extent. We observed that many of the penicillin- and multidrug-resistant strains were serotypes 23F, 19F, 14, 9V, and 6B, similar to the findings of Dagan (2004). Because these serotypes are included in the vaccine, we would expect that administration of the 7-valent polysaccharide-conjugate vaccine to Mississippi children would lead to a decrease in infections caused by these antibiotic-resistant strains proportional to the 35% decrease seen in penicillin-resistant strains alone in only 2 years (Whitney et al., 2003).
According to clinical practice guidelines published in May 2004 (AAP & AAFP, 2004), the drug of choice for treatment of acute otitis media (AOM) is amoxicillin, even in cases caused by penicillin-resistant pneumococcal strains. However, our results demonstrated that one-third of the isolates were resistant to amoxicillin, but that half of the penicillin-resistant strains were susceptible to amoxicillin. Other than vancomycin and fluoroquinolones (which are not licensed for treatment of AOM in children), the antibiotics with the highest efficacy among these isolates were clindamycin and ceftriaxone; clindamycin is used in penicillin-allergic patients, and ceftriaxone in patients who otherwise cannot tolerate amoxicillin (AAP & AAFP, 2004; Segal et al., 2005). TMP-SMX is sometimes used for AOM treatment, but, as our results have shown, there is a greater resistance problem associated with this drug than with penicillin, and it is, therefore, no longer considered an acceptable choice in the treatment of pneumococcal AOM (Leiberman et al., 2001).
Our capsule-typing results indicate that the current 7-valent conjugate vaccine would provide protection against the majority of pneumococcal strains causing AOM in Mississippi. However, even though our study only contained a few strains that were not covered by the vaccine, Porat et al. (2004) have shown that, even with the 11-valent vaccine, 20% of cases of AOM result from immunologically unrelated strains, and that these strains are developing resistance as well. Thus, while capsule-based vaccines can decrease the incidence of disease caused by certain drug-resistant serotypes, these will be replaced with other resistant, nonrelated serotypes unless a vaccine is developed that includes most capsule types. Moreover, pneumococci are able to acquire new capsule types by genetic transformation (Nesin et al., 1998). Since over 90 capsular serotypes have been identified, pneumococci with capsule-switching capability could produce an ever-changing population of pneumococci that would constantly evade the host immunity, again resulting in the need for multiple vaccines that target all capsule serotypes. Such vaccines would be difficult to produce, which in turn means they would also be too expensive for the majority of the world's population. So, while a capsule-based vaccine is effective, an ideal alternative vaccine would be noncapsule based, and would instead be based on conserved pneumococcal proteins. Not only would such a vaccine avoid the problem of antibiotic resistance [which is related to capsule type (Dagan et al., 2000)], but it would also provide wider coverage because it could be used worldwide, regardless of which serotypes most commonly cause disease in any given area.
One such virulence protein that has been widely studied for its vaccine potential is PspA. PspA plays a major role in the virulence of the pneumococcus: the absence of PspA decreases the survival of pneumococcus in the host (McDaniel et al., 1987), thereby limiting the virulence of the bacteria. Immunization with PspA provides cross-protection against PspA of different clades (McDaniel et al., 1991; Tart et al., 1996). Studies have shown that human sera obtained from adults can also cross-react with PspA of different pneumococcal strains (Virolainen et al., 2000). As all of the strains in this study belonged to PspA family 1 or 2 (or both), a PspA vaccine including at least one strain from each of clades 1–4 should elicit sufficient protection against 100% of clinical strains that cause otitis media (Hollingshead et al., 2000). The results of our survey are similar to those of a study of Colombian isolates (Vela Coral et al., 2001), further supporting the usefulness of a PspA-based vaccine in protecting against otitis media and other pneumococcal diseases, especially in light of a study that showed that PspA type is independent of serotype−another trend that we observed (Pimenta et al., 2006). In fact, a study performed by White et al. (1999) showed that immunization with PspA reduced otitis media infections in rats. Virolainen et al. (2000) demonstrated that patients with nonpneumococcal infections had higher PspA antibody titers than those with pneumococcal infection, indicating that the presence of PspA antibodies may contribute to the prevention of pneumococcal disease. This is important because there is no other protein associated with the pneumococcus that has been shown to protect against otitis media and also against more fatal invasive infections (Briles et al., 2000, 2001).
An advantage of our study was that these isolates represent all areas of the state, not just an urban academic setting, and so indicate what is seen in the many rural communities in Mississippi. This survey will add to our limited knowledge base, which needs to be expanded in order to gain an understanding of the characteristics of otitis media caused by S. pneumoniae. What we have seen in Mississippi can be applied to the south-eastern United States, and, to some extent, to the United States as a whole. Our study, like previous ones, provides evidence of the efficacy of a capsule-based vaccine, but there is still a gap in overall coverage. Our data provide further argument for the need for a noncapsule-based pneumococcal vaccine and indicate the potential utility of such a vaccine.
We would like to thank Dr Moon Nahm at the University of Alabama in Birmingham for serotype clarification of our nontypeable strains, and Stephanie Warren for her assistance with BOX PCR analysis. This study was supported in part by the National Institutes of Health, Grant AI43653.
- American Academy of Pediatrics (AAP) and American Academy of Family Physicians (AAFP) Subcommittee on Management of Acute Otitis Media (2004) Diagnosis and management of acute otitis media. Pediatrics 113: 1451–1465.
- 1997) Current Protocols in Molecular Biology. Wiley & Sons Inc., New York. , , , , , & (
- 2000) Direct expenditures related to otitis media diagnoses: extrapolations from a pediatric medicaid cohort. Pediatrics 105: 72–78. , , & (
- 2000) Immunization of humans with recombinant pneumococcal surface protein A (rPspA) elicits antibodies that passively protect mice from fatal infection with Streptococcus pneumoniae bearing heterologous PspA. J Infect Dis 182: 1694–1701. , , , , , , , & (
- 2001) The potential for using protein vaccines to protect against otitis media caused by Streptococcus pneumoniae. Vaccine 19: S87–S95. , , , & (
- 2004) The potential effect of widespread use of pneumococcal conjugate vaccines on the practice of pediatric otolaryngology: the case of acute otitis media. Curr Opin Otolaryngol Head Neck Surg 12: 488–494. (
- 2000) Acute otitis media caused by antibiotic-resistant Streptococcus pneumoniae in Southern Israel: implication for immunizing with conjugate vaccines. J Infect Dis 181: 1322–1329. , , , & (
- 1998) A multilocus sequence typing scheme for Streptococcus pneumoniae: identification of clones associated with serious invasive disease. Microbiology 144: 3049–3060. & (
- 2001) Efficacy of a pneumococcal conjugate vaccine against acute otitis media. N Engl J Med 344: 403–409. , , et al. (
- 2003) Impact of the pneumococcal conjugate vaccine on otitis media. Pediatr Infect Dis J 22: 10–16. , , , , & (
- 2005) Vancomycin-tolerance among clinical isolates of Streptococcus pneumoniae in Mississippi during 1999–2001. Am J Med Sci 330: 65–68. , , & (
- 1999) Marked differences in pneumococcal carriage and resistance patterns between day care centers located within a small area. Clin Infect Dis 29: 1274–1280. , , , & (
- 2001) Characterization of binding of human lactoferrin to pneumococcal surface protein A. Infect Immun 69: 3372–3381. , , , , & (
- 1996) Genetic relatedness within and between serotypes of Streptococcus pneumoniae from the United Kingdom: analysis of multilocus enzyme electrophoresis, pulsed-field gel electrophoresis, and antimicrobial resistance patterns. J Clin Microbiol 34: 853–859. , , , & (
- 1995) Comparative study of five different DNA fingerprint techniques for molecular typing of Streptococcus pneumoniae strains. J Clin Microbiol 33: 1606–1612. , , , , & (
- 2000) Diversity of PspA: mosaic genes and evidence for past recombination in Streptococcus pneumoniae. Infect Immun 68: 5889–5900. , & (
- 2001) Bacteriologic and clinical efficacy of trimethoprim-sulfamethoxazole for treatment of acute otitis media. Pediatr Infect Dis J 20: 260–264. , & (
- 2006) Validation of a multiplex pneumococcal serotyping assay with clinical samples. J Clin Microbiol 44: 383–388. , , , , , , & (
- 2004) Pneumococcal disease: pathogenesis, treatment, and prevention. Infect Dis Clin Pract 12: 93–98. & (
- 1987) Use of insertional inactivation to facilitate studies of biological properties of pneumococcal surface protein A (PspA). J Exp Med 165: 381–394. , , , , & (
- 1991) PspA, a surface protein of Streptococcus pneumoniae, is capable of eliciting protection against pneumococci of more than one capsular type. Infect Immun 59: 222–228. , , & (
- 1998) Comparison of the PspA sequence from Streptococcus pneumoniae EF5668 to the previously identified PspA sequence from strain Rx1 and ability of PspA from EF5668 to elicit protection against pneumococci of different capsular types. Infect Immun 66: 4748–4754. , , & (
- 2005) Acute otitis media due to penicillin-nonsusceptible Streptococcus pneumoniae before and after the introduction of the pneumococcal conjugate vaccine. Clin Infect Dis 40: 1738–1744. , , , , , , , , & (
- 2006) Enhanced protective immunity against pneumococcal infection with PspA DNA and protein. Vaccine 24: 5755–5761. , , & (
- 1998) Capsular transformation of a multidrug-resistant Streptococcus pneumoniae in vivo. J Infect Dis 177: 707–713. , & (
- 2006) Sequential multiplex PCR approach for determining capsular serotypes of Streptococcus pneumoniae isolates. J Clin Microbiol 44: 124–131. , & (
- 2005) PspA family typing and PCR-based DNA fingerprinting with BOX A1R primer of pneumococci from the blood of patients in the USA with and without sickle cell disease. Epidemiol Infect 133: 173–178. , , , , & (
- 2006) Genetic diversity of PspA types among nasopharyngeal isolates collected during an ongoing surveillance study of children in Brazil. J Clin Micro 44: 2838–2843. , , , , & (
- 2004) Four antibiotic-resistant Streptococcus pneumoniae clones unrelated to the pneumococcal conjugate vaccine serotypes, including 2 new serotypes, causing acute otitis media in Southern Israel. J Infect Dis 189: 385–392. , , , & (
- 2005) Acute otitis media – diagnosis and treatment in the era of antibiotic resistant organisms: updated clinical practice guidelines. Int J Pediatr Otorhinolaryngol 69: 1311–1319. , , & (
- 2002) Pneumococcal infections in children. Pediatr Ann 31: 241–247. (
- 1996) Truncated Streptococcus pneumoniae PspA molecules elicit cross-protective immunity against pneumococcal challenge in mice. J Infect Dis 173: 380–386. , , & (
- 1999) Pneumococcal surface protein A inhibits complement activation by Streptococcus pneumoniae. Infect Immun 67: 4720–4724. , , , & (
- 1996) Novel BOX repeat PCR assay for high-resolution typing of Streptococcus pneumoniae strains. J Clin Microbiol 34: 1176–1179. , , , & (
- 2001) Pneumococcal surface protein A of invasive Streptococcus pneumoniae isolates from Colombian children. Emerg Infect Dis 7: 832–836. , , , , & (
- 2000) Human antibodies to pneumococcal surface protein A in health and disease. Pediatr Infect Dis J 19: 134–138. , , , , & (
- 2004) Mechanisms of carriage. The Pneumococcus (TuomanenEI, MitchellTJ, MorrisonDA & SprattBG, eds), pp. 169–182. ASM Press, Washington, DC. (
- 1999) Effects of active immunization with a pneumococcal surface protein (PspA) and of locally applied antibodies in experimental otitis media. ORL J Otorhinolaryngol Relat Spec 61: 206–211. , , , & (
- 2003) Decline in invasive pneumococcal disease after the introduction of protein-polysaccharide conjugate vaccine. New Engl J Med 348: 1737–1746. , , et al. (
- 2005) Rapid multiplex assay for serotyping pneumococci with monoclonal and polyclonal antibodies. J Clin Microbiol 43: 156–162. , , , , & (