The Pneumococcus

The pneumococcus (Streptococcus pneumoniae or diplococcus) has been profoundly important in the understanding of the human response to infectious disease, the nature of genetic material, and natural transformation as a means of genetic exchange, as well as in the recognition of bacterial resistance to drugs. To develop effective diagnostic tools for the pneumococcus, one has to understand more about its nearest neighbors and include the breadth of diversity of these organisms in strain panels used to validate diagnostic tools—unfortunately, this tends to be the exception rather than the rule. The pneumolysin and autolysin genes have been flagged as potentially useful targets for molecular diagnostics for the past decade or so, and reports throughout this time have always found typical human clinical pneumococci to possess ply and lytA. This probable pathogenicity island (PPI-1) represents one example of no doubt other events that will be revealed by comparative genomics, which has contributed to the evolution of the founding pneumococcus from which the extant population has arisen. A hypothesis was constructed to account for these novel phenotypes and the role of zmpB in pneumococci. However, the development of a diagnostic test based upon single-target identification might be an ambitious hope given the genetic plasticity of the pneumococcus and its naturally transformable relatives.

As with many other gram-positive organisms, the genome of Streptococcus pneumoniae proved difficult to sequence. The majority of insertion sequence (IS) elements have undergone insertions, deletions, and/or point mutations that result in frame shifted or otherwise nonfunctional transposase genes. A primary role for the numerous repeats might be their potential contribution to genomic rearrangements using the repeats as seeds for recombination events, as discussed in this chapter. It has been noted an intriguing functional pattern for proteins that are surface-attached by alternate means. The nonshared genes represent both single-gene insertions and deletions and also a number of nonshared regions. All regions of diversity, except only region outside of the capsular region, represented genes that were present in at least one additional strain from our collection of genetically diverse strains. The mosaic distribution supports the idea that these regions are associated with genome plasticity. The significance of the high density of repeats is that they may play a significant role in generating the genome plasticity that is observed. The vast majority of clinical and nonclinical isolates of S. pneumoniae are transformable, and the frequent exchange of genetic information through transformation could permit a high degree of genetic plasticity.

Recent years have seen a renewed interest in the characterization of the capsular polysaccharides of Streptococcus pneumoniae and many other bacteria. Many of the recent advances described in this chapter have their roots in classic studies of S. pneumoniae that have historical significance both for the pneumococcal field and, often, for biology as a whole. Confirmation of linkage of the genes involved in capsule synthesis and the first maps of capsule loci were obtained through recombination experiments. The presence of insertion sequence (IS) elements and nonfunctional genes (or gene fragments) is a common finding in the capsule loci, as already noted for the type 3 locus and the cryptic type 33F locus in type 37 strains. The essential nature of the capsule in virulence was established in early studies of S. pneumoniae, as described in the chapter. A role for capsule in colonization was demonstrated using the type 2 and type 3 strains that contain defined mutations in the capsule loci and in pgm. Among clinical isolates of different capsular serotypes, levels of complement activation and deposition, as well as phagocytosis, vary in in vitro assays. As has long been the case, the S. pneumoniae capsules will continue to serve as paradigms for studies of virulence factors, immune responses, and polysaccharide biochemistry.

Pneumococci covalently link phosphorylcholine to teichoic and lipoteichoic acids found in the peptidoglycan and cytoplasmic membrane, respectively. The functions of most of the choline-binding proteins (CBPs) are unknown, but a few have been studied in some depth and it is apparent that they have a role in pathogenesis and can be protective immunogens. The presence of choline-binding domains implies that any protein expressing these domains is secreted, since choline is a constituent of teichoic and lipoteichoic acids, which are cell surface polymers. The major pneumococcal autolysin LytA contains four to six choline-binding domains but is not found in eluates of pneumococcal strain Rx1 incubated with choline-containing buffers. Pneumococcal CBPs have been shown to play a role in pathogenesis in various murine models of disease. It is likely that there is some redundancy in the function of CBPs, and this, along with the large effect of the pneumococcal capsule, may explain why more is not known about the function of CBPs in pneumococcal disease. PspA and PspC are the two most well-characterized CBPs in terms of their biological functions and roles in disease. While these two proteins can be considered paralogs based on sequence homology, they make distinct contributions to pneumococcal virulence. The major autolysin of pneumococci, designated LytA, is known to be important in remodeling the cell wall of dividing pneumococci and is the common final point for many processes which lead to cell lysis.

Streptococcus pneumoniae (the pneumococcus) makes a range of molecules that can be considered virulence factors. The organism also makes a range of protein virulence factors, including surface proteins (choline-binding proteins, LPXTG-anchored proteins, and lipoproteins) as well as a range of enzymes (superoxide dismutase, NADH oxidase, zinc metalloproteinases) and the pore-forming toxin pneumolysin (PLY). This chapter focuses on PLY and some of the other protein virulence factors of the pneumococcus. The toxin can have other effects on cells and has been shown to affect the production of proinflammatory mediators such as tumor necrosis factor alpha (TNF-α), interleukin 1β (IL-1β), and IL-6. It was shown that PLY contributes to neuronal damage in a rabbit model of pneumococcal meningitis. An effect of PLY on calcium levels has also been demonstrated in neuroblastoma cells, in which purified PLY was shown to induce apoptosis in a calcium-dependent manner. The changes in calcium levels were due to pore formation by the toxin rather than opening of voltage-gated Ca2+ channels. This study also showed that mitogen-activated protein kinase p38 was important in PLY-induced cell death. The overall virulence gene content of the pneumococcus will presumably dictate its ability to cause disease.

Many bacterial species possess more than one enzyme that hydrolyzes the same bond, a fact that complicates the determination of their biological role(s). This redundancy also contributes to the common thought of assigning a basic role to lytic enzymes in the biology of bacteria. The activity of some pneumococcal murein hydrolases (MHs) appears to be constrained by the membrane lipoteichoic acid (LTA) at the posttranslational level. Cell wall hydrolases (CWHs) of Streptococcus pneumoniae show both substrate and bond specificities. The lytA gene encodes the major S. pneumoniae autolysin (amidase) and represents the first example of a bacterial autolytic gene that was cloned and expressed. LytB is most probably a glucosaminidase capable of degrading Ch-containing cell walls. All the pneumococcal CWHs described have been shown to possess an absolute requirement for the presence of Ch for activity. The cloning of lytA has facilitated the isolation of the genes encoding the cell wall lytic enzymes from pneumococcal bacteriophages based on sequence homologies. This global analysis led the authors to propose that pneumococcal cell wall lytic enzymes could be the result of the fusion of two independent functional domains. The construction of active chimeric proteins between lysins of phage and bacteria led to new enzymes exhibiting novel properties that were, as expected, a combination of those showed by the parental enzymes.

Natural transformation, which requires a set of genes evolved for the purpose, contrasts with artificial transformation, which is accomplished by shocking cells either electrically, as in electroporation, or by ionic and temperature shifts. Although such artificial treatments can introduce very small amounts of DNA into virtually any type of cell, the amounts introduced by natural transformation are a millionfold greater, and Streptococcus pneumoniae can take up as much as 10% of its cellular DNA content. The current understanding of the mechanism of transformation and the genetics of S. pneumoniae has depended on a variety of experimental approaches: tracing of the fate of isotopically labeled DNA, analysis of genetic recombination frequencies, isolation and characterization of transformation-defective and other mutants, DNA cloning and sequencing, and identification and use of the competence-inducing peptide to characterize the regulatory aspects of transformation. Spontaneous and chemically induced mutations in many genes have been obtained; they correspond to various single-site base changes and deletions and insertions of all sizes. Markers located nearby on the chromosome will exhibit linkage, that is, show a cotransformation frequency greater than expected for two separate entry events. For transformation to occur under natural conditions, DNA must be released from donor cells as well as taken up by recipient cells. Binding of SsbB may facilitate recombination, as such proteins do in other systems. Essential to recombination, however, is the recA gene. recA expression is increased during competence.

Bacteria exist within populations, whether it is the population of pneumococci in the nasopharynx of an individual child or the isolates circulating within a local community, within a country, or globally. Multilocus enzyme electrophoresis (MLEE) is such a technique, and it has provided key insights into the population biology of many bacterial pathogens. The relationships among major lineages could be explored using a tree constructed from the concatenated sequences of all seven MLST loci. Pneumococcal clones diversify relatively rapidly, due mainly to the substantial impact of homologous recombination, presumably mediated by genetic transformation, which results in small segments of the chromosome of a recipient pneumococcus being replaced with the corresponding region from a distinct strain. Multiply antibiotic-resistant isolates of Streptococcus pneumoniae cannot have existed prior to the introduction of antibiotics into medicine and are probably less than 30 years old, yet considerable variation in the allelic profiles of the major resistant clones is observed. The carried population is thus of primary interest to the population biologist, and disease isolates need to be considered in the context of carriage. Antibiotic resistance might be expected to have arisen in those clones that were commonly carried in the nasopharynges of children. Serotypes that are rarely encountered in developed countries appear to cause a substantial amount of disease in some developing countries.

The human nasopharynx is the principal ecological niche for the heterogeneous population of Streptococcus pneumoniae (the pneumococcus), which exists as 90 different capsular types or serotypes. The focus of this chapter is to understand the behavior of S. pneumoniae and the factors that affect it in its normal habitat, the human nasopharynx. It is important to clarify that studies which have explored risk factors for pneumococcal carriage have not made a distinction between acquisition and carriage; therefore, data from these studies reflect risk factors related to the prevalence of pneumococci in carriage. There are little direct epidemiological data which show that acquired immunity plays a role in modulating carriage of pneumococci. The increase and subsequent decrease in pneumococcal carriage between 0 and 4 years of age is consistent with acquired immunity playing a role in reducing carriage, but the only direct evidence for natural immunity playing a role in preventing acquisition and thereby reducing carriage comes from a recent human challenge study. The conjugate vaccines have a marked and reciprocal effect on the acquisition and carriage of nonvaccine serotypes. Two additional beneficial effects have been observed in the human population following administration of the conjugate vaccine. The first is a significant reduction in carriage of antibioticresistant pneumococci, which is perhaps not surprising given that the major antibiotic-resistant clones are mainly of vaccine serotypes; the second is a reduction in the acquisition and carriage of vaccine-associated pneumococci in unimmunized younger siblings of vaccine recipients.

This chapter focuses on the epidemiology of pneumococcal disease in terms of pathogen characteristics and host risk factors. The clinical manifestations of Streptococcus pneumoniae infection are protean but can be classified into two major categories: invasive infections, where the organism is isolated from a normally sterile body site, such as the bloodstream or central nervous system, and mucosal infections, most often involving the upper respiratory tract. Our understanding of pneumococcal disease epidemiology is further refined through studies of specific strains of S. pneumoniae. Differences in the polysaccharide structure of the pneumococcal capsule have permitted identification of at least 90 serotypes of S. pneumoniae. Pneumococcal polysaccharide and polysaccharide conjugate vaccines protect against disease by inducing serotype-specific antibodies. Vaccination with pneumococcal conjugate vaccine causes a shift in serotypes causing otitis media. The risk and severity of pneumococcal infection is increased for persons with certain chronic medical conditions, including functional or anatomic asplenia, HIV infection, chronic obstructive pulmonary disease, asthma, cirrhosis, diabetes mellitus, chronic renal failure, cancer (particularly hematological malignancies), organ or bone marrow transplantation, nephritic syndrome, and hypogammaglobulinemia and for persons taking immunosuppressive medications, such as corticosteroid. In temperate areas, pneumococcal infections are more common during the winter months, when respiratory viral infections are most common. Vaccination promises to have the most profound impact on preventing illness and death from pneumococcal disease since the advent of antimicrobial drug therapy.

Streptococcus pneumoniae is one of many closely related oral streptococci of the mitis phylogenetic group that colonize the human oro- and nasopharynx. Recently, experimental carriage studies performed in healthy adults have offered the prospect of utilizing the natural host to further investigate this fundamental aspect of pneumococcal biology. Surface molecules that have been shown to function as adhesins to human epithelial cells include phosphorylcholine (ChoP) and CbpA. ChoP, an otherwise unusual prokaryotic structural component, is common to several other genera residing primarily in the upper respiratory tract, such as Haemophilus, Actinobacillus, Mycoplasma, and Neisseria. The expression of a surface-attached hyaluronidase (a hyaluronate lyase), Hyl, which could facilitate spread through a matrix of hyaluronan, a major polysaccharide component of host connective tissues, suggests that such a strategy may contribute to pneumococcal pathogenesis. Adherence to host structures may be particularly problematic for an encapsulated organism like the pneumococcus. Some degree of encapsulation appears to be essential for colonization, although even small amounts of capsular polysaccharide effectively block attachment to host cells. A final consideration is that pneumococcal infection frequently occurs in the setting of a recent or concurrent upper respiratory infection from common viruses.

To detect potential harmful microorganisms, higher eukaryotes have evolved two types of systems, i.e., innate immunity and adaptive immunity. Monocytes and macrophage express both CD14 and Toll-like receptors (TLRs) on their surfaces. Polymorphonuclear cells were shown to express CD14 and at least TLR2. Peptidoglycan-recognizing proteins (PGRPs) are a new family of pathogen-associated molecular pattern (PAMP)-recognizing molecules that are conserved from insects to mammals. PGRPs have several attributed functions, including direct antimicrobial activity or triggering the production of antimicrobial molecules, signaling via the TLR system, and peptidoglycan degradation. The major and most conserved constituent of the envelope of gram-positive organisms is peptidoglycan. One possible explanation is the dissimilar natures of the bacterial components used. Lipopolysaccharide (LPS) is made of noncovalently linked glycolipid subunits that are emulsified by plasma lipoproteins and lipopolysaccharide-binding protein (LBP), which presents them to the cell receptor CD14. The inflammatory activity might also depend on other constraints, including the secondary structure of the components and/or the stereochemistry of their amino acid constituents. Mesodiaminopimelic acid is a precursor of L-lysine. Therefore, it is tempting to make the provocative speculation that gram-positive animal colonizers have evolved an L-lysine peptidoglycan in order be less well detected by innate immunity. The constant inflammatory response to bacterial surface component might be profitable. As far as disease is concerned, the real problem might not be so much the ability of the host to recognize bacterial intruders as much as the capacity of pathogens to escape early recognition by innate immunity.

Thirty-three years later, the heat-stable, specific components of the opsonic interaction (antibodies) were distinguished from nonspecific, heat-labile serum proteins, now named complement. In the presence of antibody alone, the rate of neutrophil phagocytosis of serotype 3 Streptococcus pneumoniae was accelerated more than sevenfold by the addition of active complement proteins in fresh human serum. Mannose binding lectin (MBL), a member of the collectin family, exhibits the collagen-like domain and the carbohydrate recognition domain of the collectin family. The most compelling evidence for the role of innate immunity in defense against pneumococcal infection focuses on C3. Recent genetic investigations have identified a number of pneumococcal proteins that affect the activation, deposition, or cleavage of C3 so as to interfere with C3-mediated opsonization and phagocytosis. Deleting the choline-binding region of the family 2 PspA generated a mutant that failed to bind human lactoferrin but retained most of its complement deposition. Virulence was moderately attenuated. The pspC gene is present in 75% or more of pneumococcal strains, but the encoded proteins are highly polymorphic and are divided into 11 groups. Studies of type 3 pneumococci, highly resistant to phagocytosis, also pointed to absorption of factor H from plasma and led to the isolation of Hic. Pneumococcal proteins that interact with components of the classical or alternative complement pathways are predominantly surface expressed (except for Pla), related if not completely conserved (e.g., CbpA, PspA, PspC, and Hic), and potently immunogenic.

This chapter discusses clinical syndromes caused by Streptococcus pneumoniae, categorizing them based on pathogenesis and immune response. The current understanding of hematogenous infection is based on an expanded understanding of events in which pneumococci “settle out in” or “seed” various body sites. The resulting infections include meningitis, primary peritonitis, septic arthritis, osteomyelitis, and soft tissue infection. Important in the pathogenesis of acute sinusitis is congestion of the mucosal membranes caused by allergy or viral infection; resulting obstruction at the osteomeatal complex prevents clearance of bacteria. Not surprisingly, the bacteriology of acute maxillary sinusitis is similar to that of otitis media, with S. pneumoniae and/or Haemophilus influenzae being isolated in the great majority of cases. The majority of patients with pneumococcal pneumonia do not have detectable bacteremia, and it is very uncommon for the laboratory to isolate pneumococci from sputum of a patient who does not have a clear clinical picture of pneumonia or acute purulent tracheobronchitis. Empyema, the most common complication of pneumococcal pneumonia in the preantibiotic era, occurred in about 5% of cases and remains the most common today, with an incidence of approximately 2%. S. pneumoniae, despite its somewhat limited array of tissue-damaging enzymes and toxins, remains a prominent cause of infection with a surprisingly broad array of manifestations.

The description of the patient with lobar pneumococcal pneumonia is considered one of the classics of medicine. The history and physical findings usually establish the diagnosis of pneumonia. The major effect of different genetic backgrounds of animals on the course of infection has been particularly apparent in mice. CbpA is the most abundant of the choline-binding proteins and functions as an adhesin in the upper and lower respiratory tract. Of these, PAFr and epithelium derived C3 play essential roles in pneumococcal interactions with cells in the lung that lead to progression from pneumonia to bacteremia and meningitis. In leukopenic animals, the bacterial load in the lung increases but there is no decrease in the incidence of bacteremia, indicating that events in this stage of consolidation are sufficient to lead to invasion without the effects of leukocytes. In vitro studies have demonstrated that invasion of alveolar cells involves recognition of PAFr by pneumococci. Using COS cells transfected with components of the PAFr, it has been shown that the presence of PAFr is necessary for pneumococcal invasion. This choline-dependent invasion mechanism appears to apply to a large number of respiratory pathogens. Pneumococcal pneumonia is very common and causes a dramatic clinical picture due to intense inflammation in the lung.

Pneumococcal meningitis is still an unresolved problem in clinical medicine. A number of recent studies demonstrate the unfavorable outcome of pneumococcal meningitis, reporting mortality rates up to 40%. Most models are based on the injection of pneumococci or their inflammatory components into the subarachnoidal space, followed by the hallmarks of meningitis: the influx of leukocytes into CSF, development of brain edema, intracranial pressure, and typical histological changes. Inhibition of TNF-α results in decreased leukocyte invasion in pneumococcal meningitis. Deficiency in interleukin-18 (IL-18), another cytokine cleaved by caspase-1, results in reduced inflammation and better survival in experimental pneumococcal meningitis. In pneumococcal meningitis, all these steps happen in the arachnoidal microvessels and the triggers for leukocyte-endothelium interactions include histamine and cytokines. In an infant rat model of pneumococcal meningitis, Matrix metalloprotease (MMP) inhibitors decreased inflammatory alterations, including neuronal damage. Different antioxidative strategies interfering with generation of reactive oxygen species attenuate intracranial complications, including neuronal damage in pneumococcal meningitis in the adult as well as in the infant animal model. Inhibition of inducible nitric oxide synthase (NOS) ameliorates inflammation in pneumococcal meningitis. In the rabbit model, apoptosis is a predominant mechanism of neuronal damage and focuses on the dentate gyrus. Hippocampal neurons of patients who have died from pneumococcal meningitis show the morphological criteria of apoptosis and active caspase-3. Although the beneficial effect of dexamethasone has been shown in clinical trials, it should be considered that, at least in experimental models, this drug augments apoptosis in the dentate gyrus.

This chapter focuses on the rates and clinical manifestations of invasive pneumococcal disease in patients at high risk, on proposed mechanisms of impaired defense, as well as on strategies for prevention of invasive pneumococcal disease in these groups. It describes how individual host defects may each contribute independently to the increased risk of invasive disease, and observes that the potential number of immunological risk factors is greatest and rates of invasive pneumococcal disease are highest among HIV type 1 (HIV-1). The extremely high rates of invasive pneumococcal disease in apparently otherwise healthy children suggest the presence of specific defects related to this class of organisms. HIV-1-associated immunodeficiency, rather than behavioral (e.g., smoking), medical (e.g., splenectomy, liver disease), environmental (e.g., seasonality), or bacteriological features (serotypes or colonization), appears to underlie the remarkable rates of disease in this population. Despite the many fold-increased rates of pneumonia and, particularly, invasive pneumococcal disease during HIV-1 infection, no specific overriding risk or target for intervention has been confirmed. The protection provided by both the 23-valent polysaccharide vaccine in adults and the 7-valent conjugate vaccine in children is most apparent and consistent against invasive pneumococcal disease. Specific immune defects can be identified for particular groups at higher risk of invasive pneumococcal infection, whereas more generalized deficiencies in immune and nonimmune defenses underlie bacteremia in others. The development of newer vaccines which are more universally immunogenic and which combat mucosal disease more effectively is an important goal in preventing pneumococcal disease in the immunocompromised host.

This chapter describes the mechanisms and consequences for the pneumococcal population of the selective pressures imposed by antibiotics and vaccines. While inhibiting pneumococci can have effects on other species of upper respiratory tract (URT) bacteria, the chapter focuses on the selective effects of vaccines and antibiotics on pneumococci. The direct effects of antibiotics on S. pneumonia in the nasopharynx of the treated patient depend on the pharmacokinetics and pharmacodynamics of the agent. Streptococcus pneumoniae resistance to the beta-lactam antibiotic class evolved mainly by complex restructuring of the targets of the beta-lactams, the penicillin-binding proteins (PBPs). To summarize the effect of antibiotics on S. pneumoniae ecology, it is clear that antibiotic treatment in any community profoundly affects S. pneumoniae ecology. Furthermore, a significant reduction of carriage of antibiotic-resistant and multidrug resistant (MDR) S. pneumoniae in the younger siblings of the day care attendees as the result of vaccination of their older siblings was observed. The chapter emphasizes the effects of vaccines and antibiotics on the ecology of pneumococci taking place against a background of a genetically diverse population structured by direct and indirect ecological interactions between strains. The importance of understanding pneumococcal population structure has been appreciated since the earliest days. This understanding can be extended to reconcile the disparate results of clinical trials of antibiotics and vaccines, to evaluate the relative selective effect of different antibiotics for resistant strains, and to project the effects of vaccines and antibiotics on future patterns in the prevalence of serotypes and resistant strains.

Streptococcus pneumoniae is one of the most important bacterial causes of respiratory infection and invasive disease in children and adults. This chapter reviews how increased rates of antibiotic-resistant S. pneumoniae have influenced the morbidity and mortality associated with pneumococcal disease and to present optimal therapeutic approaches for management of these infections in children and adults. In an eight-center children’s hospital surveillance study of pneumococcal infections, Kaplan et al. reported the outcome of 100 children with S. pneumoniae bacteremia secondary to penicillin- and cephalosporin-nonsusceptible infections. Informative studies suggest that an immunocompetent child between the ages of 3 and 36 months with culture-proven pneumococcal bacteremia, without meningitis, due to a nonsusceptible isolate can be effectively treated with a parenteral broad-spectrum cephalosporin as an appropriate initial therapy. Prior to beginning this therapy, repeat blood cultures should be obtained to document persistent bacteremia. Clinical presentation, cerebrospinal fluid indices on admission, hospital course, morbidity rates, and mortality rates were similar for patients infected with penicillin- or ceftriaxone-susceptible versus -nonsusceptible organisms. The relatively small numbers of nonsusceptible isolates and the inclusion of vancomycin in the treatment regimen for the majority of the patients limited the power of this study to detect significant differences between the groups. Pneumococcal isolates tolerant to vancomycin have been reported in cases of meningitis associated with poor therapeutic responses. Whether the increased use of conjugate vaccines and the reduced rates of inappropriate antibiotic use will lead to decreased antibiotic resistance to the pneumococcus in the future remains to be determined.

The clinical relevance of antibiotic resistance in the treatment of pneumococcal pneumonia has recently been reviewed. This chapter updates the review and expands it to the consideration of other pneumococcal diseases such as meningitis, otitis media, sinusitis, exacerbations of chronic bronchitis, and the limited literature on other types of infection such as infections of the pleura and endocarditis. Pharmacodynamics predict that high doses of intravenous penicillin remain useful for the treatment of pneumococcal pneumonia up to MICs of 4 μg/ml. Bacteremic pneumonia caused by a resistant strain has been described following trimethoprim-sulfamethoxazole therapy in a child and an adult and following prophylaxis with this agent, suggesting that the MICs of the agent for resistant strains probably exceed the levels achievable by oral dosing. Double-tympanocentesis studies have clearly demonstrated the relevance of pharmacodynamic principles for the prediction of bacterial eradication from the middle ear. As antimicrobial penetration into the cerebrospinal fluid is limited by the blood-brain barrier, lower levels of resistance are associated with clinical failure, which has been shown to occur even with intermediately beta-lactam-resistant strains. Pharmacodynamic principles explain the clinical failures observed with the emergence of resistance in some classes of antibiotics and also explain the successful continued use of the more active drugs despite the emergence of resistance. They thus allow the development of rational guidelines for the treatment of infections caused by antibiotic-resistant pneumococci.

Penicillin resistance in Streptococcus pneumoniae has been studied in the laboratory since the early 1940s, long before it became an actual problem in clinical isolates in the late 1970s. Intraspecies gene transfer of penicillin-binding proteins (PBPs) variants between commensal streptococci and the pathogen S. pneumoniae appears to be responsible for the emergence of clinical isolates and guarantees efficient spread between clones of the pathogen. Methicillin-resistant Staphylococcus aureus organisms become completely susceptible to beta-lactams upon disruption of the fem genes (for factor essential for methicillin resistance). In view of the fact that PBP2b is encoded by an essential chromosomal gene, it may prove difficult to dilute it out of the bacterial population even without the selective pressure of beta-lactam therapy. Mosaic genes are the result of gene transfer events followed by recombination into the chromosome. In order to find genes that are potential ancestors of the mosaic blocks, pbp2x genes in closely related commensal streptococci were investigated. Examination of laboratory mutants demonstrated the complexity of resistance development with respect to PBP mutations. Two different beta-lactam antibiotics were used to select for spontaneous, independent mutant families that consisted of members with increasing resistance levels: piperacillin, a highly lytic penicillin that interacts with all PBPs at low concentrations, and cefotaxime, which does not interact with PBP2b. A non-PBP-mediated mechanism has been suggested for a high-level resistant Hungarian clone, but its molecular nature remains to be identified.

The emergence of resistance to penicillin and other beta-lactam antibiotics in pneumococci in the 1980s and 1990s led to the increased use of macrolides, fluoroquinolones, and other nonbetalactam antibiotics for pneumococcal infections. This chapter focuses on the molecular mechanisms of non-beta-lactam resistance in Streptococcus pneumoniae. Resistance to macrolides can occur by enzymatic inactivation, modification of the target by methylation or mutation, and active efflux. Proteins of the ABC transporter superfamily are organized such that two ATP-binding domains located cytoplasmically interact with two hydrophobic domains consisting of six to eight transmembrane domains. The esterase activity identified in Staphylococcus hydrolyzed 14- and 16-membered macrolides. The primary targets of other quinolones have been determined; trovafloxacin, levofloxacin, and norfloxacin target topoisomerase IV (ParC), whereas the primary target of gatifloxacin is DNA gyrase (GyrA). The use of fluoroquinolones to treat pneumococcal infections is the most important factor in the emergence of fluoroquinolone resistance in S. pneumoniae. Inducible expression allows the mRNA to be made in an inactive conformation that becomes active by the presence of an inducing macrolide. In the United States, population-based surveillance for S. pneumoniae invasive disease showed that the rapid increase in macrolide resistance was correlated with the spread of mega in pneumococci. Exciting, recent data indicate that the total burden of antimicrobial resistance can be reduced by the use of pneumococcal conjugate vaccines that reduce colonization and transmission of the vaccine serotypes.

Antibody specific for both bacterial polysaccharide and protein can be protective against extracellular bacteria, including Streptococcus pneumoniae. In this regard, polysaccharide- and protein-specific Ig isotype responses to the intact bacterium could more closely resemble those observed using conjugate vaccines consisting of covalently linked polysaccharide and protein, as opposed to purified antigens. It remains to be determined which B-cell subpopulation(s) plays the major role in IgG antipolysaccharide responses to intact extracellular bacteria, where the role may depend on cognate CD4+-T-cell help. Dendritic cells (DCs) in turn can directly stimulate activated B cells to secrete Ig in vitro and can play an active role in stimulating in vivo T-cell-independent (TI) and T-cell-dependent (TD) antipolysaccharide, as well as antiprotein, responses to intact S. pneumoniae. In vivo studies so far suggest only a limited role for endogenous NK cells in augmenting a humoral immune response, by demonstrating their ability to selectively stimulate trinitrophenyl (TNP)-specific IgG2a in response to the soluble TI antigen TNP-LPS. It is demonstrated instead that a critical balance exists between signals mediated by S. pneumoniae derived pathogen-associated molecular patterns and DC-derived interleukin-10 (IL-10) for optimization of DC induction of an in vivo humoral immune response. S. pneumoniae induces the early release of proinflammatory cytokines, the anti-inflammatory cytokine IL-10, and the type 2 cytokine IL-4. The use of intact S. pneumoniae as an immunogen to study the regulation of in vivo antipolysaccharide and antiprotein responses has revealed levels of complexity and distinct mechanistic pathways not anticipated from studies using isolated soluble antigens.

The first experimental vaccines, comprising killed whole cells, were tested in the early 1900s, albeit with inconclusive results. Polyvalent capsular polysaccharide (PS)-protein conjugates vaccines are aimed at preventing invasive diseases such as pneumonia, meningitis, and bacteremia, as well as less serious but highly prevalent infections such as otitis media. They are being targeted principally at specific groups at high risk of pneumococcal disease, particularly children under 2 years and adults over 65 years of age. The increasing prevalence of penicillin-resistant and multiply resistant pneumococci is complicating management of patients with suspected pneumococcal disease, particularly those with meningitis. Polyvalent PS-protein conjugate vaccines are very expensive to produce, and addition of further PS types or periodic reformulation to take account of altered serotype prevalence will add further to this cost. This may place the vaccine even further out of the reach of many developing countries, whose need for effective pneumococcal vaccines is greatest. Studies on the vaccine potential of pneumococcal surface protein A (PspA) have extended to human trials, and immune sera from volunteers immunized with a family 1 PspA fragment reacted with 37 different Streptococcus pneumoniae strains belonging to diverse capsular and PspA types. A further strategy under consideration for prevention of pneumococcal disease is the use of DNA vaccines. The ongoing high global morbidity and mortality associated with pneumococcal disease, and the complications caused by increasing rates of resistance to antimicrobials, have underpinned extensive efforts in recent years to develop more effective vaccination strategies against S. pneumoniae.

A consensus of the importance of the bacterial capsule for protective immunity has prevailed from very early on, based on the serotype specificity of the first whole-cell vaccines and soon after that on the protective immunity observed after vaccination with the purified polysaccharide. The evidence of serotype-specific protection against bacteremia in immunocompetent adults, starting approximately 2 weeks after a single dose of the polysaccharide vaccine, is unequivocal based on both clinical trials before vaccine licensure and the subsequent experience with the wide use of the polysaccharide vaccine. Pneumococcal pneumonia was believed to be a major cause of illness and even death in these populations. However, a vaccine effect could not be seen in either of the trials with respect to overall or serotype-specific pneumococcal pneumonia or pneumonia-associated deaths. In view of the importance of pneumonia for children in developing countries, it is regrettable that further studies to confirm or refute these findings have not been conducted in the face of a lack of funding and an expectation of an improved vaccine. In recent studies in Finland almost every child has been shown to have had at least one episode, and many several, of otitis media before the age of 2 years, with pneumococci present in the purulent middle ear fluid in 30% of these cases. The major study to test the protective efficacy of the 7-valent conjugate vaccine was carried out in California, within the framework of the large health insurance plan of Kaiser Permanente.