British Medical Bulletin 2004; 70: 1–14 British Medical Bulletin, Vol. 70 © The British Council 2004; all rights reservedDOI: 10.1093/bmb/ldh021Bacterial polysaccharide–protein conjugate vaccinesAdam FinnInstitute of Child Health, Department of Clinical Sciences South Bristol, University of Bristol, Bristol, UKFollowing demonstration that chemical conjugation of polysaccharide antigens to proteins could enhance their immunogenicity in the 1920s, interest in this approach to primary prevention of bacterial infections waned with the development and widespread use of antibiotics. Emergence of resistant bacteria rekindled interest in the late 20th century, which saw extremely rapid development and implementation of several vaccines which are already rapidly changing the epidemiology of childhood infections with Haemophilus influenzae type b, Streptococcus pneumoniae and Neisseria meningitidis. Others such as Group B streptococcus and Salmonella typhi infections may soon follow. However, several important questions about the immunology of these antigens remain unanswered and the long-term implications of reducing or eliminating the circulation of organisms which are more commonly nasopharyngeal commensals than pathogenic invaders are uncertain.IntroductionYoung children are prone to infections. There are now abundant datathat this reflects not only their immunological naïvety, but also a degreeof immunoincompetence relative to older children and adults. No doubt,to some extent structural immaturity, such as fragility of integument andmucosae, also contributes. There is a large published literature comparing,in infants and older individuals, the size or function of almost every elementof immunity, cellular and humoral, innate and adaptive; studies whichoften demonstrate the former to be the weaker1. It is hard to know whatto make of much of this observational information, but in some instancesthe implications are clear. Adaptive cell mediated immune responses to awide variety of foreign microbial antigens in newborns are both weakerand slower than in older children. Since all the necessary components forfunctional responses appear to be in place well before term2, perhaps thisCorrespondence to:Adam Finn,Professor of Paediatrics,Institute of Child Health,Level 6,UBHT Education Centre,Upper Maudlin Street,Bristol BS2 8AE, UK.E-mail:[email protected] Published online August 31, 2004Adam Finn2 British Medical Bulletin 2004;70incapacity reflects a post natal lag period in immunological regulation,still set to tolerate foreign (maternal) antigens while living in the (norm-ally) sterile environment of the uterus.If this supposition is true, it remains puzzling that a more rapid postnatal immunological ‘reboot’ has not evolved, as this relative refractori-ness often appears to persist for months, at least for some aspects ofspecific antimicrobial immunity. Interference by transplacentally acquiredIgG—a vital element of immune protection against infection duringearly life, which disappears with a half-life of approximately 28 daysafter delivery—can to some extent account for the persistent suppressionof antibody responses—particularly to protein antigens—but does notcause the down regulation of T cell function3. Nor does maternal anti-body explain the profound and long lasting failure of infants to makeantibodies to the polysaccharide (PS) capsular antigens which decoratemany bacterial pathogens4, a failure which is presumably at least partlyresponsible for the high incidence of invasive infections due to theseorganisms in this age group.Antibody responses to PS antigens are the subject of considerablemodern myth. Since they are not peptides, PSs should not, by definition,be processed and presented by MHC class II antigens, so that the recruit-ment of T cell help by this route for B cell function is not expected. Never-theless, isotype switching does occur—not only IgM but also abundantIgG and particularly IgA responses to these antigens are seen5,6. It appearsthat a distinct signalling pathway may regulate this important antibodyresponse pathway7 whose size and character vary not only betweenindividuals8 and with age, but also with previous exposure9 and betweenantigens.Complete and permanent tolerance to some bacterial capsular PS anti-gens is the rule. For example, Lancefield Group A streptococci (Strepto-coccus pyogenes) colonizing the nasopharynx, an important cause ofboth mucosal and invasive disease in children, are consistently encap-sulated, although to a variable extent. However, the PS in question ishyaluronic acid—an antigen against which antibody responses would beunwanted as it is a major component of human connective tissues10. Thecapsule of Neisseria meningitidis group B is likewise poorly immuno-genic as a vaccine11, an observation which may be explained, at least inpart, by the structural similarity of the capsular PS with polysialic acidmoieties which decorate components of mammalian tissues includingthe central nervous system12.Nevertheless, the PS capsular antigens of many other pathogenic bacteriado induce substantial protective serum antibody responses when used asvaccines. However, for most of these antigens, this is not true in youngchildren. Although the age of 2 years is often cited as that at which suchresponses start to be seen, in fact the doses of these antigens needed toConjugate vaccinesBritish Medical Bulletin 2004;70 3elicit protective responses and the average age at which such responsesbecome active vary quite substantially and predictably between antigens.For example pneumococcus type 3 capsular PS is relatively immunogenicin infants whereas types 6A and 6B are extremely poor immunogens,with other types ranging in between13.One is left wondering exactly why this is so. Presumably it is no acci-dent. This implies some relative survival advantage of this temporaryanergy, perhaps less marked than for the fully ‘invisible’ antigensdescribed above, but operative nonetheless. On the other hand, it is alsopossible that the encapsulated bacteria that infect human children todaymay not have been around long enough in evolutionary time for thesurvival advantage of vigorous early immune responses to their capsulesyet to have taken effect.In any case, faced with the real problem of the serious morbidity ofthese infections in young children and evidence that serum antibodiesto capsular antigens protect, such theoretical considerations