Sirs

In the July issue of Nature Reviews Immunology, Adkins et al.1 provided an excellent review of neonatal immunity and proposed an attractive hypothesis about the possible role of the hyporesponsiveness of neonates. They suggested that the reduced responses of neonates might prevent unwanted hyperinflammation that could be injurious to the tissues of the developing organism. The T-helper-2-cell bias of the neonatal T-cell response is certainly consistent with this idea. In addition, we have recently reported that, compared with macrophages from adults, macrophages from the neonatal spleen are defective in production of the pro-inflammatory cytokines tumour-necrosis factor, interleukin-1 (IL-1), IL-6 and IL-12 when stimulated with lipopolysaccharide (LPS)2. Moreover, in response to LPS stimulation, the neonatal macrophages produced more IL-10 than adult macrophages, which is consistent with the hypothesis of Adkins et al. that neonates favour an anti-inflammatory environment. The increased production of IL-10 is causally related to the defect in pro-inflammatory cytokine production, because IL-10-deficient neonatal macrophages produced levels of IL-1 and IL-6 that were similar to those of adult macrophages. The dysregulated production of cytokines by neonatal macrophages seemed to result from their reduced cell-surface expression of Toll-like receptor 2 (TLR2) and TLR4 (Ref. 2). Similar dysregulated production of cytokines was also observed when neonatal macrophages were stimulated with fixed Streptococcus pneumoniae bacteria (S.B. and R.L.C., unpublished observations).

Adkins et al. also proposed that the unresponsiveness of neonates to polysaccharide antigens — that is, T-cell-independent type 2 (TI-2) antigens — could be a consequence of the reduced numbers of marginal-zone B cells in neonates, as well as the reduced expression of CD21 (also known as complement receptor 2) by neonatal B cells and the relatively low levels of complement component C3 in neonatal serum1. Moreover, the authors comment that “manipulations (of antigen dose or adjuvant) that might improve other responses are unlikely to be effective for this group of antigens”1. We have previously shown that intrinsic defects in neonatal B-cell responses to polysaccharide antigens can be overcome, because they could be induced to respond to the model T-cell-independent antigen trinitrophenyl (TNP)-Ficoll in the presence of IL-1 and IL-6 (Ref. 3), two of the cytokines that were produced at reduced levels by neonatal macrophages. Moreover, we have shown that production of pro-inflammatory cytokines, and antibody responses to T-cell-independent antigens, can be partially restored by the addition of IL-10-specific antibody to cultures of neonatal splenocytes2. On the basis of these findings, we propose that neonatal unresponsiveness to polysaccharides is mainly caused by a deficiency in the production of the required cytokines by macrophages. Accordingly, we and others have shown that CpG-containing oligodeoxynucleotides — which can activate macrophages, B cells and dendritic cells — augment humoral responses in neonates4,5,6,7,8.

Although polysaccharide–protein conjugates have been extremely successful as vaccines against infection with Haemophilus influenzae or seven serotypes of S. pneumoniae, this approach is not yet applicable for all of the 23 commonly occurring serotypes of S. pneumoniae9,10. We propose that future research to improve these vaccines be directed at finding agents that can increase macrophage secretion of IL-1 and IL-6, which would then allow efficient activation of neonatal B cells.