Skip to main content

Improving vaccination in early life

Deliverables

We identified defects in newborn myeloid dendritic cells that might be relevant to the impairment of immune responses in early life. Indeed, neonatal dendritic cells were found to be deficient in maturation upon exposure to microbial products such as bacterial lipopolysaccharide or Bordetella pertussis toxin. Furthermore, they were found to produce much less interleukin-12 than adult dendritic cells. This defect was shown to be caused by a decreased transcription of the gene encoding the p35 chain of the cytokine. Further experiments established that chromatin remodelling unmasking a critical binding site for the Sp1 transcription factor in the promoter of the interleukin-12 p35 gene was deficient in neonatal dendritic cells. The defect in interleukin-12 production seems long lasting since it was still present in 12 year-old infants. Data of co-culture experiments with T cells indicate that this dendritic cell defect contributes to the impaired Th1 responses in newborns. Interestingly, the addition of recombinant interferon-gamma to neonatal dendritic cells restored their capacity to produce bioactive interleukin-12. In parallel, immature dendritic cells were shown to internalise pneumococcal polysaccharide that progress to late endosomal/lysosomal compartments after uptake. Defects in neonatal plasmacytoid cell responses to CpG oligodeoxynucleotides were also identified with a profound impairment in their capacity to produce interferon-alpha. These data contrast with those obtained in neonatal mice in which splenic dendritic cells were found to produce adult levels of interleukin-12 upon stimulation and to elicit efficient T cell responses. Taken together, these data establish major differences between dendritic cells of human and mouse newborns and identify defects which are relevant to the increased susceptibility to microbial infections in early life and to the development of new immunization strategies in the neonatal period.
Prior to testing in vivo in humans, there are no means of determining human immunological efficacy to a novel vaccine preparation, and all anticipated immunological responses are based on prior animal experiments. There is clearly a need to develop ex-vivo techniques or model systems, which could provide insights into the nature of the human immune response to novel vaccine preparations prior to their administration to humans. Over the period of this programme we have established and validated an ex-vivo assay system, based on combining antigen pulsed neonatal dendritic cells and autologous naïve CD4 T-cells, that suggests that human neonatal CD4 T cells have competent primary functional responses to both 'model' and vaccine antigens, and can develop functional antigen specific effector memory - with the proviso that these cells have been primed by mature monocyte derived DCs. We have employed two antigen systems; KLH, a prototypic antigen for priming a naïve T-cell response, and BBG2Na, a recombinant human RSV candidate vaccine. The nature of the primary neonatal T cell response is dependent on the effect of the antigenic preparation on the stimulatory capacity of the DCs. Overall, our data suggests that the capability of human neonatal CD4 T cells to respond to antigenic stimulation has been previously underestimated. Neonatal CD4 T cells are ultimately naïve and thus have an absolute requirement for stimulation by mature DCs. We conclude that if safe and effective DC-maturing adjuvant were developed, then the prospects for a successful outcome of neonatal vaccination would be greatly enhanced.
We have previously developed an antigen delivery system based on recombinant parvovirus-like particles (PPV-VLPs) formed by self-assemblage of a capsid protein of porcine parvovirus (PPV), carrying a foreign sequence at its N terminus. These pseudo-particles are produced by insertion of exogenous peptides into VP2 protein (67 KDa), which is one of three structural viral proteins of porcine parvovirus. VP2 self-assembles into 25 nm pseudo viral particles. After immunization of adult mice with PPV-VLPs carrying a CD8+ T cell epitope of the lymphocytic choriomeningitis virus nucleoprotein (PPV-VLPs-LCMV) or ovalbumin (PPV-VLPs-OVA) without adjuvant, mice developed a strong cytotoxic T lymphocyte (CTL) response against peptide-coated or virus-infected target cells. Recombinant PPV-VLPs induce a long-lasting memory CTL response which can be evidenced even 9 months after the last immunization. Besides the i.p. and i.v. routes, mucosal immunization with PPV-VLPs was also demonstrated to be efficient. Indeed, intranasal immunization of mice with PPV-VLPs induces seric as well as mucosal neutralizing IgG and IgA antibodies specific for PPV-VLPs. Intranasal immunization also induces a CTL response against LCMV peptide-coated-target cells. The CTL response induced by PPV-VLPs-OVA was shown to be B and CD4+ T cell independent, as demonstrated using knock out mice. The long-term CTL response obtained by immunization with PPV-VLPs-LCMV was shown to be protective. Indeed, mice immunized with PPV-VLPs-LCMV survived to an intra-cerebral injection of LCMV that normally killed non-immunized mice in 7 days. This protection was mediated by CD8+ T cells, as demonstrated by in vivo depletion studies and is due to the induction of high-avidity and high frequency CTL responses. One main result of Neovac concerns the demonstration of the ability of this non-replicative delivery system based on parvovirus-like particles to induce CTL responses in the neonatal period. A single immunization of 1-week-old BALB/c mice with recombinant VLP carrying a CD8+ T cell determinant from lymphocytic choriomeningitis virus (VLP-LCMV) induced antigen-specific CD8+ cytotoxic T cells that were similar to those elicited by adult immunization as assessed by cytotoxic activity, IFN-g secretion, cytotoxic precursor cell frequencies, avidity for antigen and protective activity against viral challenge. These CTL responses are elicited within 2 weeks of a single immunization, in the absence of any adjuvant and independently of the presence and help of CD4+ T cells, highlighting the potential of VLP as vaccine candidates in early life.
The development of infant and neonatal mouse models indicated that neonatal exposure to DTPw or DTPa did not induce neontal tolerance but in contrast primed for subsequent protective responses. This demonstration paves the way for clinical trial assessing whether a birth dose of pertussis vaccine could reduce the window of susceptibility to pertussis. The immune responses induced after infection of very young children or after vaccination with DTPw is qualitatively very different from that induced by DTPa vaccination, essentially with respect to the cytokine profiles, including towards tetanus toxoid or polyclonal stimulation. This may have an impact on the choice of vaccines to be used in early childhood. Alternative, novel vaccine approaches either by the use of live attenuated B. pertussis or of intranasal administration of vaccines or purified antigens in the presence of mucosal adjuvants resulted in full protection against challenge in mouse models. This may open new avenues for pertussis vaccination, in order to mimic more closely the natural route of infection and induction of protective immunity.
We identified defects in newborn myeloid dendritic cells that might be relevant to the impairment of immune responses in early life. Indeed, neonatal dendritic cells were found to be deficient in maturation upon exposure to microbial products such as bacterial lipopolysaccharide or Bordetella pertussis toxin. Furthermore, they were found to produce much less interleukin-12 than adult dendritic cells. This defect was shown to be caused by a decreased transcription of the gene encoding the p35 chain of the cytokine. Further experiments established that chromatin remodelling unmasking a critical binding site for the Sp1 transcription factor in the promoter of the interleukin-12 p35 gene was deficient in neonatal dendritic cells. The defect in interleukin-12 production seems long lasting since it was still present in 12 year-old infants. Data of co-culture experiments with T cells indicate that this dendritic cell defect contributes to the impaired Th1 responses in newborns. Interestingly, the addition of recombinant interferon-gamma to neonatal dendritic cells restored their capacity to produce bioactive interleukin-12. In parallel, immature dendritic cells were shown to internalise pneumococcal polysaccharide that progress to late endosomal/lysosomal compartments after uptake. Defects in neonatal plasmacytoid cell responses to CpG oligodeoxynucleotides were also identified with a profound impairment in their capacity to produce interferon-alpha. These data contrast with those obtained in neonatal mice in which splenic dendritic cells were found to produce adult levels of interleukin-12 upon stimulation and to elicit efficient T cell responses. Taken together, these data establish major differences between dendritic cells of human and mouse newborns and identify defects which are relevant to the increased susceptibility to microbial infections in early life and to the development of new immunization strategies in the neonatal period.