The inhibition of seroconversion after immunization in the presence of maternal antibodies is a well-documented phenomenon. The lack of generation of antibody suggests an impairment of B cell activation and development, although this was never proven. Our data demonstrate clearly that the inhibition in the generation of antibody correlates with the inhibition of B cell development, and particularly with early activation of B cells in draining lymph nodes. We have shown previously that the inhibitory action of maternal IgG is mediated through a complex of MeV vaccine/IgG cross-linking the B cell receptor (BCR) with the inhibitory receptor CD32 
. This inhibition could be partially overcome by MeV-specific IgM which crosslinks complement receptor 2/CD21 with BCR through a complex of MeV vaccine/IgM/complement protein C3d. It could also be partially overcome with viral vector systems which provided MeV hemagglutinin as antigen and induced type I interferon which is known to act through IFN-R 
. This is consistent with data demonstrating that the induction of type I interferon supports B cell differentiation 
and antibody secretion from B cells 
. Based on our data, the strong stimulatory activity of type I interferon on B cell responses seems to be due to the dual use of both CD21 and interferon receptor as functional receptors on B cells. The critical role of CD21 in the B cell response has been shown in Cr2−/− mice that are deficient in CR2 (
CD21). Cr2−/− mice lack CD21 on both B cells and follicular dendritic cells 
. They demonstrate substantial defects in antigen-specific, T cell-dependent and T cell-independent humoral immune responses 
. In addition, defects in B cell memory 
and the development of the natural antibody repertoire 
are found in Cr2−/− mice. The natural ligand of CD21 is the complement protein C3d, and the binding of C3d to CD21 stimulates B cell responses. Interferon alpha contains a peptide sequence similar to one in C3d which is located at the C3d-CR2 binding site 
and interferon alpha binds to CD21 with an affinity comparable to the natural ligand C3d. In vitro blockage of CD21 by antibody on human peripheral blood B cells diminishes the expression of interferon inducible genes 
. In an ELISPOT assay, antibody secretion from B cells by type I interferon was clearly reduced when CD21 was blocked both on cotton rat and mouse B cells. Importantly, B cells from mice with a gene deletion in the interferon receptor respond to IFN α and this response can be blocked by antibodies against CD21. These data clearly demonstrate that CD21 is a functional receptor for IFN α on B cells. It seems likely that activation of the B cell through CD21 and IFN-R by type I interferon counteracts the inhibitory signal induced by CD32 through MeV-specific IgG and leads to a net stimulatory signal.
Previously, we have shown that the provision of type I interferon by immunization with NDV-H was able to partially restore the neutralizing antibody response in the presence of maternal antibodies. NDV interacts with TLR-7 and the RIG-I pathway which leads to the induction of type I interferon in vivo
. TLRs use either the MyD88 (TLR-2, 4, 5, 7, 8, 9) or TRIF dependent signaling pathways (TLR-3, 4) (reviewed in 
). It has been suggested that the expression of different TLR ligands by pathogens might enhance immune responses by signaling via both adapter pathways 
. Here, we aimed to achieve the highest level of type I interferon by activating TLR-3 and TLR-9 through their agonists. We were able to show that a TLR-3 agonist (poly I:C) in combination with a TLR-9 agonist (ODN 2216) induced synergistically higher levels of type I interferon than either ligand alone. A combination of TLR agonists does not necessarily have a synergistic effect. Ghosh et al
compared cytokine responses of all the possible combinations of known TLR ligands in human PBMCs 
. TLR-9 agonist ODN 2216 produced high levels of interferon alpha but type I interferon induction through TLR-9 was downregulated in combination with TLR-7 or TLR-8 agonists. In contrast, TLR-3 induced significant amounts of type I interferon when used in a combination with agonists to TLR-2, 5, 7/8 
which support our data that co-stimulation of TLRs which are TRIF dependent (TLR-3) and MyD88 dependent (TLR-9) has a synergistic effect on type I interferon induction.
One possible application of our data would be the direct inoculation of IFN α with MeV vaccine in order to stimulate a better B cell and antibody response. In mice, the inoculation of a high dose of IFN α and influenza vaccine induced good T cell and B cell responses 
, but in humans this approach was not successful 
. A possible reason is the difference in the effect of IFN α on the immune system of mice and humans. In humans, IFN α drives TH1 development and acts through STAT-4 which is not the case in mice 
. The conclusion from these studies was that the mouse is not an informative animal model to study adjuvants which are targeted for use in the human respiratory system 
. In cotton rats, inoculation of MeV vaccine and cotton rat IFN α intranasally did also not result in increased B cell responses (data not shown). The failure to stimulate the B cell response might be related to technical problems. Although doses comparable to the amounts found in lung lavage after TLR agonist application were used, it is possible that higher doses of IFN α are needed, or that pegylated interferon which is more stable in vivo is required. Alternatively, the cotton rat might be an animal model better suited to test intranasal adjuvants for humans 
. This notion might be supported by the fact that in cotton rats a TLR-9 agonist optimized for human cells was most effective (ODN2216, data not shown), whereas in mice TLR-9 agonists optimized for mouse cells have to be used.
In summary, we have demonstrated that a combination of TLR-3 and TLR-9 agonists induces higher levels of type I interferon than either agonist alone. Subsequently, IFN α utilizes both IFN-R and CD21 as functional receptors for B cell stimulation and leads to restoration of B cell responses after immunization in the presence of inhibitory MeV-specific IgG.