Neutralizing antibodies are mainly against conformational epitopes

on virus surface, and are usually type specific; while non-neutralizing antibodies are mostly against linear epitopes on virus surface, and some of them have broad cross-reactivity [37], [38], [39], [40], [41], [42], [43], [44] and [45], even between distantly related types such as HPV 16 and 18 [35]. This kind of non-neutralizing cross-reactivity would provide some portion of positive signals in ELISA when detecting sera from multivalent immunized groups [46]. This might give an explanation of the difference between ELISA and neutralizing assay. Neutralizing antibody titer detection is discontinuous and gaps between detecting points increase with sera dilutions. On the contrary, percent infection inhibition at a certain dilution is a continuous GDC-0449 parameter, which provides a more detailed result when comparing two groups at a proper dilution.

In our results, percent infection inhibition and neutralizing antibody titer reflected almost the same trend: multiple VLPs co-immunization could elicit high level of neutralizing antibodies, but the neutralizing antibody levels or percent infection inhibition of trivalent groups were lower than those of corresponding monovalent PD173074 mw groups. A clinical study from Garland and Steben showed that HPV 16/18/6/11 quadrivalent vaccine and HPV 16 monovalent vaccine could induce same level of anti-HPV 16 antibodies [47]. Since the vaccines they used were formulated with relatively low dose of VLPs and were adjuvanted with Aluminium salts, these results were in accordance with our observation in adjuvant experiments. In another study, Gasparic et al. co-immunized

different types of Papillomavirus (PV) L1 DNA vaccines in mice, and observed interference between types, however, the interference they observed was due to differences of expression level [48]. In our study, VLPs were used as antigens from and influences at expression level could be ruled out, so the interference we observed indeed occurred after antigens contacted with immune system. Immune interference has been reported in many other vaccines. A lot of studies in co-immunization revealed that immune interference could happen in both antigen specific T cell responses and B cell responses [20], [21], [22], [23], [24], [25], [26], [27], [29], [46], [49], [50], [51], [52], [53], [54] and [55]. Immune interference could occur between different variants of homologous epitopes [24], [26] and [27]; and it could also happen when heterogenous antigens were immunized together [25] and [54]. The mechanism of immune interference is unclear yet. Different antigens may be interfered at different degree. A study on co-immunization of recombinant hepatitis B surface antigen (HBsAg) and inactivated hepatitis A virus (HAV) suggested that a stronger immunostimulant might be interfered less [25].