The wide variation in local immunoglobulin and antibody levels for any individual animal may have been due to the effects of the menstrual cycle as reported in macaques and women [41], [42] and [38]; however, the present study was not powered to analyse this variable. An effective vaccine will require not only sustained antibody production into mucosal fluids but the antibodies DZNeP chemical structure will need to have potent and broad virus neutralising activity. It is known that monomeric gp120 generally fails to elicit such activity [43], [44], [45] and [46] and for this reason we used a trimeric envelope immunogen, gp140, that has demonstrated remarkable stability in vitro (D. Katinger, personal learn more communication)

and is therefore more likely to mimic the native virion envelope spike [2]. Although cross-clade neutralising activity was restricted to MW965.26 and clade B SF162.LS envelope-bearing pseudoviruses and disappointingly no activity was seen against any of a broad range of clade C envelopes, this study has shown that this narrow specificity is not exclusively due to formulation of the immunogen in Carbopol and/or the vaginal route of administration,

as similar results were obtained after intramuscular immunisation in the presence of AS01 adjuvant. Moreover, as in rabbits [21], serum antibodies did not recognise the highly immunogenic gp41-ED residues 598–597 [47] (data not shown), suggesting that the gp41 region of the molecule may be occluded possibly because of the lack of membrane anchoring. Interestingly macaques have been protected against vaginal challenge with SHIVSF162 following systemic or nasal/systemic immunisation with HIV-1SF162 ΔV2 gp140 and protection was associated with serum neutralising antibody [48]. Although the restricted serum neutralising activity Immune system obtained is of questionable

relevance for a protective HIV-1 vaccine it is interesting that the correlation between anti-gp140 IgG binding antibody titre and neutralising activity seen in animals that were primed intramuscularly did not hold true for animals primed intravaginally. This observation suggests factors other than antibody titre alone may be important, including antibody subclass, avidity and fine specificity. Furthermore, we were unable to measure neutralising activity in mucosal fluids and there is a clear need for the development of micro-neutralisation assays that can be used with small volumes of biological fluid. The results obtained here inform the design of our next clinical trial that will run in parallel with a “paraclinical” macaque study that will include envelope-SHIV challenge. Through this iterative process it will be possible to cross-validate the macaque model – essential for the identification of correlates of protective immunity.