Design and Pre-Clinical Evaluation of a Universal HIV-1 Vaccine
Figure 1
(A) Localization of the 14 most highly conserved regions of the HIV-1 proteome. The numbers written vertically under each fragment boundary indicate the first and last aa positions using the HXB2 reference strain numbering (http://www.hiv.lanl.gov/content/hiv-db/LOCATE/locate.html). (B) Predicted aa sequence of the the HIVCONSV immunogen with indicated fragment numbers. (C) Summary of the fragments including: the fragment number; the protein in which it was embedded; the clade of the consensus sequence selected for inclusion in the immunogen, alternating between clades A–D; additional clades that have identical HIVCONSV; and position numbers in the chimeric vaccine. The number of additional clades with identical consensus sequences to selected clade reflects the high level of conservation in these regions, and is encouraging in terms of the global potential of the vaccine. The consensus sequences compared were to the M group consensus, clades A–K, and three very common recombinant circulating forms CRF01 (common in Asia and Africa), CRF02 (common Africa), CRF08 (common in China) retrieved from the Los Alamos database 2004 consensus alignment (http://www.hiv.lanl.gov/content/hiv-db/CONSENSUS/M_GROUP/Consensus.html). (D) Schematic representation of the HIVCONSV immunogen (not drawn to scale) indicating clade anternation (above), overlapping peptide pool derivation and protein origin by colour coding. (E) Hamming distances between the HIVCONSV antigen fragments and the global circulating viral sequences. The full M group alignment, including recombinant sequences, was used for the comparison. The Los Alamos database alignment contains only one sequence person, and contains sequences from between 600 and 1000 individuals in these proteins. The Hamming distance range for 95% of the sequences relative to the vaccine immunogen is given by the vertical lines. The distances between the full length natural proteins were then calculated relative to HXB2 reference strain Env, Vif, Gag and Pol sequences for comparison. Distance measures are minimal estimates, as gaps inserted in regions where insertions and deletions occur were not counted. (F) Numbers of known CD8+ T cell epitopes (defined to within 12 aa or less in the Los Alamos HIV-1 database) in each of the 14 conserved protein fragments included in the HIVCONSV immunogen are shown. When more than one HLA class I presenting molecules can present the same HIV-1 epitope, then each is counted as a distinct epitope; if more than one sequence variant has been described as an epitope presented by the same class I molecule, then these are counted as a distinct epitopes; however, if an HLA serotype and genotype that are potentially the same are each described as presenting the same epitope (like A2 and A*0201) they are scored as a single epitope.