| Abstract: | Expression of HLA-B57 is associated with restricted replication of human immunodeficiency virus (HIV), but the mechanism for its protective effect remains unknown. If this advantage depends upon CD8 T-cell recognition of B57-restricted epitopes, mother-to-child transmission of escape mutations within these epitopes could nullify its protective effect. However, if the B57 advantage is largely mediated by selection for fitness-attenuating viral mutations within B57-restricted epitopes, such as T242N in TW10-Gag, then the transmission of such mutations could facilitate viral control in the haploidentical infant. We assessed the consequences of B57-associated mutations on replication capacity, viral control, and clinical outcome after vertical transmission in 13 mother-child pairs. We found that expression of HLA-B57 was associated with exceptional control of HIV during infancy, even when mutations within TW10 and most other B57-restricted epitopes were transmitted, subverting the natural immunodominance of HLA-B57. In contrast, most B57-negative infants born to B57-positive mothers progressed rapidly to AIDS. The presence of T242N led to a reproducible reduction in viral fitness, as demonstrated by in vitro assays using NL4-3 constructs encoding p24 sequences from individual mothers and infants. Associated compensatory mutations within p24-Gag were observed to reverse this impairment and to influence the propensity of T242N to revert after transmission to B57-negative hosts. Moreover, primary failure to control viremia was observed in one infant to whom multiple compensatory mutations were transmitted along with T242N. These parallel in vivo and in vitro data suggest that HLA-B57 confers its advantage primarily by driving and maintaining a fitness-attenuating mutation in p24-Gag.Expression of HLA-B57 is associated with low levels of viremia and prolonged survival after human immunodeficiency virus (HIV) infection, but the mechanism underlying this association is not well understood. Because the principal role of class I HLA molecules is to bind viral peptides for presentation to cytotoxic CD8+ T cells, numerous studies have sought to identify quantitative or qualitative features of the B57-restricted CD8 T-cell response to HIV that correlate with effective viral containment. However, despite exhaustive studies of the HIV-specific CD8 response in B57-positive individuals, few differences in the frequency, function, or epitope specificity of CD8 cells have been identified that reliably distinguish B57-positive controllers from the many B57-positive individuals with progressive HIV infection (1, 28-30). Early in the course of HIV infection, CD8 cells consistently select for a T242N substitution in the B57-restricted epitope TW10, which lies in a highly conserved region of p24-Gag capsid protein. Although this mutation permits the virus to escape from CD8 T-cell surveillance, it incurs a cost to viral replicative capacity (5, 24, 26). Thus, it has been hypothesized that the B57 advantage may be an indirect effect of CD8 response, whereby viral control does not rely on active surveillance by B57-restricted CD8 T cells but instead upon an “imprint” of the B57-restricted CD8 response that forces the viral capsid protein into a less fit state. Recent studies have suggested that such escape mutations may even result in lower HIV RNA levels in the next host to whom the virus is transmitted (9, 18). Selection for T242N is frequently followed by the emergence of upstream mutations at residues H219, I223, and M228 in the cyclophilin A (CypA) binding loop that have been shown to partially compensate for its fitness defect (5, 24, 26, 32), which may explain why many B57-positive individuals ultimately develop high viral loads despite the presence of T242N. The timing with which these compensatory mutations emerge and accumulate varies among B57-positive individuals, possibly due to constraints imposed by interacting Gag residues (7), and this variability may contribute to the observed heterogeneity in clinical outcomes.Mother-to-child transmission of HIV affords a unique model for differentiating between the impact of CD8 T cells and the viral escape mutations that they induce. Because transmission occurs between a haploidentical mother-child pair, it is possible to assess the impact of CD8 T-cell responses to, and mutations within, epitopes restricted by shared and unshared HLA molecules. Each infant expresses at least three HLA molecules shared with his or her mother, and vertical transmission of maternally selected escape mutations within epitopes restricted by these shared HLA molecules may effectively eliminate these potential cytotoxic-T-lymphocyte (CTL) targets from the infecting viral inoculum. However, infants also express paternally inherited HLA molecules for which no viral “imprint” exists. If the HLA-B57 advantage depends upon active surveillance by CD8 cells, mother-to-child transmission of B57-associated escape mutations may nullify this benefit among infants who share the B57 allele with their mothers, as has been demonstrated in the case of another protective HLA class I allele B*27, where transmission of a mutation within the highly dominant KK10-Gag epitope leads to failure of viral control in B27-positive infants (20). On the other hand, if viral attenuation is a major contributor to the B57 advantage, then the advantage associated with HLA-B57 expression would likely be passed along to the child, who would benefit from the reduced replication capacity of the transmitted virus.We assessed HIV sequence evolution and clinical outcomes in 13 HIV-infected mother-child pairs, one or both of whom expressed HLA-B*B57, in order to determine the impact of viral sequence changes that occur when B57-mediated immune pressure is introduced, or removed, after transmission. |
