Philip Goulder

It is now established that cytotoxic T lymphocytes play a central role in the control of AIDS virus infections. It is becoming increasingly evident, however, that qualitative differences exist between CTL of different specificity. Most clearly this has been demonstrated in the LCMV model in mice, but the association of particular MHC class I alleles with slow (B27, B51 and B57) and rapid progression (B8 and B35) in HIV infection suggests similar differences relate to control of HIV infection also. The work described uses an approach based on the presence or absence of CTL escape mutation to help understand whether particular CTL responses are effective. It is argued that the demonstration of CTL escape is evidence for epitope-specific selection pressure mediated by CTL on the virus. The speed with which escape occurs may depend upon several factors, including purifying selection opposing amino acid change, the timing of the appearance of the response, the effectiveness of the individual CTL response, and the effectiveness of the total anti-viral immune response. It is argued that high-frequency CTL responses from which escape has not occurred that are coincident with high viral loads cannot be effective. CTL responses from which escape has not occurred that are coincident with low viral loads may either be a consequence of effective CTL, or may be the result of effective anti-viral immunity unrelated to a particular individual CTL specificity. Examples of effective and ineffective HIV-specific CTL responses are described.