Analysis of PBMC for the in vivo persistence of neo-transduced CD8⁺ gag-specific T cells. (a) Limiting dilution PCR and ³²P liquid hybridization strategy for neo DNA in PBMC lysates. (i) The titration curve in an HIV-seronegative (uninfused) individual. All lanes are negative. (ii) Illustration of the dilution curve containing DNA extracted from 0 to 5,000 cloned CD8⁺ T cells transduced with one copy of the neo gene. The cells were diluted into PBMC of an untreated donor to equalize the cellular DNA in all wells. The assay is capable of detecting as few as 5 neo copies (5 positive cells) per microgram of PBMC DNA. Independent determinations of the frequency of neo-modified cells varied up to 0.5 log. (iii) Serial dilutions of PBMC from a patient who received the second infusion of neo-modified T cells (3.3 × 10⁹ cells/m²) 24 hours earlier (patient 1). Endpoint dilution analysis indicates a titer of 2,558 infused neo-positive cells/10⁶ PBMC (1 in 391 PBMC). Negative controls consisted of DNA extracted from PBMC before CTL infusions. All samples were positive for the β-hemoglobin gene. (b) Using a single-round PCR and liquid hybridization strategy, the in vivo persistence of neo-marked T cells was assessed in peripheral blood collected before, during, and after adoptive transfer. The absolute number of CD8⁺ T cells in PBMC was determined by flow cytometry for each time point. Data points indicate mean ± SEM and represent the number of neo copies per microgram of CD8⁺ T-cell DNA. Arrows indicate the day of neo-marked T-cell infusions. All samples were run in triplicate, and all assays were repeated at least twice. (c) Comparison of PCR/liquid hybridization strategy (described above) with nested PCR in clinical samples, as described in Methods.