The mechanisms through which cellular activation results in intracellular mycobacterial killing is only partially understood. However, in vitro studies of human immunity to Mycobacterium tuberculosis have been largely modeled on the work reported by Crowle, which is complicated by several factors. The whole blood culture is simple and allows the simultaneous analysis of the relationship between bacterial killing and the effect of effector cells and humoral factors. In this study, we attempted to determine the extent to which M. tuberculosis is killed in a human whole blood culture and to explore the role of the host and microbial factor in this process.

The PPD positive subjects were compared to the umbilical cord blood and patients with tuberculosis, diabetes and lung cancer. The culture is performed using heparinized whole blood diluted with a culture medium and infected with a low number of M. avium or M. tuberculosis H37Ra for 4 days by rotating the culture in a 37℃, 5% CO₂ incubator. In some experiments, methylprednisolone- or pentoxifylline were used to inhibit the immune response. To assess the role of the T-cell subsets, CD4+, CD8+ T-cells or both were removed from the blood using magnetic beads. The Δ log killing ratio was defined using a CFU assay as the difference in the log number of viable organisms in the completed culture compared to the inoculum.

1. A trend was noted toward the improved killing of mycobacteria in PPD+subjects comparing to the umbilical cord blood but there was no specific difference in the patients with tuberculosis, diabetes and lung cancer. 2. Methylprednisolone and pentoxifylline adversely affected the killing in the PPD+subjects, umbilical cord blood and patients with tuberculosis. 3. The deletion of CD4+ or CD8+ T-lymphocytes adversely affected the killing of M. avium and M. tuberculosis H37Ra by PPD+ subjects. Deletion of both cell types had an additive effect, particularly in M. tuberculosis H37Ra. 4. A significantly improved mycobacterial killing was noted after chemotherapy in patients with tuberculosis and the Δ logKR continuously decreased in a 3 and 4 days of whole blood culture.

The in vitro bactericidal assay by human whole blood culture model was settled using a CFU assay. However, the host immunity to M. tuberculosis was not apparent in the human whole blood culture bactericidal assay, and patients with tuberculosis showed markedly improved bacterial killing after anti-tuberculous chemotherapy compared to before. The simplicity of a whole blood culture facilitates its inclusion in a clinical trial and it may have a potential role as a surrogate marker in a TB vaccine trial.