An intuitive normalization and gene identification method is proposed. After segmentation of the entire expression range into intensity intervals, the mean and standard deviation of the logarithm of expression ratios are calculated for each interval using the nearest neighbor genes. Genes with high differential expression are excluded from these calculations. For glass arrays, normalization is performed for each interval by using the mean of the logarithm of expression ratios in the interval. For nylon͞plastic membranes, the average of the means of the logarithm of ratios across the intervals of higher intensities is used for normalization.

Immunotherapy with ex vivo cultured T cells depends on a large supply of biologically active cells. Understanding the effects of culture parameters is essential for improving the proliferation and efficacy of the expanded cells. Low oxygen tension (5% pO(2)) was previously reported to improve T-cell expansion and alter cellular phenotypic characteristics compared to T cells cultured at 20% pO(2). Here we report the use of DNA-array based transcriptional analysis coupled with protein-level analysis to provide molecular insights into pO(2) and patient-variability effects on expanded primary human T cells. Analysis of seven blood samples showed that reduced pO(2) results in higher expression of genes important in lymphocyte biology, immune function, and cell-cycle progression. 20% pO(2) resulted in higher expression of genes involved in stress response, cell death, and cellular repair. Expression of granzyme A (gzmA) was found to be significantly regulated by oxygen tension with cells at 5% pO(2) having greater gzmA expression than at 20% pO(2). Protein-level analysis of gzmA was consistent with transcriptional analysis. Granzyme K (gzmK) was coexpressed with gzmA, whereas Granzyme B (gzmB) expression was found to precede the expression of both gzmA and gzmK in 15-day cultures. Temporal gene expression patterns for seven blood samples demonstrate that most genes are expressed by all patient samples in similar temporal patterns. However, several patient-specific gene clusters were identified, and one cluster was found to correlate well with cell proliferation and may potentially be used to predict patient-specific T-cell expansion.

No abstract

No abstract