Multiple sclerosis is a chronic inflammatory demyelinating disease of the CNS. Although the aetiology of multiple sclerosis is still unknown, it is widely believed that T cells play a central role in its pathogenesis. To identify and characterize disease‐relevant T cells, we analysed CD4⁺ and CD8⁺ T cells freshly isolated from the CSF and peripheral blood of 36 multiple sclerosis patients for their T‐cell receptor variable β (TCRBV) chain repertoire. In most patients, we found significant overexpression of individual TCRBV chains on CD8⁺ T cells from CSF compared with peripheral blood. In contrast, only a few multiple sclerosis patients showed differences between the two compartments in TCRBV expression on CD4⁺ T cells. The overexpression of specific TCRBV chains on CD8⁺ T cells was found to be stable over several months in selected patients and involved mainly T cells with a memory phenotype. In two patients studied, individual TCRBV chain overexpression was found to be caused by the expansion of T cell populations with identical or highly similar rearranged T‐cell receptor β‐ and α‐chain sequences, which were not found among peripheral blood CD8⁺ T cells. Our findings demonstrate selective enrichment of memory CD8⁺ T cells in the CSF of multiple sclerosis patients, suggesting a role for these CD8⁺ T cells in the pathogenesis of multiple sclerosis. Our study provides a basis for future trials to identify disease‐associated antigens and disease pathogenesis in multiple sclerosis.

Adjuvant intravesical bacillus Calmette‐Guérin (BCG) therapy is a well‐established and successful adjuvant immunotherapy in the treatment of superficial bladder cancer. Although the function of natural killer (NK) cells in other immunotherapeutic regimens (e.g., lymphokine‐activated killer [LAK] cell or interleukin‐2 [IL‐2] therapy) has been established, the contribution of NK cells to effective BCG immunotherapy is not clear. We used a human in vitro system to analyze the role of NK cells in BCG‐induced cellular cytotoxicity. After stimulation of mononuclear cells with BCG for 7 days, these BCG‐activated killer (BAK) cells displayed substantial cytotoxicity against bladder tumor cells. Magnetic depletion experiments and fluorescence activated cell sorting revealed that NK cells were the major effector cell population. To address NK cell function in vivo, we studied a syngeneic orthotopic murine bladder cancer model and compared BCG immunotherapy in C57BL/6 wild‐type mice, NK‐deficient beige mice and mice treated with anti‐NK1.1 monoclonal antibody. Four weekly instillations of viable BCG significantly prolonged survival in wild‐type mice compared with control mice treated with solvent alone. In contrast, BCG therapy was completely ineffective in NK‐deficient beige mice and in mice treated with anti‐NK1.1 monoclonal antibody. These findings suggest a key role for NK cells during BCG immunotherapy.

For heterogeneous tissues, such as blood, measurements of gene expression are confounded by relative proportions of cell types involved. Conclusions have to rely on estimation of gene expression signals for homogeneous cell populations, e.g. by applying micro-dissection, fluorescence activated cell sorting, or in-silico deconfounding. We studied feasibility and validity of a non-negative matrix decomposition algorithm using experimental gene expression data for blood and sorted cells from the same donor samples. Our objective was to optimize the algorithm regarding detection of differentially expressed genes and to enable its use for classification in the difficult scenario of reversely regulated genes. This would be of importance for the identification of candidate biomarkers in heterogeneous tissues. Experimental data and simulation studies involving noise parameters estimated from these data revealed that for valid detection of differential gene expression, quantile normalization and use of non-log data are optimal. We demonstrate the feasibility of predicting proportions of constituting cell types from gene expression data of single samples, as a prerequisite for a deconfounding-based classification approach. Classification cross-validation errors with and without using deconfounding results are reported as well as sample-size dependencies. Implementation of the algorithm, simulation and analysis scripts are available. The deconfounding algorithm without decorrelation using quantile normalization on non-log data is proposed for biomarkers that are difficult to detect, and for cases where confounding by varying proportions of cell types is the suspected reason. In this case, a deconfounding ranking approach can be used as a powerful alternative to, or complement of, other statistical learning approaches to define candidate biomarkers for molecular diagnosis and prediction in biomedicine, in realistically noisy conditions and with moderate sample sizes.