Immunofluorescent surface labelling, flow sorting and culturing of putative epidermal stem cells derived from small skin punch biopsies.

Basal keratinocytes of human epidermis strongly express the cell surface glycoprotein beta(1)-integrin, and putatively harbour epidermal stem cells. Selective sorting and culturing of keratinocyte stem cells forms the basis for studies on the role of these cells as targets for therapeutic intervention and gene therapy. Here we have studied variables which affect cell surface labelling for beta(1)-integrin, flow sorting and subsequent culturing of beta(1)-integrin-positive and beta(1)-integrin-negative keratinocytes. Keratinocytes were derived from small human skin punch biopsies (3 or 4 mm in diameter), and we tested a number of variables such as choice of proteolytic enzyme for cell isolation, cell concentration, fixation, storage of fixed cell suspensions and labelling conditions. In contrast to thermolysin treatment for cell isolation, trypsin treatment left most cell surface beta(1)-integrin molecules intact. Ethanol and paraformaldehyde fixation interfered with beta(1)-integrin detection, and unfixed cells gave the best results. Optimisation of all the individual steps resulted in a labelling protocol for reproducible staining and sorting of the cells. Sorted cells were seeded in 96-well plates (300 cells/well) and colonies were obtained in more than 50% of the wells with beta(1)-integrin-positive keratinocytes. In plates with beta(1)-integrin-negative cells, only 10% of the wells contained keratinocyte colonies. Flow sorted keratinocytes obtained by trypsin formed numerous colonies in cell culture experiments. In cell suspensions obtained with thermolysin, only sparse colonies were formed. We conclude that our methodology permits the use of small human tissue samples for cell labelling and sorting, while preserving the clonogenic potential.