Summary Hypoxia in solid tumours has been correlated with poor prognosis and resistance to radiation and chemotherapy. Hypoxia is also a strong stimulus for gene expression. We previously proposed a gene therapy approach which exploits the presence of severe hypoxia in tumours for the induction of therapeutic genes. Hypoxic cells are known to have a reduced metabolic rate, transcription and translation. These facts may prevent gene transfer and therefore warranted further investigation. In this paper the feasibility of gene delivery in vitro under tumour conditions was demonstrated. DNA was delivered in vitro using a peptide-mediated non-viral system. Across a range of oxygen tensions and mammalian cell lines (including human tumour and endothelial cells) it was shown that hypoxic cells could be transfected. Transfection efficiencies varied depending on the level of hypoxia, cell characteristics and gene promoters used. An in vitro model of hypoxia/reoxygenation, designed to mimic the variable nature of tumour hypoxia, showed that hypoxic preconditioning and reoxygenation alone did not reduce transfection efficiency significantly; only chronic anoxia reduced transfection. The fact that neither intermediate hypoxia nor intermittent anoxia significantly reduced transfection is promising for future hypoxia-targeted gene therapy strategies. 662-667 © 2000 Cancer Research Campaign doi: 10.1054/ bjoc.2000.1318, available online at http://www.idealibrary.com on List of abbreviations: HIF-1, hypoxia inducible factor 1; CMV, cytomegalovirus; SV40, simian virus 40; AnO 2 , anoxia; GFP, green fluorescent protein; FACS, fluorescent activated cell sorting Transfection efficiencies of two commonly used marker genes and viral promoters were recorded in a range of cell lines under different oxygenation conditions. A simple in vitro model of hypoxia/reoxygenation was used to mimic the variable nature of tumour hypoxia.
MATERIALS AND METHODS
Cell lines and growth mediaThe human bladder carcinoma cell line T24 (European Collection of Cell Cultures; Bubenik et al, 1973;Dirks et al, 1999), the immortalized human microvascular endothelial cell line HMEC-1 (Ades et al, 1992), human squamous carcinoma cell line of the head and neck FaDu (Rangan, 1972; American Type Culture Collection), the mouse hepatoma cell line Hepa-1 wild type (clone Hepa-1c1c7) and its derivative Hepa c4 (HIF-1β mutant; Hankinson, 1979) were utilised. Only cells tested negative for mycoplasma were used in the experiments, since it had been shown that mycoplasma interfered with transfection, especially under hypoxia (results not shown). All cells were maintained in Dulbecco's MEM (DMEM, Life Technologies, UK) with 10% fetal calf serum and L-glutamine in a 5% CO 2 humidified incubator at 37°C.Cell proliferation was monitored by cell counting using a haemocytometer and trypan blue (0.2% final concentration, Sigma) to distinguish live and dead cells.
Hypoxic conditionsCells were plated 16-24 h prior to the experiment in oxygen impermeable dishes (Permanox, Nunc, UK...