Photohaemolysis assay for drug phototoxicity complicated by 'bleaching' of released haemoglobin

The phototoxic potential of several non-steroidal anti-inflammatory drugs and quinolone antibiotics was assessed using the photohaemolysis assay. In this system, human erythrocytes are irradiated (UVA radiation 320-400 nm) from below in the presence of suspected photosensitizers. Photohaemolysis with ketoprofen, tiaprofenic acid or nalidixic acid was initially concentration dependent, but photohaemolysis apparently decreased at higher drug concentrations. As erythrocytes were irradiated from below any optical screening at high drug concentrations was discounted. Other phototoxic drugs can oxidize haemoglobin to methaemoglobin resulting in a decrease in absorption and an artificially lowered photohaemolysis level. However, in the present experiments, the use of Drabkin's solution overcame this effect as haemoglobin and most of its oxidized derivatives were converted into a single derivative, namely cyanmethaemoglobin. Further possibilities are that photosensitized damage to haemoglobin results in the formation of intracellular Heinz bodies and/or bleaching of released haemoglobin. The latter hypothesis was tested by irradiating free haemoglobin in the presence of drugs. The results suggested that certain phototoxic agents cause 'bleaching' of the haemoglobin and formation of a derivative that fails to react with Drabkin's solution. If only high concentrations of these drugs are used in the photohaemolysis assay, this increases the risk of false negative results. It is therefore suggested that both photohaemolysis and photosensitized 'bleaching' of haemoglobin should be investigated when using this assay for screening drug phototoxicity.