Matrix metalloproteinases (MMP) are a family of zinc-dependent enzymes which degrade various components of the extracellular matrix (ECM) and play an important role in facilitating tumour cell invasion of the normal brain. The family includes the gelatinases, stromelysins and collagenases. Preliminary studies have shown that there is a differential expression four metalloproteinases in human brain tumour cell lines derived from neoplasms of various histological types and grades of malignancy. Morphological and antigenic changes in human glioma-derived cell lines over many serial in vitro passages have been reported in earlier studies. When established cell lines are maintained in culture over a long period, it is possible that the secretion of enzymes such as metalloproteinases may differ according to the passage level examined. This report presents a study on the secretion of four matrix metalloproteinases - interstitial collagenase (MMP-), 72-kDa and 92-kDa gelatinases (MMP-2 and MMP-9 respectively), and stromelysin (MMP-3) - in three human brain tumour-derived cell lines at sequentially increasing passage numbers, ranging from passage 2 to passage 50; foetal astrocytes were used as a positive control. Reverse zymography and substrate degradation analysis were employed to demonstrate the presence of these enzymes in cell- conditioned culture medium. Aminophenyl mercuric acetate (APMA) was used to activate latent zymogen. Results demonstrate that there is no definite pattern of change in the levels of enzyme secretion common to all cell lines studied. Instead, the fluctuations in APMA- activated metalloproteinase activity in serial passage seems to vary considerably depending on the cell line and the type of enzyme studied. The variation in metalloproteinase expression observed on serial passage may be due to in vitro selection processes or karyotype evolution where the transcription of either the enzyme and/or its inhibitor may be affected. Thus an imbalance of the two products could be occurring in serial passage. Ideally, experiments requiring the measurement of relative enzyme activities should use cultures as near to the biopsy stage as possible, i.e. very low passages, to avoid artifacts that may arise on prolonged culturing.