Electrochemical monitoring of anticancer compounds on the human ovarian carcinoma cell line A2780 and its adriamycin- and Cisplatin-resistant variants.

A novel electrochemical technique which detects and monitors real-time changes in cell behavior in vitro has been used to examine the effects of recognized anticancer drugs on the human ovarian carcinoma cell line A2780 and its adriamycin (A2780adr)- and cisplatin (A2780cispt)-resistant variants. These cells, adherent to gold electrodes or sensors, modify the extracellular microenvironment at the cell:sensor interface, producing an electrochemical potential that is different from that of the bulk culture medium. Confluent, adherent A2780 cells produced an electrochemical signal, measured as an open circuit potential (OCP), of approximately -100 mV compared to a cell-free value of approximately -15 mV. Exposure of A2780 cells to cisplatin (range 10(-4) to 10(-6) M), adriamycin (range 10(-5) to 10(-7) M), and vinblastine (10(-6) M) all produced positive shifts in the OCP signal relative to untreated control cells during 24 h of culture, but Taxotere (range 10(-5) to 10(-7) M) had no effect. These positive shifts in OCP signal were evident well before observations of reduced cellular adhesion and viability after 24 h, as judged in parallel cultures with a plastic substratum and by scanning electron microscopy. By contrast, the same treatments applied to the A2780adr and A2780cispt variants showed that each demonstrated different sensitivities to the same drugs applied to the parental A2780 cells. The effects of the same four anticancer drugs on ovarian carcinoma (A2780) and breast carcinoma (8701-BC) cell lines showed that the former was far more responsive to adriamycin and cisplatin. Such differences in drug sensitivities between the two cell lines were subsequently confirmed using the conventional MTT assay over 5 days. Although this electrochemical technology readily detects changes in cell adhesion and viability, the modified OCP signals recorded within a few hours of anticancer drug treatments are evident well before microscopic morphological changes become apparent. It is proposed that these early changes in OCP signals, relative to control untreated cells, reflect modifications of physiological/behavioral processes manifested at the cell surface.

Electrochemical monitoring of cell behaviour in vitro: a new technology.

This article describes a novel electrochemical technique for the real-time monitoring of changes in the behaviour of adherent human cells in vitro: i.e., a biosensor that combines a biological recognition mechanism with a physical transduction technique, described collectively as Oncoprobe. Confluent viable cells adherent to gold electrodes (sensors) modify the extracellular microenvironment at the cell:sensor interface to produce a change in the electrochemical potential compared to that measured in the absence of cells. The potential was measured as an open circuit potential (OCP) with respect to a saturated calomel reference in the bulk culture medium. Typical OCP values for confluent cultures of human breast carcinoma cells, 8701-BC, approximated -100 mV compared with cell-free values of approximately -15 mV. The OCP for 8701-BC cells was modified in response to temperature changes over the range 32 to 40 degrees C and also to treatments with phytohemagglutinin (PHA, 25 microg/mL), cycloheximide (30 microM) and interleukin-1 beta (IL-1, 0.5 ng/mL) over 24 h. Cultures of synovial fibroblasts also responded to the same treatments with similar responses, producing negative shifts in the OCP signal with PHA and IL-I, but a positive shift in OCP signal with cycloheximide, all relative to the untreated control cultures. From experimental data and theoretical considerations it is proposed that the cell-derived signals are mixed electrode potentials reflecting a "conditioned," more reducing environment at the cell:sensor interface. Only viable cells caused a negative shift in the OCP signal, this being lost when cells were rendered nonviable by formalin exposure. This technology appears unique in its ability to passively "listen in" on cell surface activities, suggesting numerous applications in the fields of drug discovery, chemotherapy, and cell behaviour.

Matrix metalloproteinase and proinflammatory cytokine production by chondrocytes of human osteoarthritic cartilage: associations with degenerative changes.

OBJECTIVE: To examine by immunohistochemistry the relative distributions of 6 matrix metalloproteinases (MMPs 1, 2, 3, 8, 9, and 13) and the 2 proinflammatory cytokines interleukin-1beta (IL-1beta) and tumor necrosis factor alpha (TNFalpha) in osteoarthritic (OA) cartilage compared with normal, age-matched articular cartilage. METHODS: Articular cartilage samples were obtained from the tibial plateau of OA knees removed at arthroplasty and from normal, nonarthritic, knees obtained at autopsy. Specimens were promptly fixed in Carnoy's fixative, processed, embedded in paraffin, sectioned, and examined by immunohistochemistry for MMP and cytokine production. In addition, human articular chondrocytes (HAC) were treated in vitro with either IL-1beta, TNFalpha, or phorbol myristate acetate (PMA) to assess their potential to produce each of the MMPs, as determined by Western blotting and gelatin zymography. RESULTS: Immunodetection of the collagenases (MMPs 1, 8, and 13) and stromelysin 1 (MMP-3) was demonstrated in a proportion of chondrocytes in the superficial zone of almost all of the OA specimens that had degenerative matrix changes. The gelatinases (MMPs 2 and 9) were also demonstrated by immunohistochemistry but were not so prominent. IL-1beta- and TNFalpha-positive chondrocytes were also observed in a proportion of cells in the superficial zones of OA specimens. Much less immunostaining for MMPs and cytokines was observed in the deep zone of all OA specimens, where the cartilage matrix and chondrocyte morphology appeared normal. In contrast, full-thickness normal cartilage specimens showed virtually no immunostaining for these MMPs or cytokines. Confirmation that chondrocytes can produce these 6 MMPs was obtained from HAC cultures treated with either IL-1beta, TNFalpha, or PMA; conditioned medium from activated HAC contained all the MMPs demonstrated by immunohistochemistry. Dual immunolocalization studies of OA cartilage specimens demonstrated the coexpression of IL-1 with MMP-8 by individual chondrocytes in situ. CONCLUSION: These results indicate that the superficial zone of OA cartilage specimens, which is characterized by fibrillations, chondrocyte clusters, and degenerative matrix changes, contains a variable proportion of cells that immunostain for IL-1beta, TNFalpha, and 6 different MMPs. These observations support the concept that cytokine-MMP associations reflect a modified chondrocyte phenotype and an intrinsic process of cartilage degradation in OA.