ADENOSINE 5'-TRIPHOSPHATE (ATP) CONCENTRATIONS IN THE ENDOLYMPH AND PERILYMPH OF THE GUINEA-PIG COCHLEA

The concentration of adenosine S-triphosphate (ATP) in endolymph (EL), perilymph (PL) and cerebrospinal fluid (CSF), collected from anesthet ized guinea pigs was determined using the luciferase-luciferin reactio n. The cochlea was exposed by a ventrolateral approach and the bone ov erlying scala media of the third turn (EL) or scala vestibuli of the f irst turn (PL) was shaved to a thin layer and a small fenestrum made. For EL sampling, a double-barrelled pipette was inserted through the s piral ligament-stria vascularis complex. One barrel was filled with 15 0 mM KCl to record the endocochlear potential (EP) and upon the appear ance of the positive EP, 0.12-1.22 mu l of fluid was aspirated into th e other barrel by gentle negative pressure. For PL sampling, a single- barrelled pipette was advanced into scala vestibuli and 0.3-1.6 mu l o f fluid was collected by capillarity. CSF (0.36-1.75 mu l) was obtaine d from the cisterna magna. The cochleae were removed and processed for light microscopy to determine the extent of tissue damage from the sa mpling procedure. ATP concentrations (mean +/- SEM, nM) for EL, PL and CSF were 12.95 +/- 2.4 (n = 10), 10.5 +/- 3.9 (n = 11) and 16.1 +/- 5 .4 (n = 11) respectively. Differences in ATP concentrations among flui ds were not statistically significant. To test the effect of hypoxia o n ATP levels, a group of guinea pigs was subjected to a 90 s period of respiratory anoxia prior to sampling of EL, PL or CSF. ATP concentrat ions were 14.4 +/- 3.5 (n = 11), 20.7 +/- 4.1 (n = 10) and 13.5 +/- 4. 6 (n = 4) for EL, PL and CSF, respectively; only PL ATP concentrations were statistically different (P = 0.018, Wilcoxon rank sum test) to b asal conditions. This is the first study which demonstrates the presen ce of free ATP in cochlear fluids. The results indicate that ATP is pr esent in cochlear fluids at concentrations close to those known to cau se hair cell depolarization in vivo.