Inorganic pyrophosphate generation from adenosine triphosphate by cell-free human synovial fluid

To quantify inorganic pyrophosphate (PPi) production from extracellular adenosine triphosphate (ATP) by human synovial fluids (SF).

Serial measurements of ATP hydrolysis rate (t1/2) were performed by the luciferase method from a starting concentration of 500 nM in 21 pathologic and one normal cell-free SF samples incubated under physiologic conditions. ATP was then pumped into a sample of each fluid, using the rate constant derived from the t1/2 of that fluid, to provide steady state levels simulating those reported in SF. Trace [32P] gamma ATP was added at the start of the infusion; conversion to [32P] Pi and to [32P] PPi was determined by precipitation of Pi as reduced phosphomolybdate before and after treatment with yeast inorganic pyrophosphatase. Finally, the pumping experiment was repeated and PPi production was calculated from direct measurement of PPi at time zero and at 60 min. PPi hydrolysis was measured in each fluid by [32P] Pi precipitation from [32P] PPi tracer added at time zero.

ATP was hydrolyzed by all SF. The mean t1/2 (seconds) in 8 osteoarthritis (OA) samples was 72 s, in 5 calcium pyrophosphate dihydrate (CPPD) 30 s (p < 0.02), in 3 rheumatoid arthritis (RA) 1160 s, in normal 86 s, in 3 olecranon bursal (OB) 54 s, and in 2 total knee replacement fluid samples 17 and 121 s. The major product of ATP hydrolysis was PPi in all but 2 fluids (1 RA, 1 OB), even at lower than steady state levels. At simulated in vivo steady state ATP levels, mean conversion of APT to PPi was stoichiometric in OA and CPPD fluids. PPi hydrolysis was < 4% in all noninflammatory fluids.

PPi is the major product of extracellular ATP catabolism in most SF. Hydrolysis rates were significantly faster in SF containing CPPD crystals. Mean PPi production by these fluids at simulated in vivo steady state levels was 6-fold that of OA SF (p < 0.01). Hydrolysis of extracellular ATP by ectonucleotide pyrophosphohydrolases can account for all PPi produced by joint issues previously estimated from [32P] PPi pool and turnover studies in human knee joints.