Previous work has shown that normoxic isolated rat hepatocytes continuously produce
adenosine from
AMP and that the
nucleoside is not catabolized further but immediately rephosphorylated by
adenosine kinase [Bontemps, Van den Berghe and Hers (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 2829-2833]. We now report the effect of
anoxia on
adenosine production and on the
AMP/
adenosine substrate cycle. In cell
suspensions incubated in O2/CO2, the
adenosine concentration was about 0.4 microM. It increased 30-fold in cells incubated in N2/CO2 or with 5 mM KCN, and 20-fold in cells incubated with 2 mM
amytal.
Adenosine production, measured in hepatocytes in which
adenosine kinase and
adenosine deaminase were inhibited by
5-iodotubercidin and
deoxycoformycin respectively, was about 18 nmol/min per g of cells in normoxia; it increased about 2-fold in
anoxia, although
AMP increased 8-16-fold in this condition. From studies with inhibitors of membrane
5'-nucleotidase and of
S-adenosylhomocysteine hydrolase, it was deduced that
adenosine is produced by the latter
enzyme and by cytosolic
5'-nucleotidase in normoxia, and by cytosolic and membrane 5'-nucleotidases in
anoxia. Unlike in normoxic hepatocytes, inhibition of
adenosine kinase by
5-iodotubercidin neither elevated the
adenosine concentration nor enhanced total
purine release from
adenine nucleotides in cells treated with N2/CO2 or KCN; it had only a slight effect in cells treated with
amytal. This indicates that recycling of
adenosine is suppressed or profoundly inhibited in
anoxia. The rate of accumulation of
adenosine in
anoxia was several-fold lower than the rate of its rephosphorylation upon reoxygenation. It is concluded that the elevation of
adenosine in anoxic hepatocytes is much more dependent on decreased recycling of
adenosine by
adenosine kinase than on increased production by dephosphorylation of
AMP.