Acetyl‐CoA carboxylase was purified to homogeneity, in the presence of protein phosphatase inhibitors, from rat liver sampled without freeze‐clamping. The enzyme was in a highly phosphorylated state (4.8 mol/subunit) of low specific activity, and could be dramatically reactivated by treatment with protein phosphatase‐2A. Amino acid sequencing and fast‐atom‐bombardment mass spectrometry showed that the enzyme was phosphorylated in Ser79, Ser1200 and Ser1215, the three sites known to be phosphorylated in cell‐free assays by the AMP‐activated protein kinase.

The inactive enzyme could also be completely reactivated using a limited treatment with trypsin, which removes the N‐terminal segment containing Ser79 and reduces the phosphate content to 3.5 mol/subunit. These results strengthen previous findings that it is phosphorylation at Ser79 by the AMP‐activated protein kinase that is responsible for the inactivation, and not the phosphorylation of the 220‐kDa core fragment (which contains Ser1200 and Ser1215).

Analysis of the phosphorylation state of Ser79 in acetyl‐CoA carboxylase from rat liver showed that phosphorylation occurs post mortem if freeze‐clamping is not used. The higher phosphorylation observed in extracts made without freeze‐clamping correlates with a large increase in AMP and decrease in ATP (presumably caused by hypoxia during removal of the liver), and with increased activity of the AMP‐activated protein kinase. These results provide a rational explanation for the post mortem phosphorylation events, and re‐emphasize the point that rapid cooling of cells and tissues is essential when measuring the expressed activity of acetyl‐CoA carboxylase (as well as 3‐hydroxy‐3‐methylglutaryl‐CoA reductase).

Using the freeze‐clamping procedure, the ratio of‘expressed’ activity (measured in the presence of protein phosphatase inhibitors) to ‘total’ activity (measured after complete dephosphorylation) of rat liver acetyl‐CoA carboxylase showed a marked diurnal rhythm, changing from 50% in the active form in the middle of the dark period to less than 10% active in the middle of the light period. The very low activity in the light period was associated with a high level of phosphorylation in Ser79. This diurnal rhythm is very similar to that previously described for the phosphorylation of 3‐hydroxy‐3‐methylglutaryl‐CoA reductase, another substrate for the AMP‐activated protein kinase. Neither the activity of the AMP‐activated protein kinase nor the content of AMP, ADP or ATP changed between the dark or light periods.

Feeding of high‐fat diets (beef tallow or sunflower oil) prevented the dephosphorylation of liver acetyl‐CoA carboxylase that normally occurs during the dark period in rats on chow diet, as well as dramatically reducing the total amount of enzyme.