Normally, insulin binding to its receptor induces phosphorylation of the serine/threonine kinase Akt/protein kinase B (PKB). One of Akt/PKB's functions is to inhibit glucose output from the liver when glucose is already available from food. If this fails, hepatic glucose output increases. So, a report in Science showing that TRB3 is a negative modulator of Akt/PKB has widespread implications for type II diabetes.

Du et al. isolated TRB3 in a yeast two-hybrid screen for proteins that modulate Akt/PKB activity, and verified this interaction using in vitro assays and coimmunoprecipitations. Like its Drosophila homologue tribbles, TRB3 has a truncated kinase domain and lacks detectable catalytic activity. However, further truncation of the kinase domain weakened the interaction of TRB3 with Akt/PKB.

The effect of TRB3 on Akt/PKB was to inhibit its phosphorylation at threonine (Thr) 308 and serine (Ser) 473, thereby reducing its kinase activity — there was no change in the protein levels of Akt/PKB. Perhaps unsurprisingly, therefore, when the authors disrupted endogenous TRB3 expression in hepatocytes using RNA interference, Akt/PKB phosphorylation was potentiated in response to growth-factor signalling. So, how does this happen?

Phosphorylation-defective and phosphorylation-mimicking Akt/PKB constructs in yeast two-hybrid assays indicated that TRB3 preferentially binds to unphosphorylated Akt/PKB. Further studies showed that this was probably through TRB3 binding to, and therefore blocking, the Akt/PKB Thr308 phosphorylation site.

The authors then looked at TRB3 expression in mouse liver under feeding or fasting conditions. Hepatic TRB3 messenger RNA and protein levels were 10–20 times higher in fasted animals than fed ones, and were also higher in livers from db/db diabetic mice. When Du et al. overexpressed TRB3 in vivo, they saw increased blood glucose levels in re-fed, but not fasted, mice, indicating that TRB3 might interfere with insulin-mediated hepatic glucose release. And in a glucose-output assay, insulin could hardly inhibit glucose release from TRB3-overexpressing cells, in contrast to wild-type cells. This might be partly due to TRB's ability to block phosphorylation of Akt/PKB substrates such as glycogen synthase kinase-3β.

Under fasting conditions, TRB3-mediated hepatic glucose output might be beneficial, but pathological TRB3 overexpression after feeding might contribute to insulin resistance and hyperglycaemia. So, Du et al. have highlighted that TRB3 could be an attractive target for treating type II diabetes.