Page 78 - 72_04

SATURNINO HERRERO Y COLS. AN. R. ACAD. NAC. FARM.

FIGURE 2. A, samples of recombinant HRI wt, as a control; HRI C409A/C550A;
HRI ?8; HRI ?8/C550A; HRI ?137/C409A/C550A; HRI 138-580-C409A/C550A;
and HRI Q4 proteins were analyzed for their ability to phosphorylate purified
rabbit reticulocyte eIF2, in the presence of increasing concentrations of hemin

as indicated. Position of phosphorylated eIF2a is indicated. B, the phosphorylation
of eIF2a was estimated by quantifying the corresponding band density of the

autoradiograms. The intensity of the eIF2a band at zero concentration of hemin
was defined as 100%. The figure shows a representative experiment of at least

three independent experiments with very similar results.

replaced by the mouse GCN2 corresponding region, showed lower
eIF2a kinase activity (Fig. 2A). Probably due to its low specific
activity, HRI Q4 was more sensitive to inhibition by heme (0.4 µM
for 50% inhibition) than that of wild type HRI (Fig. 2B). This result
suggests that the kinase insert region, which does not show any
significant homology with those of the other eIF2a kinases, may not
be the important region for achieving the heme responsiveness of
HRI. Altogether, our mutational analysis indicates that the heme
regulatory region(s) are spread around the 12 conserved subdomains
characteristic of all eukaryotic Ser/Thr protein kinases.

Because the kinase catalytic region of HRI was responsible for
heme regulation, we tested whether the other eIF2a kinases, besides
mammalian HRI, could be inhibited by hemin in vitro. To this end,
we used purified eIF2a kinases from transfected HEK 293T cells:

620