Page 80 - 72_04

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

concentrations of hemin (Fig. 3A). Moreover, the extent of inhibition
of all of them by heme was similar to that of mouse HRI at all hemin
concentrations tested (Fig. 3B). We therefore conclude that the
kinase activities of different eIF2a kinases are also regulated in vitro
by hemin.

Moreover, the two novel eIF2a kinases from S. pombe (SEK1/
Hri1p and SEK2/Hri2p) were also heme-responsive in vitro (Fig. 3C).
Again, the extent of the inhibition of SEK1 and SEK2 by heme was
similar to that of mouse HRI at all concentrations tested (Fig. 3D).
Altogether, our results indicate that the responsiveness to hemin in
vitro is a common property of all the other eIF2a kinases.

We were interested in ascertaining the possible specificity for the
hemin effect reported here. In order to see if other protein kinases,
in addition to eIF2a kinases, were heme responsive, we tested the
effect of hemin on the protein kinase activity of two other Ser/Thr
protein kinases, CKII and ?PKC. The data presented in Fig. 4
demonstrated that the in vitro phosphorylation of either casein, by
CKII, or MBP, by ?PKC, was not inhibited by the presence of hemin.
We therefore conclude that the eIF2a kinases, but not other serine-

FIGURE 4. A, shown are the results from the in vitro phosphorylation of eIF2a by
recombinant mouse HRI, as a control, of casein by purified casein kinase II, and
of MBP by purified His-tagged ?PKC, in the absence or presence of the indicated
concentrations of hemin. The positions of phosphorylated eIF2a, casein and MBP
are indicated. B, the phosphorylation of eIF2a, casein and MBP was estimated as

described in Figure 2.

622