Page 83 - 72_04
P. 83

VOL. 72 (4), 611-627, 2006  HEME RESPONSIVENESS IN VITRO IS A COMMON...

In addition, some residues of this motif help to anchor ATP by
forming hydrogen bonds with either the adenine or the ribose ring
(15). Moreover, previous results indicate that the cysteine residue in
the HRM of six different hemoproteins is absolutely essential for
binding to heme (19). It has been suggested that such cysteine serves
the critical function of donating electrons to the iron atom of heme
(19). Therefore, it is possible that these three amino acids (I380,
Q381 and C385), the Cys-385 in particular, may be directly involved
in hemin binding, although the exact heme regulatory sites within
the eIF2a kinases remain to be determined.

    Collectively, our results indicate that, although the known
regulatory mechanisms of eIF2a kinases in vivo are very different,
all known eIF2a kinases are regulated in vitro by hemin. Thus, hemin
may represent a regulatory mechanism unique to eIF2a kinases.
On the other hand, heme has been shown to function as an effector
molecule that can regulate many biological processes, including
transcription, translation, protein translocation and erythroid
differentiation. Also, heme plays key roles in oxygen sensing and
utilization in all living organisms ranging from bacteria to humans
(24). Our studies provide a new example of how heme may control
the activity of a family of translational inhibitors and how a
regulatory system may be conserved from yeast to mammals to
control protein synthesis. Thus, the eIF2a kinases may provide
a feedback mechanism to coordinate the synthesis of some
hemoproteins that are essential for the function of all aerobic cells,
according to heme concentration in eukaryotic cells. When cells grow
in a limiting supply of oxygen, the low concentration of heme
synthesized would allow the activation of the eIF2a kinases, and
a low level of protein synthesis would result. In a more aerobic
environment, heme would repress the eIF2a kinases, eliciting full
induction of protein synthesis.

                               ACKNOWLEDGMENTS

    We thank Emilio Itarte and Jorge Moscat for rat CKII and rat
?PKC, respectively. We thank E. Martínez-Salas and N. Pomar for
critical reading of the manuscript. We also thank José Alcalde

                                                                                             625
   78   79   80   81   82   83   84   85   86   87   88