A divalent metal ion-stimulated protein-serine/threonine phosphatase, PP1-arch, was purified approximately 1,000-fold from the extreme acidothermophilic archaeon Sulfolobus solfataricus (ATCC 35091). Purified preparations contained 40 to 70% of total protein as PP1-arch, as determined by assay-ing sodium dodecyl sulfatepolyacrylamide gels for protein phosphatase activity. The first 25 amino acids of the protein's sequence were identified, as well as an internal sequence spanning some 20 amino acids. Using this information, we cloned the gene for PP1-arch via the application of PCR and conventional cloning techniques. The gene for PP1-arch predicted a protein of 293 amino acids that bore striking resemblance to the members of the major family of protein-serine/threonine phosphatases from members of the domain Eucarya, the PP1/2A/2B superfamily. The core of the protein, spanning residues 4 to 275, possessed 29 to 31% identity with these eucaryal protein phosphatases. Of the 42 residues found to be absolutely conserved among the known eucaryal members of the PP1/2A/2B superfamily, 33 were present in PP1-arch. If highly conservative substitutions are included, this total reached 37. The great degree of sequence conservation between molecules from two distinct phylogenetic domains implies that the members of this enzyme superfamily had evolved as specialized, dedicated protein phosphatases prior to the divergence of members of the Archaea and Eucarya from one another.Protein phosphorylation-dephosphorylation represents a ubiquitous and prominent mechanism for regulating cellular processes. Protein kinases, protein phosphatases, and the phosphoproteins that they modify have been found in virtually every living organism. A priori, such universality might suggest that protein phosphorylation-dephosphorylation constitutes a very ancient means for exerting regulatory control. However, for many years, comparisons between the components of bacterial and eucaryal protein phosphorylation networks showed them to be structurally distinct, suggesting that each arose independently, at different times and places (reviewed in reference 18). For example, while all eucaryal protein kinases could be grouped into a single giant superfamily (14), the protein kinases from bacterial organisms showed no significant resemblance to this group (9). While members of the domain Eucarya phosphorylated some proteins on tyrosine residues, members of the domain Bacteria apparently did not (13). Therefore, it was and still is widely believed that protein phosphorylationdephosphorylation constitutes a relatively recent addition to the cell's regulatory repertoire, one that arose separately in each phylogenetic domain after it had diverged from its predecessors.Recent evidence has eroded the foundation upon which this consensus is based. Tyrosine-phosphorylated proteins have been detected in several members of the Bacteria (1,10,33,34,43). Genes encoding bacterial protein-histidine kinases have been discovered in eucaryal species ranging from rats (31...

No abstract

An ATP-like affinity labeling reagent, 5'-[p-(fluorosulfonyl)benzoyl]adenosine (FSBA), was used to probe the MgATP-binding site of smooth muscle myosin light chain kinase from chicken gizzard (smMLCK) and its calmodulin (CaM) complex. Native smMLCK has an absolute requirement for the binding of the calcium complex of CaM for expression of its catalytic activity. FSBA reacted with smMLCK-CaM and with the CaM-free, inactive enzyme as well. Both reactions were dependent on time and FSBA concentration. Reaction was accompanied by the incorporation of covalently bound [14C]FSBA into smMLCK protein at a molar ratio of approximately 1:1 in each case. p-(Fluorosulfonyl)benzoic acid, an analogue of FSBA lacking the adenosine targeting group, did not react at a significant rate with either form of smMLCK. Reaction of CaM-free and CaM-bound smMLCK with FSBA displayed saturation kinetics. The first-order rate constants for the conversion of the reversible, noncovalent enzyme-FSBA complex to form the irreversibly inhibited, covalently modified enzyme were similar for both smMLCK and smMLCK-CaM, 0.15 and 0.07 min-1, respectively. The concentrations of FSBA yielding the half-maximal rate of inactivation, KI, were essentially identical--0.65 and 0.64 mM, respectively--for smMLCK and smMLCK-CaM. MgATP, but not MgGTP or a substrate peptide, potently inhibited reaction with FSBA. Inhibition by MgATP was competitive. The measured inhibitory constant for MgATP was essentially the same--33 versus 34 microM--for both smMLCK and smMLCK-CaM. It therefore is concluded that the MgATP-binding site on smMLCK remains accessible and recognizable as such when the enzyme becomes inactivated upon dissociation of CaM.