Second messengers and phosphoproteins最新文献

Two distinct types of protein kinase C cDNA clones were isolated from a Xenopus laevis oocyte cDNA library, and the complete nucleotide sequences were determined. The sequences encode a single open reading frame with a domain structure that consists of four constant (designated C1-C4) and five variable (designated V1-V5) regions. Comparison of the two sequences shows good homology at the nucleotide and deduced amino acid level. The differences reside primarily in the variable regions. Each clone encodes 671 and 676 amino acids, respectively, having extensive homology with published sequences of human, rat, and bovine protein kinase C. These results provide evidence that these two distinct types of protein kinase C are members of a multigene family in amphibian and mammalian species.

Platelet adenylate cyclase appears to be regulated through separate stimulatory and inhibitory prostaglandin receptors. To test the possibility that the inhibitory receptor represents overlap with a thromboxane A2 receptor the effect of thromboxane antagonists on prostaglandin regulation of adenylate cyclase was examined. Neither 13-azaprostanoic acid nor SQ29,548 had any effect on prostaglandin-mediated cyclic AMP regulation in intact platelets, while pinane thromboxane A2 and carbocyclic thromboxane A2 exhibited behavior consistent with these compounds acting as agonists at the inhibitory prostaglandin site. Because of the divergent behavior of the two groups of thromboxane antagonists it is unclear whether the inhibitory prostaglandin site represents a thromboxane site. It seems clear, however, that PTA2 and CTA2 represent pure agonists at the prostaglandin inhibitory site, showing little, if any, overlap with the stimulatory site, and therefore represent useful compounds for the study of prostaglandin regulation of platelet adenylate cyclase, providing additional evidence for the existence of a distinct prostaglandin inhibitory site.

Known synergism between forskolin and hormones in adenylate cyclase activation leads to the supposition that hormone might stimulate forskolin binding. That possibility was tested using intact wild type S49 cultured lymphoma cells. Using 40 nM [3H]-forskolin it was shown that the extent of forskolin binding using a filtration technique increased with the concentration of epinephrine or isoproterenol (INE). Moreover, the hormone-dependent forskolin binding was stereospecific (requiring l- rather than d-epinephrine), it was not observed in the cyc- variant and it was not inhibited by cytochalasin B. These observations lead to the conclusion that the binding is specifically associated with the adenylate cyclase system and requires a functional Gs unit. Epinephrine-stimulated forskolin binding did not correlate exactly with forskolin activation of adenylate cyclase in the presence of similar concentrations of epinephrine. It was concluded from that observation that there is not a one to one correspondence between binding and activation.

Photoaffinity labeling analogs of the adenylate cyclase activator forskolin (PF) have been synthesized, purified and tested for their effect on preparations of membrane-bound, Lubrol solubilized and forskolin affinity-purified adenylate cyclase (AC). All analogs of forskolin significantly activated AC. However, in the presence of 0.1 to 0.3 microM forskolin, the less active forskolin photoaffinity probes at 100 microM caused inhibition. This inhibition was dose-dependent for PF, suggesting that PF may complete with F for the same binding site(s). After cross-linking [125I]PF-M (see Figure 1 for structure) to either membrane or Lubrol-solubilized AC preparations by photolysis, a radiolabeled 100-110 kDa protein band was observed after autoradiography following SDS-PAGE. F at 100 microM blocked the photoradiolabeling of this protein. Radioiodination of forskolin-affinity purified AC showed several protein bands on autoradiogram, however, only one band (Mr = 100-110 kDa) was specifically labeled by [125I]PF-M following photolysis. The photoaffinity-labeled protein of 100-110 kDa of AC preparation of rat adipocyte may be the catalytic unit of adenylate cyclase of rat adipocyte itself as supported by the facts that [a] no other AC-regulatory proteins are known to be of this size, [b] the catalytic unit of bovine brain enzyme is in the same range and [c] this PF specifically stimulates AC activity when assayed alone, and weekly inhibits forskolin-activation of cyclase. These studies indicate that radiolabeled PF probes may be useful for photolabeling and detecting the catalytic unit of adenylate cyclase.

Adenosine attenuates beta-adrenergic receptor mediated activation of adenylate cyclase in myocardial membranes via adenosine Ri receptors. The effects of adenosine analogs on the binding characteristics of beta-adrenergic receptors were examined in the present study utilizing rat ventricular membranes treated with adenosine deaminase. In 125I-cyanopindolol/isoproterenol competitive binding experiments phenylisopropyladenosine (PIA) significantly increased the IC50 for isoproterenol from 48 +/- 6 nM to 140 +/- 48 nM and steepened the slope of the competition curves from -0.56 +/- 0.03 to -0.90 +/- 0.21. Computer analysis of these curves indicated that binding of isoproterenol to the high affinity state of the beta-adrenergic receptor was eliminated in the presence of PIA. PIA had no effects in the presence of GPP(NH)P. 2-chloroadenosine, a less specific Ri agonist, caused smaller increases in IC50 and slope, without significantly affecting high affinity binding. 2',5'-dideoxyadenosine, a P-site agonist, had no significant effects on isoproterenol binding. During the time course of the competitive binding experiments the membranes displayed isoproterenol-sensitive adenylate cyclase activity in the absence of added GTP. These data suggest that adenosine attenuates catecholamine-induced activation of adenylate cyclase via Ri receptors by decreasing the ability of beta-adrenergic agonists to promote the formation of a high affinity complex composed of the agonist, receptor and stimulatory guanine nucleotide binding protein.

A high performance liquid chromatographic method has been developed to separate isozymes of cyclic nucleotide phosphodiesterase. The method employs a polymer-based anion exchange column eluted with a sodium chloride gradient. Compared to traditional chromatography over DEAE-cellulose, the method is more rapid (30 min), dilutes the sample less, achieves better resolution of kinetically distinct forms, may be applied to as little as 200 micrograms of tissue protein and is appropriate for analytical use.

Somatostatin administration to female rats increased the activity of calmodulin-dependent soluble phosphodiesterase, both in pituitary and brain. This effect was also seen in homogenates of GH4C1 cells pretreated with the hormone. When assayed in the presence of EGTA no differences in rat brain and pituitary phosphodiesterase were observed between controls and somatostatin-treated, but when assayed in the presence of calcium or calcium plus calmodulin a clear increase in the activity of the enzyme was detected. In GH4C1 homogenates prepared from somatostatin-pretreated cells there was an increase in phosphodiesterase activity assayed in the presence of EGTA vs non-treated controls, which was more clear when assayed in the presence of calcium or calcium plus calmodulin. These observations suggest that somatostatin effects derive, at least in part, from increased cyclic nucleotide degradation.

In the present study the protein kinase activity was determined in surgical specimens of squamous cell carcinoma from the upper aero-digestive tract, and the activity was compared with the activity in normal mucosa obtained from the same location in patients undergoing surgery for non-neoplastic diseases. The basal protein kinase activity in the soluble fraction was about 15-fold higher in tumors as compared to in normal mucosa. The difference between protein kinase activity in the presence of Ca2+ and presence/absence of phosphatidylserine/diolein was taken as the activity of protein kinase C. Protein kinase C activity in the presence of 0.5 mM Ca2+ was 10.0 +/- 2.1 pmol 32P/mg prot. x min in tumors (n = 19) and 2.4 +/- 0.7 pmol 32P/mg prot. x min. in normal mucosa (n = 6). A similar difference was obtained with 1 mM Ca2+. An even greater difference in protein kinase C activity was seen when the soluble enzyme had been partially purified by ion-exchange chromatography on DE-52 columns. In this case a 7-fold higher protein kinase activity was found in tumors as compared to in normal mucosa. In the particulate fraction the basal protein kinase activity was about 3-fold higher in tumors as compared to in normal mucosa. Protein kinase C activity in the particulate fraction, defined as described above, was 10.7 +/- 4.1 pmol 32P/mg prot. x min in tumors and 6.1 +/- 4.4 pmol 32P/mg prot. x min in normal mucosa. These results show a significantly higher activity in both total protein kinase activity and protein kinase C activity in epithelial tumors from the upper aero-digestive tract.(ABSTRACT TRUNCATED AT 250 WORDS)

We recently described a novel isozyme of cGMP-dependent protein kinase (type I beta). It has a structure and peptide substrate specificity which is similar to that of type I alpha, but it has a different cGMP binding behavior, and autophosphorylation occurs almost entirely in serine instead of in both serine and threonine residues (Wolfe, L., Corbin, J.D., and Francis, S.H. (1989) J. Biol. Chem. 264, 7734-7741). An amino-terminal sequence of 31 amino acids derived from three proteolytic fragments of type I beta had 45% homology with a sequence beginning at type I alpha-47. However, sequences of three CNBr peptides of type I beta were identical to sequences of type I alpha beginning at type I alpha-202, -213, and -576 of 11, 27, and 30 residues. These sequences include portions of the catalytic domain and at least one cGMP-binding domain (site 1). Thus, types I alpha and I beta may be produced by alternative splicing of two unique mRNA segments to generate different amino acid sequences in the protein in a region that is amino-terminal to type I alpha-202. This segment in type I beta corresponds to the region in type I alpha that includes the major autophosphorylation site (Thr-58) which is within the domain that is proposed to inhibit catalytic activity. This region presumably interacts with the cGMP-binding site(s) to account for the differences in cGMP-binding behavior between types I alpha and I beta. Even though the sequence of type I beta in the variable region lacks the residue corresponding to Thr-58, it includes a consensus phosphorylation site (KRQAISA) beginning at type I alpha-59, which is absent in type I alpha. The results imply flexibility in the design of the autophosphorylation site and, hence, of the inhibitory domain.

A casein kinase II that is tightly bound to yeast ribosomes was partially purified and used to phosphorylate YL 44 and an unidentified 36 kDa protein in purified ribosomes. At typical cytosolic salt concentrations the phosphorylation was strongly stimulated by moderate concentrations (200 microM) of spermine or spermidine. The lowest effective concentration of spermine (20 microM, causing less than 50% stimulation) was close to that of total spermine reported in nongrowing yeast. Increases in free polyamines accompanying the 10-fold increase in total spermine and spermidine in growing yeast may therefore significantly stimulate this phosphorylation.