Microtubules prepared from chick brain homogenates by successive cycles of assembly-disassembly were found to contain two high-molecular-weight proteins, designated microtubule-associated protein, and microtubule-associated proteins. Microtubule-associated protein2 (apparent molecular weight 300,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis) was the preferred substrate for an endogenous cyclic AMPdependent protein kinase which appeared to be an integral component of the microtubules. The initial rate of phosphorylation of microtubule-associated protein2 was enhanced 4-to 6-fold by cyclic AMP, with half-maximal stimulation occurring at 2 X 10-7 M cyclic AMP. Under optimal conditions, a total of 1.0 and 1.9 mol of phosphate was incorporated per mole of microtubule-associated protein2, in the absence and presence of cyclic AMP, respectively. Cyclic AMP also stimulated the phosphorylation of tubulin, but the rate of phosphate incorporation per mol of tubulin was only 0.15% that of microtubuleassociated protein2. The data raise the possibility that the cyclic AMP-dependent phosphorylation of microtubuleassociated protein2 may play a role in microtubule assembly or function.It has been reported that cyclic AMP stimulates the phosphorylation of tubulin, the major microtubule protein (1, 2). Moreover, microtubule preparations have been found to contain a cyclic AMP-dependent protein kinase that can catalyze the phosphorylation of exogenous substrates (3-5). When microtubule protein is isolated by in vitro assemblydisassembly methods (6, 7) and the proteins are analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, two proteins of high molecular weight are found to co-purify with the tubulin (8-11). In this report, evidence is presented showing that such microtubule preparations contain a cyclic AMP-dependent protein kinase whose preferred substrate is the faster migrating of the two high molecular weight proteins that co-purify with the tubulins. This endogenous nontubulin substrate incorporates stoichiometric amounts of phosphate under the experimental conditions studied. MATERIALS AND METHODSPreparation of Microtubules. Microtubule protein was purified from brain homogenates of 1-to 3-day-old chicks by a modification of the method of Shelanski et al. (12 glycerol. The reaction tubes were then transferred to a boiling water bath for 2 min, after which 25 Ml of 0.6 M dithiothreitol were added. Aliquots (75 Ml) were subjected to electrophoresis on 5.6% polyacrylamide slab gels containing 1% sodium dodecyl sulfate (14). The gels were stained with Coomassie blue and dried. Autoradiography was carried out as described (15,16). The amount of 82p incorporated into the individual protein bands was measured either by liquid scintillation counting of gel slices (Figs. 3 and 4) or by scanning the auto-

The increase in human leukocyte adenosine 3':5'-cyclic monophosphate (cyclic AMP) levels seen in response to various substances was markedly potentiated by colchicine and other agents that affect microtubule assembly. Addition of dI-isoproterenol (2 AiM) or prostaglandin El (10 ttM), together with the phosphodiesterase inhibitor isobutylmethylxanthine (1 mM), caused a much greater increase in cyclic AMP in colchicine-pretreated cells tan in control cells. With isoproterenol (2 MM) plus isobutylmethylxanthine (1 mM), cyclic AMP levels rose about 3fold but, in combination with colchicine, these drugs caused a more than 15-fold increase in cyclic AMP. The effects of colchicine were both time-and dose-dependent; they could be seen'within 1 min after addition of colchichme or at concentrations as low as 10 nM. In addition to its potentiation of hormonally induced increases in cyclic AMP levels, colchicine also potentiated the effect of isobutylmethylxanthine alone on leukocyte cyclic AMP levels. Vinblastine, vincristine, podophyllotoxin, and oncodazole all had effects similar to those of colchicine but blmicolchicine did not. The data suggest that cytoplasmic nticrbtubules interact with the leukocyte plasma membrane to impose constraints on the expression of hormone-sensitive adenylate cyclase; the therapeutic effects of colchicine may depend in part upon the relaxation of such constraints. Moreover, the synergism described here between colchicine-like agents and hormones is of potential therapeutic importanceiln clinical conditions in which either alkaloid or hormone has been useful separately. IThe cyclic nucleotide levels of human leukocytes (both granulocytes and mononuclear cells) are controlled by a number of hormones and pharmacological agents (1-5). f3-Adrenergic agonists (isoproterenol, epinephrine), histamine, and prostaglandin E1 (PGE1) all raise adenosine 3':5'-cyclic monophosphate (cyclic AMP) levels in these cells; the muscarinic cholinergic agonists (acetylcholine, carbamylcholine) raise cyclic GMP levels. In granulocytes, a functional significance of cyclic nucleotides has been suggested by experiments showing that increases in cyclic AMP inhibit the degranulation of lysosomes that normally accompanies phagocytosis, whereas increases in cyclic GMP enhance degranulation (2, 3, 5, 6). Colchicine and vinblastine, agents that cause the disappearance of cytoplasmic microtubules by preventing their assembly (7-10), also inhibit degranulation (6, 11-13); it is through this inhibition of microtubule assembly, with consequent effects on microtubuleassociated functions, that colchicine is thought to exert its therapeutic anti-inflammatory action in acute gouty arthritis and other disorders (14). We now report that colchicine and other agents that interfere with microtubule assembly increase cyclic AMP levels in human leukocytes and potentiate the effects of hormones on cyclic AMP levels in these cells. METHODSLeukocytes were obtained from freshly drawn, heparinized blood from healthy, adult donors by de...

8-Azidoadenosine 3',5'-monophosphate (8-N3-cAMP) containing 32P has been used as a photoaffinity label specific for the adenosine 3',5'-monophosphate (cAMP) binding site(s) present in a partially purified preparation of soluble protein kinase from bovine brain. 8-N3-cAMP and cAMP were found to compete for the same binding site(s) in this preparation, as determined by a standard filter assay. When this protein preparation was equilibrated with [32P]-8-N3-cAMP, and then irradiated at 253.7 nm, the incorporation of radioactivity was predominantly into a protein with an apparent molecular weight of 49,000, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography. This labeled protein comigrated in the gel with the only protein which is endogenously phosphorylated by [gamma-32P]ATP, a protein which has been shown to be the regulatory subunit of the protein kinase (H. Maeno, P. L. Reyes, T. Ueda, S. A. Rudolph, and P. Greengard (1974), Arch. Biochem. Biophys. 164, 551). The incorporation of [32P]-8-N3-cAMP into this protein was half-maximal at a concentration of 7 x 10(-8) M. In accordance with a proposed mechanism involving the formation of a highly reactive nitrene intermediate upon irradiation of the azide, the incorporation of radioactivity into protein was maximal within 10 min of irradiation, and was almost eliminated by preirradiation of the photolabile ligand. Moreover, this incorporation was virtually abolished by a 50-fold excess of cAMP, but not by AMP, ADP, ATP, or adenosine. We suggest that 8-N3-cAMP may prove to be a useful molecular probe of the cAMP-binding site in receptor proteins and report its use in conjunction with sodium dodecyl sulfate-polyacrylamide gel electrophoresis as a highly sensitive and selective radiochemical marker for cAMP-binding proteins.

Normal human T lymphocytes incubated with adenosine (10 pAM) for 30 min at 37C show an increase in the percentage of cells expressing receptors for the Fc portion of IgG (RFc,) and the OKT8 antigen, while the proportion ofOKT4' cells decreases. These effects occur exclusively in a subset of T cells with theophylline-resistant sheep erythrocyte receptors (TR cells (2)(3)(4)(5).Although adenosine receptors have not been characterized in human T lymphocytes, many studies suggest that the pharmacologic action ofadenosine on cells requires binding ofadenosine to specific membrane receptors (6, 7). The occupied receptors activate adenylate cyclase and it is the elevated cAMP levels that mediate the transduction ofadenosine effects. Adenosine receptors capable of activating adenylate cyclase have been termed R sites and require that the ribose ring of adenosine be unmodified. R-site adenosine receptors are located on the outer surface of the cell membrane. In contrast, a second type of adenosine receptor inhibits adenylate cyclase activity. These receptors, termed P sites, require an unmodified purine ring and are thought to be located on the inner surface of the cell membrane.Systemic lupus erythematosus (SLE) is an autoimmune disease associated with impaired suppressor T-cell function during active disease. Multiple defects in immunoregulatory mechanisms have been found in SLE by using a variety of in vitro assay systems (8, 9). In particular, although concanavalin A-treated normal T cells suppress both cellular and humoral immune responses in vitro, SLE T cells do not develop suppressor activity in response to concanavalin A (8). Similarly, SLE T-helper cells do not develop suppressor activity or show changes in the proportions of cells expressing OKT4, OKT8, and RFcl, when treated with adenosine (ref. 10; unpublished results).We have investigated both the immunologic and the pharmacologic events that occur when normal and SLE T-helper/ inducer lymphocytes are treated with adenosine. Our data suggest that some of the immunoregulatory defects observed in SLE T lymphocytes may be due to abnormalities in adenosine receptor-mediated cAMP metabolism.MATERIALS AND METHODS SLE Patients and Normal Controls. Nine individuals ranging from 19 to 40 years old, with established diagnoses of SLE, were studied. The diagnosis of active SLE was confirmed clinically and serologically at the time of study and was based on the presence of (i) four or more criteria for the classification of SLE (11) The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.