Folic Acid as Second Chemotactic Substance in the Cellular Slime Moulds

IT has been known for some time that in certain species of cellular slime moulds acrasin, the substance which attracts the amoebae to central collection points during the aggregation phase, is cyclic AMP^1–4. We were also able to show that E. coli gave off another substance besides cyclic AMP (henceforth referred to as bacterial factor, or BF) which attracted the vegetative amoebae of Dictyostelium discoideum⁵. Here we demonstrate that this second attractant has the properties of folic acid or one of its derivatives. We also show that folic acid and related compounds not only attract the vegetative amoebae of D. discoideum (No. NC-4H) but also the amoebae of six other species (Dictyostelium rosarium No. CC-7; D. mucoroides No. 11; D. purpureum No. 2; D. minutum No. V-3; Polysphondylium violaceum No. 1; P. pallidum No. 2). For the latter three species cyclic AMP is not the aggregative attractant (ref. 6 and J. T. B., E. M. H., S. Noller, F. B. Oleson and A. B. Roberts, in preparation) which raises the interesting question of whether their acrasin might be related to the folates.     

Evidence for a Second Chemotactic System in the Cellular Slime Mold, Dictyostelium discoideum

An unknown substance found in bacteria (Escherichia coli) is especially effective in attracting the vegetative amoebae of the cellular slime mold, Dictyostelium discoideum. However, the aggregating amoebae are not attracted to it at all. On the other hand, the vegetative amoebae show very little chemotactic response to cyclic adenosine monophosphate (cyclic AMP), whereas the aggregating amoebae are exceptionally responsive to it. It is suggested that the new factor may be used in food seeking, whereas cyclic AMP, the chemotactic substance responsible for aggregation, is the acrasin of this species. The important point is that the amoebae are differentially stage-specific in their responses to these two chemotactic agents.     

Identification of the acrasin of Dictyostelium minutum as a derivative of folic acid

The acrasin of the slime mold Dictyostelium minutum was isolated from aggregating cells and purified. The compound was species specific and more active in the aggregative than in the vegetative stage. Three observations strongly suggest a structural relationship between the acrasin and folic acid. (1) Folic acid inhibited acrasin degradation by D. minutum. (2) Methotrexate, an antagonist of chemotaxis towards folic acid, also inhibited the response to the acrasin. (3) The chemotactic response to an excess of folic acid was delayed. The response was also delayed to simultaneously tested low amounts of a related compound, but not to unrelated compounds (Van Haastert, 1982). The response to the acrasin was observed to be delayed by excess of folic acid. The acrasinase was identified as a folic acid C9-N10 splitting enzyme. Based on chromatographic properties and biological activity of the acrasin and folate derivatives, the chemical structure of the acrasin is discussed.     

Allosteric regulation of phosphodiesterase from Portulaca callus by cGMP and papaverin

Three fractions of phosphodiesterase activity capable of hydrolysing cyclic 3′,5′-AMP and cyclic 3′,5′-GMP were purified from Portulaca callus. Hydrolysing bis-(p-nitrophenyl)-phosphate, two fractions showed linear Lineweaver-Burk plots. One fraction showed positive cooperativity. This fraction can be activated competitively by blue dextran, indicating a possible allosteric regulation by nucleotides, demonstrated by changing from being positively cooperative, to following Michaelis-Menten kinetics by cGMP and papaverin. cGMP triggers an enzyme highly active against 3′,5′cAMP and 3′5′cGMP, and papaverin triggers high activity against 2′,3′cAMP, demonstrated by two separate enzyme fractions.     

Discovery of a 3′,5′-guanosine monophosphate simulating factor in amoebae of dictyostelium discoideum

An enzyme degradable substance in an amoebae extract from Dictyostelium discoideum has been observed to exhibit properties of 3′,5′-guanosine monophosphate (cyclic GMP). This is the simplest differentiating system yet shown to contain cyclic GMP.     

Microinjection of cyclic nucleotides provides evidence for a diffusional mechanism of intraneuronal control

Cyclic 3′,5′-AMP and cyclic 3′,5′-GMP injected into large neurons of the snail Helix lucorum altered neuron activity. The effect of cAMP is usually depolarizing and that of cGMP hyperpolarizing. The results are specific for 3′,5′-cyclic nucleotides. The experiments support the hypothesis that reaction-diffusion processes involving cyclic nucleotides from the basis of an intraneuronal system of information processing.     

Human complex-forming glycoprotein, heterogeneous in charge: the primary structure around the cysteine residues and characterization of a disulfide bridge

L-929 cell surface membranes have been assayed in vitro and found to contain significant protein kinase activity. A steady-state kinetic analysis indicated that at least two distinct protein kinases were present. Plots of reaction velocity (v) against substrate (ATP) concentration were distinctly biphasic, as were Lineweaver-Burk plots of $\text{1}\text{v}$ versus $\text{1}\text{ATP}$. Michaelis constants of the two enzymes were calculated to be 22 and 173 μm, respectively. Sodium dodecyl sulfate polyacrylamide gel analysis of the phosphorylated membrane proteins provided additional support for the existence of more than one protein kinase. Different endogenous proteins were phosphorylated at 1 μm ATP compared to 1 μm ATP. Further studies of the low K_m (22 μm) enzyme suggested that it is a typical cyclic 3′,5′-AMP-independent protein kinase. Its activity was dependent on the presence of Mg²⁺, but it was not affected by cyclic 3′,5′-AMP, cyclic 3′,5′-GMP, or the heat-stable inhibitor of cyclic 3′,5′-AMP-dependent protein kinases. ATP and GTP, but not other nucleoside triphosphates, could serve as phosphoryl donor and maximum kinase activity was expressed at pH 7.0. Phosvitin and casein were superior to histones as exogenous substrates for the low K_m enzyme.     

Studies on cyclic adenosine 3' ,5'-monophosphate levels, Adenylate cyclase and phosphodiesterase activities in the dimorphic fungus Mucor rouxii.

The germination of spores of Mucor rouxii into hyphae was inhibited by 2 mm dibutyryl cyclic adenosine 3′,5′-monophosphate or 7 mm cyclic adenosine 3′,5′-monophosphate; under these conditions spores developed into budding spherical cells instead of filaments, provided that glucose was present in the culture medium. Removal of the cyclic nucleotides resulted in the conversion of yeast cells into hyphae. Dibutyryl cyclic adenosine 3′,5′-monophosphate (2 mm) also inhibited the transformation of yeast to mycelia after exposure of yeast culture to air.Since in all living systems so far studied adenylate cyclase and cyclic adenosine 3′,5′-monophosphate phosphodiesterase are involved in maintaining the intracellular cyclic adenosine monophosphate level, the activity of both enzymes and the intracellular concentration of cyclic adenosine monophosphate were investigated in yeast and mycelium extracts. Cyclic adenosine monophosphate phosphodiesterase and adenylate cyclase activities could be demonstrated in extracts of M. rouxii. The specific activity of adenylate cyclase did not vary appreciably with the fungus morphology. On the contrary, cyclic adenosine monophosphate phosphodiesterase activity was four- to sixfold higher in mycelial extracts than in yeast extracts and reflected quite accurately the observed changes in intracellular cyclic adenosine monophosphate levels; these were three to four times higher in yeast cells than in mycelium.     

The induction of filopodia in the cellular slime mold Dictyostelium discoideum by cyclic adenosine monophosphate: mechanism of aggregation.

Dictyostelium discoideum vegetative amoebae grown axenically can be induced to extend microprojections, filopodia, in response to cyclic 3′,5′-adenosine monophosphate. Cyclic 3′-5′-guanosine monophosphate, adenosine monophosphate, or adenosine diphosphate at concentrations of 1.0 mm have no effect. After incubation for 15 min, 1.0 mM adenosine triphosphate will also cause filopodial formation. Treatment with 0.1 mM 2–4 dinitrophenol or 1.0 mM sodium azide does not prevent the induction by cyclic adenosine monophosphate. The induced cells can be more extensively agglutinated with Concanavalin A at 0.5 mg/ml than noninduced cells. A model is presented that describes a possible mechanism whereby cells may aggregate via the cyclic adenosine monophosphate induced filopodia.     

The adnosine 3',5'-monophosphate and guanosine 3',5'-monophosphate phosphodiesterases from human lung tissue.

Human lung tissue contains phosphodiesterase enzymes capable of hydrolyzing both adenosine 3′,5′-monophosphate (cyclic AMP) and guanosine 3′,5′-monophosphate (cyclic GMP). The cyclic AMP enzyme exhibits three distinct binding affinities for its substrate (apparent Km = 0.4μM, 3μM, and 40μM) while the cyclic GMP enzyme reveals only two affinities (Km = 5μM and 40μM). The pH optima for the cyclic AMP and cyclic GMP phosphodiesterase are similar (pH 7.6–7.8). Both are inhibited by known inhibitors of phosphodiesterase activity (aminophylline, caffeine, and 3-isobutyl-1-methylxanthine). The divalent cations Mg²⁺ and Mn²⁺ stimulate cyclic AMP phosphodiesterase activity (in the absence of Mg²⁺) while Ca²⁺, Ni²⁺, and Cu²⁺ inhibit the enzyme. Histamine and imidazole slightly stimulate cyclic AMP hydrolytic activity. Thus, human lung tissue does contain multiple forms of both the cyclic AMP and cyclic GMP phosphodiesterase which are influenced by a variety of effectors.     

Cytochemical localization of adenylate cyclase and 3',5'-nucleotide phosphodiesterase in Dictyostelium.

Cytochemical localizations of adenylate cyclase and 3′,5′-nucleotide phosphodiesterase were performed on aggregating Dictytostelium discoideum myxamoebae. The adenylate cyclase reaction product was localized on the inner surface of the plasma membrane. The phosphodiesterase reaction product was localized on the outer surface of the plasma membrane. Differences in enzyme activity were noted according to the state of cell (isolated or aggregated) and according to the cell position in larger aggregates. Heavy precipitation indicative of adenylate cyclase activity was not observed in isolated amoebae, but was often observed in streams and in some cells of aggregates. The precipitation indicative of phosphodiesterase activity could be found in isolated amoebae and in peripheral cells of aggregates.     

5′-Nucleotidases of retinal rod membranes

5′-Nucleotidase (EC 3.1.3.5) was solubilized from rod membranes with Ammonyx LO and purified by chromatographic methods. A highly sensitive radioassay was developed. The purified enzyme behaved as a homogeneous protein of 75,000 daltons in sodium dodecyl sulfate-polyacrylamide gel electrophoresis and as a protein of 79,000 in gel filtration. Thus, the enzyme does not contain subunits. The K_m values obtained were 1.3 μm for 5′-AMP and 2.3 μm for 5′-GMP. The enzyme was inhibited by concanavalin A, wheat germ agglutinin, and Ricinus communis agglutinin. Rabbit muscle G-actin formed a complex with the enzyme and inhibited its activity. The catalytic site of the enzyme was localized on the internal surface of the disk which, in terms of membrane sidedness, corresponds to the cell surface. A soluble 5′-nucleotidase was extracted from rod membranes with Tris buffer (pH 8.0) containing EGTA in the dark; less enzyme was extracted if the membranes had been exposed to light or incubated with Ca²⁺. The extracted enzyme was partially purified. The enzyme was unstable and lost 50% of its activity in 3 days at 3 °C. The K_m values were 1.3 μm for 5′-AMP and 2.3 μm for 5′-GMP. The enzyme was inhibited by G-actin. A role for the soluble enzyme in the regulation of 5′-GMP in the rod outer segment was suggested.     

Cyclic 3',5'-adenosine monophosphate stimulates trehalose degradation in baker's yeast

The levels of cyclic 3′,5′-AMP and trehalose, as well as the specific activity of the trehalase have been investigated in cells of baker's yeast ($\text{Saccharomyces cerevisiae}$) during the lag phase preceding growth. During the first few minutes a substantial increase in the intracellular concentration of cyclic 3′,5′-AMP was observed, followed by a 6–8 fold increase in trehalase activity concomitant with the rapid degradation of trehalose. Cell free extracts prepared from resting yeast were shown to contain a cryptic trehalase, which under physiological conditions could be activated by cyclic 3′,5′-AMP to the same degree as $\text{in vivo}$. These observations suggest that in the lag phase of growth, the level of trehalose in baker's yeast is under control of a system, regulated by the level of cyclic 3′,5′-AMP.     

Cyclic guanosine 3',5'-monophosphate and phosphodiesterase activity in mitogen-stimulated human lymphocytes.

The cyclic adenosine 3′,5′-monophosphate (cyclic AMP) phosphodiesterase from human leukemic lymphocytes differes from the normal cell enzyme in having a much higher activity and a loss of inhibition by cyclic guanosine 3′,5′-monophosphate (cyclic GMP). In an effort to determine the mechanism of these alterations, we have studied this enzyme in a model system, lectin-stimulated normal human lymphocytes. Following stimulation of cells with concanavalin A (con A) the enzyme activity gradually becomes altered, until it fully resembles the phosphodiesterase found in leukemic lymphocytes. The changes in the enzyme parallel cell proliferation as measured by increases in thymidine incorporation into DNA. The addition of a guanylate cyclase inhibitor preparation from the bitter melon prevents both the changes in the phosphodiesterase and the thymidine incorporation into DNA. This blockage can be partially reversed by addition of 8-bromo cyclic guanosine 3′,5′-monophosphate (8-bromo cyclic GMP) to the con A-stimulated normal lymphocytes. These results indicate a possible role of cyclic GMP in a growth related alteration of cyclic AMP phosphodiesterase.     

Partial characterization of 5′(3′)-ribonucleotide and 3′-ribonucleotide phosphohydrolases from Tradescantia albiflora leaves

Two acid phosphomonoesterases, 5′(3′)-ribonucleotide phosphohydrolase and 3′-ribonucleotide phosphohydrolase, were isolated from Tradescantia albiflora leaf tissue and purified by ammonium sulphate precipitation, gel filtration on Sephadex G-200 and repeated chromatography on DEAE-cellulose. The enzymes differed in their sensitivity to dialysis against 1 mM EDTA; the activity of 5′(3′)-ribonucleotide phosphohydrolase was unaffected, while 3′-ribonucleotide phosphohydrolase showed an increase of 60–90%. Both enzymes were rapidly inactivated above 50°. Their ion sensitivity was identical: 1 m M Zn²⁺ and Fe²⁺ were inhibitors for both by 20–80%; while Mg²⁺, Ca²⁺, Co²⁺, K⁺, Na⁺ at 1–10 mM had no significant effect on the activity of either enzyme. Inorganic phosphate inhibited both enzymes almost completely. EDTA (1 mM) did not inhibit either enzyme; none of the divalent cations tested were enzyme activators. 3′-Ribonucleotide phosphohydrolase hydrolysed both 3′- and 5′-nucleoside monophosphates (3′-AMP, 3′-CMP, 3′-GMP, 3′-UMP, 5′-AMP, 5′-CMP, 5′-GMP, 5′-UMP). 5′(3′)-Ribonucleotide phosphohydrolase showed a preference for the 3′-nucleoside monophosphates. Adenosine 3′,5′-cyclic monophosphate, purine and pyrimidine 2′,3′-cyclic mononucleotides at 0.1–1.OmM did not inhibit the enzymes.     

Activation of murine lymphocytes by cyclic guanosine 3′,5′-monophosphate: Specificity and role in mitogen action

Cyclic guanosine 3′,5′-monophosphate (cyclic GMP) stimulates nucleic acid synthesis in lymphocytes, and has been implicated as the intracellular effector of the actions of mitogenic agents on these cells. In the present study, we examined the specificity of the mitogenic activity of cyclic GMP and of its 8-bromo (Br) derivatives, and the effects of the T cell mitogens, concanavalin A, phytohemagglutinin, and staphylococcal entertoxin B (SEB) on the cyclic GMP content and guanylate cyclase activity of mouse splenic lymphocytes. Cyclic GMP and guanosine modestly increased the incorporation of [³H]thymidine into DNA by cultured lymphocytes, but were far less effective than their 8-Br-derivatives. However, on a molar basis the mitogenic activity of both 8-Br-guanosine and 8-Br-5′-GMP exceeded that of 8-Br-cyclic GMP, when tested in the presence and absence of serum in the culture media. Combined addition of maximal doses of these nucleotides did not give additive stimulatory effects, suggesting an action on a common subpopulation of cells, and possibly a common mechanism. By contrast, cyclic AMP, 8-Br-cyclic AMP, 8-Br-adenosine, cholera toxin and prostaglandin E₁ suppressed both basal [³]thymidine incorporation and stimulation of this parameter by T-cell line mitogens and the guanosine nucleotides. Rapid effects of concanavalin A, phytohemagglutinin, SEB, guanosine, 5′-GMP, 8-Br-guanosine, and 8-Br-5′-GMP on the cyclic GMP content of murine lymphocytes could not be demonstrated. Similarly, concanalin A, phytohemagglutinin and SEB failed to alter guanylate cyclase activity when added directly to cellular homogenates or pre-incubated with intact cels. Conversely, carbamylcholine rapidly increased lymphocyte cyclic GMP but was not mitogenic.These results are consistent with the hypothesis that cyclic GMP and cyclic AMP are antagonistic in their influence on lymphocyte mitogenesis. However, they also demonstrate that related nucleotides are more potent mitogens than cyclic GMP and suggest that activation of murine lymphocytes by concanavalin A, phytohemagglutinin and SEB may not be mediated by rapid increases in cellular cyclic GMP content. Since high concentrations of exogenous cyclic GMP and related nucleotides must be used to influence DNA synthesis, the biologic significance of this effect remains uncertain.     

The Isolation and Characterization of the Acid-Soluble Nucleotides from the Tubers of Jerusalem Artichoke (Helianthus tuberosus L.)

The acid-soluble nucleotides were extracted from the tubers of Jerusalem artichoke with percbloric acid, and separated and purified by means of adsorption on and elution from active charcoal, repeated chromatography on columns of Dowex I (Cl^-), followed by paper chromatography. The following nucleotides have been characterized and/or identified: 5′-AMP, 3′-AMP, ADP, ATP, 5′-GMP, 2′-GMP, 3′-GMP, 2′,3′-cyclic GMP, GDP, GTP, 5′-UMP, UDP, UTP, NADP, UDP-glucose, UDP-galactose, UDP-fructose, UDP-N-acetylhexosamine and GDP-mannose.^** Neither cytosine ribonucleotides nor deoxyribonucleotides have been detected. The significance of these observations is discussed.     

Folic acid deaminase activity during development in Dictyostelium discoideum.

Folic acid attracts vegetative amoebae of Dictyostelium discoideum. Secreted by bacteria, it may act as a food-seeking device. The inactivation of this attractant is catalyzed by a deaminase. As assay has been developed to measure the folic acid deaminase activity. In addition to cell-surface an intracellular deaminase, the amoebae of D. discoideum release the enzyme into the medium. The pH optimum of the extracellular enzyme was 6.0, and higher for the cell-associated deaminases. The extracellular enzyme was secreted maximally by vegetative amoebae, and its activity diminished during cell differentiation. The cell-surface bound enzyme was less active than the extracellular enzyme, and its activity decreased twofold during a 6-h starvation period. The enzyme activity of homogenates and 48,000 x g pellets diminished during this period 35 to 40%. The supernatant of a homogenate had a higher deaminase activity than the homogenate itself or its pellet; this suggests the presence of an inhibitor in the particulate fraction. The underlying mechanism for inactivation of folic acid has similar characteristics as that for inactivation of cyclic adenosine monophosphate.     

Determination of the active portion of the folic acid molecule in cellular slime mold chemotaxis.

From earlier work it is known that folic acid attracts the amoebae of various species of cellular slime molds (11). Here we have tested a wide variety of pteridines, pyrimidines, and pyrazines to determine what part of the folic acid molecule is chemotactically active. It was shown that the activity lies in the pteridine ring itself. Furthermore, the cell-free supernatants of slime mold amoebae contain an enzyme that renders pterin and folic acid chemotactically inactive, which apparently increases the chemotactic sensitivity of the amoebae to those compounds. Despite the fact that slime mold amoebae secrete small amounts of folic acid-related compounds, there is no evidence that folates are acrasins; rather it is postulated that attraction to folates may be a food-seeking device for the amoebae which prey on folate-secreting bacteria in the soil.     

Properties and subcellular distribution of ribonuclease activity in Chlorella

RNase activity from Chlorella was partially purified. Two RNase activities were demonstrated, one soluble and the other ribosomal. The effects on ribonuclease activity of variations in pH and temperature, and of Mg²⁺, Na⁺, and mononucleotides were examined. The RNase activities (phosphodiesterases EC 3.1.4.23) were both endonucleolytic, releasing oligonucleotides, and cyclic nucleotide intermediates, but exhibited different specificities in releasing mononucleotides from RNA. The ribosomal activity released 3′-GMP, and after prolonged incubation 3′-UMP, but the soluble activity released 3′-GMP, 3′-AMP and 3′-UMP. Neither ofthe RNase preparations hydrolysed DNA, nor released 5′-nucleotides from RNA. Increased ribosomal RNase activity was related to dissociation of ribosomes, and latency of ribosomal RNase activity was demonstrated. The possible in vivo distribution of RNases is discussed.