A monovalent cation-sensitive actin-binding factor in a myeloid leukemia cell line

Cell extracts of a murine leukemia cell line, M1, apparently contain three kinds of actin-gelation factors; a filamin-like protein, and 38K-dimer and 105K-dimer proteins. Unlike gelation by the filamin-like protein, gelation by the latter two proteins is inhibited by low concentrations of KCl. Our study of the 38K protein has been reported elsewhere (Takagi, K. et al., J. Biochem. Tokyo 97, 605-616, 1985). We here describe the purification and characterization of the 105K protein. The 105K protein differs from the alpha-actinin group of proteins in its antigenicity, peptide components and Ca2+-insensitivity. The saturated binding ratio of the protein to purified skeletal muscle actin is 1:8; when this ratio exceeds 1:20, gelation takes place. This gelation is inhibited completely by the presence of 25 mM KCl. Electron microscopy revealed that, in the absence of KCl, the 105K protein/actin mixture forms short actin bundles that are accompanied by a meshwork of short single filaments. The presence of 25 mM KCl did not prevent actin-bundling, but the bundles became longer and the meshwork of short filaments was no longer present.     

Changes in actin-related gelation of crude cell extracts during differentiation of myeloid leukemia cells

Gelation of extracts of a myeloid leukemia cell line (Ml) was compared before and after differentiation induced with conditioned medium (CM) from rat embryo cells. Although an extract of Mml cells, a macrophage line derived from Ml line, gelled when warmed in the presence of 2 mM MgCl2, undifferentiated Ml cells gelled only after dialysis and a supplement of exogenous actin. After differentiation had been induced, an addition of exogenous actin, but not dialysis, was needed for gelation. Small amounts of KCl always inhibited the gelation of the control Ml cell extracts, but they promoted gelation of the CM-treated Ml and Mml cell extracts. Thus, the dialysis required for gelation of the control Ml cell extract appears to be necessary for the exclusion of endogenous KCl. Several possible mechanisms for the KCl control of gelation, as well as different requirements of exogenous actin needed for gelation are discussed based on the results of our experiments.     

Characterization of "abp38" in cytoskeleton of mouse myeloid leukemia cells and distribution of abp38 in various tissues

A specific antibody was prepared against "abp38," a 38 kDa-dimer protein purified from mouse myeloid leukemia cells (M1 cells), that induce gelation of actin filaments in a K+-dependent manner. Immunochemically, the total content of abp38 in undifferentiated M1 cells was found to be 0.89% of the total protein (10.7 micrograms/10(7) cells). This content increased about 8-fold (89.5 micrograms/10(7) cells) after the M1 cells had differentiated into macrophages. When the cells before differentiation were extracted with Triton solution containing 150 mM KCl, almost all abp38 in the cytoskeleton was removed, whereas in cells after differentiation, amount of abp38 remaining in the cytoskeleton was 4.5 micrograms per 10(7) cells. The amount of cytoskeleton-bound abp38 of M1 cells and mouse peritoneal macrophages decreased with increase in K+ concentration in the extraction solution. Immunoreactive molecules against abp38 antibody were present in various tissues and cultured cell lines except for skeletal muscle and erythrocytes. Furthermore, actin binding protein with a molecular size of 38 kDa was found in bovine brain. These data suggest that abp38 is a ubiquitous protein present in various tissues and species.     

Changes in contractile proteins during differentiation of myeloid leukemia cells. I. Polymerization of actin

Quantitative and qualitative changes in cellular actin were followed during differentiation of a myeloid leukemia cell line, namely Ml, which was inducible with conditioned medium (CM). During 3 d of incubation with CM, when the Ml cells differentiated to macrophages and lost their mitotic activity, the actin content, F-actin ratio in total actin, and the actin synthesis showed an increase. A greater difference before and after differentiation was found in the ability of G-actin to polymerize. Actin harvested from CM-treated cells showed a greater ability to polymerize, depending on the increased concentration of MgCl2 and/or KCl and proteins, as compared with the actin from untreated Ml cells. Actin harvested from the Mml cell line, a macrophage line, had a particularly high polymerizability with or without CM treatment. In contrast, the actin from the D- subline, which is insensitive to CM, showed almost no polymerization.     

Changes of phospholipase A2 inhibitory activity in the K+-sensitive actin gelation factor during the differentiation of myeloid leukemia cells

The phospholipase A2 inhibitory activity of a 38 kDa K+-sensitive actin gelation factor in a murine leukemia cell line (M1) was examined. A specific antibody against 38 kDa protein was found to cross-react with 37 kDa protein (lipocortin) in rat peritoneal exudates. Although the native 38 kDa protein from M1 cells did not block phospholipase A2 activity, pretreatment with alkaline phosphatase produced a form that did inhibit this enzyme. However, a purified 38 kDa protein from differentiated M1 cells blocked phospholipase A2 activity without pretreatment with alkaline phosphatase. Phospholipase A2 inhibitory activity of the 38 kDa protein was not altered by addition of actin. These findings suggest that the phospholipase A2 inhibitory of our 38 kDa protein was induced during differentiation. We also proposed that our 38 kDa protein has the same epitope as lipocortin.     

Changes in myosin during differentiation of myeloid leukemia cells

Changes in cellular myosin were followed during the differentiation into macrophages of a myeloid leukemia cell line (Ml) which can be induced by conditioned medium (CM) from a rat embryo culture. To extract the myosin, we used three different procedures, all of which gave a lower yield of myosin for the differentiated than for the undifferentiated Ml cells. This low extractability we attributed to increased binding of the myosin to the plasma membrane. Taking the different extractabilities into consideration, we calculated the myosin contents in the total cellular protein from the densitometry of SDS-polyacrylamide electrophoresis, 0.6% for the untreated Ml cells and 1.0% for the differentiated ones. The three ATPase activities of the Ml cell myosin were in the order, K+-EDTA-=Ca2+- much greater than Mg2+-ATPase in the presence of 0.6 M KCl, whether or not there was treatment with CM. Myosin was purified through fractionation with 25-55% saturated ammonium sulfate, then gel filtration with Sepharose 4B followed by affinity chromatography on F actin-Sepharose 4B. The Ml cell myosin consists of 1 heavy chain (H) and 3 light chains (L1, L2, L3), with molecular ratios of L1 + L2/H not equal to and L3/H not equal to 1. The ratio of L1/L2 was about 1.2 for the untreated Ml cells, but it decreased to about 0.7 after differentiation.     

Reversible induction of actin rods in mouse C3H-2K cells by incubation in salt buffers and by treatment with non-ionic detergents

Incubating conditions which induced actin paracrystal-like intracellular structures (actin rods) were investigated by using several cell lines. We have found that an incubation of cells of a mouse fibroblastic cell line, C3H-2K, in an isotonic solution of NaCl containing 1 mM MgCl2, 1 mM CaCl2 and 10 mM MES, pH 6.5, induced disintegration of stress fibers and formation of actin rods in the cytoplasm. Actin rods were induced also by incubating in salt buffers in which Na+ of the above solution was substituted by most cations except K+ or Rb+. When the actin rod-forming cells were transferred back to DMEM containing 10% FBS, actin rods disappeared and stress fibers subsequently re-formed within 1 h at 37 degrees C. Although the induction was observed in NaCl buffer at a wide range of pH values (5.5-10), the optimal pH was 6.5. Formation of actin rods is dependent upon cellular metabolism, as it was inhibited at 4 degrees C, or by metabolic inhibitors. Incubation in NaCl buffer induced actin rods in HeLa, L, NRK, BALB/c 3T3 and Swiss 3T3 cells, but not in CEF or MEF cells. A decrease in cell volume was observed parallel with the induction of actin rods, except for CEF and MEF cells. Alterations in intracellular concentrations of Na, K or Ca were not correlated with the induction, however. Actin rods were also induced in C3H-2K cells by a brief treatment with non-ionic detergents. Tween 80 at concentrations as low as 0.003% was effective for the induction, but did not increase the passive membrane transport of p-nitrophenylphosphate. In contrast to the induction by NaCl buffer, treatment with Tween 80 induced numerous tiny actin rods at 4 degrees C, which became larger when further incubated at 37 degrees C. Double immunofluorescence staining with anti-actin antibody and anti-vinculin antibody showed that vinculin plaques remained at least in an early stage of the actin rod formation. We discuss the mechanism for the induction of actin rods based upon the present findings.     

Lithium fluxes in Paramecium and their relationship to chemoresponse.

Paramecia respond to environmental stimuli by altering swimming behavior to disperse from or accumulate in the vicinity of the stimulus. We have found, using the T-maze assay, that treatment of paramecia with LiCl in a time- and concentration-dependent manner modifies the normal response to folate, acetate, and lactate from attraction to no response or even repulsion. Responses to NH4Cl were unaffected and to cAMP were variably affected by LiCl. Cells incubated in the presence of K+, or both Na+ and K+, but not Na+ alone reliably recovered attraction to folate. Treatment of cells with 4 mM LiCl for 1 h dramatically slowed swimming speed from about 1 mm/s in NaCl or KCl (control) to 0.18 mm/s in LiCl. Li-treated cells subsequently incubated in 4 mM NaCl, KCl or sequentially in KCl and NaCl for a total of 20 min increased their swimming speed to 0.35, 0.45 and 0.67 mm/s, respectively. Paramecia readily took up Li+ in Na(+)- and K(+)-free media reaching intracellular concentrations of 5-10 mM in 10 mM extracellular Li+. Efflux of intracellular Li+ was stimulated 35% by extracellular 10 mM NaCl and 185% by 10 mM KCl over 10 mM choline chloride. Incubation of cells in 10 mM LiCl for 1 h inhibited the rate of Ca2+ efflux by 44% compared to cells in 10 mM NaCl. This may relate to the mechanism by which Li+ perturbs chemoresponse. A mutant with defects in Ca homeostasis responds normally to NH4Cl, but not to any of the stimuli that are affected by LiCl.     

Characterization and localization of actinogelin, a Ca2+ - sensitive actin accessory protein, in nonmuscle cells

Actinogelin, which induces gelation of F-actin at Ca2+ concentrations below micromolar concentrations but not at higher concentrations, was isolated in the pure state from Ehrlich tumor cells. The protein consists of subunits of 112,000-115,000 daltons and under physiological conditions is present mostly as a dimer. Up to 1 mol of actinogelin (dimer) binds to 10-12 mol of actin monomer. The binding was slightly decreased by the presence of 50 microM Ca2+ and almost completely inhibited by 300 mM KCl. Antibodies against actinogelin giving a single precipitation line with Ehrlich cell extract and with pure actinogelin were raised in rabbits. Antibody preparations were purified before use in an affinity column containing purified actinogelin. In mouse embryo fibroblasts and 3T3 cells, staining of actin bundles by the antiactinogelin antibody usually was discontinuous or gave a striated appearance. Most of the crossing points of the actin bundles were intensively stained. In epithelial cells from mouse small intestine, actinogelin was distributed throughout the cell, with the exception of the microvilli, which were devoid of staining. In mouse peritoneal cells, the antibody staining patterns were similar to those of tetramethylrhodamine isothiocyanate-labeled heavy meromyosin, but the former usually were sharper than the latter. Intracellular localization of actinogelin was drastically altered by cytochalasin D treatment at 10 microgram/ml. We conclude that actinogelin is present in a wide variety of cell types and discuss the possible participation of actinogelin in the Ca2+-dependent regulation of microfilament distribution.     

Action of exogenous potassium and calcium ions on in vitro metacyclogenesis in Trypanosoma cruzi.

The cations Ca2+ and K+ and the anions Cl-, HCO3-, and PO4- were studied for their contribution to metacyclic trypomastigote formation of Trypanosoma cruzi in starvation media consisting of phosphate-buffered saline (PBS) + 10 mM proline + 10 mM sodium acetate as well as one of the following salts: 0.035% NaHCO3 (PBSNPA), 0.035% K2CO3 (PBSKPA) or 0.035% K2HPO4 (PBSPPA). Isolates CL and DM28c were activated to transform with 5% CO2 and the percent metacyclogenesis determined after incubation for 96 h in PBS starvation media. Maximal metacyclogenesis was found with CaCl2 and KCl. In the presence of K+, the percent transformation was highest with the phosphate salt, followed by the carbonate and the chloride salts. Cells incubated in PBSNPA and the cationic ionophores A23187 (5 x 10(-6) M), lasalocid (5 x 10(-6) M), and valinomycin (10(-8) M) do not survive; addition of 2 mM CaCl2 or 17 mM KCl to DM28c cells, reversed the lethal action of the ionophores permitting differentiation into metacyclic forms. The addition of CaCl2 to CL cells incubated in ionophores abrogated the lethal effect of the ionophores but transformation was significantly different than in control preparations. Adding KCl to ionophore incubated cells resulted in normal levels of transformation except in the case of valinomycin. DM28c and CL cells incubated in PBSKPA show significantly greater metacyclogenesis in the presence of 5 mM EGTA. These results indicate that exogenous concentrations of several cations and anions significantly influence T. cruzi metacyclogenesis and that the degree of response by the parasite to free ion levels may be strain dependent.     

Na+ and K+ binding by glycerinated muscle fibers at equal concentrations of the chlorides of these cations in the medium

The Na+ and K+ equilibrium distribution between the medium and glycerinated muscle fibres of the frog has been investigated under equal concentrations of NaCl and KCl in solutions. Concentrations of NaCl and KCl varied from 0.5-1.5 mkM till 50 mM. Ion strength (0.11) was constant owing to the imidazol--HCl buffer. The binding of Na+ and K+ by model fibres occurred in accordance with the Langmur equation. Two kinds of cation-binding sites were found. The one with a low limiting ion sorption (A infinity approximately 1.3 mmol/kg dry weight of fibres) and high affinities (-delta F0 approximately 4.3 kcal/mol) was saturated at 0.5 mM concentrations (Na+ = K+) in the medium, and the other--with A infinity exceeding the previous one by an order and low -delta F0 (2.5 kcal/mol) was discovered at Na+, K+-1-10 mM. At ion concentrations equal to 0.5-1 mM the Langmur-binding is disturbed. At Na+-K+ less than or equal to 1 mM Na+ bound:K+ bound approximately to 1:1. At higher concentrations of cations Na+ bound:K+ bound approximately equal to 3:2. It is concluded that at least part of the sites in model fibres is capable of interacting only with Na+, but not with K+. It is supposed that at equal concentrations of Na+ and K+ in the medium the cations are bound by Na+, K+-ATPase of glycerinated muscle fibres.     

Hemisodium, a novel selective Na ionophore. Effect on normal human erythrocytes.

Hemisodium is a novel Na ionophore that belongs to the class of compounds called cryptands. These compounds possess an electron-rich cavity for binding of cations and are conformationally organized during synthesis to favor the selective binding of one cation over another. In media containing 145 mM NaCl and 5 mM KCl, hemisodium (10(-5) M) increased erythrocyte Na content from 23 to 345 mmol/kg.dry cell solid (dcs) over 4 h and increased water content from 1.8 to 3.5 liter/kg.dcs over the same period. K content decreased somewhat over the same time period, but this fall in K content was prevented entirely by incubation in either low Na media (to prevent net Na entry) or in Cl free media. Thus, the decrease in K content in high NaCl media was due to cell swelling, which activated KCl cotransport, and not due to a direct action of hemisodium on K permeability. Hemisodium-mediated Na transport was conductive, because erythrocyte membrane potential (Vm), determined by diS-C3-5 fluorescence, changed from -9 to +22 mV in high Na media in the presence of hemisodium and DIDS. In cells equilibrated with sulfamate, an anion with low conductive permeability, Vm changed 54 mV per 10-fold change in external Na concentration with the addition of hemisodium. In contrast, a 10-fold change in the external concentration of K, Rb, Cs, or T1 failed to alter Vm in the presence of hemisodium, suggesting a high Na specificity of the ionophore. Na conductance determined from net fluxes increased from 0.04 to 5.2 microS/cm2 with 10 microM hemisodium, and with that concentration the ratio of Na to K conductance was 45:1. Among the Na ionophores available so far, hemisodium appears to have the greatest specificity. Hemisodium may be a valuable tool in membrane transport studies.     

Na+ and K+ binding by glycerinated muscle fibers with reciprocal cation concentrations in the medium

The binding of Na+ and K+ by glycerinated muscle fibres was observed at reserve concentrations of NaCl in the medium. Under external concentrations of Na+ of K+ up to 0.4-0.5 mM, a constant fraction (0.15-0.25 mmoles/kg dry weight of the fibres) bound by glycerinated fibres was revealed. With the increase of NaCl or KCl concentration in the medium up to 10 mM the concentration of bound cations increased too. The parameters of Na+ and K+ sorption by glycerinated models were calculated. The values of Na+ and K+ binding limits were 4.4 and 1.8 mmole/kg dry weight of the fibres and those of affinity, 3.2 and 4.1 kcal/mol, respectively. The binding of one cation took place in conditions when its concentration was 10,000-20,000 fold less than that of the other cation. This points to the fact that Na+ and K+ binding is highly specific and is carried out by different centres. It is suggested that myosin ATPase is a substratum binding Na+ and K+ in glycerinated muscle fibres at reverse ratio concentrations of these cations in the medium.     

Membrane permeability changes induce hyperpolarization in transformed lymphoid cells under high-density culture conditions

BACKGROUND: Membrane potential changes in cells from the human lymphoid B cell line, JY, evoked by increasing cell density in culture were investigated, as data published on other cell types are controversial. An attempt was also made to clear the underlying mechanism. METHODS: Nonadherent JY cells were isolated from high-density plateau-phase cultures (type A cells), medium-density log-phase cultures (type B cells), and low-density lag-phase cultures (type C cells). They were analyzed for transmembrane potential, intracellular free concentration of potassium and sodium, membrane permeability for monovalent cations, cell cycle distribution by measuring DNA content, and glucose uptake. RESULTS: C type cells proved to be relatively depolarized (-41 +/- 3 mV) and cells obtained from the highest density cultures hyperpolarized (-60 +/- 3 mV). Intracellular concentrations ([K](i) = 92-97 mM and [Na](i) = 34-35 mM) were almost identical for each type of cell. The sodium/potassium permeability constant ratio in the A and C type of cells was 0.047 and 0.094, respectively. High-density culture conditions resulted in a pronounced G(1)-phase arrest. CONCLUSIONS: Differences in the membrane potential values induced by high-density culture conditions were maintained by changes in the membrane permeability for the monovalent cations.     

Purification and characterization of an ATP-sensitive actin gelation protein from Acanthamoeba castellanii

A protein which cross-links actin filaments in a nucleotide-sensitive manner has been purified to homogeneity from Acanthamoeba castellanii. This protein, GF-210, is a slightly asymmetric molecule composed of six subunits, each with an apparent mass of 35,000 Da. As determined by the method of falling ball vicometry, GF-210 was shown to cross-link actin filaments at hexamer:actin molar ratios of 1:500, with gelation occurring at molar ratios of 1:300 and higher. Actin gels did not form in the presence of 10 microM ATP, and filament cross-linking was completely inhibited by 100 microM ATP. Although ATP was the most effective inhibitor of actin filament cross-linking, other phospho-compounds including ADP, GTP, sodium phosphate, and sodium pyrophosphate prevented gelation at concentrations lower than 1.5 mM. In contrast, 50 mM KCl was required to inhibit the formation of actin networks. Direct binding studies showed that GF-210 binds to F-actin with a KD of 1.2 microM in the absence of ATP but with a KD of 72.8 microM in the presence of 2 mM ATP. This weakening of the interaction between F-actin and GF-210 may explain the inhibition of GF-210-induced actin cross-linking by nucleotides and other phospho-compounds.     

Determination of depolarisation- and agonist-evoked calcium fluxes on skeletal muscle cells in primary culture

Changes in intracellular calcium concentration ([Ca2+]i) evoked by prolonged depolarisation (120 mM KCl) or by the application of 15 mM caffeine were measured on skeletal muscle cells in primary culture. The extrusion rate (PVmax) of calcium from the myoplasm was determined, which in turn enabled the calculation of the calcium flux (Fl) underlying the measured calcium transients. PVmax was found to increase during differentiation, from 107 +/- 10 microM/s at the early myotube stage to 596 +/- 36 microM/s in secondary myotubes. This was paralleled by a decrease in resting [Ca2+]i from 99 +/- 4 to 51 +/- 2 nM. The depolarisation-evoked Fl rose to peak and then ceased despite the continuous presence of KCl. In contrast, the caffeine-induced Fl showed a peak and a clear steady-level with a peak-to-steady ratio of 5.6 +/- 1.2. Removal of external calcium suppressed the depolarisation--induced flux by 88 +/- 5% indicating that both an influx and a release from the SR underlie the K(+)-evoked calcium transients. Subsequent applications of caffeine resulted in essentially identical fluxes indicating an efficient refilling of the internal stores. Moreover, if a depolarisation-induced calcium transient preceded the second caffeine-evoked release, the latter was significantly larger than the first suggesting that much of the calcium that entered was stored in the SR rather than extruded.     

Phalloidin enhances actin assembly by preventing monomer dissociation

Incubation of the isolated acrosomal bundles of Limulus sperm with skeletal muscle actin results in assembly of actin onto both ends of the bundles. These cross-linked bundles of actin filaments taper, thus allowing one to distinguish directly the preferred end for actin assembly from the nonpreferred end; the preferred end is thinner. Incubation with actin in the presence of equimolar phalloidin in 100 mM KCl, 1 mM MgCl2 and 0.5 mM ATP at pH 7.5 resulted in a slightly smaller association rate constant at the preferred end than in the absence of the drug (3.36 +/- 0.14 X 10(6) M-1 s-1 vs. 2.63 +/- 0.22 X 10(6) M-1 s- 1, control vs. experimental). In the presence of phalloidin, the dissociation rate constant at the preferred end was reduced from 0.317 +/- 0.097 s-1 to essentially zero. Consequently, the critical concentration at the preferred end dropped from 0.10 microM to zero in the presence of the drug. There was no detectable change in the rate constant of association at the nonpreferred end in the presence of phalloidin (0.256 +/- 0.015 X 10(6) M-1 s-1 vs. 0.256 +/- 0.043 X 10(6) M-1 s-1, control vs. experimental); however, the dissociation rate constant was reduced from 0.269 +/- 0.043 s-1 to essentially zero. Thus, the critical concentration at the nonpreferred end changed from 1.02 microM to zero in the presence of phalloidin. Dilution-induced depolymerization at both the preferred and nonpreferred ends was prevented in the presence of phalloidin. Thus, phalloidin enhances actin assembly by lowering the critical concentration at both ends of actin filaments, a consequence of reducing the dissociation rate constants at each end.     

Isolation of an actin polymerization stimulator from bovine thyroid plasma membranes

An actin polymerization stimulator was purified from bovine thyroid plasma membranes by DNase I affinity column chromatography. Although the molecular weight of the protein was about 42,000 (42K) by sodium dodecyl sulfate polyacrylamide gel electrophoresis, it did not comigrate with actin. In the presence of 30 mM KCl, the 42K protein facilitated formation of actin filaments when analyzed by a centrifugation method, accelerated the initial phase of actin polymerization as measured in an Ostwald viscometer and increased the length of filaments as shown by electron microscopy. The 42K protein also accelerated the initial phase of actin polymerization in the presence of 100 mM KCl and 2 mM MgCl₂ but did not affect the final viscosity. The effect of the 42K protein was diminished by 5 uM cytochalasin B or 1 uM cytochalasin D. This 42K protein may anchor actin filaments onto the thyroid plasma membrane.     

Ca2+-activated K channel of the BK-type in the inner mitochondrial membrane of a human glioma cell line

A single channel current was recorded from mitoplasts (i.e., inner mitochondrial membrane) of the human glioma cell line LN229 using patch-clamp techniques in the mitoplast-attached mode. We frequently found a 295 +/- 18 pS channel that showed a straight i-E relation in the range +/-60 mV in 150 mM KCl solutions on either side of the mitoplast. If KCl in the bath was exchanged against NaCl, outward currents were undetectable, indicating potassium selectivity. Channel activity determined as open probability increased with increasing Ca2+ concentrations (EC50 = 0.9 microM at 60 mV). Open probability was voltage dependent. An e-fold increase of time spent in the open state was induced by a depolarization of 10.5 mV. Open probability was decreased by charybdotoxin concentration and voltage dependently (EC50 = 1.4 nM). In conclusion, we show for the first time that the inner mitochondrial membrane in human glioma cells contains a calcium-dependent K channel of the BK-type.     

A molecular beacon strategy for real-time monitoring of triplex DNA formation kinetics

We used a molecular beacon (MB) containing a 15-mer triplex-forming oligonucleotide (TFO) to probe in real-time the kinetics of triplex DNA formation in the left side of the TCl tract (502-516) of the c-src proto-oncogene in vitro. The metal ions Na+, K+, and Mg2+ stabilized triplex DNA at this site. The pseudo-first-order rate constant (kpsi) and the second-order association rate constant (k1) for the binding of the MB to the target duplex in 10 mM sodium phosphate buffer, pH 7.3, increased from 3.2 +/- 0.9 to 15 +/- 2.8 x 10(-3) s(-1) and 6.4 +/- 1.8 to 30 +/- 5.6 x 102 M(-1) s(-1), respectively, on increasing the MgCl2 concentration from 1 to 2.5 mM. Similar values were obtained for the triplex DNA stabilized by NaCl (100-250 mM). Surprisingly, the values were around 2 times higher in the presence of KCl. The AG of triplex formation in the presence of 1 mM MgCl2, 150 mM NaCl, and 150 mM KCl were -7.8 +/- 0.3, -8.2 +/- 0.3 and -8.7 +/- 0.7 kcal/mol respectively, despite significant differences in the values of deltaH and deltaS, suggesting enthalpy-entropy compensation in the stabilization of the triplex DNA by these metal ions. These results show the utility of MBs ih probing triplex DNA formation and in evaluating kinetic and thermodynamic parameters important for the design and development of TFOs as triplex DNA-based therapeutic agents.