Demonstration of corticotropin-releasing factor binding sites on human and rat erythrocyte membranes and their modulation by chronic ethanol treatment in rats

In a previous study we reported the presence of specific corticotropin-releasing factor (CRF) binding sites in peripheral tissues of the rat (Endocrinology, 116, 2152, 1985). Using 125I-labeled rat or human CRF, specific CRF binding sites were identified on rat and human erythrocytes, but not on lymphocytes or platelets. Furthermore, identical CRF binding was observed in the presence of intact erythrocytes or lysed erythrocyte membranes. Maximal binding of 125I-CRF occurred within 25 min at 4 degrees C and was saturable. Scatchard analysis of CRF binding to erythrocyte membranes revealed the existence of a single class of binding site. Chronic exposure of rats to ethanol vapor, known to lower specific CRF binding to pituitary tissue by 35%, also decreased 125I-rat CRF binding to erythrocyte membranes by approximately 45%, which was due to a decrease in the number of CRF binding sites. The parallel decrease of CRF binding to rat-erythrocyte and pituitary membranes following chronic ethanol treatment suggests that CRF binding to erythrocyte and pituitary membranes is modulated in a similar direction, which further suggests that the determination of CRF binding to erythrocytes may provide an important clinical tool to indirectly assess CRF-receptor levels in the pituitary gland and thereby enhance our understanding of ethanol-induced disorders of the hypothalamic-pituitary-adrenal axis in patients.     

Saturable binding to cell membranes of the presynaptic neurotoxin, beta-bungarotoxin.

Brief exposure to the protein neurotoxin, beta-bungarotoxin, is known to disrupt neuromuscular transmission irreversibly by blocking the release of transmitter from the nerve terminal. This neurotoxin also has a phospholipase A2 activity, although phospholipases in general are not very toxic. To determine if the toxicity of this molecule might result from specific binding to neural tissue, we have looked for high affinity, saturable binding using 125I-labelled toxin. At low membrane protein concentration 125I-labeled toxin binding was directly proportional to the amount of membrane; at fixed membrane concentration 125I-labeled toxin showed saturable binding. It was unlikely that iodination markedly changed the toxin's properties since the iodinated toxin had a comparable binding affinity to that of native toxin as judged by competition experiments. Comparison of toxin binding to brain, liver and red blood cell membranes showed that all had high affinity binding sites with dissociation constants between one and two nanomolar. This is comparable to the concentrations previously shown to inhibit mitochondrial function. However, the density of these sites showed marked variation such that the density of sites was 13.0 pmol/mg protein for a brain membrane preparation, 2.4 pmol/mg for liver and 0.25 pmol/mg for red blood cell membranes. In earlier work we had shown that calcium uptake by brain mitochondria is inhibited at much lower toxin concentrations than is liver mitochondrial uptake. Both liver and brain mitochondria bind toxin specifically, but the density of 125I-labeled toxin binding sites on brain mitochondrial preparations (3.3 +/- 0.3 pmol/mg) exceeded by a factor of ten the density on liver mitochondrial preparations (0.3 +/- 0.05 pmol/mg). It is also shown that labeled toxin does not cross synaptosomal membranes, suggesting that mitochondria may not be the site of action of the toxin in vivo. We conclude that beta-bungarotoxin is an enzyme which can bind specifically with high affinity to cell membranes.     

Saturable binding to cell membranes of the presynanptic neurotoxin, β-Bungarotoxin

Brief exposure to the protein neurotoxin, β-bungarotoxin, is known to disrupt neuromuscular transmission irreversibly by blocking the release of transmitter from the nerve terminal. This neurotoxin also has a phospholipase A2 activity, although phospholipases in general are not very toxic. To determine if the toxicity of this molecule might result from specific binding to neural tissue, we have looked for high affinity, saturable binding using ¹²⁵I-labeled toxin. At low membrane protein concentration ¹²⁵I-labeled toxin binding was directly proportional to the amount of membrane; at fixed membrane concentration ¹²⁵I-labeled toxin showed saturable binding. It was unlikely that iodination markedly changed the toxin's properties since the iodinated toxin had a comparable binding affinity to that of native toxin as judged by competition experiments. Comparison of toxin binding to brain, liver and red blood cell membranes showed that all had high affinity binding sites with dissociation constants between one and two nanomolar. This is comparable to the concentrations previously shown to inhibit mitochondrial function. However, the density of these sites showed marked variation such that the density of sites was 13.0 pmol/mg protein for a brain membrane preparation, 2.4 pmol/mg for liver and 0.25 pmol/mg for red blood cell membranes.In earlier work we had shown that calcium uptake by brain mitochondria is inhibited at much lower toxin concentrations than is liver mitochondrial uptake. Both liver and brain mitochondria bind toxin specifically, but the density of ¹²⁵I-labeled toxin binding sites on brain mitochondrial prepartions ($\text{3.3 ± 0.3}\text{pmol/mg}$) exceeded by a factor of ten the density on liver mitochondrial preparations ($\text{0.3 ± 0.05}\text{pmol/mg}$). It is also shown that the labeled toxin does not cross synaptosomal membranes, suggesting that mitochondria may not be the site of action of the toxin in vivo. We conclude the β-bungarotoxin is an enzyme which can bind specifically with high affinity to cell membranes.     

Characterization of beta-adrenergic receptors in the rat vas deferens using [3H]-dihydroalprenolol binding

The beta-adrenergic antagonist, [3H]-dihydroalprenolol ([3H] DHA), binds to membranes prepared from the rat vas deferens in a specific and saturable manner. Scatchard and Hill plot analysis indicates a single class of binding sites with no evidence of cooperative interactions. The specific binding sites have a high affinity (Kd = 0.3 nM) and a maximal occupancy estimated to be 460 fmoles [3H]-DHA bound/g wet tissue weight. Beta-adrenergic agonists and/or antagonists inhibit [3H]-DHA binding to rat vas deferens membranes in a stereospecific manner and with a relative order of potency expected for beta-adrenergic receptors of the beta2 subtype. The receptor affinities of various beta-adrenergic antagonists in the rat vas deferens determined using inhibition of [3H]-DHA binding correlated with their receptor affinities determined physiologically using antagonism of isoproterenol-induced inhibition of neurogenic contractions in-vitro.     

Interaction of glucocorticoid hormones with mitochondrial membranes

The mechanism of 3H glucocorticoid binding with the rat liver mitochondria in vitro is investigated. The linear dependence of the amount of bound hormones on the concentration of the free ones is shown and no saturation in the region of the physiological concentrations is observed. A very low specific binding in the presence of a 100-fold excess of an unlabelled hormone is found. The outer mitochondrial membranes binds a considerably higher amount of steroids, than the inner one. The binding of steroids with the intact liver mitochondria is 2-3 times higher as compared to the binding with spheroplasts of Escherichia coli. Delipidization of mitochondria by diverse lipotropic agents differently influences the binding of steroids with the different functional groups. The interaction of steroids with mitochondria depends on the osmolarity of the incubation medium: the binding is 1.5-3 times higher in the isotonic sucrose solution, that in the hypo- or hypertonic ones. A conclusion is made about the nonspecific character of glucocorticoid binding with mitochondria caused by the interaction with hydrophobic compounds of the mitochondrial membranes. The possible chemical mechanisms for such an interaction are discussed.     

Detection and characterization of anionic polypeptide fraction binding sites in rat liver plasma membranes and cultured hepatocytes

The binding of human 125I-labeled 'anionic polypeptidic fraction' (APF) to purified rat liver plasma membranes was studied. The dissociation constant for this binding was 3.0 micrograms protein/mg membrane protein. Binding was competitively inhibited by unlabeled human APF, but not by human LDL (low density lipoproteins). When unlabeled HDL3 was added, binding of labeled APF was competitively reduced to a level between that of unlabeled APF and unlabeled LDL. Experiments with cultured rat hepatocytes confirmed those obtained with liver membranes and suggested the presence in rat liver of saturable APF-binding sites which seem to be specific for APF. The physiologic significance of these APF binding sites is discussed in relation to the fate of cholesterol in the liver.     

Characterization, size estimation and solubilization of alpha-macroglobulin complex receptors in liver membranes

Receptors for alpha 2-macroglobulin-proteinase complexes have been characterized in rat and human liver membranes. The affinity for binding of 125I-labelled alpha 2-macroglobulin.trypsin to rat liver membranes was markedly pH-dependent in the physiological range with maximum binding at pH 7.8-9.0. The half-time for association was about 5 min at 37 degrees C in contrast to about 5 h at 4 degrees C. The half-saturation constant was about 100 pM at 4 degrees C and 1 nM at 37 degrees C (pH 7.8). The binding capacity was approx. 300 pmol per g protein for rat liver membranes and about 100 pmol per g for human membranes. Radiation inactivation studies showed a target size of 466 +/- 71 kDa (S.D., n = 7) for alpha 2-macroglobulin.trypsin binding activity. Affinity cross-linking to rat and human membranes of 125I-labelled rat alpha 1-inhibitor-3.chymotrypsin, a 210 kDa analogue which binds to the alpha 2-macroglobulin receptors in hepatocytes (Gliemann, J. and Sottrup-Jensen, L. (1987) FEBS Lett. 221, 55-60), followed by SDS-polyacrylamide gel electrophoresis, revealed radioactivity in a band not distinguishable from that of cross-linked alpha 2-macroglobulin (720 kDa). This radioactivity was absent when membranes with bound 125I-alpha 1-inhibitor-3 complex were treated with EDTA before cross-linking and when incubation and cross-linking were carried out in the presence of a saturating concentration of unlabelled complex. The saturable binding activity was maintained when membranes were solubilized in the detergent 3-[(3-cholamidopropyl)dimethylammonio]propane sulfonate (CHAPS) and the size of the receptor as estimated by cross-linking experiments was shown to be similar to that determined in the membranes. It is concluded that liver membranes contain high concentrations of an approx. 400-500 kDa alpha 2-macroglobulin receptor soluble in CHAPS. The soluble preparation should provide a suitable material for purification and further characterization of the receptor.     

Specific receptors for atriopeptin III in rabbit lung

Binding studies revealed the presence of a single class of high affinity binding sites for atriopeptin III on rabbit lung membranes. An apparent dissociation constant (Kd) of 0.32 nM and a binding capacity of 166 fmol/mg protein was determined. Binding was time-dependent and saturable. The relative binding affinities of atrial peptide analogs correlated well with their potencies in eliciting relaxation of norepinephrine-contracted rabbit aorta strips. Unrelated peptide hormones did not compete for the atriopeptin binding site on rabbit lung membranes. The atrial peptide binding data are similar to those obtained for other tissues and indicate the presence of a physiologically relevant atrial peptide receptor in lung.     

High affinity peptide histidine isoleucine-preferring receptors in rat liver

Peptide Histidine Isoleucine (PHI) is generally considered a low affinity agonist for Vasoactive Intestinal Peptide (VIP) receptors. In this study, we investigated the presence and characteristics of [125I]PHI binding sites on rat liver membranes. Detergents at nonsolubilizing concentrations (1 mM CHAPS or 0.01% Tween-20) were included in the assay buffer to reduce adsorptive loss of PHI to acceptable levels and permit measurement of PHI-binding to receptors. Under these conditions, binding of PHI to liver membranes was time- and temperature-dependent, reversible and saturable. Unlabeled PHI was 9.7-fold more potent than VIP, and 357-fold more potent than secretin in displacing [125I]-PHI binding. Scatchard analysis suggested the presence of two classes of PHI receptors, with Kd 27 pM and 512 pM. The data from [125I]-PHI and [125I]-VIP binding studies suggested that one class of receptors was PHI-preferring, and the other, equally reactive with PHI and VIP. The concentration of immunoreactive PHI, measured by radioimmunoassay, in blood from the hepatic portal vein of anesthetized rats was 2-fold higher than that from the hepatic vein, suggesting uptake of circulating PHI by the liver.     

Characterization of the substance P receptor in rat brain cortex membranes and the inhibition of radioligand binding by guanine nucleotides

Rat brain cortex membranes bind to a conjugate of substance P and 125I-labeled Bolton-Hunter reagent, and this binding can be inhibited by a low concentration of substance P (Kd = 1.2 +/- 0.4 X 10(-8) M). This binding is reversible and saturable (0.5 +/- 0.1 pmol of binding sites/mg of protein). Fragments of substance P as small as the carboxyl-terminal hexapeptide can inhibit the binding although their potency decreases with the decrease in the length of the peptides. The binding affinities of smaller peptides or peptides in which the carboxyl-terminal amide or amino acids are removed are drastically reduced. Biologically active analogs of substance P, physalaemin, eledoisin, substance P methyl ester, [D-Ala0]hepta(5-11)substance P, kassinin, and the eledoisin-related hexapeptide also can inhibit the binding. However, the binding is not inhibited by polypeptides structurally unrelated to substance P or by amine hormones/neurotransmitters. The binding affinities of biologically active peptides to rat brain cortex membranes are almost identical with their affinities for rat parotid cells which we previously determined. Furthermore, the recently described substance P antagonist, [D-Pro, D-Trp]substance P, inhibits the binding of the 125I-labeled substance P derivative to brain cortex membranes and to parotid cells equally well. These results suggest that the substance P receptors in the brain cortex and the parotid gland are similar. The brain cortex membrane binding of the 125I-labeled substance P derivative can be inhibited by micromolar concentrations of GTP, GDP, and their analogs. ITP and IDP were less active. Adenine and pyridine nucleotides were inactive.     

Characterization of oxytocin receptors on isolated rat fat cells.

The binding of 3H-labelled neurohypophyseal nonapeptide hormone, oxytocin, to isolated rat fat cells has been measured under conditions where this compound elicits the known activation of glucose oxidation by these cells, called "insulin-like" action. Uptake by the cells of the [3H]peptide as a function of various concentrations of the hormone in the medium indicated the presence of two classes of binding sites with different apparent affinities and capacities. The sites of the first type exhibit a rather high affinity, but low capacity, for oxytocin (5 nM; 3 X 10(4) sited per cell) and appear to be saturable under a reversible process. Evaluation of dose-response relationships suggest that they may be directly related to the measured biological response (i.e. activation of the glucose to 14CO2 conversion). Competition experiments show that [3H]oxytocin binding to the cells remains constant within a large range of insulin concentrations. The apparent capacity of different hormone analogs to compete with oxytocin for binding to this class of receptors has been evaluated and compared with the measured insulin-like activity of these different compounds. The sites of the second category have significantly lower affinity, but higher capacity for oxytocin, and were found to be not saturable under the experimental conditions. [3H]Oxytocin uptake by ghosts prepared from the isolated fat cells showed striking similarities to the binding process described for whole cells, although the affinity and total capacity of the former were found to be slightly lower. The basal and adrenalin-stimulated adenylate cyclase of these fractions appeared to be unaffected by various concentrations of oxytocin. It is concluded that there may exist on the rat fat cell membranes a discrete number of oxytocin receptors possessing high specificity for oxytocin and exhibiting affinities and kinetic behaviour similar to those of other characterized oxytocin receptors. They are believed to be independent of the other hormonal receptors of the rat fact cells.     

Sex steroid-binding protein interacts with a specific receptor on human premenopausal endometrium membrane: modulating effect of estradiol.

Sex steroid-binding protein receptor was detected on membranes prepared from human premenopausal endometrium. The binding of sex steroid-binding protein to membranes was specific, saturable, and high affinity. Scatchard analysis showed the presence of two binding sites at different affinities. The addition of estradiol (10(-8) M) did not produce any inhibition of binding; indeed, it resulted in a modification of binding characteristics. The demonstration of sex steroid-binding protein receptor on membranes of human premenopausal endometrium indicates that the expression of receptor on membranes is not an effect of estrogen over stimulation on target tissues. Estradiol could act as a modulating factor of the binding, probably reflecting the sensitivity of tissues to different steroids.     

Selenoprotein P receptor from rat

Radioreceptor assay technology was used to show the presence in the rat of a receptor that binds selenoprotein P, a selenocysteine-containing rat plasma protein. 75Se-labeled selenoprotein P bound to testis, kidney, and liver membranes. The binding was specific in that increasing amounts of partially-fractionated rat plasma specifically displaced the binding of 75Se-labeled selenoprotein P to testis membrane in a competitive manner. 75Se-labeled selenoprotein P binding was saturable in the presence of increasing amounts of testis membranes. The binding of 75Se-labeled selenoprotein P was optimal at about pH 4.2. Several proteins and blood fractions had little or no significant effect on binding of 75Se-labeled selenoprotein P to testis membranes. All plasma sources tested specifically displaced 75Se-labeled selenoprotein P from testis membrane, indicating that selenoprotein P-related proteins may be widespread in nature. The study indicated that selenoprotein P has a receptor and is involved in selenium transport.     

Photoaffinity labelling of nucleoside-transport proteins in plasma membranes isolated from rat and guinea-pig liver.

Nitrobenzylthioinosine (NBMPR) was employed as a probe of the nucleoside transporters from rat and guinea-pig liver. Purified liver plasma membranes prepared on self-generating Percoll density gradients exhibited 16-fold (rat) and 10-fold (guinea pig) higher [3H]NBMPR-binding activities than in crude liver homogenates (3.69 and 14.7 pmol/mg of protein for rat and guinea-pig liver membranes respectively, and 0.23 and 1.47 pmol/mg of protein for crude liver homogenates respectively). Binding to membranes from both species was saturable (apparent Kd 0.14 and 0.63 nM for rat and guinea-pig membranes respectively) and inhibited by uridine, adenosine, nitrobenzylthioguanosine (NBTGR) and dilazep. Uridine was an apparent competitive inhibitor of high-affinity NBMPR binding to rat membranes (apparent Ki 1.5 mM). There was a marked species difference with respect to dipyridamole inhibition of NBMPR binding (50% inhibition at 0.2 and greater than 100 microM for guinea-pig and rat respectively). These results are consistent with a role of NBMPR-binding proteins in liver nucleoside transport. Exposure of rat and guinea pig membranes to high-intensity u.v. light in the presence of [3H]NBMPR resulted in the selective radio-labelling of membrane proteins which migrated on sodium dodecyl sulphate/polyacrylamide gels with apparent Mr values in the same range as that of the human erythrocyte nucleoside transporter (45 000-66 000). Covalent labelling of these proteins was abolished when photolysis was performed in the presence of non-radio-active NBTGR as competing ligand.     

Differential effect of ethanol on muscarinic cholinergic binding to rat and locust neural membranes

We have compared the effect of ethanol, a membrane perturbant, on the muscarinic binding sites in neural membranes from a vertebrate (rat) and an insect (locust). The binding of the muscarinic antagonist [³H]quinuclidinyl benzilate ([³H]QNB) to both rat and locust neural membranes was inhibited by ethanol at 10–500 mM concentrations; but this inhibition was greater in the locust. Ethanol (500 mM) increased the apparent dissociation constant (K d) of [³H]QNB binding to rat membranes from 0.13±0.01 nM in control to 0.20±0.02 nM; there was also an small but significant reduction in the number of binding sitesB _max. In locust, 500 mM ethanol reduced theB _max of [³H]QNB binding from 590±30 in control to 320±40 pmol/g protein; no significant alteration in theK _D was detected. The dissociation rate constant (k _off) of [³H]QNB increased from 0.020±0.003 in controls to 0.031±0.004 (min^–1) in the presence of 500mM ethanol, the association rate constant (k _on) did not change significantly. In locust, 500 mM ethanol did not affect eitherk _on ork _off. Competition experiments revealed that the binding affinities of both the agonist carbamylcholine and the antagonist atropine to the rat membranes were reduced in the presence of ethanol. In contrast, ethanol caused no alteration in the binding affinities of these ligands to the locust membranes. This differential effect of ethanol on rat and locust muscarinic binding suggests a difference in the hydrophobic domains and/or the membrane interactions of the muscarinic receptors in the two species.     

Oxidative phosphorylation. The specific binding of trimethyltin and triethyltin to rat liver mitochondria

1. The binding of trimethyltin and triethyltin to rat liver mitochondria was determined and the results were analysed by the method of Scatchard (1949). 2. One binding site (site 1) has the correct characteristics for the site to which trimethyltin and triethyltin are attached when they inhibit oxidative phosphorylation. For each compound the concentration of site 1 is 0.8nmol/mg of protein and the ratios of their affinity constants are the same as the ratio of the concentrations inhibiting oxidative phosphorylation. 3. Binding site 1 is present in a fraction derived from mitochondria containing only 15% of the original protein. In this preparation ultrasonication rapidly destroyed site 1. 4. Dimethyltin and diethyltin do not prevent binding of triethyltin to rat liver mitochondria, whereas triethyl-lead does. 5. Trimethyltin and triethyltin bind to mitochondria from brown adipose tissue and the results indicate a binding site 1 similar to that in rat liver mitochondria. 6. The advantages and limitations of this approach to the study of inhibitors are discussed.     

The interaction of adriamycin with cardiolipin in model and rat liver mitochondrial membranes

The interaction of adriamycin with cardiolipin in model membranes and in various membrane preparations derived from rat liver mitochondria was studied and the results are analyzed in the light of a possible specific interaction between adriamycin and cardiolipin. It was found that adriamycin binds to cardiolipin-containing model membranes with a fixed stoichiometry of two drug molecules per cardiolipin. Furthermore, the extent of drug complexation by mitochondria and mitoplasts (inner membrane plus matrix) is in reasonable agreement with their cardiolipin content. In contrast, adriamycin-binding curves of inner membrane ghosts and submitochondrial particles reveal considerable association to an additional site, presumably RNA. The evidence for the potential importance of RNA as a target comes from experiments on outer membranes and microsomes which both appear to bind substantial amounts of adriamycin. Removal of the major part of the RNA associated with these fractions by EDTA treatment is accompanied by a dramatic reduction of binding capacity. We propose that endogenous RNA present in mitochondria and mitoplasts is not accessible for adriamycin at low concentrations of the drug due to the presence of an intact lipid barrier. This potential site comes to expression in ghosts and submitochondrial particles, due to the absence of an intact lipid bilayer and due to the inside-out orientation of the limiting membrane, respectively. Electron microscopical studies show that adriamycin induces dramatic changes in mitochondrial morphology, similar to the uncoupler-induced effects described by Knoll and Brdiczka (Biochim. Biophys. Acta 733, 102-110 (1983). Adriamycin has an uncoupling effect on mitochondrial respiration and oxidative phosphorylation. The concentration dependence of this effect correlates with the adriamycin-binding curve for mitochondria which implies that only bound adriamycin actively inhibits respiration.     

In vitro characterization of a gamma-secretase radiotracer in mammalian brain

Inhibition of gamma-secretase is a potential therapeutic target for Alzheimer's disease (AD). The present studies have characterized the in vitro properties of a radiolabeled small molecule gamma-secretase inhibitor, [3H]compound D (Yan et al., 2004, J. Neurosci.24, 2942-2952) in mammalian brain. [3H]Compound D was shown to bind with nanomolar affinity (Kd = 0.32-1.5 nM) to a single population of saturable sites in rat, rhesus and human brain cortex homogenates, the density of binding sites ranging from 4 to 7 nM across the species. Competition studies with a structurally diverse group of gamma-secretase inhibitors with a wide range of binding affinities showed that the binding affinities of these compounds correlated well with their ability to inhibit gamma-secretase in vitro. Autoradiographic studies showed that the specific binding of [3H]compound D was widely distributed throughout adult rat, rhesus and normal human brain. There did not appear to be any difference in distribution of [3H]compound D specific binding sites in AD cortex compared with control human cortex as measured using tissue section autoradiography, nor any correlation between gamma-secretase binding and plaque burden as measured immunohistochemically. [3H]compound D is a useful tool to probe the expression and pharmacology of gamma-secretase in mammalian brain.     

High affinity thyroid hormone binding sites on purified rat liver plasma membranes.

Two orders of saturable binding sites for L-T₃ were detected on purified rat liver plasma membranes--a high affinity, low capacity binding site with a Kd of 3.2 ± 0.5 nM, and a lower affinity, higher capacity site with a Kd of 220 ± 50 nM. Competition-inhibition studies revealed that both D-T₃ and L-T₄ (two compounds with lower biological potencies than L-T₃) were also less potent than L-T₃ in competing for these binding sites. The present studies demonstrate, therefore, the presence of specific thyroid hormone binding sites on rat liver plasma membranes. In addition, they suggest that these sites may have a role both in mediating the known effects of thyroid hormones on membrane functions, and in regulating the entry of thyroid hormones into target cells.     

ATP-Mg2+ reversal of the salt activation of membrane bound carnitine palmitoyltransferase activities of liver mitochondria

The carnitine palmitoyltransferase (CPT) activities of the outer and the inner membranes of rat liver mitochondria were markedly activated by increase in the ionic strength of the assay medium. ATP at physiological concentrations in the presence of Mg2+ effectively reversed the above effect with octanoyl-CoA, but not with palmitoyl-CoA, as a substrate. Other nucleotides were unable to substitute for ATP. This ATP-Mg2+ effect on the CPT activity was not seen with mitochondria of heart or of skeletal muscles. The remarkable nucleotide, substrate and tissue specificity of these effects indicate that the above phenomenon may be functional in vivo to regulate the ability of liver mitochondria to utilize medium chain fatty acids via the carnitine-dependent route.