Purification and partial characterization of ATP pyrophosphohydrolase from fetal bovine epiphyseal cartilage

ATP pyrophosphohydrolase was partially purified from fetal bovine epiphyseal cartilage. The purification was about 10- and 100-fold over the enzyme activities of matrix vesicle fraction and cell homogenate, respectively. The pyrophosphohydrolase and alkaline phosphatase were separated by a sequential application of Sepharose CL-6B and DEAE-cellulose column chromatographies. The purified enzyme migrated as a single band corresponding to the molecular weight of 230,000 in sodium dodecyl sulfate-polyacrylamide disc gel by electrophoresis. The enzyme absolutely required Zn2+ for its activity and appeared to bind Zn2+ strongly with an apparent affinity of p[Zn2+]0.5 = 13.4. The apparent Km for ATP was 0.18 mM. The enzyme was also reactive toward various nucleoside triphosphates including GTP, CTP, and UTP. In contrast, various phosphodiesters including RNA, UDP-glucose, NAD, and bis-p-nitrophenylphosphate were 5% or less as reactive as the nucleoside triphosphates. The pyrophosphohydrolase was inactive toward adenosine 3':5'-monophosphate or various phosphonates. UDP-glucose (1 mM), NAD (1 mM), or RNA (1 mg/ml) failed to inhibit the ATP pyrophosphohydrolase activity. These observations suggest that the ATP pyrophosphohydrolase of the cartilage is probably not a phosphodiesterase I. The matrix vesicle fraction, which probably also included some plasma membrane vesiculated during collagenase digestion, contained the highest specific activity of the enzyme as compared to other subcellular fractions of either epiphyseal or articular cartilage.     

Effects of ATP on regulation of galactosyltransferase-I activity responsible for synthesis of the linkage region between the core protein and glycosaminoglycan chains of proteoglycans.

We report that ATP enhances the activity of galactosyltransferase-I, which synthesizes the linkage region between glycosaminoglycan chains and the core proteins of proteoglycans. The enzyme activity in cell-free fractions prepared from cultured human skin fibroblasts was measured by high-performance liquid chromatographic detection of galactosyl-xylosyl-(4-methylumbelliferone) produced from 4-methylumbelliferyl-beta-D-xyloside used as an acceptor. ATP at 2 mM increased the enzyme activity by about 60% in the 110 x g supernatant of the cell homogenate, but not in the supernatant or precipitate fractions obtained by 100,000 x g centrifugation. When both fractions (the 100,000 x g supernatant and precipitate) were mixed, the additional ATP increased the enzyme activity. This increase was canceled by heat treatment or trypsin digestion of the 100,000 x g supernatant. In addition, the 100,000 x g precipitate, which was prepared from the 110 x g supernatant preincubated with ATP, exhibited increased activity, and this increase was abolished by alkaline phosphatase treatment. These results suggest that a protein kinase in the 100,000 x g supernatant activates galactosyltransferase-I activity.     

Protein kinase C located in rat liver nuclei. Partial purification and biochemical and immunochemical characterization

In the rat liver homogenate, maximal protein kinase C activity was found at two calcium concentrations (1.75 and 3.5 mM). Subcellular fractionation of the liver homogenate revealed that the protein kinase C activity requiring 1.75 mM calcium was present only in the cytosolic and particulate subcellular fractions. The protein kinase C activity requiring 3.5 mM calcium concentration was mainly located in the rat liver nuclei preparation. About 19% of the liver homogenate protein kinase C activity requiring 3.5 mM calcium was present in the nuclei. Goat anti-rat brain protein kinase C antibodies revealed a single immunoreactive band at 80-82 kDa in the rat liver nuclear, particulate, or cytosolic fractions. Based on the ratio of plasma membrane marker enzyme activity determined in the nuclear preparation, the purity of the isolated nuclei was ascertained. Rat liver nuclear protein kinase C activity has been partially purified. The purification steps sequentially employed were Triton X-100 extraction of isolated nuclei, DEAE-cellulose chromatography, Phenyl-Superose, and Mono Q (fast protein liquid) chromatography. The final purification step revealed, by silver nitrate staining on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, two protein bands at 80 and 66 kDa, respectively. These findings provide definitive data regarding the nuclear location of protein kinase C. The nuclear location of protein kinase C may lead to an understanding of the molecular pathway involved in signal transduction from the plasma membrane to the nucleus.     

The effect of somatotropin on adenosine deaminase activity in tissues of rats with hypo- and hyperthyroidism

It has been shown that under hypothyrosis (thyroidectomy) the adenosine deaminase activity in the total suspension of thymocytes does not change as compared with that of intact rats. When suspension of thymocytes was separated by centrifugation in the ficoll-urografin density gradient into 6 fractions the enzyme activity has been found to increase by 83 and 31%, respectively, in light fractions with the ficoll density of 1.065 and 1.071 and to decrease by 21% in the fraction with density of 1.095 as compared with the intact rats. The enzyme activity in spleen homogenate of hypothyroid rats is higher by 72% than the norm. This activity in thymocytes and spleen of hyperthyroid rats is lower by 37% and 30% respectively. Somatotropin administration to rats normalizes these changes.     

Conversion of alpha-ketoglutarate into L-glutamic acid with urea as ammonium source using multienzyme systems and dextran-NAD+ immobilized by microencapsulation within artificial cells in a bioreactor

Urea could be effectively converted into L-glutamic acid with semipermeable nylon-polyethylenimine artificial cells containing L-glutamic dehydrogenase (EC 1.4.1. 3), yeast alcohol dehydrogenase (EC 1.1.1.1), urease (EC 3.5.1. 5) and soluble dextran-NAD(+). For batch conversion, the artificial cell suspension to total reaction volume ratios ranged from 1 in 5 to 1 in 60. From 22.6 to 53.4 mumol of L-glutamic acid could be produced by 0.4 mL artificial cell suspension within 2 h. The corresponding conversion ratios were 56.5-11. 1%. The L-glutamic dehydrogenase multienzyme system showed a good storage stability: 66.0% of the original activity was retained after 1 month of storage at 4 degrees C. A small bioreactor was prepared to contain 4.0 mL artificial cells. At a flow rate of SV = 1.5 h(-1), the maximum conversion rate was 49.6 mumol L-glutamic acid/p h. Thirty-eight percent of the maximum activity was retained when continuously used for four days at 22 degrees C. A kinetic analysis for the L-glutamic dehydrogenase multienzyme system was studied. The Michaelis constants are as follows: alpha-ketoglutarate is 0.838 mM; urea is 1.90 mM; dextran- NAD(+) is 0.345 mM; and ethanol is 5.31 mM.     

Distribution of calcium activated neutral proteinase (mM CANP) in myelin and cytosolic fractions in bovine brain white matter

The activity of calcium-activated neutral proteinase (mM CANP) was determined in homogenate, myelin and supernatant of bovine brain corpus callosum. The enzyme activity in homogenate and myelin was increased eleven and thirteen-fold respectively by Triton X-100. Myelin prepared by the method of Norton and Poduslo as well as by a modified method, was shown to contain most (more than 50%) of homogenate mM CANP activity. The specific activity was highest in myelin, and increased almost three times more than the homogenate. Supernatant only contained 17% of enzyme activity. It is concluded from these studies that mM CANP is tightly bound to the membrane and predominantly associated with the myelin sheath.     

Transglutaminase Activity in Rat Brain: Characterization, Distribution, and Changes with Age

The activity of transglutaminase was characterized in the rat brain. In adults, comparable levels of transglutaminase activity are present in all brain regions examined. The activity is present in all subcellular fractions, as studied by differential centrifugation, but the soluble fraction contains the highest specific activity. The endogenous activity (enzyme activity assayed in the absence of the exogenous substrate casein) is very low in all subcellular fractions, except in the synaptosomal fraction where its highest levels are about 40-60% of the activity assayed in the presence of casein. Furthermore, enzyme activity is present on the external surface of synaptosomes. In the soluble fraction, maximal activity can be detected between pH values of 9 and 10 when assayed in the presence of 5 mM CaCl2 (with half-maximal activity requiring 0.75 mM CaCl2) and 0.4 mM putrescine (with an apparent Km for putrescine of 0.1 mM). The activity can be partially inhibited by ZnCl2 (with an IC50 of 4.5 mM) and by AlCl3 (with an IC50 of 5.1 mM). In the cerebellum, where the full span of neuronal development can be studied after birth, the highest specific activity is observed just after birth, thereafter the activity starts to decline and by 14 days, after a reduction of about 65%, it reaches levels observed throughout life.     

Sialidase in the guinea pig pulmonary parenchyma. Increased activity in the cytosolic and microsomal subcellular fractions after stimulation with Bacillus Calmette Guérin

The sialidase activity was assayed in the guinea pig pulmonary parenchyma after removal of bronchoalveolar cells by washing. After differential centrifugation of the crude tissue homogenate, sialidase activities were measured in the subcellular fractions using the fluorogenic substrate 2-(4-methylumbelliferyl)-alpha-D-N-acetylneuraminate. Sialidase activities were found in the lysosomal-enriched (17,000 x g pellet), in the microsomal (105,000 x g pellet) and in the cytosolic (105,000 x g supernatant) fractions. Microsomal and lysosomal forms of sialidase had an optimum activity at pH 3.6-3.8, whereas the optimum for the cytosolic form was pH 4.6. The activity of all three forms was inhibited by Cu2+, whereas 1 mM Zn2+ and 0.5 mM Ca2+ activated the lysosomal and the cytosolic forms, respectively. In the crude homogenate taken from lungs of Bacillus Calmette Guérin-(BCG-) stimulated guinea pigs, the sialidase activity was increased by 43% (p = 0.025) 3 weeks after the end of the treatment. The cytosolic (+246%) and microsomal (+51%) sialidase activities were significantly increased, whereas the lysosomal sialidase activity was not changed significantly by BCG stimulation.     

Degradation of somatomedin A by various organ homogenates from rats

Degradative activities of somatomedin A (SMA) have been examined in various tissue homogenates of rat using trichloracetic acid precipitable radioactivity of 125I-SMA. Kidney and testis showed higher specific activities and liver and brain lower activities. They were dependent on SH reagents; 0.5 mM HgCl2 inhibited the degradative activity of liver completely and 1 mM dithiothreitol (DTT) augmented the activity slightly. The activities in liver were separated by differential centrifugation; about 90 per cent of the total activity in the whole homogenate was recovered in the supernatant fraction at 100,00 x g for 60 min, and 10 per cent in the precipitate. The pH profile of each fraction was different; that of the supernatant showed a single peak at pH 7.4 and that of the pellet revealed two peaks at pH 5.9 and 7.4. However, both fractions showed similar SH-dependency.     

Production and purification of L-phenylalanine oxidase from Morganella morganii.

An enzyme fraction, acting predominantly on L-phenylalanine has been purified and characterized from Morganella morganii. The total envelope was prepared by disrupting the cells with a French press followed by high speed centrifugation. After solubilization of the particulate fraction with 0.1% Triton X-100 and then centrifugation, the resulting supernatant was layered onto a DEAE-Cellulose column. Active fractions eluted were applied to a Phenyl-Sepharose CL-4B column as the final purification step. The activity of the purified enzyme to various L-amino acids in decreasing order was phenylalanine, methionine, leucine, tryptophan, and to a much lesser extent cysteine and tyrosine. At 4 degrees C in 20 mM phosphate buffer pH 7.5, the partially purified fractions collected from the DEAE-Cellulose column were stable for 120 h. On the other hand, the purified fractions obtained from the Phenyl Sepharose CL-4B column showed a drastic decrease in activity within only 24 h. Mg2+ (up to 40 mM), Mn2+ or Ca2+ (up to 10 mM) stimulated the oxidation of the purified enzyme but increases beyond such levels decreased the enzyme activity. Co2+ (0.05 mM), Cu2+ (0.5 mM) or Zn2+ (0.1 mM) decreased the enzyme activity 37, 33 and 20%, respectively.     

The aromatization of cyclohexanecarboxyl-CoA to hippuric acid by guinea pig liver mitochondria: submitochondrial localization

Summary The conversion of cyclohexanecarboxyl-CoA to hippuric acid in submit ochondrial fractions from guinea pig liver was studied using a gas chromatographic-mass spectrometric method employing selected ion monitoring. Comparison of the activities of the cyclohexanecarboxyl-CoA to hippuric acid converting system (CCoAHC-system) and marker enzymes in the various submit ochondrial fractions showed that the CCoAHC-system is localized in the mitochondrial matrix. Partial separation of the inner and outer membranes has been accomplished by treating mitochondria with digitonin in isotonic medium and fractionating the treated mitochondria by differential centrifugation. A digitonin-protein ratio of 2.6 mg of digitonin/10 mg of protein must be used in order to release significant amounts of amine oxidase activity (outer membrane marker) from low speed mitochondrial pellets. This pellet still contained most of the glutamate dehydrogenase activity and was insignificantly contaminated with adenylate kinase. Moderate concentrations of phenazine methosulfate (PMS) greatly stimulated the activity of the CCoAHC-system, even in intact mitochondria (optimal concentration of PMS: 1 mM) whilst higher concentrations (> 1 mM) decreased the activity. The formation of hippuric acid in these mitochondrial preparations was linear with time for at least 40 min and linear with respect to protein concentration up to approximately 2.0 mg mitochondrial protein·m¹.     

Protein substrate specificity of a calmodulin-dependent protein kinase isolated from bovine heart

The protein substrate specificity of a calmodulin-dependent protein kinase activity from the cytosolic fraction of bovine heart was examined. Prior to the experiments, the kinase activity was purified more than 50-fold with a recovery of greater than 10% of the homogenate activity. Two endogenous protein substrates of molecular weight 57,000 and 73,000 were phosphorylated in these kinase preparations. The kinase preparation was also able to phosphorylate exogenous synapsin, phospholamban, glycogen synthase, MAP-2, myelin basic proteins and κ-casein, but not tubulin, pyruvate kinase, the regulatory subunit of cAMP protein kinase II, myosin light chain or phosphorylase b. High levels of calmodulin were required for activation of the kinase activity toward the 57,000 and 73,000 molecular weight endogenous substrates (K_0.5 = 93 +/- 5 nM), glycogen synthase (K_0.5 = 127 +/- 10 nM), and κ-casein (K_0.5 = 321 +/- 107 nM). The kinase possessed a high affinity for glycogen synthase (half maximal activity at 0.9 +/- 0.4 μM) but a low affinity for κ-casein (21 +/- 2 μM). Sucrose density gradient centrifugation separated the calmodulin-dependent protein kinase activity into two fractions with apparent molecular weights of approximately 900,000 and 100,000. Both fractions phosphorylated the endogenous 57,000 molecular weight substrate and glycogen synthase similarly. These results indicate that cardiac calmodulin-dependent protein kinase previously observed to phosphorylate endogenous protein substrate possesses a wide range of substrate specificity.     

Partial purification and properties of UDP-N-acetylmannosamine:N-acetylglucosaminyl pyrophosphorylundecaprenol N-acetylmannosaminyltransferase from Bacillus subtilis

An enzyme which catalyzes the conversion of GlcNAc-PP-undecaprenol into ManNAc(beta 1----4)GlcNAc-PP-undecaprenol, a key lipid intermediate in the de novo synthesis of various teichoic acids, was partially purified from the 20,000 x g supernatant fraction of Bacillus subtilis AHU 1035 cell homogenate. By means of ammonium sulfate precipitation, gel chromatography, and ion-exchange chromatography, the enzyme was purified about 70-fold, giving a preparation virtually free from substances obstructive to measurement of the N-acetylmannosaminyltransferase reaction. The enzyme was shown to be specific to UDP-ManNAc. The Km value for UDP-ManNAc was 4.4 microM, and the optimum pH was 7.3. The enzyme required 10 mM MgCl2, 0.3 M KCl, 25% glycerol, and 0.1% Nonidet P-40 to function at full activity.     

An improved method for the extraction of corticosterone from cell homogenates and subcellular fractions of the rat adrenal cortex

An improved technique is described for the extraction and analysis of corticosterone (11β,21-dihydroxy-4-pregnene-3,20-dione) from homogenates and subcellular fractions of the rat adrenal cortex. Factors influencing complete extraction of corticosterone were the nature of the organic solvent system and the concentration of the tissue being extracted. The continued activity of steroidogenic enzymes during subcellular fractionation was prevented by 0.1 mM 1-benzylimidazole. For optimum extraction, homogenates were diluted 1:12 (v/v) in 0.25 M sucrose, containing 0.1 M potassium hydroxide. Dilute homogenate was mixed with absolute ethanol (1:10, v/v) and extracted three times with diethyl ether (1:5, v/v). Following extraction, corticosterone in each sample was isolated by thin-layer chromatography (TLC), quantitated by radioimmunoassay (RIA), and corrected by measuring the recovery of added ³H corticosterone. With these procedures, 90–100% of corticosterone found in extracts of adrenal homogenates was recovered in extracts of subcellular fractions of the homogenates.     

Mutation induction by rodent liver microsomal metabolities of aflatoxins B1 and G1 in Neurospora crassa.

The mutagenic activities of aflatoxins B1 and G1 were studied in the ad-3 test system of Neurospora crassa by treatment of conidia with aflatoxin and liver homogenate for 2 h. No significant increase in the ad-3 mutation frequency over the spontaneous frequency was observed when either aflatoxin or mammalian liver homogenate was omitted from the test system. The ad-3 mutation frequencies increased to between 29 and 87/10(6) survivors, which is a 73- to 217-fold increase over the average spontaneous ad-3 mutation frequency (0.4/10(6) survivors), after conidia of N. crassa were treated with 0.67 mM aflatoxin B1, hamster liver homogenate, and a NADPH generating system. A 9- to 15-fold increase in the mutation frequency over the spontaneous mutation frequency was found when 0.67 mM of aflatoxin G1 instead of aflatoxin B1 was used in the test system. Treatment of conidia with 0.44 mM aflatoxin B1 mice liver homogenate and a NADPH generating system caused a small, but significant increase in the ad-3 mutation frequencies. No significant increase in the mutation frequency was found when a single sample of human liver homogenate was used in the test system. These studies show that metabolic activation is necessary for the expression of the mutagenic activity of aflatoxins B1 and G1 in N. crassa.     

Purification of alcohol dehydrogenase from bovine liver crude extract by dye-ligand affinity counter-current chromatography

Alcohol dehydrogenase (ADH) was extracted from a crude bovine liver homogenate by dye-ligand affinity counter-current chromatography (CCC) using a cross-axis coil planet centrifuge (x-axis CPC). The purification was performed using two types of polymer phase systems composed of 4.4% polyethylene glycol (PEG) 8000-7.0% dextran T500-0.1 M potassium phosphate buffers and 16% PEG 1000-12.5% potassium phosphate buffers, both containing a procion red dye as an affinity ligand at various pH values. The best purification was achieved using the PEG 1000-potassium phosphate system at pH 7.3 containing 0.05% procion red as a ligand. The upper PEG-rich phase containing procion red was used as the stationary phase and a crude bovine liver homogenate was eluted with the potassium phosphate-rich lower phase at 0.5 ml/min. After elution of bovine liver proteins in the homogenate, ADH still retained in the stationary phase was collected from the column by eluting with the PEG 1000-rich upper phase. Collected fractions were analyzed by ADH enzymatic activity and by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) to detect contaminant proteins in the ADH fractions. The ADH was purified directly from crude bovine liver extract within 6h with minimum loss of its enzymatic activity.     

3-hydroxy-3-methylglutaryl coenzyme A reductase from rat intestine: subcellular localization and in vitro regulation

The subcellular localization of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase in rat intestine was reinvestigated. Highly enriched fractions of endoplasmic reticulum and mitochondria were prepared from mucosal cells. The highest specific activity of HMG-CoA reductase was located in the endoplasmic reticulum fraction with recovery of 25% of the total activity. The mitochondria had low specific activity and low recovery of reductase activity relative to whole homogenate (2-5%). Despite attempts to maximize cell lysis, much of the activity (about 60%) was recovered in a low speed pellet which consisted of whole cells, nuclei, and cell debris as determined by light microscopy. Taken together, the evidence strongly suggests that much of the cellular HMG-CoA reductase activity is present in the endoplasmic reticulum fraction and that mitochondria have little or no intrinsic HMG-CoA reductase. The in vitro regulation of intestinal microsomal HMG-CoA reductase was studied. The intestine possesses a cytosolic HMG-CoA reductase kinase-phosphatase system which appears to be closely related to that present in the liver. Intestinal reductase activity in microsomes prepared from whole mucosal scrapings was inhibited 40-50% by the presence of 50 mM NaF in the homogenizing buffer. It was less susceptible to the action of the kinase than liver reductase. The effects of NaF were reversed by incubation with partially purified intestinal or liver phosphatases. These results suggest that the kinase-phosphatase system could play a role in the regulation of intestinal sterol and isoprene synthesis in vivo.     

Properties of guanylate cyclase from rat kidney cortex and transplantable kidney tumors.

The subcellular distribution and properties of guanylate cyclase was examined in preparations of normal rat renal cortex and Morris renal tumors MK2 and MK3. In normal kidney cortex about two-thirds of guanylate cyclase activity of homogenates was found in soluble fractions. With renal tumors the homogenate activity was less and the enzyme was equally divided between particulate and soluble fractions. The particulate enzyme in kidney cortex and tumors was associated with all particulate fractions. Triton X-100 increased the activity of all preparations. All preparations preferred Mn2+ as the sole cation. The stimulatory effects of Ca2+ on soluble enzyme and inhibitory effects on particulate activity were similar with preparations of renal cortex and tumors. ATP inhibited all preparations. Soluble and particulate guanylate cyclases from renal cortex were activated several-fold with 1 mM NaN3. Preparations of tumor enzymes did not respond to NaN3. Thus, compared to normal renal cortex the subcellular distribution of guanylate cyclase and some of its properties are altered in preparations of renal tumors.     

Mobilization of iron from crocidolite asbestos by certain chelators results in enhanced crocidolite-dependent oxygen consumption.

The reactivity of iron on crocidolite asbestos with dioxygen was determined and compared with iron mobilized from crocidolite. Ferrozine, a strong Fe(II) chelator, was used to demonstrate that iron on crocidolite was redox active. More Fe(II) was mobilized from crocidolite (1 mg/ml) by ferrozine anaerobically (11.2 nmol/mg crocidolite/h) than aerobically (6.6 nmol/mg/h) in 50 mM NaCl, pH 7.5, suggesting that Fe(II) on crocidolite reacts with O2 upon aqueous suspension. However, suspension of crocidolite in 50 mM NaCl, pH 7.5, did not result in a measurable amount of O2 consumption. The addition of reducing agents (1 mM) increased the amount of Fe(II) on crocidolite, and addition of ascorbate resulted in 0.4 nmol O2 consumed/mg crocidolite/min. Therefore, iron on crocidolite had limited redox activity in the presence of ascorbate. However, mobilization of iron from crocidolite increased its redox activity. Citrate, nitrilotriacetate (NTA), or EDTA (1 mM) mobilized 79, 32, or 58 microM iron, respectively, in preincubations up to 76 h, and increased O2 consumption upon addition of ascorbate to 2.8, 7.6, or 22.0 nmol O2 consumed/mg/min, respectively. This activity depended only upon the presence of a component(s) mobilized from crocidolite by the chelators. Pretreatment of crocidolite with the iron chelator desferrioxamine B (10 mM) inhibited O2 consumption. The results of the present study suggest that iron on or in crocidolite is responsible for the redox activity of crocidolite, but that mobilization of iron by chelators such as citrate, NTA, or EDTA greatly enhances its redox activity. Thus, iron mobilization from crocidolite in vivo by low-molecular-weight chelators may lead to the increased production of reactive oxygen species which may damage biomolecules, such as DNA.     

Association of rapidly-labelled RNAs with actin in nuclear matrix from mouse L5178Y cells

More than 90% of rapidly-labelled nuclear RNA was associated with a nuclear matrix prepared from mouse leukemia L5178Y cells. The binding was not affected with up to 4 M NaCl; however, these RNAs were released from the nuclear matrix by treatment with a low ionic strength buffer (5 mM Tris-HCl buffer, pH 7.5, containing 1 mM ATP, 1 mM dithiothreitol, 0.2 mM ethylenediaminetetraacetic acid (EDTA) and 0.4 mM calcium chloride), without destruction of the sphere of the nuclear matrix. Actin filaments in the nuclear matrix were depolymerized with this buffer accompanied with rapidly-labelled RNAs. When the depolymerization was inhibited by slight modifications of the low ionic strength buffer (replacement of ATP by the same concentration of GTP; replacement of calcium ion by the same concentration of magnesium ion; addition of 20 micrograms/ml of phalloidine, which is a specific inhibitor of actin depolymerization), the release of rapidly-labelled RNAs from the nuclear matrix was also inhibited. The complex containing rapidly-labelled RNAs and matrix proteins was solubilized by a sonication from the nuclear matrix, and subjected to cesium chloride equilibrium centrifugation. Rapidly-labelled RNAs were concentrated on the bottom of the gradient accompanied with a small number of proteins (68K, 60K, 43K and 40K). The 43K protein was identified as actin by immunoblotting. By RNase digestion before equilibrium centrifugation, actin in the bottom fractions disappeared. These results suggest that rapidly-labelled RNAs anchor on the actin filaments in the nuclear matrix.