A rho gene product in human blood platelets. I. Identification of the platelet substrate for botulinum C3 ADP-ribosyltransferase as rhoA protein.

A substrate protein for botulinum C3 ADP-ribosyltransferase (C3 exoenzyme) in human platelets was purified to apparent homogeneity from the cytosol by ammonium sulfate fractionation and successive chromatography on columns of DEAE-Sepharose, hydroxylapatite, phenyl-Sepharose, and TSK phenyl-5PW. The purified protein yielded an amino acid sequence identical to that of rhoA protein. When platelet cytosol and membranes were incubated with C3 exoenzyme and [32P]NAD and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and isoelectric focusing, they gave only one [32P]ADP-ribosylated band on each electrophoresis that showed an M(r) of 22,000 and a pI of 6.0. The radioactive bands from the two fractions co-migrated with each other and with the [32P]ADP-ribosylated purified protein. When these radioactive products were partially digested with either alpha-chymotrypsin or trypsin and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, the same digestion pattern was found in the three samples. These results suggest that the ADP-ribosylation substrate for C3 exoenzyme in the platelet cytosol and membrane is rhoA protein and that it is the sole substrate detectable in human platelets.     

Catechol-O-methyl transferase in mouse liver plasma membranes.

Catechol-o-methyl transferase is usually localized predominantly in the cytosol fraction of cells, but fractionation of mouse liver showed plasma membranes contain ～ 70% of the total enzyme activity and have a specific activity ～ 10x greater than the cytosol fraction. Treatment of the membrane fraction with Lubrol-PX solubilized 47% of the membrane protein and 95% of the enzyme activity. A comparison of Lubrol-solubilized enzyme and [³H]norepinephrine binding activities in a variety of experimental conditions suggest binding is not related to interaction with the active site of catechol-o-methyl transferase. Isoelectric focusing of solubilized membrane proteins showed the enzyme has an isoelectric pH of 4.5-4.8.     

Cytosol promotes the guanine nucleotide-induced release of the alpha subunit of Gi2 from the membrane of mouse mastocytoma P-815 cells

Translocation of the alpha subunit of Gi2 from the membrane to the cytosol was studied in mouse mastocytoma P-815 cells. To monitor Gi2 alpha the membrane (300,000 x g pellet) was [32P]ADP-ribosylated with pertussis toxin. Incubation of the [32P]ADP-ribosylated membrane with guanosine 5'-O-(3-thiotriphosphate) (GTP gamma S) caused a small release (10%) of [32P]ADP-ribosylated Gi2 alpha from the membrane. Whereas cytosol (300,000 x g supernatant) alone had no ability to release the [32P]ADP-ribosylated Gi2 alpha from the membrane, it markedly augmented the release induced by GTP gamma S, about 50% of the total [32P]ADP-ribosylated Gi2 alpha being released by 30 min. The GTP gamma S-induced release and its enhancement by the cytosol were specific for GTP and GTP gamma S. When the cytosol was boiled this promoting activity was lost. The [32P]ADP-ribosylated Gi2 alpha released by the cytosol plus GTP gamma S from the membrane was eluted as a single peak corresponding to a molecular weight of about 100,000 from an Ultrogel AcA 44 column. In contrast, the [32P]ADP-ribosylated Gi2 alpha released by GTP gamma S alone was eluted at the position of Mr = 40,000, but it was eluted at the position of Mr = about 100,000 when it was incubated with the cytosol. Furthermore, Gi2 alpha purified from bovine lung also behaved in a similar way on gel filtration. The addition of thrombin, a stimulant of histamine secretion from mast cells, to mastocytoma cells drastically induced the translocation of Gi2 alpha from the membrane to the cytosol in a pertussis toxin-sensitive manner. These results taken together demonstrate that the cytosol contains some factor(s) that promotes the release of GTP-activated Gi2 alpha from the membrane and that the released Gi2 alpha exists in the cytosol as a soluble complex with unidentified component(s) in mastocytoma cells.     

Separation of the 24 kDa substrate for botulinum C3 ADP-ribosyltransferase and the cholera toxin ADP-ribosylation factor

Botulinum C3 ADP-ribosyltransferase modifies a approximately 24 kDa membrane protein believed to bind guanine nucleotides. Cholera toxin ADP-ribosylation factors are approximately 19 kDa GTP-binding proteins that directly activate the toxin. To evaluate a possible relationship between C3 ADP-ribosyltransferase substrate and ADP-ribosylation factor, they were partially purified from bovine brain. ADP-ribosylation factor, but not C3 ADP-ribosyltransferase substrate, stimulated auto-ADP-ribosylation of the choleragen A1 subunit whereas C3 ADP-ribosyltransferase substrate, but not ADP-ribosylation factor, was ADP-ribosylated by C3 ADP-ribosyltransferase. Thus, although both may be GTP-binding proteins, no functional similarity between ADP-ribosylation factor and C3 ADP-ribosyltransferase substrate was found.     

Purification and properties of the cytosolic substrate for botulinum ADP-ribosyltransferase. Identification as an Mr 22,000 guanine nucleotide-binding protein

The substrate for ADP-ribosyltransferase from Clostridium botulinum was purified from the cytosol of bovine adrenal gland. Purification procedures consisted of ammonium sulfate fractionation, chromatographies on columns of DEAE-Sepharose and phenyl-Sepharose, gel filtration on a TSK-gel G3000SW column, and Mono Q fast protein liquid chromatography. On DEAE-Sepharose chromatography, the substrate activity was eluted in two separate peaks, and electrophoretic analyses revealed that the substrates in the two peaks are of similar molecular weight but different isoelectric points. The major peak of the substrate was further purified. It was purified about 1,800-fold with a recovery of 2.2% by the above procedures. On sodium dodecyl sulfate-polyacrylamide gel electrophoresis, the final preparation showed a single protein band at Mr 22,000. The purified protein served as a substrate for botulinum ADP-ribosyltransferase and was maximally ADP-ribosylated to the extent of about 0.7 mol of ADP-ribose/mol of protein. A guanosine 5'-(3-O-thio)triphosphate (GTP gamma S) binding activity was co-purified with the ADP-ribosylation substrate, and the purified protein maximally bound about 0.5 mol of GTP gamma S/mol. GTP gamma S binding was effectively competed by GTP and GDP but not by GMP, ATP, and ADP. Thus, the ADP-ribosylation substrate is a GTP-binding protein. This protein, designated Gb (b for botulinum), is widely distributed in various tissues. It was rich in brain, pituitary, and adrenal glands, and poor in heart, smooth, and skeletal muscles.     

Activator protein supporting the botulinum ADP-ribosyltransferase reaction

The ADP-ribosyl moiety of NAD was transferred to proteins with Mr values of 22,000 and 25,000 when bovine brain cytosol was incubated with a botulinum ADP-ribosyltransferase C3 (BT-C3) which was purified from the culture medium of a type C strain of Clostridium botulinum. Any protein fraction eluted from a chromatographic column to which the cytosol had been applied, however, was not significantly ADP-ribosylated by BT-C3, unless the reaction mixture was further supplemented with a small amount of the cytosol. Thus, substrate protein(s) could be partially purified based on their ability to be ADP-ribosylated by BT-C3 in the presence of the cytoplasmic activator(s). The rate of ADP-ribosylation of the substrates was extremely low by itself but was increased enormously and progressively when increasing amounts of cytosol were added, affording a reliable means for assay of the activator contained therein. The activator was separated from the substrate proteins and partially purified from the cytosol by sequential chromatography steps with an anion exchanger and a gel filtration column. The activity of the partially purified activator was heat-labile and protease-sensitive, suggesting that the activator was a protein or had a protein component necessary for activity. The action of the activator protein(s) was specific for BT-C3-catalyzed ADP-ribosylation; cholera toxin-catalyzed ADP-ribosylation of GTP-binding protein (Gs) was not supported by this activator. Thus, this is the first report to show that botulinum ADP-ribosyltransferase-catalyzed reaction can proceed significantly only in the presence of other protein factor(s), just as has been observed with an ADP-ribosylation factor required for cholera toxin-induced similar reaction.     

ADP-ribosylation of platelet actin by botulinum C2 toxin

Botulinum C2 toxin is a microbial toxin which possesses ADP-ribosyltransferase activity. In human platelet cytosol a 43-kDa protein was ADP-ribosylated by botulinum C2 toxin. Labelling of the 43-kDa protein using [32P]NAD as substrate was reduced by unlabelled NAD and nicotinamide. The label was removed by treatment with snake venom phosphodiesterase. Half-maximal and maximal ADP-ribosylation occurred at 0.1 microgram/ml and 3 micrograms/ml botulinum C2 toxin, respectively. The Km value of the ADP-ribosylation reaction for NAD was about 1 microM. The peptide map of the ADP-ribosylated 43-kDa protein was almost identical with platelet actin. The ADP-ribosylated 43-kDa substrate protein bound to and was eluted from immobilized DNase I in a manner similar to G-actin. Trypsin treatment of platelet cytosol decreased subsequent ADP-ribosylation of the 43-kDa protein without occurrence of smaller labelled polypeptides. Purified platelet actin was also ADP-ribosylated by botulinum C2 toxin with similar characteristics found with actin in platelet cytosol. Phalloidin decreased the ADP-ribosylation of actin in platelet cytosol and of isolated platelet actin. Half-maximal and maximal, about 90%, reduction of actin ADP-ribosylation was observed at 0.4 microM and 10 microM phalloidin, respectively. ADP-ribosylation of purified actin, induced by botulinum C2I toxin, abolished the formation of the typical microfilament network. The data indicate that platelet G-actin but not F-actin is a substrate of botulinum C2 toxin and that this covalent modification largely affects the functional properties of actin.     

Elongation factor 2 as the major substrate for Ca2+/calmodulin-dependent protein kinase in rat adrenal glomerulosa cells

We have previously shown the existence of the major substrate protein of Mr 100,000 (substrate 100 K protein) for Ca2+/calmodulin (CaM)-dependent protein kinase in rat adrenal glomerulosa cells. In the present study, the identity of the substrate 100 K protein to elongation factor 2 (EF-2) was investigated. In a 105,000 g-supernatant fraction (cytosol), the protein of Mr 100,000 with the pI (isoelectric point) value of 6.7 was phosphorylated in the presence of calcium and CaM. The optical densities of this phosphorylated band were greatly enhanced in the presence of the EF-2 purified from pig liver (1 microgram) [20-23-fold, n = 5] when compared with those in the absence of the component. In the presence of the purified EF-2, the phosphorylation of Mr 100,000 was detected only in the presence of calcium alone or calcium plus CaM. This phosphorylation in the presence of calcium alone was completely inhibited in the presence of the CaM antagonist pimozide (500 microM), showing the existence of endogenous CaM in the cytosol. In the same fraction, the ADP-ribosylated protein of Mr 100,000 was detected in the presence of diphtheria toxin (fragment A) and (adenylate-32P) NAD, indicating the presence of EF-2 in the cytosol from rat adrenal glomerulosa cells. These results suggest that the substrate 100 K protein may be identical to EF-2 in rat adrenal glomerulosa cells.     

Chemotactic peptide receptor-supported ADP-ribosylation of a pertussis toxin substrate GTP-binding protein by cholera toxin in neutrophil-type HL-60 cells

A 40-kDa protein, in addition to the alpha-subunits of Gs (a GTP-binding protein involved in adenylate cyclase stimulation), was [32P]ADP-ribosylated by cholera toxin (CT) in the membranes of neutrophil-like HL-60 cells, only if formyl Met-Leu-Phe (fMLP) was added to the ADP-ribosylation mixture. The 40-kDa protein proved to be the alpha-subunit of Gi serving as the substrate of pertussis toxin, islet-activating protein (IAP). No radioactivity was incorporated into this protein in membranes isolated from HL-60 cells that had been exposed to IAP. Gi-alpha purified from bovine brain and reconstituted into IAP-treated cell membranes was ADP-ribosylated by CT plus fMLP. Gi-alpha was ADP-ribosylated by IAP, but not by CT plus fMLP, in membranes from cells that had been pretreated with CT plus fMLP. When membrane Gi-alpha [32P]ADP-ribosylated by CT plus fMLP or IAP was digested with trypsin, the radiolabeled fragments arising from the two proteins were different from each other. These results suggest that CT ADP-ribosylates Gi-alpha in intact cells when coupled fMLP receptors are stimulated and that the sites modified by two toxins are not identical. CT-induced and fMLP-supported ADP-ribosylation of Gi-alpha was favored by Mg2+ and allow concentrations of GTP or its analogues but suppressed by GDP. The ADP-ribosylation did not occur at all, even in the presence of ADP-ribosylation factor that supported CT-induced modification of Gs, in phospholipid vesicles containing crude membrane extract in which Gi was functionally coupled to stimulated fMLP receptors. Thus, Gi activated via coupled receptors is the real substrate of CT-catalyzed ADP-ribosylation. This reaction may depend on additional factor(s) that are too labile to survive the process of membrane extraction.     

The 65-kDa protein from pig heart. A new substrate for Clostridium botulinum ADP-ribosyltransferase (exoenzyme C3)

In the pig heart sarcolemma, a 65 kDa protein is found to be ADP-ribosylated by Clostridium botulinum ADP-ribosyltransferase (exoenzyme C3). ADP-ribosylation of this protein is regulated by guanyl nucleotides and cytosol factor in a fashion similar to that for other C3 substrates. The new exoenzyme C3 substrate was partially purified. This protein is supposed to be a GTP-binding one.     

ADP-ribosylation of a Mr 21,000 membrane protein by type D botulinum toxin

When crude membrane fraction from bovine adrenal gland was incubated with type D botulinum toxin in the presence of NAD, a membrane protein with a molecular weight of 21,000 was specifically ADP-ribosylated. This ADP-ribosylation occurred dependent on the dose of the toxin and was abolished by prior boiling ADP-ribose transfer to the membrane protein was significantly suppressed when agmatine and L-arginine methyl ester were included in the reaction mixture. Dithiothreitol stimulated this ADP-ribosylation about 3-fold. Incubation of membrane fractions from mouse brain and pancreas with this toxin also resulted in ADP-ribosylation of a protein of the same molecular weight. These results suggested that type D botulinum toxin catalyzed transfer of an ADP-ribose moiety of NAD to the specific membrane protein common to secretory cells.     

Heterogeneity of native rat liver elongation factor 2

The high heterogeneity of native rat liver EF-2 prepared from either 105000 x g supernatant or microsome high-salt extract was detected by two-dimensional equilibrium isoelectric focusing-SDS-polyacrylamide gel electrophoresis in the presence of 9.5 M urea. Five spots were always detected, all of Mr 95,000, which were not artefactual for their amount varied when EF-2 was specifically ADP-ribosylated by diphtheria toxin in the presence of NAD+, and/or phosphorylated on a threonine residue by a Ca2+/calmodulin-dependent protein kinase (most likely Ca2+/calmodulin-dependent protein kinase III described by others [(1987) J. Biol. Chem. 262, 17299-17303; (1988) Nature 334, 170-173]). Results of ADP-ribosylation and/or phosphorylation experiments with either unlabeled or labeled reagents ([14C]NAD and [32P]ATP) strongly suggest that our preparation contained native ADP-ribosylated and native phosphorylated forms which could be estimated at about 20% and 40% of the whole EF-2. Phosphorylated and ADP-ribosylated forms of EF-2 could be ADP-ribosylated and phosphorylated, respectively, but a native form both ADP-ribosylated and phosphorylated was not detected. Our results also suggest the existence of a minor native form of EF-2 and of its phosphorylated and ADP-ribosylated derivatives.     

Demonstration of GTP-binding proteins and ADP-ribosylated proteins in rat liver Golgi fraction

A Golgi-rich fraction isolated from rat liver was found to contain GTP-binding proteins with 20-25 kDa, which were tightly bound to the Golgi membrane. The Golgi fraction also contained two species of proteins which were ADP-ribosylated by bacterial toxins. Protein(s) which was ADP-ribosylated by botulinum toxin had a similar molecular mass as those with GTP-binding activity but was easily released from the membrane. Another protein with 46 kDa which was ADP-ribosylated by pertussis toxin was tightly bound to the membrane but had no significant GTP-binding activity under conditions tested here. These proteins were much less or negligible in the plasma membrane and the endoplasmic reticulum.     

Purification of the 22 kDa protein substrate of botulinum ADP-ribosyltransferase C3 from porcine brain cytosol and its characterization as a GTP-binding protein highly homologous to the rho gene product

The 22 kDa protein substrate of botulinum ADP-ribosyltransferase C3 was purified from porcine brain cytosol by acetone precipitation, CM-Sephadex, octyl-Sepharose and TSK phenyl-5PW HPLC chromatography to apparent homogeneity. ADP-ribosylation of the protein was increased by guanine nucleotides (GTP, GDP, GTP gamma S, each 100 microM) but not by GMP, ATP or ATP gamma S. The purified 22 kDa protein bound maximally 0.9 mol [35S]GTP gamma S and hydrolyzed GTP with the rate 0.007 mol per mol protein. Amino acid sequences were obtained from two tryptic peptides, selected from an in situ digestion of Immobilon electrotransferred, gel purified ADP-ribosylated substrate. The two sequences obtained, cover 23 residues from the corresponding sequences in human rho.     

A cAMP-binding phosphoprotein in Drosophila heads is similar to the regulatory subunit of the mammalian type II cAMP-dependent protein kinase

Cyclic AMP (cAMP)-dependent regulation of in vitro phosphorylation of several proteins including a cAMP-binding protein was studied with crude membrane and cytosol fractions from Drosophila heads. Phosphorylation of at least seven distinct proteins was enhanced in the presence of cAMP. Interestingly, however, the phosphorylation of a 56 kDa protein was apparently reduced by cAMP in the membrane but not in the cytosol fraction. The following data strongly indicate that the 56 kDa phosphoprotein in both membrane and cytosol fractions is a cAMP-binding protein, very similar to the regulatory subunit (RII) of a mammalian cAMP-dependent protein kinase, and that its binding to cAMP makes this protein very susceptible to the action of phosphatases: (i) cAMP highly stimulated the dephosphorylation of the 56 kDa phosphoprotein by the endogenous phosphatase in the membrane fraction. (ii) The dephosphorylation of a similar 56 kDa phosphoprotein in the cytosol fraction by an exogenous, cAMP-independent, alkaline phosphatase was also highly stimulated by cAMP. (iii) The 56 kDa phosphoprotein was covalently bound to cAMP by u.v. irradiation. (iv) The alkaline-phosphatase treatment reversibly converted this phosphoprotein to a 53 kDa non-phosphorylated protein. (v) The 53 kDa protein was selectively bound to cAMP-agarose and subsequently eluted by cAMP and high salt. (vi) This protein served as a substrate for the catalytic subunit of a mammalian cAMP-dependent protein kinase.     

Botulinum ADP-ribosyltransferase activity as affected by detergents and phospholipids

GTP-binding proteins with Mr values of 22,000 and 25,000 in bovine brain cytosol were ADP-ribosylated by an exoenzyme (termed C3) purified from Clostridium botulinum type C. The rate of C3-catalyzed ADP-ribosylation of the partially purified substrates was extremely low by itself, but was increased enormously when a protein factor(s) obtained from the cytosol was simultaneously added. The rate of the C3-catalyzed reaction was also stimulated by the addition of certain types of detergents or phospholipids even in the absence of the protein factors. The ADP-ribosylation appeared to be enhanced to an extent more than the additive effect of either the protein factors or the detergents (and phospholipids). Thus, ADP-ribosylation catalyzed by botulinum C3 enzyme was affected not only by cytoplasmic protein factors but also by detergents or phospholipids in manners different from each other.     

Subcellular fractionation of porcine neutrophils by nitrogen cavitation and sucrose-density-gradient centrifugation

Neutrophils, isolated in large quantities from porcine blood were disrupted by nitrogen cavitation and separated by differential centrifugation into a nuclear fraction and a post-nuclear supernatant. The latter was subfractionated by sucrose density gradient centrifugation into cytosol, a fraction consisting of membrane vesicles and two granule-rich fractions. The membrane fraction accounted for 1.9% of the protein in the post-nuclear supernatant, the light granule fraction for 2.2% and the dense granule fraction for 4.2%. Catalase, lactate dehydrogenase and malate dehydrogenase were largely confined to the cytosol. The dense granule fraction contained the highest quantities of the hydrolytic enzymes, although the membrane fraction was also rich in alkaline and acid phosphatase and gamma-glutamyl transpeptidase activities. Electron microscopy of the membrane fraction showed intact membrane vesicles, whereas the granular fractions consisted of electron-dense, membrane-bound granules. Two granular fractions were isolated which contained granules of differing size and density. 3H-labeled wheat germ agglutinin bound to the surface of intact neutrophils and when these were disrupted and fractionated the membrane fraction showed a specific binding activity 16-times greater than that of the cavitated sample. The membrane fraction interacted with the detergent digitonin and as a result underwent density perturbation increasing from 1.13 g X cm-3 to 1.18 g X cm-3. Dodecylsulphate-polyacrylamide gel electrophoresis showed the membrane fraction to consist of at least 40 protein bands, with relative molecular masses ranging from 200 000-16 000. The granule fractions contained less protein bands, with a protein composition quite distinct from that of the membrane fraction.     

Annexin proteins PP4 and PP4-X. Comparative characterization of biological activities of placental and recombinant proteins.

Several G-proteins (GTP-binding proteins) were identified by SDS/PAGE in the cytosol (105,000 g supernatant) and membrane fractions of the oestrogen-dependent human mammary-tumour cell line ZR-75-1. These proteins, with molecular masses in the range 18-29 kDa, specifically bind [alpha-32P]GTP, which can be displaced by unlabelled GTP, GDP and their non-hydrolysable analogues guanosine 5'-[delta-thio]triphosphate (GTP[S]) and guanosine 5'-[beta-thio]diphosphate (GDP[S]), but not by GMP, ATP, ADP, AMP and other unrelated nucleotides. The apparent dissociation constant for GTP was approx. 2 x 10(-8)M. Homogenization of ZR-75-1 cells in high-salt buffer (1 M-KCl), and successive washing of the membrane fraction, suggested that, among the major G-proteins found, the 18 kDa protein is predominantly soluble, whereas the 27-29 kDa complex is primarily bound to the membrane fraction under the experimental conditions employed. Possible translocation of these G-proteins between membrane and cytosol was analysed. No redistribution of the 27-29 kDa complex was observed, whereas GTP[S] in the presence of Mg2+ caused apparent translocation of the 18 kDa protein to the membrane fraction. This effect was specific for GTP and stable GTP analogues, whereas GDP, GMP, ATP, ADP, AMP and other unrelated nucleotides were ineffective. GTP[S] and guanosine 5'-[beta gamma-imido]-triphosphate (p[NH]ppG) were equally potent (apparent Kd approximately 5 x 10(-6)M), whereas GTP was rather weak. The nucleotide effect is temperature-, time- and concentration-dependent. The translocation process was reversible, slow, and reached its maximum between 30 and 60 min at 37 degrees C. The apparent translocation of this small G-protein from the cytosol to the membrane fraction, and the specific effect of GTP analogues, suggest that this process may have functional significance in mammary-tumour cells.     

Determination of the intracellular protein thiol distribution of hepatocytes using monobromobimane derivatisation of intact cells and isolated subcellular fractions

The derivatisation of intact rat hepatocytes with monobromobimane resulted in rapid labelling of accessible protein thiols in several subcellular fractions. The derivatisation procedure did not cause acute cytotoxicity, nor did it alter the buoyant densities of the fractions or their gross protein compositions. Quantitation of the fluorescence irreversibly associated with the fractions demonstrated considerable intracellular heterogeneity in this pool of thiols. Values were highest in cytosol (ca. 90 nmol/mg protein), intermediate in microsomes (ca. 65 nmol/mg protein) and mitochondria (ca. 45 nmol/mg protein) and lowest in a crude fraction containing both nuclei and plasma membrane (ca. 35 nmol/mg protein). Similar values were obtained from microsomes and cytosol derivatised after fractionation but there were significant increases of ca. 100% in corresponding values from isolated mitochondria and the nuclear/plasma membrane fraction. These results are discussed in terms of the dynamic fluxes in monobromobimane protein thiols during fractionation and the applicability of this noninvasive method to studies of the mechanism(s) of toxicity of reactive xenobiotics and the role(s) of protein thiols in normal cellular function.     

Identification of a novel arachidonate 12-lipoxygenase in bovine tracheal epithelial cells distinct from leukocyte and platelet forms of the enzyme

We examined the characteristics of an arachidonate 12-lipoxygenase in bovine tracheal epithelial cells in relation to the enzyme expressed in leukocytes and platelets. Homogenous preparations of intact or disrupted tracheal epithelial cells metabolized arachidonic acid predominantly to (12S)-hydroxyeicosatetraenoic acid, and subcellular fractionation by differential centrifugation demonstrated that the 12-lipoxygenase activity was localized predominantly to the 100,000 x g supernatant (cytosol fraction). Analysis of cytosolic enzymatic activity for pH dependence (maximum activity at pH 7.4-8.0), divalent cation effects (no dependence on cations), and kinetic characteristics (lag phase elimination by addition of hydroperoxide) exhibited similarity to leukocyte and platelet 12-lipoxygenases. Immunoprecipitation experiments demonstrated that the epithelial 12-lipoxygenase reacted with a monoclonal antibody (lox-2) directed against leukocyte 12-lipoxygenase but not with an antibody (HPLO-3) against the platelet enzyme. Immunoaffinity chromatography of the epithelial 100,000 x g supernatant fraction using lox-2 linked to Affi-Prep 10 yielded a single predominant protein band (Mr = 72,000) by sodium dodecyl sulfate-polyacrylamide gel electrophoresis identical in apparent mass to the bovine leukocyte lipoxygenase. Western blotting using a polyclonal antibody to leukocyte 12-lipoxygenase showed peroxidase staining of the same 72-kDa protein band. Activity assays of the purified enzymes demonstrated that substrate specificity for the epithelial 12-lipoxygenase was similar to that of the leukocyte enzyme, but the epithelial enzyme more efficiently converted 18-carbon fatty acids to the corresponding monohydroxylated conjugated dienes. We conclude that bovine tracheal epithelial cells express a 12-lipoxygenase that has immunological reactivity similar to leukocyte and distinct from platelet 12-lipoxygenase and possesses substrate specificity distinct from both enzymes. We further suggest that lipoxygenase heterogeneity may provide a basis for different functional roles for the enzyme in different cell types.