The role of eicosanoids on Rhodnius heme-binding protein (RHBP) endocytosis by Rhodnius prolixus ovaries

The participation of eicosanoids and second messengers on the regulation of RHBP endocytosis by the ovaries was investigated, using [(125)I]RHBP in experiments in vivo and in vitro. Addition of PGE(2) (one of the products of the cyclooxygenase pathway) decreased in vitro the uptake of RHBP by 35%. The rate of RHBP endocytosis increased in the presence of indomethacin, a potent cyclooxigenase inhibitor, up to 50% in vitro and up to 55% in vivo, thus giving support to the role of cyclooxygenase derivatives on endocytosis regulation. The amount of PGE(2) secreted to the culture medium by the cells of Rhodnius prolixus ovaries was 1.1 ng/ovary following RHBP uptake assay. The amount of PGE(2) decreases approximately 25% in the presence of 5 microM indomethacin. Using a scanning electron microscope we have observed that neither the surface area nor the patencies of follicle cells were affected by treatment with indomethacin, thus suggesting that, its effect is elicited in the oocyte. Finally, we have identified two ovarian peptides that were dephosphorylated after the indomethacin treatment (18 and 25 kDa). Taken together these data show that local mediators such as eicosanoids act upon the oocytes controlling RHBP endocytosis, perhaps using the protein phosphorylation signal transduction pathway.     

Characterization of leukotriene B4 synthesis in canine polymorphonuclear leukocytes.

The synthesis and release of leukotriene B4 (LTB4) from canine polymorphonuclear leukocytes (PMNs) was characterized in terms of incubation time, temperature and effects of calcium ionophore A23187 concentrations. Maximal LTB4 concentrations were determined when canine PMNs were incubated with 10 microM A23187. Increasing LTB4 concentrations were determined through 10 min incubation. The maximal LTB4 concentrations (310 +/- 30 pg LTB4/2.5 x 10(5) cells) determined at 10 min did not change through a 55 min incubation period. Greater LTB4 concentrations were synthesized by canine PMNs at 37 degrees C (268 +/- 12 pg LTB4/2.5 x 10(5) cells) than at 25 degrees C (206 +/- 11 pg LTB4/2.5 x 10(5) cells) or 5 degrees C (59 +/- 3 pg LTB4/2.5 x 10(5) cells). The synthesis of LTB4 in canine PMNs was inhibited by incubation of the cells with either of two known lipoxygenase inhibitors, BWA4C or BW755C. BWA4C inhibited LTB4 synthesis with an approximate IC50 = 0.1 microM, whereas BW755C inhibited LTB4 synthesis with an approximate IC50 = 10 microM. These results indicate canine PMNs have the capability to synthesize large quantities of LTB4 when stimulated with calcium ionophore A23187. Furthermore, the 5-lipoxygenase inhibitors BWA4C, an acetohydroxyamic acid, and BW755C, a phenyl pyrazoline, can readily inhibit LTB4 synthesis in canine PMNs.     

Products of the lipoxygenase pathway in human natural killer cell cytotoxicity

As earlier data suggested the importance of lipoxygenase activation for expression of human NK cell cytotoxicity, four different lipoxygenase inhibitors were tested for suppression of natural killer (NK) cell lysis. All inhibitors were found active at nontoxic concentrations with 50% inhibition at approximately 15 microM for nordihydroguaiaretic acid (NDGA). NK cell lysis could be reconstituted to NDGA-suppressed cells with leukotriene B4 (LTB4), the all-trans isomers 6-trans-LTB4 and 12-epi-6-trans-LTB4, and 20-COOH-LTB4. LTB4 reconstitution was best in the concentration range 1-100 pM and near control levels at both higher and lower concentrations. Herpesvirus Ateles-transformed killer T cells could also be inhibited by NDGA. These data indicate that lipoxygenase activity is required for human NK cell lysis and that several different LTB4-related products can restore NK activity in inhibited cells; they also suggest that the lipoxygenase pathway is present in the killer cell population.     

Regulation of interleukin 1 production by mouse peritoneal macrophages. Effects of arachidonic acid metabolites, cyclic nucleotides, and interferons

The effects of prostaglandin E2 (PGE2), cyclic nucleotides, leukotriene B4 (LTB4), and interferons on interleukin 1 (IL 1) production by lipopolysaccharide (LPS)-stimulated C3H/HeNCrl mouse peritoneal macrophages were studied. IL 1 production was inhibited by PGE2, the adenosine 3':5'-monophosphate analog dibutyryl cAMP, the cAMP agonist isoproterenol, and the phosphodiesterase inhibitor isobutylmethylxanthine. These agents were more inhibitory when added early in the latent phase of IL 1 synthesis following stimulation with LPS rather than just prior to release of IL 1 into the medium. Production of both the intracellular and extracellular forms of IL 1 was blocked by PGE2 and cAMP. Suppression of LPS-induced IL 1 production by PGE2 was prevented by leukocyte alpha-interferon. Moreover, alpha-interferon augmented LPS-induced IL 1 production but did not stimulate IL 1 production in the absence of LPS. Immune gamma-interferon markedly inhibited LPS-stimulated IL 1 production. The lipoxygenase inhibitor eicosa-5,8,11,14-tetraynoic acid suppressed, whereas 3-amino-1-(3-trifluoromethylphenyl)-2-pyrazoline augmented, LPS-induced IL 1 production. The opposing effects of these agents suggested that lipoxygenase metabolites do not act as inducers of IL 1 production. Purified LTB4 did not stimulate base-line or augment LPS-induced IL 1 production (both intracellular and extracellular forms). Moreover, calcium ionophore A23187 (a lipoxygenase activator) did not stimulate IL 1 production, alone or in combination with LTB4. Thus, net IL 1 production by macrophages may be regulated by a balance between the effects of PGE2, cAMP, alpha-interferon, and gamma-interferon, but not LTB4.     

Role of thymus-eicosanoids in the immune response.

The present review deals with the role(s) of thymus-eicosanoids in the immune response. It reports the production of cyclooxygenase and lipoxygenase metabolites of arachidonic acid by cells of the thymus microenvironment and the role(s) of these eicosanoids in the differentiation and the maturation of immature T-cells. The possibility that these products may be involved in tolerance to self is discussed. Briefly, it is likely that cells from the monocyte-macrophage lineage which constitute a part of the thymus microenvironment could contribute to the education of immature thymocytes by both presenting self-antigens and producing eicosanoids. Tolerance to self might result from PGE2-driven apoptosis and/or LTB4-induced generation of suppressor cells.     

Colchicine inhibits ionophore-induced formation of leukotriene B4 by human neutrophils: the role of microtubules

Neutrophils which ingest particles (serum-treated zymosan, monosodium urate crystals) or are exposed to calcium ionophore A23187 generate leukotriene B4 (LTB4). Earlier work has shown that cells exposed to colchicine before exposure to monosodium urate crystals produce less LTB4; the formation of 5-HETE is unaffected. To determine whether inhibition by colchicine of LTB4 generation was stimulus-specific and was mediated by microtubule integrity, the effects of colchicine (10 microM, 60 min) on the release of lipoxygenase products from neutrophils exposed to ionophore A23187 (10 microM, 5 min) were examined. In the presence of exogenous arachidonic acid (100 microM, 15 min), colchicine decreased LTB4 to 48% +/- 11.7 of control and 5-HETE to 60.5% +/- 5.7 of control (mean +/- SEM); 15-HETE was also decreased to 61% +/- 10.3 of control. In the absence of exogenous arachidonate, LTB4 was decreased to 22.2% +/- 11.7 of control and 5-HETE to 13% +/- 4.8 of control. Lumicolchicine did not significantly affect formation of 5-HETE or LTB4. However, vinblastine sulfate (20 microM, 60 min), another microtubule-disruptive agent, decreased the formation of both 5-lipoxygenase products. The effects of colchicine and vinblastine were not due to impairment of cell viability because the release of cytoplasmic lactic dehydrogenase was unaffected. Ultrastructural analysis of centriolar microtubules showed that decrements in microtubule numbers of colchicine- and vinblastine-treated cells paralleled decrements in 5-lipoxygenase products. These pharmacologic manipulations suggested that functional microtubules might be required for optimal lipoxygenase activity. Consequently, we prepared neutrophil-derived cytoplasts, devoid of an intact microtubule system. No significant decreases in the 5- or 15-lipoxygenase products were found when cytoplasts were exposed to colchicine in the presence of exogenous arachidonate and A23187. The data show that colchicine inhibits the formation of lipoxygenase products from neutrophils stimulated with A23187, most likely via its effect on microtubules, the integrity of which appears necessary for full expression of 5- and 15-lipoxygenases.     

Characterization and biologic properties of 5,12-dihydroxy derivatives of eicosapentaenoic acid, including leukotriene B5 and the double lipoxygenase product

Leukotriene B5 (LTB5) and three stereoisomers were prepared biosynthetically from eicosapentaenoic acid and compared with the analogous derivatives of arachidonic acid for their chemotactic and aggregating effects on human neutrophilic polymorphonuclear leukocytes. Leukotriene B4 (LTB4), LTB5, and the 6-trans-diastereoisomers of each were generated by activating polymorphonuclear leukocytes with the calcium ionophore A23187 in the presence of 14C-labeled and unlabeled arachidonic acid or 14C-labeled and unlabeled eicosapentaenoic acid, respectively. The double lipoxygenase products, (5S,12S)-6-trans-8-cis-LTB4 and (5S,12S)-6-trans-8-cis-LTB5, were generated from 5S-hydroxyeicosatetraenoic acid and racemic 5-hydroxyeicosapentaenoic acid intermediates by incubation with platelet sonicates. The products of each reaction were isolated by reverse-phase-high performance liquid chromatography and identified by their retention times relative to the appropriate totally synthetic standards, ultraviolet absorption spectra, immunoreactivity in a radioimmunoassay for LTB4, and, for all but the double lipoxygenase products, by incorporation of radiolabel from the specific polyunsaturated fatty acid source. When the concentration of LTB5 eliciting maximum chemotactic response of human polymorphonuclear leukocytes, 50 ng/ml (1.5 X 10(-7) M), and that eliciting a maximum aggregation response, 20 ng/ml (5.9 X 10(-8) M), were compared with the interpolated values of LTB4 eliciting comparable effects, the potency of LTB5 relative to LTB4 was approximately 1:8 as a chemotactic agent and about 1:20 as an aggregating agent. The double lipoxygenase products and the resolved 6-trans-diastereoisomers of the pentaene and tetraene series were about 2 logs less active as chemotactic factors than LTB4 and only (5S,12S)-6-trans-8-cis-LTB4 had even minimal aggregating activity.     

Effect of LTB4 on the inhibition of natural cytotoxic activity by PGE2.

NK activity is regulated by arachidonic acid metabolites. More precisely PGE2 and LTB4 decreases and increases respectively non-MHC-restricted cytotoxicity in humans. We have observed similar data in mice since NK activity was inhibited by PGE2 (10(-6) to 10(-8) M) and enhanced by LTB4 (10(-8) to 10(-12) M). On the other hand when PGE2 and LTB4 were combined during the same assay the lysis percentage was smaller than the one which was induced by PGE2 alone. Because PGE2 increases intracellular cyclic AMP and that LTB4 augments cyclic GMP we used a cAMP inducer (forskolin) and a cGMP analogue (8 Br-cGMP) instead of eicosanoids and we observed similar data (i.e., a decrease of natural killing) as when PGE2 was combined with LTB4. When splenocytes are cultured for 1-4 days alone, cytotoxic activity decreases unless they are cultured in the presence of indomethacin. Cytotoxic activity of spleen cells cultured in the presence of PGE2 or LTB4 is respectively decreased or increased. However, splenocytes that were cultured alone for at least 24 hr were no longer sensitive to inhibition by PGE2 but were still PGE2-sensitive when cultured in the presence of LTB4.     

Lipoxygenase-generated icosanoids inhibit glucose-induced insulin release from rat islets

Lipoxygenase-pathway metabolites of arachidonic acid are produced in pancreatic islets. They are are implicated in insulin release, since nonselective inhibitors of lipoxygenases inhibit glucose-induced insulin release. We studied the interplay in insulin release between glucose and selected icosanoids formed in 5-, 12- and 15-lipoxygenase pathways. Effects on immunoreactive insulin release of 10(7) to 10(6)-12-(R)-HETE, 12-(S)-HETE, hepoxilin A3, lipoxin B4, LTB4 or LTC4 were tested individually in 30-min incubations of freshly isolated young adult Wistar rat pancreatic islets, in the presence of 5.6 mM or 23 mM glucose. Basal insulin release (at 5.6 mM glucose) was stimulated by LTC4 and hepoxilin A3 (304% and 234% of controls at 5.6 mM glucose alone, respectively), inhibited by 12-(S)-HPETE (56%), and was not affected by 12-(R)-HETE, 12-(S)-HETE, lipoxin B4 or LTB4 (111%, 105%, 106% and 136%, respectively). Insulin release evoked by 23 mM glucose (190-320%) was inhibited (50-145%) by all icosanoids tested, except LTC4 (162%). We conclude that, among the lipoxygenase products tested, only leukotrienes and hepoxilin are candidates for a tonic-stimulatory influence on basal insulin release. Since glucose promotes icosanoid formation in islets, the observed inhibition of glucose-induced insulin release by lipoxygenase products suggests the existence of a negative-feedback system.     

Effects of cyclic AMP and dibutyryl cyclic AMP on amino acid transport in the isolated rat ovary.

Prepubertal rat ovaries were incubated in medium containing the non-utilizable amino acids alpha-aminoisobutyric acid (AIB-14C) or 1-aminocyclo-pentane-carboxylic acid (cycloleucine-14C). The rate of uptake of the two amino acids was studied in the isolated ovaries after different incubation periods. Addition of 5mM cyclic AMP (cAMP) caused a slight stimulation of the AIB-transport but in higher concentrations (10-25 mM) an inhibition was noted. With dibutyrl cyclic AMP (dbcAMP) a dose-dependent increase was seen with 0.5-5 mM concentrations with no further effect of higher concentrations. Time course studies were performed with both AIB and cycloleucine in presence of 10 mM dbcAMP and increased uptake values were noted at each time studied (30-240 min). The phosphodiesterase inhibitor aminophyline in lower concentrations did not influence AIB-transport but 5-10 mM caused increased uptake values in the ovaries. The stimulatory action of dbcAMP on amino acid transport was augmented by a low concentration of aminophylline (0.5 mM). Experiments were in addition carried out in the presence of puromycin and under these circumstances it was still possible to enhance amino acid transport by addition of dbcAMP. The results are discussed in relation to earlier reported effects of gonadotropins on ovarian amino acid transport.     

Identification of dual cyclooxygenase-eicosanoid oxidoreductase inhibitors: NSAIDs that inhibit PG-LX reductase/LTB(4) dehydrogenase.

Eicosanoids play key roles in many physiologic and disease processes, and their regulation by nonsteroidal anti-inflammatory drugs (NSAIDs) is critical to many therapeutic approaches. These autacoids are rapidly inactivated by specific enzymes such as 15-hydroxyprostaglandin dehydrogenase (15-PGDH) and 15-oxoprostaglandin 13-reductase/leukotriene B(4) 12-hydroxydehydrogenase (PGR/LTB(4)DH) that act on main series of eicosanoids (i.e., leukotrienes, prostaglandins), and recently found to act in lipoxin inactivation. Here, a panel of NSAIDs was assessed to determine each compound's ability to inhibit eicosanoid-directed activities of either the recombinant 15-PGDH or the PG-LXR/LTB(4)DH. The recombinant 15-PGDH that acts on both prostaglandin E(2) (PGE(2)) and lipoxin A(4) (LXA(4)) was not significantly inhibited by the NSAIDs tested. In contrast, several of the widely used NSAIDs were potent inhibitors of the PG-LXR/LTB(4)DH that metabolizes 15-oxo-PGE(2), and LTB(4) as well as 15-oxo-LXA(4). Diclofenac and indomethacin each inhibited PG-LXR/LTB(4)DH-catalyzed conversion of 15-oxo-PGE(2) to 13,14-dihydro-15-oxo-PGE(2) by 70 and 95%, respectively. Also, a COX-2 inhibitor, niflumic acid, inhibited the PG-LXR/LTB(4)DH eicosanoid oxidoreductase (EOR) by 80% while other COX-2 inhibitors such as nimesulide and NS-398 did not inhibit this enzyme. These results indicate that certain clinically useful NSAIDs such as diclofenac and indomethacin, in addition to inhibiting cyclooxygenases (1 and 2), also interfere with eicosanoid degradation by blocking PG-LXR/LTB(4)DH (EOR) and are members of a new class of dual cyclooxygenase (COX)-EOR inhibitors. Moreover, they suggest that the impact of NSAIDs on PG-LXR/LTB(4)DH activities as targets in the local tissue regulation of eicosanoid-mediated processes should be taken into account.     

Effects of inhibitors of eicosanoid synthesis on insulin release by neonatal pancreatic islets

In the pancreatic islet, eicosanoids may arise from both cyclooxygenase- and lipoxygenase-dependent metabolism of arachidonic acid. The inclusion of inhibitors of selective steps in these pathways indicated that in cultured neonatal rat islets, arachidonic acid may be metabolised through both pathways, concurrent with insulin release stimulated by D-glucose, D-glyceraldehyde and 2-ketoisocaproate. The effects of the inhibitors suggested that the products of the lipoxygenase pathway were necessary for the stimulatory effects of nutrients to be observed. In contrast to glucose, where insulin release was stimulated in the presence of inhibitors of cyclooxygenase, the stimulatory action of D-glyceraldehyde, 2-ketoisocaproate and melittin was only minimally affected by these inhibitors, although it was inhibited by lipoxygenase inhibition. These findings support a major stimulatory role for products of the lipoxygenase pathway of arachidonic acid metabolism in nutrient-induced secretion, and a negative or modulatory role of cyclooxygenase pathway products on glucose-stimulated insulin release in the neonatal islet.     

Uptake of Rhodnius heme-binding protein (RHBP) by the ovary of Rhodnius prolixus

The uptake of RHBP (Rhodnius heme-binding protein) by the ovaries of Rhodnius prolixus was characterized. RHBP purified from oocyte was labeled with ¹²⁵I and used to study the process of uptake by the ovary in vivo and in vitro. After injection, the [¹²⁵I]RHBP was readily removed from the hemolymph and accumulated especially in the ovary. The capacity of the ovary to take up [¹²⁵I]RHBP from the hemolymph varied during the days following blood meal. It increased up to day 2, remained stable until day 5, and then decreased up to the end of oogenesis. In vitro, the uptake of [¹²⁵I]RHBP was linear at least up to 60 min. The uptake was dependent on [¹²⁵I]RHBP concentration and showed to be a saturable process. The addition of a molar excess of non-related proteins such as Vitellin (Vt), Lipophorin (Lp), and Bovine Serum Albumin (BSA) did not reduce [¹²⁵I]RHBP uptake. Using immunogold technique the RHBP was localized at the microvilli, coated pits, and yolk granules. The main yolk protein, Vt, did not compete with RHBP for the uptake. Thus, it is discussed here that they bind to independent binding sites of the oocytes, and are directed later on to the same compartment. The need of both proteins for the completion of mature oocyte was verified in vivo. The reduction of heme-RHBP in the hemolymph, by changing the diet, decreased the number of eggs laid. Increasing the concentration of heme-RHBP in the hemolymph, the number of eggs produced increased in a dose dependent manner. In vitro, both apo-RHBP and heme-RHBP can be taken up by the oocyte. Since the mature oocyte contains only heme-saturated RHBP, the possible fate of apo-RHBP is also discussed. Arch. Insect Biochem. Physiol. 39:133–143, 1998. © 1998 Wiley-Liss, Inc.     

Role of Ca(2+)-independent phospholipase A(2) and cyclooxygenase/lipoxygenase pathways in the nitric oxide production by murine macrophages stimulated by lipopolysaccharides.

There is evidence of molecular cross talk between inflammatory mediators such as nitric oxide (NO) and prostaglandins (PG), which may regulate tissue homeostasis and contribute to pathophysiological processes. Here we examine the role of endogenous arachidonic acid (AA) and its AA metabolites in the regulation of NO release by lipopolysaccharide (LPS)-stimulated macrophages RAW 264.7. Our results suggest that bromoenol lactone-sensitive phospholipase A(2) is involved in AA release and the subsequent PG and leukotriene (LT) production. The cyclooxygenase inhibitor, indomethacin, and lipoxygenase inhibitors such as baicalein and zileuton blocked the dose-dependent PGE(2) or LTB(4) and nitrite (NO(2)(-)) production induced by LPS. Furthermore, the effects of indomethacin were reverted by exogenous PGE(2) and forskolin, whereas AH23848B, an EP(4) PGE(2) subtype receptor antagonist, decreased NO(2)(-) release. On the other hand, the effect of baicalein on NO(-)(2) production was reverted by exogenous LTB(4) and the fibrate WY 14,643, a natural and a synthetic peroxisome proliferator-activated receptor alpha (PPAR alpha), respectively. Thus, PGE(2) via EP(4) receptor/cAMP and LTB(4) via PPAR alpha may be involved in the control of NO synthesis by LPS in macrophage RAW 264.7 cultures.     

Preparation of oxygen-18-labeled lipoxygenase metabolites of arachidonic acid

Plasma pseudocholinesterase and porcine liver esterase were used to catalyse the incorporation of the stable isotope oxygen-18 into the carboxyl moiety of lipoxygenase metabolites of arachidonic acid. This simple method produces eicosanoid products containing two oxygen-18 atoms; but the enzymes studied were found to display large substrate specificity in the efficiencies at which oxygen-18 could be incorporated into the lipoxygenase metabolites. Furthermore, [18O2]LTB4 was found not to back exchange during in vitro incubation with human neutrophils. The methods involved for stable isotope incorporation are simple, efficient and produce highly enriched species in a short time. By varying the type of esterase, the amount of esterase or the length of incubation highly enriched species of all eicosanoids tested could be prepared.     

Prostaglandins, not lipoxygenase products, mediate insect microaggregation reactions to bacterial challenge in isolated hemocyte preparations

Nodulation is the predominant cellular defense reaction to bacterial challenge in insects. Eicosanoids mediate several steps in the nodulation process, including formation of hemocyte microaggregations. Isolated hemocyte preparations synthesize and secrete eicosanoids, which mediate hemocytic immune reactions. Two major groups of eicosanoids are prostaglandins (products of cyclooxygenase pathways) and various products of lipoxygenase pathways. In this study, we test the hypothesis that prostaglandins, but not lipoxygenase products, mediate hemocyte microaggregation reactions in response to bacterial challenge. Our results indicate that isolated hemocyte preparations pretreated with the cyclooxygenase inhibitors indomethacin and naproxen yielded fewer microaggregates than untreated control groups (3.7 x 10(5) microaggregates/ml hemolymph vs. 11.0 x 10(5) microaggregates/ml hemolymph). These inhibitors influence hemocyte microaggregate formation in a dose-dependent manner in treatments ranging from 0 to 200 microM. The lipoxygenase inhibitors esculetin and caffeic acid did not impact the formation of microaggregates in this system. The influence of the phospholipase A(2) inhibitor dexamethasone was reversed by amending experimental (dexamethasone-treated) preparations with prostaglandin H(2), but not prostaglandin D(2), prostaglandin E(2), nor 5(S)-hydroperoxy-6E,8Z,11Z,14Z-eicosatetraenoic acid, a product of the lipoxygenase pathway. We infer that prostaglandins are the primary mediators of microaggregation reactions to bacterial challenge in insect hemocyte preparations.     

Determination of leukotriene B4, 6-keto-prostaglandin F1 alpha and thromboxane B2 in peritoneal lavage fluid of sham-operated and adrenalectomized rats

In macrophages, isolated from the peritoneal fluid of rats, after activation, formation of metabolites of arachidonic acid occurs both by the cyclooxygenase and lipoxygenase pathways. The cells of normal animals produce mainly cyclooxygenase products. After adrenalectomy, a considerable increase occurs in the formation of lipoxygenase products, and less in those of the cyclooxygenase (1). In the experiments described here, the effect of adrenalectomy on the presence of leukotriene B4 (LTB4), 6-keto-PGF1 alpha and thromboxane B2 (TxB2) in the peritoneal fluid is determined.     

Profiles of eicosanoid production by superficial and proliferative colonic epithelial cells and sub-epithelial colonic tissue

The profile of cyclooxygenase and lipoxygenase products in normal rat colonic epithelium and subepithelium was examined. Colons were thoroughly perfused to eliminate contamination with blood. Two preparations of colonic epithelium were employed. The first consisted of intact colonic crypts and epithelial sheets. The second yielded single cell suspensions of superficial versus proliferative epithelial cells. Lipoxygenase product formation by colonic epithelium as measured by hydroxyeicosatetraenoic acid (HETE) and leukotriene B4 (LTB4) production (5-HETE greater than 12-HETE greater than 15-HETE greater than LTB4) accounted for 58% of the total colonic production of these moieties, whereas epithelium accounted for only 20% of total colonic protein. By contrast, prostaglandin (PG) E2 and PGF2 alpha production occurred predominantly (greater than 97%) in the subepithelial layers. The present studies also demonstrate markedly higher levels of accumulation of lipoxygenase products in proliferative versus superficial epithelial cells, whereas prostaglandin accumulation was greater in superficial cells. Previous studies have supported a role for lipoxygenase and cyclooxygenase products in the control of colonic secretion, inflammatory cell infiltration and proliferative activity. The present results raise the possibility that the striking differences in the sites of production of these products within the colon has functional implications.