Inhibitory effects of curcumin on in vitro lipoxygenase and cyclooxygenase activities in mouse epidermis.

Topical application of curcumin, the yellow pigment in turmeric and curry, strongly inhibited 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced ornithine decarboxylase activity, DNA synthesis, and tumor promotion in mouse skin (Huang et al., Cancer Res., 48: 5941-5946, 1988). Chlorogenic acid, caffeic acid, and ferulic acid (structurally related dietary compounds) were considerably less active. In the present study, topical application of curcumin markedly inhibited TPA- and arachidonic acid-induced epidermal inflammation (ear edema) in mice, but chlorogenic acid, caffeic acid, and ferulic acid were only weakly active or inactive. The in vitro addition of 3, 10, 30, or 100 microM curcumin to cytosol from homogenates of mouse epidermis inhibited the metabolism of arachidonic acid to 5-hydroxyeicosatetraenoic acid (5-HETE) by 40, 60, 66, or 83%, respectively, and the metabolism of arachidonic acid to 8-HETE was inhibited by 40, 51, 77, or 85%, respectively [IC50 (concentration needed for 50% inhibition) = 5-10 microM]. Chlorogenic acid, caffeic acid, or ferulic acid (100 microM) inhibited the metabolism of arachidonic acid to 5-HETE by 36, 10, or 16%, respectively, and these hydroxylated cinnamic acid derivatives inhibited the metabolism of arachidonic acid to 8-HETE by 37, 20, or 10%, respectively (IC50 greater than 100 microM). The metabolism of arachidonic acid to prostaglandin E2, prostaglandin F2 alpha, and prostaglandin D2 by epidermal microsomes was inhibited approximately 50% by the in vitro addition of 5-10 microM curcumin. Chlorogenic acid, caffeic acid, and ferulic acid (100 microM) were inactive. In vitro rat brain protein kinase C activity was not affected by 50-200 microM curcumin, chlorogenic acid, caffeic acid, or ferulic acid. The inhibitory effects of curcumin, chlorogenic acid, caffeic acid, and ferulic acid on TPA-induced tumor promotion in mouse epidermis parallel their inhibitory effects on TPA-induced epidermal inflammation and epidermal lipoxygenase and cyclooxygenase activities.

Hepatocyte modulation of Kupffer cell prostaglandin E2 production in vitro.

It is likely that dynamic interactions between hepatocytes and Kupffer cells contribute to the responses of these cell types both under normal conditions and during sepsis. In this study, we examined the influences of hepatocytes on the concentration of the inflammatory mediator PGE2 in Kupffer cell cultures. Evidence to suggest that cultured rat hepatocytes both metabolize PGE2 and produce a substance that promotes LPS-stimulated Kupffer cell PGE2 biosynthesis include the following: 1) PGE2 levels in Kupffer cell: hepatocyte coculture were lower than the levels in Kupffer cell cultures early after LPS stimulation; 2) 36 h after LPS, coculture PGE2 levels exceeded the levels in Kupffer cell cultures despite the demonstrated capacity for hepatocytes to metabolize PGE2; 3) a transferable, non-dialyzable, and heat-unstable factor in hepatocyte supernatant promoted PGE2 production when added to Kupffer cells with LPS or after LPS; 4) there was no increased PGE2 synthesis when the hepatocyte supernatant was added without LPS or if hepatocyte supernatant was preincubated with the Kupffer cells for 6 or 18 h before LPS administration; 5) there was an inability of the hepatocyte factor to promote PGE2 production in response to other macrophage-activating agents, including calcium ionophore A23187 or phorphol myristate acetate; and 6) there was no increased cell replication or protein synthesis in the Kupffer cell cultures following hepatocyte supernatant incubation. The increased Kupffer cell PGE2 production by the hepatocyte supernatant was not due to contamination of the hepatocyte supernatant by endotoxin or PGE2. These in vitro results raise the possibility that hepatocytes have the capacity to modulate local PGE2 levels by two distinct mechanisms. Hepatocytes can metabolize PGE2 as well as release factor(s) which promote LPS-induced PGE2 production by Kupffer cells.

A role for 12(S)-HETE in the response of human lens epithelial cells to epidermal growth factor and insulin.

PURPOSE: To determine whether the 12-lipoxygenase pathway of arachidonic acid metabolism is present in the human lens and whether 12(S)-hydroxyeicosatetraenoic acid (12(S)-HETE) plays a role in regulating proto-oncogene expression and DNA synthesis in human lens epithelial cells (HLECs). METHODS: Second- and third-passage primary cultures of HLECs were used for analysis. Human cataract epithelia were obtained from surgery. 12-lipoxygenase mRNA was detected by reverse transcription-polymerase chain reaction (RT-PCR), and the PCR product was sequenced. The 12-lipoxygenase protein was detected by immunoblotting. 12(S)-HETE was detected in HLEC-conditioned medium by radioimmunoassay. For studies of growth factor-induced mitogenesis, HLECs were serum starved, then stimulated with 15 ng/ml epidermal growth factor (EGF) and 1 microgram/ml insulin or with 0.3 microM 12(S)-HETE. The 12-lipoxygenase inhibitor, cinnamyl-3,4-dihydroxy-alpha-cyanocinnamate (CDC, 10 microM) was used to block endogenous 12-lipoxygenase activity. Expression of c-fos mRNA was determined by RT-PCR, and DNA synthesis was measured by 3H-thymidine incorporation. RESULTS: 12-lipoxygenase mRNA and protein were detected in HLECs and in human cataract tissues. 12(S)-HETE was released into the medium by HLECs in the presence of EGF-insulin. Stimulation of c-fos mRNA expression and DNA synthesis by EGF-insulin was inhibited when the 12-lipoxygenase pathway was blocked by CDC. This inhibition was reversed completely by exogenously added 12(S)-HETE. However, exogenous 12(S)-HETE was unable to stimulate HLEC DNA synthesis in the absence of growth factors. CONCLUSIONS: The 12-lipoxygenase pathway of arachidonic acid metabolism is present in human lens epithelial cells. 12(S)-HETE does not stimulate HLEC DNA synthesis in the absence of growth factors but enables the cellular response to EGF and insulin.

Prostaglandin biosynthesis in the rat lens.

Rat lens microsomal preparations possess the capability of converting exogenous arachidonic acid into prostaglandins (PG) E2 and F2 alpha. The low, yet measurable prostaglandin biosynthetic capacity of the lens microsomes was demonstrated by radioimmunoassay (RIA) and by separation of radiolabeled products after incubation with high specific activity U-14 [C]-arachidonic acid. Maximal formation of radioimmunoassayable PGE2 and PGF2 alpha was measured after a 15-min incubation at 37 degrees C with approximately 0.5 mg protein. Prostaglandin biosynthesis was inhibited by the nonsteroidal anti-inflammatory agents aspirin (IC50 = 52 microM) and indomethacin (IC50 = 20 microM). These results unequivocally demonstrate prostaglandin biosynthesis in rat lens.

Immunocytochemical localization of cyclooxygenase in the rat lens.

The localization of rat lens fatty acid cyclooxygenase (prostaglandin synthase) was studied using indirect immunofluorescent and indirect streptavidin-biotin immunoperoxidase staining techniques. Both methods employed monoclonal and polyclonal antibodies directed against fatty acid cyclooxygenase, the key enzyme in the conversion of arachidonic acid to prostaglandins. The immunocytochemical studies demonstrate that (1) fatty acid cyclooxygenase is present in the rat lens epithelial cell layer; (2) the enzyme appears predominantly in the cytoplasm; (3) there is an apparent higher concentration of the enzyme in the region designated as germinative and transitional zones, and meridional rows; and (4) the enzyme appears to be absent in the lens capsule and in the nucleus of the lens. The presence of the cyclooxygenase enzyme in the lens epithelium, especially the relative intense staining in the epithelial mitogenic region, suggests that oxygenation of polyunsaturated fatty acids by the fatty acid cyclooxygenase may have an important role in cellular differentiation.

Examination of mouse and rat tissues for evidence of dual forms of the fatty acid cyclooxygenase.

The possibility that the enzymatic generation of prostaglandin E2 (PGE2) and PGF2 alpha results from the catalytic activity of two distinct forms of the fatty acid cyclooxygenase was studied in microsomes prepared from kidney, lung, and brain of the mouse and rat. Three criteria established previously to detect the dual cyclooxygenase forms in the rabbit brain were used in the present study: (1) different time course profiles of microsomal PGE2 and PGF2 alpha biosynthesis from exogenous arachidonic acid; (2) elimination of the synthesis of one PG in vitro by non-steroidal anti-inflammatory drug concentrations that did not affect the synthesis of the other PG and; (3) selective autocatalytic inactivation of one cyclooxygenase by preincubation with arachidonic acid. Incubations with PGH2 endoperoxide as substrate tested whether the altered PG biosynthesis resulted from an effect on the endoperoxide utilizing enzymes and not on the cyclooxygenase. Of the six tissues examined, only the mouse brain microsomes satisfied all the criteria. The microsomes prepared from the mouse kidney produced mixed results. We conclude that the mouse brain but not the rat brain gives evidence for two distinct forms of the fatty acid cyclooxygenase. Additional distinguishing features of the different cyclooxygenases are required to determine if the cyclooxygenase forms are found in mouse kidney.

The rabbit lens epithelial cell line N/N1003A requires 12-lipoxygenase activity for DNA synthesis in response to EGF.

PURPOSE: Cultured rat lenses and primary human lens epithelial cells (HLECs) express12-lipoxygenase (12-LOX) and require a 12-LOX metabolite of arachidonic acid for growth in response to EGF and insulin. This study seeks to identify an established cell line with these characteristics. METHODS: Immunoblotting was used to screen eight lens epithelial cell lines for 12-LOX expression: the human line, HLE-B3; mouse lines alphaTN4, 17EM15, 21EM15, and MLE6, and rabbit lines N/N1003A, LEP2 and B3. DNA synthesis was measured as incorporation of 3H-thymidine into DNA. Expression of c-fos mRNA was detected by RT-PCR. The involvement of 12-lipoxygenase metabolites was determined using the lipoxygenase inhibitors baicalein, cinnamyl 3,4-dihydroxy-alpha-cyanocinnamate (CDC), or nordihydroguiairetic acid (NDGA). RESULTS: 12-LOX was detected only in the rabbit lines N/N1003A, LEP2 and B3. N/N1003A cells were chosen for further study. 12-LOX inhibitors blocked DNA synthesis in response to EGF with or without insulin. Inhibition of EGF-stimulated DNA synthesis was reversed by 0.3 microM to 3 microM 12(S)hydroxyeicosatetraenoic acid (HETE), but not by equivalent concentrations of 5(S)HETE, 8(S)HETE, 15(S)HETE, or 12(R)HETE. Baicalein prevented EGF induction of c-fos mRNA. The transformed HLEC line, HLE-B3, showed little stimulation of DNA synthesis in response to EGF and was unaffected by the presence of 12-LOX inhibitors. CONCLUSIONS: N/N1003A cells, like primary cultured human lens epithelial cells or neonatal rat lenses, require 12-LOX activity for EGF dependent growth. This line will be useful for studies of the mechanism of action of 12(S)HETE.

Correlation between red blood cell deformability and changes in hemodynamic function.

BACKGROUND: In sepsis red blood cells (RBCs) have been shown to be less deformable (i.e., more rigid) and have been implicated in decreasing nutrient blood supply and possibly leading to organ dysfunction. However, no studies have demonstrated an association between organ dysfunction and rigid RBCs. This study examined cardiovascular physiologic and histologic changes in two different models to determine whether a relationship may exist between RBC deformability and organ function. METHODS: In the following two experiments, cardiac index (CI) was continuously measured, whereas both deformability index and histology were examined at the end of the experimental periods. The first experiment studied nonanesthetized, hydrated rats after a cecal ligation and puncture (CLP), a slow-developing means of inducing RBC rigidity. In a second experiment animals were anesthetized and received a 20% total blood volume transfusion of either diamide-treated (rigid) RBCs or normal RBCs. RESULTS: CLP-treated animals' CI gradually decreased during 18 hours (232 +/- 60 ml/min/kg to 123 +/- 90 ml/min/kg; p = 0.05), with an increase in systemic vascular resistance (1459 +/- 517 dyne.sec/cm5.m2 to 2337 +/- 1213 dyne.sec/cm5.m2; p = 0.02). Diamide-treated animals had a rapid decrease in CI (86 +/- 7.0 ml/min/kg to 58 +/- 13 ml/min/kg; p = 0.05) and increase in SVR (2269 +/- 373 dyne.sec/cm5.m2 to 3897 +/- 988 dyne.sec/cm5.m2; p = 0.05) from baseline to 120 minutes after treatment respectively. The DI was significantly lower in both CLP and diamide groups (p < 0.03) when compared with control animals. Histologic evidence of subendocardial necrosis was shown in both the CLP- and Diamide-treated animals. CONCLUSIONS: These data suggest an association with RBC deformability and organ function in both septic and nonseptic animal models.

Temporal relationship of hepatocellular dysfunction and ischemia in sepsis.

To determine whether hepatic dysfunction in sepsis results from hypoperfusion or direct cellular injury, Sprague-Dawley rats underwent either cecal ligation and puncture or sham operation. After either two or six hours, effective hepatic blood flow was measured using the galactose clearance method. Hepatocytes were isolated and intracellular sodium and potassium and glucose production were measured. Hepatic blood flow in septic rats decreased as early as two hours after sepsis when compared with sham-operated rats (3.8 +/- 1.4 vs 8.7 +/- 3.1 mL/min/100 g body weight). Intracellular sodium and potassium levels and glucose production in septic rats were not significantly different when compared with controls at two hours. After six hours, hepatic blood flow remained depressed and intracellular sodium level was increased compared with sham-operated rats (41.7 +/- 10.4 vs 31.4 +/- 5.9 mmol/L [41.7 +/- 10.4 vs 31.4 +/- 5.9 mEq/L]) and potassium decreased compared with controls (90.7 +/- 7.9 vs 111.5 +/- 6.7 mmol/L [90.7 +/- 7.9 vs 111.5 +/- 6.7 mEq/L]). Glucose production was decreased in septic rats after six hours when compared with controls (4.7 +/- 1.5 vs 15.4 +/- 6.4 mumol/g hepatocytes). These data suggest that hepatic blood flow is decreased before alterations in intracellular sodium and potassium as well as glucose production.

Influence of inhibitors of eicosanoid metabolism on proliferation of rat hepatoma cells and on tumor-host interaction.

The influence of eicosanoids on the proliferation of hepatoma (HTC) cells was studied in culture and in tumor-bearing rats. The cells in culture demonstrated a capacity to metabolize arachidonic acid to eicosanoids including thomboxane B2 and the prostaglandins E2 and F2 alpha a. An effect of these eicosanoids on cell proliferation was suggested by the decreased cell division seen with an inhibitor of cyclooxygenase, flurbiprofen. A biphasic effect on the proliferation of HTC cells was observed with increasing concentrations of prostaglandin F2 alpha. These studies were extended to tumor-bearing rats where inhibitory effects on the early stages of tumor growth were seen with flurbiprofen. Bleeding times were decreased in tumor-bearing rats but were restored to control values by treatment with flurbiprofen and an inhibitor of thromboxane synthetase, OKY 046. These drugs and a thromboxane/endoperoxide receptor antagonist, SQ 29, 548, were not observed to have statistically significant effects on isotope-labeled water distribution but they had substantial effects on the maintenance of body weight by tumor-bearing rats. The data suggested that the cachexia of tumor-bearing animals may be mediated at least in part by the action of eicosanoids.

Evidence for increased activity of mouse brain fatty acid cyclooxygenase following drug-induced convulsions.

Enzymatic production of prostaglandins (PGs) from exogenous arachidonic acid was studied in brain microsomal fractions prepared from mice following pentylenetetrazol (PTZ)-induced convulsions. Prostaglandin E2 (PGE2) and prostaglandin F2 alpha (PGF2 alpha) measured either by radioimmunoassay or after incubation with [1-14C]arachidonic acid (AA) was significantly increased in microsomes from the convulsed animals. Pretreatment of the mice with the anticonvulsant ethosuximide prevented the enhanced PG production. The increased PG synthesis could not be attributed to an increased substrate availability nor to an activated phospholipase nor to a direct effect of the convulsant on the fatty acid cyclooxygenase. Evidence that a modification of the cyclooxygenase had occurred with seizure activity was obtained from kinetic analysis; the apparent Km for the AA was lowered from 30 +/- 3 microM in the controls to 12 +/- 1 microM in the PTZ-treated mice. Further evidence for a modification of the fatty acid cyclooxygenase was obtained from incubations of the microsomes with catalase to reduce peroxide formation. Limiting peroxide levels did not decrease the microsomal cyclooxygenase activity in the PTZ-treated mice to control levels. Seizure activity induced by picrotoxin and strychnine also increased the microsomal capacity of the convulsed animals to synthesize PGs. The increased brain fatty acid cyclooxygenase activity may result from a biochemical modification of the enzyme induced by seizure activity.

Inhibitory effect of curcumin and some related dietary compounds on tumor promotion and arachidonic acid metabolism in mouse skin.

Topical application of curcumin, the major yellow pigment in turmeric and curry, has a potent inhibitory effect on 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced tumor promotion in mouse skin. The structurally related compounds chlorogenic acid, caffeic acid and ferulic acid are less potent inhibitors. Curcumin is a potent inhibitor of TPA-induced ornithine decarboxylase activity and inflammation in mouse skin whereas chlorogenic acid, caffeic acid and ferulic acid are only weakly active or inactive. Curcumin is a potent inhibitor of arachidonic acid-induced inflammation in vivo in mouse skin, and this compound is also a potent inhibitor of epidermal lipoxygenase and cyclooxygenase activity in vitro. Although chlorogenic acid is only weakly active as an inhibitor of epidermal lipoxygenase activity and TPA-induced ear inflammation, it is more active than caffeic acid and ferulic acid. The inhibitory effects of curcumin, chlorogenic acid, caffeic acid and ferulic acid on TPA-induced tumor promotion in mouse skin parallel their inhibitory effects on TPA-induced epidermal inflammation and epidermal lipoxygenase and cyclooxygenase activities. Examination of the structural features of curcumin required for its biological activity indicate that free hydroxyl groups on the benzene rings are not required for inhibition of TPA-induced ornithine decarboxylase activity and inflammation in mouse skin.

The gut as source of sepsis after hemorrhagic shock.

In a model of severe hemorrhagic shock in rats, blood culture findings became positive within 2 to 4 hours of shock. The organisms cultured were primarily gram-negative. To test the hypothesis that the gut was the source of the bacteria, E. coli labeled with carbon-14 oleic acid were fed to rats undergoing hemorrhagic shock. Their plasma was then assayed for carbon-14 activity. Seven of the 14 shocked animals demonstrated increased plasma carbon-14 activity during or after shock. The mortality rate was 100 percent 80 hours postshock, and all animals had E. coli on subsequent blood culture. The seven rats without increased plasma carbon-14 activity had a survival rate of 83 percent postshock. Sham-shocked animals did not exhibit plasma carbon-14 levels greater than the background levels. These data suggest that bacterial translocation occurs during hemorrhagic shock and that the gut is the source of the bacteremia seen during hemorrhagic shock.

Organ distribution of gut-derived bacteria caused by bowel manipulation or ischemia.

Translocation of carbon-14-labeled Escherichia coli from the gut was studied at the specified times in the following groups of rats: Group 1, 5 hours after ligation of the superior mesenteric artery; Group 2, 5 hours after laparotomy and exposure of the superior mesenteric artery with gentle removal and replacement of the intestines; and Group 3, 5 hours after handling but no surgical manipulation. Both living and dead bacteria were administered by means of gavage, and the effect of viability, intestinal ischemia without reperfusion, and bowel manipulation on the translocation of enteric bacteria was assessed. We demonstrated that (1) even gentle bowel manipulation causes bacteremia as great as that associated with ligation of the superior mesenteric artery; (2) dead E. coli are absorbed into the blood in the presence of bowel manipulation or ischemia but less effectively than are live E. coli; (3) live bacteria are found in highest concentration in the lung and in descending order in the liver, kidney, heart, and spleen; (4) dead bacteria absorbed from the gut are found in highest concentration in the kidney and the liver. Lesser amounts are found in the lung, spleen, and heart.

Role of eicosanoids and white blood cells in the beneficial effects of limited reperfusion after ischemia-reperfusion injury in skeletal muscle.

Limiting the rate of reperfusion blood flow has been shown to be beneficial locally in models of ischemia-reperfusion injury. We investigated the effects of this on eicosanoids (thromboxane B2, 6-keto-PGF1 alpha, and leukotriene B4), white blood cell activation, and skeletal muscle injury as quantitated by triphenyltetrazolium chloride and technetium-99m pyrophosphate after ischemia-reperfusion injury in an isolated gracilis muscle model in 16 anesthetized dogs. One gracilis muscle in each dog was subjected to 6 hours of ischemia followed by 1 hour of limited reperfusion and then by a second hour of normal reperfusion. The other muscle was subjected to 6 hours of ischemia followed by 2 hours of normal reperfusion. Six dogs each were used as normal reperfusion controls (NR) and limited reperfusion controls (LR), with 5 dogs being treated with a thromboxane synthetase inhibitor (LR/TSI) and another five with a leukotriene inhibitor (LR/LI). LR in all three groups (LR, LR/TSI, and LR/LI) showed a benefit in skeletal muscle injury as measured by triphenyltetrazolim chloride and technetium-99m pyrophosphate when compared with NR. However, there was no significant difference between the groups with LR regarding eicosanoid levels and white blood cell activation when compared with NR. These results demonstrate that LR produces benefits by mechanisms other than those dependent upon thromboxane A2, prostacyclin, or white blood cell activation.

Thromboxane synthetase inhibition decreases polymorphonuclear leukocyte activation following hindlimb ischemia.

Ischemia of the lower extremity has been shown to cause pulmonary leukostasis and increased pulmonary artery pressure. Thromboxane (TX) has been implicated as a mediator in this process. The effect of OKY-046, a TX synthetase inhibitor, on polymorphonuclear leukocyte (PMN) production of superoxide anion (O2-) as determined by ferricytochrome reduction was examined. Fourteen dogs were subjected to 6 hours of bilateral gracilis muscle ischemia followed by 1 hour of reperfusion. O2- production from resting PMNs and PMNs stimulated with opsonized zymosan (OZ, 0.1 mg/ml) was measured prior to ischemia or drug treatment (baseline), and following reperfusion in both treated (n = 7) and control groups (n = 7). Serum TX levels were measured using a radioimmunoassay. Following reperfusion, TX levels in the treated group were decreased as compared with the control group (18 +/- 2 pg/ml vs. 72 +/- 26 pg/ml, P less than 0.05). Superoxide production by both resting and stimulated PMNs was also decreased in the treated group; from 0.98 +/- 0.16 nmol to 0.43 +/- 0.12 nmol O2- in the resting state (P less than 0.05) and from 13.3 +/- 1.5 nmol to 9.0 +/- 1.1 nmol O2- after stimulation (P less than 0.005). O2- production was increased in the control group following reperfusion as compared with baseline samples, and this increase was attenuated by treatment with OKY-046. TX synthetase inhibition decreases activation of PMNs following hindlimb ischemia.

Identification of 12(S)-hydroxyeicosatetraenoic acid in the young rat lens.

Evidence is presented indicating that young rat lens has the capacity to synthesize 12(S)-hydroxyeicosatetraenoic acid (12(S)-HETE) from exogenous arachidonic acid (AA). A 9,000 xg supernatant prepared from 15 day old rat lenses, when incubated with calcium and U-[14C]-AA, generated a radiolabelled product with a retention time identical to authentic unlabelled 12-HETE in two different HPLC solvent systems. Mass spectral analysis provided evidence that the metabolite was 12-HETE while chiral studies demonstrated the exclusive presence of the 12(S) isomer. Radiolabelled 12-HETE synthesis was inhibited by preincubating the 15 day old rat lens supernatant with 0.2 Mm curcumin, a mixed cyclooxygenase/lipoxygenase inhibitor. Radiolabelled 12(S)-HETE synthetic capacity was highest in the 4 day old rat lens, the earliest time period measured, and rapidly declined with age reaching negligible levels at about 4 months. These results suggest that the young rat lens possess the biosynthetic capacity to generate radiolabelled 12(S)-HETE from U-[14]C-AA. The profile of 12-HETE synthetic activity suggests a possible regulatory function of the hydroxylated AA metabolite in the developing lens.

Vasoactive prostaglandins in the impending no-reflow state: evidence for a primary disturbance in microvascular tone.

The impending no-reflow (NRF) state was studied in the rat hindlimb to identify possible biochemical mediators producing the no-reflow phenomenon. After 5 hours of ischemia, the venous effluents draining the ischemic limb and the contralateral nonischemic limb were collected for three 30-minute time periods. Thromboxane B2 (TxB2), prostaglandin E2 (PGE2), and 6-ketoprostaglandin F1 alpha, the stable metabolite of prostacyclin (PGI2), were measured by radioimmunoassay. Venous outflow rate, distal skin perfusion assessed by dermofluorometry, and histology of muscle and skin were examined in control limbs, ischemic limbs, and limbs with impending no reflow. The no-reflow state was characterized by a significantly decreased venous outflow (less than 0.01 ml per minute), decreased skin perfusion (index of fluorescence of 15 percent in no-reflow limbs versus 70 percent in reflow limbs), and absence of thrombosis of the vasculature. The no-reflow state also was associated with 2.4 times more thromboxane B2 and 1.5 times more 6-ketoprostaglandin F1 alpha than that observed in ischemic limbs with reflow. The biosynthesis of vasodilating prostaglandin E2 in the no-reflow state, however, was only 40 percent of the prostaglandin E2 measured in limbs with reflow. We propose that the impending no-reflow state may reflect a state of global microcirculatory "agonal" vasoconstriction, most probably due to an overabundant release of the vasoconstrictor thromboxane relative to the vasodilating prostaglandin E2 and prostacyclin. The likelihood of specific biochemical mechanisms producing the no-reflow state suggests that pharmacologic agents may be able to reverse the impending no-reflow state to improve tissue survival.