Malignant transformation and antineoplastic actions of nonsteroidal antiinflammatory drugs (NSAIDs) on cyclooxygenase-null embryo fibroblasts.

In this study, we use primary embryonic fibroblasts derived from cyclooxygenase-deficient transgenic embryos to further investigate the role of the two cyclooxygenases, cyclooxygenase 1 (COX-1) and cyclooxygenase 2 (COX-2), in the process of neoplastic transformation. Cells with either, neither, or both of the cyclooxygenases were transformed by Ha-ras and/or SV40. Our results show that when a cyclooxygenase enzyme is present, the transformed cells have marked increases in COX-2 and/or COX-1 expression. Nevertheless, each type of cell, deficient in either or both cyclooxygenases, can be readily transformed at almost equal efficiency. Different nonsteroidal antiinflammatory drugs (NSAIDs) were used to examine their possible antineoplastic effects on the transformed cells, which have various levels of expression of COX-1 or COX-2. Our results show that NSAIDs suppress the colony formation in soft agar in a dosage-dependent manner in the absence of the cyclooxygenase(s). Thymidine incorporation and apoptosis analyses further demonstrate that the NSAIDs are effective in the cyclooxygenase-null cells. Our findings with cyclooxygenase knockout cells confirm recent reports that some of the antiproliferative and antineoplastic effects of NSAIDs are independent of the inhibition of either COX-1 or COX-2. They also show that transformation is independent of the status of cyclooxygenase expression, suggesting that the involvement of the cyclooxygenases in tumorigenesis may occur at later steps.

Scanning electron microscopy of in vitro chemically transformed mouse embryo cells.

A cloned nontumorigenic control cell line of C3H mouse embryo cells (C3H/1OT1/2CL8) and two cell lines derived from it by treatment in vitro with 7,12-dimethylbenz(a)anthracene (DMBA) or 3-methylcholanthrene (MCA) were studied by scanning electron microscopy. Confluent control cells were polygonal in shape and extensively flattened with smooth surfaces. Both in vitro transformants were pleomorphic to fusiform in shape, thicker than the control cells, and lacked contact inhibition. Microvilli of variable length and small marginal ruffles were characteristic surface alterations of the MCA-transformed cells, while blebs and numerous cytoplasmic strands extending between cells were typical of the DMBA transformant. Inoculation of the DMBA-transformed cells into C3H mice and re-establishment of cells from one of the subsequent fibrosarcomas in culture revealed an increased number of microvilli on the surface of the cells and an alteration in growth pattern. Other surface characteristics remained the same. A possible relationship between surface topography and outer membrane glycolipids is discussed.

Impaired mucosal defense to acute colonic injury in mice lacking cyclooxygenase-1 or cyclooxygenase-2.

To investigate roles in intestinal inflammation for the 2 cyclooxygenase (COX) isoforms, we determined susceptibility to spontaneous and induced acute colitis in mice lacking either the COX-1 or COX-2 isoform. We treated wild-type, COX-1(-/-), COX-2(-/-), and heterozygous mice with dextran sodium sulfate (DSS) to provoke acute colonic inflammation, and we quantified tissue damage, prostaglandin (PG) E(2), and interleukin-1beta. No spontaneous gastrointestinal inflammation was detected in mice homozygous for either mutation, despite almost undetectable basal intestinal PGE(2) production in COX-1(-/-) mice. Both COX-1(-/-) and COX-2(-/-) mice showed increased susceptibility to a low-dose of DSS that caused mild colonic epithelial injury in wild-type mice. COX-2(-/-) mice were more susceptible than COX-1(-/-) mice, and selective pharmacologic blockade of COX-2 potentiated injury in COX-1(-/-) mice. At a high dose, DSS treatment was fatal to 50% of the animals in each mutant group, but all wild-type mice survived. DSS treatment increased PGE(2) intestinal secretion in all groups except COX-2(-/-) mice. These results demonstrate that COX-1 and COX-2 share a crucial role in the defense of the intestinal mucosa (with inducible COX-2 being perhaps more active during inflammation) and that neither isoform is essential in maintaining mucosal homeostasis in the absence of injurious stimuli.

Distinct roles of prostaglandin H synthases 1 and 2 in T-cell development.

Prostaglandin G and H synthases, or cyclooxygenases (COXs), catalyze the formation of prostaglandins (PGs). Whereas COX-1 is diffusely expressed in lymphoid cells in embryonic day 15.5 thymus, COX-2 expression is sparse, apparently limited to stromal cells. By contrast, COX-2 is predominant in a subset of medullary stromal cells in three- to five-week-old mice. The isozymes also differ in their contributions to lymphocyte development. Thus, experiments with selective COX-1 inhibitors in thymic lobes from normal and recombinase-activating gene-1 knockout mice support a role for this isoform in the transition from CD4(-)CD8(-) double-negative (DN) to CD4(+)CD8(+) double-positive (DP). Concordant data were obtained in COX-1 knockouts. Pharmacological inhibition and genetic deletion of COX-2, by contrast, support its role during early thymocyte proliferation and differentiation and, later, during maturation of the CD4 helper T-cell lineage. PGE2, but not other PGs, can rescue the effects of inhibition of either isoform, although it acts through distinct EP receptor subtypes. COX-dependent PG generation may represent a mechanism of thymic stromal support for T-cell development.

Allergic lung responses are increased in prostaglandin H synthase-deficient mice.

To investigate the function of prostaglandin H synthase-1 and synthase-2 (PGHS-1 and PGHS-2) in the normal lung and in allergic lung responses, we examined allergen-induced pulmonary inflammation and airway hyperresponsiveness in wild-type mice and in PGHS-1(-/-) and PGHS-2(-/-) mice. Among nonimmunized saline-exposed groups, we found no significant differences in lung function or histopathology, although PGE(2) was dramatically reduced in bronchoalveolar lavage (BAL) fluid from PGHS-1(-/-) mice, relative to wild-type or PGHS-2(-/-) mice. After ovalbumin sensitization and challenge, lung inflammatory indices (BAL cells, proteins, IgE, lung histopathology) were significantly greater in PGHS-1(-/-) mice compared with PGHS-2(-/-) mice, and both were far greater than in wild-type mice, as illustrated by the ratio of eosinophils in BAL fluid (8:5:1, respectively). Both allergic PGHS-1(-/-) and PGHS-2(-/-) mice exhibited decreased baseline respiratory system compliance, whereas only allergic PGHS-1(-/-) mice showed increased baseline resistance and responsiveness to methacholine. Ovalbumin exposure caused a modest increase in lung PGHS-2 protein and a corresponding increase in BAL fluid PGE(2) in wild-type mice. We conclude that (a) PGHS-1 is the predominant enzyme that biosynthesizes PGE(2) in the normal mouse lung; (b) PGHS-1 and PGHS-2 products limit allergic lung inflammation and IgE secretion and promote normal lung function; and (c) airway inflammation can be dissociated from the development of airway hyperresponsiveness in PGHS-2(-/-) mice.

Deletion and replacement of the mouse adult beta-globin genes by a "plug and socket" repeated targeting strategy.

We describe a two-step strategy to alter any mouse locus repeatedly and efficiently by direct positive selection. Using conventional targeting for the first step, a functional neo gene and a nonfunctional HPRT minigene (the "socket") are introduced into the genome of HPRT- embryonic stem (ES) cells close to the chosen locus, in this case the beta-globin locus. For the second step, a targeting construct (the "plug") that recombines homologously with the integrated socket and supplies the remaining portion of the HPRT minigene is used; this homologous recombination generates a functional HPRT gene and makes the ES cells hypoxanthine-aminopterin-thymidine resistant. At the same time, the plug provides DNA sequences that recombine homologously with sequences in the target locus and modifies them in the desired manner; the plug is designed so that correctly targeted cells also lose the neo gene and become G418 sensitive. We have used two different plugs to make alterations in the mouse beta-globin locus starting with the same socket-containing ES cell line. One plug deleted 20 kb of DNA containing the two adult beta-globin genes. The other replaced the same region with the human beta-globin gene containing the mutation responsible for sickle cell anemia.

Nonspecific inhibition of DNA repair by promoting and nonpromoting phorbol esters.

The effects of the tumor promoter 12-O-tetradecanoylphorbol-13-acetate (TPA) and its nonpromoting structural analogue, 4-O-methyl-12-O-tetradecanoylphorbol-13-acetate (Me-TPA), on N-acetoxy-2-acetylaminofluorene-elicited DNA repair and replicative DNA synthesis was measured in normal human fibroblasts. Both esters inhibited DNA repair synthesis, and Me-TPA was nearly as effective as TPA. In addition, TPA inhibited replicative DNA synthesis. These findings showed that inhibition of DNA repair synthesis may not be a major factor in the mechanisms of action of tumor promoters.

Cytotoxic and oncogenic activities of 1,1,1-trichloro-2,2-bis(p-chlorophenyl)-ethane and metabolites to mouse embryo cells in culture.

The cytotoxic and oncogenic activities of 1,1,1-trichloro-2,2-bis(p-chlorophenyl)ethane (DDT), and its metabolites 1,1-dichloro-2,2-bis(p-chlorophenyl)ethane (DDD), 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene (DDE), and bis(p-chlorophenyl)acetic acid, were studies at various equimolar concentrations in an in vitro mouse embryo cell culture system that was malignantly transformable. DDD was the most active compound in producing transformation, whereas DDT and DDE showed slight activity. However, none of the transformed foci had the typical malignantly transformed morphology produced by 7,12-dimethylben[a]anthracene and did not produce tumors when inoculated into syngeneic mice.

Rat bladder cell-mediated mutagenesis od Chinese hamster V79 cells and metabolism of benzo[a]pyrene.

Primary rat bladder epithelial cells were cocultivated with Chinese hamster V79 cells in the presence of carcinogens, and the induction of 6-thioguanine resistance in the V79 cells was used as a marker of cell-mediated mutagenesis. The carcinogens dimethylnitrosamine, 7,12-dimethylbenz[a]anthracene, and benzo[a]pyrene (BP) were mutagenic to V79 cells in the presence of bladder cells but not in their absence. Analysis of BP metabolites formed by bladder cells indicated that 7,8-dihydro-7,8-dihydroxybenzo[a]pyrene, 9,10-dihydro-9,10-dihydroxybenzo[a]pyrene, benzo[a]pyrene-3,6-quinone, and 9-hydroxybenzo[a]pyrene were the major organic-soluble metabolites formed. Glucuronide and sulfate conjugates of BP metabolites were also produced by bladder cells. Mutagenesis data from the rat bladder system and previous data from rat liver and lung cell-mediated mutagenesis systems indicate that the cell-mediated mutagenesis approach may provide a useful approach for studying the organotropic effect of chemical carcinogens. Furthermore, the finding that rat bladder epithelium can metabolize some carcinogens offers new possibilities for the mechanism of initiation of bladder cancer.

Stimulation of differentiated functions in human melanoma cells by tumor-promoting agents and dimethyl sulfoxide.

Treatment of cultured human HO melanoma cells with the mouse skin tumor promoter phorbol-12-myristate-13-acetate (PMA) at 5 x 10(-10) to 5 x 10(-7) M resulted in a dose-related inhibition of growth and a stimulation of differentiated functions. These included melanin synthesis and formation of dendrite-like structures. Higher doses of phorbol dibutyrate, a less potent tumor promoter, were required to produce an effect comparable to that of PMA for dendrite induction. Phorbol and two other phorbol esters, which lack tumor-promoting activity, were either inactive or elicited a poor response. In addition to morphological changes, treatment with PMA altered glucosamine incorporation into membrane gangliosides. After PMA treatment, glucosamine incorporation increased 8- to 10-fold in the GM3 ganglioside and decreased 2-fold in the GM1 ganglioside, as compared to phorbol or untreated control. Inhibition of cell growth and stimulation of melanin synthesis were also observed after treatment of the HO cells with dimethyl sulfoxide. Unlike the tumor-promoting agents, dimethyl sulfoxide did not induce the formation of dendrite-like structures in the cells. These findings indicate that HO melanoma cells can be stimulated into terminally differentiated cells after treatment with tumor-promoting agents such as phorbol diesters.

Comparison of human and rat hepatocyte metabolism and mutagenic activation of 2-acetylaminofluorene.

A method has been developed to assess the metabolism and mutagenic activation of carcinogens using human and rodent hepatocytes in vitro. A slicing technique which was especially useful for nonperfusable biopsy and resected surgical human liver tissue was used to prepare the hepatocytes. Metabolites of the model carcinogen 2-acetylaminofluorene (AAF) produced by human and rat hepatocytes were similar and consisted primarily of 2-aminofluorene with ring hydroxylated products at the 1-, 3-, 5/9-, 7-, and 8-positions produced in addition to N-hydroxy-AAF. Sulphate and glucuronide conjugates of ring-hydroxylated metabolites and 2-aminofluorene were detected. Metabolism and cell-mediated Salmonella mutagenicity illustrated interindividual variation with human hepatocytes. Levels of metabolism and mutagenesis were generally higher with human hepatocytes compared to rat hepatocyte results. The increased levels of metabolism and mutagenesis of AAF by human hepatocytes compared to rat hepatocytes probably indicates a different sensitivity to hepatocarcinogenic effects of AAF on humans as compared to rats. Understanding differences and similarities between human and rodent carcinogen activation capabilities should be useful in the extrapolation of rodent carcinogenesis data to humans.

Inhibition of benzo(a)pyrene-induced transformation of C3H/10T1/2 cells by allylisopropylacetamide and isopropylvaleramide.

Allylisopropylacetamide (AIA) and isopropylvaleramide (IVA) have been demonstrated previously to protect in vivo against the acute toxicity and adrenal necrotic effect of 7,12-dimethylbenz(a)anthracene. In the present study, the influence of these two amides on the in vitro transforming ability of two potent carcinogens, benzo(a)pyrene [B(a)P] and 7,12-dimethylbenz(a)anthracene, on C3H10T1/2 cells was investigated. Both AIA and IVA showed a dose-dependent inhibition of B(a)P-induced transformation of C3H10T1/2 cells when added simultaneously for 24 hr with the carcinogen. While pretreatment, simultaneous treatment, and posttreatment of the cells with AIA or IVA inhibited transformation, the 24-hr posttreatment was somewhat more effective. The protective effect did not appear to results from alterations in B(a)P metabolism inasmuch as aryl hydrocarbon hydroxylase activity and the metabolic products of B(a)P detected by high-pressure liquid chromatography were not changed by AIA or IVA treatment. Furthermore, AIA and IVA did not selectively kill chemically transformed C3H10T1/2 cells, as indicated by the absence of their effect on an established, chemically transformed cell line. AIA and IVA also inhibited 7,12-dimethylbenz(a)anthracene-induced transformation of C3H10T1/2 cells. These data suggest that AIA and IVA may be useful protective agents and that they presumably exert their protective effect at some stage between the activation of the carcinogen and the expression of the transformed phenotype.

Human and rat kidney cell metabolism of 2-acetylaminofluorene and benzo(a)pyrene.

The metabolism and mutagenic activation of the model carcinogens benzo(a)pyrene [B(a)P] and 2-acetylaminofluorene (AAF) by human and rat kidney cells were measured. A slicing technique followed by enzyme digestion was utilized to obtain the kidney cells. Although levels of total metabolism of B(a)P by rat and human kidney cells were similar, analysis of specific metabolites of B(a)P indicated that species differences existed. Human kidney cells produced the organic-soluble metabolites B(a)P-9,10-diol, B(a)P-4,5-diol, B(a)P-7,8-diol, B(a)P-3,6-quinone, and B(a)P-9-phenol. Rat kidney cells produced organic-soluble B(a)P-pre-9,10-diols, B(a)P-9,10-diol, B(a)P-4,5-diol, and B(a)P-6,12-quinone. Both species produced sulfate and glucuronide conjugates of all products. For AAF, kidney cells from some human tissues produced up to four times the level of total metabolites compared to rat kidney cells. Organic-soluble metabolites were qualitatively similar between the species and consisted of 2-aminofluorene (AF), N-hydroxy-AAF and ring-hydroxylated products at the 1, 3, 5/9, 7, and 8 positions. Sulfate and glucuronide conjugates of these metabolites were also detected. Human interindividual variation with kidney cells was about 2.5-fold for total AAF metabolism and up to 6-fold for individual AAF metabolites. For B(a)P metabolism, human interindividual variation in total metabolism was low while for specific metabolites there was up to a 4-fold variation. Levels of AAF and AF cell-mediated Salmonella typhimurium mutagenesis were significantly higher with human cells as compared to rat kidney cells. It appears that the differences between human and rodent kidney cell metabolism of chemical carcinogens vary with the chemical class and understanding these differences will be necessary in the extrapolation of rodent carcinogenesis data to humans.

Maintenance of adult rat hepatocytes on C3H/10T1/2 cells.

A procedure is described for maintaining primary cultures of adult rat hepatocytes on a layer of irradiated C3H/10T1/2 cells. These hepatocytes were capable of metabolizing the liver carcinogen N-2-acetylaminofluorene to water-soluble products and after 14 days in culture could still metabolize approximately 70% of the Day 1 level. Hepatocytes maintained on the C3H/10T1/2 cells were inducible for the liver-specific enzyme tyrosine aminotransferase, and exhibited approximately a 4-fold induction by hydrocortisone during a 10-day culture period. Morphologically, these hepatocytes retained many characteristics of hepatocytes in vivo. By contrast, hepatocytes maintained on plastic lost both N-2-acetylaminofluorene-metabolizing ability and tyrosine aminotransferase activity by Day 5. This was presumably due to degeneration of the hepatocytes and an overgrowth by fibroblasts. The maintenance of morphologically and biochemically functional hepatocytes in culture on feeder cells may provide a valuable approach for studying drug metabolism and liver cell transformation in vitro.

Mutagenic activities of oxidized derivatives of N-nitrosodipropylamine in the liver cell-mediated and Salmonella typhimurium assays.

The mutagenic activity of N-nitrosobis(2-oxopropyl)amine (BOP), N-nitroso(2-hydroxypropyl) (2-oxopropyl)amine (HPOP), N-nitrosobis(2-hydroxypropyl)amine (BHP), N-nitrosomethyl-2-oxopropylamine (MOP), and N-nitrosomethyl-2-hydroxypropylamine (MHP) was examined in the Ames liquid incubation assay, using hamster liver homogenate for metabolic activation, and in the hamster liver cell-mediated V79 cell assay. At similar concentrations, the cell-mediated assay showed a greater mutagenic response over background to these nitrosamines than did the bacterial assay. Also, the relative mutagenic potency in the cell-mediated assay (MOP > MHP > BOP > HPOP > BHP) correlated better than that in the Ames assay (HPOP > MHP greater than or equal to BOP = BHP = MOP) with overall carcinogenic potency in the hamster (MOP > BOP > HPOP > BHP). The liver cell-mediated assay may be an important adjunct to the battery of short-term tests for carcinogenicity prescreening.

Bovine bladder urothelial cell activation of carcinogens to metabolites mutagenic to Chinese hamster V79 cells and Salmonella typhimurium.

The ability of bovine bladder urothelial cells to activate genotoxic chemicals to mutagens was examined by cocultivating bladder cells with Chinese hamster V79 cells or Salmonella typhimurium as mutable targets. Activation of test chemicals to mutagenic intermediates by urothelial cells was detected by induction of 6-thioguanine resistance in V79 cells or by induction of histidine revertants in Salmonella. In the bladder cell-mediated V79 cell mutagenesis system, a significant increase in mutation frequency was induced by exposure to 7,12-dimethylbenz(a)anthracene and dimethylnitrosamine. The aromatic amines 2-aminofluorene, 2-acetylaminofluorene, and 4-aminobiphenyl were weakly mutagenic to V79 cells with bladder cell activation, while no mutagenic activity was detected with 1-naphthylamine, 2-naphthylamine, or benzidine. Because the mutagenic activity of the aromatic amines was low with V79 cells as the target, a bladder cell-mediated S. typhimurium system was developed for these chemicals. The aromatic amines 2-aminofluorene, 2-acetylaminofluorene, 4-aminobiphenyl and 2-naphthylamine were mutagenic to S. typhimurium TA98 and TA100 in the presence of bladder cells but not in their absence. Benzidine was mutagenic to TA98 but not to TA100. The putative noncarcinogen 1-naphthylamine was not mutagenic in the system. In contrast to the V79 data, 7,12-dimethylbenz(a)anthracene and dimethylnitrosamine were not mutagenic with either bacterial strain. Mutagenic responses were related to both the number of bladder cells used for activation and the concentration of test chemical in the Salmonella assay. The data demonstrate that bovine bladder urothelial cells can activate carcinogens from three chemical classes to mutagens and indicate the different sensitivities of V79 cells and S. typhimurium to genotoxic agents.

Carcinogenicity of N-nitrosomethyl(2-oxopropyl)amine in Syrian hamsters.

7-Methylguanine was found in hydrolysates of liver and pancreas DNA from Syrian golden hamsters given a single dose of N-[1-14C]nitrosobis(2-oxopropyl)amine (BOP). This led us to examine the carcinogenicity of a potential methylating metabolite of BOP, N-nitrosomethyl(2-oxopropyl)amine (MOP). MOP was found to be a potent pancreatic carcinogen by either single or weekly s.c. injections. A single MOP treatment (25 mg/kg body weight) induced ductular adenomas and/or adenocarcinomas in 80% of the hamsters. A higher incidence of these neoplasms was found in 93% and 87% of animals receiving, respectively, 3.5 and 1.75 mg MOP per kg body weight weekly for life. However, the lower dose (0.87 mg/kg body weight) was less effective, resulting in a 33% tumor incidence. Compared with the known potent pancreatic carcinogen BOP, MOP seemed to have a greater affinity for the pancreas since considerably lower doses were required to induce similar incidences of equivalent pancreatic tumors. Like BOP, MOP caused tumors of the liver (7 to 100% incidence), kidneys (7 to 80% incidence), and vascular system (7 to 27% incidence). However, in contrast to BOP, which was noncarcinogenic to the upper respiratory tract, MOP-treated animals developed a high incidence of nasal cavity tumors (40% after a single treatment and 27 to 100% after weekly injections). The mutagenesis studies using hamster liver cell-mediated V79 cells confirmed the stronger effect of MOP compared to BOP. The assumption that MOP might be a proximate carcinogenic metabolite of BOP could not be substantiated by our methods for determining the in vivo and in vitro metabolites of BOP.

Genetic disruption of Ptgs-1, as well as Ptgs-2, reduces intestinal tumorigenesis in Min mice.

Two isoforms of cyclooxygenase (COX) are known, and to date most studies have implicated COX-2, rather than COX-1, as the isoform involved in colon carcinogenesis. In the present study, we show that homologous disruption of either Ptgs-1 or Ptgs-2 (genes coding for COX-1 or COX-2, respectively) reduced polyp formation in Min/+ mice by approximately 80%. Only COX-1 protein was immunohistochemically detected in normal intestinal tissue, whereas both COX-1 and variable levels of COX-2 protein were detected in polyps. Prostaglandin E2 was increased in polyps compared with normal tissue, and both COX-1 and COX-2 contributed to the PGE2 produced. The results indicate that COX-1, as well as COX-2, plays a key role in intestinal tumorigenesis and that COX-1 may also be a chemotherapeutic target for nonsteroidal anti-inflammatory drugs.

Gangliosides of chemically and virally transformed rat embryo cells.

The gangliosides of control rat embryo cells, 3-methylcholanthrene, Rauscher leukemia virus, and combined 3-methylcholanthrene-Rauscher leukemia virus transformants were examined using [14 C]glucosamine as a tracer. All four cell lines exhibited a complex pattern of gangliosides. While N-acetylgalactosaminyl-(N-acetylneuraminyl)-galactosyl-glucosyl-ceramide was the major ganglioside in the control cell line, N-acetylneuraminyl-galactosyl-glucosyl-ceramide was the major ganglioside in the three transformants. The 3-methylcholanthrene transformant possessed a ganglioside pattern different from that of the Rauscher leukemia virus transformant. Hydrolysis of the gangliosides indicated that galactosamine, N-acetyl-and N-glycolylneuraminic acid were the labeled components in all cell lines.

Cyclooxygenase-1 and -2 knockout mice demonstrate increased cardiac ischemia/reperfusion injury but are protected by acute preconditioning.

BACKGROUND: The purpose of this study was to examine the effects of cyclooxygenase (COX) deficiency on baseline functional characteristics and on recovery of left ventricular developed pressure (LVDP) after 20 minutes of global ischemia and 40 minutes of reperfusion in untreated and preconditioned hearts. METHODS AND RESULTS: Compared with hearts from wild-type (WT) and COX-2(-/-) mice, baseline cardiac prostaglandin (PG) E(2) and 6-keto-PGF(1alpha) levels were significantly decreased in hearts from COX-1(-/-) mice. After ischemia, cardiac PGE(2) levels increased in WT, COX-1(-/-), and COX-2(-/-) mice (P<0.05). Recovery of function (LVDP) after global ischemia in hearts from COX-1(-/-) and COX-2(-/-) mice was significantly less than in WT hearts. Pretreatment of WT mice with indomethacin for 2 days before ischemia significantly decreased LVDP recovery; however, perfusion of WT hearts with indomethacin for 40 minutes before ischemia did not significantly alter LVDP recovery. Postischemic recovery of LVDP in COX-1(-/-) and COX-2(-/-) was unchanged by perfusion with 5 micromol/L PGE(2), PGD(2), PGF(2alpha), or carboprostacyclin. Hearts from COX-2(-/-) mice showed an increase in ischemic contracture compared with hearts from WT and COX-1(-/-) mice; however, hearts did not differ in intracellular pH, ATP, or inorganic phosphate during ischemia. Ischemic preconditioning significantly improved postischemic LVDP recovery in COX-1(-/-), COX-2(-/-), and WT mice. CONCLUSIONS: Genetic disruption or 2-day chemical inhibition of COX-1 and COX-2 decreases recovery of LVDP after ischemia; however, acute perfusion with indomethacin is not detrimental. These data are consistent with protection due to the altered expression of some protein that is modulated by COX or its metabolites.