Why do a wide variety of animals retain multiple isoforms of cyclooxygenase?

Cyclooxygenase (COX) has been cloned from the phyla Cnidaria, Mollusca, Arthropoda, and Chordata of the animal kingdom. Many organisms have multiple COX isoforms that have arisen from gene duplication. It is not well understood why there are multiple COX isoforms in the same organism, or when duplication of the COX gene occurred. Here, we summarize the current knowledge of the evolutionary history of COX in the animal kingdom and discuss the reasons why the multiple COX system has been retained so widely. The phylogenetic analysis suggests that all COX genes in animals may descend from a common ancestor and that the duplication of an ancestral COX gene might occur within each lineage after the divergence of the animal. In most instances, the expressions of multiple COX isoforms are separately regulated and these isoforms play different and important pathophysiological roles in each organism. This may be the reason why multiple COX isoforms are widely retained.

COX-2 expression in stromal fibroblasts self-limits their numbers in lymph node inflammatory responses.

We previously reported the expression of cyclooxygenase (COX)-2 in draining lymph nodes during carrageenin-induced pleurisy of rats. Here, we analyzed histological and immunohistochemical characteristics of COX-2-expressing cells. After carrageenin administration into the pleural cavity of rats, parathymic lymph nodes were enlarged beginning at 8h and peaking from 24 to 48h. Lymphatic follicles disappeared 16h after injection, and numerous macrophages and fibroblasts were observed in the cortical region. COX-2-expressing cells in the cortical region showed characteristic dendritic processes from 16 to 48h and primarily co-localized with stromal fibroblastic reticular cell markers, α-smooth muscle actin (α-SMA), and desmin. Expression of α-SMA increased following COX-2 expression. Nimesulide, a COX-2 inhibitor, increased the dendritic processes of COX-2-expressing cells as well as expression of both COX-2 and α-SMA. These results suggest that COX-2-expressing cells may be stromal fibroblastic cells, which negatively self-regulate their proliferation and modulate tissue remodeling of draining lymph nodes at inflammatory sites.

Temporal change in IL-6 mRNA and protein expression produced by cyclic stretching of human pulmonary artery endothelial cells.

The time courses of interleukin (IL)-6 gene expression and protein production were examined in human pulmonary artery endothelial cells (HPAECs) subjected to cyclic stretching. IL-6 protein was increased even in cells without stretching. Fold changes determined by dividing the level of IL-6 protein in stretched cells by that in unstretched cells at the same sampling times indicated that IL-6 protein was increased by stretching. At least 1 h of stretching was necessary to elicit an increase of IL-6 protein, and the levels peaked at 3 h after the start of stretching. After withdrawal of stretching, there was no further increase of IL-6 protein. The expression levels of the IL-6 gene were significantly increased by stretching and peaked at 30 min after the start of stretching. The difference in the peak times of IL-6 gene and protein expression likely reflects the process of protein synthesis after the appearance of IL-6 mRNA.

Contribution of the prostaglandin E2/E-prostanoid 2 receptor signaling pathway in abscess formation in rat zymosan-induced pleurisy.

Abscess formation is a classic host response to infection by many pathogenic microorganisms. Here, we studied the role of prostaglandins (PGs) and their signal transduction in abscess formation. Zymosan was injected into the pleural cavity of rats. Expression of enzymes involved in PG synthesis, their receptors, and cytokines in exudate leukocytes and abscesses were analyzed by polymerase chain reaction, Western blotting, and immunohistochemistry. Treatment with ketorolac, a cyclooxygenase (COX)-1 inhibitor, or N-[2-cyclohexyloxy-4-nitrophenyl] methanesulfonamide (NS-398), a COX-2 inhibitor, reduced the size of abscesses and the number of cells recovered from the abscess. COX-2 was detected in leukocytes of the exudate and a marginal area of abscesses. Among detected terminal PG synthases, the major one was cytosolic PGE synthase. Membrane-bound PGE synthase (mPGES)-1 was detected in cells that were similar to the COX-2-expressing cells in morphology and localization. A high level of the E-prostanoid (EP)(2) receptor and a low level of the EP(4) receptor were detected. The expression pattern of the EP(2) receptor paralleled that of COX-2 and mPGES-1. 11,15-O-Dimethyl PGE(2) (ONO-AE1-259), an EP(2) receptor agonist, and rolipram, a phosphodiesterase type-4 inhibitor, reversed the effects of COX inhibitors on abscess formation. In contrast, 16-(3-methoxymethyl) phenyl-omega-tetranor-3,7-dithia PGE(1) (ONO-AE1-329), an EP(4) receptor agonist, did not reverse the effects of NS-398. Moreover, NS-398 reduced the mRNA levels in exudate leukocytes of some proinflammatory and fibrogenic cytokines, which was reversed by ONO-AE1-259. These results suggest that PGE(2) generated via COX-1 and COX-2 may interact with the EP(2) receptor and may up-regulate in cAMP-dependent fashion the production of cytokines that promote abscess formation.

Role of the cyclooxygenase 2-thromboxane pathway in 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced decrease in mesencephalic vein blood flow in the zebrafish embryo.

Previously, we reported that 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) evoked developmental toxicity required activation of aryl hydrocarbon receptor type 2 (AHR2), using zebrafish embryos. However, the downstream molecular targets of AHR2 activation are largely unknown and are the focus of the present investigation. TCDD induces cyclooxygenase 2 (COX2), a rate-limiting enzyme for prostaglandin synthesis in certain cells. In the present study, we investigated the role of the COX2-thromboxane pathway in causing a specific endpoint of TCDD developmental toxicity in the zebrafish embryo, namely, a decrease in regional blood flow in the dorsal midbrain. It was found that the TCDD-induced reduction in mesencephalic vein blood flow was markedly inhibited by selective COX2 inhibitors, NS-398 and SC-236, and by a general COX inhibitor, indomethacin, but not by a selective COX1 inhibitor, SC-560. Gene knock-down of COX2 by two different types of morpholino antisense oligonucleotides, but not by their negative homologs, also protected the zebrafish embryos from mesencephalic vein circulation failure caused by TCDD. This inhibitory effect of TCDD on regional blood flow in the dorsal midbrain was also blocked by selective antagonists of the thromboxane receptor (TP). Treatment of control zebrafish embryos with a TP agonist also caused a reduction in mesencephalic vein blood flow and it too was blocked by a TP antagonist, without any effect on trunk circulation. Finally, gene knock-down of thromboxane A synthase 1 (TBXS) with morpholinos but not by the morpholinos' negative homologs provided significant protection against TCDD-induced mesencephalic circulation failure. Taken together, these results point to a role of the prostanoid synthesis pathway via COX2-TBXS-TP in the local circulation failure induced by TCDD in the dorsal midbrain of the zebrafish embryo.

Interaction between cyclooxygenase (COX)-1- and COX-2-products modulates COX-2 expression in the late phase of acute inflammation.

Prostanoid production depends on the activity of two cyclooxygenase (COX) isoforms. It is appreciated that COX-1 plays a role in physiological processes, whereas COX-2 acts in pathological conditions. However their roles, particularly roles of COX-1, have not yet been fully established in inflammation. Here, we examined the effects of COX inhibitors, having differential isoform selectivity, on the late phase of rat carrageenin-induced pleurisy to elucidate the role of COX-2 expressed in the draining lymph nodes and found substantial contribution of COX-1-product(s). Protein and mRNA of COX-2 were detectable with Western blotting analysis and reverse-transcription polymerase chain reaction (RT-PCR) analysis in parathymic lymph nodes, peaking at 48 h after induction of pleurisy. Microsomal prostaglandin E synthase (mPGES)-1 was detectable by immunohistochemical analysis in cells with dendritic processes, a morphological characteristic similar to that of COX-2 expressing cells. Although aspirin, indomethacin and a COX-1 inhibitor, ketorolac, significantly decreased the volume of pleural exudate, they did not affect the levels of COX-2 and mPGES-1 in the lymph node 24 h after induction of pleurisy. In contrast, COX-2 inhibitors, nimesulide and NS-398, had no effect on the exudate volume, but they increased the number of COX-2- and mPGES-1-expressing cells and extension of their dendritic processes with significant increase in the COX-2 level, which were antagonised by ketorolac. These results suggest that COX-2-expressing cells may negatively self-regulate their functions by producing PGE2 via mPGES-1: migration into the draining lymph node and their differentiation. Moreover, COX-1- and COX-2-derived prostanoids may play differential or sometimes antagonistic roles in the late phase of acute inflammation.

Local and systemic expression of inducible nitric oxide synthase in comparison with that of cyclooxygenase-2 in rat carrageenin-induced pleurisy.

Expression of inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) is up-regulated in response to inflammatory stimuli. To evaluate the extent to which local pleural inflammation involves additional site in the pleural cavity and elsewhere, we investigated the time course of the levels of iNOS and its product in the inflammatory and other sites, and compared those with a level of COX-2 in rat carrageenin-induced pleurisy. The exudate and plasma NOx levels rose, reaching peaks at 9 and 14 h, respectively. Both COX-2 and iNOS became detectable in exudate leukocytes, their levels reaching peaks at 3 and 9 h after irritation, respectively. COX-2 was detectable mainly in neutrophils, but iNOS was detectable in both neutrophils and mononuclear leukocytes. Furthermore, iNOS became detectable in neutrophils and mononuclear leukocytes in enlarged parathymic lymph nodes from 3h in addition to those in peripheral blood and Kupffer cells from 3 to 14 h, respectively. The gene product is also detectable in thymic large dendritic cells of pleurisy-induced rats as well as normal control rats. COX-2 became detectable in stellar dendritic cells of the enlarged draining lymph nodes from 14 h. Thus, these gene products were induced in the immediate proximity of regional lymph nodes, and even at a considerable distance of liver by the local inflammatory stimulus. Although their expression pattern was quite different from each other, these gene products were detectable in phagocytic cells.

Concurrent evolution and resolution in an acute inflammatory model of rat carrageenin-induced pleurisy.

Granulocyte apoptosis and subsequent clearance by phagocytes are critical for the resolution of inflammation. However, no studies have addressed how the resolution proceeds in the inflammatory site. We studied the time course of neutrophil apoptosis and the following ingestion by mononuclear leukocytes in rat carrageenin-induced pleurisy, detecting DNA fragmentation by the deoxyuridine triphosphate-biotin nick-end labeling (TUNEL) method, by acridine orange staining, and from the DNA ladder pattern on electrophoresis. Neutrophil accumulation started 3-5 h after carrageenin injection and then maintained a plateau until 24 h. Neutrophils decreased steeply between days 1 and 3. Mononuclear leukocytes started to accumulate at 5 h and reached a peak at day 2. TUNEL-positive bodies and acridine orange-positive bodies first became detectable in the cytoplasm of the mononuclear leukocytes from 24 h and 9 h, respectively. Both methods indicated that mononuclear leukocytes containing fragmented DNA increased rapidly on days 1 and 2 and reached a peak at day 3. The characteristic ladder pattern of neutrophil DNA was observed from 5 h. Tumor necrosis factor alpha was detectable on the start, and the levels of interleukin-10 and transforming growth factor-beta1 rose together with signs of neutrophil apoptosis and the following ingestion by mononuclear leukocytes. These results indicate that neutrophils start to undergo apoptosis just after the beginning of their accumulation in the inflammation site. Thus, evolution and resolution processes may proceed concurrently in acute inflammation.