Compensatory prostaglandin E2 biosynthesis in cyclooxygenase 1 or 2 null cells.

Prostaglandin E2 (PGE2) production in immortalized, nontransformed cells derived from wild-type, cyclooxygenase 1-deficient (COX-1(-/-)) or cyclooxygenase 2-deficient (COX-2(-/-)) mice was examined after treatment with interleukin (IL)-1beta, tumor necrosis factor alpha, acidic fibroblast growth factor, and phorbol ester (phorbol myristate acetate). Compared with their wild-type counterparts, COX-1(-/-) or COX-2(-/-) cells exhibited substantially enhanced expression of the remaining functional COX gene. Furthermore, both basal and IL-1-induced expression of cytosolic phospholipase A2 (cPLA2), a key enzyme-regulating substrate mobilization for PGE2 biosynthesis, was also more pronounced in both COX-1(-/-) and COX-2(-/-) cells. Thus, COX-1(-/-) and COX-2(-/-) cells have the ability to coordinate the upregulation of the alternate COX isozyme as well as cPLA2 genes to overcome defects in prostaglandin biosynthetic machinery. The potential for cells to alter and thereby compensate for defects in the expression of specific genes such as COX has significant clinical implications given the central role of COX in a variety of disease processes and the widespread use of COX inhibitors as therapeutic agents.

The isozyme-specific effects of cyclooxygenase-deficiency on bone in mice.

Prostaglandin E(2) (PGE(2)) plays a critical role in skeletal physiology and bone loss. PGE(2) production is regulated in vivo by at least two cyclooxygenase (COX) isozymes, COX-1 and COX-2. The purpose of this study was to investigate the in vivo effects of the selective deletion of COX-1 or COX-2 on bone mineral density (BMD), bone microarchitecture and bone strength in wild type (WT), COX-1(-/-) and COX-2(-/-) mice. Using a LUNAR PIXImus, BMD was measured in 18 (WT), 18 COX-1(-/-) and 16 COX-2(-/-) mice. COX-1(-/-) mice exhibited significantly higher BMD (0.0506 g/cm(2) +/- 0.0014 g/cm(2)) than either WT (0.0493 g/cm(2) +/- 0.0019, P < or = 0.05) or COX-2(-/-) (0.0473 g/cm(2) +/- 0.0034, P < or = 0.01) mice. COX-2(-/-) mice had significantly lower BMD than WT (P < or = 0.01) or COX-1(-/-) (P < or = 0.01). Flexure stress of the femurs, determined by breaking the bones with three-point bending, correlated with bone density. Although plasma levels of both Ca(2+) and PTH were comparable in wild type and COX-1(-/-) mice, both were elevated in COX-2(-/-) mice consistent with primary hyperparathyroidism. These studies suggest that COX enzymes are important regulators of BMD and bone strength in mice. The beneficial effect of absence of the COX-1 enzyme on skeletal parameters may be secondary to decreases in PGE(2). On the other hand, primary hyperparathyroidism and lower bone magnesium content may account for the lower BMD and impairments in bone strength of COX-2(-/-) mice. Further elucidation of the effects of the COX pathway on bone remodeling may provide important information on potential therapeutic targets for preventing and/or treating osteoporosis.

Synthesis of S-adenosylethionine by the gamma isozyme of methionine adenosyltransferase from Friend erythroleukemic cells.

Methionine adenosyltransferase (adenosine 5'-triphosphate-L-methionine S-adenosyltransferase, EC 2.5.1.6) was found to occur only as the gamma isozyme in Friend erythroleukemic cells. Enzyme activity in a dialyzed 105,000 X g supernatant was linear for at least 30 min and dependent on the amount of protein added. The Km for methionine was 10.6 microM. In the presence of 10% dimethyl sulfoxide, the enzyme activity was slightly inhibited and dithiothreitol was not required for maximum activity. These properties identify the methionine adenosyltransferase in Friend erythroleukemic cells as the gamma isozyme. L-Ethionine served as a substrate with a Km of 30 microM, and competitively inhibited the enzyme activity with a Ki of 150 microM. Friend erythroleukemic cells grown in the presence of ethionine accumulated S-adenosylethionine, which was dose and time dependent. Therefore, the gamma isozyme of methionine adenosyltransferase from Friend erythroleukemic cells utilize L-ethionine as substrate, resulting in an accumulation of S-adenosylethionine. These studies provide a mechanism by which ethionine through S-adenosylethionine may alter the methylation of DNA in Friend erythroleukemic cells.

Cholecalciferol induces prostaglandin E2 biosynthesis and transglutaminase activity in human keratinocytes.

In this study, we examined the effects of cholecalciferol, a primary keratinocyte metabolite and precursor of the hydroxylated form of vitamin D3, 1alpha,25-dihydroxyvitamin D3 [1alpha,25(OH)2D3], on prostaglandin E2 (PGE2) production in human keratinocytes by examining its respective effects on cyclooxygenase-1 (COX-1), cyclooxygenase-2 (COX-2), and cytosolic phospholipase A2 (cPLA2) expression, the rate-limiting enzymes regulating PGE2 biosynthesis and differentiation of keratinocytes. Cholecalciferol induced PGE2 production, whereas 1alpha,25(OH)2D3 had no effect on PGE2 production both in normal human epidermal keratinocytes and in the immortalized human keratinocyte cell line, HaCaT. In HaCaT cells, neither COX-1 mRNA nor protein was detectable without stimulation and COX-1 expression did not increase in response to cholecalciferol treatment. Although cPLA2 mRNA and protein were constitutively expressed in untreated HaCaT cells, expression levels did not increase in response to cholecalciferol treatment; however, unlike COX-1 and cPLA2 expression, COX-2 mRNA and COX-2 protein expression increased in response to cholecalciferol treatment. Calphostin C, a potent protein kinase C inhibitor, significantly reduced cholecalciferol-induced PGE2 production by inhibiting cholecalciferol-enhanced COX-2 mRNA and protein expression. These results indicate that (i) 1alpha,25(OH)2D3 does not induce PGE2 biosynthesis in keratinocytes, (ii) cholecalciferol-induced PGE2 production is primarily COX-2 dependent, and (iii) cholecalciferol enhances both COX-2 mRNA and protein expression, via a protein kinase C-dependent mechanism in human keratinocytes. Furthermore, cholecalciferol increased total cellular transglutaminase activity dose dependently, suggesting a potential role for cholecalciferol in regulating the differentiation of human keratinocytes.

Inducible nitric oxide synthase gene expression and enzyme activity correlate with disease activity in murine experimental autoimmune encephalomyelitis.

Messenger RNA encoding inducible NO synthase (iNOS) was measured by competitive reverse transcriptase polymerase chain reaction (cRT-PCR) and ribonuclease protection assays in spinal cords from mice at varying stages of experimental allergic encephalomyelitis (EAE) and from control mice. iNOS mRNA was increased in spinal cords from mice with acute EAE. cRT-PCR assays revealed a 10-20-fold increase in iNOS mRNA in spinal cords during acute EAE compared with the level observed in normal mouse spinal cords. Functional iNOS activity, as assessed by assay of calcium-independent citrulline production, was also significantly increased in spinal cords from mice with acute EAE in comparison to normal controls. The correlation of functional iNOS expression with active disease in EAE in consistent with a pathogenic role for excess NO in this model of cell-mediated central nervous system autoimmunity.

A study of the mechanism of selenite-induced hypomethylated DNA and differentiation of Friend erythroleukemic cells.

Sodium selenite is a good inducer of hemoglobin production in Friend erythroleukemic cells (FELC). At a concentration of 20 microM 70-80% of the cells produce hemoglobin and the DNA is hypomethylated. What is the mechanism for sodium selenite alteration of the DNA methylation pattern? Experiments with methionine adenosyltransferase (the enzyme that synthesizes adenosylmethionine) showed little effect of selenite on the activity of this enzyme in vitro or in vivo. Therefore, FELC are able to synthesize S-adenosylmethionine in the presence of sodium selenite. When sodium selenite was added to an in vitro assay for DNA methylase, the enzyme was non-competitively inhibited by 80% at 20 microM selenite with a Ki of 6 microM. DNA methylase isolated from control and selenite-treated FELC was purified through a DEAE-Sephacel column and no difference in activity was found. In the presence of selenite, DNA methylase is very sensitive to selenite inhibition, but removal of the selenite restores activity. However, DNA synthesized by FELC grown in the presence of selenite (no DNA methylase activity) was found to be hypomethylated. These results suggest that DNA methylase activity is inhibited in FELC grown in the presence of sodium selenite.

Inhibition of DNA methylase activity by acrolein.

Acrolein, a reactive metabolite of cyclophosphamide, may be responsible for bladder cancer induced by cyclophosphamide. DNA methylase was isolated from the liver and urothelium of rats by high salt extraction of purified nuclei. Acrolein at 10 microM inhibited liver and bladder DNA methylase activity by 30-50%. Kinetic studies with the liver enzyme showed a competitive type of inhibition with a Ki of 6.7 microM. Both dithiothreitol and glutathione afforded protection to the enzyme when added to the assay. At near equimolar concentrations of glutathione to acrolein, the methylase retained 80-90% activity. An increase in DNA had no effect on the inhibition by acrolein, whereas increased amounts of protein protected against acrolein inhibition, suggesting that acrolein reacted with the DNA methylase protein. On the other hand, DNA that had been reacted with acrolein was unable to serve as a substrate for DNA methylase. As the DNA adducts increased the methylation of the DNA decreased. Thus, acrolein has the ability to react with DNA and the DNA methylase protein, either of which results in inhibition of DNA methylation.

Gene amplification and chromosome rearrangements: a study of a single cell lineage selected for amplification and deamplification of the UMP synthase gene.

The UMP synthase gene is stably amplified in Chinese hamster lung cells selected for resistance to pyrazofurin (PF) and 6-azauridine (6AUR), inhibitors of the decarboxylase activity of the bifunctional UMP synthase enzyme. The amplified DNA is located intrachromosomally as expanded chromosomal regions (ECRs). Growth of these cells in 5-fluorouracil enables rapid selection of cells that have undergone deamplification and consequently lost resistance to PF + 6AUR. Detailed cytogenetic analyses and fluorescence in situ hybridization on three consecutive amplification-deamplification cycles in descendants of the same cloned cell showed a unique position and structure for the ECR. In the first cycle of amplification, the ECR forms a homogeneously staining region on a small marker chromosome (M3). In the second cycle of amplification, a chromosomal break was noted at the site of the endogenous UMP synthase gene on another derivative chromosome, M2, with amplification resulting in an abnormally banded region on a third marker (M3). The third cycle of amplification produced a cell line with an ECR on the distal portion of M2. This ECR was unstable, showing variations in size as well as translocations and other chromosome rearrangements. Our data, taken in its entirety, suggest a relationship between amplification as an ECR and chromosome rearrangements in Chinese hamster cells.

Nociception in cyclooxygenase isozyme-deficient mice.

Prostaglandins formed by cyclooxygenase-1 (COX-1) or COX-2 produce hyperalgesia in sensory nerve endings. To assess the relative roles of the two enzymes in pain processing, we compared responses of COX-1- or COX-2-deficient homozygous and heterozygous mice with wild-type controls in the hot plate and stretching tests for analgesia. Preliminary observational studies determined that there were no differences in gross parameters of behavior between the different groups. Surprisingly, on the hot plate (55 degrees C), the COX-1-deficient heterozygous groups showed less nociception, because mean reaction time was longer than that for controls. All other groups showed similar reaction times. In the stretching test, there was less nociception in COX-1-null and COX-1-deficient heterozygotes and also, unexpectedly, in female COX-2-deficient heterozygotes, as shown by a decreased number of writhes. Measurements of mRNA levels by reverse transcription-PCR demonstrated a compensatory increase of COX-1 mRNA in spinal cords of COX-2-null mice but no increase in COX-2 mRNA in spinal cords of COX-1-null animals. Thus, compensation for the absence of COX-1 may not involve increased expression of COX-2, whereas up-regulation of COX-1 in the spinal cord may compensate for the absence of COX-2. The longer reaction times on the hot plate of COX-1-deficient heterozygotes are difficult to explain, because nonsteroid anti-inflammatory drugs have no analgesic action in this test. Reduction in the number of writhes of the COX-1-null and COX-1-deficient heterozygotes may be due to low levels of COX-1 at the site of stimulation with acetic acid. Thus, prostaglandins made by COX-1 mainly are involved in pain transmission in the stretching test in both male and female mice, whereas those made by COX-2 also may play a role in the stretching response in female mice.

Transcriptional regulation of cyclooxygenase-2 in the human microvascular endothelial cell line, HMEC-1: control by the combinatorial actions of AP2, NF-IL-6 and CRE elements.

Interleukin-1beta (IL-1) is a potent inducer of cyclooxygenase-2 (COX-2) and prostaglandin biosynthesis in many types of cells, yet little is known about the molecular mechanisms regulating IL-1 mediated prostanoid biosynthesis in the endothelium of the microvasculature. Therefore, we examined the cis- and trans-acting factors regulating IL-1-induced COX-2 expression in the human microvascular endothelial cell line, HMEC-1. IL-1 enhanced steady state levels of COX-2 protein and mRNA synthesis by approximately 2-fold which preceded a 2-fold increase in PGF(alpha) biosynthesis. Expression of a series of COX-2 promoter-luciferase constructs in IL-1 treated HMEC-1 cells revealed that the 'full length' (-1432/+59 bp) promoter was 10 times more active than the SV-40 promoter/enhancer and that it could be further activated by IL-1. Surprisingly however, all except for the shortest COX-2 promoter construct retained the ability to respond to IL-1 and luciferase activity driven by -191/+59 bp COX-2 promoter was as responsive to IL-1 as the full-length promoter. Moreover, site-directed promoter mutagenesis and electophoretic mobility shift assays (EMSA) indicate that the combinatorial actions of AP2, NF-IL6, and CRE elements are critical for both constitutive and IL-1-inducible COX-2 promoter activity. Understanding the mechanism(s) regulating COX-2 gene expression and prostaglandin biosynthesis in the microvasculature has important implications with regard to inflammation and angiogenesis in vivo.

The genetic ablation of cyclooxygenase 2 prevents the development of autoimmune arthritis.

OBJECTIVE: To determine the effects of cyclooxygenase 1 (COX-1) and COX-2 gene deletion on collagen-induced arthritis (CIA). METHODS: Mice that were susceptible to CIA but lacked either the COX-1 or the COX-2 gene were immunized with type II collagen (CII), and the incidence and severity of arthritis were compared with findings in wild-type animals, by clinical and histologic examination. The immune response was assessed by measuring total CII IgG, IgG1, and IgG2 antibody production in sera from immunized mice. The passive transfer of arthritis, accomplished using anti-CII monoclonal antibodies, was tested in wild-type and COX-deficient (-/-) mice. Splenocytes cultured from CII-immunized wild-type and COX-/- mice were challenged with bovine alpha1(II), and cytokine production was assessed. RESULTS: COX-2 gene deletion reduced the incidence and severity of CIA compared with findings in wild-type and COX-1-/- mice. Histologic examination of joints after the onset of clinical arthritis revealed cartilage erosions, proliferation of the synovial lining, and inflammatory cell infiltration in wild-type and COX-1-/- mice, but not in COX-2-/- mice. COX-2-/- mice exhibited reduced anti-CII IgG antibody levels, indicating a decreased immune response. However, cytokine production by spleen cells from immunized mice indicated no cytokine deficiencies in COX-2-/- mice compared with wild-type or COX-1-/- mice. More important, arthritis could not be passively transferred to naive COX-2-/- mice, indicating a requirement for COX-2 in the pathogenesis of arthritis, independent of the immune response. CONCLUSION: COX-2-/- mice exhibit at least 2 defects resulting in down-modulation of the development of CIA: a reduced immune response to CII demonstrated by a markedly reduced antibody titer, and an "inflammatory" defect reflected by the inability to passively transfer arthritis to COX-2-/- mice.