The anti-inflammatory eicosanoid lipoxin A4 (LXA4), aspirin-triggered 15-epi-LXA4, and their stable analogs down-regulate IL-8 secretion and subsequent recruitment of neutrophils by intestinal epithelia. In an effort to elucidate the mechanism by which these lipid mediators modulate cellular proinflammatory programs, we surveyed global epithelial gene expression using cDNA microarrays. LXA4 analog alone did not significantly affect expression of any of the >7000 genes analyzed. However, LXA4 analog pretreatment attenuated induction of ∼50% of the 125 genes up-regulated in response to the gastroenteritis-causing pathogen Salmonella typhimurium. A major subset of genes whose induction was reduced by LXA4 analog pretreatment is regulated by NF-κB, suggesting that LXA4 analog was influencing the activity of this transcription factor. Nanomolar concentrations of LXA4 analog reduced NF-κB-mediated transcriptional activation in a LXA4 receptor-dependent manner and inhibited induced degradation of IκBα. LXA4 analog did not affect earlier stimulus-induced signaling events that lead to IκBα degradation, such as S. typhimurium-induced epithelial Ca2+ mobilization or TNF-α-induced phosphorylation of IκBα. To establish the in vivo relevance of these findings, we examined whether LXA4 analogs could affect intestinal inflammation in vivo using the mouse model of DSS-induced inflammatory colitis. Oral administration of LXA4 analog (15-epi-16-para-fluoro-phenoxy-LXA4, 10 μg/day) significantly reduced the weight loss, hematochezia, and mortality that characterize DSS colitis. Thus, LXA4 analog-mediated down-regulation of proinflammatory gene expression via inhibition of the NF-κB pathway can be therapeutic for diseases characterized by mucosal inflammation.

As an interface with the outside world, the intestinal epithelium dynamically alters its gene expression in response to the changing environments in both the intestinal lumen and the subepithelial domain. For example, in response to either lumenal enteric pathogens or subepithelial proinflammatory cytokines, the intestinal epithelium activates the expression of a panel of genes that promote an acute inflammatory response (1). As uncontrolled inflammation can result in tissue damage, the intestinal epithelium is also influenced by exogenous and endogenous anti-inflammatory mediators that attenuate proinflammatory responses. One example of this type of mediator and the focus of this study is that of the anti-inflammatory eicosanoid lipoxin (lipoxin A4 (LXA4)3).

Lipoxins such as LXA4 are derived from arachidonate as a result of its exposure to the unique combinations of lipoxygenases that occur during specific heterotypic cell-cell interactions such as those occurring in inflammation (e.g., epithelial-neutrophil interactions). LXA4-induced responses down-regulate events associated with inflammation in a variety of in vitro and in vivo models (2). Acetylation of cyclooxygenase by aspirin results in the biosynthesis of the 15-epimer of LXA4 (3). Such 15-epi-LXA4 as well synthetic analogs of LXA4 resist enzymatic degradation and thus have longer-lasting anti-inflammatory bioactivity than the native eicosanoid (4). LXA4 and its synthetic stable analogs attenuate the IL-8 expression that is induced in model epithelia in response to the gastroenteritis-causing pathogen Salmonella typhimurium and the proinflammatory cytokine TNF-α (5, 6). Lipoxin analogs also attenuate chemokine secretion by human colon, resulting in reduced neutrophil adherence and tissue damage (7).

The mechanism by which LXA4 analogs down-regulate IL-8 expression is largely unknown, although we have shown that IL-8 mRNA levels are reduced (5), implying action at the level of transcription. While this bioaction is at least somewhat specific, in that the mRNA levels of actin are not affected by LXA4 analogs, technology to broadly evaluate gene expression has not, until recently, been available. Microarray technology now permits simultaneous parallel measurement of the expression of thousands of genes, making it possible to evaluate the effect of a given mediator on global gene expression. We sought to use this technology to characterize LXA4 bioaction and perhaps better predict its in vivo behavior. In this study, we use this approach to test the hypothesis that LXA4 exerts its effects primarily on the transcriptional activation of genes involved in the proinflammatory epithelial response.

We observed that LXA4 analogs did not directly affect gene expression but broadly induced proinflammatory gene expression, particularly that regulated by NF-κB. This result is not surprising when one considers that LXA4 analogs are known to antagonize the effects of a number of proinflammatory agonists that signal through this transcription factor (as discussed above). Having identified a role for this factor, we next explored which elements of the signaling pathway were affected. Last, we tested the in vivo relevance of these findings in a mouse model of colitis.

15-(R/S)-methyl-LXA4 was synthesized by Dr. N. Petasis (University of Southern California, Los Angeles, CA) as previously described (4). 15-epi-16-parafluoro-phenoxy-LXA4 was supplied by Berlex Biosciences (Richmond, CA). Dextran sodium sulfate (DSS; m.w. ∼40,000) was obtained from ICN Pharmaceuticals (Costa Mesa, CA). MG-132 was obtained from Calbiochem (La Jolla, CA).

Polarized model intestinal epithelia were prepared via culturing T84 cells on permeable supports as previously described (8). Model epithelia were used 6–14 days after plating after verification (for T84) that they had achieved a transepithelial electrical resistance of at least 1000 Ωcm2. S. typhimurium was cultured and used to colonize model epithelia as previously described (5).

Total RNA was isolated with TRIzol (GIBCO, Gaithersburg, MD) following instructions by the manufacturer. mRNA was isolated, hybridization was performed by Incyte Genomics (Palo Alto, CA), and mRNA were analyzed as we have recently described (9). Each array condition was performed on RNA pooled from six individual 5-cm2 model epithelia so as to minimize the effect of experimental variability that might occasionally arise in individual samples. Changes in gene expression of 2-fold or more are highly likely to be significant (9).

HeLa cells (40–60% confluent) were transiently transfected using Superfect Reagent (Qiagen, Valencia, CA) with 2 μg of the reporter plasmid pIL-8-CAT (10) and variable quantities of the pCMV-myc-LXA4R expression plasmid (encoding LXA4 receptor) according to the manufacturer’s instructions. All cotransfection reactions were balanced for total amount of expression plasmid DNA with pCMV-myc vector. Approximately 16–24 h after transfection, cells were washed with HBSS and incubated with 0–100 nM 15-(R/S)-methyl-LXA4 for 1 h followed by TNF-α for 8 h. Cell lysates were prepared and assayed for chloramphenicol acetyl transferase (CAT) using the CAT ELISA kit from Roche (Basel, Switzerland).

Levels of IκBα and phospho-IκBα were assayed from whole cell lysates of model intestinal epithelia via immunoblotting using an IκBα Ab (Santa Cruz Biotechnology, Santa Cruz, CA) as previously described (11).

Intracellular Ca2+ was measured in fura 2-loaded polarized model epithelia via spectrofluorometry as previously described (11). Briefly, polarized model epithelia were prepared on customized supports permitting their insertion into a standard fluorometry cuvette. Model epithelia were incubated with 5 mM fura 2-AM (Molecular Probes, Eugene, OR) added for 60 min, and unincorporated probe was removed with a 10-min washing. Fluorescence was read with emission at 505 nm while the excitation wavelength is changed from 340 to 380 nm. Values of intracellular Ca2+ were calculated via the Grynciewitz equation (RRmin)/(RmaxR) × Kd. Rmax and Rmin are measured by adding digitonin (10 μM) and then EGTA (20 mM), respectively.

Six- to 8-wk-old BALB/c mice were obtained from The Jackson Laboratory (Bar Harbor, ME). One week following arrival, mice were given drinking water containing 4% DSS and 0.05% ethanol (vehicle) or 10 μg/ml 15-epi-16-parafluoro-LXA4 (following internal review board approved protocol). Water consumption was measured per cage and corrected for leakage/evaporation by comparison to identical water bottles placed in empty cages. Water was changed every 5 days, a time at which HPLC analysis indicated that degradation of the LXA4 analog had not occurred. Body mass was measured daily by a technician blinded to the drug protocol, who also checked mice daily for gross rectal bleeding and occult blood in stools via Hemoccult Sensa (Beckman Coulter, Fullerton, CA). Bleeding score was 0 (hemoccult negative), 1 (mild positive), 2 (strong positive), and 4 (gross blood) following the well-established procedure developed by Cooper et al. (12). We did not consistently observe diarrhea in these mice and therefore did not include this parameter in our data presentation. Mice that were judged (by the blinded technician) to be moribund were sacrificed (and classified as nonsurvivors), although only one mouse in this study fell into this category (others were found dead upon daily check). Seven days following administration of DSS, mice were switched to DSS-free water containing vehicle or LXA4 analog. Statistical significance was determined for body mass and survival data by Student’s t test with p < 0.05 termed significant.

HPLC analyses were conducted on LXA4 analog-containing drinking water via a LUNA 5 μm C18(2) column (250 × 4.60 mm) using a ProStar HPLC (Varian, Palo Alto, CA) equipped with a diode array detector.

To explore the mechanism of the anti-inflammatory bioactivity of LXA4 and its stable analogs, the effect of this eicosanoid on global gene expression in control and inflamed (S. typhimurium-infected) model intestinal epithelia was examined by high-density cDNA microarray. First, we asked whether LXA4 analogs alone globally influenced gene regulation, potentially up-regulating anti-inflammatory effector molecules. Model epithelia (6 × 5 cm2 epithelia per condition) were treated with vehicle (0.05% ethanol) or 100 nM of 15-(R/S)-methyl-LXA4 (a concentration known to attenuate agonist-induced IL-8 mRNA expression (5)) for 4 h. mRNA was isolated from each set of samples and pooled, and microarray analysis was performed by hybridization to 7075 independent cDNA targets according to the protocol of Incyte Genomics as previously described (9). The expression levels of each gene from untreated (labeled with Cy3 before hybridization) epithelia vs that of LXA4 analog-treated epithelia (Cy5 labeled) are plotted in Fig. 1. Accordingly, genes whose expression are unchanged lie on the central diagonal while, for example, genes whose expression is induced 2- to 5-fold are plotted between the upper diagonals labeled 2 and 5 (Fig. 1). Interestingly, LXA4 analog by itself did not induce any significant changes (2-fold or more (9)) in any of the genes included in the array used. These data suggest that LXA4 does not directly affect gene expression, and hence more likely modulates the signaling pathways by which proinflammatory agonists regulate gene expression.

FIGURE 1.

LXA4 analog attenuates changes in epithelial gene expression induced by S. typhimurium: scatter plot of expression levels from control and LXA4 analog-treated cells. Paired hextuplicate sets of T84 model epithelia were treated as follows for 4 h, at which time mRNA was isolated and analyzed via cDNA microarray hybridization. A, Epithelia were treated with 100 nM 15-(R/S)-methyl-LXA4 and compared with untreated (except vehicle) control. B, Epithelia were treated with vehicle for 1 h followed by apical colonization with 109 CFU S. typhimurium and compared with untreated control. C, Epithelia were treated with 100 nM 15-(R/S)-methyl-LXA4 vehicle for 1 h followed by apical colonization with 109 CFU S. typhimurium and compared with untreated control. Expression of each gene plotted as treated vs untreated so the diagonal represents no change from the control state. The diagonals represent indicated fold changes (above for positive) from control. The total number of genes between each set of diagonals is indicated.

FIGURE 1.

LXA4 analog attenuates changes in epithelial gene expression induced by S. typhimurium: scatter plot of expression levels from control and LXA4 analog-treated cells. Paired hextuplicate sets of T84 model epithelia were treated as follows for 4 h, at which time mRNA was isolated and analyzed via cDNA microarray hybridization. A, Epithelia were treated with 100 nM 15-(R/S)-methyl-LXA4 and compared with untreated (except vehicle) control. B, Epithelia were treated with vehicle for 1 h followed by apical colonization with 109 CFU S. typhimurium and compared with untreated control. C, Epithelia were treated with 100 nM 15-(R/S)-methyl-LXA4 vehicle for 1 h followed by apical colonization with 109 CFU S. typhimurium and compared with untreated control. Expression of each gene plotted as treated vs untreated so the diagonal represents no change from the control state. The diagonals represent indicated fold changes (above for positive) from control. The total number of genes between each set of diagonals is indicated.

Close modal

Next we examined the effect of LXA4 analog on the changes in epithelial gene expression that are induced by colonization with S. typhimurium (Table I). In previous experiments in which model epithelia were colonized with S. typhimurium for various lengths of time (A. Young, A. Gewirtz, and A. S. Neish, manuscript in preparation), up to 300 genes were significantly differentially and reliably regulated over a 2- to 6-h time course after attachment of this pathogen to the apical epithelial surface. The maximal number of differentially regulated genes was observed 4 h postcolonization, so this time point was selected to study potential effects of lipoxins on proinflammatory gene regulation. Thus, model epithelia were treated with vehicle (0.05% ethanol) or 100 nM of 15-(R/S)-methyl-LXA4 for 90 min, and then apically colonized with S. typhimurium for 4 h, followed by mRNA isolation and analysis as described above. All of these experiments were performed in parallel with the same pool of common control RNA derived from vehicle-treated cells. Consistent with previous data, S. typhimurium colonization of T84 model epithelia resulted in significant up-regulation of 115 genes (1.57% of the specific mRNAs assayed). Of these 115 up-regulated genes, 57 (49%) exhibited reduced (by 25% or more) induction of transcript abundance in the presence of the LXA4 analog. LXA4 analog treatment led to an increased induction (by 25% or more) of only three genes while the remaining up-regulated genes were not significantly affected by LXA4 analog. Relatively few genes (nine) exhibited decreased expression in response to S. typhimurium, and this diminution was not affected by LXA4 analog (data not shown). While conclusions regarding effects on individual specific genes require verification via other means, this technique is effective at showing the overall effect of LXA4 analog on such induced proinflammatory gene expression as is represented in Fig. 1. Many of the genes whose expression exhibited the greatest fold up-regulation in response to S. typhimurium were genes that are known to be regulated by the transcription factor NF-κB (see boldface in Table I). The induction of these NF-κB-mediated genes was attenuated by LXA4 analog (average reduction, 44%), suggesting that this eicosanoid regulates activation of this transcription factor.

Table I.

LXA4 analog attenuates activation of proinflammatory gene expressiona

Gene NameAccession No.Control Fold InductionLX Fold InductionLX/Control Induction Ratio
Small inducible cytokine subfamily A (Cys-Cys), member 20 D86955 31.8 14.2 0.45 
TNF-α-induced protein 3 M59465 23.4 17.9 0.75 
Baculoviral IAP repeat-containing 3 AI581499 22.5 9.9 0.44 
IL-1α M28983 12.3 7.8 0.63 
Superoxide dismutase 2, mitochondrial Y00472 11 6.4 0.58 
Nuclear factor of κ light gene enhancer in B-cells inhibitor M69043 10.9 7.8 0.72 
Apoptosis inhibitor 2 (baculoviral IAP repeat-contining 3) U37546 9.6 3.9 0.41 
Leukemia inhibitory factor (cholinergic differentiation factor) X13967 8.8 5.9 0.67 
Matrix metalloproteinase 7 (matrilysin, uterine) L22524 7.1 3.1 0.44 
Heat shock 70-kDa protein 1A M59828 6.3 0.59 0.09 
ESTs (chromosome 8 open reading frame 4) AA128305 6.3 7.8 1.24 
Syndecan binding protein (syntenin) AF000652 2.3 0.38 
ESTs (hypothetical protein FLJ23231AI225235 5.1 0.85 
Apoptosis inhibitor 1 (baculoviral IAP repeat-contining 2) U37547 5.9 3.3 0.56 
UDP-Gal:βGlcNac β 1,4-galactosyltransferase, polypeptide 1 X13223 5.8 2.1 0.36 
Heat shock 70-kDa protein 8 (HSC71) AL044172 5.3 0.67 0.13 
Laminin X84900 5.2 1.8 0.35 
c-fos oncogene V01512 5.2 2.9 0.56 
Protease inhibitor 3, skin-derived (SKALP) D13156 0.4 
Platelet factor 4 variant 1 M26167 4.9 3.7 0.76 
GTP-binding protein overexpressed in skeletal muscle U10550 4.7 2.5 0.53 
Phosphoprotein regulated by mitogenic pathways AJ000480 4.2 3.6 0.86 
Heat shock 70-kDa protein (HASP70B′) X51757 3.9 0.77 0.2 
Heat shock 105-kDa AB003334 3.8 0.59 0.15 
DnaJ (Hsp40) homolog, subfamily B, member 4 U40992 3.8 0.71 0.19 
Transmembrane protease, serine 2 U75329 3.8 0.79 
Prostate differentiation factor AA216685 3.8 3.4 0.89 
NK cell transcript 4 AI539055 3.7 2.5 0.68 
Proplatelet basic protein M54995 3.7 2.9 0.78 
Tissue inhibitor of metalloproteinase 3 AI245471 3.7 0.81 
IFN (α, β, ω) receptor 2 L41942 3.7 3.1 0.84 
Jagged1 (Alagille syndrome) U61276 3.6 1.7 0.47 
IFN-γR1 J03143 3.6 2.8 0.78 
Ephrin-A1 M57730 3.5 2.9 0.83 
Cytokeratin 20 X73501 3.4 1.5 0.44 
Cyclin-dependent kinase inhibitor 1A (p21, Cip1) L25610 3.4 2.1 0.62 
E74-like factor 3 (ets domain transcription factor) U66894 3.4 3.2 0.94 
ESTs (possessing ankyrin repeats) AI744478 3.3 0.91 
ESTs (dual adapter of phosphotyrosine and 3-phosphoinositides) AA149868 3.2 2.2 0.69 
Nuclear receptor subfamily 4, group A, member 1 L13740 3.2 2.3 0.72 
Down syndrome candidate region 1 U85267 3.2 3.6 1.13 
Ring finger protein Y07828 3.1 1.6 0.52 
Human Bcl-2 binding component 3 (bbc3) U82987 3.1 2.6 0.84 
Cytochrome P450, subfamily I (aromat. Comp.-inducible), polypeptide 1 K03191 3.1 3.2 1.03 
Diphtheria toxin receptor AC004634 2.4 0.8 
IFN regulatory factor 1 X14454 2.9 0.97 
SKI-like U70730 2.9 1.6 0.55 
Early growth response 1 M80583 2.9 1.9 0.66 
Low density lipoprotein receptor (familial hypercholesterolemia) L00352 2.9 2.1 0.72 
Small inducible cytokine subfamily B (Cys-X-Cys), member 10 X02530 2.8 1.5 0.54 
Stress-induced-phosphoprotein 1 (Hsp70/Hsp90-organizing protein) M86752 2.7 0.59 0.22 
Laminin, β 3 (nicein (125 kDa), kalinin (140 kDa), BM600 (125 kDa)) U17760 2.7 1.7 0.63 
Ubiquitin C AI565117 2.7 2.2 0.81 
KIAA0127 gene product D50917 2.7 2.2 0.81 
Nef-associated factor 1 AJ011895 2.7 2.5 0.93 
Heat shock 60-kDa protein 1 (chaperonin) M34664 2.6 0.71 0.27 
KIAA0291 protein (cytoplasmic linker 2) {GenBank AB006629AB006629 2.6 1.3 0.5 
Myeloid cell leukemia sequence 1 (BCL2-related) AF118124 2.6 1.5 0.58 
Chromodomain helicase DNA binding protein 4 X86691 2.6 1.6 0.62 
Growth arrest and DNA-damage-inducible, α AI634658 2.6 2.2 0.85 
CDP-diacylglycerol synthase (phosphatidate cytidylytransferase) 1 AI636223 2.5 1.7 0.68 
TNFR superfamily, member 10b AF016266 2.5 0.8 
Heat shock 90-kDa protein 1, β M16660 2.4 0.83 0.35 
KIAA0585 protein (phosphotidylserine receptor) AB011157 2.4 0.42 
   (Table continues)  
Gene NameAccession No.Control Fold InductionLX Fold InductionLX/Control Induction Ratio
Small inducible cytokine subfamily A (Cys-Cys), member 20 D86955 31.8 14.2 0.45 
TNF-α-induced protein 3 M59465 23.4 17.9 0.75 
Baculoviral IAP repeat-containing 3 AI581499 22.5 9.9 0.44 
IL-1α M28983 12.3 7.8 0.63 
Superoxide dismutase 2, mitochondrial Y00472 11 6.4 0.58 
Nuclear factor of κ light gene enhancer in B-cells inhibitor M69043 10.9 7.8 0.72 
Apoptosis inhibitor 2 (baculoviral IAP repeat-contining 3) U37546 9.6 3.9 0.41 
Leukemia inhibitory factor (cholinergic differentiation factor) X13967 8.8 5.9 0.67 
Matrix metalloproteinase 7 (matrilysin, uterine) L22524 7.1 3.1 0.44 
Heat shock 70-kDa protein 1A M59828 6.3 0.59 0.09 
ESTs (chromosome 8 open reading frame 4) AA128305 6.3 7.8 1.24 
Syndecan binding protein (syntenin) AF000652 2.3 0.38 
ESTs (hypothetical protein FLJ23231AI225235 5.1 0.85 
Apoptosis inhibitor 1 (baculoviral IAP repeat-contining 2) U37547 5.9 3.3 0.56 
UDP-Gal:βGlcNac β 1,4-galactosyltransferase, polypeptide 1 X13223 5.8 2.1 0.36 
Heat shock 70-kDa protein 8 (HSC71) AL044172 5.3 0.67 0.13 
Laminin X84900 5.2 1.8 0.35 
c-fos oncogene V01512 5.2 2.9 0.56 
Protease inhibitor 3, skin-derived (SKALP) D13156 0.4 
Platelet factor 4 variant 1 M26167 4.9 3.7 0.76 
GTP-binding protein overexpressed in skeletal muscle U10550 4.7 2.5 0.53 
Phosphoprotein regulated by mitogenic pathways AJ000480 4.2 3.6 0.86 
Heat shock 70-kDa protein (HASP70B′) X51757 3.9 0.77 0.2 
Heat shock 105-kDa AB003334 3.8 0.59 0.15 
DnaJ (Hsp40) homolog, subfamily B, member 4 U40992 3.8 0.71 0.19 
Transmembrane protease, serine 2 U75329 3.8 0.79 
Prostate differentiation factor AA216685 3.8 3.4 0.89 
NK cell transcript 4 AI539055 3.7 2.5 0.68 
Proplatelet basic protein M54995 3.7 2.9 0.78 
Tissue inhibitor of metalloproteinase 3 AI245471 3.7 0.81 
IFN (α, β, ω) receptor 2 L41942 3.7 3.1 0.84 
Jagged1 (Alagille syndrome) U61276 3.6 1.7 0.47 
IFN-γR1 J03143 3.6 2.8 0.78 
Ephrin-A1 M57730 3.5 2.9 0.83 
Cytokeratin 20 X73501 3.4 1.5 0.44 
Cyclin-dependent kinase inhibitor 1A (p21, Cip1) L25610 3.4 2.1 0.62 
E74-like factor 3 (ets domain transcription factor) U66894 3.4 3.2 0.94 
ESTs (possessing ankyrin repeats) AI744478 3.3 0.91 
ESTs (dual adapter of phosphotyrosine and 3-phosphoinositides) AA149868 3.2 2.2 0.69 
Nuclear receptor subfamily 4, group A, member 1 L13740 3.2 2.3 0.72 
Down syndrome candidate region 1 U85267 3.2 3.6 1.13 
Ring finger protein Y07828 3.1 1.6 0.52 
Human Bcl-2 binding component 3 (bbc3) U82987 3.1 2.6 0.84 
Cytochrome P450, subfamily I (aromat. Comp.-inducible), polypeptide 1 K03191 3.1 3.2 1.03 
Diphtheria toxin receptor AC004634 2.4 0.8 
IFN regulatory factor 1 X14454 2.9 0.97 
SKI-like U70730 2.9 1.6 0.55 
Early growth response 1 M80583 2.9 1.9 0.66 
Low density lipoprotein receptor (familial hypercholesterolemia) L00352 2.9 2.1 0.72 
Small inducible cytokine subfamily B (Cys-X-Cys), member 10 X02530 2.8 1.5 0.54 
Stress-induced-phosphoprotein 1 (Hsp70/Hsp90-organizing protein) M86752 2.7 0.59 0.22 
Laminin, β 3 (nicein (125 kDa), kalinin (140 kDa), BM600 (125 kDa)) U17760 2.7 1.7 0.63 
Ubiquitin C AI565117 2.7 2.2 0.81 
KIAA0127 gene product D50917 2.7 2.2 0.81 
Nef-associated factor 1 AJ011895 2.7 2.5 0.93 
Heat shock 60-kDa protein 1 (chaperonin) M34664 2.6 0.71 0.27 
KIAA0291 protein (cytoplasmic linker 2) {GenBank AB006629AB006629 2.6 1.3 0.5 
Myeloid cell leukemia sequence 1 (BCL2-related) AF118124 2.6 1.5 0.58 
Chromodomain helicase DNA binding protein 4 X86691 2.6 1.6 0.62 
Growth arrest and DNA-damage-inducible, α AI634658 2.6 2.2 0.85 
CDP-diacylglycerol synthase (phosphatidate cytidylytransferase) 1 AI636223 2.5 1.7 0.68 
TNFR superfamily, member 10b AF016266 2.5 0.8 
Heat shock 90-kDa protein 1, β M16660 2.4 0.83 0.35 
KIAA0585 protein (phosphotidylserine receptor) AB011157 2.4 0.42 
   (Table continues)  
a

Model epithelia were treated with vehicle (0.1% ethanol or 100 nM 15-(R/S)-methyl-LXA4) for 1 h, and then were colonized for 4 h with S. typhimurium at which time mRNA was isolated and analyzed via cDNA microarray hybridization. GenBank name and corresponding accession number are shown. Control fold induction and LXA4 analog induction are the fold up-regulation for each gene observed in response to S. typhimurium relative to a common control (untreated T84 cell mRNA) in the presence (LX induction) and absence (control induction) of LXA4 analog. LX/control induction ratio is the ratio of control induction to that observed in the presence of LXA4 analog. Data are in descending order of control induction beginning with the most up-regulated genes. NF-κB-regulated genes are shown in bold. Genes for heat shock proteins are italicized.

In light of both the above microarray data and our previous finding that LXA4 analogs down-regulate IL-8 secretion (5, 6) (also a NF-κB regulated gene), we next investigated whether LXA4 analogs directly attenuated epithelial cell activation of NF-κB. As polarized model epithelia do not permit direct quantitation of NF-κB promoter activity (because they are not transfectable), we used HeLa epithelial cells, which were transiently cotransfected with plasmids encoding an NF-κB-responsive reporter gene (derived from the IL-8 promoter) in the presence and absence of a plasmid encoding the LXA4 receptor. While such nonpolarized cells respond very differently to bacteria than polarized ones, both cell types respond to classic cytokine agonists such as TNF-α in a very similar manner. Thus, the effects of LXA4 analog on NF-κB-mediated transcriptional activity was determined by comparing TNF-α-induced CAT reporter activity in LXA4 analog-pretreated and untreated cells. LXA4 analog attenuated NF-κB-mediated gene expression by ∼50% (Fig. 2), consistent with the notion that attenuation of proinflammatory gene expression by LXA4 analog is mediated via signaling through this transcription factor. Such LXA4 analog attenuation of NF-κB activity was not seen in the absence of cotransfected LXA4 receptor, indicating that this receptor is specifically required for the observed anti-inflammatory bioactivity.

FIGURE 2.

LXA4 analog down-regulates epithelial activation of NF-κB. HeLa cells were transiently transfected with plasmids encoding NF-κB-CAT and empty vector or LXA4R as described in Materials and Methods, treated with 100 nM 15-(R/S)-methyl-LXA4 for 1 h, and then stimulated with TNF-α for 8 h. Cells were then lysed and CAT activity was measured via ELISA. Data are means of two separate experiments, each of which showed a similar pattern of results. ∗, Statistically significant difference (p < 0.05).

FIGURE 2.

LXA4 analog down-regulates epithelial activation of NF-κB. HeLa cells were transiently transfected with plasmids encoding NF-κB-CAT and empty vector or LXA4R as described in Materials and Methods, treated with 100 nM 15-(R/S)-methyl-LXA4 for 1 h, and then stimulated with TNF-α for 8 h. Cells were then lysed and CAT activity was measured via ELISA. Data are means of two separate experiments, each of which showed a similar pattern of results. ∗, Statistically significant difference (p < 0.05).

Close modal

We next investigated the mechanism by which LXA4 analogs attenuate NF-κB activation. NF-κB activation requires the degradation of its physically associated negative regulator IκB. Thus, we measured whether the degradation of IκBα that we have previously shown occurs in S. typhimurium-colonized or TNF-α-treated model epithelia (8) was attenuated by LXA4 analogs. Indeed, as shown in Fig. 3, LXA4 analog treatment reduced such IκB degradation induced in model epithelia by either stimulus, although the affect was more clearly visible in TNF-α-treated epithelia, likely due to the uniform kinetics of the response to this soluble agonist. This attenuation of IκBα degradation was observed in the presence of as little as 1 nM LXA4 analog, consistent with the concentration dependence of the attenuation of IL-8 secretion previously observed in the presence of LXA4 analogs (5). IκBα degradation is regulated by induced phosphorylation of serines 32 and 36. While such phospho-IκBα is normally rapidly ubiquitinated and degraded by the proteasome, it can be stabilized via inhibiting its proteolysis with pharmacologic inhibitors. Thus, using such cell-permeant inhibitors of the proteasome, we analyzed generation of phospho-IκBα induced by TNF-α in the presence of a range of concentrations of 15-(R/S) methyl-LXA4. In contrast to its inhibition of IκBα degradation, we did not observe a reduction in IκBα phosphorylation (phospho-IκBα can be distinguished from IκBα by its higher m.w.) in the presence of any of the tested concentrations of LXA4 analogs. LXA4 analogs had similar effects on IκBα phosphorylation and degradation induced by S. typhimurium or when assayed using a phospho-specific IκBα Ab (data not shown). Finally, we examined Ca2+ mobilization, one of the early signaling events that leads to IκBα phosphorylation in S. typhimurium-colonized epithelia (11). S. typhimurium-induced Ca2+ mobilization was not affected by a LXA4 analog (Fig. 4). Together, these results suggest that LXA4 analogs act subsequent to initial Ca2+ signal and resulting IκBα phosphorylation in the reduction of the IκBα degradation that attenuates proinflammatory gene activation.

FIGURE 3.

LXA4 analog reduces degradation, but not phosphorylation, of IκBα. Model epithelia were treated with vehicle (0.1% ethanol or 15-(R/S)-methyl-LXA4) for 1 h and stimulated, and whole cell lysates were assayed for IκBα levels via immunoblotting. A, Epithelia were colonized by 109 CFU S. typhimurium for 1 h in the presence or absence of 100 nM 15-(R/S)-methyl-LXA4.B, Epithelia were treated with vehicle or indicated concentration of 15-(R/S)-methyl-LXA4 in the presence or absence of the proteasome inihibitor MG-132 (50 μM), and then stimulated with TNF-α (10 ng/ml) for 45 min. Whole cell lysates were analyzed via SDS-PAGE immunoblotting using anti-IκBα. Note that phospho-IκBα is visible only upon proteasomal inhibition with MG-132. Data are individual experiments and are representative of three separate experiments.

FIGURE 3.

LXA4 analog reduces degradation, but not phosphorylation, of IκBα. Model epithelia were treated with vehicle (0.1% ethanol or 15-(R/S)-methyl-LXA4) for 1 h and stimulated, and whole cell lysates were assayed for IκBα levels via immunoblotting. A, Epithelia were colonized by 109 CFU S. typhimurium for 1 h in the presence or absence of 100 nM 15-(R/S)-methyl-LXA4.B, Epithelia were treated with vehicle or indicated concentration of 15-(R/S)-methyl-LXA4 in the presence or absence of the proteasome inihibitor MG-132 (50 μM), and then stimulated with TNF-α (10 ng/ml) for 45 min. Whole cell lysates were analyzed via SDS-PAGE immunoblotting using anti-IκBα. Note that phospho-IκBα is visible only upon proteasomal inhibition with MG-132. Data are individual experiments and are representative of three separate experiments.

Close modal
FIGURE 4.

LXA4 analog did not affect Ca2+ mobilization in response to S. typhimurium. Model epithelia were loaded with the Ca2+ indicator fura 2 (5 μM for 1 h) and intracellular Ca2+ measured by spectrofluorometry. A, Monitor of baseline Ca2+ in untreated epithelia. B and C, Fura 2-loaded epithelia were treated for 1 h with vehicle (B) (0.1% ethanol) or 100 nM 15-(R/S)-methyl-LXA4 (C) and then apically treated with 109 CFU S. typhimurium during fluorometric measurement (addition of bacteria is indicated by arrow). Data are individual experiments and are representative of three separate experiments.

FIGURE 4.

LXA4 analog did not affect Ca2+ mobilization in response to S. typhimurium. Model epithelia were loaded with the Ca2+ indicator fura 2 (5 μM for 1 h) and intracellular Ca2+ measured by spectrofluorometry. A, Monitor of baseline Ca2+ in untreated epithelia. B and C, Fura 2-loaded epithelia were treated for 1 h with vehicle (B) (0.1% ethanol) or 100 nM 15-(R/S)-methyl-LXA4 (C) and then apically treated with 109 CFU S. typhimurium during fluorometric measurement (addition of bacteria is indicated by arrow). Data are individual experiments and are representative of three separate experiments.

Close modal

Chronic inflammatory diseases of the intestine such as Crohn’s disease and ulcerative colitis (i.e., inflammatory bowel disease (IBD)) are associated with, and possibly mediated by, increased levels of proinflammatory cytokines, many of which are NF-κB regulated, in the intestinal mucosa (13). Activated NF-κB is detectable in biopsies of IBD patients (14), and therapeutic agents that are effective in IBD are known to act, at least in part, through inhibition of NF-κB activation. Because LXA4 analogs attenuated NF-κB-mediated gene expression in vitro, we next asked whether LXA4 analogs might be therapeutic for intestinal inflammation in vivo. Due to the lack of established murine models of infectious gastroenteritis (mice get systemic illness rather than intestinal inflammation from S. typhimurium), we used a well-established chemically induced murine colitis model. Specifically, we examined whether an orally administered LXA4 analog affected the colitis induced by DSS by measuring the clinical parameters that are the defined disease indicators in this widely used colitis model (Fig. 5). Colitis was induced in groups of five 8-wk-old mice via the addition of 4% DSS to their drinking water for 7 days. Simultaneous to the DSS administration, mice were also administered, via their drinking water, vehicle (0.05% ethanol) or 10 μg/ml 15-epi-16-para-fuoro-phenoxy-LXA4, an analog of LXA4 that has been shown to have local and systemic in vivo anti-inflammatory bioactivity (15). HPLC analysis of the LXA4 analog recovered from the drinking water solutions demonstrated the structural integrity of the tetraene chromophore with only ester hydrolysis byproducts seen. Daily measurements of weight, occult blood, gross bleeding, and average (per group) water consumption were made. Because each mouse drank ∼1 ml per day, the approximate daily dose of ingested compound was 10 μg per mouse.

FIGURE 5.

Oral administration of LXA4 analog attenuates DSS-induced colitis. BALB/c mice were given 4% DSS in their drinking water along with vehicle or 10 μg/ml 15-epi-16-parafluoro-phenoxy-LXA4. Body mass and bleeding (gross and occult, when stools could be obtained) were checked daily. A and B, Results (±SEM for body mass) of an experiment using six mice for each condition. The pattern is similar to two additional experiments using five mice per condition. C, Pooled results from the three replicated experiments. ∗, Statistically significant differences (p < 0.05).

FIGURE 5.

Oral administration of LXA4 analog attenuates DSS-induced colitis. BALB/c mice were given 4% DSS in their drinking water along with vehicle or 10 μg/ml 15-epi-16-parafluoro-phenoxy-LXA4. Body mass and bleeding (gross and occult, when stools could be obtained) were checked daily. A and B, Results (±SEM for body mass) of an experiment using six mice for each condition. The pattern is similar to two additional experiments using five mice per condition. C, Pooled results from the three replicated experiments. ∗, Statistically significant differences (p < 0.05).

Close modal

DSS/vehicle-treated mice began to lose weight ∼3–5 days (variance in different experiments) after the treatment began and continued to lose weight until ∼3 days after the DSS treatment was stopped. Mice that did not die during this period then stopped intestinal bleeding (as assessed by gross observation and occult blood assay) and recovered their original weights over the next several days. The DSS/LXA4 analog-treated mice lost weight with similar kinetics but on average lost significantly less weight than the DSS/vehicle-treated controls (Fig. 5). LXA4 analog-treated mice also exhibited a trend toward less intestinal bleeding than their matched controls, although the difference did not quite reach statistical significance (p = 0.062 by Wilcox signed-rank test). These differences did not result from differing levels of DSS or water consumption, as water consumption was indistinguishable between the two groups. Nor were these differences likely the result of the LXA4 analog reducing epithelial exposure to DSS via affecting epithelial chloride secretion, as LXA4 analogs have been shown not to affect this secretory pathway (6). Perhaps most importantly, DSS/vehicle-treated mice had a mortality rate of 33% (n = 15) while, in contrast, we did not observe any mortality among the DSS/LXA4 analog-treated mice (n = 15). About 3 days after DSS treatment was suspended, both DSS/vehicle- and DSS/LXA4 analog-treated mice stopped intestinal bleeding and began to recover their body weight. Excluding the DSS/vehicle-treated mice that died, the recovery rates did not differ significantly between mice receiving LXA4 analogs vs their matched controls (data not shown). In the absence of DSS, mice treated with vehicle or LXA4 analog exhibited indistinguishable slow weight gains of ∼0.5 gm per week and did not have any detectable occult (or gross) bleeding. Together, these results indicate LXA4 analogs can reduce the development of DSS colitis disease activity and its consequences on global physiological parameters of wellness, such as body weight.

Chronic inflammatory diseases of the intestine (e.g., Crohn’s disease and ulcerative colitis, collectively referred to as IBD) are a serious public health problem, particularly in the developed world. While chronic inflammatory events play a definitive role in these diseases, clinically significant flares are often associated with augmentation of the innate immune response, specifically neutrophil flux across the epithelial surface. Although the underlying causes of this response are generally unknown, proinflammatory cytokines, including those of epithelial origin, are thought to be essential mediators of this process (16). As shown in this work, stable analogs of the endogenous anti-inflammatory eicosanoid LXA4 can attenuate epithelial proinflammatory gene expression. LXA4 analog attenuation of proinflammatory gene expression was mediated through a central proinflammatory signaling pathway and lessened the disease activity in a well-established mouse model of colitis, thus suggesting that these compounds could be therapeutic for the active phase of IBD.

Acute flares of IBD resemble infectious colitis (e.g., salmonellosis) both histologically (characterized by massive neutrophil influx and transepithelial migration) and clinically (i.e., diarrhea, cramping). The clinical manifestations of salmonellosis are thought to be largely attributable to the mucosal innate (neutrophil-mediated) immune response to this organism, leading to the suggestion that IBD flares may result from an aberrant mucosal innate immune response to normally nonpathogenic gut flora (perhaps the end result of signals originating from cells of specific immunity) (1). Thus, it is encouraging that LXA4 analogs broadly attenuated the changes in epithelial gene expression induced by S. typhimurium. Interestingly, LXA4 analogs did not uniformly attenuate all such changes in gene expression but rather diminished some nearly completely while other induced changes in gene expression were unaffected. Categorization of many of these genes as well as their possible roles in inflammation are not yet well defined. One specific class of genes uniformly down-regulated by LXA4 analogs were the heat shock proteins (HSP); the up-regulation of the HSP genes induced by S. typhimurium was completely reversed in the presence of LXA4 analogs. Such regulation of HSP genes may play a role in LXA4 analog bioactivity or may simply reflect the general reduced stress level of LXA4 analog-treated epithelial cells. Consistent with the latter possibility, another family of stress-induced genes, the NF-κB-dependent genes, was also clearly down-regulated. As these genes are known to encode mediators of mucosal inflammation, this LXA4 analog bioactivity likely plays a role in the observed therapeutic effects of this eicosanoid on DSS colitis in vivo.

Several nonsteroidal anti-inflammatory drugs (NSAID) including aspirin have also been shown to attenuate activation of the proinflammatory transcription factor NF-κB (17). However, LXA4 analog attenuation of NF-κB differs from that of such agents in several important ways. LXA4 analogs act via a specific receptor, whereas NSAID primarily directly inhibit proinflammatory enzymes. While both LXA4 analogs and NSAID reduce IκBα degradation, LXA4 analogs did so at nanomolar concentrations, while NSAID require much higher concentrations for this activity. In vitro studies indicate that the mechanism by which LXA4 analog attenuated IκBα degradation also differs significantly from that of NSAID, in that LXA4 analogs did not reduce the phosphorylation of IκBα, while NSAID, like other pharmacological attenuators of NF-κB, reduce IκBα degradation by preventing this phosphorylation event (18). However, because LXA4 analogs are structurally similar to 15-epi-LXA4, which is biosynthesized by cyclooxygenase that has been acetylated by aspirin (3), some of aspirin’s attenuation of proinflammatory gene expression may yet result via this route. Mechanistically, LXA4 analog attenuation of NF-κB more closely resembles that of non-proinflammatory bacteria that also block IκBα degradation but not phosphorylation (19). Such phosphorylation-independent regulation of IκBα may thus be common to agonists that activate endogenous anti-inflammatory pathways and may hold more promise to have fewer unwanted effects than global inhibitors of IκBα kinase.

LXA4 analogs have in vivo anti-inflammatory activity when applied both locally (topically in the mouse ear) and systemically via tail vein (15, 20). Oral administration is somewhat equivalent to topical application in the gut, as it provides direct delivery of LXA4 analog to the intestinal epithelium. However, this LXA4 analog is rapidly absorbed following oral gavage in rodents with ∼17% oral availability (B. Subramanyam, W. Guilford, J. Bauman, and J. Parkinson, unpublished observations) and thus may act systemically. While LXA4 analog is short-lived in plasma (t1/2 <30 min for i.v.; B. Subramanyam, W. Guilford, J. Bauman, and J. Parkinson, unpublished observations) placement in the drinking water provides semicontinuous systemic delivery. A particularly likely target are neutrophils, which are known to be targets of LXA4 and major immune mediators of this colitis model. LXA4 analogs attenuate neutrophil chemotaxis (21), oxidative burst (22), and the release of granule proteases (23). Delivery of LXA4 analogs via tail vein showed a similar trend as oral administration on DSS colitis but did not differ statistically significantly from control (DSS-treated) mice (data not shown). This could have resulted from a less-targeted delivery of compound to the inflammatory site or may have occurred because delivery in the drinking water maintained a continued presence of the LXA4 analog while the injected compound is rapidly cleared from the circulation (24).

While it is important to sort out the precise cellular mechanism of LXA4 analog in vivo bioactivity, regardless, oral administration seems to be an effective way for this compound to down-regulate intestinal inflammation. In contrast, NSAID, including cyclooxygenase-2-specific inhibitors, tend to damage the intestinal mucosa, resulting in causation or exacerbation of intestinal inflammation, while mAb-based drugs (e.g., imflixamab), although therapeutic, must be given i.v. Thus, this novel strategy of activating endogenous anti-inflammatory pathways with stable analogs of LXA4 could be developed into an effective means of treating human colitis.

Table 1A.

Continued

Gene NameAccession No.Control Fold InductionLX Fold InductionLX/Control Induction Ratio
ESTs (GenBank N93892) N93892 2.4 1.1 0.46 
Cytochrome P450, subfamily IIIA, polypeptide 7 D00408 2.4 1.4 0.58 
ESTs (type I transmembrane protein Fn14) AI827127 2.4 1.7 0.71 
Potassium channel, subfamily K, member 1 (TWIK-1) U90065 2.4 1.9 0.79 
Territin, heavy polypeptide 1 AA102267 2.4 2.7 1.13 
Plasminogen activator, urokinase receptor AC0076953 2.4 2.9 1.21 
Tubulin, β5 X00734 2.3 1.5 0.65 
Glucosamine-6-phosphate deaminase L40636 2.3 1.9 0.83 
Ephrin-B1 U09304 2.3 2.1 0.91 
Small inducible cytokine D (Cys-X3-Cys), (fractalkine, neurotactin) AC004382 2.3 2.2 0.96 
EH domain containing 1 AF001434 2.3 2.4 1.04 
Pim-1 oncogene M54915 2.3 2.7 1.17 
Related to t-complex 1/acetyl-Coenzyme A acetyltransferase 2 X52882 2.2 0.83 0.38 
Early growth response 3 X63741 2.2 0.91 0.41 
Splicing factor proline/glutamine rich X70944 2.2 1.2 0.55 
Cdc42 effector protein 2 AF001436 2.2 1.2 0.55 
Ubiquitin B U49869 2.2 1.7 0.77 
TNF (TNF superfamily, member 2) M10988 2.2 1.8 0.82 
TGFα X70340 2.2 1.8 0.82 
Ladinin U42408 2.2 2.1 0.95 
Epiregulin D30783 2.2 4.2 1.91 
Tubulin, β polypeptide X79535 2.1 1.2 0.57 
ESTs (sirtuin 1) AI378978 2.1 1.2 0.57 
CDC28 protein kinase 2 X54942 2.1 1.2 0.57 
Putative translation initiation factor (SUI 1) AI832315 2.1 1.3 0.62 
Protein phosphatase 1, regulatory subuint 10 Y13247 2.1 1.4 0.67 
Early growth response 2 (Krox-20 (Drosophila) homolog) J04076 2.1 1.4 0.67 
Heat shock 70-kDa protein 5 (glucose-regulated protein, 78 kDa) AI878886 2.1 1.7 0.81 
Dual specificity phosphatase 5 U15932 2.1 1.7 0.81 
ESTs (transcriptional cofactor with PDZ binding motif (TAZ)) AL050107 2.1 1.9 0.9 
Epithelial membrane protein 1 U77085 2.1 0.95 
Actin, α2, smooth muscle, aorta AL048044 2.1 2.5 1.19 
Lipocalin 2 (oncogene 24p3) X99133 2.1 1.43 
FK506-binding protein 4 (59kD) M88279 0.63 0.31 
Decay accelerating factor for complement (CD55) AF052110 0.5 
ESTs (solute carrier family 7) N35555 1.1 0.55 
Chromodomain helicase DNA binding protein 2 AF006514 1.1 0.55 
Nucleoside phosphorylase AA311617 1.3 0.65 
EphA2 AA612998 1.3 0.65 
ESTs (hypothetical protein MGC11034AA775792 1.4 0.7 
B-cell translocation gene 1, anti-proliferative AI560266 1.4 0.7 
IL-2Rγ (severe combined immunodeficiency) D11086 1.5 0.75 
Carcinoembryonic Ag gene family member 6 X52378 1.5 0.75 
Carbonic anhydrase II J03037 1.5 0.75 
Mitogen-activated protein kinase kinase kinase 8 D14497 1.6 0.8 
GTP cyclohydrolase 1 (dopa-responsive dystonia) S44053 1.7 0.85 
6-Phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 AA463459 1.7 0.85 
Forkhead (Drosophila)-like 8 U59831 2.2 1.1 
Incyte EST (Incyte PD:58522)  1.9 2.1 1.11 
Ornithine decarboxylase 1 M81740 1.8 2.8 1.56 
Gene NameAccession No.Control Fold InductionLX Fold InductionLX/Control Induction Ratio
ESTs (GenBank N93892) N93892 2.4 1.1 0.46 
Cytochrome P450, subfamily IIIA, polypeptide 7 D00408 2.4 1.4 0.58 
ESTs (type I transmembrane protein Fn14) AI827127 2.4 1.7 0.71 
Potassium channel, subfamily K, member 1 (TWIK-1) U90065 2.4 1.9 0.79 
Territin, heavy polypeptide 1 AA102267 2.4 2.7 1.13 
Plasminogen activator, urokinase receptor AC0076953 2.4 2.9 1.21 
Tubulin, β5 X00734 2.3 1.5 0.65 
Glucosamine-6-phosphate deaminase L40636 2.3 1.9 0.83 
Ephrin-B1 U09304 2.3 2.1 0.91 
Small inducible cytokine D (Cys-X3-Cys), (fractalkine, neurotactin) AC004382 2.3 2.2 0.96 
EH domain containing 1 AF001434 2.3 2.4 1.04 
Pim-1 oncogene M54915 2.3 2.7 1.17 
Related to t-complex 1/acetyl-Coenzyme A acetyltransferase 2 X52882 2.2 0.83 0.38 
Early growth response 3 X63741 2.2 0.91 0.41 
Splicing factor proline/glutamine rich X70944 2.2 1.2 0.55 
Cdc42 effector protein 2 AF001436 2.2 1.2 0.55 
Ubiquitin B U49869 2.2 1.7 0.77 
TNF (TNF superfamily, member 2) M10988 2.2 1.8 0.82 
TGFα X70340 2.2 1.8 0.82 
Ladinin U42408 2.2 2.1 0.95 
Epiregulin D30783 2.2 4.2 1.91 
Tubulin, β polypeptide X79535 2.1 1.2 0.57 
ESTs (sirtuin 1) AI378978 2.1 1.2 0.57 
CDC28 protein kinase 2 X54942 2.1 1.2 0.57 
Putative translation initiation factor (SUI 1) AI832315 2.1 1.3 0.62 
Protein phosphatase 1, regulatory subuint 10 Y13247 2.1 1.4 0.67 
Early growth response 2 (Krox-20 (Drosophila) homolog) J04076 2.1 1.4 0.67 
Heat shock 70-kDa protein 5 (glucose-regulated protein, 78 kDa) AI878886 2.1 1.7 0.81 
Dual specificity phosphatase 5 U15932 2.1 1.7 0.81 
ESTs (transcriptional cofactor with PDZ binding motif (TAZ)) AL050107 2.1 1.9 0.9 
Epithelial membrane protein 1 U77085 2.1 0.95 
Actin, α2, smooth muscle, aorta AL048044 2.1 2.5 1.19 
Lipocalin 2 (oncogene 24p3) X99133 2.1 1.43 
FK506-binding protein 4 (59kD) M88279 0.63 0.31 
Decay accelerating factor for complement (CD55) AF052110 0.5 
ESTs (solute carrier family 7) N35555 1.1 0.55 
Chromodomain helicase DNA binding protein 2 AF006514 1.1 0.55 
Nucleoside phosphorylase AA311617 1.3 0.65 
EphA2 AA612998 1.3 0.65 
ESTs (hypothetical protein MGC11034AA775792 1.4 0.7 
B-cell translocation gene 1, anti-proliferative AI560266 1.4 0.7 
IL-2Rγ (severe combined immunodeficiency) D11086 1.5 0.75 
Carcinoembryonic Ag gene family member 6 X52378 1.5 0.75 
Carbonic anhydrase II J03037 1.5 0.75 
Mitogen-activated protein kinase kinase kinase 8 D14497 1.6 0.8 
GTP cyclohydrolase 1 (dopa-responsive dystonia) S44053 1.7 0.85 
6-Phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 AA463459 1.7 0.85 
Forkhead (Drosophila)-like 8 U59831 2.2 1.1 
Incyte EST (Incyte PD:58522)  1.9 2.1 1.11 
Ornithine decarboxylase 1 M81740 1.8 2.8 1.56 

We thank John Wallace of University of Calgary (Calgary, Canada) for help with preliminary experiments, Dr. Babu Subramanyam (Berlex Biosciences) for information on the pharmacokinetic absorption profile of the lipoxin analog in rodents, and Sean Lyons for outstanding technical support.

1

This work was supported by National Institutes of Health Grants DK-02792 (to A.T.G.), AR-44268 (to I.R.W.), AI-49741 (to A.S.N.), DK-35932 (to J.L.M.), and DK-47662 (to J.L.M.).

3

Abbreviations used in this paper: LXA4, lipoxin A4, CAT, chloramphenicol acetyl transferase; DSS, dextran sodium sulfate; IBD, inflammatory bowel disease; HSP, heat shock protein; NSAID, nonsteroidal anti-inflammatory drug.

1
McCormick, B., A. Gewirtz, J. L. Madara.
1998
. Epithelial crosstalk with bacteria and immune cells.
Curr. Opin. Gastroenterol.
14
:
492
2
Serhan, C. N., T. Takano, J. F. Maddox.
1999
. Aspirin-triggered 15-epi-lipoxin A4 and stable analogs on lipoxin A4 are potent inhibitors of acute inflammation: receptors and pathways.
Adv. Exp. Med. Biol.
447
:
133
3
Claria, J., C. N. Serhan.
1995
. Aspirin triggers previously undescribed bioactive eicosanoids by human endothelial cell-leukocyte interactions.
Proc. Natl. Acad. Sci. USA
92
:
9475
4
Serhan, C. N., J. F. Maddox, N. A. Petasis, I. Akritopoulou-Zanze, A. Papayianni, H. R. Brady, S. P. Colgan, J. L. Madara.
1995
. Design of lipoxin A4 stable analogs that block transmigration and adhesion of human neutrophils.
Biochemistry
34
:
14609
5
Gewirtz, A. T., B. McCormick, A. S. Neish, N. A. Petasis, K. Gronert, C. N. Serhan, J. L. Madara.
1998
. Pathogen-induced chemokine secretion from model intestinal epithelium is inhibited by lipoxin A4 analogs.
J. Clin. Invest.
101
:
1860
6
Gronert, K. G., A. T. Gewirtz, J. L. Madara, C. N. Serhan.
1998
. Identification of a human enterocyte lipoxin A4 receptor that is regulated by IL-13 and IFN-γ that inhibits TNF-α-induced IL-8 release.
J. Exp. Med.
187
:
1285
7
Goh, J., A. W. Baird, C. O’Keane, R. W. Watson, D. Cottell, G. Bernasconi, N. A. Petasis, C. Godson, H. R. Brady, P. MacMathuna.
2001
. Lipoxin A4 and aspirin-triggered 15-epi-lipoxin A4 antagonize TNF-α-stimulated neutrophil-enterocyte interactions in vitro and attenuate TNF-α-induced chemokine release and colonocyte apoptosis in human intestinal mucosa ex vivo.
J. Immunol.
167
:
2772
8
Madara, J. L., S. P. Colgan, A. Nusrat, C. Delp, C. A. Parkos.
1992
. A simple approach to measurement of electrical parameters of cultured epithelial monolayers: use in assessing neutrophil epithelial interactions.
J. Tissue Cult. Methods
15
:
209
9
Young, A. N., M. B. Amin, C. S. Moreno, S. D. Lim, C. Cohen, J. A. Petros, F. F. Marshall, A. S. Neish.
2001
. Expression profiling of renal epithelial neoplasms: a method for tumor classification and discovery of diagnostic molecular markers.
Am. J. Pathol.
158
:
1639
10
Kunsch, C., R. K. Lang, C. A. Rosen, M. F. Shannon.
1994
. Synergistic transcriptional activation of the IL-8 gene by NF-κB p65 (RelA) and NF-IL-6.
J. Immunol.
153
:
153
11
Gewirtz, A. T., A. S. Rao, P. O. Simon, Jr, D. Merlin, D. Carnes, J. L. Madara, A. S. Neish.
2000
. Salmonella typhimurium induces epithelial IL-8 expression via Ca2+- mediated activation of the NF-κB pathway.
J. Clin. Invest.
105
:
79
12
Cooper, H. S., S. N. Murthy, R. S. Shah, D. J. Sedergran.
1993
. Clinicopathologic study of dextran sulfate sodium experimental murine colitis.
Lab. Invest.
69
:
238
13
Neurath, M. F., I. Fuss, G. Schurmann, S. Pettersson, K. Arnold, H. Muller-Lobeck, W. Strober, C. Herfarth, K. H. Buschenfelde.
1998
. Cytokine gene transcription by NF-κB family members in patients with inflammatory bowel disease.
Ann. NY Acad. Sci.
859
:
149
14
Schreiber, S., S. Nikolaus, J. Hampe.
1998
. Activation of nuclear factor κB inflammatory bowel disease.
Gut
42
:
477
15
Clish, C. B., K. Gronert, C. N. Serhan.
2000
. Local and systemic delivery of an aspirin-triggered lipoxin stable analog inhibits neutrophil trafficking.
Ann. NY Acad. Sci.
905
:
274
16
Papadakis, K. A., S. R. Targan.
2000
. Role of cytokines in the pathogenesis of inflammatory bowel disease.
Annu. Rev. Med.
51
:
289
17
Kopp, E., S. Ghosh.
1994
. Inhibition of NF-κB by sodium salicylate and aspirin.
Science
265
:
956
18
Yin, M. J., Y. Yamamoto, R. B. Gaynor.
1998
. The anti-inflammatory agents aspirin and salicylate inhibit the activity of IκB kinase-β.
Nature
396
:
77
19
Neish, A. S., A. T. Gewirtz, H. Zeng, A. N. Young, M. E. Hobert, V. Karmali, A. S. Rao, J. L. Madara.
2000
. Prokaryotic regulation of epithelial responses by inhibition of IκB-α ubiquitination.
Science
289
:
1560
20
Takano, T., S. Fiore, J. F. Maddox, H. R. Brady, N. A. Petasis, C. N. Serhan.
1997
. Aspirin-triggered 15-epi-lipoxin A4 and LXA4 stable analogs are potent inhibitors of acute inflammation: evidence for anti-inflammatory receptors.
J. Exp. Med.
185
:
1693
21
Lee, T. H., C. E. Horton, U. Kyan-Aung, D. Haskard, A. E. Crea, B. W. Spur.
1989
. Lipoxin A4 and lipoxin B4 inhibit chemotactic responses of human neutrophils stimulated by leukotriene B4 and N-formyl-l-methionyl-l-leucyl-l-phenylalanine.
Clin. Sci.
77
:
195
22
Levy, B. D., V. V. Fokin, J. M. Clark, M. J. Wakelam, N. A. Petasis, C. N. Serhan.
1999
. Polyisoprenyl phosphate (PIPP) signaling regulates phospholipase D activity: a “stop” signaling switch for aspirin-triggered lipoxin A4.
FASEB J.
13
:
903
23
Gewirtz, A. T., V. V. Fokin, N. A. Petasis, C. N. Serhan, J. L. Madara.
1999
. LXA4, aspirin-triggered 15-epi LXA4, and their stable analogs selectively down-regulate PMN azurophilic degranulation.
Am J. Physiol.
276
:
C988
24
Clish, C. B., J. A. O’Brien, K. Gronert, G. L. Stahl, N. A. Petasis, C. N. Serhan.
1999
. Local and systemic delivery of a stable aspirin-triggered lipoxin prevents neutrophil recruitment in vivo.
Proc. Natl. Acad. Sci. USA
96
:
8247