A β-oxidation-resistant lipoxin A₄ analog treats hapten-induced colitis by attenuating inflammation and immune dysfunction

Abstract

Lipoxins and aspirin-triggered 15-epi-lipoxins (ATL) are counter-regulatory eicosanoids with potent antiinflammatory actions. Oral efficacy and mechanism of action of ZK-192, a β-oxidation-resistant 3-oxa-ATL analog, were examined in trinitrobenzenesulphonate (TNBS)-induced colitis. When dosed orally once daily, 300 and 1,000 μg/kg ZK-192 markedly attenuated TNBS colitis in rodents both in preventive and therapeutic regimens. ZK-192 attenuated weight loss, macroscopic and histologic colon injury, mucosal neutrophil infiltration, and colon wall thickening. ZK-192 was as effective as 3–10 mg/kg oral prednisolone. ZK-192 decreased mucosal mRNA levels for several inflammatory mediators: inducible nitric oxide synthase, cyclooxygenase 2, and macrophage inflammatory protein 2. ZK-192 also decreased mucosal mRNA and protein levels of T helper 1 effector cytokines: tumor necrosis factor α, IL-2, and IFN-γ. Systemic levels of these cytokines were also dramatically attenuated. CD3/CD28-mediated costimulation of T helper 1 effector cytokine release in lamina propria mononuclear cells was markedly inhibited by ZK-192 ex vivo and in vitro. ZK-192 also prevented colitis in lymphocyte-deficient severe combined immunodeficient mice, with ≈75% inhibition of mucosal tumor necrosis factor α and IL-2 levels. The results are further evidence that innate immune cells function as triggers for hapten-induced colitis. The combined antiinflammatory and immunomodulatory effects of ZK-192 in TNBS colitis suggest that ATL analogs may be an attractive oral treatment approach for inflammatory bowel diseases.

Crohn's disease is a relapsing disease characterized by mucosal T lymphocyte dysfunction, abnormal cytokine production, and inflammation leading to damage of the intestinal lining (1). Crohn's disease etiology remains unclear, but evidence supports a failure to attenuate immune reactivity to endogenous mucosal antigens (1). IFN-γ- and IL-2-producing type 1 helper cell (Th₁) lymphocytes dominate the mucosa in Crohn's disease patients. Central immune cell activation leads to production of an array of nonspecific inflammatory mediators, including cytokines, chemokines, and growth factors, and lipid mediators generated by arachidonic acid metabolism, including prostaglandins and leukotrienes. These amplify the local immune response and promote tissue destruction (2). Several animal models of Crohn's disease have been developed (3). Hapten-induced colitis, in which trinitrobenzene sulfonic acid (TNBS) is delivered intrarectally to rodents, displays human Crohn's disease-like features, notably predominant Th₁ activity of mucosal CD4⁺ T cells and transmural mononuclear cell-driven inflammation (4, 5). Colon inflammation develops as a result of reactivity toward TNBS-modified self-antigens involving aspects of innate and adaptive immunity (6).

Lipoxin A₄ (LX), generated during inflammation by lipoxygenase-dependent arachidonic acid metabolism, prevents acute inflammation by means of potent agonist actions at the G protein-coupled receptor ALX-R/fprl1 (7). Cyclooxygenase (COX) 2 acetylation by aspirin triggers formation of 15-epi-lipoxin (aspirin-triggered LX, ATL) that retains the antiinflammatory actions of LX and mediates, in part, aspirin's therapeutic effect (8). LX/ATL actions in counteracting inflammation are well established, but short half-life and physicochemical properties limit their therapeutic potential. LX/ATL are rapidly inactivated by prostaglandin dehydrogenase, forming metabolites with reduced affinity for ALX-R/fprl1 and decreased antiinflammatory potency (9). Recently, β-oxidation at C2–C3 was identified as a route for ATL analog metabolism in vivo, prompting design of β-oxidation-resistant 3-oxa-ATL analogs, including a 3-oxa-trienyne analog with enhanced chemical stability and oral pharmacokinetics (ZK-192, Fig. 1), which had topical antiinflammatory activity in skin (10). Here, we describe the oral efficacy profile and mechanism of action of ZK-192 in rodent TNBS colitis. Orally administered ZK-192 potently attenuated TNBS colitis. Mechanistic studies of mucosal inflammatory and immune mediators, lamina propria mononuclear cell (LPMC) effector functions, and severe combined immunodeficient (SCID) mice revealed previously undescribed actions for LX/ATL in modulating inflammation and dysfunctional immunity in a severe mucosal pathology.

Fig. 1.

3-oxa-ATL analog ZK-192 prevents and treats TNBS colitis. (a) Structure of 3-oxa-ATL analog ZK-192. (b) TNBS (2.5 mg per colon) induces severe colitis in mice, with ≈40% survival by day 7 (filled squares). Daily oral treatment with 1,000 μg/kg ZK-192 (filled circles, 80% survival) was as effective as 10 mg/kg prednisolone (open circles, 75% survival) in preventing mortality. *, P < 0.01 drug-treated vs. TNBS. n = 10 per group. (c) Established colitis induced by 1.5 mg of TNBS per colon in mice is treated by 300 and 1,000 μg/kg ZK-192 or 10 mg/kg prednisolone given orally once daily from days 3 to 14. Microscopic colon injury and mucosal neutrophil infiltration (MPO activity) were attenuated in all treated groups. Both doses of ZK-192 were equivalent to prednisolone. Data represent mean ± SEM, n = 6–8. *, P < 0.01 TNBS vs. control; Ψ, P < 0.01 drug-treated vs. TNBS.

Experimental Procedures

Synthesis of 3-oxa ATL Analog ZK-192. 3-oxa-ATL analog (5R,6R, 7E,9E,13E,15S)-16-(4-fluorophenoxy)-3-oxa-5,6,15-trihydroxy-7,9,13-hexadecatrien-11-ynoic acid, sodium salt (ZK-192, Fig. 1) was prepared as described (10).

Animals. Six- to 8-week-old female BALB/c and SCID mice (20–25 g) and male Wistar rats (175–200 g) were from Charles River Breeding Laboratories. Animal committees at the University of Perugia and the University of Calgary approved all protocols according to governmental guidelines.

Induction of Colitis and Study Design. Colitis was induced in mice and rats as described (11, 12). Fasted mice were anesthetized. and 1–2.5 mg of TNBS (Sigma) in 0.1 ml of 50% ethanol was administered into the colon lumen by catheter; control mice received 50% ethanol alone. Rats were anesthetized, and TNBS (60 mg/ml in 0.5 ml of 50% ethanol) was administered into the distal colon by cannula. Animals were monitored daily for loss of body weight and survival. At the end of the studies, surviving animals were killed, blood samples were collected by cardiac puncture, and colons were excised, weighed, and evaluated for macroscopic damage. Tissue segments from mice were used for LPMC isolation, immunohistochemical studies, cytokines, and myeloperoxidase (MPO) activity. SCID mice were treated by intracolonic injection of 2 mg of TNBS.

Vehicle (pyrogen-free water), ZK-192 (doses of 30–1,000 μg/kg in water), and prednisolone (Sigma, 3–10 mg/kg in PBS) were administered by oral gavage once daily (final volume, 100 μl). For preventive treatment, drugs were given the same day as TNBS injection and continued for 7 or 14 days. For therapeutic treatment, drugs were given to animals with established colitis starting 3 days after TNBS. Drinking water experiments were performed with 1–10 μg/ml ZK-192 provided ad libitum for 7 days, starting the same day as TNBS. Heparinized blood taken from rats (two to three per cohort) on day 14 at 1, 2, 4, and 6 h after the final dose was used to prepare frozen plasma, and ZK-192 levels were analyzed by HPLC and liquid chromatography MS methods (10).

Macroscopic and Histologic Grading of Colitis. Colons were examined with a dissecting microscope (magnification, ×5) and graded for macroscopic lesions on a scale from 0 to 10 based on criteria for inflammation, such as hyperemia, thickening of the bowel, and the extent of ulceration (4). For histology scores (5), colon specimens located 2 cm above the anal canal were fixed in 10% buffered formalin phosphate, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. Inflammation was graded semiquantitatively by the use of a BX60 microscope (Olympus, Tokyo) by a certified pathologist and was scored in a blinded fashion (0, no inflammation; 1, very low level of inflammation; 2, low level of leukocyte infiltration; 3, high levels of leukocyte infiltration and vascular density, colon wall thickening; 4, transmural infiltrations, loss of goblet cells, high vascular density, colon wall thickening).

Cytokine, MPO, and RT-PCR Assays. ELISA kits (Endogen, Woburn, MA) were used for mouse plasma, mucosa, and cell culture cytokines. Mucosal neutrophil infiltration was measured as MPO activity (6, 11). For analysis of mucosal mRNAs, mice were killed, and colons were removed, snap-frozen on liquid nitrogen, and stored at -80°C. Total RNA was isolated by using TRIzol reagent (Life Technologies), and RT-PCR was performed by using specific primers as described (6, 11).

LPMC Isolation and Cytokine Production. LPMCs were isolated as described (6, 11, 12). After excision of all visible lymphoid follicles, colons were washed in cation-free Hanks' balanced salt solution (HBSS) and treated with 1 mM EDTA in PBS for 20 min to remove epithelium, and tissue was digested with type IV collagenase (Sigma) for 20 min in a shaking incubator at 37°C. This step was repeated twice. Released cells were layered on a 40–100% Percoll gradient (Amersham Pharmacia) and centrifuged at 1,800 rpm in a Beckman JC2 centrifuge to obtain a lymphocyte-enriched population at the 40–100% interface (11, 12). LPMCs from different treatment regimens were suspended in medium (RPMI medium 1640/10% heat-inactivated FCS/3 mM l-glutamine/10 mM Hepes buffer/10 μg/ml gentamycin/100 units/ml penicillin/100 units/ml streptomycin/0.05 mM 2-mercaptoethanol) and cultured at 10⁶ cells per ml in 24-well Costar plates (Costar, Cambridge, MA). A total of 10⁶ LPMCs were cultured for 48 h in uncoated wells or wells containing immobilized murine anti-CD3ε antibody (clone 145-2C11; Pharmingen) and 1 μg/ml soluble anti-CD28 (clone 37.51; Pharmingen). Plates were coated with anti-CD3ε by using 10 μg/ml antibody in pH 9.6 carbonate buffer overnight at 4°C. Culture supernatants were harvested at 48 h and assayed for cytokines. All data were normalized to the final number of cells after 48 h.

ALX-R/fprl1 RT-PCR. Mouse ALX-R/fprl1 mRNA expression was examined by qualitative and quantitative RT-PCR using primers 5′-GATGCTAGAGGGGATGTGCAC-3′ and 5′-TCTTCAGGAAGTGAAGCC-3′, which amplify a 529-bp product identical to murine ALX-R (M. Arita, personal communications). The RT-PCR product from mouse colon was cloned into pCR2.1 vector (Invitrogen), sequenced, and confirmed to be aligned to murine ALX-R (GenBank accession no. NM_008042). Ribosomal 18S mRNA was used as a control: 5′-GCAATTATTCCCCATGAACG-3′ and 5′-GGCCTCACTA A ACCATCCAA-3′ (602 bp).

Data Analysis. All values are expressed as mean ± SEM. Variation among data sets was tested with ANOVA, and significance was tested with unpaired t tests, with a Bonferroni modification for comparison of more than two groups of data. Differences were considered significant at P < 0.05.

Results

3-oxa-ATL Analog Prevents and Treats Established TNBS-Induced Colitis. In pilot tests, daily oral dosing of 3-oxa-ATL analog ZK-192, starting on the day of TNBS administration, dose-dependently prevented TNBS colitis development at day 7 in mice (1.5 mg of TNBS per colon) and rats (30 mg of TNBS per colon). Dosed orally, 300 and 1,000 μg/kg ZK-192 significantly attenuated colon injury (macroscopic and microscopic scores) and mucosal levels of neutrophil-derived MPO activity in both species (data not shown). Lower ZK-192 doses did not attenuate the combined microscopic injury score in either species, but 30 μg/kg ZK-192 significantly attenuated colon wall thickening in rat TNBS colitis (data not shown). ZK-192 administered by drinking water (10 μg/ml) also markedly attenuated weight loss, macroscopic colon injury, and mucosal inflammation in mouse TNBS colitis. This effect was associated with decreased mucosal levels of the cytokine tumor necrosis factor α (TNF-α) and the chemokine macrophage inflammatory protein 2 (MIP-2) (Fig. 6, which is published as supporting information on the PNAS web site). ATLs are known to attenuate acute inflammation, neutrophil infiltration, and MIP-2 expression in response to exogenously added TNF-α in mice (13).

To study effects on survival, severe colitis was induced in mice by using 2.5 mg of TNBS per colon, with ≈40% animals surviving by day 7 (Fig. 1). Daily preventive treatment with 1,000 μg/kg ZK-192, given orally, improved survival to 80% (P < 0.01) and was equivalent to 10 mg/kg prednisolone (75%, P < 0.01). ZK-192 efficacy was correlated to decreased macroscopic and microscopic colon injury scores and decreased mucosal TNF-α levels (P < 0.05; data not shown). Doses of 300 and 1,000 μg/kg ZK-192 from day 3 after TNBS treated established colitis in mice. Significant effects at both doses were observed on colon injury scores and mucosal neutrophil infiltration (Fig. 1). A series of 14-day studies was performed to compare ZK-192 to prednisolone as a clinically relevant standard treatment for Crohn's disease. The ability of daily oral ZK-192 to prevent and treat established TNBS colitis in both mice and rats was highly reproducible (Tables 1 and 2, which are published as supporting information on the PNAS web site). Doses of 300–1,000 μg/kg ZK-192 were equivalent to 3–10 mg/kg prednisolone in mice, and equivalent or superior to 3 mg/kg prednisolone in rats. In rat colitis, oral ZK-192 was rapidly absorbed, with C_max plasma levels at ≈2 h reaching ≈75 and ≈250 nM at 300 and 1,000 μg/kg, respectively. By 6 h, levels declined to ≈25 and 75 nM, respectively. These low systemic drug levels attest to ZK-192 potency in vivo.

3-oxa-ATL Attenuates Mucosal Inflammation and Immune Dysfunction. To elucidate ZK-192 mechanism of action in vivo, preventive studies were performed in groups of 30 mice by using 1.5 mg of TNBS per colon, correlating colon injury scores to histology, mucosal/systemic disease biomarkers, and mucosal LPMC effector functions ex vivo. ZK-192 dose-dependently attenuated TNBS-induced weight loss, colon injury scores, and neutrophil infiltration, with 1,000 μg/kg oral ZK-192 equivalent to 10 mg/kg oral prednisolone (Fig. 7, which is published as supporting information on the PNAS web site). Fig. 2 shows transparietal colon sections from control mice (Fig. 2a), TNBS colitis plus vehicle (Fig. 2b), and TNBS colitis treated with 1,000 μg/kg ZK-192 (Fig. 2c) or 10 mg/kg prednisolone (Fig. 2d). TNBS-treated mice had transmural inflammation involving all colon wall layers, with increased muscle layer thickness and adherence to surrounding tissue. Inflammatory infiltrates consisting of monocytes, lymphocytes, and neutrophils were observed in the lamina propria with enlarged lymphoid follicles in the colon. TNBS caused diffuse mucosal inflammation characterized by epithelial cell loss, patchy ulceration, pronounced depletion of mucin-producing goblet cells, and reduced density of tubular glands. Treatment with ZK-192 and prednisolone markedly protected against TNBS-induced colon damage. The TNBS-induced increase in neutrophils and monocytes/macrophages infiltrating the lamina propria was attenuated by oral ZK-192 treatment (Table 3, which is published as supporting information on the PNAS web site).

Fig. 2.

3-oxa-ATL prevents TNBS-induced colon pathology. Transparietal colon sections from mice were obtained on day 7 and stained with hematoxylin and eosin. (a) Normal colon. (b) TNBS colitis plus vehicle. Colon wall thickening with pronounced inflammatory infiltrates in the lamina propria, and necrosis extending into the muscular and serosal layers is evident. (c) TNBS colitis plus 1,000 μg/kg ZK-192. (d) TNBS colitis plus 10 mg/kg prednisolone. Minor subepithelial edema and no inflammatory infiltrate in the mucosa/submucosa is seen in colons from drug-treated mice. (Original magnification, ×100.)

TNBS caused increases in mucosal and systemic TNF-α, IFN-γ, and IL-2 levels (Fig. 3 a–c). TNBS strongly induced mucosal mRNAs for these cytokines and IL-10, as well as the inflammatory mediators MIP-2, COX-2, and cytokine-inducible nitric oxide synthase (NOS) (Fig. 3d, lane 2). ZK-192 prevented in vivo induction of TNF-α, IFN-γ, IL-2, in the mucosa and systemically, as well as IL-10 and the inflammatory mediators MIP-2, COX-2, and inducible NOS in the mucosa (Fig. 3d, lane 3). The effect of ZK-192 on these mediators was similar to that of prednisolone (Fig. 3d, lane 4). Notably, ZK-192 did not inhibit expression of mRNAs for endothelial constitutive NOS, COX-1, or TGF-β.

Fig. 3.

3-oxa-ATL analog has combined effects on inflammation and immune dysfunction. Doses of 300 and 1,000 μg/kg oral ZK-192 were as effective as 10 mg/kg prednisolone in attenuating TNBS-induced increases in TNF-α (a), IFN-γ (b), and IL-2 (c) levels in the mucosa and plasma of mice. *, P < 0.001 TNBS vs. control; **, P < 0.001 drug-treated vs. TNBS. Data represent the mean ± SEM, n = 10–12. (d) RT-PCR analysis of mRNA for mucosal mediators. bp, bp ladder; -, negative control (water); +, positive control (mediator-positive cDNA). Lanes: 1, colon from control mouse; 2, TNBS plus vehicle; 3, TNBS plus 1,000 μg/kg ZK-192; 4, TNBS plus 10 mg/kg prednisolone. Data are representative of colons from at least four mice.

Attenuated inflammation by ZK-192 was consistent with the known antiinflammatory effects of ATL, but effects on cytokine-immune mediators suggested immunomodulatory actions on lymphocyte-driven Th₁ adaptive immune responses. To test this hypothesis, LPMC Th₁ effector cytokine release was examined ex vivo. As shown in Fig. 4, control LPMCs produced low levels of cytokines ex vivo, which were increased by T cell receptor (TCR)-directed CD3/CD28-dependent costimulation. LPMCs from TNBS-treated mice had enhanced Th₁ effector cytokine release both without and with TCR costimulation, consistent with hapten-mediated antigen-dependent priming in vivo and characteristic of adaptive Th₁ immune responses in TNBS colitis (5). In contrast, LPMCs from TNBS-treated mice receiving 300 (data not shown) and 1,000 μg/kg ZK-192 generated much less IFN-γ, TNF-α, and IL-2, even after TCR stimulation. No differences were seen in the total number of cells recovered after in vitro culture that could account for these dramatic results (data were normalized to cell numbers recovered after 48 h). These data suggested that ZK-192 treatment resulted in hyporesponsive LPMC with reduced ex vivo capacity for Th₁ effector function. Notably, ZK-192 attenuated LPMC Th₁ effector functions to an extent similar to that of prednisolone.

Fig. 4.

3-oxa-ATL analog ZK-192 attenuates TCR-directed lamina propria cell effector functions ex vivo.(a–c) Colitis induction and treatment groups are as in Fig. 3. LPMCs were isolated from colons on day 7 and cultured for 48 h with or without TCR-mediated CD3/CD28 costimulation. Basal cytokine production of unstimulated LPMCs was measured in cells from control colons (naíve LPMCs) in uncoated wells. ZK-192 and prednisolone significantly attenuated ex vivo LPMC cytokine release with and without CD3/CD28 costimulation. Data represent the mean ± SEM of six experiments. *, P < 0.01 vs. control cells; **, P < 0.01 vs. TNBS-primed untreated LPMCs. The experiment was carried out by using an equal number of cells (2 × 10⁵ LPMCs) per well. (d) Mouse ALX-R/fprl1 receptor mRNA (529 bp) is expressed in normal and inflamed colon. Lanes: 1, normal colon; 2, colon from TNBS-colitis. (e) ALX-R/fprl1 receptor expression in LPMCs obtained from TNBS colitis. Lanes 1, total LPMC; 2, CD4⁺ lymphocytes; 3, CD14⁺ monocytes. RT-PCR data are representative of four experiments. (f) ZK-192 (25 nM) attenuates CD3/CD28-dependent IFN-γ release from LPMCs isolated from mice with TNBS colitis. Data represent the mean ± SEM of four experiments. *, P < 0.05 vs. control; **, P < 0.05 vs. TNBS.

Murine ALX-R/fprl1 receptor mRNA was expressed in normal mouse colon, with slightly increased levels in colons from TNBS-treated mice (Fig. 4d). ALX-R/fprl1 was expressed in total LPMCs from TNBS-treated mice, with somewhat higher expression in CD14⁺ monocytes than CD4⁺ lymphocytes (Fig. 4e). Exposure of LPMCs from control and TNBS-treated mice to 25 nM ZK-192 in vitro markedly inhibited CD3/CD28-dependent IFN-γ release (Fig. 4f; P > 0.001 vs. CD3/CD28). Similar data were obtained for TNF-α, IL-2, and IL-12 (data not shown).

3-oxa-ATL Reduces TNBS Colitis in SCID Mice. Because ZK-192 administration reduced T lymphocyte and monocyte infiltration in the lamina propria, experiments were carried out to further define the immune compartment mediating ZK-192 actions in vivo. For this purpose, we tested ZK-192 in T and B cell-deficient SCID mice, in which innate immune cells provide important triggers for TNBS colitis (6). SCID mice developed TNBS colitis (Fig. 5) that was comparable in severity, mortality, and wasting disease to that of wild-type BALB/c mice (data not shown). Macroscopic and histology analysis of colons in SCID mouse colons showed marked inflammation and neutrophil infiltration, with large amounts of IL-2 and TNF-α. ZK-192 (1,000 μg/kg) and prednisolone (10 mg/kg) attenuated colitis in SCID mice to an extent similar to that observed in BALB/c mice (compare Fig. 5 to Figs. 1 and 3). ZK-192 inhibited mucosal TNF-α and IL-2 levels somewhat more effectively in SCID mice (≈75%) than in BALB/c mice (≈50%). These findings strongly suggest direct actions of ZK-192 on myeloid cells and other cells of the innate compartment in vivo, which are known to be major sources of T cell-directed effector cytokines such as TNF-α and IL-2.

Fig. 5.

3-oxa-ATL analog ZK-192 attenuates TNBS colitis in SCID mice. Colitis was induced in SCID mice with 2 mg of TNBS per colon and treated from day 1 to 7 with 1,000 μg/kg ZK-192 or 10 mg/kg prednisolone. Both treatments prevented TNBS-induced microscopic colon injury (Top) and mucosal levels of TNF-α (Middle) and IL-2 (Bottom). *, P < 0.01 TNBS vs. control; Ψ, P < 0.01 drug-treated vs. TNBS. n = 4 for no TNBS, n = 10 for TNBS and treatment groups.

Discussion

In immunocompetent mice, TNBS colitis evolves rapidly, with weight loss and colon injury by days 1–2 associated with neutrophil and inducible NOS-containing macrophage infiltrates (14, 15). Macrophage/dendritic cell engagement of adaptive Th₁ immunity progresses to day 7 and is sustained through day 14 (16). The therapeutic effect of the 3-oxa-ATL analog in established mucosal injury, i.e., dosing from days 3 to 14, is notable. The data contrast many previous studies in which LX/ATL effects were shown in acute models by administering LX/ATL before or at the time of insult (10, 17–19). These present studies provide clear evidence for a counterregulatory “resolving” effect of lipoxins on an established inflammatory response in a chronic disease setting. TNBS colitis responds to current Crohn's disease therapies, including 5-aminosalicylic acid, corticosteroids, and anti-TNF-α antibodies (6, 20–22), thus providing a useful model to test new therapeutic concepts for Crohn's disease. The improvement in TNBS-induced inflammatory bowel disease with ZK-192 (≈60–70%) compares favorably to anti-cytokine antibody therapies, such as anti-IL-4 in oxazolone colitis (23) and anti-IL-12, anti-α1, or anti-TNF-α in TNBS colitis (5, 6, 22). ZK-192 was more effective in preventing TNBS colitis than prednisolone in rats and equally as effective as prednisolone in mice. Efficacious doses of ZK-192 were similar to those found for a nitric oxide-releasing prednisolone derivative (11).

To identify the immune compartment mediating ZK-192 actions in vivo, we used lymphocyte-deficient SCID mice, and showed markedly attenuated colitis by ZK-192 in the absence of T and B lymphocytes. TNBS colitis is often regarded as a Th₁-mediated disease requiring T cell activation as an initiating event leading to macrophage recruitment/activation (3). Evidence that CD4⁺ Th₁ cells initiate/perpetuate disease is based on the cytokine response after CD3/CD28 costimulation of LPMC CD4⁺ T cells ex vivo and salutary effects of T cell-directed therapies on disease (5, 6, 24, 25). However, data in SCID mice show that colitis is induced in the absence of T and B lymphocytes and concur with non-T cells acting as triggers for TNBS colitis (26, 27). Within the innate compartment, neutrophils, macrophages/monocytes, dendritic cells, and natural killer cells are known to respond to LX/ATL. Indeed, data shown in Table 3 indicate that ZK-192 effectively reduced neutrophil influx in the lamina propria of TNBS colitis. Thus, whereas previous studies indicate a nonessential role for neutrophils in TNBS colitis (28), the present results suggest that inhibition of neutrophil recruitment may contribute to the antiinflammatory activity of the ATL analog in this model. In addition to neutrophils, likely targets for ZK-192 in TNBS colitis are macrophages, monocytes, and dendritic cells, which are critical innate triggers of adaptive immunity. In Toxoplasma gondii infections, 5-lipoxygenase and LX/ATL regulate dendritic cell IL-12 release, CD4⁺ Th₁ cytokine release, and selected macrophage effector functions, such as TNF-α and MIP-1α release (29). Microbial antigens prime dendritic cells to produce and use IL-2 to promote T cell-dependent adaptive immunity (30), and high TNF-α-producing dendritic cells are found in Crohn's disease patient mucosa (31). We propose that inhibition of mucosal TNF-α and IL-2 levels by ZK-192 in SCID mice with TNBS colitis is caused by direct actions on mucosal macrophages and/or dendritic cells via LX/ATL receptor(s), such as ALX-R/fprl1. LX/ATLs activate monocyte chemotaxis and may recruit myeloid cells to inflamed tissue (32). LX/ATLs stimulate actin reorganization in monocytes and macrophages, promoting pseudopodia formation and stimulating nonphlogistic phagocytosis of apoptotic neutrophils in vivo (32, 33). This is associated with TGF-β release, but not release of proinflammatory chemokines (IL-8 and MCP-1), supporting the view that LX/ATL can stimulate macrophage functions essential to resolving inflammation. Notably, mucosal TGF-β mRNA expression was not attenuated by ZK-192 in TNBS colitis.

In vivo ZK-192 treatment attenuates LPMC responsiveness to subsequent ex vivo TCR-directed stimulation. Th₁ cytokines, such as IFN-γ, play central and pathogenic roles in TNBS colitis (6, 11, 23, 34). ZK-192-dependent hyporesponsiveness in vivo suggests that LX/ATL regulate T cells directly. It was recently shown that LX/ATL potently inhibit TCR-dependent TNF-α secretion in human CD3⁺ T cells (35). Here we show that ALX-R/fprl1 is present in mouse colon and expressed in both CD14⁺ monocyte and CD4⁺ lymphocyte LPMC subsets in TNBS colitis. ZK-192 potently attenuated CD3/CD28-dependent effector cytokine release from LPMC, further supporting direct effects of ZK-192 in modulating T cell function. However, although we cannot exclude direct ZK-192 effects on CD4⁺ Th₁ cells in vivo, the SCID mice data are more consistent with secondary effects on T cells via modulation of macrophage/monocyte and dendritic cell effector functions. Further analysis of the direct actions of LX/ATL on LPMC subsets alone and in coculture are needed to unravel these important questions.

In conclusion, we have shown that a recently developed 3-oxa-ATL analog, ZK-192, has potent once-daily oral efficacy in a clinically relevant Crohn's disease model. The results significantly extend previous findings that LX/ATL have gastrointestinal protective actions in acute aspirin-induced gastritis (36) and the acute phase of dextran sodium sulfate (DSS)-induced colitis (37). The latter study showed preliminary and modest effects of an ATL analog on DSS-induced weight loss and survival, but provided no mechanistic insights on the protective actions of LX/ATL in vivo. Here, analysis of mucosal inflammation, cytokine, chemokine, and immune cell networks in immunocompetent and lymphocyte-deficient mice with TNBS colitis has revealed counterregulatory actions of ATL in vivo at multiple levels. LX/ATL exert potent antiinflammatory effects and also immunomodulatory effects on innate and adaptive immune cell-driven mucosal pathology. This broad profile suggests that ZK-192 and related 3-oxa-ATL analogs (10) provide a promising oral approach to treatment of Crohn's disease and other inflammatory bowel diseases.