Ibuprofen was purchased from Sigma (St. Louis, MO). Phospho-butanol-ibuprofen was purchased from Chem-Master International Inc. (East Setauket, NY). The chemical structure is shown in Fig. 1A. The purity of synthetized drug p-ibuprofen is >99%.

Fisher 344 male rats (135) (Harlan Sprague Dawley, Indianapolis, IN), 3–4 weeks old with an initial average body weight of 90 g, were acclimated for 1 week, divided into 6 groups: groups 1, 3 and 5 were given saline as control, 15 rats per group; and groups 2, 4 and 6 were given carcinogen, 30 rats per group. Animals received either saline or carcinogen (azoxymethane, AOM, at 15 mg/kg) by subcutaneous injection, once a week for two weeks. Drug administration was also initiated in rats via diet 10 days before subcutaneous injections: vehicle for groups 1 and 2; ibuprofen, 500 ppm, for groups 3 and 4; and phosphoibuprofen (p-ibuprofen, the same hereinafter), 900 ppm (equal molar dose with ibuprofen), for groups 5 and 6. All drugs were administered up until the end of the experiment as shown in Fig. 1B. Animals were housed and maintained according to the approved standards of Stony Brook University Institutional Animal Care and Use Committee. Animals were housed two per plastic cage with sawdust bedding and were kept under standard laboratory conditions (room temperature, 22±2°C; relative humidity, 50±5%; light/dark cycle 12/12 h). All animals had access to food and tap water ad libitum. Rats were observed daily and weighed once a week. Half of the rats from each group were euthanized at week 20 and analyzed for aberrant crypt foci (ACF). Blood was collected for pharmacokinetic (PK) analysis. The remaining rats were euthanized at week 40 and colons were dissected and analyzed for aberrant crypt foci (ACF) and tumors. Heart, lung, liver, stomach and kidney were collected from animals in the control groups and fixed in buffered-formalin for histological analysis of potential toxic effects.

For animals sacrificed at week 20, ACF were counted in colon tissues as previously described (24). Briefly, the colons were removed, rinsed with ice-cold phosphate-buffered saline (PBS), placed on filter paper, opened longitudinally, and fixed in 10% buffered formalin for 24 h. Then colon tissues were stained with 0.2% methylene blue for 3 to 5 min. The number of ACF per colon was determined by microscopic examination. ACF were distinguished from surrounding normal crypts by increased size, thickened epithelial cell lining, and enlarged cryptal area relative to surrounding normal crypts as shown in Fig. 2A.

At week 40, the remaining animals were sacrificed by CO₂ asphyxiation. Colons were removed, opened longitudinally, and rinsed with PBS. Tumors were counted and both long-diameter by short-diameter were measured to calculate tumor size. After measurements were recorded, one half of each colon sample was frozen in liquid nitrogen and stored at −80°C for further analyses. The remaining half of each colon was fixed in buffered formalin for histopathology processing. Briefly, colon tissue with tumors were equally cut into 10 pieces, and embedded in paraffin blocks. Sections (4 μm) were stained with hematoxylin and eosin to determine histopathology by pathologist, or stained by immunohistochemistry and immunofluorescence for mechanism study.

Thirty-five Fisher 344 rats (Harlan Sprague Dawley), male, 8–9 weeks old, were divided into 5 groups (7 rats per group) and treated with the following: group 1, vehicle; group 2, ibuprofen 75 mg/kg of body weight; group 3, ibuprofen 670 mg/kg; group 4, p-ibuprofen 135 mg/kg; and group 5, p-ibuprofen 1,215 mg/kg. The drugs were administered by gavage, once a day for 7 days. Animals were housed under standard conditions and euthanized by CO₂ asphyxiation 1 h after final drug administration. Blood was collected for PK studies. Stomach, small intestine and colon were checked for ulcers under magnification lens then fixed with 10% buffered formalin for histology. Heart, liver, lung and kidney were collected for toxicity analyses.

The blood samples collected from animals sacrificed at week 20 and from the toxicity study with low dose treatment were used for the PK study. Briefly, ibuprofen and its metabolites were extracted by adding a 2-fold volume of acetonitrile. After centrifugation for 10 min at 5,000 × g, the supernatants were subjected to HPLC analysis. The HPLC system consisted of a Waters Alliance 2695 Separations Module equipped with a Waters 2998 photodiode array detector (220 nm) (Waters, Milford, MA) and a Thermo BDS Hypersil C18 column (150×4.6 mm, particle size 3 μm) (Thermo Fisher Scientific, Waltham, MA). The mobile phase followed a gradient between buffer A [formic acid, acetonitrile, H₂O (95:4.9: 0.1 v/v/v)] and buffer B (acetonitrile).

To evaluate the inhibitory effect of ibuprofen or p-ibuprofen on COX-2, RAW 264.7 macrophages were pre-treated with ibuprofen or p-ibuprofen, 130 μM, for 12 h, then incubated with LPS, 100 ng/ml, overnight. The cells were collected for COX-2 measurement and analyzed by western blotting. Levels of PGE₂ in cell culture media were determined using a commercially available immunoassay kit according to the manufacturer’s instructions. Briefly, 1.5×10⁶ Raw 264.7 macrophages were pre-incubated with ibuprofen or p-ibuprofen, 130 μM, for 12 h, followed by LPS (100 ng/ml) overnight. The cultured media were collected to measure PGE₂ level using an ELISA kit (Cayman Chemical, Ann Arbor, MI).

Paraffin-embedded sections were deparaffinized, rehydrated and microwave heated for 15 min in 0.01 mol/l citrate buffer (pH 6.0) for antigen retrieval. Tissue sections were then incubated with 5% donkey serum for 30 min, then treated with primary antibodies, rabbit anti-phospho-p65, ser276 (Cell Signaling, Danvers, MA) and mouse anti-β-catenin (Millipore, Temecula, CA) or control IgG, at 4°C overnight. After washing with PBS 3×5 min, the secondary donkey anti-mouse IgG and donkey anti-rabbit IgG conjugated with fluorescents were added and incubated at room temperature for 1 h. Slides were washed thrice with PBS, mounted with media and observed under a fluorescence microscope.

RAW 264.7 macrophage (mouse leukemic moncyte) and HCT 116 colon cancer cell lines were purchased from American Type Culture Collection (Manassas, VA) and cultured in DMEM or RPMI-1640 medium, respectively.

Whole cell extracts were obtained by lysing cells in RIPA buffer [50 mM Tris-HCl (pH 7.4),150 mM NaCl, 1 mM Na₂EDTA, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1% NP-40, 0,25% sodium deoxycholate and Protease Inhibitor Cocktail 2 (Sigma-Aldrich)]. Cytoplasmic and nuclear extracts were prepared following a standard protocol (25); trypsinized cells were suspended in lysis buffer to which NP-40 was added at a subsequent step (the supernatant fraction represented the cytoplasmic extract); nuclei were washed and centrifuged, followed by resuspension in extraction buffer and pelleting. Protein extracts were analyzed by a well-established standard western blot procedure (26). Rabbit anti-COX-2 antibody was purchased from Cayman Chemical. Other primary antibodies are indicated in the immunofluorescence staining.

Data are expressed as mean ± SEM and analyzed with ANOVA. P≤0.05 was considered statistically significant.

AOM-induced ACF at week 20 and colon tumors at week 40 as shown in Fig. 2A. Histology of tumors shows well-differentiated adenocarcinoma. As shown in Table I and Fig. 2B, compared to vehicle treatment, both p-ibuprofen or ibuprofen significantly reduced the multiplicity of ACF by 31.2% (54.2±4.4 vs. 78.8±11.6, p<0.05), or 37.9% (48.9±7.9 vs. 78.8±11.6, p<0.05), respectively. However, no difference was observed between groups treated with p-ibuprofen or ibuprofen (p>0.05). Similarly, as shown in Table I and Fig. 2C, at week 40 treatment with either p-ibuprofen or ibuprofen reduced the multiplicity of colon tumors by 47.2% (2.8±0.52 vs. 5.3±0.59, p<0.01), or 56.6% (2.3±0.49 vs. 5.3±0.59, p<0.01), respectively. Again, a significant difference (p>0.05) was not observed between groups treated with p-ibuprofen and ibuprofen (Table I and Fig. 2D). Furthermore, pathological changes were not detected in heart, lung, liver, stomach and kidney tissues. These results suggest that phospho-modification did not significantly enhance the inhibitory effect of ibuprofen in AOM-induced colon cancer in rats.

As shown in Fig. 3A, a high dose of ibuprofen (670 mg/kg) administered to rats by gavage for 7 days led to stomach ulcerations in 85.7% (6 out of 7) of rats; whereas, an equal molar dose of p-ibuprofen (1,215 mg/kg) caused stomach ulceration in only 14.3% (1 out of 7) of rats. The modification of ibuprofen with the phospho-moiety remarkably reduced its GI side-effect (p<0.01). All animals with stomach ulcers also exhibited ulcerations in the small intestine. Histologically, the typical ulcer appeared with the hiatus of mucosa and ulcerative tissue in the bottom of the ulcer (Fig. 3D). Some of the ulcers exhibited inflammation and perforation (breaking through the wall) in both stomach and intestine (Fig. 3B–E). A low dose of both p-ibuprofen and ibuprofen did not lead to ulceration in either stomach or intestine. There was no obvious pathological change in heart, liver, lung and kidney. These results demonstrate that, whereas the phospho-modification of ibuprofen did not enhance ibuprofen’s inhibitory effect on AOM-induced colon cancer in rats, p-ibuprofen significantly reduced the GI toxicity of ibuprofen.

As shown in Fig. 4A, with continuous administration of ibuprofen or p-ibuprofen in the diet for 20 weeks, ibuprofen can be detected in the plasma of both treatment groups. The plasmic level of ibuprofen in animals fed a diet with p-ibuprofen is lower than that of animals fed a diet of ibuprofen. Both diet groups (p-ibuprofen and ibuprofen) exhibited the tendency of a higher plasmic ibuprofen level in AOM-treated animals as compared to saline control animals. However, no statistical difference was noted. These trends were comparable in the case where rats were on the same diet for 40 weeks. Intact p-ibuprofen was not detectable in the plasma of animals fed a diet consisting of or gavaged with p-ibuprofen. As seen in Fig. 4B, the metabolite of ibuprofen, 2-OH-ibuprofen, was detectable at a similar level as intact ibuprofen in both groups fed a diet of p-ibuprofen (102 μM) or ibuprofen (99 μM). Other metabolites were detected at very low levels, including carboxy-ibuprofen (3 and 3 μM), 1-OH-ibuprofen (4 μM, undetectable) and ibuprofen-glucuronide (1 and 1 μM); 20 weeks and 40 weeks, respectively.

We evaluated the classic pathway of NSAIDs-COX-2 expression in macrophages. Our results showed that LPS highly induced COX-2 levels and that this effect was completely blocked by both ibuprofen and p-ibuprofen (Fig. 5A). These results suggest that phospho-modification did not alter the property of ibuprofen in inhibiting COX-2 expression.

As seen in Fig. 5B, both ibuprofen and p-ibuprofen significantly suppressed PGE₂ production by 84.1% (1.7±0.5 vs. 10.7±1.1, p<0.01) and 77.4% (2.4±1.3 vs. 10.7±1.1, p<0.01) as compared to LPS-treated control, respectively. However, no significant difference between ibuprofen and p-ibuprofen (p>0.05) was observed. This result is consistent with the observed inhibition of COX-2, suggesting that phospho-modification does not decrease the ability of ibuprofen to suppress PGE₂ synthesis.

As shown in Fig. 6A, exposure to the carcinogen AOM induced β-catenin cytoplasmic accumulation and nuclear translocation. These effects were reduced by both p-ibuprofen and ibuprofen. The presence of phosphorylated NF-κB subunit p65 indicates NF-κB activation. By immunohistochemistry analysis (Fig. 6A), p-ibuprofen and not ibuprofen inhibited phosphorylation of p65 (NF-κB activation). This in vivo observation was confirmed by an in vitro study. As seen in Fig. 6B, treatment with both p-ibuprofen and ibuprofen markedly decreased β-catenin levels in the nuclear extracts of HCT116 cells as compared to control (vehicle-treated). Interestingly, p-ibuprofen inhibited the nuclear translocation of the NF-κB subunit p65 (NF-κB activation), but this inhibitory effect was not observed for ibuprofen. This additional inhibitory effect in NF-κB activation may explain why the lower plasmic level of ibuprofen after treatment with p-ibuprofen exhibited a similar tumor inhibitory effect with ibuprofen-treated group.