Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Research from our laboratory has provided evidence that the gastrointestinal (GI) mucosae may be protected from damaging agents and microorganisms in the lumen by a hydrophobic extracellular lining comprised of zwitterionic phospholipids (Butler et al., 1983; Hills et al., 1983; Lichtenberger et al., 1983; Lichtenberger, 1995). Furthermore, we have demonstrated that, in addition to inhibiting GI mucosal cyclooxygenase (COX) activity and depleting the levels of "cytoprotective" prostaglandins, nonsteroidal anti-inflammatory drugs (NSAIDs) have the capacity to topically damage the GI mucosa, in part by chemically associating with phospholipids that line the luminal aspects of the mucus gel layer, attenuating the tissue's hydrophobic barrier properties (Goddard et al., 1990; Lichtenberger et al., 1995; Giraud et al., 1999). These observations led to the development of a family of NSAIDs chemically associated with either synthetic or purified PC, which proved to be significantly less damaging to the GI mucosa in both rodents and humans (Giovannucci et al., 1995; Anand et al., 1999). We have also determined in both rats and humans that PC association did not affect an NSAID's ability to inhibit mucosal COX activity and deplete the tissue of "cytoprotective" prostaglandins. Thus, the protective nature of PC-NSAIDs appears to be unrelated to preserving the prostaglandin concentration of the GI mucosa and most likely resides in preventing the topical injurious action of NSAIDs. Preclinical studies in rats also demonstrated that PC-associated aspirin possessed a 5- to 10-fold enhancement in potency to inhibit fever in comparison with unmodified aspirin and also was more effective and potent inhibiting acute joint inflammation and pain (Lichtenberger et al., 1996b). Consistent with these findings, we also demonstrated that the anti-inflammatory/analgesic potency of indomethacin was also enhanced if the nonaspirin NSAID was chemically associated with PC (Lichtenberger et al., 1996a).

In the present study, we investigated the effects of phospholipid association on the therapeutic activity and potency of another commonly used NSAID, ibuprofen, in both rodent models of acute and chronic joint inflammation. In this endeavor, we compared several test preparations in which the ibuprofen was combined with either purified or semipurified PC from soy lecithin. We also explored the mechanism responsible for the resultant enhancement in therapeutic activity of PC-ibuprofen by investigating the drugs' bioavailability and COX-inhibitory activity in comparison with unmodified ibuprofen.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Rats, purchased from Harlan Sprague-Dawley Laboratories (Indianapolis, IN), were used in three separate studies. Male Sprague-Dawley rats weighing 150 to 200 g were used in the acute inflammation/pain and bioavailability studies, whereas female Lewis rats weighing 100-125 g were used in the adjuvant-induced arthritis study, due to the reported sensitivity of rats of this strain, age (42-51 days old), and gender to develop polyarthritis in response to a single injection of Freund's adjuvant (Wilder et al., 1982; Weichman, 1989). All animals were allowed a conditioning period of at least 1 week before the commencement of an experiment, during which time they were housed in wire mesh cages in our institution's Animal Care Center, with ad libitum access to food (Purina Lab Chow, Ralston Purina, St. Louis, MO) and water in a room, with automatic lighting control (12 h light between 7:00 AM-7:00 PM, 12 h dark, 7:00 PM-7:00 AM). The protocols outlined below received prior approval by our institution's Animal Welfare Committee, indicating that they met or exceeded both the National Institutes of Health and the American Association for the Accreditation of Laboratory Animal Care guidelines in the care and treatment of laboratory animals.

Acute Pain/Inflammation. We evaluated the analgesic activity of our ibuprofen test preparations to reduce peripheral hypersensitivity of acutely inflamed tissue to external pressure, as described in our modification of the technique of Randall and Selitto (Randall and Selitto, 1957; Lichtenberger et al., 1996). Briefly, this involved the s.c. injection of 0.05 ml of Complete Freund's Adjuvant (CFA; Sigma Chemical, St Louis, MO) into the right hind paw of an ether-anesthetized male Sprague-Dawley rat to induce an acute inflammatory response. Three days later, the rats were fasted overnight (with ad libitum access to water), and the following day the rats were intragastrically administered either 1 ml of saline or the ibuprofen test formulations. Two hours later, the animals were placed in restraining cages that had been modified to provide ready access to the rats' hind paw. External pressure was then gradually applied to both the inflamed and contralateral uninflamed paw (0-250 g, at a rate of 16 g/s) with an Analgesy meter (Life Science Institute, Woodland Hills, CA) under the control of an observer who was unaware of the treatment groups. The "pain threshold" was defined as the pressure when the animal first showed evidence of the sensation of pain, indicated by the extension of digits and/or the initial signs of paw withdrawal and/or vocalization.

Adjuvant-Induced Arthritis. The animal model of adjuvant-induced arthritis, outlined below, is a modification of a previously described technique (Weichman, 1989). Female Lewis rats received an intradermal injection (via a 25 gauge needle attached to a 2.5-ml glass syringe) at the base of the tail of 0.2 ml of CFA prepared in our laboratory as follows: 10 ml of Incomplete Freund's Adjuvant (Difco, Detroit, MI) was combined with an ampule of Microbacterium tuberculosis H37 RA (Difco), followed by mortar/pestle grinding (3 min) and sonication (1 min) until a milky suspension was obtained. The animals were then returned to the laboratory at 2-week intervals, where under light ether anesthesia their joint inflammation was measured by a spring-activated micrometer (pocket thickness gauge, Swiss Precision Instruments, Los Angeles, CA), and they were placed in restraining cages to assess pain sensitivity to externally applied pressure, as described above. Fourteen days after CFA injection, we initiated the daily treatment of the rats with saline (controls) or one of the ibuprofen test solutions, intragastrically administered as a 1-ml bolus. Rats were euthanized by CO₂ inhalation 3 to 6 weeks after the initiation of ibuprofen treatment, at which time the affected ankle joint was dissected, lanced by scalpel, and incubated in 0.1 M Tris buffer (pH 7.4) for 2 h at room temperature. The prostaglandin concentration of the collected synovial fluid effusate was then measured by radioimmunoassay, as described below.

Ibuprofen Bioavailability. Fasted male Sprague-Dawley rats were intragastrically administered 1 ml of the ibuprofen test solutions and the rats anesthetized under ether 0, 0.5, 1, 2, 4, 6, and 8 h, later, at which time blood was collected from the inferior venae cavae and immediately extracted in acetonitrile (1 part blood:3 parts acetonitrile) and vortexed. The rats were euthanized immediately thereafter by an overdose with an anesthetic agent. The extracted samples were then subjected to low-speed centrifugation, and the supernatant collected and stored at 80°C in preparation for subsequent HPLC analysis.

HPLC Analysis. For HPLC analysis, the acetonitrile-extracted blood was thawed, vortexed for ~1 min, and centrifuged for 10 min at 2500 rpm. The clear acetonitrile supernatant was removed and injected directly into the HPLC system. The HPLC system, which was based on a previously published technique (Sochor et al., 1991), consisted of a Waters solvent delivery system, a Wisp auto-sampler (710 B), a Waters 900E variable wavelength detector set at 222 nm, and a C-8 reverse-phase column (Beckman Ultrasphere octyl; 4.6 × 250 mm, 5 µ). The system operated at room temperature using a mobile phase composed of acetonitrile (+0.25% phosphoric acid) in water (45:55), with a flow rate of 1.4 ml/min. Retention time for ibuprofen was 5.6 min. The dimensions of the elution peaks of the NSAID and its metabolites were integrated and read off appropriate standard curves for quantitation of the area under the curve (AUC), C_max, T_max, rate of elimination (K_el), and circulatory half-life (t_1/2) for each drug according to standard pharmacokinetic computer formulae (Rowland and Tozer, 1989). The percentage recovery of ibuprofen from whole blood, determined at three representative concentrations over the range of the standard curve (ibuprofen), was >90%.

COX-2 Induction in Cultured Human Umbilical Vein Endothelial Cells (HUVECs). HUVECs were prepared from freshly obtained umbilical veins as previously described (Wu et al., 1988). The cells were cultured on porcine skin gelatin (1 mg/ml)-coated flasks and were maintained on Medium 199 (Invitrogen, Grand Island, NY) containing 20% heat-inactivated fetal calf serum, 50 mg/ml of endothelial growth supplement (Collaborative Biomedical Products, Bedford, MA), 100 units/ml heparin, 50 µg/ml streptomycin, 200 units/ml neomycin, and 100 units/ml penicillin (Invitrogen). We then investigated the COX-2 inhibitory activity of the ibuprofen test formulation in HUVECs in passages 2 to 4, as described below. HUVECs were cultured in a six well plate (Costar, Corning, NY) until confluent. At this time, the medium was exchanged for fresh medium containing 3% fetal calf serum for 16 h. At the completion of this incubation period, the medium was exchanged for fresh serum-containing medium supplemented with phorbol ester (100 nM phorbol-12-myristate-13-acetate; Sigma) to induce COX-2 gene expression. At the end of this induction period, the medium was replaced with fresh serum-free medium, in the absence or presence of ibuprofen and PC-ibuprofen, and the cultures incubated for an additional 30 min. The ibuprofen (±PC) was tested at a dose range between 1 µM and 1 mM. After the 30-min NSAID treatment period, arachidonic acid was added to the culture medium at a final concentration of 10 µM to initiate prostaglandin biosynthesis, and after an additional 15 min, the culture medium was collected and stored (80°C) for subsequent measurement of 6-keto PGF₂ by radioimmunoanalysis.

Prostaglandin Analysis. 6-keto-PGF₁ (metabolite of PGI₂) was measured by a highly specific enzyme immunoassay as described previously (Sanduja et al., 1991; Anand et al., 1999; Giraud et al., 1999) from medium samples derived from the above study or an effusate of synovial fluid. The latter sample was collected by dissecting the rat's hind paw or pes 1 cm above the inflamed stifle joint. Four deep lacerations (two on the ventral surface and two on the dorsal surface) were then made in the inflamed stifle joint, and the dissected tissue was then immersed in 2 ml of 50 mM Tris buffer (pH 7.4) overnight at 4°C and subsequently stored at 20°C until analyzed by prostaglandin radioimmunoassay. It should be acknowledged that the collected sample was not pure synovial fluid, and an unspecified amount of the prostaglandins analyzed were from surrounding tissue. Nonetheless, since the same technique was used for collecting samples from all groups, the component contributed by prostaglandins of nonsynovial fluid/tissue origin should not have biased the results.

Preparation of PC-Ibuprofen Formulations. In this study, we compared the therapeutic activity and bioavailability of two PC-ibuprofen formulations to that of an equivalent dose of unmodified ibuprofen. The first PC-ibuprofen formulation was prepared by initially dissolving purified PC [Phospholipon 90G (P90G), containing >93% PC, purified from soy lecithin by the Nattermann Phospholipid GmbH of Cologne, Germany] in chloroform. The organic solvent was removed by evaporation under N₂, followed by placing the vessel under vacuum for 24 to 48 h. At this time point, ibuprofen (purchased from Sigma) was dissolved in water at a predetermined concentration and added to a vessel containing the PC film, to provide a weight ratio of ibuprofen:P90G of 1:3. The vessel was then vortexed (~1 min), followed by a brief period of sonication in a bath-type sonicator to form a uniform aqueous PC-ibuprofen suspension.

Statistical Analysis. Data presented in this paper are expressed as the means ± S.E.M. Statistical comparisons between groups were made either by a Student's t test or by a one-way analysis of variance, followed by Fisher (least significant difference) multiple comparison procedure to assess significant differences between groups with p  0.05 being considered statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Evaluation of the Therapeutic Activity of PC-Ibuprofen against Acute Pain and Inflammation. In this initial study, we compared the analgesic activity of PC-ibuprofen to unmodified ibuprofen in a model of acute joint inflammation at a NSAID dose range of 10 to 50 mg/kg. In these studies, we used Phosal 35SB as the phospholipid source containing 35% PC. Figure 1 demonstrates that PC-ibuprofen was effective in significantly increasing the pain threshold of the inflamed hind paw to pressure, over saline-treated control values, at all doses tested. In contrast, the unmodified ibuprofen was not as effective as the PC-associated NSAID at increasing the pain threshold over control values, with the lowest doses (10 and 25 mg/kg) failing to have analgesic activity in this animal model.

View larger version (38K): [in this window] [in a new window]   Fig. 1.   Comparison of the dose-dependent effects of PC-ibuprofen to unmodified ibuprofen on the analgesic activity (expressed as mean ± S.E.M. of the pain threshold to pressure applied to inflamed and uninflamed contralateral hind paw) 2 h after intragastric administration of saline or the ibuprofen test formulations. Phosal 35SB was used as the source of PC. An acute inflammatory response was elicited 4 days earlier by the s.c. injection of 0.05 ml of CFA into the left hind paw. The pain threshold of the uninflamed right contralateral hind paw was also measured, and the mean ± S.E.M. is represented by the width of the dashed line. In this and subsequent figures (unless indicated otherwise), *p  0.05 versus values of the saline-treated group; p  0.05 versus values of the group treated with unmodified ibuprofen. In this and subsequent figures, analysis of variance, followed by a Fisher least significant difference procedure, was used in comparing the various test formulations with saline, whereas a Student's t test was used when comparing the values in response to ibuprofen to one of the test PC-ibuprofen formulations.

Comparison of the Efficacy of PC-Ibuprofen to Ibuprofen to Treat Adjuvant-Induced Arthritis. Two weeks after administration of adjuvant, most of the rats showed the initial signs of joint inflammation (erythema and mild swelling confined to the midfoot or ankle joint), and the treatments were started. Figure 2 demonstrates that, in comparison with the very prominent increase in ankle thickness observed in saline-treated control rats, both formulations of PC-ibuprofen (administered at a NSAID dose of 25 mg/kg) were effective in attenuating the development of joint inflammation over the 50-day study period. Interestingly, administration of unmodified ibuprofen at this dose to arthritic rats induced only a modest decrease in joint inflammation over the study period, which failed to attain statistical significance in comparison to saline-treated control values. Significant differences in the anti-inflammatory activity between the two PC-ibuprofen formulations and ibuprofen alone became apparent after 2 to 3 weeks of treatment with the test compounds. In the initial experiment, we determined that joint inflammation in arthritic rats administered the PC (35% pure):MCT vehicle in the absence of ibuprofen was comparable with that of saline-treated controls (see Table 1). Because no statistically significant differences in joint inflammation were found between these two control groups, we only used saline-treated rats as our control group in subsequent experiments.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Comparison of the anti-inflammatory activity of formulations of PC-ibuprofen (Ibu) to unmodified ibuprofen in a rodent model of adjuvant-induced arthritis. Saline (1 ml) or an equivalent volume of the ibuprofen test formulations were intragastrically administered daily to rats at a NSAID dose of 25 mg/kg, commencing 2 weeks after adjuvant challenge. Ankle thickness was measured with a spring-activated micrometer (with a 40-unit change corresponding to a 1.0-mm increase in ankle thickness) as an index of joint inflammation over the 2-month study period. n = 5 to 6 rats/group. mct, medium chain triglyceride.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Comparison of the analgesic activity of formulations of PC-ibuprofen (Ibu) to unmodified ibuprofen (intragastrically administered daily at a NSAID dose of 25 mg/kg) in a rodent model of adjuvant-induced arthritis, as described in Fig. 2. Analgesic activity was assessed by measuring the pain threshold to pressure applied to the affected hind paw. n = 5 to 6 rats/group. mct, medium chain triglyceride.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Comparison of the anti-inflammatory activity of formulation of PC-ibuprofen (Ibu) to unmodified ibuprofen (intragastrically administered daily at a NSAID concentration of 15 mg/kg) in a rodent model of adjuvant-induced arthritis, as described in Fig. 2. n = 7 rats/group.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Comparison of the analgesic activity of formulations of PC-ibuprofen (Ibu) to unmodified ibuprofen (intragastrically administered daily at a NSAID concentration of 15 mg/kg) in a rodent model of adjuvant-induced arthritis, as described in Fig. 3. n = 3 to 7 rats/group.

Bioavailability. A kinetic analysis of the appearance of ibuprofen in the blood was performed 5 to 480 min after fasted rats were intragastrically administered the ibuprofen test formulations at a dose of 100 mg/kg. The higher dose of the ibuprofen was used in these bioavailability studies to facilitate the detection and quantitation of circulating levels of the NSAID by HPLC. Whole blood concentrations of ibuprofen over the 8-h study period are depicted in Fig. 6, and values for pharmacokinetic parameters are given in Table 2. HPLC analyses failed to detect the presence of metabolites of ibuprofen in the blood of rats of any of the test groups throughout the 8-h study period.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Changes in ibuprofen (Ibu) concentration (as determined by HPLC) in blood over the 8-h study period, of rats intragastrically administered PC-ibuprofen formulations or unmodified ibuprofen at an NSAID dose of 100 mg/kg. n = 4 to 6 rats/group.

COX-Inhibitory Activity. At the completion of the adjuvant-induced arthritis study, rats in all groups were euthanized, their affected hind paws dissected, and the joints effused with buffer to measure synovial fluid prostaglandin concentration. Figure 7 demonstrates that, at a dose of 25 mg/kg, all ibuprofen test formulations had a capability to reduce the concentration of the PGI₂ metabolite, 6-keto-PGF₁, by tissues in and around the inflamed stifle joint. Although there was a trend for the PC-ibuprofen formulations to be more effective COX inhibitors than ibuprofen alone in reducing the tissues' prostaglandin levels, these differences did not attain statistical significance.

View larger version (31K): [in this window] [in a new window]   Fig. 7.   6-Keto-PGF1 concentration of effusate of tissues in and around the inflamed stifle joint of arthritic rats intragastrically administered PC-ibuprofen (Ibu) formulations or unmodified ibuprofen daily at a dose of 25 mg/kg for 64 days. n = 4 to 6 rats/group.

View larger version (16K): [in this window] [in a new window]   Fig. 8.   Comparison of the potency of ibuprofen to PC-ibuprofen to inhibit COX-2 activity of 12-O-tetradecanoylphorbol-13-acetate-activated HUVECs. COX-2 activity was assessed by measuring the conversion to arachidonic acid to PGF1 in activated HUVECs that were pretreated with the ibuprofen test formulations. *p  0.05 versus values of cells treated with the equivalent dose of ibuprofen. n = 8 dishes/group.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

NSAIDs have proven to be the drug of choice for many of the 20 to 30 million Americans suffering from rheumatoid and osteoarthritis (Gabriel and Fehring, 1992). The therapeutic activity of NSAIDs has been attributed to their established ability to inhibit COX, the key regulatory enzyme in prostaglandin biosynthesis. Although it has been presumed that tissue inflammation was specifically controlled by the COX-2 isoform, which is induced by cytokines and other inflammatory mediators (Vane, 1971; Meade et al., 1993; Mitchell et al., 1993), compelling evidence has recently been reported with the use of COX-knockout mice (Langenbach et al., 1995; Morham et al., 1995) and COX selective inhibitors (Wallace et al., 1998) to indicate that the COX-1 isoform is involved as well. One of the major problems that has beset the chronic use of NSAIDs relates to their ability to induce GI side effects, including dyspepsia, gastroduodenal ulceration, and GI hemorrhage. Because of this, it has been estimated that 20 to 40% of arthritic patients chronically taking NSAIDs have one or more gastroduodenal ulcers (Fries et al., 1989; Gabriel et al., 1991). The GI side effects of NSAIDs have been linked to both the topical damaging action of these drugs and their ability to inhibit the constitutive COX-1 isoform, which catalyzes the production of prostaglandins in the GI mucosa with established "cytoprotective" activity (Wallace, 1997). The development of the COX-2 selective inhibitors, such as celecoxib (Celebrex) and rofecoxib (Vioxx), were outgrowths of our understanding of the latter property, and recent clinical studies have confirmed that these drugs have lower GI toxicity than conventional NSAIDs when administered both acutely and chronically at equivalent therapeutic doses (Simon et al., 1998; Laine et al., 1999). However, most physicians are taking a conservative approach in prescribing COX-2 selective inhibitors, due to unresolved questions concerning their therapeutic dose range and efficacy in the treatment of chronic inflammatory disorders, their high cost, and preclinical evidence that their administration may exacerbate pre-existing GI ulcers/erosions and delay their healing (Mizuno et al., 1997; Wallace et al., 1998).

Our laboratory has taken an alternative approach in reducing the GI side effects of NSAIDs by attempting to block their topical damaging actions. This approach was rooted in our earlier observations, previously reviewed, that certain regions of the GI tract possessed a hydrophobic barrier to HCl and other toxic agents in the lumen, due to the presence of zwitterionic phospholipids, such as PC within and coating the mucus gel layer (Butler et al., 1983; Hills et al., 1983; Lichtenberger et al., 1983; Lichtenberger, 1995). NSAIDs, in turn, were demonstrated to have the capacity to chemically associate with and destabilize this phospholipid lining, resulting in an attenuation in mucosal surface hydrophobicity (Hills et al., 1983; Goddard and Lichtenberger, 1987; Goddard et al., 1990; Lichtenberger, 1995; Giraud et al., 1999). Furthermore, we demonstrated that this process together with the topical damaging action of aspirin and other NSAIDs could be significantly reduced or prevented if these drugs were preassociated with either synthetic or purified PC prior to administration (Lichtenberger et al., 1995). Evidence that such a chemical interaction between PC and NSAIDs does occur, includes our findings that PC induces alterations in the solubility, melting point, and infrared spectroscopic characteristic of an NSAID (Lichtenberger et al., 1993, 1994, 1995). In addition, using fluorescent probes, we have additional evidence of a chemical association based upon a fluorescent resonance energy transfer experiment and measuring alterations in phospholipid membrane fluidity and hydrophobicity in the presence of an NSAID (Giraud et al., 1999). Both preclinical and recently published clinical studies have demonstrated that PC-NSAIDs induced significantly fewer and less severe ulcerations and/or erosions than conventional NSAIDs, although their COX inhibitory activity was unaffected (Lichtenberger et al., 1995; Anand et al., 1999). One unexpected, although interesting, observation made during the preclinical studies was that the therapeutic efficacy and potency of aspirin and indomethacin appeared to be significantly enhanced when the drugs were administered in the PC-associated state (Lichtenberger et al., 1995, 1996b). Thus, the benefit/risk ratio is remarkably improved, both with regard to increased benefits and decreased risk, by the process of chemically associating the NSAID with PC.

Ibuprofen is an NSAID that is commonly consumed for the treatment of pain and inflammation, the use of which is limited by its low potency, requiring subjects to take doses of up to 1 to 2 g/day to obtain relief of joint pain and inflammation. Although the GI side effects of ibuprofen are considerably less than aspirin and many other NSAIDs on the market, evidence is clear that subjects will be at a significant risk for developing peptic ulceration and bleeding if they chronically take this drug at a high antiarthritic dose range (Laine et al., 1999). Because of this need, we have undertaken this study to evaluate and compare the efficacy and potency of PC-ibuprofen to conventional ibuprofen in rodent models of both acute and chronic joint inflammation/pain. Because rats are very resistant to the GI toxic actions of ibuprofen, we were not able to evaluate the GI toxicity of the ibuprofen test formulations in this study, but instead, we placed our focus on comparing the drugs' anti-inflammatory and analgesic activity.

In our initial studies, we used the acute model of joint inflammation to demonstrate that, in contrast to unmodified ibuprofen that increased the pain threshold of the sensitized hind paw at only the highest dose tested (50 mg/kg), PC-ibuprofen had significant analgesic activity at both 10 and 25 mg/kg, in addition to 50 mg/kg. Based upon this finding, we performed longer-term studies to compare the efficacy and potency of two PC-ibuprofen formulations to that of the unmodified NSAID to treat adjuvant-induced arthritis in female Lewis rats. We decided to evaluate the therapeutic activity of two PC-ibuprofen formulations of contrasting purity, cost, and consistency, with the purified PC preparation (Phospholipon 90G) being water-based and considerably more costly (~100×) than the semipurified oil-based formulation (Phosal 35). All test formulations were administered at doses of 15 and 25 mg/kg, which is between 20 to 50% of the published doses of ibuprofen to reverse arthritogenic changes in rats (Selph et al., 1993; Price et al., 1996). These values agree with the results of our acute inflammation study, where only a dose of 50 mg/kg of unmodified ibuprofen was effective in reversing pressure pain of the sensitized hind paw. We designed our experiments so that the treatment with the test formulations commenced 2 weeks after the rats were challenged with adjuvant, at the initial signs of ankle inflammation. Recently performed behavior studies indicate that the rats experience a decrease in nocturnal ambulatory activity only 1 week after adjuvant administration (data not shown).

The results presented above indicate that ibuprofen's ability to reverse hind limb inflammation and pain sensitivity associated with the development of adjuvant-induced polyarthritis was consistently enhanced if the NSAID was chemically associated with either the purified or semipurified preparation of PC. Indeed, in contrast to the two PC-ibuprofen formulations that proved to have anti-inflammatory and analgesic activity at both 15 and 25 mg/kg, unmodified ibuprofen failed significantly to reverse the previously described arthritogenic changes (in comparison with that of control arthritic rats treated with saline) at either dose.

A number of the experiments presented herein were designed to provide an explanation for the increase in therapeutic efficacy and/or potency when ibuprofen is chemically associated with PC. Pharmacokinetic analysis revealed that PC association altered both the uptake and decay of ibuprofen in the blood. The ibuprofen formulation containing the semipurified PC fraction resulted in a decrease in the initial peak ibuprofen levels of the blood attained 15 min after NSAID administration. However, in contrast to the rapid exponential decay in circulating ibuprofen levels (following the peak) observed in rats that were administered the unmodified NSAID, circulating ibuprofen levels decreased at a slower rate over the remainder of the study period in rats treated with the two PC-ibuprofen formulations. This is due to a 2- to 3-fold increase in the circulating t_1/2 of ibuprofen when administered in the PC-associated state. It also should be noted that the AUC for ibuprofen was modestly increased by 20 to 60% when the NSAID was chemically associated with PC prior to intragastric administration. The reason for these alterations in the pharmacokinetics of the drug in response to PC association and its contribution to enhanced therapeutic activity of the formulation will be the subject of future studies. However, it seems likely that the phospholipid may act as a depot and delay release of ibuprofen from the GI lumen into the blood or alter the clearance of ibuprofen. We have also obtained preliminary results that only a small fraction (<10%) of the intragastrically administered PC molecules enter systemic circulation and presumably remain associated with ibuprofen during the GI absorptive process (not shown). Future studies will attempt to confirm these findings and determine whether the small fraction of PC-ibuprofen molecules in circulation could account for the observed alterations in pharmacokinetics and therapeutic activity.

We also investigated whether PC association may enhance the COX-inhibitory activity or potency of ibuprofen, as a potential explanation for the enhanced therapeutic activity. Analysis of a fluid effusate collected from tissue in and around the inflamed joints of arthritic rats indicated that all the ibuprofen test formulations induced a significant decrease in prostaglandin concentration, with a tendency for the inhibition in COX activity to be greater in rats treated daily with the two PC-ibuprofen formulations. However, the intergroup differences (ibuprofen versus PC-ibuprofen) were not statistically significant. Because both COX isoforms contribute to synovial fluid prostaglandin biosynthesis, we have initiated in vitro studies investigating the effects of ibuprofen and PC-ibuprofen on HUVECs that have been treated with phorbol ester to selectively induce COX-2 expression and activity. These studies indicated that PC-ibuprofen was a more potent inhibitor of COX-2 in these activated cells than unmodified ibuprofen, providing a potential explanation for the enhanced therapeutic activity of the PC-associated NSAID. The molecular basis for this action is presently under study but may be attributable to the fact that the PC-ibuprofen is more membrane permeable, affording a favorable advantage for the associated NSAID to diffuse into target cells and interact with COX on either the endoplasmic reticulum or nuclear membrane.

In summary, it appears that similar to our previously published findings with aspirin and indomethacin, chemical association of ibuprofen with the zwitterionic phospholipid, PC, in either a purified or semipurified state, resulted in an increase in therapeutic effectiveness and potency of the NSAID in both acute and chronic models of joint inflammation. One interpretation of these findings is that chemical association of PC with ibuprofen would allow this commonly consumed NSAID to be administered at a lower dose, further reducing the GI side effects of the drug. The mechanism for this enhancement in therapeutic activity may relate to the NSAID being released into the circulation over a more sustained period of time with a longer circulatory half-life, when administered in a PC-associated state, and its ability to rapidly cross membranes to affect COX enzymes and other intracellular targets of NSAID action.