Materials and Methods

Sample population— Blood samples collected via jugular venipuncture from each of 8 healthy horses were used to provide equine mononuclear cells for the experiments. The use of horses in this study was approved by the University of Georgia Institutional Animal Care and Use Committee.

Aqueous BWE preparation—Aliquots of an aqueous BWE that was used to induce acute laminitis in horses in another study¹⁴ were used in the study reported here. Briefly, the heartwood of a black walnut tree was prepared as uniform shavings that were packaged in 1-kg units and stored at −20°C until used in the study. To prepare the aqueous BWE, 1 kg of shavings was thawed and transferred into 7 L of distilled water and shaken continuously for 12 hours in an orbital shaker. The resulting solution was filtered through cheesecloth, and aliquots were transferred into sterile 50-mL centrifuge tubes and stored at −20°C until further processing for the study.

Before initiating in vitro experiments with equine mononuclear cells, frozen BWE was thawed and samples were pooled and tested for evidence of bacterial contamination. Plating of the pooled sample was performed by use of a series of 10-fold dilutions of BWE on brain-heart infusion agar plates. After samples were removed for bacterial culture, the BWE was centrifuged at 3,000 × g for 20 minutes. The supernatant was removed and filtered through composite glass fiber 0.2-μm membrane filters.^a Samples of the filtered extract were evaluated for evidence of bacterial content by transferring 200-μL aliquots into 3 replicate tubes containing 10 mL of sterile brain-heart infusion broth and by directly plating 100 μL of each aliquot on 3 brain-heart infusion agar plates; both the broth and agar plates were incubated for 48 hours at 37°C. Furthermore, a 100-μL aliquot of each filtrate was analyzed for evidence of LPS-like material by use of a colorimetric Limulus amoebocyte lysate assay.^b The remaining filtrate was stored in 1-, 10-, or 25-mL aliquots at −20°C and was thawed no more than 3 times for use in the in vitro experiments.

Collection of equine mononuclear cells—Blood samples were collected into syringes containing EDTA (final EDTA concentration, 2.5mM). After collection, each syringe was placed with the syringe tip pointing upward and allowed to incubate at 22°C for 20 minutes to allow separation of the cellular components. Then, leukocyte-rich plasma was expressed from each syringe into sterile polystyrene bottles and diluted 1:2 in PBS solution^c that did not contain calcium or magnesium. The diluted cells were layered over 10 mL of mononuclear cell separation medium^d in 50-mL centrifuge tubes and centrifuged at 800 × g for 30 minutes. Mononuclear cells at the interface between the diluted plasma and mononuclear cell separation medium were collected and transferred into 15-mL conical centrifuge tubes.^e Then, mononuclear cells were pelleted by centrifugation at 800 × g for 10 minutes, suspended in PBS solution, quantified by use of a hemacytometer,^e and assessed for viability by use of trypan blue dye^f exclusion.

Assessment of cytotoxicosis induced by BWE—After quantification by use of the hemacytometer, equine mononuclear cells were assessed for viability. Mononuclear cells were maintained at 37°C for 1 hour in complete RPMI that did not contain phenol red^c but that did contain 10% fetal calf serum,^g 2mM L-glutamine,^c and 50 μg of gentamicin sulfate^f/mL. During incubation, BWE solutions were prepared by adding 400 μL of aqueous BWE to 1,600 μL of medium to make a 20% (vol/vol) BWE solution. After vortexing the 20% BWE solution, 1:2 serial dilutions were prepared until a 2.5% BWE solution was achieved. After a 1-hour incubation period, mononuclear cells were pelleted by centrifugation at 800 × g for 10 minutes. The pelleted cells were transferred and suspended in medium at a final concentration of 3 × 10⁶ cells/mL. Then, 1 mL of the mononuclear cell suspension was added to tubes containing 1-mL aliquots of each of the BWE dilutions; this yielded cell culture suspensions in medium containing 1.25% to 10% BWE. One suspension of mononuclear cells in medium, which did not include BWE, was used as a negative control sample. Tubes were incubated at 37°C for 30 minutes. Then, the cells were pelleted by centrifugation at 800 × g for 10 minutes and suspended in fresh medium. Fifty-microliter samples of the mononuclear cells from each treatment were prepared in triplicate and added to three 450-μL aliquots of 0.04% trypan blue solution. Mononuclear cells were counted by use of a hemacytometer, and viability was determined by assessing the ability of the cells to exclude the dye. The total number of cells and the number of viable cells in each sample were recorded.

To assess cellular enzymatic activity, mononuclear cells (3 × 10⁵ cells/well) were incubated in MTT; metabolically active cells convert MTT to an insoluble purple formazan compound. In preliminary experiments, we observed that the purple formazan compound was formed when BWE was incubated with MTT. Therefore, the effect of BWE on mononuclear cells was measured after the cells were exposed to BWE in medium, and then, medium was replaced with fresh medium containing MTT. Four 100-μL replicate wells of each treatment group were added to a standard flat-bottom sterile 96-well tissue culture plate^e and incubated for 4 hours at 37°C. At the end of the incubation period, the plates were centrifuged at 250 × g for 5 minutes and the medium was removed with a 12-channel pipette. The plate was gently vortexed for 3 seconds, and 100 μL of medium and 10 μL of a stock solution of MTT^h were added to each well. The plates were incubated at 37°C for an additional 4 hours. At the end of the incubation period, 100 μL of acid isopropanol^e was added to each well to dissolve crystals and generate a uniform color. Dual values measured at wavelengths of 570 and 630 nm were obtained by use of an ELISA plate reader.ⁱ The mean and SD of the results at each wavelength were calculated, and a differential value (OD 570 to 630 nm) was calculated. To ensure that the color generated was not the result of residual BWE in the wells, duplicate wells that did not contain mononuclear cells were prepared for each treatment.

Because these cells are widely used to assess toxic activity in vitro,¹⁹ the capacity of BWE to act as a toxin on WI-38 cells, which are cells of a human embryonic lung cell line, was assessed. The WI-38 cells were grown to confluence in flasks^e and removed by use of treatment with trypsin.^c The WI-38 cells were suspended in medium at a final concentration of 3 × 10⁵ cells/well and treated and tested via the trypan blue dye exclusion and cellular enzymatic activity assays as described previously for the equine mononuclear cells.

Assessment of the ability of BWE to induce nitric oxide production by mononuclear cells—Nitric oxide is produced as a gas but reacts almost immediately with water to form nitrite. Nitric oxide is measured on the basis of the formation of a colored product from the reaction of nitrite with a Griess reagent in MEM. Furthermore, MEM, rather than RPMI, must be used because MEM does not contain nitrite or nitrate.

Equine mononuclear cells were suspended at a concentration of 1 × 10⁶ cells/mL in MEM that did not contain phenol red^e but that did contain 10% fetal calf serum, 2mM L-glutamine, and 50 μg of gentamicin sulfate/mL. The suspension of cells was incubated for 1 hour at 37°C. Serial 2-fold dilutions ranging from 0.02% to 20% (vol/vol) of BWE were prepared in MEM. In addition, 3 dilutions of Escherichia coli 0111:B4 LPS^j at stock concentrations of 2, 20, and 200 ng/mL and six 2-fold dilutions of hydrogen peroxide^f (6% to 0.2%) were also prepared in MEM. Five hundred microliters of cells was added to each well of a 24-well tissue culture plate,^k and an equal volume of each treatment (ie, BWE, E coli 0111:B4 LPS, or hydrogen peroxide) was added to wells in duplicate; furthermore, 500 μL of MEM was added to a well to serve as a negative control sample. Plates were incubated at 37°C for 72 hours, then centrifuged at 250 × g. The supernatant was collected from each well and stored in 1.5-mL microcentrifuge tubes at −80°C until further processing.

Supernatant samples were thawed, and 80 μL of each was transferred into 4 replicate wells of a 96-well tissue culture plate.^l Nitrate reductase^f was thawed, diluted 1:10,000 in MEM that did not include fetal calf serum, and stored on ice until further use. Ten microliters of the nitrate reductase solution was added to each well, and the plate was incubated at 37°C for 1 hour with agitation every 15 minutes. A standard curve was generated by use of 2-fold dilutions (0.001 to 10 μg/mL) of a stock solution (100 μg/mL) of sodium nitrite. Then, 80 μL of each dilution was transferred to 4 replicate wells. Eighty microliters of fresh Griess reagent^f was added to each well. Each plate was incubated at 22°C for 20 minutes, then the development of the colored product was measured at a wavelength of 550 nm by use of an ELISA plate reader.ⁱ

Assessment of procoagulant activity and TNF-α production induced by equine mononuclear cells exposed to BWE—Five hundred microliters of equine mononuclear cells (concentration, 1 × 10⁷ cells/mL of medium) was added to each well of a 48-well tissue culture plate.^l An equal volume of LPS or BWE, which was diluted as described previously, was added to duplicate wells containing mononuclear cells. Wells that contained only medium as well as wells that contained mononuclear cells in medium without LPS or BWE were used as negative control samples. To assess the contribution of LPS on procoagulant activity, medium containing 13 U of polymyxin B^m/mL was added to wells of a duplicate 48-well tissue culture plate with wells that contained treatments similar to that of the first plate. Both plates were incubated at 37°C overnight and centrifuged at 250 × g, and the supernatants were transferred to 1.5-mL microcentrifuge tubes for storage at −80°C. After thawing, each supernatant was analyzed to determine the TNF-α concentration by use of an ELISA as previously described.²⁰ Mononuclear cells that remained on the tissue culture plates after removal of the supernatant were washed with 250 μL of PBS solution. The PBS solution was removed, 300 μL of a detergent solution (0.1% Tween-20 in PBS solution) was added to each well, and plates were stored at −80°C. After the plates were thawed at 22°C, cell lysates were evaluated for expression of procoagulant activity by use of a standard recalcification clotting assay.²¹ Retained procoagulant activity was calculated by dividing the mean number of units of procoagulant activity (relative to the equine brain reference) from cells treated with polymyxin B by the mean number of units of procoagulant activity (relative to the equine brain reference) from cells not treated with polymyxin B.

Assessment of BWE on induction of apoptosis in equine mononuclear cells—Equine mononuclear cells were suspended in medium (concentration, 1 × 10⁶ cells/mL of medium) and incubated at 37°C for 60 minutes. During the incubation of cells, dilutions of BWE were prepared in RPMI as described previously. After incubation, the cells were mixed gently and 50 μL of the cell suspension was added to each well of a 96-well tissue culture plate. Fifty microliters of RPMI containing each BWE dilution was added to 4 replicate wells containing mononuclear cells to yield final concentrations of BWE that ranged from 10% to 0.6%. Then, each plate was incubated at 37°C for 4 hours. Apoptosis of mononuclear cells was assessed by measuring the aggregate activities of caspases 3 and 7 by use of a common fluorescent substrate assay.ⁿ Briefly, 50 μL of fluorescent substrate was added to each well and plates were incubated in the dark at 22°C for 1 hour. After incubation, fluorescence, which was indicative of the combined activities of caspases 3 and 7, was measured by use of a fluorescent plate reader^o with excitation and emission settings at 485 and 538 nm, respectively.

Statistical analysis—Statistical analyses of the data were conducted by use of a statistical analysis program.^p The induction of procoagulant activity, activities of caspases 3 and 7, and TNF-α production by BWE, LPS, and hydrogen peroxide were compared via a linear best-fit slope comparison by use of a spreadsheet program.^q In addition, induction of procoagulant and caspase activities and TNF-α production by BWE were compared with the activities and production of untreated cells and with those of cells treated with LPS or hydrogen peroxide by use of a 2-tailed t test with Welch's correction^p; a similar analysis was conducted for each concentration of BWE or LPS after treatment with polymyxin B for the induction of procoagulant activity or TNF-α production. A value of P ≤ 0.05 was used to indicate significance in all analyses.

Results

Bacterial and LPS contamination of aqueous BWE—The BWE, as originally prepared, was contaminated with 1,004 bacterial colonies/mL. Bacterial growth was not observed in the brain-heart infusion broth or brain-heart infusion agar plates after incubation for 48 hours at 37°C following filtration of the BWE; however, the filtered preparation contained the equivalent of approximately 50 ng of LPS-like material/mL.

Effect of BWE on viability and metabolic activity of equine mononuclear cells—As indicated by persistence of the exclusion of trypan blue dye, BWE concentrations from 0.6% to 10% had no effect on the viability of mononuclear cells isolated from 6 horses. Wells containing only BWE, but not mononuclear cells, in medium yielded a low and inconsistent amount of MTT conversion, which typically had ODs that ranged from 0.03 to 0.05 and that were always < 0.08. When values for conversion of MTT in wells that did not contain mononuclear cells were subtracted from the activity of the mononuclear cells incubated with BWE, there was no significant effect of BWE on conversion of MTT; furthermore, the OD ranged from 0.35 to 0.62 (mean, 0.48) for all samples tested, which included control samples.

Effect of BWE on nitric oxide production by equine mononuclear cells—In preliminary experiments, we determined that incubation of equine mononuclear cells with LPS resulted in weak production of nitric oxide, generally in the range of 0.2 to 0.5 μg/mL. In contrast, incubation of mononuclear cells with hydrogen peroxide yielded > 6 μg of nitric oxide/mL (Figure 1). As a result, production of nitric oxide by mononuclear cells collected from 4 horses was determined for cells incubated with BWE, LPS, or hydrogen peroxide. Although a significantly greater amount of nitric oxide production was induced by mononuclear cells incubated with hydrogen peroxide at concentrations exceeding 0.2%, the amount of nitric oxide production induced by BWE (< 0.5 μg/mL) was comparable with that induced by LPS and was not significantly different, compared with values for the untreated control cells. Furthermore, the production of nitric oxide induced by BWE was not dose dependent. Because there was evidence of LPS in the BWE, these findings suggested that BWE induced little, if any, measurable nitric oxide activity in equine mononuclear cells.

View Full Size

Mean ± SD values for nitric oxide production by equine mononuclear cells (concentration, 1 × 10⁶ cells/mL) that were collected from 4 horses and incubated in MEM containing 0% to 6% (vol/vol) hydrogen peroxide for 72 hours. Nitric oxide was measured on the basis of the formation of a colored product from the reaction of nitrite with a Griess reagentf in MEM. Development of the colored product was measured at a wavelength of 550 nm by use of an ELISA plate reader,ⁱ and results of nitric oxide production were compared with results for a sodium nitrite standard.

Citation: American Journal of Veterinary Research 72, 3; 10.2460/ajvr.72.3.318

• Download Figure
• Download figure as PowerPoint slide

Induction of procoagulant activity in equine mononuclear cells by BWE—Incubation of mononuclear cells with BWE added to medium at concentrations of 2.5%, 5%, or 10% induced a dose-dependent expression of procoagulant activity that was significantly greater than that in untreated cells. Procoagulant activity was significantly reduced by approximately 50% when 13 U of polymyxin B/mL was added to mononuclear cells in medium. Furthermore, when polymyxin B was added to the medium, the slope of the dose-response curve was decreased by approximately 50%, which suggested that LPS in the BWE preparation acted in an additive manner with the components of BWE that did not bind polymyxin B to induce the procoagulant activity of mononuclear cells (Table 1). The addition of polymyxin B to the medium significantly reduced LPS-induced expression of procoagulant activity by > 90% in mononuclear cell samples collected from all horses. There was no significant expression of procoagulant activity by equine mononuclear cells incubated with hydrogen peroxide; < 250 procoagulant activity units were observed at a hydrogen peroxide concentration of 0.2%, and even fewer procoagulant activity units were observed with the addition of higher concentrations of hydrogen peroxide.

Approximately 22,000 procoagulant activity units were expressed by mononuclear cells incubated with BWE, whereas cells incubated with LPS at 10 and 100 ng/mL or PMA at 10⁻⁷M expressed a mean of 17,000, 35,000, and 92,000 procoagulant activity units, respectively. Analysis of these results indicated that BWE induced approximately one-fourth of the maximal capacity of mononuclear cells to express procoagulant activity, which is similar to the amount induced with PMA, and that a portion of this response appeared to be caused by interactions between mononuclear cells and components of BWE that are independent of those inhibited by the addition of polymyxin B.

Induction of TNF-α production in equine mononuclear cells by BWE—Production of TNF-α by mononuclear cells incubated with LPS, PMA, hydrogen peroxide, and BWE was assessed in cell samples collected from 6 horses. Neither PMA nor hydrogen peroxide consistently induced dose-dependent production of TNF-α. In contrast, both LPS and BWE induced dose-dependent increases in TNF-α concentrations. Approximately equivalent supernatant concentrations of TNF-α were measured over the dose ranges tested for BWE and LPS, with 10% BWE inducing maximal production of TNF-α (Figure 2). The TNF-α concentrations reached a plateau at an LPS concentration of 100 pg/mL for all cell samples tested, with either 100 or 1,000 pg of LPS/mL saturating the TNF-α production capacity of the cells.

View Full Size

Mean ± SD concentrations for TNF-α production by mono-nuclear cells (concentration, 1 × 10⁷ cells/mL) that were collected from 6 horses and incubated in MEM containing 0.6% to 10% (vol/vol) BWE with (black squares) or without (crosses) the addition of 13 U of polymyxin B/mL (A) or in MEM containing 0, 0.01, 0.10, and 1.00 ng of LPS/mL (B). Concentrations of TNF-α were determined by use of an ELISA as described elsewhere.²⁰ The production of TNF-α induced by BWE with or without the addition of polymyxin B was evaluated by fitting results for each treatment to a second-order polynomial curve; the R² for both curves is 0.92.

Citation: American Journal of Veterinary Research 72, 3; 10.2460/ajvr.72.3.318

• Download Figure
• Download figure as PowerPoint slide

Whereas incubation of mononuclear cells with 13 U of polymyxin B/mL reduced LPS-induced production of TNF-α by approximately 95%, simultaneous incubation of mononuclear cells with BWE and polymyxin B reduced the production of TNF-α by 40% to 70%. Shapes of the dose-response curves induced by BWE with or without the addition of polymyxin B were essentially identical, and the TNF-α data for both treatments were fitted to a second-order polynomial curve (R² = 0.92). These findings suggested that most of the TNF-α production induced by BWE was caused by LPS or other components of the extract that were bound by polymyxin B. Finally, it appeared that production of TNF-α induced by BWE and LPS was additive, with approximately 30% of TNF-α production caused by components of BWE that were independent of LPS.

Activation of caspase 3 or 7 (or both) by BWE in equine mononuclear cells—After incubation with BWE for 4 hours, mononuclear cells were assessed for the activities of caspases 3 and 7 by use of a fluorescent substrate assay that yields a composite signal for both enzymes. A strong third-order correlation (R² = 0.849) was observed between the concentration of BWE in the medium and the amount of fluorescent substrate conversion, with all concentrations of BWE tested inducing a significant increase in fluorescence, compared with the fluorescence in untreated cells (Figure 3).

View Full Size

Mean ± SD induction of activities of caspases 3 and 7 in equine mononuclear cells that were collected from 6 horses and incubated in MEM containing 0.6% to 10% (vol/vol) BWE. Caspase activity was measured by use of a fluorescent substrate assay.ⁿ The relationship between the induced caspase activity and concentration of BWE was fitted (R² = 0.88) to a second-order polynomial curve.

Citation: American Journal of Veterinary Research 72, 3; 10.2460/ajvr.72.3.318

• Download Figure
• Download figure as PowerPoint slide

In parallel experiments, capacities of LPS and hydrogen peroxide to induce similar caspase activity were assessed. Whereas a wide range of concentrations of LPS (1 ng/mL to 1 μg/mL) did not significantly increase this caspase activity over the value for the untreated mononuclear cells, hydrogen peroxide caused a significant dose-dependent increase in the activity of these caspase enzymes, compared with the activity in untreated cells at all concentrations tested (Figure 4). At the highest concentration (3%) of hydrogen peroxide tested, the caspase activity measured (350 AFU) was approximately twice that measured (160 AFU) after incubation with BWE at a 10% concentration of the medium.

View Full Size

Comparison of the activation of caspases 3 and 7 in equine mononuclear cells that were collected from 6 horses and incubated in MEM containing 0.2% to 3% hydrogen peroxide. Each data point represents the mean ± SD of 6 pooled measurements of caspase activity for each concentration of hydrogen peroxide. Caspase activity was measured by use of a fluorescent substrate assay.ⁿ The relationship between the activation of caspase activity and hydrogen peroxide concentration was fitted (R² = 0.97) to a second-order polynomial curve.

Citation: American Journal of Veterinary Research 72, 3; 10.2460/ajvr.72.3.318

• Download Figure
• Download figure as PowerPoint slide

Discussion

Analysis of results of the study reported here indicated that BWE preparations commonly used to induce acute laminitis in horses are not sterile and contain LPS. Furthermore, these results indicated that BWE contains components other than LPS that induce procoagulant activity and TNF-α production. In addition, BWE induced activity of caspases 3 or 7 (or both), which is associated with apoptosis, but LPS did not. The induction of apoptosis in leukocytes can alter the regulation of the inflammatory response and may help account for leukopenia observed after intragastric administration of BWE to horses. The findings of the present study indicated that BWE, when used for the induction of laminitis in horses, is capable of directly inducing inflammatory activation.

On the basis of the analysis of findings for the study reported here, it was determined that BWE was not directly toxic to mononuclear cells; incubation of mononuclear cells with concentrations of BWE ranging from 0.6% to 10% did not result in a significant loss of viability or metabolic activity as determined via trypan blue dye exclusion and MTT conversion, respectively. In contrast, medium containing BWE at concentrations > 1.25% caused a significant loss of viability and reduction in metabolic activity in WI-38 cells. Therefore, although BWE contains toxic components similar to those described elsewhere,⁹ it is not directly toxic to equine mononuclear cells in the culture system used in the present study. The BWE did induce dose-dependent activity of the caspase enzymes measured by use of a fluorescent substrate and that may have directly led to removal of some leukocytes from the circulation and induced a change in leukocyte function. At this time, we do not have data that indicate whether specific lineages of leukocytes undergo apoptosis after exposure to BWE.

In another study¹¹ performed by our laboratory group, a hydrogen peroxide-sensitive fluorogenic probe, dihydrorhodamine-123, was used to determine that a preparation of BWE, which had been used to induce acute laminitis in horses, generated free oxygen radicals in aqueous media. In the study reported here, hydrogen peroxide, but not BWE or LPS, strongly induced the production of nitric oxide by equine mononuclear cells. Investigators in another study¹⁰ reported that a family of nitroxide compounds, which function as single electron spin traps specifically for 1-oxyl-,2,6,6-tetramethyl-4-hydroxypiperidine (ie, TEMPOL), 4-amino-2,2,6,6-tetramethyl-1-piperidinyloxy (ie, TEMPAMINE), and 2-cyclohexane-5,5-dimethyloxazolidine-l-oxyl (ie, CHDO), modulate the toxic and radical-generating activity of juglone isolated from black walnut. Juglone reacts with water to form superoxide radicals and hydrogen peroxide. Nitroxide molecules, including nitric oxide, recycle juglone to its radical-generating form and thereby deplete nitroxide compounds from the system. Thus, nitric oxide produced by mononuclear cells in response to hydrogen peroxide generated by BWE may be consumed by BWE in the medium before it can be converted into nitrite by reacting with water in the medium. Because nitrite is the reaction product measured by the Griess reagent in the assay used for nitric oxide measurement, it is possible that nitric oxide is both produced and consumed in mononuclear cell cultures containing high concentrations of BWE. Furthermore, it appears that cell cultures that do not include BWE allow hydrogen peroxide, when added directly to the cell cultures, to induce the production of nitric oxide that can be measured by use of the Griess reagent.

In the present study, hydrogen peroxide and BWE, but not LPS, induced caspase activity in mononuclear cells; the induction of caspase activity is indicative of apoptosis. In the aforementioned study,¹⁰ investigators reported that the interaction of CHDO, a lipid containing nitroxide probe, with juglone was more effective than either of the polar probes (ie, TEMPOL and TEMPAMINE) in preventing the killing of bacterial cells. Thus, it was proposed that juglone and related toxins exerted their effects by altering membranes and membrane-related cellular energetics,¹⁰ which may be the mechanism responsible for the increase in the activities of caspases 3 and 7 detected in the present study. The production of radical oxygen species and membrane damage can both lead to the induction of apoptosis.²²

In the present study, expression of procoagulant activity and production of TNF-α by mononuclear cells were induced by LPS and BWE but not by hydrogen peroxide. Furthermore, the increased expression of procoagulant activity and production of TNF-α that developed after incubation of the mononuclear cells with BWE were only partly inhibited by the addition of polymyxin B to the cell cultures. Because LPS-induced production of procoagulant activity and TNF-α was completely blocked by polymyxin B, these findings suggested that BWE preparations contain components other than LPS that are capable of activating equine mononuclear cells. One such set of BWE components appears to be those that generate hydrogen peroxide.¹⁰ Although these radical generators are important for the induction of apoptosis that is independent of LPS, other components of BWE may be involved in the induction of proinflammatory mediators, such as procoagulant activity and TNF-α production; in addition, those BWE components appeared to exert their effects independent of LPS.

On the basis of the analysis of results obtained in the present study, BWE preparations used to induce laminitis in horses contained LPS that broadened the inherent proinflammatory effects of BWE on equine mononuclear cells in vitro. It should be recognized that the quantity of LPS contained in these BWE preparations was extremely small relative to the amount represented by the intestinal flora of horses. Thus, LPS in the BWE preparation is not likely to be a major source in vivo.

Analysis of results of the study reported here indicated that the direct effects of BWE on equine mononuclear cells were similar to the reported findings of proinflammatory activation in horses developing Obel grade 1 laminitis after intragastric administration of BWE.^6,16,23 Furthermore, results of the present study and these other studies^6,16,23 indicate that superoxide and hydrogen peroxide production resulting from the interaction between components of BWE and water may play an important role in the pathogenesis of laminitis. In addition, components of BWE other than those associated with direct radical generation appear to interact with mononuclear cells to induce proinflammatory events through the production of the proinflammatory mediators, procoagulant activity, and TNF-α. Collectively, the findings of the present study support a role for interactions between components of BWE and mononuclear cells in the pathogenesis of laminitis induced by the intragastric administration of BWE.