Recurrent airway obstruction is an environmental disease of mature horses characterized by periods of reversible airway (bronchi and bronchioles) obstruction associated with neutrophil accumulation in airway lumens, mucus production, and bronchospasm.1,2 Similar to human asthma, RAO involves airway hyper-responsiveness, reversible airway narrowing, and mucus hypersecretion.2 However, whereas an increase in percentage of airway eosinophils is commonly associated with exacerbation of human asthma, RAO-affected horses develop airway neutrophilia in conjunction with submucosal infiltrates composed of lymphocytes, mast cells, plasma cells, and occasionally eosinophils.2,3
A large volume of literature exists in human research regarding oxidative stress, and this phenomenon is now widely recognized as an important component of the airway inflammation in asthma.4 Oxidative stress occurs in tissues and organs when ROS are produced in excess of antioxidant defense mechanisms, resulting in harmful effects.5,6 These effects are mediated by direct actions of ROS on target cells in the airways and also indirectly via activation of signal transduction pathways and transcription factors.6 The resultant effects include increased inflammation, cell apoptosis, systemic oxidative stress, and steroid hormone unresponsiveness.6 Recently, there has been interest in the detection and use of markers of oxidative stress in horses to determine whether oxidative damage is involved in the tissue injury associated with RAO, with potential for use in the diagnosis and monitoring of this disease.7–10 The airspace epithelial surface of the lung is particularly vulnerable to oxidant damage as it directly contacts the environment.11,12 Activated inflammatory and structural cells within the airways produce ROS and cause cell damage in a self-perpetuating process.5,6,13 Oxygen radicals also mediate various other processes including increased mucous production, decreased cilia function, fibroblast injury, increased polymorphonuclear cell infiltration, gene expression of proinflammatory mediators, and impaired pulmonary function and lung repair.6,14,15
Most animal tissues that use oxygen possess enzymatic and nonenzymatic cellular defenses (ie, antioxidants) that help prevent free radical–mediated injury16,17 Enzymatic antioxidants include superoxide dismutase, catalase, GSH peroxidase, GSH S-transferase, and thioredoxin.18 Nonenzymatic antioxidants include low-molecular-weight compounds such as GSH, ascorbate, urate, α-tocopherol, bilirubin, and lipoic acid and high–molecular-weight compounds such as the proteins albumin and transferrin.18 When these cellular defenses are overwhelmed, excessive ROS production causes structural cellular damage through the oxidation of proteins, lipids, and DNA.4 Additional cellular pathways can also be triggered that compound the cellular injury.4
A tripeptide thiol, GSH is a vital intra- and extracellular protective antioxidant, the activity of which is markedly upregulated in the presence of oxidative stress.6,19,20 Reduced GSH activity acts as a substrate to detoxify peroxides and in conjunction with cGPx, results in the generation of oxidized cGPx (GSSG).19 The redox cycle is completed by GSSG reduction back to GSH by GSH reductase.12,19 Maintenance of a high (> 90%) intracellular ratio of GSH activity to GSSG concentration provides a reducing environment within the cell and represents one of the most important antioxidant defense systems in the pulmonary system.12,19 In RAO-affected horses during clinical respiratory crisis induced by exposure to dusty straw and feeding of moldy hay, significant increases in GSSG concentration and GSH redox ratio (GSSG concentration divided by the sum of GSH concentration and GSSG concentration) reportedly develop.8,21 Changes in this antioxidant defense system are also reflected in the activity of cGPx. Such activity has been measured in horses and is reportedly significantly higher in BALF from RAO-affected versus -nonaffected, healthy horses.22
A major pulmonary nonenzymatic redox system involves AA. Ascorbate is oxidized (ie, it accepts free radical electrons) to an ascorbyl radical and then to dehydroascorbate; dehydroascorbate subsequently rapidly breaks down into formoxalic acid and l-threonine.18 Quantitatively, horses have a greater nonenzymatic antioxidant capacity in the pulmonary epithelial lining fluid than do humans because of high AA concentrations, reflecting the ability of horses to synthesize AA.23 A greater oxidative load and a more prolonged inflammatory response are probably required to deplete the antioxidant capacity and cause oxidative stress.23 Compared with healthy horses, RAO-affected horses reportedly have lower AA concentrations in plasma and BALF at rest and during airway inflammation.10,24
To the authors' knowledge, concurrent measurement and comparison of the GSH and AA redox systems in RAO-affected horses has not been reported. The objective of the study reported here was to determine whether differences existed between these 2 major redox systems in airway tissues and fluids from RAO-affected and -nonaffected horses before, during, and after environmental challenge with dusty straw and hay. Bronchoalveolar fluid was collected to evaluate the local pulmonary environment by determining AA concentrations and cGPx activity. Peripherally obtained RBCs and WBCs as well as BALF cells were analyzed for cGPx activity to determine whether any differences existed between systemic and local pulmonary responses of this redox system.
Materials and Methods
Horses—Sixteen horses in pairs were used for the study: 8 horses not affected by RAO (control horses) and 8 horses with a history of RAO (RAO-affected). The control horses were selected from donation or teaching animals that had no clinical signs or history of respiratory disease, were deemed healthy during physical examinations, and did not develop RAO when housed in a barn. The RAO-affected horses were selected from a preexisting herd housed at the university that had a previous diagnosis of RAO. Diagnosis of RAO in these horses had been confirmed after a thorough review of their history and thorough diagnostic testing that included a complete physical examination, hematologic analysis, serum biochemical analysis, transendoscopic tracheal aspiration, BALF collection and analysis, thoracic ultrasonography, thoracic radiography, and pulmonary function testing. The selected horses were also known to consistently develop acute exacerbation of RAO when placed in a barn on straw bedding and fed moderately dusty alfalfa hay. This exacerbation was reversible when horses were returned to a pasture environment. The study protocol was approved by the Virginia Tech Institutional Animal Care and Use Committee.
Experimental protocol—Control and RAO-affected horses were maintained on pasture for a minimum of 8 weeks to ensure the RAO-affected horses were in complete clinical remission and to minimize exposure to respirable debris prior to collection of the first set of samples (SI; remission). In addition to pasture grazing, all horses were fed 1.8 kg of pelleted feed twicea daily.
Environmental challenge was induced 1 week later when horses were brought into the barn, housed in pairs (1 control and 1 RAO-affected horse), provided with dusty straw as bedding, and fed dusty hay ad libitum. In addition to hay, all horses were fed 1.8 kg of the same pelleted feed twice daily. This was continued until the RAO-affected horses reached a minimal clinical score of 5/8 or had stayed in the barn for > 72 hours, at which time the second set of samples (S2; challenge) was collected from the RAO-affected horse and its control horse. Horse pairs were subsequently returned to a pasture environment.
The third set of samples (S3; recovery) was collected after horses had been returned to pasture and the clinical score decreased 2 points or after 1 week on pasture, whichever happened earlier. In addition to pasture grazing, all horses were fed 1.8 kg of the same pelleted feed twice daily
Sample collection and assessment of clinical status—Once the study began, horses were physically examined and assigned a clinical RAO score twice daily (Appendix). Sample times for each horse pair were determined on the basis of the associated results2 and included collection of blood and BALF samples, as well as measurement of ΔPplmax.
Whole blood was collected into heparinized tubes via jugular venipuncture. For BALF sample collection, horses were sedated via IV administration of butorphanol tartrateb (0.01 mg/kg) and detomidine hydrochloridec (0.01 mg/kg). The nares of each horse were cleansed with gauze sponges before a sterile commercial BALF tubed was advanced blindly through the nasal passages and trachea. The tube was advanced while 30 mL of 2% sterile lidocaine solution was instilled into the bronchi and bronchioles until the tube became wedged. The cuff of the tube was inflated with 10 mL of room air, and three 100-mL aliquots of 37°C sterile saline (0.9% NaCl) solution were infused and aspirated. The aspirated fluid was pooled into a sterile specimen container and placed on ice. Collected BALF was centrifuged at 400 × g for 10 minutes, and samples containing BALF cells and cell-free supernatant were removed and frozen at −80°C until analyzed.
Measurement of ΔPplmax during tidal breathing was also made, as described elsewhere.25 Briefly, an esophageal balloon (length, 10 cm; perimeter, 3.5 cm; wall thickness, 0.06 cm) was sealed over the end of a polypropylene catheter (internal diameter, 3 mm; external diameter, 4.4 mm). The tubing was passed into the distal third of the esophagus and attached to a low-range differential pressure transducere calibrated before each study by means of a water manometer. The position of the esophageal balloon was adjusted to obtain the ΔPplmax during tidal breathing. Pleural pressure during breathing was obtained by use of a lung-function computer. At each data collection point, values of at least 15 consecutive breaths were averaged.
Cytologic assessment of BALF—Differential cell count was performed on slides prepared by use of a standard protocolf and stained with a modified Wright stain. At least 400 cells in each specimen were counted by use of immersion microscopy.
Measurement of cGPx activity in RBCs, WBCs, and BALF cells—Separation of peripheral blood cell populations within blood samples was performed, and cells were lysed as described elsewhere.26 Cells in BALF samples were lysed by freezing at −80°C and subsequent thawing. The cGPx activity of RBCs, WBCs, and BALF cells was determined by use of a colorimetric assay kit.g Once cGPx activity per milliliter of sample was determined, the value was adjusted through protein determination.h
Measurement of AA concentration—Standards of AA (0.1, 0.25, 0.5, 0.75, 1, 2.5, and 5 ppm) were prepared in 10% MPA. To 100 μL of each concentration, 234 μL of 10% MPA was added to ensure the same dilution factor as in BALE10 Two 1,000-μL aliquots of cell-free BALF (each duplicated) were used. The first aliquot was diluted with 234 mL of 10% MPA and analyzed by use of high-performance liquid chromatography for concentrations of unoxidized AA (AAfree).27,28 The second was reduced with 67 μL of dithiothreitol (1.5 mg/mL) in 0.2mM sodium phosphate (pH, 7.2).10 This mixture was kept at 25°C for 15 minutes and then quenched with 167 μL of 10% MPA. Chromatography analysis yielded the AAtotal concentration. The concentration of the oxidized form of AA, DHA, was determined by subtracting the AAfree value from the AAtotal value.
Statistical analysis—Data were analyzed by use of mixed-model repeated-measures ANOVA to test whether disease status (RAO affected or nonaffected) and sample collection time (S1, S2, and S3), as well as their interactions, had any effect on each of the indices (clinical score, ΔPplmax, BALF cytologic findings, AA concentration, and cGPx activity). Each pair of horses constituted a random block in the model. Data for BALF cGPx concentration were logarithmically transformed prior to analysis. The AA concentration and cGPx activity for RBCs and WBCs are reported as least square means ± SD to adjust for the multiple covariates within the model. Geometric least square means ± SD are reported for logarithmically transformed values of BALF cGPx activity. To analyze the effect of clinical score on ΔPplmax and results of BALF cytologic assessment, the mixed-model repeated-measures ANOVA was modified to include clinical score as a covariate. Data for neutrophil composition was logarithmically transformed prior to analysis. All analyses were performed with statistical software,i and values of P < 0.05 were considered significant.
Results
Horses—Ages ranged from 7 to 24 years (median, 13.5 years) for control horses and 10 to 25 years (median, 18 years) for RAO-affected horses. There were 3 geldings and 5 mares in the control horse group and 4 geldings and 4 mares in the RAO-affected group. The control group had even numbers of Quarter Horses and Thoroughbreds. The RAO-affected group consisted of 2 Tennessee Walking Horses, 2 Appaloosas, 1 Paint, 1 Quarter Horse, 1 warmblood horse, and 1 pony.
Clinical scores, ΔPplmax, and results of BALF cytologic assessment—All RAO-affected horses achieved a minimum RAO clinical score of 5 within 72 hours after introduction to the environmental challenge, at which point S2 was collected. A reduction in clinical score by 2 points was achieved within 1 week after horses were returned to pasture, at which point S3 was collected (Table 1). Data for ΔPplmax were unavailable for all sample times for 1 control horse because the horse resisted the measurement procedure. Cytologic analysis of BALF could not be performed for 1 RAO-affected horse at S2 because of inadequate BALF recovery.
Mean ± SD clinical scores, ΔPplmax, and results of BALF cytologic analysis of horses with (RAO affected; n = 8) and without (control; 8) RAO before (S1), during (S2), and after (S3) environmental challenge with dusty straw and hay.
| Horse group | Sample | Clinical score | ΔPplmax (cm H20) | BALF cells (cells/mL) | Neutrophils (%) | Lymphocytes (%) | Macrophages (%) | Eosinophils (%) |
|---|---|---|---|---|---|---|---|---|
| Control | S1 | 2.0 ± 0.0 | 3.9 ± 0.9 | 7.0 ± 2.6 | 1.6 ± 1.3 | 47.9 ± 9.9 | 50.4 ± 9.4 | 0.0 ± 0.0 |
| S2 | 2.1 ± 0.2 | 4.3 ± 1.3 | 6.9 ± 3.7 | 13.9 ± 7.7 | 40.1 ± 17.9 | 33.9 ± 16.0 | 0.4 ± 0.7 | |
| S3 | 2.1 ± 0.2 | 4.3 ± 2.1 | 9.7 ± 3.7 | 3.3 ±1.4 | 47.5 ±11.0 | 49.1 ± 11.2 | 0.1 ± 0.4 | |
| RAO affected | S1 | 2.3 ± 0.4 | 7.1 ± 3.4 | 4.2 ± 2.0 | 8.6 ± 7.6 | 49.4 ± 8.2 | 41.9 ± 4.2 | 0.1 ± 0.4 |
| S2 | 5.9 ± 1.0 | 26.3 ± 8.6 | 6.1 ± 4.5 | 31.9 ±28.4 | 37.9 ± 18.2 | 30.3 ± 14.5 | 0.0 ± 0.0 | |
| S3 | 3.5 ± 1.0 | 10.9 ± 7.2 | 4.9 ± 2.6 | 6.6 ± 4.6 | 57.4 ± 14.4 | 36.4 ± 12.2 | 0.0 ± 0.0 |
Data for ΔPplmax were unavailable for all sample times for 1 control horse because the horse resisted the measurement procedure. Cytologic analysis of BALF could not be performed for 1 RAO-affected horse at S2 because of inadequate BALF recovery.
In control horses, no significant associations were detected between clinical scores and ΔPplmax (P = 0.09) or BALF cytologic values (P = 0.27 for total cell concentration; P = 0.32 for neutrophil percentage among BALF cells). There was a positive linear relationship between clinical score and ΔPplmax in RAO-affected horses (P = 0.01). However, in RAO-affected horses, there was no significant association between clinical score and total cell concentration (P = 0.74) or neutrophil percentage (P = 0.32).
cGPx activity—Least square means for WBC cGPx activity in control horses were 58.04, 52.27, and 60.27 mU/mg of protein and in RAO-affected horses were 77.42, 72.81, and 70.39 mU/mg of protein at S1, S2, and S3, respectively. Least square means for RBC cGPx activity in control horses were 59.71, 57.61, and 62.34 mU/mg of protein and in RAO-affected horses were 91.01, 91.40, and 96.16 mU/mg of protein at S1, S2, and S3, respectively.
In WBCs, cGPx activity was affected by horse disease status (RAO vs no RAO; P = 0.01) but not by sample collection time (P = 0.20). The RAO-affected horses had significantly higher WBC cGPx activity than control horses at S1 (P = 0.01) and S2 (P = 0.01) but not at S3 (P = 0.17; Figure 1). In RBCs, cGPx activity was affected by horse disease status (P = 0.02) but not by sample collection time (P = 0.21). The RAO-affected horses had significantly higher RBC cGPx activity at all sample times (P = 0.03 at S1, P = 0.02 at S2, and P = 0.02 at S3; Figure 2).
Box and whisker plots of cGPx activity in peripheral WBCs from horses with (RAO affected; n = 8; gray bars) and without (control; 8; white bars) RAO before (S1), during (S2), and after (S3) environmental challenge with dusty straw and hay. The top and bottom edges of each box represent 1 SD, the central horizontal line represents the least square mean, and the whiskers represent the minimum and maximum values. *Value is significantly (P < 0.05) greater than the corresponding value for control horses.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1500
Box and whisker plots of cGPx activity in peripheral RBCs from horses with (RAO affected; n = 8; gray bars) and without (control; 8; white bars) RAO before (S1), during (S2), and after (S3) environmental challenge with dusty straw and hay. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1500
Geometric least square means for BALF cGPx activity in control horses were 0.06, 0.16, and 0.15 mU/mg of protein and in RAO-affected horses were 0.05, 0.10, 0.15 mU/mg of protein at S1, S2, and S3, respectively Alterations in BALF cGPx activity between sample collection times within each horse group were significant (P = 0.004); however, there was no significant (P = 0.45) difference between the 2 groups. The RAO-affected horses had significantly lower BALF cGPx activity at S1 versus S3 (P = 0.01). Differences between S2 and S3 (P = 0.10) or S1 (P = 0.57) in RAO-affected horses were not significant. Control horses had significantly (P = 0.002) higher BALF cGPx activity at S2 than at S3, and this difference approached significance (P = 0.06) when the S1 value was compared with the S2 value (Figure 3). Differences between S1 and S3 values were not significant (P = 0.53) in control horses.
Box and whisker plots of cGPx activity in BALF cells from horses with (RAO affected; n = 8; gray bars) and without (control; 8; white bars) RAO before (S1), during (S2), and after (S3) environmental challenge with dusty straw and hay. *Value is significantly (P < 0.05) greater than the S3 value for control horses. †Value is significantly (P < 0.05) greater than the S1 value for RAO-affected horses.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1500
BALF AA concentration—Least square means for BALF AAfree concentration in control horses were 9.47, 5.08, and 7.95mM and in RAO-affected horses were 12.12, 3.33, and 9.24mM at S1, S2, and S3, respectively (Figure 4). The AAfree concentration was affected by sample collection time (P = 0.002) but not by horse disease status (P = 0.70). The RAO-affected horses had a significantly lower AAfree concentration at S2 than at S1 (P = 0.002) or S3 (P = 0.04). There was no significant (P = 0.41) difference between S1 and S3 values of AAfree in RAO-affected horses, nor was there a difference in values between pairs of sample times in control horses (S1 vs S2, P = 0.19; S1 vs S3, P = 0.79; and S2 vs S3, P = 0.44).
Box and whisker plots of BALF AAfree concentration ([AAfree]) in horses with (RAO affected; n = 8; gray bars) and without (control; 8; white bars) RAO before (S1), during (S2), and after (S3) environmental challenge with dusty straw and hay. *Value is significantly (P < 0.05) greater than the value for RAO-affected horses at S2.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1500
Least square means for BALF DHA concentration in control horses were 5.62, 4.67, and 5.23mM and in RAO-affected horses were 5.51, 4.74, and 6.99mM at S1, S2, and S3, respectively (Figure 5). Effects for sample collection time (P = 0.17) and horse disease status (P = 0.31), as well as their interactions (P = 0.36), were not significant for BALF DHA concentration.
Box and whisker plots for BALF cell DHA concentration ([DHA]) in horses with (n = 8; gray bars) and without (8; white bars) RAO before (S1), during (S2), and after (S3) environmental challenge with dusty straw and hay.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1500
Least square means for BALF AAtotal concentration in control horses were 15.07, 9.77, and 13.18mM and in RAO-affected horses were 17.63, 8.13, and 16.23mM at S1, S2, and S3, respectively (Figure 6). The AAtotal concentration was affected by sample collection time (P < 0.001) but not by horse disease status (P = 0.53). The RAO-affected horses had significantly lower AAtotal concentrations at S2 than at S1 (P = 0.002) and S3 (P = 0.009). There was no significant (P = 0.83) difference between S1 and S3 values of AAtotal in RAO-affected horses, nor were there differences in values between pairs of sample collection times in control horses (S1 vs S2, P = 0.12; S1 vs S3, P = 0.73; and S2 vs S3, P = 0.37).
Box and whisker plots for BALF cell AAtotal concentration ([AAtotal]) in horses with (RAO affected; n = 8; gray bars) and without (control; 8; white bars) RAO before (S1), during (S2), and after (S3) environmental challenge with dusty straw and hay. See Figure 4 for remainder of key.
Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1500
Discussion
In the present study, 2 cellular redox systems (AA and cGPx) were evaluated in and compared between horses with and without a history of RAO. The experimental protocol included sample collection times based on clinical scores and not ventilatory mechanics. Clinical scoring can identify changes in respiratory function in RAO-affected horses, although there is some overlap of lung function between scores.25
The ΔPplmax values and BALF cellular composition, notably the increase in neutrophil percentage in RAO-affected horses during environmental challenge with dusty straw and hay, were consistent with recorded values for control and RAO-affected horses in other studies25,29 in which a similar scoring system was used. We detected a significant linear relationship between clinical score and ΔPplmax, consistent with other reported findings,25 but not between clinical score and BALF neutrophil composition.25 This might have reflected the fact that our study was not designed to test whether a linear relationship (or correlation) existed between clinical score and any of the BALF characteristics. Collection of samples at multiple, evenly spaced time points, including at times when horses had high clinical RAO scores, would be required to accurately evaluate these relationships.
In the present study, RAO-affected horses had significantly increased RBC cGPx activity at all measurement points, compared with control horses. This finding was in contrast to that of a previous study22 that failed to reveal a difference in RBC cGPx activity between 14 RAO-affected horses and 11 control horses. In that study, the high degree of systemic oxidative stress associated with RAO resulted in less superoxide dismutase activity but no change in RBC cGPx or catalase activity, compared with values in control horses. Increases in GSSG concentration and GSH redox ratio with no change in GSH concentration, however, have been detected in hemolysate of RAO-affected horses during remission and respiratory crisis, compared with values in control horses.8,21 This increase in GSSG concentration might be explained by the increase in RBC cGPx activity of RAO-affected horses detected in our study, since cGPx is responsible for generation of GSSG from GSH.
Although increases in RBC cGPx activity have been detected in humans and rats with non–asthma or non–chronic obstructive pulmonary disease–associated lung injury, studies30–32 of human smokers or humans with chronic obstructive pulmonary disease and asthma consistently reveal decreases in cGPx activity, with increases predicting return to function. The decreases in RBC cGPx activity detected in RAO-affected horses might reflect differences in the disease process between species or an enhanced capacity to combat oxidative stress. No significant reduction during challenge (S2) was detected in our study, suggesting that RAO-affected horses continued to have high RBC cGPx activity, which did not diminish despite the existence of acute disease. Although the degree of RBC cGPx activity was sustained during challenge, it is unknown whether this response was adequate to ameliorate the proportional increase in airway inflammation and ROS generation. Because no additional increase in RBC cGPx activity was evident when compared with values at remission (S1) and recovery (S3), this might have indicated that activity of this redox system is maximally upregulated at all times in RAO-affected horses and might not have been capable of any additional increase in function during the challenge.
The lack of difference in RBC cGPx activity between sample collection times suggested that RBC cGPx activity might not have been associated with disease severity. Alternatively, it might have indicated that the RAO-affected horses were not in remission at the start of the study (S1) or had not recovered sufficiently from the challenge by the last sample collection time (S3) for any decrease in activity to be detected. There is evidence that the clinical scoring system is an insensitive method for detecting low-grade airway obstruction.25 Therefore, our use of clinical scoring to categorize RAO-affected horses as being in remission or recovery from RAO might not have been accurate and it might have been preferable to use disease staging by use of cytologic characteristics instead. Evaluation of samples collected over a more extensive period before and after challenge, ideally with increased numbers of horses, would be required to determine the cause and extent of the sustained increased in RBC cGPx activity we observed. In addition, if RBC cGPx activity were measured in RAO-affected horses and values were comparable with those in control horses, then this might indicate a poorer prognosis because of the horses' apparent incapacity to manage oxidative stress.
A similar polarity in response in RAO-affected horses versus asthma-affected humans was detected when WBC cGPx activity was measured in the present study, with RAO-affected horses having significantly higher WBC cGPx activity at remission and challenge than control horses. Increased WBC cGPx activity has been detected in scuba divers and cyclists as an adaptive response to avoid oxidative damage and is enhanced when human athletes receive supplemental antioxidants.33,34 It is unlikely that administering antioxidants to horses will lead to an increase in WBC cGPx activity if activity is already maximally upregulated. However, when enzyme activity is limited by the availability of substrate, then antioxidant supplementation might be of benefit. Additional studies are required to evaluate the effects of antioxidant administration during all study stages. If WBC cGPx activity were measured in RAO-affected horses and values were comparable with those in control horses, this might indicate a poorer prognosis or a potential for enhancement through antioxidant supplementation.
Within the study groups, RAO-affected horses had significantly higher BALF cGPx activity during recovery than in remission. In comparison, control horses had significantly higher BALF cGPx activity at challenge than during recovery, indicating a more timely and effective response to airway irritants and inflammation. This increase in BALF cGPx activity appeared to be independent of an increasing percentage of neutrophils in BALF because no significant linear relationship was found with clinical score, which was used in the study design to determine the timing of sample acquisition. This might have indicated that increased cGPx activity was required for recovery and that the local response to oxidative stress was submaximal or ineffective in RAO, with a substantial time lag relative to the blood cellular response. Alternatively, because the present study was not designed to test linear relationships or correlations between BALF characteristics and redox systems, increased cGPX activity might have been associated with an increase in neutrophil numbers. In humans, BALF cGPx activity is variable and dependent on the type of airway irritant.35 Because specific airway irritants are difficult to isolate in horses, it is possible that the nature of the environment might influence study results, particularly with respect to the composition of contaminants in local air.
Unlike humans, horses can synthesize AA, and AA appears to be the primary antioxidant defense against inflammatory cell-derived oxidants during an acute inflammatory response in the lung.23,36 Concentrations of AA in BALF (AAtotal, AAfree, and DHA) in the present study were not significantly different in control horses, but AAtotal and AAfree values were significantly lower during challenge in RAO-affected horses than during remission or recovery. Similar reductions have been reported in RAO-affected horses after environmental challenge and with airway inflammation.10,36 Concomitant increases in the concentration of the oxidized form of AA (dehydroascorbate) might not be evident because of uptake by epithelial cells or rapid breakdown.18,23,36 Dehydroascorbate also does not appear to be recycled back to AAfree by cells within the epithelial lining fluid.36
Clear evidence exists that AA depletion and oxidative stress occurs in RAO-affected horses.7,8,10,21,36 This is in contrast to the increases in cGPx activity detected in the various biological samples collected in our study, which might have reflected the greater capacity for the GSH redox system to be upregulated with greater rapidity or magnitude than the AA system. There also might be a difference between the magnitude and speed at which the nonenzymatic antioxidant (AA) and enzymatic antioxidant (cGPx) systems can respond to challenge. It is possible that the enzymatic pathways can be upregulated faster but that acquisition of substrate for nonenzymatic pathways occurs more slowly or is depleted more rapidly.
Study horses were housed in pairs and fed identically to minimize dietary influences on AA availability Plasma concentrations of AA were not measured, but a prior study10 established that RAO-affected horses had significantly lower plasma AAtotal concentrations than control horses, regardless of whether airway inflammation existed. It has also been demonstrated that dietary supplementation with AA derivatives can increase pulmonary and systemic concentrations of AA in healthy horses despite their ability to synthesize this vitamin.24,37,38 These findings support the need for investigation into AA and other antioxidant supplementation as a means of preventing, ameliorating, or treating pulmonary disease.
In the study reported here, cGPx activity in RBCs, WBCs, or BALF cells could not consistently distinguish between remission, challenge, and recovery in RAO-affected horses. However, significant increases in cGPx activity in RBCs at all sample collection times and in WBCs at remission and challenge suggested systemic upregulation of this redox system in RAO-affected horses relative to the status in control horses. Differences in BALF cGPx activity were not identified between the 2 groups, but the delayed increase in activity detected in RAO-affected versus control horses suggested a less effective local response to oxidative stress in horses with a history of RAO.
Abbreviations
| AA | Ascorbic acid |
| AAfree | Free ascorbic acid |
| AAtotal | Total ascorbic acid |
| BALF | Bronchoalveolar lavage fluid |
| cGPx | Cellular glutathione peroxidase |
| DHA | Dehydroascorbic acid |
| ΔPplmax | Maximal change in pleural pressure |
| GSH | Glutathione |
| GSSG | Glutathione disulfide |
| MPA | Metaphosphoric acid |
| RAO | Recurrent airway obstruction |
| ROS | Reactive oxygen species |
Reliance 12P HF, Southern States Cooperative Inc, Richmond, Va.
Fort Dodge Laboratories, Fort Dodge, Iowa.
Pfizer, Exton, Pa.
Bivona tube, Bivona Medical Technology, Gary, Ind.
Validyne, model DP/45–28, Validyne Engineering Corp, Northridge, Calif.
Cytospin, Thermo, Waltham, Mass.
Bioxytecho GPx-340, Oxis International Inc, Portland, Ore.
Bio-Rad DC Protein Assay, Bio-Rad Laboratories, Hercules, Calif.
SAS, version 9.1.3, SAS Institute Inc, Cary NC.
Appendix
Scoring system used to rate degree of RAO in horses.25
| Characteristic | Score |
|---|---|
| Abdominal respiratory effort | |
| No abdominal componentto breathing | 1 |
| Slight abdominal component | 2 |
| Moderate abdominal component | 3 |
| Severe, marked abdominal component | 4 |
| Nostril flaring | |
| No flaring | 1 |
| Slight, occasional flaring of nostrils | 2 |
| Moderate nostril flaring | 3 |
| Severe, continuous flaring during each respiration | 4 |
| RAO clinical score (combination of above scores) | |
| No signs | 2 |
| Mild signs | 3–4 |
| Moderate signs | 5–6 |
| Severe signs | 7–8 |
References
- 1
Ainsworth DM, Hackett RP. Disorders of the respiratory system. In: Reed SM, Bayly WM, Sellon DC, eds. Equine internal medicine. 2nd ed. Philadelphia: Saunders, 2004;289–354.
- 3
Naylor JN, Clark EG, Clayton HM. Chronic obstructive pulmonary disease: usefulness of clinical signs, bronchoalveolar lavage, and lung biopsy as diagnostic and prognostic aids. Can Vet J 1992; 33:591–598.
- 4↑
Riedl MA, Nel AE. Importance of oxidative stress in the pathogenesis and treatment of asthma. Curr Opin Allergy Clin Immunol 2008; 8:49–56.
- 5
Macnee W & Rahman I. Is oxidative stress central to the pathogenesis of chronic obstructive pulmonary disease? Trends Mol Med 2001; 7:55–62.
- 7
Kirschvink N, Art T, de Moffarts B, et al. Relationship between markers of blood oxidant status and physiological variables in healthy and heaves-affected horses after exercise. Equine Vet J Suppl 2002;(34):159–164.
- 8
Art T, Kirschvink N, Smith NC, et al. Indices of oxidative stress in blood and pulmonary epithelium lining fluid in horses suffering from recurrent airway obstruction. Equine Vet J 1999; 31:397–401.
- 9
Deaton CM, Marlin DJ, Smith NC, et al. Antioxidant and inflammatory responses of healthy horses and horses affected by recurrent airway obstruction to inhaled ozone. Equine Vet J 2005; 37:243–249.
- 10↑
Deaton CM, Marlin DJ, Smith NC, et al. Pulmonary epithelial lining fluid and plasma ascorbic acid concentrations in horses affected by recurrent airway obstruction. Am J Vet Res 2004; 65:80–87.
- 11
Macnee W & Rahman I. Oxidants and antioxidants as therapeutic targets in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1999; 160:S58–S65.
- 12
Rahman I & MacNee W. Regulation of redox glutathione levels and gene transcription in lung inflammation: therapeutic approaches. Free Radic Biol Med 2000; 28:1405–1420.
- 13
Repine JE. Oxidative stress in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1997; 156:341–357.
- 14
Cross CE, Van de Vliet A, O'Neill CA, et al. Reactive oxygen species and the lung. Lancet 1994; 344:930–933.
- 15
Barnes PJ. Pathophysiology of asthma. Br J Clin Pharmacol 1996; 42:3–10.
- 16
Southorn PA & Powis G. Free radicals in medicine. I. Chemical nature and biologic reactions. Mayo Clin Proc 1988; 63:381–389.
- 17
Bowler RP, Crapo JD. Oxidative stress in allergic respiratory diseases. J Allergy Clin Immunol 2002; 110:349–356.
- 18↑
Kirkham P & Rahman I. Oxidative stress in asthma and COPD: antioxidants as a therapeutic strategy. Pharmacol Ther 2006; 111:476–494.
- 19↑
Rahman I & MacNee W. Oxidative stress and regulation of glutathione in lung inflammation. Eur Respir J 2000; 16:534–554.
- 20
Deneke SM, Fanburg BL. Regulation of cellular glutathione. Am J Physiol 1989; 257:L163–L173.
- 21
Kirschvink N, Smith N, Fievez L, et al. Effect of chronic airway inflammation and exercise on pulmonary and systemic antioxidant status of healthy and heaves-affected horses. Equine Vet J 2002; 34:563–571.
- 22↑
Kramaric P, Pavlica Z, Koklic T, et al. Membrane switch hypothesis. 2. Domain structure of phagocytes in horses with recurrent airway obstruction. J Chem In J Model 2005; 45:1708–1715.
- 23↑
Deaton CM. The role of oxidative stress in an equine model of human asthma. Redox Rep 2006; 11:46–52.
- 24
Deaton CM, Marlin DJ, Smith NC, et al. Antioxidant supplementation in horses affected by recurrent airway obstruction. J Nutr 2004; 134:2065S–2067S.
- 25↑
Robinson NE, Olsewski MA, Boehler D, et al. Relationship between clinical signs and lung function in horses with recurrent airway obstruction (heaves) during a bronchodilator trial. Equine Vet J 2000; 32:393–400.
- 26↑
Williams CA, Kronfield DS, Kess TM, et al. Antioxidant supplementation and subsequent oxidative stress of horses during an 80-km endurance race. J Anim Sci 2004; 82:588–594.
- 27
Levine M, Wang Y, Rumsey SC. Analysis of ascorbic acid and dehydroascorbic acid in biological samples. Methods Enzymol 1999; 299:65–76.
- 28
Behrens WA & Madere R. A procedure for the separation and quantitative analysis of ascorbic acid, dehydroascorbic acid, isoascorbic acid, and dehydroisoascorbic acid in food and animal tissue. J Liq Chromatogr 1994; 17:2445–2455.
- 29
Vrins A, Doucet M & Nunez-Ochoa L. A retrospective study of bronchoalevolar lavage cytology in horses with clinical findings of small airway disease. Zentralbl Veterinarmed A 1991; 38:472–479.
- 30
Ay H, Topal T, Uysal B, et al. Time-dependent course of hyperbaric oxygen-induced oxidative effects in rat lung and erythrocytes. Clin Exp Pharmacol Physiol 2007; 34:787–791.
- 31
Tkacova R, Kluchova Z, Joppa P, et al. Systemic inflammation and systemic oxidative stress in patients with acute exacerbations of COPD. Respir Med 2007; 101:1670–1676.
- 32
Nadeem A, Raj HG, Chhabra SK. Increased oxidative stress and altered levels of antioxidants in chronic obstructive pulmonary disease. Inflammation 2005; 29:23–32.
- 33
Ferrer MD, Sureda A, Batle JM, et al. Scuba diving enhances endogenous antioxidant defenses in lymphocytes and neutrophils. Free Radic Res 2007; 41:274–281.
- 34
Tauler P, Sureda A, Cases N, et al. Increased lymphocyte antioxidant defences in response to exhaustive exercise do not prevent oxidative damage. J Nutr Biochem 2006; 17:665–671.
- 35↑
Avissar NE, Reed CK, Cox C, et al. Ozone, but not nitrogen dioxide, exposure decreases glutathione peroxidases in epithelial lining fluid of human lung. Am J Respir Crit Care Med 2000; 162:1342–1347.
- 36↑
Deaton CM, Marlin DJ, Smith NC, et al. Effect of acute airway inflammation on the pulmonary antioxidant status. Exp Lung Res 2005; 31:653–670.
- 37
Deaton CM, Marlin DJ, Smith NC, et al. Pulmonary bioavailability of ascorbic acid in an ascorbate-synthesising species, the horse. Free Radic Res 2003; 37:461–467.
- 38
Deaton CM, Marlin DJ, Smith NC, et al. Antioxidant supplementation and pulmonary function at rest and exercise. Equine Vet J Suppl 2002;(34):58–65.
