Abstract

Broilers reared under commercial conditions inhale irritant gases and aerosolized particulates contaminated with gram-negative bacteria and bacterial lipopolysaccharide (LPS). Previous studies demonstrated that i.v. injections of LPS can trigger an increase in the pulmonary arterial pressure (PAP); however, the pulmonary hemodynamic response to aerosolized LPS entering via the most common route, the respiratory tract, had not been evaluated in broilers. In experiment 1, broilers reared on new wood shavings litter in clean environmental chambers either were not pretreated (control group) or were pretreated via aerosol inhalation of substances (food color dyes and propylene glycol) known to sensitize the airways. One day later, the broilers were anesthetized, catheterized to record the PAP, and an intratracheal aerosol spray of LPS (1 mL of 2 mg/mL of LPS) was administered. Broilers in the control group as well as broilers pretreated with aerosolized distilled water or yellow and blue food color dyes did not develop pulmonary hypertension (PH; an increase in PAP) after the intratracheal spray of LPS, whereas broilers that had been pretreated with red food color did develop PH in response to intratracheal LPS. In experiment 2, birds raised under commercial conditions on used wood shavings litter developed PH in response to intratracheal LPS regardless of whether they had been pretreated with aerosolized red food color dye. In experiment 3, broilers reared in clean environmental chambers on new wood shavings litter were used to demonstrate that Red Dye #3 and propylene glycol are capable of priming the responsiveness of the airways to a subsequent intratracheal LPS challenge. Common air contaminants such as LPS can result in PH leading to pulmonary hypertension syndrome (ascites) in broilers with appropriately primed airways.

INTRODUCTION

Bacterial lipopolysaccharide (LPS; endotoxin) is an integral component of the cell wall of gram-negative bacteria (e.g., Escherichia coli and Salmonella). Animals in commercial production facilities are chronically challenged with airborne gram-negative bacteria and LPS. Approximately 40% of the gram-negative bacteria and LPS in inhaled dust reside on particles of a respirable size (≤5 μm in diameter) capable of penetrating to the gas exchange parenchyma and triggering an inflammatory response that is profoundly deleterious to respiratory function. Symptoms of respiratory inflammation that cumulatively attenuate the growth and productivity of domesticated animals include dyspnea (labored breathing due to elevated airway resistance), hypoxemia (undersaturation of arterial blood with oxygen), and hemodynamic dysfunction including pulmonary arterial hypertension (PAH) and reduced cardiac output due to pulmonary vasoconstriction (Anderson et al., 1966; Hayter and Besch, 1974; Gross, 1990; Whyte, 1993; Sander, 1994; Brown et al., 1997; Canonico and Brigham, 1997; Parsons et al., 1997; Reynolds, 1997; Tottori et al., 1997; Fedde, 1998; Yamaguchi et al., 2000; Zucker et al., 2000; Alexander and Rietschel, 2001; Bakutis et al., 2004).

Most aerosol particulates and pathogens are trapped by mucus in the conducting airways and are prevented from entering the lung parenchyma by the mucociliary escalator. Susceptibility to airborne particulates increases when mucociliary transport is inhibited by exposure to NH3 and infectious bronchitis virus or when aerosolized particles are small enough to be conveyed with the inspired air to the terminal gas exchange surfaces. Respirable particulates reaching the lung parenchyma must penetrate the surfactant layer before being engulfed by phagocytic gas-exchange epithelial cells and translocated to the interstitium where LPS can bind to receptors on the surface of monocytes, heterophils, thrombocytes, and endothelial cells (Stearns et al., 1987; Brown et al., 1997). The ensuing cascade of intracellular signaling events culminates in the following: transcription and translation of genes associated with the innate immune response, production and local release/expression of inflammatory cytokines, production of the vasodilator NO, and production or release of vasoconstrictors including thromboxane and serotonin (Brown et al., 1997; Wideman et al., 2004).

Broiler chickens are sensitive to respiratory perturbations, because their lung capacity is only marginally adequate to support optimal performance under the best of conditions. Accordingly, broilers are susceptible to the onset of hypoxemia and PAH contributing to pulmonary hypertension syndrome (ascites) whenever vasoconstrictors released during the innate immune response overwhelm concurrently produced vasodilators. Intravenous LPS injections trigger PAH that, on average, is delayed in onset by 10 to 15 min, attains the major peak response within approximately 20 to 25 min postinjection, and thereafter gradually recedes toward the baseline pressure. These responses are not caused by contemporaneous changes in cardiac output; therefore, the hypertensive phase reflects net pulmonary vasoconstriction, whereas the subsequent recovery phase reflects net pulmonary vasodilation. Variation among broilers in their responses to LPS likely reflects variation in the relative proportions or profiles of vasoconstrictors and vasodilators produced during the inflammatory cascade (Wideman et al., 2001, 2004; Wang et al., 2002a,b, 2003; Wideman and Chapman, 2004; Chapman et al., 2005; Bowen et al., 2005a,b).

Pulmonary hemodynamic responses to airborne (aerosol) LPS may differ substantially from those triggered by i.v. LPS injections. Airway defense mechanisms potentially can minimize the interaction of inhaled LPS with the leukocytes responsible for producing key vasoconstrictors, whereas i.v. LPS administration provides immediate and intimate access to the innate immune system. A previous experiment demonstrated that respiratory function was compromised in broilers reared on floor litter where they inhaled litter dust and fumes, when compared with broilers reared in clean stainless steel cages (Wang et al., 2002b). However, prolonged exposure of broilers to floor litter is an imprecise method for evaluating time-course responses to LPS. Accordingly, the present studies were conducted to pursue the development of a suitably controllable model for exposing the conducting airways and gas exchange parenchyma of broiler lungs to aerosolized LPS.

MATERIALS AND METHODS

Bird Management

Broiler chicks were wing-banded and housed in environmental chambers (8 m2 of floor space) within the Poultry Environmental Research Lab (PERL) on the University of Arkansas Poultry Research Farm. The birds were brooded at 33°C from d 1 to 3, 31°C from d 4 to 6, 29°C from d 7 to 10, 26°C from d 11 to 14, and 24°C thereafter. Birds were fed a 23% CP corn-soybean meal-based diet formulated to meet the NRC (1994) standards for all ingredients. Feed and water were provided ad libitum. Feed was provided as crumbles throughout the experiment. Lights were on for 24 h/d through d 4 and 16 h/d thereafter.

Pilot Aerosol Delivery Technique Studies

Preliminary studies were conducted to evaluate methods for delivering aerosolized LPS into the respiratory system. All aerosol exposures to LPS were conducted inside a fume hood within the PERL. The LPS from Salmonella Typhimurium (Sigma Chemical Co., St Louis, MO) was dissolved in 0.9% sodium chloride solution (9 g of NaCl/L). Broiler weights were recorded before and 24 h after LPS was administered. Aerosolized LPS delivery was considered to be effective when the responses of the birds were typical of those known to be triggered by i.v. LPS injections, including lethargy, depression, avoidance of feed and water, and minimal or negative weight gain over a 24-h period (Xie et al., 2000; Wang et al., 2003). For each method of aerosol delivery assessed, a subset of additional experiments was conducted in which LPS was replaced with red food color dye (McCormick Food Colors and Egg Dye, McCormick & Co. Inc., Hunt Valley, MD). Necropsies were carried out on 2 birds from each treatment to qualitatively assess the penetration of the aerosolized food color into the trachea, bronchi, parabronchi, and air sacs.

Passive Aerosol LPS Inhalation.

A CompMist Piston Compressor Nebulizer (Mabis Healthcare, Waukeegan, IL) was used to vaporize the LPS solution into aerosol droplets ranging in diameter from 0.5 to 10 μm (Tiano and Dalby, 1996; Rau, 2002). During evaluations using distilled water, this nebulizer aerosolized approximately 0.25 mL of solution per min. Broilers (n = 15, 19 d of age; 357 ± 8 g of BW) were manually supported in an upright posture with their head inserted through a suitably sized opening in a Plexiglas box (30 × 20 × 30 cm; length × width × height) facing the outlet tube from the nebulizer (head-directed aerosol delivery). The 2-cm diameter outlet tube was positioned 10 cm from the face of the bird to direct vaporized LPS (6 mg of LPS/10 mL) toward the mouth and nostrils of the bird for 1 to 4 min. The Plexiglas box prevented air currents in the fume hood from diverting the aerosol away from the face of the bird. When food color was substituted for the LPS, red aerosol vapor was observed to streamline into the mouth and nostrils of the bird as it inhaled, and some red vapor exited the nostrils during expiration. In a second approach designed to permit long intervals of aerosol exposure for multiple birds, up to 4 broilers at a time were placed inside an 80-L3 plastic box (50 × 40 × 40 cm), and the outlet tube from the nebulizer was inserted through a 2-cm hole in the uppermost portion of 1 end of the box. Aerosolized LPS was allowed to suffuse the entire box for up to 40 min (whole-body aerosol delivery). The CO2 accumulation inside the box promoted vigorous panting, which facilitated deep inhalation of the fog-like LPS vapor into the respiratory tract. The nebulizer was loaded with LPS (6 mg of LPS/10 mL) and was refilled as needed to permit 40 min of whole-body aerosol LPS delivery (n = 12; 19 d of age; 596 ± 16 g of BW). For both the head-directed and whole-body LPS aerosol delivery techniques, the test broilers rarely exhibited any reduction in 24-h BW gain when compared with nonexposed flock mates (data not shown). Necropsies revealed food color aerosolized by both techniques throughout the conducting airways and within the parabronchi. It was evident throughout these pilot studies that it would be difficult to determine the quantity of LPS inhaled and retained during passive aerosol inhalation and that aerosol LPS inhalation would be difficult to administer while cardiopulmonary hemodynamic variables were being recorded in broilers.

Direct Intratracheal Aerosol LPS Injection.

For direct intratracheal administration of LPS, an IA-1B MicroSprayer (17.5-cm cannula length; Penn-Century Inc., Philadelphia, PA) connected to a 1-mL syringe was used to spray aerosolized LPS directly into the lower portions of the trachea, as described elsewhere for mammals (Wheeldon et al., 1992; Sookhai et al., 2002. Bivas-Benita et al., 2005). Broilers were restrained in dorsal recumbency inside the fume hood, the oral cavity was opened, and the larynx was visualized by carefully retracting the tongue. The cannula of the MicroSprayer was introduced into the trachea with the tip of the cannula located proximal to the syrinx, then 1 mL of LPS (2 mg/mL) was injected as an aerosol spray. Immediately after injecting the LPS, the MicroSprayer was withdrawn from the trachea, and the head of the bird was held upward with the beak held closed to ensure the LPS spray remained within the airways. Food color administered by this technique penetrated into the parabronchi. Birds receiving the intratracheal aerosol LPS spray (n = 11, 35 d of age; 1,929 ± 21 g of BW) tended to be depressed and lethargic during the ensuing 6 h, and they exhibited minimal or no 24-h BW gain (data not shown). Based on this evidence of a biological response to LPS, the intratracheal aerosol LPS delivery technique was used for subsequent experiments.

Pilot PAP Studies

Five broilers inhaled aerosolized red food color for 40 min via the whole-body delivery technique but were not used to assess the anatomical distribution of the inhaled dye. These preaerosolized broilers (36 d of age; 2,744 ± 115 g of BW) were used within 24 h of postaerosol exposure along with an equal number of their nonaerosolized control flock mates (37 d of age; 2,808 ± 87 g of BW) in pilot studies to determine if the intratracheal aerosol LPS spray would cause the PAP to increase. All birds were anesthetized using i.m. injections of 1 mL of allobarbital (5,5-diallyl-barbituric acid; 25 mg/mL; Sigma Chemical Co.) and 1 mL of ketamine HCl (100 mg/mL). They were fastened in dorsal recumbency on a surgical board. Lidocaine HCl (2%) was injected s.c. around the basilica vein, then the proximal end of a Silastic catheter (0.012 in i.d., 0.025 in o.d.) filled with heparinized saline (200 IU of heparin/mL of 0.9% NaCl) was inserted into the vein. The distal end of the catheter was attached to a blood pressure transducer interfaced through a Transbrige pre-amplifier (World Precision Instruments, Sarasota, FL) to a Biopac MP100 data acquisition system using AcqKnowledge software (Biopac Systems Inc., Santa Barbara, CA). The catheter was slowly advanced into the basilica vein, right atrium, right ventricle, and main trunk of the pulmonary artery where PAP was recorded as described previously (Guthrie et al., 1987; Wideman et al., 1996; Wideman, 1999). Birds were allowed to stabilize for a period of 8 min, and control PAP data were recorded for 4 min. One milliliter of LPS (2 mg/mL) was administered using the intratracheal MicroSprayer as described above, and the PAP was recorded for an additional 44 min. Approximately 20 min after LPS administration, the PAP increased above control levels in 4 out of 5 broilers that had been pretreated with aerosolized red food color and in only 1 out of 5 of the nonaerosolized control broilers.

These unanticipated observations led to a series of experiments designed to replicate the initial phenomenology and to identify the compound(s) in the red food color responsible for apparently sensitizing or priming the respiratory tract of broilers to LPS. The whole-body aerosol delivery technique was used for 40 min to deliver aerosolized distilled water, McCormick food colors, or saturated aqueous solutions of some of the primary ingredients listed for these food colors, including FD&C Red #3 and the carrier solvent propylene glycol (PG). Various dye compounds including FD&C Red #3 and FD&C Red Dye #40, and carrier solvents including PG, have been implicated as agents responsible for priming or amplifying respiratory inflammatory and allergic (bronchial-constrictive) responses in humans and experimental mammals (Fisherman and Cohen, 1973; Michaelsson and Juhlin, 1973; Weber et al., 1979; Pruitt, 1985; Berlin, 1997). Three experiments were conducted in which broilers were anesthetized and prepared for PAP recordings approximately 24 h postexposure to whole-body aerosol delivery of distilled water, food colors, or their ingredients.

Experiment 1.

Male broilers from a commercial source were wing-banded and reared on clean wood shavings litter in environmental chambers within the PERL. When they were 43 to 55 d of age, the broilers were randomly assigned to 1 of 5 groups: nonaerosolized broilers (NA; n = 11; 3,051 ± 75 g of BW) were not previously exposed to aerosol inhalation and were injected i.v. with 1 mL of 2 mg/mL of LPS while the PAP was recorded; water-aerosolized broilers (WA; n = 10; 3,014 ± 103 g of BW) inhaled aerosolized distilled water for 40 min then 24 h later received 1 mL of 2 mg/mL of LPS via intratracheal aerosol spray; red food color-aerosolized broilers (RA; n = 13; 2,929 ± 81 g of BW) inhaled aerosolized McCormick red food color for 40 min then 24 h later received 1 mL of 2 mg/mL of LPS via intratracheal aerosol spray; yellow and blue food color-aerosolized broilers (YBA; n = 9; 3,193 ± 107 g of BW) inhaled an aerosolized 1:1 mixture of McCormick yellow and blue food color for 40 min then 24 h later received 1 mL of 1 mg/mL of LPS via intratracheal aerosol spray; and propylene glycol-aerosolized broilers (PGA; n = 9; 3,496 ± 126 g of BW) inhaled aerosolized PG (Sigma Chemical Co., 99.5% purity) for 40 min then 24 h later received 1 mL of 2 mg/mL of LPS via intratracheal aerosol spray. The PAP was recorded as described for the pilot PAP studies.

Experiment 2.

Male broilers from a genetic line maintained on the University of Arkansas Poultry Research Farm were wing-banded and reared on previously used wood shaving litter in an open-sided poultry house equipped with Plasson drinkers. Birds from 35 to 53 d old were randomly assigned to 1 of 4 groups: nonaerosolized-saline broilers (NA-S; n = 9; 1,525 ± 87 g of BW) were not previously exposed to aerosol inhalation and received 1 mL of 0.9% NaCl via intratracheal aerosol spray; nonaerosolized-LPS broilers (NA-LPS; n = 10; 1,350 ± 65 g of BW) were not previously exposed to aerosol inhalation and received 1 mL of 2 mg/mL of LPS via intratracheal aerosol spray; nonaerosolized-LPS-i.v. broilers (NA-LPS i.v.; n = 10; 1,434 ± 74 g of BW) were not previously exposed to aerosol inhalation and were injected i.v. with 1 mL of 2 mg/mL of LPS; red food color-aerosolized broilers (RA; n = 10; 1,652 ± 60 g of BW) inhaled aerosolized red food color for 40 min then 24 h later received 1 mL of 1 mg/ mL of LPS via intratracheal aerosol spray. The PAP was recorded as described for the pilot PAP studies.

Experiment 3.

Male broilers from a different hatch of the same line that had been used in experiment 2 were wing-banded and reared on clean wood shavings litter in environmental chambers within the PERL. When they were 45 to 53 d old, the broilers were randomly assigned to 1 of 2 groups: nonaerosolized broilers (NA; n = 12; 2,611 ± 117 g of BW) were not previously exposed to aerosol inhalation and received 1 mL of 2 mg/mL of LPS via intratracheal aerosol spray, and Red Dye #3-aerosolized broilers (RD#3; n = 12; 2,647 ± 57 g of BW) inhaled aerosolized Red Dye #3 (Sigma Chemical Co.) mixed with PG (0.04 g of RD#3/mL of PG) for 40 min then 24 h later received 1 mL of 2 mg/mL of LPS via intratracheal aerosol spray. The PAP was recorded as described for the pilot PAP studies.

Data Acquisition and Statistical Analyses

The Biopac MP 100 data acquisition system continuously recorded PAP (mmHg) during 2 control intervals of 2 min each (CTL1, CTL2) and during 22 intervals of 2 min each after administration of LPS (sampling intervals 2 to 44). The PAP data were averaged electronically within each of the 2-min intervals. The protocol used for data averaging accommodates the influences of pulse pressure and respiratory cycles on PAP (Wideman et al., 1996). Individual birds were considered to be the experimental unit within each treatment group (n = number of birds per treatment group). Data from each sampling interval were pooled within each treatment. Pooled data from CTL2 were compared against the following 22 intervals within treatments using 1-way repeated-measures ANOVA by time, and pooled data from each sample interval were compared among treatments using 1-way ANOVA by group (Jandel Scientific, 1994). Means were separated by the Tukey test when the F-test from the 1-way ANOVA was declared significant (P < 0.05).

RESULTS

Experiment 1

Absolute PAP values during control sample intervals (CTL1 and CTL2) averaged approximately 22.5, 23.7, 25.5, 24, and 20.5 mmHg for the NA, WA, RA, YBA, and PGA groups, respectively (Figure 1, upper panel). None of the PAP values differed significantly among the groups during sample intervals CTL1 to 16. Intravenous LPS administration caused the PAP in the NA group to increase above the initial control values of this group by sample interval 18, and the PAP remained elevated in the NA group until sample interval 44. All groups receiving the intratracheal LPS spray tended to exhibit an increase in PAP during the ensuing 2 min (sample interval 2) coinciding with a noticeable stress response caused by inserting the cannula and introducing 1 mL of aerosol into the trachea. The NA group received LPS via an i.v. injection and exhibited no comparable increase in PAP during sample interval 2. Within 4, 6, and 6 min, the PAP values for the YBA, WA, and PGA groups returned to levels that did not differ from their respective initial control values. After sample interval 8 only, the PAP values of the RA group again increased above the initial control level (sample intervals 24 to 28). Moreover, the PAP values for the RA group were not lower than those of the NA group throughout sample intervals 20 to 38 (Figure 1, upper panel). To normalize these responses for initial numerical differences in control PAP values, all data were recalculated as the percentage change in PAP compared with the average control PAP values (CTL1 and CTL2) as shown in the lower panel of Figure 1. During sample intervals 18 to 22, the NA group exhibited a dramatically higher percentage increase in PAP attributable the i.v. LPS injection, when compared with the percentage increase in PAP by all groups that received the intratracheal LPS aerosol, regardless of aerosol pretreatment. The values for the RA and PG groups increased by sample intervals 24 and 26, respectively, to peak levels that were not significantly lower than the contemporaneous values for the NA group. In contrast, the percentage change in PAP values for the WA and YBA groups remained lower than those of the NA group until the end of the experiment, with the exception of sample interval 40 for the WA group (Figure 1, lower panel).

Experiment 2

The absolute PAP values during control sample intervals (CTL1 and CTL2) fell within a narrow range (18.5 to 20 mmHg) for all groups (Figure 2). All groups that received an intratracheal spray of saline (NA-S) or LPS (NA-LPS and RA) exhibited acute increases in their PAP values within 2 min (sample interval 2), whereas no increase in PAP was observed in the group that received LPS via an i.v. injection (NA-LPS i.v.) during the same sample interval. The initial response to the intratracheal spray had subsided by sample interval 8, and the groups did not differ again until sample interval 20 when the PAP of the NA-LPS i.v. group was higher than the PAP of the NA-S group but not higher than the values of the NA-LPS or RA groups. When compared with initial control values CTL1 and CTL2, the groups that received LPS i.v. (NA-LPS i.v.) or as an intratracheal aerosol spray (NA-LPS, RA) tended to exhibit sustained pulmonary hypertensive responses to LPS (sample intervals 20 to 38), whereas the PAP gradually declined in the NA-S group. After sample interval 40, the PAP did not differ between groups (Figure 2).

Experiment 3

Both NA and RD#3 groups increased their PAP within 2 min after the intratracheal LPS aerosol spray (Figure 3). The PAP did not differ between groups until sample interval 26 when the RD#3 group had higher PAP values than the NA group. From sample interval 26 and throughout the rest of the experiment, the PAP values of the RD#3 group remained higher than the PAP values of the NA group (Figure 3). Variation in the PAP responses among individual broilers is shown in Figure 4. The intratracheal aerosol spray caused the PAP to increase by at least 4 mmHg between sample intervals 16 to 38 in 3 of 12 broilers in the NA group (Figure 4, upper panel) and 11 of 12 broilers in the RD#3 group (Figure 4, lower panel).

DISCUSSION

The PAH responses to LPS clearly were consistently higher using the i.v. rather than the aerosol route of administration. When injected i.v., LPS molecules within the blood vessels rapidly interact with receptors on the surface of immune cells to initiate signaling cascades leading to the release of proinflammatory mediators and vaso-active factors (Wideman et al. 2001, 2004). In contrast, LPS administered via the airways must overcome multiple mucosal defenses before proinflammatory responses can be activated. The respiratory tract must guarantee effective, ongoing gas exchange while providing appropriate defense against pathogens. Mucosal defenses must provide effective surveillance and discrimination between threatening and nonthreatening agents (Ramnik and Podolsky, 2000; Granucci and Ricciardi-Castagnoli, 2003). While larger airborne particles are trapped in the nasal cavities and trachea, smaller respirable particles averaging 1.1 μm are able to reach the lung gas exchange parenchyma and abdominal air sacs (Fulton et al., 1990). Respirable particles can be heavily contaminated with a wide range of immunogenic substances including pathogens and toxins (Bakutis et al., 2004). Sustaining ongoing vigorous immune responses against these substances may incur more damage than protection to the lungs of the host (Ewaschuk and Dieleman, 2006). Accordingly, pulmonary mucosal defenses appear to be designed to bind, inactivate, or discriminate among inhaled antigens, thereby applying a pattern of tolerance of nonthreatening agents to minimize counterproductive chronic inflammatory responses within the gas exchange parenchyma.

Surfactant proteins have been proposed to modulate immune responses in the airways (Crouch, 1998; Borron et al., 2000). Mice deficient in surfactant protein A (SPA) are prone to develop a more robust immune response against LPS than are mice that have normal levels of SPA. For example, SP-A attenuates LPS-induced production of tumor necrosis factor α and macrophage inflammatory protein-2 in bronchoalveolar fluid in vivo (Borron et al., 2000). In addition, SP-A reduces the proliferative response of T cells to mitogens, a process thought to be mediated by the binding of SP-A to a 210-kDa receptor on lymphocyte and macrophage membranes (Borron et al., 1998). In mammals, SP-A is extensively present in upper and lower airways including alveoli that are known to be populated with alveolar macrophages. In birds, macrophages normally are not present within the air capillaries where gas exchange occurs, but macrophages have been detected in the atria and infundibula of the parabronchi, as well as in the larger conducting airways (Maina and Cowley, 1998; Nganpiep and Maina, 2002). Coincidentally, SP-A has not been found in the air capillaries or parabronchi (tertiary bronchi), but SP-A is localized in subsets of epithelial cells of the secondary bronchi (Zeng et al., 1998; Johnston et al., 2000) where a regulatory effect on macrophages and lymphocytes may be exerted. Evidently, in the present study, the mucosal defense mechanisms of broilers reared within environmental chambers effectively prevented aerosolized LPS from triggering counterproductive responses such as behavioral depression, reduced BW gain, and PAH except when the airways of the birds had been suitably primed or sensitized.

Consistently throughout the 3 experiments, all groups receiving the intratracheal LPS spray tended to exhibit an increase in PAP during the ensuing 2 min (sample interval 2) coinciding with a noticeable stress response caused by inserting the cannula and introducing 1 mL of aerosol into the trachea. Epinephrine is known to trigger transient pulmonary vasoconstriction and pulmonary hypertension in broilers (Wideman, 1999; Lorenzoni, 2006). Experiments 1 and 3 demonstrated that pretreating broilers with aerosolized red food color and its key ingredients, Red Dye #3 and PG, sensitized or primed the respiratory system so that the broilers more consistently exhibited a pulmonary hypertensive response to a subsequent intratracheal LPS aerosol spray. Pretreatment with aerosolized water or nonred food colors failed to sensitize the broilers in experiment 1. Ingestion of Red Dye #3 has been implicated in human cases of bronchoconstriction, urticaria, and rhinitis, and i.v. injections of the carrier vehicle PG have been associated with dermatitis, phlebitis, and increases in blood histamine levels (Fisherman and Cohen, 1973; Berlin, 1997; Doenicke et al., 1999). It is possible that Red Dye #3 and PG may interfere with the modulatory effect of surfactant proteins on the immune cells within the airways. Red Dye #3 and PG also may initiate a modest inflammatory response within the airways. Inflammation may induce production and secretion of cytokines, thereby overcoming the suppressor effect of surfactant proteins and stimulating the migration of immune cells from the interstitium into the airways. Leukocytes recruited into the airways would be directly accessible to respirable LPS, permitting an intimate interaction and thus more direct activation of the innate immune response.

In experiment 2, exposure to previously used floor litter within a curtain-sided house appeared to naturally prime or sensitize the airways of the broilers. This was the only experiment in which broilers that were not pretreated with aerosolized red food color or its key ingredients nevertheless consistently exhibited significant PAH after LPS was administered as an intratracheal spray. Perhaps a similar natural airway sensitization might be anticipated when broilers chronically inhale common airborne pollutants and gasses that normally are present in commercial poultry houses. Mechanisms by which the airways became sensitized to LPS in the present study remain to be determined. It has been demonstrated previously that intratracheal instillation of Corynebacterium parvum or E. coli effectively increased the number of phagocytes collected by lung lavage within 24 h (Toth et al., 1987). Additionally, macrophages have been reported to migrate into air capillaries in a variety of infectious diseases, including toxoplasmosis, fatal viral hydropericardium syndrome, highly pathogenic infectious bursal disease, and highly pathogenic avian influenza (Howerth and Rodenroth, 1985; Abe et al., 1998; Nakamura et al., 2001). Inhaling low doses of toxins and pathogens, or primers such as Red Dye #3 and PG, may induce subclinical levels of lung inflammation and immigration of immune cells, thereby unbalancing the mechanisms of tolerance normally exhibited by the pulmonary mucosal immune system. Actually, oxidized glutathione (an index of tissue oxidative stress) and blood CO2 (index of cardiopulmonary system performance) have been reported higher in birds reared on floor litter when compared with birds reared in clean environments (environmental chambers and stainless steel cages, respectively; Bottje et al., 1998; Wang et al., 2002b). Treatment with aerosolized Red Dye #3 and PG appears to constitute a nonpathogenic, effectively controllable experimental model for deducing the mechanisms by which air pollutants commonly present in commercial poultry houses can enhance pulmonary hypertensive responses of broilers to respirable LPS. In fact, hypertensive response to intratracheal LPS can be detected sporadically in nonaerosolized birds reared in environmental chambers (Figure 4, upper panel), and relative unresponsiveness also occurs in birds raised under commercial conditions. However, the proportion of responders is dramatically increased when birds are either reared under commercial conditions or are primed with aerosolized Red Dye #3 and PG (Figure 4, lower panel).

Figure 1

Upper panel: pulmonary arterial pressure (PAP; mean ± SEM) of male broilers reared in environmental chambers in experiment 1. The PAP was measured in non-aerosolized broilers after an i.v. injection of 1 mL of 2 mg/mL of lipopolysaccharide (LPS; NA; n = 11) and in propylene glycol-aerosolized broilers (PGA; n = 9), red food color-aerosolized broilers (RA; n = 13), yellow and blue-aerosolized broilers (YBA; n = 9), and water-aerosolized broilers (WA; n = 10) after an intratracheal spray of 1 mL of 2 mg of LPS. Data were averaged electronically during representative sample intervals before (control sample intervals CTL 1 and 2) and after the LPS administration (sample intervals 2 to 44). Asterisks (*) for the NA and RA, crosses (†) for the YBA, and number symbol (#) for the WA groups denote values higher than the pre-LPS values of the respective group (sample intervals CTL 1 and 2). Lower panel: PAP of the NA, PGA, RA, YBA, and WA groups expressed as a percentage of change in PAP compared with their respective pre-LPS values (sample intervals CTL 1 and 2). Different letters (a,b,c) designate group means that differed within a sample interval (P ≤ 0.05). Arrows indicate LPS administration. aver = average.

Figure 1

Upper panel: pulmonary arterial pressure (PAP; mean ± SEM) of male broilers reared in environmental chambers in experiment 1. The PAP was measured in non-aerosolized broilers after an i.v. injection of 1 mL of 2 mg/mL of lipopolysaccharide (LPS; NA; n = 11) and in propylene glycol-aerosolized broilers (PGA; n = 9), red food color-aerosolized broilers (RA; n = 13), yellow and blue-aerosolized broilers (YBA; n = 9), and water-aerosolized broilers (WA; n = 10) after an intratracheal spray of 1 mL of 2 mg of LPS. Data were averaged electronically during representative sample intervals before (control sample intervals CTL 1 and 2) and after the LPS administration (sample intervals 2 to 44). Asterisks (*) for the NA and RA, crosses (†) for the YBA, and number symbol (#) for the WA groups denote values higher than the pre-LPS values of the respective group (sample intervals CTL 1 and 2). Lower panel: PAP of the NA, PGA, RA, YBA, and WA groups expressed as a percentage of change in PAP compared with their respective pre-LPS values (sample intervals CTL 1 and 2). Different letters (a,b,c) designate group means that differed within a sample interval (P ≤ 0.05). Arrows indicate LPS administration. aver = average.

Figure 2

Pulmonary arterial pressure (PAP; mean ± SEM) of male broilers from a genetic line maintained on the University of Arkansas Poultry Research Farm reared on previously used floor litter in an open-sided poultry house in experiment 2. The PAP was measured in non-aerosolized broilers receiving 1 mL of NaCl via intratracheal spray (NA-S; n = 9), non-aerosolized broilers after i.v. (NA-LPS i.v.; n = 10) or intratracheal (NA-LPS; n = 10) administration of 1 mL of 2 mg/mL of lipopolysaccharide (LPS), and red food color-aerosolized broilers (RA; n = 13) after intratracheal spray of 1 mL of 2 mg/mL of LPS. Data were averaged electronically during representative sample intervals before (control, sample intervals CTL 1 and 2) and after the LPS administration (sample intervals 2 to 44). Asterisks (*) denote values higher than the pre-LPS values of the respective group (sample intervals CTL 1 and 2). Different letters (a,b) designate group means that differed within a sample interval (P ≤ 0.05). Arrow indicates LPS administration.

Figure 2

Pulmonary arterial pressure (PAP; mean ± SEM) of male broilers from a genetic line maintained on the University of Arkansas Poultry Research Farm reared on previously used floor litter in an open-sided poultry house in experiment 2. The PAP was measured in non-aerosolized broilers receiving 1 mL of NaCl via intratracheal spray (NA-S; n = 9), non-aerosolized broilers after i.v. (NA-LPS i.v.; n = 10) or intratracheal (NA-LPS; n = 10) administration of 1 mL of 2 mg/mL of lipopolysaccharide (LPS), and red food color-aerosolized broilers (RA; n = 13) after intratracheal spray of 1 mL of 2 mg/mL of LPS. Data were averaged electronically during representative sample intervals before (control, sample intervals CTL 1 and 2) and after the LPS administration (sample intervals 2 to 44). Asterisks (*) denote values higher than the pre-LPS values of the respective group (sample intervals CTL 1 and 2). Different letters (a,b) designate group means that differed within a sample interval (P ≤ 0.05). Arrow indicates LPS administration.

Figure 3

Pulmonary arterial pressure (PAP; mean ± SEM) in male broilers reared in environmental chambers in experiment 3. Non-aerosolized (NA; n = 12), and Red Dye #3 plus propylene glycol-aerosolized broilers (RD#3; n = 12) received an intratracheal spray of 1 mL of 2 mg/mL of lipopolysaccharide (LPS). Data were averaged electronically during representative sample intervals before (control, sample intervals CTL 1 and 2) and after LPS administration (sample intervals 2 to 44). Different letters (a,b) designate group means that differed within a sample interval (P ≤ 0.05). Arrow indicates LPS administration.

Figure 3

Pulmonary arterial pressure (PAP; mean ± SEM) in male broilers reared in environmental chambers in experiment 3. Non-aerosolized (NA; n = 12), and Red Dye #3 plus propylene glycol-aerosolized broilers (RD#3; n = 12) received an intratracheal spray of 1 mL of 2 mg/mL of lipopolysaccharide (LPS). Data were averaged electronically during representative sample intervals before (control, sample intervals CTL 1 and 2) and after LPS administration (sample intervals 2 to 44). Different letters (a,b) designate group means that differed within a sample interval (P ≤ 0.05). Arrow indicates LPS administration.

Figure 4

Individual responses of pulmonary arterial pressure (PAP; mean) in male broilers reared in environmental chambers in experiment 3. Upper panel: PAP in non-aerosolized (NA; n = 12) broilers during representative sample intervals before (control, sample intervals CTL 1 and 2) and after an intratracheal spray of 1 mL of 2 mg/mL of lipopolysaccharide (LPS; sample intervals 2 to 44). Lower panel: PAP in Red Dye #3 plus propylene glycol-aerosolized broilers (RD#3; n = 12) during representative sample intervals before (control, sample intervals CTL 1 and 2) and after an intratracheal spray of 1 mL of 2 mg/mL of LPS (sample intervals 2 to 44). Arrows indicate LPS administration.

Figure 4

Individual responses of pulmonary arterial pressure (PAP; mean) in male broilers reared in environmental chambers in experiment 3. Upper panel: PAP in non-aerosolized (NA; n = 12) broilers during representative sample intervals before (control, sample intervals CTL 1 and 2) and after an intratracheal spray of 1 mL of 2 mg/mL of lipopolysaccharide (LPS; sample intervals 2 to 44). Lower panel: PAP in Red Dye #3 plus propylene glycol-aerosolized broilers (RD#3; n = 12) during representative sample intervals before (control, sample intervals CTL 1 and 2) and after an intratracheal spray of 1 mL of 2 mg/mL of LPS (sample intervals 2 to 44). Arrows indicate LPS administration.

This project was supported by an animal health grant from the University of Arkansas Agricultural Experiment Station. Submitted in partial fulfillment of the requirements for a PhD by A. G. Lorenzoni.

REFERENCES

Abe
,
T.
, K. Nakamura, H. Tojo, M. Mase, and T. Shibahara.
1998
. Histology, immunohistochemistry, and ultrastructure of hydropericardium syndrome in adult broiler breeders and broiler chickens.
Avian Dis.
 
42
:
606
–612.
Alexander
,
C.
, and E. T. Rietschel.
2001
. Bacterial lipopolysaccharides and innate immunity.
J. Endotoxin Res.
 
7
:
167
–201.
Anderson
,
D. P.
, C. W. Beard, and R. P. Hanson.
1966
. Influence of poultry house dust, ammonia, and carbon dioxide on the resistance of chickens to Newcastle disease virus.
Avian Dis.
 
10
:
177
–188.
Bakutis
,
B.
, E. Monstviliene, and G. Januskeviciene.
2004
. Analyses of airborne contamination with bacteria, endotoxins and dust in livestock barns and poultry houses.
Acta Vet. (Brno)
 
73
:
283
–289.
Berlin
,
C. M.
1997
. “Inactive” ingredients in pharmaceutical products: Update (subject review).
Pediatrics
 
99
:
268
–278.
Bivas-Benita
,
M.
, R. Zwier, H. E. Junginger, and G. Borchard.
2005
. Non-invasive aerosol delivery in mice by the endo-trachial route.
Eur. J. Pharm. Biopharm.
 
61
:
214
–218.
Borron
,
P.
, F. X. McCarmac, B. M. Elhalwagi, Z. C. Chroneos, J. F. Lewis, S. Zhu, J. R. Wright, V. L. Shepherd, F. Possmayer, K. Inchley, and L. J. Fraher.
1998
. Surfactant protein A inhibits T cell proliferation via its collagen-like tail and a 210-kDa receptor.
Am. J. Physiol.
 
275
:
L679
–L686.
Borron
,
P.
, J. C. McIntosh, T. R. Korfhagen, J. A. Whitsett, J. Taylor, and J. R. Wright.
2000
. Surfactant-associated protein A inhibits LPS-induced cytokine and nitric oxide production in vivo.
Am. J. Physiol. Lung Cell. Mol. Physiol.
 
278
:
L840
–L847.
Bottje
,
W. B.
, S. Wang, F. J. Kelly, C. Dunster, A. Williams, and I. Mudway.
1998
. Antioxidant defenses in lung lining fluid of broilers: Impact of poor ventilation conditions.
Poult. Sci.
 
77
:
516
–522.
Bowen
,
O. T.
, G. F. Erf, N. B. Anthony, and R. F. Wideman.
2005a
. Pulmonary hypertension triggered by lipopolysaccharide (LPS) in ascites-susceptible and –resistant broilers is not amplified by aminoguanidine, a specific inhibitor of inducible nitric oxide synthase.
Poult. Sci.
 
85
:
528
–536.
Bowen
,
O. T.
, R. F. Wideman, N. B. Anthony, and G. F. Erf.
2005b
. Variation in the pulmonary hypertensive responsiveness of broilers to lipopolysaccharide (LPS) and innate variation in nitric oxide production by mononuclear cells.
Poult. Sci.
 
85
:
1349
–1363.
Brown
,
R. E.
, J. D. Brain, and N. Wang.
1997
. The avian respiratory system: A unique model for studies of respiratory toxicosis and for monitoring air quality.
Environ. Health Perspect.
 
105
:
188
–200.
Canonico
,
A. E.
, and K. L. Brigham.
1997
. Biology of acute injury. Pages 2475–2499 in The Lung. 2nd ed. R. G. Crystal, J. B. West, E. R. Weibel, and P. J. Barnes, ed. Lippencott-Raven, New York, NY.
Chapman
,
M. E.
, W. Wang, G. F. Erf, and R. F. Wideman.
2005
. Pulmonary hypertensive responses of broilers to bacterial lipopolysaccharide (LPS): Evaluation of LPS source and dose, and impact of pre-existing pulmonary hypertension and cellulose micro-particle selection.
Poult. Sci.
 
84
:
432
–441.
Crouch
,
E. C.
1998
. Collectins and pulmonary host defense.
Am. J. Respir. Cell Mol. Biol.
 
19
:
177
–201.
Doenicke
,
A. W.
, M. F. Roizen, R. Hoernecke, W. Lorenz, and P. Ostwald.
1999
. Solvent for etomidate may cause pain and adverse effects.
Br. J. Anaesth.
 
83
:
464
–466.
Ewaschuk
,
J. B.
, and L. A. Dieleman.
2006
. Probiotics and prebiotics in chronic inflammatory bowel diseases.
World J. Gastroenterol.
 
12
:
5941
–5950.
Fedde
,
M. R.
1998
. Relationship of structure and function of the avian respiratory system to disease susceptibility.
Poult. Sci.
 
77
:
1130
–1138.
Fisherman
,
E. W.
, and G. N. Cohen.
1973
. Aspirin and other cross-reacting small chemicals in known aspirin intolerant patients.
Ann. Allergy
 
31
:
476
–484.
Fulton
,
R. M.
, W. M. Reed, and D. B. DeNicola.
1990
. Light microscopic and ultrastructural characterization of cells recovered by respiratory-tract lavage of 2- and 6-week old chickens.
Avian Dis.
 
34
:
87
–98.
Granucci
,
F.
, and P. Ricciardi-Castagnoli.
2003
. Interactions of bacterial pathogens with dendritic cells during invasion of mucosal surfaces.
Curr. Opin. Microbiol.
 
6
:
72
–76.
Gross
,
W. B.
1990
. Factors affecting the development of respiratory disease complex in chickens.
Avian Dis.
 
34
:
607
–610.
Guthrie
,
A. J.
, J. A. Cilliers, F. W. Huchzermeyer, and V. M. Killeen.
1987
. Broiler pulmonary hypertension syndrome. II. The direct measurement of right ventricular and pulmonary artery pressures in the closed chest domestic fowl.
J. Vet. Res.
 
54
:
599
–602.
Hayter
,
R. B.
, and E. L. Besch.
1974
. Airborne-particle deposition in the respiratory tract of chickens.
Poult. Sci.
 
53
:
1507
–1511.
Howerth
,
E. W.
, and N. Rodenroth.
1985
. Fatal systemic toxoplasmosis in a wild turkey.
J. Wildl. Dis.
 
21
:
446
–449.
Jandel Scientific.
1994
. SigmaStat® Statistical Software User’s Manual. Jandel Sci. Softw., San Rafael, CA.
Johnston
,
S. D.
, S. Orgeig, O. Lopatko, and C. B. Daniels.
2000
. Development of the pulmonary surfactant system in two oviparous vertebrates.
Am. J. Physiol. Regul. Integr. Comp. Physiol.
 
278
:
R486
–R493.
Lorenzoni
,
A. G.
2006
. Effects of alpha-tocopherol and l-arginine on cardiopulmonary function in broilers. MS Diss. McGill Univ., Montreal, Canada.
Maina
,
J. N.
, and H. M. Cowley.
1998
. Ultrastructural characterization of the pulmonary cellular defenses in the lung of a bird, the rock dove, Columba livia.
Proc. R. Soc. Lond. B. Biol. Sci.
 
265
:
1567
–1572.
Michaelsson
,
G.
, and L. Juhlin.
1973
. Urticaria induced by preservatives and dye additives in food and drugs.
Br. J. Dermatol.
 
88
:
525
–532.
Nakamura
,
K.
, M. Yamada, S. Yamaguchi, M. Mase, M. Narita, T. Ohyama, and M. Yamada.
2001
. Proliferation of lung macrophages in acute fatal viral infections in chickens.
Avian Dis.
 
45
:
813
–818.
Nganpiep
,
L. N.
, and J. N. Maina.
2002
. Composite cellular defense stratagem in the avian respiratory system: Functional morphology of the free (surface) macrophages and specialized pulmonary epithelia.
J. Anat.
 
200
:
499
–516.
NRC.
1994
. Nutrient Requirements of Poultry. 9th rev. ed. Natl. Acad. Press, Washington, DC.
Parsons
,
P. E.
, G. S. Worthen, and P. M. Henson.
1997
. Injury from inflammatory cells. Pages 2439–2449 in The Lung. 2nd ed. R. G. Crystal, J. B. West, E. R. Weibel, and P. J. Barnes, ed. Lippencott-Raven, New York, NY.
Pruitt
,
A. W.
1985
. “Inactive” ingredients in pharmaceutical products.
Pediatrics
 
76
:
635
–643.
Ramnik
,
J. X.
, and D. K. Podolsky.
2000
. How to get along –friendly microbes in a hostile world.
Science
 
289
:
148
–150.
Rau
,
J. L.
2002
. Design principles of liquid nebulization devices currently in use.
Respir. Care
 
47
:
1257
–1275.
Reynolds
,
H. Y.
1997
. Integrated host defenses against infection. Pages 2353–2365 in The Lung. 2nd ed. R. G. Crystal, J. B. West, E. R. Weibel, and P. J. Barnes, ed. Lippencott-Raven, New York, NY.
Sander
,
J.
1994
. Environmental respiratory pathogens: Poultry house gases and dust. Pages 42–46 in Proc. Am. Assoc. Avian Pathol. Symp. Respir. Dis. Chickens and Turkeys, San Francisco, CA.
Sookhai
,
S.
, J. J. Wang, M. McCourt, W. Kirwan, D. Bouchier-Hayes, and P. Redmond.
2002
. A novel therapeutic strategy for attenuating neutrophil-mediated lung injury in vivo.
Ann. Surg.
 
235
:
285
–291.
Stearns
,
R. C.
, G. M. Barnas, M. Walski, and J. D. Brain.
1987
. Deposition and phagocytosis of inhaled particles in the gas exchange region of the duck, Anas platyrhynchos.
Respir. Physiol.
 
67
:
23
–36.
Tiano
,
S. L.
, and R. N. Dalby.
1996
. Comparison of a respiratory suspension aerosolized by an air-jet and an ultrasonic nebulizer.
Pharm. Dev. Technol.
 
1
:
261
–268.
Toth
,
T. E.
, P. Siegel, and H. Veit.
1987
. Cellular defense of the avian respiratory system. Influx of phagocytes: Elicitation versus activation.
Avian Dis.
 
31
:
861
–867.
Tottori
,
J.
, R. Yamaguchi, Y. Murakawa, M. Sato, K. Uchida, and S. Tateyama.
1997
. Experimental production of ascites in broiler chickens using infectious bronchitis virus and Escherichia coli.
Avian Dis.
 
41
:
214
–220.
Wang
,
W.
, G. F. Erf, and R. F. Wideman.
2002a
. Pulmonary hypertensive response to endotoxin in cellulose-primed and unprimed broiler chickens.
Poult. Sci.
 
81
:
1224
–1230.
Wang
,
W.
, G. F. Erf, and R. F. Wideman.
2002b
. Effect of cage vs. floor litter environments on the pulmonary hypertensive response to intravenous endotoxin and on blood-gas values in broilers.
Poult. Sci.
 
81
:
1728
–1737.
Wang
,
W.
, R. F. Wideman, T. K. Bersi, and G. F. Erf.
2003
. Pulmonary and hematological inflammatory responses to intravenous cellulose micro-particles in broilers.
Poult. Sci.
 
82
:
771
–780.
Weber
,
R. W.
, M. Hoffman, D. A. Raine Jr., and H. S. Nelson.
1979
. Incidence of bronchoconstriction due to aspirin, azo dyes, non-azo dyes, and preservatives in a population of perennial asthmatics.
J. Allergy Clin. Immunol.
 
64
:
32
–37.
Wheeldon
,
E. B.
, M. E. Walker, D. J. Murphy, and C. R. Turner.
1992
. Intratracheal aerosolization of endotoxin in the rat: A model of the adult respiratory distress syndrome (ARDS).
Lab. Anim.
 
26
:
29
–37.
Whyte
,
R. T.
1993
. Aerial pollutants and the health of poultry farmers.
World’s Poult. Sci. J.
 
49
:
139
–156.
Wideman
,
R. F.
1999
. Cardiac output in four, five, and six-week-old broilers, and hemodynamic responses to intravenous injections of epinephrine.
Poult. Sci.
 
78
:
392
–403.
Wideman
,
R. F.
, and M. E. Chapman.
2004
. Nω-Nitro-l-arginine methyl ester (L-NAME) amplifies the pulmonary hypertensive response to endotoxin in broilers.
Poult. Sci.
 
83
:
485
–494.
Wideman
,
R. F.
, M. E. Chapman, W. Wang, and G. F. Erf.
2004
. Immune modulation of the pulmonary hypertensive response to bacterial lipopolysaccharide (endotoxin) in broilers.
Poult. Sci.
 
83
:
624
–637.
Wideman
,
R. F.
, G. F. Erf, and M. E. Chapman.
2001
. Intravenous endotoxin triggers pulmonary vasoconstriction and pulmonary hypertension in broiler chickens.
Poult. Sci.
 
80
:
647
–655.
Wideman
,
R. F.
, Y. K. Kirby, C. D. Tackett, N. E. Marson, and R. W. McNew.
1996
. Cardio-pulmonary function during acute unilateral occlusion of the pulmonary artery in broilers fed diets containing normal or high levels of arginine-HCL.
Poult. Sci.
 
75
:
1587
–1602.
Xie
,
H.
, N. C. Rath, G. R. Huff, W. E. Huff, and J. M. Balog.
2000
. Effects of Salmonella Typhimurium lipopolysaccharide on broiler chickens.
Poult. Sci.
 
79
:
33
–40.
Yamaguchi
,
R.
, J. Tottori, K. Uchida, S. Tateyama, and S. Sugano.
2000
. Importance of Escherichia coli infection in ascites in broiler chickens shown by experimental production.
Avian Dis.
 
44
:
545
–548.
Zeng
,
X.
, K. E. Yutzey, and J. A. Whitsett.
1998
. Thyroid transcription factor-1, hepatocyte nuclear factor-3β and surfactant protein A and B in the developing chicken lung.
J. Anat.
 
193
:
399
–408.
Zucker
,
B. A.
, S. Trojan, and W. Muller.
2000
. Airborne gram-negative bacterial flora in animal houses.
J. Vet. Med. B
 
47
:
37
–46.