Serum Antibody Activity against Poly-N-Acetyl Glucosamine (PNAG), but Not PNAG Vaccination Status, Is Associated with Protecting Newborn Foals against Intrabronchial Infection with Rhodococcus equi

Prior to vaccination, antibody activities to PNAG were low among study mares and there was no significant difference (P = 0.6269) between the vaccinated mares (n = 14) and control mares (n = 8) (see Fig. S1 in the supplemental material); however, irrespective of vaccine group, antibody activities were significantly (P = 0.0208) higher for mares in 2020 than in 2018. Antibody activities against PNAG were significantly higher in serum from vaccinated mares and their foals than in the corresponding samples from controls (Table 1; Fig. 1 and 2). There was no significant effect of year (P = 0.7120) or year by group interaction (P = 0.3530) for PNAG antibody activities of mares, nor were effects of year (P = 0.5530) or year by group interaction (0.7034) significant for antibody activities of foals. C′1q deposition activity, determined as readings of optical density at 405 nm (OD₄₀₅), were significantly higher in sera from vaccinated mares and their foals than the corresponding samples from controls (Table 1; Fig. 3 and 4). There was no significant effect of year (P = 0.6503) or year by group interaction (P = 0.95554) for C′1q deposition activity of mares, nor were effects of year (P = 0.5514) or year by group interaction (0.3903) significant for C′1q deposition activity of foals.

PNAG antibody activities and C′1q deposition activity (log₁₀ transformed) from mares (Fig. 5) and foals (Fig. 6) were significantly correlated (P = 0.0035 and P = 0.0004, respectively). Antibody activities against PNAG and C′1q deposition activity in foals at the age of 1 day were strongly (R = 0.89 and 0.92, respectively) as well as significantly (P < 0.0001 for both) correlated with those of mares (Fig. S2 and S3).

The targeted number of CFU for challenging foals was 5,000 CFU, split evenly between the right and left lungs. The median challenge dose for the 22 foals was 7,200 (range, 4,600 to 15,100). The challenge dose did not differ significantly (P = 0.2891) between vaccinees (median, 7,900; range, 4,600 to 15,100) and controls (median, 6,400; range, 5,100 to 8,400). There was no significant difference (P = 0.2633) in the challenge doses administered to foals in year 1 (median, 6,400; range, 4,600 to 8,400) and year 2 (median, 7,900; range, 4,900 to 15,100). In year 1, the challenge dose did not differ significantly (P = 0.7758) between vaccinated foals (median, 6,800; range, 4,600 to 8,400) and unvaccinated foals (median, 6,200; range, 5,400 to 8,400). Similarly, in year 2, the challenge dose did not differ significantly (P = 0.2547) between vaccinated foals (median, 8,500; range, 4,900 to 15,100) and unvaccinated foals (median, 7,000; range, 5,100 to 7,900). The CFU with which foals were challenged was not significantly different (P = 0.7396) between foals that developed pneumonia (median, 7,200; range, 4,900 to 15,100; n = 16 foals) and foals that did not develop pneumonia (median, 7,299; range, 4,600 to 8,400; n = 6 foals).

The primary clinical outcome was development of pneumonia attributed to R. equi infection. The proportion of control foals that developed pneumonia (88%; 7/8) was not significantly (P = 0.3512) greater than that for vaccinees (64%; 9/14). The proportion of foals that developed pneumonia was significantly (P = 0.0124) greater in 2020 (100%; 11/11) than in 2018 (45%; 5/11). Among foals that developed pneumonia, the age at onset of pneumonia did not differ significantly between vaccinees (median, 25 days; range, 21 to 32 days) and controls (median, 27 days; range, 23 to 33 days). There were no significant differences between the vaccinated foals and the control foals for any of the findings of twice-daily physical examination, weekly thoracic ultrasonography, weekly hematology, or duration of treatment for pneumonia (Table 2). Using generalized linear modeling, foals with pneumonia had significantly (P < 0.05) lower antibody activities for PNAG and C′1q deposition activity than foals that remained healthy, irrespective of vaccination status (Table 3; Fig. 1 and 4).

All procedures for this study were reviewed and approved by the Texas A&M Institutional Animal Care and Use Committee (protocol numbers AUP IACUC 2016-0233 and IACUC 2019-0347) and the University Institutional Biosafety Committee (permit number IBC2017-105). The foals used in this study were university owned, and permission for their use was provided in compliance with the Institutional Animal Care and Use Committee procedures.

A group of 22 healthy, pregnant Quarter Horse mares and foals were used for this project. All foals were healthy, had adequate transfer of passive immunity as assessed by a semiquantitative immunoassay (SNAP foal IgG test; IDEXX, Westbrook, ME, USA), and had results of complete blood counts (CBCs) within reference ranges. Mares were randomly assigned during midgestation to 1 of 2 experimental groups: group 1 mares were vaccinated with PNAG (n = 14), and group 2 mares were sham vaccinated with 1 ml of sterile medical-grade physiological saline (i.e., 0.9% NaCl) solution (PSS). Mares in the vaccine group received 200 μg of synthetic pentamers of β-1→6-linked glucosamine conjugated to tetanus toxoid (ratio of oligosaccharide to protein, 35 to 39:1; 5GlcNH₂-TT; AV0328; Alopexx Enterprises, LLC, Concord, MA, USA) diluted to 900 μl in PSS combined with 100 μl of a water-in-oil adjuvant (Montanide Gel 01; Seppic, Inc., Courbevoie, France). Mares were vaccinated and given boosters (group 1) or sham vaccinated (group 2) at 6 and 3 weeks prior to their projected foaling dates. This sample size was based on the magnitude of prior observed effects of the vaccine (19) and the following assumptions: (i) significance of 5%; (ii) power of >80%; (iii) 75% (6/8) of control foals developing pneumonia; and (iv) 7% (1/14) of principal foals developing pneumonia. The study was conducted over 2 years: each year, we included 7 vaccinated mares and their foals and 4 unvaccinated mares and their foals.

Prior to infection with R. equi, each foal’s lungs were auscultated and examined by thoracic ultrasonography to document absence of preexisting pulmonary disease. Ultrasound examinations were repeated weekly after infection to monitor for lung lesions. At age 6 days, foals were sedated and infected transendoscopically with 15 ml of R. equi suspension infused into each mainstem bronchus (right and left).

Foals were infected with a virulent strain of R. equi originally isolated from a pneumonic foal from Texas (strain EIDL 5-331) that we have used previously to infect foals at age 28 days (19, 41, 42). This strain was streaked onto a brain heart infusion (BHI) agar plate (Bacto brain heart infusion; BD, Becton, Dickinson and Company, Sparks, MD, USA). One CFU was incubated overnight at 37°C in 50 ml of BHI broth on an orbital shaker at approximately 240 rpm. The bacterial cells were washed 3 times with 1× phosphate-buffered saline (PBS) by centrifugation for 10 min at 3,000 × g and 4°C. The final washed pellet was resuspended in 30 ml of sterile medical-grade PBS to a final concentration of approximately 167 CFU/ml, yielding a total CFU count of approximately 5 × 10³ in 30 ml. Half of this challenge dose (15 ml with 2.5 × 10³ CFU) was administered transendoscopically to the left mainstem bronchus and the other half (15 ml with 2.5 × 10³ CFU) was administered to the right mainstem bronchus. Approximately 200 μl of challenge dose was saved to confirm the concentration (dose) administered and to verify virulence of the isolate using multiplex PCR (43). This challenge dose was based on evidence that younger foals are more susceptible to infection (23), being 5-fold greater than that reported to cause pneumonia in 50% of foals infected at 4 to 7 days of age (23) and being 200-fold less than the dose we have used reliably to cause pneumonia in foals infected at approximately 28 days of age (19, 41, 42).

Blood samples were drawn into clot tubes from mares and foals between 18 and 24 h after birth to harvest serum. Blood was also collected in EDTA tubes for CBC testing at ages 18 to 24 h, 6 days (prior to infection) and then weekly after infection through the age of 55 days. Transendoscopic tracheobronchial aspiration (T-TBA) was performed at the time of onset of clinical signs for any foals developing pneumonia and at approximately age 55 days for all foals (end of study) by washing the tracheobronchial tree with sterile PBS solution delivered through a triple-lumen, double-guarded sterile tubing system (MILA International, Inc., Erlanger, KY, USA).

Humoral antibody responses were assessed in sera by indirectly quantifying concentrations by enzyme-linked immunosorbent assay (ELISA) from absorbance values of PNAG-specific total IgG. ELISA plates (MaxiSorp, Nalge Nunc International, Rochester, NY, USA) were coated with 0.6 μg/ml of purified PNAG (19) diluted in sensitization buffer (0.04 M PO₄, pH 7.2) overnight at 4°C. Plates were washed 3 times with PBS with 0.05% Tween 20, blocked with 120 μl PBS containing 1% skim milk for 1 h at 37°C, and washed again. Serum samples were added in 100-μl volumes, in duplicate, to the ELISA plate and incubated for 1 h at 37°C. Samples were initially diluted in incubation buffer (PBS with 1% skim milk and 0.05% Tween 20) to 1:100 and then serially diluted 2-fold to determine total IgG antibody activities. A positive control from a horse previously immunized with the 5GlcNH₂-TT vaccine and known to have a high titer and normal horse serum known to have a low titer were included in each assay for total IgG antibody activities. After 1 h incubation at 37°C, the plates were washed 3 times as described above. Rabbit anti-horse IgG whole molecule conjugated to alkaline phosphatase (Sigma-Aldrich, St. Louis, MO, USA) was used to detect binding. Plates were washed again, and PNPP (p-nitrophenyl phosphate) substrate (1 mg/ml) was added to plates. Plates were incubated for 15 to 60 min at 22°C in the dark. Optical densities were determined at 405 nm by using microplate readers. Total IgG endpoint antibody activities were calculated by linear regression using a final OD₄₀₅ value of 0.5 to determine the reciprocal of the maximal serum dilution reaching this value.

Quantitation of the serum antibody-mediated deposition of equine complement component Cˊ1q onto purified PNAG was determined by ELISA. ELISA plates were sensitized with PNAG, and then horse serum was added in 50-μl volumes, after which 50 μl of 6% or 10% intact, normal horse serum was added. After 60 min incubation at 37°C, plates were washed and either 100 μl of affinity-purified goat anti-human Cˊ1q, which also binds to equine Cˊ1q, diluted 1:20,000 in incubation buffer was added or the same antibody directly conjugated to alkaline phosphatase was added at 1.5 μg/ml. Plates were incubated at room temperature for 60 min. For assays with unconjugated anti-Cˊ1q, plates were washed, and 100 μl of rabbit anti-goat IgG whole molecule conjugated to alkaline phosphatase diluted 1:1,000 in incubation buffer added; a 1-h incubation at room temperature was then carried out. Washing and development of the color indicator were then carried out as described above, and values for Cˊ1q deposition were determined as the OD₄₀₅ value obtained in the highest serum concentration. This is referred to as the Cˊ1q deposition activity.

From birth until age 5 days, foals were observed twice daily for signs of disease. Beginning at age 5 days (the day prior to infection), rectal temperature, heart rate, respiratory rate, signs of increased respiratory effort (abdominal lift, flaring nostrils), presence of abnormal lung sounds (crackles or wheezes, evaluated for both hemithoraces), presence of abnormal tracheal sounds, coughing, signs of depressed attitude (subjective evidence of increased recumbence, lethargy, reluctance to rise), and nasal discharges were monitored, and results were recorded twice daily through 8 weeks (end of study). Thoracic ultrasonography was performed on the day of infection (prior to infection) and then weekly to identify evidence of peripheral pulmonary consolidation or abscess formation consistent with R. equi pneumonia. Foals were considered to have pneumonia if they demonstrated ≥3 of the following clinical signs: coughing at rest, depressed attitude (reluctance to rise, lethargic attitude, increased recumbency), rectal temperature of >39.4°C, respiratory rate of ≥60 breaths/min, or increased respiratory effort (manifested by abdominal lift and nostril flaring). Foals were diagnosed with R. equi pneumonia if they had ultrasonographic evidence of pulmonary abscessation or consolidation with a maximal diameter of ≥2.0 cm, positive culture of R. equi from T-TBA fluid and cytologic evidence of septic pneumonia from T-TBA fluid. The primary outcome was the proportion of foals diagnosed with R. equi pneumonia. Secondary outcomes included the duration of days meeting the case definition, the sum of total maximum diameters of ultrasound lesions (TMDs) observed at any single examination, and the sum of the TMDs for each examination over the study period. The TMD was determined by summing the maximum diameters of each lesion recorded in the 4th to the 17th intercostal spaces from each foal at every examination; the sum of the TMDs over time incorporates both the duration and severity of lesions. Foals diagnosed with R. equi pneumonia were treated with a combination of clarithromycin (7.5 mg/kg orally [p.o.] every 12 h [q12h]) and rifampin (7.5 mg/kg p.o. q12h) until both clinical signs and thoracic ultrasonography lesions had resolved. Foals also were treated as deemed necessary by attending veterinarians (A.I.B., N.D.C., and A.E.S.) with flunixin meglumine (0.6 to 1.1 mg/kg p.o. q12h to q24h) for inflammation and fever.

The effects of group (i.e., PNAG vaccination status of the mare), year of study, and year by group interaction on serum antibody and Cˊ1q deposition activity of mares or foals were assessed using generalized linear modeling using the glm command with Gaussian link using R statistical software (44). Endpoint antibody activities were log₁₀ transformed to meet distributional assumptions of modeling. Correlations between antibody activities of PNAG antibody and C′1q deposition activity for mares were represented graphically, and the Pearson correlation coefficient was calculated along with a P value testing the hypothesis that the Pearson correlation coefficient was 0 using the ggpubr package in R. The proportion of foals that developed pneumonia was compared between the vaccinated and unvaccinated groups using a Fisher exact test using the fisher.test command in R. Significance for all analysis was set at a P value of <0.05.