Host-directed therapy in foals can enhance functional innate immunity and reduce severity of Rhodoco

Pneumonia caused by the intracellular bacterium Rhodococcus equi is an important cause of disease and death in immunocompromised hosts, especially foals. Antibiotics are the standard of care for treating R. equi pneumonia in foals, and adjunctive therapies are needed. We tested whether nebulization with TLR agonists (PUL-042) in foals would improve innate immunity and reduce the severity and duration of pneumonia following R. equi infection. Neonatal foals (n = 48) were nebulized with either PUL-042 or vehicle, and their lung cells infected ex vivo. PUL-042 increased inflammatory cytokines in BAL fluid and alveolar macrophages after ex vivo infection with R. equi. Then, the in vivo effects of PUL-042 on clinical signs of pneumonia were examined in 22 additional foals after intrabronchial challenge with R. equi. Foals infected and nebulized with PUL-042 or vehicle alone had a shorter duration of clinical signs of pneumonia and smaller pulmonary lesions when compared to non-nebulized foals. Our results demonstrate that host-directed therapy can enhance neonatal immune responses against respiratory pathogens and reduce the duration and severity of R. equi pneumonia.

Rhodococcus equi is a facultative, intracellular bacterium that preferentially replicates within macrophages1 and causes a pyogranulomatous pneumonia in foals and immunocompromised people that resembles tuberculosis. Infection of foals with R. equi occurs during the first weeks after birth2, when foals are more susceptible to infection with this pathogen3, probably because of the naivety/immaturity of both innate and adaptive immune responses4,5,6,7,8,9. Adaptive immune responses require more time to develop such that neonates of all species have a window during which they are not able to be protected by vaccines and consequently depend heavily on innate immune responses10. Even if foals could generate an effective adaptive immune response to vaccination, no vaccine against R. equi pneumonia is commercially available despite decades of research. There is evidence that maternal vaccination against R. equi protects foals against clinical pneumonia11; that approach, however, has some limitations such as it is dependent on the dam’s immune status and adequate immune response to vaccination, ingestion of sufficient colostrum in the first 24–36 h after birth (there is no transfer of immunoglobulins through the placenta in the equine species), etc. Other preventive methods such as transfusion of R. equi-specific hyperimmune plasma are not completely effective and carry some risks for foals12,13,14,15. Chemoprophylaxis with macrolides or other classes of antimicrobials is not recommended because of emerging resistance of R. equi (and other bacteria) to these drugs16. Thus, control of R. equi pneumonia in foals relies on antimicrobial treatment of foals affected with either clinical or subclinical pneumonia17.

Macrolides have been the standard antibiotic used for decades to treat R. equi16. Resistance to macrolides in R. equi is an emerging problem in equine medicine that has been linked to the practice of widespread antimicrobial treatment of foals with subclinical pneumonia at farms with endemic R. equi17,18,19,20,21,22. Because effective alternative antimicrobials for treating foals with R. equi infections are limited18,23,24,25,26, there is an urgent need for non-antibiotic approaches such as host-directed therapy (HDT) to fight these pathogens. Immunotherapy targeting innate immune responses in neonates is an important strategy for HDT because the immune system of foals is immature4,5,6,7,8,9 and infection with R. equi appears to occur early in life2 when foals are known to be more susceptible to this bacterium3.

Toll-like receptors (TLRs) are among the principal signaling receptors for initiating innate immunity, resulting in various effects including cytokine production27. The intracellular TLR9 recognizes bacterial unmethylated cytosine-phosphate-guanine oligodeoxynucleotides (CpG-ODNs)28. We have previously demonstrated that a B class CpG-ODN induced expression of pro-inflammatory cytokines by immune cells of foals both ex vivo and in vivo4,29,30,31, including interferon-gamma (IFN-γ). Evidence exists that younger foals have lower levels of IFN-γ at birth than older foals7,8, and that stimulation of phagocytes with IFN-γ promotes killing of intracellular bacteria such as R. equi32,33 and Mycobacterium tuberculosis (Mtb)34. The TLR2 recognizes lipoteichoic acid and peptidoglycan of Gram + bacteria. TLR2−/− mice are more susceptible to both R. equi infection35 and Mtb36 indicating that activation of the TLR2 signaling pathway can be important in defense against intracellular bacteria. PUL-042 (Pulmotect, Inc., Houston, TX) is an inhaled product that combines two synthetic TLR agonists, Pam2CSK4 (ligand of TLR2/6) and a class C CpG-ODN M362 (ligand of TLR9). This compound drug has been tested in many species against different viral, bacterial, and fungal pathogens and has shown promising results of increasing host survival and reducing pathogen burden37,38,39,40,41,42.

The objective of our research was to test whether nebulization with PUL-042 in foals would improve innate immune responses and reduce the severity and duration of pneumonia due to intrabronchial challenge with R. equi of foals at 28 days of age. We demonstrate that: (1) a combination of TLR2/6 and TLR9 agonists (PUL-042) administered via nebulization at various doses and on multiple occasions is safe for neonatal foals; (2) aerosol treatment of neonatal foals with PUL-042 increased interleukin (IL)-6 and IFN-γ concentration in the supernatant of alveolar macrophages (AMs) upon ex vivo infection; (3) nebulization of PUL-042 in foals challenged intrabronchially with live virulent R. equi reduced the duration of fever, clinical pneumonia, and cough, as well as reduced the size of pulmonary lesions compared to foals that were not nebulized. Our results indicate that nebulization with PUL-042 enhanced functional responses by AMs and reduced the severity of clinical pneumonia compared to non-nebulized foals.

Nebulization with PUL-042 is safe for newborn foals

Forty-three foals (36 foals from ex vivo study, seven from in vivo challenge study; Fig. 1A,B) were nebulized multiple times with PUL-042 in this study, and none had any detectable adverse reaction to the nebulization such as increase in respiratory effort and rate, lacrimation, cough, nasal discharge, etc. The percentage of macrophages significantly increased (Fig. 2A) and the percentage of neutrophils decreased (Fig. 2B) with age in BAL cells in all foals, irrespective of treatment. The percentage of lymphocytes did not change with age or treatment group (Fig. 2C). There was no indication of systemic inflammation attributable to nebulization with PUL-042. Serum amyloid A (Fig. 2D), IL-6 (Fig. 3A) and TNF-α (Fig. 3B) decreased with age for all groups independent of their aerosolization treatment. Interleukin-10 concentrations in serum were not altered by nebulization (Fig. 3C). Plasma fibrinogen concentration did not change significantly following treatment (data not shown). These results indicated that nebulization with PUL-042 was safe for newborn foals and did not induce either systemic or local inflammation.

Nebulization with PUL-042 influences cytokine production in BAL fluid and AMs infected ex vivo with R. equi

We performed BALs in 48 foals before and after six nebulizations in the second and third week after birth with either 2.8% glycerol or one of three different doses of PUL-042, and collected both BAL cells and fluid (Fig. 1A). Foals nebulized with PUL-042 doses 2× and 4× had increased TNF-α in BAL fluid compared with foals receiving glycerol (Fig. 3D), but no other cytokine (IL-6, IL-10; Fig. 3E,F) concentrations in BAL fluid were altered by treatment. Interleukin-1α and IL-1β were not detected in BAL fluid from foals at either time-point and were not included in the analysis.

Alveolar macrophages from nebulized foals were collected and either kept in media (uninfected, baseline) or infected ex vivo with R. equi (Fig. 1A). The ratio of the concentration of infected/uninfected wells were used because of the high inter-individual variation in baseline production of cytokines. Interleukin-10 and TNF-α in infected cells increased relative to uninfected AMs in all treatment groups including the control foals (Fig. 4A,B), suggesting an age-related increase in response to infection. The AMs from foals nebulized with the two highest doses of PUL-042 that were infected ex vivo with R. equi produced greater IL-6 (Fig. 4C) and IFN-γ compared to cells from foals nebulized with 2.8% glycerol alone (Fig. 4D), indicating PUL-042 resulted in an increased immune response to infection but did not alter responses in uninfected cells. Interleukin-1α and IL-17 were not detected in the supernatant of cultured AMs at either time-point and were not included in the analysis.

Nebulization with PUL-042 does not improve ex vivo killing of R. equi by AMs

Phagocytosis of virulent R. equi by AMs from nebulized foals (Fig. 1A) significantly increased with age in animals of all groups at age 22 days compared with age 2 days (T0; Fig. 5A). Nebulization with PUL-042 did not increase phagocytosis in AMs (Fig. 5A), nor did it improve killing of R. equi intracellularly (ratio of T48/T0; Fig. 5B). Because the balance of pro- and anti-inflammatory cytokines is more important than their individual concentrations, we examined the ratio of stimulated IFN-γ expression (ratio of infected to uninfected cells) and IL-10 expression (ratio of infected to uninfected cells), and found that it was associated with killing of R. equi (Fig. 5C; R value − 0.4740).

Nebulization with PUL-042 reduces severity of pneumonia

After we ensured the safety of nebulization of PUL-042 in 48 foals, and its ability to modulate the immune responses of AMs following ex vivo infection with R. equi through both pro- and anti-inflammatory cytokine production, we examined whether nebulization with PUL-042 before and after intrabronchial infection with live, virulent R. equi decreased the severity of the resultant pneumonia in foals (Fig. 1B). All foals in all groups experimentally infected were confirmed having R. equi clinical pneumonia based on our case-definition: ≥ 2 physical signs, evidence of lung abscessation, and both presence of R. equi and cytologic evidence of suppurative pneumonia with presence of degenerate neutrophils in T-TBA samples. Twenty-two additional foals were randomly assigned to 3 groups: control not nebulized, nebulized before and after infection with 2.8% glycerol (PUL-042 diluent), and nebulized before and after infection with PUL-042. Foals that were nebulized with PUL-042 following the infection had a significantly (P < 0.05) shorter duration of clinical signs of pneumonia, coughing, and fever (Fig. 6A), shorter duration of detection (Fig. 6B), and smaller total sum (Fig. 6C) of TUS lesions when compared to foals that did not receive nebulization, but not when compared to foals nebulized with glycerol only.

In this study we found that nebulization of foals with PUL-042 after intrabronchial challenge with R. equi reduced the severity of pneumonia in foals. This may be an important finding, because it could be a potential ancillary treatment to antimicrobials for foals with natural R. equi pneumonia. It has been hypothesized that HDTs that modulate TNF-α as an adjunct to canonical chemotherapy shorten therapy duration, reduce pathology, and limit tuberculosis relapse in experimental models by decreasing exacerbated inflammation43. Additionally, as seen in mice with influenza pneumonia and antiviral therapy37, PUL-042 could have a synergistic effect when used in combination with antimicrobials, and it could potentially decrease either the dose of antibiotic or duration of treatment, aiding in the reduction of pressure for development of antimicrobial resistance. In our study, foals were nebulized before clinical pneumonia developed, thus, antimicrobial treatment and nebulization treatments were not contemporaneous. The synergy between concurrent administration of macrolides and PUL-042 in foals with pneumonia merits further evaluation. Surprisingly, we also observed that nebulization procedure with glycerol only (our nebulized control group) was beneficial to reduce severity of disease. Although duration of fever, duration of clinical signs, and duration of lung lesions tended to be higher in the glycerol only group, that difference was not significant. We are not able to explain the reason for this finding. Glycerin compounds are solvents for chemicals in e-cigarettes, and glycerin-based vapors have been shown to increase inflammatory responses in the lungs of mice44,45,46 and to increase cytokine secretion by human AMs and bronchial epithelium47,48,49,50; however, electronic cigarettes contain glycerol in much higher doses (15–100%) than used in our study (2.8%). Glycerol can stimulate pro-inflammatory cytokine secretion by human monocyte-derived macrophages and cultured airway epithelial cells51,52. To the authors’ knowledge, the effects of nebulized glycerol on innate immune responses of equine pulmonary cells have not been investigated, but it is plausible (as suggested by our findings) that nebulization of glycerol solution alone activates innate immune responses, albeit to a lesser extent than the PUL-042. It is possible that glycerol, at low concentrations such as used in this study (2.8%), induces a beneficial inflammatory response. We speculate, however, that the small sample size of the nebulized glycerol vehicle control group precluded us from detecting a significant difference between the PUL-042 and glycerol vehicle. Nevertheless, our results indicate that nebulization with PUL-042 significantly reduced the severity of disease in treated foals.

We measured different cytokines in the supernatant of AMs collected from nebulized foals and infected ex vivo with R. equi. Two important pro-inflammatory cytokines, IL-6 and IFN-γ, were only significantly increased in foals treated with PUL-042. There is evidence that young foals are deficient in IFN-γ, and this deficiency has been theorized to explain why foals are more susceptible to R. equi pneumonia7,8. We have shown previously that a B class CpG-ODN can induce INF-γ in foals, but whether this would translate to a better clinical outcome upon in vivo intrabronchial infection with R. equi, has yet to be determined. Alveolar macrophages are the main cell infected with R. equi; there is, however, evidence that pulmonary epithelial cells have an important role interacting and responding to intracellular bacteria such as R. equi and Mtb53,54,55. There is also evidence that epithelial cells are responsible for inducible resistance conferred by PUL-042 against pneumonia in mice40. Responses to PUL-042 by equine epithelial cells have not been evaluated, but they could be potentially important for lung immunity in vivo. Additionally, the concentration of the pro-inflammatory cytokine TNF-α and anti-inflammatory cytokine IL-10 increased in the fluid supernatant of infected AMs from foals from all groups.

We wanted to determine if nebulization with PUL-042 would increase bactericidal capacity of foal AMs. After collecting BAL cells from foals after six aerosol treatments with PUL-042 and infecting them with live R. equi, we did not observe either an increase in either phagocytosis or in killing of R. equi when compared with foals that received glycerol 2.8% alone. Despite our hypothesis that AMs from foals treated with PUL-042 would kill R. equi better than AMs collected from untreated foals, our negative findings were not unexpected. There is evidence that treating macrophages without the presence of epithelial cells is not sufficient to increase killing of R. equi42, and these cells were not present in our culture system where AMs were infected. Moreover, clearance of R. equi can be mediated by other innate immune cells including neutrophils11,56,57. We conclude that while nebulization with PUL-042 did not directly appear to impact killing of R. equi by AMs, it modulated immune responses to favor improved functional responses that in the pulmonary environment might enhance the capability of pulmonary cells to kill intracellular pathogens.

Because we observed effects of PUL-042 on cytokine production, and these were directly related to bacterial killing33,34,35, we wanted to determine next if cytokine production and bacterial killing were correlated. The CFU ratio (T48/T0; representing bacterial replication) was indeed associated with the ratio of IFN-γ divided by IL-10 in response to infection. The lower the CFU ratio (meaning greater bacterial killing), the higher was the cytokine ratio, suggesting that a balance between pro- and anti-inflammatory cytokines is necessary for intracellular killing of pathogens.

We also showed in this study that nebulization of PUL-042 in foals resulted in no adverse effects such as lacrimation, increased respiratory effort, or nasal discharge in any of the 43 foals nebulized multiple times with PUL-042. We also did not observe any signs of systemic inflammatory responses induced by PUL-042, because concentrations of pro- or anti-inflammatory cytokines or SAA did not increase in serum of foals after nebulization with any nebulization treatment. We found that foals had significantly higher concentrations of SAA at age 2 days than age 28 days, regardless of the treatment group. Serum amyloid A is an acute-phase protein that rapidly increases with pneumonia or when there is an inflammatory stimulus58. To our knowledge, there is no report evaluating in foals either the age-related change in SAA during the first month after birth, nor following nebulization. Possible explanations for our findings are that 2-day-old foals have higher SAA after ingesting colostrum, or that nebulization procedure per se induced inflammation. There is no evidence that nebulization increases SAA, and because we do not have age-matched control foals that were not nebulized in the ex vivo study, we cannot ascertain what caused the age-related reduction in SAA.

We observed a significant decrease in the % of neutrophils and increase in the % of macrophages in BAL fluid from nebulized foals, irrespective of treatment, while the % of lymphocytes remained unchanged. Hostetter et al. (2017) also observed that the % of neutrophils decreased and the % macrophages increased in the first month when compared to 1-week-old foals, but the differences were not significant59. In our experience, in the first 3 days after birth the % of neutrophils are higher than in 1 week-old-foals, which would explain the discrepancy between the two studies.

An increase in the % of neutrophils in BAL fluid would have indicated a transient local inflammation following nebulization, because neutrophil infiltration in the lungs is considered a marker for inflammation. In the case of infection with Streptococcus pneumoniae (Spn) in mice, however, protection against lethal Spn pneumonia conferred by a nebulized Spn lysate was independent of neutrophil recruitment to the lungs60. We do not have evidence in our study that lung inflammation occurred following nebulization, other than an increased in TNF-α in BAL fluid of foals receiving PUL-042 2× and 4× but not 6×. We consider this to be a positive feature of the treatment in newborn foals because exacerbated inflammation can worsen the severity of pneumonia. All other cytokines remained unchanged in the BAL fluid of nebulized foals.

We recognize that additional studies are necessary to optimize the timing and dose for potentially stronger effects of PUL-042 as well as to overcome anatomical challenges to nebulization of horses, such as their longer tracheas, obligate nasal breathing, etc. Nevertheless, our findings that PUL-042 stimulates innate immunity of foals are promising considering the naivety/immaturity of their immune system in the first month after birth4,5,7,8,31. It is important to note that R. equi experimental infections (including the challenge model used in our study) usually induce a more severe pneumonia than observed in natural infection, likely due to the higher number of infecting organisms used in experimental challenges. Thus, it is possible that nebulization with PUL-042 might be more effective against natural infection with lower doses. Nevertheless, our results show potential for improving neonatal immune responses against bacterial respiratory pathogens, and reducing severity of clinical pneumonia following intrabronchial challenge with R. equi.