**3. Results**

All the horses, initially asymptomatic, showed respiratory signs related to equine asthma exacerbation, after the transition from the pasture to the stable. Indeed, after 2 days from the beginning of the test (T1), clinical examination showed cough and dyspnea in all subjects. However, dyspnea showed a quick remission after environmental change and drug treatment at T3.

Histological score, and NKA and IL-8 immunohistochemistry scores of bronchial biopsies revealed no significant differences between the different biopsy sites nor in EAaffected horses in every experimental time, and in slaughtered horses (Table 1).

With the exception of the first sampling performed, endoscopic biopsies in EA-affected horses were found to be evaluable in at least one of the four samples each horse in each phase, and therefore samples of all phases in all cases were included in the statistical evaluation, both for histology and for immunohistochemistry.

The histopathology evaluation of the bronchial samples collected in the slaughtered horses ranged from a score of 2–3, allowing to consider those subjects as not affected by any respiratory problem, and to include them all as control-cases (see Supplementary Material Table S1). The histological score of EA-affected horses is detailed in Supplementary Material, Table S1. Relevant pictures of histological features are reported in Figure 3.

The comparison of histological scores of control vs EA-affected horses at four experimental points (T0; T1; T2; T3) demonstrated a significant difference at T0 (*p* = 0.001), in the asymptomatic phase; at T1 (*p* = 0.007), in the early exacerbation phase; at T2 (*p* = 0.007), in the late exacerbation phase; and at T3 (*p* = 0.007), in the remission phase (Table 2).


**Table 1.** Comparison of histological, NKA immunoreactivity score and IL-8 immunoreactivity score acquired from bronchial biopsies, obtained in four standardized position, from control horses and EA-affected horses at T0, T1, T2, T3. Median value (min-max) and *p* value—Kruskal–Wallis test.

**Figure 3.** Examples of histological samples of bronchial biopsy in the EA horses. (**<sup>a</sup>**,**b**). Low-magnification images of biopsies, obtained with tissue twisting technique with biopsy forceps before tearing. All tissue layers are observed (surface epithelium, extracellular matrix and airway smooth muscle). HE, bar 800 μm. (**c**) Marked mucosal hyperplasia (arrow), and the presence of a moderate inflammatory submucosal infiltrate of lymphocyte (asterisk). HE, bar 200 μm. (**d**) Epithelial desquamation involving over 70% of the mucous layer (arrow). HE, bar 200 μm. (**e**) Severe diffuse lymphocytic infiltrate (asterisk) and presence of mucosal gland (arrow). HE, bar 200 μm. (**f**) Severe submucosal fibrosis (asterisk). HE, bar 200 μm.


**Table 2.** Comparison of median value (min-max) of histological, NKA and IL-8 immunoreactivity scores acquired from bronchial biopsies, between control horses and EA-affected horses at T0, T1, T2, T3. Mann–Whitney test. a,b: different letters indicate significant differences (*p* < 0.05); A, B: different letters indicate significant differences (*p* < 0.01).

> Median histological scores of bronchial biopsies of EA-affected horses during the different experimental phases (T0, T1, T2, T3), evidenced a significant increase moving from the asymptomatic phase (T0) to the remission phase (T3) with significant difference between T0 and T3 (*p* = 0.01) (Table 3).

**Table 3.** Comparison of median value (min-max) of histological, NKA and IL-8 immunoreactivity scores acquired from bronchial biopsies, between EA-affected horses at T0, T1, T2, T3. Friedman test with Dunn's multiple comparison test as post hoc test. a,b: different letters indicate significant differences (*p* < 0.05).


The results for each immunohistochemical marker are reported in Tables 4 and 5. Relevant pictures of the IHC on control and EA-affected horses for NKA and IL-8 are reported in Figures 4 and 5, respectively.


T3 4, diffuse

4, diffuse

3, diffuse

3,

focal n.e. n.e. 4,

diffuse

4, diffuse

3, diffuse

4, diffuse

3, diffuse B: bronchus, P: proximal, D: distal; n.e.; not evaluable.

4, diffuse

4,

diffuse n.e. 3,

diffuse

3, diffuse

4, diffuse

4, focal

4,

focal n.e. n.e. 3,

focal

2, focal

2, focal

**Figure 4.** NKA immunohistochemical stain (brown color), bar 200 μm. (**a**) Bronchial mucosa of a control horse. Weak and diffuse cytoplasmic positivity of the respiratory epithelium. (**b**) EA-case 3, distal right bronchus, T2. Moderate and diffuse cytoplasmic positivity to the epithelial cells of the respiratory mucosa, associated with an intense nuclear positivity. (**c**) EA-case 1 horse, proximal right bronchus, T3. Intense, diffuse, cytoplasmic and nuclear positivity, mainly localized in the more superficial layers of the hyperplastic respiratory epithelium. (**d**) EA-case 3 horse, distal right bronchus, T3. Intense, diffuse cytoplasmic and nuclear localized mainly in the superficial layers of the epithelium. (**e**) EA-affected 6, proximal left bronchus, T1. Positivity confined to the luminal layer of the epithelium. (**f**) EA-case 2, distal left bronchus, T3. Predominantly nuclear positivity.

**Figure 5.** IL-8 immunohistochemical stain (brown color), bar 200 μm. (**a**) Bronchial mucosa of a control horse. Focal and weak cytoplasmic positivity to the respiratory epithelium. (**b**) EA-case 2, distal right bronchus, T3. Strong and diffuse positivity to the epithelial cells of the mucosa. (**c**) EA-case 2, proximal right bronchus, T3. Note the intense immunopositivity in all layers of the hyperplastic mucosa. (**d**) EA-case 2, distal right bronchus, T2. Note the distribution of positivity in the most superficial layers of the mucosa. (**e**) EA-affected horse 6, proximal left bronchus, T3. The positivity appears to be confined to the most superficial layer of the epithelium also in this case. (**f**) EA-case 5, proximal left bronchus, T1. Focal positivity confined to isolated respiratory epithelial cells.

In control horses, NKA immunolabelling was weak, widespread cytoplasmic in the epithelial bronchial cells. In EA-affected horses, NKA immunolabelling appeared moderate to strong staining at the cytoplasm and occasionally in the nucleus (was found in a minority of epithelial cells also in the nucleus) (Figure 4c,d). In many cases, the immunopositivity was confined to the apical portion of epithelial cells of the airway mucosa (Figure 4e).

The expression of IL-8 was cytoplasmatic. IL-8 positivity was seen in epithelial cells of control horses with a discontinuous, focal, weak staining. IL-8 immunolabelling in equine asthma instead appeared multifocal to diffuse, cytoplasmic, and moderate to intense (Figure 5b,c), in some cases confined to isolated respiratory epithelial cells (Figure 5e,f).

NKA immunopositivity scores of bronchial biopsies showed a significant increase of intensity of positivity, moving from control horses to EA-affected horses at T2 (*p* = 0.04), and at T3 (*p* = 0.04), while no significant differences were showed for signal distribution and cell localization between groups (Table 2). No differences for NKA immunohistochemistry results (intensity of positivity, signal distribution, cell localization), were recorded in EAaffected horses, among the experimental times (Table 3).

At last, IL-8 immunopositivity score, subdivided on intensity of positivity and signal distribution, recorded no significant differences between control horses and EA-affected horses (Table 2), nor in EA-affected horses in each experimental time (Table 3).
